The Primary Structure ofTrypanosoma(Nannomonas)congolenseVariant Surface Glycoproteins

The Primary Structure of Trypanosoma (Nannomonas) congolenseVariant Surface Glycoproteins

TOYOHIKO URAKAWA ,1 YUKI ESHITA,2 AND PHELIX A. O. MAJIWA3

International Livestock Research Institute, P.O. Box 30709, Nairobi, Kenya

URAKAWA , T., ESHITA, Y., AND MAJIWA, P. A. O. 1997. The primary structure ofTrypanosoma(Nannomonas) congolensevariant surface glycoproteins.Experimental Parasitology85, 215–224.The complete nucleotide sequences were determined for three transcripts each encoding a differentvariant surface glycoprotein (VSG) ofTrypanosoma(Nannomonas) congolense.The nucleotidesequence was determined also for a transcript encoding a fourth VSG, but this was truncated. The dataobtained confirm absence of the canonical polyadenylation signal, lack of conserved sequence ele-ments in the 39 untranslated region, and heterogeneity in the spliced-leader acceptor site in theT.congolenseVSG transcripts examined. A comparison of the amino acids deduced from the nucleotidesequences of the four VSGs and those of other VSGs published previously reveals a strong conser-vation of several structural domains, particularly cysteine residues located throughout most of themolecules. The majority ofT. congolenseVSGs analyzed in this study resemble most the N-terminalcysteine residue domain type B ofT. brucei,characterized by a cysteine residue located toward theN-terminal end, a cluster of cysteine residues in the central region, and at least three cysteine residuesbetween positions 250 and 300 of the molecules. One of the VSGs analyzed, ILNat3.3, did not fit intoany of the classification schemes proposed for the VSGs so far studied, and thus may represent adifferent class of these surface molecules. Unlike VSGs ofT. brucei,the T. congolenseVSGs haveno cysteine residues at the carboxy-terminal end. These data now make it possible to predict generalprimary structural features ofT. congolenseVSGs. © 1997 Academic Press

INDEX DESCRIPTORS ANDABBREVIATIONS: ILNaR, ILRAD Nannomonasantigenic repertoire; PCR,polymerase chain reaction;Trypanosoma(Nannomonas) congolense;VSG, variant surface glyco-protein.

INTRODUCTION

The surface of the African trypanosome iscovered by the variant surface glycoprotein(VSG) at two different stages of its life cycle:the bloodstream-form stage and the metacyclic-form stage. The VSG facilitates survival of theparasite in the blood of an immunocompetentmammalian host by helping the trypanosomeescape recognition by antibodies and helper T-cells (Reinitz and Mansfield 1990). A trypano-some changes its antigenic identity by the ex-pression of alternate immunologically distinct

VSGs in the course of parasitemia in an infectedhost and during growth in tissue culture asbloodstream forms (Doyleet al. 1980) or meta-cyclic forms (Prain and Ross 1988). One try-panosome can generate antigenically distinctVSGs either by a process that assorts differentsegments of other VSG genes and pseudogenes(Thonet al.1990) or by introducing changes inthe nucleotide composition of the VSG generesulting in a protein with an altered amino acidsequence (Luet al. 1993).

The amino acid composition of severalTry-panosoma bruceiVSGs has been determinedeither by direct sequencing of the purified pro-tein or by prediction from the nucleotide se-quence of the corresponding cDNAs (Carring-ton et al.1991). However, little information ex-ists on the amino acid composition of the VSGsof the other species of African trypanosomes,namelyT. vivaxandT. congolense.Since VSGsperform similar functions in different species of

1 Present address: Society for Techno-Innovation of Ag-riculture, Forestry and Fisheries, National Institute of Ani-mal Health, Laboratory of Immunogenetics, 3-1-1 Kannon-dai Tsukuba, Ibaraki 305, Japan.

2 Present address: Department of Parasitology, KurumeUniversity School of Medicine, 67 Asahi-machi, Kurume-shi, Fukuoka 830, Japan.

3 To whom correspondence should be addressed. Fax:(254-2) 631499; E-mail: [email protected].

EXPERIMENTAL PARASITOLOGY 85, 215–224 (1997)ARTICLE NO. PR964140

2150014-4894/97 $25.00Copyright © 1997 by Academic PressAll rights of reproduction in any form reserved.

the African trypanosomes, it is useful to deter-mine the extent of divergence among VSGs ofthe different species of trypanosomes and to es-tablish the level of conservation, if any, of ele-ments of the three-dimensional structure amongthem.

