Secondary Logo

Journal Logo

Basic Science

Targeting ABL1 or ARG Tyrosine Kinases to Restrict HIV-1 Infection in Primary CD4+ T‐Cells or in Humanized NSG Mice

McCarthy, Stephen D.S. PhDa,b; Leontyev, Danila PhDa,c; Nicoletti, Pauline BSd,e; Binnington, Beth MSd; Kozlowski, Hannah N. BSa,f; Ostrowski, Mario MDa,c; Cochrane, Alan PhDc,g; Branch, Donald R. PhDa,c,d,h; Wong, Raymond W. PhDc

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: December 1, 2019 - Volume 82 - Issue 4 - p 407-415
doi: 10.1097/QAI.0000000000002144



Repositioning well-tolerated kinase inhibitors, with extensive safety and efficacy for various cancer indications, offers the potential for a new class of antivirals to inhibit HIV-1 replication and target chronic inflammation unaddressed by current suppressive therapies. Some of these inhibitors, such as imatinib, have been used successfully to treat Kaposi sarcoma (KS) lesions and chronic myeloid leukemia (CML) in HIV-1 patients on combination antiretroviral therapy (cART).1,2 Unfortunately, HIV-1 patients taking cART are at higher risk for non-AIDS–defining cancers such as lung adenocarcinoma, Hodgkin's lymphoma, and anal cancer.3 It has been suggested this increased risk is due to heightened immune activation and inflammation.4 Thus, novel kinase inhibitors that reduce inflammation, target cancer, and restrict HIV-1 replication are ideal drug candidates to improve cART regimens. Investigating the use of antivirals that target host factors essential for HIV-1 replication, such as dasatinib, may pose higher barriers to drug resistance by blocking multiple stages of infection, offer unique tissue distribution, or alleviate symptoms not targeted by conventional cART.5,6

Dasatinib is an adenosine triphosphate-competitive inhibitor that targets multiple host non-receptor tyrosine kinases, in particular the SRC and ABL families of kinases.7 The ABL family consists of Abelson murine leukemia viral oncogene homolog 1 (ABL1) and a highly similar protein, Abelson-Related Gene (ARG or ABL2). Both genes encode proteins with an SH1 catalytic domain and regulatory SH2 and SH3 domains.8 Unique to the ABL family is their ability to interact with cytoskeleton via an F-actin domain.9 Distinguishing ABL1 from ARG, this kinase contains 3 nuclear localization sequences, a nuclear export signal, and a potential DNA binding domain, whereas ARG contains a microtubule domain and a second F-actin binding domain.9 ABL1 is ubiquitously expressed in the cytosol and nucleus, and has critical roles in T‐cell receptor (TCR) signaling, formation of the immunological synapse, cell adhesion, cytoskeletal rearrangement, gene transcription, cell growth, cell division, and cell differentiation.10,11 In contrast, the function and localization of ARG has been less studied in primary human T‐cells. Mouse experiments suggest ARG has redundant roles with ABL1, where its conditional knockout reduces TCR signaling, interleukin‐2 (IL‐2) production and cell proliferation in mouse T‐cells.12 Yet it remains unknown whether these 2 kinases have different functions during acute HIV-1 infection.

Shortly after HIV-1 envelope gp120 binds CXCR4 or CCR5 coreceptor, ABL1 regulates actin both upstream and downstream of Rac activation, inducing cytoskeletal changes critical for pore formation, pore enlargement, and finally HIV-1 fusion with the target T‐cell.13 Corroborating this mechanism, inhibiting ABL1 activity with dasatinib has been reported to arrest virus–cell fusion during hemifusion in U87 or TXM-BL cells.13 However, it remains untested whether dasatinib inhibits ABL1/ARG activity in a similar manner in activated primary CD4+ T‐cells, the key target cell for HIV-1 infection. ABL1/ARG may also have a role during or after HIV-1 reverse transcription. In a recent study, treating CD4+ T‐cells with dasatinib reduced phosphorylation of host SAM and HD domain‐containing deoxynucleoside triphosphate triphosphohydrolase 1 (SAMHD1) at Thr592.14 Activated SAMHD1 then reduced the cytosolic pool of dNTPs, impairing HIV-1 reverse transcription and subsequent integration.14 ABL1 can also be found in the nucleus. During HIV-1 infection, ABL1 has been shown to activate the promoter of integrated HIV-1 in the absence of viral Tat, which correlated with increased tyrosine phosphorylation of RNA polymerase II (RNA pol II).15 Hence, there are multiple mechanisms by which ABL1, and perhaps ARG, influence HIV-1 infection during virus-cell fusion, reverse transcription, or transcription of integrated virus in T‐cells.

