Regulated expression of a diphtheria toxin A-chain gene transfected into human cells: Possible strategy for inducing cancer cell suicide

Abstract

As an alternative to directing plant or bacterial toxins to surface receptors, we are investigating the possibility of killing tumor cells by the expression of an exogenously introduced toxin gene (i.e., cell suicide). Tissue-specific gene regulatory elements might thus be exploited to achieve selective killing. To assess the feasibility of such an approach, we have transfected human cells (HeLa, B-lymphoblastoid, and 293 cells) with plasmids containing the diphtheria toxin A-chain (DT-A) coding sequence. The presence of the DT-A sequence lowered the level of transient expression of chloramphenicol acetyltransferase from a cotransfected plasmid, pSV2cat. This expression level in B-cells was further diminished by the inclusion of an immunoglobulin enhancer in the DT-A plasmid. In cotransfection experiments with a DT-A plasmid lacking an enhancer, chloramphenicol acetyltransferase expression was much more strongly inhibited in 293 cells (which express adenovirus E1A and E1B products) than in the other cell types; furthermore, the presence of the DT-A sequence eliminated recovery of G418-resistant 293 cell transformants after transfection with a plasmid containing the neo selectable marker. These results suggest that cell-specific regulatory mechanisms can be exploited to achieve selective cell killing by expression of an introduced toxin gene.

Full text

1986;46:4660-4664. Published online September 1, 1986.Cancer Res
Ian H. Maxwell, Françoise Maxwell and L. Michael Glode
Cancer Cell Suicide
Transfected into Human Cells: Possible Strategy for Inducing
Regulated Expression of a Diphtheria Toxin A-Chain Gene
Updated Version http://cancerres.aacrjournals.org/content/46/9/4660
Access the most recent version of this article at:
Citing Articles http://cancerres.aacrjournals.org/content/46/9/4660#related-urls
This article has been cited by 17 HighWire-hosted articles. Access the articles at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and .pubs@aacr.orgDepartment at
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions .permissions@aacr.orgDepartment at
To request permission to re-use all or part of this article, contact the AACR Publications
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from

[CANCER RESEARCH 46, 4660-4664, September 1986)
Regulated Expression of a Diphtheria Toxin A-Chain Gene Transfected into
Human Cells: Possible Strategy for Inducing Cancer Cell Suicide1
Ian H. Maxwell,2 FrançoiseMaxwell, and L. Michael Glode
Division of Medical Oncology, University of Colorado Health Sciences Center, Denver, Colorado 80262
ABSTRACT
As an alternative to directing plant or bacterial toxins to surface
receptors, we are investigating the possibility of killing tumor cells by
the expression of an exogenously introduced toxin gene (i.e., cell suicide).
Tissue-specific gene regulatory elements might thus be exploited to
achieve selective killing. To assess the feasibility of such an approach,
we have transfected human cells (HeLa, B-lymphoblastoid, and 293 cells)
with plasmids containing the diphtheria toxin A-chain (DT-A) coding
sequence. The presence of the DT-A sequence lowered the level of
transient expression of chloramphenicol acetyltransferase from a cotrans-
fected plasmid, pSV2cat. This expression level in B-cells was further
diminished by the inclusion of an immunoglobulin enhancer in the DT-A
plasmid. In cotransfection experiments with a DT-A plasmid lacking an
enhancer, chloramphenicol acetyltransferase expression was much more
strongly inhibited in 293 cells (which express adenovirus EIA and E1B
products) than in the other cell types; furthermore, the presence of the
DT-A sequence eliminated recovery of G418-resistant 293 cell transform-
ants after transfection with a plasmid containing the neo selectable
marker. These results suggest that cell-specific regulatory mechanisms
can be exploited to achieve selective cell killing by expression of an
introduced toxin gene.
INTRODUCTION
Proteins which are highly toxic when introduced into mam
malian cells are produced by many species of plants and bacte
ria. For example, mouse L-cells are killed by the introduction
of DT-A3 at a concentration as low as one molecule/cell (1).
Attempts have been made to replace natural, B-chain-mediated
toxin entry (2) by mechanisms dependent on specific cell surface
molecules (e.g., hormone receptors or specific surface antigens)
with a view to obtaining selective lethality for certain tumor
cells. We are exploring an alternative means of exploiting a
natural toxin which does not depend on cell surface molecules.
