This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

3 1 4 A. FONTANA, F. M. VERONESE, AND E. BOCCU'

wachsenden Konzentrationen an B U d R bzw. BU steigen in beiden Fällen die Ausbeuten um den glei-chen Betrag an.

Be im B U wurden organische Peroxide nach-gewiesen, die in saurem Medium eine Zeitlang sta-bil waren und sich zur Isodialursäure ( I D S ) umwandel ten 2 . D ie bei der BUdR-Radio lyse in neutralem M e d i u m intermediär nachzuweisen-den Perox ide sind anderer Natur. Aus ihnen wird kein I D S d R gebildet. Das zum I D S d R führende Perox id konnte nicht gefaßt werden, da be im Desoxyr ibos id wegen der Labilität der gly-kosidischen Bindung nicht im Sauren gearbeitet wurde. Die kurzlebigen Perox ide des B U d R sind wahrscheinlich Perox ide der Zucker-Komponente, da ihre Ausbeute unabhängig von der BUdR-Konzen-tration ist. Ihre Vor läufer sind daher vermutlich

OH-Radikale, die am Zuckerteil angegrif fen haben. Die Ausbeute des IDS-bi ldenden BU-Peroxids steigt hingegen mit wachsender BU-Konzentration a n 2 . Dieser Anstieg bedeutet, daß die Vor läufer des BU-Peroxids an der Base angreifende e®, H- und 0 2 e - R a d i k a l e sind.

Experimentelles

Der wesentliche Teil der in dieser Arbeit verwende-ten experimentellen Technik wurde in der vorangegan-genen Arbeit 3 beschrieben.

Es wurde ausschließlich mit dem in Stellung 3 (in der Base) 14C markierten BUdR gearbeitet. Radiochro-matogramme wurden mit Hilfe eines Integralschrei-bers quantitativ ausgewertet; sehr kleine Peaks wurden außerdem noch planimetriert.

Bei den organischen Peroxiden wurden die G-Werte unmittelbar nach Bestrahlung bestimmt.

Covalent Binding of Polypeptides to Sulfonium Resins ANGELO FONTANA, FRANCESCO M . VERONESE, and ENRICO BOCCU'

Institute of Organic Chemistry and of Pharmaceutical Chemistry, University of Padova, 35100 Padova, Italy

(Z. Naturforsch. 26 b, 314—317 [1971] ; received January 9, 1971)

Basic dimethyl-arylmethylene-sulfonium resins (L. C. DORMAN and J. LOVE, J. org. Chemistry 34, 158 [1969] have been used as reactive supports for the covalent binding of N-protected as well as N-unprotected peptides through a benzyl ester linkage. Example are given of effective coupling of peptides of rather long sequence, the quantity of peptide linked being up to 0.5 ^wmole per mg of resin. The method could find practical use in the solid state degradation of polypeptides.

Sol id phase degradation, with a polypeptide an-chored to a solid support, will improve the auto-mation of the process of the peptide sequential de-gradation. The mechanics of the EDMAN'S sequena-tor 1 will be greatly simplified and the instrument will be almost the same as the one f o r peptide syn-thesis described b y MERRIFIELD et al. 2 . The solubi-lity prob lem of short-chain peptides in the sequena-tor technique should be eliminated.

Effective b inding of polypeptides has been ex-plored by several authors. In the LAURSEN'S proce-dure 3 the peptide is first acylated at the amino funct ion by reacting it with Zerf.butyloxycarbonyl

Reprints request to Dr. A. FONTANA, Institute of Organic Chemistry and of Pharmaceutical Chemistry, University of Padova, 35100 Padova, Italy.

* The abbreviations are those recommended by the IUPAC-1UB Commission on Biochemical Nomenclature, Biochemi-stry 5,1445 [1966].

( B O C ) * azide in anhydrous dimethyl formamide and then coupled to an amino resin after activation with AVV'-carbonyldi imidazole . The procedure was modi f ied by DINTZIS (cited by S T A R K 4 ) , who pro-tected the amino funct ion through reversible nitro-arylation with 2 ,6-dinitrobenzene- l ,4-disul fonate and coupled the A^-protected peptide by means o f a water-soluble carbod i imide to a hydrophi l i c amino resin.

