ABO genotyping by inverse PCR technique

Legal Medicine (Legal Med) 2000 ; 2 : 15-20

ABO genotyping by inverse PCR technique Tetsuya

KOBAYASH?~~, Atsushi

AKANE~

‘Forensic Science Laboratory, Osaka Prefectural Police Headquarters,

Osaka 541-0053,

2Department of Legal Medicine, Kansai Medical University, Moriguchi 570-8506,

Japan

Japan

(Received December 21, 1999, Accepted January 20, 2000)

Inverse PCR technique was applied to type three major alleles (A’, B and 0’) of the ABO blood

ABSTRACT

group by simultaneously detecting separated allele-determining 796th and 803rd nucleotides

nucleotides

(the 261st base in exon 6 and the

in exon 7) of the ABO gene. A sequence of about 1.7 kb from exons 6 to 7 of

each allele was amplified, both termini of the fragment ligated, and allele-typing performed PCR-restriction

fragment length polymorphism

For intramolecular

(IP-RFLP)

and allele-specific

by the inverse

inverse-PCR (ASIP) methods.

ligation, primers for the first PCR were designed to have Act I-restriction sites within the

sequences, and both termini of the 1.7-kb fragment were digested with Act I. Using the IP-RFLP method, the inverse PCR product was digested with Kpn I, NZa III and Dde I, A’, B, O’-standard (OA) and O’-variant ( OG) alleles were detected as 365-bp, 272-bp, 193-bp and 12%bp fragments, respectively. By the ASIP method using four allele-specific respectively.

primers, 222-bp, 124bp and 232-bp fragments were amplified from A’, B and 0’ templates,

These techniques

would be applicable

to detecting

separated

polymorphic

regions of some

other genes. KEY WORDS: Inverse PCR, ABO blood group system, Restriction

fragment length polymorphism,

Allele-spe-

cific PCR typed independently using two sets of primers4)6)8)10)16)18). H owever, recently, variant alleles

Introduction Since

1990, when nucleotide

sequences

of the

caused by the de novo recombination

between differ-

of the AI30

ent alleles, such as the AC allele consisting of exons

blood group system were reported1)2), ABO geno-

l-6 of the B allele and exon 7 of the A’ allele*), were

types have been determined

reported ‘WI. If such a recombination

cDNA of three major alleles (A’, B, 0’)

length

polymorphism

PCR amplification

(RFLP)‘)5)-5), allele-specific

major

independent

confor-

( SSCP)s) or amplified product

polymorphism

the three

fragment

(ASPA)@‘), single-strand

mation polymorphism length

by restriction

(APLP)

alleles

techniques@.

are based

Since

on three

nu-

had occurred,

typing of the two regions would end in

an ambiguous

result. We have explored

specific inverse-PCR (ASIP) technique,

the allele-

in which two

linked sequences are ligated and analyzed by a single procedure

using allele-specific

primers”).

In this

cleotides (deletion of the 261st nucleotide in exon 6

study, the ASIP and the inverse PCR-restriction

determining

ment length polymorphism

nucleotide

the 0’ allele and the 796th and 803rd substitutions

in exon 7 determining

specificity of A and B transferases),

the

these two regions

frag-

(IP-RFLP) analysis meth-

ods were applied to conduct

correct

typing of the

three major alleles of the ABO gene. The IP-RFLP

of the ABO gene should be analyzed for correct

method was also applied to classify the 0’ allele into

genotyping1)2), although an ABO genotyping method

two suballeles5)‘*).

by a single PCR amplification Since

the distance

between

has been reported3). these two regions

Materials and Methods

is

about 1.6 kb (Fig. l), the regions were amplified and

1.

Amplification

and ligation of a 1.7-kb pagment

(Fig.

16

March 2000

KOFSAYASHI et al.

USA), and digested with a restriction A sequence

of about

1.7

kb from

6 to the 873rd nucleotide

exon

fied using primers ABO-Fl ABO-Fl

the

5’ end

of

of exon 7 was ampli-

and ABO-RI

had five non-complementary

(Table

1).

nucleotides

endonuclease

Act I at 37°C overnight to produce complementary 5’-stalk

termini

at both

ends

Following electrophoresis

of the fragment.

in 2% agarose gel, the di-

gested fragment was cut out from the gel and puri-

(GTCGA) at the 5’ end to generate a restriction site

fied using the Wizard” PCR Preps DNA Purification

of endonuclease

System

Acc I. ABO-Rl

matched nucleotide

contained

within the sequence

a mis-

to destroy

(Promega,

Madison,

amount of the recovered

WI, USA).

