CMV and EBV Co-Infection in HIV-Infected Children: Infection Rates and Analysis of Differential Expression of Cytokines in HIV Mono- and HIV–CMV–EBV Co-Infected Groups

(1) Background: CMV and EBV co-infections can affect the HIV disease progression by modulating the immune system. The disease dynamics can differ in HIV-positive adults and children. In Pakistan, HIV is rapidly expanding, especially in children; however, the prevalence of CMV and EBV co-infection and the effect on immune modulation in HIV-positive children are not known. This study aimed to bridge this gap by estimating the rate of active CMV and EBV co-infection in HIV-positive children, followed by the analysis of differential expression of cytokines in HIV mono- and HIV/CMV/EBV co-infected children. (2) Methods: DNA samples from 319 HIV-positive children, previously recruited as part of a study to investigate the HIV outbreak in Larkana, Pakistan, in 2019, were screened for CMV and EBV through qPCR. Subsequently, differences in HIV viral loads and CD4 counts were analyzed between the HIV mono- and HIV/CMV/EBV co-infected groups. The RNA samples were used to determine the differential expression of both pro- and anti-inflammatory cytokines in the mono- and co-infected groups using RT-qPCR, while unpaired T-test and Pearson correlation test were applied to, respectively, analyze the differential cytokine expression and correlation between cytokine in the two groups. (3) Results: Of 319 samples, the rate of active EBV and CMV co-infection in HIV-positive children was observed in 79.9% and 38.9%, respectively. A significant difference was observed in HIV viral load between HIV mono- and co-infected groups. IFN-γ expression was found to be lower in the HIV mono-infected group, while higher in all other three co-infected groups. Meanwhile, mRNA expression of TGF-β1 was found to be lower in HIV mono- and HIV–CMV–EBV co-infected groups, while higher in HIV–CMV and HIV–EBV co-infected groups. IFN-γ and IL-2 exhibited a significant positive correlation in all except HIV–CMV co-infected group. (4) Conclusions: The study suggests that the presence of EBV/CMV co-infection can affect the HIV viral loads and expression of certain cytokines (IFN-γ and TGF-β1), which may affect the HIV disease dynamics in infected children.


Introduction
Since the advent of antiretroviral therapy (ART), the span and quality of life of HIVinfected patients has significantly improved. However, HIV-related morbidity and mortality are still high in certain countries, especially in the low-and middle-income countries (LMIC), primarily because of low ART coverage, poor adherence to ART, immune dysfunction, inflammation, and chronic co-infections, such as Epstein-Barr virus (EBV), cytomegalovirus (CMV) infections, etc. [1,2].
For cytokine analysis, 10 µL sample reaction mix was prepared using the following recipe: 5 µL of BlasTaq™ 2X PCR Master Mix (cat# G891, ABM, Canada), 1 µL of primer mix (10 µM forward and reverse primers; Table 1), 1 µL cDNA, and 2 µL of nuclease-free H 2 O. The qPCR reaction was performed using the following thermal-cycling protocol: 3 min at 95 • C, 40 cycles of 15 s at 95 • C, and 58 • C for 30 s. Melt curve (55-95 • C) analysis was performed at the end of 40 cycles to confirm the specificity of the PCR products. All reactions were run in duplicate. β-actin was used as a housekeeping gene to normalize the expression of cytokines. Additionally, non-template/non-primer control (NTC/NPC) was included on each plate as a negative control for each primer. This strategy has been optimized previously in our laboratory [32][33][34]. The cytokine expression was determined using the ∆Ct method [35,36]. Table 1. Name of target genes and respective primer sets used to quantify mRNA levels in qPCR.

