Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02047363 2001-02-08
RECOMBINANT THERAPIES FOR INFECTION AND
HYPERPROLIFERATIVE DISORDERS
Description
Technical Field
This invention relates to genetic engineering and the treatment of
hyperproliferative disorders and infection.
Background of the Invention
The therapy of viral infection is in its infancy. Bacterial infection is
typically treated with agents, such as antibiotics, which take advantage of
the
differences in metabolism between the infecting organism and its host.
However,
viruses largely employ the host's own enzymes to effect their replication, and
thus
leave few opportunities for pharmacological intervention. By employing strong
regulatory elements, the virus obtains transcription and translation of its
own
genes at the expense of host genes.
-2- . 20~'~0~
In mammals, viral infection is combatted naturally
by cytotoxic T-lymphocytes, which recognize viral proteins
when expressed on the surface of host cells, and lyse the
infected cells. Destruction of the infected cell prevents
the further replication of virus. Other defenses include
the expression of interferon, which inhibits protein syn-
thesis and viral budding, and expression of antibodies,
which remove free viral particles from body fluids. How-
ever, induction of these natural mechanisms require exposure
of the viral proteins to the immune system. Many viruses,
for example herpes simplex virus-1 (IiSV-1), exhibit a dor-
mant or latent phase, during which little or no protein syn-
thesis is conducted. The viral infection is essentially
invisible to the immune system during such phases.
Retroviruses carry the infectious form of their
genome in the form of a strand of RNA. Upon infection, the
RNA genome is reverse-transcribed into DNA, and is typically
then integrated into the host's chromosomal DNA at a random
site. On occasion, integration occurs at a site which trun-
Gates a gene encoding an essential cellular receptor or
growth factor, or which places such a gene under control of
the strong viral cis-acting regulatory element, which may
result in transformation of the cell into a malignant state.
Viruses may also be oncogenic due to the action of
their traps-acting regulatory factors on host cell regula-
tory sequences. In fact, oncogenesis was the characteristic
which lead to the discovery of the first known retroviruses
to infect humans. HTLV-I and HTLV-II (human T-lymphotropic
viruses I and II) were identified in the blood cells of
patients suffering from adult T-cell leukemia (ATL),. and are
believed to induce neoplastic transformation by the action
of their traps activating factors on lymphocyte promoter
-3- : 2047363
regions. HTLV-I and II preferentially infect human lympho-
cytes, and on occasion, cause their transformation into
malignancy. Since then, two additional retroviruses have
been found to infect humans: HIV-I and HIV-II, the etio-
logical agents of AIDS. However, HIV-I and II apparently
contribute to cancer only through their immunosuppressive
effects.
HIV I and II apparently infect cells which express
the CD4 surface protein. This protein is present in abun-
dance on thymocytes and some T-lymphocytes, and to a lesser
extent, on some antigen-presenting cells. HIV infection is
initially characterized by flu-like symptoms, followed by a
long latency period, which may last five to ten years. Upon
entering its active phase, HIV infection results in a rapid
decline in the population of "helper" T-lymphocytes (TH),,
which is usually recognized as a decline in the ratio of
T4+/T8+ (CD4+/CD8+) T-lymphocytes. The patient typically
experiences severe diarrhea, and if the central nervous sys-
tem is infected, exhibits a form of dementia. The depletion
of TH cells cripples the immune system, and the patient suc-
cumbs to an opportunistic infection by, for example ~
carinii or cytomegalovirus, or to Karposi's sarcoma. The
natural immune system appears wholly incapable of combatting
HIV infection, despite the typical presence of apparently
neutralizing serum antibody titers during latency.
Current therapy for HIV infection per se is
limited mainly to administration of AZT to inhibit viral
progression, although M.S. Hirsch, J Infect Dis (1988)
x:427-31 reported synergistic inhibition of HIV by AZT
with GM-CSF (granulocyte-monocyte colony stimulating factor)
or a interferon. AZT (3'-azido-3'-deoxythymidine) is rep-
resentative of a class of dideoxynucleoside (ddN) antiviral
agents. These agents rely on the ability of host DNA polym-
-4-.2047363
erases to reject a ddN, and the tendency of viral polymer-
ases to accept ddNs and incorporate them into replicating
polynucleotides. Upon incorporation, a ddN stops polymer-
ization, as it lacks the 3' hydroxyl group necessary for the
next phosphodiester linkage. The ddNs are typically
inactive in their administered form, and depend on phos-
phorylation by host cell enzymes for conversion to the
active triphosphate ddNTP.
H. Mitsuya et al, Nature (1987) x:773-78 dis-
closed the organization of the HIV genome, and suggested
various strategies for development of HIV therapies. Spec-
ulated therapies included administration of antisense RNA
(as free strands or encoded by an "antivirus") to inhibit
viral transcription, administration of glycosylation
inhibitors, administration of interferons to inhibit viral
budding, and administration of dideoxynucleoside analogues
to inhibit viral replication. Dideoxynucleoside analog
drugs useful for HIV therapy (e. g., AZT, ddC, etc.) depend
on host cell enzymes for activation by phosphorylation. D.
Baltimore, nature (1988) x:395-96 suggested treating AIDS
by removing bone marrow cells from an infected subject, and
transfecting hematopoietic cells in the bone marrow extract
with DNA or virus encoding an RNA or protein able to inter-
fere with HIV growth. The DNA could encode RNA that might
bind to HIV regulatory proteins, antisense RNA, mutant viral
polypeptides, or a viral DNA-binding protein lacking its
regulator function. The transfected cells would then be re-
introduced into the-.subject, and provided with a selective
advantage to insure dissemination.
A.I. Dayton et al, s~ (1986) x:941-47 disclosed
that the HIV tat gene is essential for viral protein,syn-
thesis and replication. Dayton suggested that one might
inhibit HIV by interference with tat, without otherwise
-5- ~ ~o~ ~.~s3
affecting the host cell. M.A. Mussing et al, Cell (1987)
g$:691-701 disclosed that the tat protein (rather than tat
mRNA alone) is essential for transactivation. Mussing also
found that a tat-art fusion (having 114 amino acids fused to
the C-terminus of tat by deletion of 7 nucleotides) retained
full tat activity. A.D. Frankel et al, Science (1988)
x:70-73 disclosed that tat forms metal-linked dimers in
vitro, and suggested that possible treatments for AIDS may
involve chelation of metal ions, or competition for tat
monomer binding. A.D. Frankel et al, Proc Nat Acad ~gci USA
(1988) $x:6297-30 disclosed the synthesis of an HIV-1 tat
fragment which retains the metal-binding properties of tat.
The fragment formed heterodimers with native tat, and can
displace tat in homodimers. Frankel suggested using the tat
fragment to inhibit tat dimerization, using liposomes to
deliver the peptides or a tat-fragment gene. The amino acid
sequence reported for tat from one HIV-1 isolate is
Met Glu Pro Val Asp Pro Arg Leu Glu Pro
Trp Lys His Pro Gly Ser Gln Pro Lys Thr
2o Ala Cys Thr Asn Cys Tyr Cys Lys Lys Cys
Cys Phe His Cys Gln Val Cys Phe Ile Thr
Lys Ala Leu Gly Ile Ser Tyr Gly Arg Lys
Lys Arg Arg Gln Arg Arg Arg Pro Pro Gln
Gly Ser Gln Thr His Gln Val Ser Leu Ser
Lys Gln Pro Thr Ser Gln Ser Arg Gly Asp
Pro Thr Gly Pro Lys Glu.
A similar protein is found in HIV-2.
The tar sequence, which acts in cis and is reg-
ulated py HIV tat, is found approximately at nucleotides +19
to +82 in the HIV-1 genome. The sequence contains two
extended, inverted repeats, and is thus predicted to be able
to form stem loop structures (Muessing, supra). HIV-2
exhibits a similar region.
Haseltine, US 4,738,922 disclosed the HTLV I and
II LTR promoter regions, and their use with traps-activators
2047363
(luk) in vectors for amplified expression of heterologous
proteins, exemplified by chloramphenicol acetyltransferase
(CAT). HTLV-I LTR apparently promotes constitutive expres-
sion, which is further amplified in the presence of HTLV-I
luk, while HTLV-II LTR apparently requires HTLV-II luk for
expression. Haseltine mentioned that viral traps-acting
factors can alter the expression of host cell genes as well
as viral genes, and Haseltine disclosed the concept of
inserting a vector having the HTLV LTR fused to a heter-
ologous gene into a cell having the HTLV genome, which
results in the traps-activation of the heterologous gene and
the overexpression of its product. Haseltine disclosed the
use of the HTLV LTR in the absence of luk to generate empty
capsid vaccines, and suggested using the HTLV LTR for
expression of antigens, to provide for destruction of the,
cell by monoclonal or polyclonal antibodies.
E.A. Dzierzak et al, Nature (1988) x:35-41
disclosed a prototypical gene therapy method in mice.
Dzierzak removed bone marrow cells from mice, exposed the
mice to lethal radiation, and introduced human 8-globin (BG)
genes into the removed marrow using a retroviral vector,
pSV(X)neo. The BG gene was inserted to read in the opposite
direction from proviral transcription, and constructs both
with and without the viral LTR enhancer regions were pre-
pared. The mice were then reconstituted with the recombin-
ant bone marrow, containing hematopoietic stem cells.
Dzierzak found that human 8-globin was expressed in the
resulting recombinant mice, and that the expression was
found only in cells of erythroid lineage. J-K Yee et al,
p__r~c Nat Acad Sci USA (1987) x:5197-201 disclosed con-
struction of retroviral vectors encoding HPRT under,control
of either the metallothionein promoter or the human cyto-
megalovirus (hCMV) promoter. Yee found that HPRT expression
-'- . 204 i 3fi3
doubled when~transcriptional regulatory sequences were
deleted from the U3 region of the retroviral LTR.
