The dietary valine requirement for prolific lactating sows does not exceed the National
Research Council estimate
A. M. Gaines, R. D. Boyd, M. E. Johnston, J. L. Usry, K. J. Touchette and G. L. Allee
J Anim Sci 2006. 84:1415-1421.
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://jas.fass.org/cgi/content/full/84/6/1415
www.asas.org
Downloaded from jas.fass.org by on November 12, 2008.
The dietary valine requirement for prolific lactating sows does not exceed
the National Research Council estimate1,2
A. M. Gaines,* R. D. Boyd,†2,3 M. E. Johnston,†3 J. L. Usry,‡
K. J. Touchette,*4 and G. L. Allee*
*Department of Animal Science, University of Missouri, Columbia 65211;
†Pig Improvement Company, Franklin, KY 42135; ‡Ajinomoto Heartland LLC, Chicago, IL 60631
ABSTRACT: Two studies were conducted to determine the effect of increasing the valine:lysine (V:L) ratio
in diets of lactating sows above the minimum proposed
by the NRC (1998). The first experiment involved 189
PIC, Camborough product sows (parity 1 to 4) that were
allotted to 1 of 3 dietary treatments. Diets were formulated to achieve total dietary V:L ratios of 0.90, 1.05,
or 1.25:1, respectively, and were corn and soybean mealbased. The second experiment involved 279 PIC, Camborough sows (parity 1 to 5) that were allotted to 1 of 4
treatments. Diets 1 and 3 were formulated using corn
and a fixed inclusion of soybean meal (16.7%), with
0.27% L-lysine HCl. The V:L ratios in diets 1 and 3 were
0.73 and 1.25:1, respectively. Diets 2 and 4 were typical
corn-soybean meal diets containing 0.05% L-lysine HCl,
with a fixed inclusion of soybean meal (22.7%). The V:L
ratios in diets 2 and 4 were 0.86 and 1.25:1, respectively.
In both experiments, each litter was standardized to a
minimum of 10 pigs, which achieved litter growth rates
of 2.22 and 2.56 kg/d in Exp. 1 and 2, respectively. In
Exp. 1, increasing the dietary V:L ratio beyond 0.90:1
did not improve (P > 0.10) the number of pigs weaned,
survival rate, or piglet growth rate, even though sows
were nursing more than 10 pigs per litter for 19 d. In
Exp. 2, total lysine intake was similar among treatments and ranged from 52.1 to 55.3 g/d. Valine intake
increased as the diet valine concentration increased
(diet 1 vs. 3 and diet 2 vs. 4, P < 0.001), ranging from
40.0 to 66.1 g/d. Litter gain tended to improve (P = 0.14)
when the 0.27% L-lysine HCl control (0.73 V:L) was
supplemented with valine to achieve a 1.25:1 V:L ratio.
In contrast, no aspect of sow or litter response was
improved when the practical control diet containing
0.05% L-lysine HCl (0.86 V:L) was supplemented with
valine to achieve a 1.25:1 V:L ratio. Collectively, this
research shows that a V:L ratio in excess of 0.86 does not
conserve maternal tissue loss or improve piglet growth
rate, but a V:L ratio of 0.73 may compromise litter
growth rate.
Key words: amino acid, lactation, valine
2006 American Society of Animal Science. All rights reserved.
INTRODUCTION
The NRC (1998) proposed a minimum dietary ratio
of valine:lysine (V:L) of 0.85:1 for high milk producing
sows. This true ileal digestible ratio is greater than
previously estimated by the ARC (1981) and Pettigrew
(1993) but lower than in a prior NRC edition (1988).
1
Research was conducted at the Pig Improvement Company Research Farm, Franklin, KY 42135. The authors recognize the technical assistance of E. Wilson and the financial support of Ajinomoto
Heartland LLC. Amino acid analysis was conducted by C. Deimerly
of the Ajinomoto Heartland analytical lab, for which we are grateful.
2
Corresponding author: [email protected]
3
Present address: The Hanor Company, Inc., P. O. Box 881, Franklin, KY 42134.
4
Present address: Cargill Animal Nutrition, Elk River, MN 55330.
Received February 2, 2005.
Accepted December 15, 2005.
J. Anim. Sci. 2006. 84:1415–1421
The estimate was based on evidence, at the whole-animal level and the tissue level (Boyd et al., 1995), that the
valine requirement is greater than would be predicted
from the amount secreted in milk (NRC, 1998). This
estimate of absorbed V:L is in relative agreement with
the initial mammary transmembrane transport ratio
estimate by Trottier et al. (1997).
A number of empirical studies have been conducted
evaluating valine nutrition of lactating sows, but there
is too much disparity among these trials to reach a
reliable conclusion. Richert et al. (1996; 1997a,b) reported that a V:L ratio of at least 1.15:1 was needed to
maximize litter growth rate. Similarly, Moser et al.