Previously we have shown that 10–100 nM of dasatinib significantly reduced infection of HXB2 (X4) and JR-FL (R5) recombinant luciferase reporter viruses in activated primary CD4+ T‐cells.16 This finding is congruent with a recent study that 75 nM of dasatinib inhibited HIV-1 p24 production in activated CD4+ T‐cells isolated from treatment naïve HIV-1+ donors.17 However, a functional role of ABL1 or ARG during the early stages of CD4+ T‐cell infection of HIV-1, and whether these kinases differ from each other, is less clear. In this work, we demonstrate that targeted siRNA knockdown of ABL1 or ARG primarily inhibits HIV-1 integration during early infection of activated, primary CD4+ T‐cells. Differentiating these 2 kinases, SAMHD1 activation increased following ARG knockdown, whereas RNA pol II activation decreased following either ABL1 or ARG knockdown. We also performed proof-of-concept experiments in a humanized NSG mouse model of HIV-1 infection,18 providing preliminary evidence for the efficacy of dasatinib treatment during acute HIV-1 infection in mice.


More details can be found in Supplemental Digital Content 1,

Cell Isolation and Cell Culture Conditions

Informed written consent was obtained from all subjects. CD4+ T‐cells were isolated from PBMCs of healthy HIV-negative donors by Ficoll gradient, and enriched with the EasySep negative selection human CD4+ T‐cell kit (StemCell, Vancouver, British Columbia, Canada). Isolated CD4+ T‐cells were grown in complete Roswell Park Memorial Institute (RPMI) 1640 media containing fetal bovine serum, penicillin, streptomycin, and gentamycin. CD4+ T‐cells were activated with 10 μg/mL anti-CD3 (UCHT-1), 2 μg/mL anti-CD28 (10R-CD28bHU) and 10 units/mL recombinant IL-2. PBMCs isolated from an HIV-1+ donor were depleted of CD8+ cells with Dynabeads CD8, followed by growth in complete RPMI media with activators.19

Western Blot and Antibodies

CD4+ T‐cells were lysed with RIPA and resolved on reducing SDS-PAGE gels with or without Stain-Free™ labeling of total proteins, and subsequently transferred to polyvinylidene difluoride membrane for Western blot detection. Anti-ABL1/ARG, anti-SAMHD1, anti-phospho Thr592 SAMHD1, anti-RNA pol II, and anti-phospho Tyr1 RNA pol II C-terminal repeat domain (CTD) were used to probe the blots. Secondary antibodies were added, followed by ECL detection.

siRNA Knockdown Experiments

Small interfering RNA (siRNA) targeting ABL1 or ARG (Dharmacon, Lafayette, CO) were electroporated into T‐cells with the Amaxa Human T‐Cell Nucleofector Kit (Lonza, Cologne, Germany), program V-024.16 Cells were activated 12 hours thereafter. Cells were harvested for qRT-PCR at 24 hours post‐transfection, or prepared for HIV-1 infection at 48 hours post-transfection.

qRT-PCR and qPCR

Twenty-four hours after siRNA electroporation, total RNA was extracted from CD4+ T‐cells from each siRNA treatment using the Trizol method. cDNA was synthesized using the iScript Reverse Transcriptase kit (BioRad, Hercules, CA). Primer sets used to detect ABL1 or ARG cDNA20,21 were cycled on a Rotor-Gene RG-3000 thermocycler (Corbett Research, Montreal, Québec, Canada). Amplicons were detected with SYBR Green I quantitative polymerase chain reaction (qPCR). ABL1 and ARG siRNA-treated groups were compared with the non‐targeting siRNA-treated group by the 2−ΔCT comparative CT method.22

In HIV-1 infection experiments with R5 (JR-FL or Ba-L) or X4 (HXB2 or IIIB) strains, CD4+ T‐cells pretreated with siRNA were harvested at 24 hours with the QIAamp DNA Blood Mini Kit (Qiagen, Toronto, Ontario, Canada). Four viral cDNA targets were amplified by SYBR Green qPCR on the Rotor-Gene RG-3000 thermocycler: early reverse transcripts (early RT), late reverse transcripts (late RT), 2-LTR circles, or integrated virus by Alu-gag nested PCR.16 Human β-globin was measured to normalize the data.

HIV-1 Infection and Luciferase Activity

We infected CD4+ T‐cells with nef-deficient, HXB2-enveloped, or JR-FL-enveloped luciferase reporter viruses, and measured luciferase activity 3 days post‐infection. Viruses were made as previously described.16

Ex Vivo Dasatinib Experiment and p24 ELISA

CD4+ T‐cells isolated from an HIV donor were grown in RPMI media ± activators and 100 nM of dasatinib (Selleckchem, Burlington, Ontario, Canada) or dimethyl sulfoxide (DMSO) vehicle. After 48 hours, Western blot detection of SAMHD1 phosphorylation was performed on cell lysates. The PBMCs of an HIV+ donor were isolated and depleted of CD8+ T‐cells, then grown in complete RPMI media with activators. Cells were given fresh media containing 10–100 nM of dasatinib or DMSO vehicle twice a week, for 21 days. Virus in the supernatant was quantified by p24gag antigen ELISA, and cell viability by the XTT assay.

