ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 298, No. 1, October, pp. 1922197, 1992

Chloroplast and Cytoplasmic Enzymes: Isolation and Sequencing of cDNAs Coding for Two Distinct Pea Chloroplast Aldolases’ ,*I3 Heidi Kuldeep Razdan,*,4 Robert L. Heinrikson,? Paul W. Morris,$ and Louise E. Anderson*y5

Zurcher-Neely,?

*Department of Biological Sciences and $Department of Biochemistry and Research Resources Center, University of Illinois at Chicago, Chicago, Illinois, 60680; and tThe Biochemistry Unit, Upjohn Company, Kalamazoo, Michigan, 49001

Received April 6, 1992, and in revised form June 5, 1992

Two cDNAs which correspond to two very similar Class I aldolases have been isolated from a pea (Pisum sativum L.) cDNA library. With the exception of one codon they match the experimentally determined N-terminal sequence of a pea chloroplast aldolase. The deduced C-terminal sequence of one of these clones is unique among Class I aldolases. The deduced C-terminus of the other is more like the C-terminus of other eucaryotic Class I aldolases. Comparisons of sequence homology suggest that the pea chloroplast isozymes are only marginally more closely related to the anaerobically induced plant aldolases than to aldolases from animals. o isea Academic

Press,

Inc.

Fructosebisphosphate aldolase (EC 4.1.2.13) in plastids is involved in both glycolysis and gluconeogenesis. In photosynthesizing chloroplasts it catalyzes the formation of fructosebisphosphate and of sedoheptulosebisphosphate. The purpose of the present experiments was to i Paper 9 of a series. Ref. 1 is the preceding paper. * Supported by grants from NSF (DMB-8417081 and DCB-9018265) and the University of Illinois-Chicago Research Board. 3 The nucleotide sequences of cpl and cp2 have been deposited with the GenBank Nucleotide Sequence Database under Accession Nos. M97476 and M97477. 4 Recipient of an Indian Government National Fellowship in Molecular Genetics. Present address, Department of Hematology, Baylor College of Medicine, One Baylor Plaza, Houston TX 77030. ’ To whom correspondence and reprint requests should be addressed at Department of Biological Sciences (m/c 066), University of IllinoisChicago, Box 4348, Chicago IL 60680. Bitnet U25317@UICVM. Fax: (312) 413-2435.

192

isolate and sequence cDNA(s) coding for this enzyme. Here we report the sequence of two cDNAs which code for two closely related pea leaf aldolases. The 5’ end of one codes for what appears to be a chloroplast transit peptide. The other is truncated at the codon corresponding to the N-terminal amino acid of the mature peptide. Both match, with the exception of one amino acid codon, the experimentally determined N-terminal sequence of the pea leaf plastid aldolase. We conclude that the first codes for a chloroplast aldolase and that the second is probably a variant form. MATERIALS

AND

METHODS

Isolation of pea leaf aldolases. Pea (Pisum sativum, L. var Little Marvel) plants were grown from seed obtained from Green Seed Co. (Kimberly, ID) for lo-12 days in Sunshine Seedling Mix No. 3 (FiscousWestern Co., Vancouver, BC, Canada) at the University of Illinois, Chicago Greenhouse. For isolation of the chloroplastic aldolase isozyme, chloroplasts were obtained by blending about 500 g pea shoots in 1 liter of 0.33 M sorbitol, 10 mM sodium pyrophosphate, 10 mM NaCl, pH 7.5 (HCI), filtering through four layers of cheesecloth and one layer of Lutrasil (Lutrasil Sales Co., Durham, NC) and pelleting by centrifugation (SOOOg,10 min). The chloroplast pellet was resuspended in an excess of the isotonic buffer and again pelleted by centrifugation (8OOOg,10 min). This washing process was repeated two more times after which the chloroplast pellet was stored frozen in the isotonic buffer (-20°C). When we wanted to prepare the chloroplastic isozyme, the broken chloroplast preparation was thawed, made 1 mM in phenylmethanesulfonyl fluoride, and centrifuged (28,OOOg.15 min) to remove particulates. The supernate was made 25% (w/v) in polyethylene glycol6000. After 40 min the precipitated proteins were removed by centrifugation (28,OOOg, 1 h) and the supernate was made 10 mM in MgClr by the addition of 1 M MgClx. After 45 min the precipitated protein was collected by centrifugation (ZS,OOOg,40 min), dissolved in 20 mM Tris-HCl, pH 7.4, and loaded onto a Pharmacia Mono Q column (HR 5/5). The column was washed with 0.13 M KC1 in the Tris buffer until the baseline stabilized, after which the enzyme was eluted with a convex 0.13-0.25 M KC1 gradient 0003sg&31/92$5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

PEA CHLOROPLAST in the Tris buffer using profile 4 from the Millipore automated gradient controller (Waters Associates, Milford, MA). The isozyme eluted in 0.19 M KCl. The yield was 8% and the enzyme was 45-fold purified. The specific activity of the purified enzyme was 5 rmol fructose bisphosphate cleaved mine1 mg protein-‘. Reported specific activities for aldolase from chloroplasts range from 3 to 8 nmol fructosebisphosphate cleaved min-’ mg protein-’ (2-4). Denaturing gel electrophoresis revealed a major protein component shadowed by a minor component which probably accounted for less than 5% of the total protein. Chromatofocusing (Pharmacia Mono P HR 5/20 column) was sometimes used as a final purification step instead of Mono Q chromatography. A similar procedure was used for the purification of the cytosolic isoenzyme, except that the supernatant solution from the chloroplast isolation (crude cytosolic extract) was the starting material, the polyethylene glycol concentration was only 20%, and chromatofocusing replaced Mono Q chromatography. The specific activity of the purified enzyme was 6 prnol fructose bisphosphate cleaved min-’ mg protein-‘, the enzyme was l7-fold purified, and the yield was 7%. The protein appeared to be electrophoretically homogeneous. Reported specific activities for cytosolic plant aldolases range from 7 to 26 nmol fructose bisphosphate cleaved mini mg protein-’ (2, 3, 5, 6). N-terminul sequencing. Prior to sequencing, the chloroplastic isozyme was dissolved in 0.1% trifluoroacetic acid, 0.1% mercaptoethanol, loaded onto an Applied Biosystems RP 300 column 6 M guanidine-HCl, (2.1 X 100 mm), and resolved in a linear gradient of acetonitrile in trifluoroacetic acid (5% acetonitrile, 0.1% trifluoroacetic to 70% acetonitrile, 0.085% trifluoroacetic acid). Automated Edman degradation was performed in an Applied Biosystems, Inc. Model 477A pulsed liquid phase sequenator or in a Model 470A gas-phase protein sequenator fitted with an on-line Model 120A PTH amino acid analyzer. Data were collected and yields calculated on a Nelson Analytical 3000 series chromatography system when the 470A system was used. Antibody production. Mice were twice injected intraperitoneally with 100 pg isozyme protein emulsified in incomplete Freund’s adjuvant with a 2-week interval between injections. Blood was collected prior to the first injection, 2 weeks after the second injection, and at irregular intervals thereafter. Sera were obtained by centrifugation after the blood had clotted. Antisera used for immunoscreening were first passed through a column of Escherichia coli protein linked to Sepharose to reduce reactivity with host bacterium protein. Like Marsh et al. (7) and Botha and Kennedy (8), we found little difference in cross-reactivity between the chloroplastic and cytosolic

10 20 GSYADELVKTAKTIASPGRCILAHDES-NWTCG X*X****X****Xt***XX********,*A*NS

Pea chloroplastic Pea cytosolic Deduced

Protein Probe

30

GSYADELVKTAKTIASPGRGILAHDES-NATCGKRLD....

3’

GRGILAHDESN CCG TCC CCA TAG A G

CAC CGA TAC A G G

CTA

CTC

ACG TT A

5’

FIG. 1. N-terminal amino acid sequence of the pea chloroplastic and cytosolic aldolase isozymes, deduced N-terminal sequence of cpl and cp2, and degenerate probe used in screening. The chloroplastic isozyme was purified as described under Materials and Methods with Mono Q anion-exchange chromatography as the final step in enzyme purification. The cytosolic enzyme was isolated as described by Anderson (25). Asterisk indicates identical residue. The sequences have been aligned for maximum homology. Since cysteine and P-serine could not be separated in the system used for this determination residue 31 of the chloroplastic isozyme could also be phosphoserine. We also obtained the same sequence (through 26 residues) for the pea leaf isozymes which had been isolated by chromatofocusing.

193

ALDOLASES El00 I

I

I

I

500 I I

E Ill

I

1000 I I

E II

E loo 500 E 1000 E 1500 E I I I I I I I II I I I II I I I II

FIG. 2.

Restriction map and sequencing strategy for cpl (upper) and cp2 (lower). E, EcoRI restriction site. Arrows indicate direction of sequencing, with each arrow indicating sequence obtained with one primer. Units are base pairs. Orientation was determined by comparison of deduced amino acid sequences with the experimentally determined Nterminal sequence and with other aldolases. The orientation of the third EcoRI restriction fragment of cp2 was determined by restriction mapping with SacI. This fragment does not match sequences in GenBank and is not part of the coding region of cp2. We assume that it is a cloning artifact. It is not included in the sequence reported in Fig. 3, but it has been deposited with the sequence of cp2 in GenBank.

isozymes and the antisera prepared against either isozyme (data not shown). We used antibody against the cytosolic enzyme for screening because more was available and because it seemed that it would be as effective as antibody against the chloroplast enzyme for this purpose. We were able to isolate two different clones which apparently code for chloroplast aldolases. Isolation of clones. Immunoscreening and plaque purification were as described by Mierendorf et al. (9). E. coli strain Y1090R- (Stratagene, La Jolla, CA) was used as host bacterium. The secondary antibody was goat antimouse IgG conjugated to alkaline phosphatase. Oligonucleotide screening was done according to the directions supplied with Colony/ PlaqueScreen Hybridization Transfer Membranes (New England Nuclear, Wilmington, DE). We used a 32P-labeled degenerate oligonucleotide designed to match part of the N-terminal portion of the chloroplastic isozyme (Fig. 1) as a probe. DNA sequencing. Xgtll phage DNA was isolated by the method of Elliot and Green (10). EcoRI digestion and subcloning into M13mp18 and M13mp19 were according to Sambrook et al. (11). The large 800bp fragment of cpl was cut with Sau3A and cloned into M13mp19 at the EcoRI-BarnHI site. Sequencing was done using the Sequenase Version 2.0 kit from United States Biochemicals (Cleveland, OH). For longer templates we used synthesized oligonucleotides as internal primers. Computer analysis. Comparisons with nucleotide (GenBank) and amino acid (Swiss-Prot) data bases were done using the FASTA program of Pearson and Lipman (12). The calculated estimate of pZ’ was done for us by Prosis. Chemicals. Oligonucleotides were synthesized in the department (Biosearch 8700 4 Column Nucleotide Synthesizer) or purchased from Operon Technologies (Alameda, CA). Restriction enzymes, T4 polynucleotide kinase, and T4 DNA ligase were obtained from Bethesda Research Laboratories (Gaithersburg, MD). Radiochemicals were from New England Nuclear (Wilmington, DE). E. coli proteins were coupled to cyanogen bromide-activated Sepharose 4B using the protocol supplied by Pharmacia (Uppsala). Other biochemicals were from Sigma Chemical Co. (St. Louis, MO). Inorganic chemicals were reagent or enzyme grade.

RESULTS

AND

DISCUSSION

N-terminal sequencing. The N-terminal sequences of the chloroplastic and cytosolic aldolase isozymes are given

194

RAZDAN

1 1 2 2

-1 +1 GLTIRAGSYADELVKTAKT CA GGC CTC ACC ATC AGA GCT GGT TCA TAT hjl MC **A * * *

11 1 ATT 2 Wt*

13 GCT GAC GAG CTC GTC AAG ACC GCG AAA ACA WCC WC* fr** 9cabk McA **A **T YWC~ Mat fcWc * * * * * * * * * A

A GCT **at *

S TCA *HI *

P G CCC GGG *atA WC* **

G GGT **jl *

I L ATT TTA **X **G * *

A GCA WCC *

M ATG Rat* *

0 GAT i&c* *

E GAG *ht *

S TCT MC *

N AAT X#r* *

A GCT atat* *

T ACA AAX *

C TGC MT *

G GGG WcA *

K AAG B** *

1R L 1 CGT TTG 2 WtG RXJr 2r; *

A GCT *A* D

S IG L TCA ATT CGG CTA fdrrt WrSx *at* **at 919~ 9~ 9~

E N GAG AAC fcak91 Mot k X

T ACC **A j,

E GAA f&G 9~

V GTT *CA A

N AAC ttk* 9~

R CGC MT *

Q CAA 9&t ft

A GCA X** 9~

W TGG h'ch at

R CGT aW& tt

T ACA JrX* h

L CTT Wt* ,t

1L Y 1 CTT TAC

Q S T T D G R K IV CAA TCC ACA ACA GAT CGA AGG AAG ATT

D V L GTC GAT GTT CTT

IE ATT

Q N I GAG CAA AAC ATT

2*

R CGT Wet *

ET AL.

2 **C **rt **at aWdt * * v * A Jr * * * Jr it h v A *

IS W C Q G L D G L 1 TCA TGG TGC CAA GGT CTA GAC GGT CTT

A S GCC TCT

R S CGG TCA

A A Y Y GCT GCA TAC TAC

**A Jr

A K W R T V V S IP GCC AAA TGG CGT ACT GTT GTG AGC ATC

N CCC AAT

G P S GGT CCT TCT

33 119

1;;

93 299

*WC *

Q

Q G CAA CAA GGT

416

A L A GCT CTT GCA

152 476

2 A?t* **i* fr*fc *X* *Rat WrfsWC***A *Xh 9c*siMC* 9&T WcGX*31f&J; f&J; A** WtG A** h'ch 2* J; * A * J; A * * * 8 x * * J; rt R * * * 1A R F 1 GCC CGT TTC

59

2 Wn't fcrk* a'~**WC**XX *ht k** WC*WCC9&h *Irk 9&h foth MC MC* aMA H& XX* *X,t MT 2* * * * * A * x* ** * xx * * ** * *

132

FIG. 3. Nucleotide and deduced amino acid sequences of cpl (upper) and cp2 (lower). Sequences are aligned for maximum indicates identity. Periods are inserted to keep codons together.

in Fig. 1. Only after residue 27 do the pea leaf isozyme sequences diverge. In contrast Lebherz et al. (3) were unable to sequence spinach cytosolic aldolase. The fact that the sequences do diverge after residue 27 indicates that in the present experiments two different isozymes were sequenced and that the pea leaf cytosolic aldolase does not have a blocked N-terminus. Cloning and sequencing of the pea chloroplast cDNAs. We first screened a pea leaf cDNA library in Xgtll(13) with antibody to the cytosolic aldolase isozyme and isolated 38 immunopositive clones. These were mixed

homology. Asterisk

and rescreened with a degenerate oligonucleotide probe designed to match part of the N-terminal sequence of the pea leaf aldolases (Fig. 1). Two of these clones, which gave different EcoRI restriction patterns (Fig. 2) and which were long enough to code for the pea leaf chloroplast aldolase, were sequenced (Fig. 3). Both cDNAs clearly code for Class I aldolases (comparisons not shown). Neither has an initial methionine codon and therefore both are probably truncated within the coding region. The sequence of the three amino acids prior to the initial glycine of the mature protein coded by cpl, Ile-Arg-Ala, matches

PEA CHLOROPLAST V GTA *9ci'; 5:

1F 1 TTT 2 **9
t ?;A* ¶t;t;t x$69; AfcJ, icfcih ic*ic &Aft i&T Jr** Aht it%at *Aat Xicxt h>tic ?ikR AA* Afcf< atAh IJ;;t

2

>t

;';

;t

IA V P I GCT GTC ccT 2 *>tA >?>t>t *AA 2n -'R 1L 1 TTG

N AAT

2 *

>t

2 n** ***

A

7t

Yt

A

G GGT it>t -'-

IM F L ATC ATG TTT TTG ic?;X 8AA ~ScA fcA>t A * A j;

A M N K S GCC ATG AAC AAA TCT

x*9< **J; *A* St

IL Q N I CTC CAA AAT

*

*

*** *

**A *

IE I GAG 2 *A* 2 A

A GCG A** "-

L L F CTC CTT TTC **j, *Jr* **>? * * j,

*** *

*

Jr

8

*

A

S Tee icAT fc

G GGT A>t?c *

G CGA *AR d

Q CAA *Ah 2

S E V E TCT GAG GTT GAA fc?i* Jthfc A>th Ati* * * 9; A

*

*

P N P W H V S F CCA AAC CCA TGG CAC GTG TCG TTC

it

?:

A

*

$1

A GCA xt?ci< ?;

T ACC ?c?;k ;';

L CTC **A *

N AAC ;t*>t A

272 836

A R A GCA AGA GCT

292 896

S Y TCA TAT

***

*7t* **St **Jc ***

***

*Ai< ***

***

n**

JCRJ,nnc it *

K AAG rtfch *

**A A

**it a

St** t** A Jr

*** 8

S TCC icicfc *

N AAC XAJ, JI

S L A TCA CTC GCT A** at*T stklt I * x

312 956

*A* nn* 9;9t >t * ?c

7ticA *** ?i *

>tA* rt

Q L G CAA CTT GGA 9~9~9~>tsCX >tkh 9, * x

Y Y TAC TAC atA* ATT ;I

G 332 GGT 1016 fcA* k

K AAG Afc3; A

1DGESEEAKKELF V KGYSY 1 GAT GGT GAA TCT GAG GAA GCC AAG AAG GAA TTG TT.T GT.C AAA GGC TAC TCT TAT 2 ?:* * :t ;1;?< ?cf"-I ,I Si:cfc >tf