Relationship between subclinical rejection and genotype, renal messenger RNA, and plasma protein transforming growth factor-beta1 levels.

Abstract

Transforming growth factor (TGF)-beta(1) is increased in allograft rejection and its production is associated with single nucleotide polymorphisms (SNPs).
The contribution of SNPs at codons 10 and 25 of the TGF-beta(1) gene to renal allograft damage was assessed in 6-month protocol biopsies and their association with TGF-beta(1) production. TGF-beta(1) genotypes were evaluated by polymerase chain reaction (PCR)/restriction fragment length polymorphism. Intragraft TGF-beta(1) messenger RNA (mRNA) was measured by real-time PCR and TGF-beta(1) plasma levels were assessed by enzyme-linked immunosorbent assay.
Eighty consecutive patients were included. Allele T at codon 10 (risk ratio, 6.7; P = 0.02) and an episode of acute rejection before protocol biopsy (risk ratio, 6.2; P = 0.01) were independent predictors of subclinical rejection (SCR). TGF-beta(1) plasma levels, but not those of TGF-beta(1) mRNA, were increased in patients with SCR (2.59 ng/mL +/- 0.91 [n = 22] vs. 2.05 ng/mL +/- 0.76 [n = 43]; P = 0.01). There was no association between allele T and TGF-beta(1) plasma or intragraft levels.
Allele T at codon 10 of the TGF-beta(1) gene is associated with a higher incidence of SCR.

Full text

Page 1

Relationship Between Subclinical Rejection and
Genotype, Renal Messenger RNA, and Plasma Protein
Transforming Growth Factor–?1Levels
Miguel Hueso,1,2Estanis Navarro,2Francesc Moreso,1Violeta Beltra ´n-Sastre,2Francesc Ventura,3
Josep M. Grinyo ´,1and Daniel Sero ´n1
Background. Transforming growth factor (TGF)–?1is increased in allograft rejection and its production is associated
with single nucleotide polymorphisms (SNPs).
Methods. The contribution of SNPs at codons 10 and 25 of the TGF-?1gene to renal allograft damage was assessed in
6-month protocol biopsies and their association with TGF-?1production. TGF-?1genotypes were evaluated by poly-
merase chain reaction (PCR)/restriction fragment length polymorphism. Intragraft TGF-?1messenger RNA (mRNA)
was measured by real-time PCR and TGF-?1plasma levels were assessed by enzyme-linked immunosorbent assay.
Results.Eightyconsecutivepatientswereincluded.AlleleTatcodon10(riskratio,6.7;P?0.02)andanepisodeofacute
rejection before protocol biopsy (risk ratio, 6.2; P?0.01) were independent predictors of subclinical rejection (SCR).
TGF-?1plasma levels, but not those of TGF-?1mRNA, were increased in patients with SCR (2.59 ng/mL ? 0.91
[n?22] vs. 2.05 ng/mL ? 0.76 [n?43]; P?0.01). There was no association between allele T and TGF-?1plasma or
intragraft levels.
Conclusions. Allele T at codon 10 of the TGF-?1gene is associated with a higher incidence of SCR.
Keywords: Protocol biopsies, Renal transplantation, Subclinical rejection, Single nucleotide polymorphisms, Trans-
forming growth factor–?1.
(Transplantation 2006;81: 1463–1466)
T
been related to the onset of allograft rejection and fibrosis
(1–4). Eight single nucleotide polymorphisms (SNPs) have
beendescribedintheTGF-?1gene(5,6),withthosedetected
in the first exon showing a relationship with allograft rejec-
tion(7–10).Exon1polymorphismsareaT-to-Ctransitionat
position?869resultinginthechangeofleucinetoprolineat
codon 10, and a G-to-C transversion at position ?915 that
resulted in a change of arginine to proline at codon 25. The
homozygous presence of the C allele at codon 10 and that of
the G allele at codon 25 have been associated with higher
TGF-?1levelsinvitroandinpatientswhohavenotreceiveda
transplant (7, 11, 12). In renal transplantation, it is not clear
whetherSNPsoftheTGF-?1geneareassociatedwithTGF-?1
productionandpathologiclesionsinpatientsinclinicallysta-
ransforming growth factor (TGF)–?1is a cytokine in-
volvedintheregulationoftheimmuneresponsethathas
ble condition. Because the presence of subclinical rejection
and chronic allograft nephropathy evaluated by means of
protocol biopsies is associated with poorer graft survival, we
evaluated the association between SNPs of the TGF-?1gene
and the presence of subclinical rejection (SCR) or chronic
allograft nephropathy (CAN) in protocol biopsies.
METHODS
Weconductedaprospective,exploratorystudyoncon-
secutive single renal transplant procedures performed be-
tween 1999 and 2001. Patients recruited fulfilled the follow-
ing criteria: serum creatinine level less than 300 ?mol/L,
proteinuria (excretion ?1 g/d), stable renal function, and
6-month protocol biopsy available. This study was approved
bytheethicscommitteeofourhospitalandwritteninformed
consent was obtained. In our center, the protocol biopsy was
performed at approximately 4 months until 2000 and at 6
months thereafter.
Two cores of tissue were obtained: one was processed
for conventional histologic examination and the other was
immediately snap-frozen in liquid nitrogen and stored at
?80°C for RNA extraction. Renal lesions were evaluated ac-
cording to 1997 Banff criteria (13). The presence of border-
line changes or acute rejection prompted a diagnosis of SCR,
and CAN was diagnosed in cases with ci plus ct scores of at
least 2. Biopsy specimens were blindly evaluated by the same
pathologisttwicetoestimatethereliabilityofbothdiagnoses.
The determination of SNPs at exon 1 of the TGF-?1
genewasperformedongenomicDNAfromperipheralwhite
blood cells with use of a polymerase chain reaction (PCR)/
restriction fragment length polymorphism method. Each
PCR sample contained 50–200 ng of genomic DNA and 0.2
?M of the sense 5?CACCACACCAGCCCTGTTC and anti-
ThisworkwassupportedbygrantstoD.S.(FISPI040086,SEN-2004,Marato ´
TV3 001210), M.H. (FIS 2000/0368), E.N. (FIS PI020766, Marato ´
TV3005310), and F.M. (FIS PI040177); and fellowships to M.H. and
V.B.S. from Fundacio ´ Catalana de Transplantament.
1IDIBELL, Department of Nephrology, Hospital Universitario Bellvitge,
L’Hospitalet de Llobregat, Barcelona, Spain.
2IDIBELL, Molecular Oncology Center, Institut de Recerca Oncolo ´gica
(COM-IRO), L’Hospitalet de Llobregat, Barcelona, Spain.
3IDIBELL,DepartmentofPhysiologicalSciences,FacultyofMedicine,Uni-
versitat de Barcelona, L’Hospitalet de Llobregat, Barcelona, Spain.
Address correspondence to: Daniel Sero ´n, M.D., IDIBELL, Departament de
Nefrologı ´a,Hospital Universitario
L’Hospitalet de Llobregat, Barcelona E08907, Spain.
E-mail: 17664dsm@comb.es
Received 31 August 2005.
Accepted 19 December 2005.
Copyright © 2006 by Lippincott Williams & Wilkins
ISSN 0041-1337/06/8110-1463
DOI: 10.1097/01.tp.0000206102.67063.24
Bellvitge.,FeixaLlargas/n,
Transplantation • Volume 81, Number 10, May 27, 2006
1463

