Association between Heat Shock Protein 70/Hom Genetic Polymorphisms and Uveitis in Patients with Sarcoidosis

Abstract

purpose. The predisposition to sarcoidosis, a systemic granulomatous disorder of unknown etiology, is genetically determined, and genetics appears also to drive the disease down distinct phenotypic pathways. This study was undertaken to test the hypothesis that sarcoidosis-related uveitis represents a genetically distinct disease subset, by investigating single nucleotide polymorphisms (SNPs) in the HSP-70/1 and HSP-70/Hom genes. HSP70 molecules play a key role in the immune response by functioning both as chaperones and as inducers of proinflammatory cytokine secretion.

methods. By sequence-specific primers–polymerase chain reaction (SSP-PCR) five SNPs were evaluated in 270 white patients with sarcoidosis, including 88 with sarcoid-related uveitis, and in 347 matched control subjects. One hundred twenty-five patients with idiopathic anterior uveitis (IAU) and 56 with idiopathic intermediate uveitis (IIU) were also included in the study as disease control subjects.

results. The HSP-70/Hom rs2075800 G allele frequency was higher in the sarcoid-uveitis group than in both the sarcoid-non-uveitis and control groups (83% vs. 71%, OR = 2.00, P _c = 0.01; and 83% vs. 66%, OR = 2.45, P _c = 0.00005, respectively). Similar results were observed when considering the carriage frequency of the associated haplotype (HSP-70 haplotype 2) across the three study groups (47% vs. 29%, OR = 2.17, P _c = 0.03; and 47% vs. 21%, OR = 3.26, P _c = 0.0003, respectively). In addition, the carriage frequency of the HSP-70 haplotype 2 discriminated among sarcoid-related uveitis, IAU, and IIU (47% vs. 19%, OR = 3.26, P _c = 0.001; and 47% vs. 23%, OR = 2.81, P _c = 0.04, respectively).

conclusions. A strong association was found between HSP-70/Hom rs2075800 G and uveitis in patients with sarcoidosis. Further studies are needed to understand the molecular mechanisms underlying this association.

Sarcoidosis is a systemic inflammatory disorder characterized by the accumulation of CD4⁺ T-lymphocytes and macrophages, in association with noncaseating epithelioid granulomas, in affected organs. ^{1 2} The cause remains uncertain, but it is commonly accepted that the disease results from a complex interaction between environmental factors and host-specific genetic profiles. Genes likely to be pathogenetic may have major or minor effects. ^{3 4 5}  

Based on clinical and outcome data, it is often argued that sarcoidosis is not a single entity but a “family” of clinically and genetically different phenotypes. ^{3 6 7 8 9} Thus, sarcoidosis-related uveitis may represent a separate disease subset characterized by a specific genetic profile. Sarcoid uveitis is especially common in Japan, with more than half of patients having ocular disease, compared with one-third of patients or fewer in Europe. ¹⁰ In addition, ocular sarcoidosis is more prevalent in blacks than in whites, ¹¹ lending further support to a genetic susceptibility hypothesis. ³  

The heat shock proteins (HSP), a group of highly conserved proteins classified by molecular weight and grouped into families of different size, ¹² function as intracellular chaperones, controlling protein folding, unfolding, and translocation under physiological conditions and after physical or chemical stress. ¹³ Extracellular HSPs process and present immunogenic peptides to cytotoxic T cells, activate the innate immune system, and induce proinflammatory cytokine secretion from monocytes. ^{14 15} In humans, three members of the HSP-70 gene family have been mapped on 6p21.3 in the human leukocyte antigen (HLA) class III region. ¹⁶ HSP-70/1 (HSPA1A) and HSP-70/2 (HSPA1B) encode an identical protein, the major heat-inducible HSP-70, and differ only in their regulatory region, whereas HSP-70/Hom (HSPA1L) has an amino acid sequence almost identical to HSP-70 but is constitutively expressed. 

Single nucleotide polymorphisms (SNPs) in the HSP-70 genes are associated with several diseases. ^{17 18 19 20 21} However, the reported associations have often been inconclusive because of small patient samples and high and variable linkage disequilibrium (LD) in the HLA area, preventing assignation of the precise etiologic locus. Nonetheless, the importance of HSP70 in protecting cells from stress-induced apoptosis implicates this gene family in disease pathogenesis in sarcoidosis. ²² Given the importance of apoptosis in the pathogenesis of uveitis, we examined the hypothesis that sarcoid uveitis is associated with polymorphisms in HSP-70 genes using a cohort of phenotypically well-defined white patients and normal control subjects. Two sets of patients with idiopathic forms of uveitis were also included in the study as disease control subjects. 

Subjects and Methods

Clinical Cohorts

Two hundred seventy white patients, including 88 with clinical features of sarcoidosis-related uveitis, were included in the study. They were recruited from the Sarcoidosis Clinic of Royal Brompton Hospital (London, UK), a tertiary referral center taking patients mainly from the southeast of the United Kingdom. In all patients, sarcoidosis was diagnosed histologically and according to the criteria defined in the ATS/ERS/WASOG Statement on Sarcoidosis. ² Clinical features of the patients with sarcoidosis are summarized in Table 1 . 

Patients with ocular involvement were assessed at a specialist uveitis clinic (Moorfields Eye Hospital) by one of the authors (SL) for further evaluation. Twenty-seven (31%) patients presented with panuveitis, 22 (25%), with anterior uveitis, 22 (25%) with posterior uveitis, and 17 (19%) with intermediate uveitis. In addition, 34 patients (39%) were treated with either topical or systemic steroids. Full details of the ophthalmic examination were available for 62 (70%) of the patients with sarcoid-related uveitis, as some were more historic cases. Ocular examination revealed cystoid macular edema in 29 (47%) subjects, signs of vasculitis in 18 (29%), snowballs in 8 (13%), and optic nerve involvement in 6 (10%). Written consent was obtained from all patients. The Ethics Committees of the Royal Brompton Hospital gave authorization for the study. 

One hundred twenty-five subjects with idiopathic anterior uveitis (IAU) and 56 with idiopathic intermediate uveitis (IIU) were also investigated. They were all collected from the Uveitis Clinic (SL) at Moorfields Eye Hospital. Anterior uveitis was defined as inflammation confined to the anterior chamber and/or anterior vitreous, whereas intermediate uveitis was defined as inflammation involving vitreous, pars plana, and peripheral retina. ^{23 24} Idiopathic disease was identified when there were no symptoms, signs, or investigation results suggestive of an alternative diagnosis. Therefore, patients with uveitis of infectious etiology or uveitis associated with known systemic disease (e.g., multiple sclerosis, ankylosing spondylitis, Reiter syndrome, Behçet disease, or inflammatory bowel disease) were not included in the study. The clinical phenotype of the patients was recorded at the time of blood collection. After verbal and written informed consent was obtained, demographic details and detailed medical and ocular histories were recorded for each of the patients. All patients had a complete ocular examination, including best corrected visual acuity, slit lamp examination, applanation tonometry, and dilated fundus examination. Fluorescein angiography or optical coherence tomography was performed if macular edema was considered likely on clinical examination. The Ethics Committees of Moorfields Eye Hospital gave permission for the study and it was adherent to the tenets of the Declaration of Helsinki. 

Control Subjects

The control population included 347 subjects collected from the southeast of the United Kingdom and matched, as closely as possible, by sex, age, and ethnicity. They were selected at random from 10,000 healthy individuals attending general practice health-screening clinics. ²⁵ They were all healthy, as judged by regular checks (including medical history, physical examination, and routine laboratory blood testing). 

Sequence-Specific Primers and Polymerase Chain Reaction

Single nucleotide polymorphisms (SNPs) were determined using sequence-specific primers (SSPs) and polymerase chain reaction (PCR) that utilizes SSPs with 3′-end mismatches and identifies the presence of specific allelic variants through PCR amplification. Patients and control subjects were also genotyped for the HLA-B27 allele, as previously described. ²⁶ All PCR reactions were performed in identical conditions. ²⁶  

To minimize the genotyping error rate, all samples identified as possessing rare haplotypes (frequency <1%) were regenotyped. 

The primer sequences used to detect HSP-70 polymorphisms are shown in Table 2 . 

Data Analysis

Genotype and allele frequencies were determined by direct counting. The haplotypes were deduced by PHASE, version 2, a statistical haplotype-reconstruction method. ²⁷  

Data were analyzed by χ² contingency tables or Fisher’s exact test, as appropriate. Adjustment for multiple tests was made using the formula P _c = P × n, where P _c is the corrected value, P is the uncorrected value, and n is the number of tests performed (Bonferroni method). P < 0.05 was considered significant. An odds ratio (OR) with 95% confidence interval (CI) was also calculated. Statistical analyses were performed with commercial software (SPSS, ver. 11.5; SPSS Inc., Chicago, IL). 

Throughout the article, the deduced haplotypes encompassing polymorphisms in both HSP-70/1 and HSP-70/Hom genes are referred to as HSP-70 haplotypes. 

According to normal convention, HLA data and HSP-70 haplotypes are analyzed by carriage frequency, whereas HSP-70/1 and HSP-70/Hom allele data are analyzed by allele frequency, unless otherwise stated. 

