Plant Molecular Biology 35: 1053–1056, 1997. 1053c 1997 Kluwer Academic Publishers. Printed in Belgium.

Short communication

Isolation and characterisation of tobacco (Nicotiana tabacum) cDNA clonesencoding proteins involved in magnesium chelation into protoporphyrin IX

Elisabeth Kruse, Hans-Peter Mock and Bernhard Grimm�

Institut fur Pflanzengenetik und Kulturpflanzenforschung, Corrensstrasse 3, 06466 Gatersleben, Germany(�author for correspondence)

Received 29 July 1997; accepted 28 August 1997

Key words: chlorophyll synthesis, porphyrins, RACE, chloroplast, Fe chelatase

Abstract

We have identified cDNA clones encoding the two Mg chelatase subunits CHL I and CHL H from tobacco (Nicotianatabacum) by screening a cDNA library with homologous cDNA fragments from Arabidopsis thaliana. A full-lengthChl I cDNA clone encodes a peptide with 426 amino acids. The entire cDNA sequence encoding 1382 amino acidlong CHL H was obtained by extension of a truncated cDNA fragment using the ‘rapid amplification of cDNAends’ (RACE) method. Both genes Chl I and Chl H were strongly expressed in young leaves and to a lesser extentin mature leaves. Only traces of both transcripts were found in flowering organs. Southern blot analysis suggeststhat CHL I is encoded by a single gene and CHL H most likely by several genes.

Introduction

Mg chelatase catalyses the insertion of Mg2+ intoprotoporphyrin IX which is the first step directingtetrapyrroles to the (bacterio)chlorophyll biosynthet-ic pathway unique to photosynthetic organisms [8].Mutation of one of three open reading frames desig-nated bchD, bchH and bchI of the 45 kb photosynthet-ic gene cluster of Rhodobacter sphaeroides [3] abol-ished the Mg chelatase activity and led to accumula-tion of protoporphyrin IX [1]. These three proteins areencoded in the R. sphaeroides gene cluster for photo-synthetic proteins [6, 18]. The three homologous pro-teins CHL D, H and I from Synechocystis sp. PCC6803expressed in Escherichia coli were also able to recon-stitute Mg chelatase activity in vitro [11]. In contrast tothe complex structure of Mg chelatase, ferrochelatase,the other metal chelating enzyme at the branch pointof tetrapyrrole biosynthesis, consists only of a singlesubunit [16].

The nucleotide sequence data reported will appear in the EMBL,GenBank and DDBJ Nucleotide Sequence Databases under theaccession numbers AF014053 (Chl I) and AF014054 and AF014052(Chl H).

The first plant genes encoding polypeptides forMg2+ chelation have been identified by T-DNA tag-ging or transposon mediated mutagenesis and were ini-tially non-committally assigned as proteins involved inchlorophyll synthesis. A gene from Arabidopsis thali-ana designated cs42 showed sequence similarity tobchI [12], and the oli gene from Antirrhinum majus washomologous to bchH [9]. These sequence comparisonssuggested a structural similarity between bacterial andplant Mg chelatase. Transcript levels of the homolog-ous barley Chl I and Chl H genes which are related tothe Xan-h and Xan-f locus, respectively, are altered inthe three chlorophyll-deficient mutant Xantha-f, -g and-h [10]. Here we report the identification and charac-terisation of cDNA sequences encoding the CHL I andCHL H proteins from tobacco.

A � ZAP cDNA library from tobacco leaf (SR1 strain; Stratagene, La Jolla, CA) was screened bystandard procedures [14] using 32P-labelled Arabidop-sis cDNA inserts encoding either CHL I [12] or CHLH (generous gift from Dr C. Koncz, MPI fur Pflan-zenzuchtung, Cologne). Chl I and Chl H clones wereisolated from the library and sequenced on both strands

1054

by the dideoxy chain termination method [15] usinginternal primers.

