Pharmacokinetics of intranasally-administered dihydroergotamine in the rat

Intranasal dosing of dihydroergotamine (DHE) allows convenient self-administration and provides an alternate route of administration for the treatment of migraine in addition to the existing parenteral dosage forms. In this study, the pharmacokinetics of 3H-DHE were investigated following intravenous and intranasal dosing (0.343 mg DHE/animal) in the rat. Intranasal administration of DHE resulted in rapid absorption. The extent of absorption of the radiolabeled dose was approximately 45%-60%. Absolute bioavailability of the parent drug was 35%-40%, as determined by deconvolution and by the ratios of AUC0-infinity following intranasal and intravenous dosing. Due to the limited capacity of the nostrils, approximately half of the intranasal dose was swallowed into the gastrointestinal tract. Biliary excretion was found to be the predominant pathway of radioactivity excretion following both routes of administration. The results from this study suggest that intranasal administration provides a viable means of delivering DHE into the systemic circulation.

Luting cement-metal surface physicochemical interactions on film thickness

Low film thickness is critical to the clinical success of cemented castings. This study investigated the effect of luting agent-metal physico-chemical surface interactions on film thicknesses of representative luting agents. Control group luting agents were placed between two glass plates, as described by American Dental Association specifications 8, 61, and 66, and test group luting agents were positioned between glass and metal plates. The materials selected were zinc phosphate cement, polycarboxylate cement, glass ionomer cement, glass ionomer-composite resin hybrid cement and a resinous cement, with a type III gold alloy, a noble metal ceramic alloy, and a base metal ceramic alloy. A two-way analysis of variance and follow-up tests were done. The effects of the type of metal surface, type of cement, and their statistical interaction significantly affected film thickness (p < 0.0001). The type of cement had a greater affect on film thickness than the type of metal. A glass ionomer cement produced lower overall film thicknesses than other cement types, and a noble metal ceramic alloy created lower overall film thicknesses than other types of metal. American Dental Association specifications for cement film thickness did not accurately reflect normal cement use.

Diminished luminal release of esophageal epidermal growth factor in patients with reflux esophagitis

OBJECTIVES

It has recently been demonstrated that human esophageal mucosa, containing numerous submucosal mucous glands, has the ability to elaborate significant amounts of esophageal epidermal growth factor (eEGF). Because of its role in the maintenance of the integrity of the esophageal mucosa, we elected to study the rate of secretion of eEGF in patients with reflux esophagitis (RE), compared with controls, using our newly developed esophageal perfusion model.

METHODS

Fourteen healthy asymptomatic volunteers and 14 patients with endoscopically confirmed esophagitis underwent esophageal perfusion with saline, HCl (0.01 M, pH 2.1) HCl/pepsin (0.5 mg/ml of HCl), and ending NaCl solution during four consecutive 8-min perfusion periods. All perfusates were assayed for EGF by RIA (Amersham). Results are expressed as mean +/- SEM. Student's t test was used for statistical analysis.

RESULTS

The basal rate of luminal EGF release in patients with RE was 3.78 +/- 0.29 ng/min. This value significantly declined (2.27 +/- 0.27 ng/min; p < 0.001) during mucosal exposure to HCl but was significantly enhanced when the HCl perfusing solution was supplemented with pepsin (4.20 +/- 0.29; p < 0.001 vs. HCl). Introduction of saline during the last perfusion period maintained a rate of luminal EGF release similar to that observed during the initial esophageal perfusion with saline. Luminal release of EGF in patients with RE was significantly lower, compared with corresponding values recorded in controls during perfusion with saline (3.78 +/- 0.29 vs. 14.1 +/- 1.25 ng/min; p < 0.00001), with HCl (2.27 +/- 0.27 vs. 5.95 ng/min; p < 0.0001), with HCl/pepsin solution (4.2 +/- 0.29 vs. 11.7 +/- 1.88 ng/min; p < 0.0001), and during the final perfusion period with saline (3.73 +/- 0.25 vs. 15.1 +/- 1.1 ng/min; p < 0.00001). Therefore, the rate of luminal EGF release in controls was 4-fold, 3-fold, 3-fold, and 4-fold higher than that of patients with RE during perfusion with initial saline, HCl, HCl/pepsin and final saline, respectively.

CONCLUSIONS

1) Decreased esophageal EGF in patients with RE may facilitate the development or delay the healing of mucosal injury. 2) Depletion of EGF from the mucus layer covering the epithelium under the impact of refluxed luminal acid/pepsin may be considered as one of the potential underlying mechanisms leading to damage of the esophageal mucosa during gastroesophageal reflux episodes.

Depressed [Ca2+]i responses to isoproterenol and cAMP in isolated cardiomyocytes from experimental diabetic rats

Experimentally, diabetic rat hearts are characterized by diminished responses to beta-adrenergic stimulation. Among the aberrant responses are diminished beta-adrenoceptor number and depressed contractile protein activity. In this study, intracellular Ca2+ concentration ([Ca2+]i) was determined by microfluorescence in response to beta-adrenergic stimulation to understand the basis for the changes in the beta-adrenergic pathway in diabetic myocardium. In quiescent myocytes, isoproterenol caused a decrease in [Ca2+]i, which was blocked by timolol and thapsigargin. This suggests that the beta-agonist-induced [Ca2+]i changes are mediated, in part, by sarcoplasmic reticulum Ca-adenosinetriphosphatase. Diabetic myocytes showed a blunted response to isoproterenol, which was reversed by insulin treatment. In electrically stimulated myocytes, isoproterenol and 8-bromo-adenosine 3',5'-cyclic monophosphate (cAMP) increased [Ca2+]i and contraction in a concentration-dependent manner. Electrically stimulated diabetic myocytes demonstrated a depressed maximum [Ca2+]i response to isoproterenol and 8-bromo-cAMP without a change in sensitivity. These data suggest that in addition to alterations in beta-adrenoceptor function there are postreceptor defects in diabetic myocardium that may impair the regulation of [Ca2+]i in diabetic myocardium.

An ultra-sensitive electrochemical enzyme immunoassay for thyroid stimulating hormone in human serum

A sensitive heterogeneous electrochemical enzyme immunoassay has been developed for thyroid stimulating hormone (TSH) by modifying a commercially available two-site immunoenzymometric assay. p-Aminophenyl phosphate (PAPP) was used as the substrate of alkaline phosphatase, and hydrolysed to p-aminophenol (PAP). The amount of PAP produced from the assay was proportional to the amount of TSH in the sample. Detection of PAP was done by oxidative amperometry in a flow injection system. The working electrode was a glassy carbon electrode whose potential was held at +325 mV (vs Ag/AgCl). The amperometric detection of PAP required only 1 microliter of sample (the range of linearity: 50.0 fmol-100 pmol PAP, the limit of detection: 10.9 fmol PAP). Intra-assay precision over the assay range of linearity (0.02-60 mIU l-1 or 0.02-60 pIU TSH) showed a maximum RSD of 8.0%, and a low detection limit of 0.01 mIU l-1 or 0.01 pIU TSH. The study also indicates that this two-site electrochemical enzyme immunoassay correlates well with the Bio-Rad's immunoradiometric assay currently used in our medical center (r = 0.992, slope = 1.53, n = 43) and a highly sensitive immunochemiluminometric assay in the Nichols Institute (r = 0.986, slope = 0.499, n = 23).

Cellular functions of diabetic cardiomyocytes: contractility, rapid-cooling contracture, and ryanodine binding

To study the mechanisms of cardiac dysfunction in experimental diabetes, adult rat cardiomyocyte shortening (measured with a video edge-detector system), the sarcoplasmic reticulum (SR) Ca2+ content [assessed by rapid-cooling contracture (RCC) and caffeine contracture (CC)] was examined. Ryanodine binding to the SR Ca(2+)-release channel of myocardium homogenate was also studied. Myocytes from diabetic rats showed depressed shortening (44% decrease compared with controls), reduced maximum rates of shortening and relengthening (58 and 56% decrease, respectively), and prolonged time to peak shortening (47% increase). RCCs and CCs from diabetic cells were 68 and 75% of the control values, respectively. Most of these cardiomyocyte abnormalities were corrected by daily insulin treatment in the diabetic rats. Ryanodine binding parameters indicated that the number of high-affinity binding sites was decreased in diabetic hearts. These data suggest that changes in contractile parameters as measured in diabetic myocytes are in good agreement with data obtained from intact heart or cardiac tissue preparations. Decreased SR Ca2+ content and reduced ryanodine binding sites indicate that the SR functions of storage and release of Ca2+ were depressed. This consequently may cause depressed contraction in diabetic hearts.

Pharmacokinetics of SDZ 64-412, a novel antiasthmatic agent, following intravenous, oral, and inhalation dosing in the rat

The pharmacokinetics of SDZ 64-412, an antiasthmatic agent, were investigated following intravenous, oral, and inhalation dosing in rats. 14C-SDZ 64-412 was administered intravenously (2.75 mg kg-1) and orally (5.5 mg kg-1, 110 mg kg-1), whereas non-radiolabeled drug (5.04 mg kg-1) was administered using nose-only inhalation chambers. Radioactivity and parent drug concentrations in blood, lung, and excreta were determined at designated times post-dose. SDZ 64-412 was rapidly and extensively (approximately 80%) absorbed following both oral doses, although absorption appeared to be prolonged with increasing dose. The absorbed drug was shown to undergo extensive and saturable first-pass metabolism. The bioavailability of the parent drug, calculated by dose-normalized AUC and deconvolution methods, was only 10-15% from the low dose, but increased to approximately 40% following the high dose. Following inhalation dosing, SDZ 64-412 concentrations in blood and lungs increased rapidly, and did not decline immediately after termination of dosing. The inhalation dose yielded a bioavailability of approximately 40%, and AUC of the drug in lungs was approximately 25 times greater than in blood. In general, SDZ 64-412 was extensively distributed and rapidly eliminated from the systemic circulation. Biliary excretion was the predominant route of radioactivity excretion. The present findings suggest that inhalation administration provides a viable means of delivery of SDZ 64-412.

In vitro characterization of nonpeptide irreversible inhibitors of HIV proteases

The irreversible inhibition of human immunodeficiency virus type 1 (HIV-1) and type 2 (HIV-2) proteases by 1,2-epoxy-3-(p-nitrophenoxy)propane (EPNP) and eight haloperidol derivatives has been studied. EPNP specifically inhibits HIV-1 and HIV-2 proteases with a stoichiometry of one EPNP molecule/dimeric enzyme. The site of modification of HIV-2 protease by EPNP has been unambiguously identified as Asp-25 using high performance tandem mass spectrometry. The haloperidol derivatives assayed consist of epoxides, ynones, and alpha,beta-unsaturated ketones. The Kinact values for these haloperidol derivatives range from 10.7 to 521 microM for HIV-1 protease and from 8.6 to 283 microM for the HIV-2 enzyme, being in some cases approximately 1000-fold more potent irreverisble inhibitors of HIV proteases than EPNP. This potency results from the haloperidol character of the compounds and the chemical reactivity of the groups capable of forming a covalent bond with the enzyme. Covalent modification of HIV-2 protease by a radiolabeled epoxide derivative of haloperidol, UCSF 84, is prevented by EPNP and the peptidomimetic transition state analog U-85548. In similar experiments, incorporation of UCSF 84 into HIV-1 protease is partially prevented by these active-site inhibitors. In contrast, a mutant HIV-1 protease, HIV-1 PR C95M, in which Cys-95 has been replaced by Met, is labeled 50% less than HIV-1 protease and is fully protected by EPNP and U-85548. These results indicate the presence of 2 reactive residues in HIV-1 protease: Cys-95 and another located in the active site of the enzyme. The alpha,beta-unsaturated ketone derivative of haloperidol, UCSF 191, which is stable over a broad pH range, was used to study the pH profile of inactivation of HIV-1 and HIV-2 proteases. Comparison of the profiles of inactivation of wild-type HIV-1 protease, HIV-1 PR C95M, and HIV-1 PR C67L as well as HIV-2 protease (which has no cysteine residues) reveals the contribution of Cys-95 to the reactivity of these irreversible inhibitors. The inhibitors UCSF 70, UCSF 84, UCSF 115, UCSF 142, and UCSF 191 reduce p55gag polyprotein processing when assayed in a mammalian cell line that produces HIV-1 viral particles lacking the envelope.

Impact of acid and pepsin on human esophageal prostaglandins

OBJECTIVES

Although the prostaglandin-mediated mucosal protection within the gastric compartment has been well established, its potential role in the maintenance of integrity of the esophageal mucosa in humans has not been explored due to the lack of appropriate methodology.

METHODS

We have recently developed an esophageal perfusion catheter, equipped with two balloons, compartmentalizing a 7.5-cm segment of the esophageal lumen. Using this catheter, we studied the impact of the luminal perfusion with saline, HCl (0.01 M, pH 2.1), and HCl/pepsin solutions (0.5 mg/ml) on esophageal luminal release of PGE2 in 21 asymptomatic, presumably healthy volunteers (12 M, 9F; mean age 40 yr). The content of PGE2 in its methyl oximated form was measured by RIA (Amersham, IL), using a novel iodinated label. Results are expressed as mean +/- SEM. Student's t test was used for statistical analysis.

RESULTS

Perfusion of the esophageal lumen resulted in continuous release of PGE2 into the perfusate at the rate of 1880 +/- 393 pg/min during the first 8-min perfusion period. During continuation of perfusion with saline, the luminal release of PGE2 was maintained at the rate of 1820 +/- 640 pg/min during the second 8-min perfusion period. This rate declined (although in nonsignificant fashion; p < 0.2) during the third perfusion period, reaching a plateau of 1220 +/- 473 pg/min and maintained during the last (period IV) perfusion period with saline. Introduction of acid during the perfusion period II in the second group of investigated subjects resulted in a rapid and statistically significant decline of the luminal release of PGE2 to the value of 1020 +/- 167 ng/min (p < 0.01). Continuation of esophageal perfusion with acid during the next 8-min perfusion period further diminished the luminal release of PGE2 to the value of 520 +/- 73; p < 0.001. The significant decline in the rate of luminal PGE2 release was still maintained despite the replacement of acid with saline during the ending 8-min perfusion (period IV; 560 +/- 80 ng/min; p < 0.001). Esophageal perfusion with HCl/pepsin solution, in group III subjects, potentiated luminal release of PGE2, reaching the value of 1553 +/- 340 pg/min, which is 3 times higher than the value of PGE2 observed during corresponding perfusion with HCl (period III; p < 0.03). This significant impact of HCl/pepsin solution was still maintained despite the substitution of HCl/pepsin with NaCl during the last perfusion period, and was still significantly higher (1260 +/- 220 pg/min; p < 0.02) than the corresponding value during the ending perfusion with NaCl after HCl (group II). This study for the first time demonstrates that luminal release of PGE2 in humans remains under a significant impact of luminal chemical factors such as acid and pepsin.

CONCLUSION

The modulatory effect of acid and pepsin on esophageal mucosal prostaglandin release may play a role in the development of reflux-related mucosal pathology.

In vivo microleakage of luting cements for cast crowns

Standardized tooth preparations were completed on previously intact human molars in vivo, and castings were made with a precious metal ceramic alloy by conventional techniques. The castings were randomly assigned to the following luting agents: zinc phosphate, composite resin-glass ionomer hybrid, and a composite resin-glass ionomer hybrid with a dentinal bonding agent and were cemented in a standardized manner to periodontally compromised molars. After 6 months the teeth were carefully extracted, stained, embedded, and sectioned, and the in vivo microleakage was measured. ANOVA disclosed significant differences between groups, and a multiple comparisons test revealed that the zinc phosphate group leaked significantly more than other cement groups.

Determination of amide-type local anaesthetics by direct injection of plasma in a column-switching high-performance liquid chromatographic system using a pre-column with a semipermeable surface

A high-performance liquid chromatographic method using column switching was applied to the direct determination of two local anaesthetics, ropivacaine and bupivacaine, in human plasma. The method is intended to be used in a combined LC-GC system; here only the LC-part is described. After addition of internal standard, the samples were filtered and directly injected into a semipermeable surface (SPS) pre-column where the analytes were strongly retained and separated from many endogenous compounds by a short washing step. The retained analytes were transferred by a buffered methanol phase from the pre-column into a carbonaceous HPLC column and they were detected by UV detection at 254 nm. The SPS pre-column could withstand numerous (> 200) direct injections of plasma samples (10 microliters). The method has a detection limit of 8.2 ng and requires a total assay time of 15 min per plasma sample. Quantitative recoveries were obtained over the range 3.3-114 micrograms/ml with inter-day precisions of 1.6-5.2% (C.V.).

Impairment of salivary epidermal growth factor secretory response to esophageal mechanical and chemical stimulation in patients with reflux esophagitis

OBJECTIVES

It has been demonstrated recently that salivary epidermal growth factor (sEGF) output in healthy individuals is strongly and significantly influenced by esophageal intraluminal mechanical and chemical stimuli. Therefore, we have studied the impact of intraesophageal mechanical and chemical stressors on the rate of secretion of sEGF in 14 patients with reflux esophagitis (RE), and compared these results with corresponding parameters measured in 14 sex- and age-matched controls.

METHODS

EGF was assessed in saliva collected during basal conditions, chewing of parafilm, placement of esophageal tubing, inflation of intraesophageal balloons, and perfusion with NaCl, HCl, and HCl/pepsin solutions. The concentration of sEGF was measured with an RIA kit from Amersham (Arlington Heights, IL).

RESULTS

The concentrations of sEGF were (mean +/- SEM) 2.50 +/- 0.32 ng/ml and 2.00 +/- 0.37 ng/ml in basal saliva and during stimulation by chewing the parafilm, respectively. Basal sEGF value appeared to be significantly higher than in controls (2.50 +/- 0.32 vs. 1.90 +/- 0.22 ng/ml, p < 0.05, in one-tailed t test). Placement of intraesophageal tubing resulted in a significant decline of sEGF concentration, compared with parafilm-stimulated conditions (1.25 +/- 0.12 vs. 2.00 +/- 0.37 ng/ml, p < 0.0001) and corresponding tubing-stimulated sEGF value in controls (1.25 +/- 0.12 vs. 1.52 +/- 0.16 ng/ml, p < 0.05). sEGF concentrations after inflation of intraesophageal balloons and subsequent perfusion with initial saline, HCl, HCl/pepsin, and ending saline were also highly significantly lower (1.05 +/- 0.18 ng/ml, p < 0.001; 1.10 +/- 0.20 ng/ml, p < 0.001; 1.10 +/- 0.18 ng/ml, p < 0.001; 1.10 +/- 0.19 ng/ml, p < 0.001; and 1.05 +/- 0.18 ng/ml, p < 0.001, respectively) than sEGF concentration recorded during stimulation with parafilm. Concentrations of sEGF during esophageal perfusion with HCl, HCl/pepsin, and ending saline were also significantly lower than corresponding values in controls (1.10 +/- 0.18 vs. 1.49 +/- 0.11 ng/ml, p < 0.05; 1.10 +/- 0.19 vs. 1.59 +/- 0.11 ng/ml, p < 0.05; and 1.05 +/- 0.18 vs. 1.65 +/- 0.13 ng/ml, p < 0.01, respectively). The rate of sEGF output, which was 1.30 +/- 0.24 ng/min during basal conditions, increased significantly during stimulation with parafilm (2.30 +/- 0.38 ng/min, p < 0.05). Both basal and parafilm-stimulated sEGF outputs were somewhat higher, although nonsignificantly, than corresponding values recorded in healthy individuals. Mechanical and chemical stimulation (initial NaCl, HCl, and ending NaCl) failed to evoke a significant increase in sEGF output over the value observed during parafilm stimulation in patients with RE, although such a significant increase was clearly demonstrated in healthy individuals. Therefore, sEGF output in patients with RE remained significantly lower than corresponding values recorded in controls during an entire mechanical stimulation (2.65 +/- 0.35 vs. 4.60 +/- 0.85 ng/min, p < 0.001, after placement of intraesophageal tubing and 2.80 +/- 0.54 vs. 5.15 +/- 0.70 ng/min, p < 0.001, after inflation of balloons). sEGF output in patients with RE remained also significantly lower than adequate control values during chemical stimulation (3.65 +/- 0.64 vs. 5.20 +/- 0.60 ng/min, p < 0.05, during perfusion with initial saline; 3.70 +/- 0.70 vs. 5.20 +/- 0.60 ng/min, p < 0.05, during perfusion with HCl; 3.70 +/- 0.52 vs. 5.55 +/- 0.72 ng/min, p < 0.01, during perfusion with HCl/pepsin, and 3.30 +/- 0.56 vs. 5.80 +/- 0.86 ng/min, p < 0.001, during ending saline).

CONCLUSION

Impairment in sEGF secretion during mechanical and chemical intraesophageal stimulation, mimicking the natural scenario occurring during gastroesophageal reflux, may facilitate the development of esophageal mucosal pathology and delay the healing of already developed mucosal injury.

[Serological investigation on intrafamilial transmission of HCV infection]

To clarify the intrafamilial transmission of HCV infection, the antibody to HCV was assayed in 124 serum samples from the family members of 60 HCV-Ab (+) index cases (group C), and compared with that from 83 family members of 40 HCV-Ab (-) index patients with hepatitis B (group B), Nine of 124, including 2 parents, 6 spouse and 1 granddaughter, were positive for anti-HCV. The prevalence of anti-HCV was 7.3% (9/124) in total, and 1.85% (2/108) in the subjects who had no history of blood donation. None of 83 was positive for anti-HCV in group B. On the other hand, the prevalence of HBV infection was 25.33% in the group of HCV infection only, and 40.91% in the group with HBV/HCV double-infection or HBV infection. It indicates that the risk of HBV intrafamilial transmission is higher than that of HCV, and the risk of transmission of HCV from mother to infant may be lower than that of sexual transmission.

[Atrial specific granules are an intracellular calcium store]

In order to elucidate whether the secretion of atrial natriuretic peptide (ANP) is correlated to the release of calcium (Ca) from the atrial specific granules (ASG), it is necessary to explore whether the ASG contain highly concentrated Ca and how the high Ca is maintained in the ASG. The present study was designed to determine Ca in the ASG with the quantitative electron microscope X-ray microanalysis and Ca(2+)-ATPase with electron microscopic (EM) cytochemical technique. The ultrathin cryosections of rapid frozen fresh rat auricles were used for measuring Ca concentration with a JEM-1200EX electron microscope equipped with a Link AN 10,000 energy dispersive X-ray spectroscope. The measurement showed that the Ca concentration in the ASG was quite high, being 81 +/- 15 mmol/kg (n = 10), comparable with that found in the sarcoplasmic reticulum. With the Ca(2+)-ATPase EM cytochemical technique, the reaction products proved to be deposited on the membrane of the ASG. It was postulated that the Ca(2+)-ATPase on the membrane of ASG pumped Ca2+ out of the cytosol into the ASG and thus maintained a high Ca concentration inside the ASG. Therefore, the ASG might be considered to be a Ca store in atrial cardiocytes.

Peptide-antibody conjugates for tumour therapy: a MHC-class-II-restricted tetanus toxin peptide coupled to an anti-Ig light chain antibody can induce cytotoxic lysis of a human B-cell lymphoma by specific CD4 T cells

Anti-idiotype antibody therapy of B-cell lymphomas, despite numerous promising experimental and clinical studies, has so far met with limited success. Tailor-made monoclonal anti-idiotype antibodies have been injected into a large series of lymphoma patients, with a few impressive complete tumour remissions but a large majority of negative responses. The results presented here suggest that, by coupling to antilymphoma idiotype antibodies a few molecules of the tetanus toxin universal epitope peptide P2 (830-843), one could markedly increase the efficiency of this therapy. We show that after 2-hr incubation with conjugates consisting of the tetanus toxin peptide P2 coupled by an S-S bridge to monoclonal antibodies directed to the lambda light chain of human immunoglobulin, human B-lymphoma cells can be specifically lysed by a CD4 T-lymphocyte clone specific for the P2 peptide. Antibody without peptide did not induce B-cell killing by the CD4 T-lymphocyte clone. The free cysteine-peptide was also able to induce lysis of the B-lymphoma target by the T-lymphocyte clone, but at a molar concentration 500 to 1000 times higher than that of the coupled peptide. Proliferation assays confirmed that the antibody-peptide conjugate was antigenically active at a much lower concentration than the free peptide. They also showed that antibody-peptide conjugates required an intact processing function of the B cell for peptide presentation, which could be selectively inhibited by leupeptin and chloroquine.(ABSTRACT TRUNCATED AT 250 WORDS)

Chemically-modified tetracycline normalizes collagen metabolism in diabetic rats: a dose-response study

An earlier study indicated that a chemically-modified non-antimicrobial tetracycline (4-de-dimethylaminotetracycline; CMT-1) can inhibit excess collagenase activity in the connective tissues of diabetic rats, however, the optimum oral dose and resulting serum concentration were not determined. In the current study, adult male Sprague-Dawley rats (body weight approx. 350 g) were made diabetic by streptozotocin injection and administered by oral gavage either 0, 1, 2, 5, or 10 mg CMT-1 per day. After 3 weeks of drug therapy, the rats were killed and gingiva, skin, and serum collected. The tissues were 1) extracted, partially purified and analyzed for collagenase activity using [3H-methyl] collagen as substrate and SDS-PAGE/fluorography; 2) extracted in neutral salt and dilute acid solutions (4 degrees C) to assess collagen solubility; and 3) analyzed for hydroxyproline to determine tissue (skin) collagen mass. Serum was analyzed for glucose and CMT-1 concentration, the latter by HPLC. Inducing diabetes dramatically increased both gingival and skin collagenase activity and reduced skin collagen mass by 69.8%. Increasing the oral dose of CMT-1 progressively increased the serum concentration of the drug from 0.6-6.5 micrograms/ml and progressively decreased the excessive collagenase activity in gingiva and skin (p < 0.01 vs untreated diabetics). Although skin collagen mass tended to be increased at all oral doses of CMT-1, only the 5 mg dose effect was statistically significant (p < 0.01). The diabetes-induced reduction in collagen solubility, a classic abnormality (reflecting excessive collagen crosslinking) of this disease, was also normalized by CMT-1 therapy.(ABSTRACT TRUNCATED AT 250 WORDS)