Forensic

Science
ELSEVIER
Forensic Science International
75 (1995) 85-93
Internatiii

Km genotyping by polymerase chain reaction

(PCR) using allele-specific amplification primers

Yuji Yamamoto*, Hideo Ishizu

Department of Legal Medicine, Okayma University Medical School, 2-5-l Shikata-cho,
Okayama 700, Japan

Received 9 January 1995; accepted 30 March 1995

Abstract

We developed a method for identifying genetic polymorphisms of the human immunoglob-

ulin kappa light chain, namely the Km allotypes, by allele-specific amplification by means of
the polymerase chain reaction (PCR). This procedure consisted of first and second PCR. In
the first PCR, the 353 bp fragment of the human immunoglobulin kappa light chain constant
gene was amplified without differentiating individual alleles. In the second PCR, the specific
sequence in each of the three alleles Km*l, Km* 1,2 and Km*3, in the first PCR product was
specifically amplified using allele-specific primers. The product of the second PCR was
separated by electrophoresis on a polyacrylamide gel and the band was observed by means
of UV trans-illumination after staining with ethidium bromide. From DNA extracted from
lymphocytes, the specific sequence of each Km allele was selectively amplified, and the Km
genotype was identified. Moreover, the Km genotype could be ascertained from whole blood,
saliva and hair roots without DNA extraction. No genetic contradiction was found in the
results of Km genotyping among parents and their children. The estimated gene frequency of
115 Japanese individuals living in Okayama Prefecture, Japan, was: Km* 3 = 0.739 and
Km*l,2 = 0.261.

Keywords: Polymorphism; Deoxyribonucleic acid (DNA); Human immunoglobulin kappa

light chain; Km allotypes; Polymerase chain reaction (PCR); Allele-specific amplification

* Corresponding author.

0379-0738/95/$09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved

SSDI 0379-0738(95)01775-E
86 Y. Yamamoto, H. Ishizu /Forensic Science International 75 (1995) 85-93

1. Introduction

The allotype of the human immunoglobulin kappa light chain (x-chain) was first
described by Ropartz et al. [l] in 1961 and was named as Inv factor. The genetic
polymorphism seen in the K-chain is called Km (kappa marker) at present. In Km
allotypes Km(l), Km(2) and Km(3) antigenic specificities are known and amino
acid substitutions in the ~-chain constant region are involved in the manifestation
of these specificities [2,3]. The K-Chain locus is located on chromosome 2. The locus
comprises a single constant gene (CK gene), five joining genes and about 50 variable
genes [4]. For the CK gene, the alleles Km*l, Km*1,2 and Km*3, manifest the Km
allotypes. Among them base substitutions occur on codons, which correspond to
amino acid residues 153 and 191 from the N terminal of K-chain protein [5,6].
Namely, the base sequence of the codon at 153 is GCC in Km*3 and Km*1,2 and
it is GTC only in Km*l. By substitution of the second base of the codon, alanine
is generated in Km*3 and Km* 1,2 and valine is generated in Km*l. At codon 191,
base substitution occurs on the first base of the codon as G in Km*3 and C in
Km*1,2 and Km*l. As a result, amino acid substitutions occur as valine in Km*3
and leucine in Km*1,2 and Km* 1. As to the Km*1 allele, however, the base
sequence has not yet been determined and it is inferred from the base sequence
estimated from amino acid substitution at the polymorphic sites.
Based on the findings on the CK gene, Kurth et al. [6] reported in 1991, the
specific amplification of the CK gene by means of the polymerase chain reaction
(PCR) and a means of detecting Km alleles by dot hybridization using allele-specific
probes. However, this method requires radiolabeled probes with which to detect
Km alleles and, therefore, it is considered that a more simplified test method is
needed for forensic application. We, therefore, improved the method of Kurth et al.
[6] as follows. After the specific amplification of the CK gene, we applied semi-
nested PCR using allele-specific amplification primers to detect Km alleles. We
established a simple and sensitive method with which to identify the Km genotypes.

