CLEAVAGE O F HISTIDYL PEPTIDE BONDS
Sept. 2c), 1963
chair and boat forms of ring A are listed in Table V. These nonbonded interactions are considerably decreased in the planar forms. Planar l is more favorable than planar 2 since a stronger interaction between the 19-methyl and 4P-methyl groups is present in the latter case. TABLE V Conformation
I m p o r t a n t nonbonded interactions
Chair C Boat 1 B - l Boat 2 B-2
20-Br: 19-Me:@-Me CO: 19-Me:Za-Br:4a-Me 19-Me:4p-Me:Bm-Br:5m-H
We thus find that the conclusions based on dipole moment data and n.m.r. spectra are in substantial agreement with the results of conformational analysis. Each of these methods has important limitations and the results derived from a consideration of only one of these methods could be unreliable. Bond angle distortions of a few degrees can occur a t different points throughout the molecule, and consequently one cannot expect dipole moment or other physical measure[ CONTRIBUTIOS FROM
THE
2799
ments to correspond exactly to the calculated value. In the cases studied here, however, all of these methods essentially pointed to the same stereochemistry. These findings can therefore be accepted with confidence as a first approximation to the real state of affairs. Acknowledgment.-UTe are thankful to Professor A. Vystrril and Dr. J. Klinot for samples of allobetulone derivatives and to Professor G. O ~ r i s s o n , 'Dr. ~ J. M . Lehn, and Dr. 9.L. Allinger for valuable discussions. The computations were programmed by Dr. P. Funke and carried out a t the Computer Centerz0of Stevens Institute of Technology, for which we record our appreciation. This work was supported in part by a grant (G-13290) from the Kational Science Foundation and a grant (CX-05079) from the Cancer Institute of the U. S. Public Health Service. (19) I n a paper presented before the Second International Symposium on the.Chemistry of Katural Products, Prague, 1902, J . hf. Lehn, J . Levisalles, G. Ourisson, and P. Witz described their studies on triterpenes which are parallel t o ours. (20) Supported in part by a grant from the S a t i o n a l Science Foundation.
DEPARTMEST O F BIOPHYSICS, \ v E I Z M A N N INSTITUTE O F
SCIEXCE,
REHOVOTH, ISRAEL]
Cleavage of Histidyl Peptide Bonds by N-Bromosuccinimidel BY SHMUEL SHALTIEL'
AND
ABRAHAM PATCHORNIK
RECEIVED APRIL23. 1963
A nonenzymatic method for cleavage of C-histidyl peptide bonds was developed. The imidazole ring of histidine residues is oxidized by X-bromosuccinimide ( K B S ) in pyridine acetate buffers ( p H 3-4) a t room temperature. After destruction of excess S B S the reaction mixture is heated for 1 hr. a t looo. Yields of cleavage were found t o be 50-65% in histidyl dipeptide models and 30-55ch in several synthetic polypeptides consisting of 5-10 amino acids. The histidyl-prolyl bond in sperm-whale myoglobin was cleaved in 53yoyield. Tryptophyl and tyrosyl peptide bonds are also cleaved by S B S . By adjustment of reaction conditions and the use of masking groups, differentiation of cleavage of peptide bonds next to tryptophan, tyrosine, and histidine was achieved.
Introduction In the last few years a new approach to sequence analysis and fragmentation of proteins has been developed. The specific chemical reactivity of certain amino acid residues, permits cleavage of peptide bonds a t these sites by nonenzymatic reagents. In many cases, the amino acid residue is selectively modified prior to cleavage. This modification in itself is of potential value in studying the correlation between structure and biological function of proteins. Several nonenzymatic methods for selective cleavage of peptide bonds are reported in the literature. Some of these methods were successfully applied to polypeptides and protein^.^-^^ The work done in this field was recently reviewed by LVitkop.'j (1) Presented in part in the 1961 and 1963 Sfeetings of the Israel Chemical, Society ( S . Shaltiel and A. Patchornik, Bull Res. Council I s r a e l , lOA, 48 ( I Q F l ) ,and "Proceedings of the X X X I I Meeting of the Israel Chem Soc.," 1963, p. 19). (2) P a r t of a P h . D thesis t o be submitted t o the Hebrew University, Jerusalem (3) A. Patchornik, W B. Lawmon, E . Gross, a n d B. Witkop, J . A m . Chem. Soc , 82, ,5923 ( I Q i i O ) . (-1) I>. K Ramachandran a n d B. Witkop, ibid., 81, 4028 (1959). (,j)E . Gross a n d B. Witkop, J Bioi. Chem., 237, 1856 (1962). (6) A Patchornik a n d h f . Sokolovsky, "Proceedings of t h e Vth European Peptide Symposium," 1962, in press. (7) S Sarid and A. Patchornik, "Proceeiings of t h e XXXII Meeting of the Israel Chem. Soc ," 1963, p . 18 (8) F A Quiocho, 11. O'llell, and F. Friedberg, E x p e r i e n l i a , 17, 217 (1961). (9) T . Peters, Compl rend f ~ a vlab. . C a r k b e v g , 31, 227 (1939). ( I O ) L. K . Ramachandran, Biochim. Biophys. A d a , 41, 524 (1960); J . Sri and I n d . Resenvch, 2lc, 111 11902). (11) I Bernier and P. Jolles, Compl. r e n d . acad. sci., Paris, 263, 743 (1961). (12) G . F. Grannis, Avch. Biochem. Biophys., 91,255 (1960). (13) J Schultz, H Allison, and Lr.Grice, Biochem., 1, 694 (1962). ( 1 4 ) D. T. Gish, Biochim. Biophys A c l a , 3 5 , 557 (1959); J . A m . Chem! Soc., 82, 0329 (1900).
We wish to report here an oxidative method for cleavage of histidyl peptide bonds. No proteolytic enzyme is known to cleave these bonds selectively. Furthermore, histidyl residues are thought to be involved in the biological activity of several enzymes.Ifi Therefore, a selective method of cleavage may serve as a tool for the determination of sequence near the active sites of these proteins. Histidyl residues (I) have a double bond in the 7-6 position relative to the carbonyl group of the peptide bond. A similarly located double bond is found in tryptophyl (11), tyrosyl (111))and phenylalanyl (IV) residues. n
-NHCHC~
-NHCHC /
GH,
CHI
I
I1
-
"0
-NHCHC /
CHz
I11
-
$o
- NHCHC -
a0
/
CH2
IV
When brominating reagents, such as N-bromosuccinimide (NBS), N-bromoacetamide, or bromine, react with tryptophyl or tyrosyl residues, cleavage of their ~ ~ ~ mechanism ~ ~ ~ ~ sugC-peptide bond O C C U ~ S . The (15) B. Witkop in "Advances in Protein Chemistry," Vol. 16, C. B . Anfinsen, M. L. Anson, K . Bailey, a n d J. T. Edsall, E d , Academic Press, I n c . , Kew York, N . Y . , 1961, pp. 221-321. (IS) E . A. Barnard a n d W D. Stein in "Advances in Enzymology," Vol. X X , F. F. S o r d , Ed., Interscience Publishers, I n c , A-ew York, S . Y . , 1958, pp. 51-110. (17) G . Schmir, L. Cohen, and B. Witkop, J . A m . Chem. Soc., 81, 2228 (1959). (18) E . J. Corey and L. F. Haefele, i b i d . , 81,2225 (1959).