S.-F. Yu et al, Proc Nat Acad Sci USA (1986)
$,x:3194-98 disclosed the construction of self-inactivating
("SIN") retroviral gene transfer vectors. SIN vectors are
created by deleting the promoter and enhancer sequences from
the U3 region of the 3' LTR. A functional U3 region in the
5' LTR permits expression of the recombinant viral genome in
appropriate packaging cell lines. However, upon expression
of its genomic RNA and reverse transcription into cDNA, the
U3 region of the 5' LTR of the original provirus is deleted,
and is replaced with the U3 region of the 3' LTR. Thus,
when the SIN vector integrates, the non-functional 3' LTR U3
region replaces the functional 5' LTR U3 region, and renders
the virus incapable of expressing the full-length genomic
transcript. Yu constructed a recombinant virus using the
Mo-MuLV LTR regions and packaging (psi) sequence, and
inserted a neomycin resistance (Neo) gene under control of
either a metallothionein promoter (virus MT-N), an HSV tk
promoter (virus TK-N), or a SV40 promoter (virus SV-N) as a
selectable marker. Yu also inserted a human c-~,g gene
under control of a human metallothionein promoter (hMT) into
TK-N, and demonstrated inducible transcription of c-~,g in
NIH 3T3 cells after infection with the recombinant virus.
S.L. Mansour et al, ature (1988) x:348-52 dis-
closed a method for selecting cells after homologous recomb-
ination with a linear transfecting vector. A linear vector
was prepared having a region homologous to a target gene, a
neomycin resistance gene inserted in an exon of the homol-
ogous region, and an HSV-tk gene outside the homologous
region. Upon specific homologous recombination, the.trans-
formed cell displays a phenotype negative for the target
region and HSV-tk, and positive for neomycin resistance
- g -
~ ~4 ~3fi3
(homologous recombination into the target site disrupts the
target site gene, and fails to incorporate the tk gene).
The phenotype distinguishes cells having homologous recomb-
ination from non-specific integration, as the latter cells
will display a phenotype positive for the target gene,
HSV-tk, and neomycin resistance. Neomycin resistance is
tested by culturing cells in neomycin, while tk+ is tested
by culturing cells in gancyclovir (which is converted to a
toxic product by tk).
R.D. Palmiter et al, Cell (1987) $Q:435-43 dis-
closed the preparation of transgenic mice having a DNA con-
struct encoding the diphtheria A chain under control of the
elastase promoter. The elastase promoter is active only in
pancreatic acinar cells, and promotes the expression of
elastase. The DNA constructs were microinjected into mouse
eggs, and the resulting progeny examined. Transgenic mice
in which the construct was active failed to develop normal
pancreatic tissue.
M.E. Selsted et al, J Clin Invest (1985) 7~:1436-
39 disclosed the primary amino acid sequence for three
related human cytotoxic effector polypeptides, termed human
neutrophil antimicrobial peptides (HNPs). The three HNPs
have 29-30 amino acid residues and exhibit activity against
bacteria, fungi, and herpes simplex virus (T. Gantz et al,
Clin Invest (1985) x:1427-35).
T.L. Wasmoen et al, J Hiol Chem (1988) x:12559-
63 disclosed the primary amino acid sequence of human
eosinophil granule major basic protein (M8P). MBP is an
effector polypeptide having 117 amino acid residues, a pI of
10.9, and exhibiting cytotoxic activity against mammalian
cells and parasites.
M.A. Adam et al, J Virol (1988) ~x:3802-06 dis-
closed retroviral vectors which exhibit efficient RNA pack-
- 9 -
2 04 7363
aging, having a psi+ sequence. R.D. Cone et al, Mol Cell
Biol (1987) 7:887-97 disclosed the construction of retro-
viral vectors (pSVX) including a human B-globin gene, and
the use of the recombinant vectors to obtain human B-globin
expression in a murine erythroleukemia cell line. A.D.
Miller et al, Mol Cell Biol (1986) x:2895-902 disclosed cell
lines useful for packaging replication-defective retroviral
vectors.
Guild et al, J Virol (1988) ~x,:3795-801 disclosed
retroviral vectors using the Mo-MuLV LTRs and psi (packag-
ing) sequence, useful for transfer of entire genes into mam-
malian cells. Guild employed B-actin and histone promoters
(which are active in essentially all cells) to obtain tran-
scription of neon (as a selectable marker). Expression of
the neon was demonstrated in vivo, after virus-infected bone
marrow cells were used to reconstitute lethally-irradiated
mice.
A.D. Friedman et al, Nature (1988) x:452-54 dis
closed transfection of tk mouse cells with plasmids expres
sing HSV-1 tk, and HSV-1 VP16. VP16 acts as a traps act
ivator for transcription of the immediate early genes in
herpes simplex virus (HSV-1). On the VP16 plasmid, the pro-
moter was replaced with the Mo-MSV promoter, and the carboxy
terminal of the VP16 was deleted. The resulting plasmid
codes for a mutant VP16 protein which competes with wild
type VP16 for DNA binding. Transfected cells exhibited
resistance to HSV-1 replication upon later infection.
Friedman suggested that one might induce resistance to HIV
by transfecting with a dominant mutant of the HIV trans-
activator protein (tat).
J. Sodroski et al, Science (1985) x:74-77 dis-
closed the location of the HIV tat gene. J. Sodroski et al,
Science (1985) x:171-73 disclosed construction of a plas-
2 04 73s3
- 10 -
mid having CAT under control of the HIV LTR. The HIV LTR
contains the transactivating region (tar) gene which is
induced by tat. Sodroski discovered that transactivating
factors would induce transcription of genes under control of
the HIV LTR. B.M. Peterlin et al, Proc Nat Acad Sci USA
(1986) $x:9734-38 disclosed plasmids having the HIV tar
gene, and the effects of orientation and position of tar on
transactivation by tat. Peterlin found that tar functions
best when downstream from a promoter and an enhancer. Act-
ivation with tat increased transcription to RNA. G.J. Nabel
et al, Science (1988) x:1299-302 disclosed that a HIV
tat-III-CAT fusion plasmid could be activated by HSV and
adenovirus trans-acting factors.
B.K. Felber et al, Science (1988) x$:184-87 dis-
closed an assay for HIV, using CD4-expressing cell lines,
transfected with the chloramphenicol acetyltransferase (CAT)
gene under control of the HIV LTR. Upon infection with HIV,
the transfected cell lines expressed CAT in proportion to
the amount of virus present. Felber suggested using the
assay as a means for screening possible anti-HIV drugs for
their ability to inhibit viral growth, and for identifying
anti-tat drugs.
nicniosure of the Invention
One aspect of the invention is a method for treat-
ing host cells for an infection or a hyperproliferative dis-
order..~rhich is characterized by expression of regulatory
factors capable of regulating transcription of DNA, by
inserting into the cells a polynucleotide construct having a
regulatory region which is activated by the regulatory fac-
tor, and an effector gene under control of the regulatory
region which renders said cell susceptible to protection or
destruction. The gene product may destroy the cell dir-
-11- ~ 2047363
ectly, as in the case of cytotoxins, or may increase the
cell's susceptibility to destruction by pharmacologic
agents. Alternatively, the gene product may directly
inhibit the infectious or malignant agent, e.g., by com-
petition for binding sites, by binding as antisense RNA, by
expression of protein inhibitors or antibodies, by expres-
sion of sequence-specific ribozymes, by expression of
enzymes which activate anti-viral compounds, and the like.
For example, the activation region may be homologous to the
HIV tar region, and the effector gene may encode ricin A or
HSV-1 thymidine kinase. Upon infection with HIV, the HIV
tat protein activates the tar region, and induces tran-
scription and expression of ricin A, resulting in cell
death, or of HSV-1 tk, resulting in cytotoxicity when
treated with dideoxynucleoside agents such as gancyclovir,.
Another aspect of the invention is a DNA construct
which accomplishes the method of the invention.
Another aspect of the invention is a composition
useful for delivering the DNA constructs of the invention to
host cells.
Another aspect of the invention is a method for
protecting specific cell populations within an organism from
viral infection, by inserting into the cells of the popula-
tion a polynucleotide construct comprising a cis-acting
sequence which promotes expression of a nearby gene only in
the presence of trans-acting factors found substantially
only i~~the selected cell population: and under control of
the cis-acting sequence, an effector gene which protects the
cell or renders it susceptible to protection. Preferably,
the cis-acting sequence is derived from the host. The
effector gene is preferably a nucleoside kinase, such as
HSV-1 thymidine kinase.
- lla-
According to a first aspect of the invention, there is provided a
method for treating a host cell ex vivo to render the host cell protected from
a
hyperproliferative disorder, or to destroy the host cell when said
hyperproliferative
disorder arises, wherein said hyperproliferative disorder is associated with
the
expression in said host cell of a trans-acting regulatory factor, and wherein
said
trans-acting regulatory factor is capable of regulating expression of genes
under
the control of a cis-acting regulatory sequence, which method comprises:
inserting
into said host cell a polynucleotide construct comprising said cis-acting
regulatory
sequence, and an effector gene under the control of said cis-acting regulatory
sequence, wherein said effector gene is expressed in the presence of the trans-
acting regulatory factor and provides an effector gene product that renders
said
host cell protected from said hyperproliferative disorder, or that
participates in the
destruction of said host cell when said hyperproliferative disorder arises,
said gene
product being selected from the group consisting of a cytokine, a ribozyme, a
glycosylation inhibitor, a myristilation inhibitor, and an RNase.
According to a second aspect of the invention, there is provided a
polynucleotide construct which comprises: a cis-acting regulatory sequence
which
is controllable by a trans-acting regulatory factor, wherein the presence of
said
trans-acting regulatory factor in a mammalian host cell is associated with
that host
cell having a hyperproliferative disorder or an infection by an infectious
agent; and
an effector gene under the control of said cis-acting regulatory sequence,
said
effector gene encoding a gene product for rendering said host cell protected
from
said hyperproliferative disorder or said infection, or a gene product for
destroying
said host cell when said hyperproliferative disorder or said infection arises,
said
CA 02047363 2000-07-13
- 1 lb -
gene product being a member of the group consisting of a cytokine, a protease
inhibitor, a glycosylation inhibitor, a myristilation inhibitor and an RNase.
