PROTEOLYTIC PROCESSING OF THE N-TERMINUS OF CITRUS TRISTEZA
VIRUS OPEN READING FRAME 1
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To my loving family
TABLE OF CONTENTS
LIST O F TABLES ....................................................... v
LIST O F FIG URES ..................................................... vi
A B STRA C T ........................................................... vii
1. IN TRO D U CTIO N ................................................. 1
Citrus Tristeza Virus: Classification and Genome Structure ........... 2
Viral Expression Strategies .................................... 6
Closterovirus Expression Strategy ............................... 7
Virus Population Structure .................................... 10
Different Types of Proteases are Encoded in Viral Genomes ......... 12
Viral Cysteine Proteases: Leader versus Main Proteases ............ 13
Closterovirus Protease Classification ........................... 16
O bjectives ................................................ 17
2. SEQUENCE ANALYSIS AND DETERMINATION OF
AUTOPROTEOLYTIC ACTIVITIES OF CITRUS TRISTEZA VIRUS
PUTATIVE CYSTENE PROTEASE DOMAINS ....................... 18
Closterovirus Genomes Encode for Papain-Like Cysteine Proteases ... 18
M aterials and M ethods ....................................... 23
V irus Isolate ......................................... 23
RNA Isolation and Complementary DNA (cDNA) Synthesis .. 23
Polymerase Chain Reaction. Amplification of Protease
Domains and Cloning Strategy .................... 23
In vitro Transcription and Translation ..................... 25
Protease Inhibitor Assays ............................... 26
R esults ................................................... 26
CTVL 1 is Proteolytically Active ......................... 26
No Proteolytic Activity was Detected for the CTVL2 Protein .. 27
Processing Pattern of the N-terminal Region of CTV ORF 1
Revealed Three Cleavages in the Region ............ 30
Sensitivity to Protease Inhibitors ......................... 30
D iscussion ................................................ 32
3. SITE DIRECTED MUTAGENESIS OF PUTATIVE ACTIVE AND
CLEAVAGE SITES OF CTVL 1 AND CTVL2: THE N-TERMINAL
PROCESSING DYNAMICS ........................................ 36
Materials and Methods ...................................... 38
Site Directed Mutagenesis .............................. 39
In vitro Transcription and Translation ..................... 40
Results ................ .............................. 40
Cysteine 403 and Histidine 1 are Involved in CTVL I Activity.. 40
CTVL 1 Might be Responsible for More Than One Cleavage
Within the CTV ORFI N-Terminal Region .......... 41
CTVL2 is an Active Protease ........................... 42
CTVL 1 Might be an Additional Substrate for CTVL2 ........ 42
Mutations at the P1 Position of PCS1 and PCS2 are not Well
Tolerated ..................................... 45
D iscussion ................................................ 46
4. CIS TRANS ACTIVITY OF CTVL I AND CTVL2 .................... 54
M aterials and M ethods ....................................... 57
Virus Isolate, cDNA Synthesis and Cloning ................ 57
Post-Translational Cis/Trans Activity Assays ............... 57
Co-Translational CisiTrans Activity ...................... 57
R esults ................................................... 58
CTVL 1 Encodes a Cis Acting Protease Tested in an
In vitro A ssay ........................................ 58
CTVL 1 -L2-PCS2 did not Complement the Proteolytic Reaction
in a Post-Translational Assay ..................... 59
CTVL2 may have Trans Activity as Shown by a Co-Translational
Trans Proteolytic Assay .......................... 62
D iscussion ................................................ 65
5. SUMMARY AND CONCLUSIONS ................................. 69
6. REFEREN CES .................................................. 70
7. BIOGRAPHICAL SKETCH ........................................ 79
LIST OF TABLES
2.1: Sequence of oligonucleotides used to amplify PCP domains from ORF 1 of
CTV strain T2K .................................................. 25
3.1: Oligonucleotide primer sequences used to introduce mutations to the putative
active sites ...................................................... 39
3.2: Oligonucleotide primer sequences used to introduce mutations into the
putative cleavage sites ............................................. 40
LIST OF FIGURES
* 1.1: Citrus tristeza virus symptoms in different host species ................ 3
* 1.2: Schematic representation of citrus tristeza virus genome organization and
expression strategies ............................................... 5
0 1.3: Electron micrographs of citrus tristeza closterovirus particles ............ 7
S 1.4: General mechanism of action of cysteine proteases ................... 14
0 2.1: Protein sequence analysis of described and putative cysteine proteases of
Potyvirus and Closterovirus ......................................... 22
0 2.2: Schematic representation of CTV ORF I N-terminal region and expression
constructs of CTV PCPs ........................................... 24
* 2.3: Analysis of proteolytic activity of the N-terminal region of CTV ORF1... 29
* 2.4: Effect of chemical protease inhibitors in the proteolytic activity of CTVL 1
and CTV L2 ..................................................... 31
0 3.1: Site directed mutagenesis of predicted catalytic amino acids of CTVL 1 and
C TV L2 ........................................................ 44
* 3.2: Site directed mutagenesis of PCS 1. Localization of CTVL2 substrate site 44
0 3.3: SDS-PAGE autoradiogram of CTVL1-L2-PCS2 proteins mutagenized at the
P 1 position of their putative cleavage sties ............................. 46
0 4.1: CTVL1-PCS 1 trans complementation of proteolytic activity ........... 60
* 4.2: CTVLI-L2-PCS2 tans proteolytic activity .......................... 61
0 4.3: Assessment of cis-trans activity of CTVL 1 and CTVL2 .............. 64
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
PROTEOLYTIC PROCESSING OF THE N-TERMINUS OF CITRUS TRISTEZA
VIRUS OPEN READING FRAME 1
Chairperson: Charles L. Niblett
Major Department: Plant Molecular and Cellular Biology Program
Citrus tristeza virus (CTV) causes one of the most economically important
diseases in commercial citrus worldwide. CTV is a member of the genus Closterovirus in
the Closteroviridae family of positive stranded plant RNA viruses. It is characterized by
long flexuous rod-shaped particles, with a genome of- 19 Kb with 12 open reading
frames (ORFs). ORF1 encodes a polyprotein with two putative papain-like cysteine
protease domains (PLPs) designed as CTVL1 and CTVL2, a helicase, a methyl
transferase and a RNA dependant RNA polymerase (RdRp) expressed by a +1 ribosomal
In the present study, the N-terminal region of ORF1 was cloned and used as a
template in a transcription and translation assay. The presence of two cysteine proteases
was demonstrated in vitro. The putative catalytic amino acids for both cysteine protease
domains of CTV were previously identified by sequence alignment with the papain-like
protease of beet yellows virus. Site directed mutagenesis confirmed C3 H596 and C896 -
H-16 as the residues in the active sites for CTVL1 and CTVL2, respectively. Proteolytic
processing of the region includes a unique cleavage by the CTVL1 domain at the first
putative cleavage site (PCS) G484-G485, located at its C-terminus. A second cleavage is
mediated by CTVL2 at the second PCS G976-G977. These proteolytic activities released
both proteases from the polyprotein, and they were independent of each other. An
additional cleavage is mediated by CTVL2 at a previously unreported within the first
protease. This second activity of CTVL2 is dependent on CTVL1 activity. All the
proteolytic activities were insensitive to a protease inhibitor cocktail of broad specificity
for serine, cysteine and aspartic proteases.
Under the conditions tested, we were not able to detect trans activity for CTVLl;
however, we obtained some evidence, which indicates that CTVL2 may be able to act in
trans at the newly detected cleavage site. A working model includes independent
autocatalytic releases of CTVL1 and CTVL2 from the polyprotein and a trans cleavage of
CTVL1 mediated by CTVL2. The biological significance of the processing of the ORFI
N-terminal region remains to be determined.
Citrus tristeza closterovirus (CTV) causes "tristeza," the most economically
important viral disease of citrus (Kitajima et al., 1964; Rocha-Pefia et al., 1995).
"Tristeza," which means sadness in Spanish, describes a decline disease caused by CTV
and occurs on citrus scions that are propagated on sour orange rootstocks (Bar-Joseph et
al., 1989). Sour orange rootstock has been widely used in the citrus industry due to the
tolerance of this rootstock towards different pathogens such as Phytophtora (Klotz,
1978), and several graft-transmissible pathogens. Also, sour orange rootstocks are
adaptable to different soils and are compatible with most citrus (Rocha-Pefia et al., 1995).
Citrus tristeza virus causes a variety of symptoms, depending both on the infecting
strain as well as the infected host/rootstock combination (Rocha-Pefia et al., 1995).
Symptoms include quick decline on sour orange rootstock (Figure 1. IA), honeycombing
at the bud union (Figure 1.1 B), stem pitting of sensitive cultivars (Figure 1.1 C), leaf vein
corking of sensitive cultivars (Figure 1.1 D), reduced fruit size and quality (Figure 1.1 E)
and vein clearing and leaf cupping (Figure 1.1F).
As an indicator plant, Mexican lime (Citrus aurantifolia (Christm.) Swingle) is
the most sensitive to CTV, but the severity of the symptoms does not necessarily
correlate with those observed in other hosts. Decline inducing (DI) strains are detected in
sweet orange (Citrus sinensis (L.) Osbeck) grafted onto sour orange seedlings. When
CTV-DI infected budwood is propagated onto sour orange (Citrus aurantium (L.))
seedlings, severe stunting results. Seeding yellows (SY) strains are identified on sour
orange, acid lemon (Citrus limon (L.) Burm), and grapefruit (Citrus paradisi (Macf.))
seedlings, which suffer chlorosis and stunting. For the strains which cause stem pitting
(SP) on grapefruit and/or sweet orange, there is an indexing system on Duncan grapefruit
and Madame Vinous sweet orange seedlings. CTV-SP-infected plants in the field show
longitudinal pits formed in the wood of the stems/branches of the scions, independent of
the rootstock. There is loss of plant vigor and yield reduction. Also vein clearing and
vein corking of leaves of sweet orange have been observed as symptoms for SP strains
(Rocha-Pefia et al., 1995).
Citrus Tristeza Virus: Classification and Genome Structure
CTV belongs to the order Nidovirales and is a member of the Closteroviridae
family of positive single-stranded RNA viruses according to the International Union of
Microbiology Societies classification (van Regenmortel et al., 2000).
The initial characterizations of the virus included the detection of thread-like
particles associated with the disease symptoms. These particles were about 2,000 pLm in
length and 10-12 Lm in diameter, and they resembled those of beet yellows virus
(Kitajima et al., 1964). The CTV genome was described as single-stranded positive sense
RNA (+ssRNA) by Bar-Joseph and Lee (1989) of approximately 20Kb in length (Bar-
Joseph et al., 1985). Once the complete genome of the T36 isolate was sequenced, it was
determined to have a 19,296 nt sequence, with 12 open reading frames (ORFs) that
potentially coded for 17 products (Pappu et al., 1994; Karasev et al., 1995).
Figure 1.1: Citrus tristeza virus symptoms in different host species. (A) Quick decline of
a sweet orange tree on sour orange rootstock. (B) Honeycombing of the bud union of a
sour orange rootstock. (C) Stem pitting in stems of Mexican lime seedlings. (D) Vein
corking symptoms on leaves of Mexican lime Citrus aurantifolia. (E) Grapefruit from a
tree infected with a stem pitting isolate of CTV; the fruit on the right comes from an
uninfected tree. (F) Vein clearing and cupping in leaves of Mexican lime.
Pictures taken from www.ecoport.org Pictures A, B and E by Lee, R. Pictures C, D, F by
Roistacher, C. N.
A diagram of the CTV genome with the putative open reading frames (ORFs) is
shown in Figure 1.2A and the expression strategies employed by CTV are shown in
Figure 1.2B. Sequence analysis allowed the assignment of potential functions for most of
the ORFs (Karasev et al., 1995). Currently, experimental evidence is being obtained to
corroborate some of these putative functions. The genome organization includes a 5'
proximal polyprotein of a calculated molecular mass of 349 KDa. This polyprotein
contains domains for two putative papain-like cysteine proteases (CTVLI and CTLV2), a
methyl transferase (MT), a helicase (HEL), and a RNA dependant RNA polymerase
(RdRp) that is expressed via a +1 ribosomal frameshift (;evik, personal communication)
that defines the ORFIa and ORFIb (Karasev et al., 1995).
The functions of genes located in the 3' region of the genome include a heat shock
protein 70 homolog (HSP70h) which has been detected by antibodies in CTV- (Rosales,
personal communication) and BYV- infected tissues (Napuli et al., 2000) by tissue
printings, as well as in association with the virion (Figure 1.3B). A duplicated coat
protein (p27) is expressed in CTV- infected tissues (Febres et al., 1994), and both the coat
protein (CP) and its diverged copy form part of the virion (Figure 1.2A; Febres et al.,
1996). This coat protein duplication is common in the Closteroviridae family, and it has
been suggested that the duplication occurred before the separation of BYV and CTV from
their common ancestor (Boyko et al., 1992).
Some of the functions of the 3' ORFs of closteroviruses have been elucidated
through the construction of a full length BYV cDNA infectious clone. That study
revealed that the gene product of the ORF 1 a/b is sufficient for RNA replication and
0k 10 15 20
PRO-1 MMX HEL p33 p65 P27 p18 p20
PFD-2 RdRp p6 p61 CP p13 p23
01" la lb 2 3 4 5 6 7 8 910 11
Viral cellular Ribosomal Subgenomic
proteases Pro teases? + 1 frameshift RNAs
Figure 1.2: Schematic representation of the citrus tristeza virus genome organization and
expression strategies. Panel A represents the open reading frames and the putative
proteins encoded. Panel B represents the genomic and sub genomic RNAs as single lines.
The putative proteins translated from each RNA are shown as boxes (Solid boxes
represent proteins of demonstrated activity or presence in CTV-infected tissue).
Figure from Manjunath et al., (2000).
transcription. This ORF la/b self-replicating clone was used as a vehicle to evaluate the
effects of adding specific 3' ORFs to the complex. These experiments revealed that the
ORF that encodes the p21 protein functions as an enhancer of genome amplification
(Peremyslov et al., 1998). Other functions have been revealed for other 3' ORFs, e.g. a 3'
end proximal ORF encodes a 23 kDa protein with the ability to cooperatively bind single-
stranded and double-stranded RNA in a non-sequence specific dependant manner (L6pez
et al., 2000). A protein, designated as p20, has been reported to interact strongly with
itself in a yeast two hybrid system and apparently forms amorphous inclusion bodies in
infected protoplasts, but no biological function has been assigned to it (Gowda et al.,
2000). The expression and function of the remaining ORFs in the CTV genome remain to
be identified and characterized.
Viral Expression Strategies
Among the positive-sense RNA viruses, a general classification into two big
"super groups," picorna- and alpha-like, can be made according to the expression
strategies used in the viral infection-replication cycle. The picoma-like viruses include
viruses in the Potyviridae, Comoviridae and Sequiviridae families. The genomes of these
viruses encode a long polyprotein that is proteolytically processed into smaller functional
domains by virus-encoded proteases. This expression strategy produces equimolar
amounts of all the viral proteins (Krdusslich and Wimmer, 1988). The expression of a
polyprotein allows the temporal and spatial control of the activity of the specific domains
as they become available in the form of individual proteins or processing intermediates
(Garcia et al., 1999).
The second class is comprised of the alpha-, como- and sobemo-like groups and
includes the Closteroviridae family. This group includes viruses that use a variety of
expression strategies. The synthesis of a polyprotein that is proteolytically processed into
non-structural proteins, including the RNA polymerase, mediates the synthesis of
transcripts known as sub-genomic RNAs (sgRNA). These sgRNAs encode the ORFs
downstream from the polyprotein, and these include the viral structural proteins. This
strategy allows the differential expression of enzymatic non-structural proteins versus the
Figure 1.3: Electron micrographs of citrus tristeza closterovirus particles. (A) Viral
particle immunolabeled with a gold-conjugated rabbit polyclonal antiserum against the
diverged copy of the coat protein, showing its localization at one end of the particle. (B)
Viral particle immunolabeled with a gold-conjugated chicken polyclonal antiserum
against the HSP7Oh protein, showing its interaction with the viral particle. Gold labeled
particles are shown with an arrow.
Picture A by Febres, V. (1996). Picture B by Rosales, M. (2001).
increased need for structural proteins (Krausslich and Wimmer, 1988). Polyprotein
expression is frequently combined with other strategies such as alternative translation
initiation sites, frameshifting, and readthrough of suppressible termination codons (Garcia
et al., 1999). The low frequency with which these alternative mechanisms occur
represents an additional mechanism to control the differential expression of different
types of proteins (KrAusslich and Wimmer, 1988).
