Radiolabeled Constructs for Evaluation of the Asialoglycoprotein

Bioconjugate Chem. 2003, 14, 997−1006

997

Radiolabeled Constructs for Evaluation of the Asialoglycoprotein Receptor Status and Hepatic Functional Reserves Michio Abe,† Jing Lai,† Zbigniew P. Kortylewicz,† Hideo Nagata,‡ Ira J. Fox,‡ Charles A. Enke,† and Janina Baranowska-Kortylewicz*,† University of Nebraska Medical Center, Departments of Radiation Oncology and Surgery, J. Bruce Henriksen Laboratories for Cancer Research, Omaha, Nebraska 68198. Received May 14, 2003; Revised Manuscript Received August 13, 2003

Transplantation of isolated hepatocytes may eventually replace a whole liver transplantation for the treatment of selected liver metabolic disorders and acute hepatic failure. To understand the behavior of transplanted hepatocytes, methods for longitudinal assessment of functional activity and survival of hepatocyte transplants must be developed. Targeting of asialoglycoprotein receptor (ASGPr) with various radiolabeled or Gd-labeled constructs of asialofetuin (AF) is expected to allow noninvasive and quantitative assessments of the ASGPr status in functioning hepatocytes before and after the transplant. Six new constructs of 125I-, 99mTc-, 153Gd-, and 111In-radiolabeled AF with distinct stabilities and clearance rates were prepared and evaluated in vitro in mice, rat, porcine, and human hepatocytes, and in vivo in mice and rats. The blood and organ clearance rates, as well as liver and spleen uptake, were measured. Even extensive chemical modifications of AF with poly-L-lysine and various chelating agents do not appear to diminish AF’s binding to ASGPr. Binding to isolated hepatocytes and the in vivo liver uptake studies indicate unimpaired functional activity of AF as evidenced by the rapid (88%. The purity of the hepatocyte preparation was judged to be over 98% by immunofluorescence for class II bearing nonparenchymal cells. Isolated hepatocytes were suspended in ViaSpan and shipped to the University of Nebraska Medical Center by overnight delivery. Conjugation of DTPA with p-Lys [p-Lys-(DTPA)20-30] and Asialofetuin [AF-(DTPA)7]. All conjugates were prepared using methods described previously for modification of antibodies (17). Briefly, when SCN-DTPA was used, 0.1 mM solution of p-Lys hydrobromide (MW 53 900) in 0.05 M sodium carbonate buffer, pH 8.3, was reacted with 5 mM aqueous solution of SCNDTPA at 1:300 molar ratio. The mixture was allowed to react for 2 h at room temperature. An aliquot (0.01 mL) of a crude reaction mixture was reserved for determination of the substitution ratio. The remaining mixture was transferred into an ultrafiltration device (Centricon-30) and unreacted SCN-DTPA is removed during four buffer exchanges with 1 mL of 0.05 M phosphate buffered saline, pH 7.2 (PBS). During the final buffer exchange with 0.1 M PBS, pH 7.5, the volume of the sample was reduced to achieve a concentration of p-Lys of approximately 0.2 mM. The p-Lys-DTPA conjugate was transferred into metal-free vials and stored frozen at -20 °C until ready to use. Typically 20-30 DTPA residues were associated with each molecule of p-Lys as determined using the instant thin-layer chromatography (ITLC) method (17). Reactions with DTPA dianhydride were conducted as described above with the following modification: dianhydride was dissolved in anhydrous dimethyl sulfoxide (DMSO) and used immediately. Unbound DTPA was then removed by gel filtration on Sephadex G-50 column using 0.1 M sodium phosphate buffer containing 0.1 M sodium chloride, pH 7.5. Fractions containing p-Lys-DTPA conjugate were combined and concentrated using Centricon-30. Identical methods were employed in the preparation of AF-(DTPA)7. Typically seven DTPA residues were incorporated into each molecule of AF when the reaction was conducted at 20:1 molar ratio. Determination of the DTPA-Protein Substitution Ratio (17). To an aliquot of the reaction mixture containing protein, its SCN-DTPA conjugate, and unreacted (S)-1-p-isothiocyanatobenzyl-DTPA, 1 mol equiv of carrier-added 111InCl3 in 0.15 M acetate buffer, pH 5.3, is added. The mixture is allowed to react for 45-60 min, and a sample is withdrawn for ITLC analysis as described below. The substitution ratio is calculated using the following formula: number of DTPA molecules conjugated to one molecule asialofetuin ) fraction indium bound (radioactivity associated with the origin) multiplied by the total molar equivalents of DTPA used in the conjugation reaction. The remaining reaction mixture is treated with 1000 mol equiv of diethylenetriaminepentaacetic acid, incubated for additional 5 min and analyzed

