Association of HLA-DRB1*15 and HLADQB1*06 with SLE in Saudis.

BACKGROUND AND OBJECTIVES: Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by humoral autoimmunity. The etiology of SLE is thought to be multifactorial including environmental, hormonal, and genetic factors. The human leukocyte antigen (HLA) has extensively been associated with the susceptibility to SLE; however, the association is heterogeneous among different ethnic groups. The aim of this study was to determine the association of HLA-A, HLA-B, HLA-DRB1, and HLA-DQB1 with SLE susceptibility in the Saudi population. DESIGN AND SETTINGS: A total of 86 consecutive SLE patients attending the rheumatology clinic at King Abdulaziz Medical City, Riyadh, were recruited for this study.
METHODS: HLA types were determined by the polymerase chain reaction sequence-specific oligonucleotide (PCR-SSP) method in 86 patients and 356 control subjects.
RESULTS: The following HLA alleles were found to be positively associated with SLE: HLA-A*29 (OR=2.70; 95% CI=1.03-7.08; P=.0035), HLA-B*51 (OR=1.81; 95% CI=1.17-2.79; P=.0066), HLA-DRB1*15 (OR=1.45; 95% CI=0.98-2.29; P=.063), and HLA-DQB1*06 (OR=1.67; 95% CI=1.19-2.36; P=.0032), whereas HLA-DRB1*16 was negatively associated with the disease (OR=0.18; 95% CI=0.02-1.3; P=.055). HLA-DRB1*15 haplotypes were significantly associated with SLE (OR=2.01, 95% CI=1.20-3.68, P=.008); this was mainly due to the HLADRB1*15-DQB1*06 association.
CONCLUSIONS: Our data suggest an association between MHC class I and class II (HLA-A*29, HLA-B*51, HLA-DRB1*15, and HLA-DQB1*06) and susceptibility to SLE in the Saudi population. HLA-DRB1*15-DQB1*06 haplotype showed the highest risk factor for the disease that is similar to what was seen in the African American patients, suggesting shared susceptibility genetic factors among these ethnic groups.

Systemic lupus erythmatosus (SLE) is a chronic autoimmune disease characterized by inflammation of various tissues and organs of the body due to the production of autoantibodies.1 SLE mostly affects the heart, lung, skin, joints, kidney, liver, blood vessels, and nervous system. Diagnosing SLE can be difficult as the symptoms come and go unpredictably. The disease is treatable mostly with corticosteroids and immunosuppressants, but thus far it is incurable and can be fatal. The prevalence of SLE varies among ethnicity and gender. It has been found to occur more frequently among African Americans and those of non-European descent. Also it is more common in women than in men.2

The etiology of SLE is unknown; however, several factors that have been considered to trigger as well as exacerbate the disease are hormones,3 pathogens,4 medications, and UV light.5 Moreover, genetic factors have been strongly suggested to contribute to the disease. These were mainly shown by studies on identical twins6 in addition to observing a higher disease frequency in the relatives of patients with SLE7 and children of mothers with SLE.8 Extensive studies have linked SLE susceptibility to genes of the human leukocyte antigen (HLA) region. However, the extent of this association varies among different population.

The aim of this study is to investigate the association between HLA alleles and SLE in our Saudi population.

METHODS

Patients and controls

A total of 86 consecutive SLE patients attending the rheumatology clinic at King Abdulaziz Medical City, Riyadh, were recruited for this study. All patients met at least 4 of the 11 American College of Rheumatology criteria.9 Patients were consented and file review was conducted to collect all clinical and laboratory data. HLA results were compared with 356 ethnically matched controls.10

HLA typing

A total of 5 mL peripheral blood was collected in EDTA. DNA was prepared from blood leukocytes using the salting out procedure. White cells were separated using Ficoll Hypaque followed by lysis of erythrocytes in red blood cell lysis buffer and protein digestion in proteinase K solution. Finally, DNA was extracted by precipitating proteins in a saturated salt solution using the QIAamp DNA Blood Mini Kit from Qiagen (Valencia, California). All individuals were DNA typed for HLA-A, HLA-B, HLA-DRB1, and HLA-DQB1 using polymerase chain reaction-sequence specific primer (PCR-SSP) (Deutsche Dynal AG, Hamburg, Germany) using low-resolution typing method.

