| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Abstracts of Oak Ridge Posters |
1
BioPAL, 10 New Bond St., Worcester, MA 01606
a author for
correspondence: fax 508-852-8118, e-mail EVGroman{at}aol.com
Since the initial studies of receptor-mediated endocytosis on LDL (1)(2) and asialoglycoprotein (ASGP) (3)(4), many internalized ligands and their receptors have been identified and characterized (5). It was soon recognized that these receptors could be exploited for pharmaceutical and medical imaging applications (6). In the case of the ASGP receptor (ASGP-R), delivery of diagnostic or therapeutic agents to hepatocytes was achieved by attachment of the agent to a variety of macromolecules possessing the terminal galactose signal necessary to bind the ASGP-R (6). These moieties include radioactive isotopes (7), drugs (8), and DNA (9).
Considering the effort devoted to developing drug delivery systems utilizing the ASGP-R, surprisingly little effort has been devoted to developing simple, easy-to-use assays to measure this receptor activity in vivo. Much of the technology associated with receptor-based diagnostic agents has arisen in fields associated with magnetic resonance and single photon emission computed tomography imaging. One class of magnetic resonance compounds includes iron chelates, which appear to be absorbed by hepatocytes by an unknown biochemical mechanism (10). Saini et al. (11) and others have demonstrated the promise of ferrite particles as a magnetic resonance contrast agent for liver imaging. These commercially available micron-sized particles are taken up by the Kupffer cells of liver and do not measure ASGP-R activity.
Sato et al. (12) reported on the ability of the ASGP-R to take up galactose-terminated, protein-coated ferrite particles in vivo. These particles are taken up preferentially by hepatocytes (70%) and largely avoid uptake by the reticular endothelial system. For a crystal-based material to bind the ASGP-R of hepatocytes, it must penetrate pores of the 100-nm fenestrae characteristic of endothelial cells that lie between the hepatocyte and the circulatory compartment (13). Particles smaller than 100 nm are needed to serve as hepatocyte-directed diagnostic agents.
Josephson et al. (14) developed the findings of Sato et al.
(12) by synthesizing a superparamagnetic iron oxide colloid
(
50 nm) coated with the polygalactosylated polysaccharide
arabinogalactan. This colloid cleared rapidly from the vascular system;
its clearance was inhibited by asialofetuin but not by fetuin,
indicative of ASGP-R interaction. More than 90% of the particles were
taken up in hepatocytes.
The most thoroughly studied ASGP-R-directed agents are those based on galactosylated albumin labeled with radioactive technetium (99mTc) (15)(16)(17)(18)(19). Although the use of Tc-galactosylated albumin offers a unique and elegant approach to the study of liver biology, its application in the research laboratory and the clinic is limited. Many investigators have limited access to 99Tc generators and would prefer to avoid the use of radioactivity. In the clinic, radioactivity presents a potential hazard to the patient and medical personnel. 99Tc scintigraphy also relies on expensive instrumentation.
Stable-isotope labeling offers a cost-effective technology to deal with measuring the ASGP-R activity and biodistribution in vivo (20). This report explores the feasibility of using stable-isotope labeling of ASGP-R-directed agents for measuring ASGP-R activity in vivo. Two examples of such labels are reported—arabinogalactan labeled with samarium and gold colloids coated with lactosylated albumin—and their specificity for the ASGP-R and sensitivity are explored.
Unless stated otherwise, all reagents were obtained from Sigma-Aldrich. The diethylenetriaminepentaacetic acid (DTPA) conjugate of arabinogalactan was prepared as described previously (21). The final product contained 4.7 moles of DTPA per mole of arabinogalactan. The samarium chelate of arabinogalactan (Sm-arabinogalactan) was prepared from a solution of samarium chloride (0.3 mmol), adjusted to pH 2, that was added to DTPA-arabinogalactan (0.1 mmol), also adjusted to pH 2, in saline. The solution was adjusted to pH 5, mixed for 30 min, and adjusted to pH 7.4. PolyGalactoseGoldTM colloidal gold (20 nm) and lactosylated albumin (1 g/L) were incubated for 2 h at room temperature in 0.01 mol/L bicarbonate buffer, pH 8.5. Colloidal gold was isolated by centrifugation and suspended in phosphate-buffered saline. Albumin-coated gold colloid was prepared by substituting bovine albumin for lactosylated albumin. The coating and stability of each colloid was evidenced by the suspension remaining red in the presence of saline (a light-scattering color consistent with a 20-nm particle size). In contrast, uncoated colloidal gold suspended in saline turned purple and a precipitate formed.
The stability of Sm-arabinogalactan was evaluated in rat serum by incubation for 60 min at 37 °C followed by analysis using size-exclusion chromatography (Sephadex G-50). All of the samarium was retained in the fractions corresponding to arabinogalactan.