Page 2

sense5=TCACCAGCTCCATGTGGATAGprimers.DNAwas
amplified for 35 cycles at 94°C for 45 seconds, 58°C for 45
seconds,and72°Cfor2minutes,followedbyafinalextension
at72°Cfor7minutes.APCRproductof178bpwasobtained,
theidentityofwhichwasconfirmedbysequencing.Todeter-
mine the allelic variants of the TGF-?1gene, 10 ?L of the
amplified DNA were digested with 10 U of MspA1I (codon
10)orBglI(codon25)for2hoursat37°C.Digestedproducts
werefractionedin10%polyacrylamidegelsandvisualizedby
ethidium bromide staining.
Isolation of total RNA and reverse transcription were
done as previously described (14). Steady-state levels of
TGF-?1were determined in triplicate by real-time PCR (ABI
Prism 7700; Applied Biosystems; Bedford, MA) on 0.5 ?g of
complementary DNA with use of the following primers: for-
ward primer
300nM,reverseprimer5=GCCGACTACTACGCCAAGGAat
900 nM, and the TaqMan probe5=FAM-TGGGTTTCCAC-
CATTAGCACGCG-TAMRA at 200 nM. As internal control
fornormalization,weamplifiedafragmentoftheglyceralde-
hyde-3-phosphatedehydrogenasetranscriptwithacommer-
cial pair of primers (PDARS ID number, 4310884E; Applied
Biosystems).TodetermineabsolutelevelsofTGF-?1messen-
ger RNA (mRNA), we constructed a standard curve with a
cloned fragment of the TGF-?1transcript that included the
sequence amplified. In subsequent quantifications, this plas-
mid was used to create an absolute standard curve that was
runinparallelwiththeamplificationofallunknownsamples.
TGF-?1mRNA was expressed as log number of copies.
Fasting plasma samples were obtained at the time of
protocol biopsy and total TGF-?1protein levels were mea-
suredinplatelet-poorplasmabyenzyme-linkedimmunosor-
bent assay (Quantikine; R&D Systems, Abingdon, UK). The
intraassay coefficient of variation of duplicated samples was
6%.
Results were expressed as means ? SD. Differences be-
tween categoric data were evaluated with a 2?2 contingency
5=TGCTTGAACTTGTCATAGATTTCGT at
tableandtheFisher’sexacttests.Student’sttestandtheanal-
ysis of variance were employed to compare normally distrib-
uted quantitative data. The Mann-Whitney U test and
Kruskal-Wallis test were applied for nonparametric data.
Univariate and multivariate logistic regression were em-
ployedtoidentifyvariablesassociatedwithhistologicdiagno-
sis.The?statisticwasemployedtoevaluatethereproducibil-
ity of histologic diagnoses. P values were corrected for the
number of variables compared according to the Bonferroni
method.
We analyzed protocol biopsies from 80 recipients of a
cadaveric graft. Recipient age was 47 years ? 13, 52 of the 80
patients were men, and there were 11 repeat transplantation
cases. Thirty-three patients received cyclosporine and pred-
nisone with antilymphocytic antibodies (n?7), azathioprine
(n?1), mycophenolate mofetil (n?17), or sirolimus (n?4).
The other 47 patients received tacrolimus and prednisone
with mycophenolate mofetil (n?44) or sirolimus (n?3).
TGF-?1SNP distribution in 72 cadaveric donors served as
controls (age 44 y ? 16; 62% male patients).
Protocol biopsies were performed at 168 days ? 58.
Histologic diagnoses were normal (n?28), borderline
changes (n?8), acute rejection (n?5), CAN (n?24), CAN
with borderline changes (n?12), and CAN with acute rejec-
tion (n?3). The CAN was mild in 29 cases, moderate in nine
cases, and severe in one case. Patients with borderline
changes (n?20) or acute rejection (n?8) were categorized
as having subclinical rejection (SCR; 35%). There were 39
patients(49%)diagnosedwithCANregardlessofSCR.The?
statistics were 0.83 for the diagnosis of SCR (95% CI, 0.69–
0.96; P?0.001) and 0.67 for CAN (95% CI, 0.51–0.84;
P?0.001).
RESULTS
Characteristics of patients according to TGF-?1geno-
types are shown in Table 1. The distribution of SNP at codon
TABLE 1.
Baseline characteristics of patients according to TGF-?1genotypes at codons 10 and 25a
CharacteristicCodon10Codon25
TT (n?26)CT (n?33)CC (n?18)GG (n?68)CG (n?9)CC (n?3)
Sex(M/F)
Recipient
Donor
Age(y)
Recipient
Donor
Transplantationprocedure
First
Repeat
HLAA?B?DRmismatches
Panel reactive antibodies >20% (no/yes)
Cold ischemia time (h)
Antirejectiontreatment
Cyclosporine
Tacrolimus
Delayedgraftfunction(no/yes)
Acuterejection(no/yes)
20/6
21/5
20/13
25/8
11/7
11/7
46/22
53/15
6/3
6/3
0/3
0/3
46?9
38?11
45?15
35?16
50?15
46?19
46?13
37?14
50?16
46?21
45?17
43?17
24
2
27
6
16
2
60
7
5
4
3
0
3.1?1.0
25/1
19?5
3.1?1.1
30/3
20?5
3.3?1.0
18/0
21?3
3.2?1.0
64/4
20?5
2.6?1.0
8/1
19?3
3.3?0.6
3/0
22?5
915
18
27/6
28/5
825
43
5
4
3
017
21/5
20/6
10
17/1
15/3
57/11
56/12
8/1
8/1
3/0
1/2
aCorrected P Values were not significant for all comparisons.
1464
Transplantation • Volume 81, Number 10, May 27, 2006

