mature Inhibitor 11. The Inhibitor I1 sequence exhibits two domains, one domain having a trypsin inhibitory site and the other a chymotrypsin inhibitory site, ap-.
Vol. 260,No. 11. Issue of June 10,pp. 65614561,1985 Printed in U.S.A.
THEJOURNAL OF B~OLOGICAL CHEMISTRY
0 1985 by The American Society of Biological Chemists, Inc.
Wound-inducedProteinase Inhibitors from TomatoLeaves 11. THE cDNA-DEDUCED PRIMARY STRUCTURE OF PRE-INHIBITOR 11* (Received for publication, May 11,1984)
John S. Graham$, G. Pearce, J. Merryweatherg, Koiti Titanili, Lowell H. Ericssonli, and Clarence A. Ryan11 From the Institute of Biological Chemistry and Biochemistry/Biophysics Program, Washington State University, Pullman, Washington 99164, SChiron Corporation, Emeryville, California94608, and the VDepartment of Biochemistry, University of Washington, Seattle, Washington 98195
A cDNA containing the complete amino acid-coding7,800 with a specificity toward chymotrypsin-like enzymes region of wound-induced tomato Inhibitor I1 was con- (21), but it accumulates at a rate of only about half that of structed in the plasmid pUC9. The open reading frame Inhibitor I (8).Like Inhibitor I, Inhibitor I1 is synthesized as codes for 148 amino acids including a 25-amino acid a preprotein and is finally compartmentalized as a mature signal sequence preceding the N-terminal lysine of the protein in the centralvacuole of tomato leaf cells. mature Inhibitor11. The InhibitorI1 sequence exhibits In thiscommunication we present the complete nucleotide two domains, one domain having a trypsin inhibitory sequence of a cDNA, prepared from wound-induced mRNA, site and the other a chymotrypsin inhibitory site, ap- from which we deduce the complete amino acid sequence of parently evolved from a smaller gene by a process of pre-Inhibitor 11.This sequence reveals that Inhibitor Ievolved gene duplication and elongation. The amino acid sequence of tomato leaf Inhibitor I1 exhibits homology from a much smaller gene by duplication and elongation. The with two small proteinase inhibitors isolated from po- inhibitor exhibits amino acid sequence homology with two much smaller trypsin and chymotrypsin inhibitors previously tato tuber and an inhibitor from eggplant. The small potato tuber inhibitors are homoglous with 33 amino isolated from potato tubers andwith a small trypsin inhibitor acids of the N-terminal domain and 19 amino acids from eggplant. from the C-terminal domain. Two identical nucleotide EXPERIMENTAL PROCEDURES sequences of Inhibitor I1cDNA in the 3' noncoding region were present that were also found in an Inhib- Materials-Tomato plants (Solanum esculentum L. var. Bonnie Best) were grown in growth chambers with a 18-h day (150 X 10' itor I cDNA. These include an atypical polyadenylation signal, AATAAG, anda 10-base palindromic sequence, microeinstein/m2s, 31 'C) and a 6-h night (21 "C). All plants used for experimentation were at the 2-leaf stage (14 days after planting). CATTATAATG, for which no function is yet known.
