Molecular analysis of Tripterygium hypoglaucum (level) Hutch-induced mutations at the HPRT locus in human promyelocytic leukemia cells by multiplex polymerase chain reaction
Sheng Xue Liu
Department of Hygiene Toxicology, College of Preventive Medicine, Third Military Medical University
People's Republic of China
The genotoxicity and cytotoxicity of a Chinese medicinal herb, Tripterygium hypoglaucum (level) Hutch (THH), was investigated in human promyelocytic leukemia (HL-60) cells using the hypoxanthine-guanine phosphoribosyltransferase mutation assay. THH showed clear cytotoxicity and mutagenicity in HL-60 cells at concentrations between 6.7 and 20.0 mg/ml. When the mutants were characterized by techniques based on multiplex PCR, 46.6% of induced mutants were found to have deletions, whereas only 7.7% of spontaneous mutants showed deletions. The rest were not characterized, but were assumed to be mainly point mutations. Mapping of all intragenic deletion breakpoints showed a random distribution of breakpoints in nine exons. Deletion of exon 1 appeared as the only whole gene deletion, while deletions of exon 7/8 and 9 often occurred concomitantly (71.4%). It is concluded that THH is mutagenic in HL-60 cells, predominantly inducing deletions. Since this herb is widely used as a traditional medicine, its genotoxicity should be assessed in vivo in treated humans.
PCR (Saiki et al., 1985; Park et al., 1995) has been used for
the analysis of exon deletions in different human cells (Gibbs
et al., 1989, 1990), Chinese hamster cells (Yang,J.-L. et al.,
1989; Rossiter et al., 1991) and other cell lines (Yu et al.,
1992; Elisabetta et al., 1995; Pluth et al., 1998). These studies
have indicated a wide hypoxanthine-guanine
phosphoribosyltransferase gene (hprt) mutation spectrum in various
mammalian cell lines induced by physical and chemical mutagens.
This technique is, therefore, a good method to understand in
more detail the molecular mutation spectrum.
Tripterygium hypoglaucum (level) Hutch (THH) is a
traditional Chinese herb belonging to the genus Celastraceae.
Its main chemical components are alkaloids, terpenes and
pigments. THH has been used widely in traditional Chinese
medicine for the treatment of various human autoimmune
diseases, such as rheumatic arthritis, lupus erythematosus,
hyperthyroidism, psoriasis and so on. It has also been reported
that THH shows antitumor activity (Luo et al., 1988;
Wang,S.M. et al., 1989). In their basic studies on THH,
Wang,S.M. et al. (1989) confirmed that THH has a strong
ability to induce chromosomal non-disjunction, chromosomal
aberrations and aneuploidy in mice. Wang et al. (1993) found,
in addition, that THH can induce C-mitotis, malsegregation
and sister chromatid exchange (SCE) in mice. In our laboratory
we used fluorescent in situ hybridization (FISH) with mouse
minor centromeric and telomeric DNA probes and CREST
antibodies to study the chromosomal composition of
micronuclei (MN). We found that 6070% of MN induced by
THH contained whole chromosomes and, in addition, that
THH showed a very strong ability to induce apoptosis in
Chinese hamster embryo, mouse NIH3T3 and human
lymphoma Jurkat cell lines (Cao et al., 1997, 1998; Cao and
Nusse, 1999). It is interesting that Jurkat tumor cells were
found to be more sensitive (~10- to 20-fold) in terms of
apoptosis as compared with non-tumor cells. All these results
indicated that THH has an ability to induce chromosomal
damage and aneuploidy and that it is also an inducer of
apoptosis. There is, however, no previous information on the
ability of THH to induce gene mutations in mammalian cells.
In this paper, the multiplex PCR molecular analysis method
for HPRT gene mutations in human promyelocytic leukemia
cells (HL-60) was used to analyze the mutation spectra induced
by THH and the mechanism of genotoxicity.