Human Leukocyte Reconstitution of NSG Mice

Animal procedures were approved by the University of Toronto. We characterized leukocyte subpopulations in a published humanized mouse model suitable for HIV-1 infection studies.18,23 Four-week old male and female NOD/LtsZ-scidIL2Rγnull (NSG) mice were purchased from Jackson Labs, Farmington, CT. Mice were anesthetized with isoflurane for all procedures. Mice were injected with 50 mg/kg myeloablative busulfan by 2 intraperitoneal (i.p.) injections with a 12-hour delay. Twenty-four hours thereafter, 250,000 human CD34+ cells pooled from cord blood hematopoietic progenitor cells (CB-HPCs) were transplanted into mice via tail-vein injection. Multi-lineage reconstitution of human leucocytes was confirmed by flow cytometry 19 weeks after engraftment.

Ba-L Infection of Humanized NSG Mice

HIV-1 Ba-L was injected into 19 humanized mice by i.p. Starting at week 5 of infection, blood plasma samples were taken weekly to analyze viral RNA with the m2000 RealTime HIV-1 Viral Load Assay (Abbott, Mississauga, Ontario, Canada). Between weeks 5–7 of infection, mice were sub-Q injected daily with T-20, dasatinib, or vehicle diluent. At the experimental endpoint, all animals were anesthetized and humanely euthanized by exsanguination.

Flow Cytometry and Data Acquisition

Nineteen weeks post‐engraftment, we collected mouse PBMCs from saphenous vein for flow cytometry. PBMCs were stained with anti-mouse CD45, anti-human CD45, anti-human CD33, anti-human CD3, anti-human CD4, anti-human CD8β, anti-human TCRαβ, anti-human CD19, anti-human CCR5, and live/dead fixable aqua. Cells were acquired using a LSRFortessa (BD Biosciences, Mississuaga, Ontario, Canada).

Statistical Analysis

Means were compared using a two-tailed, unpaired Student's t test. Fluorescent activated cell sorting data was analyzed by FlowJo version10 software (Treestar). A two-way analysis of variance was performed in Figure 5B. For all figures, an asterisk (*) denotes a P value <0.05, 2 asterisks (**) denotes a P value <0.01 and 3 asterisks (***) denotes a P value <0.001. Error bars shown are the SEM.


ABL1 and ARG Are Highly Expressed in Activated CD4+ T‐cells

We first compared ABL1 and ARG protein expression after T‐cell activation in enriched CD4+ T‐cells (Fig. 1A) isolated from healthy human donors. Antibody detecting ABL1/ARG suggests marked increase in their expression in CD4+ T‐cells after activation with IL-2, anti-CD28, and anti-CD3 in cell culture. We also performed targeted siRNA knockdown of ABL1 or ARG expression in CD4+ T‐cells.

ABL1/ARG protein expression after siRNA knockdown in CD4+ T‐cells. 1.0 × 107 enriched CD4+ T‐cells isolated from a healthy donor were electroporated with of nontargeting (NT), ABL1- or ARG-targeting siRNA. A, Western blot of ABL1/ARG and total protein from cell lysate of 1 million enriched CD4+ T‐cells, 48 hours post-siRNA knockdown. Cells were cultured in RPMI without (lane 1) or with T‐cell activators (lanes 2–4) for 36 hours post-siRNA knockdown. B, ABL1 and ARG mRNA expression 24 hours post-siRNA knockdown (see Figure S1, Supplemental Digital Content 1, for qRT-PCR optimization). Shown are ABL1 or ARG mRNA expression relative to GAPDH by qRT-PCR. Means are of 3 biological replicates, and error bars are the SEM. These experiments are representative of 2 independent experiments.

Because of high protein sequence similarity between ABL and ARG proteins,24 we opted to document siRNA knockdown of both kinases through mRNA expression using qRT-PCR (Fig. 1B). Twenty-four hours post-siRNA knockdown, we could consistently suppress ABL1 or ARG mRNA expression by 70%–75%, without significantly altering expression of the other kinase and had no appreciable effect on cell viability (data not shown).

siRNA Knockdown of ABL1 or ARG Reduce in vitro HIV-1 Infection at the Level of Viral Integration

With enriched CD4+ T‐cells expressing less ABL1 or ARG, we then infected them with a luciferase reporter virus 2 days post-siRNA knockdown (Figs. 2A, B). Luciferase detection of JR-FL (R5) or HXB2 (X4) was performed 3 days after infection. At this time-point, ARG siRNA knockdown reliably reduced luciferase activity of both viruses by 54%–77%. However, knockdown of ABL1 could only reduce luciferase activity of JR-FL infection by 64%. To determine which stage during the early HIV lifecycle these kinases have their effect, we repeated the previous experiment and extracted viral cDNA from cell lysates 24 hours post‐infection (Figs. 2C–F). Two R5 viruses (JR-FL, Ba-L) or 2 X4 viruses (HXB2, IIIB) were tested. Unexpectedly, we detected no significant change of early or late reverse transcripts after ABL1 or ARG siRNA knockdown. Rather, siRNA knockdown of ABL1 or ARG consistently reduced viral integration by 2- to 12-fold, and increased 2-LTR circle formation by 2- to 4- fold. Given the similar effects observed between the R5 and X4 HIV-1, this data suggests siRNA knockdown of ABL1 or ARG in activated CD4+ T‐cells reduces HIV-1 infection ex vivo at a step after reverse transcription.