Page 3

10 was not different between transplant recipients and con-
trolsubjects(CC,n?18;CT,n?33;TT,n?26vs.CC,n?11;
CT,n?23;TT,n?22).Similarly,therewerenodifferencesin
the distribution of SNPs at codon 25 (CC, n?3; CG, n?9;
GG, n?68 vs. CC, n?2; CG, n?9; GG, n?47). The distribu-
tion of genotypes of the TGF-?1gene followed Hardy-Wein-
berg equilibrium.
The prevalence of SCR was different in transplant re-
cipientsaccordingtoTGF-?1SNPatcodon10(TT,46%;CT,
46%; CC, 8% in grafts with SCR vs. TT, 28%; CT, 41%; CC,
31%ingraftswithoutSCR;P?0.04)butnotatcodon25(CC,
4%; CG, 14%; GG, 82% in SCR vs. CC, 4%; CG, 10%; GG,
86% in grafts without SCR). Conversely, the prevalence of
CANwasnotassociatedwithSNPatcodon10(TT,31%;CT,
39%;CC,30%ingraftswithCANvs.TT,37%;CT,46%;CC,
17%in grafts without CAN) or codon 25 (CC, 3%; CG, 13%;
GG, 84% in CAN vs. CC, 5%; CG, 10%; GG, 85% in grafts
without CAN).
To further evaluate the association between SCR and
SNPatcodon10,westudiedriskfactorsassociatedwithSCR.
Multivariate logistic regression analysis showed that allele T
(riskratio,6.7;95%CI,1.3–35.7;P?0.02)andacuterejection
before the protocol biopsy (risk ratio, 6.2; 95% CI, 1.5–25.0;
P?0.01) were independent predictors of SCR, adjusting for
the type of anticalcineurinic treatment.
TGF-?1plasma levels were determined in 65 patients
andintragraftmRNATGF-?1renallevelsweredeterminedin
61 patients. TGF-?1plasma levels were increased in patients
with SCR (2.59 ng/mL ? 0.91 [n?22] vs. 2.05 ng/mL ? 0.76
[n?43];P?0.01)andinpatientsreceivingcyclosporinecom-
pared with tacrolimus (2.58 ng/mL ? 0.98 [n?20] vs. 2.08
ng/mL ? 0.74 [n?45]; P?0.02). Two-way analysis of vari-
ance showed that only SCR was independently associated
withTGF-?1plasmalevels(P?0.03).Tothecontrary,mRNA
TGF-?1renallevelswerenotsignificantlydifferentinpatients
with versus without SCR (2.58 log copies ? 0.76 [n?23] vs.
2.31logcopies?0.65[n?38];P?0.15)orinpatientstreated
with cyclosporine or tacrolimus (2.42 log copies ? 0.67
[n?23] vs. 2.41 log copies ? 0.73 [n?38]). Allele T at codon
10wasnotassociatedwithplasmaTGF-?1levels(2.14ng/mL
? 0.83 [n?50] vs. 2.45 ng/mL ? 0.87 [n?13]) or mRNA
TGF-?1levels (2.41 log copies ? 0.67 [n?49] vs. 2.50 log
copies ? 0.85 [n?11]).
Inthepresentstudy,weobservedanindependentasso-
ciation between recipient allele T at codon 10 and SCR at 6
months.Thisassociationisinagreementwiththeobservation
that this allele is also more prevalent in other inflammatory
diseases and in liver allograft rejection (10, 15–17). In renal
transplant recipients, an association between donor or recip-
ient SNP at codon 10 and an increased incidence of clinical
acute rejection has been reported (8, 9). These data suggest
thatSNPatcodon10mayinfluencegraftinflammation,even
though the relative contribution of the donor or recipient is
not well established at present. To the contrary, in a large
epidemiologic study (18), there was no association between
SNPs at codons 10 or 25 and 3-year renal allograft survival.
Because the detrimental effect of SCR on graft failure can be
demonstrated in only patients who undergo long-term fol-
low-up, it cannot be discarded that SNPs at codon 10 may
have an influence on long-term graft outcome (19).
IthasbeenproposedthatTGF-?1genotypeatcodon10
is associated with TGF-?1production. However, TT geno-
type has been associated with higher or with lower TGF-?1
mRNA and protein levels (7, 11, 12, 20). In renal allograft