2800
SHMUEL SHALTIEL AND ABRAHAM PATCHORNIK
-
60
50 -
w0'
9b
Vol. 85
1
-
40
30-
0
ae
-
20
10-
2
I
3
.,
5
4
PH* Fig. 1.-Effect of pH on cleavage of P\'(a)-Z-histidyl-glgcine treated with 3 moles of NBS/mole: I , a t room temperature; 11, after heating for 1 hr. a t 100". The solvents for the reaction were prepared by adding pyridine to 25% aqueous acetic acid. T h e experiment at pH 1 (curve I ) was performed in 0.1 LY HCI.
gested3 l7 l 9 for this cleavage involves the formation of a five-membered imino-lactone ring (VII) which is readily hydrolyzed.
I
2
3
4
5
6
MOLES NBS/ MOLE PEPTIDE, Fig. 2.-Cleavage of X( a ) Z-histidyl glycine as a function of addition of XBS ( I ) at room temperature; (11) after heating for 1 hr. a t 100' (in pyridine-acetic acid-water, 1 : 10: 19, v . / v . , p H 3.3).
moles of NBS/mole of peptide brought about optimal cleavage (Fig. 2, curve I). Similar yields of cleavage (18-28%) were obtained when N(,,-carbobenzyloxyhistidyl-glycine (and other histidine-containing dipeptides) were cleaved in 50yoaqueous acetic acid (Table 1). T A B L EI YIELDSOF CLEAVAGE OF HISTIDYL PEPTIDES
VI Peptide
17+ RCONHCH-C~NHR~ /
+
\
CH,
0
\ C-C/ /
H
RCONHCH- C=O / \ --t CHz 0 HtO
1
\
-Br
\
/
/
C-
/
H C-Br \
VI1
Phenylalanyl peptide bonds undergo a similar cleavage20 after partial reduction of the benzene ring. Since imidazole and its derivatives are known to react readily with brominating reagents,21 an analogous reaction could be expected with histidine residues.
Results m'hen N(,)-carbobenzyloxy-histidyl-glycine was treated with NBS in acidic medium, glycine was liberated in low yield. I n order to find the optimal conditions for cleavage, we treated histidyl dipeptide models with various molar ratios of NBS in different solvents and hydrogen ion concentrations. Figure 1 (curve I) describes the pH-dependency of the cleavage when the reaction is performed a t room temperature. From this curve i t is evident that very low pH values are required in order to attain maximal yields of cleavage (20%). As for the amount of reagent required, we found that 3 (19) K Izumia, J I 3 Francis, A . V . Robertsun, and B Witkop, J A , n . C h e m Sor , 84, 1702 i l Y i i 2 ) (20) SI 1Vilchek and .4 Patchurnik, ibzif , 84, 4 l i l 9 (IOfi2). ( 2 1 ) I< S Schipper and A. R 1)ay in "Heterricyclic Compounds," Vol 3 , K C Elderfield. E d , John 1Viley and Sons, Inc , New York, N Y , 1 9 J 7 . p 207.
Cleavage product
Imidazole-3-propionyl-glycineGly K;ra,-Z-L-Histidyl-glycine?z Gly Nc,,-Z-L-Histidyl-glycine Gly-OEt OEt GI Y K;ra,-Z-L-Histidyl-L-alanine Ala S;ral-Z-L-Histidyl-L-leucine Leu S~,~-Z-L-Histidyl-i.-phenylalanine Phe Kca.,-Z-L-Histidyl-DL-phenylalanine P he S~o,-Z-L-Histidyl-L-phenylalanine amide Phe-NH? Kc,.,-Z-L-Histidyl-L. ( e - Z b lysine (c-Z)-Lys
+
---Cleavage, 50% AcOH 1 hr. Room t e m p b 100'
Yc-Pyr.-acet." Room 1 hr. temp.' 100°
24 20 17
63 59 55
3 2 0
65 ,5Y
21 23
65 GO
2 2
03 58
18
53
2
56
50
19
50
1
50
18
53
0
50
28
GO
3
a8
Yield Pyridine-acetic acid-water, 1 :10: 19 v . / v . ( p H 3.3). of cleavage was determined after 30 min. (see Experimental). a
An attempt was made to replace NBS by other brominating or oxidizing reagents. These included Nbromoacetamide, iodine, hydrogen peroxide, and performic acid. Imidazole-propionyl-glycine was treated with 3 moles/'mole of each of these reagents in 50yoacetic acid. The yields of cleavage were found to he essentially the same with NBS, N-bromoacetarnide, and bromine (247,, 22%, and l S % , respectively), whereas the other reagents cleaved the peptides to a negligible extent only (less than 5%). iVhen N(,)-carbobenzyloxy-histidyl-glycine (2-HisGlyOH)'? was treated with 3 moles,'mole of NBS in 50% aqueous acetic acid and the reaction mixture was allowed to stand a t room temperature, we observed that after the initial cleavage of 20yo (Fig. 3) the yield rose slowly with time and reached 50yo after G days. This ('22) Z = carbnbenzyloxy
Sept. 20, 1963
L,---ii
CLEAVAGE OF HISTfDYL P E P T I D E
60
30
90
min
.
2801
BONDS
-
120
12 hours.
24
120
TIME,
Fig. 3.-Release of glycine with time from N( 0 1 ) Z-histid~-l-glycine treated with 3 moles of SBS/mole in 50% aqueous acetic acid.
0
0.5
1.0
1.5
2.0
2.5
30
MOLES N B S / MOLE C B Z TRYPTOPHAN. Fig. 5.-Competition between tryptophan and histidine residues for R’BS a t pH 4.0: I , spectrophotometric “titration” of Z tryptophan with NBS; 11, “titration” of 2-tryptophan in t h e presencr of 5 moles of N(a)-2-histidine.
