1984. Alpha-amanitin-insensitive transcription of variant surface glycoprotein genes provides further evidence for discontinuous transcription in trypano- somes.
MOLECULAR AND CELLULAR BIOLOGY, Feb. 1990, p. 720-726 0270-7306/90/020720-07$02.00/0 Copyright C 1990, American Society for Microbiology
Vol. 10, No. 2
Genomic Organization, Chromosomal Localization, and Developmentally Regulated Expression of the GlycosylPhosphatidylinositol-Specific Phospholipase C of Trypanosoma brucei KOJO MENSA WILMOT,* DALE HERELD, AND PAUL T. ENGLUND
Department of Biological Chemistry, The Johns Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, Maryland 21205 Received 11 September 1989/Accepted 2 November 1989
The surface of the bloodstream form of the African trypanosome, Trypanosoma brucei, is covered with about 107 molecules of the variant surface glycoprotein (VSG), a protein tethered to the plasma membrane by a glycosyl-phosphatidylinositol (GPI) membrane anchor. This anchor is cleavable by an endogenous GPI-specific phospholipase C (GPI-PLC). GPI-PLC activity is down regulated when trypanosomes differentiate from the bloodstream form to the procyclic form found in the tsetse fly vector. We have mapped the GPI-PLC locus in the trypanosome genome and have examined the mechanism for this developmental regulation in T. brucei. Southern blot analysis indicates a single-copy gene for GPI-PLC, with two allelic variants distinguishable by two NcoI restriction fragment length polymorphisms. The gene was localized solely to a chromosome in the two-megabase compression region by contour-clamped homogeneous electric field gel electrophoresis. No rearrangement of the GPI-PLC gene occurs during differentiation to procyclic forms, which could potentially silence GPI-PLC gene expression. Enzymological studies give no indication of a diffusible inhibitor of GPI-PLC activity in procyclic forms, and Western immunoblot analysis reveals no detectable GPI-PLC polypeptide in these forms. Therefore, it is highly unlikely that the absence of GPI-PLC activity in procyclic forms is due to posttranslational control. Northern (RNA) blot analysis reveals barely detectable levels of GPI-PLC mRNA in procyclic forms; therefore, regulation of GPI-PLC activity in these forms correlates with the steady-state mRNA level.
fore, it may be the prototype for a new class of enzymes, the GPI-dependent phospholipases. The biological function of GPI-PLC is not known. One possibility is that it catalyzes a slow release of VSG from the plasma membrane of actively dividing bloodstream-form trypanosomes (34, 42); cultured bloodstream-form trypanosomes are known to release sVSG into the medium very slowly (5). A second possibility is that GPI-PLC serves to release the VSG coat during or after differentiation of bloodstream-form trypanosomes to procyclic forms, either in culture at 27°C (29) or in the midgut of the tsetse fly
Glycosyl-phosphatidylinositols (GPIs) serve as anchors to the plasma membrane for numerous functionally distinct hydrolytic enzymes, antigens, and cell adhesion molecules (14, 24). Examples of GPI-anchored proteins include acetylcholinesterase, CD14, neural cell adhesion molecule, and scrapie prion protein. The Trypanosoma brucei variant surface glycoprotein (VSG) is the most thoroughly characterized GPI-anchored protein. The structure and some aspects of the biosynthetic pathway of the VSG GPI have recently been elucidated (14, 26). Trypanosomes contain a highly active GPI-specific phospholipase C (GPI-PLC), which cleaves dimyristoyl glycerol from the VSG GPI anchor, thereby converting the membrane form of VSG (mfVSG) to a soluble form (sVSG) (8). GPI-PLC, purified recently (7, 15, 19) as a 37- to 40kilodalton polypeptide, is a Ca'-independent phospholipase which is stimulated by chelating agents and inhibited by some sulfhydryl agents. GPI-PLC is highly specific; it efficiently cleaves the VSG GPI anchor (7, 15, 19) and GPI biosynthetic precursors which contain a glycan linked to phosphatidylinositol, (26), but barely cleaves unmodified phosphatidylinositol (19). GPI-PLC behaves as an integral membrane protein, even though the nucleotide sequence encoding the protein does not suggest the existence of an obvious hydrophobic membrane-spanning domain (9, 18); the mechanism of its association with membranes remains to be defined. The sequence of GPI-PLC is not homologous to that of other known eucaryotic phospholipase C's; there-
*