According to a third aspect of the invention, there is provided a
composition which comprises: a polynucleotide construct as described above;
and
a pharmaceutically acceptable carrier.
According to a fourth aspect of the invention, there is provided a
method for treating a host cell ex vivo to render the host cell protected from
infection by an infectious agent, or to destroy the host cell when infected by
said
infectious agent, wherein said infection by said infectious agent is
associated with
the expression in said host cell of a trans-acting regulatory factor, and
wherein
said trans-acting regulatory factor is capable of regulating expression of
genes
under the control of a cis-acting regulatory sequence, which method comprises:
inserting into said host cell a polynucleotide construct comprising said cis-
acting
regulatory sequence and an effector gene under the control of said cis-acting
regulatory sequence, wherein said effector gene is expressed in the presence
of
the trans-acting regulatory factor and the effector gene product renders said
host
cell protected from infection by said infectious agent, or participates in the
destruction of said host cell when said host cell is infected by said
infectious
agent, said effector gene product being selected from the group consisting of
a
cytokine, a ribozyme, a glycosylation inhibitor, a myrisylation inhibitor, and
an
RNase.
According to a fifth aspect of the invention, there is provided a
polynucleotide construct which comprises: a cis-acting regulatory sequence,
said
cis-acting regulatory sequence controllable by a trans-acting regulatory
factor,
CA 02047363 2000-07-13
- l lc -
wherein the presence of said trans-acting regulatory factor in a human host
cell is
associated with said human host cell being infected by a human disease-causing
infectious agent; and an effector gene under the control of said cis-acting
regulatory sequence and expressible in the presence of said trans-acting
regulatory factor, said effector gene encoding a gene product capable of
rendering
said human host cell protected from infection by said infectious agent, or
said
gene product capable of participating in the infectious agent-specific
destruction of
said human host cell when said human host cell is infected by said infectious
agent, said gene product further being a member of the group consisting of a
cytokine, a protease inhibitor, a glycosylation inhibitor, a myristylation
inhibitor,
and an RNase.
According to a sixth aspect of the invention, there is provided a
composition which comprises: a polynucleotide construct as described above;
and
a pharmaceutically acceptable carrier.
According to a seventh aspect of the invention, there is provided a
polynucleotide construct which comprises a cis-acting regulatory sequence,
said
cis-acting regulatory sequence controllable by a trans-acting regulatory
factor that
is not encoded within said construct, wherein the presence of said traps-
acting
regulatory factor in a human host cell is associated with said human host cell
being infected by a human disease-causing infectious agent; and an effector
gene
under the control of said cis-acting regulatory sequence and expressible in
the
presence of said traps-acting regulatory factor, said effector gene encoding a
gene product capable of rendering said human host cell protected from
infection
by said infectious agent, or said gene product capable of participating in the
CA 02047363 2000-07-13
- 1 ld -
infectious agent-specific destruction of said human host cell when said human
host cell is infected by said infectious agent, said gene product further
being
selected from the group consisting of a ribozyme, a glycosylation inhibitor, a
myristylation inhibitor, and an RNase.
According to an eighth aspect of the invention, there is provided a
polynucleotide construct for disease-specific expression which comprises a cis-
acting regulatory sequence, said cis-acting regulatory sequence controllable
by a
traps-acting regulatory factor that is not encoded within said construct,
wherein the
presence of said traps-acting regulatory factor in a human host cell is
associated
with said human host cell being infected by a human disease-causing infectious
agent; and an effector gene under the control of said cis-acting regulatory
sequence and expressible in the presence of said traps-acting regulatory
factor,
said effector gene encoding a cytokine capable of rendering said human host
cell
protected from infection by said infectious agent, or said gene product
capable of
participating in the infectious agent-specific destruction of said human host
cell
when said human host cell is infected by said infectious agent.
According to a ninth aspect of the invention, there is provided a
polynucleotide construct which comprises at least two tandem copies of a cis-
acting regulatory sequence, said cis-acting regulatory sequence controllable
by a
traps-acting regulatory factor that is not encoded within said construct,
wherein the
presence of said traps-acting regulatory factor in a human host cell is
associated
with said human host cell being infected by a human disease-causing infectious
agent; and an effector gene under the control of said cis-acting regulatory
sequence and expressible in the presence of said traps-acting regulatory
factor,
CA 02047363 2000-07-13
- l le -
said effector gene encoding a gene product capable of rendering said human
host
cell protected from infection by said infectious agent, or said gene product
capable
of participating in the destruction of said human host cell when said human
host
cell is infected by said infectious agent, said gene product further being
selected
from the group consisting of a prodrug converting enzyme, a cytokine, a
protease
inhibitor, a ribozyme, a glycosylation inhibitor, a myristylation inhibitor,
and an
RNase.
According to a tenth aspect of the invention, there is provided a
polynucleotide construct for inhibiting infection by an infectious agent in a
host
cell, said construct comprising at least two copies of a polynucleotide
binding
sequence that competes for binding of a trans-acting regulatory factor against
a
cis-acting regulatory site that is homologous to said polynucleotide binding
sequence, and wherein transcription from the cis-acting regulatory site is
associated with an infection by said infectious agent.
CA 02047363 2000-07-13
-12- - 2047363
Another aspect of the invention is a polynucleo-
tide construct useful for protecting selected cell popula-
tions from infection, which comprises a cis-acting sequence
which promotes expression of a nearby gene only in the pres-
s ence of traps-acting factors found substantially only in the
selected cell population; and under control of the cis-
acting sequence, an effector gene which protects the cell or
renders it susceptible to protection. Preferably, the cis-
acting sequence is derived from the host. The effector gene
is preferably a nucleoside kinase, such as HSV-1 thymidine
kinase.
R_ri_ef Description of the Drawings
Figure 1 is a diagram of a generic polynucleotide
construct of the invention.
Figure 2 is a diagram of the vector pTHl.
Figure 3 is a diagram of the vector pMXSVNeo-
tar/tk.
Figure 4 graphically illustrates the results of
the experiment described in Example 5.
Figure 5 graphically illustrates the results of
the experiment described in Example 4.
A. Definitions
The term "treatment" as used herein refers to
reducing or alleviating symptoms in a subject, preventing
symptoms from worsening or progressing, inhibition or elim-
ination of the causative agent, or prevention of the infec-
tion or disorder in a subject who is free therefrom.. Thus,
for example, treatment of a cancer patient may be reduction
of tumor size, elimination of malignant cells, prevention of
-13- 204763
metastasis, or the prevention of relapse in a patient who
has been cured. Treatment of infection includes destruction
of the infecting agent, inhibition of or interference with
its growth or maturation, neutralization of its pathological
effects, and the like.
The term "infection" as used herein includes
infection by viruses, bacteria, fungi, and other parasites,
such as leishmania and malarial parasites. "Infectious
agents" within the scope of this invention include viruses
such as HIV-I, HIV-II, HTLV-I, HTLV-II, herpes simplex virus
(HSV), cytomegalovirus (CMV), Epstein-Barr virus (EBV),
human papilloma viruses (HPV), hepatitis B virus (HBV),
hepatitis C virus (HCV), polio virus, and the like: bac-
teria such as B. pertussis, and the causative agents of
tetanus, diphtheria, cholera, and the like: mycobacteria.
such as ~. tuberculosis, and the causative agent of leprosy;
yeasts and fungi such as C. albicans and p. carinii: para-
sites such as malarial Plasmodia, giardia, and the like.
Certain methods and constructs of the invention are most
suitable for intracellular infectious agents, such as
viruses, malaria and the like, while other methods and con-
structs are applicable to extracellular infections.
The term "hyperproliferative disorder" refers to
disorders characterized by an abnormal or pathological pro-
liferation of cells, for example, cancer, psoriasis, hyper-
plasia and the like.
The term "cis-acting regulatory sequence" refers
to a polynucleotide sequence which is capable of responding
to a trans-acting factor, and enhancing transcription of
cis-located genes. Most appropriate cis-acting regulatory
sequences may be derived from the infectious agent they will
be used to combat. For example, the tar region of the HIV-1
LTR is a suitable cis-acting regulatory sequence for treat-
-14- _ 204363
ment of HIV-1. Where cell-type specific expression is
desired, one may employ cis-acting sequences such as, for
example, (for T-cells) the CD2 antigen (Genbank HUMATCCD2),
IL-2 (Genbank HUMIL2A), IL-2 receptor (Genbank HUMIL2R1),
CD1 antigen (Genbank HUMHTAl), CD3 antigen (Genbank
HUMATCT31), CD4 antigen (Genbank HUMATCT4), T-cell protease
(Genbank MUSSPTCS): for B-cells, IgG (Genbank HUMIGCA1),
MHC-1 antigen (Genbank HUMMFiA2): for macrophages, Mac-1
antigen (Genbank HUMLAP), IL-1 (Genbank HUMILiP), and the
like. The cis-acting regulatory sequence may be used in
multiple tandem copies, in order to increase its competition
with the endogenous cis-acting regulatory site. Suitable
sequences for treatment of hyperproliferative disorders
(especially cancers) are obtainable from the binding sites
of known oncoproteins, or by identification of a known pro-
tein whose expression is altered by an oncogene. The cis-
acting regulatory sequence must be capable of providing suf-
ficient expression of the effector gene to confer suscepti-
bility to protection or destruction of the cell when act-
ivated by the traps-acting factor, without allowing sub-
stantial constitutive expression in the absence of the
traps-acting factor. The choice of cis-acting sequence will
depend upon the traps-acting factor associated with the
infection or disorder to be treated, and on the effector
gene selected. For example, the HIV LTR contains both the
tar region, which is highly selective for HIV tat, and also
a region activated by the endogenous nuclear factor NF-KB
( the .LTR has tandem-. NF-KH binding regions ) . Although the
tar sequence strongly suppresses expression in the absence
of tat (see for example Muesing, Peterlin, supra), the cis-
acting elements upstream of the tar sequence influence the
degree of constitutive expression ("leakiness") that occurs
in the absence of tat. The degree of leakiness in the
-15- 2p~73fi3
absence of tat can be controlled by deletion or substitution
of the cis-acting elements (e. g., NF-KH binding sites)
within the HIV-1 LTR. In order to use the tar sequence with
effector genes such as ricin A (which is effective at very
low intracellular concentration), one would remove the NF-
KH binding sites using restriction endonucleases prior to
inserting the vector into host cells. In contrast, if the
effector gene encoded a product which was not particularly
toxic to the host cell (for example, the rpt-1 protein found
in resting T-lymphocytes: see Patarca, supra), but which
requires higher concentration for effective viral inhib-
ition, one must employ a cis-acting sequence which provides
for strong expression upon induction, but may allow some
low-level constitutive expression.
