Gutell 063.jmb.1997.269.0203

212 views

Published on

Published in: Technology, Health & Medicine
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
212
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Gutell 063.jmb.1997.269.0203

  1. 1. Regulation and Trafficking of Three Distinct 18 SRibosomal RNAs During Development of theMalaria ParasiteJun Li1, Robin R. Gutell2,3, Simon H. Damberger2, Robert A. Wirtz4Jessica C. Kissinger1, M. John Rogers1, Jetsumon Sattabongkot5and Thomas F. McCutchan1*1Growth and DevelopmentSection, Laboratory of ParasiticDiseases, National Institute ofAllergy and Infectious DiseasesNational Institutes of HealthBethesda, MD 20892-0425USA2Department of MCD BiologyCampus Box 347, University ofColorado, Boulder, CO 80309-0347, USA3Department of Chemistry andBiochemistry, Campus Box 215University of Colorado, BoulderCO 80309-0215, USA4Entomology Branch, Divisionof Parasitic Diseases-NCIDCenters for Disease Control andPrevention, 4770 BufordHighway NE, AtlantaGA 30341-3724, USA5Entomology Department, USArmy Medical ComponentBangkok, ThailandThe human malaria parasite Plasmodium vivax has been shown to regulatethe transcription of two distinct 18 RNAs during development. Here weshow a third and distinctive type of ribosome that is present shortly afterzygote formation, a transcriptional pattern of ribosome types that relatesclosely to the developmental state of the parasite and a phenomenon thatseparates ribosomal types at a critical phase of maturation. The A-typeribosome is predominantly found in infected erythrocytes of the ver-tebrate and the mosquito blood meal. Transcripts from the A gene arereplaced by transcripts from another locus, the O gene, shortly after ferti-lization and increase in number as the parasite develops on the mosquitomidgut. Transcripts from another locus, the S gene, begins as the oocystform of the parasite matures. RNA transcripts from the S gene are prefer-entially included in sporozoites that bud off from the oocyst and migrateto the salivary gland while the O gene transcripts are left within theoocyst. Although all three genes are typically eukaryotic in structure,the O gene transcript, described here, varies from the other two in coreregions of the rRNA that are involved in mRNA decoding and transla-tional termination. We now can correlate developmental progression ofthe parasite with changes in regions of rRNA sequence that are broadlyconserved, where sequence alterations have been related to function inother systems and whose effects can be studied outside of Plasmodium.This should allow assessment of the role of translational control in para-site development.# 1997 Academic Press LimitedKeywords: small subunit (SSU) rRNA; ribosome; development; malaria;translational control*Corresponding authorIntroductionThe Apicomplexa comprises a group of parasiticprotozoa with increasing medical importance,which includes Cryptosporidium, Toxoplasma, Thei-leria, Babesia and Plasmodium. The control of devel-opment of these parasitic protozoa is unusual andappears to involve a broad control mechanismlying at the center of translational machinery, ribo-somal RNA. At least in the malaria parasite, Plas-modium species, structurally distinct ribosomes areactive during different stages of development anddifferentiation (McCutchan et al., 1995). In mosteukaryotic organisms there is a correlation betweenproliferative states of development and enhancedrRNA transcription and ribosome production(Sollner-Webb & Tower, 1986). The increase in ri-bosome production is occasionally accompanied bymodi®cation to the ribosome (Etter et al., 1994).Plasmodium appears to respond to developmentalstimuli in a similar fashion except that points of de-velopmental commitment and cell proliferation areaccompanied not only by increased ribosome pro-duction but also by changes in the probable cataly-tic moiety of the ribosome, the rRNA. MalariaAbbreviations used: ITS1, 3H-internal transcribedspacer.J. Mol. Biol. (1997) 269, 203±2130022±2836/97/220203±11 $25.00/0/mb971038 # 1997 Academic Press Limited
  2. 2. parasites maintain a life cycle which includes threedifferent asexual multiplicative stages and one sex-ual reproductive stage, all with different rates ofreplication in the vertebrate and mosquito host(Figure 1). Asexual reproduction takes place in theliver (exoerythrocytic schizogony) and the blood ofthe vertebrate host (erythrocytic schizogony), andalso on the mosquito midgut (sporogony). Sexualreproduction initiates in the vertebrate blood(gametogenesis) and ®nishes by fertilization of thefemale macrogamete with the ex¯agellated malemicrogamete in the mosquito midgut after engor-gement of a blood meal. In previous studies, twodistinct 18 S rRNA (small subunit rRNA) geneswere identi®ed from Plasmodium vivax (Li et al.,1994b). Transcripts from the type A gene are domi-nant in erythrocytic stages and gametocytes, whilethe type S gene seems to be transcribed in the spor-ozoite (Figure 1). However, neither of these rRNAtypes could be detected in the infected mosquitoduring the early stage of P. vivax development (Liet al., 1994b). The gap between termination of Agene transcripts and expression of the S geneprompted a more exhaustive investigation for acorrelation between development and the ex-pression of rRNAs. Here we describe a third typeof 18 S rRNA (type O) which is associated with oo-cyst development in infected mosquitoes (Figure 1)Figure 1. Schematic representation of malaria life cycle and stage-speci®c ribosomal RNAs. Human infection by Plas-modium begins when an infected female anopheline mosquito inoculates sporozoites into the bloodstream duringfeeding. The sporozoites invade liver cells and transform into trophozoites. In six to eight days one mature schizontwill release thousands of liver-stage merozoites into the bloodstream (exoerythrocytic schizogony). The secondasexual proliferative stage (erythrocytic schizogony) is initiated when the liver-stage merozoites invade the erythro-cytes giving rise to blood stage trophozoites. About 14 to 16 erythrocytic merozoites are generated in a 48-hour cyclefor re-infection. The merozoites may alternately differentiate into single gametocytes, the initial stage of the sexualreproduction (gametogenesis). Mosquito infection begins when the gametocytes are drawn in the blood meal, and themale microgametocyte ex¯agellates into individual microgametes and fertilizes the female macrogamete. The result-ing zygote transforms into a motile ookinete, which penetrates the mosquito midgut and rounds up as an oocyst onthe external surface. After a period of 9 to 14 days, thousands of sporozoites are differentiated in the mature oocyst(sporogony), the only multiplicative stage in the mosquito.204 Ribosomal RNAs During Development of the Malaria Parasite
  3. 3. and show the existence of a mechanism for sortingribosomes in the oocyst which selects the type S ri-bosome for inclusion into the maturing sporozoite,with the concomitant exclusion of the type O ribo-some.Exploring the multi-rRNA system in Plasmodiumspecies may both elucidate a basic mechanism ofdevelopmental control and reveal targets for meta-bolic intervention of parasite growth. For example,it has been shown that the ribosomal GTPase cen-ter, a site of antibiotic interaction, produced in theasexual blood stage is different from that found inthe sporozoite stage (Rogers et al.,1996). One mustnow also entertain the idea that there is a differ-ence in af®nity of the type O ribosome for aspeci®c subset of mRNAs. Finally the addition ofthe third type of rRNA enhances the possibilitythat the direction of events leading to the evolutionof this multi-gene family, perhaps even the para-site, can be determined.ResultsThree rRNA genes are associated withP. vivax infectionWe have monitored Plasmodium ribosomal RNAsequences from the blood of malaria patients as anindicator of the species of infecting parasite (Liet al., 1997). Investigation of patients in Thailand,who were initially diagnosed by microscopy ashaving P. vivax infection, revealed three distincttypes of P. vivax 18 S rRNA genes associated withthe infection. Partial 18 S rRNA genes were ampli-®ed from DNA isolated from infected blood usingPCR primers that are conserved in the genus Plas-modium (Li et al., 1995). Sequence analysis of theproducts revealed three genes. Two of these geneshave been reported as the A and S genes of P.vivax (Li et al., 1994a). The third type of gene is re-ferred to as type O for reasons that will be de-scribed below. Sequence information from afragment of this gene allowed the design of Ogene-speci®c oligonucleotides (Figure 2) for furtherinvestigation, as the PCR from patients blood wasdesigned with primers to conserved regions whichpresumably amplify all the P. vivax 18 S rRNAgenes.To con®rm association of the O gene with P.vivax, we collected P. vivax-infected blood frompatients in different locations of central Thailand.Ampli®cation by RT/PCR of RNA from patientsyielded products that hybridized to the probesspeci®c to the type A gene of P. vivax but not tothose designed to detect the other human malariaparasites, including P. falciparum, P. malariae and P.ovale (Li et al., 1995). Volunteer patients also al-lowed laboratory-reared mosquitoes, Anophelesdirus, to feed on their arms before they were trea-ted for malaria. The species of parasite maturing inthe mosquitoes was also determined using speciesspeci®c monoclonal antibodies directed to the P.vivax circumsporozoite protein, which is the majorsurface protein expressed in sporozoites (Wirtzet al.,1991). The four samples collected frompatients did not have mixed infections as deter-mined to the limits of our detection techniques(data not shown). To further substantiate the as-sociation of the O gene with P. vivax, the complete18 S rRNA O gene was isolated from a laboratory-maintained strain of P. vivax, Sal-1, originally iso-lated in El Salvador. The O gene was produced intwo overlapping fragments (1.6 kb and 1.9 kb) am-pli®ed with speci®c primer pairs, 705/683 and706/573 (Figure 2). Hybridization shows that bothfragments are recognized by an O gene speci®coligo-probe (741) targeted to the central overlap-ping regions (Figure 2). The cloned fragments con-tain the entire 18 S rRNA genes and the ITS1region.The type O transcript appears soon afterfertilization in developing ookinetes andoocysts but is segregated frombudding sporozoitesThe temporal and developmental pattern ofrRNA expression was determined for the period ofparasite development within the mosquito(Figure 3a). For this purpose, mosquitoes were col-lected at different time intervals after the infectiousblood meal and total RNA was isolated from theseparated thorax and abdomen. The