Irreversible inhibition of serine proteases by peptidyl derivatives of α-aminoalkylphosphonate diphenyl esters

Vol. 161, No. 1, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

May 30, 1989

Page 143-149

IRREVERSIBLE

INHIBITION

DERIVATIVES

OF

Oleksyszyn

March

31,

PROTEASES

of Chemistry,

BY

PEPTIDYL

DIPHENYL

ESTERS

and James C. Powers*

Georgia

Atlanta, Received

SERINE

a-AMINOALKYLPAOSPHONATE

J6zef School

OF

Institute

Georgia

of Technology

30332

1989

SUNMARY : Peptidyl a-aminoalkylphosphonate diphenyl esters have been synthesized and shown to be effective inhibitors of serine proteases. Extending the peptide chain from a single a-aminoalkylphosphonate residue (kobs/[I]=2.5-268 M-1,-1 ) to a tripeptide or tetrapeptide derivative (kobs/[I]=7,000-17,000 M-ls-1) resulted in 65-2800 improvement in inhibitory potency and increased specificity. The rate of inactivation of chymotrypsin by MeO-Sue-Ala-Ala-Pro-HNCH(CH2Ph)P(O) (OPh)2 was decreased 5 fold in the presence of the substrate Sue-Val-Pro-Phe-NA (0.119 mM). Phosphonylated serine proteases are extremely stable since the half-life for reactivation >48 hrs for the inhibited elastases and at least 10 hrs for chymotrypsin. 0 1989 Academic PIBSS, Inc.

Phosphorylating classical the

inhibitors

for

identification

mechanism (2)

agents serine

of inactivation covalent

phosphorus

bond

to be good

transition

state

phosphorus

atom

phosphorus

(4) (5).

esters,

atom are

p-nitrophenyl

esters

phosphate

derivatives

extremely

toxic

*To whom reprint Abbreviations:

nitroanilide;

able

between

active

in

the

site crystal

analogues

due to the

proved

to be useful

reported

inhibitors

only

to react

with

should

serine

active

site

structure of the for

good site (6),

X-ray

proteases

leaving serine. but

in The serine

and the of are

considered

tetrahedral and 31P NMFI

organophosphorus

with

active

(1).

serine

presence

was observed

with be

the

diagnostically

enzymes

enzymes

tools are

compounds

nonspecific

due to reaction requests

the

found

some selectivity are

used

of the

The phosphorylated

since

been

phosphorylation

(3).

and have Most

and have

(DFP) are

of new proteolytic

atom has been trypsin

or p-nitrophenyl

proteases

involves

diisopropylphosphoryl

investigations

as diisopropylfluorophosphate

and classification

and a single

inhibitor

such

was

fluoridates groups In the

generally

at the case

of

organic

inhibitors

and

iPr, iso-propyl; elastase.

NA,

acetylcholinesterase. addressed.

HLE, human leukocyte Ph, phenyl; PPE, porcine

elastase; pancreatic

4-

0006-291x/89 143

$1.50

Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form resewed.

Vol.

161,

No.

1, 1989

To enhance

BIOCHEMICAL

the

selectivity

Lamden

synthesis of some phosphonate representative phenylalanine an inactivation 160 M-ls-1

rate with

fluorine

bond

unstable

and it

continuation

of this with

tetrahedral an extended

peptide significantly

constants

for

peptides

with

the

study,

leukocyte

proteases elastases,

with

reported

the

the

place

only

sequences esters

bovine

(8)

reported

of the

12-27

MATERIALS

AND

In a

synthesis

of

incorporating

group

M-ls-1.

specifically

Here

side

pancreatic

a of

of the rate

we report

and irreversibly inhibitor-enzyme

group

inhibition

C-terminal

porcine

only

assays.

carbonyl

and the

containing

stable

the

leaving

potency

but extremely

analogs

scissile

at the

chymotrypsin,

extremely

enzyme

and peptide

had

of a phosphorus-

derivative

during

inhibitory were

diphenyl

give

COMMUNICATIONS

chymotrypsin,

acid

The peptide

related

to

groups

in the

tetrapeptides

including

(7)

The presence

et al.,

leaving

decreased

aminoalkylphosphonate serine

Bartlett

moiety

substrate

M-ls-1

hydrolysis

substrate.

inhibitor

180,000 elastase.

rapid

different

phosphorus

RESEARCH

and Bartlett

a-aminoalkylphosphonic

undergoes

phosphonates

of

pancreatic

this

BIOPHYSICAL

derivatives of amino acids where the related structure Z-NH-CH(CH2Ph)P(O)(O-i-Pr)F

constant

porcine makes

AND

that

ainactivate and human complexes.

