|
|
|
|||||||
|
|||||||||
(Received for publication, October 12,
1994) From the
In this report we describe a novel gene delivery system using
malaria circumsporozoite (CS) protein as a specific ligand. The CS
protein covers the entire surface of sporozoites of malaria parasites.
Previous studies have demonstrated that intravenously injected CS
protein binds specifically to the basolateral surface of hepatocytes
within minutes, indicating the high hepatocyte specificity of CS
protein. This characteristic of CS protein prompted us to explore the
possibility of using this protein as a liver-specific ligand for
hepatic gene delivery vehicle. As an initial step, we investigated the
efficacy of CS protein-mediated gene transfer into primary hepatocytes
as well as established cell lines. Recombinant CS proteins were
chemically conjugated to poly(L-lysine). The CS conjugates
were complexed with recombinant plasmid DNA carrying a reporter gene.
When the DNA complex was used to transfect primary hepatocytes, a very
low level of expression of the reporter gene was observed. The level of
expression was greatly enhanced when the cells were cotransfected with
adenovirus, which presumably releases the internalized DNA from
endosomal entrapment. The CS-mediated gene transfer into the cells
required region II+, an evolutionarily conserved amino acid
sequence conferring the binding of CS protein to its receptor. CS
protein also efficiently mediated gene transfer into a number of cell
lines, i.e. HepG
There has been an increasing interest in the development of gene
therapy using receptor-mediated gene delivery systems in recent years (1) . Receptor-mediated gene targeting takes advantage of the
selective uptake of macromolecules by receptor-mediated endocytosis in
almost all eukaryotic cells. Because of the cell type-specific
distribution of various receptors, the delivery can be highly
specific(2) . Furthermore, in comparison with viral delivery
systems(1) , receptor-mediated gene delivery allows a greater
flexibility of DNA in terms of size and sequence selections because the
DNA to be delivered does not need to be packaged into viral capsids,
avoiding tedious clonal selection and virus production processes. These
characteristics make the system attractive to human gene therapy. Receptor-mediated gene targeting vehicles consist of two components:
a cell receptor-specific ligand and a polycationic moiety, e.g. polylysine. The polycationic moiety serves as an intermediate for
electrostatic interaction with DNA, resulting in the formation of
toroid complex when the negative charge of the DNA molecule is
completely neutralized. The toroid complex can be internalized via
normal receptor-mediated endocytosis. Several ligands have been used
for receptor-mediated gene transfer. The most extensively characterized
ligands are asialoorosomucoid (ASOR) ( To fully develop the
capacity of receptor-mediated gene targeting, it is necessary to
explore as many cell-specific ligands as possible. The malaria
circumsporozoite (CS) protein is a potential candidate for hepatic gene
delivery. Malaria is transmitted by the bite of infected female
anopheles. Minutes after infection, the malaria sporozoites invade
hepatocytes. This liver-specific invasion is mainly attributed to the
circumsporozoite protein (CS), which densely coats the outer surface of
sporozoite(7, 8) . CS protein recognizes its receptor
that is predominantly distributed at the basolateral domain of
hepatocytes (9) . The specificity of cell recognition and rapid
invasion prompted us to explore the possibility of using this protein
as a liver-specific ligand for hepatic gene delivery vehicle. The
initial work has to establish the formulations of CS-polylysine carrier
for the efficient delivery of DNA into primary hepatocytes. We report
here that CS conjugates can efficiently deliver recombinant plasmid DNA
into primary hepatocyte cultures as well as into a number of
established cell lines, demonstrating the potential application of this
molecular conjugate for in vitro and in vivo gene
delivery.
CHO, HeLa, NIH3T3, and HepG2
cells were grown in Dulbecco's modified Eagle's medium
supplemented with 10% FCS. HL-60 and K562 cells were maintained in RPMI
1640 medium containing 10% FCS. Cells were maintained in a 37 °C
incubator containing 5% CO
Prior to protein
uptake assay, the culture medium was replaced with 0.5 ml of serum-free
medium. 2.5 µg of [ To determine uptake of CS
conjugate-DNA complex, pCMV-
Figure 1:
Schematic representation of the
recombinant CS proteins. The major functional regions of CS protein,
region I (RI), region (II+), and repeat region (REPEATS), which contains 21 (NANP) amino acid residues are
indicated. The numbers shown above the constructs represent
amino acid sequences from P. falciparum CS protein.
DHFR-CSF1-His
Figure 2:
Purification of CS/polylysine and
transferrin/polylysine conjugates. CS27IVC-His
Aliquots from each
fractions were determined by acid-urea polyacrylamide gel
electrophoresis. Under these electrophoretic conditions, free
polylysine and polylysine-polylysine conjugates (not shown) migrated
more rapidly than did unconjugated CS protein and CS-polylysine
conjugate (Fig. 2D), because of their high contents of
positive charges. As shown in Fig. 2D, fractions eluted
in the front peak (fractions 7-10) of the CS conjugate sample
contained no detectable free polylysine, whereas, as expected,
polylysine and/or polylysine-polylysine conjugates were only seen in
the second peak (fractions 11-20). Proteins eluted in the front
peak had a very slow mobility in this gel system, i.e. barely
entering the stacking gel. These results suggest that the molecular
size of these protein conjugates are rather large. The second peak also
contained CS-polylysine conjugates, as suggested from their mobilities
in the acid-urea gel electrophoresis (Fig. 2D, arrows). However, under the optimized transfection conditions
(see below), only the conjugates from the front peak showed
transfection activities (Fig. 2E). The exact
molecular mass of the conjugates in these two peaks have not been
vigorously determined. The molar ratios of CS:polylysine in these types
of conjugate preparations were not known. However, the results shown
here were highly reproducible (five independent experiments).