biopsy specimens obtained in pediatric patients with chronic
allograft dysfunction, Melk et al. (20) failed to observe any
association between TGF-?1SNPs and their intrarenal ex-
pression. In our study, performed in stable grafts, we also
failed to observe this association. However, we observed that
the presence of SCR and immunosuppressive treatment was
associated with circulating TGF-?1protein levels, suggesting
thatSNPsmayplayonlyamodestroleinTGF-?1production
in renal transplantation. In addition, we cannot discard that
the association between allele T and SCR may be caused by
other genes in linkage disequilibrium with TGF-?1.
CONCLUSION
In summary, we suggest that TGF-?1genotype may
influence the susceptibility for SCR in renal transplantation.
ACKNOWLEDGMENTS
The authors thank Dr. Jordi Bover for discussions on the
topic of this work.
REFERENCES
1.Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth
factor ? in human disease. N Engl J Med 2000; 342: 1350.
Border WA, Noble NA. Transforming growth factor ?1in tissue fibro-
sis. N Engl J Med 1994; 331: 1286.
Baboolal K, Jones GA, Janezic A, et al. Molecular and structural conse-
quences of early renal allograft injury. Kidney Int 2002; 61: 686.
Wahl SM. Transforming growth factor ?1: the good, the bad and the
ugly. J Exp Med 1994; 180: 1587.
Cambien F, Ricard S, Troesch A, et al. Polymorphisms of the trans-
forminggrowthfactor-?1geneinrelationtomyocardialinfarctionand
blood pressure. The Etude Cas-Te ´moin de L’Infarctus du Myocarde
(ECTIM) study. Hypertension 1996; 28: 881.
Pociot F, Hansen PM, Karlsen AE, et al. TGF-?1 gene mutations in
insulin-dependent diabetes mellitus and diabetic nephropathy. J Am
Soc Nephrol 1998; 9: 2302.
Awad MR, El Gamel A, Hasleton P, et al. Genotypic variation in the
transforminggrowthfactor-beta1gene:Associationwithtransforming
growthfactor-beta1production,fibroticlungdisease,andgraftfibrosis
after lung transplantation. Transplantation 1998; 66: 1014.
Alakulppi NS, Kyllo ¨nen LE, Ja ¨ntti VT, et al. Cytokine gene polymor-
phisms and risk of acute rejection and delayed graft function after kid-
ney transplantation. Transplantation 2004; 78: 1422.
Hoffmann S, Park J, Jacobson LM, et al. Donor genomics influence
graft events: the effect of donor polymorphisms on acute rejection and
chronic allograft nephropathy. Kidney Int 2004; 66: 1686.
Warle MC, Farhan A, Metselaar HJ, et al. Cytokine gene polymor-
phisms and acute liver graft rejection. Liver Transpl 2002; 8: 603.
Suthanthiran M, Li B, Song JO, et al. Transforming growth factor-?1
hyperexpression in African-American hypertensives: a novel mediator
of hypertension and/or target organ damage. Proc Natl Acad Sci U S A
2000; 97: 3479.
YamadaY,MiyauchiA,GotoJ,etal.Associationofapolymorphismof
the transforming growth factor-?1 gene with genetic susceptibility to
osteoporosis in postmenopausal Japanese women. J Bone Miner Res
1998; 13: 1569.
Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classifica-
tion of renal allograft pathology. Kidney Int 1999; 55: 713.
Hueso M, Beltran V, Moreso F, et al. Splicing alterations in human
renalallografts:detectionofanewsplicevariantofproteinkinasePar1/
Emk1 whose expression is associated with an increase of inflammation
in protocol biopsies of transplanted patients. Biochim Biophys Acta
2004; 1689: 58.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
© 2006 Lippincott Williams & Wilkins
1465
Brief Reports

