Molecular & Biochemical Parasitology 119 (2002) 121–124

Short communication

Surface expression of the conserved ribosomal protein P0 onparasite and other cells

Subhash Singh, Alfica Sehgal, Sanjeev Waghmare, Tirtha Chakraborty,Arunava Goswami 1, Shobhona Sharma *

Department of Biological Sciences, Tata Institute of Fundamental Research, Homi Bhabha Road, Mumbai 400 005, India

Received 9 June 2001; received in revised form 6 September 2001; accepted 12 September 2001

Keywords: Ribosomal phosphoprotein P0; Mammalian cells; Toxoplasma ; Yeast; Surface expression

www.parasitology-online.com.

Invasion-blocking antibodies against the ribosomalphosphoprotein P0 of the human malarial parasitePlasmodium falciparum (PfP0) have been identifiedthrough a differential immunoscreen using immune andpatient sera from malaria endemic regions of EasternIndia [1]. Purified IgG from rabbit sera, raised againstdifferent domains of the PfP0 protein, show no effecton the intraerythrocytic stages of Plasmodium falci-parum in culture, but inhibit the invasion of erythro-cytes by merozoites in a concentration dependentmanner [2,3]. The same antibodies against the PfP0protein protect mice against challenge with a virulentstrain of Plasmodium yoelii [4]. We have recentlydemonstrated this protein to be present on the surfaceof merozoites and gametocytes of Plasmodium [3,4].The phosphoprotein P0 is highly conserved across eu-karyotic organisms [5], and is related to the family ofphosphoproteins P1 and P2, due to the highly ho-mologous carboxy-terminal 22 amino acid domain [6].The ribosomal function of P0 is mediated through apentameric P12·P0·P22 complex, which forms the stalkof the large ribosomal subunit at the GTPase center [7].

P0 interacts with the elongation factor eEF2 [8], and isessential for the ribosomal activity and cell viability inyeast [9]. Through deletion analysis of the P0 protein,the ribosomal function has been mapped to amino acidposition 185–230 in yeast P0 [10].

Since P0 is a very conserved protein [5], we wonderedwhether the surface expression of P0 is an exclusiveproperty of Plasmodium cells, or if it is a generalproperty of the P0 protein in other organisms as well.The presence of this protein on the host cell surface willhave serious implications with the usage of this proteinin a malaria vaccine. This question also has a directbearing on the mechanism of translocation of thisprotein to the surface, as to whether it is a uniquephenomenon for Plasmodium. Therefore, we decided toinvestigate the presence of this protein on different celltypes. In this paper we report the localization of P0protein on the surface of another Apicomplexan para-site Toxoplasma, yeast and mammalian cell lines usingmono-specific cross-reactive anti-PfP0 antibodies.

In order to test for cross-reactivity of anti-PfP0 serawith different cell types, Western Blot analysis wascarried out with total protein extract of these cell types.Anti-PfP0 antisera were raised in rabbits against recom-binant GST-fusion proteins, PfP0N (amino acid 17–61)and PfP0C (amino acid 61–316), and purified as de-scribed earlier [3]. Specific single band reactivities of 38,33, 34 and 38 kD proteins were observed with Toxo-plasma gondii, yeast (Saccharomyces cere�isiae, strainEG103), Chinese Hamster Ovary (CHO) cells and hu-man leukocyte protein preparations respectively, using

Abbre�iations: GST, glutathione-S-transferase; IPTG, isopropyl-thio-galactosidase; PfP0, Plasmodium falciparum phosphoriboproteinP0; PfP0N, amino-terminal domain of PfP0; PfP0C, carboxy-terminaldomain of PfP0.

* Corresponding author. Tel: +91-22-215-2971/2979 x 2570; fax:+91-22-215-2110/2181.

E-mail address: sharma@tifr.res.in (S. Sharma).1 Present address: Lecturer, Agricultural Science Unit, Indian

Statistical Institute, 203, B.T. Road, Calcutta 700 035, India.

0166-6851/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 6 -6851 (01 )00394 -2

S. Singh et al. / Molecular & Biochemical Parasitology 119 (2002) 121–124122

both anti-PfP0N and anti-PfP0C antibodies (Fig. 1).These molecular weights are the expected molecularweights of the respective P0 proteins. Since both anti-PfP0N and anti-PfP0C sera identify the same singleprotein band in each of the organisms tested, these resultsshow that the sera are specifically recognizing the ho-mologous P0 proteins.

