Targeted in vitro construction of conditional lethal vaccinia virus mutants

MISSING IMAGE

Material Information

Title:
Targeted in vitro construction of conditional lethal vaccinia virus mutants
Physical Description:
viii, 212 leaves : ill. ; 29 cm.
Language:
English
Creator:
Hassett, Daniel E., 1964-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Vaccinia virus -- genetics   ( mesh )
Mutagenesis, Site-Directed -- methods   ( mesh )
Genes, Structural, Viral   ( mesh )
Telomere   ( mesh )
Alanine   ( mesh )
Molecular Sequence Data   ( mesh )
Amino Acid Sequence   ( mesh )
Department of Molecular Genetics and Microbiology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Molecular Genetics and Microbiology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1995.
Bibliography:
Bibliography: leaves 195-211.
Statement of Responsibility:
by Daniel E. Hassett.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002306734
oclc - 50397208
notis - ALR0052
System ID:
AA00009025:00001


This item is only available as the following downloads:


Full Text














TARGETED IN VITRO CONSTRUCTION OF CONDITIONAL LETHAL
VACCINIA VIRUS MUTANTS














By
DANIEL E. HASSETT








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
1995






























This work is dedicated to my parents, Alan and Noel and Miss

Loretta.














ACKNOWLEDGMENTS


I wish to offer my sincere thanks and gratitude to Rich Condit

whose friendship, support and unending patience has made my

graduate career a very enjoyable experience.

I thank all of my labmates in the Condit and Moyer labs for

their interest in my work, suggestions and friendship. Thanks go to

Ms. Jackie Lewis for her help in cloning and sequencing some of the

D1R mutants. Special thanks go to Ms. Joyce Conners for her helpful

advice and friendship over the last five years.

I would like to thank my family for their love and support

especially my parents, they deserve much of the credit for making

me the person I am today. I would like to thank the Donald who

helped to make Florida my home. I would also like to thank Elinor

and Bruce Cuvilier who helped me at a time in my life when I was

unable to help myself, their belief in me meant more to me than

either of them will ever know. Lastly I would like to offer my

thanks to N. Loretta Sukhu for her love, patience and perseverance.














TABLE OF CONTENTS


ACKNOWLEDGEMENTS ..................................................................................iii

ABSTRACT .......................................................................................................vi

CHAPTERS

1 INTRODUCTION AND BACKROUND......................................................
Genome and Genomic Organization...........................................6...
Viral Life Cycle..................................................................................9...
Viral Genetics...................................................................................33
Clustered Charge to Alanine Mutagenesis........................... 37
Transient Dominant Selection...................................................42
Capping of mRNA and the Vaccinia mRNA Capping
Enzym e............................................................................................. 50

2 MATERIALS AND METHODS............................................................60
Recombinant DNA Techniques..................................................60
Virological Techniques................................................................. 72

3 TESTING CLUSTERED CHARGE TO ALANINE MUTAGENESIS
AND TRANSIENT DOMINANT SELECTION ON
THE VACCINIA VIRUS GENE G2R..................................................87
Introduction......................................................................................87
R esults................................................................................................. 90
Discussion.........................................................................................109









4 CONSTRUCTION AND CHARACTERIZATION OF A NOVEL
CONDITIONAL LETHAL MUTANT IN THE LARGE SUBUNIT
OF THE VACCINIA VIRUS mRNA CAPPING ENZYME...........112
Introduction.................................................................................... 112
R esults................................................................................................ 118
D iscussion......................................................................................... 167

5 SUMMARY.............................................................................................182
Clustered Charge to Alanine Mutagenesis...............183
Transient Dominant Selection.................................................. 188
Future Experiments with E793.............................................. 190

LIST OF REFERENCES...............................................................................195

BIOGRAPHICAL SKETCH.........................................................................212














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


TARGETED IN VITRO CONSTRUCTION OF CONDITIONAL
LETHAL VACCINIA VIRUS MUTANTS

By

Daniel E. Hassett

December, 1995




Chairperson: Richard C. Condit Ph.D.
Major Department: Molecular Genetics and Microbiology


The goal of this study has been to develop a general directed

genetic method for the creation of site-specific, conditionally lethal,

temperature-sensitive vaccinia virus mutants. The techniques of

clustered charge to alanine mutagenesis and transient dominant

selection were tested as a mutagenesis strategy and as a means of

targeted gene replacement, respectively. The efficiency of clustered

charge to alanine mutagenesis and transient dominant selection were









initially assessed by the creation of nine novel alleles of the vaccinia

virus G2R gene because a simple drug selection can be used to select

for mutations that alter the function of G2R. Three of the G2R

mutants were conditionally lethal as measured by one step growth

experiments. These data indicate that 33% of the clustered charge to

alanine alleles were conditionally lethal in vivo. The experiments

with G2R also revealed that on average 11% of the viral progeny

isolated during the transient dominant selection contained the

clustered charge to alanine mutation.

To demonstrate the generality of the method, nine clustered

charge to alanine mutations were created in the large subunit of the

heterodimeric vaccinia virus capping enzyme, gene DIR. In addition

to catalyzing the formation of a cap 0 structure on newly synthesized

mRNA, the vaccinia capping enzyme has been shown to be an early

gene transcription termination factor and an intermediate gene

transcription initiation factor. One of the nine mutations created in

D1R was temperature-sensitive for growth as measured by one-step

growth experiments. In vivo characterization of the conditional

lethal DIR allele revealed that protein, DNA and mRNA synthesis

appear relatively normal at the nonpermissive temperature. The









mutant is, however, defective in the resolution of viral telomeres and

blocked at an early stage of viral assembly. Methyltransferase

assays on purified virions revealed that the mutant capping enzyme

protein is thermolabile for methyltransferase activity in vitro.

The results with the vaccinia virus genes G2R and D1R

demonstrate that clustered charge to alanine mutagenesis and

transient dominant selection can be used as a general directed

genetic method for the creation of conditionally lethal, temperature-

sensitive vaccinia virus mutants.


viii













CHAPTER 1
INTRODUCTION AND BACKGROUND





The science of modern molecular biology has relied heavily on

genetics to gain insight into the molecular mechanisms of living

systems. Many of the fundamental biochemical mechanisms used by

living systems have been initially uncovered and understood through

the genetic analysis of eukaryotic viruses. The small genome size

and quick replicative cycle of viruses have made them attractive

model systems with which to study the processes of DNA and RNA

replication, transcription, translation and post translational protein

processing. The genetic analysis of the orthopoxvirus vaccinia has

been at the forefront of many of these investigations.

Orthopoxviruses, whose prototypic member is vaccinia, are

characterized by their large (192-300 kb) double-stranded DNA

genome, cytoplasmic site of replication and packaging within their

virions of all the enzymes necessary for early gene transcription











(reviewed in 78). Medical interest in vaccinia began because of the

virus' ability to provide protective immunity against variola virus,

the causative agent of smallpox. Although the global eradication of

smallpox was completed in 1979, largely due to the use of vaccinia as

a live virus vaccine, medical interest continues because of vaccinia's

potential use as a recombinant live virus vaccine vector (30, 36, 42,

88). The ability to genetically manipulate vaccinia and the capability

of the virus to express foreign genes when placed under the control

of appropriate viral regulatory signals are essential aspects of

recombinant vaccinia vaccine design. Therefore successful vaccine

design requires detailed knowledge of the general biology of the

poxvirus replication cycle. The basic biology of vaccinia is also of

interest, primarily due to the cytoplasmic site of viral replication. To

replicate within the cytoplasm vaccinia must encode most if not all of

the genetic information necessary for DNA replication, transcription

and controlled gene expression (78). The cytoplasmic site of

replication, packaging of viral enzymes within the viral particle and

the ease with which the viral genome may be genetically

manipulated has made vaccinia an excellent model system with

which to study many aspects of eukaryotic biology.











Most of the genetic analysis of vaccinia virus has focused on

the isolation and characterization of randomly generated viral

mutants (reviewed in 15). The characterization of these mutants has

proven quite useful in deciphering the general mechanisms involved

in the control of eukaryotic gene expression, DNA replication and

mRNA processing. However, the applicability of vaccinia virus

genetics has been limited by 1) the inability to target specific viral

genes for mutagenesis and 2) the laborious screening procedures

necessary for the isolation of viral mutants.

The purpose of this dissertation is to develop a directed genetic

method for the creation of conditionally lethal temperature-sensitive

mutations in specific vaccinia virus genes. Directed genetics of

vaccinia virus involves the in vitro mutagenesis of cloned copies of

viral genes followed by recombination-mediated gene replacement,

to transfer these mutations onto the viral chromosome. Two

problems must be addressed in order to successfully apply directed

genetics to vaccinia virus: 1) the in vitro mutagenesis scheme must

produce a high proportion of conditional lethal mutants in vivo; 2)

the mutant alleles must be efficiently recombined back onto the viral

chromosome. Clustered charge to alanine mutagenesis and transient











dominant selection were tested as a means of making site-specific

mutations in vitro that are temperature-sensitive in vivo and as a

strategy for targeted gene replacement, respectively. The mutation

of charged amino acids (Asp, Glu, His, Lys, Arg), closely spaced within

the primary sequence of a polypeptide, to alanines has been shown

to produce a high proportion of temperature-sensitive mutants when

these mutations are assayed in vivo (25, 130). Recombination

mediated gene replacement involving the dominant selectable

marker xanthine-guanine phosphoribosyltransferase from

Escherichia coli (ecogpt) was then used to transplace the wild type

chromosomal copy of the gene of interest with a clustered charge to

alanine mutant copy (32).

Clustered charge to alanine mutagenesis and transient

dominant selection were tested by creating mutations in the vaccinia

virus genes G2R and DIR. The efficiency of the vaccinia virus

directed genetic method was initially assessed by creating mutations

in G2R, because mutations in G2R produce viruses with selectable

phenotypes (70). The ability to select rather than to screen for

viruses containing clustered charge to alanine mutations which

altered G2R function made it possible to easily determine the











phenotypes of individual mutations. As a result of transient

dominant selection, individual plaques are produced which contain a

heterogeneous population of wild type and mutant viruses. The

selectability of the G2R mutants also made it possible to determine

what percentage of viruses within this heterogenous population

contain the mutation of interest. The general applicability of the

directed genetic method was assessed by creating mutations in DIR,

which encodes the large subunit of the heterodimeric vaccinia mRNA

capping enzyme. The vaccinia virus mRNA capping enzyme is

encoded by genes D1R (large subunit) and D12L (small subunit) (77,

86). In addition to its role in mRNA 5' end formation, the capping

enzyme also catalyzes the termination of early gene transcription

and the initiation of intermediate gene transcription (109, 114, 126).

The multifunctional nature of the capping enzyme and the lack of

temperature-sensitive mutations in the large subunit make the

capping enzyme an excellent target for directed genetics. In

addition, the in vivo analysis of a capping enzyme mutant would

provide more information about the role the mRNA capping enzyme

plays in vivo during the viral life cycle.










Genome and Genomic Organization




The vaccinia virus genome (Copenhagen strain) is a double-

stranded DNA molecule of 191,636 bp in length (41, 54). The ends of

the genome terminate in incompletely base-paired hairpin loops

which serve to covalently crosslink each strand of the DNA duplex

thereby generating a single contiguous polynucleotide chain.

Nucleotide sequencing of the hairpin termini indicate that they exist

in two isomeric forms, the sequences of which are inverted and

complementary with respect to each another (4). Adjacent to each

hairpin termini are 12 kb inverted terminal repeat sequences that

contain within them numerous smaller direct repeat elements (41).

