Changes of glutathione metabolism during feline immunodeficiency virus infection.

Abstract

To the Editor: Glutathione (GSH) is the most abundant acid-soluble thiol in biological fluids, where it is involved as a cofactor in several functions, ranging from glycolysis to regulation of the cell cycle (1,2). As far as detoxication of reactive intermediates is concerned, GSH may act as a nucleophile, forming conjugates, or as reductant, being oxidized to glutathione disulfide (GSSG) (3). Most of these reactions, which may protect cells from the harmful effects of oxidative stress, are catalyzed by glutathione S-transferases (GSH-T) and GSH peroxidases (GSH-Px), whereas GSH-reductase (GSH-R) is the main enzyme involved in the reduction of GSSG (3-5). A role for GSH in the progression of HIV-induced pathology has been inferred from the repeated findings of decreased GSH levels in plasma, lung epithelial lining fluid, erythrocytes, lymphocytes, and T-cell subsets in HIV-infected patients (6). In view of the relevant redox buffering capability of GSH, decreased GSH concentration may suggest that an increased oxidative stress could be relevant for the pathogenesis of virus-induced disease as well as immunosuppression because low GSH levels inhibit many T- and B-cell functions, including the effector phase of cytotoxic T-cell responses and the IL-2-dependent cell activities (7). Despite the elevated levels of lipid peroxidation byproducts, such as malondialdehyde, found in HIV-infected patients, the nature as well as the biological meaning of the changes involving the antioxidant systems remain quite controversial because some of the contrasting findings reported so far could depend on technical difficulties to obtain and express comparable data (8-10). To investigate whether changes in the GSH-dependent antioxidant system also may develop during feline immunodeficiency virus (FIV) infection and FIV-induced pathology, both the concentration of GSH and the activity of some enzymes involved in GSH metabolism were assessed and compared in naturally and experimentally infected cats as well as in FIV-infected cultures of the Crandell feline kidney (CrFK) cell line. Naturally infected field cats and their controls were selected (with the owners' consent) from those referred to the Animal Clinic of the University of Pisa and found positive for FIV antibody in serum by a commercial test (Cite-combo FIV-FeLV, Agritech System, Portland, ME, U.S.A.); at sampling, none of the FIV-positive cats had symptoms compatible with overt feline AIDS-related complex. The FIV-seronegative controls were healthy cats referred to the clinic for unrelated problems and matched insofar as possible to the infected cats for breed, age, gender, and weight. None of the animals from either group was positive for feline leukemia virus antigen. Specific pathogen-free (SPF) cats were purchased from Iffa Credo (L'Asbrege, France) and housed individually in climatized animal facility under EC law conditions. Experimental FIV infection was performed when the cats were aged 6 months with the Pisa-M2 isolate; SPF-FIV+ cats were sampled 28 months after infection, when they displayed an evident loss of CD4+ lymphocytes. Uninfected control animals were age- and sex-matched SPF cats maintained under identical conditions but in a separate quarter. Blood specimens were collected, with the cats under slight anesthesia, in tubes containing EDTA from the jugular vein. Peripheral blood mononuclear cells (PBMC) were harvested by density gradient centrifugation on Histopaque (Sigma, St. Louis, MO, U.S.A.); after washing, counting, and CD4+ enumeration by the fluorescence-activated cell sorter (FACS) (11), they were pelleted for biochemical determinations. GSH and GSSG content was measured by highperformance liquid chromatography (HPLC) assay using 2,4-dinitro-fluorobenzene according to Reed et al. (12). The activity of GSH-T, GSH-Px, and GS-R as well as the protein concentration within each sample were measured and expressed as detailed in Pratesi et al. (13). Differences between experimental groups were calculated by either one-way analysis of variance or Student's t test using a p value of 0.05 as significance level. The main results obtained by measuring GSH/GSSG concentration as well as the activity of the relevant enzymes involved in the consumption and recycling of this acid-soluble thiol within PBMC from FIV-positive cats are shown in Table 1. As far as the field cats are concerned, no changes were seen in either GSH or GSSG contents of the PBMC; also, the activities of GSH-T and GSH-R were unchanged. The only difference between controls and FIV-infected animals was a borderline decrease (-39.9%; p = 0.049) of the activity of GS-R; accordingly, the capability of cell to recycle GSSG to GSH should be slightly decreased; on the other hand, the small increase in intracellular GSSG observed in FIV-infected field cats when expressed on a “per cell” basis (1.1 ± 0.6 nmol GSSG/107 cells versus 0.6 ± 0.3 of the controls) did not reach statistical significance. It should be noted that the time of seroconversion of the field cats was unknown; furthermore, the cats were in good health as shown from the absence of clinical symptoms and CD4+ count changes. Because, in addition to the FIV infection, FIV-induced pathology also could play a role in modifying the overall GSH metabolism, all relevant determinations of GSH status were repeated on SPF cats experimentally infected with FIV Pisa-M2 strain from 28 months. At this time, a significant decrease of both CD4+ count and of GSH content of the PBMC was observed without relevant changes of the plasma concentration of GSH (2.6 ± 0.3 nmol/ml in FIV-positive cats versus 2.7 ± 0.5 in controls). The GSSG content of PBMC was unchanged with respect to controls, whereas GSSG plasma concentration was somewhat increased (2.9 ± 0.9 nmol/ml in FIV-positive cats versus 1.9 ± 0.1 in the controls, p < 0.05). Among the glutathione-related enzymes, GSH-T was not affected by FIV infection, whereas both GSH-Px and GSH-R underwent a significant decrease with respect to controls. Because the effects of FIV infection on the GSH-based antioxidant systems seem enhanced from the FIV-induced cell damage, GSH content as well as GSH-related enzyme activity were assayed in FIV permissive CrFK cells at varying times after FIV infection, that is, in the absence as well as in the presence of FIV-induced cytopathology (namely, syncytia). The Pisa-M2 FIV isolate was propagated in CrFK cells adapted to grow in the presence of 0.5% fetal calf serum and highly susceptible to syncytium formation by this virus (14). Biochemical determinations were performed 2, 4, and 6 days after infection. The decrease of cell GSH (10.9 ± 1.7 nmol/mg protein versus 15.5 ± 1.9 of the controls, p < 0.05) and of GSH-Px activity (31 ± 6.1 nmol NADPH oxidized/min/mg protein versus 14.5 ± 3.6 of the controls, p < 0.05) reached its maximum 6 days after infection, when the number of FIV-induced syncitia was very high (14); a decrease of GSH-R was evident from 2 days postinfection (15.1 ± 2.7 nmol NADPH oxidized/min/mg protein versus 29 ± 6.2 of controls, p < 0.05), whereas GSH-T always showed no changes. To summarize the results we obtained, we point out that FIV infection is followed by multiple alterations of GSH metabolism. A major determinant of these changes, which are consistent with an overall decrease in GSH availability within the cell, however, seems to be FIV-induced cytopathology rather than FIV infection as such. Data obtained from FIV-infected cats are consistent with similar findings obtained from HIV-infected patients and suggest the FIV model as being quite suitable for a deeper insight into the mechanism(s) of HIV-induced pathology in regard to oxidant stress-related disturbances. The FIV model lends itself to direct experimental testing of the hypotheses under investigation and could help both to clarify some of the many conflicting results so far obtained in HIV-infected patients and to evaluate the possibility of counteracting disease progression by combining antiviral and antioxidant therapy. Acknowledgment: This work was supported by grants from Ministero della Sanità, Istituto Superiore di Sanità, “Progetto Allestimento Modelli Animali per l'AIDS” and Ministero della Università e della Ricerca Scientifica, Roma, Italy. *Aldo Paolicchi; *Patrizia Tonarelli; *Solange Silva; ‡Patrizia Bandecchi; *†Gino Malvaldi *Department of Biomedicine; †Retrovirus Center; ‡Department of Animal Pathology; University of Pisa; Pisa, Italy