Oncogene (2003) 22, 2452–2456
& 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00
www.nature.com/onc
Separation of long-range human TERT gene haplotypes by
transformation-associated recombination cloning in yeast
Jung-Hyun Kim1,2, Sun-Hee Leem1,2, Yangil Sunwoo2 and Natalay Kouprina1
1
Laboratory of Biosystems and Cancer, National Cancer Institute (NIH), Bethesda, MD, USA; 2Department of Biology, Dong-A
University, Pusan 604-714, Korea
The hTERT gene encoding a catalytic subunit of human
telomerase contains four blocks of variable number of
tandem repeats (VNTRs) – two in intron 2 and two in
intron 6. The segregation of hTERT VNTRs was analysed
in families, revealing that all of them were transmitted
through meiosis following a Mendelian inheritance. The
work reports a further characterization of the minisatellites in hTERT. We employed transformation-associated
recombination (TAR) cloning to isolate parental hTERT
alleles and determined the specific combination of
minisatellites at each of the polymorphic sites. A longrange haplotyping of hTERT determined by TAR cloning
was verified by classical Mendelian analysis. Since such a
strategy can be applied for any chromosomal locus, we
conclude that recombinational gene capture could greatly
facilitate haplotypes analysis.
Oncogene (2003) 22, 2452–2456. doi:10.1038/sj.onc.1206316
Keywords: hTERT; VNTR; polymorphism; haplotypes;
TAR cloning
Telomeres and their maintenance by telomerase comprise a biological clock that influences cellular lifespan
in mammals (Chiu and Harley, 1997). Telomerase
expression is confined primarily to the germ line and
regenerating tissues of the adult soma (Kim et al., 1994;
Yan et al., 1999; Yoshida et al., 1999). In other cells,
telomerase is turned off and telomeres progressively
shorten with each division. Once telomeres shorten
below a critical length, a DNA damage checkpoint is
activated and cellular senescence ensues. In most human
tumors, however, telomerase is reactivated and telomere
function is maintained indefinitely (Shay and Wright,
1999; Yoshida et al., 1999).
In a recent work, we described the isolation and
characterization of the complete genomic copy of the
human telomerase gene, hTERT (Leem et al., 2002).
Sequence analysis allowed to determine two blocks of
variable number of tandem repeats within intron 2
*Correspondence: N Kouprina, Laboratory of Biosystems and Cancer, National Cancer Institute, NIH, Bldg. 37, Room 5032, Bethesda,
MD 20892-4471, USA;
E-mail: [email protected]
Received 11 October 2002; revised 5 December 2002; accepted 10
December 2002
(VNTR 2-1st and VNTR 2-2nd), two within intron 6
(VNTR 6-1st and VNTR 6-2nd), and a nonpolymorphic
minisatellite within intron 12 (TR12) (Szutorisz et al.,
2001; Leem et al., 2002). We found that VNTRs 2-1st
and 2-2nd have minisatellites with 42 and 61 bp periods,
respectively. The number of repeats in the six alleles of
hTERT-VNTR 2-1st ranges from 40 to 111, with 40 the
most common. The number of repeats in the four alleles
of hTERT-VNTR 2-2nd ranges from 40 to 44, with 44
the most common, and a corresponding degree of
heterozygosity of 0.476. hTERT-VNTR 6-1st and 62nd have minisatellites with 38 and 36 bp periods,
respectively. hTERT-VNTR 6-1st has eight alleles with
18–38 copies of the repeat and a degree of heterozygosity of 0.62. In all, 30 alleles of hTERT-VNTR 62nd have 23–88 copies of the repeat, with 22 and 47 the
most common. The repeats diverge by approximately
10%. Analysis of segregation of the minisatellites in
families revealed that all the four VNTRs are transmitted
in a Mendelian inheritance pattern. Comparison of normal
and cancer tissue from patients revealed rearrangements of
at least two VNTRs, suggesting that the minisatellites
might be associated with activation of telo-merase
expression in cancer cells. The work reported here
describes further the characterization of hTERT-VNTRs,
including isolation of parental alleles of the human
telomerase gene using in vivo recombination in yeast.