METHODS

Human leukocyte elastase (HLE) was supplied to us by Dr. James Travis and his research group at the University of Georgia. Porcine pancreatic elastase (PPE) was purchased from Worthington, bovine chymotrypsin from the Sigma Chemical Company. N-(2-hydroxyethyl)-1-piperazineethane sulfonic acid (HEPES), triphenyl phosphite, benzyl carbamate, aldehydes and all common chemicals and solvents were obtained from the Aldrich Co., Milwaukee WI. MeOSue-Ala-Ala-Pro-Val-NA (9), Sue-Ala-Ala-Ala-NA, and Sue-Val-Pro-Phe-NA (10) were prepared as previously described. Synthesis of Inhibitors. Diphenyl a-(N-benzyloxycarbonylamino)alkylphosphonates were obtained by the method described earlier (11) and converted by the dicyclohexylcarbodiimide coupling method to the desired triand tetrapeptide inhibitors. The final products were purified by thin layer chromatography (TLC) on silica gel. The purity of each compound was checked by lH-NMR, 31P-NMR, IR, mass spectroscopy, TLC and elemental analysis, and in each case the analytical results were consistent with the proposed structure. Racemic diphenyl a-aminoalkylphosphonates were used for the synthesis of the peptide derivatives and the products are mixtures of diastereromers. Full details of the syntheses will be presented elsewhere. The W spectrum and 31P NMR of several inhibitors (1, 3, 4, & 5) in HEPES buffer containing 10% Me2SO or Me2SO-d6 showed no changes after three days incubation indicating no hydrolysis. Enzyme Inactivation-Incubation method. Inactivation was initiated by adding a 25-50 )lL aliquot of inhibitor in Me2SO to 0.5 ml of a buffered enzyme solution (0.1-2.0 w) such that the final Me2SO concentration was 5-10 % v/v at 25OC. Aliquots were removed periodically and diluted into substrate solution (40-200 fold dilution). The residual enzyme activity was measured spectrophotometrically in 0.1 M HEPES, 0.5 M NaCl, pH 7.5 using the following substrates: chymotrypsin, Sue-Val-Pro-Phe-NA (0.476 mM) (10); HL elastase, MeO-Sue-Ala-Ala-Pro-NA (0.482 mM) (10); PP elastase, Sue-Ala-Ala-Ala-NA (0.714 mM) (10) - Peptide p-nitroanilide hydrolyses were monitoring at 410 run (12). First order inactivation rate constants (kobs) were obtained from plots of ln vt /v, vs. time. Inactivation rate constants shown in Table 1 are typically 144

Vol.

161,

No.

1, 1989

the average concentration

BIOCHEMICAL

AND

BIOPHYSICAL

of the duplicate or triplicate are shown in Table 1.

RESEARCH

experiments

COMMUNICATIONS

and inhibitor

Dephosphonylation Kinetics. The dephosphonylation rate constants of inactivated enzyme (X5% of activity in most cases) were measured after removal of excess inhibitor by centrifuging with Centriconmicroconcentrators twice at 0 OC for 1 hr after the addition of fresh buffer. The solution was warmed to 25 OC, aliguots were removed at intervals, and the enzymatic activity was assayed as described above. The half time for dephosphonylation was obtained from plots of In (vo-vt) vs. time where v. is the enzymatic rate under the same conditions for the uninhibited enzyme. Determination of Inactivation Rates in the Presence of SubstrateProgress Curve Method. In some cases, kobs/[I] values were determined the presence of substrate as described by Tian and Tsou (13). For example inactivation of chymotrypsin (0.109 J.lMM)by MeO-Suc-Ala-Ala-Pro-NHCH(CH2Ph)P(O) (OPh)2 (1.5-4.0 w) in the presence of 0.476, 0.357, and 0.119 Sue-Val-Pro-Phe-NA was measured by addition of a 0.01 mL aliquot of enzyme a substrate and inhibitor solution containing 10% of Me2SO. The increase absorbance was monitored (410 nM) with time until no further release of pnitroaniline was observed. The kobs/[I] values were calculated from plots log([Pl,-[Pit) vs. time, where [Ploo and [P]t are the concentration of pnitroaniline after total inactivation and at time t, respectively.

