169
Molecular and Biochemical Parasitology, 46 (1991) 169-180 © 1991 Elsevier Science Publishers B.V. / 0166-6851/91/$03.50 ADONIS 0166685191000500 MOLBIO 01521
Stable DNA transfection of a wide range of trypanosomatids Cara M. C o b u r n 1, Karen M. O t t e m a n 1, Tessie M c N e e l y 2, Salvatore J. Turco 2 and Stephen M. Beverley 1 Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA, U.S.A.; and 2Department of Biochemistry, University of Kentucky, Lexington, KY, U.S.A. (Received 15 October 1990; accepted 28 November 1990)
We have shown that the Leishmania major transfection vector pR-NEO (or derivatives thereof) can be introduced and stably maintained in four species complexes of pathogenic Leishmania (L. tropica, L. mexicana, L. donovani, L. braziliensis), and the genera Endotrypanum and Crithidia; transfection of Trypanosoma cruzi or Trypanosoma brucei was not successful. Quantitative plating assays showed that the transfection efficiencies were high in L. major and Leishmania amazonensis (5×10-5/ce11) and about 10fold less for Leishmania panamaensis and Crithidia. Leishmania donovani transfected with pR-NEO retained the ability to infect hamsters, and amastigotes recovered after 2 months yielded G418-resistant promastigotes which retained high levels of extrachromosomal pR-NEO DNA. In promastigotes, the transfected DNAs existed as extrachromosomal circles, and expressed the predicted 2.4-kb hybrid NEO/DHFR-TS mRNA bearing the trans-spliced miniexon. Large quantitative differences were observed only in Crithidia: relative to transfected Leishmania species, the copy number of pR-NEO was elevated 20-fold, while the levels of the NEO/ DHFR-TS mRNA or Escherichia coli ~-galactosidase (synthesized from the expression vector pX-[~GAL) were reduced 80 and more than 1000-fold, respectively. Thus, genetic signals derived from L. major DNA that mediate RNA expression or stability are recognized by the heterologous Leishmania species but less efficiently by Crithidia. These studies suggest that pR-NEO derived vectors may be applied to the study of genes expressed throughout the life cycle in a wide range of pathogenic trypanosomatids. Key words: Extra-chromosomal DNA; ~-Galactosidase; Hybrid RNAs; Amastigote; Leishmania; Crithidia; Endotrypanum
Introduction
Genetic approaches constitute powerful tools for probing the unusual aspects of gene expression and maintenance in parasites, and for dissecting the biology of the infectious cycle. Unfortunately, classical genetic methods have proven to be of limited use in the trypanosomatid protozoa, due to the lack of or difficulties encountered in manipulating conventional sexual cycles [1-4] and the small repertoire of mutant lines. Recent advances in transfecCorrespondence to: Stephen M. Beverley, Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, 250 Longwood Ave., Boston, MA 02115, U.S.A. Abbreviations: DHFR-TS, dihydrofolate reductase-thymidylate synthase; NEO, neomycinresistance gene; NPT, neomycin phosphotransferase; PCR, polymerase chain reaction.
tion of exogenous genes have lead to alternate strategies of genetic analysis. Expression from DNAs introduced transiently by electroporation has now been accomplished in Leptomonas [5], Leishmania [6; unpublished data] and Trypanosoma brucei [7,8], and stable transfection has been reported for two species of Leishmania, Leishmania major [9; 10] and Leishmania enri etti [11 ]. Our laboratory has focused on stable transfection vectors derived from the amplified circular dihydrofolate reductase-thymidylate synthase (DHFRTS) gene present in certain methotrexate-resistant lines of L. major [9,10,12]. The prototypic vector, pR-NEO, contains the entire 30-kb amplified region with the coding region of neomycin phosphotransferase (NEO) substituted for that of DHFRTS. pR-NEO can be introduced into L. major promastigotes with a high efficiency (up to 10-4 using 100 Bg DNA) and confers resistance to the amino
170
glycoside G418 by expression of a chimeric NEO/DHFR-TS mRNA. This mRNA arises from correct processing at the normal DHFR-TS sites for miniexon addition and polyadenylation, and the steadystate mRNA levels (per gene copy) are similar to those of the DHFR-TS mRNA in amplified lines. pR-NEO usually resides as a free episome in 2-20 gene copies and can be recovered by transfection into E. coli, thus constituting a shuttle vector. Recently smaller derivatives of pR-NEO have been developed that transfect Leishmania with efficiencies comparable to pR-NEO. One of these, pX, contains only 2.3 kb of L. major DNA, and a polylinker site into which foreign genes can be inserted and stably expressed, thereby constituting a general expression vector [ 10]. A construct bearing the E. coli lacZ gene (pX-13GAL) directed the synthesis of up to 1% of the total L. major cellular protein as active ~-galactosidase. pX has also been utilized to express the GP46/M-2 protein of Leishmania amazonensis in both L. major and L. amazonensis [ 10]. In this report we show that the pR-NEO and pX vectors can be efficiently introduced into a wide range of Leishmania species, including members of the four major human pathogenic species complexes and the related genera Endotrypanum and Crithidia. In Leishmania high levels of foreign gene expression were observed, whereas in Crithidia reduced levels of expression were found. Materials and Methods
Cells. The following cell lines were utilized: L. major, the CC-1 clonal derivative [9]; L. amazonensis, LTB0016 (D. McMahon-Pratt); Leishmania panamaensis, clone B3-202C of strain MHOM/CO/81/CLO-64 (D. McMahon-Pratt); Leishmania donovani, the clonal DI-700 line (B. Ullman) and the Ld4 population of the 1S2D line [13]; Endotrypanum shaudinni, strain LV88 (D. McMahon-Pratt; ref. 14); Crithidia fasciculata, CFC-1 (L. Simpson); Trypanosoma cruzi, Tul strain (D. McMahon-Pratt, and T. brucei rhodesiense, YTAT 1.1 (E. Ullu). Cells were grown in the media listed in Table I. Transfection. Plasmid DNAs were prepared and transfection performed with late log phase cells as
described by Kapler et al.[9], using 4x107 cells per electroporation and single pulses of 500/xF and 2.25 kV cm -~ in a BioRad Gene Pulser apparatus. Plating on M199 medium/agar plates was performed as described [9] using G418 (Geneticin; BRL) concentrations indicated in Table II.
