Proc. Nati. Acad. Sci. USAVol. 88, pp. 10941-10945, December 1991Plant Biology

Molecular cloning of a putative plant endomembrane proteinresembling vertebrate protein disulfide-isomerase and aphosphatidylinositol-specific phospholipase C

(endoplasmic reticulum/Medicago sadva/plant cell culture/tunicamycin)

BASIL S. SHORROSH AND RICHARD A. DIXON*Plant Biology Division, The Samuel Roberts Noble Foundation, P.O. Box 2180, Ardmore, OK 73402

Communicated by J. E. Varner, September 16, 1991 (received for review November 9, 1990)

ABSTRACT cDNA clones containing sequence similarityto the multifunctional vertebrate protein disulfide-isomerase(PDI, EC 5.3.4.1) were isolated from an alfalfa (Medicagosaiva L.) cDNA library by screening with a cDNA sequenceencoding human PDI. The polypeptide encoded by a clonedesignated B2 consisted of 512 amino acids and was charac-terized by a 24-amino acid hydrophobic leader sequence, tworegions with absolute identity to the vertebrate PDI active site(Ala-Pro-Trp-Cys-Gly-His-Cys-Lys), and a C-terminal endo-plasmic reticulum retention signal (Lys-Asp-Glu-Leu). Theoverall identity of the B2 sequence to that of human PDI was35% at the amino acid level (79% when conservative substi-tutions were included) and 39% at the nucleotide level; thisincluded homology between B2 and the region of human PDIbelieved to be involved in binding estrogens. The deducedamino acid sequence of B2 was also 35% identical to that of arat form I phosphatidylinositol-specific phospholipase C. Ly-sates from Escherichia coli cells harboring an expression plas-mid bearing the B2 sequence contained significantly elevatedlevels of PDI activity. Southern analysis indicated the presenceof a small PDI-related gene family in alfalfa, of which B2appeared to correspond to a single gene. An -2-kilobase B2transcript was expressed in all alfalfa organs tested. In alfalfacell suspension cultures, B2 transcripts were strongly inducedby tunicamycin but not by exposure to fungal elicitor.

Protein disulfide-isomerase (PDI, EC 5.3.4.1) catalyzes for-mation of the disulfide bonds that stabilize the tertiary andquaternary structures of many extracellular proteins (1, 2).PDI is localized in the endoplasmic reticulum of plant (3) andvertebrate (4) cells, where it appears to be loosely associatedwith the lumenal surface. The vertebrate enzyme is a dimerconsisting of two identical subunits of -57 kDa, for whichcDNA and genomic sequences have been obtained (5-7). Itcontains two nearly identical, putative active-site regionscontaining the consensus sequence Ala-Pro-Trp-Cys-Gly-His-Cys-Lys, a sequence closely related to the active site ofthe thioredoxins (5). PDI is identical to a number of otherendomembrane proteins (or components of them): the j8subunit of prolyl 4-hydroxylase (PH) (6, 7), the thyroid-hormone-binding protein (8), and the lower molecular weightsubunit of the microsomal triacylglycerol transfer proteincomplex (9). Its amino acid sequence is very similar to thatof the glycosylation site-binding protein (GSPB) of the oligo-saccharyltransferase complex (10). In addition, vertebratePDI shares substantial similarity in its amino acid sequence(including the two active sites) with rat phosphatidylinositol-specific phospholipase C (PI-PLC) (11) and contains tworegions with sequence similarity to hormone-binding do-

mains of the human estrogen receptor (12). Recently, aGSBP/PDI clone was isolated from yeast (13).Apart from its suggested endomembrane location (3), little

is known about plant PDI. The enzyme has been purifiedfrom the unicellular green alga Chlamydomonas reinhardtii(14), from which it appears to be a dimer of a 60-kDa subunitantigenically unrelated to the vertebrate enzyme. It has beenreported that human PDI sequences do not detect comple-mentary sequences in a C. reinhardtii cDNA library (14).Antibody crossreactivity studies have provided conflictingevidence ofa role for PDI in the Chlamydomonas PH (14, 15).PI-PLCs have been partially purified from plants (16), buttheir amino acid sequences, and therefore possible relation-ships to PDI, have yet to be determined.We here report the isolation of an alfalfa cDNA clonet

whose deduced amino acid sequence exhibits extensive sim-ilarity to vertebrate PDI and PI-PLC and which exhibits PDIactivity on expression in Escherichia coli. We describe thegenomic organization of alfalfa PDI-related sequences, thelevels of PDI-related transcripts in tissues of mature alfalfaplants, and changes in PDI-related transcript levels in alfalfacell suspension cultures following treatment with tunicamy-cin or fungal elicitor.

