The viral thymidine kinase gene as a tool for the study of mutagenesis in
Trypanosoma brucei
The viral thymidine kinase gene as a tool for the study of mutagenesis in Trypanosoma brucei
Jesús
Valdés
+
,
Martin C.
Taylor
[sect]
,
Michael A.
Cross
,
Marjolijn J. L.
Ligtenberg
,
Gloria
Rudenko
and
Piet
Borst*
Division of Molecular Biology, The Netherlands Cancer Institute, Plesmanlaan
121, 1066 CX
Amsterdam
,
The Netherlands
Received March 4, 1996;
Revised and Accepted April 1, 1996
ABSTRACT
We have tested the use of thymidine kinase as a negative selection system for
Trypanosoma brucei.
To this end we have targeted a construct containing a Herpes simplex virus
thymidine kinase (TK) gene into the ribosomal DNA array of procyclic
T.brucei
. This resulted in TK activity 30-50-fold above background and in susceptibility to the nucleoside
analogues ganciclovir, ethyl-deoxyuridine and 1-[2-deoxy,2-fluoro-8-D-arabinofuranosyl]-5-iodouracil, all of which have no effect on wild-type trypanosomes.
TK
+
trypanosomes, however, reverted to a ganciclovir resistant phenotype at a rate of 10
-6
per cell-generation. A similar reversion rate was observed using the Varicella-zoster virus TK gene. Loss of TK activity was not due to detectable
DNA rearrangements or a decrease in TK mRNA. Sequence analysis of the revertant
genes demonstrated, however, the occurrence of point mutations and frameshifts. One revertant line had a mutation in the thymidine
binding site leading to the substitution of a conserved arginine by a glycine.
Other mutations included single base insertion, single base deletion and the
introduction of a premature termination codon by point mutation.
INTRODUCTION
Trypanosoma brucei
sp. is a parasitic protozoan responsible for sleeping sickness in man and
Nagana in domestic animals. In recent years, the study of this organism at the
molecular level has been facilitated by the development of both transient and
stable transfection techniques. Stable transformation has been obtained with
positively selectable marker genes encoding resistance to the antibiotics
neomycin/geneticin (
1
), hygromycin (
2
) and phleomycin/bleomycin (
3
) and in both major life-cycle stages (
4
). As yet, however, no negatively selectable marker has been shown to work in
T.brucei.
The thymidine kinase-thymidylate kinase (TK) gene from the Herpes simplex type 1 virus (HSV-1) has been widely used as a negatively selectable marker in
mammalian transfection systems. The TK gene has been used to enrich for
correctly targeted integrations of DNA into mammalian genomes and for
elimination of certain cell types from a population. It is the promiscuity of
the TK enzyme that allows it to be used as a negatively selectable marker. The
enzyme will phosphorylate a wide variety of modified nucleosides and nucleoside
analogues which remain unphosphorylated in cells lacking the viral TK. The
phosphorylated analogues then act as competitive inhibitors of DNA polymerase
or as DNA chain terminators. These analogues include the antiviral therapeutic
agents acyclovir and ganciclovir, and less well-known compounds such as 1-[2-deoxy,2-fluoro-8-D-arabinofuranosyl]-5-iodouracil (FIAU). The HSV-1 TK gene has been
demonstrated to work as a negative selection system in conjunction with
ganciclovir in
Leishmania major
, a protozoan related to
T.brucei
(
5
).
The experiments described below show that a viral TK gene can be used as a
negatively selectable marker in African trypanosomes. However, the gene is
readily inactivated by mutation. This has allowed us to estimate the minimal
rate of mutation in this organism to be 10
-9
per base pair per cell generation. The rate of mutation may have implications
for the evolution of the variant surface glycoprotein (VSG) genes on which the
trypanosome depends for survival.