Closterovirus Expression Strategy
Sequence analysis of the BYV genome revealed the presence of a polyprotein
encoded by the ORF ua/b (Agranovsky et al., 1994). The Closterovirus polyprotein
encodes one or two papain-like cysteine proteases, a MT and a HEL (Agranovsky et al.,
1994; Karasev et al., 1995; Jelkmann et al., 1997; Zhu et al., 1998). A dual activity has
been shown for the BYV L-pro. First, auto-proteolytic release from the polyprotein was
demonstrated in an in vitro transcription and translation assay (Agranovsky et al., 1994)
as well as in an in vivo assay, where it was demonstrated that the proteolytic release of L-
pro was essential for RNA replication (Peremyslov et al., 1998).
Analysis of BYV L-pro mutants revealed that this protein had a high tolerance to
structural changes in its N-terminal region, with the exception of a 54 amino acid stretch
at the 5' region of the ORF, which was important for virus viability. It also was
demonstrated that this protein was not essential for viral genome amplification, but its
activity increased the viral RNA level 1,000-fold when compared to the basal level
(Peremyslov et al., 1998; Peng and Dolja, 2000).
Further maturation of the closterovirus polyprotein has been determined by the
use of monoclonal antibodies against the BYV methyl transferase and helicase, which
have been detected as individual proteins in BYV-infected tissue (Erokhina et al., 2000).
Similarly, the presence of the RdRp as an individual protein in CTV-infected tissue has
been reported ((;evik, personal communication). The nature of the protease(s)
responsible for the release of these proteins from the polyprotein remains to be identified.
The expression of the polyprotein also includes a ribosomal + 1 frameshift, from
which the RdRp is expressed. This event has been experimentally demonstrated with
both in vivo and in vitro assays, and it was estimated to occur in 1-5 % of the translation
events ((;evik, personal communication). The remainder of the closterovirus ORFs are
likely to be expressed via 3'-co-terminal sgRNAs. The presence of six subgenomic RNA
species has been detected in BYV-infected tissues, whereas there is a great variability in
the number of sgRNA species detected in CTV-infected tissues (Agranovsky, 1996).
Studies on the kinetics of accumulation of CTV RNAs revealed temporal control of the
synthesis of the different sgRNAs both in host and non-host protoplasts (Navas-Castillo
et al., 1997).
This temporal control was further studied with the use of a BYV infectious clone
in which individual genes were tagged with bacterial P-glucuronidase (GUS). Analysis
of the results revealed that the temporal regulation of gene expression included early
expression of HSP70h, CP and its diverged copy, as well as the p21 protein, while the
expression of two other ORFs (p64 and p20) was related to the late phase of viral
infection. This study also revealed that the expression of the 3' ORFs can affect
transcription of sgRNA species since the deletion of six of the 3' ORFs resulted in the up
regulation of the remaining sgRNAs in the deleted construct. This pattern of temporal
regulation of multiple transcriptional units is unique among RNA viruses (Hagiwara et
During infection with CTV, the presence of RNA molecules that consisted of
different deletions of the CTV genome and included variable portions of the 5'- and 3'-
terminal regions was observed (Mawasssi et al., 1995a; b). These defective RNA (D-
RNA) molecules seem to be replicated via a double-stranded RNA (dsRNA) intermediate
using a replicase-driven template switching mechanism (Ayll6n et al., 1999). The
presence of D-RNAs has been related to reduced accumulation of the helper virus,
inducing attenuation of symptoms in tomato bushy stunt virus-infected plants (Scholthof
et al., 1995) and to the exacerbation of symptoms in broad bean mottle virus infections,
according to the infected host (Romero et al., 1993). Their presence also has been found
not to interfere with symptoms in clover yellow mosaic virus- (White et al., 1991),
cucumber mosaic cucumovirus- (Graves et al., 1995), and CTV- (Ayll6n et al., 1999)
The presence of a D-RNA of almost identical sequence in two CTV isolates (T317
and T318) that greatly differed in their pathogenicity suggested that this D-RNA was not
the cause of the increased pathogenicity observed for the T318 isolate (Ayll6n et al.,
1999). Alternatively, D-RNAs with 5'- regions larger than 4,000 nt present in the CTV
isolate VT have been implicated in the suppression of seedling yellows symptoms on
specific hosts (Yang et al., 1999).
Virus Population Structure
The long term survival of viruses in their host (persistence) is affected by genetic,
ecological, and environmental factors. There are three mechanisms that describe the
persistence of a virus in its host. These include the following: balance between number
of infected cells and multiplication of the host cells; persistence in a limited number of
cell types; and continuous infection of a susceptible host with or without persistence in
infected cells (Domingo et al., 1998).
For the tristeza disease, the perennial nature and the susceptibility of citrus hosts
result in persistence of the virus in the infected plants. A consequence derived from this
is that the host is susceptible to multiple infections with the virus over the years, resulting
in a viral population that can be a mixture of different genotypes (Manjunath et al., 2000).
The presence of different genotypes in CTV-infected plants has been experimentally
demonstrated by the use of single-stranded conformational polymorphism (SSCP) of the
genes that encode the coat protein and its diverged copy. These experiments showed that
CTV isolates were composed of a population of genetically related variants (haplotypes),
having a predominant one in the population, although there was also a case of two
haplotypes with high divergence in the same isolate (Kong et al., 2000; Niblett et al.,
The presence of viral haplotypes in CTV isolates revealed the possibility of their
existence as quasispecies in the infected plant (Manjunath et al., 2000). Due to the
absence of proofreading activity of viral RNA polymerases (Drake and Holland, 1999),
RNA virus populations consist of complex distributions of genomes carrying different
mutations forming the population structure known as quasispecies (Domingo et al.,
1998). These quasispecies are represented by molecular variants of a genotype in the
range of I or 2 % difference between them (Davis, 1999).
The genetic organization of the viral population as a quasispecies represents an
adaptive strategy as it constitutes the raw material on which selective forces and random
sampling events act in the molecular evolution of RNA viruses (Domingo et al., 1998). A
study of sequence identity of five mild CTV isolates (asymptomatic in field trees and
causing only weak symptoms on the indicator plant, Mexican lime) from different
geographic and host origins revealed little variation among these isolates (Albiach-Marti
et al., 2000). That study reveals the importance of the interactions between specific viral
and host determinants. Most of the mutations among the five isolates were silent
mutations or changes that resulted in similar amino acids, suggesting that this CTV
genotype is well adapted to its hosts, and has not changed in several hundred years.
Another study of the haplotype distribution of CTV isolates after host change or aphid
transmission revealed changes in the populations of the original isolate and successive
subisolates. The extent of those changes was greater than that observed between isolates
from different geographical locations. These results suggested that adaptation to a new
host changed the haplotype distribution, this change being more important than the
geographical origin of the isolate (Ayll6n et al., 1999) even though this is not a constant
outcome of the adaptation process.
Different types of Proteases are Encoded in Viral Genomes.
Proteases are encoded in the genomes of diverse viral groups, which include non-
enveloped single-stranded RNA (ssRNA) viruses, enveloped ssRNA viruses, non-
enveloped double-stranded DNA (dsDNA) viruses and enveloped dsDNA viruses (Bab6
and Craik, 1997). Proteolysis of viral polyproteins is found primarily in positive-sense
(+) ssRNA viruses and retroviruses (Krdusslich and Wimmer, 1988).
Various studies have revealed several common characteristics for the proteases of
this group (Ryan and Flint, 1997 and references therein). They are commonly found as a
domain of a larger protein, which can represent alternative processing products, and their
activity can depend on the specific processing, intermediate location, and also can be
modified by interaction with other proteins or RNA. Viral proteases also can cleave host
proteins in trans, thus modifying host functions and can be regulated until a particular
cellular environment is encountered (Ryan and Flint, 1997).
Cellular proteases have been classified according to their active site nucleophiles
into serine-, cysteine-, aspartyl- and metallo- proteases (Ryan and Flint, 1997). Serine-
proteases are characterized by having an aspartate, a histidine and a serine as a catalytic
triad (Ryan and Flint, 1997). Cysteine-proteases have a catalytic dyad of cysteine and
histidine. On the carboxyl-side of the catalytic cysteine, there is a conserved aromatic
residue that is characteristic of all papain-like cysteine-proteases (Ziebuhr et al., 2000).
The aspartic or acid proteases are characterized by the presence of two aspartic acid
residues in their active site, and the metallo-proteases perform their nucleophilic attack
using a metal cation (Ryan and Flint, 1997). The general mechanism of catalysis by
cysteine proteases is through an acid-base formation of an acyl-thiol intermediate
followed by a hydrolysis reaction (Figure 1.4).
The use of chemical inhibitors has helped in the classification of the proteases. As
examples, aspartic-proteases are specifically inhibited by pepstatin; cysteine-proteases by
trans-epoxysuccinyl-L-leucylamido-(4-guanidino) butane (E-64) and cystatins; metallo-
proteases by chelating agents and seine proteases by diisopropyl fluorophosphate and
phenylmethylsulphonyl fluoride (Kay and Dunn, 1990).
Viral Cysteine Proteases: Leader versus Main Proteases
The presence of papain-related cysteine proteases has been reported for several +
ssRNA viruses. Protein alignments of the PLPs from different + ss RNA viruses has
revealed low conservation of the sequences of these proteases. Only the CW(Y)
dipeptide, which includes the catalytic cysteine, is conserved, and there is a much lower
conservation in the area surrounding the catalytic histidine.
F S" *Him + RCOOH
3" P, "Him -
S-C=o NH2 m
Figure 1.4: General mechanism of action of cysteine proteases. The acyl-thiol
intermediate occurs by the formation of a non-covalent complex between the enzyme and
the substrate. This complex undergoes acylation, forming and releasing an amine R'-
NH2; this is followed by a deacylation step, which releases the second product,
regenerating the free enzyme. Im and +Him refer to imidazol and protonated imidazol,
respectively. (Rao et al., 1998).
This same protein alignment prompted the grouping of these viral proteases into
two classes; leader and main proteases (Gorbalenya et al., 1991). Leader proteases can
mediate a single cleavage at their own C-termini, or a double cleavage at sites located in
the amino-terminal half of the polyprotein. The cleavage site for leader proteases has at
least a small residue, and their active site is composed of a cysteine and a downstream
histidine. The main proteases represent proteins that mediate the processing of the non-
structural proteins, being in some cases the only protease encoded in the viral genome
(Gorbalenya et al., 1991; Ziebuhr et al., 2000). Sequence alignments revealed that the
main protease group possesses an additional conserved domain, the "X" domain
(Gorbalenya et al., 1991) and its activity was linked to the ability to perform trans-, but
not cis-, proteolytic cleavages for the rubella virus non-structural protease (Liang et al.,
Even though the sequence alignments provided some insight for the classification
of PLPs, experimental confirmations have not always been as straightforward. Rubella
virus (RUB) is an enveloped + ssRNA virus of the family Togaviridae. Part of its
genome encodes a polyprotein with an RdRp, a helicase and a papain-like cysteine
protease (RUB NSP). In vitro transcription and translation assays and site directed
mutagenesis allowed the identification of the catalytic amino acids as well as the residues
at a cleavage site. Co-expression of substrates and enzymes failed to complement the
reaction in trans, resulting in RUB NSP being reported only as a cis-acting protein. This
observation disagreed with the "main protease" prediction made by Gorbalenya et al.
(1991), but there were also characteristics such as the median location of the protease in
the polyprotein instead of N-terminal location and its post-translational, instead of co-
translational, maturation that also differed from the characteristics typical of leader
proteases (Chen et al., 1996).
Further research into the activity of RUB NS protease revealed that this protein
was able to perform trans-proteolysis (Yao et al., 1998), and it required the presence of
divalent cations for this activity. This protein was reclassified as a main protease, but
again, there were several characteristics such as its cleavage at a single site, and a space
of about 40 residues between the active site and cleavage site as for the typical leader
proteases, which did not fit with the description typical of main proteases. This suggested
that the classification of the viral papain-like protease family was more complicated than
initially thought (Liu et al., 1998).
The characterization of the metal ion binding activity of RUB NS protease
revealed the critical residues for the activity and showed that Zn> did not have a major
effect on the secondary structure of the protein. It also was shown that this proteolytic
activity could be blocked by metal ion chelators and the metallo-protease inhibitor
captopril. This finding suggested that RUB NS protease is not a papain-related cysteine
protease, but a novel metallo-protease (Liu et al., 2000).
Other proteases that have been classified as main proteases are the alphavirus
nsP2 protease (Hardy and Strauss, 1989), foot and mouth protease (Kirchweger et al.,
1994) and coronavirus PLP- I protease (Bonilla et al., 1997).
Closterovirus Protease Classification
The proteases of CTV have been classified in the MEROPS database as belonging
to the CA clan, family C42 (Rawlings and Barrett, 2000). Currently the two PLP of CTV
and the BYV L-pro are the only members of that family in the MEROPS database. BYV
L-pro is the type member of the C42 family. Sequencing of the BYV genome revealed
the presence of a putative papain-related cysteine protease and its activity was
corroborated in an in vitro transcription and translation assay (Agranovsky et al., 1994).
The major objectives of this investigation are the characterization and study of the
proteolytic activities of the N-terminal region of CTV ORF la/b. This will contribute to a
better understanding of the biology of CTV and aid in the development of more effective
control measures. The specific objectives of this research are the following:
1. Determine the proteolytic activity ofCTVL I and CTVL2.
2. Determine their sensitivity to chemical protease inhibitors.
3. Determine the effect of site directed mutagenesis of the putative amino acids
on the active sites of both CTVL 1 and CTVL2.
4. Site directed mutagenesis of the amino acid that occupies the P 1 position for
both putative cleavage sites.
5. Determine potential cis and trans proteolytic activities of CTVL I and CTVL2
using CTV ORF I N-terminal region as substrate.
6. Establish a working model for the CTV ORF 1 N-terminal proteolytic
SEQUENCE ANALYSIS AND DETERMINATION OF AUTOPROTEOLYTIC
ACTIVITIES OF CITRUS TRISTEZA VIRUS PUTATIVE CYSTEINE PROTEASE
Closterovirus Genomes Encode for Papain-Like Cysteine Proteases
The presence of genes encoding classical cysteine proteases related to cellular
papain-like proteases has been reported in the genome of several positive-stranded RNA
viruses (Gorbalenya et al., 1991). Beet yellows closterovirus (BYV) is the type member
of the Closteroviridae, a group of plant viruses with flexible filamentous particles and a
single stranded (ss) positive sense (+) RNA genome.
Sequence analysis of the complete genome revealed similarities with the genomes
of Tobraviridae. This similarity allowed the identification of a helicase, a methyl
transferase and an RNA-dependant RNA polymerase (RdRp) domain within the first
open reading frame (ORF) of BYV, which was designated as the replication complex
(Agranovsky et al., 1994).
Alignments of the replication complexes of the Closteroviridae and Tobraviridae
viral groups reveal two unique regions for BYV (Closteroviridae) that account for the
difference of 1387 and 1328 residues with respect to tobacco rattle virus (TRV,
Tobraviridae) and pea early browning virus (PEBV, Tobraviridae) ORF I translation
products, respectively. These unique sequences did not produce any significant result
when compared against the database, but a motif analysis identified the sequence GLCY,
which resembles the sequences around the active site of papain-like thiol proteases
(Agranovsky et al. 1994).
The region that encodes the putative papain-like cysteine protease in the BYV
genome was cloned and used as a template for an in vitro transcription and translation
reaction (Agranovsky et. al, 1994). This experiment confirmed the presence of
proteolytic activity within the complex. The active amino acids involved in the
proteolytic activity were identified by site-directed mutagenesis and were mapped to the
conserved residues Cys509 and His569 as well as the cleavage site at Gly88-Gly59. Point
mutations of any of these amino acids resulted in the loss of proteolytic activity, whereas
mutations in non-conserved amino acids near the predicted active site had different
effects. Substitution of His556 was tolerated, while substitutions of Cys517 and Cys518
drastically reduced the proteolytic activity but did not abolish it completely (Agranovsky
et al., 1994).
Sequencing of the citrus tristeza virus (CTV) genome revealed the presence of
two putative papain-like cysteine proteases, named CTVL I and CTVL2 (Karasev et al.,
1995). Pairwise comparisons revealed similarity between the two putative cysteine
protease domains of CTV ORF I with the unique papain-like cysteine protease domain of
BYV. This allowed the prediction of the putative amino acids in the active sites as well
as the putative cleavage sites (Karasev et al., 1995). The positions of the putative
cleavage sites predict two proteins of 484 (CTVL 1) and 492 (CTVL2) residues. When
the complete amino acid sequence of each predicted protein was compared with that of
the BYV L-pro sequence, the putative cysteine protease domain was mapped to the C-
terminal end of each protein and encompassed approximately 30 % of that sequence.
