| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Webb-Waring Institute for Cancer, Aging, and Antioxidant Research, University of Colorado Health Sciences Center, Denver, Colorado 80262
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Thiol compounds acting as redox modulators may affect the proliferation rates of actively growing cells, such as cancer and transformed cells. The Jurkat T-cell leukemia cell line grows in culture without stimulation, as cancerous and transformed cells do. The redox status of transformed cells has been correlated with their content of antioxidants. Oberley and Oberley (2) and Sun (3) have found low levels of the mitochondrial manganese SOD23 in nearly all of the cancer cell lines studied, as well as in the primary malignant tissues. Furthermore, in childhood acute lymphoblastic leukemia, SOD levels have been reported to be about one-half those of normal cells (4 , 5) . These deficiencies are presumably associated with an increase in the oxidative status of the cell, but the mechanisms that account for the regulation of these important antioxidants enzymes in mammalian cells are still unknown.
Mutations in SODs have been associated with human diseases. Seventy-one different mutations in the cytoplasmic SOD1 have been correlated with amyotrophic lateral sclerosis (6)
. A specific mutation (R213G) in extracellular SOD3 that decreases binding affinity for extracellular surfaces is found in 
7% of the population at large, but in 16% of patients with renal failure (7)
.
The roles of thiols such as NAC have been controversial with regard to cellular responses such as proliferation and apoptosis. NAC and GSH, presumably acting as antioxidants, have been used to rescue T-lymphocytes from apoptosis (8) . However, later studies demonstrated that some thiol compounds could paradoxically elicit apoptosis by acting as pro-oxidants (9) . To understand how changes in the redox status of cells affect their proliferative rates, we conducted experiments using the human T-ell leukemia cell line Jurkat. The cells were subjected to a shift in their redox status by treatment with thiol agents such as MPG, GSH, and NAC. We hypothesized that the redox status of lymphocytes or Jurkat cells would determine whether they proliferate or die by apoptosis or necrosis. Thus, a thiol-induced shift toward a more reducing environment in proliferating lymphocytes might be expected to stop growth and induce quiescence. In fact, the response of Jurkat cells to increasing concentrations of thiols proved to be far more complex.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines and Cell Cultures.
The human T-cell leukemia cell line Jurkat (clone E6–1) was purchased from the American Type Culture Collection. Jurkat cells were grown and maintained in RPMI 1640 supplemented with 10% heat-inactivated FBS. The cells were cultured at 37°C under 5% CO2 atmosphere and 100% humidity and subcultured every 4 days. Jurkat cells (7.5 x 104) were placed into 24-ell plates to a final volume of 1 ml/well under various experimental conditions. Growth was measured by direct counting with a hemocytometer. Viability was assessed by trypan blue exclusion.
HPBL Isolation.
HPBLs were purified using Ficoll gradients (Accu-Prep, Accurate Chemical, and Scientific Corp.). The blood was centrifuged for 20 min at 2200 rpm and the leukocyte "buffy coat" layer resuspended in 30 ml of Hank’s buffer. The cell suspension was underlaid carefully with 20 ml of Ficoll-Hypaque and centrifuged for 30 min at 1750 rpm. The layer containing the HPBLs at the interphase of the gradient was recovered and resuspended in Hank’s buffer to 40 ml. The cell suspension was centrifuged for 10 min at 800 rpm and the pellet washed twice with Hank’s buffer, then resuspended in 20 ml of Hank’s buffer and underlaid with 10 ml of FBS (non-heat-inactivated) containing 3 mM EDTA. After 20 min of centrifugation, the supernatant was decanted and the platelet-free HPBL pellet resuspended in 50 ml of OPTI-MEM (Life Technologies, Inc.) containing 5% autologous serum (reserved from the whole-blood sample) and 1% L-glutamine/penicillin/streptomycin solution (Sigma Co.) The cells were adjusted to 5 x 105/ml and transferred to T-150 flasks. The cell suspensions were incubated at 37°C in 5% CO2, 95% air and 100% humidity for 2 h and transferred to new T-150 flasks and incubated under the same conditions for an additional 4 h. If red cell contamination was evident in the HPBL pellet, 9 ml of sterile distilled H2O were added and resuspended, passing through a sterile transfer for 20 s. One ml of 10x Hank’s buffer was added and the final volume adjusted to 20 ml with Hank’s buffer. The cell suspension was centrifuged for 10 min at 800 rpm, and the final red cell-free pellet was resuspended in 50 ml of OPTI-MEM with 4% FBS (heat-inactivated) and 1% L-glutamine/penicillin/streptomycin solution. The final lymphocyte population was more than 95% pure as judged by light microscopy.
Thiol Treatment.
Stock solutions of GSH, MPG, and NAC (100 µM) were freshly prepared in RPMI before the start of the experiments, and 7.5 x 104 cells were grown in the presence of increasing thiol concentrations in a volume of 1 ml/well. Growth and viability were measured as described above.
Apoptosis and Necrosis.
Apoptosis and necrosis were determined after 24 and 48 h of incubation by morphological assessment with acridine orange-ethidium bromide staining and fluorescence microscopy (11)
. Cells were centrifuged in a volume of 100 µl for 15 s at 12,000 rpm. The supernatant was aspirated and the pellet resuspended in 20 µl of PBS and 2 µl of a mixture containing 100 µg/ml each of acridine orange and ethidium bromide. Aliquots of 10 µl of stained cells were observed with fluorescence microscopy, and the number of apoptotic and necrotic cells were determined by counting 100 cells in each of five different, randomly selected fields.
SOD and Catalase Activity Assays.
After a 24-h incubation in the presence of various MPG concentrations, 107 cells were washed once, resuspended in chilled 10 mM phosphate buffer (pH 7.0) containing 0.15 M NaCl, and sonicated for 30 s on ice. Aliquots were used for measuring enzyme activities. SOD activity was measured using the cytochrome c reduction assay as described by McCord and Fridovich (12)
. Catalase activity was evaluated according to Bergmeyer (13)
, using a freshly prepared stock solution of hydrogen peroxide. The protein content of crude extracts was measured using the method described by Lowry et al. (14)
SOD Activity Gels.
Aliquots of the cell lysates of 3 µl or less containing 0.1 unit of total SOD activity were loaded on agarose gels (Universal Gel/8; Helena Laboratories, Beaumont, TX) and electrophoresed in 20 mM Tris-glycine buffer (pH 8.2) at 200 mA for 15 min. The gel was stained for SOD activity with nitroblue tetrazolium-riboflavin-EDTA and then developed under fluorescent light as described by McCord (15)
.
SSCP.
Genomic DNA was extracted from Jurkat cells by phenol/chloroform and ethanol precipitation. SSCP analysis was performed on genomic DNA samples as described in the FMC BioProducts (Rockland, ME) protocol by using primers for a portion of exon 3 of the sod2 gene and the Mutation Detection Enhanced gel system. The primer sequences were: WV5 5' (5'CAGTGGTTGAAAAAGTAGGAG3') and WV31 3' (5'TTAGGGCTGAGGTTTGTCCA3'). This amplification yielded a 148-bp PCR product.
RT-PCR and Sequencing of Jurkat sod2 cDNA.
Total RNA was purified from cell extracts using the total RNA purification system (Promega, Madison, WI). Samples of total RNA were reverse transcribed and amplified by PCR using the SF3 5' (5'-CCCCTGCAGATGTTGAGCCGGGCAGTGTGCGGCACCAGC-3', and the SF4 3' (5'-CCCTCTAGATTACTTTTTGCAAGCCATGTATCTTTCAGT-3') primers with the Access RT-PCR system (Promega) as described by the manufacturer. The PCR products were purified and cloned into the TA cloning kit from Invitrogen Life Technologies (Carlsbad, CA). Colonies were screened by PCR, followed by restriction analysis with HaeII and MboII. Clones with inserts of the right size, orientation, and restriction sites were sequenced with Ampli Taq DNA polymerase FS with the dRhodamine terminator cycle sequencing-ready reaction kit at the DNA sequencing and analysis core facility of the University of Colorado Cancer Center. Results were analyzed with PCGENE software.
Statistical Analyses.