MATERIALS AND METHODS
Constructs
Plasmid pSPIR
[alpha][beta]:
The [alpha][beta] tubulin intergenic region was obtained by PCR with primers: 5'-gctCTAGAAAGTGTGACAACGTC and 5'-gccggtcGACTATTTTCTTTGATG (lower case
letters indicate bases added to produce useful restriction sites). The fragment was
cloned into the
Xba
I and
Sal
I sites of pSP72 (Promega).
Plasmid pSPIR
[alpha][beta]
*
:
pSPIR[alpha][beta] was digested with
Sma
I and
Xba
I. The
Xba
I end was made blunt with Klenow enzyme, and the plasmid religated.
Plasmid pSN
: The 1.2 kb partial
Pst
I fragment of pRK20(+) containing the
neo
gene and the [beta][alpha] intergenic region of the tubulin array (
6
), was cloned into the
Pst
I site of pSPIR[alpha][beta]*.
Plasmid p
[beta][alpha]
TKO
: The 176 bp
Spe
I fragment from pSPir[beta][alpha] (
7
) containing the [alpha]-tubulin
trans
-splice acceptor site was cloned into the
Xba
I site of pTKO (a kind gift from Dr Hein te Riele, Netherlands Cancer
Institute), which contains the HSV-1 TK gene.
Plasmid pSTKN
: The [alpha]-tubulin splice acceptor-TK fragment was isolated from p[beta][alpha]TKO by digestion with
Hin
dIII followed by end repair with Klenow enzyme, and a further digestion with
Eco
RI. The fragment was subcloned into the
Eco
RI and
Sma
I sites of pSN.
Plasmid pBTKN
: A 4.3 kb
Eco
RI-
Bgl
II fragment bearing the
T.brucei
ribosomal promoter from plasmid pR4 (
8
) was cloned into the
Eco
RI and
Bgl
II sites of pSTKN.
Plasmid pVTK
: This construct is analogous to pBTKN but contains the VZV TK gene and the
hygromycin resistance selectable marker. It was derived from pBluescript KS(+)
and has the following sequences inserted in the polylinker (5' -> 3'): a 4.3 kb
Eco
RI-
Bgl
II fragment bearing the
T.brucei
ribosomal promoter (as in pBTKN); the 402 bp intergenic region between actin
genes 1 and 2 as splice acceptor site; the VZV TK gene; the 270 bp [beta][alpha] intergenic region of the tubulin array as polyadenylation/splice
acceptor site; the hygromycin resistance gene; the 330 bp [alpha][beta] intergenic region of the tubulin array as polyadenylation site. A
detailed description of how this plasmid was constructed can be obtained from
the authors upon request.
Growth and transformation of trypanosomes
All studies were carried out using strain 427-60 cultured procyclic forms of
T.brucei
(
9
). The trypanosomes were cultured in a semi-defined medium as described (
9
). Constructs pBTKN (see Fig.
1
and below) and pVTK were linearised by digestion with
Cla
I. Trypanosomes (2.5 * 10
7
) were electroporated with 5 [mu]g of linearised pBTKN or pVTK DNA as described (
10
). Transformants were selected by growth in medium supplemented with 20 [mu]g/ml G418 (Gibco-BRL) for pBTKN and with 20 [mu]g/ml hygromycin for pVTK and thereafter continuously maintained on
selection. Transfection efficiency was measured and transformant lines were
cloned, by limiting dilution, mixing in wild-type trypanosomes during the initial phase of culturing. For measurement
of growth rates cloned lines were inoculated at 2.5 * 10
5
cells/ml. Growth was measured as the absorbance at 550 nm, of 500 [mu]l aliquots taken at regular time intervals. Nucleoside analogues were stored
as concentrated aqueous stock solutions at -20oC. The lyophilised sodium salt of ganciclovir was obtained from
Sarva-Syntex Nederland B.V, Rijswijk; FIAU was a kind gift of Dr Anton Berns,
The Netherlands Cancer Institute, Amsterdam.
