Transplant Immunology 1998; 6: 193-197

Genotyping for polymorphisms in

interferon-y, interleukin-10, transforming
growth factor+1 and tumour necrosis
factora genes: a technical report
Chris Perrey, Vera Pravica, Paul J Sinnott and Ian V Hutchinson
School of Biological Sciences, University of Manchester; Manchester
Received 6 April 1998; accepted 21 April 1998

Abstract: Polymorphic variants of cytokine genes are associated with acute and chronic transplant rejec-
tion. In this technical report, the methods currently used in our centre to genotype individuals for inter-
feron-y, interleukin-10, transforming growth factor-b1 and tumour necrosis factor-u are described in detail.
The DNA sequences of primers and probes, and conditions for polymerase chain reactions are given, and
the allele and genotype frequencies in our control populations are summarized.

Introduction development and will be reported elsewhere, as will the method-

ology for the typing of other cytokine gene polymorphisms.
Allelic variants of cytokine genes are associated with higher or
lower production of cytokines, both in vitrr? and in vivo.' We
have shown that the genotype encoding high production of Methods
tumour necrosis factor-a (TNF-a) is articularly associated with
acute rejection of heart?’ kidne $ ,’ and liver” transplants. DNA extraction
Paradoxically, the high producer allele of interleukin-10 (IL-lo) For this study, genomic DNA from whole blood or frozen cells
is associated with protection against rejection of transplanted was obtained by phenol extraction and ethanol precipitation
hearts”’ but is associated with worse acute rejection of kidney following proteinase K (Boehringer Mannheim, Lewes, UK)
transplants.g Interferon? (EN-y) genotype is related to acute digestion. Commercially available DNA extraction kits have
kidney graft rejection” and to the development of fibrosis in been used suaxssWy in other studies.
lung all~grafts.‘~ FinaUy, transkrrnkg grcnvth f&tor_B (TGF-
pl) genotype is signikmtly associated with fibr&j and chron-
ic rejection of lung tranuplants3~‘2 and may have a bearing on DNA was ampWed using PCR performed on a PTC-100 ther-
chronic rejection of kidney &too. mal cycler (Genetic &search Instrumentation Ltd, Dunmoor,
The important porymorphirianslie in the promoter reqions of UK). Each 30 @ rcxtion n&ure contained 2 pl of test DNA
IL-lo’ and TNF-a,” in the first intron of the IFN-y gene and in (50-200 ng), 50 mM KCl, 10 mM ‘I&HCl, 0.1% lfiton X-100,
the signal sequence of TGF-p1.3 This technical paper summarizes 200 pM each dATp dm, dGTP and dTTP (Gibco BFU), 2.5
the methods we currently use to type polymorphisms in these four mM MgClz (except for TGF-pl where 1.5 mM MgCl, was used),
cytokine genes. These involve the polymerase chain reaction 1 M betaine mono&We @@a, Poole, Dorset, UK), 0.5 pM
(PCR) and either sequence-spec& o&x&e&& probing or each primer and 1 U ?&g p@mcmse (Gibco BRL, Paisley, UK).
the determination of alleles -ding to t&e size of the PCR After an initial time of 5 min, samples were subjected
product by gel electrophoresis. A&native methods are under to 30 rounds at !XPCfar 30 8, @mealing temperature (depend-
ingonthesetofprimcns)fbr3Osand724Iforlmin,witha
fmal extension time of5 min at 72°C. The primers used in each
Address for correspondence: IV Hut&bon, Immunology Research
Group, School of Biological Sciences, University of Manchester, reaction and their anneaiing temperatures are shown in Table
Stopford Building 3.239, Oxford Road, Manchester Ml3 9PT, UK E- 1. The amplified products were monitored by electrophoresis on
mail: [email protected] a 2% agarose gel with ethidium bromide (10 mg/l).
0 Arnold 1998 0966_3274(98)TI225OA
194 C A3rey et al.