Materials and methods
Water extracts from THH
The dry root of THH was provided by Kunming Medicine Co. (Yunnan,
Peoples Republic of China). Samples of 20 g of the herb were kept in 400 ml
of distilled water overnight and then boiled three times, the extract was
concentracted to 30 ml and the sediment was removed by centrifugation and
filtration (1 ml of water extract is equal to 0.67 g THH). At present, the
chemical components of this water extract are thought to include only
alkaloids, terpenes and pigments.
HL-60 is a human acute promyelocytic leukemia cell line described earlier
by Collins et al. (1978). HL-60 cells were maintained as an asynchronous,
exponentially growing population in RPMI 1640 medium (Sigma, St Louis,
MO) supplemented with 10% fetal bovine serum (SJQ, Hangzhou, Peoples
Republic of China), 100 U/ml penicillin (Sigma), 100 g/ml streptomycin
(Sigma) and 2 mM L-glutamine (Gibco, Carlsbad, USA) at 37C in an
atmosphere of 5% CO2. Before treatment the cells were incubated for 1 day
in complete medium supplemented with 106 M aminopterin (Gibco), 104 M
hypoxanthine (Sigma) and 105 M thymidine (Sigma) (HAT culture medium)
to remove pre-existing HPRT mutants that cannot live in HAT culture medium.
Then the medium was replaced with complete medium supplemented with
105 M thymidine and 104 M hypoxanthine. Two days later, this medium
was removed and the cells were incubated in normal medium for 710 days
To measure the cytotoxicity of THH, exponentially growing HL-60 cells were
treated with different concentrations of THH in culture medium for 4 h. Initial
cell numbers per treatment were fixed at 5.010 6 cells. Sterile distilled water
was used as a negative control and N-ethyl-N-nitrosourea (Shanren, Tokyo,
Japan) was used as a positive control. At the sampling time the cells were
harvested and washed twice with D-Hanks medium (Hanks buffer without
Ca2 and Mg2 ) at 37C and afterwards diluted in normal culture medium.
The cells were counted, diluted and transferred to 96-well microwell plates
(Gibco) at an average of 1 cell/200 l medium/well. After incubation for 7
days, wells containing colonies were counted and the plating efficiency (PE)
PE [ln(no. of negative wells/no. of all wells)]/[no. of cells per well].
After expression of gene mutations (8 days), approximate cell numbers per
treatment dose were 4.8510 7, 3.6510 7, 3.1210 7, 2.3110 7, 1.0110 7
Primer sequence (53)
and 0.6910 7 in the 0, 3.33, 6.67, 10.00, 13.33 and 20.00 mg/ml groups,
respectively. For cloning efficiency (CE) 1 cell/well was transferred to the
96-well plates and for assay of mutant frequency (MF) 110 4 cells were
added to each well in 200 l medium with 1 g/ml 6-thioguanine (6-TG)
(Sigma). Plates were analyzed for colony presence 7 days after seeding for
CE and 8 days after for MF.
[ln(no. of negative wells/no. of all wells)]/[no. of cells per wellCE].
Screening, extension and DNA isolation
A single positive clone was transferred from the 96-well plate to a 24-well
microwell plate (Gibco) to continue culture for an additional 12 days. Each
well contained 1 ml screening medium including 2 g/ml 6-TG. Some of the
cloned cells were then transferred to a new 24-well plate which contained
HAT culture medium at 103 cells/well and cultured for 13 days. If the cloned
cells in a well were obviously dead they were identified as mutated clones
and the remaining cloned cells in the original 24 wells were transferred to
culture bottles for extension expression. DNA isolation and purification
from wild-type cells and HPRT mutant cells were performed according to
Design, synthesis and appraisal of primers
Eight pairs of oligonucleotide primers were designed by computer software
with a small modification according to Wei et al. (1996). The synthesis and
appraisal of the eight pairs of primers were completed by different laboratories
(Beckman Co., Beijing; Cybersyn B.J., USA; Institute of Cellular Biology of
Chinese Academy of Science, Shanghai).