Lj
A-I
/A-A
I
I
0 30
90
I
150 T I M E (M I N).
I
I
I
210
Fig. 4.-Yield of cleavage of S (ol)-Z-histidyl-glycine (in aqueous AcOH) on heating the reaction mixture for different times: I, with 2 moles of SBS/mole of peptide; 11, with 2 moles of SBS/mole of peptide; 111, with 3 moles of SBS/mole of peptide.
slow rise in yield cannot be attributed to hydrolysis of the histidyl peptide since no free glycine was detected in a parallel control experiment (without NBS), Similar high yields of cleavage were obtained when, instead of the long incubation period, the reaction mixture was heated to 100” for 1 hr. (Fig. 4). By this procedure, yields up to 6576 were obtained with histidyl dipeptide models (Table I ) . The fact t h a t the reaction does not proceed to^completion (Fig. 4) is not due, in this case, to partial destruction by NBS23,24 of the released amino acid. On total hydrolysis (with 6 -VHC1) of aliquots from the same reaction mixture y e recovered 90-96y0 of the glycine. If excess KBS is used (more than 3 moles:’ mole of peptide), yields of cleavage are lowered (Fig. 2 ) owing to partial oxidation23of the free amino acid released. By destroying excess NBS (before heating) with formic acid,3sodium thiosulfate, or imidazole, this difficulty can be overcome without affecting the yields of cleavage. Heating of peptides or proteins in aqueous acetic acid is liable to cause some hydrolysis of other peptide bonds.25 However, when the cleavage was performed in a buffer consisting of pyridine, acetic acid, and water (1: 10: lSv.,:v.,pH 3.3),yieldswerenotlowered (TableI) and no hydrolysis was detected. (23) E . W. Chappelle a n d J Ll. Luck, J . B i d . Chem., 229, 171 (1957). (24) S . Konigsberg, G . Stevenson, and J. M. Luck, ibid., 236, 1341 (1960) (25) A. Yaron, A. Patchornik, a n d M . Sela; see ref. 15, p. 229.
Selectivity.-Since tryptophyl and tyrosyl peptide bonds are also cleaved by NBS3.I7in acidic media, it was necessary to find a way for differentiating between cleavage next to tryptophyl and tyrosyl residues and cleavage next to histidines. Among these residues, tryptophans react fastest with NBS. This reaction is accompanied by characteristic spectral changes. On adding NBS to carbobenzyloxytryptophan (2-Try) a decrease in absorption a t 280 mp is observed that reaches a minimum when the amount of NBS a4ded is 1.53moles/mole. On further addition of XBS there is a rise in the absorption (Fig. 5 , curve I). In order to determine the competition between tryptophyl and histidy1 residues for NBS, we followed the reaction between 2-Try and NBS spectrophotometrically, in the presence of increasing amounts of Y(,,-carbobenzyloxyhistidine26 (2-His). As shown in Fig 5 , the reaction between 2-Try and the first 1.5 moles of NBS is so fast that even 5 moles of 2-His do not interfere with i t However, the reaction with additional amounts of XBS is markedly retarded by the presence of 2-His. Tyrosyl residues react with IVBS more slowly than tryptophyl residues. The reaction between tyrosine and NBS can also be followed spectrophotometically by measuring the increase in optical density a t 260 mP.l7 N-Carbobenzyloxytyrosine (2-Tyr) was treated with NBS in the presence of increasing amounts of 2-His and it was found (Fig. 6) that histidyl residues interfere with this reaction to a great extent. An even greater competition of the histidyl residues for NBS has been observed in mixtures of tryptophyl, tyrosyl, and histidyl dipeptides as well as in oligopeptides containing these residues. Thus, the synthetic peptide His-Phe-Arg-Try-GlyOHZ7 consumed 1.9 moles of NBS (instead of 1.53 moles) for maximal decrease of indole absorbance a t 280 m p . Similarly, a tyrosine-histidine competition was observed in the pep(26) Histidyl residues or their oxidation products do not absorb a t this wave length and t h u s their presence (even in excess) does not interfere with this determination. (27) This peptide was k i n d l y supplied by P r o f . K . H o f m a n n , University of Pittsburgh.
2802
SHMUEL SHALTIEL AND ABRAHAM PATCHORNIK 20 I
I
I
I
I
I
I
I
TABLE I1
1
I
CLEAVAGE OF MIXTURES O F TRYPTOPHYL, TYROSYL, A N D HISTIDYL PEPTIDES~
Peptide mixtureb
a
0 5
Vol. 85
./*e--
I
1
I
05
IO
1.5
I 2.0
I 2.5
I 3.0
I
I
3.5
4.0
MOLES N B S I M O L E CBZ TYROSINE,
Fig. 6.-Competition between tyrosine and histidine residues for S B S a t pH 3.0. I , spectrophotometric "titration" of N Ztyrosine; 11, 111, and I\-, "titration" of S-Z-tyrosine in the presence of 1, 3 , and 5 moles (respectively) of S(m)-Z-histidine.
tide 2-Val-Tyr-Val-His-Pro-PheOMe.28 This peptide required 4.2 moles of NBS (instead of 3 moles) to reach the theoretical spectral changes a t 260 mp. I n this connection it should be mentioned that histidine residues with a protected carboxyl group differ in reactivity from histidine residues with a free carboxyl g r o ~ p . ~Thus ~ , ~ Z-His ~ absorbs 2 moles of NBS in about 2 min. and a third mole within 45 min., while a histidine residue in peptides, such as 2-His-Gly-OH or 2-His-Phe-XH2,absorbs 3 moles of NBS within 30 sec. For this reason, when we studied the competition between tryptophan, tyrosine, and histidine residues we always compared analogous models, ;.e. 2-Try (or 2-Tyr) was compared with 2-His, and tryptophyl (or tyrosyl) peptides were compared with histidyl peptides. In spite of the competition between tryptophyl, tyrosyl, and histidyl residues for NBS, i t is possible to differentiate between the cleavage of tryptophyl, tyrosyl, and histidyl peptide bonds. By performing the reaction with XBS a t room temperature in pyridine-acetic acidwater (1: 10:19 v. '%,), rupture of tryptophyl and tyrosyl linkages occurs, whereas histidyl residues are oxidized but not cleaved. Xfter destroying the excess NBS and heating the reaction mixture, cleavage of histidyl linkages occurs without increase in the yields of cleavage of the other peptide bonds. This was proved using equimolar mixtures of tryptophyl, tyrosyl, and histidyl dipeptides (Table 11). The extent of cleavage of each of the peptides could be determined by paper chromatography since the dipeptides were chosen so that a different amino acid was liberated from each of the peptides in the mixtures. Differentiation between cleavage next to tryptophyl and tyrosyl residues was also attempted. Cleavage of peptide bonds next to tyrosine can be prevented by 0-carbobenzyloxylation' or 0-acetylation of these residues. Thus, when 0-carbobenzyloxy tyrosine or N ,0-dicarbobenzyloxy-tyrosyl-glycine was treated with SBS, no increase in the optical density a t 260 mp 128) This peptide was kindly supplied by Dr. B. Iselin of C I B A , Ltd., Basle Io C.'. ref. 16, p . 58. 0 ) As determined by iodometric titrations; c ~ f . PA. Z. Barakat and 11. F. Abd El-\Vahab, A n d . Chem., 26, 1973 (1954).
