Effect of host genetics on CMV retinitis occurrence in patients with AIDS

Abstract

Background

Cytomegalovirus (CMV) retinitis is a common opportunistic infection among patients with AIDS and still causes visual morbidity despite the wide spread usage of highly active antiretroviral therapy (HAART). The ubiquitous CMV pathogen contains a human interleukin-10 (IL-10) homolog in its genome and utilizes it to evade host immune reactions through an IL-10 receptor mediated immune-suppression pathway.

Methods

Effects of IL-10R1, IL-10 and previously described AIDS restriction gene variants are investigated on the development of CMV retinitis in the Longitudinal Study of the Ocular Complications of AIDS (LSOCA) cohort (n=1284).

Results

In Europen Americans (n=750), a haplotype carrying an amino acid changing variation in the cytoplasmic domain (S420L) of IL-10R1 can be protective (OR = 0.14, CI: 0.02–0.94, P = 0.04) against, whereas another haplotype carrying an amino acid changing variation in the extracellular domain (I224V) of IL-10R1 can be more susceptible (OR = 6.21, CI: 1.22–31.54, P = 0.03) to CMV retinitis. In African Americans (n=534), potential effects of IL-10 variants are observed.

Conclusion

Host genetics may have a role in the occurrence of CMV retinitis in patients infected with HIV.

INTRODUCTION

Human cytomegaloviruses (HCMVs) are ubiquitous β-herpes virus pathogens that infect 80% of the US adults by age 40 usually with little or no clinical symptoms. However, in neonates, allogenic transplant recipients and immunocompromised people such as patients with AIDS, HCMV is a substantial cause of morbidity and mortality [1]. Even after the introduction of highly active antiretroviral therapy (HAART), HCMV continues to be one of the most frequent opportunistic pathogens in patients with AIDS [2].

CMV retinitis was the main reason (>90%) of vision loss among patients with AIDS in the pre-HAART era [3, 4]. Although immune recovery in response to HAART reduced the incidence of CMV retinitis cases, the expected decline in the number of new cases of CMV retinitis has leveled off [2]. CMV retinitis related visual morbidity continues to be a problem in the era of HAART due to HAART intolerant and/or unresponsive patients, resistance development to long term CMV treatment, and improved survival rate increasing the population who remain at risk for CMV retinitis [2]. Moreover, there are several case reports of patients with elevated CD4+ T-cell counts who nevertheless experience a relapse of CMV retinitis [5–8]. Clearly, non-genetic factors such as poor general health, availability of sufficient health care and adherence to therapy are important factors in CMV retinitis development [9–11]. However, the effect of host genetics in CMV activation and progression to CMV retinitis has been largely unexplored.

The HCMV genome contains several genes that can evade the host immune system [1, 12]. One of these genes is a human interleukin-10 (IL-10) homolog called cmvIL-10 [13]. IL-10 is a pleiotropic cytokine that inhibits inflammatory and cell-mediated immune responses through suppression of production of proinflammatory cytokines and the expression of major histocompatibility complex class II and co-stimulatory molecules [14]. The receptor complex for IL-10 is composed of two subunits, IL-10R1 and IL-10R2. The binding of IL-10 to its receptor complex activates the Janus kinase (JAK)-STAT3/STAT1 signal transduction pathway. IL-10R1 (Jak1 associated), is the crucial ligand (IL-10) binding and signal transducing unit of the IL-10 receptor complex [15, 16]. The cmvIL-10 can bind to IL-10R1, initiate signaling through IL-10R complex, and exhibit nearly identical immunosuppressive properties similar to that of human IL-10 [17, 18]. Genetic polymorphisms of IL-10R1 have been shown to diminish the inhibitory effects of IL-10 on monocytes [19], and to reduce STAT3 and STAT1 activated signaling [20]. Moreover, IL-10R1 polymorphisms can decrease signaling activity of cmvIL-10 through the IL-10 signaling pathway [21]. Potential roles for IL-10 receptor polymorphisms in infectious disease outcomes have also been suggested. Specifically, effects of IL-10R1 variants in hepatitis C [22, 23] and progression to AIDS [24]. On the other hand, IL-10 variants that modulate IL-10 transcript production also regulate AIDS progression in HIV-1 infected individuals [24–27].