The phrase "susceptible to protection or destruc-
tion" means that upon infection or occasion of a hyperpro-
liferative disorder, the presence of the effector gene
within the host cell renders the cell either (a) capable of
inhibiting the infectious agent or hyperproliferative con-
dition , or (b) the effector gene kills the host cell, or
renders it sensitive to an additional exogenous toxic agent.
The additional "toxic agent" does not include the host
immune system or antibodies, as immunity is often suppressed
or ineffective in cases of infection or hyperproliferative
disease. Inhibition of infection may be accomplished by
reducing intracellular levels of nutrients or metabolites
needed~by the infectious agent (e. g., purine or pyrimidine
nucleotides, carbohydrates, phosphates, etc.), down-reg-
ulating host enzymes required by the infectious agent (for
example ribbsomal enzymes, endogenous proteases, protein
folding enzymes, transport proteins, and the like), down-
regulating host cell regulatory factors employed by the
infectious agent (for example, the NF-KB nuclear factor
-16- 204,~~3
found in activated lymphocytes which up-regulates HIV-1
transcriptionl, up-regulating viral or host cell factors
which suppress viral gene expression, by locking the host
cell in a static phase of the mitotic cycle (in the case of
certain viruses and hyperproliferative disorders), and the
like.
Inhibition may also be accomplished by interfer-
ence with the infectious agent life cycle, for example by
expressing factors which bind to important infectious agent
regulatory factors and cause inhibition or deactivation, by
expressing host or infectious agent regulatory factors which
down-regulate the agent's expression (for example, viruses
having a latent phase must have factors which prevent con-
stitutive expression due to their strong regulatory
regions), by expressing non-infectious defective mutants of
the coat, envelope, or capsid proteins of the infectious
agent, by encoding multiple copies of the polynucleotide
binding sequence (such that the sequences compete for bind-
ing with the infectious agent regulatory agent, and thus
limit transcription of the agent), by expression of factors
which attack infectious agent proteins, lipids or carbo-
hydrates, (such as neutrophil antimicrobial peptides,
eosinophil granule major basic protein, and the like) or
which inhibit or prevent processing by infectious agent
enzymes. Alternatively, one can encode cytokines which may
interfere with or inhibit infectious agents, or which may be
cytotoxic, for example, interferons, interleukins, tumor
necrosis factors, colony-stimulating factors, transforming
growth factors (a and 8), epidermal growth factors, and the
like. Cytokine expression is also useful in the treatment
of hyperproliferative disorders. For example, cytokines may
inhibit or kill tumors directly, or may induce differentia-
tion into a final (non-neoplastic) state. Hyperprolifer-
-1~- _ 2047363
ative disorders may also be treated using any of the other
methods noted above which are appropriate. For example,
inhibition of cell functions (such as mitotic cycle, protein
expression, and the like) can inhibit proliferation.
Cytotoxic techniques may involve the direct
expression of a cellular toxin, such as ricin A, diphtheria
toxin, and the like, or may employ one of the above-noted
methods to a degree which results in cell death (for
example, complete inhibition of cellular respiration).
l0 Alternatively, one may express an enzyme or protein which
renders the cell susceptible to an additional agent, for
example, elevated expression of a nucleoside kinase may
render the host cell susceptible to the action of gancyclo-
vir or acyclovir, or may increase the potency of an anti-
viral agent such as AZT. Dideoxynucleoside analogues (ddNs)
such as AZT, dideoxycytidine, acyclovir, gancyclovir, and
the like, are relatively non-toxic in their non-phosphoryl-
ated form. However, specific viral or intracellular enzymes
may convert the ddNs to the corresponding triphosphates, at
which point the agent acts as a chain termination agent dur-
ing transcription or replication. The utility of ddNs is
based on the fact that (1) viral polymerises exhibit a
higher affinity for ddNs than mammalian polymerises, and (2)
some viral nucleoside kinases exhibit a higher rate of phos-
phorylation for ddNs than mammalian kinases. Thus, viral
enzyaes (and consequently viral genes) are more affected by
chain termination than are mammalian enzymes and genes. The
HSV-1 thymidine kinase (HSV-1 tk) is more efficient at con-
verting ddNs to the corresponding triphosphates, thus ren-
dering cells having active HSV-1 tk more susceptible to
ddNs. .
The term "effector gene" refers to a polynucleo-
tide sequence which, when expressed (due to action by the
18 _ 204 T363
traps-acting. factor on the cis-acting regulatory sequence),
renders the host cell susceptible to protection or des-
truction, as defined above. Cytotoxic proteins within the
scope of this invention do not include possibly toxic viral
proteins under control of its normal viral regulatory
region. The effector gene must be a gene which is not
naturally regulated by the cis-acting regulatory region
employed. one class of suitable effector genes includes
genes encoding a nucleoside kinase capable of phosphorylat-
ing dideoxynucleoside analogues at a greater rate than host
cell nucleoside kinases, for example HSV-1 thymidine kinase,
guanine kinase, plant nucleoside phosphotransferase, Leish-
mania donovani purine 2'-deoxyribonucleosidease, L. donovani
hypoxanthine-guanine phosphoribosyltransferase, and other
suitable enzymes of nucleoside metabolism capable of act-.
ivating nucleoside antiviral or chemotherapeutic agents.
Another class of effector genes includes genes functional at
the mRNA level, such as antisense mRNA, and ribozymes (V.
Walbot et al, Nature (1988) ~g:196-97). Another class of
effector genes includes genes encoding cytokines useful for
antiviral or anti-hyperproliferative disorders, such as
tumor necrosis factor, a interferon, 8 interferon, gamma
interferon, transforming growth factor-B, inhibitory pep-
tides, and interleukin-2. Other effector genes within this
invention include protease inhibitors, which may inhibit
essential protein processing, and inhibitors of glycosyl-
ation, phosphorylation, or myristylation of viral proteins.
RNase is also suitable. one may alternatively "titrate'~ the
traps-acting factor by providing a plurality of cis-acting
sequences in conjunction with a strong termination sequence.
In the context of cell-specific expression, the
effector gene must confer susceptibility of the cell to pro-
tection, but not destruction. Thus, one avoids deleting an
-19- x.204 i~~~
entire class of cells, such as the class of TH cells. This
technique can be employed to increase the therapeutic ratio
of certain antiviral and chemotherapeutic agents, such as
AZT. For example, AZT inhibits HIV replication, but is rel-
y atively toxic to bone marrow cells. T-lymphocytes are com-
monly infected in HIV infection, and are more tolerant of
AZT. Hy transfecting a bone marrow aspirate with a poly-
nucleotide construct which expresses, for example, HSV-1 tk
only in T-lymphocytes (for example, under control of the CD4
antigen promoter), reintroducing the cells into the subject,
and administering an amount of AZT less than the normal dos-
age, one can increase the intracellular concentration of
phosphorylated AZT in tolerant T-lymphocytes, without
increasing the phosphorylated AZT levels in sensitive stem
cells (in bone marrow). .
Polynucleotide constructs of the invention may
also be designed to treat autoimmune disorders, by using a
cis-acting sequence which responds to a trans-acting factor
characteristic of those specific immune cells which par-
ticipate in the particular autoimmune disorder treated, and
an effector gene which suppresses the activity of those
cells.
The term "antisense mRNA" refers to mRNA which is
complementary to and capable of forming dsRNA complexes with
a "sense" strand of mRNA. Hybridization of mRNA inhibits
its translation into proteins, thus, antisense mRNA acts as
a very specific protein synthesis inhibitor. Antisense mRNA
may be protective, if, it complements for example a viral
protein mRNA or mRNA transcribed from an active oncogene.
Antisense mRNA may be destructive, for example where it com-
plements an essential host cell enzyme, such as a house-
keeping enzyme with no effective alternate pathway, or ren-
ders the host cell susceptible to agents which inhibit the
-20- 2o4~~s3
alternate pathways. See also G. Zon et al, EP 288,163,
which disclosed the use of oligodeoxynucleotides for inhib-
ition of retroviral replication and oncogenic proliferation.
The term "ribozyme" refers to a catalytic RNA
polynucleotide capable of catalyzing RNA cleavage at a
specific sequence. Ribozymes are useful for attacking par-
ticular mRNA molecules. For example, in chronic myelogenous
leukemia, a chromosomal translocation involving the genes
bcr and abl (Philadelphia chromosome) results in expression
of a bcr-abl fusion protein, which is believed to result in
abnormal function of the abl oncoprotein. Because the
fusion between the bcr and abl genes occurs at points within
one of two introns, the spliced bcr-abl fusion transcript
contains only two possible sequences at the splice junction
between the bcr and abl exons. As the bcr-abl mRNA will .
only occur in lymphoid cells which have undergone this
oncogenic chromosome translocation, a ribozyme specific for
either of the two bcr-abl fusion mRNA splice junctions may
be prepared, and thus may inhibit expression of the
corresponding oncoprotein.
