https://doi.org/10.

1079/BJN19930162 Published online by Cambridge University Press

British Journul of Nutrition (1993), 70, 661-678 667

Protein and energy relationships in the broiler chicken

11. Effects of protein quantity and quality on metabolism*
BY R. W. R O S E B R O U G H A N D J. P. M c M U R T R Y
Non-ruminant Animal Nutrition Laboratory, Livestock and Poultry Sciences Institute, United States
Department of Agriculture-Agricultural Research Service, Beltsville Agricultural Research Center,
Beltsville, MD20 705, U S A

(Received 3 August 1992 - Accepted 4 December 1992)

Male broiler chickens growing from 7 to 35 d were fed on a diet containing 150 g crude protein
(N x 625)/kg diet supplemented with lysine to equal that in diets containing 166, 183 and 200 g crude
protein/kg diet (Expt 1).A second group of male broiler chickens growing over the same period were fed
on a diet containing 120 g crude protein/kg supplemented with lysine, arginine, tryptophan, threonine
and isoleucine equal to that in diets containing 144,172 and 200 g crude protein/kg diet (Expt 2). Growth
was improved by lysine supplementation but not to the level attained by feeding 200 g crude protein/kg
(Expt 1). Lysine, arginine, tryptophan, threonine and isoleucine supplementation of a low-protein diet
also improved growth, but growth again fell short of that attained by feeding a diet containing 200 g
crude protein/kg. Plasma insulin-like growth factor-1 and thyroxine concentrations increased and
triiodothyronine decreased as the crude protein level increased from 150 to 200 g/kg diet. Supplemental
lysine did not affect plasma levels of these hormones. Although dietary crude protein levels noticeably
changed rates of in vitro lipogenesis, changing either the level of a single limiting amino acid or the levels
of several limiting amino acids did not change lipogenesis.
Lipogenesis: Protein quality : Enzyme activity : Broiler chickens

Both the quantity (crude protein (N x 6.25) concentration) and quality (amino acid
composition relative to the required balance) of the dietary protein affect the body
composition of chickens (Khalil et af. 1968). For example, diets with small energy:protein
values promote lean broiler carcasses (Donaldson et af. 1956; Thomas & Combs, 1967)
while diets containing large energy :protein values promote high rates of in vitro lipogenesis
(Rosebrough & Steel, 1985) and de n o w carcass lipid synthesis by the liver (Donaldson,
1985). Recently, the effects of different energy: protein values have been studied and
different possible interpretations of results are possible (Rosebrough & Steele, 1985). For
example, compared with a diet containing a large energy: protein value, feeding a very small
energy : protein diet will result in a very lean carcass when crude protein is expressed relative
to dry matter. On the other hand, if carcass protein is expressed as g protein amassed over
time, the diet containing a larger energy : protein value resulted in more total carcass
protein. It is possible that diets containing very small energy: protein values promote lean
broilers by restricting energy consumption (Bartov, 1979). Excretion of surplus amino
acid-N may also require metabolic energy that would not be available for fat synthesis
(Buttery & Boorman, 1976).
Dietary protein quality reflects a balance of amino acids required for maximum growth
and lean tissue synthesis and the limiting amino acid levels in a protein source (Fisher et
* Mention of a trade name, proprietary product or vendor does not constitute a guarantee or warranty of the
product by USDA or imply its approval to the exclusion of other suitable products or vendors.
23-2
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
668 R . W. R O S E B R O U G H A N D J. P. M c M U R T R Y

al. 1959). Dietary protein level can be described by the lysine content of the diet relative to
the crude protein level. For example, Yeh & Leveille (1969) proposed that there should be
60 g lysine/kg protein in the diet of the chicken. Although the relationship of a limiting
amino acid to crude protein may regulate lipogenesis, it is not known if the effect is due to
the presence of a limiting amino acid at the ribosomal level or to a shift in cellular
metabolism caused by a need to process excess amino acid-C. In the process of converting
gluconeogenic amino acid-C into glucose-C, reducing equivalents (NADPH) normally
required for the addition of acetyl residues during de n o w lipogenesis are utilized and may
be unavailable for lipogenesis.
The purposes of the experiments described in the present report were : (1) to examine
lipid metabolism in chickens fed on additional amounts of the limiting amino acid lysine
as either lysine hydrochloride or soya-bean meal, (2) to continue the previously described
investigation in chickens fed on a lower level of crude protein, but supplemented with the
purported first five limiting amino acids (lysine, arginine, tryptophan, threonine and
isoleucine). The null hypothesis tested was that amino acid supplementation of low-protein
diets would influence metabolism similarly to a diet containing a greater amount of crude
protein. These regimens allow the testing of hypotheses relating to either the first limiting
amino acid in lipid metabolism or natural v. synthetic amino acids as improvers of protein
quality. In the first regimen a maize-soya-bean-meal basal diet was formulated to be
adequate for all essential amino acids except lysine. Lysine hydrochloride was then added
to meet requirements. In the second regimen an essential amino acid-deficient basal diet
was formulated, and then blended with either a diet containing 200 g protein/kg or a low-
protein dizt supplemented with amino acids to be equal to the diet containing 200 g
protein/kg.
MATERIALS A N D METHODS
Animals and diets
Expt 1. Ross male broiler chicks (7-d-old; n 336) were assigned to one of seven dietary
treatments (Table 1). Each treatment consisted of eight pen replicates of six chickens per
pen. The dietary treatments consisted of three stock diets: 150 g crude protein and 8 g
lysine/kg (diet A); 150 g crude protein and 12 g lysine/kg (diet G) and 200 g crude protein
and 12 g lysine/kg (diet D). Diets B and C were prepared by blending diets A and D, and
diets E and F were prepared by blending diets A and G. Diet A was formulated as adequate
in all required amino acids except lysine (National Research Council, 1984). Diet G was
formulated by adding lysine hydrochloride to the first diet so that this diet was now
adequate for all required amino acids, but contained a low level of crude protein. Diet D
met all amino acid needs for the 7- to 28-d-old chickens and served as a control diet.

