Bioconjugate Chem. 1994, 5, 612-620
612
Targeted Delivery of DNA Using YEE(GalNAcAH)3, a Synthetic Glycopeptide Ligand for the Asialoglycoprotein Receptor J u n e Rae Merwin,* G. Stephen Noell, Wendy L. Thomas, Henry C. Chiou, Mary E. DeRome, Timothy D. McKee, George L. Spitalny, a n d Mark A. Findeis TargeTech, Inc., Meriden, Connecticut 06450. Received May 24, 1994@
I n vivo gene therapy shows promise as a treatment for both genetic and acquired disorders. The hepatic asialoglycoprotein receptor (ASGPr) binds asialoorosomucoid-polylysine-DNA (ASOR-PLDNA) complexes and allows targeted delivery to hepatocytes. The tris(N-acetylgalactosamine aminohexyl glycoside) amide of tyrosyl(glutamy1) glutamate [YEE(G~~NACAH)~] has been previously reported to have subnanomolar affinity for the ASGPr. We have used a n iodinated derivative of YEE(GalNAcAH13 linked to polylysine and complexed to the luciferase gene (pCMV-Luc) in receptorbinding experiments to establish the feasibility of substituting ASOR with the synthetic glycopeptide for gene therapy. Scatchard analyses revealed similar & values for both ASOR and the glycopeptide. Binding and internalization of 1251-S~~-YEE(GalNAcAH)3 were competitively inhibited with either unlabeled ASOR or glycopeptide. The reverse was also true; lZ5I-ASORbinding was competed with unlabeled YEE(GalNAcAH)3suggesting specific binding to the ASGPr by both compounds. Examination of i n vivo delivery revealed that the lZ5I-labeledglycopeptide complex mimicked previous results observed with lZ5I-ASOR-PL-DNA. CPM in the liver accounted for 96% of the radioactivity recovered from the five major organs (liver, spleen, kidney, heart, and lungs). Cryoautoradiography displayed iodinated glycopeptide complex bound preferentially to hepatocytes rather than nonparenchymal cells. I n vitro, as well as in vivo, transfections using the glycopeptide-polylysine-pCMV-luciferase gene complex (YG3-PL-Luc) resulted in expression of the gene product. These data demonstrate that the YEE(GalNAcAH)3 synthetic glycopeptide can be used as a ligand in targeted delivery of DNA to the liver-specific ASGPr.
INTRODUCTION
The introduction of exogenous genes into cells is the basis for gene therapy. This expanding area of research encompasses widespread opportunities to treat both genetic and acquired disorders. Most of the gene therapy protocols currently in the clinic utilize an engineered viral vector andor ex vivo manipulation of autologous cells followed by reintroduction into the donor (reviews, 1-5). The lack of cellular specificity and the impracticality of treating large populations with ex vivo protocols indicate the need for improved gene-transfer technologies. Targeted delivery of therapeutic genes via receptor-mediated endocytosis has the potential to address these concerns (6, 7). Ehrlich conceptualized targeted delivery and coined the term “magic bullet” early in the 20th century as a result of his pioneering research in immunology and medicinal chemistry (8). However, it was not until the 1980’s that this principle was extended to targeted delivery of DNA. Research in this area was initiated by Wu and Wu (9) using the asialoglycoprotein receptor (ASGPrIl which is
* Present address: Advanced Technologies-Cell Biology, The West Company, 101 Gordon Street, P.O. Box 645, Lionville, PA 19341-0645. Tel: 215-594-2900. Fax: 215-594-3000. Abstract published in Advance ACS Abstracts, September 15, 1994. Abbreviations: ASGPr, asialoglycoprotein receptor; ASOR, asialoorosomucoid; ASOR-PL-DNA, asialoorosomucoid-polylysine-DNA complex; EDC, 1-(3-(dimethylamino)propyl)-l’ethylcarbodiimide; Gal, Galactose; GalNAc, N-acetylgalactosamine; HSA, human serum albumin; O N , overnight; PBS, phosphate buffered saline; rt, room temperature; YEE(Ga1NAcAHh, tridN-acetylgalactosamine aminohexyl glycoside) amide of tyrosyl(glutamy1)glutamate; YG3-PL-DNA, YEE(Ga1NAcAH)Z-polylysine-DNA complex; YG3-PL-Luc, YEE(GalNAcAHIz-polylysine-luciferase DNA complex. @
