Molecular characterization of a gene encoding a membrane protein of Spiroplasma citri

Gene 189 (1997) 95–100

Molecular characterization of a gene encoding a membrane protein of Spiroplasma citri Fengchun Ye a,1, Ulrich Melcher b, Jacqueline Fletcher a,* a Department of Plant Pathology, Oklahoma State University, 110 Noble Research Center, Stillwater, OK 74078, USA b Department of Biochemistry and Molecular Biology, Oklahoma State University, Stillwater, OK 74078, USA Received 13 March 1996; revised 5 November 1996; accepted 6 November 1996

Abstract A 9.6-kb genomic DNA segment, previously cloned from the phytopathogen Spiroplasma citri BR3-3X [Fletcher et al. (1981) Phytopathology 71, 1073–1080], contained several open reading frames including one encoding a 58-kDa protein. In this work, the transcription initiation site of the P58 mRNA was mapped and part of the gene was expressed in Escherichia coli as a fusion protein. A synthetic peptide, whose sequence is included in the fusion protein, was produced. Antibodies against both the fusion protein and the peptide reacted with a 60-kDa protein in a S. citri total protein extract. Hydrophobicity characteristics of this protein and its fractionation into the detergent phase indicated that P58, which shares limited sequence similarity with the adhesin of Mycoplasma hominis and the attachment protein of M. genitalium, is an integral membrane protein. [ Elsevier Science B.V. All rights reserved. Keywords: Adhesin; Mycoplasma, primer extension; Recombinant protein; Western blot

1. Introduction The phytopathogenic prokaryote Spiroplasma citri (Saglio et al., 1971) is transmitted from plant to plant via the leafhopper, Circulifer tenellus ( Kaloostian et al., 1979). Like other members of the class Mollicutes, such as mycoplasmas, spiroplasmas have no cell wall but only a trilaminar cell membrane (Saglio et al., 1971). Membrane proteins, especially surface proteins, have been reported to be very important for the interaction * Corresponding author. Tel. +1 405 7449948; Fax +1 405 7447373; e-mail: [email protected] 1 Present address: Department of Surgery, Urology Division, McGill University and Montreal General Hospital Research Institute, 1650 Avenue Cedar, Montreal, Quebec, Canada. Abbreviations: aa, amino acid(s); Ap, ampicillin; CCR, central coding region; cDNA, DNA complementary to RNA; EDTA, disodium ethylene diamine tetraacetate; IgG, immunoglobulin G; IGL, intergenic region of the left end; IGR, intergenic region of the right end; IPTG, isopropyl-b--thiogalactopyranoside; kb, kilobase(s) or 1000 bp; malE, maltose-binding protein gene of E. coli; nt, nucleotide(s); ORF, open reading frame(s); PAGE, polyacrylamide gel electrophoresis; PCR, polymerase chain reaction; PMSF, phenylmethylsulfonyl fluoride; SD, Shine-Dalgarno (sequence); SDS, sodium dodecyl sulfate; TE, 10 mM Tris, 1 mM EDTA, pH 7.5. 0378-1119/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved PII S 03 7 8 -1 1 1 9 ( 9 6 ) 0 0 8 40 - 2

between many species of mycoplasmas and their hosts (for reviews, see Wieslander et al., 1992; Wise, 1993). For instance, the P1 surface protein of the human pathogen Mycoplasma pneumoniae is a mediator for the attachment to and invasion of its host cells (Baseman et al., 1982). Similar adhesin proteins have been found for a number of other species of mycoplasmas (Dallo et al., 1989; Henrich et al., 1993; Tham et al., 1994). Little is known about how spiroplasmas interact with their parasitized plant or insect cells, and very few surface membrane proteins of spiroplasmas have been characterized (Fletcher et al., 1989). We previously mapped the genomes of S. citri insect transmissible lines, BR3-3X and BR3-T ( Wayadande et al., 1995), and the insect non-transmissible line, BR3-G, which was derived from long-term maintenance of BR3-3X by grafting in plants ( Wayadande et al., 1995). A large chromosomal inversion and a 10-kb deletion near each of the inversion borders were found in the genome of BR3-G; a 9.6-kb genomic DNA segment corresponding to one of the deletions in the BR3-G genome was cloned from BR3-3X and sequenced ( Ye et al., 1996). In that work, at least two copies of the CCR of this fragment were found in genomes of BR3-3X and BR3-T, but one copy was deleted in BR3-G. The CCR of this segment con-