Computer assisted sequence analysis of the viral genome

suggests a coding capacity of 198 intronless genes (41). By

convention open reading frames are designated by their relative

position with respect to Hindll restriction sites. (65, 99). Digestion

of the vaccinia virus chromosome with HindIII produces 16

fragments ranging in size form 49 kb (A fragment) to 273 bp (P

fragment) (41) (Fig. 1).












co
4-o







0
o c
..4


4) Cu0



Gn
-~ 4) 8 0 -
0 '* 4)*U


4) a
.co o^>- '*



OC o-g

w0 c
0 t3 (L> *


1-4 ~ i 1-
go Cu









on>
4 U
_a -a
oa I --4




fe)0n







4)C#)


t~o8
23 .= a) 3
4J .O cu0






o C4 m-
.^ ^ E c3





0 -
g i sI S








.- 0 4)











on *n
0 d -
8 > a
to a 6 5
4) 4)- 4)e
U 4) BO 4)'
=o"c "ro












on 8
.5 L oo 0



*-0 Cu* *;-' o- -
U .



















0
0
0-




C

0
mEP




O
tm



C
(S
(0






C








C
(0

0






0



C





e
0

0
"0
0



C


c
0
c



0




C
0

0
1


0
0


0









9

The vaccinia virus genome can also be functionally divided into

three regions, a central core containing genes required for replication

in tissue culture flanked on either side by two regions which contain

genes nonessential for tissue culture replication. The central core,

which extends from the HindIII E fragment through the left hand

two thirds of the HindIII A fragment contains approximately 100

genes (41). The genes within the central core are highly conserved

among other members of the orthopoxviridae and encode the

enzymatic activities necessary for the survival and reproduction of

the virus within the infected cell. In contrast the nonessential genes

encode proteins necessary for pathogenesis and spread within the

infected host organism (41).

Viral Life Cycle




The critical events in the life cycle of vaccinia virus which

include attachment and penetration of the infecting virion, early

intermediate and late gene expression, DNA replication, virion

morphogenesis and release are diagrammatically represented in Fig.

2 (reviewed in 78). Attachment to and penetration of the host cell

are poorly understood processes primarily due to a lack of any




























* 0











PMl
O
o>
U >
* *




c-o m 0

,-4 0












> 4-
uu
0















-o 4) _
- U -8
4 -

















'-4 .
o *























NCA
U.-
E -o ^Q
0 0 0





a|.








I -





a ca <
Se > '8w
.*C 13 ca-


t>


** c *























z
E 2>
o16Q-
CU0


6<


E
8 E


0


E2>


0-
2o >
a5-
S|&>











definitive information about the identity of the cellular or viral

receptors involved. Following attachment and penetration, the

infecting virus reaches the cytoplasm which is the site of all

subsequent viral transcription and replication. Almost immediately

after penetration there is a partial uncoating of the virion (uncoating

I) that serves to facilitate the onset early viral transcription.

Early Transcription



The transcription of early genes is directed by an RNA

polymerase, an early transcription factor and an early gene specific

RNA polymerase associated protein packaged within the virion (79).

One-half of the viral genome is transcribed during the early phase of

the infection (78). Early genes encode proteins such as the core RNA

polymerase subunits, viral mRNA capping enzyme and polyA

polymerase as well as enzymes involved in DNA replication and

nucleic acid biosynthesis (78).

The DNA dependant RNA polymerase activity isolated from

virion particles contains at least 8 subunits and has a molecular mass

in excess of 450 kDa (78). This core RNA polymerase contains all of

the catalytic activities necessary for RNA polymerization from single











stranded DNA, but it lacks initiation specificity and the ability to

transcribe a double-stranded DNA template (reviewed in 79).

Recently two distinct populations of RNA polymerase activity

have been observed within purified virions. These polymerase

populations differ in their association with a 94 kDa protein termed

Rap94 for RNA polymerase associated protein of 94 kDa. Rap94 is

encoded by the H4 gene of vaccinia virus (1, 55). Rap94 confers

early promoter specificity on the core RNA polymerase. Although

essential for early gene transcription, Rap94 is not required for the

transactivation of late gene transcription in vitro (132).

Specific transcription of a double-stranded DNA template

containing an early promoter can be conferred on the core RNA

polymerase-Rap94 complex by the addition of a heterodimeric

protein referred to as yaccinia early transcription factor (vETF)

(reviewed in 79). vETF is composed of an 82 kDa protein (the

product of the A7L gene) containing leucine zipper and zinc finger

motifs and a 70 kDa protein (the product of the D6R gene) which

contains motifs associated with helicases and ATPases (8, 40).

Purified vETF has been shown to be a sequence specific DNA binding

protein as well as a DNA dependant ATPase (9). vETF recognizes cis








14

acting sequences contained within early viral promoters, since point

mutations in a consensus early promoter which abrogated vETF DNA

binding abolished early transcription activity (134). VETF has also

been shown to bend DNA around the region of the transcriptional

start site (8).

The Rap 94-containing core polymerase and vETF within the

infecting virion are sufficient for the transcription of early genes

(79). These early mRNAs are terminated, capped and polyadenylated

by virally encoded enzymes also contained within the virion (79).

Early Transcription Termination



The vaccinia virus mRNA capping enzyme catalyzes the

termination of early gene transcription in response to the cis acting

sequence UUUUUNU in the nascent mRNA (114). Early transcription

termination occurs 70 to 75 nucleotides downstream of a TTTTTNT

element in the sense DNA strand (96, 135). Both mRNA capping

enzyme subunits are required for early transcription termination

activity; however, neither functional guanyltransferase nor (guanine-

7)methyltransferase active sites are necessary for termination (63).

A model for early transcription termination has been described that








15

calls for the capping enzyme to bind to mRNA as it is extruded from

the back of the elongating polymerase (63). Upon encountering the

UUUUUNU signal the capping enzyme catalyzes the termination of

transcription via protein-protein contacts with the elongation

complex. Consistent with this data is the observation that the large

subunit of the capping enzyme is capable of binding RNA. Although

the RNA binding of the DIR protein is nonspecific, RNA binding can

be preferentially competed with poly U, indicating that this RNA

binding domain may be involved in recognition of the UUUUUNU

sequence in vivo (64). The domain responsible for the interaction of

the capping enzyme with the early transcription elongation complex

is presently unknown.

Following termination early transcripts, which have been

capped and polyadenylated, are extruded from the viral core into the

cytoplasm (78). The accumulation of early gene products which

encode transcription factors specific for intermediate gene

transcription and the onset of viral DNA replication lead to the

second transcriptional stage of the viral life cycle the transactivation

of intermediate genes.











Intermediate Transcription



The transcription of intermediate and late vaccinia genes

occurs after the onset of DNA replication (79). The two classes of

post replicative genes can be distinguished by their relative

transcriptional activity in the presence inhibitors of DNA synthesis.

Intermediate genes are defined as those post replicative genes

capable of being expressed from plasmids transfected into infected

cells that have been treated with a DNA synthesis inhibitor (127).

Therefore all of the trans-acting factors necessary for intermediate

gene transcription are present in the infected cells prior to DNA

replication (79). Five intermediate genes have been identified to

date, three of which function as late transcription factors (56, 127).

Transcription from intermediate promoters requires the core

RNA polymerase complex, rpo 30 (gene E4L), an undefined cellular

factor or factors termed VITF-2 or vaccinia intermediate

transcription factor 2. and the vaccinia mRNA capping enzyme, (97,

98, 126). The capping enzyme forms a stable complex with the core

RNA polymerase and complex formation is required for binding of

the polymerase at the intermediate promoter (126). Therefore the











participation of vaccinia capping enzyme in intermediate

transcription initiation is at a level other than that of mRNA capping

(44). The mRNA capping enzyme domain necessary for the

transactivation of intermediate transcription is unknown.

Late Transcription



With continuing DNA synthesis and the buildup of intermediate

proteins, some of which encode late transcription factors, the last

transcriptional phase of the viral life cycle begins. Late genes encode

structural proteins of the viral particle, the early transcriptional

machinery packaged within the developing virion and enzymes

involved in morphogenesis (78).

A genetic approach has been used to identify some of the

intermediate gene products required for late gene expression.

Intermediate and late transcription is blocked in the presence of DNA

synthesis inhibitors (127). However, intermediate promoters are

utilized if they are contained on plasmids transfected into the

inhibitor treated infected cells (127). Late transcription factors,

encoded by intermediate genes, were identified by transfecting

cloned vaccinia genes into DNA synthesis inhibitor treated cells and








18

assaying for transcription of a late promoter containing reporter gene

construct (133). This methodology identified three intermediate

genes, AlL, A2L and G8R that function as late transcription factors

(56). The exact role each of these proteins plays in the transcription

of late genes is unknown. Recently a factor identified as P3 has been

shown to also be required for high level in vitro transcription of a

late promoter containing reporter construct. The P3 late

transcription factor activity is present in infected cells prior to the

onset of DNA replication; it is, however, not detected in uninfected

cells. It remains to be determined if P3 is a cellular factor activated

by the infection or is virally encoded (58).

Structure of Intermediate and Late Transcripts



Two distinguishing features of intermediate and late mRNA

transcripts are the presence of nontemplated polyadenylate residues

at their 5' ends and the lack of discrete 3' ends (79). The

nontemplated adenylate residues at the 5' end of capped

intermediate and late mRNAs are thought to arise via a slippage

mechanism by the RNA polymerase (79). The average length of the

polyA head is between 35 to 50 nucleotides (79). Interestingly, in











the absence of a poly A head, the translational start site of most

intermediate and late mRNAs would be coterminal with the 5' end.

Therefore the polyA head may function in a manner analogous to a 5'

untranslated region, increasing the efficiency of translation initiation

by facilitating the binding of either translation initiation factors or

the 40s ribosomal subunit (79).

Transcripts derived from individual intermediate and late

genes are heterogeneous in length at their 3' ends due to a lack of

discrete sites of transcription termination. The termination of

intermediate and late transcription appears to occur randomly,

generating a population of mRNA molecules of different lengths. This

heterogeneity in size can be detected as a smear upon northern

analysis of individual intermediate or late viral mRNAs.

DNA Replication



Orthopoxvirus DNA replication occurs in areas of the cytoplasm

known as virosomes or viral factories. Because replication of

vaccinia DNA can occur in enucleated cells, it assumed that all of the

replication machinery is virally encoded (92). Many of the viral

enzymes necessary for DNA replication have been identified; these











include a DNA polymerase, a nicking-joining enzyme, a DNA

topoisomerase, a dUTPase, and a DNA binding protein (reviewed in

78, 123).

Several previously mentioned structural features of the

vaccinia virus genome are shown in Fig. 3. The ends of the genome

consist of incompletely base paired hairpin termini which covalently

crosslink each strand of the genome (Fig. 3,1). The DNA sequences of

the hairpin ends are inverted and complementary isomers of each

other termed flip and flop or fast and slow based upon their relative

mobilities in nondenaturing agarose gels. On either side of the viral

chromosome, adjacent to the hairpin ends are located 12 kb inverted

terminal repeats (Fig. 3,2). Within these inverted terminal repeats

are numerous small tandemly arranged direct repeat elements. The

direct repeat sequences begin about 100 base pairs from the hairpin

ends and extend inward for approximately 3 kb into the inverted

terminal repeats (Fig. 3,3) (5).

The exact mechanism of poxvirus DNA replication is unknown;

however, the available data suggests that replication occurs in a

semiconservative fashion and begins with a site-specific nick within

























Fig. 3. Structural elements of the vaccinia virus genome.
1) Nucleotide sequence of the maximum base paired structure of the
vaccinia virus hairpin ends flip (S) and flop (F). Below is the
nucleotide sequences of flip arranged 5'-3' and flop arranged 3'-5'
demonstrating the inverted complementary nature of the hairpin
isomers. Adapted from Baroudy, B.M., S. Venkatesan and B. Moss
1983. Structure and replication of vaccinia virus telomeres.
Cold.Spring.Harbor.Symp.Quant.Biol. 47 Pt 2 :p 726.
2) Vaccinia genome showing the location of the 12 kb inverted
terminal repeats. 3) A single inverted terminal repeat
demonstrating the position of the direct repeats.











