Published online before print January 8, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Wadsworth Center, New York State Department of Health, Albany, New York, USA; and
Department of Medicine, Division of Rheumatology, University of Massachusetts Medical School, Worcester, Massachusetts, USA
1 Correspondence: Biggs Laboratory, Wadsworth Center, New York State Department of Health, Empire State Plaza, Albany, NY 12201-0509, USA. E-mail: lawrenced{at}wadsworth.org
ABSTRACT
Although the exact etiology of rheumatoid arthritis (RA) remains unknown, there is increasing evidence that reactive oxygen species and a pro-oxidant/antioxidant imbalance are an important part of the pathogenesis of joint tissue injury. Flow cytometry was used to evaluate the thiol status [surface-thiols and intracellular glutathione (iGSH)] of leukocytes from RA patients and controls. Levels of surface-thiols and iGSH of leukocytes from RA patients were significantly lower than of leukocytes from controls. CD53, a glycoprotein of the tetraspanin superfamily, which coprecipitates with the GSH recycling enzyme 
-glutamyl transpeptidase, was elevated significantly on leukocytes from RA patients compared with leukocytes from controls. Surface-thiols and GSH play important roles in redox buffering of cells, providing protection from oxidative stress. The chronic inflammation of RA has been associated with oxidative stress, which is shown to cause a decline in the levels of cellular antioxidant sulfhydryls (R-SH). As antioxidant-protective levels also decline with age, the problem is compounded in older RA patients, who did have fewer R-SH. Chronic stress can also have an effect on telomere lengths, determining cell senescence and longevity. Although telomeres shorten with increasing age, our flow cytometry studies indicate that accelerated shortening in telomere lengths occurs with increasing age of RA patients, suggesting premature cellular aging. The paradox is that lymphocytes from RA patients are believed to resist apoptosis, and we suggest that the elevated expression of CD53, which results from the increased oxidative stress, may protect against apoptosis.
Key Words: glutathione • CD53 • flow cytometry • telomeres
INTRODUCTION
Rheumatoid arthritis (RA) is a chronic, systemic, inflammatory, autoimmune syndrome, which produces degradation of articular cartilage and bone erosion. The long-term outcomes of this progressive disease are significant morbidity, loss of functional capacity, and increased mortality [1 , 2 ]. RA affects 1–2% of the general population worldwide [3 ], and the occurrence in women is three times greater than in men. Although the onset of RA can occur at any age, the incidence increases with age. The exact etiology of RA remains unknown.
The formation and scavenging activity of free radicals in biological systems have been linked closely to a number of pathological conditions. In healthy individuals, reactive oxygen species (ROS) and associated oxidative stresses are kept in check by a combination of antioxidant activities [4 , 5 ]. Human cells have developed a formidable antioxidant defense against oxidant reactions. In particular, they possess enzymatic and nonenzymatic antioxidant molecules, including thiols [mainly glutathione (GSH)], for defense. One key chemical barrier against stress-induced damage is the redox equilibrium of sulfhydryl (SH)/disulfides, by which low molecular weight thiols can be oxidized reversibly to disulfides and/or protein mixed disulfides in response to an oxidative stress [6 7 8 ]. There is increasing evidence that ROS and the resulting pro-oxidant/antioxidant imbalance play a major role in RA, as well as in other disease states [9 10 11 ]. It has been shown that lymphocytes, which are highly sensitive to thiol modification, are impaired in many of their immune functions when exposed to oxidative stress or SH modifiers such as those found in the cellular microenvironment [12 ].
