Pii: s0166-6851(02)00238-4

Molecular & Biochemical Parasitology 125 (2002) 211 /216 Short communication Targeting of a tetracycline-inducible expression system to the transcriptionally silent minichromosomes of Trypanosoma brucei Bill Wickstead 1, Klaus Ersfeld, Keith Gull School of Biological Sciences, University of Manchester, 2.205 Stopford Building, Oxford Road, Manchester M13 9PT, UK Received 15 July 2002; received in revised form 10 September 2002; accepted 10 September 2002 Keywords: Trypanosoma ; Gene regulation; Tetracycline; Minichromosomes; Transcription endogenous mRNAs are specifically ablated uponinduction of ectopic double-stranded RNA production), Tetracycline-regulated ectopic gene expression some groups report bleed-through of characteristics of has come to form the backbone of transgenic manipula- RNAi-induction in non-induced cells. Others have tion of Trypanosoma brucei . It has been used for the noted a failure to obtain stable transformants when expression of toxic gene products for conditional inducible RNAi is directed against certain genes. Both knock outs and in RNA interference The these phenomena are thought be the result of ‘leaky' (i.e.
system has been used to induce expression levels higher undesirably high) expression of the regulated genes in even than those of the most active endogenous loci and the absence of induction.
has been shown to regulate protein levels 104-fold in the Here we report that loci on the endogenous mini- most favourable cases .
chromosomes of T. brucei may be targeted for trans- In spite of many successes, tet-responsive gene genics as a means to improve the regulation of a tet- expression in trypanosomes has also encountered limita- tions. Gene regulation over the huge range seen in theinitial report has been difficult to recapture */mostnotably, lower levels of regulation have been observed 2. Ectopic expression from minichromosomes in the cell lines now most commonly used in transgenicstudies. Secondly, the levels of non-induced and, to a In Trypanosomatidae, choice of integration site for lesser extent, induced expression appear to be acutely inducible vectors is hampered by an unusual gene clone-specific. For example, Biebinger et al. described organisation and the promiscuous nature of RNA vectors for inducible expression of toxic gene products; polymerase II (pol II). The majority of genes of these the vector producing the clone with greatest regulation organisms are transcribed processively by pol II from (700-fold at the protein level) also showed 6-fold or very long polycistrons found on the megabase-sized lower regulation in five of ten clones analysed. Finally, chromosomes (MBCs). To facilitate this, the polymerase in inducible RNAi knock-down studies (in which seems able to initiate transcription from very many sitesin the genome, with an apparently low level of DNAsequence specificity. As a result */for tight down- Abbreviations: BSF, bloodstream-form; GFP, green fluorescent regulation of non-induced cells */integration must avoid not only active genes, but also chromatic regions, which chromosome; MC, minichromosome; PCF, procyclic-form; PCR,polymerase chain reaction; pol I, RNA polymerase I; pol II, RNA may initiate pol II transcription, or be party to its read- polymerase II; tet, tetracycline; rDNA, ribosomal RNA genes; TUB , a/b-tubulin gene locus; VSG, variable surface glycoprotein.
The standard target sequence for inducible vectors in * Corresponding author. Tel.: /44-1865-285-455; fax: /44-1865- T. brucei has come to be the region upstream of the promoter for the 18S rRNA gene */a site referred to E-mail address: (K. Gull).
1 Present address: Sir William Dunn School of Pathology, here as the rDNA spacer. This MBC region has been University of Oxford, South Parks Road, Oxford, OX1 3RE, UK shown to be quiescent by nuclear run-on analysis and 0166-6851/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 6 - 6 8 5 1 ( 0 2 ) 0 0 2 3 8 - 4 B. Wickstead et al. / Molecular & Biochemical Parasitology 125 (2002) 211 /216 is targeted such that inducible genes run in the opposite We used these vectors to stably transform T. brucei direction to that of neighbouring endogenous genes .
strain 427-derived cell lines expressing tetracycline Even with this level of theoretical protection, in some repressor (PCF) or tetracycline repressor and T7 RNA instances the level of regulation of constructs targeted to this site is low.
authors) or the ‘single-marker cell line' described by There are, of course, other potential integration sites Wirtz et al. respectively. Transformation was within the T. brucei genome which might reasonably be carried out as in Positive transformants were expected to give good down-regulation of inducible selected by growth for 14 days in the presence of vectors. Prominent amongst these are loci on the induction (1 mg ml1 tetracycline PCF; 1 mg ml1 minichromosomes (MCs): the population of 100 or so doxycycline BSF) and selective drug (20 mg ml1 small linear chromosomes maintained by the organism hygromycin B PCF; 10 mg ml1 hygromycin B BSF), as a reservoir for VSG genes (review Unlike the as well as drugs for maintenance of background (2 mg gene-dense MBCs, MCs are 30  ml1 phleomycin PCF; 2 mg ml1 G418 BSF). For consist predominantly of a repeat of  analysis, independent clones were grown with back- which is highly enriched in MCs and also found on the ground selection for a further 7 days with or without intermediate chromosomes (ICs; 200  sequence is believed to form a large repetitive central Each targeting sequence tested yielded positive trans- core to MCs which is flanked by non-repetitive sub- formants in both PCF and BSF cells. Southern hybri- telomeric regions containing VSG genes (VSG s) and disation to whole chromosomes separated by pulsed- capped by telomeres * field gel electrophoresis demonstrated that VSG -tar- model we have con- firmed by mapping several MCs (data not shown).