origin of theparasite RNA could be determined, since maturesporozoites migrate from developed oocysts on themidgut wall to the salivary glands of the thorax(Figure 3a). RNA originating from mature sporo-zoites in salivary gland was distinguished fromthat found in the oocyst by severing the mosquitoat the junction separating the abdomen and thethorax. Only RNA from mature sporozoites isfound in the thorax sample while the abdomensample contains RNA from both parasites in theblood meal and those proceeding through develop-ment to the mature oocyst stage. The RNA fromthe thorax and abdomen was then analyzed for ex-pression of the distinct genes during the timecourse of development in the mosquito (Figure 3b).The results indicate that the A-type transcript wasthe dominant rRNA detected in infected blood andin the gut of engorged mosquitoes. The bloodstage rRNA decreased rapidly and was detectedonly in low levels in the mosquito gut at 24 hoursafter feeding (Figure 3b). The novel type of rRNA,which we present here, was identi®ed after the dis-appearance of the A gene transcript and remainedthroughout oocyst development (Figure 3b). Byconvention with nomenclature, the third type ofgene is referred to as the O gene as it re¯ects de-velopment starting in the ookinete stage and conti-nuing through the oocyst stage (Figure 1). Type SrRNA was not detected until day 6 (Figure 3b),when the oocyst is close to mature and sporozoitesare differentiated (Li et al., 1994b). The increasingRibosomal RNAs During Development of the Malaria Parasite 205
  4. 4. signal after this point may relate to an increase inthe mass and number of parasites as they devel-op into mature sporozoites. Clearly, the type Oand S genes are transcribed during the laterperiod of oocyst development but they are distin-guished by the pattern and location of ex-pression. The type O gene is transcribed earlierand limited to the abdomen of the infected mos-quito (Figure 3b), and is not transferred to thethorax portion of the mosquito. The type S rRNAFigure 2. Sequence alignment of three 18 S rRNA genes from P. vivax Sal-1 strain. The coding region for the maturerRNAs starts from the 5Hend and is followed at the 3Hend by the internal transcribed spacer ITS1 (open box) and the5Hend of the 5.8 S rRNA gene (shaded box). The upper line shows the O gene sequence; the middle and lower linesrepresent the S and A gene, respectively, and are shown only where the sequences differ from the O gene. Dashesrepresent gaps where the sequences cannot be directly aligned. Oligonucleotides complementary to the sequencesused for cloning and analysis of the gene expression are indicated by shaded boxes, with numbers and direction. TheGenBank accession numbers for the P. vivax 18 S rRNA gene sequences are: type A U07367, type O U93095 and typeS (previously C) U07368.206 Ribosomal RNAs During Development of the Malaria Parasite
  5. 5. peaks around day 10 to day 12 when rapidnuclear division and differentiation of sporozoitesin oocysts occur. On day 14 S-type rRNA wasdetected in the thorax, suggesting that it is thedominant type of rRNA in sporozoites that mi-grate from the maturing oocyst on the midgut inthe abdomen to the salivary glands in the thorax.This association was also supported by the factFigure 3. Developmentally regulated transcription of the rRNA genes in P. vivax. a, Schematic view of parasite devel-opment in mosquitoes following an infectious blood meal (in days, where D 0 ˆ day 0, etc.). The progressive courseof differentiation and growth stages is drawn appropriate to days following the blood meal. R, ring form; T, tropho-zoite; Sc, schizont; M, merozoite; G, gametocytes (male and female); Z, zygote; K, ookinete; O, oocyst and S, sporo-zoites. Stages are colored for ease of identi®cation, and a line indicates separation of thorax and abdomen RNAsamples. b, Autoradiograph of a Northern blot analysis of total RNA prepared from infected mosquitoes at differentdays after an infectious blood meal. The RNA, from either thorax or abdomen, was ®rst separated by electrophoresison an agarose gel and immobilized on a nylon membrane. Then, triplicates of the blots were individually probedwith 32P-labeled oligonucleotides 741, 742 and 743, which are speci®c for type A, type O and type S rRNAs, respect-ively. Each slot represents the average signal of RNA from one tenth of a ten-mosquito pool collected at each timepoint. The sample from day 0 (D0) was collected two hours after the mosquito fed on the infected patient. Lanes Band M represent RNA from P. vivax infected patient blood and uninfected mosquito, respectively. The middle panelwas autoradiographed for 12 hours while the top and lower panels were autoradiographed for three hours.Ribosomal RNAs During Development of the Malaria Parasite 207
  6. 6. that the S-type rRNA is the only type detected insporozoites puri®ed from the salivary glands(data not shown).The secondary structure of the type O rRNAdiffers from the other 18 S rRNAs in regionsassociated with translational functionSequence alignment of the three P. vivax rRNAgenes (Figure 2) indicates that the O gene is differ-ent from the other two genes, although they sharesigni®cant homology. Comparison of the ITS1 se-quences shows little similarity among the threegenes (Figure 2), further indicating three distincttranscription units. The secondary structures forthe three 18 S rRNAs were derived by the com-parative method. By searching for positional co-variances secondary structure models for the typeO, A and S rRNAs were determined, based on theexisting Eukarya 18 S rRNA and Eubacterial 16 SrRNA models (R.R.G., data not shown; availablefrom http://pundit.colorado.edu:8080/root.html).Analysis of the O gene sequence shows that it is,overall, very similar in sequence and structure tothe A and S genes (Figure 4). The structural analy-sis shows that it forms all of the expected domainsfound in the Eukarya 18 S rRNA consensus,although, as discussed below, there are anomalousinsertions and deletions. The secondary structureof the O gene has many compensatory base-pairchanges from the A and S genes and the overallconservation with the Eukarya, together with itsstage speci®c expression (Figure 3), stronglysuggests an active and functional role. In sharpcontrast, there are about 36 positions where the Ogene is different from the Apicomplexa and Eukar-ya consensus. There are an additional 18 positionswhere the O gene differs from the Apicomplexaconsensus. Overall, about 12 positions are inser-tions in the O gene, with three occurring in highlyvariable regions and probably not signi®cant (de-noted in green, Figure 4). Nine insertions are pro-minent as they occur at positions well conserved inthe Eukarya, and in many cases these are con-served in all rRNAs. The size of the insertionranges from a single nucleotide to more than 20nucleotides in the O gene (Figure 4). There are alsoabout nine signi®cant deletions in the O gene, rela-tive to the Eukarya consensus, which are at severalhighly conserved regions of the 16 S-like rRNA(Figure 4). It is noteworthy that the sequence of theO gene obtained from ampli®cation of genomicDNA was identical to that derived from rRNAtranscripts (data not shown). Northern blots werealso probed with oligonucleotides complementaryto the unique insertions in the O gene to con®rmexpression of these regions in the mosquito stagesand, hence, these sequences are present in the ma-ture O gene product (Figure 3b). The very long``inserted variable region (approximately 1135 inEscherichia coli numbering) is unusual in the Ogene. It is longer than the equivalent region in theA and S genes and in the O gene forms a verylong, extended helix with few internal loops andmismatches. In contrast, the equivalent helix con-tains several internal loops in the A and S genes,making their helices more irregular (R.R.G., datanot shown).Of the anomalous differences in the O gene, themost signi®cant are discussed here. There are tworegions in the O-type rRNA that vary at positionsthat have been shown in other studies to be in-volved in the decoding step in protein synthesis(Figure 4); these are unique variations in the Ogene as the A and S genes resemble all other Eu-karya. One of these regions is near the 3H-end,shown in red (Figure 4; 1400 region in E. coli num-bering). This area is usually termed the decodingsite as it is in close proximity to tRNA at the A, Pand E sites in the ribosome and also associateswith the mRNA (Zimmerman, 1996). The Sal-1 Ogene has both a 22 nt insertion and a deletion of acore conserved base-pair (Figure 4) (correspondingto the conserved 1399Á1504 base-pair in E. coli)which disrupts the O gene transcripts secondarystructure in this region. Another region altered inthe O-type rRNA corresponds to helix 34(nucleotides 1046 to 1067 and 1189 to 1211 in E. colinumbering). This helix is also in intimate contactduring protein synthesis as it is in close proximitywith mRNA and tRNA (Zimmerman, 1996). A del-etion at position 1054 was originally isolated as asuppresser of UGA termination codons (Murgolaet al., 1988), although subsequent analysis hasshown a wider effect on translational accuracy inE. coli and Saccharomyces cerevisiae (Chernoff et al.,1996; Goringer et al., 1991; Moine & Dahlberg,1994). The type O SSU rRNA in this region hasthree insertions in the 3H-half of helix 34, one ofwhich corresponds to an insertion at position 1200(Figure 4). Mutation at this position in E. coli hassigni®cant effects on translational accuracy, caus-ing read-through and frameshifting (Moine &Dahlberg, 1994). The insertions in the 3H-side ofhelix 34 in the type O rRNA also correspond to asite of rRNA cleavage of the P. falciparum type ArRNA that degrades the transcript during early de-velopment in the mosquito (Waters et al., 1989),and the possible signi®cance of this is discussedbelow. Another insertion occurs at positions corre-sponding to the 912 region in E. coli. Mutation atthis position also affects translational accuracy inE. coli and S. cerevisiae (Liebman et al.,1995;Lodmell et al.,1995). Hence it is likely that the Ogene transcripts have an altered speci®city in trans-lation or decay of certain mRNA species. ThesemRNAs are likely to be critical for development ofthe parasite. The O gene has other ``non-intronand anomalous insertions that occur in highly con-served regions of this rRNA have, to date, onlybeen documented in this O gene sequence (shownin red; Figure 4). All of these insertions should beconsidered signi®cant and most unusual. There area few insertions that occur in less conserved re-gions of this rRNA (green boxes; Figure 4). These208 Ribosomal RNAs During Development of the Malaria Parasite
  7. 7. occur in the O gene relative to the A and S genes.In summary, there are nine signi®cant (red) inser-tions and three less important (green) insertions. Inaddition, there are several nucleotides deleted in theO gene, relative to the Eukarya consensus. Thesepositions are denoted with a large red dot (Figure 4).Figure 4. Comparison of the secondary structures of P. vivax 18 S rRNAs: type O compared with type A and S. Thehelices are represented according to Watson-Crick, wobble (G*U) and unusual (G*A) base-pairing. Regions of vari-able sequence are left mostly as nucleotides in line format. Insertions and missing nucleotides relative to the A and Sgenes are color coded, as discussed in the text, and insertions in variable and conserved regions in the O gene arecolor shaded. Compensatory base-pairs and differences in the O gene sequences are indicated by the symbols (!*;see the legend on the Figure). Secondary structures for the P. vivax rRNAs are available from the World Wide Website for RNA secondary structures (http://pundit.colorado.edu:8080)Ribosomal RNAs During Development of the Malaria Parasite 209
  8. 8. These all occur in conserved regions of the 18 SrRNA and should be considered signi®cant.Phylogenetic analysis suggests an order ofdivergence of the three rRNA genesPlasmodium are members of the Apicomplexa,with Dino¯agellates as the likely sister group(Allsopp et al., 1994; Escalante & Ayala, 1995). Therelationship of the three P. vivax rRNA genes toeach other and other relevant Apicomplexa was es-timated by maximum parsimony analysis(Swofford, 1993). The sequences used for analysisare conserved regions that can be unambiguouslyaligned among all sequences compared (R.R. Gu-tell, unpublished). The results (Figure 5) show thatPlasmodium species are well separated from theother Apicomplexans but that the relationships ofthe different Apicomplexa lineages to each otherare less certain, with no bootstrap value over 60%.The rRNA genes of other members of the Apicom-plexa are more distantly related (Sarcocystis, Toxo-plasma and Cryptosporidium). As expected,Dino¯agellata form the sister group (Prorocentrumand Symbiodinium), and Ciliophora (Parameciumand Oxytrichia) are outgroups in the phylogram(Figure 5). Among the P. vivax rRNA genes therehave been two independent gene duplicationevents; the ®rst leading to the O gene and the ante-cedent of the A/S lineage. The second duplicationevent occurred later and lead to the A and S genes.The duplication and divergence leading to the Aand S genes is not a recent event because when theA-type genes from the monkey malarias (P. fragile,P. knowlesi, P. simium) and the human parasite P.malariae are added to the data set, they form amonophyletic group. This demonstrates that the Aand S genes diverged before speciation. The Ogene has been detected by expression in at leasttwo other Plasmodium species (J. L. et al., unpub-lished), and is therefore likely to be a common orperhaps universal feature of the genus.DiscussionPlasmodium species exhibit remarkable controlover ribosome production during parasite develop-ment. The rRNA genes are few in number, be-tween four and eight rDNA units (McCutchan et al.,1995), and are genetically unlinked, being dis-persed throughout the genome (Wellems et al.,1987). The genes appear to accumulate mutationsindependently of each other and hence sequencedifferences between the units do not occur at thesame rate as in other organisms (Rogers et al.,1995). The presence of structurally distinct rRNAgenes and the physical separation of the transcrip-tion units on different chromosomes may be in-volved in selectively accessing and expressingstage-speci®c rRNA during different periods of theparasite development. The controlled switching ofrRNA gene transcription in Plasmodium wassuggested in earlier studies (Gunderson et al.,1987;McCutchan, 1986; McCutchan et al.,1988; Waterset al., 1989). The exact timing of the transcriptionalswitches varies somewhat with the species of para-site, and even the temperature at which the para-site is being maintained, but the sequence of eventsis thought to be uniform. The type A rRNA is thedominant transcript in erythrocytic stages includ-Figure 5. Phylogram of 18 rRNA genes. Horizontal branch lengths between nodes correspond to the number ofshared derived changes. Bootstrap percentages are indicated above each branch. Analysis parameters are de®ned inMaterials and Methods. For ease of interpretation, the Phyla are differentiated by the intensity of backgroundshading.210 Ribosomal RNAs During Development of the Malaria Parasite
  9. 9. ing gametocytes, the initial form of sexual stages.Here we show that after zygote formation in themosquito midgut, the type A rRNA is replaced bythe type O rRNA. Transcription of the O gene con-tinues in the oocyst through the entire develop-ment of the parasite in the mosquito. The S typerRNA is initiated about a week after the infectiousblood meal, when differentiation of sporozoites be-gins in maturing oocysts. The rapid increase of theS-type rRNA corresponds with the period fordifferentiation of sporozoites, which can migratefrom the mature oocyst to the salivary gland,ready for infection of humans. The S type rRNA isreplaced by the type A rRNA when merozoitesdifferentiate in maturing liver stage schizonts (Liet al, 1991). The merozoites released from matureschizonts will initiate the erythrocytic stage of theparasites in which the A gene dominates. Thus, theswitch of rRNA types correlates more with the pro-gression of distinct stages of development thanwith the presence of the parasite in one host or theother (summarized in Figure 1), suggesting thatregulated transcription of stage-speci®c rRNAcould be an integral part of the developmental con-trol of the Plasmodium species.The possible functional signi®cance of maintain-ing alternative forms of rRNA is indicated by thedevelopmental regulation of their expression andby structural differences in core regions known tobe associated with biological function in other or-ganisms (Figure 4). The three distinct rRNA genesidenti®ed from P. vivax each correspond to a dis-crete proliferative stage during parasite develop-ment. The O gene is signi®cantly different in thecore regions which are associated with the accu-racy center in the ribosome (Liebman et al., 1995).Changes in sites universally identi®ed with mRNAassociation and decoding suggest that differencesin ribosome af®nity for subsets of mRNA or sup-pression of translation termination are involved inthe developmental progress. There also appear tobe active mechanisms involved in the turnoverfrom one type of ribosome to another. The ®rst in-dication that a distinction was being made betweenribosomal types co-existing within the developingparasites in the mosquito came from a study show-ing a preferential breakdown of type A transcriptsin the zygote (Waters et al., 1989). This was initiallydif®cult to understand, since the point of cleavagewithin the type A transcript occurs in a core regionof the RNA that is common to eukaryotic smallsubunit rRNAs. We present two possible expla-nations for this ®nding. The O and A gene tran-scripts vary in sequence at the processing site andspeci®c cleavage could be responsible for the stab-ility of the O ribosome in the presence of degrad-ing A ribosomes. Physical separation of theribosomal types may also account for the selectivestability of the O type ribosome. Localization oftranscripts during the period of oocyst develop-ment indicates that there are mechanisms involvedin physically separating one type of ribosome fromthe other. This occurs despite the fact that the cyto-plasm of the oocyst is known to be open to traf®ck-ing of organelles to the maturing sporozoites untilthe point when the sporozoite buds off into thehemocoel. Here we show that although the cyto-plasm is open to traf®cking, the type O ribosomesare found only in oocysts and do not go into thesporozoite. Hence there is a mechanism keepingthe newly forming translational system separatefrom the existing one.Having three distinct types of rRNAs associatedwith a single organism permits one to study theprogression of events associated with the evolutionof a new ribosome and the global effects associatedwith such a change, which is unprecedented. Anumber of initial conclusions can be reached withthese data. The multiple ribosome system rep-resents a relatively recent adaptation after thedivergence of Plasmodium from all other Api-complexa that have been analyzed to date, sinceother members of the genus maintain complex lifecycles and do not have a multiple ribosome sys-tem. We show that the genes themselves appear tohave evolved via two gene duplication events. TheO gene, speci®c for oocyst development, divergesbefore the other two genes in P. vivax (A and S),when other Apicomplexa are used to root the phy-logenetic tree. Either the O gene or the A/S pro-genitor most closely resembles the ancestral state.The most recent duplication and divergence gaverise to the A and S genes. Neither of these dupli-cations occurs very deep on the Plasmodiumbranch, suggesting that their origin is fairly recent.There are numerous examples of genes originat-ing by duplication of a parent gene, sequence driftand co-option (Li & Graur, 1991). This scenario forthe generation of a multiple ribosome system, in-volving changes in core regions of rRNA, wouldindicate a more complicated process. This wouldinclude global changes involving compensatory al-teration of molecules associated with the ribosomecomplex (e.g. ribosomal proteins and mRNAs). Theother scenario which has resulted in different cyto-plasmic ribosomes being expressed in a single cellis secondary endosymbiosis, where one eukaryoteengulfs another, permanently retaining a second-ary nucleus, the nucleomorph, surrounded by itsown cytoplasm (Palmer & Delwiche, 1996). Endo-symbionts are essentially a cell within a cell whichmaintain separated cytoplasmic compartmentswith different ribosomes. The possibility that Plas-modium species are derived from an endosymbioticevent has been indicated (Wilson et al., 1994). Thework described here does not indicate cytoplasmiccompartmentalization in the oocyst, as there is noindication of this from ultrastructural studies(Sinden & Strong, 1978). The presence of segre-gated ribosomes may prompt re-investigation ofthe question and presents the tools for such in-quiry. This presents a challenge to the molecularphylogeneticist, who, when other related genes areavailable, should be able to relate ribosomalchange to the highly successful adaptation of thisparasite to its present niche. For those who studyRibosomal RNAs During Development of the Malaria Parasite 211
  10. 10. the machinery involved in protein synthesis theopportunity is presented to understand the adap-tation of an essential and complex system in thepresence of an already successful one.Materials and MethodsAmplification of parasite rDNAOligonucleotides used for ampli®cation and detectionof DNA fragments are designated by number andshown in Figure 2. Genomic DNA of P. vivax was pre-pared from four Thai isolates, which were con®rmed bythe CS protein based-ELISA test (Wirtz et al., 1991), andone laboratory strain Sal-1 (El Salvador strain) as de-scribed previously (Li et al., 1994b). The complete 18 SrRNA gene, including the 3Hinternal transcribed spacer(ITS1) region, was ampli®ed by polymerase chain reac-tion (PCR) from the DNA of the Sal-1 strain with oligo-nucleotides 705 and 573 as the 5Hand 3H-end primers,respectively. Partial 18 S rDNA fragments were pro-duced from the four Thai isolates with a pair of primers566 and 570, approximately 140 and 40 nucleotidesshorter from the 5Hand 3H-ends of the coding region. Thereaction was in a volume of 100 ml containing 20 to 50 ngDNA, 200 mM of each dNTP, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2 mM MgCl2, and 2.5 units Taq DNA poly-merase (Perkin Elmer Cetus; Norwalk, CT) in a PerkinElmer DNA Thermal Cycler with the following par-ameters: 94C/one minute, 55C/one minute, 72C/twoto three minutes and a total of 30 cycles.Cloning and sequence analysisThe procedure for cloning ampli®ed rRNA genes hasbeen described (Li et al., 1994b). The complete sequencesof the three 18 S rRNA genes and the ITS regions weredetermined in both directions from the Sal-1 strain andThai isolates. Sequence alignment was performed withLasergene software (DNASTAR Corp., Madison, WI)and AE2 software (T. Macke, Scripps Clinic; availablefrom Ribosome Database Project at http://rdp.life.uiu-c.edu). Secondary structure models were inferred fromcomparative sequence analysis. The sequences of P. vivaxrRNAs were aligned with those of the available 18 SrRNAs, and the models were constructed according tomaximum similarity in both primary and secondarystructures (Gutell et al., 1994), since identical structurecan be folded from the same type of rRNA with differentprimary sequences. Our comparative studies have alsoidenti®ed tertiary interactions (Gautheret et al., 1995).The structures were drawn with the computer programXRNA developed by B. Weiser and H. Noller (Universityof California, Santa Cruz; available at ftp://fangio.ucs-c.edu/pub/XRNA).Isolation of RNA from infected mosquitoesLaboratory-reared mosquitoes, Anopheles dirus, werefed on P. vivax-infected Thai patients. The engorged mos-quitoes were held at 26C and ten mosquitoes were re-moved and frozen at À70C at intervals beginning twohours after the blood meal and thereafter every two daysexcept day 4. The development of the parasite in themosquito was followed by microscopic examination ofmidguts for oocysts and salivary glands for sporozoites.Preparation and analysis of total RNA from the P. vivaxinfected blood and mosquitoes were based on the meth-od previously described (Chomczynski Sacchi, 1987;Li et al., 1994b).Phylogenetic analysis18 RNA sequences were aligned based on the con-served sequences and corresponding secondary struc-tures (Gutell et al ., 1994). Regions of the genes that werenot unambiguously aligned were not used in the analy-sis. Sequences were analyzed by the parsimony method(Swofford, 1993) using 100 heuristic random addition re-plicatives and 100 bootstrap replicates. Uninformativecharacters were ignored. Genbank accession numbers forthe sequences used in this study are: P. vivax (Sal-1)U07367, U03768 and U93095 (this study); P. vivax (Thai)U93233, U93234 and U93235 (this study); Plasmodium cy-nomolgi L08241 and L08242; Babesia equi Z15105; Theileriabuffeli Z15106; Sarcocystis gigantia L24384; Toxoplasmagondii M97703; Cryptosporidium wrairi U11440; Prorocen-trum micans M14649; Symbiodinium microadriaticumM88521; Paramecium tetraurelia X03772; Oxytricha novaM14601.Nucleotide sequence data reported in this paper havebeen reported to GenBank with the accession numbersU93233, U93234, U93235 and U93095.AcknowledgementsProcedures for drawing blood and feeding mosquitoeson patients were approved by the human use commit-tees of the Ministry of Public Health, Thailand, and USArmy Medical Research and Development Command.The work was supported by a World Health Organiz-ation grant, 890093, and by NIH grant GM48207(awarded to R.R.G.) and the NSF/Sloan Foundation(J.C.K.). R.R.G. and S.H.D. thank the W.M. Keck fortheir generous support of RNA Science on the BoulderCampus.ReferencesAllsopp, M. T., Cavalier-Smith, T., De Waal, D. T. Allsopp, B. A. (1994). Phylogeny and evolution ofthe piroplasms. Parasitology, 108, 147±152.Chernoff, Y. O., Newnam, G P. Liebman, S. W.(1966). The translational function of nucleotideC1054 in the small subunit rRNA is conservedthroughout evolution: genetic evidence in yeast.Proc. Natl Acad. Sci. USA, 93, 2517±2522.Chromczynski, P. Sacchi, N (1987). Single-stepmethod of RNA isolation by acid guanidinium thio-cyantate-phenol-choroform extraction. Anal. Bio-chem. 162, 156±159.Escalante, A. A. Ayala, F. J. (1995). Evolutionary ori-gin of Plasmodium and other Apicomplexa basedon rRNA genes. Proc. Natl Acad. Sci. USA, 92,5793±5797.Etter, A., Bernard, V., Kenzelmann, M., Tobler, H Muller, F. (1994). Ribosomal heterogeneity fromchromatin diminution in Ascaris lumbricoids. Science,265, 954±956.Gautheret, D., Damberger, S. H. Gutell, R. R. (1995).Identi®cation of base-triples in RNA using com-parative sequence analysis. J. Mol. Biol. 248, 27±43.Goringer, H. U., Hijazi, K. A., Murgola, E. J. Dahlberg, A. E. (1991). Mutations in 16S rRNA that212 Ribosomal RNAs During Development of the Malaria Parasite