S
40** o o c A
.. . ... .. .. . ..
A-T-C-*-A-T-eA-C -I-A -Al-A-T-A-T-A-T-T A-A-A-A-TA-TTJ.A-A--TI-AA-A-I7 -A-AA-A-iA-4-4-A-A-?-4.4.A-T.C.-
T TT C C O T A T-V
0 ** 7* sO0 O** 1

F
o** 40* so to
A A A c A T A-A
T-A-0-T-A-At4-A-A-A-T- -fA-T-A-TA-T-A-. T -T-T-T-. A-T-A-A-A T-A (T-T-T-T-A -r4-r-r-- .r--4--r-C-r-
I -.. *. ,** . ...,.. .. . .
T-T-C-A-A-T-C-A-T-T-T-A-A-T-A-T-A-T-A-T.A-T-T-A-A-A-A-T-A-T-T-A-A-T-T-A-A-A-T-T-A-A-A-A-T A V- 4-4-A--4-4r-A-T-C-A-C-A4*
s** o ** 0o y



IAAAfITTA IATAt TA T TACATGATIT GAArTATA T 4ATIATAAT AATGA GTT cAAICAT ATIA ATATAATAT IAAA TA TAATAAA IGAT

V 00 so 0 70 so0 SO 40 30 20 1
IAT" TAAAAfATAATA TAATGTACTAAAACI TATATATAT TATAITTIATCTAACTAAAGT TAG1AAATTATATAAT A IAAA TIATA AATI TATTIACI,%A

2. 12 kb Inverted terminal repeats


i l i i I I I I i
20 40 60 80 100 120 140 160 180

kllobase pairs



Hairpin end Hairpin end




3.

C IIIllllIlll Illllllll I I
I I I I I
2 4 6 8 10
direct repeats











the inverted terminal repeat. The isolation of replicative

intermediates of rabbitpox virus and vaccinia which consist of

concatameric forms of genomic DNA arranged in a head to head or

tail to tail linear array (concatamers consisting of a left end of the

genome fused to a left end of another genome or a right end of the

genome fused to another right end) have lead to the formulation of a

model for DNA replication outlined in Fig. 4 (5, 82). In this model

DNA replication begins with a nick within the inverted terminal

repeats proximal to the hairpin termini. The nick generates a free 3'

hydroxyl capable of acting as a primer for DNA synthesis. Elongation

of the free 3' end generates a replication intermediate that is slightly

longer than genome length (Fig. 4,2). The self complementarity of

opposing strands of the duplex allows the DNA strands to fold back

upon themselves forming hairpin ends (Fig. 4,3). One hairpin

contains a free 3' hydroxyl and thus can act as a primer for the

duplication of the entire genome (Figs. 4,4 and 4,5). The resulting

molecule consists of two genomes arranged such that the right end of

one genome is fused to the right end of another genome (tail to tail)

with a nick at one end (Fig. 4,5). Continuation of this process yields a

tetrameric molecule consisting of genomes arranged head to head


























Fig. 4. Self-priming model for vaccinia virus DNA
replication. Sequences of the inverted terminal repeats and their
complements are indicated by uppercase and lowercase letters
respectively. F and S denote flip and flop hairpin ends. Newly
replicated DNA is represented by dashed lines. Shown is a scheme in
which replication is initiated by a nick at one end of the genome.
Adapted from Baroudy, B.M., S. Venkatesan and B. Moss 1983.
Structure and replication of vaccinia virus telomeres.
Cold.Spring.Harbor.Symp.Quant.Biol. 47 Pt 2:p 725,


















c
'C


C

AC
A


0 E F


I d


do FE

D F Ied


d s f


FED 0'


c

>
*


c
. A C 0DE I ______b
T oab cd e F oe


0 E" F nock








A C
. 0 F f d c

d Ft d
A -------- -----









.-7"----------- y


--------------------











and tail to tail (not shown). Support of this model includes the

isolation of just such tetrameric replication intermediates (82).

Additional support of the model has been provided by temperature

shift experiments with replication deficient mutants which during

the initial phases of DNA replication incorporate radiolabeled DNA

precursors near the ends of the genome (94).

Telomere Resolution



The vaccinia replication scheme outlined in Fig. 4 leads to two

copies of the viral genome fused at a duplex copy of a hairpin loop

(Fig. 4,5). The duplicated form of the hairpin loop is called the

concatamer junction. The resolution of concatameric DNA into

genome length units, containing terminal hairpin ends (Fig. 4,6), a

process known as telomere resolution, has been described as a site-

specific recombination event that requires an as yet unidentified

transacting telomere resolvase and two identical inverted copies of a

cis element known as the telomere resolution target (reviewed in

23). Although the exact mechanism is unknown, during telomere

resolution a cruciform structure is formed by intrastrand base

pairing between flip and flop sequences within the concatamer








27

junction (21). Cruciform structures have been observed in electron

micrographs of concatamer junction containing plasmids (21). The

arms of the cruciform consist of flip and flop sequences and the base

forms a holiday structure. Site-specific nicking, strand exchange and

ligation then serve to reconstitute the terminal hairpin ends (73).

Circular plasmids containing a concatamer junction and

surrounding flanking sequences will resolve, when transfected into

infected cells, into linear minichromosomes with covalently closed

hairpin ends (21, 73). This in vivo assay has been used to determine

the sequence requirements for telomere resolution (73). Optimal

resolution of circular plasmids into linear minichromosomes in vivo

requires two inverted copies of a 53 base pair sequence, flanking

inverted repeats capable of intrastrand base pairing. Two inverted

copies of the telomere resolution target are present in each

concatamer, one on either side of the concatamer junction. Each

telomere resolution target consists of a 20 base pair core domain

absolutely required for resolution and two auxiliary domains which

when present serve to enhance the resolution efficiency (21). The

sequence of the core element is absolutely conserved among

divergent members of the orthopoxviridae, implying that there may











be common transacting factors recognizing this sequence during the

resolution event (23). Although the sequences of the auxiliary

domains are less highly conserved the relative positions of purines

and pyrimidines in these regions is maintained among the

orthopoxviridae (23, 71).

It has been shown that the telomere resolution target functions

as a promoter during the latter stages of the vaccinia virus life cycle

(89, 119). The RNA transcript from the telomere region appears not

to code for protein (53). Sequence analysis of the telomere

resolution target indicates that the critical region appears to be a

good candidate for a late viral promoter (119). Whether the

telomere resolution target functions as an intermediate or late

promoter is unknown, as is the role that transcription plays in the

resolution event (G. McFadden, personal communication).

Transacting Factors Required for Resolution



There is very little information regarding the protein factors

involved in the resolution of vaccinia telomeres. The minimal

enzymatic activities required for resolution are a site-specific

endonuclease and a DNA ligase although proteins capable of











catalyzing strand exchange are also probably involved (118). A

nicking-joining enzymatic activity isolated from purified virions has

been shown to preferentially cleave within the core domain of the

telomere resolution target (61, 72). The nicking-joining activity

comigrates with a 50 kDa polypeptide which has been previously

shown to be a single strand specific endonuclease (61). Although

definitive proof is still lacking, this enzymatic activity is a good

candidate for the site-specific endonuclease necessary for telomere

resolution. A genetic approach has also been taken to identify

factors involved in telomere resolution. Six noncomplementing

temperature-sensitive mutants have been identified which are

incapable of resolving DNA concatamers at the nonpermissive

temperature (10, 74). Five of the six complementation groups have a

corresponding defect in the synthesis of late viral proteins. The

observation that mutants defective in late protein synthesis are also

defective in telomere resolution suggests that a critical factor

synthesized late during infection is required for telomere resolution

(74).

The single temperature-sensitive mutant which is incapable of

telomere resolution but not defective in late gene expression maps to











the small subunit of the vaccinia virus capping enzyme (10).

Therefore, genetic evidence implicates the small subunit of the

vaccinia virus capping enzyme (gene D12L) in the resolution process.

Although no direct biochemical role in resolution has been

demonstrated for the mRNA capping enzyme, it is interesting to

speculate in light of the capping enzyme's function as an initiation

factor for intermediate transcription that transcription of the

telomere resolution target may play a role in resolution (10).

Transcription through the telomere resolution target may open up

the DNA strands and facilitate the formation of a unique secondary

structure that is recognized by the enzymes involved in resolution.

Alternatively, by unwinding the DNA strands, transcription may

indirectly aid the binding of the DNA resolvase (10).

Virion Morphogenesis and Release



Much of the information available on the morphogenesis of

poxviruses comes from electron microscopic examination of cells

infected with viral mutants (20). These observations have led to the

model for morphogenesis depicted in Fig. 2. Morphogenesis begins

within an electron dense region of the cytoplasm known as











viroplasm. Viroplasm is thought to consist of nucleoprotein

complexes which are the result of viral DNA replication, although

direct evidence for this is lacking. The first distinct structure in the

morphogenesis pathway is the appearance of viral crescents.

Crescents consist of two closely opposed lipid bilayer membranes

derived from the host cells intermediate compartment (116). Spike

like projections known as spicules radiate from the convex surface of

the crescent. Spicules are thought to give the crescent its rigid

convex shape (78). Crescents envelope a portion of the viroplasm

producing a spherical membranated structure, containing within it a

granular electron dense material. The next stage in morphogenesis is

thought to be the condensation of the granular material within the

spherical particle to produce an extremely electron dense material

presumably composed of nucleoprotein. There is some evidence that

condensed nucleoprotein may be injected into the spherical particle

before it is sealed, although at present this conclusion is somewhat

speculative (76). Following the condensation or injection of the

nucleoprotein, there occurs a series of complex morphological

changes within the developing particle that results in the formation

of the first infectious form of the virus. The first infectious form is a











membranated brick shaped virion particle that surrounds a

biconcave nucleoprotein containing core. Between the core walls and

the inner surface of the viral membrane are two granular structures

of unknown function known as lateral bodies. Packaged within the

core is the viral DNA and early transcription apparatus (78).

A majority of these infectious particles, known as intracellular

membranated virus (IMV), remain within the cytoplasm. However, a

small subset of IMV particles are actively transported out of the

viroplasm. The IMV particle obtains two additional lipid bilayer

membranes which are derived from the host cell golgi apparatus (49,

101). This triply membranated intracellular enveloped virion (IEV)

is then transported to the plasma membrane. Fusion of the

outermost viral membrane with the plasma membrane liberates the

now doubly membranated viral particle from the cytoplasm into the

extracellular milieu (91). The doubly membranated form, known as

extracellular enveloped virus (EEV) is the form responsible for the

dissemination of the infection in vivo and plaque formation in vitro

(90).











Viral Genetics



The relatively small genome size, lack of introns and

recombinagenic nature of poxviruses make them very amenable to

genetic study (reviewed in 15). The isolation and characterization of

poxvirus mutants has been of fundamental importance in

understanding the biology of the viral infection. These mutants have

proven valuable for studying all aspects of viral replication. In

addition, the isolation of viral mutants has provided poxvirologists

with the reagents necessary for the later biochemical analysis of

specific enzymatic activities encoded by the virus.

There are presently six reported collections of viral mutants,

each of which has been characterized to varying degrees (reviewed

in 15). This discussion will focus on three collections of vaccinia

virus temperature-sensitive mutants, that of Dales and coworkers,

Ensinger and coworkers, and Condit and coworkers.

The Dales collection was isolated in vaccinia virus strain IHD-W

(International Health Department) by chemical mutagenesis of a

wild-type viral stock followed by screening individual plaque

isolates for temperature sensitivity. Ninety temperature-sensitive











mutants were isolated and characterized by electron microscopic

examination of mutant infected cells at the nonpermissive

temperature (40oC). This collection was divided into 17 classes based

upon the stage in viral morphogenesis reached under nonpermissive

conditions (20). Complementation and recombination mapping has

been reported on subsets of this collection, however there has been

no attempt to map the complete collection by either

complementation analysis or marker rescue (31, 69).