2. Materials and methods

2.1. Specimens
Lymphocytes were separated from the peripheral blood of 115 Japanese individ-
uals using Ficoll-Paque (Pharmacia LKB, Uppsala Sweden). After rinsing with
sterilized physiological saline, 100 ~1 of lysis buffer (10 mM Tris-HCI pH 8.3, 50
mM KCI, 0.1% Triton X-100, 200 pg/ml proteinase K) was added to the
lymphocyte pellet (about 1 000 000 cells), then heated at 65C for 3-5 h and further
at 100C for 5 min. The lymphocyte lysate was used for the first PCR. Further-
more, 0.5 ~1 of whole blood, 0.1 ~1 of saliva and 3 mm of the root portion of hairs
from ten Japanese adults were used for the first PCR without previous DNA
extraction procedures including lysis. For the second PCR, the amplified product of
the first PCR was diluted ten times with sterilized distilled water.
 Y. Yamamoto, H. Ishizu / Forensie Science International 75 (1995) 85-93 87

Table I
Oligonucleotide primers synthesized for Km genotyping by semi-nested PCR

Primer Sequence (5-3) Notes

Name Symbol

KMPI ACTGTGGCTGCACCATCTG 5 Cli primer

KMP2 GAACTGAGGAGCAGGTGGG 3 CK primer
KM153A 0 GAGTTACCCGATTGGAGGG ASA (Km*l,2; Km*3) primer
KM153V GAGTTACCCGATTGGAGGA ASA (Km*l) primer
KM191V : AGACTACGAGAAACACAAAG ASA (Km*3) primer
KM191L @ AGACTACGAGAAACACAAAC ASA (Km*l; Km*1,2) primer

Symbols for the primers are used in Table 2 and Fig. 1

bASA represents allele-specific amplification.

2.2. Primers
Table 1 shows the primers for Km genotyping by semi-nested PCR. KMPI and
KMP2 are a primer set for amplification of the 353 bp fragment including
allele-specific base sequence sites of the CK gene [6] and KM1 53A, KM153V
KM 19 1V and KM 19 1L are allele-specific amplification primers for the Km alleles.
These primers were synthesized using a DNA Synthesizer 380B (Applied Biosys-
terns, Foster City, CA) and purified briefly using a SepPak C,, Cartridge (Millipore,
Milford, MA).

2.3. Principle of Km genotyping

Table 2 shows the primer sets used for the first PCR and the second PCR (four
reactions from @ - 0) of the semi-nested PCR. The target DNA fragment (target
allele) in each reaction and the predicted size of the fragment are also given.
The method consists of the first and second PCR. In the first PCR, 353 bp
fragments including the base substitution sites at codons 153 and 191 of the CK
gene were amplified without discriminating each allele. In the second PCR, some of
the amplified products from the first PCR were used as specimens and four
independent PCRs of @, @, 0 and 0 having specificity to each allele were
performed in parallel (Fig. 1). Each of the products of the second PCR from @ - 0

Table 2
Primer sets for the first and second PCR

PCR Primer set Predicted product (bp) Target gene or alleles

First PCR (2%@ 353 CK

Second PCR
Reaction @ a-0 152 Km* 1,2;Km*3
Reaction @I @-0 152 Km*1
Reaction 0 0-B 126 Km*3
Reaction 0 a-Q 126 Km*l; Km*l,2
88 Y. Yamamoto, H. Ishizu /Forensic Science International 75 (199.5) 85-93

1. First PCR
Cadon No. 153 191
Km1 ----------------T--------------C--------------
K,,,.l,z ----------------c--------------C--------------
Km3 ----------------C--------------G---------------
a a
-----------------------------------
-----------------------------------
4 -__-____----_____-----------------
-----------------------------------
----, -----_____----____________________

2. Second PCR

Km. 1 Kml, 2 Km3

+ chain 5,-------T-------C-------3 5,-------C-------C--------3 5,-------C-------G-------3
- chain 3,-------A-------G-------5 3,-------G-------G-------5 3---___- G------C --____- .j

0 0, G--____-
-%
0 0--, c --_____ G9
+ c------
---------- 8 %

Fig. 1. Strategy for identifying Km alleles by allele-specific amplification with PCR.

was separated by electrophoresis and the genotype was identified from the migra-
tion profiles of the amplified fragments.

