Chapter 7 Biology of HIV-1 transmission
Introduction
HIV-1 has spread rapidly around the globe, and in some parts of sub-Saharan Africa infects up to one-third of adults aged 15–49. Indeed, the effects of HIV-1 have been most devastating in the developing world: of the estimated 2.6 million new infections occurring each year, nearly 70% take place there (http://www.unaids.org). While there remains continued spread of HIV-1 globally, recent estimates suggest that the number of new infections are decreasing; there were ~ 20% fewer infections in 2009 than in 1999 (http://www.unaids.org).
HIV-1 can be transmitted sexually, from mothers to their infants, and via contaminated blood. Globally, heterosexual transmission accounts for the vast majority of new cases of HIV-1 infection, and the epidemic has had similar impact on men and women, with cases in women on the rise [1]. Since sexually infected women who become pregnant can in turn transmit the virus to their infants (called vertical transmission), preventing sexual transmission is viewed as key to slowing the HIV-1 pandemic.
Despite the remarkable spread of HIV, the risk of transmission per exposure is low; estimates are on the order of 0.1% per contact for heterosexual transmission. The per-contact risk is higher (~ 1%) for male-to-male sexual transmission, and for blood exposures via contaminated needles (called parenteral transmission) [2]. These numbers may underestimate the risk for persons who have other endogenous or exogenous risk factors that increase their susceptibility, as host factors of the source partner as well as those of the exposed individual are known to alter transmission risk. Certain features of the virus may also influence its fitness for transmission. Thus, the per-contact risk should be considered an average estimate that may be much higher (or lower) in certain circumstances, as discussed below [3].
HIV-1 can be transmitted from an infected mother to her infant in utero, during delivery, or through breastfeeding [4, 5]. In the absence of any interventions to reduce transmission, approximately one-third of infants born to HIV-1-infected mothers will become infected. In breastfeeding populations, the risk of HIV-1 infection is almost double the risk in non-breastfeeding populations. While it may seem reasonable to therefore recommend against breastfeeding for all HIV-1-infected mothers, this must be balanced with the potential for increased mortality due to other infectious diseases, which can often occur in regions where access to clean water is limited. Fortunately, mother-to-child transmission of HIV-1 can be lowered considerably by antiviral treatment. In developed countries where state-of-the-art treatment is available, and HIV-1 infected mothers do not breastfeed, transmission is as low as 1–2%. There is also increasing access to antivirals to prevent mother-to-child transmission in developing countries, leading to decreased transmission rates.
Factors in the Infecting Partner that Determine the Likelihood of Transmission
Higher virus levels in the infecting host (also called index case or source partner) are correlated with infection [2]. This is perhaps unsurprising, as one might expect that exposure to a higher dose of virus would increase the likelihood of transmission. In most studies, viral levels have been defined by measuring systemic HIV-1 RNA in plasma, even though plasma may not be the major bodily fluid to which the person is exposed to during sexual contact. These findings could therefore reflect the fact that the levels of virus in blood plasma are correlated with the viral levels in other body fluids, such as genital secretions. Indeed, one recent study provided direct evidence that genital virus levels predict HIV-1 transmission, and this was true even after accounting for plasma virus levels [6]. The presence of sexually transmitted diseases (STDs) has been shown to increase risk of transmission. Many STDs increase genital HIV-1 levels, which could in turn increase risk by increasing HIV-1 exposure.
Virus levels are highest during acute (primary) HIV-1 infection, before the virus is contained by the host, and this is thought to be the time when a person is most infectious [2]. Viral levels drop after primary infection resolves, and then slowly and steadily increase over time. Thus, the advanced stage of HIV-1 infection, when CD4 counts are low, is also a time when a person is potentially highly infectious [3].
There are several risk factors common to both vertical transmission and sexual transmission, including the levels of plasma virus in the index case [2, 4]. In the case of vertical transmission, it has been shown that the levels of maternal breast-milk virus and genital virus correlate with the risk of infant infection. Poor breast health in a breastfeeding mother, particularly mastitis, increases infant risk of infection.
Factors in Viral Selection
HIV-1 is highly genetically variable and it continually evolves and adapts in the infected host [7]. HIV-1 seems to undergo a selective bottleneck during transmission because very few viruses are apparently transmitted from one host to another. It is possible that this bottleneck is at least partially a result of stochastic events that reflect the low frequency at which HIV-1 is transmitted. But it may also indicate that there is selection for particular variants with certain properties. The major lines of evidence to support selective transmission include the observations that (1) the early virus population is often genetically less diverse than the source-virus population; and (2) viruses present early in infection tend to infect cells using one particular co-receptor (CCR5).