(2000) reported an improvement in litter growth rate
as the V:L ratio increased from 0.89 to 1.33:1 in sows
that were nursing large litters (≥10 pigs/litter). In contrast, Carter et al. (2000) reported that a V:L ratio of
0.89:1 was sufficient to meet the need for litter growth
of sows nursing 10 or more pigs. Southern et al. (2000)
1415
Downloaded from jas.fass.org by on November 12, 2008.
1416
Gaines et al.
also reported no improvement in sow productivity when
the V:L ratio was increased from 0.85 to 0.94:1, using
hydrolyzed feather meal as a source of valine. The 2
latter reports are in agreement with the NRC (1998);
however, the reason for the discrepancy between these
data and the earlier reports is not clear. With the exception of the data by Moser et al. (2000), the studies where
an increase in the V:L ratio beyond 1:1 proved beneficial
to litter growth involved diets containing substantial
amounts of crystalline AA (Richert et al., 1996;
1997a,b).
This research was conducted to determine if a dietary
V:L ratio in excess of that suggested by the NRC (1998)
conserves maternal body tissue or increases litter
growth rate in sows nursing more than 10 pigs. Practical corn-soybean meal diets were used. Two of the semisynthetic diets, used by Richert et al. (1996), were reproduced to see if we could duplicate the response. The
PIC pig research and care committee approved animal
use protocols in this research.
MATERIALS AND METHODS
Experiment 1
One hundred eighty-nine PIC, Camborough product
sows (parity 1 to 4) were used to determine the effect
of increasing the V:L ratio (dose-response assay) for
lactating sows. Crystalline valine was added to practical corn-soybean meal diets. Sows were assigned on d
112 of pregnancy, within parity, to 1 of 3 diets, which
were fed throughout a 19-d lactation period. Sows were
housed in conventional farrowing crates in an environmentally regulated research facility. Temperature was
maintained at 21°C (21.6 ± 1.5°C), and lights were on
from 0600 to 1500. Within 24 h of farrowing, all litters
were standardized to a minimum of 10 pigs each.
Dietary treatments were formulated to achieve total
dietary V:L ratios of 0.90, 1.05, or 1.25:1. The lysine
concentration was slightly below the estimate of lysine
requirement (0.90% total lysine) determined using a
regression equation that was developed by regressing
total lysine intake of lactating sows on litter growth
rate (Pettigrew, 1993). Diets were based on corn and
soybean meal and contained 12% wheat middlings and
0.068% crystalline L-lysine HCl (Table 1). Dietary valine concentration was increased by substituting Lvaline for an equivalent amount of corn. Calculated
ratios of other essential AA to lysine exceeded NRC
recommendations (1998) for the level of litter growth
that was anticipated and subsequently achieved. The
respective ratios of threonine, methionine + cysteine,
and tryptophan to lysine were 0.68, 0.61, and 0.22:1, respectively.
Feed was provided to each sow weekly in closed plastic containers. Each sow was fed ad libitum from their
respective container, which held a known amount of
feed. Apparent feed disappearance was computed
weekly by subtracting residual from initial container
Table 1. Diet composition of lactation diets varying in
valine:lysine ratio, Exp. 1 (as-fed basis)
Total dietary valine:lysine ratio
Ingredient, %
Corn
Soybean meal, 47.5% CP
Wheat middlings
Choice white grease
Dicalcium phosphate
Limestone
Salt
Vitamin trace mineral premix1,2
L-Lysine HCl
L-Valine
Calculated composition
ME,3 Mcal/kg
CP, %
True ileal digestible lysine,4 %
Total lysine, %
Total valine, %
Valine:lysine
Threonine:lysine
Isoleucine:lysine
Methionine + cysteine:lysine
Calcium, %
Available phosphorus, %
Chemical composition
CP, %
Lysine, %
Methionine + cysteine, %
Valine, %
Tryptophan, %
Valine:lysine ratio
Threonine, %
Isoleucine, %
0.90
1.05
1.20
60.88
19.81
12.00
3.12
2.15
0.75
0.46
0.75
0.068
0.000
60.76
19.82
12.00
3.09
2.15
0.75
0.46
0.75
0.068
0.137
60.65
19.84
12.00
3.05
2.16
0.75
0.46
0.75
0.068
0.274
3.32
16.3
0.80
0.90
0.81
0.90
0.68
0.73
0.60
0.90
0.50
3.32
16.3
0.80
0.90
0.95
1.05
0.68
0.73
0.61
0.90
0.50
3.32
16.3
0.80
0.90
1.08
1.20
0.68
0.73
0.61
0.90
0.50
15.9
0.86
0.51
0.76
0.18
0.88
0.57
0.62
16.2
0.88
0.52
0.90
0.19
1.02
0.56
0.62
16.2
0.89
0.52
1.06
0.18
1.19
0.57
0.64
1
Premix supplied per kilogram of diet: Vitamin A, 9,900 IU; vitamin
D3, 1,760 IU; vitamin E, 66 IU; vitamin K (menadione activity), 4.4
mg; riboflavin, 9.9 mg; D-pantothenic acid, 38 mg; niacin, 44 mg;
vitamin B12, 37.4 ␮g; D-biotin, 0.22 mg; folic acid, 1.32 mg; choline,
0.66 g; thiamine, 2.2 mg; pyridoxine, 3.3 mg; Zn, 125 mg (ZnSO4);
Cu, 15 mg (CuSO4); Fe, 100 mg (FeSO4); Mn, 50 mg (MnSO4); I, 0.35
mg (EDDI); and Se, 0.30 mg (Na2Se).