in mM to in of

RESULTS Inactivation for

kinetics.

inactivation

of chymotrypsin,

aminoalkylphosphonate first-order lives.

The

Inactivation

plots of the

are

reported

remained

the

was very

phenylalanine

constant

three

analog

constants

serine

for

1.

However,

4 is most reacted

clear

selectivity

reactive

slowly

with

with

Inactivator

[II WI

3 4 5 6

'Vonditions described 1.7 uM.

were 0.1 Hepes, under Materials dEnzyme concentration

lOPh)2 (OPh)2 (OPh)2 (OPh)2 (OPh12 (OPh)p

four a-(N-

of Val

chymotrypsin

HLE (6.0

halfand Phe

was observed M-ls-l),

with

since a rate

and did

PROTEASES C-TERMINAL

not

BY DIPHENYL DIPHENYL

PPEC

Chymotrypsinb

Z-HN-CH(i-Pr)P(O) Z-Ala-Ala-HN-CH(i-Pr)P(O) MeO-Sue-Ala-Ala-Pro-HN-CH(i-Pr)P(O) Z-NH-CH(CH2Ph)P(O) MeO-Sue-Ala-Ala-Pro-HN-CH(CH2Ph)P(O) Z-Phe-Pro-HN-CH(CH2Ph)P(O)

than

by diphenyl

TABLE 1. RATE CONSTANTS (&,be/[I]) FOR THE INACTIVATION OF SERINE l-(N-BENZYLOXY-CARB~NY~IN~)ALKYLPHOSPHONATES AND PEPTIDEs WITH l-AMINOALKYLPHOSPHONATE RESIDUESa

1 2

The psuedo

greater

proteases

(k,bs/[Il)

a-

1 and 4 (analogues

slow.

of 260 M-ls-l,

rate in Table

linear

benzyloxycarbonylamino)alkylphosphonates respectively)

order

PPE, and HLE by diphenyl

derivatives

inactivation

second

k,&

(11

(M-l-l)

50 240

180 58 11.0

11,000

5.0

17,000

k,bs/ [II (M-h-1)

III (I(M)

k,bs/ [II

2.5

26 14.0 4.9

90 1300

0.4

50

2.4

24.0

340

9

7100 NI

21 260

[II WM)

HLEd

58 220 93

0.34

NI

7100

30

6.0

120 51

7.2

0.5 M N&l, pH 7.5 at 25 OC. Rate constants were measured as bEnzyme concentration 1.9 JIM. CEnzyme concentration and Methods. less than 5% inhibition after 4 hrs. 0.32 FM. NI,

145

(M-is-l)

1.5

Vol. 161, No. 1, 1989 react

with

PPE.

elastase

BIOCHEMICAL In contrast,

(90 M-ls-1)

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

valine

the

and PP elastase

analog

(2.5

is

1

~-1,-l),

most

but

reactive

is

a very

inhibitor for chymotrypsin (0.4 M-ls-1). The tripeptide Ala-Ala-HN-CH(i-Pr)P(O) (OPh)2 2 was a much more effective and had rate

constants

to 1) and 1300 aminoacid respectively inhibited used

M-l.s-l

residue

tetrapeptide

of 340 M-ls-1 with

again

derivative

2800

and 79 fold

chymotrypsin

inhibitors 17,000

rate

inhibit

when

M-ls-lfor

than

compared

phosphonate

to the

inhibitors

and inhibits

only

amino

are

quite

HLE very

Substrate

analogs with

are

respectively

acid

derivative

specific

Val-Pro-Phe-NA M-ls-l

in

Addition

900,

absence

1070

of the

Dephosphonylation

after

removal

centrifuging

twice

at 0 OC.

