Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Nerve growth factor (NGF)¹ is required by neural crest-derived sensory and sympathetic neurons for survival during embryonic and early postnatal life. NGF is produced and secreted in the target tissues of peripheral neurons, and exerts its effects on neuronal survival by terminal uptake and retrograde transport (Angeletti and Levi-Montalcini, 1970; Thoenen and Barde, 1980). The recent discovery and characterization of additional trophic factors, brain-derived neurotrophic factor, neurotrophin 3, and neurotrophin 4/5, which are structurally related to NGF but have different neuronal specificities (Barde, 1989; Hohn et al., 1990; Berkemeier et al., 1991; Henderson et al., 1993; Davies, 1994), indicate that competition for and survival dependence on tissue-derived trophic factors are critical events in the assembly and function of the peripheral nervous system (PNS; Purves and Lichtman, 1985). Additional responses induced by NGF administered in sensory and sympathetic neurons in vitro and in vivo include chemotaxis (Campenot, 1982; Hoyle et al., 1993; Levi-Montalcini, 1983), peripheral axonal branching (Diamond et al., 1987), dendritic and terminal arborization (Snider, 1988; Hoyle et al., 1993), regulation of neurotransmitter production, including catecholamine synthesis and regulation of neuropeptide levels (Kessler and Black, 1980a,b), establishment of functional synaptic connections (Diamond et al., 1987; Hoyle et al., 1993), and long-term control of metabolic functions and cellular size (Thoenen and Barde, 1980; Levi-Montalcini, 1987; Rush et al., 1995).

The physiological effects of NGF are mediated by binding to specific cell surface high- and low-affinity membrane receptors expressed by the responsive tissues. The low-affinity receptor is a 75-kDa glycoprotein (p75), which binds to all members of the neurotrophin family (Rodriguez-Tebar et al., 1990, 1992; Escandon et al., 1993, 1994). A 140-kDa receptor tyrosine kinase (trkA) constitutes the high-affinity receptor (Kaplan et al., 1991; Barbacid, 1994). p75 and trkA NGF receptors are expressed in most adult sympathetic and many primary sensory dorsal root ganglia (DRG) neurons and their fiber processes (Verge et al., 1992; McMahon et al., 1994; Shelton et al., 1995; Wetmore and Olson, 1995; Wright and Snider, 1995). Based on the action of NGF on these tissues, a potential therapeutic role has been proposed for recombinant human nerve growth factor (rhNGF) in peripheral neuropathies, which are characterized by dysfunction of the small unmyelinated fibers and sympathetic neurons.

In a recent phase II clinical trial, rhNGF was demonstrated to be active and improved some of the sensory impairments present in diabetic neuropathy when administered three times a week at doses of 0.1 and 0.3 µg/kg for 6 months. In this study, Apfel et al. (1998) measured the following endpoints for efficacy: 1) the quantitative neurologic examination (Neuropathy Impairment Score in the Lower Limbs), 2) quantitative measures of sensory function (cooling detection threshold, vibratory detection threshold, and heating pulses: 5.0), and 3) the symptom questionnaires (Neuropathy Symptom Profile, Neuropathy Symptoms and Change, and the global symptom assessment). Although numerous studies have investigated the in vivo and in vitro properties of NGF, studies characterizing the in vivo distribution of rhNGF are lacking. Here we present the characterization of organ and tissue disposition and pharmacokinetic (PK) analysis of rhNGF after single and multiple administration in cynomolgus monkeys. NGF levels in plasma were determined by trichloroacetic acid (TCA) precipitation of labeled material, a two-site enzyme immunoassay, and a novel bioassay procedure. Radioactive tissue distribution was assessed by radioanalysis, anti-NGF immunoprecipitation followed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), and emulsion microautoradiography. These studies demonstrated that after s.c. administration of rhNGF at relatively low doses (0.8 µg/kg) in monkeys, rhNGF is capable of accessing and localizing in the target tissues intended for NGF therapy in diabetic neuropathy.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Iodination of rhNGF. rhNGF produced in Chinese hamster ovary cells were purified as a dimer consisting of two 118 amino acid polypeptides, formulated in 10 mM sodium acetate/140 mM NaCl, pH 5.5 and stored at 4°C (Genentech, Inc., South San Francisco, CA). rhNGF was iodinated with sodium I-125 (NEN Life Science Products, Boston, MA) using the lactoperoxidase method. Twenty micrograms of rhNGF was labeled in sodium acetate buffer, pH 5.5 with 1 U/ml lactoperoxidase and 2 mCi sodium I-125. The reactions were initiated by the addition of 15 µl of H₂O₂ diluted 1:174,000. After a 5-min incubation at room temperature, 15 µl of H₂O₂ was added and the reaction was stopped 5 min later by adding 15 µl of 20 mM N-acetyl-L-tyrosine. The iodinated proteins were separated from unincorporated I-125 using PD-10 size exclusion columns (Pharmacia, Uppsala, Sweden). ¹²⁵I-rhNGF was stored at 4°C in a buffer of 10 mM sodium acetate/140 mM NaCl, pH 5.5 containing 1 mg/ml human serum albumin and 0.5 mg/ml protamine sulfate. The final material was >98% TCA-precipitable and SDS-PAGE of the test material revealed one single radioactive band with a molecular mass of approximately 13 kDa.

Animals Husbandry. Eight male, mature adult cynomolgus monkeys (Macaca fascicularis), weighing between 3.3 and 4.9 kg, were obtained from Hazleton Research Products, Inc. (Madison, WI). Animals were maintained in a controlled environment (19-26°C, 50 ± 20% relative humidity, and a 12-h light/dark cycle). During the acclimatization period, certified primate diet no. 5048 (PMI Feeds, Inc., Richmond, IN) and water were provided ad libitum. Studies in animals were approved by the Institutional Animal Care and Use Committee and were performed in accordance with the guidelines of the American Association for Accreditation of Laboratory Animal Care.

Treatment Protocol and PK Study Designs. To reduce uptake of I-125 to the thyroid, monkeys received 10 mg of sodium iodide by oral administration at 48, 24, and 1 h before administration of test material. Doses (s.c.) of radiolabeled and nonradiolabeled rhNGF in normal saline solution were administered via bolus injection between the scapulae. Table 1 summarizes the study design. Group 1 animals received ¹²⁵I-rhNGF (49 µCi/kg, 0.8 µg/kg) + unlabeled rhNGF (2.0 mg/kg), and group 2 animals received a single dose of ¹²⁵I-rhNGF (83 µCi/kg, 0.8 µg/kg). Group 1 animals continued to receive 10 doses of rhNGF every other day (2.0 mg/kg, doses 2 through 11). To identify any potential changes in PK parameters on multiple dosing, a relatively high dose of 2 mg/kg rhNGF was chosen. Dose 12 was a mixture of ¹²⁵I-rhNGF (71 µCi/kg, 0.8 µg/kg) and rhNGF, followed by two additional doses of rhNGF every other day (2.0 mg/kg, doses 13 and 14). The last dose (dose 15) was 66 µCi/kg ¹²⁵I-rhNGF. Dose volume was 1.0 ml/kg for group 1 animals and 0.5 ml/kg for group 2 animals. Doses and dose volumes were adjusted weekly based on body weight.

Blood Collection. Blood samples were collected from group 1 animals (n = 4) for single dose PK assessment through 48 h at dose 1 and for multiple dose PK assessment through 48 h at dose 12. Blood samples (~3.0-ml) were collected via the femoral vein in EDTA at 15 to 30 min predose; 10, 20, and 30 min; and 1, 2, 3, 4, 6, 8, 12, 16, 20, 28, 36, and 48 h after doses 1 and 12. Plasma was harvested and stored frozen at 70°C. Fecal and urine samples were collected predose (~ 24-0 h), and 0 to 24, 24 to 48, 48 to 72, 72 to 96, and 96 to 120 h postdose from group 1 animals at doses 1 and 12.

Termination and Tissue Collection. For group 1 at dose 15 (multiple dose), and group 2 after the single dose, two animals were sacrificed at 8 and 24 h postdose by exsanguination under sodium pentobarbital anesthesia. At sacrifice, blood was collected via cardiac puncture into tubes containing EDTA. Animals were intracardially perfused with normal saline. Various tissues and organs were excised, rinsed with water, and split into two samples: one was stored frozen at 70°C for radioanalysis and TCA precipitation, and the other was prepared in tissue-embedding material, frozen in an isopentane/dry ice bath, and stored frozen at 70°C for emulsion microautoradiography.

TCA Precipitation. The total radioactivity of plasma, urine, and tissue homogenate supernatants was quantitated in a gamma counter (MinAxi Auto Gamma 5000 series; Packard Instrument Company, Elk Grove, IL). Two hundred microliters of cold (4°C) 1% BSA, pH 7.2 was added to 20 µl of EDTA plasma or 50 µl of tissue homogenate supernatant and vortexed. The samples were treated with 50 µl of cold (4°C) 50% TCA, vortexed, and incubated for 30 min on ice. After an additional 500 µl of cold (4°C) 10% TCA was added, samples were vortexed and centrifuged at 14,000 rpm for 5 min. The supernatants were aspirated and the radioactivity of the precipitated protein pellet was quantified in a gamma counter. The data are expressed as nanogram equivalents of ¹²⁵I-rhNGF per gram of wet tissue or nanogram equivalents of ¹²⁵I-rhNGF per milliliter of plasma, assuming 1 g = 1 ml. This value was calculated by dividing the TCA-precipitable radioactivity per gram of wet tissue or TCA-precipitable radioactivity per milliliter of plasma (cpm/g and cpm/ml, respectively) by the specific activity of the dosed material (cpm/ng).

Plasma Analysis. Monkey plasma samples were analyzed for the immunoreactive NGF concentration by enzyme-linked immunosorbent assay (ELISA). The two-site ELISA used polyclonal antibodies raised to rhNGF in rabbits as described previously (Bennett et al., 1990). The assay range was from 0.4 to 6.0 ng/ml. Bioactive NGF concentration was quantified by a kinase receptor activation assay (KIRA). This assay detected NGF-induced trkA activation by measuring receptor phosphorylation as described previously (Sadick et al., 1997). Data are presented as bioactive NGF concentration with an assay range of 1.2 to 20 ng/ml.

PK Analysis. Individual NGF concentrations were used to construct semilogarithmic plasma concentration-versus-time profiles. Systemic clearance was assessed after s.c. dosing using noncompartmental methods (linear trapezoidal) by Win-nonlin, Version 1.1 (Scientific Consulting Inc., Apex, NC). The PK parameters were derived from data obtained by four different detection methods: 1) total radioactivity (ng · eq/ml), 2) TCA-precipitable radioactivity (ng · eq/ml), 3) ELISA concentrations (ng/ml), and 4) KIRA concentrations (ng/ml) versus time (h). The PK parameters used as the best estimates of PK disposition were those derived from ELISA concentrations. The KIRA results were based on a smaller data set, thus giving less well defined PK parameters. Due to a relatively high metabolism of the I-125 moiety in plasma in vivo, the total radioactivity values were not used to perform PK analysis. The most accurate estimates of mean (±S.D.) PK parameters, derived by ELISA, are displayed in Table 2. The NGF plasma concentration data are plotted in Fig. 1, displaying ELISA, TCA-precipitable, and KIRA results. Statistical analyses were not performed because of the small sample sizes studied.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   PK profile of NGF in plasma after single or multiple administration in monkeys. Male cynomolgus monkeys were given s.c. doses of NGF (2 mg/kg) three times a week for 4 weeks. A full plasma PK profile was obtained after doses 1 and 12. Blood was collected in EDTA 10, 20, and 30 min; 1, 2, 3, 4, 6, 8, 12, 16, 20, 28, 36, and 48 h postdose after doses 1 and 12. Data represent the mean ± S.D. plasma concentration-time data from four animals according to measurements by ELISA and KIRA (A) and TCA precipitation (B).

Immunoprecipitation and SDS-PAGE/Film Autoradiography. Tissues stored at 70°C were thawed on ice, blotted dry, and weighed; the amount of radioactivity associated with each tissue was quantitated in a gamma counter. Tissues were homogenized using a probe-type tissue homogenizer (Tekmar Tissumizer) at 1.0 g of tissue/10 ml of a lysis buffer containing 20 mM EDTA/0.1% SDS/1% Triton X-100 in PBS, pH 7.4. The tissue lysates were centrifuged (2000g for 10 min, 4°C). Four milliliters of the tissue homogenate supernatant and 3 ml of undiluted plasma were immunoprecipitated by adding a 1:200 dilution (20 µl) with anti-mNGF- antiserum (rabbit anti-mouse 2.5S NGF, no. 40015; Collaborative Biomedical Products, Bedford, MA) overnight at 4°C with gentle agitation. Protein-A Sepharose in PBS (50-µl packed volume, P-3391; Sigma, St. Louis, MO) slurry was added to the samples and incubated for 4 h at 4°C with gentle agitation. The Protein-A Sepharose pellets were washed three times with lysis buffer. After centrifugation, the supernatants were aspirated and the pellets were resuspended in 80 µl of SDS-PAGE sample buffer. Samples were boiled at 100°C for 10 min, vortexed, and the pellet was spun down at 10,000g for 1 min. Forty microliters of the supernatants were analyzed on a 16% Tris-Glycine gel (Novex, San Diego, CA) under reducing conditions (5% -mercaptoethanol). Gels were subjected to autoradiography using Kodak X-Omat film (Eastman Kodak, Rochester, NY) and intensifying screens at 70°C.

Emulsion Microautoradiography. Two 3- to 5-µm cryosections of the collected tissues were dipped in Kodak NTB3 emulsion (no. 165-4441; Eastman Kodak) and allowed to develop in the dark at 4°C for 1 or 2 weeks, respectively. At the end of the exposure period, the sections were developed using Kodak D19 developer (no. 146-4593; Eastman Kodak) and then fixed using Kodak Fixer (no. 190-2485; Eastman Kodak). After routine H&E counterstaining, the sections were evaluated using bright and dark field microscopy.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Pharmacokinetics of rhNGF after Single and Chronic s.c. Administration. In this study, NGF plasma concentrations were evaluated after s.c. administration using four detection procedures. Total radioactivity was measured, and plasma samples were subjected to TCA precipitation to calculate nanogram per milliliter equivalents of ¹²⁵I-rhNGF. In addition, equivalent plasma aliquots were subjected to an ELISA. In some plasma samples, levels of NGF were also characterized by a novel bioassay detection (KIRA) that measured activation of the NGF high-affinity receptor, trkA system (Sadick et al., 1997). Mean ± S.D. plasma concentration-time curves for NGF obtained from TCA precipitation, ELISA, and KIRA analyses are shown in Fig. 1.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Stability of 125I-rhNGF in plasma after single s.c. administration. Plasma samples collected from the PK arm of the study (125I-rhNGF + 2 mg/kg rhNGF at dose 1, group 2) at 8, 16, 20, and 24 h (terminal bleeding) were immunoprecipitated with anti-NGF- antiserum followed by a Protein A-Sepharose step as described in Materials and Methods. Immunoprecipitates were applied to 16% SDS-PAGE and film autoradiography. Arrow indicates position of control 125I-rhNGF (molecular mass ~13 kDa).

Analysis of ¹²⁵I-rhNGF Stability in Plasma after Single s.c. Administration (PK Analysis dose 1, group 1). Thirty minutes after single s.c. administration, 73.7 ± 4.2% of the circulating radioactivity was TCA-precipitable. This value decreased to 31.3 ± 3.6% at 12 h postdose. At the last timepoint (48 h), only 9.27 ± 3.3% of the material was TCA-precipitable (data not shown). Therefore, TCA-precipitable radioactivity rather than total radioactivity was used as a measure of ¹²⁵I-rhNGF concentrations in plasma. Only a small fraction of radioactivity in blood (7.74-16.46%) was associated with the cellular fraction. To characterize the integrity of the labeled material in plasma, aliquots were subjected to immunoprecipitation using anti-NGF- antiserum followed by Protein A-Sepharose. The immunoprecipitated plasma samples were separated by SDS-PAGE and visualized by film autoradiography (Fig. 2). In agreement with the TCA-precipitable plasma time-concentration profile (Fig. 1B), a time-dependent decrease in the density of the 13-kDa band, which corresponded to the molecular mass of monomeric NGF was observed. Interestingly, no lower molecular mass degradation products were observed in the plasma samples, indicating long-term stability of this molecule.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   Nanogram equivalents of 125I-rhNGF TCA-precipitable radioactivity in non-neuronal tissues 8 and 24 h after single and multiple s.c. administration. Figure 3 shows the amount of 125I-rhNGF ng equivalents expressed as TCA-precipitable cpm/g tissue at 8 and 24 h after a single dose (A; group 2) and 15 doses (B; group 1). TCA precipitation was performed on the tissue lysate supernatants and nanogram equivalents were calculated as described in Materials and Methods. Data represent the mean of two animals per timepoint. The error bars represent each value. P-A, peripheral-axillary; cort., cortical; med., medullary; S-M, submandibular-maxillary; par., parotid; vent., ventricle; sept., septum; mes., mesenteric.

Organ Distribution of Radioactivity after s.c. Administration of ¹²⁵I-rhNGF.

Non-neuronal Organs were collected at 8 and 24 h after single (dose 1, group 2) and multiple (dose 15, group 1) ¹²⁵I-rhNGF administration. Samples from the tissue lysates were analyzed by TCA precipitation. Nanogram equivalents of ¹²⁵I-rhNGF were calculated by dividing the TCA-precipitable radioactivity per gram of wet tissue (cpm/g) by the specific activity of the administered drug (cpm/µg ¹²⁵I-rhNGF). Figure 3, A and B (doses 1 and 15, respectively), show the results of this analysis. Table 3 shows the actual values in ¹²⁵I-rhNGF nanogram equivalents per gram of wet tissue. After s.c. administration, the test material was observed in most tissues analyzed. The non-neuronal tissues with the highest mean concentration of ¹²⁵I-rhNGF radioactivity at 8 h postdose in decreasing order were: thyroid, injection site, peripheral-axillary lymph nodes, adrenals, kidney, liver, and spleen. Significant amounts of TCA-precipitable radioactivity were also found in the lymph nodes draining the injection site. The injection site and the thyroid were the tissues with the highest amount of radioactivity at both timepoints. The high levels of TCA-precipitable material in the thyroid were unexpected. With the exception of the thyroid, the disposition of ¹²⁵I-rhNGF to non-neuronal tissues decreased at 24 h. The highest mean concentration of radioactivity in non-neuronal tissues at 24 h postdose were found in the thyroid, injection site, peripheral-axillary lymph nodes, adrenals, kidney, liver, and spleen.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Nanogram equivalents of 125I-rhNGF radioactivity in neuronal tissues 8 and 24 h after single and multiple s.c. administration. Figure 4 shows the amount of 125I-rhNGF nanogram equivalents expressed as radioactivity per gram of tissue at 8 and 24 h after a single dose (A) and 15 doses (B). Tissues were harvested and processed for quantitation of radioactivity as described in Materials and Methods. SC, spinal cord; SCG, sympathetic ganglia.

Neuronal. The disposition of labeled rhNGF to neuronal tissues is shown in Fig. 4, A and B (doses 1 and 15, respectively). Nanogram equivalents of ¹²⁵I-rhNGF per gram of wet tissue are shown in Table 4. ¹²⁵I-rhNGF distributed to several neuronal tissues, including sensory and sympathetic ganglia in the periphery. Disposition of radioactivity to neuronal tissues at 8 h postdose were in decreasing order: pituitary gland, sympathetic ganglia, sciatic, and tibial nerves. In contrast, distribution of ¹²⁵I-rhNGF to the central nervous system (CNS) was minimal (spinal cord) or negative (brain). This was expected because NGF does not cross the blood-brain barrier. The levels of ¹²⁵I-rhNGF in the thoracic-lumbar and cervical sympathetic ganglia increased considerably from 8 to 24 h (1.26-16.7 and 0.39-6.43 ng equivalents/g, respectively). The concentration of ¹²⁵I-rhNGF in DRG (lumbar) also rose from 0.47 to 1.42 ng equivalents/g. This is consistent with the expression of specific NGF receptors in these cellular populations. With the exception of the injection site and thyroid, disposition of labeled rhNGF to sympathetic ganglia at 24 h after dose 1 was the highest of all tissues analyzed. The amount of labeled material localized in sensory DRG and sympathetic neurons at 24 h after dose 15 was considerably less than at 24 h after dose 1. At dose 15, the concentrations of rhNGF in the thoracic-lumbar and cervical sympathetic ganglia rose from 0.30 to 1.32 and from 0.27 to 3.96 ng equivalents/g, respectively. The concentration of radioactivity in the cervical-distal DRG also rose to 0.21 ng equivalents/g tissue. Figure 5, A and B, shows the 24- to 8-h ratios of ¹²⁵I-rhNGF TCA-precipitable nanogram equivalents per gram of wet tissue in neuronal tissues after single and multiple administration, respectively.

View larger version (49K): [in this window] [in a new window]   Fig. 5.   Ratios of 125I-rhNGF 24 h/8 h radioactivity in neuronal tissues after single and multiple s.c. administration. Figure 5 shows the 24 h/8 h ratios of radioactivity (cpm/g tissue) after single (A) and multiple (B) dosing. Tissues were harvested and processed for quantitation of radioactivity as described in Materials and Methods. SC, spinal cord; SYMP, sympathetic ganglia.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   Immunoprecipitation of 125I-rhNGF in non-neuronal tissues at 8 and 24 h after single (dose 1) and multiple (dose 15) s.c. administration. Tissues were harvested and homogenized and lysates were immunoprecipitated with anti-NGF- antiserum followed by a Protein A-Sepharose step as described in Materials and Methods. Immunoprecipitates were applied to 16% SDS-PAGE followed by Kodak X-Omat film autoradiography and exposed at 70°C for 10 days. Arrows indicate the position of control 125I-rhNGF (molecular mass ~13 kDa).

SDS-PAGE Autoradiographic Analysis of ¹²⁵I-rhNGF Stability.

Non-neuronal To further qualitatively characterize the metabolic fate and specificity of the labeled material in vivo, tissue lysates were subjected to an immunoprecipitation step using anti-NGF antibodies, as described in Materials and Methods. SDS-PAGE analysis of the immunoprecipitated samples is shown in Fig. 6. Our results demonstrated that specific ¹²⁵I-rhNGF signals consisted mostly of a single band with various amounts of lower molecular mass degradation products. The non-neuronal tissues that showed the highest levels of ¹²⁵I-rhNGF processing indicated by the relative amounts of full-length ¹²⁵I-rhNGF versus lower molecular mass bands were kidney, liver, and spleen.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Disposition of 125I-rhNGF to lymph nodes at 8 and 24 h after single (dose 1) and multiple (dose 15) s.c. administration. Peripheral and mesenteric lymph nodes were dissected and processed for immunoprecipitation as described in Materials and Methods. Immunoprecipitates were applied to 16% SDS-PAGE followed by Kodak X-Omat film autoradiography and exposed at 70°C for 10 days. Arrows indicate the position of control 125I-rhNGF (molecular mass ~13 kDa). P-A, peripheral-axillary; mes, mesenteric.

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Disposition of 125I-rhNGF to the CNS. Different regions from the CNS were dissected and processed for immunoprecipitation as described in Materials and Methods. Immunoprecipitates were applied to 16% SDS-PAGE followed by Kodak X-Omat film autoradiography and exposed at 70°C for 10 days. Arrows indicate the position of control 125I-rhNGF (molecular mass ~13 kDa). S-cord = spinal cord.

Neuronal CNS. Various regions from the brain and the spinal cord were dissected and subjected to immunoprecipitation as described previously. Figure 8 shows the SDS-PAGE film autoradiography of immunoprecipitated samples. Disposition of ¹²⁵I-rhNGF to the CNS was minimal and the lowest amount of all analyzed tissues. Minimal ¹²⁵I-rhNGF signal was found in the cerebellum, frontal cortex, and hypothalamus. Trace amounts of radioactivity were detected in the brainstem-pons and the midbrain. Detectable ¹²⁵I-rhNGF signal was observed in the cervical spinal cord at dose 1 (8 h) and in the thoracic and lumbar regions of the spinal cord at dose 1 (24 h).

Neuronal PNS. Figure 9A shows the SDS-PAGE film autoradiography of peripheral nerves. ¹²⁵I-rhNGF was present in the radial nerve at 24 h and the sciatic nerve at 8 and 24 h. The material present in these neuronal structures was mostly intact ¹²⁵I-rhNGF (molecular mass ~13 kDa). Figure 9B shows the intense localization of labeled rhNGF into sympathetic and DRG. SDS-PAGE film autoradiographic analysis showed that ¹²⁵I-rhNGF disposition to these peripheral neurons increased with time after doses 1 and 15. In general, there was also a decreased disposition of labeled material to the tissues analyzed at dose 15 than at dose 1. This decrease in tissue disposition after chronic administration correlated with an increase in peak plasma concentrations of labeled rhNGF. Specific accumulation of ¹²⁵I-rhNGF in these tissues known to express high amounts of NGF receptors is indicative of a receptor-mediated uptake mechanism. In addition, there was evidence of active processing and metabolism of ¹²⁵I-rhNGF indicated by the presence of lower-molecular-mass radioactive bands. A high-molecular-mass ¹²⁵I-rhNGF complex was also observed at the top of the SDS-PAGE gel.

View larger version (28K): [in this window] [in a new window]   Fig. 9.   Disposition of 125I-rhNGF to the PNS. Different regions from the PNS as indicated in each panel were dissected and processed for immunoprecipitation as described in Materials and Methods. Immunoprecipitates were applied to 16% SDS-PAGE followed by Kodak X-Omat film autoradiography and exposed at 70°C for 10 days. Arrows indicate the position of control 125I-rhNGF (molecular mass ~13 kDa).

Excretion of Radioactivity. The amount of radioactivity recovered in feces, urine, and total excretion (feces and urine, including cage wash and cage wipe) for group 1 animals after doses 1 and 12 is shown in Fig. 10, A and B, respectively. In group 1 animals after dose 1, 90.6% of the administered radioactive dose was recovered in the excreta. Urinary excretion was the major route of elimination, accounting for a mean of 79.1% of the administered radioactive dose, whereas radioactivity in the feces accounted for only 1.7% of the administered dose. The pattern of excretion of radioactivity was similar in the group 1 animals after dose 12 with an average of 90.9% of the administered radioactive dose being recovered in the excreta. Urinary excretion was the major route of elimination (77.1% of the administered radioactive dose), whereas feces contained 1.8% of the administered dose. The cage wash and cage wipe collected at the conclusion of the study represented 11.3 and 0.8% of the administered dose, respectively.

View larger version (20K): [in this window] [in a new window]   Fig. 10.   Cumulative radioactivity of rhNGF in feces and urine. Mean cumulative amount of radioactive dose in feces and urine at specified intervals postdose for group 1 monkeys dosed s.c. with 125I-rhNGF and rhNGF (2 mg/kg) at doses 1 (A) and 12 (B). Data represent the mean ± S.D. of four animals.

Emulsion Microautoradiography Analysis. To assess the cellular disposition of ¹²⁵I-rhNGF after s.c. administration, frozen sections of selective tissues were processed for emulsion microautoradiography. Tissue localization of ¹²⁵I-rhNGF was frequently found associated with nerves and neurons of the sympathetic nervous system (Fig. 11, A, B, and E). Large neurons within the sympathetic ganglia were specifically labeled (Fig. 11, A and B) as well as neurons with the DRG (data not shown). Figure 11, C and D, shows disposition of ¹²⁵I-rhNGF to perivascular nerves.

View larger version (74K): [in this window] [in a new window]   Fig. 11.   Emulsion microautoradiography of 125I-rhNGF disposition to sensory and sympathetic neurons. Frozen tissue sections (3-5 µm) were processed for emulsion microautoradiography analysis. Figure shows dark field (A) and bright (B) disposition of radioactivity to sympathetic neuronal cell bodies 24 h postdose. C and D, 125I-rhNGF presence in a perivascular nerve by dark and bright field analysis, respectively. E, 125I-rhNGF disposition to the sympathetic fibers. Calibration bars are 90 (A and B), 45 (C and D), and 95 µm (E).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

To date, there is limited data concerning the in vivo distribution of rhNGF (Pradier et al., 1994; Tria et al., 1994). A characterization of the tissue disposition and in vivo PK properties of rhNGF in nonhuman primates would provide relevant information to better understand the pharmacological effects of rhNGF as a potential therapeutic agent. We report here the kinetics and tissue uptake of ¹²⁵I-rhNGF after single and multiple s.c. administration in cynomolgus monkeys.

There are two receptor classes for NGF with different affinities and functions (Sutter et al., 1979; Raffioni et al., 1993; Dechant et al., 1994). After ligand binding, the NGF-receptor complex is internalized and transported retrogradely to the cell soma (Thoenen and Barde, 1980). Both p75 and trkA receptors undergo ligand-induced dimerization, activating several intracellular signal transduction pathways (Chao, 1992; McDonald and Chao, 1995). The K_d value for the functional high-affinity receptor is approximately 10 pM, corresponding to ~0.25 ng/ml rhNGF. Our determination of nanogram equivalents per gram tissue ¹²⁵I-rhNGF indicated that after a single bolus s.c. administration of ¹²⁵I-rhNGF (0.8 µg/kg) in nonhuman primates, the labeled material distributed into highly perfused organs. Elevated TCA-precipitable values of ¹²⁵I-rhNGF in the peripheral-axillary lymph nodes compared with other tissues were presumably due to these lymph nodes being involved in draining the injection site. The assumption is supported by the relative position of the lymph node to the injection site and by low levels observed in the mesenteric lymph nodes. Disposition of ¹²⁵I-rhNGF to the adrenals confirm the results of Otten et al. (1977), reporting the localization of ¹²⁵I-rhNGF in adrenal chromaffin cells of adult rats after i.v. administration. Urinary excretion was the major route of elimination of I-125 radioactivity. In addition, the high levels of TCA-precipitable radioactivity found in the kidneys indicate that this organ may play an important role in the clearance of ¹²⁵I-rhNGF. In general, the mean concentration of radioactivity in non-neuronal tissues declined at 24 h postdose.

The increased concentrations of radioactivity in parts of the PNS at 24 h over 8 h postdose is indicative of specific uptake in these target tissues known to express specific receptors for NGF. The concentrations of ¹²⁵I-rhNGF present in sensory and sympathetic ganglia are above the K_d value for the high-affinity functional NGF-receptor complex. This indicates that at this s.c. dose, NGF reaches the necessary concentrations in vivo to elicit its various biological responses. This is important because the s.c. dose administered (0.8 µg/kg) in this preclinical model is related to the doses used in the clinical trials (0.1 or 0.3 µg/kg) for the use of rhNGF in the treatment of peripheral neuropathies. The lowest levels of TCA-precipitable radioactivity were found in the brain, consistent with the inability of highly basic proteins such as NGF to cross the blood-brain barrier. The noticeable decrease in ¹²⁵I-rhNGF disposition present primarily in the DRG contrasts with an increase in the ¹²⁵I-rhNGF C_max (1.3 ± 0.12 and 4.1 ± 1.2 µg/ml at doses 1 and 12, respectively) and AUC (14.5 and 37.6 µg · h/ml for doses 1 and 12, respectively). The decrease in ¹²⁵I-rhNGF neuronal disposition and increase in ¹²⁵I-rhNGF plasma levels after chronic administration (2 mg/kg) may indicate receptor down-regulation or receptor saturation. Additional experiments need to be conducted to directly address these possibilities. Currently, clinical data that establishes comparisons after chronic dosing of rhNGF is not available.

Histological analysis of emulsion autoradiography indicated that specific labeling was found in association with nerves most frequently localized adjacent to large vessels in sections of the kidney, spleen, liver, and salivary gland. Large neurons within the sympathetic ganglia were also labeled, as well as cells within the DRG. Specific binding at the injection site was localized to nerves and vessel walls. Immunoprecipitation analysis of dissected organs, tissues, and plasma with anti-NGF antiserum confirmed a widespread disposition of full-length ¹²⁵I-rhNGF after s.c. administration as well as specific localization at 24 h postdose in sensory and sympathetic neurons. The observed predominant band of 13 kDa corresponds to the position of the NGF monomer. These experiments also indicated active metabolism of ¹²⁵I-rhNGF in the kidney, liver, and most abundantly in the target tissues for NGF uptake. In contrast, ¹²⁵I-rhNGF remains highly stable in plasma for as long as 24 h postdose after single administration. Decreases in the percent TCA precipitability of radioactivity in plasma over time reflect dehalogenation of the I-125 moiety. The immunoprecipitation and SDS-PAGE analysis also revealed that the high amounts of TCA-precipitable radioactivity measured in the thyroid do not represent true levels of ¹²⁵I-rhNGF but rather indicate metabolism and processing of the I-125 moiety.

We have demonstrated that after a s.c. administration at relevant clinical doses, ¹²⁵I-rhNGF is taken up by tissues, selectively accumulating into sensory and sympathetic neurons. These findings are consistent with the improvements in sensory function observed in the phase II clinical trial for the use of rhNGF in the treatment of diabetic peripheral neuropathy.