Effects of ABL1 or ARG siRNA knockdown on HIV-1 luciferase reporter activity or viral integration. Forty-eight hours post-siRNA nucleofection and T‐cell activation, 5 × 105 enriched CD4+ T‐cells were infected with either R5 or X4 HIV-1 strains. A and B, Luciferase activity was measured from cell lysates 3 days post‐infection with JR-FL (R5) or HXB2 (X4) viruses. C and D, In separate experiments, DNA was extracted from T‐cells after 24 hours of JR-FL or HXB2 infection, to measure early reverse transcripts (Early RT), late reverse transcripts (Late RT), 2-LTR circle formation or integrated virus. E and F, Other T‐cells were infected with Ba-L (R5) or IIIB (X4) viruses. qPCR of HIV-1 cDNA was normalized to genomic β-globin, and graphed relative to the non‐targeting (NT) siRNA treatment set as 1-fold change (x-axis). The means shown are the average of biological triplicates. Error bars are SEM. These experiments are representative of 2 independent experiments.

siRNA Knockdown of ABL1 and ARG Alter the Activation of SAMHD1 and RNA Polymerase II

To explore potential downstream mechanisms of ABL1 or ARG knockdown in activated CD4+ T‐cells, we quantified the phosphorylation status of SAMHD1 (Thr592) and RNA polymerase II CTD Tyr1. Activating CD4+ T‐cells alone reduced SAMHD1 activation (increased Thr592 phosphorylation) by half, as expected for proliferating cells (Fig. 3A). ARG knockdown, but not ABL knockdown (P = 0.27), showed a significant increase in SAMHD1 activation. These results are consistent with the effects of the ABL family kinase inhibitor dasatinib (Fig. 3B). With respect to activation of RNA polymerase II at the CTD, siRNA knockdown of either kinase moderately reduced Tyr1 phosphorylation (Fig. 3C). This reduced phosphorylation was strongest in T‐cells pretreated with ABL1 siRNA. Collectively, these findings demonstrate a role for ABL1 or ARG signaling in the activation of 2 known cellular restriction factors of HIV-1, complementing the effects observed on viral cDNA in Figure 2.

Phosphorylation of SAMHD1 or RNA pol II after ABL1 or ARG knockdown or kinase inhibition. A, Enriched CD4+ T‐cells from a healthy donor were electroporated with NT, ABL1, or ARG siRNA. Cells were cultured for 36 hours either without T‐cell activators or with anti-CD3, anti-CD28, and IL‐2. One million cells were lysed and prepared for phospho-Western blots detecting Thr592-phosphorylated and total SAMHD1. B, Enriched CD4+ T‐cells from a healthy donor were incubated with 100 nM of dasatinib or DMSO solvent control. Forty-eight hours post‐drug incubation and T‐cell activation, lysed cells were prepared for phospho-Western blot of SAMHD1 activation. C, Detection of Tyr1-phosphorylated CTD and total RNA pol II, from cell lysates treated as described in (A). Phospho-protein in each lane was normalized to total protein (not shown). The means displayed are the average of 3 independent experiments on 3 different blood donors in (A and C), or one donor in (B). Error bars are SEM. Statistical comparisons were made to activated T-cells treated with NT siRNA.

The ABL1/ARG Kinase Inhibitor Dasatinib Significantly Reduces HIV-1 p24 Production

We and others have reported the use of inhibiting HIV-1 production with the ABL1/ARG kinase inhibitor dasatinib.16,17 To confirm and extend these findings, we isolated PBMCs from a treatment-naïve HIV-1+ donor, depleted them of CD8+ cells, and dosed the cells with 10–100 nM of dasatinib for 3 weeks (Fig. 4). Activated cells began showing detectable p24 levels by week 3. This was inhibited by 58.9% with a 10 nM dose of dasatinib administered twice a week. p24 levels were nearly undetectable at either the 50 nM or 100 nM dose of dasatinib, inhibiting p24 in the supernatant by 99.8% relative to the negative control group. A minimal reduction in cell viability was observed at the 50 nM dose, but not at the 100 nM dose (Fig. 4). Thus, these results support our rationale to test the efficacy of dasatinib in an animal model of HIV-1 infection.

p24 measurements from PBMCs of an HIV-1+ donor treated with dasatinib ex vivo. CD8-depleted cells were activated in cell culture with anti-CD3, anti-CD28, and IL-2, and administered dasatinib (10 nM, 50 nM or 100 nM), or DMSO solvent control twice a week for 21 days. Virus in supernatant was quantified weekly by p24 ELISA. Cell viability was assessed on day 21 by the XTT assay. Means are the average of 3 biological replicates. Error bars are the SEM.

