Mould exposure at home relates to inflammatory markers in blood

Transcription

1 Er Respir J 2003; 21: DOI: / Printed in UK all rights reserved Copyright #ERS Jornals Ltd 2003 Eropean Respiratory Jornal ISSN Mold exposre at home relates to inflammatory markers in blood L. Beijer, J. Thorn, R. Rylander Mold exposre at home relates to inflammatory markers in blood. L. Beijer, J. Thorn, R. Rylander. #ERS Jornals Ltd ABSTRACT: Living in damp bildings has been associated with airway symptoms, sspected to be de to inflammatory reactions. The relationship between home exposre to mold and signs of inflammation was, therefore, stdied. Nonsmoking sbjects with a high (G-high, w4.0 ng?m -3, n=17) or low (G-low, v2.0 ng?m -3, n=18) amont of airborne b(1r3)-d-glcan, an indicator of mold biomass, in the home were recrited. Blood samples were analysed for granlocytic enzymes, T-cell sbsets and the secretion of cytokines from in vitro incbated peripheral blood mononclear cells (PBMCs). In the G-high grop, the proportion of cytotoxic T-cells (CD8zS6F1z) was lower and secretion of tmor necrosis factor-a from PBMCs higher than in the G-low grop. There were no significant differences in secretion of interferon gamma and interlekin (IL)-4 from PBMCs between the two grops. Among nonatopic sbjects, the ratio between interferon gamma and IL-4 was significantly higher in the G-high grop than in the G-low grop and was related to the amont of b(1r3)-d-glcan in the home. No significant differences were fond regarding secretion of IL-10 or IL-1b from PBMCs, eosinophil cationic protein or myeloperoxidase in serm, or differential cell conts in blood. The effects fond on inflammatory markers in relation to b(1r3)-d-glcan in the home sggest preglation of some parts of the inflammatory/immnological system de to mold exposre. Er Respir J 2003; 21: Dept of Environmental Medicine, Sahlgrenska Academy, Göteborg University, Göteborg, Sweden. Correspondence: L. Beijer, Dept of Environmental Medicine, Box 414, SE Göteborg, Sweden. Fax: lena.beijer@envmed.g.se Keywords: Cytotoxic T-cell b(1r3)-d-glcan interferon gamma interlekin-4 mold tmor necrosis factor-a Received: October Accepted after revision: Agst This stdy was spported by the Vårdal Fondation, Stockholm, Sweden, (grant A96 082) and the Swedish Heart Lng Fondation, Stockholm, Sweden, (grant ). Nmeros reports describe an association between living in damp bildings and the presence of symptoms in the airways as well as general fatige and headache [1 3]. Some stdies have shown an increased risk of asthma in children and adlts [4 6]. Increased indoor hmidity creates the conditions for growth of microorganisms, and the presence of mold growth has often been reported in association with symptoms. The mechanism behind these symptoms cold ths be related to an increased inflammatory reaction in the airways in response to inhaled microorganisms or sbstances derived from these [7]. Most previos stdies on health effects associated with living or working in damp and/or moldy bildings have sed qestionnaires to assess the prevalence of symptoms. A few stdies have performed objective measrements of health effects sch as changes in lng fnction, airway reactivity or markers of inflammation in body compartments. In atopic children with asthma, peak expiratory flow variability was associated with the amont of b(1r3)-d-glcan relative to the living room floor area [8]. Increased levels of tmor necrosis factor-a (TNF-a), interlekin (IL)-6 and nitric oxide in the nasal lavage flid of school staff were related to staying in a mold-contaminated school [9]. Myeloperoxidase (MPO) levels in serm were increased among sbjects living in hoses with w1 ng airborne b(1r3)-d-glcan?m -3 compared to sbjects exposed to v1 ng airborne b(1r3)-d-glcan?m -3 [10]. Exposre to indoor molds has also been associated with effects on the immne system and is a possible risk factor for atopic sensitisation to airborne allergens. In a stdy among children, an increased nmber of memory T-cells and a redced CD4/CD8 ratio were related to mold exposre in the home [11]. In a stdy of children from Germany, reporting of dampness and mold in the dwelling was associated with a higher freqency of atopic sensitisation [12]. Indoor exposre to aspergills was a risk factor for atopy in a cohort of Astralian children [13] and a stdy from Finland showed an increased prevalence of atopy among moldexposed schoolchildren [14]. In order to frther evalate the effects of mold exposre at home on the inflammatory and immne system, a stdy was performed among persons living in homes with high (G-high) or low (G-low) levels of airborne b(1r3)-d-glcan. This polyglcose compond, present in the cell wall of fngi, was sed as an indicator of the total airborne mold biomass. Clinical effects were assessed by measring levels of inflammatory markers and nmbers of lymphocyte sbsets in blood and secretion of cytokines from cltivated peripheral blood mononclear cells (PBMCs).