Radiolabeled Constructs for ASGP Receptors

on ITLC plates to determine any nonspecific retention of indium at the plate’s origin. Modification of Asialofetuin and p-Lys(DTPA)20-30 with 3-(2-Pyridyldithio)propionate. AF and p-Lys are referred below as proteins. Methods used for their modification were identical. N-Succinimidyl 3-(2pyridyldithio)propionate was dissolved in 95% ethanol and slowly added to proteins in 0.1 M PBS, pH 7.5, at molar ratios of 4:1 for AF and 10:1 for p-Lys-(DTPA)20-30. Reaction mixtures were left for approximately 30 min at room temperature. Unbound SPDP was removed by gel filtration on Sephadex G-50 column. Fractions containing protein were combined, and 0.1 mL aliquot was used to determine the number of 2-pyridyl disulfide residues. One milliliter of PBS was added to the SPDP-protein conjugate, and absorbance at 343 nm was recorded (A1). Dithiothreitol was added directly into the cuvette containing SPDP conjugates to give a final concentration of 25 mM. The reduction was allowed to proceed for 15 min, and the absorbance at 343 nm was recorded (A2). The molar extinction coefficient of 8080 M-1 cm-1 was used to calculate the SPDP substitution ratio as follows: substitution ratio ) concentration of 2-thiopyridone released on reduction [(A2 - A1)/8080] divided by the concentration of protein. Modification of Asialofetuin with N-Sulfosuccinimidyl 4-(p-Maleimidophenyl)butyrate. AF was reacted with N-sulfosuccinimidyl 4-(p-maleimidophenyl)butyrate in 20 mM sodium phosphate buffer containing 0.15 M sodium chloride, pH 7.0, at a molar ratio of 1:10 AF to SMPB. The mixture was allowed to react for 30 min at room temperature. Unbound SMPB was removed on a Sephadex G-50 column. Modification of p-Lys-(DTPA)20-30 with the Traut’s Reagent. Traut’s reagent (2-iminothiolane) in 0.16 M borate buffer, pH 8.0, was added to p-Lys-(DTPA)20-30 also in 0.16 M borate buffer, pH 8.0, at a molar ratio of 20:1. The mixture was reacted for 45 min at room temperature under nitrogen. Iminothiolated protein was separated from the excess 2-iminothiolane on a Sephadex G-50 column equilibrated and eluted with PBS-EDTA. Preparation of 3-Mercatopropionate-Modified Asialofetuin and p-Lys-(DTPA)20-30. SPDP conjugates prepared above were transferred into Centricon30, and the buffer was exchanged for 0.1 M sodium acetate buffer containing 0.1 M sodium chloride, pH 4.5. DTT dissolved in water was added to a final concentration of 25 mM to liberate free thiol groups. After 30-min reduction, thiolated proteins were purified on a Sephadex G-50 column eluted with 0.1 M PBS, pH 7.5, and used immediately in the preparation of AF-p-Lys conjugates as described below. Conjugation of 3-Mercatopropionate-Modified Asialofetuin with 3-(2-Pyridyldithio)propionateModified p-Lys-(DTPA)20-30 [AF-SPDP-SPDP-pLys-(DTPA)20-30]. 3-Mercatopropionate-modified AF in 0.1 M PBS, pH 7.5, was added to 3-(2-pyridyldithio)propionate-modified p-Lys-(DTPA)20-30 in the same buffer at a molar ratio of 1:1. The reaction progress was followed by measuring 2-thiopyridone released in a thiol-disulfide exchange reaction. After approximately 18 h at 4 °C, the absorbance at 343 nm reached a plateau indicating that the conjugation was complete. AF-SPDP-SPDP-p-Lys(DTPA)20-30 conjugate was purified using a Centricon100 at 1000g with three changes of PBS. Conjugation of Maleimide-Activated Asialofetuin with 3-Mercatopropionate-Modified p-Lys(DTPA)20-30 [AF-SMPB-SPDP-p-Lys-(DTPA)20-30]. Maleimide-activated asialofetuin in 20 mM sodium phos-

Bioconjugate Chem., Vol. 14, No. 5, 2003 999

phate buffer containing 0.15 M sodium chloride, pH 7.0, was added to freshly prepared 3-mercatopropionatemodified p-Lys-(DTPA)20-30 in 0.1 M PBS, pH 7.5 at a molar ratio of 1:1. The reaction mixture was kept for 18 h at 4 °C. AF-SMPB-SPDP-p-Lys-(DTPA)20-30 conjugate was separated from the unconjugated modified proteins using a Centricon-100 at 1000g with three buffer exchanges with 0.1 M sodium acetate, 0.1 M sodium chloride, pH 4.5. An additional purification step included the removal of potential disulfide-linked p-Lys aggregates (usually 95% after a 45-min reaction. The radiolabeled product was isolated on a BioRad desalting column using PBS, pH 7.2, as the eluant with an average yield of recovery >70%. This radiolabeling protocol produces AF-(99mTc-SHNH)6 with a specific activity of about 2-2.5 µCi 99mTc/µg construct. Radioiodination of Asialofetuin. The iodogen method (19) was used to prepare 125I-AF. AF in PBS, pH 7.2 was transferred into a glass tube coated with 0.1 mg iodogen. Sodium [125I]iodide was added (1 mCi/0.1 mg AF), and the mixture was reacted for 15 min. The progress of radioiodination was monitored using ITLC. The purification was accomplished on a Bio-Rad desalting column (Hercules, CA). Quality Controls. HPLC Analysis of Radiolabeled Conjugates. Size-exclusion HPLC analysis with a Bio-sil SEC-250 column (300 × 7.8 mm) with a Bio-sil SEC guard column (10 × 7.8 mm) was used to determine the molecular weight and purity of the radiolabeled conjugates. A 0.05-mL sample of radiolabeled conjugate was injected into the column. The column was eluted with a buffer containing 50 mM NaH2PO4, 50 mM Na2HPO4, 100 mM NaCl, and 0.05% NaN3, pH 6.7, at a flow rate of 0.5 mL/min. The conjugates were detected using a dual detection: the absorbance at 280 nm and the radioactivity. Instant Thin Layer Chromatography (ITLC). Radiolabeled conjugates were also analyzed on ITLC plates using 0.9% saline or methanol/water (1:4; v/v) as developing solvents for 125I-labeled constructs, or 0.15 M sodium acetate, pH 5.3, for 99mTc- and 111In-labeled constructs. In all solvent systems, the radiolabeled protein construct remains at the origin while free radioisotopes move with the solvent front. The distribution of radioactivity on ITLC strips was monitored using a radiochromatography scanner (Vista-100, Packard Instruments). Stability of Constructs in Liver Homogenates. Swiss Webster mice were anesthetized with CO2 and sacrificed by cervical dislocation. Liver (1.2 g) was homogenized in either 4.0 mL of 0.2 M acetate buffer, pH 5.0, or 4.0 mL of PBS, pH 7.2. Crude homogenates were sonicated on ice at 400 W for 1 min. The radiolabeled conjugate (0.025 mCi) was added to 1 mL of liver homogenate, and 0.1-mL portions were transferred into microcentrifuge tubes and incubated in a tissue culture incubator (5% CO2) at 37 °C for 5, 15, 30 min, and 1, 2, and 24 h. At designated time points, liver homogenates were centrifuged at 1000g for 10 min, and the supernatant was analyzed for the presence of a low and high molecular weight soluble degradation products using ITLC. Total radioactivity associated with the pellet and the supernatant was determined in a gamma counter. The supernatants collected after 2 h incubation were also analyzed on size the exclusion HPLC columns.

