Role of apolipoprotein E epsilon 4 allele on chronic allograft nephropathy after renal transplantation

Role of apolipoprotein E epsilon 4 allele on chronic allograft nephropathy after renal transplantation

Role of Apolipoprotein E Epsilon 4 Allele on Chronic Allograft Nephropathy After Renal Transplantation D. Hernández, E. Salido, J. Linares, M.A. Cobo,...

65KB Sizes 0 Downloads 49 Views

Role of Apolipoprotein E Epsilon 4 Allele on Chronic Allograft Nephropathy After Renal Transplantation D. Hernández, E. Salido, J. Linares, M.A. Cobo, Y. Barrios, M. Rufino, S. Garcı´a, B. Martı´n, V. Lorenzo, J.M. González-Posada, A. González-Rinne, and A. Torres

ABSTRACT Lipid abnormalities may contribute to chronic allograft nephropathy (CAN). Apolipoprotein E (ApoE) gene polymorphism regulates lipoprotein metabolism, but little is known about an association between CAN and this polymorphism. The ApoE gene (E3/E4) polymorphism was typed by PCR assay (99 E3/E3, 28 E3/E4, 1 E4/E4) on 128 consecutive renal transplant patients with functioning grafts for more than 3 years (6.7 ⫾ 2.8 years). Twenty-eight patients with histological CAN were compared with 100 patients who had no clinical evidence of chronic rejection (no proteinuria and sCr ⬍ 2.5 mg%). As expected, univariate analysis revealed that patients with CAN experienced a greater acute rejection rate (78% vs 21%; P ⫽ .001), a higher serum creatinine (3.6 ⫾ 1.7 vs 1.4 ⫾ 0.5 mg%; P ⫽ .0001), and an older organ donor (43 ⫾ 20 vs 29 ⫾ 13 years; P ⫽ .0001). The lipid profiles (total cholesterol and triglycerides levels) were similar in both groups with 60% in each group receiving anti-lipemic drugs. Interestingly, the ApoE epsilon 4 allele was overrepresented in the group with CAN (39% vs 17%, P ⫽ .019). Logistic regression analysis showed that the epsilon 4 allele was an independent predictor of CAN (OR: 3.4; CI 95%: 1.07 to 11; P ⫽ .040) as were donor age and acute rejection episodes. In conclusion, an interaction between risk factors and genetic factors may determine CAN in this population. This finding may help to target prophylactic interventions in these recipients.


IPID ABNORMALITIES may contribute to atherosclerosis and chronic allograft nephropathy (CAN) in both the general population and in renal transplant recipients.1,2 Apolipoprotein E (ApoE), a protein consisting of 299 amino acids, not only contributes to the clearance of lipoprotein particles from the plasma but also plays a key role in the atherosclerotic process and progression of renal failure.3 In addition, ApoE gene polymorphism may modulate lipoprotein metabolism; the ApoE epsilon 4 (␧4) allele has been associated with a more atherogenic profile in uremic patients.3,4 Moreover, the ApoE ␧4 allele has been found to be a genetic marker of global atherosclerosis in renal transplant recipients.5 Taken together, these findings suggest that the ApoE polymorphism may have a deleterious effect, but whether this predisposes to CAN, which resembles many aspects of atherosclerosis, remains to be elucidated. This cross-sectional study was performed to assess the influence of Apo-E ␧4 allele on chronic allograft nephropathy after renal transplantation.

PATIENTS AND METHODS We selected 128 consecutive patients according to the following inclusion criteria: functioning graft over 3 years (mean: 6.7 ⫾ 2.8 years) and immunosuppressive therapy with cyclosporine (CsA). Kidney transplantations were performed in a regional center and immunosuppression consisted of sequential induction therapy with antithymocyte globulin (ATGAM, Upjohn, Kalamazoo, Mich, USA) or OKT3, as well as CsA, prednisone, and azathioprine or mycophenolate mofetil for maintenance. Episodes of acute rejec-

From the Department of Nephrology, University Hospital of the Canary Islands, Research Institute Reina Sofia, La Laguna, Tenerife, Spain. Supported by grant (PI 2003/008) from Consejeria de Educación, Cultura y Deporte (Gobierno de Canarias) and FIS (02/1350 and C03/03) from Spanish Ministry of Health. Address reprint requests to Domingo Hernández, MD, Department of Nephrology, University Hospital of the Canary Islands, Ofra s/n 38320, La Laguna, Tenerife, Spain. E-mail: [email protected]

0041-1345/04/$–see front matter doi:10.1016/j.transproceed.2004.10.038

© 2004 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710


Transplantation Proceedings, 36, 2982–2984 (2004)



tion were initially treated with three intravenous boluses of 500 mg methylprednisolone.

Endpoints When clinically indicated an allograft biopsy was performed. Histological evidence of CAN was defined according to the BANFF classification.6 Patients with histologically documented CAN (n ⫽ 28) were compared with 100 recipients (control group) without clinical evidence of this entity, that is, with stable serum creatnines (⬍2.5 mg%) and no proteinuria during follow-up.