2 318 L. BEIJER ET AL. Sbjects Methods Sbjects for the stdy were recrited from a previos investigation in an area of terraced hosing [10] and by advertising in the local press, seeking persons with mold problems in their homes. Inclsion criteria were an age of yrs, nonsmokers and the absence of physician-diagnosed asthma or diseases reqiring medication. The Ethics Committee of the Faclty of Medicine at Göteborg University, Göteborg, Sweden, approved the stdy and all participating sbjects gave their informed consent. The amont of airborne b(1r3)-d-glcan in the sbjects9 homes was measred and two grops were defined within the recrited poplation, those with v2.0 ng b(1r3)-d-glcan?m -3 in their homes (G-low) and those withw4.0 ng b(1r3)-d-glcan?m -3 (G-high), 35 sbjects in all. The limits sed for high and low exposre were based on the reslts from the previos terraced hosing stdy, in which 30% of the 75 hoses examined had a b(1r3)-d-glcan level of w4 ng?m -3 and 30% of v2.0 ng?m -3 [10]. One person was recrited from each home, except for three homes in the G-low grop and two in the G-high grop, where coples participated. Of the 21 persons recrited by advertising, 43% were classified into the G-high grop, and, of the 14 persons recrited from the previos stdy, 57% were classified into the G-high grop according to the criteria described above. Table 1 shows home b(1r3)-d-glcan levels and sbject characteristics. Two blood samples were taken from each sbject for analysis of serm marker levels and cytokine secretion from PBMCs. The samples were taken at an interval of 2 3 weeks and the mean of the two measrements sed for each person. Lymphocyte sbtyping was performed sing a single blood sample from each sbject. Determination of b(1r3)-d-glcan levels Airborne dst was created sing a machine designed to whirl p dst approximately eqivalent to that generated by a few people moving abot in a room [15]. The dst generator was rn for 5 min and this was followed by a 10-min pase. Two sch periods were sed for the sampling. Dplicate air samples Table 1. Home b(1r3)-d-glcan levels and sbject characteristics G-low G-high Sbjects n Homes n b(1r3)-d-glcan ng?m ( ) 6.0 ( ) Females n (%) 9 (50) 6 (35) Atopics n (%) 6 (33) 5 (29) Age yrs 52 (26 66) 45 (19 59) Data are presented as absolte vales with or withot percentages in parenthesis or as median (range). were taken in this 30 min period by drawing air throgh polycarbonate filters with a pore size of 0.8 mm (Isopore1; Millipore, Inc., Boston, MA, USA) monted side-by-side 0.8 m above the floor at a rate of 5 L?min -1. Two rooms were investigated in each home. The bedroom was sampled in 25 of the 30 homes and the living room in 26 homes. Other rooms sampled were workrooms, dining rooms and basements. The filters were shaken for 10 min in 10 ml 0.3 M sodim hydroxide in pyrogen-free water on ice in order to nwind the triple helical strctre of the b(1r3)-d-glcan and make it water-solble. The soltion was then analysed sing a b(1r3)-d-glcanspecific Limls lysate [16]. Filter extract samples of 50 ml were placed in a microwell plate and 50 ml specific b(1r3)-d-glcan lysate (Fngitic G Test1; Seikagak Corp., Tokyo, Japan) were added. The plate was incbated in a spectrophotometer (Well reader1; Scinics Corp., Tokyo, Japan), and the kinetics