-Glutamyl transpeptidase (GGT) is a cell-surface enzyme in the recycling pathway of GSH, part of the antioxidant defense mechanism. Through GGT cleavage of -glutamyl from GSH, the cysteinylglycine dipeptide is released, thus eventually increasing the supply of cysteine available to the cell [13
]. Studies indicate that the metabolism of extracellular GSH by GGT is also important in T cell signaling and in the activation of transcription factors [14
]. CD53, a glycoprotein of the tetraspanin superfamily, has been reported to coprecipitate with GGT activity [15
]. The functional significance of this association is still not fully understood; the association may be important in regulating local intracellular redox potential by playing a role in the binding or recycling of the end products of the GGT reaction. It has been proposed that tetraspanins act as molecular facilitators, grouping specific cell-surface proteins together, thereby increasing the formation of stable and functioning signaling complexes [16
]. Overexpression of CD53 in stably transfected cells resulted in elevated levels of GSH and reduced levels of peroxides [17
]. Increases in mRNA transcripts and protein expression of CD53 have been associated with increased resistance to H2O2, UVB, -irradiation, and apoptosis [18
]. The ligation of CD53, with an associated reduction in caspase activation, triggers a survival response and reduces the number of cells that enter apoptosis [19
]. It has been known for some time that resistance to apoptotic cell death through low-level, proapoptotic or high-level, antiapoptotic stimuli can initiate and perpetuate autoimmune diseases, including RA [20
].
Several studies have demonstrated links between chronic oxidative stress, as occurs with chronic inflammation, and shortened telomeres, which are determinants of cell senescence and accelerated aging [21 , 22 ]. Telomeres are unique DNA-protein complexes consisting of G-rich hexanucleotide repeats, located at the termini of eukaryotic chromosomes. These complexes play an important role in maintaining chromosomal integrity. In the absence of compensatory mechanisms, they are not fully replicated with each cell division, as a result of limitations of DNA polymerase to complete replication to the ends of a linear chromosome, and they, therefore, shorten with each replication. When the telomeres have shortened sufficiently, the cell is arrested in senescence [23 ]. In a study using normal human endothelial cells, mild, chronic oxidative stress, induced by interference with GSH-dependent, antioxidant defenses, accelerated telomere erosion and the onset of replicative senescence. Aging is closely associated with an oxidative shift in the thiol/disulfide redox state of the intracellular GSH (iGSH) and cysteine pools, with a decline in cellular thiol concentration [24 ].
In light of the "inflamm-aging" model of longevity [25 ], we determined if the chronic inflammation of RA contributes to a faster decline of aging-associated biomarkers. Our analysis of the thiol status (surface-thiols and iGSH) of peripheral blood leukocytes (PBL) from RA patients revealed significant differences compared with the cellular thiols status of leukocytes from a normal, non-RA control group. The decline of cellular thiols of leukocytes from RA patients inversely affected the cell-surface levels of CD53. In addition, the mean length of telomeres (adjusted for age) was shorter in RA patients than in non-RA control subjects, an indication that chronic stresses accelerate cellular aging.
MATERIALS AND METHODS
Specimen collection
EDTA venous blood samples were obtained with informed consent from healthy volunteers and from RA patients who fulfilled the 1987 criteria for RA by the American Rheumatism Association and were in functional Class I, II, or III (Table 1
), according to the revised criteria of the American College of Rheumatology [26
]. Patients with an active disease duration of at least 6 months, as manifest by at least three joints that were swollen and six joints that were tender at the time of the blood donation, were accepted for the study (Table 2
). In addition, RA patients had an erythrocyte sedimentation rate 
28, a CRP 1.4, or morning stiffness of at least 45 min in duration. All standard therapy for RA including nonsteroidal anti-inflammatory drugs, disease-modifying antirheumatic drugs (DMARDs), and combinations of DMARDs were allowed as long as doses were stable for 2 months prior to participation in the study. Specimens were randomized blindly for the different analyses, and not all specimens collected were used for all analyses; no analyses were excluded. The New York State Department of Health Institutional Review Board (IRB; Protocol #04-025), University of Massachusetts Medical School IRB, and the New England IRB approved participation in this study.