Table I shows the sequences of the eight pairs of oligonucleotide primers.
Exons 7 and 8 were amplified simultaneously using the same primers, because
they are only 163 bp apart. All primers except the exon 1-specific ones
enabled amplification of the corresponding exons by multiplex PCR. It was,
however, difficult to include exon 1 primers within the remaining set of all
primers without spurious synthesis of a non-specific signal. In our preliminary
experiments with several primer pairs in one PCR reaction it was difficult to
control and optimize the reaction conditions. In addition, insertions and
deletions within exons could occur, therefore we restricted the number of
primer pairs in a single PCR reaction in order to confirm the distances of
PCR products according to their molecular weights. This reduced the number
of false negative and false positive results. Therefore, after several preliminary
experiments, the eight pairs of primers were divided into three groups: one
multiplex PCR included exons 2, 5, 6 and 7/8, the second included exons 3,
4 and 9 and exon 1 was amplified separately. Seventy-one mutants were
analyzed by this multiplex PCR method.
For amplification of HPRT exons, genomic DNA template (3650 ng) was
mixed with 50 pmol of each primer pair in a total reaction volume of 50 l
containing 50 mM KCl, 10 mM TrisHCl (pH 8.8), 0.31.05 mM MgCl2,
0.2 mM dNTPs and 2.5 U AmpliTaq DNA polymerase (Shenggong, Shanghai,
Peoples Republic of China). After initial denaturation of the template DNA
at 98C for 7 min, a total of 40 PCR cycles were performed with denaturation
at 94C for 1.5 min, annealing at 52C for 1.5 min and extension at 72C for
2.0 min. Exon 1 was synthesized individually under modified conditions: a
total of 30 PCR cycles were performed with denaturation at 95C for 0.5 min,
Fig. 1. The relationships between plating efficiency (PE), mutant frequency
(MF), cloning efficiency (CE), deletion percentage (DP) and THH dose.
annealing at 64C for 1.0 min and extension at 72C for 1 min. The last cycle
was finished with a 7 min extension at 72C. The PCR product (10 l) was
used for analysis by 3% agarose gel electrophoresis or by PAGE.
Cytotoxicity and mutagenicity of THH
Figure 1A shows the cytotoxicity of THH to HL-60 cells. The
PE gradually decreased with increasing concentration of THH.
The doseresponse relationship could be expressed by the
equation y 96.3e0.03x (P 0.01). There was a significant
effect on PE at THH concentrations of 5 mg/ml.
Figure 1B shows the mutagenicity of THH in HL-60 cells.
A linear increase in MF with increasing concentration of THH
was found. This doseresponse relationship could be expressed
by the equation y 3.34 0.76x (P 0.01). At 6.67
20 mg/ml THH, MF was 10.6- to 43.8-fold that in the control
Tripterygium hypoglaucum Hutch-induced mutations
No. showing PCR changes
No. showing no change
Spontaneous THH (mg/ml) 3.33 6.67
0.05 versus control group.
Multiplex PCR analysis
Thirteen spontaneous and 58 THH-induced HPRT mutants
were characterized by multiplex PCR. According to the
electrophoresis pattern of PCR products, 43 (60.6%) of 71 mutants
analyzed were found to exhibit no abnormal band in any of
the nine exons. This indicated that these mutants had point
mutations and not exon deletion or insertion. In 21 of 71
mutants there were less than eight bands for each locus, which
showed partial deletion of exons. The remaining seven mutants
had no PCR products, which meant that all exons studied were
deleted. Of all mutants analyzed, 39.4% (28 of 71) had partial
or whole deletions.
Molecular spectrum of HPRT gene
Table II shows the changes of spontaneously derived and
THH-induced mutants at the HPRT locus. The electrophoresis
patterns of mutants mainly consisted of three types: the normal
pattern including point mutations, total deletions and partial
deletions. THH-induced mutant cells (6.6720 mg/ml) showed
mutation spectra that were significantly different from the
spectra of spontaneous mutations. No spontaneous mutants
showed total exon deletions, while THH-induced mutants did.