Ensinger and coworkers have isolated a set of 100

temperature-sensitive mutants in the WR (western reserve) strain of

vaccinia virus by chemical mutagenesis of a wild-type stock. A total

of 62 mutants have been mapped by complementation analysis or

marker rescue (27-29).

Condit and coworkers have isolated a series of 65 temperature-

sensitive mutants in the WR strain of vaccinia virus comprising 32

separate complementation groups (13, 14). Of this collection, 29

complementation groups have been assigned a position on the viral

chromosome by marker rescue (14, 104, 120). The position of 20 of

these complementation groups are shown in Fig. 1. As expected from

mutations which affect essential genes, mutants in this collection








35

cluster within the centrally conserved core. To localize a mutant to a

particular open reading frame, fine mapping has now been done on a

number of these mutants (12, 34, 50, 70, 87, 95, 121).

Mutant Classification



In addition to the morphological classification of Dales et al.,

biochemical analysis of the phenotypes of temperature-sensitive

viral mutants has been used as a means of classification. Condit and

coworkers divided their collection into four groups, normal, DNA-,

defective late, and abortive late (reviewed in 15). The normal

phenotype includes all viral mutants which display normal patterns

of protein and DNA synthesis at the nonpermissive temperature.

Viruses classified as DNA- are those that are incapable of

[3H]thymidine incorporation at the nonpermissive temperature. The

defective late mutants comprise a very heterogeneous group of

viruses, which at the nonpermissive temperature, display defects in

the synthesis of late viral proteins. The abortive late phenotype is

characterized by an abrupt cessation of late protein synthesis at

approximately 8 hours post infection.











Directed Viral Genetics




The Condit temperature-sensitive mutant collection represents

mutations in only 20% of the essential virus genes (45). With the

publication of the complete sequence of vaccinia virus, it became

clear that there are a number of genes of both known and unknown

function for which there are no temperature-sensitive alleles (41,

54).

The traditional approach to generating temperature-sensitive

vaccinia virus mutants has involved random chemical mutagenesis of

viral stocks followed by screening for temperature sensitivity by

either assaying individual plaques or through a plaque enlargement

protocol. The utility of both methods is limited by requiring the

investigator to screen large numbers of plaques isolates. An

additional problem with the traditional approach to vaccinia virus

genetics is that the random mutagenesis methods employed do not

allow an investigator to target specific open reading frames for

mutational analysis. It is for these reasons that a more direct

approach for the creation of temperature-sensitive mutants must be

found. The goal of directed viral genetics is to target specific viral











open reading frames for mutagenesis in vitro and then place the

mutant allele back into its normal location in the viral genome (15).

For directed genetics to be applied to a complex DNA virus such

as vaccinia, two important criteria must be met. The difficulty in

successfully screening large numbers of viruses necessitates that the

mutagenesis scheme chosen produce a high proportion of

conditionally lethal mutants. The second obstacle that must be

overcome involves the targeted insertion of the mutant allele back

into the viral chromosome, along with the displacement of the wild-

type genomic copy of the gene.

This dissertation attempts to address the question of whether

or not clustered charge to alanine mutagenesis and transient

dominant selection can be used as a general method for the creation

of temperature-sensitive mutations in specific vaccinia virus genes.

Clustered Charge to Alanine Mutagenesis




Clustered charge to alanine mutagenesis is actually a

refinement of a technique known as alanine scanning mutagenesis,

first reported by Cunningham and Wells as an in vitro mutagenesis

method that caused minimal disturbances to the resulting protein's











secondary structure (19). In this report individual amino acids

within a previously defined receptor binding domain of human

growth hormone (hGH) were systematically mutated to alanines. The

authors chose alanine because it is a small aliphatic amino acid

commonly found both on the surface, as well as buried within the

hydrophobic core of a variety of proteins and in a variety of

secondary structures. Therefore the authors felt that single alanine

substitutions would be unlikely to cause any extreme electrostatic or

steric disturbances in the mutant protein, and thus would be unlikely

to grossly alter the mutant protein's secondary structure.

Elimination of amino acid side chains beyond the 6 carbon, by

replacement with alanine, would aid in the identification of those

amino acids whose side chains participate in protein-protein

interactions between hGH and the hGH receptor. Recognition of the

recombinant alanine substitution hGH mutants by monoclonal

antibodies specific for the native structure of hGH demonstrated that

most of these mutants retained the secondary structure of the wild-

type protein (19). The initial reporting of alanine scanning was











followed by a number of other reports which demonstrated the

utility of this method for the examination of both interprotein and

intraprotein interactions (2, 6, 11).

Alanine scanning was later modified by Bennett et al., who

mutated clusters of charged amino acids (Glu, Asp, His, Lys, Arg) to

alanines in an effort to define functional determinants on the surface

of human tissue plasminogen activator protein (htPA) (7). Clusters of

charged amino acids were targeted for mutagenesis because the

authors theorized that highly charged regions in the primary

sequence of a protein would be located on the surface, rather than

buried within the interior of the folded protein. In addition, charged

surface residues are also important mediators of electrostatic

protein-protein interactions, therefore loss of the surface charge

could disrupt these interprotein interactions. Charged clusters were

mutated in groups of 1 to 4 amino acids and clusters spanning

cysteine residues were avoided in order to refrain from disrupting

disulfide bonds which act to stabilize protein secondary structure.

The recombinantly expressed mutant proteins were tested in vitro

for a variety of functional activities and compared to wild-type htPA

protein. These data helped to localize some of the functional domains











involved in htPA's diverse biological activities, which include

activation of plasminogen to plasmin, stimulation of tPA by fibrin,

and inhibition of tPA by tissue plasminogen activator inhibitor

protein (7).

Most of the reports which used clustered charge to alanine

mutagenesis involved examining recombinantly expressed protein or

peptide fragments in vitro. However in 1993 Wertman, Drubin and

Botstein demonstrated that clustered charge to alanine mutagenesis

could be used to create mutations that were temperature-sensitive in

vivo (130). In a mutational analysis of the Saccharomyces cerevisiae

actin gene (ACT1) 16 out of the 34 mutations constructed were

conditionally lethal for colony formation. Of the 16 conditional lethal

mutations, 9 were both cold sensitive and heat sensitive and 6 were

strictly heat sensitive conditional lethals. Examination of the three

dimensional crystallographic structure of yeast actin confirmed that

81% of the clustered charge to alanine mutations created in this

study targeted residues located on the surface of the folded protein

(130).

Clustered charge to alanine mutagenesis has also been used by

Diamond and Kierkagaard to create temperature-sensitive mutations











in the RNA dependent RNA polymerase protein of polio virus (25).

Of the 27 mutations constructed, 9 (35%) yielded viruses which the

authors classified as temperature-sensitive. These viruses were

classified as temperature-sensitive based upon either a small plaque

phenotype or decreased accumulation of viral RNA at the

nonpermissive temperature (39.5oC) (25). Interestingly, all of these

viruses are capable of producing infectious progeny at the

nonpermissive temperature, therefore none of these viruses are

conditional lethal temperature-sensitive mutants.

Following the initial demonstration that clustered charge to

alanine mutagenesis can be used to create temperature-sensitive

mutations in vivo, other investigators have used this technique to

make temperature-sensitive alleles in a number of diverse proteins

including the HIV type 1 integrase gene and the yeast ubiquitin-

conjugating enzyme CDC34 (UBC3) (93, 131). These results have

served to demonstrate that clustered charge to alanine mutants can

be created in proteins that serve structural, as well as enzymatic,

functions.











Transient Dominant Selection




The ability of homologous viral and plasmid DNA sequences to

recombine in vivo has been used to create deletion and insertion

mutants as well as viruses expressing foreign genes (80).

Recombination can also be used as a means of targeted gene

replacement to introduce site-specific mutations into the viral

genome (33). The principal limitation of this technology has been a

low proportion of recombinants (1:1000) among a background of wild-

type nonrecombinant viruses (33). A number of selection and

screening methods have been devised to aid in the identification of

viral recombinants (32, 38, 66). One of the most widley used

techniques for the identification of viral recombinants has been a

transient dominant selection method which relies upon the ability of

recombinant viruses expressing the Escherichia coli xanthine-guanine

phosphoribosyltransferase protein (ecogpt) to grow and form plaques

in the presence of mycophenolic acid (MPA). The antibiotic

mycophenolic acid blocks inosine monophosphate dehydrogenase, a

cellular enzyme in the de novo guanine monophosphate synthesis

pathway (83). The selection for viral recombinants is made possible








43

by the different catalytic activities of the mammalian and bacterial

enzymes hypoxanthine phosphoribosyltransferase and xanthine-

guanine phosphoribosyltransferase. The bacterial enzyme (ecogpt)

will efficiently use xanthine as a substrate for purine biosynthesis

via the salvage pathway, whereas the mammalian enzyme will not

(83). Therefore, when provided with xanthine and grown on an

appropriate cell line, only viruses expressing ecogpt will form

plaques in the presence of MPA.

During this study, this methodology was applied as a means of

targeted gene replacement, the salient features of which are

illustrated in Fig. 5. Infected cells were transfected with a plasmid

that contained tandemly arranged single copies of a mutant vaccinia

gene and the ecogpt gene. The ecogpt gene was under the

transcriptional control of a vaccinia virus promoter (p7.5k) which

drives expression of ecogpt at both early and late times after

infection (33). Single crossover homologous recombination events

integrate the plasmid into the genome and generate a virus which

contains the ecogpt cassette flanked on either side by a copy of the

gene of interest (33). Viruses which have integrated the plasmid

into the genome can be selected for by plating the lysate from the

























Fig. 5. Recombination events occurring within infected cells
during transient dominant selection. A transfected plasmid
containing both the ecogpt gene and a mutant vaccinia gene will
homologously recombine into the viral genome during the infection
transfection experiment. The mycophenolic acid resistant
recombinant virus produced contains the ecogpt gene flanked on
either side by a copy of the gene of interest. During the growth of
each mycophenolic acid resistant plaque, the virus will resolve the
gene duplication generating viruses which have either retained
(bottom left) or lost (bottom right) the mutation.












vaccinia gene


wild type
vaccinia virus
chromosome




K-


I
infection-transfection



mycophenolic acid resistant recombinant


resolution


wild type vaccinia gene


FIIIIIIIIIIIIII


mutant vaccinia gene











infected, transfected cells in the presence of MPA (33). Duplication

of the cloned gene makes these mycophenolic acid resistant viruses

unstable, so that during the growth of each MPA resistant plaque,

resolution will occur via recombination between the two genomic

copies of the gene of interest. The products of the resolution event

will be viruses which have lost the ecogpt gene and contain only a

single copy of the gene of interest. Resolution occurs during the

growth of each MPA resistant plaque such that each MPA resistant

plaque contains a mixture of viruses, some of which have a mutant

vaccinia gene derived from the transfected plasmid, while others

retain a wild-type copy of the vaccinia gene. Unresolved viruses

which retain the ecogpt gene will presumably serve a helper

function for growth of resolved viruses in the presence of MPA.

This study will address a number of important questions about

transient dominant selection including: 1) Can transient dominant

selection be used to insert site-specific mutations into essential

vaccinia genes? 2) What is the percentage of mutant viruses within

each MPA resistant plaque?; 3) How many total viruses need to be

screened for temperature sensitivity before an unknown mutation

can be adjudged to not confer a temperature-sensitive phenotype?








47

A protocol, which is outlined in Fig. 6, has been developed to answer

these questions by creating mutations in the vaccinia virus gene G2R,

a gene with a selectable phenotype. The advantages of using the

vaccinia G2R gene to determine the efficiency of both clustered

charge to alanine mutagenesis and transient dominant selection is

discussed in detail in other sections of this dissertation (see chapter 1

and introduction to chapter 3). Briefly, a transient dominant

selection experiment was done with a plasmid containing a known

conditional lethal, temperature-sensitive G2R allele (Cts56).

