Plant Physiol. (1980) 66, 897-9020032-0889/80/66/0897/06/$00.50/0

Peptide Mapping Reveals Considerable Sequence Homologyamong the Three Polypeptide Subunits of Gi Storage Proteinfrom French Bean Seed1

Received for publication January 8, 1980 and in revised form June 30, 1980

YU MA, FREDRICK A. BLISS, AND TIMOTHY C. HALLDepartment of Horticulture, University of Wisconsin, Madison, Wisconsin 53706

ABSTRACT

The major storage protein, GI globulin, of bean (cv. Tendergreen) seedswas subjected to limited proteolysis with trypsin, chymotrypsin, papain,proteinase K, and protease V8 and to cleavage with cyanogen bromide and2-(2-nitrophenylsulfanyl)-3-methyl-3'bromoindolenine. Mapping ofpeptidesseparated from each of the three Gl subunits by polyacrylamide gelelectrophoresis revealed that many proteolytic cleavage sites were presentat similar positions on the subunits. Evidence was adduced that the Glsubunits are homologous in amino acid sequence for about 61% of theirlength. The remaining region (possibly COOH-terminal) of the subunitsappears to be heterologous, with the a subunit bearing an additionalmethionine residue.

Proteolysis, followed by separation of the fragments to yield adistinctive fingerprint, has been used in the characterization ofmany proteins, including those from seeds (8). The procedure forelectrophoretic separation of peptides obtained by partial prote-olysis on SDS gels introduced by Cleveland et aL (1) is alsoproving to be of broad application (10). It is valuable for analysisof radioactive polypeptides synthesized in vitro using a singleradioactive amino acid substrate because the fragments obtained,being larger than those obtained by limit digestion procedures, aremore likely to contain a labeled residue. This procedure is espe-cially useful for proteins that, like G 1 globulin, the major storageprotein of French beans (11, 12), consist of subunits which arevery difficult to separate except on SDS-containing gels.

Here, the application of the Cleveland (1) mapping techniquein the further characterization of the G I subunits is described andevidence that they contain extensive regions of homology is pro-vided.

MATERIALS AND METHODS

Plant Material and Gl Purification. Seeds of Phaseolus vulgarisL. cv. Tendergreen were purchased from Olds Seed Company,Madison, WI. Cotyledons were obtained by removing the seedcoat and embryonic axis; G I protein was isolated as described bySun et al. (18), resuspended in 0.5 M NaCl, and stored at -27 C.The protein concentration was measured by using freeze-dried G 1

'This work was supported by a grant from the Herman Frasch Foun-dation, National Science Foundation Grant PCM 78-11804, United StatesDepartment of Agriculture-Science and Education Administration Grants5901-0410-9-0357-0 and 5901-0410-8-0053-0, and by the Research Divi-sion, College of Agricultural and Life Sciences, University of Wisconsin.

as a standard; E18V' for 1 mg/ml 0.5 M NaCl = 0.515.Electrophoretic Procedures. Prior to electrophoresis, G 1 protein

in 0.5 M NaCl was dissociated by heating to 100 C for 3 min in abuffer containing 0.625 M Tris-HCl (pH 6.8), 2% SDS, 40osucrose, 1% 2-mercaptoethanol, 2 mm EDTA-Na2, and 0.01% (w/v) bromophenol blue. The final protein concentration was ad-justed to 10 mg/ml; typically, 40,g were loaded into each gel slot.The discontinuous SDS-PAGE2 system of Laemmli (9), as modi-fied by Ma and Bliss (11), was used; for 0.75 mm thick, 13% (w/v) acrylamide slab gels (16), an initial current of 7.5 mamp/gelwas used and this was increased to 25 mamp/gel after all sampleshad moved into the stacking gel. Following electrophoresis, pro-tein was stained with 0.1% (w/v) Coomassie brilliant blue R250(19); the G1 protein was seen to be resolved into three discretebands.For mapping individual peptides, two cycles of electrophoresis