In this study, we analysed seven families and
additional 26 unrelated individuals for VNTR polymorphisms and found five new minisatellite variants
(Tables 1 and 2), bringing the total number of identified
alleles for VNTR 2-1st and 6-2nd to eight and 33,
respectively. (The degree of heterozygosity was 0.595
and 0.951, respectively, based on the analysis of a total
of 126 unrelated individuals.) Analysis of the hTERT
sequence revealed a sixth minisatellite segment at the 30
end of the gene (positions 70167–70409 in AY007685)
that consisted of six repeat units of 40 bp each. Analysis
of the individuals showed that the minisatellite was not
polymorphic (data not shown). Segregation analysis of
hTERT VNTRs in seven new families revealed that all
the VNTR 2-1st, 2-2nd, 6-1st, and 6-2nd alleles were
meiotically stable and inherited in a Mendelian fashion
(Table 1), in agreement with previous data (Szutorisz
et al., 2001; Leem et al., 2002). Figure 1 illustrates the
hereditary segregation of four VNTRs during three
generations in family VII.
Separation of long-range human TERT gene haplotypes
Jung-Hyun Kim et al
2453
Table 1
Inheritance of polymorphic minisatellites of hTERT
Locus
Family
VNTR 2-1st
VNTR 2-2nd
VNTR 6-1st
VNTR 6-2nd
VI. Father
Mother
Child
78, 78
90, 78
78, 90
44, 44
43, 44
44, 43
27, 35
22, 22
27, 22
64, 62
71, 74
64, 71
VII. Grandfather
Grandmother
Father
Mother
Child 1
Child 2
40,
40,
40,
40,
78,
78,
44,
44,
44,
44,
44,
44,
22,
22,
22,
22,
22,
22,
78
78
78
85
85
40
44
44
44
42
42
44
22
22
22
35
35
22
64, 66
110, 105
64, 110
69, 60
110, 60
110, 69
New alleles of VNTR 2-1st and VNTR 6-2nd identified in families VI and VII are underlined and in bold. No new alleles of VNTRs were detected
in families I, II, III, IV, and V. Values represent number of repeats
Table 2 Polymorphism of hTERT-VNTRs
Locus
Subj. no.
VNTR 2-1st
VNTR 2-2nd
VNTR 6-1st
VNTR 6-2nd
78, 80
90, 78
40, 78
44, 44
42, 43
42, 44
23, 23
23, 23
23, 35
64, 64
62, 66
19, 19
8
14
19
26 unrelated individuals were analysed for VNTR polymorphisms. New alleles were identified in individuals 8, 14, and 19. New alleles of VNTR21st and VNTR6-2nd are underlined and in bold. Values represent number of repeats
a
1
2
3
4
5
6
b
1
2
3
4
5
6
3 kb
c
1 kb
Figure 1 Gel electrophoresis showing Mendelian inheritance in family VII of hTERT-VNTR 2-1st (a), VNTR 2-2nd (b), VNTR 6-1st
(c), and VNTR 6-2nd (d). We used PCR primers specific to VNTRs (Leem et al., 2002) to analyse VNTR length in DNA from family
members. Lane 1, paternal grandfather; lane 2, paternal grandmother; lane 3, father; lane 4, mother; lanes 5 and 6, two children. Two
previously unknown alleles of VNTR 6-2nd with the number of repeats equal to 105 and 110 were detected. The first and last lanes
correspond to ladders
We identified the hTERT-VNTR alleles and traced
their transmission in family VI (Table 1). That enabled
us to haplotype the child’s hTERT, that is, to link each
VNTR in the gene to its parental allele. The results
enabled us to propose the following combination of
VNTRs in the child: VNTR 2-1st, 78 repeats; VNTR 22nd, 44 repeats; VNTR 6-1st, 27 repeats; and VNTR 6-
2nd, 64 repeats for the paternally transmitted allele and
VNTR 2-1st, 90 repeats; VNTR 2-2nd, 43 repeats;
VNTR 6-1st, 22 repeats; and VNTR 6-2nd, 71 repeats
for the maternally transmitted allele. Since long-range
haplotyping by classical Mendelian analysis is subject to
interference by meiotic recombination, we re-investigated the proposed haplotypes by physically separating