M o r e recently, WIELAND and c o w o r k e r s 5 re-ported a procedure in which BOC-peptides are firstly linked as amides with the aminomethyl group of azoarylsulfonic acids. The co loured peptide deri-

1 P . EDMAN a n d C . BEGG , E u r o p e a n J. B i o c h e m . 1 , 8 0 [ 1 9 6 7 ] .

2 R . B . MERRIFIELD, J. M . STEWART , a n d N . JERNBERG, A n a l y t . C h e m . 3 8 , 1 9 0 5 [ 1 9 6 6 ] .

3 R . A . LAURSEN, J. A m e r . c h e m . S o c . 8 8 , 5 3 4 4 [ 1 9 6 6 ] . 4 G. R. STARK, Advances Protein Chem. 24, 261 [1969].

POLYPEPTIDES AND SULFONIUM RESINS 3 1 5

vatives are then fixed as anions at acidic alumina and subjected to solid state E d m a n degradation.

However , these procedures to link peptides to solid support can be accompanied b y undesired side reactions involving unprotected hydroxyl and carb-oxyl g r o u p s ; moreover , the reversible blocking of the amino group is required.

In view of the actual importance of the sol id phase degradation 3 ~ 8 , w e wish to report our findings dealing with the use of basic dimethyl-(arylmethylene) -sul fonium resins ( 1 ) 9 as reactive supports f o r the covalent b inding of polypeptides. These resins have been previously used in the solid-phase peptide synthesis 9 to link BOC-amino acids to the styrene-divinylbenzene copo lymer resin 9 . The reaction involves salt f o rmat ion ( 3 ) between the sul fonium resin ( 1 ) and the carboxyl i c g r o u p ( 2 ) and subsequent esterification ( 4 ) by heating in vacuo.

CH 0 R

(R) - /~\ -CH2S^ 3 HC03e + HO—C—CH—NH-w CH3

1 2

( 2 > 0 - C ^ ( 3 eO-C-CH-NH^v+ H20 +C02 CH3

3 i Cj> R

® - f j V c H 2 0 - C - C H - N H ^ , + CH3SCH3

4

Experimental

Preparation of the sulfonium, resin: The chloro-methylated copolystyrene — 2% divinylbenzene, 200 — 400 mesh, was obtained from Bio-Rad Laboratories (0.9 meq — CH2C1 per gram), or prepared by chloro-methylation of the polymer as described by MERRI-FIELD 10. The samples of the sulfonium resin, chloride form, were transformed to the bicarbonate form fol-lowing the procedure of DORMAN and LOVE 9. The wet resin (30 — 50% water) was stored in the refrigerator.

The peptides used in this work were synthesized in this laboratory, or were Mann Res. Lab. compounds, whose purity was ascertained by thin layer chromato-

5 C. BIRR, C. JUST, a n d TH. WIELAND , L i e b i g s A n n . C h e m . 736, 88 [1970].

6 A . DIJKSTRA, H . A . BILLIET, A . H . VAN DONINCK , H . VAN VELTHUYZEN , L . M A A T , a n d H . C . BEYERMANN, R e c u e i l Trav. chim. Pays-Bas 86, 65 [1967].

7 A . SCHELLENBERGER, H. JESCHKEIT, R . HENKEL, a n d H . LEHMAN, Z. Chem. 7, 191 [1967] ; C. A. 67, 6107 [1967].

8 G . MANECKE a n d G . GUNZEL, N a t u r w i s s e n s c h a f t e n 5 5 , 8 4 [ 1 9 6 8 ] .

graphy or amino acid analysis. The peptides cor-responding to partial sequences of glucagon were a generous gift of Dr. E. WÜNSCH.

