Development of synthetic peptide substrates for the poliovirus 3C proteinase


Material Information

Development of synthetic peptide substrates for the poliovirus 3C proteinase
Physical Description:
ix, 122 leaves : ill. ; 29 cm.
Weidner, Jeffrey Robert, 1962-
Publication Date:


Subjects / Keywords:
Research   ( mesh )
Cysteine Endopeptidases   ( mesh )
Endopeptidases   ( mesh )
Polioviruses   ( mesh )
Chromatography, High Pressure Liquid   ( mesh )
Oligopeptides -- diagnostic use   ( mesh )
Recombinant Proteins -- diagnostic use   ( mesh )
Substrate Specificity   ( mesh )
Department of Biochemistry and Molecular Biology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Biochemistry and Molecular Biology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1989.
Bibliography: leaves 114-120.
Statement of Responsibility:
by Jeffrey Robert Weidner.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001473005
oclc - 25072841
notis - AGY4765
System ID:

This item is only available as the following downloads:

Full Text



Jeffrey Robert Weidner




This dissertation is dedicated to
my mother and father, Susan and Gary Weidner.


I would like to express my appreciation to my advisor,

Dr. Ben M. Dunn, for his support and patience over the last

seven years. In addition, I express my gratitude to the

members of my committee for their advice and assistance in

the preparation of this dissertation. I thank Dr. Ray

Roberts for his advice and a most enjoyable collaboration.

I owe a huge debt to Mr. Benne Parten and Mrs. Alicia

Alvarez for their technical assistance in peptide sequencing

and synthesis as well as their friendship. Special thanks

to Dave Jewell, a kindred spirit, for the many stimulating

hours of discussion as well as his appreciation for well-

brewed beer. Finally, I would like to thank the following

people for their help and/or support: Dr. Jan Pohl, Dr.

Mike Pennington, Paula Scarborough, Dr. Nancy Denslow,

Phillip Hartzog, Ian Hornstra, Dr. Peter McGuire, Kim

McCormick, and Barry Woods.




ABSTRACT . ............ viii



Proteolytic Processing of Viral Proteins .. 1
Picornaviruses . . 5
Polyprotein Processing in Picornaviruses 11


Introduction . . 26
Experimental Procedures . .. 27
Results . ... . 34
Discussion . ... . 68


Introduction ... . 75
Experimental Procedures . ... 78
Results . ... . 80
Discussion ... . 103



REFERENCES .. . . 114



AcOH acetic acid

C,1 octadecyl silica

C-terminal carboxy terminal

CM carboxy methyl

CPM 7-diethylamino-3-(4'-


4-methyl coumarin

DABITC 4-dimethylaminophenylazophenyl-4'-


DCM dichloromethane

DEAE diethylaminoethyl

dHO, distilled water

DIEA diisopropylethylamine

DMF dimethylformamide

DMSO dimethylsulfoxide

DTT dithiothreitol

EDTA ethylenediaminetetra-acetic acid

EMCV encephalomyocarditis virus

FMDV foot and mouth disease virus

g grams

h hours

HC1 hydrochloric acid







2-ethanesulfonic acid

anhydrous hydrogen fluoride

high performance liquid


concentration to achieve 50%


inhibitor binding constant

Michaelis constant for an enzyme

catalyzed reaction




morpholinoethanesulfonic acid

microgram (106 grams)

microliter (106 liters)

micromolar (106 molar)

milligram (103 grams)

millimolar (103 molar)


nanomole (10' mole)

nuclear magnetic resonance



-log (H']

picomole (10.12 moles)















PTH phenylthiohydantoin

TFA trifluoroacetic acid

Tris tris hydroxymethyll) aminomethane

Vmax maximal velocity for an enzyme

catalyzed reaction

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Jeffrey Robert Weidner

December, 1989

Chairman: B.M. Dunn, Ph.D.
Major Department: Biochemistry and Molecular Biology

Picornaviruses, such as polio, translate their entire

genome as a single polyprotein which must be proteolytically

processed to produce the mature viral proteins. This

processing is essential to viral replication and is entirely

mediated by the virus. A majority of the cleavages are

catalyzed by the virally encoded cysteine proteinase, 3C.

We report here the design and synthesis of a series of

oligopeptide substrates, for an HPLC assay for poliovirus 3C

proteinase activity, based upon native 3C cleavage sites.

These substrate peptides, in conjunction with recombinant

polio 3C proteinase, were used to develop a quantitative in

vitro assay for this enzyme. Such a quantitative assay is

essential for the detailed biochemical characterization of

the proteinase. The pH dependence of the enzyme as well as


interactions with potential inhibitors has been examined.

These peptide substrates can also serve as probes to

investigate structural requirements for substrate binding to

the 3C proteinase. Our data support the premise that there

is a conformational component required for substrate

recognition by 3C proteinase.

Finally, we report the development of a quenched

fluorescent peptide substrate for the assay of polio 3C

activity, Polio 1.2CD. This is the first reported substrate

suitable for the continuous monitoring of picornaviral 3C

proteinase activity. This assay greatly facilitates the

rapid evaluation of potential inhibitors as well as new

substrates. Quantitative evaluation of the inhibition of

polio 3C proteinase by egg white cystatin obtained with

Polio 1.2CD is in agreement with data obtained with other

unlabelled peptide substrates.



Proteolytic Processing of Viral Proteins

Proteolytic processing occurs in a wide range of

biological events including zymogen activation, hormone

synthesis, protein trafficking, antigen processing, and

enzyme cascades. Specific cleavage of proteins may be the

most prevalent post-translational modification of protein

structure. A number of viruses have evolved which are

dependent upon proteolytic processing of virally encoded

proteins at some point in their replication cycle (Table 1-


As Table 1-1 shows, virally encoded proteinases are

often associated with RNA viruses. Krausslich and Wimmer

(1988) suggest that the synthesis of polyproteins may have

arisen in these viruses in response to pressures to minimize

the size of the genome of these viruses. RNA viruses must

rely upon RNA-dependent polymerases for the replication of

their genomes. These enzymes lack the proofreading

functions inherent in most DNA polymerases and are therefore

subject to higher mutational rates. By encoding all or part

of their genomes as a polyprotein, these viruses can

eliminate many extraneous non-coding regions associated with

Table 1-1. Virus Associated Proteinases

group genetic origin of protease
material protease type

retroviruses RNA virus aspartic

alphaviruses RNA virus serine

flaviviruses RNA virus serine

togaviruses RNA virus serine

picornaviruses RNA virus cysteine

comoviruses RNA virus cysteine

nepoviruses RNA virus cysteine

potyviruses RNA virus cysteine

myxoviruses RNA host ?

paramyxoviruses RNA host ?

arenaviruses RNA host ?

coronaviruses RNA host ?

tymoviruses RNA virus ?

adenoviruses DNA virus serine

poxviruses DNA virus ?


gene regulation, transcription, or translation. Any

reduction in the size of the genome would thus reduce the

likelihood of deleterious mutations. DNA viruses, which

make use of the more meticulous DNA-dependent polymerases,

are not subject to the same pressures to minimize the size

of their genomes. The synthesis of groups of proteins as

polyproteins also provides a mechanism for the production of

stoichiometric quantities of interacting proteins.

Polyprotein synthesis requires the presence of a highly

specific proteinase in order to release the individual

proteins from their precursors. These proteinases, in

contrast to degradative enzymes, must be highly specific in

order to insure that proper processing occurs. In addition,

since viruses are dependent upon a number of host cell

systems for their replication, these proteinases must not

indiscriminately cleave host cell proteins. The high level

of specificity for these proteinases, when coupled with

their necessity for viral replication, makes these enzymes

excellent targets for antiviral therapies. Virally encoded

proteinases are optimal targets since disruption of

proteinase activity has a reduced chance of disrupting host

cell functions.

Most virally encoded proteinases are synthesized by

positive-stranded RNA viruses and retroviruses (Krausslich

and Wimmer, 1988). These viruses all have genomic RNA,

which is the same polarity as viral mRNA. Proteolytic


processing of glycoproteins from negative-stranded RNA

viruses also occurs; however, these cleavages are generally

mediated by host cell enzymes and occur in vesicular


Viral polyproteins can be grouped into three classes.

The most basic type is the expression of the entire genome

as a single polyprotein precursor. This is characteristic

of picornavirus, flavivirus, and potyvirus replication

(Krausslich and Wimmer, 1988; Sonenberg, 1987; Wellink and

van Kammen, 1988). In this scheme all viral proteins are

produced in equimolar amounts.

Togaviruses and some plant RNA viruses synthesize a

polyprotein from the 5' end of the genomic RNA. This

polyprotein encodes for nonstructural proteins including a

virus-specific RNA polymerase. This polymerase then

synthesizes mRNA transcripts from the 3' end of the genome.

These structural proteins can then be synthesized in amounts

exceeding the synthesis of nonstructural proteins.

Retroviruses synthesize two polyproteins from the 5'

end of their genome. The smaller contains all of the

proteins from the gag region of the genome, while the larger

is a gag-pol fusion polyprotein. This latter polyprotein is

synthesized as the result of either a ribosomal

frameshifting event or the read through of a single

termination codon. Synthesis of the gag-pol fusion


polyprotein is thus a rare event allowing for differential

expression of the gag proteins and the protease and reverse

transcriptase which are expressed only in the larger fusion


Most of the information available on viral protein

processing has been obtained by observing changes in viral

protein sizes over time with SDS-PAGE. Only a handful of

viral proteinases have been isolated at this time. Studies

of viral replication in the presence of proteinase

inhibitors in cell culture as well as analysis of nucleic

acid sequence data have provided much of the information

about viral proteinase structure, mechanism, and

specificity. A more detailed biochemical characterization

of these enzymes is essential for understanding viral

replication and for using these enzymes as targets in

antiviral therapies. Currently efforts are underway in a

number of laboratories to characterize these systems, with

most of the attention being focused on the retroviruses and



The family picornaviridae consists of related small RNA

viruses and contains a number of human and animal pathogens.

As Table 1-2 shows, this family can be further divided into

4 genera: enteroviruses (poliovirus, hepatitis virus A,

coxsackieviruses, etc.), cardioviruses (encephalomyocarditis

virus [EMCV], mengovirus, etc.), rhinoviruses, and


-4 t
0 O*.-4 a

0 OW q0 0
r> >c N r m HI0

r. -i to W4 4 40
0 I 0 40 I

)a 4a 0 ( 1
fi ~4- 0 0
4 o -4 4 C O

W (0 OO (4 0

H > W 4 U

0 0 w *. ,

S -,. 3 1 0 0H
0 04J

S4. 0 40( 0
> > a 0 c

0 O 4

to I 04 0 .
to e o i, :n

0O 02 V 9
C A 0> 4 U
4) 3JC 40 r.

00 .4 C) 0
0 0O 2 O tm
C 00 N, .0 > m 0 4

E1 0
m 44 0
'm 4

04 )
S1 0*H (0 4
to 0 4-1 > 4

0> 4 0 0) 0- 0
o 0 r r- a)
OO 0 0 B0 C. 0 n
0 4 uo 4 40 o -H

EH 0 z0 0 z 0 H H
m c i >i CT>
S U ).0 N C. X> U 4-

0g V

E- a a)
01 0

.0 0 C -4 -H O
0 a 0 0 0 0 -H 0

E-i U v, M E Q

aphthoviruses [foot and mouth disease viruses (FMDV)].This

family has been well studied since it contains a large

number of important pathogens.

The picornavirus genome consists of a single, positive-

strand RNA molecule ranging in size from 7102 bases for

human rhinovirus 2 (Skern et al., 1985) to 8450 bases in

FMDV-A10 (Carroll et al., 1984). This monocistronic RNA has

a poly(A) tail at the 3' end and is capped at the 5' end by

the virally encoded VPg protein (Lee et al.,1976), thus

resembling eukaryotic mRNA. The viral genome can be

translated directly by host cell ribosomes. The VPg moiety

is not present on viral mRNA molecules and is not necessary

for infectivity or translation (Flanegan et al., 1977). In

addition, the cardioviruses and the aphthoviruses possess a

tract of poly(C) residues near the 5' end of the genome

(Chumakov and Agol, 1976; Harris and Brown, 1976).

Picornaviral RNA contains a single, long, open reading

frame encoding a viral polyprotein of approximately 250 KD.

Proteolytic cleavage of this polyprotein yields all of the

picornaviral proteins. The polyprotein can be divided into

three regions in order from the N-terminus, P1, P2, P3.

The P1 region is comprised of four proteins (1A, 1B,

1C, and 1D) which are the viral capsid proteins. In the P2

region, protein 2A has been identified as the proteinase in

the enteroviruses (Toyoda et al., 1986) and the rhinoviruses

(Sommergruber et al., 1989) responsible for the cleavage at

the P1-P2 junction. However, protein 2A is too small, 25

amino acids in encephalomyocarditis virus (EMCV) (Palmenberg

et al., 1984), to function as a proteinase in either the

cardioviruses or the aphthoviruses. No function has yet

been assigned to either of the other two P2 proteins. The

P3 region encodes for four proteins including VPg; 3C, the

proteinase responsible for most polyprotein cleavages; and

the RNA polymerase 3D.

The replication cycle of a typical picornavirus is

reviewed by Rueckert (1985) and illustrated in Figure 1-1.

Infection begins with virus binding to cellular receptors

located on the plasma membrane. These receptors mediate

delivery of the genome to the cytoplasm. This RNA molecule

is then used to direct synthesis of the polyprotein which is

processed to yield all the viral proteins including the RNA

polymerase. This polymerase is used for the synthesis of a

negative-stranded RNA molecule.

The newly synthesized RNA molecule can then serve as

the template for the synthesis of additional positive-

stranded RNA via a multistranded replicative intermediate.

Initially most of the positive stranded RNA is used as viral

mRNA to direct protein synthesis. As the protein

concentration increases, more of these RNA molecules are

covalently attached to VPg at their 5' end and go on to be

packaged into virions. These completed virions remain

14S fTAMER t,




Figure 1-1. Overview of the picornaviral replication cycle.

Figure 1-2. Diagrammatic representation of the picornaviral
capsid. The external proteins, VP1 (ID), VP2 (1B), and VP3
(1C), form an icosahedron. The protomer subunit is
designated by the thick lines.


within the host cell until released by infection-mediated

cell lysis.

The capsid is an icosahedral shell, approximately 25 nm

in diameter, composed of 60 identical subunits known as

protomers (Finch and Klug, 1959). Protomers are composed of

a single copy of each of the viral capsid proteins, 1A, 1B,

1C, and ID (Rueckert et al., 1969). A typical picornavirus

capsid structure is illustrated in Figure 1-2. The crystal

structure for the capsids of both poliovirus (Hogle et al.,

1985) and mengovirus (Luo et al., 1987) has been solved to a

2.9 A resolution.

Polyprotein Processing in Picornaviruses

The existence of large polyproteins and their

proteolytic processing into smaller proteins was first

detected in enteroviruses (Sumers and Maizel, 1968; Holland

and Kiehn, 1968; Jacobson and Baltimore, 1968). Highly

unusual conditions, such as growth in the presence of amino

acid analogs (Jacobson and Baltimore, 1968) or Zn2 ions to

inhibit the viral proteinases (Butterworth and Korant,

1974), are required for the detection of the intact


Figure 1-3 shows that all picornaviruses share a common

polyprotein structure and processing scheme. While there

seem to be three distinct proteolytic activities, all

cleavages appear to be mediated by viral molecules and not

host cell proteinases (Shih et al., 1979; Wellink and

I C I, bl : I I 3D POLIO

Figure 1.3. Cleavage sequences within picornaviral

polyproteins. Representative picornaviral genomes (polio,
rhino-14, EMCV, FMDV-AIO, and hepatitis-A) are presented
schematically. Proteolytic cleavage occurs between the
indicated amino acid pairs. FMDV cleavages between tandemly
L A I IC 13 3r0rrH' IC 38 "WM

III IC 1i3 a 5 l 30 33 ) 4 HI ?I0I A *

Figure 1.3. Cleavage sequences within picornaviral
polyproteins. Representative picornaviral genomes (polio,
rhino-14, EMCV, FMDV-A10, and hepatitis-A) are presented
schematically. Proteolytic cleavage occurs between the
indicated amino acid pairs. FMDV cleavages between tandemly
linked 3B proteins occur at identical EG dipeptides.
Question marks denote sites whose sequence is not precisely
known, or whose position is suggested on the basis of
homologous protein alignments. Protease 3C catalyzed sites
are represented by the heavy lines. Proteinase 2A and 1AB
cleavage sites are the lighter vertical lines. sites for
which the proteinase has not been identified are enclosed in

van Kammen, 1988; Krausslich and Wimmer, 1988). In the

polio polyprotein there are three types of cleavage sites:

nine Gln-Gly sites which are cleaved by the 3C proteinase,

two Tyr-Gly sites cleaved by the 2A proteinase, and an Asn-

Ser site in the capsid which appears to be cleaved by an

RNA-mediated activity (Palmenberg (1987). N-terminal

protein sequencing of the final protein products in

conjunction with some C-terminal sequence determination and

analysis of the polyprotein sequence as derived from the

genome sequence were used by Pallansch et al. (1984) to

determine all of the cleavage sites in the polio

polyprotein. They also established that there is no N-

terminal or C-terminal trimming of these proteins by

aminopeptidases or carboxypeptidases. Polio remains the

only picornavirus with all of its cleavage sites determined

experimentally, though many individual sites from other

polyproteins have also been confirmed by sequence analysis

(Figure 1-3). The remaining sites are inferred from

analysis of predicted polyprotein sequences and the size of

processed proteins and by homology to other picornaviruses.

As Figure 1-3 shows, while the overall scheme of polyprotein

processing is conserved among the picornaviruses, there is

some variation in the actual cleavage sites.

The entire polyprotein is approximately 250 KD in size

but is almost never seen. The initial proteolytic cleavage

which separates the capsid proteins from the rest of the


polyprotein occurs while the nascent polyprotein is still on

the ribosome. In the enteroviruses and rhinoviruses this

cleavage occurs between proteins 1D and 2A (Campbell and

Jackson, 1983; Pallansch et al., 1984; Callahan et al.,

1985; Sommergruber et al., 1989). However, by a careful

kinetic analysis of the appearance of the precursors and

final products of EMCV polyprotein processing in vitro, Jackson

(1986) determined that the primary cleavage event occurred

between proteins 2A and 2B in cardioviruses.

Immunoprecipitation with antibodies to 2A from FMDV also

shows that the 2A/2B junction is the primary cleavage site

in the aphthoviruses (Vakharia et al., 1987). Thus in the

aphthoviruses and cardioviruses the first cleavage produces

a protein with the structure L-1ABCD-2A, while in

enteroviruses and rhinoviruses the analogous protein is

1ABCD. As mentioned previously, sequence analysis reveals

that the 2A proteins from aphthoviruses and cardioviruses

are not homologous to the 2A proteins from enteroviruses and


Toyoda et al. (1986) using an E.coli expression system

demonstrated that the cleavage of the polio 1D/2A junction

was dependent upon expression of 2A. Truncations,

deletions, or insertions within the 2A coding region all

abolished cleavage at the 1D/2A junction. Analogous

experiments, conducted with human rhinovirus 2 sequences,


also show that 2A is required for rhinovirus processing at

the 1D/2A site (Sommergruber et al., 1989).

Nicklin et al. (1987) have produced a stable protein

spanning the 1D/2A junction by making either a deletion or

an insertion into the 2A region. This protein could be

cleaved at the 1D/2A site by exogenous 2A proteinase,

demonstrating that 2A can carry out the cleavage in an

intermolecular (trans) reaction. Currently most researchers

feel that this cleavage takes place in vivo initially via an

intramolecular (cis) interaction due to its rapid nature,

occurring while the polyprotein is still on the ribosome,

though there is no good evidence for this as yet.

An additional role in the termination of host cell

protein synthesis has been proposed for the 2A proteinase.

Etchison et al. (1982, 1984) have demonstrated that viral

infection by polio induces the cleavage of the mRNA cap

binding protein p220 in host cells. Cleavage of this

protein appears to be associated with the shut down of host

cell protein synthesis. Mutation of the 2A region

eliminates the cleavage of p220, and the reduction of host

protein synthesis in these cells is not as efficient as in

those cells infected with wild type virus (Bernstein et al.,

1985). The mechanism of this shutdown has not been

completely characterized at this time. Krausslich et al.

(1987) have determined that, while cleavage of p220 is


induced by 2A, 2A does not catalyze the cleavage of p220


The 2A proteinase cleaves a Tyr-Gly bond in

enteroviruses and rhinoviruses and is highly specific,

cleaving only one bond in most viruses. Type 1 poliovirus

has an additional 2A cleavage site in the 3D region which

generates the proteins 3C' and 3D' and may serve to regulate

the amounts of active polymerase in the cell. However, only

two of ten Tyr-Gly bonds are cleaved. All 2A cleavage sites

contain the sequence Thr-Tyr-Gly (Nicklin, et al., 1986).

Mutation of the threonine at the 3C'/3D' site to an alanine

abolished cleavage at this site yet produced viable virus

(Lee and Wimmer, 1988). Substitution of phenylalanine for

tyrosine at the P1 position did not abolish cleavage. Other

viruses show even more variation in this P2-P1-P1' sequence.

Coxsackieviruses B1 and B3 cleave at Thr-Thr-Gly and Asn-

Thr-Gly sequences respectively (lizuka et al., 1987;

Lindberg et al., 1987). Rhinovirus 2A has been shown to

cleave Thr-Ala-Gly, Ser-Tyr-Gly, and Asn-Val-Gly according

to the strain (Duechler et al., 1987; Kowalski et al.,


Konig and Rosenwirth (1988) have purified poliovirus 2A

and have used its esterase activity towards a small peptide

substrate, Gly-Leu-Gly-Gln-Met-OCH,, to characterize its

activity. They have shown that this activity is inhibited

by iodoacetamide and N-ethylmaleimide, classical sulfhydryl

modifying reagents, as well as Zn 2 and have classified this

enzyme as a cysteine proteinase. This is supported by the

sequence alignment studies of Bazan and Fletterick (1988,

1989), who designate Cys-109 as the catalytic residue acting

in concert with His-20 and Asp-38 to form the catalytic

triad of a cysteine proteinase. At this time, these

predictions have not been confirmed experimentally.