Oncogene
Separation of long-range human TERT gene haplotypes
Jung-Hyun Kim et al
2454
the parental hTERT alleles. We accomplished that by
transformation-associated recombination (TAR) cloning (Larionov et al., 1996; Larionov et al., 1997;
Kouprina and Larionov, 1999), a method developed
for the selective isolation of chromosomal segments and
genes up to 250 kb long from complex genomes. For this
purpose, we constructed a TAR vector, pVC-TERT,
containing two unique targeting sequences homologous
to the 30 and 50 regions of hTERT (Figure 2a). To isolate
hTERT genomic regions, we presented the linearized
TAR vector (Figure 2a) to yeast spheroplasts along with
genomic DNA prepared from a blood sample taken
from the child (family VI). As shown in Figure 2b,
homologous recombination between the hooks in the
vector and homologous sequences in the hTERT
genomic fragment led to the rescue of the entire gene
as a circular yeast artificial chromosome (YAC). To
identify clones containing the hTERT gene, we used the
polymerase chain reaction (PCR) and screened the yeast
transformants for the presence of exon sequences using
specific primer pairs (http://www.cephb.fr/poltel). Six
clones were positive for all exon sequences (Figure 2c),
and we characterized each for VNTRs 2-1st, 2-2nd, 61st, and 6-2nd. Four contained VNTRs with lengths
corresponding to a maternal allele and two contained
VNTRs with lengths corresponding to a paternal allele
(Figure 3), permitting us to conclude with confidence
that the child inherited the parental alleles intact.
There are several indications that VNTR polymorphism affects gene expression (Paquette et al., 1998; Fuke
et al., 2001). The high level of polymorphism and
consequent heterozygosity pose a problem as to what
haplotypes are present in a given cell and which of them
is responsible for gene expression. The separated alleles
may provide good material for addressing this question.
So far, several approaches have been proposed for
separation of haplotypes in multiply heterozygous
individuals. Classically, haplotypes are resolved by
pedigree analysis. That strategy is limited, however,
because of the problem of collecting DNA samples from
appropriate family members and the interference of
meiotic recombination. Individual chromosomes can
a
b
HIS3
CEN6
pVC-TERT vector
5’
3’
X
X
hTER
ERT
196 bp
135 bp
200 kb
c
150 kb
100 kb
0.5 kb
50 kb
5’ 1
2
3
4
5 6
7
8 9 10 11 12 13 14 15 16 3’
Figure 2 Direct resolution of the hTERT haplotypes of an individual. (a) Diagram showing isolation of hTERT by TAR cloning
(Kouprina and Larionov, 1999). The method exploits a high level of recombination between homologous DNA sequences during
transformation in the yeast Saccharomyces cerevisiae. The TAR cloning vector pVC-TERT is a derivative of the basic TAR cloning
vector pVC604 (Kouprina and Larionov, 1999). The vector contains a 196 bp KpnI–ApaI fragment specific to a promoter region
(positions 18176–18371; Accession no. AY007685) and a 135 bp XhoI–EcoRI fragment from the 30 region (positions 69601–70500;
Accession no. AY007685). The targeting sequences were cloned into the polylinker of pVC604. For hTERT capture, genomic DNA
isolated from the child in family VI was transfected into yeast spheroplasts along with the TAR vector. Recombination between the
vector and the hTERT-containing DNA genomic fragment results in establishment of a circular yeast artificial chromosome (YAC).
