Tetherin/BST-2: Restriction Factor or Immunomodulator?

Abstract

Background: Cell-mediated immune (CMI) responses are critical for the control of HIV-1 infection and their importance was highlighted by the existence of viral proteins, particularly Vpu and Nef, that antagonize these responses. Pandemic HIV-1 Vpu counteracts Tetherin/BST-2, a host factor that could prevent the release of HIV-1 virions by tethering virions on the cell surface, but a link between Tetherin and HIV-1 CMI responses has not yet been demonstrated in vivo. In vitro, the virological and immunological impact of Tetherin-mediated accumulation of virions ranged from enhanced or diminished cell-to-cell spread to enhanced recognition by virus-specific antibodies for natural killer cellmediated lysis. However, Tetherin-restricted virions could be internalized through an endocytosis motif in the Tetherin cytoplasmic tail.

Methods: Given the uncertainties on which in vitro results manifest in vivo and the dearth of knowledge on how Tetherin influences retroviral immunity, in vivo retrovirus infections in mice encoding wild-type, null and endocytosis-defective Tetherin were performed. Here, we review and highlight the results from these in vivo studies.

Results: Current data suggests that endocytosis-defective Tetherin functions as a potent innate restriction factor. By contrast, endocytosis-competent Tetherin, the form found in most mammals including humans and the form counteracted by HIV-1 Vpu, was linked to stronger CMI responses in mice.

Conclusion: We propose that the main role of endocytosis-competent Tetherin is not to directly restrict retroviral replication, but to promote a more effective CMI response against retroviruses.

Figure 1

Structural features of tetherin. Tetherin in both humans and mice exist as a dimer consisting of an N-terminal transmembrane (TM) domain and C-terminal GPI anchor. The extracellular coiled-coil domain is glycosylated. The N-terminal cytoplasmic tail encodes a dual tyrosine motif (YxY) that facilitates endocytosis. In contrast to mouse tetherin, human tetherin has a 5 amino acid deletion (Δ) in the cytoplasmic tail that conferred NF-κB activation properties.

Figure 2

Fate of cell surface-tethered virions. The accumulation of virions on the infected cell surface due to tetherin activity could have several consequences. (A) Aggregated virions on the cell surface may either promote or decrease cell-to-cell spread. (B) The accumulation of envelope glycoproteins can enhance recognition of infected cells by env-specific antibodies, which could induce NK cell mediated antibody-dependent cellular cytotoxicity through Fc-antibody interactions.

Figure 3

Cell surface expression of long versus short-form tetherins. Long-form tetherins encode a cytoplasmic tail with an endocytosis motif, which facilitates its recycling and/or degradation through the endocytic pathway. In contrast, short-form tetherins lack this endocytosis motif, leading to higher expression levels on the cell surface. Most mammalian tetherins are of the long-form type and therefore endocytosis-competent. Endocytosis-defective, short-form tetherins were found in felids and ovines, in humans due to a leaky Kozak sequence and in the NZW/LacJ (NZW) strain of mice. Long-form tetherin in humans also have NF-κB signaling properties.

Figure 4

Test cross to investigate the role of the NZW Tetherin SNP in acute FV infection. Fv1, Fv2, H-2 and Apobec3/Rfv3 govern FV resistance and susceptibility in mice [69, 91]. Fv1 confers a Gag-dependent post-entry block whereas Fv2 confers susceptibility to splenomegaly. These 4 genes, as well as Tetherin, were encoded in different chromosomes, thus allowing for straightforward Mendelian genetic crosses. Fv1 was controlled using a dual-tropic FV strain, whereas the backcrossing approach yielded progeny mice that were all Fv2-susceptible. Three cohorts of B₁ mice were assembled and FV levels at 7 days post-infection (dpi) were evaluated. (1) In Apobec3/Rfv3 resistant mice, the NZW Tetherin SNP had no impact on acute FV replication, revealing a ‘dominance hierarchy’ between Apobec3 and tetherin antiretroviral activity in vivo. Apobec3/Rfv3-susceptible mice were further subdivided into their H-2 genotypes. (2) In H-2^z/z mice, endocytosis-defective Tetherin was more protective, correlating with its more potent antiretroviral activity in vitro. (3) However, in H-2^b/z mice, endocytosis-competent Tetherin was more protective, and this protection correlated with stronger NK cell responses.

Figure 5

Working model on how Tetherin-mediated virion endocytosis promotes NK cell and T cell responses. FV can infect antigen-presenting cells (APC) including macrophages and dendritic cells. (1) In the presence of tetherin, virions are prevented from being released and are brought back into the infected cell through endocytosis. (2) These virions are retained in endosomes, potentially allowing more time for endosomal proteases to degrade the virion. (3) Released viral RNA could activate TLR3, resulting in APC activation such as the upregulation of the costimulatory molecule CD80 and the release of cytokines. IL15, in particular, could promote NK cell development and activity. (4) Virion proteolysis results in an increase in viral peptides for loading to MHC-II, which in conjuction with increased costimulatory molecule expression, would (5) promote CD4+ T cell responses. (6) Degraded virion peptides may be cross-presented to MHC-I to stimulate CD8+ T cell responses. Moreover, (7) released cytokines such as IL-2 from NK cells and CD4+ T cells may help stimulate CD8+ T cell responses.