2.4. Conditions for PCR

2.4.1. The first PCR. A PCR mixture containing 30 ~1 of 10 mM Tris-HCl pH
8.3, 50 mM KCI, 1.5 mM MgCI,, 100 PM of each dNTP, 0.1% Triton X-100, 1.5
U of Taq DNA polymerase (Promega, Madison, WI) and 0.5 ,LLM each of primers
@ and Q were dispensed into a 500 ~1 tube on crushed ice. To each tube, 1 ~1 of
the specimen was added and the mixture was covered with about 50 ~1 of mineral
oil. The tube was then placed in a Program Temp Control System PC-700 (Astec,
Fukuoka, Japan). After heating at 95C for 5 min, 35 thermal cycles of PCR (each
cycle: 40 s at 94C, 40 s at 60C and 40 s at 72C) were performed.
2.4.2. The second PCR. To each of the tubes on crushed ice, PCR mixtures
containing 30 ~1 of 10 mM Tris-HCI pH 8.3, 50 mM KCI, 1.5 mM MgCI,, 100
PM of each dNTP, 0.1% Triton X-100 1.5 U of Taq DNA polymerase and primers
of each set from the reactions @ - 0 shown in Table 2 at 0.5 PM each were
dispensed. To each tube, 1 ~1 of a diluted solution of the products of the first PCR
was added. About 50 ~1 of mineral oil was overlaid and this was allowed to react.
The thermal PCR cycle was set to heating at 95C for 3 min and to 25 cycles (each
cycle: 40 s at 94C 40 s at 63C and 30 s at 72C).
 Y. Yamamoto, H. Ishizu / Forensic Science International 75 (1995) 85-93 89

Table 3
Km genotyping based upon the results of electorophoretic analysis of the second PCR products

Genotype Results of the second PCRs

@ @ 0 0
Km*l/Km*l + +
Km*l/Km*1,2 + + +
Km*l/Km*3 + + + +
Km*l,2/Km*1,2 + +
Km*1,2/Km*3 + + +
Km*3/Km*3 + +

+, The predicted product was amplified by the second PCR.

2.5. Electrophoresis
PCR amplification products (9 ~1 each) were resolved by electrophoresis on an
8% polyacrylamide gel at 350 V for 40 min. After staining with ethidium bromide
(EtBr), bands of the amplified DNA fragments were visualized under UV trans-illu-
mination. The Km genotype was identified from the migration profiles of the
second PCR products using the criteria shown in Table 3.

2.6. Direct sequencing

To verify that this PCR method correctly detected the Km alleles, the first PCR
product from individuals who had been typed as homozygotic for the Km*1,2 or
Km*3 alleles using this procedure was sequenced by means of dideoxy chain
termination using a Silver Sequencekit (Promega, Madison, WI), with @ and Q as
the primers. Twenty ~1 of the first PCR products were separated by 1.5% NuSieve
GTG (FMC, Rockland, ME) agarose gel electrophoresis and the predicted 353 bp
band was recovered. The band was purified twice using a Centricon- concentra-

132 / 1,2 I,2 / 3 3/3

-l-----l-
M@@OO@@@O@@@O

Fig. 2. Electrophoretic separation of the second PCR products from individuals typed as having each of
the three common Km genotypes. Products of the second PCRs were separated by means of polyacry-
lamide gel electrophoresis at 350 V for 40 min, then stained with ethidium bromide.
90 Y. Yamamoto, H. Ishizu /Forensic Science International 75 (1995) 85-93

tor (Amicon, Beverly, MA) and suspended in 50 ~1 of distilled water. Five 1.11of the
purified DNA was used as a template for the sequencing reaction. The cycle
sequencing reaction was performed according to the manufacturers instructions
and the products of the reaction were resolved by electrophoresis on a 4%
polyacrylamide gel containing 7 M urea. The gel was then stained with Silver Stain
Plus (Bio-Rad, Hercules, CA) and the developed bands were read on a light box at
visible wavelengths.