were used. The samples were applied across the top of the stackinggel (not in individual lane slots) so that the stained G1 polypep-tides appeared as three parallel lines across the gel after electro-phoresis. Strips containing the individual subunits were excisedand four to six pieces (each 8 mm long) were inserted into thelane slots of a second gel after equilibration with buffer A [125mM Tris-HCl, 0.1% SDS, 1 mm EDTA-Na2, and 40%o sucrose (pH6.8)]. The use of several strips is necessitated by the effectivedilution of the protein sample on proteolytic digestion when asingle polypeptide band subsequently resolves into several peptidefragment bands. Accommodation of multiple gel slices was facil-itated by widening the stacking gel compartment to 1.5 mm, butleaving the width of the separating gel at 0.75 mm. Proteolyticdigestion was carried out essentially as described by Cleveland etal. (1): after protein from the gel slices and the protease (added ontop of the slices) had migrated to the interface between thestacking and separating gels, the current was turned off for 15min. The peptide digestion products were subsequently resolvedby further electrophoresis in the 8.5-cm long separation gel. Forchemical hydrolysis, the gel strips were soaked in appropriatesolutions and equilibrated with buffer A, and the digested peptideswere resolved in the normal manner. The equilibration step isessential (as is the application of similar numbers of gel slices toeach sample well) to prevent uneven banding patterns. Whenprotein standards are to be run in these second dimension gels,they must also be polymerized into gel strips, equilibrated, andapplied to the gel in a fashion identical to that for the sampleproteins.

Proteolytic Procedures. Papain and bovine pancreas a-chymo-trypsin were purchased from Sigma; protease V8 (Staphylococcus

2 Abbreviations: PAGE, polyacrylamide gel electrophoresis; BNPS-skatole, 2-(2-nitrophenylsulfanyl)-3-methyl-3'-bromoindolenine; kd, kilo-daltons; CNBr, cyanogen bromide.

897

MA, BLISS, AND HALL

aureus) was from Miles; trypsin treated with L-(tosylamido 2-phenyl) ethyl chloromethyl ketone (bovine pancreas) was fromWorthington, and proteinase K (Tritirachium album Limber) wasfrom E. Merck. CNBr was purchased from Aldrich Co. andBNPS-skatole was from Pierce Co., Rockford IL. For kineticexperiments, the proteins (50 ,ug) were digested in 20 ,ul solutionas described previously (6), and 8-,ul samples were removed atappropriate intervals and dissociated by heating to 100 C for 3min with 2,l solution containing 10o (v/v) 2-mercaptoethanoland 2% (w/v) SDS. For mapping of individual polypeptides, theoverlaying procedures of Cleveland et al. (1) were followed. ForCNBr cleavage, slices were cut from first-dimension gels andsoaked in 2 ml 70% (v/v) formic acid containing 2% (w/v) CNBrfor 48 h at room temperature in the dark. The soaked slices werewashed thoroughly with a large quantity of distilled H20 and thenequilibrated with buffer A. For BNPS-skatole cleavage, slicesfrom the first dimension gel were soaked with 3 ml 50%o (w/v)glacial acetic acid, 1 ml glacial acetic acid (containing 50 mmBNPS-skatole) was added, and the mixture was stirred at roomtemperature for 24 h. The slices then were washed and equilibratedas described for CNBr-treated material.

RESULTS

Kinetic Digestion. In agreement with the observation ofRomeroand Ryan (15), it was found here that Gl protein was rapidlydigested by trypsin in the absence of SDS; when SDS was presentin the reaction at levels similar to those within dissociating gels,the trypsin was inactivated in less than 1 min (Fig. 1). Chymo-trypsin generated a characteristic series of peptides from Glprotein, but it also was rapidly inactivated in the presence of SDS(Fig. 2A). Papain (Fig. 2B) and protease V8 (Fig. 2C) wererelatively resistant to SDS dissociation, the digestion to smallerpeptides continuing for over 30 min.From Figures 1 and 2, it can be inferred that each of the three

G 1 subunits contains sites for cleavage by trypsin, chymotrypsin,papain, and protease V8. This follows from the fact that theintensity of each of the stained bands representing the a, f,, andy subunits of GI decreased during proteolysis. Further, since thisdecrease in intensity occurred uniformly for each subunit (e.g.note the complete disappearance of all three polypeptides betweenI and 5 min in the absence of SDS in Fig. 1), it is likely that thereare similar numbers ofcleavage sites on each subunit and probablythat they are at similar locations along each of the polypeptidechains. The latter deduction is based on the fact that differentialpositions of the cleavage sites would result in differential accessi-bility for the protease (hence, differential susceptibility to cleav-

FIG. 1. Kinetic analysis of digestion of GI protein by trypsin. To the

left, the extensive hydrolysis in the absence of SDS is seen; the rapidinactivation of trypsin by a level of SDS similar to that present in a

denaturing acrylamide gel is shown to the right.