2
Calcium and rice hull carrier of vitamin trace mineral premix
contained 5.20% calcium.
3
Computed from NRC (1998).
4
Computed using the NRC (1998) coefficient for true ileal digestible
lysine.
weight and dividing this value by 7. This estimate was
subsequently refined by correcting for inevitable feed
loss during the eating process (through the floor). Feed
loss was estimated from 15 sows and averaged 5.0
(±1.9)%. Feed intake was estimated by applying the
correction factor to all sows in both experiments (i.e.,
feed intake was computed by multiplying daily feed
disappearance by 0.95) and was the basis for the values
reported in this paper.
Experiment 2
Two hundred seventy-nine PIC, Camborough product
sows (parity 1 to 5) were used to verify the adequacy
of the NRC minimum recommended V:L ratio (0.86 to
1.25:1), using practical corn-soybean meal diets con-
Downloaded from jas.fass.org by on November 12, 2008.
Valine requirement of lactating sows
Table 2. Diet composition of lactation diets varying in
valine:lysine ratio and level of soybean meal (16.73%,
diets 1 and 3 vs. 22.68%, diets 2 and 4), Exp. 2 (as-fed basis)
Diet:
Dietary valine:lysine ratio:
Ingredient, %
Corn
Soybean meal (47.5% CP)
Choice white grease
Dicalcium phosphate
Ground limestone
Salt
Vitamin trace mineral premix1
L-Lysine HCl
L-Threonine
DL-Methionine
L-Tryptophan
L-Isoleucine
L-Valine
Calculated composition
ME,2 Mcal/kg
CP, %
True ileal digestible lysine,3 %
Total lysine, %
Total valine, %
Valine:lysine
Threonine:lysine
Isoleucine:lysine
Methionine + cysteine:lysine
Calcium, %
Available phosphorus, %
Chemical composition
Crude protein, %
Lysine, %
Threonine, %
Isoleucine, %
Valine, %
Tryptophan, %
Valine:lysine ratio
Methionine + cysteine, %
1
0.73
2
0.86
3
1.25
4
1.25
73.71
16.73
5.00
2.71
0.59
0.50
0.15
0.275
0.14
0.07
0.04
0.09
0.00
68.47
22.68
4.98
2.63
0.54
0.50
0.15
0.05
0.00
0.00
0.00
0.00
0.00
73.23
16.73
5.00
2.71
0.59
0.50
0.15
0.275
0.14
0.07
0.04
0.09
0.48
68.11
22.68
4.98
2.63
0.54
0.50
0.15
0.05
0.00
0.00
0.00
0.00
0.36
3.43
14.2
0.82
0.91
0.66
0.73
0.72
0.70
0.60
0.88
0.52
3.43
16.2
0.81
0.91
0.78
0.86
0.68
0.72
0.60
0.85
0.51
3.43
14.2
0.81
0.91
1.14
1.25
0.72
0.70
0.60
0.89
0.53
3.43
16.2
0.81
0.91
1.13
1.25
0.68
0.72
0.60
0.87
0.52
13.9
0.88
0.64
0.63
0.65
0.19
0.74
0.54
15.9
0.89
0.60
0.65
0.75
0.18
0.85
0.53
14.0
0.87
0.64
0.62
1.06
0.18
1.22
0.55
15.8
0.89
0.60
0.65
1.08
0.18
1.21
0.54
1
Premix supplied per kilogram of diet: vitamin A, 11,000 IU; vitamin D3, 1,760 IU; vitamin E, 77 IU; vitamin K (menadione activity),
4.0 mg; riboflavin, 11.0 mg; D-pantothenic acid, 33 mg; niacin, 55 mg;
vitamin B12, 44.0 ␮g; D-biotin, 0.28 mg; folic acid, 1.21 mg; choline,
0.66 g; thiamine, 3.0 mg; pyridoxine, 3.0 mg; Zn, 125 mg (ZnSO4);
Cu, 15 mg (CuSO4); Fe, 100 mg (FeSO4); Mn, 35 mg (MnSO4); I, 0.35
mg (EDDI); and Se, 0.30 mg (Na2Se).
2
Computed from NRC (1998).