The term "vector" as used herein refers to any
polynucleotide construct capable of encoding the cis-acting
regulatory sequence and the effector genes) selected, and
capable of transferring these genes into suitable target
cells. Vectors may be linear or circular, double stranded
or single stranded. Vectors may comprise DNA, RNA, DNA/RNA
hybrids; and DNA and/or RNA polynucleotides having chem-
ically modified bases~.~ Suitable vectors within the scope of
this invention include linear dsDNA segments, plasmids,
recombinant viruses (e. g., recombinant vaccinia, adenovirus,
adeno-associated viruses, and the like), replication-
defective retroviral vectors (including human, simian,
murine, and the like), suitably non-pathogenic derivatives
-21- 2047363
of replication-competent retroviral vectors, and the like.
Replication defective retroviral vectors are presently pre-
ferred.
B. General Method
Preparation of polynucleotide constructs is con-
ducted using methods known generally in the art. Once an
infectious agent or hyperproliferative disorder has been
selected for treatment, the first step is to identify an
associated traps-acting factor, an appropriate cis-acting
sequence, and an effector gene.
Some viral traps-acting factors are already known
and characterized. Other viral traps-acting factors may be
identified by deletion analysis of the cloned viral genome.
For example, a new virus may be cloned and sequenced using
standard techniques, and the open reading frames identified.
Then, deletion mutants are prepared using restriction
enzymes, and the mutant viruses assayed for transcriptional
competence in suitable host cells. Mutants which exhibit
very low mRNA transcription probably lack either a gene
encoding a traps-acting factor, or a cis-acting regulatory
sequence. Whether the deletion affects the traps-acting
factor or the cis-acting sequence can generally be deter-
mined by examination of the position and content of the gen-
omic sequence (e. g., cis-acting viral regulatory sequences
are generally found upstream of open reading frames). The
sequence.,of the cis-acting viral region may be compared for
homology to known cisjacting regions. As homologous cis-
acting sequences may react with the same traps-acting fac-
tor, appropriate endogenous traps-acting factors may thus be
identified. Alternatively, viral proteins may be expressed
recombinantly in appropriate hosts (e. g., bacteria, yeast,
mammalian cell culture), and purified extracts labeled and
- 22 -
' '_ 2 p4 73fi3
screened for binding to the viral genome library. In this
manner, appropriate traps-acting factor/cis-acting sequence
pairs may be identified. The suitability of the pair may be
evaluated using the CAT expression assay detailed below and
in the examples. In these assays, the cis-acting sequence
is cloned into a suitable vector, and a suitable reporter
gene (e. g., CAT, B-galactosidase, etc.) ligated to the cis-
acting sequence to provide for cis-control. The test con-
struct is then transferred into a suitable host cell (e. g.,
mammalian cell culture) and assayed for reporter gene
expression in the presence and absence of the traps-acting
factor. Suitable effector genes are known in the art, or
may be cloned using standard techniques.
Intracellular parasites (including viruses) often
induce interferon expression. Thus, isolation of the inter-
feron cis-acting sequence should allow identification of
traps-acting factors involved in the anti-parasite response,
and preparation of appropriate polynucleotide constructs of
the invention.
Hyperproliferative disorders are characterized by
inappropriate gene regulation or gene product activity, typ-
ically involving cell growth or differentiation factors, and
occasionally genes encoding structural proteins. Inappro-
priate gene regulation may result from the expression of a
dysfunctional traps-acting factor (e. g., a factor in which
truncation or mutation has eliminated inhibition of the fac-
tor, or which binds irreversibly, etc.), from the over-
expression of a traps-acting factor or receptor (e.g., due
to the translocational juxtaposition of a strong promoter),
or other defects. In any case, the disorder characterist-
ically exhibits a traps-acting factor which is either dif-
ferent from those found in normal cells, or is present in
abnormally large quantities. Traps-acting factors encoded
- 23 -
z~4i363
by viruses associated with hyperproliferative disorders
(e.g., HTLV-I, HTLV-II, the E6 and E7 genes of HPV type 16
and type 18) may also be used to regulate expression of an
effector gene, as described herein. Once the affected cis-
acting sequence is identified, a suitable effector gene may
be selected. As the characteristic traps-acting factors are
generally at least somewhat homologous to normal traps-act-
ing factors, the cis-acting sequence is likely to exhibit
some regulation by endogenous traps-acting factor in normal
cells. Thus, preferred effector genes for treatment of
hyperproliferative disorders are those which are relatively
non-toxic to the host cell at low levels. However, the act-
ivity of the traps-acting factors may be constitutively high
in the hyperproliferative cells, whereas the activity may be
lower and fluctuating (as during the cell cycle) in normal
cells. This difference may be exploited to allow accumula-
tion of high levels of effector gene product in hyperplastic
cells.
Similarly, many cell-type specific traps-acting
factors and cis-acting regulatory sequences have been dis-
covered and described in the literature. Additional cis-
acting regulatory sequences may be determined by the methods
outlined above. It should be noted that in some instances,
a traps-acting factor associated with a hyperproliferative
disorder and its cis-acting regulatory sequence will be
identical to normal, cell-type specific traps-acting factors
and regulatory sequences. In such cases, the hyperprolif-
erative disorder is typically caused by an excess concen-
tration of traps-acting factor. Accordingly, the effector
gene must be carefully selected.
Referring to Figure 1, a generic retroviral vector
of the invention is depicted. The vector comprises a 5'
retroviral LTR and primer binding site (~), a psi encapsid-
~ 04 X363
- 24 -
ation (packaging) signal sequence (~), an optional 3' RNA
processing signal sequence (~,), an effector gene (4), a 5'
cis-acting regulatory sequence (,~) including a promoter,
optionally a promoter ($) and selectable marker gene (Z),
and a 3' retroviral LTR and primer binding site ($).
Sequences ~-$ constitute the retroviral portion of the vec-
tor, while sequence $ is typically derived from a plasmid,
and provides for maintenance of the vector in cell culture.
The 5' LTR (~) and 3' LTR ($) provide for reverse tran-
scription and integration of the vector into the host cell
genome. Suitable LTRs include those derived from Moloney
murine leukemia virus (Mo-MuLV) (Shinnick et al, Nature
(1981) x:543), Harvey murine sarcoma virus (Ha-MSV), (Van
Beveran et al, ~ (1981) x:97), HTLV-I, HTLV-II, HIV-1,
and HIV-2. In replication-defective retroviruses, the U3.
region of the 3' LTR ($) is inactivated. Psi sequence
must be included in order for the vector to be included in
viral capsids generating in the packaging cell line, but is
otherwise inactive once integrated into the host genome.
Suitable psi sequences have been described by Cepko et al,
Cell (1984) x:1053-62: Guild et al, supra: and Kriegler
et al, Cell (1984) $$:483. Where the vector is nonretro-
viral, and is to be inserted by transfection rather than
infection, the LTR sequences and psi sequence may be omit-
ted. The effector gene (4) is as described above. In
"internal promoter" recombinant retroviral vectors, the
effector gene (4) is provided with its own promoter/cis-
acting,regulatory sequence ($) and terminator sequence/poly-
adenylation sequence (4) (although the terminator and PA
sequences may be derived from the effector gene).
The effector gene may be oriented in either dir-
ection, but is usually oriented in the direction opposite to
the LTR reading frame. In "enhancer replacement" recombin-
2 ~4 X363
- 25 -
ant retroviral vectors, the effector gene (g) is oriented in
the same direction as the LTR, and is not provided with a
separate cis-acting regulatory sequence (~) or terminator
sequence/polyadenylation sequence (4). Instead, the cis-
acting regulatory sequence is provided within the 3' LTR ($,)
U3 region, so that the effector gene is controlled by the
cis-acting regulatory region only after reverse transcrip-
tion.
The selectable marker (Z), if present, encodes a
characteristic which enables cells expressing the marker
sequence to survive under conditions selecting for trans-
fected cells. The selectable marker typically encodes an
enzyme conferring resistance to an antibiotic, for example
chloramphenicol, neomycin (Southern et al, ~f Mol AnDl Gen
(1982) x,:327-41), or hygromycin (Gritz et al, Gene (1983)
x:179-88). The selectable marker, when employed, is
usually provided with its own promoter (~) which allows
expression of the marker gene during selection of trans-
fected/infected cells. Suitable marker gene promoters
include the histone promoter, HSV-1 tk promoter, and the
metallothionein promoter.
Sequence ~, is a polynucleotide maintenance
sequence, which provides for the stable maintenance of the
vector within producer cells. Typically, sequence ~, is
derived from a plasmid, and provides an origin of repli-
cation (such as pBR322 ori, or the yeast 2p origin) and
(usually) an antibiotic resistance marker. Suitable main-
tenance sequences include pXf3, pBR322, pUCl8, pML, and the
like. In the case of linear DNh segments used for targeted
integration, the vector may be linearized at a point within
sequence ~. The LTR regions ,~ and $ are replaced with
sequences homologous to the sequence of desired targeted
recombination. Sequence ~ is deleted. Sequence ~ includes
-26- X2047363
polynucleotide sequences which provide for maintenance and
replication in microbial hosts, and additionally includes a
counter-selection marker (which provides for deletion of
transfected host cells which undergo non-specific integra-
tion).
np~ptnnment of drug activation systems
As a prototype system for activation of cytotoxic
drugs, we chose the thymidine kinase gene from Herpes Sim-
plex Virus type 1 (HSV-1 tk). HSV-1 tk can activate a var-
iety of dideoxynucleoside analogues (ddNs) by conversion of
these drugs to the (5') monophosphate species. The ddNMPs
can act as competitive inhibitors of cellular nucleotide or
nucleoside kinases, causing depletion of cellular nucleo-
tide pools: following conversion to triphosphate species by
15 cellular enzymes, ddNTPs can be incorporated into nascent
DNA (and possibly RNA) strands, causing chain termination.