For this, study we took a candidate gene approach to investigate potential effects of host genetics on CMV retinitis development. We focused primarily on IL-10R1 polymorphisms, but also explored polymorphisms in IL-10. The study participants were HIV-infected European American and African American patients enrolled in the Longitudinal Study of the Ocular Complications of AIDS (LSOCA) cohort. As HCMV-related organ diseases are seen typically in the later years of HIV infection, correlate with AIDS progression (such as CD4+ T-cell counts), and HAART treatment, we also examined variants in genes [28, 29] that have been shown to influence HIV-1 infection, AIDS progression, therapy response, and antiviral drug metabolism to account for potential genetic confounding factors.

PATIENTS and METHODS

Study population and Clinical Assessment of CMV retinitis

Study patients included 750 European American and 534 African American individuals enrolled in the Longitudinal Studies of the Ocular Complications of AIDS (LSOCA) cohort. LSOCA is a prospective multicenter observational study of incident or prevalent cases diagnosed with AIDS according to the 1993 surveillance case definition of the Centers for Disease Control and Prevention. Eligible patients included both incident and prevalent cases of CMV retinitis as well as those without CMV retinitis. The first patient was enrolled into the study in September 1998 and enrollment was still continuing through 30 Sep 2009. Each pateint provided a written consent for study participation. Further information on the study design and implementation and ophthalmologic examinations have been described in previous publications [4, 10, 11].

Genotyping and Haplotype construction

The IL-10R1 region was genotyped for high frequency (≥ 5%) functional and haplotype tagging variants (SNPs) of rs3135932 (Exon-4, replacement, also reported as SNP3 in the literature [19]), rs2228055 (Exon-5, replacement, also reported as SNP6 in the literature [19]), rs4252279, rs4252314, rs4252286 (Intron-6), rs2229113 (Exon-7, replacement, also reported as SNP4 in the literature [19]), and rs2229114 (Exon-7, replacement, also reported as SNP8 in the literature [19]). Additionally, 11 haplotype tagging SNPs (promoter region: rs17351243, rs4072227, rs6667202, rs1800890, rs1800896 and rs1800894; intronic: rs3021094, rs3024508; 3′ UTR: rs3024496, rs3024498 and rs3024500) covering the IL-10 region were selected. Supplemental figure 1 shows individually genotyped SNPs and inferred haplotypes for IL-10R1 in European and African Americans. Previously identified [28–31] functional polymorphisms for AIDS restriction and therapy response genes were genotyped for CCR5Δ32 (rs333), CCR2-64I (rs1799864), CCR5 P1 (rs1799988), SDF-3′A (rs1801157), RANTES -403A (rs2107538), RANTES -28G (rs2280788), RANTES-In1.1C (rs2280789), CX3CR1-V249I (rs3732379), CX3CR1-T280M (rs3732378), IFNG-179T (rs2069709), MDR1-C3435T (rs1045642), and MCP-1364G(rs2857657, intronic 767G, representative of haplotype 7). All SNPs were genotyped using the ABI-TaqMan method (Applied Biosystems, Foster City, CA, USA).

All haplotypes were inferred by the expectation maximization algorithm with SAS Genetics (SAS Institute, Cary, NC, USA) and the HaploView software [32]. The presence of CCR5_59353C (rs1799988) in the absence of CCR2-64I and CCR5Δ32 defines the CCR5 P1 promoter haplotype +.P1.+. [31] The RANTES -403A, RANTES -28G, RANTES-In1.1C genotypes define the RANTES haplotypes. RANTES -H1=G-C-T, RANTES -H2=A-C-T and RANTES -H3=A-C-C (low producer haplotype) [30, 33].