The cDNA clone NtCHLI (deposited in the Gen-Bank under the accession number AF014053) con-tains the entire open reading frame for tobacco CHLI and encodes a protein of 426 residues with a pre-dicted molecular mass of 46.6 kDa. The sequenceflanking the initiation codon (AAAACAAAATGGC)resembles the consensus sequence for dicotyledonousplants (AAAAAAAA/CATGGC) [2]. A putative poly-adenylation motif (AATTA) [19] is present in a 198 bplong 30-untranslated region 55 nucleotides upstreamof the poly(A) tail. The deduced amino acid sequenceof tobacco CHL I contains an N-terminal extensionwith characteristics of a plastid targeting signal [17].The tobacco precursor protein shows the best matchto the soybean CHL I with 81% identical amino acidresidues [13]. Comparison of the amino acid sequenceof the tobacco protein with other CHL I/bchI pro-teins revealed the following relative overall sequencesidentities (in descending order): 78% to ArabidopsisCHL I [15], 72% to Synechocystis PCC6803 chlI [11]and 48% to Rhodobacter bchI. The overall similarityamong all known CHL I/bchI proteins was 66%.

Screening of the tobacco cDNA library for theChl H sequence resulted in a 3839 bp long sequencewhich included an uninterrupted open reading framecoding for 1221 amino acid residues and which lackeda putative initiation codon. cDNA fragments encod-ing the 50 end of the Chl H transcript were obtainedby PCR amplification of cDNA [5] prepared frompoly(A)+ RNA of tobacco leaves (Samsun NN, IPKGatersleben) using the Marathon 50/30-RACE system(Clontech, CA) and the gene-specific antisense primerCTC- ATCACCTCAGGCATTGAT. The PCR frag-ments were ligated into the pCR II vector (Invitrogen,San Diego, CA). The sequence of several circa 750–800 bp long inserts contained the coding sequencefor the 177 N-terminal amino acids and belonged totwo different subgroups of Chl H transcripts. The twogroups of subclones showed 94% identity at the nuc-leotide level. The overlapping nucleotide sequence ofone group (NtCHLH-1) and the NtCHLH sequencewas identical, whereas the 50 extended fragment ofsubgroup 2 (NtCHLH-2 was deposited in the GenBankunder the accession number AF014051) contained onebase pair substitution. The NtCHLH-1 fragment wasfused in frame to the 3839 bp long NtCHLH sequenceby PCR as follows: The insert of NtCHLH-1 was amp-lified by PCR using the antisense primer describedabove and the M13 universal primer, the sequence

of NtCHLH was amplified by means of the com-plementary sense primer ATCAATGCCTGAGGTG-ATGAG and the pBluescript SK primer (Stratagene).Both PCR fragments were isolated, combined in aPCR reaction with SK and universal primer, and theresulting extended PCR product was subcloned intothe pCR II vector. The intactness of the combinedcDNA was confirmed by nucleotide sequencing andin vitro transcription/translation. The entire nucleotidesequence of 4578 bp comprises 257 nucleotides ofthe 50-untranslated region, the open reading frameand 176 nucleotides of the 30-non-coding sequence(deposited in the GenBank under the accession numberAF014052). Differences in the sequence of subclonesNtCHLH-1 and NtCHLH-2 which includes the openreading frame of the first 177 amino acid residues,resulted in amino acid substitutions at the follow-ing positions: 13 (N!S), 37 (I!T), 92 (S!A), 106(Q!K), 125 (T!A) and 155 (S!T). The tobacco ChlH sequence encodes a protein of 1382 amino acids witha putative N-terminal plastid targeting sequence. Thecalculated molecular mass of 154 kDa corresponds tothe size of the homologous precursor CHL H proteinfrom higher plants [7, 9, 10]. Based on sequence com-parison with plant as well as prokaryotic counterparts,we tentatively assess the proteolytic cleavage site ofthe CHL H transit sequence between amino acid 52 (A)and 53 (I). In comparison to the known bchlH/CHL Hsequences, tobacco CHL H shares 91, 84, 82 and 65%of invariant amino acid residues with the protein fromAntirrhinum [9], Arabidopsis [7], barley [10] and Syn-echocystis sp. PCC 6803 [11], respectively. All knownCHL H/bchH proteins share a 60% overall similarity.The cobN gene product, a protein required in coen-zyme B12 formation of Pseudomonas denitrificans andinvolved in the chelation of cobalt into porphyrins, isthe closest related protein to the CHL H family. CobNbinds the tetrapyrrole substrate [4], a function which isalso attributed to CHL H [6].

Total RNA isolated from tobacco leaves was sub-jected to northern blot analysis. The Chl I and the ChlH cDNA hybridised to transcripts of apparent molecu-lar sizes of 1.4 kb and 4.5 kb, respectively (Figure 1,left panel). These sizes correspond to the length of ourcDNA sequences described. Tissue specific expressionof the two Mg chelatase transcripts was analysed inyoung and fully expanded mature leaves (leaf 5 or 10of 6-week-old plants), in roots and in flowering organs(Figure 1, right panel). The Chl I and Chl H RNAshowed the same tissue specific expression pattern.Thetranscript levels were slightly higher in young than in

1055

Figure 1. Northern blot analysis. Left panel: 10 �g of total RNAisolated from tobacco leaves were separated on a formaldehyde-agarose gel and transferred onto nylon membrane. Filters wereprobed with the cDNA inserts of the clones NtCHLH and NtCHLI,respectively, radio-labelled by random primer insertion. Right pan-el: tissue-specific expression of Mg chelatase transcripts. Total RNAwas extracted from young (YL) and mature (ML) leaves (leaf 5 and10, respectively, counting from top to base), roots (R), petals (Pe)pistils (Pi) and stamen (St) and subjected to northern blot analysisas described above. Equal loading of RNA was verified by ethidiumbromide staining and probing with an actin cDNA sequence.

old leaves, thus paralleling the higher chlorophyll syn-thesising capacity during leaf development. The RNAlevels decreased in full-size leaves. Low amounts ofmRNA could be detected in petals, pistils and stamen,which is conceivably due to the low chlorophyll con-tent of flowering organs. Mg chelatase transcripts werebelow the detection limit in roots. For comparison, thesame RNA filter was probed with a cDNA insert encod-ing tobacco ferrochelatase (E. Schmidt and B. Grimm,unpublished). Its steady-state mRNA level was modi-fied relative to those of the transcripts of the Mg che-latase subunits. However, Fe chelatase did not stronglydiffer in the tissues analysed, which might be explainedby a constant need of haem in different organs relativeto the predominant need of chlorophyll in photosyn-thetically active tissue. Actin mRNA provides controlfor the loading of the RNA samples. Southern blotanalyses were performed using tobacco genomic DNAdigested with BamHI, DraI, EcoRV, HindIII, XbaI andKpnI, respectively. Filters were probed with the cDNAinserts of NtCHLH and NtCHLI and washed underhigh stringency conditions. Taking the number and thesize of the hybridising bands into account (Figure 2),it is suggested that CHL I is encoded by a single-copygene and CHL H by a small gene family or differ-ent alleles due to the allotetraploid nature of tobacco.The latter assumption would be consistent with the

Figure 2. Southern blot analysis of tobacco Chl H and Chl I. Ten �gof genomic DNA from tobacco were digested with BamHI (B), DraI(D), EcoRV (E), HindIII (H), XbaI (X) and KpnI (K) and subjectedto Southern blot analysis using 32P-labelled cDNA inserts of clonesNtCHLH and NtCHLI.

limited heterogeneity of the two subclones obtainedby the 50 RACE approach. The identification of nuc-leotide sequences encoding two proteins involved inmagnesium protoporphyrin IX chelation allows to per-form future experiments on the control of Mg chelataseactivity using transgenic plants with modulated expres-sion and function of these proteins.

Acknowledgements

The work was supported by a grant from the DeutscheForschungsgemeinschaft to B.G. (Gr 936/4-1).

References

1. Bollivar DW, Suzuki JK, Beatty JT, Dobrowolski JM, BauerCE: Directed mutational analysis of bacteriochlorophyll a bio-synthesis in Rhodobacter capsulatus. J Mol Biol 237: 622–640(1994).

2. Cavener DR, Ray S: Eukaryotic start and stop translation sites.Nucl Acids Res 19: 3185–3192 (1991).

3. Coomber SA, Chaudhri M, Connor A, Britton G, Hunter CN:Localized transposon Tn5 mutagenesis of the photosyntheticgene cluster of Rhodobacter sphaeroides. Mol Microbiol 4:977–989 (1990).

4. Debussche L, Couder M, Thibaut D, Cameron B, Crouz-et J, Blanche F: Assay, purification and characterisation ofcobaltochelatase, a unique complex enzyme catalyzing cobaltinsertion in hydrogenobyrinic acid a,c-diamide during coen-zyme B12 biosynthesis in Pseudomonas denitrificans. PlantPhysiol 67: 246–249 (1992).

5. Frohman MA, Dush MK, Martin GR: Rapid production of full-length cDNA from rare transcripts: amplification using a singlegene-specific oligonucleotide primer. Proc Natl Acad Sci USA85: 8998–9002 (1992).

1056

6. Gibson LCD, Willows RD, Kannangara CH, von WettsteinD, Hunter CN: Magnesium-protoporphyrinogen chelatase ofRhodobacter sphaeroides: reconstitution of activity by com-bining the products of the bchH, -I, and -D genes expressedin Escherichia coli. Proc Natl Acad Sci USA 92: 1941–1944(1995).

7. Gibson LCD, Marrison JL, Leech RM, Jensen PE, BasshamDC, Gibson M, Hunter CN: A putative Mg chelatase subunitfrom Arabidopsis thaliana cv. C24. Plant Physiol 111: 61–71(1996).

8. Gorchein A: Magnesium protoporphyrin chelatase activity inRhodopseudomonas spheroides: studies with whole cells. Bio-chem J 127: 97–106 (1972).

9. Hudson A, Carpenter R, Doyle R, Coen ES: Olive: a key generequired for chlorophyll biosynthesis in Antirrhinum majus.EMBO J 12: 3711–3719 (1993).

10. Jensen PE, Willows RD, Petersen BL, Vothknecht UC, Stum-mann BM, Kannangara CG, von Wettstein D, Henningsen KW:Structural genes for Mg-chelatase subunits in barley: Xantha-f,-g and -h. Mol Gen Genet 250: 383–394 (1996).

11. Jensen PE, Gibson LCD, Hennigsen K, Hunter CN: Expressionof the chlI, chlD, and chl H genes from the cyanobacteriumSynechocystis PCC6803 in Escherichia coli and demonstrationthat the three cognate proteins are required for magnesium-protoporphyrin chelatase activity. J Biol Chem 271: 16662–16667 (1996).

12. Koncz C, Meyerhofer R, Koncz-Kalman Z, Nawrath C, ReissB, Redei GP, Schell J: Isolation of a gene encoding a novel

chloroplast protein by T-DNA tagging in Arabidopsis thaliana.EMBO J 9: 1337–1346 (1990).

13. Nakayama M, Masuda T, Sato N, Yamagata H, Bowler C,Ohta H, Shioi Y, Takamiya K: Cloning, subcellular localizationand expression of CHL I, a subunit of magnesium-chelatasein soybean. Biochem Biophys Res Commun 215: 422–428(1995).

14. Sambrook F, Fritsch EF, Maniatis T: Molecular Cloning: ALaboratory Manual, 2nd ed. Cold Spring Harbor Laboratory,Cold Spring Harbor, NY (1989).

15. Sanger F, Nicklen S, Coulson AR: DNA sequencing with chainterminating inhibitors. Proc Natl Acad Sci USA 74: 5463–5467(1977).

16. Smith AG, Santana MA, Wallace-Cook ADM, Roper JM,Labbe-Bois R: Isolation of a cDNA encoding chloroplast fer-rochelatase from Arabidopsis thaliana by functional comple-mentation of a yeast mutant. J Biol Chem 269: 13405–13413(1994).

17. von Heijne G, Steppuhn J, Herrmann RG: Domain structureof mitochondrial and chloroplast targeting peptides. Eur J Bio-chem 180: 535–545 (1989).

18. Willows RD, Gibson LCD, Kannangara CG, Hunter CN, vonWettstein D: Three separate proteins constitute the magnesiumchelatase of Rhodobacter sphaeroides. Eur J Biochem 235:438–443 (1996).

19. Wu L, Uede T, Messing J: The formation of mRNA 30-ends inplants. Plant J 8: 323–329 (1995).