Statistical analysis

The maximum likelihood estimates of allele frequencies and haplotype frequencies were computed using an expectation maximization algorithm by the Arlequin software.11 To compare the differences between the allele frequencies in the controls and SLE groups, a 2×2 contingency table analysis was performed using the Pearson chi-square tests with Fisher exact test, when the expected value for an HLA marker was <5. The strength of association between HLA alleles and SLE was estimated by odds ratios (OR) and 95% confidence intervals (95% CI). P<.05 was considered to be statistically significant. For the 2-locus haplotypes, the the standardized disequilibrium coefficient (D′) and the chi-square values were also calculated.

RESULTS

We investigated HLA-A, HLA-B, HLA-C, HLA-DRB1, and HLA-DQB1 genes in 86 Saudi SLE patients and compared their results with 356 healthy controls. The female gender dominated showing a ratio of female to male as 10.7:1, with the mean age of 26.1 years at onset and the mean disease duration of 8.8 years (Table 1). Arthritis was the most common clinical presentation, followed by renal involvement, malar rash, leucopenia, and alopecia (Table 2). Most patients presented with ANA (98.9%), anti-DNA (98.9%), and low complement C3 and C4 (93.0%), (Table 3).

Table 1

Demographics of the SLE patients.

Characteristics
Total patient number	86
F:M	10.7:1
Age range of patients	9–60 y
Mean age at onset (SD)	26.1 10.2 y
Disease duration mean (SD)	8.8 (5) y

SLE: Systemic lupus erythematosus, F:M: female to male ratio, SD: standard deviation.

Table 2

Clinical presentation of SLE patients.

Clinical presentation	No. (%)
Arthritis	71 (82.6)
Renal involvement	30 (34.9)
Alopecia	19 (22.1)
Raynaud phenomenon	7 (8.1)
Photosensitivity	11 (12.8)
Hemolytic anemia	13 (15.1)
CNS involvement	4 (4.7)
Malar rash	25 (29.1)
Oral ulcers	13 (15.1)
Leukopenia	25 (29.1)
Thrombocytopenia	9 (10.5)
Serositis	4 (4.7)
Pleuritis	4 (4.7)

SLE: Systemic lupus erythematosus, CNS: central nervous system.

Table 3

Number and percentage of autoantibody positive SLE patients.

Autoantibody	No. (%)
ANA	85 (98.9)
Anti-DNA	85 (98.9)
Low C3 or C4	80 (93.0)
RNP antibodies	12 (14.0)
Anti-Sm	5 (5.8)
Anti-cardiolipin	39 (45.3)
Anti-Ro	23 (26.7)
Anti-La	10 (11.6)

SLE: Systemic lupus erythematosus.

Tables 4 to 7 show the HLA class I and class II allele frequencies in both SLE cases and controls. The following HLA types were significantly increased in cases versus controls: HLA-A*29 (OR 2.70, 95% CI 1.03–7.08, P=.035) and HLA-DQB1*06 (OR 1.67, 95% CI 1.19–2.36, P=.032). However, HLA-B*51 (OR 1.81, 95% CI 1.17–2.79, P=.0066) and HLA-DRB1* 15 (OR 1.49, 95% CI 0.98–2.29, P=.063) were marginally significant. HLA-DRB1*01 (OR 0.16, 95% CI 0.02–1.2, P=.041) was protective, whereas HLA-A*02 (OR 0.69, 95% CI 0.47–1.02, P=.06) and HLA-DRB1*16 (OR 0.18, 95% CI 0.02–1.3, P=.055) were marginally protective.

Table 4

HLA-A associations with SLE in Saudi patients.