Exp 2. Ross male broiler chicks (7-d-old; n 336) were assigned to one of seven dietary
treatments (Table 1). Each treatment consisted of eight pen replicates of six chickens per
pen. These treatments were again obtained by blending three diets (H, M and D). Diet H
was formulated to be deficient in the following amino acids, lysine, arginine, tryptophan,
leucine and isoleucine. Diet M was a modification of diet H such that the amino acids
mentioned were added to reach the following levels (g/kg diet): lysine 12, tryptophan 2.1,
threonine 5.7, isoleucine 7.8, arginine 14.2. Diet D was the same as in Expt 1 and served
again, as the control diet.
The chickens were housed in battery brooders in an environmentally controlled room
maintained at 23" with a 12 h lightdark cycle (06.00-18.00 hours light). Treatments were
randomly assigned to pens in each battery. Both feed and water were apportioned on an
ad lib. basis. For purposes of statistical analyses, the observation was the pen mean.
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
CHICKEN HEPATIC METABOLISM I N VITRO 669

Table 1. Expt 1. Composition of the diets

-
Diet ... A B C D E F G

Ingredient (g/kg)
Maize meal 800 766 733 700 800 800 800
Soya-bean meal 0 33 66 100 0 0 0
Soya-bean oil 20 22 23 25 20 20 20
Soya-bean protein* 85 90 95 100 85 85 85
L-Lysine hydrochloridet 0 0 0 0 2 4 6
L-Methioninet 5 5 5 5 5 5 5
Dicalcium phosphate 40 40 40 40 40 40 40
Limestone 10 10 10 10 10 10 10
Se premix1 1 1 1 1 1 1 1
Mineral premix5 1 1 1 1 1 1 1
Vitamin premix 11 5 5 5 5 5 5 5
Iodized salt 3 3 3 3 3 3 3
Sand 30 23 17 10 28 26 24
Composition (g/kg diet)
Crude protein (N x 6.25) 150 170 183 200 150 150 150
Methionine + cystine 9.5 9.9 10.4 10.8 9.5 9.5 9.5
Lysine 8.2 9.4 10.6 11.9 9.4 10.7 12.0
Tryptophan 1.5 1.7 1.9 2.2 1.5 1.5 1.5
Threonine 6. I 6.8 7.5 8.2 6.1 6. I 6. I
Isoleucine 7. I 8. I 9.1 10.1 7.1 7. I 7. I
Arginine 10.1 11.5 12.9 14.3 10.1 10-1 101

* Soya-bean protein grade 11 (900 g crude protein/kg); Nutritional Biochemicals, Cleveland, Ohio 44122,
USA.
t US Biochemicals, Cleveland, Ohio 44122, USA.
1 Provided 0.2 mg Se/kg diet.
5 Provided (mg/kg diet): Mn 100, Fe 100, Cu 10, Co 1, I 1, Zn 100, Ca 89.
11 Provided (mg/kg diet): retinol 3.6, cholecalciferol 0.075, biotin 1, vitamin E 10, riboflavin 10, pantothenic
acid 20, choline 2 g, niacin 100, thiamin 10, pyridoxine 10, menadione sodium bisulphite 1.5, cyanocobalamin 0.1,
folk acid 2, ethoxyquin 150.

Metabolic studies
In vitro lipogenesis
Chickens were randomly selected from each pen at 35 d and killed by cervical dislocation
at 09.00 hours. The livers were rapidly excised, weighed and placed in individual vessels
containing 10 m~-N-2-hydroxyethylpiperazine-"-ethane sulphonic acid (HEPES) and
155 mM-NaC1 (pH 7.5). A portion of the liver was sliced with a MacIlwain tissue chopper
(50-75 mg/explant) and duplicate explants were incubated for 2 h at 37" in 25 ml
Erlenmeyer flasks containing 3 ml Hanks' balanced salts (Hanks & Wallace, 1949)
supplemented with 10 mM-HEPES (pH 7.4) and 10 g bovine serum albumen/l. Under the
assay conditions described previously, reactions are linear from 1 to 4 h with substrate
concentrations from 5 to 40 mM (Rosebrough & Steele, 1987). Explants were incubated in
the presence of 10 mM-sodium [2-14C]acetate(18 disintegrations/min per nmol). Following
a 2 h incubation period, the liver explants were extracted for 24 h in 15 ml chloroform-
methanol (2 : 1, v/v) and then partitioned into aqueous and lipid phases with 3 ml 155 mM-
KCl (Folch et al. 1957). The bottom phase was evaporated to dryness, dispersed in
liquid-scintillation cocktail and counted by liquid-scintillation spectroscopy.
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
670 R. W . R O S E B R O U G H A N D J. P. M C M U R T R Y

Table 2. Expt 2. Composition of the diets

Diet ... H I J D K L M

Ingredient (g/kg)
Maize meal 800 766 733 700 800 800 800
Soya-bean meal 40 60 80 100 40 40 40
Soya-bean oil 30 28 27 25 20 20 30
Soya-bean protein* 25 50 75 100 45 45 25
L-Lysine hydrochloridet 0 0 0 0 2.5 5 7.5
L- Argininet 0 0 0 0 2.5 5 7.5
L-Methioninet 5 5 5 5 5 5 5
L-Tryptophant 0 0 0 0 0.3 0.6 1
L-Threoninet 0 0 0 0 0.3 0.6 1
L-Isoleucinet 0 0 0 0 0.9 1.8 2.8
Dicalcium phosphate 40 40 40 40 40 40 40
Limestone 10 10 10 10 10 10 10
Se premix$ 1 1 1 1 1 1 1
Mineral premix5 1 1 1 1 1 1 1
Vitamin premix11 5 5 5 5 5 5 5
Iodized salt 3 3 3 3 3 3 3
Sand 40 30 20 10 28 26 24
Composition (g/kg diet)
Crude protein (N x 6.25) 120 144 172 200 123 I29 134
Methionine + cystine 8.3 8.8 9.4 10.8 8.4 8.4 8.4
Lysine 5.9 7.9 9.9 11.9 7.9 9.8 12-0
Tryptophan 1.2 1.5 1.9 2.2 1.5 1.8 2.1
Threonine 4.8 5.9 7.5 8.2 5.0 5.4 5.7
Isoleucine 5.2 6.8 9.1 10.1 6.0 6.9 7.8
Arginine 7.3 9.6 12.9 14.3 9.6 11.9 14.2
~-