Statistical significance was determined by one-way ANOVA using the Neuman-Keuls test by StatMost software for PC.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. After 24 h, low MPG concentrations ranging from 0.03 mM to 1.0 mM inhibited proliferation. Higher MPG concentrations of 3.0 mM and 10.0 mM, however, produced no inhibitory effects. We also quantified the percentage of apoptotic and necrotic cells under each condition, as well as cell viability (Fig. 1b)
. Surprisingly, we found that only the intermediate concentration of MPG (0.1 mM) induced the highest percentages of apoptosis (40%). In cultures incubated with 0.3 mM and 1.0 mM MPG, the cells died mainly by necrosis. Cell viability assessed by trypan blue exclusion was minimal at concentrations between 0.1 and 1.0 mM, essentially mirroring the necrosis. The highest concentrations of 3.0 mM and 10.0 mM MPG produced no observable adverse effects; by all criteria, these cells resembled control cells (Fig. 1b)
. Thus, the growth rate of Jurkat cells is affected by increasing concentrations of MPG in a complex and multiphasic manner.
|
shows the proliferation of Jurkat cells incubated with increasing concentrations of NAC. Cells incubated with 0.03 mM NAC showed no effects, but concentrations ranging from 0.1 mM to 10 mM inhibited the proliferation at 24 h. Cells incubated with concentrations from 0.3 mM to 3 mM of NAC showed the highest percentages of apoptosis (Fig. 1d)
. The percentage of necrotic cells was high only in the cells incubated with 1.0 mM NAC, again reflected as low viability assessed by trypan blue exclusion.
When GSH concentration was very low (0.03 mM or 0.1 mM) or very high (3.0 mM or 10.0 mM), there were no differences in growth compared with control (Fig. 1e)
. Cells incubated with intermediate concentrations of 0.3 mM and 1.0 mM, however, showed diminished viability and growth. The percentage of apoptotic cells was highest at a GSH concentration of 1.0 mM (Fig. 1f)
. The percentage of necrosis was highest in cells incubated with 0.3 mM and 1.0 mM GSH, correlating inversely with the percentage of viable cells (Fig. 1f)
.
We then evaluated the effects of MPG on the specific activity of SOD, in intact cells and in cell-free extracts. Fig. 2a
shows that SOD specific activity decreased about 75% compared with control, as MPG in the culture medium was increased to 0.1, 1, and 3 mM. Also, extracts from control cultures grown in the absence of MPG were treated in vitro with various concentrations of MPG. A dose-dependent loss of SOD2 activity was observed (Fig. 2b)
.
|
shows that the major band of SOD1 is not different from the control (HPBLs), although an additional weaker, faster moving band was visible in the Jurkat extract. The SOD2 of Jurkat cells, however, differs from that of normal lymphocytes quantitatively as well as qualitatively. The SOD2 activity is lower in Jurkat cells, as described previously for transformed and cancerous cells. In addition, the electrophoretic mobility is slightly faster than the SOD2 of normal HPBLs, suggesting the possibility of a mutated or differently processed SOD2 in Jurkat cells.
|
, strongly suggesting a mutation in the gene.
The Jurkat sod2 cDNA was cloned and sequenced. Five of six independent clones (designated Jurkat Allele 1, Fig. 4
) contained the C1183T substitution (which encodes the A-9V variant) in exon 2, previously reported by Heckl (16)
. These five clones contained wild-type sequence in the entire region coding the mature SOD2. In one of six independent clones, however, we found a cDNA sequence presumably reflecting the other Jurkat allele (designated Jurkat Allele 2, Fig. 4
). This sequence was wild type at the C1183 locus, but contained a new point mutation, C5782T, in exon 3, that will produce a L60F mutation in the enzyme. Thus, Jurkat T lymphocytes appear to be heterozygous in the sod2 gene at two loci: the previously described -9 locus and at a previously undescribed locus, 60.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ß, c-jun, c-fos; Refs. 18
, 19
). Some of these proteins have been shown to be redox controlled, and their modification by oxidants may activate the cells to proliferate. However, the original premise holds true as well; high concentrations of superoxide can cause cell death (1)
by inducing lipid peroxidation, protein inactivation, and DNA damage at the molecular level, as well as by inducing the phenomena of apoptosis and necrosis.
Jurkat T-cells proliferate in culture without activation, a characteristic common among immortalized cells. The effects of several thiol-reducing agents on growth and proliferation were evaluated, with results that appear to be complex, multiphasic, and paradoxical. The growth of Jurkat cells progressively slows as thiol concentrations increase toward 1 mM (Figs. 1, a, c, and e)
but is dramatically restored to normal at 3 mM or higher. Fig. 1b
sheds more light on what is happening in the growth-inhibitory range of thiol concentrations by revealing that the cells are dying almost entirely by apoptosis at 0.1 mM MPG, but that the mode of death shifts to predominantly necrosis at 1 mM MPG. This progressive behavior of "proliferation to apoptosis to necrosis" is reminiscent of how primary human peripheral lymphocytes behave in the face of increasing rates of superoxide production (1)
. The apparent paradox is that our previous study observed this progressive behavior as cells were exposed to increasing oxidative stress brought about by the redox-cycling agent paraquat, whereas the present study sees the same behavior as cells are exposed to increasing concentrations of a reducing agent, MPG.
We begin to understand the paradox with the data seen in Fig. 2a
. Very low concentrations of MPG in the growth medium markedly decrease the SOD activity in Jurkat cells, and similar concentrations inhibit the enzyme in cell-free extracts (Fig. 2b)
. This inhibition of SOD2 by thiols was unexpected because Matsuda et al. (20)
have reported that the activity of wild-type SOD2 is unaffected by thiols. We have found, however, that a mutant form of the enzyme, I58T, is quite sensitive to inhibition by thiol reagents (21)
, probably through the formation of mixed disulfides. This mutant form of SOD2 has been characterized by Borgstahl et al. (22)
and found to exist largely as a dimeric structure rather than the normal tetramer. In the dimer, each subunit exposes cysteine-140, a residue normally buried in the tetrameric structure. Thus, we wondered whether Jurkat cells may contain the I58T mutant form of SOD2 (or a similar mutation), which might explain the abnormal sensitivity to thiol reagents. If this were true, low concentrations of thiols could paradoxically create a condition of oxidative stress by virtue of SOD2 inhibition, resulting in increased numbers of apoptotic and necrotic cells. At high concentrations of the thiol reagents (3–10 mM), the antioxidant protection provided by the thiols might make up for the loss of SOD2 activity or, at the very least, drive the protein mixed-disulfides back to their reduced states to reactivate the SOD2, decreasing oxidative stress and allowing restoration of normal growth. The formation of protein-mixed-disulfides (or S-thiolation) under conditions of mild to moderate oxidative stress is well documented, and a number of metabolic enzymes are known to be inactivated by such modification (23)
. If a mutant SOD2 were susceptible to inactivation by S-thiolation, the resulting elevation in oxidative stress would only serve to promote additional S-thiolation reactions, leading to a possible vicious cycle within the cell.
An examination of the electrophoretic behavior of Jurkat SOD2 (Fig. 3a)
was consistent with our hypothesis that the SOD2 might be a mutant form. The electrophoretic pattern differed from that seen in normal HPBLs, both quantitatively and qualitatively. There was less SOD2 activity, and its mobility was clearly faster than that of the wild type. This difference in mobility could reflect a coding mutation, or it could be evidence of S-thiolation of the enzyme, or both. Confirmation of a mutation was provided by SSCP analysis (Fig. 3b)
. The pattern presented by an amplicon containing exon 3 (where the I58T mutant is located) is clearly different from wild type. The pattern is consistent with a heterozygous genotype because the wild-type pattern is present, together with a similar but faster moving pattern, presumably contributed by the mutant allele.
Cloning and sequencing of the Jurkat sod2 cDNA (Fig. 4)
clearly revealed that the cells are indeed heterozygous. One allele contained the common alanine-to-valine mutation at residue -9 (in the leader sequence) in exon 2, but was wild type in exon 3. The other allele was wild type at -9, but contained a new mutation, leucine-to-phenylalanine, at position 60 in exon 3. This L60F mutation is only two residues downstream from the I58T mutation previously studied by Borgstahl et al. (22)
The location of the I58T mutation is in a two-helix bundle that extends finger-like from each subunit (Ref. 24
; the protein data bank coordinates were visualized with the software Rasmol). The monomer-monomer interface of human SOD2 is typical of globular protein subunit interactions, involving a substantial area of contact. The dimer-dimer interface, however, is quite atypical, involving a small area of contact between two pairs of these finger-like two helix bundles. These areas of contact involve only about eight residues contributed by each subunit. The eight contact residues include both the isoleucine residue at position 58 and the leucine residue at position 60. Thus, it seems highly likely that the L60F mutation that we report here from Jurkat cells will be functionally similar to the previously studied, tetramer-destabilizing I58T mutation, (22)
, which renders the enzyme thermally unstable and vulnerable to inactivation by S-thiolation reactions (21)
.
