Aquaculture nutrition, tập 16, số 6, 2010

Aquaculture Nutrition
2010 16; 559–568

..........................................................................................

doi: 10.1111/j.1365-2095.2009.00685.x

Research Institute of Fish Culture and Hydrobiology, University of South Bohemia, Vodnany, Czech Republic

Due to distinctive feeding habits and digestive physiology
determination of nutrient digestibility is more demanding with
crustacean than fish species. A study was conducted to validate
the use of linear prediction equations established with fish
species to predict apparent digestible protein (ADP) and lipid
(DL) contents [g kg)1 dry matter (DM)] of feed ingredients
(ADP = )10.0731 + 0.8942 CP, DL = )1.5824 + 0.8654
CL) and compound diets (ADP = )51.4001 + 0.9872 CP,
DL = )2.7303 + 0.9123 CL) from dietary crude protein
(CP) and lipid (CL) contents for crustaceans (shrimp, lobster,
crab). Observed values (n = 91) obtained in 17 studies, which
evaluated CP digestibility of 26 feed ingredients for eight
crustacean species, presented a linear relationship
(R2 = 0.9213, RMSE = 54.1245) with predicted values.
Predicted values overestimated observed values with a mean
prediction error (MPE) of 0.1315. However, observed DL
values of 17 feed ingredients (n = 31; studies = 5; species = 5) were overestimated with a MPE of 0.4500. Similar
trends than above were found with prediction of the ADP
(n = 185; studies = 32; species = 11; R2 = 0.8047;
MPE = 0.1002) and DL (n = 64; studies = 11; species = 3;
R2 = 0.6907; MPE = 0.2372) contents of compound diets.
KEY WORDS:

WORDS:

crustaceans, digestible protein, digestible lipid,

prediction
Received 26 June 2008, accepted 27 February 2009
Correspondence: James Sales, Research Institute of Fish Culture
and Hydrobiology, University of South Bohemia, Zatisi 728, 38925
Vodnany, Czech Republic. E-mail:

Measurement of nutrient digestibility of diets and feed
ingredients, which is of utmost importance to nutritionists

..............................................................................................

Ó 2009 Blackwell Publishing Ltd

and feed formulators to optimize nutritional value and cost
of diets (Smith et al. 2007) and in the application of waste
management (Cousin et al. 1996), is more problematic with
crustaceans than fish. Compared to fish, crustaceans are
slow eaters (Ishikawa et al. 1997), resulting in leaching
losses of nutrients before feed is consumed. Furthermore,
collection of faeces by stripping or dissection, commonly
used with fish, are no feasible methods with crustaceans
(Smith & Tabrett 2004). The conventional ÔGuelphÕ system,
based on settlement of faeces (Cho et al. 1982) and widely
used in digestibility studies with fish, creates difficulties
with crustaceans due to their feeding habits and coprophagy (Martı´ nez-Palacios et al. 2001). Several crustacean
species, such as the common prawn (Palaemon serratus)

and spot shrimp (Pandalus platyceros) (Forster & Gabbott
1971), hybrid lobster (Homarus sp.) (Bordner et al. 1983),
American lobster (Homarus americanus) (Leavitt 1985), red
swamp crayfish (Procambarus clarkii) (Brown et al. 1986)
and common yabby (Cherax destructor) (Jones & De Silva
1997a), have the ability to selectively partition some components of its diet during digestion, complicating the use of
markers in digestibility studies. In addition, the regurgitation of part of the indigestible feed after a meal has been
reported in some crustaceans (Forster & Gabbott 1971;
Newman et al. 1982).
Recommendations for the study of feed digestibility in
crustaceans include in vitro techniques applied as screening
devices of the suitability of feed ingredients for inclusion in
shrimp diets (Lee & Lawrence 1997; Lazo et al. 1998).
Digestive proteases from the animal under study, and
enzymes extracted from animals fed the same diet that
will be evaluated, can better access the digestibility of
protein than commonly used or commercial enzymes that
are not present in the shrimpÕs digestive system, or are
acting on a different pH than in the shrimpÕs digestive
gland (Lan & Pan 1993; Divakaran et al. 2004; Lemos
et al. 2004). The pH-stat method, based on the degree
of hydrolysis using digestive enzymes, has the potential for


estimating the digestibility of alternative protein sources
for inclusion in shrimp feeds (Ezquerra et al. 1997), but
was found to be inadequate with shrimp feeds that have
partially been hydrolysed (Co´rdova-Murueta & Garcı´ aCarren˜o 2002).
However, it was found with fish that apparent digestible
protein (ADP; Sales 2008) and lipid (DL; Sales 2009) contents in feed ingredients and compound diets can be predicted with high accuracy from its dietary contents across a

wide range of species, feed types, nutrient levels, life stages
and rearing conditions with the use of linear regression
techniques. With crustaceans a similar analysis is hampered
by a lack of studies presenting suitable values. Despite the
assumption that static empirical models should only be
applied within ranges, thus also species, used for development (Sales 2008, 2009), the present study was aiming at
validation of prediction equations established for the prediction of ADP and DL contents of fish feeds for crustacean
species.

Databases were created from studies (Tables 1 & 2) presenting dietary contents and digestibility coefficients of crude
protein (CP) and crude lipid (CL) for feed ingredients
(Table 3) and compound diets (Table 4) evaluated with several crustacean species.
Characteristics of datasets are presented in Table 5. Tuan
et al. (2006) used three, four and five replicates in the same
study to evaluate CP digestibility of different feed ingredients
with mud crabs. Irvin & Williams (2007) collected faeces in
tropical spiny lobsters with a balloon glued to surround the
anal pore, as described by Irvin & Tabrett (2005). Whereas
Brown et al. (1986) and Akiyama et al. (1989) utilized single
protein source diets to determine apparent protein digestibility
of feed ingredients in red swamp crayfish and Pacific white
shrimp, respectively, all other studies applied the reference

Table 1 Studies used for information on dietary contents and apparent digestibility of crude protein and crude lipid in feed ingredients
evaluated with crustaceans

Reference

Shrimp
Akiyama et al. (1989)

Bautista-Teruel et al. (2003)
Cruz-Sua´rez et al. (2001)
Cruz-Sua´rez et al. (2007)
Davis et al. (2002)
Ezquerra et al. (1997)
Herna´ndez et al. (2008)
Kumaraguru Vasagam
et al. (2007)
Merican & Shim (1995)
Smith et al. (2007)

Water
type

Temperature
(°C)

Feed
habit2

Pacific white shrimp
Black tiger shrimp
Blue shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Black tiger shrimp

Litopenaeus vannamei

Penaeus monodon
Litopenaeus stylirostris
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Penaeus monodon

Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt

25–29
22–28
28
27–31
28
25
28
28

O
O
O
O

O
O
O
O

22.3
14.5
2.7
1.6–2.0
8.0–10.0
3.5–4.0
5.2
4.0

Black tiger shrimp
Black tiger shrimp

Penaeus monodon
Penaeus monodon

Salt
Salt

26–27
29

O
O

20.7

23.5

Protein
Protein
Protein
Protein
Protein
Protein
Protein
Protein,
lipid
Lipid
Protein

7
2

Red swamp crayfish
Australian redclaw

Procambarus clarkii
Fresh
Cherax quadricarinatus Fresh

24–27
27

O
O

14.1
3.6

Protein
Protein

7

Australian redclaw

Cherax quadricarinatus Fresh

27

O

3.6

8

Tropical spiny lobster

Panulirus ornatus

28

C

700.0

7
3

Australian redclaw
Red swamp crayfish

Cherax quadricarinatus Fresh
Procambarus clarkii
Fresh

27
22

O
O

94.0
40–50 mm

9

Mud crab

Scylla serrata

Salt

–

C

148.0

3
5

Chinese hairy crab
Mud crab

Eriocheir sinensis
Scylla serrata

Fresh
Salt

28
29

O
C

5.1
96.0

5
12

Pavasovic et al. (2007)
Reigh et al. (1990)

Crab
Catacutan et al. (2003)
Mu et al. (2000)
Tuan et al. (2006)
2

Scientific name

7
1
6
2
1
7
2
6

Crayfish
Brown et al. (1986)
Campan˜a Torres
et al. (2005)
Campan˜a Torres
et al. (2006a)
Irvin & Williams (2007)

1

Common name

n1

Salt

Size (g)

Nutrient

Lipid
Protein,
lipid
Protein
Protein
Protein,
lipid
Protein
Protein

Number of means from study.
C, carnivorous; O, omnivorous.