Rats (300–375 g) used in all studies were male hooded/BBZDR
(BioMedical Research Models). For each set of experiments, three rats
were anesthetized with Nembutal. The carotid artery and jugular vein
were exposed, and the animal received via the jugular vein an injection
of Sm-arabinogalactan (4 mg/kg), PolyGalactoseGold (50 µg/kg), or
albumin-coated gold (50 µg/kg). A second set of animals received
injections of asialofetuin (100 mg/kg) followed by Sm-arabinogalactan
or PolyGalactoseGold. At various times, blood samples were removed from
the carotid artery. After 60 min, the animals were sacrificed, and
1-g amounts of various tissues were recovered for quantification of
samarium or gold. Tissues were blotted to remove blood and clots. No
additional processing of tissue such as perfusion was necessary to
perform neutron activation analysis.
The samarium and gold content in all samples was quantified by neutron activation by BioPAL (Worcester, MA; www.biopal.com). Samples were placed in 2-mL polypropylene tubes free of trace element contaminants and dried at 70 °C for a minimum of 12 h. An internal standard of tungsten, to correct for variations in neutron flux, was added to each sample. Samples were activated for 15 min in a neutron field created by a 2-MW nuclear reactor. Short-lived activated products, principally resulting from sodium and chloride, were allowed to decay for 2 days, and the remaining radioactivity from activated samarium or gold was counted using a high-resolution gamma spectrometer.
The fate of Sm-arabinogalactan after intravenous injection into rats
was examined by obtaining the biodistribution of the conjugate 60 min
after injection. Biodistribution of the conjugate was inferred by the
presence of samarium in tissues after the injection of
Sm-arabinogalactan. Sm-arabinogalactan was present in the liver (51%
of injected dose) and urine (35% of injected dose). Less than 1% of
the injected dose was present in spleen or blood. When asialofetuin, a
ligand for the ASGP-R of hepatocytes, was co-injected with
Sm-arabinogalactan, hepatic uptake decreased to 3%, whereas the urine
fraction increased to 52%. Asialofetuin also increased blood
concentrations of Sm-arabinogalactan from 1% to 17%, which increased
urinary elimination. Biodistribution data and inhibition of hepatic
uptake by asialofetuin indicated that the hepatic clearance of
Sm-arabinogalactan was mediated by the ASGP-R. Renal clearance of
Sm-arabinogalactan is consistent with the conjugate’s low molecular
mass (40 kDa) and compact globular shape allowing excretion
by glomerular filtration.
The fate of PolyGalactoseGold after intravenous injection into rats and
mice was examined by obtaining the biodistribution of the conjugate 60
min after injection. Biodistribution of the conjugate was inferred by
the presence of gold in tissues after the injection of
PolyGalactoseGold. PolyGalactoseGold was present in the liver
(100% of injected dose) but not in the spleen, adrenals, lung,
kidney, heart, marrow, brain, muscle, or urine. When asialofetuin, a
ligand for the ASGP-R of hepatocytes, was co-injected with
PolyGalactoseGold, blood clearance was substantially slower (Fig. 1
). When albumin-coated colloidal gold was substituted for
PolyGalactoseGold, blood clearance for albumin-coated gold was
also substantially slower than that seen with PolyGalactoseGold,
demonstrating the specific role of galactose for clearing
PolyGalactoseGold. These results are consistent with reports in the
literature that lactosylated albumin is a ligand of the ASGP-R and
indicate that the hepatic clearance of PolyGalactoseGold was mediated
by the ASGP-R. The lack of renal clearance of PolyGalactoseGold is
consistent with its size (20 nm), which corresponds to a molecule
approximately the size of IgM. With the activation procedure described
above, 1 ng of gold corresponding to 800 dpm could be easily
distinguished from background (5 dpm).
|
We hypothesize that a compound consisting of a stable isotope bound to a receptor-directed reagent can be assayed with neutron activation technology to study and evaluate hepatic function. As a first step we characterized two ASGP-R-directed reagents labeled with a stable isotope of either samarium or gold, each possessing excellent properties for neutron activation. This approach presents an alternative to existing methods used to measure ASGP-R activity using radioactive, magnetic, or dye-labeled agents.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or presence of asialofetuin (•) or with
gold colloid coated with albumin (
). Blood samples were drawn at the
indicated times, and the percentage of gold remaining in the blood
(normalized for gold concentration at time zero) was determined using
neutron activation analysis. Values presented represent the mean value
obtained with three rats. Mean (SD) values for samples were calculated
using the percentages of gold remaining in the blood. Standard
deviations are therefore not calculated for values of 100%. Mean (SD)
values at 5, 10, and 15 min were as follows: lactosylated
albumin-coated gold colloid, 33% (6.1%) and 8% (1.0%);
lactosylated albumin-coated gold colloid in the presence of
asialofetuin, 95% (1.7%), 43% (3.6%), and 9% (2.0%);
albumin-coated gold colloid, 100%, 75% (5.2%), and 42% (6.2%).