Increased HIV-1 transcriptional activity and infectious burden in peripheral blood and gut-associated CD4+ T cells expressing CD30

Abstract

HIV-1-infected cells persist indefinitely despite the use of combination antiretroviral therapy (ART), and novel therapeutic strategies to target and purge residual infected cells in individuals on ART are urgently needed. Here, we demonstrate that CD4+ T cell-associated HIV-1 RNA is often highly enriched in cells expressing CD30, and that cells expressing this marker considerably contribute to the total pool of transcriptionally active CD4+ lymphocytes in individuals on suppressive ART. Using in situ RNA hybridization studies, we show co-localization of CD30 with HIV-1 transcriptional activity in gut-associated lymphoid tissues. We also demonstrate that ex vivo treatment with brentuximab vedotin, an antibody-drug conjugate (ADC) that targets CD30, significantly reduces the total amount of HIV-1 DNA in peripheral blood mononuclear cells obtained from infected, ART-suppressed individuals. Finally, we observed that an HIV-1-infected individual, who received repeated brentuximab vedotin infusions for lymphoma, had no detectable virus in peripheral blood mononuclear cells. Overall, CD30 may be a marker of residual, transcriptionally active HIV-1 infected cells in the setting of suppressive ART. Given that CD30 is only expressed on a small number of total mononuclear cells, it is a potential therapeutic target of persistent HIV-1 infection.

Fig 1. Assessment of CD30 expression on CD4⁺ T cells.

(A) Percentage of CD4⁺ T cells expressing CD30 in samples from HIV-1-uninfected (n = 10) ART suppressed (n = 17) and viremic donors (n = 9) are shown. Surface expression was higher in ART suppressed and viremic HIV-1-infected individuals (p = 0.0023 and p = 0.045 respectively). (B) sCD30 (IU) in plasma was highest in viremic (n = 4) compared with HIV-1-uninfected (n = 10) and ART suppressed donors (n = 9) (p<0.001 and p<0.001 respectively). (C) No significant correlations were identified between the percentage of CD4⁺CD30⁺ T cells and sCD30 levels (r = 0.20, P = 0.35 by Spearman rank correlation analysis). The percentage of CD4⁺ T cells co-expressing CD69 in each cohort are shown in (D) and the percentage of CD69⁺CD4⁺ T cells expressing CD30 are shown in (E). The percentage of CD4⁺ T cells co-expressing HLA-DR (F) and HLA-DR⁺CD4⁺ T cells expressing CD30 (G) are shown. Despite significant increases in CD30⁺ cells expressing CD38/HLA-DR, there were no significant differences in the frequency of CD30 expression in CD38/HLA-DR-expressing CD4⁺ T cell populations. (H) The highest frequency of CD30⁺ T cells observed in HIV-1-infected individuals were of transitional/effector memory phenotype, compared to naïve T cells in healthy controls. (I) CD30⁺CD4⁺ and CD30^-CD4⁺ T cells from viremic donors expressed significantly more PD1 than healthy controls (p = 0.02 and p = 0.02 respectively), CD30⁺CD4⁺ and CD30^-CD4⁺ T cells from ART suppressed donors also expressed significantly more PD1 than healthy controls (p = 0.002 and p = 0.002 respectively). (J) Conversely, very few CD4⁺PD1⁺ T cells co-expressed CD30, but ART suppressed donors had significantly higher expression compared to healthy controls (p = 0.01). T Bars represent median ± interquartile range for all data. *p<0.05; **P < 0.01; ***P < 0.001. Significant intergroup differences were determined using rank Kurskal-Wallis tests incorporating Dunn's tests for multiple comparisons. Wilcoxon matched-pairs signed rank tests were used to determine statistical significant between CD30 intergroup, paired samples.

Fig 2. HIV-1 is enriched in CD30 expressing CD4⁺ T cells.

Cell-associated HIV-1 unspliced RNA (A) and DNA (B) in sorted CD4⁺ T cell populations, in samples from ART suppressed (n = 17) and viremic donors (n = 9) are shown. HIV-1 RNA is significantly enriched in CD30 expressing cells from ART suppressed donors (p = 0.008) and viremic donors (p = 0.007). (C) The contribution of CD30⁺CD4⁺ T cells to total CD4⁺ T cell population, and the contribution of HIV-1 RNA (D) and DNA (E) from CD30⁺CD4⁺ sorted cells to total HIV-1 RNA and DNA burden in CD4⁺ T cells, are shown. Despite the rarity of CD30⁺ T cells, a large contribution of HIV-1 DNA and RNA are attributed to these cells in some individuals. The percentage of total HIV-1 RNA (F) but not DNA (G) found in CD30⁺CD4⁺ T cells was significantly higher in African American (n = 13) than white (n = 12) participants (p = 0.0103). Bars represent mean ± standard deviation; *P < 0.05; **P < 0.01; Significant intergroup differences were determined using Kruskal-Wallis tests incorporating Dunn's tests for multiple comparisons. Wilcoxon matched-pairs signed rank tests were used to determine statistical significant between CD30 intergroup, paired samples.

Fig 3. HIV-1 RNA and DNA quantification in rectal tissue-derived CD4⁺ T cell subsets.

HIV-1 RNA and DNA levels for each CD4⁺ T cell subset (based on co-expression of CD30 and CD32) are shown in A and B, respectively. (C) A significantly higher mean HLA-DR MFI was observed in CD30⁺CD32⁺ cells compared with CD30^-CD32^-CD4⁺ T cells (P = 0.004 by Friedman test with Dunns correction for multiple comparisons). In contrast, no significant intergroup differences were observed in cell-associated HIV-1 RNA or DNA between cohorts. Bars represent mean ± standard deviation.

Fig 4. Cell-associated HIV-1 RNA and CD30 RNA co-localize in human rectal and ileal tissue from HIV-1-infected individuals.

In situ chromogenic hybridization staining of HIV-1 RNA (pink) and CD30 RNA (black) was performed in ileum and rectum from HIV-1-infected individuals. Less than 0.1% of uninfected GALT cells expressed CD30 RNA (A). Although HIV-1 RNA⁺ staining alone was identified in sections from GALT from viremic HIV-1-infected individuals (n = 3) (B), 33% of HIV-1 RNA⁺ cells also expressed CD30 (C). In GALT from ART-suppressed individuals (n = 3), 88% of all HIV-1 RNA positive cells co-expressed CD30 (D). Percentages of HIV-1 RNA⁺ expressing CD30 RNA are shown in (E). 308 HIV-1 RNA⁺ cells were observed in sections from all viremic samples and 25 RNA⁺ cells were observed in tissues sections from suppressed individuals. Nearly all cells with high levels of CD30 RNA also expressed HIV-1 RNA, and minimal background CD30 RNA staining was observed in each field.

Fig 5. Anti-CD30 therapy reduces total HIV-1 DNA burden ex vivo.

Following ex vivo culture of PBMC in the presence of antiretroviral drugs (raltegravir, 3TC) and brentuximab vedotin for 5 days, we observed a significant reduction in the mean total cell-associated HIV-1 DNA from seven ART-suppressed individuals, despite no change in viable cell numbers. Significant overall decreases were found within samples at the higher input concentrations (100μg/ml p = 0.0469, 500μg/ml p = 0.0390). Significant intergroup differences were determined using paired, nonparametric Friedman tests incorporating Dunn's tests for multiple comparisons. All experimental wells with and without brentuximab vedotin were subject to the same input cell number and culture conditions.