Furthermore, similar results were obtained in the preparations of
functional transferrin- (Fig. 2C, F, and G) and ASOR-poly-L-lysine (
To demonstrate that the CS-mediated DNA transfection seen in the
cultured primary hepatocytes was ligand-dependent and to optimize
conditions for the transfection efficiency. Two parameters were
investigated: CS conjugate to DNA ratio and the concentration of
adenovirus. Six µg of reporter DNA were complexed with various
amounts of CS conjugate. Hepatocyte cultures were transfected with
these complexes. Fig. 3shows that a maximal transfection
efficiency of about 80% was achieved when 0.14 nM of CS
conjugate was used, corresponding to a CS conjugate:DNA ratio of 116:1.
This optimal ratio of CS conjugate to DNA was comparable with those
published using transferrin (4) and ASOR (15) conjugates. In contrast, within the same ranges of
carrier:DNA ratios, BSA conjugates prepared under the same conditions
showed no better than 10% of transfection efficiency. Likewise, less
than 10% of transfection efficiency was seen with polylysine as a
carrier. These results suggested that the transfection efficiency seen
in the CS carrier is ligand-dependent.
Figure 3:
Effects of CS-polylysine conjugate
concentrations on the expression of
It is of importance to
investigate whether conjugates prepared by various
ligand:polylysine:EDC ratios would affect the biological activities. To
this end, we carried out conjugation reactions with fixed concentration
of CS protein with increased concentrations of polylysine. To derive
the reactions to completion because of increased polylysine, EDC
concentrations were increased proportionally. Table 1showed that
conjugates prepared at the 1:1 to 1:2.5 molar ratios of CS
protein:polylysine, greater than 80% of transfection efficiencies were
achieved. However, the transfection efficiency was drastically reduced
at molar ratio 1:5.0. The reason of this reduced transfection
efficiency could be due to overmodification of the CS protein with
polylysine or EDC, thereby reducing the affinity of CS protein to its
receptors.
To determine the optimal viral concentration for
transfection, conjugate:DNA complex at a ratio of 116 to 1 was added to
primary hepatocyte cultures followed by addition of different amounts
of virus particles. As shown in Fig. 4, increasing adenovirus
particles resulted in increased transfection efficiency with optimal
amount of viral concentration of 4
Figure 4:
Effects of adenovirus concentration on the
expression of
Since substantial cell death was found at the
optimal adenovirus concentration of CS-mediated gene transfer, it was
of importance to investigate whether there is a causal relationship
between adenovirus-induced cytotoxicity and the reporter gene
expression. To this end, we carried out the following kinetics study.
Primary hepatocytes in cultures were transfected with CS conjugate-DNA
complex in the presence of adenovirus. At different time intervals, the
cultures were terminated and expression of reporter gene and cell
viability were determined. As shown in Fig. 5, expression of
Figure 5:
Kinetics of
Figure 6:
Expression of
To confirm the importance of region
II+ in CS-mediated gene delivery, a peptide containing region
II+ sequence of CS protein (E35) was used as a competitor for the
CS-mediated gene transfer. Different amounts of E35 peptide, ranging
from 0.031 to 0.248 µM, were added to the transfection
mixture. The efficiencies of transfection were analyzed in comparison
with those using an unrelated peptide (A128) with different sequence
but the same length as competitor. In the presence of 0.125 µM E35 peptide, greater than 50% of
Figure 7:
Influence of region II CS peptide on the
expression of
Figure 8:
Expression of
The inability of CS
protein-mediated gene transfer into HL-60 cells could be due to the
absence of either adenovirus receptor or CS receptor, or both. To
address the adenovirus receptor issue, we transfected HL-60 cells with
recombinant To address the issue of CS receptors in HL-60
cells, we determined the rates of
[
Figure 9:
Uptake of CS27IVC-His
Figure 10:
Uptake of DNA complex by primary
hepatocytes (A) and HL-60 cells (B).
CS27IVC-His
Previous study has demonstrated
that CS protein binds to heparan sulfate proteoglycans associated with
the surface membrane of hepatocytes, and the binding can be abolished
by heparinase treatment (26, 27) . These observations
suggested that the CS receptor(s) may be heparan sulfate-related
proteoglycans. In an attempt to determine whether the CS-mediated gene
transfer into primary hepatocytes could be blocked by heparin and
dextran sulfate (by competing CS-conjugate binding to receptors),
however, we found that these chemicals cause dissociation of DNA from
the CS conjugate, as judged by gel-retardation assay (not shown).