3
Computed using the NRC (1998) coefficient for true ileal digestible
lysine.
taining 0.05% L-lysine HCl. We also determined if we
could duplicate the litter growth response to an increase
in the analyzed V:L ratio (i.e., 0.73 increased to 1.25)
that was reported previously by Richert et al. (1996),
whose diets involved extensive crystalline AA addition
to the initial diet (0.73 V:L).
Four dietary treatments were used with 2 diets (diets
1 and 3) composed as described by Richert et al. (1996)
with a minor exception that valine replaced corn rather
than cornstarch. The diets were formulated to contain
0.91% total lysine and 0.275% crystalline L-lysine HCl
(Table 2). The level of soybean meal was held constant
(16.7%) for both diets and crystalline L-valine was
added to increase the dietary V:L ratio from 0.73 to
1417
1.25:1, respectively. The dietary valine concentration
was increased by substituting crystalline L-valine for
an equivalent amount of corn. Crystalline L-threonine,
DL-methionine, L-tryptophan, and L-isoleucine were
added to achieve ratios to lysine of 0.72, 0.60, 0.22,
and 0.70:1, respectively. Two diets (diets 2 and 4) were
formulated similar to those of Exp. 1 and based on corn
and soybean meal without extensive crystalline AA use
(Table 2). The level of soybean meal was greater than
for diets 1 and 3, and was held constant (22.7%), thus
achieving dietary V:L ratios of 0.86 and 1.25:1 in diets
2 and 4, respectively. Dietary valine concentration was
increased by replacing corn with L-valine.
Sows were assigned on d 112 of pregnancy, within
parity, to 1 of the 4 diets, which were fed throughout a
19-d lactation period. Sows were housed in conventional
farrowing crates in an environmentally regulated farrowing facility with the light pattern and litter standardization as described in Exp. 1. In contrast to Exp.
1, diets were fed ad libitum beginning on d 2 and continuing through d 12 of lactation, after which a maximum
allowance of 7.27 kg/sowⴢd−1 was set. This was done
to prevent the daily lysine intake from exceeding the
requirement based on the expected litter growth rate.
The minimum ideal ratio of valine could only be estimated where lysine intake was in the linear portion of
the response curve. Provision of an excess of the reference AA would cause the ideal valine ratio to be inaccurate and low.
Sow and Litter Criteria
To calculate sow weight change during lactation,
sows were weighed on d 1 of lactation and at the time
of weaning in both experiments. In Exp. 1, to calculate
change in loin and backfat depth (wean value − d 1
value), loin depth and backfat depth at the 10th and
last rib were measured by the same technician using
real-time ultrasound (Aloka 500; Aloka, Ithaca, NY).
Change in loin and fat depth were used as an indication
of body protein and lipid mobilization, respectively, in
support of lactation.
To calculate litter growth rate, piglets were weighed
on d 1 of lactation and at weaning. All pigs were processed according to routine management practices,
which included tail docking, subcutaneous iron dextran
injection, and castration. The sow was the sole source
of nutrients provided to the litter.
Diet Analysis
Crude protein was determined by Kjeldahl analysis
(N × 6.25), with tryptophan used as the internal standard for nitrogen recovery (AOAC, 1996). The AA content of the diets was determined using gas liquid chromatography (Beckman Model 6300; Beckman Instruments, Inc., San Ramon, CA) and by AOAC procedure
994.12, alternative 1 (AOAC, 1996). Performic acid was
used before hydrolysis to oxidize cysteine and methio-
Downloaded from jas.fass.org by on November 12, 2008.
1418
Gaines et al.
nine to cysteic acid and methionine sulfone, respectively. Amino acids were liberated from protein by hydrolysis with 6 N HCl at 110°C for 24 h. Samples were
cooled to room temperature, titrated with 50% NaOH
to a pH of 2.20, and diluted to the desired volume with
sodium citrate buffer. Norleucine was added to the hydrolysate to achieve 0.1 mmol/L in the final sample.
Samples were filtered through a 0.2-␮m syringe filter
before injection. Amino acids were separated using a
3-buffer gradient-elution system on an ion-exchange
column (4.6 mm i.d. × 120 mm) packed with sulfonated
polystyrene divinylbenzene. Detection was by postcolumn derivatization with ninhydrin, using ultraviolet
spectroscopy at 440 and 570 nm. Tryptophan was determined using gas liquid chromatography after alkaline
hydrolysis using 60% barium hydroxide under anaerobic environment for 16 h. Conditions and use of the
internal standard were as described.
Statistical Analysis
Data were analyzed by ANOVA using GLM procedures of SAS (SAS Inst., Inc., Cary, NC), and LS mean
comparisons were made using Fisher’s protected least
significant difference test (P < 0.05). The statistical
model included parity, diet, and parity × diet interactions. The initial analysis indicated that there were no
parity × diet interactions (P > 0.10) in either experiment; thus, the interaction term was dropped from the
models. Sow data involving BW and carcass measurements contained their respective initial values as covariates in the model. Litter growth data were adjusted
for the number of pigs nursed and lactation length by
covariate analysis. In Exp. 1, polynomial coefficients
were used to determine linear and quadratic effects of
increasing dietary valine. In Exp. 2, preplanned nonorthogonal contrasts were used to test the effects of increasing dietary V:L ratio (diet 1 vs. 3; diet 2 vs. 4)
within diet type. The effect of diet type was also compared at the 1.25:1 V:L ratio (diet 3 vs. 4). Sow and litter
were used as the experimental unit for respective data.
RESULTS
Experiment 1
Increasing the dietary V:L ratio beyond 0.90:1 did
not affect the number of pigs weaned, survival rate,
litter weight gain, or piglet growth rate even though
sows were nursing >10 pigs per litter for 19 d (Table
3). Sow feed intake, corrected for feeding waste, was
not affected by dietary valine concentration and averaged 7.22 kg/d (Table 4). Lysine intake was similar
across treatments and averaged 65.0 g/d. Daily valine
intake increased (linear, P < 0.001) from 58.5 to 78.1
g/d as dietary valine concentration increased. Despite
an increase in relative valine intake, there were no
differences in the amount of sow BW or loin depth loss
(Table 4). Fat depth declined at both the 10th and last
rib but was similar among treatments. The concomitant
loss of loin and fat depth would suggest that sow BW
loss consisted of both body protein and lipid.
Experiment 2
Total litter gain tended to improve (P = 0.14) when
the 0.275% L-lysine HCl control was supplemented with
valine to achieve a 1.25:1 V:L ratio, diets 1 vs. 3, respectively (Table 5). It is noteworthy that the litter growth
advantage for the 1.25 V:L group was achieved despite
the latter having a lower estimated feed intake (P <
0.05). It is not clear whether the 0.73 V:L control group
exhibited an increase in intake due to deficient AA intake or whether the 1.25 V:L test group simply consumed less. The former is sometimes observed in growing pigs but under more extreme protein or AA deficiencies. To the contrary, no aspect of sow or litter
response was improved when the corn-soybean meal
control diet (0.05% L-lysine HCl) was supplemented
with valine to achieve a 1.25:1 V:L ratio, diets 2 vs. 4,
respectively. The litter gains achieved by all treatments, except the 0.73 V:L control, were remarkably
similar and averaged approximately 44.8 kg (ca. 257 g
of gainⴢpig−1ⴢd−1).
The 0.73 V:L control group also had an increase lysine
intake when compared with the 1.25 V:L counterpart
(P < 0.03). Otherwise, estimated daily lysine intake was
similar among treatments (diets 2, 3, 4) and ranged
from 52.1 to 53.4 g/d. Valine intake increased predictably as dietary valine concentration increased (diets 1
vs. 3, P < 0.001; diet 2 vs. 4, P < 0.001) and ranged from
40.1 to 66.1 g/d. There were no differences in either
feed or valine intake with diet form (diet 3 vs. 4), and
neither of the 1.25 V:L ratio diets improved litter gain,
piglet growth rate, or maternal BW loss compared with
the corn-soybean meal control diet that was formulated
to the NRC (1998) recommendation.
DISCUSSION
This research shows that a V:L ratio in excess of 0.86
does not conserve maternal tissue loss or improve piglet
growth rate for the prolific, high milk production sow.
This suggests that the NRC (1998) estimate is not too
low, and it agrees with subsequent studies by Carter
et al. (2000) and Southern et al. (2000). These results
are in contrast to earlier reports suggesting that the
optimum V:L ratio was in excess of 1.0 (Richert et al.,
1996, 1997a,b; Moser et al., 2000). We do agree that
piglet growth rate seems to be compromised by a dietary
V:L ratio of 0.73. This apparent response is consistent
with but less remarkable than previously reported by
Richert et al. (1996). In total, there is now sufficient
information in the literature to conclude that the dietary V:L ratio requirement does not exceed the NRC
estimate (1998). It is not clear whether the NRC estimate can be further refined downward and whether the
V:L relationship is dynamic over a wide range of body
protein mobilization (Kim et al., 2001).
Downloaded from jas.fass.org by on November 12, 2008.