MATERIALS AND METHODS

Library Screening. A 546-base-pair (bp) internal Xho Ifragment from the human PDI (PH 1 subunit) cDNA (6) wasused to screen a AZAPII (Stratagene) cDNA expressionlibrary constructed from poly(A)+ RNA isolated from analfalfa cell suspension culture at 2, 3, and 4 hr after exposureto fungal elicitor. GeneScreenPlus nylon membranes con-taining plaque lifts were air-dried for 5-10 min, autoclaved at120°C for 3 min, and then washed at 42°C for 2 hr in 0.1%(wt/vol) SDS/1 M NaCl/1 mM EDTA/50 mM Tris-HCI, pH8.0. Prehybridization in the presence of 50% (vol/vol) form-amide was performed, according to standard procedures (17),for 6 hr at 37°C. The probe [labeled by random oligonucle-otide priming (18)] was then added and the membranes wereincubated overnight at 37°C. The membranes were washed at60°C in 2x standard saline citrate (SSC)/1% SDS and then atroom temperature in 0.lx SSC prior to autoradiography.Positive plaques were removed and phage were eluted andreplated twice, according to standard procedures (17).Template Isolation and DNA Sequence Analysis. The pBlue-

script vector was in vivo excised, in the presence ofthe helperphage R408 (Stratagene protocol), from AZAPII to yield a

Abbreviations: IPTG, isopropyl P-D-thiogalactopyranoside; PDI,protein disulfide-isomerase; PI-PLC, phosphatidylinositol-specificphospholipase C; PH, prolyl 4-hydroxylase; GSBP, glycosylationsite-binding protein.*To whom reprint requests should be addressed.tThe sequence reported in this paper has been deposited in theGenBank data base (accession no. M72335).

10941

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Janu

ary

5, 2

021

10942 Plant Biology: Shorrosh and Dixon

phagemid containing the cloned insert. After propagation inE. coli XL1-Blue cells, insert cDNA was subjected to double-stranded DNA sequencing using dideoxynucleotide chaintermination (19) with T7 DNA polymerase and pBluescriptuniversal primers. Subsequent sequencing reactions wereprimed with synthetic oligonucleotides. End deletions of thecDNA insert for sequencing were generated by mung beanexonuclease (Promega). Deduced amino acid sequences werealigned by the CLUSTAL program (20). A score of 8 or greaterin a Dayhoff PAM matrix of amino acid similarity was usedas a cutoff to indicate amino acids with similar physicochem-ical properties or encoded by codons having one base ex-change.Growth and Elicitation of Cell Cultures. Cell suspension

cultures of alfalfa (Medicago sativa L.) cv. Apollo wereinitiated and maintained in a modified Schenk and Hilde-brandt medium as described (21). Cells were treated withelicitor from the cell walls of Colletotrichum lindemuthianum(60 ug of glucose equivalents per ml of culture) (21) or withtunicamycin (10 pug/ml final concentration), frozen in liquidN2, and stored at -70'C.RNA Isolation and Northern Blot Analysis. Total RNA (15

,ug) was isolated from alfalfa cell suspension cultures or planttissues by the method of Logemann et al. (22), fractionatedin 1% agarose gels containing 2.2 M formaldehyde, andblotted onto nitrocellulose membranes. Blots were prehy-bridized for 6 hr at 420C in the presence of 50% formamide.Labeled probe was added and blots were incubated overnightat 420C. Blots were washed at 650C in 0.2x SSC/0.1% SDSand then in 0.2x SSC at room temperature prior to autora-diography.

Southern Blot Analysis. High molecular weight genomicDNA (17) from alfalfa leaves was digested with restrictionendonucleases, electrophoresed in a 0.7% agarose gel, andblotted onto a GeneScreenPlus nylon membrane. Prehybrid-ization and hybridization conditions were as described abovefor library screening, except that hybridization was at 42°C.The blot was washed at 65°C in 2 x SSC/1% SDS and then in0.1 x SSC at 65°C prior to autoradiography.Measurement of Transcript Levels by RNase Protection. A

subclone containing the first 342 bp of the 5' end of the B2cDNA insert in pBluescript was digested with Nhe I atposition 208 and then transcribed from the T7 promoter in thepresence of [a-[35S]thio]UTP. The radiolabeled antisenseRNA (134 bases) was used to detect B2-related transcripts inalfalfa by RNase protection (23).