NBS, moles
Bond cleaved
Cleavage,c (2 Room 1 hr. temp.d a t 1000
2-L-Tryptoph yl-glycine 6 N ( a )-L-Z-Histidyl-L-phenylanine
Tq-Gly His-Phe
41 2
44 40
Z-L-Tryptophyl-L-alanine X ( a )-2-L-Histidyl-glycine
6
Try-i\la His-Gly
52
2
53 55
?;-Z-L-Tvrosyl-glycine N(a)-Z-L-Histidyl-L-(e-carhobenzyloxy)-lysine
6
Tyr-Gly
42
45
N-Ac-~-Tyrosyl-~-alanine S ( a-Z-~-Histidyl-glycine )
6
2-L-Tr yptophyl-glycine Iv-hc-L-Tyrosyl-L-alanine X ( a )-Z-L-Histidyl-L-(e-carhohenzyloxy )-lysine Z-~-Tryptophyl-~-alanine Ij-Z-L-Tyrosyl-glycine S(*) -Z-L-Histidyl-L-leucine
10
9
His-Lys
3
50
Tyr-Ala His-Gly
41 1
44
50
T r y-Gly Tyr-Ala
43 $2
46 43
His-Lys
2
4i
Try-Ala Tyr-Gly His-Leu
55 48 1
53 40 55
In pyridine-acetic acid-water (1:10: 19, v . / v . , p H 3.3). Equimolar mixtures of the peptides were used. I-ield of cleavage was determined by paper chromatography of the reaction mixture and a quantitative ninhydrin assay (see Experiniental). Extent of cleavage was determined after standing for 30 min. a t room temperature.
(characteristic of lactone formation and of cleavageI7) could be detected. The above-mentioned peptide as well as N,O-dicarbobenzyloxy-tyrosyl-alanine benzyl ester were not cleaved by KBS (in pyridine-acetic acid-water, 1 : 10: 19, v., v.) even after heating for 1 hr. a t 100" in the same solvent. However, after removal of the 0-carbobenzyloxy group (by treatment with alkali (pH 13) for 5 min.), cleavage by NBS proceeded in 45-50yo yield. The masking and unmasking of tyrosyl residues was followed spectrophotometrically. 31 Cleavage in Oligopeptides and in Sperm-Whale Myoglobin.-The applicability of this method to larger molecules was tested by performing the cleavage on four synthetic polypeptides and one protein, spermwhale myoglobin.32 Yields of cleavage were found to be 30-5376 (see Table 111). The extent of cleavage was measured by three different methods. IVhen as a result of the cleavage a free amino acid was released, we used paper chromatography and quantitative ninhydrin assay (e.g., cleavage of the his-leu bond in Val"-Hypertensin I ) . T o estimate the cleavage of his-pro bonds we used the enzyme proline i m i n o - p e p t i d a ~ e ~(PIP) ~ to cleave the new K-terminal proline which was then determined c ~ l o r i m e t r i c a l l y . ~The ~ extent of cleavage of the his-phe bond in the peptide His-Phe-Xrg-Try-GlyOH was determined by quantitative dinitrophenylation34 (DNP). I n this case, the a-amine of histidine was acetylated before the cleavage to prevent possible side reactions between the a-amine and NBS or products of oxidized imidazole. Vsing our method of cleavage we were able to confirm the existance of a his-pro bond in sperm-whale myoglobin. The primary sequence of this protein was (31) S . Shaltiel a n d A . Patchornik, in preparation (32) A sample of this protein was kindly supplied by I)r J C Kendrow, Cavendish Laboratory. Cambridge, Eng. ( 3 3 ) S Sarid, A Berger. a n d E Katchaliki, J A d . Chem , 234, 1710 (195~); a37, 2207 ( 1 9 ~ 2 ) . (34) H . Fraenkel-Conrat, J. I . Harris, and A . 1,. L e v y in "hlethods of Biochemical Analysis," T'ol. 2 , Interscience Publisher\, Inc., S e w Y o r k , N . Y., 1955, pp. 339-425.
iw,
CLEAVAGE OF HISTIDYLPEPTIDE BONDS
Sept. 20, 1963
2803
TABLE 111 CLEAVAGE OF SYXTHETIC POLYPEPTIDE MODELS’
Peptide
XBS, moles/mole pep
---Cleavage, Room temp
Bond cleaved
%--1 hr a t 100°
Method ot determn
1
.~c-His-Phe-.~rg-Tr~--G1yOH”2’
Z-\~al-Tyr-\~al-Hi~-Pro-PheOMe~~~ I Asp( S H p ~ - A r g - ~ ~ a I - T y r - ~ a l - H i s - P r o - P h e O H 3 ~ [ \-alj-H>-pertensinII-asp-/5amide)
6 6
6
3
His-Phe Try-GI>-
45
30 47
DSP SIN
Tyr-Val His-Pro
24
0
28 42
DSP PIP
Tyr-\*al His-Pro
15d 2
..
48
DSP PIP
;Isp( SHp)-.~rg-\~al-Tyr-\~al-His-Pro-Phe-His-LeuOH35 10
Tyr-\‘a1 .. . . P 1.P His-Pro 3 45 XIS i \.aI5-H\-pertensin I ) His-Leu 3 53 ‘l In pyridine-acetic acid-water (1: 10:19, v . / v . ) , p H 3.3. Ac = acetyl; 2 = carbobenzl-loxy. X I S = ninhydrin; D S P = dinitrophenylation; P I P = proline imino-peptidase. Determined by G. L. Schmir and L. Cohen ( ( I f . ref. 36) S o correcExtent of cleavage was not determined tion factor for hi-drolytic or chromatographic losses has been applied to this result.
recently elucidated using chemical37 and X-rays8 methods. .According to these studies there are four proline residues in the protein: one is situated after isoleucine, two residues after alanine, and one after histidine. M7henthe protein was treated with KBS a t room temperature (pH 3.3) no terminal proline could be detected, indicating that a try-pro or tyr-pro sequence is unlikely. However, on heating the reaction mixture for 1 hr. a t S5-90° (after destruction of excess NBS with imidazole) a new terminal proline was found. After incubation of the cleaved protein with proline iminopeptidase we determined 53% cleavage of the his-pro bond (assuming one his-pro bond in the molecule).
