Interactions between human immunodeficiency virus and herpes viruses within the oral mucosa


Corresponding author and reprint requests: F. X. Mbopi-Keou, Faculty of Medicine and Biomedical Sciences, Department of Microbiology and Infectious Diseases, University of Yaounde I, PO Box 1206, Yaounde, Cameroon


There is evidence from clinical case reports and epidemiological studies that human immunodeficiency virus (HIV) can be transmitted through oral sex. Herpes viruses that appear in the oral mucosa might influence the oral replication of HIV. A review of data suggesting that interactions occur between HIV and herpes viruses indicates that such interactions might operate in the oral mucosa. Defining the mechanisms by which herpes viruses interact with HIV in the oral mucosa should permit intervention measures to be targeted more precisely.

There is growing evidence that human immunodeficiency virus (HIV) can be transmitted through oral sex. HIV-1 RNA may be present in up to 90% of whole saliva samples and in 50% of gingival crevicular fluid samples [1]. Although the virus burden of saliva is usually less than that of plasma and semen, it has been reported that HIV RNA levels in saliva in some patients may be up to five-fold higher than in plasma [2]. Although many other viruses can also be found in the oral mucosa, herpes viruses have been identified in a wide range of HIV-related oral lesions, e.g., herpes simplex virus (HSV) in periodontal disease [3], cytomegalovirus (CMV) in aphthous-like ulceration [4], and human herpes virus 8 (HHV-8) in Kaposi's sarcoma [5]. It is possible that asymptomatic carriage of herpes viruses may significantly influence the oral carriage of HIV (and vice versa) [6]. Importantly, such viruses have also been identified in HIV-infected individuals free of oral disease, e.g., Epstein–Barr virus (EBV) [7] and HHV-8 [8]. These viruses possess a variety of mechanisms that might heighten the local carriage of HIV.

There is extensive evidence that HSV-1 reactivation in the female genital tract may increase during HIV replication in co-infected individuals [9]. The HIV-1 long terminal repeat (LTR) can be transactivated by HSV-1 immediate–early (IE) gene products. The ICP0 and ICP4 proteins are thought to be important mediators of this process, which is known to involve induction of the cellular activators NFκB and Sp1 [10]. It is conceivable that recurrent repeated episodes of HSV-1 reactivation might directly and regularly increase the HIV load in oral fluid. HIV-1 virions have been observed to infect keratinocytes of HSV-1, which, because they lack the CD4 molecule, are normally incapable of being infected by HIV-1 [11].

CMV can drive HIV production via heterologous transactivation by CMV IE gene products of the HIV-1 LTR, as might occur in cells co-infected by both HIV and CMV, or cells sharing the same internal milieu [12]. In particular, CMV encodes IE2-86, a DNA-binding protein that acts as a promiscuous transactivator of many virus and cellular promoters [13], and also IE1-72, which stimulates transcription from a CMV promoter, as well as heterologous viral and cellular promoters (thereby upregulating ΝFκΒ) that play essential roles in virus life cycles and orchestrate virus-dependent cytokine expression [14]. To effect transactivation, CMV IE products may or may not synergise with the HIV Tat protein, an important transcriptional transactivator of the LTR [15].

The oral mucosa, particularly the tongue, is the site at which EBV undergoes lytic replication. EBV vegetative replication occurs in the tongue during the early stages of HIV disease, well before oral hairy leukoplakia becomes clinically apparent [16]. Infection of primary CD4+ and CD8+ T-lymphocytes by EBV has been shown to enhance HIV-1 expression [17]. In addition, it has been shown that, following HIV-1 infection of CD4+/CR2+ cell lines transformed by EBV, the cells showed evidence of HIV-1 upregulation [18]; additional evidence was also provided that the EBV protein EBNA-2 is required by Tat to transactivate the HIV-1 LTR [18]. Lytic EBV infection of B-cells can be associated with superantigen-like expression in these cells, leading to the expansion of Vβ12 T-cell receptor-bearing CD4+ T-cells [19]. Such an expansion potentially provides a pool of cells susceptible to HIV infection.

HHV-6 can upregulate the production of CD4 in T-lymphocytes [20]. Increased expression of CD4 can also be induced by HHV-6 in γ/δ T-cells, a subset of T-cells involved in the protective immune response against specific microorganisms, rendering them susceptible to HIV-1 infection [21]. Furthermore, HHV-6 and HIV-1 can co-infect human CD4+ T-cells, thereby leading to accelerated cell death, and factors in HHV-6-infected cells stimulate HIV-1 LTR-directed gene expression, mediated by HHV-6-induced nuclear protein(s) that specifically bind to the ΝFκΒ motifs of the HIV-1 enhancer region [22]. However, HHV-6 also appears to be able to downregulate the HIV-1 co-receptor CXCR4 [23]. How these apparently disparate HHV-6-associated influences relate to HIV activation in vivo has not yet been comprehensively studied. Nevertheless, there is evidence that the overall influence of HHV-6 is to increase the systemic HIV-1 viral load [24]. In contrast, HHV-7 appears to exert effects on HIV that oppose the effects of HHV-6. HHV-7 downregulates CD4 expression in T-lymphocytes [25] and interferes with HIV-1 interaction with CD4 in terminally differentiated monocytes [26].

The mouth is a major site of HHV-8 carriage [5]. Among HHV-8-seropositive/HIV-1-seronegative male homosexuals in the USA without Karposi's sarcoma [27], HHV-8 DNA was detected in 30% of oropharyngeal samples, compared with 1% of anal and genital samples, with the HHV-8 titre from the oral cavity being up to 2.5-fold higher than those from other sites; in-situ hybridisation studies indicated that HHV-8 DNA and mRNA can be found in oral epithelial cells [27]. HHV-8 encodes chemokine-like proteins (vMIP-I and vMIP-II) that may interact with HIV-1. vMIP-II has been shown to block HIV-1 infection in a CD4+ cell line expressing CCR3 and, to a lesser extent, in a cell line expressing CCR5, and both vMIP-I and vMIP-II partially inhibited HIV infection of peripheral blood mononuclear cells [28]. Such results would imply that HHV-8 exerts an anti-HIV-1 effect. However, other studies suggest otherwise. Thus, the HHV-8 latency-associated nuclear antigen (LANA), some other HHV-8-encoded proteins from open reading frames 50 and 57, and the KIE2 immediate–early gene product, are able to interact with the HIV transactivator protein to transactivate the HIV-1 LTR [29].

In conclusion, herpes viruses that appear in the oral cavity appear to be able to influence oral HIV replication. The clinical, and certainly the virological, influences of herpes virus carriage on the infectivity of HIV in the mouth have not yet been fully addressed. There is no published research on the impact of asymptomatic herpes virus carriage on the HIV load in the different compartments of the mouth, despite substantial in-vitro evidence to suggest that herpes viruses have the potential to enhance HIV carriage in the mouth, and perhaps vice versa. Defining the mechanisms by which herpes viruses interact with HIV in the oral mucosa should permit intervention measures to be targeted more precisely.