Deregulated hepsin protease activity confers oncogenicity by concomitantly augmenting HGF/MET signalling and disrupting epithelial cohesion.

Hepsin belongs to a family of cell-surface serine proteases, which have sparked interest as therapeutic targets because of the accessibility of extracellular protease domain for inhibitors. Hepsin is frequently amplified and/or overexpressed in epithelial cancers, but it is not clear how enhanced hepsin expression confers a potential for oncogenicity. We show that hepsin is consistently overexpressed in more than 40% of examined breast cancers, including all major biological subtypes. The effects of doxycycline-induced hepsin overexpression were examined in mammary epithelial organoids, and we found that induced hepsin acutely downmodulates its cognate inhibitor, hepatocyte growth factor (HGF) activator inhibitor type 1 (HAI-1). Hepsin-induced depletion of cellular HAI-1 led to a sharp increase in pericellular serine protease activity. The derepressed hepsin proteolytically activated downstream serine proteases, augmented HGF/MET signalling and caused deterioration of desmosomes and hemidesmosomes; structures important for cell cohesion and cell-basement membrane interaction. Moreover, chronic induction of hepsin considerably shortened the latency of Myc-dependent tumourigenesis in the mouse mammary gland. The serine protease and uPA system inhibitor WX-UK1, identified as a micromolar range hepsin inhibitor, prevented hepsin from augmenting HGF/MET signalling and disrupting desmosomes and hemidesmosomes. The findings suggest that the oncogenic activity of hepsin arises not only from elevated expression level but also from depletion of HAI-1, events which together trigger gain-of-function activity impacting HGF/MET signalling and epithelial cohesion. Thus, hepsin overexpression is a major oncogenic conferrer to a serine protease activity involved in breast cancer dissemination.

MCT1, MCT4 and CD147 gene polymorphisms in healthy horses and horses with myopathy.

Polymorphisms in human lactate transporter proteins (monocarboxylate transporters; MCTs), especially the MCT1 isoform, can affect lactate transport activity and cause signs of exercise-induced myopathy. Muscles express MCT1, MCT4 and CD147, an ancillary protein, indispensable for the activity of MCT1 and MCT4. We sequenced the coding sequence (cDNA) of horse MCT4 for the first time and examined polymorphisms in the cDNA of MCT1, MCT4 and CD147 of 16 healthy horses. To study whether signs of myopathy are linked to the polymorphisms, biopsy samples were taken from 26 horses with exercise-induced recurrent myopathy. Two polymorphisms that cause a change in amino acid sequence were found in MCT1 (Val(432)Ile and Lys(457)Gln) and one in CD147 (Met(125)Val). All polymorphisms in MCT4 were silent. Mutations in MCT1 or CD147 in equine muscle were not associated with myopathy. In the future, a functional study design is needed to evaluate the physiological role of the polymorphisms found.

Effects of training on equine muscle fibres and monocarboxylate transporters in young Coldblooded Trotters.

REASONS FOR PERFORMING STUDY: Muscular changes caused by training are breed-specific and studies on the Norwegian-Swedish Coldblooded Trotter (NSCT) are limited. Knowledge about lactate-transporters in muscle in this light draught breed used for harness racing is lacking. OBJECTIVES: To identify muscular changes associated with training in young NSCTs and investigate muscular distribution of the monocarboxylate transporter 1 (MCT1) and its ancillary protein CD147, which facilitate lactate transport across membranes. METHODS: Nine horses were followed from the start of their training period until the end of their 3-year-old season. A biopsy sample of the middle gluteal muscle was collected on 4 occasions. On the last 3 sampling occasions, individual V(La4)-values (the speed corresponding to a blood lactate concentration of 4 mmol/l) were determined in an incremental exercise test on a high-speed treadmill. One horse was excluded due to lameness. Histochemical and immunohistochemical analyses were performed on all muscle samples to determine fibre types (I, IIA, IIAX, IIX), oxidative capacity (NADH) and the expression of MCT1 and CD147. The activity of selected metabolic enzymes in the muscle before and after training was determined. RESULTS: The percentage of type IIX fibres decreased with training while the percentage of type IIAX fibres increased. The activity of citrate synthase and 3-OH-acyl-CoA-dehydrogenase increased with training. The expression of MCT1 was lower in membranes and cytoplasm of type IIX fibres compared to all other fibre types both before and after training. The antibody against CD147 stained membranes and cytoplasm of all fibres. The first V(La4)-value was lower than the last 2 in all horses. CONCLUSIONS: Muscular changes with training of NSCTs were similar to those reported in Standardbreds, indicating fibre type transitions and increased oxidative capacity. Expression of MCT1 differed among fibre types and was related to the oxidative capacity of the fibres.

Identification of inhibitors of the nicotine metabolising CYP2A6 enzyme--an in silico approach.

The compulsive nature of tobacco use is attributable to nicotine addiction. Nicotine is eliminated by metabolism through the cytochrome P450 2A6 (CYP2A6) enzyme in liver. Inhibition of CYP2A6 by chemical compounds may represent a potential supplement to anti-smoking therapy. The purpose of this study was to rationally design potent inhibitors of CYP2A6. 3D-QSAR models were constructed to find out which structural characteristics are important for inhibition potency. Specifically located hydrophobic and hydrogen donor features were found to affect inhibition potency. These features were used in virtual screening of over 60,000 compounds in the Maybridge chemical database. A total of 22 candidate molecules were selected and tested for inhibition potency. Four of these were potent and selective CYP2A6 inhibitors with IC(50) values lower than 1 muM. They represent novel structures of CYP2A6 inhibitors, especially N1-(4-fluorophenyl)cyclopropane-1-carboxamide. This compound can be used as a lead in the design of CYP2A6 inhibitor drugs to combat nicotine addiction.

MCT1 and CD147 gene polymorphisms in standardbred horses.

REASONS FOR PERFORMING STUDY: Transport of lactate across membranes is facilitated by proton-monocarboxylate transporters (MCT). The most widely distributed isoform is MCT1, which needs an ancillary protein CD147. Studies on erythrocytes have shown that high activity of MCT1 is inherited as the dominant allele and that activity is regulated through CD147. Mutations of human MCT1 have been described that appear to impair lactate transport in muscles and cause exertional rhabdomyolysis. There are no reports of this potential relationship in the horse. OBJECTIVES: To obtain sequences of equine MCT1 and CD147 to examine differences between horses with high and low lactate transport activity in their erythrocytes. METHODS: Muscle biopsy samples were taken from 3 healthy Standardbred horses and from 7 horses which according to the owners had signs of myopathy after intense exercise. DNA and RNA were isolated and PCR analysis and sequencing performed. RESULTS: Currently, PCR fragments covering 100% of MCT1 and 70% of CD147 coding region are retained and sequence analysis has demonstrated one single nucleotide polymorphism (SNP) in the C-terminal area of MCT1 and one SNP in the extracellular domain of CD147. Both cause an amino acid change. The SNPs found are not related to lactate transport activity in erythrocytes or signs of myopathy. CONCLUSIONS: More samples need to be analysed to make conclusions on the significance of the polymorphisms found. Furthermore, full sequence coverage of the coding region of CD147 is needed. POTENTIAL RELEVANCE: The molecular probes produced could be used as tools to study gene regulation of lactate transport.

Monocarboxylate transporters (MCT) as lactate carriers in equine muscle and red blood cells.

REASONS FOR PERFORMING STUDY: Monocarboxylate transporters (MCT) facilitate the transport of lactate across membranes. In red blood cells (RBC) the transport activity varies interindividually due to differences in the amount of an ancillary protein CD147. Similar variations in muscles could have a great influence on lactate accumulation during exercise. OBJECTIVES: To study the expression of MCT isoforms and CD147 in the middle gluteal muscle. METHODS: Venous blood and muscle biopsy samples were taken from 14 Standardbred horses. Lactate transport activity in RBC and the amounts of MCT1, 2, 4 and CD147 were measured. RESULTS: In muscle MCT1, MCT4 and CD147 were found. Amount of MCT1 was variable and not dependent on age or training. Expression of MCT4 increased with age and correlated positively with CD147. CD147 in muscle correlated with that in RBC. MCT4 in muscle and CD147/MCT1 in RBC were higher in race fit than in moderately trained horses. CONCLUSIONS: MCT isoform profile in equine muscle is similar to that in man. The correlation between CD147 in muscle and RBC supports the view that lactate transport activity in muscles may vary interindividually as with RBC. POTENTIAL RELEVANCE: A larger number of horses need to be analysed to confirm the relationship of CD147 in muscle and RBC; and to allow the use the lactate transport activity in RBC as an indicator of the respective activity in muscles.

More potent inhibition of human CYP2A6 than mouse CYP2A5 enzyme activities by derivatives of phenylethylamine and benzaldehyde.

1. A rapid 96-well plate assay method was developed and validated to measure liver microsomal coumarin 7-hydroxylation in vitro. 2. The method was used to test inhibition of human and mouse CYP2A enzymes by three phenylethylamine derivatives 2-(p-tolyl)-ethylamine, amphetamine, 2-phenylethylamine and benzaldehyde, and two of its derivatives, 4-methylbenzaldehyde and 4-methoxybenzaldehyde. 3. The benzaldehyde derivatives were more potent inhibitors of CYP2A5 than the phenylethylamines. The K(ic) value of 4-methylbenzaldehyde was 3.4 micro M and for 4-methoxybenzaldehyde it was 0.86 micro M for CYP2A5. 4. Amphetamine is a weak inhibitor of CYP2A6, whereas benzaldehyde is a suicide inhibitor with K(inact) = 0.16 min(-1) and K(I) = 18 micro M. The K(ic) values of 2-phenylethylamine, 2-(p-tolyl)-ethylamine, 4-methylbenzaldehyde and 4-methoxybenzaldehyde were 1.13, 0.23, 0.36 and 0.73 micro M for CYP2A6, respectively. 5. Novel potent inhibitors were found for CYP2A6 and, except for 4-methoxybenzaldehyde, all the compounds inhibited CYP2A5 and CYP2A6 enzymes differentially. These data add to the refinement of CYP2A enzyme active sites and provide chemical leads for developing novel chemical inhibitors of the CYP2A6 enzyme.

Effect of exercise on plasma ferritin concentrations: implications for the measurement of iron status.

Iron is of key importance for aerobic metabolism, and natural feeds of the horse are fairly rich sources of iron. Accordingly, the known incidence of iron deficiency anaemia is apparently rare in performance horses; despite this, iron deficiency in performance horses continues to be of concern to trainers and veterinarians. Effects of exercise on plasma ferritin concentrations were therefore studied in Standardbreds, Finnhorses and half-bred riding horses. Blood samples were taken after a moderate exercise test on a racetrack, a competition exercise test on a treadmill and a race. Even moderate exercise caused an increase in plasma ferritin concentration, with the increase being greater as the intensity and duration of exercise increased. Return to the basal level was slower after maximal-intensity exercise than after moderate exercise. In conclusion, although ferritin is a useful marker of low iron stores, samples should be taken only after at least 2 days rest following strenuous exercise.

Induction of heat shock protein 72 mRNA in skeletal muscle by exercise and training.

In response to stress, cells synthesise heat shock proteins (HSP) to maintain protein homeostasis. To study whether exercise and training induce expression of HSP72 in the middle gluteal muscle, 10 Finnhorses performed a submaximal 60 min exercise test on a treadmill. Test A was performed after 3 months of training, and the other two tests 2 (B) and 5 (C) weeks later. Blood samples were taken during and after the tests, and biopsy samples before, immediately after and 23 h after each test. HSP72 mRNA was analysed using a digoxigenin-labelled probe. Blood lactate concentration in the 3 tests varied between 7.2 and 10.2 mmol/l. Training increased HSP72 mRNA, as indicated by increases in samples taken at rest (A

Muscle fibre growth in undernourished reindeer calves (Rangifer tarandus tarandus L.) during winter.

To study whether moderate under-nutrition causes muscle wasting, reindeer (Rangifer tarandus tarandus L.) calves were fed either pelleted reindeer feed ad libitum (n=8) or restricted amounts of lichens (n=8). The restricted amount was 60% of ad libitum intake of lichens, and the feeding period was 6 weeks preceded by a 2-week adjustment period. Biopsy samples from the middle gluteal muscle (M. gluteus medius) for the analysis of fibre composition and area, as well as for the activity of cathepsin B were taken before the restriction period in November and January, and after the restriction period in April. In all calves the muscle fibre composition remained unchanged during the winter. In the lichen group, the fibre size also remained unchanged, whereas in control calves the cross sectional area of type I and type IIA fibres increased significantly from November to April. Cathepsin B activity decreased in all calves from November to January and remained at that low level for the rest of the study period, which suggests an attenuated rate of protein degradation. These results can be taken as an indication that moderate under-nutrition causes no muscle wasting in reindeer calves, and the decreased availability of nitrogen is partially compensated for by adaptive decrease in protein degradation. Interestingly the adaptive changes in protein metabolism are equally well seen in the well-fed controls as in the undernourished lichen-fed reindeer.

A comparative molecular field analysis of cytochrome P450 2A5 and 2A6 inhibitors.

Structure-activity relationships of 23 P450 2A5 and 2A6 inhibitors were analysed using the CoMFA and GOLPE/GRID with smart region definition (SRD). The predictive power of the resulting models was validated using five compounds not belonging to the model set. All models have high internal and external predictive power and resulting 3D-QSAR models are supporting each other. Both Sybyl and GOLPE highlight properties near lactone moiety to be important for 2A5 and 2A6 inhibition. Another important feature for pIC50 was the size of the substituent in the 7-positon of coumarin. The models suggest that the 2A5 binding site is larger that that of 2A6 due to larger steric regions in the CoMFA coefficient maps and corresponding GOLPE maps. In addition, the maps reveal that 2A6 disfavours negative charge near the lactone moiety of coumarin.

Pronounced differences in inhibition potency of lactone and non-lactone compounds for mouse and human coumarin 7-hydroxylases (CYP2A5 and CYP2A6).

1. The structural requirements for a compound to be a potent inhibitor for mouse CYP2A5 and human CYP2A6 enzymes catalysing coumarin 7-hydroxylase activity have been studied. 2. The IC50 of 28 compounds for the pyrazole-treated male DBA/2 mouse and human liver microsomal coumarin 7-hydroxylation were determined at 10 microm coumarin concentration 15 times over Km of coumarin. 3. The three most potent inhibitors for CYP2A5 were gamma-nonanoic lactone, gamma-decanolactone and gamma-phenyl-gamma-butyrolactone with an IC50 = 1.9+/-0.4, 2.1+/-0.2 and 2.4+/-0.3 microM and for CYP2A67-methylcoumarin, butylcyclohexane and indan with an IC50. = 30+/-3.2, 43+/-9 and 50+/-11 microM. 4. Among the 28 compounds studied, only 2-benzoxazolinone, 2-indanone and gamma-valerolactone showed similar inhibitory activity in both species. Indan had a lower IC50 for human than for mouse coumarin 7-hydroxylation, whereas the IC50 of 24 other compounds was higher for human than for mouse coumarin 7-hydroxylation. 5. The largest difference in IC50 between mouse and human activity was observed with 5-substituted phenyl, pentyl, hexyl, heptyl or octyl gamma-lactones or 6-substituted delta-lactones. IC50 of gamma-undecanolactone and gamma-decanolactone was 500 times lower for mouse than human coumarin 7-hydroxylation. 6. The difference in the IC50 between human and mouse coumarin 7-hydroxylation decreased substantially with the corresponding compounds without the lactone ring. 7. It is concluded that certain 5- or 6-position substituted gamma- and delta-lactones are potent inhibitors for mouse CYP2A5 but not for the orthologous human CYP2A6 and that the active site of CYP2A6 could be smaller than the active site of CYP2A5.

Metabolic response in skeletal muscle fibres of standardbred trotters after racing.

Histochemical and biochemical analyses were performed on muscle biopsies obtained after racing from the gluteus muscle of 18 standardbred trotters. Fibre type composition and enzyme activities varied among the horses. The percentage of type IIB fibres showed a positive correlation to the lactate dehydrogenase activity and a negative correlation to the citrate synthase activity. ATP concentrations in whole muscle after racing showed a negative correlation to both lactate and IMP concentrations. Within individual fibres, ATP concentrations varied markedly, with some type II fibres having values as low as 1-5 mmol/kg d.w. and some fibres having values as high as 40-58 mmol/kg d.w., whereas mean ATP concentration for whole muscle was 18.3 +/- 7.7 mmol/kg d.w. Some fibres with low ATP concentrations revealed high IMP concentrations. Blood samples taken after racing showed high values for lactate, ammonia, and uric acid in plasma. Muscle AMP and ADP concentrations after racing were related to the horses placing in a race, with higher concentrations giving a lower placing. The results of this study show that adenine nucleotide breakdown in muscle is of great importance for energy release during racing, and that ATP and IMP concentrations may very markedly among individual fibres. Thus, metabolite analyses on whole muscle must be evaluated with caution, as this only represents a mean value for metabolic responses in different fibres during racing.

Resynthesis of glycogen in skeletal muscle from standardbred trotters after repeated bouts of exercise.

OBJECTIVE: To determine glycogen resynthesis rate and changes in plasma metabolite concentrations in horses before and after repeated exercise. ANIMALS: 6 clinically normal Standardbred trotters. PROCEDURE: Horses trotted distances of 3,000, 3,000, and 2,000 m (trial A) and 3 days later, trotted 2,100, 2,100, and 1,600 m (trial B). Horses had 1 hour rest periods between bouts of exercise. Trotting speed was increased with each exercise bout, up to a near maximal. Muscle biopsy specimens and venous blood samples were obtained before each trial and 0, 4, 24, 48, and 72 hours after the third bout. Blood samples were also taken between exercise bouts. Muscle glycogen content and plasma glucose, glycerol, nonesterified fatty acid, and triglyceride concentrations were determined. RESULTS: Muscle glycogen content was significantly decreased immediately after exercise from 473 +/- 45 to 329 +/- 79 mmol/kg of dry weight in trial A, and from 472 +/- 128 to 347 +/- 59 mmol/kg in trial B. Further decreases were measured 4 hours after exercise. Glycogen resynthesis was negligible 24 hours after exercise. Basal muscle concentrations of glycogen were obtained 72 hours after exercise in trial A (472 +/- 128 mmol/kg), but not in trial B (279 +/- 52 mmol/kg). Plasma concentrations of glucose were greater than or equal to before-exercise values. Plasma concentrations of lipid metabolites, glycerol, triglycerides, and nonesterified fatty acids, were less than before-exercise values 2 to 72 hours after exercising. CONCLUSIONS: Repeated bouts of exercise decrease glycogen repletion rate, which is not attributable to hypoglycemia, but may be influenced by limited availability of lipids for energy production.

Androgen receptors and skeletal muscle composition in trotters treated with nandrolone laureate.

To study the effects of nandrolone laureate (19-nortestosterone) on muscle hypertrophy and concentration of androgen receptors (AR), biopsy specimens were taken from the middle gluteal muscle of 6 Finnhorse trotters (geldings and mares) undergoing training before, immediately after, and 13 weeks after a 14-week treatment with nandrolone. Another 6 similarly trained horses served as controls. An additional 10 mares and 10 geldings were used to study annual variation in muscle concentration of AR. AR was immunohistochemically localized in the nuclei. AR concentration remained constant during the first 14 weeks of the study, but increased significantly during the 13-week follow-up period in both groups. This finding can be explained by the annual variation in AR. In the anabolic steroid (AS)-treated horses, but not in the controls (C), the cross-sectional area of the type I fibres increased significantly during the treatment period, and the percentage of type IIA fibres correlated positively with AR concentration at the end of nandrolone treatment. In the AS group, the concentration of DNA decreased during the 13-week follow-up period, and the percentage of H-chains in the isoenzymes of LDH increased. Protein concentration increased in both groups during the follow-up period. Glycogen content and the activity of citrate synthase in muscle during the study remained unchanged. It can thus be concluded that AS produce differing effects on type I and type II fibres, and the AR concentration in equine muscle may contribute to the change observed in the middle gluteal muscle.

Exercise-induced changes in the activities of beta-glucuronidase and N-acetyl-beta-D-glucosaminidase in plasma and muscle of standardbred trotters.

The activities of lysosomal enzymes, such as beta-glucuronidase and N-acetyl-beta-D-glucosaminidase, have been shown to increase in muscle after endurance exercise. We examined whether measurable activities of lysosomal enzymes are present in equine plasma and whether the exercise-induced changes in the muscle are reflected in plasma. Six trained Standardbred trotters performed three exercise bouts with 1 h intervals and the same procedure was repeated 3 days later. Venous blood samples and muscle biopsies from the middle gluteal muscle were taken before and after exercise. The activities of beta-glucuronidase and N-acetyl-beta-D-glucosaminidase were measured both from plasma and muscle specimens. Cell infiltration into the muscle after exercise was evaluated by the DNA content and histochemically by haematoxylin stain. The activity of N-acetyl-beta-D-glucosaminidase in plasma was increased immediately after exercise, but had returned to the basal level at 4 h. N-acetyl-beta-D-glucosaminidase in muscle and beta-glucuronidase in muscle and plasma increased 2 days after exercise and returned to the basal level on day 3. A similar pattern was seen when the exercise protocol was repeated 3 days later, except that the activities continued to increase during the 3 days after exercise. The DNA content in muscle correlated with beta-glucuronidase in muscle and plasma and with the N-acetyl-beta-D-glucosaminidase in muscle indicating that the activities reflect the infiltration of phagocytes into the exercise-injured muscle. It can be concluded that the activities of the lysosomal enzymes in plasma increase after exercise and that the changes are mainly due to a simultaneous increase in the number of neutrophils. Therefore, plasma activities of the lysosomal enzymes are poor indicators of exercise-induced muscle damage.

An empirical and theoretical study on mechanisms of mutagenic activity of hydrazine compounds.

Hydrazine compounds are important industrial and laboratory chemicals. Many of them are carcinogenic in animal tests. Although the carcinogenicity is well established, the results of mutagenicity tests performed on alkylhydrazines vary greatly in different studies. In an attempt to clarify the situation we have applied Salmonella typhimurium TA102 tests to hydrazine and its mono- and dimethyl derivatives. These compounds were also tested by an Escherichia coli DNA repair-assay. The results of the repair tests indicate that unsymmetrically alkylated hydrazines can cause DNA-lesions which are lethal in repair-deficient strains. Finally QSAR (Quantitative Structure Activity Relationships) was used to develop a model to describe the genotoxic mechanism of hydrazine compounds, taking advantage of the results of previous mutagenicity studies. Energy of the lowest unoccupied molecular orbital together with octanol-water partition coefficient explains nearly completely the mutagenic activity of alkylated hydrazine compounds included in the analysis. The mutagenic activity of unsubstituted hydrazine is apparently based on different mechanisms.

Competitive inhibition of coumarin 7-hydroxylation by pilocarpine and its interaction with mouse CYP 2A5 and human CYP 2A6.

1. We have shown earlier that pilocarpine strongly inhibits mouse and human liver coumarin 7-hydroxylase activity of CYP 2A and pentoxyresorufin O-deethylase activity of CYP 2B in vitro. Since pilocarpine, like coumarin, contains a lactone structure we have studied in more detail its inhibitory potency on mouse and human liver coumarin 7-hydroxylation. 2. Pilocarpine was a competitive inhibitor of coumarin 7-hydroxylase in vitro both in mouse and human liver microsomes although it was not a substrate for CYP 2A5. Ki values were similar, 0.52 +/- 0.22 microM in mice and 1.21 +/- 0.51 microM in human liver microsomes. 3. Pilocarpine induced a type II difference spectrum in mouse, human and recombinant CYP 2A5 yeast cell microsomes, with Ka values of 3.7 +/- 1.6, 1.6 +/- 1.1 and 1.5 +/- 0.1 microM, respectively. 4. Increase in pH of the incubation medium from pH 6 to 7.5 increased the potency of inhibition of coumarin 7-hydroxylation by pilocarpine. 5. Superimposition of pilocarpine and coumarin in such a way that their carbonyls, ring oxygens and the H-7' of coumarin and N-3 of pilocarpine overlap yielded a common molecular volume of 82%. 6. The results indicate that pilocarpine is a competitive inhibitor and has a high affinity for mouse CYP 2A5 and human CYP 2A6. In addition the immunotype nitrogen of pilocarpine is coordinated towards the haem iron in these P450s.

Accumulation of allantoin and uric acid in plasma of exercising trotters.

Plasma concentrations of hypoxanthine, uric acid, and allantoin, which are breakdown products of adenine nucleotides, were measured in Standardbred and Finnhorse trotters during and after an exercise test on a high-speed treadmill, after an incremental exercise test performed on a racetrack, and after a racing competition. Fiber-type composition of the middle gluteal muscle and the muscle concentrations of adenine nucleotides and inosine monophosphate were measured after the racetrack test. Changes in the concentration of hypoxanthine were not observed in any of the tests. Peak concentration of uric acid was measured between 5 and 30 minutes after exercise, and it was three- to tenfold higher than the value at rest. The variability can be explained by intensity of the exercise test and variation among horses. The concentration of allantoin after exercise was 2 to 3 times as high as that at rest, depending on the intensity of the exercise, although the absolute increase was about 10 times as high as the increase in the concentration of uric acid. Peak values of allantoin for the treadmill and the racetrack tests were obtained 4 to 6 minutes after exercise and < 30 minutes after the races. Peak concentration of allantoin correlated positively with the percentage of type-II (IIA+IIB) fibers in the middle gluteal muscle. Significant correlations were not observed between plasma concentration of uric acid or allantoin and muscle concentrations of adenosine triphosphate (ATP) or inosine monophosphate. It can be concluded that in horses, breakdown of ATP during and after exercise continues until allantoin is produced.(ABSTRACT TRUNCATED AT 250 WORDS)