Abe et al.

In Vitro Binding and Degradation of 125I-AF. Isolated hepatocytes (5 × 105) were suspended in 1 mL of Delbecco’s Minimum Essential Medium (DMEM) supplemented with 10% heat inactivated fetal bovine serum (FBS). Twenty nanograms of 125I-AF was added to each tube and cells were incubated at 37 °C on a multipurpose shaker rotating at 6 rpm. At designated time points, hepatocytes were centrifuged at 3000 rpm for 5 min, and supernatants were aspirated. Cell pellets were washed twice with 1 mL of ice-cold PBS. The cellassociated radioactivity was measured in a gamma counter. The nonspecific binding was determined in the presence of a 20 000-fold (0.4 mg) excess of native, unlabeled AF. For metabolic studies, the rate of degradation was measured as the increase in the acid-soluble radioactivity accumulating in the culture medium in the presence of hepatocytes. 125I-AF-containing cells, washed with ice-cold PBS, were resuspended in a prewarmed to 37 °C DMEM supplemented with 10% heat inactivated FBS. To the cell suspension sampled immediately after mixing (time 0) and at designated time points, an equal volume of ice-cold 20% trichloroacetic acid was added. After 30 min on ice, the precipitated proteins were separated from the acid-soluble fraction by centrifugation at 4000 rpm for 10 min. The supernatant- and pelletassociated radioactivity was measured in a gamma counter. Biodistribution. Female, six-week-old Swiss Webster mice (Charles River Laboratories, Wilmington, MA) were used after one-week acclimation in the UNMC facilities. Radiolabeled conjugates with the specific activities g1 mCi/mg AF were administered via the tail vein at a dose of 2 µCi/mouse (111In-labeled agents) or 5 µCi/mouse (125IAF). Mice were randomized into groups of four per time point. At 5, 15, 30, 60, 120, 180, and 1440 min after injection, mice were killed and the necropsy was performed. Tissues and blood were weighed, and their radioactive content was determined in a gamma counter. Tissue- and blood-associated radioactivity was normalized to the injected dose and the weight of the sample and expressed as percent of the injected dose per gram of tissue (%ID/g). Localization indices were calculated as tissue-to-blood ratios. Tissue and blood clearance curves were analyzed using a numerical module of a SAMM II program (SAMM Institute, University of Washington). Gamma Camera Imaging. Scintigraphic images were acquired using a Technicare pinhole gamma camera in a 128 × 128 matrix using a 10-min scan for each time point. The dose of 0.04 mCi of 111In-labeled agents was administered iv via a tail vein. Mice were kept under a light ketamine/xylazine anesthesia during the image acquisition. RESULTS

Chemistry. Two types of AF constructs were prepared: derivatives that incorporate radiolabel directly into the AF molecule and derivatives that carry additional substituents for indirect radiolabeling (Schemes 1 and 2). In directly modified AF derivatives the radiolabeling was accomplished by either radioiodination of AF with radioiodine or through an appropriate chelating group reacted with lysine residues of AF for radiometal labeling, DTPA for 111In and 153Gd, and SNHN for 99mTc. The indirectly modified AF derivatives were radiolabeled with 111In and 153Gd through the p-Lys-(DTPA)20-30 portion of the AF constructs. Chemical structures of all tested AF constructs are shown in Schemes 1 and 2. For the purpose of indirect radiolabeling with 111In several methods of the conjugate



