Polymorphisms of interferon γ and NKG2D receptor genes in predicting vertical transmission of HIV/HCV coinfection

Introduction

Infections caused by human immunodeficiency virus (HIV) and hepatitis C virus (HCV) remain one of the major global health problems. According to 2017 data, 2.3 million people living with HIV were simultaneously infected with HCV [1]. Particular attention of researchers in this area is focused on the problem of HIV/HCV co-infection among women of fertile age [2][3].

Data from a meta-analysis, including more than 900 sources from around the world, revealed that on average the proportion of pregnant women additionally infected with HCV was 2.4% in the general population of HIV-infected persons, [4]. In accordance with modern methodological approaches, determining the risk of vertical transmission of HCV in coinfected women requires regarding the HCV viral load and HIV status. In relation to HIV, it should be noted that the administration of highly active antiretroviral therapy and a low viral load are important prognostic factors for reducing the vertical transmission of this virus in recent years [5]. Some authors reported that the proportion of children born to mothers with HIV/HCV coinfection and infected with HCV was 5.2% in developed countries [6].

Currently, the role of the placenta as an epidemic risk factor for the vertical transmission of HIV and HCV in comorbid conditions is widely discussed [7]. The fact is that 15–35% of cases of perinatal transmission of the virus occur through transplacental pathways, including intrauterine and antenatal penetration ones; 50–75% of cases are due to infection during childbirth, and 10–20% during breastfeeding [8].

From the point of view of the significance of transplacental transmission of pathogens, the immune aspect in women deserves special attention. In particular, that is stipulated by the fact that HIV and, to a lesser extent, HCV belong to the virus category, which can exhibit an immunotropic effect [9][10]. Moreover, psychoactive substances, the abuse of which is a common concomitant of HIV and HCV infection, suppress immune responses and can affect viral pathogenesis [11].

In the light of investigation of the immune mechanism that is most important in the transplacental transmission of viruses, the population of innate immune cells, namely natural killer cells, deserves special attention. In the female reproductive organs, in particular, in the uterus (uNK) and in the decidua (dNK) at pregnancy, these lymphocytes produce a pool of tissue-resident natural killer cells with distinctive characteristics. This is related to the peculiarities of the immune system response to the development of pregnancy; since the fetus is semi-allogeneic for the mother, the pregnancy must induce mechanisms preventing its rejection [12].

The main functions of dNK are the control of trophoblast ingrowth into the decidua through the cytotoxic effects of these cells [13], as well as the production of cytokines, growth factors, angiogenic and other factors, which determine neoangiogenesis, tissue remodeling, and placental development. The importance of a high level of IFNγ produced by dNK is stipulated by its involvement in the formation of fetal resistance to viral infection [14].

The identical aspect of tissue-resident natural killer cells localized in the decidua of the uterus in the gestational process, which determines the state of the mother-fetus interface, has so far been beyond the attention of researchers. Meanwhile, the current state of scientific knowledge on this problem has been significantly expanded with findings suggesting the importance of this immune mechanism both in maintaining the physiological course of pregnancy and in the cellular immune protection of the fetus from the infection possibility [15].

The NKG2D receptor is one of the sensitive elements for the activation of natural killer cells, including those in the decidua of the uterus during pregnancy. In this case, its ligands are stress-induced molecules of the trophoblast and placenta, especially at their attack by viruses. In this context, IFNγ is the main secretory product of dNK [16].

Currently, there is evidence that polymorphism of the NKG2D gene affects susceptibility to infectious diseases, cancer, and autoimmune disorders [17], while the polymorphism of the allelic variant rs2617160 (TT genotype) of the NKG2D gene is the only marker and independent factor associated with susceptibility to chronic hepatitis B. Individuals with the TT rs2617160 genotype may demonstrate a significant decrease in natural cytotoxic activity during the elimination of hepatitis B virus (HBV) compared with carriers of the TA or AA genotypes, which contributes to the persistence of the virus and leads to chronic HBV infection [18].