40

A~~~~~~~~ 4W^:5

.30 60 15024f.-..i

-

B

:Z Z S 6jiiB5~~~~~~~~~~~~~~~~~~~~~~4i

HS1_w w o a"S;~~~~~~~~~~C.

c _50.1 u5 e ; ;20ti330 60120 1

FIG. 2. Kinetic analysis of G 1 protein digestion by three proteolyticenzymes. Digestion is shown with chymotrypsin (A), with papain (B) andwith protease V8 (C). Undigested GI protein run in the same 16% gel isshown to the left of each panel, and samples of each proteolytic enzymeare shown to the right of each panel. The quantity of G I protein and ofenzyme added to each lane are indicated at the top of each panel.

898 Plant Physiol. Vol. 66, 1980

1^ t, - e ;

...ddf-__snF r

PEPTIDE MAPPING OF G1

age) and, consequently, varying intensities of staining for each ofthe G1 polypeptides and their degradation products throughoutthe kinetic analysis. Inspection of the peptide profiles generatedby proteolysis with three different enzymes (Figs. I and 2) revealsthat there is very little such variation in intensity, and, conse-quently, that the proteolytic sites are at similar locations. Fromthis, it can be inferred that a considerable degree of amino acidsequence homology exists for each of the polypeptides. Evidencefor homology between the G I subunits has also been obtained byhybrid-arrested translation (14) of G1 mRNA, both of the radio-active polypeptide bands corresponding to G I mRNA translationproducts being decreased when G1 mRNA is complexed withcloned G1 DNA (5).Mapping of Individual Subunits of Gl. The reproducibility

found in the kinetics of proteolytic digestion indicates that certainsites on the G 1 polypeptides are more susceptible or accessible tocleavage than are others. To compare the digestion of the a, f,and y polypeptides, they were separated by one cycle of electro-phoresis and then gel slices containing individual subunits wereinserted into a second gel for digestion by the Cleveland (1)procedure. Papain and protease V8 were chosen (Fig. 3) becausethey were effective in digesting G1 polypeptides, yet resistant toinactivation by SDS. An analogous procedure was followed forchemical cleavage of the G 1 polypeptides with BNPS-skatole andCNBr (Fig. 4).

At low concentrations of papain (Fig. 3A, 0.2 ,ug/gel slot), aseries of polypeptides was generated, particularly interesting beingthe doublet (marked with arrows) seen for each of the a, fi, and-y subunits. Because the doublets did not migrate the same distancein each lane, it followed that they have different mol wt. Analysisof these data required that the mol wt of the G 1 subunits, and ofthe fragments generated by partial proteolysis, be determined asaccurately as possible. When the G I polypeptides were subjectedto electrophoresis through acrylamide gels of different porosities,the apparent mol wt for each subunit changed slightly (Fig. 5A).The presence of sugar residues in these peptides (5) probablycontributed to this. Because all the Cleveland maps were obtainedwith 16% acrylamide gels, the apparent mol wt of the G I subunits(a, 51.0 kd; f, 48.0 kd; y, 45.5 kd) and of the standard proteins insuch a gel (Fig. 5B) were used to calculate the molecular sizes offragments obtained on proteolysis.The doublet of papain polypeptides (see arrows, Fig. 3A) were

found to have mol wt of 40.0 and 38.0 kd for the a subunit, 37.0and 35.0 kd for the I? subunit, and 34.5 and 32.5 kd for the ysubunit. These are represented diagrammatically by the dashedlines in Figure 6A. From that figure, it can be seen that cleavageat a common point, 11.0 kd from the left-hand end, would giverise to the upper component of the doublet for each subunit andanother homologous cleavage at 13.0 kd would yield the lowercomponent of the doublet. The "stepped" appearance of thedoublet then is rationMlized since cleavage at common points in ahomologous sequence will result in the generation of a series ofpeptides from each G1 subunit, each differing in size by the sameamount as do the native polypeptides. Similar stepped peptidefragments are shown in Figure 6, B to F, and a summary of thecommon cleavage points revealed by these analyses is presentedas Figure 7. No stepped appearance was apparent on digestionwith protease V8 (Fig. 3B), possibly reflecting the ability of thisenzyme to cleave within the putative heterologous region.