Expression of B2 in E. coli. The B2 cDNA insert wasamplified by PCR in 10 mM Tris HCl, pH 8.3/50 mM KCl/3mM MgCl2/0.01% gelatin/100 ,uM each dNTP containing 0.1nmol of each primer (I, 5'-CGCGCCCATGGAGGAATCAT-CAACTGACGC-3'; II, KS primer in pBluescript), "500 ngof template DNA, and 2.5 units of Amplitaq DNA polymer-ase (Perkin-Elmer/Cetus) in a final volume of 100 ,u. Am-plification (30 cycles) was at 94°C for 1.5 min, 55°C for 2 min,and 72°C for 2 min, with the final 72°C incubation for 10 min.Amplified DNA was isolated by electrophoresis onto What-man 3MM paper, digested with Nco I/HindlI, and sub-cloned into the Nco I/HindIII sites of expression vectorpSE380 (24). The construct was used to transform E. colistrain SB221 (25), and the cells were grown overnight at 37°Cin 5 ml M9 minimal medium supplemented with 0.8% (wt/vol)glucose, 12 ,ug of thiamin per ml, 0.3 mM leucine, 0.2 mMtryptophan, 4 mM MgSO4, 0.2 mM CaCl2, and 50 ,ug ofampicillin per ml. After 3 hr growth in fresh medium (OD6N0.3), cells were incubated for 3 hr in the presence of 5 mMisopropyl 3-D-thiogalactopyranoside (IPTG), collected bycentrifugation at 40C, and lysed at room temperature for 10min in 400 p.l of 0.1 M sodium phosphate, pH 7.5/5 mMEDTA/0.1% lysozyme. Cell debris was removed by centrif-

ugation at 40C and the protein concentration of the superna-tant was determined by the Bradford assay (26).

Assay of PDI Activity. PDI activity was assayed by amodification of the glutathione: insulin transhydrogenaseassay (27). Each reaction mixture (200 14) contained 111.6 ILI(:600 tug of protein) of E. coli lysate, 0.1 M sodium phos-phate buffer (pH 7.5), 5 mM EDTA, 1 mM reduced glu-tathione, 50 pug of insulin, and 0.1 p.Ci (37 kBq) of porcineinsulin monoiodinated with 1251I at tyrosine A14 (371 .uCi/,ug;NEN). The mixture was incubated at 370C for 40 min, treatedwith 500 jul of 15% (wt/vol) trichloroacetic acid and 300 .ul of2.5% (wt/vol) bovine serum albumin, incubated on ice for 15min, and finally centrifuged at 14,000 x g for 30 min at 40C.Radioactivity in a 900-pl aliquot of the supernatant wasdetermined by y counting.

RESULTSThree hundred thousand plaques from an alfalfa cDNAlibrary were screened at moderate stringency with alabeled human PDI cDNA. Thirty positive clones wereidentified after two rounds of plaque purification, and threeof these clones (B2, L1, and G1) were subjected to full DNAsequence analysis. The 1770-bp insert in clone B2 consists ofa 1536-bp open reading frame, 6 bp of 5' untranslated region,and 228 bp of 3' untranslated region (Fig. 1). Five shortsequence motifs are repeated within the open reading frame,and there are also two pairs of inverted repeats. Although the

CCMCA 6

ATGGCGAAAAACGTTGCGATTTTCGGCTTATTG=---CTCTTCTTGTGTTGGTrCCTTCTCAGATCTTCG 76

CTGAGGMTCATCAACTGACGCTAAGGMTTTGTTCTTACATTGGATAACACTAA-TTCCATGACACTGT 146Rl

TAAGAAGCACGAtTTCATCGTCGTTGMTTCTAC | ACC AAGCTAGCCC

IR1CAGAGTATGAGAAGGCTGCTTCTATCTTGAGCACTCACGAGCCACCAGTTGTTTTGGCTAAAGTTGATGC

AGGAATGGTGGAAAGAATATTCAAGAATACAAAGGTCCCCGTGAAGCTGAAGGTATTGTTGAGTATTTGAIR2

AAAAACAAAGTGGCCCTGCATCCACAGAAATTAAATCTGCTGATGATGCGACCGCTTTTGTTGGTGACAASL

1R2CAAAGTTGTTATTGTCGGAGTTTTCCCTAATTTTCTGGTGAGGAGTACGATAACTTCATTGCATTAGCA

GAGAAGTTGCGTTCTGACTATGACTTTGCTCACACTTTGAATGCCAAACACCTTCCAAAGGGAGACTCAT

R3TGTAGAAGCTCTAGAGAAATTCATTGAAGAATCCAGTACCCCAATTGTGACTGTCTTCAACAATGAGCCT

IRV' IRV'AGCAATCACCCTTTTGTTGTCAAATTCTTTACTCTCCCAACGCAAAGGCTATGTGTCATCAACTTA

R4CTACCGAAGGTGCTGAATCTTTCAAAACAAAATACCATGAAGTTGCTGAGCAATACAAACAACAGGGAGT

TAGCTTCTTGTTGGAGATGTTGAGTCTAGTCAAGGTGCCTTCCAGTATTTTGGACTGAAGGAAGAACAA

GTACCTCTAATTATTATTCAGCATAATGATGGCAAGAAGTTTTTCAAACCCAATTTGGAACTTGATCAAC

TCCCAACTTGGTTGAAGGCATACAAGGATGGCAAGGTTGAACCATTTGTCAAGTCTGAACCTATTCCTGA

AACTAACAACGAGCCTGTTAAAGTGGTGGTTGGGCAAACTCTTGAGGACGTTGTTTTCAAGTCTGCGMGRl'

R4'AATGTTTTGATAGAGTTTTAT GCTCCTTGGTGTGGTCACTGCAAG CAGTTGGCTCCAATCTTGGATGA

R2'

CCAAGGAGACA 1 1T1GAT1GT1CCAAGCT1AT1CCAACCT1TGT1ACT1T1CABGGTGA@GAAGTGAAAACT1AT1CACR3'

AATACGACGGTGGTAGGACAAAGGAAGACATCATAGAATTCATTGAAAAGAACAAGGATAAAACTGGTGCR5 R5'

TGCTCAACAAGAAGTAGAACAACCAAAAGCTGCTGCTCAGCCAGAAGCAGAACAACCAAAAGATGAGCTTSL

TGAAAAGTTCCGCTTGGAGGATATCGGCACACAGTCATCTGCGGGCTTTACAACTCTTTTGTATCTCAGA

ATCAGAAGTTAGGAAATCTTAGTGCCAATCTATCTATTTTTGCGTTTCATTTTATCTTTTTGGTTTACTCPA

TAATGTATTACTGAATAATGTGAGTTTTGGCGGAGTTTAGTACTGGAACTTTTGTTTCTGTAAAATTTAA

PATCCGAGATAAATTAGTTG

214

284

354

424

494

564

634

704

774

844

914

984

1054

1124

1194

1262

1332

1402

1472

1542

1612

1682

1752

1770

FIG. 1. Nucleotide sequence of the alfalfa B2 cDNA insert. Twoactive-site regions (boxed), repeats (R), inverted repeats (IR), po-tential stem-loop regions (SL), and putative polyadenylylation sites(PA) are indicated. The 5' and 3' untranslated regions are underlined.

Proc. Natl. Acad. Sci. USA 88 (1991)

Dow

nloa

ded

by g

uest

on

Janu

ary

5, 2

021

Plant Biology: Shorrosh and Dixon

A.

B2 MIAKVVAIFGLLFSLLVLVPSQIFkESSTDAKEFVLTLDNTNFHDTVKKHDFI\V-EFYAP 602-PH MLRR -------- ALLCL-PWXALVRfAPEEEDHVLVLRKSNFAEALAAHKYPPVEFHAP 51

..~ ~ ~ ~ .T......**.**. *.*B2 WCGHCKKLAPEYEKAASILSTHEPPVVLAKVDANEEHNKDLASEN4'~KGFPTIKIFRNGG 120B-PH WCGHCKALAPEYAKAAGKLKAEGSEIRLAKVDATEE--SDLAQQYGVRGYPTIKFFRNGD 109

82 KNI-QEYKGPREAEGIVEYLKKQSGPASTEIKSADDATAFVGDNKVVIVGVFPKFSGEEY 1792-PH TASPKEYTAGREADDIVNWLKKRTGPAATTLPDGAAAESLVESSEVAVIGFFKDVESDSA 169

P2 rNFIALAEKLRSDYDFAHTLNAKHLPKGDSSVSGPVVRLFKPFDELFVDSKDFNVFA - - 2362-PH KQFLQAAEAI -DDIPFGITSNSDVFSKYQLDKDGVV- -LFKKFDE---GRNNFEGEVTKE 223

B2 - LEKFIEESSTPIVTVFNNEPSNHPFVVKFFNSPNAKAMLFINFTTEGAESFKTKYHEVA 2952-PH NLLDFIKHNQLPLVIEFTEQTAPKIFGGEI - ---KTHILLFLPKSVSDYDGKLSNFKTAA 279*.**...* * *..,.. * .. ... **. .,... .. .....

B2 EQYKQQGV-SFLVGDVESSQGAFQYFGLKEEQVPLI-IIQHNDGKKFFKPN--- LELDQL 3512-PH ESFKGKILFIFIDSDHTDNQRILEFFGLKKEECPAVRLITLEEEMTKYKPESEELTAERI 339

B2 PTWLKAYKDGKVEPFVKSEPIPET-NNEPVKVVVGQTLEDVVFKSAKNVLIEFYAPWCGH 4102-PH TEFCHRFLEGKIKPHLMSQELPEDWDKQPVKVLVGKNFEDVAFDEKKNVFVEFYAPWCGH 399

B2 CKQLAPILDEVAVSFQSDADVVIAKLDATANDIPTDTFDVQGYPTL-YFRSASGK-LSQY 4682-PH CKQLAPIWDKLGETYKDHENIVIAKMDSTANEV- -EAVKVHGFPTLGFFPASADRTVIDY 457

;@^A-Aw*A * ***** * *** * *A *** * *

B2 DGGRTKEDIIEFIEKNKDKTGAAQQEVEQPKAAAQPEAEQP ------ KDEL 5122-PH NGERTLDGFKKFLESGGQDGAGDVDDLEDLEEAEEPDMEEDDDQKAVKDEL 508

B.