Furthermore, it was found that the polymorphism of the allelic variant rs2430561 can affect the expression of the IFNγ gene [19], while the existence of alleles T and A at position +874 from the translation start site is associated with high and low production of IFNγ, respectively [20]. The latter fact was discovered already at the beginning of the 21st century. Polymorphisms in the non-coding regions of the IFNγ gene are associated with autoimmune disorders [21] and chronic infectious diseases [22], including the risk of chronic viral hepatitis B and C [23][24].

Data on assessing the significance of these single nucleotide polymorphisms in the transplacental transmission of HIV/HCV coinfection from mother to child could not be found in the available scientific literature.

The purpose of the study was the examination of single nucleotide polymorphisms of the rs2430561 allele of the IFNγ gene and the rs2617160 allele of the NKG2D gene as potential risk factors for intrauterine infection of a child born to a mother coinfected with HIV/HCV.

Materials and methods

The study included 65 mother-child pairs registered in the general outpatient and pediatric outpatient departments of the Moscow City AIDS Prevention and Control Center (MGC AIDS) of the Moscow State Budgetary Healthcare Institution “Infectious Clinical Hospital No. 2 of the Moscow Health Department” in the period from 2004 to 2021.

All women under clinical supervision were warned about participation in the study and signed informed voluntary consent in accordance with the principles of the Helsinki Declaration of the World Medical Association.

The age of mothers was from 17 to 49 years, and the age of children was from 6 months to 18 years. All mothers, without exception, were coinfected with HIV/HCV, HIV infection was at stage 3-4B. The mother-child pairs included in the study were divided into 4 groups according to perinatal transmission of viruses to the child:

• group 1: mother-child pairs with perinatal infection of the child with HIV and HCV, 9 pairs (14%);
• group 2: mother-child pairs with perinatal HIV infection of the child, 12 pairs (18%);
• group 3: mother-child pairs with perinatal HCV infection of the child, 17 pairs (26%);
• group 4: mother-child pairs with an inconclusive test for HIV in the child and the absence of HCV, 27 pairs (42%).

Dependence on taking injectable psychoactive drugs, including a history of it, in each group was revealed by 21% to 31% of women.

The study was prospective-retrospective in nature. The study protocol is presented in Figure 1.

All mothers and their children were assessed for single nucleotide polymorphisms of the IFNγ and NKG2D genes using multiplex polymerase chain reaction (PCR). PCR was carried out on the basis of the PCR analyzer “Abbott m2000rt” (Abbott Molecular Inc., USA) and the automated station for sample preparation and isolation of nucleic acids and proteins “QI Asymphony SP” (QIAGEN GmbH, Germany) in accordance with the user's manual for the devices.

At the preanalytical stage, venous blood was collected into tubes with the potassium salt of ethylenediaminetetraacetic acid (EDTA) to prevent coagulation; after mixing, 250 μl of whole blood was collected into labeled tubes and the fraction of leukocyte cells was washed using the reagent “HEMOLYTIC” (RU No. FSR 2010/09505 dated 09/04/19, AmpliSens LLC). The resulting cell suspension was taken to isolate total DNA using the AmpliPrime DNA-sorb-V reagent kit (RU No. FSR 2009/05220 dated 05/03/2019, NextBio LLC).

After DNA extraction, a study was carried out to identify single nucleotide polymorphisms using multiplex PCR. To determine polymorphisms in the NKG2D gene, the “TaqMan” reagent kit manufactured by Applied Biosystems was used; for genotyping polymorphic markers in the IFNg gene, the “rs2430561 IFNG gene” reagent kit manufactured by TestGen (series: 202111-411) was used.

This study analyzed retrospective data from outpatient monitoring cards of mothers in each trimester of pregnancy to identify their HIV and HCV viral loads assessed by PCR. In addition, using the results of an immunological study carried out by flow cytometry, the absolute number of lymphocytes with phenotypes CD3⁺CD4⁺ (T-helper) and CD3⁺CD8⁺ (cytotoxic T-lymphocytes, CTL) was determined to calculate the immunoregulatory index (IRI) as the ratio of these cells (CD3⁺CD4⁺/CD3⁺CD8⁺).

Results

The results of the investigation of the representation of various variants of single nucleotide polymorphism in the allelic variant rs2430561 of the IFNγ gene in both mother and child are presented in Table 1.

As it follows from the data in Table 1, determining the prognostic value of single nucleotide polymorphism of one of the alleles of the IFNγ gene showed ambiguous results. In particular, intergroup comparisons of the frequency of occurrence of this polymorphism in pregnant women coinfected with HIV/HCV revealed the absence of influence of this factor on the perinatal infection of the child. Meanwhile, the same single nucleotide polymorphism in a child turned out to be significant in the development of perinatal infection with HIV or HCV, including the two viruses simultaneously.

Thus, infection of a child with HIV as well as HIV/HCV is associated with the frequency of occurrence of the polymorphism of the rs2430561 allele in the IFNγ gene. In infected children, the AA variant was registered significantly less often than in children born healthy. The occurrence of the AT variant in infected children in both groups was higher. Approximately a quarter of children infected with HIV had a TT variant that was not found in healthy children, and therefore, the existence of this variant can presumably be considered a risk factor for perinatal HIV infection.

The identical analysis using comparative statistics showed that in the group of children infected with HCV, the AA variant was also registered less frequently than in healthy children; the TT variant was not registered at all in any of the comparison groups; the AT variant was determined 2.2 times more often in the group children infected with HCV, and presumably, it can serve as a nonspecific marker of infection with this virus.

Further, single nucleotide polymorphisms of the rs2617160 allele of the NKG2D gene were investigated in the mother and child by comparing data from examined groups consisting of the infected children (groups 1, 2, 3) with group 4, which did not contain perinatally infected children. The results of this study are presented in Table 2.

As it follows from Table 2, there is reason to suggest a relationship between the single nucleotide polymorphism of the rs2617160 allelic variant of the NKG2D gene and perinatal infection of the child with HIV and/or HCV if the mother has HIV/HCV co-infection. This polymorphism is important for both the mother and the child.

Indicatively, one-way analysis of variance revealed a relationship between single nucleotide polymorphism of the rs2617160 allele of the NKG2D gene in the mother and perinatal infection of the child with HIV and HCV simultaneously, as well as HIV alone. In this respect, the AA variant was recorded only in mothers with HIV/HCV co-infection, whose children were perinatally infected, and its frequency of occurrence was approximately a third of women in group 1 and a half of women in group 2 who were under observation. The remaining women were carriers of the AT variant, the frequency of which was not significantly different from that in group 4 consisted of healthy born children. The TT variant was recorded only in group 4.

The existence of AT and, to a lesser extent, AA genotypes of allele rs2617160 of the NKG2D gene was also typical for children perinatally coinfected with HIV/HCV, although this fact had no prognostic significance.

Thus, perinatal infection of a child with HIV (isolated or simultaneously with HCV) was associated with the existence of a single nucleotide polymorphism in the form of AT or AA genotypes of the rs2617160 allele of the NKG2D gene in a mother coinfected with HIV/HCV, which can be interpreted as a possibility of predicting the risk of a child becoming infected with HIV or HIV/HCV even before pregnancy.

A summary diagram of the relationship between the examined polymorphisms and the possibility of a child becoming infected with HIV and HCV is presented in Figure 2. The figure illustrates that the examined polymorphism of the IFNγ gene is significant only when detected in a child, that is, essentially it does not have prognostic significance without an invasive method of sampling for research. From this point of view, the NKG2D gene polymorphism is more promising, since it is informative when detected in both mother and child that requires further confirmation at the population level.

In order to elucidate the main mechanisms that determine the prognostic significance of these polymorphisms for intrauterine infection of the fetus, the viral load of HIV and HCV, as well as the absolute content of subpopulations of T-helper cells (CD3⁺CD4⁺) and cytotoxic T-lymphocytes (CTL, CD3⁺CD8⁺) were estimated in the blood of mothers of the examined groups in each trimester with the subsequent calculation of the immunoregulatory index as the ratio CD3⁺CD4⁺/CD3⁺CD8⁺.

