Discovery at the VICB
Targeting Phospholipase D to Suppress HIV Infection
By: Carol A. Rouzer, VICB Communications
Published: May 29, 2015
A selective inhibitor of phospholipase D1 suppresses replication of the human immunodeficiency virus by blocking deoxyribonucleotide biosynthesis.
Despite substantial progress in the development of antiretroviral therapies, the human immunodeficiency virus (HIV) remains a major worldwide public health problem. The causative agent of acquired immunodeficiency syndrome (AIDS), HIV has infected approximately 1.2 million people in the United States, with 50,000 new cases each year. Once an almost certain death sentence, AIDS has been transformed into a chronic condition with the use of antiretroviral drugs. However, continual mutation of the virus leading to acquired drug resistance necessitates the constant search for new drugs with novel mechanisms of action. Thus, the discovery by Vanderbilt Institute of Chemical Biology members Alex Brown and Craig Lindsley, and their collaborators Harry E. Taylor and Richard T. D’Aquila at the Northwestern University HIV Translational Research Center, that inhibitors of phospholipase D (PLD) suppress HIV-1 replication is a welcome and exciting advance [H. E. Taylor, et al. (2015) PLoS Pathogens, 11, e1004864].
HIV-1 infects CD4+ T lymphocytes by binding first to CD4 proteins on the cell surface (Figure 1). The viral core gains entry into the cell and begins to reverse-transcribe its RNA genome into double-stranded cDNA (dscDNA) using primarily viral-encoded proteins. The dscDNA copies then gain entry into the nucleus, where they incorporate into the host genome or form a variety of stable circular forms. At this point, the virus remains latent until the T cell is activated. In response to activation, levels of transcription factors (NF-κB, SP1, NFAT, Tat) required for viral transcription increase substantially. As a result, new viral genomes and virally encoded mRNA are produced. Translation of the mRNA yields new viral proteins, which assemble with the genomes into complete particles that are then released from the cell to infect additional cells.
Figure 1. Life cycle of HIV-1. The gp120 protein on the virus surface interacts with CD4 and either CXCR4 or CCR5 on the T cell surface and then gains entry into the cell. The viral core contains two copies of the RNA genome and proteins necessary to reverse-transcribe the RNA to DNA. The resulting double-stranded cDNA (dscDNA) gains entry into the cell nucleus through the nuclear pore complex (NPC). There, the DNA is integrated into the host DNA, establishing a latent infection for future activation. Additional DNA copies are also synthesized and converted to circular non-integrated forms. The virus remains quiescent until the T cell is activated. Activation provides key transcription factors (NF-κB, SP1, NFAT, Tat) needed for transcription of the viral DNA. This produces both new viral RNA genomes and mRNA for the production of viral proteins which are then assembled into complete viral particles and released from the cell. Reprinted by permission from Macmillan Publishers Ltd. from M. Coiras, et al. (2015) Nat. Rev. Microbiol., 7, 298. Copyright 2015.
A careful study of the lifecycle of HIV-1 reveals that the virus does not replicate immediately after infecting a resting T cell. The replication restriction is particularly notable prior to integration of the virus into the host cell genome. Activation of the cell, however, leads to rapid replication and integration. One mechanism that explains this phenomenon is that resting T cells have very low levels of the deoxyribonucleotide triphosphate (dNTP) precursors required for DNA synthesis. Activation of the cells is accompanied by a marked increase in dNTP pools. Similarly, addition of dNTPs to resting infected T cells results in rapid viral replication, while blockade of dNTP synthesis using hydroxyurea, an inhibitor of ribonucleotide reductase, suppresses viral replication. These observations led to the hypothesis that hydroxyurea might be a useful antiviral agent; however attempts to use it as a drug in the clinic led to unacceptable toxicity.
The c-Myc transcription factor regulates the expression of multiple genes required for dNTP synthesis. These include enzymes required for pyrimidine and purine synthesis along with transporters for glucose and amino acids that serve as biosynthetic precursors (Figure 2). C-Myc activity is, in turn, regulated by the Ras/ERK pathway, and inhibition of either ERK or c-Myc suppresses HIV-1 replication. PLD, which converts phosphatidylcholine to phosphatidic acid (PA) and choline, plays a key role in the regulation of Ras/ERK and c-Myc activity during T cell activation. This led investigators in the Brown lab to hypothesize that inhibition of PLD should suppress HIV-1 replication in activated T cells.
Figure 2. Regulation of dNTP synthesis in activated T cells. Stimulation of the T cell receptor (TCR) leads to activation of PLD with production of PA. PA activates ERK and mTORC1, which in turn, activate c-Myc. C-Myc stimulates transcription of the genes for amino acid transporters (Box 1), glucose transporters (Box 2), and enzymes of dNTP synthesis (Box 3). As a result, cells have both the precursors and the enzymes needed for dNTP synthesis. Figure reproduced under the Creative Commons Attribution License (CC BY 3.0) from H. E. Taylor, et al. (2015) PLoS Pathogens, 11, e1004864.
To test their hypothesis, the investigators activated CD4+ T cells with phytohaemagglutinin (PHA) in the presence or absence of VU0359595 (PLD1i) a selective inhibitor of the PLD1 isoform synthesized in the Lindsley laboratory. They found that PLD1i reduced the PHA-stimulated activation of ERK1 and ERK2 and the subsequent phosphorylation of the ERK target, ribosomal S6 protein. Since S6 is also a target of the mTOR signaling pathway and PLD1 has been shown to activate mTOR, the researchers investigated the effects of PLD1i on the mTOR pathway in PHA-activated T cells. The results confirmed mTOR activation in PHA-treated T cells, as indicated by phosphorylation of target proteins, including the translation repressor protein 4E-BP1 and the CAD complex comprising carbamoyl-phosphate synthestase 2, aspartate transcarbamoylase, and dihydroorotase. Phosphorylation of these proteins in PHA-treated T cells was blocked by inhibitors of mTOR, ERK, and PLD1. These three inhibitors also blocked PHA-mediated induction of c-Myc expression and the expression of the c-Myc targets thymidylate synthase (TS), the large subunit of ribonucleotide reductase (RRM1), and the small catalytic subunit of ribonucleotide reductase (RRM2).
As noted above, c-Myc regulates the expression of key transporters of dNTP biosynthetic precursors. These include transporters of glucose (Glut1), glutamine (SNAT1 and SNAT2), and large neutral amino acids (Slc7a5/LAT1). Activation of T cells with α-CD3/α-CD28 treatment resulted in increased levels of all these proteins. These increases were blocked by PLD1i as well as inhibitors of ERK, mTOR, and c-Myc. Together, the results support the hypothesis that T cell activation stimulates PLD1, which in turn, activates ERK and mTOR, ultimately leading to increased expression of c-Myc and its target genes that are required for dNTP synthesis (Figure 2). This hypothesis was supported by a 3.7-, 1.6-, and 9-fold decrease in dATP, dCTP, and dTTP pools in PLD1i-treated activated T cells as compared to T cells activated in the absence of PLD1i.
Increased biosynthesis of dNTPs is required to support the proliferation of activated T cells. Consistently, the investigators found that PLDi treatment delayed DNA synthesis in α-CD3/α-CD28, resulting in an accumulation of cells in the G1b stage of the cell cycle. The net effect was a concentration-dependent inhibition of proliferation of the activated T cells by PLD1i. This current work on the role of PLD1 in HIV replication builds upon previous studies from the Brown lab showing that PLD activity modulates pyrimidine biosynthesis via an mTOR-dependent pathway (T. P. Matthews, et al. (2015) ACS Chem. Biol., published February 17, DOI:10.1021/cb500772c).
The ability of PLD1i to suppress dNTP synthesis in activated T cells suggested that it should be an effective block to HIV-1 replication. Indeed, PLD1i reduced the efficiency of HIV-1 infection by nearly 75% in T cells from four different human donors. The infection efficiency was rescued to varying degrees by the addition of dNTPs. During HIV-1 infection, many more cDNA transcripts are synthesized than are needed to integrate into the host genome. The excess transcripts undergo various processes of self-integration, recombination, and host-mediated DNA repair. The result is a variety of circular forms containing one or two viral long-terminal repeats (LTRs) (Figure 3). To assess the effects of PLD1 inhibition on the synthesis of these unintegrated viral reverse transcripts in infected T cells, the researchers performed RT-PCR to quantitate the viral DNA. The results demonstrated that PLD1i had little effect on the level of early transcripts (ERTs), but markedly suppressed levels of late transcripts (LTRs and 2-LTRs). These findings were consistent with the proposed mechanism of PLD1i. Synthesis of ERTs occurs very early after viral entry and initiation of reverse transcription. This low level of DNA synthesis does not require large dNTP pools. However, synthesis of LTRs and 2-LTRs is dependent on the availability substantial quantities of dNTPs. Addition of exogenous dNTPs to the cells restored LTR but not 2-LTR synthesis by PLD1i-treated infected cells. This finding suggested that PLD1i may suppress HIV-1 replication by one or more mechanisms in addition to its effects on dNTP pools.
Figure 3. Unintegrated viral DNA forms in HIV-1-infected cells. Excess viral DNA that is not integrated into the host genome can undergo multiple processes, leading to various circular forms that contain one or two viral long terminal repeats (LTRs). Figure reproduced under the Creative Commons Attribution License from R.D. Sloan and M.A. Wainberg, (2011) Retrovirology, 8, 52.
These findings suggest that PLD1 inhibition is an exciting and novel approach to the control of HIV-1 replication. Thus far, preclinical studies have revealed remarkably low toxicity for PLD1 inhibitors in vivo. Clearly, further exploration of the clinical utility of PLD1i and similar inhibitors for the treatment of HIV infection is warranted.