geted constructs were integrated into MCs in all clones MCs are potentially good sites for inducible vector analysed (n /14). Constructs directed toward the 177 bprepeat were integrated into MCs ( integration for several reasons. Most importantly, they /80%, 30 of 38) or are transcriptionally silent * /15%, six of 38) */consistent with the distribution /no active genes are present of the repeat in the genome * along the entire chromosome. It follows that insertion of integrated into MBCs or as episomes (  ectopic DNAs to MCs will not disrupt endogenous gene 38, in both cases).
expression (the same can not be said of insertion into These data illustrate that loci on normally silent MCs MBCs). MCs contain a large amount of repetitive DNA can support ectopic expression in both PCF and BSF which is likely to be heterochromatic in nature. It is lifecycle stages.
anticipated that a more closed chromatin structuremight promote better regulation of integrated vectors.
Finally, endogenous MCs are stable with respect tomitosis .
3. Regulation of inducible expression from different sequences */the commonly used rDNA spacer and a/b-tubulin gene array locus (TUB ), both found on MBCs, To assess the transcriptional activity of vectors and three minichromosomal sites: the 177 bp repeat, integrated into different genomic locations, we em- VSG -G4 and VSG -S8. Both minichromosomal VSG s ployed the technique of quantitative real-time reverse- are subtelomeric and can be found on different MCs transcription PCR. Total mRNA was harvested from actively dividing trypanosomes (High-Pure RNA isola- shows the structure of the inducible vectors tion kit, Roche) and cDNA created by reverse-transcrip- used in this study. The plasmids are essentially deriva- tion (Omniscript kit, Qiagen) using an oligo-(dT)15 tives of pHD430 described by Wirtz and Clayton , primer. Relative amounts of GFP cDNA were then though a number of changes have been made. They were assessed by quantitative PCR employing a 5?-nuclease designed for expression of GFP and a hygromycin drug- assay, in which the extent of amplification is assessed by resistance marker under the action of a single tet- release of a fluorophore from a labelled oligonucleotide responsive promoter. In PCF cells this was a derivative probe (Taqman probes, Applied Biosystems). GFP of an endogenous procyclin EP gene promoter, recruit- mRNA levels were normalised against a separate ing endogenous pol I In BSF cells a tet-inducible assessment of constitutive g-tubulin mRNA to control derivative of a T7 RNA polymerase promoter was used for loading, RNA quality and RT efficiency. No- A further vector was made lacking either promoter reverse-transcription and no-template controls were (not shown). This vector was not inducible and was included to check for genomic DNA contamination targeted to the TUB locus, relying on pol II read- and mis-priming. We estimate this system to be capable through for expression. We can supply further sequence of quantitation across more than six orders of magni- data and full lineage for these vectors on request.
tude of mRNA concentration in trypanosomes.

B. Wickstead et al. / Molecular & Biochemical Parasitology 125 (2002) 211 /216 Fig. 1. Inducible expression from different sites in the T. brucei genome. (A) Anatomy of inducible expression vectors. Targeting sequences are:rDNA spacer, 177 bp repeat, and the minichromosomal VSG s G4 and S8. Vectors for expression in procyclic-form (PCF) and bloodstream-form(BSF) cells differ only in the inducible promoters used: tetracycline-inducible procyclin EP1 promoter (Ti-PEP1) for PCF and tetracycline-inducibleT7 polymerase promoter (Ti-PT7) for BSF. (B) Flow cytometric analysis of GFP levels for stably transformed cell lines grown for 7 days in thepresence (thin solid line) or absence (thick solid line) of induction (1 mg ml1 tetracycline PCF; 1 mg ml1 doxycycline BSF). Untransformed cells(dotted line) are also shown.