The finding of a destabilizing SOD2 mutation in a leukemia-derived cell line may not be coincidental. Nearly all cancerous tissues, including leukemia lymphocytes (4) , have decreased SOD2 (and increased oxidative stress) leading some to regard sod2 as a tumor suppressor gene (25) . Exposure of mitotically competent cells such as fibroblasts (26) or lymphocytes (1) to increased production of superoxide induces the cells to proliferate. Thus, a nearly universal feature of malignantly transformed cells may be either an increase in production of superoxide or a decrease in endogenous SOD levels (27) , with the latter seemingly more common. The decrease in SOD activity could occur via misregulation of the gene (resulting, e.g., from mutations in the promoter region of the gene) or via coding mutations such as L60F that may result in a thermally unstable enzyme or in an enzyme that is susceptible to inhibition by the formation of mixed-disulfides involving sensitive thiol groups exposed by the weakened tetrameric interface. It is interesting to note that SOD1 of chickens and of humans exists as S-thiolated derivatives, reflecting mixed disulfides with GSH (28) . It is also noteworthy that the bizarre behavior toward thiol reagents that we describe here for Jurkat cells is not unique, but has been reported by Morse et al. (9) for CEM cells, another human leukemia-derived cell line. Thiols represent an important class of antioxidants that contribute to cellular homeostasis in many ways. We report here an unusual mutation that renders SOD2, a vital antioxidant enzyme, sensitive to inactivation by these normally protective compounds.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by a grant from the Gustavus and Louise Pfeiffer Foundation, and the Centro de Investigaciones Biologicas del Noroeste, La Paz, BCS, Mexico. 
2 To whom requests for reprints should be addressed, at The University of Colorado Health Sciences Center, Box C-321, 4200 East Ninth Avenue, Denver, CO 80262. Phone: (303) 315-6257; E-mail: joe.mccord{at}uchsc.edu
3 The abbreviations used are: SOD, superoxide dismutase; MPG, N-2-mercaptopropionyl glycine; GSH, glutathione; NAC, N-acetylcysteine; SSCP, single-stranded conformational polymorphism; HPBL, human peripheral blood lymphocyte; FBS, fetal bovine serum.
Received 5/30/02. Accepted 10/30/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B transcription factor and HIV-1. EMBO J., 10: 2247-2258, 1991.[Medline]
b and AP-1 in intact cells—AP-1 as secondary antioxidant-responsive factor. EMBO J., 12: 2005-2015, 1993.[Medline]
This article has been cited by other articles:
|
|
 |
|
  B. Bogunovic, M. Stojakovic, L. Chen, and M. Maric An Unexpected Functional Link between Lysosomal Thiol Reductase and Mitochondrial Manganese Superoxide Dismutase J. Biol. Chem., April 4, 2008; 283(14): 8855 - 8862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. W. Oberley Therapeutic Manipulation of the Intracellular Redox State Am. Assoc. Cancer Res. Educ. Book, April 1, 2006; 2006(1): 255 - 258. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Hermann, Y. Li, M. B. Ray, J. M. Wo, and R. C. G. Martin II Association of Manganese Superoxide Dismutase Expression With Progression of Carcinogenesis in Barrett Esophagus Arch Surg, December 1, 2005; 140(12): 1204 - 1209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. T. Teoh, P. V. Turner, and D. H. Evans Tumorigenic Poxviruses Up-Regulate Intracellular Superoxide To Inhibit Apoptosis and Promote Cell Proliferation J. Virol., May 1, 2005; 79(9): 5799 - 5811. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. St. Clair Manganese Superoxide Dismutase: Genetic Variation and Regulation J. Nutr., November 1, 2004; 134(11): 3190S - 3191S. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