Scheme 1. Directly Radiolabeled Asialofetuin Derivatives

Scheme 2. Indirectly Radiolabeled Constructs of Asialofetuin with p-Lys

preparation were explored. The formation of disulfide linkages was accomplished via AF-conjugated 3-mercatopropionate displacement of 2-thiopyridine from the 3-(2pyridyldithio)propionate portion of p-Lys conjugate. This approach was also reversed, i.e., 3-mercatopropionate generated from SPDP on p-Lys was used to displace 2-thiopyridine on SPDP-modified AF. The use of Nsuccinimidyl 4-(4′-maleimidophenyl)butyrate (SMPB)modified AF allowed preparation of two constructs using the reaction of the maleimide double bond with the thiol group either generated from SPDP on p-Lys or introduced as 2-iminothiolane (Traut’s reagent) on p-Lys. The synthesis of (S)-1-p-isothiocyanatobenzyl-diethylenetriaminepentaacetic acid (SCN-DTPA) was accomplished according to the method of Corson and Mears (14). The chemical modification of AF was done using either SCN-DTPA or DTPA cyclic anhydride at the protein-to-ligand ratio of 1:20. The typical degree of substitution, as determined through the carrier-added labeling of the crude reaction mixture, was found to be seven DTPA residues per one molecule of AF. The

radiochemical yield of labeling with 111InCl3 and 153GdCl3 was routinely >90%. The radioiodinated AF was prepared using the iodogen method (19) in nearly 100% radiochemical yield. For labeling with 99mTc, AF was modified with N-succinimidyl 6-hydrazinonicotinate hydrochloride (13). On average a 6:1 molar ratio of SHNH to AF was obtained. Radiochemical yield of AF(99mTc-SHNH)6 constructs was routinely >70% (corrected for decay). All constructs were purified on a size exclusion column. AF constructs with p-Lys (MW 53 900) were modified with SCN-DTPA at a substitution ratio as high as 70 DTPA residues per p-Lys molecular without any deleterious effects on the biological activity of AF. However, for studies described here, all AF constructs were prepared to contain on an average 20-30 DTPA groups to facilitate comparison of their biological properties. Radiochemical yields for 111In labeling of p-Lys-(DTPA)20-30 constructs with AF were typically >90%. All purifications were done on a size exclusion column much like directly radiolabeled AF. Additionally, p-Lys-AF conjugates were subjected

1002 Bioconjugate Chem., Vol. 14, No. 5, 2003

Abe et al.

Table 1. Expression of Asialoglycoprotein Receptors in Hepatocytes from Various Species Measured at 37 °C Using 125I-AF no. of ASGPr per cell average (standard deviation)

species

2.44 × 105 (0.22 × 105) 4.08 × 105 (0.14 × 105) 4.00 × 105 (0.10 × 105) 2.1 × 105 (0.5 × 105)b 7 × 104 c

mouse rat pig rata a

Reference 20.

b

Fresh hepatocytes. c Incubated hepatocytes.

to an ultrafiltration step to remove any residual unreacted p-Lys that may have been collected with AF constructs from the size-exclusion column. In the synthesis of maleimide-based AF constructs, which requires generation of free thiols and a prolonged conjugation step, protein-containing fractions collected from the sizeexclusion column went through an additional purification step after treatment with 25 mM DTT to reduce any disulfide bonds that may have contributed to the formation of p-Lys dimers and other aggregates. In Vitro Studies. Metabolic processing and binding studies in vitro were conducted to assess biological behavior of chemically modified AF and to determine differences, if any, in the recognition of the ASGP receptor. The rate of metabolic processing of each AF construct was evaluated. The stability of radiolabeled AF constructs was tested in mouse liver homogenates and in the presence of intact hepatocytes derived from mice, rat, and swine livers. Isolated fresh hepatocytes were also used to determine binding properties of AF constructs, the expression of ASGPr and the intracellular uptake of AF constructs. Hepatocytes were isolated using a perfusion technique of Barry and Friend (15) as modified by Seglen (16). Mice and rat hepatocytes were prepared on site (UNMC) and used within a couple of hours of preparation. The viability of mice hepatocytes was routinely >95%. Porcine hepatocytes were isolated and suspended in ViaSpan for shipment to UNMC by the overnight delivery. Viability as assessed by trypan blue staining was routinely >88%. Table 1 summarizes ASGP receptor recognition by AF-(99mTc-SHNH)6 in cells derived from mouse, rat, and pig livers. The measured number of ASGPr per cells is in good agreement with the published data (20; 2.10 × 105 ( 0.50 × 105 for incubated hepatocytes and 7 × 104 for fresh hepatocytes). The stability and binding of AF constructs in liver homogenates proved to be a good predictor of the in vivo properties. This point is illustrated for AF-SPDP-SPDPp-Lys-(111In-DTPA)20-30 and AF-SMPB-SPDP-p-Lys(111In-DTPA)20-30 in Figures 1A and 1B and later in Figure 7. Fresh liver homogenates diluted with PBS at 1-to-4 ratio (v/v) were incubated with radiolabeled 111In constructs. At designated time points, distribution of radioactivity associated with pellets and supernatants was determined (Figure 1A). In both cases there is a binding of AF constructs to ASGPr in liver homogenate as indicated by disappearance of 111In from the soluble fraction. In the case of AF-SPDP-SPDP-p-Lys-(111InDTPA)20-30, a large fraction of bound radioactivity returns into the supernatant within 30 min. Further analyses of supernatants using ITLC indicated that the majority of 111 In was still associated with larger molecular weight species that were not eluted from the origin. Sizeexclusion HPLC revealed that 111In recovered in supernatants from liver homogenates incubated with AFSMPB-SPDP-p-Lys-(111In-DTPA)20-30 for 120 min was in the form of the intact construct whereas incubation of

Figure 1. (A) Stability of AF-SPDP-SPDP-p-Lys-(111InDTPA)20-30 (O) and AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30 (b) in mouse liver homogenates diluted with PBS. (B) Radioactive HPLC profile of mouse liver homogenate supernatants collected after 120 min incubation with AF-SMPB-SPDP-pLys-(111In-DTPA)20-30 (2) and AF-SPDP-SPDP-p-Lys-(111InDTPA)20-30 (4). Peaks with the retention time (tR) of 17 min correspond to the intact construct (MW approximately 110 000); a peak at tR of 30 min corresponds to low molecular weight (1000 < MW < 2500 Da) degradation products.