Individually isolated MPA resistant plaques were then tested for the

presence of the Cts56 allele by plating at 31oC and 40oC in the

presence and absence of isatin-B-thiosemicarbazone (IBT). Wild-

type viruses will not grow in the presence of IBT. Loss of G2R

function results in viruses which are IBT dependent, therefore Cts56

is dependent upon IBT for growth at 40oC. The Cts56 virus is also

resistant to IBT at 31oC. The phenotype of individual IBT resistant

plaques can then be determined by re-plating them at 31oC and 40oC

in the presence and absence of IBT. The isolation of viruses with the

Cts56 phenotype proves that transient dominant selection can be

























Fig. 6. Testing the efficiency of transient dominant selection
by making mutations in G2R. Mycophenolic acid resistant
plaques are isolated from a transient dominant selection experiment
done with wild type vaccinia virus and a plasmid which contains the
ecogpt gene and a mutant copy of G2R. Viruses within the
mycophenolic acid resistant plaque that contain a mutation that
alters G2R function can be selected for by plating individual
mycophenolic acid resistant plaques in the presence of IBT. The
efficiency of transient dominant selection can be measured by
determining the percentage of viruses containing the mutation (+IBT)
among the total population (31oC -IBT).











Ecogpt


infection transfection


wild type
vaccinia virus


Mutant vaccinia
gene


I select for recombinants
with mycophenolic acid


pick and test
individual plaques


-IBT 310o C +IBT 31o C +IBT 400 C -IBT 40 oC




Test the phenotype of individual IBT
resistant plaques by replating at 31 C
and 400C + and IBT











used to insert site-specific conditionally lethal mutations into the

viral genome. In addition, the proportion of IBT resistant viruses

within individual MPA resistant plaques will indicate how many total

viruses from each transient dominant selection experiment involving

a clustered charge to alanine mutant of unknown phenotype will

need to be screened for temperature sensitivity. This information

will be used to determine how to modify the protocol in order to

screen clustered charge to alanine mutations in genes without

selectable phenotypes. In these cases progeny from individual MPA

resistant plaques will be isolated by plated at 31oC and screened

individually by plating at 31oC and 40oC.

Capping of mRNA and the Vaccinia mRNA Capping Enzyme




The nucleus is the site of eukaryotic mRNA synthesis and

maturation (75). Initially mRNAs are synthesized by RNA

polymerase II in the form of precursor molecules known as pre-

mRNA or heterogeneous nuclear RNA (hnRNA). Pre-mRNAs undergo

extensive posttranscriptional modifications within the nucleus,

including 5' terminal capping, 3' terminal polyadenylation, splicing

and internal methylations, before being transported to the cytoplasm











and entering the pool of mature translationally competent mRNA

(75). The modifications occurring to the 5' end of eukaryotic mRNAs

are of particular interest because mRNA is synthesized, translated

and possibly degraded in a 5' to 3' direction (75).

The 5' end of a mature eukaryotic mRNA molecule consists of a

cap structure, m7GpppN, in which the first encoded nucleotide is

linked to a 7-methylguanosine residue by a unique 5'-5'

phosphodiester bond (105). The cap structure of mRNA is necessary

for efficient translation initiation and protects the mRNA from

degradation by 5'-3' exonucleases (105, 106). In addition, the

presence of a cap structure on nuclear mRNA facilitates mRNA

splicing and nuclear export (43, 57, 59). Capping of pre-mRNA

occurs cotranscriptionally and is one of the initial steps in the

posttranscriptional mRNA maturation pathway (75).

The formation of the mRNA cap structure (m7GpppA) occurs in

three sequential enzymatic reactions (reviewed in 75, 105). The 5'

end of the triphosphorylated primary mRNA transcript is first

cleaved to a diphosphorylated form by an mRNA triphosphatase and

then capped with guanosine monophosphate by an mRNA

guanyltransferase. During the guanyltransferase reaction, a stable











intermediate forms via a covalent phosphoamide bond between a

lysine residue in the guanyltransferase active site and the phosphate

moiety of guanosine monophosphate (75, 110). Following

guanylation the guanine cap is methylated at the N7 position by a

(guanine-7)methyltransferase (75). This general mechanism for

capping mRNA occurs in vaccinia, reovirus, cytoplasmic polyhedrosis

virus and probably most eukaryotic cells, although cap formation in

vesicular stomatitis virus is thought to occur by the transfer of a

guanosine diphosphate to the 5' end of a monophosphorylated

primary mRNA transcript (reviewed in 75).

Methylation of the nucleotides adjacent to the 7-

methylguanosine residue occurs to varying extents, and this provides

a basis for the following cap nomenclature system: m7GpppN (cap 0),

m7GpppNm (cap 1), m7GpppNmpNm (cap 2), m7GpppNmpNmpNm (cap 3)

and m7GpppNmpNmpNmpNm (cap 4). Generally these methylation

reactions are thought to occur sequentially, although there is

evidence that at least in the case of vesicular stomatitis virus and

cytoplasmic polyhedrosis virus the 2'O-methylation of the

penultimate nucleotide can occur prior to the methylation of the

guanosine cap (reviewed in 75). The cap 0 structure, which is the








53

minimal cap structure required to facilitate translation initiation, is

found on the 5' end of mRNA isolated from yeast, slime molds and

some viral mRNAs (37, 62, 128), whereas mRNAs containing a cap 1

structure have been isolated from slime molds, animal cells and

many animal viruses (62). The cap 2 structure is found on mRNA

from insects, vertebrates and their viral parasites (62).

Interestingly, only cap 1 mRNA is found in the nucleus, implying that

the methyltransferases responsible for forming the other cap

structures are confined to the cytoplasm (62).

Because the enzymatic machinery necessary for capping mRNA

is localized within the nucleus, many cytoplasmic viruses including

vaccinia have evolved their own mRNA capping enzymes (75). All of

these viruses package capping enzymes within the viral particle,

along with a virally encoded RNA polymerase which ensures the

capping of mRNAs synthesized during the early phases of the viral

life cycle. The packaging of a viral capping enzyme and a viral RNA

polymerase within the virion has been used to develop in vitro

transcription/capping assays, whereby the mechanism of mRNA

capping has been examined in detail (reviewed in 39, 75). In the

case of vaccinia, incubation of purified virus particles in vitro with a











nonionic detergent, a reducing agent, all four ribonucleoside

triphosphates, magnesium and S-adenosyl-L-methionine results in

the formation of early viral mRNA molecules which contain cap 0 or

cap 1 structures at their 5' ends (62). The vaccinia mRNA capping

enzyme catalyzes the formation of the cap 0 structure, and the

methylation of the penultimate nucleotide is done by the small

subunit of the vaccinia polyA polymerase protein (102, 115, 124).

The multifunctional nature of many cytoplasmic viral capping

enzymes is in contrast to cellular capping enzymes. The L protein of

vesicular stomatitis virus is thought to contain the guanyltransferase,

(guanine-7)methyltransferase and (nucleoside-2'O-)

methyltransferase active sites, in addition to serving as a subunit of

the RNA dependent RNA polymerase (128). The NS3 protein of

flaviviruses functions as a protease, a helicase and an mRNA

triphosphatase (129). Among avian reoviruses, a single protein

encodes both RNA dependent RNA polymerase and guanyltransferase

activities (117). As previously discussed, the vaccinia capping

enzyme functions as an early transcription termination factor and an

intermediate transcription initiation factor. The biochemical

activities of cellular capping enzymes seem to be confined to capping








55

mRNA. In rat liver, yeast, and brine shrimp the RNA triphosphatase

and guanyltransferase activities remain tightly associated during

multiple chromatographic purification steps. However (guanine-

7)methyltransferase activity does not coelute with the RNA

triphosphatase-guanytransferase containing fractions (75). Single

polypeptides with RNA triphosphatase and guanyltransferase

activities have been isolated from brine shrimp and rat liver (75). In

contrast, the yeast Saccharomyces cerevisiae contains two tightly

associated polypeptides, an 80 kDa RNA triphosphatase and a 52 kDa

guanyltransferase (75).

The only nonviral capping enzymes which have been cloned

are those from Saccharomyces cerevisiae (CEG1) and

Schizosaccharomyces pombe (PCE1) (103). The close association of

the RNA triphosphatase and guanyltransferase activities in many

eukaryotic capping enzymes may indicate that the RNA

triphosphatase and guanyltransferase reaction are coupled in vivo.

Vaccinia Virus mRNA Capping Enzyme



Although the specific functions of the vaccinia virus mRNA

capping enzyme have been previously discussed in detail in earlier








56

sections of this dissertation it may be useful to review what is known

about the vaccinia virus mRNA capping enzyme. The vaccinia virus

mRNA capping enzyme is a heterodimer composed of a 97 kDa large

subunit and a 33 kDa small subunit encoded by the vaccinia genes

D1R and D12L respectively (77, 86). The heterodimer contains

mRNA triphosphatase, guanyltransferase and (guanine-

7)methyltransferase enzymatic activities (108, 115, 124). The

domain structure of the enzymatic activities involved in capping

mRNA are fairly well defined, with the mRNA triphosphatase and

guanyltransferase active sites being localized to the amino terminal

60 kilodaltons of DIR, and the (guanine-7)methyltransferase domain

being contained within a stable heterodimer formed between the

carboxyl terminal portion of DIR and an intact D12L protein (107,

112).

During the guanyltransferase reaction, guanosine

monophosphate becomes covalently linked to a lysine residue (K260)

in the large subunit forming a enzyme-GMP intermediate. The

guanyltransferase active site has been mapped to a conserved KTDG

motif at amino acids 260-263 in DIR (17, 84). The KTDG motif is part











of a larger, complex motif associated with nucleotidyl transferases

such as capping enzymes and DNA and RNA ligases (111).

Heterodimerization between DIR and D12L and wild-type

levels of methyltransferase activity have been detected by

coexpression in bacteria of a plasmid encoding amino acids 498-844

of DIR and a full length D12L protein (46). However, the detection of

low levels of methyltransferase activity by expression of only amino

acids 540-844 of DIR indicates that the (guanine-7)

methyltransferase active site resides solely within the carboxyl

terminal 304 amino acids of DIR (16). Nevertheless, it is clear from

this data that heterodimerization between DIR and D12L is required

for full methyltransferase activity. A methyltransferase motif

UU[D/E]xGxGxG, where U is an aliphatic amino acid and x is any

amino acid, was identified in DIR that bears homology to motifs

found in the methyltransferases of flaviviruses, reoviruses and the

Hhal methyltransferase (68). All of these methyltransferases utilize

S-adenosyl-L-methionine as a substrate, and this motif is presumed

to be involved in binding S-adenosyl-L-methionine. The vaccinia

D1R sequence AIDFGNG (amino acids 596-602) lacks the third glycine

residue of the consensus sequence, however this sequence has










nevertheless recently been shown to be essential for

methyltransferase activity in vitro (68). A number of

alaninescanning mutants have been generated in recombinantly

expressed D1R and examined for DIR-D12L subunit association and

methyltransferase activity (68). One of these mutations

(H682A,Y683A) is defective in methyltransferase activity, but not

subunit association, implying that these two amino acids form part of

the methyltransferase active site. S-adenosyl-L-methionine has also

been covalently photocrosslinked to amino acids 499-579 and 806-

844, implicating these residues in the binding of S-adenosyl-L-

methionine (48).

The vaccinia virus mRNA capping enzyme also functions as an

early gene transcription termination factor and an intermediate gene

transcription initiation factor (109, 126). Transcription termination

of early genes requires a full length intact heterodimer and the

sequence UUUUUNU in the nascent mRNA (64, 114). The mechanism

of early termination is not known but it is thought that the capping

enzyme catalyzes termination by protein-protein interactions with

the viral transcription elongation complex. A recombinantly

expressed guanyltransferase defective capping enzyme mutant is










able to catalyze early transcription termination in vitro, therefore

termination is independent of cap formation. The capping enzyme

forms a stable complex with the viral RNA polymerase prior to

binding of the polymerase complex at the intermediate promoter

(126). The formation of the capping enzyme-RNA polymerase

complex is necessary for transcription initiation (126).