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd


..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd
1
2

2
9
1
1
12

Crayfish
Campan˜a Torres et al. (2005)
Irvin & Williams (2007)
Jones & De Silva (1997a)
Jones & De Silva (1997b)
Pavasovic et al. (2006)
Reigh et al. (1990)
Ward et al. (2003)

2

Number of means from study.
C, carnivorous; O, omnivorous.

10
1
8
4
10
2
2
5
22
4

4
6
8
1
6
5
6
2
7
6
18
4
5
5
6

Shrimp
Ashmore et al. (1985)
Brunson et al. (1997)
Cabanillas-Beltra´n et al. (2001)
Catacutan (1991)
Colvin (1976)
Cruz-Sua´rez et al. (2001)
Cruz-Sua´rez et al. (2007)
Eusebio (1991)
Fenucci et al. (1982)
Forster et al. (2003)
Goytortu´a-Bores et al. (2006)
Herna´ndez et al. (2008)
Kumaraguru Vasagam et al. (2005)

Kumaraguru Vasagam et al. (2007)
Lee & Lawrence (1985)
Lin et al. (2004)
Lin et al. (2006)
Martı´nez-Palacios et al. (2001)
Ostrowski-Meissner et al. (1995)
Rivas-Vega et al. (2006)
Smith et al. (1985)
Sudaryono et al. (1996)
Sudaryono et al. (1999a)
Sudaryono et al. (1999b)
Taechanuruk & Stickney (1982)

1

n1

Reference

Australian redclaw
Tropical spiny lobster
Yabby
Yabby
Australian redclaw
Red swamp crayfish
Southern rock lobster

Giant river prawn
Atlantic white shrimp
Pacific white shrimp

Black tiger shrimp
Indian white shrimp
Blue shrimp
Pacific white shrimp
Black tiger shrimp
Blue shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Black tiger shrimp
Black tiger shrimp
Atlantic white shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Pacific white shrimp
Black tiger shrimp
Black tiger shrimp
Black tiger shrimp
Giant river prawn

Common name

Cherax quadricarinatus
Panulirus ornatus
Cherax destructor
Cherax destructor
Cherax quadricarinatus

Procambarus clarkii
Jasus edwardsii

Macrobrachium rosenbergii
Penaeus setiferus
Litopenaeus vannamei
Penaeus monodon
Penaeus indicus
Litopenaeus stylirostris
Litopenaeus vannamei
Penaeus monodon
Litopenaeus stylirostris
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Penaeus monodon
Penaeus monodon
Penaeus setiferus
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Litopenaeus vannamei
Penaeus monodon
Penaeus monodon
Penaeus monodon
Macrobrachium rosenbergii

Scientific name

Fresh
Salt
Fresh
Fresh
Fresh
Fresh
Salt

Fresh
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt
Salt

Salt
Salt
Salt
Fresh

Water
type

27
28
21–24
–
26–28
22
18

27
25–30
22
27–29
28–30
28
27–31
25–28
26–29
27
28
28
28
28

28
29–31
24–27
28
27
27
27
25–28
28–30
28
27–31

Temperature
(°C)

O
C
O
O
O
O
C

O
O
O
O
O
O
O

O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O

Feed
habit2

3.6
700.0
8.3–16.0
7.0–15.0
29.6
40–50 mm
8.0–13.0

40–50
7.7
10.2
30–40
0.4–1.1
2.7
1.6–2.0
8.7
1.2–11.9
5.1
3.3
5.2
3.3
4.0
3.7, 9.8
11.5
8.6
10.2
1.5
15.4
8.4–11.0
4.7
4.3
4.1
>30

Size (g)

Table 2 Studies used for information on dietary contents and apparent digestibility of crude protein and crude lipid in compound diets evaluated with crustaceans

lipid

lipid
lipid
lipid

lipid
lipid
lipid
lipid

lipid
lipid

Protein
Protein, lipid
Protein
Protein
Protein
Protein
Protein

Protein
Protein
Protein,
Protein,
Protein
Protein
Protein
Protein

Protein
Protein
Protein,
Protein,
Protein,
Protein,
Protein
Protein,
Protein,
Protein,
Protein
Protein
Protein,
Protein
Protein
Protein
Protein

Nutrient


Table 3 Dietary crude protein (CP) and lipid (CL) contents (g kg)1 dry matter; range) and apparent protein (APDC) and lipid (ALDC)
digestibility coefficients (%; range) of individual feed ingredients
Ingredient

CP

APDC

Animal protein

Blood meal
Casein
Crab meal
Crustacean meal
Fish meal
Krill meal
Meat meal
Mussel meal
Poultry meal
Shrimp meal
Squid meal

875.0
880.0–913.0
394.0
378.0
602.0–766.6
642.0
468.0–596.0
478.0
692.0–693.6
436.0–704.6
720.0–785.0

93.5
90.9–99.1
66.4
85.3
47.4–95.0
88.7

66.2–95.0
88.8
87.2–90.4
74.6–94.9
59.3–97.6

Plant protein
Canola meal
Copra meal
Cow peas
Lupin
Mung beans
Peas
Soybean meal
Soy concentrate

394.0–441.0
219.0
242.0–293.0
308.0–546.0
249.0–284.0
211.0–252.5
404.9–551.0
495.0–816.0

80.0–91.0
94.3
60.0–83.0
85.9–96.8
65.0–83.0

79.8–92.7
80.6–98.7
90.9–96.4

486.0
78.0
120.0–192.0

92.6
96.4
76.4–94.2

114.8–139.0
212.0
759.0
166.0–184.1

57.7–95.2
95.7
95.0–98.0
92.5–96.0

Grains
BrewerÕs yeast
Maize
Rice bran
Sorghum
Wheat
Wheat bran
Wheat gluten

Wheat midlings

substitution method (Cho et al. 1982), with from 150 to
500 g kg)1 of the reference diet replaced with the feed ingredient. In the former two studies diets contained 60 and 40 g
kg)1 dry matter (DM), respectively, of frozen squid blended in
water as an attractant, which might have contributed to protein content. Eleven of the 19 studies indicated in Table 1 have
incorporated the relative contribution of the nutrient to the
combined diet in calculation of feed ingredient digestibility, as
described by Forster (1999). Although crustacean meal,
mussel meal and mung beans were absent in the development
of the equation to predict ADP contents of feed ingredients for
fish (Sales 2008), and crustacean meal, krill meal, mussel meal,
copra meal, cow peas, lupin and mung beans were not included
in the equation for DL prediction (Sales 2009), values for these
ingredients were included in the present evaluation.
Studies using purified and semi-purified diets (Bordner
et al. 1983; Shiau & Chou 1991; Shiau et al. 1991a,b; Shiau &
Peng 1992; Koshio et al. 1993; Cousin et al. 1996; Gonza´lezPen˜a et al. 2002; Guo et al. 2006) were excluded from the
dataset on compound diets. With ADP contents of purified

CL

ALDC

46.0
48.0
59.0–132.0
161.0
96.0
53.0

92.1
53.2
42.0–83.5
66.0
87.2
63.7

39.0
44.5–170.0

87.2
41.4–87.8

64.0
11.0–15.0
95.0
8.3–11.0

95.0
77.0–106.0
33.2
70.0–130.0

13.0–192.5

40.5–92.1

47.0
130.0

53.0
15.0–35.0

94.5
93.3
85.6
90.6–95.0

proteins usually higher than that of practical protein sources
(Shiau et al. 1992), the former have no application in practical diet formulation on an industrial scale. In studies evaluating feed ingredient digestibility through substitution only
the reference diet was included in the compound diet dataset
if containing practical feed ingredients. Several studies evaluating CP digestibility of compound diets used three, four
and five replicates (Fenucci et al. 1982; Smith et al. 1985;
Jones & De Silva 1997b), whereas the study of Pavasovic
et al. (2006) has 10 replicates per treatment. Martı´ nezPalacios et al. (2001) used both siphoning and settling
(modified Guelph system) to collect faeces for determination
of CP and CL digestibility.
All values were converted to a DM basis. Although DM
loss of diets has been determined in several studies (Fenucci
et al. 1982; Taechanuruk & Stickney 1982; Brown et al. 1986;
Reigh et al. 1990; Brunson et al. 1997; Sudaryono et al.
1999a,b; Cruz-Sua´rez et al. 2001, 2007; Herna´ndez et al.
2008), uncorrected digestibility estimates were used in datasets.