This method proved adequate when tested on peptides (see Tables I1 and 111). As for the competition between tryptophan and tyrosine, i t was reported3 36 that by using a limited amount of NBS, tryptophyl residues may be cleaved predominantly. Our results indicate t h a t tryptophyl bonds cannot by cleaved in high yields withGut causing some cleavage of tyrosyl bonds. For instance, when increasing amounts of S B S were added to an equimolar mixture of tryptophyl, tyrosyl, and histidyl peptides and the cleavage of each peptide was followed (Fig. 7 ) . we
Discussion m’hen a peptide or a protein is treated with NBS, several amino acid residues are liable t o be attacked. These include : cysteine, cystine, methionine, tryptophan, tyrosine, and histidine. Cysteine and cystine residues react rapidly with NBS, and different products may be formed1° with different ‘amounts of the reagent. However, by prior oxidation of the protein with performic acid,39--11 cysteines and cystines are converted quantitatively into cysteic acid residues which are resistant to NBS. Methionines are oxidized by excess NBS into methionine sulfone residuesY6; therefore., in the determination of bonds cleaved, a sulfone marker should be included. Tryptophyl, tyrosyl, and histidyl peptide bonds are all cleaved by NBS. The method we propose does not, make i t possible to cleave histidyl bonds without cleaving tryptophyl and tyrosyl bonds as well. From the available experimental evidence, i t seems that also tryptophyl and tyrosyl bonds cannot be cleaved without, at least, partial oxidation of histidyl residues. In a protein, where certain residues may be activated while others are inaccessible to the reagent, such competition for NBS is liable to cause misleading results. Nevertheless, it is possible to differentiate between cleavage of tryptophyl and tyrosyl peptide bonds, on the one hand, and of histidyl peptide bonds on the other. This is achieved by performing the cleavage in two steps: (a) cleavage of tryptophyl and tyrosyl bonds a t room temperature with concomitant oxidation of histidyl residues, (b) cleavage of histidyl peptide bonds by heating. (3:) l h i s peptide was kindly supplied b y Dr R. Schwyzer of C I B A Ltd , Bade ( 3 0 ) G I. Schmir and I.. A Cohen, J . A i n . Chem. Soc., 83, 723 (l9fi1). ( 3 i ) A . B Edmundson a n d C H UT Hirs, S a l w e , 190, 663 (1961). (38) J . C K e n d r e a , e l a l , i b i d , 190, 6 6 0 (1901) (:is) F Sanger. Bzorhein J . . 4 4 , 120 (1949). (40) J h l . M u e l l e r . J G Pierce, H. Davoll, and V du Vigneaud, J Bid. Chem., 191, 809 ( l Y 5 l ) . (41) C H Hirs, ibiii , 219, 6 1 1 (1956).
t I m R
I
L
3
6
9
3
6
9
3
6
9
NBS (Moles1
Fig. 7.-Cleavage of an equimolar mixture of the peptides % tryptophylglycine, S-Ac-tyrosi-1-alanine, and I\;(a)-histidyl(c-Z)-lysine with increasing amounts of S B S . The solvent was pyridine-acetic acid-water, 1: 10: 19, v . / v . , pH 3.3: .I, extent of cleavage of the tr)--gly bond a t room temperature, -O--, and after heating, -0-; B, extent of cleavage of t h e tyr-ala bond a t room temperature, -0-, and after heating, -0-; C, extent of cleavage of the his-1)-s bond a t room temperature, -0-, anti after heating, -0-.
found that in order to achieve maximal cleavage of the tryptophyl bond (42Yc) G moles of XBS were needed. Under these conditions the tyrosyl bond was cleaved to some extent (ljyG) and the histidyl residues were partially oxidized (an aliquot of the same reaction mixture was heated and the histidyl bond was cleaved in 24yL yield). 0-Carbobenzyloxylation or 0-acetylation may be a means of preventing cleavage of tyrosyl linkages.’ A drawback of this masking group is that it usually renders the protein less soluble. Moreover, the literature concerning the reactivity of phenolic hydroxyls on acetylation is contradictory. Some authors12 state t h a t phenolic hydroxyls are acetylated more readily than aliphatic hydroxyls, while othersd3insist that phenolic (42) H S Olcott and H Fraenkel-Conrat, Chem R e t ’ , 41, L j l ( 1 W i ) (43) J . Salak and 2. Vodrazka. Biochrin B i o p h u s . A r l a , 65, 11.5 (19(j2).
2804
SHMUEL SHALTIEL AND ABRAHAM PATCHORNIK
Vol. 85
hydroxyls are not affected by acetylation. The applicability of this method for masking tyrosins should be further investigated. Preliminary experiments indicate that cleavage of histidyl peptides is feasible on paper chromatograms (see Experimental). This method may therefore be of use for quick identification and determination of histidyl peptides in partially hydrolyzed or enzymatically digested proteins. The products formed on treatment of various histidine models with NBS and the mechanism suggested for this cleavage will be the subject of a forthcoming publication.
4
m d
Experimental
C
5
Synthesis of Model Compounds.-N(,) ,N,I,,-Dicarbobenzyloxy-histidyl amino acid esters were synthesized from dicarbobenzyIoxy-L-histidine4$ and t h e appropriate amino acid ester4G by the dicyclohexylcarbodiimide rr1ethod.~7-j' The peptides prepared and the corresponding analytical data are described in Table I V . Sonaqueous titrations were used for the characterization of these peptides and for the determination of their purity4j>j2 (see Methods). N(a,-CarbobenzyloxyhistidylPeptides.-Dicarbobenzyloxyhistidyl peptide esters (3-5 mmoles) were suspended in methanol ( 2 5 ml.) and 2 AV S a O H ( 2 . 5 equiv.) was added, whereupon the suspension cleared. The reaction mixture was allowed to stand a t room temperature for 1 hr. and then i t was cooled ( a " ) , diluted with water (25 ml.), and acidified with 1 S HCI until crystallization started ( p H 5.0-5.5). If crystallization did not start within 30 min., the solution was concentrated i n Y U C U O and t h e peptide was precipitated by addition of acetone. The peptides were recrystallized, dried over PaOj ( a t 50' for 2-3 hr.), and charcterized by nonaqueous titrations (see Methods). N!,)-Carbobenzyloxy-L-histidyl-L-phenylalanine Amide.Absolute methanol (150 ml.) was saturated with ammonia a t 0". To this soluti,on, 14 g. (24 mmoles) of dicarbobenzyloxyL-histidyl-L-phenylalanine methyl ester was added. The reaction mixture was allowed to stand a t room temperature ( i n a tightly stoppered flask) for 4 days. The precipitate formed was filtered, washed with cold methanol, and recrystallized from 50Fc aqueous methanol. After drying over P 2 0 5 for 2 hr. ( a t 60") the yield was 8.5 g. (827,), m . p . 214-216' (lit.j5 203-204"). Anal. Calcd. for C23H2jS504: C, 63.43; H , 5.79; S , 16.08; neut. equiv., 435. Found: C, 63.27; H, 5.89; N, 15.80; neut. equiv., 438 (with HCIOc in glac. acetic acid) and 436 (with HClO4 in dioxane). N(,)-Carbobenzyloxy-L-histidyl-glycine Ethyl Ester.