3. Results and discussion

The 353 bp fragment of the CK gene was amplified from the specimens by the
first PCR using @ and Q primers as predicted from the sequence homology
between the CK gene and the primers.

192 / 3 3/3 132 / 3

III Annealing
@@0aD@@0@@@0@ temperature

60 C

63 C

6 6 C

Fig. 3. Electrophoretic separation of the second PCR products. The second PCR products from three
Japanese individuals were separated by means of polyacrylamide gel electrophoresis at 350 V for 40 min,
then stained with ethidium bromide.
 Y. Yamamoto, H. Ishizu / Forensic Science Inrernarional 75 (1995) 85-93 91

Fig. 2 shows the results of electrophoresis of the second PCR products from the
specimens of three individuals. Each of the specimens were typed as Km*1,2/
Km*1,2, Km*l,2/Km*3 and Km*3/Km*3. The bands of target fragments were
recognized in the lanes of reactions @ and 0 in Km*1,2/Km*1,2, in those of
reactions @, 0 and 0 in Km*1,2/Km*3 and in those of reactions @ and 0 in
Km*3/Km*3.
Faint mis-primed bands also appeared in the lanes of the mismatch primer (with
mismatching allele-specific primers and specimen alleles) in the second PCR, i.e. in
the @ lane of each type or 0 lane of Km*1,2/Km*1,2. The appearance of these
mis-primed bands was influenced by the annealing temperature in the second PCA.
When the annealing temperature was 60C intense mis-primed bands appeared in
the lanes of reactions @ and 0 with mismatched primers and allele base sequences.
When the annealing temperature was raised to the optimal 63C allele specificity
increased, the mis-primed bands almost completely disappeared and the genotype
could be accurately identified (Fig. 3).
Direct sequencing of the first PCR product from two individuals who had been
typed as Km*3/Km*3 homozygotes by the new method revealed complete homol-
ogy with the published sequence of the Km*3 allele [4]. The sequence determined
for two Km*1,2/Km*1,2 homozygotic individuals was also identical to the Km*3
sequenceexcept for a single nucleotide substitution from G to C at the first base of
the codon 191 as reported [6]. These results show that the new method can
accurately differentiate the Km*3 and the Km*1,2 alleles.
By omitting the DNA extraction, we attempted direct genotyping by means of
PCR using 0.1 ,ul of whole blood, 0.1 ~1 of saliva and 3 mm of hair roots from ten
adults. We identified the Km genotype and the same genotype in different speci-
mens of the same individual (Fig. 4). In the second PCR of these specimens, 1 ~1
of a lo-fold dilution of the products of the first PCR was usually applied, but better
results were obtained using 1 ,~l of a loo-fold dilution of hair roots. Fig. 5 shows
the result of application of this method to a paternity case, in which a father-daugh-
ter relationship was affirmed with a probability of 99.91% by Essen-Mbllers

Whole blood Sal iva Hair root

(0.1 WI> (0.1 bI> (3 ml>

Fig. 4. Km genotyping from whole blood saliva and hair root samples obtained from a Km*1,2/Km*3
type individual.
92 Y. Yamamoto, H. Ishizu 1 Forensic Science International 7.5 (1995) 85-93

1 2 3

I1I
@~o@@cBo~@@lo@

3/3 192 / 3 192 / 3

O10
0
Fig. 5. Mendelian inheritance of Km alleles in a paternity case.

equation by 24 kinds of the conventional blood group tests including the HLA test.
The mother had Km*3/Km*3 type and the daughter had Km*1,2/Km*3 type and
it was considered that the Km*1,2 of the daughter had been transmitted from her
father. The Km genotype of the alleged father was Km*1,2/Km*3 type and he had
the Km*1,2 allele. There was no contradiction if this was the true father of the child
and the probability of paternity calculated on the Km genotype was 65.4%. Also,
no contradiction in the Mendelian hereditary law was found in the results of the
test of Km genotype in seven other paternity tests, in which paternity was
confirmed by testing genetic blood markers.
Table 4 summarizes the results of genotyping 115 individuals living in Okayama
Prefecture, Japan. The allele frequency was estimated at Km*3 = 0.739 and
Km*1,2 = 0.261. The distribution of the observed and predicted values of genotype
frequencies was in good agreement with Hardy-Weinberg equilibrium (x2 = 0.1596,
0.50 < P < 0.75). Using Inv(2) and Inv(3) antisera, Steinberg and Matsumoto [7]
have determined the Km serotypes of 748 unrelated Japanese from Osaka and
reported that the allele frequency was 0.302 in Inv2 and 0.698 in Inv3. Compared
 Y. Yamamoto, H. Ishizu / Forensic Science International 75 (1995) 85-93 93

Table 4
Distribution of the Km genotypes in Okayama Prefecture, Japan

Genotype Number observed (%) Number predicted Allele frequency

Km*3/Km*3 62 (53.91) 62.83 Km*3 = 0.739

Km*l,2 = 0.261
Km*l,2/Km*3 46 (40.00) 44.35 x2=0.1596
d.f. = I
Km*1,2/Km*l,2 7 (6.09) 7.82 0.50 <P < 0.75
Total 115 (100.00) 115.00

with the Inv3 frequency in Osaka, the Km*3 frequency of Japanese from Okayama
estimated by our method was somewhat lower.
Among the 115 cases studied here, we did not find the rare Km* 1 allele and,
therefore, the ability to detect the Km*1 allelic sequenceof our method has not yet
been proven. In northern European populations, the frequency of the Km*1 allele
is about 0.001-0.006 [S]. Judging from the data reported by Steinberg and
Matsumoto [7] that the distribution of the Inv or Km phenotypes among 176
offspring of 250 members of 87 Japanesefamilies agreed with the expectation based
on the assumption of the alleles InvZ and Inv3, the Inv allele must also be very rare
among Japanese.
The simple method of identifying Km genotypes by allele-specific amplification
with PCR, is sensitive and can achieve Km genotyping even from a very small
quantity of specimens without DNA extraction. Accordingly, it should be a useful
means of paternity testing and personal identification.

References

[I] C. Ropartz, J. Lenoir and L. Rivat, A new inheritable property of human sera: the InV factor.
Nature, 189 (1961) 586.
[2] C.P. Milstein, A.G. Steinberg, C.L. McLaughlin and A. Solomon, Amino acid sequence change
associated with genetic marker Inv(2) of human immunoglobulin. Nature, 248 (1974) 160-161.
[3] A.G. Steinberg, C.P. Milstein, C.L. McLaughlin and A. Solomon, Structural studies of an Inv(l,-2)
kappa light chain. Immunogenetics, 1 (1974) 108-117.
[4] H.G. Klobeck, A. Solomon and H.G. Zachau, Contribution of human V,,, germ-line genes to
light-chain diversity. Nature, 309 (1986) 73-76.
[5] P.A. Hieter, E.E. Max, J.G. Seidman, J.V. Maize1 Jr and P. Leder, Cloned human and mouse
kappa immunoglobulin constant and J region genes conserve homology in functional segments.
Cell, 22 (1980) 197-207.
[6] J.H. Kurth, A.M. Bowcock, H.A. Erlich, S. Nevo and L.L. Cavalli-Sforza, Km typing with PCR:
Application to population screening. Am. J. Hum. Genet., 48 (1991) 613-620.
[7] A.G. Steinberg and H. Matsumoto, Studies on the Gm, Inv, Hp and Tf serum factors of Japanese
populations and families. Hum. Biol., 36 (1964) 77-85.
[8] A.G. Steinberg and C.E. Cook, The distribution of the human immunoglobulin allotypes. Oxford
Monographs on Medical Genetics, Oxford University Press, Oxford, 1981.