intermediate host. During this transformation the VSG polypeptide is shed, in part as a proteolytic fragment (5, 34), and replaced by procyclin, a surface protein characteristic of procyclic cells (33). Interestingly, the loss of VSG and the appearance of procyclin are accompanied by a 103-fold decrease in GPI-PLC activity (7). Finally, it is possible that GPI-PLC plays some role in antigenic variation, a process by which the surface coat of bloodstream-form trypanosomes is replaced with an antigenically distinct one (8, 13, 22). In this work we have mapped the GPI-PLC gene locus and demonstrated restriction fragment length polymorphisms associated with it. We have also identified the chromosome containing the gene. Lastly, we have studied the mechanism of the developmental regulation of GPI-PLC enzyme activity. MATERIALS AND METHODS Growth of trypanosomes. T. brucei bloodstream forms (strain TREU 667; from S. Hajduk, University of Alabama at Birmingham) were grown in Wistar rats, immunosuppressed
Corresponding author. 720
VOL. 10, 1990
DEVELOPMENTAL REGULATION OF T. BRUCEI GPI-PLC
by irradiation (800 rads) prior to infection. Trypanosomes were isolated from rat blood, at a parasitemia of 108 cells per ml, by chromatography on a DE-52 column (23). Procyclic trypanosomes (also TREU 667) were grown in semidefined medium (12) and harvested by centrifugation at a density of 107 cells per ml (27). Isolation of DNA and analysis by Southern blotting. Genomic DNA was isolated from procyclic and bloodstream forms following lysis (108 cells per ml at 450C for 1 h) in 0.5% sodium dodecyl sulfate (SDS)-100 ptg of proteinase K (Boehringer Mannheim Biochemicals) per ml-100 mM NaCl-25 mM EDTA-10 mM Tris hydrochloride (pH 8.0). Extractions with phenol-chloroform-isoamyl alcohol (25: 24:1), ethanol precipitations, and RNase A digestion were performed as described previously (2, 25). After dialysis against TE (10 mM Tris hydrochloride [pH 8], 1 mM EDTA), the DNA was concentrated by Centricon-30 (Amicon Corp.) ultrafiltration. The DNA was digested with restriction enzymes (under conditions described by the supplier), electrophoresed in a 1% agarose gel at 250C in lx TBE (89 mM Tris, 89 mM boric acid, 2 mM EDTA) (25), and then transferred to a Hybond-N membrane (Amicon) (2). DNA was fixed to the membrane by UV irradiation, and the membrane was prehybridized overnight in hybridization buffer, consisting of Sx SSPE (0.9 M NaCl, 50 mM sodium phosphate [pH 7.0], 5 mM EDTA), 50% formamide, 5x Denhardt solution (2), and 100 ,ug of Escherichia coli tRNA per ml. Hybridization with random-primed probes was carried out in hybridization buffer at 42°C for 18 h; the final posthybridization wash was in 0.1x SSC (lx SSC is 0.15 M NaCl plus 0.015 M sodium citrate [pH 7.1])-0.1% SDS-1 mM sodium phosphate (pH 7.0) at 65°C for 1 h. Hybridization with oligonucleotide probes was carried out in 5 x SSC-20 mM sodium phosphate (pH 7.0)-10x Denhardt solution-7% SDS-100 ,ug of tRNA per ml at 60°C for 18 h (44). The final posthybridization wash was at 60°C in lx SSC-1.6% SDS-3.3 mM sodium phosphate (pH 7.0). Membranes were autoradiographed by using XAR film (Eastman Kodak Co.). Purification of RNA and analysis by Northern (RNA) blotting. Poly (A)' RNA was isolated from freshly harvested trypanosomes (3). Cells (2 x 109) were lysed for 2 h at 45°C in 20 ml of buffer containing 200 mM NaCl, 200 mM Tris hydrochloride (pH 7.5), 1.5 mM MgCl2, 2% SDS, and 200 ,ug of proteinase K per ml. After the salt concentration was adjusted to 1 M with 5 M NaCl, poly(A)+ RNA was isolated on an oligo(dT)-cellulose column (Pharmacia, Inc.; 0.1 g of lyophilized resin for lysate from 109 cells). Poly(A)+ mRNA eluted from the oligo(dT)-cellulose column with diethylpyrocarbonate-treated water was evaporated to dryness and resuspended in 100 ,ud of diethylpyrocarbonate-treated water. After ethanol precipitation to remove excess SDS, poly(A)+ RNA was electrophoresed in a 1.2% agarose gel at 25°C and 100 V for 3 h in lx MOPS (morpholinepropanesulfonic acid) buffer (20 mM MOPS-NaOH [pH 7.0], 5 mM sodium acetate, 1 mM EDTA) (2) and then transferred to a Hybond-N membrane as described above. Hybridization and washing conditions were identical to those described for Southern blots. Hybridization probes. See Fig. 2B for the location of probes on the GPI-PLC map. An 829-base-pair NdeI-SalI restriction fragment from pDH4 (pBS [Stratagene Inc.] containing a 1.44-kilobase-pair [kb] GPI-PLC cDNA [18] in its EcoRI site) was purified by agarose gel fractionation and electroelution of the DNA band with the ELUTRAP system (Schleicher & Schuell, Inc.). The fragment, derived com-
721