Figure 2. Stability of 125I-AF (20 ng/5 × 105 cells) incubated for 60 min in the presence of freshly prepared hepatocytes from mice and pigs (black bars); general stability of 125I-AF was evaluated under identical set of conditions in the absence of hepatocytes (white bars). The gray bar represents 125I-AF incubated with mice hepatocytes in the presence of 20 000-fold excess unlabeled AF sufficient to block all AF binding sites.

AF-SPDP-SPDP-p-Lys-(111In-DTPA)20-30 for 120 min resulted in substantial degradation of the construct (Figure 1B). The in vivo fate of these constructs parallels in vitro findings (see below). Processing of 125I-AF in mice and porcine hepatocytes was rapid (Figures 2 and 3). 125I-AF purified on a sizeexclusion column was incubated in the presence of freshly prepared hepatocytes from mice and 24-h-old porcine hepatocytes. Control samples consisted of 125I-AF incubated under identical set of conditions in the absence of hepatocytes. A second control consisted of hepatocytes treated with the 20 000-fold excess of nonradioactive AF (0.4 mg/5 × 105 cells). This amount is sufficient to block AF binding sites. It is apparent that isolated hepatocytes are able to process 125I-AF in a receptor-dependent manner. The lesser metabolic activity observed with porcine hepatocytes is probably related to the age of these cells (prepared 24-36 h prior to use). The excess native AF inhibited binding of 125I-AF to porcine hepatocytes in the same way as in the case of mice hepatocytes (Figure 3). A similar result was obtained with freshly isolated rat hepatocytes, but immortalized rat hepatocytes and immortalized human hepatocytes bound neither 125I-AF nor (99mTc-SHNH)-AF indicating the loss of ASGPr in these cells (data not shown). The average number of AF binding sites is similar in hepatocytes from all analyzed species (Table 1). Competition binding studies conducted with 111In- and



Figure 3. Binding and metabolic processing of 125I-AF in porcine hepatocytes. (A) Fresh hepatocytes were incubated with 125I-AF (20 ng/5 × 105 cells). (B) The degree of protein degradation was determined in medium collected from hepatocytes incubated with 125I-AF at 37 °C for 60 min (/ P < 0.001). Proteins were precipitated with ice-cold 10% TCA. Table 2. Pharmacokinetics of Hepatic Uptake and Elimination of Radiolabeled Asialofetuin and Its Constructs in Mice construct

absorption t1/2 (min)

postabsorption average t1/2 (std dev) (min)

elimination average t1/2 (std dev) (h)

AF-(99mTc-SHNH)6 AF-(111In-DTPA)7 AF-SPDP-SPDP-Lys-(111In-DTPA)20-30 AF-SMPB-Traut-Lys-(111In-DTPA)20-30 AF-SMPB-SPDP-Lys-(111In-DTPA)20-30

nd nd 6.3 nd 3.8 3.3

5.1 (0.05) nd 29.1 (5.1) 14.4 (7.1) 9.1 (0.1) 7.9 (1.8)

2.8 (0.06) 14.7 (14.7)a 41.0 (0.43) 19.0 (3.52) 44.3 (3.08) 36.0 (9.12)

125I-AF

a

Corrected for decay.

Table 3. Tissue-to-Blood Ratios Derived from Biodistribution Data of Radiolabeled Asialofetuin and Its Constructs in Mice time after administration (min) construct p-Lys-(111In-DTPA)20-30 control AF-(111In-DTPA)7 AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30 AF-SMPB-Traut-p-Lys-(111In-DTPA)20-30 AF-SPDP-SPDP-p-Lys-(111In-DTPA)20-30 125I-AF (99mTc-SHNH)6-AF p-Lys-(111In-DTPA)20-30 control AF-(111In-DTPA)7 AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30 AF-SMPB-Traut-p-Lys-(111In-DTPA)20-30 AF-SPDP-SPDP-p-Lys-(111In-DTPA)20-30 125I-AF (99mTc-SHNH)6-AF p-Lys-(111In-DTPA)20-30 control AF-(111In-DTPA)7 AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30 AF-SMPB-Traut-p-Lys-(111In-DTPA)20-30 AF-SPDP-SPDP-p-Lys-(111In-DTPA)20-30 125I-AF (99mTc-SHNH)6-AF a

5

15

30

A. Liver-to-Blood Ratios 0.4 0.5 0.8 58.0 130.5 257.3 19.6 30.6 40.3 38.5 94.3 134.5 1.19 1.87 2.09 75 nd 3.2 71.6 nd 63.0 B. Kidney-to-Blood Ratios 2.5 6.5 9.4 24.5 69.7 142.9 2.3 5.2 8.22 3.1 8.0 13.6 0.7 2.5 4.4 0.54 nd 0.15 1.12 nd 2.6 C. Spleen-to-Blood Ratios 0.1 0.15 0.2 5.4 13.2 20.4 4.4 4.6 5.7 9.2 19.5 22.4 0.21 0.29 0.37 1.9 nd 0.56 4.4 nd 7.0

60

120

1440

1.0 323.4 50.2 116.5 2.52 3.1 69.4

1.2 718.7 68.3 147.0 5.63 2.4a 54.3

5.6 830.4 194.3 353.8 nd nd nd

19.9 175.8 12.2 14.7 4.7 0.14 4.2

26.3 416.4 17.8 17.6 13.3 0.13a 4.4

203.6 673.9 69.4 54.6 nd nd nd

0.3 28.9 6.8 20.1 0.36 0.56 27.5

0.3 61.4 8.3 19.4 0.57 0.48a 25.2

2.9 74.0 40.1 46.0 nd nd nd

180 min.

