P H A S E V A R I A T I O N OF LPS IN H. INFLUENZA nan, 1982; Raetz and Dowhan, 1990). If adenylate kinase has similar functions in H. influenzae as in E. coil, regulation of adk expression in lie3 in conjunction with galE seems reasonable. It provides an attractive mechanism by which the major integral components of the outer membrane, LPS and phospholipid, could be coordinately regulated and related to cellular energy balance and growth rate. The ability to respond rapidly to appropriate environmental signals with rapid replication would be advantageous in promoting survival.

Variation de phase dn lipopolyoside chez Haemophilus influenzae Les caractdristiques structurelles du lipopolyoside (LPS) de Haemophilus infl,tenzae, d~finies l'aide d'anticorps monoclonaux, pcuvent montrer des variations de phase d'une g6n6ration /l l'autr¢. Plnsieurs locus g6n6tiques sont impliqu6s dans la biosynth6se du LPS par H. influenzae. Nous d6crivons ici trois loci lic qui jouent un r61e dans la variation de phase du LPS. Les loci licl et lie3 on~ 6t6 enti/~rement sdquenc6s, et lic2 a 6t6 partiellement s6quenc6. Chaque locus consiste en de multiples s6quences ORF (open reading frames) et chacun contient une s6quence r6pgtitive dans la partie terminale 5' de la premi6re ORF qui peut ~tre impliqude dans la variabilit6 de phase. Les g6nes/l l'intgrieur de licl et de lic2 sont directement impliqugs dans l'expression des 6pitopes de la phase variable, mais le rfle des g6nes /l l'int6rieur de lic3 se situe /l un niveau plus complexe. Mots-clds : LPS, Haemophilus influenzae, Variation de phase, ORF; gal2, adk, Loci, Epitopes.

References Adhya, S. (1987), The g~lactose operon, in "Escherichia coil and Salmonella typhimurium: cellular and molecular biology" (F.C. Neidhart, J.L. Ingraham, K. Brooks Low, B. Magasanik, M. Schaechter & H.E. Umbarger) (pp. 1503-1512). American Society for Microbiology, Washington. Brune, M., Schumann, R. & Wittinghofer, F. (1985), Cloning and sequencing of the adenylate kinase gene (adk) of Escherichia coil. Nucl. Acids Res., 13, 7139-7151. Butler, P.D. & Moxon, E.R. (1990), The physical map of the genome of Haemophilus influenzae type b. J. gen. Microbiol., 136, 2333-2342. Cronan, J.E. Jr., Ray, T.K. & Vagelos, P.R. (1970), Selection and characterization of an E. coil mutant defective in membrane phospholipid biosynthesis. Proc. nat. Acad. Sci. (Wash.), 65, 737-744. Glaser, M., Nuity, W. & Vagelos, P.R. (1975), Role of adenylate kinase in the regulation of macromolecular biosynthesis in a putative mutant of Escherichia coil defective in membrane phospholipid biosynthesis. J. Bact., 123, 128-136. Goelz, S.E. & Cronan, J.E. Jr. (1982), Adenylate kinase of Escherichia coil: evidence for a functional interaction in phospholipid synthesis. Biochemistry, 21, 189-195. Herriott, R.M., Meyer, E.M., Vogt, M.J. & Modan, M. (1970), t~efined medium for growth of Haemophilus influenzae. J. Bact., 101, 513-516. Hitchcock, P.J., Leive, L., Makela, P.H., Rietschel, E.T., Strittmatter, W. & Morrison, D.C. (1986), Lipopolysaccharide nomenclature - - past, present and future. J. Bact., 166, 699-705. Kimura, A. & Hansen, E J. (1986), Antigenic and phenotypie variations of Haemophilus influenzae type b lipnnolysaccharide and their relationshipto virulence. Infect. Immun., 51, 69-79. Kimura, A., Patrick, C.C., Miller, E.E., Cope, L.D., McCracken, G.H. Jr. & Hansen, E.J. (1987), Haemophilus influenzae type b lipooligosaccharide: stability of