pletely from the GPI-PLC coding sequence, was random primed and labeled with [a-32P]dATP by using the Multiprime DNA-labeling system (Amersham Corp.). Oligonucleotide probes as241 and as1254 hybridize specifically to nucleotides 241 to 270 and to nucleotides 1254 to 1284, respectively, on the coding strand of GPI-PLC (18). Oligonucleotide probes were end labeled with [_y-32P]ATP (32). pTbac4T-1 (37), containing the genes for T. brucei a- and P-tubulin, was a gift from N. Agabian, University of California, San Francisco. It was digested with HindIII and labeled by random priming. Immunoblotting. A 1% n-octyl glucoside extract of trypanosome membranes (108 cell equivalents) was prepared as previously described (19), and protein was quantitatively precipitated from the extract with organic solvents (43). Following SDS-polyacrylamide gel electrophoresis (10% gel; Bio-Rad Mini Protean II), the proteins were electrotransferred, at 40C, to a nitrocellulose membrane in a buffer containing 6.25 mM Tris and 48 mM glycine at 0.11 A (70 V) for 15 h; the electrotransfer buffer was replaced with fresh solution, and the electrotransfer was continued for 7 h at 0.2 A. The nitrocellulose was then analyzed for total protein by India ink staining (2) and for GPI-PLC polypeptide by probing with antibody. Primary antibody was from polyclonal antiserum raised against GPI-PLC protein (18); alkaline phosphatase-conjugated goat anti-rabbit immunoglobulin G served as the secondary antibody. Antigen-antibody interactions were performed at 250C in 10 mM Tris hydrochloride (pH 7.5)-0 mM NaCl containing 3% bovine serum albumin and 1% goat serum (1). The color was developed with 5-bromo-4-chloro-3-indolylphosphate p-toluidine and Nitro Blue Tetrazolium chloride, under conditions specified by the supplier (Promega Biotec). Chromosomal fractionation by contour-clamped homogeneous electric field (CHEF) gel electrophoresis. Trypanosome minichromosomes and intermediate-sized chromosomes were resolved in 1% agarose (0.5x TBE at 160 mA and 200 V, with a 60-s pulse for 20 h at 9°C) (10, 40). RESULTS Analysis of the GPI-PLC gene by Southern blotting. One of our primary objectives was to map the GPI-PLC gene in the T. brucei genome. Southern blot analysis of genomic DNA from bloodstream-form trypanosomes, by using restriction enzymes which do not cleave the GPI-PLC coding region (HaeIII, HindIII and PvuII), revealed in each case a single restriction fragment (1.8, 17, and 20 kb, respectively) to which the GPI-PLC probe hybridized (Fig. 1A). These results were consistent with either a single-copy gene or a tandem duplication (the coding region of GPI-PLC is 1.07 kb) of the gene. In contrast, NcoI, which also does not cleave the GPI-PLC coding sequence, released two hybridizing fragments (20 and 23 kb, present in roughly equal amounts) (Fig. 1B). The persistence of both NcoI fragments at several digestion times, even in the presence of a large excess of enzyme, confirmed that the larger fragment was not due to partial digestion. Results from the NcoI digest could be explained by the presence of two copies of the gene. To determine the genomic organization of the GPI-PLC gene, we first digested DNA with enzymes which cleave within the GPI-PLC coding sequence (HindII, NdeI, and AccI) (Fig. 1C; see map in Fig. 2B). We then double digested the DNA with these enzymes in combination with several other enzymes which do not cleave within the gene (see Fig.
722
MENSA-WILMOT ET AL.
A
MOL. CELL. BIOL.
C
A s ..1 N'
A I
I
A
R I
D H I I
EDH H E 11 I I
RDH III
A
kb
23.1
-
9.4
-
--N _
_
65 _ 4.3 _
A
I
a
2.3 2.0 _
R
DH
II I- I
A E DH HE I I I I
N
R DH III
2
I
A
1 kb
-
13.3 . 1.1
-
B
B 1
as241
2
3
4
5
*
5
g
3
6
_-
Ndel II
Hind 11 I
Hind II Acc
Sal
as1254
1. I-M
I
kb 23.1--
9.4
-
6.5
-
-100 bp
FIG. 1. Southern blot analysis at the GPI-PLC locus of T. brucei. Bloodstream-form genomic DNA (5 fig) was digested to completion with restriction enzyme, fractionated on a 1% agarose gel, and transferred to a Hybond-N membrane. The fragments were probed with the NdeI-Sa1I fragment (see Fig. 2B for a map). There are no sites within the GPI-PLC coding sequence for any of the nucleases used in the experiments summarized in panels A and B. (A) Analysis of bloodstream-form DNA. Lanes: 1, no digestion; 2, HaeIII; 3, HindIII; 4, PvuII; 5, HaeIII and HindIII; 6, HaeIII and PvuII; 7, PvuII and HindIII. (B) Digestion of bloodstream-form DNA (2 pFg) with NcoI. Reactions were terminated by addition of 20%o SDS (prewarmed to 700C) to a final concentration of 1% followed by incubation at 70'C for 15 min. Lanes: 1, no enzyme addition; 2, digestion for 15 min; 3, digestion for 30 min; 4, digestion for 60 min; 5, digestion for 15 h followed by digestion for 3 h with a fresh volume of enzyme (doubling the original enzyme concentration); 6, digestion for 15 h. (C) Enzymes used cleave within the coding region of the GPI-PLC gene; see Fig. 2B for a map. Digests were fractionated on a 1% agarose gel and transferred to Hybond-N membrane. Lanes: 1, no enzyme added; 2, HindII; 3, NdeI; 4, AccI. Fragment sizes are indicated.
