Introduction

Top Abstract Introduction Methods Results Discussion References

Initial interest in SRIF, a cyclic tetradecapeptide isolated from ovine hypothalamus (Brazeau et al., 1973), centered on its ability to inhibit GH release from the pituitary gland. Many other actions of SRIF, including inhibition of secretion of thyrotropin, prolactin, insulin, gastrin and glucagon hormones and a reduction in gastric acid secretion, are now known (see review in Moreau and DeFeudis, 1987). SRIF also is thought to function as a neurotransmitter or neuromodulator (Reubi et al., 1981; Srikant and Patel, 1981; Epelbaum, 1986) because of its ubiquitous central nervous system distribution and its association with neuronal SRIF receptors.

High-affinity SRIF receptors are present on SRIF target tissues such as brain and pituitary (Heiman et al., 1987; Srikant and Patel, 1985; Epelbaum et al., 1982) adrenal cortex and medulla (Srikant and Patel, 1985; Epelbaum et al., 1995), pancreas (Zeggari et al., 1986; Srikant and Patel, 1986) and gastrointestinal tract (Gu et al., 1995; Reubi et al., 1990). A wide variety of human tumors express high-affinity SRIF receptors also. The incidence of tumors expressing these receptors is especially high among neuroendocrine tumors, including pituitary adenomas, carcinoids, islet cell carcinomas, pheochromocytomas, medullary thyroid carcinomas and SCLC. In addition, SRIF receptors are often found in astrocytomas, neuroblastomas, meningiomas, lymphomas and renal cell carcinomas (Reubi et al., 1993). In those tumors, which do have SRIF receptors, receptor density often is quite high (Reubi et al., 1993; Taylor, 1993).

Recently, five SRIF receptor subtypes have been cloned, and their tissue distribution has been studied. This progress has been reviewed by Viollet et al. (1995). All five SRIF receptor subtypes are members of the G protein-coupled superfamily, possessing seven transmembrane regions (Yamada et al., 1992; Viollet et al., 1995). They are closely related to the opiate receptors (Evans et al., 1992; Viollet et al., 1995), and ligands for SRIF and opiate receptors may share similarities as well. Indeed, the potent mu ligand CTP and the SRIF receptor ligand RC-102 (sequences shown in table 1) are each composed of eight amino acid residues and differ only in the replacement in the RC-102 compound of the Phe3 residue with a tyrosine and of the Cys7 residue with a penicillamine (Cai et al., 1986; Kramer et al., 1989).

Based on sequence homology, it is possible to classify the five cloned SRIF receptor subtypes into two categories. SSTR1 and SSTR4 types are 71% homologous, and the SSTR2/SSTR5/SSTR3 group has 65% to 75% homology. Homologies in transmembrane regions 2, 3, 5 and 7 are even higher within these categories (Viollet et al., 1995; Hoyer et al., 1995). Furthermore, classification according to ligand binding data shows that the SSTR1/SSTR4 receptors have very low affinity for the SRIF analogs BIM-23014, Octreotide and Seglitide, whereas the SSTR2/SSTR5/SSTR3 subtypes have high-to-moderate affinity for these three ligands whose structures are shown in table 1 (Bruns et al., 1996; Viollet et al., 1995; Hoyer et al., 1995).

There are numerous disease conditions for which SRIF analogs are being evaluated for their therapeutic potentials and are already in clinical testing. SRIF, however, is too enzymatically unstable to be of value for clinical use except when administered by continuous intravenous infusion (Moreau and DeFeudis, 1987). The SRIF analog BIM-23014 (Lanreotide, Somatuline) and others have been successfully used in the treatment of pituitary disorders such as acromegaly (Moreau and DeFeudis, 1987; Morange et al., 1994; Wen and Loeffler, 1995; Caron et al., 1995; Marek et al., 1994) in which there is hypersecretion of growth hormone, in TSH-secreting adenomas (Wen and Loeffler, 1995; Gancel et al., 1994) and in the management of carcinoid tumors (Anthony et al., 1993; Scherubl et al., 1994).

SRIF analogs inhibit in vitro cell proliferation in a variety of human cancer cell lines, including SCLC (Taylor et al., 1988), pancreatic carcinoma (Liebow et al., 1989) and breast carcinoma (Setyono-Han et al., 1987). Thus, SRIF analogs may have a direct antiproliferative effect on these cancers and may act by inhibiting secretion of autocrine growth factors. BIM-23014 is active in vivo against several tumor types such as SCLC (Taylor et al., 1988; Prevost et al., 1994; Bogden et al., 1990b; Anthony et al., 1993), breast carcinoma (Thomas et al., 1992), prostate cancer (Bogden et al., 1990a; Maulard et al., 1995), carcinoids (Anthony et al., 1993; Scherubl et al., 1994) and cholangiocarcinoma (Tan et al., 1995), and ¹²⁵I-labeled BIM-23014 has been used for intraoperative detection of occult gastrinomas (Woltering et al., 1994).

Based on the clinical applications and widespread use in SRIF receptor research, it is important to establish the in vivo distribution of BIM-23014 (Lanreotide, Somatuline) to target and nontarget tissues. In this study, we determined the biodistribution (after subcutaneous injection) in male Sprague-Dawley rats, of ¹²⁵I-labeled BIM-23014 and two newly developed SRIF analogs, BIM-23190 and BIM-23197.

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals. Male Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, IN) weighing 200 to 300 g were used. Rats were housed under standard 12-hr light/12-hr dark conditions (lights on at 7:00 a.m.) and received food and water ad libitum before and during the experiments. Animal protocols for the state handling and treatment of these rats were approved by The University of Arizona Institutional Animal Care and Use Committee (IACUC).

Peptide iodinations. All peptides for iodinations were obtained from Biomeasure (Milford, MA). Peptides were radiolabeled with ¹²⁵I using the chloramine T procedure described previously (Schetz et al., 1995). The iodinated peptides were purified by HPLC on a Beckman Instruments (Fullerton, CA) ultrasphere ODS column (0.46 × 25 cm) using eluants of 0.1% TFA in H₂O and 0.1% TFA in ACN. A linear gradient of 15% to 40% ACN in 25 min and a column flow rate of 1.5 ml/min were used. One-minute fractions were collected, and 5-µl portions were counted on a Beckman 5500 series -counter. Iodinated peptides were well resolved from their unlabeled precursors and eluted ~4 min later than the unreacted peptide. The two consecutive fractions of greatest activity eluting after unlabeled peptide were pooled for each iodination and combined with identical fractions from additional iodinations. Pooled, purified peptide was concentrated under N₂ purge to remove ACN and was diluted with a mixture of nine parts sterile saline (0.9%) and one part ethanol before storage at 16°C.

Biodistribution. Before peptide drug injections, the stock peptide drug solution was diluted with sterile saline to yield a solution for injection that contained 5% ethanol. Dilutions were prepared every 1 to 2 working days, and three 10-µl aliquots of each dilution were counted before rat injections. BIM-23190 and BIM-23197 in solutions used for subcutaneous injections were >95% pure and BIM-23014 was 90% pure as determined by reversed-phase HPLC with online radioactivity detection.

Tissue extractions. Selected tissues in which radioactivity was comparatively high at 30 min were noted. At the 60-min time point, these tissues were harvested and placed immediately on ice, and portions were quickly and thoroughly homogenized with a Tekmar Tissumizer (Tekmar Instruments, Cincinnati, OH) in 5 ml of cold extraction solvent per gram of tissue. Ten percent TFA in water was used as the extraction solvent for tissues obtained from rats injected with BIM-23190 and BIM-23197. For rats injected with BIM-23014, it was necessary to extract tissues with a solvent containing a higher proportion of organic component; therefore, these tissues were extracted with 75% ACN 25% H₂O. The homogenates were centrifuged for 20 min at 28,000 rcf, and the supernatants were separated and stored at 20°C. Later, the extracts were thawed, centrifuged to remove solid matter that formed during storage, pooled and concentrated on a Savant Speed Vac concentrator (Savant Instruments, Farmingdale, NY) before HPLC analysis to determine the proportion of intact peptide remaining.

Data analysis. Tissue radioactivity levels for each of the three peptide drugs studied are reported as averaged values from three to six animals per time point (mean ± S.E.M.) and are expressed as percent total cpm injected per gram of tissue or milliliter of plasma. These values were analyzed by one-way analysis of variance for each tissue at each time point. When a significant difference was found, mean values were subsequently compared by the Newman-Keuls test. Plasma terminal elimination half-life values were calculated after determination of the rate constant for elimination of radioactivity from plasma (k) by the procedure for first-order drug decay. All statistical tests were performed according to the methods of Tallarida and Murray (1987).

Radioligand binding assays. Membranes from CHO-K1 cells stably expressing the different hSSTR subtypes were prepared by cell homogenization in ice-cold 50 mM Tris-HCl buffer. The buffer for the binding assays consisted of 50 mM HEPES (pH 7.4), containing bovine serum albumin (10 mg/ml), MgCl₂ (5 mM), aprotinin (200 KIU/ml), bacitracin (0.02 mg/ml) and phenylmethylsulfonyl fluoride (0.02 mg/ml). For the hSSTR1, 3, 4 and 5 binding assays, 0.05 nM [¹²⁵I-Tyr¹¹]SRIF-14 was incubated at 37°C for 30 min in the presence of increasing concentrations of unlabeled SRIF analogs. For assay of hSSTR2 ligand binding, 0.05 nM [¹²⁵I-MK-678 was incubated at 25°C for 90 min in the presence of increasing concentrations of the unlabeled analogs. Immediately after incubation, samples were rapidly filtered through GF/C filters that had been presoaked with 0.3% polyethyleneimine. Each tube and filter were washed three times with portions of ice-cold assay buffer. Specific binding was defined as total radioligand bound minus that bound in the presence of 1 µM SRIF-14 (SSTRs 1, 3, 4 and 5) or 1 µM MK-678 for SSTR2.

	Results

Top Abstract Introduction Methods Results Discussion References

After subcutaneous injection of [¹²⁵I] BIM-23014, plasma radioactivity quickly rose to a maximum value at 5 min, which persisted until 20 min before declining. The time of peak plasma concentrations for the iodinated SRIF analogs BIM-23190 and BIM-23197, 30 min, was considerably longer, and the maximum plasma concentrations obtained were 1.7 to 1.9 times greater than that of BIM-23014 (fig. 1). To determine the level of plasma radioactivity as intact peptide drug, plasma extracts of rats killed 1 hr after peptide drug injection were analyzed by HPLC with on-line real-time radioactivity detection (fig. 2). Of the total radioactivity present in the plasma 1 hr after peptide drug injection, 10%, 86% and 81% remained as intact drug for BIM-23014, BIM-23190 and BIM-23197, respectively. Note that in the BIM-23014 plasma extract, most of the radioactivity eluted as peptide metabolite or metabolites at the column void volume. Plasma terminal elimination half-lives of 3.76 hr for BIM-23190 and 3.26 hr for BIM-23197 have been calculated, based on the data shown in figure 1. No half-life is presented for BIM-23014 because it was rapidly metabolized, and plasma radioactivity at 1 hr represents a preponderance of metabolites.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Plasma radioactivity time course after subcutaneous injection of 125I-labeled SRIF analogs in adult male rats. Blood was removed at the indicated times and immediately centrifuged (0.1 ml of plasma from each rat was counted). Each data point represents the mean ± S.E. from three to six rats.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   HPLC/Flo-One radiochromatograms of plasma extracts after subcutaneous injection of 125I-SRIF analogs. Plasma samples obtained 60 min after peptide drug injection were extracted with ACN. Extracts were pooled and lyophilized before HPLC analysis. A, BIM-23014. B, BIM-23190. C, BIM-23197. Arrows indicate elution time of intact peptide drug.

Comparative distributions of radioactivity to target tissues and urinary bladder (plus contents) for each of the three SRIF analogs are shown in figures 3 and 4. A clear relationship in distribution patterns was evident, with the tissue concentrations of BIM-23190 and BIM-23197 associated radioactivity being generally similar and the tissue concentrations of BIM-23014 associated radioactivity being generally dissimilar to the former SRIF analogs. BIM-23014, in accord with its short time to peak plasma concentration, also was delivered to tissues and organs more rapidly than the other analogs. Its time to peak tissue concentration was 20 min, with the exception of pituitary (10 min), stomach and small intestine (30 min) and urinary bladder (240 min). Maximum tissue concentrations of BIM-23190 and BIM-23197, in contrast, were attained at 30 to 120 min, with the only exception being BIM-23197, in which a maximum liver concentration was found at 20 min after drug injection.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Tissue distribution of 125I-labeled SRIF analogs at 5 through 30 min after peptide drug injection. Each plot represents the mean ± S.E of tissue samples from three to six individual rats. Values are expressed as the percentage total radioactivity injected per gram of tissue. Note the break in the x axis that separates the data on adrenals and urinary bladder from other tissues. Statistically significant differences between BIM-23190 or BIM-23197 and BIM-23014 (*P < .05, **P < .01) are noted. Statistically significant differences between BIM-23197 and BIM-23190 (P < .05, P < .01) are shown.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Tissue distribution of 125I-labeled SRIF analogs at 60 through 480 min after peptide drug injection. Each plot represents the mean ± S.E of tissue samples from three to six individual rats. Values are expressed as the percentage total radioactivity injected per gram of tissue. Note the break in the x axis that separates the data on adrenals and urinary bladder from other tissues. Statistically significant differences between BIM-23190 or BIM-23197 and BIM-23014 (*P < .05, **P < .01) are noted. Statistically significant differences between BIM-23197 and BIM-23190 (P < .05, P < .01) are shown.

Each of the peptide drugs and their metabolites were excreted in urine; therefore, the urinary bladder and contents were highly concentrated in radioactivity. For tissues other than the bladder, BIM-23014-associated radioactivity concentrates maximally in the small intestine, followed by the liver, adrenals and kidney. The relative tissue affinities were considerably different for BIM-23190 and BIM-23197. Maximal tissue concentrations of these drugs were found in the adrenal gland, followed by the kidney, pituitary and pancreas. For the adrenal, kidney, pituitary, pancreas, seminal vesicles and stomach, radioactive content after injection of BIM-23190 or BIM-23197 was several-fold to many fold higher at most time points than that obtained on injection of BIM-23014. Liver radioactivity was similar for all three analogs, but BIM-23014 radioactivity was more than double that observed for the other analogs in the small intestine at 30 and 60 min.

In the pituitary, it is notable that BIM-23014 concentration was maximal at 10 min and then rapidly declined, whereas for BIM-23190 and BIM-23197, peak radioactivity was delayed to 60 min (BIM-23190) or 120 min (BIM-23197) but persisted for a long period. At 480 min, pituitary radioactivity was 7.8%, 62.2% and 68.3% of the maximum for BIM-23014, BIM-23190 and BIM-23197 respectively. BIM-23190- and BIM-23197-associated radioactivity in stomach, pancreas and adrenals also was persistent, maintaining substantial concentrations until 240 min.

Urine and organ extracts were analyzed by HPLC to characterize the in vivo stability of the peptide drugs studied (table 2). BIM-23190 and BIM-23197 have much greater in vivo stability than BIM-23014 in plasma, urine, kidney, stomach and pancreas. The stability of each peptide drug analog is comparable in liver. The lower stability of BIM-23014 should be considered when examining the tissue distribution data, especially at longer time points.

Peptide biodistribution to additional organs and tissues are shown in tables 3 and 4. Distribution to the hypothalamus and to other regions of the brain was minimal. Little distribution of the SRIF analogs was also seen in the eyes, fat, muscle and testes. Substantial tissue radioactivity occurred in the caecum, large intestine, heart, lungs and vas deferens.

In the caecum, large intestine, heart and lungs, the patterns observed in figures 3 and 4 were repeated. Peak tissue concentrations for BIM-23014 were achieved at 20 min, whereas the other analogs peaked at 30 min. Radioactivity content for the vas deferens was maximal at 240 min for BIM-23014 and at 30 to 60 min for the other analogs. Significantly higher tissue distribution was achieved for BIM-23190 and BIM-23197 than for BIM-23014 in the caecum, large intestine and vas deferens. In the heart, BIM-23190 associated radioactivity was unique, concentrating to a much greater extent than either of the other analogs.

The comparative distribution pattern obtained for BIM-23014 in lung was unusual. At time periods of 120 min, BIM-23014 concentration was lower than that of either of the other analogs except at 20 min. At 20 min, BIM-23014 mean lung concentration was 4.1 and 3.5 times greater than that of BIM-23190 and BIM-23197, respectively.

For each peptide drug, radioactivity accumulated abundantly in gastrointestinal tract contents but was negligible in excreted feces except for the 0- to 8-hr time interval. The percentage of total cpm injected, which was recovered in GI contents, is shown in table 5. It is clear that BIM-23014-associated radioactivity accumulates in GI contents much more rapidly and extensively than is the case for the other analogs. At 8 hr, the percentage of BIM-23014-associated radioactivity present in the GI contents is about twice that observed for the other peptides, and at shorter times the differences are even greater. The percentage injected cpm in the entire excreted feces for the 0- to 8-hr interval was (mean ± S.E.M.) 3.41 ± 3.20, 0.144 ± 0.048 and 2.83 ± 1.65 for BIM-23014, BIM-23190 and BIM-23197, respectively.

In the urine (table 6), radioactivity accumulated more rapidly and extensively after injection of either BIM-23190 or BIM-23197 than it did after injection of BIM-23014. For the 0 to 8 hr urine collections of BIM-23190- and BIM-23197-injected rats, recoveries of radioactivity were approximately three times greater than for BIM-23014, and these differences were even greater at 0 to 120 min. At the 0- to 8-hr time interval, radioactivity distributed equally between urine and GI contents for both BIM-23190- and BIM-23197-injected rats. However, for BTM-23014, there was a 6-fold greater amount of radioactivity in the GI contents than in the urine. The greater urinary excretion of BIM-23190 and BIM-23197 than that of BIM-23014 was also seen in the urinary bladder data (figs. 3 and 4).

In table 7, the binding potency of four SRIF analogs is compared for the five human SSTR subtypes. All four ligands were characterized by selectivity for hSSTR2, 3 and 5 subtypes. Within this group of subtypes, each analog consistently displayed an affinity rank order of hSSTR2 > hSSTR5 > hSSTR3. BIM-23190 and BIM-23197 had lower K_i values at hSSTR2 than the other analogs studied and displayed considerably greater selectivity for binding to the type 2 receptor over type 5 compared with the other analogs tested.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The peak tissue concentrations of a drug and its half-life after its subcutaneous injection will be influenced by its in vivo stability and its liability to rapid excretion. [¹²⁵I]BIM-23014 was rapidly absorbed after subcutaneous injection, quickly concentrated in the liver and then rapidly eliminated, presumably via biliary secretion into the small intestine. This is evident by the high recovery of radioactivity in GI contents at early time points. BIM-23014 is a relatively lipophilic peptide, and this likely contributes to its rapid biliary secretion. Iodination of the peptide drug increases its lipophilicity and may therefore further promote its biliary excretion. Rapid and high biliary clearance has also been observed when a related, radiolabeled (³H or ¹⁴C) SRIF analog, Sandostatin (Octreotide), was injected intravenously in rats. Ninety percent of biliary excretion occurred within the first 2 hours and accounted for 53% of the injected dose (Lemaire et al., 1989). Octreotide tissue concentrations were highest in kidney, liver and skin at 30 min after injection.

The possible linkage between SRIF analog lipophilicity and elevated biliary excretion has been examined by others. SRIF analogs are transported into the liver from blood in part via the bile acid transporter before biliary excretion. This is the same transporter that the hepatotoxic peptide phalloidin uses; thus, phalloidin-induced hepatotoxicity can be reduced by saturating the transporter with competing substrates such as SRIF analogs. When proline in a series of SRIF analogs was replaced with the more hydrophobic phenylalanine residue, IC₅₀ values decreased by nearly 5-fold. This finding was regarded as consistent with a role for hydrophobicity (Ruwart, 1995).

BIM-23190 [4-(2-Hydroxyethyl)-1-piperazinylacetyl-D-Phe-cyclo[Cys-Tyr-D-Trp-Lys-Abu-Cys]-Thr-NH₂] and BIM-23197 [4-(2-Hydroxyethyl)-1-piperazine-2-ethanesulfonyl-D-Phe-cyclo[Cys-Tyr-D-Trp-Lys-Abu-Cys]-Thr-NH₂] are closely related peptides that were obtained by modification of the amino terminus of another SRIF analog, BIM-23023 [D-Phe-cyclo[Cys-Tyr-D-Trp-Lys-Abu-Cys]-Thr-NH₂]. The intention was to reduce analog lipophilicity by increasing local hydrophilicity at the amino terminus and thereby diminish its biliary secretion (Moreau et al., 1996). In comparison with BIM-23014 [D-Nal-cyclo[Cys-Tyr-D-Trp-Lys-Val-Cys]-Thr-NH₂] BIM-23190 and BIM-23197 appear to have reduced biliary secretion as judged by decreased radioactivity recovered in GI contents and a corresponding increase in peak plasma peptide levels. Furthermore, their much longer in vivo stability helps to prolong plasma and tissue concentrations of intact drug.

Important factors affecting the tissue distribution patterns of an SRIF analog will be its relative affinity for the different somatostatin receptor subtypes as well as the tissue distribution and densities of the SRIF receptor isoforms. The three SRIF analogs examined in the biodistribution study displayed negligible binding to hSSTR1 and 4 but bound to hSSTR2 with high affinity, displaying subnanomolar K_i values. BIM-23014 was 7-fold more selective for binding to the human type 2 receptor than type 5, but with BIM-23190 and BIM-23197 this selectivity was further enhanced, achieving selectivity ratios of 32 and 52 for hSSTR2/hSSTR5 binding. These data suggest that type 2 and type 5 receptors may substantially control in vivo distribution of these analogs. Although comprehensive binding data for these SRIF analogs to rat SSTRs have not been reported, it may be that binding affinities for the rat receptors are similar to those obtained for the human receptors because rat and human SSTRs have high sequence homologies. Sequence homology between human and rat for equivalent receptor subtypes ranges from 82% to 94% for the five receptors (Patel et al., 1995). Indeed, K_i values for BIM-23014 binding to rat and human SSTR2 are 0.34 and 0.86 nM, respectively (Coy and Taylor, 1996).

SRIF receptor subtype distribution in brain and periphery has been examined by numerous investigators. Yamada et al. (1992), examined the distribution of SRIF receptor subtypes SSTR1 and SSTR2 in human tissues and found SSTR1 mRNA highest in jejunum, stomach and pancreatic islet cells, with lower amounts present in colon, colon carcinoma and kidney. Human SSTR2 mRNA was high in cerebrum and kidney, whereas lower levels were found in jejunum, colon, colon carcinoma, liver, hepatoma and pancreatic islets. Adrenal, pituitary and pancreas were not examined. mRNA from each receptor type has been found in the human adrenal, with SSTR2 and SSTR4 being most abundant (Epelbaum et al., 1995). SSTR5 mRNA is low in abundance in human tissues, but O'Carroll et al. (1994), using a sensitive RT-PCR procedure, were able to demonstrate its presence in heart, small intestine, adrenal, pituitary, cerebellum, skeletal muscle and placenta.

A useful compilation of SSTR isoform distribution in the rat is that of Patel et al., 1995 (table 8). In the periphery, SSTR2 is very highly expressed in adrenals, pancreatic islets and pituitary and present at lower levels in kidney and several other organs. SSTR4 is unique in its abundant expression in heart. SSTR5 mRNA is prominent in the pituitary and small intestine and detectable in pancreatic islets and spleen. Tissues notable in that only one SSTR isoform was found are lung (SSTR4) and liver (SSTR3). Other investigators, however, have obtained conflicting results. Raulf et al. (1994) have found SSTR types 1, 3 and 4 in rat lung and all except the SSTR2 isoform in rat liver, and they did not detect SSTR4 in rat heart. In rat pituitary, the rank order of expression of the subtypes is SSTR2 > SSTR1 = SSTR3 > SSTR5 > SSTR4 (Bruno et al., 1993). Pancreatic islets express SSTR2 > SSTR1 = SSTR4 > SSTR3 = SSTR5 (Patel et al., 1995).

BIM-23014 was less stable in vivo than the other SRIF analogs we tested. Ten percent of radioactivity in the 1-hr plasma extract eluted as authentic BIM-23014 on HPLC. Relatively lower stability of BIM-23014 was also found in tissue extracts; therefore, the earlier time points of 10 to 30 min are optimal for examining distribution of intact BIM-23014. During this time interval, peak tissue concentrations are found in the small intestine (1.23%/g, 30 min), liver (0.97%/g, 20 min), adrenals (0.79%/g, 20 min), kidney (0.70%/g, 20 min), pituitary (0.49%/g, 10 min) and stomach (0.44%/g, 30 min). For BIM-23190 and BIM-23197 at 20 to 30 min after injection, both the rank order of tissue distribution and SRIF analog concentration in the tissues are much different than those found for BIM-23014. The newer SRIF analogs had a rank order of distribution to tissues as adrenals > kidney > pituitary = pancreas > stomach > liver. Distribution to these tissues with the exception of the liver was 3-fold to many fold higher (P < .01) than that of BIM-23014.

The preferential distribution of BIM-23190 and BIM-23197 to adrenal, pituitary and pancreas correlates much more closely to the abundance of SSTR2 mRNA in these tissues than does the distribution of BIM-23014 to the same tissues. This correlation, however, does not indicate exclusive binding to this receptor isoform. The prolongation of high tissue radioactivity in these tissues is noteworthy, especially in the pituitary, in which high binding was still evident at 8 hr. This prolonged binding in the pituitary is associated with prolonged biological activity. Moreau et al. (1996) observed potent inhibition in vivo (rat) of (D-Ala²-GRF)-stimulated GH release at 2 to 8 hr after the subcutaneous administration of BIM-23190 or BIM-23197. The ED₅₀ values for these two peptides at the 8-hr time period are 6-fold lower than those for Octreotide. However, the prolonged elevation of tissue radioactivity may be due in part to receptor internalization after ligand binding (Patel et al., 1996).

Brain content of each of the SRIF analogs was minimal. This reflects the limited permeability of the blood-brain barrier to peptide based drugs and is confirmed by the recent findings of Abbruscato et al. (1997), who used an in situ model to show lower blood-brain barrier permeability for CTAP, a SRIF analog, than for morphine. Moreover, the low SRIF analog content of brain may be due in part to a peptide transporter system at the capillary bed that can transport the SRIF analog RC-160 from brain to blood, as shown previously by Banks et al. (1994).

In lung, BIM-23014 was transiently elevated at 20 min to a value much higher than either of the other SRIF analogs. This elevated mean value of 1.5%/g appears to reflect two distinct receptor subpopulations. Lung radioactivity for each of four rats at this time point was <0.33%/g, whereas lung radioactivity for the other two rats was 3.0% and 5.0%/g. Rat lung ordinarily contains mostly SSTR4, with lesser amounts of SSTR3 and SSTR1. It is possible that the two rats with exceedingly elevated lung concentrations of BIM-23014 have a differential ratio of SSTR receptors.

The comparatively high levels of BIM-23190 and BIM-23197 associated radioactivity relative to that of BIM-23014 in heart was not anticipated. Rat heart expresses SSTR4 mRNA at a high level and SSTR1 and SSTR3 at moderate levels (Bruno et al., 1993; Patel et al., 1995). The SRIF analogs tested have negligible affinity for hSSTR4. Therefore, high BIM-23190 and BIM-23197-associated radioactivity levels in the rat heart may be due to binding to the SSTR3 receptor subtype.

In the present study, distribution of BIM-23190 and BIM-23197 to seminal vesicles and vas deferens was markedly elevated. The tissue harvested as seminal vesicles also includes the prostate gland. Rat prostate has been examined for its SSTR subtype content. Of the five subtypes studied, only SSTR3 was detected in the prostate (Raulf et al., 1994). Human primary prostate cancer reportedly expresses SSTR1 but not SSTR2 or SSTR3 mRNA (Reubi et al., 1996). Nevertheless, BIM-23014, which preferentially binds to SSTR2, 3 and 5 subtypes, inhibits prostate tumor growth (subcutaneous administration) in a rat prostate tumor model (Bogden et al., 1990a), inhibits prostate tumor growth in nude mice bearing human prostate cancer xenografts when topically applied on the wound (Bogden et al., 1996) and, when BIM-23014 was administered by intramuscular injection into men with prostate carcinoma, it inhibited prostate tumor growth (Maulard et al., 1995).

The increased plasma levels of BIM-23190 and BIM-23197 relative to BIM-23014, enhanced in vivo stability and greater selectivity for SSTR2 receptors resulted in a much greater distribution of these two newly developed peptide drug analogs to tissues containing SSTR2 receptors than was obtained with BIM-23014. The markedly increased and prolonged distribution of BIM-23190 and BIM-23197 relative to BIM-23014, in pituitary, adrenal, pancreas and prostate tissues suggest a role for these new SRIF analogs in the treatment of cancers common to these tissues and in the treatment of acromegaly. Further studies are necessary to evaluate these potential clinical applications, as well as applicability in the treatment of small cell lung cancers. It is worth noting that based on pharmacokinetic considerations, long-acting sustained release preparations of BIM-23014 were designed and used (Moreau et al., 1996; Morange et al., 1994). Perhaps this will not be necessary with the new generation of novel and stable SRIF analogs, BIM-23190 and BIM-23197.