Toxoplasma gondii tachyzoites, yeast spheroplasts,CHO cells, the human cell lines K562 (an ery-throleukemic cell line), U937 (a lymphoma cell line),Daudi (a Burkitt lymphoma cell line) and SCC (anepithelial cell line), and peripheral human blood cellswere subjected to solution immunofluorescence assay(SIFA). Toxoplasma tachyzoites were harvested by filter-ing the infected HFF cell lysate through a 3 �m filter andused for SIFA [11]. Yeast cells were grown with 1%glucose and harvested at an OD600 value of 1–2 (�107

cells). Yeast spheroplasts were prepared by treatmentwith zymolyase (Seikagaku Corp., Japan) [12], and usedfor SIFA. CHO cells were subcultured in 35 mm2 Petriplates and the adherent CHO cells were used for SIFAassay after 24 h of growth. A distinct surface staining wasobserved with every tachyzoite of Toxoplasma, verysimilar to that shown by the surface marker SAG1 (giftfrom David Roos) (Fig. 2A). Yeast spheroplasts alsoshowed distinct surface reactivity with anti-PfP0 serasamples, but the number of cells that reacted were30–50% of the total cell population (Fig. 2B). With CHOcells, the number of cells found to be reactive was about2–10% (Fig. 2C). In some of the reactive CHO cells, thesurface reactivity could be detected at the membraneextensions quite distinctly (Fig. 2C, right panel). Humanperipheral blood cells showed no SIFA activity (Fig. 2D).Approximately 100,000 total blood cells were scanned for

each antibody tested, but no reactivity was detected witheither red cells or leukocytes. Red cells do not possess anyP0 protein, as fixed red cells do not show any IFAreactivity [2]. To test the leukocytes specifically, 105 cellsof the buffy coat samples were collected from threedonors and each sample was tested with the antibodies,and no surface reactivity was detected in any of thesesamples (data not shown). Each of these donors washealthy local (area of low malaria incidence) individual.Control monoclonal antibodies against glycophorin A(Pharmingen, USA) and HLA Class I (gift from S.Chiplunkar) reacted as expected with red cells andleukocytes (Fig. 2D). However, even though peripheralblood cells did not show any surface reactivity, hemopoi-etic cell lines K562, U937 and Daudi cells showed distinctcell staining (Fig. 2E). The epithelial cell line SCC alsoshowed surface staining (Fig. 2E, d). The frequency ofcells staining for these cell lines was low, varied with celllines, and with different experiments, but was alwayswithin 2–5%. For each cell type, SIFA experiments wereperformed at least three times. Preimmune sera did notshow any fluorescence activity with any of these differentcell types used. The numbers and the patterns of reactiv-ity were found to be similar irrespective of whetheranti-PfP0N or anti-PfP0C serum sample was used.

The PfP0N and PfP0C possess no overlapping aminoacid sequence. We have shown earlier that the anti-PfP0N and anti-PfP0C antibodies do not cross-react witheach other [3]. On Western Blots both these sera light upspecifically only the homologous P0 protein. Since bothanti-PfP0N and anti-PfP0C antisera react to the cellsurfaces with a similar pattern, it is unlikely that thesurface reactivity recorded is due to non-specific P0-likecross-reactive domains located on the surface of theparasite.

About 10–15% of patients suffering from an autoim-mune disorder, Systemic Lupus Erythematosus (SLE),possess autoantibodies against the conserved carboxy-terminal 22 amino acids of the P-proteins [13]. SLE serumsample has been used for immunoprecipitation study ofP. falciparum protein extract, and distinct cross-reactivitywith PfP0 has been observed [14]. It has been reportedearlier that SLE sera do not react to the amino terminaldomain of the P-proteins [13]. We have also recentlyreported that 10% of SLE patient sera samples cross-re-act with P. falciparum P0 protein, but only with PfP0Cand not PfP0N protein [15]. We have shown that thesePfP0C positive SLE sera samples inhibit the growth ofP. falciparum through specific reactivity with PfP0C [15].These results confirm the conserved structures of PfP0and human P0.

So far only one report documents the surface localiza-tion of human P0-like protein, and this has been reportedon the plasma membrane of human HepG2 and neuronalcell lines using purified SLE sera [16]. Results presentedin this paper show that the surface expression of P0 is

Fig. 1. Western blot analysis of total protein extracts from differentorganisms. Lane A contains: 106 cells of Toxoplamsa tachyzoites; B:108 cells of Saccharomyces cere�isiae ; C: 106 cells of CHO cells; andD: 106 cells of human white blood cells. In each panel, the left-handstrip was treated with preimmune serum, the middle strip withanti-PfP0N and the right-hand strip with anti-PfP0C antisera. In eachcase, the serum was used at a dilution of 1:100, except for strips ofpanel A, where the sera were used at 1:200 dilution. Positions ofmolecular weight standards used are shown on the left. The arrow-heads indicate the P0 bands.