We present here nucleotide sequences of full-length transcripts which encode three differ-ent VSGs ofT. congolenseclones derived fromtwo distinct antigenic repertoires: ILRADNannomonasantigenic repertoire (ILNaR)2 andILNaR3. Additionally, we show thatT. congo-lensetranscripts, like those ofT. brucei,havedegenerate splice acceptor sites for theminiexon (Layden and Eisen 1988). But, unlikeT. brucei VSG transcripts, theT. congolenseVSG transcripts lack a 14-mer of nucleotidesconserved within the 39 untranslated region(Rice-Fichtet al.1981; Matthyssenset al.1981;Majumder et al. 1981). Alignment of the de-duced amino acid sequences of theT. congo-lenseVSGs revealed that the longest homolo-gous stretches between any two of the mol-ecules comprise at most seven amino acidresidues. Structurally, the majority of theT.congolenseVSGs resemble most the VSGswithin the N-terminal cysteine domain type B ofT. brucei(Carringtonet al. 1991).

The average number of amino acid residuesin each of the VSGs analyzed in the presentstudy is 440; this corresponds to a protein ofMr

48.9. The predicted molecular weights of theseVSGs are within the range of sizes reported fornative VSGs ofT. congolenseIL3000 metacy-clic forms (Fishet al. 1989), the VSGs pro-ducedin vitro by translation ofT. congolensemRNA (Cook et al. 1985), and the sizes pre-dicted from nucleotide sequences ofT. congo-lense YNat1.1 and YNat1.3 (Strickleret al.1987). The size differences between the mo-lecular weights calculated from the nucleotidesequences and either the purified antigens orthose detected in the lysates of trypanosomes bymonospecific antibodies can be accounted forby the in vivo modification, i.e., glycosylationand processing of both amino- and carboxy-termini. On the average, the VSGs ofT. congo-lenseare smaller than those ofT. brucei,which

are 60–65 kDa. Other than the conserved struc-tural motifs consisting primarily of cysteineresidues, there are no other significant similari-ties between VSGs ofT. congolenseand thoseof T. brucei.The data presented here togetherwith those already published (Strickleret al.1987; Eshitaet al. 1992; Rauschet al. 1994)make it possible to includeT. congolenseVSGsin considerations for the prediction of generalstructural features of the VSGs of African try-panosomes.

MATERIALS AND METHODS

cDNA cloning.The cloning and characterization of par-tial transcripts encoding ILNat2.1 VSG, mVSG1, andmVSG2 expressed by trypanosome clones in the ILNaR2have been described (Majiwaet al. 1985; Eshitaet al.1992). Sequences which encode the IL3000 VSG, ILNat3.2,and ILNat3.3 were similarly identified in cDNA librariesmade from mRNA of the trypanosome clones homoge-neously expressing the respective VSGs. They were eachshown to contain sequences encoding bloodstream-formvariant-specific mRNAs by the differential colony hybrid-ization (Grunstein and Hogness 1975) and specific hybrid-ization with only mRNA (Thomas 1980) prepared frombloodstream-form trypanosome clones expressing the re-spective variant antigenic types. Additionally, cell-freetranslation products made from mRNAs affinity selected bythe cDNA inserts could be specifically immunoprecipitatedonly by the monospecific antisera raised against the VSGsof the respective trypanosome clones.

PCR and sequencing methodologies.Recombinant plas-mids containing the VSG cDNAs were digested with re-striction enzymes to produce smaller insert fragments whichwere subcloned and then sequenced on both strands; alter-natively, the recombinant plasmids were linearized, unidi-rectionally deleted with exonuclease III, recloned in thesame pUC vector (Vieira and Messing 1988), and se-quenced on both strands using the dideoxynucleotide chaintermination method (Sangeret al. 1977).

For the cDNA clones representing incomplete transcripts,sequences at the 59 and 39 ends of the transcripts wereobtained by PCR amplification of single-stranded cDNAusing, in each case, an oligonucleotide primer appropriatelylocated at an internal position in the insert and either oli-go(dT) or an oligonucleotide complementary to theminiexon ofT. congolense.