B2

GI

PDI/C

2-PH

GSBP

dCK

PI - PLC

STD

B2

GI

PDI/C

2-PH

GSBP

iCK

PI-PLC

C.

B2

PI-PLC

PDI/C

D.

ER/H

PDI/H

B2

I V|\VE F Y A P W C G H C K|K|L A P E Y

A L V E F Y A P W C G H C K K L A P E Y

L L V E F Y A P W C G H C K A L A P E Y

P P V E FiIA P W C G H C K A L A P E Y

L A V E F Y A P W C G H C K A L A P Y

V L L E F Y A P W C G H C K PA P E Y

M L V E FF A P W C G H C K R A P E Y

S MJ;jW A P W C GFCKL I VI

V LII E F Y A P W C G H C K Q L A P I jL

VIL VE FP A P W C G H C K L A P IL

V4JV E F Y A P W C C H C K Q L A P I W

V FJE F Y A P W C G H C K Q L A P I W

I E F Y A P W C G H C K Q L A PMW

I L V|E F Y A P W C G H C KIL A P E Y

0LQ

H

365 F V K S E P I P E T N N E - P V K V V V

363 Y L K S E P I P E T N E G - P V K V V V

353 H L E/DL/RP/AE/GDW D K PVV K_ V1

305-349 NSLALSLTADQMVSA--LLDAEPPI--LYSEYDP---TRP-FSEASHMGLLTNI182-230 DDIIPFGITSNSDVFSKYQLDKDGVV- -LFKKFDE - - GRNNFEGEVTKENLLDE.9*1^2 .**S..T.. * ..P .V*V* ***L*** ....D. * . * ....

191-239 SDYDFAHTLHAKHLPKGDSSVSGPVVRLFKPFDELFVDSKDPNVEAL-EKFIEE

ER/H 350-392 ADRELVHMINWAKRVPGFVDLTLHDQVH--- LLECAWLEILMIGLV,s* ** ** * A***. *.*. * **s

PDI/H 120-163 AGREADDIVNWLKKRTGPAATTLPDGAAAESLVESS- -EVAVIGFF*s** ** ** *** * * * * * *

.... ... ...

B2 128-171 GPREAEGIVEYLICKQSGPASTEIKSADDATAFVGDN- -KVVIVGVF

FIG. 2. Amino acid similarities between alfalfa B2 and G1 poly-peptides and related eukaryotic sequences. (A) Alignment of theamino acid sequence deduced from the B2 cDNA insert with that ofhuman PDI [PH ,( subunit (3-PH)]. Identical amino acids in the twosequences are starred. Dots indicate pairs ofamino acid residues thathave a similarity score of .8 in a Dayhoff PAM matrix of amino

Proc. Natl. Acad. Sci. USA 88 (1991) 10943

B2 cDNA insert does not contain a poly(A) tail, two putativepolyadenylylation sites are located in the 3' untranslatedregion. Clone Li appears to encode an allelic variant of B2.The deduced amino acid sequence ofB2 exhibits an overall

identity of 35% to that of the human PH f8-subunit/PDImultifunctional polypeptide (Fig. 2A). Particularly striking isthe sequence identity of B2, vertebrate PDIs, chicken GSPB,rat PI-PLC, and the endomembrane protein ERp72 (28, 29) inthe two putative active-site regions (Fig. 2B). A singleactive-site region of close similarity to B2 is found in thelower molecular weight thioredoxins (Fig. 2B), but the alfalfaB2 sequences are clearly more characteristic of PDI than ofthioredoxin. Outside the active-site regions, alfalfa B2 andhuman PDI share less sequence identity, although bothcontain hydrophobic N-terminal signal peptides predicted bythe Von Heijne method (30), and terminate in the Lys-Asp-Glu-Leu endoplasmic reticulum retention sequence (31). Thesequence Leu-Phe-Ser-Leu-Leu-Val-Leu-Val in the pre-dicted B2 signal peptide is also found in the signal peptides ofpotato endochitinase (32) and rat pancreatic RNase (33).Alfalfa B2 contains a potential N-glycosylation site (Asn-Phe-Thr) at amino acid 278.Rat PI-PLC exhibits strong sequence similarity to human

PDI, especially with respect to the two active-site regions(11), and the deduced alfalfa B2 polypeptide shows an overallamino acid identity of 35% to the phospholipase (full se-quence comparisons not shown). Interestingly, the B2 se-quence contains a region of9 amino acids (starting at position365) which is identical to a correspondingly placed sequencein rat PI-PLC but which is not conserved in human, bovine,chicken, or rat PDIs (Fig. 2C).