Luria-Delbrück fluctuation test
The rate of reversion to a TK
-
phenotype was measured using the Luria-Delbrück fluctuation test essentially as described (
11
). Stock cultures of different transformants were inoculated into medium
containing 20 [mu]g G418 per ml at 10
3
cells/ml. For each transformant, 2 * 10
2
trypanosomes were seeded in each of 150 wells in microtitre plates, and a bulk
culture containing the same number of trypanosomes as had been seeded in the
entire microtitre plate was set up as a control. Cultures were incubated at 28oC for 3-6 days depending on the cell line, until the density reached ~1-2 * 10
6
trypanosomes per well.
Replica cultures were then set up in which the entire contents of the wells were
transferred to 2.5 ml of fresh medium supplemented with 20 [mu]g G418 per ml and 100 [mu]M ganciclovir in 24 well plates. At this stage, the trypanosomes from
the control bulk culture were also put on G418 and ganciclovir selection and
were divided between 150 cultures. These cultures should contain equal numbers
of mutants thus resulting in a similar rate of growth to each other in the
presence of ganciclovir, and providing a suitable endpoint to the test. The
plates were sealed with parafilm and incubated at 28oC. Growth was scored by absorbance of 0.5 ml aliquots at day 10-12; for those wells showing no growth additional checks were done at
days 14, 16 and 18. The frequency of reversion was then assessed by
substituting experimental results into the equation:
{roman a} = {{- l n ^ P o . l n {roman 2}} over N}
where
P
0
is the proportion of wells without cell growth and
N
is the final number of trypanosomes per culture upon addition of ganciclovir.
For the VZV TK transformant, the fluctuation test was performed as described
above except that 20 [mu]g/ml hygromycin and 20 [mu]g/ml FIAU were used to select for TK
-
mutants.
Thymidine kinase assay
Thymidine kinase activity was measured in crude soluble protein preparations.
Trypanosomes (50 ml) in late logarithmic phase culture were pelleted and washed
in cold PSG (0.59 M Na
2
HPO
4
, 0.03 M NaH
2
PO
4
, 0.044 M NaCl, 1% w/v glucose). All procedures in lysate preparation after this
stage were carried out at 4oC. The cell pellet was resuspended in 1 ml 50 mM sodium phosphate buffer
(pH 7.5). Seventy-five [mu]l 10% NP-40, 1.5 [mu]l 100 mM
N
[alpha]-
p
-tosyl-L-lysine chloromethylketone (TLCK) and 1.5 [mu]l 1 M dithiothreitol were added and the suspension shaken
several times until clear. The lysate was freeze-thawed three times in liquid nitrogen. It was then centrifuged for 15 min
at top speed in a microfuge and the supernatant was cleared in a Beckman SW
50.1 rotor at 30 000
g
for 35 min. The protein content was assayed using the Bradford method (BioRad).
Lysates were stored at -70oC until required.
For the TK assay the volume of lysate equivalent to 10 [mu]g protein was mixed in a final volume of 25 [mu]l with 200 mM Tris-HCl, pH 7.5, 10 mM NaF, 2 mM MgCl
2
, 5 mM ATP and 40 [mu]M [
3
H]thymidine (6.7 Ci/mmol, NEN-Dupont). All assays were done in duplicate. The reaction was incubated for
30 min at 37oC. Two 11 [mu]l aliquots were taken from each tube and spotted onto half-discs of Whatman DE81 filters. One half was dried immediately and
the second washed three times in ethanol. The percentage of label remaining
bound to the second filter is then a measure of the thymidine kinase activity
in that sample. The length of time of incubation and the amount of protein used
were shown to be within the linear range of the assay in control experiments,
as was the ATP dependency of the measured activity (data not shown).
DNA and RNA analyses
DNA and RNA preparation, blotting and hybridisation were performed as described
(
10
,
13
). Probes were prepared by random priming. Blots were washed at a stringency of
0.1* SSC, 0.1% SDS at 65oC.
Rescue of plasmid DNA from trypanosome lines
Total genomic DNA from the transformed lines was digested with
Cla
I. The DNA was diluted to 200 ng/ml to enhance the chance of intramolecular
ligation, and circularised with T4 DNA ligase (Boehringer Mannheim). The
ligated DNA was introduced into
Escherichia coli
DH5[alpha] by electroporation as described (
12
), using a Gene Pulser with pulse controller (BioRad). Ampicillin resistant
colonies were picked and plasmid DNA isolated by standard methods (
13
). Colinearity of rescued plasmids with input DNA was checked by restriction
enzyme analysis. The rescued plasmids were retransfected into wild-type trypanosomes as above. After selection cloned lines from each
transfection were assayed for TK enzyme activity and for expression of TK mRNA.
Sequence analysis of plasmid DNA
DNA sequencing was performed using the dideoxy chain termination method with the
Sequenase
TM
enzyme as described by the manufacturer (US Biochemicals).
PCR amplification from trypanosome genomic DNA and linear amplification
sequencing
Regions identified as containing putative mutations by plasmid sequencing were
amplified and sequenced directly using the polymerase chain reaction (PCR). The
amplification conditions were as follows: 2 [mu]g genomic DNA was mixed with 1* PCR buffer containing 200 [mu]M dNTPs, 3.0 mM MgCl
2
, 5% dimethylsulphoxide and 160 ng of each primer in a total volume of 50 [mu]l. This mix was incubated at 94oC for 5 min, then allowed to cool to 50oC for 2 min after which 0.5 U
Taq
polymerase was added to each tube. The tubes were incubated at 72oC for 1 min and then taken through 25 cycles of 30 s at 94oC, 30 s at 45oC, 1 min at 72oC and a final cycle of 1 min at 94oC, 1 min at 55oC and 5 min at 72oC. The product was precipitated with PEG-8000 for linear amplification
sequencing (LAS).
For the LAS reaction the DNA pellet was resuspended in 10 [mu]l H
2
O and divided between the four dideoxynucleotide reactions. For this reaction
the primer was end-labelled using T4 polynucleotide kinase and [[gamma]-
32
P]ATP. The sequencing reaction was performed as described (
13
). The cycling conditions were: 30 cycles of 1 min at 94oC, 1 min at 55oC and 1 min at 72oC. The reactions were fractionated on 7% polyacrylamide
sequencing gels and autoradiographed at -70oC.
RESULTS
Stable integration and expression of TK gene
The TK construct was designed with the
neo
gene downstream of the HSV-1 TK gene such that the trypanosomes could not switch off transcription of
TK without losing G418 resistance (Fig.
1
). The cells were continuously maintained on low-level G418 selection (20 [mu]g/ml) to prevent repeat-mediated excision of the construct, or gene conversion. The
trans
-splicing and polyadenylation signals used were from the tubulin gene array
since these signals allow RNA processing in both life-cycle stages. The ribosomal DNA array was used as the recipient site since
this is constitutively active and gives high-level expression of integrated genes (
8
).
The
Cla
I linearised construct pBTKN was targeted to the ribosomal array of insect-form trypanosomes by standard transfection procedures (
10
). As a control the pUCTbNeo3 construct, containing only a
neo
gene, was transfected into wild-type trypanosomes (
1
,
15
). The pBTKN construct integrated at a frequency of 1.3 * 10
-4
per cell as measured by outgrowth of dilution series. This was somewhat less
efficient than the integration of our standard control construct pUCTbNeo3
targeted to the tubulin array but was as expected for a single-crossover integration (
1
,
15
). Four of the lines obtained (5.3, 5.7, 5.8 and 5.31) were analysed in detail.
Correct targeting of the construct to a ribosomal array in each cell line was
confirmed by Southern hybridisation (data not shown).
Growth and reversion of parasites in the presence of toxic nucleoside analogues
The growth of wild-type trypanosomes was unaffected by concentrations of ganciclovir <= 1 mM (data not shown). The TK
+
transformants, in contrast, were inhibited and the degree of growth inhibition
was dependent upon the concentration of ganciclovir. After an interval,
however, trypanosomes began to grow out at a rate which matched the wild-type, suggesting reversion of the cells to a TK
-
phenotype. This phenomenon was also observed with other nucleoside analogues
(ethyl deoxyuridine and FIAU; data not shown), thus suggesting that the reversion mechanism is independent of the type of nucleoside analogue involved. When these resistant trypanosomes
were seeded into fresh medium containing ganciclovir and G418 there was no lag
phase and the rate of growth was identical to that of wild-type trypanosomes (Fig.
2
B). Growing the resistant trypanosomes in the absence of ganciclovir selection
for four passages did not result in a return to ganciclovir sensitivity
suggesting that the reversion was a permanent mutation rather than a transient
adaptation.
Levels of TK mRNA and TK activity
RNA blotting (Fig.
3
) showed the expected 2.1 kb transcript in all transfectants including those
that had reverted to a TK
-
phenotype. In addition to this, a second higher molecular weight transcript was
observed which corresponds to an unspliced precursor RNA. This RNA was also
seen by Rudenko
et al.
(
8
) with other plasmid transformants containing this version of the ribosomal
promoter. Interference with splicing could be due to secondary structure
formation in the 5' part of the primary transcript, which is derived from the rDNA unit and
therefore not normally spliced, blocking access of the spliceosome to the
tubulin splice acceptor site.
Sequence analysis of inactivated TK genes
The TK genes from four revertants were sequenced and the results are summarised
in Table
1
. Each gene was found to contain a mutation which should inactivate the gene
product. Since we also identified mutations that arose during the rescue in
E.coli
, the trypanosomal origin of the mutations was confirmed by direct linear
amplification sequencing from trypanosome genomic DNA or by sequencing the
entire ORF and confirming that only one mutation was present (data not shown).
.
Mutations found in the revertant-derived TK genes
Mutation
Position in gene
Effect on TK protein
C deleted
212
Frameshift at codon 70
C -> G
487
Arg -> Gly in thymidine binding site
C inserted
532
Frameshift at codon 177
C -> T
748
Creation of stop codon at codon 250
The TK genes from the revertant lines were rescued and sequenced. When a
putative inactivating mutation was identified this region of the TK gene was
amplified and sequenced directly from trypanosomal genomic DNA to prove that
the mutation was present in the revertant cells and not introduced during post-rescue manipulation in
E.coli
. The nucleotide positions are numbered from the adenine of the translational
start codon.
The most interesting mutation was that found in revertant 5.7r3. In this case a
C -> G transversion had occurred within the sequence encoding the thymidine
binding site of the enzyme. This change resulted in the substitution of an
arginine residue that is absolutely conserved in the TK genes of all members of
the herpesvirus family studied so far, by a glycine. A similar mutation in
which this arginine is replaced by glutamine has been shown to correlate with
acyclovir resistance and loss of TK activity in Varicella Zoster virus (
18
,
19
). In our (limited) analysis no common mutation was found although all mutations
were single base-pair changes.
Measurement of inactivation rate
No data are available on the natural mutation rate in trypanosomes or related
organisms and it would therefore be of interest to determine the rate of
inactivation by the Luria-Delbrück fluctuation test. The results are shown in Table
2
. This analysis indicated that both transformant lines tested reverted to a TK
-
phenotype at approximately the same rate, giving a mean value of 8.9 * 10
-7
cell-generation
-1
. The similarity of the reversion rates of two individual cell lines suggests
that the inactivation mechanism is characteristic for the TK gene in
trypanosome DNA and not influenced by incidental factors such as the particular
ribosomal array targeted. The inactivation rate could, however, be due to some
specific property of the HSV-1 TK gene. Since this gene is unusually GC rich (65%; ref.
20
), it is possible that the trypanosomal DNA polymerase might have problems
replicating it. To address this possibility, we tested the Varicella-Zoster virus TK gene (VZV TK) which has a GC content of 46% (
21
), similar to the 48% mole GC reported for
T.brucei
(
22
), and is of similar length to the HSV-1 TK gene.
Inactivation of trypanosomes expressing a VZV TK gene
Trypanosomes were transformed with a construct designed similarly to pBTKN but
containing the VZV TK gene and the hygromycin resistance selectable marker.
This DNA was also targeted to the ribosomal array and transformant cell lines
were characterised for correct integration (data not shown). Cells expressing
VZV TK were found to be resistant to ganciclovir but sensitive to FIAU. This
result was not unexpected since VZV is much less susceptible than HSV-1 to acyclovir, a compound closely related to ganciclovir (
23
). As had been observed for trypanosomes expressing the HSV-1 TK gene, the VZV TK transformants reverted back to a FIAU resistant
phenotype and grew out at a rate which matched the wild-type, indicating mutation of the VZV TK gene (data not shown).
To see if the inactivation rate of the two different TK genes in
T.brucei
was the same, we performed a fluctuation test on both transformants in
parallel. FIAU instead of ganciclovir was used to select TK
-
mutants as both VZV and HSV-1 TK expressing cell lines were equally sensitive to this nucleoside
analogue. Both transformant lines reverted to FIAU resistance at a similar rate
(Table
2
). The reversion rate of HSV-1 TK cell line TKN5.3 was found to be slightly lower in this experiment
than in the previous one, being 1.2 * 10
-7
cell-generation
-1
as opposed to 8.3 * 10
-7
cell-generation
-1
. However, different nucleoside analogues were used for the selection of
revertants in the two experiments (FIAU as opposed to ganciclovir), and this
may influence the rate estimate to a minor extent.
.
Experimental data for the Luria-Delbrück fluctuation analysis of reversion in TK transformed trypanosome
lines
TK transformant
TKN 5.3
TKN 5.7
TKN 5.3
VTK 1
Selective agent
Ganciclovir
Ganciclovir
FIAU
FIAU
Initial number of trypanosomes,
N
i
200
200
200
200
Final number of trypanosomes,
N
f
2.7 * 10
6
1.1 * 10
6
7 * 10
6
6.6 * 10
6
Total number of cell generations per culture,
3.9 * 10
6
1.6 * 10
6
1 * 10
7
9.5 * 10
6
(
N
f
- N
i
)
[divide] ln 2
Number of experimental cultures
120
120
90
96
Cultures without growth
5
28
28
13
Proportion of cultures with no mutants,
P
o
0.04
0.22
0.31
0.14
Average number of mutants per culture,
3.22
1.51
1.17
1.97
m
= -ln
P
o
Rate of reversion,
8.3
*
10
-7
cell-
9.4
*
10
-7
cell-
1.2
*
10
-7
cell-
2.1
*
10
-7
cell-
m
[divide]
total number of generations
generation
-1
generation
-1
generation
-1
generation
-1
The rate of reversion to ganciclovir- or FIAU-resistance for HSV-1 TK (TKN) and VZV TK (VTK) transfectants was calculated using
the Luria-Delbrück fluctuation test equation described in Materials and Methods.
DISCUSSION
We have targeted the herpesvirus TK gene into the ribosomal array of
T.brucei
. The gene is initially expressed in a functional form which renders the
trypanosome susceptible to antiviral nucleoside analogues. The activity of TK
found in our transformants is consistently much higher than the activity of the
putative trypanosomal TK homologue in our assay system. Since the enzyme assay
is performed on crude soluble protein extracts, which do not contain detectable
TK inhibitors, it is unlikely that the trypanosomal TK activity measured is
significantly underestimated.