The vector pVC-TERT was linearized by XhoI before use to release the targeting sequences. CEN corresponds to the yeast
chromosome VI centromere and HIS3 is a marker for selection in yeast. We have also used a radial TAR cloning method (Kouprina
et al., 1998) for isolation of the hTERT gene. In that case, the TAR vector contained a 30 hTERT-specific sequence and, as a second
targeting hook, an Alu human repeat (Leem et al., 2002). (b) Electrophoretic profile of two TAR YAC isolates containing different
alleles of the hTERT gene. Chromosomal-size DNA was isolated from transformants containing hTERT YACs, exposed to a low dose
of gamma rays (5 krad), separated by transverse alternating field electrophoresis, and blot hybridized with an hTERT-specific probe.
The signals at the starting wells correspond to large circular molecules. The bands at approximately 60 kb correspond to YAC DNA
molecules linearized by radiation. (c) PCR analysis of one of the YAC TAR isolates containing hTERT. Presence of the entire hTERT
gene in the YAC clone was confirmed by PCR with primer pairs specific for 16 hTERT exons and two additional pairs specific for 50
and 30 junction sequences. The PCR products were sequenced to confirm that they were truly hTERT exon sequences. The first and last
lanes correspond to ladders
Oncogene
Separation of long-range human TERT gene haplotypes
Jung-Hyun Kim et al
2455
a
1
2
3
4
5
3
4
5
b
1
2
3
1
2
3
4
5
4 kb
c
1
2
4
5
1.5 kb
Figure 3 Long-range haplotyping of the human TERT gene. We performed VNTR analysis of family VI members and two TAR
clones obtained from the child. The length of VNTRs was analysed with PCR primers specific to VNTRs (Leem et al., 2002). The
VNTR lengths in the TAR clones correspond to those of the maternal and paternal alleles. (a) VNTR 2-1st, (b) VNTR 2-2nd, (c)
VNTR 6-1st, and (d) VNTR 6-2nd. Lane 1, father; lane 2, mother; lane 3, child; lanes 4 and 5, respectively, maternal and paternal
alleles of the child. The first and last lanes correspond to ladders
also be separated by construction of hybrid cell lines,
microdissection of chromosomes (Ludecke et al., 1989),
or amplification of spermatocyte DNA (Li et al., 1988).
Those procedures, too, are laborious and time-consuming, and they cannot be used for large numbers of
individuals. Another proposed approach is based on
single-molecule dilution (SMD) of a genomic template
and amplification via biphasic PCR (Jeffreys et al., 1990;
Ruano et al., 1990). Though the SMD approach has the
advantage of obviating the need for a pedigree, it has
two major disadvantages. First, it is problematic to
obtain a single, large DNA molecule by dilution, and
second, single molecules are difficult to use for multiple
PCR reactions.
In this study, we demonstrated that TAR cloning was
a suitable approach for separating long-range haplotypes. Complete sequencing of TAR YAC isolates
containing parental hTERT alleles can help identify
other polymorphic regions, including single-nucleotide
polymorphisms and polymorphisms within VNTRs
themselves. Physical analysis of TAR isolates can be
simplified by retrofitting the circular YACs into
bacterial artificial chromosomes (BACs) with a BAC
cassette and then transferring them into E. coli cells
(Kouprina and Larionov, 1999). Recently, we found
that up to 15% DNA divergence does not preclude
efficient gene isolation by TAR (Noskov et al., 2003).
Such tolerance to DNA divergence allows the isolation
of highly polymorphic alleles with the same TAR
vector.
It is worth noting that TAR cloning could become a
routine procedure in any laboratory with adherence to a
few guidelines described in our previous publications
(Noskov et al., 2001, 2002) and a recent paper analysing
optimum conditions for selective isolation of genes from
complex genomes (Leem et al., 2003). Since TAR
cloning can be applied for any chromosomal region,
such a new strategy is well suited for a large-scale
analysis of long-range haplotypes in multiply heterozygous individuals to reveal haplotypes that contribute
to diseases.
