Plant Molecular Biology 20: 459-465, 1992. © 1992 Kluwer Academic Publishers. Printed in Belgium.

459

Purification and characterization of seven chloroplast ribosomal proteins: evidence that organelle ribosomal protein genes are functional and that NH2-terminal processing occurs via multiple pathways in chloroplasts J. Schmidt, E. Herfurth and A.R. Subramanian* Max-Planck-Institut fiir Molekulare Genetik, Abt. Wittmann, Ihnestrasse 73, Berlin 33 (Dahlem), Germany (* author for correspondence) Received 23 March 1992; accepted in revised form 2 June 1992

Key words: rpL20, L32, L33, L36, S 12, $16, S19; formyl-methionine processing; RP-HPLC and HPEC; spinach (Spinacia oleracea)

Abstract

Putative genes for 21 ribosomal proteins (RPs) have been identified in the chloroplast D N A of four plants by nucleotide sequencing and homology comparison but few of the gene products have been characterized. Here we report the purification and N-terminal sequencing of seven proteins from the spinach chloroplast ribosome. The data show them to be the homologues ofEscheriehia coli RPs L20, L32, L33, L36, $12, S16 and S19, and thus support the view that their genes identified in the chloroplast D N A represent functional genes. The initiating methionine residue was not detected in the mature protein in most cases but it was present in S 16, indicating that only the formyl group is removed in this case. This result and the previously reported finding of N-methyl alanine at the N-terminus of chloroplast L2 indicate the existence of multiple N-terminal processing pathways in the chloroplast.

Introduction

The chloroplast genomes of three plants which have been completely sequenced each contain 34-35 stable RNA genes and 78-80 polypeptide genes [1]. Twenty-one of the polypeptide genes encode proteins similar to Escherichia coli ribosomal proteins and are, by homology, believed to be components of the chloroplast ribosome [2]. In rice and tobacco, they include 12 RPs of the small ribosomal subunit and 9 RPs of the large ribosomal subunit; the corresponding numbers are 11 and 10 in the case of Marchantia [1]. As

discussed elsewhere [2], nearly all the organelleencoded RPs are homologues of the early assembly proteins, as deduced from the work on the E. coli ribosome. The 21 chloroplast RP genes, all of which have also now been sequenced in maize ([2,3,4]; unpublished results from our lab), occur mainly as clusters of 2-12 genes. In several cases the arrangement of the genes is similar to that found in E. coli or Bacillus species [2]. For example, two of the clusters resemble the SIO and spc operons of E. coli [2] with some genes of the bacterial clusters now located in the nucleus (e.g. rpslO, 13,

460

14, 129, S17 of the SIO operon) or translocated to another location in the chloroplast genome (e.g. rpsl4 of the spc operon). In at least 3 other cases a novel gene arrangement (introgression) is found [2]: an RP gene is introduced into a different gene complex. An example is the gene cluster rps2atpI, H, F, A which has recently been shown to be also a functional transcription unit. Transcription analyses in tobacco [6], spinach [7], and maize ([2-5]; unpublished results) have shown that all 21 of the RP genes are transcribed. Where examined in detail, the RP clusters are found to be functional transcription units (operons) directed by one or more promoters and terminated by specific secondary structures [2, 6, 7]. Thus, evidently, these genes are not silent, but it is not clear whether they are all translated into the corresponding polypeptides or are, in some cases, pseudogenes represented by functional nuclear genes. A Shine-Dalgarno sequence is generally found in the immediate 5' upstream region of the assumed translational start codons of the chloroplast RP genes [8], but there are several instances where such a ribosome-binding signal is missing (e.g. in maize rpll6, rp132, rps3, rps7 and exon 1 ofrpsl2). In spinach, the rpl23 gene of the SIO operon, although transcribed, has been shown to be a pseudogene [7]. It is therefore necessary to analyse for the gene products at the polypeptide level for a functional appraisal. Such work will also provide additional information on post-translational processing and modification, and can be useful vis-a-vis the reported R N A editing in the chloroplast [9]. In this paper we describe the purification and identification of 3 proteins of the small subunit and 4 proteins of the large subunit of the spinach chloroplast ribosome. From the N-terminal sequence, they are evidently homologues of the E. coli RPs S12, S16, S19, L20, L32, L33 and L36. The protein data support the view that the genes for these proteins, all identified in the chloroplast DNA, represent functional genes. In six cases the initiating methionine residue was not detected but in one case, i.e. S 16, only the formyl group is removed. Interestingly, the S16 protein in E. coli also begins with N-terminal methionine.

Materials and methods

Spinach (Spinacia oleracea, cv. Alwaro) chloroplast RPs reported in this paper were purified from pools (numbers 37, 71 and 110) of a previous large-scale prefractionation involving selective removal of protein L12 [ 10], gel filtration and ion exchange chromatography [11]. The pools were kept at -80 °C until use.

Prow& purification Protein components of the above pools were purified by either reversed-phase (RP) H P L C or high-performance electrophoresis chromatography (HPEC). For H P L C separation (Waters Associate) Vydac C4 and C18 (300 A, 5 #m) columns, 250 m m x 4.6 m m with a gradient of 0.1 ~o trifluoroacetic acid (TFA) in H 2 0 (solvent A) and 0.1 ~o TFA in either isopropanol or acetonitrile (solvent B) were used, at the combined flow rate of 0.5 ml/min. For H P E C separation an Applied Biosystems Instrument (model 230 A) with a vertical 50 m m x 2.5 m m gel of 10~o polyacrylamide was used. The upper and lower electrode buffers were Tris-phosphate pH 7.0 and Tris-C1 pH 7.5, respectively. Aliquots from fractions containing UV-absorbing peaks were analysed on SDS slab gels [ 11] to estimate the protein amount/purity and approximate molecular weights. The material was then desalted (from H P E C runs), vacuum-dried and used for sequenator runs.

Two-dimensional gel electrophoresis Two-dimensional gel electrophoresis of the pools or of purified material was done as described [ 12].

Amino acid sequencing Proteins were degraded with phenyl isothiocyanate in the horizontal flow-through reactor of a modular Knauer Sequencer (model 810) by the wet-phase filter technique [ 13 ] employing polyvi-

461 nylidene difluoride (PVDF) membranes without polybrene. Some of the purified proteins were sequenced at the Protein/Nucleic Acid Shared Facility of the Medical College of Wisconsin (NIH grant RR-03326).

were made up of two components but individual fractions were obtained giving unambiguous sequence results. The peak marked L35 contained the homologue of E. coli L35. It was previously obtained from another pool and the data therefrom were used to obtain the c D N A clone of this nuclear-encoded protein [ 14]. Protein S 12 was obtained pure in one of the peaks (Fig. 1A); its gene in the chloroplast D N A is a divided gene [ 1 ] with one of the exons occurring widely separated and in a different transcription unit with the catalytic subunit (clpP) of an ATP-dependent protease (e.g. [3]). Figure 1 also shows several peaks marked A1, A2 etc. or ( - ). The latter are peaks with no Coomassie blue staining material, while the former contain polypeptides whose N-terminal sequences revealed that they are not encoded in the plastic genome. Data on these presumably nuclear-encoded chloroplast ribosomal proteins will be reported in a separate communication. Figure 2 shows the elution profiles of pool 37 by R P - H P L C and by HPEC. This pool contained mainly two components, both of very similar size and net charge, as revealed by two-dimensional analysis (inset in Fig. 2B). Both H P L C and H P E C failed to resolve these two components completely, but the latter procedure gave individual fractions that contained only a single compo-

Computer analysis Computer analysis was software (version6.2) using NBRF/SwissProt databanks with a VAX

based on the U W G C G sequence comparisons or RIBO (this institute) 8600/VMS computer.

Results

Purified proteins Figure 1 shows the R P - H P L C elution profiles of two pools from which six of the chloroplast RPs reported here (S12, S16, $19, L32, L33 and L36) were obtained. Protein L32 appeared in pure form in two peaks (Fig. 1B); this may indicate, as in the similar case of L2 reported earlier [11], that it exists in two conformations or in two processed/ modified forms (with the N-terminal sequence retained). Two of the peaks (L33 (L36), S16/$19)

Pool 71

Pool llO

% B(---)

.10

.10

100

,

Fi .08

80

"i" .08

60

,g .o6

.0t~

~0

< .04

.02

20

.02

E

@ .06

Purification and characterization of seven chloroplast ribosomal proteins: evidence that organelle ribosomal protein genes are functional and that NH2-terminal processing occurs via multiple pathways in chloroplasts.

Putative genes for 21 ribosomal proteins (RPs) have been identified in the chloroplast DNA of four plants by nucleotide sequencing and homology compar...
538KB Sizes 0 Downloads 0 Views