expression and association with virulence. Infect. lmmun., 55, 1979-1986. Lemaire, H.-G. & Muller-Hill, B. (1986), Nucleotide sequences of the galE gene and the gaiT gene of E. coil Nucl. Acids Res., 14, 7705-7711. Levinson, G. & Gutman, G.A. 0987), Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol. Biol. Evol., 4, 203-221. Lund, B., Lindberg, F., Marklund, B.-I. & Normark, S. (1987), The papG protein is the --D-galactopyranosyl-(I--+4)-[~-Dgalactopyrnoose-binding adhesin of uropathogenic Escherichia coil Proc. nat. Acad. Sci. (Wash.), 84, 5898-5902. Moxon, E.R., Smith, A.L., Averill, D.R. & Smith, D.H. (1974), Haemophilus influenzae meningitisin infant rats after intranasal inoculation. J. infect. Dis., 129, 154-162. Raetz, C.R.H. & Dowhan, W. (1990), Biosynthesis and function of phospholipidsin Escherichia coll. J. biol. Chem., 265, 1235-1238. Smith, A.L., Smith, D.H., Averill, D.H., Marino, J. & Moxon, E.R. (1973), Production of Haemophilus influenzae b meningitis in infant rats by intraperitoneal inoculation. Infect. Immun., 8, 278-290. Spinola, S.M., Kwaik, Y.A., Lesse, A.J., Campagnari, A.A. & Apicella, M.J. (1990), Cloning and expressionin Escherichia coil of a Haemophilus influenzae type b lipooligosaccharide synthesis gene(s) that encodes a 2-keto-3-deoxyoctulosonic acid epitope. Infect. Immun., 58, 1558-1564. Stern, A. & Meyer, T.F. (1987), Common mechanism controlling phase and antigenic variation in pathogenic neisseriae. Mol. Microbiol., 1, 5-12. Syrogiannopoulos, G.A., Hansen, E.J., Erwin, A.L., Munford, R.S., Rutledge, J., Reisch, J.S. & McCracken, G.H. Jr. (1988), Haemophdus influenzae type b lipooligosaccharide induces meningeal inflammation. J. infect. D/~., 157, 237-244. Virji, M., Weiser, J.N., Lindberg, A.A. & Moxon, E.R. (1990), Antigenic similarities in lipopolysaccharides of Haemophilus and

D.J. M A S K E L L E T A L .


Neisseria and expression of a digalactoside structure also present on human cells. Microb. Pathogen. (in press). Weiser. J.N., Lindberg, A.A., Manning, E.J., Hansen, E.J. & Moxon, E.R. (1989a), Identification of a chromosomal locus for expression of lipopolysaccharide epitopes in Haemophih~s influenzae. Infect. Immun., 57, 3045-3052. Weiser, J.N., Love, J.M. & Moxon, E.R. (1989h), The molecular mechanism of phase variation of H. influenzae lipopolysaccharide. Cell, 59, 657-665. Weiser, J.N., Maskell, D.J., Butler, P.D., Lindberg, A.A. & Moxon, E.R. (1990), Characterization of

repetitive sequences controlling phase variation of Haemophilus influenzae lipopolysaccharide. J. Bact., 172, 3304-3309. Wispelway, B., Lesse, A.J., Hansen, E.J. & Scheld, W.M. (1988), Haemophilus influenzae lipopolysaccharide-induced blood brain barrier permeability during experimental meningitis in the rat. J. Clin. invest., 82, 1339-1346. Zamze, S.E., Ferguson, M.A.J., Moxon, E.R., Dwek, R.A. & Rademacher, T.W. (1987), Identification of phosphorylated 3-deoxy-manno-octulosonic acid as a component of Haemophilus influenzae lipopolysaccharide. Biochem. J., 245, 583-587. Zamze, S.E. & Moxon, E.R. (1987),

Composition of the lipopolysaccharide from different capsular serotype strains of Haemophilus

influenzae. J. gen. Microbiol., 133, 1443-1451. Zwahlen, A., Rubin, L.G., ConneIly, C.J., Inzana, T.J. & Moxon, E.R. (1985), Alteration of the cell wall of Haemophilus influenzae type b by transformation with cloned DNA: association with attenuated virulence. J. infect. Dis., 152, 485-492. Zwahlen, A., Rubin, L.G. & Moxon, E.R. (1986), Contribution of lipopolysaccharide to pathogenicity of Haemophilus influenzae: c o m p a r a t i v e virulence of genetically-relatedstrains in rats. Microb. Pathogen., I, 465-473.