CS protein has been known to play an important role in the
invasion of liver by malaria parasite(7, 8) . It is
known that this hepatic invasion is nonpathogenic and malaria symptom
manifests during erythrocyte cycle(8) . Much attention to this
protein in the malaria field is in its potential use as a target for
intervention of malaria infection, particularly in the vaccine
development(28) . The specificity of CS protein for hepatocytes
described in those studies (7, 29) raised a distinct
possibility that CS protein may be used as an effective vehicle for
hepatic gene targeting. For this reason, we initiated the present study
using primary hepatocyte cultures. We believe that this in vitro study is a prerequisite for future in vivo experimentation. Using this in vitro system, we have made
several important observations. First, we demonstrated that CS
protein can be utilized as a carrier to deliver recombinant DNA into
primary hepatocytes. The expression of Second, we
demonstrated that CS-mediated gene transfer into primary hepatocytes
require the presence of region II+ but not the NANP repeats. These
observations have important implications in the use of CS protein as a
liver-specific gene targeting carrier. These repeats are immunodominant
and confer the major antigenicity of CS protein. Human sera from P.
falciparum malaria parasite-infected individuals in endemic areas
contained mainly antibodies against the CS
repeats(7, 8) . On the other hand, antibody against
region II+ has not been detected in the human sera of
malaria-infected individuals. Moreover, antibodies to region II have
been difficult to elicit by immunization of mice and
rabbits(31) . The fact that deleting the immunodominant domain
did not compromise the ability of gene transfer (Table 2)
suggests that repeat-free CS protein can be used as a ligand for gene
delivery with reduced immunogenic complications to the hosts.
Region
II+ is the epitope for receptor binding(7) . It contains
evolutionarily conserved amino acid residues with homology to cell
adhesive motif found in several other proteins, including
thrombospondin(32) , properdin(33) , and a blood-stage
antigen of P. falciparum(33) . The observation that
region II+ peptide can block the CS-mediated gene transfer (Fig. 7) raises the possibility that the synthetic peptide
containing region II+ residues may be used as a ligand for gene
targeting via CS receptor. Experiments testing this possibility are
under way. Third, we showed that in addition to primary hepatocytes,
CS-protein could elicit gene transfer into many established cell lines,
including HeLa, NIH3T3, CHO, HepG2, and K562 cells (Fig. 8). We
believe that this is not due to alteration of CS protein configuration
by polylysine conjugation that resulted in changes in ligand
specificity for the following reasons. (i) The transfection is
ligand-dependent; very low transfection efficiency was seen using BSA
conjugates or polylysine as carriers (Fig. 5). (ii) The
transfection efficiencies in different cell types were correlated with
the levels of cellular receptor(s); very low transfection efficiency
was found in HL-60 cells in which receptor levels were low ((25) , and Fig. 9and Fig. 10). (iii) The
transfection efficiencies were region II+-dependent, consistent
with the binding specificity of CS protein to its receptor(7) .
These observations suggest a broad applicability of CS protein as gene
delivery vehicle for cultured cells. In this regard, CS protein may be
utilized for gene transfer into patient-derived cell cultures in the ex vivo gene therapy. The observation that CS protein could
effectively deliver recombinant DNA into primary hepatocytes described
in this report suggests that it can be a potential carrier for hepatic
gene targeting. Although the present study showed that the efficient
delivery requires adenovirus with the titer as high as 10 Additional concern regarding hepatic DNA delivery systems is the
structural requirements of the delivery vehicles. In the in vivo setting, the size of the DNA complex must be <200 nm to pass
the fenestrations in the liver parenchyma. Although the exact size of
CS carrier-DNA complexes used in the present study was not determined,
previous study using ASOR-polylysine-DNA complex prepared under the
similar procedure gave molecular sizes of 80-100 nm(15) .
This size range may be acceptable for in vivo trails. Perales et al.(6) have recently reported that, by changing
salt conditions, it is possible to modulate the sizes of galactosylated
poly(L-lysine)-DNA complex (to about 10 nm). The formulation
of such small DNA delivery system was found to correlate with the
prolonged expression of transgene in the livers of intact animals. Most of our current knowledge about receptor-mediated gene targeting
to livers has been derived from the studies of ASOR receptors. Using
this delivery system, Wu and associates (44) have demonstrated
a partial correction of genetic analbuminemia in Nagase rats by
intravenous injection of complexed DNA containing the human serum
albumin gene under the control of the mouse albumin enhancer/promoter.
In another study, target expression of low density lipoprotein receptor
gene into the receptor-deficient rabbits was demonstrated(45) .
The levels of expression of the transgenes in these animals remained
relatively low. It is conceivable that the efficacy of ASOR-mediated
hepatic gene transfer is influenced by the physiology of ASOR and its
receptors(46, 47) . The synthesis and bioavailability
of ASOR receptors and their circulating ligands can drastically affect
the efficacy of gene delivery. The use of CS protein, which is not
normally present in the host, may minimize possible interference by the
endogenous ligand (if there is any), thereby enhancing the targeting
efficacy. The CS protein carrier may be particularly useful in
individuals with hepatitis(48) , liver cirrhosis(49) ,
hepatocellular carcinomas(49, 50) , and diabetes
mellitus(51) . These individuals suffer from accumulation of
ASOR due to down-regulation of ASOR receptor synthesis. The
development of CS conjugate as DNA carrier, in complement to the
existing ASOR carrier system, should allow one to test the synergism in
the targeting specificity and enhancement of expression of the dual
delivery systems. The availability of additional hepatic DNA targeting
system would allow one to distribute the ``payloads'' into
different vehicles and thus facilitate the efficacy of targeting. From
the economic point of view, the production of CS protein carriers
should be considerably less expensive than the ASOR carrier. The gene
encoding the CS protein has been cloned and bacterially expressed
recombinant proteins are available. In contrast, because of the low
abundance of ASOR in blood, a large volume of blood is required to
extract sufficient quantities of the ligand. This not only increases
the cost of production, it also creates the potential risk of viral
contamination in the ligand preparations, as in the case of many other
blood-derived products. The use of CS conjugates should impose minimal
risk to patients and to the general population. In conclusion, the
present study established that the CS protein can be an effective
system for gene transfer into primary hepatocyte cultures as well as
into many different cultured cell lines. These findings provide a basis
for further in vivo studies of hepatic gene delivery. These
experiments are currently under way.
Volume 270,
Number 8,
Issue of February 24, 1995 pp. 3667-3676
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
, HeLa, NIH3T3, and K562, but not
HL-60, which contains low levels of receptor. Thus, the CS conjugate
can be used to deliver DNA into many different cultured cells. Most
importantly, the CS conjugate has a potential to be further developed
into a liver-specific gene delivery vehicle in vivo.
)(3) and
transferrin(4) . Recently, a synthetic neoglycoprotein that
recognizes the same receptor as ASOR has been used as a gene delivery
vehicle(5, 6) . In vivo, the distributions of
ASOR and transferrin receptors are very different. Transferrin
receptors are found in many different cell types, while ASOR receptors
are almost exclusively distributed on the sinusoidal domain of the
hepatocytes(3) . Because of its unique distribution, much
attention has been focused on the development of ASOR as a hepatic gene
delivery vehicle; however, the efficiency of ASOR-mediated hepatic gene
delivery system has not been perfected.
Recombinant CS Proteins and Peptides
Bacterially
derived CS recombinant proteins CS27IVC-His
(27-123(NANPNVDP)
(NANP)
300-411),
CS27IC-His(27-123(NANPNVDP)(NANP)300-411)
containing P. falciparum CS sequences from the T4 isolate were
prepared according to the procedure previously
described(10, 11) . The recombinant plasmids were
constructed by inserting the corresponding encoded DNA sequences into
PDS56/RBSII vector. A synthetic oligonucleotide encoding hexahistidine
was ligated to the 3` end of the CS genes to facilitate the recombinant
protein purification by metal chelate affinity chromatography. The
constructs were transformed into Escherichia coli strain M15.
Recombinant protein DHFR-CSF1-His ((DHFR)6-146),
which contains a mouse dihydrofolate reductase (DHFR) cDNA linked to
amino acid residues 6-146 of the CS sequence and the engineered
hexahistidine, has been described previously(11) . Peptide E35
(EWSPCSVTCGNGIZVRIKPGSAN) and A128 (GNEIEPGNNAYGSQSDTDASELT), both
having 23 amino acid residues, were synthesized on a Vega Coupler 250C
synthesizer using tert-butocycarbonyl chemistry and
deprotected as described previously(12) . The peptides were
purified by high performance liquid chromatography. The amino acid
composition of the synthesized peptides were confirmed by amino acid
analysis.Cells and Cell Culturing
Hepatocytes were isolated
from adult male C57BL/6 mice by the procedure previously described (13) with slight modification. Briefly, the livers were
perfused with a solution containing collagenase (100 units/ml), trypsin
inhibitor, 120 µg/ml (both from Worthington) and Hank's A
balanced salts by cannulating the vena cava and releasing the perfusate
via the portal vein. The resulting crude hepatocytes were suspended in
36% Percoll and further purified by centrifugation at 20,000 
g for 20 min(14) . The purified hepatocytes usually
had >95% viability at the time of plating as determined by trypan
blue exclusion. The cells were suspended in Waymouth's MD 705/1
medium (Life Technologies, Inc.) containing 10% fetal calf serum (FCS)
(Life Technologies, Inc.), penicillin (100 units/ml) and streptomycin
(100 mg/ml) and plated onto Costar six-well tissue culture plate 3506
at 3 10
cells/well. in air.Preparation of CS Conjugate
CS protein was
conjugated to poly(L-lysine) (molecular mass = 26.3
kDa, Sigma) by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC,
Sigma) according to the protocol described by Cristiano et
al.(15) . Standard reaction mixture (1.5 ml) contained 1
ml of CS protein (in 100 mM MOPS buffer, pH 7.3) and 0.5 ml of
poly(L-lysine) also in 100 mM MOPS, pH 7.3. EDC
powder was then directly added to the mixture so that the final
concentrations of CS protein, poly(L-lysine), and EDC were
94.3 µM, 141.05 µM, and 90.90 mM,
respectively. This molar ratio is comparable to those used by other
investigators in the preparation of ASOR
conjugates(3, 15) . FITC-polylysine (3.72 mg) was
dissolved in 100 mM MOPS, pH 7.3, and EDC was added to the
sample to a final concentration of 90.90 mM to mimic the
conditions for the preparation of the CS conjugate. The reaction
mixture was incubated at 25 °C for 18 h. The mixture was applied to
a Superose 6 column (16 30 cm) pre-equilibrated in HBS (150
mM NaCl, 20 mM Hepes, pH 7.3). The samples were
eluted with the same buffer at the velocity of 0.17 ml/min and
collected in 1-ml fractions. The chromatography was monitored by UV
absorption at 280 nm. Protein concentrations in each fraction were
determined by Bio-Rad protein assay system according to vendor's
specifications, using immunoglobulin G (IgG) as reference. The amounts
of conjugates used throughout this study were based on this protein
determination unless otherwise indicated. Conjugates were analyzed by
acid-urea gel electrophoresis according to the procedure described by
Meckee et al.(16) . Transferrin, ASOR, and BSA (Sigma)
conjugates were prepared by the same procedure.Preparation of Adenovirus
Adenovirus dl312 was
propagated in 293 cell and purified twice on CsCl density gradient
according to the procedure described(17) . The purified virus
