Protein status of exercising Arabian horses fed diets containing 14% or 7.5% crude protein fortified with lysine and threonine

Reviewed by Equine Nutrition and Physiology Society

PROTEIN STATUS OF EXERCISING ARABIAN HORSES FED DIETS CONTAINING 14% OR 7.5% CRUDE PROTEIN FORTIFIED WITH LYSINE AND THREONINE P.M. Graham-Thiers, PhD i ; D.S. Kronfeld, PhD, DVMi; K.A. Kline, MSi; DJ. Sklan, PhD2; P.A. Harris, PhD3

SUMMARY Ten Arabian horses (5 mares and 5 geldings averaging 436± 17 kg) were randomly assigned to one of two dietary treatments: LP (7.5% CP fortified with 0.5% lysine and 0.3% threonine) or HP (14.5% CP). Diet composition and nutrient content are shown in Table 1. Horses were conditioned for nine weeks, then an exercise test was performed. It consisted of a warm-up followed by six, one-minute sprints at 10 mls separated by four minutes of walking on a 6% slope. It concluded with a 3D-minute recovery at the walk. Blood samples were taken every two weeks during the conditioning period as well as at rest, during the last 15 seconds of each sprint and at 5, 10, 20 and 30 minutes of recovery during the exercise test. Urine samples were obtained from mares every two weeks. Blood samples were analyzed for albumin, total protein, plasma urea-N (PUN) and creatinine. Urine was analyzed for urea, uric acid and creatinine. Horses were observed daily for clinical signs of protein deficiency. Effects of diet and time were evaluated by analysis of variance with repeated measures. During the conditioning period, there was no effect of diet on plasma albumin (P = .25), total protein (P = .72) or creatinine (P = .21). All values were within the normal ranges reported for horses. There was an effect of diet on PUN (P = .0001) with horses in the high protein group exhibiting greater PUN levels than horses in the low protein group. No difference in urine creatinine levels (P = .78) were observed. Urine urea (P = 0.011) as well as uric acid (P = 0.0001) were lower in the low protein group than in the high protein group. These differences are expected as a reflection of the different protein levels in the diet. During the exercise Authors' addresses: 'Virginia Polytechnic Institute and State University, Blacksburg, VA 24601-0606; 2 Hebrew University, Rehovot, Israel; 3Equine Studies Group, WALTHAM Centre for Equine Nutrition and Care, Melton Mowbray, England, UK, LE14 4RT

516

test, no differences in plasma albumin (P = .32), total protein (P = .81) or creatinine (P = .39) were observed. A greater PUN persisted in the high protein group (P = .0001). This was expected due to the difference in dietary nitrogen. No detrimental effect of the lower level of protein, fortified with amino acids, on protein status was observed during this experiment based on the measurements in this study. These results indicate that the restricted protein level fortified with limiting amino acids was adequate for conditioning and exercise over the nine weeks of the experiment. In a companion study, the lower level of protein fortified with amino acids moderated the acid-base responses to repeated sprints. i

INTRODUCTION An optimal range of dietary protein has not been developed for the exercising horse. 2 Increased needs for protein during exercise include maintenance of new muscle mass, repair of existing muscle mass and replacement of any nitrogen that would be lost in sweat. 3 The upper limit of dietary protein for the exercising horse should minimize thermogenic, ureogenic and acidogenic effects, but the lower limit must avoid protein deficiency. The current dietary protein recommendations are 9.8, 10.4 and 11 % crude protein (CP) for horses in light, moderate and intense exercise, respectively.4 Surveys in MichiganS,6 and Australia7 revealed that Thoroughbreds and Standardbreds in training were fed an average of 14% CPo Another study in central North CarolinaS found that 70% of the horses were being fed protein in excess of NRC recommendations. We propose that a low protein diet fortified with limiting amino acids may reduce undesired effects of protein without risking protein deficiency. This study evaluated the effects of dietary protein level on protein status of Arabian horses during interval training and a repeated sprint test. A companion study evaluated the acid-base status of these horses. i JOURNAL OF EQUINE VETERINARY SCIENCE

