Modeling Mucosal Cell-Associated HIV Type 1 Transmission in Vitro

Abstract

Human immunodeficiency virus type 1 (HIV-1) can efficiently spread by direct cell-to-cell contact, a mechanism termed cell-associated HIV transmission. By some estimates, cell-associated HIV transmission is 10–1000-fold more effective than cell-free HIV infection. Mucosal cell-associated HIV transmission may occur when HIV-bearing cells in mucosal secretions from an HIV-infected donor transfer virus directly to recipient target cells in or below the mucosal epithelium, or through HIV transcytosis across the mucosal epithelium of a noninfected host. This mechanism may play an important role in the sexual and vertical transmission of HIV-1, yet most in vitro tests of vaccine and microbicide efficacy assess cell-free virus transmission. This article reviews in vitro assays that have been used to model mucosal cell-associated transmission, including microscopy, immune cell cocultures, use of HIV-infected cells in epithelial cell transcytosis assays, and cell-associated infection of mucosal tissue explants. Assays that authentically simulate mucosal cell-associated HIV transmission could provide valuable insight into mechanisms and molecules that can potentially be targeted for HIV prevention, as well as critical models for testing novel HIV prevention strategies for efficacy against cell-associated HIV transmission.

Many viruses, including human immunodeficiency virus type 1 (HIV-1), can spread (1) as cell-free virions that bud from infected cells and encounter target cells via diffusion through the extracellular milieu or (2) by infected cells through direct cell-to-cell contact [1, 2]. Cell-to-cell HIV transmission, also known as cell-associated HIV transmission, has been shown to be 10-fold to >1000-fold more efficient than cell-free transmission in vitro [3, 4]. This striking difference in efficiency between these 2 modes of HIV transmission has been ascribed to a number of factors: (1) proximity of the cell-associated virus to its target, (2) receptor clustering at points of cell-to-cell contact, (3) increased multiplicity of infection due to the targeted budding of virus at synapses formed between cells, and (4) the relative resistance of cell-associated transmission to a number of factors that inhibit the infectiousness of cell-free virions, such as neutralizing antibodies, and host restriction factors, such as tetherin and TRIM 5-α [5]. Cell-to-cell HIV transmission among cells in lymphoid organs and possibly other sanctuary sites is thought to underlie HIV persistence in vivo [5].

There is mounting evidence that cell-associated HIV transmission could play a role in sexual and vertical transmission of HIV [6]. Since the mechanisms underlying cell-associated HIV transmission differ from those of cell-free transmission, cell-associated transmission could require different prevention strategies. For example, some broadly neutralizing antibodies and antiretroviral therapies are less effective at blocking cell-associated HIV transmission than cell-free transmission [7, 8]. On the other hand, strategies that block cell attachment and the formation of viral synapses may be particularly effective at blocking cell-associated transmission [9]. Through this review article I hope to bring attention to this understudied field by presenting the strengths and weaknesses of in vitro cell-associated HIV transmission assays that have been used to date to assess mucosal cell-associated transmission and test HIV prevention strategies.