One of the best-characterized ddNs which can be
activated by HSV-1 tk is acyclovir (acyclo-G). The safety
and efficacy of acyclovir derives primarily from its selec-
tive activation by HSV-1 tk: acyclo-G is converted to
acyclo-GMP several thousandfold more efficiently by HSV-1
thymidine kinase (tk) than by cellular thymidine kinases (Km
for HSV-1 tk = 0.005X Km Vero tk: Vrel HSV-1 tk = 3,OOO,OOOX
Vrel Vero tk). Cellular enzymes then convert acyclo-GMP to
acyclo-GTP, which is incorporated into DNA by HSV-1 DNA
polyaerase causing termination of viral DNA synthesis.
Acyclo-GTP inhibits HSV-1 DNA synthesis at about 1/20 the
concentration required'for similar inhibition of cellular
DNA synthesis (P. A. Furman et al, J Virol (1979) x:72-77).
once a cis-acting regulatory sequence has been
identified, it is tested for effect in a model system. For
example, the cis-acting regulatory sequence may be cloned
into a plasmid with a suitable reporter gene, such as CAT or
-27_ X2047363
8-galactosidase. The plasmid is then transfected into a
suitable cell line (e.g., CHO cells, HeLa, HUT78, and the
like). The traps-acting factor may be supplied by selecting
a host cell line which expresses the traps-acting factor, or
by cotransfecting the host cell line with a plasmid encoding
the traps-acting factor under control of a different pro-
moter (either inducible or constitutive). The transfected
host cells are then assayed for expression of the reporter
gene.
The suitability of an effector gene such as HSV-1
tk is assessed by cloning into a suitable plasmid under con-
trol of the selected cis-acting regulatory region. The
plasmid also preferably contains a selectable marker, allow-
ing one to select those cells which have become genetically
transformed by the vector.
~pnera~ structure of th~,~tidine kinase vectors
The general structure of retroviral vectors which
can be used to generate recombinant retroviruses containing
an effector gene linked to cis-acting regulatory elements is
shown in Figure 1. The basic retroviral vectors for expres-
sion of HSV-1 tk are derived from Moloney Murine Leukemia
Virus (Mo-MuLV). All of the Mo-MuLV genes (gag, pol and
env) have been removed from the vectors in order to generate
completely replication-defective viruses. The remaining
components are required for expression and packaging of
viral RNA, reverse transcription and integration, and tran-
scription of the tk gene (and marker gene, if desired) in
the target cells. These components consist of the Mo-MuLV
Long Terminal Repeats (LTRs), the plus- and minus-strand
primer binding sites, and the RNA encapsidation signal (Psi
sequence).
In order to regulate the expression of HSV-1 tk in
a cell type-specific manner, a hybrid transcription unit is
-2$- ~20473~3
constructed in which the HSV-1 tk coding sequence replaces
the coding sequence of a cell type-specific transcription
unit.
Expression of the HSV-1 tk gene via these vectors
can be accomplished in two ways. The first type of vector
uses a separate enhancer/regulatory element to regulate
expression of HSV-1 tk. In this type of vector, transcrip-
tion of HSV-1 tk proceeds in the opposite orientation to the
plus-sense of the provirus (as shown in Figure 1). The tk
gene is provided with its own polyadenylation signal, and,
if desired, also can be provided with an intron between the
tk coding sequence and polyadenylation signal. Potential
transcriptional interference in this first class of vectors
can be reduced by removing the enhancer/regulatory element
sequences from the 3' LTR of the vector. Because only the
U3 sequences of the 3' LTR are transcribed into viral RNA,
these sequences form both LTRs in the resultant provirus.
This allows integration of the proviral cDNA, but results in
loss of the enhancers and regulatory elements from both LTRs
of the provirus.
In the second type of vector (enhancer replacement
vector), the Mo-MLV LTR provides the enhancer and regulatory
element sequences that control expression of the HSV-1 tk
gene. The Mo-MLV LTR can drive expression of heterologous
genes in a variety of cell types (excluding hematopoietic
stem cells), but is most efficient in T-cells. Control of
cell type-specific gene expression of HSV-1 tk may be
achieved by substitution of specific enhancer/regulatory
element sequences for the enhancer/regulatory element
sequences within the 3' Mo-MLV LTR. The 5' Mo-MLV LTR is
retained in the vector to allow efficient expression of
viral RNA in the packaging cell line used to generate infec-
tious virus.
-29- '. 204~3fi3
A selectable marker gene can be incorporated into
these vectors 3' to the HSV-1 tk gene in order to allow in
vitro selection of transformed cells. Expression of the
marker gene is provided by its own regulatory element. In
general, this regulatory element would be derived from a
moderately active cellular gene that is constitutively
expressed in all tissues. Some examples of such regulatory
elements would be the cytoplasmic B actin promoter, histone
promoters, and promoters for glycolytic enzymes.
The second type of vector uses a separate
enhancer/regulatory element to regulate expression of HSV-1
tk. In this type of vector, transcription of HSV-1 tk pro-
ceeds in the opposite orientation to the plus-sense of the
provirus. The tk gene is provided with its own polyadenyl-
ation signal, and, if desired, also can be provided with~an
intron between the tk coding sequence and polyadenylation
signal. Potential transcriptional interference in this sec-
ond class of vectors can be reduced by removing the
enhancer/regulatory element sequences from the 3' LTR of the
vector. This allows integration of the proviral cDNA, but
results in loss of the enhancers and regulatory elements
from both LTRs of the provirus.
A selectable marker can be provided, as in the
first class of vectors, and is transcribed in the opposite
direction from HSV-1 tk. Transcription thus initiates in
the center of the integrated provirus and proceeds in op-
posite directions, in a manner analogous to the transcrip-
tion of the E2/E3 genes of human adenoviruses.
Formu~atson and Administr~ ion
The polynucleotide constructs of the invention may
be formulated, depending on the form of construct. .For
example, where the vector is a competent (infectious) virus,
it may be formulated in the same manner as live virus vac-
-30- ~20473fi3
cines. Such. formulations are typically buffered, physio-
logic saline for injection, or buffered oral formulations,
either of which may optionally contain antibiotics. Lipo-
some formulations may be prepared by standard methods, for
example by suspending lipids in chloroform, drying the
lipids onto the walls of a vessel, and hydrating the lipids
with a solution containing the polynucleotide construct.
Suitable lipids are known in the art, including phosphatidyl
serine, phosphatidyl glycerol, lethicin, and the like. A
synthetic lipid particularly useful for polynucleotide
transfection is N-[1-(2,3-dioleyloxy)propyl]-N,N,N-tri-
methylammonium chloride, which is commercially available
under the name Lipofectin~ (available from HRL, Gaithers-
burg, MD), and is described by P.L. Felgner et al, Proc Nat
Acad Sci USA (1987) $4:7413.
Recombinant retroviral vectors may also be formu-
lated for in vivo administration (if replication defective).
Vectors which are based on HIV-1 or HIV-2 will demonstrate
the same cellular tropisms as the native virus, thus pro-
2o viding an efficient and comprehensive means for targeting
the appropriate cell populations in vivo when treating HIV-
1 or HIV-2 infection. Vectors based on HIV may employ the
HIV LTR promoters, tar sequence, and the tat gene for con-
trol of expression of heterologous genes. One may refine
the target cell specificity of the vector by an appropriate
choice of packaging constructs (e.g., env gene). For
example; the isolate HIV-1"1~, (M. Quiroga et al, "Modern
Approaches to New Vaccines" (1989, Cold Spring Harbor Lab-
oratories) p.. 80) is naturally selective for macrophage
infection: thus, by packaging the vector in an HIV-1"1"-
derived envelope, targeting to the macrophages can be
effected. Similarly, packaging the vectors using recom-
binant envelope genes which incorporate sequences derived
-31- r 2047363
from envelopes or capsids derived of hepatitis B virus (or
HAV, or HCV) to obtain targeted delivery to hepatocytes. By
using a packaging cell line expressing envelope proteins
derived from HIV-2, one can obtain an effective method for
targeting the vectors of the invention for protection of
TH-cells, due to the affinity of gp160"" for CD4+ (Smith et
al, science (1987) ~,~$:1704-06).
The mode of administration of polynucleotide con-
structs of the invention depends on the nature of the vec-
l0 tor. For example, retroviral vectors may be administered by
inoculation, parenterally or orally. Where the vector is an
infectious recombinant virus, administration may be by par-
enteral injection, oral or intranasal administration, and
the like. Liposome formulations are preferably administered
by intranasal spray or intravenous injection. The presently
preferred method of administration is by incubating defec-
tive retroviral vectors with aspirated autologous bone mar-
row cells or T lymphocytes ex vivo, followed by reintroduc-
tion of the treated cells by standard techniques. Briefly,
bone marrow cells are aspirated by techniques known in the
art for bone marrow transplants, and are generally aspirated
from the pelvis. If desired (particularly in the treatment
of leukemia and other hyperplastic disorders), the subject
may be treated with chemotherapy or radiotherapy following
bone marrow aspiration, in order to reduce the burden of
infected or hyperproliferating cells. The aspirate may be
screened to remove undesirable material (e. g., tumor cells,
bacteria, viral particles, and the like), for example by
immunoprecipitation, immunoadsorption, immunoreaction with
complement'fixation, fluorescence-activated cell sorting
(FMCS), and the like. Ideally, hematopoietic stem cells are
labeled and isolated using stem cell-specific Mobs. The
screened aspirate is then transfected or infected with a
-32- 204763
construct of the invention. One method of infection is to
maintain the aspirated cells in culture in contact with
infectious titers of replication-defective retroviral vec-
tors for a period sufficient to insure efficient inocula-
tion. Growth factors for hematopoietic cells (for example
IL-3) may be added during cocultivation (E. A. Dzierzak,
supra). The calcium phosphate transfection techniques known
for murine cells may be used (see for example, E.A.
Dzierzak, supra: S-F. Yu et al, Proc Nat Acad Scy~USA
(1986) $;~:3194-98). Constructs may also be introduced by
electroporation (S. Mansour, supra). Alternatively, one may
employ a transfection agent such as Lipofectin'. Where the
vector includes a marker, the infected cells may be screened
or selected, if desired. The infected cells are then
reintroduced into the subject using methods employed for
autologous bone marrow transplantation, typically by intra-
venous infusion or injection. One may also transfect or
infect lymphocytes either in vivo or ex vivo using con-
structs of the invention. Infection using MLV-based con-
structs is preferably conducted ex vivo, as the envelope is
inactivated by human complement. Specific subsets of the
lymphocyte population may be selected by employing mono-
clonal antibodies to separate the desired subset, or by
other methods known in the art: see for example P.W.
Kantoff et al, proc Nat Acad Sci USA (1986) $x:6563-67.
If desired, one may convert some of the lympho-
cytes or bone marrow cells to packaging cells, by inserting
independent constructs capable of expressing the viral genes
encoding gag-pol and env, followed by insertion of a vector
of the invention. O. Danos et al, Proc Nat Acad Sci USA
(1988) $x:6460-64. When the autologous packaging cells are
reintroduced into the subject, continuous expression of rep-
lication-defective retrovirus during the cell lifetime
-33- _ 2047363
results in transfer of the therapeutic construct of the
invention to a large number of host cells.
C . E;~;amnles
The examples presented below are provided as a
further guide to the practitioner of ordinary skill in the
art, and are not to be construed as limiting the invention
in any way.
EXAMPLE 1
Recombinant Retrovirai Vectors
Retroviral vectors were constructed using standard
techniques for manipulation of recombinant DNA (T. Maniatis
et al., Molecular Cloning: A Labor~tr"-y M~ (New York,
Cold Spring Harbor Laboratory, 1982)). Plasmids used in~
construction of retroviral vectors were obtained from the
following sources: pMX1112SVNeo was a gift from M. McMann
(UCSF), SV-N was a gift from E. Gilboa (constructed as
described by Yu, supra).
2o Plasmid pMX1112SVNeo is constructed by inserting a
Cla-Cla fragment of the SV40 early promoter linked to the
neon gene into the plasmid pMXi112 (described by A.M.C.
Brown et al, "Retroviral Vectors", pp. 189-212, in "DNA
Cloning: a practical approach, vol. 3 (D.M. Glover, ed, IRL
Press, 1987)). The SV40-neon gene is positioned between the
psi packaging sequence and the 3' Mo-MuLV LTR region on the
plasmid. Plasmid pMXSVNeol8 was prepared from pMX1112SVNeo
by removing the EcoRI site upstream of the 5' Mo-MuLV LTR,
inserting an EcoRI linker at the XhoI site, and adding the
pUCl8 polylinker between the new EcoRI site and the HindIII
site. Plasmid pTARl contains the HIV 5' LTR (including the
tar sequence), a gene encoding CAT, and an SV40 polyadenyl-
ation signal and t intron, and was constructed by cloning
-34- _ 20473fi3
the XhoI-NarI fragment of HIV-1 (SF2) (R. Sanche2-Pescador
et al, science (1982) 22.2:484-92) into pML. A synthetic
polylinker was added at the NarI site, and the Stul-BamHI
fragment of pSV2CAT (C.M. Gorman et al, Proc Nat Acad Sci
~ (1982) 22:6777-81) containing the CAT gene and SV40 RNA
processing signals was inserted to provide pTARi. The pTARl
was cleaved with Asp718 and XhoI, and the resulting fragment
cloned into pMXSVNeol8 to produce pTBi, shown in Figure 2.
Plasmid pTATl was constructed with the HIV tat gene under
control of the SV40 EP, and flanked by the SV40 polyadenyl-
ation signal (Peterlin et al, supra). Plasmid pRT contains
the HSV-1 tk gene, promoter, and RNA processing signals.
The plasmid ptar/tk was prepared by cloning the BglII-EcoRI
fragment of pRT (containing the tk coding sequence and poly-
adenylation signal) into pTARi in place of the CAT gene.
Plasmid pMXSVNeo-tar/tk was constructed by cloning the
Asp718-EcoRI portion of ptar/tk into the pMXSVNeol8 poly-
linker (pMXSVNeo-tar/tk is depicted in Figure 3).
Amphotropic retroviruses capable of infecting
human cells were prepared by standard techniques (R. Mann et
al, Cell (1983) x:153-59). Briefly, the amphotropic pack-
aging cell line PA-317 (Miller et al, Mol Cell Biol (1986)
x:2895-902: ATCC CRL 9078) was transfected with plasmid
vector using the CaP04 coprecipitation technique (R. Graham
et al, Virol (1973) x:456-67). 10 ~g of plasmid vector
plus 25 ~g salmon sperm DNA carrier were applied as a CaPO4
coprecipitate to 5 x 105 PA-317 cells in a 60 mm tissue cul-
ture dish for 8-16 h. The precipitate was removed and the
monolayer was. rinsed with Dulbecco's phosphate-buffered
saline. Fresh medium (Dulbecco's modified Eagles medium,
DMEM, supplemented with 10% fetal bovine serum and antibiot-
ics) was applied and the cells were allowed to grow and
recover for 2 d. The transfected cells were then trypsin-
~ 2 04 X363
- 35 -
ized and transferred to a 150 mm2 T-flask and grown for 10-
14 d in medium containing 0.8 mg/mL 6418. The resulting
6418-resistant colonies were trypsinized and cloned by lim-
iting dilution in 96-well tissue culture plates. Recombin-
ant retrovirus produced by individual clones was character-
ized by infection of human cell lines (HeLa, ATCC CCL 2:
HUT78, ATCC TIB 161) to determine viral titer and generation
of the correct proviral structure.
Ecotropic recombinant retroviruses were prepared
as described above, except that the ecotropic packaging cell
line Psi-2 (R. Mann, supra) was used instead of PA-317.
EXAMPLE 2
Tnfoe~t;~., of cells with recombinant retroviruses
Cell lines were infected with recombinant retro-
viruses using standard procedures (R. Mann, supra).
Briefly, 5 x 106 cells were plated into 60 mm tissue culture
dishes and allowed to grow for 16 h. Cell-free supernatants
from packaging cells containing plasmid vectors was applied
to the target cells for 6-8 h in the presence of 8 ~g/mL
polybrene. In the case of viruses carrying the 6418-resis-
tance marker, cells were grown, selected, and cloned as des-
cribed for preparation of packing clones, above. Retro-
viruses carrying HSV-1 tk, but lacking the 6418-resistance
marker were titered using tk- cell lines: L-M (tk-) (ATCC
CCL 1.3) for ecotropic viruses, 143 H cells (ATCC CRL 8303)
for amphotropic viruses. Selection for tk+ colonies was
performed using HaT medium.
-36- 204 7363
EXAMPLE 3
,ssay For Traps-activatiQnby HIV tat
Plasmids pTARl, pTBl, and pTH2 (having the HIV LTR
region in the opposite orientation from pTBl), encoding CAT
under control of the HIV tar cis-acting regulatory sequence,
were transfected into HeLa cells. Plasmid pTATl, expressing
the HIV tat traps-acting agent, was transfected into half of
the HeLa cells. After 48 hours, the cells were lysed, and
14C-chloramphenicol was incubated with the lysates. The
products were extracted with EtOAc, and analyzed by thin-
layer chromatography.
The results indicated that CAT was expressed in
HeLa cells which contained both a tar-encoding plasmid and a
tat-encoding plasmid, but not in HeLa cells lacking the
pTATi plasmid. Although the level of constitutive expres-
sion (leakiness) varied little between pTARi, pTHl, and
pTB2, CAT expression in response to HIV tat was higher in
cells containing pTBl and pT82 than in cells containing
pTARl (lacking the Mo-MuLV LTRs and SV40 enhancer).
EXAMPLE 4
~.,t; v; ra t Cytotoxic Ef factor Genes
Amphotropic MXSVNeo-tar/tk retrovirus was pro-
duced, and used to infect HUT78 cells (ATCC TIB 161).
Transformed cells ("HUT-tk cells") were selected using
1 mg/tdI. 6418. The sensitivity of mass-cultured HUT-tk cells
to acyClovir was determined by growth in medium containing
various concentrations~of drug. HUT78 cells containing
pMXSVNeo-tar/tk ("HUT-tk cells") were pelleted and exposed
to polybrene (2 ~g/mL) at 37'C for 30 minutes. Cells were
then pelleted and resuspended at 105 cells per mL, and
exposed to HIV-1 (SF2) for 1.5 hours at 37°C. The cells
were then pelleted and resuspended in RPMI 1640 medium con-
- 37 - . 2 04 7363
taining 10%~fetal calf serum and antibiotics at 104 cells
per 50 ~L. Fifty ~L of cell suspension was then distributed
to wells of a 24 well plate containing 1 mL of medium with
45, 100, or 200 ~M acyclovir. HIV-SF2 was added (sufficient
to provide 3 OD in a p25 gag ELISA after 7 days of incuba-
tion), and was allowed to propagate for 7 days, and then
assayed by anti-p25gag ELISA. Normal HUT78 cells served as
controls for the effect of HSV-1 tk. The results of these
experiments are shown in Figure 5.
Although HIV replication in HUT78 cells was not
significantly affected by acyclovir, replication was
strongly inhibited by acyclovir in HUT-tk cells (about 60%
with 100 ~M acyclovir). Cytopathic/cytotoxic effects were
apparent in both cell types (data not shown). Although
analysis of single cell clones of HUT-tk showed variation in
sensitivity to acyclovir, the mass-cultured HUT-tk cells
were capable of significantly restricting viral replication.