Though 2A does not function as a proteinase in the

cardioviruses and aphthoviruses, Strebel and Beck (1986)

have demonstrated that the L protein of FMDV has proteolytic

activity. Using an in vitro transcription and translation

system, they noted that L apparently mediated the cleavage

at the L-01 junction, and this cleavage was eliminated by

mutations in the L protein but not by mutations in P1.

However, L does not catalyze the primary cleavage event

between 2A and 2B (Vakharia et al., 1987; Clarke and Sangar,

1988). The catalytic entity for this event remains unknown

at this time.

The majority of cleavages in picornaviral polyproteins

are catalyzed by the 3C proteinase. The first clues that 3C

might be involved in polyprotein processing came from Pelham

(1978). Processing of EMCV polyprotein beyond the initial

cleavage event required nearly complete synthesis of the

polyprotein (Pelham 1978; Shih et al., 1979). In addition,

no processing occurs in arrested translation systems until

an exogenous unarrested translation mixture was added.

Cleavage of the L-P1-2A precursor was used to follow the

purification of a proteolytic activity from EMCV-infected

cell cultures. This 22KD proteinase copurified with the

viral 3C protein (Palmenberg et al., 1979; Gorbalenya et

al., 1979; Svitkin et al., 1979).

Identification of 3C as a proteinase in polio did not

occur until 1982. Hanecak et al. (1982) noted that

antibodies raised against polio 3C inhibited processing of

the polyprotein, while preimmune serum and 2C antibodies did

not. In addition, only Gln-Gly cleavages were affected and

not the initial Tyr-Gly cleavage. Further evidence that 3C

is a proteinase has come from bacterial expression systems.

A polyprotein containing 3ABC and part of the 3D sequence

could be processed to yield the correct proteins (Hanecak et

al., 1984). Insertions into the 3C region abolished this

processing, again implicating 3C as the proteinase. In

addition, pulse chase experiments in this system indicated

that the 3B/3C junction was cleaved more efficiently than

3C/3D. Similar experiments have been used to demonstrate

that 3C is a processing enzyme in FMDV and in human

rhinovirus (Grubman and Baxt, 1982; Klump et al., 1984;

Decock and Billiau, 1986).

Engineered mRNA has been used to obtain stable

polyprotein precursors which can be used as substrates for


the 3C proteinase (Nicklin et al., 1987; Ypma-Wong and

Semler, 1987a; Clarke and Sangar, 1988). These substrates

are cleaved properly by exogenous 3C, demonstrating that 3C

can act in trans to process the polyprotein, indicating an

intermolecular reaction. Palmenberg and Rueckert (1982)

studied the processing of the EMCV polyprotein following in

vitro translation. They noted that processing to yield 3C was

only partially sensitive to dilution. They demonstrated

that, while intermolecular (dilution sensitive) cleavages do

occur, intramolecular (dilution insensitive) processing must

also occur.

Though 3C is an active proteinase alone, it may not be

the only significant form of the proteinase invivo. Using

engineered protein substrates, Ypma-Wong and Semler (1987a)

have demonstrated that for polio the protein 3CD is able to

process the capsid proteins much more efficiently than 3C

alone. While polio 3C and 3CD were equally efficient in

their activities towards the P2 and P3 proteins, 3C was only

able to cleave the 1C/1D junction. This has led them to

propose that 3CD rather than 3C is the molecule responsible

for processing the capsid proteins (see Figure 1-4).

Constructs containing up to 75% of the 3D sequence were

still incapable of processing P1 (Ypma-Wong et al., 1988;

Jore et al., 1988). Recently Takahara et al. (1989) were

able to observe a limited amount of cleavage at the 1B/1C

Figure 1-4. Proposed model illustrating the cleavage
specificities of the poliovirus 3C and 3CD proteinases. The
3C proteinase is depicted as a niched circle while the
auxiliary 3D sequence is depicted as a thick line. In this
model, the 3D sequence is required to interact with P1 in
order to position the P1 Q-G pairs (indicated by *) so that
they can be recognized by the 3C moiety of the 3CD protein.
The other Q-G cleavage sites (indicated by *) can be cleaved
by 3C or 3CD.


junction alone. However, Parks et al. (1989) have

recently demonstrated that 3C is sufficient for cleavage of

EMCV capsid precursors. They noted that 3C as well as its

precursors (P3, 3ABC, and 3CD) were all able to cleave

capsid precursors and, while they may not be essential,these

precursors could play a significant role in the invivo

processing of the polyprotein. This is further supported by

the kinetic studies of Jackson (1986) who noted that initial

processing of the capsid region began before free 3C could

be detected.

Inhibitor studies place 3C into the family of cysteine

proteinases. Pelham (1978) showed that polyprotein

processing could be inhibited by thiol modifying reagents

such as iodoacetamide and N-ethylmaleimide. Chicken egg

white cystatin, a cysteine proteinase inhibitor, has also

been shown to interfere with the intracellular processing of

poliovirus proteins (Korant et al., 1985). Sequence

analysis led Argos et al. (1984) to propose Cys-147 and His-

161 as catalytic residues in polio 3C. Site-directed

mutagenesis of polio 3C in an E. coli expression system has

been used to demonstrate the importance of these residues

(Ivanoff et al., 1986). Mutation of Cys-147 to Ser or His-

161 to Gly eliminates 3C activity, while mutation of the

nonconserved Cys-153 to Ser had no apparent affect upon 3C



Sequence analysis shows that 3C is related to both

cellular cysteine proteinases and serine proteinases.

Gorbalenya et al. (1986) showed homology between 3C and

Cathepsin H as well as trypsin. More recently Bazan and

Fletterick (1988, 1989) have suggested that viral cysteine

proteinases may be structurally more related to serine

proteinases such as trypsin than cysteine proteinases. They

also predict a catalytic triad composed of His-40, Asp 85,

and Cys-147. This differs from earlier predictions that

His-161 is catalytically important (Argos et al., 1984;

Werner et al.,1986).

As noted before, 3C is a highly specific enzyme. Polio

3C shows the greatest stringency, cleaving only Gln-Gly

bonds. Gln-Gly is not sufficient for cleavage, as only 9 of

13 potential cleavage sites are recognized. Examination of

3C cleavage sites in other polyproteins reveals that the Pl-

P1' specificity is relaxed in other viruses (see Figure 1-

3). While a majority of these sites are Gln-Gly, Glu and

Val occur naturally in P1, and Ser, Ala, Thr, Leu, Ile, Met,

and Val have all been reported to occupy P1' (Krausslich and

Wimmer, 1988). Mutagenesis has been used to begin to

explore the specificity of 3C. Parks and Palmenberg (1987)

reported that Ala and Ser were acceptable substitutions for

the P1' Gly in the EMCV 1C/1D cleavage site. Later studies

showed that for EMCV 1C/1D cleavage site Gln and Glu are

both acceptable in P1, while Lys, Arg, Leu, and Pro in this


position abolished cleavage (Parks et al., 1989). The P1'

site could accommodate Gly, Ser, Cys, and Ala but not Thr,

Val, Ile, Glu, or Tyr.

There appears to be some specificity beyond the PI-PI'

sequence. In fact, Pallai et al. (1989) reported that

purified 3C was unable to hydrolyze the dipeptide, Gln-Gly.

Most 3C cleavage sites in EMCV are flanked in either the P2

or P2' position by a Pro residue (Palmenberg et al., 1984),

but this requirement is not absolute. Displacement of Pro

from the 1C/1D cleavage site reduced but did not eliminate

cleavage at this junction in vitro (Parks and Palmenberg,


The P4 position also appears to be conserved. Alanine

or some other neutral aliphatic amino acid is almost always

found in this position (Nicklin et al., 1986). Substitution

of Lys into the P4 position of the polio 3B/3C site

eliminated cleavage as did insertion of Thr into the P4

position at the 3C/3D junction (Semler et al., 1987).

Replacement of Thr with Ala at the P4 position enhanced the

cleavage of a synthetic peptide substrate approximately 500-

fold (Pallai et al., 1989).

An obvious additional determinant for cleavage is

accessibility. Cleaved bonds must be on the surface of the

protein, where they can be recognized by the proteinase.

Crystal structure data for the capsid proteins show that


uncleaved Gln-Gly sequences lie within structurally rigid

a-barrel domains (Arnold et al., 1987). Proper folding has

also been implicated in cleavage site recognition.

Baltimore (1971) noted that processing did not occur in

polyprotein synthesized at 41"C. This polyprotein was

unable to undergo processing even when shifted to 37C.

Truncated P1 was also unable to serve as a substrate though

it contained authentic 3C cleavage sites (Nicklin et al.,

1987; Ypma-Wong and Semler, 1987a, 1987b). Finally, the

specificity of 3C is further demonstrated by its inability

to cleave cellular proteins (Korant et al., 1980a).

The final cleavage in picornaviruses occurs within the

assembled virion. This cleavage occurs between proteins 1A

and lB. Neither 2A nor 3C seem to be involved in this

cleavage which takes place at an Asn-Ser bond in polio. It

also appears that this site is located on the interior side

of the viral capsid where it would be inaccessible to

external proteinases (Rossmann et al., 1985). This has led

to the proposal of an autocatalytic mechanism for this

cleavage involving the P-hydroxyl of Ser-10 of 1B (Rossmann

et al., 1985; Arnold et al., 1987). Since there does not

appear to be a suitable proton-extracting residue in the

protein, the viral RNA has been suggested to act as the base

and activate the serine (Arnold et al., 1987). While this

mechanism has not been proven, Arnold et al. (1987) have

demonstrated that some small diamino compounds such as


hydrazine and ethylene diamine can induce the cleavage of a

1AB protein synthesized in viro.

Thus all proteolytic events in picornaviral protein

processing are mediated by virally encoded activities.

While these activities have not yet been rigorously

characterized, a strong foundation of work exists to

continue with the study of these activities.




Currently a number of in viro assay systems exist for

both 2A and 3C (Parks et al., 1986; Nicklin et al., 1987;

Ypma-Wong and Semler, 1987a; Vakharia et al., 1987). These

systems all use engineered protein precursors as substrates

with cleavage being followed by SDS-PAGE. These systems

have the advantage of using substrates which are as close to

the native substrate as possible. In addition, mutations

allow for modification of the cleavage site. However, it is

difficult to obtain quantitative data from these systems,

and alteration of the cleavage sites can sometimes be very

tedious. Synthetic peptide substrates have proved useful in

a number of proteinase systems. The cleavage of these

substrates can be accurately quantitated and it is

relatively easy to modify the cleavage site through

additional synthesis.

The 3C proteinase was chosen for this study because it

catalyzes the majority of the polyprotein cleavages and thus

presents more possible sequences for the generation of

synthetic peptide substrates. In addition, the existence of


bacterial expression systems provides a ready source of

enzyme (Ivanoff et al., 1986; Nicklin et al., 1988; Takahara

et al., 1989).

It is known that a dipeptide of Gln-Gly will not be

cleaved by the 3C proteinase (Pallai et al., 1989) and Gln-

Gly is not sufficient to define a cleavage site as only 8 of

13 Gln-Gly bonds in the polio polyprotein are readily

cleaved (Nicklin et al., 1986). Table 2-1 lists all of the

polio 3C cleavage sites. Examination of these sites reveals

little homology beyond the P1-PI' residues, indicating that

there may be a structural component necessary for

recognition of the substrate by 3C. Nicklin et al. (1986)

have observed a general preference for alanine at the P4

position which has been verified with protein substrates

(Parks and Palmenberg, 1987; Semler et al., 1987). Beyond

this there are few guidelines for the synthesis of potential

peptide substrates.

Experimental Procedures

Materials. The pEXC expression system (Ivanoff et al.,

1986) was a gift from Dr. B. Korant (Dupont, Wilmington,

DE). Extracts from polio infected HeLa cells were obtained

from Dr. J.B. Flanegan (University of Florida, Gainesville,

FL). Amino acid derivatives were purchased from Applied

Biosystems (Foster City, CA) or Bachem Inc. (Torrance, CA).

All other reagents and materials were of the finest

commercial grade available.

Table 2-1. Sequences Surrounding Polio (Mahoney) 3C
Cleavage Sites.

Cleavage P Site
Site 10 9 8 7 6 5 4 3 2 1-1'2'3'4'516'7'8'9'10'










Synthesis of Peptide Substrates. All amino acids used

were of the L-configuration with the exception of glycine

and were the a-N-t-butyloxycarbonyl derivatives.

Conventional solid-phase peptide synthesis proceeded from

the C-terminal amino acid which was attached via a

phenylacetamidomethyl linker (Mitchell et al., 1978) to a

polystyrene resin. Automated, step-wise synthesis of the

peptide was performed on an Applied Biosystems 430A peptide

synthesizer according to standardized protocols (Kent et

al., 1984). All amino acids were added as symmetric

anhydrides with the exception of Arg, Asn, and Gin which

were added as the HOBT esters. Recouplings were routinely

performed for all amino acids in order to maximize the yield

of desired peptide. Peptides were simultaneously

deprotected and removed from the resin by treatment with

anhydrous HF. The peptidyl resin was mixed with 10 ml of

anhydrous HF and 1.5 ml of anisole per gram of peptidyl

resin at 0C for 1 hour. Following removal of the HF from

the resin at reduced pressure, the resin was washed with a

1:1 mixture of ethyl ether and ethyl acetate to remove

anisole and other organic byproducts. The peptide was then

extracted from the resin with a 10% acetic acid solution in

water. This extract was then lyophilized and the resulting

peptides were stored desiccated at 4'C.


Purification of Synthetic Peptides. All peptides were

purified by reversed phase HPLC using either MBondaPak C18

or NovaPak C18 columns from Waters. Initially peptides were

separated using a gradient of 10% to 60% acetonitrile in

0.1% TFA over 50 minutes. Conditions were subsequently

optimized for the purification of each peptide. Purified

peptides were lyophilized and stored desiccated at 4*C.

Amino Acid Analysis. Aliquots of approximately 50

nmoles of peptide were placed into pyrex tubes for

hydrolysis and dried prior to the addition of 1 ml ultrapure

6 N HC1 and 10 pl liquified phenol. The tube was evacuated,

sealed, and heated to 110C for 24 hours. The hydrolysate

was dried, the sample dissolved in 0.2 M citric acid, and

quantitative amino acid analysis was performed on a Beckman

6300 amino acid analyzer. Peptidyl resins were hydrolyzed

in a 1:1 mixture of propionic acid and 12 N HC1 (Scotchler

et al., 1970). Cysteine determinations were made following

the conversion of cysteine to the more stable cysteic acid

derivative by hydrolysis in 6 N HC1, 0.35 M DMSO (Spencer

and Wold, 1969).

Protein Seauence Analysis. Protein sequences were

determined by sequential Edman degradation on an Applied

Biosystems 470A gas-phase protein sequencer with an on-line

120 A PTH-AA analyzer following protocols supplied by the

manufacturer (Strickler et al., 1984).


Partial Purification of Polio 3C from HeLa Cells. All

purification was carried out at 4'C. Clarified cytosolic

extracts from polio infected HeLa cells were loaded onto a

DEAE-Sephadex column equilibrated with 10 mM Tris, pH 8.5;

14 mM 2-mercaptoethanol; 1 mM EDTA and the flow through peak

collected. This material was dialyzed against 4 liters of 10

mM Bis-Tris, pH 6.8; 14 mM 2-mercaptoethanol; 2 mM EDTA.

Following dialysis, the retentate was loaded onto a CM-

Sephadex column equilibrated with the same buffer. Polio 3C

proteinase was eluted with the equilibration buffer

supplemented with 0.35 M NaCl.

Expression of pEXC. Expression of the pEXC plasmid was

carried out as outlined by Ivanoff et al., 1986. E.coli

HB101 containing the pEXC expression vector was used to

inoculate 50 ml of LB media (Maniatis et al. 1982)

supplemented with 0.1 mg/ml of ampicillin and incubated at

37C with shaking at 300 rpm for 12 hours. This culture was

then diluted into 1 liter of M9 minimal media (Maniatis et

al. 1982) to induce expression of the plasmid. The bacteria

were grown in the expression media for 6-8 hours before


Purification of recombinant 3C. Bacteria were

harvested by centrifugation at 5,000 x g for 10 min. The

cells were resuspended in 3 ml of 50 mM Tris, pH 8.5; 150 mM

NaCl; 5 mM DTT; 2 mM EDTA (Buffer H) per g of cells and


centrifuged at 5,000 x g for 10 min. The cell pellet was

resuspended in 4 ml buffer H per g of cells and lysed by

sonication, 5 x 15 seconds. The insoluble pellet was

collected by centrifugation at 10,000 x g for 10 minutes at

4'C. The protein in this pellet was solubilized in 20 ml of

8 M urea; 50 mM Tris, pH 8.5; 5 mM DTT; 2 mM EDTA per g of

cells and centrifuged at 13,000 x g for 30 minutes and the

pellet discarded. All subsequent steps of the purification

were carried out at 4C. The denatured 3C was then refolded

by dialysis against 1 liter of 40 mM Tris, pH 8.5; 5 mM DTT;

2 mM EDTA for 3 hours with buffer changes at 1 hour


The refolded 3C proteinase was then loaded onto a

column of DEAE Sephadex equilibrated with 40 mM Tris, pH

8.5; 5 mM DTT; 2 mM EDTA. The flow through peak was

collected and the protein precipitated by dialysis overnight

against a solution of (NH4)2SO4 to give a final concentration

of 0.55 g/l of (NH4)2SO4. The precipitate was collected by

centrifugation at 10,000 x g for 10 min. and dissolved in 10

ml of 25 mM MES, pH 6.4; 2 mM EDTA; 1 mM DTT (buffer C).

This protein solution was then dialyzed for 1 hr against 1

liter of buffer C and loaded onto a column of CM Sephadex

equilibrated with buffer C. After washing the column with

100 ml of buffer C, the 3C proteinase was eluted with a

gradient of 0 to 0.5 M NaCl in 600 ml of buffer C and could

be stored at 0'C for up to 10 days.

SDS Gel Electrophoresis. Samples of approximately 5 to

25 fg of protein per lane were separated on a 16 cm x 16 cm,

12.5% polyacrylamide gel containing 0.1% SDS according to

the method of Laemmli (1970) at 15 mA constant current for

12 hours.

Electroblotting onto PVDF Membranes. The SDS-PAGE gel

containing separated proteins for sequence analysis was

washed in transfer buffer, 10 mM MES, pH 6.5; 10% methanol,

for 5 minutes to remove excess Tris and glycine. The

transfer buffer was replaced and the protein bands from the

gel were electroblotted onto a PVDF membrane at 0.5 A for 30

minutes (Matsudaira, 1987). The membrane was stained with

0.1% Coomassie Blue R-250 and destined in 50% methanol in

order to localize the protein bands.

HPLC Assay for Polio 3C Activity. Polio 3C and

substrate peptide were incubated in 50 mM phosphate, pH 7.5,

containing 100 mM NaCl, 2 mM EDTA, and 1 mM DTT at 37*C.

Several aliquots were removed from the reaction mix at timed

intervals and the reaction quenched by the addition of 2

volumes of 6 M GdnHC1. These quenched reaction mixtures

were then separated by reversed phase HPLC and the amounts

of substrate and cleavage products were quantified by

integration of peak areas determined from absorbance at 254

nm. Initial cleavage rates were then determined from those

aliquots representing less than 10% total substrate




Synthesis of Peptide Substrates. Initially a series of

9 putative substrates were synthesized based upon the 8 well

recognized polio 3C cleavage sites. The sequences of these

peptides are listed in Table 2-2. The peptide, Polio 1.2,

was synthesized as an analog of Polio 1 to examine the

feasibility of replacing methionine residues with

norleucine. Peptides were generally at least 90% pure

following HF cleavage, though sequences high in glutamate

tended to generate a cleavage byproduct. These byproducts

could be identified by their anomalous UV absorbance spectra

which indicated that benzyl group migration may have

occurred in < 5% of the peptides. Each peptide was purified

to homogeneity by HPLC and the amino acid composition of

each was confirmed by amino acid analysis.

Cleavage of Polio 1. Polio 1 was chosen as the initial

sequence to examine since the cleavage site it represents,

2C/3A, is very readily cleaved in the polyprotein and the

proline in the P2' position could provide some

conformational restriction to the substrate. Figure 2-1

shows that this peptide could be cleaved by partially

purified polio 3C proteinase obtained from infected HeLa

cells but not by a similar preparation from mock infected

HeLa cells. While this cleavage was slow, requiring several

hours for significant hydrolysis to occur, it was

reproducible. The slow cleavage rates observed in these


a a p 0 0 b
< < U l E 0 P0 4



w Ek >

0 0 E > E0 0 z
I. P P. H (A H o P
j I I I In I H
a- I I I I I I I I Io .

to 4 H 44 4 4 4H 4 4)

I I I I I I I r- I
t C C C 4 C. C

1 1 < 3 0 0 10 0 P

4 CD H 0 C CD C m
o H Ho H r- D H 4 H HH CI
5.. fl 4 .D r

0 II
u in $4 $4 >, M >1 k

0 4) C 3 0 C 3 0 E-
*- Io C H H 5 H CD 5 C
H w m s m P. Ev (9 en m M

o N0 I ( m0 0 II
S > > N >O O % > CDB
4 4 U) H O H 'I H -H E-H CI
o l i q I

to o 0 0 0 H
E- C 54 5 M N al N 4N


N N 5

CD -I 0 0 0 0 0 0 0 0 0 I

CD CDC 0 0 0 0 0 0 0 00 Hr
P.2 P. a P. P. P. P. P. P. P. 5

Time (min)

Figure 2-1. Reversed phase chromatograms showing the
cleavage of 150 AM Polio 1 after an 18 hour incubation with
partially purified 3C proteinase from polio infected HeLa
cells. A similar preparation from mock infected HeLa cells
failed to cleave the peptide. Separation was performed on a
Waters ABondapak C18 column using a linear gradient of 10 to
50% acetonitrile in 0.1% TFA over 30 min at 1.5 ml/min.


extracts were thought to reflect the limited amounts of

proteinase available from this source. This proteolytic

activity was inhibited in the presence of Zn*2 ions (data

not shown). While only a single product peak is apparent at

280 nm, two product peaks can be observed at 220 nm (not

shown). Amino acid analysis of the product peaks confirmed

that cleavage had occurred at the predicted Gln-Gly bond.

Though not conclusive, these data are consistent with

cleavage mediated by polio 3C; however, the limited amounts

of 3C proteinase recovered were not sufficient for the

proposed studies.

Expression of Polio 3C in E. oli. The pEXC polio 3C

expression system (Ivanoff et al., 1986) was obtained to

serve as a source for the 3C proteinase. While this had

been used to generate in vivo 3C proteinase activity, it had

not been used as a source of 3C proteinase for in vitro

studies. Thus it was necessary to demonstrate that polio 3C

proteinase activity could be recovered in significant

amounts for in vitro studies. The pEXC vector (Figure 2-2)

utilizes a trp promoter to regulate expression of a fusion

protein which contains the entire 3C protein in addition to

an N-terminal methionine and 25 residues of the 3D sequence

at the C-terminus. Proteolytic activity could be detected

in vivo by monitoring the appearance of the processed mature

3C protein on SDS-PAGE gels. Expression is induced by


___________ _________a~i-Y (ITA

> (M)

Figure 2-2. Schematic drawing of the pEXC clone and the
recombinant proteinase gene. The poliovirus-specific
proteins are indicated with arrows (<->). P, precursor
polypeptide; I, internal initiation product; M, mature

placing the cells into minimal media and starving the

bacteria for tryptophan. When induced, polio 3C can

represent up to 10% of the total cellular protein (Ivanoff

et al., 1986).

Extracts from E. coli HB101 with no plasmid (not shown),

the pEXC vector, and the pEXC vector in which the active

site Cys-141 has been changed to Ser-141 and activity

abolished were incubated with both Polio 1 and Polio 1.2

(data not shown) and examined for their ability to cleave

these peptides. Only extracts from bacteria containing the

pEXC vector were able to cleave these peptides (Figure 2-3).

The products of the cleavage of the peptide, Polio 1, had

identical retention times and amino acid compositions to

those generated by cleavage catalyzed by extracts from

polio infected HeLa cells. Thus pEXC is a viable source for

the 3C proteinase. In addition, there was no significant

difference in the rates of cleavage of Polio 1 and Polio

1.2, indicating that norleucine was an acceptable

substitution for methionine.

Purification of pEXC Proteinase. Initial experiments

revealed that the majority of expressed 3C proteinase could

be found in an insoluble form in the bacteria. This

material could be solubilized in 8 M urea or 3 M GdnHCl and

activity recovered after refolding by dialysis to remove

denaturant, so this was chosen as the starting material for

, 0.03

0.02 P 1 1 eo
pEXC induced

0.10 o mutant induced

0 C- 5 mutant uninduced
0 10 20 30
TIME (min)

Figure 2-3. Cleavage of peptide Polio 1 by the pEXC
proteinase. Cytosolic extracts from induced cells
containing the pEXC expression vector can cleave Polio 1.
Similar extracts from uninduced bacteria or bacteria with a
mutant 3C gene which produces inactive 3C fail to cleave the
peptide. Complete cleavage of 150 gM Polio 1 could be
observed in less than 6 hours. Separation was performed on
a Waters iBondapak C18 column using a linear gradient of 10
to 50% acetonitrile in 0.1% TFA over 30 min at 1.5 ml/min.

further purification. Precipitation of protein during

refolding could be eliminated by decreasing the protein

concentration by dilution with additional denaturant. This

necessitated the use of urea as the denaturant in order to

insure the low ionic strength essential for optimal

purification on the anion exchange column. The purification

of 3C is documented in Table 2-3 and Figure 2-4. As Figure

2-4 shows this purification results in 2 primary bands of 24

KD and 29 KD on a silver stained SDS-PAGE gel. These bands

correspond in size to the precursor and mature forms of the

3C proteinase, respectively, as expressed by pEXC.

Protein Sequence Analysis. The two proteins remaining

in the proteinase positive peak were separated by SDS-PAGE

and electroblotted onto a PVDF membrane prior to sequence

analysis. The protein bands were then cut out of the

membrane for further analysis. One set of protein bands was

hydrolyzed to quantify the amount of protein loaded per

lane. Approximately 200 pmol of blotted protein was then

loaded onto the microsequencer for protein sequence

determination. The sequence data for the 24 KD band is

shown in Figure 2-5.

Protein sequence analysis positively identified both

protein bands from the SDS gel as products from the pEXC

expression system (Figure 2-6). The band corresponding to

processed 3C showed the inserted N-terminal methionine and

then matched the sequence of polio 3C for 28 of 32 residues


4M o r co L

0. .
r. dP v 0 o -


>0 coI r- to

B *44-H

M OD r- r-

0n > H (n in H o
H .

0H O, Q rl
*l U1 0 cC l N H

O m 0 MI ,--l


o -o

0- N L 4
t o to o
0 o0 04 H


\ N 0 H o 0

aH 0


0 4. .1-4 0 r-i




4 20

, 14


Figure 2-4. SDS polyacrylamide gel separation of fractions
from a typical preparation of polio 3C from the pEXC
expression system. Protein bands were visualized by silver
staining. The lanes are labelled as follows: MW (molecular
weight markers), RF refoldedd protein from the urea
extract), DEAE (DEAE-Sephadex flow through fraction), and CM
(CM-Sephadex column fraction). CM fractions were loaded at
two concentrations to check the purity of the enzyme. The
effect of mercury on enzyme stability was also examined
(+HG, -HG). The bands at 29 KD and 24 KD are the expected
sizes for the precursor and mature forms of 3C,
respectively, as produced by pEXC.

- .0


o- -M G A
Y*e *A

o I

S40 NG G
T F e
20 R -* .

0 5 10 15 20 25 30 35

Figure 2-5. N-terminal sequence determination for the 24 KD
protein. A plot of PTH amino acid recovered versus cycle
number as well as the identity of each residue is shown.
Analysis of alanine recoveries indicated an initial yield of
85 pmoles and a repetitive yield of 91.9%.


5 10 15 20
25 30 35

Figure 2-6. N-terminal Protein Sequences of pEXC Products.
Protein sequence data positively identifies the two proteins
isolated as the precursor and mature form of the polio 3C
proteinase as expressed in the pEXC system.


(4 residues were not assigned). The precursor, which has a

25 amino acid C-terminal extension, had the identical N-

terminal sequence to the mature form of the proteinase, as


The ability of the precursor to process itself shows

that it does possess 3C activity. Larger forms of 3C have

been shown to possess significant activity both invivo and in

vitro (Ypma-Wong and Semler, 1987a; Parks et al., 1989).

Though polio 3CD is believed to have enhanced activity

towards the cleavage sites in the capsid region, constructs

containing up to 75% of the 3D sequence had activities

identical to 3C alone (Ypma-Wong and Semler, 1987). For

these reasons this mixture of mature and precursor 3C was

deemed suitable for all experiments within the scope of this

work. Most recently a preparation of 3C which contained

almost exclusively the mature form of the proteinase was

obtained from bacteria grown in minimal media supplemented

with 2% CAS-amino acids (Figure 2-7). All Vmax data were

obtained with this enzyme preparation.

HPLC Assay for Peptide Cleavage. Preliminary studies

showed that the peptide, Polio 1.2, could be cleaved by a

proteinase 3C activity obtained from polio infected HeLa

cells. This peptide could also serve as a substrate for a

mixture of 3C and a slightly larger precursor derived from a

bacterial expression system. In addition, this expression


66 I

45 40 -

36 '*, I ,

29 -

Figure 2-7. SDS polyacrylamide gel separation of 3C
proteinase obtained from the pEXC expression system when
supplemented with 2% CAS amino acids. Protein bands were
visualized by silver staining. The lanes are labelled as
follows: MW (molecular weight markers), UREA (urea
extract), RF refoldedd proteins from the urea extract), DEAE
(DEAE-Sephadex fraction), CM (CM-Sephadex fraction). The 24
KD band corresponds to the mature form of the 3C proteinase.


system was able to supply proteinase in sufficient amounts

that cleavage was complete within 1 hour. These data

indicated that the in vitro HPLC experiments were feasible.

Cleavage of peptide substrates could be quantified by

HPLC through the appearance of cleavage products, as well as

the depletion of substrate. Integration of the appropriate

peaks allowed the percentage of peptide cleaved to be

determined and then plotted versus reaction time (Figure 2-

8). Initial rates were then determined by linear regression

analysis of those peaks representing less than 10% total

peptide cleavage. These rates could be converted to

appropriate units after the determination of the

concentration of stock peptide solutions by amino acid

analysis. Optimally, at least 4 points within the initial

10% of the reaction were used for rate determinations;

however, this was not always feasible.

Initial experiments with Polio 1.2 showed that this

substrate was recognized by 3C with an apparent Km of 1 mM

(Figure 2-9). Because of the presence of at least 2

(possibly 3, precursor, mature enzyme, and internal

initiation product) active polio 3C proteinase species, Vmax

and kcat were not determined in these experiments. This

apparent Km value of 1 mM was used as the basis for the

design of screening assays for the remaining peptide

substrates with purified polio 3C proteinase.



0 30 60 90 120

TIME (min)

Figure 2-8. Cleavage of the substrate peptide, Polio 1.2,
as measured by HPLC. Linear regression analysis of those
points representing less than 10% total cleavage was used to
approximate the initial rate of the reaction. A rate of 170
AM/min was determined for the cleavage of this 930 AM Polio
1.2 solution.

-2 -1 0 1 2 3 4 5
1/[S] (1/M x 10E-3)

Figure 2-9. A Lineweaver-Burke plot of the reciprocal of
the observed rate of cleavage of Polio 1.2 by 3C versus the
reciprocal of the substrate concentration reveals that polio
3C binds to Polio 1.2 with an apparent Km of 1 mM.

Screening of Peptides Polio 1-8. Each putative

substrate was assayed at a concentration of 1 mM for its

ability to be cleaved by 3C at an enzyme concentration of 2

Ag/ml. Aliquots were removed from each assay at timed

intervals and analyzed by HPLC for any cleavage which may

have occurred. These conditions were generally sufficient

to cleave 50% of Polio 1.2 in 2 hours. Cleavage was

quantitated by integration of substrate and product peak

areas and plotted versus time. The results of one such

experiment are shown in Figure 2-10. Relative rates of

cleavage were determined by comparing the times necessary to

achieve 10% cleavage of the substrate peptide. Appendix I

shows that this time is proportional to (Km + 1)/Vmax. The

polio 3C proteinase preparations showed a marked tendency to

degrade with time, probably due to self-cleavage since the

enzyme does contain an internal cleavage site. These

changes in active enzyme concentration required that any

comparative studies had to be carried out within as short a

span of time as possible, so that all assays were conducted

with the same concentration of active enzyme. For this

reason, relative rates, rather than absolute amounts of

cleavage, were used when comparing data from different

experiments. Table 2-4 lists the relative rates determined

for each of the 8 peptides. Cleavage of individual peptides

was also confirmed by amino acid analysis of the products.


The peptides could be grouped into three classes

according to the rates of cleavage. Polio 1.2 and Polio 7

were the best substrates and appeared to be cleaved with

equal efficiency. The second group consisted of Polio 4,

Polio 6, and Polio 8 and was cleaved at approximately 25% of

the rate of cleavage of Polio 1.2. Finally, Polio 2, Polio

3, and Polio 5 showed little or no cleavage by polio 3C


Following these initial experiments, coincubation

studies similar to those of Pallai et al. (1989) were

conducted. In these experiments, equimolar amounts of Polio

1.2 and a second peptide were assayed in the same reaction

mixture. Those peptides which are cleaved by the 3C

proteinase will act as alternative substrates to Polio 1.2.

In this case both peptides will bind to and be cleaved by

the 3C proteinase. It is possible to calculate the relative

rates of cleavage from the fraction of each substrate

cleaved at a given time using the following equation:

(Vmax/Km),/(Vmax/Km)2 = log(l-F,)/log(l-F2)

where F is the fraction of substrate converted to product at

a given time (O'Leary and Baughn, 1972; Pallai et al.,

1989). It is also possible that those peptides which were

not cleaved in the previous experiments were binding to the

enzyme but rate of conversion of the enzyme-substrate

complex was very slow or zero. This would be detected in

the coincubation experiments as a decrease in the rate of


50 0

o 40 g

-I 30



0 1 2 3 4

TIME (hours)

Figure 2-10. Comparison of the cleavage of 1 mM solutions
of peptides, Polio 1.2 (0), Polio 2 (<>), Polio 3 (E),
Polio 4 (V), Polio 5 (D), Polio 6 (a ), Polio 7 (0), and
Polio 8 (Y), by 2 Ag/ml polio 3C proteinase. Amounts of
cleavage were quantified by HPLC following quenching of the
reaction by 6M GdnHC1.

Table 2-4. Cleavage of Polio Peptides by 3C Proteinase.

Peptide Cleavage P site Rel.
Site 9 8 7 6 5 4 3 2 1-1'2'3'4'5'6'7'8' Rate'

Polio 1.2 2C/3A R C LnE A L F Q-G P L Q Y K D 1.00

Polio 2 2A/2B Y E E E A LnE Q-G I T N Y I E S 0.04

Polio 3 2B/2C I P Y V I K Q-G E S W L K K 0.00

Polio 4 1B/1C R N Y T L P R L Q-G L P V LnN T Y 0.17

Polio 5 1C/1D Y E Q K A L A Q-G L G Q LnL E Y 0.07

Polio 6 3A/3B K L F A G H Q-G A Y T G L P 0.25

Polio 7 3B/3C Y R T A K V Q-G P G F D Y 1.05

Polio 8 3C/3D R S L F T Q S Q-G E I P W LnR 0.27

Ln = norleucine

1Relative rates of cleavage were determined by comparing the
time to achieve 10% cleavage of the peptide as determined by
the experiment shown in Figure 2-10. Relative rates in this
table are the mean of 3 determinations.


cleavage of the peptide Polio 1.2 since binding of the

second peptide would lower the concentration of free enzyme.

In this case the second peptide would be acting as a

competitive inhibitor. The coincubation experiments

confirmed that Polio 1.2 and Polio 7 were cleaved at nearly

identical rates. The other peptides were cleaved at rates

similar to those determined in the initial experiments. In

addition, none of the peptides which were not cleaved were

able to inhibit the cleavage of Polio 1.2, indicating that

they are failing to bind to the enzyme at the site at which

Polio 1.2 binds.

Characterization of Polio 3C. Further experiments to

characterize the proteolytic activity of polio 3C were

undertaken using the substrates Polio 1.2 and Polio 7.

Polio 1.2 was used to determine the pH stability of the

enzyme (Figure 2-11) as well as an activity profile to

determine the optimal pH for assays (Figure 2-12).

Nonlinear regression analysis of this data revealed apparent

pKa values of 6.8 and 8.2 with an optimal pH of 7.5. These

apparent pKa's are consistent with the expected values for

an active site histidine and cysteine, respectively;

however, much more rigorous kinetic studies will be required

in order to interpret the pH behavior of this enzyme.

Several diagnostic proteinase inhibitors were examined

for their ability to interact with polio 3C and these

results are summarized in Table 2-5. Classical cysteine




0 0 0


0.25 -

0.00 -1- I
5 6 7 8 9 10 11


Figure 2-11. A pH profile for the stability of the polio 3C
proteinase. The enzyme was incubated at the indicated pH
for 48 hours at 4 *C. Single point assays for activity were
carried out with 0.5 mM Polio 1.2 at a pH of 7.5. The
amount of product formed in 30 min was used as the measure
of activity. All values are reported relative to the sample
stored at pH 7.5.



+ +

0.00 -r-r
5.10 .20 6. 7,00 ,7.0 7.10 1.2 8.60 .00

Figure 2-12. A pH profile for the cleavage of Polio 1.2 by
3C. The observed rate of cleavage as determined by product
appearance in 30 min is plotted against the pH of the assay
buffer. Substrate concentration was 0.5 mM in all cases.
Apparent pKa's of 6.7 and 8.2 were determined by nonlinear
regression analysis.

Table 2-5. Inhibition of Polio 3C Proteinase.

Inhibitor [I] % Inhibition'

lodoacetic Acid 1 96

N-Ethyl Maleimide 1 100

Hg*2 1 100

Zn+2 1 97

Egg White Cystatin 0.01 27

Human Cystatin 0.02 0

PMSF 0.10 6

o-Phenanthroline 0.10 0

'Inhibition of the initial rate of cleavage of 1 mM Polio
1.2 as determined by HPLC as in Figure 2-8.


0 0.74 mM Polio 7
0.37 mM Polio 7 O

. -4
E4-- Ki = 18 /M


0 I
-20 -10 0 10 20

Egg White Cystatin (AM)

Figure 2-13. A Dixon plot of the inverse rate of cleavage
of Polio 7 by 3C proteinase versus the concentration of egg
white cystatin. These rates were determined at substrate
concentrations of 0.37 mM (0) and 0.74 mM (0). The lines
were determined by a linear least squares fit to the data.
The intersection of these lines indicates a Ki of 18AM.


proteinase inhibitors such as mercury, zinc, iodoacetic

acid, and N-ethylmaleimide all inhibited activity.

Inhibition by mercury could be reversed by excess thiol

reagents such as DTT or 2-mercaptoethanol. Chicken egg

white cystatin, a protein inhibitor of cysteine proteinases,

could inhibit polio 3C activity; however, human cystatin C

failed to inhibit the enzyme. Polio 3C was also impervious

to inhibition by PMSF, a serine proteinase inhibitor, or the

metallo-proteinase inhibitor,o-phenanthroline. Previous

inhibition studies of polyprotein processing were conducted

invivo or in crude cytosolic extracts where the target of the

inhibitor was unidentified. These data confirm the

classification of 3C as a member of the family of cysteine

proteinases as well as the most likely target of inhibition

for those earlier studies.

The inhibition of polio 3C by egg white cystatin was

further examined using the peptide, Polio 7. A Dixon plot

of the inhibition of cleavage of this peptide by the 3C

proteinase (Figure 2-13) indicates that the inhibitor binds

with a Ki of 18 MM. While cystatin generally acts as a

competitive inhibitor of cysteine proteinases (Barrett,

1987), it is not possible to distinguish between competitive

and non-competitive inhibition on the basis of a Dixon plot

(Purich and Fromm, 1972). Recently, Orr et al. (1989)

reported that the related rhinovirus proteinase was not

inhibited by egg white cystatin.


Rhinovirus 3C Peptide Substrates. Recently this

laboratory has also synthesized an analogous series of

peptide substrates for the human rhinovirus 3C proteinase

(Orr, et al., 1989). These substrates were analyzed for

their ability to serve as substrates for the poliovirus 3C

proteinase. Cleavage assays identical to those performed

with the Polio series of substrates were carried out with

these Rhinovirus peptides and the results are given in

Figure 2-14 and Table 2-6. It should be noted that similar

nomenclature has been used for equivalent peptides (i.e.

Polio 1 and Rhino 1 both correspond to the 2C/3A cleavage

site for the respective viruses).

In general the data obtained with the Rhinovirus

peptides agreed with the data from the Polio series of

substrates. Two peptides, Rhino 3 and Rhino 5, were not

cleaved at all. These peptides represent cleavage sites at

anomalous bonds, Gln-Ala and Glu-Gly respectively. Rhino 2,

Rhino 4, and Rhino 8 were each cleaved at approximately one

third the rate of the best substrates. Finally, Rhino 1,

Rhino 6, and Rhino 7 were all cleaved readily by polio 3C at

rates comparable to those observed for Polio 1 or Polio 7.

Inspection of the sequences for both the Polio and

Rhino series of peptides reveals that only those substrates

with a proline in the P2' position [Polio 1 and Polio 7

(Table 2-4) and Rhino 1, Rhino 6, and Rhino 7 (Table 2-6)]

are readily cleaved. This proline may provide necessary


I o
30 60 90 12

TIME (min)

Figure 2-14. Comparison of the cleavage of 1 mM solutions
of peptides, Rhino 1 (0), Rhino 2 ( ), Rhino 3 ([), Rhino
4 ([), Rhino 5 (A), Rhino 6 (A), Rhino 7 (p), and Rhino
8 (y), by 3 Mg/ml polio 3C proteinase. Amounts of cleavage
were quantified by HPLC following quenching of the reaction
by 6M GdnHC1.

Table 2-6. Cleavage of Rhino Peptides by 3C Proteinase.

Peptide Cleavage P-site Rel.
Site 8 7 6 5 4 3 2 1-1'2'3'4'5'6'7'8'9' Rate'

Rhino 1 2C/3A D S L T G L F Q-G P V Y K 1.00

Rhino 2 2A/2B E A I A EE Q-G LS DY IT 0.35

Rhino 3 2B/2C VP Y I E R Q-A N D G W F R K 0.00

Rhino 4 B/C R S K S I V P Q-G L P T T T Y 0.38

Rhino 5 1C/1D S Q T V A L T E-G L G D E LE E Y 0.00

Rhino 6 3A/3B K L F A Q T Q-G P T S GN P 0.80

Rhino 7 3B/3C Y R P V V V Q-G P N EE F 1.00

Rhino 8 3C/3D K Q Y F V E K Q-G V I A R 0.31

iRelative rates of cleavage were determined by comparing the
time to achieve 10% cleavage of the peptide as determined by
the experiment shown in Figure 2-14. Relative rates in this
table are the mean of 3 determinations.

conformational restrictions for the substrate peptides. Of

particular interest is the comparison of peptides Polio 6

and Rhino 6. Both of these peptides correspond to the 3A/3B

cleavage site, yet Rhino 6 is hydrolyzed much faster than

Polio 6. Examination of these sequences revealed several

differences though perhaps the most suggestive one was the

lack of a proline in the P2' position of Polio 6.

Polio 6 Analogs. The experiments with the Rhinovirus

peptides suggested that an important structural component

for the recognition of synthetic peptide substrates by the

polio 3C proteinase might be the presence of proline in the

P2' position. Accordingly a Polio 6 analog was synthesized

in with the P2' alanine was replaced with a proline.

However, this analog, Polio 6-2'P, was not cleaved by 3C in

any detectable amounts. Further comparison of the sequence

of Polio 6 with 30 other picornaviral cleavage sites

revealed that it uniquely contains a glycine at the P3

position. Since P3 is often occupied by a hydrophobic

residue, it was decided to introduce a valine into this

position. Two additional Polio 6 analogs were synthesized.

Polio 6-3V had the single substitution of valine for glycine

at the P3 position, while Polio 6-3V2'P was a double

substitution with valine at P3 and proline in the P2'


Kinetic data for the cleavage of the Polio 6 series of

peptides as well as Polio 1.2, Polio 7, and Rhino 6 are

Figure 2-15. Rate versus substrate concentration for A,
Polio 1.2; B, Polio 7; C, Polio 6; D, Polio 6-3V; E, Polio
6-3V2'P; and F, Rhino 6. Kinetic parameters are given in
Table 2-7.



0 0

0 *

10 r
e -4


0 0 0

pr p1


presented in Figure 2-15 and table 2-7. This data was

obtained with an enzyme preparation containing almost

exclusively mature 3C (Figure 2-7). Vmax values rather than

rate constants are reported, since there is no current

method available to measure the active enzyme concentration.

Total enzyme concentration was determined by amino analysis

of an aliquot of the enzyme sample used for these


The concentration range for these experiments was

limited in some cases by the solubility of the peptides to

values significantly below Km. This is apparent in Figure

2-15 as a lack of curvature in the rate plots and is also

reflected in Table 2-7 as larger uncertainty values. In

these cases only the ratio of Vmax/Km can be accurately

determined. Curvature in the rate plots is essential for

accurate determination of kinetic parameters. Ideally these

experiments should be conducted over a range of substrate

concentrations from 0.2 to 5 times Km. Thus the values

reported here should be regarded as approximations of the

true values.

Both Polio 6 analogs with valine in the P3 position

were readily cleaved by 3C as were Polio 1.2, Polio 7, and

Rhino 6. It also appears that the relatively slow rate of

cleavage of Polio 6 is due to decreased binding to the

enzyme rather than a large change in the rate of hydrolysis

of the Gln-Gly bond. The best substrates all have Km's of


approximately 1 mM while the Km for Polio 6 is 7-fold



These peptides, although based upon cleavage site

sequences, are not direct copies of these sequences. The

sequences were altered in order to produce substrates more

suited to the type of in vitro assay desired while striving to

maintain the essential properties for recognition by the 3C

proteinase. The criteria which were used for the design of

these peptides are discussed below in detail.

First, due to the proposed structural requirement for

substrate recognition, short sequences which might lack the

components necessary to assume essential conformations were

judged to be inappropriate for this study. All peptides

synthesized were 13 to 17 residues in length. This size was

deemed sufficient for short range conformational

interactions without being too long to discourage facile

synthesis of putative substrates.

In order to assure the ready detection of substrate and

product by HPLC analysis, each peptide contains at least one

tyrosine or tryptophan residue. In some cases it was

necessary to insert tyrosine into the sequence; in these

instances the tyrosines were placed as far from the cleavage

site as feasible and they generally replaced naturally

occurring hydrophobic residues such as isoleucine. The


absorbance of these aromatic residues at 280 nm allows them

to be easily distinguished from non-peptide components of

the assay mixture and simplifies the monitoring of the


With the exception of Polio 1, methionine residues were

replaced with the isosteric amino acid, norleucine. This

eliminated any possibility of inhomogeneity due to oxidation

of the sulfur atom of methionine.

Finally, in order to promote solubility, as much charge

as possible was incorporated into the molecule. In some

cases this necessitated the incorporation of a terminal

arginine residue. Again, these substitutions were done as

far as possible from the Gln-Gly cleavage site.

Peptide substrates generated in this study were able to

be used in the development of a in vitro assay for the polio 3C

proteinase. While these peptides were not suitable for a

truly quantitative assay, they do provide an improved assay

for quantifying polio 3C proteinase activity when compared

to the SDS gel systems. In addition these peptide

substrates have provided some information about the

structural requirements for peptide substrates of this

enzyme. In general, linear peptides of the size used in

this study (13 to 17 residues) do not possess a great deal

of ordered structure due to the limited number of intrinsic

constraints (e.g., those provided by hydrogen bonding, salt


bridges, etc.) available in these short sequences. They

instead presumably have a great deal of conformational

freedom. This freedom may actually impair their ability to

be recognized as substrates by the 3C proteinase and may

account for the relatively large Km values observed. The

enzyme may be able to induce the proper conformation for

cleavage in the bound peptide substrates; however, peptides

whose conformations are somewhat constrained so that at

least part of the substrate pool is in a conformation

resembling that of the native substrate may bind more

efficiently to the enzyme than less constrained peptides.

The large Km values observed may then reflect a

conformational equilibrium in which only a small pool of

substrate peptide with the proper conformation is actually

being recognized by the enzyme at a higher affinity than the

apparent Km suggests.

There is a common belief that there is a strong

structural component involved in the recognition of native

substrates by 3C. Evidence for this at present is limited

but will be briefly summarized here. First is the high

specificity of 3C itself. Although present in the

cytoplasm, 3C only cleaves 9 Gln-Gly bonds within the

polyprotein and fails to significantly cleave any other

viral or host sequences (Korant et al., 1980). Proper

folding of the polyprotein also appears to be essential for

recognition of the cleavage sites. Ypma-Wong and Semler


(1987b) demonstrated that deletion of the C-terminal region

of P1 eliminated processing of this precursor protein.

Later it was shown that a 4 amino acid insertion 159

residues from the 3C cleavage site was able to eliminate

processing of P1 by 3C (Ypma-Wong et al., 1988). In

addition the polio 3C proteinase is unusually rigid in its

specificity for Gln-Gly bonds. Parks et al. (1989) have

shown that the EMCV 3C proteinase is able to cleave at least

5 different PI-P1' sequences. All of this seems to indicate

that a structural component in conjunction with the primary

sequence is necessary for substrate recognition by the 3C


The strong preference for the P2' proline exhibited

among the peptide substrates examined in this study supports

the view that conformation of the cleavage site is important

for its recognition by 3C. Proline residues, because of

their cyclic nature, are conformationally much more

restricted than other amino acids and tend to be associated

with kinks or bends in the protein structure. Palmenberg et

al. (1984) noted that proline flanks nearly all of the EMCV

3C cleavage sites at either the P2 or P2' position and

suggested that a a-turn conformation may exist at 3C

cleavage sites. Proline residues are known to flank a

number of 3C cleavage sites in addition to those in EMCV.

Ypma-Wong et al. (1988) have also suggested that 3C cleavage

sites are located in loops between structural domains of the

polyprotein. Initially only those peptides containing

proline in the P2' site were readily cleaved by 3C, strongly

supporting the idea that a bend or loop may be required for

recognition by the enzyme. However, a Gln-Gly-Pro sequence

alone is not sufficient for cleavage as Polio 6-2'P


The other position examined in this study was the P3

position. It would appear that the presence of a glycine at

this position, in the absence of other strong structural

constraints, permits too much conformational freedom. All

other 3C cleavage sites examined tend to contain much more

steric bulk in the side chain at this position, although

there is no other apparent homology at this position beyond

an absence of aromatic residues. Replacement of the glycine

in the P3 position of Polio 6 with a valine was sufficient

to enhance cleavage rates to match the best peptide

substrates yet prepared, even in the absence of proline at

the P2' position. Inclusion of a proline at the P2'

position did not appear to enhance cleavage further though

more study is required to confirm this.

Recently another group has reported the use of

synthetic peptide substrates for in vitro assay of polio 3C

activity (Pallai et al., 1989). They have synthesized

peptides based upon the polio (Sabin strain) polyprotein and

observed cleavage of these peptides with polio (Sabin) 3C

I 0 C 4
o o *
* 0
o o V 0


m H

> 0
a Ep
in >1

0 U

. Fo
, iiC

< rt

mto (
H m q N m w it) %D

proteinase. Their results, summarized in Table 2-8, are in

good agreement with the results from this study. The

majority of cleaved peptides contained proline at the P2'

position, though again this requirement was not absolute.

In addition, the preference for alanine was investigated.

Substitution of an alanine for a threonine residue at the P4

position of one peptide resulted in a greater than 40-fold

increase in the cleavage rate.

Finally, it is interesting to note that the 2C/3A and

3B/3C cleavage sites were always well recognized as

substrates. This is interesting since these two sites are

cleaved very rapidly in the polyprotein (Pallansch et al.,

1984; Jackson, 1986). It is tempting to speculate that

these sites are so readily cleaved in vivo because they have

conformations that are more efficiently recognized and

cleaved by 3C and some of these conformational determinants

are carried over to the peptide substrates. However, this

speculation requires more structural studies of the

polyprotein as well as the peptide substrates before it can

be tested.




Recently synthetic peptides have been used as

substrates for polio 3C in an HPLC assay (Nicklin et al.,

1988; Pallai et al., 1989; Weidner and Dunn, manuscript in

preparation). These peptide substrates represent an advance

in that they allow assays with well defined, readily

synthesized substrates and feedback can generally be

obtained faster than with SDS-PAGE assays. The HPLC assays

allow for quantitation of enzymatic activity which is

difficult with polyprotein substrates and gel


However, there are still disadvantages with the HPLC

assay. The primary disadvantage stems from the fact that

these assays, like the SDS-PAGE assays, are still

discontinuous, providing no information between time points.

Initial rates for kinetic analysis can be difficult to

determine in a discontinuous assay. In addition, rate

determinations by HPLC assay involve several injections

during the time course and tedious calculations without

immediate feedback. Development of a continuous assay would

eliminate these problems as well as allow for more rapid

screening of new substrates and inhibitors.

Continuous assays for proteolytic enzymes often involve

substrates in which the chromophore itself is altered by

cleavage of the substrate, thus producing a change in its

chromogenic properties which can be followed. This approach

was ruled out for the 3C proteinase. The extreme

specificity of this enzyme makes it unlikely that a

chromophore could be incorporated into the Gln-Gly cleavage

site without eliminating hydrolysis of the substrate. The

necessity of removing the reporter moieties from the

immediate vicinity of the cleavage site suggested that

fluorescence resonance energy transfer could be the basis

for a continuous assay for the 3C proteinase.

Fluorescence quenching by nonradiative transfer of

energy from a fluorescent donor to an chromogenic energy

acceptor was first described by Forster (1948). The

efficiency of this transfer can be described by the

following equation:

E = r /(r + Ro) (1)

where r is the distance of separation between the energy

donor and the energy acceptor and Ro is the distance at

which the energy transfer is 50% efficient. Quenched

fluorescent substrates for proteinases based upon such

energy transfers were proposed ten years ago by Yaron et al.

(1979). As equation (1) shows, the efficiency of quenching


is inversely dependent upon the sixth power of the distance

of separation between the donor and acceptor. If a donor

and acceptor are placed upon opposite sides of the bond to

be cleaved, cleavage would allow the donor and acceptor to

diffuse away from each other, increasing the separation.

This would then result in a relief of fluorescence quenching

due to energy transfer. This increase in fluorescence would

be proportional to the amount of cleavage and would thus

provide a convenient assay for proteolysis.

The distance, R,, is dependent upon the nature of the

donor-acceptor pair and is described below:

Ro = 9.79 x 10' (Jn'iKQ)16 (2)

in which J is the integral of spectral overlap for the

fluorescence emission spectra of the donor and the

absorbance spectra of the acceptor, n is the refractive

index of the medium, K describes the orientation of the

donor and acceptor, and Q is the quantum yield of the energy

donor. A good donor-acceptor pair should have a large

overlap of the donor's fluorescence emission spectrum with

the absorbance spectrum of the acceptor. In addition the

orientation of the donor and acceptor moieties should be


One fluorescent peptide substrate has been reported in

the literature (Nicklin et al., 1988). This peptide was

labelled at the amino terminus with a dansyl group to

simplify the detection of the peptide. However, this


substrate was only useful for discontinuous assays and would

not be suitable for a continuous assay.

A series of oligopeptide substrates for the polio 3C

proteinase was recently synthesized in this laboratory

(Weidner and Dunn, manuscript in preparation) and one of the

best substrates, Polio 1.2, was chosen to be the basis for

the development of a continuous fluorescence assay for

enzyme activity. The sequence of Polio 1.2 is given below:


where the Gln-Gly bond represents the site of cleavage by

the 3C proteinase. This peptide is readily cleaved by polio

3C with a Km of 1 mM. In addition this peptide contains a

cysteine and a lysine residue on opposite sides of the

cleaved bond and near the termini. The thiol and amine side

chains of these residues present excellent targets for

differential labelling of the peptide.

Experimental Procedures

Materials. The pEXC expression system (Ivanoff, 1986)

was a gift from Dr. B. Korant (Dupont, Wilmington, DE).

Amino acid derivatives were purchased from Applied

Biosystems (Foster City, CA) or Bachem Inc. (Torrance, CA).

CPM and DABITC were purchased from Molecular Probes (Eugene,

OR). All other reagents and materials were of the finest

commercial grade available.


Acetylation of Peptidvl Resins. The N-terminal tBOC

protecting group was removed by exposure to 100%

trifluoroacetic acid (TFA) for 15 minutes with mixing. The

resin was then washed three times with dimethylformamide

(DMF) and once with diisopropylethylamine (DIEA) to

neutralize the N-terminal amine and washed 3 times with DMF.

The peptidyl resin was then exposed to 50% acetic anhydride

in dichloromethane (DCM) for 15 minutes with constant mixing

and washed 3 times with DMF followed by methanol. The

resulting acetylated peptidyl resin was then dried in a

vacuum desiccator prior to deprotection and HF cleavage.

Synthesis of Polio 1.2-CD. Two equivalents of 7-

diethylamino-3-(4'-maleimidylphenyl)-4-methyl coumarin (CPM)

in DMF were added to the cysteine of acetylated Polio 1.2 (1

mg/ml) in 10 mM phosphate, pH 7.5; 20% DMF and mixed for 1

hour. Excess CPM was removed by HPLC and the resulting

crude CPM derivatized peptide, Polio 1.2C, was

rechromatoghaphed to yield pure Polio 1.2C. This peptide

was then lyophilized and redissolved in 5 mM Na2B4O,, pH 9.5

to yield a solution of approximately 0.2 mg/ml Polio 1.2C.

This was mixed with an equal volume of acetonitrile

containing 5 equivalents of 4-dimethylaminophenylazophenyl-

4'-isothiocyanate (DABITC) for 1 hour. Excess DABITC was

removed by heptane extraction and the crude peptide purified

by HPLC to yield Polio 1.2CD.


Fluorescence Measurements. All fluorescence

measurements were made in solutions of 50 mM sodium

phosphate, 1 mM DTT, 1 mM EDTA, pH 7.5 on an Aminco-Bowman

or a Perkin Elmer MP-4 spectrofluorometer using an

excitation wavelength of 390 nm. All fluorescence assays

for 3C activity were performed at room temperature.


Synthesis of Polio 1.2CD. The initial step in the

synthesis was the acetylation of the N-terminus of the

peptide in order to prevent this amine from reacting with

the derivatizing agents at later steps in the synthesis.

This was easily accomplished while the peptide was still

attached to the resin. At this stage the N-terminal amine

can be selectively deprotected while all other reactive side

chains are protected to eliminate the possibility for any

unwanted reactions. Following HF cleavage and deprotection,

the peptide, Polio 1.2Ac, was reduced and purified to assure

that all cysteines would be in the thiol form. Polio 1.2Ac

was then derivatized at the cysteine thiol with CPM at pH

7.5. This pH was low enough to insure that the e-amino

group of the lysine was fully protonated and unable to

compete as a nucleophile with the cysteine thiol. This

reaction could by monitored by following the coumarin

fluorescence at an excitation wavelength of 390 nm and an

emission wavelength of 480 nm, since the fluorescence yield

of the CPM increases as the reaction proceeds. Following


N o\ 0 H H I O N 0- H H 0
o 0 0 0 0 I O1 0 0p 0 C- 0

Hu N 0 I H H H H1

a t0 00 I o 0 0
ON H O H I N H 0 H 0

. 0 < U i 0 E1.4 0 I|













1.2 Polio 1.2C




0.4 Polio 1.2CD


200 300 400 500 600

Figure 3-1. The UV-VIS absorbance spectra for peptides
Polio 1.2C and Polio 1.2CD. The absorbance maximum at 390
nm for Polio 1.2C is indicative of the addition of a
coumarin moiety to the molecule. Polio 1.2CD shows an
additional absorbance peak centered on 470 nm resulting from
the addition of the DABTC group. Spectra were taken in 50
mM sodium phosphate (pH 7.5), ImM DTT, 1 mM EDTA with 50 lM
Polio 1.2C and 15 MM Polio 1.2CD.

o 0.50-- --0.02 2

L 0

0.25- -0.01

300 400 500 600
Wavelength (nm)

Figure 3-2. Coumarin fluorescence and DABTC absorbance
spectra. Fluorescence excitation (A) and emission (B)
spectra of 2 AM CPM-mercaptoethanol in 50 mM sodium
phosphate (pH 7.5), 1mM DTT, 1 mM EDTA. (C) Absorbance
spectrum of 1.5 AM Polio 1.2CD C-terminal cleavage peptide
containing the DABTC group in the same buffer.

purification of this peptide, Polio 1.2C, its composition

was verified by amino acid analysis (Table 3-1.) and further

characterized by UV-VIS absorbance (Figure 3-1.) and

fluorescence spectroscopy. The DABITC group was then

attached to the E-amino group of the lysine at a higher pH,

9.5, to yield the DABTC derivative. Again the composition

was confirmed by amino acid analysis (Table 3-1.), UV-VIS

absorbance (Figure 3-1.) and fluorescence spectroscopy. The

UV-VIS spectra of this peptide, Polio 1.2CD, shows that the

DABTC moiety absorbs maximally at 470 nm which is very close

to the maximum emission wavelength of 480 nm for coumarin

fluorescence. This is further demonstrated in Figure 3-2

which shows an overlay of the coumarin fluorescence emission

and excitation spectra with the absorbance spectra of DABTC

derivatized peptide. Table 3-1 shows the amino acid

composition for each of the 3 peptides. The disappearance

of cysteine from the analysis following derivatization with

CPM confirms that this was the target of the reaction. The

DABTC derivatization of lysine is reversible under the

conditions of acid hydrolysis used but the depressed lysine

in the analysis of peptide Polio 1.2CD suggests that it has

been modified, since this reversal is not quantitative.

Cleavage of Polio 1.2CD by 3C Proteinase. Figure 3-3

shows the fluorescence emission spectra of a 10 pM solution

of Polio 1.2CD, with excitation at 380 nm, prior to and

following exposure to polio 3C proteinase. The approximate



20 uncleaved


400 500 600 700

Wavelength (nm)

Figure 3-3. Fluorescence emission spectra of Polio 1.2CD.
The fluorescence emission (excitation at 390 nm) of a 10 AM
solution of Polio 1.2CD increases approximately 10-fold
following exposure to 3C proteinase. Spectra were taken in
50 mM sodium phosphate (pH 7.5), 1mM DTT, 1 mM EDTA.

40.1 min.

S24.4 min.
C 10 hr.

2 hr.
1 hr.
0 hr.

0 10 20 30 40
TIME (min)

Figure 3-4. Separation of peptides generated by the action
of polio 3C proteinase on peptide Polio 1.2CD. The bottom
chromatogram was obtained before the addition of enzyme.
Additional chromatograms were obtained after incubation with
enzyme for 1, 2, and 10 hours. The original substrate
peptide eluted at 40.1 min; 2 cleavage products eluted at
24.4 min and 33.7 min. Reversed phase HPLC was carried out
on a Waters Novapak C18 column using a gradient of 10 to 60%
acetonitrile in 0.1% TFA over 50 min.

Figure 3-5. Absorbance spectra of cleavage products
obtained from the action of polio 3C proteinase upon Polio
1.2CD. Peaks were collected from HPLC separation of the
products as described in Figure 3-4, dried, redissolved in
water, and neutralized with ammonia prior to spectroscopy.

Table 3-2. Amino Acid Composition of Polio 1.2CD Cleavage

Amino Acid 24.2 min. 33.7 min. 40.1 min.

































Nle = norleucine

10-fold increase in fluorescence shown in Figure 3-3

occurred following overnight exposure to the enzyme. The

change in fluorescence was continuous as observed by

monitoring the fluorescence emission at 480 nm (data not


This increase in fluorescence occurred concomitant with

cleavage of Polio 1.2CD as demonstrated by HPLC. Figure 3-4

shows the appearance of two new product peaks with time

following addition of the enzyme. These product peaks were

collected from the HPLC for identification. Figure 3-5

shows that 24.4 minute peak had an absorbance spectrum

consistent with DABTC and should thus represent the C-

terminal cleavage peptide. A typical coumarin absorbance

spectrum is associated with the 33.7 minute peak which

correspond to the N-terminal cleavage peptide. The identity

of these peaks was further confirmed by amino acid analysis

as shown by Table 3-2. The amino acid composition further

demonstrates that cleavage occurred at the Gln-Gly bond as

would be expected if cleavage were mediated by the polio 3C


Inner-Filter Effects. A plot of fluorescence yield

versus concentration of Polio 1.2CD (Figure 3-6) fails to

show a linear increase in fluorescence for either the

uncleaved peptide or the cleaved products and is suggestive

of inner-filter phenomena. Chromophores within the solution

can decrease the measured fluorescence by absorbing light at



o0 0 0

Polio 1.2CD (MM)

Figure 3-7. Plot of fluorescence yield versus peptide
concentration for Polio 1.2CD alone (0) and following
cleavage by polio 3C proteinase (0). Inner-filter effects
become significant above 2 MM.

either the excitation or emission wavelengths or both. Most

fluorometers measure fluorescence form the center of the

cuvette and any absorbance at the excitation wavelength will

reduce the amount of light reaching the center of the

cuvette, additionally emitted light can be absorbed while

travelling through the sample to the detector. If the

absorbances of the sample solution at the excitation and

emission wavelengths are known, it is possible to correct

for these inner-filter effects (Lakowicz, 1983; Street,

1985). The following equation (Melhuish, 1961) has been

used to correct for inner filter effects in this study:

Fc = Fo x antilog[(A, + A.)/2)] (3)

where Fc is the corrected fluorescence, Fo is the observed

fluorescence, A, is the absorbance of the sample at the

excitation wavelength, and A, is the absorbance at the

emission wavelength. The measured absorbances are divided

by 2 since the path length to the center of the cuvette is

0.5 cm. Figures 3-7 and 3-8 show the magnitude of inner-

filter effects which can be corrected for by equation (3).

The absorbance of the DABTC at both the excitation and

emission wavelengths (Figure 3-2) indicates that inner

filter effects in solutions of Polio 1.2CD should be

significant at even low concentrations of peptide. All

subsequent fluorescence measurements reported are corrected

by this method for inner-filter effects.

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EVKZ42A6R_PTLH6C INGEST_TIME 2012-09-24T13:00:56Z PACKAGE AA00011811_00001