‘Ihble 1 Primers and annealing temperatures for cytokine gene polymorphisms

Primer pairs Sequence T (“C)

TGF-Sl sense S-CITCACCAGCTCCATGTCGATAG-3 60
TGF-j31 antisense 5’-ACTGCGCCCPK’ICCCTG-3
TNF-a -308 sense 5’-ACTCAACACAGCITIT CCCTCCA-3 66
TNF-a -308 antisense 5’-TCCTCCCI’GCTCCGATTCCG-3
IL-10 sense 5’-ATCCAAGACAACACTACTAA-3 55.5
IL-10 antisense 5’-TAAATATCCC-3
IPN-y sense 5’-GCTGTCAT~TAATA’ITCAGAC-3 56
IFN-yantisense 15’~CGAGClT&MAGATAGITCG3’
Regions amplified: TGF-01: +691 to +%5, W-a; -331 to -224, IL-l& -1115 to -528, IFN-T +812 to +932.

TN&, IL10 and TGF-@l promoter region were baked in an oven at WC for 10 min and the DNA was immo-
polymorphism analysis biied onto membrane by crosshnking in a UV Stratalinker
Sequence-speci$c oligonucleotide pminhg (SOP). After specific (Stratagen Ltd, Cambridge, UK) at 120 mJ/cm2. The membranes
PCR reactions were performed, two 5’ biotinylated oligo- were incubated in 50 ml tubes (Falcon, Becton Dickinson, New
nucleotide probes (Genosys, UK, Pampisford, UK) were used to Jersey, USA) with 10 ml of 5x SSC hybridization buffer (where
positively identify each polymorphism in cytokine gene promot- 5x SSC is 0.75 M NaCl and 0.075 M sodium citrate) with 0.5%
ers by a dot blot technique. A total of 2 pl of PCR product was blocking agent (milh powder), 0.1% ZV-lamylsarcosine and 0.02%
blotted onto Hybond N+ nylon transfer membrane (Amersham sodium dodecyl sulphate (SDS) for 30 min at 42.5”C. Following
International, Slough, UK). The double-stranded DNA was sep the addition of 400 ng of a specific probe (see Table 2) to each tube,
arated by treating membranes with denaturing solution (0.5 M membranes were ahowed to hybridize for 90 min at 42.5”C. The
NaOH and 1.5 M NaCI) for 5 mm and then neutralizing solution membranes were washed twice in 5x SSC containing 0.1% SDS
(1.5 M NaCl and 0.5 M Tiis, pH 7.5) for 1 min. The membranes for 5 min at room temperature and then stringency washed in lx

lbble 2 Probes and stringency temperatures used in cytokine gene polymorphisms

Probe Seauence T (“0

TGF-Bl codon lO*C coding strand 5’-GCTGCTGC~GCTGCTGC-3 58
TGF-pl codon lO*T coding strand 5’-GCTGCTG~GCTGCTGCT-3’
TGF-Sl codon 25*G coding strand 5’-GCCTGGCC($GCCGGCCG-3’ 62
TGF-or codon 25*C coding strand S-GCCTGGCC~GCCGGCCG-3
TN&a: -308*G coding strand 5’-GAGGGGCATGQGGACGG-3 56
TN&a: -308*A noncoding strand 5’-CCCGTC~CATGCCCCTC-3 59
IL-10 -1082”G coding strand 5’~lTCl-TTGGGA~GGGGAAG-3 47
IL-10 -1082*A noncoding strand 5’-ACITCCC~CCCAAAGAA-3
IL-10 -819*C coding strand 5’-CAGGTGATGTAACATCETGTGC-3 61
IL-10 -819*T noncoding strand 5’-GCACAGAGATA~ACATCACCKZ3
IL-10 -592*C coding strand S-CCGCCTGT~CT~AGGAA-3 48.5
IL-10 -592*A noncoding strand S-TTCCTACAG~ACAGGCGGG-3

Codon 10 Coaon ZU

F‘lgure 1 Autoradiography of TGF-& hybridization probes. Poiymerase chain reaction products from individuals were placed in the same position
on each membrane. Photographs show positive results for C and/or T for codon 10 and G and/or C for codon 25, respectively.

ltansplant immunology 1998; 6: 193-197

 Genotypingfor polymorphisms: a technical report 195

2f3 315 3/3 2f2 3/3 a3 m 30 2f3 3f4 3/4 313 2/s z4 2i3

Figure 2 Separation of the polymerase chain reaction products containing polymorphic dinucleotide repeat on the polyacrylamide gel. Allele #2
has 12 CA repeats, #3 has 13, #4 has 14 and #5 has 15 CA repeats.

SSC with 0.1% SDS for 30 min at the temperature appropriate for approach is cumbersome for individual samples, and for this
each cytokine, as stated in ‘Iable 2. The membranes were washed reason we are developing amplification refractory mutation sys-
for 1 min in 0.15 M NaCl and 0.1 M ‘Iris buffer, pH 7.5, incubat- tem (ARMS)-PCR methodologies for genotyping.
ed for 30 min at room temperature in the same buffer containing The identification of the IFN-y CA-repeat polymorphisms are
0.5% milk powder, as a blocking agent to reduce nonspecific bind- shown in Figure 2. A single band represents a putative homoxy-
ing, and then incubated with strept peroxidase gote, so that the sample fourth from the left is a #2/#2 homozy-
conjugate (Amersham International, Slough, UK) for 30 min at gote and the third sample, with a slower band, is a #3/#3
room temperature before detection by chemiluminescence using homoxygote. Heterozygotes have two bands, as illustrated in the
the ECLR system (Amersham International, Slough, UK). X-ray three samples on the right, which are identified as having alle-
llhns were exposed and binding of allele-specific probes was used les #2l#3, #2/#4 and #2/#5, respectively (see Figure 2). An
to determine genotypes (Figure 1). example of the band identifying allele #l, which is very rare and
runs below the allele #2 band, is not shown on this gel. At pre-
Analysjs of microsatelIite polymorphism in the first sent, we have no simple manual alternative to the PAGE
intron of IF3I-y gene approach to determine the number of CA repeats in the first
After specific PCR reactions were performed, 8 pl of each sam- intron of the IFN-7gene. This method allowed us to type up to
ple was mixed with 3 pl of 5x ‘Iris-borate electrophoresis load- 30 samples.
ing buffer (Tris 0.89 M, Boric acid 0.89 M, EDTA 0.02 M, pH The distribution of alleles and genotypes in our control pop-
8, 49% glycerol, 0.1% SDS, 1% bromophenol blue) and sepa- ulations are shown in Tables 3-6. It should be noted that these
rated by polyacrylamide gel electrophoresis (PAGE) (using a gel are principally Caucasian healthy volunteers or cadaveric renal
containing 12% acrylamidebis acryhunide, 19:l) at 35 mA for transplant donors. Any study should have a set of matched con-
4 h. Following electrophoresis, gels were stained with ethidium trols, particularly in studies of other ethnic groups. We have rea-
bromide (10 mg/l), visualized with UV light on a transillumina- son to suggest that allele distribution may be quite different in
tor and photographed for the definition of alleles (Figure 2). other populations.
In general terms, we can ascribe high, intermediate and low
cytokine-producer status according to zygosity, being homozy-
Results and discussion gotes high/high, heteroxygotes high/low and homozygotes
low/low, respectively. The high responder alleles are TNF-a -
The methods describe here have been used successfully to type 308 *A; TGF-81 codon 10 *T (Leu), codon 25 *G (Arg); IFN-
cytokine gene polymorphisms for interferon-y, IL-IO, TNF-cz y allele #2; IL-10 -1082 *G. These are indicated in Tables 3-6,
and TGF-81, cytokines known to be associated with acute and and have a bearing on the interpretation of cytokine gene inher-
chronic human organ graft rejection.5-‘3 itance in transplant patients.
An example of the PCR-SSOP results of the detection of It is worth emphasizing that the cytokine gene polymorphisms
polymorphisms in the TGF-81 leader sequence is shown in described in this technical paper are known to have functional
Figure 1. The dot in each corresponding position on all four relevance, either because they directly affect gene transcription
membranes represents the DNA from an individual. Hence, the or protein synthesis, or because they are in very close linkage
individual whose DNA is at the top right-hand comer is codon with a functional mutation. In particular, they are strongly sta-
10 *C positive and *T negative (i.e. homozygous C/C) and tistically associated with post-transplant outcome, and may serve
codon 25 *G positive and *C positive (i.e. heteroxygous G/C). as useful prognostic indicators of transplant rejection and
The PCR-SSOP method has the advantages that large batches immunosuppression.
(40-80) can be done at one time, and that alleles are positively
identified by the binding of specific probes so that homoxygotes
can be defined with certainty. However, the PCR-SSOP

Transplant Immunology 1998; 6: 193-197

196 C l+rrey et al.

lhble 3 Frequences of TGF-gl genotypes for codon 10 and codon 25 and haplotype inheritance in corn&r

TGF-/31 poiymorphisms Genotype ControIs (n = 107) Frequency (%)

Ahele (phenotype)
Codon lO*Leu (high) T 139 65
Codon lO*Pro (low) ,C 75 35
Codon 25*Arg (high) G 193 90
Codon 25*Pro (low) C 21 10
Genotype (phenotype)
Codon 10
Leu/Leu (high) Tn. 44 41
Leu/Pro (intermediate) CfI 51 48
Pro/Pro (low) c/c 12 11
Codon 25
Arg/Arg (high) G/G a7 81
ArgRro (intermediate) G/C 19 18
Pro/Pro (low) c/c 1 1
Haplotype inheritance
LeuiLeu ArgMrg “r/T G/G 44 41
LeuiPro ArglArg “TIC G/G 38 36
Leu/Pro Arg/Pro bT/C G/C 13 12
Pro/Pro Argkg bC/C G/G 5 5
LeulLeu ArgPro t’IR G/C 0 0
Pro/Pro Arg/Pro “C/C G/C 6 6
Pro/Pro Pro/Pro ‘C/C G/C 1 1
LeulLeu Pro/Pro WT c/c 0 0
Leu/Ro Fro/Pro “r/c c/c 0 0

High TGF-gl production in vivo and in vitro is associated with codon 1O’Leu and codon 25*Arg. Putative ‘high, bintermediate and low producer
haplotypes are indicated.

‘Ibble 4 Frequencies of TNF-a alleles and genotypes in controls. Allele A+ (TN&) individuals produce more TNF-a upon in vitro stimulation

TNF-o position -308 polymorphism Controls (n = 106) Frequency (%)

Ahele (phenotype)
G (low) 154 73
A (high) 58 27
Genotype (phenotype)
GG (low) 65 61
GA (high) 24 23
AA (high) 17 16

‘Ihble 5 Frequencies of IL-10 alleles at position -1082 and frequency distribution of genotypes, including polymorphisms at positions -819 and
-592 in controls

IL-10 polymorphisms Controls (n = 330) Frequency (%)

Allele (phenotype)
-1082*G (high) 336 51
-1082*A (low) 324 49
-819*C 509 77
-819*T 151 23
-592*C 509 77
-592*A 151 23
Haplotypea (phenotype)
GCC (high) 336 51
ACC (intermediate) 181 27
ATA (low) 143 22
Haplotype inheritance (phenotype)
GCUGCC (high) 99 30
GCC/ACC (intermediate) 68 21
GCC/ATA (intermediate) 70 I 21
ACUACC (low) 27 8
ACC/ATA (low) 41 12
ATAIATA (low) 25 8
‘Three putative haplotypes have been identified: GCC (at positions -1082*G, -819*C and -592*C, respectively), ACC and ATA.

Transplant Immunology 1998; 6: 193-197

 Genotyping for polymorphisms: a technical report 197

%ble 6 Frequencies of IFN-7 phenotypes and frequency distribution

4 Wilson AG, di Giovine FS, Blakemore M, Duff G. Single-base poly-
of the alleles of CA repeats in controls
morphism in the human tumour necrosis factor-a (TNF-a) gene
detectable by Ncol restriction of the PCR product. Hum Mol Genet
IJ?N-~po1ym0fphisnls Controls (n = 164) Frequency (%)
1992; 1: 353.
Allele (phenotype) 5 ‘lhgore A, Gonsalkorale WM, Pravica V et al. Interleukin- 10
#l (unknown) 0 0 (IL-lo) genotypes and serum IL-10 levels in idiopathic inflamma-
#2 (high) 158 48 tory bowel diseases Clin Exp Zmmunol (submitted).
#3 (low) 140 43 6 ‘lamer DM, Sankaran D, Grant SCD et al. The effect of polymor-
#4 (low) 14 4 phism in the TNF-a and IL-10 genes on heart transplant rejection.
#S (low) 16 5 Int J Heart Lung Tmnsplunt 1997; 16:108.
Genotype (phenotype) 7 lkner DM, Grant SCD, Yonan N et al. Qtokine genotypes and
#2/#2 (high) 34 20.7 heart transplant rejection. Tmnspkzntation1997;64: 776-78.
#2/#3 (intermediate) 77 46.9 8 Sankaran D, lamer DM, Johnson RW et al. Tnterleukin-10 and
#2/#4 (intermediate) 5 3 tumour necrosis factor alpha gene polymorphisms predict renal
#2/#5 (intermediate) 7 4.3 transplant outcome. Eur JZmmunogenet 1997; 24: 65.
#3/#3 (low) 24 14.6 9 Sankaran D, Ashraf S, Asderakis A et al. Recipient tumour necrosis
#3/#4 (low) 7 4.3 factor-a and interleukin-10 gene polymorphisms predict acute graft
#3/#5 (low) 9 5.5 rejection following renal transplantation. TmnsZmmunol(submitted).
#4/#4 (low) 1 0.6 10 Pravica, V, Perrey C, Plevris J et al. Elevated frequency of TNF-cx -
308 A allele in patients requiring liver transplantation. Immunology
1997;92 (suppl 1): 13.3.
11 Asderakis A, Pravica V, Dyer PA, Sinnott PJ, Hutchinson IX CA-
References repeat polymorphism in the first intron of the interferon-y (IFN-y)
gene. Zmmunofogy1997;92 (suppl 1): 15.8.
1 Thmer DM, Williams DM, Sankaran D, Lazarus M, Sinnott PJ, 12 Awad MR, Pravica V, El-Game1 A, Hasleton P, Sinnott PJ,
Hutchinson IV. An investigation of polymorphism in the interleukin- Hutchinson IV. CA-repeat allele in the first intron of the IFN-y
10 gene. Eur J Immwwgen 1997; 24: l-8. gene is associated with lung allograft fibrosis. Immunology 1997;92
2 Pravica V, Asderakis A, Perrey C, Hajeer A, Sirmott PJ, Hutchinson (suppl 1): 13.9.
IV. In Vito production correlates with CA repeat polymorphism in 13 Awad MR, El-Game1 A, Simm E, Hasleton P, Yonan N, Deiraniya
the human interferon-ygene. Eur JZmmunogenetics (in press). AK et al. Genotypic variation in the transforming growth factor-beta
3 Awad MR, lkrner DM, Sinnott PJ, Hutchinson IV Polymorphism 1 gene: association with TGF-fil production, fibrotic lung disease and
in the TGF-pl gene. Eur J Immunogenetics 1997;24145. graft fibrosis after lung transplantation. Transplantation(in press).

Tmnspknt Immunology 1998; 6: 193-197