Acknowledgements
We thank Miriam Bloom (SciWrite Biomedical Writing &
Editing Services) for professional editing. S-H Leem and J-H
Kim were partially supported by Grant No. R05-2000-00000151-0 from the Korea Science & Engineering Foundation.
References
Chiu CP and Harley CB. (1997). Proc. Soc. Exp. Biol., 14, 99–
106.
Fuke S, Suo S, Takahashi N, Koike H, Sasagawa N and
Ishiura S. (2001). Pharmacogenomics J., 2, 152–156.
Jeffreys AJ, Neumann R and Wilson V. (1990). Cell, 60, 473–
485.
Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD,
Ho PL, Coviello GM, Wright WE, Weinrich SL and Shay
JW. (1994). Science, 266, 2011–2019.
Kouprina N, Annab L, Graves J, Afshari C, Barrett JC,
Resnick MA and Larionov V. (1998). Proc. Natl. Acad. Sci.
USA, 95, 4469–4474.
Kouprina N and Larionov V. (1999). Current Protocols in
Human Genetics, vol. 1. Boyle AL, ed. John Wiley and Sons,
Inc.: New York, pp. 5.17.1–5.17.21.
Larionov V, Kouprina N, Graves J, Chen XN, Korenberg JR
and Resnick MA. (1996). Proc. Natl. Acad. Sci. USA, 93,
491–496.
Oncogene
Separation of long-range human TERT gene haplotypes
Jung-Hyun Kim et al
2456
Larionov V, Kouprina N, Solomon G, Barrett JC and
Resnick MA. (1997). Proc. Nat. Acad. Sci. USA, 94,
7384–7387.
Leem SH, Londono-Vallejo JA, Kim JH, Bui H, Tubacher E,
Solomon G, Park JE, Horikawa I, Kouprina N, Barrett JC
and Larionov V. (2002). Oncogene, 21, 769–777.
Leem SH, Noskov VN, Park JE, Kim SI and Kouprina N.
(2003). Nucl. Acids Res., 31, in press.
Li H, Gyllensten UB, Cui X, Saiki RK, Erlich HA and
Arnheim N. (1988). Nature, 335, 414–417.
Ludecke H, Senger G, Claussen U and Horsthemke B. (1989).
Nature, 338, 348–350.
Noskov V, Koriabine M, Solomon G, Randolph M, Barrett
JC, Leem SH, Stubbs L, Kouprina N and Larionov V.
(2001). Nucl. Acids Res., 29, E62.
Noskov V, Kouprina N, Leem SH, Koriabine M, Barrett JC
and Larionov V. (2002). Nucl. Acids Res., 30, E8.
Oncogene
Noskov VN, Leem SH, Solomon G, Mullokandov M, Chae JI,
Yoon YH, Shin YS, Kouprina N and Larionov V. (2003).
J. Mol. Evol., 56, in press.
Paquette J, Giannoukakis N, Polychronakos C, Vafiadis P and
Deal C. (1998). J. Biol. Chem., 273, 14158–14164.
Ruano G, Kidd KK and Stephens JC. (1990). Proc. Natl.
Acad. Sci. USA, 87, 6296–6300.
Shay JW and Bacchetti S. (1997). Eur. J. Cancer, 33, 787–791.
Szutorisz H, Palmqvist R, Roos G, Stenling R, Schorderet DF,
Reddel R, Lingner J and Nabholz M. (2001). Oncogene, 20,
2600–2605.
Yan P, Saraga EP, Bouzourene H, Bosman FT and Benhattar
J. (1999). J. Pathol., 189, 207–212.
Yoshida R, Kiyozuka Y, Ichiyoshi H, Senzaki H, Takada H,
Kioki K and Tsubura A. (1999). Anticancer Res., 19,
2167–2172.