Supplementary material

Genetic association of PRDM1-ATG5 intergenic regionand autophagy in systemic lupus erythematosis

Xu-jie Zhou1, Xiao-lan Lu2, Ji-cheng Lv1, Hai-zhen Yang3, Lian-xiang Qin1, Ming-hui Zhao1, Yin Su2, Zhan-guo Li2, Hong Zhang1

Sup. Table1. Result of logistic regression analysis of 9 SNPs for the development of SLE

SNP / P value by different model
Recessive / Additive / Dominant
rs811925 / 0.836 / 0.999 / 0.782
rs573869 / 0.078 / 0.193 / 0.666
rs548234 / 0.676 / 7.20×10-5 / 3.89×10-5
rs6937876 / 0.251 / 0.014 / 0.074
rs6568431 / 0.052 / 0.053 / 0.318
rs4945747 / 0.060 / 0.515 / 0.993
rs2245214 / 0.137 / 0.450 / 0.433
rs2757133 / 0.194 / 0.846 / 0.844
rs573775 / 0.615 / 0.864 / 0.450

P values for each SNP under the recessive, additive, or dominant model were calculated by logistic regression analysis controlled for the 9 SNPs, gender and age.

Sup. Table2.Primers for mRNA expression in vitro validation

Gene / Forward / Backward
PRDM1 / CCAGCTCTCCAATCTGAAGG / GGCATTCATGTGGCTTTTCT
ATG5 / TTTGCATCACCTCTGCTTTC / TAGGCCAAAGGTTTC AGCTT
ATG3 / TGCTATAAGCGGTGCAAACA / CGGCTTCCGTTATTCCTGTA
IRF5 / CATTACTGTACAGGTGGTGC / AGATGTGATGGAGCTCCTTG
GAPDH / TGAACGGGAAGCTCACTGG / TCCACCACCCTGTTGCTGTA

Interaction analysis by logistic regression and chi-square test

To detect the possible genetic interaction effect among the associated autophagy genes, data for the 5 SLE associated SNPs were compiled.

Multiplicative interactive effect of the SNPs was estimated by a multiple logistic regression model[1-3]. For each individual key variables were defined as (1) a binary variable indicating case-control status, (2) 5 SNP variables ranging from 0-2 indicating the number of minor alleles that the individual has. For each SNP pair, a logistic regression model was built to predict case-control status based on the indicator variables (gender, age) and the two SNP variables (a total of 4 variables and an intercept). We tested whether the log-likelihood of the model was significantly improved by adding an additional multiplicative pairwise interaction term for those two SNPs. A total of 10 tests (5×4/2) were conducted. An interaction term was considered significant only if p<5×10-3 (=0.05/10).

To test for additive interactions, direct counting and chi-square tests were used[4, 5].

Sup. Table3.Additive interaction analysis of genes involved in SLE in genotype combinations by chi-square test