Dasatinib Significantly Reduces Ba-L Infection of Humanized NSG Mice

We optimized an established humanized mouse model of HIV-1 infection. Approximately 20 NSG mice were used to determine the best method of myeloablation (irradiation vs. busulfan) and to characterize human cell populations in PBMCs isolated from mice 19‐weeks post‐engraftment of human CD34+ pluripotent cells (see Figure S2, Supplemental Digital Content 1, At this time-point of maximal reconstitution, 93.2% of lymphocytes expressed human CD45 (panel A). Of these lymphocytes, 43.7% expressed the human B-cell receptor CD19, whereas 31% expressed the human T‐cell receptor CD3 (panel B). The human T‐cell sub-population consisted predominantly of cells expressing human CD4 (58.9%), and fewer cells expressing human CD8 (36.1%, panel C). Human myeloid cells were also reconstituted, as shown by the 14.2% of lymphocytes expressing human surface CD33 (panel D). In the human CD4+ T‐cell subset, 29.9% of them expressed a TCR (panel E), with 9.99% coexpression of human CCR5 receptor. Hence, our mouse model demonstrates broad reconstitution of human cells of myeloid and lymphoid lineages, and human CD4+ T‐cells that express CCR5 at levels similar in PBMCs isolated from healthy human donors.16

Further optimization experiments determined that HIV-1 infection with Ba-L (R5) at a dose of 10,000 TCID50, lead to a reproducible course of acute HIV-1 infection (Fig. 5A). Infected mice typically demonstrated 6–7 weeks of exponential increase of plasma viral p24, which would plateau at week 7, then gradually descend. Daily treatment with 25 mg/kg sub-Q injections of T-20 fusion inhibitor between weeks 5–7 of infection, blunted this exponential increase in viremia, reducing plasma viral load to nearly undetectable levels by week 7 (97.6% inhibition). Because of variations in human cell repopulation after 19 weeks of engraftment, mice showed variable viral RNA plasma levels 5 weeks post‐infection. These could range over 3 orders of magnitude. Thus, a baseline viremia of Ba-L was set at week 5 in our next experiment (Fig. 5B). At this time-point, mice were treated daily with 25–30 mg/kg of dasatinib by subcutaneous (sub‐Q) injections, or injected with vehicle control. We determined this range to be the maximum tolerated dose (MTD) of dasatinib. Similar to the T-20 treatment, dasatinib significantly attenuated the increase in viremia during weeks 5–7 post-infection. By week 7, plasma viral load was reduced by 95.1% (20-fold reduction) in dasatinib-treated mice, relative to vehicle-treated mice.

Treatment of humanized NSG mice infected with Ba-L with T-20 or dasatinib. Nineteen weeks post-CB-HPC transplant, mice were injected with Ba-L (TCID50 = 10,000) sub-Q. A, HIV-1 RNA viral load in the plasma was quantified 3–7 weeks post‐infection. At 5 weeks post-infection (arrow), 2 mice were given daily sub-Q injections of T-20 at 25 mg/kg, until week 7. Another 2 mice were injected daily with equivalent volume of PEG-400 + Tween-80 vehicle in PBS, to serve as negative controls. Error bars represent the range. B, Baseline plasma viral load of Ba-L was measured and normalized at week 5 post-infection in 15 mice. After week 5, 7 of these mice were given daily sub-Q injections of dasatinib at 25–30 mg/kg, until week 7. The other 8 mice were injected daily with vehicle and served as negative controls. Bars labeled by the same letter on the graph are not significantly different from each other by two-way analysis of variance. Bars labeled by a different letter on the graph are statistically significantly different from each other (P < 0.05) by two‐way analysis of variance. Means are the averages of mice from 5 independent experiments (1–3 mice per experiment). Error bars are the SEM.


Through 3 complimentary modes of investigation, our study of ABL family kinases and the potent ABL family inhibitor dasatinib reveal new biological insights into HIV-1 and host T‐cells it infects.

We first demonstrated that siRNA knockdown of ABL1 or ARG had stronger effects against the R5 viruses tested (JR-FL, Ba-L) compared with the X4 viruses examined (HXB2, IIIB). This was true whether we used luciferase activity or qPCR as our readout of HIV-1 infection (Fig. 2). We previously reported that dasatinib and other SRC family kinase inhibitors exhibit more potent effects against JR-FL than HXB2 infection in primary CD4+ T‐cells.16 Although the role of ABL1 has been well-defined downstream of TCR engagement,25 whether ABL1 or ARG signaling differs upon HIV-1 binding to its co-receptor CXCR4 or CCR5 warrants further research.

We also assessed whether siRNA knockdown of ABL1 or ARG inhibited HIV-1 infection post‐entry in primary CD4+ T lymphocytes. Contrary to the report of Harmon et al., and consistent with the findings of Bermejo et al., we determined ABL1 and ARG siRNA knockdown have effects post‐viral fusion during acute HIV-1 infection.13,14 qPCR of viral DNA 24-hours post‐infection suggest their effects occur before 2-LTR circle formation and viral integration (Fig. 2). These highlight, for the first time, new roles that these kinases may play during HIV-1 infection.