Extent of coupling: The extent of coupling to the resin was determined by acid hydrolysis of a sample of the peptide-resin in dioxane : water : HCl ( 3 : 2 : 5 ) at 110° for 36 hours in a sealed tube, followed by amino acid analysis (C. Erba analysator, Model 3A27) . Alternatively, the peptide-resin (1 — 4 mgs) was suspended in 3 ml of purified tetrahydrofuran : metha-nol : 1 N NaOH ( 7 : 2 : 1 ) in a cuvette. After 30 min. at room temperature and occasional shaking, the amount of peptide in solution was determined spectro-photometrically at 280 mju (em 5600 for tryptophan derivatives at 293 m// (£M 2500 for tyrosine deriva-tives) , or at 380 m// (eji 3950 for the 2-nitrophenyl-sulfenyl (NPS) -derivatives).

Example 1: The peptide BOC-Trp-Phe (12.8 m g ; 26.5 //moles) was dissolved in 2 ml of dioxane in a test tube and then 50 mg (20.6 //moles H C O a e ) of the sulfonium bicarbonate resin was added. After vigorous shaking for 10 min. with a tube vibromixer, the resin was filtered off and washed with dioxane. In the com-bined washes the quantity of the peptide was 7.0 / /moles (spectrophotometric determination, £M 5600 at 280 m//) . corresponding to 19.5 //moles linked to the resin by salt formation (95% binding). The peptide-resin salt was heated in a drying pistol at 78° in vacuo over P 2 0 5 for 4 hours. The resin product was then washed with dioxane: I N HCl (9 : 1) in order to decouple the peptide not esterified and dried in vacuo at 78° for one hour. In the mother liquor wash the quantity of peptide was 2.4/ /moles, which make 17.1/ /moles of peptide linked to the resin through benzyl ester bond, corresponding to 83% esterification. The peptide content of the resin determined by spectrophotometric reading of the released peptide by saponification was 0.42 / /moles/mg of resin.

Analogously, the synthetic peptide NPS-Thr(tBu)-Ser- (tBu) -Asp (OtBu) -Tyr(tBu)-Ser-(tBu)-Lys(BOC)Tyr-(tBu) -Leu-Asp (OtBu) -OH protected 7 — 15 sequence of glucagon was coupled in dimethylformamide to the sulfonium resin to an extent of 0.22 / /moles/mg of resin.

Example 2: The peptide Ala-Gly (11.7 mg ; 79.0 //moles) was dissolved in 1 ml of dioxane : water (7 : 3) in a test tube with triethylamine (25 / / l ; 180

//moles) and then 2-nitrophenylsulfenyl chloride (NPS-C1) 12 (22.8 mg ; 120 //moles) in 0.3 ml of dioxane were added under vigorous shaking (vibro-mixer). After 10 min., 10 / / I of glacial acetic acid were added and the complete disappearance of the

9 L . C . DORMAN a n d J. LOVE, J. o r g . C h e m i s t r y 3 4 , 1 5 8 [ 1 9 6 9 ] .

1 0 R . B . MERRIFIELD, J. A m e r . c h e m . S o c . 8 5 , 2 1 4 9 [ 1 9 6 3 ] . 1 1 E . WÜNSCH , A . ZWICK , a n d A . FONTANA , C h e m . B e r . 1 0 1 ,

3 2 6 [ 1 9 6 8 ] . 1 2 L . ZERVAS, D . BOROVAS, a n d E . GAZIS, J. A m e r . c h e m . S o c .

8 5 , 3 6 6 0 [ 1 9 6 3 ] .

316 A. FONTANA, F. M. VERONESE, AND E. BOCCU'

free peptide Ala-Gly confirmed by thin layer chromato-graphy in ethyl acetate : pyridine : water : acetic acid (60 : 20 : 14 : 6 ) . The solution was filtered from the precipitate 2,2'-dinitrophenyl disulfide arising from excess reagent1 2 , lyophilized and then redissolved in 4 ml of dioxane. The sulfonium bicarbonate resin (101.6 mg ; 50.8 / /moles of H C 0 3 e ) was added and the mixture vigorously stirred for 10 minutes. The additional handling of the peptide-resin salt was the same as in Example 1 ; the dioxane : 1 N HCl ( 9 : 1 ) wash was substituted with dioxane : acetic acid (9.5 : 5 .5) . The peptide carboxylate-sulfonium resin step was 97% and the final esterification 87% (calculated on the basis of the amount of resin used). The peptide content of the resin was 0.44 / /mole/mg of resin, as determined by spectrophotometric reading (NPS-chromophore) of the released peptide by saponification.