A l/100

sample was added to the

the restriction site of NZuIII at the 861st to 864th po-

ligation mixture (100 pl of the final volume) consist-

sitions of exon 7. The PCR mixture (50 ~1) consisted

ing of 66 mM Tris-HCl (pH 7.6), 6.6 mM MgCl,, 10

of 10 mM Tris-HCl

mM dithiothreitol,

(pH 8.3), 50 mM KCl, 0.2 mM

dNTP, 1.5 mM MgC&, 0.04 U/p1 AmpliTaq

GoldTM

66 PM adenosine-5’-triphosphate

and 0.02 U/p1 T4 ligase (Toyobo, Osaka, Japan), and

DNA polymerase (Perkin Elmer, Norwalk, CT, USA),

the

0.5 PM each of primers and 150 ng of template DNA.

overnight.

Following

2. IPRFXP analysis

heating

at 95°C for 9 min, 40 cycles at

95°C for 1 min, 60°C for 1 min and 72°C for 2 min were performed,

followed by incubation

at 72°C for

ligation

reaction

was performed

at

16°C

The self-ligated 1.7-kb circular template DNA was heated at 95°C for 30 min and chilled rapidly in ice-

7 min. The PCR product was purified using a micro-

water before the second amplification

concentrator

ABO-F2 and ABO-R2 (Table 1) to improve the effi-

Microcon@ 100 (Amicon, Danvers, CT,

using primers

ciency of PCR 13)14).The PCR conditions same as described exon 6 A’ B:G

01{g;

exon 7

lntron

703 771 796 603 GCCG A C A C

261 : G

was digested with restriction sisting of 10 mM Tris-HCl

661-664

CATG Act I site

I site >

PCR amplification (1.7 kb)

and 1 mM dithiothreitol.

(pH 7.5), 10 mM MgCl*, Next, 2 /.~lof 1 M NaCl so-

further digested by a restriction enzyme Da!eI at 37°C overnight. The digested sample was electrophoresed

Act I digestlon Intramolecular ligation

in 2% agarose gel.

\1 IP-RFLP /ASIP I IP-Sequencing A6042 ABO-F2 ABO-ABR ABO-AOF ABO-BF -ABO-OR 703

enzymes Kpn I and Nla

lution was added to the buffer, and the product was

77 Primer ABO-R m

A~IsitehnerABO-FI

the

(5 ~1)

III at 37°C overnight in a 20 ~1 restriction buffer con(N/a III site)

GCCG GTCG

$

were

above. The PCR product

79% 603

3. ASIP analysis Allele-specific

PCR amplification

using the self-ligated

was performed

1.7-kb circular

template after

261

Table

1.

Sequences

of primers

Filet PCR Fig.

1. Strategy

for ABO genotyping

technique.

The

determining

nucleotides

and the reverse

T to destroy

digestion

of

determining method.

in exons 6 and 7 of the ABO

IP-RFLP

both

primer

ABO-Fl

Act I site at the 5’ terminus

primer

nucleotide

intramolecular

ABO-Fl ABO-Rl

involving

using the forward

with a non-complementary

PCR

allele-

1.7 kb fragment

gene are amplified

by inverse

ABO-RI

with a mismatched

the Nla III site. After termini

ligation nucleotides

of

is performed by the

the

fragment,

to detect IP-RFLP

Act I allele-

or ASIP

ABO-FP ABaR

5’-gtc gaC ACT AGG AAG GAT GTC CTC GTG GT-3’ 5’GTG GTC CGG TtC TAC CAG CTG3’ 5’-TTC TAC GGA AGC AGC CGG GAGS’ 5’-AAT GTC CAC AGT CAC TCG CG3’

ASIP ABO-AOF 5’-TCG GCA CCC TGC ACC CCGJ’ ABO-BF 5’GAT TTC TAC TAC ATG GGG GG3’ ABO-ABR 5’-CAG CC4 AGG GGT CAC CA-3 ABO-OR

5’-tgt taG GGA GCC AGC CAA GGG GTA-3’

Small letters indicate non-complementary sequences.

ABO genotyping

Legal Med Vol. 2 No. 1

17

by inverse PCR

heating at 95°C for 30 min and chilling in ice-water.

gion were cut with Nla III, exons 6 and 7 ligated in-

PCR

tramolecularly

was performed

using

four

allele-specific

would be separated again, the allele-

primers, ABO-AOF detecting the A’- and O’-specific

determining

703rd G, ABO-BF detecting

independently,

the B-specific

796th A

regions at exons 6 and 7 would be typed and inverse

PCR would become

and 803rd C, and ABO-ABR and ABO-OR detecting

meaningless. Fortunately, the Nla III site exists in the

the presence or absence of the 261st G, respectively

annealing

(Table 1). ABO-OR had five non-complementary cleotides

at the 5’ end to make different

nu-

fragment

site of the primer ABO-Rl.

Accordingly,

the Nla III site was destroyed by changing cleotide

in the sequence

of ABO-Rl

one nu-

(Table

1). By

length from the PCR product of A’ allele, according

this modification,

to the APLP technique 8). The PCR conditions

allele was cut to 93bp and 272-bp fragments with Nlu

were

the inverse-PCR product of the B

the same as described above, but the PCR cycle was

III, whereas the products

repeated

were not.

30 times.

trophoresed

The

PCR product

in 12% polyacrylamide

was elec-

gel and stained

to Akane

et aLs), the 0’

alleles

are

known to be subtyped as two alleless)s)12~,which are

with ethidium bromide.

called OA and OG in this study. Between these alleles,

4. IP-Sequmcing From the 1.7-kb circular template, lele-specific

According

of the A’ and 0’ alleles

an A’- or B-al-

365-bp or O’-allele-specific

364bp

frag-

several base substitutions have been found in exons 6 and 712).In this IP-RFLP method, a nucleotide substi-

ment was amplified using primers ABO-F2 and ABO-

tution from the 771s.t C (O*) to T (OG) was checked

R2 (Table

by further digestion with Dde I; the 364bp

1). The fragment

was purified

Microcon

100, and the nucleotide

termined

by the biotinylated

using a Sequencing

using a

sequence was de-

terminator

method

inverse-

PCR product of OGwas cut with both Kpn I and Dde I, while that of O* was digested only by Kpn I. Thus, A’, B, OA and OG alleles were typed by de-

high -Plus- kit (Toyobo) .

tecting

Results

365-bp,

272-bp,

193-bp

and 128-bp

ments, respectively, in the electrophoresed

1. IP-RFLP analysis (Fig. 2)

gel.

2. ASIP analysis (Fig. 3)

The 365-bp or 364bp inverse-PCR products ampli-

For ASIP analysis, primers with high allele-speci-

fied with primers ABO-F2 and ABOR2 were typed by

ticity should be designed. The allele-specificity

RFLP analysis. The 0’ allele was determined

by di-

primer is dependent

gestion with Kpn I, which identifies

of the

primer and PCR conditions,

deletion

261st G in exon 6. The B allele was determined gestion with NZa III, which identifies 796th A. However,

all alleles

along the 861stS64th

by di-

the B-specific

A’A’

and the complementary

BB

A’A’ 0’0’

IP-RFLP enzymes

shows the molecular ladder (Pharmacia,

base of the template DNA. It

analysis

BO’ BB

C%Y’ OGOG

$I 232 bp 222 bp

93 bp 65 bp 2.

of the primer

have an NZa III site

89 :g 128 bp

Fig.

but also on the combi-

nucleotide

positions of exon 7. If the re-

365 bp 272 bp

restriction

of a

not only on the length of the

nation of the 3’ terminal

M

M

frag-

of ABO

genotypes

f with

Kpn I, Nla III and Dde I. Lane

M

weight standard marker, the lOO-bp Uppsala, Sweden).

124 bp

Fig. 3. ASIP analysis of Al30 genotypes. the lOO-bp ladder marker.

Lane M shows

KOBAYASHIet al.

18

was reported

that the combination

of A/G, G/A,

C/C or A/A may induce high allele-specificity, other combinations

fication15). Based on the report, the nucleotides

Discussion

but

may result in non-specific amplispe-

2000

March

In this study, the inverse PCR method, the two distinct regions determining

in which

the three major

cific to the ABO alleles were not suitable for design-

alleles are interligated, was applied to the simultane-

ing highly-specific

ous identification

primers,

four

primers.

allele-specific

After testing primers,

various

ABO-AOF,

intramolecularly.

allele typing. Since ABO-AOF identifies

PCR fragments

nucleotide

while ABO-BF identifies

the 796th and

803rd bases, the ASIP product of the B allele could be easily differentiated

from those of the A’ and 0’

alleles because of a great (98 bp) difference plified length. However, the difference between A’ and 0’ (@)

in am-

in sequence

is the presence or absence of

For the allele

typing, the first amplified fragment must be ligated

ABO-BF, ABO-ABR and ABO-OR were selected for the 703rd

of the nucleotides. If intermolecular from different

nantly, the genotyping

ligation between

alleles occurs domi-

will yield an ambiguous

re-

sult. In our previous study’l), in which the ASIP method was applied to genotype

the MN blood group, the

first PCR products were used directly striction digestion)

(without re-

for the ligation reaction.