Attention has been drawn to the similarity of two regionswithin human PDI to the hormone-binding domains of thehuman estrogen receptor (12). The amino acid sequencesimilarity in these regions is shown in Fig. 2D. Although thecorresponding regions in alfalfa B2 are less similar to theestrogen receptor sequences than is human PDI, alfalfa B2nevertheless exhibits regions of identity to the human PDIthroughout these regions.The deduced amino acid sequence of the alfalfa cDNA

clone G1 also contains the two characteristic Ala-Pro-Trp-Cys-Gly-His-Cys-Lys active-site regions (Fig. 2B), and ahydrophobic N-terminal signal peptide. However, G1 resem-bles the mammalian endomembrane protein ERp72 (28, 29)more closely than it does PDI. Details of the sequence andexpression of G1 will be presented elsewhere.

Expression of the B2 sequence in the vector pSE380 in E.coli cells resulted in increased PDI activity. This activity wasinducible by IPTOG, and was similar to the IPTG-inducedactivity in E. coli cells harboring an expression vectorcontaining a mouse PDI sequence. In the absence of IPTG,leaky expression of B2 led to an increase of 18,200 cpm in

acid similarity. Conserved amino acids in human, bovine, chicken,and rat PDIs are underlined on the human (3-subunit sequence. Apotential glycosylation site in the alfalfa B2 polypeptide is markedwith a plus sign. Arrows mark the proposed sites of cleavage of thesignal peptides. Gaps are indicated by dashed lines. An unidentifiedamino acid in the human sequence is represented by X. (B) Alignmentof the proposed active-site regions (a and a') of alfalfa B2 and G1polypeptides with related eukaryotic sequences. The sequences arethe PDI consensus [human, bovine, chicken, and rat (PDI/C)], the( subunit of human PH (p-PH), chicken GSBP, human ERp72(initially described as deoxycytidine kinase; ref. 28) (dCK), rat formI PI-PLC, and spinach thioredoxin (STD). (C) Sequence similarity ofa region of alfalfa B2, rat PI-PLC, and the PDI consensus (PDI/C).Invariant amino acids in human, bovine, chicken, and rat PDIs areunderlined. (D) Sequence similarity of regions of alfalfa B2, humanPDI (PDI/H), and the hormone-binding domain of the human estro-gen receptor (ER/H).

F

E

Dow

nloa

ded

by g

uest

on

Janu

ary

5, 2

021

10944 Plant Biology: Shorrosh and Dixon

Ekb8.1 -

6.1

H

a-

P X E H P X

ofs

4.1 -

3.1 - a-so

ON

a

0o

C

CD

c

t:a)

2.0 -

1.6-

-p

1.0 -

a.

0.5 -

FIG. 3. Genomic organization of alfalfa B2 sequences. Southernblots of alfalfa lepfgenomic DNA (10 .tg) digested with EcoRI (lanesE), HindIII (lanes H), Pst I (lanes P), orXba I (lanes X) were probedwith the 207-bp EcoRV fragment from the 3' untranslated region ofB2 (Left) or the 41.8-kilobase (kb) B2 cDNA (Right). DNA molec-ular size markers are indicated.

solubilized 125I insulin A chains per mg of lysate proteinabove that in the lysate from cells harboring expressionvector alone. This value increased to 47,600 cpm/mg withlysates from ceilsincubated for the same time in the presenceof IPTG. Corresponding values for expression of mouse PDIin the same E. coli cells were 8600 cpm/mg (without IPTG)and 49,400 cpm/mg (with IPTG). SDS/PAGE analysis of E.coli lysates confirmed the appearance of a new 52-kDaprotein (B2 minus signal sequence) in IPTG-treated cellsharboring the B2 expression plasmid (data not shown).

Southern blot analysis indicated that the B2 cDNA inserthybridized to probably no more than two related genes in the

U,a))

4.) C.)

0 D)L CL

CD CD (n ~U)> 0E°E Oa> CD 0

-j a_ En cc

0

a)

z

< --E(h) e +E(h) Q a

zr- ---1 0

c 0 2 4 6 8 10 12 24 c 0 2 4 6 8 10 12 24 Cl &

- 4t asl..J

IQ

I1( 12 4 a

Time (h)

FIG. 5. RNase protection analysis ofB2 transcripts in control (o)and Colletotrichum elicitor-treated (o) alfalfa cell suspension cul-tures. The normalized graphical representation of the data wasobtained by scanning the autoradiograph with an Ambis radiodetec-tor; the autoradiograph underestimates the signal for the 10-, 12-, and24-hr elicited (+E) samples.

alfalfa genome (Fig. 3 Right). Probing of the same blots witha 207-bp probe from the 3' end of B2, which does not sharesequence similarity with the Gi clone, revealed that B2 isitself probably encoded by a single gene (Fig. 3 Left).B2 transcripts (2 kb) were constitutively expressed in all

mature alfalfa tissues tested, with highest transcript levelsfound in flowers (Fig. 4). Similar results were observed whenblots were probed with a cDNA segment specific for the 3'region of B2.

In view of the induction ofPH activity by fungal elicitor inalfalfa cell suspension cultures (data not shown), and the factthat B2 was isolated from acDNA library containing elicitor-induced sequences, we investigated whether exposure ofcultures to elicitor from C. lindemuthianum resulted in in-creased B2 transcript levels. RNase protection analysis (Fig.5) indicated a possible perturbation of B2 transcript levels asa function of time in elicited cells compared with unelicitedcells but did not show any major induction above controllevels. In contrast, B2 transcripts were strongly induced onexposure of the cells to the protein glycosylation inhibitortunicamycin (Fig. 6), a treatment which induces PDI tran-scripts in Chinese hamster ovary cells (34).

DISCUSSIONWe have identified a small gene family in alfalfa encodingproteins closely related to vertebrate PDI and PI-PLC. Oftheclones isolated to date, B2 is the most likely candidate toencode the alfalfa homolog of PDI in view of the followingfeatures: (i) its overall sequence similarity to PDI [the de-duced amino acid sequence of B2 is more similar to verte-brate PDI than the recently cloned yeast GSPB/PDI (13)],including the presence oftwo active-site regions separated by-340 amino acids; (ii) its expression ofPDI activity in E. coli,

0.24-

FIG. 4. Northern blot analysis of B2 transcripts in tissues ofmature alfalfa plants. TotalRNA (15 Ag per lane) from various tissuesof plants of the same age was fractionated in a 1% agarose/formaldehyde gel, transferred to nitrocellulose, and probed with thefull-length B2 cDNA insert. RNA molecular size markers are indi-cated.

Oh 2h 4.6h 8r 2AF---- - lI F-m+ + -e ++

FIG. 6. Northern blot analysis of B2 transcript levels in alfalfacell suspension cultures treated without (-) or with (+) tunicamycin(10 ,ug/ml) for 0-24 hr.

kb4.4-

2.4-

1.4-

Proc. Natl. Acad. Sci. USA 88 (1991)

Dow

nloa

ded

by g

uest

on

Janu

ary

5, 2

021

Proc. Natl. Acad. Sci. USA 88 (1991) 10945

(iii) its apparent molecular weight of 54,509 and predicted pIof 4.68; (iv) its highly acidic C terminus ending in theLys-Asp-Glu-Leu endoplasmic reticulum retention signal,and (v) induction of its transcripts by tunicamycin. Of the 30positive clones identified by our screening strategy, noneencoded a polypeptide more likely to be a candidate for thealfalfa homolog of PDI than B2. Restriction analysis and/orsequencing from their 5' and 3' ends suggested that the otherclones represented either B2, G1, or Li (data not shown).The presence of a region of identity to rat PI-PLC (Lys-

Ser-Glu-Pro-Ile-Pro-Glu-Thr-Asn) which is, however, notconserved in vertebrate or yeast PDIs, suggests the intriguingpossibility that B2 may be a multifunctional PDI/PI-PLC.PI-PLCs from plants so far appear to be associated predom-inantly with the plasma membrane (16).A number of polypeptides containing active-site sequences

of the type (Trp/His)-Cys-Gly-(His/Pro/Lys)-Cys-(Lys/Arg/Asp) exhibit PDI activity. These include PDI itself (themost active), thioredoxin, and the f8 subunits of the gona-dotropic hormones lutropin and follitropin (35). The posses-sion of PDI activity is therefore predictable for a protein ofthe deduced amino acid sequence of B2.Human PDI is selectively inhibited by estrogens, and it has

been proposed that the estrogen receptor sequences in PDIarose from the hormone receptor gene by exon shuffling (12).Although the region of B2 including amino acids 128-171exhibits less sequence similarity to the human estrogenreceptor region than does the corresponding region in humanPDI, this region of the B2 sequence nevertheless showsconsiderable similarity to human PDI (16 out of 44 aminoacids identical overall, and 12 out of 21 amino acids identicalin the 128-148 region, with 4 conservative changes). A similarsituation exists for the other hormone-binding domain (Fig.2D). These findings either place doubt on the exon-shufflingmodel or suggest that steroid hormone receptor bindingregions may have, or may have had, some function in plants.

Unlike PH activity, B2 transcripts were not significantlyinduced in elicitor-treated alfalfa cells. Induction of PDItranscripts by tunicamycin has been proposed to be a re-sponse to the accumulation of underglycosylated and/orincorrectly folded proteins in the endoplasmic reticulum (34).Further studies are needed to investigate the role of B2 inprotein processing in alfalfa and its relation to PH.

We thank Dr. Taina Pihlajaniemi (Oula, Finland) for the gift of thehuman PH P-subunit cDNA clone, Dr. Karen Dalkin (Noble Foun-dation) for construction of the alfalfa cDNA library, Dr. MichaelGreen (St. Louis University Medical Center) for supplying murinePDI cDNA, Dr. Masayori Inouye (University of Medicine andDentistry of New Jersey) for supplying E. coli strain SB221, andAllyson Wilkins for preparation of the manuscript.

1. Freedman, R. B. (1987) Nature (London) 329, 196-198.2. Freedman, R. B. (1989) Cell 57, 1069-1072.3. Roden, L. T., Miflin, B. J. & Freedman, R. B. (1982) FEBS

Lett. 138, 121-124.

4. Andreae, M., Blankenstein, P., Zhang, Y.-H. & Robinson,D. G. (1988) Eur. J. Cell Biol. 47, 181-192.

5. Edman, J. C., Ellis, L., Blacher, R. W., Roth, R. A. & Rutter,W. J. (1985) Nature (London) 317, 267-270.

6. Pihlajaniemi, T., Helaakoski, T., Tasanen, K., Myllyla, R.,Huhtala, M.-L., Koivu, J. G. & Kivirikko, K. I. (1987) EMBOJ. 6, 643-649.

7. Parkkonen, T., Kivirikko, K. I. & Pihlajaniemi, T. (1988)Biochem. J. 256, 1005-1011.

8. Yamauchi, K., Yamamoto, T., Hayashi, H., Koya, S.,Takikawa, H., Toyoshima, K. & Horiuchi, R. (1987) Biochem.Biophys. Res. Commun. 146, 1485-1492.

9. Wetterau, J. R., Combs, K. A., Spinner, S. N. & Joiner, B. J.(1990) J. Biol. Chem. 265, 9800-9807.

10. Geetha-Habib, M., Noiva, R., Kaplan, H. A. & Lennarz, W. J.(1988) Cell 54, 1053-1060.

11. Bennett, C. F., Balcarek, J. M., Varrichio, A. & Crooke, S. T.(1988) Nature (London) 334, 268-270.

12. Tsibris, J. C. M., Hunt, L. T., Ballejo, G., Barker, W. C.,Toney, L. J. & Spellacy, W. N. (1989) J. Biol. Chem. 264,13967-13970.

13. LaMantia, M., Miura, T., Tachikawa, H., Kaplan, H. A.,Lennarz, W.-J. & Mizunaga, T. (1991) Proc. Natl. Acad. Sci.USA 88, 4453-4457.

14. Kaska, D. D., Kivirikko, K. I. & Myllyla, R. (1990) Biochem.J. 268, 63-68.

15. Kaska, D. D., Gunzler, V., Kivirikko, K. I. & Myllyla, R.(1987) Biochem. J. 241, 483-490.

16. Tate, B. F., Schaller, G. E., Sussman, M. R. & Crain, R. C.(1989) Plant Physiol. 91, 1275-1279.

17. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) MolecularCloning:A Laboratory Manual (Cold Spring Harbor Lab., ColdSpring Harbor, NY).

18. Feinberg, A. P. & Vogelstein, B. (1983) Anal. Biochem. 132,6-13.

19. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.Acad. Sci. USA 74, 5463-5467.

20. Higgins, D. G. & Sharp, P. M. (1988) Gene 73, 237-244.21. Dalkin, K., Edwards, R., Edington, B. & Dixon, R. A. (1990)

Plant Physiol. 92, 440-446.22. Logemann, J., Schell, J. & Willmitzer, L. (1987) Anal. Bio-

chem. 163, 16-20.23. Mavandad, M., Edwards, R., Liang, X., Lamb, C. J. & Dixon,

R. A. (1990) Plant Physiol. 94, 671-680.24. Brosius, J. (1989) DNA 8, 759-777.25. Coleman, J., Inukai, J. & Inouye, M. (1985) Cell 43, 351-360.26. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254.27. Carmichael, D. F., Morin, J. E. & Dixon, J. E. (1977) J. Biol.

Chem. 252, 7163-7167.28. Huang, S.-H., Tomich, J. M., Wu, H., Jong, A. & Holcenberg,

J. (1991) J. Biol. Chem. 266, 5353.29. Mazzarella, R. A., Srinivasan, M., Haugejorden, S. M. &

Green, M. (1990) J. Biol. Chem. 265, 1094-1101.30. Von-Heijne, G. (1986) Nucleic Acids Res. 14, 4683-4690.31. Munro, S. & Pelham, H. R. B. (1987) Cell 68, 899-907.32. Gaynor, J. J. (1988) Nucleic Acids Res. 16, 5210.33. McDonald, R. J., Story, S. J. & Swift, G. H. (1982) J. Biol.

Chem. 257, 14582-14585.34. Dorner, A. J., Wasley, L. C., Raney, P., Haugejorden, S. M.,

Green, M. & Kaufman, R. J. (1990) J. Biol. Chem. 265,22029-22034.

35. Hawkins, H. C., Blackburn, E. C. & Freedman, R. B. (1991)Biochem. J. 275, 349-353.

Plant Biology: Shorrosh and Dixon

Dow

nloa

ded

by g

uest

on

Janu

ary

5, 2

021