99mTc-labeled constructs of AF indicated that binding and cell uptake of these agents are similar to 125I-AF data shown in Figure 3. Modification of AF with p-Lys did not impair binding of 111In-labeled constructs to hepatocytes. However, the catabolic processing of radiometal-labeled constructs in the presence of viable hepatocytes is considerably slower than degradation of 125I-AF. A 60min incubation time sufficient to degrade approximately 20% of 125I-AF results in less than 1% degradation of 111In- or 153Gd-labeled directly labeled AF constructs under identical conditions. In addition to controls containing only buffer and radiolabeled AF constructs, fixed hepatocytes were also used as a control to verify that binding of radiolabeled

AF to its receptor does not contribute to the degradation of this protein. Stability of radiolabeled AF in the presence of fixed hepatocytes was similar to that measured in incubation medium in the absence of cells. While binding of 125I-AF to fresh porcine hepatocytes and to fixed cells was very similar, 95.14 ( 4.47% versus 82.90 ( 4.94, respectively, within the scope and timeline of the experiments the degradation of 125I-AF in the presence of fixed porcine cells was not facilitated by the presence of fixed hepatocytes. The very slow loss of radiolabel was related only to the radiolysis or chemical degradation. Likewise, fresh murine hepatocytes metabolized on average 21.05 ( 0.02% 125I-AF per hour, but the loss of


Figure 4. Liver uptake of 125I-AF in rats and mice at various times after administration. Rats received 0.05-mCi dose of 125IAF; mice received 0.005-mCi dose. At designated times animals were killed and necropsy performed.

Abe et al.

Figure 6. Organ-to-blood ratio in normal mice treated with 0.05 mCi 99mTc-(SHNH)6-AF administered IV in 0.2 mL of normal saline. At indicated time points mice (n ) 5) were killed, and necropsy was performed.

Figure 5. Biodistribution of (99mTc-SHNH)6-AF in mice: b liver; O spleen; 0 kidney; 3 blood. Mice (n ) 5) received 0.05 mCi dose iv via a tail vein in normal saline (data is corrected for decay).

radioactivity from 125I-AF in the presence of fixed mouse hepatocytes was less than 0.5% during the same time. In Vivo Studies. Radioiodinated 125I-AF rapidly localizes in liver wherein it is promptly deiodinated (Figure 4). Analyses of serum samples from mice and rats treated with 125I-AF revealed that at 5 and 10 min postinjection, 46% (n ) 1 rat) and 54.8% ( 3.5% (n ) 4 mice) of the total circulating radioactivity is already in the form of free [125I]-iodide. A similar analysis of the 180-min serum sample, indicated that only 2.95% (n ) 1 rat) and 2.74 ( 0.47% (n ) 4 mice) of circulating 125I was still in the form of intact 125I-AF. The hepatic postabsorption half-life was 5.1 ( 0.05 min (Table 2). Hepatic uptake of 99mTc-(SHNH)6-AF in mice is rapid and highly selective (Figure 5). Five minutes after injection nearly 85% of the injected dose is localized in liver, and this level is maintained up to 3 h after injection. Some uptake of AF-(99mTc-SHNH)6 is also observed in spleen, with 0.83 ( 0.40% injected dose at 30 min and 2.00 ( 0.98% at 180 min after injection. The liver-to-blood ratio and spleen-to-blood ratios are 144.3 ( 18.4 and 28.0 ( 9.3, respectively, 1 h after injection (Figure 6, Tables 3A, 3C). The liver uptake of 111In-labeled AF constructs indicated that all tested reagents specifically targeted hepatocytes and recognized AFGPr irrespective of the chemical modification of AF. However, systemic and liver

Figure 7. Tissue clearance curves for AF-SMPB-SPDP-pLys-(111In-DTPA)20-30 (black symbols, solid lines) and AFSMPB-Traut-p-Lys-(111In-DTPA)20-30 (white symbols, dotted lines) in selected tissues. Data for liver, spleen, and femur were fitted into a biexponential model and are indicated with solid and dotted lines.

clearance rates were strongly influenced by the chemical structure of the conjugates. p-Lys-(DTPA)20-30 constructs labeled with 111In were used as a nonspecific control for all 111In-labeled AF-p-Lys conjugates. Tables 3A, 3B, and 3C show tissue-to-blood ratios for studied constructs. The uptake and in vivo processing of two AF constructs radiolabeled with 111In is shown in Figures 7. There is a favorable clearance from systemic circulation and normal spleen accompanied by a considerable retention of the released radioisotope in liver and kidneys (up to 24 h). Constructs containing SPDP linkages appear to be less stable, and 60 min after administration the radioactivity associated with kidneys is clearly visible in planar images (Figure 9A). The degradation of maleimide-based constructs is slower (Figures 8, 9B, 10). The Traut’s reagentand SPDP-based maleimide constructs have similar degradation rates and pharmacokinetics, but the liver uptake of AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30 is lower and 111In released from this construct appears in



Figure 8. Comparison of distribution of various AF constructs in mice 60 min after intravenous administration.

Figure 9. Planar images in mice acquired 60 min after iv injection (tail) of 40 µCi 111In-labeled (A) AF-SPDP-SPDP-pLys-(111In-DTPA)20-30 and (B) AF-SMPB-SPDP-p-Lys-(111InDTPA)20-30. Images were acquired using a Technicare pinhole gamma camera; acquisition matrix 128 × 128; time of acquisition 600 s.

systemic circulation faster. Liver-to-blood ratios at 30 min after injection are 40 and 135 for AF-SMPB-SPDP-pLys-(111In-DTPA)20-30 and AF-SMPB-Traut-p-Lys(111In-DTPA)20-30, respectively. The kidney-to-blood ratio for AF-SMPB-SPDP-p-Lys(111In-DTPA)20-30 is measured at 30 min after administration at 8 and for AF-SMPB-Traut-p-Lys-(111InDTPA)20-30 at nearly 14. The lack of the uptake of the radiometal released from AF constructs in the bone and/ or bone marrow indicates that the metabolites released from the liver do not contain free metal (Figure 7; femur). In vitro stability studies in liver homogenates and human serum confirmed the absence of free metal in incubated reaction mixtures (Figures 1A, 1B). Hepatic half-lives for all constructs in mice are summarized in Table 2. Three reagents had a measurable absorption phase with a halflife < 6 min. The postabsorption and elimination halflives appeared to be related to the rate of hepatic degradation of a given construct. The shortest half-life was measured for 125I-AF at 5 min. Of the DTPA-modified AF, AF-SPDP-SPDP-p-Lys-(111In-DTPA)20-30 construct was eliminated from liver approximately twice as fast as the other three reagents.

Figure 10. Liver images acquired after iv administration of 40 µCi AF-SMPB-SPDP-p-Lys-(111In-DTPA)20-30. Data were analyzed using a Scion Image software. DISCUSSION

The ability of asialofetuin to target asialoglycoprotein receptor has long been recognized as a convenient tool for the delivery of various vectors such as liposomes, recombinant lipoproteins, and polymers for drug or gene delivery to the liver, especially to hepatocytes (21, 22). The results of our studies indicate that appropriately radiolabeled AF can also be used to monitor the metabolic and functional activity of hepatocytes before and after transplant. The rapid metabolic removal of radioiodine


from AF and its rapid clearance should allow for a simple and noninvasive evaluation of metabolic activity of hepatocytes pre- and posttransplant. Kinetic measurements of 125I plasma levels can be supplemented with a more traditional evaluation of the ASGPr status with either 99mTc- or 111In-AF constructs. It is expected that because of such rapid kinetics of 125I-AF its concentrations in liver and spleen as a function of time postadministration will provide good estimates of activities and function of the spleen-transplanted hepatocytes. The evaluation of tissue-to-blood ratios supports this conclusion. Using the more stable AF constructs labeled with indium-111, the in vivo quantification of ASGPr, the total receptor concentrations per liter of hepatic plasma or per volume of transplanted hepatocytes in the spleen, can be measured. One 111In-labeled AF construct, AF-SMPB-SPDPp-Lys-(111In-DTPA)20-30 appears to have a superior liver targeting and metabolic stability. Images acquired 60 min postinjection indicate that the majority of the radiotracer is localized in liver. This may be a reagent of choice for determination of ASGPr concentration in transplanted hepatocytes. As shown in the images in Figure 10 the hepatocyte uptake of AF-SMPB-SPDP-p-Lys-(111InDTPA)20-30 is virtually 100% with undetectable redistribution into systemic circulation and other organs. Normal spleen is not apparent in the images, and the uptake measured during the necropsy is also insignificant (see spleen-to-blood ratio in Table 2C), indicating that a presence of a viable population of transplanted hepatocytes in the spleen should be detectable in planar images particularly in larger animals with impaired liver function. These data appear to indicate that several of the proposed constructs can also be utilized for functional MR imaging. In preliminary experiments, conjugation ratios of up to 60-70 DTPA residues per one poly-L-lysine molecule were achieved. This number of paramagnetic chelates should be sufficient to bring about measurable T1 relaxation time changes. The in vivo studies of carrieradded 153Gd-DTPA-AF constructs (20 DTPA residues/pLys) in mice indicate that nearly 100% of the injected radioactivity localizes in the liver at 5 min after the intravenous administration (data not shown). Studies to correlate the in vitro ASGPr status and function with the survival of intrasplenic hepatocyte transplants are in progress. ACKNOWLEDGMENT

This research was supported in part by the LB595 grant from the Nebraska Department of Health. LITERATURE CITED (1) Bohnen, N. I., Charron, M., Reyes, J., Rubinstein, W., Strom, S. C., Swanson, D., and Towbin, R. (2000) Use of indium111-labeled hepatocytes to determine the biodistribution of transplanted hepatocytes through portal vein infusion. Clin. Nucl. Med. 25, 447-450. (2) Gupta, S., Lee, C. D., Vemuru, R. P., and Bhargava, K. K. (1994) 111Indium labeling of hepatocytes for analysis of shortterm biodistribution of transplanted cells. Hepatology 19, 750-757. (3) Gupta, S., Yerneni, P. R., Vemuru, R. P., Lee, C. D., Yellin, E. L., and Bhargava, K. K. (1993) Studies on the safety of intrasplenic hepatocyte transplantation: relevance to ex vivo gene therapy and liver repopulation in acute hepatic failure. Hum. Gene. Ther. 4, 249-257. (4) Hawkins, R. A., Hall, T., Gambhir, S. S., Busuttil, R. W., Huang, S. C., Glickman, S., Marciano, D., Brown, R. K., and Phelps, M. E. (1988) Radionuclide evaluation of liver transplants. Semin. Nucl. Med. 18, 199-212.

Abe et al. (5) Sakahara, H., Kiuchi, T., Nishizawa, S., Saga, T., Nakamoto, Y., Sato, N., Higashi, T., Tanaka, K., and Konishi, J. (1999) Asialoglycoprotein receptor scintigraphy in evaluation of auxiliary partial orthotopic liver transplantation. J. Nucl. Med. 40, 1463-1467. (6) Kudo, M., Todo, A., Ikekubo, K., Yamamoto, K., Vera, D. R., and Stadalnik, R. C. (1993) Quantitative assessment of hepatocellular function through in vivo radioreceptor imaging with technetium 99m galactosyl human serum albumin. Hepatology 17, 814-819. (7) Miki, K., Kubota, K., Inoue, Y., Vera, D. R., and Makuuchi, M. (2001) Receptor measurements via Tc-GSA kinetic modeling are proportional to functional hepatocellular mass. J. Nucl. Med. 42, 733-737. (8) Pimstone, N. R., Stadalnik, R. C., Vera, D. R., Hutak, D. P., and Trudeau, W. L. (1994) Evaluation of hepatocellular function by way of receptor-mediated uptake of a technetium99m-labeled asialoglycoprotein analog. Hepatology 20, 917923. (9) Rogers, J. C., and Kornfeld, S. (1971) Hepatic uptake of proteins coupled to fetuin glycopeptide. Biochem. Biophys. Res. Commun. 45, 622-629. (10) Morell, A. G., Irvine, R. A., Sternlieb, I., and Schneiberg, I. H. (1968) Physical and chemical studies on ceruloplasmin: V. Metabolic studies on sialic acid-free ceruloplasmin in vivo. J. Biol. Chem. 243, 155-159. (11) Morell, A. G., Gregoriadis, G., and Scheinberg, I. H. (1971) The role of sialic acid in determining the survival of glycoprotein in the circulation. J. Biol. Chem. 246, 1461-1467. (12) Roseng, L., Tolleshaug, H., and Berg, T. (1992) Uptake, intracellular transport, and degradation of poly(ethylene glycol)-modified asialofetuin in hepatocytes. J. Biol. Chem. 267, 22987-22993. (13) Abrams, M. J., Juweid, M., Tenkate, C. I., Schwartz, D. A., Hauser, M. M., Gaul, F. E., Fuccello, A. J., Rubin, R. H., Strauss, H. W., and Fischman, A. J. (1990) Technetium-99mhuman polyclonal IgG radiolabeled via the hydrazino nicotinamide derivative for imaging focal sites of infection in rats. J. Nucl. Med. 31, 2022-2028. (14) Corson, D. T., and Meares, C. F. (2000) Efficient multigram synthesis of the bifunctional chelating agent (S)-1-p-isothiocyanatobenzyl-diethylenetriaminepentaacetic acid. Bioconjugate Chem. 11, 292-299. (15) Berry, M. N., and Friend, D. S. (1969) High yield preparation of isolate rat liver parenchymal cells: A biochemical and fine structural study. J. Cell Biol. 43, 506-520. (16) Seglen, P. O. (1976) Preparation of isolated rat liver cells. Methods Cell. Biol. 13, 29-83. (17) Baranowska-Kortylewicz, J., Dalrymple, G. V., Quadri, S. M., and Harrison, K. A. (1995) Radiolabeling of antibodies for therapy and diagnosis. Methods Mol. Biol. 51, 423-439. (18) Claessens, R. A., Boerman, O. C., Koenders, E. B., Oyen, W. J., van der Meer, J. W., and Corstens, F. H. (1996) Technetium-99m labeled hydrazinonicotinamido human nonspecific polyclonal immunoglobulin G for detection of infectious foci: a comparison with two other technetium-labeled immunoglobulin preparations. Eur. J. Nucl. Med. 23, 414421. (19) Fraker, P. J., and Speck, J. C., Jr. (1978) Protein and cell membrane iodinations with a sparingly soluble chloroamide, 1,3,4,6-tetrachloro-3a,6a-diphenylglycoluril. Biochem. Biophys. Res. Commun. 80, 849-857. (20) Weigel, P. H., and Oka, J. A. (1983) The surface content of asialoglycoprotein receptors on isolated hepatocytes is reversibly modulated by changes in temperature. J. Biol. Chem. 258, 5089-5102. (21) Wu, J., Nantz, M. H., and Zern, M. A. (2002) Targeting hepatocytes for drug and gene delivery: emerging novel approaches and applications. Front. Biosci. 7d, 717-725. (22) Fiume, L., Di Stefano, G., Busi, C., Mattioli, A., Bonino, F., Torrani-Cerenzia, M., Verme, G., Rapicetta, M., Bertini, M., and Gervasi, G. B. (1997) Liver targeting of antiviral nucleoside analogues through the asialoglycoprotein receptor. J. Viral. Hepat. 4, 363-370.

BC034081A

Radiolabeled Constructs for Evaluation of the Asialoglycoprotein

Recommend Documents