Molecular biological techniques. Pulsed field electrophoresis of chromosomes was performed using a contour-clamped homogeneous electric field (CHEF) apparatus as described [9]. DNAs in agarose plugs were prepared for restriction digestion by washing at least three times (for at least 2 h each wash) in 10 mM Tris/1 mM EDTA, pH 7.4 (TE), washing twice in TE and 1 mg ml -~ phenyl methane sulfonate (PMSF, freshly added from a 100 mg ml -~ stock in isopropanol, 2 h), and finally three TE washes. For digestion, a washed plug was suspended for 2 h in the appropriate restriction enzyme digestion buffer (lacking reductants and albumin); the buffer was then removed and replaced with 200/.~1 complete digestion buffer and at least 20 units of enzyme. Digestion proceeded overnight at 37°C, and the plug was washed in 0.2 M Tris/0.1 M EDTA, pH 7.4 prior to electrophoresis. Polyadenylated RNAs were made and Southern and Northem blot hybridizations were performed as described [9]. Polymerase chain reaction analysis. Primer extension was performed as previously described (Kapler et al.), using 0.2 ~tg poly(A) + RNA and oligonucleotide primer G G G A A T T C G G A T C C A T CAGAGCAGCCGATTGTCTG containing 22 nucleotide (nt) from the neomycin phosphotransferase gene (nt 108-87) and a 13-nt 5' extension encoding EcoRI and BamHI restriction sites (bold face), cDNAs were subjected to one or two rounds of PCR amplification as previously described [9,15], using the primer described above in combination with either miniexon primer A ( G G G A A T TCGGATCAACGCTATATAAGTTATCAG) which contains nt 5-23 oftheL, major miniexon and the same 13-nt 5' extension described above, or miniexon primer B (TCAGTTTCTGTACTTTATT G) which contains nt 16-35 of the L. major miniexon [15]. ~-Galactosidase assay.
~-Galactosidase was as-
171 sayed using the fluorescent substrate methylumbelliferyl-[3-D-galactoside (Sigma). Cells were collected by centrifugation, washed once in Hanks' balanced saline solution (HBSS; Gibco) and resuspended in 80/xl TPB buffer (50 mM Tris, pH 8.3/1 mM EDTA/150/~g ml -l benzamidine/20/xg m1-1 leupeptin/200 ~g m1-1 1,10-phenanthroline/50/xg m1-1 soybean trypsin inhibitor/100/xg m1-1 PMSF, added just before use). The cell suspensions were placed on ice and broken using a Branson cup horn sonifier, and centrifuged 15 min in a microcentrifuge at 4°C. 80/zl of clear supernatant was transferred to a 320/~1 of [~-galactosidase reaction mix (23 mM Tris, pH 7.5/125 mM NaC1/2 mM MgCIJ 12 mM mercaptoethanol/0.3 mM 4-methylumbelliferyl-[3-D-galactoside) and incubated at 37°C for 30 min. The reaction was stopped by the addition of 2 ml glycine-carbonate buffer (133 mM glycine/3 mM Na2CO3,pH 10.7) and the fluorescence measured in a Hoefer TKO 100 minifluorimeter.
Isolation andinfection ofmonocytes. Humanperipheral blood monocytes were isolated using Sepracell-MN (Sepratech Corp., Oklahoma City, OK) as described [16]. From 100 ml fresh human blood, 2-3x107 monocytes were obtained. Monocytes were cultured in Teflon vials at 37°C in medium 199 (Whittaker M.A. Bioproducts, Walkersville, MD) supplemented with 10 mM Hepes (pH 7.2), penicillln (60/zg ml ), gentamlcm (80/zg ml ) and 10% heat-denatured fetal bovine serum. Monocytes were infected with Ld4 L. donovani and parasites monitored as described [16]. •
-1
.