Many studies of the past two decades have shown that the virus population early in infection, which is presumably very similar to the virus that was transmitted, is genetically more homogeneous than the virus population that is present during chronic infection [8]. The viral sequences present during the early stages of infection are often remarkably homogeneous, which suggest that a single virus was transmitted. This transmitted viral sequences, which can only be inferred from the sequence detected weeks later when HIV-1 reaches high enough levels to detect, has recently been dubbed a founder virus. However, studies in women first suggested that multiple viral sequences are sometimes transmitted, and this has now been observed in other populations and linked to the presence of biological cofactors such as other STDs [8]. Even in cases where the virus is genetically heterogeneous early in infection, it is generally less diverse than what would be expected during chronic infection, suggesting that only a subset of variants are successfully transmitted. More recent detailed studies of viruses in both the index case and their newly infected partner (transmission pairs) near the time of HIV-1 acquisition provide direct evidence for this transmission bottleneck [9]. However, in some situations it is possible that the limited diversity of transmitted strains indicates that the source partners harbored a virus population of limited diversity, perhaps because they transmitted during their primary infection, which is a time of high infectivity.
No matter what the complexity of the viral genotype, the viruses present within the first few months after infection almost invariably require the CCR5 co-receptor for entry (these are called R5 viruses) [10]; CCR5 is one of two major HIV-1 co-receptors (the other being CXCR4), and the co-receptor, along with the CD4 receptor, is critical for HIV-1 entry into cells. The observation that most recently transmitted viruses are R5 viruses suggests that CCR5 variants are favored for transmission. This apparent selection for R5 viruses occurs during all routes of transmission, including sexual, vertical, and parenteral routes. In support of this model, it has been shown that individuals who do not express cell surface CCR5 due to a specific genetic polymorphism are less susceptible to HIV infection (see also below).
Despite the fact that transmitted viruses share a common co-receptor requirement, the viruses transmitted from different individuals are quite genetically distinct. This diversity has made vaccine development a daunting prospect. Thus, there has been considerable interest in defining common features among transmitted viral strains. Signature sequence characteristics have been noted among viruses present early in infection, at least in some populations [8], and they may provide insights into which biological properties of viruses increase their fitness for transmission. Many studies have attempted to identify biological characteristics that confer the selective advantage for transmitted HIV-1 variants, but none have identified a clear biological phenotype common to all transmitted strains that may explain their selection.
Endogenous Host Factors
Host genetics
Multiple host genetic polymorphisms have been linked to HIV-1 susceptibility [11–13]. The mutations that have been identified derive largely from targeted studies focused on genes that code for host factors known to be critical for HIV-1 replication. For example, many studies have focused on allelic variation within co-receptor genes, or genes coding for ligands that bind the HIV-1 co-receptors (e.g. CCL5/RANTES for CCR5 and CXCL12/SDF1 for CXCR4) and thus potentially compete for HIV-1 entry. Therefore, alterations in the expression or function of the proteins encoded by these genes could impact HIV-1 replication at the cellular level. Overall, studies of host genetic factors have provided a somewhat complex view of the effects of host genetics on HIV-1 transmission, as consistent results have not always been found across studies. This may partially be due to the complexity of the interactions between the different alleles, as well as differences in allele frequency and other factors in the populations examined. Moreover, with many of the mutations it is unclear whether they actually affect protein levels or function, or whether they were detected because they are genetically linked to other mutations in nearby genes, which play a more direct role in transmission.
Genetic variations in loci encoding molecules that play a role in acquired immunity have also been associated with HIV-1 transmission risk. Several studies suggest that human leukocyte antigen (HLA) allele concordance between the index case and the uninfected partner may increase the risk of transmission. HLA proteins are acquired on the virus as it buds from the host cell, and it has been postulated that discordance of HLA may mark the infectious virus as more immunologically foreign, and thus decrease transmission. Genes involved in innate immunity—the killer cell-immunoglobulin-like receptor (KIR) genes that bind to HLA proteins and modulate natural killer (NK) cell activity—have also been implicated in HIV-1 susceptibility. NK cells play a central role in the initial antiviral response. Both allelic differences and expression differences in KIR have been implicated in HIV susceptibility, although these findings are somewhat preliminary [11].
To date, host genes identified as risk factors for HIV-1 acquisition have primarily been uncovered because these genes/proteins were known to play a role in HIV-1 biology. It is possible that a more global genomic approach, which would sample a larger number of genes independently of whether they have an established link to HIV-1 replication or immunity, could yield a much longer list of polymorphisms involved in HIV-1 susceptibility. Indeed, there is considerable interest in these genome-wide approaches to studying HIV resistance in individuals who are HIV-1 negative despite being highly exposed to HIV-1 [12]. It is thought that examining a large group of highly exposed HIV-1 negative individuals may reveal other host genes important in HIV-1 susceptibility. However, identifying such cohorts is complicated because HIV-1 transmission is a somewhat rare event that can be influenced by many other factors, making it hard to truly define those who are more resistant to infection versus those who are simply beating the odds.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