The PCR amplification reactions were performed in a50-ml reaction containing 10 mM Tris z HCl, pH 8.3, 1.5mM MgCl2, 50 mM KCl, 150 mM each deoxynucleotidetriphosphate, 100 ng of each oligonucleotide primer, and 1.0unit of Taq DNA polymerase (Promega Biotech, U.S.A.).The reaction mixtures were layered with mineral oil andplaced in a microprocessor-controlled thermal cycler pro-grammed to incubate the samples at 94°C for 1 min, 55°C

URAKAWA, ESHITA, AND MAJIWA216

for 1 min, when using the miniexon primer, or 45°C whenusing the oligo(dT) primer. In each case, the synthesis re-action was performed at 72°C for 1.5 min. At the end of 30cycles, the synthesis reaction was continued for 10 min andthen cooled to 4°C. The PCR products were recovered byphenol extraction and ethanol precipitation, before cloninginto a T-vector. At least three independent clones of thePCR product were subjected to nucleotide sequence deter-mination on both strands.

RESULTS AND DISCUSSION

The nucleotide sequences of four cDNAclones of transcripts, each encoding a differentbloodstream form VSG ofT. congolense,wereanalyzed (Table I). Each of the cDNAs had aminiexon at the 59 end and two of them, i.e.,those encoding ILNat2.1 and ILNat3.2, con-tained poly(A) at the 39 end. Although the se-quence encoding IL3000 VSG had four A resi-dues at its 39 end, it was not possible to deter-mine whether this was an authentic portion ofthe poly(A) tract since no other cDNA clones ofthis VSG were found that had longer tracts ofpoly(A) residues, and other VSG transcripts, forexample, ILNat3.2, were found which had vari-able lengths of A residues interrupted by othernucleotides before the poly(A) homopolymer atthe 39 terminus (data not shown). Attempts toobtain a cDNA containing the 39 end of thetranscript encoding ILNat3.3 were unsuccess-ful; thus, the sequence as presented for this

cDNA is incomplete. These sequences havebeen deposited with the GenBank and the Ac-cession No. assigned to each one is indicated inTable I.

In T. brucei,there exists a 14-mer of nucleo-tides which forms the core of a conserved se-quence element within the 39 UTR of all theVSG transcripts examined so far in this speciesof the African trypanosomes (Rice-Fichtet al.1981; Majumderet al. 1981; Matthyssenset al.1981). Although no specific function has beenascribed to this element, circumstantial evi-dence suggests that it is a component of a largerelement with a role in the stage-specific regu-lation of VSG mRNA abundance (Berberofetal. 1995). The nucleotide sequences locatedwithin the 39 end vicinity of transcripts encod-ing T. congolenseVSGs were examined forpresence of any conserved sequence elementsbut none was found (Fig. 1). Instead, theT.congolenseVSG transcripts were found to pos-sess unusually short untranslated regions be-tween the termination codon and the poly(A)tail. In the transcript which encodes theILNat2.1 VSG, for example, there was no UTR,and the termination codon appeared to be a partof the poly(A). In the mVSG1 transcript, theUTR is only 7 nt long; in the mVSG2 transcript,the UTR is 24 nt long and in the one encoding

TABLE IT. congolenseVSGs Analyzed to Date

VSG Serodeme Stage

Number of

Predictedsize (kDa)a Accession No.

Aminoacids

Cysteineresidues

Potential N-linkedglycosylation sites

mVSG1 ILNaR2 Metacyclic 435 8 3 46.2 M74802mVSG2 ILNaR2 Metacyclic 451 10 1 48.1 M74803IL3000 bVSG ILNaR2 Bloodstream 436 8 4 47.0 U07140ILNat2.1 ILNaR2 Bloodstream 445 9 4 47.6 U07141ILNat3.2 ILNaR3 Bloodstream 444 8 4 47.7 U07142ILNat3.3 ILNaR3 Bloodstream (365)b (11) (3) ? U07143BENat1 BENaR1 Bloodstream 429 11 5 45.6 X79399BENat1.2 BENaR1 Bloodstream 449 8 3 47.7 X79400BENat1.3 BENaR1 Bloodstream 428 7 3 45.4 X79401YNat1.1 YNaR1 Bloodstream 418 9 7 44.9 M15112YNat1.3 YNaR1 Bloodstream 412 9 5 43.8 M15113

a Predicted size of preprotein.b Incomplete sequence.

Trypanosoma congolenseVARIANT SURFACE GLYCOPROTEINS 217

ILNat3.2, the UTR is 25 nt long. This observa-tion appears to be true for the transcripts encod-ing the YNat1.1, YNat1.3, BENat1, andBENat1.2 VSGs (Strickleret al. 1987; Rauschet al. 1994). Thus, in general, the UTRs ofT.congolenseVSG transcripts tended to be shorterthan those of theT. bruceiVSG-encoding tran-scripts.

In trypanosomatids, the selection of the poly-adenylation site of the pre-mRNA in a polycis-tronic transcript is coupled totrans-splicing ofthe miniexon to the transcript of the gene lo-cated downstream (Ulluet al. 1993; LeBowitzet al. 1993). However, correct polyadenylationof the individual transcripts within the polycis-tron does not require the presence of splice ac-ceptor site of a downstream gene (Hehlet al.1994). From the data shown in Fig. 1, it is ap-parent that inT. congolense,as inT. brucei,themRNAs are processed and polyadenylated

without an apparent AAUAAA hexamer withintheir 39 untranslated regions (Huang and van derPloeg 1991; Schu¨rch et al. 1994). Although thesequence AAUAAA, which is involved in thepolyadenylation of mRNAs (McLauchlanet al.1985), was present in some of the VSG-encoding sequences, for example, in ILNat3.2(Fig. 1), none of them was located within theUTRs. Furthermore, no such sequences wereobserved in the transcripts which encode threeother T. congolenseVSGs: IL3000, ILNat2.1,and ILNat3.3. It is therefore unlikely that theseare the sequences that ensure accurate polyad-enylation of the transcripts. Although trypano-some transcripts do not possess an apparentAAUAAA sequence, other sequences withintheir 39 end segments most likely mediate theregulatory interactions that achieve correctpolyadenylation of the mRNA transcripts. Suchsequences remain to be functionally identified

FIG. 1. Sequences encodingT. congolenseVSGs lack the 14 nucleotide residues conserved in the sequencesencodingT. bruceiVSGs. Shown are both the nucleotide and the corresponding amino acid sequences markingthe 39 terminal end of transcripts and ending in the poly(A) tracts. The termination codons are underlined andindicated with a star in each case. The 14-mer conserved in the sequences of the transcripts encodingT. bruceiVSGs is underlined. Sequence encodingT. b. brucei ILTat1.3 is from Rice-Fichtet al. (1981; GenBankAccession No. J01221) and that encodingT. b. gambienseis from Thi et al. (1991; GenBank Accession No.M62629).


since they are not obvious from inspection ofthe nucleotide sequences.

All messenger RNAs of trypanosomatidspossess at their 59 ends 39 nt constituting theminiexon, or spliced leader, which is added bytrans-splicing to each one during the maturationprocess (for review see Borst 1986). Theminiexon and the main exon mRNAs are tran-scribed from genes located on different chro-mosomes (Guyauxet al.1985). In general, thereappears to be no strict requirement for an ac-ceptor site at which the miniexon must be in-serted on the pre-mRNA (Tschudiet al. 1985;Layden and Eisen 1988; Kapleret al. 1990);therefore, provided other appropriate recogni-tion signals and the AG dinucleotide are pre-sent, the miniexon can be added anywherewithin 30–70 bases from the start of the trans-lation codon.

We examined the generality of this observa-tion by determining the location of the mini-

exon relative to the presumed translation startcodon of mRNAs encoding VSGs and a cyste-ine protease ofT. congolense.The cysteine pro-tease is strongly conserved among different iso-lates ofT. congolense,irrespective of the anti-genic repertoire (Authieet al.1992; Jaye 1993);in contrast the transcripts encoding VSGs arehighly variable among different variant anti-genic types ofT. congolensewithin a singleantigenic repertoire (Majiwaet al.1985, 1986).Figure 2 shows the nucleotide sequences of theregion spanning the translation start codon (atposition +1) to the first nine nucleotides of theminiexon in the PCR products of transcriptswhich encode mVSG2, ILNat2.1, and aT. con-golensecysteine protease (GeneBank Acces-sion No. L25130). The miniexon appears tohave been added at several different locations inthe 59 region on each of the transcripts exam-ined. Furthermore, each of the transcripts ana-lyzed contained more than one potential 39 ac-