enzyme-inhibitor activity slowly

respectively). enzyme

With

of

complexes

with

These

inhibit

were

of

substrate

11,000

the

most

Mdlsml

for

increases

in

peptide

not

react

with

to

the

PPE

enzyme

decreases in inactivation ) for the inactivation of

and 2200

of 0.476, M-IS-~

0.357

rate

and 0.119

respectively

Less

of excess This

than

inhibitor

mM Suc-

compared

is

elastases

consistent

all

reactivation

5% of

the

to

enzymatic

by diluting with

adduct. Chymotrypsin upon further standing

the

not

11,000

substrate.

Kinetics.

was regained covalent regained

3 also

3 did

6 does

both)

concentrations

42 and 65 fold 4.

for

Mmlsml

at the

constants

since

j.lM) by 5 in presence

were the

and the

of phenylalanine

rate

compared

of another

potency

(9 or 4.9 w),

mixture resulted in significant The rate constants (kobs/[I] (0.109

but

Z-

slowly.

Protection.

chymotrypsin

improvement

The tetrapeptide

rate,

poor

Addition

inhibitory

1.

elastases

6, which

fold

PPE and HLE (7100

lower

of chymotrypsin

chymotrypsin

incubation constants.

better the

The peptide

effective 5 and

3 inhibited

HL

valine derivative elastase inhibitor

improvement). improved

at a 340 fold

to effectively

PPE (136

HLE (14 fold

significantly

valine

chymotrypsin.

with

toward

the

gives

than

and

of a

by 4 and 5 10 and 16 hrs

very

greater

activity

buffer

formation

inactivated 25 OC (tl/2=

inhibitors

half-lives

with

stable

inhibitor-

48 hrs.

DISCUSSION There

are

at

new therapeutically inhibitors:

stability

have

are

stability

increased toward

phosphorus fast

factors

determinated inhibitor

hydrolysis.

nonenzymatic

should

with

in

For

very

electronegative

example,

proteases, attachment inhibitors,

Decreasing 146

in designing proteases

Reactivity

electrophilicity

serine

some excellent

hydrolysis.

serine

and specificity.

by the

toward

be considered

irreversible

stability

reactivity

has resulted

which

organophosphorus chemical

Organophosphorus

atom.

very

three

useful

reactivity,

chemical will

least

but

of the

phosphorus

leaving

groups

also

decreased

of a fluorine but

and

these

the electronegativity

atom

also

to

undergo of the

Vol. 161, No. 1, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Active Site Substrate Binding Site JJ

Active

zE--His,,

3

R-CO-NH-dH-;:;;;

*

/Serl95 0 I, ‘OPh 0 ,. ‘, y ‘! N N

R-CO-NH-dH-P’

0

Figure

1.

peptidyl

leaving

group

but

at the

are

structural

protease

Proposed derivative

and the

ability

but

esterase.

organophosphorus

inhibitor

factors

The peptide reported

here

groups

are

electrophilic proteases. bond

one solution

phosphonate

moiety

to undergo

reaction

and this inhibitor case

slowly

Peptide

with

or 31~-NMR

stable

serine

alkylphosphonate chymotrypsin corresponding poorly partially

with

but useful moiety

inhibitors peptide the

by the

rate fact

between

the

active

with

incubation are

in

reported too

compounds.

serine by the

The rate

In addition, elastases

constants that

the

of 12-27

buffer

earlier

However, M-ls-1.

147

diphenyl

a-

of

of

a substrate

the

enzyme>48 hrs).

=

activity

diphenyl

observed

phosphoramides,

of a covalent

In

was very

10 hrs.

hydrolytically Tetrapeptides

(8).

of serine

(tl/z

for (7)

esters

in either are

unstable with

potent

inhibitors

irreversible KM value

be

for

as

the

chymotrypsin explained

a carboxylic

of

of

to be considered the a-amino

they inhibit This can unlike

are

the UV spectrum The fluoridates

3 days.

inside of the peptide chain are and have KI values close to the substrate

two phenoxy

of inactivation

presence

by 4 and 5, enzyme

were

of the

formation

serine in the

of about

esters

atom sufficient

site

consistent 1).

several

an ideal

The presence

(Figure

protease

diphenyl

phosphorus

site

such

and reactivity.

of a-aminoalkylphosphonic

acids

proteinases

balance

the

active

inactivated

after

a-aminoalkylphosphonic therapeutically

is

and no changes

spectrum

serine

of designing

significantly

a half-lives

derivatives

hydrolytically

hydrolases

active site involvement. extremely stable with the

of chymotrypsin

regained

serine

serine

problem

with

of the

decreased

demonstrates complex is

different

specificity

makes

derivatives

by 5 is

between

inhibitor.

evidence

due to phosphonylation

aminoalkylphosphonate

the

to the

protease

and could

and other

a delicate its

inhibitor,

serine

inhibitor,

irreversible

a

when there

a natural

of a-aminoalkylphosphonate

The available

chymotrypsin

only

an ideal

determines

serine

on the

Thus,

derivatives

organophosphorus

not

represents

which

results

between

with

of the

Specificity

proteases

oxyanion hole

stability

organophosphorus

serine

H-His5,

a serine protease acid diphenyl ester.

chemical

similarities

potential

between

of

rates.

to discriminate

also

as acetylcholine competing

the

inactivation

and stereochemical

the

proteases

increases

of lower

substrate

provide

scheme for the reaction of an a-aminoalkylphosphonic

substantially

cost

Site

‘1

amide,

very

Vol. 161, No. 1, 1989 prefers

the

similarity

conformation

between

scissile

bond.

phosphonic the

cis

BIOCHEMICAL

the

of the

inhibitor the

alkylester

electrophilicity

moiety

of the

nucleophiles,

and results

The inhibition aminoalkylphosphonic

presence

of the

derivatives

diphenyl

For example,

those observed with chloromethyl ketones chloromethyl

ketones

the

toxicity

probable

various

cellular

derivatives with after

3 hrs

suffer

at the

with

PPB.

300-500

The tripeptide has been

3 is

Racemic

reacts correct

are

reacting slower

isomer

enzyme once the

and thus pure

are

than

because

of

to alkylate

show low than

for

reactivity

5% inhibition

1, 2,

& 5). by of the

for

chymotrypsin.

which (17),

is

is

preferred

indeed

specific

since

esters peptide

reacts

with

with

not

to valine

than

best

rate

and does

related

elastases

it

in

the

and has a inhibition

chymotrypsin

target

structurally

specific

a sequence

quite

better

better

agents

a Pl phosphonate

with

are

determined mainly the Sl pocket (16)

discovered

with

shown

the

react

(2 & 3) are

chymotrypsin.

irreversible

inhibitors

a-aminoalkylphosphonates in Table

with

only

or not

rate

diastereromers

is

It

which

rate

constants

constants are

the

the

acid

other

reported

148

in

table

in

that

the

second

isomer that

residue) isomer

1 should

is

the

of

suggest

due to non-productive

isolated.

used

mixtures

likely

experiments

as an L-amino that

is

while

Preliminary

possible

were

nonequivalent

one stereoisomer,

at all. It

inhibition

1 are

by 31P NMH measurements.

(same stereochemistry

the

is with

diphenyl

reactive

of diphenyl

peptides

proteases. the

1

Table

enzymes

also

(71, but protease

PPE (15).

mixtures

more

in

contains

than

one of the

as shown

serine

decreasing

more

for

and the

diastereromers

is

slower

phosphonate fold

reported

synthesis

It

times

Peptide

at least

enzymes

M-is-l.

10,000

related

such as peptide 15). Peptide

and also

of chymotrypsin-like we have

the

a-aminoalkylphosphonate

as 4, 5, & 6, are

6, which

with

ability

(less

with

toward

a-

rates

esterase

inhibitors residue

of the decreases

by Bartlett serine

their

limitation

esters

which

of 17,000

HLE at least

this

such

inhibitor

constant

Diphenyl

region or the

compared

inhibitors M-ls-l,

from

given

diphenyl

by a number

chymotrypsin

(15).

of peptide PI amino acid

by

as therapeutic

result

the

constants.

inhibition

potential

should

acetylcholine

derivative

substrates

little

from

to phenylalanine,

The tripeptide

rate

reported irreversible

the

the

reactivity

may be moderate

elastase

concentrations

The peptide

related

with

have

hydrolase

The specificity interaction of the enzyme.

with

in

destroys

significantly

proteases

other peptidyl irreversible (kobs/[I] values of l-1560

nucleophiles

serine

inactivation

fluorides peptidyl

This

phosphonamidate

decreases

of serine

esters

which

do not

the

atom,

low

constants

hydrolytically unstable phosphonyl quite respectably compared to other inhibitors.

these

(14).

at least

in

phosphorus in the

rate

PO and NH groups

and a substrate

Additionally,

acid

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

the

reacts

faster

effectively binding

to the

be improved

Vol.

161,

No.

It esters

is are

BIOCHEMICAL

1, 1989

clear useful

therapeutic

that serine

agents.

should

derivatives

protease

They.are

irreversibly inhibited of the diastereromers sequences

peptide

group

ACKNOWT.EDGMEiNT.

Institutes

is

close This

of Health

to that research

and have stable

with several of peptides

improve

RESEARCH

COMMUNICATIONS

of a-aminoalkylphosphonic

hydrolytically

inhibitors. Additional specificity phenoxy groups with other alcohols leaving

BIOPHYSICAL

inhibitors

derivatives and synthesis

significantly

AND

the

diphenyl

potential

utility

and form

extremely

serine proteases. with more specific

potency

and specificity

could also be introduced or phenols as long as the

as stable

Resolution amino acid of the

by replacing pK, of the

the

of phenol. was supported

by a grant

from

the

National

(HL29307).

REFERENCES 1.

2. 3. 4. 5. 6. I. a. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Hartley, B. S. (1960) Ann. Rev. Biochem. 29, 45-67. Schaffer, N. K., May, C. S., and Summerson, W. H. (1953) J. Biol. Chem. 202, 67-76. Stroud, R. M., Kay, L. M., and Dickerson, R. E. (1974) J. Mol. Biol. 83, 185-208 Kossiakoff, A. A., and Spencer, S. A. (1981) Biochemistry 16, 654-664. Gorenstein, D. G. (1984) In P-31 NMR: Principles and Application (D. G. Gorenstein, Ed.), pp. 130-131. Academic Press, New York. Nayak, P. L., and Bender, M. L. (1978) Biochem. Biophys. Res. Commun. 83, 1178-1182. Lamden, L. A., and Bartlett, P. A. (1983) Biochem. Biophys. Res. Coaunun. 112, 1085-1090. Bartlett, P. A., and Lamden, L. A. (1986) Bioorq. Chem. 14, 356-377. Nakajima, K., Powers, J. C., Ashe, B. M., and Zimmerman, M. (1979) J. Biol. Chem. 254, 4027-4032. Tanaka, T., Minematsu, Y., Reilly, C. F., Travis, J., Powers, J. C. (1985) Biochemistry 24, 2040-2047. Oleksyszyn, J., Subotkowska, L.,and Mastalerz, P. (1979) Synthesis 1979, 985-986. Erlanger, B. F., Kokowsky, N., and Cohen, W. (1961) Arch. Biochem. Biophys. 95, 271-278. Tian, W.-X., and Tsou, C.-L. (1982) Biochemistry, 21, lOLE-1032. du Plessis, M.P., Modro, T.A., and Nassimbeni, L.R. (1982) J. Org. Chem. 47, 2313-2318. Powers, J. C., and Harper, J. W. (1986) In Proteinase Inhibitors (A. J. Amsterdam. Barrett and G. Salvesen, Eds.), pp. 55-152. Elsevier, Nomenclature of Schechter, I., and Berger, A. (1967) Biochem. Biophys. Res. Commun. 27, 157-162. Powers, J. C., Tanaka, T., Harper, J. W., Minematsu, Y., Barker, L., Lincoln, D., Crumley, K. V., Fraki, J. E., Schechter, N. M., Lazarus, G. K., Nakashino, K., Neurath, H., and Woodbury, R. G. (1985) G., Nakajima, Biochemistry 24, 2048-2058.

149