* Soya-bean protein grade I1 (900 g crude protein/kg) ; Nutritional Biochemicals, Cleveland, Ohio 44122,
USA.
t US Biochemicals, Cleveland, Ohio 44122, USA.
3 Provided 0.2 mg Se/kg diet.
5 Provided (mg/kg diet): Mn 100, Fe 100, Cu 10, Co 1, I 1, Zn 100, Ca 89.
(1 Provided (mg/kg diet): retinol 3.6, cholecalciferol 0.075, biotin I , vitamin E 10, riboflavin 10, pantothenic
acid 20, choline 2 g, niacin 100, thiamin 10, pyridoxine 10, menadione sodium bisulphite 1.5, cyanocobalamin 0.1,
folk acid 2, ethoxyquin 150.

Enzyme assays
Remaining liver tissue was homogenized in 100 mM-HEPES (pH 7.5)-3.3 mhl-p-mer-
captoethanol (1 : 10, w/v) and centrifuged at 50000 g for 60 min (Rosebrough et a(. 1988).
The supernatant fractions were kept at 0" until analysed for the activities of
+
malate : NADP oxidoreductase (decarboxylating) (EC 1 . 1 . 1 .40; MDH-NADP), iso-
citrate : NADP + oxidoreductase (decarboxylating) (EC 1 . 1 . 1 .42 ; ICD-NADP) and glu-
tamic-oxaloacetic aminotransferase (EC 2 . 6 . 1 . 1; GOT).
MDH-NADP activity was determined by a modification of the method of Hsu & Lardy
(1969). The reaction contained 50 mM-HEPES (pH 7 . 9 , 1 mM-NADP, 5 mM-MnC1, and
the substrate, 2.2 mM-L-malate (disodium salt). A 50 ,ul portion of the 50000 g supernatant
fraction was pre-incubated for 15 min in the presence of the first three ingredients. The
reaction was initiated by adding the substrate and following the rate of reduction of NADP
at 340 nm at 25". The reaction was found to proceed linearly for at least 60 min providing
that the reaction contained no more than 100 ,ug supernatant fraction protein.
ICD-NADP activity was determined by a modification of the method of Cleland et al.
(1969). The reaction contained 50 mM-HEPES (pH 7.5), 1 mM-NADP, 5 mM-MnC1, and
the substrate, 4.4 mM-DL-isocitrate.A 25 p1 portion of the 50000 g supernatant fraction was
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
CHICKEN HEPATIC METABOLISM I N VITRO 67 1
pre-incubated for 15 min in the presence of the first three ingredients. The reaction was
initiated by adding the substrate and following the rate of reduction of NADP at 340 nm
at 25". The reaction was found to proceed linearly for at least 60 min providing that the
reaction contained no more than 50 pg supernatant fraction protein.
GOT was determined by a modification of the method of Martin & Herbein (1976). The
reaction contained 50 mM-HEPES, 200 mM-L-aspartate, 0.2 mM-NADH, 1000 units
ma1ate:NAD + oxidoreductase (EC 1 . 1 . l . 37)/1 and the substrate, 15 m~-2-oxoglu-
tarate. A 25 pl portion of the 50000g supernatant fraction was pre-incubated for 15 min
in the presence of the first four ingredients. The reaction was initiated by adding the
substrate and following the rate of oxidation of NADH at 340 nm at 25". The reaction was
found to proceed linearly for at least 30 min providing that the reaction contained no more
than 50 pg supernatant fraction protein. Enzyme activities are expressed as pmol product
formed/min under the assay conditions (Rosebrough & Steele, 1985a).

Hormone assays
Both triiodothyronine (T,) and thyroxine (T,) concentrations were estimated with a
double-antibody procedure that has been outlined by May (1978). Plasma insulin-like
growth factor- 1 (IGF-1) was estimated with a heterologous radioimmunoassay as
previously described (Ballard et al. 1990). IlZ5-labelled IGF- 1 was purchased from
Amersham Corp. and recombinant chickens IGF-1 for the standard was supplied by
Gropep Ltd, Adelaide, SA, Australia. Primary antisera (rabbit anti-human IGF-1) was
kindly provided by Dr Geoff Francis, CSIRO, Adelaide, SA, Australia. All hormone assays
were conducted as single batches to remove inter-assay variation.

Statistical analyses
Both experiments were replicated twice with each treatment replicated a total of eight times.
Data were analysed as a randomized-block design with each replicate being considered as
a block. All interactions involving the blocking factor were pooled with the residual. Before
analyses, data for in vitro lipogenesis were ranked according to the magnitude of values.
The rank value was then used in an analysis of variance to decide statistical significance.
This transformation was necessary because of a lack of homogeneity of error variances.
Although standard errors are presented in the respective tables, mean separation was
accomplished by using the transformed value. The general linear models procedure (GLM)
of the analysis of variance was used to test the overall null hypothesis of inequality of
means. The Newman-Keuls range statistic was used to decide significance of pair-wise
comparisons (Remington & Schork, 1970).