Page 4

15.Sugiura Y, Niimi T, Sato S, et al. Transforming growth factor-?1 gene
polymorphism in rheumatoid arthritis. Ann Rheum Dis 2002; 61: 826.
BuckwalterMS,Wyss-CorayT.Modellingneuroinflammatorypheno-
types in vivo. J Neuroinflammation 2004; 1: 10
Chan CC, Reed GF, Kim Y, et al. A correlation of pregnancy term,
disease activity, serum female hormones, and cytokines in uveitis. Br J
Ophthalmol 2004; 88: 1506.
Mytilineos J, Laux G, Opelz G. Relevance of IL10, TGF?1, TNF? and
16.
17.
18.
IL4R? gene polymorphisms in kidney transplantation: a collaborative
transplant study report. Am J Transplant 2004; 4: 1684.
Choi BS, Shin MJ, Shin SJ, et al. Clinical significance of an early proto-
colbiopsyinliving-donorrenaltransplantation:ten-yearexperienceat
a single center. Am J Transplant 2005; 5: 1354.
Melk A, Henne T, Kollmar T, et al. Cytokine single nucleotide poly-
morphism and intrarenal gene expression in chronic allograft ne-
phropathy in children. Kidney Int 2003; 64: 314.
19.
20.
1466
Transplantation • Volume 81, Number 10, May 27, 2006