TGF-β1 Gene Polymorphisms and Peritoneal Equilibration Test Results in CAPD Patients

Abstract

Transforming growth factor-β1 (TGF-β1) stimulates the expression of collagen mRNA in cultured human peritoneal mesangial cells, which may predispose them to developing peritoneal fibrosis. Polymorphisms in the signal sequence genetically may be responsible for increased TGF-β1 production (i.e., a substitution at amino acid position 10 and 25, +869 Leu¹⁰–Pro and +915 Arg²⁵–Pro, respectively). The aim of this study was to find out whether there is any relation between peritoneal equilibration test (PET) results and TGF-β1 gene polymorphism. Thirty-two CAPD patients and 72 healthy subjects were included into the study. Each CAPD patient had undergone two PET with a two-year interval. The patients were classified according to the results of a baseline PET as high (high-high average) and low (low-low average) transporters. In high transporters group (n = 20), the genotype frequencies were found as 45% Leu/Leu, 55% Leu/Pro for codon 10; and 85% Arg/Arg, 15% Arg/Pro for codon 25. In low transporters group (n = 12), the genotype frequencies were detected as 66.7% Leu/Leu and 33.3% Leu/Pro for codon 10; and 83.3% Arg/Arg, 16.7% Arg/Pro for codon 25. The distribution of the TGF-β1 genotypes in our control population was compatible with a Hardy-Weinberg equilibrium. We found no relation between TGF-β1 genotypes and peritoneal transport group (χ² test, p > 0.5). There was no relation between TGF-β1 genotype and longitudinal change in peritoneal transport. This study is the first study analyzing the possible link between TGF-βl gene polymorphisms and the characteristics of peritoneal transport and longitudinal change of peritoneal transport characteristics in CAPD patients. Further work is needed to clarify the functional importance of these two polymorphisms in TGF-β1 production and in the development of peritoneal fibrosis.

Keywords:

• transforming growth factor-β1
• peritoneal fibrosis
• peritoneal dialysis
• gen polymorphism

INTRODUCTION

Peritoneal fibrosis is one of the most serious complications after long-term continuous ambulatory peritoneal dialysis (CAPD) that causes the loss of dialytic function.[1],[2] Some factors like duration of dialysis, the dialysate glucose concentration, and peritonitis have been reported to lead the peritoneal fibrosis and the loss of dialytic function. The incidence of ultra filtration (UF) failure truly increases with time spent on CAPD, suggesting that structural and functional changes occur within the peritoneal membrane.[3] Mesothelial cells (MCs) of patients undergoing peritoneal dialysis are continuously exposed to non-physiological composition of dialysate characterized by low pH, hyperosmolality, and high glucose as well as lactate content. Continued exposure of the peritoneum to this nonphysiological dialysate and acute peritonitis is known to result in structural and functional alterations of the peritoneum.[3],[4]

Although many factors seem to be responsible for the inadequacy of peritoneum and peritoneal fibrosis, some patients have not been affected by these factors, and a great inter-patient variability of peritoneal equilibration test (PET) results has been observed. PET provides a simple quantitative assessment of peritoneal membrane function, and its results may change due to peritoneal fibrosis. A high glucose environment and cytokines released due to inflammation of the peritoneal membrane stimulate transforming growth factor-βl (TGF-βl) expression in various cell types, and this up-regulated TGF-βl synthesis has been implicated in the development of peritoneal fibrosis.[4]. TGF-βl is one of the cytokine growth factors that are known to play a central role in extra cellular matrix regulation.[5] TGF-β1-induced fibroblast proliferation, stimulation of collagen synthesis, and enhanced matrix accumulation contribute to the development of fibrogenesis.[6],[7] TGF-βl gen polymorphisms in the signal sequence genetically may be responsible for increased TGF-βl production. Seven polymorphisms in the TGF-βl gene have been reported in a great number of studies.[[8] At present, the basis for increased TGF-βl production in humans is unknown; however, recent evidence suggests that genetic factors may have a significant role to play in the increased TGF-βl production.[7] The rate of increase in TGF-βl production in response to different stimuli such as high glucose or inflammatory cytokines may be different in each patient. TGF-β1 gen polymorphisms may be responsible for this difference.

The aims of this study were to determine TGF-β1 gene polymorphisms in codon 10 and 25 and analyze the influence of TGF-β1 gene polymorphisms on the peritoneal transport rate, as well as to demonstrate whether there is any relationship between peritoneal longitudinal change and TGF-β1 gene polymorphisms.