|
View this table: [in a new window] |
Table 1. American College of Rheumatology Revised Criteria for Classification of Functional Status in RAa
|
|
View this table: [in a new window] |
Table 2. Baseline Characteristics of the RA Cohort
|
For determination of iGSH levels, PBMC were isolated on a Histopaque-1077 (Sigma Chemical Co.) gradient, washed twice, and resuspended in PBS (Sigma Chemical Co.) at 107cells/ml. Aliquots (100 µl) were stained with one or more of the following mAb: CD3-FITC, CD4-PE, CD8-PE, CD19-PE, CD16/56-PE, and CD14-PE (BD Biosciences). The specimens were incubated 30 min on ice, washed thrice, and resuspended in 100 µl PBS. Cells were fixed with 1% paraformaldehyde (Polyscience, Inc., Warrington, PA, USA) for 10 min on ice, washed once, and resuspended in 5 mM N-ethylmaleimide (NEM; Invitrogen)/Cytofix/Cytoperm solution (BD Biosciences). The cells were incubated overnight at 4°C, washed twice with 1x PermWash solution (BD Biosciences), and resuspended in 100 µl 1x PermWash containing 2 µg AlexaFluor 488-conjugated 8.1-GSH mAb {anti-GSH adduct with NEM (anti-GS-NEM); refs. [27 , 28 ]}. After 30 min incubation on ice, the suspension was washed twice with PermWash and resuspended in 300 µl PBS. The specimens were analyzed on a FACSCalibur flow cytometer using CellQuest software.
For CD53 surface level analysis, whole blood (100 µl) was stained with one or more of the following mAb: CD3-PerCP, CD45-APC, CD4-PE, CD8-PE, CD19-PE, CD56/16-PE, CD14, CD53-FITC, or mouse IgG1
-FITC (BD Biosciences). The specimens were incubated with the mAb 15 min at room temperature, lysed with a 1x solution of FACSLyse (BD Biosciences), and run on a FACSCalibur flow cytometer using CellQuest software.
Telomere length analysis
Previously isolated and frozen PBMC were thawed, and PBMC (1x106) were washed and centrifuged (10 min; 200 g). The cell pellets were resuspended in hybridization buffer containing 70% deionized formamide (Sigma Chemical Co.), 10 mM Tris, pH 7.0, 10% FCS, and 0.3 µg/ml telomere-specific, FITC-conjugated peptide nucleic acid probe (FITC-OO-CCCTAACCCTAACCCTAA-COOH; Applied Biosystems, Foster City, CA, USA). Samples were heat-denatured at 82°C for 10 min, followed by hybridization in the dark for 16 h at room temperature. The cells were washed (PBS, Sigma Chemical Co.), centrifuged (10 min; 200 g) twice, and then resuspended in PBS with 10% FBS, RNase (10 µg/ml; Sigma Chemical Co.), and propidium iodide for 2 h at room temperature. Cells were analyzed on a FACSCalibur (Becton Dickinson, San Jose, CA, USA) flow cytometer. The telomere fluorescence signal is defined as the mean fluorescence signal in G0/G1 cells after subtraction of the background fluorescence signal [fluorescence in situ hybridization (FISH) procedure without probe]; results are expressed in molecular equivalents of soluble fluorochrome units. Human T cell lines (Jurkat and CEM-CCRF) were used as controls for short and long telomeres, respectively. Telomere length was calculated [29
] from the molecules of an equivalent, soluble fluorochrome units standard curve (Quantum Beads, Bangs Labs, Fischers, IN, USA) using the following equation: telomere length (kb) = [Flchannel #–Flchannel # (blank)] x 0.019 x 0.02604/slope.
Statistical analysis
Statistical analysis was performed using Sigma Plot 9.0 software. In all tests, comparisons with associated P values less than 0.05 were considered significant. Age and gender correlations in each study group were quantified by the Pearson correlation test.
RESULTS
RA patients have decreased levels of surface-thiols on PBL
Earlier studies (e.g., ref. [30
]) have shown that there are at least 15 different cell-surface proteins, which may contain free SH groups, depending on the redox status of the cells. To measure the overall level of cell surface R-SH, we took advantage of the readily available AlexaFluor-maleimide reagents. The nucleophilic maleimide covalently couples to an available SH moiety. Maleimide is effective in its ability to react with thiols and is considered to be highly specific for thiol groups. The Alexa dyes that are coupled to maleimide do not interfere with the maleimide-thiol interaction and as they are charged molecules, do not enter the cell, thus enabling a FACS-detectable, overall cell surface-thiol determination.
For assessment of the surface-thiol levels of RA patients, whole blood specimens were treated as described in Materials and Methods, and the measured levels were compared with the levels assessed for a healthy control population (Fig. 1 ). Overall, RA patients had significantly lower levels of surface-thiols on most of their PBMC subpopulations as compared with the control population. The relative quantity of the surface-thiols in both groups followed a pattern seen consistently among the analyses within this study, as well as in our previous surface-thiol study [11 ]. The ordering seen, namely (CD19+) B cells > (CD8+) T cells > (CD4+) T cells, appears to correlate well with the differing sensitivities of the cells to radiation [31 ] and thiol-reactive chemicals [32 ]. In addition, in RA patients, there was a trend toward decreasing overall surface-thiol levels with increasing age (Fig. 2 ). This trend was not seen in the healthy, aging, control population; in fact, surface-thiols increased on CD4+ and CD8+ lymphocytes with age (Fig. 3 ). The decrease with leukocytes from RA patients is consistent with the idea that a decrease in oxidative stress management as a function of age is compounded by chronic inflammatory disease states, as in RA [33 , 34 ]. There was no significant correlation between surface-thiol levels and gender in the patient group or the control group.
 View larger version (19K): [in a new window] |
Figure 1. Surface-thiol levels on leukocyte subsets. Patients with RA (n= 93) have significantly lower levels of surface-thiols on most leukocyte subpopulations than do the non-RA healthy controls (n=26). The rank order by relative quantity of the surface-thiols in both groups was monocytes (Monos) > granulocytes (Grans) > lymphocytes (CD19>CD8>CD4). The levels of thiols on the surface of leukocytes are presented as units of mean fluorescence intensity (MFI). The data are mean ± SD; *, significant difference (P<0.001) from the appropriate control subset.
|
 View larger version (42K): [in a new window] |
Figure 2. Correlation of surface-thiols and age on leukocyte subsets from RA patients. The surface-thiols of CD4+ lymphocytes, NK cells, and monocytes declined significantly with age.
|
 View larger version (21K): [in a new window] |
Figure 3. Correlation of surface-thiols and age on leukocyte subsets from healthy, control donors. Unlike the CD4+ lymphocyte population from RA patients, this subset showed a significant increase of surface-thiols with age. The surface-thiols of CD8+ lymphocytes also increase with age, whereas all of the other subsets had no significant changes with age.
|
 View larger version (22K): [in a new window] |
Figure 4. Levels of iGSH in leukocyte subsets. Patients with RA (n=44) have significantly lower mean levels of iGSH, as measured by AlexaFluor647-8.1GSH (anti-GS-NEM) intracellular staining, after fixation and NEM treatment, than do the normal, healthy controls (n=17). Levels are expressed as the MFI of the AlexaFluor-647. In terms of relative quantity of iGSH, the pattern seen is monocytes > CD8 > CD4 > CD19; this pattern differs from that seen for surface-thiol levels. The data are mean ± SD; *, significant difference (P<0.05) from the appropriate control subset.
|
 View larger version (27K): [in a new window] |
Figure 5. Patients with RA show a trend toward decreasing iGSH levels with increasing age, although the trend reached significance only for monocytes and B cells. This trend was not seen in a healthy, aging population; there was no significant positive or negative correlation of iGSH and age in the control population (data not shown).
|
 View larger version (19K): [in a new window] |
Figure 6. CD53 expression by leukocyte subsets. Patients with RA (n=26) had significantly higher mean levels of CD53 on the cell surface of most leukocyte populations than did the healthy controls (n=24). The CD53 levels are expressed as MFI. It is interesting that the relative expression of CD53 follows the same pattern as for surface-thiols (monocytes>CD19+>CD8+>CD4+). The data are mean ± SD; *, significant difference (P<0.05) from the appropriate control subset.
|
 View larger version (23K): [in a new window] |
Figure 7. Teleomere length analyses. PBMC from RA patients have significantly shorter telomeres than do PBMC from the non-RA controls (A). Two cell lines (Jurkat and CEM-CCRF) were used as internal controls for short and long telomere controls. The telomere fluorescence signal is defined as the mean fluorescence signal in G0/G1 cells after subtraction of the background fluorescence signal (FISH procedure without probe); results are expressed in molecular equivalents of soluble fluorochrome units. Telomere length was calculated [29
] as described in Materials and Methods. The data are mean ± SD; *, significant difference from the controls (P=0.003). When adjusted for age, the mean length of telomeres of PBMC is shorter in RA patients than in the non-RA controls, and the difference is more pronounced after the second decade (B).
|
The need for an intact redox balance in cells has led to the evolution of several effective, intracellular, antioxidant defense systems, which can sense elevated levels of oxidative stress and swiftly reinstate a healthy redox environment [4 , 5 ]. The response of a cell to stress often involves changes in cell thiol content. Thiols are first consumed in reactions that protect the cell by removing deleterious compounds; the thiols are then replaced through enzymatic reduction of disulfides or de novo synthesis. It has been suggested that the pro-oxidant/antioxidant imbalance seen in cells from patients with RA, resulting from accumulation of ROS, is a result of acceleration of some cellular reaction or an impaired antioxidant defense system [9 , 10 ].
In this study, we evaluated the thiol status (surface-thiols and iGSH) of PBL from patients with RA. We found consistently that RA patients had significantly lower levels of surface-thiols and iGSH than did the control group. During the maintenance of a balanced redox environment, an initial mild stress causes an elevation in thiols and iGSH, which provide protection against more severe forms of stress. Cellular GSH levels are maintained by GSH reductase activity, which converts GSSG back to GSH, and by an up-regulation of GSH synthesis [5 ]. Severe and/or chronic oxidative stress, such as that resulting from inflammation and tissue damage in RA, leads to a decline in these defense mechanisms [35 , 39 ]. Continuous exposure to even low levels of oxidants can eventually cause depletion of GSH, by depleting the substrate required to replenish GSH. Decreased protection leads to DNA damage, a rise in intracellular-free Ca2+ and iron, damage to proteins, and lipid peroxidation, resulting in cell toxicity and cell death [39 , 40 ]. The status of GSH in a cell may reflect the ability of a cell to protect itself against oxidative injury. Patients with RA exhibit significantly higher than normal plasma levels of a variety of oxidant end products and of altered plasma thiol patterns [41 , 42 ], suggesting the presence of increased oxidative stress. The significantly reduced levels of surface-thiols and iGSH in PBMC of RA patients, which we observed, may represent the constant efforts of the cells to function in the presence of pro-oxidant compounds in the cellular microenvironment [43 , 44 ]. It is interesting that a significant increase in erythrocyte GSH level in RA patients has been reported, suggesting an additional protective response against continuous ROS production [41 ].
The available amount of iGSH depends on the equilibrium between processes, during which GSH is expended and regenerated, and the process of its biosynthesis. These processes are associated closely with Meister’s -glutamyl cycle, in which the membrane enzyme GGT plays a pivotal role in the salvage pathway of extracellular GSH. GGT initiates the breakdown of GSH into its amino acid components with cysteine transported into the cell for GSH biosynthesis [13
, 45
]. Studies have shown that elevated GGT activity is involved in the prevention of NO-induced apoptosis in human Th2 cells [46
] and the U937 cell line [47
], regardless of the iGSH levels. In RA, the mechanisms that elicit and/or propagate chronic inflammation remain unclear, but accumulating evidence indicates that insufficient apoptosis represents at least one underlying process [48
, 49
]. Although not designed specifically to promote apoptosis, several medications currently used in the treatment of RA may function in part through their induction of apoptosis [50
]. As it was beyond the scope of this study to quantify GGT activity directly, as a result of time restrictions in specimen arrivals, we chose to examine CD53 levels. GGT is physically associated with the TM4 protein CD53, as indicated by the coimmunoprecipitation of GGT activity with CD53, although little is known about this association [15
]. In addition, a CD53 gene has been identified among the set of those genes that regulates apoptosis [18
]. Ligation of CD53, through interactions with the extracellular environment, triggers a survival response and reduces the number of cells that enter apoptosis, most likely, as a result of the transient activation of the c-Jun N-terminal kinase [19
]. Increasing evidence implicates c-Jun in the protection of cells against stress-induced apoptosis [51
]. A previous study implicated CD53 in adhesion by showing that stimulation of rat B cells with antibodies to CD53 triggered a homotypic adhesion reaction related to cellular migration [52
]. In the present study, we have demonstrated that RA patients have significantly higher levels of CD53 on the majority of PBL subpopulations than do control subjects. In light of what is currently known about the function of CD53, it is plausible that this protein plays important roles, not only in the migration of PBL into synovia of RA joints, but also in establishing the apoptosis-resistant nature of these cells.
Cumulative oxidative effects play a significant role in the aging process [33 ]. A host of diseases, including cardiovascular and neurodegenerative disorders, as well as RA, certainly increase in frequency exponentially with age; however, the basis for the steep rise in disease incidence with age is unexplained. An underlying factor common to aging and disease states is inflammation [53 ], which is most likely related to the increase of ROS and a decrease in efficiency of the redox balance mechanisms. The results of the work presented here show that a decline in surface-thiols and iGSH levels with increasing age is restricted to the RA patients. If we take thiol levels as an indicator of redox balance, we can posit that the accelerated decrease in thiol levels, as a result of chronic inflammation in RA patients, is compounded by normal age-related decreases.
With the ever-increasing interest in the aging process, research has turned its attention toward telomeres as biomarkers for aging and age-related diseases such as RA [21 , 22 ]. Telomeres in most human cells shorten with each round of DNA replication, as a result of decreased telomerase activity; however, telomerase activity is not the only determinant of rate of telomeric DNA loss. Damage to DNA as a result of oxidative stress is repaired less well in the telomeric region than elsewhere on the chromosome, and the result is that telomere loss is accelerated, and cellular, replicative senescence is triggered [54 , 55 ]. Given our hypothesis that cellular thiol levels are indicative of cellular oxidative stress, we looked at telomere length in PBMC of RA patients and control subjects. Although the correlation between thiol levels and reduced telomere length was not significant (>0.05), our results were consistent with findings of other studies that indicated a premature erosion of telomere length in RA [56 ].
In conclusion, surface-thiols and iGSH play important roles in redox buffering of cells, providing protection from oxidative stress and the resultant cellular damage. However, chronic stress, such as the inflammation associated with RA, leads to a decline in the levels of this protection. We hypothesize that the loss of cellular thiols of lymphocytes from the RA patients causes increased expression of GGT with a concomitant increase of CD53. Paradoxically, CD53 has been suggested to interfere with apoptosis [18 , 19 ], which should be increased as a result of a cell’s loss of reducing equivalents [5 ]. Apoptotic resistance has been suggested to be related to the pathogenesis of RA [20 ]. Currently, our aim is to examine adjunct therapies with antioxidants, which may aid in maintaining or increasing iGSH and decreasing CD53 expression, thus promoting the potential for lessening the functional changes that occur as the result of the severe stress of inflammation and aging.
ACKNOWLEDGEMENTS
Studies were, in part, supported by National Institutes of Health grant RO1-AT00309. The authors thank the staff of the Immunology Core of the Wadsworth Center for their assistance with the flow cytometry.
Received August 28, 2006; revised October 14, 2006; accepted November 8, 2006.
REFERENCES
1 Proc. Natl. Acad. Sci. USA 89,300-304-glutamyl transpeptidase Biochemistry 32,6302-6306[CrossRef][Medline]-Glutamyl transpeptidase activity alters the T cell response to oxidative stress and FAS-induced apoptosis Int. Immunol. 15,17-27-Glutamyl transpeptidase, an ecto-enzyme regulator of intracellular redox potential, is a component of TM4 signal transduction complexes Eur. J. Immunol. 28,4123-4129[CrossRef][Medline]-glutamyl transpeptidase Int. Immunol. 13,519-528-glutamyl transpeptidase activity is involved in prevention of apoptosis and maintenance of proliferation in U937 cells FASEB J. 13,69-79This article has been cited by other articles:
 |
|
 |
|
  M. J. Mattapallil, A. Augello, C. Cheadle, D. Teichberg, K. G. Becker, C.-C. Chan, J. J. Mattapallil, G. Pennesi, and R. R. Caspi Differentially Expressed Genes in MHC-Compatible Rat Strains That Are Susceptible or Resistant to Experimental Autoimmune Uveitis Invest. Ophthalmol. Vis. Sci., May 1, 2008; 49(5): 1957 - 1970. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