Potato tuber Inhibitor I1 was prepared by the method of Bryant et al. (3). Rabbit anti-potato Inhibitor I1 antibodies were prepared as reported previously (4, 18). Tomato Inhibitor I1 was prepared from wounded leaves by the method of Plunkett et al. (21). All other As part of our studies of the wound-induced regulation of materials were obtained from previously described sources (7). expression of two proteinase inhibitorgenes in tomato plants, Amino Acid Sequence Determination-The amino sequence of the studies to determine the structure and organization of these N-terminal 75 residues (Lys-26 to Arg-100) of tomato Inhibitor I1 genes were undertaken. We have reported the isolation and was determined as described in the accompanying article (7). The characterization of a cDNA prepared from wound-induced procedures of Edman and Begg (5) programmed according to Brauer Inhibitor I mRNA (7)and have deduced the amino acid et al. (1)were used. Polybrene was used (26) and the products were identified by high performance liquid chromatography (2,6). sequence of Inhibitor I and of its processing after synthesis. Poly(A+) mRNA Isolation-Isolation of Poly(A+) mRNA was perTomato leaf Inhibitor I1 is a proteinase inhibitor that also formed as previously described (18).Fractions of Poly(A+) mRNA accumulates in tomatoleaf cells in response to severe wounds, enriched in Inhibitor 11 mRNA were obtained by sedimentation of such as those resulting from insect attacks (24). Inhibitor 11, Poly(A+) RNA on 10-30% sucrose gradients and identified by cella 12,300 M, protein, is a potent inhibitor of endopeptidases free translation and immunoprecipitation of individual fractions (19). Construction of cDNA Library-Methods used in the construction and has a dual specificity towards the pancreatic proteinases of the cDNA library were described in the accompanying article (7). trypsin and chymotrypsin (21). Although members of the Screening of Tomato cDNA Library for Inhibitor II SequencesInhibitor I1 family have been well characterized (21), a com- Fifteen hundred recombinant clones were initially screened as deplete amino acid sequence has not been reported. Inhibitor I1 scribed (7). A second screening of 130 wound-positive clones was accumulates in leaves coordinately with Inhibitor I, M , = performed utilizing a 17-nucleotide oligomer mixture prepared as previously described and corresponding to the 64 possible coding sequences of an amino acid sequence of tomato Inhibitor I1 including *This research was supported by National Science Foundation Grant PCM 8023285, United States Department of Agriculture Co- amino acid residues 35-40. The oligonucleotide probes were radiolaoperative States Research Service Competitive Grant 81-CRCR-1- beled and utilized for in situ colony hybridization (7) except that 06797, and in part by National Institutes of Health Grant GM-15731. incubation and washes were at 25 'C. This is Scientific Paper 6831, Project 1791, from the College of Positive Hybrid Select Translation-Wound-positive clones, seAgriculture Research Center, WashingtonState University. The costs lected by use of the synthetic oligonucleotide mixture, were divided of publication of this article were defrayed in part by the payment of into 5 groups of 10 clones each and hybrid select translations were page charges. This article must therefore be hereby marked "adver- performed on each group (7). Polyacrylamide gel electrophoresis of tisement" in accordance with 18U.S.C. Section 1734 solelyto indicate the translation products was accomplished by the method of Swank this fact. and Munkres (25). $ Washington State University Foundation Fellowship Awardee. DNA Sequence Determination-Selected restriction fragments of I1 To whom reprint requests should be addressed. pT2-47 were isolated, end labeled, processed, and sequenced as de-
