a molecular aspect of symbiotic interactions between the weevilsitophilus oryzaeand its...

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 239, 769–774 (1997)ARTICLE NO. RC977552

A Molecular Aspect of Symbiotic Interactions betweenthe Weevil Sitophilus oryzae and Its EndosymbioticBacteria: Over-expression of a Chaperonin

Hubert Charles, Abdelaziz Heddi,1 Josette Guillaud, Christianne Nardon, and Paul NardonLaboratoire de Biologie Appliquee, INSA 406, UA-INRA 203, SDI-CNRS 5128,20, Avenue Albert Einstein, 69621 Villeurbanne Cedex, France

Received September 5, 1997

ogy. They supply the weevil with several vitamins, in-Specific proteins of symbiosis were analyzed by the cluding riboflavin, pantothenic acid and biotin (2), they

comparison of two-dimensional electrophoresis pro- interfere with the amino acid metabolism (3), and theytein patterns of symbiotic and aposymbiotic strains of increase the mitochondrial oxidative phosphorylationthe weevil Sitophilus oryzae. One protein was shown (4). Hence, the SOPE are largely responsible for theto be exclusively expressed in the aposymbiotic strain destructive potential of this cereal pest weevil (5).and three proteins, including a chaperonin, were char-

The integration of the bacteria in the bacteriocyteacterized in the symbiotic strain pattern. The groE-physiology evolved probably from a long period of inter-like operon, encoding the two chaperonins groES andaction and coevolution between these prokaryotic andGroEL-like proteins of the endocytobiotes, was se-eukaryotic cells. The integration process involves notquenced. It was found to be very similar to the groEonly metabolic but also genetic interactions (6). As aoperon of Escherichia coli (82% identity). In vitro andresult, the SOPE have totally lost their ability to growex vivo experiments of protein labelling demonstratedin vitro and seem to have undergone some chromo-that almost 40% of the endocytobiote protein synthesissomal rearrangements (7). On the other hand, viableex vivo is focused on the GroEL-like protein. Finally,aposymbiotic strains of S. oryzae can be obtained, withwe showed by northern blotting that heat shock at

387C results in groEL mRNA accumulation inside the special care, by heat treatment of the symbiotic strainendocytobiotes. This work supports the hypothesis (8). Successive back crosses between symbiotic malesthat chaperonins could have an essential physiological and aposymbiotic females of S. oryzae lead, after a fewfunction in the maintenance of the symbiotic associa- generations, to a decrease in the fitness traits of thetion. q 1997 Academic Press weevil (6). The molecular mechanisms of this nucleocy-

Key Words: Sitophilus oryzae; symbiosis; two-dimen- toplasmic incompatibility are still unknown, but shouldsional gel electrophoresis; groE operon; chaperonin. be related to the expression of specific symbiotic genes.

The aim purposes of this work are the screening andthe characterization of gene product differences be-tween symbiotic and aposymbiotic strains of S. oryzae.

The weevil Sitophilus oryzae (Coleoptera, Curcu- We have first compared the two-dimensional electro-lionidae) is devoided of gut bacteria but harbors intra- phoresis protein patterns of these strains in order tocellular symbiotic bacteria (endocytobiotes) in special- detect proteins that are induced or inhibited by symbio-ized cells called bacteriocytes. These cells form an or- sis. We then focused our analysis on the endocytobioticgan (bacteriome) arranged as a packsaddle around the chaperonin (GroEL-like protein, [9, 10]), showing withmidgut of the larvae. The adult bacteriomes are located protein labelling experiments that it represents theat the apex of the mesenteric caeca and at the apex of most expressed protein by the SOPE. The genes encod-the ovaries. In the host cells, the bacteria lie free in ing the GroEL and GroES-like proteins were sequencedthe cytosol and are transmitted to the progeny through and their expression was shown to be heat inducible.the oocyte (1). S. oryzae principal endocytobiotes(SOPE) are very well integrated in their host physiol- EXPERIMENTAL PROCEDURES

Insects and bacteria. S. oryzae were reared on wheat at 27.57Cand 75% relative humidity (11) and endocytobiotes were isolated1 To whom correspondence should be addressed. Fax: 04 72 43 85

34. E-mail: [email protected]. using the procedure described in 12.

0006-291X/97 $25.00Copyright q 1997 by Academic PressAll rights of reproduction in any form reserved.

769

AID BBRC 7552 / 693c$$$541 10-21-97 10:11:55 bbrcg AP: BBRC

Vol. 239, No. 3, 1997 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

One-dimensional gel electrophoresis of proteins. Larvae were ex- in Figure 1A. Differences with the aposymbiotic straintracted from inside the grain and homogenized (0.1 mg/ml) in electro- are highlighted in Figure 1B. More than 10 differentphoresis buffer [100 mM TrisHCl (pHÅ6.5), 30% (v/v) glycerol, 4% (w/

batches of each of the symbiotic and the aposymbioticv) Sodium Dodecyl Sulfate (SDS), 100 mM b-mercaptoethanol, 0.05%samples were studied (coloured with Coomassie blue(w/v) bromophenol blue]. Electrophoresis was performed as described

by Laemmli (13) at 200 V using a 12.5% acrylamide slab gel. or silver stained) and the results were consistent. OnTwo-dimensional gel electrophoresis of proteins. Larvae were ho- a typical gel, over 150 spots could be identified. Four

mogenized in the insect physiologic medium Y (14) containing prote- main differences (a, b, g, d) were found between the twoase inhibitors [5.5 mM phenyl methyl sulfonyl fluoride (PMSF), 160 strains. The protein patterns of purified endocytobiotesmM ethylenediamine tetraacetic acid (EDTA), 30 mM pepstatine, 35

were analyzed as well (Figure 1C) and only the spotsmM antipain]. After a first centrifugation at 800g (3 min, 47C) pro-teins were precipitated overnight in 5 volumes of acetone at 0307C. g and d were found on the gels. In vivo labelling ofAfter a second centrifugation at 9,000g (10 min, 47C), the protein proteins was also performed by microinjecting 35S-me-pellet was lyophilized and stored at 0207C. Electrophoresis was then thionine in S. oryzae larvae and this confirmed theseperformed, according to the method described by O’Farrell (15) with

four differences (data not shown). Among these four50 to 100 ng of lyophilised proteins in 25 ml of solubilising solutionproteins, one has been immunologically identified (data[9 M urea, 1% (v/v) triton X-100, 5% (v/v) dithiothreitol, 1.6% (v/v)

ampholyte 5-7 and 0.4% (v/v) ampholyte 3-10 (LKB)]. not shown). This protein (d) belongs to the chaperoninDNA purification, PCR amplification, and sequencing of groE-like family (Hsp60) and it has been previously character-

coding region. Endocytobiote DNA was purified by proteinase K ized in three Sitophilus species (14). This protein isdigestion and phenol/chloroform extraction. The degenerated prim-

produced by the endocytobiotes and the correspondingers used for PCR amplification were MES2F: 5*ATGAATATTCGC-spot in the aposymbiotic strain (Figure 1B3) should beCATTRCAYGA3* and SEL2R: 5*CATCATNCCNCCCATNCCNCC3*.

Reaction cocktails for PCR amplification consisted of 1.5 U of Taq the mitochondrial Hsp60 protein. The co-chaperoninpolymerase (Appligene), 1.5 mM MgCl2, 0.2 mM deoxynucleoside (GroES-like protein) was not detected on bidimensionaltriphosphate, 0.6 mM primers and 100 ng of DNA template in a final electrophoresis gels but was detected in the symbioticvolume of 50 ml. The PCR parameters were 947C for 5 min followed

strain by Western blotting (data not shown).by 35 cycles of 947C for 45 sec, 557C for 45 sec and 727C for 45 sec(Perkin Elmer 2400). PCR with DNA extracted from the symbioticstrain of S. oryzae resulted in a product of the expected size (about Sequencing of the groE-like Coding Region of the2,000 pb) as determined by agarose gel electrophoresis, while DNA SOPEextracted from the aposymbiotic strain failed to yield PCR products.These results indicate that the products obtained from the symbiotic A 1,971 pb DNA fragment of the SOPE was amplifiedstrain originated from the SOPE. PCR product was directly cloned by PCR and sequenced on both strands (Figure 2). It(plasmid Sym III) in a pMOSBlue vector (Amersham) and double

was found that the groEL and groES-like genes consiststranded sequenced by the Genome express company (Grenoble) inan applied Biosystems apparatus. Two other subclones (Sym I and of 1,632 and 291 pb, suggesting that the GroEL andSym II) were also sequenced to control PCR misincorporations. The GroES-like monomers are composed of about 544 andgroE operon sequence of the SOPE was deposited in GenBank under 97 amino acids, respectively. Molecular weights de-accession number AF005236.

duced from the amino acid sequence (57.18 kDa for theEx vivo and in vitro protein labelling of the SOPE. For ex vivoGroEL and 10.36 kDa for the GroES-like proteins) were(with bacteriocyte) and in vitro (with bacteria) labelling of proteins,

dissected bacteriomes or purified endocytobiotes were incubated in consistent with molecular mass estimated from proteinthe physiologic insect medium Y (methionine free) containing 45 mCi/ electrophoresis gels. The amino acid sequence of theml of L-[35S]-methionine (1mCi/mmol, Amersham). After a one hour GroEL-like protein was aligned with that of other chap-incubation (277C), bacteriomes or endocytobiotes were homogenized

eronins found in GeneBank data base and phylogeneticin electrophoresis buffer. Proteins were then separated on a 12.5%analysis (data not shown) confirmed that the SOPEgel and detected by fluorography.

Northern blotting. The total RNA of S. oryzae larvae was extracted belong to the Enterobacteriaceae family. The groELwith the TRIzol reagent (GIBCO BRL). The RNAs (15 mg/well) were and groES-like genes were about 82 and 86% identicalseparated by electrophoresis on an 1.2% agarose gel and blotted on to the E. coli groEL and groES genes, respectively (93Hybond N/ (Amersham) nylon membranes. The groE probe used for

and 91% similarity were found at the amino acid se-hybridization was the Xba I - EcoR I fragment of the Sym III plasmid(2.0 kb) corresponding to the groE-like operon of the SOPE. Probes quence level). Figure 2 shows the amino acid sequenceswere labelled with 32P-dCTP (3 000 Ci/mmol, Amersham) by random of the GroES and GroEL-like proteins and the corre-priming (Nonaprimer kit, Appligene). Prehybridization and hybridiza- sponding amino acid substitutions in E. coli chaper-tion were performed as described in 16. Membranes were then washed

onin. More than 50% of amino acid substitutions werethree times (15 min) at 577C in 21, 11 and 0.51 SSC and exposed tofound to be conservative. Finally, a G/C content of 55%Hyperfilm MP (Amersham). The amount of RNA blotted for each sam-

ple was determined by film scanning (NIH-image software) and nor- was found for the amplified fragment that is consistentmalized by hybridization of the 16SrDNA probe of the SOPE obtained with the 55% of the genomic G/C content of the SOPEby PCR with eubacterial universal primers.

described in 17.RESULTS

In Vitro and ex Vivo Protein Labelling of SOPETwo-Dimensional Gel Electrophoresis of CytosolicProteins Endocytobiotes and bacteriocytes isolated from S. or-

yzae larvae, when incubated in Y-medium at 277C, in-The electrophoresis protein pattern of cytosolic pro-teins of the symbiotic strain of S. oryzae is presented corporate 35S-methionine actively. The comparison be-

770



FIG. 1. Two-dimensional electrophoresis protein pattern of the symbiotic strain of S. oryzae (A). Differences observed in the aposymbioticstrain are reported in the squares B1, B2 and B3. (C) protein pattern of isolated endocytobiotes.

tween the in vitro and the ex vivo protein profiles dem- observed: (1) the level of a 1.8 kb transcript was shownto increase with the heat shock treatment; (2) the 1.6onstrates that almost all the bacterial metabolism ex

vivo is focused on the production of the GroEL-like pro- kb mRNA level increased dramatically after 60 mincontinuous heat shock. It is noteworthy that the 2.1 kbtein (Figure 3). Densitometric analysis of protein bands

reveals that this protein represents more than 10 and mRNA level does not seem to be sensitive to the heatshock.40% of the total neosynthesised endocytobiote proteins,

in vitro and ex vivo respectively.

DISCUSSIONHeat Shock Induction of the groEL-like Gene

ExpressionSpecific proteins of symbiosis were analyzed by pro-

tein pattern comparison of symbiotic and aposymbioticNorthern analysis of RNA, extracted from S. oryzaelarvae without heat shock (at 277C), showed the exis- strains of S. oryzae and four proteins were character-

ized. The a protein (30 kDa, pHi Å 4.6) is specificallytence of two major transcripts of 1,600 and 2,100 bpthat hybridize with the groE probe (Figure 4, first lane). expressed in the symbiotic strain. It is not found on the

protein profiles of isolated endocytobiotes and thereforeAfter the heat shock (387C) two striking results were

771



FIG. 2. Amino acid sequences of the GroES and GroEL proteins of S. oryzae. Substituted amino acids in E. coli GroES and GroELproteins are reported above. Conservative substitutions calculated with MacMollyTetra software (31) are indicated with a star below thesequence. Bold amino acids are those that are conserved in most g-proteobacteria.

seems to be expressed by the host genome in response confirmed that the SOPE belong to the Enterobacteria-ceae and are very close to E. coli (93 and 91% aminoto the endocytobiote presence. The b protein (33 kDa,

pHi Å 6.1) is specific to the aposymbiotic strain. Its acids identity for the GroES and GroEL-like proteins,respectively). It is noteworthy that more than 50% ofexpression may be inhibited in the symbiotic strain,

probably by the endocytobiotes. These two non identi- E. coli amino acid substitutions are conservative in theGroEL-like sequence, indicating that the protein prob-fied proteins revealed the influence of endocytobiotes

on the host genome expression. However, we should ably conserves its chaperonin function inside the endo-cytobiotes.keep in mind that heat treatment of the symbiotic

strain (in order to obtain the aposymbiotic strain) Ex vivo labelling of proteins shows that the majorityof the endocytobiote protein neosynthesis is directedmight be responsible for the selection of different indi-

viduals among the symbiotic population. Other strains, towards the GroEL-like protein. As a comparison,GroEL represents about 1 and 10% of the neosynthe-from different regions, have to be tested to confirm

these results. sised proteins in E. coli when cultivated at 37 and 437C,respectively (19). This selective induction suggests thatThe two other proteins (g and d) are of bacterial ori-

gin, the g protein is not identified and the d protein intracellular living conditions could generate an im-portant stress for the symbiotic bacteria. Finally,is a chaperonin. ELISA experiments (data not shown)

revealed that this GroEL-like protein is not secreted Northern blotting experiments demonstrated clearlythat heat shock involves groEL-like mRNA accumula-in the insect haemolymph, as has been described pre-

viously in the pea aphid symbiosis (18). The co-chaper- tion inside the symbiotic strain of S. oryzae. The groEoperon of the SOPE could be regulated by several pro-onin (GroES-like) was also immunodetected in the

SOPE and the genes encoding both proteins were se- moters (leading to the accumulation of several mRNA)or could be post-transcriptionally regulated by RNAquenced. The groES and groEL-like genes are arranged

in an operon very similar to the E. coli groE operon. processing. Present results do not enable us to drawany conclusion about this regulation. Further experi-Phylogenetic analysis based on the two gene sequences

772



ments are currently underway to elucidate the molecu-lar regulation of this gene in SOPE.

Chaperonins are stress proteins constitutively ex-pressed but also heat inducible. These proteins wereoriginally recognized as factors required for bacterio-phage assembly in Escherichia coli (9). They are nowknown to be involved in the folding of nascent polypep-tides, the assembly of oligomeric protein complexes andprotein export from bacteria (10,19). Over-productionof chaperonin was described in parasitic bacteria, suchas Salmonella typhi (20), Chlamydia trachomatis (21) FIG. 4. Northern blot experiments. mRNA that hybridize withor Mycobacterium paratuberculosis (22). It has been the groEL-like probe are indicated by their size (2.3, 1.8 and 1.6 kb).supposed that GroEL-like proteins are required for the Heat shocks were performed from 0 to 120 min. Membranes were

also hybridized with a 16S rDNA probe in order to normalize theparasite survival inside the hostile host cytosol byamount of transcripts. The blots were exposed for 24 h and 6 h forchaperoning essential bacterial proteins or virulencegroEL-like and 16S rDNA probes, respectively.factors (19, 23). In long term associations (i.e., symbi-

otic associations), overproduction of chaperonin wasalso described (24, 25, 26). In the amoeba symbiosis, it

used the term symbionin to qualify the GroEL-like pro-has been shown that a specific promoter located insidetein produced by the primary endocytobiotes of the peathe groES-like sequence is responsible for the groEL-aphid. The accuracy of this term is now highlighted bylike gene expression (27). In the aphid symbiosis, athe new results obtained in the weevil and in otherPs32 heat shock promoter is located in the upstreamendocytobiosis models, such as the amoebae (30) andregion of the groE operon, but the groEL-like gene doesthe tsetse fly (26).not seem to be heat inducible (28). In both models, the

physiological function of these proteins is not known.ACKNOWLEDGMENTSWhatever the situation, what remains interesting for

the symbiologists to understand is why a chaperonin isWe thank Chaque Khatchadourian for her technical assistanceselectively produced by the endocytobiotes while other and Valerie James for correcting the English.

stress proteins, like Hsp70, are not. This preferentialinduction differentiates the stress induced by tempera- REFERENCESture from the stress due to the intracellular conditions.Moreover, it could reveal an important and specific 1. Nardon, P., and Wicker, C. (1981) Ann. Biol. 4, 329–373.function for the chaperonin in the regulation of the 2. Wicker, C. (1983) Comp. Biochem. physiol. 76A, 177–182.symbiotic association. Fifteen years ago, Ishikawa (29) 3. Gasnier-Fauchet, F., and Nardon, P. (1987) Insect Biochem. 17,

17–20.4. Heddi, A., Lefebvre, F., and Nardon, P. (1993) Insect Biochem.

Mol. Biol. 23, 403–411.5. Grenier, A. M., Nardon, P., and Bonnot, G. (1986) Acta Oecolog-

ica 7, 93–110.6. Nardon, P., and Grenier, A. M. (1988) in Cell to Cell Signals in

Plant, Animal and Microbial Symbiosis (Scannerini, S., Ed.), pp.255–270, NATO ASI Series H17 Springer-Verlag, Berlin.

7. Charles, H., Condemine, G., Nardon, C., and Nardon, P. (1997)Insect Biochem. Mol. Biol. 27, 345–350.

8. Nardon, P. (1973) C. R. Acad. Sci. Paris 277D, 981–984.9. Georgopoulos, C. P., Hendrix, R. W., Casjens, S. R., and Kaiser,

A. D. (1973) J. Mol. Biol. 76, 45–60.10. Gatenby, A. A., and Viitanen, P. V. (1994) Annu. Rev. Plant Phys-

iol. Plant Mol. Biol. 45, 469–491.11. Laviolette, P., and Nardon, P. (1963) Bull. Biol. Fr. Belg. 97,

305–33.12. Heddi, A., Lefebvre, F., and Nardon, P. (1991) Endocytobiosis

and Cell. Res. 8, 61–73.13. Laemmli, U. K. (1970) Nature, Lond. 227, 680–685.14. Charles, H., Ishikawa, H., and Nardon, P. (1995) C. R. Acad. Sci

Paris 318, 35–41.15. O’Farrel, P. H. (1975) J. Bio. Chem. 250, 4007–4021.FIG. 3. (A) In vitro and (B) ex vivo (i.e., endocytobiotes inside

the isolated bacteriocytes) labelling of proteins. The position of the 16. Heddi, A., Lestienne, P., Wallace, D. C., and Stepien, G. (1993)J. Biol. Chem. 268, 12156–12163.GroEL-like protein is indicated with an arrow head.

773



17. Dasch, G. A., Weiss, E., and Chang, K. P. (1989) in Bergey’s Man- 24. Jeon, K. W. (1983) Int. rev. Cytol. 14, 29–47.ual of Systematic Bacteriology, pp. 811–833, Williams and Wil- 25. Ishikawa, H. (1989) in Insect Endocytobiosis: Morphology, Physi-kins, Baltimore, MD. ology, Genetics, Evolution (Schwemmler, W., and Gassner, G.,

Eds.), pp. 123–143, CRC Press, Washington.18. Van den Heuvel, J. M., Verbeek, M., and Van der Wilk, F. (1994)J. Gen. Vir. 75, 2559–2565. 26. Aksoy, S. (1995) Insect Mol. Biol. 4, 23–29.

19. Van der Vies, S., and Georgopoulos, C. (1996) in The Chaperonin, 27. Ahn, T. I., Lim, S. T., Leeu, H. K., Lee, J. E., and Jeon, K. W.pp. 137–165. Academic Press, New York. (1994) Gene 128, 43–49.

20. Lindler, L. E., and Hayes, J. M. (1994) Microb. Pathology 17, 28. Sato, S., and Ishikawa, H. (1997) J. Bact. 179, 2300–2304.271–275. 29. Ishikawa, H. (1982) Insect Biochem. 12, 613–622.

21. Ho, Y., and Zhang, Y. X. (1994) Gene 141, 143–144. 30. Jeon, K. W. (1995) Trends Cell. Biol. 5, 137–140.22. Colston, A., Mcconnell, I., and Bujdoso, R. (1994) Microbiology- 31. Vahrson, U., and Wittig, U. (1994) Analysis and Interpretation of

UK 140, 3329–3336. DNA and Protein Sequence using MacMollyTetra, Karoi-Verlag,Berlin.23. Kaufmann, S. H. E. (1992) Experimenta 48, 640–643.

774


a molecular aspect of symbiotic interactions between the weevilsitophilus oryzaeand its...

Documents