Comparisons of upstream sequences ofCTVL1, CTVL2, BYV L-pro and little cherry
virus (LChV, Closteroviridae) did not produce any significant alignment among them
(Karasev et al., 1995; Jelkmann et al., 1997; Zhu et al., 1998). When compared, the
papain-like cysteine protease (PLP) upstream sequences of LChV and lettuce infectious
yellows virus (LIYV, Closteroviridae) additional conserved sequences, interrupted by
deletions or insertions of different lengths were found, but they were not identified with
any known domain (Jelkmann et al., 1997).
A protein sequence alignment of the C-terminal regions of BYV L-pro, CTVL 1,
and CTVL2 with the helper component protease (HC-pro) of seven different potyviruses
is shown in Figure 2.1A. In this alignment it can be observed that although there is little
conservation among the sequences of closteroviruses when compared to potyviruses, the
putative and already confirmed active amino acids are conserved in all the proteins. The
sequences of BYV L-pro vs. CTVL1 showed a 14 % identity and 25 % similarity; BYV
L-pro vs. CTVL2 showed 16 % identity and 26 % similarity; and CTVL1 vs. CTVL2
showed 17 % identity and 28 % similarity.
The position of the putative cleavage site for each protease differed considerably
between the potyvirus and closterovirus groups, but in all the cases the Gly-Gly pair was
present. A dot matrix generated by Align X of the putative CTVL1 and CTVL2 cysteine
proteases and BYV L-pro is shown in Figure 2. 1B. In this matrix it can be observed that
the more conserved regions are in the areas surrounding the catalytic amino acids, which
are located at the beginning and the end of the plots.
The sequence conservation among the two putative cysteine proteases of CTV as
well as their similarity in size have suggested that these proteins evolved in the CTV
genome by a tandem duplication phenomenon (Karasev et al., 1995). Even though gene
duplication with subsequent divergence is a common evolutionary mechanism observed
in DNA genomes, only a few examples are reported for viral RNA genomes (Boyko et
al., 1992). The presence of a duplicated papain-like thiol protease also has been
predicted or identified for the genomes of Coronaviridae and Arteriviridae of the order
Nidovirales (Kanjanahaluethai and Baker, 2000; Zieburhr et al., 2000; Tijms et al., 2001
and references therein).
Gene duplication is also reported for the coat protein gene of the Closteroviridae
family; where a diverged copy of the coat protein (p27) appears in the genome (Boyko et
al., 1992; Febres et al., 1994; Agranovsky et al., 1995). It has been suggested that the
CP duplication occurred in the common ancestor of BYV and CTV (Boyko et al., 1992).
For this duplicated CP the degree of conservation among both proteins was 30 %
similitude (Boyko et al., 1992).
In the present study, the proteolytic activity of the N-terminal region of the CTV
ORF I translation product was demonstrated by in vitro transcription and translation
assay. Expression constructs encoding either the CTVL I and/or CTVL2 were generated
and used in a wheat germ-coupled transcription and translation system. Translation
products were subjected to polyacrylamide gel electrophoresis analysis, and the
proteolytic activity was determined based on the migration patterns of the products
observed in the gels.
20 40 60
g HD&F;@ j -----------RRADLSURRALG PTVGAFKTYL*EYGR
MK------- R-R!FRIRDVDLGPF PRIWFHRLERLYGK
P A. LC AO ---------- .TFR ............T KL
1 ~ ~ r -- -
/Y / /2o
* "J/ 1 / ,
7 / // 5
'V // '7 /
/, /~ /,
// ", ,'
L -, : ;,z i i ._
r' - -
/ "' i
,- *. /
:/ ., ,
Figure 2.1: Protein sequence analysis of described and putative cysteine proteases of
potyviruses and closteroviruses. (A) Protein alignment of C-terminal cysteine protease
regions. Closterovirus: T2KPI, citrus tristeza virus CTVL 1; T2KPII, citrus tristeza virus
CTVL2; BYV, beet yellows virus L-pro (Accession Number X73476). Potyvirus:
GVLRV, grapevine leafroll virus (AF039204); PPV, plum pox virus (AJ243957); PVY,
potato virus Y (AF229174); SMV, soybean mosaic virus (S42280); TEV, tobacco etch
virus (NP062908); BCMV, bean common mosaic virus (U19287); TVMV, tobacco vein
mottle virus (X04083); ZYMV, zucchini yellow mosaic virus (L31350). (B) Dot matrix
generated by Align X of the putative CTVL I and CTVL2 cysteine proteases and BYV L-
pro. The conditions to generate the matrices were 30 % stringency with a window size of
80 100 120
DSLKFMR(;T:TFSVF .LStES4-DLRS PNHHLVGG-----------------------
AASRYGVRGYYSAPRCF CYNDSPK-PLTS YHNG G-------------------------
SALKHCVRGRlVSRSLF DVASAFS#PFYS 'FIG -----------------------
RSLC ILYGAYTSR(G DYDAKF]KDLR SAVIAGKDG EVV2SDTPAMXQKTIEAVY
RNARJ LDHEAKIF DIG, ATQ 14NQ ,S#ASD-ThOSSM4TXLvGG--
HDAS I#VDHDTQT. SzGiZTT, SVYSQ#.LAND-ELESIIHYRVGG--
RflARM LVD ACOTM GI04WND SKXYR~
.RAE~LVDHDNKT -R!T TSQLXEVHS-;LESEKXYNVGG;--
RSAR2DUV~iASQM DS GGfNQL.JQNrASl- DLEGENXHYRVGG~-
ASA ILYAKT TT ?jSQL;EKIASN-TLESPNAQYKVGG--
Materials and Methods
The Florida grapefruit stem pitting CTV isolate 3800 was used as the RNA
source. This isolate is believed to contain at least two distinct viral strains designated as
T2K and T38K (Manjunath et al., 2000). Sequences of the T2K strain were used in the
RNA Isolation and Complementary DNA (cDNA) Synthesis
Double stranded (dsRNA) replicative forms of CTV were isolated from bark
tissue of grapefruit plants following the protocol described by Valverde et al., (1990).
The dsRNA was denaturated by incubation at 70 'C for 5 minutes and quickly transferred
to ice. The oligonucleotide 5'-GTCAAACGAGATATCTTTGTCGAGG-3' was used to
specifically prime the Thermoscript (Gibco BRL) mediated reverse transcription of the
Polymerase Chain Reaction. Amplification of Protease Domains and Cloning Strategy
Polymerase chain reaction (PCR) was used to amplify different portions of the
coding region for the protease domains of ORF 1. Amplification was performed using
2.5u/100 gl of expand polymerase (Boheringer) in reactions containing 50 mM KCI, 10
mM Tris-HC1 pH 9.0, 0.1% Triton X-100, 2.5 mM MgC12, 0.1 mM of each
deoxyribonucleotide triphosphate (dNTPs), 100 pmol of each primer, and 2 5 ptl of
cDNA. The PCR parameters consisted of 92 "C for 2 min., followed by 30 cycles of 30
sec. at 92"C, 30 seconds at 50*C and 1 minute/Kb at 72C.
Figure 2.2 shows a schematic representation of the expression constructs used in
these experiments. Translations of these constructs were expected to contain both
unprocessed products with the full size protein as well as processed products with lower
molecular mass as a result of in vitro proteolytic cleavage.
The primers used to amplify each construct are listed in Table 2.1. Either ApaI
(GGGCCC) or XhoI (CTCGAG) recognition sequences were included within the primer
sequence to facilitate further cloning and manipulation.
G484 G485 G976 G977
CTVL1-L2-PCS2 113 KDa
C VL1-PCS1 59 I I I 1KDa
PCTS-CTVL2PCS2 64 KlDa
Figure 2.2: Schematic representation of CTV ORF I N-terminal region and expression
constructs of CTV papain-like cysteine proteases. The putative cleavage sites that delimit
each protease are denoted by G4"4-G4"5 and G976-G977. Construct CTVL I -PCS I starts with
the first Met of CTV ORFI and ends 40 amino acids after the first putative cleavage site
(PCS). Construct PCS I -CTVL2-PCS2 starts 40 amino acids before the first PCS and
ends 40 amino acids after the second PCS. Construct CTVL I-L2-PCS2 starts at the first
Met of CTV ORF 1 and ends 40 amino acids after the second PCS.
(*) Marks the position of the putative active amino acids C3-H' and C896-H956.
Table 2.1: Sequence of oligonucleotides used to amplify papain-like cysteine protease
domains from ORF I of CTV strain T2K. Restriction sites are shown in bold italic.
CTVL 15' 5'-AAAGGGCCCACCATGTCGAAACTCAGAGGAAGCTT-3'
CTVL1-S1 3' 5'-AAACTCGAGTCAGAGATAACCATCACGTCGCAAGCG-3'
PCS I-CTVL2- 5'-AAAGGGCCCACCATGGACTCTCTTAAGGTTCCTATG-3'
PCS 1-CTVL2- 5'-AAACTCGAGTCAGTCCTTTTCCACAGACCGAATC-3'
PCSI-CTVL2 3' 5'-AAACTCGAGTCAACCCATATTATGGTACTTATTTAA-3'
CTVL2-PCS2 5' 5'-AAAGGGCCCACCATGGGGACTTCTGCACACGTCTTAAT
Following PCR, the products were GeneClean (Bio 101, Inc.) gel purified and
cloned into the pGEM-T vector (Promega). Selected clones were subjected to restriction
enzyme analysis, and those which were oriented under the control of the SP6 promoter
were submitted for DNA sequencing.
In vitro Transcription and Translation
Selected constructs were sequenced and used as templates for in vitro
transcription and translation reactions. The TNT coupled in vitro transcription and
translation wheat germ extract system (Promega) was used following the manufacturer's
instructions. Briefly, 10 il reactions contained 2 ig of plasmid DNA/pil and were
incubated at 30'C for 1 hour in the presence of 0.5 IiCi/ml [3H] leucine. After
incubation, the reactions were subjected to sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) in 7.5 % gels for one hour at 200V. For detection of the
radiolabeled proteins, the method described by Bonner and Laskey (1974) was followed.
Briefly, the gels were fixed in dimethyl sulfoxide (DMSO) and DMSO- 2,5-
diphenyloxazole (PPO). Following hydration-PPO precipitation, the gels were dried and
exposed to x-ray film overnight at -80'C, and developed using an automatic X-ray film
developer (Kodak X-Omat Clinic 1 Processor).
Protease Inhibitor Assays
The sensitivity of the proteolytic reactions was tested by the addition of different
chemical protease inhibitors to the translation mixture. In vitro transcription and
translation reactions were performed as described above with the addition of one of the
following cysteine protease inhibitors: 100 [.M trans-epoxysuccinil-L-leucylamido-(4-
guanidino) butane (E-64); 2.5 mM N-ethylmaleimide (NEM); or a broad spectrum
protease inhibitor cocktail containing 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride
(AEBSF), 0.8 .LM aproptinin, 20 pM leupeptin, 36 p[M bestatin, 15 pM pepstatin A and
14 pM E-64.
The translation mixtures were incubated at 30 'C for 1 hour and the products
analyzed by SDS-PAGE.
CTVL I is Proteolytically Active
Analyses of the proteolytic activity of the two putative cysteine proteases of CTV
ORF I were performed by the construction of three overlapping cDNA constructs that
spanned 640 amino acids from the N-terminal region of ORFI to include the two PCPs
and the predicted cleavage sites. These cDNAs were cloned into the pGEM-T vector
under the control of SP6 polymerase, subjected to coupled in vitro transcription and
translation, and the resulting products were analyzed by SDS-PAGE and
autoradiography. The first construct, CTVL I-PCS 1 (Figure 2.2), started at methionine 1
of CTV ORFI and ended 40 amino acids downstream of the first PCS. The expected
product from this translation was a full length protein with a mass of approximately 58
KDa. If this protein had autoproteolytic activity, it was expected to liberate the 40
downstream amino acids of the PCS G8-G4, producing an additional product of 54
KDa, corresponding to the mature proteolytic product. The SDS-PAGE autoradiogram
of the CTVL 1 -PCS 1 translation product is shown in Figure 3B. Two proteins are present
which correspond to the predicted full size protein (58 KDa) and the mature product of
CTVL 1-PCS 1 (54 KDa). This result confirmed the proteolytic activity of the first PLP
domain of CTV ORF 1.
No Proteolytic Activity was Detected for the CTVL2 Protein
To test the proteolytic activity of the second PLP of CTV ORF 1, the construct
PCS1-CTVL2-PCS2 was designed (Figure 2.2). To design this construct, we had to take
into consideration that CTVL2 is an internal domain of the polyprotein, and that it may
be released from it by proteolysis. This raised three main concerns for the design of this
construct. First was the possible effect on the activity of the protein by the introduction
of an extra amino acid in order to have methionine as the first amino acid for its
expression; second was that the second PLP domain was delimited by a PCS at its N and
C terminal ends, and both PCSs had to be considered as potential substrates; and third
was the length of the amino acid chains that were either before or after the PCS,
designated as the substrates.
These substrates were designed to determine proteolytic activity for each PLP,
based on size differences as determined by the difference in electrophoretic mobilities
between the unprocessed and mature proteins. To detect proteolytic activity, the mass of
the substrate needed to be detectably different when compared to the size of the protein
carrying the proteolytic activity. Due to the C-terminal proximity of the PLP on each
protein, the substrate peptide also needed to be small enough so that it would not include
the first PCP active site in the construct for the second PCP (see Figure 2.2). Therefore, a
length of 40 amino acids was chosen to exclude the first catalytic domain from the
substrate of PCS1-CTVL2-PCS2. To be consistent with this length limitation, this same
length of amino acids was used in the design of the rest of the constructs.
Translation of PCS 1 -CTVL2-PCS2 was expected to produce a full length product
of approximately 64 KDa. Proteolytic processing at both PCSs was expected to produce a
54 KDa protein. Also, it was anticipated that an intermediate protein of 58 KDa
corresponding to partially processed substrates at either PCS might be present.
The SDS-PAGE autoradiogram of PCS 1 -CTVL2-PCS2 translation products is
shown in Figure 2.3C, lane 1. Only one product of approximately 64 KDa was obtained
for this construct. The estimated molecular mass for this protein corresponded to the
predicted full size for PCS 1-CTVL2-PCS2. This result seemed to indicate that there was
no proteolytic activity associated with CTVL2 at either of the PCSs. To rule out the
requirement of a cleavage at the first PCS in order to activate the second PLP domain, the
construct CTVL2-PCS2 (Figure 2.3 C, lane 2) was translated.
A 44 G45 G976 0977 kDa
Molecuar mass 216
113 KDa CTVL1-L2-PCS2 132
59 KDa CTVLI-PCS I 78
54 KDa CTVL1
64 KDa PCSI-CTVL2-PCS2 52
59 KD& CGVL2-PCS2 'A 46
59 KDa PCSI CTVL2 32
54 gDa CTVL2
Figure 2.3: Analysis of proteolytic activity of the N-terminal region of CTV ORFI. (A)
Schematic representation of the expected proteins from the translation constructs used for
in vitro transcription and translation and the possible mature products, intermediates and
expected molecular masses. (B-D) SDS-PAGE autoradiograms of [3H] Leu labeled wheat
germ coupled in vitro transcription and translation products of CTV PCP constructs. (B)
CTVL1. (C) PCSI-CTVL2-PCS2, CTVL2-PCS2 (D) CTVL1-L2-PCS2.
Construct CTVL2-PCS2 represented the beginning of the second PCP, starting
with an additional methionine followed by the glycine that occupies the P1' position at
the first PCS. It also included the second PCS. This construct was intended to simulate
the protein after being cleaved at PCS 1 by CTVL 1. Translation of construct CTVL2-
PCS2 produced a full sized protein of 58 KDa, but not a mature 54 KDa product (Figure
2.3C, lane 2). Again, this second construct, which was designed to demonstrate
proteolytic activity in the CTVL2 protein failed to demonstrate this activity. Even
though proteolytic activity from CTVL2 was not detected, it can not be assumed that this
protein is proteolytically inactive.
Processing Pattern of the N-terminal Region of CTV ORF I Revealed Three Cleavages in
To study the proteolytic activity of the whole N-terminal region of CTV ORF 1 as
a unit, we translated the construct CTVLl-L2-PCS2. This plasmid encodes both
proteolytic domains as well as both cleavage sites. The predicted products from this
experiment are the full size protein of an estimated molecular mass of 113 KDa and, after
full processing at the PCSs, two proteins of undistinguishable molecular mass of
approximately 54 KDa corresponding to CTVL 1 and CTVL2. Also, depending on the
maturation pattern that this region follows, processing intermediates of CTVL2-PCS2
can be expected to be released after a first cleavage at PCS 1 or of CTVL I -L2 after a first
cleavage at PCS2 (see Figure 2.3A).