The vaccinia capping enzyme has also been genetically

implicated in the resolution of vaccinia virus telomeres (10). A

temperature-sensitive mutation mapped to the small subunit of the

vaccinia virus mRNA capping enzyme (gene D12L) which exhibits

normal protein and DNA expression under nonpermissive conditions

is unable to resolve viral telomeres at the nonpermissive

temperature (40oC) (10). Although the phenotype of this mutant

genetically implicates the vaccinia mRNA capping enzyme in

telomere resolution, there is no biochemical evidence that the

capping enzyme participates either directly or indirectly in the

resolution of viral telomeres.













CHAPTER 2
MATERIALS AND METHODS





Recombinant DNA Techniques




Bacterial Strain and Methods



Escherichia coli strain JM101 (SupE, thiA(lac-proAB)

F'[traD36,proAB+,laclq lacZAM15] was used as host to propagate all

plasmid, phage and phagemid constructs. Standard techniques that

have been previously described were followed for the growth,

maintenance and isolation of all plasmids, phage and phagemid

constructs (100).

Electroporation of DNA into JM101



Electrocompetent bacterial cells were prepared from log phase

cultures of JM101 (OD600 between 0.5-1.0) grown in Lennox L broth.








61

Cells were pelleted at 5,000 x g for 15 min in a GS3 rotor (Sorvall) at

4oC, washed twice with 1 liter of ice cold distilled water, re-pelleted

at 5000 x g, resuspended in 5 ml of cold 20% glycerol and stored at -

70oC. Electroporations were done with a Biorad gene pulser set at 2.5

kV, 25 igF, 200 Q (average time constant of 4-5 msec, field strength

of 12.5 kV) in 1 cm2 cuvettes, using a protocol recommended by the

manufacturer.

Construction of G2R and DIR Vectors



Mutant G2R containing plasmids for transient dominant

selection were constructed using two similar protocols. For wild-type

G2R, AS1, and Cts56, a plasmid pUC118G was constructed, which

consists of the phagemid vector pUC118 (125), a cassette containing

the constitutive p7.5k vaccinia promoter upstream from the

Escherichia coli guanine phosphoribosyltransferase (ecogpt) gene

(obtained as a 1.1 kb Hpal, Dral fragment from pTKgpt-Fls (33)),

and a 982 base pair PvuII, EcoRV fragment derived from the

vaccinia virus HindIII G fragment, containing the wild-type G2R gene

with 82 nucleotides of 5' flanking sequence and 236 nucleotides of 3'

flanking sequence (70). The G2R containing DNA was placed adjacent








62

to and upstream from the p7.5k promoter, with the G2R sequences in

the opposite transcriptional orientation relative to the p7.5k

promoter. Single stranded pUC118G phagemid DNA (125) was used

as a template for mutagenesis to construct mutants AS1 and Cts56,

and transient dominant selections were done using double-stranded

plasmid DNA containing either wild-type or mutant G2R sequences.

For AS2 through AS9, mutagenesis of G2R was done using single

stranded DNA from an M13mpl9 G2R clone, provided by J. Lewis,

containing only the coding sequence of G2R obtained by PCR

amplification. Mutant G2R genes were then subcloned into a plasmid

containing pGEM3ZF- (Promega) vector DNA and the p7.5k ecogpt

cassette described above for conferring resistance to mycophenolic

acid. The G2R gene was placed adjacent to and upstream from the

p7.5k promoter, with the G2R sequences in the same transcriptional

orientation relative to the 7.5K promoter. The transcription start site

of G2R has been previously mapped to base 10,335 of the vaccinia

virus strain WR HindlII G fragment, therefore none of the cloned G2R

sequences used in this study contain the promoter for the G2R gene

(70).











The phage M13mpl8DIRNco- used as a template for all D1R

mutagenesis experiments was a gift from E. Niles. The phage

M13mpl8DIRNco- contains the complete DIR open reading frame in

which an internal Ncol restriction endonuclease site at position

+2,235, relative to the transcriptional start site, has been removed by

site directed mutagenesis using an oligonucleotide which disrupts the

Ncol recognition sequence without altering the DIR coding sequence

(47). The plasmids used in the transient dominant selection

experiments with DIR were made using the following protocol. The

D1R gene was subcloned as a 2.8 kb PstI SacI fragment from

appropriately mutagenized M13mpl8DIRNco- clones into the 7.5K-

ecogpt containing vector pBSgptB cut with PstI and Sacl. The

plasmid pBSgptB, a gift from P. Turner, was made by inserting a 2 kb

EcoRI fragment, containing the p7.5k ecogpt cassette from pTK61-gpt

(33) into the EcoRI site of pBluescriptKS(+) (Stratagene). Mutant

copies of DIR were inserted upstream from and in the same

transcriptional orientation as the p7.5k ecogpt cassette. None of the

cloned DIR sequences used in this study contained the promoter for

the DIR gene (85).











Mutagenesis of G2R and D1R



Clustered charge to alanine mutations were created in cloned

copies of G2R and DIR using oligonucleotide directed mutagenesis.

The sequence of all G2R and DIR mutagenic oligonucleotides are

shown in Tables 1 and 2, respectively. All G2R mutations were made

using the Amersham oligonucleotide mutagenesis kit according to the

manufacturers instructions. Oligonucleotide directed mutagenesis on

D1R was performed using the T7-gen in vitro mutagenesis kit

according to the manufacturers instructions (U. S. Biochemicals).

Mutagenic oligonucleotides, synthesized by the DNA synthesis

core facility of the University of Florida, were designed to include 12

nucleotides 5' and 9 nucleotides 3' of the mismatched region (19).

DNA Sequencing



All DNA sequencing was done using the Sequenase version 2.0

kit according to the manufacturers instructions (U.S. Biochemicals)

and appropriate oligonucleotide primers. The complete G2R sequence

of each mutant phage or phagmid was confirmed byDNA sequence

analysis. The G2R gene from mutant viral DNA was amplified by









65


O






>O Z > .
o g | a g|
5 6 !sA 5dlS
4-0


oo
0
CU)


IU A.





aS -iS


0 U ;

o
J o O S *



a 0




U u U U U "U
o < < 3








a 4) 0 5 <
3 0 u H 0
0 0 .nf
C3 u : 9






S- a a a' a *






o m w r -



0E
r5 -
C' ^ S C', S C', S^^ 0




















<< f

A A

2 S

Q n


0
.0


0
0


0



so



(U
c
0,



0o





U
.2.
0


.0>

0S -.
ll









0


S.-
-.





-.
0o





JO
*



0
-p.0

2 2U
0










PCR, and appropriate selected regions of the G2R PCR product were

sequenced to confirm the presence of the desired mutations in Cts56,

ASI, AS4, AS6, and AS9 viruses. The presence of the mutant

sequence in the DIR phage clones was confirmed by DNA sequencing

of selected regions of D1R surrounding the mutation of interest. DNA

sequence analysis of PCR amplification products from portions of the

D1R open reading frame confirmed the presence of the E516 and

E793 mutation in E516 and E793 viral DNA, respectively. The E126

and D326 mutations created novel NheI and NotI restriction enzyme

sites respectively, in DIR. Portions of the DIR gene from either E126

or D326 viral DNA were amplified by PCR and subjected to

restriction enzyme digestion and agarose gel electrophoresis to

confirm the presence of the appropriate mutations in the E126 and

D326 viral DNA (data not shown) (100).

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis



Protein samples in Ix Laemmli sample buffer were boiled and

separated on 10% SDS polyacrylamide slab gels as described (60).

Following electrophoresis, the gels were stained with Coomassie blue,

destined in 7.5% glacial acetic acid, 5% methanol, and dried










according to standard techniques (100). Autoradiograms of

radiolabeled protein samples were made using Kodak XAR film at -

70oC.

Northern Blot Analysis



Five ug of each purified RNA sample was denatured in

formamide and separated on a 1.2% formaldehyde agarose gel using

standard techniques (100). The RNA was transferred from the gel

onto a GeneScreen membrane in phosphate buffer (136 mM NaH2PO4,

6.4 mM Na2HPO4) overnight, according to the method of Southern

(117). The RNA was fixed to the membrane by irradiation at 254 nm

in a Stratalinker (Stratagene) for 1 minute. The membrane was

prehybridized in 7 ml of a solution containing 50% formamide, 10%

dextran sulfate, 1.45 M NaCI, 100 ug/ml denatured salmon sperm

DNA, 2X P buffer (IX P buffer contains 1% bovine serum albumin

(BSA, Sigma fraction V), 1% polyvinylpyrrolidone (MW 40,000), 1%

ficoll (MW 40,000), 0.25 M Tris-HCl [pH 7.5], 0.5% sodium phosphate,

5% SDS) for 3 hours at 55oC. Hybridizations were done by adding

1x106 cpm of the appropriate radiolabeled riboprobe to the

prehybridizing membrane and incubating at 55oC overnight.










Nonspecifically bound radioactivity was removed by washing the

membrane once in 0.1X SSC (IX SSC is 150 mM NaCI, 15 mM sodium

citrate), 0.1% SDS for 15 minutes at room temperature and twice for

30 minutes each in 0.1X SSC, 1% SDS at 65oC. Autoradiograms were

made using Kodak XAR film at -70oC.

Preparation of Radiolabeled Riboprobes



Plasmids used to prepare antisense riboprobes specific for the

vaccinia virus CliR, G8R and F17R mRNA transcripts were a gift

from B. Moss, and their construction has been described elsewhere

(3). Briefly, 678, 530 and 682 nucleotides of coding sequence

corresponding to the 5' ends of vaccinia virus genes CllR, G8R and

F17R respectively, were each inserted into the vector plasmid

pGEM3ZF (Promega). The plasmids were linearized by restriction

enzyme digestion with EcoRI (C11R and G8R) or BamHIl (F17R) and

subjected to transcription with T7 RNA polymerase (Promega) to

generate antisense RNA molecules of 316 (C11R), 300 (G8R), or 126

(F17R) nucleotides in length. Each 20 ul transcription reaction

contained 1 ug of linearized plasmid DNA, 40 mM Tris-HCl [pH 7.5], 6

mM MgCl2, 2 mM spermidine, 10 mM NaCl, 5 mM dithiothreitol, 0.5










mM each of ATP, GTP and UTP, 12 uM unlabeled CTP, 50 uCi

[a32P]CTP (3,000 Ci/mmol, Amersham), 20 units of RNasin (Promega),

and 20 units of T7 RNA polymerase (Promega). Following a 60

minute incubation at 37oC, 20 units of RNAse free DNAse (Promega)

was added to the reaction and the 37oC incubation was continued for

an additional 15 minutes. The labeled RNA was purified by

extraction with phenol-chloroform (1:1), precipitated with

ammonium acetate and isopropanol and quantitated by liquid

scintillation counting according to standard techniques (100).

Western Blot Analysis of E793 Virions



Transfer of proteins from 10% sodium dodecyl sulfate

polyacrylamide gels to nitrocellulose was done according to the

method of Towbin (122). Rabbit polyclonal antiseras for D1R and

D12L, kindly provided by E. Niles and S. Shuman, respectively, were

each used at a dilution of 1:2000. The primary antibody was detected

by using horseradish peroxidase linked polyclonal anti-rabbit

antisera (Amersham) at a dilution of 1:10,000 and visualized by

Amersham's enhanced chemiluminescence western blotting kit

according to the manufacturers instructions.








71

Virion Directed In Vitro Transcription and Methyltransferase Assays



Purified vaccinia virus virions were preincubated at 30oC, 35oC

or 40oC in 100 ul reactions containing 13 ug/ml virus, 0.01 M Tris-

HCL [pH 8.5], 0.01% nonidet P40, 0.02 M dithiothreitol, 0.2 M MgCI2.