FIG. 2. Diversities in the miniexon splice acceptor site inT. congolensetranscripts. Nucleotide sequences weredetermined for the different clones of PCR products which contained the miniexon and an internal segment ofthe respective cDNA. The ATG initiation codon is highlighted and the A residue within this codon is assignedposition +1. Nine nucleotide residues constituting part of the 39 end of the miniexon are aligned and shown onthe left of each of the sequences. Gaps were introduced to retain alignment of both the miniexon (on the left)and the ATG initiation codons (on the right) in the sequences of the different recombinant clones. For eachtranscript, numbers on the right designate the actual number of the recombinant clone whose nucleotide sequencewas determined. Different numbers of clones had identical nucleotide sequences. The AG residues, which havea role in the splicing process, are underlined.


ceptor site (AG). Sixty-seven percent of themVSG2 cDNA clones had the miniexon in-serted at the same position compared with 56%of the ILNat2.1 cDNAs and 50% of the cysteineprotease cDNAs. Among the transcripts ana-lyzed, the largest number of nucleotides be-tween the initiator methionine codon and theminiexon was 115 nt in the cysteine protease,and the shortest was 18 nucleotides, found insome of the transcripts which encode mVSG2.

There was no apparent pattern to the pre-ferred splice acceptor site for the miniexon (Fig.2). Such a flexibility in the addition of theminiexon to kinetoplastid mRNAs has been re-ported for transcripts which encode calmodulinsin T. brucei(Tschudiet al. 1985), VSGs ofT.equiperdum(Layden and Eisen, 1988), and thedihydrofolate reductase–thymidylate synthaseof Leishmania major(Kapleret al.1990). Thusin T. congolense,as in other trypanosomatids,the miniexon is added at variable distances rela-tive to the predicted initiator methionine codonin any group of transcripts encoding a protein.

Since VSGs are a family of surface proteinswith a similar function, they are expected tohave elements of conserved structure, irrespec-tive of species of the trypanosome from which aparticular VSG is isolated (Blumet al.1993). Inorder to clarify the nature of such elements, acomparative multiple alignment was performedof the predicted amino acid sequences of 11T.congolenseVSGs, 4 of which were determinedin the present study and 7 others for which se-quences are available in public domain data-bases (Strickleret al. 1987; Eshitaet al. 1992;Rauschet al. 1994). As shown in Fig. 3, thereare several amino acid residues whose relativepositions are conserved in all theT. congolenseVSGs analyzed so far. One possibly significantblock of conserved amino acid residues has anapproximate consensus GxSLVnDLLCLC andis located within the central domain of theseproteins. On average, the majority of the con-served amino acids are cysteine residues whichare necessary for the formation of disulfidebonds and are thus essential for stabilization ofsecondary structure (Strickleret al.1987; Blumet al.1993). The predicted amino acid sequence

of ILNat3.3 VSG displays features which differfrom those observed in the other VSGs withwhich it was compared. A typical example ofthis is the absence of the CLC motif found in thecentral domain of a majority of theT. congo-lenseVSGs. This motif appears as CTC in theILNat3.3 and is located not centrally but atamino acid residue position 113–115 of themolecule. Furthermore, in the region betweenamino acid residues 200 and 300, this VSG hasa higher proportion of proline residues than theothers. Whether ILNat3.3 represents a class ofVSGs different from those proposed for theVSGs analyzed to date must await further in-vestigations. However, conservation in the rela-tive positions of the amino acid residues thoughtto be essential for integrity of the three-dimensional structures of these proteins (Figs. 3and 4) suggests that the VSGs in general have asimilar structure, a feature that will be deter-mined conclusively when a number representa-tive of each group have been crystallized.

The VSGs are anchored to the cell membranethrough covalent attachment of the C-terminalamino acid residue to a glycosylphosphatidyli-nositol (GPI; Fergusonet al.1988), added post-translationally after cleavage of the carboxy-terminal signal sequence. In order to predict thesite of cleavage of the signal sequence and ad-dition of the GPI anchor, we applied thev,v + 2 rule (Kodukulaet al. 1993) to the de-duced amino acid sequence of theT. congolenseVSGs. In four of the VSGs, the amino acidresidues expected at thev andv + 2 sites werefound within less than 30 residues from the car-boxy-terminal end. These are SNS in mVSG1,SGS in IL3000 and YNat1.1, and SSG inBENat1.1. In each case these residues are simi-lar or identical to those which have been experi-mentally determined in other GPI-anchoredproteins (Kodukulaet al. 1993). Although thesame amino acid residues are not evident insimilar relative positions in the otherT. congo-lenseVSGs examined, there are in these posi-tions amino acids with similar properties andwhich therefore fulfill the requirements for ad-dition of the GPI anchor. The deduced N-terminal leader sequences of theT. congolense


VSGs appeared to be typical of secreted pro-teins in possessing, within the N-terminal signalpeptide, hydrophobic amino acids between hy-drophilic residues and a basic amino acid resi-due toward the N-terminus. The site for cleav-age of the predicted N-terminal signal peptide

would be expected to occur within the first 40amino acid residues from the putative initiatormethionine.