Detection of the homologous peptide fragments depends ontheir size since those of 12.0 to 10.0 kd or less do not resolve in the16% SDS gels. In the case of papain digestion, an 1 l.0-kd fragmentcould not be identified reliably, but the 13.0-kd fragment resultingfrom the cleavage giving the lower band of the doublet wasresolved. As the concentration (or, under favorable conditions,duration) ofdigestion was increased, more cleavages took place,resulting in the generation of additional peptide fragments, many

A

C By

B

PAPAIN0.2.g /Slot 2.0 Jig/ Slot

O B Y cL 3 YcdBy

PROTEASE V80.2 Jjg/SJot

Vrz ,VW

Enzyme-'- 4.ik, .-

24 5 kd --23.520 -19.5 --17 -

145 -

4NW.

. I'

c oL B y

FIG. 3. Digestion of individual G1 subunits by papain and proteaseV8. G 1 protein was electrophoretically separated into its componentsubunits; slices containing individual subunits were added to a second gel,overlayered with proteolytic enzyme, and subjected to a second cycle ofelectrophoretic separation as described in the text. The separation of thefragments resolved after papain hydrolysis is shown in Panel A. Twoconcentrations of papain were used; at a low concentration (0.2 jg papain/slot), high mol wt fragments were obtained, including a doublet ofpeptides(arrows) that show a stepped appearance in mol wt analogous to that ofthe undigested G I subunits. At a 10-fold higher concentration of papain,smaller peptides were generated and a similar peptide profile was obtainedfrom each each of the G I subunits. Panel B shows the pattern of peptidesobtained after digestion with protease V8.

being of similar length and thus migrating to similar positions ineach lane of the gel (Fig. 3A; papain concentration, 2 ,ug/slot).Peptides having similar lengths are diagrammatically mapped tothe left of each G1 polypeptide in the analyses of Figure 6, A toD. Unlike the case for the stepped peptides, it is impossible to besure that each of the homologous polypeptides actually retains theleft-hand terminus since more than one cleavage event may haveoccurred.

Plant Physiol. Vol. 66, 1980 899

MA, BLISS, AND HALL

BNPS- SKATOLE

A...

f.f

I(V)av-

x-

4..

S.~ ~ ~ 0C

50 -

*

47 -

44dL B y

.000

o0*1

10 1 3 16

Acrylamide (% .w/v)

CYANOGEN BROMIDE

0

x-'

3:1

W. ~ __

M, :!i#:

;ic

C d d..

Y

.4 .6 .8Rf

FIG. 4. Cleavage ofG I subunits by BNPS-skatole and CNBr. Slices ofgel containing G I subunits were subjected to chemical cleavage as denotedin the text; the resulting peptides were separated by a second cycle ofelectrophoresis. Panel A shows that BNPS-skatole released only one largefragment (f) from each of the intact (i) G I subunits; the pattern of CNBrfragments is shown in Panel B, lane C being a control containing uncleavedprotein.

DISCUSSION

The agents used are capable of cleaving polypeptides at a widerange of sites. Trypsin cleaves mainly at arginine and lysine;chymotrypsin cleaves preferentially at tryptophan, phenylalanine,and tyrosine and also at leucine, methionine, asparagine, andhistidine; papain has a wide specificity but cleaves mainly atarginine, lysine, and glycine sites (20). Protease V8 specificallycleaves peptide bonds on the COOH-terminal side of glutamicacid (7). Proteinase K cleaves at many sites, especially peptidebonds adjacent to the carboxylic group of aliphatic and aromaticamino acids (2). CNBr specifically degrades methionine residues(4), and BNPS-skatole is specific for tryptophan (3).A more extensive kinetic analysis might reveal the relative

location of the homologous regions inasmuch as the sum of

FIG. 5. Apparent mol wt of G 1 subunits. In Panel A, the valuesobtained for the a, ,B, and y subunits are plotted for three acrylamideconcentrations. In Panel B, the migration of the a, fi, and y subunits ofG I

protein in a 16% acrylamide gel are shown relative to the following mol wt

standards: bovine plasma albumin, 66 kd; ovalbumin, 45 kd; hog stomachmucosa pepsin, 34.7 kd; bovine pancreas trypsinogen, 24 kd; bovine milkf?-lactoglobulin, 18.4 kd; and egg-white lysozyme, 14.3 kd.

fragment lengths after each cleavage should equal the total poly-peptide length. The extent of homology indicated by the cleavagesites mapped from the left-hand side of Figure 6 is, therefore,conservative. Using common sites indicated by mapping thestepped peptides, the maximum extent of homology appears to beabout 31.0 kd (see Fig. 6B). An 18.0-kd fragment is generatedfrom only the a subunit by CNBr cleavage (Figs. 4B and 6D).This corresponds to 33.0 kd from the left-hand end and may wellsuggest that the transition from homologous to heterologous se-

quence occurs between 31.0 and 33.0 kd, taking 31 kd as the upperlimit to the homology between the three subunits of 61% for thea polypeptide (the homologous region representing 31.0 kd of thetotal 51.0 kd), 65% for the ,B subunit, and 68% for the y subunit.Amino acid analysis of G 1 protein revealed that there are only

AI..4 kd "--o

48kd~8d -.,

Cl..

(3

B

900 Plant Physiol. Vol. 66, 1980

5 3 A

i i1.

f

PEPTIDE MAPPING OF G 1

PAPAIN

0cc______ I-

... .. L1@**.ww ......s@*v

~

Ul= == = == = ====LO== = -

4i Cs'onCII

D CYANOGEN BROMIDE(0 cIq ~~~~C.) '., co

CV) y~~~UlOUl-~~~~~~

LOLO.(.--- - - -

S aureus V8

Li -A -&-A PM

\n 4 n LnI II\\11/ 1

....

0C0

N

U:

7---~~~~

LOU,i

BNPS -SKATOLE

PROTEINASE K

I ~ -A -& &-A N

-n \-A\-t/en -/-!\\\I/ 1//

co-M-~~~~~~~* - --------------

wvvX... ........ LO

id

CHYMOTRYPSIN

-tAO LO LOU, LOU

......................\II/.............. .'m------------------------------------------

LO

- - - - - - - - - - - - - -.......................

LO0LO"I' 1f _ _C4_

- - - - - - - - - -

01---

FIG. 6. Analysis of cleavage sites on the G I subunits. In each panel, the upper solid line denotes the a subunit (51 kd); the middle line, the subunit(48.0 kd); and the lower line, the -y subunit (45.5 kd). As discussed in the text, cleavage sites were arranged to maximize homology among the subunits.*.... and - - -, indicate major bands seen on electrophoresis of cleavage products;. (to the left) lie entirely within the constant or homologousregion; - - -, peptide sequences that extend into the heterologous region at the right. Numbers denote fragment lengths in kilodaltons. Panels A and Bare plotted from gels shown in Figure 3 A and B, and Panels C and D are from gels shown in Figure 4, A and B, respectively. The gels used to calculatedata of Panels E and F are not shown.

3 tryptophan residues/G monomer (17). The profile shown inFigure 4A indicates that there is 1 tryptophan residue/subunitand mapping analysis showed that it lies approximately 6 kd fromthe end of the homologous region (Fig. 6C). A value of approxi-mately 10 methionine residues was obtained by amino acid anal-ysis (17), and the mapping analysis of Figure 6D indicates thatthe a subunit contains 4 methionine residues, whereas the ,8 andy subunits each have 3 methionine residues.

It appears likely that the heterologous region of the G 1 subunitslies at one end of the amino acid sequence, although whether thisis NH2- or COOH-terminal has not been determined. Evidencefor regions of homology at the NH2 terminus of polypeptidescomprising the I IS protein from soybean seed has been obtainedfrom sequencing studies (13). Maps, such as those shown inFigures 6, C and D, and 7, can be used to locate methionine and

tryptophan residues within a polypeptide chain, and the use ofadditional procedures that cleave at specific amino acids wouldbe valuable in further characterizing polypeptide chains of un-

known amino acid sequence. It is estimated that errors of ±2 kdin positioning amino acid locations may arise as a result ofdifficulties in precisely determining apparent mol wt of the peptidebands after migration through the second dimension gel. Furthercomplications, common to any peptide mapping procedure, maybe caused by incomplete cleavage, or cleavage at residues otherthan those predicted. Nevertheless, because nucleic acid sequenc-ing is now much more rapid than amino acid sequencing andsince cloned mRNA sequences do not necessarily contain eitherthe 3' or 5' termini, maps such as those obtained here should beparticularly valuable in cases (like GI) where the NH2 and COOHtermini of the polypeptide are hard to sequence.

A

p

C

Plant Physiol. Vol. 66, 1980 901

F

MA, BLISS, AND HALL Plant Physiol. Vol. 66, 1980

IS-SK ATOLE C:(HYMOTRPPSIN V:s aeQs VS 5. HALLTC, SM SUN, BU BUCHBINDER, JW PYNE, FA BLISS,JD KEMP 1980 BeankNOGEN BROMIDE K: PROTEINASE p: PAPAIN seed globulin mRNA: translation, characterization, and its use as a probe

towards genetic engineering of crop plants. In CJ Leaver, ed, Genome Orga-nization and Expression in Plants. Plenum Press, New York, pp 259-272

6. HALL TC, Y MA, BU BUCHBINDER, JW PYNE, SM SUN, FA BLISS 1978 MessengerRNA forGI protein of French bean seeds: cell-free translation and product

P P M M C CKCVC characterization. Proc Natl Acad Sci USA 75: 3196-32007. HOUMARD J,GR DRAPEAU 1972 Staphylococcal protease: a proteolytic enzyme

'specific for glutamoyl bonds. Proc Natl Acad Sci USA 69: 3506-35078. JACKSON P, JM MILroN, D BOULTER 1967 Fingerprint patterns of the globulin

fraction obtained from seeds of various species of the Fabaceae. New Phytol66: 47-56

:S3?...1 . 9. LAEMMLI UK 1970 Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 227: 680-685

10. LAMB RA, PW CHOPPIN 1978 Determination by peptide mapping of the unique.____________________ .____'____.____*____.___ polypeptides in Sendai virions and infected cells. Virology 84: 469-478:I::.:::: - 11. MA Y. FA BLISS 1978 Seed proteins of common bean. Crop Sci 17: 431-437

12. McLEEs[ER RC, TC HALL, SM SUN, FA BLISS 1973 Comparison of globulin* - proteins from Phaseolus vulgaris with those from Vicia faba. Phytochemistry

A I l I 12: 85-9310 20 30 40 50 13. MOREIRA MA, MA HERMODSON, BA LARKINS, NC NIELSEN 1979 Partial char-

Mol. Wt.(x o3) acterization of the acetic and basic polypeptides of glycinin. J Biol Chem 254:9921-9926

nmary of cleavage sites on theG I subunits compiled from 14. PATERSON BM, BE ROBERrs, EL KuFF 1977 Structural gene identification and

Cleavage positions are shown only for those fragments that mapping by DNA-messenger RNA hybrid-arrested cell-free translation. ProcNatl Acad Sci USA 74: 4370-4374

utative heterologous (right-hand) region and, hence, are 15. ROMERO J, D RYAN 1978 Susceptibility of the major storage proteins of the bean,end of the native polypeptide chains, whereas fragments of Phaseolus vulgaris L., to in vitro enzymatic hydrolysis. J Agric Food Chem 26:

may either contain the homologous (left-hand) terminus or 784-788

m internal locations. 16. STUDIER FW 1973 Analysis of bacteriophage T, early RNAs and proteins on

slab gels. J Mol Biol 79: 237-248LITERATURE CITED 17. SUN SM 1974 Isolation and physical and chemical characterization of globulins

from Phaseolus vulgaris seed. PhD thesis. University of Wisconsin-MadisonDW, SG FISCHER, MW KIRSCHNER, UK LAEMMLI 1977 Peptide 18. SUN SM, TC HALL 1975 Solubility characteristics of globulins from Phaseolusty limited proteolysis in sodium dodecyl sulfate and analysis by gel seeds in regard to their isolation and characterization. J Agric Food Chem 23:

oresis. J Mol Biol 252: 1102-1106 184-189N HENNRICH, M KLOCKOw, H METz, HD ORFH, H LAND 1974 19. WEBER K, M OSBORN 1969 The reliability of molecular weight determinationsK from Tritirachium album Limber. Eur J Biochem 47: 91-97 by dodecyl sulfate polyacrylamide gel electrophoresis. J Biol Chem 244: 4406-1972 Modification of tryptophan with BNPS-skatole. Methods 4412

25: 419-423 20. WHITE A, P HANDLER, EL SMIrH 1968 Principles of Biochemistry, 4th Ed.

57 The cyanogen bromide reaction. Methods Enzymol 11: 238-255 McGraw-Hill, New York

902

TI BNP

M: cItA

M T

-I-

FIG. 7. SurnFig. 6, A to F.contain the pilocated at onesimilar lengthbe cleaved fror

1. CLEVELANDmapping belectropho

2. EBELING W,Proteinase

3. FONTANA AEnzymol 2

4. GROSS E 196