We investigated levels of expression in three indepen- To our surprise, the regulation of constructs targeted dent randomly-selected clones from each of the stable to the rDNA spacer was very poor (on average 7-fold).
transformations of PCF cells targeting the sites: rDNA Two out of three clones failed to be down-regulated spacer, 177 bp repeat, VSG -G4 and VSG -S8. Three even to the level of the TUB read-through construct (3- clones expressing GFP by read-through at the TUB fold regulation) and the best regulated clone only locus were also included in the analysis for PCF cells.
showed 16-fold induction. For all clones analysed, For BSF cells, only rDNA spacer, 177 bp repeat and regulation was vastly improved when minichromosomal VSG -G4 were targeted.
loci were targeted instead of the rDNA spacer */mean The effect of different integration targets on the regulation of /170-fold regulation was observed for relative cellular concentration of GFP mRNA is shown 177 bp-targeted clones; targeting minichromosomal in In PCF cells containing tet-inducible EP1 VSG -G4 and -S8 resulted in around 600-fold and 120- promoter, fully-induced constructs targeted to the fold regulation, respectively.
rDNA spacer resulted in GFP mRNA levels about 5- BSF cells were stably transformed using constructs fold above those produced by pol II transcription from similar to those used for PCF cells except that a the TUB locus */consistent with the results of other tetracycline-responsive T7 promoter was used in place studies. These mRNA levels were matched in induced of the procyclin EP1 promoter. Again three independent clones in which minichromosomal VSG s had been clones were selected at random for each site targeted.
targeted. The concentration of GFP mRNA in induced These clones showed a slightly different pattern of fully- 177 bp-tagged clones was somewhat lower */at or just induced mRNA levels to those observed in PCF above the level of pol II read-through */indicative, equivalents; the rDNA spacer was in this case the target perhaps, of the more condensed nature of chromatin site displaying weakest transcription, followed by the at these loci.
177 bp repeat and with VSG -G4 the most active site.
B. Wickstead et al. / Molecular & Biochemical Parasitology 125 (2002) 211 /216 Fig. 2. Regulation of tetracycline-responsive expression from different sites in the T. brucei genome. Procyclic-form (PCF) and bloodstream-form(BSF) cells were stably transformed with the vectors shown in . Results are shown for three independent randomly-selection clones for eachintegration target site grown for 7 days in the presence ( /) or absence (m) of induction (1 mg ml1 tetracycline PCF; 1 mg ml1 doxycycline BSF).
Hatched box represents levels of expression from three independent clones tagged with a promoter-less construct targeted to the a/b-tubulin genearray (TUB ). (A) Effect of integration target site on GFP mRNA levels. Relative mRNA concentration was assessed by quantitative real-timereverse-transcription PCR. Background represents the maximum amplification from no-RT and no template controls. (B) Effect of integration targetsite on GFP levels. Green fluorescence was assessed by flow cytometry and is displayed as units above fluorescence of untransformed (non-GFPexpressing) cells. Background represents the error in measurement of fluorescence of untransformed cells.
B. Wickstead et al. / Molecular & Biochemical Parasitology 125 (2002) 211 /216 Unfortunately, little can be inferred from the differences interested in the application of MC-targeting to the in expression levels between PCF and BSF cells in which burgeoning field of RNAi in African trypanosomes. To the same sequences were targeted, since the inducible this end we have developed a modified version of the genes in the two life-cycle stages are transcribed by p2T7Ti plasmid which replaces the rDNA spacer different polymerases. It is interesting, however, that in targeting sequence with 177 bp repeat sequence (p2T7Ti- neither life-cycle stage did the minichromosomal VSG s 177). Initial data obtained using this construct are show a noticeable degree of telomeric-silencing. More- promising. We have used integration at 177 bp repeat over, high-level ectopic expression could be induced loci in BSF cells to drive inducible lethal RNAi knock- from a normally silent MC-VSG locus (although not, it down of a protein involved in transcriptional regulation.
should be noted, of the VSG itself) in BSF cells Our previous attempts to generate similar RNAi cell expressing a different VSG from an expression-site on lines by integration at the rDNA spacer had failed. The vector p2T7Ti-177 can be obtained from the authors on In BSF clones, as with PCF clones, the regulation of GFP mRNA from the rDNA spacer was extremely poor( /3-fold) and again regulation was greatly increased bytargeting integrative constructs to minichromosomal loci. Regulation was not significantly different between177 bp and VSG -G4 targeted constructs (mean induc- We thank Dr Miguel Navarro for advice on the tion of 600- and 400-fold, respectively), but targeting the experiments. The work in our lab is supported by a 177 bp repeat gave lower levels of mRNA in the non- Wellcome Trust program and equipment grant and a induced state.
BBSRC postgraduate stipend to B. Wickstead.
We corroborated the results of the regulation of GFP mRNA by looking at regulation of the protein by flowcytometry. B shows the levels of green fluorescence seen in seven representative clones in which differentgenomic sites have been targeted for the insertion of [1] Gossen M, Bujard H. Tight control of gene expression in inducible vectors. B shows the effect of integration mammalian cells by tetracycline-responsive promoters. Proc target site on the average cellular green fluorescence of Natl Acad Sci USA 1992;89:5547 /51.
[2] Wirtz E, Clayton C. Inducible gene expression in trypanosomes the clones analysed in A. For clones showing mediated by a prokaryotic repressor. Science 1995;268:1179 /83.
greater regulation of mRNA, levels of GFP in the non- [3] Clayton CE, Michels P. Metabolic compartmentation in African induced state were difficult to assess since fluorescence trypanosomes. Parasitol Today 1996;12:465 /71.
was very close to that of untransformed cells (‘back- [4] Ochatt CM, Butikofer P, Navarro M, Wirtz E, Boschung M, ground' in Interestingly, however, for those Armah D, Cross GAM. Conditional expression of glycosylpho-sphatidylinositol phospholipase C in Trypanosoma brucei . Mol clones safely within the window of resolution, regulation Biochem Parasitol 1999;103:35 /48.
of protein levels was in all cases better than that seen for [5] LaCount DJ, Bruse S, Hill KL, Donelson JE. Double-stranded mRNA. This is particularly pronounced in BSF cells, RNA interference in Trypanosoma brucei using head-to-head where rDNA spacer-tagged clones which are regulated promoters. Mol Biochem Parasitol 2000;111:67 /76.
[6] Shi HF, Djikeng A, Mark T, Wirtz E, Tschudi C, Ullu E. Genetic /3-fold at the mRNA level, show a interference in Trypanosoma brucei by heritable and inducible change in average cellular green fluorescence. The origin double-stranded RNA. RNA 2000;6:1069 /76.
of this difference is unknown. None-the-less, patterns of [7] Biebinger S, Wirtz LE, Lorenz P, Clayton C. Vectors for inducible regulation observed at the level of mRNA were also seen expression of toxic gene products in bloodstream and procyclic for protein levels * Trypanosoma brucei . Mol Biochem Parasitol 1997;85:99 /112.
regulation was observed when constructs were integrated into minichromosomal [8] White TC, Rudenko G, Borst P. Three small RNAs within the 10 kb trypanosome rRNA transcription unit are analogous to loci rather than the rDNA spacer.
domain VII of other eukaryotic 28S rRNAs. Nucleic Acids Res In summary, constructs were developed to test a tet- 1986;14:9471 /89.
inducible system integrated at different genomic loca- [9] Rudenko G, Blundell PA, Taylor MC, Kieft R, Borst P. Vsg gene tions in T. brucei . Ectopic expression could be achieved expression site control in insect form Trypanosoma brucei . EMBO not only from MBC loci (rDNA spacer and TUB ), as J 1994;13:5470 /82.
[10] Ersfeld K, Melville SE, Gull K. Nuclear and genome organization described previously, but also 177 bp repeat and VSG of Trypanosoma brucei . Parasitol Today 1999;15:58 /63.
loci on MCs. Regulation of vectors incorporated into [11] Sloof P, Bos JL, Konings AFJM, Menke HH, Borst P, Gutteridge MCs was found to be much greater than for those WE, Leon W. Characterization of satellite DNA in Trypanosoma targeted to the rDNA spacer, due to a better down- brucei and Trypanosoma cruzi . J Mol Biol 1983;167:1 /21.
regulation of expression in the non-induced state. The [12] Williams RO, Young JR, Majiwa PAO. Genomic environment of T. brucei Vsg genes: presence of a minichromosome. Nature system has yet to prove itself in a sufficiently large 1982;299:417 /21.
number of transgenic studies, but its potential for many [13] Weiden M, Osheim YN, Beyer AL, Van der ploeg LHT.
applications can be clearly seen. We are particularly Chromosome structure: DNA nucleotide-sequence elements of a B. Wickstead et al. / Molecular & Biochemical Parasitology 125 (2002) 211 /216 subset of the minichromosomes of the protozoan Trypanosoma [16] Wirtz E, Hoek M, Cross GAM. Regulated processive transcrip- brucei . Mol Cell Biol 1991;11:3823 /34.
tion of chromatin by T7 RNA polymerase in Trypanosoma brucei .
[14] Zomerdijk JCBM, Kieft R, Borst P. A ribosomal RNA gene Nucleic Acids Res 1998;26:4626 /34.
promoter at the telomere of a minichromosome in Trypanosoma [17] Wirtz E, Leal S, Ochatt C, Cross GAM. A tightly regulated brucei . Nucleic Acids Res 1992;20:2725 /34.
inducible expression system for conditional gene knock-outs and [15] Alsford S, Wickstead B, Ersfeld K, Gull K. Diversity and dominant-negative genetics in Trypanosoma brucei . Mol Biochem dynamics of the minichromosomal karyotype in Trypanosoma Parasitol 1999;99:89 /101.
brucei . Mol Biochem Parasitol 2001;113:79 /88.

Source: http://users.path.ox.ac.uk/~kgull/pdf/2002_wickstead.pdf