-Ilicarbobenzyloxy-L-histidyl-glycine ethyl ester (1 g., 2 mmoles) was suspended in 10 ml. of absolute ethanol and 1.9 ml. of 0.55 LV NaOC2Hj was added with vigorous stirring. After 30 inin., the reaction mixture was neutralized with 1 HCI and evaporated to dryness. The peptide was extracted into hot chloroform. The extract mas washed with water and dried over Sa2S04. On treatment with petroleum ether a solid paste was precipitated which was recrystallized from ethanol and dried over P205;yield 0.57 g. ( 7 5 ( x ) ,m . p . 113" (lit.j5 11:3-114'). Anal. Calcd. for ClEHzlSlOs: C , 57.8Y; H , 5.67; N, 15.01; neut. equiv., 373. Found: C , 58.35; H, 5.82; K , 11.91; neut. equiv., 376 (with HCIO, in glac. acetic acid) and 368 (with HC104 in dioxane). 4(or 5)-Imidazole-propionic Acid.-A sample of u-chloroimidazolepropioiiic acid56 (10 g., 57 mmoles) was dissolved in 150 ml. of 57; aqueous HCI, t o which 1 g. of 5YGPd-C catalyst was added. The reaction mixture was transferred into a Parr apparatus for reduction. After 24 hr., the catalyst was filtered off and washed ( 1 4 ) All melting points are uncorrected. (4.5) A . Patchornik, A . Berger, and E . Katchalski, J . A m . Chem. SOL., 79, (j1l(i ( 1 9 5 7 ) . (40) J P Greenstein and hl Winitn in "Chemistry of t h e Amino Acids," L'ol. 2 . John Wiley and Sons, l n c . , S e w York, N. Y . , 1901, p p . 924-913. 147) J. C . Sheehan and G . P. Hess, J . .4m.Chem. Soc , 77, lO(i7 (19%). (18) S. Akabori, K . Okawa, a n d F Sakiyama, :Valid;.e. 181, 772 (1958). (49) F. S a k i y a m a , K . Okawa, T. Yamakawa, and S Akabori, Bull ('hem. SOC.Japa>a. 31, 926 (1958). (50) F. S a k i y a m a , ibid , 36, 1913 (1962). (.>I) K . Inouye and H Otauka, J Ovg Cheiw , 27, 42:jC (l(i2) !.52) A Patchornik and S Shaltiel, Bull. Rei. Coa,rcii I s u n e l , 11A, 224
R
\V, Holley a n d E. Sondheimer, J . A m C h e m SO( , 76, 1326
(1951).
K . Hofmann, el a i . , i b i d , 79, 1641 (1937). (RS) N. C Davis, J . B i d C/ze,iz., 223, 985 (195(:). (56) S. Edlhacher a n d H . von Bidder, Z . physzol. Chem., 276, 120 ( 1 9 1 2 ) .
(;14)
Sept. 20, 1963
CLEAVAGE OF HISTIDYL PEPTIDE BONDS
with water.
The filtrate and the washings were concentrated On cooling, a heavy white precipitate was formed which was recrystallized twice from aqueous acetone; m.p. 130" (lit.57105'), yield 9.7 g. (87.570). Anal. Calcd. for CpHsX2O2.HC1.Hz0: C, 37.02; H , 5.70; K , 14.10; C1, 18.22. Found: C , 36.98; H , 5.70; N, 14.30; C1, 18.65. The hydrochloride hydrate (3.9 g., 20 mmoles), was dissolved in a minimum volume of hot methanol. On addition of 2.8 ml. of (C2Hj)ON(20 mmoles) precipitation began. After filtering and washing with absolute ethanol, t h e sample was recrystallized by dissolving in water and dilution with acetone. The sample was dried over P205 a t 60" for 2 h r . ; m.p. 212' (lit.58212'), yield 2.2 g. (785;). N~l,,-Carbobenzyloxy-imidazole-propionyl-glycinep - Nitrobenzyl Ester .-Imidazole-propionic acid hydrochloride hydrate (9.7 g., 50 mmoles) was dissolved in 100 ml. of 1 N XaOH, and 75 ml. of 2 N Sa2C03 was added. T h e solution was cooled t o 0" and carbobenzyloxy chloride (8.5 ml., 50 mmoles) was added portionwise with vigorous shaking during 20 min. The reaction mixture was allowed t o stand for 30 min. and then it was acidified until a gelatinous mass precipitated. This was extracted into 100 ml. of ethyl acetate. An aliquot of this solution was titrated with HClOl in glacial acetic acid (using thymol blue as indicator) and thus t h e concentration of S(I,,-carbobenzyloxyimidazole-propionic acid in t h e ethyl acetate was determined t o be 30 mmoles/100 ml. (yield 607c). This compound reacted with 30 mmoles of glycine p-nitrobenzyl ester (prepared from 7.4 g. of its hydrobromide b y (CYH,)tK) in t h e presence of dicyclohexylcarbodiimide according t o t h e method described for the synthesis of dicarbobenzyloxy-histidylpeptide esters. 49 S,I,&xbobenzyloxy-imidazole-propionyl-glycine p-nitrobenzyl ester was recrystallized from ethyl acetate-petroleum ether; m . p . 118", yield-i.1 g. Anal. Calcd. for C23H22S407: C, 59.22; H , 4.75; 9,12.01; neut. equiv., 466. Found: C, 59.55; H , 4.80; S , 12.02; neut. equiv., 4T6 (with HClO, in glac. acetic acid). 'imidazole - propionyl-glycine.-S(~m~ - Carbobenzyloxy-imidazole-propionyl-glycine p-nitrobenzyl ester (2.3 g., 5 mmoles) was dissolved in 25 ml. of methanol T o this solution 6 ml. of 2 S KOH was added. The reaction was allowed t o proceed for 30 min. a t room temperature and the p H was adjusted t o 5.0 with 105; HCIO,. The precipitate which formed immediately (KCIO,) was filtered and washed with methanol. The filtrate and the washings were concentrated in V ~ C U Oand t h e resulting oil was dissolved in 10 ml. of methanol. Part of the substance (residual KCI04) which did not go into solution was filtered off. On addition of ether t o the filtrate, a white paste was precipit a t e d , This paste was crystallized twice from a minimal volume of hot methanol, and dried over PZOSa t 60"; yield 0.42 g. (42y0), m.p. 209". Anal. Calcd. for C & I I N & ~ : C, 48.72; H , 5.62; K , 21.31; neut. equiv., 197. Found. C, 48.42; 13, 5.42; S , 21.03; neut. equiv., 201 (with HCIOj in glac. acetic acid) and 200 (with KaOCH3 in methanol-benzene). Tryptophyl and tyrosyl peptides were synthesized by the dicyclohexyl carbodiimide method: Z-L-tryptophylglycine ethyl ester, m . p . 119" (lit.55 120'); 2-L-tryptopliyl-L-alanine benzyl ester, m.p. 150" (lit.5s 153"); S,0-Z2-L-tyrosylglycine ethyl ester, m.p. 164' (lit.60164-165"). By saponification of these peptides we obtained: 2-L-tryptophylglpcine, m.p. 157" (lit.61 158-159'); 2-L-tryptophyl-L-alanine, m.p. 156' (lit.62155'); 5-2-L-tyrosylglycine, m . p . 98" ( lit.63 100"); K-Ac-L-tyrosyl-Lalanine, m . p . l10°.64 Methods. Characterization and Determination of Peptides Containina Histidine by Nonaqueous Titrations.-The titrimetric properties of imidazoles and a c y l i m i t l a ~ o l e swere ~ ~ used for characterization and determination of peptides containing histidine. Each of t h e peptides was titrated with three reagents: ( a ) HClOl in glacial acetic using crystal violet ( C . V . ) as indicator, ( b ) HClOa in dioxane using thymol blue ( T . B . ) a s indicator, ( c ) NaOCH1 in methanol-benzene using thymol blue as indicator. The titrations were performed on samples of 0.1-0.15 mmoles, dissolved in the appropriate solvent. The solvent was previously neutralized with the reagent and a n aliquot was set aside as a standard for color change. Standard solutions (0.1 N ) of t h e reagents were used.
in vacuo until t h e first crystals appeared.