Res. Microbiol. 1991, 142, 725-730


Antigenic variation in Trypanosoma equiperdum C. Roth (1) (*), C. J a c q u e m o t (l), C. G i r o u d (2), F. Bringaud (2), H. Eisen (3) and T. Baltz (2)

(/) Unitd d'Immunoparasitologie, URA 361, CNRS, Institut Pasteur, 75724 Paris Cedex 15, ¢2) Laboratoire d'Immunologie et Biologie parasitaire, Universitie de Bordeaux II, 146 rue Leo Saignat, 33076 Bordeaux Cedex (France), and (3) Division o f Basic Sciences, Fred Hutchinson Cancer Research Center, 1124 Columbia Street, Sealtle, ~,~A 92011 (USA)


Trypanosoma equiperdum is an African trypanosome that causes dourine in horses. Like the other African trypanosomes, T. equiperdum escapes elimination by the immune system of its host by using an elaborate system of antigenic variant. The trypanosomes are covered by a coat consisting of a single protein called the variable surface glyceprotein (VSG) that acts as the major trypanosome immunogen. As the host responds to one VSG, trypenosomes covered with another VSG become dominant. There is a loose order of appearance of these VSG during the infection. The factors that affect the timing of VSG expression and the effective size of the VSG repertoire in T. equiperdum are reviewed. The VSG genes are generally activated by a process of duplicative transposition involving the duplication of a silent VSG gene and inserting a copy of the gene into an expressien site. The order of VSG expression is related to the amount of homology between the silent gene and the expression site. The genes expressed late in infection lack extensive homology with the expression site and depend on homology with the gene in the expression site. The genes coding for VSG expressed late in infection are hybrid genes because of this mode of transfer. This transfer mechanism allows the trypanosome to create complex VSG genes from parts of several different silent genes that are each pseudogenes. Additionally, data are presented showing that only a limited portion of the VSG is actually seen by the host immune system. These factors indicate that the effective VSG repertoire is greater than the number of VSG genes in the trypanosome genome.

Key-words: Dourine, Trypanosome equiperdum, Antigenicity; Variation, Epitopes, VSG, Gene expression.


Trypanosoma equiperdum is a member of the African trypanosomes which are best known as the causative agent of "sleeping sickness" in man. T. equiperdum

causes a related disease in horses but differs from the T. brucei group in the fact that it is transmitted venereally rather than by the tse-tse fly, As such, T. equiperdum remains in the trypomastigote or "blood stage" form and does not

(*) To whomcorrespondenceshould be addressed.

undergo the promastigote portion of the life cycle. Like the other African trypanosomes, T. equiperdum lives flee in the blood stream of the infected host and evades destruction by the host immune system through the use of a corn-

726 plex system of antigenic variation. The trypanosomes are covered by a dense coat consisting of a single glycoprotein that acts as the major trypanosome immunogen. As the host reacts immunologically to the surface proteins, it provides the selective pressure for the expansion of a subpopulation that has changed its variable s u r f a c e glycoprotein (VSG). The antibody response is not causing the variation; rather, it is selecting the small number of pre-existing variants in the population. Capbern et al. (1977) showed that more than 100 immunologically dist;nct VSG are expressed during an experimental T. equiperdum infection of a rabbit. Furthermore, these authors demonstrated that there is a loose order of VSG expression. Some VSG are preferentially expressed early in infection, while other VSG are only observed later in infection. This observation indicates that the trypanosomes can "reset" their VSG clock whenever they enter an immunologically naive host but in a manner that does not lock them into a rigid pattern of VSG expression that could be easily blocked. We have been particularly interested in the molecular mechanisms involved in the timing of VSG gene expression - - why are some VSG expressed early in infection and others only later ? This interest has involved related issues such as defining the region of the VSG seen by the host immune system and defining the "effective" size of the VSG repertoire. If only a small portion of the VSG is exposed, then the parasite might be able change only small portions of the gene to alter its immunologic identity. The following discussion is a short review of what is known about the timing of VSG gene expression in T. equiperdum, the extent of the VSG repertoire and the region of the VSG exposed on the surface of the trypanosome. The authors appologize in advance to workers in this area whose work is not discussed due to our emphasis and the limited size of the review.

C. R O T H E T AL.

Timing of VSG gene expression During the trypomastigote or blood stage of the trypanosome, the VSG genes can be activated by several different mechanisms. Active VSG genes are always located in sites referred to as expression sites. The active site is always adjacent to a chromosome telomere and preceded by a region called the barren region because of its relative sparsity of sites recognized by restriction endonucleases and containing a tandem series of 76-bp repeats (Campbell et al., 1984; Florent er al., 1987). The barren region is preceded by a number of genes coding, for the most part, for products of unknown function. These genes are referred to as ESAG for expression-site-associated genes. Several reports indicate t h a t in T. brucei, t h e promoter used for the transcription of the VSG genes lies upstream of these ESAG and about 50 kb from the VSG gene (Pays et al., 1989; Zomerdijk et al., 1990). The promoter has not been identified for a T. equiperdum VSG gene. Though only one expression site is generally active at a time, there are several potential expression sites (Longacre et al., 1983; Buck et aL, 1984a,b) and, at least in vitro, two expression sites can be activated at the same time (Baltz et al., 1986). A VSG gene may be activated by activating the expression site occupied by the VSG and inactivating the previously used site. In the mammalian host the preferred mechanism is to replace the VSG gene in the active expression site with a copy of one of the many silent VSG genes which are located either near telomeres or internally on one of the chromosomes. This latter process is referred to as duplicative transposition. A schematic representation of the duplicative t r a n s p o s i t i o n mechanism is shown in figure 1. Whether the VSG gene is activated early or late during infection,

the silent "basic copy" (BC) of the gene remains in its normal chromosomal position and a new copy of the gene is observed in the expression site. Though the original paper by Capbern et aL (1977) classified the VSG into three groups, i.e. early, middle and late, we have chosen, for reasons of simplicity, to consider the differences between those expressed early and late in infection. One difference between "early" and "late" genes is that the entire coding region of the silent gene is generally transferred to the expression site early in infection. In the case of the T. equiperdum VSGI gene, the EcoRI-restriction sites located outside the coding region of the gene are always transferred to the expression site when the gene is activated (Longacre et al., 1983). Though most VSG genes do not have such conveniently positioned restriction enzyme recognition sites, Longacre and Eisen 0987) were able to show by an RNA protection technique that the coding region of the genes for several different early VSG were the same, or nearly the same size in the silent copy and in the expression-linked copy. Despite the difference in terminology, these results are equivalent to those of Bernards et al. (1981) in T. brucei showing that the site of recombination for an "early" VSG gene was in the 3' non-translated region of the expression-linked copy of the gene. In contrast to the result with the early genes, Longacre and Eisen (1987) found that the expressionlinked copies of several late VSG genes were unique to the expression site. This result suggested that the expression linked copy of these late genes was a hybrid gene.

Structure of late VSG genes The observations of Longacre and Eisen were extended by Roth et al. (1986, 1989) and by Thon et al. (1989). Studying the nucleotide sequence of the BC genes related to VSG78 and the expression-


Silent 78


ESAG "C" __:.-_-_-_-.-=





Barren Region "76 bp Repeats"


Origin of Transcription

ESAG "D. . . .

Barren Region Ec~,,i 76 bp Repeats"jllnlll

Active 1


Barren Region "76 bp Repeats"

Active 78

Fig. 1. Schematic representation of VSG gene replacement by duplicative transposition early and late in infection. The top line represents silent BC genes for VSGI and 78 where the silent gene coding for VSGI is located near a telomere (the open circle) and preceded by 76-bp repeats (the broken line). The middle line represents a typical expression site in which the VSG "X " is being expressed. The expressed VSG gene is preceded by a barren region containing 76-bp repeats and expression site associated genes (ES.~.G). A single origin of transcription is assumed to be upstream of the ESAG genes. If the VSG1 gene invades the expression site (early), the copy of the VSGI gene replaces gene X by recombinations in the 76-bp repeat region and the near the telomere. If the VSG78 gene invades the site (late), the copy of the VSG78 gene depends on homology between the gene X and itself to make the necessary recombinations, thus creating a hybrid VSG gene.