2A). Fragments were detected by hybridization either with a probe which spans the region between the NdeI and Sall sites within the coding sequence, or with oligonucleotide probes which map to either end of the coding sequence (see map in Fig. 2B for location of probes). Based on Southern blots shown in Fig. 1 and numerous others that are not shown, the restriction map of the GPI-PLC locus is summarized in Fig. 2A. We detected two allelic variants of the gene, which were distinguishable by two NcoI restriction fragment length polymorphisms. One allele has an NcoI site about 7.2 kb upstream of the AccI site within the gene, whereas the second allele has lost this site. The second allele has an NcoI site located 6.3 kb 3' to the reference AccI site; this latter
FIG. 2. (A) Restriction map of the GPI-PLC locus in T. brucei. The map was obtained as outlined in the text. Abbreviations: A, AccI; D, NdeI; E. HaeIII; H, HindIl; R, EcoRI; N, NcoI; S, Sall. The restriction fragment length polymorphism described in the text is marked by possession of only one (N' or N2) of the two NcoI cleavage sites indicated, on either allele. The direction of transcription deduced from the protein coding sequence is shown by the arrow. The black area indicates the location of the protein-coding sequence. The 3'-most HindIl site within the gene is located 1 nucleotide 3' to the AccI site within the gene; the Sall site precedes the AccI site by 1 nucleotide. The 5'-most HindIl site within the gene is absent from the published cDNA (prepared from T. brucei subsp. rhodesiense, WRATat 1.1, RNA) sequence (18); this site could arise from a single A-to-T transversion 400 base pairs (bp) 5' from the AccI site. (B) Location of DNA probes on GPI-PLC gene. Symbols: =, noncoding region of GPI-PLC cDNA; X, GPIPLC coding region; _, location of 30-base oligonucleotide probes (as1254 and as241); the arrowheads point in the 5'-to-3' direction of the oligonucleotide.
NcoI site is absent from the first allele (Fig. 2A). The maps in Fig. 2A span about 17 kb, and there is no evidence for tandem repetition of the region. Chromosomal localization of the GPI-PLC gene. The chromosomal location of the GPI-PLC gene was determined by CHEF agarose gel electrophoresis. Using a 1% gel and a 60-s pulse time, we localized the gene within the two-megabase compression region in both procyclic and bloodstream-form trypanosomes (Fig. 3). No hybridization was detected to intermediate or minichromosomes or to chromosomes remaining in the slot. Comparison of the GPI-PLC locus in bloodstream-form and procyclic trypanosomes. Down regulation of the GPI-PLC activity during transformation of bloodstream-form trypanosomes to procyclic forms could be explained by genomic rearrangement of the GPI-PLC locus. However, from Southern blot analysis, we could detect no differences in the GPI-PLC locus in DNA from the two stages of the trypanosome life cycle (compare Fig. 1A and C with Fig. 4A and B). Also, the chromosomal locations of the gene in the two forms are indistinguishable (Fig. 3). Therefore, there is no
DEVELOPMENTAL REGULATION OF T. BRUCEI GPI-PLC
VOL. 10, 1990
A
A
B 1
1
2
3
4
2
3
4
B 5 5
7
1 2
4 kb
23.1 -
-
i.6 M
11 Mb
_
9.4
645
723
_
3 4
kb 23.1- * 9.4-
-
6.5 -
Q0
4.3 -_ 4.3-
-
0)7
Mb intermediate chromosomes
2.3 2.0
-
m 1.3 -
Qm
2.2. 2.0 .
1.1 -
Mini-
/ Chromosomes -
48 Kb
FIG. 3. Chromosomal localization of the GPI-PLC gene by CHEF agarose gel electrophoresis. Intact chromosomal DNA of trypanosomes was prepared in low-melting-point agarose plugs by digestion with proteinase K in Sarkosyl (40, 41). After the plugs had been washed in 50 mM EDTA, the chromosomal DNA was electrophoresed by using the CHEF system (1% agarose-0.5x TBE, with a 60-s pulse for 20 h at 160 mA and 200 V). (A) Autoradiograph of a Southern blot of the gel depicted in panel B, probed with the NdeI-SalI fragment (see Fig. 2B). (B) Photograph of gel stained with ethidium bromide. Lanes (for both panels): 1, ladder of lambda DNA concatemers (FMC Bioproducts); 2, bloodstream-form chromosomes; 3, procyclic chromosomes; 4, Saccharomyces cerevisiae chromosomes (FMC Bioproducts).
evidence that genomic rearrangement plays a role in the regulation of GPI-PLC activity during transformation. The steady-state level of GPI-PLC mRNA is reduced in procyclic trypanosomes. To explore the mechanism of differential GPI-PLC gene expression during development, we measured the steady-state level of its mRNA. We easily detected stable GPI-PLC mRNA in bloodstream-form trypanosomes (Fig. 5A, lane 1) but could barely measure these molecules in procycic trypanosomes (Fig. 5A, lane 2). Under identical experimental conditions, the mRNAs for aand P-tubulin were present in comparable quantities (Fig. SB) in both bloodstream-form (lane 1) and procyclic (lane 2) trypanosomes. These results are similar to those described by Carrington et al. (9). The GPI-PLC mRNA in bloodstream-form trypanosomes reproducibly appeared as a diffuse band, with the largest transcript at about 3.7 kb; the majority was in the 1- to 2-kb size range. The fact that a small amount of GPI-PLC mRNA was reproducibly detected in procyclic trypanosomes raised the possibility that some GPI-PLC polypeptide exists in procyclic trypanosomes in which the enzymatic activity is regulated posttranslationally. We investigated this possibility in the experiments described in the next two sections. Procyclic trypanosomes lack a detectable negative regulator of GPI-PLC activity. Transformation of bloodstream-form trypanosomes to procyclic forms is accompanied by a virtually complete loss of GPI-PLC activity (6) (Table 1, experiments 4 and 6). If the small amount of mRNA in procyclic forms is translated, the absence of activity could be explained by the presence of a regulatory molecule which
0.4 -
FIG. 4. Southern blot analysis of the GPI-PLC gene in procyclic trypanosomes. Genomic DNA from procyclic trypanosomes was analyzed by using the protocols described for Fig. 1. (A) Enzymes are identical to those used in Fig. 1A; (B) enzymes are identical to those used in Fig. 1C.
suppresses the GPI-PLC enzymatic activity. This hypothesis was tested by mixing homogeneous GPI-PLC with crude extracts of procyclic trypanosomes and examining the effect of these extracts on the activity of the purified enzyme; no inhibition of GPI-PLC activity by lysates or detergent extracts from procyclic trypanosomes was detected, even when a 10-fold excess of extract (in terms of cell equivalents) was added (Table 1, experiments 1 to 3). Our results indicate that a stable and diffusible inhibitor, present in excess, does not contribute to the suppression of GPI-PLC activity in procyclic trypanosomes. The GPI-PLC polypeptide is detectable in bloodstreamform but not procyclic trypanosomes. Cognizant of the fact that the steady-state level of mRNA for a given protein in T. brucei does not always correlate with the detectable level of enzyme activity (30), we assayed for the GPI-PLC polypeptide in procyclic trypanosomes. The GPI-PLC protein, like the enzyme activity, is detectable only in bloodstream-form trypanosomes (Fig. 6). Similar results were obtained with whole trypanosomes lysed by boiling in 1% SDS, indicating that the results shown in Fig. 6 are not an artifact of the detergent extraction protocol. Procyclic trypanosomes, therefore, do not synthesize detectable amounts of GPI-PLC protein from the small amount of detectable steady-state mRNA.
DISCUSSION Our goals for undertaking these studies were twofold: first, we wanted to characterize the genomic organization of the GPI-PLC gene, and second, we wanted to study the
MENSA-WILMOT ET AL.
724
A
MOL. CELL. BIOL.
A
B
B
2
2
2
kb -
Fps kd tZ 219 6-
9.5
1
C 2
1
2
M.r e
1004-- 4.4
680-E 427 _E
- 2.4
..
UI
- 1.4
274-EU 18.4 _
:
FIG. 5. Measurement of steady-state levels of GPI-PLC mRNA in bloodstream-form and procyclic trypanosomes. Poly(A)+ RNA (15 ptg) was size fractionated on a formaldehyde-agarose gel, transferred to a Hybond-N membrane, and cross-linked by UV irradiation. (A) The membrane was probed with the NdeI-SalI GPI-PLC restriction fragment (see map in Fig. 2B), washed, and examined by autoradiography (24-h exposure). (B) The GPI-PLC probe was removed from the membrane shown in panel A by boiling in 1% SDS and rehybridized to a probe specific for both a- and 13-tubulin mRNAs. The membrane was autoradiographed (15-min exposure).
mechanism of down regulation of GPI-PLC activity during transformation of bloodstream-form trypanosomes to the procyclic forms. Excluding the genes for the VSG, relatively few other trypanosome structural genes have been well characterized. Multiple copies of the genes for the housekeeping proteins actin, tubulin, calmodulin, and ubiquitin (4, 35, 37, 39) occur in tandem, varying in copy number from 3 for actin to 13 to 17 copies per haploid genome for both a and 13 genes of tubulin. Multicopy tandem arrangements of genes exist for the glycosomal enzymes fructose-bisphosphate aldolase TABLE 1. Effect of crude procyclic and bloodstream-form fractions on the activity of homogeneous GPI-PLCa
Expt Purified
Trypanosome
no.no. GPI-PLC GPI-PLC
extract
1 2 3
+ + +
None Procyclic, lysate Procyclic, n-octyl glucoside
4 5
-
Procyclic, lysate Procyclic, n-octyl glucoside
6 7
-
Cell equiva- GPI-PLC in extract (106) activity (units)'
~~~lents
3 1
0.5 0.8 0.5
3 1
0.0 0.0
3 1
120.0 57.0
extract -
extract
Bloodstream form, lysate Bloodstream form, n-octyl glucoside extract
GPI-PLC activity was assayed under standard conditions (19). Reaction mixtures contained homogeneous GPI-PLC (0.05 ng, derived from 2 x 105 trypanosomes [19]) and either a hypotonic lysate or a 1% n-octyl glucoside extract of bloodstream- or procyclic-form trypanosomes (cell equivalents are
indicated). b The
indicated GPI-PLC activity is an average of duplicate determinations
(±0.2 U).
FIG. 6. The GPI-PLC polypeptide is detectable in bloodstreamform trypanosomes but not in procyclic forms. Protein from a 1% n-octyl glucoside extract of trypanosome membranes (108 cell equivalents) was quantitatively precipitated with organic solvents (43). Following SDS-polyacrylamide electrophoresis, the proteins were electrotransferred to a nitrocellulose membrane and analyzed for total protein or GPI-PLC polypeptide. (A) Photograph of membrane stained with 0.1% India ink (2). Lanes: 1, proteins from bloodstream-form trypanosomes; 2, proteins from procyclic trypanosomes. (B) A membrane identical to that used in panel A was assayed for GPI-PLC polypeptide by using polyclonal antibody raised against GPI-PLC and an alkaline phosphatase-based detection system. Lanes: 1, protein from bloodstream-form trypanosomes; 2, protein from procyclic trypanosomes. (C) A membrane identical to those in panels A and B was used, except that preimmune serum was used. Protein sizes (in kilodaltons [kd]) are indicated.
(four or five copies) (11) and glyceraldehyde-3-phosphate dehydrogenase (four copies) (28). The gene for triose-phosphate isomerase, another glycosomal enzyme, is found as a single copy per haploid genome (36), as is that for ornithine decarboxylase (31). Our analysis of the developmentally regulated GPI-PLC gene, in both bloodstream and procyclic trypanosomes, showed that it is present as a single copy per haploid genome (Fig. 1, 2, and 4) and that it is located exclusively on a chromosome in the two-megabase compression region (Fig. 3); no copies of the GPI-PLC gene exist on minichromosomes, intermediate-sized chromosomes, or chromosomes remaining in the slot on CHEF electrophoresis (Fig. 3). Genomic restriction analysis reveals two restriction fragment length polymorphisms for NcoI (Fig. 1B and 2A) and another for PvuII (data not shown). From this analysis, we conclude that T. brucei TREU 667 is heterozygotic at the GPI-PLC locus as a result of having two allelic variants of the GPI-PLC gene. The genes for phosphoglycerate kinase, triosephosphate isomerase, and the tubulin gene cluster in T. brucei are also surrounded by polymorphic restriction sites (16). A myriad of mechanisms are used by eucaryotic cells for the control of gene activity. These mechanisms include regulation of (i) the initiation of pre-mRNA synthesis, and (ii) the maturation and/or degradation of the mRNA. Posttranscriptional regulation mechanisms can also control the utilization of mRNA for protein synthesis. Protein activity can further be regulated by covalent modification or by binding of regulatory subunits or cofactors. To understand the basis of the developmental regulation of GPI-PLC of T.
VOL. 10, 1990
DEVELOPMENTAL REGULATION OF T. BRUCEI GPI-PLC
brucei, we tested several of these different levels of regulation of gene activity. The steady-state GPI-PLC mRNA level is much lower in procyclic trypanosomes than in bloodstream forms (Fig. SA). In bloodstream-form trypanosomes, the largest transcript is about 3.7 kb, but a large fraction of the mRNA molecules migrate as a smear in the 1.0- to 2.2-kb range (a 1.1-kb sequence is sufficient to encode the 358-amino-acid GPI-PLC polypeptide [9, 18]). Although the heterogeneity of the GPI-PLC mRNA could be explained by mRNA breakdown during isolation, the use of several mRNA isolation procedures and different RNase inhibitors failed to significantly alter the observed pattern. Furthermore, probing of the same blot for tubulin (Fig. 5B) revealed the expected 1.7and 1.8-kb mRNA for a- and P-tubulin, respectively (37), as the predominant transcripts. It is possible that the tubulin mRNAs are intrinsically more stable than GPI-PLC mRNA. Using ILTAR trypanosomes, Carrington et al. (9) recently reported a homogeneous GPI-PLC mRNA (3.4 kb); the difference between their results and ours, on this issue, could also be due to a relatively higher turnover rate of GPI-PLC mRNA in TREU 667 trypanosomes. Nuclear run-on experiments were conducted to estimate the relative rates of GPI-PLC transcription in bloodstreamform and procycic trypanosomes (our unpublished results, with T. brucei stock 427-60, in collaboration with H. M. Chung and L. Van der Ploeg). Owing to the low rates of GPI-PLC transcription, the results were not definitive, giving a-amanitin-sensitive transcription signals that were only twofold above the background. This transcription of the GPI-PLC gene occurred in both procyclic and bloodstream forms. These observations suggest that procyclic trypanosomes maintain a low level of GPI-PLC mRNA by specifically catabolizing the mRNA. If confirmed, this mode of regulation of GPI-PLC steady-state mRNA level would be similar to the posttranscriptional control mechanism described for the differential expression of the phosphoglycerate kinase gene of T. brucei (17), as opposed to the posttranscriptional regulation mechanism for T. brucei cytochrome c. Although the mRNA for cytochrome c is found in both bloodstream-form and procyclic trypanosomes, the protein is detected only in procyclic forms (38). Some strains of T. brucei, including TREU 667 (used in this study), can express levels of some enzyme activities which do not correlate with the steady-state level of their mRNAs (30). If the small amount of GPI-PLC mRNA in procyclic trypanosomes resulted in the production of a significant amount GPI-PLC polypeptide, regulation of enzyme activity might also occur posttranslationally. To test this hypothesis, we assayed for an inhibitor of the enzyme in procyclic trypanosomes; we found no such factor present in excess (Table 1). Also, using Western blots, we could detect no GPI-PLC polypeptide in procyclic trypanosomes (Fig. 6). These two observations diminish the potential role of posttranslational control mechanisms in the down regulation of GPI-PLC activity during the transformation of bloodstreamform to procyclic trypanosomes. The steady-state mRNA levels of both GPI-PLC and VSG (21) are reduced after differentiation of bloodstream-form trypanosomes to the procycic state. However, it is unlikely that the two genes are coordinately regulated. Whereas our preliminary nuclear run-on experiments indicate that the transcription rates of GPI-PLC genes are similar in both bloodstream-form and procyclic trypanosomes (H. M. Chung and L. Van der Ploeg, unpublished data), the VSG transcription rate is drastically reduced in procycic trypano-
725
somes (20). In addition, GPI-PLC transcription is partially sensitive to a-amanitin (1 mg/ml) (Chung and Van der Ploeg, unpublished), but VSG transcription is resistant (20). ACKNOWLEDGMENTS We thank members of the our laboratory for valuable discussions and suggestions, and we appreciate the outstanding administrative support provided by Shirley Skiles. We thank L. Van der Ploeg and H. M. Chung, who performed the nuclear run-on assays. For comments on the manuscript we thank Abram Gabriel, Gerald W. Hart, David Perez-Morga, and Lex Van der Ploeg. This work was supported by Public Health Service grant A121334 from the National Institutes of Health and by a grant from the McArthur Foundation. K.M.-W. is supported by a postdoctoral fellowship from the Rockefeller Foundation. LITERATURE CITED 1. Alfano, C., and R. McMacken. 1989. Ordered assembly of nucleoprotein structures at the bacteriophage lambda replication origin during the initiation of DNA replication. J. Biol. Chem. 264:10699-10708. 2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. D. Seidman, J. A. Smith, and K. Struhl. (ed). 1989. Current protocols in molecular biology. John Wiley & Sons, Inc., New York. 3. Badley, J. E., G. A. Bishop, T. St. John, and J. A. Frelinger. 1988. A simple, rapid method for the purification of polyA' RNA. Biotechniques 6:114-116. 4. Ben Amar, M. F., A. Pays, P. Tebabi, B. Dero, T. Seebeck, M. Steinert, and E. Pays. 1988. Structure and transcription of the actin gene of Trypanosoma brucei. Mol. Cell. Biol. 8:2166-2176. 5. Bfilow, R., C. Nonnengasser, and P. Overath. 1989. Release of the variant surface glycoprotein during differentiation of bloodstream to procyclic forms of Trypanosoma brucei. Mol. Biochem. Parasitol. 32:85-92. 6. Billow, R., and P. Overath. 1985. Synthesis of a hydrolase for the membrane-form variant surface glycoprotein is repressed during transformation of Trypanosoma brucei. FEBS. Lett. 187:105-110. 7. Bfilow, R. and P. Overath. 1986. Purification and characterization of the membrane-form variant surface glycoprotein hydrolase of Trypanosoma brucei. J. Biol. Chem. 261:11918-11923. 8. Cardoso de Almeida, M. L., and M. J. Turner. 1983. The membrane form of the variant surface glycoproteins of Trypanosoma brucei. Nature (London) 302:349-352. 9. Carrington, M., R. Billow, H. Reinke, and P. Overath. 1989. Sequence and expression of the glycosyl-phosphatidylinositolspecific phospholipase C of Trypanosoma brucei. Mol. Biochem. Parasitol. 33:289-296. 10. Chu, G., D. Vollrath, and R. W. Davis. 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Science 234:1582-1585. 11. Clayton, C. E. 1985. Structure and regulated expression of genes encoding fructose biphosphate aldolase in Trypanosoma brucei. EMBO J. 4:2997-3003. 12. Cunningham, I. 1977. New culture medium for maintenance of tsetse tissues and growth of trypanosomastids. J. Protozool. 24:325-329. 13. Donelson, J. E. 1988. Unsolved mysteries of trypanosome antigenic variation, p. 371-400. In P. T. Englund, and A. Sher (ed.), The biology of parasitism. Alan R. Liss, Inc., New York. 14. Ferguson, M. A., and A. F. Williams. 1988. Cell-surface anchoring of proteins via glycosyl-phosphatidylinositol structures. Annu. Rev. Biochem. 57:285-320. 15. Fox, J. A., M. Duszenko, M. A. Ferguson, M. G. Low, and G. A. Cross. 1986. Purification and characterization of a novel glycanphosphatidylinositol-specific phospholipase C from Trypanosoma brucei. J. Biol. Chem. 261:15767-15771. 16. Gibson, W. C., K. A. Osinga, P. A. M. Michels, and P. Borst. 1985. Trypanosomes of the subgenus Trypanozoon are diploid for housekeeping genes. Mol. Biochem. Parasitol. 16:231-242. 17. Gibson, W. C., B. W. Swinkels, and P. Borst. 1988. Posttran-
726
18.
19.
20.
21.
22. 23. 24.
25.
26.
27. 28.
29.
30.
