Dietary Soybean Protein Decreases Plasma Taurine in Cats1»2

Transcription

1 Nutrient Metabolism Dietary Soybean Protein Decreases Plasma Taurine in Cats1»2 SUHGWK W. KIM, JAMS G. MRRIS3 AND QÅ NTN R. RGRS Department of Molecular Biosciences, School of Veterinary Medicine, University of California, Davis, CA ABSTRACT Commercial dry and canned diets fed to cats cause ~two- and fourfold increase in the taurine requirement, respectively, as compared with that ob served for purified diets. In two experiments, the effect of source and level of protein and fiber in the diet on the concentration of taurine in plasma and whole blood of cats was studied. All diets contained 1 g taurine/kg dry matter. When a casein-based diet containing either 25% or 50% protein was given to cats for 6 wk, no difference in plasma taurine concentration was ob served; however, substituting soybean protein for ca sein resulted in a significant (P < 0.01) decrease in plasma taurine concentration of cats in the 50% soy bean protein group, but not in the 25% soybean protein group. In xperiment 2, the food intake of cats was limited [26 g dry matter/(kg body weight â d), and the protein was 30 or 60% of the diet. Cats fed 60% soybean protein or casein diets had significantly lower plasma taurine concentrations than cats fed a 30% casein diet, with the 60% soybean protein diet causing the greater decrease. There was no effect of either 2 or 4% soybean fiber on plasma taurine concentration as compared with the same diet without the added fiber. The taurine con centration in plasma was higher (P < 0.05) in male cats than in female cats. Protein source, amount in the diet and gender did not affect the whole blood taurine concentration. Cats given diets containing 60% casein or soybean protein diets excreted a greater amount of fecal total bile acid and total taurine than cats given a 30% casein diet. Cats with higher plasma concentra tions of taurine excreted a greater amount of free taurine in urine, and a lesser amount of taurine and bile acids in feces. These results show that although protein source (soybean protein) and the quantity in the diet have a significant effect on the excretion pattern of tau rine by cats, these effects are not sufficient to account for the marked increase in the taurine requirement found when canned heat-processed diets are fed. J. Nutr. 125: , INDXING â cats â casein â taurine KY WRDS: soybean protein â fiber Taurine (2-aminoethanesulfonic acid) is synthesized from the sulfur amino acid, cysteine. Taurine is an essential dietary nutrient for cats (Hayes et al. 1975) because the limited rate of synthesis does not meet the rate of loss. The major pathway of taurine metab olism is conjugation with bile acids to form bile salts (Rabin et al. 1976). Cats use taurine to conjugate bile acids but, unlike many other mammals, cats and dogs cannot substitute glycine for taurine when taurine is limiting. No endogenous degradation of taurine has been demonstrated in mammalian tissues (Remtulla et al. 1977). xcess absorbed taurine is excreted in the urine. The obligatory demand for taurine to produce bile salts and the lack of complete recycling of taurine in the enterohepatic circulation result in a continual fecal loss from the body. arly occurrences of feline central retinal degeneration (Aquirre 1978, Hayes et al. 1975) were found in cats receiving commercial diets that were based largely on plant ingredients and were low in dietary taurine. More recently, very low plasma taurine concentrations have been found in cats fed heat-processed commercial diets presumably contain ing adequate amounts of taurine (Morris et al. 1990). The cause of taurine depletion in cats fed heat-pro cessed commercial diets has not been elucidated. Hickman et al. (1990) found that 9% of a pulse-labeled dose of 14C-taurine added to a heat-processed com mercial diet was recovered in the expired air of cats as 14C2, and suggested that the 14C2 was a product of the catabolism of taurine by intestinal bacteria. s timates of the bile salt turnover in humans are in the range of five to 15 times a day (Berry and Reichen 1983); thus even a small increase in the proportion of taurine degraded during each enterohepatic cycle 1Supported by a gift from Mark Morris and Associates, Topeka, KS. 2 The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact. 3 To whom correspondence should be addressed /95 S3.00  1995 American Institute of Nutrition. Manuscript received 4 August Initial review completed 29 August Revision accepted 29 June