To our knowledge, our experiments provide the first measurement of the mutation
rate in a kinetoplastid protozoan. We think that the rate measured here
reflects the mutation rate of trypanosome genes in general. Our arguments for
this are as follows. All transformant lines tested gave similar growth curves
in toxic nucleoside analogues (data not shown), and the fluctuation test showed
similar rates of inactivation in two different lines. It is unlikely,
therefore, that inactivation is a consequence of the integration of the TK gene
into a particular ribosomal array influencing adjacent genes. Although we
cannot rule out that the position of the TK gene in the genome affects the
mutation frequency, it seems unlikely that ribosomal arrays are more prone to
point mutation than any other locus. The growth curves showing reversion to
resistance were similar with different nucleoside analogues and different
concentrations of any one nucleoside, and reversion occurred with every
nucleoside tested. This indicates that the rate of mutation is dependent on the
number of generations and not due to specific mutagenic effects of a particular
nucleoside analogue. It is also unlikely that high level expression of TK
results in an imbalance in dNTP pools at the replication fork because of
increased synthesis of dTMP and dCMP. Evidence from other systems suggests that
the dCTP:dTTP ratio rather than the absolute pyrimidine dNTP concentration is
important and as long as both nucleotide levels increase proportionally there
is no mutagenic effect (
24
).
The high GC content of the HSV-1 TK gene does not seem to contribute to the rate of mutation. None of the
mutations identified occur within poly G runs and therefore polymerase slippage
does not appear to cause the mutations. Moreover, the less GC rich VZV TK gene
is inactivated at a similar rate. It is therefore probable that we are
measuring the basal mutation rate in the trypanosome and not a gene-specific event.
The overall mutation rate in an organism is commonly expressed as the number of
mutations per base pair per cell generation. This is estimated by dividing the
rate of inactivation of a given gene by the length of that gene. In
E.coli
, inactivation of genes of the
lac
and
his
operons gives a mutation rate of 5 * 10
-10
(
25
). Using the same approach, we estimate the mutation rate of
T.brucei
to be ~10
-9
per base pair per cell generation. Drake (
25
) observed that a diverse set of microbes, which use DNA to encode their genes
and which includes the eukaryotes
Saccharomyces cerevisiae
and
Neurospora crassa
, exhibit widely varying spontaneous mutation rates per base pair. However, when
expressed as mutations per genome, the rates are remarkably similar at 0.003
mutations per genome per DNA replication. Drake concluded that such a common
mutation rate must have been shaped in response to evolutionary forces of a
very general nature. Based upon the inactivation of a TK gene, we calculate the
mutation rate in
T.brucei
to be 12-fold higher, at 0.037 mutations per genome per DNA replication. Why such a
high mutation rate has arisen during the evolution of
T.brucei
is unclear, but it may be related to the requirement for diversity among the
VSG genes on which the trypanosome depends for survival. In this context it is
of interest that Lu
et al
. (
26
) have reported evidence for a highly mutagenic DNA copying process in African
trypanosomes. This process is limited to the duplicational transposition of one exceptional
VSG gene in bloodstream form trypanosomes and is thought to involve mutation
rates in excess of 10
-2
per copying event. This is clearly unrelated to the basal mutation rate studied
here.
We have nevertheless tested whether the TK gene can be used as a negative
selection system for studies of the mechanism of VSG expression site switching.
We have inserted the HSV-1 TK gene in the active VSG expression site in bloodstream form
T.brucei
, and have found that it is possible to select trypanosomes which have changed
their VSG coat in the presence of nucleoside analogues (unpublished results).
Thus, in spite of the high inactivation frequency, it appears that the TK
system will prove useful not only for the study of recombination and gene
conversion but also in the analysis of the regulation of gene expression in
African trypanosomes.
ACKNOWLEDGEMENTS
We thank our colleagues Dr Steve Hajduk, Dr Ronald Plasterk, Dr Wilbert Bitter,
Fred van Leeuwen and Pat Blundell for critical reading of the manuscript. We
are grateful to Drs Anton Berns and Hein te Riele of the Netherlands Cancer
Institute for their gifts of FIAU and the HSV-1 TK gene respectively, and to Dr E. Littler (Wellcome Research
Laboratories) for the VZV TK gene. We also thank Dr Piet van de Putte for
discussions about mutation frequencies. This work was funded in part by The
Netherlands Foundation for Chemical Research (SON) with financial aid from The
Netherlands Organisation for Scientific Research (NWO). J.V. was funded by the
E.E.C, M.C.T. was funded by a Wellcome Trust travelling fellowship, and M.A.C.
was funded by an EMBO long term fellowship.
REFERENCES
1 ten Asbroek, A.L.M.A., Ouellette, M. and Borst, P. (1991) Nature348, 174-175.MEDLINE Abstract
2 Lee, M.G-S. and van der Ploeg, L.H.T.M. (1991) Science250, 1583-1587.MEDLINE Abstract
3 Jefferies, D., Tebabi, P., Le Ray, D. and Pays, E. (1993) Nucleic Acids Res. 21, 191-195.MEDLINE Abstract
4 Carruthers, V.B., van der Ploeg, L.H.T.M. and Cross, G.A.M. (1993) Nucleic Acids Res. 21, 2537-2538.MEDLINE Abstract
5 LeBowitz, J.H., Cruz, A. and Beverley, S.M. (1992) Mol. Biochem. Parasitol. 58, 321-325.
6 Zomerdijk, J.C.B.M., Kieft, R. and Borst, P. (1991) Nature353, 772-775.MEDLINE Abstract
7 Ligtenberg, J.L.M., Bitter, W., Kieft, R., Steverding, D., Janssen, H., Calafat, J. and Borst, P. (1994) EMBO J. 13, 2565-2573.MEDLINE Abstract
8 Rudenko, G., Chung, H-M.M., Pham, V.P. and van der Ploeg, L.H.T.M. (1991) EMBO J. 10, 3387-3397.MEDLINE Abstract
9 Brun, R. and Schonenberger, M. (1979) Acta Trop. 36, 289-292.MEDLINE Abstract
10 Zomerdijk, J.C.B.M., Ouellette, M., ten Asbroek, A.L.M.A., Kieft, R., Bommer, A., Clayton, C. E. and Borst, P. (1990) EMBO J. 9, 2791-2801.MEDLINE Abstract
11 Luria, S.E. and Delbrück, M. (1943) Genetics28, 491-511.
12 Zabarovsky, E.R. and Winberg, G. (1990) Nucleic Acids Res. 18, 5912.MEDLINE Abstract
13 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbour Laboratory Press, Cold Spring Harbour, NY.
18 Roberts, G.B., Fyfe, J.A., Gaillard, R.K. and Short, S.A. (1991) J. Virol. 65, 6407-6413.MEDLINE Abstract
19 Saywer, M.H., Inchauspe, G., Biron, K.K., Waters, D.J., Straus, S.E. and Ostrove J.M. (1988) J. Gen. Virol. 69, 2585-2593.MEDLINE Abstract
20 Wagner, M.J., Sharp, J.A. and Summers, W.C. (1981) Proc. Natl Acad. Sci.78, 1441-1445.MEDLINE Abstract
21 Davison, A.J. and Scott, J.E. (1986) J. Gen. Virol. 67, 1759-1816.MEDLINE Abstract
22 Borst, P. and Fairlamb, A.H. (1976) In van den Bossche, H. (ed.), Biochemistry of Parasites and Host-Parasite Relationships. Elsevier/North-Holland Biomedical Press, Amsterdam, pp. 169-191.
23 Machida, H. (1986) Antimicrobial Agents and Chemotherapy29, 524-526.MEDLINE Abstract
24 Kunz, B.A. and Kohalmi, S.E. (1991) Ann. Rev. Genet. 25, 339-359.
*
To whom correspondence should be addressed
Present addresses:
+
Departamento de Fisiologia, CINVESTAV-México, Apartado Postal 14-740, México DF 07000, México,
[sect]
Applied Molecular Biology Unit, Department of Medical Parasitology, London
School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK and
[Dagger]
Department of Human Genetics, Postbus 9101, 6500HB Nijmegen, The Netherlands
