UvA-DARE (Digital Academic Repository) Translational studies in X-linked adrenoleukodystrophy Engelen, M. Link to publication

Transcription

1 UvA-DARE (Digital Academic Repository) Translational studies in X-linked adrenoleukodystrophy Engelen, M. Link to publication Citation for published version (APA): Engelen, M. (2012). Translational studies in X-linked adrenoleukodystrophy General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 24 Nov 2017

2

3 Translational studies in X-linked adrenoleukodystrophy

4 Translational studies in X-linked adrenoleukodystrophy Thesis, University of Amsterdam, Amsterdam ISBN: Cover: Joo Yeon Engelen - Lee Lay-out: Stephan Kemp Printed by Ipskamp Drukkers B.V. Copyright 2012 Marc Engelen, The Netherlands All rights reserved. No part of this thesis may be reproduced, stored in a retrieval system or transmitted in any way or by any means without prior permission of the author.

5 Translational studies in X-linked adrenoleukodystrophy Academisch proefschrift ter verkrijging van de graad van doctor aan d e Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Aula der Universiteit op vrijdag 16 november 2012, te 13:00 uur door Marc Engelen geboren te Nijmegen

6 Promotiecommissie Promotores: Prof. dr. B.T. Poll The Prof. dr. M. de Visser Co-promotores: Dr. S. Kemp Dr. B.M. van Geel Overige leden: Prof. dr. F. Baas Prof. dr. J.B. van Goudoever Prof. dr. M.S. van der Knaap Prof. dr. F.A. Wijburg Prof. dr. M.A.A.P. Willemsen Faculteit der Geneeskunde The work described in this thesis was carried out at the laboratory Genetic Metabolic Diseases and the Departments of Neurology and Pediatric Neurology of the Academic Medical Center, University of Amsterdam, The Netherlands. The research was financially supported by the Departments of Neurology and Pediatric Neurology of the Academic Medical Center and grants from the European Leukodystrophy Association ( I5 & I4), the European Union Framework Programme 7 (LeukoTreat ), the Stop ALD Foundation and the Netherlands Organization for Scientific Research ( ).

7 Gastopponent: Prof. dr. P.A. Aubourg Printing of this thesis was financially supported by the Academic Medical Center, University of Amsterdam, the Department of Neurology of the Academic Medical Center, Actelion Pharmaceuticals, Genzyme, Shire and the Stop ALD Foundation.

8 When a thing is done, it s done. Don t look back. Look forward to your next objective. George C. Marshall

9 Voor Joo Yeon en Philip

10 Table of contents Part I Introduction Chapter 1 X-linked adrenoleukodystrophy (X-ALD) clinical 11 presentation and guidelines for diagnosis, follow-up and management Orphanet Journal of Rare Diseases (2012) 7: 51 Part II In vitro studies Chapter 2 Cholesterol-deprivation increases mono-unsaturated 37 very long-chain fatty acids in skin fibroblasts from patients with X-linked adrenoleukodystrophy Biochimica et Biophysica Acta (2008) 1781: Chapter 3 Bezafibrate lowers very long-chain fatty acids in 51 X-linked adrenoleukodystrophy fibroblasts by inhibiting fatty acid elongation Journal of Inherited Metabolic Diseases, in press Part III Clinical trials Chapter 4 Lovastatin in X-linked adrenoleukodystrophy 65 The New England Journal of Medicine (2010) 362: Chapter 5 Bezafibrate for X-linked adrenoleukodystrophy 75 PLoS ONE (2012) 7:e41013 Part IV Extension of the phenotype Chapter 6 X-linked adrenomyeloneuropathy due to a novel missense 81 mutation in the ABCD1 start codon presenting as demyelinating neuropathy Journal of the Peripheral Nervous System (2011) 16: Chapter 7 The clinical, biochemical and genetic spectrum of X-linked 85 adrenoleukodystrophy in women: a cross-sectional cohort study manuscript in preparation Part V Summary and general discussion Chapter 8 Summary, general discussion, future research and 105 implications for clinical practice

11 Part VI Nederlandse samenvatting Chapter 9 Van gen naar ziekte; X-gebonden adrenoleukodystrofie 109 Nederlands Tijdschrift voor Geneeskunde (2008) 152: Chapter 10 Samenvatting in het Nederlands en beschouwing 119 Appendix Acknowledgements / Dankwoord 125 Curriculum vitae 131 List of publications 133

12

13 Introduction Chapter 1 1 X-linked adrenoleukodystrophy (X-ALD): clinical presentation and guidelines for diagnosis, follow-up and management Marc Engelen 1,3, Stephan Kemp 2,3, Marianne de Visser 1, Björn M. van Geel 1,4, Ronald J.A. Wanders 2, Patrick Aubourg 5,6 and Bwee Tien Poll-The 1,3 1 Department of Neurology; 2 Department of Clinical Chemistry, Laboratory Genetic Metabolic Diseases; 3 Department of Pediatric Neurology/Emma Children s Hospital, Academic Medical Center, University of Amsterdam, The Netherlands. 4 Department of Neurology, Medical Center Alkmaar, Alkmaar, The Netherlands. 5 Assistance Publique des Hôpitaux de Paris, Department of Pediatric Neurology, Hôpital Kremlin-Bicêtre; 6 INSERM UMR745- Université Paris-Descartes, Paris, France. Orphanet Journal of Rare Diseases (2012) 7(1):51 11

14 Chapter 1 Abstract 1 X-linked adrenoleukodystrophy (X-ALD) is the most common peroxisomal disorder. The disease is caused by mutations in the ABCD1 gene that encodes the peroxisomal membrane protein ALDP which is involved in the transmembrane transport of very long-chain fatty acids (VLCFA; >C22). A defect in ALDP results in elevated levels of VLCFA in plasma and tissues. The clinical spectrum in males with X-ALD ranges from isolated adrenocortical insufficiency and slowly progressive myelopathy to devastating cerebral demyelination. The majority of heterozygous females will develop symptoms by the age of 60 years. In individual patients the disease course remains unpredictable. This review focuses on the diagnosis and management of patients with X-ALD and provides a guideline for clinicians that encounter patients with this highly complex disorder. 12

15 Introduction Definition X-ALD is a metabolic disorder characterized by impaired peroxisomal beta-oxidation of very long-chain fatty acids (VLCFA; > C22), which is reduced to about 30% of control levels (Kemp et al 2004; Singh et al 1984). Consequently, there is an accumulation of VLCFA in plasma and all tissues, including the white matter of the brain, the spinal cord and adrenal cortex (Moser et al 1981). It is caused by mutations in the ABCD1 gene located on the X-chromosome (Mosser et al 1993). Mutations in this gene cause the absence or dysfunction of ALDP, a peroxisomal transmembrane protein that transports VLCFacyl-CoA esters from the cytosol into the peroxisome (van Roermund et al 2008; Ofman et al 2010). 1 Epidemiology With an estimated birth incidence of 1 in 17,000 newborns (male and female) (Bezman and Moser, 1998), X-ALD is the most common peroxisomal disorder. It occurs in all regions of the world (Kemp et al 2001). Now that newborn screening has become technically feasible and may be implemented in some parts of the world (Hubbard et al 2009) the true prevalence might be even higher. Clinical features of X-ALD A short history In retrospect, the first cases of X-ALD were probably described in the late 19 th century. In 1897 Heubner described a young boy with rapidly progressive neurologic deterioration consistent with X-ALD, classified as having diffuse sclerosis on autopsy (Heubner, 1897). This designation was used at the time for any disease of the white matter causing hardening of the tissue. Cases of diffuse sclerosis that resemble X-ALD were also described in 1899 by Ceni (Ceni, 1899) and in 1910 by Haberfield and Spieler (Haberfeld and Spieler, 1910). Shortly thereafter Schilder suggested that diffuse sclerosis was too vague and proposed a more accurate pathological classification of the leukodystrophies (Schilder, 1912; Schilder, 1913; Schilder, 1924). He described several cases with lesions in the cerebral white matter and perivascular inflammation that he named encephalitis periaxialis diffusa. The syndrome of rapidly progressive cerebral demyelination with inflammatory changes in the white matter on autopsy became known as Schilder s disease. Many cases were described in the following fifty years (Hoefnagel et al 1962). Nevertheless, Siemerling and Creutzfeldt are often credited for having described the first case of X-ALD because they were the first to describe the association of acute cerebral demyelination with clinical and pathological signs of Addison s disease (Siemerling and Creutzfeldt, 1923). X-ALD has remained an enigmatic disease for most of the 20 th century, although some suspected it might be a metabolic disorder affecting both the central nervous system and adrenal glands (Blaw, 1970; Forsyth et al 1971). Based on the presence of lipid inclusions in adrenal glands, testis, brain and Schwann cells, Schaumburg and Powers suggested the term adrenoleukodystrophy for the disorder and speculated it might be a lipid storage 13

16 Chapter 1 1 disorder (Schaumburg et al 1972). This hypothesis was confirmed when post-mortem biochemical analysis of brain, adrenal glands and serum of seven patients revealed the presence of cholesterol esters with high amounts of fatty acids longer than C22. These fatty acids were referred to as very long-chain fatty acids (VLCFA) (Igarashi et al 1976). X-linked adrenoleukodystrophy now became a distinct clinical entity with an associated biochemical marker that could be used to confirm the diagnosis in readily accessible materials like blood cells and plasma (Moser et al 1981). Already in 1910 Addison s disease associated with spastic paraplegia was described (von Neusser and Wiesel, 1911). Familial Addison disease with or without spastic paraplegia was usually considered to be a variant of hereditary spastic paraplegia (Harris-Jones and Nixon, 1955; Penman, 1960). In 1976, an X-linked adult onset progressive myelopathy that was often associated with Addison s disease was reported (Budka et al 1976). A year later, five more cases were described and the term adrenomyeloneuropathy (AMN) was proposed because of the involvement of the adrenal cortex, spinal cord and peripheral nerves (Griffin et al 1977). With plasma VLCFA analysis as newly identified biomarker, AMN was soon found to be a form of X-ALD thus extending the phenotypic spectrum of X-ALD (Moser et al 1981). Now, we classify the symptomatology of X-ALD in children and adults in several phenotypes (Table 1). They will be discussed here in more detail. Phenotypes in male X-ALD patients Addison-only Adrenocortical insufficiency (or even an Addisonian crisis) can be the presenting symptom of X-ALD in boys and men, years or even decades before the onset of neurological symptoms. X-ALD is a frequent cause of Addison s disease in boys and adult males (Hsieh and White, 2011; Laureti et al 1996a), in particular when circulating adrenocortical autoantibodies are absent (Laureti et al 1996b). A percentage as high as 35 has been reported (Laureti et al 1996a). However, this figure may be overestimated. Indeed, in a study of 40 Dutch men with adrenocortical insufficiency, no patients with X-ALD were identified (B.M. van Geel and J. Assies, unpublished data). Recognizing that Addison s disease is due to X-ALD has important implications for genetic counseling but also management. It is therefore important to consider X-ALD in any boy or adult male presenting with Addison s disease. Adrenocortical insufficiency initially affects the glucocorticoid function of the adrenal, but the mineralocorticoid function is ultimately deficient in approximately half of the X-ALD patients (Laureti et al 1996a). Cerebral ALD (childhood, adolescent and adult) These phenotypes are the most rapidly progressive and devastating phenotypes of X-ALD. They most frequently present in childhood (childhood cerebral ALD; CCALD), however never before the age of 2.5 years (van Geel et al 1997; Moser et al 2001). The onset of CCALD is insidious, with deficits in cognitive abilities that involve visuospatial and visuomotor functions or attention and reasoning. In boys and adolescents it initially results in a decline of school performance. These early clinical symptoms are often misdiagnosed as attention deficit hyperactivity disorder and can delay the diagnosis of CCALD. As the disease progresses, more overt neurologic deficits become apparent, which include withdrawn or hyperactive 14

17 Introduction Progression Endocrine disorder Peripheral neuropathy Behavioral and cognitive disorder Myelopathy White matter lesions on brain MRI Table 1: The X-ALD phenotypes % Age of onset CCALD Extensive + - often AD rapid Extensive + rare often AD rapid AdolCALD Possible at a preclinical stage ACALD 2-5 > 21 + or - Extensive + possible often AD rapid AMN often AD and slow (no cerebral testicular disease) insufficiency - Sensorymotor, mostly axonal, rarely demyelinating > 18 + Wallerian degeneration of the corticospinal tracts in brainstem, pons and internal capsules rapid often AD and testicular insufficiency + Sensory-motor, mostly axonal > 18 + parieto-occipital, frontal, or involving the centrum semiovale 20% of AMN patients over a period of 10 years AMN (cerebral disease) > AD - slow Very rare +/- AD rare (< 1%) + Very rare Wallerian degeneration of the corticospinal tracts in brainstem, pons and internal capsules is less common than in males with AMN highly variable, mostly > 40 Addison only Decreasing with age Women with Unknown X-ALD how many are symptomatic - = absent; + = present; CCALD = childhood cerebral ALD; AdolCALD = adolescent cerebral ALD; ACALD = adult cerebral ALD; AMN = adrenomyeloneuropathy; AD = Addison-disease 1 15

18 Chapter 1 1 Figure 1: MRI of the brain in a case of childhood cerebral ALD showing characteristic extensive white matter changes in the parieto-occipital region and internal capsules on FLAIR sequences (A). This area is initially affected in about 80% of cases of cerebral ALD. The rim enhances after administration of gadolinium on T1 sequences (B). In about 20% of cases the site of initial involvement in cerebral ALD is the frontal white matter as shown on this FLAIR image of a different patient with cerebral ALD (C), with prominent rim enhancement after administration of gadolinium on a T1 weighted image (D). behavior, apraxia, astereognosia, auditory impairment ( word deafness reflecting impairment in acoustic analysis of word sounds), decreased visual acuity, hemiparesis or spastic tetraparesis, cerebellar ataxia and seizures. At this stage progression is extremely rapid and devastating. Affected boys can lose the ability to understand language and walk within a few weeks. Eventually, patients are bedridden, blind, unable to speak or respond, 16

19 Introduction requiring full-time nursing care and feeding by nasogastric tube or gastrostomy. Usually death occurs two to four years after onset of symptoms, or - if well-cared for - patients may remain in this apparent vegetative state for several years (Moser et al 2001). The rapid neurologic decline of CCALD is caused by a severe inflammatory demyelination process primarily affecting the cerebral hemispheres. Postmortem histopathological examination of brain tissue reveals extensive demyelination with perivascular infiltration of lymphocytes and macrophages that resembles, to some extent, the demyelinating lesions seen in multiple sclerosis (Powers et al 1992). In 80% of patients the initial demyelinating lesion is localized in the splenium of the corpus callosum and progresses to involve the adjacent parieto-occipital white matter (van der Knaap and Valk, 2005). Alternatively, the initial demyelinating lesions may occur in the genu of corpus callosum and then progress symmetrically or asymmetrically to the white matter of the frontal lobes (van der Knaap and Valk, 2005). The lesions may also initially involve the pyramidal tracts within the pons or the internal capsules and then extend into the white matter of the centrum semiovale. Brain MRI shows abnormal signal intensities (increased signal on T2 and FLAIR sequences, decreased signal on the T1 sequence) in the corpus callosum, parieto-occipital or frontal white matter or pyramidal tracts within the brainstem, pons and internal capsules (Figure 1). Atypical patterns of demyelination, for instance highly asymmetrical disease have been described (Wang et al., 2006). Initially, the demyelinating lesions do not show enhancement on T1 sequences after intravenous gadolinium administration. The presence of gadolinium enhancement of demyelinating lesions occurs usually in a second stage when the disease starts to progress rapidly, reflecting severe inflammation and disruption of the blood brain barrier. This transition to the rapidly progressive neuroinflammatory stage may occur very early, even when the demyelinating lesions are restricted to the corpus callosum or pyramidal tracts or later, once the demyelinating lesions have already significantly extended into the cerebral white matter. In the absence of biomarkers to predict this evolution, brain MRI remains the only tool to detect this evolution in an early stage. A scoring system to grade the demyelinating abnormalities on brain MRI has been developed by Loes (Loes et al., 1994). This score correlates well with neurologic symptoms when the demyelinating lesions involve the parieto-occipital white matter. However, lesions in the frontal white matter can produce severe symptoms (especially behavioral), while the Loes score is still low. Less frequently, cerebral demyelination as the presenting phenotype of X-ALD occurs in adolescence (AdolCALD) or adulthood (ACALD). The symptomatology in these patients strongly resembles CCALD, but the initial progression of symptoms usually is slower. In adults, the early cognitive decline is rarely recognized by their families and friends or at work. As the disease progresses, psychiatric disturbances mimicking schizophrenia or psychosis are not uncommon (Garside et al 1999). In those cases, the diagnosis of X-ALD is often delayed; particularly when no family history of X-ALD is present and when clinical symptoms of Addison s disease are absent. 1 Approximately 10% of boys or adolescents with cerebral ALD may not develop the rapidly progressive neuroinflammatory stage of the disease. The same may occur in men with ACALD or in men with AMN who develop secondary cerebral demyelination (see below). This cerebral demyelinating form of X-ALD is often referred to as chronic or arrested cerebral X-ALD. The cerebral demyelinating process arrests spontaneously and the patient can remain stable for a decade or even longer. But even after a year period of stability, 17

20 Chapter 1 sudden onset of rapid neurologic deterioration may occur, reflecting a full progression to the neuroinflammatory stage of the disease. 1 Adrenomyeloneuropathy (AMN) Virtually all patients with X-ALD who reach adulthood develop AMN, usually in the 3 rd and 4 th decade. Initial symptoms are limited to the spinal cord and peripheral nerves. Patients develop gradually progressive spastic paraparesis, sensory ataxia with impaired vibration sense, sphincter dysfunction (mostly urinary), pain in the legs and impotence (Moser et al 2007). The clinical burden of peripheral nerve involvement is usually moderate and difficult to assess because of prominent spinal cord symptoms. If polyneuropathy is investigated electrophysiologically, an axonopathy is found in the majority of the patients (van Geel et al 1996). Rarely, the initial symptomatology may be that of a peripheral neuropathy, either demyelinating or axonal (Kararizou et al 2006; Engelen et al 2011). Before the availability of MR imaging, AMN was often misdiagnosed as multiple sclerosis or hereditary spastic paraparesis (HSP). The pathological basis of AMN is a noninflammatory distal axonopathy that involves the long tracts of the spinal cord and, to a lesser extent, the peripheral nerves (Powers et al., 2000). This phenotype is in most cases slowly progressive, causing severe motor disability of the lower limbs over one or two decades but mild or no significant deficits in arms and hands. Brain MRI is normal or may show subtle abnormalities such as moderately increased signal intensities of the pyramidal tracts in brainstem, pons and internal capsules on FLAIR and T2 sequences that reflect likely Wallerian degeneration in patients with longstanding symptoms of AMN (Figure 2). These abnormalities are not considered manifestations of cerebral ALD. However, if the increased signal of the pyramidal tracts becomes more intense and extends beyond the internal capsules into the white matter of the centrum semiovale, this is considered as manifestations of cerebral ALD (Figure 3). Figure 2: MRI of the brain in a patient with AMN showing increased signal in the pyramidal tracts on T2-weighed coronal (A) and axial (B) images indicative of Wallerian degeneration. 18

21 Introduction MRI of the spinal cord eventually shows non-specific atrophy, but no demyelination or gadolinium enhancement as seen in multiple sclerosis. In contrast, magnetization transfer or diffusion tensor-based imaging sequences show significant abnormalities of the spinal cord tracts (Dubey et al., 2005; Fatemi et al., 2005; Smith et al., 2009). A retrospective study revealed that over a period of 10 years, approximately 20% of AMN patients developed additional cerebral demyelination (van Geel et al., 2001). After an initial progression demyelinating lesions can stabilize spontaneously leading to moderate cognitive deficits. However, once the cerebral demyelinating lesions have entered the active phase of neuroinflammation with gadolinium enhancement, the prognosis is as poor as in CCALD. The neurologic symptoms of AMN patients that develop cerebral ALD are identical to those observed in adults with adult cerebral ALD, with additional symptoms of the pre-existing myelopathy. Approximately 70% of AMN patients have adrenocortical insufficiency (Moser et al., 1991). An equal percentage of affected males have subclinical signs of testicular insufficiency (Assies et al., 1997). Clinical symptoms of testicular insufficiency are rare. Hair of patients with AMN is often thin and sparse; these patients often show balding at an early age (Figure 4). This typical scanty scalp hair was first described in 1955 (Harris-Jones and Nixon, 1955). 1 Women with X-ALD As in many X-linked diseases, it was assumed that female carriers remain asymptomatic. However, many women develop AMN-like symptoms (O Neill et al., 1984). Physical examination of a large group of female carriers attending family conferences in the United States revealed that more than 50% had some kind of abnormality on neurologic examination (Moser et al., 2001). An increasing number of symptomatic heterozygous women are identified as the first member of their family to be affected by X-ALD and the real incidence of AMN in heterozygous women is likely to be close to 65% by the age of 60 years (Engelen et al, manuscript in preparation (Chapter 7)). Onset of neurologic symptoms mainly occurs between the 4 th and 5 th decade and they are very similar to those observed in adult males with AMN. Sensory ataxia, fecal incontinence and pain in the legs are however often more prominent in symptomatic women with AMN. Cerebral involvement and adrenocortical insufficiency are rare, 2% and 1%, respectively (el-deiry et al., 1997; Moser et al., 2001). So far, neither have been documented in the Netherlands and France. It is speculated that skewed X-inactivation in neuronal cells may contribute to the manifestation of neurologic symptoms in X-ALD carriers (Maier et al., 2002; Naidu et al., 1997). In a recent study the association of skewed X-inactivation and symptomatic status could not be confirmed (Salsano et al., 2012). However, this may be attributable to differences in mean age between the symptomatic and asymptomatic group in this study. Asymptomatic and pre-symptomatic patients A diagnosis of X-ALD must be followed by extended family screening together with a geneticist. This enables the detection of: 1) heterozygous women who can be offered prenatal or in some countries preimplantation diagnosis for future pregnancies, and 2) boys or adult males who are asymptomatic but at risk to develop cerebral demyelination or adrenocortical insufficiency (Kemp et al 2001). It is important to detect these complications as early as possible for treatment, as described in the section on clinical management. 19

22 Chapter 1 1 Figure 3: MRI of the brain (T2 (A) and FLAIR (C) images; T1 with gadolinium (B, D)) of a patient with AMN who rapidly deteriorated clinically with new symptoms of cognitive decline. On MRI extensive white matter changes were seen in the parieto-occipital white matter and corpus callosum (A), but no enhancement of the lesion after administration of gadolinium (B). A follow-up MRI about 3 months later shows progression of the white matter lesion (C) and there is now faint enhancement of the rim of the lesion after gadolinium administration (D). 20

23 Introduction 1 Figure 4: Thin and scanty scalp hair in a man with X-ALD (AMN phenotype). Evolution over time X-ALD phenotypes are not static. Presymptomatic males are nearly all at risk to develop neurologic (cerebral ALD, AMN) or endocrinologic (Adisson s disease) symptoms. Addisononly males can develop AMN or cerebral ALD and AMN males can develop cerebral demyelination. It is estimated that over a period of 10 years about 20% of patients with AMN will progress to a cerebral phenotype (van Geel et al 2001). Presymptomatic women with X-ALD are mostly at risk to develop AMN. Progression of X-ALD in a specific individual can not be predicted. X-ALD is characterized by the absence of a genotype-phenotype correlation. In spite of identical ABCD1 gene mutations, patients can have markedly divergent neurological and neuropathologic characteristics (Berger et al 1994; Kemp et al 1994). There is ample evidence that other genetic and/or environmental factors may influence the clinical presentation of X-ALD. Segregation analysis suggests that the phenotypic variability is due to at least one autosomal modifier gene (Maestri and Beaty, 1992; Smith et al 1991). Head trauma may be an environmental factor triggering the onset of cerebral ALD (Raymond et al 2010). Other brain lesions, like a stroke, may also trigger cerebral demyelination in patients with X-ALD. Etiology and Pathophysiology X-ALD is caused by mutations in the ABCD1 gene located on the X-chromosome. So far, 600 different mutations have been identified (see ABCD1 encodes ALDP, 21

24 Chapter 1 1 a peroxisomal transmembrane protein involved in the transport of VLCFA-CoA esters from the cytosol into the peroxisome (van Roermund et al 2008; Ofman et al 2010; Kemp et al 2011). ALDP deficiency leads to impaired VLCFA beta-oxidation (about 30% of normal) and the accumulation of VLCFA-CoA esters in cells. The VLCFacyl-CoA esters in the cytosol are incorporated into various lipid fractions and are also substrate for further elongation to even longer fatty acids (Kemp et al 2005; Ofman et al 2010). ELOVL1 (elongation of verylong-chain-fatty acids) is the single elongase catalyzing the synthesis of both saturated VLCFA (C26:0) and mono-unsaturated VLCFA (C26:1). ELOVL1 expression is not increased in X-ALD fibroblasts, but increased synthesis of VLCFA is due to increased substrate availability (Ofman et al 2010). Various in vitro experiments have demonstrated that VLCFA accumulation is toxic. VLCFA are extremely hydrophobic and their rate of desorption from biological membranes is about 10,000 times slower than that of long-chain fatty acids causing disruptive effects on the structure, stability and function of cell membranes (Ho et al 1995; Knazek et al 1983). Excess of VLCFA in cultured cells decreases the ACTH-stimulated cortisol release by human adrenocortical cells (Whitcomb et al 1988) and causes cell death in astrocytes and oligodendrocytes (Hein et al., 2008). In vivo, VLCFA cause oxidative stress and oxidative damage to proteins (Fourcade et al., 2008; Powers et al., 2005) microglial activation and apoptosis (Eichler et al., 2008). VLCFA-induced oxidative stress may contribute to axonal damage in the spinal cord of Abcd1 knock-out mice that develop an AMN-like phenotype (Fourcade et al., 2008; Pujol et al., 2002). In addition, a dysregulation of oxidative phosphorylation, adipocytokine and insulin signaling pathways, and protein synthesis was recently shown in the spinal cord of Abcd1 knock-out mice (Schluter et al., 2011). The neuropathological hallmark of AMN is an axonopathy with microgliosis but without significant myelin changes. It is therefore likely that the primary consequence of VLCFA accumulation impairs the capacity of oligodendrocytes and Schwann cells to sustain axonal integrity, resulting in axonal damage. The VLCFA homeostasis in X-ALD is disturbed (Kemp et al., 2005). This may contribute to the destabilization of the myelin sheath and impair the function of astrocytes and microglia which play an important role in myelin integrity (Eichler et al., 2008; Hein et al., 2008). Not all males with X-ALD develop cerebral ALD corroborating the notion that additional triggers, genetic, epigenetic and/or environmental, modify this process. In the absence of a relevant X-ALD mouse model that develops cerebral demyelination with neuroinflammation, the pathogenic processes that result in cerebral demyelination and subsequently severe neuroinflammation remains therefore poorly understood. Biochemical and Molecular Diagnosis Newborn screening is now technically feasible (Hubbard et al 2009). It is based on the measurement of C26:0 lysophosphatidylcholine (26:0-lyso-PC) in dried blood spots. It will lead to identification of pre-symptomatic patients with X-ALD. Whether this screening is implemented is a matter of national policy and dependent on ethical considerations. 22

25 Introduction If X-ALD is suspected in a male with neurological symptoms (with or without typical brain MRI abnormalities) or Addison s disease, demonstration that VLCFA are elevated in plasma confirms the diagnosis. For women with X-ALD, the diagnostic test of choice is mutation analysis of the ABCD1 gene, because 15% of women with X-ALD have normal plasma VLCFA levels (Moser et al 2007). Family screening follows the same recommendations. The ABCD1 gene is the single causative gene for X-ALD (Mosser et al 1993). It is an X-linked inherited disorder. Therefore all daughters of an affected male are obligate carriers whereas his sons can never be affected. When a woman carries the gene for X-ALD, there is a 50% probability for each pregnancy that the gene is transmitted to a son or daughter. The frequency of de novo mutations in the index case is estimated to be around 4% (Wang et al 2011) which indicates that the ABCD1 mutation occured in the germ line. There is evidence of gonadal or gonosomal mosaicism in less than 1% of patients which means an increased risk of an additional affected offspring. 1 Table 2: Differential diagnosis of chronic myelopathy with normal MRI Deficiencies: Vitamin B12, folic acid, copper Hereditary spastic paraparesis with amyotrophy Infections (e.g., HTLV-1, HIV) Primary lateral sclerosis Radiation myelopathy Cerebrotendinous xanthomatosis Metachromatic leukodystrophy Krabbe disease Differential diagnosis Symptoms In boys and men presenting with Addison s disease, including those who have only a glucocorticoid deficiency, and in whom there are no detectable steroid-21-hydroxylase or other organ-specific antibodies, X-ALD should be considered and determination of plasma VLCFA levels must be performed (Laureti et al 1998; Mukherjee et al 2006). Young boys and adult males presenting with cognitive and neurological symptoms with (usually enhancing) white matter lesions on brain MRI should be tested for X-ALD. In adult men, the most common presenting symptom of X-ALD is a chronic myelopathy. In the past, AMN was often misdiagnosed as primary progressive multiple sclerosis or hereditary spastic paraparesis. After ruling out a compressive myelopathy by MRI of the spinal cord and other common causes of chronic myelopathy (some possible diagnoses for chronic myelopathy with a (near) normal MRI are summarized in Table 2) (Wong et al 2008), X-ALD should be considered. A clinical clue to the diagnosis can be the presence of adrenocortical insufficiency and early baldness. However, even in the absence of clinical signs of adrenocortical insufficiency AMN should be considered. 23

26 Chapter 1 1 A category that remains underdiagnosed relatively often are women with AMN. Physicians familiar with X-ALD are aware that it is not uncommon that women with X-ALD develop a myelopathy (Moser et al 1991). However, most neurology textbooks do not list X-ALD in the differential diagnosis for chronic myelopathy in women. The diagnosis can only be excluded by ABCD1 mutation analysis. As mentioned before, there is an increasing number of X-ALD families in which the index case is a woman with clinical symptoms of AMN. Increased VLCFA Increased plasma VLCFA are not pathognomonic for X-ALD or might even be false positive, because of hemolysis of the sample or dietary causes, for example a ketogenic diet (Theda et al 1993). Alternatively, some dietary products like rapeseed oil or mustard seed oil that are rich in erucic acid (C22:1) can result in the lowering of C26:0 therefore causing a false negative result (Ann Moser, personal communication). If metabolic screening reveals increased VLCFA, the next step is to confirm the diagnosis by performing ABCD1 mutation analysis. If this is negative, it is important to consider other peroxisomal disorders, such as 1) peroxisomal biogenesis disorders with late onset of symptoms (Regal et al 2010; Sevin et al 2011) (Ebberink et al 2012), and 2) peroxisomal acyl-coa oxidase 1 (ACOX1) or D-bifunctional enzyme (DBP) deficiency with late onset of symptoms. This requires further testing for bile acid intermediates, phytanic and pristanic acids in plasma, and plasmalogens in erythrocytes (Wanders et al 2010). A further metabolic screening of peroxisomal dysfunction must also be performed on skin fibroblasts, at least by performing catalase immunofluorescence after culturing cells at 37 C, but also at 40 C (Wanders and Waterham, 2006). White matter changes on MRI In males with confluent white matter changes, X-ALD should be considered, especially when there is increased signal intensity on T2-weighed and FLAIR sequences in the parietooccipital region and the splenium of the corpus callosum (van der Knaap and Valk, 2005). Rim enhancement of demyelinating lesions is usually observed when boys or adult males present with overt neurological symptoms from cerebral ALD. Therefore, intravenous administration of gadolinium is advised if X-ALD is considered. However, in about 20% of males with X-ALD, the white matter changes occur predominantly in the genu of corpus callosum and frontal lobes, or involve the pyramidal tracts with extension in the white matter of the centrum semiovale (van der Knaap and Valk, 2005). The brain MRI pattern of cerebral X-ALD is usually readily recognized by neuroradiologists. A helpful algorithm for diagnostic work-up of patients with white matter changes on brain MRI is suggested by Schiffmann and Van der Knaap (Schiffmann and van der Knaap, 2009). There are other peroxisomal disorders that present with MRI abnormalities resembling X-ALD, in particular peroxisomal biogenesis disorders and ACOX1 or DBP deficiency with late onset of symptom (Barth et al 2001). Although the radiologic findings might suggest X-ALD, the clinical presentation is very different and therefore these disorders are easily differentiated. Genetic counseling and prenatal diagnosis Genetic counseling must be offered to the parents of affected boys, adult males and women 24

27 Introduction with X-ALD and their family to detect: 1) carriers who can be offered prenatal diagnosis, and 2) asymptomatic or pre-symptomatic men or women who can benefit from therapeutic interventions. Regular follow-up in presymptomatic males can prevent serious morbidity and mortality. ABCD1 mutational analysis can be performed either on a fresh chorionic villus sample at weeks of pregnancy or on amniotic cells obtained from amniotic fluid after centrifugation at weeks of gestation (Rahil et al 2002). In some countries, pre-implantation genetic diagnosis is available. If the fetus is a female, there is no consensus with respect to prenatal diagnosis and termination of pregnancy, due to the highly variable expression of disease in women with X-ALD. Cases will be evaluated on an individual basis. 1 Clinical Management The flowchart in Figure 5 summarizes the recommendations for follow-up of boys and men with X-ALD. Boys and adult males with X-ALD Follow-up in boys and men with X-ALD is important for two reasons: 1) early detection of adrenocortical insufficiency and 2) early detection of cerebral ALD to propose allogeneic hematopoietic stem cell transplantation (HCT) if a HLA-matched donor or cord blood is available. Despite significant mortality risk, allogeneic HCT remains the only therapeutic intervention that can arrest the progression of cerebral demyelination in X-ALD, provided the procedure is performed very early, i.e., when affected boys or men have no or minor symptoms due to cerebral demyelinating disease (Shapiro et al 2000; Miller et al 2011). In the future, transplantation of autologous hematopoietic stem cells that have been genetically corrected with a lentiviral vector before re-infusion might become an alternative to autologous HCT, once the very encouraging results obtained in the first two treated patients will have been extended to a larger number of patients with cerebral X-ALD (Cartier et al 2009). If boys or men do not have Addison s disease it is recommended that they are evaluated yearly by an endocrinologist for adrenocortical dysfunction by measuring the plasma ACTH levels and performing an ACTH stimulation test (Mukherjee et al 2006). Steroid replacement therapy can then be initiated if necessary. Boys without neurological deficits should be monitored closely for radiological signs of cerebral ALD. CCALD has not been reported before the age of 2.5 years (Kemp et al 2001). We recommend an MRI of the brain every 6 months in boys aged 3 to 12 years old to screen for early signs of CCALD. If symptoms occur suggestive of cerebral ALD (for instance declining school performance) the MRI should be performed at the earliest available opportunity, but it is our experience that the detection of brain MRI abnormalities precedes any detectable cognitive dysfunction by at least 6 months to 1 year. After the age of 12 years, the incidence of CCALD decreases, but an MRI scan must be performed yearly or earlier if new symptoms occur. It is important to detect cerebral ALD as early as possible, preferably in the 25

28 Chapter 1 1 asymptomatic stage with only moderate radiological abnormalities to discuss the possibility to perform allogeneic HCT. Accordingly, if a brain MRI shows abnormalities, even very limited such as an increased signal intensity on T2 or FLAIR sequences in the splenium or genu of the corpus callosum, brain MRI must be repeated within 3 months to evaluate disease progression and in particular to identify the presence of gadolinium rim enhancement of lesions. Because the disease can be very rapidly progressive, it is strongly advised to discuss the possibility of allogeneic HCT as soon as brain MRI abnormalities typical of cerebral ALD are detected. After a successful transplant, the lesions on MRI stabilize and even regress. Treatment results are better the earlier treatment is started (Miller et al 2011). For adult men with or without signs of AMN, we advise evaluation by a neurologist yearly or bi-annually to screen for symptoms of AMN and to administer symptomatic treatment if necessary (for instance, medication against spasticity). Referral to a rehabilitation physician and urologist will often become necessary. Adult men can develop cerebral ALD and in our centers, we offer an MRI of the brain every single year (Moser et al 2001; van Geel et al 2001). There is no proven treatment for cerebral ALD in adults. It seems likely that allogeneic HCT is also effective in adults with early stage cerebral ALD, but there are no published studies or cases describing this treatment. We tend to consider allogeneic HCT in an adult patient with early stage cerebral ALD, after carefully counseling the patient about the lack of evidence for the treatment and the risk of the procedure which is significantly higher than in boys. Whereas the onset of demyelinating lesions involving the corpus callosum and adjacent parieto-occipital or frontal white matter leaves no doubt about the onset of cerebral ALD, the situation is different when there are only slightly increased signal abnormalities in the pyramidal tracts of AMN patients that gradually become more intense and involve the white matter of the centrum semiovale. This can herald the onset of cerebral ALD, but can also reflect Wallerian degeneration in severe AMN. For AMN there is no effective disease modifying therapy available yet. Although Lorenzo s oil (LO) had great promise, several open-label trials have shown that the disease progresses even when plasma VLCFA are normalized by LO treatment (Aubourg et al 1993; van Geel et al 1999). A large randomized placebo-controlled clinical trial was designed to provide a definitive answer, but was unfortunately aborted before completion by the safety monitoring board because of presumed side effects of the placebo treatment. There is also a retrospective study suggesting that if presymptomatic boys are started on LO, it may delay the onset of neurological symptoms (Moser et al 2005). We consider the scientific evidence to support the efficacy of LO weak, and do not offer this treatment to our patients. Regular follow-up in AMN remains important, however, mainly to provide symptomatic treatment. Lovastatin also lowered plasma VLCFA (Singh et al 1998), but a placebo-controlled trial revealed that lovastatin did not have an effect on the C26:0 levels in peripheral blood lymphocytes and erythrocytes nor on the VLCFA content of the low-density lipoprotein fraction (Engelen et al 2010). More research and new treatments strategies are desperately needed, especially for those affected by AMN, which is relentlessly progressive and causes severe disability. Antioxidants reduce markers for oxidative stress and axonal degeneration in the spinal cord of Abcd1 26

29 Introduction 1 Figure 5: Flowchart describing the outpatient management of X-ALD. *If there is no gadolinium enhancement present, consider arrested cerebral ALD and repeat the MRI in 3 months. knockout mice (Lopez-Erauskin et al 2011). Based on this observation a clinical trial with anti-oxidants in X-ALD is ongoing in Spain. Women with X-ALD Women with X-ALD should be evaluated for the development of neurologic symptoms. Since women with X-ALD very rarely develop adrenocortical insufficiency or cerebral involvement, periodic evaluation of adrenocortical function and brain MRI is not mandatory (Moser et al 2007). Greater awareness among physicians that women can develop neurologic symptoms is important for counseling but also to prevent unnecessary diagnostic tests and erroneous diagnosis. We know of cases of women with X-ALD who underwent cervical laminectomy for presumed cervical spondylogenic myelopathy. For symptomatic women with X-ALD, we advise (as for men with AMN) a yearly evaluation by a neurologist to discuss the indication of rehabilitation, the referral to an urologist and treatment of spasticity and neuropathic pain. 27

30 Chapter 1 Prognosis 1 It is debatable whether men with X-ALD can remain really asymptomatic for life. It seems probable that there is only presymptomatic X-ALD in men. Although there is no large prospective cohort study to accurately determine the natural history of the disease, several general observations can be made. It is likely that all patients with X-ALD if they survive into adulthood eventually develop myelopathy, i.e. AMN. Usually, symptoms and signs occur from the 3 rd decade of life, but much earlier, or much later, is possible. The severity and progression cannot be predicted for individual patients. There is marked variability ranging from men with X-ALD that are wheelchair bound by the age of 25 and others who are able to walk with a cane while in their seventies. This is important when counseling patients: symptoms will occur, but only time will tell how severely affected an individual will be. Not every boy or man with X-ALD will develop cerebral ALD. About 35-40% of boys with an ABCD1 mutation will develop CCALD before reaching adulthood. It can still not be predicted who will develop this devastating manifestation of the disease. Previously it was believed that after reaching adulthood this complication was very rare. However, it is now well established that at least 20% of adult males with the AMN phenotype will develop cerebral demyelination later in life (van Geel et al 2001). If this occurs, these patients have the same poor prognosis as boys with inflammatory cerebral ALD. It is well documented that women develop symptoms that resemble AMN (Moser et al 1991), but there is no prospective study with adequate numbers to really estimate what percentage of women become symptomatic. We recently completed a large prospective cohort study to describe the symptomatology of X-ALD in women (Engelen et al, manuscript in preparation (Chapter 7)). Unresolved questions and future research For the majority of patients with X-ALD there is currently no curative or preventive treatment. However, several promising new approaches will hopefully come to fruition in the future. For example, it has been demonstrated in X-ALD cells that small interfering RNA (sirna)-mediated inhibition of ELOVL1 reduces VLCFA synthesis and levels (Ofman et al., 2010). Compounds that can inhibit ELOVL1 are therefore interesting candidates for new preventive treatments. We demonstrated that bezafibrate which is both an effective, safe and well-tolerated compound lowers the levels of VLCFA in X-ALD cells by inhibiting VLCFA synthesis (Engelen et al 2012). A clinical trial to evaluate its in vivo efficacy in X-ALD patients is ongoing. As mentioned before, it is likely that several modifier genes play a role in the onset and severity of AMN, the onset of cerebral ALD and the progression to the rapidly progressive neuroinflammatory stage. However, so far no major modifier genes have been identified that could lead to the development of a genetic test to predict phenotype and disease evolution. The eventual phenotype of X-ALD in an individual will most likely be determined by the combination of several epigenetic and environmental modifiers, most of which have not been identified. Much research is currently focused on identifying these modifiers to 28

31 Introduction better predict clinical outcome in individual patients. An important issue is the systematic characterization of disease manifestations in women with X-ALD. Clinical research has always focused on boys and men and women were often considered carriers with no or minimal symptoms. For adequate counseling this information is important and should be available. 1 For clinicians caring for patients with AMN it is well known that cerebral demyelination can occur. It is important to systematically study if an allogeneic HCT can be successful in these cases and even arrest the progression of myelopathy. Conclusions Pediatricians, endocrinologists, neurologists and psychiatrists may encounter X-ALD, which is a relatively common metabolic disorder, in their practice. The disorder is associated with severe morbidity and mortality in the majority of affected patients. Recognition of X-ALD is highly important, since in some cases treatment is available, such as allogeneic HCT in the early stage of CCALD and endocrine replacement therapy for adrenocortical insufficiency. Furthermore, prenatal testing to prevent unnecessary new cases of this devastating disease is available. List of abbreviations ACALD (Adult cerebral adrenoleukodystrophy), ACOX1 (acyl-coa oxidase 1), ACTH (adrenocorticotropic hormone), AdolCALD (Adolescent cerebral adrenoleukodystrophy), ALDP (adrenoleukodystrophy protein), AMN (adrenomyeloneuropathy), CCALD (childhood cerebral adrenoleukodystrophy), DBP (D-bifunctional protein), FLAIR (fluid attenuated inversion recovery), HCT (hematopoietic stem cell transplantation), LO (Lorenzo s oil), MRI (magnetic resonance imaging), VLCFA (very long-chain fatty acids), X-ALD (X-linked adrenoleukodystrophy). Acknowledgements This work was supported by a grant from the Netherlands Organization for Scientific Research ( to SK). References Assies J, Gooren LJ, Van GB, Barth PG Signs of testicular insufficiency in adrenomyeloneuropathy and neurologically asymptomatic X-linked adrenoleukodystrophy: a retrospective study. Int J Androl 20: Aubourg P, Adamsbaum C, Lavallard-Rousseau MC, Rocchiccioli F, Cartier N, Jambaque I, Jakobezak C, Lemaitre A, 29

32 Chapter 1 Boureau F, Wolf C, Bougneres PF A two-year trial of oleic and erucic acids ( Lorenzo s oil ) as treatment for adrenomyeloneuropathy. N Engl J Med 329: Barth PG, Gootjes J, Bode H, Vreken P, Majoie CB, Wanders RJ Late onset white matter disease in peroxisome biogenesis disorder. Neurology 57: Berger J, Molzer B, Fae I, Bernheimer H X-linked adrenoleukodystrophy (ALD): A novel mutation of the ALD gene in 6 members of a family presenting with 5 different phenotypes. Biochemical and Biophysical Research Communications 205: Bezman L, Moser HW Incidence of X-linked adrenoleukodystrophy and the relative frequency of its phenotypes. Am J Med Genet 76: Blaw ME Melanodermic type leukodystrophy (adreno-leukodystrophy). In: Vinken PJ, Bruyn CW, editors. Handbook of Clinical Neurology. New York: American Elsevier. p Budka H, Sluga E, Heiss WD Spastic paraplegia associated with Addison s disease: adult variant of adrenoleukodystrophy. J Neurol 213: Cartier N, Hacein-Bey-Abina S, Bartholomae CC, Veres G, Schmidt M, Kutschera I, Vidaud M, Abel U, Dal-Cortivo L, Caccavelli L, Mahlaoui N, Kiermer V, Mittelstaedt D, Bellesme C, Lahlou N, Lefrere F, Blanche S, Audit M, Payen E, Leboulch P, l Homme B, Bougneres P, Von KC, Fischer A, Cavazzana-Calvo M, Aubourg P Hematopoietic stem cell gene therapy with a lentiviral vector in X-linked adrenoleukodystrophy. Science 326: Ceni C Über einen interessanten Fall von gliomatöser Infiltration bei der Großhirnhemisphäre. Arch Psychiatr Nervenkr 31: Dubey P, Fatemi A, Huang H, Nagae-Poetscher L, Wakana S, Barker PB, van ZP, Moser HW, Mori S, Raymond GV Diffusion tensor-based imaging reveals occult abnormalities in adrenomyeloneuropathy. Ann Neurol 58: Ebberink MS, Koster J, Visser G, Spronsen Fv, Stolte-Dijkstra I, Smit GP, Fock JM, Kemp S, Wanders RJ, Waterham HR A novel defect of peroxisome division due to a homozygous non-sense mutation in the PEX11β gene. J Med Genet 49: Eichler FS, Ren JQ, Cossoy M, Rietsch AM, Nagpal S, Moser AB, Frosch MP, Ransohoff RM Is microglial apoptosis an early pathogenic change in cerebral X-linked adrenoleukodystrophy? Ann Neurol 63: el-deiry SS, Naidu S, Blevins LS, Ladenson PW Assessment of adrenal function in women heterozygous for adrenoleukodystrophy. J Clin Endocrinol Metab 82: Engelen M, Ofman R, Dijkgraaf MGW, Hijzen M, van der Wardt LA, van Geel BM, de Visser M, Wanders RJA, Poll-The BT, Kemp S Lovastatin in X-Linked Adrenoleukodystrophy. N Engl J Med 362: Engelen M, van der Kooi AJ, Kemp S, Wanders RJ, Sistermans EA, Waterham HR, Koelman JT, van Geel BM, de VM X-linked adrenomyeloneuropathy due to a novel missense mutation in the ABCD1 start codon presenting as demyelinating neuropathy. J Peripher Nerv Syst 16: Engelen M, Schackmann MJ, Ofman R, Sanders RJ, Dijkstra IM, Houten SM, Fourcade S, Pujol A, Poll-The BT, Wanders RJ, Kemp S Bezafibrate lowers very long-chain fatty acids in X-linked adrenoleukodystrophy fibroblasts by inhibiting fatty acid elongation. J Inherit Metab Dis. DOI /s Fatemi A, Smith SA, Dubey P, Zackowski KM, Bastian AJ, van Zijl PC, Moser HW, Raymond GV, Golay X Magnetization transfer MRI demonstrates spinal cord abnormalities in adrenomyeloneuropathy. Neurology 30

33 Introduction 64: Forsyth CC, Forbes M, Cumings JN Adrenocortical atrophy and diffuse cerebral sclerosis. Arch Dis Child 46: Fourcade S, Lopez-Erauskin J, Galino J, Duval C, Naudi A, Jove M, Kemp S, Villarroya F, Ferrer I, Pamplona R, Portero- Otin M, Pujol A Early oxidative damage underlying neurodegeneration in X-adrenoleukodystrophy. Hum Mol Genet 17: Garside S, Rosebush PI, Levinson AJ, Mazurek MF Late-onset adrenoleukodystrophy associated with longstanding psychiatric symptoms. J Clin Psychiatry 60: Griffin JW, Goren E, Schaumburg H, Engel WK, Loriaux L Adrenomyeloneuropathy: a probable variant of adrenoleukodystrophy. I. Clinical and endocrinologic aspects. Neurology 27: Haberfeld W, Spieler F Zur diffusen Hirn-Ruckenmarksklerose im Kindesalter. Dt Z Nervheilk 40: Harris-Jones JN, Nixon PG Familial Addison s disease with spastic paraplegia. J Clin Endocrinol Metab 15: Hein S, Schonfeld P, Kahlert S, Reiser G Toxic effects of X-linked adrenoleukodystrophy-associated, very long chain fatty acids on glial cells and neurons from rat hippocampus in culture. Hum Mol Genet 17: Heubner O Über diffuse Hirnsclerose. Charité Ann 22: Ho JK, Moser H, Kishimoto Y, Hamilton JA Interactions of a very long chain fatty acid with model membranes and serum albumin. Implications for the pathogenesis of adrenoleukodystrophy. J Clin Invest 96: Hoefnagel D, Van den Noort S, Ingbar SH Diffuse cerebral sclerosis with endocrine abnormalities in young males. Brain 85: Hsieh S, White PC Presentation of primary adrenal insufficiency in childhood. J Clin Endocrinol Metab 96:E925-E928. Hubbard WC, Moser AB, Liu AC, Jones RO, Steinberg SJ, Lorey F, Panny SR, Vogt RF, Jr., Macaya D, Turgeon CT, Tortorelli S, Raymond GV Newborn screening for X-linked adrenoleukodystrophy (X-ALD): validation of a combined liquid chromatography-tandem mass spectrometric (LC-MS/MS) method. Mol Genet Metab 97: Igarashi M, Schaumburg HH, Powers J, Kishmoto Y, Kolodny E, Suzuki K Fatty acid abnormality in adrenoleukodystrophy. J Neurochem 26: Kararizou E, Karandreas N, Davaki P, Davou R, Vassilopoulos D Polyneuropathies in teenagers: a clinicopathological study of 45 cases. Neuromuscul Disord 16: Kemp S, Ligtenberg MJL, vangeel BM, Barth PG, Wolterman RA, Schoute F, Sarde CO, Mandel JL, Vanoost BA, Bolhuis PA Identification of A 2 Base-Pair Deletion in 5 Unrelated Families with Adrenoleukodystrophy - A Possible Hot-Spot for Mutations. Biochemical and Biophysical Research Communications 202: Kemp S, Pujol A, Waterham HR, van Geel BM, Boehm CD, Raymond GV, Cutting GR, Wanders RJ, Moser HW ABCD1 mutations and the X-linked adrenoleukodystrophy mutation database: role in diagnosis and clinical correlations. Hum Mutat 18: Kemp S, Valianpour F, Mooyer PA, Kulik W, Wanders RJ Method for Measurement of Peroxisomal Very-Long- 31

34 Chapter 1 Chain Fatty Acid {beta}-oxidation in Human Skin Fibroblasts Using Stable-Isotope-Labeled Tetracosanoic Acid. Clin Chem 50: Kemp S, Valianpour F, Denis S, Ofman R, Sanders RJ, Mooyer P, Barth PG, Wanders RJA Elongation of very long-chain fatty acids is enhanced in X-linked adrenoleukodystrophy. Molecular Genetics and Metabolism 84: Kemp S, Theodoulou FL, Wanders RJ Mammalian peroxisomal ABC transporters: from endogenous substrates to pathology and clinical significance. Br J Pharmacol 164: Knazek RA, Rizzo WB, Schulman JD, Dave JR Membrane microviscosity is increased in the erythrocytes of patients with adrenoleukodystrophy and adrenomyeloneuropathy. J Clin Invest 72: Laureti S, Casucci G, Santeusanio F, Angeletti G, Aubourg P, Brunetti P. 1996a. X-linked adrenoleukodystrophy is a frequent cause of idiopathic Addison s disease in young adult male patients. J Clin Endocrinol Metab 81: Laureti S, Falorni A, Volpato M, Casucci G, Picchio ML, Angeletti G, Luthman H, Brunetti P, Betterle C, Santeusanio F. 1996b. Absence of circulating adrenal autoantibodies in adult-onset X-linked adrenoleukodystrophy. Horm Metab Res 28: Laureti S, Aubourg P, Calcinaro F, Rocchiccioli F, Casucci G, Angeletti G, Brunetti P, Lernmark A, Santeusanio F, Falorni A Etiological diagnosis of primary adrenal insufficiency using an original flowchart of immune and biochemical markers. J Clin Endocrinol Metab 83: Loes DJ, Hite S, Moser H, Stillman AE, Shapiro E, Lockman L, Latchaw RE, Krivit W Adrenoleukodystrophy: a scoring method for brain MR observations. AJNR Am J Neuroradiol 15: Lopez-Erauskin J, Fourcade S, Galino J, Ruiz M, Schluter A, Naudi A, Jove M, Portero-Otin M, Pamplona R, Ferrer I, Pujol A Antioxidants halt axonal degeneration in a mouse model of X-adrenoleukodystrophy. Ann Neurol 70: Maestri NE, Beaty TH Predictions of a 2-locus model for disease heterogeneity: application to adrenoleukodystrophy. Am J Med Genet 44: Maier EM, Kammerer S, Muntau AC, Wichers M, Braun A, Roscher AA Symptoms in carriers of adrenoleukodystrophy relate to skewed X inactivation. Ann Neurol 52: Miller WP, Rothman SM, Nascene D, Kivisto T, DeFor TE, Ziegler RS, Eisengart J, Leiser K, Raymond G, Lund TC, Tolar J, Orchard PJ Outcomes after allogeneic hematopoietic cell transplantation for childhood cerebral adrenoleukodystrophy: the largest single-institution cohort report. Blood 118: Moser HW, Moser AB, Frayer KK, Chen W, Schulman JD, O Neill BP, Kishimoto Y Adrenoleukodystrophy: increased plasma content of saturated very long chain fatty acids. Neurology 31: Moser HW, Moser AB, Naidu S, Bergin A Clinical aspects of adrenoleukodystrophy and adrenomyeloneuropathy. Dev Neurosci 13: Moser HW, Smith KD, Watkins PA, Powers J, Moser AB X-linked adrenoleukodystrophy. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The Metabolic and Molecular Bases of Inherited Disease: McGraw Hill, New York. p Moser HW, Raymond GV, Lu SE, Muenz LR, Moser AB, Xu J, Jones RO, Loes DJ, Melhem ER, Dubey P, Bezman L, Brereton NH, Odone A Follow-up of 89 Asymptomatic Patients With Adrenoleukodystrophy Treated With Lorenzo s Oil. Arch Neurol 62:

35 Introduction Moser HW, Mahmood A, Raymond GV X-linked adrenoleukodystrophy. Nat Clin Pract Neurol 3: Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL, Aubourg P Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: Mukherjee S, Newby E, Harvey JN Adrenomyeloneuropathy in patients with Addison s disease : genetic case analysis. J R Soc Med 99: Naidu S, Washington C, Thirumalai S, Smith KD, Moser HW, Watkins PA X-chromosome inactivation in symptomatic heterozygotes of X-linked adrenoleukodystrophy. Ann Neurol 42:498a. O Neill BP, Moser HW, Saxena KM, Marmion LC Adrenoleukodystrophy: clinical and biochemical manifestations in carriers. Neurology 34: Ofman R, Dijkstra IM, van Roermund CW, Burger N, Turkenburg M, van Cruchten A, van Engen CE, Wanders RJ, Kemp S The role of ELOVL1 in very long-chain fatty acid homeostasis and X-linked adrenoleukodystrophy. EMBO Mol Med 2: Penman RW Addison s disease in association with spastic paraplegia. Br Med J 1:402. Powers JM, Liu Y, Moser AB, Moser HW The inflammatory myelinopathy of adreno-leukodystrophy: cells, effector molecules, and pathogenetic implications. J Neuropathol Exp Neurol 51: Powers JM, DeCiero DP, Ito M, Moser AB, Moser HW Adrenomyeloneuropathy: a neuropathologic review featuring its noninflammatory myelopathy. J Neuropathol Exp Neurol 59: Powers JM, Pei Z, Heinzer AK, Deering R, Moser AB, Moser HW, Watkins PA, Smith KD Adreno-leukodystrophy: oxidative stress of mice and men. J Neuropathol Exp Neurol 64: Pujol A, Hindelang C, Callizot N, Bartsch U, Schachner M, Mandel JL Late onset neurological phenotype of the X-ALD gene inactivation in mice: a mouse model for adrenomyeloneuropathy. Hum Mol Genet 11: Rahil H, Solassol J, Philippe C, Lefort G, Jonveaux P Rapid detection of common autosomal aneuploidies by quantitative fluorescent PCR on uncultured amniocytes. Eur J Hum Genet 10: Raymond GV, Seidman R, Monteith TS, Kolodny E, Sathe S, Mahmood A, Powers JM Head trauma can initiate the onset of adreno-leukodystrophy. J Neurol Sci 290: Regal L, Ebberink MS, Goemans N, Wanders RJ, De ML, Jaeken J, Schrooten M, Van CR, Waterham HR Mutations in PEX10 are a cause of autosomal recessive ataxia. Ann Neurol 68: Salsano E, Tabano S, Sirchia SM, Colapietro P, Castellotti B, Gellera C, Rimoldi M, Pensato V, Mariotti C, Pareyson D, Miozzo M, Uziel G Preferential expression of mutant ABCD1 allele is common in adrenoleukodystrophy female carriers but unrelated to clinical symptoms. Orphanet J Rare Dis 7:10. Schaumburg HH, Richardson EP, Johnson PC, Cohen RB, Powers JM, Raine CS Schilder s disease. Sex-linked recessive transmission with specific adrenal changes. Arch Neurol 27: Schiffmann R, van der Knaap MS Invited article: an MRI-based approach to the diagnosis of white matter disorders. Neurology 72: Schilder PF Zur Kenntnis der sogenannten diffusen Sklerose (über Encephalitis periaxialis diffusa). Z Gesamte Neurol Psychiatr 10:

36 Chapter 1 Schilder PF Zur Frage der Encephalitis periaxialis diffusa (sogenannten diffusen Sklerose). Z Gesamte Neurol Psychiatr 15: Schilder PF Die encephalitis periaxialis diffusa (nebst Bemerkungen über die Apraxie des Lidschlusses). Arch Psychiatr Nervenkr 71: Schluter A, Espinosa L, Fourcade S, Galino J, Lopez E, Ilieva E, Morato L, Asheuer M, Cook T, McLaren A, Reid J, Kelly F, Bates S, Aubourg P, Galea E, Pujol A Functional genomic analysis unravels a metabolic-inflammatory interplay in adrenoleukodystrophy. Hum Mol Genet. Sevin C, Ferdinandusse S, Waterham HR, Wanders RJ, Aubourg P Autosomal recessive cerebellar ataxia caused by mutations in the PEX2 gene. Orphanet J Rare Dis 6:8. Shapiro E, Krivit W, Lockman L, Jambaque I, Peters C, Cowan M, Harris R, Blanche S, Bordigoni P, Loes D, Ziegler R, Crittenden M, Ris D, Berg B, Cox C, Moser H, Fischer A, Aubourg P Long-term effect of bone-marrow transplantation for childhood-onset cerebral X-linked adrenoleukodystrophy. Lancet 356: Siemerling E, Creutzfeldt HG Bronzekrankheit und sklerosierende Enzephalomyelitis. Arch Psychiatr Nervenkr 68: Singh I, Moser AE, Moser HW, Kishimoto Y Adrenoleukodystrophy: impaired oxidation of very long chain fatty acids in white blood cells, cultured skin fibroblasts, and amniocytes. Pediatr Res 18: Singh I, Khan M, Key L, Pai S Lovastatin for X-Linked Adrenoleukodystrophy. N Engl J Med 339: Smith KD, Sack G, Beaty TH, Bergin A, Naidu S, Moser A, Moser HW A genetic basis for the multiple phenotypes of X-linked adrenoleukodystrophy. Am J Hum Genet 49:165. Smith SA, Golay X, Fatemi A, Mahmood A, Raymond GV, Moser HW, van Zijl PC, Stanisz GJ Quantitative magnetization transfer characteristics of the human cervical spinal cord in vivo: application to adrenomyeloneuropathy. Magn Reson Med 61: Theda C, Woody RC, Naidu S, Moser AB, Moser HW Increased very long chain fatty acids in patients on a ketogenic diet: a cause of diagnostic confusion. J Pediatr 122: van der Knaap MS, Valk J X-linked adrenoleukodystrophy. In: Heilmann U, editor. Magnetic Resonance of Myelination and Myelin Disorders. Berlin-Heidelberg-New York: Springer. p van Geel BM, Koelman JH, Barth PG, Ongerboer de Visser BW Peripheral nerve abnormalities in adrenomyeloneuropathy: a clinical and electrodiagnostic study. Neurology 46: van Geel BM, Assies J, Wanders RJ, Barth PG X linked adrenoleukodystrophy: clinical presentation, diagnosis, and therapy. J Neurol Neurosurg Psychiatry 63:4-14. van Geel BM, Assies J, Haverkort EB, Koelman JH, Verbeeten B, Wanders RJ, Barth PG Progression of abnormalities in adrenomyeloneuropathy and neurologically asymptomatic X-linked adrenoleukodystrophy despite treatment with Lorenzo s oil. J Neurol Neurosurg Psychiatry 67: van Geel BM, Bezman L, Loes DJ, Moser HW, Raymond GV Evolution of phenotypes in adult male patients with X-linked adrenoleukodystrophy. Ann Neurol 49: van Roermund CWT, Visser WF, IJlst L, van Cruchten A, Boek M, Kulik W, Waterham HR, Wanders RJA The human peroxisomal ABC half transporter ALDP functions as a homodimer and accepts acyl-coa esters. FASEB J 34

37 Introduction 22: von Neusser E, Wiesel J Die Erkrankungen der Nebennieren. Wien-Leipzig: HöIder Wanders RJ, Waterham HR Biochemistry of mammalian peroxisomes revisited. Annu Rev Biochem 75: Wanders RJ, Ferdinandusse S, Brites P, Kemp S Peroxisomes, lipid metabolism and lipotoxicity. Biochim Biophys Acta 1801: Wang S, Wu JM, Cheng YS Asymmetric cerebral lesion pattern in X-linked adrenoleukodystrophy. J Chin Med Assoc 69: Wang Y, Busin R, Reeves C, Bezman L, Raymond G, Toomer CJ, Watkins PA, Snowden A, Moser A, Naidu S, Bibat G, Hewson S, Tam K, Clarke JT, Charnas L, Stetten G, Karczeski B, Cutting G, Steinberg S X-linked adrenoleukodystrophy: ABCD1 de novo mutations and mosaicism. Mol Genet Metab 104: Whitcomb RW, Linehan WM, Knazek RA Effects of long-chain, saturated fatty acids on membrane microviscosity and adrenocorticotropin responsiveness of human adrenocortical cells in vitro. J Clin Invest 81: Wong SH, Boggild M, Enevoldson TP, Fletcher NA Myelopathy but normal MRI: where next? Pract Neurol 8:

38 Chapter

39 In vitro studies Chapter 2 2 Cholesterol-deprivation increases mono-unsaturated very longchain fatty acids in skin fibroblasts from patients with X-linked adrenoleukodystrophy Marc Engelen 1,2, Rob Ofman 1, Petra A.W. Mooijer 1, Bwee Tien Poll The 2, Ronald J.A. Wanders 1 and Stephan Kemp 1,2 1 Department of Clinical Chemistry, Laboratory Genetic Metabolic Diseases; 2 Department of Pediatric Neurology/ Emma Children s Hospital, Academic Medical Center, University of Amsterdam, The Netherlands Biochimica et Biophysica Acta (2008) 1781(3):

40 Chapter 2 Abstract 2 X-linked adrenoleukodystrophy (X-ALD) is the most common peroxisomal disorder and is characterized by a striking and unpredictable variation in phenotypic expression. It ranges from a rapidly progressive and fatal cerebral demyelinating disease in childhood (CCALD), to the milder slowly progressive form in adulthood (AMN). X-ALD is caused by mutations in the ABCD1 gene that encodes a peroxisomal membrane located ABC half-transporter named ALDP. Mutations in ALDP result in reduced beta-oxidation of very long-chain fatty acids (VLCFA, >22 carbon atoms) in peroxisomes and elevated levels of VLCFA in plasma and tissues. Previously, it has been shown that culturing skin fibroblasts from X-ALD patients in lipoprotein-deficient medium results in reduced VLCFA levels and increased expression of the functionally redundant ALD-related protein (ALDRP). The aim of this study was to further resolve the interaction between cholesterol and VLCFA metabolism in X-ALD. Our data show that the reduction in 26:0 in X-ALD fibroblasts grown in lipoprotein-deficient culture medium (free of cholesterol) is offset by a significant increase in both the level and synthesis of 26:1. We also demonstrate that cholesterol-deprivation results in increased expression of stearoyl-coa-desaturase (SCD) and increased desaturation of 18:0 to 18:1. Finally, there was no increase in [1-14 C]-26:0 beta-oxidation. Taken together, we conclude that cholesterol-deprivation reduces saturated VLCFA, but increases mono-unsaturated VLCFA. These data may have implications for treatment of X-ALD patients with lovastatin. 38

41 In vitro studies Introduction X-linked adrenoleukodystrophy (X-ALD; MIM300100) is a severe metabolic disorder characterized by impaired peroxisomal beta-oxidation of very long-chain fatty acids (VLCFA, >22 carbon atoms) and elevated VLCFA levels in plasma and tissues of patients (Moser et al 1981). The clinical presentation of X-ALD is highly variable, even among affected members of the same family (Moser et al 2001). In the most severe phenotype, affected patients develop cerebral demyelination with inflammation which is rapidly progressive and usually fatal within 2 years after onset. This phenotype presents most frequently in early childhood (childhood cerebral ALD; CCALD), but can also occur in adolescence or adulthood. The adrenomyeloneuropathy (AMN) phenotype usually develops between the 3rd and 4th decade of life and is gradually progressive. The main symptoms are spastic paraparesis and incontinence, caused by progressive myelopathy and peripheral neuropathy. Adrenocortical and testicular insufficiency can occur in isolation (Addison-only), but also in combination with any of the other phenotypes. 2 X-ALD is caused by mutations in the ABCD1 gene (MIM ) (Mosser et al 1993), that encodes a peroxisomal transmembrane protein named the adrenoleukodystrophy protein (ALDP), which is classified as a member of the ATP-binding cassette subfamily D of transporters (Mosser et al 1993; Mosser et al 1994). Although the precise function of ALDP is not clear, absence or dysfunction of ALDP causes impaired peroxisomal beta-oxidation of VLCFA in human skin fibroblasts (Singh et al 1981). Two other peroxisomal transmembrane proteins belonging to the class of ATP-binding cassette transporters have been identified in mammals. These include ALD-related protein (ALDRP) (Lombard-Platet et al 1996) and PMP70 (Kamijo et al 1990). It has been well established that these proteins show functional redundancy, at least partially, and have overlapping substrate specificities. Over-expression of either PMP70 or ALDRP in X-ALD fibroblasts corrects VLCFA beta-oxidation (Kemp et al 1998; Netik et al 1999). Furthermore, in the Abcd1 knockout mouse in which ALDRP was overexpressed the biochemical abnormalities normalized and the neurological AMN-like phenotype was reversed (Pujol et al 2004). This prompted research with the aim to identify ways to induce the expression of ALDRP, encoded by the ABCD2 gene, which is normally expressed at low levels in most tissues (Berger et al 1999). It has been shown that 4-phenylbutyrate (4-PBA) can induce ALDRP expression in human and mouse primary fibroblasts (Kemp et al 1998). Unfortunately, the dosage needed to obtain a biological effect in humans makes it unpractical for clinical application (Kemp et al 1998; Moser et al 2001). In 1998 Singh et al (Singh et al 1998a) reported that beta-oxidation of VLCFA was increased markedly by incubating fibroblasts from X-ALD patients with lovastatin, a cholesterol lowering drug that belongs to the class of HMG-CoA-reductase inhibitors. Later it was shown that culturing fibroblasts from X-ALD patients in medium with lipoprotein-deficient fetal calf serum resulted in a reduction in 26:0 levels and increased ABCD2 expression (Weinhofer et al 2002). Since cholesterol depletion leads to activation of the SREBP-pathway (Brown and Goldstein, 1997), a likely mechanism for the observed effects on 26:0 metabolism would be the increased expression of ALDRP. Subsequent studies showed that the promoter of the 39

42 Chapter 2 2 ABCD2 gene indeed contains a functional sterol regulatory element (SRE) (Weinhofer et al 2002). Based on these observations patients with X-ALD have been treated with lovastatin and a low-fat diet in a small open-label study. An approximately 40% reduction in plasma 26:0 levels was observed in 7 out of 12 patients (Singh et al 1998b; Pai et al 2000). In another study in which patients were treated with simvastatin but without dietary fat restriction, no reduction in plasma VLCFA was observed (Verrips et al 2000). Still, based on the aforementioned observations many patients with X-ALD are being treated with lovastatin. Unfortunately, further studies showed that in Abcd1 knockout mice lovastatin (Yamada et al 2000) and simvastatin (Cartier et al 2000) were unable to reduce VLCFA. On the contrary, even higher levels of VLCFA in tissues, notably the brain, were found. With respect to lovastatin treated patients no data is available on the effect on VLCFA levels in tissues other than plasma and erythrocytes. There are, however, indications that interspecies differences are important. Mouse fibroblasts cultured in lipoprotein-deficient culture medium showed a 40% reduction in 26:0 levels compared to a 80% reduction in human fibroblasts (Weinhofer et al 2005). Therefore, the data from mice experiments cannot be extrapolated to humans directly. The aim of the experiments described in this study was to further resolve the interaction between cholesterol and VLCFA metabolism in X-ALD using skin fibroblasts from patients with X-ALD and controls. Materials and Methods Fatty acid substrates Deuterium-labeled free fatty acids 16,16,16-D 3-16:0 and 17,17,17,18,18-D 5-18:0 were purchased from CDN isotopes (Pointe-Claire, Canada) and 12.5 mm stock solutions in DMSO were prepared. Prior to usage stock solutions were vortex mixed and diluted in Ham F-10 tissue culture medium to their final concentration. All chemicals used were of analytical grade. Cell lines and cell culture Human primary skin fibroblasts cell lines were obtained from patients with X-ALD through the neurology outpatient clinic of the Academic Medical Centre. From each patient written informed consent was obtained. Five cell lines from X-ALD patients were used. The clinical phenotype of the patient, the ABCD1 mutation and the effect of the mutation on ALDP expression by means of immunofluorescence were: ALD005 (AMN, p.ser290x, absent); ALD009 (AMN, p.glu471fs, absent); ALD011 (Addison-only, p.arg464x, absent); ALD013 (AMN, p.met1val, absent) and ALD014 (AMN, p.leu220pro, reduced). Fibroblasts from anonymous controls were obtained from the laboratory cell bank. Cells were grown in HAM F-10 tissue culture medium supplemented with 10% fetal calf serum, penicillin (100 U/mL), streptomycin (100 U/mL) and amphotericin B (250 ng/ml) or in Dulbecco s Modified Eagle Medium (DMEM), additionally supplemented with HEPES (25 mmol/l). All cell lines were used with passage numbers below 20. For cholesterol deprivation 10% lipoprotein-deficient fetal calf serum, containing less than 3 µmol/l of cholesterol, was used (Bodinco B.V., Alkmaar, The Netherlands). For comparison, regular fetal calf serum contains approximately 40

43 In vitro studies 1.5 mmol/l of cholesterol. For cholesterol loading, the lipoprotein-deficient fetal calf serum was supplemented with 10 µg cholesterol (dissolved in EtOH) and 1 µg 25-OH-cholesterol (dissolved in EtOH) per ml serum. Culture media were refreshed every 5 days. Fatty acid synthesis and fatty acid desaturation Cells were seeded in T75 tissue culture flasks at approximately 50% confluency and the medium was refreshed the next day. VLCFA synthesis was measured by incubating the cells for 10 days in the presence of increasing concentrations of D 3-16:0. Culture media and D 3-16:0 were refreshed after 5 days. For the measurement of long chain fatty acid desaturation activity, cells were incubated for 24 hours in the presence of 100 µm D 5-18:0. After incubation, cells were harvested with trypsin, washed twice with phosphate-buffered saline (PBS), once with 0.9% NaCl and taken up in 200 µl deionized water. After sonication protein concentration was determined using BCA (Smith et al 1985). Total cellular fatty acids were analyzed using the electrospray ionization mass spectrometry (ESI-MS) method described previously (Valianpour et al 2003; Kemp et al 2004). 2 Measurement of VLCFA beta-oxidation Cells were cultured in the different media for 7 days and the effect on the peroxisomal betaoxidation activity was measured using [1-14 C]-26:0 as described previously (Wanders et al 1995). Measurements were performed in triplicate for each cell line. As an internal control, mitochondrial beta-oxidation was determined using [1-14 C]-16:0 (Wanders et al 1995). Quantitative RT-PCR analysis of SCD mrna levels The expression levels of SCD mrna were related to the expression levels of GAPDH (Glyceraldehyde-phosphate-dehydrogenase) mrna using the LightCycler system (Roche, Mannheim, Germany). Total RNA was isolated from primary skin fibroblasts growing in log phase with TRIzol reagent (Invitrogen, Carlsbad, USA) and cdna was synthesized using the first-strand cdna synthesis kit (Roche, Mannheim, Germany). Quantitative real-time PCR analysis of SCD in skin fibroblasts was performed using the LightCycler FastStart DNA Master SYBR green I kit (Roche, Mannheim, Germany). The following primers for SCD were used: forward 5 -CACCCAGCTGTCAAAGAGAA-3 and reverse 5 -TCACCCACAGCTCCAAGT-3. For GAPDH the following primers were used: forward 5 -ACCACCATGGAGAAGGCTGC-3 and reverse 5 -CTCAGTGCCCAGGATGC-3. The specificity of amplification was confirmed by agarose gel electrophoresis. Data were analyzed using LightCycler software version 3.5 (Roche) and the LinReg PCR program version 7.5 for analysis of RT-PCR data (Ramakers et al 2003). Results Effect of cholesterol deprivation on endogenous VLCFA Culturing fibroblasts from X-ALD patients and control subjects in lipoprotein-deficient medium resulted in a 0.29 nmol/mg protein (35%) reduction of 26:0 levels (Fig. 1A and B). Interestingly, these conditions caused a 1.03 nmol/mg protein (70%) increase in the 26:1 levels in fibroblasts from X-ALD patients, but not in control fibroblasts (Fig. 1C and D). This 41

44 Chapter 2 2 experiment was repeated with cells cultured in DMEM instead of HAM F-10 to compare our results with the data from Weinhofer et al who did not investigate the effect on 26:1 levels (Weinhofer et al 2002). The data in Fig. 1B and Fig. 2A show a comparable effect on 26:0 levels: deprivation of cholesterol resulted in a 40% reduction in 26:0 levels in X-ALD cells independent of the type of medium used. Under both culture conditions there was a significant increase in 26:1 (Fig. 1D and 2B). Remarkably, both the 26:0 as well as the 26:1 levels were higher when cells were cultured in DMEM. Compared to Ham F10, 26:0 levels were approximately 1.5-fold higher (compare Fig. 1B with Fig. 2A), and 26:1 levels were about 4-fold higher (compare Fig. 1D with 2B). This could be related to differences in composition between DMEM and Ham F10 culture medium. For example there are clear differences in the concentration of D-glucose (25 mmol/l versus 6 mmol/l) and L-glutamine (1 mmol/l versus 4 mmol/l). Effect of cholesterol deprivation on VLCFA synthesis from D3-16:0 and 26:0 beta-oxidation To gain more insight into the mechanism by which cholesterol deprivation influences the levels of saturated and mono-unsaturated VLCFA, we measured the effect of cholesterol deprivation on the synthesis of both D 3-26:0 and D 3-26:1 from D 3-16:0. Furthermore, VLCFA beta-oxidation was measured. To this end, X-ALD fibroblasts were cultured in the presence of D 3-16:0 for 10 days and the amount of D 3 -VLCFA was determined. Cholesterol deprivation did not affect net synthesis of saturated VLCFA from D 3-16:0 (Fig. 3A). However, the synthesis Figure 1: Endogenous levels of 26:0 (A and B) and 26:1 (C and D) in fibroblasts from controls (A and C) and X-ALD patients (B and D). Fibroblasts were cultured for 10 days in HAM F10 medium supplemented either with 10% fetal calf serum (open bars) or with 10% lipoprotein-deficient fetal calf serum (black bars). Fatty acid levels are expressed as nmol/mg protein. Statistical significance was determined with Student s unpaired t-test. Error bars represent the standard deviation: (*) p < 0.05; (**) p < 0.01 and (***) p <

45 In vitro studies of D 3-26:1 from D 3-16:0 was found to be increased 7-fold under low-cholesterol conditions (Fig. 3B). When X-ALD fibroblasts were grown in lipoprotein-deficient medium, the rate of [1-14 C]-26:0 beta-oxidation was not affected (Fig. 4). Furthermore, the beta-oxidation of [1-14 C]-16:0 was not different between X-ALD fibroblasts and controls and between the different culture conditions (data not shown). Effect of cholesterol deprivation on SCD expression and desaturation activity In mammalian cells, the synthesis of mono-unsaturated fatty acids from saturated fatty acids is catalyzed by stearoyl-coa-desaturase (SCD), an enzyme that introduces a cis -double bond in the ω-9 position (Ntambi and Miyazaki, 2003). The main substrates for this enzyme are palmitoyl-coa (16:0-CoA) and stearoyl-coa (18:0-CoA). We investigated whether the increase in mono-unsaturated VLCFA could be explained by an increased expression and activity of SCD. To this end, we performed qpcr analysis using control and X-ALD cell lines cultured on standard HAM F10 medium, lipoprotein-deficient medium, or lipoproteindeficient medium supplemented with free cholesterol and 25-OH-cholesterol. When 2 Figure 2: Endogenous 26:0 (A) and 26:1 levels (B) in fibroblasts from X-ALD patients cultured for 10 days in DMEM supplemented either with 10% fetal calf serum (open bars) or with 10% lipoproteindeficient fetal calf serum (black bars). Fatty acid levels are expressed as nmol/mg protein. Statistical significance was determined with Student s unpaired t-test. Error bars represent the standard deviation: (*) p < 0.05; (**) p < Figure 3: Net synthesis of D 3-26:0 (A) and D 3-26:1 (B) from D 3-16:0 in fibroblasts from X-ALD patients. Fibroblasts were incubated with increasing concentrations D 3-16:0 for 10 days in HAM F10 medium supplemented either with 10% fetal calf serum (open squares) or with 10% lipoprotein-deficient fetal calf serum (grey circles). Fatty acid levels are expressed as nmol/mg protein. Error bars represent the standard deviation. 43

46 Chapter 2 2 Figure 4: The effect of cholesterol-deprivation on VLCFA (26:0) beta-oxidation in fibroblasts from X- ALD patients. Fibroblasts were cultured for 7 days in HAM F10 medium supplemented either with 10% fetal calf serum (open bars) or with 10% lipoprotein-deficient fetal calf serum (black bars). Control cells cultured under standard conditions were included for comparison. Peroxisomal 26:0 beta-oxidation is expressed as pmol/h/mg protein. Error bars represent the standard deviation. n.d. = not determined. Figure 5: The effect of cholesterol-deprivation on the expression of SCD was analyzed by quantitative PCR. Control (A) and X-ALD (B) cells were cultured in HAM F10 supplemented either with 10% fetal calf serum (open bars), with 10% lipoprotein-deficient fetal calf serum (black bars), or with 10% lipoprotein-deficient fetal calf serum with 10 μl cholesterol and 1 μl 25-OH-cholesterol (grey bars). Statistical significance was determined with Student s unpaired t-test. Error bars represent the standard deviation: (*) p < fibroblasts from both controls and X-ALD patients were cultured in lipoprotein-deficient medium, the SCD mrna levels increased 4 to 5-fold (Fig. 5). This increase was abolished by the addition of cholesterol and 25-OH-cholesterol to the lipoprotein-deficient culture medium (Fig. 5). Subsequently, in order to obtain information on the activity of SCD, we incubated X-ALD fibroblasts and controls with labeled D 5-18:0 and the amount of D 5-18:1 present after 24 hours was measured. Figure 6 shows that the synthesis of D 5-18:1 was increased under conditions of cholesterol-deprivation which was reversed by the addition of cholesterol and 25-OH-cholesterol to the lipoprotein-deficient culture medium. 44

47 In vitro studies 2 Figure 6: The effect of cholesterol-deprivation on the activity of SCD was analyzed by measuring the amount of D 5-18:1 formed from D 5-18:0. Control (A) and X-ALD (B) cells were cultured for 24 hours in HAM F10 with 100 μm D 5-18:0 supplemented either with 10% fetal calf serum (open bars), with 10% lipoprotein-deficient fetal calf serum (black bars), or with 10% lipoprotein-deficient fetal calf serum with 10 μl cholesterol and 1 μl 25-OH-cholesterol (grey bars). Fatty acid levels are expressed as nmol/ mg protein. Statistical significance was determined with Student s unpaired t-test. Error bars represent the standard deviation: (*) p < 0.05; (**) p < Discussion The data presented in this paper confirm the earlier observation that X-ALD fibroblasts grown in lipoprotein-deficient culture medium have a reduced level of 26:0 (Fig. 1B) (Weinhofer et al 2002). However, we also found a large increase in the 26:1 level (Fig. 1D). This has not been reported before. VLCFA are mostly derived from endogenous synthesis and not from dietary sources (Moser et al 2001), and it has been shown previously that X-ALD fibroblasts have increased 26:1 synthesis from oleate (18:1) (Kemp et al 2005). Saturated VLCFA are synthesized from palmitate (16:0) by chain elongation (Jakobsson et al 2006), whereas elongation of palmitoleate (16:1) yields mono-unsaturated VLCFA. Mono-unsaturated fatty acids in humans are synthesized by SCD, an enzyme that introduces a cis double bond at the ω-9 position. The preferred substrates for SCD are palmitoyl-coa (16:0-CoA) or stearoyl-coa (18:0-CoA), and not fatty acyl-coas with longer chain length (Ntambi and Miyazaki, 2003). Our results show that cholesterol depletion of the culture medium results in increased 26:1 levels (Fig. 1D) and increased synthesis of D 3-26:1 from D 3-16:0 in X-ALD fibroblasts (Fig. 3B). Interestingly, this increase in 26:1 was accompanied by an increase in the expression of SCD and increased conversion of D 5-18:0 to D 5-18:1. SCD is regulated at the transcriptional level and it is a known target of the SREBP class of transcription factors (Ntambi and Miyazaki, 2003). 25-OH-cholesterol is a potent inhibitor of SREBP activation (Brown and Goldstein, 1997), and addition of cholesterol and 25-OH-cholesterol will abolish SREBP activation and can serve as a control condition. The fact that the SREBP signaling pathway is activated by cholesterol depletion (Brown and Goldstein, 1997), may well explain the increase in SCD mrna levels when X-ALD fibroblasts were cultured in lipoprotein-deficient medium (Fig. 5A). This finding is supported by the increased conversion of D 5-18:0 to D 5-18:1 by SCD (Fig. 6A). The observation that both SCD mrna levels and increased conversion of D 5-18:0 to D 5-18:1 can be reversed by adding free cholesterol and 25-OH-cholesterol to the lipoproteindeficient medium strongly suggests that the observed effects are indeed attributable to 45

48 Chapter 2 cholesterol deprivation and not to other differences between the media. 2 The findings described in this paper may well be relevant for research aimed towards the generation of new therapeutic approaches for X-ALD currently ongoing and also to clinical practice. It has been shown that increased expression of ALDRP could reverse the deficient VLCFA beta-oxidation in X-ALD fibroblasts (Kemp et al 1998). Furthermore, the compound 4-phenylbutyric acid (4-PBA) reduced VLCFA levels in both human X-ALD fibroblasts as well as in the brain and adrenals of Abcd1 knockout mice. Even though the mode of action has not yet been resolved, this may be caused by peroxisome proliferation, increased ALDRP expression (Kemp et al 1998) or increased mitochondrial beta-oxidation (McGuinness et al 2003). In Abcd1 knockout mice in which ALDRP is overexpressed, there is normalization of the biochemical abnormalities and also reversal of the AMN-like phenotype occurring at the age of 18 months (Pujol et al 2004). This data supports the hypothesis that ALDRP overexpression might be of clinical benefit in X-ALD. Unfortunately, 4-PBA mediated pharmacological overexpression of ALDRP proved unsuitable for application in humans, due to the short half-life of 4-PBA (Moser et al 2001). Therefore, other means to upregulate ALDRP in humans as a treatment for X-ALD are being sought. It has been suggested that cholesterol lowering with lovastatin might accomplish this (Weinhofer et al 2002). In 1998 the reports that lovastatin reduced the 26:0/22:0 ratio in cultured X-ALD fibroblasts (Singh et al 1998a) and also reduced plasma VLCFA (reported as the sum of 22:0 + 24:0 + 26:0) in X-ALD patients (Singh et al 1998b) offered new hope for a widely applicable and effective treatment for X-ALD. Unfortunately, a follow-up paper from the same group showed less striking effects on 26:0 levels (Pai et al 2000) and a study with simvastatin failed to show any effect on 26:0 (Verrips et al 2000). Later work by Weinhofer et al was aimed to identify the mechanism by which lovastatin reduced the 26:0/22:0 ratio in X-ALD skin fibroblasts (Weinhofer et al 2002). To study this, X-ALD skin fibroblasts were cultured in lipoprotein-deficient medium containing low levels of cholesterol. When grown in this medium, X-ALD fibroblasts showed an increase in ABCD2 expression, as well as a decrease in 26:0 levels (Weinhofer et al 2002). The observed upregulation of ALDRP was taken as a plausible explanation for the observed effect of lipoprotein-deficient culture medium and lovastatin on 26:0 levels in X-ALD fibroblasts. Our findings, however, shed new light on the feasibility of lowering cholesterol as a therapeutic approach. The decrease in 26:0 induced by culturing X-ALD skin fibroblasts in lipoprotein-deficient medium is offset by a significant increase in 26:1. On the whole, there is no decrease in the total VLCFA pool: the 0.3 nmol/mg protein decrease in 26:0 (Fig. 1A) is smaller than the 1 nmol/mg protein increase in 26:1 (Fig. 2A). The observation that 26:0 beta-oxidation is not increased in lipoprotein-deficient medium is in line with this finding (Fig. 4). The role of VLCFA accumulation in the pathogenesis of X-ALD is still largely unknown. Since VLCFA accumulation (predominately 26:0) is the biochemical hallmark of X-ALD it seems plausible that this accumulation is somehow related to the development of symptoms. This is supported by experiments by Whitcomb and colleagues showing a direct toxic effect of 26:0 on adrenocortical cells (Whitcomb et al 1988). Adrenocortical cells cultured in the presence of increasing concentrations of 26:0 showed a decreased response to ACTH stimulation. Also, 26:0 has disruptive effects on cell membrane structure and function (Ho 46

49 In vitro studies et al 1995). Whether 26:1 has different effects on cell functioning and membrane properties than 26:0 and whether it is less toxic remains to be resolved, and should be investigated. The data presented here indicate that it is unclear if 26:0 reduction through cholesterol lowering is a feasible therapeutic approach. Caution should be taken in prescribing lovastatin to X-ALD patients, since data on for example the possible accumulation of mono-unsaturated VLCFA or VLCFA accumulation in tissues is lacking. Currently, we are conducting a clinical trial to evaluate the effect of lovastatin versus placebo on VLCFA in plasma, lymphocytes and erythrocytes of X-ALD patients. 2 Acknowledgements The authors thank R. Sanders, I.M.E. Dijkstra, F. Stet and H. van Lenthe for technical assistance. This work was supported by grants from the European Leukodystrophy Association: ELA I5 and ELA I4. References Berger J, Albet S, Bentejac M, Netik A, Holzinger A, Roscher AA, Bugaut M, Forss-Petter S The four murine peroxisomal ABC-transporter genes differ in constitutive, inducible and developmental expression. Eur J Biochem 265: Brown MS, Goldstein JL The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89: Cartier N, Guidoux S, Rocchiccioli F, Aubourg P Simvastatin does not normalize very long chain fatty acids in adrenoleukodystrophy mice. FEBS Lett 478: Ho JK, Moser H, Kishimoto Y, Hamilton JA Interactions of a very long chain fatty acid with model membranes and serum albumin. Implications for the pathogenesis of adrenoleukodystrophy. J Clin Invest 96: Jakobsson A, Westerberg R, Jacobsson A Fatty acid elongases in mammals: their regulation and roles in metabolism. Prog Lipid Res 45: Kamijo K, Taketani S, Yokota S, Osumi T, Hashimoto T The 70-kDa peroxisomal membrane protein is a member of the Mdr (P-glycoprotein)-related ATP-binding protein superfamily. J Biol Chem 265: Kemp S, Wei HM, Lu JF, Braiterman LT, McGuinness MC, Moser AB, Watkins PA, Smith KD Gene redundancy and pharmacological gene therapy: implications for X-linked adrenoleukodystrophy. Nat Med 4: Kemp S, Valianpour F, Mooyer PA, Kulik W, Wanders RJ Method for measurement of peroxisomal very-longchain fatty acid beta-oxidation in human skin fibroblasts using stable-isotope-labeled tetracosanoic acid. Clin Chem 50: Kemp S, Valianpour F, Denis S, Ofman R, Sanders RJ, Mooyer P, Barth PG, Wanders RJ Elongation of very longchain fatty acids is enhanced in X-linked adrenoleukodystrophy. Mol Genet Metab 84: Lombard-Platet G, Savary S, Sarde CO, Mandel JL, Chimini G A close relative of the adrenoleukodystrophy (ALD) gene codes for a peroxisomal protein with a specific expression pattern. Proc Natl Acad Sci U S A 93:

50 Chapter 2 McGuinness MC, Lu JF, Zhang HP, Dong GX, Heinzer AK, Watkins PA, Powers J, Smith KD Role of ALDP (ABCD1) and mitochondria in X-linked adrenoleukodystrophy. Mol Cell Biol 23: Moser HW, Moser AB, Frayer KK, Chen W, Schulman JD, O Neill BP, Kishimoto Y Adrenoleukodystrophy: increased plasma content of saturated very long chain fatty acids. Neurology 31: Moser HW, Smith KD, Watkins PA, Powers J, Moser AB X-linked adrenoleukodystrophy. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The Metabolic and Molecular Bases of Inherited Disease. New York: Mc Graw Hill. p Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL, Aubourg P Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: Mosser J, Lutz Y, Stoeckel ME, Sarde CO, Kretz C, Douar AM, Lopez J, Aubourg P, Mandel JL The gene responsible for adrenoleukodystrophy encodes a peroxisomal membrane protein. Hum Mol Genet 3: Netik A, Forss-Petter S, Holzinger A, Molzer B, Unterrainer G, Berger J Adrenoleukodystrophy-related protein can compensate functionally for adrenoleukodystrophy protein deficiency (X-ALD): implications for therapy. Hum Mol Genet 8: Ntambi JM, Miyazaki M Recent insights into stearoyl-coa desaturase-1. Curr Opin Lipidol 14: Pai GS, Khan M, Barbosa E, Key LL, Craver JR, Cure JK, Betros R, Singh I Lovastatin therapy for X-linked adrenoleukodystrophy: clinical and biochemical observations on 12 patients. Mol Genet Metab 69: Pujol A, Ferrer I, Camps C, Metzger E, Hindelang C, Callizot N, Ruiz M, Pampols T, Giros M, Mandel JL Functional overlap between ABCD1 (ALD) and ABCD2 (ALDR) transporters: a therapeutic target for X-adrenoleukodystrophy. Hum Mol Genet 13: Ramakers C, Ruijter JM, Deprez RH, Moorman AF Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci Lett 339: Singh I, Moser HW, Moser AB, Kishimoto Y Adrenoleukodystrophy: impaired oxidation of long chain fatty acids in cultured skin fibroblasts an adrenal cortex. Biochem Biophys Res Commun 102: Singh I, Pahan K, Khan M. 1998a. Lovastatin and sodium phenylacetate normalize the levels of very long chain fatty acids in skin fibroblasts of X- adrenoleukodystrophy. FEBS Lett 426: Singh I, Khan M, Key L, Pai S. 1998b. Lovastatin for X-linked adrenoleukodystrophy. N Engl J Med 339: Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC Measurement of protein using bicinchoninic acid. Anal Biochem 150: Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S Analysis of very long-chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 79: Verrips A, Willemsen MA, Rubio-Gozalbo E, De JJ, Smeitink JA Simvastatin and plasma very-long-chain fatty acids in X-linked adrenoleukodystrophy. Ann Neurol 47: Wanders RJ, Denis S, Ruiter JP, Schutgens RB, van Roermund CW, Jacobs BS Measurement of peroxisomal fatty acid beta-oxidation in cultured human skin fibroblasts. J Inherit Metab Dis 18 Suppl 1: Weinhofer I, Forss-Petter S, Zigman M, Berger J Cholesterol regulates ABCD2 expression: implications for the therapy of X-linked adrenoleukodystrophy. Hum Mol Genet 11: Weinhofer I, Forss-Petter S, Kunze M, Zigman M, Berger J X-linked adrenoleukodystrophy mice demonstrate abnormalities in cholesterol metabolism. FEBS Lett 579:

51 In vitro studies Whitcomb RW, Linehan WM, Knazek RA Effects of long-chain, saturated fatty acids on membrane microviscosity and adrenocorticotropin responsiveness of human adrenocortical cells in vitro. J Clin Invest 81: Yamada T, Shinnoh N, Taniwaki T, Ohyagi Y, Asahara H, Horiuchi, Kira J Lovastatin does not correct the accumulation of very long-chain fatty acids in tissues of adrenoleukodystrophy protein-deficient mice. J Inherit Metab Dis 23:

52 Chapter

53 In vitro studies Chapter 3 3 Bezafibrate lowers very long-chain fatty acids in X-linked adrenoleukodystrophy fibroblasts by inhibiting fatty acid elongation Marc Engelen 1,3, Martin J.A. Schackmann 2, Rob Ofman 2, Robert-Jan Sanders 2, Inge M.E. Dijkstra 2, Sander M. Houten 2, Stéphane Fourcade 4, Aurora Pujol 4,5, Bwee Tien Poll The 1,3, Ronald J.A. Wanders 2 and Stephan Kemp 2,3 1 Department of Neurology; 2 Department of Clinical Chemistry, Laboratory Genetic Metabolic Diseases; 3 Department of Pediatric Neurology/Emma Children s Hospital, Academic Medical Center, University of Amsterdam, The Netherlands. 4 Neurometabolic Diseases Laboratory, The Bellvitge Institute of Biomedical Research (IDIBELL), Center for Biomedical Research on Rare Diseases (CIBERER), Barcelona, Spain. 5 ICREA (Institució Catalana de Recerca i Estudis Avançats), Barcelona, Spain. Journal of Inherited Metabolic Diseases (2012) in press 51

54 Chapter 3 Abstract 3 X-linked adrenoleukodystrophy (X-ALD) is caused by mutations in the ABCD1 gene encoding ALDP, an ATP-binding-cassette (ABC) transporter located in the peroxisomal membrane. ALDP deficiency results in impaired peroxisomal beta-oxidation and the subsequent accumulation of very long-chain fatty acids (VLCFA; > C22:0) in plasma and tissues. VLCFA are primarily derived from endogenous synthesis by ELOVL1. Therefore inhibiting this enzyme might constitute a feasible therapeutic approach. In this paper we demonstrate that bezafibrate, a PPAR pan agonist used for the treatment of patients with hyperlipidaemia reduces VLCFA levels in X-ALD fibroblasts. Surprisingly, the VLCFA-lowering effect was independent of PPAR activation and not caused by the increase in either mitochondrial or peroxisomal fatty acid beta-oxidation capacity. In fact, our results show that bezafibrate reduces VLCFA synthesis by decreasing the synthesis of C26:0 through a direct inhibition of fatty acid elongation activity. Taken together, our data indicate bezafibrate as a potential pharmacotherapeutic treatment for X-ALD. A clinical trial is currently ongoing to evaluate the effect in patients with X-ALD. Take-home message: Pharmacological inhibition of ELOVL1 with bezafibrate lowers very long-chain fatty acids in X-linked adrenoleukodystrophy fibroblasts and might constitute a new approach to treatment. 52

55 In vitro studies Introduction X-linked adrenoleukodystrophy (X-ALD: OMIM ) is an inherited metabolic disorder characterized by impaired peroxisomal β-oxidation of very long-chain fatty acids (VLCFA; C22) and accumulation of VLCFA (mainly C26:0) in plasma and tissues of patients (Moser et al 2001). It is caused by mutations in the ABCD1 gene ( encoding a peroxisomal transmembrane protein named ALD protein (ALDP: OMIM ) (Mosser et al 1993). Clinically, X-ALD is characterized by a striking and unpredictable variation in phenotypic expression, ranging from the rapidly progressive childhood cerebral form (CCALD) to the more slowly progressive adult form adrenomyeloneuropathy (AMN) and variants without neurological involvement ( Addison-only phenotype) (Moser et al 2001). Experimental pharmacotherapy in X-ALD was aimed at normalizing VLCFA beta-oxidation and VLCFA levels. Over the years several compounds have been investigated, such as Lorenzo s oil (Aubourg et al 1993; van Geel et al 1999), 4-phenylbutyrate (Kemp et al 1998), and lovastatin (Singh et al 1998; Engelen et al 2010). These treatments were shown to be either unpractical or ineffective in clinical trials and therefore other drugs are needed. Fenofibrate (a PPAR-α agonist) was shown to induce expression of ALDR (Abcd2) in the liver of Abcd1 -/- mice (Netik et al 1999). ALDRP is a functional homolog of ALDP (Kemp et al 1998). However, fenofibrate has no effect on Abcd2 expression in the brain of Abcd1 -/- mice, possibly because it is a substrate for the Mdr1 transporter at the blood brain barrier and therefore does not penetrate into the brain very effectively (Berger et al 1999). For this reason, we investigated the effect of several other drugs known to activate PPAR on VLCFA metabolism in cultured skin fibroblasts from patients with X-ALD. The results described in this paper show that bezafibrate (BF), but not fenofibrate, clofibrate or other PPAR agonists, could reduce VLCFA in cultured fibroblasts from patients with X-ALD. The VLCFA which accumulate in X-ALD, are partly absorbed from the diet (Kishimoto et al 1980), but mostly result from endogenous synthesis through elongation of long-chain fatty acids (Tsuji et al 1981). Recently, we identified ELOVL1 as the key enzyme responsible for the synthesis of VLCFA and demonstrated that knock-down of ELOVL1 resulted in lower VLCFA synthesis and reduced levels of VLCFA in cultured X-ALD fibroblasts (Ofman et al 2010). Hence, inhibiting fatty acid elongation (for example by inhibition of ELOVL1) by pharmacological means could be a potential treatment for X-ALD. Here, we show that BF lowers VLCFA in X-ALD fibroblasts by direct inhibition of fatty acid elongation. 3 Materials and Methods Chemicals Deuterium-labeled palmitate-16,16,16-d 3 acid (D 3 C16:0) was purchased from CDN isotopes (Pointe-Claire, Canada). A 12.5 mm stock solution in dimethyl sulfoxide (DMSO) was prepared. BF, fenofibrate and clofibrate, WY14643, GW501516, and rosiglitazone were purchased from Sigma-Aldrich (St. Louis, MO, USA). Stock solutions in DMSO were made of 400 mm (BF and clofibrate), 50 mm (fenofibrate), 10 mm (WY14643 and rosiglitazone), 500 µm (GW501516). MK-886 was purchased from Cayman Chemical (Ann Arbor, MI, USA) and a stock solution of 50 mm in DMSO was used. Prior to usage the stock solutions were vortex mixed and diluted in HAMF10 tissue culture medium to the final concentration. All chemicals used were of analytical grade. 53

56 Chapter 3 Cell lines and cell culture Primary human skin fibroblasts were obtained from X-ALD patients through the Neurology Outpatient Clinic of the Academic Medical Center. From each patient written informed consent was obtained. X-ALD diagnosis was confirmed by VLCFA and ABCD1 mutation analysis. Control fibroblasts were from male anonymous volunteers. Cells from patients with a peroxisomal biogenesis disorder were obtained from the laboratory cell bank. Cells were grown in HAMF10 supplemented with 10% fetal calf serum, 2.5 mm HEPES, 100 U/ml penicillin, 100 U/ml streptomycin and 2 mm glutamine. Cells were used between passage numbers 6 and 20. Culture media were refreshed every 5 days. 3 Fatty acid synthesis Synthesis of D 3 -VLCFA in intact cells was measured using D 3 -C16:0. Assays were performed in triplicate. Cells were seeded at 40% confluency in T75 flasks. The next day, medium was replaced by fresh medium supplemented with D 3 -C16:0 (dissolved in DMSO) at a final concentration of 50 µm. After 72 hr, cells were harvested and VLCFA analyzed as described (Valianpour et al 2003). Measurement of fatty acid beta-oxidation Mitochondrial beta-oxidation activity of intact fibroblasts was measured by quantifying the production of 3H 2 O from [9,10-3 H(N)] oleic acid as described previously (Moon and Rhead 1987). Peroxisomal beta-oxidation activity of intact fibroblasts was measured using [1-14 C]- 26:0 as described previously (Wanders et al 1995). All measurements were performed in triplicate for each cell line. Immunofluorescence and counting of peroxisomes Fibroblasts where grown on glass microscopy slides in 6-well plates with or without 400 µm BF. Cells were fixed with paraformaldehyde and permeabilized with Triton X-100. Peroxisomes were visualized by catalase immunofluorescence microscopy as described previously (Kemp et al 1996). To count peroxisomes, we made images of immunofluorescence-stained cells after focusing on the cell nucleus, and determined the peroxisome number per cell with the aid of a colony counter. For each cell line and condition, 10 cells were counted at random. Quantitative RT-PCR analysis ELOVL1, ELOVL4, ELOVL6, ACOX1, ABCD3 and CPT1a mrna levels in control and X-ALD fibroblasts growing in log phase were determined as described (Engelen et al 2008), with primer sets presented in Table S1 (Supporting Information Table S1, online only). Purification of mouse liver microsomes Microsomes were isolated from livers from wild type and transgenic ELOVL1 over-expressing mice (Kemp et al, manuscript in preparation) by differential centrifugation as described by Baudhuin et al (Baudhuin et al 1964), with minor modifications. Livers were washed with ice-cold homogenization buffer containing 250 mm sucrose, 2 mm EDTA, 2 mm DTT and 5 mm MOPS (ph 7.4), minced and homogenized using a potter tissue grinder with Teflon pestle with 5 strokes at 500 rpm. A post-nuclear supernatant was produced by centrifugation at 600 g for 10 min. The supernatant was centrifuged at 22,500 g for 10 min and the pellet was discarded. Next, the supernatant was centrifuged for 1 h at 100,000 g to obtain 54

57 In vitro studies a microsomal fraction. The pellet was resuspended in homogenization buffer containing 10 mg/ml methyl-beta-cyclodextrin and sonicated for 4 times 5 seconds at 8 W with a 1 minute interval. The microsomal membranes were collected by centrifugation at 100,000 g for 1 h. Finally, the microsomes were resuspended in homogenization buffer and stored at -80 C until further use. All steps were carried out at 4 C. Protein concentrations were determined using BCA as described (Smith et al 1985). Fatty acid elongation assay The fatty acid elongation assay was carried out using a method adapted from Nagi et al (Nagi et al 1989). The reaction mixture contained 50 mm potassium phosphate buffer (ph 6.5), 10 mg/ml α-cyclodextrin, 1 mm NADPH, 5 µm rotenone, 60 µm [2-14 C] malonyl-coa (6.5 dpm/pmol) (American Radiolabeled Chemicals, St Louis, MO) and 20 µm C16:0-CoA or 20 µm C22:0-CoA (Avanti Polar Lipids, Alabaster, AL), in a total volume of 200 μl. The reaction was started by adding 100 µg protein of the microsomal fraction and allowed to proceed for 30 min at 37 C. NADPH dependency was tested by performing the reaction without NADPH. Reactions were carried out with or without BF or BF-CoA ( µm). The reaction was stopped by adding 200 μl 5 M KOH in 10% methanol, saponified at 65 C for 1 h and acidified by adding 200 μl 5 N HCl and 200 μl 96% ethanol. Fatty acids were extracted 3 times with 1 ml hexane and the hexane phases were collected in a scintillation vial to which 10 ml scintillation cocktail (Ultima-Gold, Perkin Elmer) was added and radioactivity counted. 3 Synthesis of bezafibroyl-coa (BF-CoA) BF-CoA was synthesized by a method adapted from Rasmussen et al (Rasmussen et al 1990). Dichloromethane (DCM) and tetrahydrofuran (THF) (Merck) were dried with molecular sieve deperox (Fluka). Triethylamine and ethylchloroformate (Merck) were diluted to 1 M with dry DCM. The reaction contained 36 μmol BF dissolved in 1.4 ml DCM/THF (5:2) and 40 μl 1 M triethylamine. The reaction mixture was incubated at room temperature for 10 min under constant stirring and under an atmosphere of nitrogen. After 10 min, 40 μl 1 M ethylchloroformate was added and the incubation was continued for 45 min. After the incubation, the mixture was dried under nitrogen and dissolved in 0.5 ml tert-butanol. Next, 40 μmol CoA trilithium salt (Sigma-Aldrich) dissolved in 0.5 ml 0.4 M potassium bicarbonate was added and the sample was incubated for 30 min at room temperature. The reaction was stopped by adding 100 μl 0.1 N HCl. BF-CoA was purified using a C18 solid phase extraction column (JTBaker). The column was eluted by a gradient of acetonitrile and 40 mm ammonium acetate, starting with 10% acetonitrile and 90% 40 mm ammonium acetate increasing to 50% acetonitrile and 50% 40 mm ammonium acetate. Acetonitrile was evaporated and purity was checked by HPLC. BF-CoA was quantified using 5,5 -dithiobis-(2-nitrobenzoic) acid (DTNB) (Sigma-Aldrich). The method used was an adaptation of that of Ellman (Ellman 1959). BF-CoA was diluted in 20 μm MES buffer ph 6.0, an equal volume of 2 M NaOH was added and the sample was incubated for 30 min at 50 C. The reaction was neutralized with 2 M HCl. Absorbance at 412 nm was measured and the concentration was determined using a calibration curve of CoA trilithium salt (Sigma-Aldrich). 55

58 Chapter 3 Results 3 Effect of BF on endogenous VLCFA levels and de novo C26:0 synthesis in fibroblasts from patients with X-ALD We tested the effect of several drugs from the fibrate class, but found that only BF reduces C26:0 levels, both fenofibrate and clofibrate being ineffective (Fig. 1A). BF is a PPAR panagonist activating all three PPARs. To determine whether the effect of BF on C26:0 levels is mediated by activation of either PPARα, PPARβ/δ or PPARγ, X-ALD fibroblasts were incubated with the PPARα ligand WY14643, PPARβ/δ ligand GW501516, or the PPARγ ligand rosiglitazone either alone or in all possible combinations. Only treatment with BF reduced C26:0 levels by about 30% after 7 days (Fig. 1B). To examine if longer incubations with BF would result in a further decrease of C26:0 levels, we cultured X-ALD fibroblasts for up to 21 days. This resulted in a small additional decrease of C26:0 of 10% after 14 days and 15% after 21 days, respectively (Fig. 1C). No signs of cyto- Figure 1: Only BF, and not other drugs from the fibrate class, reduces C26:0 in X-ALD fibroblasts. (A) C26:0 levels in 3 X-ALD cell lines cultured for 7 days without drugs (black bar), in the presence of BF (grey bar), fenofibrate (white bar), or clofibrate (hatched bar). (B) C26:0 levels in 3 X-ALD cell lines cultured for 7 days without drugs (black bar), in the presence of BF (grey bar), agonists of PPAR-α (WY 14643), PPAR-β/δ (GW ) and PPAR-γ (rosiglitazon), or with different combinations of PPAR agonists (white bars). (C) C26:0 levels in 8 control (white bar) and 8 X-ALD cell lines cultured without (black bar) or with 400 µm BF (grey bars) for up to 3 weeks. Fatty acid levels are in nmol/mg protein. Data are mean ± SD. * = p < 0.05, ** = P < 0.01, *** = P < by ANOVA followed by Dunnett s multiple comparison test compared with untreated X-ALD cells. 56

59 In vitro studies toxicity or impaired growth as determined by the MTS cell proliferation assay (CellTiter 96 Aqueous One Solution Cell Proliferation Assay) were observed (data not shown). Previously, we validated the use of stable-isotope labeled fatty acids to study VLCFA de novo synthesis in whole cells and demonstrated that the synthesis of D 3 -C26:0 from D 3 -C16:0 is elevated in X-ALD fibroblasts (Ofman et al 2010). Earlier work has shown that BF is a potent inhibitor of long-chain fatty acid elongation, while clofibrate is not (Sanchez et al 1993; Vazquez et al 1995). Therefore, we tested the effect of BF, clofibrate and fenofibrate on C26:0 de novo synthesis. Of all fibrates tested, only BF affected the de novo D 3 -C26:0 synthesis (Fig 2A). BF reduced D 3 -C26:0 synthesis in a concentration-dependent manner. At 400 µm BF, the synthesis of D 3 -C26:0 was reduced by 75% compared with untreated X-ALD fibroblasts (Fig. 2B). At this concentration, there was no difference in the amount of newly synthesized D 3 -C26:0 between X-ALD and control cells. We performed further experiments with BF, and not the other fibrates or PPAR agonists, to identify by which mechanism BF reduces C26:0 levels in fibroblasts from patients with X-ALD. Figure 2: BF, but not other fibrates, inhibits D 3 -C26:0 synthesis. (A) D 3 -C26:0 synthesis from D 3 -C16:0 in 5 X- ALD cell lines cultured for 3 days without drugs (black bar) and in the presence of BF (grey bar), fenofibrate (FF, white bar), or clofibrate (CF, hatched bar). (B) D 3 - C26:0 analysis from D 3 -C16:0 in 4 control (white bar) and 4 X-ALD cell lines cultured for 3 days without (black bar) or with increasing concentrations of BF (grey bars). Fatty acid levels are in nmol/mg protein. Data are mean ± SD. ** = P < 0.01, *** = P < by ANOVA followed by Dunnett s multiple comparison test compared with untreated X-ALD cells. 3 The effect of BF on C26:0 levels is independent of induction of mitochondrial or peroxisomal beta-oxidation BF and PPAR ligands in general are known to induce mitochondrial and peroxisomal betaoxidation (Cabrero et al 2001; Islinger et al 2007; Bonnefont et al 2009; Pyper et al 2010). The effect of BF treatment on the rate of mitochondrial and peroxisomal fatty acid betaoxidation was determined in X-ALD fibroblasts. Exposure of X-ALD fibroblasts to BF caused a 50% increase in C16:0 beta-oxidation (Fig 3A), and a 35% increase in C26:0 beta-oxidation (Fig. 3B). Since it is known that fibrates induce peroxisome proliferation in rodents (Islinger et al 2007), we determined the amount of peroxisomes in X-ALD fibroblasts incubated for 10 days with BF. We did not find any evidence for an effect of BF treatment on peroxisome abundance (Fig. 3C). 57

60 Chapter 3 3 Figure 3: Effect of BF on mitochondrial and peroxisomal beta-oxidation and peroxisome number. Measurement of (A) C16:0 beta-oxidation and (B) C26:0 beta-oxidation activity in 5 control cell lines (white bar) and 5 X-ALD cell lines cultured without (black bar) or with 400 µm BF (grey bar) for 7 days. Activities are in pmol/mg/hour. Data are mean ± SD. * = P < 0.05 and *** = P < by ANOVA followed by Dunnett s multiple comparison test compared with untreated X-ALD cells. (C) Peroxisome number in 10 X-ALD cell lines cultured without (white bar) or with 400 µm BF (grey bar) for 3 days. Mean and quartiles are indicated, the error bars represent the range. Figure 4: C26:0 reduction by BF is not mediated by increased peroxisomal beta-oxidation. Analysis of de novo D 3 -C26:0 synthesis in 3 X-ALD (black bar) cell lines and 3 cell lines from patients with a peroxisomal biogenesis disorder (PBD, white bar). Cells were incubated with 50 µm D 3 -C16:0 for 3 days without or with 400 µm BF (crosshatched bars). Fatty acid levels are in nmol/mg protein. Data are mean ± SD. *** = p < by student s unpaired t-test. BF reduced C26:0 levels in cultured X-ALD fibroblasts, while other drugs of the fibrate class or other synthetic PPAR agonists could not. This strongly suggested that the effect of BF on VLCFA levels is not dependent solely on induction of mitochondrial or peroxisomal betaoxidation, since the other compounds should have been effective as well then (Blaauboer et al 1990; Kemp et al 2011). BF treatment resulted in a 35% increase in the peroxisomal C26:0 beta-oxidation capacity (Fig 3B). To test if the induction of peroxisomal beta-oxidation by BF caused the reduction in C26:0 levels, we incubated fibroblasts from patients with a peroxisomal biogenesis disorder (PEX1, PEX6 and PEX26) with BF and measured the effect on D 3 -C26:0 de novo synthesis. The synthesis of D 3 -C26:0 from D 3 -C16:0 was reduced roughly 50% in peroxisome-deficient fibroblasts incubated with BF (Fig 4). This clearly indicated that the effect of BF on D 3 -C26:0 synthesis and C26:0 levels is only partially mediated by an induction of the peroxisomal beta-oxidation capacity. 58

61 In vitro studies Figure 5: C26:0 reduction by BF is not mediated by increased mitochondrial beta-oxidation. (A) C18:1 beta-oxidation in 3 X-ALD cell lines cultured without (black bar), or for 48 hours with 400 µm BF (grey bar), or with BF and increasing concentrations of MK-886 (white bars). Activities are in pmol/mg/ hour. Data are mean ± SD. (B) De novo D 3 -C26:0 synthesis in 4 X-ALD cell lines incubated with 50 µm D 3 - C16:0 without (black bar) or with 400 µm BF (grey bar), 50 µm MK-886 (white bar), or 400 µm BF and 50 µm MK-886 (cross-hatched bar). Fatty acid levels are in nmol/mg protein. Data are mean ± SD. ** = p < 0.01; *** = p < by ANOVA followed by Dunnett s multiple comparison test compared with untreated X-ALD cells. 3 To investigate whether the induction of mitochondrial beta-oxidation by BF is responsible for the reduction in C26:0, we used the PPARα inhibitor MK-886 (Kehrer et al 2001). The induction of mitochondrial beta-oxidation by BF in X-ALD fibroblasts could be reversed with 50 µm MK-886, suggesting that PPARα activation is completely blocked at this concentration (Fig 5A). Next, we measured D 3 -C26:0 de novo synthesis in X-ALD fibroblasts incubated with BF in the presence of MK-886. Addition of MK-886 did not affect the reduction of D 3 -C26:0 synthesis (Fig 5B). Combined, these data are highly suggestive that the reduction of C26:0 levels by BF is PPAR independent and can not be explained by the increase in mitochondrial beta-oxidation capacity and only partially by the increase in peroxisomal beta-oxidation capacity. This suggests that BF inhibits the formation of D 3 -C26:0. BF directly inhibits fatty acid elongation The amount of D 3 -C26:0 present is the net result of the elongation of D 3 -C16:0 to D 3 -C26:0 by ELOVL6 and ELOVL1, respectively (Ofman et al 2010) and the degradation of D 3 -C26:0 by peroxisomal beta-oxidation. The inhibitory effect of BF on the formation of D 3 -C26:0 could either be indirect by affecting gene expression or direct by inhibition of key enzymes involved in VLCFA de novo synthesis (Ofman et al 2010). To investigate the effect of BF on gene expression levels of several key enzymes involved in VLCFA synthesis, peroxisomal betaoxidation and mitochondrial beta-oxidation we performed quantitative RT-PCR. As shown in Fig 6, there is only a small reduction in the mrna levels of ELOVL6, which is not statistically significant. There was no increase in the expression of ACOX1. This is in line with previously published data (Blaauboer et al 1990) and our own data (Fig 3C) showing that there is no peroxisome proliferation, at least in cultured human cells, treated with fibrates, in contrast to rodents (Blaauboer et al 1990). BF did not induce the expression of ABCD3. Induction of this gene was detected in mice treated with fenofibrate, and considered to be the mecha- 59

62 Chapter 3 3 nism by which fibrates might be useful in correcting the metabolic defect in X-ALD (Berger et al 1999; Netik et al 1999). Our data show that induction of ABCD3 did not occur in human X-ALD fibroblasts upon exposure to BF. In line with previous data (Djouadi et al 2005), BF treatment resulted in increased expression of CPT1a. The previous experiments suggested that the effect of BF on C26:0 levels and C26:0 synthesis could be mediated by a direct inhibiting effect on fatty acid elongation. VLCFA are synthesized by the concerted action of ELOVL6 and ELOVL1 (Ofman et al 2010). ELOVL6 elongates C16:0 to C22:0 and ELOVL1 elongates C22:0 to C26:0. We measured the effect of free BF and BF esterified to coenzyme CoA (BF-CoA) on C16:0-CoA and C22:0-CoA elongation. In the elongation assay BF had no effect. However, BF-CoA inhibited the chain elongation activity of both C16:0 and C22:0 in a concentration dependent manner (Fig. 7). Both fenofibrate and clofibrate did not inhibit C22:0-CoA elongation (Fig. 7B). These data demonstrate that BF- CoA lowers C26:0 levels by direct inhibition of fatty acid chain elongation. Discussion Synthetic PPAR alpha ligands, like fibrates, are potentially interesting compounds to investigate as therapeutic agents in X-ALD because they are known to activate mitochondrial and peroxisomal fatty acid beta-oxidation (Vazquez et al 2001). They might therefore reduce VLCFA accumulation by increasing VLCFA degradation. Indeed, Brown and colleagues demonstrated that treatment of two CCALD patients with clofibrate resulted in a reduction in VLCFA (Brown et al 1982). However, this reduction was not sustained. More recent experiments showed that fenofibrate induced expression of both ALDRP (ABCD2) and PMP70 (ABCD3) in the liver of Abcd1-deficient mice, but not in brain (Berger et al 1999). In Mdr1 -/- knockout mice induction of ALDRP and PMP70 in brain did occur, suggesting that fenofibrate is indeed cleared from the brain by Mdr1 (Berger et al 1999). These studies, however, did not report the effect on VLCFA levels in tissues. In this paper, we studied the effect of several classical fibrates and other synthetic PPAR ligands in a cell model for X-ALD, and show that BF but not the other fibrates reduced endogenous C26:0 levels. This C26:0 reducing effect could not be mimicked by other PPAR-ligands which means that the effect is PPAR independent. BF has been demonstrated to induce mitochondrial beta-oxidation (Djouadi et al 2005; Bonnefont et al 2009). Blocking the induction of mitochondrial beta-oxidation Figure 6: Effect of BF on gene expression of genes involved in fatty acid metabolism. Expression levels of genes involved in fatty acid metabolism in 3 X-ALD cell lines cultured without (black bars) or with 400 µm BF (grey bars) for 48 hours were analyzed by quantitative PCR. Data are mean ± SD. * = p < 0.05 by student s unpaired t-test. 60

63 In vitro studies 3 Figure 7: BF inhibits fatty acid elongation. (A) Fatty acid elongation activity of C16:0-CoA (a substrate of ELOVL6) in the presence of an increasing concentration of BF (o) and BF-CoA ( ). (B) Fatty acid elongation activity of C22:0-CoA (a substrate of ELOVL1) in the presence of an increasing concentration of BF (o) and BF-CoA ( ) or with an increasing concentration of fenofibrate (FF, ) or clofibrate (CF, ). Fatty acid elongation activity was measured using 20 µm C16:0-CoA or C22:0-CoA as substrate with concentrations of the inhibitor up to 400 µm. Error bars represent the standard deviation. with MK-886 did not prevent the reduction of D 3 -C26:0 de novo synthesis in X-ALD fibroblast by BF. We also showed that the peroxisomal C26:0 beta-oxidation capacity in X-ALD skin fibroblasts increased with 35% upon treatment with BF. However this does not seem to be the only mechanism of reduction of VLCFA in fibroblasts incubated with BF, because in cells from patients with a peroxisome biogenesis disorder in which peroxisomal beta-oxidation is completely deficient, BF lowered D 3 -C26:0 levels as well. This strongly suggested that BF might reduce C26:0 levels primarily by inhibiting C26:0 synthesis. To test this we measured de novo synthesis of D 3 -C26:0 from D 3 -C16:0 in X-ALD fibroblasts. BF indeed decreased de novo synthesis of D 3 -C26:0 in a concentration-dependent manner. At a concentration of 400 µm BF, D 3 -C26:0 levels in X-ALD fibroblasts were at the level of control fibroblasts. The rate limiting enzymes involved in synthesis of C26:0 from C16:0 are ELOVL6 (elongation of C16:0 to C22:0) and ELOVL1 (elongation of C22:0 to C26:0) (Kemp and Wanders 2010; Ofman et al 2010). By qpcr we showed that expression levels of these enzymes are not affected in fibroblasts incubated with BF suggesting a direct inhibition of VLCFA synthesis. Previously using rat liver microsomes other investigators showed that BF inhibits palmitoyl-coa (C16:0- CoA) elongation in an in vitro assay (Sanchez et al 1993; Vazquez et al 1995). We used purified microsomes from wild type and ELOVL1 over-expressing mice to test the effect of BF on elongation of both long-chain fatty acids (LCFA) and VLCFA. Our results demonstrate that BF-CoA, but not free BF, is a potent inhibitor of both LCFA and VLCFA elongation. It should be noted, however, that these results do not allow us to demonstrate at which level BF inhibits VLCFA synthesis. Fatty acid elongation requires four sequential reaction steps: (i) condensation between the fatty acyl-coa and malonyl-coa to form 3-ketoacyl-CoA; (ii) reduction using NADPH to form 3-hydroxyacyl-CoA; (iii) dehydration to trans-2-enoyl-coa; 61

64 Chapter 3 and (iv) reduction to fully elongated fatty acyl-coa. The initial condensation reaction is catalyzed by the enzyme referred to as elongation of very long-chain fatty acids (ELOVL) and is considered to be rate limiting (Cinti et al 1992). While seven elongases have been identified in mammals (designated ELOVL1-ELOVL7)(Jakobsson et al 2006), only a single enzyme has been identified yet for the subsequent reaction step (Jakobsson et al 2006). Identification of the specific enzyme(s) affected by BF is not a trivial thing and requires detailed analysis of all enzymes involved, including: 3-ketoacyl-CoA reductase (HSD17B12), 3-hydroxyacyl dehydratase (HACD3) and the trans-2,3-enoyl-coa reductase (TECR). This will be the subject of future studies. 3 Concluding remarks The work described in the paper shows that inhibition of VLCFA synthesis by pharmacological means could be a feasible treatment option for X-ALD. BF is a good candidate for this approach. BF lowers the levels of C26:0 by a direct inhibition of the synthesis. Mouse studies to evaluate the in vivo effect of BF treatment on VLCFA in X-ALD mice would be interesting; however, the effect of fibrates is quite different in rodents and humans. Watanabe et al demonstrated that rats and mice are unusable as a model system (for primates) to the study the effect of BF (Watanabe et al 1989). BF has a proven safety profile for (long-term) use in humans. With a daily dose of 200 mg of BF peak plasma levels of 50 µm can be reached, with a maximum daily dose of 800 mg of BF therapeutic levels might be reached in plasma (Miller and Spence 1998). A small scale proof of principle clinical trial is currently ongoing to evaluate the effect in X-ALD patients. Acknowledgements We thank Femke Stet, Petra Mooyer, and Henk van Lenthe for expert technical assistance. This research was supported by the Netherlands Organization for Scientific Research (VIDIgrant number ), the European Union Framework Programme 7 (grant number LeukoTreat ). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. References Aubourg P, Adamsbaum C, Lavallard-Rousseau MC et al (1993) A two-year trial of oleic and erucic acids ("Lorenzo's oil") as treatment for adrenomyeloneuropathy. N Engl J Med 329: Baudhuin P, Beaufay H, Rahman-Li Y et al (1964) Tissue fractionation studies. 17. Intracellular distribution of monoamine oxidase, aspartate aminotransferase, alanine aminotransferase, D-amino acid oxidase and catalase in rat-liver tissue. Biochem J 92: Berger J, Albet S, Bentejac M et al (1999) The four murine peroxisomal ABC-transporter genes differ in constitutive, inducible and developmental expression. Eur J Biochem 265: Blaauboer BJ, van Holsteijn CW, Bleumink R et al (1990) The effect of beclobric acid and clofibric acid on peroxisomal beta-oxidation and peroxisome proliferation in primary cultures of rat, monkey and human hepatocytes. 62

65 In vitro studies Biochem Pharmacol 40: Bonnefont JP, Bastin J, Behin A, Djouadi F (2009) Bezafibrate for an inborn mitochondrial beta-oxidation defect. N Engl J Med 360: Brown FR, Van Duyn MA, Moser AB et al (1982) Adrenoleukodystrophy: effects of dietary restriction of very long chain fatty acids and of administration of carnitine and clofibrate on clinical status and plasma fatty acids. Johns Hopkins Med J 151: Cabrero A, Alegret M, Sanchez RM, Adzet T, Laguna JC, Vazquez M (2001) Bezafibrate reduces mrna levels of adipocyte markers and increases fatty acid oxidation in primary culture of adipocytes. Diabetes 50: Cinti DL, Cook L, Nagi MN, Suneja SK (1992) The fatty acid chain elongation system of mammalian endoplasmic reticulum. Progress in Lipid Research 31: Djouadi F, Aubey F, Schlemmer D et al (2005) Bezafibrate increases very-long-chain acyl-coa dehydrogenase protein and mrna expression in deficient fibroblasts and is a potential therapy for fatty acid oxidation disorders. Hum Mol Genet 14: Ellman G (1959) Tissue sulfhydryl groups. Arch Biochem Biophys 82: Engelen M, Ofman R, Mooijer PAW, Poll-The BT, Wanders RJA, Kemp S (2008) Cholesterol-deprivation increases mono-unsaturated very long-chain fatty acids in skin fibroblasts from patients with X-linked adrenoleukodystrophy. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 1781: Engelen M, Ofman R, Dijkgraaf MGW et al (2010) Lovastatin in X-Linked Adrenoleukodystrophy. N Engl J Med 362: Islinger M, Luers GH, Li KW, Loos M, Volkl A (2007) Rat liver peroxisomes after fibrate treatment. A survey using quantitative mass spectrometry. J Biol Chem 282: Jakobsson A, Westerberg R, Jacobsson A (2006) Fatty acid elongases in mammals: Their regulation and roles in metabolism. Progress in Lipid Research 45: Kehrer JP, Biswal SS, La E et al (2001) Inhibition of peroxisome-proliferator-activated receptor (PPAR)alpha by MK886. Biochem J 356: Kemp S, Mooyer PA, Bolhuis PA et al (1996) ALDP expression in fibroblasts of patients with X-linked adrenoleukodystrophy. J Inherit Metab Dis 19: Kemp S, Theodoulou FL, Wanders RJ (2011) Mammalian peroxisomal ABC transporters: from endogenous substrates to pathology and clinical significance. Br J Pharmacol doi: /j Kemp S, Wanders R (2010) Biochemical aspects of X-linked adrenoleukodystrophy. Brain Pathol 20: Kemp S, Wei HM, Lu JF et al (1998) Gene redundancy and pharmacological gene therapy: implications for X- linked adrenoleukodystrophy. Nat Med 4: Kishimoto Y, Moser HW, Kawamura N, Platt M, Pallante SL, Fenselau C (1980) Adrenoleukodystrophy: evidence that abnormal very long chain fatty acids of brain cholesterol esters are of exogenous origin. Biochem Biophys Res Commun 96: Miller DB, Spence JD (1998) Clinical pharmacokinetics of fibric acid derivatives (fibrates). Clin Pharmacokinet 34: Moon A, Rhead WJ (1987) Complementation analysis of fatty acid oxidation disorders. J Clin Invest 79:

66 Chapter 3 Moser HW, Smith KD, Watkins PA, Powers J, and Moser AB (2001). X-linked adrenoleukodystrophy. In Scriver,C.R., Beaudet,A.L., Sly,W.S., and Valle,D. (Eds.), The Metabolic and Molecular Bases of Inherited Disease,. McGraw Hill, New York, pp Mosser J, Douar AM, Sarde CO et al (1993) Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: Nagi MN, Cook L, Suneja SK et al (1989) Evidence for two separate beta-ketoacyl CoA reductase components of the hepatic microsomal fatty acid chain elongation system in the rat. Biochem Biophys Res Commun 165: Netik A, Forss-Petter S, Holzinger A, Molzer B, Unterrainer G, Berger J (1999) Adrenoleukodystrophy-related protein can compensate functionally for adrenoleukodystrophy protein deficiency (X-ALD): implications for therapy. Hum Mol Genet 8: Ofman R, Dijkstra IM, van Roermund CW et al (2010) The role of ELOVL1 in very long-chain fatty acid homeostasis and X-linked adrenoleukodystrophy. EMBO Mol Med 2: Pyper SR, Viswakarma N, Yu S, Reddy JK (2010) PPARalpha: energy combustion, hypolipidemia, inflammation and cancer. Nucl Recept Signal 8: e002. Rasmussen JT, Borchers T, Knudsen J (1990) Comparison of the binding affinities of acyl-coa-binding protein and fatty-acid-binding protein for long-chain acyl-coa esters. Biochem J 265: Sanchez RM, Vinals M, Alegret M et al (1993) Fibrates modify rat hepatic fatty acid chain elongation and desaturation in vitro. Biochem Pharmacol 46: Singh I, Khan M, Key L, Pai S (1998) Lovastatin for X-Linked Adrenoleukodystrophy. N Engl J Med 339: Smith PK, Krohn RI, Hermanson GT et al (1985) Measurement of protein using bicinchoninic acid. Anal Biochem 150: Tsuji S, Sano T, Ariga T, Miyatake T (1981) Increased synthesis of hexacosanoic acid (C26:0) by cultured skin fibroblasts from patients with adrenoleukodystrophy (ALD) and adrenomyeloneuropathy (AMN). J Biochem (Tokyo) 90: Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S (2003) Analysis of very long-chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 79: van Geel BM, Assies J, Haverkort EB et al (1999) Progression of abnormalities in adrenomyeloneuropathy and neurologically asymptomatic X-linked adrenoleukodystrophy despite treatment with "Lorenzo's oil". J Neurol Neurosurg Psychiatry 67: Vazquez M, Alegret M, Lopez M et al (1995) Different effects of fibrates on the microsomal fatty acid chain elongation and the acyl composition of phospholipids in guinea-pigs. Br J Pharmacol 116: Vazquez M, Roglans N, Cabrero A et al (2001) Bezafibrate induces acyl-coa oxidase mrna levels and fatty acid peroxisomal beta-oxidation in rat white adipose tissue. Mol Cell Biochem 216: Wanders RJ, Denis S, Ruiter JP, Schutgens RB, van Roermund CW, Jacobs BS (1995) Measurement of peroxisomal fatty acid beta-oxidation in cultured human skin fibroblasts. J Inherit Metab Dis 18 Suppl 1: Watanabe T, Horie S, Yamada J et al (1989) Species differences in the effects of bezafibrate, a hypolipidemic agent, on hepatic peroxisome-associated enzymes. Biochem Pharmacol 38:

67 Clinical trials Chapter 4 4 Lovastatin in X-linked adrenoleukodystrophy Marc Engelen 1,2, Rob Ofman 3, Marcel G.W. Dijkgraaf 4, Michiel Hijzen 3, Lucinda A. van der Wardt 5, Björn M. van Geel 1,6, Marianne de Visser 1, Ronald J.A. Wanders 3, Bwee Tien Poll-The 1,2 and Stephan Kemp 2,3 Departments of Neurology 1, Pediatric Neurology 2, Laboratory Genetic Metabolic Diseases 3, Clinical Epidemiology, Biostatistics & Bio-informatics 4 and Nutrition 5, Emma Children s Hospital/Academic Medical Center, University of Amsterdam, The Netherlands. Department of Neurology 6, Medical Center Alkmaar, Alkmaar, The Netherlands. The New England Journal of Medicine (2010) 362(3): (published in abbreviated form) 65

68 Chapter 4 Abstract 4 Background X-linked adrenoleukodystrophy (X-ALD) is a peroxisomal disorder and is characterized by rapidly progressive cerebral demyelination or a gradually progressive myelopathy with or without adrenocortical insufficiency. The disease is caused by mutations in the ABCD1 gene resulting in impaired peroxisomal beta-oxidation and the subsequent accumulation of very long-chain fatty acids (VLCFA; >C22) in plasma and tissues. In 1998 it was reported in this Journal that lovastatin reduces total plasma VLCFA in X-ALD patients. This finding led to treatment with lovastatin in many patients with X-ALD despite contradicting data on the effect of lovastatin on plasma VLCFA. The aim of this clinical trial was to confirm that in patients with X-ALD lovastatin treatment can reduce C26:0 levels in plasma, but also in lymphocytes and erythrocytes. Methods The study was designed as a randomized double-blind placebo controlled cross-over trial (ISRCTN ). With a total of 14 patients, a 50% reduction of C26:0 levels can be detected with a power of 80%. The primary outcome measures were reduction of LDL-cholesterol in plasma, C26:0 levels in plasma and LDL-lipoprotein particles, lymphocytes and erythrocytes. Results Fourteen male patients with X-ALD were enrolled in the trial. LDL-cholesterol in plasma was reduced by 40%. Plasma C26:0 levels were reduced by almost 20% (but this effect was also observed for non-vlcfa). There was no effect on C26:0 levels in lymphocytes, erythrocytes and LDL-lipoprotein particles. Conclusion The reduction of plasma C26:0 is non-specific and linked to a reduction in LDL-lipoprotein particles. Lovastatin is not a treatment option for X-ALD. 66

69 Clinical trials Introduction X-linked adrenoleukodystrophy (X-ALD) is the most common peroxisomal disorder with a birth incidence of 1: (Bezman et al 2001). It is characterized by impaired peroxisomal beta-oxidation of very long-chain fatty acids (VLCFA; > C22:0), which accumulate in all tissues (Moser et al 2001). X-ALD is caused by mutations in the ABCD1 gene (see also the X-ALD database at resulting in the absence or dysfunction of a peroxisomal transmembrane protein, the ALD protein (ALDP) (Mosser et al 1993). The disease is characterized by a highly variable clinical expression, even within families (Kemp et al 2001). Often, the first manifestation of the disease in childhood is isolated adrenocortical insufficiency ( Addison-only phenotype) without neurological dysfunction. About 40% of affected male patients develop rapidly progressive cerebral demyelination causing severe disability and death usually within 2 years after symptom onset (Moser et al 2001). This usually occurs between the age of 3 and 10 years (childhood cerebral ALD; CCALD), but is more rarely also seen in adolescence (adolescent cerebral ALD; AdolCALD), or adulthood (adult cerebral ALD; ACALD) (Moser et al 2001). In adulthood the most common phenotype is adrenomyeloneuropathy (AMN), a gradually progressive myelopathy and peripheral neuropathy, causing severe disability. Symptoms usually appear in the 3 rd or 4 th decade. Patients with AMN can also develop secondary cerebral demyelination ( AMN cerebral )(Van Geel et al 2001). Currently, treatment options are very limited and are mostly symptomatic. Only in a very early stage of CCALD progression can be halted or reversed by hematopoietic stem cell transplantation (Peters et al 2004). Treatment with Lorenzo s oil, a 4:1 mixture of C18:1 and C22:1 in combination with a low-fat diet rapidly normalizes plasma VLCFA, but seems to have no effect on disease progression in several open-label clinical trials (Aubourg et al 1993; van Geel et al 1999). Previously, it was reported that lovastatin lowers VLCFA in patients with X-ALD (Singh et al 1998b; Pai et al 2000). This finding, however, could not be reproduced with simvastatin (Verrips et al 2000). Subsequently performed animal experiments showed that lovastatin has no effect on brain and adrenal VLCFA levels in Abcd1 knockout mice (Yamada et al 2000), and simvastatin even caused accumulation of VLCFA in these tissues (Cartier et al 2000). In vitro experiments with cultured skin fibroblasts grown in cholesterol-depleted culture medium showed a reduction in VLCFA levels (Weinhofer et al 2002). We recently demonstrated that this reflects a shift to increased synthesis of mono-unsaturated VLCFA (Engelen et al 2008). This trial was designed to investigate if lovastatin indeed has a biochemical effect in vivo in patients with X-ALD, and to help us decide whether a large scale trial with clinical outcome parameters is warranted. 4 Methods Eligible patients Patients were recruited from the AMC neurology outpatient clinic which is a referral center for peroxisomal disorders and through the patient support organization. Men with the AMN phenotype of X-ALD, confirmed by plasma VLCFA analysis and/or ABCD1 mutation analysis, were eligible. Exclusion criteria were the use of Lorenzo s oil and/or cholesterol lowering therapy, contraindications for lovastatin use (liver- or kidney failure), or impairment so se- 67

70 Chapter 4 vere that the patient was unable to visit the outpatient clinic. The study protocol was approved by the local institutional review board. The trial was registered with the Dutch Trial Register (#852) and Current Controlled Trials (ISRCTN ). 4 Treatment The study was designed as a randomized double-blind cross-over placebo controlled trial (Figure 1). Eligible patients were randomized and started on a run-in phase with a low fat diet. A randomization list was created by the Biostatistics department. Prepackaged medication sets with identical appearance were created by the Stichting Haarlemse Ziekenhuisapotheken (GCP certified foundation of hospital pharmacies). Patients entering the trial were given consecutive trial numbers, matching a set of prepackaged medication (twelve bottles numbered consecutively). After a one month run-in phase in which a standard low fat diet was administered by the dietician (as recommended by the American Heart Association) (Lichtenstein et al., 2006), medication or placebo was started, followed by washout and switch to the other treatment arm. Adverse effects were recorded. Treatment assignments were concealed from all investigators. Blood samples were taken at 0, 4, 12, 26, 30, 38 and 52 weeks. The code was broken after all patients had completed the follow-up visits and the data analyzed. Laboratory studies At each visit four standard Vacutainer tubes (2 with heparin, 2 with EDTA) were collected. Routine measurements (creatine kinase and transaminase activities in serum, and a plasma lipid spectrum) were performed at the Department of Clinical Chemistry. The samples for VLCFA measurements were processed the same day at the Laboratory for Genetic Metabolic Diseases. Blood was centrifuged in Leukoprep tubes for 20 minutes at 2000 rpm. The plasma fraction was stored in separate tubes at -80 C after snap freezing in liquid nitrogen. The lymphocyte pellet was resuspended in lysisbuffer twice to remove erythrocytes, and then washed with cold saline and stored at -80 C. The erythrocytes were washed twice with cold saline and stored at -80 C. VLCFA were measured as described previously (Kemp et al 2005). Lipoprotein fractions were isolated from plasma as described previously (Innis-Whitehouse et al 1998). Outcome measures Endpoints were plasma total cholesterol, LDL-cholesterol, C18:1, C24:0 and C26:0 levels, lymphocyte and erythrocyte C26:0 levels, and C18:1, C24:0 and C26:0 content of the LDL- 68

71 Clinical trials lipoprotein fraction at 22 weeks after start of treatment. For plasma total cholesterol, LDLcholesterol, C18:1, C24:0 and C26:0 levels, and lymphocyte C26:0 levels intermediate endpoints at 8 weeks after start of treatment were assessed as well. Sample size It was assumed that the mean baseline C26:0 level prior to treatment would be about 2.94 µmol/l with a standard deviation of 0.87 µmol/l (Valianpour et al 2003). It was further assumed that lovastatin treatment might result in a 50% drop of this level. Based on the crossover design, all patients were to be analyzed within their randomization groups by comparing the change scores over the treatment periods (see below). The change scores in the groups would then be plus 1.47 (placebo minus lovastatin) and minus 1.47 (lovastatin minus placebo) respectively. Assuming a conservatively estimated zero correlation between the outcome measurements for the first and the second treatment period within each group, the standard deviation of the change scores in each group would be Using a two groups t-test with a 0.05 two-sided significance level to detect a difference in these change scores of 2.94 assuming a common standard deviation of 1.739, 80% power would be achieved with a sample size of 7 in each group of the 2x2 cross-over design (or 14 in total). 4 Statistical analysis Available pre-diet score distributions for plasma total cholesterol, LDL-cholesterol, C18:1, C24:0, C26:0 and lymphocyte C26:0 were assessed for normality by one sample Kolmogorov-Smirnov tests. Given the small number of patients in both trial arms, the pre-diet scores were tested for inequality with two-sample two-sided Student s t-tests. Based on the cross-over design, all patients were analyzed within their randomization groups. The analysis was performed with correction for non-significant pre-treatment baseline differences following the run-in phase with standard low fat diet, that were in the same direction as the expected differences between placebo and lovastatin under the alternative hypothesis that there is a significant difference between lovastatin and placebo treatment. The correction was performed by using the changes from baseline in each treatment period. In the group receiving placebo followed by lovastatin, we first subtracted baseline from outcome scores during placebo treatment as well as during treatment with lovastatin. Subsequently, change scores over treatment periods were determined by subtracting the observed changes under the lovastatin regimen from the observed changes under the placebo regimen. In the group receiving lovastatin followed by placebo, we first subtracted baseline from outcome scores during treatment with lovastatin as well as during placebo. Subsequently, change scores over treatment periods were determined by subtracting the observed changes under the placebo regimen from the observed changes under the lovastatin regimen. Data was analyzed using the SPSS software for Windows, version 16 (SPSS, Chicago, IL). P- values below 0.05 indicated statistical significance. Graphs were made in GraphPad Prism (GraphPad Software, La Jolla, CA). Results Patients Twenty-two patients were considered for inclusion in the study. Five were excluded from 69

72 Chapter 4 participation (one was on Lorenzo s oil, 2 were already on statin therapy, 2 were too severely impaired to complete the trial). Three patients declined to participate. Fourteen patients were randomized. No significant differences were observed for the randomized groups. There were no serious adverse events. All patients completed the study. 4 Biochemical results In Table 1 data and change scores from baseline for 8 weeks and 22 weeks following start of treatment are presented for the all outcome parameters. The data is further summarized in scatter plots (Figure 2). At the first (8 weeks of treatment) and second (22 weeks of treatment) time point, there is a significant decrease in total and LDL-cholesterol levels in plasma of 38%. The plasma C24:0 and C26:0 levels are decreased initially by 20% and 18% respectively, although the difference is no longer statistically significant at the second time point for C26:0. There is a reduction of 17% in C18:1 levels at the first and second time point. There is no change in C26:0 levels in erythrocytes or lymphocytes at either time point. The C18:1, C24:0 and C26:0 content of LDL lipoprotein particles are unchanged. Discussion This study shows that lovastatin initially reduces C26:0 levels by about 20% in the plasma of patients with X-ALD, but fails to do so in lymphocytes or erythrocytes. It has been reported previously that lovastatin lowers plasma VLCFA in patients with X-ALD (Singh et al 1998b; Pai et al 2000). This prompted many patients worldwide to take lovastatin, even though no clinical trial with clinical endpoints showing a beneficial effect has been performed and both in vitro and in vivo data on results with lovastatin and cholesterol-lowering are conflicting (Singh et al 1998a; Cartier et al 2000; Yamada et al 2000; Weinhofer et al 2002; Engelen et al 2008). A trial with clinical endpoints is difficult to perform, considering the low incidence of X-ALD and the unpredictable disease course, requiring large groups and a long follow-up. It seemed premature to perform this trial for lovastatin since the biological plausibility is low, considering the conflicting data in the literature mentioned above. Therefore, our aim was to first perform a proof of principle trial with biochemical endpoints. Compliance in the trial was excellent, since there was a clear decrease in LDL-cholesterol levels of almost 40% during the lovastatin treatment period. There was also no drop-out in this small trial with highly motivated participants. There were no adverse effects. We demonstrated a decrease in C26:0 levels in plasma of roughly 20%. However, even with this decrease, levels remain between 2 and 3 times above the control level of 0.67 ± 0.13 µmol/l (Valianpour et al 2003) and the effect seems to diminish over time (the decrease is no longer statistically significant after 22 weeks of treatment). The decrease is not specific for VLCFA since C18:1 (a long chain mono-unsaturated fatty acid) is also reduced in plasma by lovastatin. It is of note that treatment with lovastatin does not affect VLCFA at the cellular level, since C26:0 levels in erythrocytes and lymphocytes were unchanged. Since VLCFA are virtually water insoluble and only a small fraction binds to albumin (Ho et al 1995), most of the plasma VLCFA is transported as cholesterol-esters in lipoprotein particles like LDL. To establish whether plasma VLCFA reduction is linked to a reduction in LDL cholesterol we measured the C18:1, C24:0 and C26:0 content of the LDL lipoprotein fraction. Lovastatin treatment did not decrease the C18:1, C24:0 or C26:0 content of LDL-cholesterol (Table 1). Based on these observations we conclude that lovastatin leads to a small decrease in C24:0 and C26:0 levels in plasma which 70

73 Clinical trials Table 1: Major outcome measures after 8 weeks (if applicable) and 22 weeks of treatment with lovastatin at a dose of 40 mg once daily* 8 wk of treatment 22 wk of treatment 95% CI P value Mean change 95% CI P value Baseline Mean change Plasma Total cholesterol ± to 2.02 < ± to 1.98 <0.01 LDL ± to 1.87 < ± to 1.82 <0.01 C18: ± to ± to C24: ± to 18.5 < ± to C26: ± to ± to Lipoprotein Fraction C18:1 in LDL 1540 ND NA NA -84 ± to C24:0 in LDL 64 ND NA NA 3.4 ± to C26:0 in LDL 9.7 ND NA NA -1.9 ± to Cells C26:0 (lymphocytes) ± to to C26:0 (erythrocytes) 0.18 ND NA NA ± to * Plus minus values are means ±SE. The mean reduction indicates the absolute change from baseline levels after treatment. P values were calculated with the use of a two-sided, unpaired Student s t-test. ApoB denotes apolipoprotein B, CI confidence interval, C18:1 oleic acid, C24:0 tetracosanoic acid, C26:0 hexacosanoic acid, NA not applicable, and ND not determined. Equal variances were not assumed. 4 has to be considered a non-specific result of the LDL-cholesterol decrease. This is corroborated by the finding that C18:1 is also reduced and further supported by the lack of effect on C26:0 levels in peripheral blood lymphocytes and erythrocytes, and in LDL-lipoprotein fraction VLCFA content. 71

74 Chapter 4 4 Figure 2: Scatter plots of the effect of lovastatin on levels of plasma total cholesterol (A), LDLcholesterol (B), plasma C24:0 (C) and C26:0 (D), C24:0 (E) and C26:0 (F) in LDL lipoprotein particles, C26:0 in lymphocytes (G) and C26:0 in erythrocytes (H). TP1 and TP2 are 8 or 22 weeks. Units are mmol/l (A and B), µmol/l (C and D), pmol/mmol ApoB (E and F) and nmol/mg protein (G) and % of total fatty acids (H). Error bars indicate the mean and standard deviation. *** p <

75 Clinical trials Conclusion It seems unnecessary to invest a great deal of resources and time in a trial with clinical endpoints. Physicians should not prescribe lovastatin as a VLCFA lowering therapy to patients with X-ALD, since evidence does not support it. Acknowledgements The authors gratefully acknowledge the patients who were willing to participate in this trial. The work described here was supported by grants from the European Leukodystrophy Association (ELA: I5 to BPT and I4 to SK) and the Netherlands Organization for Scientific Research (NWO-VIDI-grant to SK). References 4 Aubourg P, Adamsbaum C, Lavallard-Rousseau MC, Rocchiccioli F, Cartier N, Jambaque I, Jakobezak C, Lemaitre A, Boureau F, Wolf C, A two-year trial of oleic and erucic acids ( Lorenzo s oil ) as treatment for adrenomyeloneuropathy. N Engl J Med 329: Bezman L, Moser AB, Raymond GV, Rinaldo P, Watkins PA, Smith KD, Kass NE, Moser HW Adrenoleukodystrophy: incidence, new mutation rate, and results of extended family screening. Ann Neurol 49: Cartier N, Guidoux S, Rocchiccioli F, Aubourg P Simvastatin does not normalize very long chain fatty acids in adrenoleukodystrophy mice. FEBS Lett 478: Engelen M, Ofman R, Mooijer PA, Poll-The BT, Wanders RJ, Kemp S Cholesterol-deprivation increases monounsaturated very long-chain fatty acids in skin fibroblasts from patients with X-linked adrenoleukodystrophy. Biochim Biophys Acta 1781: Ho JK, Moser H, Kishimoto Y, Hamilton JA Interactions of a very long chain fatty acid with model membranes and serum albumin. Implications for the pathogenesis of adrenoleukodystrophy. J Clin Invest 96: Innis-Whitehouse W, Li X, Brown WV, Le NA An efficient chromatographic system for lipoprotein fractionation using whole plasma. J Lipid Res 39: Kemp S, Pujol A, Waterham HR, van Geel BM, Boehm CD, Raymond GV, Cutting GR, Wanders RJ, Moser HW ABCD1 mutations and the X-linked adrenoleukodystrophy mutation database: role in diagnosis and clinical correlations. Hum Mutat 18: Kemp S, Valianpour F, Denis S, Ofman R, Sanders RJ, Mooyer P, Barth PG, Wanders RJ Elongation of very longchain fatty acids is enhanced in X-linked adrenoleukodystrophy. Mol Genet Metab 84: Lichtenstein AH, Appel LJ, Brands M, Carnethon M, Daniels S, Franch HA, Franklin B, Kris-Etherton P, Harris WS, Howard B, Karanja N, Lefevre M, Rudel L, Sacks F, Van HL, Winston M, Wylie-Rosett J Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation 114: Moser HW, Smith KD, Watkins PA, Powers J, Moser AB X-linked adrenoleukodystrophy. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The Metabolic and Molecular Bases of Inherited Disease. New York: Mc Graw Hill. p

76 Chapter 4 Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL, Aubourg P Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: Pai GS, Khan M, Barbosa E, Key LL, Craver JR, Cure JK, Betros R, Singh I Lovastatin therapy for X-linked adrenoleukodystrophy: clinical and biochemical observations on 12 patients. Mol Genet Metab 69: Peters C, Charnas LR, Tan Y, Ziegler RS, Shapiro EG, DeFor T, Grewal SS, Orchard PJ, Abel SL, Goldman AI, Ramsay NK, Dusenbery KE, Loes DJ, Lockman LA, Kato S, Aubourg PR, Moser HW, Krivit W Cerebral X-linked adrenoleukodystrophy: the international hematopoietic cell transplantation experience from 1982 to Blood 104: Singh I, Pahan K, Khan M. 1998a. Lovastatin and sodium phenylacetate normalize the levels of very long chain fatty acids in skin fibroblasts of X- adrenoleukodystrophy. FEBS Lett 426: Singh I, Khan M, Key L, Pai S. 1998b. Lovastatin for X-linked adrenoleukodystrophy. N Engl J Med 339: Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S Analysis of very long-chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 79: van Geel BM, Assies J, Haverkort EB, Koelman JH, Verbeeten B, Jr., Wanders RJ, Barth PG Progression of abnormalities in adrenomyeloneuropathy and neurologically asymptomatic X-linked adrenoleukodystrophy despite treatment with Lorenzo s oil. J Neurol Neurosurg Psychiatry 67: Van Geel BM, Bezman L, Loes DJ, Moser HW, Raymond GV Evolution of phenotypes in adult male patients with X-linked adrenoleukodystrophy. Ann Neurol 49: Verrips A, Willemsen MA, Rubio-Gozalbo E, De JJ, Smeitink JA Simvastatin and plasma very-long-chain fatty acids in X-linked adrenoleukodystrophy. Ann Neurol 47: Weinhofer I, Forss-Petter S, Zigman M, Berger J Cholesterol regulates ABCD2 expression: implications for the therapy of X-linked adrenoleukodystrophy. Hum Mol Genet 11: Yamada T, Shinnoh N, Taniwaki T, Ohyagi Y, Asahara H, Horiuchi, Kira J Lovastatin does not correct the accumulation of very long-chain fatty acids in tissues of adrenoleukodystrophy protein-deficient mice. J Inherit Metab Dis 23:

77 Clinical trials Chapter 5 5 Bezafibrate for X-linked adrenoleukodystrophy Marc Engelen 1,3, Luc Tran 2, Rob Ofman 2, Josephine Brennecke 2, Ann B. Moser 4, Inge M.E. Dijkstra 2, Ronald J.A. Wanders 2, Bwee Tien Poll-The 1,3, Stephan Kemp 2,3 1 Department of Neurology, 2 Department of Clinical Chemistry, Laboratory Genetic Metabolic Diseases, and 3 Department of Pediatric Neurology/Emma Children s Hospital, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 4 Department of Neurogenetics, Kennedy Krieger Institute, Baltimore, MD, USA. PLoS ONE (2012) 7(7): e

78 Chapter 5 Abstract 5 X-linked adrenoleukodystrophy (X-ALD) is caused by mutations in the ABCD1 gene and is characterized by impaired beta-oxidation of very-long-chain fatty acids (VLCFA) and subsequent VLCFA accumulation in tissues. In adulthood X-ALD most commonly manifests as a gradually progressive myelopathy, (adrenomyeloneuropathy; AMN) without any curative or disease modifying treatments. We recently showed that bezafibrate (BF), a drug used for the treatment of hyperlipidaemia, reduces VLCFA accumulation in X-ALD fibroblasts by inhibiting ELOVL1, an enzyme involved in the VLCFA synthesis. We therefore designed a proof-of-principal clinical trial to determine whether BF reduces VLCFA levels in plasma and lymphocytes of X-ALD patients. Ten males with AMN were treated with BF for 12 weeks at a dose of 400 mg daily, followed by 12 weeks of 800 mg daily. Every 4 weeks patients were evaluated for side effects and blood samples were taken for analysis. Adherence was good as indicated by a clear reduction in triglycerides. There was no reduction in VLCFA in either plasma or lymphocytes. Plasma levels of BF did not exceed 25 µmol/l. We concluded that BF, at least in the dose given, is unable to lower VLCFA levels in plasma or lymphocytes in X-ALD patients. It is unclear whether this is due to the low levels of BF reached in plasma. Our future work is aimed at the identification of highly-specific inhibitors of ELOVL1 that act at much lower concentrations than BF and are well tolerated. BF appears to have no therapeutic utility in X-ALD. 76

79 Clinical trials Introduction X-linked adrenoleukodystrophy (X-ALD) is a peroxisomal disorder characterized by impaired beta-oxidation of very long-chain fatty acids (VLCFA) and accumulation of these VLCFA in tissues (Moser et al 2007). It is caused by mutations in the ABCD1 gene ( (Mosser et al 1993). The disease is highly variable in clinical expression, however, in adulthood it most frequently manifests as a gradually progressive myelopathy and peripheral neuropathy (adrenomyeloneuropathy phenotype or AMN) (Moser et al 2007). Treatment for AMN is purely symptomatic and currently there is no proven intervention that can halt progression of the disease (Moser et al 2007). We identified ELOVL1 as the enzyme responsible for the synthesis of VLCFA (Ofman et al 2010), and demonstrated that sirna-mediated knockdown of ELOVL1 lowers VLCFA levels in X-ALD fibroblasts (Ofman et al 2010). Next, we showed that bezafibrate (BF) reduces VLCFA levels in X-ALD fibroblasts by directly inhibiting ELOVL1 (Engelen et al 2012). BF is a drug of the fibrate class for the treatment of dyslipidaemia and has a proven safety profile for (long-term) use in humans (Miller and Spence, 1998). We therefore designed a proof of principal clinical trial to test whether BF can reduce VLCFA levels in the plasma and lymphocytes of patients with X-ALD. 5 Figure 1: Schematic representation of the BEZA trial design. Methods The protocol for this trial and supporting CONSORT checklist are available as supporting information; see Checklist S1 and Protocol S1 (online only). The BEZA trial study protocol was approved by the Institutional Review Board (Medisch Ethische Toetsings Commissie) of the Academic Medical Center. The trial is registered at clinicaltrials.gov (NCT ). Adult men with biochemically and genetically proven X-ALD without contra-indications for the use of BF were eligible for inclusion. All participating patients were evaluated at baseline for eligibility and received trial medication after written informed consent was obtained. They were evaluated at intervals of 4 weeks until the end of the trial at 24 weeks. The initial dose of BF was 400 mg per day, which was subsequently increased to 800 mg per day at week 12 (Figure 1). At each visit side effects were monitored, a general physical examination including weight was performed and blood samples taken. Blood samples were taken in the morning after an overnight fast before the first medication dose. Blood samples were analyzed at the laboratory for clinical chemistry for routine laboratory tests. VLCFA and BF levels were analyzed as previously described (Masnatta et al 1996; Valianpour et al 2003). Lysophosphatidylcholine-C26:0 (C26:0 lysopc) was analyzed in bloodspots (Hubbard et al 2009). Data were analyzed with PASW statistics, version 18 (IBM). Statistical significance was evaluated with Student s paired t-test. 77

80 Chapter 5 Results Ten males with AMN participated in the trial. No side effects that necessitated discontinuation of the trial medication occurred. Body weight was unchanged (Table 1). There was a clear reduction in plasma triglycerides (1.34 mmol/l to 0.70 mmol/l at BF 400 mg and 0.71 mmol/l at BF 800 mg), and to a lesser extent a decrease in total cholesterol and LDL-cholesterol. There was also an increase in HDL-cholesterol (Table 1). These are known effects of BF and confirm patient adherence. There was no consistent reduction in C26:0 in plasma or lymphocytes, neither at 400 nor at 800 mg BF per day (Table 1). We observed an increase in plasma C22:0 and C24:0 at a dose of 800 mg BF per day. The amount of C26:0 lysopc was unchanged in blood spots after 24 weeks of treatment with BF. The plasma level of BF did not exceed 25 µmol/l at the highest dose of 800 mg BF per day. 5 Table 1: Summary of the different parameters measured at the indicated time point in the trial. Plasma Baseline BF 400 mg BF 800 mg Total cholesterol (mmol/l) 5.57 ± ± 0.88** 4.85 ± 0.84* LDL (mmol/l) 3.67 ± ± 0.80** 2.90 ± 0.73* HDL (mmol/l) 1.39 ± ± 0.25*** 1.64 ± 0.26*** TG (mmol/l) 1.34 ± ± 0.31** 0.71 ± 0.22* C22:0 (µmol/l) ± ± ± 9.74** C24:0 (µmol/l) ± ± ± 12.50*** C26:0 (µmol/l) 3.26 ± ± 0.50** 2.99 ± 0.49 C26:0/C22:0 ratio ± ± ± 0.012** Bezafibrate (µmol/l) n.d. n.d ± 6.7 Lymphocytes Baseline BF 400 mg BF 800 mg C22:0 (nmol/mg) 5.89 ± ± ± 1.47 C24:0 (nmol/mg) 6.03 ± ± ± 1.44 C26:0 (nmol/mg) 0.35 ± ± ± 0.11 C26:0/C22:0 ratio 0.06 ± ± ± 0.03 bloodspots Baseline BF 400 mg BF 800 mg C26:0 lysopc 2.84 ± ± ± 1.18 Weight (kg) 86.7 ± 9.3 n.d ± 10.5 Summary of the different parameters measured at the indicated time point in the trial. Values are mean ± the standard deviation. Statistically significant differences from the baseline value are indicated. * p < 0.05, ** p < 0.01, *** p < n.d. = not determined. 78

81 Clinical trials Discussion The pathophysiology of X-ALD is not well understood, although it seems likely that accumulation of VLCFA is toxic and related to neurodegeneration (Kemp et al 2012). Therefore drugs that reduce the level of VLCFA might be effective in halting or slowing progression of the disease. Recently, we showed that it is possible to reduce VLCFA in fibroblasts from X-ALD patients by inhibiting the synthesis of VLCFA by the enzyme ELOVL1 (Ofman et al 2010). We later showed that this can also be accomplished by incubating fibroblasts from X-ALD patients with BF (Engelen et al 2012). BF is a drug that has been in use for decades for the treatment of hypertriglyceridaemia and has an excellent safety profile (Miller and Spence, 1998). Therefore we decided to initiate this small scale proof of principle clinical trial to investigate whether BF reduces VLCFA in plasma and lymphocytes of X-ALD patients. In a previous clinical trial with lovastatin we demonstrated that reduction of plasma VLCFA can be an artifact of LDL reduction and does not reflect a reduction in blood cells (Engelen et al 2010). Unfortunately, we could not show a reduction of plasma or lymphocyte VLCFA levels. Conversely, there was an unexpected increase in C22:0 and C24:0 levels in plasma. We did not observe this in blood cells or bloodspots. Our results show that there is no rationale for a large follow-up trial with clinical endpoints utilizing this compound. The concept of treating X-ALD patients with an inhibitor of VLCFA synthesis remains a feasible option. It seems that BF is simply not efficacious enough. Our previous work suggests that BF is a competitive inhibitor of ELOVL1 (Engelen et al 2012). In our cell culture experiments a high concentration of BF of 400 µmol/l was required to achieve a maximal effect on the level of VLCFA. At this concentration the de novo VLCFA synthesis was reduced to the level in control cells. It is likely that even with the high dose of 800 mg BF per day, the intracellular levels of BF remained inadequate. Indeed, at the highest BF dosage plasma levels did not exceed 25 µmol/l with an average of 10 µmol/l (Table 1). These levels are not peak levels, but rather residual plasma levels. It is unlikely that concentrations even approaching 400 μmol/l were reached. This may explain the lack of in vivo efficacy of BF on our outcome parameters. To achieve the effect of VLCFA reduction, significantly higher BF concentrations are necessary as compared to concentrations indicated for reduction of TG. Future research will be focused on the identification of specific inhibitors of ELOVL1 that act at much lower concentrations than BF and are well-tolerated. In conclusion, BF appears to have no therapeutic utility in X-ALD. 5 Acknowledgements This research was supported by the Netherlands Organization for Scientific Research (VIDI ), the European Union Framework Programme 7 (LeukoTreat ) and The Stop ALD Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 79

82 Chapter 5 References Engelen M, Schackmann MJ, Ofman R, Sanders RJ, Dijkstra IM, Houten SM, Fourcade S, Pujol A, Poll-The BT, Wanders RJ, Kemp S Bezafibrate lowers very long-chain fatty acids in X-linked adrenoleukodystrophy fibroblasts by inhibiting fatty acid elongation. J Inherit Metab Dis. Engelen M, Ofman R, Dijkgraaf MGW, Hijzen M, van der Wardt LA, van Geel BM, de Visser M, Wanders RJA, Poll-The BT, Kemp S Lovastatin in X-Linked Adrenoleukodystrophy. N Engl J Med 362: Hubbard WC, Moser AB, Liu AC, Jones RO, Steinberg SJ, Lorey F, Panny SR, Vogt RF, Jr., Macaya D, Turgeon CT, Tortorelli S, Raymond GV Newborn screening for X-linked adrenoleukodystrophy (X-ALD): validation of a combined liquid chromatography-tandem mass spectrometric (LC-MS/MS) method. Mol Genet Metab 97: Kemp S, Berger J, Aubourg P X-linked adrenoleukodystrophy: Clinical, metabolic, genetic and pathophysiological aspects. Biochim Biophys Acta 1822: Masnatta LD, Cuniberti LA, Rey RH, Werba JP Determination of bezafibrate, ciprofibrate and fenofibric acid in human plasma by high-performance liquid chromatography. J Chromatogr B Biomed Appl 687: Miller DB, Spence JD Clinical pharmacokinetics of fibric acid derivatives (fibrates). Clin Pharmacokinet 34: Moser HW, Mahmood A, Raymond GV X-linked adrenoleukodystrophy. Nat Clin Pract Neurol 3: Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL, Aubourg P Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: Ofman R, Dijkstra IM, van Roermund CW, Burger N, Turkenburg M, van Cruchten A, van Engen CE, Wanders RJ, Kemp S The role of ELOVL1 in very long-chain fatty acid homeostasis and X-linked adrenoleukodystrophy. EMBO Mol Med 2: Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S Analysis of very long-chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 79:

83 Extension of the phenotype Chapter 6 X-linked adrenomyeloneuropathy due to a novel missense mutation in the ABCD1 start codon presenting as demyelinating neuropathy 6 Marc Engelen 1,2, Anneke J. van der Kooi 1, Stephan Kemp 2, Ronald J.A. Wanders 2, Erik A. Sistermans 3, Johannes T.M. Koelman 1, Hans R. Waterham 2, Björn M. van Geel 1, 4, Marianne de Visser 1 1 Department of Neurology and 2 Laboratory Genetic Metabolic Diseases, Academic Medical Center, Amsterdam; 3 Department of Clinical Genetics, VU Medical Center, Amsterdam; and 4 Department of Neurology, Medical Center Alkmaar, The Netherlands Journal of the Peripheral Nervous System (2011) 16:

84 Chapter 6 6 X-linked adrenoleukodystrophy (X-ALD) is caused by mutations in the ABCD1 gene which encodes the adrenoleukodystrophy protein (ALDP), a peroxisomal transmembrane protein (Mosser et al 1993; Moser et al 2001). Several often overlapping phenotypes can be distinguished. Approximately, half of all patients develop adrenomyeloneuropathy (AMN), characterized by slowly progressive spastic paraparesis, peripheral neuropathy, sphincter dysfunction, and adrenocortical insufficiency (Moser et al 2001). We present a patient with an unusual neuropathy who was shown to have X-ALD caused by a mutation in the ABCD1 gene. A 27-year-old man presented with exercise-related weakness and fatigue in the legs which progressed slowly over the last 2 years. Three years prior to presentation he shortly used testosterone replacement therapy for decreased libido. Family history was negative for neuromuscular and endocrinological disorders. Physical examination revealed slight generalized wasting and weakness of the legs and decreased sensation for light touch at the dorsum of both feet with decreased position sense of the toes. Pes cavus was not present. Deep tendon reflexes were normal, and plantar reflexes were flexor. Electrophysiologic findings were compatible with a mild, symmetric demyelinating neuropathy confined to the lower limbs (Table 1). Serum creatine kinase activity was elevated (270 U/L, normal <193 U/L). Normal or negative results were obtained for thyroid stimulating hormone, sodium and potassium, and vasculitis parameters (anti-neutrophil cytoplasmic antibody, antinuclear antibody and complement levels). Cerebrospinal fluid examination showed elevated protein (0.66 g/l, upper limit of normal 0.49). DNA analysis revealed no mutations in the PMP22 gene. Refsum s disease was considered, but the concentration of phytanic acid in plasma was normal. When phytanic acid levels are determined in our laboratory, we routinely include very long-chain fatty acids (VLCFA) as well. The VLCFA plasma C26:0 concentration was 4.32 μmol/l (normal μmol/l), and C26:0/C22:0 ratio was 0.09 (normal 0.02). These results prompted mutation analysis of the ABCD1 gene, which revealed a missense mutation (c.1a>g) in the start codon of the ABCD1 gene, resulting in the substitution of the initiator methionine to a valine residue (p.met1val). The effect of the mutation on the stability of ALDP was investigated in cultured skin fibroblasts by means of immunofluorescence (IF) microscopy analysis and western blot analysis (Ligtenberg et al 1995). Both IF and western blot analysis showed no detectable ALDP (Figure 1). These data indicate that most likely translation is not initiated. Endocrinological investigations revealed mild adrenocortical insufficiency with a dehydroepiandrosteron sulphate level of 5.9 µmol/l (normal 8 17) that did not increase after administration of tetracosactide. The basal morning cortisol level was 250 μmol/l (normal ) with a normal increase after administration of tetracosactide. The adrenocorticotropic hormone level was 14 ng/l (normal <55). There was also testicular insufficiency with a testosterone level of 7.9 nmol/l (normal ), sex hormone binding globulin 18 nmol/l (normal 16 38), free androgen index 59.5 (normal 49 89), luteinizing hormone (LH) 4.0 U/L (<15), and follicle stimulating hormone (FSH) 3.1 U/L (normal ). MRI of the cerebrum (Loes X-ALD severity score 0) (Loes et al., 1994) and spinal cord was normal. Additional DNA analysis revealed that the patient s mother was an asymptomatic carrier for X-ALD. Recently, the patient s tendon reflexes have become more brisk, suggestive of spinal cord involvement. X-ALD is a clinically heterogeneous disorder. Most patients either have childhood cerebral adrenoleukodystrophy or the AMN phenotype (Moser et al 2001). Our patient s presentation is unusual in that he presented with exercise-related weakness and fatigue in the legs and was found to have a demyelinating neuropathy, only in the legs. We found one similar case in a series of teenagers with polyneuropathy (Kararizou et al 2006). Detailed 82

85 Extension of the phenotype Table 1: Electroneurographical features. Sensory nerve conduction velocity (m/s) Sensory nerve action potential (µv) Distal latency (ms) Sensory nerve Motor nerve conduction velocity (m/s) Proximal compound muscle action potential (mv) Distal compound muscle action potential (mv) Distal latency (ms) Motor nerve Median L Median L Ulnar R Ulnar R Peroneal L Sural R Tibial R H-waves F-waves Nerve M-wave (ms) H-wave (ms) Nerve Muscle Minimal latency (ms) Tibial L Median L Abductor pollicis brevis 29.6 Tibial R Ulnar R Abductor digiti quinti Peroneal L Extensor digitorum brevis Skin temperature was kept at 32 C. Abnormal figures are shown in bold. L, left; R, right. 6 genetic analysis and nerve conduction findings were not reported. Recently a patient with pes cavus, spasticity and a primarily demyelinating neuropathy with variable, focal slowing of motor nerve conduction velocities (NCVs) but relative preservation of motor and sensory 83

86 Chapter 6 6 Figure 1: The p.met1val mutation affects ALDP stability. The level of ALDP in fibroblasts from controls and X-ALD patients was determined by western blot analysis as described previously (Wanders et al., 1995). A monoclonal antibody, directed against an epitope between the amino acids 279 and 482 of the human ALDP was used. Lanes 1 and 4, protein extracts from two different control subjects; lane 2, the index patient; lane 3, protein extract from an X-ALD patient with a p.arg554his mutation that results in no detectable ALDP ( amplitudes was described who was found to have both a PMP22 gene deletion consistent with a hereditary neuropathy with liability to pressure palsies and a c.1166g>a mutation in the ABCD1 gene, causing an amino acid substitution at codon 389 (p.arg389his) (Hodapp et al 2006). Whether polyneuropathy in AMN patients is primarily demyelinating or axonal by nerve conduction studies is controversial. We have reported previously that polyneuropathy in this condition is predominantly axonal on electrophysiological examination (van Geel et al 1996), whereas Chaudhry et al. have reported that demyelination occurs more frequently (Chaudhry et al 1996). This is corroborated by histological studies of nerve biopsies of patients with AMN in whom also either a predominantly demyelinating or axonal neuropathy can be observed even within the same family (Schröder et al 1996). It is important to realize that X-ALD should be considered in the differential diagnosis of polyneuropathy at presentation, whether it is predominantly axonal or demyelinating. References Chaudhry V, Moser HW, Cornblath DR Nerve conduction studies in adrenomyeloneuropathy. J Neurol Neurosurg Psychiatr 61: Hodapp JA, Carter GT, Lipe HP, Michelson SJ, Kraft GH, Bird TD Double trouble in hereditary neuropathy: concomitant mutations in the PMP-22 gene and another gene produce novel phenotypes. Arch Neurol 63: Kararizou E, Karandreas N, Davaki P, Davou R, Vassilopoulos D Polyneuropathies in teenagers: a clinicopathological study of 45 cases. Neuromuscul Disord 16: Ligtenberg MJ, Kemp S, Sarde CO, van Geel BM, Kleijer WJ, Barth PG, Mandel JL, van Oost BA, Bolhuis PA Spectrum of mutations in the gene encoding the adrenoleukodystrophy protein. Am J Hum Genet 56: Loes DJ, Hite S, Moser H, Stillman AE, Shapiro E, Lockman L, Latchaw RE, Krivit W Adrenoleukodystrophy: a scoring method for brain MR observations. AJNR Am J Neuroradiol 15: Moser H, Smith K, Watkins P, Powers J X-linked adrenoleukodystrophy. In: The Metabolic & Molecular Bases of Inherited Disease. Scriver R, Beaudet AL, Sly WS, and Valle D (Eds). McGraw-Hill Inc., New York, pp Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL, Aubourg P Putative X-linked adrenoleukodystrophy gene shares unexpected Homology with ABC transporters. Nature 361: Schröder JM, Mayer M, Weis J Mitochondrial abnormalities and intrafamilial variability of sural nerve biopsy findings in adrenomyeloneuropathy. Acta Neuropathol 92: van Geel BM, Koelman JH, Barth PG, Ongerboer de Visser BW Peripheral nerve abnormalities in adrenomyeloneuropathy: a clinical and electrodiagnostic study. Neurology 46: Wanders RJ, Dekker C, Ofman R, Schutgens RB, Mooijer P Immunoblot analysis of peroxisomal proteins in liver and fibroblasts from patients. J Inherit Metab Dis 18 (Suppl1):

87 Extension of the phenotype Chapter 7 The clinical, biochemical and genetic spectrum of X-linked adrenoleukodystrophy in women: a cross-sectional cohort study 7 Marc Engelen 1,2, Mathieu Barbier 3,4, Inge M.E. Dijkstra 5, Remmelt Schür 2, Rob M.A. de Bie 1, Camiel Verhamme 1, Marcel G.W. Dijkgraaf 6, Patrick A. Aubourg 3,4, Ronald J.A. Wanders 5, Björn M. van Geel 7, Marianne de Visser 1, Bwee Tien Poll The 1,2 and Stephan Kemp 2,5 Departments of 1 Neurology and 2 Pediatric Neurology/Emma Children s Hospital, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 3 Assistance Publique des Hôpitaux de Paris, Department of Pediatric Neurology, Hospital Kremlin-Bicêtre, Paris, France. 4 INSERM UMR745- University Paris-Descartes, Paris, France. 5 Laboratory Genetic Metabolic Diseases and 6 Clinical Epidemiology, Biostatistics & Bio-informatics, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 5 Department of Neurology, Medical Center Alkmaar, Alkmaar, The Netherlands. manuscript in preparation 85

88 Chapter 7 Abstract X-linked adrenoleukodystrophy is a peroxisomal disorder caused by mutations in the ABCD1 gene. It is characterized by impaired very long-chain fatty acid (VLCFA) beta-oxidation and VLCFA accumulation in plasma and tissues. Clinically, the disease has been well characterized in boys and men. However, women can develop symptoms and are not merely carriers. In this cross-sectional cohort study we investigated symptoms and prevalence of these symptoms in women with X-ALD. A secondary goal was to determine if the X-inactivation pattern of the ABCD1 gene was associated with symptomatic status. We found that women with X-ALD develop signs and symptoms of myelopathy (38/46, 70%) and/or peripheral neuropathy (17/46, 37%). Especially striking was the occurrence of fecal incontinence (13/48, 28%). The prevalence of symptomatic women increased sharply with age (from 36% in women under 40 to 88% in women over 60 years of age). Most X-ALD carriers had increased VLCFA in plasma and/or fibroblasts, and/or decreased VLCFA beta-oxidation in fibroblasts. We did not find an association between the X-inactivation pattern and symptomatic status. X-ALD carriers developed an adrenomyeloneuropathy-like phenotype and there is a strong association between symptomatic status and age. Clinicians should consider X-ALD in women with a myelopathy (especially with early fecal incontinence). 7 86

89 Extension of the phenotype Introduction X-linked adrenoleukodystrophy (X-ALD) is a peroxisomal disorder characterized by deficient beta-oxidation of very long-chain fatty acids (VLCFA; C22:0) and accumulation of VLCFA in plasma and tissues (Moser et al 2007). It is caused by mutations in the ABCD1 gene (Mosser et al 1993). In men with X-ALD, disease course and symptomatology have been studied extensively, as reviewed by Moser et al (Moser et al 2007). In adult men with X-ALD, the most frequent phenotype is adrenomyeloneuropathy (AMN), with or without adrenocortical insufficiency (Moser et al 2007). It is characterized by a slowly progressive myelopathy and peripheral neuropathy, manifesting as a spastic paraparesis, sensory disturbance in the lower extremities and incontinence. Onset is typically in the 3 rd decade of life, but can be sooner or much later (Moser et al 2007). However, the phenotype of X-ALD in women has not been systematically investigated in a large prospective study. There have been several studies and observations concerning women with X-ALD, although these are rare compared to the many studies on X-ALD in men. In 1984, 21 obligate carriers of X-ALD were described and several were shown to have neurological symptoms, mostly signs and symptoms of myelopathy (O Neill et al 1984). It was estimated that that roughly 50% of women with X-ALD develop symptoms at some point in their life (Moser et al 1991). In a small study 8 women with X-ALD were clinically characterized and were found to have a myelopathy (Schmidt et al 2001). In a retrospective study from the Netherlands many women were found to show neurological symptoms and the prevalence of symptoms increased with age (van Geel, 2001). It is known that adrenal insufficiency is rare in women with X-ALD (el-deiry et al 1997), and there are only a few reports of cerebral X-ALD in women (Pilz et al 1973; Jung et al 2007). These studies provide no information on the prevalence of neurologic symptoms in the entire population of women with X-ALD, since data was acquired in small groups and may be biased towards women with symptoms. The spectrum in women with X-ALD may be wider than in affected men, ranging from asymptomatic to severely affected. Some studies suggest this could be related to the pattern of X-inactivation, while others can not confirm this correlation (Migeon et al 1981; Watkiss et al 1993; Maier et al 2002; Salsano et al 2012; Naidu et al 1997). 7 To summarize, the studies done so far are prone to selection bias since they favor inclusion of those women with symptoms and might overestimate the true prevalence of neurologic symptoms in women with X-ALD. The primary goal of our study was to determine the proportion of women with X-ALD that have symptoms and to characterize these symptoms both clinically and with ancillary investigations. The secondary goal was to investigate whether symptomatic women differ biochemically from women without a clinical phenotype. Methods Research population The study protocol was approved by the Institutional Review Board of the Academic Medical Center. This study was designed as a prospective cross-sectional cohort study. All female 87

90 Chapter 7 carriers over the age of 18 years were eligible to participate. Women with X-ALD were recruited from the outpatient clinics of the Academic Medical Center (a reference center for peroxisomal disorders in the Netherlands) and Medical Center Alkmaar. To prevent bias for symptomatic women, we tried to examine all female relatives with X-ALD of the women participating. Through the Dutch X-ALD patient organization letters to invite women to participate were sent to all members. After informed consent was obtained, women visited the outpatient clinic for assessment, including questionnaires, neurological examination, neurophysiological tests, blood samples and skin biopsy. 7 Clinical assessment A careful history with emphasis on neurologic complaints and neurological examination was performed. In particular, symptoms of incontinence, gait disorder and maximum walking distance, and sensory disturbance were recorded. Urinary incontinence was defined as urge incontinence. Stress incontinence was not considered to be a symptom of myelopathy in this study. A gait disorder was considered present if the walking distance was significantly reduced or running was not possible. Sensory complaints were recorded if there were paraesthesias or numbness in the lower extremities. Sensory disturbances on examination were considered present if there was reduced sensation to touch or pinprick, vibration- or position sense in the lower extremities. Muscle strength was rated using the MRC scale, and spasticity was rated using the Ashworth scale. An EDSS value (Kurtzke, 1983) was scored based on the information in the medical records. Patients with a peripheral neuropathy or myelopathy were considered symptomatic. A myelopathy was considered to be present if 1) there were symptoms of myelopathy (for instance sphincter disturbances) and 2) signs of myelopathy on neurological examination (pyramidal tract or dorsal column signs) were present. A peripheral neuropathy was considered present if the NCV and/or myography were abnormal (as described below). All participants completed the SF36 (quality of life assessment) and the AMC Linear Disability Scale (ALDS) (Weisscher et al 2007). SF36 values can be compared to reference values for the Dutch population, matched for gender and age. The ALDS values are compared to those of patients with Parkinson s disease from the CARPA study (Post et al 2011). Neurophysiological testing Nerve conduction studies and myography were performed according to a fixed protocol, allowing comparison with reference values and possibly follow-up in the future (Verhamme et al 2009). An axonal peripheral neuropathy was diagnosed if in two separate nerves at least 1 parameter was outside the 95% confidence interval and the criteria for demyelinating peripheral neuropathy were not fulfilled (van Asseldonk et al 2005). Somatosensory evoked potentials (SSEP) of left median and left and right posterior tibial nerve (Aramideh et al 1992; Aalfs et al 1993) and brainstem auditory evoked potentials (BAEP) on the left and right side were registered, conforming to local protocols. Blood samples Venous blood samples were taken and plasma and lymphocytes were isolated and stored as described previously (Engelen et al 2010). VLCFA were determined in these samples as described previously (Valianpour et al 2003). 88

91 Extension of the phenotype Cell culture and biochemical analysis From skin biopsy material primary fibroblast cell lines were isolated. Cell lines were cultured in HAMF10 culture medium supplemented with 10% fetal calf serum. For all subjects for whom material was available VLCFA were determined in plasma and fibroblasts as described previously (Valianpour et al 2003). Peroxisomal beta-oxidation activity was measured as described previously by incubating cells with deuterium labeled C22:0 (Kemp et al 2004). ALDP levels were determined by immunofluorescence (IF) as well as by quantitative immunoblot (Kemp et al 1996; Zhang et al 2011). Genetic analysis and ABCD1 allelic specific expression in fibroblasts Mutation analysis of the ABCD1 gene was performed as described previously (Boehm et al 1999). The ABCD1 allelic specific expression (ABCD1 ASE) of 38 carriers was measured by pyrosequencing. The RNA from primary fibroblasts was extracted. The cdna synthesis has been described elsewhere (Bieche at al 2001). The ratio of wild-type versus mutant allele expressed was measured by pyrosequencing. We PCR-amplified the cdna using biotin-labeled primers and performed quantitative pyrosequencing using a PyroMark Q96 ID instrument (Qiagen, Venlo, The Netherlands). Biotin-labeled single stranded amplicons were isolated according to protocol using the Qiagen Pyromark Q96 Work Station and underwent pyrosequencing with sequencing primer. The list of PCR- and sequencingprimers is indicated in supplemental table S1 (online only). Each sample was pyrosequenced in upper direction (with lower biotin-labeled amplicons) and in lower direction (with upper biotin-labeled amplicons). The ratio of wild-type and mutant allele was measured by the Pyro Mark ID software (Qiagen). For each carrier, the ABCD1 ASE value represents the mean of upper and lower reads (two independent experiments). For included samples, variations between upper and lower reads were 10%. For a subset of samples (n=6), we repeated the experiments using different PCR-primers or different cdnas. The ABCD1 ASE values were similar (variations 4%). For five patients it was not technically possible to determine the ABCD1 ASE with this technique due to complex mutational events. For these five cases, values were calculated using ALDP immunofluorescence and determining the ratio of ALDP positive and negative cells with IF-microscopy. We defined the ratio of ABCD1 allele expressed as: severely skewed when the ratio of normal or mutated allele was between 0% and 10%; moderately skewed when the ratio of normal or mutated allele was between 11% and 25%; randomly when the ratio of normal or mutated allele was between 26% and 74%, similar to a previous report (Miozzo et al 2007). In the manuscript ABCD1 ASE will be referred to as X-inactivation. 7 Data entry and statistical analysis Data were entered in and analyzed with IBM SPSS statistics version 19 (IBM Inc, New York, U.S.A.) and Prism version 5 (GraphPad Software, La Jolla, U.S.A.). The relationship between symptomatic status, age and X-inactivation pattern (ABCD1 ASE) was analyzed by logistic regression. Results Demographic and clinical characteristics A total of 46 women from 26 kindreds were enrolled in the study. Of these participants 89

92 Chapter 7 Table 1: Summary of demographic, clinical and genetic data of all the women participating in the study. 7 Family Age U Inc F Inc Gait Sens C Sens D Sp Par P Ref EDSS Mutation ALDP A 56 Yes Yes No No No No No Yes 1.5 p.pro480thr absent A 44 No No Yes No No No No Yes 1.0 p.pro480thr absent AA 45 No No No No No No No Yes 0 p.arg660trp absent AA 59 Yes No Yes No No No Yes Yes 3.5 p.arg660trp absent AA 75 Yes No Yes No Yes Yes Yes Yes 6.0 p.arg660trp absent B 42 Yes Yes Yes No Yes Yes Yes Yes 4.0 p.leu220pro reduced B 44 No No No No No No No No 0 p.leu220pro reduced B 44 No No No No No No No No 0 p.leu220pro reduced B 51 No No No Yes Yes No No Yes 1.0 p.leu220pro reduced B 59 No No No Yes Yes No Yes No 2.0 p.leu220pro reduced C 44 No No No No No No No No 0 p.gln133* absent D 38 Yes Yes Yes No Yes Yes Yes Yes 6.0 p.leu654pro absent D 57 Yes No Yes Yes Yes No No Yes 5.5 p.leu654pro absent E 31 No No No No No No No No 0 p.arg74trp absent E 37 No No No No No No No No 0 p.arg74trp absent E 60 No No Yes No Yes Yes Yes Yes 5.5 p.arg74trp absent F 35 No No No No No No No No 0 p.met1val absent G 42 No Yes No No No No No No 1.0 p.ala245asp present H 61 Yes Yes Yes Yes Yes No No Yes 3.5 exon8-10del absent I 71 No No No No Yes No No Yes 2.0 p.glu609lys absent J 42 No No No No Yes No No Yes 1.5 p.glu90* absent K 31 No No No No No No No No 0 p.pro543leu absent K 48 Yes No No No Yes No No Yes 2.5 p.pro543leu absent Age: age at examination, U Inc: urinary incontinence, F Inc: fecal incontinence, Gait: gait disorder, Sens C: sensory complaints, Sens D: sensory symptoms on neurological examinations, Sp: spasticity, Par: paresis of the lower extremities, P Ref: pathological reflexes (lower extremities), EDSS: expanded disability status scale, Mutation: mutation in ABCD1, ALDP: effect of mutation in ABCD1 on ALDP as determined by Western blot. & mutation inferred from mutation identified in father and uncle. 90

93 Extension of the phenotype Family Age U Inc F Inc Gait Sens C Sens D Sp Par P Ref EDSS Mutation ALDP K 57 No No Yes Yes Yes No Yes Yes 3.5 p.pro543leu absent K 60 Yes No No No Yes No No Yes 3.5 p.pro543leu absent L 51 Yes No Yes No Yes Yes Yes Yes 6.5 p.ile657del absent M 22 No No No No No No No No 0 p.ser149asn reduced M 40 No No No No No No No No 0 p.ser149asn reduced N 29 No No No No No No No No 0 p.arg389his reduced N 45 Yes No Yes Yes No No No No 2.0 p.arg389his reduced N 57 Yes Yes Yes Yes Yes No No No 3.5 p.arg389his reduced N 70 No No Yes No Yes No Yes Yes 3.5 p.arg389his reduced O 40 Yes Yes Yes Yes Yes No No Yes 3.5 p.glu609lys absent P 59 Yes Yes Yes Yes Yes Yes Yes Yes 6.0 p.leu215* absent Q 39 No Yes Yes No Yes No No No 3.0 p.val208trpfs absent R 28 No No No No No No No No 0 p.pro480thr absent S 35 No No No No No No No No 0 p.his283tyr reduced S 76 Yes No Yes No Yes No No Yes 2.0 p.his283tyr reduced T 51 Yes Yes Yes No Yes No Yes Yes 4.0 p.gln177* absent U 47 Yes Yes No No Yes No No No 2.0 p.arg464* absent V 56 Yes No Yes No Yes No Yes Yes 2.5 p.asp442glyfs absent W 45 Yes Yes Yes Yes Yes No No Yes 3.5 p.ala616thr absent W 65 No Yes No No No No No No 2.0 p.ala616thr absent X 47 Yes No Yes No No No No Yes 2.5 p.arg113alafs absent Y 24 No No No No No No No No 0 p.glu609gly & absent & Z 50 No No No No Yes No No Yes 2.0 p.ser633argfs absent 7 91

94 Chapter 7 Figure 1: Age distribution of the cohort were recruited from the outpatient clinic, 34 were female relatives of these women or responded to the letter sent to members of the patient organization. Complete clinical and electrophysiological data are available from 46 women, blood samples are available from 45 women and skin biopsies were obtained from 43 women (one obligate carrier refused venapuncture, two refused skin biopsy, one skin biopsy failed due to a yeast infection). The age of the women included in the study ranged from 22 to 76 years (average 48 ± 13 years). The age distribution is shown in Figure 1. Mutation analysis was done in 45 women (as for one obligate carrier no leukocytes were available), the mutation in that subject was considered to be the mutation found in her father and paternal uncle. Complaints of incontinence (both urinary and fecal), abnormal gait, and sensory symptoms were common in the group studied, as listed in Table 1. Data is summarized in Table 2. There was a myelopathy (as defined in the methods section) in 32/46 (70%) of the women. Of the women in the cohort 34/46 (74%) were considered symptomatic because of either a myelopathy and/or a peripheral neuropathy. Demographic and clinical data is shown in Figure 1 and Table 1. Women who have signs and symptoms suggestive of myelopathy increase in frequency with age (Table 2). The percentage of women that are symptomatic increases from 4/11 (36%) in the youngest age group, and increases to 30/35 (86%) in the older age groups. Logistic regression shows a strong relationship between age and symptomatic status, with the probability of being symptomatic increasing strongly with age (Wald 9.620, p = 0.002). The SF-36 physical and mental compound scores and the ALDS score are reported for the entire cohort by age group (Table 2). Electrophysiological testing Nerve conduction studies and myography Of all participants NCV, EMG and evoked potentials (BAEP and SSEP) were available. Of the women 17/46 (37%) had a sensorimotor axonal neuropathy according to local electrophysiological criteria (Table 3). 92

95 Extension of the phenotype Table 2: Symptoms and signs by age group yrs yrs > 60 yrs all ages Incontinence (urine) 1/11 (9) 15/27 (56) 5/8 (63) 21/46 (46) Incontinence (fecal) 2/11 (18) 9/27 (33) 2/8 (25) 13/46 (28) Gait disorder 2/11 (18) 15/27 (56) 5/8 (63) 22/46 (48) Sensory complaints 1/11 (9) 12/27 (44) 2/8 (25) 15/46 (33) Sensory disturbance 2/11 (18) 12/27 (44) 6/8 (75) 23/26 (50) Spasticity 1/11 (9) 3/27 (11) 1/8 (13) 5/46 (11) Weakness 1/11 (9) 7/27 (26) 2/8 (25) 10/46 (22) Pathological reflexes 2/11 (18) 19/27 (70) 6/8 (75) 27/46 (59) Abnormal history 3/11 (27) 21/27 (78) 6/8 (75) 30/46 (65) Abnormal examination 4/11 (36) 22/27 (82) 6/8 (75) 32/46 (70) Symptomatic 4/11 (36) 23/27 (85) 7/8 (88) 34/46 (74) EDSS 0.82 (1.94) 2.41 (1.83) 3.50 (1.56) 2.22 (1.99) SF-36 physical 50.0 (10.6) 44.0 (15.1) 44.4 (13.4) 45.3 (13.7) SF-36 mental 51.4 (4.3) 51.5 (8.1) 51.1 (5.8) 51.4 (6.87) ALDS 89.1 (0.7) 87.4 (4.2) 84.4 (7.0) 87.4 (4.4) Symptoms and signs (reported as absolute number (%)), EDSS, SF-36 and ALDS scores (reported as mean ± SD) for the entire cohort, and stratified by age group. SF-36 physical: SF-36 physical compound score, SF-36 mental: SF-36 mental compound score. 7 Somatosensory evoked potentials (median nerve and posterior tibial nerve) The SSEP from the median nerve was abnormal in 3/42 (6.5%) of cases (4 registrations failed because of technical reasons), of the tibial nerve 14/44 (30.4%) was abnormal (2 registrations failed). Of the female carriers who were asymptomatic on neurological examination (11/46), the median nerve SSEP was normal in all cases and the SSEP of the posterior tibial nerve was abnormal in only 1/11 (9%). Of the women who were symptomatic on neurological examination (35/46), 3/38 (8.6%) had an abnormal median nerve SSEP and 13/35 (37.1%) had an abnormal tibial nerve SSEP. The average latency of the cortical peak increases with age for median nerve and tibial nerve SSEP. The percentage of abnormal SSEP increases with age. The findings from the SSEP from median and tibial nerve for the complete cohort are reported in Table 4. Brainstem auditory evoked potentials In the entire cohort BAEP measurements were done in 45 women. The measurement failed for technical reasons in one participant. BAEP results are shown in Table 4. For the entire group 26/46 (41.3%) women had an abnormal BAEP, mostly consisting of an increase I V and I III interval. There were no significant left right differences. Of the women who were symptomatic, 23/35 (65.7%) had an abnormal BAEP result. In the asymptomatic group this proportion is 3/7 (27.3%). 93

96 Chapter 7 Table 3: Summary of nerve conduction studies. Mean (SD) Reference interval Median nerve dlt (ms) 3.6 (0.8) CMAP (mv) 6.2 (2.9) MNCV (m/s) 58.4 (11.6) Posterior tibial nerve dlt (ms) 5.1 (1.0) CMAP (mv) 10.0 (5.5) MNCV (m/s) 41.6 (6.4) Sural nerve dlt (ms) 3.9 (0.5) SNAP (uv) 7.2 (4.9) Summated CMAP 19.8 (8.4) Summated SNAP 39.8 (18.1) Table 4: Results of somatosensory evoked potentials (median and tibial nerve) and brainstem auditory evoked potentials yrs yrs > 60 yrs all ages SSEP arm N9 (ms) 9.50 (0.69) (0.93) 10.4 (0.85) 9.99 (0.90) N13 (ms) (0.82) (1.40) (0.97) (1.29) N20 (ms) (1.12) (1.50) (1.22) (1.44) abnormal 0/10 (0) 3/26 (11.5) 0/6 (0) 3/42 (6.5) SSEP leg N35 Left (ms) (4.44) (10.32) (5.00) (8.56) P37 Left (ms) (5.25) (7.19) (5.28) (6.78) N35 Right (ms) (4.59) (11.75) (3.08) (9.52) P37 Right (ms) (4.19) (9.78) (4.49) (8.29) abnormal 1/10 (10.0) 11/27 (40.7) 2/7 (28.6) 14/44 (30.4) BAEP I III Left (ms) 2.10 (0.25) 2.31 (0.32) 2.24 (0.24) 2.25 (0.30) I V Left (ms) 4.16 (0.43) 4.38 (0.47) 4.19 (0.23) 4.29 (0.44) I III Right (ms) 2.29 (0.21) 2.38 (0.20) 2.24 (0.14) 2.33 (0.21) I V Right (ms) 4.23 (0.33) 4.46 (0.42) 4.23 (0.32) 4.36 (0.40) abnormal 4/11 (36.4) 17/26 (65.4) 5/8 (62.5) 26/45 (57.8) 94

97 Extension of the phenotype Table 5: Results of plasma and fibroblast studies. Plasma Mean (SD) Reference interval C26:0 (µmol/l) 2.26 (0.69) C26:0/C22:0 ratio (0.012) Fibroblasts C26:0 (nmol/mg protein) 0.75 (0.32) C26:0/C22:0 ratio 0.22 (0.10) D 3 -C16:0/D 3 -C22:0 ratio 0.73 (0.41) Biochemical and genetic analysis Plasma and leukocytes For 45/46 participant blood samples were obtained and VLCFA levels in plasma were determined. The data are summarized in Table 5. VLCFA were clearly increased with an average of 2.26 µmol/l ± 0.69 µmol/l (reference values µmol/l). 31/46 women (68.9%) had abnormal plasma VLCFA levels. VLCFA levels, ALDP levels and beta-oxidation activity in fibroblasts VLCFA levels, VLCFA beta-oxidation in skin fibroblasts, ABCD1 mutation analysis, fibroblast ALDP levels (as determined by immunoblot) and fibroblast ALDP immunofluorescence were determined if material was available. In 37/43 (86%) C26:0 levels in fibroblasts were increased, and 36/43 the C26:0/C22:0 (84%) ratio was abnormal. In 26/43 (60%) the D 3 - C16:0/D 3 -C22:0 was decreased, indicating reduced peroxisomal VLCFA beta-oxidation. Residual levels of ALDP were determined by immunoblot and correlate with residual VLCFA beta-oxidation and C26:0 levels, i.e. the more residual ALDP the higher the VLCFA betaoxidation and the lower fibroblast C26:0 levels. Results are summarized in Table 5 and Figure 2. 7 ABCD1 allelic specific expression (ABCD1 ASE) studies in fibroblasts and correlation of X-inactivation with symptomatic or asymptomatic status. Data for ABCD1 ASE is shown in Figure 3. It is randomly distributed and correlates well with beta-oxidation activity and ALDP expression in cultured fibroblasts (Figure 4). This suggests the methodology used to determine ABCD1 ASE is valid. To correlate ABCD1 ASE with symptomatic status it is important to correct for age, since there is a strong correlation between age and symptomatic status (Table 2). There seems to be no correlation between symptomatic status and the ABCD1 ASE pattern. Discussion Women with X-ALD can develop neurological symptoms. In fact, one of the earliest descriptions of an AMN-like phenotype in X-ALD concerns a female patient (Penman, 1960). 95

98 Chapter 7 7 Figure 2: ALDP levels as determined by Western blot correlate with the residual peroxisomal beta-oxidation activity (A), C26:0 synthesis (B), C26:0 levels (C) and the C26:0/C22:0 ratio (D) in skin fibroblasts from female carriers. Fatty acids are in nmol/mg protein. This is, however, not so obvious for most physicians (Jangouk et al 2011). We know of several cases in which women with X-ALD underwent cervical laminectomy for suspected cervical spondylogenic myelopathy (van Geel et al 2007). We found in the largest prospective cohort of female carriers to date that neurological abnormalities are found in approximately 70% of female carriers and that the frequency increases with age. The main symptoms are consistent with a myelopathy, as in men with X-ALD and the AMN phenotype. It is striking how often fecal incontinence is reported, often as an early symptom. One should bear in mind that it is often not voluntarily reported because it is felt to be embarrassing. To our knowledge, this is not reported often in men 96

99 Extension of the phenotype Figure 3: Distribution of ABCD1 allele specific expression in skin fibroblasts from female carriers. with AMN or in other degenerative myelopathies, for instance hereditary spastic paraplegia (Salinas et al 2008). In fact, in our cohort sphincter disturbance is a frequent and early symptom, and even occurs in a few symptomatic carriers without other signs of myelopathy. Signs of peripheral neuropathy were mostly absent or minor clinically, and do not seem to contribute much to the overall disability. However, on electrophysiological testing, though, about 37% of the women in this cohort fulfilled the criteria for an axonal sensorimotor neuropathy. Signs and symptoms present in our cohort caused reduced quality of life. Comparing the SF36 scores to the Dutch norm population, symptomatic women show a reduced quality of life in all domains. The compound physical score was lower than that of age matched controls, corresponding to the signs of myelopathy frequently present in our cohort. The compound psychological score was reduced even more strikingly. Possibly this is related not only to the distress caused by the physical disability also by the fact that many women with X-ALD suffered the loss of children to X-ALD in their families. Furthermore, not only was quality of life reduced, analysis of the ALDS scale revealed that women with X-ALD in the symptomatic category had the same levels of disability as patients with Parkinson s disease three years after the diagnosis. The group of patients with Parkinson s disease is significantly older (average age 67 versus 48 years) than our cohort but still levels of disability were similar, suggesting that the burden of disease in women with X-ALD is considerable. 7 SSEP was reported to be very sensitive in detecting abnormalities in women with X-ALD in a presymptomatic stage (Restuccia et al 1997). In our cohort SSEP of the arms (median nerve) were normal in all asymptomatic women, and were abnormal in only 9% of the symptomatic women. This is not surprising, since in none of the symptomatic women were the arms affected clinically. SSEP of the legs (posterior tibial nerve) were only abnormal in 27% of the asymptomatic women and in 67% of the symptomatic women. Therefore, we conclude that SSEP is not superior to clinical examination in detecting myelopathy in our cohort. BAEP were abnormal in a large proportion of all women in our cohort, as described previously (Restuccia et al 1997). A larger proportion of the symptomatic women has an abnormal BAEP test result (66% versus 27%). The findings were aspecific, mostly consisting of an increased 97

100 Chapter

101 Extension of the phenotype I V or III V interval. This indicates a slowing of conduction in the brainstem. For clinical practice BAEP has no value as a diagnostic tool. As has been known for several decades, 15 20% of heterozygotes have normal plasma VLCFA (Moser et al 1999). In our cohort we found that a slightly higher proportion (30%) had normal VLCFA. Furthermore, even in cultured fibroblasts the VLCFA levels can be normal (Moser et al 1983). Combining VLCFA test result in plasma and fibroblasts marks a higher proportion of the women in our cohort as biochemically abnormal. In this cohort, by combining VLCFA levels in plasma and fibroblasts 41 of 43 women (95.3%) for whom plasma and fibroblasts were available could be identified. This corresponds to an earlier observation that by combining VLCFA measurements in plasma and fibroblasts 93% of carriers can be identified. We also performed a beta-oxidation assay in cultured fibroblasts. For the VLCFA beta-oxidation results in fibroblasts vary from completely normal to levels as low as those in men with X-ALD. By adding this parameter, 44 of 45 women with X-ALD (97.8%) for whom plasma and/or fibroblasts were available could be identified. However, this still does not reliably identify all women with X-ALD. In women the diagnostic test of choice is ABCD1 mutation analysis. We show that the biochemical abnormality in fibroblasts correlates with the X-inactivation pattern, i.e. the more skewed to the mutant allele, the more abnormal the fibroblasts are biochemically. Our study shows that in fibroblasts, ABCD1 allele specific expression (ABCD1 ASE) predicts the biochemical phenotype. This validates our methodology. X-inactivation was considered to be skewed preferentially to the mutant allele (Migeon et al 1981). That finding cannot be reproduced in our data set (Figure 4). In fact, the median skewing is 49.3, suggesting a random ABCD1 ASE pattern. For over 2 decades there has been controversy over whether X-inactivation predicts the symptomatic status of women with X-ALD. In 1993 it was reported that symptomatic status and X-inactivation in fibroblasts do not correlate in a small sample of 12 women (Watkiss et al 1993). A more recent report suggests that skewed X-inactivation patterns in leukocytes do correlate with symptoms in women with X-ALD (Maier et al 2002). However, the assay used in that study (HUMARA) has limitations. It is now known that X-inactivation is not uniform through the entire X-chromosome, but can differ from locus to locus (Carrel et al 2005). The most recent study compares a group of symptomatic carriers and asymptomatic carriers and cannot find an association between X-inactivation in fibroblasts and symptomatic status (Salsano et al 2012). It should be noted that there is an average age difference of more than 15 years between the groups of symptomatic and asymptomatic women with X-ALD, making a comparison difficult considering that our data show that age is a strong predictor for symptomatic status in women with X-ALD. In our study we can not establish a link between X-inactivation as determined by ABCD1 ASE) and symptomatic status when taking age into account. This still does not exclude X-inactivation as an important modifier. A major weakness, as in the other studies, is that non-neural tissue was used to obtain material for X-inactivation studies. Also, even though this is the largest cohort of women with X-ALD studied systematically so far, 7 Figure 4: ABCD1 ASE correlates with ALDP levels as determined with IF (A) or Western blot (B), the residual peroxisomal beta-oxidation activity (C), C26:0 synthesis (D), C26:0 levels (E) and the C26:0/ C22:0 (F) ratio in skin fibroblasts from female carriers. 99