2 2832 KIM T AL. would result in a substantial loss of taurine from the total pool. Soybean protein and dietary fiber are known to be hypocholesterolemic and affect bile acid metabolism in rats (Sugano et al. 1990); commercial cat foods vary in the quantity and quality of protein (both amino acid composition and digestibility). For these reasons, in this paper we describe experiments which examine the effect of soybean protein vs. casein and the addition of soybean fiber in the diet on the taurine status of cats. MATRIALS AND MTHDS Animals. Specific pathogen-free cats (mean ±SM body weight, 2763 ±74 g for female and 3336 ±87 g for male) from the Nutrition and Pet Care Center at the University of California, Davis were used. Both male and female cats were housed in individual me tabolism cages in a temperature-controlled room (22 ±3 C)with a 14-h light: 10-h dark cycle. Cats were maintained according to a protocol approved by the University of California, Animal Use and Care Ad ministrative Advisory Committee and according to the standards of the Guide for the Care and Use of Lab oratory Animals (1985) of the National Institutes of Health. Diets. In xperiment 1, pelleted type diets were used. The composition of the diets is shown in Table 1. For xperiment and K-carrageenan 2, gel type diets made from with different water contents starch were used to give diets of similar texture; 45% water was used in the 30% casein diets (0, 2 or 4% isolated soy bean fiber, abbreviated 30% Cas, SF2 and SF4, respec tively) and 55% water was used in the 60% casein (60% Cas) and 60% soybean protein (60% Soy) diets. The composition of diets used in xperiment 2 is shown in Table 2. All diets met or exceeded the requirements given by the National Research Council (1986). The taurine concentration was 1 g/kg dry matter for both experiments. For each experiment, a low casein diet with no fiber was used to accustom cats to a purified diet. xperimental design. In xperiment 1, eight cats were adapted for 2 wk to a diet containing 25% casein (25% Cas). Four cats (two female and two male) were randomly assigned to a diet of 50% casein (50% Cas) while the other four cats were fed the 25% Cas diet. After 6 wk, the casein component in the diets was changed to isolated soybean protein. Both groups of cats remained at their respective levels of protein (25 and 50%) and were given these diets for another 6 wk. Animals were given free access to food and water. Because the effect of protein level was minimal in xperiment 1, the level of dietary protein for the high protein group was raised to 60%, in xperiment 2 TABL 1 Composition of diets used in xperiment 1 Ingredient 25% Cas 50% Cas 25% Soy 50% Soy Casein1Soybean protein1starch3dextroseanimal tallow4mineral mix5vitamin mix5choline-clsodium S/kg acetateamino mixtaurine250â â â 1_ acid 1 CP 85%, New Zealand Milk Products, Petaluma, CA. 2 CP 90%, Isolated soybean protein, Ardex F-disp., SPI Group, San Leandro, CA. 3 Melojel, food-grade cornstarch, National Starch and Chemical, Bridgewater, NJ. 4 Rendered animal tallow, Florin Tallow, Dixon, CA. 5 For composition, see Williams et al. (1987). 6 Arg, 2.5 Thr, 1; Cys, Met, 2.5; Thr, 1 Cys, 2.5. leaving what was considered a satisfactory level of carbohydrate (11%) in the diet. The low protein diet was raised to 30% to avoid having to supplement the diet with L-threonine. The experiment was also ex tended to 14 wk. Thirty-six cats (18 female and 18 male) were fed the 30% Cas diet for 4 wk. Thirty cats (15 female and 15 male) were selected and each gender was divided into three blocks of five cats. Within gen der, cats were stratified on the basis of taurine con centration of whole blood. Cats within blocks were randomly allocated (one from each block) to the five treatments (i.e., n = 6, 3 of each gender for 30% Cas, SF2, SF4, 60% Cas and 60% Soy). To equalize the tau rine intake, cats were fed a set amount of each diet [26 g dry matter/(kg initial body weight â d)] for 14 wk. Sampling and analytical procedures. Blood samples were taken in heparinized syringes from the jugular vein of unanesthetized cats without prior food deprivation. For xperiment 2, three initial and three final blood samples were collected within a week to reduce the variance for the mean initial and final plasma taurine concentrations. A portion of each blood sample was centrifuged immediately after collection, and the plasma was separated. Plasma and whole blood samples were stored at -20 C until analyzed. The plasma was deproteinized with an equal volume of 60 g/l sulfosalicylic acid and centrifuged at 10,000 X g for 15 min at 4 C.The supernatant was analyzed for taurine using an amino acid analyzer (Beckman 121MB model, Beckman Instruments, Fullerton, CA). To lyse blood cells, whole blood was frozen and thawed twice;

3 SYBAN PRTIN DCRASS PLASMA TAURIN IN CATS 2833 TABL 2 Composition of diets used in xperiment 2 Ingredient SF* SF2 SF4 60% Cas 60% Soy Casein1Soybean protein'soybean fibersucrosestarch1animal tallow1mineral mix1vitamin (60% Cas and 60% Soy) had higher weight gains than mix1choline-cll-arginine-hcll-methioninetaurine/(-carrageenan300â â â â â â 1 See Table 1. 2 Fibrim 1250, Ralston-Purina, St. Louis, M. * SF = 30% casein diets. then an equal volume of distilled water was added. Lysed whole blood samples were deproteinized and analyzed for taurine in the same manner as for plasma samples. In xperiment 2, urine and feces were col lected for 5 d during the final week of the experiment and stored at -20 C until analyzed. For urine samples, free and total taurine (before and after acid hydrolysis) were analyzed, and for feces samples, total taurine was analyzed. Bound taurine was calculated as total minus free taurine. For the total taurine analysis, equal vol umes of urine or feces and 12 mol/l HCl were sealed within glass ampules and heated at 110 Cfor 24 h for hydrolysis. Fecal total bile acids were extracted for analyses according to the method of Ide and Horii (1987) and determined using an assay kit (no. 450, Sigma Chemical, St. Louis, M). Statistical analysis. Two-way ANVA (and where appropriate, repeated measures) was used to examine the effects of diet and time (or gender in x periment 2) on taurine status using SAS General Linear Models (SAS 1985). The differences between means were assessed by Fisher's protected LSD test (Steel and Torrie 1980). P < 0.05 was considered significant. Lin ear and exponential equations were fitted to the rela tionship between urinary free taurine and plasma free taurine concentrations collected during the 14th wk. higher and ended lower than those of cats in the 25% Cas group (Fig. 1). Switching the protein source in the diets from casein to soybean protein caused a signifi cant (P < 0.01) decrease in plasma taurine concentra tion of cats fed the high protein diet during the second 6 wk. There was no significant change of the plasma taurine concentration of cats fed the low protein diets throughout the experiment. Because there were only two cats of each gender per group, no gender effect was examined in xperiment 1. Individual body weight gains in xperiment 2 varied from 100 to 500 g. Cats fed the high protein diets cats receiving the 30% casein (30% Cas, SF2 and SF4) diets. According to the repeated measures ANVA, there were no significant time effects on plasma taurine concentration, but all male groups had a higher (P < 0.05) plasma taurine concentration than corre sponding female groups after 6 wk (Fig. 2). At the last week, cats fed 60% Soy and 60% Cas diets had sig nificantly lower (P < 0.05) plasma taurine concentra tions than cats given the 30% Cas diet. The 60% Soy diet also caused a lower (P < 0.05) plasma taurine con centration than the 60% Cas diet. The diet and gender interaction was not significant (P = 0.64). There was no significant effect of the addition of either 2 or 4% soybean fiber to the diet on plasma concentration of taurine (results not shown in tables or figures). Initial plasma values of 82 /xmol/l (SF), 85  tmol/l(sf2) and 72 fimol/l (SF4) were maintained [87  tmol/l(sf), 83 l/l (SF2) and 75 ^mol/l (SF4)] until the end of the e ni z ff < I % Protein 50% Protein Casein j. Soybean protein WK RSULTS The plasma taurine concentrations of cats given the 25 or 50% casein diets in xperiment 1 were not sig nificantly different for the first 6 wk, although plasma concentrations of cats in the 50% Cas group started FIGUR 1 Plasma concentration of taurine in cats fed a diet containing 1000 mg taurine and either 250 or 500 g protein supplied as either casein or isolated soy protein/kg diet, (xperiment 1). ach value represents the mean ±SM of four cats (two males and two females). Values for groups at weeks 0, 6 and 12 not having a common letter are signif icantly different, P < 0.05.

4 2834 KIM T AL. M 30% Cas M 60% Cas M 60% Soy â -â â F 30% Cas â -Aâ F 60% Cas F 60% Soy xcretion TABL 4 of fecal bile acids and total taurine by cats fed high protein diets (xperiment if'2 e ui Bile acids Total taurine Diet Male Female Male Female 30% Cas60% Cas60% 9b182.6a e«77.9b*117.3a*2. SoyPooled à a e132. SM45.1e74.8b170.3a e70.8b99 Values represent the mean of three animals. 2 Within a column, values not sharing a common superscript are significantly different, P < * Significantly different than male, P < WK FIGUR 2 Plasma concentration of taurine of male and female cats given a diet containing 1000 mg taurine and either 300 or 600 g casein or 600 g soybean protein/kg diet, (xperiment 2). M and F represent male and female, respec tively. ach value represents the mean ±SMof three cats. From 5 wk all male groups are significantly different from corresponding female groups, P < Cats given the 600 g soybean protein/kg diet had significantly lower final plasma taurine concentrations than cats given the 600 g casein/kg diet, and both groups had significantly lower final plasma taurine concentrations than the group given the 300 g casein/ kg diet. experiment. Similarly, whole blood concentration of taurine was not affected by either level of soybean fi ber. Protein source and amount in the diet did not affect the mean group whole blood taurine concentra tion. Initial mean values were 571 /imol/l (30% Cas), 562 fimol/l (60% Cas) and 569 »mol/L(60% Soy), which were not significantly lower at the end of the experiment, i.e., 513 jimol/l (30% Cas), 518  unol/l (60% Cas) and 484 //mol/l (60% Soy). Although female cats had slightly lower whole blood taurine concen tration than males, (results not shown), the effect of gender was not significant. The quantities of taurine excreted in urine are given in Table 3, and the fecal excretion of taurine and total bile acids is presented in Table 4. Male cats excreted significantly greater amounts of fecal total taurine than female cats. Bile acid excretion also was greater for male than for female cats, but the difference was sig nificant only for cats given the 60% Soy diet, Table 4. Female cats given the 60% Cas and 60% Soy diets ex creted less taurine in their urine compared with cats given the 30% Cas diet. Male cats followed the same trend as females, but the differences were not signif icant. Urinary taurine excretion of cats given the 60% Cas and 60% Soy diets were similar, but fecal excretion of taurine by cats given the 60% Soy diet was signif icantly greater than that of cats given either the 30 or 60% Cas diet. The relationship between plasma con centration of taurine and urinary free taurine excretion (R2 = 0.52, P < 0.01) from xperiment 2 is shown in TABL 3 Urinary excretion of free, bound and total (free + bound) taurine and the percentage of taurine in the bound state by cats given diets containing two levels of protein {xperiment -?''-' Free Bound Total i Bound Diet Male Female Male Female Male Female Cas60% 30% Cas60% SoyPooled SM146b67a47a1080a«46a'b22b b55b* a78b77b.948 ' Values represent the mean of three animals. 2 Within a column, values not sharing a common superscript are significantly different, P < * Significant difference between males and females, P < Male 1156 ± 471± ± Female ±2972 ±17

5 SYBAN PRTIN DCRASS PLASMA TAURIN IN CATS 2835 Figure 3. The mean percentage of taurine in the bound state in urine varied from 46 to 72% (Table 3) but did not appear to be related to diet. DISCUSSIN It is apparent that the variations in taurine concen tration are greater in the plasma of the groups with the highest plasma taurine concentrations (Fig. 1, 2). We have noted this routinely in our laboratory. This greater variance appears to be the result of excess di etary taurine and the resultant down-regulation of taurine transport by the kidney, with the subsequent effect of each meal causing an increase in plasma tau rine before it is lost in the urine. When taurine is not in excess, the continuous drain on body taurine by the liver in making taurocholate for biliary secretion and an up-regulation of taurine transport into various tis sues appear to dampen the increase of taurine in plasma after a meal. Thus, the variance is smaller when plasma taurine concentrations are below 70  tmol/lin samples taken from animals having free access to food. The results presented demonstrate that the source and level of protein in a purified diet affect plasma concentration of taurine in cats. In xperiment 1, a diet of 50% soybean protein produced a marked lower plasma concentration of taurine compared with a 25% soybean protein diet. No significant differences were apparent after 6 wk in cats fed 50% and 25% casein diets, although the plasma concentration of taurine in cats fed the 50% casein diet started higher and ended lower than that of cats fed the 25% casein diet. A longer period of time may have resulted in a significant effect. In xperiment 2, at the end of 14 wk, significantly lower plasma concentrations of taurine occurred in cats given the 60% casein diet than in those given the 30% casein diet. The different response in plasma concentrations of taurine in cats receiving the 60 and 50% casein diets is probably due to the higher level of casein, the longer period of diet consumption and perhaps the type of diet used (gel vs. dry). The lower plasma taurine concentration in cats fed the 60% Soy diet than in cats fed the 60% Cas diet ap pears to be the result of increased fecal excretion of taurine and bile acids (Table 4). Sugano et al. (1990) reported that rats fed partially digested fractions (hydrophobic peptides) of soybean protein (prepared after digestion by pepsin or microbial protease) had significantly greater fecal excretion of steroids. These hydrophobic peptides bind bile acids and may decrease the enterohepatic circulation of taurine conjugated bile acids, thereby causing a greater loss of taurine. These undigested fractions may also in duce changes in microflora, leading to increased deconjugation and degradation of taurine. Rats and 200i III z D W III u. It Z SF â SF2 SF4 â 60% Cas + 60% Soy a eo PLASMA TAURIN (fimol/l) FIGUR 3 Relationship between urinary free taurine excretion in nmo\/d (Y) and plasma concentration of taurine in  iiriol/l(x) in cats given various diets. SF, SF2, and SF4 represent diets containing 300 g casein and 0, 20 or 40 g of soybean fiber/kg diet; 60% Cas and 60% Soy represent diets containing either 600 g casein or 600 g soybean protein/kg diet, n = 30. Linear regression: Y = 1.21, X = 4.54, R1 = 0.49 (P < ); exponential, Y = â I'89455'-3", R2 = 0.52 (P < 0.001). rabbits fed diets containing soybean protein excreted more bile acids and cholesterol through feces, and had lower serum or plasma cholesterol concentra tions than animals fed diets containing casein (Choi et al. 1989, Huff and Carroll 1980, Schrijver 1990). Apparently, increased bile salts synthesis caused a decrease in serum cholesterol concentration (Huff and Carroll 1980). It is also reported that increased intestinal transit time of taurocholate was observed in rats fed a soybean protein diet compared with a casein diet (Iwami et al. 1990). The cause of taurine depletion in cats fed heat-pro cessed commercial diets has not been fully elucidated. Because all of the taurine is recovered from the pro cessed diet by water extraction, the problem is not one of destruction or binding during processing. A substance(s) produced during the heat processing of com mercial canned diets or the presence of certain peptides during digestion may decrease the taurine status of cats by increasing the enterohepatic recycling of tau rine conjugated bile acids, or by changing the intestinal microflora. Intestinal bacteria deconjugate bile salts, and in humans, about 7% of the taurine-conjugated bile acids are deconjugated at each enterohepatic cycle (Hepner et al. 1973). Because little taurine is metab olized by routes other than conjugation with bile acids, it should be possible to undertake a balance trial be tween dietary taurine intake plus synthesis and taurine excretion in feces and urine. However, such balance studies consistently show that less than half of the

6 2836 KIM T AL. taurine ingested is recovered in feces and urine. These balance studies indicate that taurine is degraded by microflora in the intestine. Isethionate, sulfite and ac etate have been proposed as end products of the bac terial degradation of taurine (Goodman et al. 1967, Kondo and Ishimoto 1975), of which acetate and isethionate are absorbed from the gut (Goodman et al. 1967, McNeil et al. 1978). Measurements of the end products of taurine degradation in cats given high soybean protein or heat-processed commercial diets may give a more direct explanation of the fate of taurine. Bacterial overgrowth in the intestine of humans has been associated with reduced levels of plasma taurine and rats with surgically created self-filling blind loops of the intestine have lower than normal plasma and retinal taurine 1981). Bacterial overgrowth concentrations (Sheikh resulting in greater microbial degradation of taurine may occur if sub stances produced during heat processing (such as gelatin from animal byproducts) provided a greater level of available substrate for the intestinal microflora. Alternatively, gelatin and soybean protein are comparatively high in arginine and may cause greater cholecystokinin (CCK) release in cats than casein which is much lower in arginine. Greater CCK re lease would increase secretion of bile salts into the intestine, thereby providing more taurocholate to be subjected to microbial degradation. There have been no previous reports of a gender effect on plasma taurine concentration of cats. In x periment 2, male cats had higher concentrations of taurine in plasma than female cats. Because food intake per kilogram of body weight was the same for all cats in this experiment, male and female cats had the same taurine intake per unit of body weight. The cause of this gender difference is not known, but it appears to result from greater synthesis of taurine in males than females, because male cats generally excreted a greater amount of taurine in their urine (Table 3) and fecal total bile acids (Table 4) than female cats. The impor tance of this finding is that experiments on plasma taurine status should take into account the gender of the cat. In xperiment 2, cats fed the 30% Cas excreted greater amounts of free taurine in urine than cats fed 60% Cas or 60% Soy. Because urinary free taurine ex cretion is related to the plasma taurine status (Fig. 3), cats that had higher plasma concentrations of taurine excreted greater amounts of free taurine. This result is similar to that reported by Hickman et al. (1990) and is consistent with earlier work (Park et al. 1989) showing that the cat kidney adapts to low plasma tau rine concentrations by increasing taurine rã sorption. However, the percentage of bound taurine in the urine did not correlate with the concentration of plasma taurine, gender, type of dietary protein or fiber. Al though various conjugates of taurine such as bile salts, guanidotaurine (Thoai and Robin 1954), carbamyltaurine (Schrà mand Crokaert 1957), quinaldylglycyltaurine (Kaihara and Price 1961) and retinotaurine (Skare et al. 1982) are known to exist, the composition and importance of the bound taurine in urine of cats are unknown. Hemicellulose fed at high dietary concentration has been reported to increase fecal excretion of bile salts in humans (Forman et al. 1968). In xperiment 2, we tested the effect of a soybean fiber isolate, added in excess of that found in heat-processed commercial diets, on taurine status. Although this soybean fiber isolate is mainly hemicellulose, we found no effect of either 2 or 4% soybean fiber in the diet on plasma taurine concentration. Ide et al. (1989) reported that pectin at either 6 or 10% of the diet resulted in significantly increased secretion of biliary bile salts and decreased hepatic taurine con centration of rats, whereas rats fed diets containing pectin at either 1 or 3% in the diet did not show any differences compared with rats fed fiber-free diets. The lack of a response to dietary soybean fiber in xperiment 2 appears to be the result of the low lev els of dietary fiber tested. xperiments with humans have often used 10 or even 20% of dietary fiber to achieve an effect on bile salts metabolism. Thus, be cause fiber content is low in commercial diets, it is unlikely that soluble fiber is the cause of the in creased fecal loss of taurine in cats fed heat-pro cessed commercial diets. Whole blood concentrations of taurine were not af fected by protein source or amount in the diet in x periment 2. Because the plasma concentration of tau rine represents a small pool, and tissue removes taurine from plasma against a concentration gradient, plasma is more sensitive to small changes in taurine balance than whole blood. In contrast, whole blood concen tration of taurine changes more slowly because it is a larger pool and is present at high concentrations in certain cells such as platelets and granulocytes. The large pool size and relatively short experimental period could explain the lack of significant changes in whole blood concentrations of taurine with increased dietary protein. Apparently, after a decrease in the amount lost in urine and an increase in the uptake into tissues, one gram of taurine per kilogram of diet is sufficient to maintain adequate taurine homeostasis of cats fed the 60% soybean protein diet. If these cats had been fed the diets for longer periods, or if dietary taurine had been decreased somewhat, there probably would have been an effect on whole blood concentration of taurine. In conclusion, this study shows that the type and quantity of the protein in the diet affect the taurine status in cats. Cats fed high soybean protein diets had significantly lower plasma taurine concentrations than cats fed low casein diets, apparently caused by an in creased fecal loss of taurocholate.

7 SYBAN PRTIN DCRASS PLASMA TAURIN IN CATS 2837 ACKNWLDGMNTS The authors thank Hoffman-La Roche Inc., Nutley, NJ, for the gift of the vitamin mixture and Ralston- Purina Co., St. Louis, M, for providing the soybean fiber. LITRATUR CITD Aquirre, G. D. (1978) Retinal degeneration associated with the feeding of dog foods to cats. J. Am. Vet. Med. Assoc. 172: Berry, W. & Reichen, J. (1983) Bile acid metabolism: its relation to clinical disease. Seminars in Liver Disease 3: Choi, Y. S., Goto, S., Ikeda, I. &. Sugano, M. (1989) Interaction of dietary protein, cholesterol and age on lipid metabolism of the rat. Br. J. Nutr. 61: Forman, D. T., Gavin, J.. &. Forestner, J.. (1968) Increased ex cretion of fecal bile acids by an oral hydrophilic colloid. Proc. Soc. xp. Biol. Med. 127: Goodman, H.., Wainer, A., King, J. S., Jr. & Thomas, J. J. (1967) 35S 2-Hydroxyethanesulfonic acid (isethionic acid) in urine of human subjects given 35Staurine. Proc. Soc. xp. Biol. Med. 125: Hayes, K. C., Carey, R.. & Schmidt, S. Y. (1975) Retinal degen eration associated with taurine deficiency in the cat. Science (Washington, DC) 188: Hepner, G. W., Sturman, J. A., Hofmann, A. F. & Thomas, P. J. (1973) Metabolism of steroids and amino acid moieties of con jugated bile acids in man. III. Choletaurine (taurocholic acid). J. Clin. Invest. 52: Hickman, M. A., Rogers, Q. R. & Morris, J. G. (1990) ffect of processing on fate of dietary [14C)taurine in cats. J. Nutr. 120: Huff, M. W. & Carroll, K. K. (1980) ffects of dietary protein on turnover, excretion oxidation, in rabbits. and absorption of cholesterol J. Lipid Res. 21: and on steroid Ide, T. & Horii, M. (1987) A simple method for the extraction and determination of non-conjugated and conjugated luminal bile acids in rats. Agrie. Biol. Chem. 51: Ide, T., Horii, M., Kawashimi, K. & Yamamoto, T. (1989) Bile acid conjugation and hepatic taurine concentration in rats fed on pec tin. Br. J. Nutr. 62: Iwami, K., Kitagawa, M., Nagasaki, T. & Ibuki, F. (1990) Compar ison of intestinal taurocholate uptake in rats given soy proteinor casein-based diet. Nutr. Res. 10: Kaihara, M. & Price, J. M. (1961) Quinaldylglycyltaurine: a urinary metabolite of quinaldic acid and kynurenic acid in the cat. J. Biol. Chem. 236: Kondo, H. &. Ishimoto, M. (1975) Purification and properties of sulfoacetaldehyde sulfo-lyse, a thiamine pyrophosphate-dependent enzyme forming sulfite and acetate. J. Biochem. 78: McNeil, N. I., Cummings, J. H. & James, W. P. T. (1978) Short chain fatty acid absorption by the human large intestine. Gut 19: Morris, J. G., Rogers, Q. R. & Pacioretty, L. M. (1990) Taurine an essential nutrient for the cat. J. Small Anim. Pract. 31: National Research Council (1986) Nutrient Requirements of Cats, rev. ed. National Academy Press, Washington, DC. Park, T., Rogers, Q. R., Morris, J. G. &.Chesney, R. W. (1989) ffect of dietary taurine on renal taurine transport by proximal tubule brush-border membrane vesicles in the kitten. J. Nutr. 119: Rabin, B., Nicolosi, R. J. &. Hayes, K. C. (1976) Dietary influence on bile acid conjugation in the cat. J. Nutr. 106: Remtulla, M. A., Applegarth, D. A., Clark, D. G. &.Williams, I. H. (1977) Analysis of isethionic acid in mammalian tissues. Life Sci. 20: SAS Institute Inc. (1985) SAS/STATâ Guide for Personal Com puters, version 6.0. SAS Institute Inc., Cary, NC. Schrà m,. & Crokaert, R. (1957) Radiochromatographic study of substances formed from 35S]taurine in the rat. Biochem. J. 66: Schrijver, R. D. (1990) Cholesterol metabolism in mature and im mature rats fed animal and plant protein. J. Nutr. 120: Sheikh, K. (1981) Taurine deficiency and retina defects associated with small intestinal bacterial overgrowth. Gastroenterology 80: 1363 abs.). Skare, K. L., Sietsema, W. K. & DeLuca, H. F. (1982) The biological activity of retinotaurine. J. Nutr. 112: Steel, R. G. D. & Torrie, J. H. (1980) Principles and Procedures of Statistics, 2nd à d.,pp McGraw-Hill, New York, NY. Sugano, M., Goto, S., Yamada, Y., Yoshida, K., Hashimoto, Y., Matsuo, T. &. Kimoto, M. (1990) Cholesterol-lowering activity of various undigested fractions of soybean protein in rats. J. Nutr. 120: Thoai, N. & Robin, Y. (1954) Mà tabolismedes derives guanidyles. II. Isolement de la guanidotaurine (taurocyamine) et de l'acide guanidoacetique (glycocyamine) des vers marins. Biochim. Biophys. Acta 13: Williams, J. M., Morris, J. G. & Rogers, Q. R. (1987) Phenylalanine requirement of kittens 117: and the sparing effect of tyrosine. J. Nutr.