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd


Table 4 Frequency and level (range) of inclusion of major protein

supplying ingredients (protein database) and lipid sources (lipid
database) in compound diets

Ingredient
Protein database
Animal protein ingredients
Crabmeal
Crayfish meal
Crustacean meal
Fish meal
Fish solubles
Krill hydrolysate
Krill meal
Lobster meal
Meat meal
Mussel meal
Scallop meal
Shrimp meal
Snail meal
Squid meal
Plant protein ingredients
Cowpeas
Kelp meal
Lupin
Lupin concentrate
Maize gluten
Peanut meal
Soybean concentrate
Soybean lechitin
Soybean meal

Wheat gluten
Lipid database
Lipid sources
Coconut
Cod liver
Fish
Palm
Peanut
None

Table 5 Description of databases
Number of studies
Ingredients

Inclusion
frequency
(% of diets)

Inclusion
level (g kg)1)

2.2
1.1
6.5
93.5
21.6
1.1
1.1
0.5
4.9

6.5
0.5
60.5
0.5
51.4

20.0–147.0
100.0–359.1
50.0–51.0
32.0–778.0
20.0
60.0
330.0
83.0
100.0–378.0
10.0–50.0
130.0
29.0–729.0
304.9
5.0–127.0

3.8
4.3
4.3
0.5
3.2
5.9
22.2
14.6
39.5

17.3

150–295.0
33.9–59.4
100.0–705.0
240.0
32.0–37.7
257.6–290.0
2.5–124.0
4.2–60.8
43.8–895.9
22.0–160.0

3.1
34.4
45.3
3.1
3.1
10.9

50.0–70.0
3.0–30.0
8.2–70.0
50.0–70.0
50.0–70.0

ADP and DL contents (g kg)1 DM) were predicted from
dietary CP and CL contents (g kg)1 DM) according to Sales
(2008, 2009), respectively:
Feed ingredients

ADP = )10.0731 + 0.8942 CP
DL = )1.5824 + 0.8654 CL
Compound diets
ADP = )51.4001 + 0.9872 CP
DL = )2.7303 + 0.9123 CL
Simple linear regression analysis, conducted as described
by Sales (2008, 2009) with the software STATISTICA (data
analysis software system, Version 7.1; StatSoft, Tulsa, OK,

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd

Compound diets

Characteristic

Protein

Lipid

Protein

Lipid

Studies
Number of means
Number of species
Replicates
Two

Three
Four
Five
Marker
Chromic oxide
Total collection
Yttrium oxide
Faeces collection
Balloon
Filtering net or
scooping
Settling column
Siphoning
Tweezer
Not indicated
Protein determination
Dumas (Ebeling 1968)
Kjeldahl (AOAC 1990)
Lowry method
(Lowry et al. 1951)
Not indicated
Lipid determination
Chloroform/metanol
(Folch et al. 1957;
Bligh & Dryer 1959)
Ether (AOAC 1990)
Not indicated

17
91

8

5
31
5

32
185
11

11
64
3

8
5
3

3
1
1

1
17
6
4

13
2
2

4

1
2

1

11
2
1

3
1

1

9
1

28
1
3

8

1
4

1

4
19
1
2

1
7
1

1
11

1
20
1

5

10

3

4

5

1

3

3

USA), was used to evaluate the relation between predicted
(y) and observed (x) values.
A further measurement of the error of predicted relative to
observed values (Theil 1966) was done by calculation of the
mean square prediction error (MSPE):
MSPE ¼

n
X

ðOi À Pi Þ2 =n

i¼1

where n is the number of experimental observations, and Oi
and Pi the observed and predicted values, respectively.
The accuracy of prediction for different equations was
evaluated by the mean prediction error (MPE):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
MSPE
MPE ¼
O
where O is the mean of the observed values.


2

ECT ¼ ðX P À X O Þ

ER ¼ ðSP À r  SO Þ2

ED ¼ ð1 À r2 Þ Â SO 2
with X P and X O the mean predicted and observed values,
respectively, SP and SO the standard deviations of the predicted and observed values, respectively, and r the correlation
coefficient between predicted and observed values.

Although differences, most often contradictory, in CP
digestibility between plant and animal protein sources have
been reported with crustaceans (Brown et al. 1986; Reigh
et al. 1990; Ahamad Ali 1992; Catacutan et al. 2003;
Campan˜a Torres et al. 2006b), no improvement of the
accuracy of prediction equations for ADP or DL contents
was found by Sales (2008, 2009) when distinguishing between
dietary plant and animal feed ingredients. Due to this, and
the limited number of values available, no separation
according to type of feed ingredient was done.
Notwithstanding an intercept and slope different from 0
and 1, respectively, 92% of the variation in ADP content
could be explained by a linear regression model (Fig. 1).
Limited dispersion of points will cause small standard errors
and high values for test statistics calculated for the intercept
and slope. This will result in values that are likely to be
significant from zero and one, respectively (Mitchell 1997).
However, the latter author also stated that the variation
explained by the regression (R2) is of no relevance to validation, seen that there is no aim to predict from the fitted
line.
In contrast, deviations, calculated as prediction minus
observation, will give an indication of how far the model fails

Predicted apparent digestible protein (g kg–1 DM)

The MSPE can be differentiated (Benchaar et al. 1998)
into: (1) error in central tendency (ECT) as measurement of
the deviation of the mean of predicted values from the mean
of observed values, (2) error due to regression (ER) presenting the difference of the least squares regression coefficient from one, and (3) error due to disturbance (ED)
illustrating the variation in observed values that is not
accommodated by a least squares regression of observed on
predicted values.

900

y = 40.2352* + 0.9123*x
R 2 = 0.9213, RMSE = 54.1245

800
700
600
500
400
300
200
100
0

0

100 200 300 400 500 600 700 800 900 1000
Determined apparent digestible protein (g kg–1 DM)

Figure 1 Linear relationship between predicted and observed
apparent digestible protein values for feed ingredients (n = 91).
*
Different (P < 0.05) from 0 for intercept and 1 for slope.

to simulate observed values (Mitchell 1997). This is accommodated in MPE analysis, which account for deviations of
predicted values from observed values caused by mean bias,
linear bias and random variation (Oldick et al. 1999). A
MPE of 0.1315 derived with the ADP equation (Table 6) is
comparable to a value of 0.1064 obtained for this measurement when evaluating the prediction equation with independent studies on fish species (Sales 2008). With most of the
MSPE attributed to the ED (>99%), the error was caused by
a failure to predict the pattern of fluctuations across observed
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
values. With a mean overestimation ( MSPE) of 56.4118 g
kg)1 DM (Table 6), overestimations at both ends of the data
range have been found, with neutral or underestimated values in the middle (Fig. 1).
Predicted DL values showed a strong linear relationship
(R2 value above 0.8000) with observed values, with an
intercept and slope not significantly different from 0 and 1,
respectively (Fig. 2). However, a MPE of 0.4500 (Table 6)
presented evidence of a high overestimation of observed
values by the prediction equation, with a relative high proportion of the MSPE assigned to deviation from the regression slope (ER). Due to a lack of values from independent
studies the DL prediction equation has not been validated to
observed values with fish species (Sales 2009).

Similar to tendencies found with feed ingredients, predicted
ADP values for compound diets were highly related to
observed values (Fig. 3), presented an overestimation of

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd


Table 6 Mean prediction errors (MPEs) and components of the mean square prediction error (MSPE) between predicted and observed values
(g kg)1 dry matter)
Proportion of MSPE

Ingredients
Protein
Lipid
Compound diets
Protein
Lipid

MPE

Bias1

Error in central
tendency

Error due to
regression

Error due to
disturbance

56.4118

20.5422

0.1315
0.4500

2.6075
8.1650

0.0021
0.1580

0.0011
0.3496

0.9967
0.4924

30.8516
14.0427

0.1002
0.2372

3.3632
8.3252

0.0119
0.3515

0.1274

0.0019

0.8607
0.6466

Predicted, observed.

Predicted apparent digestible lipid (g kg–1 DM)

180

140

y = 1.6331 + 1.1431x

Predicted apparent digestible lipid (g kg–1 DM)

1

pffiffiffiffiffiffiffiffiffiffiffiffiffi
MSPE

R 2 = 0.8117, RMSE = 18.8371

160
140
120
100
80
60

40
20
0

0

20

40

60

80

100

120
–1

Determined apparent digestible lipid (g kg

140

Predicted apparent digestible protein (g kg–1 DM)

y = 16.5420 + 0.9572x
R 2 = 0.8047, RMSE = 30.7087

500

400

300

200

100

0
0

100

200

300

400

500

120

100

80

60

40

20

0
0

20

40

60

80

100

120

Determined apparent digestible lipid (g kg–1 DM)

DM)

Figure 2 Linear relationship between predicted and observed
apparent digestible lipid values for feed ingredients (n = 31).

600

y = 25.1376* + 0.7160*x
R 2 = 0.6907, RMSE = 9.8839

600

Determined apparent digestible protein (g kg–1 DM)

Figure 3 Linear relationship between predicted and observed
apparent digestible protein values for compound diets (n = 185).

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd

Figure 4 Linear relationship between predicted and observed
apparent digestible lipid values for compound diets (n = 64).
*
Different (P < 0.05) from 0 for intercept and 1 for slope.

observed values, and have a relatively low MPE value with
the ED contributing to most to the MSPE (Table 6).
However, the linear relationship between predicted and
observed DL values was characterized by an R2 value of less
than 0.7000 (Fig. 4), and a MPE value of almost 0.2400
(Table 6). With 35% of the MSPE caused by deviation of the
mean predicted value from the mean of observed values,
elimination of a portion of the error by a mean correction
factor would be possible.
As also valid with feed ingredients, the inadequacy of
equations to accurately predict DL values in compound diets
for crustacean species could partly be attributed to a limited
number of independent values and studies available, causing
bias in the evaluation. A further contributor could have been

the extrapolation of equations outside ranges used for
development, although this had no effect on the prediction of


ADP content. However, Ishikawa et al. (1997) stated that the
digestion and absorption of dietary lipids might be different
between crustaceans and fish. Unlike vertebrates, crustaceans
are unable to synthesize cholesterol (Kanazawa et al. 1971)
and bile acids from acetate and cholesterol (Holwerda &
Vonk 1973), assuming that the mechanism for assimilation of
dietary lipids is different from that found in other animals
(Teshima & Kanazawa 1983). Furthermore, assimilation of
free fatty acids is more effective than that of triacylglycerols
(Glencross et al. 1997).

This study presents evidence that the ADP contents of a wide
variety of feed ingredients and compound diets could be
predicted in crustacean species, reared under a range of different dietary, environmental and physiological conditions,
with a high degree of accuracy from dietary CP contents with
the use of a linear prediction equations developed with fish
species. This permits easy and rapid obtainable results, which
could be used in practical diet formulation for crustacean
species, and will eliminated the use of lengthy, tedious and
troublesome digestibility experiments that could be subjected
to considerable error. However, prediction equations for DL
contents established with fish species were found to be
unsuitable for crustaceans, probably related to differences in
the mechanisms of lipid digestibility between fish and crustaceans.

This study was financial supported by Research Plan No.

MSM 6007665809 of the University of South Bohemia Cˇeske´
Budeˇjovice, Research Institute of Fish Culture and Hydrobiology in Vodnˇany, Czech Republic.

Ahamad Ali, S. (1992) Evaluation of some animal and plant protein
sources in the diet of the shrimp Penaeus indicus. Asian Fish. Sci.,
5, 277–289.
Akiyama, D.M., Coelho, S.R., Lawrence, A.L. & Robinson, E.H.
(1989) Apparent digestibility of feedstuffs by the marine shrimp
Penaeus vannamei Boone. Nippon Suisan Gakk., 55, 91–98.
AOAC (1990) Official Methods of Analysis, 15th edn. Association of
Official Analytical Chemists, Arlington, VA, USA.
Ashmore, S.B., Stanley, R.W., Moore, L.B. & Malecha, S.R. (1985)
Effect on growth and apparent digestibility of diets varying in
grain source and protein level in Macrobrachium rosenbergii.
J. World Maricult. Soc., 16, 205–210.
Bautista-Teruel, M.N., Eusebio, P.S. & Welsh, T.P. (2003) Utilization of feed pea, Pisum sativum, meal as a protein source in

practical diets for juvenile tiger shrimp, Penaeus monodon. Aquaculture, 225, 121–131.
Benchaar, C., Rivest, J., Pomar, C. & Chiquette, J. (1998) Prediction
of methane production from dairy cows using existing mechanistic
models and regression equations. J. Anim. Sci., 76, 617–627.
Bligh, E.G. & Dryer, W.J. (1959) A rapid method of total lipid
extraction and purification. Can. J. Biochem. Physiol., 37, 911–917.
Bordner, C.E., DÕAbramo, R. & Conklin, D.E. (1983) Assimilation
of nutrients by cultured hybrid lobsters (Homarus sp.) fed experimental diets. J. World Maricult. Soc., 14, 11–24.
Brown, P.B., Williams, C.D., Robinson, E.H., Akiyama, D.M. &
Lawrence, A.L. (1986) Evaluation of methods for determining in
vivo digestion coefficients for adult red swamp crayfish Procambarus clarkii. J. World Maricult. Soc., 17, 19–24.
Brunson, J.F., Romaire, R.P. & Reigh, R.C. (1997) Apparent
digestibility of selected ingredients in diets for white. shrimp

Penaeus setiferus L. Aquac. Nutr., 3, 9–16.
Cabanillas-Beltra´n, H., Ponce-Palafox, J.T., Martı´ nez-Palacios,
C.A., Cha´vez-Sa´nchez, M.C. & Ross, L.G. (2001) Comparison of
the digestibility of diets based on fish meal and soybean meal in
Litopenaeus vannamei Boone 1931, using different temperatures
and salinities for culture. Cienc. Mar., 27, 577–593.
Campan˜a Torres, A., Martı´ nez Co´rdova, L.R., Villarreal
Colmenares, H. & Civera Cerecedo, R. (2005) In vivo dry matter
and protein digestibility of three plant-derived and four animalderived feedstuffs and diets for juvenile Australian redclaw,
Cherax quadricarinatus. Aquaculture, 250, 748–754.
Campan˜a Torres, A., Martı´ nez Co´rdova, L.R., Villarreal
Colmenares, H. & Civera Cerecedo, R. (2006a) Carbohydrate and
lipid digestibility of animal and vegetal ingredients and diets for
juvenile Australian redclaw crayfish, Cherax quadricarinatus.
Aquac. Nutr., 12, 103–109.
Campan˜a Torres, A., Martı´ nez Co´rdova, L.R., Villarreal Colmenares, H. & Civera Cerecedo, R. (2006b) Apparent dry matter and
protein digestibility of vegetal and animal ingredients and diets for
pre-adult Australian redclaw crayfish Cherax quadricarinatus (von
Martens 1858). Hidrobiolo´gica, 16, 103–106.
Catacutan, M.R. (1991) Apparent digestibility of diets with various
carbohydrate levels and the growth response of Penaeus monodon.
Aquaculture, 95, 89–96.
Catacutan, M.R., Eusebio, P.S. & Teshima, S.-I. (2003) Apparent
digestibility of selected feedstuffs by mud crab, Scylla serrata.
Aquaculture, 216, 253–261.
Cho, C.Y., Slinger, S.J. & Bayley, H.S. (1982) Bioenergetics of
salmonid fishes: energy intake, expenditure and productivity.
Comp. Biochem. Physiol., 73B, 25–41.
Colvin, P.M. (1976) Nutritional studies on penaeid prawns: Protein
requirements in compounded diets for juvenile Penaeus indicus

(Milne Edwards). Aquaculture, 7, 315–326.
Co´rdova-Murueta, J.H. & Garcı´ a-Carren˜o, F.L. (2002) Nutritive
value of squid and hydrolyzed protein supplement in shrimp feed.
Aquaculture, 210, 371–384.
Cousin, M., Cuzon, G., Guillaume, J. & AQUACOP (1996)
Digestibility of starch in Penaeus vannamei: in vivo and in vitro
study on eight samples of various origin. Aquaculture, 140, 361–
372.
Cruz-Sua´rez, L.E., Ricque-Marie, D., Tapia-Salazar, M., McCallum,
I.M. & Hickling, D. (2001) Assessment of differently processed
feed pea (Pisum sativum) meals and canola meal (Brassica sp.) in
diets for blue shrimp (Litopenaeus stylirostris). Aquaculture, 196,
87–104.
Cruz-Sua´rez, L.E., Nieto-Lo´pez, M., Guajardo-Barbosa, C., TapiaSalazar, M., Scholz, U. & Ricque-Marie, D. (2007) Replacement

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd


of fish meal with poultry by-product meal in practical diets for
Litopenaeus vannamei, and digestibility of the tested ingredients
and diets. Aquaculture, 272, 466–476.
Davis, D.A., Arnold, C.R. & McCallum, I. (2002) Nutritional value
of feed peas (Pisum sativum) in practical diet formulations for
Litopenaeus vannamei. Aquac. Nutr., 8, 87–94.
Divakaran, S., Forster, I.P. & Velasco, M. (2004) Limitations on the
use of shrimp Litopenaeus vannamei midgut gland extract for the
measurement of in vitro protein digestibility. Aquaculture, 239,
323–329.

Ebeling, M.E. (1968) The Dumas method for nitrogen in feeds.
J. Assoc. Off. Anal. Chem., 51, 766–770.
Eusebio, P.S. (1991) Effect of dehulling on the nutritive value of
some leguminous seeds as protein sources for tiger prawn, Penaeus
monodon, juveniles. Aquaculture, 99, 297–308.
Ezquerra, J.M., Garcı´ a-Carren˜o, F.L., Civera, R. & Haard, N.F.
(1997) pH-Stat method to predict digestibility in white shrimp
(Penaeus vannamei). Aquaculture, 157, 251–262.
Fenucci, J.L., Casal de Fenucci, A., Lawrence, A.L. & Zein-Eldin,
Z.P. (1982) The assimilation of protein and carbohydrate from
prepared diets by the shrimp, Penaeus stylirostris. J. World
Maricult. Soc., 13, 134–145.
Folch, J., Lees, M. & Sloane-Stanley, G.H. (1957) A simple method
for the isolation and purification of total lipids from animal tissues. J. Biol. Chem., 226, 497–509.
Forster, I. (1999) A note on the method of calculating digestibility
coefficients of nutrients provided by single ingredients to feeds of
aquatic animals. Aquacult. Nutr., 5, 143–145.
Forster, J.R.M. & Gabbott, P.A. (1971) The assimilation of nutrients
from compounded diets by the prawns Palaemon serratus and
Pandalus platyceros. J. Mar. Biol. Assoc. UK, 51, 943–961.
Forster, I.P., Dominy, W., Obaldo, L. & Tacon, A.G.J. (2003)
Rendered meat and bone meals as ingredients of diets for shrimp
Litopenaeus vannamei (Boone 1931). Aquaculture, 219, 655–670.
Glencross, B.D., Smith, D.M. & Williams, K.C. (1997) Effect of
dietary phospholipids on digestion of neutral lipid by the prawn
Penaeus monodon. J. World Aquac. Soc., 29, 365–369.
Gonza´lez-Pen˜a, M. del C., Anderson, A.J., Smith, D.M. & Moreira,
G.S. (2002) Effect of dietary cellulose on digestion in the prawn
Macrobrachium rosenbergii. Aquaculture, 211, 291–303.
Goytortu´a-Bores, E., Civera-Cerecedo, R., Rocha-Meza, S. &

Green-Yee, A. (2006) Partial replacement of red crab (Pleuroncodes planipes) meal for fish meal in practical diets for the white
shrimp Litopenaeus vannamei. Effects on growth and in vivo
digestibility. Aquaculture, 256, 414–422.
Guo, R., Liu, Y.-J., Tian, L.-X. & Huang, J.-W. (2006) Effect of
dietary cornstarch levels on growth performance, digestibility and
microscopic structure in the white shrimp, Litopenaeus vannamei
reared in brackish water. Aquac. Nutr., 12, 83–88.
Herna´ndez, C., Olvera-Novoa, M.A., Aguilar-Vejar, K., Gonza´lezRodrı´ guez, B. & Abdo de la Parra, I. (2008) Partial replacement
of fish meal by porcine meat meal in practical diets for
Pacific white shrimp (Litopenaeus vannamei). Aquaculture, 277,
244–250.
Holwerda, D.A. & Vonk, H.J. (1973) Emulsifiers in the intestinal
juice of crustacea. Isolation and nature of surface-active substances from Astacus leptodactylus Esch. and Homarus vulgaris L.
Comp. Biochem. Physiol., 45B, 51–58.
Irvin, S.J. & Tabrett, S.J. (2005) A novel method of collecting fecal
samples from spiny lobsters. Aquaculture, 243, 269–272.
Irvin, S.J. & Williams, K.C. (2007) Apparent digestibility of selected
marine and terrestrial feed ingredients for tropical spiny lobster
Panulirus ornatus. Aquaculture, 269, 456–463.

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd

Ishikawa, M., Teshima, S.-i., Kanazawa, A. & Koshio, S. (1997)
Measurements of digestibilities of cholesterol and fatty acids using
5a-cholestane as an inert marker in the tilapia, Oreochromis
niloticus, and the freshwater prawn, Macrobrachium rosenbergii.
J. Appl. Ichthyol., 13, 31–35.
Jones, P.L. & De Silva, S.S. (1997a) Influence of differential movement of the marker chromic oxide and nutrients on digestibility

estimations in the Australian freshwater crayfish Cherax destructor. Aquaculture, 154, 323–336.
Jones, P.L. & De Silva, S.S. (1997b) Apparent nutrient digestibility of
formulated diets by the Australian freshwater crayfish Cherax
destructor Clark (Decapoda, Parastacidae). Aquac. Res., 28, 881–891.
Kanazawa, A., Tanaka, N., Teshima, S. & Kashiwada, K. (1971)
Nutritional requirements of prawn. II. Requirement for sterols.
Bull. Jpn. Soc. Sci. Fish., 37, 211–215.
Koshio, S., Teshima, S.-i., Kanazawa, A. & Watase, T. (1993) The
effect of dietary protein content on growth, digestion efficiency and
nitrogen excretion of juvenile kuruma prawns, Penaeus japonicus.
Aquaculture, 113, 101–114.
Kumaraguru Vasagam, K.P., Ramesh, S. & Balasubramanian, T.
(2005) Dietary value of different vegetable oil in black tiger shrimp
Penaeus monodon in the presence and absence of soy lecithin
supplementation: effect on growth, nutrient digestibility and body
composition. Aquaculture, 250, 317–327.
Kumaraguru Vasagam, K.P., Balasubramanian, T. & Venkatesan, R.
(2007) Apparent digestibility of differently processed grain legumes,
cow pea and mung bean in black tiger shrimp, Penaeus monodon
Fabricius and associated histological anomalies in hepatopancreas
and midgut. Anim. Feed Sci. Technol., 132, 250–266.
Lan, C.C. & Pan, B.S. (1993) In-vitro digestibility simulating the
proteolysis of feed protein in the midgut gland of grass shrimp
(Penaeus monodon). Aquaculture, 109, 59–70.
Lazo, P.B., Romaire, R.P. & Reigh, R.C. (1998) Evaluation of three
in vitro enzyme assays for estimating protein digestibility in Pacific
white shrimp Penaeus vannamei. J. World Aquacult. Soc., 29, 441–
450.
Leavitt, D.F. (1985) An evaluation of gravimetric and inert marker
techniques to measure digestibility in the American lobster.

Aquaculture, 47, 131–142.
Lee, P.G. & Lawrence, A.L. (1985) Effects of diet and size on growth,
feed digestibility and digestive enzyme activities of the marine
shrimp, Penaeus setiferus Linnaeus. J. World Aquacult. Soc., 16,
275–287.
Lee, P.G. & Lawrence, A.L. (1997) Digestibility. In: Crustacean
Nutrition, Advances in World Aquaculture, Vol. 6 (DÕAbramo,
L.R., Conklin, D.E. & Akiyama, D.M., eds), pp. 194–260, World
Aquaculture Society, Louisiana, USA.
Lemos, D., Navarrete del Toro, A., Co´rdova-Murueta, J.H. &
Garcia-Carren˜o, F.L. (2004) Testing feeds and feed ingredients for
juvenile pink shrimp Farfantepenaeus paulensis: in vitro determination of protein digestibility and proteinase inhibition. Aquaculture, 239, 307–321.
Lin, H.Z., Guo, Z., Yang, Y., Zheng, W. & Li, Z.J. (2004) Effect of
dietary probiotics on apparent digestibility coefficients of nutrients
of white shrimp Litopenaeus vannamei Boone. Aquac. Res., 35,
1441–1447.
Lin, H.-Z., Li, Z.-J., Chen, Y.-Q., Zheng, W.-H. & Yang, K. (2006)
Effect of dietary traditional Chinese medicines on apparent
digestibility coefficients of nutrients for white shrimp Litopenaeus
vannamei Boone. Aquaculture, 253, 495–501.
Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J. (1951)
Protein measurement with the Folin-phenol reagents. J. Biol.
Chem., 193, 265–275.


Martı´ nez-Palacios, C.A., Cabanillas-Beltra´n, H., Ponce-Palafox, J.,
Cha´vez Sa´nchez, Ma. C. & Ross, L.G. (2001) A modified chamber
designed for estimation of digestibility in shrimp. N. Am. J.
Aquacult., 63, 252–255.
Merican, Z.O. & Shim, K.F. (1995) Apparent digestibility of lipid

and fatty acids in residual lipids of meals by adult Penaeus monodon. Aquaculture, 133, 275–286.
Mitchell, P.L. (1997) Misuse of regression for empirical validation of
models. Agr. Syst., 54, 313–326.
Mu, Y.Y., Lam, T.J., Guo, J.Y. & Shim, K.F. (2000) Protein
digestibility and amino acid availability of several protein sources
for juvenile Chinese hairy crab Eriocheir sinensis H. MilneEdwards (Decapoda, Grapsidae). Aquac. Res., 31, 757–765.
Newman, M.W., Lutz, P.L. & Snedaker, S.C. (1982) Temperature
effects on feed ingestion and assimilation efficiency of nutrients by
the Malaysian prawn, Macrobrachium rosenbergii (de Man).
J. World Maricult. Soc., 13, 95–103.
Oldick, B.S., Firkins, J.L. & St-Pierre, N.R. (1999) Estimation of
microbial nitrogen flow to the duodenum of cattle based on dry
matter intake and diet composition. J. Dairy Sci., 82, 1497–1511.
Ostrowski-Meissner, H.T., LeaMaster, B.R., Duerr, E.O. & Walsh,
W.A. (1995) Sensitivity of the Pacific white shrimp, Penaeus vannamei, to aflatoxin B1. Aquaculture, 131, 155–164.
Pavasovic, A., Richardson, N.A., Mather, P.B. & Anderson, A.J.
(2006) Influence of insoluble dietary cellulose on digestive enzyme activity, feed digestibility and survival in the red claw
crayfish, Cherax quadricarinatus (von Martens). Aquac. Res., 37,
25–32.
Pavasovic, A., Anderson, A.J., Mather, P.B. & Richardson, N.A.
(2007) Effect of a variety of animal, plant and single cell-based feed
ingredients on diet digestibility and digestive enzyme activity in
redclaw crayfish, Cherax quadricarinatus (Von Martens 1868).
Aquaculture, 272, 564–272.
Reigh, R.C, Braden, S.L. & Craig, R.J. (1990) Apparent digestibility coefficients for common feedstuffs in formulated diets for
red swamp crayfish, Procambarus clarkii. Aquaculture, 84, 321–
334.
Rivas-Vega, M.E., Goytortu´a-Bores, E., Ezquerra-Brauer, J.M.,
Salazar-Garcı´ a, M.G., Cruz-Sua´rez, L.E., Nolasco, H. & CiveraCerecedo, R. (2006) Nutritional value of cowpea (Vigna unguiculata L. Walp) meals as ingredients in diets for Pacific white
shrimp (Litopenaeus vannamei Boone). Food. Chem., 97, 41–49.

Sales, J. (2008) The use of linear regression to predict digestible
protein and available amino acid contents of feed ingredients and
diets for fish. Aquaculture, 278, 128–142.
Sales, J. (2009) Linear models to predict the digestible lipid content
of fish diets. Aquac. Nutr. DOI: 10.1111/j.1365-2095.2008.00620.x.

Shiau, S.-Y. & Chou, B.-S. (1991) Effects of dietary protein and
energy on growth performance of tiger shrimp Penaeus monodon
reared in seawater. Nippon Suisan Gakk., 57, 2271–2276.
Shiau, S.-Y. & Peng, C.-Y. (1992) Utilization of different carbohydrates at different dietary protein levels in grass prawn, Penaeus
monodon, reared in seawater. Aquaculture, 101, 241–250.
Shiau, S.-Y., Lin, S.-F. & Lu, L.-J. (1991a) Effects of different types
of wheat flour in feed for grass prawn Penaeus monodon. Nippon
Suisan Gakk., 57, 705–710.
Shiau, S.-Y., Kwok, C.-C. & Chou, B.-S. (1991b) Optimal protein
dietary level of Penaeus monodon reared in seawater and brackish
water. Nippon Suisan Gakk., 57, 711–716.
Shiau, S.-Y., Lin, K.P. & Chiou, C.L. (1992) Digestibility of different
protein sources by Penaeus monodon raised in brackish water and
in sea water. J. Appl. Aquacult., 1, 47–53.
Smith, D.M. & Tabrett, S.J. (2004) Accurate measurement of in vivo
digestibility of shrimp feeds. Aquaculture, 232, 563–580.
Smith, L.L., Lee, P.G., Lawrence, A.L. & Strawn, K. (1985) Growth
and digestibility by three sizes of Penaeus vannamei Boone: effects
of dietary protein level and protein source. Aquaculture, 46, 85–96.
Smith, D.M., Tabrett, S.J., Glencross, B.D., Irvin, S.J. & Barclay,
MC. (2007) Digestibility of lupin kernel meals in feeds for the
black tiger shrimp, Penaeus monodon. Aquaculture, 264, 353–362.
Sudaryono, A., Tsvetnenko, E. & Evans, L.H. (1996) Digestibility
studies on fisheries by-product based diets for Penaeus monodon.