S. Singh et al. / Molecular & Biochemical Parasitology 119 (2002) 121–124 123

Fig. 2. Solution immunofluoresence assay (SIFA) of different cell types using different antisera, used at 1:100 dilutions (3); followed by secondaryantibody treatment with goat-anti-rabbit IgG, or goat anti-mouse IgG, conjugated with FITC or rhodamine (Boehringer Mannheim, Germany)diluted in RPMI 1640 medium. Panels A: Toxoplasma tachyzoites; B: yeast spheroplasts; C: CHO cells; D: human blood cells; E: human cell lines(a: K562 cell; b: Daudi cells; c: U937 cell and d: SCC cell). In each panel ‘Br’ shows the bright field of the corresponding dark field shown nextto it. Additional dark fields are shown on the right-hand side in panels B and C. Treatment with N: anti-PfP0N antibodies; C: anti-PfP0Cantibodies; PI: Preimmune sera; and surface markers M: anti-SAG1 antibody; RM: anti-glycophorin A antibody; LM: anti-HLA Class I antibody.In panel A, the lowest row shows Toxoplasma tachyzoites double stained with anti-PfP0N antibodies and anti-SAG1 antibodies. In panels C andD: the arrows indicate the SIFA active subset of CHO cells and leukocytes, respectively. The bar corresponds to 5 �m.

regulated. While every Plasmodium [3] and Toxoplasmacell showed surface staining, the frequency for surfacereactivity was 30–50% for yeast cells, and as low as2–10% in mammalian cell lines. Peripheral blood cells(red cells and leukocytes) did not show any surfacereactivity. The reason for this variation in SIFA reactiv-ity is unclear. The low frequency of positively reactingcells is unlikely to be dead cells in the case of leukocytes,CHO, K562 and Daudi cell lines as we routinely per-formed the Trypan Blue exclusion assay to assess cellviability. We have also performed fixed sample analysisof these cell types, and all cells show the same patternsof reactivity with P0 and other markers such as beta-tubulin and ER-marker. Whether the SIFA reactive cellsbelong to an early stage of apoptosis remains to be tested.

From the cross-reactivity of anti-PfP0 antibodies it isclear that large protein domains of PfP0 are not suitablefor a malaria vaccine. For vaccination purposes, smaller

stretches of PfP0 with no cross-reactivity with the humanP0 protein need to be identified. Passive immunizationswith purified IgG obtained from immune adults, how-ever, have been performed earlier on malaria patientswith no apparent ill effects [17,18]. The experimentalmice, which recovered completely in our passive immu-nization experiments, were healthy with no pathologicalsymptoms of SLE [4]. The observation that peripheralblood cells do not show surface expression of P0 proteinis an encouraging observation for passive immun-otherapy. However, other human tissues also need to beexamined for surface PfP0 expression. A large number(87%) of the malaria immune adults from Eastern-India possessed anti-PfP0 antibodies, but did notexhibit any symptoms of SLE as documentedthroughextensive questionnaire [1]. Indeed, it has been notedearlier that the proportion of SLE patients in malariaendemic areas is lower than that in areas with no malaria[19].

S. Singh et al. / Molecular & Biochemical Parasitology 119 (2002) 121–124124

The biology of the P0 protein is complex and intrigu-ing. It is apparent that it plays an essential role in theribosomes [7,8] and also exhibits apurinic/apyrimidinicendonuclease activity in the nucleus [20]. Unlike mostother ribosomal proteins, the level of P0 mRNA andprotein are regulated, as has been shown in carcino-genic and apoptotic cells [21,22]. In flies, P0 also playsregulatory roles [23,24]. The ribosomal P2 protein ho-mologue, L12, of Neisseria gonorrhoeae has recentlybeen shown to play a role in gonococcal invasion ofhuman endometrial Hec1B cells [25]. How P0 proteingets transported on the surface remains a question, asthe deduced P0 protein sequences do not possess anycanonical signal sequence or transmembrane motif. It ispossible that the P0 protein is secreted out in a complexwith other protein(s). Future studies centered on cellbiological analysis of the membrane transportation ofP0, along with attempts to map the various functions ofP0 on sub-domains of the protein, will help in under-standing the diverse functions of this protein.

Acknowledgements

We are indebted to David Roos (University of Penn-sylvania, Philadelphia, PA) for antibodies to SAG1antigen, and for the methodology with Toxoplasma. Wealso thank Shubha Chiplunkar (Cancer Research Insti-tute, Mumbai) for providing us with the anti-HLAantibodies and the human cell lines and Satyajit Mayor(National Centre for Biological Sciences, Bangalore)for the CHO cells.

References

[1] Lobo CA, Kar SK, Ravindran B, Kabilan L, Sharma S. Novelproteins of Plasmodium falciparum identified by differential im-munoscreening using immune and patient sera. Infect Immun1994;62:651–6.

[2] Goswami A, Singh S, Redkar VD, Sharma S. Characterizationof P0, a ribosomal phosphoprotein of Plasmodium falciparum. JBiol Chem 1997;272:12138–43.