1419
Valine requirement of lactating sows
Table 3. Effect of dietary valine:lysine ratio on litter growth (Exp. 1)1
Valine:lysine
Probability
Item
0.90
1.05
1.25
Diet
Linear
Quadratic
SEM
No. of sows
Lactation length, d
Avg. parity
Litter2
No. of pigs placed/sow
Initial litter weight, kg
Pigs weaned/sow
Piglet survival, %
Litter wean wt, kg
Litter gain,3 kg
Piglet daily gain, g/d
61
19.2
2.5
62
19.5
2.5
66
19.5
2.6
—
0.60
0.94
—
0.40
0.79
—
0.56
0.84
—
0.25
0.14
10.9
17.8
10.2
94.3
60.0
42.2
222
10.7
17.9
10.0
93.2
58.1
40.2
218
10.9
17.8
10.2
94.7
59.8
42.0
219
0.68
0.96
0.58
0.65
0.62
0.59
0.87
0.97
0.98
0.90
0.79
0.95
0.90
0.75
0.38
0.79
0.31
0.37
0.33
0.31
0.67
0.15
0.28
0.13
0.01
1.52
1.50
6.05
1
Data represent least squares means for each treatment. Actual valine:lysine ratio was determined through
chemical analysis to be 0.88, 1.02, and 1.19 for the 0.90, 1.05, and 1.25 groups, respectively.
2
Lactation length and number of pigs nursed were used as covariates in the analysis.
3
Initial measurement used as a covariate.
Lysine intake (g/d) was greater than anticipated in
Exp. 1. A possible criticism of this experiment is that
lysine intake was in excess of the requirement, which
would result in an inaccurate estimate of the minimum
V:L need. A fundamental tenet to establishing an ideal
pattern among AA is that the reference AA, lysine, must
be marginally deficient and in the linear portion of the
response curve. This concern was addressed in Exp. 2
by setting an upper limit on feed intake of 7.27 kg/d,
beginning on d 12 of lactation. It is noteworthy that
sows in both experiments lost BW. The fact that both
fat and loin eye depth was lost during lactation (Exp.
1) suggests that tissue mobilization involved a net loss
from lipid and protein pools; the latter implies an AA
deficit.
We concluded that the reference AA, lysine, was marginally deficient and is in the linear portion of the response curve for Exp. 1 and 2, using published estimates of lysine need. We calculated the lysine requirement using an updated equation (Boyd et al., 2000)
that relates litter growth rate to lysine intake of the
lactating sow in the following manner:
Total dietary lysine, g/d = 0.0266
(litter growth rate, kg/d) − 7.55 (R2 = 0.63).
This relationship was originally proposed by Pettigrew (1993). Our estimate was computed using litter
growth rate (indirect function of milk output) and maintenance (obligatory loss). However, we included mater-
Table 4. Effect of the dietary valine:lysine ratio on nutrient intake and sow performance
(Exp. 1)1
Probability
Item
Dietary valine:lysine
Nutrient intake
Feed intake,3 kg/d
Lysine intake,3 g/d
Valine intake,3 g/d
Sow
Initial BW, kg
Weight loss,4 kg
Initial fat depth 10th rib, mm
Fat loss 10th rib,4 mm
Initial fat depth last rib, mm
Fat loss last rib,4 mm
Initial loin depth 10th rib, mm
Loin depth loss 10th rib,4 mm
Initial loin depth last rib, mm
Loin depth loss last rib,4 mm
2
Diet
Linear
Quadratic
SEM
0.90
7.22
65.0
58.5
1.05
7.22
65.0
68.6
1.25
7.23
65.1
78.1
0.99
0.99
0.001
0.89
0.89
0.001
0.95
0.95
0.68
0.06
0.59
0.62
215.8
−10.2
17.9
−2.5
19.4
−1.93
48.2
−2.7
47.2
−1.1
218.2
−9.2
17.1
−1.7
18.7
−1.59
47.6
−1.7
47.4
−2.1
218.5
−9.4
17.5
−1.8
18.2
−1.73
47.5
−2.3
47.9
−2.1
0.71
0.86
0.56
0.17
0.37
0.74
0.71
0.42
0.80
0.30
0.45
0.67
0.62
0.16
0.16
0.64
0.45
0.58
0.52
0.19
0.73
0.74
0.33
0.20
0.95
0.52
0.74
0.23
0.87
0.39
2.52
1.41
0.53
0.31
0.57
0.31
0.69
0.53
0.73
0.52
1
Data represent least squares means for each treatment. Actual valine:lysine ratio was determined through
chemical analysis to be 0.88, 1.02, and 1.19 for the 0.90, 1.05, and 1.25 groups, respectively.
2
Lactation length and number of pigs nursed were used as covariates in the analysis.
3
Calculation based on an estimate of 5% for feed wastage.
4
Initial measurement used as a covariate.
Downloaded from jas.fass.org by on November 12, 2008.
1420
Gaines et al.
Table 5. Effect of dietary valine:lysine ratio on sow and litter performance (Exp. 2)1
Diet
Item
L-LysineⴢHCl,
%
Dietary valine:lysine ratio
No. of sows
Lactation length, d
Average parity
Litter
No. of pigs placed/sow
Initial litter weight, kg
Pigs weaned/sow2
Piglet survival,2 %
Litter wean wt,2 kg
Litter gain,2,3 kg
Piglet daily gain,2 g
Sow
Initial weight, kg
Lactation weight loss,3 kg
Feed intake,2,4 kg/d
Lysine intake,2,4,5 g/d
Valine intake,2,4,5 g/d
Probability
1
2
3
4
Diet
1 vs. 3
2 vs. 4
3 vs. 4
SEM
0.27
0.73
70
19.4
2.34
0.05
0.86
73
19.5
2.35
0.27
1.25
69
19.7
2.25
0.05
1.25
67
19.5
2.41
—
0.89
0.91
—
0.47
0.68
—
0.99
0.78
—
0.56
0.47
—
0.27
0.16
10.5
19.7
9.9
93.5
62.5
42.7
250.0
10.6
19.5
9.9
93.3
63.7
44.5
258.8
10.7
19.3
10.0
94.5
64.7
45.1
257.1
10.8
19.8
10.0
94.8
64.5
44.9
256.2
0.72
0.83
0.72
0.70
0.53
0.45
0.72
0.29
0.51
0.48
0.52
0.18
0.14
0.38
0.18
0.59
0.36
0.32
0.63
0.81
0.75
0.34
0.38
0.96
0.83
0.91
0.90
0.91
0.12
0.47
0.12
0.02
1.24
1.21
6.09
234.6
−6.7
6.08
55.3
40.1
232.5
−9.0
5.87
53.4
45.8
230.5
−10.5
5.72
52.1
65.2
232.4
−10.3
5.85
53.2
66.1
0.78
0.36
0.20
0.15
0.001
0.31
0.11
0.03
0.03
0.001
0.89
0.57
0.90
0.90
0.001
0.64
0.93
0.45
0.70
0.57
3.04
1.81
0.12
1.12
1.25
1
Data represent least squares means for each treatment. Lactation diets varied in dietary valine:lysine
ratio (0.73, 0.86, 1.25, and 1.25) and soybean meal level. Refer to Table 2.
2
Lactation length and number of pigs nursed were used as covariates in the analysis.
3
Initial BW used as a covariate.
4
Calculation based on a 5% estimate for feed wastage.
5
Calculation based on calculated AA values.
nal lysine mobilization (described by the negative intercept) in the calculated estimate. We did not consider
maternal lysine mobilization to be obligatory unless
dietary intake fails to adequately meet the lysine requirement. This outcome is both more important and
more difficult to achieve for the first litter sow because
conservation of body protein mass during the first lactation is important to second litter size (Boyd et al., 2000).
In Exp. 1, the lysine required during lactation was estimated to be 66.9 g/d. This assumes 2.23 kg of litter
growth/d (computed from Table 3) and that an additional 7.55 g of lysine/d was required to prevent body
lysine mobilization for maintenance (intercept). This
compares to an estimated lysine intake of 65.0 g/d (Table 4). In Exp. 2 the lysine requirement was computed to
be 75.9 g/d (2.57 kg litter gain/d), which is considerably
greater than actual lysine intake (<55 g/d).
The fact that there was no response by the addition
of valine to diets having a marginal deficiency of lysine
and other AA led to the conclusion that the valine content was adequate in the corn-soybean control diets
that were used in Exp. 1 and 2. This line of reasoning
is most convincing for Exp. 2 because average valine
intake was less than 50 g/d (Table 5). The discrepancy
between this and early studies on the requirement for
valine cannot be attributed to extreme differences in
either the number of pigs nursed or litter growth rate,
because both are similar across studies. The level of
valine provided by the corn-soybean control diet in Exp.
2 is relatively similar to that provided by the control
diets of Richert et al. (1996, 1997a,b) and Moser et al.
(2000), in the range of 40 to 48 g/d. The level achieved
in the dose response assay (Exp. 1; Table 3) was also
similar (60 to 70 g/d). The possibility that diet composition was involved was also considered. With the exception of the report by Moser et al. (2000), supplemental
valine improved litter growth rate in trials involving
high (0.24 to 0.38%) levels of L-lysine HCl (Richert et
al., 1996, 1997a,b). Two of the diets were virtually duplicated in Exp. 2 (0.27% L-lysine HCl). Although valine
addition (1.25 V:L) improved litter growth rate when
compared with the 0.73 V:L control, it performed equivalent to the corn-soybean control (0.86 V:L) and test
(1.25 V:L) diets, containing 0.05% L-lysine HCl.
The concept of a dynamic ideal AA pattern was extended to the lactating sow by the elegant work of Kim
et al. (2001). The work of Kim et al. (2001) is relevant
to the current study and to future AA research. They
proposed that the order of limiting dietary AA may
differ with increasing body protein mobilization (or
weight loss). This is best illustrated with threonine,
where the threonine:lysine ratio was estimated to increase in increments from 0.59 to 0.75 as the sow experienced progressively greater BW loss and protein mobilization. Substantial increases in BW loss occur with extreme heat stress, for example. It is noteworthy that
the V:L ratio was estimated to increase from 0.77 to only
0.78 over this same range of weight loss. The weight loss
that we observed for sows in Exp. 1 and 2 was moderate
(9.7 and 9.1 kg respectively) and, based on their research, would require a minimum V:L ratio of 0.78.
Further research is required to determine whether the
Downloaded from jas.fass.org by on November 12, 2008.
Valine requirement of lactating sows
V:L requirement is closer to the pattern in milk (0.78)
or the NRC estimate (0.85). It seems that the ARC
(1981) estimate of 0.70 for the V:L ratio is deficient,
based on the report of Richert et al. (1996) and Exp. 2.
IMPLICATIONS
This research demonstrates that the dietary valine
requirement is not greater than the NRC (1998) proposed minimum (0.855 total or true ileal digestible valine:lysine) for the sow that is nursing a large litter
(greater than 10 pigs). This means that crystalline lysine can be added to grain, soybean meal diets for lactating sows containing approximately 0.90% total dietary
lysine without compromise of maternal body tissue or
litter growth. Further refinement of the minimum valine:lysine ratio (between 0.70 and 0.85) is required before we can predict the second limiting amino acid and
to take advantage of greater crystalline amino acid use.
LITERATURE CITED
AOAC. 1996. Official Methods of Analysis. 16th ed. Assoc. Off. Anal.
Chem., Arlington, VA.
ARC. 1981. Page 164 in The Nutrient Requirements of Pigs: Agricultural Technical Review. Rev. ed. Commonw. Agric. Bureau,
Slough, England.
Boyd, R. D., R. E. Kensinger, R. Harrell, and D. E. Bauman. 1995.
Nutrient uptake and endocrine regulation of milk synthesis by
mammary tissue of lactating sows. Page 35 in Second Int. Lactation Biol. Symp. H. A. Tucker, and J. A. Pettigrew, ed. J. Anim.
Sci. 73(Suppl. 3).
1421
Boyd, R. D., K. J. Touchette, G. C. Castro, M. E. Johnston, K. U. Lee,
and I. K. Han. 2000. Recent advances in amino acid and energy
of prolific sows: Review. Asian Aus. J. Anim. Sci. 13:1638–1652.
Carter, S. D., G. M. Hill, D. C. Mahan, J. L. Nelssen, B. T. Richert,
and G. C. Shurson. 2000. Effects of dietary valine concentration
on lactational performance of sows nursing large litters. J. Anim.
Sci. 78:2879–2884.
Kim, S. W., D. H. Baker, and R. A. Easter. 2001. Dynamic ideal
protein and limiting amino acids for lactating sows: The impact
of amino acid mobilization. J. Anim. Sci. 79:2356–2366.
Moser, S. A., M. D. Tokach, S. S. Dritz, R. D. Goodband, J. L. Nelssen,
and J. A. Loughmiller. 2000. The effects of branched-chain amino
acids on sow and litter performance. J. Anim. Sci. 78:658–667.
NRC. 1988. Page 52 in Nutrient Requirements of Swine. 9th ed. Natl.
Acad. Press, Washington, DC.
NRC. 1998. Page 40 in Nutrient Requirements of Swine. 10th ed.
Natl. Acad. Press, Washington, DC.
Pettigrew, J. E. 1993. Amino acid nutrition of gestating and lactating
sows. BioKyowa Technical Review No. 5. Nutri-Quest, Inc.,
Chesterfield, MO.
Richert, B. T., R. D. Goodband, M. D. Tokach, and J. L. Nelssen.
1997a. Increasing valine, isoleucine, and total branched-chain
amino acids for lactating sows. J. Anim. Sci. 75:2117–2128.
Richert, B. T., M. D. Tokach, R. D. Goodband, J. L. Nelssen, R. G.
Campbell, and S. Kershaw. 1997b. The effect of lysine and valine
fed during lactation on sow and litter performance. J. Anim. Sci.
75:1853–1860.
Richert, B. T., M. D. Tokach, R. D. Goodband, J. L. Nelssen, J. E.
Pettigrew, R. D. Walker, and L. J. Johnston. 1996. Valine requirement of the high-producing lactating sow. J. Anim. Sci.
74:1307–1313.
Southern, L. L., F. M. LeMieux, J. O. Matthews, T. D. Bidner, and
T. A. Knowles. 2000. Effect of feather meal as a source of valine
for lactating sows. J. Anim. Sci. 78:120–123.
Trottier, N. L., C. F. Shipley, and R. A. Easter. 1997. Plasma amino
acid uptake by the mammary gland of the lactating sow. J. Anim.
Sci. 75:1266–1278.
Downloaded from jas.fass.org by on November 12, 2008.
References
This article cites 9 articles, 8 of which you can access for free at:
http://jas.fass.org/cgi/content/full/84/6/1415#BIBL
Downloaded from jas.fass.org by on November 12, 2008.