Resistance to simian immunodeficiency virus low dose rectal challenge is associated with higher constitutive TRIM5α expression in PBMC

Abstract

Background: At least six host-encoded restriction factors (RFs), APOBEC3G, TRIM5α, tetherin, SAMHD1, schlafen 11, and Mx2 have now been shown to inhibit HIV and/or SIV replication in vitro. To determine their role in vivo in the resistance of macaques to mucosally-acquired SIV, we quantified both pre-exposure (basal) and post-exposure mRNA levels of these RFs, Mx1, and IFNγ in PBMC, lymph nodes, and duodenum of rhesus macaques undergoing weekly low dose rectal exposures to the primary isolate, SIV/DeltaB670.

Results: Repetitive challenge divided the monkeys into two groups with respect to their susceptibility to infection: highly susceptible (2-3 challenges, 5 monkeys) and poorly susceptible (≥6 challenges, 3 monkeys). Basal RF and Mx1 expression varied among the three tissues examined, with the lowest expression generally detected in duodenal tissues, and the highest observed in PBMC. The one exception was A3G whose basal expression was greatest in lymph nodes. Importantly, significantly higher basal expression of TRIM5α and Mx1 was observed in PBMC of animals more resistant to mucosal infection. Moreover, individual TRIM5α levels were stable throughout a year prior to infection. Post-exposure induction of these genes was also observed after virus appearance in plasma, with elevated levels in PBMC and duodenum transiently occurring 7-10 days post infection. They did not appear to have an effect on control of viremia. Interestingly, minimal to no induction was observed in the resistant animal that became an elite controller.

Conclusions: These results suggest that constitutively expressed TRIM5α appears to play a greater role in restricting mucosal transmission of SIV than that associated with type I interferon induction following virus entry. Surprisingly, this association was not observed with the other RFs. The higher basal expression of TRIM5α observed in PBMC than in duodenal tissues emphasizes the understated role of the second barrier to systemic infection involving the transport of virus from the mucosal compartment to the blood. Together, these observations provide a strong incentive for a more comprehensive examination of the intrinsic, variable control of constitutive expression of these genes in the sexual transmission of HIV.

Figure 1

Outcome of repeated, low-dose rectal challenges with SIV/DeltaB670. Indian-origin rhesus macaques received weekly low dose rectal challenges. Animals were challenged with 1cc of culture supernatant containing 250 TCID₅₀ SIV/DeltaB670 for the first challenge and 2500 TCID₅₀ for the remaining 5 challenges. Monkey 705 received a 7th challenge with 2.5 ×10⁵ TCID₅₀ 72 days after the 6th challenge. Blood was collected twice a week and copies of viral RNA were quantified in plasma by qRT-PCR using external standards. Animals were divided into two groups based on the number of challenges required for systemic infection: highly susceptible (red); 2–3 challenges (R700, R701, R702, R703, R704) and poorly susceptible (blue); ≥6 challenges (R697, R698, R705). Red dotted line at 10⁴ viral RNA copies indicates the threshold virus load (VL) associated with clinical disease in rhesus macaques infected with SIV/DeltaB670. Arrows indicate time points of challenge. Χ = sacrifice for tissue collection.

Figure 2

Basal ISG expression in PBMC, ILN and duodenum. Relative mRNA levels of (A) Mx1, (B) TRIM5α, (C) Tetherin, (D) A3G, (E) Mx2, (F) SAMHD1, (G) SCHL11, and (H) IFNγ was determined in PBMC, inguinal lymph node mononuclear cells (ILN), and duodenal tissue biopsies of the eight macaques. All samples were obtained 4 days prior to the first exposure (basal levels) on all animals except monkey R700 from which PBMC samples were obtained on day 0 (day of first exposure). NormFinder [64] was used to determine the most stable endogenous control (TBP, HPRT, β2M, βGus) among the three tissues. All gene expression values were normalized to TBP, the most stable endogenous control. Relative expression was determined using the ΔCt method and the following formula: 1000 x 2^-ΔCt. Horizontal lines denote mean expression levels in each tissue. Each symbol indicates the mean expression value for one animal. Significance was calculated using the Friedman test with Dunn's multiple comparison test. Asterisks indicate significance at P < 0.05.

Figure 3

Statistical analysis of basal ISG expression and susceptibility to infection. Relative basal mRNA levels of (A) Mx1, (B) TRIM5α, (C) Tetherin, (D) A3G, (E) Mx2, (F) SAMHD1, (G) SCHL11 were measured in the PBMC, ILN, and duodenum of the eight macaques, highly susceptible (R700-R704) and poorly susceptible animals (R697, R698, R705). Levels of basal ISG expression for each group were calculated as described in Figure 2. Each symbol indicates the mean expression value for one animal. Significance was calculated by the Mann–Whitney test. Asterisks indicate significance at P < 0.05.

Figure 4

Linear regression analysis of basal RF expression and resistance to infection. Basal ISG mRNA levels in (A) PBMC, (B) Duodenum, (C) ILN were calculated as described in Figure 2 and plotted versus the number of exposures required for systemic infection. Each dot indicates mean expression value of one animal. Lines depict linear regression analysis with r² and P values indicated next to each gene. Asterisks indicate significance with P < 0.05.

Figure 5

Basal TRIM5α mRNA levels over time. Relative basal mRNA levels of TRIM5α was measured 4, 26–33, and 289–298 days prior to the first challenge and plotted for the highly susceptible (R700-R702, R704) and poorly susceptible animals (R697, R698, R705). Values for R703 were omitted due to lack of sample at all 3 time points. GeNorm analysis by the GenEx 6 MultiD software (TATA Biocenter, Sweden) chose TBP and βGus as the most stable genes and all gene expression values were normalized to them using the ΔCt method and the following formula: 1000 × 2^-ΔCt. Significance was calculated by 2way ANOVA with Sidak’s multiple comparison test. Asterisks indicate significance at P < 0.05. The magnitude of the observed differences seen for TRIM5α is such that we have power >0.8 to state that these significant differences are genuine, based on determination of Cohen’s d as a measure of the Effect Size (ES) seen [55].

Figure 6

ISG induction in PBMC of macaques highly susceptible to infection. The fold change in the relative expression of each RF, Mx1, and IFNγ post-exposure are plotted for susceptible animals (A) R700, (B) R701, (C) R702, (D) R703, (E) R704. Each gene is depicted by a different colored bar as indicated in the figure legend. Plasma virus loads (VL) are shown by the black line. All samples were obtained 4 days prior to the first exposure (basal levels) on all animals except monkey R700 from which PBMC samples were obtained on day 0 (day of first exposure). Fold change in gene induction was calculated by dividing the relative mRNA level at each time point by the basal level obtained prior to first rectal challenge. Values were normalized to the endogenous controls TBP and HPRT. The black horizontal line at a 2-fold change indicates the threshold for the observed induction to be true. Asterisks next to day numbers indicate days of rectal challenge.

Figure 7

ISG induction in PBMC of macaques poorly susceptible to infection. The fold change in the relative expression of each RF, Mx1, and IFNγ are plotted for resistant animals (A) R697, (B) R698 as described in Figure 6.

Figure 8

ISG induction in PBMC of the elite controller. The fold change in the relative expression of each RF, Mx1, and IFNγ are plotted for the elite controller, R705, as described in Figure 6.

Figure 9

ISG induction in the duodenum during early challenges. The fold change in the relative expression of each RF, Mx1, and IFNγ are plotted for the duodenum of animals that became infected during the first 3 challenges (A) R700, (B) R701, (C) R702 as described in Figure 6.