Scientists have identified a protein that effectively strips dendritic cells (DCs) and monocyte-derived macrophages (MDMs) of the intracellular deoxynucleoside triphosphates (dNTPs) needed by invading HIV-1 to synthesize its own DNA and replicate. An international team led by researchers at the Institut Cochin in Paris, New York University School of Medicine, and the University of Rochester Medical Center sites say SAMHD1 expressed by HIV-1’s target myeloid cells hydrolyzes intracellular dNTP, essentially starving the virus of the building blocks required to support viral reverse transcription, and thus blocking lentiviral infection.
In contrast, the researchers also found that viruses of the HIV-2-simian immunodeficiency virus lineage ((SIVsm)-HIV-2) express a virion-packaged accessory protein, Vpx, which degrades SAMHD1. This acts to prevent the protein from depleting dNTPs and restores the permissiveness of myeloid cells to infection, but doesn’t appear to lead to increased virulence. Reporting their results in Nature Immunology, the Institut Cochin’s Florence Margottin-Goguet, Ph.D., and colleagues, suggest that therapeutic strategies that deplete intracellular dNTP pools might represent a promising approach to treating lentivus infection.
“The implications of depleting the nucleotide pool as a host-defense mechanism are potentially far reaching," they write. The team’s published paper is titled “SAMHD1 restricts the replication of human immunodeficiency virus type 1 by depleting the intracellular pool of deoxynucleoside triphosphates.”
SAMHD1 was first identified as an ortholog of a mouse gene induced by interferon-γ in response to viral infection, and has been proposed to operate as a negative regulator of the innate immune response. Prior work has also shown that mutations in conserved regions of the SAMHD1 protein inactivate its lentivirus-restricting activity.
Sequence comparison studies by Dr. Margottin-Goguet demonstrated weak homology between SAMHD1 and EF1143, which suggested the protein may have a role in regulating intracellular dNTP pools in myeloid cells. This notion was confirmed by the finding that intracellular dNTP concentrations in human monocyte cells were increased when SAMHD1 was knocked down using an shRNA, and the observation that recombinant SAMHD1 hydrolyzed all four dNTPs in vitro.
Primary MDMs take up exogenous deoxynucleosides and directly convert them into dNTPs, and the team confirmed that treating healthy donor-derived MDMs either with deoxynucleosides or Vpx-containing virus-like particles led (VLPs) to a 5- to 33-fold increase in the concentration of dNTPs, and enhanced HIV-1 infection. Interestingly, the nucleoside treatment was more effective in terms of enhancing HIV-1 infection, probably because these cells continued to hydrolyze the dNTPs, while cells exposed to Vpx-containing VLPs degraded SAMHD1, preventing dNTP hydrolysis.
Further tests were carried out in an SAMHD1-deficient monocyte cell line (U937) transduced to express either human, mouse, rhesus macaque, or inactive SAMHD1. The cells were treated with the protein kinase C activator Phorbol 12-myristate 13-acetate (PMA), which induces their differentiation into macrophages. These assays demonstrated that in the absence of PMA, intracellular dNTP concentrations were relatively unaffected by the various SAMDH1 constructs. However, after treatment of the cells with PMA, expression of SAMHD1 was accompanied by a considerable decrease in the dNTP pool relative to cells transduced with a control lentiviral vector, or with the inactive SAMHD1.
The deficiency of SAMHD1 in unmodified U937 cells means this cell line is normally highly susceptible to HIV-1 infection, but this susceptibility was markedly reduced on expression of SAMHD1. Treating human- or rhesus-SAMHD1-expressing cells with Vpx-VLPs blocked this resistance to infection, due to degradation of SAMHD1, although Vpx-VLPs appeared to have no effect on mouse SAMHD1.
Unlike MDMs, activated primary CD4+ T lymphocytes are normally fully susceptible to HIV-1. The team also demonstrated that Vpx-containing VLPs had only a marginal effect on dNTP concentrations, which suggests that in these cells SAMHD1 plays only a minor role in controlling dNTP concentration. In fact, overexpressing SAMHD1 in a CD4+ human lymphoma T cell line had no effect on HIV-1 infection relative to control cells, whereas pharmacologically blocking ribonucleotide reductase in the SAMHD1-expressing T cells did lead to increased resistance to HIV-1 infection. “These results suggested that the high rate of dNTP production in the T cells rapidly replenishes the dNTPs that are hydrolyzed by SAMHD1, limiting its antiviral activity in these cells,” the authors write.
“SAMHD1-mediated depletion of the intracellular dNTP pool provides a means by which the cell establishes an antiviral state without the need for an inhibitor that interacts directly with a specific viral component,” they conclude. “Furthermore, it provides a rationale for why SAMHD1 restriction operates only in terminally differentiated and nondividing cells such as MDMs and DCs. Those cells, in contrast to dividing cells, do not need to maintain a high concentration of dNTPs and therefore are not harmed by the catalytic activity of SAMHD1...Thus, this mechanism may serve to restrict a diverse range of retroviruses.”
The team also points out that the use of Vpx to degrade SAMHD1 is an important strategy used by lentiviruses to prevent cellular antiviral defenses. “We don’t know precisely how SAMHD1 and Vpx affect the virulence of HIV-1 and HIV-2, but it's something we’re actively exploring," notes Baek Kim, Ph.D., co-corresponding author at the University of Rochester Medical Center. “The work suggests new ways to target virus replication in macrophages, a critically important cell population that serves as a key reservoir of virus infection and a contributor to HIV-induced disease.”