was dialyzed against HBS buffer (three times, 1 liter each). Viral DNA
was extracted from aliquots of viral preparation by 1 h of SDS (0.1%)
and proteinase K (100 µg/ml) digestion, followed by
phenol/chloroform extraction and ethanol precipitation. The yield of
DNA was spectrophotometrically determined at 260 nm absorbance, and
used for calculation of the number of viral particles(15) . The
viruses were then aliquoted, and stored in HBS containing 10% glycerol
at -80 °C until use. Recombinant adenoviruses harboring
-galactosidase linked to the cytomegalovirus promoter (18) were prepared by the same procedure.Formation of DNA Complex and Transfection
The
reporter construct, pCMV--gal(19) , which contains a
bacterial -galactosidase gene under the transcriptional control of
cytomegalovirus enhancer/promoter was used for transfection assay.
Plasmid DNA was prepared by 2 CsCl gradient
ultracentrifugation. Six µg of pCMV--gal DNA in 350 µl of
HBS was mixed with various amounts of CS protein conjugate in 150
µl of HBS. The mixture was incubated at 25 °C for 30 min.
Before transfection, culture medium was removed and replaced by 0.5 ml
of the medium containing 2% FCS. CS protein conjugate-DNA complex was
added to the cells, immediately followed by addition of an appropriate
amount of adenovirus. The cells were incubated at 37 °C for 2 h.
After addition of 2 ml of complete medium supplemented with 10% FCS,
cells were incubated for an additional 16 h.-Galactosidase Activity
Assay-Galactosidase activity was analyzed by the methods
described by McGregor et al.(20) . Briefly, The
medium was removed from transfected cells. The cells were washed three
times with phosphate-buffered saline. The expression of
-galactosidase were determined either by histochemical staining
(X-gal stains) or by measuring the enzymatic activity of cell extracts
using ONPG as substrate. The unit of ONPG assay was defined as
described by McGregor et al.(20) .Uptakes of CS27IVC-His
CS27IVC-His and Conjugate-DNA
Complex was tritium-labeled by sodium
cyanoboro[H]hydride according to the procedure
previously described(21) . Briefly, 1 mg of CS protein in 1 ml
of HBS was added to 10 mCi of sodium
cyanoboro[H]hydride in the presence of 12 mM formaldehyde. The reaction proceeded at 4 °C for 30 min.
Labeled CS protein was purified by Sephadex G-25 chromatography. The
specific activity of labeled protein was 68,000 cpm/µg counted in a
Beckman LS 3801 liquid scintillation counter.H]CS27IVC-His were added to the culture. The cells were incubated at 37 °C
in 5% C0 incubator. At indicated time intervals, the medium
was removed. The cells were washed three times with phosphate-buffered
saline and lysed with 0.5 ml of 0.5 N NaOH. The radioactivity
of cell lysate was counted in Scintiverse (Fisher) by a liquid
scintillation counter (Beckman LS3801).-gal plasmid DNA was nick-translated
with [
P]dCTP using nick translation kit (Life
Technologies, Inc.) according to the protocols provided by the vender.
The labeled DNA was purified by Sephadex G-25 chromatography. Ten ng of P-labeled DNA was mixed with 2.99 µg of unlabeled
pCMV--gal DNA. The DNA was then mixed with 3.7 µg of CS
conjugate in a total volume of 500 µl of HBS. The formation of DNA
complex was checked by gel retardation assay on a 0.8% agarose gel. The
DNA complex was then diluted to 1 ml with serum-free medium and added
to 5 10 primary hepatocytes or HL-60 cells. The
cells were incubated at 37 °C in 5% CO incubator. At
different time intervals, the medium was removed and the
radioactivities in the cultured cells were determined.
Preparation of CS-Poly(L-lysine)
Conjugate
When H-recombinant CS27IVC-His protein (molecular mass = 53 kDa, Fig. 1) was
passed through a Superose 6 column and the eluents were monitored by
radioactivity, two distinct peaks were resolved (Fig. 2A, dashed line). These two peaks may
correspond to oligomeric (but predominantly dimeric) and monomic CS
proteins, respectively, since it has been reported that CS protein can
oligomerized by cysteine formation(9) . Recombinant
CS27IVC-His protein was conjugated to
poly(L-lysine) (26.3 kDa) by EDC, and fractionated by the same
column. Two distinct peaks were also eluted (Fig. 2A, solid line). The front peak had an apparent molecular mass of
greater than 100 kDa, whereas the second peak, approximately 80 kDa,
using the eluting profiles of presumptive multimeric CS protein (Fig. 2A) and unconjugated transferrin (80 kDa) (Fig. 2C, broken line) as references. These
results suggested that the front peak contained CS conjugate with very
little contamination of free CS protein. When poly(L-lysine)
alone was treated with EDC and fractionated through the same column,
there was a shift in the eluting profile as compared with that of the
untreated sample (Fig. 2B), indicating the formation of
polylysine-polylysine conjugates. However, both the conjugated
polylysine and free polylysine were eluted behind the front peak of the
CS conjugate (compare panels A and B). These results
suggest that the front peak in the CS conjugate preparation should
contain only negligible amount of polylysine-polylysine conjugates.
Thus, these chromatographic conditions apparently are effective in the
preparation of CS conjugates without significant contamination of free
CS protein and self-conjugated polylysine. We consider this very
important, because these contamination would have an adverse effect on
the efficacy of receptor-mediated gene delivery.
is the fusion protein of mouse DHFR and CS
peptides. The 18 amino acid residues shown below RII+ are the
epitope recognized by receptor.
/polylysine
and transferrin/polylysine were conjugated by EDC. The reaction
mixtures were then passed through a Superose 6 column. Details are
described under ``Materials and Methods.'' A,
chromatographic eluting profiles of
[H]CS27IVC-His and
CS27IVC-His/polylysine conjugate. B, eluting
profiles of FITC-polylysine and FITC-polylysine conjugated by EDC. The
fluorescence intensity was determined by a fluorometer at excitation
490 nm and emission 530 nm. C, the elution profiles of
transferrin and transferrin/polylysine conjugate. D and F, acid-urea gel electrophoretic analyses of proteins eluted
from the CS/polylysine and transferrin/polylysine, respectively.
Unconjugated CS protein transferrin (Tf), polylysine (PLL), and a mixture of CS and PLL (CS/PLL)
were run as references. E and G, transfection
efficiencies of protein fractions corresponding to the eluting profiles
shown in D and F. Six µg of pCMV--gal
plasmid DNA were complexed with 5.8 µg of CS-PLL conjugate from
each fraction in D and 4.2 µg of transferrin-PLL from each
fraction in E. DNA complexes were used to transfected HepG2
cells in the presence of adenovirus (5 10 particles/cell). The results were calculated from duplicated
samples, and 800-1,000 cells in three random fields were
counted.
)conjugates.
Thus, we conclude that this simple gel filtration chromatographic
technique can be used to prepare functional molecular carriers for gene
delivery.Optimization of Transfection Conditions for Primary
Hepatocytes
To test whether the CS-polylysine conjugates could
deliver recombinant DNA into primary hepatocytes, we carried out a set
of pilot experiments. CS27IVC-His protein conjugates from
the front peak were complexed with recombinant pCMV--gal DNA. The
DNA complexes were used to transfect primary hepatocyte cultures.
Eighteen hours after transfection, cells were stained with X-gal. Less
than 0.1% of positive cells were found (data not shown). It is possible
that the DNA complexes were internalized but not able to escape from
endosomal entrapment. Previous studies of ASOR and transferrin carriers
demonstrated that adenovirus infection could destabilize endosomes,
presumably allowing endosomally contained plasmid DNA to
escape(15, 22, 23, 24) . Therefore,
we cotransfected hepatocytes with a replication defective adenovirus
(dl312). The application of adenovirus greatly enhanced the frequency
of X-gal-positive cells, indicating that the efficiency of CS-mediated
gene transfer also required endosomal destabilizing agents. Protein
conjugates from the fractions of second peak were similarly analyzed,
however, no more than 20% of blue cells were seen under various
``optimization'' conditions (see below). Thus, all the data
presented here were using CS27IVC-His conjugates collected
from the front peak of the Sepharose 6 column (for simplicity,
hereafter referred to as CS conjugate unless other specifications).
-galactosidase in primary
hepatocytes. Different amounts of CS conjugate, BSA conjugate, and
polylysine were complexed with 6 µg of pCMV-b-gal DNA as described
under ``Materials and Methods.'' Primary hepatocytes (3
10) were transfected with the complexes in the
presence of adenovirus (4 10
particles/cell). The
cells were stained with X-gal, and percentage of blue cells were
determined by visually counting 800-1000 cells from three random fields
for each duplicated samples.
10 particles/cell under the transfection conditions. Additional
viral particles resulted in decreased transfection efficiency, probably
due to excess cytotoxicity of viral infection. The viabilities of
transfected cells decreased as the viral concentrations increased.
Under the optimal transfection conditions, about 58% of cells were
viable (or 66% after correcting against the plating efficiency, which
was about 90%).
-galactosidase in primary hepatocytes.
CS27IVC-His conjugate (7.4 µg) was complexed with 6
µg of pCMV--gal DNA. Indicated amount of adenovirus was added
to the cells after the addition of the complex. Eighteen hours after
transfection, cell viability, and -galactosidase activity were
determined by trypan blue exclusion and X-gal stains, respectively.
Data were collected by counting 800-1000 cells from three random fields
for each triplicated sample.
-galactosidase appeared 6 h after transfection, reached its
maximum around 15 h, and decreased thereafter. On the other hand, cell
viability decreased as the transfection time increased in a linear
pattern. At the time when the reporter gene started to express (6 h
after transfection), 80% of cells were still viable, while at the
maximal levels of expression, 60%. Virtually no viable cells were
detected after 48 h of transfection. These results suggest that the use
of adenovirus as an enhancer for reporter gene expression in
receptor-mediated gene delivery caused important toxicity.
-galactosidase expression
and cell viability after transfection with DNA complex.
CS27IVC-His conjugate (7.4 µg) was complexed with 6
µg of pCMV--gal DNA. The transfection of primary hepatocytes
was initiated (time 0) by the addition of the DNA complexes and
adenovirus (4 10 particles/cell). At different time
intervals, percentage of blue cells (-galactosidase expression)
and cell viability were determined as described in the legend to Fig. 4. The plating efficiencies of hepatocytes were also
included. The data were not normalized against the plating
efficiencies.
Dependence of CS Protein-mediated Gene Delivery upon the
Evolutionarily Conserved Region II+ Epitope
To investigate
the functional domain in CS protein for mediating the DNA transfer, we
tested recombinant constructs containing deletions in various regions
of the CS protein (Fig. 1). Recombinant CS27IC-His,
which contains regions I and II+ but only one copy of the NANP
repeat (instead of the 21 copies as in CS27IVC-His), had
84% transfection efficiency in the primary hepatocyte cultures (Table 1). This transfection efficiency was consistent with that
using the CS27IVC-His conjugates (Table 1). However,
recombinant DHFR-CSF1 (Fig. 1), which lacks region II+ and
the entire NANP repeats, but retains only region I, showed poor gene
transfer ability (Table 1). Histochemical staining of the
transfected cells using these various conjugates is shown in Fig. 6. These results suggest that region II+ is important
for CS-mediated gene delivery.
-galactosidase in
primary hepatocytes after transfected with DNA complexes containing
different recombinant CS proteins. CS conjugates containing
CS27IVC-His (b), CS27IC-His (c), and DHFR-CSF1-His (d) were
complexed with 6 µg of pCMV--gal DNA. The amount of CS
conjugate used were experimentally determined in pre-experiments.
Transfection was carried out in the presence of adenovirus (4
10 particles/cell). -Galactosidase was determined by
X-gal staining. Six µg of pCMV--gal DNA was used as control (a).
-galactosidase activity was
inhibited and the activity was completely ablated when 0.248 µM of E35 was added. However, at the comparative concentrations,
peptide A128 showed no competitive effect (Fig. 7). These
results further supported the notion that region II+ of CS protein
contains functional residues for gene transfer.
-galactosidase in primary hepatocytes.
CS27IVC-His conjugate (7.4 µg) was conjugated to 6
µg of pCMV--gal DNA. Different amounts of E35 and A128
peptides were added to cells after the addition of DNA complexes and
adenovirus (4 10 particles/cell).
-Galactosidase activity was quantitated using ONPG as a substrate.
Results were from triplicate experiments.
Dependence of CS Protein-mediated Gene Delivery upon Its
Receptors
Although in vivo study has demonstrated that
CS protein almost exclusively binds to hepatocytes, using an cell
adhesion assay, Rich et al.(25) implicated that CS
receptors are present in a broad range of established cell lines,
particularly CEM, HSB-2, K562, and KG-1 lymphocytic cell lines.
Therefore, we investigated whether the efficiency of CS-mediated gene
transfer could be correlated with the presence of receptors on these
various cell lines. We chose HepG2 (7) and CHO (26, 27) cells, which have been reported to contain
high levels of CS receptor, and HL-60 cells, which contain low levels
of receptors(25) . pCMV--gal reporter gene was transfected
into these cells using conjugates prepared from CS27IVC-His protein. Under the optimal conditions, CHO and HepG2 cells
exhibited -galactosidase activities 130 times higher than that of
the HL-60 cells (Fig. 8).
-galactosidase in
different cell lines transfected with CS carrier. CS27IVC-His conjugate (7.4 µg) was complexed with 6 µg of
pCMV--gal DNA. The DNA complex was used to transfect HeLa, NIH3T3,
CHO, Hep G2, and HL-60. The viral concentration for each cell line was
optimized. -Galactosidase activity was quantified as described
under ``Materials and Methods'' using ONPG as a substrate.
Data were average of triplicated
experiments.
-galactosidase adenovirus (ranging from 5
10 to 1 10
particles/5 10 cells). We found that this recombinant adenovirus has very poor
transfection efficiency (<1%) in HL-60 cells, in comparison with
greater than 95% transfection efficiency in HepG2 cells (data not
shown). Although there may be other reasons for the inability of the
expression of reporter gene in the adenovirus-infected HL-60 cells, the
possibility of lacking adenovirus receptors in these cells could not be
formally ruled out.H]CS27IVC-His uptake in primary
hepatocytes and in HL-60 cells. Since it has been demonstrated that
multimeric CS proteins were more efficiently uptaken than monomeric
proteins by the liver parenchyma(29) , H-labeled
multimeric CS fractions (the first peak, Fig. 2A, dashed line) were used for the uptake study. As shown in Fig. 9, the levels of labeled CS protein uptake by HL-60 cells
were significantly lower than those in the primary hepatocytes,
suggesting that HL-60 cells containing reduced levels of CS receptors.
To substantiate these observations, we also measured the uptakes of DNA
mediated by CS protein conjugates in these two different cell types,
using transferrin-conjugates and BSA conjugates as positive and
negative controls, respectively. As shown in Fig. 10, the
transferrin-DNA complexes were efficiently uptaken by both primary
hepatocytes and HL-60 cells. However, the levels of CS-DNA uptake were
significantly lower in HL-60 cells than those in primary hepatocytes.
These results are consistent with the findings that the cell-type
specificities of CS-mediated gene transfer are correlated with the
presence of functional CS receptors.
by
primary hepatocytes and HL-60 cells.
[H]CS27IV-His (2.5 µg) was mixed
with 0.5 ml of serum-free medium and added to 5 10 cells. At different time intervals, the medium was removed. The
cells were washed three times with phosphate-buffered saline and lysed
with 0.5 ml of 0.5 N NaOH. The radioactivity was measured by a
liquid scintillation counter.
(3.7 µg), transferrin (2.9 µg), and BSA
(2.75 µg) conjugates were complexed with 3 µg of P-labeled DNA prepared by mixing 10 ng of nick-translated
DNA with 2.99 µg of cold DNA. The complexes were added to 5
10 cells. Data were from triplicate
samples.
Gene Delivery by CS Conjugates in Various Cultured
Cells
Since CS receptor has been found broadly distributed in
established cell lines(25) , CS conjugate may be used as a
conventional tool for DNA transfer in different cell lines. To explore
this possibility, we chose NIH3T3 (normal fibroblast cell line), HeLa
(tumor cell line), and K562 (suspension culture) for transfection using
CS conjugates. HeLa, NIH3T3, and K562 all showed high transfection
efficacies (Fig. 8). However, a number of human melanoma cell
lines (A-375, HT-144, and WM115), and freshly prepared human bone
marrow cells (gifts of Dr. Albert Deiseroth, M. D. Anderson Cancer
Center) showed very low levels of transfection efficiencies (data not
shown). Whether the poor transfection efficiencies in these cells were
due to the lack of adenovirus and/or CS receptors have not been
determined. Nonetheless, these results demonstrated the CS conjugate
can be used as a gene delivery carrier in some cultured cells.
-galactosidase reporter
gene required the presence of adenovirus. Adenovirus is believed to
function as an endosomal disrupting tool, facilitating the release of
internalized DNA after endocytosis. These results strongly suggest that
CS protein, like the other molecular carriers such as ASOR and
transferrin, undergoes a similar endosome-lysosome pathway after
internalization by endocytosis. Although it has been generally believed
that malaria infection is receptor-mediated, there is no direct
evidence regarding how the sporozoites invade hepatocytes.
Understanding this endocytic pathway of invasion may facilitate
research aiming at control of malaria infection.
cytotoxic particles/cell, the same efficacy of gene delivery may
be achieved by coupling of adenovirus to DNA
complex(22, 24, 34, 35) , or using
cripple adenovirus particles (36) or less toxic adenovirus (37) thereby reducing the requirement of viral particles for
endosomal lysis. Another concern of using adenovirus is the safety
issue and should be critically evaluated, although the recent study
showed that no major cytotoxic effects were found in the rat liver
after adenovirus-mediated in vivo gene transfer(38) .
However, a recent human trial of recombinant adenovirus containing the
human cystic fibrosis transmembrane conductance regulator gene into the
CF respiratory epithelium with dose up to 2 10 plaque-forming units showed toxic effects to one of the four
patients, including local and systemic inflammatory
syndrome(39) . Alternatively, viral fusogenic peptides can be
used for substitution of adenovirus, since influenza hemagglutinin HA-2
terminal peptides have been demonstrated to augment gene transfer
efficiency greater than 100-fold by the receptor-mediated gene transfer
into HeLa cells(40, 41) . In addition, recent studies
in protein trafficking and endosomal physiology have identified several
endogenous proteins associated with vesicle budding, membrane fusion (42) , and protein translocation(43) . Upon further
development, these proteins could be potential tools for enhancing the
expression of transgenes in the receptor-mediated gene delivery system.
)-D-galactoside;
X-gal, 5-bromo-4-chloro-3-indolyl -D-galactoside; FCS,
fetal calf serum; BSA, bovine serum albumin; CHO, Chinese hamster
ovary; FITC, fluorescein isothiocyanate; HBS, Hepes-buffered saline.)
We are grateful to Dr. T. C. Liang for valuable
comments on the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y.-C. Lin-Lee, S. Nakamura, V. Gandhi, S. A. Curley, D. Stuber, T. R. Burkot, and M. Tien Kuo Prolonged Stability and Sustained Prodrug Cell Killing Activity Using Receptor-mediated Delivery of Malarial Circumsporozoite-Cytosine Deaminase Fusion Protein into Liver Cancer Cells Mol. Cancer Ther., May 1, 2002; 1(7): 461 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Song, M. Ikeguchi, G. Zhou, and M. T. Kuo Identification and Characterization of a Hepatoma Cell-specific Enhancer in the Mouse Multidrug Resistance mdr1b Promoter J. Biol. Chem., October 27, 1995; 270(43): 25468 - 25474. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |