Eur. J. Biochem. 241, 17-24 (1996) 0 FEBS 1996

Cloning, purification and characterization of the protein subunit of ribonuclease P from the cyanobacterium Synechocystis sp. PCC 6803 Alberto PASCUAL and Agustin VIOQUE

lnstituto de Bioquimica Vegetal y Fotosintesis, Facultad de Biologid, Universidad de Sevilla-CSIC, Sevilla, Spain

(Received 22 May 1996) - EJB 96 0751/2

The rnpA gene from the cyanobacterium Synechocystis sp. PCC 6803, which codes for the protein subunit of ribonuclease P (RNase P), has been cloned by functional complementation of an Escherichia coli mutant. This protein had previously been characterized only in proteobacteria and gram-positive bacteria. rnpA and the closely linked rpmH gene, which code for the large subunit ribosomal protein L34, have been sequenced. The Syneclzocystis 6803 L34 protein is more similar to the homologous protein from some non-green chloroplasts than to the L34 protein from other bacteria. The protein subunit of RNase P from Synechocystis 6803 has been overexpressed in E. coli and purified to homogeneity. Anti- bodies raised against the Synechocystis 6803 RNase P protein did not recognize the homologous protein from E. coli (C5 protein). Similarly, antibodies raised against the E. coli C5 protein did not recognize significantly the Synechocystis 6803 protein. In spite of the lack of immunological cross-reactivity and the low level of sequence identity, the E. coli and Synecfzocystis 6803 proteins are functionally inter- changeable. In enzymatic assays using either an E. coli precursor tRNATy' or a Synechocystis 6803 precur- sor tRNAG'" as substrates, we have detected RNase P activity with holoenzymes reconstituted with the RNA subunit from E. coli and the protein subunit from Synechocystis 6803 or with the RNA subunit from Synechocystis 6803 and the protein subunit from E. coli. The relative efficiency of cleavage of the different substrates is dependent on the origin of the protein subunit used to reconstitute the holoenzyme.

Keywords: ribonuclease P ; Synechocystis 6803 ; protein-RNA interaction; precursor tRNA processing ; cyanobacteria.

Ribonuclease P (RNase P) is an endonuclease responsible for the generation of the mature 5' end of tRNAs from precur- sors of tRNAs (Altman, 1989; Pace and Smith, 1990). Every cell and every cellular compartment that synthesizes tRNAs has been shown to contain an RNase P activity. The enzyme is com- posed of RNA and protein in most of the RNaseP activities investigated so far. In many eubacteria, the RNA component of the enzyme has been shown to be catalytically active in vitro in the absence of the protein subunit (Guerrier-Takada et al., 1983) under appropriate buffer conditions. In vivo, the protein subunit is essential. It has been proposed that the protein functions as an electrostatic shield (Reich et al., 1988) and that it also facilitates product release (Tallsjo and Kirsebom, 1993). The catalytic RNA subunit has been characterized in a few strains of cyano- bacteria (Banta et al., 1992; Vioque, 1992; Pascual and Vioque,

Correspondence to A. Vioque, Instituto de Bioquimica Vegetal y Fotosintesis, Facultad de Biologia, Universidad de Sevilla-CSIC, Apartado 11 13, E-41080 Sevilla, Spain

Fax: +34 5 4620154. E-mail: vioque@cica.es Abbreviations. IPTG, isopropyl thio-P-~-galactoside; C5 protein,

protein subunit of Escherichia coli RNase P; M1 RNA, RNA subunit of E. coli RNase P ; pre-tRNA, precursor tRNA; RNase P, ribonuclease P; Synechocystis RNase P RNA, RNA subunit of Synechocystis 6803 RNase P; Synechocystis RNase P protein, protein subunit of Synecho- cysfis 6803 RNase P.

Enzyme. Ribonuclease P (EC 3.1.26.5). Note. The novel nucleotide sequence data published here have been

deposited with the EMBL sequence data bank and are available under accession number X8198Y.

1994). However, the protein subunit has not been studied in cyanobacteria.

In some instances it has been possible to isolate the rnpA gene by functional complementation of an Escherichiu coli mutant. A thermosensitive mutation has been isolated in the rnpA gene (Schedl and Primakoff, 1973). The mutant (rnpA49) has a single amino acid substitution in the RNase P amino acid sequence (Kirsebom et al., 1988). The effect of the mutation (Arg46-His46) is not clear but it has been suggested that the mutation affects the assembly of the holoenzyme at the re- strictive temperature (Baer et al., 1989). This idea is supported by the fact that overexpression of the RNA subunit (Motamedi et al., 1984) or of the mutant protein subunit itself (Vioque et al., 1988) can suppress the thermosensitive phenotype. Expres- sion in an E.coli rnpA49 strain of heterologous RNase P proteins can also rescue the mutant phenotype. Thus, the mutant rnpA49 strain can be used for the screening of libraries and the isolation of heterologous rnpA genes. This strategy has been successful for the cloning of Streptomyces RNase P protein (Morse and Schmidt, 1992).

In spite of their low sequence similarity, it is clear that the diverse bacterial RNase P proteins are functionally equivalent and can probably adapt a similar three-dimensional structure. This statement is supported by the fact that it is possible to reconstitute an heterologous enzyme in many instances in vitro: the RNA subunit from Salmonella typhimurium, Thermus uqua- ticus, and Thermotogu maritimu can form a functional holoen- zyme with the E. coli protein subunit (Baer and Altman, 1985; Brown et al., 1993). The protein and RNA subunits of E. coli

18 Pascual and Vioque (EM J. Biochem. 241)

and B. subtilis can also be exchanged to form heterologous func- tional enzymes (Guerrier-Takada et al., 1983).

In this work, we characterize the RNase P protein from a cyanobacterium and show, using either an E. coli or a Synecho- cystis 6803 precursor tRNA, that a functional holoenzyme can be reconstituted by mixing E. coli and Synechocystis 6803 sub- units, despite their low sequence similarity and lack of inmuno- logical cross-reactivity.