Aquaculture, 143, 331–340.
Sudaryono, A., Tsvetnenko, E. & Evans, L.H. (1999a) Evaluation of
potential of lupin meal as an alternative to fish meal in juvenile
Penaeus monodon diets. Aquac. Nutr., 5, 277–285.
Sudaryono, A., Tsvetnenko, E., Hutabarat, J., Supriharyono &
Evans, L.H. (1999b) Lupin ingredients in shrimp (Penaeus
monodon) diets: influence of lupin species and types of meals.
Aquaculture, 171, 121–133.
Taechanuruk, S. & Stickney, R.R. (1982) Effects of feeding rate and
feeding frequency on protein digestibility in the freshwater shrimp
(Macrobrachium rosenbergii). J. World Maricult. Soc., 13, 63–72.
Teshima, S.-i. & Kanazawa, A. (1983) Digestibility of dietary lipids
and cholesterol in the prawn. Bull. Jpn. Soc. Sci. Fish., 49, 963–
966.
Theil, H. (1966) Applied Economic Forecasting. North-Holland
Publishing Company, Amsterdam, The Netherlands.
Tuan, V., Anderson, A., Luong-van, J., Shelley, C. & Allan, G.
(2006) Apparent digestibility of some nutrient sources by juvenile
mud crab, Scylla serrata (Forskal 1775). Aquac. Res., 37, 359–365.
Ward, L.R., Carter, C.G., Crear, B.J. & Smith, D.M. (2003) Optimal
dietary protein level for juvenile southern rock lobster, Jasus edwardsii, at two lipid levels. Aquaculture, 217, 483–500.

..............................................................................................

Aquaculture Nutrition 16; 559–568 Ó 2009 Blackwell Publishing Ltd


Aquaculture Nutrition
doi: 10.1111/j.1365-2095.2009.00694.x

2010 16; 569–581

..........................................................................................

Aqua Research Lab, Department of Zoology, University of Delhi, Delhi, India

Digestive enzymes of Cirrhinus mrigala (Ham.) were studied
during ontogenic development. Specific amylase activity was
detected in first feeding fish. The enzyme activity decreased
up to day-18 and then it increased with the age of fish to
reach the highest level on day-34. Protease activity was
28.61 ± 8.90 mU mg protein)1 min)1 on day-4 and
increased with the age throughout the study period. Trypsin
activity was 31.86 ± 1.12 mU mg protein)1 min)1 on day-4.
The activity decreased up to day-10 and from day-12
onwards increased up to day-26. Chymotrypsin activity
was 14.56 ± 2.74 mU mg protein)1 min)1on day-4 and
constantly increased up to day-26. A significant increase in
lipase activity was observed between days-24 and 34.
SDS-PAGE and substrate SDS-PAGE showed the diversity
of protein (17.4–127.8 kDa) and protease activity bands
(16.6–88.8 kDa) during ontogenesis. Soybean trypsin inhibitor, phenyl methyl sulphonyl fluoride, N-a-p-tosyl-L-lysine
chloromethylketone and N-tosyl-L-phenylalanine chloromethylketone inhibited the protease activity up to 79.72–97.21,
65.55–94.83, 45.41–75.31 and 40.78–64.72%, respectively.
Inhibition study in substrate SDS-PAGE revealed the
abundance of serine proteases and the presence of isoforms
of trypsin and chymotrypsin. Ethylenediamine-tetraacetate
showed 5.56–22.78% inhibition of metal ion-specific enzyme
activity.
Cirrhinus mrigala, digestive enzyme, inhibition

study, ontogenesis, protease, substrate sodium dodecyl sulfate polyacrylamide gel electrophoresis

KEY WORDS:

Received 18 February 2009, accepted 12 May 2009
Correspondence: Prof. Rina Chakrabarti, Aqua Research Lab, Department
of Zoology, University of Delhi, Delhi 110007, India. E-mail:


..............................................................................................

Ó 2009 Blackwell Publishing Ltd

The onset of digestive system and its physiological consequence in the fish larvae seems to be a key parameter to
understand the basic aspects of larval nutrition. Digestive
process of fish displays ontogenic patterns that are related to
digestive system morphogenesis and appear to reflect evolutionary adaptive measures to specific diets and changing
nutritional requirement (Buddington 1985). With differentiation of digestive tract, the fish change slowly to their speciesspecific feeding habits, a development that also includes the
adaptation of digestive enzymes (Kumar & Chakrabarti
1998). Enzymes involved in the digestion of food are of
particular interest because of their intimate participation in
the hydrolysis of nutrients necessary for adequate growth
and survival. A detailed knowledge of developmental changes occurring during the early-life stages of fish is essential
for the design of adequate larval rearing and feeding strategies and for formulation of dry diets (Verreth & Segner
1995). The knowledge of temporal appearance of key
enzymes in the gut of cultivable species in aquaculture is
essential to understand age-specific formulation of feeds
that contributes to rapid and efficient growth rates
(Tengjaroenkul et al. 2002). Besides this, comprehensive
analysis of enzymes associated with digestion and assimilation in early development of larvae is essential for identifying

the limiting factors during larval rearing and feeding
practices.
It is stated that the problems encountered in feeding of
formulated feed to fish larvae probably do not result from a
lack of general digestive capability but whether they can
adapt their digestive secretions to different diets. Various
studies revealed that fish larvae could modulate their enzyme
secretion in response to nature (Albertini-Berhaut 1978;
Hofer & Nasiruddin 1985) and quantity and composition of
food ingested (Cousin et al. 1987; Segner et al. 1989;


Zambonino Infante & Cahu 1994). The development of larva
to a fingerling relies on a proper development of digestive
functions during larval life, and the maturation of digestive
tract can be altered by diet composition (Ma et al. 2005).
Feeding must be initiated on digestible diets before or soon
after depletion of the endogenous energy sources, yolk and
oil for better survival of larvae (Kim et al. 2001). The
digestion of food in some marine fish larvae depends on its
own produced enzymes rather than those supplied by exogenous sources (Moyano et al. 1996). During early development, lack of the true stomach in fish exhibits regional
differentiation of the gastro-intestinal tract and pancreatic
digestive enzyme secretion (Tanaka 1971; Stroband & Kroon
1981). Therefore, considerable attention has been paid to
evaluate the functional characteristics of the digestive enzymes that play an important role in the ontogenesis of larva.
Information on the intestinal enzyme activities during
ontogeny is essential to know the digestive efficiency of the
cultured species at particular stages of development. This will
ensure that the fish culturist develops proper feeding strategies for a particular species during early ontogenesis. Biochemical studies of the digestive enzyme classes and their
changes during the fish ontogenesis can give valuable information on stomach functionality and can be very useful for

assessing the optimal moment for weaning and also for
developing appropriate feeding strategies for fish larvae
(Yetty et al. 2004). The knowledge and characterization of
the digestive enzymes will also help for a better understanding of their course of action in the digestive physiology
of the cultured organisms.
Indian major carp Cirrhinus mrigala (Ham.) which is
commonly known as mrigal, is an important commercial fish
in India. Mrigal is a stomachless fish and detritivore (in adult
stage). There is no data available on the pattern and
appearance of digestive enzymes during ontogenesis of mrigal. The present study aims at quantitative assay of digestive
enzymes that play a vital role in digestive physiology of fish
during early and crucial days of development. In this study,
amylase, protease, trypsin, chymotrypsin and lipase activities
were taken into account. An effort was also made to
characterize and know the class of functional proteases that
appeared during the ontogenesis of mrigal.

Four-day-old mrigal with an initial average weight of
1.06 ± 0.06 mg and length 6.4 ± 0.1 mm were procured

from a commercial fish farm. Larvae were cultured under
recirculating systems at a stocking density of 45 000 m)3.
Three replicates were used in this study. Water temperature
and pH ranged from 27.6 to 30.8 °C and 7.1 to 8.9, respectively during the study period. Dissolved oxygen was maintained above 5 mg L)1 with the help of aerators. Fish were
fed with zooplankton ad libitum consisting primarily of
Ceriodaphnia spp., Mesocyclops spp. and Brachionus spp. up
to day-14. Mixed feeding of live food and artificial diet was
started from day-14 onwards. Artificial diet (400 g kg)1
protein) was prepared by using fish meal (512.5 g kg)1),
wheat flour (367.5 g kg)1), cod liver oil (100 g kg)1) and

vitamins and mineral premix (20 g kg)1). Artificial diet was
given at the rate of 4% of body weight of fish. Duration of
the experiment was 30 days.

