Genetic Characterization and Susceptibility for Sarcoidosis in Japanese Patients: Risk Factors of BTNL2 Gene Polymorphisms and HLA Class II Alleles

Abstract

Purpose.: Sarcoidosis is a heterogeneous and multisystem granulomatous disorder. The etiology still is uncertain, but the disease currently is thought to be triggered by various genetic as well as environmental factors. Recently, an association between sarcoidosis and the butyrophilin-like 2 (BTNL2) gene located in close proximity to the HLA-DRB1 gene was reported. The purpose of our study was to verify the relationship between BTNL2 and HLA risk alleles for the susceptibility to sarcoidosis, and to assess whether the BTNL2 association is independent of the HLA risk alleles.

Methods.: In our study, 11 single nucleotide polymorphisms (rs28362677, rs2076533, rs2076530, rs2076529, rs2294881, rs3763304, rs2076523, rs28362682, rs3806156, rs9268499, rs3763317), including the functional rs2076530 (G > A) of the BTNL2 gene, and HLA-DRB1 and -DQB1 alleles, were genotyped in 237 Japanese patients diagnosed with sarcoidosis and 287 healthy Japanese control subjects.

Results.: In the patient group, the HLA-DRB1*08:03 (P = 6.15 × 10⁻⁵, odds ratio [OR] = 2.43) and BTNL2 rs2076530_A (P = 6.90 × 10⁻⁶, OR = 1.84) were associated with disease susceptibility. Upon stratification analysis in search for a synergistic effect given the extensive linkage disequilibrium between BTNL2 rs2076530_A and HLA-DRB1*08:03, our results suggested that the risk-bearing allele of these two loci interact negatively. No significant differences were observed in allele frequencies for alleles in patients with ocular and other systemic sarcoidosis.

Conclusions.: Our studies implicated that the HLA-DRB1 allele is a major contributing genetic factor in the development of sarcoidosis in Japan. However, further studies are needed to verify how HLA or BTNL2 alleles confer the disease phenotype, severity of sarcoidosis.

Introduction

Sarcoidosis is a chronic systemic disease characterized by marked macrophage and CD4+ T-cell activity, and the accumulation of noncaseating granulomas in a wide range of organs, such as lungs, lymph nodes, eye, skin, and heart. ¹ In the current survey of uveitis in Japan, the most frequent inflammatory disease was sarcoidosis (10.6%), followed by Vogt-Koyanagi-Harada disease (7.0%), acute anterior uveitis (6.5%), scleritis (6.1%), herpetic iridocyclitis (4.2%), Behçet's disease (3.9%), bacterial endophthalmitis (2.5%), masquerade syndrome (2.5%), Posner-Schlossman syndrome (1.8%), and retinal vasculitis (1.6%). ² The etiology of sarcoidosis still is uncertain, but the disease currently is thought to be triggered by various genetic as well as environmental factors. It is well established that sarcoidosis is associated with the human leukocyte antigen (HLA) class II genes, especially HLA-DRB1 and HLA-DQB1 genes, in several ethnic groups. ^3–7 However, to our knowledge it has not yet been clarified whether HLA genes themselves are the pathogenic gene related directly to sarcoidosis or are associated with the disease only because of linkage disequilibrium with some other genes. Recently, a single nucleotide polymorphism within butyrophilin-like 2 (BTNL2), rs2076530, has been implicated as a risk factor for sarcoidosis. ⁸ The BTNL2 gene, located only 170 kilobases (kb) from the HLA-DRB1 gene telomerically on chromosome 6, is a member of the immunoglobulin superfamily with likely costimulatory activities in T-cell activation on the basis of amino acid homology to B7 molecules. ⁹ The G/A transition of rs2076530 causes a premature truncation of protein, disrupting the membrane localization of the protein and a necessary process for down-regulating activated T-cells (Th1). ^8,10 The truncated protein increases the risk of developing sarcoidosis independent of HLA-DRB1 risk alleles. ^8,11 However, subsequent studies have showed inconsistent results on a sarcoidosis risk factor of BTNL2 rs2076530_A and its independent effect of HLA-DRB1. ^12,13 In other diseases, such as type 1 diabetes (T1D), ¹⁴ rheumatoid arthritis (RA), ¹⁵ systemic lupus erythematosus (SLE), ¹⁴ multiple sclerosis (MS), ¹⁵ Graves' disease (GD), ¹⁶ chronic beryllium disease (CBD), ¹⁷ and ulcerative colitis (UC), ¹⁸ a significant association with BTNL2 polymorphisms appeared to be secondary to the primary HLA-DRB1 association because of its strong linkage disequilibrium (LD) with DRB1 alleles. 