At 15 minute intervals 10 ul samples were removed and held on ice

until the conclusion of the preincubation time course. For the

transcription reactions, 10 ul of a solution containing 10 mM ATP, 2

mM GTP, 2 mM CTP, 0.66 mM unlabeled UTP, 0.4 uCi [3H]UTP (14

Ci/mmol, Amersham) was added to each preincubated sample and

transcription reactions were carried out at 30oC for 30 minutes. For

methyltransferase reactions, 10 ul of a solution containing 10 mM

ATP, 2 mM GTP, 2 mM CTP, 2 mM unlabeled UTP, 0.66 uCi [3H]S-

adenosyl-L-methionine (15 Ci/mmol, Amersham) was added to each

preincubated sample and methyltransferase reactions were carried

out at 30oC for 30 minutes. The reactions were terminated by

spotting 15 ul onto a DE81 filter (Whatman). Unincorporated

radioactivity was removed by washing in 0.5 M Na2HPO4, after which

the filters were analyzed by liquid scintillation counting according to

established methods (100). Wild-type and E793 transcription and










methyltransferase reactions were normalized to identical reactions

carried out with wild-type or E793 virus that had not been

preincubated and the results expressed as the percent of activity

relative to non-preincubated controls.



Virological Techniques




Cells and Virus



BSC40 and CVI cells were routinely maintained at 37oC in 5%

CO2. BSC40 cells were grown in Dulbecco's Modified Eagle medium

(DME, Life Technologies) with 10% Hyclone (Hyclone Laboratories).

CV1 cells were maintained in CV1 medium, which is Dulbecco's

modified Eagle medium supplemented with 1% nonessential amino

acids (Sigma), 1% sodium pyruvate (Sigma) and 10% fetal calf serum

(Life Technologies). Stocks of wild-type vaccinia virus strain WR

were grown on BSC40 cells at 37oC. Temperature-sensitive vaccinia

virus mutants were grown on BSC40 cells at 31oC. The isatin-B-

thiosemicarbazone dependent mutant G2A was grown on BSC40 cells

at 37oC in the presence of 45 tM IBT (Pfaltz and Bauer).











Preparation of IBT



A 1 mg/ml stock of IBT in acetone was diluted with 4 volumes

of 0.25 N NaOH and further diluted 1:100 in DME+10% Hyclone+l%

Noble agar (Difco) for a final concentration of 45 jtM. IBT containing

media was added to infected BSC40 cells following the 60 minute

absorption period and remained in the media throughout the course

of the infection.

Plaque Titration



Viral stocks were thawed, sonicated in a Vibra cell sonicator

with a cuphorn attachment (Sonics and Materials) for 1 minute and

serially diluted in PBSAM (170 mM NaCI, 3.4 mM KCI, 10 mM

Na2HPO4, 1.8 mM K2HPO4 [pH 7.2], 10 mM MgCl2, 0.01% BSA).

Confluent monolayers of BSC40 cells on 60 mm dishes were infected

with 0.5 ml aliquots of each viral dilution for one hour at either 31oC

or 400C. The inocula were removed and each dish was overlaid with

4 ml of DME+10% Hyclone containing 1% molten noble agar (Difco),

allowed to solidify and incubated at 31oC, or 40oC for 6 days. The

plaques were stained by overlaying the cells with 4 ml of DME+10%











Hyclone, 1% molten Noble agar, 0.005% neutral red (3-Amino-7-

dimethylamino-2-methylphenazine hydrochloride, Sigma) and

incubating at 31oC or 40oC from 3 hours to overnight. For permanent

records of wild-type and mutant plaques, the Noble agar overlay was

removed after staining with neutral red and the cell monolayers

were stained with a solution of 0.25% crystal violet (Sigma) in 20%

ethanol at room temperature for a few hours. The dishes were then

rinsed with distilled water and air dried.

Viral Plaque Isolation



Infected monolayers of BSC40 or CV1 cells were stained with

neutral red as described above to visualize individual plaques. Well

isolated single plaques were picked with a sterile pasteur pipette and

placed in 3 ml of PBSAM, frozen and thawed 3 times and then plaque

titrated as previously described.

Preparation of Viral Stocks



Confluent monolayers of BSC40 cells, on 100 mm dishes, were

infected with the appropriate virus at a low multiplicity of infection

(moi) (between 0.01 and 0.1 plaque forming units per cell) and











incubated at either 37oC (wild-type) or 31oC (mutants) until the

cytopathic effect was complete. Cells were scraped from the dishes

into the media and pelleted by centrifugation at 400 x g for 5

minutes at 4oC. The cell pellets were resuspended in 0.1-1.0 ml of

PBSAM/lx107 infected cells, frozen and thawed 3 times and stored

at -70oC.

Transient Dominant Selection



Transient dominant selection was performed essentially as

described by Falkner and Moss (33). CV1 cells on 60 mm dishes

were pretreated overnight in CV1 media containing MPA, (15 ug/ml

hypoxanthine, 250 ug/ml xanthine, 25 ug/ml mycophenolic acid,

Sigma) infected with wild-type vaccinia virus (moi of 0.05 plaque

forming units per cell), transfected with 20 ug of plasmid DNA using

Transfectase (Life Technologies) according to the manufacturers

instructions, and incubated at 37oC for three days. Infected,

transfected cells were harvested by scraping from the dish into the

media, pelleted by centrifugation at 400 x g for five minutes,

resuspended in 1 ml of PBSAM, frozen and thawed three times and

stored at -70oC. The samples were plaque titrated on CV1 cells in the











presence of MPA for 3 to 5 days. The plaque titrations were done

essentially as described above except that CV1 cells and media were

used and 0.5% Sekem HGT agarose was used in place of 1% Noble

agar.

For the G2R mutant screen, 10 individual MPA resistant

plaques from each transfection were picked and plaque titrated on

BSC40 cells in the presence and absence of 45 jtM IBT at 31oC and

400C. To determine their phenotypes, individual IBT resistant

plaques were picked, then re-plaqued in the presence and absence of

IBT at 31oC and 40oC. Individual G2R mutant viruses were then re-

plaque purified, and viral stocks were grown as described above. For

the DIR mutant screen, five individual MPA resistant plaques from

each transfection were plaque titrated on BSC40 cells at 31oC. Fifty

individual plaques, ten from each original MPA resistant plaque,

were picked, then re-plaqued at 31oC and 40oC. Individual plaques

from dishes showing either heterogeneity in plaque size or reduced

plaque size in comparison to wild-type were re-plaque purified at

31oC, and viral stocks prepared as previously described.










One Step Growth Experiments



Confluent monolayers of BSC40 cells on 60 mm dishes were

infected at 31oC or 40oC with the indicated virus at an moi of 6. At

various times post infection, infected cells were scraped from the

dishes into the media and frozen and thawed 3 times. Viral yields

were determined by plaque titration on BSC40 cells at 31oC. The

data is expressed as the number of plaque forming units produced

per infected cell (pfu/cell).

Marker Rescues



Confluent monolayers of BSC40 cells on 60 mm dishes were

infected with the mutant virus indicated in the figure, incubated at

310C for four hours, transfected with 20 ug/ml of the appropriate

DNA in HBS (125 mM CaCI2, 150 mM NaCl, 0.7 mM Na2HPO4, 5 mM

KCI, 20 mM N-[2-hydroxyethyl] piperazine-N-[2-ethanesulfonic acid]

(HEPES), 6 mM Dextrose) and shifted to 40oC for three days. The cells

were stained with crystal violet as previously described. The DNA

constructs used in the marker rescue experiments are described in

the figure legends.










Purification of Vaccinia Virus Virions



5x108 BSC40 cells, on 10-150 mm dishes, were infected at an

moi of between 0.01 and 0.1 and incubated at either 37oC (wild-

type) or 31oC (E793) until complete cytopathic effect was observed.

Cells were scraped from the dishes into the media, pelleted by

centrifugation at 400 x g for 10 minutes, washed with 20 ml of PBS

(170 mM NaCl, 3.4 mM KCI, 10 mM Na2HPO4, 1.8 mM K2HPO4 [pH

7.2]) and resuspended in 20 ml of 10 mM Tris-HCl [pH 7.5], 5 mM

MgCl2. The cells were lysed by dounce homogenization and the

nuclei were removed by pelleting at 400 x g. After sonication of the

supernatant fraction, the virus within the sample was pelleted

through a 36% w/v sucrose, 10 mM Tris-HCl [pH 7.5] cushion by

centrifugation at 13,500 rpm in an SW28 rotor (Beckman) at 4oC for

40 minutes. Viral pellets were resuspended in 6 ml of 10 mM Tris-

HCL [pH 7.5], resonicated and spun through a 25%-40% w/v sucrose,

10 mM Tris-HCl [pH 7.5] gradient at 13,500 rpm in an SW28 rotor at

4oC for 40 minutes. Viral bands were collected, diluted with 2

volumes of 10 mM Tris-HCl [pH 7.5] and pelleted at 15,000 rpm in an

SW28 rotor at 4oC for 60 minutes. The pellet was resuspended in 4










ml of 10 mM Tris-HCl [pH 7.5]. The purified virus was quantitated

by Bradford assay using a Biorad protein assay kit, according to the

manufacturers instructions, and absorbance measurements at 260

nm (1 OD260 = 64 ug virus). The infectivity of the purified virus was

assayed by plaque titration on BSC40 cells at 31oC. There was no

significant difference in either the quantity or infectivity of the E516

or E793 virus preparations compared to wild-type preparations.

In Vivo Protein Pulse Labeling of Infected Cells



Confluent monolayers of BSC40 cells on 35 mm dishes were

infected at an moi of 15 with the indicated virus at 31oC or 400C.

After a 30 minute absorption the inoculum was removed, 2 ml of

medium prewarmed to the appropriate temperature was added to

each dish, and the incubations were continued. At various times

after infection the media was removed and the infected cells were

washed three times with 2 ml of prewarmed PBS. Proteins within

the infected cells were pulse labeled for 15 minutes with 10 uCi of

Trans[35S]-label (1,153 Ci/mmol, ICN Biochemical) in 0.5 ml of

prewarmed PBS. The label was removed and the cells were lysed in











0.3 ml of Laemmli sample buffer (60), scraped from the dishes,

boiled and frozen at -20oC.

Viral DNA Accumulation in Infected Cells



Confluent monolayers of BSC40 cells on 60 mm dishes were

infected at a moi of 15 with either wild-type or E793 virus and

incubated at 31oC or 40oC. At 2 hour intervals after infection, the

cells were washed with 1 ml of PBS, scraped from the dishes into 1

ml of PBS, and pelleted by centrifugation at 400 x g for 10 minutes.

The cells were resuspended in 0.3 ml of 10X SSC containing 1 M

ammonium acetate. Samples were frozen and thawed three times

and diluted with 1.2 ml of 10X SSC. 100 ul of each diluted sample

was denatured by the addition of 2.5 ul of 10 N NaOH and 10 ul of 5

M NaCI (final concentration 0.25 N NaOH, 0.5 M NaCl) for 10 minutes

at room temperature. After denaturation 1 ul was removed, diluted

1:100 in 0.1X SSC, 0.125 N NaOH and applied to a GeneScreen

membrane (New England Nuclear), using a Schleicher and Schuell slot

blotting apparatus according to manufacturers instructions. DNA was

fixed to the membrane by irradiation at 254 nm in a Stratalinker

(Stratagene) for 1 minute. The membrane was prehybridized in 10











ml of prehybridization buffer (100 mg/ml dextran sulfate (MW

50,000), 1% SDS, 1.45 M NaCL) at 65oC for 4 hours. Hybridizations

were done overnight at 65oC, by the addition to the prehybridizing

membrane of 1 ml of probe mix. Nonspecifically bound radioactivity

was removed by washing the membrane twice for 5 minutes in 2X

SSC at room temperature and once for 30 minutes at 60oC in 2X SSC,

1% SDS. Radioactivity bound to the membrane was quantitated with

an Image Quant phosphorimager according to the manufacturers

instructions.