Results

Five biallelic SNPs were investigated: rs1008438 A/C (HSP-70/1, promoter), rs1043618 C/G (HSP-70/1, promoter), rs1043620 C/T (HSP-70/1, exon, coding sequence [no amino change]), rs2227956 T/C (HSP-70/Hom, exon, coding sequence [Met/Thr]), rs2075800 G/A (HSP-70/Hom, exon, coding sequence [Gly/Lys]), according to the publicly available SNP database. ²⁸ Seven haplotypes were deduced, including two rare haplotypes (haplotypes 6, 7). The computer program PHASE was run separately for patients and control subjects and revealed identical HSP-70 haplotype structure. For each SNP, the proportion of heterozygous and homozygous genotypes was consistent with Hardy-Weinberg equilibrium in cases and control subjects. 

HSP-70/1 and HSP-70/Hom Alleles and Haplotypes Associated with Sarcoidosis as a Whole

Analysis of the whole sarcoidosis cohort by SNP showed a significant increase in the rs1043620 C allele frequency in the patient group compared with the frequency in control subjects (91% vs. 85%, OR = 1.78, 95% CI = 1.25–2.53, P _c = 0.006; Table 3 ). The G allele rs2075800 was also overrepresented in the sarcoidosis group compared with its presence in control subjects (75% vs. 66%, OR = 1.49, 95% CI = 1.16–1.92, P _c = 0.007; Table 3 ). Additional investigated variants rs1008438, rs1043618, and rs2227956 showed no differences in genotype, phenotype, and allele frequency between patients and control subjects. The genotype frequencies of the investigated HSP-70/1 and HSP-70/Hom polymorphisms for patients and control subjects are summarized in Table 4 . 

Further analysis by haplotypes demonstrated that the HSP-70 haplotype 2 carriage frequency was significantly overrepresented in the patient group compared with its frequency in control subjects (34% vs. 21%, OR = 1.97, 95% CI = 1.37–2.82, P _c = 0.003; Table 5 ). By contrast, haplotype 3 was significantly overrepresented in the control group (43% vs. 56%, OR = 0.59, 95% CI = 0.43–0.82, P _c = 0.01; Table 5 ). In addition, haplotype 4 carriage frequency was significantly lower in the patient group (18% vs. 28%, OR = 0.57, 95% CI = 0.39–0.84, P _c = 0.03; Table 5 ), whereas the carriage frequency of haplotype 5 was significantly increased in the sarcoidosis group compared with its frequency in control subjects (16% vs. 7%, OR = 2.32, 95% CI = 1.39–3.87, P _c = 0.01; Table 5 ). 

HSP-70/1 and HSP-70/Hom Alleles and Haplotypes Associated with Sarcoid-Related Uveitis

Patients with sarcoidosis were subsequently divided in two groups according to the presence or absence of ocular involvement. Analysis by SNPs showed that the rs1043620 C allele was significantly overrepresented in the non-uveitis group compared with its presence in control subjects (91% vs. 85%, OR = 1.81, 95% CI = 1.20–2.73, P _c = 0.02; Table 6A ). In contrast, the rs2075800 G allele frequency was significantly higher in the sarcoid uveitis group compared with its frequency in both sarcoid non-uveitis (83% vs. 71%, OR = 2.00, 95% CI = 1.28–3.15, P _c = 0.01) and control subjects (83% vs. 66%, OR = 2.45, 95% CI = 1.61–3.72, P _c = 0.00005; Table 6A ). None of the other investigated SNPs revealed differences in genotype, phenotype, and allele frequency across the three groups. In addition, in the sarcoid uveitis cohort subgroup analysis revealed no genotype–phenotype associations with any of the investigated SNPs. The genotype frequencies of the investigated HSP-70/1 and HSP-70/Hom polymorphisms in patients and control subjects are summarized in Table 7 . 

Analysis by haplotypes showed that HSP-70 haplotype 2 carriage frequency was significantly increased in the sarcoid uveitis group compared with the frequency in both sarcoid non-uveitis (47% vs. 29%, OR = 2.17, 95% CI = 1.28–3.67, P _c = 0.03) and control subjects (47% vs. 21%, OR = 3.26, 95% CI = 2.0–5.32, P _c = 0.0003; Table 6B ). By contrast, HSP-70 haplotype 3 carriage frequency was significantly underrepresented in the sarcoid uveitis group compared with its presence in both control subjects (28% vs. 56%, OR = 0.31, 95% CI = 0.19–0.53, P _c = 0.0006) and the sarcoid non-uveitis group (28% vs. 49%, OR = 0.41, 95% CI = 0.24–0.71, P _c = 0.01; Table 6B ). In addition, HSP-70 haplotype 5 carriage frequency was significantly higher in the sarcoid uveitis group than in control subjects (18% vs. 7%, OR = 2.76, 95% CI = 1.42–5.37, P _c = 0.02; Table 6B ). No other significant differences in the HSP-70 haplotype carriage frequency were seen. 

In the sarcoid uveitis cohort, no associations were observed between disease phenotypes and any of the deduced haplotypes. However, our study was not sufficiently powered to perform a subgroup analysis on genotype/phenotype. 

HSP-70/1 and HSP-70/Hom Alleles and Haplotypes Associated with Sarcoidosis-Related Uveitis and Idiopathic Uveitis

To determine whether the HSP-70 association was specific to sarcoid-related uveitis, for further analysis we considered only patients with sarcoidosis with ocular involvement and compared them with patients with idiopathic forms of uveitis—namely, idiopathic anterior uveitis (IAU) and idiopathic intermediate uveitis (IIU). The rs2075800 G allele was strongly overrepresented in the sarcoid-uveitis group compared with its presence in both IAU (83% vs. 62%, OR = 3.00, 95% CI = 1.88–4.78, P _c = 0.00001) and control subjects (83% vs. 66%, OR = 2.45, 95% CI = 1.61–3.72, P _c = 0.00008; Table 8A ). In addition, in the IAU group an increase in the rs1043620 C allele frequency was observed compared with that in control subjects (94% vs. 85%, OR = 2.78, 95% CI = 1.60–4.84, P _c = 0.0007; Table 8A ). 

Analysis by haplotype showed an increased carriage frequency of the HSP-70 haplotype 2 in the sarcoid uveitis group compared with its frequency in both IAU (47% vs. 19%, OR = 3.62, 95% CI = 1.97–6.64, P _c = 0.001) and IIU (47% vs. 23%, OR = 2.81, 95% CI = 1.34–5.89, P _c = 0.04; Table 8B ). By contrast, HSP-70 haplotype 3 was underrepresented in the sarcoid uveitis group compared with its presence in both IAU (28% vs. 60%, OR = 0.27, 95% CI = 0.15–0.48, P _c = 0.0006) and IIU (28% vs. 52%, OR = 0.37, 95% CI = 0.19–0.75, P _c = 0.04; Table 8B ). In addition, HSP-70 haplotype 4 carriage frequency was lower in the IAU group than in control subjects (12% vs. 28%, OR = 0.35, 95% CI = 0.20–0.63, P _c = 0.005; Table 8B ). 

No other differences in the HSP-70 haplotype carriage frequency were seen. 

HLA-B27 Carriage Frequency in Sarcoidosis-Related Uveitis, Non-uveitis Sarcoidosis, IAU, and IIU Groups and in Control Subjects

HLA-B27-associated anterior uveitis is a distinct clinical entity with wide-ranging medical significance due to its ocular, systemic, immunologic, and genetic features. In addition, it is common, and is associated with major ocular morbidity due to its typically recurrent attacks of inflammation. ^{29 30} To exclude the possibility that the HSP70 associations seen in this study were secondary to linkage disequilibrium with HLA-B27, we determined whether carriage of the HLA-B27 allele contributed to ocular disease in sarcoidosis. No association with sarcoidosis-related uveitis was identified (sarcoid uveitis 2%, sarcoid non-uveitis 7%, control subjects 9%), whereas an increased HLA-B27 carriage frequency in the IAU (68%; OR = 21.95, P < 0.0001 vs. control subjects; OR = 91.54, P < 0.0001 vs. sarcoid uveitis; OR = 7.72, P < 0.0001 vs. IIU), and IIU (22%, OR = 2.84, P = 0.005 vs. control subjects; OR = 11.86, P = 0.0002 vs. sarcoid uveitis) groups was observed, as expected (Table 9) . 

Subsequently, to verify a potential HLA-B27 allele effect on the association between HSP-70 and sarcoidosis-related uveitis, we performed multivariate logistic regression. The HSP-70 haplotype 2 remained significantly associated with sarcoid uveitis compared with sarcoid non-uveitis (OR = 1.84, 95% CI = 1.06–3.17, P = 0.03), IAU (OR = 3.9, 95% CI = 1.62–9.41, P = 0.002), IIU (OR = 2.67, 95% CI = 1.20–5.91, P = 0.015) and control subjects (OR = 3.23, 95% CI = 1.91–5.44, P < 0.0001), after adjustment for the B27 allele carriage. 

Discussion

In this study, we showed that a non-synonymous variation in HSP-70/Hom (rs2075800 G) and its related haplotype (HSP-70 haplotype 2) is associated with uveitis in patients with sarcoidosis. In addition, we used individuals with idiopathic forms of uveitis (IAU and IIU) as disease control subjects and demonstrated that the HSP-70 association is specific to sarcoid uveitis. Furthermore, because of the known associations between HLA-B27 and uveitis, ³⁰ we performed logistic regression analysis and confirmed that HSP-70/Hom rs2075800 G is a risk factor for sarcoid uveitis independent of HLA-B27. These results suggest that, because uveitis is likely to result from a variety of different immunopathogenetic processes, the predisposing genes vary considerably, depending on the underlying disease. 

HSP-70 haplotype 2—tagged by the rs2075800 G allele—is associated with sarcoid uveitis; by contrast, the HSP-70 haplotype 3—tagged by the rs2075800 A allele—is protective. Because haplotypes 2 and 3 differ only in the rs2075800 allele, we speculate that the phenotypic effect is attributable to this non-synonymous variant rather than to the associated haplotype. Nevertheless, because of the strong and variable LD existing across the HLA region, we cannot exclude that the association between HSP-70/Hom and sarcoid uveitis is secondary to closely linked polymorphisms. 

The HSP70/Hom rs2075800 A allele has been reported to be associated with rheumatoid arthritis (RA). ¹⁹ HSP-70/Hom rs2227956 has been recently associated with both Löfgren’s syndrome, a benign form of sarcoidosis, and non-Löfgren sarcoidosis in a cohort of Polish patients. ²¹ Our data set did not include patients presenting with classic Löfgren’s syndrome. However, we were unable to reproduce similar findings in our cohort of non-Löfgren sarcoidosis (n = 270). This discrepancy is most likely due to the small number of patients in the Polish study (Löfgren’s syndrome n = 13, non-Löfgren sarcoidosis n = 29), rather than ethnic variation. Indeed, the allelic frequency of the mutant C allele was not dissimilar between the U.K. and Polish control groups (19% and 12%, respectively). 

Because of the lack of functional studies, the biological consequences of the association between HSP70/Hom rs2075800 and sarcoidosis-related uveitis can only be speculated. The G-to-A polymorphism rs2075800 leads to a novel amino acid at residue 602 of HSP70/Hom. According to a model structure generated from the corresponding region of Escherichia coli DnaK, ³¹ human HSP-70/Hom residue 602 corresponds to position 592 of DnaK, an α-helix within the C-terminal domain that modulates substrate binding and interaction with co-chaperones. This non-synonymous SNP may affect the expression and functional efficiency of HSP-70/Hom, thus causing the accumulation of abnormal proteins within the cell. 

A perturbation in protein folding and assembly is only one of the numerous potential mechanisms through which HSP-70 can lead to ocular involvement in sarcoidosis. One intriguing hypothesis is that the HSP-70/Hom associated variant modulates risk of sarcoid uveitis, not because it affects the HSP-70 chaperoning ability, but because it influences the extent of the uveal inflammatory process. HSP-70 acts as a very early co-inducer of a non-specific proinflammatory response by stimulating monocytic cells to produce IL-1, IL-6, and TNF-α. ¹⁴ In addition, HSP-70 induces the production of IL-12, hence priming T-cells to produce more IFN-γ and TNF-α, which have been demonstrated to play pivotal roles in anterior uveitis immunopathogenesis. ³² Whereas HSP-70/1 and HSP-70/2 are stress-inducible genes, HSP-70/Hom is constitutively expressed in various organs, including the eye. ³³ Therefore, altered expression of HSP70Hom may influence the ability of the cell to withstand inflammation or alter the apoptotic threshold of cell within the eye. 

In conclusion, our results show that the HSP70/Hom rs2075800 G allele (or other genes in LD in the HLA region) is associated with the presence of uveitis in patients with sarcoidosis and further strengthen the prevailing hypothesis that the broad variety of sarcoidosis phenotypes is accounted for, at least in part, by specific genetic profiles. This association is specific to sarcoidosis, as the carriage of the rs2075800 G allele discriminates between sarcoidosis-related and idiopathic forms of uveitis. However, a major limitation of this and many other studies of gene polymorphisms is that the understanding of the biological relevance of SNPs in the coding sequence is still incomplete. Our study also highlights the importance of having made a precise phenotypic classification, as genetic studies are more likely to succeed if a very specific disease phenotype is addressed. Nevertheless, the association between HSP-70/Hom rs2075800 and sarcoidosis-related uveitis must be verified in populations of different ethnicities. 

²  Contributed equally to the work and therefore should be considered equivalent authors.

 Supported by the Asmarley Charitable Trust.

 Submitted for publication December 15, 2006; revised February 27 and March 2, 2007; accepted April 27, 2007.

 Disclosure: P. Spagnolo, None; H. Sato, None; S.E. Marshall, None; K.M. Antoniou, None; T. Ahmad, None; A.U. Wells, None; M.A. Ahad, None; S. Lightman, None; R.M. du Bois, None; K.I. Welsh, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Clinical Features of Patients with Sarcoidosis

	Sarcoidosis Patients (n = 270)
Gender	Male 125 (46%)
	Female 145 (54%)
Age at diagnosis	38.2 (16–75)
Smoking history	96/240 (40%)
Treatment	195/240 (81%)
Major organ involvement*	58/240 (24%)
Skin involvement	40/240 (17%)
Bronchoalveolar lavage (BAL), †
BAL Macrophages	69.5 (16–97)
BAL Lymphocytes	18.8 (2–73)
BAL Neutrophils	2.3 (0–70)
Lung function (% predicted)
FEV1	88.5 (25–132)
FVC	94.5 (40–141)
DLCO	76.7 (23–121)
Radiologic stage of disease
0	32 (13%)
1	42 (18%)
2	49 (20%)
3	20 (9%)
4	97 (40%)

 Clinical details were available for 240 patients. Data are given as absolute numbers with percentages in parentheses. For age at diagnosis, BAL cellularity and lung function data medians are indicated in parentheses.

*  Major organ involvement includes neurologic, cardiac, renal or liver disease.

†  BAL data was available for 87 patients.

Table 2.

 

Primer Sequences Used for PCR-SSP Detection of HSP-70/1 and HSP-70/Hom Polymorphisms

Allele	Locus	Sense Primer 5′-3′	Antisense Primer 3′-5′	Amplicon Size (bp)
rs1008438 C	Promoter	CAG GAC GGG AGG CGA AAC	CAC AGG TTC GCT CTG GGA A	219
rs1008438 A		CAG GAC GGG AGG CGA AAA	CAC AGG TTC GCT CTG GGA A	219
rs1043618 C	Promoter	GCT CGG TGA TTG GCT CAG AA	CTG CTC TCT GTC GGC TCG	282
rs1043618 G		GCT CGG TGA TTG GCT CAG AA	CTG CTC TCT GTC GGC TCC	282
rs1043620 C	Exon, coding sequence	TTG ACG CGA AGC GGC TGA TC	ATG GAC GTG TAG AAG TCG ATG	688
rs1043620 T		TTG ACG CGA AGC GGC TGA TT	ATG GAC GTG TAG AAG TCG ATG	688
rs2227956 T	Exon, coding sequence	ACC TCC AGC CAC GAA AGG A	GGT ATT CTC AAT GTC ACA GCC AT	299
rs2227956 C		ACC TCC AGC CAC GAA AGG A	GTA TTC TCA ATG TCA CAG CCA C	298
rs2075800 G	Exon, coding sequence	GAT AGG GTT ACA CAT CTG CTC	CCT GAG CAA GGA GGA GAT TG	292
rs2075800 A		TGA TAG GGT TAC ACA TCT GCT T	CCT GAG CAA GGA GGA GAT TG	293

Table 3.

 

HSP-70/1 and HSP-70/Hom Allelic Frequencies

SNP Position	Allele	Sarcoidosis (n = 270)	Controls (n = 347)	OR
HSP-70/1	C	214 (40%)	246 (35%)	1.19
rs1008438	A	326 (60%)	448 (65%)
HSP-70/1	C	210 (39%)	242 (35%)	1.19
rs1043618	G	330 (61%)	452 (65%)
HSP-70/1	C	490 (91%)*	587 (85%)	1.78
rs1043620	T	50 (9%)	107 (15%)
HSP-70/Hom	T	446 (83%)	560 (81%)	1.13
rs2227956	C	94 (17%)	134 (19%)
HSP-70/Hom	G	403 (75%), †	460 (66%)	1.49
rs2075800	A	137 (25%)	234 (34%)

*   P _c = 0.006; 95% CI, 1.25–2.53.

†   P _c = 0.007; 95% CI, 1.16–1.92.

Table 4.

 

HSP-70/1 and HSP-70/Hom Polymorphisms Genotype Frequencies

SNP Position	Genotype	Sarcoidosis (n = 270)	Controls (n = 347)
HSP-70/1	C/C	38 (14%)	51 (15%)
rs1008438	C/A	138 (51%)	144 (41%)
	A/A	94 (35%)	152 (44%)
HSP-70/1	C/C	37 (14%)	51 (15%)
rs1043618	C/G	136 (50%)	140 (40%)
	G/G	97 (36%)	156 (45%)
HSP-70/1	C/C	220 (81%)*	248 (72%)
rs1043620	C/T	50 (19%)	91 (26%)
	T/T	0	8 (2%)
HSP-70/Hom	T/T	180 (67%)	224 (65%)
rs2227956	T/C	86 (32%)	112 (32%)
	C/C	4 (1%)	11 (3%)
HSP-70/Hom	G/G	153 (57%), †	153 (44%)
rs2075800	G/A	97 (36%)	154 (44%)
	A/A	20 (7%)	40 (12%)

*   P _c = 0.02.

†   P _c = 0.009.

Table 5.

 

HSP-70 Haplotype Carriage Frequencies

Haplotype	rs1008438	rs1043618	rs1043620	rs2227956	rs2075800	Sarcoidosis (n = 270)	Controls (n = 347)	OR
1	C	C	C	T	G	169 (63%)	190 (55%)	1.38
2	A	G	C	T	G	93 (34%)*	73 (21%)	1.97
3	A	G	C	T	A	115 (43%), †	193 (56%)	0.59
4	A	G	T	C	G	50 (18%), ‡	99 (28%)	0.57
5	A	G	C	C	G	43 (16%), §	26 (7%)	2.32
6	C	C	C	T	A	5 (2%)	1 (0%)	NS
7	C	G	C	T	G	4 (1%)	4 (1%)	NS

*   P _c = 0.003, 95% CI, 1.37–2.82.

†   P _c = 0.01, 95% CI, 0.43–0.82.

‡   P _c = 0.03, 95% CI, 0.39–0.84.

§   P _c = 0.01, 95% CI, 1.39–3.87.

Table 6.

 

HSP-70/1 and HSP-70/Hom Allelic Frequencies and Haplotype Carriage Frequencies

A. Allelic Frequencies
SNP Position	Allele	Sarcoid Non-uveitis (n = 182)	Sarcoid Uveitis (n = 88)	Controls (n = 347)
HSP-70/1	C	146 (40%)	68 (39%)	246 (35%)
rs1008438	A	218 (60%)	108 (61%)	448 (65%)
HSP-70/1	C	144 (40%)	66 (38%)	242 (35%)
rs1043618	G	220 (60%)	110 (63%)	452 (65%)
HSP-70/1	C	331 (91%)*	159 (90%)	587 (85%)
rs1043620	T	33 (9%)	17 (10%)	107 (15%)
HSP-70/Hom	T	303 (83%)	143 (81%)	560 (81%)
rs2227956	C	61 (17%)	33 (19%)	134 (19%)
HSP-70/Hom	G	257 (71%)	146 (83%), † , ‡	460 (66%)
rs2075800	A	107 (29%)	30 (17%)	234 (34%)

B. Haplotype Carriage Frequencies
Haplotype	Sarcoid Non-uveitis (n = 182)	Sarcoid Uveitis (n = 88)	Controls (n = 347)
1	116 (64%)	53 (60%)	190 (55%)
2	52 (29%)	41 (47%), § , ∥	73 (21%)
3	90 (49%)	25 (28%), ¶ , #	193 (56%)
4	33 (18%)	17 (19%)	99 (28%)
5	27 (15%)	16 (18%), **	26 (7%)
6	4 (2%)	1 (1%)	1 (0%)
7	2 (1%)	2 (2%)	4 (1%)

*   P _c = 0.02, OR: 1.81, 95% C.I.: 1.20–2.73, sarcoid non-uveitis vs. controls.

†   P _c = 0.01, OR: 2.00, 95% C.I.: 1.28–3.15, sarcoid uveitis vs. sarcoid non-uveitis.

‡   P _c = 0.00005, OR: 2.45, 95% C.I.: 1.61–3.72, sarcoid uveitis vs. controls.

§   P _c = 0.03, OR: 2.17, 95% C.I.: 1.28–3.67, sarcoid uveitis vs. sarcoid non-uveitis.

∥   P _c = 0.0003, OR: 3.26, 95% C.I.: 2.00–5.32, sarcoid uveitis vs. controls.

¶   P _c = 0.01, OR: 0.41, 95% C.I.: 0.24–0.71, sarcoid uveitis vs. sarcoid non-uveitis.

#   P _c = 0.0006, OR: 0.31, 95% C.I.: 0.19–0.53, sarcoid uveitis vs. controls.

**   P _c = 0.02, OR: 2.76, 95% C.I.: 1.42–5.37, sarcoid uveitis vs. controls.

Table 7.

 

HSP-70/1 and HSP-70/Hom Polymorphisms Genotype Frequencies

SNP Position	Genotype	Sarcoidosis Non-uveitis (n = 182)	Sarcoidosis Uveitis (n = 88)	Controls (n = 347)
HSP-70/1	C/C	26 (14%)	12 (14%)	51 (15%)
rs1008438	C/A	94 (52%)	44 (50%)	144 (41%)
	A/A	62 (34%)	32 (36%)	152 (44%)
HSP-70/1	C/C	25 (14%)	12 (14%)	51 (15%)
rs1043618	C/G	94 (52%)	42 (48%)	140 (40%)
	G/G	62 (35%)	34 (39%)	156 (45%)
HSP-70/1	C/C	149 (82%)*	71 (81%)	248 (71%)
rs1043620	C/T	33 (18%)	17 (19%)	91 (26%)
	T/T	0	0	8 (2%)
HSP-70/Hom	T/T	124 (68%)	56 (64%)	224 (65%)
rs2227956	T/C	55 (30%)	31 (35%)	112 (32%)
	C/C	3 (2%)	1 (1%)	11 (3%)
HSP-70/hom	G/G	90 (49%), †	63 (72%), §	153 (44%)
rs2075800	G/A	77 (42%), ‡	20 (23%), ∥	154 (44%)
	A/A	15 (8%)	5 (6%)	40 (12%)

*   P _c = 0.04 sarcoid non-uveitis vs. controls.

†   P _c = 0.003 sarcoid non-uveitis vs. sarcoid uveitis.

‡   P _c = 0.008 sarcoid non-uveitis vs. sarcoid uveitis.

§   P _c = 0.00002 sarcoid uveitis vs. controls.

∥   P _c = 0.001 sarcoid uveitis vs. controls.

Table 8.

 

HSP-70/1 and HSP-70/Hom Allelic Frequencies and Haplotype Carriage Frequencies in Sarcoid Uveitis, Idiopathic Anterior Uveitis (IAU), Idiopathic Intermediate Uveitis (IIU) and Controls

A. Allelic Frequencies
SNP Position	Allele	Sarcoid Uveitis (n = 88)	IAU (n = 125)	IIU (n = 56)	Controls (n = 347)
HSP-70/1	C	68 (39%)	102 (41%)	45 (40%)	246 (35%)
rs1008438	A	108 (61%)	148 (59%)	67 (60%)	448 (65%)
HSP-70/1	C	66 (38%)	95 (38%)	37 (33%)	242 (35%)
rs1043618	G	110 (63%)	155 (62%)	75 (67%)	452 (65%)
HSP-70/1	C	159 (90%)	235 (94%)*	100 (89%)	587 (85%)
rs1043620	T	17 (10%)	15 (6%)	12 (11%)	107 (15%)
HSP-70/Hom	T	143 (81%)	219 (88%)	93 (83%)	560 (81%)
rs2227956	C	33 (19%)	31 (12%)	19 (17%)	134 (19%)
HSP-70/Hom	G	146 (83%), † , ‡	154 (62%)	77 (69%)	460 (66%)
rs2075800	A	30 (17%)	96 (38%)	35 (31%)	234 (34%)

B. Haplotype Carriage Frequencies
Haplotype	Sarcoid uveitis (n = 88)	IAU (n = 125)	IIU (n = 56)	Controls (n = 347)
1	53 (60%)	73 (58%)	33 (59%)	190 (55%)
2	41 (47%), § , ∥	24 (19%)	13 (23%)	73 (21%)
3	25 (28%), # , **	75 (60%)	29 (52%)	193 (56%)
4	17 (19%)	15 (12%), ††	12 (21%)	99 (28%)
5	16 (18%)	16 (13%)	7 (12%)	26 (7%)
6	1 (1%)	8 (6%)	0	1 (0%)
7	2 (2%)	7 (6%)	8 (14%)	4 (1%)

*   P _c = 0.0007, OR, 2.78, 95% CI, 1.60–4.84, IAU vs. controls.

†   P _c = 0.00008, OR, 2.45, 95% CI, 1.61–3.72, sarcoid uveitis vs. controls.

‡   P _c = 0.00001, OR, 3.00, 95% CI, 1.88–4.78, sarcoid uveitis vs. IAU.

§   P _c = 0.001, OR, 3.62, 95% CI, 1.97–6.64, sarcoid uveitis vs. IAU.

∥   P _c = 0.04, OR, 2.81, 95% CI, 1.34–5.89, sarcoid uveitis vs. IIU.

#   P _c = 0.0006, OR, 0.27, 95% CI, 0.15–0.48, sarcoid uveitis vs. IAU.

**   P _c = 0.04, OR, 0.37, 95% CI, 0.19–0.75, sarcoid uveitis vs. IIU.

††   P _c = 0.005, OR, 0.35, 95% CI, 0.20–0.63, IAU vs. controls.

Table 9.

 

HLA-B27 Allele Carriage Frequencies in Sarcoidosis as a Whole, Sarcoid Non-uveitis, Sarcoid Uveitis, IAU, IIU, and Controls

	Sarcoid Non-uveitis (n = 182)	Sarcoidosis (n = 270)	Sarcoid Uveitis (n = 88)	IAU (n = 125)	IIU (n = 56)	Controls (n = 347)
HLA-B27	13 (7%)	15 (6%)	2 (2%)* , †	84 (68%), ‡ , §	12 (22%), ∥	31 (9%)

 HLA-B27 data were available for 265 sarcoidosis (178 sarcoid non-uveitis and 87 sarcoid uveitis), 123 idiopathic anterior uveitis, 55 idiopathic intermediate uveitis and 347 controls.

*   P < 0.0001, OR, 91.54, 95% CI 21.42–391.23, IAU vs. sarcoid uveitis.

†   P = 0.0002, OR, 11.86, 95% CI 2.54–55.39, IIU vs. sarcoid uveitis.

‡   P < 0.0001, OR, 7.72, 95% CI 3.67–16.24, IAU vs. IIU.

§   P < 0.0001, OR, 21.95, 95% CI 12.93–37.28, IAU vs. controls.

∥   P = 0.005, OR, 2.84, 95% CI 1.36–5.95, IIU vs. controls.