In our study, to verify the association between BTNL2 and sarcoidosis, and independence from the HLA-DRB1 risk alleles, we performed a case-control association study of the BTNL2 polymorphisms and HLA-DRB1 alleles in Japanese sarcoidosis patients. 

Materials and Methods

Subjects

The total subject group consisted of 239 Japanese patients with sarcoidosis and 287 healthy Japanese controls from Yokohama City University, Hokkaido University, Fujita Health University, Tokyo University, Keio University, the Kumamoto City hospital, and the Yuasa Eye Clinic. All 239 patients had chronic sarcoidosis, which was diagnosed according to the “diagnostic criteria and guidelines for sarcoidosis” developed by the Japanese Society of Sarcoidosis and Other Granulomatous Disorders (JSSOG 2007). ¹⁹ The diagnostic criteria used currently in Japan include histologic and clinical diagnosis. The key criteria are clinical features from laboratory examination, radiologic features, pathologic findings (noncaseous epithelioid cell granuloma in biopsy specimens), and exclusion of other diseases. The concepts agree with the ATS/ERS/WASOG statement on sarcoidosis (ATS/ERS/WASOG 1999). ²⁰ A clinical diagnostic group presenting no histologic evidence of granuloma, but demonstrating clinical features suggesting sarcoid lesions in two or more organs, and supported by the evidence from two or more characteristic clinical and radiologic features include bilateral hilar lymphadenopathy (BHL) on chest x-ray and/or findings of diagnostic imaging, such as CT/HRCT or bronchoscopy, elevated serum ACE, negative tuberculin test, abnormal uptake on Gallium-67 citrate scintigraphy, and an increase in lymphocyte count or an elevated CD4/CD8 ratio in bronchoalveolar lavage fluid. 

Among the patients with sarcoidosis, 198 had ocular (uveitis), 138 lung, 47 skin, 48 cardiac, and 10 nerve lesions (Table 1). All control subjects were unrelated healthy volunteers who were ethnically similar to the patients. Details of the study were explained to all patients and controls, and written informed consent was obtained for genetic screening. The study obtained the approval of the Ethics Committee of Yokohama City University School of Medicine, and was in compliance with the guidelines of the Declaration of Helsinki. 

Table 1. 

 

Characteristics and Frequencies of Clinical Phenotype

	Sarcoidosis	Controls
N	239	287
Age at onset, mean	49.6 ± 16.2	33.1 ± 9.4
Sex, men/women	0.245/0.755	0.456/0.544
	n	Frequencies
Diagnostic criteria
Clinical	135	0.565
Histologic	104	0.435
Lesions
Lung	138	0.577
Ocular, uveitis	198	0.828
Skin	47	0.197
Heart	48	0.201
Nerve	10	0.042
Chronic	239	1.000
Lofgren syndrome	0	0.000

Genomic DNA was extracted from peripheral blood cells using the QIAamp DNA Blood Maxi Kit (Qiagen, Tokyo, Japan). 

Genotyping of Single-Nucleotide Polymorphisms (SNPs) in BTNL2 Region

We examined 11 SNPs: rs3763317, rs9268499, rs3806156, rs28362682, rs2076523, rs3763304, rs2294881, rs2076529, rs2076530, rs2076533, and rs28362677 selected based on information from public sources, including the NCBI dbSNP, ABI, and HapMap databases, and a previous report. ⁸ Genotyping of all SNPs was performed by the TaqMan 5′ exonuclease assay using primers supplied by ABI (Foster City, CA). 