of the ensing color reaction read photometrically, transformed into absorbance nits and compared to a standard crve created sing the Fngitech G TE Reference Standard (Seikagak Corp.). The reslts were expressed in nanograms per millilitre of liqid and the detection limit was 20 pg?ml -1. The airborne concentration of b(1r3)-d-glcan was calclated from the airflow throgh the filter. The mean exposre determined from the two sampling sites was sed in the analysis. Symptoms The sbjects answered a qestionnaire on symptoms present dring the preceding 3 months. The symptoms addressed were cogh (dry or with phlegm), chest tightness, shortness of breath, irritation in the eyes, nose or throat, and nasal congestion and itchy nose. Qestions were also posed on headache, nsal tiredness, wheezy chest and skin problems. For the prevalence of each symptom, a scale of 1 4 was sed: 1: never or very seldom; 2: a few times a month; 3: a few times a week; and 4: daily. Blood cell conts Venos blood was collected in tbes containing ethylene diamine tetra-acetic acid and a differential cell cont performed on smears stained with May- Grünwald Giemsa and conted sing a light microscope. Lymphocyte sbsets were qantified sing flow cytometry. Helper/indcer (CD3zCD4z) and cytotoxic/sppressor (CD3zCD8z) T-cell poplations were identified. CD3zCD8z cells were also analysed in combination with a monoclonal antibody binding S6F1, an epitope of the a-chain of lekocyte fnctionassociated antigen-1 complex (CD11a). CD8z T-cells expressing S6F1 have been classified as cytotoxic cells, whereas CD8z T-cells showing weak staining for S6F1 are either naive cells or sppressor effector cells [17]. Owing to technical problems, nmbers of cytotoxic

3 MOULD AND INFLAMMATION 319 T-cells (CD8zS6F1z) cold not be determined in two G-low and three G-high sbjects. means sed for analysis. The median intrasbject variation for all of the cytokine analyses was y30%. Markers of granlocyte activation Eosinophil cationic protein (ECP) was assayed in serm sing a florescent enzyme immnoassay techniqe (CAP ECP FEIA; Pharmacia Diagnostics, Uppsala, Sweden). MPO was assayed sing a radioimmnoassay techniqe (CAP MPO RIA; Pharmacia Diagnostics). Atopy Atopy was determined by measring the concentration of specific serm immnogloblin E antibodies directed against 10 airborne allergens (tree and grass pollen, domestic animal, hose dst mite and mold) sing a florescent enzyme immnoassay techniqe (CAP Phadiatop1 FEIA; Pharmacia Diagnostics). The reslts were expressed as positive (atopic) or negative (nonatopic) sing a specific immnogloblin E calibrator with a ct-off vale of 0.35 ku?l -1 as reference, in accordance with the manfactrer9s instrctions. Cytokine prodction Venos blood was collected in vactainer tbes containing sodim citrate, a gel barrier and density gradient flid (CPT TM tbes; Becton Dickinson, Franklin Lakes, NJ, USA). After centrifgation in a horizontal rotor at 1,5006g and room temperatre for 20 min, the PBMCs were collected and washed twice in Hank9s salt soltion spplemented with 10% homologos serm. The PBMCs were then sspended in AIM V1 medim (Gibco BRL, Life Technologies, Paisley, UK) spplemented with M 2-mercaptoethanol to a final concentration of cells?ml -1. The cells were incbated with or withot phytohaemaggltinin (PHA, Mrex Diagnostics Ltd, Dartford, UK) at a final concentration of 250 ng?ml -1 or lipopolysaccharide (E.coli 026:B6; Difco Laboratories, Detroit, MI, USA) at a final concentration of 500 pg?ml -1.PHAwassedas the major T-cell stimls for measrement of interferon gamma (IFN-c) and IL-4 secretion and lipopolysaccharide as the major monocyte stimls for the measrement of TNF-a secretion. After incbation for 48 h in 5% CO 2 at 37C, the spernatants were collected and stored at -25C before cytokine analysis. TNF-a from nstimlated PBMCs was analysed sing an enzyme-linked immnosorbent assay (ELISA) kit with a sensitivity of 0.1 pg?ml -1 (Qantikine high sensitive; R&D Systems, Abingdon, UK). IL-1b, IL-4, IL-10, IFN-c and TNF-a from stimlated PBMCs were analysed sing PeliKinecompact TM hman cytokine ELISA (Central Laboratory of the Netherlands Red Cross Blood Transfsion Service, Amsterdam, the Netherlands). Cytokines from PBMCs were analysed for all of the G-low sbjects and 14 of the G-high sbjects. Cytokine secretion from stimlated and nonstimlated PBMCs was measred on two separate occasions and the Statistical analyses For dplicate measrements on the same person, the coefficient of variation (CV) was calclated in order to evalate the variability between the first and second measrements. The reslts are presented as grop median with 10th and 90th percentiles in parenthesis. The Mann-Whitney U-test was sed to test differences in blood analyses between the G-high and G-low grops. Differences in symptoms were tested sing the Chi-sqared test. The sbgrop of nonatopic sbjects was also analysed. The low nmber of atopic sbjects did not warrant separate statistical analysis of this grop. The correlation between continos variables was calclated sing the Spearman rank correlation test. Reslts Table 2 shows the reslts of the lymphocyte typing. There was a trend toward lower nmbers of CD8z cells as well as of cytotoxic T-cells (CD8zS6F1z) in the G-high grop. The proportion of cytotoxic T-cells (CD8zS6F1z) was significantly lower in the G-high grop compared with the G-low grop. Similar reslts were seen in separate analysis of nonatopic sbjects (data not shown). No difference was fond for CD4z T-cells between the G-low and G-high grops. Secretion of TNF-a from nonstimlated PBMCs was significantly greater in the G-high grop compared with the G-low grop (10.7 ( ) verss 7.3 ( ) pg?ml -1,pv0.05). A similar tendency was fond in the separate analysis of nonatopic sbjects, althogh the difference was not significant (data not shown). The median intrasbject CV between dplicate TNF-a analyses in the G-high grop was 31% (maximm 84%) and in the G-low grop 21% (maximm 52%). When only sbjects with a CV of v20% between dplicate samples were compared, a significant difference between the grops was still fond (9.3 Table 2. Nmbers and proportions of different CD8z T-lymphocytes in the blood of persons with low (G-low) and high (G-high) levels of (1R3)-D-glcan in their homes G-low G-high CD8z Sbjects n Nmber 10 9 cells?l ( ) 0.35 ( ) Proportion % 29.5 ( ) 23.5 ( ) CD8zS6F1z Sbjects n Nmber 10 9 cells?l ( ) 0.14 ( ) Proportion % 22.5 ( ) 12.5 ( )* Data are presented as median with the 10th and 90th percentiles in parenthesis. G-low: v2.0 ng?m -3 ; G-high: w4.0 ng?m -3.*:pv0.05 verss G-low.