MATERIALS AND METHODS

Library construction and complementation assays. Cells of Synechocystis sp. PCC 6803 were grown in medium BGll (Rippka et al., 1979j, and total DNA was isolated as described before (Cai and Wolk, 1990). The DNA was partially digested with HpuII and fractionated by centrifugation at lO00OOg for 21 h through a sucrose gradient (Sambrook et al., 1989). The fractions containing fragments of 4-6 kb were pooled and li- gated to pBluescript SK+ (Stratagene) that had been digested with ClaI and dephosphorylated. Plasmid isolation from E. coli, restriction endonuclease digestion, and agarose gel electrophore- sis were carried out by standard methods (Sambrook et al., 1989). The library was used to transform E. coli strain NHY322 (Kirsebom et al., 1988). This strain is thermosensitive because it carries the rnpA49 mutation (Schedl and Primakoff, 1973). The transformants were plated on ampicillin-containing Luria- Bertani agar plates and incubated overnight at 43 "C. Clones able to grow at 43 "C were selected for further analysis.

DNA sequencing. Fragments from the recombinant plas- mids that contained the DNA of interest were subcloned in pBluescript SK+ (US Biochemicals) and both strands were sequenced using the M13IpUC sequencing primer and the M13 reverse sequencing primer (Boehringer). DNA sequencing was performed with double-stranded DNA using the dideoxynucleo- tide chain-termination method with Sequenase version 2.0 (US Biochemicals) and ["S]dATP[uS]. When necessary, a set of nested deletions was generated with exonuclease I11 and nuclease S1. A kit from Pharmacia was used for this purpose.

Sequence analysis was performed using the UWGCG soft- ware package (Devereux et al., 1984). Protein sequences were aligned using Clustal W (Thompson et al., 1994).

Protein overexpression and purification. A BslI fragment extending from the beginning of the Synechocystis 6803 rnpA gene to 15 nucleotides after the termination codon was purified, and its ends were made blunt with the exonuclease activity of Klenow enzyme. The fragment was cloned into the NdeI site of pT7-3a (Rosenberg et al., 1987), which was made blunt with the polymerase activity of Klenow, to generate plasmid pARA2. The expected protein synthesized when the T7 promoter i n pARA2 is induced corresponds exactly to the protein subunit of Synechoc.ystis 6803 RNase P (Synechocystis RNase P protein).

To overexpress the RNase P protein, plasmid pARA2 was introduced by transformation in E. coli strain BL21(DE3). The transformation mix was used to directly inoculate 1 1 M9 mini- mal medium supplemented with glucose (2 mg/ml), thiamine ( 1 pg/ml), and ampicillin (100 pg/rnl). The culture was incu- bated until the absorbance at 550 nm was 0.5 ; isopropyl thio-p- D-galactoside (IPTG) was added to a final concentration of 2 mM. Cells were harvested 3 h after induction with IPTG.

The initial steps of the Synechocystis RNase P protein purifi- cation were similar to those previously used for the purification of E. coli C5 protein (Vioque et al., 1988). A 30000 g superna- tant (S30j was prepared from l l of cells by grinding with alu- mina and treatment with deoxyribonuclease I in buffer A

[50 mM Tris-HCI, pH 7.5, 60 mM NH,CI, 10 mM Mg(Ac)J as described (Robertson et al., 1972) and centrifugation at 30000 g. The S30 fraction was centrifuged for 2 h at I00000 g in a Beck- man Ti50 rotor. The supernatant was removed and the pellet was resuspended in 1 ml buffer A containing 1 M NH,CI. The suspension was gently agitated for 2 h at 4°C and centrifuged again at 100000g. The supernatant (ribosomal wash) was ap- plied to an 80-ml Sephadex G-50 column (Pharmacia) equili- brated in the same buffer. The fractions containing the RNase P protein were pooled and applied to a 1 -ml nickel chelate column (His-Bind Resin, Novagen) equilibrated in the same buffer. The columnn was washed with 20 ml of the same buffer, 20 ml of the same buffer containing in addition 7 M urea, and the protein was eluted in the same buffer containing 7 M urea and 60 mM imidazole.

E. coli CS protein was purified as described previously (Vioque et al., 1988).

Preparation of RNase P RNAs. The RNA subunit of E. coli RNase P (M1 RNA) was prepared by in vitro run-off transcription of plasmid pJA2 digested with FokI as described (Vioque et al., 1988). To generate the RNA subunit of Synecho- cystis 6803 RNase P (Synechocystis RNase P RNA), plasmid pT76803 was constructed containing the coding sequence of the Synechocystis 6803 rnpB gene under the control of the T7 pro- moter. The cloned Synechocystis 6803 rnpB gene (Vioque, 1992) was amplified by PCR with a forward primer (5'-GGAAT- TCTAATACGACTCACTATAGAGAGTTAGGGAGG-3') that contains an EcoRI site, the T7 promoter and that overlaps the 5' end of the coding sequence and a reverse primer (5'- CCCAAGCTTTAAAAAAAGAGAGTTAGTC-3') that contains a DraI site overlapping the 3' end of the coding sequence and a Hind111 site. A PCR fragment of the expected size was obtained, purified, digested with EcoRI and HindIII and ligated into pUC19 that had been treated with the same enzymes. The con- struction was confirmed by sequencing. The plasmid obtained after digestion with DruI, was used for in virrn run-off transcrip- tion with T7 RNA polymerase. The RNA obtained would con- tain the same ends and sequence as the Synechocystis RNase P RNA.

The RNase P RNAs were resuspended in assay buffer and renatured by incubation at 65°C for 5 min and slowly cooled to room temperature.

Preparation of precursor tRNAs. Labelled E. coli pre- tRNATy' (Altman and Smith, 1971) was prepared by in vitro run-off transcription with T7 RNA polymerase in the presence of [(x-"P]GTP as described (Vioque et al., 1988).