(57) 4.Windaus and W Vogt, Bed? C h e m . Phys. Palh., 11, 408 (1908).
(AS) E W Rugeley and T. B Johnson, J . A m . Chem. SOL., 47, 2995 (192.5). (59) M . IVilchek and A Patchornik, J . Ovg. Chem., 28, 1874 (1963). (DO) R . Schwyzer, hI. Feurer, a n d B. Iselin, Helz.. C h i m . Acla, 38, 83 (19%). (61) K Hofmann, e l al , J . A m . Chem. SOL.,80, 1486 (1958). (62) E. L Smith, J . B i d Chem., 176, 39 (1948). (03) > I . Bergmann a n d J . S . F r u t o n , i b i d . , 118,40.5 (1937). (641 A samnle of this Deptide was kindly supplied b y M r . M . Wilchek.
2805
The peptides t h a t we synthesized fall into three groups, t h e titrimetric properties of which are summarized in Table 1'. D a t a concerning molecular weight determinations of these peptides are given in Table 11'. TABLE V
TITRIMETRIC PROPERTIES
O F PEPTIDES
General formula
2- NH-CH-CO-NH-
I
COL T A I N l N G HISTIDINE
HClOi (AcOH), moles/ mole
HCIO4 (dioxane), moles/ mole
YaOCHs (CH,OH), m(,leii mule
CH-COORZ
PHZ
I Ri
1
0
00
1
1
0
1
1
1
/=(
NvN-z 2-NH-
CH-CO-NH-CH-COORz I I Ri
2-NH-CH-GO-NH-CH-COOH I
I
FHZ
Ri
/=( N\ N H
v
Indicator Color change
-
C.V.' Violet green
TB.' Yellow red
-
T.B.' Yellow hlue
-
a NaOCH3 removes the carbobenzyloxy group from the irnidazole ~ a t a l y t i c a l l y . ~ C.V.-crystal ~ violet; T.B.-thymol blue.
Paper chromatography was carried out on Whatman No. 1 chromatography paper by the descending technique. The solvent used was usually 1-butanol-acetic acid-water (25: 6 : 25, v./v.). Amino acids or amino acid esters were revealed b y spraying with a solution of 0.570 ninhydrin in 85Yt aqueous acetone. Proline was detected on paper by spraying with a solution containing 154 ninhydrin and li?; trichloroacetic acid in ethanol.33 High-voltage e1ectrophoresis5j was performed in ppridine-acetic acid buffer ( p H 3.5) a t a potential gradient of 40 v./cm. Ultraviolet spectra were taken in a Beckman model D K 1 recording spectrophotometer. Total hydrolysis of peptides was carried out with 6 N HCI in sealed tubes a t 110' for 22 hr. Determination of Yields of Cleavage in Dipeptides.-The method used for the determination of yields of cleavage in dipeptide models is exemplified in the following experiment: Solutions of 2-His-Gly (0.05 M) and S B S (0.05 M ) in pyridine-acetic acid-water (1 : 10: 19, v./v. j were prepared. To 1 rnl. of the peptide solution was added 3 ml. of the S B S solution (3 moles/mole) (solution A ) . Another portion of the peptide solution was diluted with 3 ml. of the solvent (solution B ) . A 0.125 JM solution of glycine in the same solvent was prepared (solution C). The concentration of the glycine in solution C was equal to the concentration of the peptide in solutions A and B , After standing for 30 min. a t room temperature, a I-ml. sample of each of solutions A, B , and C was transferred into test tubes which were sealed and heated t o 100' for 1 hr. Band-like spots (length 2 cm.) of each of the unheated solutions A, B, and C were put on chromatography paper (Whatman S o . I ) . T h e spots from solutions X and B were of 50X whereas from solution C there were spo of 10h, 20X, 30X, and 10X corresponding t o 20, 40, 60, and 80 of t h e glycine in solutions -1and B . .\nother chromatogram \ s likewise "spotted" with the heated soiutions A , B, and C. The two chromatograms were run for 12 hr. with butanolacetic acid-water. After drying for 30 min. in a ventilated oven a t OO", the chromatograms were dipped in a 0.5(,; ninhydrin solution i n 859; aqueous acetone and heated again for 20 min. a t the same temperature. T h e spots formed were cut and eluted with 7 ml. of aqueous ethanol 75'1. The color intensity was determined in a Klett-Summerson photoelectric colorimeter w i t h a filter S o . 56. The results are summarized in Table 1'1. .I calibration curve was drawn for each Chromatogram (Fig. 8) from which the percentage of cleavage was determined. It should be mentioned t h a t different calibration curves are obtained for different chromatograms owing t o several factors, such as t h e duration and temperature of drl-ing and developing with ninhydrin. In our experiment t h e freed amino acid (for which t h e calibration curve was drawn) underwent a similar procedure t o t h a t of the peptide in order t o minimize t h e sources of (65) A .\I Katz, U'. J . Dreyer, and C B Anfinsen, J . B i d . C'hem., 234, 2897 (1959).