4 320 L. BEIJER ET AL. ( ) pg?ml -1 in G-high grop (n=6) verss 4.2 ( ) pg?ml -1 in G-low grop (n=5), pv0.01). Stimlation of the PBMCs with lipopolysaccharide broght abot a marked increase in TNF-a secretion bt no difference was fond between the G-low and G-high grops. Median secretion of IFN-c was higher and secretion of IL-4 lower from PHA-stimlated PBMCs in the G-high grop compared to the G-low grop, bt no significant differences between the grops were fond (IFN-c: 1,305 (163 6,520) pg?ml -1 in G-low grop verss 1,783 (202 10,935) pg?ml -1 in G-high grop; IL-4: 7.3 ( ) pg?ml -1 in G-low grop verss 6.6 ( ) pg?ml -1 in G-high grop). The median CV between dplicate samples for IFN-c was 34% (maximm 129%) in the G-low grop and 27% (maximm 133%) in the G-high grop. For IL-4 the CV was 33% (maximm 126%) in the G-low grop and 43% (maximm 112%) in the G-high grop. The consistent relationship between the grops motivated an analysis of the ratio between the two cytokines, as an assessment of the balance of T-helper cell (Th) type 1- and 2-associated activation. There was a nonsignificant tendency for a higher IFN-c/IL-4 ratio in the G-high grop compared to the G-low grop (228 (86 675) verss 129 (54 554)). When nonatopic sbjects were analysed separately, the IFN-c/IL-4 ratio was significantly higher in the G-high grop compared to the G-low grop (247 ( ) verss 126 (34 514), pv0.05). A significant relationship was also fond between the IFN-c/IL-4 ratio and the amont of b(1r3)-d-glcan measred in the home (r=0.47, pv0.05). Figre 1 illstrates this relationship, exclding one sbject with a very high amont of b(1r3)-d-glcan at home (51.7 ng?m -3, IFN-c/IL-4=172; r=0.50, pv0.05). No significant differences were fond between the G-high and G-low grops regarding secretion of IL-10 or IL-1b from PBMCs, ECP or MPO levels in serm, differential cell conts in blood, or the prevalence of symptoms. IFN-g/IL b(1 3)-D-glcan in homes ng m -3 Fig. 1. Relationship between home b(1r3)-d-glcan levels and the ratio between interferon gamma (IFN-c) and interlekin(il)-4 secretion from phytohaemaggltinin-stimlated peripheral blood mononclear cells in nonatopic sbjects, exclding one sbject with a very high amont of b(1r3)-d-glcan at home (51.7 ng?m -3, IFN-c/IL-4=172). Discssion The recrited persons represent a biased poplation sample as they were aware of problems related to mold exposre at home or answered an advertisement reqesting the participation of persons with a mold problem at home. The poplation recrited from the previos stdy [10] reported a slightly higher prevalence of some symptoms sch as those involving the nose and central nervos system-related symptoms as compared to the overall poplation in the previos stdy. For other symptoms sch as throat irritation and cogh, the prevalences were similar. Regarding the poplation recrited by advertising, it is reasonable to presme a higher prevalence of symptoms in this grop compared to a random poplation sample. The assignment into exposre grops, however, was made according to measrements of b(1r3)-d-glcan in the sbjects9 homes and not according to sbjective reporting of mold problems at home. There was a similar distribtion of sbjects into the G-high and G-low grops from the two recritment grops (see Methods section) and there were no differences in symptom prevalence between the G-high and G-low grops. Taken together, it is ths nlikely that a selection bias according to symptoms or rote of recritment acconted for the differences fond in inflammatory marker levels in the blood between the G-low and G-high grops. The reslts sggest an effect on lymphocytes as the G-high grop showed a lower proportion of cytotoxic CD8zT-cells in their blood. A previos stdy showed that only CD8zT-cells expressing S6F1 cold migrate throgh an endothelial layer [18], and another stdy demonstrated that CD8zS6F1z T-cells exhibited a cytotoxic effect [17]. The redced proportion of these cells in blood from the G-high grop cold ths be the reslt of increased trapping in the lng de to inflammatory activity in the airways. Indeed, FIREMAN et al. [19] fond an inverse relationship in the nmber of CD8zS6F1z T-cells when they compared blood and bronchoalveolar lavage flid from patients with idiopathic plmonary fibrosis. The relationship between mold exposre at home and effects on blood lymphocytes has been explored in a previos stdy [11]. This stdy examined children, and, contrary to the present reslts, the data showed a slightly redced nmber of CD4z T-cells and slightly increased nmber of CD8z T-cells, reslting in a lower CD4z/CD8z ratio among mold-exposed children. The reason for this discrepancy is not clear; differences in exposre or in the reaction of adlts and children are possible explanations. Some stdies have shown a relationship between living in a hose with mold and/or dampness and increased risk of atopy in children [12 14]. This relationship is spported by data from an experimental stdy in mice, where exposre to b(1r3)-d-glcan indced increased expression of messenger riboncleic acid encoding IL-10, a Th2-driving cytokine, in lng tisse cells, whereas expression of IL-12, promoting IFN-c prodction, was depressed [20]. In the present stdy, however, the nonatopic G-high grop exhibited a higher IFN-c/IL-4 ratio than the nonatopic G-low grop and this ratio correlated with the level of

5 MOULD AND INFLAMMATION 321 b(1r3)-d-glcan in the home. This sggests that, in nonatopic sbjects, b(1r3)-d-glcan exposre at home is not associated with a Th2 pattern of cytokine secretion. The reason why a different ratio between the G-high and G-low grops was fond only among nonatopic sbjects cold be that, among atopic sbjects, secretion of IFN-c as well as IL-4 has a tendency to be increased in the G-high grop compared to the G-low grop. In the present stdy, effects on secretion of TNF-a from nonstimlated PBMCs and changes in blood T-cell sbsets and the IFN-c/IL-4 ratio in relation to b(1r3)-d-glcan levels in the sbjects9 homes were fond. In accordance with a previos stdy in an area of terraced hosing [10], the G-high grop exhibited a higher concentration of MPO in their serm than the G-low grop, bt the difference was not significant. For the other inflammatory markers stdied, no significant effects were fond. This sggests that there is no massive inflammatory response among sbjects exposed to b(1r3)-d-glcan in their homes, at the levels present in this stdy, a conclsion spported by the lack of an association between the extent of reported symptoms and b(1r3)-d-glcan levels in the home. The relationships between airborne b(1r3)-d-glcan and inflammatory cytokines do not necessarily imply that mold is the casative agent. The conditions reqired for mold growth also promote the growth of other microorganisms and the formation of volatile organic componds from bilding material [21]. In previos stdies, a relationship between the amont of endotoxin from Gram-negative bacteria in floor dst and the severity of asthma was fond [22 24]. Endotoxin was not analysed in the present stdy since it is the athors9 experience from earlier stdies that airborne endotoxin levels in dwellings, as sampled sing the present techniqe, are generally below the detection limit of the assay sed (0.6 pg?m -3 ). Conversely, mold cells contain a nmber of biologically active agents. Mycotoxins from Mycobacterim mrale and Stachybotrys atra have been related to symptoms and disease [25, 26]. The cell wall sbstance b(1r3)-d-glcan itself has important effects on the immne system [27]. Frther work is needed to assess the role of these potentially inflammagenic agents from microorganisms in symptoms related to indoor exposres. In smmary, the present reslts sggest a relationship between living in homes with higher levels of b(1r3)-d-glcan and changes in some parts of the inflammatory and immnological systems. In nonatopic persons, a dose/response relationship was fond between the interferon gamma/interlekin-4 ratio and the amont of b(1r3)-d-glcan measred in the home. The increased tmor necrosis factor-a secretion from blood cells and redced proportion of cytotoxic CD8z T-cells in the blood sggest preglated inflammatory activity in the airways related to mold exposre. Acknowledgements. The athors thank D. Söderberg, G. Arvidsson, E. Kerekes and R.M. Olofsson for skilfl technical management. References 1. Platt SD, Martin CJ, Hnt SM, Lewis CW. Damp hosing, mold growth, and symptomatic health state. BMJ 1989; 298: Dijkstra L, Hothijs D, Brnekreef B, Akkerman I, Boleij JSM. Respiratory health effects of the indoor environment in a poplation of Dtch children. Am Rev Respir Dis 1990; 142: Brnekreef B. Associations between qestionnaire reports of home dampness and childhood respiratory symptoms. Sci Total Environ 1992; 127: Jaakkola JJK, Jaakkola N, Rotsalainen RH. Home dampness and molds as determinants of respiratory symptoms and asthma in pre-school children. J Expo Anal Environ Epidemiol 1993; 3: H FB, Persky V, Flay BR, Richardson J. An epidemiological stdy of asthma prevalence and related factors among yong adlts. J Asthma 1997; 34: Williamson IJ, Martin CJ, McGill G, Monie RD, Fennerty AG. Damp hosing and asthma: a casecontrol stdy. Thorax 1997; 52: Rylander R. Sick bilding syndrome. In: Basomba A, Sastre J, eds. XVI Eropean Congress of Allergology and Clinical Immnology. Mondzzi Editore, Bologna, 1995; pp Dowes JA, Zidhof A, Doekes G, et al. (1R3)-b-Dglcan and endotoxin in hose dst and peak flow variability in children. Am J Respir Crit Care Med 2000; 162: Hirvonen MR, Rotsalainen M, Roponen M, et al. Nitric oxide and proinflammatory cytokines in nasal lavage flid associated with symptoms and exposre to moldy bilding microbes. Am J Respir Crit Care Med 1999; 160: Thorn J, Rylander R. Airways inflammation and glcan in a rowhose area. Am J Respir Crit Care Med 1998; 157: Dales R, Miller D, White J, Dlberg C, Lazarovits A. Inflence of residential fngal contamination on peripheral blood lymphocyte poplations in children. Arch Environ Health 1998; 53: Schafer T, Kramer U, Dockery D, Vielf D, Behrendt H, Ring J. What makes a child allergic? Analysis of risk factors for allergic sensitization in preschool children from East and West Germany. Allergy Asthma Proc 1999; 20: Garrett MH, Rayment PR, Hooper MA, Abramason MJ, Hooper BM. Indoor airborne fngal spores, hose dampness and associations with environmental factors and respiratory health in children. Clin Exp Allergy 1998; 28: Savilahti R, Uitti J, Roto P, Laippala P, Hsman T. Increased prevalence of atopy among children exposed to mold in a school bilding. Allergy 2001; 56: Rylander R, Persson K, Goto H, Yasa K, Tanaka S. Airborne b,1 3 glcan may be related to symptoms in sick bildings. Indoor Environ 1992; 1: Tamra H, Arimoto Y, Tanaka S, Yoshida T, Obayashi T, Kawai T. Atomated kinetic assay for endotoxin and (1R3)-b-D-glcan in hman blood. Clin Chim Acta 1994; 226: Morimoto C, Rdd CE, Letvin NL, Schlossman SF. A novel epitope of LFA-1 antigen which can

6 322 L. BEIJER ET AL. distingish killer effector and sppressor cells in hman CD8 cells. Natre 1987; 330: Berman JS, Mahoney K, Sakkonen JJ, Masyama J. Migration of distinct sbsets of CD8z blood T cells throgh endothelial cell monolayers in vitro. J Lekoc Biol 1995; 58: Fireman E, Vardinon N, Brke M, et al. Predictive vale of response to treatment of T-lymphocyte sbpoplations in idiopathic plmonary fibrosis. Er Respir J 1998; 11: Wan G-H, Li C-S, Go S-P, Rylander R, Lin R-H. An airborne mold-derived prodct, b-1,3-d-glcan, potentiates airway allergic responses. Er J Immnol 1999; 29: Korpi A, Pasanen AL, Pasanen P. Volatile componds originating from mixed microbial cltres on bilding materials nder varios hmidity conditions. Appl Environ Microbiol 1998; 64: Michel O, Ginanni R, Dchatea J, Vertongen F, Le Bon B, Sergysels R. Domestic endotoxin exposre and clinical severity of asthma. Clin Exp Allergy 1991; 21: Michel O, Kips J, Dchatea J, et al. Severity of asthma is related to endotoxin in hose dst. Am J Respir Crit Care Med 1996; 154: Rizzo MC, Naspitz CK, Fernández-Caldas E, Lockey RF, Mimiça I, Solé D. Endotoxin exposre and symptoms in asthmatic children. Pediatr Allergy Immnol 1997; 8: Andersson MA, Niklin M, Koljalg U, et al. Bacteria, molds, and toxins in water-damaged bilding materials. Appl Environ Microbiol 1997; 63: Etzel RA, Montaña E, Sorenson WG, Kllman GJ, Allan TM, Dearborn DG. Acte plmonary hemorrhage in infants associated with exposre to Stachybotrys atra and other fngi. Arch Pediatr Adolesc Med 1998; 152: Rylander R, Lin R-H. (1R3)-b-D-Glcan relationship to indoor air-related symptoms, allergy and asthma. Toxicology 2000; 152: