Methionine Requirement of Channel Catfish Fed
Soybean Meal-Corn-Based Diets1
Yongjiu Cai2 and Gary J. Burtle
Animal and Dairy Science Department, The University of Georgia,
Coastal Plain Experiment Station, Tifton 31793-0748
ABSTRACT: A soybean meal-corn-based diet was
used to determine dietary methionine (Met) required
by 14-g channel catfish ( Ictalurus punctatus) in a
42-d experiment at 25°C. The basal diet with balanced
limiting amino acids relative to the catfish whole-body
amino acid profile contained 277 g of CP, 3.6 g of Met,
4.0 g of cystine (Cys), and 10 MJ of DE/kg of DM. DL-
Methionine was added to the basal diet from 0 to 12.0
g/kg in 2-g intervals at the expense of L-glutamic acid
to produce seven isonitrogenous and isocaloric diets. A
reference diet contained 331 g of CP, 8 g of Met, 5 g of
Cys, and 10 MJ of DE/kg of DM (included 8% fish
meal). Seven graded Met levels resulted in quadratic
responses ( P < .01) of weight gain, specific growth
rate, feed or GE intake, feed or energy efficiency,
protein or energy retention, protein efficiency ratio,
Key Words: Aquaculture, Catfish, Growth, Methionine, Sulfur Amino Acids
Introduction
The methionine requirement of channel catfish
( Ictalurus punctatus) recommended by NRC (1983,
1993) was based on one published experiment (Hard-
ing et al., 1977) in which purified diets contained a
calculated amino acid composition similar to whole-
chicken egg protein in the absence of cystine. As a
usual practice in fish amino acid nutrition, the
determined methionine (or an amino acid) require-
ment has been extrapolated to fish diets with various
CP levels, assuming those essential amino acid
requirements are in proportion to levels of protein
1The authors thank David H. Baker at University of Illinois for
reviewing a draft of the manuscript and his helpful suggestions for
discussion of the sulfur amino acid requirement, Harry K. Dupree at
the National Biological Service, and Philip R. Utley at The
University of Georgia for critical reviews of the manuscript.
2Present address: Dept. of Fisheries and Wildlife, Univ. of
Minnesota, 200 Hodson Hall, 1980 Folwell Ave, St. Paul
55108-6124.
Received June 12, 1995.
Accepted November 22, 1995.
514
and apparent net protein or energy utilization. Chan-
nel catfish required 9.4 g of Met/kg of DM (34.1 g/kg
of CP) with a total 11.3 g/kg of calculated digestible
sulfur-containing amino acids based on multiple
regression dose-response models or 270 mg of Met/kg
of fish per day based on a broken-line response of
protein gain to Met intake. At the adequate Met level,
catfish with the lowest ( P < .05) liver lipids showed
feed intake and protein or energy utilization efficiency
similar ( P > .05) to that of catfish fed the reference
diet. Catfish fed all-plant-protein diets require more
dietary methionine than previously reported. Catfish
fed corn-soybean meal diets fortified adequately with
methionine result in performance that approaches
that of fish fed a fish meal-based diet.
J. Anim. Sci. 1996. 74:514–521
intake. However, recent studies show the relationship
not to be linear but to be exponential (Finke et al.,
1987) and the fish body amino acid profile to be a
better reference than chicken-egg protein (Ketola,
1982; Wilson, 1989).
In general, a single methionine requirement level is
recommended based on weight gain of catfish when
fed diets containing 8 to 10% fish meal with 32 to 36%
CP. There is an increasing need for least-cost-
formulated diets with minimal fish meal and maximal
plant proteins. Studies have shown that diets includ-
ing large amounts of plant proteins contain antinutri-
tional compounds (Grant, 1989; NRC, 1993), which
may limit bioavailability of amino acids. Although
crystalline DL-methionine (Met) was found effective
for use by channel catfish (Robinson et al., 1978), it is
not commonly used in commercial catfish diets.
Methionine alone can generally meet the require-
ment for sulfur-containing amino acid ( SAA) (Baker,
1987) and all-plant-protein diets are most limiting for
SAA. The objective of this study was to determine the
dietary methionine requirement in fingerling channel
catfish fed a 27% CP, soybean meal-corn diet using
multiple response criteria. The requirement level was
 METHIONINE REQUIREMENT OF CHANNEL CATFISH 515
expressed as a percentage of dietary protein on a DM Table 1. Formulation and composition of basal
basis and as milligrams/kilogram of fish BW gain per (Diets 1) and reference (Diet 8) dietsa
day.
Diet
Item 1 - Basal 8 - Reference
Materials and Methods
g/kg
Fish Ingredient (as-fed basis)
Soybean meal (48% CP) 422.9 437.7
Thirty channel catfish, Tifton strain at the Univer- Menhaden fish meal — 80.0
Ground corn 297.0 292.7
sity of Georgia, were acclimated in each of 24 tanks
Wheat middlings — 100.0
(75-L volume) for 2 wk before the experiment. The Dextrin 125.0 —
fish were fed Diet 1 (Table 1 ) twice daily, 2% BW for Menhaden fish oil 48.0 35.0
each meal at approximately 0830 and 1600. After Calcium phosphate 33.3 23.6
acclimation, 20 fish in each tank, excluding the CaCO3 3.0 —
DL-Methionine 0.0 2.0
extreme sizes, were randomly selected for the feeding
L-Glutamic acid 12.0 —
experiment. The initial body weight was 14.36 ± .14 g L-Amino acid mixtureb 29.8 —
(mean ± SD). Five fish were taken as initial samples Vitamin premixc 3.0 3.0
that were stored at −10°C until carcass analysis. Mineral premixd 5.0 5.0
During the 6-wk experiment, all fish were counted Ascorbic acid 1.0 1.0
Binder (lignocellulose) 20.0 20.0
and weighed at the end of every other week. Before
Composition (DM basis)
counting, the fish were anesthetized in water contain-
DM 950.5 971.1
ing 100 mg/L of tricaine methanesulfonate (FIN- CP ( N × 6.25) 276.0 331.1
QUEL, Argent Chemical Laboratories, Redmond, GE, MJ/kg 17.38 18.45
WA). After weighing in water, the fish were placed in DE, MJ/kge 10.08 10.03
a 5 g/L of NaCl solution for approximately 4 min to Ether extracte 66.4 63.9
Crude fibere 41.4 31.4
revive before being returned to the tank. At the end of
Available Pe 5.5 5.7
wk 6, five fish from each tank were taken for analysis Cystine 4.0 5.0
of body composition and an additional five fish were Methionine 3.6 8.0
sampled for liver tissue. The methods of this study Lysine 19.5 20.0
were approved by the University of Georgia Office of aDiets 2 to 7 were formulated by adding 2.0 to 12.0 g of DL-
Animal Care and Use. methionine/kg at 2-g intervals to Diet 1. DL-methionine was added
isonitrogenously and L-glutamic acid and dextrin levels were ad-
justed to ensure energy similarity.
Diets bComposition as gram/kilogram of mix (as-fed): arginine, 40.3;
isoleucine, 33.6; leucine, 10.1; lysine, 174.5; threonine, 77.2; valine,
Diets 1 to 7 (Tables 1 and 2 ) were formulated to 83.9; and cellulose as the carrier, 580.4.
cContributed in milligrams per kilogram of dry diet: choline
meet the requirements for amino acids (NRC, 1983; chloride, 487; inositol, 32; niacin, 29; Ca-d-pantothenate, 43;
Wilson, 1991) rather than for CP, as is more common pyridoxine HCl, 6; riboflavin, 13; thiamine HCl, 6; folic acid, .58;
in fish studies. A methionine-deficient basal diet was biotin, .13; vitamin B12, .007; retinyl acetate, 5; cholecalciferol, .54;
DL-a-tocopherol acetate, 130; menadione, 11; and cellulose as the
formulated by using soybean meal and corn as sources carrier, 2481.
of protein (Table 1). L-glutamic acid was included on dContributed in milligram per kilogram of dry diet: KH PO ,
2 4
a nitrogen basis to replace the graded levels of 1297; Ca(H 2 PO4) 2·H2O, 732; MgSO4·7H2O, 714; NaH2PO4·H2O,
472; NaCl, 236; FeC6H5O7·H2O, 160; ZnSO4·7H2O, 28; CoCl2·6H2O,
supplemented DL-Met. Menhaden fish oil (U.S. Bi- 5; AlCl3·6H2O, 81; KI, 81; CuCl, 1; MnSO4·H2O, 4; selenium premix
ochemical, Cleveland, OH) was used as an energy and (.2% Se), 60; and CaCO3 as the carrier, 1,758.
eBased on calculation (NRC, 1983).
an essential fatty acid source. Both vitamin and
mineral premixes were prepared in reference to catfish
test diets by Wilson (1991).
Essential amino acid concentrations in the basal 6.0, 8.0, 10.0, and 12.0 g/kg of diet at the expense of L-
diet were calculated based on their estimated availa- glutamic acid, resulting in total methionine concentra-
bility (Wilson, 1991) from each protein source. tions ranging from 3.5 to 15.5 g/kg in 2-g intervals.
Dietary quantities of essential amino acids were The dietary DL-Met treatments were expected to
adjusted by adding amino acids to simulate the catfish result in quadratic responses covering definite defi-
body amino acid profile reported by Wilson and Poe cient, optimum, and excessive ranges. Seven diets
(1985), except for methionine. The amino acid mix- were isonitrogenous by analysis and isocaloric by
ture contained six essential amino acids (Table 1 ) and calculation (Table 1).
cellufil (U.S. Biochemical, Cleveland, OH) as a In order to verify adequate culture conditions, Diet
carrier. 8 (Table 1 ) was formulated as a reference to
In preparation of Diets 1 to 7, the basal diet was approximate the conventional requirement for CP. It
supplemented with synthetic DL-Met at 0, 2.0, 4.0, contained fish meal and was similar to a practical
516 CAI AND BURTLE

Table 2. Effects of graded levels of DL-methionine on performance of

channel catfish fed all-plant-protein diets after 2-week intervalsa

Dietary methionine, g/kgb

Response 3.5 5.5 7.5 9.5 11.5 13.5 15.5 Reference SEM
Body weight, g/fish
Initial 14.36 14.39 14.28 14.43 14.50 14.26 14.32 14.51 .06
Finalcd 28.60 30.78 31.94 33.83 32.81 30.56 29.01 37.05 .48
Specific growth rate, %/de
0 to 2 wk 2.12 2.26 2.28 2.34 2.25 2.17 1.80 2.56 .11
0 to 4 wkcf 1.74* 2.04 2.12 2.18 2.14 1.97* 1.70* 2.36 .07
0 to 6 wkcd 1.63 1.81 1.92 2.03 1.94 1.81 1.64 2.23 .04
Weight gain, mg/d
0 to 2 wkcf 356 383 384 402 384 363 299* 446 22
0 to 4 wkcf 324* 397* 414 434 424 376* 317* 485 16
0 to 6 wkcd 339 390 420 462 436 388 350 537 11
Feed intake, mg/d
0 to 2 wk 584 602 593 614 622 576 596 595 10
0 to 4 wk 633 684 692 716 714 665 640 705 24
0 to 6 wkcf 671* 732 746 781 775 717* 692* 790 22
Gain:feed, g/kg
0 to 2 wkcf 605* 636 646 652 617 626 500 750 29
0 to 4 wkcd 505 580 598 605 593 565 494 686 19
0 to 6 wkcd 503 533 563 591 562 542 507 679 16
aData represent means of three tanks of fish per treatment, each tank held 20 fish for a 42-d period.
bAs formulated per kilogram of diets, the basal diet provided 3.5 g of methionine (Met), 4.0 g of cystine, and 12.0 g of L-glutamic acid
(Glu) and supplemented with 0, 2.0, 4.0, 6.0, 8.0, 10.0, and 12.0 g of DL-Met at the expense of Glu, respectively.
cQuadratic effect among seven dietary Met treatments ( P ≤ .01) except feed intake ( P < .05).
dAll means differed from the reference ( P < .05).
eSpecific growth rate = [(log BW
e final − loge BWinitial)/42] × 100.
fMeans in the same row were different from the reference ( *P ≤ .05), but any paired adjacent points among the seven DL-Met levels were
not compared statistically (Baker, 1986).

channel catfish diet (NRC, 1983) with modification by
adjusting values to 10 MJ/kg of DE, 5.7 g/kg of
available P, and adding 2.0 g/kg of synthetic DL-Met.
After mixing, the eight diets were individually
adjusted to pH 7.0 by adding 5 mol/L NaOH (Wilson
et al., 1977). The diets were passed through a
2-mm-diameter meat grinder die then dried at 70°C to
contain 8% moisture. Extrusions were crumbled to
approximately 4-mm pellets. The dried pellets were
bagged and frozen at −5°C until use, 1 to 9 wk.
Design and Maintenance
Because the fish were homogeneous in original BW,
a completely randomized design was used to assign
treatments to tanks. Seven treatments of graded
levels of dietary methionine and one reference treat-
ment were each assigned to three replicate tanks. The
experiment covered a 42-d period.
Fish were fed the diets twice daily as during the
acclimation period. For near satiation feeding, the
ration size was calculated so that a few pellets
remained in the bottom of tanks after each meal. The
diet allowance was adjusted weekly based on average
fish weight within each treatment. At the end of wk 1,
3, and 5, fish weights in each tank were estimated
based on feed efficiency obtained in the original and
previous weeks, respectively. In the first 3 wk, fish
were fed 2% BW for each meal. Beginning the 4th wk,
fish were fed 2% BW for the morning meal and 1.6%
BW for the afternoon meal to reduce diet waste. Any
uneaten pellets were counted and removed at 15 to 20
min after diets were offered. Actual diet intake was
the difference between the quantity offered and the
number of uneaten pellets multiplied by the average
pellet weight.
Photoperiod was timed at a 14:10 h (light:dark)
cycle. Each tank was supplied with filtered and
aerated well water at a rate of 1.2 L/min and an
airstone so that the minimum dissolved oxygen in any
tank was measured at > 6.3 mg/L. Water temperature
was controlled throughout the experiment by heating
to 24.8 ± .9°C.
Chemical Analysis
Fish samples were pooled on a per-tank basis for
analysis of DM (105°C for 24 h), nitrogen, fat (ether
extract), and energy by methods of AOAC (1984).
Nitrogen was determined by the macro Kjeldahl
method with a Kjeltec System 1026 Distilling Unit
(Tecator, Herndon, VA). Energy was determined with
air-dry samples by bomb calorimetry (Parr Instru-
ments, Moline, IL) following the manufacturer’s
procedure.
 METHIONINE REQUIREMENT OF CHANNEL CATFISH
Livers from five fish of each tank and the fish used
for analysis were individually blotted carefully to
remove moisture before weighing. The weighed liver
samples were pooled, chopped, and divided for tripli-
cate extraction of total lipids (Bligh and Dyer, 1959).
Total lipids were expressed as a percentage of the wet
tissue weight.
Feed samples were analyzed by methods of AOAC
(1984) for proximate composition of DM, GE, and CP
( N × 6.25). Analyses of amino acids in diets were
performed with a Beckman System 6300 High Perfor-
mance Amino Acid Analyzer (Beckman Instruments,
Palo Alto, CA) equipped with a HP 3390A Reporting
Integrator (Hewlett-Packard, Avondale, PA). Each
sample was prepared for analysis in triplicate before
hydrolysis by weighing and adding L-norleucine (N-
6877, Sigma Chemical, St. Louis, MO) as an internal
standard. The samples were hydrolyzed with 6 N HCl
under nitrogen for 22 h at 110°C for analysis of most
amino acids (Gehrke et al., 1985). For analysis of
SAA, the samples were oxidized with performic acid
(MacDonald et al., 1985) then hydrolyzed by the
same method as for other amino acids, measuring
methionine as methionine sulfone and cystine as
cysteic acid. Amino acids were separated on a
Beckman Na High Performance Column with Na
eluent and detected by colorimeter after postcolumn
reaction with ninhydrin. All buffer solutions and
reagents used for amino acid analysis were purchased
from Beckman Instruments.
Statistical Methods
Performance data in response to Diets 1 to 7 were
subjected to the GLM procedure (SAS, 1988) ap-
propriate for a completely randomized design with
tank mean as the experimental unit. Orthogonal
polynomial contrasts in equal spaces were made to
test for linear, quadratic, and cubic effects of dietary
DL-Met treatments. Any paired adjacent points
among the seven DL-Met levels were not compared for
statistical difference (Baker, 1986). Quadratic regres-
sion models were used to calculate the maximum
responses. The optimum methionine concentration
was the average of those where 95% of the highest
responses were calculated.
All response means, including Diet 8, were evalu-
ated by ANOVA (SAS, 1988). Orthogonal contrasts
were used separately to compare the means of data
from the reference treatment with data from the seven
DL-Met treatments. Significant effects were consi-
dered at P ≤ .05.
Results
Analytical and calculated compositions of ex-
perimental diets were similar (Table 1, data on Diets
2 to 7 are not shown). As expected by the experiment
design, DL-Met was present at equal intervals accom-
517
panying a decrease in concentrations of L-glutamic
acid in Diets 1 to 7 from 59.8, 57.7, 55.7, 53.5, 51.1,
49.1, to 46.9 g/kg, respectively. Cystine was constant
within the seven diets.
Seven dietary DL-Met concentrations resulted in
quadratic ( P < .01), but not linear or cubic ( P > .10),
responses in most of the measured growth criteria
(Table 2). When catfish were fed diets with increasing
DL-Met levels, final body weight, specific growth rate
( SGR) , weight gain, feed intake, and the gain to feed
ratio increased uniformly to maxima at Diet 4, then
declined to minima at Diet 7. There were no such
quadratic effects found in SGR and ADG during the
period from wk 0 to 2, and in ADFI during the period
from wk 0 to 2 or to 4. Methionine supplementation
above requirement levels decreased weight gain but
did not cause mortality. Only three fish died through-
out the experiment, presumably injured during han-
dling at weighing.
Body content of protein, fat, and energy increased
but water content decreased during the experimental
period compared with initial values (Table 3). The
seven dietary Met treatments resulted in quadratic
effects on fat and energy contents ( P < .01). Diet 4
(led to maximum responses) resulted in the lowest
liver lipids among all treatments ( P < .05). Based on
tissue analyses, such calculated responses as protein,
fat, or energy gain, CP or GE intake, protein efficiency
ratio ( PER) , apparent net protein utilization ( AN-
PU) , and apparent net energy utilization ( ANEU)
increased quadratically ( P < .01), but not linearly ( P
> .05), with increasing dietary Met concentrations.
The maximum responses uniformly appeared at Diet
4.
The dietary methionine requirement was compre-
hensively determined by main regression models
listed in Table 4. In addition, all quadratic responses
confirmed that criteria measured did not increase
after the 9.5 g/kg level (Diet 4). The methionine
requirement was calculated as 95% of the maximum
responses. As a result, the methionine requirement
was 9.4 g/kg of dry diet (34.1 g/kg of dietary protein)
in the presence of 4.0 g of cystine/kg of dry diet.
Based on carefully measured feed intake data, an
absolute methionine requirement was determined as
270 mg/kg of fish weight per day from a broken-line
plot of net protein gain against methionine intake
(Figure 1). With increasing DL-Met intake, net
protein gain increased from 2.8 to 3.4 g/kg of fish, then
reached a plateau. Net protein gain decreased sharply
between the 430- and 500-mg methionine intake.
Discussion
Experimental Approach
Multiple response criteria used in this study
provided more information for the estimation of
methionine requirement than weight gain data alone.
518 CAI AND BURTLE

Table 3. Effects of dietary methionine concentrations on tissue composition

and growth performance criteria for channel catfisha

Dietary methionine, g/kgb

Response 3.5 5.5 7.5 9.5 11.5 13.5 15.5 Reference SEM
Composition, % (except as noted) c
Water 75.07 73.91 73.74 73.64 73.80 74.12 75.47 74.20 .27
Protein 17.01 17.14 17.53 17.26 17.41 17.89 17.27 17.31 .16
Fatde 6.71 7.08 7.56 7.82 7.46 6.41 5.88* 7.28 .21
Energy, KJ/gde 6.12 6.49 6.50 6.63 6.54 6.39 5.85* 6.46 .10
Liver lipidse 2.68 2.72 2.82 2.56* 2.90 2.78 3.01 3.09 .09
Daily gain
Protein, mgdf 61 70 78 83 80 75 64 97 2
Fat, mgde 40* 46* 52 57 52 41* 35* 58 2
Energy, KJde 2.75* 3.34* 3.54* 3.91 3.68* 3.24* 2.63* 4.27 .11
Daily intake
CP, mgdf 185 204 205 217 212 196 190 262 3
GE, KJde 11.66* 12.95* 13.30* 13.76 13.34* 12.36* 11.86* 14.58 .22
Efficiency
PERdeg 1.82* 1.92 2.05 2.13 2.05 1.99 1.84 2.05 .04
ANPU, %deh 32.5* 34.4 38.0 38.4 37.6 38.2 33.7 36.9 .9
ANEU, %dei 23.5* 25.7* 26.6 28.4 27.5 26.3 22.2* 29.2 .7
aData represent means of three tanks of fish per treatment, each tank held 20 fish for a 42-d period. Protein was calculated as N × 6.25.
Fat was obtained by ether extract.
bAs formulated per kilogram of diets, the basal diet provided 3.5 g of methionine (Met), 4.0 g of cystine, and 12.0 g of L-glutamic acid
(Glu) and was supplemented with 0, 2.0, 4.0, 6.0, 8.0, 10.0, and 12.0 g of DL-Met at the expense of Glu, respectively.
cValues for body water, protein, fat, and energy in the initial fish were 78.96, 16.19, 1.73%, and 4.15 KJ/g, respectively.
dQuadratic effect ( P < .01) among seven dietary methionine treatments.
eMeans in the same row were different from the reference mean ( *P < .05), but any paired adjacent points among the seven DL-Met levels
were not compared statistically (Baker, 1986).
fMeans differed from the reference ( P < .05).
gProtein efficiency ratio (PER) = (BW gain)/(protein consumed).
hApparent net protein utilization (ANPU) = (protein gain in fish body)/(protein consumed).
iApparent net energy utilization (ANEU) = (energy gain in fish body)/(GE consumed).

Both feed intake and feed conversion efficiency have

been rarely considered as response criteria for deter-
mining amino acid requirements of fish (Lovell,
1989). Of the amino acids required for protein
synthesis, methionine is clearly the most toxic
(Benevenga and Steele, 1984). Therefore, graded
levels of DL-Met in diets should result in responses for
multiple criteria ranging from deficiency through
adequacy to excess (Baker, 1987). Fed the seven
diets, catfish in our study responded to multiple
criteria in a uniform and quadratic manner. As shown
in Tables 2 and 3, the best performance occurred at a
similarly adequate methionine concentration (Diet 4).
In most nutritional requirement studies with fish,
the duration of experiments is typically from 6 to 24
wk, whereas nutritional studies with other vertebrates
can be shorter, 1 to 3 wk. In some amino acid
requirement studies with fish, statistically significant
differences in gross measurement alone, such as
weight gain and feed efficiency, may not become
apparent until later in the study. Gradations in all Figure 1. Relationship between net protein gain and
measured criteria of our study became obvious begin- methionine intake in channel catfish fed soybean meal-
ning at 4 wk (at 2 wk for gain/feed) for growth corn-based diets for 42 d. An absolute methionine
responses, except feed intake. Moreover, all criteria requirement is determined as 270 mg/kg of fish weight
measured responded to Diet 4 independent of response per day excluding an excessive intake observation (500
criterion and time interval. mg/kg of fish per day).
 METHIONINE REQUIREMENT OF CHANNEL CATFISH 519
Table 4. Estimation of percentage of dietary methionine required for 95% of the
maximum responses based on quadratic equations of performance in channel
catfish fed diets with seven graded concentrations of dietary methioninea

Requirement estimated
for 95% of the
Response R2 P< maximum response
Weight gain, mg/d .951 .002 .938
Protein gain, mg/d .986 .001 .958
Fat gain, mg/d .926 .006 .882
SGR, %/db .983 .001 .929
DMI, mg/d .922 .006 .946
GE intake, KJ/d .945 .003 .916
Gain/feed, g/kg .951 .002 .936
PER, g/gc .961 .002 .946
ANPU, %d .877 .015 .988
ANEU, %e .928 .005 .915
Mean (SD = .028): — — .935
aData were based on means of three tanks of fish per treatment, each tank held 20 fish for a
42-d period. Protein was calculated as N × 6.25.
bSpecific growth rate = [(log BW
e final − loge BWinitial)/42] × 100.
cProtein efficiency ratio (PER) = (BW gain)/(protein consumed).
dApparent net protein utilization (ANPU) = (protein gain in fish body)/(protein consumed).
eApparent net energy utilization (ANEU) = (energy gain in fish body)/(GE consumed).

The performance of catfish fingerlings in our study
reflects the transition to a plant protein-based diet.
Although experimental catfish in our study were fed
soybean meal-corn-based diets, the weight gain of fish
receiving near optimum DL-Met (Diet 4 ) was 235% of
their initial weight after d 42, and even the most
deficient or excessive DL-Met treatment doubled the
initial weight (Table 2). Compared with other amino
acid experiments in channel catfish cited by NRC
(1993), average daily gain of catfish fed Diet 4 was
higher than in those studies of catfish fed purified
diets at 26.7°C. The feed efficiency (59%) in our study
(soybean-corn-based diet fed to satiation) was inferior
to that noted in some studies for catfish amino acid
requirements (casein-gelatin-based diets and res-
tricted feeding) but superior to 48, 54, and 57% in
studies for arginine (Robinson et al., 1981),
phenylalanine (Robinson et al., 1980b), and lysine
(Robinson et al., 1980a), respectively. In our study,
catfish fed Diet 4 had 14% lower ADG and 13% lower
gain/feed than fish fed the reference diet, but the
differences were much lower than those differences
between the best performance and reference diets that
were shown in 56-d amino acid experiments with
channel catfish cited by NRC (1983, 1993) except the
42-d methionine experiment conducted by Harding et
al. (1977). The bioavailabilities of amino acids from
our all-plant-protein diets were probably lower than
those from our reference diet, which contained fish
meal.
Methionine Requirement
The dietary methionine requirement for channel
catfish is recommended as 5.6 g/kg diet or 23.4 g/kg of
dietary protein (NRC, 1983, 1993). That recommen-
dation is based on one published trial of Harding et al.
(1977) in which experimental diets were formulated
with purified ingredients containing a calculated
amino acid pattern similar to whole-egg protein but
lacking cystine and containing 24% CP.
Based on all performance criteria measured in our
study, fish fed the 9.4 g/kg level of methionine intake
demonstrated maximum growth with the greatest
nitrogen retention and feed efficiency. That methio-
nine requirement (9.4 g/kg of diet or 34.1 g/kg of
dietary protein) was derived using mathematical
models that are used more for animal nutrition
studies (Baker, 1986) than for fish studies. Also, the
total digestible SAA requirement (11.3 g/kg diet) can
be estimated for channel catfish from our data (Table
5). In particular, an absolute methionine intake
requirement level was determined as 270 mg of Met/
kg of fish BW per day, which agrees with a dietary
requirement level if catfish were fed a restricted rate
of 3% of BW/d. The total dietary methionine require-
ment derived from our study was higher than that
previously reported for catfish by Harding et al.
(1977), probably because of differences in the amino
acid availability of the used protein sources. The
methionine digestibility for channel catfish was
reported as 84.6% for soybean meal and 70.5% for corn
(Wilson, 1991), whereas it is probably greater from
purified ingredients. Also, the dietary requirement
from our study is in the range of methionine
requirement (20 to 40 g Met/kg dietary protein) for
seven other species of fish reviewed by Wilson (1989).
Effects of Dietary Methionine Concentrations
The highest DL-Met treatment (Diet 7), which
decreased growth rate, was analyzed as 16.9 g/kg diet,
520 CAI AND BURTLE

Table 5. Calculation of the digestible sulfur-containing amino acid (SAA)

requirement in channel catfish fed the experimental diets

g/kg of
Bases of calculation dry diet
L-methionine (Met) from the basal diet: 3.6 × .82a 3.0
L-cystine (Cys) from the basal diet: 4.0 × .82a 3.3
DL-Met needed for the requirement level: 9.4 − 3.6b 5.8
Estimated total digestible SAA requirement 12.1
Of the total digestible SAA requirement (12.1)
Needed digestible Met: 12.1 × (1.0 − .6) c 4.8
Needed digestible Cys: 12.1 × .6c 7.3
Of the DL-Met (5.8) supplemented, all digestible
DL-Met as Met: 4.8 − 3.0 1.8
DL-Met to Cys: 7.3 − 3.3 4.0
Actual yield of Cys from DL-Met supplementation: 4.0 × .8d 3.2
Digestible SAA from both DL-Met supplementation and the basal diet
From the basal diet (Met + Cys): 3.0 + 3.3 6.3
From the DL-Met supplementation: 1.8 + 3.2 5.0
Total digestible SAA requirement: 6.3 + 5.0 11.3
aAverage essential amino acid digestibility in soybean meal and corn is estimated as 82% for channel
catfish based on data of Wilson (1991).
bSynthetic DL-Met was reported to be 100% available for channel catfish (Robison et al., 1978).
cDietary cystine can save 60% of methionine requirement in channel catfish (Harding et al., 1977).
dDL-Methionine converts to cystine with 80% efficiency via trans-sulfuration on a weight or concentra-
tion basis (Chung and Baker, 1992).

or approximately 80% higher than the requirement
level of 9.4 g/kg. The decreased growth rate was
similar to that observed in rats (Benevenga and
Steele, 1984), chicks, and pigs (Baker, 1987, 1989;
Han and Baker, 1993) fed excess methionine but was
contradictory to the finding of Harding et al. (1977)
for catfish. In the study of Harding et al., catfish fed
diets containing 7.5 to 17.5 g/kg (2.5 g/kg intervals) of
methionine had similar weight gain and feed effi-
ciency responses. Possibly, restriction of food intake
when feeding 17.5 g of methionine/kg did not cause a
decrease in growth rate in that study. Baker (1987)
indicated that a methionine level twice the require-
ment is generally well tolerated by animals but at
threefold or higher will result in toxicity. In our study,
the decrease in most performance criteria occurred
generally within the twofold range, and protein gain
decreased beginning at the diet containing 50% more
DL-Met than the requirement (Figure 1).
Similar liver total lipids among treatments confirm
no adverse lipotropic effect of the seven dietary DL-
Met levels. That evidence indicates that soybean-corn
diets had little effect on catfish liver lipid accumula-
tion. Diet 4 liver lipid was lower than the value for the
reference diet, which supports the positive lipotropic
value of adequate methionine supplementation.
Potential of Using All-Plant Proteins
in Catfish Diets
Our results demonstrate the potential use of an all-
plant-protein diet when crystalline amino acids are
added in amounts similar to the whole-body amino
acid composition of channel catfish. Alternatively, a
certain amount of fish meal with soybean meal could
actually provide an amino acid pattern similar to the
pattern used in our study. An example is the study
conducted by Mohsen and Lovell (1990) for partial
substitution of soybean meal in channel catfish diets
with different animal protein sources. These research-
ers found that increases in weight gain produced by
including fish meal were related to increasing dietary
levels of the most limiting amino acids. However,
because of relatively lower availability of amino acids
than in fish meal, the increase in catfish weight gain
affected by including other animal protein sources did
not parallel higher dietary amino acid concentrations.
Another case was reported in which growth of rainbow
trout was improved when the trout were fed soybean
meal-based diets supplemented with essential amino
acids to simulate those in the trout egg or body protein
profile (Rumsey and Ketola, 1975).
With soybean-corn diets, catfish grew rapidly with
daily SGR of 2%, as is normal for young fish. In
practice, SGR is used to compare the instantaneous
rates of fish growth on a unit time basis (Ricker,
1979). As our results indicated, catfish fed an all-
plant-protein diet with adequate methionine sup-
plementation (Diet 4 ) had maximum 38.4% ANPU
that was higher than previous nitrogen retention
values of 30 to 35% for channel catfish (Lovell, 1989)
and 28.4% ANEU that approximated 29% dietary
energy utilization efficiency for growth in carnivorous
species of fish (Brett and Groves, 1979). When catfish
were fed Diet 4, maximum energy retention, feed and
energy intakes, PER, ANPU, and ANEU were identi-
cal to those of fish fed Diet 8 (reference) in spite of
20% lower CP intake than the reference.
 METHIONINE REQUIREMENT OF CHANNEL CATFISH
Feed intake results verify that an all-plant-protein
diet with adequate supplementation of DL-Met was
palatable without obvious adverse effects for catfish.
As shown in Table 2, feed consumption was similar in
all eight diets from wk 0 to 4. During the period from
0 to 6 wk, feed intakes for catfish fed Diet 2, 3, 4, or 5
were similar to those of catfish fed Diet 8 that
contained 8% fish meal.
In conclusion, all-plant-protein soybean meal-corn-
based diets should contain 9.4 g of methionine (or 34.1
g/kg of CP) in the presence of cystine at 4 g/kg of dry
diet ( a minimal total digestible SAA 11.3 g/kg) to
provide 270 mg of Met/kg of fish per day for the best
growth and highest nitrogen retention when fed to
fingerling channel catfish. The responses in feed
intake, energy retention and efficiency, PER, ANPU,
and ANEU to the adequately supplemented methio-
nine diet with low protein (27.6%) were identical to
those of a conventional diet containing fish meal with
higher protein (33.1%). The results indicate that an
all-plant-protein diet formulated with an amino acid
pattern similar to the carcass protein has much
promise in channel catfish diet formulation. However,
additional data must be collected to determine the
availability of amino acids from plant materials for
channel catfish.
Implications
The results of this experiment indicate that catfish
fed diets with reduced fish meal and increased plant
proteins have a higher dietary methionine require-
ment than previously reported for catfish fed semipu-
rified diets. Our study indicates that channel catfish
can perform better on lower crude protein, limiting
amino acid-supplemented diets than previously sug-
gested and supports the catfish industry trend toward
reduced dietary animal protein. The dietary methio-
nine requirement should be established as 9.4 g/kg
when soybean meal-corn diets are formulated for
channel catfish.
Literature Cited
AOAC. 1984. Official Methods of Analysis (14th Ed.). Association of
Official Analytical Chemists, Arlington, VA.
Baker, D. H. 1986. Problems and pitfalls in animal experiments
designed to establish dietary requirements for essential
nutrients. J. Nutr. 116:2339.
Baker, D. H. 1987. Construction of assay diets for sulfur-containing
amino acids. Methods Enzymol. 143:297.
Baker, D. H. 1989. Amino acid nutrition of pigs and poultry. In: W.
Haresign and D.J.A. Cole (Ed.) Recent Advances in Animal
Nutrition. p 245. Butterworth Heinemann, London.
Benevenga, N. J., and R. D. Steele. 1984. Adverse effects of exces-
sive consumption of amino acids. Annu. Rev. Nutr. 4:157.
Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid
extraction and purification. Can. J. Biochem. Physiol. 37:911.
Brett, J. R., and T.D.D. Groves. 1979. Physiological energetics. In:
521
W. S. Hoar, D. J. Randall, and J. R. Brett (Ed.) Fish Physiol-
ogy. Vol. 8. p 279. Academic Press, New York.
Chung, T. K., and D. H. Baker. 1992. Maximal portion of the young
pig’s sulfur amino acid requirement that can be furnished by
cystine. J. Anim. Sci. 70:1182.
Finke, M. D., J. R. DeFoliart, and N. J. Benevenga. 1987. Use of
simultaneous curve fitting and a four parameter logistic model
to evaluate the nutritional value of protein sources at growth
rates of rats from maintenance to maximum gain. J. Nutr. 117:
1681.
Gehrke, C. W., L. L. Wall, J. S. Absheer, F. E. Kaiser, and R. W.
Zumwalt. 1985. Sample preparation for chromatography of
amino acids: Acid hydrolysis of protein. J. Assoc. Off. Anal.
Chem. 68:811.
Grant, G. 1989. Anti-nutritional effects of soybean: A review. Prog.
Food & Nutr. Sci. 13:317.
Han, Y., and D. H. Baker. 1993. Effects of excess methionine or
lysine for broilers fed a corn-soybean meal diet. Poult. Sci. 72:
1070.
Harding, D. E., O. W. Allen, and R. P. Wilson. 1977. Sulfur amino
acid requirement of channel catfish: L-methionine and L-cys-
tine. J. Nutr. 107:2031.
Ketola, H. G. 1982. Amino acid nutrition of fishes: Requirements
and supplementation of diets. Comp. Biochem. Physiol. 73B:17.
Lovell, T. 1989. Nutrition and Feeding of Fish. Van Nostrand Rein-
hold, New York.
MacDonald, J. L., M. W. Krueger, and J. H. Keller. 1985. Oxidation
and hydrolysis determination of sulfur amino acids in food and
feed ingredients: Collaborative study. J. Assoc. Off. Anal.
Chem. 68:826.
Mohsen, A. A., and R. T. Lovell. 1990. Partial substitution of soy-
bean meal with animal protein sources in diets for channel
catfish. Aquaculture 90:303.
NRC. 1983. Nutrient Requirements of Warmwater Fishes and Shell-
fishes. National Academy Press, Washington, DC.
NRC. 1993. Nutrient Requirements of Fish. National Academy
Press, Washington, DC.
Ricker, W. E. 1979. Growth rates and models. In: W. S. Hoar, D. J.
Randall, and J. R. Brett (Ed.) Fish Physiology. Vol. 8. p 677.
Academic Press, New York.
Robinson, E. H., O. W. Allen, Jr., W. E. Poe, and R. P. Wilson. 1978.
Utilization of dietary sulfur compounds by fingerling channel
catfish: L-methionine, DL-methionine, methionine hydroxy
analogue, taurine and inorganic sulfate. J. Nutr. 108:1932.
Robinson, E. H., R. P. Wilson, and W. E. Poe. 1980a. Re-evaluation
of the lysine requirement and lysine utilization by fingerling
channel catfish. J. Nutr. 110:2313.
Robinson, E. H., R. P. Wilson, and W. E. Poe. 1980b. Total aromatic
amino acid requirement, phenylalanine requirement and tyro-
sine replacement value for fingerling channel catfish. J. Nutr.
110:1805.
Robinson, E. H., R. P. Wilson, and W. E. Poe. 1981. Arginine
requirement and apparent absence of a lysine-arginine an-
tagonist in fingerling channel catfish. J. Nutr. 111:46.
Rumsey, G. L., and H. G. Ketola. 1975. Amino acid supplementation
of casein in diets of Atlantic salmon fry and of soybean meal for
rainbow trout fingerlings. J. Fish. Res. Board Can. 32:422.
SAS. 1988. SAS/STAT User’s Guide (Release 6.03). SAS Inst. Inc.,
Cary, NC.
Wilson, R. P. 1989. Amino acids and protein. In: J. E. Halver (Ed.)
Fish Nutrition. p 111. Academic Press, New York.
Wilson, R. P. 1991. Channel catfish. In: R. P. Wilson (Ed.) Hand-
book of Nutrient Requirements of Finfish. p 35. CRC Press,
Boca Raton, FL.
Wilson, R. P., D. E. Harding, and D. L. Garling, Jr. 1977. Effect of
dietary pH on amino acid utilization and the lysine require-
ment of fingerling channel catfish. J. Nutr. 107:166.
Wilson, R. P., and W. E. Poe. 1985. Relationship of whole body and
egg essential amino acid patterns to amino acid requirement
patterns in channel catfish. Comp. Biochem. Physiol. 80B:385.