SHMUEL SHALTIEL AND ABRAHAM PATCHORNIK
2806
VOl. 85
Cleavage of the His-Pro Bond in Val5-Hypertensin 11-asp+ amide.-The procedure described above was applied t o 9 ptnoles (9.3 m g . ) of this peptide with 6 moles/mole of NBS. Excess of the reagent was destroyed (before heating) with imidazole. The yield of cleavage was found t o be 48'.; after heating and 2' before heating. I t should be mentioned t h a t in this case some hydrolysis (13$) of t h e his-pro bond was detected in t h e control 200 experiment. Cleavage of the His-Pro Bond in Val6-Hypertensin I.-The procedure described above was applied t o 3.3 pmoles (4.27 mg.) of this peptide with 10 moles/mole of S B S . Excess of the reagent was destroyed (before heating) with imidazole. The yield of cleavage of t h e his-pro bond was 45( Iafter heating and I / / 34; before heating ( a t room temperature). ,e ' Cleavage of the Tyr-Val Bond in 2-Val-Tyr-Val-His-Pro-Phe OMe.-.L\ sample of 9.02 mp. (10 umoles) of this DeDtide was treated with 60 pmoles of %BS in' pyridine acetale,' pH 8.3. The reaction mixture was divided into two parts: one of these was heated for I hr. a t 100" and then lyophilized, and the other was lyophilized without heating. Both samples were treated with fluorodinitrobenzene in bicarbonate solution,3' The resulting mixture was hydrolyzed a t 110" for 16 h r . Dinitrophenol was removed by sublimation and t h e residual yellow material was separated by two-dimensional chromatography with valine as a standard. The amount of D S P valine was determined by elution of t h e spots and spectrophotometric assay. The yield 20 40 60 00 of cleavage was 28uc in the heated sample and 24(&in the sample which was not heated. These results were corrected for hydroGLYCINE RELEASED, %. lytic and chromatographic losses. Correction factors were estiFig. 8.---Calibration curve for the determination of >-ield of mated from a parallel control e ~ p e r i m e n t . ~ ~ Cleavage of the His-Phe Bond in His-Phe-Arg-Try-Gly .-.I cleavage in Nf ol)-%-histidyl-gl?-cine. sample of 7 mg. (10 pmoles) of this peptide was dissolved in bicarbonate solution (400 pmoles/ml.) and 0.01 ml. of acetic anerror. Results of cleavage mere found t o be reproducible within hydride (100 pmoles) was added under cooling (0'). After 1 hr, i5((. the reaction mixture was acidified to p H I with dilute HCI. Cleavage of a Peptide on a Chromatogram.--X spot of 100k Ll-ophilization was carried out and the resulting acetylated pepO.Ot23 >lf Z-His-GlyOH (in 0.1 .V HCI) was put on a strip of tide was dissolved in 10 ml. of pyridine acetate ( p H 3.2); 5 mi. chromatography paper (Whatman S o . 1) which was then dipped of this solution was treated with 10 ml. of a 3 fimoles/inl. soluti(in for 1 min. in a solution of bromine in petroleum ether. After of S B S (6 moles of NBS/mole of peptide). Cleavage and airing and drying, the paper was put in a 100' oven over water dinitrophenylation were carried out as described above. The vapors for 1 hr. On developing we found glycine. his-phe bond was cleaved in 3c3 yield a t room temperature and TABLE VI in 30yc yield on heating. Cleavage of the Try-Gly Bond in His-Phe-Arg-Try-Gly and DETERXISATIOK OF CLEAVAGE IS Z-His-G1y His-Leu Bond in Val5-Hypertensin I.-By cleavage of these --At room temp.---After heating-bonds, a free amino acid is liberated. The method described Volume Color Color for the determination of yields in dipeptides was applied in these "spotted.' intensity, Cleavage. intensity, Cleavage cases. Yields of cleavage are given in Table 111. K . U 'I c7 Solution x K L' ';70 Cleavage of the His-Pro Bond in Sperm-Whale Myoglobin.-I The sample on which the cleavage was performed was a paste A . 50 2 160 58 precipitated by ( NH4j:SO4 which contained ca. 4 0 c i protein B 30 0 0 0 0 according t o Dr. J . C. Kendrew. This was confirmed by colori10 66 20 20 C 54 metric determination6e,67 of the iron content in our sample (0. t2'!;, 20 130 10 104 40 Re as compared mith 0.3152 Fe in the pure protein'^.^^ C A sample of 125.8 mg. of myoglobin paste3' ( 2 . 8 pmoles) was 30 193 GO 168 C 60 suspended in 10 ml. of pyridine-acetate buffer ( p H 3.33. .Isolu215 10 266 80 80 C tion of S B S in the same solvent (:35 mg. in 3 xnl.) was added K . C = Klett units. portionwise with vigorous stirring. After standing for 15 tnin. a t room temperature, excess S B S was destroyed by adding crystals Cleavage of the His-Pro Bond in 2-Val-Tyr-Val-His-Pro-Phe of imidazole until no more S B S could be detected o n KI-starch 0Me.-.I satnple of the peptide (3.65 mg., 4 pmoles) was dispaper. The reaction mixture was heated for 1 hr. a t 8.5-90". solved in 4 ml. of pyridine acetate buffer ( p H 3 . 3 , solution A , ) . After lyophilization, 20 ml. of water was added, the p H of the .Iti ptnoles/ml. solution of S B S was prepared by dissolving 10.7 mixture was adjusted t o 8.5, and most of the substance was tng. of the reagent in 10 nil. of the same buffer (solution B ) . dissolved. The color of the solution was red while t h a t o f the T o a test tube containing 2 ml. of solution A we added 2 ml. of precipitate was dark red. After centrifugation the precipitate solution B ( 6 tnoles SBSitnole peptide). The turbidity and was washed with water, and the washings combined \vith the yellow color initially formed disappeared on standing a t room solution were lyophilized again to yield a !-ellowish po\vder. teni:jerature. After 13 min. no S B S could be detected in the By incubation with proline iminopeptidase, it was determined reaction mixture by KI-starch paper; 2 ml. of t h e reaction mixt h a t the powder contained 1.48 pmoles of proline or t h a t cleavage ture \vas transferred into another test tube, sealed, and heated in 5355 yield occurred. for 1 hr. a t 100". After cooling, the contents were washed into I n a parallel experiment in which the reaction mixture \vas not a round-bottomed flask and lyophilized; 4 ml. of water was added heated no proline was detected. and 2 aliquots of the resulting so!ution (each of 1 ml. containing Acknowledgments.-\Ye thank Professor E. Kat11.25 pmole of the original peptide) were taken out. These were incubated for 20 hr. a t 37", one of them with proline iminopeptiSarid chalski for his interest in this work and Dr. atid the other without the enzyme. In the portion which for her collaboration in the quantitative determination was digested with the enzyme 0.105 pmole of proline was deterof N-terminal proline residues. This investigation was mined colorimetrically,using acidic ninhydrin. This corresponds supported by Grants A-3171 and AM-.j098 from the to 12' cleavage. The following control experiments were perfurmed: -In aliNational Institutes of Health, United States Public quot of t h e reaction mixture (containing 0.23 pmole of peptide) Health Service. which was oxidized with NBS but not heated, and an aliquot of (06) F. Scheihl a n d U . Saffer, Z p h y s z o ! Cheni , B98, 1 7 2 ( 1 9 2 ) . the peptide which was not oxidized but only heated ( 1 hr. a t (07) W e t h a n k RIrs S. Rogozinski for performing this determination 100°, pH :3.3),. underwent the same procedure (with P I P ) . ( 6 8 ) Assuming a molecular weight of 18,000, Ci. ref :17 So proline was detected. I
L
I
I
c
/
l
'
I
f
s.