unique to hepatocytes and binds branched galactoseterminal glycoproteins, such as asialoorosomucoid (ASOR, 10). Liganareceptor complexes have been shown to internalize via receptor-mediated endocytosis engaging the endosomal-lysosomal pathway (11). I n vitro DNA delivery has been accomplished by targeting the more ubiquitous transferrin receptor (12, 13). Previous research has also included i n vitro gene transfer using insulin-polylysine conjugates that are internalized by hepatocytes (14) and antibody conjugates specific to antigen-bearing cells (15). In general, the complex used for delivery consists of a covalently linked ligandpolycation conjugate which binds DNA in a n electrostatic manner (16). To substantiate transcription and translation, the gene product of choice was measured. These gene products included membrane bound, intracellular, or secreted proteins (17-19). Recently reported synthetic ligand-based conjugates for targeted delivery of DNA have demonstrated the viability of using galactose-containing ligands smaller than proteins to achieve receptor binding and internalization of the ligand-bound DNA (20,21). The use of synthetic and semisynthetic ligands in DNA delivery is a relatively unexplored field. We show herein that the use of the synthetic peptide YEE(GalNAcAH)3parallels the biological activity of ASOR (a natural ligand for the ASGPr) in binding, uptake, and competition studies. We have also demonstrated i n vitro and i n vivo targeting and gene expression using a YEE(GalNAcAH)3-polylysine-DNA complex (YG3-PL-DNA). EXPERIMENTAL PROCEDURES
Synthesis. General. Poly-L-lysine and human serum albumin (HSA; -97% pure) were purchased from Sigma, and l-(3-(dimethylamino)propyl)-l’ethylcarbodiimide (EDC) was purchased from Aldrich. Synthetic Cluster
1043-1802/94/2905-0612$04.5OiO 0 1994 American Chemical Society
Bioconjugate Chem., Vol. 5, No. 6,1994 613
YEE(GalNAcAH)3 Synthetic Peptide in Gene Therapy A. 1
3
2 /
Table 1. List of Compounds and Their Descriptions"
SDS-PAGE 4
compd YEE(GalNAcAH)3 Suc-YEE(GalNAcAH)s HSA-Suc-YEE(GalNAcAH)3
5
*
Luc-PL-HSA-Suc-YEE(GalNAcAH)3abbreviated as YG3-PL-Luc ASOR-PL-LUC
c
E. ACID UREA GEL 1 2 3 4
C. RETARDATION GEL 1
2
. )
Figure 1. Electrophoresis of material. (A) 4-20% gradient SDS-PAGE: lanes 1 and 5, molecular weight markers 200, 97.4,69,and 46 kD (small arrows); lane 2,YEE(GalNAcAH)s; lane 3, PL-HSA-Suc-YEE(GalNAcAH)3; lane 4, HSA. (B) Acid-urea gel: lane 1, 4 kD polylysine; lane 2, purified PLHSA-Suc-YEE(GalNAcAH)s conjugate; lane 3, HSA, lane 4, ASOR. (C) Gel retardation assay: lane 1, DNA alone (arrowhead); lane 2,purified DNA fully retarded in the well due to complexation with conjugate (arrow).
Ligands. The synthesis (22)and a n alternative stepwise synthesis of YEE(GalNAcAH)3, along with the synthesis of succinyl-YEE(GalNAcAH)3 (23), have been described. Polylysine-HSA -Suc -YEE(GalNAcAH)3. SUC -YEE (GalNAcAH)3(25 mg, 17.3 pmol) and HSA (50 mg) were dissolved in 10 mM MES Biological Buffer (EM Scientific Corp.), pH 5 . EDC (33 mg, 172 pmol) was added and the resulting mixture stirred for 2 h at room temperature (rt). Polylysine (4 kD average MW, 50 mg) and additional EDC (25 mg) were added to the reaction mixture which was stirred for a n additional 2 h. The reaction mixture was dialyzed sequentially: two times against 1M guanidine hydrochloride, one time against 1M NaC1, two times against water, and then lyophilized. The final lyophilized yield was 53.6 mg (-43% of reactants). An amino acid analysis revealed a n average of 6.9 mol of YEE(Ga1NAcAH)dmol HSA and 0.7 mol of PUmol of HSA. A 4-20% gradient SDS-PAGE was performed on SucYEE (GalNAcAH)3, purified PL -HSA- Suc-YEE( GalNAcAH)~conjugate, and HSA (Figure 1A). Suc-YEE(GalNAcAH)3 ran through the gel due to its small MW (-1400 D; lane 2); while HSA alone displayed a prominent band at -65 kD as well as higher molecular weight bands suggesting impurities in the HSA sample (lane 4). The PL-HSA-Suc-YEE(GalNAcAH)3 conjugate, due to the high positive charge and possible aggregation, remains primarily at the top of the gel (lane 3). A small amount of conjugate entered the gel and is shown at -70 kD which would be the expected MW when considering the addition of Suc-YEE(GalNAcAH)3 and polylysine to HSA. Due to the difficulty in deciphering the conjugate on SDS-PAGE, it was necessary to perform acid urea gel electrophoresis on free polylysine, the purified con-
explanation basic triantennary cluster ligand cluster ligand plus succinyl group cluster ligand plus human serum albumin linked via succinyl group cluster ligand plus human serum albumin linked via succinyl group, polylysine covalently linked to HSA, and complexed plasmid DNA (pCMV-LUC) asialoorosomucoid plus polylysine and complexed plasmid DNA (pCMV-Luc) (positive control)
a The basic cluster ligand is used as the core molecule with additions of 1251,a succinyl group, human serum albumin, polylysine, pCMV-Luc, or a combination thereof.