RESULTS
Body weights, feed efficiencies and in vitro lipogenesis values for Expt 1 are presented in
Table 3. Chickens fed on diet A (150 g crude protein and 8 g lysine/kg) were equal in body
weight to chickens fed on diets B (166 g crude protein/kg) or E (150 g crude protein and
9.4 g lysine/kg diet). These chickens were lighter (P< 0.05) than chickens fed on diets G
(150 g crude protein and 12 g lysine/kg) or D (200 g crude protein/kg) which were the
heaviest chickens in the present experiment). Feed conversion efficiency was poorest in
chickens fed on diet A and best in the group fed on diet D. All levels of lysine
supplementation (diets E, F and G) resulted in feed conversion efficiencies which were
poorer than that of diet D.
The greatest rate of lipogenesis (P< 0.05) was noted in chickens fed on diet A. There was
a significant ( P < 0.05) linear decrease in lipogenesis as the dietary crude protein level
increased (diets B, C and D). In contrast, additions of lysine hydrochloride to diet A to
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
672 R. W. R O S E B R O U G H A N D J . P. MCMURTRY

Table 3. Expt 1. The effects of dietary crude protein (nitrogen x 6.25) and lysine
supplementation* on chicken body weight and in vitro lipogenesis at 35 d of &geS
(Mean values with their standard errors for eight pen means per dietary treatment)

Feed efficiency Lipogenesis

Wt (8) (gain/feed) @mol/g liver)

Diets Mean SE Mean SE Mean SE

A 896" 15.0 0.43" 0.013 43.2Cd 2.44

B 960" 63.6 0.49' 0.026 36.6'" 2.74
C 1156"" 44.2 0.50" 0020 30.lab 1.67
D 1244' 49.4 0.59" 0.017 23.7" 1.67
E 913" 9.2 0.5Ob 0.025 48.3'' 3.03
F 1116" 11.2 O.5lb 0.028 53.Se' 3.23
G 1136" 14.6 0.52' 0.013 59.5' 3.20

Mean values with unlike superscript letters were significantly different (P < 0.05).
* Chickens (7-d-old; average of 135 g) were assigned to one of the following dietary treatments (crude
protein-lysine; g/kg diet): A 15W3; B 170-9.4; C 183-10.6; D 200-1 1.9; E 150-9.4; F 150-10.7; G 15&12.0; for
details of composition, see Table 1. Chickens were fed for a 28 d experimental period and then selected from each
treatment to determine the effects of these diets on intermediary metabolism.
t In vitro lipogenesis was determined by culturing liver explants for 2 h in the presence of 10 mM sodium [2-
14C]acetateand by noting incorporation of acetate into hepatic lipids. Values are expressed as pmol substrate
incorporated per g liver.

Table 4.Expt 1. The effects of dietary crude protein (nitrogen x 6 2 5 ) level and lysine
supplementation* on metabolic hormone levels in chickens
(Mean values with their standard errors for eight pen means per dietary treatment,
expressed as ng/ml plasma)

IGF-1 T* T3 T, :T3

Diets Mean SE Mean SE Mean SE Mean SE

A 16.8" 0.60 6.8" 0.69 2.1" 0.12 2.9" 0.34

B 16.8" 0.92 8.2" 0.67 1.7ab 0.14 4.8"" 0.77
C 19.0" 1.04 9.0"" 0.76 1,6&' 0.08 5.6" 0.79
D 21-7' 0.89 10.9" 0.69 1.4" 0.06 7Pd 0.85
E 18.6" 1.10 8.2" 1.11 1.8b' 0.10 4.6ab 0.68
F 17.4" 0.92 10.8" 157 1.4" 0.14 7.7d 1.33
G 17.5" 0.67 7.0" 0.58 1.7"' 0.13 4.1"" 0.54

Mean values with unlike superscript letters were significantly different (P < 0.05).
IGF-I, insulin-like growth factor-1 ; T,, triiodothyronine; T4, thyroxine.
* Chickens (7-d-old; average of 135g) were assigned to one of the following dietary treatments (crude
protein-lysine; g/kg diet): A 150, 8; B 170, 9.4; C 183, 10.6; D 200, 11.9; E 150, 9.4; F 150, 10.7; G 150, 12; for
details of composition, see Table 1. Chickens were fed for a 28 d experimental period and then selected from each
treatment to determine the effects of these diets on intermediary metabolism.

equal those at the higher levels of crude protein were accompanied by an increase (P <
0.05) in lipogenesis.
Plasma IGF-1 was greatest (P < 0.05) in chickens fed on diet D (Table 4). There were no
significant differences between the other treatment groups. Plasma T, was also greater (P
< 0.05) in chickens fed on diet D when compared with values obtained from chickens fed
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
C H I C K E N H E P A T I C METABOLISM I N V I T R O 673

Table 5. Expt 2. The effects of dietary crude protein (nitrogen x 6.25) level and amino acid
supplementation* on chicken body weight and in vitro lipogenesisf at 35 d of age
(Mean values with their standard errors for eight pen means per dietary treatment)

Feed efficiency Lipogenesis

Weight (8) (gain/feed) (pmol/g liver)

Diet Mean SE Mean SE Mean SE

H 668" 14.1 0,27a 0.016 28,5d 2.11

I 859b 20.2 0,37b 0,013 24.7cd 2.03
J 93Vd 11.2 0.43Cb 0,033 14.6"b 1.75
D 1083" 21.4 0.46' 0.022 9.4" 2.19
K 894b" 25.7 0.41Cb 0.019 17.9h 3.84
L 916'' 21.8 0.46' 0,010 19.6hc 3.59
M 952' 15.7 0.46' 0.027 19.6bC 3.59

a-e Mean values with unlike superscript letters were significantly different (P< 0.05).
* Chickens (7-d-old; average of 135 g) were assigned to one of the following dietary treatments: D, H-J,
deficient in lysine, arginine, tryptophan, threonine and isoleucine but with increasing levels of crude protein (200,
120, 144, 172 g/kg respectively; K-M, supplemented with increasing amounts of lysine, arginine, tryptophan,
threonine and isoleucine, so that the levels in diet M were equal to those in diet D, but with low crude protein
levels (123, 129, 134 g/kg respectively); for details of composition, see Table 2. Chickens were then fed for a 28 d
experimental period and then selected from each treatment to determine the effects of dietary treatments on
intermediary metabolism.
7 In vitro lipogenesis was determined by culturing liver explants for 2 h in the presence of 10 mM sodium [2-
"Clacetate and by noting incorporation of acetate into hepatic lipids. Values are expressed as pmol substrate
incorporated per g liver.

on the diets containing either 166 crude protein/kg diet (diet B) or the two higher levels of
lysine supplementation (diets F and G). Plasma T, was lowest in chickens fed on diet D and
greatest in chickens fed on diet A. Lysine supplementation to give either 10.7 or 12 g
lysine/kg diet (diets F and G respectively) also decreased plasma T,. Likewise, increasing
the dietary crude protein level decreased plasma T,. Plasma T,:T, increased as dietary
crude protein increased ; however, analysing this trend was not as simple when lysine was
added to basal diet. For example, the addition of lysine hydrochloride to give 10.7 g/kg diet
(diet F) significantly increased (P < 0.05) the ratio (from 2.9 to 7.7); however, a further
addition of lysine hydrochloride to give 12 g/kg diet (diet G) decreased the ratio (from 7.7
to 4.1) compared with diet A.
The effects of feeding diets containing either increasing levels of crude protein or low-
protein diets supplemented with lysine, arginine, tryptophan, threonine and isoleucine on
growth, feed efficiency and in vitro lipogenesis are shown in Table 5. There was, again, a
significant (P < 0.05) linear effect of crude protein on body weight. The addition of lysine,
arginine, tryptophan, threonine and isoleucine also resulted in a linear increase in growth.
Feed conversion efficiency was also improved by amino acid supplementation, although the
two higher levels of supplementation gave the same efficiency value. As in Expt 1, there was
a linear improvement in efficiency as the crude protein level increased.
As in Expt 1, increasing dietary crude protein described (P < 0.05) in vitro lipogenesis.
Compared with diet D, amino acid supplementation resulted in higher rates of lipogenesis;
however, these same diets did result in rates lower than the unsupplemented low-protein
basal diet (diet H).
Table 6 shows the effects of diet upon plasma hormone levels. There were no significant
differences in plasma T, values between chickens consuming diet H (120 g crude
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
674 R. W. R O S E B R O U G H A N D J. P. M c M U R T R Y

Table 6. Expt 2. The efsects of dietary crude protein (nitrogen x 6.25) level and amino acid
supplementation* on metabolic hormone levels in chickens
(Mean values with their standard errors for eight pen means per dietary treatment,
expressed as ng/ml plasma)

IGF-I T* T3 T, : T,

Diets Mean SE Mean SE Mean SE Mean SE

H 18.6" 0.40 9.1" 0.83 1.9" 010 5.0" 0.39

1 20.3'" 0.72 10.9abc 0.92 1.6' 0.13 7.2" 0.99
J 19.5" 0.68 12.4"' 0.58 1.3"" 0.07 9.4' 0.44
D 21.5" 062 12.9' 0.66 1.2" 0.08 11.8' 1.27
K 18.5" 0.74 107ab 0.49 1.4abc 0.07 8.1br 0.62
L 20.1"" 0.93 9.3a 0.59 1.4abc 0.09 6.9'" 0.55
M 19.9"" 079 12.0"' 0.94 1.5"' 0.09 7.9hc 0.79

Mean values with unlike superscript letters were significantly different (P < 0.05).
IGF-1, insulin-like growth factor 1 ; T,, triiodothyronine; T,, thyroxine.
* Chickens (7-d-old; average of 135 g) were assigned to one of the following dietary treatments: D, H-J,
deficient in lysine, arginine, tryptophan, threonine and isoleucine but with increasing levels of crude protein (200,
120, 144, 172 g/kg respectively); K-M, supplemented with increasing amounts of lysine, arginine, tryptophan,
threonine and isoleucine, so that the levels in diet M were equal to those in diet D, but with low crude protein
levels (123, 129, 134 g/kg respectively); for details of composition, see Table 2. Chickens were then fed for a 28 d
experimental period and then selected from each treatment to determine the effects of dietary treatments on
intermediary metabolism.

Table 7. Expt 2. The effects of dietary crude protein (nitrogen x 6.25) level and amino acid
supplementation* on hepatic enzyme activities? in chickens
(Mean values with their standard errors for eight pen means per dietary treatment,
expressed as units/g liver)

MDH-NADP GOT ICD-NADP

Diets Mean SE Mean SE Mean SE

H 17.0" 0.74 50.2ab 5.96 23.4ab 0.74

I 15.4" 1.01 49.5a" 2.84 21.9" 1.75
J 13.1b 1.16 55.0"" 4.98 26.0"" 2.09
D 8.6" 1.32 68.2' 3.46 31.5' 1.32
K 23.6' 1.95 57.0b 5.24 28.4bC 2.76
L 19.4" 2.71 48.7"b 3.44 22.8" 1.83
M 21.0"' 1.83 46.6" 2.39 23.7'" 1.83

Mean values with unlike superscript letters were significantly different ( P < 0.05).
MDH-NADP, malate: NADP+ oxidoreductase (decarboxylating) (EC 1.1.1.40); GOT, glutamic-oxaloacetic
aminotransferase (EC 2 . 6 . 1 . 1 ) ; ICD-NADP, isocitrate: NADP + oxidoreductase (decarboxylating)
(EC 1.1.1.42).
* Chickens (7-d-old; average of 135 g) were assigned to one of the following dietary treatments: D, H-J,
deficient in lysine, arginine, tryptophan, threonine and isoleucine but with increasing levels of crude protein (200,
120, 144, 172 g/kg respectively; K-M, supplemented with increasing amounts of lysine, arginine, tryptophan,
threonine and isoleucine, so that the levels in diet M were equal to those in diet D, but with low crude protein
levels (123, 129, 134 g/kg respectively); for details of composition, see Table 2. Chickens were then fed for a 28 d
experimental period and then selected from each treatment to determine the effects of dietary treatments on
intermediary metabolism.
t One unit is that amount of enzyme resulting in the production of 1 pmol oxidized or reduced NAD(P)/min
at 25".
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
CHICKEN HEPATIC METABOLISM I N V I T R O 675
protein/kg) and diets I (144 g crude protein/kg), K and L (first two increments of amino
acid supplementation). Both diet D (200 g crude protein/kg) and diet M (highest rate of
amino acid supplementation) increased T, compared with diet H. All dietary treatments
decreased T, relative to diet H. Increments of lysine, arginine, lysine, tryptophan, threonine
and isoleucine (diets K and L respectively) gave similar plasma IGF-1 values, as did
increases in crude protein (diets I and J respectively). Compared with diet H, the greatest
IGF-1 values were noted in chickens fed on diet D (200 g crude protein/kg diet).
The activities of certain hepatic enzymes are presented in Table 7. The activities of GOT
and ICD-NADP were greater in the group fed on diet D than in any of the other treatment
groups. Although amino acids were added to the basal diet (H) to equal those in diet D,
the activities of these two enzymes were not decreased compared with those obtained with
diet H. MDH-NADP activity was decreased by increasing the dietary crude protein level
from 120 to 200 g/kg diet (diets H > I > J > D). Adding methionine, lysine, tryptophan,
threonine and isoleucine to diet H to equal those in the control diet (D) did not change
MDH-NADP activity.