HLA-DRB1 and HLA-DQB1 Genotyping

Genotyping of HLA-DRB1 and -DQB1 alleles was performed by Luminex reverse sequence-specific oligonucleotides using bead kits from One Lambda, Inc. (Canoga Park, CA). Additional PCR sequence-specific primer of SBT reactions were used to determine high-resolution alleles, if necessary. 

Statistical Analysis

Allele frequencies of all detected SNPs were tested for Hardy-Weinberg equilibrium (HWE). Differences in allele frequencies between cases and controls were assessed with the χ² test or Fisher's exact test. The Haploview 3.32 program was used to compute pairwise LD statistics. ²¹ Standardized disequilibrium D' and r ² were calculated. LD blocks were defined according to the criteria of Gabriel et al. ²² Haplotype frequencies were estimated with an accelerated expectation-maximization algorithm, similar to the partition–ligation–expectation–maximization method described previously. ²³ All P values were derived from a 2-sided test, and P values < 0.05 were considered statistically significant. To obtain a measure of significance corrected for multiple testing we applied Bonferroni's correction. 

Results

The Case-Control Association Study of BTNL2 Polymorphisms

Eleven biallelic SNPs in the BTNL2 gene were genotyped in all the sarcoidosis patients and control subjects (Fig. 1). Among 11 SNPs, the SNP 9 (rs2076530)_A allele showed extremely strong association with susceptibility to the disease (P = 6.9 × 10⁻⁶, odds ratio [OR] = 1.84, Table 2). All 11 SNPs were in HWE between cases and controls. 

Figure 1. 

 

Location of 11 SNPs in the BTNL2 used in our study and genomic organization of 257 kb of the MHC region of human chromosome 6, at 6p21.31, showing the relative location of BTNL2 and HLA-Class II genes.

Table 2. 

 

Allele Frequencies of SNPs of the BTNL2 Gene between Sarcoidosis Patients and Controls

SNP	db SNP	Allele	Risk	Risk Allele Frequency, n (%)		P	Pc	OR (95% CI)
Name			Allele	Cases (n = 237)	Controls (n = 287)
S1	rs3763317	A/G	G	162 (34.2)	162 (28.2)	0.038
S2	rs9268499	C/T	C	274 (57.8)	288 (50.2)	0.014
S3	rs3806156	A/C	A	267 (56.3)	314 (54.7)
S4	rs28362682	T/A	A	121 (25.5)	113 (19.7)	0.024
S5	rs2076523	G/A	G	267 (56.3)	314 (54.7)
S6	rs3763304	G/A	A	129 (27.2)	142 (24.7)
S7	rs2294881	G/A	A	243 (51.3)	285 (49.7)
S8	rs2076529	A/G	A	299 (63.1)	322 (56.1)	0.022
S9	rs2076530	A/G	A	355 (74.9)	355 (61.8)	0.0000069	0.0003	1.84 (1.41–2.40)
S10	rs2076533	G/A	G	298 (62.9)	322 (56.1)	0.026
S11	rs28362677	G/A	G	342 (72.2)	376 (65.5)	0.021

 CI, confidence intervals.

The haplotype frequencies and structures were estimated and constructed from the 11 SNPs using the Haploview software (Table 3). Three LD blocks were induced (Fig. 2) and no significant differences were observed in the haplotype frequencies of each block between patients and controls groups. However, the haplotype distributions in blocks 2 and 3 were different between the DRB1*08:03-positive patient group and the all patients group, and also the control group. The third frequent haplotype of block 2 in the control and all patients groups, A–A–G was the most frequent haplotype in DRB1*08:03-positive patients (57.9%), and the second most frequent haplotype of block 3, A–G–A–A–G–G was the most frequent haplotype in DRB1*08:03-positive patients (53.6%). 

Figure 2. 

 

Structure of LD plotted for 11 SNPs in the BTNL2. The D' values corresponding to each SNP pair are expressed as a percentage and shown within the respective square. The haplotype blocks were determined using the Haploview 4.2 software. The gradient of D' value (low to high value) is indicated by different shades of red (light to dark, respectively).

Table 3. 

 

Haplotye Structure and Frequency in Three Blocks of BTNL2 Observed in Controls, Patients, and DRB1*08:03-Positive Patients

	Block 1		Freq.	Block 2			Freq.	Block 3						Freq.
	SNP1	SNP2		SNP3	SNP4	SNP5		SNP6	SNP7	SNP8	SNP9	SNP10	SNP11
Controls, n = 287	A	T	0.496	C	T	A	0.453	G	A	A	A	G	G	0.329
	G	C	0.280	A	T	G	0.256	A	G	A	A	G	G	0.159
	A	C	0.222	A	A	G	0.197	G	G	G	A	A	A	0.131
								G	G	G	G	A	A	0.125
								G	A	G	G	A	G	0.094
								A	G	G	G	A	A	0.089
								G	A	A	G	G	G	0.073
Patients, n = 239	A	T	0.419	C	T	A	0.435	G	A	A	A	G	G	0.298
	G	C	0.339	A	T	G	0.308	A	G	A	A	G	G	0.165
	A	C	0.239	A	A	G	0.253	G	G	G	A	A	A	0.108
								G	G	G	G	A	A	0.064
								G	A	G	G	A	G	0.049
								G	A	A	A	A	G	0.046
								A	G	G	G	A	A	0.045
DRB1*08:03-Positive Patients, n = 96	A	T	0.469	A	A	G	0.579	A	G	A	A	G	G	0.536
	G	C	0.280	A	T	G	0.221	G	G	G	A	A	A	0.171
	A	C	0.222	C	T	A	0.200	G	A	A	A	G	G	0.143
								A	G	G	G	A	A	0.079
								G	A	G	G	A	G	0.050
								G	A	A	G	G	G	0.021

 Freq., frequency.

Association Analysis of HLA-DRB1 and HLA-DQB1 Alleles

To determine whether there were different allelic distributions in sarcoidosis patients and controls, we performed an association test for each allele group of the DRB1 and DQB1 alleles. The frequency of the HLA-DRB1*08:03 allele was significantly different between sarcoidosis patients and their matched controls (Table 4, P = 6.90 × 10⁻⁶, OR = 2.43). Among DRB1 alleles, two risk alleles, DRB1*08:03 and DRB1*09:01, were observed in the disease susceptibility. No statistically significant differences were detected between cases and controls of DQB1 alleles. 

Table 4. 

 

Distribution of HLA-DRB1 and -DQB1 Alleles in Patients with Sarcoidosis and in Healthy Control Subjects