The proportions of deletion mutations were very different
between spontaneous and THH-induced mutants. About 25
50% of mutations found in THH-induced mutants were
deletions while the proportion in spontaneous mutants was only
7.7%. The proportion of the normal pattern was very high
(92.3%) in spontaneous mutants, compared with only 5075%
in THH-induced mutants. A clearer doseresponse relationship
was seen in induction of partial and whole deletion mutation
than in induction of total mutations.
Analysis of deletion breakpoints
Figure 2 shows the distribution of the deletions in the nine
exons of the HPRT gene found in the 71 mutants analyzed
during this experiment. Neither an obvious difference among
absolute numbers of mutations in the nine exons nor a clear
hot-spot were found. Deletion mutations were found in all
nine exons of the HPRT gene, while single deletions per
mutant only in exon 1, 7/8 or 9 were not found. Deletion in
exon 1 was observed only when total gene deletion occurred,
and most of the deletions in exons 7/8 and 9 were concomitant
(linked deletions, 71.4%).
It has been shown that THH not only affects tumor growth,
but also induces non-disjunction, aneuploidy and chromosomal
aberrations. Recently, Cao and Nusse (1999) reported that a
water extract of THH could also induce apoptosis in cultured
cells in vitro. However, no evidence has so far been reported
Fig. 2. Schematic diagram of the distribution of deletions (black bars)
within the nine exons of the human HPRT gene. (A) Distribution of deleted
exons in spontaneous mutants; (B) distribution of deleted exons in
THHinduced mutants. *, Intragenic deletion; **, intragenic insertion.
on the mutagenicity of THH. Our study has demonstrated that
water extracts of THH are cytotoxic and mutagenic to cultured
HL-60 cells in vitro. The present results on the induction of
HPRT mutations by THH may help in understanding the
pharmacological and toxicological effects of THH and further
suggest that the use of this herb may have a genotoxic risk.
Based on 13 spontaneous and 58 THH-induced HPRT
mutants characterized by multiplex PCR, distinct differences
in the mutation spectra were found between control and
induced mutants. Among the 13 spontaneous mutants no total
deletion mutations were found and only one mutant showed a
partial deletion. It is known that spontaneous mutation at the
HPRT locus in many kinds of cells mainly involves point
mutations (Tomita et al., 2000), which could not be
distinguished using the multiplex PCR method alone. However,
2550% of THH-induced mutants had exon deletions.
It is of interest that the highest fractions of partial and
whole deletions were not found at the highest concentration
(20 mg/ml) but at lower concentrations (1013.3 mg/ml).
Yamada et al. (1996) also reported a similar effect when
studying X-ray-induced HPRT gene mutations in primary
human skin fibroblasts. They found that the highest fraction
of whole deletions did not appear at the highest dose (4 Gy)
but at a lower dose (2 Gy). The reason for this effect is
probably that higher doses (or in our case higher concentrations
of THH) induce serious exon deletions so that these cells are
not able to survive.
We have reported that THH induces a high frequency of
MN harboring whole chromosomes at all concentrations tested
(3.33, 6.67 and 13.33 mg/ml) and produces a dose-dependent
increase in fragment-containing MN, indicating that THH has
both aneugenic and clastogenic potential (Yang and Cao,
2001). We intend to further characterize the THH-induced
mutants by DNA sequencing, to better understand the
mutagenic mechanism of THH. Various studies have shown that
THH has multiple genotoxic potential, inducing gene
mutations, chromosome breakage and aneugenic events. It is,
therefore, important to study whether genotoxic effects can be
detected in patients treated with THH.
The authors would like to thank Dr. Makoto Hayashi of NIHS, Japan and Dr.
J. Fitzgerald of Department of Human Services of South Australia for their
helpful comments on this manuscript. This research was supported by NSFC
contract 39970650, 30100153 and 30100241.