Statistical Analysis
An unpaired T-test was applied to measure the significant difference in the mean HIV viral load, CD4+ cell count, and cytokine gene expression between each of the four groups. Similarly, Pearson correlation was employed to analyze the correlation between cytokine expressions in each group. SPSS version 20 was used for all statistical analyses, where a p < 0.05 was considered to be significant.
The highest mean HIV viral load was observed in the HIV-EBV co-infected group, while the lowest mean HIV viral (341 copies/mL) load was observed in CMV co-infected patients ( Table 2). The CD4 count in all four groups ranged from 1094 to 1144 cells/ mm 3 ( Table 2). The statistical analysis showed no significant difference in CD4 count in all four groups.

Correlation between Differentially Expressed Cytokines in HIV Mono-and Co-Infected Groups
Since IFN-γ and TGF-β1 were found to be differentially expressed in mono-and coinfected groups, in the next step, we determined the correlation between the gene expression of IFN-γ and TGF-β1 and other cytokines in all four groups independently in order to identify the influence (positive or negative) of one cytokine to another [37,38]. In the HIV mono-infected group, a significant positive correlation was observed between IFN-γ and IL-2 (r = 0.37, p = 0.008), and IFN-γ and IL-10 (r = 0.30, p = 0.033). Similarly, expression of TGF-β1 and IL-4 (r = 0.63, p = 0.00), TGF-β1 and IL-10 (r = 0.54, p = 0.00), and TGF-β1 and TNF-α (r = 0.48, p = 0.00) were also positively correlated (Table 3). Table 3. Correlation of eight cytokines in HIV mono-and co-infected groups. Each column shows the R-value (the coefficient of correlation). Correlations with p < 0.05 are indicated with * and p < 0.01 are indicated with **.

HIV+ Mono-Infected Group
In  (Table 3). No significant positive or negative correlation was observed in the HIV/CMV group.

