Lactosamination of liposomes and hepatotropic targeting research

Abstract

Key Words: liposomes, asialoglyco-protein, liver, interferon-alpha, antigens, viral, drug carriers, drug therapy, rats

INTRODUCTION

Site-specific delivery of therapeutic drugs to their target cells is a major s cientific challenge for the pharmaceutical sciences. It offers a number of advantages over conventional drug administration. With drug targeting, high local conc entrations of the drug can be achieved, thus circumventing many unwanted side ef fects. Various carriers have been suggested for the delivery of drugs, including liposomes[1-5] and (neo) glycoproteins[6-8]. The asialoglycopr otein receptor (ASGP-R) has frequently been utilized for targeting drugs to the parenchymal liver cell[6-12]. Liposomes have several advantageous cha racteristics as drug carrier, and particularly, ligand-tacked liposomes achieve a highly effective targeting[13]. Hara et al[14] reported that asialofetuin (AF)-tacked liposomes distributed to rat hepatocytes selectively in vivo, and ASGP-R mediated the uptake of AF-liposomes encapsulating IFN-γ by isolated rat hepatocytes in vitro[15]. Lacto saminated human serum albumin (L-HSA) is a neoglycoprotein taking number of galactose residue as terminal sugar[6].

In this paper, we studied the preparation and rat hepatocyte uptake of the conjugate of L-HSA and liposomes, and the inhibitory effect of L-HSA-liposomes containing IFN-α on replication of hepatitis B virus (HBV) on 2.2.15 cells line.

MATERIALS AND METHODS

Materials

Sodium cyanoborohydride was purchased from Aldrich, Chemical Co., Milwaukee, WI, USA, α-lactose, phosphatidylcholine (PC), cholesterol (Chol), and N-succini midyl-S-acetylthioacetate (SATA) were obtained from Sigma, St. Louis, MO, USA. Maleimido-4-(p-phenylbutyryl) phosphatidylethanolamine) (MPB-PE) was pu rch ased from Avanti Polar lipids, Birmingham, AL, USA. Dulbecco’s modified eagle medium, penicillin and streptomycin were obtained from GIBCO, Grand Island, NY, USA. MTT was from Boehringer Mannheim, Germany. Fetal calf serum was a product of Hyclone, USA. Other reagents were of analytical grade.

Methods

Lactosamination of human serum albumin Lactosamination was performed according to Schwartz & Gray[16]; human se rum albumin (HSA) 500 mg was allowed to react at 37 °C with 1.0 g α-lactose and 1.0 g NaBH3CN in 20 mL 0.05 mol/L potassium phosphate buffer solution (PB S), pH 8.0. The reaction was continued for 144 h. And the mixture was dialysed against double distilled water. The lactose content of lactosaminated human serum albumin (L-HSA) was determined by the phenol-sulphuric acid method of Dubois et al[17].

Liposomes preparation PC, Chol, and MPB-PE were mixed in a molar ratio of 23:16:1 and dissolved in ether, dried by a rotary evaporator, dissolved in HN-buffer (10 mmol/L HEPES, 135 mmol/L NaCl, pH 6.7) and lyophilized. Phospholipid phosphorus of liposome preparation was measured by phosphate assay after perchloric acid destruction. Conventional liposomes were prepared with PC and Chol in a molar ratio of 23:16.

Coupling of L-HSA to MPB-PE containing liposomes (MPB-liposomes) L-HSA was coupled to MPB-liposomesbyasulfhydryl -maleimidecouplingtechni que according to Derksen et al[18]. Using SATA as a heterobifunctional reagent, free sulfhydryl groups were introduced in L-HSA[19]. After separation of SATA from the protein by Sephadex G-25 gel permeation chromatog raphy, the acetylthioacetate-L-HSA was deacetylated by a freshly prepared solution of 0.5 mol/L hydroxylamine-HCl, 0.5 mol/L HEPES, 25 mmol/L EDTA, pH 7.0. After deacetylation, the thioacetyl-L-HSA was allowed to react with the MPB-PE containing liposomes for 4 h at room temperature, in a ratio of 3 mg of prote in per mg liposomes. N-ethylmaleimide was added to cap un-reacted sulfhydryl groups. Liposomes were separated from unconjugated protein by Sepharose 4B infiltrat i on. The L-HSA- liposomes conjugates were characterized by determining protein a nd phospholipid phosphorus content, lyophilized, and stored at 4 °C.

Tissue distribution of L-HSA-liposomes L-HSA-liposomes were labled with ¹³¹I by Iodogen’s method. Male Sprague-Dawley rats weighing 200-250 g were anaesthetized by intraperitoneal injection of 20-25 mg sodium pentobarbital. Radiolabled liposomes were injected via the penile vein and the abdomen was opened. At the end of the experiment, liver lobules and other tissues were removed and weighed. Radioactivity was determined by γ-counter. And the results were registered as cpm per gram tissue.

Inhibitory effect of L-HSA-liposomes encapsulating interferon-α on HBV replication To assess the effects of L-HSA- liposomes en c apsulating interferon-α on HBV replication, 2.2.15 cells were plated at a dens ity of 2 × 10⁵ per 17 mm culture dish and pre-incubated in Dulbecco’s modified eagle medium containing 10% fetal calf serum for 24 h. After being washed with phosphate-buffered saline, they were cultured at 37 °C for 12 d, with fresh medium changed every 3 d supplemented with L-HSA-liposomes encapsulating interferon-α (LL-IFN), conventional liposomes encapsulating interferon-α (CL-IFN), and free interferon-α (IFN) at an appropriate concentr ation. The culture medium was collected every 3 days and the cells were used in subsequent experiments. The viability of cells was examined spectrophotometrically by the MTT methods, and the cytotoxicity of the compounds was also monitored by the MTT assay.

RESULTS

Synthesis and characterization of L-HSA-liposomes

Having reacted with lactose, the HSA and lactose mixture was dialysed against water for 3 days. There was no lactose in the final dialysis solution, suggesting the L-HSA had been purified completely. Each HSA molecule was modified with about 17 molecules of lactose. When L-HSA was coupled to MPB-liposomes, the amount of L-HSA that could be coupled to 1 μmol MPB-liposomes was 538.7 μg. A number of about 107000 L-HSA molecules per liposome particle were calculated, assuming that the molecular weight of L-HSA was 7600 and the average diameter of MPB-liposomes was 400 nm. The liposomes conjugate was stored at 4 °C for at least 8 weeks and filtered by Sepharose 4B. The infiltration figure showed as a single absorbing peak, suggesting that the liposomal conjugate was stable at 4 °C.

Tissue distribution of L-HSA-liposomes

40 min after i.v. injection of ¹³¹I labled L-HSA-liposomes, r adioactivity in liver was higher than the other organs (P≤ 7.84 × 10^-6) (Table 1). Spleen uptake of L-HSA-liposomes was less than half of that of liver, and the uptake by kidney, heart, and lung was 1/4 to 1/9.

Table 1 Tissue distribution of ¹³¹I-labeled L-HSA-liposomes in rat.

Rat No.	Tissue distribution in vivo (cpm per gram tissue)
	Liver	Spleena	Kidneyb	Heartc	Lungd
1	31777.2	15394.2	7912.1	4427.5	9568.5
2	30476.1	15464.1	5424.5	3180.8	6044.1
3	30013.1	14041.4	7932.3	3777.5	5886.1
4	36099.3	10817.5	6372.3	3722.7	8717.3
5	36479.8	17882.6	7228.6	3885.5	7695.8
Average	32969.1	14719.9	6974.1	3798.8	7582.4

¹³¹I labeled L-HSA-liposomes were injected into rats. Radioactivity in different tissues was determined 40 min after injection as described in Materials and Methods. Compared with liver:

^aP = 7.84 × 10^-6;

^bP = 1.06 × 10^-7;

^cP = 2.98 × 10^-8;

^dP = 2.1 × 10^-7

When L-HSA was pre-injected, the liver uptake decreased significantly, showing no statistic difference with spleen (P = 0.38) (Table 2). However, ¹³¹I labled L-HSA-liposomes uptake was also observed in the other organs.

Table 2 Tissue distribution of ¹³¹I labeled L-HSA-liposomes in rat, after L-HSA pre-injection.

Rat No.	Tissue distribution in vivo (cpm per gram tissue)
	Liver	Spleena	Kidneyb	Heartc	Lungd
1	34682.0	35187.6	5535.7	2474.2	6368.0
2	20587.4	19791.1	4541.8	2277.2	8507.7
3	24288.8	27297.1	4564.4	2089.5	7280.7
4	33733.1	17637.5	5560.3	2698.3	12561.1
5	25600.9	18994.5	6999.2	3227.0	11087.1
Average	27778.4	23781.6	7699.1	2553.2	8679.4

Fifteen min after L-HAS pre-injection, ¹³¹I labeled L-HSA-liposomes were injected into rats. Radioactivity in different tissues was determined 40 min af ter injection as described in Materials and methods. Compared with liver:

^aP = 0.38;

^bP = 4.35 × 10^-5;

^cP = 1.66 × 10^-5;

^dP = 2.52 × 10^-4

Anti-HBV activity of L-HSA-liposomes encap-sulating interferon-α

Table 3 shows the antiviral activity of L-HSA-liposomes encapsulating interferon-α tested against HBV in vitro. The liposomes and L-HSA-l iposomes did not show any anti- HBV effect. However, when entraped in L-HSA-liposomes, int erferon was 3 to 5 times more effective against HBV replication in 2.2.15 cells than CL-IFN and IFN (P value was 0.015 and 0.003, respectively). To achieve similar antiviral effects, the dose of LL-IFN was only 1/4, 1/8 of CL-IFN and IFN, respectively. The treatments did not appear to be cytotoxic to 2.2.15 cell s at the concentrations used in the anti-HBV assay.

Table 3 Inhibitory effect on HBeAg expression in 2.2.15 cells of di fferent types of interferons.

Treatment	Inhibitory effect on HBeAg expression (%)
	3 days	6 days	9 days	12 days
Liposomes	0	1.8	0	0
L-HSA-Liposomes	0	0	0	0
IFN 0.08 MUa	0	4.7	12.7	2.3
CL-IFN 0.08 MUb	0	0	22.8	5.8
LL-IFN 0.08 MU	44.8	60.7	43.5	45.5
LL-IFN 0.04 MU	6.8	47.4	28.6	12.1
LL-IFN 0.02 MU	0	24.6	23.8	4.6
LL-IFN 0.01 MU	0	4.8	15.4	5.2

IFN: interferon; CL-IFN: interferon entraped in conventional liposomes; LL-IFN: interferon entraped in L-HSA-liposomes. Compared with LL-IFN 0.08 MU:

^aP = 0.003,

^bP = 0.015

DISCUSSION

Worldwide, HBV infection is the main cause of chronic liver disease. HBV carrie r s are at risk for chronic hepatitis, cirrhosis, and hepatocyte carcinoma. Previous studies have indicated the correlation of HBVX gene to hepatocyte carcinoma[20]. An effective treatment for HBV infection is therefore urgently needed. Interferon-α, nucleotide analogy, Chinese medicinal herbs, and other i mmunological approaches have shown promising results in a subset of patients treated for prolonged periods[21-24]. But overall response rates have been unsatisfactory. Selective drug delivery to the parenchymal liver cell by drug conjugation to a carrier would improve curative effect, reduce therapeutic dosage and side effects of drugs. In this report we describe the preparation of a liposomal drug carrier system, which could target the therapeutic drug to liver via the asialoglycoprotein receptor. Lactose was coupled to human serum albumin by reductive lactosamination in the presence of sodium cyanoborohydride. Then L-HSA was covalently coupled with liposomes using the heterobifuntional reagent N-Succinimidyl-S-acetylthioacetate. This is a well established method for coupli ng proteins to liposomes, based on the reaction of thiolated proteins and liposo mal maleimido-4-(p-phenylbutyryl) phosphatidylethanolamine[10]. T hio lation of L-HSA, in which about 17 of the free ε-amino groups of lysine were derivatized with lactose, still allowed the introduction of several sulfhydryl molecules in this protein. This indicates that SATA is suitable for coupling deri vatized protein to liposomes.

Mammalian liver contains a unique asialoglycoprotein receptor responsible for the rapid serum clearance and lysosomal catabolism of desialylated glycoproteins bearing terminal, non-reducing galactose residues. Previous studies demonst rated that, for bovine serum albumin, at least 13 lactosyl groups were needed for high affinity recognition by the ASGP-R[25]. In our study, about 107000 molecules L-HSA perliposome were coupled, and each HSA molecule was modifi ed with 17 lactose molecules, hence the liposomal conjugate had a high affinity to ASGP-R. The in vivo study indicated that the liver uptake of liposomal conjugate was higher than other organs, which was about 2 times higher than that of spleen, and 4 to 9 times higher than kidney, heart, and lung. L-HSA pre-injection almost prevented the selective liver delivery of liposomal conjugate, suggesting that the galactose-specific nature of hepatic uptake. Modified with 17 lactose molecules, L-HSA was capable of inhibiting the hepatic uptake of liposomal conjugate, owing to the high affinity for the hepatic galactose-recognizin greceptor. Another study also showed that liposomes modified with several galatose residues had a similar liver targeting[26].

2.2.15 cell is a stable expression system of transfetced HBV DNA, which could produce HBV particle, and is often taken as a cell model for screening anti-hepatitis B virus drugs[27,28]. Interferon had the inhibitory effect on HBV replication in this system[29]. Wu et al[30] proved that this system contained ASGP-R. In vitro, interferon-α entraped in liposomal conjugate had a higher activity against HBV free than interferon and than interf eron entrapped in conventional liposomes. To achieve similar antivirus effects, the doses of LL-IFN capable of inhibiting virus growth were 4 to 8 times less than that of CL-IFN and IFN.

In conclusion, the results obtained in this study indicated that the conjugate of lactosaminated human serum albumin and liposomes achieved good liver targeting, and allowed the development of a potent liver-targeting drug carrier system.