Based upon the distribution of conserved cys-teine residues,T. bruceiVSGs can be classifiedinto N-terminal or C-terminal structural domain

FIG. 3. Amino acid residues conserved amongT. congolenseVSGs. The amino acid residues predicted fromthe nucleotide sequences were aligned using Clustal V program (Higginset al. 1992) with the followingparameters: fixed gap penalty, 10; floating point penalty, 10; protein weight matrix, PAM 250. Some of theamino acid residues which appear in the same relative position in at least three of the different VSGs are shaded.Sequences of BENat1, BENat1.2, and BENat1.3 are from Rauschet al.1994; those of YNat1.1 and YNat1.3 arefrom Strickleret al. 1987.


types (Carringtonet al. 1991). TheT. congo-lenseVSGs analyzed in this study and thoseanalyzed previously (Strickleret al. 1987;Rauschet al. 1994) belong to the N-terminalcysteine domain type B. The VSGs in this do-main type are characterized by the presence of acysteine residue toward the N-terminal end, acluster of cysteine residues in the central region,and proline residues in the central part and thecarboxy-terminal half of the molecules. UnlikeT. brucei VSGs, however, theT. congolenseVSGs have no cysteine residues at the extremecarboxy-terminal end (Fig. 4). Collectively,these data make it possible to includeT. congo-lenseVSGs in considerations that lead to theprediction of general structural features of theAfrican trypanosome VSGs.

REFERENCES

AUTHIE, E., MUTETI, D. K., MBAWA, Z. R., LONSDALE-ECCLES, J. D., WEBSTER, P., AND WELLS, C. W. 1992.Identification of a 33-kDa immunodominant antigen ofTrypanosoma congolenseas a cysteine protease.Molecu-lar and Biochemical Parasitology56, 103–116.

BERBEROF, M., VANHAMME , L., TEBABI, P., PAYS, A., JEF-FERIES, D., WELBURN, S., AND PAYS, E. 1995. The 39-

terminal region of the mRNAs for VSG and procyclin canconfer stage specificity to gene expression inTrypanosomabrucei. European Molecular Biology Organization Jour-nal 14, 2925–2934.

BLUM, M. L., DOWN, J. A., GURNETT, A. M., CARRINGTON,M., TURNER, M. J., AND WILEY, D. C. 1993. A structuralmotif in the variant surface glycoproteins ofTrypano-soma brucei. Nature362,603–609.

BORST, P. 1986. Discontinuous transcription and antigenicvariation in trypanosomes.Annual Review of Biochemis-try 55, 701–732.

CARRINGTON, M., MILLER, N., BLUM, M., RODITI, I., WILEY,D., AND TURNER, M. 1991. Variant specific glycoproteinof Trypanosoma bruceiconsists of two domains eachhaving an independently conserved pattern of cysteineresidues.Journal of Molecular Biology221,823–835.

COOK, G. A., HONIGBERG, B. M., AND ZIMMERMANN , R. A.1985. Isolation and cell-free synthesis of variant surfaceglycoproteins fromTrypanosoma congolense. Molecularand Biochemical Parasitology15, 281–293.

DOYLE, J. J., HIRUMI, H., HIRUMI, K., LUPTON, E. N., AND

CROSS, G. A. M. 1980. Antigenic variation in clones ofanimal-infectiveTrypanosoma bruceiderived and main-tainedin vitro. Parasitology80, 359–369.

ESHITA, Y., URAKAWA , T., HIRUMI, H., FISH, W. R., AND

MAJIWA, P. A. O. 1992. Metacyclic-form-specific variantsurface glycoprotein-encoding genes ofTrypanosoma(Nannomonas) congolense. Gene113,139–148.

