Dimethyl sulfoxide

Objectives.-To determine the relationship, if any, between reagents that modulate survival of T-cells and replication of human immunodeficiency virus I (HIV-1) and to determine the effects of the solvent dimethyl sulfoxide (DMSO) and drugs such as cyclosporin A and all-trans retinoic acid on HIV-1 replication.

Design.-To first establish the direct effects of solvent alone (ie, DMSO) at various concentrations on HIV-1 replication, followed by the ability of various compounds such as the caspase inhibitor N-benzyloxycarbonyl-val-ala-aspfluoromethylketone (z-VAD-fmk), cyclosporin A, and alltrans retinoic acid on HIV-11 replication. Next, to determine if HIV-1 induces T-cell apoptosis using TUNEL (TdTmediated dUTP-biotin nick end-labeling) assays and DNA fragmentation and poly-(ADP-ribose)-polymerase (PARP) cleavage, and then to examine how the various compounds influence T-cell survival after HIV-1 exposure.

Methods.-The human T-cell line, CEM cells, were exposed to HIV^sub IIIB^ and viral replication monitored using reverse transcription assays at 3, 6, and 9 days following infection. Cells were pretreated with various compounds dissolved in DMSO over a wide range of concentrations, and DMSO itself was also examined. T-cell death and apoptosis were assessed using TUNEL staining to detect 3'-OH DNA strand breaks and agarose gel electrophoresis to detect DNA fragmentation (laddering). Furthermore, PARP cleavage implicated in the apoptotic process was also examined.

Results.-At very low levels, such as 0.002%, DMSO it

self appears to enhance HIV-1 replication at 6 and 9 days after infection. At low levels of cyclosporin A, such as 0.01 (mu)g/mL, HIV-1 replication was further enhanced above the solvent effect, but at 1 (mu)g/mL, cyclosporin A strongly inhibited HIV-1 replication. Retinoic acid between 0.01 and 1 (mu)g/mL did not influence HIV-1 replication. in addition, a discrepancy was noted in that HIV-1-infected T-cells were TUNEL positive, indicating DNA strand breaks; however, more complete DNA fragmentation was not detected nor was PARP cleavage identified. The induction of TUNEL positivity was blocked by the caspase inhibitor z-VAD-fmk but not by DMSO or cyclosporin A. Even though z-VADfmk blocked the appearance of TUNEL-positive T-cells, there was not a consistently observed increase in HIV-1 replication.

Conclusion.-Low levels of DMSO and cyclosporin A can enhance HIV-1 replication in CEM cells. At higher levels, cyclosporin A inhibits HIV-1 replication with no significant effects by all-trans retinoic acid. No evidence for classic apoptosis was detected in CEM cells after HIV-1 infection, although DNA strand breaks may be present as revealed by TUNEL positivity. There was no correlation between levels of HIV-1 replication and T-cell survival or death. The mechanism of T-cell death after HIV-1 infection requires further study, and investigators who add compounds dissolved in DMSO must include controls to carefully examine the direct effects of even trace levels of this solvent on HIV-1 replication.

(Arch Pathol Lab Med. 2000;124:240-245)

The regulation of human immunodeficiency virus 1 (HIV-1) replication in CD4+ T-cells is complex and multifactorial.1 Moreover, the mechanisms of CD4+ T-cell death during HIV-1 infection are also incompletely understood2 and the relationship between HIV-1 replication and apoptosis is unclear. Two reports document enhanced HIV-1 replication in T-cells when apoptosis is inhibited,3,4 but another report did not find statistically significant increases in HfV-1 replication when T-cell death was inhibited.5 Furthermore, there are also conflicting reports of the molecular mechanism underlying HIV-1-induced apoptosis. Focusing on a central converging mediator of apoptosis triggered by multiple stimuli, the interleukin 1(beta)-converting enzyme protease substrate that becomes cleaved (ie, poly-(ADP-ribose)-polymerase [PARP]),6 one group reported PARP cleavage induced by HIV-1,(4) whereas another group failed to detect PARP cleavage.5 One possible explanation for these discrepancies may reflect different experimental approaches, including the use of pharmacological agents that have been solubilized in the polar solvent dimethyl sulfoxide (DMSO).7

Not only are many nonpolar drugs mixed with DMSO for experimental and therapeutic drug delivery of aqueous solutions, but also DMSO is generally used in the cryopreservtion of numerous cell types and cell lines. Since DMSO itself may influence both HIV-1 replication and apoptosis,8-10 we sought to examine the direct effects of DMSO and several pharmacological agents mixed with DMSO on HIV-1 replication and its induction of apoptosis in a CD4+ T-cell line. To perform these studies, we used DMSO itself, as well as several other drugs previously examined in the context of HIV-1 replication and apoptosisrelated research, ie, cyclosporin A, all-trans retinoic acid, and z-VAD-fmk.4,11-14 Cyclosporin A, all-trans retinoic acid were selected for this study because we were also interested in exploring whether these 2 agents, which are used by dermatologists to treat several skin disorders in patients without acquired immunodeficiency syndrome (AIDS) such as psoriasis, could directly influence HIV-1 replication and/or apoptosis of CD4+ T-cells in individuals with AIDS.

In this report, we document that even very low levels of DMSO enhance HIV-1 replication in the T-cell line, CEM, with slight enhancement by low levels of cyclosporin A but not all-trans retinoic acid or z-VAD-fmk. At higher doses, cyclosporin A inhibits HIV-1 replication. We also observed induction of TUNEL (TdT-mediated dUTPbiotin nick end-labeling) positivity (which detects 3'-OH DNA strand breaks seen early in induction of apoptosis15-18) in nuclei of HIV-1 T-cells but no DNA laddering. Thus, it appears that DNA fragmentation is not a feature of the T-cell death pathway stimulated by FHV-1.(19) Even though z-VAD-fink strongly blocked induction of cell death by HIV-1, there was no correlation between the extent of cell death and HIV-1 replication under these experimental conditions, since DMSO enhanced HIV-1 replication but did not prevent induction of TUNEL-positive T-cells after HIV-1 infection. Attempts to define the molecular mechanism by which HIV-1-induced cell death of CEM T-cells do not support a role for PARP cleavage. Taken together, these results indicate that careful consideration of possible solvent effects on HIV-1 replication need to be monitored, the mechanism of cell death induced by HIV-1 requires further evaluation, and inhibition of apoptosis does not necessarily lead to enhanced HIV-1 replication in this T-cell line.

MATERIALS AND METHODS

Cell Culture and Viral infection

CEM cells (kindly provided by Drs Clive Woffendin and Gary Nabel, University of Michigan) were maintained in RPMI-1640 containing 10% heat-inactivated fetal bovine serum (FBS), 2 mmol/L glutamine, 100 U/mL of penicillin, 100 (mu)g/mL of streptomycin sulfate, and 50 Rg/mL of gentamycin sulfate. A hightiter stock of HIV^sub IIIB^ (kindly provided by Drs Clive Woffendin and Gary Nabel) was prepared by passaging HIV^sub IIIB^ on CEM cells.

The median tissue culture infectious dose (TCID^sub 50^) titer (10^sup 6^/ mL) was calculated as terminal dilution-producing syncytia. CEM cells were incubated with 0.01 multiplicity of infection of HIV^sub IIIB^ for 4 hours at 37 deg C. After the incubation, cells were washed with RPMI-1640 and resuspended at 3 x 10^sub 5^ cells/mL in RPMI-1640 supplemented with 10% FBS in the presence or absence of different drugs. Virus production was monitored every third day by measuring the culture supernatant reverse transcriptase (RT) activity as previously described.20

DNA Isolation and Electrophoresis

DNA was isolated from cell samples as previously described.19 Briefly, 5 X 10^sup 6^ cells were pelleted and resuspended in 100 (mu)L of cell lysis buffer (10 mmol/L EDTA, 50 mmol/L Tris [pH 8.0] containing 0.5% [wt/voll N-lauroylsarcosine and I mg/mL proteinase K) and then incubated for 1 hour at 50 deg C. RNase A was added to each sample at a final concentration of 0.2 mg/mL, and the samples were incubated for another hour at 50 deg C. Cell lysates were diluted to 250 (mu)L by the addition of TE buffer, extracted twice with a phenol-chloroform-isoamyl alcohol (24:24:1) mixture, and then precipitated overnight at -20 deg C by addition of 2.5 volumes of ethanol. DNA was subjected to electrophoresis in 1.5% agarose gel and visualized by ethidium bromide staining.

TUNEL Assay

An in situ apoptosis peroxidase detection kit, based on the TUNEL assay, was used (Apo Tag kit, Oncor, Gaithersburg, Md). Cytospins of T-cells cultured in the presence or absence of DMSO, HIV-1, and/or z-VAD-fmk were prepared and the cells fixed and stained according to the manufacturer's instructions. Nonrandom oligonucleosomal DNA fragmentation, which is associated with apoptosis, creates 3'-OH DNA ends that can be labeled with dUTP.5 Previous studies have shown apoptotic cells labeled with dUTP, whereas necrotic cells and DNA damaged by irradiation are not labeled.17.18

Western Blot Analysis

Protein extracts from cell cultures were prepared as previously described.21 A total of 50 (mu)g of each protein was electrophoresed under reducing conditions through a 12.5% sodium dodecyl sulfate-polyacrylamide gel and electroblotted to nitrocellulose membrane (Schleicher & Schuell, Keene, NH). PARP was detected using a mouse monoclonal antibody (Pharmingen, San Diego, Calif) followed by goat anti-mouse immunoglobulin C-peroxidase (Jackson Immunoresearch, West Grove, Pa). Protein expression was detected by a chemiluminescent detection system (ECL Kit, Amersham, Arlington Heights, 111). Rainbow molecular weight markers (Amersham) were used to estimate the size of the immunoreactive bands.

RESULTS

Initially, the direct effects of DMSO on HIV-1 replication were explored over a wide range of concentrations, including 0.002%, 0.02%, 0.1%, and 0.2%, in a total of 8 different experiments. Figure 1, A demonstrates a representative experiment in which very low levels of DMSO (ie, 0.002%) enhanced HIV-1 replication (although this difference was not statistically significant in all experiments) on days 6 and 9 after infection compared with the untreated control T-cells infected with HIV-1. Note that on day 3 there was an approximately equivalent increase in HIV-1 replication under all the experimental conditions. At higher doses of DMSO, such as 0.1%, preincubating the T-cells did not significantly enhance HIV-1 replication at day 6 but did enhance it at day 9 compared with reactions devoid of DMSO. Addition of the DMSO at the same time as the HIV-1 did not influence the subsequent RT levels compared with T-cells that received DMSO 2 hours before HIV-1 addition (data not shown). In several experiments, concentrations of DMSO of 0.02% and 0.2% produced apparently similar results as shown for DMSO levels of 0.002% and 0.1% (Figure 1, A).

Having established the importance of solvent effects (ie, DMSO) in this system, the influence of cyclosporin A on HIV-1 replication was investigated. Cyclosporin A is typically dissolved in DMSO,22 and for each concentration of cyclosporin A used, separate reactions with equivalent levels of DMSO without cyclosporin A were included. Figure 1, B depicts a representative experiment in which it can be seen on day 6 that cyclosporin A at a concentration of 0.01 (mu)g / mL enhanced HIV-1 replication above the level seen with an equivalent level of DMSO (0.002%), whereas little effect of cyclosporin A was observed at 0.1 and 0.5 (mu)g / mL. However, cyclosporin A at 1 (mu)g / mL had a profound inhibitory effect on HIV-1 replication even though the equivalent level of DMSO in which it was dissolved (0.2%) significantly enhanced HIV-1 replication. In contrast, addition of 0.01, 0.1, 0.5, or 1 @Lg/mL of all-trans retinoic acid in DMSO (0.01%) did not induce any significant increase in HIV-1 replication under similar conditions (data not shown).

Further studies were performed to determine if addition of another DMSO-soluble compound that can influence apoptosis (ie, z-VAD-fmk) modulates HIV-1 replication. Figure 1, C portrays a representative result for 4 different experiments in which 100 (mu)mol/L z-VAD-fmk was added in 0.1% DMSO and RT activity measured. Compared with untreated control cells or T-cells exposed to DMSO alone, the z-VAD-fmk enhanced HIV-1 replication on day 3 (although this was not statistically significant when examining all 4 experiments) but did not increase HIV replication on day 9.

To directly assess the influence of the solvent DMSO, cyclosporin A, and z-VAD-fmk on the cytopathic effects of HIV-1 on T-cells, several studies were performed. After 5 days in culture, T-cells maintained either in RPMI-10% FBS with or without DMSO (0.1%) but without exposure to HIV-1 have a 3% to 9% TUNEL-positive population (Figure 2). Note the presence of individual viable cells with only occasional cells having a TUNEL-positive nucleus. In contrast, 5 days after exposure to HIV-1 there are numerous small and large syncytia forming, and 54% of the T-cells appear to be TUNEL positive. The presence of DMSO (0.1%) did not significantly influence the number or size of syncytia or the percentage of TUNEL-positive cells (46%). In contrast, addition of z-VAD-fmk did not influence the number or size of syncytia but dramatically reduced the number of TUNEL-positive T-cells after HIV1 infection to below 5%. Thus, z-VAD-fmk but not DMSO alone had the potential to reduce the number of TUNELpositive T-cells following HIV-1 infection. Because we have previously observed a discrepancy between the number of TUNEL-positive cells and cells actually undergoing apoptoSiS,23 DNA fragmentation was assessed by agarose gels. Since TUNEL-positive nuclei only reflect the presence of DNA strand breaks, complete engagement of the apoptotic program results in complete DNA fragmentation, producing a defined ladder like configuration. Figure 3 demonstrates that under these experimental conditions no DNA fragmentation is apparent in the T-cells 3 or 5 days following HIVA infection. A positive control is included (Jurkat cells treated with anti-CD95 antibody) that shows DNA laddering under our experimental conditions.

Given the discrepancy between the TUNEL-positive staining of T-cells after HIV-1 infection and the lack of DNA fragmentation, we sought additional evidence for engagement of the apoptotic program focusing on PARR Figure 4 reveals the absence of PARP cleavage in T-cells either 3 or 5 days following infection with HIV-1. In addition, the presence of either DMSO (0.2%) or cyclosporin A (1 (mu)g/ml,) did not lead to detectable PARP cleavage, even though the jurkat cells treated with anti-CD95 antibody underwent PARP cleavage.

COMMENT

Many AIDS patients present to dermatologists with various infections and immunological disorders such as psoriasis. In patients not infected with HIV-1, certain immunomodulatory drugs, such as cyclosporin A or all-trans retinoic acid, may be prescribed, but little is known regarding the influence of such drugs on HIV-1 infection and apoptosis. This study was designed to begin to determine if drugs such as cyclosporin A (as well as the solvent it may be dissolved in-DMSO) influence the ability of HIV-1 to infect and/or kill T-cells. In this study, relatively low levels of DMSO (approximately 0.002%) were shown to enhance HIV-1 replication. Furthermore, cyclosporin A at low levels (0.01 (mu)g/mL) also enhanced HIV-1 replication, whereas at higher levels (1 (mu)g /mL) cyclosporin A strongly inhibited HIV-1 replication. Careful studies of HIV-1-negative patients treated with therapeutic doses of cyclosporin A have revealed that the tissue level of cyclosporin A in the skin is apparently approximately 1 (mu)g/ mL.24 Thus, it is possible that the exact tissue level in the dermis or epidermis of the skin containing HIV-1-infected T-cells and dendritic cells may be differently influenced by cyclosporin A (ie, at very low levels of cyclosporin A HIV-1 replication may be enhanced, whereas at high tissue levels HIV-1 replication may be suppressed). In any event, there appears to be a solvent effect with DMSO, and our results confirm an earlier report9 and highlight the importance of performing carefully controlled experiments that include both the drug of interest and the solvent it may be dissolved in.

The current results indicate that, in our system, HIV-1 infection of T-cells does not induce DNA fragmentation typically seen in cells undergoing apoptosis.19 This is not the first time we have noted a discrepancy between cells that are TUNEL positive but do not contain DNA fragmentation.23 In psoriasis, numerous epidermal keratinocytes are TUNEL positive, but there is no DNA fragmentation.23 Caution is therefore warranted in evaluating TUNEL-positive nuclei to the complex process of apoptosis. Furthermore, the compound z-VAD-fmk, which is a potent inhibitor of apoptosis, did not influence HIV-1 replication but did reduce the number of TUNEL-positive T-cells after HIV-1 infection. Since we could not detect either DNA fragmentation or PARP cleavage, the killing mechanism involved in HIV-1-mediated T-cell death remains unanswered.

Comparing our results to previously published reports, several similarities and differences are apparent. The resuits in this report regarding the lack of DNA fragmentation and PARP cleavage in T-cells infected with HIV-1 are in agreement with Kolesnitchenko et al.5 However, although we agree with the report by Chinnaiyan et al4 that z-VAD-fmk can reduce the number of TUNEL-positive Tcells following HIV-1 infection, the marked increase (greater than threefold) in HIV-1 replication induced by zVAD-fmk they reported was not seen in any of our exper:ents. It is possible that some of the increase noted in this earlier report was due to the presence of DMSO, although specific levels of this solvent were not mentioned, and we have never seen such dramatic effects of DMSO on HIV replication in our laboratory.

In conclusion, it is clear that DMSO, a solvent commonly used to dilute various drugs and cryopreserve cells, has direct effects on HIV-1 replication. There was no consistent correlation between regulation of cell death and the overall level of HIV-1 replication, and the mechanism by which HIV-1 induces T-cell death remains to be defined.

This study was supported in part by National Institutes of Health grant RO1-AR43962 (B.J.N.).

We thank Dr Clive Woffendin for providing valuable assistance during the early phase of this project.

References

1. Fauci AS. Multifactorial nature of human immunodeficiency virus disease: implications for therapy. Science. 1993;262:1011-101 S.

2. Ameisen JC. From cell activation to cell depletion: the programmed cell death hypothesis of AIDS pathogenesis. Adv Exp Med Biol. 1995;374:139-163.

3. Antoni BA, Sabbatini P, Rabson AB, White E. Inhibition of apoptosis in human immunodeficiency virus-infected cells enhances virus production and facilitates persistent infection.] Virol. 1995;69:2384-2392.

4. Chinnaiyan AM, Woffendin C, Dixit VM, Nabel GJ. The inhibition of proapoptotic ICE-like proteases enhances HIV replication. Nat Med. 1997;3:333337.

5. Kolesnitchenko V, King L, Riva A, Tani Y, Korsmeyer SJ, Cohen DI. A major human immunodeficiency virus type 1-initiated killing pathway distinct from apoptosis. I Virol. 1997;71:9753-9763.

6. Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science. 1995;267:1456-1462.

7. van Rijswijk MH, Ruinen L, Donker AJ, cle Blecourt JJ, Mandema E. Dimethyl sulfoxide in the treatment of AA amyloidosis. Ann N Y Acad Sci. 1983; 411:67-83.

8. Trubiani 0, Ciancarelli M, Rapino M, Di Primio R. Dimethyl sulfoxide induces programmed cell death and reversible Gl arrest in the cell cycle of human lymphoid pre-T cell line. Immunol Lett. 1996;50:51-57.

9. Seki 1, Ikeda R, Hoshino H. Dimethyl sulfoxide and related polar compounds enhance infection of human T cells with HIV-1 in vitro. Biochem Biophys Res Commun. 1996;227:724-729.

10. Lin CK, Kalunta Cl, Chen FS, Nguyen TT, Kaptein JS, Lad PM. Dimethyl sulfoxide suppresses apoptosis in Burkitt's lymphoma cells. Exp Cell Res. 1995; 216:403-410.

11. Franke EK, Yuan HE, Luban J. Specific incorporation of cyclophilin A into HIV-1 virions. Nature. 1994;372:359-362.

12. Thali M, Bukovsky A, Kondo E, et al. Functional association of cyclophilin A with HIV-1 virions. Nature. 1994;372:363-365.

13. Thomson AW, Bonham CA. Inhibition of T lymphocyte activation and apoptotic cell death by cyclosporin A and tacrolimus (FK506). its relevance to therapy of HIV infection. Adv Exp Med Biol. 1995;374:211-216.

14. Yang Y, Bailey J, Vacchio MS, Yarchoan R, Ashwell JD. Retinoic acid inhibition of ex vivo human immunodeficiency virus-associated apoptosis of peripheral blood cells. Proc Natl Acad Sci U S A. 1995;92:3051-3055.

15. Compton MM. A biochemical hallmark of apoptosis: internucleosomal degradation of the genome. Cancer Metastasis Rev. 1992;11:105-119.

16. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. j Cell Biol. 1992; 119:493-501.

17. Gorczyca W, Bigman K, Mittelman A, et al. Induction of DNA strand breaks associated with apoptosis during treatment of leukemias. Leukemia. 1993; 7:659-670.

18. Darzynkiewicz Z, Li X, Gong J. Assays of cell viability: discrimination of cells dying by apoptosis, Methods Cell Biol. 1994;41:15-38.

19. Martin Sj, Matear PM, Vyakarnam A. HIV-1 infection of human CD4+ Tcells in vitro. Differential induction of apoptosis in these cells. J. Immunol. 1994; 152:330-342.

20. Dayton Al, Sodroski ), Rosen CA, Goh WC, Haseltine WA. The trans-activator gene of the human T cell lymphotropic virus type III is required for replication. Cell. 1986;44:941-957.

21. Liles WC, Ledbetter )A, Waltersdorph AW, Klebanoff Sj. Cross-linking of CD45 enhances activation of the respiratory burst in response to specific stimuli in human phagocytes. I ImmunoL 1995; 155:2175-2184.

22. Nickoloff Bj, Fisher GJ, Mitra RS, Voorhees jj. Additive and synergistic antiproliferative effects of cyclosporin A and gamma interferon on cultured human keratinocytes. Am j Pathol. 1988; 131:12-18.

23. Wrone-Smith T, Mitra RS, Thompson CB, Jasty R. Castle VP, Nickoloff Bj. Keratinocytes derived from psoriatic plaques are resistant to apoptosis compared with normal skin. Am I Pathol. 1997;151:1321-1329.

24. Fisher Gj, Duel] EA, Nickoloff Bj, et al. Levels of cyclosporin in epidermis of treated psoriasis patients differentially inhibit growth of keratinocytes cultured in serum free versus serum containing media. j Invest Dermatol. 1988;91:142146.

Brunella Taddeo, PhD; Brian J. Nickoloff, MD, PhD; Kimberly E. Foreman, PhD

Accepted for publication July 20, 1999.

From the Department of Pathology and Skin Cancer Research Laboratories, Cardinal Bernardin Cancer Center, Loyola University Medical Center, Maywood, Ill.

Reprints: Brian J. Nickoloff, MD, PhD, Department of Pathology, Skin Cancer Research Laboratories, Loyola University Oncology Institute, Room 301, 2160 S First Ave, Maywood, IL 60153-5385 (e-mail: BNICKOL@LUC.EDU).

Copyright College of American Pathologists Feb 2000
Provided by ProQuest Information and Learning Company. All rights Reserved