Plasmid pT7Gln was prepared for the synthesis of Synecho- cystis 6803 pre-tRNAC"" by PCR amplification from genomic DNA with primers based on the published sequence of the Synechocystis 6803 trnQ gene (Mayes et al., 1993). The forward primer (5'-CGACGGGATCCTAAATACGACTCACTATAGGT- TAATCAATGGGGTG-3') contains a BamHI site, the T7 pro- moter sequence, and overlaps the 5' end of the pre-tRNA sequence. The reverse primer (S-GTCCCAAGCTTGGATG- GACGCCTGGCTGGGGTGCTAGGATTCG-3') overlaps the 3' end of the pre-tRNA sequence, and contains a FokI, il BstNI, and a HindIII site. A PCR product of the expected size was purified, digested with HindIII and BamHI, and ligated into pUC19 treated with the same enzymes. The clone was confirmed by sequencing. Run-off transcription of FokI-digested plasmid with T7 RNA polymerase would generate a pre-tRNA with the expected sequence of the endogenous pre-tRNA"'" and lacking the 3'-end CCA sequence. In the case of .Synechocystis 6803 tRNA"'", the CCA sequence at the 3' end of the tRNA is not encoded in the gene and must be added post-transcriptionally (Mayes et al., 1993). Fig. 1 shows the predicted secondary struc-

Pascual and Vioque ( E m J. Biochem. 241) 19

A 0' 'A C A C - 0 C - 0 U-0 U-A . .. C-0 C

ppp0-CAOOCCAOUAAAAAaCAuUACCCC0-C

A:: R N a s e P a-c U - A

B A C U A R N a s e P

0 - C "-P

4 0 pppOOUUAAUCAAU-A

0 -C 0 - C

a-c

a-c

c -a a-c

A - U a-c

u *

:-U

U A

0 - C a-c

u u u 0

U O

E. cvii Synechocystis ptRNATYr ptRNAGln

Fig. 1. Nucleotide sequence and secondary structure models of E. coli pre-tRNAnr (A) and Synechocystis 6803 pre-tRNAG1" (B) tran- scripts.

tures of E. coli pre-tRNAT3" and Synechocystis 6803 pre- tRNAG1".

RNase P assays. Holoenzyme was reconstituted by direct mixing of the renatured RNA subunit with a tenfold molar ex- cess of the protein subunit in assay buffer [I0 mM Hepes, pH7.5, 10 mM Mg(OAc),, 400mM NH,OAc, and 0.01% Nonidet P-401 (Talbot and Altman, 1994) and incubation for 5 min at 37°C immediately before the assays. Assays were per- formed as described (Vioque et al., 1988) by incubating at 37°C an aliquot of the reconstituted holoenzyme with 100nM IZP- labeled substrate in assay buffer. Under these conditions, the RNA subunits have no detectable enzymatic activity in the ab- sence of protein. The amount of enzyme and time of reaction was adjusted so that the percentage of cleavage obtained was in the linear range of the kinetics of the reaction. Reaction products were analyzed by electrophoresis on 8 % acrylamide/7 M urea gels and autoradiography. Quantitative analysis of enzymatic reactions was carried out with an InstantImager (Packard).

Preparation of antibodies. Synechocystis RNase P protein purified up to the ribosomal wash step was dialyzed against buffer A. A precipitate containing the RNase P protein was collected by centrifugation and resuspended in SDS electropho- resis buffer. The sample was applied to a preparative SDS/ acrylamide gel and the band corresponding to the RNase P pro- tein was cut from the gel and crushed with a glass rod and used directly for immunization.

Antibodies were prepared in one New Zealand White male rabbit. The rabbit was immunized at four subcutaneous sites with a total of 1 mg protein in 500 p1 incomplete Freund's adju- vant. At three weeks after the initial immunization, the rabbit was given a subcutaneous booster injection of 0.5 mg protein in incomplete Freund's adjuvant. The rabbit was bled 15 days after the last injection. The serum was heated 10 min at 55"C, clari- fied by centrifugation, and used directly for Western blots. For immunoprecipitation, the antibodies were further purified on a Hi-Trap protein-A-Sepharose FPLC column (Pharmacia) as specified by the manufacturer.

Antibodies against C5 protein were a generous gift of Drs. Paul Eder and Sidney Altman. In both cases, a preimmune serum was obtained and treated in exactly the same way.

Western blot analysis. Total E. coli proteins from cells overexpressing either the E. coli C5 protein or the Syneclzocystis RNase P protein were prepared by resuspending a cell pellet directly in SDS loading buffer. Total Synechocystis 6803 or

Anabaena sp. PCC 7120 proteins were prepared by sonication of cell suspensions and extraction of pigments with methanol. The protein samples were electrophoresed in SDS/12 % polyacrylamide minigels (Bio-Rad) and transferred to nitro- cellulose by electroblotting. The nitrocellulose filter was blocked overnight with 2% nonfat milk in Tris/NaCl (15 mM Tris/HCI, pH 7.5, 200 mM NaC1) and incubated with rabbit antiLC5 serum or rabbit anti-Synechocystis RNase P protein serum for 1 h and developed with goat alkaline-phosphatase-conjugated anti-rabbit antibodies (Sigma) as described (Sambrook et al., 1989).

Depletion of RNase P activity by immunoprecipitation. Immunodepletion was performed essentially as described (Mamula et al., 1989). FPLC-purified antibodies were incubated with protein-A-Sepharose 4B (Sigma) overnight at 4°C with gentle agitation and the beads were washed three times in assay buffer. Identical amounts of RNase P holoenzyme reconstituted from E. coli or Synechocystis 6803 subunits were incubated with either preimmune, an t i43 serum or anti-Synechocystis RNase P protein serum at 4°C with gentle agitation and centrifuged 1 min. An aliquot of the supernatant was used for RNase P assays.

RESULTS Cloning of the Synechocystis 6803 rnpA gene. Transformation of the thermosensitive E. coli strain NHY322 with the Synecho- cystis 6803 plasmid library resulted in the selection of a clone that was able to grow at 43°C. The selected clone contained a plasmid with a 6.6-kb insert of Synechocystis 6803 genomic DNA. The origin of the insert was confirmed by Southern analy- sis (data not shown). A 1.4-kb HindIII-Xbal fragment derived from the original 6.6-kb fragment, was still able to complement the mutation in NHY322. This fragment was sequenced (Fig. 2) and two ORFs were identified with homology to bacterial and chloroplast ribosomal protein L34 and to the protein subunit of RNase P, respectively. Deletions generated with exonuclease 111 that truncated the second ORF suppressed the ability to comple- ment the mutation in NHY322. Thus, we conclude that this ORF corresponds to the Synechocystis 6803 rnpA gene and the first ORF corresponds to the rpmH gene. The rnpA coding sequence starts with a GTG codon only nine nucleotides downstream from the end of the rpmH gene. A GTG initiation codon seem to be a general feature of the rnpA gene in eubacteria. A third ORF partially overlapping the rnpA gene was also identified (ORFI, Fig. 2). ORFI shows no homology with proteins in the databases and is unrelated to the 7-kDa protein that is coded in other eubacteria by an ORF overlapping the rnpA gene.