MATERIALS AND METHODS Horses. Ten Arabian horses (5 mares, 5 geldings, age 7 to 10 years) were used in the experiment. General health and feed intake were observed daily, and body weight, body condition, and coat condition were monitored every two weeks. The protocol was approved by the Institutional Animal Care and Use Committee. Diets. Horses were randomly assigned t6 two complete feeds formulated to provide 3.3 Mcal DE/kg dry matter (DM) (Table 1) and were similar except for two protein levels: HP, 14.5% CP and LP, 7.5% CP, supplemented with .5% crystalline lysine and .3% threonine' to match levels in the HP diet. Conditioning. All horses underwent nine weeks of physical conditioning using a high-speed treadmill. b The interval training protocol consisted of three phases, each lasting 3 weeks (Table 2). Horses were interval trained twice a week and were walked at 1.5 mls for 30 minutes on non-exercised days on a mechanical walker. Standard Exercise Test. After the nine weeks of conditioning, all horses performed a standard exercise test (SET). The SET consisted of six minutes at the walk (1.5 mis, three minutes on zero slope and three minutes on a 6% slope), a five-minute warm up at the trot (3.5 mls) followed by six, one-minute sprints at 10 mls on a 6% slope separated by walking (1.5 mls) for four minutes, and followed by a 30minute recovery period at the walk (1.5 mls) with zero slope. Horses were fasted overnight but had access to water before

Table 1. Ingredient and chemical composition of experimental diets, as-fed basis' Item, %

High Protein

Low Protein

45 17

45

10

10 5

Orchardgrass hay Oats Corn oil Molasses

5

Soybean meal

20

17

0

0 17

1.5

.75

Limestone

0.1

Salt

0.5

0.55 0.5

Corn grain Dicalcium Phosphate

Vitamin/Mineral mix Lysine Threonine Moisture (%)

the SET. All tests were performed in a climate-controlled bam with temperature set at 24°C and relative humidity approximating 50%. Heart rate was recorded using a digital monitor. c A 14-guage, '150 cm polyethelene catheterd was placed in the left jugular vein approximately 60 minutes before the SET. The area over the left jugular vein was surgically prepared by shaving and washing with an iodine solution (7.5% iodine )! A small incision was made through the skin over the area for placement of the catheter in the jugular vein after the area had been anaesthetized with a subcutaneous injection (lidocaine hydrochloride injectable 2%).f A lO-gauge needle was inserted into the jugular vein and the catheter was passed into the pulmonary artery. Placement of the end of the tubing in the pulmonary artery was ensured by observing pressure waves on an oscilloscope attached to a pressure transducer (Propaq 140).g Once the catheter was placed, it was secured by sutures. h A 5-mL extension seti was attached to the catheter to facilitate sampling during exercise. The catheter was kept patent with heparinizedi salinef (10 units/mL). Sampling and Analysis. During the conditioning period, weight, body condition score, blood and urine samples were taken every two weeks approximately three hours after the morning meal. Horses were weighed on an electronic scale. k Body condition score (BCS) was evaluated using a standardized system. 9 Blood samples were taken via jugular puncture into heparinized tubes. l Plasma was separated and frozen for later analysis. Urine samples were taken from mares only. A catheter was placed into the urethra after washing the mare with an iodine solution (7.5% iodine)e and urine flow was initiated by suction from a syringe placed on the end of the catheter. Urine was collected, divided into aliquots and frozen for later analysis. During the SET, 30 mL of mixed venous blood was taken at rest, during the last 15 seconds of each sprint and at 5, 10, 20 and 30 minutes of recovery. Blood samples were taken using syringes, m placed in heparinized tubes l and centrifuged to yield plasma. Aliquots were frozen for later analysis. Table 2. Protocol for interval training over nine weeks. Each horse exercised twice weekly and each phase is additive to the one before. Phase one was weeks 0-3, phase two weeks 4-6 and phase three weeks 7-9.