Example 3: The peptide Leu-Tyr-Leu (16.6 mg; 42.8 / /moles) was dissolved in 1 ml of 0.1 N HCl and then lyophilized. The peptide hydrochloride was dis-solved in 3 ml of dimethylformamide in a test tube and then the wet sulfonium bicarbonate resin (78.8 mg; 39.9 / /moles of H C 0 3 e ) was added. After vigorous shaking for 10 min. with a tube vibro-mixer, the resin was filtered under suction in a sintered-glass funnel and washed with dimethylformamide. In the combined washes the quantity of the peptide was 36.0 //moles (spectrophotometric determination), corresponding to 6.8 / /moles linked to the resin by salt formation (17% binding). After heating of the peptide-resin salt and successive washing with dioxane: 1 N HCl (9 : 1) as in Example 1, the additional peptide recovered was 2.8 / /moles, which make a total of 4.0 //moles of peptide linked to the resin through benzyl ester bond, cor-responding to 10% esterification. Quantitative amino acid analysis of an acid hydrolyzate of the peptide-resin gave a peptide content of 0.05 / /mole/mg of resin.

Following analogous procedure the following pep-tides have been bound to the sulfonium resin: Leu-Tyr -HCl (0 .136/ /moles /mg resin, Phe-Val-GIN-Trp-Leu-Met-AsN-Thr • HBr (0 .048/ /moles /mg of resin).

Results

Several N-protected peptides of rather long se-quence have been effectively coupled to the sulfo-nium resin, the degree of substitution being 0 .22 — 0 . 4 4 / / m o l e s / m g of resin. The amount of peptide linked to the resin was usually determined by amino acid analysis after acid hydrolysis or by saponifica-tion in tetrahydrofuran : methanol : 1 N (or 0.1 ~n) N a O H ( 7 : 2 : 1 ) of the benzyl ester linkage and spectrophotometric determination (Trp , Tyr , NPS-derivatives) of the released peptide. This solvent

1 3 D . LEVY a n d F . H . CARPENTER, B i o c h e m i s t r v 6 , 3 5 5 9 [1967] .

mixture was f o u n d the most effective (ti/t 1 — 7 min. depending on the peptide-resin) in produc ing a fast release of the peptides f r o m the resins, whereas in mixtures, e.g., like dioxane : methanol : 0 .1 N N a O H or tetrahydrofuran : 1 N N a O H the benzyl ester bond was cleaved at much slower rate or it was al-most completely resistant.

In addition to the BOC-groaup 3 ' 5 , also the 2-ni-trophenylsulfenyl ( N P S ) - g r o u p 1 2 was shown to be advantageous in the reversible protection of the aminogroups . Whi le the introduction of the BOC-group is rather difficult with peptides and requires BOC-azide in anhydrous d i m e t h y l f o r m a m i d e 3 ' 1 3 , the NPS-group can be introduced in aqueous tetra-hydrofuran or dioxane by means of NPS-C1, the reaction of the peptide being istantaneous and quan-titative. The cleavage of the NPS-group is then easily accomplished by washing the NPS-peptide-resin with gaseous HCl in organic solvents or by mercaptans 1 4 . A s an example, the peptide Ala-Gly was protected b y reaction with NPS-C1, and after re-moval of the protecting group , a degradation cycle with methyl isothiocyanate was per formed , fol-lowing essentially the procedure of LAURSEN 3 . On the basis of the amino acid analysis of the peptide resin after the degradation process, the removal of the TV-terminal Ala was 9 5 per cent.

Since the esterification of peptides to the resin occurs without intermediate activation of the carb-oxyl group but through carboxylate-sul fonium group ion pair f ormat ion , the reversible protection of the amino group seems to be not very important. Indeed the yields of coupl ing to the resin are lower but, the quantity of the linked peptide ( 0 . 0 5 — 0 .1 / / m o l e / m g of resin) is still large and amenable f o r solid state degradation purposes.

In the present method of course the co-carboxvl groups of the aspartic and glutamic acid residues are also esterified to the solid support. In fact, in identical experiments using the same sul fonium resin bicarbonate f o r m , the peptide Z-Trp-Phe was esteri-fied to an extent of 0 .25 /<mole /mg of resin, whereas

C O O H the peptide Z-Glu-Trp-OMe 5 . 1 8 /<mole /mg.

If the peptide-resins will be used f o r solid-state sequential degradation no thiazolinone or thio-hydantoin will be released when the degradation

1 4 A . FONTANA, F . MARCHIORI, L . MORODER , a n d E . SCOF-FONE, Tetrahedron Letters [London] 26, 2985 [1966] ; Gazz. chim. ital. 96, 1313 [1966] .

POLYPEPTIDES AND SULFONIUM RESINS 317

reaches such a residue. Therefore , if several aspartic or glutamic acid residues are present in the peptide, the procedure can be complicated.

In an attempt to find proper condit ions for the preferential covalent binding of the a -carboxy group in respect to the co-carboxy-group the fol-lowing preliminary experiment has been performed. The sul fonium resin, bicarbonate f o r m , was firstly

C O O H treated with Z-Glu-OBz in d ioxane and then to this carboxylate-sulfonium resin were added in separate

C O O H experiments Z -Trp-Phe and Z-Glu-Trp-OMe. After heating of the peptide-resin in vacuo at 7 8 ° f o r 4 hours, the quantity of the peptide bound to the resin was 0 . 2 4 and 0 . 0 6 / / m o l e / m g of resin respec-tively. This result indicates that the more acidic a - carboxy group displaces preferentially the co-carb-oxy group of a protected glutamyl resin salt, al-lowing a preferential covalent b inding o f the pep-tide through a-benzyl ester linkage.

Discussion

The results of the present study indicate that aryl-methylene sul fonium resins are very effective reagents f o r the covalent b inding of N-protected as well as N-unprotected polypeptides through a ben-zyl ester linkage.

15 J. K . INMAN and H. M . DINTZIS, Biochemistry 8 , 4 0 7 4 [ 1 9 6 9 ] ,

16 G. S. OMENN and C. B. ANFINSEN, J. A m e r . chem. Soc . 90, 6 5 7 1 [ 1 9 6 8 ] .

The quantity of linked peptide ( 0 . 0 5 — 0 .5 /<mole/mg o f resin) cou ld be unpractical f o r solid state degradation purposes ; since the usual small amounts of peptide available, resins o f lower capa-city should be used or the peptide-resins have to be diluted with glass beads or simply with cross-linked divinylbenzene-polystyrene po lymer .

The method here described has definite advan-tages over the available techniques, since reversible protection of amino groups and activation of the carboxyl group are no more necessary. The use of the sulfonium resins is also not restricted to the polystyrene type resins, but different matrix may be used, like properly derivatized Polyacrylamide beads 15 , which have the property of swell in aqueous solvents.

Finally, it should be mentioned that the proce-dure of attaching peptides to solid supports will be of significance in the solid phase coupl ing of pep-tides of synthetic 16 as well as of n a t u r a l 1 7 ' 1 8 or igin. Experiments on these lines are in progress.

The authors are grateful to Prof. E. SCOFFONE and Prof. C. A. BENASSI for their interest to this work and to N. BERGAMO for skillful technical assistance.

This work was supported by the Istituto di Chimica delle Macromolecole and by the Centro Nazionale di Chimica del Farmaco e dei Prodotti Biologicamente Attivi of the Consiglio Nazionale delle Ricerche.

17 R. E. OFFORD, Nature [ L o n d o n ] 221, 37 [ 1 9 6 9 ] . 1 8 J. BURTON a n d S . LANDE, J. A m e r . c h e m . S o c . 9 2 , 3 7 4 6

[ 1 9 7 0 ] .