linked copy of the gene, Roth et el. (1986) showed that the active gene copy was a hybrid gene. The 3'-most 250-bp VSG78 gene were donated by one BC gene, the 3' donor, while the rest of the gene was derived from a gene family u~related to the 3' donor gene except in the region of recombination. Surprisingly, not only was the VSG78 gene a hybrid gene constructed from unrelated genes, but even the 5' portion of the active gene contained the sequences from different members of the silent gene family (Roth et el., 1986, 1989). Additionally, it was shown that all of the genes contributing to the active VSG78 gene are pseudogenes and none of them singly

cou'd code for a complete VSG (Roth et al., 1989). When the copy o f the VSG78 gene expressed in different infections was examined, it was always hybrid, but the mosaic 5' region was different in each case (fig. 2). Then et el. (1989) examined the structure of the gene coding for another late antigen, VSG20, and found essentially the same results as was observed for VSG78. The 5' end of the gene was a mosaic of sequences from two re!ated silent genes, while the 3' end was derived from an unrelated gene. Again, all the genes donating sequences to the active gene are pseudogenes. The observation that the genes used to form these active late VSG genes

are pseudogenes explains why they are expressed as hybrid genes. The astounding fact is that they are formed and expressed in nearly every infection (Capbern et eL, 1977). There must be both a high rate of recombination or gene conversion and a strong selective pressure in the host to explain these observations. The results suggest that in the case of VSG expressed early in infection the silent gene related to the active gene is located near a telomere and is surrounded by sequences similar to those in the expression site. There are barren region sequences 5' of the silent gene and telomcre related sequences 3' of the silent gene. Thus,

C. R O T H


0 I






0.5 I





1.0 I









1.5 kb I


BC-B Bc-c

I ~ ~ ~~ ~ ~ ~










7820 78 b's ~


78 'er


ensures that an entire functional gene will be inserted and expressed. One can imagine that this property is especially important early in infection when the parasite load is relatively low and the host's immune system is unimpaired. Later in infection, the host is less healthy and its immune system has been partially impaired due to the infection. At this later time, the extent of the repertoire is more important than the efficiency. The trypanosome can now afford the luxury of trying different gene combinations to find a VSG that functions well and possibly to create new VSG genes. This line of reasoning suggests that the VSG repertoire may not be limited by the number of silent VSG genes but rather by the trypanosome's ability to combine portions of different silent genes, a process that could greatly expand the effective repertoire.

~ Exposed

VSG epitopes

Fig. 2. Schematic model of the VSG78 gene family. The silent genes that are always present in the genome are labelled BC-A, -B and -C. The vertical lines within these blocks represent stop codons in the reading frame used for the active VSG78 genes. The expressed VSG genes are labelled 78, 782°, 78his and 78ter. The different colors represent the probable silent gene that donated that region of the gene.

Table 1. Summary of the characteristics of variant VSG78 genes from isolates resistant to mAb killing. I







+ +

+ -

+ +






Ser192 ~ Arg Gene conversion at 5' end and Thr220 -. Ala Gln172 -- Glu

H3R, H7R and H21R representthe strains resistantto mAb H3, H7 or H21, respectively.ColumnslabelledI-IVindicatethe abilityof mAb in that group to kill the different resistantstrains.The last columnindicatesthe amino acid changethought to be responsible for the resistancephenotype.

if a region slightly larger than the VSG gene is transposed to the expression sight, there will be small

regions of homology that allow the incoming gene to recombine with the expression site. This procedure

The ability of the trypanosomes to form mosaic genes suggested that they might be able to change individual surface epitopes by replacing parts o f the active VSG gene with related segments of the silent genes. To test this hypothesis, monoclonal antibodies (mAb) that kill trypanosomes expressing VSG78 were isolated. Since the trypanosomes are killed by the mAb, the recognized epitopes must be exposed on the surface of the trypanosome. Some of the mAb were used to select trypanosome variants that were not killed by a particular mAb. Most of these variants had simply replaced the VSG78 gene with the gene for an early VSG. However, some variants that had altered the active VSG78 gene were isolated and cloned. The VSG78 variants remained sensitive to most of the mAb not used ir~ the selection. This property was used to group the different mAb (table I). Two independent variants resistant to a group one mAb were examined by

Antigenic variation in Trypanosoma equiperdum.

Trypanosoma equiperdum is an African trypanosome that causes dourine in horses. Like the other African trypanosomes, T. equiperdum escapes elimination...
363KB Sizes 0 Downloads 0 Views