1

Training

0 0 7.9

(week) 0-3

Exercise (minutes)

Speed (m/s)

Slope (%)

1.5 1.5

0

7.3

0-3

0-2.5 2.5-5

Gait Walk Walk

5-9

Trot

3.5

6 6

.5 .3

6

DE (Mcal/kg)'

3.25

3.22

0-3

Ether Extract (%)'

13.1

13.3

0-3

9-12

Canter

7.0

7.6

0-3

12-15

Trot

3.5

6

0.61±.023 0.52±.014

4-6

15-18

Canter

8.0

6

18-21

Trot

3.5

Canter Trot

9.0 3.5

6 6

Walk

1.5

Crude Protein (%)'

13.5

L-Lysine (%)'"

0.69±.039

L-Threonine (%)"*'

0.59±.036

·Standard analytical procedures were followed by the Virginia Tech Forage Laboratory; N=4 •• N=4 "dry matter basis

Volume 20, Number 8, 2000

4-6 7-9 7-9 0-9

21-24 24-27 Additional 2 minutes

6 0

517

. , Albumin

0

~HP

Total Protein

•:

4.00 . - - - - - - - - - - - - - - - - , 8.50 ..J "tI

m 3.50

8.00

:&

a: a..

7.50

3.00

c(

c(

o

iii

..J

C w

C

7.00

~

..J

a..

C

0.09

iii LX.la...t:.4....l>......I<4.....,..L.J«.............,"'--'.:..L...J

14

::::l

C

..J

2.00

'-'-

o

14

28

42

56

'-'- 6.50 63

a..

TIME, days

Figure 1. Effect of nine weeks of conditioning on plasma albumin and total protein. There was no effect of diet (P=.53 and P=.93, respectively); however, there was an effect of training (P=.005 and P=.021, respectively).

28

42 56 TIME, days

-a

E

w z 1.25

D

Z

i= cw a:: (.)

- - Albumin

High

Ci

z

Low Protein

..

- -

jjj

8.00

o

14

28

42

56

63

30 , . - - - - - - - - - - - - - - , DIET·TRAINING P=.004 _

High

..."

25

Protein

CI

20

DLow Protein

u

iii

rn

E

15

a.

10

0

~

14

a f!28

42

~ ~

56

DIET

P=.OOOl

0

..J

C

c(

7.50

I-

0

I-

3.25

c

::E

I/)

7.00 R S1 S2 S3 S4 S5 S6 R1 R2 R3 R4 REST

Figure 2. Effect of nine weeks of conditioning on plasma creatinine. There was no effect of diet (p=.29); however, there was an effect of training (p=.033).

l-

a: a.

3.50

3.00 L..IIiL...J..........I....L.........-L-J........J.....JI!llil.....JC-JOIIIL...1....J

-a

Z

3.75

c(

TIME, days

SPRINTS

c· ..J a.

RECOVERY

Figure 5. Effect of six one-minute repeated sprints (81-86) and a 30-minute recovery (R1-R4) on plasma albumin and total protein. There was no effect of diet (P=.32 and P=.81, respectively); however, there was an effect of exercise (P=.001 and P=.004, respectively).

All plasma samples were analyzed for total protein (TP) (Proc. No. 541),° albumin (Proc. No. 631),° creatinine (Proc. No. 557)° and urea-N (Proc. No. 67-UV).0 Urine aliquots were analyzed for urea (Proc. No. 640),° creatinine (Proc. No. 557)° and uric acid (Proc. No. 685-11 ).0 Statistics. Data were summarized as least squares means and standard errors. Analysis of variance with repeated measures was used to evaluate the effects of diet and time (conditioning) during the conditioning period, and diet, time (exercise and/or recovery) and interactions during the SET using the GLM procedure of SAS.IO

63

TIME, days

Figure 3. Effect of nine weeks of conditioning on plasma urea-No There was an effect of diet (p=.0001) as well as a diet x time interaction (P=.004).

518

..J

..J

iii

a..

a. 0.75

~ ~ a:

..J

..J

Z ::::l

::::l III

m

C

c

::::l

8.50

i

:&

1.00

::E

E

0.05

E

Total Protein

c(

I/)

;:

0.11

aI

Figure 4. Effect of nine weeks of conditioning on urine urea-N and uric acid as expressed in a ratio to creatinine. There was an effect of diet (p=.011 and P=.0002, respectively) as well as a diet x training interaction for uric acid (P=.056).

Protein

C

"cc

0.17

c o

"C:I

_

Ii

aI

--

63

4.00 ..J

o

0.23

..J

1.50 , . - - - - - - - - - - - ,

. E

0.03

a: 0.00

:& m

i

4D

0.29

E 0.06

I-

2.50

c

0.12 , . - - - - - - - - - - - - - - , 0.35

E

o

c(

4D

.E

CI

I-

:&

m

u

z W I-

~

DLP ~HP _LP uric uric urea

;:

z

::::l

..•.

urea

RESULTS Horses remained in good health throughout the study. No signs of protein deficiency (poor coat, poor condition, JOURNAL OF EQUINE VETERINARY SCIENCE

-

... -

II

c c

....

;:

CD

-

"

18

-

!--e..-i -, - ---e--e--e ..&.

~ ..

14

-i

• •a..

Creatinine 2.S0

CD

W D: ::J

"6,,

Low Protein

,, '-

E E

-

22

II

u

High Protein

,~ ..

T

... "" -

.J.

-.L

E

T

2.00

.1

;:

T ~"tt 1"

'~ ..~

i

-" at

..... .. -

...I

T

"~

1.S0

w

Z Z

•i= •en • W D:

T,'

CJ

10

1.00

:IE

:IE til

6

...I

O.SO

R S1S2S3S4SSS6R1R2R3R4 REST SPRINTS RECOVERY

...I Q.

Figure 6. Effect of six one-minute sprints (81-86) and a 30-minute recovery (R1-R4) on plasma urea-N as expressed in a ratio to creatinine. There was an effect of diet (P=.0001) as well as an effect of exercise (P=.001). 30 ..I "0

Q

26

T

T

T

f-y-!

~

T

-

J.

T T ,e--t- _"_e __ e ,.-e' ~ T !

1

T.. J.

J.

E

c

w

22

a: J

c

::E

c

..

- HIGH

--

PROTEIN LOW PROTEIN

18

rn

..I Q.

-

DIET P=.OOO1

14 10

R S1 S2S3S4SSS6R1 R2R3R4 REST

SPRINTS

RECOVERY

Figure 7. Effect of nine weeks of conditioning on plasma urea-No There was an effect of diet (P=.0001) but no effect of exercise.

anemia or weakness) were observed. Average starting weight was 436± 17 kg and ending weight was 445± 17 kg, with no differences between diets (P = .53). Body condition scores (BCS) were 5.2±.4 initially and 5.7±.4 at the end of the study with no difference between diets (P = .33). Feed intake was not different between diets (P = .28) but due to protein level differences, average CP intake was 578 ± 29 gld on the LP diet and 1,117 ± 65 gld on the HP diet (P = .001). Conditioning. Training resulted in a 14% increase in plasma concentrations of albumin (P = .005) (Fig. 1, Table 3) and a 9% increase in creatinine (P = .033) (Fig. 2, Table 3). Total protein was decreased by 11 % as a result of training (P = .021) (Fig. 1, Table 3). Over the course of training, there was a training x diet interaction in plasma urea-N (P = .004) with the level increasing in the HP group and decreasing in the LP group over the course of the experiment (Fig. 3, Table 3). A diet x training interaction also existed for urine uric acid (P = .056) (Fig. 4) concentration increasing in the HP group Volume 20, Number 8, 2000

Table 3. Data are listed to show changes due to conditioning over nine weeks as well as the changes due to exercise during the standard exercise test. Training Plasma albumin (g/dL) Plasma total protein (g/dL) Plasma creatinine (mg/mL) Plasma urea-N (mg/dL) HP LP

Day 0 2.93±.04 7.92±.17 1.09±.04

Day 63 3.36±.09 7.0S±.14 1.19±.06

P value .005 .021 .033

20.09±1.27 23.73±1.07 .004 20.26±1.27 12.96±1.07 .004 Peak of exercise P value Rest (S6} 3.69±.11 .004 3.37±.OS 7.S9±.11 .001 7.1S±.14 1.6S±.OS .001 1.3±.07

Exercise Test Plasma albumin (g/dL) Plasma total protein (g/dL) Plasma creatinine (mg/mL) Plasma urea-N (mg/mL) HP 25.72±1.1 LP 16.04±1.1 Plasma urea-N (mg/mg creatinine) HP 20.76±.54 LP 11.91±.54

27.44±1.0 14.70±1.0

NS NS

16.S±.44 S.60±.44

.001 .001

but decreasing in the LP group. No effects of training on urine creatinine and urine urea levels were observed. SET exercise. Exercise resulted in a 10% increase in plasma total protein (P = .001) (Fig. 5, Table 3), a 29% increase in creatinine (P = .001) (Fig. 6, Table 3) and a 9.4% increase in plasma albumin (P = .004) (Fig. 5, Table 3). Plasma urea-N was relatively unchanged in the HP group, increasing by 7%, and in the LP group decreasing by 8% (Fig. 7, Table 3); however, the urea-N:creatinine ratio was decreased by 19% and 28%, respectively, for HP and LP during exercise (P = .001) (Fig. 6, Table 3). This difference was due to the increase in creatinine since when urea-N alone was reviewed the change in its level during exercise was much less. Since albumin and total protein were both increased by approximately 10%, a 10% loss of water from plasma can be estimated from this data. Heart rates averaged 195 bpm over the six sprints and 62 bpm at 30 minutes of recovery. There were no differences in heart rates due to diet. Diet. During the conditioning period, no effect of diet was observed in plasma albumin (P = 0.53) (Fig. 1), total protein (P = 0.93) (Fig. 1) or creatinine (P = 0.29) (Fig. 2). All values were within reference ranges. ll ,12 Plasma urea-N concentration and the urea-N:creatinine ratio were greater (P = 0.0001) in the HP group compared to the LP group (Fig. 3). Urine urea and uric acid concentrations as well as urine urea:creatinine and uric acid: creatinine ratios were greater (P = .011 and .0002, respectively) in the HP group compared to the LP group (Fig. 4). No differences were noted in urine creatinine levels (P = .78). During the exercise test, no differences were found in plasma concentrations of albumin (P = .32) (Fig. 5), total protein (P = .81) (Fig. 5) or creatinine (P = .39) (Fig. 6). Plasma urea-N (Fig. 7) as well as the urea-N:creatinine ratio (Fig. 6) were greater (P = .0001) in the HP group compared to the LP group. 519

DISCUSSION Feeding a 7.5% CP amino acid fortified diet revealed no detrimental effect on protein status during nine weeks of conditioning and exercise tests. Protein levels below current recommendations4 have been fed to exercising horses without detrimental effects. Inadvertently, 8.4% CP was fed to Arabian horses undergoing similar conditioning and SET to this study without any untoward effects. 13 A crude protein level as low as 5.5 % CP was suggested to be adequate for exercising horses by others. 14 This conclusion was based on normal albumin and total protein levels as well as a lack of outward signs of protein deficiency. However, the authors recommended a protein level of 8.5% CP, since the horses fed this level of protein appeared to have better protein reserves as indicated by increased plasma urea levels during a four-day fast. A level of 6% dietary CP was found to be adequate for exercising two-year-olds compared to non-exercising twoyear-olds suggesting improved use of the dietary protein with exercise. 15 In another study an increase in nitrogen retention in exercising horses was observed during conditioning as well as following the period of conditioning. 16 It was concluded that more protein (above maintenance) was needed to build and support muscle hypertrophy seen with exercise training. Plasma total protein and albumin in this study were not affected by diet but albumin increased and total protein decreased with conditioning. Since the main two components of plasma total protein being albumin and globulins, an overall decrease in TP and increase in albumin indicate a decrease in globulin level which has been observed as a result of exercise stress. Albumin is not regarded as a sensitive index of protein status in the horse; however, the length of the experiment (nine weeks) would appear to have been long enough to observe any adverse effect of dietary protein on albumin synthesis since the half life of albumin is about 18 days. I? The changes that occurred during exercise in this study show an increase in plasma albumin, TP, urea-N and creatinine with a decrease during recovery. These changes are most likely due to decreased plasma volume with increased hematocrit during strenuous exercise. The increase in plasma creatinine and its failure to return to resting values by 30 minutes of recovery probably reflect an increase in muscle activity as a result of exercise. The lack of increase in urea-N during exercise and recovery indicates that energy and protein were adequate, and did not require additional catabolism of muscle protein. Excess protein leads to several potentially detrimental effects including an increase in water requirements associated with increased urea production and excretion as well as an increase in urine ammonia increasing respiratory stress from exposure to the fumes in the environment as a result of stall

520

confinement. 3 The levels of urine urea in the HP horses are reflections of the level of nitrogen in the diets and demonstrate the increased excretion of urea in this treatment. Specific detrimental effects of high protein for the exercising horse have not been reported IS ,19,20; however, it seems logical that high protein intake producing more heat, urea and acid, would contribute to earlier fatigue. Those researchers determined that higher levels of protein did not effect heart rates and lactate levels. However, other parameters of exercise such as acid-base balance may be affected by high protein. I Protein metabolism is not only a function of quantity but depends on the amino acid profile, as well as digestibility. The addition of lysine and threonine to the LP diet improved the amino acid profile of that diet in relation to requirements. Lysine and threonine have been shown to be the limiting amino acids for growth in the horse.21,2~ Exercise has shown to increase protein digestibility in the horse and increased protein efficiency for gain in two-year-olds. 23 , 15 The adequacy of the LP diet in this study may have been due to the improved amino acid profile of the LP diet or to the improved utilization of the dietary protein due to exercise. Fat level may also affect protein digestibility. Improved ileal digestibility of amino acids was observed in growing pigs fed a fat supplemented diet. One possible explanation is that fat delays gastric emptying and thus improves digestibility.24 Since the diets in this study were high in fat content, this may have contributed to a more efficient use of the protein through improved digestibility.

CONCLUSION Overall, the results indicate that the LP diet supplemented with limiting amino acids supported normal protein status during a nine-week conditioning program and a repeated sprint test. Exercise and fat adaptation may have improved the availability of amino acids. Fortification of the LP diet with lysine and threonine appears to have prevented any negative effects on health and performance due to excess dietary amino acids associated with high protein intakes. A companion study suggested a favorable effect of the low protein diet fortified with limiting amino acids on acid-base responses in the sprint tests. 1

FOOTNOTES aHeartland Lysine Inc" Chicago, IL bMustang 2000, Kagra Inc" Fahrwangen, Switzerland cPolar Pacer, Port Washington, NY dlntramedic polyethylene tubing, Becton Dickinson, Sparks, MD eOperand, Redi Products, Inc" Prichard, WV IVEDCO, SI. Joseph, MO 9Protocol Systems Inc, hZ-O, 60mm, Dermalon, Davis & Geck iBaxter Healthcare Corp., Deerfield, IL

JOURNAL OF EQUINE VETERINARY SCIENCE

iElkins-Sinn Inc., Cherry Hill, NJ kEZ-weigh, Dyco, Cave Creek, AZ IVacutainers, Becton Dickinson, Rutherford, NJ mSherwood Medical, St.Louis, MO nSigma Diagnostics, St. Louis, MO

REFERENCES 1, Graham-Thiers PM, Kronfeld DS, Kline KA: Dietary protein influences acid-base responses to repeated sprints. Equine Vet J suppl. V30: in press. 2. Kronfeld DS, Ferrante PL, Grandjean D: Optimal nutrition for athletic performance, with emphasis on fat adaptation in dog and horses. J Nutr 1994; 124:2745S-2 2753S. 3. Meyer H: Nutrition of the equine athlete. In: Equine Exercise Physiology. Eds: J. Gillespie, N. Robinson. Ann Arbor, MI: Edward Brothers, 1987:p 644-673. 4. NRC: Nutrient requirements of horses. 5th edition, Washington, DC: National Academy Press, 1989. 5. Gallagher K, Leech J, Stowe H: Protein, energy and dry matter consumption by racing thoroughbreds: a field survey. J Equine Vet Sci 1992a;12(1):43-48. 6. Gallagher K, Leech J, Stowe H: Protein, energy and dry matter consumption by racing Standardbreds: a field survey. J Equine Vet Sci 1992b;12(6):382-388. 7. Southwood L, Evans D, Bryden W, Rose R: Nutrient intake of horses in Thoroughbred and Standardbred stables. Aust Vet J 1993;70: 164-168. 8. Honore EK, Uhlinger CA: Equine feeding practices in central North Carolina: a preliminary survey. J Equine Vet Sci 1994; 14(8):424-429. 9. Henneke DR, Potter GD, Kreider JL, Yeates BF: Relationship between condition score, physical measurements and body fat percentage in mares. Equine Vet J 1983;15:371-372. 10. SAS: SASISTAT User's Guide (Release 6.11) Cary, NC: SAS Inst Inc, 1991.

Volume 20, Number 8, 2000

11. Wilson E, Cumberland P, Green R: Chemistry reference values for domestic animals. The Southwestern Veterinarian 1986;37(2):125-127. 12. Weidemen H, Wilke p, DeWaal A: Effect of exercise and training on blood and biochemical indices in Thoroughbred horses. IntJAnimSci 1995;10:9-12. 13. Taylor LE, Kronfeld DS: Acid-base variables during incremental exercise in sprint trained horses fed a high fat diet. J Anim Sci 1995;73:2009-2018. 14. Patterson P, Coon C, Hughes I: Protein requirements of mature working horses. J Anim Sci 1985:61 (1):187-194. 15. Orton RK, Hume ID, Leng RA: Effects of level of dietary protein and exercise on growth rates of horses. Equine Vet J 1986;17:381-385. 16. Freeman D, Potter G, Schelling G, Kreider J: Nitrogen metabolism in mature horses at varying levels of work. J Anim Sci 1988;66:407-412. 17. Young VR, Marchini JS, Cortiella J: Assessment of protein nutritional status. J Nutr1990;120:1496-1502. 18. Miller PA, Lawrence LM: The effect of dietary protein level on exercising horses. J Anim Sci 1988;66:2186-2192. 19. Frank NB, Meacham TN, Fontenot JP: Effect of protein on performance and nutrition. J Equine Vet Sci 1987;7:321-322. 20. Miller-Graber PA, Lawrence LM, Foreman JH, Bump KD, Fisher MG, Kurcz EV: Dietary protein level and energy metabolism during treadmill exercise in horses. J Nutr 1991 ;121 :1462-1469. 21. Ott EA, Asquith RL, Feaster JP: Lysine supplementation of diets for yearling horses. J Anim Sci 1981 :53:1496-1504. 22. Graham PM, Ott EA, Brendemuhl JH, TenBroeck SH: The effect of supplemental lysine and threonine on growth and development of yearling horses. J Anim Sci 1994;72:380-386. 23. Worth MJ, Fontenot JP, Meacham TN: Physiological effects of exercise and diet on metabolism in the equine. Proc ofthe 121' Equine Nutrition and Physiology Symposium, 1987:p145-151. 24. Li S, Sauer WC:The effect of dietary fat content on amino acid digestibility in young pigs. J Anim Sci 1994;72:1737-1743.

521