[3] Chatterjee S, Singh S, Sohoni R, et al. Characterization ofdomains of the phosphoriboprotein P0 of Plasmodium falci-parum. Mol Biochem Parasitol 2000;107:143–54.

[4] Chatterjee S, Singh S, Sohoni R, et al. Antibodies against theribosomal phosphoprotein P0 of Plasmodium falciparum protectmice against challenge with P. yoelii. Infect Immun2000;68:4312–8.

[5] Goswami A, Chatterjee S, Sharma S. Cloning of a ribosomalphosphoprotein P0 gene homologue from Plasmodium falci-parum. Mol Biochem Parasitol 1996;82:117–20.

[6] Rich BE, Steitz JA. Human acidic ribosomal phosphoproteinsP0, P1 and P2: Analysis of cDNA clones, in vitro synthesis and

assembly. Mol Cell Biol 1987;7:4065–74.[7] Uchiumi T, Kominami R. Direct evidence for interaction of the

conserved GTPase domain within 28S RNA with mammalianribosomal acidic phosphoproteins and L12. J Biol Chem1992;267:19179–85.

[8] Justice MC, Ku T, Hsu MJ, et al. Mutations in ribosomalprotein L10e confer resistance to the fungal-specific eukaryoticelongation factor 2 inhibitor sordarin. J Biol Chem1999;274:4869–75.

[9] Santos C, Ballesta JPG. Ribosomal protein P0, contrary tophosphoproteins P1 and P2, is required for ribosome activityand Saccharomyces cere�isiae viability. J Biol Chem1994;269:15689–96.

[10] Santos C, Ballesta JPG. The highly conserved protein P0 car-boxyl end is essential for ribosome activity only in the absence ofproteins P1 and P2. J Biol Chem 1995;270:20608–14.

[11] Roos DS, Donald GK, Morrissette NS, Moulton LC. Moleculartools for genetic dissection of the protozoan parasite Toxoplasmagondii. Methods Cell Biol 1994;45:27–63.

[12] Pringle JR, Adams AEM, Drubin DG, Haarer BK. Immu-nofluorescence methods for yeast. Methods Enzymol1991;194:565–602.

[13] Elkon K, Skelly S, Parnassa A, Moller W, Danho W, WeissbachH, Brot N. Identification and chemical synthesis of a ribosomalprotein antigenic determinant in systemic lupus erythematosus.Proc Natl Acad Sci USA 1986;83:7419–23.

[14] Francoeur AM, Peebles CLJ, Heckman K, Lee JC, Tan EM.Identification of ribosomal protein autoantigens. J Immunol1985;135:2378–84.

[15] Singh S, Chatterjee S, Sohoni R, Badakare S, Sharma S. Serafrom lupus patients inhibit growth of Plasmodium falciparum inculture. Autoimmunity 2001;33:253–63.

[16] Koren E, Reichlin MW, Kosec M, Fugate RD, Reichlin M.Autoantibodies to the ribosomal P proteins react with a plasmamembrane-related target on human cells. J Clin Invest1992;89:1236–41.

[17] Cohen S, Mcgregor IA, Carrington SP. Gammaglobulin andacquired immunity to human malaria. Nature 1961;192:733–7.

[18] Sabchareon A, Burnouf T, Quattara D, et al. Parasitological andclinical response to immunoglobulin administration in P. falci-parum malaria. Am J Trop Med Hyg 1991;45:297–308.

[19] Greenwood BM. Autoimmune disease and parasitic infections inNigerians. Lancet 1968;2:380–2.

[20] Yacoub A, Kelley MR, Deutsch WA. Drosophila ribosomalprotein P0 contains apurinic/apyrimidinic endonuclease activity.Nucleic Acids Res 1996;24:4298–303.

[21] Kondoh N, Wakatsuki T, Ryo A, et al. Identification andcharacterization of genes associated with human hepatocellularcarcinogenesis. Cancer Res 1999;59:4990–6.

[22] Brockstedt E, Rickers A, Kostka S, et al. Identification ofapoptosis-associated proteins in a human Burkitt lymphoma cellline. Cleavage of heterogeneous nuclear ribonucleoprotein A1 bycaspase 3. J Biol Chem 1998;273:28057–64.

[23] Frolov MV, Birchler JA. Mutation in P0, a dual functionribosomal protein/apurinic/apyrimidinic endonuclease, modifiesgene expression and position effect variegation in Drosophila.Genetics 1998;150:1487–95.

[24] Craig TL, Denlinger DL. Sequence and transcription patterns of60S ribosomal protein P0, a diapause-regulated AP endonuleasein the flesh fly, Sarcophaga crassipalpis. Gene 2000;255:381–8.

[25] Spence JM, Clark VL. Role of ribosomal protein L12 in Gono-coccal invasion of Hec1B cells. Infect Immun 2001;68:5002–10.