Three sequences similar to the dnaA box, 5'-TTAT(C/ A)CA(C/A)A-3' (Yoshikawa and Ogasawara, 1991), were iden- tified upstream of the rpmH gene.

Sequence of the Synechocystis RNase P protein. The Synecho- cystis RNase P protein is only 20-30% identical to RNase P proteins of other eubacteria. The known bacterial RNase P pro- teins do not have many similarities, except for their size and positive charge. The central portion of the protein is the only highly conserved region and is very rich in basic residues (Fig. 3A). It has been suggested that this region may represent an RNA-binding domain (Morse and Schmidt, 1992). This re- gion of the Synechocystis 6803 protein conserves the almost in- variant RNRLKR motif. Despite the low sequence conservation, there is some resemblance between the hydrophobicity profiles of E. coli and Synechocystis 6803 proteins except around the amino and carboxy ends (Fig. 3B).

Purification of Synechocystis RNase P protein. The Synecho- cystis RNase P protein was cloned under the control of the T7

20 Pascual and Vioque ( E m J. Biochem. 241)

- ? U T T ~ ~ T ~ T ~ ~ ~ ~ ~ T A T ~ ~ ~ ~ C C ~ M ~ ~ f f i R ~ ~ T - ~ ~ A T 1 170 K U I X H P R C I C ~ W C D I V V T L W D ~ ~ R L C L R A U

~ ~ M ~ R D C O O E M C r r T A ~ K T A T A ~ f f i M C C f f i ~ ~ ~ K ~ ~ ~ ~ C C C C R ~ ~ T A ~ K C C ~ R f f i 1260 P N F R C S L '

ma1 ~ M ~ T M ~ ~ T B T A ~ M T C ~ K ~ R ~ C C C C ~ ~ ~ ~ ~ ~ ~ ~ M T A R T A r r T ~ 1348

Fig. 2. Nucleotide sequence of the Synechocystis 6803 rpmH and rnpA regions. The deduced amino acid sequences are given below the nucleotide sequence. The HindIII-Xhal fragment can complement the E. mli mnpA49 mutation in NHY322 as well as an exunuclease-111-generated fragment extending from the Hind111 site to position 952 (indicated by V). A fragment extending from the Hind111 site to position 818 (indicated by Y) fails to complement. Possible ribosome-binding sites upstream of the initiation codons of protein L34 and ORFl are underlined. Arrows indicate the position and orientation of possible dnaA boxes.

promoter and overexpressed in E. coli. The overexpressed pro- tein was purified up to the ribosomal wash step using the same procedure described for the purification of E coli C5 protein (Vioque et al., 1988). The next step would normally involve a precipitation by dialysis against a low salt buffer. However, this step cannot be used with the Synechocystis 6803 protein be- cause, in contrast to the E. c-oli protein, it precipitates irrevers- ibly and cannot be brought into solution even in the presence of 7 M urea. The protein was successfully purified by sequential gel filtration and chromatography on a nickel chelating column (Fig. 4). Proteins containing a stretch of several histidine resi- dues bind strongly to the nickel chelate column. This property has been used to tag proteins for easy and fast purification (Schmitt et al., 1993). The Synechocystis RNase P protein binds to the His-Bind resin in the absence of a tag with moderate affinity and is eluted at low imidazole concentration (60 mM), but the binding is specific enough to allow its successful purifi- cation. It is possible that this protein has affinity for the nickel chelating column through the C-terminal region, which has three closely spaced histidines (DHGHSRNHLL). About 0.5 mg pure active protein is obtained from 1 1 of induced culture. There ex- ists the possibility that the E. coli C5 protein copurifies with the Syneclzocy.stis 6803 protein. However, the endogenous levels of CS protein are very low (Vioque et al., 1988) and, in addition, C5 protein is not expected to bind to the nickel chelate column because it contains only four histidine residues and these resi- dues are not close to each other. We have attempted to detect contamination of our pure Synechocystis RNase P protein with the E. coli C5 protein by Western blot analysis with antibodies specific for the E. coli protein (see below). No contamination was detected using these antibodies within a resolution limit of 0.5% (data not shown).

The RNase P protein can also be purified by hydrophobic chromatography (data not shown). The protein binds strongly to phenyl-Sepharose and elutes only in the presence of 7 M urea. This indicates that the protein, as predicted from its sequence, has a highly hydrophobic behavior that could explain the irreversible precipitation observed in low salt buffer.

Enzymatic activity of the purified protein. The function of the Synechocystis RNase P protein was analyzed by the reconstitu- tion of the holoenzyme either with the E. coli RNase P RNA (MI RNA) or with the Synechocystis RNase P RNA. When E. coli pre-tRNAQ' was used as substrate, it was cleaved by the Synechocystis 6803 holoenzyme (Fig. 5A, lane 4) as well as by holoenzyme reconstituted from E. coli M I RNA and Synecho- cystis RNase P protein (Fig. S A , lane 3) or from Synechocystbv RNase P RNA and C5 protein (Fig. SA, lane 5).

The same result was obtained with a pre-tRNA substrate from Synechocystis 6803 (Fig. 5 B). The Synechocystis 6803 pre- tRNA"'" was synthesized, based on the published gene sequence, with a 10-bp 5' leader sequence and lacking the 3'end CCA, which is not encoded in the gene. This previously uncharac- terized substrate was cleaved at the expected position by endo- nucleolytic action. The position of the cleavage site was con- firmed by analysis of the cleavage products on a sequencing gel and by primer extension (data not shown).

