| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Communications |
Selective Genetics Inc., San Diego, California 92121, USA; and
a Dyax Corporation, Cambridge, Massachusetts 02115, USA
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: phage display • transfection • receptor-mediated • ligand
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alternative ligands could greatly improve existing vectors for therapeutic gene delivery by targeting specific cells, thus reducing toxicity and allowing vectors to be administered systemically (11) . In an effort to increase vector selectivity and potency, growth factors and other cell-specific ligands that bind cell surface receptors have been used for targeting both nonviral and viral vectors (12) . For example, we used fibroblast growth factor (FGF2)1 to target DNA condensates (13) and adenovirus 14, 15) to cells that overexpress FGF receptors, such as occurs in certain tumor cells and after injury. Targeting can even be applied to a single-stranded DNA bacterial virus, which normally has no tropism for mammalian cells. Recently, we showed that a modified filamentous bacteriophage can be specifically targeted to FGF receptor-bearing cells with noncovalently attached FGF2, resulting in successful foreign transgene expression (16) . Based on these observations, we reasoned that if targeted phage transduction could also be accomplished by genetically displaying a targeting ligand on the phage surface, it would then be feasible to identify novel ligands by direct selection of those that mediate gene transfer.
In the present study, we transduced COS-1 cells with filamentous phage displaying a known gene targeting ligand, FGF2. We constructed FGF2 display phage by fusing the FGF2 gene to the pIII gene encoding the minor coat protein. The phage were modified for transduction of mammalian cells by inserting a green fluorescent protein (GFP) reporter gene transcriptionally driven by the cytomegalovirus immediate early promoter (CMV) into the phage genome. Specificity of the FGF2-phage-mediated transduction was demonstrated by inhibition of transduction in the presence of excess free FGF2. These data support our hypothesis that phage transduction can be used to screen display libraries for novel ligands capable of gene transfer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Modification of phage for mammalian cell reporter gene
expression
A mammalian cell expression cassette consisting of the SV40
origin of replication (SVori), the CMV immediate early promoter, a
multiple cloning site, and the bovine growth hormone (BGH)
polyadenylation and transcriptional termination signals was inserted
into an ampicillin-resistant M13 phage vector, MANP, to create the MCMV
vector. These elements were assembled using splice overlap
extension-polymerase chain reaction (SOE-PCR) (18)
. Briefly, the primer
pairs [1, 2], [3, 4], and [5, 6] (Table 1
) were used to amplify SVori and CMV DNA fragments from pEGFP-N1
(Clontech, Palo Alto, Calif.), and the primer pair [7, 8] (Table 1)
was used to amplify the BGH DNA terminator fragment from pcDNA3.1
(Invitrogen, Carlsbad, Calif.). DNA fragments were gel purified,
combined, and used as template for amplification using flanking primers
1 and 8. The resulting 1 kb product was subcloned into the unique
PvuII site of the MANP vector. The EGFP gene is human codon
optimized and has a mutation, described by Cormack et al. (19)
, that
results in a 35-fold increase in fluorescence intensity over wild-type
GFP. The EGFP gene was amplified from pEGFP-N1 (Clontech) by PCR with
primers 9 and 10 (Table 1)
, subcloned into SmaI digested
pBS+ (Stratagene, San Diego, Calif.). After sequencing, the EGFP gene
was removed with BssHII and XbaI and inserted
into the multiple cloning site of the BssHII/XbaI
digested MCMV phage vector to create MG3. This vector encodes a short
peptide extension on the amino terminus of pIII (CGPSPPVRWC).
|
Construction of FGF2 display phage
To avoid potential folding and aggregation problems that might
occur as a result of expressing the FGF2 (C96S) mutant with an unpaired
cysteine (C78), we created DNA encoding a 155 amino acid double mutant
of mature FGF2 (C78S,C96S) and inserted it in frame with the pIII gene
of M13, using the NcoI and PstI sites in the MG3
vector (Fig. 1
) such that the insert is downstream from the pIII signal peptide
encoding sequence. The DNA sequence between the NcoI and
PstI sites encoding the peptide (CGPSPPVRWC) was replaced
with the FGF2 gene. The double mutant was created by site-directed
mutagenesis of a FGF2 C96S mutant (17)
. SOE-PCR was used to amplify DNA
from the FGF2 template using primer pairs [11, 12], and [13, 14]
(Table 1)
. Amplified DNA fragments were purified by gel electrophoresis
and used as template for amplification with primers 11 and 14 (Table 1)
. With this strategy, a cysteine to serine mutation was made at amino
acid 78 of FGF2, the termination codon was removed, a
gly-gly-gly-gly-ser linker was added, and NcoI and
PstI sites were introduced for subcloning. Sequencing of the
FGF2-phage vector indicated that a single base pair error was
introduced during assembly, which resulted, however, in a silent
mutation.
|
Purification of phage particles and determination of titer
Phage were purified by standard polyethylene glycol (PEG)
precipitation (1)
. Briefly, ~20 ampicillin-resistant colonies of
phage infected Top10F' host bacteria (Invitrogen) were seeded into one
liter of LB media containing 100 µg/ml ampicillin and grown for
16 h at 37°C with shaking. Bacteria were removed by
centrifugation (20 min at 11,800 g) and supernatant was
transferred to fresh bottles. AEBSF (4-(2-aminoethyl)-benzene sulfonyl
fluoride) was added to a final concentration of 0.2 mM. Phage particles
were precipitated by addition of 1/6th volume of 30% PEG 8000 in 1.5M
NaCl, incubation at 4°C for 2 h, and centrifugation (30 min at
11,800 g). The pellet was resuspended in 30 ml of PBS
containing 0.2 mM AEBSF at 4°C for 30 min with occasional vortexing.
Residual bacteria were removed by centrifugation (20 min at 26,000
g) and supernatant transferred to fresh tubes. Phage
particles were again precipitated by addition of 1/6th volume of 30%
PEG 8000 in 1.5M NaCl, incubation at 4°C for 30 min, and
centrifugation (30 min at 26,000 g). The final phage pellet
was resuspended in 10 ml PBS containing 0.2 mM AEBSF, titered, and
stored at 4°C.
For CsCl purified phage, 4 g of CsCl were added to 2 ml of phage in PBS and 6 ml TE (10 mM Tris pH 8.0, 1 mM EDTA). After 16 h of centrifugation at 211,000 g at 15°C, the upper translucent band was removed from the top with a 18 g needle and 1 ml syringe. Phage were dialyzed against PBS overnight at 4°C and titered in XL-1Blue MRF' host cells (Stratagene) using standard protocols (1) . The smaller plaques generated by the FGF2 display phage were counted to determine titer for mammalian cell transduction assays. Some preparations contained large revertant plaques that resulted from phage containing deletions of all or part of the FGF2 gene. Only preparations containing >95% small plaques were used.
Analysis of FGF2 expression on display phage
CsCl purified phage were boiled in 1x Tris-glycine-sodium
dodecyl sulfate (SDS) sample buffer (Novex, San Diego, Calif.)
containing 5% 2-mercaptoethanol for 3 min prior to loading on 8%
polyacrylamide Tris-glycine gels (Novex) for immunoblot analysis. After
separation, proteins were transferred to pure nitrocellulose filters
(0.2 µM pore) (Bio-Rad, Richmond, Calif.) in 96 mM glycine, 12 mM
Tris Base, 0.01% SDS and 20% methanol. The filter was blocked in 1%
bovine serum albumin (BSA) in PBST (phosphate-buffered saline with
0.05% Tween 20) for 1 h at RT and probed with mouse anti-FGF2
antibody (Transduction Laboratories, Lexington, Ky.) at a 1:500
dilution, or mouse anti-pIII antibody (MoBiTec, Gottingen, Germany) at
a 1:1000 dilution, in 1% BSA in PBST for 1 h at room temperature.
After five 5 min washes with PBST, filters were incubated with
biotinylated donkey anti-mouse antibody (Jackson Immunoresearch, West
Grove, Pa.) at a 1:5000 dilution in 1% BSA in PBST for 1 h. After
five 5 min washes with PBST, filters were incubated with a horseradish
peroxidase (HRP) streptavidin conjugate (Amersham Life Sciences,
Arlington Heights, Ill.) at a 1:20,000 dilution in 1% BSA in PBST for
15 min. Filters were washed five times for 5 min with PBST and
developed with chemiluminescent substrates, SuperSignal or SuperSignal
ULTRA (Pierce Chemical, Rockford, Ill.), for 1 min, then exposed to
film.
A sandwich enzyme-linked immunoassay (ELISA) was used to detect immunologically reactive FGF2 on recombinant phage. Rabbit anti-fd phage antisera and mouse anti-FGF2 antibody (#B1786 and #F3393, respectively; Sigma, St. Louis, Mo.) in 0.1 M carbonate/bicarbonate buffer (pH 9.6) were adsorbed onto 96-well plates overnight. Wells were washed three times with PBST, blocked with Superblock (Pierce Chemical) for 3 h, and washed three times with PBST prior to addition of phage dilutions (50 µl/well). Phage were incubated overnight at 4°C, washed eight times with PBST, and incubated with HRP-conjugated anti-M13 antibody (Pharmacia Biotech, Piscataway, N.J.) at a 1:5000 dilution for 1 h. After five washings with PBST, signal was developed with 50 µl/well o-phenylenediamine dihydrochloride substrate (Sigma) for ~5 min. All incubations were done at room temperature. Reactions were stopped with 20 µl 3M HCl, optical density (OD) was read at 490 nm, and data were analyzed with SoftMax Pro software (Molecular Devices, Sunnyvale, Calif.).
Phage transfection and detection of reporter gene
The ability of FGF2 displaying phage particles to
transduce cells was evaluated using the COS-1 cell line (ATCC,
Rockville, Md.) that was grown in Dulbecco's modified Eagle's medium
(DMEM) containing 10% FCS, 2 mM L-glutamine, pyruvate, and
nonessential amino acids. The COS-1 cell line is a derivative of CV-1
monkey kidney cells and carries a single copy of the SV40 T-antigen,
which permits high copy replication of DNAs that contain an SV40 origin
of replication. Cells were transferred into six-well plates at
densities of 40,000 cells 24 h prior to phage particle addition.
The phage were added at 1011 pfu/ml and incubated
with the cells (~75,000/well) for 4 h at 37°C in DMEM
containing 2% BSA (bovine serum albumin) as a blocking agent. In some
instances, free FGF2 was added to a final concentration of 2 µg/ml or
blocking phage were added at a 100-fold excess at the time of phage
addition. The blocking phage particles (ß-gal phage) do not display
FGF2 or contain a GFP gene. They were prepared from pRC-CMVß
phagemid-transformed host bacteria by rescue with M13 helper phage as
described previously (16)
. After washing to remove unbound phage, the
cells were returned to the incubator for an additional 72 h.
Nongenetic phage targeting was performed by assembling a phage-FGF2 receptor complex at the cell surface as described previously (16) . Briefly, the cells were washed twice with ice-cold PBS/FBS (PBS with 2% fetal bovine serum), followed by the sequential addition of biotinylated FGF2, neutravidin (Pierce Chemical), and biotinylated anti-M13 antibody on ice for 15 min each, with a cell wash in between each step. Fifteen minutes after adding biotinylated anti-M13 antibody, the cells were washed and phage was added in PBS/FBS to form a complex at the cell surface. After 1 h on ice, the cells were washed, put into fresh culture media and returned to the incubator at 37°C to permit internalization. GFP expression was measured 72 h later.
Transduction was measured by counting all of the GFP-positive autofluorescent cells in each well. Transfections were done in triplicate and performed at least twice. The cells were washed twice in PBS/FBS and 1 ml of Dulbecco's PBS was added to each well. Fluorescent cells were detected directly in the culture plates using an epifluorescent inverted microscope (Nikon Diaphot) equipped with a fluorescein isothiocyanate filter set.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
This GFP expression cassette included an origin of replication from
SV40 to ensure high copy number replication in SV40
T-antigen-containing cells. COS-1 cells were transfected with
double-stranded phage DNA by CaPO4
coprecipitation to confirm that the GFP expression cassette was
functional (not shown).
Display of FGF2 on MG3-phage
Although our strategy for expressing FGF2 on the surface of
filamentous phage was similar to one used to display IL-3 (5)
, it was
unclear how the inclusion of the CMV-GFP expression cassette would
affect display. DNA encoding FGF2 was fused 5' to DNA encoding the
full-length minor coat protein pIII, so that three to five copies of
targeting ligand fused at the amino terminus of the coat protein were
expected to be displayed per phage (Fig. 1B
). Because
earlier attempts to display FGF2 (C96S) were less successful, we used
an FGF2 gene derivative that contained two mutations, resulting in the
conversion of the two free surface cysteines to serines to reduce the
probability of misfolding and aggregation. The (C96S,C78S) mutant of
FGF2 has previously been produced in bacteria and shown to have
equivalent biological activity as recombinant wild-type FGF2 (20)
.
Recombinant phage were identified by the smaller plaque size, probably
caused by the foreign pIII extension reducing phage productivity.
The FGF2-pIII fusion was detected on phage particles by immunoblot
analysis of FGF2-phage. An immunoblotted fusion protein of ~85 kDa
was detected in the lane loaded with an extract of
1010 FGF2-phage but not in the lane containing an
equivalent amount of parent phage (lanes a and b, Fig. 2
). This immunoreactive protein migrates at the predicted combined
molecular mass of FGF2 (~18kDa) and pIII (migrates anomalously at
65–70 kDa). When an identical blot was prepared in parallel and probed
with an anti-pIII antibody, a protein band corresponding to native pIII
was seen in both FGF2-phage and in the parent MG3-phage extracts (lanes
c and d, Fig. 2
). An additional higher molecular mass but less abundant
protein migrating at the expected molecular mass of the FGF2-pIII
fusion protein was detected only in the extracts from FGF2-phage (lane
c). The smaller immunoreactive pIII protein in the FGF2-phage lane was
not detected by anti-FGF2 antibody and migrated slightly faster than
the MG3-phage encoded pIII, indicating that it was probably the result
of cleavage within the amino terminus of pIII. These data suggest that
both truncated pIII and FGF2-pIII fusion protein were present on the
FGF2-phage particles. We estimated from a dilution series on additional
immunoblots (not shown) that the abundance of the fusion coat protein
was ~1/50th of the free pIII protein. If 3–5 pIII molecules are
present on each phage, then 6–10% of phage would display FGF2.
|
We also assessed the presence of FGF2 on the phage coat by sandwich
ELISA. There was a significant increase in FGF2-phage retained on the
ELISA plate by FGF2 antibody compared with the MG3 parent phage
(Fig. 3
A). There was little difference in the amount of either phage
retained on the ELISA plate by anti-M13 antibody (Fig. 3B
).
|
Transduction of COS-1 cells with FGF2-phage
FGF2 display phage and MG3 control phage were compared for their
ability to transduce COS-1 cells. Transfection with the FGF2-phage
resulted in about a 10-fold greater transduction efficiency than the
control phage (Fig. 4
A). These findings indicate that the FGF2 ligand on the
surface of the phage particles results in FGF2-dependent binding and
internalization of phage because expression of the phage DNA encoded
GFP requires nuclear transport, transcription, and translation in the
target cell. The specificity of the FGF2-phage-mediated transduction
for the FGF receptor complex was demonstrated by successful inhibition
of transduction by these phage in the presence of an excess of free
FGF2 (Fig. 4A
). Competition with FGF2 had no effect on
nonspecific transduction by MG3 control phage.
|
We further demonstrated the specificity of the targeting pathway of
FGF2-phage by adding an excess of phage particles with no GFP gene and
no targeting ligand to the incubation media. These `blocking phage'
were expected to interfere with any nonspecific uptake of GFP
containing phage, thus reducing the transduction efficiency of
untargeted phage. As anticipated, transduction of cells by MG3 control
phage was significantly inhibited by a 100-fold excess of blocking
phage. Moreover, the presence of excess blocking phage did not inhibit
FGF2 targeting and, if anything, slightly increased FGF2-phage
transduction (Fig. 4B
).
It was also important to rule out the possibility that the increased
transduction efficiency of FGF2-phage was due to structural differences
in the phage particles themselves or, alternatively, to differences in
the amount of transduction competent particles in the phage
preparation. To investigate this possibility, we compared the
transduction ability of FGF2-phage and MG3 control phage by
nongenetically targeting both phages. Equivalent titers of each phage
were used to transfect COS-1 cells using avidin-biotin FGF2 targeting
(16)
, and there was no significant difference in transduction between
FGF2-phage and control phage (Fig. 4C
). Thus, both
FGF2-phage and MG3 control phage preparations have the same capacity to
transduce COS-1 cells, once appropriately targeted.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
phage 24, 25)
, but there was no attempt to target
the phage particles in these experiments. Recently, filamentous phage
have been targeted to the cell surface and shown to be internalized
with RGD peptides (26)
or selected random peptides (9)
. Mammalian cell
transfection has been reported with an RGD tail tube modified phage
(27)
, however, this was achieved by growing cells directly on RGD-phage
that had been adsorbed to the culture dish. We show here that
filamentous phage containing a CMV-driven GFP gene and genetically
displaying the growth factor ligand, FGF2, can transduce COS-1 cells in
an FGF2-specific manner. To our knowledge, this is the first report to
demonstrate mammalian cell gene transfer by genetically targeted
filamentous phage. These data are consistent with and extend our
previous results demonstrating that filamentous phage that normally
have no tropism for mammalian cells can be adapted for cell-specific
mammalian cell transduction when appropriately targeted. Consistent with previous reports describing the functional display of various small peptide cytokines and growth factors as pIII fusion proteins on the surface of M13 phage 2-8) , we found that FGF2 can be displayed on the phage surface as a pIII fusion. Early attempts to express FGF2 fused to the pIII or pVIII coat protein in other vector systems designed to include expression of both fusion protein and native pIII coat protein (1) resulted in only native pIII and no detectable FGF2 fusion protein by immunoblot analysis (data not shown). We therefore attempted expression of an FGF2-pIII fusion onto the phage coat by expressing it in the phage vector described here, which contains only the FGF2-fused copy of the pIII gene. Even in this system, the FGF2-pIII fusion represented only ~2–3% of the total pIII protein present on phage particles, indicating low but detectable expression of FGF2-pIII. Together, these data suggest that there is strong selective pressure against bacterial expression of the FGF2 fusion in the periplasm where phage assembly occurs, thus resulting in more of the free pIII protein on the phage particle. The free pIII presumably is the result of proteolytic cleavage of the FGF2 fusion protein.
The significant albeit low transduction efficiency obtained with FGF2-phage could be explained by the expression of a single FGF2 on 6–10% of phage particles, which is similar to what is obtained for other ligands with vectors designed for monovalent phage display (1) . Since heparin-induced FGF2 dimerization is thought to activate FGF2 (28) , this process, and therefore biological activity, might be increased by multivalent display. Sequences in the phage vector itself could also contribute to the lower transduction efficiency. Noncovalent FGF2 targeting in the present study resulted in about a 12-fold lower transduction efficiency than expected from our previous results with a smaller phagemid vector system (16) . Therefore, we expect that more efficient ligand expression on the phage surface and further modifications of the vector will improve transduction efficiency. This increased efficiency could potentially exceed that which we have previously reported for nongenetic targeting [~1%, (16) ].
Our ELISA data showed binding of a neutralizing anti-FGF2 antibody to the displayed FGF2, suggesting that the FGF2 on the phage surface is properly folded and therefore active. This notion is further supported by the observation that significant transduction of COS-1 cells with FGF2-phage was obtained relative to the parent phage, MG3. Moreover, the FGF2-phage-mediated transduction was specific for FGF receptors, as shown by the substantial inhibition of transduction in the presence of excess free FGF2. The low level of nonspecific transduction by the control MG3 phage was not affected by the presence of excess FGF2, suggesting that simple adhesion to the cell surface and subsequent internalization were not mediated by FGF receptors. However, the presence of a 100-fold excess of nontransducing phage particles significantly inhibited this nonspecific transduction but not the specific FGF2-phage transduction, providing further evidence of specificity. These data indicate that FGF2 display phage can transduce mammalian cells, that the transduction is specifically mediated by the displayed ligand, and that FGF2-phage were internalized by a distinct pathway from the untargeted phage.
Our results extend the proposal that peptide display phage might themselves be used as gene delivery vehicles 9, 16) . Indeed, in many ways, bacteriophage fit the description of the ideal gene therapy vector (11) . Though immunogenic, like animal viral vectors, phage might be safer because they are much less likely to generate a replication-competent virus in animal cells. Phage would also be less likely to infect nontargeted tissues because of their natural lack of tropism for eukaryotic cells. Thus, the targeted phage particles resemble synthetic DNA conjugates that can be biologically produced. The results described here suggest that development of an improved gene delivery vehicle could be achieved by further modification and selection of genetically altered phage, resulting in safe and efficient gene transfer into mammalian target cells.
Having demonstrated transduction by phage genetically displaying a known gene targeting ligand, it should be possible to apply selective phage transduction and the power of combinatorial phage libraries to the discovery of new gene targeting ligands. Current phage display methods for identifying cell- or organ-specific targeting ligands select only for phage binding and/or internalization 9, 10) and require recovery of infective phage. Noninfective phage that are, for example, uncoated and trafficked to the nucleus for expression, would be missed by existing screening methods. Therefore, selection of ligand targeted phage using phage internalization and subsequent expression of a selectable reporter gene could directly identify novel ligands capable of both cell-specific internalization and trafficking required for gene delivery. We anticipate that the discovery of such ligands will greatly improve the efficacy of existing gene therapy vectors.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Abbreviations: AEBSF, 4-(2-aminoethyl)-benzene sulfonyl
fluoride; BGH, bovine growth hormone; BSA, bovine serum albumin; DMEM,
Dulbecco's modified Eagle's medium; ELISA, enzyme-linked immunoassay;
FBS, fetal bovine serum; FGF2, fibroblast growth factor; GFP, green
fluorescent protein; HRP, horseradish peroxidase; OD, optical density;
PBST, phosphate-buffered saline with 0.05% Tween 20; PEG, polyethylene
glycol; SDS, sodium dodecyl sulfate; SOE-PCR, splice overlap
extension-polymerase chain reaction.
Received for publication November 9, 1998.
Revision received February 5, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
plac gene expression in GMI-gangliosidosis fibroblasts. Proc. Natl. Acad. Sci. U. S. A. 72,3531-35350This article has been cited by other articles:
|
|
 |
|
  C. N. Zanghi, R. Sapinoro, B. Bradel-Tretheway, and S. Dewhurst A tractable method for simultaneous modifications to the head and tail of bacteriophage lambda and its application to enhancing phage-mediated gene delivery Nucleic Acids Res., April 3, 2007; 35(8): e59 - e59. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. N. Zanghi, H. A. Lankes, B. Bradel-Tretheway, J. Wegman, and S. Dewhurst A simple method for displaying recalcitrant proteins on the surface of bacteriophage lambda Nucleic Acids Res., October 13, 2005; 33(18): e160 - e160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Burg, K. Jensen-Pergakes, A. M. Gonzalez, P. Ravey, A. Baird, and D. Larocca Enhanced Phagemid Particle Gene Transfer in Camptothecin-treated Carcinoma Cells Cancer Res., February 1, 2002; 62(4): 977 - 981. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. ODERMATT, A. AUDIGÉ, C. FRICK, B. VOGT, B. M. FREY, F. J. FREY, and L. MAZZUCCHELLI Identification of Receptor Ligands by Screening Phage-Display Peptide Libraries Ex Vivo on Microdissected Kidney Tubules J. Am. Soc. Nephrol., February 1, 2001; 12(2): 308 - 316. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Eguchi, T. Akuta, H. Okuyama, T. Senda, H. Yokoi, H. Inokuchi, S. Fujita, T. Hayakawa, K. Takeda, M. Hasegawa, et al. Protein Transduction Domain of HIV-1 Tat Protein Promotes Efficient Delivery of DNA into Mammalian Cells J. Biol. Chem., July 6, 2001; 276(28): 26204 - 26210. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