6561
Tomato Inhibitor I I cDNA
6562
scrihed hv Maxam and Gilhert (15. 16). Fragments laheled with '*P on only one end were obtained hv cleavage with a second, asymmetrically cleaving restriction enzyme and sequenced directly. For dideoxy sequencing, the pT2-47 insert was restricted with HpaII. HpaII/ HindIll and HpnII/EcoRI subfragments were isolated as previously reported (7). The suhfragments werecloned into the AccI/HindIII and AccI/EcoRI sites of M13mp9 and M13mp8,respectively. A Snu3A/SnlI suhfragment of "2-47 was generated by cleavage with the appropriate restriction enzymes and suhcloned into the HarnHIl Snll site of M13mpll (Fig. 4). Conditions of cloning, transformation, propagation of MI.?, isolation of both replicative form and singlestranded DNA, and dideoxy sequencing were by puhlished procedures (17, 28). RESULTS AND DISCUSSION
A cDNA library of wound-induced tomato leaf mRNAs,
A B C D
Inh. IIPre
-18.4K -12.3 K -6.2 K
constructed in the plasmid pUC9, was screened for identification and selection of Inhibitor I1 cDNA inserts. Possible Inhibitor I1 cDNAs were identified among the 130 selected colonies using a 17-mer synthetic oligonucleotide probe mixture shown in Fig. 1. The probes were prepared (27) from all possible codons that code for amino acid residues 35-40 of tomato Inhibitor 11, obtained by automated Edman degradation of pure InhibitorI1 isolated from wounded tomato leaves. FIG. 2. Cell-freetranslation of mRNAselectedbyfilterThe end labeled oligomer hybridized strongly with 50 of the bound plasmid. Poly(A') R N A fromwounded tomato leaves was hybridizedwith Inhihitor 11-specific (pT2-47) cloned DNA. Hound 130 wound-induced cDNAclones (data not shown), suggesting that theprobe was nonspecificfor InhibitorI1 cDNA-contain- RNA translated in a rahbit reticulocyte lysate system in the presence of ["Slmethionine, and the laheled products electrophoresed in a ing colonies. Hybrid select translations were performed on SDS,12.5%polyacrylamide gel containing 8 M urea. Imnr A . no antithe 50 wound-induced clones which hybridized with the syn- inhibitor 11 added: ImnP R. anti-inhihitor I I added: l a n e (', total thetic probe in groups of 10 clonesto identify groups contain- translation products; Lane D,molecular weight markers detected by ing Inhihitor I1 cDNA inserts. One of the groups immunopre- Coomassiehlue staining, top to hottorn: ovalhumin, 43,000: n-chycipitated a hand which migrated on sodium dodecyl sulfate- motrypsinogen, 25.700; /+lactoglobulin, 18.400; lysozyme, 14.300; cypolyacrylamide gels at the position of pre-Inhibitor I1 (19). tochrome c, 12.300; hovine t m s i n inhihitor, 6,Z)O;insulin, 3.000. Hybrid select translations of individual clonesfrom this group revealed that a single clone was responsiblefor the bindingof Inhibitor I1 mRNA, as shown in Fig. 2. As additional proof that clone pT2-47 contained InhibitorI1 mRNA sequence, we purified plasmid DNA from this clone and blottedDNA from I clone pT2-47 in triplicate onto nitrocellulose filters, as well I00 200 300 400 500 600 as 4 additional clones whichhybridized with the synthetic FIG. 3. Relevant restriction endonuclease sites and neoligomer mixture in the original in situ colony hybridization quencing strategy used to establish the nucleotide sequence of experiments. The filter was challenged with radiolabeled oli- Inhibitor I1 cDIVA. Nucleotides are numhered from the 5' proximal gomer a t 25 "C overnight. A more