EXAMPLE 5
Assay for inhibition of HIV reQlication
HIV-infectable cells transfonaed to an HSV-1 tk
positive phenotype can be used for screening nucleoside ana-
logues for antiviral activity and cellular cytotoxicity
(Mitsuya, supra).
HUT78 cells containing pMXSVNeo-tar/tk ("HUT-tk
cells") were cloned by limiting dilution and assayed for
sensitivity to varying concentrations of acyclovir as a
measure of constitutive tk expression. An especially sen-
sitive clone was designated TK.S, was chosen to test the
ability of cells expressing high levels of tk to activate
AZT and resist HIV infection.
HUT78, mass-cultured HUT-tk (from Example 4), and
TK.S cells were pelleted and exposed to polybrene (2 ~g/mL)
-38- 26~+~363
at 37°C for 30 minutes. Cells were then pelleted and resus-
pended at 105 cells per mL, and exposed to HIV-1 (SF2) for
1.5 hours at 37°C. The cells were then pelleted and resus-
pended in RPMI 1640 medium containing serum and antibiotics
at 105 cells per 50 ~L. Fifty uL of cell suspension was
then distributed to individual wells of a 24 well plate con-
taining 1 mL of medium with 0, 1, 5, or 10 ~M AZT. The
infected cells were culture for 7 days, and virus replica-
tion was then assayed by anti-p25gag ELISA. The results are
set forth in Figure 4.
EXAMPLE 6
~onstruct~on of specific thymidine kinase vectors
The structure of various retroviral vectors con-
taining HSV-1 tk is shown in Figure 1. The functional char-
acteristics and therapeutic applications of these vectors
are described below.
SIN-tar/tk-H4Neo: This vector employs the self-
inactivating vector described by Yu, using the HIV tar cis-
acting sequence and the HSV-1 tk effector gene, and con-
taining neomycin resistance under control of the H4 histone
promoter. This plasmid confers susceptibility to ddNs on
cells infected with HIV.
SIN-CD4/tk (-H4Neo): This vector is identical to
the SIN-tar/tk-H4Neo vector described above, except that the
tar cis-acting sequence is replaced with the cis-acting
sequence which promotes expression of CD4 antigen in TH
cells. The histone promoter/neomycin resistance marker is
optional. This vector confers susceptibility to ddNs on all
CD4+ cells. As the CD4 antigen is currently believed to be
important in the mode of entry for HIV, this vector would
also be useful in the treatment of HIV infection.
39 2 04 7363
SIN-Macl/tk (-H4Neo): This vector is like SIN-
tar/tk-H4Neo, with the HIV tar cis-acting sequence replaced
by the cis-acting sequence regulating Macl antigen expres-
sion in macrophages. The histone promoter/neomycin resist-
s ance marker is optional. This vector confers susceptibility
to ddNs on all macrophages, which also serve as host cells
to HIV.
SIN-tar/rA-H4Neo: This vector employs the HIV-1
tar cis-acting sequence, controlling the ricin A effector
gene, using neon as a selectable marker.
SIN-fos/ppt: This vector employs the self-inact-
ivating vector described by Yu, using the ~g oncogene cis-
acting sequence (R. Treisman Cell (1986) g,~:567-74) using a
plant phosphotransferase effector gene. This vector, in
combination with a ddN, would be useful in the treatment of
~-type malignancies.
SIN-pcna/tnf: This vector uses the cis-acting
sequence regulating Proliferating Cell Nuclear Antigen (des-
cribed by J.E. Selis, Leukemia (1988) x,:561-601) and a tumor
necrosis factor effector gene. Although PCNA is expressed
in all cells, the expression is transient, and occurs only
during cell division. Thus, normal cells infected with the
vector would not produce toxic concentrations of TNF,
whereas neoplastic hyperproliferating cells which are con-
stantly in cell division would express toxic concentrations.
SIN-acg/ifn: This vector employs the cis-acting
regulatory sequence which provides expression of a chorionic
gonadotropin (S. E. Goelz, Science (1985) x,:187-90) in com-
bination with a 8-interferon effector gene. This vector,
like the preceding vector, relies on the fact that acg is
infrequently expressed in normal cells, and would be useful
in cancer treatment.
o- ~ 204 73fi3
SIN-IgG/Rbz: This vector employs the cis-acting
sequence from IgG, along with a ribozyme effector gene spe-
cific for the bcl/abl splice site which occurs in chronic
myelogenous leukemia (CML). As IgG is expressed in e-cells,
cell-type protection against CML is obtained.
SIN-e7/Ld: This vector uses the cis-acting
sequence responding to the E7 gene of human papilloma virus
type 16 (HPV16) in conjunction with a r.~ishmania donovani
purine 2'-deoxyribonucleosidease gene. As expression of
HPV16 E7 gene is associated with carcinoma of the uterine
cervix (Phelps et al, ~ (1988) x:539-47), this vector
(in combination with the drug 6-methylpurine 2-deoxyribo-
side) destroys cells in which the E7 gene is expressed.
HIV-2/tk: This vector is derived from HIV-2 from
which all of the internal genes are deleted except for the
portion of the gag gene which overlaps the psi (encapsid-
ation) signal and the rev-response element from the env
gene. The HSV-1 tk gene is inserted 3' from the psi
sequence, and is expressed under control of the HIV-2 LTR.
If desired, the HIV-1 tar sequence can be substituted for
the HIV-2 tar sequence. This vector may be used to express
HSV-1 tk from a genome-length RNA by deleting or altering
the HIV gag sequence ATG codons (Guild et al, supra),
obtaining expression of the HSV-1 tk gene from its own ATG
start codon. Alternatively, one may include the HIV-2 env
splic~ acceptor just 5' of the HSV-1 tk gene, without alter-
ing tha~gag start codons, thus obtaining expression of the
HSV-1 tk via a spliced'subgenomic mRNA. In either case,
expression of HSV-1 tk is dependent upon complementation by
tat from either a resident HIV-1 provirus or a
superinfecting HIV-1 virus. Expression of tk in combination
with administration of ACV or GCV then activates toxic
levels of the antiviral agent, thus destroying the infected
-41- 2p47363
cell. These constructs may be packaged in either HIV-2
envelopes or in HIV-1$,I" envelopes, the latter being useful
for targeting and destroying infected macrophages.
HIV-2/TCR-tk: This construct incorporates the T-
cell receptor promoter internally to drive expression of
HSV-1 tk, and contains the RRE from env and psi signal from
the 5' end of the HIV-2 gag gene. The 3' LTR of the vector
is modified by inserting the TCR a constant region enhancer
sequence (described by Ho et al, Proc Nat Acad Sci USA
(1989) $x:6714), according the principle disclosed by J.
Overhauser et al, ~ Virol (1985) x:133-34 for construction
of a glucocorticoid-responsive MoMLV. The HIV-2/TCR-tk vec-
tor may be used to infect and genetically transform CD4+
cells in vivo. As neither the vector nor the T4 cell nor-
orally contains tat, no expression of transcripts initiating
in the viral LTR occurs in the absence of infection with
HIV. Initiation of transcription (expression of tk) occurs
from the internal TCR promoter. Expression from this pro-
moter is enhanced by the presence of the TCR a enhancers
incorporated into the proviral LTRs. This vector provides
constitutive expression of tk in the genetically transformed
TH lymphocytes, which in turn activates an antiviral agent
(e.g., ACV, ddC and the like). This permits the use of ACV
and related compounds in cells in which they would normally
not be effective. It also increases the effectivness of AZT
at low concentrations as shown in Figure 4. The ability to
employ~'a variety of antiviral agents should reduce the pos-
sibility of development of drug resistance by HIV. An
analogous vector may be prepared by substituting MLV virus
components~for the HIV sequences used, and employing an MLV
amphotropic packaging cell line.
- 42 -
~ 04 7363
EXAMPLE 7
rn vivo Administration of Constructs
(A) Packaging cell lines for HIV-based vectors
are constructed as follows:
The gag-pol coding sequence is isolated from HIV-
1 or HIV-2, and is inserted into an expression vector under
control of the human cytomegalovirus (CMV) major immediate
early (MIE) promoter. A suitable polyadenylation sequence
is provided 3' from the gag-pol insertion, for example, the
l0 SV40 polyA coding sequence. Hoth the gag-pol and the env
sequences must contain the rev-response element (RRE) pres-
ent in the env coding region, in order to obtain nuclear
export (M. H. Malim et al, Nature (1989) x$:254-57).
The env sequence (including the RRE), which will
determine the ultimate specificity of the finished vector,
is obtained from a selected HIV-l, HIV-2 or SIV-1 isolate,
and is inserted into a separate expression vector under con-
trol of the CMV MIE promoter. In this construction, a B-
globin polyA sequence is employed. An intron may be
inserted 5' to the env coding sequence, optionally contain-
ing a coding sequence for a marker or reporter gene (e. g.,
neor).
Separate similar vectors are constructed to insert
tat cDNA and rev cDNA, under control of the SV40 early pro-
motet. These vectors employ SV40 polyA sequences. By sep-
arating the coding sequences for gag-pol, env, tat, and rev,
the possibility of recombination to regenerate a functional
HIV virus is minimized.
ThE expression vectors are then used to construct
a packaging cell line, by transfection of suitable cell
line, for example NIH-3T3, HeLa, or the like. Upon trans-
fection with a retroviral vector of the invention such as
HIV-2/TCR-tk, a protective construct of the invention is
43
prepared which is capable of infecting the population of
host cells that is infected during bona fide HIV infection
(mainly TH cells). When administered in combination with
suitable antiviral agents such as AZT, ACV, ddC or the like,
this treatment protects infected cells from infectious HIV
replication.
(B) Macrophage/monocyte-trophic constructs are
prepared following the procedure of part A above, but
employing the HIV-1"", env gene to provide env protein.
Upon transfection with a vector of the invention such as
HIV-2/tk, an ablative construct is provided which is capable
of superinfecting and destroying host cells which are reser-
voirs of HIV and infection, particularly macrophages and
monocytes.