FERGUSON, M. A. J., HOMANS, S. W., DWEK, R. A., AND

RADEMACHER, T. W. 1988. Glycosylphosphotidylinositol

FIG. 4. T. congolenseVSGs belong to cysteine residue domain type B. Line diagram summary of the primarystructure ofT. congolenseVSGs, indicating the position of the potential N-linked glycosylation sites (filledcircles) and conserved cysteine residues (vertical bars). Similar data compiled from amino acid sequences ofT. bruceiVSGs are shown for comparison.


moiety that anchorsTrypanosoma bruceivariant surfaceglycoprotein to the membrane.Science239,753–759.

FISH, W. R., MURIUKI, C. W., MUTHIANI , A. M., GRAB,D. J.,AND LONSDALE-ECCLES, J. D. 1989. Disulfide bondinvolvement in the maintenance of the cryptic nature ofthe cross-reacting determinant of metacyclic forms ofTrypanosoma congolense. Biochemistry28, 5414–5421.

GRUNSTEIN, M., AND HOGNESS, D. 1975. Colony hybridiza-tion: A method for the isolation of cloned DNAs thatcontain a specific gene.Proceedings of the NationalAcademy of Sciences USA72, 3961–3965.

GUYAUX , M., CORNELISSEN, A. W. C. A., PAYS, E., STEIN-ERT, M., AND BORST, P. 1985.Trypanosoma brucei:Asurface antigen mRNA is discontinuously transcribedfrom two distinct chromosomes.European Molecular Bi-ology Organization Journal4, 995–998.

HEHL, A., VASSELLA, E., BRAUN, R., AND RODITI, I. 1994. Aconserved stem–loop structure in the 39 untranslated re-gion of procyclin mRNAs regulates expression inTry-panosoma brucei. Proceedings of National Academy ofSciences USA91, 370–374.

HIGGINS, D. G., BLEASBY, A. J., AND FUCHS, R. 1992.CLUSTAL V: Improved software for multiple sequencealignment.Computer Applications in Biological Sciences8, 189–191.

HUANG, J.,AND VAN DER PLOEG, L. H. T. 1991. Maturationof polycistronic pre-mRNA inTrypanosoma brucei:Analysis of trans-splicing and poly (A) addition to na-scent RNA transcripts from thehsp70 locus.Molecularand Cellular Biology11, 3180–3190.

JAYE, A. 1993. Characterisation of aTrypanosoma(Nannomonas) congolense-specific antigen: Identifica-tion as a thiol protease precursor. Ph.D. thesis, BrunelUniversity, Middlesex, UK.

KAPLER, M. G., ZHANG, K., AND BEVERLY, S. M. 1990.Nuclease mapping and DNA sequence analysis of tran-scripts from the dihydrofolate reductase-thymidylate syn-thase (R) region ofLeishmania major. Nucleic Acids Re-search18, 6399–6408.

KODUKULA, K., GERBER, L. D., AMTHAUER, R., BRINK, L.,AND UDENFRIEND, S. 1993. Biosynthesis of glycosylphos-phatidylinositol (GPI)-anchored membrane proteins in in-tact cells: Specific amino acid requirements adjacent tothe site of cleavage and GPI attachment.The Journal ofCell Biology120,657–664.

LAYDEN, R. E.,AND EISEN, H. 1988. Alternatetrans-splicingin Trypanosoma equiperdum:Implications for splice siteselection.Molecular and Cellular Biology8, 1352–1360.

LEBOWITZ, J. H., SMITH, H. Q., RUSCHE, L., AND BEVERLY,S. M. 1993. Coupling of poly(A) site selection andtrans-splicing inLeishmania. Genes and Development7, 996–1007.

LU, Y., HALL , T., GAY, L. S., AND DONELSON, J. E. 1993.Point mutations are associated with a gene duplicationleading to the bloodstream re-expression of a trypano-some metacyclic VSG.Cell 72, 397–406.

MAJIWA, P. A. O., MATTHYSSENS, G., WILLIAMS , R. O.,AND

HAMERS, R. 1985. Cloning and analysis ofTrypanosoma(Nannomonas) congolenseILNat2.1 VSG gene.Molecu-lar and Biochemical Parasitology16, 97–108.

MAJIWA, P. A. O., YOUNG, J. R., HAMERS, R., AND MAT-THYSSENS, G. 1986. Minichromosomal variable surfaceglycoproteins and molecular karyotypes ofTrypanosoma(Nannomonas) congolense. Gene41, 183–192.