In this

the 261st G alone, which causes only one base differ-

study, however, intermolecular

ence in amplified length. So according to the APLP

tween the B and 0’ alleles were produced when PCR

method*), five non-complementary

products

nucleotides were

from the BOQype

ligation products besample with primers

added to the 5’ end of O’-specific ABO-OR, and the

ABO-Fl and ABO-Rl were used directly for the liga-

sequence of ABO-ABR for detecting A’ and B alleles

tion. Whether

was designed to be as short as possible, inducing a

occur

19-bp difference

length and concentration

in the length of ASIP products be-

dominantly

intermolecular

tween the A’ and 0’ alleles. Thus, A’, B and 0’ alleles were identified

as 222-

The O* and OG alleles could not be typed by the When a forward primer identifying

generally

ligation will

depends

upon

the

of the DNA fragments”);

ligation tends to occur dominantly

with increase in the DNA concentration. concentration

bp, 124bp and 232-bp fragments, respectively. ASIP method.

inter- or intramolecular

The DNA

in the ligation mixture in this study

was so low that intramolecular to occur dominantly.

ligation was expected

One of the reasons why in-

the 771st T specific to OG (ABO-GF) was mixed with

tramolecular

four other primers, two forward primers, ABO-AOF

attributed to the conditions of the termini of the first

and ABO-GF, annealed

PCR products.

with the OG fragment,

then two ASIP products corresponding

and

to these two

ligation did not occur dominantly was T4 ligase, which was used in this

study, can ligate fragments

with not only stalk, but

primers were detected in the gel. Moreover, the for-

also blunt termini.

mer product was stronger than the latter. Generally,

DNA polymerase

the outer primer may suppress amplification

with a

(mostly, dATP) at the 3’ end of the amplified frag-

nested primer, when both primers are added to the

ment with a frequency from 15 to 90 % according to

same PCR mixture’@. Since @-specific

If non-ternplated the 3’ end sequences 18)-*o).

nucleotides

However, it is known that Taq

adds a non-ternplated

nucleotide

adenyla-

were also specific to A’ and B alleles, 0’ subtyping by

tion occurs at one terminus of the fragment, but not

the ASIP method alone was abandoned.

at the other, both ends become non-complementary

products

of the 0’ alleles

The ASIP

should be subtyped by

to each other,

inhibiting

intramolecular

ligation.

RFLP, SSCP or nested ASPA methods.

However, there would be no interference

3. IP-Sequencing

molecular ligation between the non-adenylated

The nucleotide

substitutions or base deletion spe-

cific to A’, B, OA and OG alleles were verified by sequencing.

with interends

of different fragments. To counter the inhibition

of self-ligation

by non-

ternplated adenylation, both ends of the 1.7-kb fragment amplified

with primers ABO-Fl

and ABO-RI

Legal Med Vol. 2 No. 1

ABO genotypingby inverse PCR

were digested with a restriction endonuclease be stalk and complementary restriction

to each other. Since the

site of Act I does not exist within the se-

quences

from exon 6 to the 867th

exon 7, but exists at the 868th-873rd 7, the sequence plementary Since

Act I to

of ABO-Rl

to the region

nucleotide position

was designed containing

of

of exon

to be com-

the Act I site.

no Act I site was found in exon 6, the Act I site

was added

at the 5’ terminus

1.7-kb fragment

of the fragment

of intramolecular

ligation,

type sample to be typed correctly Inverse

PCR, which

rated regions, quences

Thus,

had two Act I sites at both

Act I digestion ciency

of ABO-Fl.

involves

was explored

flanking

a region

We have applied

termini.

may also be applicable groups

forensic-type

specimens

ing. BioTechniques

of ABO alleles on

using rapid ABO genotyp-

1995; 18: 478-483.

7) Matsuki T, Nakajima T, Furukawa K. O-gene detection by allele

specific

amplification

in the ABO

blood group system. Jpn J Hum Genet

1994; 39:

293-7. 8) Watanabe G, Umetsu K, Yuasa I, Suzuki T. Amplified product

length

polymorphism

(APLP):

A novel

strategy for ABO blood group genotyping.

the BO’-

Genet 1997; 99: 34-7.

as shown in Fig.3.

of known

se-

to analyze separated of the

polymor-

phisms.

recombination

in the ABO blood group gene and of recombination

prod-

ucts. Hum Genet 1997; 99: 454-61. 19) Watanabe

G, Umetsu K, Sato M, Suzuki T. Three gene

in the ABO

blood

group system. Jpn J Legal Med 1997; 51: 205-10. 11) Li ZX, Yoshimura

S, Kobayashi K, Akane A. Allele-

specific inverse-PCR

amplification

MN blood group. BioTechniques

for genotyping 1998; 25: 358-62.

12) Olsson ML, Chester MN. Frequent

Acknowledgment

A,

G, Ito S, Matsui K, Miyazaki T. A de novo

cases of recombinated

of other

of linked

Ishimoto

evidence for the occurrence

sequence’5)‘4).

to the genotyping

Hum

9) Suzuki K, Iwata M, Tsuji H, Takagi T, Tamura

of sepa-

in the genes

and haplotyping

6) Crouse C, Vincek V. Identification

enabling

MN”) and ABO blood group systems. This technique blood

chain reaction. J Forensic Sci. 1992; 37: 1269-75.

the effi-

to analyze unknown

sequences

5) Lee JCI, Chang JG. ABO genotyping by polymerase

improved

the ligation

this technique

allele-determining

the

19

occurrence

of a

variant 0’ gene at the blood group ABO locus. VOX This work was supported the

Ministry

Culture

in part by a grant from

of Education,

Science,

Sports

and

of Japan.

reaction to amplify cellular DNA adjacent to an inte-

1) Yamamoto

F, Clausen

Hakomori

S. Molecular

14) Triglia T, Peterson J,

basis of the histo-

side the boundaries

of known sequences.

T, Marken

2) Yamamoto F, Marken J, Tsuj T, White T, Clausen H, S. Cloning and characterization

complementary transferase)

of DNA

to human UDP-GalNAc: Fuc al-+2

GalNAc transferase mRNA.

J

(histo-blood

Biol

Chem

group A

1990;

265:

Nucleic

Acids Res 1988; 16: 8186. 15) Kwok S, Kellogg DE, McKinne N, Spasic D, Goda L, Levenson mismatches

C, SninskyJ.

Effects of primer-template

on the polymerase

human immunodeficiency

chain

reaction:

virus type 1 model stud-

ies. Nucleic Acids Res 1990; 18: 999-1005. 16) Zhu KY, Clark JM. Addition of a competitive primer

1146-51. 3) Akane A, Yoshimura S, Yoshida M, Okii Y, Watabiki

can dramatically improve the specificity of PCR am-

T, Matsubara K, Kimura K. ABO genotyping follow-

plification

ing a single PCR amplification.

21: 58690.

J Forensic

Sci 1996;

of specific alleles. BioTechniques

1996;

17.

17) Dugr&zyk A, Boyer HW, Goodman HM. Ligation of

41: 272-4. 4) Fukumori

for

of DNA segments that lie out-

H, White genetic

MG, Kemp DJ. A procedure

in uivo amplification

blood group ABO system. Nature 1990; 345: 229-33.

Galal-+

13) Silver J, Keerikatte V. Novel use of polymerase chain grated provirus. J Virol 1989; 63: 1924-8.

References

Hakomori

Sang 1996; 70: 2630.

Y, Ohnoki

S, Shibata

Nishimura H. Genotyping

H, Yamaguchi

H,

of ABO blood groups by

PCR and RFLP analysis of 5 nucleotide positions. Int 1 Legal Med 1995: 107: 179-82.

EcoR I endonuclease-generated

DNA fragments into

linear and circular structures. J Mol Biol 1975; 96: 171-184. 18) Brounstein

Ml, Carpten D, Smith 1R. Modulation

of

20

KOBAYASHI

non-ternplated

nucleotide

addition

merase: primer modifications ing. BioTechniqes

that facilitate genotyp nucleotide

catalyzed by procaryotic

DNA polymerases.

Nucleic

9677-86. 20) Smith JR,

Carpten

JD,

Brownstein

M, Ghosh

S,

Magnuson VL, Gilbert DA, Trent JM, Collins FS. An

1996; 20: 1004-10.

19) Clark JM. Novel non-ternplated reactions

by Tag poly-

March 2000

et al.

addition

and eucaryotic

Acids Res. 1988;

16:

approach

to genotyping

plated nucleotide

errors caused by non-tem-

addition by Tag DNA polymerase.

Genome Res 1995; 5: 312-7.