stringent wash using 2 X Hindlll site. Arrouts indicate the extent of each sequence determiSSC (0.015 M sodium citrate, pH 7.2,0.15M sodium chloride) nation. Arroux originating from own circles represent 3'-end laheled fragments. Thearrow originating from the o p n s q u o r ~represents 5 ' at 28 "C showed that clone pT2-47 does contain the cDNA end laheled fragments. Arrows originating from rlosrd clrrlrs represequence that is complementary to the oligonucleotide mixsent dideoxy sequence determinations. The h r a p linr represents the ture (data not shown). coding region. The light line represents untranslnted regions. The DNA Sequence Analysis-The pT2-47 plasmid DNA was h o t c h d areas represent pUC9 DNA. purified and analyzed by restriction mapping. The cDNA insert was found to he about 710 base pairs in length and contained a complete coding sequence for pre-Inhibitor 11. A the derived amino acid sequence of the Inhibitor I1 protein are shown in Fig. 4. The cDNA insert is 711 bases in length. restriction map of the cDNA and the sequencing strategy employed are shown in Fig. 3. The nucleotide sequence and It has a 5' C tail of 23 bases that is a result of the cloning procedure. The open reading framebegins at position 72 and ends at position 497. A sequence of AATAAG is present 15 Protein --Lys-Gly-Cys-Asn-Tyr-Tyr-35 39 38 37 36 40 bases upstream from the polyadenylation site and is assigned as a putative polyadenylation signal. Although this sequence mRNA 5'-AA$GGN-TGg-AA:-TAg-TAi-3' deviates from most eukaryotic polyadenylation signals (USUally AATAAA), it is identical to, and in a similar position as, 17-mer those found in the zein gene from maize (10) andin Inhihitor mixture 3'-TT:-CCN-AC$TTt-AT$-AT-5' FIG. 1. Oligonucleotide mixture used in selecting Inhibitor I mRNA (7). Another possible DNA sequence of interest is the 10-base I1 cDNA clones.Amino acids35-40 of the mature InhihitorI1 (Figs. 5 and 6 ) were rhosen for oligonucleotide synthesis as they represented palindromic sequence XATTATAATG- which terminates 64 minimal codon redundancy.The DNA sequence corresponding to the base pairs upstream from the polyadenylation site (Fig. 4). I t mRNA is shown in the midd/e. Nucleotide symhols ahow and b ~ l o n ~ is identical to a sequence found in Inhibitor I cDNA ( 7 ) that the line indicate the amhipuous positions. The final 17-mer mixture is 72 hase pairsupstream from the polyadenylation site. ronsisted of'all 64 possihle sequences which could encode thepeptide. The third haseof Tyr 40 codon was not included in the final synthesis Although it is not known if this sequence has a function, the to minimize complexity. probability of the sequence occurring by chance a t similar I
Tomato Inhibitor II cDNA
6563
(~)GAccccAAAATTAAAAGAAGACAAGGT*CTIGT**TCAA 11
6"
40
Ah w I+,$ L Y S GLI UCI A W UCI AI^ Tvr ~ e uLeu 1 T A T C C A T C A T G GCT CTT C A C A A G GAA G?T A A T T T T GTC GCT T A C C T A C T A
me
PO
ay Pne Leu rVr&I h P A* GGA A T G T T T C T A T A T GTT,G$T GCC
L a
AI
d
c
110
mO L z A $ % m ~ r r~ 100
Cy,
Pro A~
~
q
ser ~
pro G
A,W
C v s Glv
~
A s n Leu G ~ Y Phr
h
k
on ~
L Y Ph ~
Ala ~
C v s Thr
A
~
L v r Ser ~
Leu
G
Phe Aon C I , ~ ~
GI
le
c
A
eo A
Pro
~
11' R O Thr ~
i,n
11-I1
~
~
~
~
~
~
m
~
~
~
:
l',l-l
:
A
~
k
AW
GI
:
/'I 1
He C n
~
~
CIS T h .
~
-
CVS
c
T h
i
T
i
.
n
I I'
~
:
A
~
" ,
Ala TVr ~ m
\2$ r
S
Thr
~
r
1 1 - 1 ,
~
h D
~
~
;,"GI%,
%
1 I A&CE&S srr A C A A T TGT TGCf?
I10
Gln
~
70
y+++GL T
~~~
:
I I'
GI"
A!
i
5
Gf*A
:
1'1 I
c $HR-#~$gh& -f
gA++&j$&&%
G % c q n ~ & ~ s
!Jc:
30
s rh,
?
50
40
0"
I/-I,
I./ 1.1
Leu
GTr
IO0
8a
g G + F s
I~Q MI
i
r
r
~
~
~
~
~
m
~
~
~
~
i
~
l
~
.
1'11-1
;
1.11
:
~
,
~
~
~
~
~
I 1'
t 1 0
1 ~ sA l a A m M e t Tvr Pro UCI M e t . A T C A I CCC A A G ? m A T x n A F n ' F t b R Pro
FIG.5. Comparison of amino acid sequences of tomato leaf Inhibitor I1 (TI-ZI)with small polypeptide inhibitors (PCZ-I) 5eo 540 sno and (PTI)isolated from potato tuber and with a trypsin inhibitor from eggplant (EP).Regions of amino acid identity ~,,~l~~ocAlCAAGllclGlGTlATC~GTAATTACTAA~TATCT~~~~AGAAlAGATAlCCC l A A iisolated Al e.0 >a are boxed. PCA, pyrrolidonecarboxylic acid Numbering on top of the FIG.4. Complete DNA sequence of the pT2-47 cDNA insert. amino acid sequences begins with the N-terminal amino acid found The numbering of amino acids appears on top. The numbering of in tomato pre-Inhibitor 11. Lower numbers are the amino acid posinucleotides appears underneath each line of sequence data. The tions numbered from the N-terminal amino acid of PCI-1, PTI, and presumptive "signal sequence" is not underlined. The amino acid EP. sequence of the mature Inhibitor I1 is shown by I I B . The putative polyadenylation site is shown by I~IIIIII~~IIII~~~~~. A 10-bp palindromic sequence is shown by 1HB. TGA C ~ ~ ~ A G A C l l G T C C A l C l T C l ~ ~ ~ l ~ ~
~:~:CAAAA~~AAG~AA~~AA~GTATGAAA~AAAAGGATGCACACT~A~A~AA~GACA~GCTAA~.~A~.~A,~AA
*,ti
hRO
locations in two distinct mRNAs, which show no other homology, is small and its presence in both Inhibitor I and I1 mRNAs suggests that itmay have some role in the expression of these two coordinately regulated inhibitor genes. Amino Acid Sequence of Inhibitor ZZ-The amino acid sequence of pre-Inhibitor I1 shows the presence of a 25-residue signal sequence. Previous research had shown that newly translated InhibitorI1 wasabout 3kDa larger than thenative protein that is sequestered in plant vacuoles (18), and the sequence in Fig. 4 confirms the existence of a precursor form. The 25-residue N-terminal extension, preceding the N-terminal Lys-Ala-Cys sequence of the matureInhibitor I1 is typical of other eucaryotic signal sequences (29). It possesses a central core region of 15 hydrophobic amino acids flanked by charged amino acids. We suspect that thesite of cleavage of the signal sequence is at the Ala-Lys peptide bond, since this site would produce the N-terminal lysine found in the mature Inhibitor I1 protein. Additionally, alanine is a favored residue for the signal peptide PI site of cleavage and theAlaLys bond is the only site in this region that satisfies the specificity requirements of known eucaryotic signal peptidases (20). The processed form of Inhibitor I1 contains 123 amino acid residues with a molecular weight of 13,500.This isslightly larger than the molecular weight reported by Plunkett et aE. (21) of 12,300. Inhibitor I1 shares considerable amino acid sequence identity with some very small species of inhibitors from potato tuber andthe eggplant (cf. Fig. 5). It is not known if the smaller inhibitions are post-translational productsof a pre-Inhibitor I1 moleculeor if the potato andeggplant possess genes for these small inhibitors. Analysis of the Inhibitor I1 genes and gene products in potato and eggplant tissue should help answer this question. A search for intermediate forms of Inhibitor I1 in these plants is in progress. Inspection of the amino acid sequence of Inhibitor I1 reveals an example of gene duplication-elongation of an ancestral gene resulting in proteinase Inhibitor I1 having two domains (Fig. 6). These two domains are located within amino acids 26-82 and 83-148 of pre-Inhibitor I1 (Fig. 5). Inhibitor I1 is known to be a potent inhibitor of both chymotrypsin and
83
- A ~ ~ t ~ ~ P h . - M ~ ~ ~ . p - P ~ ~ ~ I l ~ - A l ~ - ~ ~ - S ~ c - A ~ a ~ ~ ~
- s ~ r ~ ~ l e - r ~ ~ - P r ~ - n r - c l ~ ~ I h ~ - T h ~ ~ T h r f C l y - l ~ r - L y ~ - C l ~ ~ ~ ~ 148
~ C l y - L y ~ - A s p ~ ~ v ~ l f C ~ ~ ~ l ~ - C l ~ ~ l ~ - S ~ ~ - ~ p ~ l ~ ~ A l
FIG. 6. The two-domain structure of tomato Inhibitor I1 based on amino acid sequence. Numbering of amino acid begins with the first amino acid of the mature form of tomato Inhibitor 11. Regions of comparison include amino acids 1-57 and 58-123 of mature tomato Inhibitor I1 (residues 26-82 and 83-148 of the preInhibitor 11). Regions of amino acid identity between the two domains are boxed.
trypsin (21) and isconsidered to be a "double-headed" inhibitor (23). Early attempts to determine the amino acid sequence of Inhibitor I1 isolated from potato tubers resulted in only partial sequences (11).This may have been due to the internal repeats in the sequence (Figs. 4 and 6) and the presence of multiple forms of Inhibitor I1 in tubers (3). Inhibitor I1 also exhibits amino acid sequence homology with two small polypeptide inhibitors isolated from S. tuberosum tubers, that have been termed PCI-1 and PTI (9) and an inhibitor from the eggplant called EP (22). These polypeptides have inhibitory specificities towards either chymotrypsin or trypsin, but not both (9). The identities among the small inhibitors and tomato Inhibitor 11are extensive (Fig. 6). These identities overlap the two domains of Inhibitor 11; the amino acid sequence of PTI can be found within amino acid positions 25-106of maturetomatoInhibitor I1 (or 50-100 of preInhibitor 11) and the two inhibitors exhibit 73% amino acid identity in this region. A PCI-1 sequence is also located within amino acids 50-101 of pre-Inhibitor I1 and they have 75% amino acid identity. Inhibitor I1 and the eggplant inhibitor have 58% identity (22). These homologies also help explain the dual specificities of Inhibitor I1 toward trypsin and chymotrypsin. The reactive site (-Pl-Pl'-)of the eggplant has been established at residues 38 and 39 (numbered below the sequences) as Arg-Asn (22). This is consistent with the known
6564
Tomato Inhibitor11cDNA
mechanisms (13) and indicates that thePI-PI'sites of PCI-1 and PTI are Leu-Asn and Arg-Asn, respectively, again consistent with their chymotrypsin (PCI-1) and trypsin (PTI) specificities. If the active site residues are conserved at the comparable locations in Inhibitor 11, then residues 87 and 88, Phe-Asn, contain a reactive site that accounts for the chymotrypsin specificity of the inhibitor. The region which contains this site is repeated at residues 30 and 31 (Fig. 7) and contains an Arg-Glu. This site could account for the trypsin inhibitory activity of the molecule. A Glu at the PI' site has been found in reactive site sequences of avian ovomucoid third domains (12) and satisfies known mechanistic rules for inhibitor reactive sites at this position (13). In a comparison of the amino acid codons found in the open reading frames of tomato Inhibitor I(7) and tomato Inhibitor I1 cDNAs, a bias in codon usage for at least 12 amino acids seems to exist. For example, 16 of 19 glutamic acid residues within both inhibitors are encoded by the codon GAA while 12 of 14 asparagine residues are encoded by the codon AAT. This has not been generally found among codons from other plant families (14). The construction and charaterization of Inhibitor I1 cDNA has provided new information concerning the relationships between it and the coordinately regulated mRNA coding for Inhibitor I (7). The two tomato proteinase inhibitor mRNAs are now known to share a variant polyadenylation site and a conserved 3' untranslated 10-base palindromic region. The availability of cDNA probes for two distinct tomato proteinase inhibitors, which are induced to accumulate as a response to the wounding of tomato leaf tissue, will enable us to initiate a study of the regulation of mRNA biosynthesis, and more importantly, will provide the means for isolating and characterizing inhibitor genes from tomatoandpotato genomic libraries. Since the possibility exists that control regions responsible for the regulation of inhibitor expression are located near or within the inhibitor genes, a knowledge of these regions will be crucial in determining the biochemical basis for the regulation of inhibitor gene expression. In addition, theseregulatory regions can be utilized for the construction of chimeric plant genes responsive to environmental signals. Acknozuledgments-We wish to thank Dr. Thomas Okita for the gift of pUC9 and his helpful discussions during the course of this work, Ed Foxfor his technical advice, and Richard Hamlin for growing the plants. REFERENCES 1. Brauer, A. W., Margolies, M. N., and Haber, E. (1975) Biochemistry 14, 3029-3035 2. Bridgen, P. J., Cross, G.A.M., and Bridgen, J. (1979) Nature ( L o n d . ) 263, 613-614 3. Bryant, J., Green, T. R., Gurusaddaiah, T., and Ryan, C.A.
(1976) Biochemistry 1 5 , 3418-3424 4. Campbell, D. H., Garvey, J. S., Cremer, N. E., and Sussdorf, D. H. (1964) Methods in Zmmumbgy, pp. 114-129, W. A. Benjamin, Inc., New York 5. Edman, P., and Begg, G. (1967) Eur. J. Biochem. 1,80-91 6. Ericsson, L. H., Wade, R.D., Gagnon, J., McDonald, R.M., Granberg, R. R., and Walsh, K. A. (1977) in Solid Phase Methods in Protein Sequence Analysis (Previero, A., and Coletti-Previero, M. A., eds) pp. 137-142, Elsevier/North Holland, Amsterdam 7. Graham, J. S., Pearce, G., Merryweather, J., Titani, K., Ericsson, L. H., and Ryan, C. A. (1985) J. BWl. Chem. 2 6 0 , 6555-6560 8. Gustafson, G., and Ryan, C. A. (1976) J.Bwl. Chem. 251,70047010 9. Hass, M.G., Hermodson, M.A., Ryan, C.A., and Gentry, L. (1982) Biochemistry 2 1,752-756 10. Hu, N.-T., Peifer, M. A., Heidecker, G., Messing, J., and Rubenstein, I. (1982) EMBO J. 1 , 1337-1342 11. Iwasaki, T., Wada, J., Kiyohara, T., and Yoshikawa, M. (1977) J.Biochem. (Tokyo) 82,991-1004 12. Laskowski, M., Jr., Empie, M. W., Kato, I., Kohr, W. J., Arde, H. W., Bogard, W. C., Jr., Wever, E., Papmokos, E., Bode, W., and Huber, R. (1981) in Structural and Functional Aspects of Enzyme Catalysis (Eggerer, H., and Huber, R., ed) pp. 136-152, Springer-Verlag, Berlin 13. Laskowski, M., Jr., and Sealock, R. W. (1971) in The Enzymes (Boyer, P., ed) Vol. 3, p. 375, Acadmeic Press, New York 14. Lvcett. FEBS " . G. W.. Delaunev. A. J.. and Crov. R. R. D. (1983) ~,~ Lett. 153,43-46 15. Maxam. A.M.. and Gilbert., W. .(1977) , Proc. Natl. Acad. Sci. U. S. A. 74,560-564 16. Maxam, A.M., and Gilbert, W. (1980) Methods Enzymol. 6 5 , 499-560 17. Messing, J., Crea, R., and Seeburg, P. H. (1981) Nucleic Acids Res. 9,309-321 18. Nelson, C. E., and Ryan, C. A. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,1975-1979 19. Nelson, C . E., Walker-Simmons, M., Makus, D., Zuroske, G., Graham, J., and Ryan, C.A. (1983) in Plant Resistance to Insects (Heding, P. A., ed), ACS Symposium Series 208, pp. 103-122, American Chemical Society, Washington, D. C. 20. Perlman, D., and Halvorson, H. P. (1983) J. Mol. Biol. 167, 391409 21. Plunkett, G., Senear, D. F., Zuroske, G., and Ryan, C. A. (1982) Arch. Bwchem. Biophys. 213,463-472 22. Richardson, M. (1979) FEBS Lett. 1 0 4 , 322-326 23. Ryan, C. A. (1983) in The Biochemistry of Plants (Stumpf, P.K., and Conn, E. E., ed) Vol. 6, pp. 351-371, Academic Press, New York 24. Ryan, C. A. (1983) in Variable Plants and Herbivores in Natural and Managed Systems (Denno, R. F., and McClure, M. S., eds) pp. 43-60, Academic Press, New York 25. Swank, R. T., and Munkres, K. D. (1971) Anal. Biochem. 3 9 , 462-477 26. Tarr, G. E., Beecher, J. F., Bell, M., and McKean, D. J. (1978) Anal. Biochem. 84,622-627 27. Urdea, M. S., Merryweather, J. P., Mullenbach, G. T., Coit, D., Heberlein, U., Valenzuela, P., and Barr, P. J. (1983) Proc. Natl. Acad. Sci. U. S. A. 8 0 , 7461-7465 28. Vieira, J., and Messing, J. (1982) Gene (Amst.) 1 9 , 259-268 29. Von Heijne, G. (1983) Eur. J.Biochem. 1 3 3 , 17-21 "
I
_
I
