Experimental Physiology

Transcription

1 66 Exp Physiol 99.1 (2014) pp Symposium Report Endoplasmic reticulum stress in vasopressin neurons of familial diabetes insipidus model mice: aggregate formation and mrna poly(a) tail shortening Hiroshi Arima, Yoshiaki Morishita, Daisuke Hagiwara, Masayuki Hayashi and Yutaka Oiso Department of Endocrinology and Diabetes, Nagoya University Graduate School of Medicine, Nagoya, Japan Experimental Physiology New findings What is the topic of this review? Familial neurohypophysial diabetes insipidus (FNDI) is caused by a mutation in the vasopressin (AVP) gene locus. While mutant AVP precursors are reported to accumulate in the endoplasmic reticulum of AVP neurons, it is not clear how AVP neurons cope with the accumulation of misfolded proteins. What advances does it highlight? We show that AVP mrna poly(a) tail length is shortened and the mrna expression is decreased in the FNDI mouse model. The shortening of the mrna poly(a) tail could be a novel unfolded protein response by which aggregate accumulation is reduced. The immunoglobulin heavy chain binding protein (BiP) is an endoplasmic reticulum (ER) chaperone, which binds to newly synthesized secretory and transmembrane proteins to facilitate protein folding. BiP mrna is expressed in the arginine vasopressin (AVP) neurons in the supraoptic nucleus of wild-type mice even in basal conditions, and the expression levels increase in response to dehydration. These data suggest that AVP neurons are subjected to ER stress. Familial neurohypophysial diabetes insipidus (FNDI) is caused by mutations in the gene locus of AVP. The mutant proteins could accumulate in the ER and possibly increase ER stress in the AVP neurons. We bred mice possessing a mutation causing FNDI, which manifested progressive polyuria, as do the patients with FNDI. Electron microscopic analyses demonstrated that aggregates accumulated in the ER of AVP neurons in FNDI mice. Despite polyuria, which could potentially induce dehydration, AVP mrna expression was decreased in the supraoptic nucleus, and the AVP mrna poly(a) tail length was shortened in FNDI mice compared with wild-type mice. Incubation of hypothalamic explants of wild-type mice with ER stressors caused shortening of the poly(a) tail length of AVP mrna, accompanied by decreases in the expression. These data revealed a mechanism by which ER stress decreases poly(a) tail length of AVP mrna, and this reduces the load of unfolded proteins that form the aggregates in ER of the AVP neurons in FNDI mice. (Received 27 July 2013; accepted after revision 7 October 2013; first published online 11 October 2013) Corresponding author H. Arima: 65 Tsurumai-cho, Showa-ku, Nagoya , Japan. arima105@med.nagoya-u.ac.jp DOI: /expphysiol

2 Exp Physiol 99.1 (2014) pp Endoplasmic reticulum stress in vasopressin neurons 67 Introduction Arginine vasopressin (AVP), which plays a crucial role in water balance as an antidiuretic hormone, is synthesized in the magnocellular neurons of the supraoptic nucleus (SON) and paraventricular nucleus (PVN) of the hypothalamus. The AVP gene consists of three exons, which encode the signal peptide, AVP, neurophysin II (NPII) and glycoprotein. The prepro-avp is translated from the mrna and converted to pro-avp by removal of its signal peptide within the endoplasmic reticulum (ER), where the pro-avp folding occurs. The pro-avp is then packaged into the granules via the Golgi apparatus. The AVP, NPII and glycoprotein are yielded during axonal transport to the posterior pituitary, from which AVP is released into the circulation. Deficiency of AVP leads to a disorder called diabetes insipidus, in which urine volume and water intake are increased while the urine osmolality is decreased. Diabetes insipidus can occur on a familiar basis, a disorder called familial neurohypophysial diabetes insipidus (FNDI), which is inherited mainly in an autosomal dominant manner. Although the carriers are normal at birth, the symptoms, such as polyuria and polydipsia, appear several months or years later, despite the existence of one normal allele. While more than 60 point mutations in FNDI have been reported in the AVP gene so far, most mutations exist in the domain of NPII, which functions as a carrier protein of AVP (Christensen & Rittig, 2006). The AVP gene expression levels in the SON and magnocellular PVN are high even in basal conditions, and can be further upregulated by only 1 2% increases in plasma osmolality (Arima et al. 1999). In addition, various proteins are reportedly expressed in the AVP neurons. All these data suggest that AVP neurons are subjected to large demands for protein synthesis. In these conditions, perturbation of the ER homeostasis could occur, which possibly leads to the accumulation of unfolded proteins in the ER lumen (i.e. ER stress), and consequently triggers responses known as the unfolded protein response (UPR; Ron & Walter, 2007). The immunoglobulin heavy chain binding protein (BiP) is one of the most abundant ER chaperones (Haas & Wabl, 1983). BiP binds not only to newly synthesized secretory and transmembrane proteins to facilitate protein folding, but also to misfolded proteins in order to prevent their aggregation (Gething, 1999). In situ hybridization demonstrated that BiP mrna was highly expressed in the SON and PVN in the hypothalamus in basal conditions (Hagiwara et al. 2012). Analyses of adjacent slices for the mrna expression of BiP and AVP showed that the distribution of BiP mrna overlapped that of AVP mrna in the SON and PVN. The dual in situ hybridization demonstrated that 98 and 99% of AVP neurons in the SON and PVN, respectively, expressed BiP mrna. Both AVP and BiP mrna expression levels in the SON and PVN were significantly increased after water deprivation (Hagiwara et al. 2012). In non-stress conditions, BiP is bound to the ER stress transducers, such as inositol-requiring protein- 1, activating transcription factor-6 and protein kinase RNA-like ER kinase, whereas in ER stress conditions it preferentially binds to unfolded proteins and dissociates from the transducers, leading to the activation of the UPR. Given that the expression of BiP is reportedly increased during ER stress (Gething, 1999), the induction can serve as a general indicator of ER stress as well as UPR being triggered. The expression of BiP in the AVP neurons could therefore lead to the interpretation that stimuli for AVP synthesis per se cause an ER stress, as it is reported that a certain percentage of the newly synthesized wild-type protein molecules could fail to mature and be degraded (Petaja-Repo et al. 2000). Alternatively, the expression of BiP mrna might reflect the activity of UPR in the AVP neurons, as UPR is actively involved in ensuring proper function in some secretory cells (Wu & Kaufman, 2006). In any case, it is plausible that the BiP expression is related to ER stress and that BiP might contribute to the protein quality control system as an ER chaperone in the AVP neurons. The mechanisms by which polyuria is caused in FNDI have been explored both in cell lines (Ito et al. 1993, 1997, 1999; Castino et al. 2005) and in animal models (Si-Hoe et al. 2000; Russellet al. 2003). Given that mutant AVP precursors are reportedly trapped within the ER, one hypothesis was that the aggregated mutant proteins in the ER are toxic and cause a loss of AVP neurons. If this is true, progressive polyuria could be explained by gradual loss of AVP neurons. In this regard, it is of note that autophagy was induced and possibly involved in both cell survival and apoptosis in Neuro2a cells expressing a mutant that causes FNDI (Castino et al. 2005). Another hypothesis is that the function of normal NPII is hampered in the presence of mutant NPII (so-called dominant negative effects ). Familial diabetes insipidus model mice In order to gain a better understanding of the mechanisms underlying progressive polyuria in FNDI, we made knock-in mice having a mutation of NPII (Cys98stop), which causes FNDI in humans (Hayashi et al. 2009). Immunohistochemical analyses with antibody for mutant NPII revealed that mutant protein is expressed in the SON as well as in the PVN in heterozygous (FNDI) mice, but not in wild-type mice. Immunohistochemical analyses with antibody for normal NPII (PS41) showed that normal NPII is expressed in the SON as well as in the PVN in both wild-type and FNDI mice. However, while normal

3 68 H. Arima and others Exp Physiol 99.1 (2014) pp NPII is expressed in both the cell bodies and the axons in wild-type mice, in the FNDI mice it is mainly expressed in the cell bodies, with diminished axonal staining. This suggests that the trafficking of normal NPII (or AVP) is disturbed in the presence of mutant NPII, which appears consistent with the dominant negative effects hypothesis. Urine volume in FNDI mice significantly increased compared with that in wild-type mice at 1 month of age, and further increased progressively until 12 months. The increases in urine volume were accompanied by increases in water intake and decreases in urine osmolality. Of note, female FNDI mice had more apparent phenotypes than males, although such differences between genders have not been reported in FNDI patients. It is also shown that round inclusions appeared in the SON and PVN in FNDI mice. Both the number and the size of the inclusions increased with age in male FNDI mice. In contrast, while their size increased with age in the female mice as well, there were fewer inclusion bodies at 12 than at 6 months. Progressive polyuria in the absence of AVP neuronal loss in FNDI mice The analysis of cells expressing AVP mrna in the SON showed that the number of AVP cells did not differ between genotypes in male mice at any time point examined (until 12 months). In female mice, however, while the numbers of AVP cells did not differ between genotypes at 1 and 6 months, they were significantly diminished at 12 months in FNDI mice compared with wild-type mice. The neuronal loss in the SON of 12-month-old female FNDI mice was confirmed by counting the number of cells stained with Cresyl Violet in the SON. Neither active caspase-3 nor terminal deoxynucleotidyl transferase dutp nick end labeling (TUNEL)-positive neurons were observed in 12-month-old female heterozygous mice. Furthermore, the morphological characteristics of apoptosis, including chromatin condensation or nuclear fragmentation, were not detected by the electron microscopy analysis. Thus, we did not find any evidence of apoptosis in the AVP neurons of FNDI mice. Overall, the data in male mice indicate that progressive polyuria could occur even in the absence of AVP neuron cell death. In contrast, the data in female mice indicate that neuronal loss of AVP could occur in the late phase of the disease. Accumulation of aggregates in the ER lumen of AVP neurons in FNDI mice Electron microscopic analyses demonstrated that, while aggregates were already present in the ER lumen of AVP neurons in 1-month-old FNDI mice, such aggregates were not observed in the SON of wild-type mice. In 12-monthold FNDI mice, the lumens of ER in the AVP cells were almost entirely occupied by the aggregates. In some cells, there were massive aggregates that appeared to exist in an enlarged lumen of the ER. In others, round-shaped aggregates, which possibly correspond to inclusion bodies observed in light microscopic analyses, occupied the cytoplasm (Fig. 1). In summary, we succeeded in making FNDI model mice, in which the phenotypes of FNDI patients are mimicked. Furthermore, it was demonstrated that aggregates were accumulated in the ER lumen of AVP neurons in FNDI mice. Given that the transcriptional activities of the AVP gene in the magnocellular neurons are high, the aggregate accumulation in the ER would easily induce cell death. However, our analyses showed that this was not the case. We therefore hypothesized that there might be some protective mechanisms by which aggregate accumulation was reduced in the AVP neurons in FNDI mice. Shortening of AVP mrna poly(a) tail length in FNDI mice In situ hybridization demonstrated that the levels of AVP heteronuclear (hn)rna expression in the SON, a sensitive indicator for gene transcription, were not significantly different between wild-type and FNDI mice in the basal conditions (Morishita et al. 2011). The levels of AVP hnrna expression in the SON were increased to a similar extent after osmotic stimuli in both wild-type and FNDI mice. These data suggest that transcriptional activities of the AVP gene are maintained in FNDI mice. In contrast, the expression levels of AVP mrna in the SON were significantly reduced in FNDI mice compared with those of wild-type mice in basal conditions. Furthermore, while AVP mrna levels in both types of mice were also significantly increased in response to dehydration, the levels were significantly lower in FNDI mice than in wildtype mice. Using cdna from the hypothalamus of FNDI mice, RT-PCR revealed that the expression of normal mrna and mutant mrna of AVP was similar in FNDI mice. These data exclude the possibility that the mrna expression is reduced by the nonsense-mediated decay mechanism. Northern blot analyses demonstrated that the size of AVP mrna was reduced in FNDI mice in comparison to the wild-type mice (Fig. 1). In contrast, removal of the poly(a) tail resulted in the same size of mrna in both types of mice. The sizes of oxytocin (OT) with or without the poly(a) tail were not different between wild-type and FNDI mice. Thus, it is suggested that AVP mrna is decreased via shortening of the mrna poly(a) tail length in FNDI mice. Effects of ER stress on mrna poly(a) tail length To determine whether or not ER stress could decrease expression levels and poly(a) tail length of AVP mrna,

4 Exp Physiol 99.1 (2014) pp Endoplasmic reticulum stress in vasopressin neurons 69 Figure 1. Aggregate formation in endoplasmic reticulum (ER) and arginine vasopressin (AVP) mrna poly(a) tail length in mice with familial neurohypophysial diabetes insipidus (FNDI) Expression of mutant neurophysin II (NPII) results in the formation of aggregates in the ER lumen of AVP neurons. The AVP mrna poly(a) tail length is shortened in the FNDI mouse model, and this could reduce accumulation of aggregates by decreasing the translation of mutant protein. Abbreviations: Ag, aggregates; and Nu, nucleus. the effects of ER stressors on mrna expression and poly(a) tail length were examined in hypothalamic organotypic cultures of wild-type mice. Incubation with the ER stressors thapsigargin or tunicamycin increased BiP mrna expression and phosphorylation of protein kinase RNA-like ER kinase, and induced spliced X- box binding protein 1 (XBP1) mrna expression in the hypothalamic cultures, indicating that ER stressors indeed induced UPR in the mouse hypothalamus. Northern blot analyses demonstrated that these ER stressors significantly reduced the peak position of AVP mrnas and OT mrnas, while the removal of the poly(a) tail resulted in similarsizesofavporotmrna.furthermore,real-time quantitative RT-PCR demonstrated that treatment with ER stressors reduced the expression levels of both AVP and OT mrnas. In contrast, incubation of hypothalamic explants with tauroursodeoxycholic acid, a molecular chaperone reducing ER stress, significantly increased the peak position of AVP mrnas and OT mrnas, while removal of the poly(a) tail resulted in similar sizes of AVP or OT mrna. Furthermore, real-time quantitative RT-PCR demonstrated that the incubation with tauroursodeoxycholic acid significantly increased both AVP and OT mrna expression levels. These data suggest that ER stress could reduce the expression levels of AVP and OT mrna by shortening the length of mrna poly(a) tails. The mrna poly(a) tail length of some secreted or transmembrane proteins is reported to change in several conditions. This is also true for neurohypophysial hormones, as follows: osmotic stimulus induced elongation of AVP and OT mrna poly(a) tails, accompanied by increases in the mrna expression in the SON (Carrazana et al. 1988; Zigg et al. 1988; Carter & Murphy, 1991); hyponatraemia induced shortening of the AVP and OT mrna poly(a) tail (Chooi et al. 1992; Svane et al. 1995); and the circadian peak of AVP mrna expression in the suprachiasmatic nucleus was accompanied by elongation of the AVP mrna poly(a) tail length (Robinson et al. 1988). These data suggest that the length of the poly(a) tail is a crucial determinant of AVP and OT mrna stability. The finding that the transcriptional activity of AVP neurons was maintained in FNDI mice suggests that changes in AVP mrna poly(a) tail length are not likely to be due to decreased cell viability. Decreases in mrna poly(a) tail length are known to be accompanied by decreases in mrna stability as well as the efficacy of translation (Bernstein & Ross, 1989; Wakiyama et al. 1997; Kuraishi et al. 2002). Givne that shortening of the poly(a) tail length of AVP mrna is found in FNDI model mice, it is possible that the reduction in the translation might be one of the causes of the AVP depletion responsible for polyuria in FNDI. Another interpretation is that the shortening of the mrna poly(a) tail might comprise one of the cellular protective mechanisms (or UPR) by which accumulation of mutant proteins would be reduced (Fig. 1).

5 70 H. Arima and others Exp Physiol 99.1 (2014) pp Conclusion and perspectives Our study using FNDI model mice revealed that aggregates were accumulated in the ER lumen of AVP neurons and that the length of the AVP mrna poly(a) tail was shortened. As demonstrated in hypothalamic organotypic cultures, ER stress decreases the poly(a) tail length of AVP mrna as well as its expression levels; therefore, the shortening of poly(a) tail length could be a novel UPR to reduce the accumulation of aggregates in the ER. In contrast, as the size of the aggregates increases over time and is correlated with the increase in urine volumes in the FNDI mice, it is important to clarify in future how the aggregate formation affects the function and viability of AVP neurons in FNDI mice. Call for comments Readersareinvitedtogivetheiropiniononthisarticle. To submit a comment, go to: submit/expphysiol;99/1/66. References Arima H, Kondo K, Kakiya S, Nagasaki H, Yokoi H, Yambe Y, Murase T, Iwasaki Y & Oiso Y (1999). Rapid and sensitive vasopressin heteronuclear RNA responses to changes in plasma osmolality. J Neuroendocrinol 11, Bernstein P & Ross J (1989). Poly(A), poly(a) binding protein and the regulation of mrna stability. Trends Biochem Sci 14, Carrazana EJ, Pasieka KB & Majzoub JA (1988). The vasopressin mrna poly(a) tract is unusually long and increases during stimulation of vasopressin gene expression in vivo. MolCellBiol8, Carter DA & Murphy D (1991). Rapid changes in poly (A) tail length of vasopressin and oxytocin mrnas form a common early component of neurohypophyseal peptide gene activation following physiological stimulation. Neuroendocrinology 53, 1 6. Castino R, Isidoro C & Murphy D (2005). Autophagy-dependent cell survival and cell death in an autosomal dominant familial neurohypophyseal diabetes insipidus in vitro model. FASEB J 19, Chooi KF, Carter DA & Murphy D (1992). Decrease in hypothalamic vasopressin mrna poly(a) tail length following physiological stimulation. Cell Mol Neurobiol 12, Christensen JH & Rittig S (2006). Familial neurohypophyseal diabetes insipidus an update. Semin Nephrol 26, Gething MJ (1999). Role and regulation of the ER chaperone BiP. Semin Cell Dev Biol 10, Haas IG & Wabl M (1983). Immunoglobulin heavy chain binding protein. Nature 306, Hagiwara D, Arima H, Morishita Y, Goto M, Banno R, Sugimura Y & Oiso Y (2012). BiP mrna expression is upregulated by dehydration in vasopressin neurons in the hypothalamus in mice. Peptides 33, Hayashi M, Arima H, Ozaki N, Morishita Y, Hiroi M, Ozaki N, Nagasaki H, Kinoshita N, Ueda M, Shiota A & Oiso Y (2009). Progressive polyuria without vasopressin neuron loss in a mouse model for familial neurohypophysial diabetes insipidus. Am J Physiol Regul Integr Comp Physiol 296, R1641 R1649. Ito M, Jameson JL & Ito M (1997). Molecular basis of autosomal dominant neurohypophyseal diabetes insipidus. Cellular toxicity caused by the accumulation of mutant vasopressin precursors within the endoplasmic reticulum. J Clin Invest 99, ItoM,OisoY,MuraseT,KondoK,SaitoH,ChinzeiT,Racchi M & Lively MO (1993). Possible involvement of inefficient cleavage of preprovasopressin by signal peptidase as a cause for familial central diabetes insipidus. JClinInvest91, Ito M, Yu RN & Jameson JL (1999). Mutant vasopressin precursors that cause autosomal dominant neurohypophyseal diabetes insipidus retain dimerization and impair the secretion of wild-type proteins. J Biol Chem 274, Kuraishi T, Mizoguchi Y, Sun Y, Aoki F, Imakawa K & Sakai S (2002). The casein mrna decay changes in parallel with the poly(a) tail length in the mouse mammary gland. Mol Cell Endocrinol 190, Morishita Y, Arima H, Hiroi M, Hayashi M, Hagiwara D, Asai N, Ozaki N, Sugimura Y, Nagasaki H, Shiota A, Takahashi M & Oiso Y (2011). Poly(A) tail length of neurohypophysial hormones is shortened under endoplasmic reticulum stress. Endocrinology 152, Petaja-Repo UE, Hogue M, Laperriere A, Walker P & Bouvier M (2000). Export from the endoplasmic reticulum represents the limiting step in the maturation and cell surface expression of the human δ opioid receptor. JBiol Chem 275, Robinson BG, Frim DM, Schwartz WJ & Majzoub JA (1988). Vasopressin mrna in the suprachiasmatic nuclei: daily regulation of polyadenylate tail length. Science 241, Ron D & Walter P (2007). Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev MolCellBiol8, Russell TA, Ito M, Ito M, Yu RN, Martinson FA, Weiss J & Jameson JL (2003). A murine model of autosomal dominant neurohypophyseal diabetes insipidus reveals progressive loss of vasopressin-producing neurons. JClinInvest112, Si-HoeSL,DeBreeFM,NijenhuisM,DaviesJE,HowellLM, Tinley H, Waller SJ, Zeng Q, Zalm R, Sonnemans M, Van Leeuwen FW, Burbach JP & Murphy D (2000). Endoplasmic reticulum derangement in hypothalamic neurons of rats expressing a familial neurohypophyseal diabetes insipidus mutant vasopressin transgene. FASEB J 14, Svane PC, Thorn NA, Richter D & Mohr E (1995). Effect of hypoosmolality on the abundance, poly(a) tail length and axonal targeting of arginine vasopressin and oxytocin mrnas in rat hypothalamic magnocellular neurons. FEBS Lett 373,

6 Exp Physiol 99.1 (2014) pp Endoplasmic reticulum stress in vasopressin neurons 71 Wakiyama M, Futami T & Miura K (1997). Poly(A) dependent translation in rabbit reticulocyte lysate. Biochimie 79, Wu J & Kaufman RJ (2006). From acute ER stress to physiological roles of the Unfolded Protein Response. Cell Death Differ 13, Zingg HH, Lefebvre DL & Almazan G (1988). Regulation of poly(a) tail size of vasopressin mrna. J Biol Chem 263, Additional Information Competing interests None declared. Funding This research was supported by Grants-in-Aid for Scientific Research (Research on Hypothalamo- Hypophyseal Disorders) from the Ministry of Health, Labor, and Welfare of Japan.