•
-1
Infection of hamsters and recovery of amastigotes. Female Golden Syrian hamsters (21-28 days old) were injected intraperitoneally with 109 stationary phase promastigotes in 2-3 ml sterile phosphatebuffered saline. Hamsters were killed by etherization after 9-11 weeks, and their spleens were blotted on microscope slides and stained with DiffQuik to monitor the infection. Amastigotes were isolated from spleens homogenized in PSGEMKA buffer (20 mM NazHPO4, pH 7.3/104 mM NaC1/10 mM MgClJl 0 mM KC1/0.5 mM EDTA/5.5 mM Dglucose/0.02% bovine serum albumin; ref. 17). The homogenate was centrifuged twice at 100xg for 15 min, and the pellets discarded. The supernatant was centrifuged at 1200×g for 15 min, and the pellet re-
suspended in PSGEMKA containing 0.05% saponin for 6 min. Amastigotes were pelleted at 1200×g for 15 min, washed once with PSGEMKA, and then suspended in d 10 medium [ 16] at 22°C and 5 % CO2 and allowed to transform into promastigotes, occurring within 3 days. pR-NEO-transfected amastigotes were transformed in d 10 medium containing 50 /xgm1-1 G418. Results
In vitro G418 sensitivity. The sensitivities of L. amazonensis, Leishmania braziliensis, L. donovani, E. shaudinni, Crithidia fasciculata, Trypanosoma cruzi and T. brucei to the aminoglycoside G418 were measured during liquid culture in vitro (Table I). The concentration required for 50% inhibition in cell ~rowth (ECs0) ranged between 2 and 40 p.g G418 ml-'. In general, Leishmania species were more sensitive than the other genera tested. Some of the differences observed among species may reflect differences in the media employed, since the ECs0 for L. panamaensis in Schneider's medium was 50 /zg m1-1 vs. 6/xg ml -t in M199 medium (Table I). Transfection studies in liquid culture. For heterologous species we initially attempted transfection in liquid culture, due to the variable ability of the species or isolates tested to yield colonies by plating. Cells were electroporated in the presence or absence of 50/.~g pR-NEO, and after an overnight incubation without drug, diluted 1:10 into varying concentrations of G418. Typically, in the first passage cells grew regardless of the G418 concentration used, probably due to the high cell density (>106/ml). Upon further passages at a 1:10 dilution above a minimal G418 concentration (Table I), growth was observed only in cells electroporated in the presence of pR-NEO. Two passages in drug were usually required to observe pR-NEO transfection-dependent growth, at G418 concentrations minimally 2-3 fold greater than the ECs0 (Table I). Similar data were reported previously forL. major [9]. For all genera except Trypanosoma, the G418-resistant lines obtained contained pR-NEO DNA as discussed below. For both Trypanosoma cruzi and T.brucei, several different electroporation parameters were tested (125, 250 and 500/xF for T. brucei; 25, 125
172 TABLE I G418 sensitivity and selective conditions in liquid culture Species, line
Medium
L. major L. donovani DI-700 Ld4 L. amazonensis L. panamaensis
M 199
ECs0 qxg m1-1) 1-2
M 199 d 10c M 199 M 199 Schneider's M 199 M 199 Cunningham's Cunningham's
E. shaudinni C.fasciculata T. cruzi T. brucei rhodesiense
8 3 10 6 50 30 40 25 4-8
Minimal" selective conc. (/zg m1-1) 8
Selectiveb passages 2
16-32 10 30 ND ND 60 40-80 d d
2 2 2 ND ND 2 2 __ __
aMinimal G418 concentration used to obtain differential survival of control and lines transfected with 40-50/.Lg pR-NEO. bNumber of serial passages of cells required to show differential survival of pR-NEO transfectants propagated in the G418 concentrations listed in the adjacent left column. CRef. 16. dDifferential survival of control and transfected lines was not observed in these species. ND, not done. TABLE II Quantitative transfection efficiency on agar plates Species
DNA a
G418 Relative " G418 R (/zg ml -~) survival b colonies c
Uncorrected
Corrected d
L. major
pR-NEO pX-GP46 pX-~GAL
16 16 16
52 52 52
230+ 15 1080+34 1070+82
5.7 × 10 -6 2.7 × 10-5 2.7x 10 5
1.1 x 10-5 5.2x 10 5 5.2x 10-5
L. amazonensis
pX-GP46A pX-~GAL
80 50
68 48
1240+ 117 610 + 36
3.1 x 10 5 1.5 x 10 5
4.6 x 10-s 3.3 x 10-5
L. panamaensis
pX-GP46A pX-~GAL
25 25
97 37
113 + 53 24+4
2.8 × 10 6 6× 10 7
2.9 x 10-6 1.6x 104
C.fasiculata
pX-~GAL
120
34
42+7
1.1 × 10 -6
3.2x 10 6
(%)
Transfection frequency (G418 Rcolonies/cell)
a40 ~tg DNA of the indicated construct was utilized.
bRelative survival measures both the intrinsic plating efficiency and survival of cells following electroporation, and is defined as the per cent colonies obtained/cells plated. It is measured by plating an aliquot of 104 cells (taken from 4 x 107 cells passed through the complete electroporation protocol) on media lacking G418, compared to control cells that were not electroporated but were otherwise similarly treated. c -t. . . . Average _ standard deviation of trlphcate determinations. dCorrected for relative survival. a n d 5 0 0 / x F f o r T. cruzi, b o t h at 2.25 k V c m - l ) . U n d e r all o f t h e s e c o n d i t i o n s t h e r e w a s s i g n i f i c a n t c e l l let h a l i t y , r a n g i n g f r o m 1 0 % to m o r e t h a n 9 0 % n o n m o t i l e c e l l s at t h e h i g h e s t c a p a c i t a n c e s e t t i n g . F o r T. cruzi n o D N A - d e p e n d e n t G 4 1 8 r e s i s t a n c e w a s o b s e r v e d at 25 o r 4 0 / ~ g G 4 1 8 m l -~. F o r T. brucei, p R -
N E O d e p e n d e n t G 4 1 8 r e s i s t a n c e w a s o b s e r v e d in t h e 125 ~ c u l t u r e , a f t e r 8 1:10 s e r i a l d i l u t i o n s in 3,5,10, 15 a n d t h e n 3 0 / x g G 4 1 8 m1-1. H o w e v e r , t h e r e s i s t a n c e o f this l i n e a p p a r e n t l y a r o s e i n d e p e n d e n t l y o f p R - N E O , as S o u t h e r n b l o t h y b r i d i z a t i o n w i t h a N E O - s p e c i f i c p r o b e f a i l e d to r e v e a l t h e p r e s -
173
ence of the pR-NEO plasmid (data not shown).