MENSA-WILMOT ET AL. scriptional control of the differential expression of phosphoglycerate kinase genes in Trypanosoma brucei. J. Mol. Biol. 201: 315-325. Hereld, D., G. W. Hart, and P. T. Englund. 1988. cDNA encoding the glycosyl-phosphatidylinositol-specific phospholipase C of Trypanosoma brucei. Proc. Natl. Acad. Sci. USA 85:8914-8918. Hereld, D., J. L. Krakow, J. D. Bangs, G. W. Hart, and P. T. Englund. 1986. A phospholipase C from Trypanosoma brucei which selectively cleaves the glycolipid on the variant surface glycoprotein. J. Biol. Chem. 261:13813-13819. Kooter, J. M., and P. Borst. 1984. Alpha-amanitin-insensitive transcription of variant surface glycoprotein genes provides further evidence for discontinuous transcription in trypanosomes. Nucleic Acids Res. 12:9457-9472. Kooter, J. M., H. J. van der Spek, R. Wagter, C. E. d'Oliveira, F. van der Hooven, P. J. Johnson, and P. Borst. 1987. The anatomy and transcription of a telomeric expression site for variant-specific surface antigens in T. brucei. Cell 51:261-272. Lamont, G., R. Tucker, and G. Cross. 1986. Analysis of antigen switching rates in Trypanosoma brucei. Parasitology 92:355367. Lanham, S. M., and D. G. Godfrey. 1970. Isolation of salivarian trypanosomes from man and other mammals using DEAEcellulose. Exp. Parasitol. 28:521-534. Low, M. G. 1989. Glycosyl-phosphatidylinositol: a versatile anchor for cell surface proteins. FASEB J. 3:1600-1608. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Masterson, W. J., T. L. Doering, G. W. Hart, and P. T. Englund. 1989. A novel pathway for glycan assembly: biosynthesis of the glycosyl-phosphatidylinositol anchor of the trypanosome variant surface glycoprotein. Cell 56:793-800. Michelotti, E. F., and S. L. Hajduk. 1987. Developmental regulation of trypanosome mitochondrial gene expression. J. Biol. Chem. 262:927-932. Michels, P. A., A. Poliszczak, K. A. Osinga, 0. Misset, J. Van Beeumen, R. K. Wierenga, P. Borst, and F. R. Opperdoes. 1986. Two tandemly linked identical genes code for the glycosomal glyceraldehyde-phosphate dehydrogenase in Trypanosoma brucei. EMBO J. 5:1049-1056. Overath, P., J. Czichos, U. Stock, and C. Nonnengaesser. 1983. Repression of glycoprotein synthesis and release of surface coat during transformation of Trypanosoma brucei. EMBO J. 2: 1721-1728. Parsons, M., and T. Hill. 1989. Elevated phosphoglycerate kinase mRNA but not protein in monomorphic Trypanosoma brucei: implications for stage-regulation and post-transcrip-
MOL. CELL. BIOL.
tional control. Mol. Biochem. Parasitol. 33:215-227. 31. Phillips, M. A., P. Coffino, and C. C. Wang. 1987. Cloning and sequencing of the ornithine decarboxylase gene from Trypanosoma brucei. Implications for enzyme turnover and selective difluoromethylornithine inhibition. J. Biol. Chem. 262:87218727. 32. Richet, E., P. Abcarian, and H. A. Nash. 1986. The interaction of recombination proteins with supercoiled DNA: defining the role of supercoiling in lambda integrative recombination. Cell 46: 1011-1021. 33. Roditi, I., H. Schwarz, T. W. Pearson, R. P. Beecroft, M. K. Liu, J. P. Richardson, H. J. Buhring, J. Pleiss, R. Bulow, R. 0. Williams, and P. Overath. 1989. Procyclin gene expression and loss of the variant surface glycoprotein during differentiation of Trypanosoma brucei. J. Cell Biol. 108:737-746. 34. Shapiro, S. Z. 1986. Trypanosoma brucei: release of variant surface glycoprotein during the parasite life cycle. Exp. Parasitol. 61:432-437. 35. Swindle, J., J. Ajioka, H. Eisen, B. Sanwal, C. Jacquemot, Z. Browder, and G. Buck. 1988. The genomic organization and transcription of the ubiquitin genes of Trypanosoma cruzi. EMBO J. 7:1121-1127. 36. Swinkels, B. W., W. C. Gibson, K. A. Osinga, R. Kramer, G. H. Veeneman, J. H. van Boom, and P. Borst. 1986. Characterization of the gene for the microbody (glycosomal) triosephosphate isomerase of Trypanosoma brucei. EMBO J. 5:1291-1298. 37. Thomashow, L. S., M. Milhausen, W. J. Rutter, and N. Agabian. 1983. Tubulin genes are tandemly linked and clustered in the genome of trypanosoma brucei. Cell 32:35-43. 38. Torri, A. F., and S. L. Hajduk. 1988. Posttranscriptional regulation of cytochrome c expression during the developmental cycle of Trypanosoma brucei. Mol. Cell. Biol. 8:4625-4633. 39. Tschudi, C., A. S. Young, L. Ruben, C. L. Patton, and F. F. Richards. 1985. Calmodulin genes in trypanosomes are tandemly repeated and produce multiple mRNAs with a common 5' leader sequence. Proc. Natl. Acad. Sci. USA 82:3998-4002. 40. Van der Ploeg, L. H., A. W. Cornelissen, J. D. Barry, and P. Borst. 1984. Chromosomes of kinetoplastida. EMBO J. 3: 3109-3115. 41. Van der Ploeg, L. H., A. W. Cornelissen, P. A. Michels, and P. Borst. 1984. Chromosome rearrangements in Trypanosoma brucei. Cell 39:213-221. 42. Weitz, B. G. F. 1960. A soluble protective antigen of Trypanosoma brucei. Nature (London) 185:788-789. 43. Wessel, D., and U. I. Flugge. 1984. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 138:141-143. 44. Zeff, R. A., and J. Geliebter. 1987. Oligonucleotide probes for genomic DNA blots. Focus 9:1-2.
