| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
"G. Fornaini" Institute of Biological Chemistry, University of Urbino, 2-61029 Urbino, Italy.
2
Nephrology and Dialysis Unit, "R. Silvestrini"
Hospital, 06100 Perugia, Italy.
3
Department of Molecular and Cellular Biology, University
of Perugia, 06100 Perugia, Italy.
a Address correspondence to this author at: Istituto di Chimica Biologica "G. Fornaini", Via Saffi, 2-61029 Urbino (Ps), Italy. Fax 39-722-320188; e-mail galli{at}uniurb.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: In this study, we characterized the expression of GST in erythrocytes of 118 uremic patients under different clinical conditions. The mechanisms responsible for the regulation of protein expression and enzyme activity were investigated in light of different dialysis approaches, oxidative stress, uremic toxins, erythrocyte age, and erythropoietin (EPO) supplementation.
Results: Mean GST activity in uremic patients was highly overexpressed with respect to controls, and this phenomenon was exclusively attributable to an increased expression of GST. Overexpression of GST did not appear to be dependent on oxidative stress and was not influenced by vitamin E supplementation. In the same manner, both erythrocyte age and EPO supplementation apparently did not interfere with the GST concentrations, which were the same in controls and patients. Preliminary experiments suggested that high-molecular weight or protein-bound toxins could play some role in the overexpression of GST.
Conclusions: GST expression may be a useful marker for the individual accumulation of uremic toxins as well as of the efficiency of new dialysis strategies in removing them.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In human erythrocytes, GST is present in large amounts (5)
and in two forms: a highly cationic enzyme (designated with the Greek
letter 
), which accounts for <5% of the total GST activity; and
the main anionic enzyme corresponding to the P form (
)
(6)(7). Because of its abundance and overlapping
substrate specificity with the form, the GST P1-1 dimer is often
considered the sole GST enzyme in the erythrocyte
(2)(7).
In humans, increased concentrations of erythrocyte GST have been observed in uremia (8), where the role of GST expression may well be of relevance from a clinical standpoint. In fact, the exposure of cells to a uremic environment seems to determine the onset of various complications (9)(10), the intensity and characteristics of which differ from patient to patient as a function of the clinical conditions and of the efficiency/physiologic compatibility ratio of the dialysis procedures used, in which a critical prooxidant/antioxidant balance can be present (11).
The aim of the present study was to investigate GST expression in uremic patients under different clinical conditions. One of these was the use of a vitamin E-modified filter, which was expected to be both effective against oxidative stress-related toxicity and highly biocompatible (12). The role of a "protein-leaking" dialyzer able to eliminate high-molecular weight toxins, some of which have an inhibitory effect on erythropoiesis (13), was also investigated. Finally, we analyzed other aspects affecting GST activity, including erythropoietin (EPO) supplementation and dialysis fluid composition and biocompatibility in chronic ambulatory peritoneal dialysis (CAPD) patients.
These clinical and therapeutic approaches may influence the detoxification metabolism and, as a consequence, GST expression in erythrocytes and other cell types. In this context, the possibility of using GST expression either as an index of uremic toxicity or as a specific tool to evaluate the physiologic compatibility of some dialysis therapies in uremic patients is discussed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
clinical trials
Clinical trial 1: HD with vitamin E-modified filters.
In a
subgroup of patients in S-HD (n = 15), the effect of the treatment
with two different types of dialysis membrane on GST expression was
evaluated. The first was a cuprammonium rayon-based membrane, which was
used for at least 1 month; after this period the same subjects were
treated with a multilayer vitamin E-coated membrane (12),
which was used for 3 months. Both membranes were provided by Terumo
(Japan).
Clinical trial 2: HD with high-flux (protein-leaking) filters
(BK-F).
Four patients undergoing chronic HD with standard
polymethylmethacrylate-based low-flux filters [two with normal GST
activity and two with GST activity >3.5 U/g hemoglobin (Hb)] were
treated for a period of 3 months with high-flux protein-leaking
dialyzers (BK-F), which have a nominal cutoff of
Mr 70 000 (13). The
different permeabilities of these two classes of membrane (pore
diameter <30 Å for low-flux filters and 100 Å for BK-F,
respectively) gave us preliminary information on the effect of toxins
with molecular weights below and above 70 000 on GST expression.
Clinical trial 3: evaluation of GST expression in CAPD patients as
a function of peritoneal dialysis fluid composition and EPO
supplementation.
In a subgroup of CAPD patients (n = 10), the
effect of EPO therapy on GST expression was assayed as a function of
red blood cell (RBC) age. At the same time, these patients (n =
21), all of whom had been stabilized with lactate-based peritoneal
dialysis fluid (PDF), were shifted to treatment with bicarbonate-based
PDF for a period of 3 months and then returned to treatment with
lactate-based fluid for an additional 3 months. GST and GSH were
measured at all the three steps.
laboratory techniques
Blood sampling and RBC preparation.
Ten milliliters of venous
blood (for predialysis or CAPD patients and controls) or arteriovenous
blood (for HD patients) was drawn from either the antecubital vein or
the arteriovenous fistula into heparin-containingVacutainer
Tubes. The cells and plasma were separated by centrifugation and
treated as previously described to isolate the RBCs
(14). An aliquot of plasma was stored at -20 °C until
assayed in the kinetic experiments with purified GST; the remaining
aliquot was used immediately for the other analyses described below,
including assays of vitamin E and fatty acid content.
Human RBCs were fractionated by a slightly modified version of the density gradient procedures described by Rennie et al. (15). Cell counting and correction for the Ht values in each fraction were used to determine the amount of young and old cells. The different ages of the cell fractions were confirmed by evaluating the metabolic activity of the cells and the activity of the K+/Cl- cotransport system, as well as the values of medium corpuscular volume and medium corpuscular hemoglobin content.
The reticulocyte count was performed on whole blood by optical microscopy after specific staining, and the concentrations of EPO in the serum were measured using a RIA method, as described previously (14)(16).
Analysis of the catalytic properties of GST.
GST activity and
kinetic parameters, namely maximal activity
(Vmax), apparent Michaelis constants
(Km) for GSH and
chlorodinitrobenzene (CDNB), and optimum pH, were determined according
to the method of Habig et al. (1) with some minor
modifications as described by Carmagnol et al. (8).
GST was purified from a crude RBC lysate by the method described by
Awasthi et al. (6) and affinity chromatography with a
hexylglutathione gel (Sigma Chemicals). Alternatively, a purified
preparation of human GST- was purchased from Sigma. Kinetic,
electrophoretic, and immunoblotting data confirmed the presence of
GST- in the purified enzyme preparations; specific activity was
measured using CDNB as substrate.
The effect of the toxins contained in the uremic plasma, serum, and
dialysis fluid samples on the kinetic parameters of the GST purified
from human RBCs was also determined. These samples, obtained at
different times of dialysis, were incubated for 30 min at 37 °C in
the presence of partially purified GST-. One volume of plasma or
serum was mixed with 1 volume of enzyme suspension (0.1 g/L) in a final
volume of 1 mL. Before GST activity was measured, the mixture was
chromatographed on a Sephadex G-25 column (Pharmacia Biotech), with 10
mmol/L phosphate buffer (pH 7) as elution buffer. The fractions
containing the enzyme were recovered and concentrated to a final
protein concentration of 0.1 g/L. Enzyme activity was measured as
above, with the exception that the concentration of CDNB was 0.5
mmol/L, the Km.
In some experiments, samples of plasma were subjected to ultrafiltration on PM10 membranes (Amicon) and/or deproteinization by boiling for 10 min or by extraction with 50 g/L trichloroacetic acid followed by decantation and filtration through 0.22 µm filters.
Analysis of GST protein expression.
GST expression was
measured as described previously (17) by immunoblotting on
the RBC lysate, which was first subjected to a partial purification of
the main protein (hemoglobin) by chloroform-ethanol extraction
(18) to increase the sensitivity of the procedure. The
clarified extract was subjected to sodium dodecyl
sulfate-polyacrylamide gel electrophoresis using 10%
polyacrylamide gels, with 15 µg of protein loaded in each lane.
Proteins were blotted onto a nitrocellulose sheet (Bio-Rad
Laboratories), and a sheep anti-rabbit polyclonal antibody
(Calbiochem-Novabiochem) was used to detect the amount of GST protein
by the ECL enhanced chemiluminescence detection system
(Amersham-Pharmacia Biotech). The signal was recorded with photographic
films (Kodak; supplied by Sigma-Aldrich) and measured by laser scanning
densitometry.
Analysis of free thiols, vitamin E, free fatty acids, and
malondialdehyde.
The presence of oxidative stress was tested by
assay of the lipoperoxidative end-product malondialdehyde (MDA) as free
fraction in the total RBC lysate and plasma. The thiobarbituric acid
test was used, but its specificity was evaluated in some experiments by
direct HPLC analysis of MDA as described previously (19),
with only minor modifications. Briefly, RBC suspensions at a Ht of 30%
were subjected to hypotonic lysis, and after perchloric acid
extraction (20), the supernatant was filtered through 0.22
µm pore filters and subjected to HPLC. An LC-18T reversed-phase
column (15 cm x 4.6 mm i.d.; Supelco) and isocratic conditions
were used. The mobile phase was a mixture of 500 mL/L
NaH2PO4, pH 8, and 500 mL/L
methanol, and the flow rate was 1 mL/min. The elution time for MDA was
16.3 ± 0.6 min. An MDA calibrator was prepared by acidic
hydrolysis of a solution of 1,1,3,3-tetrahydroxypropane.
In the S-HD subgroup treated with the vitamin E-modified filter, vitamin E and lipid composition in terms of the essential free fatty acids (arachidonic, linoleic, and linolenic acid), were measured in plasma and RBC lipid extracts, as described previously (12).
Free thiols in the RBC, of which GST made up >95%, and in the plasma, which were mainly protein thiols and to a lesser extent non-protein thiols such as GSH, were assayed with Elman's reagent (14)(16).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and in Table 1
. Healthy subjects displayed a mean control value of 2.40
± 0.89 U/g Hb (range, 1.22–4.00 U/g Hb). In the predialysis patients,
a mean GST activity of 3.17 ± 1.17 U/g Hb was observed (range,
1.60–5.10 U/g Hb). This value was not statistically different compared
with controls. On the other hand, in the patients undergoing different
dialysis procedures, GST activity increased significantly with respect
to the control values (P <0.01). The rise in the mean GST
activity in the dialysis patients followed this order: S-HD >
CAPD > D-HD. These three subgroups showed mean GST activities (in
U/g Hb) of: 5.13 ± 1.60 (range, 2.11–9.10), 5.07 ± 1.61
(range, 1.28–8.50), and 3.98 ± 1.41 (range, 1.56–6.89),
respectively. The total population of dialysis patients (n = 103)
displayed a GST activity of 4.77 ± 1.62 U/g Hb (range, 1.28–9.10
U/g Hb).
|
|
Regardless of the type of dialysis therapy, 65% of the patients studied had GST activity higher than the cutoff line value of 4.2 U/g Hb, which corresponded to the mean control value plus 2 SD. As for the mean activity, in the dialysis patients the relative incidence of GST overexpression, calculated as the fraction of patients in each subgroup with GST activity above the cutoff value, followed the order: S-HD (72%) > CAPD (57%) > D-HD (43%). The mean GST activity in the D-HD patients was significantly lower than in the other two groups of dialysis patients (P <0.05 in both cases).
GST activity was not correlated with GSH concentrations in the RBCs of the uremic patients or with hemoglobin concentrations and RBC number (not shown). In a similar way, no correlation was found between GST activity and the clinical and anthropological characteristics recorded, e.g., patient age, length of time in dialysis, or creatinine and blood urea nitrogen concentrations.
expression of gst protein
The immunoblotting of the RBC lysate of controls and two groups of
patients representative of the whole population studied is shown in
Fig. 2
. The analysis revealed the presence of two bands of
Mr ~31 000 and 245 000,
respectively. The Mr 31 000 protein,
expressed in the same amount in both patients and controls, was
demonstrated by further analysis to correspond to the cytosolic enzyme
carbonic anhydrase (not shown).
|
A strict correlation between GST activity and protein expression in the RBC cytosol was observed in both the controls and patients (r = 0.889). Therefore, the increase in GST activity observed in the subgroups of uremic patients studied appears to be almost entirely a consequence of an increased GST protein expression.
kinetic analysis of gst
Total erythrocyte GST activity and GSH concentrations in the S-HD
patient group, as well as Vmax and
Km for CDNB and GSH of the partially
purified enzyme, were compared to those of the control group (Table 2
). In the patients, in the presence of an increased total
activity and of a slightly increased GSH concentration, the kinetic
behavior of GST remained unmodified.
|
Erythrocyte GSH was also slightly increased in the S-HD patient group in comparison to controls.
gst inhibition by the plasma of patients overexpressing gst
A dose-dependent inhibition of GST activity was observed in the
plasma of patients overexpressing GST in their RBCs (activity, 5 U/g
Hb), whereas the plasma obtained from healthy controls and from S-HD
patients with a GST activity near the control interval (activity, 2.5
U/g Hb) did not exert this effect (not shown). After a HD session with
conventional low-flux dialyzers (nominal cutoff 
30 000), the
inhibitory activity of the uremic plasma from subjects overexpressing
GST decreased by 35% ± 23% (n = 14; P
>0.05). Ultrafiltration of the plasma drawn from the same
subjects before the HD session with a membrane with a cutoff of 10 000
removed 27% ± 16% of the inhibitory activity (n = 3).
Deproteinization by boiling or by acidic extraction permitted the
recovery of 23% ± 9% and 31% ± 16%, respectively, of the
inhibitory activity present in the whole plasma (n = 5).
lipoperoxidation and gsh concentrations in uremic patients
The concentrations of the lipoperoxidative subproduct MDA showed a
broad distribution in the RBCs of the four subgroups, and the mean
concentration was not statistically different from that of the controls
(Table 3
). However, in the S-HD group, a positive correlation between
GST activity and MDA concentration was observed (r =
0.619; P >0.05).
|
GSH concentrations in the RBC of the four subgroups examined did not differ or were slightly increased compared with those of the controls.
effects of treatment with vitamin e-modified dialyzers on blood
vitamin e, fatty acid composition, gsh, and gst expression in s-hd
patients
In the subgroup of S-HD patients treated with cuprammonium
rayon-based filters, normal or slightly decreased concentrations of
vitamin E and high concentrations of GST activity were observed. After
the 3-month period of treatment with a vitamin E-modified filter, an
increase in vitamin E from 0.30 ± 0.11 to 0.64 ± 0.31 mg/L
RBCs was observed (113%; P <0.0001). A comparable increase
in the cell membrane content of polyunsaturated lipids, in particular
arachidonic acid, was observed (not shown). These changes did not
affect or induced only a slight decrease in the GST activity [from
3.8 ± 1.4 to 3.3 ± 1.2 U/g Hb (14%; P >0.05)]
and the GSH concentration (from 8.6 ± 1.1 to 8.9 ± 1.7
µmol/g Hb) in the RBCs.
effects of epo supplementation and rbc age on gst expression in
capd patients
Younger cells had significantly higher GST activity and GSH
concentrations but lower MDA concentrations with respect to older cells
in both the controls and patients (Fig. 3
). Regardless of the EPO supplementation, GST activity in the
young and middle fractions obtained from the patients was higher than
in healthy controls (P <0.01 in both). However, the amount
of EPO supplied to the patients did not correlate with the GST
concentrations observed in the unfractionated cells even in the
presence of a number of young cells (12% ± 5% higher than that of
controls, not shown). The mean GST concentration displayed by the
EPO-treated patients was not significantly higher than that of patients
not treated with EPO (not shown).
|
In a subgroup of patients (n = 7) not responding to EPO therapy (Ht gain at the end of the protocol <1%), a comparison with patients (n = 21) who responded to this therapy (Ht gain of 6% ± 4% after supplementation; range, 2–8%), demonstrated that these two subgroups of patients did not differ significantly in GST expression even in the presence of a significantly different reticulocyte count (respectively, 0.6% ± 0.5% and 2.1% ± 1.6%; P <0.01).
effect of pdf composition on gst expression
Regardless of the type of PDF used, GST remained significantly
higher in the patients vs controls over the entire prospective trial
(not shown). GSH remained in the control range in all three steps of
the study.
effect of bk-f filters on erythrocyte gst expression
In three subjects, two with normal starting activity and one with
high GST activity, the loss of high-molecular weight substances during
dialysis caused a decrease in GST activity by ~30% at all
experimental times considered (Table 4
). A comparable decrease in the in vitro inhibitory activity of
their plasma on the GST was observed (not shown). In the patients
displaying the highest GST activity, the BK-F treatment did not modify,
or only slightly increased, the activity observed during the treatment
with the low-flux filter.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study involving 118 subjects, we found that erythrocyte GST was overexpressed by >50% of the uremic population in various stages of uremia and undergoing different types of dialysis. Moreover, our results definitively demonstrate that this increased GST activity is the consequence of an increased expression of the protein rather than a kinetic modulation of the enzyme.
Regardless of the procedure examined, dialysis was always associated with a more severe overexpression of GST, in terms of both activity and number of subjects affected (65% of the dialysis patients), if compared with patients in conservative predialysis therapy, who displayed normal or slightly increased GST expression. The D-HD group overexpressed GST to a lesser degree than did the S-HD and CAPD groups. In fact, a daily dialysis schedule has been considered to have a more favorable dialysis efficiency/physiologic compatibility ratio with respect to other techniques, especially S-HD (22).
A possible element responsible for the high GST expression in CAPD patients might be the bioincompatibility of the PDF used, which can impair RBC metabolism either directly or via a challenge of the peritoneal tissue (23). However, in this study, we demonstrated that bicarbonate-based PDF, even if considered the most biocompatible PDF, induced the same GST overexpression in vivo as lactate-based PDF, thus suggesting that other factors can affect expression of the enzyme.
RBC age (7) can influence GST expression in dialysis patients. In fact, GST concentrations were higher in young cells, which also had higher GSH and lower MDA concentrations than did older cells. However, the profile of the GST activity in the fractionated RBCs was modified in the S-HD patients overexpressing GST. Interestingly, the younger fractions in the RBCs of these patients displayed GST activities higher than those of the controls, whereas older fractions had the same activity as the control samples. This evidence seems to suggest that in the uremic environment, GST could be progressively inactivated during the RBC life span; thus, its overexpression could be a response to low efficiency of the GST-dependent detoxification system, which tends to become defective during the aging of uremic RBCs.
Supplementation with EPO, which can increase the number of young cells in circulation, did not interfere with the mean concentrations of GST in EPO-responsive patients, who had a GST expression profile (as a function of cell age) comparable to that of patients not treated with EPO and had mean GST concentrations comparable to those of EPO-unresponsive patients.