96

F. Ye et al. / Gene 189 (1997) 95–100

tains four putative open reading frames (ORFs), P58, P12, P54 and P123 (Fig. 1A), for which the gene products and their functions were unknown. In this study, we report the identification and cell membrane localization of the P58 protein, and its limited sequence similarity with the adhesin protein, P50, of Mycoplasma hominis ( EMBL accession No. X73834) and with the attachment protein, MG191, of Mycoplasma genitalium (Fraser et al., 1995).

2. Experimental and discussion 2.1. Nucleotide (nt) and deduced amino acid (aa) sequences of P58, a putative membrane protein of S. citri The nt sequence of the ORF for P58 and its deduced aa sequence are shown in Fig. 1B. Nine nt upstream of the start codon ATG is GAGG, which may serve as a Shine-Dalgarno (SD) sequence. About 60 residues

Fig. 1. (A) Genetic organization of the 9.6 kb genomic segment of S. citri BR3-3X. Arrows indicate transcription directions of genes. Tn∞ase: SpV1-related transposase gene; P58, P12, P54, P123, P18: unknown S. citri protein genes; SpV1 coat-sim: interrupted and incomplete SpV1 virus coat protein gene. LCR, left coding region; IGL, left intergenic region; CCR, central coding region; IGR, right intergenic region; RCR, right coding region. At least two copies of the CCR of this segment are present in the genomes of BR3-3X and BR3-T, but one copy was naturally deleted in BR3-G. The two deletion borders were located within IGL and IGR. (B) Nt and deduced aa sequences of the ORF encoding protein P58 of S. citri BR3 (GenBank accession No. U44405). The putative SD site, the start codon, the −35 and −10 Pribnow box ( TATAAAT ) regions are underlined. The transcription start, determined by primer extension (data not shown), and the stop codon are indicated with an asterisk, and the primer used for the extension is shown with an arrow. The two inversely repeated sequences downstream of the coding region are indicated with arrows. A possible transmembrane helix and a possible ‘leucine-zipper’ (dimerization) domain are indicated, with the repeated leucine/isoleucine residues underdotted. Potentially antigenic regions were predicted using MacVector ( Kodak, New Haven, CT, USA). ‘Predict Protein’ (Rost et al., 1994) was used to predict regions of loop structure that had a high probability of being exposed. Sequence selected for synthesis of a peptide (boxed ) was identified consistently by both methods to be antigenic. The sequence selected for the MBP-P58 fusion protein in E. coli is single underlined. Amino acid residues in the alignment identical or similar to those of P50 of Mycoplasma hominis are indicated with dark and light shading, respectively.