Statistical Analysis Categorical variables were analyzed using chi-square or Fisher exact probability tests as appropriate. For univariate comparisons of numerical variables a t test was used. Stepwise logistic regression was performed to identify predictor factors for CAN. Values are expressed as means ⫾ SD, with a P value ⬍ .05 considered significant. All computations were made using the SPSS 12.0 (Chicago, Ill, USA).

RESULTS Predictor Variables Demographic and clinical data recorded at the time of renal transplant included: recipient age, gender, body mass index (BMI, g/m2), cause of renal disease, donor age, prior transplant, previous blood transfusions, panel reactive antibodies, number of HLAABDR mismatches, delayed graft function, and smoking history. Following renal transplant, we recorded chemical data and major events such as acute rejections and the development of posttransplant diabetes, hyperlipidemia, or hypertension defined according to standard criteria. We also recorded clinical and biochemical data prior to this study. Biochemical measurements were performed using standard enzymatic methods.

Genomic Typing of the ApoE Gene Polymorphism Genomic DNA was isolated from whole blood anticoagulated with EDTA. The genotypes (E3/E3, E3/E4, E4/E4) of ApoE polymorphism were ascertained by polymerase chain reaction amplification of the CfoI polymorphic site of ApoE gene, as previously reported.7 The primers used for amplification were 5= GCCCGGCTGGGTGCGGACATG 3= and 5= GAGGTGCACCCGCAGCTCCTC 3=.

The distribution of the E3/E3, E3/E4, and E4/E4 genotypes in our patients was 77% (n ⫽ 99), 22% (n ⫽ 28), and 1% (n ⫽ 1), respectively. This genotype distribution was similar to other Caucasian populations. As expected, univariate analysis revealed that patients with CAN showed a higher acute rejection rate and older donor organs than those without this complication. Likewise, serum creatinine levels prior to the study were greater in CAN patients as compared with the control group. Lipid profiles (total cholesterol and triglycerides levels) were not different between the groups, but a similar proportion of patients with and without this complication were receiving lipid-lowering agents (60%). Other known CAN risk factors were similar between the two groups. Interestingly, the ApoE ␧4 allele was overrepresented in the group with CAN (Table 1). By multivariate logistic regression analysis, the ␧4 allele was an independent risk factor for CAN, increasing the probability of this entity by three times compared with the ␧3 allele (Table 2).

Table 1. Clinical Characteristics and Biochemical Data in Both Groups

Recipient age (years) Male gender (%) Donor age (years) BDR mismatches Acute tubular necrosis (%) Serum creatinine at discharge (mg/dL) Serum creatinine prior to study (mg/dL) Posttransplant BMI (kg/m2) Posttransplant hypertension (%) Posttransplant diabetes (%) *Posttransplant hyperlipidemia (%) Total serum cholesterol prior to study (mg/dL) Triglycerides prior to study (mg/dL) Smoking (%) CMV infection (%) Duration of RT (years) Acute rejections rate (%) Acute rejections number CsA levels prior to study (ng/mL) ACEI treatment (%) ␧4 allele (%)

Patients with CAN (n ⫽ 28)

Patients without CAN (n ⫽ 100)

P value

48.6 ⫾ 12 82 43 ⫾ 20 1.6 ⫾ 0.9 38 1.9 ⫾ 0.7 3.6 ⫾ 1.7 27.2 ⫾ 5 96 14 64 218 ⫾ 48 139.5 ⫾ 65 14 43 6.9 ⫾ 3.7 78 1.3 ⫾ 1 193 ⫾ 78 25 39

49.6 ⫾ 13.6 68 29 ⫾ 13.6 1.7 ⫾ 1 26 1.7 ⫾ 0.7 1.4 ⫾ 0.5 27.5 ⫾ 5 84 15 731 216 ⫾ 39 139.4 ⫾ 75 17 46 6.5 ⫾ 4 21 0.6 ⫾ 0.8 207 ⫾ 46 30 17

.723 .165 .000 .791 .229 .162 .000 .762 .118 .416 .479 .830 .986 .732 .768 .493 .001 .000 .247 .606 .019

Abbreviations: BMI, body mass index; CMV, cytomegalovirus; ACEI, angiotensin-converting enzyme inhibitors. *Pretransplant and posttransplant hyperlipidemia includes high cholesterol and/or triglycerides levels.



Table 2. Risk Factors for CAN After Renal Transplantation* Variable

Odds ratio

95% confidence interval

Donor age Acute rejection ␧4 allele

1.04 2.1 3.4

1.07–1.1 1.2–3.5 1.07–11

P value

.005 .010 .040

Abbreviation: CAN, chronic allograft nephropathy. *Stepwise logistic regression model (n ⫽ 128 transplants and 28 CAN patients). An odd ratio greater than 1.00 indicates a higher risk for CAN. Other variables included in the model, but not in not in the table, were recipient age, BDR mismatches, posttransplant body mass index, renal function, CsA levels, duration of renal transplant, posttransplant diabetes, smoking, CMV infection, and postransplant hypertension and dyslipidemia (each with P ⬎ .5).