Combination / Number (%) / p vs. con / OR (95% CI)
Controls / SLE
rs548234/rs6937876-1/1 / 398(28.6) / 298(25.3) / 1.00
rs548234/rs6937876-1/2 / 364(26.1) / 252(21.4) / 0.485 / 0.925(0.742-1.152)
rs548234/rs6937876-2/1 / 241(17.3) / 179(15.2) / 0.949 / 0.992(0.777-1.267)
rs548234/rs6937876-2/2 / 389(27.9) / 449(38.1) / 2.68×10-5 / 1.524(1.259-1.887)
overall χ2=30.572, P=1.05×10-6,RERI=0.607
rs548234/rs13361189-1/1 / 235(16.9) / 150(12.7) / 1.00
rs548234/rs13361189-1/2 / 524(37.8) / 401(34.0) / 0.143 / 1.199(0.941-1.528)
rs548234/rs13361189-2/1 / 203(14.6) / 178(15.1) / 0.030 / 1.374(1.031-1.831)
rs548234/rs13361189-2/2 / 425(30.6) / 449(38.1) / 4.84×10-5 / 1.655(1.297-2.113)
overall χ2=20.528, P=1.32×10-4,RERI=0.082
rs548234/rs10065172-1/1 / 245(17.7) / 155(13.1) / 1.00
rs548234/rs10065172-1/2 / 512(36.9) / 396(33.6) / 0.101 / 1.223(0.962-1.554)
rs548234/rs10065172-2/1 / 199(14.3) / 187(15.9) / 6.13×10-3 / 1.485(1.119-1.972)
rs548234/rs10065172-2/2 / 431(31.1) / 441(37.4) / 8.73×10-5 / 1.617(1.271-2.058)
overall χ2=18.820, P=2.98×10-4
rs548234/rs11706903-1/1 / 398(28.6) / 232(19.8) / 1.00
rs548234/rs11706903-1/2 / 363(26.1) / 314(26.8) / 4.65×10-4 / 1.484(1.189-1.852)
rs548234/rs11706903-2/1 / 313(22.5) / 279(23.8) / 2.63×10-4 / 1.529(1.217-1.922)
rs548234/rs11706903-2/2 / 317(22.8) / 346(29.5) / 2.80×10-8 / 1.872(1.499-2.339)
overall χ2=31.851, P=5.63×10-7
rs6937876/rs13361189 -1/1 / 210(15.1) / 137(11.5) / 1.00
rs6937876/rs13361189- 1/2 / 430(30.9) / 344(28.9) / 0.121 / 1.226(0.948-1.587)
rs6937876/rs13361189 -2/1 / 232(16.7) / 194(16.3) / 0.090 / 1.282(0.961-1.709)
rs6937876/rs13361189 -2/2 / 519(37.3) / 516(43.3) / 8.09×10-4 / 1.524(1.190-1.952)
overall χ2=12.897, P=4.87×10-3,RERI=0.016
rs6937876/ rs10065172 -1/1 / 219(15.7) / 149(12.5) / 1.00
rs6937876/ rs10065172 -1/2 / 419(30.1) / 332(27.9) / 0.238 / 1.165(0.904-1.500)
rs6937876/ rs10065172 -2/1 / 229(16.5) / 195(16.4) / 0.119 / 1.252(0.944-1.660)
rs6937876/ rs10065172 -2/2 / 524(37.7) / 516(43.3) / 2.58×10-3 / 1.447(1.137-1.842)
overall χ2=10.915, P=0.012,RERI=0.030
rs6937876/ rs11706903 -1/1 / 320(23.0) / 202(17.1) / 1.00
rs6937876/ rs11706903 -1/2 / 320(23.0) / 275(23.2) / 0.011 / 1.361(1.072-1.728)
rs6937876/ rs11706903 -2/1 / 391(28.1) / 312(26.4) / 0.046 / 1.264(1.004-1.592)
rs6937876/ rs11706903 -2/2 / 361(25.9) / 394(33.3) / 2.04×10-6 / 1.729(1.378-2.169)
overall χ2=23.590, P=3.04×10-5,RERI=0.104
rs13361189/ rs10065172 -1/1 / 346(25.0) / 222(18.6) / 1.00
rs13361189/ rs10065172 -1/2 / 95(6.9) / 109(9.1) / 3.83×10-4 / 1.788(1.295-2.469)
rs13361189/ rs10065172 -2/1 / 101(7.3) / 122(10.2) / 6.66×10-5 / 1.883(1.377-2.574)
rs13361189/ rs10065172 -2/2 / 844(60.9) / 739(62.0) / 1.78×10-3 / 1.365(1.123-1.659)
overall χ2=22.502, P=5.13×10-5
rs13361189/ rs11706903 -1/1 / 197(14.2) / 142(12.0) / 1.00
rs13361189/ rs11706903 -1/2 / 242(17.4) / 187(15.8) / 0.636 / 1.072(0.804-1.430)
rs13361189/ rs11706903 -2/1 / 511(36.8) / 373(31.5) / 0.923 / 1.013(0.786-1.305)
rs13361189/ rs11706903 -2/2 / 437(31.5) / 481(40.7) / 9.43×10-4 / 1.527(1.187-1.964)
overall χ2=23.582, P=3.05×10-5,RERI=0.442
rs10065172/ rs11706903 -1/1 / 214(15.4) / 154(13.0) / 1.00
rs10065172/ rs11706903 -1/2 / 231(16.7) / 188(15.9) / 0.394 / 1.131(0.852-1.501)
rs10065172/ rs11706903 -2/1 / 495(35.7) / 361(30.5) / 0.916 / 1.013(0.791-1.298)
rs10065172/ rs11706903 -2/2 / 447(32.2) / 481(40.6) / 1.19×10-3 / 1.495(1.172-1.908)
overall χ2=20.517, P=1.33×10-4,RERI=0.351

Interactions were conducted by direct countings and chi-square tests.

Genotype combinations were conducted under the dominant model. The overall significance for the difference in risk genotype counts between patients and controls in all groups was high. And the risk genotype combination contributed the most to the overall interaction.

RERI: The relative excess risk due to interaction

Significant interactions were shown in bold and RERIs were calculated.

rs548234: ATG5, 1: TT; 2: CC+CT;

rs6937876: ATG5, 1: AA; 2: GG+AG;

rs13361189: IRGM, 1: TT; 2: CC+CT;

rs10065172: IRGM, 1: CC; 2: TT+TC;

rs11706903:ATG7, 1: CC; 2: AA+AC.
97 gene expressions examined in downstream ATG5 activation

We examined 97 genes as follows:

11 autophagy related genes[6, 7]: SQSTM1, BECN1, APG3L, APG4A, APG4B, APG4C, APG4D, APG7L, APG10L, APG12L, and APG16L1;

39 type I IFN related genes[8-10]:IRF5, MDA5, IFNAR2, IFNB1, IFNA1, IFNA2, IFNA4, IFNA5, IFNA6, IFNA7, IFNA8, IFNA10, IFNA13, IFNA14, IFNA16, IFNA17, IFNA21, OAS1, OAS2, IFIT4, IFI16, ISG20, STAT1, STAT2, STAT3, STAT4, STAT5A, STAT5B, IRF2BP1, IRF2BP2, IRF1, IRF2, IRF3, IRF4, IRF6, IRF7, RSAD2, PLSCR1, and TMPO;

4 IL-12 related genes[10]:IL12A, IL12B, IL12RB1,IL12RB2;

24 apoptosis related genes[11, 12]: PDCD1, PDCD1LG1, PDL2,CDKN1A, TNFRSF6, TNFSF6, CTLA4, FAF1, FAIM2,FASTK, CFLAR, CIAS1, CASP1,CASP2, CASP3, CASP4, CASPR4, CASP5, CASPR5, CASP6, CASP7, CASP8, CASP9, and CASP10;

19 NF-κB pathway related genes[13-15]: NFKB1, NFKB2, NKAP, NFKBIA, TNIP1, TNIP2, TNIP3, TNFAIP3, RELA, RELB, REL, TRAF3IP1, TRAF1, TRAF2, TRAF3, TRAF4, TRAF5, TRAF6, and TNF.

Sup. Table 4. List of genes analysed whose expression correlated with ATG5 in 210 unrelated HapMap samples in genome-wide significance.

Genes / The Pearson correlation coefficient / P value
autophagy related genes
BECN1 / 0.449 / 8.18×10-12
APG3 / 0.645 / 4.35×10-26
APG4B / 0.330 / 1.00×10-6
APG7L / 0.487 / 6.43×10-14
APG10L / 0.465 / 1.14×10-12
APG12L / 0.488 / 5.98×10-14
APG16L / 0.472 / 4.87×10-13
type I IFN related genes
IRF5 / 0.426 / 1.13×10-10
IFNA21 / 0.358 / 9.48×10-8
IFI16 / 0.381 / 1.17×10-8
STAT1 / 0.462 / 1.64×10-12
TMPO / 0.607 / 1.60×10-22
IFNB1 / 0.378 / 1.50×10-8
IRF2BP1 / 0.471 / 5.12×10-13
apoptosis related genes
PDL2 / 0.462 / 1.68×10-12
TNFRSF6 / 0.435 / 4.16×10-11
FAF1 / 0.453 / 5.06×10-12
CASP3 / 0.407 / 8.91×10-10
CASP5 / 0.414 / 4.44×10-10
CASP6 / 0.458 / 2.86×10-12
CASP8 / 0.561 / 8.84×10-19
NF-κB related genes
NFKB1 / 0.568 / 2.63×10-19
NFKB2 / 0.387 / 6.42×10-9
NKAP / 0.575 / 7.30×10-20
NFKBIA / 0.550 / 5.07×10-18
TNIP1 / 0.484 / 9.95×10-14
TNFAIP3 / 0.620 / 1.08×10-23
TRAF3IP1 / 0.509 / 2.90×10-15
TRAF1 / 0.470 / 6.39×10-13
TRAF2 / 0.600 / 5.76×10-22
TRAF6 / 0.452 / 6.01×10-12

Among the 97 genes analysed, 31 genes whose expression correlated with ATG5 in genome-wide significance (P1.06×10-6), including 7/11 of autophagy related genes, 7/39 of type I IFN related genes, 7/24 of apoptosis related genes,10/19 of NF-κB pathway related genes.

References:

1.Karlson EW, Chang SC, Cui J, et al. Gene-environment interaction between HLA-DRB1 shared epitope and heavy cigarette smoking in predicting incident rheumatoid arthritis. Ann Rheum Dis 2010;69:54-60.

2.Gateva V, Sandling JK, Hom G, et al. A large-scale replication study identifies TNIP1, PRDM1, JAZF1, UHRF1BP1 and IL10 as risk loci for systemic lupus erythematosus. Nat Genet 2009;41:1228-33.

3.Raychaudhuri S, Thomson BP, Remmers EF, et al. Genetic variants at CD28, PRDM1 and CD2/CD58 are associated with rheumatoid arthritis risk. Nat Genet 2009;41:1313-8.

4.Kallberg H, Padyukov L, Plenge RM, et al. Gene-gene and gene-environment interactions involving HLA-DRB1, PTPN22, and smoking in two subsets of rheumatoid arthritis. Am J Hum Genet 2007;80:867-75.

5.Perdigones N, Vigo AG, Lamas JR, et al. Evidence of epistasis between TNFRSF14 and TNFRSF6B polymorphisms in patients with rheumatoid arthritis. Arthritis Rheum 2010;62:705-10.

6.Munz C. Enhancing immunity through autophagy. Annu Rev Immunol 2009;27:423-49.

7.Virgin HW, Levine B. Autophagy genes in immunity. Nat Immunol 2009;10:461-70.

8.Rullo OJ, Woo JM, Wu H, et al. Association of IRF5 polymorphisms with activation of the interferon alpha pathway. Ann Rheum Dis 2010;69:611-7.

9.Kawasaki A, Kyogoku C, Ohashi J, et al. Association of IRF5 polymorphisms with systemic lupus erythematosus in a Japanese population: support for a crucial role of intron 1 polymorphisms. Arthritis Rheum 2008;58:826-34.

10.Lee HK, Lund JM, Ramanathan B, et al. Autophagy-dependent viral recognition by plasmacytoid dendritic cells. Science 2007;315:1398-401.

11.Yousefi S, Simon HU. Apoptosis regulation by autophagy gene 5. Crit Rev Oncol Hematol 2007;63(3):241-4.

12.Pyo JO, Jang MH, Kwon YK, et al. Essential roles of Atg5 and FADD in autophagic cell death: dissection of autophagic cell death into vacuole formation and cell death. J Biol Chem 2005;280:20722-9.

13.Criollo A, Senovilla L, Authier H, et al. IKK connects autophagy to major stress pathways. Autophagy 2010;6:189-91.

14.Criollo A, Senovilla L, Authier H, et al. The IKK complex contributes to the induction of autophagy. EMBO J 2010;29:619-31.

15.Shi CS, Kehrl JH. TRAF6 and A20 regulate lysine 63-linked ubiquitination of Beclin-1 to control TLR4-induced autophagy. Sci Signal 2010;3(123):ra42.