IN VITRO ASSAYS USED TO MODEL MUCOSAL CELL-ASSOCIATED HIV TRANSMISSION

Microscopy

Much of what we know about cell-associated HIV transmission has been learned from microscopy image analysis. Early evidence for virus transmission between cells was provided by electron microscopy images showing directional budding of virus at cell-cell contacts [10, 11]. Subsequently, use of sophisticated fluorescence microscopy imaging techniques enabled visualization of the recruitment of HIV and its receptors to intercellular junctions, the formation of virologic synapses [12], and the direct transfer of HIV via synapses from infected to uninfected cells [13]. Microscopy has also been used to identify components of virologic synapses that could be targeted to block cell-associated HIV transmission. HIV gp160 env, CD4, and chemokine coreceptors play crucial roles in synapse formation; ICAM-1/LFA-1 adhesion molecules stabilize cell-cell contacts; and actin and cytoskeleton proteins are remodeled during cell-associated HIV transmission [14]. The targeted use of antibodies and other antagonists to many of these structures inhibits cell-associated HIV transmission [15, 16]. Microscopy has also been used for testing the efficacy of HIV broadly neutralizing antibodies and microbicides in blocking cell-to-cell HIV transfer and provided the first evidence that cell-associated transmission may be resistant to patient serum and broadly neutralizing monoclonal antibodies [4, 17]. Although this approach is labor intensive and has been largely supplanted by other methods (described below), a related quantitative flow cytometry method using infectious green fluorescent protein–labeled HIV for assessment of cell-associated HIV transmission was recently published [18] and shows promise as a platform for screening vaccine-induced antibodies and microbicides.

Immune Cell Coculture Assays

Cell-to-cell HIV transmission is commonly demonstrated in cell suspensions containing infected T cells [19, 20], macrophages [21], or HIV-bearing dendritic cells [22] as virus donors and uninfected T cells or highly permissive reporter cell lines as target cells [1]. In these cocultures, cells form transient adhesive structures called virologic synapses through which virus is directionally transmitted to adjacent cells [4]. The efficiency of cell-associated transmission in coculture assays can vary depending on the types of infected and target cells used [7]. Cell-associated HIV transmission is difficult to quantify in coculture systems because of simultaneous infection of target cells with cell-free HIV. Various approaches have been used to reduce this possibility, including comparison of virus transmission in static versus gently shaken cultures, to differentiate cell-free from cell-associated transmission [21]; treatment of infected cells with mitomycin C or antiviral drugs such as saquinavir (protease inhibitor), to block new virus assembly [7, 23]; use of activated primary CD4⁺ T cells as targets instead of engineered reporter cell lines (eg, TZM-bl) that are highly susceptible to cell-free infection [7]; and use of specialized reporter vectors [3].

An early hypothesis predicted that virus transmitted via the cell-associated pathway would be shielded from the effects of neutralizing antibodies [2, 24]. However, studies to determine the relative efficacy of HIV broadly neutralizing antibodies in cell-associated versus cell-free HIV transmission assays have produced mixed results (Table 1). Several of the major studies have concluded that antibodies directed against the CD4 binding site on HIV effectively neutralize cell-associated HIV transmission, whereas others have shown diminished neutralizing activity with CD4bs antibodies but effective neutralization with antibodies directed against the gp41 MPER region. Antibodies and other antagonists directed against cell HIV receptors (CD4 and CCR5) appear to also be effective [8, 25–28]. Discordance in these data may be due to differences in (1) timing of antibody treatment (eg, CD4bs antibodies can affect synapse formation but may show reduced efficiency once synapses are formed), (2) source and types of virus and antibodies, (3) effector:target cell systems, and (4) possibly other factors, such as background levels of cell-free HIV transmission [7]. It underscores the need for much more systematic research in this field to fully understand the effects of HIV-directed antibodies and other factors on cell-cell HIV transmission.

Table 1.

Neutralization Activity of Broadly Neutralizing Human Immunodeficiency Virus (HIV) Antibodies in Cell-Associated HIV Transmission Assays

Reference	System	CD4bs (VRCO1, b12)	gp41 MPER (4 E10, 2F5)
Frankel et al [25]	DC-T	+	+
Abela et al [26]	PBMC-TZMbl	−	+
Malbec et al [27]	T-T	+	−
Sagar et al [22]	DC-T	-	+
Duncan et al [28]	MDM-T	+	−
Martin et al [29]	Jurkat/A301.R5	+	+
Zhong et al [30]	Jurkat/SupT	ND	+

Abbreviations: +, strong neutralization in both cell-free and cell-associated HIV transmission assays; −, poor neutralization activity in cell-associated HIV transmission assays.