Downstream of ABL1 or ARG signaling, SAMHD1 is a major restriction factor that blocks HIV-1 reverse transcription in resting CD4+ T‐cells depleting the cytosolic pool of dNTPs available to the virus, and degrading viral RNA. Its activity and tetrameric stability is tightly regulated by cyclin A2, CDK1 and CDK2, which phosphorylate SAMHD1 at threonine 592.26,27 The reported potent effects of dasatinib reducing SAMHD1 activity during HIV-1 infection could stem from ABL family kinases that phosphorylate CDKs.28 In addition to this mechanism in resting T‐cells, another mechanism of SAMHD1 viral restriction dependent on Thr592 phosphorylation is independent of dNTPase activity in activated CD4+ T‐cells.29,30 These findings are consistent with our results of ARG siRNA knockdown having no effect on the synthesis of early and late HIV-1 reverse transcripts (Fig. 2), yet significantly increasing SAMHD1 activation (Fig. 3).

Regarding the nuclear effects of ABL1 or ARG, it has been reported that ABL1 shuttles from the cytoplasm to the nucleus and becomes activated within minutes of DNA damage, by either ATM or DNA-PK phosphorylating ABL at serine 465.31 Cells deficient in ATM or DNA-PK demonstrate abnormal junctions between HIV-1 provirus and host DNA, and these DNA double-strand break (DSB) repair enzymes are believed to have multiple roles during HIV-1 integration.32 Moreover, it has been found that inhibiting the catalytic subunits of DNA-PKs only has modest effects on 2-LTR circle formation, which is primarily regulated by enzymes in the nonhomologous end-joining pathway.33 This is congruent with our finding that ABL1 siRNA knockdown reduced viral integration and led to a concomitant increase in 2-LTR circle formation. In comparison, less is known about nuclear ARG kinase activity. The kinase has 8 isoforms, of which one localizes specifically to the nucleus.34 It has been suggested that this truncated ARG isoform lacks 2 putative nuclear export signals that would allow it to shuttle back to the cytoplasm.34 In the nucleus, ARG phosphorylates and associates with Rad51 in response to DNA damage, and may have a role in homologous recombination DNA repair.35 Rad51 is stimulated downstream of NF-kB and interacts with the HIV-1 promoter,36 providing another potential mechanism by which the ABL family of kinases may promote HIV-1 replication, which can be targeted for pharmaceutical intervention.

To provide evidence for preclinical evaluation of the ABL family kinase inhibitor dasatinib, we tested 3 doses of the drug on CD8-depleted PBMCs isolated from a treatment-naïve, HIV-1+ donor (Fig. 4). 58.9% inhibition of p24 levels with 10 nM of dasatinib is comparable to another study that investigated ex vivo inhibition of dasatinib on HIV-1 production after 10 days of cell growth.17 Although dasatinib likely has pleiotropic effects on other cell kinases, it is also possible that ABL1 and ARG inhibition could reduce RNA pol II activation, as demonstrated in Figure 3. Before Tat protein synthesis, it has been reported that nuclear ABL1 phosphorylates the C-terminal repeated domain of RNA pol II, activating the HIV-1 promoter and stimulating transcription elongation.15 After Tat synthesis, ABL1 signals downstream of Tat in a p73 pathway that may have a role in Tat-mediated neurological damage during HIV-1 infection.37 Taken together, our ex vivo ABL1/ARG siRNA knockdown and dasatinib findings provided strong rationale for testing of dasatinib in a humanized mouse model of HIV-1 infection.

NSG mice engrafted with human CD34+ cells demonstrated high levels of human leucocyte reconstitution in peripheral blood 19-weeks post‐transplant (Fig S2). Although this HIV-1 model has limitations, such as purportedly reduced anti-HIV-1 humoral responses or proper B-cell maturation, it offers a sophisticated model of CD4 and CD8 T‐cell populations with normal T‐cell function and lymphoid organogenesis.18 It also mimics the time-course of human HIV-1 replication, as shown in Figure 5A. Our proof-of-principle experiment of administering 25–30 mg/kg of dasatinib daily demonstrated a 95.1% reduction in plasma viral load by week 7 of Ba-L infection, comparable to mice treated with the fusion inhibitor T-20. While promising, future experiments should examine oral dosing regiments of dasatinib, its minimum effective dose, and its potential immunosuppressive effects on B-cells.38 We are optimistic that future HIV-1 animal model experiments will investigate the potential of cART regiments in combination with dasatinib, since dasatinib has shown synergy in inhibiting other viruses such as HCV when combined with sofosbuvir.39

In conclusion, the findings of this work demonstrate new roles of the ABL family kinases during early HIV-1 infection of CD4+ T‐cells, and evaluate a kinase inhibitor that inhibits both of these targets in vivo. For treating second-line CML, dasatinib is a safe and well-tolerated kinase inhibitor at a daily oral dose of 100 mg.40,41 Moreover, CML in HIV-1 infected patients has been treated successfully with no additional myelosuppression reported.42 Our previous work has shown a significant role for c-SRC in driving HIV-1 integration,16 and dasatinib is known to inhibit c-SRC and ABL. Our work herein, coupled with our previous report implicating a significant role for c-SRC in HIV-1 infection, suggests that c-SRC, ABL1, and ARG are all important kinases for productive HIV-1 infection, and that inhibition of all 3 might be necessary. Indeed, the related SRC/ABL inhibitor, imatinib, has been used to safely treat KS or CML cancers in HIV-1 patients on fully suppressive cART, improving long-term survival.1,42,43 The clinical application of ABL-family kinase inhibitors of known specificity, biodistribution, pharmacokinetics, toxicity profile, and ease of administration suggest dasatinib could be repositioned as a novel inhibitor of HIV-1 infection, and should be explored further in animal models of HIV-1 disease progression.