F. Ye et al. / Gene 189 (1997) 95–100

upstream there is a putative promoter sequence, the Pribnow box ( TATAAAT ) preceded by TGTTGAGAAATA at the correct spacing to be a −35 region. Dot matrix analysis suggested a hairpin immediately downstream of the coding region, a possible factorindependent terminator of transcription. Northern blot hybridization of S. citri BR3 total RNAs using CCR of the 9.6-kb segment as probe revealed four mRNA bands with sizes close to those of the expected P58, P12, P54 and P123 transcripts (data not shown), suggesting that these genes are transcribed independently rather than as a single operon. Primer extension (data not shown) indicated that transcription of the P58 gene started with the base G that is 12 nt downstream of the Pribnow box. The predicted aa sequence of P58 is 494 residues long (58 329 Da). No repeating units of primary structure were found. An analysis of hydrophobicity revealed that the protein is mostly hydrophilic, but the N-terminus (11 residues) is hydrophobic and a second hydrophobic stretch exists between residues 54 and 75. These observations suggest that this is a putative membrane protein. That the N-terminus is hydrophobic suggests that it may be a signal sequence for secretion or membrane insertion, and the hydrophobic stretch between residues 54 and 75 is of sufficient length to be a transmembrane a-helical segment. A ‘leucine zipper’like structure between residues 341 and 363 suggests that the protein may be present as a dimer in spiroplasma cells. 2.2. Product identification of the ORF encoding P58 protein and its surface membrane localization To identify the protein product of the ORF encoding the putative membrane protein, P58, two specific antibodies were raised in mice. The first antibody was against a synthetic peptide chemically linked to keyhole limpet hemocyanin. The synthesized peptide, NH -EILSSEYPDQIG-COOH, corresponding to resi2 dues 198–210 of the P58 protein, was predicted (see legend to Fig. 1) to be an antigenic domain of P58. Because spiroplasmas use the codon UGA for tryptophan rather than as a stop codon (Renaudin et al., 1986), a spiroplasma gene such as the P58 protein gene, with 9 UGA codons, cannot be fully expressed in Escherichia coli. To overcome this problem, we isolated, by polymerase chain reaction (PCR), a UGA codonfree region (aa residues 145–263, about 15 kDa) of the P58 gene and inserted it into the reading frame of the maltose binding protein (MBP, 42 kDa) gene (malE) of the E. coli expression vector pMAL-c2 (New England Biolabs; Fig. 2A). The over-expressed and purified 57-kDa MBP-P58 fusion protein (Fig. 2B) was then used as antigen to make the second antibody in mice. The anti-MBP-P58 serum was cross-absorbed with E. coli cells over-expressing the MBP before being used in

97

Western blots. By Western blotting, both antibodies strongly recognized a 60-kDa protein from total protein preparations of all S. citri BR3 lines, BR3-3X, BR3-T and BR3-G. This size is very close to the presumed size of the P58 protein. Both antibodies also detected several other, weaker bands ranging from 30 to 120 kDa in size. The anti-peptide serum also strongly reacted with the 57 kDa MBP-P58 fusion protein (data not shown), a result that was expected because the peptide sequence (aa residues 198–210) was included within the stretch (aa residues 145–263) selected for the MBP-P58 fusion protein. This confirmed that the selected sequence of P58 was correctly expressed in E. coli upon recombination with the malE gene. The fact that the 60 kDa band was revealed by both antibodies strongly suggested that it corresponded to the P58 protein of S. citri. Because the titer of the anti-peptide antibody was much lower than that of anti-fusion protein antibody, the latter was used to determine the cellular localization of the P58 protein in S. citri. Membrane and cytoplasmic proteins were prepared and subjected to SDS-polyacrylamide gel electrophoresis (PAGE ) and Western blotting. As shown in Fig. 3, the 60 kDa band was detected in both cytoplasmic and membrane fractions, indicating that the P58 protein is indeed a membrane protein. Detection of the P58 protein in the cytoplasmic preparation might be due to contamination of the cytosolic fraction by the membrane fraction, because the centrifugation speed (40 000×g) used for the fractionation after sonication was lower than the speed (100 000–130 000×g) generally recommended for membrane and cytosolic fractionation of mollicutes (Rottem et al., 1968). The isolated S. citri membranes were fractionated using 1% Triton X-114. Western blotting showed that the 60 kDa band was present only in the detergent phase and not in the aqueous phase (Fig. 4). The amphiphilic nature of this protein suggests that it is an integral membrane protein. The possibility that this protein is exposed on the surface of S. citri was examined by trypsin treatment of intact BR3-3X cells. Immunoblotting with the antiMBP-P58 antibody showed that the 60 kDa protein band was missing in the membrane preparation after trypsin treatment (data not shown), suggesting that at least part (aa residues 145–263) of the P58 was trypsinsensitive and exposed on the cell surface of the spiroplasmas. 2.3. Limited aa sequence similarity of P58 to two mycoplasma adhesins BLITZ (Sturrock and Collins, 1993) and BLAST (Altschul et al., 1990) searches of protein sequence databases revealed no definitive homologue for P58. However, among the highest scoring proteins was the M. hominis adhesin protein, P50, with 20.2% identity and 51.1% similarity over 183 residues. A GRASTA

98

F. Ye et al. / Gene 189 (1997) 95–100

Fig. 2. (A) Strategy for construction of the plasmid expressing MBP-P58 fusion protein in E. coli. The two primers used for PCR amplification of the P58 gene are indicated with arrows. The asterisk-labelled nucleotides were changed in the primer sequences to create the EcoRI and HindIII sites. A TGA codon before the HindIII site serves as stop for the MBP-P58 translation. The sequence designated malE represents part of the malE gene plus the polylinker. (B) Over-expression and purification of the MBP-P58 fusion protein. Proteins were stained with Coomassie brilliant blue. 1, Total proteins of pMAL-P58 transformed E. coli before adding IPTG; 2, total proteins of pMAL-P58 transformed E. coli protein after adding IPTG; 3, the purified 57-kDa MBP-P58 fusion protein. Method: Plasmid pMAL-c2 was obtained from New England Biolabs (Beverly, MA, USA). PCR amplification of the UGA-free part of the P58 gene was carried out under following conditions: 100 ng total DNA of S. citri BR3-3X, 0.1 mM of each primer and 1 unit Taq DNA polymerase (Promega, Madison, WI, USA), with 36 cycles of reaction: 92°C, 1 min; 72°C, 45 s; 56°C, 1 min. The pMAL-P58 recombinant plasmid was transferred to E. coli DH5a by electroporation. Correct insertion of the PCR fragment into the vector was confirmed by DNA sequencing of the recombinant plasmid. Over-expression and affinity chromatography purification of the fusion protein using amylose resin (New England Biolabs) were conducted as recommended by the manufacturers.

search of proteins encoded by the M. genitalium genome (Fraser et al., 1995) identified MG191, the attachment protein or adhesin, among the highest scoring proteins. The M. genitalium region identified was near the C-terminal end of MG191. P58, P50 and the C-terminal 500 residues of MG191 were submitted to multiple sequence alignment using the program PIMA (Smith and Smith, 1992). The significance scores (the number of standard deviations that the score for the aligned actual sequences is above the mean of scores for a series of scrambled alignments) for the alignment of P58 with P50 was 11.3 while that for MG191 with P58 was 8.2. Because both adhesin proteins of mycoplasmas function as receptors for the organisms to attach to and then invade their host cells, the similarity of P58 with these proteins suggests the possibility of a similar function for the spiroplasma protein.

3. Conclusions (1) The CCR of the 9.6 kb sequence from S. citri BR3-3X, one copy of which was deleted in the

insect non-transmissible line, BR3-G, contained several genes, among which was the gene encoding P58, a putative membrane protein. This gene had its own promoter and terminator signal sequences for transcription and SD sequence for translation. (2) Both antibodies, obtained using a synthetic peptide and an over-expressed fusion protein as antigens, recognized the 60 kDa band. Therefore, this band must correspond to the P58 protein. Data from both aa sequence analysis and Western blotting indicated that P58 was indeed a surface-exposed membrane protein of S. citri. Its limited aa sequence similarity with two mycoplasma adhesins suggests that this spiroplasma protein may have similar functions. (3) Whether P58 protein is involved in spiroplasmainsect vector interactions remains to be determined. However, the loss of one copy of the P58 gene in the insect non-transmissible mutant line, BR3-G, may not account for the loss of its insect transmissibility. The undeleted copy of this gene in BR3-G seems functional, since the P58 protein was detected in this line.

F. Ye et al. / Gene 189 (1997) 95–100

99

Acknowledgement

Fig. 3. Detection by Western blot of the P58 protein in membranes and cytoplasmic fractions of S. citri. Lane 1: size marker; lanes 2, 3, 4: BR3-3X, BR3-T and BR3-G cytoplasmic proteins; lanes 5, 6, 7: BR3-3X, BR3-T and BR3-G membrane proteins. Methods: Membrane and cytoplasmic proteins were prepared from 200 ml LD8 broth culture (Davis, 1979) of each cell line. After centrifugation (12 000×g, 30 min), S. citri cells were resuspended in 1 ml lysis buffer (50 mM Tris-HCl, pH 8.0, 1 mM PMSF, 10 mM EDTA) and then sonicated in ice-cold water (10 periods of 30 s each, with intervals of 15 s, at an output of 40 W ). After centrifugation (12 000×g, 30 min) to eliminate unlysed cells, the supernatants were subjected to high speed centrifugation (40 000×g, 1 h). The soluble cytoplasmic proteins (supernatant) were removed, and the insoluble membranes (pellets) were washed once with lysis buffer and collected by recentrifugation (40 000×g, 1 h). Membranes and cytoplasmic proteins were SDSPAGE separated and transferred to nitrocellulose. After incubation with the anti-MBP-P58 as first antibody and alkaline phosphatase conjugated anti-mouse IgG (Sigma, St. Louis, MO, USA) as second antibody, the blots were developed with nitroblue tetrazolium (5-bromo4-chloro-3-indolyl phosphate). Prestained protein standards (Bio-Rad, Melville, NY, USA): phosphorylase, 95 kDa; bovine serum albumin, 76 kDa; carbonic anhydrase, 29 kDa.

Fig. 4. Triton X-114 partitioning and immunodetection of S. citri membrane proteins. Lane 1: prestained protein standards (see Fig. 3); lanes 2 and 3: total protein extracts of BR3-T and BR3-G; lanes 4 and 5: membrane proteins of BR3-T and BR3-G from detergent ( Triton X-114) phase; lanes 6 and 7: proteins of BR3-T and BR3-G from the aqueous phase. Method: S. citri membranes prepared as described in the legend to Fig. 3 were resuspended in 150 ml lysis buffer (50 mM Tris-HCl, pH 8.0, 1 mM PMSF, 10 mM EDTA) containing 1% Triton X-114, and then subjected to three cycles of TX-114 phase fractionation (Rosengarten and Wise, 1991). Proteins of the final detergent phase and the aqueous phase were separately precipitated with equal volumes of acetone at −20°C overnight, pelleted (10 000×g at 4°C for 15 min) and air-dried, and resuspended in 20 ml sample loading buffer (62.5 mM Tris-HCl, pH 6.8, 10% glycerol, 2% SDS and 0.001% bromophenol blue) for SDS-PAGE. Immunodetection using antiMBP-P58 antibody was conducted as described in the legend to Fig. 3.

This work was supported by the National Science Foundation (Grant EHR 9108771) and the Oklahoma Agricultural Experiment Station (Project 2052). Special thanks to Mr. Muralidhar Bandla for the preparation of antibodies and technical assistance, and to Dr. Steven White of the OSU Recombinant DNA/Protein Resource Facility for the synthesis of peptides. We also thank Dr. Yinghua Huang and Dr. Alejandro Penaloza-Vazquez for their kindness to review this paper prior to submission.

References Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403–410. Baseman, J.B., Cole, R.M., Krause, D.C. and Leith, D.K. (1982) Molecular basis for cytadsorption of Mycoplasma pneumoniae. J. Bacteriol. 151, 1514–1522. Dallo, S.F., Horton, J.R., Su, C.J. and Baseman, J.B. (1989) Homologous regions shared by adhesin genes of Mycoplasma pneumoniae and Mycoplasma genitalium. Microb. Pathogen. 6, 69–73. Davis, R.E. (1979) Spiroplasmas: Helical cell wall-free prokaryotes in diverse habitats. In: Proc. ROC-US Coop. Sci. Seminar on Mycoplasma Diseases of Plants, Natl. Sci. Council, Republic of China, Taiwan, pp. 59–64. Fletcher, J., Schultz, G.A., Davis, R.E., Eastman, C.E. and Goodman, R.M. (1981) Brittle root disease of horseradish: evidence for an etiological role of Spiroplasma citri. Phytopathology 71, 1073–1080. Fletcher, J., Wills, J.W. and Denman, S.E. (1989) Identification of surface proteins of Spiroplasma citri with protease and antibody probes. Curr. Microbiol. 19, 383–391. Fraser, C.M., Gocayne, J.D., White, O., Adams, M.D., Clayton, R.A., Fleischmann, R.D., Bult, C.J., Kerlavage, A.R., Sutton, G., Kelley, J.M., Fritchman, J.L., Weidman, J.F., Small, K.V., Sandusky, M., Fuhrmann, J., Nguyen, D., Utterback, T.R., Saudek, D.M., Phillips, C.A., Merrick, J.M., Tomb, J.F., Dougherty, B.A., Bott, K.F., Hu, P.C., Lucier, T.S., Peterson, S.N., Smith, H.O., Hutchison, C.A. and Venter, J.C. (1995) The minimal gene complement of Mycoplasma genitalium. Science 270, 397–403. Henrich, B., Feldmann, R.C. and Hadding, U. (1993) Cytoadhesins of Mycoplasma hominis. Infect. Immun. 61, 2945–2951. Kaloostian, G.H., Oldfield, G.N., Pierce, H.D. and Calavan, E.C. (1979) Spiroplasma citri and its transmission to citrus and other plants by leafhoppers. In: K. Maramorosch and K.F. Harris (Eds.), Leafhopper Vectors and Plant Disease Agents, Academic Press, New York, NY, pp. 447–450. Renaudin, J., Pascarel, M.C., Saillard, C., Chevalier, C. and Bove, J.-M. (1986) Chez les spiroplasmes le codon UGA n’est pas non sens et semble coder pour le tryptophane. C.R. Acad. Sci. Paris 303, 539–540. Rosengarten, R. and Wise, K.S. (1991) The Vlp system of Mycoplasma hyorhinis: Combinatorial expression of distinct size variant lipoproteins generating high-frequency surface antigenic variation. J. Bacteriol. 173, 4782–4793. Rost, B., Sander, C. and Schneider, R. (1994) PHD-An automatic mail server for protein secondary structure predication. Comput. Appl. Biosci. 10, 53–60. Rottem, S., Stein, O. and Razin, S. (1968) Reassembly of mycoplasma membranes disaggregated by detergents. Arch. Biochem. Biophys. 125, 46–56.

100

F. Ye et al. / Gene 189 (1997) 95–100

Saglio, P., Lafleche, D., Bonissol, C. and Bove, J.-M. (1971) Culture in vitro des mycoplasmes associe´s au stubborn des agrumes et leur observation au microscope e´lectronique. C.R. Acad. Sci. Paris Ser. D 272, 1387–1390. Smith, R.F. and Smith, T.F. (1992) Pattern-induced multi-sequence alignment (PIMA) algorithm employing secondary structure-dependent gap penalties for comparative protein modelling. Protein Engin. 5, 35–41. Sturrock, S.S. and Collins, J.F. (1993) MPsrch version 1.3. Biocomputing Research Unit, University of Edinburgh, UK. Tham, T.N., Ferris, S., Bahraoui, E., Canarelli, S., Montagnier, L. and Blanchard, A. (1994) Molecular characterization of the P1-like adhesin gene from Mycoplasma pirum. J. Bacteriol. 176, 781–788. Wayadande, A.C. and Fletcher, J. (1995) Transmission of Spiroplasma

citri lines and their ability to cross gut and salivary gland barriers within the leafhopper vector Circulifer tenellus. Phytopathology 85, 1256–1259. Wieslander, A., Boyer, M.J. and Wroblewski, H. (1992) Membrane protein structure. In: J. Maniloff, R.N. McElhaney, L.R. Finch and J.B. Baseman ( Eds.), Mycoplasmas: Molecular Biology and Pathogenesis. American Society for Microbiology, Washington, DC, pp. 93–112. Wise, K.S. (1993) Adaptive surface variation in mycoplasmas. Trends Microbiol. 1, 59–63. Ye, F., Melcher, U., Rascoe, J.E. and Fletcher, J. (1996) Extensive chromosome aberrations in Spiroplasma citri strain BR3. Biochem. Genet. 34, 269–286.