Instead, DNA coding for DT-A is introduced into cells in
constructs containing tissue-specific, cis-acting, transcriptional
regulatory elements with the intention of achieving lethal
expression in cells containing the factors to which these ele
ments respond.
The production of enzymically active toxin in mammalian
cells by expression of the DT-A gene should result in the
inhibition of further protein synthesis (by ADP ribosylation of
elongation factor 2) (2) and consequent cell death. In this paper,
we present evidence that transfection with DT-A expression
plasmids produces toxic effects which can be modulated in a
cell-specific manner. By enabling the elimination of specific
target cell populations, this approach should find significant
applications in developmental and cell biology and eventually
in cancer therapy.
Received 2/14/86; revised 5/20/86; accepted 5/30/86.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported by grants from Joyce Calvert, the Kyle Dudley Cancer Research
Foundation, and the Tad Beck Foundation; by Grant 84-37 from the Milheim
Foundation for Cancer Research, and by American Cancer Society Institutional
Grant IN-5Y.
1To whom requests for reprints should be addressed.
3The abbreviations used are: DT-A, diphtheria toxin A-chain; CAT, chlor
amphenicol acetyltransferase; MT, metallothionein.
MATERIALS AND METHODS
Plasmids and Cell Lines. DT-A plasmid, pi) 1201 (3), was supplied
by J. Murphy, pSV2/3-globin (4) by Paul Berg, p84H (containing MT
HA) (5) by M. Karin and A. Haslinger, pSER.Alu.a (6) by A. Hayday,
and pSV2cat (7) by B. Howard. 293 cells (8) were from J. Alwine and
the GM4025 B-lymphoblastoid line was from the Human Mutant Cell
Genetic Repository, Camden, NJ. HeLa and 293 cells were maintained
in Dulbecco's modified Eagle's medium with 10% newborn bovine
serum. GM4025 cells were maintained in RPMI 1640 medium with
10% fetal bovine serum. Media also contained gentamicin sulfate (SO
Mg/ml).
Plasmid Constructions (Fig. 1). A parental expression plasmid,
pXMT, was constructed by cloning part of the human metallothionein
IIA (MT IIA) promoter into a pSV2 (4) derivative as follows. pSV2-
327-/3G was constructed from pSV2/3-globin (4) by substituting a
pBR327 DNA fragment (Sall-Pstl) for pBR322 DNA (Pvull-Pstl), thus
removing the poison sequence (9) and introducing an Ym<;lll site. The
BamHl site in the SV40-derived DNA was mutated by partial BamHl
cleavage, filling in, and ligation. An Xmalll-Bglll MT IIA fragment
(—71to +272, relative to the cap site) (5) from plasmid p84H was
cloned into the corresponding sites of the above vehicle to give pXMT.
In the DT gene, a \i/«.V\Isite occurs at the boundary of the A and B
coding regions and a Hhal site is present in the first and second codons
of the mature A-chain (10, 11). pDT201 (3) consists of an 831-base
pair Sau3Al DNA fragment including the DT-A coding sequence,
cloned into pUC8. We cleaved pDT201 with Hhal, blunted the 3'-
overhang with T4 DNA polymerase and then, after Sau3M cleavage,
isolated the »580-base pair DT-A fragment. This was cloned into
pXMT, which had been cleaved at the unique l!am\\\ site (overlapping
the ATG at position +73 of MT IIA), filled in, and then cleaved with
Bglll (at the MT IIA-SV40 junction), to give pTH-1. The blunt end
ligation at the 5'-end joined the MT initiator ATG in-frame to the DT-
A coding sequence and regenerated the BamHl site. To obtain the
frame-shift mutant pTH-2, pTH-1 was cleaved at the unique Acci site
in the DT-A coding sequence (35 codons from the NH2 terminus), filled
in with Klenow DNA polymerase, and religated, resulting in a 2-base
pair insertion. Constructs pTH-3 and pTH-4, containing the 279-base
pair Alul immunoglobulin heavy chain enhancer fragment (6) in the
EcoRl site were made by recombining Pstl DNA fragments from pTH-
1 or pTH-2 and pSER.Alu.a (6). The neo transcriptional unit (contain
ing SV40 promoter, splicing, and polyadenylation signals), isolated
from a derivative of pSV2neo (4), was cloned into the EcoRI site of
pTH-1 or pTH-2 using linkers. pTH-lneo and pTH-leon differ only
in the orientation of the neo unit.
Plasmids were isolated using an alkaline sodium dodecyl sulfate lysis
procedure (12) which efficiently removed bacterial chromosomal DNA.
In most preparations, ribosomal RNA was then removed by precipita
tion with NaCl (~4 M at 0°Covernight) and, after phenol plus chloro
form extractions, plasmid was excluded from Sepharose CL2B. In some
preparations, plasmids were purified by isopyknic centrifugation in
CsCl plus ethidium bromide (13). Equivalent results were obtained in
transfections with the same plasmids purified by either procedure.
Transfections. HeLa and 293 cells were transfected (5 x 10s cells/
transfection) with plasmid DNA mixtures using calcium phosphate
coprecipitation (7, 14), including, for HeLa, a 20% v/v glycerol shock
(7). GM4025 cells were transfected (1 x IO7 cells/transfection) using
DEAE-dextran, followed by exposure to chloroquine (15, 16).
For assays of transient expression of CAT activity, cells were har
vested 48 h after transfection and extracts were prepared by freezing
and thawing 3 times. CAT activity was determined (7) by acetylation
of [l4C]chloramphenicol (0.1 nC\ at «45fÃ-Ci/^mol/150 n\ reaction)
4660
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from

TOXIN GENE TRANSFECTION AND EXPRESSION
using samples containing 100-200 ¿igprotein (Bio-Rad assay). The
percentage of acetylation of the [l4C]chloramphenicol was determined
by scintillation counting of spots excised from the thin layer Chromat
ographie plate.
For selection of G418-resistant transformants, cells were trypsinized
2 days after transfection and replated in medium containing G418 (4)
(nominal concentration of active fraction, 400 ^g/ml). Macroscopic
colonies were counted after 17 days.
RESULTS
We constructed DT-A expression plasmids using a truncated
metallothionein promoter for DT-A transcription, to allow the
possibility of heavy metal-regulated expression. Similar trun
cation of the MT IIA promoter was reported to minimize basal
transcription while retaining metal response (5); however, as
described below, we obtained evidence that the basal transcrip
tion level from this promoter was sufficient for significant
expression of DT-A without exposure of transfected cells to
heavy metal ions.
Plasmid constructs are shown in Fig. 1. The mature DT-A
coding sequence from the plasmid pDT201 (3) was inserted
into an expression plasmid to give the construct pTH-1. Deriv
atives of pTH-1 were constructed (see Fig. 1) containing either
an immunoglobulin heavy chain enhancer (pTH-3, pTH-4), or
a neo transcription unit to permit selection with the antibiotic
G-418. Derivatives designated with even numbering (pTH-2,
pTH-4, pTH-2neo) contain a frameshift mutation in the DT-A
coding sequence.
* MT IIA »DT-A
pSV2ß-globin ¥ SV2-327-BG » pXMT » pTH~l
Promoter
(Hh)
XBAUlr>T
A(Bg)(B)
|
ATG
nTH-7 FRAMESHIFT
' (A)
AATAAA
pTH-3 ENHANCER I
DT-A
pTH-4 FRAMESHIFT
(A) ENHANCER [
DT-A J_
Fig. 1. DT-A expression plasmids. Plasmids with odd numbering (pTH-1,
pTH-3) contain the wild-type DT-A coding sequence; even numbering (pTH-2,
pTH-4) indicates a frame-shift mutation has been introduced into the DT-A. The
coding region for mature DT-A is joined to a 5'-truncated metallothionein
promoter DNA fragment (from the human MT IIA gene) (5) which also supplies
the ATG translation initiation codon. pTH-1 codes for a DT-A product differing
from authentic DT-A (193 amino acids) by 4 amino acid residues (2 at each
terminus), pilli and pTH-4 contain a 279-base pair Alul fragment from human
immunoglobulin heavy chain DNA shown to have enhancer activity in B-cells
(6). pTH-1 neo and pTH-2neo (not shown) contain the neo gene to permit selection
of transformants resistant to G418 (4). , pBR327 I )N.\: •.MT HA promoter
region: O, DT-A DNA; D, SV40 DNA. Positions of the initiation codon (ATG),
SV40 splice (A.), and polyadenylation signal (AATAAA), and the ampicillin
resistance gene (AMP') are indicated, as are relevant restriction sites (X, A'molll;
Hh, Hhai; B, BamW; A, Acci; Bg, Bglll; S, Sau3M; E, EcoRl). Sites destroyed
in the constructions are shown in parentheses.
We first sought evidence for a toxic effect of the transfected
DT-A gene using transient expression systems. Since routinely
available procedures result in transfection of a relatively small
fraction of cells, we used cotransfection with a plasmid contain
ing the cat gene (7). Expression of DT-A should result in
inhibition of protein synthesis and therefore of CAT expression
in cotransfected cells. Similar experiments have been used to
study activation (14, 17) or inhibition (14, 18) of transcription
from specific promoters by other fra/iî-actingproducts. As
shown in Fig. 2a and Table 1, the level of CAT expression in
HeLa cells two days after transfection was inhibited by 80-97%
when 5 ng pSV2cat (7) was cotransfected with an equal mass
of a DT-A-expressing plasmid (pTH-1, pTH-lneo, or pTH-3).
Concentration-dependent inhibition was observed with these
plasmids (Fig. 2a, tracks 3-5), inhibition being substantial with
as little as 0.3 ¿¿g(Table 1). In contrast, cotransfection with a
plasmid lacking the DT sequence (Fig. la, tracks 6 and 7), or
with the frame-shift mutants pTH-2 or pTH-4, did not inhibit
CAT expression from pSV2cat (Table 1). These mutants were
constructed by introducing a frame shift into the early part of
the DT-A coding sequence (see Fig. 1) to eliminate production
of active DT-A. Failure of these mutants to inhibit CAT expres
sion supports the conclusion that the inhibition observed with
the corresponding DT-A expressing plasmids was due to active
DT-A production, since the mutants differed in structure only
by a 2-base pair insertion.
To assess activity of the truncated promoter, a construct
analogous to pTH-1 (designated pCAT-1) was made in which
the cat gene replaced the DT-A coding sequence. Transfection
of this construct into HeLa cells indicated a significant basal
level of expression (0.5-3% of that from the SV40 early pro
moter, as determined in parallel transfections with pSV2cat)
(results not shown). Exposure of the cells to cadmium chloride
(2.5-5 MM)during the transient expression period stimulated
CAT expression 1.5- to 6-fold. As shown above, we observed
strong inhibition of CAT expression in cotransfection experi
ments with pTH-1 without exposure of the cells to heavy metal
ions; this inhibition was increased only slightly (1- to 2-fold) by
cadmium (not shown). The basal activity of the truncated MT
IIA promoter therefore appeared sufficient to generate inhibi
tory levels of DT-A. Although cadmium response was minimal,
we were able to demonstrate other means of regulation of DT-
A expression from the truncated MT IIA promoter in the
following experiments.
Several cellular genes are associated with enhancers (cis-
acting DNA sequences) which activate transcription in response
to tissue-specific frans-acting factors (19-22). We wished to
determine whether linkage with an enhancer could direct tissue-
specific expression of the DT-A gene; therefore, we transfected
a human B-lymphoblastoid cell line with pSV2cat together with
pTH-1 or pTH-2 or their derivatives into which a human Ig H
chain gene enhancer (6) had been cloned (pTH-3 and pTH-4;
see Fig. 1). Immunoglobulin enhancers are preferentially active
in B-cells and are among the best characterized cellular ele
ments regulating tissue-specific gene expression (6, 23, 24). To
confirm the effectiveness of the enhancer in the B-cell line we
made a plasmid derivative (pCAT-3) corresponding to pTH-3
in which the cat gene replaced the DT-A sequence (again using
the truncated MT promoter). CAT expression from pCAT-3 in
the B-cells was >25-fold that from pCAT-1 at 48 h posttrans-
fection (results not shown). In contrast, pCAT-1 and pCAT-3
gave approximately equal expression when transfected into
HeLa cells (not shown).
The effects of cotransfecting B-cells with pSV2cat and pTH-
4661
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from

TOXIN GENE TRANSFECTION AND EXPRESSION
(a) HeLa
% ACETYLATED
83 91 33 19 7 83 80
(b)B-cells
51 43 30
3450.3
1pg pTH-lneo56
70.5
5ugpSV2-327neo
%ACETYLATED
24 8 22 9 0.8 33 39
Table 1 Transient CAT activity obtained after cotransfection ofpSV2cat with
plasmids containing wild-type or frame-shift mutant DT-A gene
Cells of each type were transfected as described in "Materials and Methods"
with 5 tig pSV2cat together with the indicated amounts of other plasmids. Results
are from multiple experiments using two independent preparations each of pTH-
1, -2, -3, and -4. Values are percentage of control CAT activity observed in the
same experiment with pSV2cat alone. Where >4 determinations were made, the
mean ±SD is shown.
HeLa cells, % of control CAT activitywithpSV2-327neo
pTH-2neo
pTH-2
pTH-4
pTH- 1neo
pTH-I
pTH-30.3
»g38
42 ±15 (8)°
56 ±23 (7)0.5«94
106,9023,46,
17
541
Pg65,
120
84
19
46± 12(11)
38 ±9(11)s««91137,
52
121, 179
89, 104,81
8, 7, 3, 3
10, 17, 15, 16
11,13,20B
cells, % of control CAT activitywithpTH-2
pTH-4
pTH-I
pTH-30.1
Mg65,
147
47,700.3
Mg148107
62 ±10(7)
33 ±13(7)1
Mg70,
105
83, 129
30 ±16 (7)
16 ±11 (6)3*ig69
56
15,21, 17
6,7,11293
cells, % of control CAT activitywithpTH-2
pTH-10.03
/ig180.1 Mg60.3 «97
5,3,3Ifg90 0.4, 1
" Numbers in parentheses, number of determinations.
•1•
•2
30.1pg•
•t•50.3
1
pTH-1•
60.1pg•7•8•
90.3
1 Ipg
pTH-3 pTH-2•
10Ipg
pTr
(c) 293 cells
ACETYLATED
28 2.1 1.7 32 30 27
®
0.03 0.1
pg pTH-1 0.3 0.3 1
pg pTH-2
Fig. 2. Inhibition of CAT expression from pSV2cat in (a) HeLa cells, (e) B-
lymphoblastoid cells, and (c) 293 cells by cotransfection with plasmids containing
the wild-type DT-A coding sequence. Cells were transfected with DNA mixtures
containing 5 «KpSV2cat and the indicated amounts of other plasmids or with 5
pg pSV2cat alone (tracks 1 and 2 in a and h and truck* 1 and 5 in c). The
autoradiogram of the thin layer chromatogram from the CAT assay is shown.
1, -2, -3, or -4 are shown in Fig. 2b and Table 1. The H-chain
enhancer increased inhibition of CAT expression such that ~3-
fold more pTH-1 than pTH-3 was required for the same level
of inhibition (Table 1; and compare Fig. 2b tracks 3-5 with 6-
8). Neither pTH-2 nor its derivative pTH-4 (DT-A frame-shift
control plasmids with or without enhancer) was inhibitory when
cotransfected in amounts up to 1 /¿g:the relatively minor
inhibition apparent with 3 /<gof these plasmids (Table 1) might
result from competition for cellular transcript ¡miaifactors (25,
26). In similar cotransfections of HeLa cells pTH-3 was no
more effective than pTH-1 in inhibiting CAT expression from
pSV2cat (Table 1). We conclude that the expression of a toxin
gene can be regulated in a cell-specific manner by linkage with
an enhancer. We are currently investigating whether preferen
tial DT-A expression in B-cells can be increased by substituting
an immunoglobulin promoter (27-29) for the metallothionein
promoter in pTH-3.
To explore further the possibility that trans-acting factors
modulating promoter or enhancer activity might serve to re
strict toxin gene expression to specific cell types, we performed
cotransfection experiments with 293 cells, a human embryonic
kidney cell line expressing the EIA and El B products of
adeno virus type 5 (8). The EIA product increases the transcrip-
tional activity of adenovirus early and certain cellular promoters
(14, 30) while repressing certain other promoters (14, 18). As
shown in Fig. 2c and Table 1, pTH-1 inhibited CAT expression
from cotransfected pSV2cat much more strongly in 293 than
in HeLa cells (Fig. 2a). As little as 30 ng pTH-1 inhibited CAT
expression from 5 ng pSV2cat by 80% (Fig. 2c, track 2), while
up to 1 ng of the frame-shift DT-A control plasmid pTH-2 did
not inhibit significantly (tracks 6 and 7). This suggests that the
greater inhibition observed in 293 cells resulted from increased
transcription from the MT HA promoter relative to the SV40
early promoter of pSV2cat so that the cells expressed a higher
ratio of DT-A to CAT enzyme. These effects may involve
differential activation/repression of the 2 promoters by the EIA
4662
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from

TOXIN GENE TRANSFECTION AND EXPRESSION
Table 2 Frequency ofC4I8 resistant transformants obtained after transfection of
293 cells with neo plasmids containing wild-type orframe-shift mutant DT-A
gene
293 cells were transfected with plasmids (10 pg) containing the neo transcrip
tion unit. Plasmids used (see Fig. 1) were: controls, pSV2-327neo (no DT-A) and
pTH-2neo (frame-shift mutant DT-A); with wild-type DT-A, pTH-1 neo and pTH-
leon (pTH-1 derivatives differing only in the orientation of the neo transcription
unit).
PlasmidNone
pSV2-327neo
pTH-lneo
pTH-leon
pTH-2neoNumber
ofcolonies064
0,0
0
79,71
product (14, 18), in addition to the unusual stabilization of
transfected plasmid DNA observed in 293 cells (14).
To obtain additional evidence for toxicity resulting from
expression of the DT-A gene, we transfected 293 cells with neo-
containing derivatives of pTH-1 or pTH-2 and selected for
transformants resistant to G418 (4). As shown in Table 2, no
transformants were recovered after transfection with the DT-
A-expressing plasmids pTH-lneo and pTH-leon; in contrast,
the DT-A frame-shift mutant pTH-2neo gave a similar trans
formation frequency («1.4x 10~4)to that observed with a neo
plasmid containing no DT-A sequence. We infer that failure to
recover transformants from the pTH-1 derivatives resulted from
a lethal level of expression of DT-A in the transfected cells;
however, in similar experiments with HeLa cells (results not
shown), transformants were obtained with pTH-lneo with an
average frequency approximately one-half of that found with
pTH-2neo (which was ~1 x 10~5);thus, a minimal level of DT-
A expression such as that obtained by transcription from the
truncated MT HA promoter in HeLa may not necessarily be
incompatible with cell survival. This conclusion suggests the
feasibility of protecting non-target cells from toxicity.
DISCUSSION
Our findings of inhibition of gene expression from a cotrans-
fected plasmid and lowered transformation frequency are con
sistent with the conclusion that transfection of an efficiently
expressed DT-A gene is lethal. We have demonstrated two ways
in which the expression of a transfected DT-A gene can be
modulated in different cell types: (a) expression was increased
in B-cells but not in HeLa cells when an immunoglobulin
enhancer was present in the construct; (b) expression from a
weak promoter, lacking known enhancer elements, was much
stronger in 293 cells than in HeLa or B-cells. We have so far
demonstrated preferential rather than specific expression of
DT-A in different cell types. Rapidly increasing knowledge of
tissue-specific gene control mechanisms will be directly appli
cable to designing efficient means of directing the expression
of exogenous genes including DT-A to specific cell types. At
tenuated DT-A mutants are available (1, 2), should control of
basal expression of the wild-type gene prove inadequate to avoid
toxicity in non-target cells. The achievement of tightly con
trolled tissue-specific lethality following introduction of a toxin
gene into cells is therefore a realistic near-term goal.
With regard to the eventual application of toxin gene therapy
to cancer, our results indicate that immunoglobulin enhancers
might be used to contribute to target specificity in treatment of
B-cell neoplasms. Although normal B-cells would also be sus
ceptible, the toxicity to other dividing cell populations seen
with current therapeutic agents would be avoided. Tissue-spe
cific enhancers from other genes coding for "tumor-specific"
products might similarly be harnessed in directing expression
of a toxin gene specifically in tumor cells. Promising candidates
would include genes coding for products normally expressed
only in fetal tissues but frequently expressed ectopically in
certain tumors (31). The enhanced DT-A expression we ob
served in 293 cells probably resulted at least partly from the
presence of adenovirus EIA products. It may therefore also be
possible to exploit the abundant expression in certain tumors
(32) of proteins functionally analogous to EIA (e.g., the c-myc
product) (33, 34) to achieve selective killing by a toxin gene.
These concepts potentially offer a novel approach to cancer
therapy less subject to undesirable side effects than current
procedures. It is clear that the success of such therapy will
depend on the development of efficient means of delivering
exogenous genes to tumor cells both in terms of access to and
efficient uptake by these cells. Intensive work on recombinant
viruses has produced vehicles capable of efficient transduction
of foreign genes into cells In vitro (35-38). Such defective
recombinant viruses should be applicable to the delivery of
toxin gene therapy to accessible populations of cancer cells,
e.g., in bone marrow, or in skin or bladder cancers. Although
systemic delivery of toxin gene vectors to solid tumors or
métastaseswill be more difficult, this problem may also even
tually be overcome using viral vehicles by exploiting the mech
anisms evolved by many viruses for efficiently infecting tissues
remote from their site of entry into the organism.
ACKNOWLEDGMENTS
We thank Paula Lane for skillful assistance.
REFERENCES
1. Yamaizumi, M., Mekada, E., Uchida, T., and Okada, Y. One molecule of
diphtheria toxin fragment A introduced into a cell can kill the cell. Cell, 15:
245-250, 1978.
2. Pappenheimer, A. M., Jr. Diphtheria toxin. Ann. Rev. Biochem., 46:69-94,
1977.
3. Leong, I>., Coleman, K. I)., and Murphy, J. R. Cloned fragment A of
diphtheria toxin is expressed and secreted into the periplasmic space of
Escherichia coli K12. Science (Wash. DC), 220: 515-517, 1983.
4. Southern, P. J., and Berg, P. Transformation of mammalian cells to antibiotic
resistance with a bacterial gene under control of the SV40 early region
promoter. J. Mol. Appi. Genet., /.- 327-341, 1982.
5. Karin, M., Haslinger, A., Holtgreve, H., Richards, R. I., Krauter, P., West-
•phal,H. M., and Beato, M. Characterization of DNA sequences through
which cadmium and glucocorticoid hormones induce human metallothionein-
HAgene. Nature (Lond.), 308: 513-519, 1984.
6. Hayday, A. C, Gillies, S. p., Saito, H., Wood, C., Wiman, K., Hayward, W.
S., and Tonegawa, S. Activation of a translocated human c-myc gene by an
enhancer in the immunoglobulin heavy-chain locus. Nature (Lond.), 307:
334-340, 1984.
7. Gorman, C. M., Moffat, L. F., and Howard, B. H. Recombinant genomes
which express chloramphenicol acetyltransferase in mammalian cells. Mol.
Cell. Biol., 2:1044-1051, 1982.
8. Graham, F. L., Smiley, J., Russell, W. C., and Nairn, R. Characteristics of a
human cell line transformed by DNA from human adenovirus type 5. J. Gen.
Viro!., 56:59-72, 1977.
9. Lusky, M., and Botchan, M. Inhibition of SV40 replication in simian cells
by specific pBR322 DNA sequences. Nature (Lond.), 293: 79-81, 1981.
10. Greenfield, L., Björn,M. J., Horn, G., Fong, D., Buck, G. A., Collier, R. J.,
and Kaplan, D. A. Nucleotide sequence of the structural gene for diphtheria
toxin carried by corynebacteriophage ß.Proc. Nati. Acad. Sci. USA, 80:
6853-6857, 1983.
11. Kaczorek, M., Delpeyroux, F., Chenciner, N., Streeck, R. E., Murphy, J. R.,
Boquet, P., and Tiollais, P. Nucleotide sequence and expression of the
diphtheria w.v228 gene in Escherichia coli. Science, 227:855-858, 1983.
12. Ish-Horowicz, D., and Burke, J. F. Rapid and efficient cosmici cloning.
Nucleic Acids Res., 9: 2989-2998, 1981.
13. T. Maniatis, E. F. Fritsch, and J. Sambrook (eds.). Molecular Cloning, p.
93. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982.
14. Alwine, J. C. Transient gene expression control: effects of transfected DNA
stability and trans-activation by viral early proteins. Mol. Cell. Biol., 5:1034-
1042, 1985.
15. McCutchan, J. H., and Pagano, J. S. Enhancement of the infectivity of
4663
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from

TOXIN GENE TRANSFECTION AND EXPRESSION
simian virus 40 deoxyribonucleic acid with diethylaminoethyl-dextran. J.
Nail. Cancer Inst., 41: 351-357, 1968.
16. Luthman, II., and Magnusson, G. High efficiency polyoma DNA transfection
of chloroquine treated cells. Nucleic Acids Res., 11: 1295-1308, 1983.
17. Kingston, R. E., Baldwin, A. S., Jr., and Sharp, P. A. Regulation of heat
shock protein 70 gene expression by c-myc. Nature (Lond.), 312: 280-282,
1984.
18. Velcich, A., and Ziff, E. Adenovirus Eia proteins repress transcription from
the SV40 early promoter. Cell, 40: 705-716, 1985.
19. Gillies, S. D., Folsom, V., and Tonegawa, S. Cell type-specific enhancer
element associated with a mouse MHC gene, Eß.Nature (Lond.), 310: 594-
597, 1984.
20. Walker, M. D., Edlund, T., Boulet, A. M., and Rutter, W. J. Cell-specific
expression controlled by the 5'-flanking region of insulin and chymotrypsin
genes. Nature (Lond.), 306: 557-561, 1983.
21. Ornitz, D. M., Palmiter, R. D., Hammer, R. E., Brinster, R. L., Swift, G.
II., and MacDonald, R. J. Specific expression of an elastase-human growth
hormone fusion gene in pancreatic acinar cells of transgenic mice. Nature
(Lond.), 5/5:600-602,1985.
22. Ephrussi, A., Church, G. M., Tonegawa, S., and Gilbert W. B-Lineage-
specific interactions of an immunoglobulin enhancer with cellular factors in
vivo. Science (Wash. DC), 227: 134-140, 1985.
23. Gillies, S. D., Morrison, S. L., Oi, V. T., and Tonegawa, S. A tissue-specific
transcription enhancer element is located in the major intron of a rearranged
immunoglobulin heavy chain gene. Cell, 33: 717-728, 1983.
24. Picard, D., and Schaffner, W. A lymphocyte-specific enhancer in the mouse
immunoglobulin k gene. Nature (Lond.), 307: 80-82, 1984.
25. Mercóla, M., Goverman, J., Mirell, C., and CaÃ-ame,K. Immunoglobulin
heavy-chain enhancer requires one or more tissue-specific factors. Science
(Wash. DC), 227: 266-270, 1985.
26. Seguin, C., Felber, B. K., Carter, A. D., and Hamer, D. H. Competition for
cellular factors that activate metallothionein gene transcription. Nature
(Lond.), 312: 781-785, 1984.
27. Foster, J., Stafford, J., and Queen, C. An immunoglobulin promoter displays
cell-type specificity independently of the enhancer. Nature (Lond.), 315:423-
425, 1985.
28. Mason, J. O., Williams, G. T., and Neuberger, M. S. Transcription cell type
specificity is conferred by an immunoglobulin v,, gene promoter that includes
a functional consensus sequence. Cell, 41:479-487, 1985.
29. Grosschedl, R., and Baltimore, D. Cell-type specificity of immunoglobulin
gene expression is regulated by at least three DNA sequence elements. Cell,
47:885-897, 1985.
30. Stein, R., and Ziff, E. B. HeLa cell /3-tubulin gene transcription is stimulated
by adenovirus 5 in parallel with viral early genes by an E1a-dependent
mechanism. Mol. Cell. Biol., 4: 2792-2801, 1984.
31. Mclntire, K. R. Tumor markers: how useful are they? Hosp. Pract., 19: 55-
68, December 1984.
32. Little, C. D., Nau, M. M., Carney, D. N., Gazdar, A. F., and Minna, J. D.
Amplification and expression of the e-myc oncogene in human lung cancer
cell lines. Nature (Lond.), 306: 194-196,1983.
33. Land, H., Parada, L. F., and Weinberg, R. A. Tumorigenic conversion of
primary embryo fibroblasts requires at least two cooperating oncogenes.
Nature (Lond.), 304: 596-602, 1983.
34. Ruley, H. E. Adenovirus early region 1A enables viral and cellular transform
ing genes to transform primary cells in culture. Nature (Lond.), 304: 602-
606, 1983.
35. Tratschin, J.-D., West, M. H. P., Sandbank, T., and Carter, B. J. A human
parvovirus, adeno-associated virus, as a eukaryotic vector: transient expres
sion and encapsidation of the prokaryotic gene for chloramphenicol acetyl-
transferase. Mol. Cell. Biol., 4: 2072-2081, 1984.
36. Karlsson, S., Humphries, R. K., Gluzman, Y., and Nienhuis, A. W. Transfer
of genes into hematopoietic cells using recombinant DNA viruses. Proc.
Nati. Acad. Sci. USA, 82: 158-162, 1985.
37. Miller, A. D., Law, M.-F., and Verma, I. M. Generation of helper-free
amphotropic retroviruses that transduce a dominant-acting, methotrexate-
resistant dihydrofolate reducÃ-asegene. Mol. Cell. Biol., 5:431-437, 1985.
38. Cone, R. D., and Mulligan, R. C. High-efficiency gene transfer into mam
malian cells: generation of helper-free recombinant retrovirus with broad
mammalian host range. Proc. Nati. Acad. Sci. USA, 81: 6349-6353, 1984.
4664
American Association for Cancer Research Copyright © 1986 on July 10, 2011cancerres.aacrjournals.orgDownloaded from