The authors thank K. Johnson-Henry and S. Robinson (P.M. Sherman's lab) and Nades Palaniyar of The Hospital for Sick Children for use of a CFX96 and ChemiDoc MP.


1. Koon HB, Bubley GJ, Pantanowitz L, et al. Imatinib-induced regression of AIDS-related Kaposi's sarcoma. J Clin Oncol. 2005;23:982–989.
2. Hagiwara S, Yotsumoto M, Odawara T, et al. Non-AIDS-defining hematological malignancies in HIV-infected patients: an epidemiological study in Japan. AIDS. 2013;27:279–283.
3. Worm SW, Bower M, Reiss P, et al. Non-AIDS defining cancers in the D:A:D Study—time trends and predictors of survival: a cohort study. BMC Infect Dis. 2013;13:471.
4. Kowalkowski MA, Day RS, Du XL, et al. Cumulative HIV viremia and non-AIDS-defining malignancies among a sample of HIV-infected male veterans. J Acquir Immune Defic Syndr. 2014;67:204–211.
5. Readinger JA, Schiralli GM, Jiang JK, et al. Selective targeting of ITK blocks multiple steps of HIV replication. Proc Natl Acad Sci U S A. 2008;105:6684–6689.
6. Wrasidlo W, Crews LA, Tsigelny IF, et al. Neuroprotective effects of the anti-cancer drug sunitinib in models of HIV neurotoxicity suggests potential for the treatment of neurodegenerative disorders. Br J Pharmacol. 2014;171:5757–5773.
7. Lombardo LJ, Lee FY, Chen P, et al. Discovery of N-(2-chloro-6-methyl- phenyl)-2-(6- (4-(2-hydroxyethyl)- piperazin-1-yl)-2-methylpyrimidin-4- ylamino)thiazole-5- carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays. J Med Chem. 2004;47:6658–6661.
8. Mayer BJ, Baltimore D. Mutagenic analysis of the roles of SH2 and SH3 domains in regulation of the Abl tyrosine kinase. Mol Cell Biol. 1994;14:2883–2894.
9. Wang Y, Miller AL, Mooseker MS, et al. The Abl-related gene (Arg) nonreceptor tyrosine kinase uses two F-actin-binding domains to bundle F-actin. Proc Natl Acad Sci U S A. 2001;98:14865–14870.
10. Sirvent A, Benistant C, Roche S. Cytoplasmic signalling by the c-Abl tyrosine kinase in normal and cancer cells. Biol Cell. 2008;100:617–631.
11. Hardin JD, Boast S, Schwartzberg PL, et al. Abnormal peripheral lymphocyte function in c-abl mutant mice. Cell Immunol. 1996;172:100–107.
12. Zipfel PA, Zhang W, Quiroz M, et al. Requirement for Abl kinases in T cell receptor signaling. Curr Biol. 2004;14:1222–1231.
13. Harmon B, Campbell N, Ratner L. Role of Abl kinase and the Wave2 signaling complex in HIV-1 entry at a post-hemifusion step. PLoS Pathog. 2010;6:e1000956.
14. Bermejo M, Lopez-Huertas MR, Garcia-Perez J, et al. Dasatinib inhibits HIV-1 replication through the interference of SAMHD1 phosphorylation in CD4+ T‐cells. Biochem Pharmacol. 2016;106:30–45.
15. Baskaran R, Escobar SR, Wang JY. Nuclear c-Abl is a COOH-terminal repeated domain (CTD)-tyrosine (CTD)-tyrosine kinase-specific for the mammalian RNA polymerase II: possible role in transcription elongation. Cell Growth Differ. 1999;10:387–396.
16. McCarthy SD, Sakac D, Neschadim A, et al. c-SRC protein tyrosine kinase regulates early HIV-1 infection post-entry. AIDS. 2016;30:849–858.
17. Pogliaghi M, Papagno L, Lambert S, et al. The tyrosine kinase inhibitor Dasatinib blocks in-vitro HIV-1 production by primary CD4+ T‐cells from HIV-1 infected patients. AIDS. 2014;28:278–281.
18. Singh M, Singh P, Gaudray G, et al. An improved protocol for efficient engraftment in NOD/LTSZ-SCIDIL-2Rgammanull mice allows HIV replication and development of anti-HIV immune responses. PLoS One. 2012;7:e38491.
19. Wong RW, Balachandran A, Ostrowski MA, et al. Digoxin suppresses HIV-1 replication by altering viral RNA processing. PLoS Pathog. 2013;9:e1003241.
20. Perego RA, Bianchi C, Brando B, et al. Increment of nonreceptor tyrosine kinase Arg RNA as evaluated by semiquantitative RT-PCR in granulocyte and macrophage-like differentiation of HL-60 cells. Exp Cell Res. 1998;245:146–154.
21. Anunobi CC, Koli K, Saxena G, et al. Expression of the SIBLINGs and their MMP partners in human benign and malignant prostate neoplasms. Oncotarget. 2016;7:48038–48049.
22. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc. 2008;3:1101–1108.