The results of this study suggest that the ApoE ␧4 allele is an important risk factor for developing CAN in renal transplant recipients. To our knowledge, no previous studies have determined the possible association between this polymorphism and CAN. CAN remains the leading cause of late allograft failure. There is growing information that atherogenesis is involved in its development.8,9 Lipid disorders may contribute to CAN, but the exact mechanism of hyperlipidemia-induced CAN is unclear.2 Individual differences to develop CAN in response to similar lipid abnormalities suggest that genetic factors are implicated in the pathogenesis of this complication. ApoE gene polymorphism has a lipid-modulating effect. There is accumulating evidence that the ApoE ␧4 allele has atherogenic potential. The ApoE ␧4 allele is commonly associated with an increased prevalence of atherosclerotic vascular lesions, independent of plasma lipid levels.1,10 CAN resembles atherosclerotic vascular disease and ApoE is abundantly synthesized in the human kidney. Thus, the ApoE ␧4 allele may play an important role in the development of CAN. Indeed, the ApoE ␧4 allele was more prevalent in our patients with histologically documented CAN, displaying an independent effect in the multivariate analysis. Similarly to previous reports,5 there were no differences in lipid profile and lipid-lowering therapies in both groups, a finding that bears out the influence of the ␧4 allele on CAN. Obviously, prospective studies are needed to confirm these interesting findings. In any case, our finding argues in favor of using prophylactic interventions, such as statins, on ApoE ␧4 allele patients to minimize this complication. The fact that pravastatin was able to inhibit chronic rejection in an animal transplant model11 supports this point of view. There are possible reasons to explain these findings. First, immunosuppressants contribute importantly to the posttransplant lipidic disorders.12 CsA induces hyperlipidemia by increasing production and/or reducing catabolism of lipoproteins. In addition, CsA accelerates low-density lipoprotein (LDL) oxidation in vitro.13 The ApoE ␧4 allele

not only influences plasma lipoprotein levels but also induces down-regulation of LDL receptor on the cell surface.3 It is plausible that the ApoE ␧4 allele may have a permissive role in the development of CAN when a hyperlipemic stimulus is present. In other words, CsA-induced hyperlipidemia may unmask the effects of the ApoE ␧4 allele following transplantation. Second, a more pronounced negative effect of the ApoE ␧4 allele on allograft might be expected in renal transplant recipients who are known to have multiple risk factors. In this setting, our findings support the argument that CAN is a consequence of an interaction between environmental and genetic factors. Further studies are warranted to better understand this interaction. The present study has several limitations. It was a retrospective study. The number of transplant patients and events (CAN) may be too small to adequately assess our findings. Inclusion of a larger sample population is required in future studies to confirm a causal relationship between CAN and ␧4 allele. We did not determine the ApoE ␧2 allele, which may be another potential limitation. However, ␧2 is underrepresented in the Spanish population14 and it is unlikely to have had an effect in our study, which was specifically aimed to assess the association between the CAN and the ␧4 allele, the one that has been reported to be atherogenic. In conclusion, the ApoE ␧4 allele is an independent risk factor for CAN. This finding may help to target prophylactic interventions in renal transplant recipients.

REFERENCES 1. Wilson PF, Myers RH, Larson MG, et al: JAMA 272:1666, 1994 2. Fried LF, Orchard TJ, Kasiske BL, et al: Kidney Int 59:260, 2001 3. Liberopoulos E, Siamopoulos K, Elisaf M: Am J Kidney Dis 43:223, 2004 4. Lim PS, Liu CS, Hong CJ, et al: Nephrol Dial Transplant 12:1916, 1997 5. Rodrigo E, Gonzalez-Lamuno D, Ruiz JC, et al: Am J Transplant 2:343, 2002 6. Racusen LC, Solez K, Colvin RB, et al: Kidney Int 55:713, 1999 7. Corbo RM, Vilardo T, Mantuano E, et al: Clin Genet 52:77, 1997 8. Halloran PF, Melk A, Barth C: J Am Soc Nephrol 10:167, 1999 9. Fink JC, Onuigbo MA, Blaut SA, et al: Am J Kidney Dis 39:1096, 2002 10. McCarron MO, Delong D, Alberts MJ: Neurology 53:1308, 1999 11. Ji P, Si M-S, Podnos Y, et al: Transplantation 74:821, 2002 12. Quaschning T, Mainka T, Nauck M, et al: Kidney Int 56 (Suppl 71):S235, 1999 13. Moore R, Hernandez D, Valantine H: Drug Saf 24:755, 2001 14. Muros M, Rodriguez-Ferrer C: Atherosclerosis 121:13, 1996