r

Indian Journal of Chemistry Vol. 38B, July 1999, pp. 783;788

Solid phase synthesis of diazepam binding inhibitor fragments 32-38 and 45-50 on a flexible butanediol dimethacrylate crosslinked polystyrene support

P K Ajikumar & K S Devaky*

School of Chemical Sciences, Mahatma Gandhi University, Kottayam, Kerala, Ind ia 686 560

Received 25 A I/gust 1998; accepted (revised) 22 April 1999

Two short sequences of the neuropeptide Diazepam Binding Inhibitor (DB I) KQATVGD (32-38) and GLLDLK (45-50) have been synthes ized on a butanediol dimethacrylate (BDDMA) cross lin ked polystyrene support by the solid phase technique. The BDDMA crosslinked polystyrene support is prepared by thc radical initiated suspension polymerization of the monomers. These supports are found to posscss very good mechanical and physical properties. The solvation properties of the new polymer support arc found to hc superior to the commonly employed styrene-DYB res ins. A 2 mole percent crosslinked polymcr support is found to be ideal for peptide synthes is. The BDDMA-PS res in is functionalised by chloromcthylation. Po lymcr support with a chlorine capacity 1.8 mmol/g is used for peptide synthesis. The first amino acid is appc ndcd to the polymer support by Gisin's cesi um salt method . The syntheses of the target peptidcs arc carri cd a lit hy thc usual so lid phase strategy using Boc amino acids. After synthes is the peptides are cleaved fro m thc support using tritluoroacetic ac id . The peptides are obtained in high yield and thc purity of the peptides arc chcckcd by TLC and HPLC.

Over the years, Merrifield 's method of so lid phase peptide synthesis I has undergone many strateg ical improvementsv . The ri gidity, hydrophobicity and physicochemical incompatibility of the styrene-DYB resin necess itated the development of new polymer supports with optimum physicochemical characteristics and a number of polymer supports with different structural characteri sti cs have been developed for peptide synthes is4

.5

. Herein we report the use of butanediol dimethac rylate-cross linked polystyrene res in for the solid phase synthes is of peptides. The solid phase technique plays a key ro le in the synthesis of biologicall y and pharmaco logically important peptides6

.7

. A number of bi oactive peptides hllve been synthes ized by thi s strategy so far. The paper describes preparati on and characteri zati on of the polymer support , its fun cti onali sati on, swelling studies and the solid phase synthes is of a few peptides. Three mode l peptides and two acti ve sequences of Diazepam Binding Inh i bitor~'(), a bioactive po lypeptide, were synthes ized using the new support. Diazepam binding inhibitor is a neuropeptide found in mammalian brain and it di splaces benzodiazepine from its binding s ites. The fragments 32-38 and 45-50 are two biologicall y important sites of OBI. These two fragments were synthesised 111

good yie ld with high purity, using the new support.

Results and discussion

Butanediol dimethacrylate (BDDMA) cross lin ked po lystyrene supports were pre pared by the rad ica l initi ated suspension po lymeri sat ion 10 of the monomers, styrene 1 and BDDMA 2 (Scheme I).

Polymers (3a-3d) with 2%, 4%, 6% and 89'0 cross link densities were prepared by vary ing the monomer mole ratios. The res ins were obtained as spherical beads. The po lymer beads were characterized by infrared spectroscopy and 11C CP­MAS NMR spectroscopy. IR spect rum shows a sharp band at 1720 cm-I corresponding to the ester carbonyl group of the cross linking agent. So lid state J1C CP­MAS NMR spectrum (Figure 1) shows an intense peak at 130.4 ppm corresponding to aromatic polystyrene carbon and a small peak at 148.2 ppm ari sing from C-3 carbon of the styrene. T he backbone methylene carbon of the pol ymer appea rs at 42.7ppm.