Transfection efficiencies determined by quantitative plating studies. Theefficiencyoftransfection in L. major, L. amazonensis, L. panamaensis and C. fasciculata was measured by quantitative plating studies using three different molecular constructs, pR-NEO [9], pX-I~GAL and pX-GP46A [10]. The efficiency of transfection of L. amazonensis was similar to that of L. major, about 1-5x10 -5 per cell surviving electroporation and plating (Table II). This may be compared to the previously reported -5 . . . . value o f 7 x l 0 forL. maJor, m studies which used more than twice as much pR-NEO DNA (88 vs. 40 ~g in present study; [9]). In contrast, transfection frequencies for L. panamaensis and C. fasciculata were about 10-fold lower, from 1-3xl0-6/cell (Table II). Plating studies were not performed for E. shaudinni and L. donovani, however the rapidity with which G418 resistance emerged in liquid culture suggested that the transfection efficiencies were at least as high as those observed for L. panamaensis and C.fasciculata. pR-NEO DNA in transfected G418 R lines. DNAs from G418 resistant lines were analyzed for the presence of the pR-NEO DNA using Southern blot analysis and a NEO-specific hybridization probe. Chromosomes from L. major, L. donovani, L. amazonensis, Endotrypanum and Crithidia were examined following separation by pulsed-field electrophoresis. No hybridization was observed to the untransfected parental lines (Fig. 1A, lanes marked by '-'), whereas in each transfectant line the NEO probe recognized a DNA of similar mobility, identical to that of monomeric supercoiled circular pRNEO (Fig. 1A, lanes marked by '+'; ref. 9). Less abundant, more slowly migrating DNAs were also evident, which may be oligomeric forms of pRNEO [9,18]. The mobility properties of these hybridizing DNAs were characteristic of circular DNAs, i.e., pulse-time dependent relative mobility and small distortions in the path of migration (data not shown; [18,19]). Hybridization to the well was observed which may be nicked circular pR-NEO DNA [ 12]; this hybridization was especially prominent in Crithidia (Fig. 1A). No alterations in karyotype were evident in the ethidium bromide stained gel (not shown).
A
I
L.maJ: L.amaz L.don. Endo. Crith. II II II i( -
+
-
+
-
+
-
+
-
+
--Well -C
--800
-200
B.
L.don. L.amaz Endo. Crith.
L.maj i
II
ii -
+
--
+
II --
+
II --+
--
+
-24 -15 -94 -6.7 -43
Fig. 1. Presence of pR-NEO in transfected lines of Leishmania, Endotrypanum and Crithidia. Chromosome samples were prepared from control parental lines (-) or pR-NEO transfected derivatives (+) grown in the minimal selective concentration of G418 listed in Table I. The DI700 line of L. donovani was utilized. (A) Molecular karyotype analysis. Chromosomes were separated by CHEF electrophoresis (18 h with 160 s pulse; 18 h with 80 s pulse) and analyzed by Southern blot hybridization with a probe specific for the NEO gene (the 0.9kb SpeI NEO fragment from pSpe-NEOA; [9]). The arrow indicates the position of the circular pR-NEO plasmid in the transfected lines. The positions of the well and compression (C) are marked, as are molecular weight markers (kb). (B) Southern blot analysis. Chromosomal preparations were digested with BgllI, separated on a 0.8% agarose gel, and analyzed by Southern blot hybridization with the NEO-specific probe described in panel A. The arrow marks the position of the expected 9.4-kb fragment from pR-NEO; hybridizing fragments of higher molecular weight have sizes predicted for partial digestion products. The positions of molecular weight markers (kb) are shown.
174
A,
B.
lowing transfection of L. major, which arises from correct utilization of the normal RNA processing L. omozonensis sites of the DHFR-TS gene [9]. Thus, it appears that /_.m~i L.don Cr/'th Endo. -Hf' II in these four heterologous species the L. major sigII II _i_I -+ I A- A+IIA- A+ r nals responsible for directing the sites of 5' and 3' end formation were correctly recognized. In some pR-NEO transfected lines, the sample well exhib~5 m ited a significant amount of hybridization (Crithi75 i !iii!i!i!i~i!i~i~ii~ii~i~!¸~¸ ~ii!i dia, L. amazonensis; Fig. 3A,B). Additional studies !!~i!iii~!iiiii!!i!i!i!i!i~iiii~iiii~: will be required to determine whether this material represents residual contaminating DNA, which usually fractionates with poly(A) RNA (Fig. 3B), or possibly higher molecular weight RNAs some2 4 z ii ,iii i i¸iiiiiii '¸ how arising from pR-NEO. The 5' end of all trypanosomatid mRNAs is formed by trans-splicing of a 39 nucleotide RNA 3ziii! ii!!i!i iiiiiiiii! termed the miniexon [20]. To test whether the chimeric NEO/DHFR-TS mRNA synthesized in heterologous species bore the miniexon, PCR amplification was employed, using either of two oligonucleotide primers specific for the miniexon (whose sequence is identical inLeishmania and CriFig. 2. Northernblotanalysisoftransfectedlines.RNAswere thidia; refs.21-24; J. Miller, unpublished data) and subjectedto Northernblot hybridizationusingthe NEO-specione specific for NEO as described previously [9]. fic probedescribedin Fig. 1A. The arrowmarksthe positionof The two miniexon primers A and B were comthe 2.4-kbNEO/DHFR-TShybridmRNA,and the positionsof plementary to the 5' and 3' portions of the miniexon molecularweightmarkers(kb) are shown.(A) Poly(A)÷RNAs from the species indicated, either parental (-) or pR-NEO and constructed so that mRNAs bearing the minitransfected(+). (B) Poly(A)÷and poly(A)-RNA fromparental exon should yield PCR products differing by 29 nt in or pR-NEOtransfectedL. amazonensis,as indicated. size, 452 and 423 bases respectively [15]. Accordingly, using cDNA derived from pR-NEO transIn Southern blot analysis of these same lines the fected L. major and Crithidia, identical products NEO-specific probe identified a 9.4-kb BgllI fragwhose sizes were estimated as 440 and 403 bp were ment, only in the pR-NEO transfected lines (Fig. obtained using miniexon primer A (Fig. 3A, lanes 1B); 9.4 kb is the size of the NEO-hybridizing frag4,5) and B (Fig. 3A, lanes 6,7), respectively. The ment present in pR-NEO cut with BgllI. Quantitasmall differences (
Molecular and Biochemical Parasitology, 46 (1991) 169-180 © 1991 Elsevier Science Publishers B.V. / 0166-6851/91/$03.50 ADONIS 0166685191000500 MOLBIO 01521
Stable DNA transfection of a wide range of trypanosomatids Cara M. C o b u r n 1, Karen M. O t t e m a n 1, Tessie M c N e e l y 2, Salvatore J. Turco 2 and Stephen M. Beverley 1 Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA, U.S.A.; and 2Department of Biochemistry, University of Kentucky, Lexington, KY, U.S.A. (Received 15 October 1990; accepted 28 November 1990)
We have shown that the Leishmania major transfection vector pR-NEO (or derivatives thereof) can be introduced and stably maintained in four species complexes of pathogenic Leishmania (L. tropica, L. mexicana, L. donovani, L. braziliensis), and the genera Endotrypanum and Crithidia; transfection of Trypanosoma cruzi or Trypanosoma brucei was not successful. Quantitative plating assays showed that the transfection efficiencies were high in L. major and Leishmania amazonensis (5×10-5/ce11) and about 10fold less for Leishmania panamaensis and Crithidia. Leishmania donovani transfected with pR-NEO retained the ability to infect hamsters, and amastigotes recovered after 2 months yielded G418-resistant promastigotes which retained high levels of extrachromosomal pR-NEO DNA. In promastigotes, the transfected DNAs existed as extrachromosomal circles, and expressed the predicted 2.4-kb hybrid NEO/DHFR-TS mRNA bearing the trans-spliced miniexon. Large quantitative differences were observed only in Crithidia: relative to transfected Leishmania species, the copy number of pR-NEO was elevated 20-fold, while the levels of the NEO/ DHFR-TS mRNA or Escherichia coli ~-galactosidase (synthesized from the expression vector pX-[~GAL) were reduced 80 and more than 1000-fold, respectively. Thus, genetic signals derived from L. major DNA that mediate RNA expression or stability are recognized by the heterologous Leishmania species but less efficiently by Crithidia. These studies suggest that pR-NEO derived vectors may be applied to the study of genes expressed throughout the life cycle in a wide range of pathogenic trypanosomatids. Key words: Extra-chromosomal DNA; ~-Galactosidase; Hybrid RNAs; Amastigote; Leishmania; Crithidia; Endotrypanum
Introduction
Genetic approaches constitute powerful tools for probing the unusual aspects of gene expression and maintenance in parasites, and for dissecting the biology of the infectious cycle. Unfortunately, classical genetic methods have proven to be of limited use in the trypanosomatid protozoa, due to the lack of or difficulties encountered in manipulating conventional sexual cycles [1-4] and the small repertoire of mutant lines. Recent advances in transfecCorrespondence to: Stephen M. Beverley, Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, 250 Longwood Ave., Boston, MA 02115, U.S.A. Abbreviations: DHFR-TS, dihydrofolate reductase-thymidylate synthase; NEO, neomycinresistance gene; NPT, neomycin phosphotransferase; PCR, polymerase chain reaction.
tion of exogenous genes have lead to alternate strategies of genetic analysis. Expression from DNAs introduced transiently by electroporation has now been accomplished in Leptomonas [5], Leishmania [6; unpublished data] and Trypanosoma brucei [7,8], and stable transfection has been reported for two species of Leishmania, Leishmania major [9; 10] and Leishmania enri etti [11 ]. Our laboratory has focused on stable transfection vectors derived from the amplified circular dihydrofolate reductase-thymidylate synthase (DHFRTS) gene present in certain methotrexate-resistant lines of L. major [9,10,12]. The prototypic vector, pR-NEO, contains the entire 30-kb amplified region with the coding region of neomycin phosphotransferase (NEO) substituted for that of DHFRTS. pR-NEO can be introduced into L. major promastigotes with a high efficiency (up to 10-4 using 100 Bg DNA) and confers resistance to the amino
170
glycoside G418 by expression of a chimeric NEO/DHFR-TS mRNA. This mRNA arises from correct processing at the normal DHFR-TS sites for miniexon addition and polyadenylation, and the steadystate mRNA levels (per gene copy) are similar to those of the DHFR-TS mRNA in amplified lines. pR-NEO usually resides as a free episome in 2-20 gene copies and can be recovered by transfection into E. coli, thus constituting a shuttle vector. Recently smaller derivatives of pR-NEO have been developed that transfect Leishmania with efficiencies comparable to pR-NEO. One of these, pX, contains only 2.3 kb of L. major DNA, and a polylinker site into which foreign genes can be inserted and stably expressed, thereby constituting a general expression vector [ 10]. A construct bearing the E. coli lacZ gene (pX-13GAL) directed the synthesis of up to 1% of the total L. major cellular protein as active ~-galactosidase. pX has also been utilized to express the GP46/M-2 protein of Leishmania amazonensis in both L. major and L. amazonensis [ 10]. In this report we show that the pR-NEO and pX vectors can be efficiently introduced into a wide range of Leishmania species, including members of the four major human pathogenic species complexes and the related genera Endotrypanum and Crithidia. In Leishmania high levels of foreign gene expression were observed, whereas in Crithidia reduced levels of expression were found. Materials and Methods
Cells. The following cell lines were utilized: L. major, the CC-1 clonal derivative [9]; L. amazonensis, LTB0016 (D. McMahon-Pratt); Leishmania panamaensis, clone B3-202C of strain MHOM/CO/81/CLO-64 (D. McMahon-Pratt); Leishmania donovani, the clonal DI-700 line (B. Ullman) and the Ld4 population of the 1S2D line [13]; Endotrypanum shaudinni, strain LV88 (D. McMahon-Pratt; ref. 14); Crithidia fasciculata, CFC-1 (L. Simpson); Trypanosoma cruzi, Tul strain (D. McMahon-Pratt, and T. brucei rhodesiense, YTAT 1.1 (E. Ullu). Cells were grown in the media listed in Table I. Transfection. Plasmid DNAs were prepared and transfection performed with late log phase cells as
described by Kapler et al.[9], using 4x107 cells per electroporation and single pulses of 500/xF and 2.25 kV cm -~ in a BioRad Gene Pulser apparatus. Plating on M199 medium/agar plates was performed as described [9] using G418 (Geneticin; BRL) concentrations indicated in Table II.
Molecular biological techniques. Pulsed field electrophoresis of chromosomes was performed using a contour-clamped homogeneous electric field (CHEF) apparatus as described [9]. DNAs in agarose plugs were prepared for restriction digestion by washing at least three times (for at least 2 h each wash) in 10 mM Tris/1 mM EDTA, pH 7.4 (TE), washing twice in TE and 1 mg ml -~ phenyl methane sulfonate (PMSF, freshly added from a 100 mg ml -~ stock in isopropanol, 2 h), and finally three TE washes. For digestion, a washed plug was suspended for 2 h in the appropriate restriction enzyme digestion buffer (lacking reductants and albumin); the buffer was then removed and replaced with 200/.~1 complete digestion buffer and at least 20 units of enzyme. Digestion proceeded overnight at 37°C, and the plug was washed in 0.2 M Tris/0.1 M EDTA, pH 7.4 prior to electrophoresis. Polyadenylated RNAs were made and Southern and Northem blot hybridizations were performed as described [9]. Polymerase chain reaction analysis. Primer extension was performed as previously described (Kapler et al.), using 0.2 ~tg poly(A) + RNA and oligonucleotide primer G G G A A T T C G G A T C C A T CAGAGCAGCCGATTGTCTG containing 22 nucleotide (nt) from the neomycin phosphotransferase gene (nt 108-87) and a 13-nt 5' extension encoding EcoRI and BamHI restriction sites (bold face), cDNAs were subjected to one or two rounds of PCR amplification as previously described [9,15], using the primer described above in combination with either miniexon primer A ( G G G A A T TCGGATCAACGCTATATAAGTTATCAG) which contains nt 5-23 oftheL, major miniexon and the same 13-nt 5' extension described above, or miniexon primer B (TCAGTTTCTGTACTTTATT G) which contains nt 16-35 of the L. major miniexon [15]. ~-Galactosidase assay.
~-Galactosidase was as-
171 sayed using the fluorescent substrate methylumbelliferyl-[3-D-galactoside (Sigma). Cells were collected by centrifugation, washed once in Hanks' balanced saline solution (HBSS; Gibco) and resuspended in 80/xl TPB buffer (50 mM Tris, pH 8.3/1 mM EDTA/150/~g ml -l benzamidine/20/xg m1-1 leupeptin/200 ~g m1-1 1,10-phenanthroline/50/xg m1-1 soybean trypsin inhibitor/100/xg m1-1 PMSF, added just before use). The cell suspensions were placed on ice and broken using a Branson cup horn sonifier, and centrifuged 15 min in a microcentrifuge at 4°C. 80/zl of clear supernatant was transferred to a 320/~1 of [~-galactosidase reaction mix (23 mM Tris, pH 7.5/125 mM NaC1/2 mM MgCIJ 12 mM mercaptoethanol/0.3 mM 4-methylumbelliferyl-[3-D-galactoside) and incubated at 37°C for 30 min. The reaction was stopped by the addition of 2 ml glycine-carbonate buffer (133 mM glycine/3 mM Na2CO3,pH 10.7) and the fluorescence measured in a Hoefer TKO 100 minifluorimeter.
Isolation andinfection ofmonocytes. Humanperipheral blood monocytes were isolated using Sepracell-MN (Sepratech Corp., Oklahoma City, OK) as described [16]. From 100 ml fresh human blood, 2-3x107 monocytes were obtained. Monocytes were cultured in Teflon vials at 37°C in medium 199 (Whittaker M.A. Bioproducts, Walkersville, MD) supplemented with 10 mM Hepes (pH 7.2), penicillln (60/zg ml ), gentamlcm (80/zg ml ) and 10% heat-denatured fetal bovine serum. Monocytes were infected with Ld4 L. donovani and parasites monitored as described [16]. •
-1
.
•
-1
Infection of hamsters and recovery of amastigotes. Female Golden Syrian hamsters (21-28 days old) were injected intraperitoneally with 109 stationary phase promastigotes in 2-3 ml sterile phosphatebuffered saline. Hamsters were killed by etherization after 9-11 weeks, and their spleens were blotted on microscope slides and stained with DiffQuik to monitor the infection. Amastigotes were isolated from spleens homogenized in PSGEMKA buffer (20 mM NazHPO4, pH 7.3/104 mM NaC1/10 mM MgClJl 0 mM KC1/0.5 mM EDTA/5.5 mM Dglucose/0.02% bovine serum albumin; ref. 17). The homogenate was centrifuged twice at 100xg for 15 min, and the pellets discarded. The supernatant was centrifuged at 1200×g for 15 min, and the pellet re-
suspended in PSGEMKA containing 0.05% saponin for 6 min. Amastigotes were pelleted at 1200×g for 15 min, washed once with PSGEMKA, and then suspended in d 10 medium [ 16] at 22°C and 5 % CO2 and allowed to transform into promastigotes, occurring within 3 days. pR-NEO-transfected amastigotes were transformed in d 10 medium containing 50 /xgm1-1 G418. Results
In vitro G418 sensitivity. The sensitivities of L. amazonensis, Leishmania braziliensis, L. donovani, E. shaudinni, Crithidia fasciculata, Trypanosoma cruzi and T. brucei to the aminoglycoside G418 were measured during liquid culture in vitro (Table I). The concentration required for 50% inhibition in cell ~rowth (ECs0) ranged between 2 and 40 p.g G418 ml-'. In general, Leishmania species were more sensitive than the other genera tested. Some of the differences observed among species may reflect differences in the media employed, since the ECs0 for L. panamaensis in Schneider's medium was 50 /zg m1-1 vs. 6/xg ml -t in M199 medium (Table I). Transfection studies in liquid culture. For heterologous species we initially attempted transfection in liquid culture, due to the variable ability of the species or isolates tested to yield colonies by plating. Cells were electroporated in the presence or absence of 50/.~g pR-NEO, and after an overnight incubation without drug, diluted 1:10 into varying concentrations of G418. Typically, in the first passage cells grew regardless of the G418 concentration used, probably due to the high cell density (>106/ml). Upon further passages at a 1:10 dilution above a minimal G418 concentration (Table I), growth was observed only in cells electroporated in the presence of pR-NEO. Two passages in drug were usually required to observe pR-NEO transfection-dependent growth, at G418 concentrations minimally 2-3 fold greater than the ECs0 (Table I). Similar data were reported previously forL. major [9]. For all genera except Trypanosoma, the G418-resistant lines obtained contained pR-NEO DNA as discussed below. For both Trypanosoma cruzi and T.brucei, several different electroporation parameters were tested (125, 250 and 500/xF for T. brucei; 25, 125
172 TABLE I G418 sensitivity and selective conditions in liquid culture Species, line
Medium
L. major L. donovani DI-700 Ld4 L. amazonensis L. panamaensis
M 199
ECs0 qxg m1-1) 1-2
M 199 d 10c M 199 M 199 Schneider's M 199 M 199 Cunningham's Cunningham's
E. shaudinni C.fasciculata T. cruzi T. brucei rhodesiense
8 3 10 6 50 30 40 25 4-8
Minimal" selective conc. (/zg m1-1) 8
Selectiveb passages 2
16-32 10 30 ND ND 60 40-80 d d
2 2 2 ND ND 2 2 __ __
aMinimal G418 concentration used to obtain differential survival of control and lines transfected with 40-50/.Lg pR-NEO. bNumber of serial passages of cells required to show differential survival of pR-NEO transfectants propagated in the G418 concentrations listed in the adjacent left column. CRef. 16. dDifferential survival of control and transfected lines was not observed in these species. ND, not done. TABLE II Quantitative transfection efficiency on agar plates Species
DNA a
G418 Relative " G418 R (/zg ml -~) survival b colonies c
Uncorrected
Corrected d
L. major
pR-NEO pX-GP46 pX-~GAL
16 16 16
52 52 52
230+ 15 1080+34 1070+82
5.7 × 10 -6 2.7 × 10-5 2.7x 10 5
1.1 x 10-5 5.2x 10 5 5.2x 10-5
L. amazonensis
pX-GP46A pX-~GAL
80 50
68 48
1240+ 117 610 + 36
3.1 x 10 5 1.5 x 10 5
4.6 x 10-s 3.3 x 10-5
L. panamaensis
pX-GP46A pX-~GAL
25 25
97 37
113 + 53 24+4
2.8 × 10 6 6× 10 7
2.9 x 10-6 1.6x 104
C.fasiculata
pX-~GAL
120
34
42+7
1.1 × 10 -6
3.2x 10 6
(%)
Transfection frequency (G418 Rcolonies/cell)
a40 ~tg DNA of the indicated construct was utilized.
bRelative survival measures both the intrinsic plating efficiency and survival of cells following electroporation, and is defined as the per cent colonies obtained/cells plated. It is measured by plating an aliquot of 104 cells (taken from 4 x 107 cells passed through the complete electroporation protocol) on media lacking G418, compared to control cells that were not electroporated but were otherwise similarly treated. c -t. . . . Average _ standard deviation of trlphcate determinations. dCorrected for relative survival. a n d 5 0 0 / x F f o r T. cruzi, b o t h at 2.25 k V c m - l ) . U n d e r all o f t h e s e c o n d i t i o n s t h e r e w a s s i g n i f i c a n t c e l l let h a l i t y , r a n g i n g f r o m 1 0 % to m o r e t h a n 9 0 % n o n m o t i l e c e l l s at t h e h i g h e s t c a p a c i t a n c e s e t t i n g . F o r T. cruzi n o D N A - d e p e n d e n t G 4 1 8 r e s i s t a n c e w a s o b s e r v e d at 25 o r 4 0 / ~ g G 4 1 8 m l -~. F o r T. brucei, p R -
N E O d e p e n d e n t G 4 1 8 r e s i s t a n c e w a s o b s e r v e d in t h e 125 ~ c u l t u r e , a f t e r 8 1:10 s e r i a l d i l u t i o n s in 3,5,10, 15 a n d t h e n 3 0 / x g G 4 1 8 m1-1. H o w e v e r , t h e r e s i s t a n c e o f this l i n e a p p a r e n t l y a r o s e i n d e p e n d e n t l y o f p R - N E O , as S o u t h e r n b l o t h y b r i d i z a t i o n w i t h a N E O - s p e c i f i c p r o b e f a i l e d to r e v e a l t h e p r e s -
173
ence of the pR-NEO plasmid (data not shown).
Transfection efficiencies determined by quantitative plating studies. Theefficiencyoftransfection in L. major, L. amazonensis, L. panamaensis and C. fasciculata was measured by quantitative plating studies using three different molecular constructs, pR-NEO [9], pX-I~GAL and pX-GP46A [10]. The efficiency of transfection of L. amazonensis was similar to that of L. major, about 1-5x10 -5 per cell surviving electroporation and plating (Table II). This may be compared to the previously reported -5 . . . . value o f 7 x l 0 forL. maJor, m studies which used more than twice as much pR-NEO DNA (88 vs. 40 ~g in present study; [9]). In contrast, transfection frequencies for L. panamaensis and C. fasciculata were about 10-fold lower, from 1-3xl0-6/cell (Table II). Plating studies were not performed for E. shaudinni and L. donovani, however the rapidity with which G418 resistance emerged in liquid culture suggested that the transfection efficiencies were at least as high as those observed for L. panamaensis and C.fasciculata. pR-NEO DNA in transfected G418 R lines. DNAs from G418 resistant lines were analyzed for the presence of the pR-NEO DNA using Southern blot analysis and a NEO-specific hybridization probe. Chromosomes from L. major, L. donovani, L. amazonensis, Endotrypanum and Crithidia were examined following separation by pulsed-field electrophoresis. No hybridization was observed to the untransfected parental lines (Fig. 1A, lanes marked by '-'), whereas in each transfectant line the NEO probe recognized a DNA of similar mobility, identical to that of monomeric supercoiled circular pRNEO (Fig. 1A, lanes marked by '+'; ref. 9). Less abundant, more slowly migrating DNAs were also evident, which may be oligomeric forms of pRNEO [9,18]. The mobility properties of these hybridizing DNAs were characteristic of circular DNAs, i.e., pulse-time dependent relative mobility and small distortions in the path of migration (data not shown; [18,19]). Hybridization to the well was observed which may be nicked circular pR-NEO DNA [ 12]; this hybridization was especially prominent in Crithidia (Fig. 1A). No alterations in karyotype were evident in the ethidium bromide stained gel (not shown).
A
I
L.maJ: L.amaz L.don. Endo. Crith. II II II i( -
+
-
+
-
+
-
+
-
+
--Well -C
--800
-200
B.
L.don. L.amaz Endo. Crith.
L.maj i
II
ii -
+
--
+
II --
+
II --+
--
+
-24 -15 -94 -6.7 -43
Fig. 1. Presence of pR-NEO in transfected lines of Leishmania, Endotrypanum and Crithidia. Chromosome samples were prepared from control parental lines (-) or pR-NEO transfected derivatives (+) grown in the minimal selective concentration of G418 listed in Table I. The DI700 line of L. donovani was utilized. (A) Molecular karyotype analysis. Chromosomes were separated by CHEF electrophoresis (18 h with 160 s pulse; 18 h with 80 s pulse) and analyzed by Southern blot hybridization with a probe specific for the NEO gene (the 0.9kb SpeI NEO fragment from pSpe-NEOA; [9]). The arrow indicates the position of the circular pR-NEO plasmid in the transfected lines. The positions of the well and compression (C) are marked, as are molecular weight markers (kb). (B) Southern blot analysis. Chromosomal preparations were digested with BgllI, separated on a 0.8% agarose gel, and analyzed by Southern blot hybridization with the NEO-specific probe described in panel A. The arrow marks the position of the expected 9.4-kb fragment from pR-NEO; hybridizing fragments of higher molecular weight have sizes predicted for partial digestion products. The positions of molecular weight markers (kb) are shown.
174
A,
B.
lowing transfection of L. major, which arises from correct utilization of the normal RNA processing L. omozonensis sites of the DHFR-TS gene [9]. Thus, it appears that /_.m~i L.don Cr/'th Endo. -Hf' II in these four heterologous species the L. major sigII II _i_I -+ I A- A+IIA- A+ r nals responsible for directing the sites of 5' and 3' end formation were correctly recognized. In some pR-NEO transfected lines, the sample well exhib~5 m ited a significant amount of hybridization (Crithi75 i !iii!i!i!i~i!i~i~ii~ii~i~!¸~¸ ~ii!i dia, L. amazonensis; Fig. 3A,B). Additional studies !!~i!iii~!iiiii!!i!i!i!i!i~iiii~iiii~: will be required to determine whether this material represents residual contaminating DNA, which usually fractionates with poly(A) RNA (Fig. 3B), or possibly higher molecular weight RNAs some2 4 z ii ,iii i i¸iiiiiii '¸ how arising from pR-NEO. The 5' end of all trypanosomatid mRNAs is formed by trans-splicing of a 39 nucleotide RNA 3ziii! ii!!i!i iiiiiiiii! termed the miniexon [20]. To test whether the chimeric NEO/DHFR-TS mRNA synthesized in heterologous species bore the miniexon, PCR amplification was employed, using either of two oligonucleotide primers specific for the miniexon (whose sequence is identical inLeishmania and CriFig. 2. Northernblotanalysisoftransfectedlines.RNAswere thidia; refs.21-24; J. Miller, unpublished data) and subjectedto Northernblot hybridizationusingthe NEO-specione specific for NEO as described previously [9]. fic probedescribedin Fig. 1A. The arrowmarksthe positionof The two miniexon primers A and B were comthe 2.4-kbNEO/DHFR-TShybridmRNA,and the positionsof plementary to the 5' and 3' portions of the miniexon molecularweightmarkers(kb) are shown.(A) Poly(A)÷RNAs from the species indicated, either parental (-) or pR-NEO and constructed so that mRNAs bearing the minitransfected(+). (B) Poly(A)÷and poly(A)-RNA fromparental exon should yield PCR products differing by 29 nt in or pR-NEOtransfectedL. amazonensis,as indicated. size, 452 and 423 bases respectively [15]. Accordingly, using cDNA derived from pR-NEO transIn Southern blot analysis of these same lines the fected L. major and Crithidia, identical products NEO-specific probe identified a 9.4-kb BgllI fragwhose sizes were estimated as 440 and 403 bp were ment, only in the pR-NEO transfected lines (Fig. obtained using miniexon primer A (Fig. 3A, lanes 1B); 9.4 kb is the size of the NEO-hybridizing frag4,5) and B (Fig. 3A, lanes 6,7), respectively. The ment present in pR-NEO cut with BgllI. Quantitasmall differences (