jugate, HSA, and ASOR (Figure 1B). This protocol, as described by McKee et al. (24), separates bands primarily on the basis of charge as well as size, due to the high concentration of urea and the acidic conditions. As a result, the ASOR molecule (lane 4), which is acidic in nature, does not migrate into the gel as far as HSA (lane 3). The rationale for this procedure was to determine the purity of the conjugate as far as the amount of free polylysine remaining in the conjugate sample. The arrow points out the prominent band for the purified conjugate (lane 2). A small amount of free polylysine within the sample appeared at the bottom of the gel (lane 1, arrowhead). The high MW banding pattern of the purified conjugate (lane 2) is similar to that of HSA alone (lane 3 and Figure 1A) which may be due to the impurities of the HSA sample as well as cross linking of the molecules. However, they do not appear to interfere with the ability of the conjugate to complex DNA. The asialoorosomucoid-polylysine conjugate was prepared as described previously using 5 kD MW polylysine (24). Plasmid Construct. pCMV-Luc was a kind gift from Dr. James Economou (UCLA Medical Center, Los Angeles, CA). This plasmid contains the firefly luciferase gene derived from plasmid pXPl(25) under the control of the cytomegalovirusimmediate early promoter. The plasmid was maintained and propagated in E. coli strain DH5a. Isolation of plasmid DNA was accomplished by standard alkaline/detergent lysis of saturated bacterial cultures grown in Terrific Broth (26). Supercoiled plasmids were then purified by double banding on cesium chloride gradients. All DNAs were EtOH precipitated and stored in 100 mM Tris pH 7.5, 10 mM EDTA at 4 "C. Complex Formation. A YG3-PL-DNA complex was formed as described (24).Briefly, 3.0 mg of purified conjugate (as a 5 mg/mL aqueous solution) was slowly added to 1mg of pCMV-Luc DNA resulting in 300 pg of DNA/mL in 0.15 M NaC1. The complex was then filtered through a 0.45 pm filter (Acrodisc, Gelman Sciences), and the final concentration was found to be 177 pg/mL by U V absorbance. An aliquot of the complex was then run on a n 1% agarose gel to confirm full retardation of the DNA (Figure IC). Once the DNA is fully complexed with conjugate, the DNA will retard in the well and not enter the gel bed (lane 2, arrow). This sample retarded at a 3:1conjugate:DNA weight:weight ratio. A complete list of compounds used in our experiments is shown in Table 1. Cell Surface Binding Analyses. Binding Studies. HUH-7 human hepatoma cells (generous gift of Dr. T. J. Liang, Massachusetts General Hospital, Boston, MA) were plated at 0.25 x lo6cells/well in 24 well plates and grown overnight ( O N ) to -75% confluence in MEM medium with 10% fetal calf serum. Suc-YEE(Ga1NAcAH)~was iodinated as described (27,28). Half of the
Merwin et al.
614 Bioconjugafe Chem., Vol. 5, No. 6, 1994
a.
OH
0
N 0H -
AcH& OH
1, R = H : YEE(GalNAcAH)3
2, R = COCH2CH,CO2H : Suc-YEE(GalNAcAH),
b.
Figure 2. Chemical structures. (A) Structure of the trivalent cluster glycoside YEE(GalNAcAH)3(1)and its N-succinyl derivative (2). (B)Structure of polylysine-HSA-Suc-YEE(GalNAcAH)s. The number of equivalents of cluster ligand (XIand polylysine (y) bound to HSA will vary with the conditions under which the cross linking reactions are conducted.
samples were pretreated with 150-fold molar excess unlabeled Suc-YEE(GalNAcAH)3 for 1h at 4 "C enabling us to ascertain nonspecific binding. The cells were incubated with varying amounts of 1251-Suc-YEE(GalNAcAH)~(specific activity = 1.5 x 1O1O CPWpmol) in serum-free medium for 2 h a t 4 "C. Free ligand was removed from the cells with three washes of phosphate buffered saline (PBS) and the cells lysed with 0.5 mL of 0.2 N NaOH. The lysate was measured for radioactivity in a y counter. Similar analyses were performed on ASOR to compare the binding properties of both ASGPr ligands in our hands. Internalization Studies. To distinguish endocytosed from externally bound ligand, the cells were incubated with 0.1pM radiolabeled YGS-PLDNA complex (specific activity = 1.7 x lo6CPWpg DNA) for up to 3 h in serum-free medium containing 3.8 mM Ca2+a t 37 "C. The negative control was 1251-orosomucoid which does not bind the ASGPr. The monolayers were rinsed three times with PBS and incubated with 10 mM EDTMBS, pH 2.5 for 5 min a t rt to dissociate externally bound ligand (29, 30). Background binding was determined by preincubating cells with 0.1% sodium azide for 1 h a t 4 "C to inhibit internalization and then treating
the cells with the experimental ligand for an additional hour a t the same temperature. Samples were washed, and externally bound receptorAigand complexes were dissociated as stated above. The remaining background counts were subtracted from the total radioactivity and the final value considered as ligand taken up by the cells. Competition Studies. To determine receptor specificity, HUH-7 cells were incubated simultaneously with 1pM radiolabeled YEE(GalNAcAH13 plus varying concentrations of unlabeled glycopeptide or ASOR for 2 h a t 4 "C. The reverse experiment was also performed using radiolabeled ASOR and increasing amounts of unlabeled YEE(GalNAcAH)3. Representative experiments are shown in the figures; however, binding, internalization, and competition studies were repeated using triplicate samples in three or more assays. In Vivo Targeting and Clearance. Balb/C mice were tail vein injected with lo6 CPM of lZ5I-YG3-PLLuc complex in 0.5 mL of PBS. Animals were sacrificed by cervical dislocation 5 min post-injection. Major organs were excised and, along with the remainder of the carcass, were counted in the y counter to determine the amount of radioactivity distributed throughout the body.
Bioconjugate Cbem., Vol. 5, No. 6, 1994 615
YEE(GalNAcAH)3 Synthetic Peptide in Gene Therapy
A.
-
1600
1400 -
1.4k
1000 800 1200
1.2
A
.-
0 Y
P 400 - I 600
y
r,\
0.6j 0.44 0.24 0.0
0.010
0.000
-
200
0
0.020
nM Bound
--
I
I
4
I
1
2
3
1
0
4
5
uM
B. 3000
-
2500
-
2000
-
1500:
1000
-
500
-
f.
o!'
0.0
nM Bound '
I
0.1
'
I
0.2
'
I
0.3
.
,
t
0.4
0.5
.
1
0.6
uM Figure 3. ASGPr/glycopeptide binding analyses. Binding studies were performed in which half of the samples were pretreated (A) or with 150-fold molar excess unlabeled ligand and then incubated with various concentrations of 1251-S~~-YEE(GalNAcAH)3 1251-ASOR (B) as described in the Experimental Procedures. Free ligand was removed, the cells lysed, and radioactivity counted. The data revealed dose-dependent binding curves which approached saturation of receptors. Scatchard plots are shown in the insets.
Cryoautoradiography. Frozen Section Preparation. 1251-YG3-PL-Luc complex (4 x lo6 CPM) in 0.5 mL of PBS was tail vein injected into Balb/C mice which were sacrificed 5 min post-injection. The liver was perfused with 3% paraformaldehyde, excised, cut into 2-3 mm pieces, and post fixed in the same fixative for 30 min at rt. The tissue was rinsed two times with PBS and infused with 0.5 M sucrosePBS O/N at 4 "C for cryoprotection, after which it was infiltrated with two parts 20% sucrose/PBS to one part OCT embedding medium (Miles, Diagnostic Division, Elkhart, IN) in a cryomold for 30 min at rt. The tissue was rapidly submerged in liquid nitrogen cooled isopentane and stored a t -80 "C until sectioned a t 3 pm on a cryostat, mounted onto chrom aludgelatin subbed slides, air dried a t rt, and stored a t 4 "C until used. Autoradiography. Under safelight conditions, the slides were dipped vertically into Kodak NTB-3 liquid emulsion film (International Biotechnologies, Inc., New
Haven, CT) at 43 "C, after which the backs of the slides were wiped clean. The slides were air dried for 1.5 h and placed into black slide boxes containing tissue-wrapped desiccant. The boxes were sealed with black electrical tape and stored on end with the emulsion facing upward a t 4 "C for 3-7 days. After exposure, the slides were developed under red safelight conditions in Kodak Dektol (1:1 with distilled water) for 2 min, submerged in Kodak fix for 5 min, and rinsed with distilled water for 5 min (all a t 14 "C). The slides were air dried in a dust free environment followed by staining with 0.5% toluidine blue in 1%benzoic acid for 1-2 min, rinsed three times in distilled water, and air dried. Coverslips were mounted onto slides with Cytoseal60 (Stephens Scientific, Riverdale, NJ) and the slides viewed under a microscope. Negative controls consisted of livers excised from normal mice which were handled and treated in the same manner as experimental samples. In Vitro Transfections. HUH-7 hepatocytes were
616 Sioconjugate Chem., Vol. 5, No. 6,1994
Merwin et al.
A.
a I m
!W2
600000
-
500000
-
400000
-
300000
-
200000
-
100000
0
n0
ina I
W
T
T
...... ...... :;:;:;:;:;:; ...... ...... ...... ...... ...... ...... ...... ...... ...... ......
...... ............ ...... ......
...... ...... ...... ...... ...... ...... ...... ...... ...... ......
TIME
T
....... ............. ....... ...... ....... ...... ....... ...... ....... ...... ....... ...... ....... ...... ....... ...... ....... ...... ....... ...... ......
...... ...... ...... ...... ...... ...... ............ ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ............ ...... ...... ...... ...... ...... ......
(mlnutes)
B. 3.5
-
32.5
Y EE(GelNAcAH)3 Complex
21.5
Orosomucoid
-
1-
TIME
(minutes)
Figure 4. Internalization study. HUH-7 cells were incubated with 0.1 pM iodinated Suc-YEE(GalNAcAH)3 complex for increasing time periods. External receptor-bound ligand was dissociated and nonspecific background binding subtracted. (A) The assay revealed that YG3-PL-Luc complexes were taken up by the cell with the amount internalized increasing over time. (B)After 3 h of incubation, the amount of internalized YEE(GalNAcAH)3complex was '3.5 times background.
seeded a t 1.25 x lo6 celldl00 mm tissue culture dish and grown O/Nto 40-50% confluence in MEM medium with 10% fetal calf serum. The monolayers were rinsed three times with PBS and incubated in medium containing 100 pM chloroquine for 1 h a t 37 "C and the rinse step repeated. Chloroquine is a lysosomatropic agent which has been shown to enhance the efficiency of receptormediated gene transfer in vitro (31). Fresh medium containing YG3-PL-Luc complex (12 pg of DNA) was added. The concentration of CaClz was increased to 3.8 mM as the transfections are mediated via receptor binding and endocytosis which is a calcium-dependent process. After 1.5-3 h, the cells were washed two times in PBS and the medium changed. At 48 h r the cells were harvested for quantitation of luciferase expression (32) using an AutoLumat Luminometer LB953 (EG & G Berthold, Pittsburgh, PA). In Vivo Gene Expression. One h prior to IV injection of YG3-PL-Luc complex, Balb/C mice were treated with colchicine and chloroquine by ip injection (0.75 and 40 mgkg, respectively). Intravenous injection of YG3-PLLuc complex was performed via the tail vein with a
standard dose of 300 pg complexed DNA in 1mL of saline containing 0.15 M NaC1. Three days post-injection, the mice were sacrificed. Approximately 0.4 g of liver was excised, homogenized, and processed, and the quantity of intracellular luciferase was determined as stated above (32). RESULTS AND DISCUSSION
YEE(GalNAcAH)sSynthesis and Formation of the DNA Complex. The cluster glycoside YEE(GalNAcAH)3 (Figure 2A) is the highest affinity synthetic ligand of the ASGPr reported to date (22). In binding assays using rat hepatocytes, this compound was reported to effect a 50% inhibition a t a concentration of 0.2 nM of ASOR binding (22). Lee and Lee obtained this high specificity through the use of N-acetylgalactosamine as the carbohydrate moiety which has higher affinity for the receptor in comparison with galactose. The use of a flexible branched peptide backbone provides a triantennary array of GalNAc groups to allow tight binding to the multivalent ASGPr. In light of the demonstrated affinity of this
Bioconjugate Chem., Vol. 5, No. 6, 1994 617
YEE(GalNAcAH)3 Synthetic Peptide in Gene Therapy
A. n
z
A.
30007
IT
100 -I
W H
2500
%
asoravg Yee avg
2000
r
-I
1500
s-0
E w
1000
w
7
y)
so0
N r
0 0
0.
37.5
75
150
300
Competitor (pM)
Organs
3000
B. 2500 n
2000
0
v
e
-p:
In N
1500
1000
r
500
0 0
0.7
1.4
2.8
5.6
Competition with Unlabeled YEE(Ga1NAcAH) 3
bM]
Figure 5. Competition studies. HUH-7 cells were incubated with 1 pM iodinated Suc-YEE(GalNAcAH)3 and varying concentrations of unlabeled glycopeptide or ASOR (A). The reverse competition study was also performed (B) using IZ5I-ASOR competed with 300-fold molar excess unlabeled Suc-YEE(GalNAcAH13which is the same maximum molar increase used in A.
compound for the ASGPr, we sought to evaluate it as a ligand for targeted delivery to liver cells. To facilitate the formation of conjugates between YEE(GalNAcAHI3and amine-containing macromolecules, the N-succinyl derivative, SUC--YEE(G~~NACAH)~, was prepared by reaction of the tyrosyl N-terminal amino group with succinic anhydride. The free carboxyl group allows carbodiimide-mediated cross linking in aqueous solution between the fully deprotected cluster glycoside and polylysine or HSA. Polylysine-HSA-Suc-YEE(Ga1NACAH)~ was used directly to form a targetable electrostatic complex with DNA (Figure 2B). Binding Analysis. To address whether YEE(Ga1NACAH)~ could replace ASOR as a n ASGPr ligand, a series of binding studies were performed. lZ5I-SucYEE(GalNAcAH13 was shown t o bind to HUH-7 hepatocyte monolayers in a dose dependent manner approaching saturation of receptors a t 2 pM (Figure 3A). Nonspecific background binding determined by incubation with 150-fold excess unlabeled peptide was subtracted from total binding. Duplicate experiments were performed using lZ5I-ASOR (Figure 3B). Scatchard analyses of both 1251-S~~-YEE(GalNAcAH)3 and lZ5I-ASOR binding revealed similar results; namely, Kd values of 6 x and 2 x M, respectively (Figure 3A,B). Lee
Organ liver spleen kidney heart lung pancreas stomach intestine uterus and ovaries tail throat hind legs front legs body: 1 body: 2 brain head blood (100 ul) blood clot total body
b of Total Counts
81.04 0.92 1.11 0.09 1.43 0.13 0.30 0.97 0.15 0.43 0.83 0.65 0.42 1.90 1.75 0.03 1.10 0.14 0.20 93.59
Figure 6. In vivo targeting and clearance analyses. lo6 CPM of 1251-Suc-YEE(GalNAcAH)3 complex was tail vein injected into Balb/C mice. The major organs were removed at 5 min postinjection and radioactivity counted to determine percent of labeled complex targeted to the liver. The remaining carcass was dissected and counted to analyze total biodistribution.
and Lee (22) have shown that the glycopeptide without a succinyl group exhibited subnanomolar affinity for the ASGPr; however, we observed nanomolar range affinities for both ASOR and Suc-YEE(GalNAcAH)3. The differences in binding affinities may be attributed to the experimental protocol as well as the culture systems used (human hepatocyte cell line used in house versus freshly isolated hepatocytes or soluble ASGPr; 22). The object of this project was to compare Suc-YEE(Ga1NAcAH)B to ASOR using both in vivo and in vitro assays. While the affinity of Suc-YEE(GalNAcAW3 for ASGPr appears to be less than previously reported, our data a E r m that Suc-YEE(GalNAcAH)3 possesses similar ASGPr binding properties compared to ASOR in our hands. Incubating the ASGPr-negative hepatocyte cell line, SK-Hep-1, with 1251-Suc-YEE(GalNAcAH~3resulted in only minimal binding. In competition studies using 100fold molar excess unlabeled Suc-YEE(GalNAcAH)3,no change was seen in the SK-Hep-1 binding results, suggesting nonspecific binding rather than ASGPr engagement (data not shown). Internalization Studies. In order to validate YEE(GalNAcAH)3usage as a ligand for targeted delivery of
618 Bioconjugate Chem., Vol. 5, No. 6,1994
Merwin et al.
Figure 7. Hepatocyte localization of iodinated glycopeptide. Perfused livers were removed from Balb/C mice injected with 4 x
lo6
counts of iodinated YG3-PL-Luc complex and processed for autoradiography as stated in the Experimental Procedures. Staining of negative control livers was negligible (B) while 1251-Suc-YEE(GalNAcAH)~complex injected livers exhibited positive silver grains located in the parenchymal hepatocytes (C, D). A section from the same liver was processed for immunofluorescence and stained with an antibody specific to Kupffer cells. Vascular cells gave no evidence of iodinated glycopeptide uptake (C). V = vasculature, P = parenchyma, arrows = endothelial cells. Original magnification (A, B, C) = 600 x , (D) = 1000x .
DNA to the ASGPr, it was necessary to establish that the glycopeptide-PL-DNA complex was able to be taken up by hepatocytes. We established internalization by incubating HUH-7 cells with iodinated YG3-PL-Luc complex. Following incubation, half of the samples were treated to dissociate external ligandlreceptor binding. As seen in Figure 4A, there was an escalating amount of internalization of complex over time. After 3 h of incubation, the amount of complex taken up by the cell was 3.5-fold (Figure 4B). Iodinated orosomucoid was used as a negative control and was not internalized by the hepatocytes. Our results verify that Suc-YEE(GalNAcAH)3 complexes were capable of binding the ASGPr and being endocytosed. Competition Studies. To demonstrate receptor specificity, competition assays were performed using HUH-7 cells incubated with 1251-S~~-YEE(GalNAcAH)3and increasing concentrations of unlabeled Suc-YEE(Ga1NAcAH)~or ASOR (Figure 5A). As shown, both the unlabeled peptide and native ligand were able to compete for ASGPr binding with 1251-Suc-YEE(GalNAcAH)3. The reverse experiment, unlabeled peptide competing with iodinated ASOR, confirmed ASGPr specific binding (Figure 5B). In Vivo Targeting and Clearance. I n vivo studies demonstrated highly selective targeting and clearance of the 1251-YG3-PL-Luc complex to the liver. Within 5 min post-injection, -96% of the radioactivity recovered
from the five major organs was located in the liver: lungs = 1.2%,heart = 0.3%, spleen = 1.7%,and kidneys = 0.7% (Figure 6A). The remainder of the carcass contained minimal amounts of the radioactivity recovered from the body (= -94% of total radionucleotide injected; Figure 6B). These in vivo data corroborate our i n vitro results showing 1251-S~~-YEE(GalNAcAH)3 complex uptake via the liver specific ASOR receptor. Cryoautoradiography was used to visualize labeled complex within the liver 5 min post-injection. Positive silver grains were observed located in and surrounding hepatocytes (Figure 7C,D) while nonparenchymal endothelial cells (arrows, Figure 7C) and control cultures (Figure 7B) exhibited only background amounts of radioactivity. In order to distinguish nonparenchymal Kupffer cells from hepatocytes, sections from the same liver were stained with antibodies specific for this macrophage. A representative photograph shown in Figure 7A revealed a pattern of staining markedly different from that of the silver grains which depicted localization of the YG3-PL-Luc complex in liver parenchyma. In Vitro Transfections. In order to show that YEE(GalNAcAH)3 could be used for gene therapy, we transfected cells with a YG3-PL-Luc complex to demonstrate expression of the transfected gene. ASOR-PL-Luc transfections constituted the positive control while ASGPr-negative 3T3 fibroblasts from ATCC were used
Bioconjugate Chem., Vol. 5, No. 6, 1994 619
YEE(GalNAcAH)3Synthetic Peptide in Gene Therapy
(3) Mulligan, R. C. (1991) Gene transfer and gene therapy:
Table 2. In Vitro Transfectionsa
cell type HUH-7
YG3-PL-Luc 197 319 228 873
3T3
69 91
ASOR-PL-pCMV-Luc 667 077 515 464 51 63
a HUH-7 cells and ASGPr-negative 3T3 fibroblasts were transfected with both YG3-PL-Luc and ASOR-PL-Luc. Values from the 3T3 cells were at background levels (-50 RLU).
Table 3. In Vivo Expressiona
animal mouse 1 mouse 2 mouse 3 mouse 4
RLU 8836 12 097 1943 116 772
a Four mice were injected with YG3-PL-Luc complex and the livers excised 3 days later. Luciferase activity was measured from each animal with detection of gene expression in all cases (background -50 RLU).
as the negative control. The results, presented in Table 2, revealed positive transfections by both YG3-PL-Luc and ASOR-PL-Luc complexes. Values equaling background were observed with the receptor-minus fibroblast cells. A 3T3 cell line stably transformed with the ASGPr (33)was also used for in vitro transfections and resulted in positive transfections by both ASOR and YEE(Ga1NAcAH)~complexes. However, the values for the transformed 3T3 fibroblasts were lower than the HUH-7 hepatocytes (data not shown). We have shown, herein, that YEE(GalNAcAH13 can substitute for ASOR in a ligand-PL-DNA complex for in vitro delivery of DNA to hepatocytes via the ASGPr. In Vivo Gene Expression. To further illustrate the use of YEE(GalNAcAH13,we injected the YG3-PL-Luc complex into mice. Two days later, the livers were removed and homogenized, followed by determination of luciferase activity. As shown in Table 3, all four mice were positive for luciferase activity. Mouse 4 expressed an unusually high value, while mice 1-3 are within a 6-fold difference in RLU. Summary. We have demonstrated the use of the synthetic glycopeptide YEE(GalNAcAHl3 as a ligand for targeted delivery of DNA via the ASGPr. Specific binding to this receptor has been validated by binding, competition, and internalization studies. Functional confirmation was provided by in vitro transfections as well as in vivo targeting, cryoautoradiography, and expression studies. YEE(GalNAcAH)3 complexed to DNA was able to duplicate the results obtained in similar experiments using ASOR-PL-DNA. Taken together, these data demonstrate the use of the synthetic cluster ligand, YEE(GalNAcAHI3, for the delivery of DNA specifically to hepatocytes via the ASGPr. ACKNOWLEDGMENTS
The authors would like to thank Mr. Thomas Ardito (Yale School of Medicine) for his assistance with microscopic techniques and Dr. T. J. Liang (Massachusetts General Hospital) and Dr. J. Economou (UCLA Medical Center) for their generous gifts of HUH-7 cells and pCMV-Luc, respectively. LITERATURE CITED (1) Anderson, W. F. (1992) Human gene therapy. Science 256, 808-813. (2) Mulligan, R. C. (1993) The basic science of gene therapy. Science 260, 926-932.
Principles, prospects, and perspective. Etiology of human disease a t the DNA level. (J.Lindsten, and U. Petterson, Eds.) pp 143-189, Raven Press, New York. (4) Thompson, L. (1992) At age 2, gene therapy enters a growth phase. Science 258, 744-746. (5) Miller, A. D. (1992) Human gene therapy comes of age. Nature 357, 455-460. (6) Findeis, M. A., Menvin, J. R., Spitalny, G. L., and Chiou, H. C. (1993) Targeted delivery of DNA for gene therapy via receptors. Trends Biotechnol. 11, 202-205. (7) Chiou, H. C., Spitalny, G. L., Merwin, J . R., and Findeis, M. A. (1994)I n vivo gene therapy via receptor-mediated DNA delivery. Gene Therapeutics: Methods and Applications and Direct Gene Transfer. (J.A. Wolff, Ed.) pp 143-156, Birkhauser Publishers, Boston. (8) Baumler, E. (1984) Paul Ehrlich, Scientist for Life, p 108, Holmes & Meier Publishers, New York. (9) Wu, G. Y., and Wu, C. H. (1987) Receptor-mediated in vitro gene transformation by a soluble DNA carrier system. J.Biol. Chem. 262, 4429-4432. (10) Ashwell, G., and Morell, A. G. (1974) The role of surface carbohydrates in the hepatic recognition and transport of circulating glycoproteins. Adv. Enzymol. Relat. Areas Mol. Biol. 41, 99-128. (11) Wall, D. A., Wilson, G., and Hubbard, A. L. (1980) The galactose-specific recognition system of mammalian liver: the route of ligand internalization in rat hepatocytes. Cell 21,7983. (12) Wagner, E., Zenke, M., Cotten, M., Beug, H., and Birnstiel, M. L. (1990) Transferrin-polycation conjugates as carriers for DNA uptake into cells. Proc. Natl. Acad. Sci. U S A . 87,34103414. (13) Wagner, E., Plank, C., Zatloukal, K., Cotten, M., and Birnstiel, M. (1992) Influenza virus hemagglutinin HA-2 N-terminal fusogenic peptides augment gene transfer by transferrin-polylysine/D" complexes: Towards a synthetic virus-like gene transfer vehicle. Proc. Natl. Acad. Sci. U.S.A. 89, 7934-7938. (14) Rosenkranz, A. A., Yachmenev, S. V., Jans, D. A., Serebryakova, N.V., Murav'ev, V. I., Peters, R., and Sobolev, A. S. (1992) Receptor-mediated endocytosis and nuclear transport of a transfecting DNA construct. Exp. Cell Res. 199,323329. (15) Trubetskoy, V. S., Torchilin, V. P., Kennel, S. J., and Huang, L. (1992) Use of N-terminal modified poly@-lysine)antibody conjugate as a carrier for targeted gene delivery in mouse lung endothelial cells. Bioconjugate Chem. 3,323-327. (16) Wu, G. Y., and Wu, C. H. (1988) Receptor-mediated gene delivery and expression in vivo. J. Biol. Chem. 263, 1462114624. (17) Wilson, J. M., Grossman, M., Wu, C. H., Chowdhury, N. R., Wu, G. Y., and Chowdhury, J. R. (1992) Hepatocytedirected gene transfer in vivo leads to transient improvement of hypercholesterolemia in low density lipoprotein receptordeficient rabbits. J. Biol. Chem. 267, 963-967. (18) Wu, G. Y., and Wu, C. H. (1988) Evidence for targeted gene delivery to Hep G2 hepatoma cells in vitro. Biochemistry 27, 887-892. (19) Wu, G. Y., Wilson, J. M., Shalaby, F., Grossman, M., Shafritz, D. A., and Wu, C.H. (1991) Receptor-mediated gene delivery in vivo: Partial correction of genetic analbuminemia in Nagase rats. J. Biol. Chem. 266, 14338-14342. (20) Plank, C., Zatloukal, K., Cotten, M., Mechtler, K., and Wagner, E. (1992) Gene transfer into hepatocytes using asialoglycoprotein receptor mediated endocytosis of DNA complexed with an artificial tetra-antennary galactose ligand. Bioconjugate Chem. 3, 533-539. (21) Haensler, J., and Szoka, F. C., Jr. (1993) Synthesis and characterization of a trigalactosylated bisacridine compound to target DNA to hepatocytes. Bioconjugate Chem. 4(1), 8593. (22) Lee, R. T., and Lee, Y. C. (1987) Preparation of cluster glycosides of N-Acetylgalactosamine that have subnanomolar binding constants toward the mammalian hepatic GaVGalNAc-specific receptor. Glycoconjugate J. 4, 317-328.
620 Sioconjugate Chem., Vol. 5, No. 6, 1994 (23) Findeis, M. A. (1994) Stepwise synthesis of a GalNAccontaining cluster glycoside ligand of the asialoglycoprotein receptor. Intl. J. of Pept. Prot. Res. 43, 477-485. (24) McKee, T. D., DeRome, M. E., Wu, G. Y., and Findeis, M. A. (1994) Preparation of asialoorosomucoid-polylysine conjugates. Bioconjugate Chem. 5, 306-311. (25) Nordeen, S. K. (1988) Luciferase reporter gene vectors for analysis of promoters and enhancers. BioTechniques 6(5), 454-456. (26) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular cloning, A laboratory manual, Appendix 2, Cold Spring Harbor Laboratory Press, New York. (27) Greenwood, F. C., and Hunter, W. M. (1963) The preparation of '31I-labeled human growth hormone of high specific radioactivity. Biochem. J . 89, 114-123. (28) DeLarco, J. E., Preston, Y. A., and Todaro, G. J. (1981) Properties of a sarcoma-growth-factor-like peptide from cells transformed by a temperature-sensitive sarcoma virus. J . Cell. Physiol. 109, 143-152. (29) Weigel, P. H., and Oka, J. A. (1982) Endocytosis anddegradation mediated by the asialoglycoprotein receptor in
Merwin et ai. isolated rat hepatocytes. J. Biol. Chem. 257, 1201-1207. (30) Ghinea, N., Hai, M. T. V., Groyer-Picard, T-T., Houllier, A., Schoevaert, D., and Milgrom, E. (1992) Pathways of internalization of the hCG/LH receptor: Immunoelectron microscopic studies in Leydig cells and transfected L-cells. J . Cell Biol. 118, 1347-1358. (31) Cotten, M., Langle-Rouault, F., Kirlappos, H., Wagner, E., Mechtler, K., Zenke, M., Beug, H., and Birnstiel, M. L. (1990) Transferrin-polycation-mediated introduction of DNA into human leukemic cells: Stimulation by agents that affect the survival of transfected DNA or modulate transferrin receptor levels. Proc. Natl. Acad. Sci. U.S.A. 87, 4033-4037. (32) Brasier, A. R. (1993) Introduction of DNA into mammalian cells. Current Protocols in Molecular Biology (F.M. Ausubel et al., Eds.) pp 9.6.10-19.6.14, John Wiley & Sons, Inc., Boston. (33) Shia, M. A., and Lodish, H. F. (1989) The two subunits of the human asialoglycoprotein receptor have different fates when expressed alone in fibroblasts. Proc. Natl. Acad. Sci. U.S.A. 86(4), 158-162.