Oxidative stress could be included in the factors responsible for an overexpression of GSH-dependent enzymes in dialysis patients. In fact, lipophilic substances released during the oxidative damage to polyunsaturated lipids, such as short-chain aldehydes and alkenals, have been demonstrated to be substrates for GST (2) and to accumulate in the plasma and RBCs of dialysis patients (16)(19). In this study, we measured the concentration of free MDA in the RBCs as an index of lipoperoxidation and found that MDA did not accumulate in the RBCs of the dialysis patients to a greater extent than that observed in the controls. Moreover, only a weak correlation between GST expression and MDA was found in S-HD, which is thought to be associated with a greater susceptibility to oxidative stress with respect to other types of dialysis, such as CAPD (16). Furthermore, in a subgroup of patients treated with vitamin E-modified dialysis filters and showing GST overexpression, although a significant increase in plasma and RBC vitamin E occurred, GST concentrations were only slightly lower. These data suggest that GST expression in uremic patients probably does not depend on a compensatory response of GST expression against oxidative stress and particularly lipoperoxidation. However, the data presented in this study do not exclude the presence of oxidative stress in dialysis patients, at least in some subjects undergoing HD procedures using poorly biocompatible materials (12).
Intriguingly, GST inhibitors are present in uremic plasma, and this could justify an overexpression of GST in erythroid cells. In in vitro experiments, we observed that whole plasma of patients with high GST concentrations exerted an inhibitory activity on GST purified from human RBCs. A major fraction of this inhibitory activity (>60%) was not removed by membranes with a cutoff of Mr 10 000, and treatment with conventional (low-flux) HD filters eliminated the same amount of inhibitory activity.
The use of high-flux (protein-leaking) dialyzers, tested herein in a pilot study on four patients, slightly decreased (<30%) the GST activity of three patients regardless of the starting GST concentrations shown during treatment with low-flux dialyzers; a comparable decrease in the inhibitory activity of their plasma was observed.
Taken together, these findings are in agreement with previous observations that low-molecular weight substances present in the uremic blood and bound with high affinity to plasma proteins may be responsible for inhibition of hepatic GST (24). Among these endogenous ligands, one has been identified as 3-carboxy-4-methyl-5-propyl-furanopropanoic acid (24), which also inhibits mitochondrial respiration and acts as a hypoproliferative agent on the erythroid stem cells (25)(26). Many other low-molecular weight toxins (27) are able to bind with high affinity to plasma proteins and could be responsible for an induction of the detoxification system. In fact, the possibility that during dialysis the steady concentration of the free form of these solutes undergoes rapid and significant changes able to affect GST expression in the erythroid cells cannot be ruled out. In this context, high-flux dialyzers might represent a solution for the removal of these protein-bound toxins and high-molecular weight toxins overall (28).
In conclusion, this study provides new information on the mechanism of the overexpression of GST in uremic patients undergoing dialysis. The overexpression of GST and its related enzyme activity do not appear to be influenced by the presence of oxidative stress and do not depend on cell age in patients and controls or on supplementation and refractory responses to EPO. It seems probable that other factors linked to uremic toxicity are responsible for this phenomenon. In particular, some low-molecular weight toxins with electrophilic properties and bound to plasma proteins may play a key role in this context. Thus, the assay of GST expression can be useful to evaluate the individual accumulation of these molecules and to test the efficiency of new dialysis strategies in removing them.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. Santangelo, V. Witko-Sarsat, T. Drueke, and B. Descamps-Latscha Restoring glutathione as a therapeutic strategy in chronic kidney disease Nephrol. Dial. Transplant., August 1, 2004; 19(8): 1951 - 1955. [Full Text] [PDF]
|
|
|
|
|
|
A. Floridi, F. Antolini, F. Galli, R. M. Fagugli, E. Floridi, and U. Buoncristiani Daily haemodialysis improves indices of protein glycation Nephrol. Dial. Transplant., May 1, 2002; 17(5): 871 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Klemm, C. Voigt, M. Friedrich, R. Funfstuck, H. Sperschneider, E.-G. Jager, and G. Stein Determination of erythrocyte antioxidant capacity in haemodialysis patients using electron paramagnetic resonance Nephrol. Dial. Transplant., November 1, 2001; 16(11): 2166 - 2171. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
; n = 5) and
S-HD patients (•; n = 8) showing GST activities of 2.2 ±
0.6 and 4.8 ± 1.0 U/g Hb, respectively. *, P
<0.01 for dialysis patients vs controls. Bars, SD.