Variant analysis of CARD14 in a Chinese Han population with psoriasis vulgaris and generalized pustular psoriasis.

TO THE EDITOR

Psoriasis is a common, chronic, inflammatory, organ-specific autoimmune skin disease with a complex genetic background (Nestle ; Zhang ). Psoriasis vulgaris (PsV) is the most common type, accounting for approximately 85–90% of all psoriasis patients, and characterized by raised, well-demarcated, erythematous oval plaques with adherent silvery scales (Nestle ). Generalized pustular psoriasis (GPP) is the least prevalent form of psoriasis and is considered to be a potentially life-threatening, multi-systemic disease (Körber ), characterized by the sudden eruption of generalized sterile pustules in a wave-like manner (Mengesha and Bennett, 2002).

Genetic factors have been shown to have a critical role in the pathogenesis of psoriasis, and several genes and genomic regions have been reportedly associated with some clinical forms of this disease (http://omim.org/), including the gene for caspase recruitment domain family member 14 (CARD14), which was identified by fine-mapping of psoriasis susceptibility locus 2 (PSORS2) in European and Taiwanese patients (Hwu ; Jordan ). To date, 23 variants have been reported in the CARD14 gene (Supplementary Table S3 online), mainly in patients with PsV (Jordan ; 2012b; Körber ). However, most studies of CARD14 have been carried out in European populations, and studies in Chinese patients are rare.

We performed direct DNA sequencing in 236 psoriasis patients (174 PsV and 62 GPP patients) and 365 controls to investigate the prevalence of CARD14 variants in the Chinese Han population. We also compared CARD14 variants between patients with GPP and PsV and between pediatric- and late-onset PsV.

A total of four new, rare heterozygous missense variants (allele frequency <1%) were found in the CARD14 exon, all with frequencies of 0.2% in psoriasis patients: p.Met119Val (c.355A>G) and p.Arg166His (c.497G>A) were only seen in GPP with PsV patients, and p.Ala216Thr (c.646G>A) and p.Thr591Met (c.1772C>T) (rs200102454) were only seen in PsV patients. We predicted the effects of the four variants on protein function using the Sorting Intolerant From Tolerant (SIFT) tool (http://sift.bii.a-star.edu.sg). Only p.Thr591Met was predicted to have damaging effects (score=0.03), whereas the other three were predicted to be tolerated (Supplementary Table S4 and Supplementary Figure S1 online).

An additional known rare heterozygous variant p.Arg682Trp (c.2044C>T) (rs117918077), which was reported in 1.3% of 2169 European psoriasis patients (Jordan ), was only found in one GPP with PsV patient (0.2% in psoriasis patients) in this study (Supplementary Table S4 and Supplementary Figure S1 online).

The overall frequency of these five rare variants was 1.1% in psoriasis patients (0.6% in PsV patients and 2.4% in GPP patients) but 0% in controls. A significant association between CARD14 rare variants and GPP was observed using Fisher's exact test (corrected-P=0.03), but there was no significant association with either psoriasis or PsV (corrected-P=0.09, 0.998, respectively). We also conducted a gene-burden test using the sequence kernel association test package to compare the distributions of the five rare missense variants in psoriasis patients and controls, which showed a P-value of 0.00172. Subtype analysis of PsV versus GPP and pediatric- versus late-onset PsV revealed no significant differences in the frequencies of the five rare variants (corrected-P=0.438, 1, respectively; Table 1).

Table 1

Analysis of five rare CARD14 variants using Fisher's exact test

	p.Met119Val				p.Arg166His				p.Ala216Thr				p.Thr591Met				p.Arg682Trp				Combination of five variants
	AA	Aa	aa	Allele frequency (a)	AA	Aa	aa	Allele frequency (a)	AA	Aa	aa	Allele frequency (a)	AA	Aa	aa	Allele frequency (a)	AA	Aa	aa	Allele frequency (a)	AA	Aa	aa	Allele frequency (a)
P-PsV	94	0	0		94	0	0		93	1	0	0.005	94	0	0		94	0	0		93	1	0	0.005
L-PsV	80	0	0		80	0	0		80	0	0		79	1	0	0.006	80	0	0		79	1	0	0.006
PsV	174	0	0		174	0	0		173	1	0	0.003	173	1	0	0.003	174	0	0		172	2	0	0.006
GPP	61	1	0	0.008	61	1	0	0.008	62	0	0		62	0	0		61	1	0	0.008	59	3	0	0.024
Psoriasis	235	1	0	0.002	235	1	0	0.002	235	1	0	0.002	235	1	0	0.002	235	1	0	0.002	231	5	0	0.011
Control	365	0	0		365	0	0		365	0	0		365	0	0		365	0	0		365	0	0
P (Psoriasis versus control)				0.393 (0.561)				0.393 (0.561)				0.393 (0.561)				0.393 (0.561)				0.393 (0.561)				0.009 (0.09)
P (PsV versus GPP)				0.263 (0.438)				0.263 (0.438)				1 (1)				1 (1)				0.263 (0.438)				0.115 (0.438)
P (PsV versus control)				1 (1)				1 (1)				0.322 (0.998)				0.322 (0.998)				1 (1)				0.104 (0.998)
P (GPP versus control)				0.145 (0.29)				0.145 (0.29)				1 (1)				1 (1)				0.145 (0.29)				0.003 (0.03)
P (P-PsV versus L-PsV)				1 (1)				1 (1)				1 (1)				0.46 (1)				1 (1)				1 (1)

Abbreviations: GPP, generalized pustular psoriasis; L-PsV, late-onset psoriasis vulgaris; P-PsV, pediatric-onset psoriasis vulgaris; PsV, psoriasis vulgaris (including P-PsV and L-PsV); Ps, psoriasis (including PsV and GPP).

In the table, there are two types of P-values—nominal and corrected, of which the latter are in parentheses, calculated by false-discovery rate.

One known low-frequency variant, p.Asp176His (c.526G>C) (rs144475004), was detected with frequencies of 1.9% in psoriasis patients and 1.8% in controls. Three common single-nucleotide polymorphisms (SNPs), rs2066964, rs34367357, and rs11652075, were also detected in our samples, with frequencies of 44.1 vs. 7.4 vs. 44.3% in psoriasis patients and 46.0 vs. 5.6 vs. 48.2% in controls, respectively. There were no differences between the groups in terms of these four variants (Supplementary Table S5 and Supplementary Figure S1 online).

CARD14 is located within PSORS2 and encodes a nuclear factor (NF)-κB activator. Variants in CARD14 have recently been detected in association with psoriasis, mostly in patients with PsV (Jordan ). Particular rare variants within CARD14 could lead to psoriasis by upregulating psoriasis-associated genes in keratinocytes (Jordan ).

The present study identified five rare variants that, in combination, were more common in psoriasis patients compared with controls. Regarding the predicted effects of these rare variants on protein function, p.Thr591Met and p.Arg682Trp were predicted to be damaging, possibly by regulating the activation of NF-κB, leading to inflammation and epidermal hyperplasia. The other variants were not predicted to be damaging, and their functions remain unknown.

GPP is considered to differ from PsV in terms of its etiology, especially regarding variants in the IL36RN gene (Körber ; Li ; Sugiura ). In our study, variants within CARD14 significantly associated with GPP, compared with controls. However, we failed to find any difference between GPP and PsV, which might be attributed to the insufficient sample size. To elucidate whether the gene CARD14 is a specific susceptibility gene for PsV or GPP, further study with a large sample size is needed.

Different genes have been reported to be responsible for diseases with different onset ages (Swanbeck ). Cheng ) performed an association analysis in patients with early-onset psoriasis (onset age <40 years) and identified significant associations with the SNP rs4649203 in IL28RA and rs2303138 in LNPEP (rs4649203, P=0.0191, odds ratio (OR)=0.87; rs2303138, P=0.00391, OR=1.15). However, the current study failed to distinguish between pediatric- and late-onset PsV in terms of CARD14 variants. We therefore conclude that the CARD14 gene variant cannot explain age of onset in this cohort of psoriasis patients.

In conclusion, we performed an association analysis between CARD14 variants and psoriasis in a Chinese Han population. The results suggest that rare CARD14 variants may have an important role in the pathogenesis of GPP.

This study was approved by the human medical and ethics committee of Shandong Provincial Institute of Dermatology and Venereology and was conducted according to the Declaration of Helsinki Principles. After informed consent, genomic DNA was extracted from patients' peripheral blood samples using a QuickGene DNA whole blood kit L (Kurabo Industries, Osaka, Japan). All exons of the CARD14 gene were amplified by PCR, and the products were sequenced on an ABI 3130xl Genetic Analyser (Applied Biosystems ABI, Carlsbad, CA). (For specific details about materials and methods see Supplementary Data online).