102 Chapter 7 the number of participants might be too small to detect the relationship. In future studies, we will keep expanding the cohort and possibly extend X-inactivation studies to different tissues. In this cohort imaging of the brain was not performed routinely since brain involvement is rare and this has been studied by others (Fatemi et al 2003). MRI of the spinal cord of men with AMN and symptomatic women with X-ALD shows abnormalities on magnetization transfer imaging (Dubey et al 2005). This is compatible with the clinical findings in our cohort. An MRI of the brain and spinal cord was only performed in three women to exclude other disorders, and was normal in each case. 7 A limitation of our study is that not all women with X-ALD in the Netherlands were examined. Even though we tried to minimize bias by including not only women from the outpatient clinic (around 75% of the participants were recruited through the X-ALD patient organization), it is still possible that the participants are not a totally random sample of women with X-ALD. Based on the prevalence it can be estimated there are at least 200 women with X-ALD in the Netherlands. Our population therefore probably contains no more than 25 to 30% of all women with X-ALD in the Netherlands. It is therefore possible that those without symptoms were less likely to participate. Although there was no cohort of age matched controls, it is unlikely that the symptoms described are normal aging. Myelopathy and peripheral neuropathy can not be considered a part of healthy aging. Co-morbidity, like diabetes, was rare in the cohort. In summary, women with X-ALD are highly likely to develop symptoms during their lifetime. The most important predictor is age, with virtually all women with X-ALD having some clinical manifestation if older than 60. Clinical manifestations are mostly related to myelopathy. Especially striking is the high incidence of fecal incontinence. Peripheral neuropathy is not prominent clinically, although based on nerve conduction studies 37% has a sensorimotor axonal peripheral neuropathy. Biochemical abnormalities are common, but cannot exclude X-ALD with certainty in women. The biochemical abnormalities in fibroblasts in women with X-ALD are clearly related to the X-inactivation pattern. We were not able to show a link between X-inactivation and symptomatic status, but this may be related to the limitations stated above. Women with X-ALD should not be considered merely carriers, but they constitute a distinct AMN-like phenotype of X-ALD. Physicians should consider the diagnosis in women with a chronic myelopathy and ABCD1 mutation analysis should have a place in diagnostic protocols for chronic non-compressive myelopathy. References Aalfs CM, Koelman JH, Aramideh M, Bour LJ, Bruyn RP, Ongerboer de Visser BW Posterior tibial nerve somatosensory evoked potentials in slowly progressive spastic paraplegia: a comparative study with clinical signs. J Neurol 240(6): Aramideh M, Hoogendijk JE, Aalfs CM, Posthumus Meyjes FE, de VM, Ongerboer de Visser BW Somatosensory evoked potentials, sensory nerve potentials and sensory nerve conduction in hereditary motor and sensory neuropathy type I. J Neurol 239(5):

103 Extension of the phenotype Van Asseldonk JT, Van den Berg LH, Kalmijn S, Wokke JH, Franssen H Criteria for demyelination based on the maximum slowing due to axonal degeneration, determined after warming in water at 37 degrees C: diagnostic yield in chronic inflammatory demyelinating polyneuropathy. Brain 128(Pt 4): Bieche I, Parfait B, Tozlu S, Lidereau R, Vidaud M Quantitation of androgen receptor gene expression in sporadic breast tumors by real-time RT-PCR: evidence that MYC is an AR-regulated gene. Carcinogenesis 22(9): Boehm CD, Cutting GR, Lachtermacher MB, Moser HW, Chong SS Accurate DNA-based diagnostic and carrier testing for X-linked adrenoleukodystrophy. Mol Genet Metab 66(2): Carrel L, Willard HF X-inactivation profile reveals extensive variability in X-linked gene expression in females. Nature 17; 434(7031): Dubey P, Fatemi A, Huang H, Nagae-Poetscher L, Wakana S, Barker PB, van Zijl P, Moser HW, Mori S, Raymond GV Diffusion tensor-based imaging reveals occult abnormalities in adrenomyeloneuropathy. Ann Neurol 58(5): Engelen M, Ofman R, Dijkgraaf MG, Hijzen M, van der Wardt LA, van Geel BM, de Visser M, Wanders RJ, Poll The BT, Kemp S Lovastatin in X-linked adrenoleukodystrophy. N Engl J Med 362(3): el-deiry SS, Naidu S, Blevins LS, Ladenson PW Assessment of adrenal function in women heterozygous for adrenoleukodystrophy. J Clin Endocrinol Metab 82(3): Fatemi A, Barker PB, Ulug AM, Nagae-Poetscher LM, Beauchamp NJ, Moser AB et al MRI and proton MRSI in women heterozygous for X-linked adrenoleukodystrophy. Neurology 22;60(8): Jangouk P, Zackowski KM, Naidu S, Raymond GV. Adrenoleukodystrophy in female heterozygotes: underrecognized and undertreated. Mol Genet Metab 2012 February;105(2): Jung HH, Wimplinger I, Jung S, Landau K, Gal A, Heppner FL Phenotypes of female adrenoleukodystrophy. Neurology 20;68(12): Kemp S, Mooyer PA, Bolhuis PA, van Geel BM, Mandel JL, Barth PG et al ALDP expression in fibroblasts of patients with X-linked adrenoleukodystrophy. J Inherit Metab Dis 19(5): Kemp S, Valianpour F, Mooyer PA, Kulik W, Wanders RJ Method for Measurement of Peroxisomal Very-Long- Chain Fatty Acid beta-oxidation in Human Skin Fibroblasts Using Stable-Isotope-Labeled Tetracosanoic Acid. Clin Chem 50(10): Kurtzke JF Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 33(11): Maier EM, Kammerer S, Muntau AC, Wichers M, Braun A, Roscher AA Symptoms in carriers of adrenoleukodystrophy relate to skewed X inactivation. Ann Neurol 52(5): Migeon BR, Moser HW, Moser AB, Axelman J, Sillence D, Norum RA Adrenoleukodystrophy: evidence for X linkage, inactivation, and selection favoring the mutant allele in heterozygous cells. Proc Natl Acad Sci U S A 78(8): Miozzo M, Selmi C, Gentilin B, Grati FR, Sirchia S, Oertelt S et al Preferential X chromosome loss but random inactivation characterize primary biliary cirrhosis. Hepatology 46(2): Moser AB, Kreiter N, Bezman L, Lu S, Raymond GV, Naidu S et al Plasma very long chain fatty acids in 3,

104 Chapter 7 peroxisome disease patients and 29,000 controls. Ann Neurol 45(1): Moser HW, Moser AE, Trojak JE, Supplee SW Identification of female carriers of adrenoleukodystrophy. J Pediatr 103(1):54-9. Moser HW, Moser AB, Naidu S, Bergin A Clinical aspects of adrenoleukodystrophy and adrenomyeloneuropathy. Dev Neurosci 13(4-5): Moser HW, Mahmood A, Raymond GV X-linked adrenoleukodystrophy. Nat Clin Pract Neurol 3(3): Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H et al. Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 1993 February 25;361(6414): Naidu S, Washington C, Thirumalai S, Smith KD, Moser HW, Watkins PA. X-chromosome inactivation in symptomatic heterozygotes of X-linked adrenoleukodystrophy Ann Neurol 42:498a. O Neill BP, Moser HW, Saxena KM, Marmion LC. Adrenoleukodystrophy: clinical and biochemical manifestations in carriers. Neurology 1984 June;34(6): Pilz P, Schiener P. Kombination von Morbus Addison und Morbus Schilder bei einer 43 jahrigen Frau. Acta Neuropathol (Berl) 1973;26: Penman RW. Addison s disease in association with spastic paraplegia. Br Med J 1960 February 6;1(5170): Post B, Muslimovic D, van GN, Speelman JD, Schmand B, de Haan RJ. Progression and prognostic factors of motor impairment, disability and quality of life in newly diagnosed Parkinson s disease Mov Disord 26(3): Restuccia D, Di Lazarro V, Valeriani M, Oliviero A, Le Pera D, Colosimo C, Burdi N, Cappa M, Bertini E, Di Biase A, Tonali P Brain 120 (pt 7): Salinas S, Proukakis C, Crosby A, Warner TT. Hereditary spastic paraplegia: clinical features and pathogenetic mechanisms Lancet Neurol 7(12): Salsano E, Tabano S, Sirchia SM, Colapietro P, Castellotti B, Gellera C et al. Preferential expression of mutant ABCD1 allele is common in adrenoleukodystrophy female carriers but unrelated to clinical symptoms Orphanet J Rare Dis January 26;7(1):10. Schmidt S, Traber F, Block W, Keller E, Pohl C, von OJ et al Phenotype assignment in symptomatic female carriers of X-linked adrenoleukodystrophy. J Neurol 248(1): Valianpour F, Selhorst JJ, van Lint LE, van Gennip AH, Wanders RJ, Kemp S Analysis of very long-chain fatty acids using electrospray ionization mass spectrometry. Mol Genet Metab 79(3): van Geel BM, Assies J, Wanders RJ, Barth PG X-Linked adrenoleukodystrophy: clinical presentation, diagnosis, and therapy. J Neurol Neurosurg Psychiatry 63(1): van Geel BM Carrier state of X-linked adrenoleukodystrophy. Ned Tijdschr Geneesk 144(37): Verhamme C, van Schaik IN, Koelman JH, de Haan RJ, de VM The natural history of Charcot-Marie-Tooth type 1A in adults: a 5-year follow-up study. Brain 132(Pt 12): Watkiss E, Webb T, Bundey S Is skewed X inactivation responsible for symptoms in female carriers for adrenoleucodystrophy? J Med Genet 30(8):

105 Extension of the phenotype Weisscher N, Post B, de Haan RJ, Glas CA, Speelman JD, Vermeulen M The AMC Linear Disability Score in patients with newly diagnosed Parkinson disease. Neurology 69(23): Zhang X, De Marcos LC, Schutte-Lensink N, Ofman R, Wanders RJ, Baldwin SA et al Conservation of targeting but divergence in function and quality control of peroxisomal ABC transporters: an analysis using cross-kingdom expression. Biochem J 436(3):

106 Chapter

107 Summary and General Discussion Chapter 8 Summary, general discussion, future research and implications for clinical practice 8 105

108 Chapter 8 Summary of the thesis X-linked adrenoleukodystrophy (X-ALD) is a relatively common metabolic disorder (Moser et al 2007). It causes severe disability in the majority of affected men and women. For most patients there is currently no curative treatment. We review current recommendations for management in Chapter 1. The biochemistry of X-ALD has been studied extensively. Although there are still unresolved issues, there is a reasonable understanding of the molecular mechanisms of this disease at the cellular level, as reviewed by Kemp and Wanders (Kemp and Wanders 2010). This, and an excellent in vitro model system available for research (i.e. cultured primary skin fibroblasts) offers possibilities to screen for compounds that might be of therapeutic use in this disorder. In Chapters 2 and 3 we describe the use of this in vitro model to test potential very long-chain fatty acid (VLCFA) lowering compounds. 8 It is important to stress that the hypotheses generated by in vitro experiments always need to be validated in clinical trials, maybe first with biochemical endpoints, but eventually with relevant clinical endpoints. Patients with X-ALD are often desperate for a cure and willing to take medication based on pathophysiological theories or inconclusive and poorly designed clinical trials. If one claims a compound might be effective based on in vitro data there is a moral obligation to validate these claims in a clinical trial. We show that even if compounds or strategies seem promising in the laboratory (and even in laboratory animals), they can be totally ineffective in a clinical trial. We performed two clinical trials to test if lovastatin (Chapter 4) or bezafibrate (Chapter 5) can lower VLCFA in X-ALD. Lovastatin was widely believed to lower VLCFA, but we show that the VLCFA lowering in plasma by lovastatin is an artefact. Bezafibrate is also ineffective, probably because plasma concentrations required for a VLCFA lowering effect can not be reached even with the maximum tolerable dose. Because a biochemical effect was absent further studies with clinical endpoints seem unnecessary. As with all disorders, the clinical spectrum often expands with time as more atypical cases are diagnosed and described. In Chapter 6 we describe a man with X-ALD presenting with signs and symptoms of a demyelinating peripheral neuropathy. Usually, the peripheral neuropathy in X-ALD is not prominent and classified as an axonal sensomotor peripheral neuropathy (van Geel et al 1997). We conclude that in demyelinating peripheral neuropathy without a specific cause X-ALD can be considered. In Chapter 7 we describe a large cohort of women with X-ALD. It is clear that women with X-ALD are not merely carriers but develop signs and symptoms of myelopathy and peripheral neuropathy. Symptomatic status correlates with age, after 60 years of age most women can be considered symptomatic. There is a striking phenotypic heterogeneity; we could not find an association between the X-inactivation pattern in fibroblasts and symptomatic status. As we discuss in the paper, however, it is possible that expression in other (neural) tissues does correlate. These tissues are not easily accessible and so have not been studied by us. General discussion and future research Research to find the cause and possibly a cure for rare diseases is sometimes considered less relevant on the basis of the relatively small amount of people that would benefit. However, studying these disorders can sometimes lead to new insights in human physiology that 106

109 Summary and General Discussion might be applicable to other disorders. Also, even though the number of affected individuals is small, the burden of disease is not for those affected and their families. We will continue with our efforts to improve the outcome in this devastating disease. Unfortunately, the compound we identified in vitro as a VLCFA lowering drug, is not effective in a clinical trial. The concept, however, of decreasing VLCFA synthesis pharmacologically remains a plausible strategy for future studies. Currently, we are trying to identify inhibitors of ELOVL1 (the rate limiting enzyme involved in VLCFA synthesis) in vitro. We hope this will yield new drugs to stabilize the disease in the future. Clinical trials in X-ALD to test clinical efficacy are difficult. The disease course is highly variable, age of onset and progression differ enormously between patients. To determine if a new drug offers clinical benefit trials with large groups of patients are necessary. This is a considerable challenge in a rare disease. Furthermore, long follow-up of at least 3 years is necessary because the disease is usually gradually progressive. Effects on surrogate endpoints (like reduction of plasma VLCFA) should not be considered proof of clinical efficacy. Lorenzo s oil normalizes VLCFA, but does not prevent progression of disease in X-ALD (Aubourg et al 1993; van Geel et al 1999). Still, clinical trials with biochemical endpoints can be useful as pilot studies. Because trials with clinical endpoint take enormous effort potential new drugs can first be evaluated in small clinical trials with biochemical endpoint. If there is an effect on the surrogate endpoints a follow-up trial with relevant clinical endpoints can be performed. It is important to stress that plasma VLCFA can be misleading. In Chapter 4 we show that if plasma LDL is reduced, plasma VLCFA are lower because of reduced VLCFA transport capacity of the blood. It is vital to also determine VLCFA in leukocytes and/or erythrocytes. In a university hospital patient care and clinical research are sometimes intertwined. We are planning to create a large cohort of men and women with X-ALD and follow them prospectively according to a fixed protocol. Our new collaboration with the VUmc will allow us to implement new MRI techniques in the follow-up, that could lead to earlier detection of the onset of cerebral ALD. This will provide well documented information on the natural history of X-ALD and might also be useful in future therapeutic research. 8 Research into the pathophysiology of cerebral ALD has been seriously hampered by the lack of a suitable mouse model. The current X-ALD mouse develops an AMN-like phenotype. We hope that the new mouse currently developed by Dr. S. Kemp will develop full blown cerebral ALD. This would provide many new research opportunities to unravel the pathophysiology of cerebral ALD. The clinical course in X-ALD is unpredictable, there is no genotype-phenotype correlation (Kemp et al 2012). The identification of genetic or environmental modifiers could help in identifying those patients at risk for cerebral ALD. This would improve management, because a tailor made follow-up program can then be offered. New genetic modifiers have been discovered in collaboration with Professor Aubourg that bring this goal a step closer. In the women with X-ALD we hoped to be able to predict symptomatic status by analyzing the X-inactivation pattern in fibroblasts. We were unable to show an association, as described in Chapter 7. We are planning to analyze the X-inactivation pattern in different tissues and to increase the number of patients in our analysis by collaborating with other research groups. 107

110 Chapter 8 Implications for clinical practice X-linked adrenoleukodystrophy is a complex disorder. Patients with X-ALD benefit from careful follow-up, as described in Chapter 1. Timely identification of adrenocortical insufficiency and hormonal substitution therapy reduces morbidity and mortality. There is a narrow window of opportunity for bone marrow transplantation after the onset of cerebral ALD and follow-up with routine MRI scans is recommended. Furthermore, virtually all patients will gradually develop neurological deficits and symptomatic treatment of bladder- or bowel dysfunction and spasticity should be started. This follow-up should be offered in a large center by physicians with experience in this field. Ideally, this follow-up should be done according to a fixed protocol. This will most likely result in better patient care and allows us to systematically study the natural history of the disorder. We are currently working on improving the X-ALD outpatient clinic. References Aubourg P, Adamsbaum C, Lavallard-Rousseau MC, Rocchiccioli F, Cartier N, Jambaqué I, Jakobezak C, Lemaitre A, Boureau F, Wolf C, et al. A two-year trial of oleic and erucic acids ( Lorenzo s oil ) as treatment for adrenomyeloneuropathy. N Engl J Med. 9;329(11): Moser HW, Mahmood A, Raymond GV X-linked adrenoleukodystrophy. Nat Clin Pract Neurol 3(3): Kemp S, Wanders R Biochemical aspects of X-linked adrenoleukodystrophy. Brain Pathol 20(4): Kemp, S., Berger, J. & Aubourg, P X-linked adrenoleukodystrophy: Clinical, metabolic, genetic and pathophysiological aspects. Biochim Biophys Acta 1822(9): van Geel BM, Koelman JH, Barth PG, Ongerboer de Visser BW Peripheral nerve abnormalities in adrenomyeloneuropathy: a clinical and electrodiagnostic study. Neurology 46(1): van Geel BM, Assies J, Haverkort EB, Koelman JH, Verbeeten B Jr, Wanders RJ, Barth PG Progression of abnormalities in adrenomyeloneuropathy and neurologically asymptomatic X-linked adrenoleukodystrophy despite treatment with Lorenzo s oil. J Neurol Neurosurg Psychiatry. 67(3):

111 Nederlandse versie Chapter 9 Van gen naar ziekte; X-gebonden adrenoleukodystrofie Marc Engelen, Stephan Kemp en Björn M. van Geel 9 Afdeling Neurologie en het Laboratorium voor Genetische Metabole Ziekten, Academisch Medisch Centrum/ Universiteit van Amsterdam, Amsterdam. Afdeling Neurologie, Medisch Centrum Alkmaar, Alkmaar. Nederlands Tijdschrift voor Geneeskunde (2008) 152(14):

112 Chapter 9 De ziekte X-gebonden adrenoleukodystrofie (X-ALD) is de frequentst voorkomende peroxisomale stofwisselingsziekte met als biochemisch kenmerk een stapeling in plasma en weefsels van zeerlangeketenvetzuren (ZLKV), dit zijn vetzuren met meer dan 22 koolstofatomen (Moser et al 2001). De symptomen zijn zeer variabel, ook bij familieleden met dezelfde mutatie (Kemp et al 2001). Vooralsnog is het niet mogelijk de klinische manifestaties bij een individuele patiënt te voorspellen. Er wordt een aantal fenotypen onderscheiden (Tabel). Cerebrale X-ALD. De cerebrale vormen van X-ALD zijn het ernstigst. Bij ongeveer 30% van alle mannelijke patiënten ontwikkelt dit fenotype zich op de kinderleeftijd, meestal tussen het 3 e en het 10 e levensjaar. Deze vorm wordt ook wel childhood-onset cerebrale ALD (CCALD) genoemd. Soms manifesteert de ziekte zich op hogere leeftijd en heet dan adolescent cerebral ALD of adult cerebral ALD (Moser et al 2001). Deze vormen hebben een sluipend begin met vaak als eerste verschijnselen gedragsverandering en afname van schoolprestaties, gevolgd door visusproblemen, doofheid, spastische tetraparese en epilepsie. De verschijnselen zijn snel progressief; na twee jaar verkeren de meeste patiënten in een vegetatieve toestand of zijn overleden. De klachten worden veroorzaakt door een inflammatoire reactie met demyelinisatie in de hersenen (Powers et al 1992). Bij MRIonderzoek van de hersenen worden kenmerkende verschijnselen gevonden (Figuur 1) (Loes et al 1994). Een groot deel van de jongens heeft bijnierschorsinsufficiëntie. 9 Figuur 1: MRI-afbeeldingen van het cerebrum van een jongen met childhood-onset cerebrale adrenoleukodystrofie: (a) T2-gewogen opname met uitgebreide hyperintense wittestofafwijkingen pariëto-occipitaal; (b) T1-gewogen opname na intraveneus contrast met aankleuring van de rand van de laesie. Adrenomyeloneuropathie (AMN). De meeste patiënten met X-ALD die de adolescentie gepasseerd zijn, krijgen adrenomyeloneuropathie, doorgaans tussen het 20e en het 40e levensjaar (Moser et al 2001). AMN wordt gekenmerkt door langzaam progressieve spastische paraparese, gestoorde sensibiliteit in de benen en mictie- en defecatieproblemen. Daarnaast heeft 80% van de mannen bijnierschorsinsufficiëntie en 75% tekenen van 110

113 Nederlandse versie Overzicht van de fenotypen van X-gebonden adrenoleukodystrofie (X-ALD) met daarbij vermeld de frequentie van vóórkomen en belangrijke klinische verschijnselen ALD AMN Addison-only draagsters met cerebrale afwijkingen zonder cerebrale afwijkingen adult cerebral adolescent cerebral childhood onset neemt toe met leeftijd neemt af met leeftijd frequentie (%) van de AMN patiënten (in 10 jaar) 2, > 21 > 18 > 18 > 2 variabel, > pariëto-occipitaal - zeer zeldzaam uitgebreid uitgebreid uitgebreid Wallerse degeneratie van piramidebanen leeftijd in jaren bij ontstaan symptomen piramidebaansyndroom wittestofafwijkingen op MRI zeer zeldzaam gedrags- en cognitieve stoornissen - + / - sensomotorisch, voornamelijk axonaal polyneuropathie - zeldzaam mogelijk sensomotorisch, voornamelijk axonaal altijd AD < 1% vaak AD en gonadale insufficiëntie meestal AD meestal AD meestal AD vaak AD en gonadale insufficiëntie endocriene functiestoornissen progressie snel snel snel langzaam snel - langzaam + = aanwezig; = afwezig; ALD = adrenoleukodystrofie; AMN = adrenomyeloneuropathie; AD = ziekte van Addison

114 Chapter 9 gonadale insufficiëntie (Assies et al 1997; Moser et al 2001). Kenmerkend is de spaarzame beharing van het hoofd (Figuur 2). De diagnose wordt vaak pas laat gesteld (van Geel et al 1993). Pathologisch is er axonale degeneratie in het ruggenmerg en in de perifere zenuwen, zonder ontstekingscomponent (van Geel et al 1996; Powers et al 2000). Patiënten met AMN hebben een grote kans (20% in 10 jaar) op additionele cerebrale demyelinisatie (van Geel et al 2001). Hoewel de verdeling van de cerebrale laesies bij AMN er iets anders uitziet, is de prognose net zo slecht als die van CCALD (van Geel et al 2001). 9 Figuur 2: Typische spaarzame beharing van het hoofd van een patiënt met adrenomyeloneuropathie. Addison-only -fenotype. Sommige patiënten hebben aanvankelijk bijnierschorsinsufficiëntie zonder neurologische verschijnselen. De symptomen zijn hetzelfde als die van de ziekte van Addison (Moser et al 2001). Hyperpigmentatie van de handlijnen is bijvoorbeeld een uiting van deze bijnierschorsinsufficiëntie (Figuur 3). Bij jongens en mannen met een primaire bijnierschorsinsufficiëntie dient onderzoek te worden ingezet naar X-ALD, zeker bij afwezigheid van antilichamen tegen bijnierschorscellen (Mukherjee et al 2006). Patiënten bij wie afwijkingen aan het zenuwstelsel aanvankelijk ontbreken, kunnen deze gedurende het leven alsnog krijgen, zodat het ene fenotype in het andere kan overgaan. Uiteindelijk ontstaan bij de meeste patiënten met dit fenotype toch neurologische uitvalsverschijnselen. Asymptomatische en atypische vormen. Bij screenen van de familie van indexpatiënten kunnen mannen gevonden worden die tot op hoge leeftijd geen of nauwelijks klachten krijgen. Ook zijn er patiënten beschreven met een atypisch klachtenpatroon, zoals de in Azië beschreven spinocerebellaire varianten (Ohno et al 1984). 112

115 Nederlandse versie Draagsters. Zoals vaker bij X-chromosomaal overervende ziekten is waargenomen, kunnen ook draagsters symptomen krijgen. Op patiëntenbijeenkomsten in de Verenigde Staten zijn draagsters onderzocht en van hen bleek 53% neurologische afwijkingen te hebben (Moser et al 1991). In Nederland bleek 48% van de draagsters verschijnselen van AMN te hebben. Dit percentage neemt toe met de leeftijd en in de leeftijdscategorie jaar heeft zelfs 79% klachten die samenhangen met X-ALD (van Geel, 2000). Bijnierschorsinsufficiëntie en cerebrale afwijkingen zijn bij draagsters zeer zeldzaam (Moser et al 2001). Figuur 3. Gehyperpigmenteerde hand-lijnen bij bijnierschorsinsufficiëntie als gevolg van adrenoleukodystrofie. Het gen en het eiwit X-ALD wordt gekenmerkt door een verhoging van ZLKV in plasma en weefsels (Moser et al 1981). De afbraak van ZLKV via beta-oxidatie is een peroxisomale functie. Al sinds 1984 is bekend dat de peroxisomale beta-oxidatie in cellen van patiënten met X-ALD sterk verminderd is, tot 25-30% van normaal (Singh et al 1984). In 1993 is het gen geïdentificeerd dat verantwoordelijk is voor het defect bij X-ALD (Mosser et al 1993). Dit ABCD1-gen codeert voor een peroxisomaal transmembraaneiwit dat geclassificeerd is als een ATP- binding cassette (ABC)- half transporter. De functie is nog niet onomstotelijk vastgesteld, maar verondersteld wordt dat het eiwit een rol speelt bij het transport van ZLKV over de peroxisomale membraan (Figuur 4) (Kemp and Wanders, 2007). 9 De cel ZLKV accumuleren in alle cellen en weefsels, maar uiteindelijk ontstaat bij X-ALD schade aan het zenuwweefsel en de steroïdhormoonproducerende weefsels (Moser et al 2001). Specifieke celpopulaties die te gronde gaan, zijn de hormoonproducerende cellen in de zona fasciculata en reticularis van de bijnierschors (en minder in de zona glomerulosa), de leydigcellen in de testes en oligodendrocyten in het centrale zenuwstelsel (Moser et al 2001). 113

116 Chapter 9 figuur 4. Schematisch overzicht van het biochemisch defect in X-gebonden adrenoleukodystrofie (X-ALD). Mutaties in het ABCD1-gen veroorzaken een afwezig of defect ALD-proteïne (ALDP). Zeerlangeketenvetzuren (ZLKV) worden gemaakt door enzymen in de membraan van het endoplasmatisch reticulum uit langeketenvetzuren. Als door een defect ALDP de afbraak van ZLKV verminderd is, ontstaat stapeling ervan in de cel. 9 Hoe ZLKV-accumulatie leidt tot celdood is vooralsnog niet opgehelderd. In 1988 is een toxisch effect van ZLKV op hormoonproducerende cellen in de bijnier aangetoond (Whitcomb et al 1988). Ook is bekend dat ZLKV veranderingen in de microviscositeit van celmembranen veroorzaken, waardoor signaaltransductie verstoord kan worden (Ho et al 1995). Pathologisch zijn er 2 duidelijk verschillende patronen. Enerzijds is er de niet-inflammatoire axonopathie in ruggenmerg en perifere zenuwen (Powers et al 2000), die gevonden wordt bij AMN, en de atrofie van de bijnieren (Powers et al 1980). Anderzijds ontwikkelt zich bij een deel van de patiënten een hevige inflammatoire reactie in het centrale zenuwstelsel met perivasculaire lymfocytaire infiltraten en snel progressieve cerebrale demyelinisatie (Powers et al 1992). De rol van ZLKV in de pathofysiologie hiervan is onduidelijk. Het is onbekend hoe het komt dat bij sommige patiënten cerebrale demyelinisatie ontstaat en bij anderen niet. Relaties tussen fenotype en ABCD1-mutatie (Kemp et al 2001) of plasma-zlkv-spiegels zijn tot op heden niet gevonden (Moser et al 1999). Waarschijnlijk zijn er modificerende genen die het uiteindelijke fenotype mede bepalen (Asheuer et al 2005). 114