MAJUMDER, H. K., BOOTHROYD, J. C.,AND WEBER, H. 1981.Homologous 39-terminal regions of mRNAs for surfaceantigens of different antigenic variants ofTrypanosomabrucei. Nucleic Acids Research9, 4745–4753.

MATTHYSSENS, G., MICHIELS, F., HAMMERS, R., PAYS, E.,AND STEINERT, M. 1981. Two variant surface glycopro-teins of Trypanosoma bruceihave a conserved C-terminus.Nature293,230–233.

MCLAUCHLAN, J., GAFFREY, D., WHITTON, J. L., AND CLE-MENS, J. B. 1985. The consensus sequence YGTTYY lo-cated downstream from AATAAA signal is required forefficient formation of mRNA 39 termini. Nucleic AcidsResearch13, 1347–1368.

PRAIN, C. J.,AND ROSS, C. A. 1988.Trypanosoma congo-lense:Interactions between trypanosomes expressing dif-ferent metacyclic variable antigen typesin vitro and invivo. Parasitology97, 277–286.

RAUSCH, S., SHAYAN , P., SALNIKOFF, J., AND REINWALD, E.1994. Sequence determination of three variable surfaceglycoproteins fromTrypanosoma congolense. EuropeanJournal of Biochemistry223,813–821.

REINITZ, D. M., AND MANSFIELD, J. M. 1990. T-cell-independent and T-cell-dependent B-cell responses to ex-posed variant surface glycoprotein epitopes in trypano-some-infected mice.Infection and Immunity58, 2337–2342.

RICE-FICHT, A. C., CHEN, K. K., AND DONELSON, J. E. 1981.Sequence homologies near the C-termini of the variablesurface glycoproteins ofTrypanosoma brucei. Nature294,53–57.

SANGER, F., NICKLEN, S., AND COULSON, A. R. 1977. DNAsequencing with chain-termination inhibitors.Proceed-ings of the National Academy of Sciences USA74,5463–5467.

SCHURCH, N., HEHL, A., VASSELLA, E., BRAUN, R., AND

RODITI, I. 1994. Accurate polyadenylation of procyclinmRNAs inTrypanosoma bruceiis determined by pyrim-idine-rich elements in the intergenic region.Molecularand Cellular Biology14, 3668–3675.

STRICKLER, J. E., BINDER, D. A., L’I TALIEN, J. J., SHIMA -MOTO, G., WAITE, S. W., DALHEIM , L. J., NOVOTNY, J.,RADDING, J. A., KONIGSBERG, W. H., ARMSTRONG,M. Y. K., RICHARDS, F. F.,AND LALOR, T. M. 1987.Try-panosoma congolense:Structure and molecular organi-zation of the surface glycoproteins of two early blood-stream variants.Biochemistry26, 796–805.

THI, D., STEINERT, M., PAYS, E., AND AERTS, D. 1991. Highhomology between variant surface glycoprotein gene ex-pression sites ofTrypanosoma bruceiandTrypanosoma


gambiense. Molecular and Biochemical Parasitology48,199–210.

THOMAS, T. 1980. Hybridization of denatured RNA andsmall DNA fragments transferred to nitrocellulose.Pro-ceedings of the National Academy of Sciences USA77,5201–5205.

THON, G., BALTZ, T., GIROUD, C., AND EISEN, H. 1990.Trypanosome variable surface glycoproteins: Compositegenes and order of expression.Genes and Development4,1374–1383.

TSCHUDI, C., YOUNG, A. S., RUBEN, L., PATTON, C. L., AND

RICHARDS, F. F. 1985. Calmodulin genes in trypanosomesare tandemly repeated and produce multiple mRNAs with

a common 59 leader sequence.Proceedings of the Na-tional Academy of Sciences USA82, 3998–4002.

ULLU, E., MATHEWS, K. R., AND TSCHUDI, C. 1993. Tempo-ral order of RNA-processing reactions in trypanosomes:Rapid trans-splicing precedes polyadenylation of newlysynthesized tubulin transcripts.Molecular and CellularBiology 13, 720–725.

VIEIRA, J., AND MESSING, J. 1988. Production of single-stranded plasmid DNA.Methods in Enzymology153,3–11.

Received 8 February 1996; accepted with revision 16 De-cember 1996


The Primary Structure ofTrypanosoma(Nannomonas)congolenseVariant Surface Glycoproteins

Documents

Transcript of The Primary Structure ofTrypanosoma(Nannomonas)congolenseVariant Surface Glycoproteins