The results of the estimation of the viral load of HIV and HCV in the blood of pregnant women of different examined groups in the form of 95% confidence intervals of the indicators are presented in Figure 3.

The data show that the viral load of HIV or HCV in pregnant women with HIV/HCV coinfection seems to have a prognostic value although it manifests itself differently for each virus and depends on the pregnancy trimester.

In the second trimester of pregnancy, the viral load of HIV in groups where the child was perinatally infected with HIV, HCV, or both viruses (group 1–3) was statistically different from this indicator in group 4, where vertical transmission of viruses from the mother did not occur. Thus, HIV viral load indicators in a woman, coinfected with HIV/HCV, in the second trimester of pregnancy make it possible to determine the risk of infection of the unborn child if the HIV viral load is above 950 copies/ml. In the first and third trimesters, at the same criterion level of viral load (>950 copies/ml), the prognostic value of this indicator applies only to group 1, in which children coinfected with HIV/HCV were born.

Indicators of the HCV viral load in a coinfected woman were informative for the prognosis of perinatal HCV infection of the child when they exceeded 620,000 copies/ml in the first trimester of pregnancy or exceeded 975,000 copies/ml in the second trimester; in the third trimester of pregnancy if these values were above 120,000 copies/ml, any variant of infection was possible for the child.

Characteristics of immunological data, namely subpopulations of T-lymphocytes in the blood of pregnant women with HIV/HCV coinfection in accordance to the examined groups as 95% confidence intervals of indicators are given in Figure 4.

The main feature of immunological changes in pregnant women coinfected with HIV/HCV related to the ability to provoke intrauterine infection of the fetus is the absence of intergroup differences for each indicator in the first trimester of pregnancy.

Thus, in mothers co-infected with HIV/HCV, various changes were found in both the viral load of HIV and HCV, and immunological parameters, which differed significantly among each other in the examined groups and manifested in various manners in the trimesters of pregnancy.

Summarizing the findings, it becomes obvious that several factors are important for the perinatal infection of a child. They include, in particular, the state of the mother’s body, as evidenced by shifts in immunological indicators that affect the pattern of infection in different groups. In addition, this process depends on the state of the interface “mother-fetus” (placenta), and, perhaps, on the condition of the fetus’s body, capable of counteracting the transition of viral agents from the mother to fetus in accordance with the level of viral load of pathogens.

In accordance with the assumptions made, parallels were drawn between single nucleotide polymorphisms of the IFNγ and NKG2D genes and the identified shifts in the studied parameters. First, this was done in relation to the viral load of HIV and HCV, and second, in relation to immunological parameters (Figure 5).

As shown above, single nucleotide polymorphism of the IFNγ gene in the allelic variant rs2430561, which was considered in the context of predicting the risk of having an infected child in a mother co-infected with HIV/HCV, did not have a prognostic value when determined in a pregnant woman. Meanwhile, the existence of this single-nucleotide polymorphism in a child was connected with the possibility of infection. Nevertheless, this fact cannot have a prognostic value, as determining this trait in the fetus during pregnancy would require an invasive intervention that would contribute to its infection. However, in the aspect of elucidating the mechanism of a child infection from an infected mother, this is of some interest. Immunological indicators of a mother as a risk factor for child infection were not considered, since the cellular elements of the immune systems of the mother and fetus function separately due to the placental barrier, although indirectly the immune factors have an impact on the viral load with HIV and HCV. In the authors’ opinion, investigation of the viral load of HIV and HCV by itself in the mother’s blood was appropriate in this case, taking into account the capacity of the viruses for vertical transmission.

As follows from Figure 4, coinfection of a child having the TT and AT genotypes of the rs2430561 allele in the IFNγ gene with HIV/HCV was highly probable if the mother, who was coinfected with HIV/HCV, had an HIV viral load of at least 2530 copies/ml during the entire pregnancy. These conditions must be added by the HCV viral load in the mother exceeding 1,450,000 copies/ml in the first trimester and 120,000 copies/ml in the third trimester.

The risk of a child becoming infected with HIV at the same single nucleotide polymorphism variant was associated with other conditions. In particular, when in the second trimester of pregnancy, the maternal HIV viral load was more than 950 copies/ml and in the third trimester, the HCV viral load exceeded 120,000 copies/ml. In this regard, it is interesting that HIV infection required not only a certain load of this virus in the mother’s blood in the second third of the gestational process but also a certain level of coinfecting hepatitis C virus in the blood in the later stages of pregnancy.

In this study, the risk of HCV infection in a child, having the TT and AT genotype of the rs2430561 allele of the IFNγ gene, was identified against the background of the HIV viral load above 950 copies/ml in the second trimester, as well as the HCV viral load above 975,000 copies/ml in the second trimester and above 360,000 copies/ml in the third trimester.

Discussion

The data obtained attest to the prognostic significance of single-nucleotide polymorphisms of the NKG2D gene, which partly reflects the level of the activation processes of natural killer cells in pregnant women coinfected with HIV/HCV. The results largely confirm the working hypothesis that the features of this important population of innate immune lymphocytes, which play a significant role in the formation of the placenta as an interface in the mother-fetus system, give an idea of the state of the barrier for the penetration of viral pathogens into the child’s body. In addition, the results demonstrate the importance for the infection of a child of such indicators in the blood of a pregnant woman as the viral loads of HIV and HCV, and the level of reduction in the number of CD3⁺CD4⁺ lymphocytes, which are predominantly impaired by HIV.

A number of the noted phenomena require interpretation, which is not entirely easy to do on the basis of the contradictory data available in the scientific literature. This is especially true when a child is infected with HIV or HCV against the background of an increase in the viral load of both pathogens simultaneously.

At present, there is information available that HIV significantly increases the vertical transmission rate of HCV. Evidence of the effect of HCV on the progression of HIV infection is less convincing. In particular, there are certain studies, showing that coinfection with HCV reduces the efficacy of combination antiretroviral therapy [25]. Since in this study, pregnant women received antiretroviral therapy, which may be one of the explanations for the established fact of a child infection with HIV when the HCV viral load increased.

The statement about the role of an increase in the viral load of HIV, along with HCV, for the infection of a child with the latter pathogen is largely contradictory. In particular, the risk of vertical transmission of HCV is believed to be associated with high maternal HCV RNA levels, but not with HIV load [26]. However, there is evidence that the incidence of hepatitis C virus transmission from women coinfected with HIV/HCV is reduced when the HIV viral load is well controlled [27]. All this information was obtained by researchers without regard to the polymorphism of the IFNγ gene in infected children, which is yet more proof of the significance of data revealed in this study and their prospects for further investigation.

Data on the prognostic value of the NKG2D gene polymorphism in pregnant women coinfected with HIV/HCV for infection of the child are more certain, especially in terms of the identified correspondence between this polymorphism and the level of reduction in the amount of CD3⁺CD4⁺ lymphocytes in the blood. There is scientific information about the role of a decrease in the number of these cells (below 350) in the vertical transmission of the pathogen [28][29]. It is debatable why, if a pregnant woman with HIV/HCV co-infection has a polymorphism of this gene, the child can be infected with HIV/HCV or only HIV, but not HCV. Meanwhile, when analyzing immunological data without their connection with the polymorphism of the NKG2D gene, it was noted that infection of a child with HCV was possible under the condition of a decrease in cytotoxic T-lymphocytes (CD3⁺CD8⁺) in a pregnant woman. The authors did not find available information that explains this phenomenon. One can assume that control of HCV is carried out both by the antiviral adaptive immune response with CTLs and by the innate immunity with natural killer cells. A decrease in the amount or functional activity of one of these cellular elements disrupts their ratio, which entails a diminution in control over HCV reproduction.

Conclusions