The translation products of CTVL 1 -L2-PCS2 are shown in Figure 2.3D. The
banding pattern revealed four distinctive proteins with approximate molecular masses of
120, 60, 57 and 54 KDa. These proteins can be assigned to the full sized unprocessed
protein (120 KDa), proteolytic intermediate CTVL2-PCS2 (60 KDa) and fully processed
CTVL 1 and CTVL2 (57 KDa). The fourth protein of 52 KDa that appears in the gel is the
predominant product in the translation mixture, and its size can not be related to any of
the predicted proteolytic products.
Sensitivity to Protease Inhibitors
Protease inhibitors have been used in the characterization of the enzymatic
activity of proteases. To test the sensitivity ofCTVL I and CTVL2 to chemical
inhibitors, we performed in vitro transcription and translation assays in the presence of
cysteine protease specific inhibitors and a broad spectrum protease inhibitor cocktail.
As shown in Figure 2.4, the presence of the cysteine protease inhibitors E-64
(Figure 2.4) or NEM (data not shown) did not affect any of the proteolytic reactions at
the concentrations tested. No effect of the protease inhibitor cocktail was observed as
well. The decrease in transcription-translation efficiency in the presence of protease
inhibitors reflects their interference with the translation system.
Figure 2.4: Effect of chemical protease inhibitors on the proteolytic activity of CTVL 1
and CTVL2. In vitro transcription and translation of CTVL 1 -PCS 1 and CTVL 1 -L2-
PCS2 in the presence of a broad spectrum protease inhibitor cocktail (1 mM AEBSF,
0.8ViM aproptinin, 20 V.M leupeptin, 36 [iM bestatin, 15 VM pepstatin A and 14 vM E-64)
or the cysteine protease specific inhibitor E-64 (100 pM). The presence of the protease
inhibitors did not change the processing pattern of these constructs, but it did affect the
transcription-translation efficiency of the system.
Sequence analysis has revealed the presence of PLPs in different members of the
family Closteroviridae (Agranovsky et al., 1994; Karasev et al., 1995; Klaassen et.. al.,
1995; Jelkmann et al., 1997 and Zhu et al., 1998). For all PLPs reported for
Closteroviridae, L-pro from BYV is the only case for which the PCP has been
experimentally tested and its proteolytic release from the polyprotein demonstrated in an
in vitro transcription and translation assay (Agranovsky et al., 1995).
The data presented here corroborates the catalytic activity of the N terminal
region of CTV ORF1. Fully processed products of CTVLI-L2-PCS2 showed at least
four major processed proteins, accounting for three cleavages within this region. There
were also several minor proteins that were present to some extent in all the translation
reactions. The origin of these minor proteins has not been determined. N-terminal
sequencing of those products could help to elucidate if they are produced by proteolysis
or not. Alternatively, these minor proteins may not be proteolytic products but rather be
artifacts of the system where early termination and internal initiation is known to occur.
The same concern was reported during in vitro translation assays of the equine arteritis
virus papain protease (Snijder et al., 1992), where similar minor proteins were
disregarded as artifacts of the system since these proteins also occurred in the presence of
protease inhibitors. In our case, the presence and quantity of these bands also varied
accordingly to the age of the DNA used in the translation reaction (data not shown).
The proteolytic activity for CTVL 1 was demonstrated by in vitro transcription
and translation. For this protein a mature product appeared to be produced by proteolysis
at the predicted cleavage site G4-G"'5, based on the expected molecular masses for the
products present in the reaction. Based on these results, autoproteolytic release of
CTVL I from the polyprotein is comparable to that reported for BYV L-pro (Agranovsky
et al., 1994). So far we have expressed CTVL2 in the presence of a PCS at both N- and
C- termini, and we removed the N-terminal PCS to simulate the cleavage of PCS I by
CTVL 1. In neither case was proteolytic activity observed. However, the lack of activity
of CTVL2 in our system does not necessarily rule out catalytic capacity for this protein.
The first explanation could be that CTVL2 might not be functional at either of the
predicted cleavage sites. Alternatively, other factors might be necessary to modulate the
activity of this protein. It has been shown that activity of PCPs can be greatly influenced
by the length of the substrates and of the protein itself (Teng et al., 1999).
There is a similar report of inactivity of a second PLP in the murine coronavirus
mouse hepatitis virus (MHV) ORF 1. In this case, the lack of activity of the MHV PLP-2
protease was due to the experimental design which was testing for activity on a substrate
described for the other protease in the viral genome. It was not until the substrate for this
protein was identified hundreds of amino acids downstream of the PLP2 domain (Schiller
et al., 1998) that the activity of this protein was finally characterized (Kanjanahaluethai
and Baker, 2000). A similar explanation may pertain for CTVL2 in our experiments.
The substrates presented to this protease were only the predicted cleavage sites G484-G485
and G976-G977. An alternative possibility could be that the substrate for this enzyme is not
any of the PCSs predicted by Karasev et al. (1995) but an unreported site elsewhere in
the polyprotein, or possibly in any other protein of viral or cellular origin.
The translation of CTVL 1 -L2-PCS2 evidenced at least three cleavages within the
area expressed. From these data, we can not determine if all three cleavages were
mediated by CTVL 1, or if there was participation of CTVL2 in the maturation process.
The molecular mass (52 KDa) of the principal product in the translation reaction
did not correspond to the products predicted after cleavage at both PCSs. Thus the
appearance of this protein may result from cleavage at a previously unreported site. The
presence of a duplicated PCP has been reported for other (+) ssRNA viruses besides
closteroviruses (Lee et al., 1991; Shapira and Nuss, 1991; Godeney et al., 1993).
Although it seems to be a common phenomenon, there is not yet an explanation of a
selective advantage for this duplication (Garcia et al., 1999). Another common finding is
the presence of more than one catalytic type of protease encoded in the same viral
genome, performing sequential cleavages based on different affinities for different
substrates that might be modulated by accessibility and local conformation (Bergmann &
James, 1999; de Groot et al., 1990). The present study was only on the proteolytic
activity within the N-terminal region of CTV ORF 1. There is evidence that the
polyprotein encoded by closterovirus ORF I is processed in a pattern similar to that of the
precursor of nonstructural proteins of alphaviruses and I a/ lb polyprotein of corona-like
viruses (Erokhina et al., 2000). The presence of methyl transferase and helicase domains
as single proteins released from the polyprotein was demonstrated by the use of
monoclonal antibodies in BYV infected tissue. This indicated that closterovirus ORFI
protein undergoes multiple proteolysis in vivo (Erokhina et al., 2000). The RNA
dependant RNA polymerase of CTV has also been demonstrated to be released from the
polyprotein. Using antibodies, Cevik (personal communication), was able to locate CTV
RdRp in cell fractions of CTV-infected tissue. The nature of the proteins that mediate the
maturation of the C-terminal region of closterovirus ORF l a/b have not been identified,
but it has been suggested that this processing might be mediated by a trans activity of L-
pro (or L-pro-like) or by a host enzyme (Erokhina et al., 2000). The results obtained with
the protease inhibitor study revealed that both CTVL 1 and CTVL2 were insensitive to
the inhibitors tested under the conditions present in our system. It has been reported that
protease inhibitors are capable of blocking the translation reaction (Pieroni et al., 1997),
and several investigations took advantage of post-translational processing or trans
proteolytic activity assays to perform these studies (Hahm et al., 1995; Pieroni et al.,
1997; Sircar et al., 1998; Hata et al., 2000). For CTVLI and CTVL2, it seems to be a co-
translational process, and even though we observed decreased concentration of the
translation products in the presence of protease inhibitors, we were able to observe
mature translation products in assays with two different concentrations of the E-64
inhibitor. The insensitivity of some viral cysteine proteases to the E-64 inhibitor has
been reported previously (Tihanyi et al., 1993; Sircar et. al., 1998; Andr6s et al., 2001).
A study of the activity of the African swine fever virus protease revealed its insensitivity
to E-64, but it was susceptible to the cysteine protease inhibitor NEM (Andr6s et al.,
2001). In the case of CTVL 1 and CTVL2, both were insensitive to E-64 and NEM or to
any of the protease inhibitors present in the cocktail used in our experiment. This
insensitivity to protease inhibitors may add to the characteristic properties of these viral
SITE DIRECTED MUTAGENESIS OF PUTATIVE ACTIVE AND CLEAVAGE SITES
OF CTVL 1 AND CTVL2: THE N-TERMINAL PROCESSING DYNAMICS.
Catalysis by the cysteine proteases proceeds via the formation of a covalent
intermediate and involves a cysteine and a histidine residue (Polgar and Halasz, 1982).
Usually, for papain-like cysteine proteases, the catalytic cysteine is flanked at the C-
terminal site by tryptophan, a bulky hydrophobic amino acid (Gorbalenya et al., 1991).
A comparison between cellular and viral cysteine proteases showed that the
spacing between the two catalytic residues differs considerably. For cellular papain-like
proteases, the space between these amino acids is approximately 130 to 160 residues,
contrasting with the viral proteases having approximately 60 80 residues (Snijder et al.,
1992). These characteristics and other properties derived from sequence alignments of
cellular and viral proteases have allowed the identification of new putative papain-related
thiol proteases encoded by the genomes of the positive-strand RNA viruses (Gorbalenya
et al., 1991).
In the identification of the closterovirus BYV L-pro cysteine protease, sequence
alignments revealed the JxxxxGOCYU motif (J= aromatic, X= any amino acid, 0= bulky
aliphatic or aromatic, U= bulky) to be the consensus sequence for the closterovirus and
potyvirus families (Agranovsky et al., 1994). The consensus sequence BGxCYUxH was
reported for the Closteroviridae family (Karasev et al., 1995). From this analysis, the
most conserved region around the catalytic cysteine was the dipeptide CW(Y), being
characteristic of these kinds of proteases (Gorbalenya et al., 1991). The other amino acid
in the catalytic dyad is histidine, and the sequence conservation around it is lower than
that found around the catalytic cysteine, making predictions for this amino acid position
more difficult (Gorbalenya et al., 199 1; Agranovsky et al., 1994).
Protein alignments for closterovirus sequences had identified the presence of a
PCP domain in the N-terminal region of BYV ORF 1 translation product, identified by its
similarity to those PLP from potyviruses. Site directed mutagenesis of C509 and H596
completely abolished proteolytic activity of BYV-L-pro demonstrating the involvement
of these residues in the proteolytic reaction (Agranovsky et al., 1994). Sequence analysis
of the CTV genome revealed two PCP domains encoded by ORF I that represent a gene
duplication when compared with the genome of BYV (Karasev et al., 1995). From the
sequence alignments with BYV L-pro, C4"3 H4" were predicted to be the active amino
acids for the first proteolytic domain and C896 H956 for the second domain in the CTV
In the description of the interaction between a protease and its substrate, the
conventional nomenclature refers to the protease subsites as "S" and the substrate amino
acids as "P". The amino acids of the N-terminal side of the scissile bond are numbered
P3, P2, P1 and those of the C-terminal side are numbered PI', P2', P3'..., where the bond
between PI-Pl' is the scissile bond. Similarly, the protease subsites that complement the
substrate binding site are numbered S3, S2, S1, S l', S2', S3' (Schechter and Berger, 1967).
Generally, leader proteases have been reported to cleave between two small amino
acid residues (Carrington et al., 1989; Gorbalenya et al., 1991; Kirchweger et al., 1994).
With the exception of the conserved dipeptide CW(Y), alignment of viral cysteine
proteases revealed little conservation (Gorbalenya et al., 1991). The amino acid
conservation around the putative cleavage sites of closterovirus cysteine proteases is
limited to the P2, P 1 and P4' positions, where a bulky hydrophobic residue, a Gly residue
and a negatively charged residue are usually found (Jelkmann et al., 1997; Zhu et al.,
Studies on the putative substrate for BYV Lpro, revealed that mutations at the
predicted P 1 position G588 abolished the proteolytic processing of the N-terminal region
of BYV (Agranovsky et al., 1994). By comparison with the BYV sequence, two putative
cleavage sites were predicted at G484-G485 and G976-G 977 for the CTV ORF I protein
(Karasev et al., 1995).
In this study, site directed mutagenesis was performed to both putative active site
amino acids for each proteolytic domain, as well as for the amino acid that occupies the
P1 position at each putative cleavage site. These constructs were used in an in vitro
transcription and translation assay to determine the effects of the mutations on the
proteolytic processing observed for the N-terminal region of CTV ORF 1.
Materials and Methods
The CTV strain used and the general RNA reverse transcription and DNA
amplification techniques were as described in Chapter 2.
Site Directed Mutagenesis
Site directed mutagenesis (SDM) was performed using a PCR amplification
technique. Two partially overlapping complementary oligonucleotides carrying the
desired mutation were used to prime individual amplifications with the corresponding 5'
or 3' oligonucleotide pair. Overlapping PCR mutated products were mixed and
reamplified using the external primers to generate the complete coding sequence for the
desired protein. The primers used to introduce the mutations in the putative active sites
are shown in Table 3.1.
Table 3.1: Oligonucleotide primer sequences used to introduce mutations to the putative
active sites. The mutagenized codons are shown in bold italic letters.
SDM C403S+ 5'-CGGTCAGA GC'ATGTCCGTCACGTGTTC-3'
SDM C403S- 5'-CGGACATAGCTCTGACCGTCGCGAACTTTAGC-3'
SDM C896S+ 5'-ATCCCTGAAGGAAGGATATAGCTACATTCGC-3'
SDM C896S- 5'-GCGAATGTAGCTATATCCTTCCTTCAGGGAT-3'
SDM H646E+ 5'-GGTAGTGTTTTTGAATGCTTGTCA-3'
SDM H646E- 5'-TGACAAGCATTCAAAAACACTACC -3'
SDM H956E+ 5'-CCACGCTGCTTCGAATGTTGCTAC-3'
SDM H956E- 5'-GTAGCAACATTCGAAGCAGCGTGG-3
Several mutations were introduced in the P 1 position of both PCSs. SDM was
performed using a common antisense oligonucleotide that did not carry any mutation and
had an overlap with conserved regions of the mutagenic sense primers. These primers are
shown in Table 3.2.
Table 3.2: Oligonucleotide primer sequences used to introduce mutations into the
putative cleavage sites.
5 '-AACCATCATTTAGTTA TGGGGACTTCT-3'
In vitro Transcription and Translation
Mutagenized proteins were used in an in vitro transcription and translation assay
as described in Chapter 2.
Cysteine and Histidine are Involved in CTVL 1 Activity
Using an in vitro transcription and translation assay, the autoproteolytic activity of
CTVL 1 over the PCS 1 in the CTVL 1 -PCS I construct was demonstrated (Chapter 2).
This construct produced two proteins one "precursor" with molecular mass of
approximately 59 KDa and a mature protein of 54 KDa (Figure 3. 1B).
To identify the amino acids involved in the proteolytic activity of CTVL 1, we
performed SDM of the predicted catalytic residues C'3 and H', producing constructs
carrying individual mutations at either amino acid. Translation from CTVLl c403A-PCS 1
(Figure 3.1B) as well as CTVLl H403E-PCS1 (data not shown) produced a unique band with
a molecular mass corresponding to that of the precursor or unprocessed protein. Even
though we can not unequivocally conclude that C43 and H4' are the catalytic amino
acids, the loss of activity in the mutants suggests their relevance with respect to the
catalytic activity of CTVL 1.
CTVL I Might be Responsible for More Than One Cleavage Within the CTV ORF I N-
After demonstrating that mutations at either C"3 or H41 inactivated CTVL 1
proteolytic activity, the same mutations were introduced into the CTVL 1-L2-PCS2
construct to study the overall effect of CTVL 1 on the processing of the CTV ORF I N-
It was demonstrated in Chapter 2 that there were at least three proteolytic
cleavages within the region, producing four distinct products (Figure 3. 1B). Mutations at
either active amino acid of CTVL 1 completely changed the processing pattern observed
for the wild type protein. These constructs produced a pattern of one doublet with
molecular mass of approximately 120 KDa that corresponded to the predicted mass of the
unprocessed protein encoded in these constructs and the intermediate molecule after
PCS2 cleavage. These results demonstrated the activity of CTVL2 on PCS2, releasing it
from the polyprotein, and also showing that CTVL 1 might have a more extensive role in
the proteolytic processing of the CTV ORF I N-terminal region. An alternative
explanation is that CTVL I might initiate a proteolytic activating cascade, and its activity
is required to activate downstream proteolytic events mediated by CTVL2.
