Acta Tropica, 52(1992)175-187 © 1992 Elsevier Science Publishers B.V. All rights reserved 0001-706X/92/$05.00
175
ACTROP 00249
Evolution of the retrotransposons TRS/ingi and of the tubulin genes in trypanosomes R i c h a r d Braun a, K a t h r y n Behrens a, Andreas Glauser a and Reto Brun b alnstitute for General Microbiology, University (?f Bern, Baltzer-Strasse 4, CH-3012 Bern, Switzerland and bSwiss Tropical Institute, Socinstrasse 57, CH-4002 Basel, Switzerland
(Received 25 June 1992: accepted 9 September 1992) The African trypanosomes have genomes of high plasticity, as demonstrated for instance by their ability to shuffle their genes around, coding for variant-specific surface glycoproteins (VSGs). Another indication of their genome plasticity is the presence of multiple retroelements. The retrotransposon-like element TRS/ingi is present in many copies in the genome of trypanosomes. One particular derivative of TRS/ ingi, called TUBIS, had previously been found to interrupt a tubulin gene in a particular strain of T. brucei. Here both TRS/ingi and TUBIS were studied by hybridizing genomic DNA of various strains and species of trypanosomes with suitable probes in order to elucidate the evolution of this family of retroelements. The TSR/ingi elements are highly repeated and have very long open reading frames, while TUBIS clearly is a truncated, inactivated form of this element, found in only one particular chromosomal location. Both elements were shown to be present in several strains and species of the subgenus Trypanozoon, in particular in T. brueei hruceL T. gambiense, T. rhodesiense, T. equiperdum and T. evansi. They could not be detected in species of other subgenera, in particular in T. congolense and T. cruzi. These findings suggest that the retrotransposon TRS/ingi was acquired by trypanosomes only after divergence of present day subgenera. The TUBIS element was found in exactly the same chromosomal location (at the 3' end of the tubulin gene cluster) in many different strains and species of the subgenus Trypanozoon. This shows that the element was transposed to this location before speciation of the subgenus. Although TRS/ingi is unlikely to be involved directly in VSG switching, it may have contributed to the genome plasticity of trypanosomes. Key words: Trypanosoma brucei; Retrotransposon; Evolution; Tubulin genes; TUBIS; Genome organization
Introduction T h e A f r i c a n t r y p a n o s o m e s o f the s u b g e n u s T r y p a n o o z o o n are a h i g h l y successful g r o u p o f parasites. M a n y m o l e c u l a r m e c h a n i s m s a l l o w t h e m to e v a d e the h o s t defenses, n o t a b l y t h e i r ability to f r e q u e n t l y c h a n g e t h e i r s u r f a c e c o a t by n e w l y a c t i v a t i n g o n e o f a l a r g e n u m b e r o f v a r i a n t - s p e c i f i c s u r f a c e g l y c o p r o t e i n genes. T h e s e V S G switches, the p r e s e n c e o f a large n u m b e r o f m i n i c h r o m o s o m e s a n d the f r e q u e n t c h a n g e s in c h r o m o s o m e size testify to a h i g h d e g r e e o f g e n o m i c plasticity (for r e v i e w s see Borst, 1986; V a n d e r P l o e g , 1990). C o r r e l a t e d w i t h this plasticity is the p r e s e n c e o f a c o n s i d e r a b l e n u m b e r o f m o b i l e genetic e l e m e n t s , w h o s e e x i s t e n c e in g e n e r a l has Correspondence to." R. Braun, Institut ffir Allgemeine Mikrobiologie, Universit~it Bern, Baltzer-Strasse 4, CH 3012 Bern, Switzerland.
176
not found a satisfactory explanation in any organism. They may function to increase the mutation rate in their carrier species and thereby provide them with a selective advantage. Genetic elements resembling retrotransposons have been found in many species of eukaryotes from yeast to man (Fanning and Singer, 1987; Kingsman and Kingsman, 1988; Doolittle et al., 1989). In trypanosomes, too, several different retroelements have been found. One group resembles retrons, poly-A tailed mRNAlike sequences with flanking repeats at either end. To this group belong the RIMEs, 500 nucleotide long elements found in the ribosomal RNA genes of Trvpanosoma brucei, the causative agent of sleeping sickness in Africa (Hasan et al., 1984) and an element of similar structure recently found in the glyceraldehyde-3-phosphate dehydrogenase gene cluster of T. cruzi, the parasite of Chagas disease in South America (Kendall et al., 1990). A second group of retroelements is like the human LINES (Singer and Berg, 1991) to be placed in the retrotransposon category, since they are much larger, namely 5 7 kb long, have short insertion site duplications and one or two very long open reading frames resembling those of retroviruses. This group contains the SLACS, with all nine copies exclusively in the spliced leader sequence gene cluster (miniexon gene cluster) of T. brucei gambiense (Aksoy et al., 1987; Aksoy et al., 1990), and also the TRS/ingi of T. brucei brucei. TRS/ingi was discovered at the same time by two different groups and given the name TRS for Trypanosome Repetitive Sequence by one of the groups (Murphy et al., 1987) and ingi (Kiswahili for many) by the other group (Kimmel et al., 1987). The element occurs in about 100-200 copies per genome. It is 5.2 kb long and shows marked sequence similarities with reverse transcriptase, tether, RNase and Zn-fingers (Doolittle et al., 1989). There is a half of a RIME element (about 250 nucleotides) at each end of TRS/ingi. In the context of our long-standing interest in the tubulin gene cluster of T. brucei (Seebeck et al., 1983; Imboden et al., 1986; Imboden et al., 1987) we have become involved with TRS/ingi. We recently showed that one particular member of the TRS/ ingi family interrupts the tubulin gene cluster of T. brucei at its 3' end (Affolter et al., 1989). It is located in the middle of a/~-tubulin gene, thereby of course inactivating this gene into a pseudogene. We have called this element of the TRS/ingi family TUBIS for tubulin insertion sequence. The fact that TUBIS interrupts an otherwise intact/Ltubulin gene is a good indication that TRS/ingi elements actively transposed at some time in the past. In the following we address the question of whether the insertion of TUBIS in the tubulin gene cluster is strain-specific or whether it is found in different strains and different species. The answer to this question will bear on the evolution of this retroelement.
Materials and Methods
Growth of trypanosomatids Procyclic trypomastigotes of Trypanosoma brucei brucei, T.b. rhodesiense and T.b. gambiense were grown at 27°C in SDM-79 supplemented with 10% heat inactivated (56°C for 30 rain) fetal bovine serum (Gibco) and 100 units per ml penicillin and 100 ~tg per ml streptomycin (Brun and Schoenenberger, 1979). When not specified, the strain of T.b. brucei used was STIB 366; occasionally S 427 was employed.
177 The two strains of T.b. rhodesiense used were STIB 389 and STIB 704, both isolated from sleeping sickness patients in Tansania; the T.b. gambiense cultivated was strain TH-1/78E(031), isolated from a patient in the Ivory Coast. Trypanosoma congolense procyclic trypomastigotes, clone STIB 223V isolated in 1971 from a hyaena in Tansania, were cultured at 27°C in SM medium buffered with 25 mM Hepes and supplemented as in the case of T. brucei with serum and antibiotics (Cunningham, 1977). Trvpanosoma evansi strain STIB 779 (KETRI 2443), obtained from Dr. E. Zweygarth, Nairobi, Kenya, and Trypanosoma equiperdum strain BoTat-78, obtained from Dr. H. Eisen (Roth et al., 1989) or strain STIB 818, isolated in China, were propagated in mice or rats. The blood was obtained by cardiac puncture and the trypanosomes separated from the blood cells by DEAE 53 chromatography. Epimastigotes of Trypanosoma cruzi Y strain were grown in a simple monophasic medium (Mattei et al., 1977) at 27°C.
Analytical procedures DNA from protozoa was isolated as described by Seebeck et al. (1983), except that the CsCI step was omitted. From the tsetse fly, Glossina morsitans centralis, DNA was prepared as follows. The puparium of 30 pupae was cut open and discarded, and the soft tissue lysed in 4 ml buffer (100 mM Tris-HC1 (pH 8.0), 50 mM NaCI, 50raM EDTA, 1% SDS, 0.15mM spermine, 0 . 5 m M spermidine). To this 20gl proteinase K solution (10 mg/ml) were added and the mixture incubated 3 h at 37°C. Before and after incubation the mixture was passed through a Potter homogenizer. This treatment was followed by one extraction each with phenol, phenol/chloroform/ isoamyl alcohol and chloroform/isoamyl alcohol. The DNA was then precipitated with alcohol and re-dissolved in 3 ml TE buffer. Restriction enzyme digestion, gel electrophoresis, transfer, labelling by nick translation and hybridization were done under standard conditions (Maniatis et al., 1982). For subcloning the Bluescript vector BS + was used. As a size marker in electrophoreses, bacteriophage lambda DNA cut with HindIII was used (Boehringer Mannheim). The hybridization solution contained 50% formamide, 0.25 M NaH2PO4, 0.25 M NaC1, 7% SDS, 1 mM EDTA, pH 7.2. Hybridization was done overnight at 37°C on Zeta-Probe blotting membranes after 1 h prehybridization. Thereafter, the filters were rinsed briefly in 2 x SSC and washed at room temperature for 15 rain each with 2 x SSC/0.1% SDS and 0.5 x SSC/0.1% SDS.
Hybridization probes The following plasmids, also marked in Fig. 1, were used as hybridization probes: pTBtu9B covers an entire 3.6 kb repeat of the tubulin coding and the intergenic regions from one EcoRI site to the next. There is one single EcoRI site per tubulin repeat (Seebeck et al., 1983). pTRS1.6-BS 1.2 covers the reverse transcriptase region from nucleotide 2583 (SalI) to nucleotide 3753 (BamHI) of TRS 1.6 (Murphy et al., 1987). pTBtu9BI covers the first two-thirds of a /~-tubulin gene from nucleotide 125 (PstI) to nucleotide 906 (PstI) of T. brucei, as numbered by Kimmel et al. (1985). pTBtu3.3 (9CV) covers the first 330 nucleotides of TUBIS from nucleotide 48 (HaelII) to nucleotide 378 (HindIII) as numbered by Affolter et al. (1989).
178 pTBlu 9B [
P
A. I
I
N
H
I
E
I
I
I
B-tubulin I
(1 -tubulin
M
pTBlu 3.4
P
B. J
I
N
I Nc I
H
I
I
B-pseudogene I
M
I
pTBIu 3.5
II H I
TUBIS
I
I
Ha
Hf
I
I
M
I
Hf
I pTBIu 3.3
I
1 kb
I
Fig. [. Maps of the tubulin gene cluster and its 3' end in T. brucei. Restriction sites, P: Pstl, N: Narl, H: HindIII, M: MboI, E: EcoRI, Nc: Ncol, Hf: Hinfl, Ha: HaeIlI. The blocks represent coding regions or putative coding regions. The brackets over and below the maps indicate cloned sequences used as probes in hybridization experiments. The clone pTBtu9B is not shown: it covers an entire 3.6 kb repeat from one EcoRI site to the next. Also not shown is a TRS clone. (A) A single 3.6 kb repeat coding for ~- and //-tubulin. The intergenic region between ~ and/~ is arbitrarily shown at the 3' end of the gene. (B) The 3' end of the tubulin gene cluster. The TRS/ingi-like element TUBIS is found after 901 nucleotides of/~-tubulin pseudogene. pTBtu3.5 (9CII) covers 1025 nucleotides o f T U B I S from nucleotide 425 (Hinfl) to nucleotide 1450 (HinfI) as n u m b e r e d by Affolter et al. (1989). pTBtu3.4 ( 9 C I I I ) covers 220 nucleotides o f T U B I S from nucleotide 158 (NeoI) to nucleotide 378 (HindIII) as n u m b e r e d by Affolter et al. (1989). The n u m b e r i n g o f T U B I S starts a r b i t r a r i l y at nucleotide 868 o f the /%tubulin p s e u d o g e n e (Affolter et al., 1989).
Results
The tubulin gene cluster In T. brucei brucei the tubulin genes have been shown to be a r r a n g e d in a single large cluster with a b o u t l0 repeats o f 3.6 kb length each (Seebeck et al., 1983; T h o m a s h o w et al., 1983). Restriction enzymes that cut only a few times in the repeat give a simple digestion pattern, when genomic S o u t h e r n blots are p r o b e d with a cloned tubulin sequence. F i g u r e 2 confirms this result for T. brucei brucei and shows that a very similar p a t t e r n is o b t a i n e d with a second strain o f T. brucei as well as with T. rhodesiense a n d T. gambiense. The t r y p a n o s o m a l species T. evansi too shows virtually the same pattern, while the p a t t e r n for T. cruzi is very different ( F i g u r e 3). This leads to the conclusion that the tubulin genes are a r r a n g e d in a very similar gene cluster within the g r o u p o f T. brucei species. T h e flanking regions o f the tubulin cluster are also a r r a n g e d very similarly in diverse m e m b e r s o f the T. brucei group, as is seen from the results o f restriction m a p p i n g experiments. In the 5'-flanking region there is an E c o R I site 12 kb u p s t r e a m o f the end o f the coding region in all
179
A
B
Fig. 2. Hybridization of genomic DNAs with a tubulin probe. DNA was digested with EcoRI. Lane l, T. hrucei strain 366; lane 2, T. hrucei strain 427; lane 3, T. ~amhiense; lane 4, lambda HindllI marker: lane 5, T. rhodesiense strain 389; lane 6, T. rhodesiense strain 704: lane 7, 7". gamhiense. The tubulin gene repeat of 7". brucei is marked with an arrowhead. (A) Ethidium bromide stained gel. (B) Hybridization with //-tubulin probe pTBtu9Bl.
species tested (Fig. 2). In both strains o f T. rhodesiense examined there is a second cutting site at 10 kb. The two different size T. rhodesiense restriction fragments are together about as intense on the a u t o r a d i o g r a p h s as the single fragment in T. brucei or in T. gambiense. These observations are most easily explained by the presence of two different allelic E c o R I sites in the 5'-flanking region o f the tubulin gene cluster of T. rhodesiense. In the T-flanking region T. brucei and T. evansi show similar patterns when genomic D N A is cut with HindIII or E c o R I and when the transferred D N A is probed with cloned tubulin sequences (Fig. 3). The next HindlII site is a b o u t 0.3 kb downstream o f the cluster, i.e., 0.3 kb beyond the ,q-tubulin pseudogene, giving a fragment o f about 0.8 kb. This same band was observed with T. brucei strains 366 and 427, T. gambiense and T. rhodesiense strains 389 and 704 (data not shown). After this site, the next following HindIII site is another 5 to 6 kb further downstream as seen by hybridization with pTBtu3.5 (data not shown). The first E c o R I restriction site in the 3'-flanking region o f the cluster is 3 kb beyond the/3-tubulin pseudogene. This manifests itself as a 6 kb band in Figs. 2 and 3. This same band also lights up with a TUBIS-specific probe (see later). The D N A o f T. equiperdum also shows a very similar restriction pattern in the tubulin coding area as T. brucei. The repeat is also 3.6 kb long and is cut within the cluster to the same size fragments by the restriction enzymes BgIII and PstI, as is the D N A o f T. brucei (data not shown).
180
Fig. 3. Hybridization of genomic DNAs with a tubulin probe. DNA was digested either with EcoRl (lanes 1, 3, 5 and 7) or with HindlII (lanes 2, 4, 6 and 8). Lanes 1 and 2, T. brucei; lanes 3 and 4, T. cruzi; lanes 5 and 6, T. evansi; lanes 7 and 8, Glossina morsilans. Hybridization with fl-tubulin probe pTBtu9Bl. Note a band of 0.8 kb in lanes 2 and 6 marked with an asterisk: this derives from the 3' flanking region of the cluster and according to sequence data from T. brucei is 785 nucleotides long. The tubulin gene repeat is indicated by an arrowhead. The position of the lambda HindIII markers is given on the right. This clearly d e m o n s t r a t e s that the t u b u l i n genes are arranged in the same way in T. equiperdum as in T. brucei and the other species of the subgenus T r y p a n o z o o n . The TRS/ingi retrotransposon Previous studies on the d i s t r i b u t i o n of the a b u n d a n t T R S / i n g i r e t r o t r a n s p o s o n family in different species of t r y p a n o s o m e s were confirmed a n d extended. P r o b i n g genomic blots of restriction enzyme cut D N A shows that T R S / i n g i is present in multiple copies in T. brucei a n d also in T. evansi, but it c a n n o t be detected either in T. cruzi or in the insect vector Glossina morsitans (Fig. 4). Neither could it be detected in T. congolense (data not shown). This mobile element therefore seems to be confined to the subgenus t r y p a n o z o o n . However, its genomic s u r r o u n d i n g s are quite different in different species. The patterns obtained with T. brucei a n d T. evansi D N A show that the two species can easily be differentiated using a T R S / i n g i D N A probe. This shows that T R S / i n g i sequences can be used for species identification in the same
181
1 2
3
4 ..J--..£
m
m
V'l Fig. 4. Hybridization of genomic DNAs with a TRS/ingi probe. DNA was digested with EcoRI (lanes 1, 3, 6 and 8) or HindIII (lanes 2, 4, 7 and 9). Lane 5, lambda HindIII marker. The DNA came from the following species: lanes I and 2, T. brucei; lanes 3 and 4, T. cruzi; lanes 6 and 7, T. evansi; lanes 8 and 9, Glossina morsitans. Hybridization was with pTRS1.6-BS 1.2 from the putative reverse transcriptase region of TRS/ingi.
way as o t h e r r e p e a t e d D N A elements are used as d i a g n o s t i c tools for t r y p a n o s o m e s (Hide et al., 1990) a n d m a n y o t h e r parasites (for review see Barker, 1990). TUBIS
The m a i n p o i n t o f this p a p e r is to study the e v o l u t i o n a r y stability o f T U B I S , the T R S / i n g i - r e l a t e d insertion in the 3' end o f the tubulin gene cluster. As m e n t i o n e d earlier, T U B I S was originally f o u n d to be located in the middle o f the l a s t / % t u b u l i n gene o f the cluster, t u r n i n g this gene into a t r u n c a t e d / 3 - t u b u l i n p s e u d o g e n e , at least in a p a r t i c u l a r strain o f T. brucei. A l t h o u g h T U B I S a n d T R S / i n g i are clearly related,
182 the difference in their sequences is large enough to allow differential hybridization. Blotting experiments with a TUBIS-specific probe tell one clearly, that there are far fewer T U B I S sequences than TRS/ingi elements. This is seen when comparing genomic D N A hybridizations using either cloned T U B I S and TRS/ingi sequences as probes (Figs. 4 and 5A). W h e n hybridizing with the T U B I S probe pTBtu3.4, genomic digests o f 4 different species, each show one and the same predominant band. With E c o R I this band is at 6 kb, which agrees with the results shown in Fig. 2: there this same band was labelled with a tubulin probe and thereby defined the 3' end o f the cluster. The c h r o m o s o m a l locations o f the other, minor bands hybridizing with the T U B I S probes are not known. Pulsed field gel electrophoreses indicate that they are not in minichromosomes (data not shown). It has already been shown above, that restriction sites in the flanking regions of the tubulin cluster are quite similar in various species o f the subgenus t r y p a n o z o o n to those in the strain o f T. brucei, in which T U B I S was originally characterized. This already points to the probable presence o f T U B I S in other species than T. b r u c e i brucei. In the following we will describe Southern blotting experiments where
A 1 2 3
4 5 6 7 891011
B 1 2
"k
Fig. 5. Hybridization of genomic DNAs with TUBIS probes. (A) DNA was digested with EcoRl (lanes 1, 3, 5, 9 and ll) or HindIIl (lanes 2, 4, 6, 8 and 10). Lane 7, lambda HindIll marker. DNA came from the following species: lanes I and 2, T. brucei, lanes 3 and 4, T. congolense; lanes 5 and 6, T. evansi; lanes 8 and 9, T. rhodesiense; lanes 10 and 1 l, T. gambiense. Hybridization was with the TUBIS probe pTBtu3.4. (B) Separate gel electrophoresis. DNA from T. evansi was digested with HindIII (lane 2). Marker: lane I. Hybridization with the TUBIS probe pTBtu3.3, which includes part of RIME A. Note that in all the HindIII digests a band of 0.8 kb, marked by an asterisk, is visible. This is or corresponds to the 785 nucleotide joining fragment sequenced in T. brucei.
183
sequences were specifically studied which span the TUBIS insertion site at the 3' end of the tubulin gene cluster. For these experiments three different restriction enzymes were used, namely HindIII, MboI and NarI. They all cut the fl-tubulin pseudogene and they also cut TUBIS within a few hundred nucleotides of its 5' end. Table 1 gives the sizes of the expected fragments based on published T. brucei sequence data. With HindIII a minor genomic fragment of 0.8 kb hybridized with both a tubulinspecific probe (Fig. 3) and also with a TUBIS-specific probe (Figs. 5A and 5B). These hybridizations define the 785 bp transition fragment. Two other restriction enzymes can also be used for the same purpose of identifying transition fragments starting in the fl-tubulin pseudogene and going into the TUBIS sequence. MboI and NarI both lead to a restriction fragment of about 1.3 kb, which hybridizes strongly with a tubulin-specific probe and which is marked by an asterisk in Fig. 6A. Despite some background in the lower third, the band is discernible in all lanes. Within the tubulin cluster itself, the restriction enzyme NarI has only one cutting site per repeat, leading to the prominent repeat band of 3.6 kb in lanes 2, 4, 6 and 9 in the upper part of Fig. 6A. The restriction enzyme MboI, on the other hand, cuts at many sites in the tubulin repeat, thereby giving numerous small fragments that hybridize with the tubulin probe. Some of these are clearly seen at about 0.5 kb in lanes 1, 3, 5 and 8. They may contribute to the background smear in the left half of Fig. 6A. Note that the MboI-generated fragments from the 3' end are always slightly larger than the ones obtained with NarI, in agreement with the calculated sizes in Table 1. Although hybridization is, as expected, less strong with a TUBIS-specific probe (Figure 6B), the same 1.3 kb fragments clearly hybridized with this probe, when the genomic DNA was derived either from T. brucei brucei, T. rhodesiense, T. gambiense or T. evansi. Additionally the situation with T. equiperdum is of particular interest: here also the same transition fragments could be identified (Fig. 7). In the experiment shown in this figure, the restriction enzyme Sau3a was used instead of MboI: it may be recalled that the two enzymes have the same cutting sequence. The DNA of T. equiperdum is clearly cut in the same way as that of T. brucei in the transition region from the fl-tubulin to the TUBIS sequence. Taken together, experiments with three different restriction enzymes show that the 3'-flanking sequences of the tubulin cluster are virtually identical in all these species and that TUBIS is in exactly the same genomic location in all.
TABLE l The joining region between the fl-tubulin pseudogene and TUBIS The coordinates of the restriction sites are derived from the published sequences of the tubulin genes (Kimmel et al., 1985) and of TUBIS (Affolter et al., 1989). The numbering system of TUBIS used in this communication and in Affolter et al. (1989) is the same. Restriction enzyme
HindllI Mhol NarI
Cutting sites in
Fragment length of
fl-tubulin
TUBIS
fl-tubulin
insert
Total fragment length
460 402 286
378 835 686
441 499 615
344 801 652
785 1300 1267
184
B
A 12
1 2
3 4 5 6 7 8 9
34
5 6789
w
8
m B
Fig. 6. Hybridization of genomic DNAs with tubulin and TUBIS probes. DNA was digested with MboI (lanes 1, 3, 5 and 8) or NarI (lanes 2, 4, 6 and 9). Lane 7, lambda HindIII marker, DNA came from the following species: lanes 1 and 2, T. brucei; lanes 3 and 4, T. evansi; lanes 5 and 6, T. gambiense; lanes 8 and 9, T. rhodesiense, (A) Hybridization was with the/%tubulin probe pTBtu9BI. Note labelled bands at 1.3 kb, marked with an asterisk. (B) Same filter as in (A), but hybridized with the TUBIS probe pTBtu3.4. Note the same labelled bands at 1.3 kb as in (A).
Discussion
Previous observations on the presence of TRS/ingi in various species of trypanosomes has been extended (Murphy et al., 1987; Kimmel et al., 1987). This retroelement occurs only in the subgenus Trypanozoon, but not in others. For T. vivax the published data are contradictory. The tubulin genes are arranged in a single cluster in T. brucei (Seebeck et al., 1983; Thomashaw et al., 1983). We show here that this is also the case for the other species of the subgenus Trypanozoon, but not for other subgenera. In these other subgenera the tubulin genes are more dispersed. In addition, as is shown here, wherever a tubulin gene cluster is found, this cluster is flanked by TUBIS at its 3' end. This means that TRS/ingi jumped into the tubulin cluster early in its development in the ancestor of the subgenus Trypanozoon and stayed in that location. One may wonder whether TRS/ingi in general increased the formation of gene clusters, perhaps through an increased mutation or recombination rate. Since recombination is important in the switching of VSG genes to their expression site, one may be tempted to speculate that TRS/ingi might be involved in this process. However, there is no evidence for this. In fact, some observations point against a direct involvement of TRS/ingis in this process of recombination, although this retroele-
185
A 1
2
3
B 4
5
6
-
1
2
3
4
5
6
Do lIRa*
E
e
Fig. 7. Hybridization of genomic DNA with tubulin and TUBIS probes. D N A was digested with EcoRI (lanes 1 and 2), H&dlII (lanes 3 and 4) and Sau3a (lanes 5 and 6). D N A came from T. brucei (lanes 1, 3 and 5) or from T. equiperdum (lanes 2, 4 and 6). Asterisks and arrowheads identify bands that light up with both probes. (A) Hybridization with ~-tubulin probe pTBtugBI, (B) Same filter as in (A), but hybridized with the TUBIS probe pTBtu3.4.
ment or fragments of it have been found in the proximity of several VSG genes (Pays et al., 1989; Matthews et al., 1990; Smiley et al., 1990). Firstly, some species of trypanosomes such as T. congolense show some form of switching of their VSGs, but have no easily identifiable TRS/ingi elements. Secondly, there is no evidence at all that the switching occurs via an RNA copy of a gene, and thirdly, there are no obvious preferences of TRS/ingi to be located in or close to the VSG storage or expression sites. How could this narrow distribution, limiting TRS/ingi to only a few species, have arisen? Several possibilities can be envisaged, of which the most likely is that the subgenus trypanozoon acquired the element from somewhere else before it diverged into different species. In this case two options should be considered. The first is that of a retroviral origin. Although many viruses with a similar genome organization to that of TRS/ingi carrying gag-pol-env genes are known (Doolittle et al., 1989), none has been found in parasitic protozoa. However, the trypanosomes and their ancestors may well have come into contact with retroviruses of many very different sources. These could have been transmitted like any other virus through air, water etc. The second possibility is less orthodox and would involve horizontal gene transfer from host to parasite without the intermediary of a virus. In such a scenario a gene transfer might have occurred inside the host (either the vertebrate or the insect vector). Maybe occasionally trypanosomes take up a fragment of host D N A and incorporate it into their own genome. It is in this context that G. morsitans was
186 analyzed for the presence o f TRS/ingi. However, the o u t c o m e was negative, as reported above. In prokaryotes horizontal gene transfer t h r o u g h plasmids is well established (for review see, for example, D a y et al., 1990). This seemingly esoterical question o f horizontal gene transfer from eukaryotic host to eukaryotic parasite could have considerable practical implications for infectious diseases, particularly for intracellutar parasite infections in the tropics. If such exchanges occurred, they would seem more likely to happen with intracellular rather than with extracellular parasites, also more likely within members o f the eukaryotes than between eukaryotes and parasitic prokaryotes. A prime case to study is malaria: do species o f Plasmodium have genes or gene fragments derived from their specific hosts? Consequences for the development o f vaccines immediately spring to mind. Very little is k n o w n about the time-scale o f the evolution o f trypanosomes. Comparisons o f kinetoplast ribosomal R N A sequences and considerations o f the fossil record o f their insect hosts suggest that the ancestors o f T. brucei (belonging to the African Salivaria group) and T. cruzi (South American Stercocaria group) separated some 100 M y r ago (Lake et al., 1988). The two Salivarian subgenera T r y p a n o z o o n (with T. brucei) and N a n n o m o n a s (with T. congolense) obviously only separated later, but no geological time has so far been attached to this event. With more sequence data available, this question m a y become more accessible to study in the near future. Molecular studies on the evolution o f parasites and the coevolution with their hosts will make major contributions to the understanding o f h o w the complex host-parasite relationships came to be.
Acknowledgements We thank T. Seebeck for m a n y helpful discussions. Supported by the Swiss National Science Foundation.
References Affolter, M., Rindisbacher, L. and Braun, R. (1989) The tubulin gene cluster of Trypanosoma brucei starts with an intact/3-gene and ends with truncated fl-gene interrupted by a transposon-like sequence. Gene 80, 177 183. Aksoy, S., Lalor, T.M., Martin, J., Van der Ploeg, L.H.T. and Richards, F. (1987) Multiple copies of a retroposon interrupt spliced leader RNA genes in the African trypanosome, Trypanosomagambienese. EMBO J. 6, 3819-3827. Aksoy, S., Williams, S., Chang, S. and Richards, F.F. (1990) SLACS retrotransposon from Trypanosoma brucei gambiense similar to mammalian LINEs. Nucleic Acids Res. 18, 785-792. Barker, R.H. (1990) DNA probe diagnosis of parasitic infections. Exp. Parasitol. 70, 494-499. Borst, P. (1986) Discontinuous transcription and antigenic variation in trypanosomes. Annu. Rev. Biochem. 55, 702-732. Brun, R. and Schoenenberger, M. (1979) Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semidefined medium. Acta Trop. 36, 289-292. Cunningham, I. (1977) New culture medium for maintenance of tsetse tissue and growth of trypanosomatids. J. Protozool. 24, 325-329.
187 Day, M.J., Bale, M.J. and Fry, J.C. (1990) Plasmid transfer in a freshwater environment. In J. Fiksel and V.T. Covello (eds.) Safety assurance for environmental introductions of genetically engineered organisms, pp. 181-197. Springer, Berlin. Doolittle, R.F., Feng, D.F., Johnson, M.S. and McClure, M.A. (1989) Origins and evolutionary relationships of retroviruses. Q. Rev. Biol. 64, 1-30. Fanning, T.G. and Singer, M.F. (1987) LINE-l: a mammalian transposable element. Biochim. Biophys. Acta 910, 203-212. Hasan, G.M., Turner, M.J. and Cordingley, ,J.S. (1984) Complete nucleotide sequence of an unusual mobile element from Trypanosoma brucei. Cell 37, 333 341. Hide, G., Cattand, P., LeRay, D., Barry, J.D. and Tait, A. (1990) The identification of Trypanosoma brucei subspecies using repetitive DNA sequences. Mol. Biochem. Parasitol. 39, 213 226. Imboden, M., Blum, B., DeLange, T., Braun, R. and Seebeck, T. (1986) Tubulin mRNAs of Trypanosoma brucei. J. Moi. Biol. 188, 393 402. Imboden, M., Laird, A.P.W., Affolter, M. and Seebeck, T. (1987) Transcription of the intergenic region of the tubulin gene cluster of Trypanosoma brucei: evidence for a polycistronic transcription unit in a eukaryote. Nucleic Acids Res. 15, 7357 7368. Kendall, G., Wilderspin, A.F., Ashall, F., Miles, M.A. and Kelly, J.M. (1990) Trypanosoma cruzi glycosomal glyceraldehyde-3-phosphate dehydrogenase does not conform to the 'hotspot' topogenic signal mode. EMBO J. 9, 2751 2758. Kimmel, B.E., ole-Moiyoi, O. and Young, J.R. (1987) Ingi, a 5.2-kb Dispersed sequence element from Trypanosoma brucei that carries half of a smaller mobile element at either end and has homology with mammalian LINEs. Mol. Cell. Biol. 7, 1465 1475. Kimmel, B.E., Samson, S. and Wu, J., Hirschberg, R. and Yarbrough, R. (1985) Tubulin genes of the African trypanosome Trypanosoma brucei rhodesiense: nucleotide sequence of a 3.7-kb fragment containing genes for :~ and fl tubulin. Gene 35, 237 248. Kingsman, A.J. and Kingsman, S.M. (1988) Ty: a retroelement moving forward. Cell 53, 333 335. Lake, J.A., de la Cruz, V.F., Ferreira, P.C.G., Morel, C. and Simpson, L. (1988) Evolution of parasitism: Kinetoplast protozoan history reconstructed from mitochondrial rRNA gene sequences. Proc. Natl. Acad. Sci. USA 85, 4779 4783. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Mattei, D.M., Goldenberg, S. and Morel, C. (1977) Biochemical strain characterization of Trypanosoma cruzi by restriction endonuclease cleavage of kinetoplast DNA. FEBS Lett. 74, 264-268. Matthews, K.R., Shiels, P.G., Graham, S.V., Cowan, C. and Barry, J.D. (1990) Duplicative activation mechanisms of two trypanosome telomeric VSG genes with structurally simple 5' flanks. Nucleic Acids Res. 18, 7219-7227. Murphy, N.B., Pays, A., Tebabi, P., Coquelet, H., Guyaux, M. and Steinert, M. (1987) Trypanosoma brucei repeated element with unusual structural and transcriptional properties. J. Mol. Biol. 195, 855 871. Pays, E., Tebabi, P., Pays, A., Coquelet, H., Revelard, P., Salmon, D. and Steinert, M. (1989) The genes and transcripts of an antigen gene expression site from T. brucei. Cell 57, 835-845. Roth, C., Bringaud, F., Layden, R.E., Baltz, T. and Eisen, H. (1989) Active late-appearing variable surface antigen genes in Trypanosoma equiperdum are constructed entirely from pseudogenes. Proc. Natl. Acad. Sci. USA 86, 9375-9379. Seebeck, T., Whittaker, P.A., Imboden, M., Hardman, N. and Braun, R. (1983) Tubulin genes of Trypanosoma brucei: A tightly clustered family of alternating genes. Proc. Natl. Acad. Sci. USA 80, 4634-4638. Singer, M. and Berg, P. (1991) Genes and genomes, a changing perspective. University Science Books, Mill Valley, CA. Smiley, B.L., Aline, R.F., Myler, P.J. and Stuart, K. (1990) A retroposon in the 5' flank ofa Trypanosoma brucei VSG gene lacks insertional terminal repeats. Mol. Biochem. Parasitol. 42, 143-151. Thomashow, L.S., Milhausen, M., Rutter, W.J. and Agabian, N. (1983) Tubulin genes are tandemly linked and clustered in the genome of Trypanosoma brucei. Cell 32, 35-43. Van der Ploeg, L.H.T. (1990) Antigenic variation of African trypanosomes: genetic recombination and transcriptional control of VSG genes. In B. Hames and D.L. Glover (eds). Frontiers in molecular biology: genome rearrangements and amplification, pp. 51-97, IRL Press, Oxford.
175
ACTROP 00249
Evolution of the retrotransposons TRS/ingi and of the tubulin genes in trypanosomes R i c h a r d Braun a, K a t h r y n Behrens a, Andreas Glauser a and Reto Brun b alnstitute for General Microbiology, University (?f Bern, Baltzer-Strasse 4, CH-3012 Bern, Switzerland and bSwiss Tropical Institute, Socinstrasse 57, CH-4002 Basel, Switzerland
(Received 25 June 1992: accepted 9 September 1992) The African trypanosomes have genomes of high plasticity, as demonstrated for instance by their ability to shuffle their genes around, coding for variant-specific surface glycoproteins (VSGs). Another indication of their genome plasticity is the presence of multiple retroelements. The retrotransposon-like element TRS/ingi is present in many copies in the genome of trypanosomes. One particular derivative of TRS/ ingi, called TUBIS, had previously been found to interrupt a tubulin gene in a particular strain of T. brucei. Here both TRS/ingi and TUBIS were studied by hybridizing genomic DNA of various strains and species of trypanosomes with suitable probes in order to elucidate the evolution of this family of retroelements. The TSR/ingi elements are highly repeated and have very long open reading frames, while TUBIS clearly is a truncated, inactivated form of this element, found in only one particular chromosomal location. Both elements were shown to be present in several strains and species of the subgenus Trypanozoon, in particular in T. brueei hruceL T. gambiense, T. rhodesiense, T. equiperdum and T. evansi. They could not be detected in species of other subgenera, in particular in T. congolense and T. cruzi. These findings suggest that the retrotransposon TRS/ingi was acquired by trypanosomes only after divergence of present day subgenera. The TUBIS element was found in exactly the same chromosomal location (at the 3' end of the tubulin gene cluster) in many different strains and species of the subgenus Trypanozoon. This shows that the element was transposed to this location before speciation of the subgenus. Although TRS/ingi is unlikely to be involved directly in VSG switching, it may have contributed to the genome plasticity of trypanosomes. Key words: Trypanosoma brucei; Retrotransposon; Evolution; Tubulin genes; TUBIS; Genome organization
Introduction T h e A f r i c a n t r y p a n o s o m e s o f the s u b g e n u s T r y p a n o o z o o n are a h i g h l y successful g r o u p o f parasites. M a n y m o l e c u l a r m e c h a n i s m s a l l o w t h e m to e v a d e the h o s t defenses, n o t a b l y t h e i r ability to f r e q u e n t l y c h a n g e t h e i r s u r f a c e c o a t by n e w l y a c t i v a t i n g o n e o f a l a r g e n u m b e r o f v a r i a n t - s p e c i f i c s u r f a c e g l y c o p r o t e i n genes. T h e s e V S G switches, the p r e s e n c e o f a large n u m b e r o f m i n i c h r o m o s o m e s a n d the f r e q u e n t c h a n g e s in c h r o m o s o m e size testify to a h i g h d e g r e e o f g e n o m i c plasticity (for r e v i e w s see Borst, 1986; V a n d e r P l o e g , 1990). C o r r e l a t e d w i t h this plasticity is the p r e s e n c e o f a c o n s i d e r a b l e n u m b e r o f m o b i l e genetic e l e m e n t s , w h o s e e x i s t e n c e in g e n e r a l has Correspondence to." R. Braun, Institut ffir Allgemeine Mikrobiologie, Universit~it Bern, Baltzer-Strasse 4, CH 3012 Bern, Switzerland.
176
not found a satisfactory explanation in any organism. They may function to increase the mutation rate in their carrier species and thereby provide them with a selective advantage. Genetic elements resembling retrotransposons have been found in many species of eukaryotes from yeast to man (Fanning and Singer, 1987; Kingsman and Kingsman, 1988; Doolittle et al., 1989). In trypanosomes, too, several different retroelements have been found. One group resembles retrons, poly-A tailed mRNAlike sequences with flanking repeats at either end. To this group belong the RIMEs, 500 nucleotide long elements found in the ribosomal RNA genes of Trvpanosoma brucei, the causative agent of sleeping sickness in Africa (Hasan et al., 1984) and an element of similar structure recently found in the glyceraldehyde-3-phosphate dehydrogenase gene cluster of T. cruzi, the parasite of Chagas disease in South America (Kendall et al., 1990). A second group of retroelements is like the human LINES (Singer and Berg, 1991) to be placed in the retrotransposon category, since they are much larger, namely 5 7 kb long, have short insertion site duplications and one or two very long open reading frames resembling those of retroviruses. This group contains the SLACS, with all nine copies exclusively in the spliced leader sequence gene cluster (miniexon gene cluster) of T. brucei gambiense (Aksoy et al., 1987; Aksoy et al., 1990), and also the TRS/ingi of T. brucei brucei. TRS/ingi was discovered at the same time by two different groups and given the name TRS for Trypanosome Repetitive Sequence by one of the groups (Murphy et al., 1987) and ingi (Kiswahili for many) by the other group (Kimmel et al., 1987). The element occurs in about 100-200 copies per genome. It is 5.2 kb long and shows marked sequence similarities with reverse transcriptase, tether, RNase and Zn-fingers (Doolittle et al., 1989). There is a half of a RIME element (about 250 nucleotides) at each end of TRS/ingi. In the context of our long-standing interest in the tubulin gene cluster of T. brucei (Seebeck et al., 1983; Imboden et al., 1986; Imboden et al., 1987) we have become involved with TRS/ingi. We recently showed that one particular member of the TRS/ ingi family interrupts the tubulin gene cluster of T. brucei at its 3' end (Affolter et al., 1989). It is located in the middle of a/~-tubulin gene, thereby of course inactivating this gene into a pseudogene. We have called this element of the TRS/ingi family TUBIS for tubulin insertion sequence. The fact that TUBIS interrupts an otherwise intact/Ltubulin gene is a good indication that TRS/ingi elements actively transposed at some time in the past. In the following we address the question of whether the insertion of TUBIS in the tubulin gene cluster is strain-specific or whether it is found in different strains and different species. The answer to this question will bear on the evolution of this retroelement.
Materials and Methods
Growth of trypanosomatids Procyclic trypomastigotes of Trypanosoma brucei brucei, T.b. rhodesiense and T.b. gambiense were grown at 27°C in SDM-79 supplemented with 10% heat inactivated (56°C for 30 rain) fetal bovine serum (Gibco) and 100 units per ml penicillin and 100 ~tg per ml streptomycin (Brun and Schoenenberger, 1979). When not specified, the strain of T.b. brucei used was STIB 366; occasionally S 427 was employed.
177 The two strains of T.b. rhodesiense used were STIB 389 and STIB 704, both isolated from sleeping sickness patients in Tansania; the T.b. gambiense cultivated was strain TH-1/78E(031), isolated from a patient in the Ivory Coast. Trypanosoma congolense procyclic trypomastigotes, clone STIB 223V isolated in 1971 from a hyaena in Tansania, were cultured at 27°C in SM medium buffered with 25 mM Hepes and supplemented as in the case of T. brucei with serum and antibiotics (Cunningham, 1977). Trvpanosoma evansi strain STIB 779 (KETRI 2443), obtained from Dr. E. Zweygarth, Nairobi, Kenya, and Trypanosoma equiperdum strain BoTat-78, obtained from Dr. H. Eisen (Roth et al., 1989) or strain STIB 818, isolated in China, were propagated in mice or rats. The blood was obtained by cardiac puncture and the trypanosomes separated from the blood cells by DEAE 53 chromatography. Epimastigotes of Trypanosoma cruzi Y strain were grown in a simple monophasic medium (Mattei et al., 1977) at 27°C.
Analytical procedures DNA from protozoa was isolated as described by Seebeck et al. (1983), except that the CsCI step was omitted. From the tsetse fly, Glossina morsitans centralis, DNA was prepared as follows. The puparium of 30 pupae was cut open and discarded, and the soft tissue lysed in 4 ml buffer (100 mM Tris-HC1 (pH 8.0), 50 mM NaCI, 50raM EDTA, 1% SDS, 0.15mM spermine, 0 . 5 m M spermidine). To this 20gl proteinase K solution (10 mg/ml) were added and the mixture incubated 3 h at 37°C. Before and after incubation the mixture was passed through a Potter homogenizer. This treatment was followed by one extraction each with phenol, phenol/chloroform/ isoamyl alcohol and chloroform/isoamyl alcohol. The DNA was then precipitated with alcohol and re-dissolved in 3 ml TE buffer. Restriction enzyme digestion, gel electrophoresis, transfer, labelling by nick translation and hybridization were done under standard conditions (Maniatis et al., 1982). For subcloning the Bluescript vector BS + was used. As a size marker in electrophoreses, bacteriophage lambda DNA cut with HindIII was used (Boehringer Mannheim). The hybridization solution contained 50% formamide, 0.25 M NaH2PO4, 0.25 M NaC1, 7% SDS, 1 mM EDTA, pH 7.2. Hybridization was done overnight at 37°C on Zeta-Probe blotting membranes after 1 h prehybridization. Thereafter, the filters were rinsed briefly in 2 x SSC and washed at room temperature for 15 rain each with 2 x SSC/0.1% SDS and 0.5 x SSC/0.1% SDS.
Hybridization probes The following plasmids, also marked in Fig. 1, were used as hybridization probes: pTBtu9B covers an entire 3.6 kb repeat of the tubulin coding and the intergenic regions from one EcoRI site to the next. There is one single EcoRI site per tubulin repeat (Seebeck et al., 1983). pTRS1.6-BS 1.2 covers the reverse transcriptase region from nucleotide 2583 (SalI) to nucleotide 3753 (BamHI) of TRS 1.6 (Murphy et al., 1987). pTBtu9BI covers the first two-thirds of a /~-tubulin gene from nucleotide 125 (PstI) to nucleotide 906 (PstI) of T. brucei, as numbered by Kimmel et al. (1985). pTBtu3.3 (9CV) covers the first 330 nucleotides of TUBIS from nucleotide 48 (HaelII) to nucleotide 378 (HindIII) as numbered by Affolter et al. (1989).
178 pTBlu 9B [
P
A. I
I
N
H
I
E
I
I
I
B-tubulin I
(1 -tubulin
M
pTBlu 3.4
P
B. J
I
N
I Nc I
H
I
I
B-pseudogene I
M
I
pTBIu 3.5
II H I
TUBIS
I
I
Ha
Hf
I
I
M
I
Hf
I pTBIu 3.3
I
1 kb
I
Fig. [. Maps of the tubulin gene cluster and its 3' end in T. brucei. Restriction sites, P: Pstl, N: Narl, H: HindIII, M: MboI, E: EcoRI, Nc: Ncol, Hf: Hinfl, Ha: HaeIlI. The blocks represent coding regions or putative coding regions. The brackets over and below the maps indicate cloned sequences used as probes in hybridization experiments. The clone pTBtu9B is not shown: it covers an entire 3.6 kb repeat from one EcoRI site to the next. Also not shown is a TRS clone. (A) A single 3.6 kb repeat coding for ~- and //-tubulin. The intergenic region between ~ and/~ is arbitrarily shown at the 3' end of the gene. (B) The 3' end of the tubulin gene cluster. The TRS/ingi-like element TUBIS is found after 901 nucleotides of/~-tubulin pseudogene. pTBtu3.5 (9CII) covers 1025 nucleotides o f T U B I S from nucleotide 425 (Hinfl) to nucleotide 1450 (HinfI) as n u m b e r e d by Affolter et al. (1989). pTBtu3.4 ( 9 C I I I ) covers 220 nucleotides o f T U B I S from nucleotide 158 (NeoI) to nucleotide 378 (HindIII) as n u m b e r e d by Affolter et al. (1989). The n u m b e r i n g o f T U B I S starts a r b i t r a r i l y at nucleotide 868 o f the /%tubulin p s e u d o g e n e (Affolter et al., 1989).
Results
The tubulin gene cluster In T. brucei brucei the tubulin genes have been shown to be a r r a n g e d in a single large cluster with a b o u t l0 repeats o f 3.6 kb length each (Seebeck et al., 1983; T h o m a s h o w et al., 1983). Restriction enzymes that cut only a few times in the repeat give a simple digestion pattern, when genomic S o u t h e r n blots are p r o b e d with a cloned tubulin sequence. F i g u r e 2 confirms this result for T. brucei brucei and shows that a very similar p a t t e r n is o b t a i n e d with a second strain o f T. brucei as well as with T. rhodesiense a n d T. gambiense. The t r y p a n o s o m a l species T. evansi too shows virtually the same pattern, while the p a t t e r n for T. cruzi is very different ( F i g u r e 3). This leads to the conclusion that the tubulin genes are a r r a n g e d in a very similar gene cluster within the g r o u p o f T. brucei species. T h e flanking regions o f the tubulin cluster are also a r r a n g e d very similarly in diverse m e m b e r s o f the T. brucei group, as is seen from the results o f restriction m a p p i n g experiments. In the 5'-flanking region there is an E c o R I site 12 kb u p s t r e a m o f the end o f the coding region in all
179
A
B
Fig. 2. Hybridization of genomic DNAs with a tubulin probe. DNA was digested with EcoRI. Lane l, T. hrucei strain 366; lane 2, T. hrucei strain 427; lane 3, T. ~amhiense; lane 4, lambda HindllI marker: lane 5, T. rhodesiense strain 389; lane 6, T. rhodesiense strain 704: lane 7, 7". gamhiense. The tubulin gene repeat of 7". brucei is marked with an arrowhead. (A) Ethidium bromide stained gel. (B) Hybridization with //-tubulin probe pTBtu9Bl.
species tested (Fig. 2). In both strains o f T. rhodesiense examined there is a second cutting site at 10 kb. The two different size T. rhodesiense restriction fragments are together about as intense on the a u t o r a d i o g r a p h s as the single fragment in T. brucei or in T. gambiense. These observations are most easily explained by the presence of two different allelic E c o R I sites in the 5'-flanking region o f the tubulin gene cluster of T. rhodesiense. In the T-flanking region T. brucei and T. evansi show similar patterns when genomic D N A is cut with HindIII or E c o R I and when the transferred D N A is probed with cloned tubulin sequences (Fig. 3). The next HindlII site is a b o u t 0.3 kb downstream o f the cluster, i.e., 0.3 kb beyond the ,q-tubulin pseudogene, giving a fragment o f about 0.8 kb. This same band was observed with T. brucei strains 366 and 427, T. gambiense and T. rhodesiense strains 389 and 704 (data not shown). After this site, the next following HindIII site is another 5 to 6 kb further downstream as seen by hybridization with pTBtu3.5 (data not shown). The first E c o R I restriction site in the 3'-flanking region o f the cluster is 3 kb beyond the/3-tubulin pseudogene. This manifests itself as a 6 kb band in Figs. 2 and 3. This same band also lights up with a TUBIS-specific probe (see later). The D N A o f T. equiperdum also shows a very similar restriction pattern in the tubulin coding area as T. brucei. The repeat is also 3.6 kb long and is cut within the cluster to the same size fragments by the restriction enzymes BgIII and PstI, as is the D N A o f T. brucei (data not shown).
180
Fig. 3. Hybridization of genomic DNAs with a tubulin probe. DNA was digested either with EcoRl (lanes 1, 3, 5 and 7) or with HindlII (lanes 2, 4, 6 and 8). Lanes 1 and 2, T. brucei; lanes 3 and 4, T. cruzi; lanes 5 and 6, T. evansi; lanes 7 and 8, Glossina morsilans. Hybridization with fl-tubulin probe pTBtu9Bl. Note a band of 0.8 kb in lanes 2 and 6 marked with an asterisk: this derives from the 3' flanking region of the cluster and according to sequence data from T. brucei is 785 nucleotides long. The tubulin gene repeat is indicated by an arrowhead. The position of the lambda HindIII markers is given on the right. This clearly d e m o n s t r a t e s that the t u b u l i n genes are arranged in the same way in T. equiperdum as in T. brucei and the other species of the subgenus T r y p a n o z o o n . The TRS/ingi retrotransposon Previous studies on the d i s t r i b u t i o n of the a b u n d a n t T R S / i n g i r e t r o t r a n s p o s o n family in different species of t r y p a n o s o m e s were confirmed a n d extended. P r o b i n g genomic blots of restriction enzyme cut D N A shows that T R S / i n g i is present in multiple copies in T. brucei a n d also in T. evansi, but it c a n n o t be detected either in T. cruzi or in the insect vector Glossina morsitans (Fig. 4). Neither could it be detected in T. congolense (data not shown). This mobile element therefore seems to be confined to the subgenus t r y p a n o z o o n . However, its genomic s u r r o u n d i n g s are quite different in different species. The patterns obtained with T. brucei a n d T. evansi D N A show that the two species can easily be differentiated using a T R S / i n g i D N A probe. This shows that T R S / i n g i sequences can be used for species identification in the same
181
1 2
3
4 ..J--..£
m
m
V'l Fig. 4. Hybridization of genomic DNAs with a TRS/ingi probe. DNA was digested with EcoRI (lanes 1, 3, 6 and 8) or HindIII (lanes 2, 4, 7 and 9). Lane 5, lambda HindIII marker. The DNA came from the following species: lanes I and 2, T. brucei; lanes 3 and 4, T. cruzi; lanes 6 and 7, T. evansi; lanes 8 and 9, Glossina morsitans. Hybridization was with pTRS1.6-BS 1.2 from the putative reverse transcriptase region of TRS/ingi.
way as o t h e r r e p e a t e d D N A elements are used as d i a g n o s t i c tools for t r y p a n o s o m e s (Hide et al., 1990) a n d m a n y o t h e r parasites (for review see Barker, 1990). TUBIS
The m a i n p o i n t o f this p a p e r is to study the e v o l u t i o n a r y stability o f T U B I S , the T R S / i n g i - r e l a t e d insertion in the 3' end o f the tubulin gene cluster. As m e n t i o n e d earlier, T U B I S was originally f o u n d to be located in the middle o f the l a s t / % t u b u l i n gene o f the cluster, t u r n i n g this gene into a t r u n c a t e d / 3 - t u b u l i n p s e u d o g e n e , at least in a p a r t i c u l a r strain o f T. brucei. A l t h o u g h T U B I S a n d T R S / i n g i are clearly related,
182 the difference in their sequences is large enough to allow differential hybridization. Blotting experiments with a TUBIS-specific probe tell one clearly, that there are far fewer T U B I S sequences than TRS/ingi elements. This is seen when comparing genomic D N A hybridizations using either cloned T U B I S and TRS/ingi sequences as probes (Figs. 4 and 5A). W h e n hybridizing with the T U B I S probe pTBtu3.4, genomic digests o f 4 different species, each show one and the same predominant band. With E c o R I this band is at 6 kb, which agrees with the results shown in Fig. 2: there this same band was labelled with a tubulin probe and thereby defined the 3' end o f the cluster. The c h r o m o s o m a l locations o f the other, minor bands hybridizing with the T U B I S probes are not known. Pulsed field gel electrophoreses indicate that they are not in minichromosomes (data not shown). It has already been shown above, that restriction sites in the flanking regions of the tubulin cluster are quite similar in various species o f the subgenus t r y p a n o z o o n to those in the strain o f T. brucei, in which T U B I S was originally characterized. This already points to the probable presence o f T U B I S in other species than T. b r u c e i brucei. In the following we will describe Southern blotting experiments where
A 1 2 3
4 5 6 7 891011
B 1 2
"k
Fig. 5. Hybridization of genomic DNAs with TUBIS probes. (A) DNA was digested with EcoRl (lanes 1, 3, 5, 9 and ll) or HindIIl (lanes 2, 4, 6, 8 and 10). Lane 7, lambda HindIll marker. DNA came from the following species: lanes I and 2, T. brucei, lanes 3 and 4, T. congolense; lanes 5 and 6, T. evansi; lanes 8 and 9, T. rhodesiense; lanes 10 and 1 l, T. gambiense. Hybridization was with the TUBIS probe pTBtu3.4. (B) Separate gel electrophoresis. DNA from T. evansi was digested with HindIII (lane 2). Marker: lane I. Hybridization with the TUBIS probe pTBtu3.3, which includes part of RIME A. Note that in all the HindIII digests a band of 0.8 kb, marked by an asterisk, is visible. This is or corresponds to the 785 nucleotide joining fragment sequenced in T. brucei.
183
sequences were specifically studied which span the TUBIS insertion site at the 3' end of the tubulin gene cluster. For these experiments three different restriction enzymes were used, namely HindIII, MboI and NarI. They all cut the fl-tubulin pseudogene and they also cut TUBIS within a few hundred nucleotides of its 5' end. Table 1 gives the sizes of the expected fragments based on published T. brucei sequence data. With HindIII a minor genomic fragment of 0.8 kb hybridized with both a tubulinspecific probe (Fig. 3) and also with a TUBIS-specific probe (Figs. 5A and 5B). These hybridizations define the 785 bp transition fragment. Two other restriction enzymes can also be used for the same purpose of identifying transition fragments starting in the fl-tubulin pseudogene and going into the TUBIS sequence. MboI and NarI both lead to a restriction fragment of about 1.3 kb, which hybridizes strongly with a tubulin-specific probe and which is marked by an asterisk in Fig. 6A. Despite some background in the lower third, the band is discernible in all lanes. Within the tubulin cluster itself, the restriction enzyme NarI has only one cutting site per repeat, leading to the prominent repeat band of 3.6 kb in lanes 2, 4, 6 and 9 in the upper part of Fig. 6A. The restriction enzyme MboI, on the other hand, cuts at many sites in the tubulin repeat, thereby giving numerous small fragments that hybridize with the tubulin probe. Some of these are clearly seen at about 0.5 kb in lanes 1, 3, 5 and 8. They may contribute to the background smear in the left half of Fig. 6A. Note that the MboI-generated fragments from the 3' end are always slightly larger than the ones obtained with NarI, in agreement with the calculated sizes in Table 1. Although hybridization is, as expected, less strong with a TUBIS-specific probe (Figure 6B), the same 1.3 kb fragments clearly hybridized with this probe, when the genomic DNA was derived either from T. brucei brucei, T. rhodesiense, T. gambiense or T. evansi. Additionally the situation with T. equiperdum is of particular interest: here also the same transition fragments could be identified (Fig. 7). In the experiment shown in this figure, the restriction enzyme Sau3a was used instead of MboI: it may be recalled that the two enzymes have the same cutting sequence. The DNA of T. equiperdum is clearly cut in the same way as that of T. brucei in the transition region from the fl-tubulin to the TUBIS sequence. Taken together, experiments with three different restriction enzymes show that the 3'-flanking sequences of the tubulin cluster are virtually identical in all these species and that TUBIS is in exactly the same genomic location in all.
TABLE l The joining region between the fl-tubulin pseudogene and TUBIS The coordinates of the restriction sites are derived from the published sequences of the tubulin genes (Kimmel et al., 1985) and of TUBIS (Affolter et al., 1989). The numbering system of TUBIS used in this communication and in Affolter et al. (1989) is the same. Restriction enzyme
HindllI Mhol NarI
Cutting sites in
Fragment length of
fl-tubulin
TUBIS
fl-tubulin
insert
Total fragment length
460 402 286
378 835 686
441 499 615
344 801 652
785 1300 1267
184
B
A 12
1 2
3 4 5 6 7 8 9
34
5 6789
w
8
m B
Fig. 6. Hybridization of genomic DNAs with tubulin and TUBIS probes. DNA was digested with MboI (lanes 1, 3, 5 and 8) or NarI (lanes 2, 4, 6 and 9). Lane 7, lambda HindIII marker, DNA came from the following species: lanes 1 and 2, T. brucei; lanes 3 and 4, T. evansi; lanes 5 and 6, T. gambiense; lanes 8 and 9, T. rhodesiense, (A) Hybridization was with the/%tubulin probe pTBtu9BI. Note labelled bands at 1.3 kb, marked with an asterisk. (B) Same filter as in (A), but hybridized with the TUBIS probe pTBtu3.4. Note the same labelled bands at 1.3 kb as in (A).
Discussion
Previous observations on the presence of TRS/ingi in various species of trypanosomes has been extended (Murphy et al., 1987; Kimmel et al., 1987). This retroelement occurs only in the subgenus Trypanozoon, but not in others. For T. vivax the published data are contradictory. The tubulin genes are arranged in a single cluster in T. brucei (Seebeck et al., 1983; Thomashaw et al., 1983). We show here that this is also the case for the other species of the subgenus Trypanozoon, but not for other subgenera. In these other subgenera the tubulin genes are more dispersed. In addition, as is shown here, wherever a tubulin gene cluster is found, this cluster is flanked by TUBIS at its 3' end. This means that TRS/ingi jumped into the tubulin cluster early in its development in the ancestor of the subgenus Trypanozoon and stayed in that location. One may wonder whether TRS/ingi in general increased the formation of gene clusters, perhaps through an increased mutation or recombination rate. Since recombination is important in the switching of VSG genes to their expression site, one may be tempted to speculate that TRS/ingi might be involved in this process. However, there is no evidence for this. In fact, some observations point against a direct involvement of TRS/ingis in this process of recombination, although this retroele-
185
A 1
2
3
B 4
5
6
-
1
2
3
4
5
6
Do lIRa*
E
e
Fig. 7. Hybridization of genomic DNA with tubulin and TUBIS probes. D N A was digested with EcoRI (lanes 1 and 2), H&dlII (lanes 3 and 4) and Sau3a (lanes 5 and 6). D N A came from T. brucei (lanes 1, 3 and 5) or from T. equiperdum (lanes 2, 4 and 6). Asterisks and arrowheads identify bands that light up with both probes. (A) Hybridization with ~-tubulin probe pTBtugBI, (B) Same filter as in (A), but hybridized with the TUBIS probe pTBtu3.4.
ment or fragments of it have been found in the proximity of several VSG genes (Pays et al., 1989; Matthews et al., 1990; Smiley et al., 1990). Firstly, some species of trypanosomes such as T. congolense show some form of switching of their VSGs, but have no easily identifiable TRS/ingi elements. Secondly, there is no evidence at all that the switching occurs via an RNA copy of a gene, and thirdly, there are no obvious preferences of TRS/ingi to be located in or close to the VSG storage or expression sites. How could this narrow distribution, limiting TRS/ingi to only a few species, have arisen? Several possibilities can be envisaged, of which the most likely is that the subgenus trypanozoon acquired the element from somewhere else before it diverged into different species. In this case two options should be considered. The first is that of a retroviral origin. Although many viruses with a similar genome organization to that of TRS/ingi carrying gag-pol-env genes are known (Doolittle et al., 1989), none has been found in parasitic protozoa. However, the trypanosomes and their ancestors may well have come into contact with retroviruses of many very different sources. These could have been transmitted like any other virus through air, water etc. The second possibility is less orthodox and would involve horizontal gene transfer from host to parasite without the intermediary of a virus. In such a scenario a gene transfer might have occurred inside the host (either the vertebrate or the insect vector). Maybe occasionally trypanosomes take up a fragment of host D N A and incorporate it into their own genome. It is in this context that G. morsitans was
186 analyzed for the presence o f TRS/ingi. However, the o u t c o m e was negative, as reported above. In prokaryotes horizontal gene transfer t h r o u g h plasmids is well established (for review see, for example, D a y et al., 1990). This seemingly esoterical question o f horizontal gene transfer from eukaryotic host to eukaryotic parasite could have considerable practical implications for infectious diseases, particularly for intracellutar parasite infections in the tropics. If such exchanges occurred, they would seem more likely to happen with intracellular rather than with extracellular parasites, also more likely within members o f the eukaryotes than between eukaryotes and parasitic prokaryotes. A prime case to study is malaria: do species o f Plasmodium have genes or gene fragments derived from their specific hosts? Consequences for the development o f vaccines immediately spring to mind. Very little is k n o w n about the time-scale o f the evolution o f trypanosomes. Comparisons o f kinetoplast ribosomal R N A sequences and considerations o f the fossil record o f their insect hosts suggest that the ancestors o f T. brucei (belonging to the African Salivaria group) and T. cruzi (South American Stercocaria group) separated some 100 M y r ago (Lake et al., 1988). The two Salivarian subgenera T r y p a n o z o o n (with T. brucei) and N a n n o m o n a s (with T. congolense) obviously only separated later, but no geological time has so far been attached to this event. With more sequence data available, this question m a y become more accessible to study in the near future. Molecular studies on the evolution o f parasites and the coevolution with their hosts will make major contributions to the understanding o f h o w the complex host-parasite relationships came to be.
Acknowledgements We thank T. Seebeck for m a n y helpful discussions. Supported by the Swiss National Science Foundation.
References Affolter, M., Rindisbacher, L. and Braun, R. (1989) The tubulin gene cluster of Trypanosoma brucei starts with an intact/3-gene and ends with truncated fl-gene interrupted by a transposon-like sequence. Gene 80, 177 183. Aksoy, S., Lalor, T.M., Martin, J., Van der Ploeg, L.H.T. and Richards, F. (1987) Multiple copies of a retroposon interrupt spliced leader RNA genes in the African trypanosome, Trypanosomagambienese. EMBO J. 6, 3819-3827. Aksoy, S., Williams, S., Chang, S. and Richards, F.F. (1990) SLACS retrotransposon from Trypanosoma brucei gambiense similar to mammalian LINEs. Nucleic Acids Res. 18, 785-792. Barker, R.H. (1990) DNA probe diagnosis of parasitic infections. Exp. Parasitol. 70, 494-499. Borst, P. (1986) Discontinuous transcription and antigenic variation in trypanosomes. Annu. Rev. Biochem. 55, 702-732. Brun, R. and Schoenenberger, M. (1979) Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semidefined medium. Acta Trop. 36, 289-292. Cunningham, I. (1977) New culture medium for maintenance of tsetse tissue and growth of trypanosomatids. J. Protozool. 24, 325-329.
187 Day, M.J., Bale, M.J. and Fry, J.C. (1990) Plasmid transfer in a freshwater environment. In J. Fiksel and V.T. Covello (eds.) Safety assurance for environmental introductions of genetically engineered organisms, pp. 181-197. Springer, Berlin. Doolittle, R.F., Feng, D.F., Johnson, M.S. and McClure, M.A. (1989) Origins and evolutionary relationships of retroviruses. Q. Rev. Biol. 64, 1-30. Fanning, T.G. and Singer, M.F. (1987) LINE-l: a mammalian transposable element. Biochim. Biophys. Acta 910, 203-212. Hasan, G.M., Turner, M.J. and Cordingley, ,J.S. (1984) Complete nucleotide sequence of an unusual mobile element from Trypanosoma brucei. Cell 37, 333 341. Hide, G., Cattand, P., LeRay, D., Barry, J.D. and Tait, A. (1990) The identification of Trypanosoma brucei subspecies using repetitive DNA sequences. Mol. Biochem. Parasitol. 39, 213 226. Imboden, M., Blum, B., DeLange, T., Braun, R. and Seebeck, T. (1986) Tubulin mRNAs of Trypanosoma brucei. J. Moi. Biol. 188, 393 402. Imboden, M., Laird, A.P.W., Affolter, M. and Seebeck, T. (1987) Transcription of the intergenic region of the tubulin gene cluster of Trypanosoma brucei: evidence for a polycistronic transcription unit in a eukaryote. Nucleic Acids Res. 15, 7357 7368. Kendall, G., Wilderspin, A.F., Ashall, F., Miles, M.A. and Kelly, J.M. (1990) Trypanosoma cruzi glycosomal glyceraldehyde-3-phosphate dehydrogenase does not conform to the 'hotspot' topogenic signal mode. EMBO J. 9, 2751 2758. Kimmel, B.E., ole-Moiyoi, O. and Young, J.R. (1987) Ingi, a 5.2-kb Dispersed sequence element from Trypanosoma brucei that carries half of a smaller mobile element at either end and has homology with mammalian LINEs. Mol. Cell. Biol. 7, 1465 1475. Kimmel, B.E., Samson, S. and Wu, J., Hirschberg, R. and Yarbrough, R. (1985) Tubulin genes of the African trypanosome Trypanosoma brucei rhodesiense: nucleotide sequence of a 3.7-kb fragment containing genes for :~ and fl tubulin. Gene 35, 237 248. Kingsman, A.J. and Kingsman, S.M. (1988) Ty: a retroelement moving forward. Cell 53, 333 335. Lake, J.A., de la Cruz, V.F., Ferreira, P.C.G., Morel, C. and Simpson, L. (1988) Evolution of parasitism: Kinetoplast protozoan history reconstructed from mitochondrial rRNA gene sequences. Proc. Natl. Acad. Sci. USA 85, 4779 4783. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Mattei, D.M., Goldenberg, S. and Morel, C. (1977) Biochemical strain characterization of Trypanosoma cruzi by restriction endonuclease cleavage of kinetoplast DNA. FEBS Lett. 74, 264-268. Matthews, K.R., Shiels, P.G., Graham, S.V., Cowan, C. and Barry, J.D. (1990) Duplicative activation mechanisms of two trypanosome telomeric VSG genes with structurally simple 5' flanks. Nucleic Acids Res. 18, 7219-7227. Murphy, N.B., Pays, A., Tebabi, P., Coquelet, H., Guyaux, M. and Steinert, M. (1987) Trypanosoma brucei repeated element with unusual structural and transcriptional properties. J. Mol. Biol. 195, 855 871. Pays, E., Tebabi, P., Pays, A., Coquelet, H., Revelard, P., Salmon, D. and Steinert, M. (1989) The genes and transcripts of an antigen gene expression site from T. brucei. Cell 57, 835-845. Roth, C., Bringaud, F., Layden, R.E., Baltz, T. and Eisen, H. (1989) Active late-appearing variable surface antigen genes in Trypanosoma equiperdum are constructed entirely from pseudogenes. Proc. Natl. Acad. Sci. USA 86, 9375-9379. Seebeck, T., Whittaker, P.A., Imboden, M., Hardman, N. and Braun, R. (1983) Tubulin genes of Trypanosoma brucei: A tightly clustered family of alternating genes. Proc. Natl. Acad. Sci. USA 80, 4634-4638. Singer, M. and Berg, P. (1991) Genes and genomes, a changing perspective. University Science Books, Mill Valley, CA. Smiley, B.L., Aline, R.F., Myler, P.J. and Stuart, K. (1990) A retroposon in the 5' flank ofa Trypanosoma brucei VSG gene lacks insertional terminal repeats. Mol. Biochem. Parasitol. 42, 143-151. Thomashow, L.S., Milhausen, M., Rutter, W.J. and Agabian, N. (1983) Tubulin genes are tandemly linked and clustered in the genome of Trypanosoma brucei. Cell 32, 35-43. Van der Ploeg, L.H.T. (1990) Antigenic variation of African trypanosomes: genetic recombination and transcriptional control of VSG genes. In B. Hames and D.L. Glover (eds). Frontiers in molecular biology: genome rearrangements and amplification, pp. 51-97, IRL Press, Oxford.