Preparation of probe mix. lx107 cpm/ml of radiolabeled wild-

type vaccinia virus DNA was added to 1 ml of a 1 mg/ml solution of

salmon sperm DNA, boiled for five minutes, cooled on ice for 15

minutes and used immediately.

Preparation of randomly labeled vaccinia virus DNA. Wild-

type vaccinia DNA was digested with HindIII and radiolabeled with

[a32P] dCTP (3000Ci/mmol, Amersham) by "oligo-labeling" according

to the method of Feinberg and Vogelstein (35). Unincorporated

nucleotides were removed by column chromatography through

Sephadex G50 as described (100). Incorporated radioactivity was

determined by precipitation of a 1 ul aliquot of the probe reaction










onto filter paper (Whatman) with trichloroacetic acid followed by

liquid scintillation counting, as previously described (100).

Preparation of RNA From Infected Cells



Confluent monolayers of BSC40 cells on 100 mm dishes were

infected at an moi of 15 with wild-type or E793 virus and incubated

at either 31oC or 40oC. At various times the cells were washed once

with cold PBS, lysed in 1 ml of GT buffer (4M guanidinium

isothiocyanate, 0.1 M Tris-HCl [pH 7.5], 0.5% Sarkosyl, 0.1 M 2-

mercaptoethanol), and scraped from the dishes. Samples were

pelleted through a CsCl (5.7 M CsCl, 0.01 M

ethylenediaminetetraacetic acid (EDTA) [pH 7.5]) cushion for

approximately 24 hours at 18oC in a SW-55 Ti rotor (Beckman) at

35,000 rpm. The RNA pellet was dissolved in distilled water,

precipitated with two volumes of ethanol and resuspended in

distilled water. The RNA was quantitated by absorbance at 260 nm

(1 OD260 = 40 ug/ml of RNA) (100).










Electron Microscopy of Infected Cells



Confluent monolayers of BSC40 cells on 60 mm dishes were

infected with either wild-type or E793 at an moi of 10 at 31oC or

40oC. At 24 hours post infection, cells were washed with 1 ml of PBS,

scraped from the dish into 1 ml of PBS and pelleted by centrifugation

at 400 x g. The cell pellet was resuspended in 1 ml of 2%

glutaraldehyde in PBS, fixed overnight at room temperature,

repelleted by centrifugation at 400 x g and post-fixed in an aqueous

solution of 1% OS04 for 1 hour. The cells were repelleted at 400 x g,

rinsed in distilled water, stained in 1% aqueous uranyl acetate for 30

minutes, rinsed again in distilled water and dehydrated in 25%, 50%,

75% and 100% ethanol for 10 minutes each. After dehydration the

cells were washed twice for 15 minutes in 100% acetone then

embedded in Spurr's epon resin which was then polymerized for 2

days at 60oC. After thin sectioning the samples were mounted and

viewed with a Hitachi H7000 transmission electron microscope.











Southern Analysis of Concatameric and Monomeric Vaccinia DNA

Isolated From Infected Cells



Confluent monolayers of BSC40 cells on 60 mm dishes were

infected with either wild-type, Cts53 or E793 virus at an moi of 10 at

31 oC or 40oC. At 6 and 24 hours post infection the cells were scraped

from the dish into the media, pelleted by centrifugation for 10

minutes at 400 x g, washed once with PBS and resuspended in 50 ul

of 0.15 M NaCl, 0.02 M Tris-HCI [pH 8.0], 0.01 M EDTA. The 50 ul

samples were added to 250 ul of 0.02 M Tris-HCI [pH 8.0], 0.01 M

EDTA, 0.75% SDS containing 0.8 mg/ml proteinase K (Boehringer

Mannheim) and incubated at 37oC for 6 hours. The samples were

passed through a 25 gauge needle 5 times, extracted twice with

phenol chloroform (1:1) and twice with chloroform before

precipitation with 0.1 volume of 3 M sodium acetate and 2 volumes

of ethanol at room temperature for 30 minutes. The precipitate was

pelleted by centrifugation in a microfuge at 12,000 rpm for 15

minutes, washed with 1 ml of 70% ethanol air dried, resuspended in

50 ul of TE (10 mM Tris-HCl [pH 8.0] 1 mM EDTA) and quantitated by

measurement of absorbance at 260 nM. 1.5 micrograms of total










nucleic acid was digested with 20 units of BstEII (New England

Biolabs) overnight at 60oC and separated on a 1.5% agarose gel in

TAE buffer (0.04 M Tris-HCL [pH 8.0], 0.001 M EDTA). Southern

alkaline transfer of the DNA onto a GeneScreen membrane was done

according to a protocol provided by the manufacturer. The DNA was

fixed to the membrane by uv irradiation at 254 nm in a Stratalinker

for 1 minute. The membrane was prehybridized in 7 ml of

prehybridization buffer (100 mg/ml dextran sulfate (MW 50,000),

1% SDS, 1.45 M NaCL) at 65oC for 4 hours. Hybridization was done

overnight at 65oC, by the addition to the prehybridizing membrane

of 1 ml of pBD probe mix. Nonspecifically bound radioactivity was

removed by washing the membrane twice for 5 minutes in 2X SSC at

room temperature and once for 30 minutes at 60oC in 2X SSC, 1% SDS.

Radioactivity bound to the membrane was quantitated with an

Image Quant Phosphorimager according to the manufacturers

instructions. The resolution efficiency was calculated by determining

the ratio of the intensity of the 1.3 kb resolved telomere band to the

intensity of the 2.6 kb concatamer junction band.

Preparation of pBD probe mix Approximately 1x108 cpm of

radiolabeled pBD plasmid DNA was added to 1 ml of a 1 mg/ml








86

solution of salmon sperm DNA, boiled for five minutes, cooled on ice

for 15 minutes and added to the prehybridizing membrane

immediately.

The construction of plasmid pBD, which contains a 2.6 kilobase

BstEII-BstEII vaccinia virus concatamer junction insert was a gift

from M. Merchlinsky and has been previously described (73).

Briefly, the 2.6 kb vaccinia virus insert within plasmid pBD consists

of a double-stranded copy of the terminal hairpin loop (208 base

pairs) which is the concatamer junction and 2.4 kilobases of inverted

terminal repeat sequences. Plasmid pBD was radiolabeled with

[ac32P] dCTP (3000 Ci/mmol, Amersham) by "oligo-labeling" according

to the method of Feinberg and Vogelstein (35). Unincorporated

nucleotides were removed by column chromatography through

Sephadex G50 as described (100). Incorporated radioactivity was

determined by precipitation of a 1 ul aliquot of the probe reaction

onto filter paper (Whatman) with trichloroacetic acid followed by

liquid scintillation counting, as previously described (100).













CHAPTER 3
TESTING CLUSTERED CHARGE TO ALANINE MUTAGENESIS AND
TRANSIENT DOMINANT SELECTION ON THE VACCINIA VIRUS GENE
G2R


Introduction




Traditional approaches to viral genetics, which involve random

mutagenesis of the virus followed by screening for phenotypes of

interest, have proven to be quite useful in the genetic and

biochemical characterization of vaccinia virus (reviewed in 5).

Random mutagenesis methods, however, are limited by their

inability to target specific open reading frames and by the laborious

screening methods required to identify mutants of interest. It is for

these reasons that a more direct approach for the creation of

temperature-sensitive virus mutants, in specific viral genes, must be

found. Practical application of directed genetics to a complex DNA

virus such as vaccinia, requires an in vitro mutagenesis scheme

which produces a high proportion of conditionally lethal mutants











combined with an efficient method for targeted replacement of the

wild-type viral gene with the mutant allele. The object of this work

was to determine if clustered charge to alanine mutagenesis and

transient dominant selection could be used as a practical method for

the creation of conditionally lethal temperature-sensitive vaccinia

virus mutants.

Clustered charge to alanine mutagenesis involves replacing

clusters of charged amino acids in a protein with alanines. Charged

residues which are clustered in the primary sequence of a

polypeptide often cannot be efficiently accommodated in the

hydrophobic core of the protein, therefore they are most likely to be

found on the surface of the folded protein (7). The replacement of

multiple charged residues with alanines has been shown to disrupt

protein-protein interactions in vitro, while not severely altering

protein secondary structure (7, 19). When this approach was applied

to the Saccharomyces cerevisiae actin (ACTI) gene, of 34 mutations

constructed, 44% had a temperature-sensitive phenotype in vivo

(130). Likewise, clustered charge to alanine mutagenesis of the

poliovirus 3D polymerase resulted in a high proportion (37% of 27








89

mutants) of temperature-sensitive mutants (25). Thus alterations of

a protein's surface charge may disrupt tertiary or quaternary

interactions and result in temperature sensitivity in vivo.

Vaccinia has proven to be an excellent model system for

studying transcription, DNA replication, and the regulation of gene

expression (78). The genome of vaccinia consists of approximately

192 kilobases of double-stranded DNA with covalently closed ends.

The poxviruses are unique among DNA viruses in that they replicate

in the cytoplasm of infected cells (78). Thus the virus encodes

virtually all of the enzymes required for viral RNA and DNA

metabolism. Numerous enzymatic activities common to other

eukaryotic organisms have been shown to be encoded by the virus

(41, 78). Genetic analysis of vaccinia has provided many of the tools

necessary for understanding the viral life cycle. Previous genetic

studies in the Condit laboratory led to the isolation of conditionally

lethal temperature-sensitive mutations in 30 separate vaccinia virus

complementation groups (13). These mutants have proven valuable

for studying all aspects of viral replication. However, vaccinia codes

for approximately 150 essential genes, therefore this collection








90

represents mutations in only 20% of the essential virus genes (41). A

directed genetic method that produces conditionally lethal

temperature-sensitive mutations in specific genes would allow one to

genetically study many of the viral genes in which there are no

conditionally lethal mutants.

Results




To evaluate the potential of clustered charge to alanine

mutagenesis as a means of creating conditionally lethal vaccinia

viruses, a series of nine clustered charge to alanine mutations were

created in the G2R gene (Fig. 7). G2R was chosen for the test because

mutations in this gene confer easily selectable phenotypes (70), thus

circumventing the necessity for screening large numbers of

individual plaque isolates to determine the phenotype conferred by a

particular mutation. The selection is based on the sensitivity of the

virus to the anti-poxviral drug isatin-B-thiosemicarbazone (IBT). As

shown in Fig. 8, plaque formation by wild-type vaccinia virus is

inhibited by IBT, deletion mutants in G2R such as G2A are dependent















rsi m ^ iC c/


^in
^oCl
W. co g3 C
u o -fi s yt-


a



2 o o '" r
.I 2 s II"

.5 & on o .
g )











.- o n c -5 e i.
C8Uran oE


>. <2oM g **


t- g ; *c A I
w o eC *;"" <*

uon co 82cp
m4)















C = o< -


- g ^ -
c ". cla
o s 3 a
.. QcN

0c: 0
4) 0o *c *- a




0












0 *0 .
-o^"
i..s~
3. 03 c
|u|^4 *|




.*- cf 4) _a.
* c2 aS 2 .


cu e u S *- < 1
4) I -1 i1















(4




>4
?H
>-I
E-1



CO
H
N


0
44

u
U)
F-i


Ul)

1-i
C'Jr1 I





IC/






fri

1-i
fri


i-i
z
0
0
fil
>4








U




E-4
E>4
fri

H
H



Pfr









-11
W












>
1-o


u

U)
I-i
H
fri





U)
H
H




r,04
E-1








Ul)
14N
H





P4
H
Mfi

>4
EN
H-


W H

-I-)


U')
ul


14Q
?4
Oll
0

fr1
fr


0
>4
H




N
H

M
u


en



z
-il



H




H

U)
H




E>4



Ln


r1-
0


E-4




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EWIL9RCIP_D8ET52 INGEST_TIME 2012-02-20T21:24:50Z PACKAGE AA00009025_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES