Nitrogen Balance, Plasma Free Amino Acid ...

Nutrient Metabolism

Nitrogen Balance, Plasma Free Amino Acid Concentrations and urinary Orotic Acid Excretion during Long-Term Fasting in Cats1-4 VINCENT BIOURGE,5 JOSEPH M. GROFF,* CINDY FISHER, DEBORAH BEE, JAMES G. MORRIS AND QUINTON R. ROGERS6 Department of Molecular Biosciences and *Department Medicine, University of California, Davis, CA 95616

and then the net nitrogen losses decreased to a plateau (-303 ±52 mg-d-^kg-2/3) after 4 wk. Fasting was associated with a decrease in plasma concentration of essential amino acids. When plasma amino acid concen trations were considered individually, concentrations of alanine, methionine, taurine, citrulline, arginine and tryptophan decreased the most (>50%), whereas concentra tions of glutamine, glutamate and ornithine significantly increased. Orotic acid was not detected in the urine during the fast. After 1 wk of fasting, obese cats had reduced nitrogen excretion, but not to the same extent as has been shown in obese humans or obese rats. It is suggested that the availability of several amino acids may become limiting for liver protein synthesis during fasting and that these deficiencies may contribute to the development of hepatic lipidosis. Orotic acid was not linked to hepatic lipidosis caused by fasting in cats. J. Nutr. 124: 1094-1103, 1994.

1Presented at the Waltham Symposium of the Nutrition of Companion Animals, September 23-25, 1993, Adelaide, Australia (Biourge, V., Groff, J. M., Morris, J. G. &. Rogers, Q. R. (1993) Nitrogen balance, plasma free amino acid concentrations and urinary orotic acid excretion during long-term voluntary fasting in obese cats. Proceedings of the Waltham Symposium on the Nutrition of Companion Animals, p. 67 (abs.)j. ^Supported by a gift from Alpo Pet-Food, Allentown, PA, and a

INDEXING KEY WORDS:

feline nitrogen balance

grant from the Companion Animal Fund, School of Veterinary Medicine, University of California, Davis, CA. •'Thisreport represents a portion of a thesis submitted by the first author to the graduate school as a partial fulfillment of the requirement for the Ph.D. degree. *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

Obese cats may not be able to endure food depri vation as well as obese people and dogs (Benedict 1915, Biourge et al. 1993, de Bruijne 1979, Owen et al. 1969). We reported earlier that rapid weight loss, most probably the result of minimal food intake, was as sociated with an outbreak of hepatic lipidosis in a 0022-3166/94 $3.00 ©1994 American Institute of Nutrition. Manuscript received 9 September 1993. Initial review completed

solely to indicate this fact. 5Hill's Fellow in Nutrition. 6To whom correspondence should be addressed. Abbreviations used: ALT, alanine amino transferasc; AST, aspartate amino transferase; FIHL, feline idiopathic hepatic lipidosis; SAP, serum alkaline phosphatase.

20 October 1993. Revision accepted 7 February 1994. 1094

Downloaded from jn.nutrition.org by guest on October 14, 2011

group of laboratory cats (Biourge et al. 1993). Briefly, six cats showed clinical signs similar to those reported for cats with feline idiopathic hepatic lipi dosis (FIHL)7 after 5 to 6 wk of rapid weight loss (Barsanti et al. 1977, Jacobs et al. 1989). The onset of weight loss coincided with a change from a com mercial to a purified diet. All six cats were obese (38 ± 6% overweight) before the dietary change. The purpose of this study was to confirm that longterm fasting induces hepatic lipidosis in obese cats and to examine the effect of long-term fasting on nitrogen balance, plasma free amino acid concentra tions and urinary orotic acid excretion. Nitrogen and amino acid metabolism are known to be different in cats as compared with other laboratory animals and humans (Rogers et al. 1977), but the effect of fasting, to the best of our knowledge, has not been reported.

ABSTRACT The purpose of this study was to ascertain the changes in nitrogen balance, plasma free amino acid concentrations, urinary orotic acid excretion and body weight during long-term fasting in adult obese cats. Results from eight cats that fasted rather than eat an unpalatable diet are reported. After 5 to 6 wk of weight loss, six of the eight cats developed signs of hepatic lipidosis, and the livers of all cats were severely in filtrated with lipids. Cats lost (mean ±SE) 33.2 ±1.4% of their pre-fasting body weight. Mean nitrogen balance (±SE)was -547 ±54 mg-d-i-kg"2/3 for the first week,

•fasting •cats •hepatic lipidosis •amino acids

of Medicine, School of Veterinary

NITROGEN

METABOLISM DURING

Orotic acid in the diet induces fatty liver in rats (Durschlag and Robinson 1980) and has been proposed as a cause of FIHL (Burrows et al. 1981); therefore we examined urinary orotic acid excretion during the course of the fast. The changes associated with fasting on hematology, serum chemical constituents and liver histology have been reported elsewhere (Biourge 1993).

MATERIALS AND METHODS

Diet. For at least 6 mo before the study, cats were given free access to commercial canned (Alpo PetFood, Allentown, PA) and dry cat food (Alpo Pet-Food and IAMS Cat Food, Dayton, OH). After 9 d for adaptation to the metabolism cages, the purified diet was offered without restriction (Table 1). This pu rified diet was nutritionally complete but was offered as a powder that had a low palatability. Cats had free access to water at all times. Protocol. After adaptation to metabolism cages and baseline measurements, cats were fed only the pu rified diet. The duration of the voluntary fast was limited to 7 wk, but the fast was terminated if cats became hyperbilirubinemic (serum total bilirubin concentration >7 ^mol/L) or reached a minimum body weight determined for each cat at the beginning of the study by comparison to a lean cat of the same length and body type. Cats were maintained ac cording to the Guide for the Care and Use of Laboratory Animals (NRC 1985) and the Animal Welfare Act. The protocol was approved by the Com mittee for Care and Use of Laboratory Animals, University of California, Davis.

FASTING IN CATS

1095

TABLE 1 Composition of the purified diet Amount

Ingredient

g/kg diet Soybean protein1 Cornstarch2 Sucrose^ Hydrogenated beef tallow4 Taurine5 Choline chloride6 Mineral mix7 Vitamin mix7

350.0 285.5 200.0 100.0 1.5 3.0 50.0 10.0

'Archer Daniels Midland (Decatur, IL). ^National Starch and Chemical (Bridgewater, NJ). 3Staley (Decatur, IL). 4Bunge Edible Oils (Fort Worth, TX). 5Taisho Pharmaceutical (Terranee, CA). 6Du Pont (Highland, IL). 7Williams et al. (1987).

Measurements and collections. Food intakes were measured daily, and body weights were recorded weekly. Urine was collected in beakers containing 5 mL of H2SÛ4 and 2 mL of xylene. The urine volume was measured and feces collected between 0800 and 1000 h daily. Urine and feces were stored at 4°Cuntil the end of the week, then pooled and kept at -20°C pending analysis. Two 6-mL blood samples were taken from the jugular vein of each cat on wk 0 (baseline), 2, 4, 5 and 6 of fasting. The first blood sample was collected in a heparinized syringe, and the plasma was separated by centrifugation at 4°Cand kept on ice for up to 2 h until an aliquot was deproteinized with an equal volume of sulfosalicylic acid (275 mmol/L) in aqueous solution. Deproteinized plasma was kept at -20°C for up to 3 wk pending analysis for free amino acid concentrations. The other blood sample was drawn into a non-heparinized sy ringe and divided in two aliquots. One aliquot was placed in a 5-mL tube containing 22 /¿molof EDTA, and the other was allowed to clot. Plasma was sepa rated from red blood cells and serum from clot by centrifugation. An aliquot of serum was sent to a commercial laboratory (California Veterinary Diag nostic Laboratories) for determination of serum alanine amino transferase (ALT, EC 2.6.1.2), aspartate amino transferase (AST, EC 2.6.1.1) and alkaline phosphatase (SAP, EC 3.1.3.1) activities and serum total bilirubin, direct bilirubin, glucose, urea, protein and albumin concentrations. The remaining EDTAtreated plasma and serum were stored at -80°C pending serum samples all food

further analysis for plasma free fatty acid and hydroxybutyrate concentrations. Blood were taken between 0800 and 1000 h, and was removed the evening before sampling.

Downloaded from jn.nutrition.org by guest on October 14, 2011

Animals. Obese adult specific-pathogen-free cats were obtained from the Nutrition and Pet Care Center, University of California, Davis. Fifteen cats (eight spayed females, two intact females and five castrated males) of 2 to 9 y of age were initially included in the original study (Biourge 1993). Results from only eight of 15 cats (three castrated males, five spayed females) are included in this report. The reasons for not using the results collected on the seven remaining cats were as follows: urine was not collected in two cats (pilot study), four cats ate more than 6 g/d of the purified diet (see below) and one cat had elevated serum urea nitrogen [7.8 mmol/L vs. 5.8 mmol/L (normal upper value; California Veterinary Diagnostic Laboratory, West Sacramento, CA)] and serum creatinine concentrations [0.20 mmol/L vs. 0.16 mmol/L (normal upper value; California Veterinary Diagnostic Laboratories)], indicating com promised renal function. All eight cats were obese [6.4 ±0.2 (SE)kg, >40% overweight] at the beginning of the study. Cats were housed in metabolic cages for 9 d for adaptation and base-line measurements before the low palatability diet was offered. The room was maintained at 22°Con a 14-h light, 10-h dark cycle.

VOLUNTARY

1096

BIOURGE ET AL.

calibrated with a commercial standard (Nitrogen in Orchard Leaves #502 055; Leco; recovery = 100 + 1%). The blank was determined by running the analyzer without samples. Hair loss was not accounted for in N balance determinations because of its small contri bution and difficulties in complete collection and separation. Orotic acid in the urine was determined by HPLC using a modification of the method of Brusilow and Hauser (1989). Briefly, orotate is resolved from inter fering urine components by strong anión exchange HPLC (Radial Pak SAX #87752, Waters, Millipore Division, Milford, MA) using a solution of 0.9 mmol/ L ammonium formate buffer (550 mL/L) and methanol (450 mL/L), adjusted to pH 2.8, and eluted at 2 mL/min. Before analysis, urine samples were centrifuged at 16,000 x g and the supernatant filtered (Acrodisc 12 CR PTFE syringe filter #4423, 0.2 ¿un; Gelman, Ann Arbor, MI). The urine filtrate was then diluted 1:10 with the ammonium formate-methanol liquid phase described above. Diluted filtered urine (15 /iL) was placed on the column. Serum hydroxybutyrate and plasma (obtained using EDTA as an ticoagulant) free fatty acid concentrations were deter mined enzymatically using commercially available kits (hydroxybutyrate kit no. 310-A, Sigma Chemical, St. Louis, MO; NEFA C no. 990-75401, Wako Chem icals, Dallas, TX). Water was used as a blank, and sample concentrations were obtained by comparison to a linear standard curve (hydroxybutyrate = standard range 0.0 to 4.8 mmol/L, R2 = 0.997; free fatty acid = standard range 0.0 to 2.0 mmol/L, R2 = 0.998).

Statistics. The significance (P < 0.05) of the changes between wk 0 and the end of fast for serum ALT, AST and SAP activities, serum bilirubin, glucose, urea, protein, albumin and hydroxybutyrate concentrations and plasma free fatty acid concentrations were tested by paired t tests. The significance of the changes over time for nitrogen balance, urinary and total nitrogen losses, and concentrations of plasma free amino acid concentrations were tested using a repeated-measure ANOVA with time (weeks of fast starting at wk 0) as the repeated factor (Cody and Smith 1987). Means during the fast were compared with the mean before the initiation of the fast and with the mean of the first week of fasting (nitrogen balance only) using the contrast method (Cody and Smith 1987). A logarithmic transformation was used for the analysis of serum liver enzyme activities, total bilirubin con centrations and free amino acid concentrations. All statistical analyses were performed using the SAS statistical package (PC-SAS, version 6.04, SAS In stitute, Cary, NC). All results are expressed as means ±SEM.

RESULTS The eight cats retained to evaluate nitrogen metab olism during fasting ate less than 1 g/d of the purified diet (mean = 0.90 ±0.45 g/d; range 0.0-2.2 g/d; l g diet = 47.8 mg nitrogen). These intakes were probably an overestimation because of the difficulty in meas uring spillage. Fasting resulted in rapid weight loss (Fig. 1). The average body weight for the eight cats at the end of the fast was 4.3 ±0.1kg, representing a loss of 33.2 ±1.4% from the initial body weight. The rate of weight loss decreased from 9.7 ±0.6%/wk for the first week to 4.0 ±0.3%/wk at the end of the fast. Commercial food was returned after 5 wk to five cats, one cat because it had reached its predetermined minimum body weight, the four other cats because of hyperbilirubinemia. The three remaining cats were offered commercial food after 6 wk, one cat because it had reached its predetermined minimum body

8Serum chemistries were performed on a Hitachi 747-200 auto mated analyzer (EM Science, Exton, PA). Methods used for the analytical determinations can be found in the following references: serum ALT activities: Bergmeyer, H. U. & Horder, M., Clin. Chem. Acta (1980) 105: 147F; serum AST activities: Bergmeyer, H. U. et al., Clin. Chem. Acta (1977) 80: F21; SAP activities: Bowers, G. N. &. McComb, R. B. Clin. Chem. (1975) 21: 1988; total and direct serum bilirubin concentrations: Walters, M. & Gerarde, H., Microchem. J. (1970) 15: 231; serum glucose concentrations: Cooper, G. R., CRC Crit. Rev. Clin. Lab. Sci. (1973) 4 (2): 101; serum urea concentrations: Sampson, E. J. et al., Clin. Chem. (1975) 26 (7): 816; serum total protein concentrations: Comali, A. G. et al., J. Biol. Chem. (1949) 177: 751; serum albumin concentrations: Webster, D., Clin. Chem. (1977) 23: 663.

Downloaded from jn.nutrition.org by guest on October 14, 2011

Percutaneous ultrasound-guided liver biopsies (Strombeck and Guilford 1990) were obtained before and at the end of the voluntary fast, using a 18-g biopsy needle (Biopty®needle, Bard Biopty Cut #001418, Bard Urological Division, Covington, GA). Biopsy specimens were fixed in 100 mL/L neutral buffered formalin. After paraffin embedding or freezing, sections were stained with Harris' hematoxylin and eosin, periodic acid-Schiff or oil red O (Humason 1979, Luna 1968). Analytical methods. Plasma amino acid concentra tions were determined on a 7300 Beckman Amino Acid Analyzer (Beckman Instruments, Palo Alto, CA). Before analysis, the deproteinized plasma (see above) was adjusted to pH 2.2 with lithium hydroxide, and the equivalent of 20 ¿iLof plasma was placed on the column of the analyzer. Serum ALT, AST and SAP activities and serum total bilirubin, direct bilirubin, glucose, urea, protein and albumin concentrations were determined on an automated chemistry analyzer at a commercial clinical veterinary laboratory (California Veterinary Diagnostic Laboratories).8 Nitrogen in urine, feces and diets was determined by the Kjeldahl method (AOAC 1975) or with an automatic nitrogen analyzer (Leco FP-248 model 601-700, Nitrogen Determinator; Leco, St. Joseph, MI) according to the manufacturer's instructions and

NITROGEN

METABOLISM DURING

VOLUNTARY

FASTING IN CATS

1097

1.05 1.00 0.95 0.90 0.85

DtetCKY Change

0.80

I

0.75 0.70

Hyperbilirubinemia 0.65 0.60

-2-10123456 Time (weeks)

=v

-v -w.

weight, and the other two cats because of hyper bilirubinemia. The cats seemed to tolerate the fast very well and were alert and playful during most of the study. However, a few days before hyperbilirubinemia was observed, cats appeared less alert and showed mild signs of dehydration. Severe muscle wasting, as deter mined by palpation of muscle masses around the spine and the large muscle groups of the legs, was obvious at the end of the fast, especially in cats with hyperbilirubinemia. At the end of the fasting period, all eight cats had elevated SAP activities (Table 2). Seven of the eight cats had elevated serum ALT ac tivities, and six had an increase in serum AST ac tivities and elevated serum bilirubin (total and direct) concentrations. The changes in ALT, AST and SAP activities and serum bilirubin concentrations between baseline and the end of the fast were statistically significant (P < 0.05, Table 2). When offered com mercial cat food, five of the six cats with hyper bilirubinemia refused to eat and had to be tube-fed until voluntary consumption resumed. Fasting was associated with a significant (P < 0.05) decrease in serum urea, protein and albumin concentrations and a nonsignificant decrease (P = 0.07) in serum glucose concentration. Fasting resulted in a significant (P < 0.05) elevation of serum hydroxybutyrate and plasma free fatty acid concentrations (Table 2). Microscopic examination of liver biopsies revealed that fasting was associated with an accumulation of lipid within the hepatocytes (Fig. 2). Oil red O and periodic acidSchiff stainings confirmed the lipid content of the

üfr

^^^V)V'?K^W. 7*» vA-