The relative efficiency of cleavage of the two pre-tRNA sub- strates used here was dependent on the source of the protein subunit used to reconstitute the holoenzyme. For instance, the Synechocwis RNase P RNA cleaved E. coli pre-tRNA"' better in the presence of the E. coli C5 protein than in the presence of the Synechocystis RNase P protein (Fig. 5A, lanes 4 and 5), but the opposite was observed with the S.ynechocysti.s 6803 pre-

Pascual and Vioque (Eur: J . Biochem. 241) 21

A Streptomyces b i k i n i e n s i s P T R A G F V V S - X A V G - G A V V R N Q V K R R L R H L V C D R L S A L P P - Streptomyces c o e l i c o l o r R T R A G F V V S - K A V G - V A V V R N K V K R R L R H L M R D R I D L L P P - Micrococcus l u t e u s R P R A G F V V S - K A V G - N A V T R N R V K R R L R A V V A E - Q M R L P P L Hycobacter im 2eprae A P H V G L I I A - K T V G - S A V E R H R V A R R L R H V A R T M L G E L G G - Baci l lus s u b t i l i s - L R V G L S V S - K K I G - N A V M R N R I K R L I R Q A F L E E K E R L K E - Hycoplasmd capricolum Y L K Y G I S V G - K K I G - N A V I R N K V K R Q I R M I L K Q N I S E I G T - nycoplasma genf ta l fum T W R V A I S I A - K T K Y K L A V Q R N L I K R Q I R S I F Q Q I S N N L E P - Eschericia col i - - R I G L T V A - K K N V R R A H E R N R I K R L T R E S F R L R Q H E L P A - HaentoPhilus influenzae - - R L G L T V A - K K H L K R A H E R N R I K R L V R E S F R L S Q H R L P A - Proteus mtrkbilis H P R I G L T I A - K K N V K R A H E R N R I K R L A R E Y F R L H Q H Q L P A - Pseudomonas putida H P R L G L V I G - K K S V K L A V Q R N R L K R L M R D S F R L N Q Q L L A G - Buchnera aphidicola H P R L G L S I S - R K N I K H A Y R R N K I K R L I R E T F R L L Q H R L I S - Coxie l la burneti i H S R L G V V A S - K R N V R K A V W R N R V R R V V K E A F R I R K K D L P A -

- - R F G I T V S Q K - V S K K A T V R N R L K R Q I R A V I N H F Q P Q I K P - Synechocystis 6803 ' * * * * I * I

3 i

-3 I

I I 50 100

Amino acid residue number Fig. 3. Comparison of the RNase P amino acid sequences from eubacteria. (A) Alignment of the central conserved region of the known bacterial RNase P protein subunits. The sequences used were from Strepromyces bikiniensis (Morse and Schmidt, 1992), Streptomyces coelicolor (Calcutt and Schmidt, 1992), Micrococcus luteus (Fujita et al., 1990), Mycobacterium leprae (database accession number L39923), Bacillus subtilis (Ogasawara and Yoshikawa, 1992), Mycoplasnza cupricolum (Miyata et al., 1993), Mycoplasma genitalium (Fraser et al., 1995), Escherichia coli (Hansen et al., 1985), Haemophilus influenzae (Fleischmann et al., 1995). Proteus mirubilzs (Skovgaard, 1990), Pseudomonas putida (Ogasawara and Yoshikawa, 1992), Bwhnera aphidicola (Lai and Baumann, 1992), Coxiella burnetii (Suhan et al., 1994), and Synechocystis 6803 (this work). Positions where a positively charged residue is present in most of the proteins are indicated *. Highly conserved hydrophobic residues are also shown (#). Proteins were aligned using Clustal W. (B) Hydrophobicity plot of Synechocystis RNase P protein (-) and E. col i C5 protein (---) calculated with a window of 11 amino acids (Kyte and Doolittle, 1982). The black bar indicates the position of the conserved RNRLKR sequence.

Fig. 4. Purification of Synechocystis RNase P protein. The Synecho- cystis RNase P protein was purified from E. coli extracts that overex- press this protein as described in the Materials and Methods section. Lane 1, 30000 g supernatant; lane 2, ribosomal pellet; lane 3, ribosomal wash: lane 4, peak fractions from the G-50 column; lane 5 , peak fraction from the His-Bind resin column. Molecular mass standards are indicated (kDa).

tRNA"'" (Fig. 5 B, lanes 4 and 5). Exactly the same reconstituted holoenzymes were used with both substrates.

Inmunological analysis. The reconstitution assays described above indicate that the protein subunits from E. coli and Synechocystis 6803 are functionally interchangeable, despite their low sequence similarity. The structural relationship be- tween both proteins was studied using antibodies generated against the C5 protein and Synechocystis RNase P protein. Fig. 6 shows that the antibodies generated against the E. coli C 5 pro-

tein do not recognize the Synechocystis protein, even when it is present in large amounts (Fig. 6B). Similarly, antibodies gener- ated against the Synechocystis RNase P protein do not recognize significantly the E. coli C5 protein (Fig. 6C). Only a faint band can be seen at the C5 protein position. The antibodies obtained against the Synechocystis RNase P protein recognized a protein of the expected size in protein extracts from Synechocystis 6803 (Fig. 6 D , lane 1). A protein of similar size was also detected, although weakly, in extracts from the filamentous heterocyst- forming cyanobacterium Anabuena sp. PCC 71 20 (Fig. 6 D, lane 2) which suggests that the homologous protein from Ariabaeizcz 7120 is recognized by this antibody.

The result of the western analysis was confirmed by an immunodepletion experiment. Anti-CS serum cannot immuno- deplete the activity of holoenzyme reconstituted with the Sy- nechocystis 6803 protein while it can immunodeplete the holo- enzyme reconstituted with the CS protein (Fig. 7). A comple- mentary result was obtained with the anti-Syneci2ocy.l.ti.r RNase P protein. These antibodies cannot immunodeplete the activity reconstituted with the E. coli C5 protein, but they immunode- plete significantly the activity of the holoenzyme reconstituted with the Synechocystis RNase P protein and the Synrchocysfis RNase P RNA. However, when M1 RNA was used in the recon- stitution with the Synechocystis RNase P protein, no significant immunodepletion was observed by the Synechocystis 6803 pro- tein antibodies (Fig. 7). When the protein was incubated with the antibody before reconstitution of the holoenzyme, a significant reduction in activity was observed (data not shown). This result suggests that the binding of the protein subunit to M1 RNA

22 Pascual and Vioque (Eur: J. Biochem. 241)

Fig.5 Enzymatic activity of RNase P holoenzyme reconstituted with subunits from E. coli and Synechocystis 6803. "P-labeled E. coli pre- tRNAly' (A) or Syrzechocystis 6803 pre-tRNA"'" (B) were incubated as described in the Materials and Methods section in buffer alone (lane 1) or with Synechocystis RNase P protein (lane 2), reconstituted holoeiizyme from E. coli MI RNA and Synrclzocystis RNase P protein (lane 3), reconstitu- ted holoenzyme from Synechocystis RNase P RNA and Synechocystis RNase P protein (lane 4) or Synechocysfis RNase P RNA and E. coli CS protein (lane 5). The positions of Ihe prc-tRNA, 5'-cnd-maturcd tRNA and 5'-leader are indicated. In B, the short S'-leader has run off the gel.

Fig. 6. Imrnunoblot analysis. (A) Coomassie-brilliant-blue-stained gel of total protein from E. coli cells overexpressing the E. coli C5 protein (lane 1) or the Swwchocystis RNase P protein (lane 2). Arrowheads indi- cate the overexpressed protein. Molecular mass standards are indicated (kDa). (B) Western blot of a duplicate of the gel in A with antibodies against the E. c d i C5 protein diluted X104. (C) Western blot of a dupli- cate of the gel in A with antibodies against the Syrzechocystis RNase P protein diluted 5x10' times. (D) Western blot of total protein from Syrzechocysris 6803 (lane 1) and Anabaena 7120 (lane 2) with antibodies against the Synechocystis RNase P protein diluted X10'. About 10 pg and 30 pg protein were loaded in lanes 1 and 2, respectively. The arrow points to the band of the expected size for the RNase P protein. The broad, diffuse band above the hypothetical RNase P protein band in lane 2 is an artil'act due to the high amount of phycobiliproteins present in the extract.

protects the protein against the interaction with the antibodies, while the binding of the protein to the Synechocystis RNase P RNA does not.

DISCUSSION In this work, we describe the cloning of the gene coding for

the protein subunit of RNase P from a cyanobacterium. The gene coding the RNase P protein from Synechocystis 6803 was iso- lated from a plasmid library by its ability to complement a muta- tion in the E. coli RNase P protein. This result clearly indicates that the RNase P proteins from E. coli and Synechocystis 6803 are functionally interchangeable in vivo. We have confirmed this by enzymatic assays in vitro.

In all eubacteria analyzed so far, the rnpA gene is down- stream and closely linked to the rpmH gene, which codes for ribosomal protein L34. In this respect, the gene organization of Synechocyds 6803 is conserved. The rnpA gene starts with a

RNA E S E S E S E S

Antibodies E. coli Synechocystis 6803

Fig. 7. Depletion of RNase P activity with antibodies. RNase P holoen- zyme reconstituted from E. coli (E) or Synechocystis 6803 (S) protein or RNA subunits as indicated was incubated for 2 h with protein-A-conju- gated antibodies prepared against the E. coli C5 protein (hatched bars) or against the Synechocystis RNase P protein (open bars) as described in the Materials and Methods section. RNase P activity was assayed in aliquots of the supernatant with pre-tRNA-lY' as substrate. The activity is expressed as the percentage of the activity obtained when immunodeple- tion was performed with the corresponding preinmune antibodies in exactly the same conditions. The result of a typical experiment is shown. Similar qualitative results were obtained when the incubation time with antibodies or the amount of antibodies was changed.

GTG initiation codon nine nucleotides after the stop codon of the rpmH gene. The rnpA gene lacks a ribosome-binding site, so it is possible that the protein is synthesized by translational coupling with L34. The lack of a ribosome-binding site and the GTG initiation codon suggest an inefficient translation. In E. coli, which has a similar genetic organization, the amount of C5 protein is very low, while ribosomal protein L34 accumulates to high levels as expected for a ribosomal protein. Something simi- lar might happen in Synechocystis 6803. To detect the protein in crude extracts of Synechocystis 6803 with the corresponding antibody, it is necessary to load large amounts of protein and use the antibodies at a low dilution (Fig. 6 D , lane 1).

The rpmH-rpA region is transcribed divergently from the dnaA gene in most eubacteria. Except in Enterohacterictceae, the origin of chromosomal replication (oriC) is close to the dnuA gene (Ogasawara and Yoshikawa, 1992). However, in Syn- echocystis 6803 the dnaA gene maps in a different region of the chromosome (Richter and Messer, 1995) and oriC is not in the dnaA region. The three sequences similar to dnaA boxes upstream of rpmH could be part of the Synechocystis 6803 oriC.

Pascual and Vioque ( E m J. Biochem. 241) 23

The rpmH gene is found in the chloroplast of the unicellular algae Odontella sinensis (Kowallik et al., 1995), Porphyra purpurea (Reith and Munholland, 1995), and Olisthodiscus luteus (database accession number Z35718), and in the cyanelle of Cyanophora paradoxa (Stirewalt et al., 1995). In Porphyra and Cyanophora, a gene homologous to the bacterial RNase P RNA (rnpB) is also identified (Reith and Munholland, 1995; Shevelev et al., 1995). In Cyanophora, it has been shown that the rnpB transcript is part of a functional RNase P enzyme (Baum et al., 1996) but nothing is known about the protein sub- unit(s). However, in higher plants RNase P seems to be a purely proteinaceous enzyme, which lacks an essential RNA compo- nent (Wang et al., 1988). A gene homologous to the rnpA gene is not detected either downstream of the rpmH gene or in another region of the chloroplast genome in Odontella, Porphyru or Cyanophora. If the encoded protein exists, it might be possible to detect it with the antibodies described here against the Synechocystis RNase P protein.

The Synechocystis RNase P protein is very rich in basic amino acids (23.5%) and has some highly hydrophobic seg- ments such as the sequence FQPQIKPGFDVVIIVLPQGIG between positions 74 and 95. The presence of these long hy- drophobic regions could explain its low solubility under non- denaturing conditions and the fact that, once precipitated, it is difficult to get the protein back in solution. Hydrophobic interac- tions are important in the RNA-protein interaction in the E. coli RNase P holoenzyme (Talbot and Altman, 1994). The same could be true in Synechocystis 6803.

The functional equivalence between E. coli and Synecho- r,,ysti.s RNase P protein and the similarity in the hydrophobicity plots suggest that both proteins can attain similar structures and interact with the catalytic RNA subunits in similar ways. How- ever the details of the interaction of the Synechocystis RNase P protein and the RNA subunit from E. coli and Synechocystis must be different as suggested by the differential sensitivity to immunodepletion by Synechocystis RNase P antibodies of the enzyme reconstituted with M1 RNA or with Synechocystis RNase P RNA.

In this work, we describe the synthesis of a new pre-tRNA substrate from Synechocystis 6803 that allows us to study the cyanobacterial RNase P activity in a completely homologous system. This substrate is functional and will be a tool for future studies on the details of the interaction between enzyme and substrate in the cyanobacterial system. In this study, we show data that indicate that the origin of the protein subunit affects the efficiency of cleavage of the substrate. In the presence of the E, coli C5 protein, the E. coli pre-tRNATY' is cleaved better, while in the presence of the Synechocystis RNase P protein the Synechocysris 6803 pre-tRNAF1" is cleaved better (Fig. 5). The differential effect of the protein on the rate of cleavage of different substrates has already been suggested by in vitro (Guerrier-Takada et al., 1983) and in vivo studies in E. coli (Kirsebom et al., 1988).

A detailed study of the protein-RNA interactions in the sys- tem described here will help to provide constraints and facilitate the development of three-dimensional models for this ribo- nucleoprotein enzyme.

We thank Dr Leif Kirsebom for strain NHY322 and Drs Paul Eder and Sidney Altman for the gift of E. coli C5 protein antibodies. We are also grateful to Francisco J. Florencio for critical reading of the manuscript. Part of this work was perfomied by Agustin Vioque in Prof. Sidney Altman's laboratory (Yale University). The hospitality and support of Dr Altman and members of his laboratory is acknowledged. This work was supported by grants from the Direccidn General de Inves- tigacicin Cient@ca y Te'cnica (grant PB91-0104 to A. V.) and from the Junta de Andalucia (PA13283 to A. V.). A. P. was supported by a fellow-

ship from the Spanish Ministry of Education. The work performed by A. V. at Yale University was supported by the Spanish Ministry of Education.

REFERENCES Altman, S . & Smith, J. D. (1971) Tyrosine tRNA precursor molecule

polynucleotide sequence, Nature New Biol. 233, 35 -39. Altman, S. (1989) Rihonuclease P: an enzyme with a catalytic RNA

subunit, Adv. Enzymol. Relat. Areas Mol. Biol. 62, 1-36. Baer, M. & Altman, S. (1985) A catalytic RNA and its gene from

Salmonella typhimurium, Science 228, 999 - 1002. Baer, M., Wesolowski, D. & Altman, S. (1989) Characterization in vitro

of the defect in a temperature sensitive mutant of the protein subunit of RNase P from Escherichia coli, J. Bacteriol. 171, 6862-6866.

Banta, A. B., Haas, E. S., Brown, J. W. & Pace, N. R. (1992) Sequence of the ribonuclease P RNA gene from the cyanobacterium Anacystis nidulans, Niicleic Acids Res. 20, 911.

Baum, M., Cordier, A. & Schon, A. (1996) RNase P from a photosyn- thetic organelle contains an RNA homologous to the cyanobacterial counterpart, J. Mol. Biol. 257, 43-52.

Brown, J. W., Haas, E. S. & Pace, N. R. (1993) Characterization of rihonuclease P RNAs from thermophilic bacteria, Nucleic Acids Rex

Cai, Y. & Wolk, C. P. (1990) Use of a conditionally lethal gene in Anabaena sp. PCC 7120 to select for double recombinants and to entrap insertion sequences, J. Bucteriot. 172, 3138-3145.

Calcutt, M. J. & Schmidt, F. J. (1992) Conserved gene arrangement in the origin region of the Streptomyces cnelicnlor chromosome, J. Buc- teriol. 174, 3220-3226.

Devereux, J . , Haeberli, H. & Smithies, 0. (1984) A comprehensive set of sequences analysis programs for the VAX, Nucleic Acids Res. 12, 387 -395.

Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A,, Kirkness, E. F., et al. (1995) Whole-genome random sequencing and assembly of Haemophilus injluenzue Rd, Science 269, 496-5 12.

Fraser, C. M., Gocayne, J. D., White, O., Adams, M. D., Clayton, R. A,, et al. (1995) The minimal gene complement of Mycuplusma geni- talium, Science 270, 397-403.

Fujita, M. Q., Yoshikawa, H. & Ogasawara, N. (1990) Structure of the dnaA region of Micrococcus luteus: conservation and variation among eubacteria, Gene (Amst.) 93, 73-78.

Guerrier-Takada, C., Gardiner, K., Marsh, T., Pace, N. & Altman, S . (1983) The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme, Cell 35, 849-857.

Hansen, E G., Hansen, E. B. & Atlung, T. (1985) Physical mapping and nucleotide sequence of the rnpA gene that encodes the protein component of ribonuclease P in Escherichia culi, Gene (Amst.) 38. 85-93.

Kirsebom, L. A., Baer, M. F. & Altman, S. (1988) Differential effects of mutations in the protein and RNA moieties of RNase P on the efficiency of suppression by various tRNA suppressors, J. Mol. Biol.

Kowallik, K. V., Stoebe, B., Schaffran, I., Kroth-Pancic, P. & Freier, U. (1995) The chloroplast genome of a chlorophyll a+c-containing alga, Odontella sinensis, Plant Mol. B id . Rep. 13, 336-342.

Kyte, J. & Doolittle, R. F. (1982) A simple method for displaying the hydropathic character of a protein, J . Mol. B id . 157, 105-132.

Lai, C.-Y. & Baumann, P. (1992) Genetic analysis of an aphid endosym- biont DNA fragment homologous to the rnpA-rpmH-dnnA-dnaN- gyrB region of Eubacteria, Gene (Amst.) 113, 175-181.

Mamula. M. J., Baer, M., Craft, J. & Altman, S. (1989) An immunologi- cal determinant of RNase P protein is conserved between Escheri- chia coli and humans, Proc. Natl Atzld. Sci. USA 86, 8717-8721.

Mayes, S. R., Dalla Chiesa, M., Zhang, Z. & Barber, J. (1993) The genes aroA and frnQ are located upstream of psbO in the chromosome of Synechocystis 6803, FEBS Lett. 325, 255-261.

Miyata, M., Sano, K., Okuda, R. & Fukumura, T. (1993) Mapping of replication initiation site in Mycoplusma cupricolum genome by two- dimensional gel-electrophoresis analysis, Nucleic Acids Res. 21, 4816-4823.

21, 671-679.

204, 879-888.

24 Pascual and Vioyue (Eur: 1. Biochem. 241)

Morse, D. P. & Schmidt, F. J. (1992) Sequences encoding the protein and RNA components of ribonuclease P from Streptumyces biki- tiiensis var. zorbunensis, Gene (Amst.) 117, 61 -66.

Motamedi, H., Lee, Y. & Schmidt, F. J. (1984) Tandem promoters pre- ceding the gene for the M1 RNA component of Escherichicc coli rihonuclease P, Proc. Nut1 Acad. Sci. USA 70, 2091 -2095.

Ogasawara, N. & Yoshikawa, H. (1992) Genes and their organization in the replication origin region of the bacterial chromosome, Mol. Microbiol. 6, 629-634.

Pace, N. R. & Smith, D. (1990) Ribonuclease P: function and variation, J . B id . Clieni. 265, 3587-3590.

Pascual, A. & Vioyue, A. (1994) Sequence and structure o l the RNA subunit of RNase P from the cyanobacteriuin Pseudanahaenrt sp. PCC6903, Biochinz. Biuphys. Actu 1218, 463-465.

Reich, C., Olsen, G. J., Pace, B. & Pace, N. R. (1988) Role of the protein moiety of ribonuclease P, a ribonucleoprotein enzyme, Science 239, 178- 181.

Reith, M. & Munholland, J. (1995) Complete nucleotide sequence of the Porph?ra purpurea chloroplast genome, Plunt Mol. Hiol. Rep. 13, 333-335.

Richter. S. & Messer, W. (1 995) Genetic structure of the dnaA region of the cyanobacterium Swuclzttcysti.s .sp. strain PCC6803. J. Bucteriol.

Rippka, R.. Deruelles, J., Waterbury, 1. B., Herdman, M. & Stanier, R. Y. (1 979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria, .I. Gen. Microbiol. 111, 1-61.

Robertson, H. D., Altman, S. & Smith, J. D. (1972) Purification and properties of a specific Escizerichia coli ribonuclease which cleaves a tyrosine transfcr ribonucleic acid precursor, J. Bid . Chem. 247, 5243 - 5251.

Rosenberg, A. H., Lade, B. N., Chui, D., Dunn, J. J. & Studier, F. W. (1987) Vectors for selective expression of cloned DNAs by T7 RNA polymerase. Gene (Amst.) 56, 125 - 135.

Samhrook, J. , Fritsch, E. F. & Maniatis, T. (1989) Molecular cloning: u laboratory manual, 2nd edn, Cold Spring Harbor Laboratory, Cold Spring Harhor NY.

Schedl, P. & Primakoff, P. (1973) Mutants of Escherichia coli thermo- sensitive for the synthesis of transfer RNA, Proc. Nut1 Acad. Sci. USA 70, 2091 -2095.

177,4245-4251.

Schmitt, J., Hess, H. & Stunnenberg, H. G. (1993) Affinity purification of histidine-tagged proteins, Mol. H i d Rep. 18, 223 -230.

Shevelev, E. L., Bryant, D. A,, Loeffelhardt, W. & Bohnert, H. J. (1995) Ribonuclease-P RNA gene of the plastid chromosome from Cyano- phura parudoxu, DNA Res. 2, 231 -234.

Skovgaard, 0. (1990) Nucleotide sequence of a Protects mirahilis DNA fragment homologous to the 6OK-rnpA -rpmH-dizaA-dizaN-rei~F-~yrB region of Escherichia coli, Gene (Amst.) 93, 27-34.

Stirewalt, V. L., Michalowski, C. B., Loffelhardt, W., Bohnert, H. .I. & Bryant, D. A. (1995) Nucleotide sequence of the cyanelle genotne from Cyunophoru paradoxu, P l m t Mol. B id . Rep. 13, 327- 332.

Suhan, M., Chen, S.-Y., Thompson, H. A., Hoover, T. A., Hill. A. & Williams, J. (1 994) Cloning and characterization of an autonomous replication sequence from Cuxiella hurnetii, J . Bucteriol. 176, 5233 -5243.

Talbot, S. J . & Altman, S. (1994) Kinetic and thermodynamic analysis of RNA-protein interactions in the RNase P holoenzyme from Esche. riclzia coli, Biochenzistry 33, 1406- 141 1 .

Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice, Nucleic Acids Rex 22, 4673-4680.

Tilllsjij, A. & Kirsebom, L. A. (1093) Product release is a rate-limiting step during cleavage by the catalytic RNA subunit of Escheric,hia coli RNase P, Nucleic Acids Re$. 21, 51 -57.

Vioyue, A. (1992) Analysis of the gene encoding the RNA subunit of' ribonuclease P from cyanobacteria, NurIeic, Acids Rus. 20, 6331 - 6337.

Vioyue, A., Arnez, J. & Altman, S. (1988) Protein-RNA interactions in the RNase P holoenzyme from Escherichia culi, .I. Mol. Riol. 202.

Wang, M. J., Davis, N. W. & Gegenheimer, P. (1988) Novel mechanisms for maturation of chloroplast transfer RNA precursors, EMBO J. 7,

Yoshikawa, H. & Ogasawara, N. (1991) Structure and function of DnaA and the DnaA-box in eubacteria: evolutionary relationships of bacte- rial replication origins, Mol. Micrubiol. 5, 2589 -2597.

835 - 848.

1567-1514.