SLE	Controls				OR	95% CI	P
	N	Frequency	N	Frequency
HLA-A*01	17	0.099	51	0.072
HLA-A*02	40	0.233	217	0.305	0.69	0.47–1.02	.06
HLA-A*03	7	0.041	46	0.065
HLA-A*11	8	0.047	27	0.038
HLA-A*23	10	0.058	38	0.053
HLA-A*24	16	0.093	53	0.074
HLA-A*25	0	0.000	1	0.001
HLA-A*26	11	0.058	33	0.046
HLA-A*29	7	0.041	11	0.015	2.70	1.03–7.08	.035
HLA-A*30	7	0.041	39	0.055
HLA-A*31	16	0.093	50	0.070
HLA-A*32	4	0.023	37	0.052
HLA-A*33	7	0.041	43	0.060
HLA-A34	2	0.012	0	0.003
HLA-A66	1	0.006	1	0.001
HLA-A68	15	0.081	55	0.077
HLA-A69	2	0.012	0	0.000
HLA-A74	2	0.012	8	0.011

SLE: Systemic lupus erythematosus, OR: odds ratio, HLA: human leukocyte antigen.

Table 5

HLA-B associations with SLE in Saudi patients.

SLE	Controls				OR	95% CI	P
	N	Frequency	N	Frequency
HLA-B*07	20	0.116	69	0.097
HLA-B*08	13	0.076	57	0.08
HLA-B*13	1	0.006	10	0.014
HLA-B*14	1	0.006	7	0.01
HLA-B*15	7	0.041	32	0.045
HLA-B*18	1	0.006	22	0.031
HLA-B*27	0	0	6	0.008
HLA-B*35	14	0.081	58	0.081
HLA-B*37	2	0.012	8	0.011
HLA-B*38	4	0.023	9	0.013
HLA-B*39	5	0.029	6	0.008
HLA-B*40	6	0.035	11	0.015
HLA-B*41	9	0.052	28	0.039
HLA-B*42	3	0.017	8	0.011
HLA-B*44	4	0.023	26	0.037
HLA-B*45	0	0	2	0.003
HLA-B*46	0	0	2	0.003
HLA-B*47	0	0	1	0.001
HLA-B*49	1	0.006	27	0.038
HLA-B*50	23	0.134	137	0.192
HLA-B*51	35	0.203	88	0.124	1.81	1.17–2.79	.0066
HLA-B*52	3	0.017	11	0.015
HLA-B*53	7	0.041	31	0.044
HLA-B*54	0	0	1	0.001
HLA-B*55	2	0.012	5	0.007
HLA-B*56	3	0.017	0	0
HLA-B*57	2	0.012	15	0.021
HLA-B*58	6	0.035	27	0.038
HLA-B*67	0	0	2	0.003
HLA-B*73	0	0	5	0.007
HLA-B*78	0	0	1	0.001

SLE: Systemic lupus erythematosus, OR: odds ratio, HLA: human leukocyte antigen.

Table 6

HLA-DRB1 associations with SLE in Saudi patients.

SLE	Controls				OR	95% CI	P
	N	Frequency	N	Frequency
HLA-DRB1*01	1	0.006	25	0.035	0.16	0.02–1.20	.041
HLA-DRB1*15	35	0.203	104	0.146	1.49	0.98–2.29	.063
HLA-DRB1*16	1	0.006	23	0.032	0.18	0.02–1.30	.055
HLA-DRB1*03	30	0.174	103	0.145
HLA-DRB1*04	17	0.099	114	0.16
HLA-DRB1*11	12	0.07	49	0.069
HLA-DRB1*12	0	0	5	0.007
HLA-DRB1*13	29	0.169	94	0.132
HLA-DRB1*14	1	0.006	7	0.01
HLA-DRB1*07	33	0.192	144	0.202
HLA-DRB1*08	5	0.029	7	0.01
HLA-DRB1*09	1	0.006	0	0
HLA-DRB1*10	7	0.041	37	0.052

SLE: Systemic lupus erythematosus, OR: odds ratio, HLA: human leukocyte antigen.

Table 7

HLA-DQB1 associations with SLE in Saudi patients.

SLE	Controls				OR	95% CI	P
	N	Frequency	N	Frequency
HLA-DQB1*02	60	0.349	245	0.344
HLA-DQB1*03	34	0.198	164	0.23
HLA-DQB1*04	3	0.017	19	0.027
HLA-DQB1*05	5	0.029	77	0.108
HLA-DQB1*06	70	0.407	207	0.291	1.67	1.19–2.36	.0032

SLE: Systemic lupus erythematosus; OR: odds ratio, HLA: human leukocyte antigen.

Table 8 describes the association between HLA-DRB1* 15 haplotypes and SLE. Apparently, HLA-DRB1* 15-DQB1*06 haplotype carried a significant risk for SLE (OR 2.01, 95% CI 1.20–3.68, P=.008) in our Saudi population.

Table 8

Haplotypes in association with SLE.

SLE	Controls
DRB1*15 Haplotype	N	Frequency	N	Frequency
HLA-A*02 HLA-B*07 HLA-DRB1*15 HLA-DQB1*06	8	0.047	27	.038
HLA-A*01 HLA-B*51 HLA-DRB1*15 HLA-DQB1*06	4	0.023	0	0
HLA-A*02 HLA-B*51 HLA-DRB1*15 HLA-DQB1*06	3	0.017	8	.011
HLA-A*31 HLA-B*35 HLA-DRB1*15 HLA-DQB1*06	3	0.017	6	.008
HLA-A*02 HLA-B*50 HLA-DRB1*15 HLA-DQB1*06	2	0.012	1	.001

SLE: Systemic lupus erythematosus, human leukocyte antigen.

Note: Having DRB1*15 haplotypes in SLE compared to controls OR=2.01, 95% (1.20–3.68), P=.008.

DISCUSSION

We investigated the association of HLA genes in a Saudi cohort of SLE patients. This is the first description of immunogenetics of SLE in Saudi Arabia. The age at onset and the preponderance of females over males in this cohort were similar to other populations.12

Two major HLA haplotypes have been shown repeatedly to be associated with SLE worldwide: HLA-DR3 and HLA-DR2 (DR15 and DR16) bearing haplotypes.13–16 Different HLA-DR alleles were reported in different ethnics groups: HLA-DRB1* 0301 with Caucasians, HLA-DRB1*1503 with African Americans, and HLA-DRB1*08 alleles with Hispanics. In our population, HLA-DRB1*15 haplotypes were found to be associated with SLE in Saudis, while HLA-DRB1*16 was protective. In Mexicans, DR15 haplotypes were found to be associated with risk for SLE,17 while, 1 study showed that HLA-DRB1*16 was associated with chronic discoid lupus in Mexicans.18 HLA-DRB1*04 was protective in this Saudi population; the same result was observed in patients from Northwest Spain.19

Several studies analyzed the MHC region for genetic risk of SLE. Graham et al20 narrowed the disease-associated haplotypes HLA-DRB1*1501-HLA-DQB1*0602 and HLA-DRB1*0801-HLA-DQB1* 0402 to a region of 500 kb. Fernando et al21 using British SLE families and TdT analysis, narrowed the susceptibility region in MHC to 180 kb that involved the HLA-DRB1*0301-HLA-DQA1*0501-HLA-DQB1* 0201. Our own results suggested that HLA-DRB1* 15-HLA-DQB1*06 haplotype is a risk factor for SLE in Saudis; however, looking at the allele frequencies we find that the frequency of HLA-DRB1*15 is nearly 20% whereas that of HLA-DQB1*06 is 40%, suggesting that HLA-DQB1*06 is associated with SLE independent of HLA-DRB1*15. Thus narrowing the risk area of SLE to the DQB1 region, it still remains elusive whether HLA-DQB1*06 is the culprit or another gene polymorphism is in linkage disequilibrium with it.

One third of our patients have renal involvement; whereas, in other Asian populations, renal involvement ranged from 18% to 100%, majority reporting >50% of their patients.22 In Italians, lupus nephritis was found to be associated with the HLA-DR15-bearing haplotypes; 23 this was also reported in other studies.24–26 In our patients, there was no association between HLA-DR15-bearing haplotypes and lupus nephritis (data not shown). Alarcón et al27 analyzed factors influencing the development of lupus nephritis. Their results suggested that younger, hypertensive, and of African American or Hispanic ethnicity were predictors of lupus nephritis risk. Moreover, end-stage renal disease was also predicted by the presence of homozygosity for the valine allele of FcγRIIIa (FCGR3A*GG).27 This finding suggested that HLA-DR15 is not the only predictor of lupus nephritis risk and thus further analysis is required to determine the risk factors for the development of lupus nephritis in the Saudi patients.

In conclusion, this is the first study to show HLA-DRB1 and HLA-DQB1 associations with SLE in the Saudi population.