DISCUSSION
A logical progression in the study of the role of protein per se in the regulation of
metabolism involves further experiments concerning protein quality (amino acid
composition). The first experiment in the present study addresses lysine as a limiting factor
in protein quality. The second experiment involves supplementation of a low-protein diet
with several amino acids to obtain both the proper balance and quantity. The results of the
present study concerning metabolism in chickens fed on diets containing a marginal crude
protein level both complement and conflict with previous work (Rosebrough e f al. 1990)
concerning the feeding of lysine-adequate and -inadequate diets. In that study a marginal
level of crude protein (adequate in all amino acids with exception of lysine) was fed to
growing chickens. It was shown that a diet containing a marginal level of crude protein
(150 g/kg) could be supplemented with lysine to give a growth rate similar to that attained
with a diet containing a higher level of crude protein (200 g/kg). In contrast, changes in
intermediary metabolism such as oxidation, lipogenesis and liver glucose production,
however, were not equal under these conditions. In the present study growth was not
completely restored by feeding this level of lysine supplementation. It should be pointed
out, however, that different age birds were used in the study of Rosebrough et ul. (1990;
28-d-old v. 35-d-old chickens in the present study).
The present study also expands upon our previous efforts by showing that graded
amounts of lysine or protein will improve both growth and feed efficiency, although it
should be noted that lysine supplementation to equal that in the control diet did not result
in equal growth. Likewise, a very low level of crude protein (120 g/kg diet) was used to test
the effect of the supplementation of several amino acids. In this case chicken growth
performance was also improved compared with the basal, unsupplemented diet. Again,
growth did not equal that in the group fed on the control diet.
Although growth could be improved by incremental increases in either lysine (Expt 1) or
a balanced mix of limiting amino acids (Expt 2), plasma IGF-1 differed only when control
values (200 g crude protein) were compared with basal, unsupplemented diets (1 50 g crude
protein in Expt 1 and 120 g crude protein in Expt 2). The usefulness of assaying plasma IGF-
1 concentrations as an indicator of the metabolic status of animals has met with varying
degrees of success. For example, in comparing plasma IGF- 1 concentrations among animal
species, lower concentrations were noted in the chicken (Leung et al. 1986) than in the
growing rat (Prewitt et ul. 1982), although trends associated with growth are similar in both
species. Growth retardation in chickens may be due also to reduced systemic IGF-1 as well
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
676 R. W. R O S E B R O U G H A N D J . P. M c M U R T R Y

as increased catabolic hormones. Buyse et a/. (1986) used exogenous corticosterone to