23. Xu SX, Leontyev D, Kaul R, et al. Neisseria gonorrhoeae co-infection exacerbates vaginal HIV shedding without affecting systemic viral loads in human CD34+ engrafted mice. PLoS One. 2018;13:e0191672.
24. Kruh GD, Perego R, Miki T, et al. The complete coding sequence of arg defines the Abelson subfamily of cytoplasmic tyrosine kinases. Proc Natl Acad Sci U S A. 1990;87:5802–5806.
25. Gu JJ, Ryu JR, Pendergast AM. Abl tyrosine kinases in T‐cell signaling. Immunol Rev. 2009;228:170–183.
26. St Gelais C, Kim SH, Ding L, et al. A putative cyclin-binding motif in human SAMHD1 contributes to protein phosphorylation, localization, and stability. J Biol Chem. 2016;291:26332–26342.
27. St Gelais C, Kim SH, Maksimova VV, et al. A cyclin-binding motif in human SAMHD1 is required for its HIV-1 restriction, dNTPase activity, tetramer formation, and efficient phosphorylation. J Virol. 2018;92:e01787–17.
28. Zukerberg LR, Patrick GN, Nikolic M, et al. Cables links Cdk5 and c-Abl and facilitates Cdk5 tyrosine phosphorylation, kinase upregulation, and neurite outgrowth. Neuron. 2000;26:633–646.
29. White TE, Brandariz-Nunez A, Valle-Casuso JC, et al. The retroviral restriction ability of SAMHD1, but not its deoxynucleotide triphosphohydrolase activity, is regulated by phosphorylation. Cell Host Microbe. 2013;13:441–451.
30. Welbourn S, Dutta SM, Semmes OJ, et al. Restriction of virus infection but not catalytic dNTPase activity is regulated by phosphorylation of SAMHD1. J Virol. 2013;87:11516–11524.
31. Tang J, Wang JY, Parker LL. Detection of early Abl kinase activation after ionizing radiation by using a peptide biosensor. Chembiochem. 2012;13:665–673.
32. Sakurai Y, Komatsu K, Agematsu K, et al. DNA double strand break repair enzymes function at multiple steps in retroviral infection. Retrovirology. 2009;6:114.
33. Kilzer JM, Stracker T, Beitzel B, et al. Roles of host cell factors in circularization of retroviral DNA. Virology. 2003;314:460–467.
34. Bianchi C, Torsello B, Di Stefano V, et al. One isoform of Arg/Abl2 tyrosine kinase is nuclear and the other seven cytosolic isoforms differently modulate cell morphology, motility and the cytoskeleton. Exp Cell Res. 2013;319:2091–2102.
35. Li Y, Shimizu H, Xiang SL, et al. Arg tyrosine kinase is involved in homologous recombinational DNA repair. Biochem Biophys Res Commun. 2002;299:697–702.
36. Kaminski R, Wollebo HS, Datta PK, et al. Interplay of Rad51 with NF-kappaB pathway stimulates expression of HIV-1. PLoS One. 2014;9:e98304.
37. Bagashev A, Mukerjee R, Santerre M, et al. Involvement of miR-196a in HIV-associated neurocognitive disorders. Apoptosis. 2014;19:1202–1214.
38. Dargart JL, Fish K, Gordon LI, et al. Dasatinib therapy results in decreased B cell proliferation, splenomegaly, and tumor growth in a murine model of lymphoma expressing Myc and Epstein-Barr virus LMP2A. Antiviral Res. 2012;95:49–56.
39. Xiao F, Fofana I, Thumann C, et al. Synergy of entry inhibitors with direct-acting antivirals uncovers novel combinations for prevention and treatment of hepatitis C. Gut. 2015;64:483–494.
40. Shah NP, Guilhot F, Cortes JE, et al. Long-term outcome with dasatinib after imatinib failure in chronic-phase chronic myeloid leukemia: follow-up of phase 3 study. Blood. 2014;123:2317–2324.
41. Shah NP, Rousselot P, Schiffer C, et al. Dasatinib in imatinib-resistant or -intolerant chronic-phase, chronic myeloid leukemia patients: 7-year follow-up of study CA180-034. Am J Hematol. 2016;91:869–874.
42. Patel M, Philip V, Fazel F, et al. Human immunodeficiency virus infection and chronic myeloid leukemia. Leuk Res. 2012;36:1334–1338.
43. Schlaberg R, Fisher JG, Flamm MJ, et al. Chronic myeloid leukemia and HIV-infection. Leuk Lymphoma. 2008;49:1155–1160.

dasatinib; drug repurposing; protein tyrosine kinases; proto-oncogene protein c-abl; Abl-related gene tyrosine kinase; CD4 positive T lymphocytes

Supplemental Digital Content

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.