CTVL2 is an Active Protease
Previously we were unable to demonstrate any proteolytic activity for CTVL2
when studied with the PCS 1-CTVL2-PCS2 and CTVL2-PCS2 constructs (Chapter 2). To
determine whether CTVL2 has an active role in the proteolytic processing observed in the
CTVL 1 -L2-PCS2 construct, we introduced mutations at each of the two CTVL2 putative
catalytic amino acids in individual clones. As mentioned above, CTVL2 was shown to
perform the proteolytic processing of PCS2, evidenced when CTVL 1 was inactivated. To
determine if the predicted amino acids were the catalytic residues, we performed SDM of
C896 and H956 These substitutions when introduced changed the processing pattern
observed for the translation products when compared to the CTVL1-L2-PCS2 wild type
(Figure 3.1, lanes 4 and 5). Both mutations produced the same pattern, and it differed
from the wild type protein with respect to the absence of the 53KDa protein. This protein
corresponds with the extra protein that was not predicted for the translation of CTVL 1-
L2-PCS based on the presence of two putative cleavage sites. This suggests that CTVL2
is an active protease, which cleaves at both PCS2 and at an unpredicted site. The putative
active site amino acids for CTVL2, C896 and H956, seem to have a key role in the
proteolytic activity of the enzyme.
CTVL I Might be an Additional Substrate for CTVL2
A second activity of CTVL2 was demonstrated by an additional change in the
processing pattern of CTVL 1 -L2-PCS2 when CTVL2 putative active amino acids were
mutated. The substrate for CTVL2 might be within the area of CTVL 1 since its activity
was not detected when it was tested on the PCS 1-CTVL2-PCS2 construct and was
evident only when both proteases were expressed together.
To study the characteristics of the putative cleavage sites, we introduced the
mutation G484S in the P1 position of PCS1 in the CTVLI-PCSI construct. Translation
from this mutant produced a processing pattern similar to the one observed for CTVL I-
PCS1, having an unprocessed protein of 59 KDa and a mature protein of 54 KDa. In
addition to these two proteins, a third protein of 53KDa appeared for the CTVL 1-PCS484S
translation product (Figure 3.2, lane 2). This additional protein co-migrated with the
unpredicted 53KDa band in the CTVL1-L2-PCS2, seemingly present as a product of the
activity of CTVL2. If these two co-migrating bands are the same protein, this result
indicates that a cleavage site for CTVL2 is within the area of CTVL 1. Even though
CTVLI does not normally cleave at this site, it is not surprising to see proteolytic activity
at an alternative cleavage site. When the wild type substrate is modified, especially if
both proteolytic activities are derived from a gene duplication event, they might still have
overlapping activities as they diverge from each other
G"G485 G9176 Gg77
59-5 kDa CTVLI-PCS!
59 kDa CTVLlm-PCSI
113 kDa CTVLI L2-PCS2
I113 kDa CTVLI. L2LPCS2
113 ~ ~ ~ ~ k: 7L-~iPS
- '- 59
Figure 3.1: Site directed mutagenesis of predicted catalytic amino acids of CTVL1 and
CTVL2. (A) Schematic representation of the constructs used in this experiment.
Constructs carrying individual mutations were C403S, C403A, H646E, C896S, C896A,
H956E. (B) SDS-PAGE autoradiograms of [3H]Leu labeled wheat germ coupled in vitro
transcription and translation products of mutagenized CTVL I-L2-PCS2 constructs.
Mutations shown correspond to C403S and C896S.
G464 C485 G976 G977
59 kDa CTVLI-PCSI
59 kDa CTVL I-PCSIl,,,
113 kDa CVL I-L2,-PCS2
Figure 3.2: Site directed mutagenesis of PCS 1. Localization of CTVL2 substrate site. (A)
Schematic representation of the constructs used for in vitro transcription and translation.
(B) SDS-PAGE Autoradiograms of [3H] Leu labeled wheat germ coupled in vitro
transcription and translation products.
Mutations at the P1 Position of PCS1 and PCS2 are not Well Tolerated
Mutations at either PCS produced similar phenotypes to those observed when the
active amino acids were mutated, but they did not completely abolish processing in the
majority of the cases (Figure 3.3). Glycine at the P1 position of each PCS was substituted
by an alanine, serine, methionine, phenylalanine and aspartic acid on the CTVL 1 -L2-
PCS2 construct. Mutation G484A at PCS1 reduced the processing of the whole product,
causing an increase in the unprocessed full size protein when compared to wild type
(Figure 3.3, lane 1). Curiously, an important reduction in the activity of CTVL2 was also
observed. A more drastic effect on the proteolytic processing was caused by the less
conservative mutations G484F (Figure 3.3, lane 2), G484M and, G484E (data not shown).
In these cases, the effect of the mutation resembled the CTVL I active site mutations, as
the processing of the protein was drastically reduced for both CTVL 1 and CTVL2
activities.Mutations of PCS2 had the pattern characteristic of CTVL2 active site mutants,
carrying drastically reduced CTVL2 activity. When compared to wild type, the 59 KDa
band that corresponds to CTVL2-PCS2 is in greater abundance in the mutant, evidencing
the reduced processing at the PCS2 (Figure 3.3, lanes 3-6). These constructs also showed
a reduced CTVL2 activity, suggesting that an efficient cleavage at the PCS2 is required
for activation of CTVL2. Mutations G976M and G976E abolished almost all CTVL2
activity (Figure 3.3, lanes 3 and 4), and G976F (Figure 3.3, lane 5), G9976A and G976S
(data not shown) reduced CTVL2 activity but some 53 KDa product was still evident.
1 2 3 4 5 6 7 kDa
Figure 3.3: SDS-PAGE autoradiogram of CTVL 1-L2-PCS2 proteins mutagenized at the
P 1 position of their putative cleavage sties. Lanes I and 2 are mutations of PCS 1. Lanes
3 through 5 are mutations of PCS2. The following lanes contain results with the
indicated mutations: 1) G484A; 2) G484F; 3) G976M; 4) G976E; 5) G976F; 6) CTVL 1-
L2-PCS I Wild type; 7) CTVL 1-L2896A-PCS2.
Identification of the residues involved in the catalytic activity of the active
residues of BYV L-pro by SDM (Agranovsky et al., 1994) facilitated the prediction of
putative catalytic amino acids for both putative papain-like cysteine proteases of CTV.
This identification was accomplished based upon sequence alignments of ORF I a of both
CTV and BYV (Karasev et al., 1995). Even though the sequence conservation was
limited between these proteases, the catalytic dyads appeared to be Cys3 and His' for
CTVL 1 and Cys89 and His956 for CTVL2. Protein alignments also enabled the prediction
of cleavage sites for both proteolytic domains. However, there was no experimental
evidence for the identity of the active amino acids for both CTV PCP domains.
In this study we introduced mutations for each of the putative catalytic amino
acids for either PCPD in the CTV ORF I N-terminal region. To identify the catalytic
cysteines, we introduced the conservative mutations C403S and C896S. The other
catalytic position was represented by the residues H' and H956 for CTVL 1 and CTVL2,
respectively. These amino acids were substituted with aspartic acid, which is the most
common substitution of non-conserved histidines in the closterovirus genomes
(Agranovsky et al., 1994).
All the substitutions of the putative catalytic amino acids in each protease changed
the processing pattern observed for the wild type protein. This indicates the importance
of each of these amino acids in the proteolytic processing of CTV ORFI N-terminal
region Even though the identity of these amino acids as the catalytic residues cannot be
unequivocally assured, there are several characteristics that support this conclusion.
Sequence alignment with previously identified catalytic amino acids of the
"closely related" BYV-LPro protease and the change in the processing pattern when
these residues were substituted constitute the strongest line of evidence. There are other
characteristics that are shared with several viral cysteine proteases. These characteristics
include an overall size less than 155 amino acids versus the over 200 amino acids for
cellular proteases. The distance between catalytic amino acids is smaller for viral than for
cellular proteases, and the relative placement of the catalytic amino acids is towards the
C-terminal region of the proteolytic domain (Oh and Carrington, 1989; Gorbalenya et al.,
1991). The demonstration of proteolytic activity in the CTV ORF I N-terminal region
confirms the prediction of the presence of at least one proteolytic enzyme encoded by
CTV. Other members of the Closteroviridae family also were predicted to carry PCP
within their genomes. Sequence alignment for little cherry virus (LChV), lettuce
infectious virus (LIV) and sweet potato sunken vein virus (SPSVV) leader proteins did
not reveal sequence conservation apart from the C-terminal PCP domain, even though
there was stronger conservation among PCP of these Closterovirus or Closterovirus-like
genomes when compared to those of BYV and CTV. Another conserved domain also
was found upstream of the PCP for these three viruses. These data agree with the
grouping derived from the RNA polymerase alignments, suggesting that these two
proteins evolved as a single entity (Jelkman et al., 1997).
When CTVL 1 was mutated in the CTVL 1-L2-PCS2 construct, all major cleavages
were absent, leaving only the processing at the PCS2. This result suggests that CTVL2
has proteolytic activity over the PCS2, being revealed only when CTVL 1 was inactivated.
Previously, we did not find any obvious activity of CTVL2 when it was expressed as an
individual protein out of its polyprotein context. A similar case of a duplicated cysteine
protease is found in the genome of the equine arteritis virus (EAV). For this virus,
expression of single protease domains out of the polyprotein context drastically reduced
their activity, evidencing the importance of the conformation adopted when expressed as
part of a larger protein (den Boon et al., 1995). A similar situation was found for the
murine coronavirus MHV, where the PLP- 1 activity seems to be optimal when both
enzyme and substrate are expressed as a unit or as part of a large protein. Protein
sequences at both the N- and the C-terminal positions of the proteolytic domain affected
its catalytic efficiency (Teng et al., 1999). With CTV we may have a similar situation
when CTVL 1 -PCS I is expressed. The catalytic reaction proceeds, but further incubation
of the product after stopping the translation reaction does not result in all the precursor
being processed into mature protein (data not shown), evidencing the inefficiency of the
process. Similarly PCS1-CTVL2-PCS2 did not show any activity when expressed by
itself, but it revealed proteolytic activity when expressed as part of a larger protein.
From these results we can infer a processing pattern in which CTVL 1 and CTVL2
cleave themselves from the polyprotein at PCS 1 and PCS2, respectively. Cleavage at
PCS 1 activates CTVL2 to perform a second cleavage within the area of CTVL 1. Thus in
this system, CTVL 1 co-translationally cleaves itself from the polyprotein and activates a
downstream event. CTVL2 catalyzes at least two proteolytic cleavages; one at its own C-
terminal end on the PCS2 in a CTVL 1 independent fashion and a second cleavage at the
N-terminal region of CTVL 1 and in a CTVL 1 dependent fashion.
Duplicated cysteine proteases have been reported in other viral groups (Lee et al.,
1991; Shapira and Nuss, 1991; Godeney et al. 1993). Arterivirus replicase ORFIa
protein encodes two cysteine protease domains that are located in similar positions to
those of CTV within the viral polyprotein. These PCPDs are designated as nsp 1 c and
nsp 1[3. In this system, nsp Ia releases itself from the polyprotein cleaving the
nsp 1 a/nsp I 3 bond. Nsp I 3 releases itself from the rest of the polyprotein cleaving the
nsp I P/polyprotein bond. The individual cleavages of these two proteins are independent
of each other (den Boon et al., 1995). These results correlate with the autoproteolytic
release of CTVL I and CTVL2 from the polyprotein independent of each other.
A well documented case of cysteine protease duplication is that of murine
coronaviruses. When aligned with that of other coronaviruses and cellular papain
proteases, the sequence of murine hepatitis virus (MHV) revealed the presence of two
putative papain-like protease domains in the viral genome (Lee et al., 1991). When
comparing the position of the active amino acids of the MHV proteases with those of
CTV, we found that the positions of the catalytic amino acids for MHV Prol were C..2'
and H 27, with 157 residues between them, in contrast to the 61 amino acids for CTVL 1.
Similarly, MHVPro2 has 157 residues between its catalytic amino acids while CTVL2
has only 60. Even though there are differences between the lengths of the spacers for
these proteases when compared between the viral groups, the difference is conserved
within the same viral genome, thus being characteristic for the entire duplicated gene.
The separation between the catalytic centers of both PCP is also greater for MHV; 595
residues between the catalytic cysteines compared to the 493 residues for CTV.
One proteolytic activity was reported with the translation of murine coronavirus
gene 1 polyprotein. This protein underwent proteolytic processing releasing a protein of
28 KDa, and that activity was linked to the Cys"3 residue. This activity was detected
only when the whole polyprotein was expressed, revealing the importance of the overall
conformation (Baker et al., 1990). Extensive site directed mutagenesis of the putative
p28 cleavage site and surrounding amino acids revealed the importance of the P1, P2 and
P5 positions in the maturation of this protein. The cleavage site was mapped to the G247-
Val1248 dipeptide, with Gly247 and Arg246 also being major determinants for the recognition
of this cleavage site (Dong and Baker, 1994; Hughes et al., 1995; Dong et al., 1995).
Deletion analysis studies of the p28 processing showed that processing at this cleavage
site was affected by sequences upstream to the PCPD, down regulating its activity. In
addition, some of the deletion constructs revealed a different processing pattern that
coincided with a 65 KDa protein (p65) that was present in virus-infected cells. These
observations suggested the presence of another cleavage site in the region. The catalytic
His1272 was identified, and the activity of the proteolytic domain was demonstrated on
both cleavage sites (Bonilla et al., 1995). The second substrate for the protease, which
released the p65 protein, was characterized and compared to the p28 cleavage site. Both
sequences had a conserved basic amino acid at the P5 position that play an important
processing role, with the cleavage occurring between two small neutral amino acids
(Bonilla et al., 1997).
Based on sequence characteristics of both cleavage sites, Bonilla et al. (1997)
searched the polyprotein sequence for the presence of the consensus cleavage sequence
P5-(R,K)xxx(G, A) I (G, A, V)-PI'. They found another putative cleavage site that would
produce a protein that corresponded with the observed 50 KDa (p50) protein in infected
cells. If that cleavage occurs, it might be a way to regulate the polyprotein processing
since the PCS would interrupt the active site of this protease. We searched for other
putative cleavage sites within the CTVL 1 region, but we did not find any sequence
similar to the PCS 1 or PCS2 positions or to the reported "conserved" closterovirus
cleavage site sequence P2(bulky)P I (G)XXXP4'(negatively charged) reported by Jelkman
et al., (1997). A further option is the possibility of more cleavage sites for CTV that have
not been reported within the polyprotein or other viral/cellular proteins. Studies that
spanned the 400 KDa of MHV ORF 1 a evidenced the presence of other cleavage sites
within the polyprotein of this virus (Schiller et al., 1998). This led to the detection of the
activity of a second putative cysteine protease domain, which had been previously
undetected. PLP2 has a substrate downstream from its physical location in the
polyprotein, and it has different substrate determinants than those of PLP 1, suggesting
that the overall conformation of the protein might affect the efficiency of the processing
that could regulate different activities of the replicase complex (Kanjanahaluethai and
Baker, 2000). Infectious bronchitis virus (IBV) a member of the Coronaviridae family,
also contains two cysteine protease domains. Based on the complete viral sequence
(Boursnell et al., 1987), two papain-like cysteine proteases and a picomavirus 3C-like
protease domain were identified (Gorbalenya et al., 1989). In vitro transcription and
translation of the ORF L a sequence offered the first evidence of proteolytic activity when
an 87 KDa (p87) protein was released from the polyprotein. The molecular mass of this
protein corresponded to that of a protein observed in infected Vero cells (Liu et al., 1995).
This p87 was determined to be the product of the activity of two overlapping papain-like
proteases (Liu et al., 1995). Further characterization of this activity revealed that only the
first proteolytic domain was responsible for the p85 release, and the catalytic amino acids
were mapped to C274 and His437 and the cleavage site to Gly673-Gly674 (Lin and Liu,
1998). When the activity of IBV PLPD- 1 was compared to that of the human coronavirus
229 PLPD and mouse hepatitis virus strains JHM and A59 in terms of the composition of
the catalytic residues and the cleavage specificity, there were similarities between the
proteins, but there also were three major differences. These included no protease activity
for IBV PLPD-1 when expressed in an in vitro system, a different specificity for the
cleavage site recognition at the P5 position where a valine instead of a basic amino acid
was located for the IBV sequence, and the final difference found for the IBV protein was
the lack of trans cleavage for both in vivo and in vitro assays (Lin and Liu, 1998).
Another activity for the IBV protease was the cleavage of a second dipeptide bond
between Gly2265-Gly2266. Both cleavage products were identified in IBV infected cells
(Lim et al., 2000). When the activity of PLP I was studied in human coronavirus 229E
another member of the Coronaviridae family, the activity was similar to that described
for the murine hepatitis virus. MHV cleaves at G141_V241 to form p28, and HCV produces
p9 after a cleavage at Gly"1-Asn"2 reflecting different positions of the cleavage sites. In
contrast with MHV, HCV229E PCP 1 does not mediate more cleavages within the
replicase complex, and it shows trans activity (Herold et al., 1998).
CIS TRANS ACTIVITY OF CTVL 1 AND CTVL2
Cis-proteolysis occurs when the activity is performed in an intramolecular
fashion, whereas trans-activity refers to intermolecular reactions. These and other
characteristics have been used to classify viral cysteine proteases into leader or main
proteases (Gorbalenya et al., 1991; Chen et al., 1996). An initial classification of leader
proteases included those proteases from poty-, bymo- and aphtoviruses. The general
characteristics of these proteases were that they mediated a single cleavage event at their
own C-terminus, being described as accessory proteases. The other group, the main
proteases, encompassed those from a- and rubiviruses. This class of proteases represents
cases in which a single enzyme is capable of performing several or all of the cleavages
during the processing of the polyprotein (Gorbalenya et al., 1991).
Semliki forest virus (Alphavirus, Togaviridae) encodes a main cysteine protease,
nsP2A, which is the only protease needed to process the SFV P1234 polyprotein. The
maturation pathway for this polyprotein is a series of proteolytic processing that includes
both cis only and trans only reactions mediated by nsP2 at specific cleavage sites (Merits
et al., 2001).
Murine hepatitis virus (MHV, Coronaviridae) encodes two papain-like proteases,
PLP-1 and PLP-2. Characterization of the activity of PLP-1 revealed that this protein is
capable of cleaving at different positions of the polyprotein and possesses the ability to
cleave in trans, being compared to the activities presented by the alphavirus nsP2
protease (Bonilla et al., 1997). This characteristics showed strong dependance on the
incubation temperature and the size of the substrates presented to the protease, where low
temperatures and larger substrates resulted in more efficient cleavages, evidencing the
importance of the overall folding of the polyprotein in the process (Teng et al., 1999).
The other protease, PLP-2, encoded in the MHV genome also has been demonstrated to
be a trans-active protease, having its substrate site more than 1000 residues downstream
from its active site (Kanjanahaluethai and Baker, 2000). These two proteases showed
characteristics of main cysteine proteases.
Besides the proteolytic domain, an "X" domain was found, exclusively for the
main proteases (Gorbalenya et al., 1991). This domain was later related with the trans-
proteolytic activity of RUB NS protease (Liang et al., 2000). Other examples of main
proteases include proteases encoded by foot and mouth disease virus (Kirchweger et al.,
1994) and rubella virus (Liu et al., 1998). In all the cases, these enzymes cleave both in
cis and trans and in different sites apart from the catalytic center within the polyprotein.
There are several examples of leader proteases encoded in the genomes of
different viral groups. In the translation of the major large dsRNA of the hypovirulence
associated virus of the chestnut blight fungus it was found that a 29 KDa protein was
released by cotranslational proteolysis. This proteolytic activity was associated with the
p29 protein, and its sequence characteristics had similarities with the potyviral HC pro
(Choi et al., 199 1b) relating it to the leader proteases (Choi et al., 1991 a).
For the arterivirus replicase ORFIa protein there have been described either one
PLP for equine arteritis virus (EAV) or two PLP domains, PCPa and PCP3 for lactate
dehydrogenase elevating virus (LDV) and porcine reproductive and respiratory virus
(PRRSV). In either case, the proteases released themselves from the polyprotein in a cis
autocatalytic reaction and did not show any further processing of the polyprotein (Den
Boon et al., 1995).
Even though there is some variability among the characteristics of the viral
proteases, there are features that are constant among them. Within the order Nidovirales,
the family Artriviridae and Coronaviridae contain both main and accessory proteases.
The accessory proteases share characteristics as the recognition of 1 or 2 cleavage sites,
and they are located at the N-terminal half of the polyprotein. There is at least one small
amino acid at the scissile bond, and there are cysteine and a histidine residues at the
catalytic dyad (Ziebuhr et al., 2000).
Besides the common characteristics of the accessory proteases for this order, there
are specific attributes for each family. The arterivirus accessory proteases have a short
spacer distance between their catalytic residues, which is similar to those found in other
PCPs (Gorbalenya et al., 1991); they are located at an amino terminal position within the
polyprotein; and they cleave downstream of their catalytic domain (Ziebuhr et al., 2000).
Examples of these proteases are found in EAV, LDV and PRRSV (Den Boon et al.,
1991). The coronavirus accessory proteases have a spacer between their catalytic
residues that is almost as twice as long as those of the arteriviruses. The proteases are
separated by at least 1000 residues, which gives them a more central location within the
polyprotein, and they perform proteolytic cleavages upstream of the catalytic domain.
Viruses that encode proteases in this group include murine hepatitis virus (MHV), human
coronavirus (HcoV), transmissible gastroenteritis virus (TGEV) and infectious bronchitis
virus (IBV) (Ziebuhr et al., 2000).
Materials and Methods
Virus Isolate, cDNA Synthesis and Cloning
The CTV strain and general techniques were described in Chapter 1.
Descriptions of the cDNA constructs used to demonstrate cis and trans activity of CTVL 1
and CTVL2 are found in Chapters 2 and 3.
Post Translational Cis/Trans Activity Assays
In vitro transcription and translation was performed as described in Chapter 2.
Constructs used as substrate were translated in the presence of [3H]Leu. Constructs used
to synthesize enzymes were translated in the presence of a complete amino acid mix with
no radio-labeled amino acid present. After the reaction was completed, RNase was added
to the reactions, and equal amounts of enzyme and substrate were mixed and incubated
for an additional hour under the same conditions. The reactions were performed at both
22 and 30 'C. Following the incubation, the samples were analyzed by SDS-PAGE as
Co-Translational Cis/Trans Activity
Enzyme and substrate cDNA clones were co-translated in the same reaction mix.
The reaction conditions were those used for the single translations, except that both
plasmid DNAs that were being tested were added.
CTVL 1 Encodes a Cis Acting Protease Tested in an In vitro Assay
The experimental design that we used to demonstrate trans activity of CTVL 1 and
CTVL2 included the production of inactivated proteases by SDM of an amino acid in the
active site as described in chapter 3. The inactive proteins were then used as substrates to
demonstrate complementation in trans of the proteolytic reactions when incubated in the
presence of an active protease. The in vitro transcription and translation reactions to
produce the substrate proteins were performed in the presence of a radio-labeled amino
acid, whereas those to be tested for enzyme activity were synthesized in the absence of
the radio-labeled amino acid.
We previously demonstrated that the translation product of CTVL 1 -PCS 1 yielded
both full length size protein of 59 KDa and a mature protein of 54 KDa, evidencing the
autoproteolytic cleavage at the PCS I, mediated by CTVL 1. Figure 4.4.1B shows a
complementation test in which we used unlabeled CTVL 1-PCS I as the source of the
proteolytic activity. To evaluate CTVL 1 trans activity at the PCS 1, we tested the
substrate site in the PCS 1 -CTVL2-PCS2 construct. Previously we showed that this
construct was proteolytically inactive at the PCS 1. Therefore, if it showed any
processing, the proteolytic activity must originate from the trans supplemented enzyme.
CTVL 1 failed to cleave the PCS I when tested in trans under these conditions.
Since processing of PCS 1 in the PCS I -CTVL2-PCS2 construct was never
observed, there was no evidence that this site was correctly folded when presented in this
construct. This may be the reason for the trans cleavage failure. To evaluate this
possibility we used the CTVL lm-PCS I construct, which basically is the inactivated
version of the protease that was in the assay. The result of this experiment also did not
indicate trans proteolytic activity of CTVL 1 (Figure 4.1 B, lanes 5 and 6).
Additional tests of trans proteolytic activity for CTVL 1 used the CTVL I m-L2-
PCS2 translation product as a substrate. This protein represents approximately the first
110 KDa of the CTV ORFI N-terminal region, including both proteolytic domains and
putative cleavage sites for these proteases. Trans complementation with the CTVL 1-PCS
translation product did not change the molecular mass of the protein as shown in Figure
4. 1D, lane 1. To confirm that the unlabeled enzyme was being translated, an aliquot of
the unlabeled translation reaction was incubated in the presence of [3H]Leu. The result is
shown in Figure 4.1 D, lane 3, corroborating the synthesis of the active protease.
CTVL 1 -L2-PCS2 did not Complement the Proteolytic Reaction in a Post-Translational
The translation product of CTVL I -L2-PCS2 was shown in Chapter 3 to have
proteolytic activity and caused at least 3 cleavages within the region. To test if any of the
remaining mature proteins had trans proteolytic activity we tested these enzymes with all
the available constructs that carried any of the substrate sites.
Figure 4.2 shows the results of incubations with the PCS I -CTVL2-PCS2 as
enzyme and, CTVL I-PCS 1, CTVL I m-PCS I and CTVL I m-L2-PCS2 as substrates. In
none of the cases tested did incubation with the unlabeled proteolytic active product
change the electrophoretic pattern observed for both substrates.
G46 G4076 0937
2 3 4 5 6
Crvu I. jj 59-54- kfl
~1CrVLI PCSI S SkD. 7
NUMBER ENZ*AME SUBSTRATE
2 CTVLI-PCSI PCSI-CTVLI-PCS2
4 CTVLI.PCS1 CTVLI-PCSI 32
6 CTVLI PCSI CTVLI-PCSI
1 2 3 4 kDa
G4 6 5 006 9G
CTVLI PCS1 39 kD.
CTVLI L2"PCS2 113 kDa
NUMBER CNZYM B rSA1Z
I CTVLI PCS1
2 CTVLI.-L2-PCS2 -46
3 CTVL I -PCS I
4 CIVLI-PCSI CTVL -L2- PCS2 32
Figure 4.1: CTVL I-PCS 1 trans complementation of proteolytic activity. (A) Schematic
representation of the proteins used in the trans-complementation assay shown in panel B.
The putative cleavage sites are indicated. Mutated CTVL 1 -PCS 1 carried either C403A or
C403S substitutions to inactivate its auto-proteolytic activity and was used as a substrate
for the trans reaction. Radio-labeled substrates were incubated for one hour with
unlabeled active enzyme and then subjected to electrophoresis. (B) SDS-PAGE
autoradiogram of in vitro transcription and translation [3H]Leu labeled products as
described for panel A. (C) Schematic representation of CTVL 1-PCS I trans proteolytic
activity assay. Mutated CTVLl-CTVL2-PCS2 carrying either C403A or C403S
substitutions was used as a substrate for trans proteolytic complementation of CTVL 1.
Putative cleavage sites and expected molecular mass of unprocessed products are
indicated. (D) SDS-PAGE Autoradiogram of in vitro transcription and translation
[3H]Leu labeled products as described for panel C. Lane 1 and 2 show the enzyme and
substrate used in the assay, respectively. Lane 3 after mixing the enzyme and the
substrate, unlabeled enzyme was labeled with [3H]Leu to corroborate its translation.
Lane 4 shows the complementation assay as described for panel C.
04 1 2 a 4 5 6 7 8
CTVLI-PCSI 5"- ~ 78
CTVLI.mPCS1 S9 kD.4
PCSI '1- -PCS2 CSll2 A 63 kD.
-YTL -:,-CTVL2 PCS2 I 11 kD.
Figure 4.2: CTVL 1 -L2-PCS2 trans proteolytic activity. (A) Schematic representation of
the translation products of the constructs used in the in vitro transcription and translation
assay. (B)SDS-PAGE autoradiogram of the trans proteolytic complementation assay.
Lane 1 and 2 PCS 1-CTVL2-PCS2 with and without the CTVL1-L2LPCS translation
product. Lane 3 5, CTVLI-PCS1, CTVLlm-PCS1 and CTVLlm-L2-PCS2. Lane 6 8
same proteins as in 3 5 supplemented with unlabeled CTVL 1 -L2-PCS2.
This suggests that trans proteolytic activity is not associated with the mature
translation products from the CTVL 1 -L2-PCS2 cDNA construct. So far, we have
evaluated the ability of both CTVL 1 and CTVL2 to perform trans proteolytic reactions
over the known cleavage sites. The enzymes were prepared from different constructs in
all the combinations where we had observed proteolytic activity. Substrates were
presented under conditions that simulated the native conformation or where they were
previously cleaved when the proteolytic domain was active. Although no other changes
were introduced in the substrate proteins apart from the inactivation of the catalytic
domain, we cannot rule out a conformational change that could affect the folding and
hence the recognition of the cleavage site in the trans reaction.
CTVL2 may have Trans Activity as Shown by a Co-Translational Trans Proteolytic
To test trans proteolytic activity we used translation products that we knew were
proteolytically active because they had experienced auto-proteolytic maturation. Thus,
we had done experiments where both protease and substrates were independently
translated and then mixed to assess trans proteolysis in a post-translational fashion. This
approach was a convenient way to study the system because it provided the possibility of
having only the substrate labeled, making the analysis of the results more forthright. The
problem with this approach is that even though it was a clean way to observe the results,
it did not provide any evidence that the proteolytic domains remained active after being
processed, and that might have introduced an artifact into our results.
To further test the absence of trans proteolytic activity of CTVL 1 and CTVL2 we
performed a trans co-translational assay where both enzyme and substrate constructs
were translated together in the same tube, and the patterns observed were compared to
those of the individual constructs while trying to differentiate between superimposition of
patterns or patterns produced by the interaction between the proteins translated from both
constructs. The results of these experiments can be observed in Figure 4.3. Panel A
shows the electrophoretic pattern that each individual cDNA translation product produced
and which has been reported in previous chapters. Lanes I and 2 evaluated the trans-
activity of CTVL 1 when presented from the CTVL 1 -PCS I construct on the CTVL I m-L2-
PCS2 and CTVL 1-L2m-PCS2 substrates. In neither case did we see a different pattern
from that of the superimposition of both translation products, indicating no trans-
proteolytic reaction in the system.
Lane 3 and 4 evaluated the trans-activity for both proteases when presented in the
CTVL 1-L2-PCS2 construct. To discriminate between any possible trans-activity of
either protease contained in this construct, we used as a substrate the CTVL1-L2-PCS2
constructs with mutations in either CTVL 1 or CTVL2. Trans-activity in this case would
be evidenced by the complementation of the particular mutation, producing the expected
wild type electrophoretic pattern. Under these conditions, Figure 4.3B lane 3 shows the
complementation experiment to evaluate the trans activity of CTVL 1. This experiment,
as with the previous ones done using CTVL 1 as the enzyme, did not show any evidence
of trans activity, indicating that this protease is probably a cis acting protein only.
The use of the CTVL 1-L2m-PCS2 construct allowed us to evaluate the trans
activity of CTVL2 when supplemented as part of the CTVL1-L2-PCS2 proteolytic
construct. Under these conditions we were able to obtain some evidence suggesting that
CTVL2 might be a trans-acting protein at the unpredicted cleavage site. Even though the
experimental design did not allow us to discriminate between the bands that corresponded
to each of the cDNA translation products, the major processing product corresponds to
that of the pattern produced when both proteases are active in the construct (Figure 4.3B
lane 4). Based on this result, it seems that the pattern observed is different from the result
of superimposition of the individual translation products for each construct in the
reaction. When this result is compared with that of the other experiment where we used
the same substrate but complementation did not occur (Figure 4.3B lane 2), we can see
that the typical pattern produced by the inactivation of CTVL2 is reduced for the trans-
Enzyme: CTVL1-PCS 1
1 2 3 4 kDa
m m --- 52
Figure 4.3: Assessment of cis-trans activity of CTVLI and CTVL2. (A) SDS-PAGE
autoradiograms of [3H]Leu labeled wheat germ coupled in vitro transcription and
translation reactions of cDNA constructs used to test cis-trans activities of CTVL 1 and
CTVL2. (B) Co-translation of enzyme substrate cDNA constructs described in part A.
(C) Schematic representation of the cDNA constructs used in these experiment.
G44 G,465 G976 Gg77
4 + l 4llllllllll
The purpose of this study was to determine the possible cis-trans characteristics of
both CTVL1 and CTVL2. The nature of the system allowed a simpler experimental
design to test for CTVL I activity, since it seemed to be independent of other proteolytic
reactions within the polyprotein. For CTVL2 we have already determined that it requires
CTVL 1 activity to trigger a second cleavage within the CTVL 1 region. Therefore, in
order to be able to evaluate this activity, both proteins had to be expressed together.
To assess these characteristics for either protease we tried to study them as
individual proteins (CTVL 1) or as a region with different proteolytic activities (CTVLI-
L2-PCS2). Trans-activity of CTVL I was evaluated in different ways by attempting to
simulate the conditions where the cleavage occurs at PCS 1. As enzyme sources we used
constructs for which we had previously observed proteolytic activity (CTVL1-PCS1 and
CTVL1-L2-PCS2). The substrates used in the assay included PCS 1-CTVL2-PCS2,
CTVL I m-PCS I and CTVL 1 m-CTVL2-PCS2. Complementation of CTVL 1 proteolytic
activity on the PCS 1 was not observed for any of the combinations tested. This may
indicate that CTVL 1 is a cis-acting protease which releases itself from the polyprotein, a
characteristic of all the described leader proteases.
Even though we have not been able to demonstrate trans-activity for CTVL 1,
there are many factors that could have influenced the result. In the characterization of the
trans-activity of a papain-like protease of the marine coronavirus MHV, it was found that
the PLP- 1 protein that had been previously described as cis acting only (Baker et al.,
1989) was able to act in trans as well (Bonilla et al., 1997). The newly described trans-
activity for this protease was evidenced by the expression of deletion constructs of the
substrate. This suggested that the observed trans activity was due to the increased
availability of the substrate to the trans activity of PLP- 1 protease, rather than the product
of altered catalytic properties of the enzyme (Bonilla et al., 1997).
More evidence of the importance of protein folding in the trans activity of PLP-1
was obtained though a series of experiments performed at different temperatures. Trans-
proteolysis at two cleavage sites was significatively more efficient when performed at
22C rather than 30 'C. Further observation revealed a more efficient cleavage at 16 'C,
and reactions performed at 37 'C were even less efficient than those performed at 30 'C.
It has been suggested that this temperature dependence for trans-cleavage could be a
requirement for a specific folding of the polyprotein for the recognition of the trans-
cleavage sites, and that it is achieved only at low temperatures in an in vitro assay (Teng
et al., 1999). To test temperature dependence in our system, we performed the trans
complementation reactions at both 22 'C and 30 'C. However, we did not observe any
differences under these temperature conditions, with the CTVL 1 reaction being cis-acting
only under both temperature conditions tested.
PLPD- I from the coronavirus avian infectious bronchitis virus (IBV) has been
demonstrated to be a cis-acting only protease when it is expressed in vitro in rabbit
reticulocyte lysates. When the protein is expressed in intact cells, it can act in trans on
the same substrate (Lim et al., 2000). The trans-activity of CTVL 1 cannot be properly
evaluated in vivo due to the unavailability of an infectious cDNA clone in our laboratory,
which would allow us to perform in vivo complementation assays for these reactions.
Another factor which needs to be taken into consideration when analyzing the cis-
trans proteolytic activity of these proteases is that they are being expressed in the N-
terminus of the polyprotein in an in vitro translation system. The length of the substrate
was demonstrated to be important for the trans-activity of MHV-A59 PLP-1. When
different sizes of the substrate were tested, it was found that no proteolytic activity was
detected with substrates smaller than 301 residues, and that approximately 44% of the
protein was processed when the 622 residue size was achieved, reaching maximum
processing at the 867 amino acid length (Teng et al., 1999).
Another example which demonstrates the importance of the substrate
conformation in a trans-proteolytic reaction, is the rubella virus (RV) protease. Co-
expression of a construct of approximately 2000 amino acids which included the
proteolytic region bearing mutations at the cleavage site to serve as a protease and a
construct with mutations at the active site to serve as a substrate did not reveal trans-
cleavage for this protein (Chen et al., 1996). However, when the whole ORFI of RV was
expressed, it was demonstrated that the cysteine protease encoded within it was capable
of performing proteolytic cleavages both in cis and in trans, suggesting that either
conformational changes of smaller constructs or the requirement for an intact ORF I
translation product could explain previous failures to detect trans proteolysis (Yao et al.,
Our experimental system only represents an in vitro translation of 1016 N-
terminal amino acids of the possible 3124 residues encoded in CTV ORF l a. To date we
do not know about other possible cleavage sites within the polyprotein or the effects of
expressing these proteins out of the polyprotein context. There is some evidence that the
polyprotein might undergo further proteolytic processing in vivo, due to the presence of
free helicase and methyl transferase domains in BYV infected tissues (Erokhina et al.,
2000) and free RdRp in CTV infected tissue (1evik, personal communication). The
origin of the protease activities responsible for these cleavages remains to be determined.
With our current information on CTVL 1 and CTVL2 we can not reach any conclusion
about the involvement of these proteins in those proteolytic reactions. Other proteases
have been described whose characteristic do not agree completely with the description of
the leader proteases. For blueberry scorch carlavirus and turnip yellow mosaic virus, the
cysteine proteases involved in the processing of the polyproteins are in both cases cis-
acting only enzymes as the leader proteases, but their substrates are located
approximately 400 amino acids downstream from the catalytic dyad (Bransom and
Dreher, 1994; Lawrence et al., 1995), contrasting with the typical 40 residues found for L
proteases (Gorbalenya et al., 199 1). CTV proteases have been predicted to be a
duplicated papain-like leader protease similar to L-pro from BYV (Karasev, et al., 1995).
The results of the co-translational trans proteolytic assay did not reveal any evidence of
trans activity for CTVL 1. The results observed in the complementation of CTVL2
activity suggest that this protease its able to cleave its substrate in trans. If CTVL2 its
confirmed to be a trans active protein, then its classification as a leader protease should
SUMMARY AND CONCLUSIONS
The N-terminal region of citrus tristeza closterovirus ORF1 strain T2K was cloned
and used for in vitro transcription and translation assays to determine the mechanism of
proteolytic processing in the region. From the results obtained, there are at least three
proteolytic cleavages in the region, under the conditions studied. These proteolytic
activities were insensitive to a broad spectrum protease inhibitor cocktail as well as to the
specific cysteine protease inhibitors E64 and N-ethylmaleimide.
The putative catalytic amino acids C"3- H464 and C896- H956 were experimentally
confirmed by site directed mutagenesis as the active residues of CTVL1 and CTVL2,
respectively. The importance of the amino acid that occupies the P1 position in both
putative cleavage sites was established as well, being determined that changes in this
position were not well tolerated by the system, resulting in inhibition or reduction of the
proteolytic processing of the area.
CTVL1 cotranslationally cleaved itself from the rest of the polyprotein at the
PCS 1. Cis-trans complementation assays failed to demonstrate trans activity for this
protein under the conditions tested. These results jointly with the sequence characteristics
such as the relative position of the active amino acids between each other and within the
proteolytic domain, as well as the position of the cleavage site for this protease, suggested
the CTVL1 might belong to the leader proteases group. CTVL2 was able to
cotranslationally release itself from the polyprotein cleaving at the PCS2. This activity
was dependent on the length of the protein since it was not observed until the CTVL1-L2-
PCS2 was expressed as a unit. Also it was demonstrated that this cleavage was
independent of the CTVL1 activity since it also occurred when the first proteolytic
domain was inactivated.
A second activity was described for CTVL2, hydrolysis of a cleavage site within
the area of CTVL1. This activity was CTVLl-dependant since inactivation of this protein
eliminated the second cleavage performed by CTVL2. From cis-trans complementation
assays, co-translation of the substrate and the enzyme constructs suggested that this
second activity of CTVL2 occurs in trans. This latest characteristic does not agree with
the definition of a typical leader protease.
With these data we developed a working model in which CTVL1 and CTVL2 co-
translationally release themselves from the polyprotein in a cis acting proteolytic function.
The release of both proteases allow CTVL2 to act in trans in a region within the area of
CTVL1. Sequencing of the proteolytic products is necessary to confirm the exact location
of this cleavage site, and its biological significance remains to be determined. Processed
proteins may perform other proteolytic or non-proteolytic activities in the cell during the
course of the viral infection.
The biological significance of these studies performed in vitro remains to be
confirmed in an in vivo system using an infectious clone of CTV.
Agranovsky, A.A. (1996) Principles of molecular organization, expression, and evolution
of closteroviruses: over the barriers. Adv Virus Res 47, 119-58.
Agranovsky, A.A., Koenig, R., Maiss, E., Boyko, V.P., Casper, R. and Atabekov, J.G.
(1994) Expression of the beet yellows closterovirus capsid protein and p24, a capsid
protein homologue, in vitro and in vivo. J Gen Virol 75(Pt 6), 1431-9.
Albiach-Marti, M.R., Mawassi, M., Gowda, S., Satyanarayana, T., Hilf, M.E., Shanker,
S., Almira, E.C., Vives, M.C., Lopez, C., Guerri, J., Flores, R., Moreno, P., Gamsey,
S.M. and Dawson, W.O. (2000) Sequences of Citrus tristeza virus separated in time and
space are essentially identical. J Virol 74(15), 6856-65.
Andres, G., Alejo, A., Simon-Mateo, C. and Salas, M.L. (2001) African swine fever virus
protease, a new viral member of the SUMO- 1- specific protease family. J Biol Chem
Ayllon, M.A., Rubio, L., Moya, A., Guerri, J. and Moreno, P. (1999) The haplotype
distribution of two genes of citrus tristeza virus is altered after host change or aphid
transmission. Virology 255(1), 32-9.
Babe, L.M. and Craik, C.S. (1997) Viral proteases: evolution of diverse structural motifs
to optimize function. Cell 91(4), 427-30.
Baker, S.C., La Monica, N., Shieh, C.K. and Lai, M.M. (1990) Murine coronavirus gene
1 polyprotein contains an autoproteolytic activity. Adv Exp Med Biol 276, 283-9.
Baker, S.C., Shieh, C.K., Soe, L.H., Chang, M.F., Vannier, D.M. and Lai, M.M. (1989)
Identification of a domain required for autoproteolytic cleavage of murine coronavirus
gene A polyprotein. J Virol 63(9), 3693-9.
Bar-Joseph, M., Gumpf, D. J., Dodds, J. A., Rosner, J. A., and Ginzberg, I. (1985) A
simple purification method for citrus tristeza virus after prolonged lag period in Israel.
Phytopathology 68, 1110-1111.
Bar-Joseph, M., and Lee, R. F. (1989) Citrus tristeza virus. AAB Description of plant
viruses, No. 353 (No. 33 revised). Commonwealth Mycol. Inst./Assoc. Apple. Biol. Kew,
Surrey. 7 pp.
Bar-Joseph, M., Marcus, R., and Lee, R. F. (1989) The continuos challenge of citrus
tristeza virus control. Annu. Rev. Phytopathol. 27, 291-316.
Bergmann, E. M., and James, M. N. (1999) Proteolytic enzymes of the viruses of the
family Picornaviridae. Proteases of infectious agents (Dunn, B. Editor). Academic Press,
San Diego. pp. 282.
Bonilla, P.J., Pinon, J.L., Hughes, S. and Weiss, S.R. (1995) Characterization of the
leader papain-like protease of MHV-A59. Adv Exp Med Biol 380, 423-30.
Bonilla, P.J., Hughes, S.A. and Weiss, S.R. (1997) Characterization of a second cleavage
site and demonstration of activity in trans by the papain-like proteinase of the murine
coronavirus mouse hepatitis virus strain A59. J Virol 71(2), 900-9.
Bonner, W.M. and Laskey, R.A. (1974) A film detection method for tritium-labelled
proteins and nucleic acids in polyacrylamide gels. Eur J Biochem 46(1), 83-8.
Boyko, V.P., Karasev, A.V., Agranovsky, A.A., Koonin, E.V. and Doija, V.V. (1992)
Coat protein gene duplication in a filamentous RNA virus of plants. Proc Natl Acad Sci U
S A 89(19), 9156-60.
Bransom, K.L. and Dreher, T.W. (1994) Identification of the essential cysteine and
histidine residues of the turnip yellow mosaic virus protease. Virology 198(1), 148-54.
Carrington, J. C., and Dougherty, W. G. (1987) Small nuclear inclusion protein encoded
by a plant potyvirus genome is a protease. J Virol 61(8), 2540-2548.
Carrington, J.C., Cary, S.M., Parks, T.D. and Dougherty, W.G. (1989) A second
proteinase encoded by a plant potyvirus genome. EMBO J 8(2), 365-70.
Chen, J.P., Strauss, J.H., Strauss, E.G. and Frey, T.K. (1996) Characterization of the
rubella virus nonstructural protease domain and its cleavage site. J Virol 70(7), 4707-13.
Choi, G.H., Pawlyk, D.M. and Nuss, D.L. (1991 a) The autocatalytic protease p29
encoded by a hypovirulence-associated virus of the chestnut blight fungus resembles the
potyvirus-encoded protease HC-Pro. Virology 183(2), 747-52.
Choi, H.K., Tong, L., Minor, W., Dumas, P., Boege, U., Rossmann, M.G. and Wengler,
G. (1991 b) Structure of Sindbis virus core protein reveals a chymotrypsin-like seine
proteinase and the organization of the virion. Nature 354(6348), 37-43.
Davis, G.L. (1999) Hepatitis C virus genotypes and quasispecies. Am J Med 107(6B),
de Groot, R.J., Hardy, W.R, Shirako, Y. and Strauss, J.H. (1990) Cleavage-site
preferences of Sindbis virus polyproteins containing the non-structural proteinase.
Evidence for temporal regulation of polyprotein processing in vivo. Embo J 9(8), 2631-8.
De Mejia, M. V. G., Hiebert, E., Purcifull, D. E., Thornburry, D. W., and Pirone, T. P.
(1985) Identification of potyviral amorphous inclusion protein as a non-structural virus-
specific protein related to helper component. Virology 142, 34-43.
den Boon, J.A., Faaberg, K.S., Meulenberg, J.J., Wassenaar, A.L., Plagemann, P.G.,
Gorbalenya, A.E. and Snijder, E.J. (1995) Processing and evolution of the N-terminal
region of the arterivirus replicase ORF 1 a protein: identification of two papainlike
cysteine proteases. J Virol 69(7), 4500-5.
Dolja, V.V., Hong, J., Keller, K.E., Martin, R.R. and Peremyslov, V.V. (1997)
Suppression of potyvirus infection by coexpressed closterovirus protein. Virology 234(2),
Domingo, E., Baranowski, E., Ruiz-Jarabo, C.M., Martin-Hernandez, A.M., Saiz, J.C.
and Escarmis, C. (1998) Quasispecies structure and persistence of RNA viruses. Emerg
Infect Dis 4(4), 521-7.
Dong, S., Gao, H.Q. and Baker, S.C. (1995) Proteolytic processing of the MHV
polymerase polyprotein. Identification of the P28 cleavage site and the adjacent protein,
P65. Adv Exp Med Biol 380, 431-5.
Dong, S. and Baker, S.C. (1994) Determinants of the p28 cleavage site recognized by the
first papain- like cysteine proteinase of murine coronavirus. Virology 204(2), 541-9.
Drake, J.W. and Holland, J.J. (1999) Mutation rates among RNA viruses. Proc Natl Acad
Sci U S A 96(24), 13910-3.
Erokhina, T.N., Zinovkin, R.A., Vitushkina, M.V., Jelkmann, W. and Agranovsky, A.A.
(2000) Detection of beet yellows closterovirus methyltransferase-like and helicase-like
proteins in vivo using monoclonal antibodies. J Gen Virol 81 Pt 3, 597-603.
Febres, V.J., Pappu, H.R., Anderson, E.J., Pappu, S.S., Lee, R.F. and Niblett, C.L. (1994)
The diverged copy of the citrus tristeza virus coat protein is expressed in vivo. Virology
Febres, V. J., Ashoulin, L., Mawassi, M., Frank, M., Bar-Joseph, M., Manjunath, K. L.,
Lee, R. F., and Niblett, C. L. (1996) The p27 protein is present at one end of citrus
tristeza virus particles. Mol Plant Pathol 86(12), 1331-1335.
Garcia, J. A., Fernndez-Fernindez, M. R., and L6pez-Moya, J. J. (1999) Proteases of
infectious agents (Dunn, B. Editor). Academic Press, San Diego. pp. 282.
Godeny, E.K., Chen, L., Kumar, S.N., Methven, S.L., Koonin, E.V. and Brinton, M.A.
(1993) Complete genomic sequence and phylogenetic analysis of the lactate
dehydrogenase-elevating virus (LDV). Virology 194(2), 585-96.
Gorbalenya, A.E., Donchenko, A.P., Blinov, V.M. and Koonin, E.V. (1989) Cysteine
proteases of positive strand RNA viruses and chymotrypsin-like serine proteases. A
distinct protein superfamily with a common structural fold. FEBS Lett 243(2), 103-14.