  11. 11. affect UGA (stop codon)-directed translationtermination. Proc. Natl Acad. Sci. USA, 88, 6603±6607.Gunderson, J. H., Sogin, M. L., Wollett, G., Hollingdale,M., de la Cruz, V. F., Waters, A. P McCutchan,T. F. (1987). Structurally distinct, stage-speci®c ribo-somes occur in Plasmodium. Science, 238, 933±937.Gutell, R. R., Larsen, N. Woese, C. R. (1994). Lessonsfrom an evolving rRNA: 16S and 23S rRNA struc-tures from a comparative perspective. Microbiol.Rev. 58, 10±26.Li, J., Zhu, J. D., Appiah, A., McCutchan, T. F., Long,G. W., Milhous, W. K. Hollingdale, M. R. (1991).Plasmodium berghei: quantitation of in vitro effects ofantimalarial drugs on exoerythrocytic developmentby a ribosomal RNA probe. Exp. Parasitol, 72, 450±458.Li, J., McConkey, G. A., Rogers, M. J., Waters, A. P. McCutchan, T. F. (1994a). Plasmodium: the devel-opmentally regulated ribosome. Exp. Parasitol. 78,437±441.Li, J., Wirtz, R. A., McConkey, G. A., Sattabongkot, J. McCutchan, T. F. (1994b). Transition of Plasmodiumvivax ribosome types corresponds to sporozoitedifferentiation in the mosquito. Mol. Biochem. Parasi-tol. 65, 283±289.Li, J., Wirtz, R. A., McConkey, G. A., Sattabongkot, J.,Waters, A. P., Rogers, M. J. McCutchan, T. F.(1995). Plasmodium: genus-conserved primers forspecies identi®cation and quantitation. Exp. Parasi-tol. 81, 182±190.Li, J., Wirtz, R. A. McCutchan, T. F. (1997). Analysisof malaria parasite RNA from decade old Giemsa-stained blood smears and dried mosquitoes. Am. J.Trop. Med. Hyg., in the press.Li, W.-H. Graur, D. (1991). In Fundamentals of Molecu-lar Evolution (Li, W.-H. Grauer, D., eds), pp. 136±167, Sinauer Associates, Inc., Sunderland, MA.Liebman, S. W., Chernoff, Y. O. Liu, R (1995). The ac-curacy center of a eucaryotic ribosome. Biochem.Cell. Biol. 73, 1141±1149.Lodmell, J. S., Gutell, R. R. Dahlberg, A. E. (1995).Genetic and comparative analyses reveal an alterna-tive secondary structure in the region of nt 912 ofEscherichia coli 16S rRNA. Proc. Natl Acad Sci. USA,92, 10555±10559.McCutchan, T. F. (1986). The ribosomal genes ofPlasmodium. Int. Rev. Cytol. 99, 295±309.McCutchan, T. F., de al Cruz, V. F., Lal, A. A.,Gunderson, J. H., Elwood, H. J. Sogin, M. L.(1988). Primary sequences of two small subunitribosomal RNA genes from Plasmodium falciparum.Mol. Biochem. Parasitol, 28, 63±68.McCutchan, T. F., Li, J., McConkey, G. A., Rogers, M. J. Waters, A. P. (1995). The cytoplasmic ribosomalRNAs of Plasmodium spp. Parasitol. Today, 11, 134±138.Moine, H Dahlberg, A. E. (1994). Mutations in helix34 of Escherichia coli 16S ribosomal RNA have mul-tiple effects on ribosome function and synthesis.J. Mol. Biol. 243, 402±412.Murgola, E. J., Hijazi, K A., Goringer, H. U. Dahlberg,A. E. (1988). Mutant 16S ribosomal RNA: a codon-speci®c translational suppressor. Proc. Natl Acad.Sci. USA, 85, 4162±4165.Palmer, J. D. Delwiche, C. F. (1996). Second-Handchloroplasts and the case of the disappearingnucleus. Proc. Natl Acad. Sci. USA, 93, 7432±7435.Rogers, M. J., McConkey, G. A., Li, J. McCutchan, T.F (1995). The ribosomal DNA loci in Plasmodium fal-ciparum accumulate mutations independently. J. Mol.Biol. 254, 881±891.Rogers, M. J., Gutell, R. R., Damberger, S. H., Li, J.,McConkey, G. A., Waters, A. P. McCutchan, T. F.(1996). Structural features of the large subunitrRNA expressed in Plasmodium falciparum sporo-zoites that distinguish it from the asexuallyexpressed large subunit rRNA. RNA, 2, 134±145.Sinden, R. E. Strong, K. (1978). An ultrastructuralstudy of the sporogonic development of Plasmo-dium falciparum in Anopheles gambiae. Trans. Roy.Soc. Trop. Med. Hyg. 72, 477±491.Sollner-Webb, B. Tower, J. (1986). Transcription ofcloned eukaryotic ribosomal RNA genes. Annu. Rev.Biochem. 55, 801±830.Swofford, D. L. (1993). PAUP: Phylogenetic analysisusing parsimony, Version 3.1.1. Distributed by theIllinois Natural History Survey. Champaign, IL.Waters, A. P., Syin, C. McCutchan, T. F. (1989).Developmental regulation of stage-speci®c ribosomepopulations in Plasmodium. Nature, 342, 438±440.Wellems, T. E., Walliker, D., Smith, C. L., Do Rosario,V. E., Maloy, W. L., Howard, R. J., Carter, R. McCutchan, T. F. (1987). A histidine-rich proteingene marks a linkage group favored strongly in agenetic cross of Plasmodium falciparum. Cell, 49,633±642.Wilson, R. J. M., Williamson, D. H. Preiser, P. (1994).Malaria and other Apicomplexans: the ``plantconnection. Infect. Agents Dis. 3, 29±37.Wirtz, R. A., Charoenvit, Y., Burkot, T. R., Esser, K. M.,Beaudoin, R. L., Collins, W. E. Andre, R. G.(1991). Evaluation of monoclonal antibodies againstPlasmodium vivax sporozoites for ELISAdevelopment. Med. Vet. Entomol. 5, 17±22.Zimmerman, R. A. (1996). The decoding domain. InRibosomal RNA: Structure, Evolution, Processing, andFunctional in Protein Biosynthesis (Zimmerman,R. A. Dahlberg, A. E., eds), pp. 277±309, CRCPress, Boca Raton, FL.Edited by D. E. Draper(Received 27 January 1997; received in revised form 18 March 1997; accepted 18 March 1997)Ribosomal RNAs During Development of the Malaria Parasite 213

×