PATIENTS AND METHODS

Thirty-two CAPD patients (22 male, 10 female) and 72 healthy subjects (40 male, 32 female, mean age: 44 ± 12 years) were studied. Mean age for CAPD patients was 45.3 ± 11.6 years (range 24–71 years) and the median duration on CAPD was 24 months (range 6–72 months). The underlying renal disease was chronic tubulointerstitial nephritis in 6, chronic glomerulonephritis in 5, amyloidosis in 4, hypertensive nephrosclerosis in 4, and unknown in 13 patients. The overall peritonitis incidence was 0.7 episodes/patient-year in our CAPD population. Our patients were performing their CAPD treatment using lactate-containing standard dialysis solution for four or five exchanges. The glucose concentration for each exchange was calculated. For example, for an individual who was using 4 × 2 L exchanges (2 × 1.36%, 1 × 2.27%, and 1× 3.86%), there would be 54.4 + 45.4 + 77.2 = 176.8 g of glucose per day.

Peritoneal transport was evaluated using a standard 4-hour PET with 2 L dialysate containing 2.27% glucose.[9] All patients were followed prospectively for two years. Each patient had undergone two PET with a two-year interval. The patients were divided into two groups as a high transporter (high/high average) and low transporter (low/low average) according to the baseline PET results.

Venous blood samples from 32 CAPD patients were collected in EDTA coated tubes. Two biallelic polymorphisms of the TGF-β1 gene were studied at position +869, codon 10 (leucine to proline substitution) and position +915, codon 25 (arginine to proline substitution). All patients gave informed consent.

Detection of TGF-β1 DNA Polymorphism

DNA was prepared from whole blood employing a simple salting-out procedure. PCR amplifications of the related regions were carried out in 50 μl volumes of reaction mixtures containing 75mM Tris-HCL (pH 8.8), 100 mM NH₂SO₄, 2 mM MgCl₂, 50 mM of each dNTP, 50 pmol of each set of specific primers, 0.5 U of Taq DNA polymerase, and 0.5 μg DNA sample. The cycling conditions were as follows: 94^oC for three minutes, followed by 40 cycles of 94^oC for 30 sec, 60^oC for 30 sec, and 72^oC for 45 sec in an automated thermal cycler (Perkin Elmer Cetus Model 9600, Norwalk, Connecticut, USA). The amplified products were analyzed by electrophoresis on 2% agarose gel. TGF-β resulted in 195 bp amplicon. Homozygote LL, homozygote PP, and heterozygotes produce 1, 1, and 2 bands, respectively.

Statistical Analysis

Data are shown as mean ± SD or as percentages. Comparison between groups was performed by chi-square test or one-way analysis of variance (ANOVA). Hardy-Weinberg's equilibrium was checked with the chi-square test. The effect of TGF-β1 genotype on the longitudinal change in peritoneal transport characteristics was analyzed by ANOVA for repeated measures, with TGF-β1 genotype as the independent grouping factor. p < 0.05 was considered significant.

RESULTS

We studied 32 CAPD patients and 72 healthy controls. Table 1 shows TGF-β1 genotype distribution in the CAPD patients and control group. There were no significant differences in the TGF genotype for the patients and controls. The distribution of the TGF-β1 genotypes in our control population was compatible with a Hardy-Weinberg equilibrium. There were 20 high transporters and 12 low transporters CAPD patients. Mean D/P creatinine ratio at 4 hours was 0.68 ± 0.29, glucose ratio at 4 hours was 0.33 ± 0.10, and mean Kt/v was 2.10 ± 1.35. Distribution of TGF-β1 gene polymorphisms and peritoneal transport group of the patients are listed Table 2. In the high transporters group, the genotype frequencies were found as 45% Leu/Leu, 55% Leu/Pro for codon 10; and 85% Arg/Arg, 15% Arg/Pro for codon 25. In the low transporters group, the genotype frequencies were detected as 66.7% Leu/Leu and 33.3% Leu/Pro for codon 10; and 83.3% Arg/Arg, 16.7% Arg/Pro for codon 25. There was no relation between TGF-β1 genotype and peritoneal transport group (chi-square test, p = 0.63 for codon 25, p = 0.21 for codon10 (see Table 2). There were no differences in D/P creatinine and glucose ratio at 4 hours, Ktv, or Ccr among the genotypic distributions for codon 10 and 25 (see Table 3). TGF-β1 genotype did not affect longitudinal change in D/P creatinine and glucose ratio at 4 hours, Ktv, or Ccr, as listed in Table 4.

Table 1 TGF-βl gene polymorphisms in patients and control subjects

Polymorphisms	CAPD n (%)	Control n (%)	p
Codon 10			0.6
LL	17 (53.1%)	36 (50%)
LP	15 (46.9%)	36 (50%)
Codon 25			0.7
A/A	27 (84.4%)	63 (87.5%)
A/P	5 (15.6%)	9 (12.5%)

Table 2 TGF-βl genotype and peritoneal transport group (n = 32)

	Codon 25 (n = 32)		Codon 10 (n = 32)
Transporter type	A/A	A/P	L/L	L/P
High	17	3	9	11
Low	10	2	8	4

Evaluation of synergism between TGF-βl genotype and peritoneal transport group by chi-square test.

Abbreviations: High = high and high-average, Low = low and low-average

Table 3 Comparisons of peritoneal transport characteristics between TGF-βl gene polymorphisms groups (n = 32)

	Codon 25			Codon 10
	A/A	A/P	p	L/L	L/P	p
D/P glu	0.34 ± 0.11	0.32 ± 0.08	.69	0.33 ± 0.13	0.34 ± 0.07	.91
D/P cre	0.69 ± 0.31	0.62 ± 0.13	.60	0.61 ± 0.18	0.76 ± 0.37	.18
Kt/v	2.15 ± 1.39	1.76 ± 1.17	.65	2.49 ± 1.73	1.68 ± 0.56	.14
Ccr	75.76 ± 27.82	92.50 ± 17.00	.26	78.6 ± 28.57	77.38 ± 26.08	.91

Table 4 The effect of TGF-βl genotype on the longitudinal change in peritoneal transport characteristics (n = 32)

	0 Mon	24 Mon	0 Mon	24 Mon	p
Codon 25	A/A		A/P
Kt/V	2.44 ± 0.80	2.18 ± 1.41	2.43 ± 0.49	1.76 ± 1.17	.66
D/P glu	0,34 ± 0.10	0.64 ± 0.94	0.26 ± 0.05	0.28 ± 0.29	.44
D/P cre	0,72 ± 0.33	0.71 ± 0.33	0.66 ± 0.05	0.80 ± 0.18	.93
Ccr	76.61 ± 29.86	63.94 ± 22.07	98.00 ± 15.87	89.87 ± 21.78	.08
Codon 10	L/L		L/P
Kt/V	2.40 ± 0.61	2.49 ± 1.73	2.49 ± 0.94	1.7 ± 0.58	.28
D/P glu ratio	0.34 ± 0.13	0.60 ± 0.90	0.32 ± 0.07	0.60 ± 0.90	.97
D/P cre ratio	0.62 ± 0.17	0.64 ± 0.16	0.81 ± 0.39	0.81 ± 0.41	.14
Ccr	82.00 ± 30.86	68.15 ± 23.50	76.09 ± 28.10	66.04 ± 24.09	.67

DISCUSSION

In this study, we found that TGF-βl gene polymorphisms were not associated with peritoneal characteristics or longitudinal change in peritoneal transport in CAPD patients.

Although CAPD patients were exposed to the same factors such as peritonitis, high glucose concentration, and long duration of dialysis, they did not demonstrate similar peritoneal transport characteristics. After exposure to these factors, the response rate of TGF-βl production may be different, and this difference may be responsible for the varying characteristics of peritoneal membrane transport.

TGF-βl, a multifunctional cytokine, has a role in fibrosis of multiple organs. It is known that the expression of TGF-βl is increased in some CAPD patients.[10] It is also suggested that increased TGF-βl levels are related to the peritoneal fibrosis in these patients. There are several in vivo and in vitro studies supporting this hypothesis.[1],[4],[10],[11] TGF-βl decreases both extracellular matrix accumulation and degradation and increases type 1–3 collagen synthesis. The reasons for high production of TGF-βl in peritoneum are still unclear in peritoneal dialysis patients.[10]

TGF-βl gene polymorphism may have an effect on the increase in the TGF-βl levels.[[12] A study by Award et al. found that there is an increase in both TGF-βl serum levels and pulmonary fibrosis due to TGF-βl gene polymorphism.[12] In addition, Arwight et al. and Gamel et al. also reached the same results.[[13] According to these studies, as TGF-βl gene polymorphism increases TGF-βl expression, it may have a direct relation to the increased fibrosis. However, no such study seems to exist.

The increase in TGF-βl expression in response to different stimuli such as high glucose, the dialysate used, or inflammatory cytokines may be different in each patient. The level of TGF-β1 may be the result of the individual production and secretion capacity for TGF-β1, which is genetically controlled. A high glucose environment is known to stimulate TGF-βl expression in various cell types.[4],[5],[11] TGF-β1 expression by MCs exposed to a high glucose milieu (15–45 times higher than normal) in vitro has been studied to evaluate the role of high glucose-induced TGF-β1 synthesis in the development of peritoneal fibrosis.[14]

Kumano et al. reported that a high glucose content increased the TGF-β1 gene expression of rat MCs in association with decreased MC proliferation and regeneration.[15] They suggested that TGF-β1 played a central role in mediating the biological effect of high glucose dialysate on peritoneal MCs, leading to peritoneal fibrosis. On the other hand, chronic exposure to glucose-containing dialysate may result in the deposition of advanced glycation end-products (AGEs) in peritoneal tissues.[16] The increased peritoneal deposition of AGEs might correlate with the development of peritoneal fibrosis and ultra filtration failure.[17]

In CAPD patients, peritoneal MCs are not only continuously bathed in nonphysiological dialysate with high glucose, but are also subjected to recurrent inflammatory episodes.[18],[19] These microenvironments can result in structural damage to MCs. Dobbie et al. reported thickening and diabetiform changes of MCs,[19] and others observed the patchy destruction of the MC layer and a considerable increase in the thickness of the fibrous layer of mesothelium in long-term CAPD patients.[20],[21]

TGF-βl gene polymorphisms in the signal sequence genetically may be responsible for increased TGF-βl production. Seven polymorphisms in the TGF-βl gene have been reported in many studies.[8]

In our study population, there is no correlation between TGF-βl gene polymorphisms and longitudinal change in peritoneal transport. It is unlikely to be a result of inadequate power, as median duration of dialysis was not long enough and it has a relatively small sample size. The relationship between the duration of peritoneal dialysis and ultrafiltration failure has been confirmed.[22] The pathophysiological features of the gradual changes in peritoneal membrane transport are not well understood. Recent studies of human and animal peritoneal biopsy samples revealed that the density of the peritoneal microvasculature and generalized thickening of the submesothelial collagenous zone of the peritoneum, connective tissue deposition, and microvascular sclerosis increased with the increase in the dialysis time.[22] In addition, it has been noted that dialysate TGF-β1 concentrations increased with increased duration of dialysis and correlated with peritoneal membrane solute transport in peritoneal dialysis patients in stable condition.[22] Systemic processes, such as diabetes mellitus, atherosclerosis, systemic inflammation, and malnutrition, may also play a role in membrane dysfunction.[22] For this reason, we excluded all patients who had these conditions.

Unfortunately, neither circulating nor peritoneal TGF-β1 levels were measured in our study. Therefore, it is difficult to clarify the relation between TGF-β1 levels and peritoneal transport characteristics. We cannot confirm the effects of codon 10 and 25 polymorphism of TGF-β1 gene on the production of TGF-β1 protein.[8],[22] It might be that the increased TGF-βl expression by MCs induced by high glucose or inflammatory state play an important role in peritoneal fibrosis and loss of function as an ultrafiltration barrier in chronic PD patients.[4],[11]

There are two important limitations in this study: a lack of the peritoneal biopsy that evaluated peritoneal fibrosis, and a lack of the results of plasma TGF-β1 levels.

In the literature, the TGF-β1 codon 10 and codon 25 TGF-β1 genotypes' relation to lung fibrosis after lung transplantation, graft vascular disease after heart transplantation, end-stage heart failure caused by dilated cardiomyopathy, renal failure after heart transplantation, hypertension in blacks, and gingival overgrowth due to cyclosporine A and calcium channel blocker in renal transplant patients has been described.[6],[7],[12],[13,], [23–25] It is not clear yet which alleles cause higher amounts of the TGF-β1 protein.

This study is the first study analyzing the possible link between TGF-βl gene polymorphisms and the characteristics of peritoneal transport and longitudinal changes of peritoneal transport characteristics in CAPD patients. Clearly, further studies are needed to establish whether the polymorphisms of TGF-βl gene affect the function and synthesis of the TGF-βl protein and whether these polymorphisms cause peritoneal fibrosis. Moreover, peritoneal biopsy might be useful for the diagnosis of peritoneal fibrosis.