Gorbalenya, A. E., and Snijder, E. J. (1996) Viral cysteine proteases. Perspectives in
Drug Discovery and Design 6, 86.
Gorbalenya, A.E., Koonin, E.V. and Lai, M.M. (1991) Putative papain-related thiol
proteases of positive-strand RNA viruses. Identification of rubi- and aphthovirus
proteases and delineation of a novel conserved domain associated with proteases of rubi-,
alpha- and coronaviruses. FEBS Lett 288(1-2), 201-5.
Gowda, S., Satyanarayana, T., Davis, C.L., Navas-Castillo, J., Albiach-Marti, M.R.,
Mawassi, M., Valkov, N., Bar-Joseph, M., Moreno, P. and Dawson, W.O. (2000) The p20
gene product of Citrus tristeza virus accumulates in the amorphous inclusion bodies.
Virology 274(2), 246-54.
Graves, M.V. and Roossinck, M.J. (1995) Characterization of defective RNAs derived
from RNA 3 of the Fny strain of cucumber mosaic cucumovirus. J Virol 69(8), 4746-51.
Hagiwara, Y., Peremyslov, V.V. and Dolja, V.V. (1999) Regulation of closterovirus gene
expression examined by insertion of a self-processing reporter and by northern
hybridization. J Virol 73(10), 7988-93.
Hahm, B., Han, D.S., Back, S.H., Song, O.K., Cho, M.J., Kim, C.J., Shimotohno, K. and
Jang, S.K. (1995) NS3-4A of hepatitis C virus is a chymotrypsin-like protease. J Virol
Hardy, W.R. and Strauss, J.H. (1989) Processing the nonstructural polyproteins of sindbis
virus: nonstructural proteinase is in the C-terminal half of nsP2 and functions both in cis
and in trans. J Virol 63(11), 4653-64.
Hata, S., Sato, T., Sorimachi, H., Ishiura, S. and Suzuki, K. (2000) A simple purification
and fluorescent assay method of the poliovirus 3C protease searching for specific
inhibitors. J Virol Methods 84(2), 117-26.
Herold, J., Gorbalenya, A.E., Thiel, V., Schelle, B. and Siddell, S.G. (1998) Proteolytic
processing at the amino terminus of human coronavirus 229E gene 1-encoded
polyproteins: identification of a papain-like proteinase and its substrate. J Virol 72(2),
Hughes, S.A., Bonilla, P.J. and Weiss, S.R. (1995) Identification of the murine
coronavirus p28 cleavage site. J Virol 69(2), 809-13.
Jelkmann, W., Fechtner, B. and Agranovsky, A.A. (1997) Complete genome structure and
phylogenetic analysis of little cherry virus, a mealybug-transmissible closterovirus. J Gen
Virol 78(Pt 8), 2067-71.
Kanjanahaluethai, A. and Baker, S.C. (2000) Identification of mouse hepatitis virus
papain-like proteinase 2 activity. J Virol 74(17), 7911-21.
Karasev, A.V., Boyko, V.P., Gowda, S., Nikolaeva, O.V., Hilf, M.E., Koonin, E.V.,
Niblett, C.L., Cline, K., Gumpf, D.J., Lee, R.F. (1995) Complete sequence of the citrus
tristeza virus RNA genome. Virology 208(2), 511-20.
Kay, J. and Dunn, B.M. (1990) Viral proteinases: weakness in strength. Biochim Biophys
Acta 1048(1), 1-18.
Kirchweger, R., Ziegler, E., Lamphear, B.J., Waters, D., Liebig, H.D., Sommergruber,
W., Sobrino, F., Hohenadl, C., Blaas, D., Rhoads, R.E. (1994) Foot-and-mouth disease
virus leader proteinase: purification of the Lb form and determination of its cleavage site
on eIF-4 gamma. J Virol 68(9), 5677-84.
Kitajima, E. W., Silva, D. M., Oliveira, A. R., Miller, G. D., and Costa, A. S. (1964)
Thread-like particles associated with tristeza disease of citrus. Nature 201, 1011-1012.
Klaassen, V. A., Boeshore, M., Koonin, E. V., Tian, T., and Falk, B. W. (1995) Genome
structure and phylogenetic analysis of lettuce infectious yellows virus, a whitefly
transmitted, bipartite closterovirus. Virology 208, 99-110.
Klotz, L. J. (1978) Fungal, bacterial and non-parasitic diseases and injuries in the seed
bed nursery and orchard. The citrus industry, Vol IV (Calavan, E. C. and Carman, G. E.
Eds.) Univ. Calif., Div. Agri, Sci. Berkeley, CA.
Kong, P., Rubio, L., Polek, M. and Falk, B.W. (2000) Population structure and genetic
diversity within California citrus tristeza virus (CTV) isolates. Virus Genes 21(3), 139-
Koonin, E.V., Choi, G.H., Nuss, D.L., Shapira, R. and Carrington, J.C. (1991) Evidence
for common ancestry of a chestnut blight hypovirulence- associated double-stranded
RNA and a group of positive-strand RNA plant viruses. Proc Natl Acad Sci U S A
Krdiusslich, H.G. and Wimmer, E. (1988) Viral proteinases. Annu Rev Biochem 57, 701-
Lawrence, D.M., Rozanov, M.N. and Hillman, B.I. (1995) Autocatalytic processing of the
223-kDa protein of blueberry scorch carlavirus by a papain-like proteinase. Virology
Lee, H.J., Shieh, C.K., Gorbalenya, A.E., Koonin, E.V., La Monica, N., Tuler, J.,
Bagdzhadzhyan, A. and Lai, M.M. (1991) The complete sequence (22 kilobases) of
murine coronavirus gene 1 encoding the putative proteases and RNA polymerase.
Virology 180(2), 567-82.
Liang, Y., Yao, J. and Gillam, S. (2000) Rubella virus nonstructural protein protease
domains involved in trans- and cis-cleavage activities. J Virol 74(12), 5412-23.
Lim, K.P., Ng, L.F. and Liu, D.X. (2000) Identification of a novel cleavage activity of the
first papain-like proteinase domain encoded by open reading frame I a of the coronavirus
Avian infectious bronchitis virus and characterization of the cleavage products. J Virol
Lim, K.P. and Liu, D.X. (1998) Characterization of the two overlapping papain-like
proteinase domains encoded in gene I of the coronavirus infectious bronchitis virus and
determination of the C-terminal cleavage site of an 87-kDa protein. Virology 245(2), 303-
Liu, D.X. and Brown, T.D. (1995) Characterization and mutational analysis of an ORF
la-encoding proteinase domain responsible for proteolytic processing of the infectious
bronchitis virus la/lb polyprotein. Virology 209(2), 420-7.
Liu, X., Yang, J., Ghazi, A.M. and Frey, T.K. (2000) Characterization of the zinc binding
activity of the rubella virus nonstructural protease. J Virol 74(13), 5949-56.
Liu, X., Ropp, S.L., Jackson, R.J. and Frey, T.K. (1998) The rubella virus nonstructural
protease requires divalent cations for activity and functions in trans. J Virol 72(5), 4463-
Lopez, C., Navas-Castillo, J., Gowda, S., Moreno, P. and Flores, R. (2000) The 23-kDa
protein coded by the Y-terminal gene of citrus tristeza virus is an RNA-binding protein.
Virology 269(2), 462-70.
Manjunath, K. L., Lee, R. F., and Niblett, C.L. (2000) Citrus tristeza virus. Recent
advances in the molecular biology of citrus tristeza closterovirus. Fourteenth IOCV
Conference Citrus tristeza. 1-11.
Mawassi, M., Mietkiewska, E., Hilf, M.E., Ashoulin, L., Karasev, A.V., Gafny, R., Lee,
R.F., Gamsey, S.M., Dawson, W.O. and Bar-Joseph, M. (1995a) Multiple species of
defective RNAs in plants infected with citrus tristeza virus. Virology 214(1), 264-8.
Mawassi, M., Gafny, R., Gagliardi, D. and Bar-Joseph, M. (1995b) Populations of citrus
tristeza virus contain smaller-than-full-length particles which encapsidate sub-genomic
RNA molecules. J Gen Virol 76(Pt 3), 651-9.
Merits, A., Vasiljeva, L., Ahola, T., Kaariainen, L. and Auvinen, P. (2001) Proteolytic
processing of Semliki Forest virus-specific non-structural polyprotein by nsP2 protease. J
Gen Virol 82(Pt 4), 765-73.
Napuli, A.J., Falk, B.W. and Dolja, V.V. (2000) Interaction between HSP70 homolog and
filamentous virions of the Beet yellows virus. Virology 274(1), 232-9.
Navas-Castillo, J., Albiach-Marti, M.R., Gowda, S., Hilf, M.E., Garnsey, S.M. and
Dawson, W.O. (1997) Kinetics of accumulation of citrus tristeza virus RNAs. Virology
Niblett, C.L., Genc, H., Cevik, B., Halbert, S., Brown, L., Nolasco, G., Bonacalza, B.,
Manjunath, K.L., Febres, V.J., Pappu, H.R. and Lee, R.F. (2000) Progress on strain
differentiation of Citrus tristeza virus and its application to the epidemiology of citrus
tristeza disease. Virus Res 71(1-2), 97-106.
Oh, C.S. and Carrington, J.C. (1989) Identification of essential residues in potyvirus
proteinase HC-Pro by site-directed mutagenesis. Virology 173(2), 692-9.
Pappu, H.R., Karasev, A.V., Anderson, E.J., Pappu, S.S., Hilf, M.E., Febres, V.J.,
Eckloff, R.M., McCaffery, M., Boyko, V., Gowda, S. (1994) Nucleotide sequence and
organization of eight 3' open reading frames of the citrus tristeza closterovirus genome.
Virology 199(1), 35-46.
Peng, C.W. and Dolja, V.V. (2000) Leader proteinase of the beet yellows closterovirus:
mutation analysis of the function in genome amplification. J Virol 74(20), 9766-70.
Peremyslov, V.V., Hagiwara, Y. and Dolja, V.V. (1998) Genes required for replication of
the 15.5-kilobase RNA genome of a plant closterovirus. J Virol 72(7), 5870-6.
Pieroni, L., Santolini, E., Fipaldini, C., Pacini, L., Migliaccio, G. and La Monica, N.
(1997) In vitro study of the NS2-3 protease of hepatitis C virus. J Virol 71(9), 6373-80.
Plagemann, P.G. and Moennig, V. (1992) Lactate dehydrogenase-elevating virus, equine
arteritis virus, and simian hemorrhagic fever virus: a new group of positive-strand RNA
viruses. Adv Virus Res 41, 99-192.
Polgar, L. and Halasz, P. (1982) Current problems in mechanistic studies of serine and
cysteine proteinases. Biochem J 207(1), 1-10.
Rao, M.B., Tanksale, A.M., Ghatge, M.S. and Deshpande, V.V. (1998) Molecular and
biotechnological aspects of microbial proteases. Microbiol Mol Biol Rev 62(3), 597-635.
Rawlings, N.D. & Barrett, A.J. (2000) MEROPS: the peptidase database. Nucleic Acids
Res. 28, 323-325.
Rocha-Pefia, M. A., Lee, R. F., Ldstra, R., Niblett, C. L., Ochoa-Corona, F. M., Garnsey,
S. M., and Yokomi, R. K. (1995) Citrus tristeza virus and its aphid vector Toxoptera
citricida. Plant disease, 437-445.
Romero, J., Huang, Q., Pogany, J. and Bujarski, J.J. (1993) Characterization of defective
interfering RNA components that increase symptom severity of broad bean mottle virus
infections. Virology 194(2), 576-84.
Ryan, M.D. and Flint, M. (1997) Virus-encoded proteinases of the picornavirus super-
group. J Gen Virol 78(Pt 4), 699-723.
Schechter, I., and Berger, A. (1967) On the size of the active site in proteases. Biochem.
Biophys. Res. Com. 27, 157-162.
Schiller, J.J. and Baker, S.C. (1998) Maturation of the polymerase polyprotein of the
coronavirus MHV strain JHM involves a cascade of proteolytic processing events. Adv
Exp Med Biol 440, 135-9.
Scholthof, K.B., Scholthof, H.B. and Jackson, A.O. (1995) The effect of defective
interfering RNAs on the accumulation of tomato bushy stunt virus proteins and
implications for disease attenuation. Virology 211(1), 324-8.
Shapira, R. and Nuss, D.L. (1991) Gene expression by a hypovirulence-associated virus
of the chestnut blight fungus involves two papain-like protease activities. Essential
residues and cleavage site requirements for p48 autoproteolysis. J Biol Chem 266(29),
Sircar, S., Ruzindana-Umunyana, A., Neugebauer, W. and Weber, J.M. (1998)
Adenovirus endopeptidase and papain are inhibited by the same agents. Antiviral Res
Snijder, E.J., Wassenaar, A.L. and Spaan, W.J. (1992) The 5' end of the equine arteritis
virus replicase gene encodes a papainlike cysteine protease. J Virol 66(12), 7040-8.
Snijder, E.J., Wassenaar, A.L. and Spaan, W.J. (1994) Proteolytic processing of the
replicase ORFla protein of equine arteritis virus. J Virol 68(9), 5755-64.
Teng, H., Pinon, J.D. and Weiss, S.R. (1999) Expression of murine coronavirus
recombinant papain-like proteinase: efficient cleavage is dependent on the lengths of both
the substrate and the proteinase polypeptides. J Virol 73(4), 2658-66.
Tihanyi, K., Bourbonniere, M., Houde, A., Rancourt, C. and Weber, J.M. (1993) Isolation
and properties of adenovirus type 2 proteinase. J Biol Chem 268(3), 1780-5.
Tijms, M.A., van Dinten, L.C., Gorbalenya, A.E. and Snijder, E.J. (2001) A zinc finger-
containing papain-like protease couples subgenomic mRNA synthesis to genome
translation in a positive-stranded RNA virus. Proc Natl Acad Sci U S A 98(4), 1889-94.
van Regenmortel, M.H.V., Fauquet,C.M., Bishop, D.H.L., Carstens, E.B, Estes, M.K.,
Lemon, S.M., Maniloff, J., Mayo, M.A., McGeoch, D.J., Pringle, C.R., Wickner, R.B.
(2000). Virus Taxonomy, VIIth report of the ICTV. Academic Press, SanDiego, 1167pp.
Vardi, E., Sela, I., Edelbaum, 0., Livneh, 0., Kuznetsova, L. and Stram, Y. (1993) Plants
transformed with a cistron of a potato virus Y protease (NIa) are resistant to virus
infection. Proc Natl Acad Sci U S A 90(16), 7513-7.
White, K.A., Bancroft, J.B. and Mackie, G.A. (1991) Defective RNAs of clover yellow
mosaic virus encode nonstructural/coat protein fusion products. Virology 183(2), 479-86.
Yang, G., Che, X., Gofman, R., Ben-Shalom, Y., Piestun, D., Gafny, R., Mawassi, M. and
Bar-Joseph, M. (1999) D-RNA molecules associated with subisolates of the VT strain of
citrus tristeza virus which induce different seedling-yellows reactions. Virus Genes 19(1),
Yao, J., Yang, D., Chong, P., Hwang, D., Liang, Y. and Gillam, S. (1998) Proteolytic
processing of rubella virus nonstructural proteins. Virology 246(1), 74-82.
Zhu, H.Y., Ling, K.S., Goszczynski, D.E., McFerson, J.R. and Gonsalves, D. (1998)
Nucleotide sequence and genome organization of grapevine leafroll- associated virus-2
are similar to beet yellows virus, the closterovirus type member. J Gen Virol 79(Pt 5),
Ziebuhr, J., Snijder, E.J. and Gorbalenya, A.E. (2000) Virus-encoded proteinases and
proteolytic processing in the Nidovirales. J Gen Virol 81 Pt 4, 853-79.
Jorge Vdzquez-Ortiz was born in Caracas, Venezuela, in 1970. He obtained a
bachelor's degree in biology in 1994 from the "Universidad Central de Venezuela" in
Caracas. After his graduation he worked at the "Universidad Central de Venezuela" as an
assistant professor in the human physiology department, medical school. In 1996 he
started his studies for a Ph.D. in plant molecular and cellular biology at the University of
Florida, from which he graduated in 2001.
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
C. Niblett, Chair
Professor of Plant Molecular and Cellular
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
B. M. Dunn
Distinguished Professor of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
J. M. avis
Associate Professor of Forest Resources and
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Professor of Plant Pathology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Professor of Horticultural Science
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
M. L. Wayne
Assistant Professor of Zoology
This dissertation was submitted to the Graduate Faculty of the College of
Agricultural and Life Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
Dean, College of Agricult1 d Life
Dean, Graduate School