	Frequency, n (%)				P	OR (95% CI)
	Cases (n = 237)		Controls (n = 287)
HLA-DRB1*
01:01	8 (1.7)		29 (5.1)		0.0037	0.32 (0.14–0.70)
03:01	0 (0.0)		2 (0.4)
04:01	4 (0.8)		1 (0.2)
04:03	13 (2.7)		18 (3.2)
04:05	63 (13.3)		73 (12.9)
04:06	15 (3.2)		20 (3.5)
04:07	1 (0.2)		3 (0.5)
04:09	1 (0.2)		0 (0.0)
04:10	8 (1.7)		9 (1.6)
08:02	17 (3.6)		34 (6.0)
08:03	78 (16.5)		43 (7.6)		6.9E-06	2.43 (1.64–3.61)
08:09	1 (0.2)		0 (0.0)
09:01	99 (20.9)		80 (14.1)		0.0039	1.61 (1.16–2.23)
11:01	7 (1.5)		11 (1.9)
12:01	24 (5.1)		18 (3.2)
12:02	9 (1.9)		10 (1.8)
13:01	3 (0.6)		2 (0.4)
13:02	15 (3.2)		42 (7.4)		0.0026	0.41 (0.22–0.75)
13:03	0 (0.0)		1 (0.2)
13:07	0 (0.0)		1 (0.2)
14:03	9 (1.9)		6 (1.1)
14:05	5 (1.1)		11 (1.9)
14:06	2 (0.4)		12 (2.1)		0.027	0.20 (0.04–0.88)
14:07	3 (0.6)		0 (0.0)
14:09	0 (0.0)		2 (0.4)
14:12	1 (0.2)		1 (0.2)
14:54	20 (4.2)		22 (3.9)
15:01	25 (5.3)		37 (6.5)
15:02	43 (9.1)		75 (13.2)		0.039	0.66 (0.44–0.97)
16:02	0 (0.0)		5 (0.9)
HLA-DQB1
DQB1*0201	0	0.0	2	0.4
DQB1*0301	53	11.2	60	10.8
DQB1*0302	43	9.1	62	11.2
DQB1*0303	101	21.3	82	14.7	0.0069	1.57 (1.14–2.16)
DQB1*0401	59	12.4	67	12.1
DQB1*0402	21	4.4	26	4.7
DQB1*0501	8	1.7	30	5.4	0.0015	0.30 (0.14–0.66)
DQB1*0502	6	1.3	15	2.7
DQB1*0503	22	4.6	25	4.5
DQB1*0601	119	25.1	115	20.7
DQB1*0602	23	4.9	31	5.6
DQB1*0603	3	0.6	1	0.2
DQB1*0604	15	3.2	40	7.2	0.0050	0.42 (0.23–0.77)
DQB1*0622	1	0.2	0	0.0

LD Analysis between BTNL2 and DRB1 Alleles

Examination of haplotype frequency between DRB1–BTNL2 also revealed that the patient group was significantly higher than the control group at rs2076530_A and DRB1*08:03 haplotype (P = 7.72 × 10⁻⁶, OR = 2.42). Almost 85% were rs2076530_A propositi in the DRB1*08:03-positive patient groups. 

The haplotype rs2076530_A and DRB1*09:01 showed weaker association with the disease susceptibility (P = 0.005) than the rs2076530_A and DRB1*08:03 haplotype (Table 5). D' (degree of LD) value between DRB1*08:03 or DRB1*09:01 and rs2076530_A was 0.92 and 0.84 in patients, and 0.90 and 0.83 in controls, respectively. 

Table 5. 

 

Linkage Disequilibrium between BTNL2 rs2076530_A and HLA-DRB1*08:03 or -DRB1*09:01

Samples	HLA-DRB1	rs2076530_A (SNP9)
		D'	r 2
Total, n = 547	*08:03	0.92	0.05
	*09:01	0.84	0.07
Controls, n = 287	*08:03	0.90	0.04
	*09:01	0.83	0.06
Patients, n = 237	*08:03	0.92	0.05
	*09:01	0.84	0.06

We investigated whether these two loci contributed independently to susceptibility for sarcoidosis or the possible confounding effect is related. In this analysis, we stratified the study cohort into HLA-DRB1 risk alleles (DRB1*08:03 and DRB1*09:01) carriers and HLA-DRB1 risk alleles noncarriers (Table 6). No significant association with the disease was observed for rs2076530_A in noncarriers of DRB1 risk alleles. However, for the carriers of DRB1 risk alleles, the BTNL2 risk allele showed weak association with sarcoidosis (P = 0.011). Meanwhile, for the carriers of the rs2076530_A allele, HLA-DRB1 risk alleles showed an extremely strong association with sarcoidosis. 

Table 6. 

 

Effects of LD between DRB1 Risk Alleles and rs2076530_A Allele on Association Results

	Patients		Controls		P	OR (95% CI)
Stratified with HLA-DRB1 risk alleles
DRB1 risk alleles*
rs2076530_A frequency, n (%)
Carrier	n = 134	239 (89.2)	n = 111	180 (81.1)	0.01	1.92 (1.15–3.21)
Noncarrier	n = 103	116 (56.3)	n = 176	175 (49.7)	0.13	1.30 (0.92–1.84)
Stratified with rs2076530_A allele
rs2076530_A
DRB1 risk alleles* frequency, n (%)
Carrier	n = 199	173 (43.5)	n = 217	120 (27.6)	1.83 × 10−6	2.01 (1.51–2.69)
Noncarrier	n = 38	4 (5.3)	n = 70	3 (2.1)	0.22	2.54 (0.55–11.65)

*   DRB1*08:03 and DRB1*09:01.

We also carried out two-locus analyses to detect the alleles that showed the strongest association among the neighboring loci (Table 7), according to the method of Svejgaard and Ryder. ²⁴ Tests 1 and 2 investigated whether the DRB1* risk alleles are deviating in the rs2076530_A positives and negatives. Conversely, the rs2076530_A was tested for independent association against the DRB1* risk alleles in tests 3 and 4 in the presence and absence of DRB1* risk alleles, respectively. In test 7, it was investigated whether the DRB1* risk alleles and rs2076530_A association differ, and in test 8 the combined association of the DRB1* risk alleles and rs2076530_A was compared relative to their absence. 

Table 7. 

 

Two-Locus Analysis of Genetic Risk of DRB1 Risk Alleles and rs2076530_A Allele to Sarcoidosis

DRB1 Risk Alleles*	rs2076530_A Allele	Cases, n = 237	Controls, n = 287
Observed data
+	+	130	109
+	−	4	2
−	+	69	108
−	−	34	68
		OR	P Value
Two–by–two comparisons (DRB1 risk alleles and rs2076530_A allele)
T1	++ vs. −+	1.87	0.002
T2	+− vs. −−	3.13	0.222
T3	++ vs. +−	0.60	0.856
T4	−+ vs. −−	1.28	0.346
T5	+− vs. −+	4.00	0.348
T6	++ vs. −−	2.39	0.000037

 T1 and T2, tests for independent DRB1 risk alleles; T3 and T4, tests for independent rs2076530_A allele; T5, tests for difference between DRB1 risk alleles and rs2076530_A allele to sarcoidosis; T6, combined association.

*   DRB1*08:03 and DRB1*09:01.

As shown in Table 7, the DRB1* risk alleles contributed to the susceptibility independently of the rs2076530_A (test 1), whereas the rs2076530_A showed no significant association in DRB1* risk alleles –positives (test 3) or –negatives (test 4). The synergistic effect between two alleles was not observed (test 6), because the OR (2.39) for persons with both alleles was not exactly higher than a risk allele at one locus (Tables 2 and 4, OR was 2.43 for the DRB1*08:03 and 1.84 for the rs2076530_A). 

Discussion

SNPs in the BTNL2 gene have been reported to be associated with the disease susceptibility for sarcoidosis. ^8,11,25 Especially, a functional BTNL2 polymorphism rs2076530_A allele has been discussed on disease susceptibility conferred dependently or independently with HLA-DRB1 alleles. ^8,11–13 This reflects the extensive linkage disequilibrium across the HLA region and the difficulties to assign primary associated locus within this genomic region. ^14–16 Multivariable logistic regression analyses showed that BTNL2 effects are independent of human leukocyte antigen class II genes in Caucasians, but may interact antagonistically in African Americans. ¹¹ In our study, strong LD was observed between DRB1* alleles (DRB1*08, DRB1*09, DRB1*11, DRB1*12, DRB1*14, DRB1*15) and BTNL2 rs2076530_A, particularly for DRB1*08:03 or the DRB1*09:01 allele and BTNL2 rs2076530_A. Haplotype structures and frequencies in the BTNL2 gene were not different between the patient and control groups. In contrast, patients carrying DRB1*0803 showed different structure and frequencies of haplotypes in blocks 2 and 3. These haplotypes were specific for the DRB1*0803 allele due to strong LD. Taken together, our findings supported no evidence of an independent genetic association between BTNL2 polymorphisms rs2076530_A and the susceptibility to sarcoidosis. 

Numerous studies have suggested that there is an association between certain HLA alleles and sarcoidosis. ^26–30 However the susceptible allele was different (DRB1*03, DRB1*11, DRB1*12, DRB1*14, and DRB1*15 on HLA class II gene) in various countries. ^4–7,27,31 This may be influenced by different diagnosis and clinical characteristics of the study subjects, and the ethnicity studied. No consensus was established about which HLA locus is involved directly in the pathogenesis of sarcoidosis, but the most recent studies focused mainly on the HLA-DRB1 genes. It is probable to suppose that the HLA-DRB1 alleles have been the best studied candidate genes in sarcoidosis because DR proteins are responsible for the presentation of specific antigens that are able to start and maintain the immunopathologic process. The loading of a particular peptide onto an HLA class II molecule is dependent strictly on the characteristic of the peptide-binding groove of the molecules. ^32–34 In HLA-DR molecules, pockets P1, P4, P6, P7, and P9 are present within the binding groove. ³³ Among the 5 pockets, the amino acid epitopes in pockets 4, 6, and 7 seem to have a role in the susceptibility of sarcoidosis. ^4,5,28,35 The combination of specific amino acids residues at 11, 47, and 71 positions in three pockets appears to be implicated in susceptible or protective determinants for disease progression. The risk allele DRB1*08:03 in Japanese patients with sarcoidosis carries HLA-DRB1-S¹¹, HLA-DRB1-Y⁴⁷, and HLA-DRB1-R⁷¹. 

These associations also could be involved in the specific forms, ^35,36 clinical features, ^37,38 progression, ^39,40 and prognosis ⁴¹ of the disease. Our data clearly showed the existence of an association between HLA-DRB1*08:03 and sarcoidosis in general (regardless of the onset of the disease and subtypes of the disease). 

BTNL2 is a member of the immunoglobulin superfamily and a role as co-stimulatory receptor involved in modulation of T-cell response on the basis of the amino acid homology of B7 (CD80 and CD86) proteins by experiments in mice. ¹⁰ A possible modulatory role for the protein in intestine inflammation has been proposed in this animal model. However, the function of human BTNL2 differs significantly from that of its murine ortholog in a number of respects. ^8,10,42 In addition, sarcoidosis, tuberculoid leprosy, tuberculosis, and Crohn's disease are Th1-mediated diseases characterized by granuloma formation. In Th1-dominated granulomatous diseases, the truncating BTNL2 SNP (rs2076530) was not supported to be important. ⁴³  

Therefore, to our knowledge the physiologic role of human BTNBL2 has not yet been elucidated, and the role of the protein in the pathogenesis of autoimmune diseases remains to be revealed. Taken together, we concluded that there probably is no major role for BTNL2 in the pathogenesis of sarcoidosis from our results presented. 

Recent studies have highlighted the effecter role of Th17 cells in pathologic conditions, including autoimmunity and inflammation. ⁴⁴ An elevation of Th17 cells was demonstrated in the peripheral blood and bronchoalveolar lavage of patients with active sarcoidosis. ⁴⁵ Restrict inflammation caused by a joint Th1/Th2 response might be involved in the pathogenesis of granuloma formation in sarcoidosis. 

Accurate clinical diagnosis, especially where disease heterogeneity is known to exist, large sample sizes, and the precise mapping of the HLA region will provide additional power to dissect gene effects in determining susceptibility to sarcoidosis. 

Footnotes

 Supported by Grant-in-Aid for the National Cerebral and Cardiovascular Center (ID:20-5) from the Ministry of Health, Labour and Welfare of Japan.

Footnotes

³  These authors contributed equally to this work presented here and therefore should be regarded as equivalent authors.

Footnotes

 Disclosure: H. Suzuki, None; M. Ota, None; A. Meguro, None; Y. Katsuyama, None; T. Kawagoe, None; M. Ishihara, None; Y. Asukata, None; M. Takeuchi, None; N. Ito, None; E. Shibuya, None; E. Nomura, None; R. Uemoto, None; T. Nishide, None; K. Namba, None; N. Kitaichi, None; S. Morimoto, None; T. Kaburaki, None; Y. Ando, None; S. Takenaka, None; J. Nakamura, None; K. Saeki, None; S. Ohno, None; H. Inoko, None; N. Mizuki, None