Discussion
In this study, we investigated the rate of active EBV and CMV infection in the samples collected from HIV-positive children during the 2019 Larkana outbreak. Subsequently, we performed a comparative assessment of cytokine expression in HIV mono-infected and HIV/CMV/EBV co-infected samples.
The majority (80%) of the HIV-positive children were found to be co-infected with EBV, while~40% with CMV, and, out of these, 18.5% with both EBV and CMV. Variable prevalence of CMV in HIV-positive patients has been reported from different parts of the world, for example, 32.4% (adults; age: 19.5-41.5 years) in India [39], 94% (both children and adults; age: 3-58 years) in Iran [40], 12.1% (infants) in Nigeria [41], 79% (age: 6-weekold infants) in Zimbabwe [42], and 10.3% (neonates) in France [43]; however, contrary to reported prevalence worldwide, we observed a CMV infection rate of about 40% in HIV-positive children from our cohort. Conversely, a high rate (80%) of EBV co-infection was observed in our cohort, which matches the rates reported in North India (62%) [44] and the Netherlands (64%) [45], while it is higher than the rates reported in Kenya (38.6%) [46]. Previous studies from the US and Zimbabwe have reported co-infection of CMV and EBV with HIV separately [47,48]. A research study conducted on Kenyan infants (HIVinfected) observed 93.9% of the infants to be simultaneously co-infected with CMV and EBV, pointing to common transmission risk factors [42]. In Pakistan, few studies have reported HIV prevalence in children [23,25]; however, to the best of our knowledge, no study has reported the rate of active CMV and EBV infection in HIV-positive children. Our study, therefore, is the first report to describe high EBV and CMV infection rates in HIV-infected Pakistani children.
In the next step, we analyzed the differences in HIV viral load and CD4 count between HIV mono-and co-infected groups. The CD4 counts were comparable between the monoand co-infected groups; however, the HIV viral load was found to be significantly lower (p-value < 0.0001) among HIV-CMV and HIV-CMV-EBV as compared to the HIV monoinfected group. Santos et al. reported a high prevalence of EBV in HIV-seropositive individuals and showed that HIV viral load was a key factor for EBV (type 1 and 2) coinfection [49]. Similarly, another study found HIV viral load to be the risk factor for CMV co-infection [50]. It is speculated that, during herpesvirus/HIV co-infection, CD4 T cell proliferation increases, thereby expanding the target cell type susceptibility to HIV infection, resulting in a high HIV viral load [51][52][53][54].
To date, limited studies have analyzed the differential expression of cytokines in HIVinfected children co-infected with CMV and EBV. Therefore, in this study, we investigated the differential expression of cytokine transcripts (both pro-and anti-inflammatory) in HIV mono-infected and co-infected with CMV and EBV children. We found the expression level of IFN-γ to be significantly decreased in HIV mono-infection groups, while it was increased in HIV co-infected children. It is hypothesized that infants/neonates have reduced IFN-γproducing cells and IFN-γ levels [55], which, however, increases with age in HIV-infected patients [56]. IFN-γ is essential for the regulation of chronic and latent infection of herpes virus (alpha, beta, and gamma) [57][58][59]. Infection with CMV, in particular, has been shown to induce a significant expression of IFN-γ and other Th1 cytokines by effector CMVspecific effector T cells [60]. Similarly, studies have also shown a correlation between increased IFN-γ levels and EBV reactivation [61,62]. Limited studies have reported the expression of IFN-γ in HIV and CMV/EBV co-infection. The decreased IFN-γ observed in the mono-infected group may be attributed to the acute or early chronic phase of HIV infection [63], while the presence of herpes virus co-infection may lead to increased IFN-γ and Th-1 response, which is also supported by correlation analysis, where IFN-γ expression correlated with IL-2 and TGF-β1. Interestingly, in the HIV mono-and HIV-CMV-EBV triple co-infection group, TGF-β1 expression levels were significantly lower, but they were higher in HIV-CMV and HIV-EBV co-infection groups. TGF-β1 production in HIV infection has varied kinetics depending on the cell type, which raises the possibility that TGF-β1 might play both positive and deleterious functions during infection. TGF-β1 has shown to be rapidly and systemically generated after acute HIV-1 infection and is maintained at an elevated level [64]. The concentration of TGF-β1 is linked with HIV disease progression; as the disease progresses, the TGF-β1 levels also increase [65,66]. CMV infection increases immunological tolerance in an immunocompromised environment by boosting TGF-1 transcription and release and suppressing cytotoxic Th1 cells [67][68][69]. Similarly, EBV cell lines generate TGF-1 and are resistant to TGF-1-mediated apoptosis and growth inhibition, which aids in the proliferation of EBV-infected cells [70]. It has been reported that other co-infection with HIV causes a significant increase in TGF-β1 levels [71,72]. Our study also showed an increase in TGF-β1 levels in the HIV/CMV/EBV co-infected group, which may enhance the CMV/EBV infection [73].
We identify certain limitations of this study. Firstly, the sample size of HIV/CMV co-infected patients was very low in this cohort. A comparison with a higher number of HIV/CMV samples might affect the outcomes related to the markers of HIV disease progression (CD4 count and viral loads) and/or cytokine expression. Secondly, due to the availability of only DNA and RNA samples (and the absence of serum/plasma samples), serological detection of CMV and EBV could not be performed. Due to the same reason, we could only analyze the mRNA expression of cytokines and not the protein expression. However, it is important to note that numerous studies have investigated changes in cytokine expression at the mRNA levels only and have reported differential cytokine gene expression [74,75].

Conclusions
In conclusion, the high co-infection with CMV and EBV in HIV-positive children may affect the HIV viral loads and expression of certain cytokines (IFN-γ and TGF-β1), which may affect the HIV disease dynamics. Further mechanistic understanding of the involvement of herpes viruses in HIV-positive children may provide insights into disease pathogenesis.

Informed Consent Statement:
The study was conducted after obtaining written informed assent from participants and informed consent from the parents/guardians. Written informed consent has been obtained from the patient(s) to publish this paper.
Data Availability Statement: The data are available within the manuscript or its Supplementary Files.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.