Statistical Analyses

Each SNP and haplotype found at ≥ 1% frequency in the study population was evaluated as a risk factor for CMV retinitis by three different models of inheritance: allelic, dominant, and codominant (additive). Allelic analyses examined individual allele effects. Genotypes were coded as 0, 1 and 2 copies of the rare allele for the codominant (additive) model. Dominant model analyzed genotypes as absence or presence of the rare alleles. Odds Ratios (OR) for the codominant and dominant models were calculated by logistic regression. Nominal P values were reported. IL-10R1, and IL-10 haplotypes were further analyzed with a more powerful haplotype trend regression approach [34], where an estimated haplotype matrix of posterior probabilities for each individual was used in the regression models. The whole effect of haplotypes was assessed by the global tests followed by individual tests for each haplotype. Further, a stepwise regression approach was used to obtain the most parsimonious model. All regression models were adjusted for square root of nadir CD4+ T-cell count, log₁₀HIV-1 load, age, gender and HAART. All analyses were performed with SAS version 9.1 (SAS Institute, Cary, NC, USA). Throughout the manuscript cases represent study patients diagnosed with CMV retinitis and controls represent study patients who did not develop CMV retinitis.

Structural analysis of IL-10R1 amino acid changes was carried on the X-ray structure of human IL-10/IL-10R1 complex, pdb code 1Y6K [35].

RESULTS

Within European Americans (n = 750), patients with CMV retinitis were slightly younger (42.2 vs. 44.0, P = 0.01), had higher initial log₁₀HIV viral load (5.21 vs. 5.03, P = 0.03), and lower CD4+ T-cell counts (32 vs. 79, P < 0.001) compared to patients without CMV retinitis (Table 1). Similar trends were observed for African American patients (n = 534), where patients with CMV retinitis were younger (39.0 vs. 42.1, P = 0.004), had higher log₁₀HIV viral load (5.48 vs. 5.06, P < 0.001), and lower CD4+ T-cell counts (28 vs. 67, P < 0.001) compared to patients without CMV retinitis. Also, the fraction of African American patients on HAART was lower in the CMV cases (67% vs. 82%, P = 0.001) compared to the controls (Table 1). In addition, there were significant differences (P ≤ 0.01) in terms of male gender percentage (91% vs. 67%), age (43.5 vs 41.5), nadir CD4+ T-cell counts (67.6 vs. 55.1), and time since AIDS diagnosis before study enrollment (5.2 vs 4.0) between European American and African American study patients. Due to significant differences of these clinical covariates between and within groups, the dominant and codominant association models were adjusted for nadir CD4+ T-cell count, log₁₀HIV viral load, age, gender and HAART.

TABLE 1.

Clinical aspects of European American and African American LSOCA patients used in this study

Variable	European Americans				African Americans
	Cases (n=200)		Controls (n=550)		Cases (n=110)		Controls (n=424)
	Mean±SD	Median (25th%, 75th%-tile)	Mean±SD	Median (25th%, 75th%-tile)	Mean±SD	Median (25th%, 75th%-tile)	Mean±SD	Median (25th%,75th%-tile)
Male gender (%)	92		91		64		68
Age at study entry (yrs)a	42.2±7.2	41.0 (37, 47)	44.0±8.6	43.0 (38, 49)	39.0±8.2	38.0 (34,44)	42.1±8.4	42.0 (37, 47)
Nadir CD4+ T-cell count (cells/mL)a	32.0±75.1	10.0 (5.25)	79.3±92.5	50.0 (18, 120)	28.4±66.9	9.0 (3, 22)	61.6±70.2	32.0 (10, 95)
Baseline HIV viral load (log10 copies/mL)a	5.2±1.0	5.5 (4.9, 5.9)	5.0±0.9	5.3 (4.6, 5.7)	5.5±0.5	5.6 (5.2, 5.9)	5.1±0.9	5.3 (4.7, 5.7)
Time since AIDS diagnosis (yrs)b	5.5±3.3	5.2 (3.3, 7.3)	5.1±3.7	4.7 (2.1, 7.2)	3.8±2.9	3.5 (1.5, 5.7)	4.0±3.6	3.1 (1.0, 6.2)
HAART use (%)c	81		84		67		82

^aSignificantly different between CMV retinitis cases and controls in European Americans and African Americans (see Results section for details)

^bYears before study entry

^cSignificantly different between CMV retinitis cases and controls in African Americans (P = 0.001)

European American Analyses

In SNP-based analyses, the S (serine) to L (leucine) amino acid changing (replacement) mutation (S420L), rs2229114, that is present in the cytoplasmic domain of IL-10R1 was over-represented in patients without CMV retinitis in all three models of association tests (P = 0.03–0.04; Table 2). Moreover, none of the homozygote carriers of this mutation developed CMV retinitis. In contrast, another replacement (isoleucine to valine change) mutation, rs2228055, and an intronic variant, rs4252286, showed increased frequency in CMV cases, but the association tests were not significant (Table 2). Previously reported strong linkage disequilibrium between rs3135932 (SNP3) and rs2229113 (SNP4) also was observed in our data (D′ = 0.90, P = 10⁻⁵⁵). However, rs2229114, rs2228055, and rs4252286 did not show significant linkage disequilibrium with each other or with any other SNP (Supplemental Figure 1).

Table 2.

Allelic distribution and association tests of IL-10R1 polymorphisms in CMV retinitis cases and controls

Gene	SNP	European Americans								African Americans
		Allele Frequency (%)		Allelic		Codominanta		Dominanta		Allele Frequency (%)		Allelic		Codominanta		Dominanta
		Cases(n=200)	Controls (n=550)	OR	P	OR	P	OR	P	Cases(n=110)	Controls (n=424)	OR	P	OR	P	OR	P
IL10-R1b
	rs3135932 (A/G)	17	14	1.21	0.23	1.27	0.18	1.39	0.11	2	3	0.72	0.49	0.72	0.54	0.72	0.54
	rs2228055 (A/G)	7	5	1.49	0.11	1.55	0.14	1.55	0.16	0.5	2	0.24	0.13	0.24	0.20	0.24	0.20
	rs4252279 (C/T)	9	11	0.81	0.31	0.79	0.26	0.87	0.60	11	11	1.00	0.99	0.99	0.97	0.98	0.94
	rs4252314 (A/G)	3	4	0.68	0.26	0.66	0.27	0.59	0.21	0.5	1	0.48	0.48	0.56	0.61	0.56	0.61
	rs4252286 (G/A)	4	2	1.61	0.15	1.95	0.08	1.95	0.08	0.9	0.4	2.50	0.28	1.22	0.84	1.22	0.84
	rs2229113 (G/A)	30	29	1.03	0.81	1.08	0.61	1.19	0.35	19	20	1.07	0.74	1.02	0.92	1.07	0.75
	rs2229114 (C/T)	2	5	0.44	0.03	0.42	0.03	0.42	0.04	2	0.6	3.25	0.06	1.61	0.60	1.61	0.60

^aLogistic regression model results adjusted for square root of nadir CD4+ T-cell count, log₁₀HIV-1 load, age, gender and HAART.

^brs3135932, rs2228055, rs2229113, rs2229114 are also know as SNP3, SNP6, SNP4 and SNP8, respectively in literature [19]. rs2229113 and rs2229114 were previouslt examined for AIDS progression association [24].

We were concerned that the rs2229114-T CMVR protective effect could also be associated with AIDS progression. So, we compared the allelic distribtuion of rs2229114-T in the patients with the highest and lowest 25% percentile CD4+ T-cell counts and HIV viral load among the CMV-negative patients. The rs2229114-T frequency was not significantly different (P = 0.44) between the highest and lowest 25% percentile CD4+ T-cell groups (5.4% vs. 4.3%), and was not significantly different (P = 0.78) between the highest and lowest 25% percentile HIV viral load groups (4.2% vs. 4.4%). Similar analyses with the low and high 50% percentiles also did not show significant rs2229114-T frequency distribution difference among these groups (CD4+ T-cell: 5.0% vs. 3.0%, P = 0.23; HIV viral load: 5.3% vs. 3.2%, P = 0.61).

None of the examined AIDS progression and therapy response related gene variants had signifiant association with CMV retinitis (Supplemental Table 1). Although an allelic association test indicated CMV retinitis susceptibility for the CCR5 promoter haplotype, +.P1.+, the effect was conflicting (codominant vs. dominant model) after correcting for the clinical covariates (Supplemental Table 1).

Analyses of haplotypes also indicated that certain IL-10R1 haplotypes were associated significantly with CMV retinitis development. In the haplotype trend regression analysis, eight parsimonious haplotypes were identified of which the AACAGGT was protective (OR = 0.14, P = 0.04) against, whereas the AGCAGGC was susceptible (OR = 6.21, P = 0.03) to CMV retinitis (Table 3). When compared against the most common and wild type haplotype AACAGGC, the protective effect of AACAGGT was still evident in a dominant (OR = 0.38, P = 0.03) and in an additive model (OR = 0.38, P = 0.03). Also, patients with the AGCAGGC haplotype had higher odds of CMV retinitis development compared to the patients with the wild type haplotype in a dominant (OR = 1.72, P = 0.05) and in an additive model (OR = 1.92, P = 0.05) analysis. All haplotype associations were significant irrespective of the clinical covariate corrections. Moreover, the haplotype analyses results agreed with the SNP-based analyses. The protective haplotype AACAGGT had only a single variant difference compared to the wild type haplotype AACAGGC, which was the protective SNP rs2229114. Similarly, the susceptible haplotype AGCAGGC had only one variant difference compared to the wild type haplotype, that is, the rs2228055 SNP that displayed a non-significant trend towards CMV retinitis susceptibility.

Table 3.

Haplotype Trend Regression Analyses of single nucleotide polymorphisms (SNP) in IL-10R1.

Gene, haplotypes	European Americans					African Americans
	Haplotype Frequency (%)		P		OR (95% CI)	Haplotype Frequency (%)		P		OR(95% CI)
	Cases (n=200)	Controls (n=550)	Global	Haplotype		Cases (n=110)	Controls (n=424)	Global	Haplotype
IL-10R1			0.02					0.08
AACAGGC	49.3	51.2		0.94	0.98 (0.58–1.65)	60.4	68.9		0.05	0.51 (0.25–1.00)
AACAAGC	3.6	1.9		0.12	4.19 (0.69–25.49)	0.9	0.4		0.85	1.45 (0.03–63.85)
AACAGAC	13.3	13.9		0.87	1.07 (0.49–2.32)	18.9	15.7		0.20	1.77 (0.74–4.25)
AACAGGT	1.7	4.1		0.04	0.14 (0.02–0.94)	2.2	0.4		0.25	7.82 (0.24–260)
AATAGGC	8.6	10.1		0.35	0.67 (0.28–1.56)	12.1	10.2		0.55	1.35 (0.50–3.63)
AGCAGGC	5.7	3.5		0.03	6.21 (1.22–31.54)	0.9	1.6		0.36	0.22 (0.01–5.78)
GACAGAC	12.1	9.2		0.12	1.96 (0.84–4.56)	2.2	2.0		0.70	1.65 (0.13–20.42)
GACGGAC	2.7	3.6		0.40	0.51 (0.11–2.45)	1.8	0.6		0.27	6.72 (0.23–193)

Haplotypes were estimated using the ordered SNPs: rs3135932, rs2228055, rs4252279, rs4252314, rs4252286, rs2229113, rs2229114.

Regression models are adjusted for square root of nadir CD4+ T-cell count, log₁₀HIV-1 load, age, gender and HAART.

We also evaluted potential structural implications that may be associated with the replacement (isoleucine (Ile) to valine (Val) change) SNP rs2228055. Ile224 is located in the proximity of the membrane and included in the large internal hydrophobic core in its bottom part formed by Leu132, Ile134, Ile139, Val203, Phe201 and Leu226. The replacement mutation Ile to Val (I224V, rs2228055) is a change of one hydrophobic to another hydrophobic residue. However, the volume of Val residue is much less than that of Ile (Supplemental Figure 2).

Omnibus CMV association analyses of IL-10 haplotypes were not significant (results not shown). Previously reported [26, 36] IL-10 proximal promoter haplotypes, formed by SNPs rs1800872, rs1800871 and rs1800896 (such as ATA, ACC and GCC) and associated with differential IL-10 production, were inferred with proxy SNPs rs1800872 and rs1800896 because of the high linkage disequilibrium between rs1800871 and rs1800872. Individual IL-10 haplotypes or grouping of these proximal promoter haplotypes did not have a significant effect on the development of CMV retinitis (results not shown).

African American Analyses

None of the examined IL-10R1 SNPs showed any statistically significant association with CMV retinitis (Table 2). Two IL-10 SNPs, rs3024500 and rs3024496, also showed a trend for increased susceptibility for CMV in models adjusted for the clinical covariates (P = 0.02–0.05; Supplemental Table 2). There was a strong linkage disequilibrium between these two SNPs (D′ = 0.9, P = 10⁻⁷⁸), which implies that the susceptibility signal may be driven by just one of these SNPs or by an untyped SNP in this haplotype group. Moreover, the IL-10 rs4072227 variant showed a significant increase in CMV retinitis risk in an allelic model, but that effect was not as significant after correcting for clinical covariates (Supplemental Table 2). As CD4+ T-cell count, HIV viral load, and HAART have significant effects on CMV, we tested whether the aforementioned IL-10 SNPs had any effect on these clinical covariates but we did not observe any significant effects. Finally, SNP and haplotype analyses of the examined AIDS progression and therapy response related gene variants did not show any significant effect on CMV retinitis or on any clinical covariates tested (Supplemental Table 1).

Global CMV association tests for IL-10R1, and IL-10 haplotypes were not significant in African Americans (Table 3; results not shown). However, when haplotypes were examined individually, the IL-10R1 AACAGGC haplotype trended towards protection against CMV retinitis.

DISCUSSION

In this study we used a candidate gene approach to examine the potential effects of IL-10R1 mutations on CMV retinitis development. Upon observation of significant associations of the examined IL-10R1 variants with CMV outcome, we extended our analyses to IL-10 variants. Moreover, we examined variants in genes that have been shown to influence HIV-1 infection, AIDS progression, therapy response and antiviral drug metabolism to account for potential genetic confounding factors. Our results suggest that a haplotype carrying an amino acid changing variation in the cytoplasmic domain of IL-10R1 can be protective against, whereas another haplotype carrying an amino acid changing variation in the extracellular domain of IL-10R1 can be more susceptible to CMV retinitis. Overall, our results indicate that IL-10R1 amino acid changing mutations show a significant association with onset of CMV retinitis even after adjustment for the other genes and clinical covariates examined in European Americans. However, we note that none of our nominally significant P values will pass conservative Bonferroni multiple test correction.

The observed effects of IL-10R1 variants on development of CMV retinitis were specific to European Americans. In African American patients, similar effects of IL-10R1 variants were not observed, possibly because the derived allele and the associated haplotype frequencies were too low for any statistical power. A potential increased CMV retinitis susceptibility, specific to African Americans, was suggested by two IL-10 variants (that are in high linkage disequilibrium) after adjusting for the effects of AIDS related clinical covariates. The IL-10 rs3024500-C variant, over represented in African American patients with CMV retinitis, had previously been shown to increase the risk for severe complications of human ocular Chlamydia trachomatis infection in a Gambian population [37]. Moreover, rs3024500-C allele was part of a higher IL-10 transcript producing haplotype that has been associated with susceptibility to sequelae of human ocular chlamydial infection [38]. Individual IL-10 polymorphism effects on the development of CMV retinitis in LSOCA African American patients may result from complex interactions between human IL-10 production, HIV infection progression, and HCMV activity. Statistically powerful modeling of potential effects of African American IL-10 variants (and their haplotypes) on the development of CMV retinitis requires more patients than were available for this study.

IL-10/IL-10R signaling path is a key pathway for viral persistence. Animal models lacking IL10 exhibit enhanced pathogen clearance [39–41] and blockage of IL-10/IL-10R signaling pathway leads to resolution of chronic viral infection [42, 43]. Thus, it is not surprising that human CMV uses its IL-10 homolog to bind to IL-10R1 [13] and to lead to immunosuppression [18]. The cmvIL-10 shows only 27% amino acid similarity to human IL-10, yet it has similar IL-10R1 affinity as does that of human IL-10. However, cmvIL-10 has a different interdomain angle compared to human IL-10 that changes the orientation of IL-10R1 in the putative cell surface complex [17]. Due to this structural difference, naturally occurring IL-10R1 amino acid polymorphisms may destabilize the cmvIL-10/IL-10R1 complex. Indeed, the common IL-10R1 variants rs3135932 (SNP3) and rs2229113 (SNP4) differentially reduce the signaling activity of cmvIL-10 [21]. In this study, we did not observe significant effects of these two SNPs on the development of CMV retinitis. Although rs3135932 and rs2229113 variants have been shown to reduce cmvIL-10 (and human IL-10) induced signaling activity, they do not totally diminish the signaling through the IL-10 receptor complex [20, 21]. These two SNPs may specifically affect the time to CMV-related disease. However, as our analyses concentrated on the end-point CMV retinitis, we might not have captured the potential effects of rs3135932 and rs2229113. The SNPs rs2229114 and rs2228055 and the associated haplotypes that showed an effect on the development of CMV retinitis in this study have not been investigated in any infectious disease outcome so far. Although the mutation Ile to Val (I224V, rs2228055) is a change of hydrophobic to another hydrophobic residue, the Ile is a much larger amino acid than the Val. For this reason the I224V change may affect the conformation of IL-10R1 structure in the proximity of the membrane, which in turn may affect the transduction of downstream signaling of the IL-10R complex. It is much harder to model the potential effects of the S420L (rs2229114) substitution, because this change involves a residue that is in the cytoplasmic domain of IL-10R1, for which no published crystal structure is available. However, 420S residue is a highly conserved residue in chimpanzee, mouse, rat, dog, and cow IL-10R1. Moreover, the cytoplasmic residues are thought to be involved in STAT3 binding and signal transduction. Therefore, S420L change may lead to a reduction in the IL-10 signaling pathway. Antibody blockade of the IL-10/IL-10R pathway in vitro has been shown to enhance CD4+ T-cell responses in samples from patients chronically infected with HIV [44]. Our findings suggest that a similar IL-10/IL-10R blockade may be useful in controlling chronic HCMV infections.

No world population, whether industrialized or not, is immune to HCMV infection. Recent surveys show that the prevalence of HCMV reaches up to 90–100% in some populations (i.e., Africa and southeast Asia) where incidence of HIV infections is also high [45]. Although usually associated with mild or no clinical symptoms, HCMV infections can lead to drastic end stage organ diseases which may be fatal when activated in immunocompromised patients. Our findings suggest that host genetics may play a crucial role in the development of one of these end stage organ diseases, CMV retinitis, in patients with AIDS. Previously, autopsy studies showed high incidence of other organ CMV infections in patients with CMV retinitis [46]. Indeed, 13% of the LSOCA patients who developed CMV retinitis in this study were also diagnosed with various CMV-related diseases including pneumonitis, pancreatitis, gastritis, and colitis, indicating the system wide disease of patients that have CMV activation. Moreover, patients co-infected with HIV and CMV progress more rapidly to AIDS and death than HIV infected patients who are not infected with CMV [11, 47–49]. Our results should stimulate further research, especially with longitudinal follow-up studies, to elucidate the role of IL-10 signalling and other host genetics in initiation and progression of CMV-related diseases.