A 2% cross linked po lymer with a bead s ize of 200-400 mesh was chosen for the synt hes is of the target peptides. The polymer was functionali sed with chloromethyl groups via Friedel -Crafts reaction ll

(Scheme II). By adjusting the reacti on cond itions resins with varying chlorine capac ities were prepared. The functional group capacity was estimated by the pyridine fusion method l 2

. T he IR spectrum of the

784 IND IAN J. C HEM., SEC B, JULY 1999

CH3 CH3

I I C~=C-C-O-C~-C~-C~-C~-O-C-C=C~

II II . o 2 0

Suspension polymerisation

r r -CH,-CH-~-C-~-Cl+-C~-CH-~-C-~-CH-~-

~ ro~ ~ ro~ l;J I l;J l0J I ~~ r r r r r r

" I " " I 0 " c=0 c=

3a- 2% BDDMA-PS 3b- 4% BDDMA-PS 3c- 6% BDDMA-PS 3d- 8% BDDMA-PS

YI fy' yl fy' I I

-CH,-CH-CH,-C-~-CH-C~-O+-~-C-~-CH-~-I I 0i:J 0i:J

Schem e I- Prepara ti on of styrene- BDDM A copolymcr by sli spension polymerization.

2 1 +CH-CHz-h; 3 4

cd Scheme I1--Chl orolTlclh ylati o ll of Rc~ ill Ja

250 200 150 100 50 o -50

Figure I- Solid state Lic CP-M AS NMR spectrum o f styrene­BDDMA copo lymer

crosslinking and nature of the so lven\. T he swe lling characteri stics of the BDDMA-styrenc copolymers (3a-3d) were studied in a va ri ety of so lvents commonly employed in peptide synthesis. The swelling behaviour of a 2% BDDMA-cross lin ked chl oromethylpolys tyrene res in was compared with that of a 2% DYE-cross lin ked chloromethylpoly­styrene resin . The results obtained are given in Tab les Ia and Ib

ch loromethylated resin showed the characteristi c C-CI band at 680 em-I.

Swelling studies

The solvat ion properties of the polymer matrix is very important in SPPS since the reaction is taki ng place in a heterogenic medium lJ

. The extent of solvat ion of the solid phase determines the diffusion of soluble reagents, rates of different reacti ons taking pl ace and the yie ld and purity of the synthes ized peptides. The ex tent of swelling of cross linked polymers is deternlined by facto rs like polarity of the backbone, nature of the cross linking agent, degree- of

Synthesis of peptides A 2% crosslinked pol ymer support wi th a capac ity

of 1.8 mmol CI/g was selected for the peptide synthesis. Using this hi gh capacit y resi n, we have synthesized three model peptides and two acti ve sequences of the biologica l polypeptide DB II.J. The peptides were assembled on the resin by manual so li d phase strategy, using Boc am ino ac id chemi stry. The C-terminal amino ac ids Boc-Gly, Boc-Lys( -2CI-Z)­OH and Boc-Asp(-OBzl)-OH were incorporated as the benzyl ester to the chl oromethylated res in by the Gi sin 's cesium salt method I). The substituti on level was determined by Gi sin's pic ri c ac id method and was

..

r

AJIKUMAR el al .. : SOLID PHASE SYNTHESIS OF PEPTIDES 785

Table Ia-:- Swelling properties of resins 3a-3d

So lvation mUg resin*

Crosslink densi ty 2% 4% 6% 8%

Dich loromethane 7.43 5.34 4.29 2.76

Chloroform 7.92 5.59 4.32 2.82

N-Methylpyrrolidone 6.97 ·4.74 4.04 2.75

N.N-Dimethylformamide 6.82 4.32 3.84 2. 13

Tetrahydrofuran 8.2 1 5.86 4.40 2.85

Toluene 7. 13 4.80 3.83 2.47

Methanol 3.30 2.25 1.97 1.55

'Based on weight increase after equilibrating I g dry resin for 24 hr.

Table Ib--Comparison of solvat ion properties o f DYB and BDDMA-crossl inked chloromethyl polystyrene resins

Solven t Solvation (mUg resin) *

2% DYB-PS 2%BDDMA-PS resin resin

Dichloromethane 5.05 7.05

Chloroform 5. 12 7.35

N-Methylpyrrolidone 4.32 6 .56

N.N-Dimethylformamide 3.44 6 .20

Tetrahydrofuran 4.98 8. 10

Toluene 4.63 6.98

Methanol 2.02 3. 15

*Based on weight increase afte r equilibrating Ig dry resin for 24 hr.

found to be quantitative. Three model peptides (Gly­Leu-Gly, Phe-Gly-Lys and Phe-Thr-Gl y-Lys) were assembled by the DCC coupling method . The Boc group was removed by exposing the peptidyl res in to 30% TFA in DCM and subsequent neutralisation with 5% DIEA in DCM. The successive amino acids were then incorporated by the DCC mediated coupling. The extent of the coupling reaction was monitored by the semi-quantitative ninhydrin reaction l6

. For all the coupling steps double coupling was employed to ensure completion of the reaction . A two-fold molar excess of the Boc amino acid was used for the coupling reaction.

For the synthesis of biological peptides DBI 32-38 (Lys-Gln-Al a-Thr-Val-Gl y-Asp) and 45-50 (G ly-Leu­Leu-Asp-Leu-Lys), HOBt active ester coupling in NMP was performed . All the amino acylation reactions were almost complete at the first coupling step. Here a lso doubl e coupling was employed . The final cleavage of the peptides was effected by neat TFA. The purity of the crude peptide was checked by TLC and HPLC. The TLC of the peptides in the solvent system pyridine: acetic acid: water (50:35: 15)

showed a single spot. HPLC profiles of the two DBI fragments are shown in Figures 2a and 2b . The results of the amino acid analysis of the DBI fragments are given in Table II . The values agree with the theoretical amino ac id compositi on of the peptides.

This manuscript evaluates the effi c iency of the new BDDMA crosslinked polystyrene res in for the stepwi se solid phase synthes is of peptides. T hi s copolymer can be eas il y prepared with good mechanical stability and morphology. Moreover, the BDDMA crosslinked res in exhibits better so lvat ion properties both in polar and non-polar solve nts. The support is found to be very effi c ient for peptide synthesis including biologica l peptides

Experimental Section

General: Side chain protected Boc amino acids were purchased from Mis Si gma C hemical Co. USA and the other Boc amino ac ids were prepared by Schnabel's method I ? Polyvinyl a lcohol, BDDM A, TFA and thioani sole we re supplied by Ald rich Chemical Co. USA and styrene by Fluka AG, Switzerland . AR grade so lvents were used after distillation and purification. IR spectra were recorded on a Shimadzu IR 470 spectrophotometer using KBr pellets and the 13C CP-MAS NM R measurements were carried out on Bruker 300MSL CP-MAS instrument. HPLC was performed using Shimadzu SPD-IOA UV -VIS detec tor HPLC. Ami no acid compositions of peptides were determined by amino ac id analysis on Shimadzu amino acid analyser system.

Synthesis of BDDMA crosslinked polystyrene support

A mixture of styrene (22. 6 mL, 98 moI9'o), BDDMA (0.88 mL, 2 mol%), to luene ( 10 mL ) and benzoyl peroxide (500 mg) was suspended in a so lution of polyvinyl alcohol (PYA) (3 .5 g, 72000-100000 mo\.wt. ) dissolved in water 350 mL ( 1% PY A) and kept mechanicall y stirred at 600 rpm under an inert nitrogen atmosphere at a temperature of 85°C. After 6 hr the beaded resin was filtered and was hed with hot water to remove PY A, then it was soxhletted with acetone followed by methanol to remove all linear polymers and low molecular weight products . Polymers with crosslink densities 4%, 6%, and 8% were also prepared by the above procedure by varying the monomer mole ratios.

786 INDIAN 1. CHEM., SEC B, JULY 1999

~3'06 3·1

2·80

]\ 3·57 '"' ~4.26 "---

2'63 0' 00 Min 5·25 I I I I I I I I

0·00 308 Min 6·16

Time (min) Time (min)

Figure 2--HPLC profiles of the crude peptides. (a) Lys-Gln-Ala-Thr- Val-Gly-Asp (b) Gly-Leu-Leu-Asp-Leu-Lys on a rp C- t 8 column; solvent-O.I % TFA contail1ing acetonitrile; flow rate I mUmin.

Table II-Amino acid analysis of the peptides DBI 32-38 and 45-50

Amino acid DBI 32-38 DB145-50

Exp Obs Exp Obs

L 3 3.31

G . l.12 1.18

D l.01 l.01

K 0.93 0.95

Q l.11

A l.12

T 0.83

V 0.93

Chloromethylation of the styrene-BDDMA copolymer

The dry resin (l g) was swelled in DCM (10 mL) and refluxed with chloromethylmethylether (6 mL) and fresh ly prepared ZnCh in THF (0.3 mL) at 50°C for 4 hr with intermittent shaking. The resin was filtered, washed with THF, THF: H20 ( I : I), THFI 2N HCI (I: 1) and hot water till free from chloride and finally with methanol. Further purification of the resin was done by soxhelet extraction with THF. To estimate the chlorine capacity the resin (50 mg) was digested with pyridine (4 mL) for 6 hr and free chloride ions thus replaced were estimated by Volhard's titrimetric method . The calculated value was 1.8 mmole Cl/g.

Attachment of first amino acid to the chloromethylated resin

(a) Attachment of Boc-Gly. Boc-Gly (63 mg, 0.36 mmole) was dissolved in

minimum amount of ethanol and pH of the system

was adjusted between 7-8 by adding aq. solution of Cs2CO, . The solution was then co-evaporated with benzene and dried as the Boc-Gly-O-Cs salt powder and kept in vacuo over P20 5. The Cs-salt of Boc-Gly was then dissolved in minimum quantity of NMP and to this solution chloromethy lated resin (100 mg, 0.18 mmole CI) was added . The mixture was kept at 50-60°C for 32 hr with occasional shaking. The resin was filtered, washed with NMP, NMP:H20 (1 :1), methanol and DCM. The Boc amino acid appended resin was dried in vacuo. The amino group capacity was determined by picric acid method and was found to be 1.65 mmole NH2/g.

(b) Attachment of Boc-Lys (-2CI-Z)-OH and Boc­Asp( -OBzl)-OH

For appending Boc-Lys( -2CI-Z)-OH and Boc­Asp(-OBzl)-OH the procedure was simi lar to that adopted for Boc-Gly. Here, for 100 mg of resin (0.18 mmole Cl) the amount of amino acid used was 0.45 mmole 186 mg of Boc-Lys(-2CI-Z)-OH and 0.45 mmol, 145 mg of Boc-Asp(-OBzl)-OH. The amino capacity was estimated to be 1.11 and 1.35 mmole NH2/g for Lys and Asp respectively.

Peptide synthesis

Synthesis of Leu-Gly-Gly

Boc-Gly resin (100 mg) was placed in a solid phase shaking vessel and the Boc group was removed by treating with 30% TFA in DCM (4 mL x 30 min) followed by neutralization with 5% DIEA in DCM (4 mL, 2 x 5). The resin was then filtered and washed with DCM. The ninhydrin test was performed for the

-I

)

AJIKUMAR et al.. : SOLID PHASE SYNTHESIS OF PEPTIDES 787

indication of the free amino group. The second amino acid Boc-Gly (0.33 mmole, 57 .8 mg) was then coupled to the deprotected resin by the DCC method. Coupling of the third amino acid Boc-Leu was carried out after deprotection . Double coupling was employed at all stages to ensure 100% completion of reaction . After the TFA cleavage the peptide was obtained as a white powder (24 mg).

Synthesis of Phe-Gly-Lys and Phe-T hr-Gly- Lys

The Boc-Lys( -2CI-Z)-OH res in ( 100 mg) was taken in a solid phase reaction vessel and the tri - and tetra-peptides were synthesized by the same protocol employed in the synthesis of Gly-Leu-Gly. For each coupling two-fold molar excess amino acid was taken. The amounts of peptide obtained after the cleavage of 100 mg peptidyl reS \J1 were 28 and 30 mg respecti vel y.

Synthesis of L)ls-GIIl-Ala-T hr- Val-Gly-Asp (DBI 32-38)

Boc Asp(-OBzl)-OH res in ( 100 mg) was washed with DCM and the Boc group was removed foll owed by neutrali sation. Here in the synthesis of thi s peptide, instead of DCC coupling HOBt active es ter method was employed. The active ester of the amino ac id with HOBt was prepared in NMP using DCC. After the removal of DCU by filtrati on, the filtrate was used for the coupling. The amount of peptide obtained was 35 mg.

Synthesis of Gly-Leu-Leu-Asp-Leu-Lys.(DBI 45-50)

Starting from Boc Lys( -2CI-Z)-OH res in (100 mg) the amino acids assembled by the same protocol employed in the synthesis of the peptide DBl 32-38.Yield of the peptide obtained was 37 mg.

The protocol for the synthesis consisted of the following steps.

I. washed the res1l1 with DCM (2.5 mL, 5x l min .)

II. Boc deprotected using 30% TFA in DCM (4 mL x 30min).

111. washed with DCM (2.5 mL,5x I min).

IV. neutrali sed using DIEA in DCM (4 mL; 2x5min).

v. washed with NMP (2.5 mL, 5x I min).

VI. added active ester of Boc amino acid 111

NMP, shaken for 45min .

Vll. added DMSO (15 % of total volume of NMP) shaken for 15min.

viii. added DIEA (3.8 meg. ), shaken for 5min . IX. washed with methanol in DCM 33 :67

(2.5 mL, 5x lmin .). Cleavage of the peptide from the resin

The peptidyl resin ( 100 mg) was suspended in TFA (5 mL) containing thioani sole (0.5 mL) and kept at room temperature for 24 hr. The res in was filtered off using a sintered funnel and the filtrate was then rotaevaporated to remove TFA . To the res idue co ld ether (10 mL) was added to prec ipitate the peptide. The precipitated peptide was washed with ether till the peptide was free of any contaminating agent.

Amino acid analysis

The peptide samples ( I mg) were hydrolysed with TFA- 6N HCI ( 1:2 ) in evacuated sea led tubes at IIODC for 24 hr. The tube was broken and dri ed over NaOH and P20 S ill vacuo. The res idue was di ssolved in citrate buffer and loaded in the analyser.

Solvation studies Solvation studies were carried out using a sintered

crucible. The dry polymer ( I g) was placed in a sintered glass crucible (poros ity 3) and the res in was equilibrated with the so lvent for 24 hr. Excess so lve nt was removed by applying mild suction and the weight of the crucible with the swo llen res in was determined. The process was repeated till a constant weight fo r the swollen polymer was obtained. A blank experi ment was carried out with the empty crucible. Finally the solvation of the polymers were ca lcul ated as the volume of the solvent absorbed by unit weight of dry resin ( mUg) .

Acknowledgement .

The authors thank DST, New Delhi for finan cial support to KSD and CSIR, New Delhi for the award of a Junior Research Fell owship to PKA.

References

Merrifield R B, J Am Chelll Soc. 85. 1963,2 141). 2 Pill ai V N R & MUlter M, Tal' Cllrr Chelll . 106. 1982 . II I).

3 Arlherlon E & Sheppard R C. Solid I,I/{Ist' Ilclllide srnthesis: A practical approach, (IRL Press. Oxford) . 19!i9.

4 Renil M, Nagaraj R & Pill ai V N R. Tetrah edron. 50. 1994. 6681.

5 Jaya T Varkey & V N R Pillai . .I Peptide Res. 5 1. 1998.49.

6 Charmiehael J A Wall ace & lanel ark Lewi s. Biochelllistry, 36, 1997, 14733.

788 INDI AN J. CHEM ., SEC B, JULY 1999

7 Stephane Peluso, Pascal Dumy, Ian M. Eggleston, Patrick Garrouste & Manfred Muller, Tetrahedron, 53, 21 , 1997, 7231 .

8 Zhdanoya I Y, Kordzadze R N & PlyashkiYich Ya G (lP Parloy Inst. Physiol. Leningrad USSR): BYIIIl Eksp BioI Med ,1 I 1(2), 1991,165.

9 Birgi lla Agerberth, Anita Boman, Mats Anderson, Hans Jornuall , Yiktor Mull & Hans G Boman, Eli I' J Biochell1 , 216, 1993,623.

10 Sherington D C & Hodge P, (Eds) PolYlll er slIpported reactions in organic synthesis, (John Wiley & Sons, New York), 1980, P. 469.

II Fei nberg R S & Merrifield R B, Tetrahedron. 30, 1974,3209. 12 Stewart J M & Young J D, Solid phase l' elJlide sYlllhesis ,

(Pierce Chemical Co, Ill inois), 1984, p.54. 13 Gregg B Fields & Cynthi a G Fields . .I Alii Cifelli Soc. I 13,

1991,4202. 14 Erminco Costa & Alessandro Guidotti. Lifl' Sciences . 49 .

1991 ,325. 15 Gisin B F, Helv Acta, 56. 1973, 142. 16 Kaiser E, Cilescoll R L, Biss inger C D & Cook P I. Anal

Biochelll , 34, 1970, 27R. 17 Schnabel E,AnaIChelll , 702 , 1967, IXX .