reduce growth and increase body fat of chickens and found a decrease in plasma IGF-1
concentration. Huybrechts et al. (1985) surveyed plasma IGF-1 concentrations in growing
meat-and egg-type chickens and found an age-dependent decrease in plasma IGF-1 in the
egg-type chicken but not in the meat-type chicken. The latter observation is not surprising
because the meat-type chicken has been intensively selected for rapid growth.
Several recent studies provide evidence for nutrient control of plasma IGF- 1. Refeeding
a diet containing non-essential amino acids increased plasma IGF- 1 in rats to a lesser extent
than refeeding a diet containing essential amino acids (Isley et al. 1984; Clemons et al.
1985a, 6). Prewitt et al. (1982) also noted a linear increase in plasma IGF-1 with increasing
protein intake coupled to a moderate energy restriction (75% of ad lib. intake). A more
recent study (Lauterio & Scanes, 1987) compared plasma IGF-1 concentrations in chickens
as a function of dietary protein and showed a decrease in plasma IGF-1 when chicks were
switched from a diet containing 200 g crude protein/kg to one containing 50 g crude
protein/kg. Correspondingly, the opposite feeding regimen increased plasma IGF- 1. The
change in plasma IGF-1 in this study was attributed to changes in protein nutritional
status. An examination of feed intakes in the latter study cast some doubt on this
hypothesis because feed intake was not controlled. Thus, protein intake per se could not be
separated from the energy intake of the chickens in this study. In a later study (Rosebrough
et al. 1988) growing chickens were used to test the relationship between protein and energy
intakes and various indices of growth; in this study a feed restriction regimen was practised
so that fixed quantities of both protein and energy were fed. Under the condition of a
restricted energy intake (70% of ad lib. intake) and two levels of crude protein intake,
plasma IGF-1 was greater in those chickens given the greater amount of protein. Plasma
IGF-1 concentrations matched both growth and relative breast muscle size, suggesting that
this hormone regulated lean tissue development. It is tempting to speculate that the increase
in muscularity (increase in relative breast muscle size) in chickens given a higher daily
protein allotment may relate to IGF-1. It was concluded from the Rosebrough et al. (1988)
study that plasma IGF-1 concentration was a useful indicator of dietary protein adequacy
given that energy intake was also considered.
Analyses of data also showed that thyroid hormone values may be of limited usefulness
in determining the protein nutritional status of chickens. In the present study large
differences among dietary treatments were required because of the large variation in values.
Yang et al. (1987) indicated that, although a reduction in carbohydrate energy decreased
body weight, there were no changes in either T, or T,. The findings of Yang et al. (1987)
suggested that: (1) growth could be separated from thyroid function and (2) growth
retardation was related to the quantity of the dietary energy rather than its composition.
Few studies have presented any biochemical logic for the decrease in lipogenesis
accompanying the feeding of high-protein diets. Bartov (1979) stated that excess dietary
protein also forced the chicken to use energy to excrete excess N as uric acid. Thus, less
energy would be available for lipogenesis (Buttery & Boorman, 1976). Yeh & Leveille
(1969) found an inverse relationship between the level of the dietary protein and the ensuing
rate of in vivo lipogenesis. They thought that an increase in the dietary protein level
decreased the flow of substrates through glycolysis and increased the production of glucose
from substrates formerly in the pathways leading to fat synthesis. Reducing equivalents
produced through the normal operation of the Krebs Cycle would be used for glucose
production because of a higher state of reduction compared with glucose precursors (Yeh
& Leveille, 1969). It has also been hypothesized that under conditions of feeding high-
protein diets transamination of amino acids would deplete 2-oxoglutarate and force citrate
from lipid synthetic pathways to 2-oxoglutarate production to support transamination
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
CHICKEN H E P A T I C METABOLISM IN V l T R O 677
(Rosebrough & Steele, 1986b). In addition, malate transport from the mitochondrion
requires 2-oxoglutarate influx. Thus, malate availability for the reaction catalysed by
MDH-NADP may depend on citrate utilization and the production of 2-oxoglutarate.
The enzyme activities in the present study also suggest that ICD-NADP may function in
both lipid and protein metabolism by providing a residual capacity for the production of
reducing equivalents during a period of decreased MDH-NADP activity. Intracellular
competition may exist between acetyl-CoA carboxylase (EC 6 . 4 . 1 .2) and the aconitase
(EC 4.2.1.3)-isocitrate dehydrogenase (EC 1.1.1.42) pathway for limited cytoplasmic
citrate. The requirement for 2-oxoglutarate (a product of the reaction catalysed by ICD-
NADP as a co-reactant for transamination, occurring during increased dietary y-otein
intake, depresses citrate levels to that activation of acetyl-CoA carboxylase would not
occur. Clark et al. (1979) reported that avian acetyl-CoA carboxylase was particularly
sensitive to citrate levels. The latter report seems to offer some support for the role of high-
protein diets as regulators of lipogenesis via citrate availability.
The results of the present study support those of Tanaka et a[. (1983) that show that in
the face of a constant energy intake an increase in protein energy will result in a decrease
in lipogenic enzyme activities. Although MDH-NADP may provide the necessary NADPH
for lipogenesis, the enzyme may not strictly regulate lipogenesis according to the findings
in the present study. For example, the lowest noted activity (attained by feeding a diet
containing 200 g crude protein/kg diet) would result in the formation of nearly five times
the NADPH required for the de n o w synthesis of the fatty acids noted for this group.
In summary, the present study using growing chickens was designed to test the
relationship between either crude protein or protein quality and various indices of growth
and intermediary metabolism. Plasma IGF- 1 concentrations paralleled growth and may
either regulate lean tissue development or reflect changes in lean tissue synthetic rates. It
is likely that plasma IGF-1 concentrations can be used as an indication of dietary protein
adequacy given dietary treatment effects are large and that energy intake is considered with
both protein quantity and quality. In contrast, an apparent improvement in the protein
quality of a low-protein diet did not depress lipogenesis when compared with feeding an
unimproved diet.
The hypothesis of the present study was that an improvement in protein nutriture,
attained by adding either a limiting amino acid to an otherwise adequate diet or by adding
several limiting amino acids to a marginal level of crude protein, would repartition energy
away from lipogenesis. This hypothesis was not substantiated by the present findings.
Therefore, the regulation of lipogenesis in the broiler is more complex than merely meeting
some amino acid need for lean tissue synthesis, and thereby, directing energy away from
fat synthesis.

REFERENCES
Ballard, F. J., Johnson, R. J., Owens, P. C., Francis, G. L., Upton, F. M., McMurtry, J. P. & Wallace, J. C.
(1990). Chicken insulin-like growth factor-I : Amino acid sequence, radioimmunoassay, and plasma levels
between strains and during growth. General and Comparative Endocrinology 79, 459469.
Bartov, I. (1979). Nutritional factors affecting quantity and quality of carcass fat in chickens. Federafion
Proceedings 38, 2627-2639.
Buttery, P. J. & Boorman, K. N. (1976). The energetic efficiency of amino acid metabolism. Protein M e t u b n h n
und Nurrifion, pp. 197-204 [D. J. A. Cole, editor]. London: Butterworths.
Buyse, J., Decuypere, E., Huybrechts, L. M., Kuhn, E. R. & Whitehead, C. (1986). Effect of corticosterone on
circulating concentrations of corticosterone, prolactin, thyroid hormones and somatomedin-C and on fattening
in broilers selected for high or low fat content. Journal of’Eizdoocrinologjj 112, 229-237.
Clark, S. D., Watkins, P. A. & Lane, M . D. (1979). Acute control of fatty acid synthesis by cyclic A M P in the
chick liver cell: possible site of inhibition of citrate formation. Journal qf Lipid Research 20, 974985.
Cleland, W. W., Thompson, V. M. & Barden, R. E. (1969). Isocitrate dehydrogenase (TPN specific) from pig
heart. Mefhods iiz Enzymilogy 13, 30-33.
 https://doi.org/10.1079/BJN19930162 Published online by Cambridge University Press
678 R. W. R O S E B R O U G H A N D J. P. M c M U R T R Y

Clemons, D. R., Seek, M. M. & Underwood, L. E. (1985a). Supplemental essential amino acids augment the
somatomedin-C/insulin-like growth factor response to refeeding after fasting. Melabolism 34, 39 1-395.
Clemons, D. R., Underwood, L. E., Dickerson, R. N., Brown, R. O., Hak, L. J., MacPhee, R. D. & Heizer, W. D.
(1985b). Use of plasma Somatomedin-C/insulin-like growth factor I measurements to monitor the response to
nutritional repletion in malnourished patients. American Journal of Clinical Nutrition 41, 191-198.
Donaldson, W. E. (1985). Lipogenesis and body fat in chicks: Effects of ca1orie:protein ratios and dietary fat.
Poultry Science 64,1199-1204.
Donaldson, W. E., Combs, G. F. & Romoser, G. L. (1956). Studies on energy levels in poultry rations. 1. The effect
of calorie-protein of the ration on growth, nutrient utilization and body composition of chicks. Poultry Science
35, 1100-1204.
Fisher, H., Griminger, P. & Leveille, G. A . (1959). Protein depletion and amino acid requirement in the growing
chick. Journal of Nutrition 69, 117-123.
Folch, J., Lees, M. & Sloane-Stanley, G. H. (1957). A simple method for the isolation and purification of total
lipids from animal tissues. Journal of Biological Chemislry 226, 497-509.
Hanks, J . H. & Wallace, R. E. (1949). Relation of oxygen and temperature in the preservation of tissues by
refrigeration. Proceedings of the Society of Experimental Biology and Medicine 71, 196200.
Hsu, R. Y. & Lardy, H. A. (1969). Malic enzyme. Methods in Enzymology 13, 23&235.
Huybrechts, L. M., King, D. B., Lauterio, T. J., Marsh, J. & Scanes, C. G. (1985). Plasma concentrations of
somatomedin-C in hypophysectomized, dwarf and intact domestic fowl as determined by heterologous
radioimmunoassay. Journal of Endocrinology 104, 233-239.
Isley, W. L., Underwood, L. E. & Clemons, D. R. (1984). Changes in plasma somatomedin-C in response to
ingestion of diets with variable protein and energy content. Journal of Purenteral and Enteral Nutrition 8,
40741 1,
Khalil, A. A,, Thomas, 0. P. & Combs, G. F. (1968). Influence of body composition, methionine deficiency or
toxicity and ambient temperature on feed intake in the chick. Journul of Nutrition 96, 337-341.
Lauterio, T. J. & Scanes, C. G. (1987). Hormonal responses to protein restriction in two strains of chickens with
different growth characteristics. Journal oj' Nutrifion 17, 758-763.
Leung, F. C., Taylor, J. E., Wein, S. & Van Iderstine, A. (1986). Purified chicken growth hormone and a human
pancreatic GH-releasing hormone increase body weight gain in chickens. Endocrinology 118, 1961-1965.
Martin, R. J. & Herbein, J. H. (1976). A comparison of the enzyme levels and in vitro utilization of various
substrates for lipogenesis in pair-fed lean and obese pigs. Proceedings of the Society for Experimental Biology
and Medicine 151, 231-235.
May, J. (1978). A radioimmunoassay for 3,5,3-triiodothyronine in chicken serum. Poultry Science 57, 243-260.
National Research Council (1984). Nutrient Requirefnents of Poultry. Washington, DC: National Academy Press.
Prewitt, T. E., D'Ercole, A. J . , Switzer, B. R. & Van Wyk, J . J. (1982). Relationship of serum immunoreactive
somatomedin-C to dietary protein and energy in growing rats. Journal of Nutrition 112, 144150.
Remington, R. D. & Schork, M. A. (1970). Statistics with Applications to the Biological and Health Sciences.
Englewood Cliffs, N.J. : Prentice-Hall.
Rosebrough, R. W., McMurtry, J. P., Mitchell, A. D. & Steele, N. C. (1988). Protein and energy restrictions in
the broiler chicken. 6. Effect of dietary protein and energy restrictions on carbohydrate and lipid metabolism
and metabolic hormone profiles. Comparative Biochemistry and Physiology 90, 31 1-31 6.
Rosebrough, R. W., Mitchell, A. D., Von Vleck, M. F. & Steele, N. C. (1990). Protein and energy relations in the
broiler chicken. 8. Comparisons involving protein- and iysine-adequate diets on lipid metabolism. British
Journal of Nutrilion 64, 5 15-523.
Rosebrough, R. W. & Steele, N. C. (1985). Energy and protein relations in the broiler. 1. Effect of protein levels
and feeding regimes on growth, body composition, and in vitro lipogenesis of broiler chicks. Poultry Science 64,
119-126.
Rosebrough, R. W. & Steele, N. C. (19860). Protein and energy relations in the broiler. 3. Growth and in vitro
metabolism in male and female chickens used as parent stock. Growth 50, 63-75.
Rosebrough, R. W. & Steele, N. C. (1986b). Energy and protein relations in the broiler. 4. Role of sex, line. and
substrate on in ritro lipogenesis. Growth 50, 461471.
Rosebrough, R. W. & Steele, N. C. (1987). Methods to assess glucose and lipid metabolism in avian liver explants.
Comparative Biochemistry and Physiology 88A, 1041-1049.
Tanaka, T., Ohtani, S. & Shigeno, K. (1983). Effect of increasing dietary energy on hepatic lipogenesis in growing
chicks. 11. Increasing energy by carbohydrate supplementation. Poultry Science 62, 44504518.
Thomas, 0. P. & Combs, G. F. (1967). Relationship between serum protein level and body composition in the
chick. Journal of Nutrition 91, 468472.
Yang, H., Cree, T. C. & Schalch, D. S. (1987). Effect of a carbohydrate-restricted, calorie-restricted diet on the
growth of young rats and on serum growth hormone, somatomedins, total thyroxine and triiodothyronine, free
T, index and corticosterone. Metubolism 26, 794798.
Yeh, Y . Y. & Leveille, G. A. (1969). Effect of dietary protein on hepatic lipogenesis in the growing chick. Journal
of Nutrition 98, 356-366.
Printed in Great Britain