Coculture assays are currently being used to screen microbicide compounds for efficacy against cell-associated HIV transmission. The Southern Research Institute in the United States has developed a comprehensive HIV screening algorithm for the discovery and preclinical testing of topical microbicides that includes cell-associated HIV transmission assays [23]. For this purpose, they use HIV-infected H9 or MOLT cells, as well as GHOST target cells. Their algorithm calls for a pH transition to simulate vaginal conditions, as well as the addition of 12.5% seminal plasma. R. J. Shattock's laboratory in the United Kingdom uses HIV-infected PM1 cells and TZM-bl (ie, HeLa cervical carcinoma derived) reporter cells [31]. The Virology Unit at the Institute of Tropical Medicine in Antwerp, Belgium, routinely screens antiretrovirals and microbicides in a cell-associated HIV transmission assay consisting of HIV-infected leukocytes and T-cell lines (R5 and X4 MaRBLE) containing a firefly luciferase reporter gene [32].

It could be argued that coculture assays such as the ones described above simulate HIV transmission that potentially occurs between HIV-bearing immune cells in mucosal secretions and uninfected HIV target cells residing in the mucosal epithelium. HIV-infected T cells and macrophages have been isolated from semen obtained from HIV-positive men [33], and infected leukocytes have also been detected in cervicovaginal secretions and breast milk from HIV-infected women [34, 35]. Uninfected counterparts of these cells along with dendritic cells could serve as HIV target cells in genital and gastrointestinal epithelia, especially under inflammatory conditions [36]. HIV-infected leukocytes from semen, cervicovaginal secretions, or breast milk could encounter uninfected target cells residing on the surface of the epithelium or within the stratum corneum, the superficial layer of the vaginal and penile epithelium that is devoid of tight junctions [37, 38], or they could transmigrate into the epithelium to encounter target cells within or below the mucosal layer [6]. However, it should be kept in mind that HIV-infected cells and target cells in mucosal tissues have distinct characteristics that distinguish them from PBMCs or cell lines that are normally used in coculture assays. T lymphocytes at mucosal sites are predominately differentiated memory T cells, and tissue macrophages usually have an M2 phenotype [36, 39]. In addition, the Langerhans cells and numerous dendritic cells in mucosal tissues could play a major role in cell-associated HIV transmission [22, 40]. Finally, cell-associated HIV transmission in vivo could be affected by the degree of allogenicity between donor and host.

Cell-Associated HIV Epithelial Transcytosis Assays

Various polarized epithelial cell monolayers have been used to study HIV mucosal transmission. It has been known for some time that contact between HIV-infected cells and epithelial cells results in a massive and rapid budding of HIV virions toward the epithelium [10, 11]. This is followed by the internalization of virions into endosome-like structures and their passage across the epithelial barrier via a characteristic epithelial transcellular vesicular pathway, a process termed transcytosis [41]. After their passage, the virions are capable of infecting target cells residing below the epithelium [41]. Efficient HIV transcytosis has been demonstrated to occur across polarized monolayers of immortalized vaginal cells [42], as well as transformed intestinal [43], endometrial [44], and cervical epithelial cells [45, 46]. In most transcytosis models, HIV-infected leukocytes are more efficient than cell-free virus in producing infection of subepithelial target cells. Synapse formation between HIV-infected lymphocytes and epithelial cells has been demonstrated by electron microscopy [47, 48]. Whereas HIV transcytosis has been readily demonstrated with transformed epithelial cell models, the physiological relevance of this cell-associated HIV transmission mechanism is unclear; HIV transmission was inefficient when infected cells were added to the apical surface of polarized primary cultures of human ectocervical and endocervical epithelia [49, 50], ectocervical and endocervical epithelial sheets [49], and a reconstructed vaginal epithelial model [51].

Epithelial transcytosis assays have been used to test the efficacy of mucosal and monoclonal antibodies to prevent cell-associated HIV transmission. Mucosal HIV antibodies, especially immunoglobulin A, efficiently block HIV transcytosis from infected cells [52], whereas several neutralizing monoclonal antibodies against HIV have failed to effectively block cell-associated HIV transcytosis [53]. Cell-associated HIV transcytosis assays have also been used to test microbicide candidates [31, 54–56].

Cell-Associated HIV Infection of Mucosal Tissue Explants

Mucosal tissue explants, comprising an intact epithelium and resident mucosal HIV target cells, have also been used for studies of cell-associated HIV transmission. In one study [57], cell-free HIV or infected cells were placed on the luminal side of polarized ectocervical explant tissue, and viral transmission was detected by measuring the HIV load in the lower chamber at different time points. The addition of X4 cell-associated HIV and both X4 and R5 cell-free HIV resulted in transmission of the virus across the mucosa. In another study [58], labeled viable cells from semen were shown to bind to and penetrate the ectocervical epithelium but failed to bind to endocervical explants because of their entrapment in mucus secreted by these cells. Two studies have demonstrated efficient cell-associated HIV transmission across inner but not outer foreskin tissue in vitro [48, 59]; cell-free HIV was ineffective in this ex vivo foreskin model. Tugizov et al [60] recently demonstrated that HIV-infected macrophages but not T cells were able to transverse fetal oral and intestinal epithelia but not adult epithelia. In contrast, another group reported that HIV-infected CD4⁺ T cells transmigrate across human colonic explant tissues to infect target cells below [61].

Mucosal tissue explants have the advantage of containing an intact epithelial layer, authentic mucosal HIV target cells, and other factors in the mucosal environment. However, they have several disadvantages, including (1) marked interdonor variation in hormone status, inflammation, sexually transmitted disease history, and HIV target cell populations, making assay reproducibility difficult; and (2) rapid deterioration of tissue structure in vitro [62]. Nonetheless, mucosal tissue explants have been widely used to screen microbicide candidates for efficacy against cell-free HIV infection but rarely against cell-associated infection [31, 54, 63].

ADDITIONAL CONSIDERATIONS

For enhanced authenticity, cell-associated HIV transmission models should incorporate mucosal tissue-derived cells as infection sources and/or targets and should include other elements of the mucosal environment, such as semen, cervicovaginal secretions, a pH range, genital pathogens, and microflora [64]. This poses several challenges: few viable immune cells are normally recovered from genital secretions by noninvasive techniques, and these cells vary in phenotype, activation state, and function, depending on local conditions such as inflammation, infections/microflora, and hormone status [65, 66]. Furthermore, the molecular composition of genital secretions and the nature of tight junctions in genital epithelium may also be affected by these and other factors [67, 68]. Little research has been done to determine whether such factors affect cell-associated HIV transmission or the efficacy of antibodies and other compounds being tested in cell-associated transmission assays.

Another relevant topic is immune defense against infected cells that mediate cell-associated HIV transmission. Three principal effector mechanisms are known to target HIV-infected cells: cell-mediated immunity, carried out by CD8⁺ T cells; natural killer (NK) cell immunity, mediated by NK cells; and antibody dependent cellular cytotoxicity (ADCC), mediated by Fc-bearing immune cells and antibodies [69, 70]. Vaccine and microbicide candidates should also be tested in these assays because it should be determined that prevention strategies do not interfere with these important defense mechanisms. The ideal prevention strategy would potentiate these mechanisms. A few laboratories have begun to systematically test monoclonal antibodies and vaccine antisera to determine which antibody specificities, subclasses, and isotypes are most effective in mediating ADCC killing of HIV infected cells [71]. This work should continue.

CONCLUSIONS

Cell-associated HIV transmission is increasingly recognized as a potentially important mechanism underlying HIV transmission across mucosal surfaces. However, the focus for microbicide and vaccine screening remains cell-free HIV infection assays. Relatively little effort has been spent on developing authentic mucosal cell-associated HIV transmission assays, and as such much more work is needed to expand, refine, and validate these assays.