Larvae were collected on every other day randomly from
three tanks at 9 a.m. before morning feeding. Fish were
washed properly with the help of a sieve and were preserved
immediately at )20 °C. Fish were dissected on a glass plate
maintained at 0 °C under dissecting microscope. Gut of
individual frozen fish was taken out; 100 mg of gut tissue was
homogenized with 1 mL of chilled, distilled water. The
number of fish sampled initially for 100 mg tissue was
286 ± 5. Subsequently, the number was reduced to 20 ± 2
as the fish grew. The homogenate was centrifuged at 10 000 g
at 4 °C for 10 min and the supernatant was immediately used
for analysis. Three replicates were used for each sample.
Changes in length and weight were recorded throughout the
study period. Total soluble protein was measured by the
method of Bradford (1976). Bovine serum albumin (Sigma,
St Louis, MO, USA) was used as a standard against the
sample protein.

Amylase was assayed by measuring the increase in reducing power of buffered starch (SRL, Mumbai, India) solution with 3,5-dinitro salicylic acid (SRL, Mumbai, India) at
540 nm according to the method of Bernfeld (1955) using
UV Spectrophotometer (UV-1601, Shimadzu, Kyoto,
Japan). Specific amylase activity was expressed in terms of
mg of maltose liberated per mg of protein per hour at
37 °C.
Total protease activity was measured by using 1%
azocasein (Sigma, St Louis, MO, USA) as substrate

in 50 mM Tris–HCl (SRL, Mumbai, India), pH-7.5

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


(Garcia-Carreno 1992) and the change of absorbance was
recorded at 366 nm (UV-1601, Shimadzu, Kyoto, Japan).
The enzyme activity was recorded in mU per mg of protein
per min.
Trypsin and chymotrypsin activities were measured by
taking N-a-benzoyl D,L-arginine-p-nitoanalidine (Sigma, St
Louis, MO, USA) and Suc-Ala-Ala-Pro-Phe-p-nitroanalidine (Sigma, St Louis, MO, USA), as substrates, respectively.
Change of absorbance was recorded under kinetic mode for
3 min at 410 nm by using UV-1601 Spectrophotometer
(Erlanger et al. 1961). Activity units were expressed as
change in absorbance mg protein)1 min)1. Activity units
were calculated by the following equation:

Activity units ¼

ðAbs410 minÀ1 Þ Â 1000 Â mL of reaction mixture
:
8800 Â mg protein in reaction mixture

The activity was expressed in mU per mg of protein per
min. The molar extinction coefficient of para-nitroanalidine
is 8800.
Lipase activity was measured according to the method of

Winkler & Stuckman (1979). The principle of the assay is the
colorimetric estimation of para-nitrophenol released because
of enzymatic hydrolysis of para-nitrophenyl palmitate (Sigma,
St Louis, MO, USA) at 410 nm (UV-1601, Shimadzu,
Kyoto, Japan). One enzyme unit is defined as 1 nmol of
p-nitrophenol enzymatically released from the substrate
mL)1 min)1. The enzyme activity was expressed in mU per
mg of protein. The extinction coefficient of para-nitrophenol
is 15 000 cm)2 mg)1.

Protease inhibition assay was carried out to evaluate the
protease class by treating the enzymes with different specific
inhibitors as described by Garcia-Carreno & Haard (1993).
Phenyl methyl sulphonyl fluoride (PMSF; 100 mM in
2-propanol; Sigma, St Louis, MO, USA) and Soybean
trypsin inhibitor, (SBTI; 250 lm in distilled water; Sigma,
St Louis, MO, USA) were used as serine protease inhibitors.
N-a-p-tosyl-L-lysine chloromethylketone (TLCK; 1 mM in
HCL; Sigma, St Louis, MO, USA) and N-tosyl-L-phenylalanine chloromethylketone (TPCK; 5 mM in methanol;
Sigma, St Louis, MO, USA) were used as specific inhibitors
for trypsin and chymotrypsin, respectively. Ethylenediaminetetraacetate (EDTA; 20 mM in distilled water, Himedia,
Mumbai, India) was used for inhibiting metal ion-specific
proteases. All these inhibitors were preincubated at 1 : 1
ratio for 1 h at room temperature.

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd

Assay of protein patterns in the enzyme extracts was performed by using 12% SDS-PAGE (Laemmli 1970); 20 lg of

protein for each enzyme extract was loaded on to each well at
controlled temperature of 4 °C. After electrophoresis, the
gels were stained with Coomassie brilliant blue (SRL,
Mumbai, India) for 2 h and then destained.
The protease composition and evaluation of their classes
were carried out by substrate SDS-PAGE (Garcia-Carreno
et al. 1993). Enzyme extract containing 5 mU of activity was
loaded on to each well. After electrophoresis, gel was
immersed in 3% casein (SRL, Mumbai, India) in 50 mM
Tris–HCl, pH 7.5 for 30 min at 5 °C to allow the substrate to
diffuse into the gel. Then the gel was incubated for 60 min at
25 °C. The gel was washed, stained and destained. Clear
bands with blue background were indicated as protease
activity bands. The gels were documented in calibrated
densitometer (GS-800, Bio-Rad, CA, USA) with the help of
Quantity one – 4.5.1 software (Hercules, California, USA).
For protease class evaluation, enzyme extract having
5 mU activity were incubated with different inhibitors like
SBTI (250 lM), PMSF (100 mM), TLCK (10 mM), TPCK
(5 mM) and EDTA (20 mM) prior to loading on to the wells.
Then they were subjected to substrate SDS-PAGE. The
bands were compared with activity bands without inhibition.

Data of digestive enzyme activities were analyzed by using
one-way ANOVA and regression analysis using SPSS statistical
package (SPSS Inc., Chicago, IL, USA). Statistical significance was accepted at P < 0.05 level.

The initial total length of mrigal was 0.64 ± 0.01 cm
(Fig. 1). The total length was doubled on day-20
(1.28 ± 0.09 cm) and became three times on day-26. The fish

attained the length of 2.4 ± 0.07 cm on day-34. The initial
average weight of mrigal was 1.06 ± 0.07 mg on day-4. The
weight was more than three times higher on day-16 compared
with day-4 (Fig. 2). There was a 22-times increase in weight
on day-28 compared with day-16. Significantly (P < 0.05)
higher weight was recorded on day-34 compared with the rest
of the days. Both the length and weight exhibited exponential
relationships with increasing age of mrigal.


3

2
1.5
1
0.5

Amylase activity
(mg maltose/mg protein/h)

R 2 = 0.9457

2.5
Length in cm

1.2

y = 0.4236e0.0538x

0

y = –5E – 06x6 + 0.0003x5 – 0.0051x4 +
0.0478x3 – 0.2084x2 + 0.3179x + 0.5806
R 2 = 0.9495

1
0.8
0.6
0.4
0.2
0

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)

Figure 1 Total length of Cirrhinus mrigala during ontogenic development. Each value represents mean ± SE (n = 3).

4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)

Figure 3 Specific amylase activity showing polynomial (sixth degree)
relationship with ontogenic development of Cirrhinus mrigala. Each
value represents mean ± SE (n = 3).

300

y = 0.2733e0.2029x
R 2 = 0.9491

200

150
100
50
0

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)
Figure 2 Average weight of Cirrhinus mrigala during ontogenic
development. Each value represents mean ± SE (n = 3).

Specific amylase activity in 4-day-old mrigal was recorded as
0.722 ± 0.05 mg maltose mg protein)1 h)1. There was no
significant difference (P > 0.05) in amylase activity between
4-day and 6-day-old fish. The activity decreased (32%) significantly (P < 0.01) on day-12 (0.487 ± 0.01 mg maltose mg protein)1 h)1) compared with day-4. A 7% increase
in specific activity was found on day-16 compared with day12 (Fig. 3). Then a 9% decrease in enzyme activity was
recorded on day-18 compared with day-16. On day-22, the
specific amylase activity (0.69 ± 0.04 mg maltose mg protein)1 h)1) increased (45%) significantly (P < 0.01) compared with day-18. There was no significant (P > 0.05)
difference in specific amylase activity between days-22 and
24. From day-24 onwards, the specific amylase activity increased constantly to reach the highest level on day-34
(0.971 ± 0.06 mg maltose mg protein)1 h)1). There was a
47% increase in activity on day-34 compared with day-24.

5000
Protease activity
(mUnits/mg protein/min)

Weight in mg

250

4000

y = –0.0247x4 + 1.6595x3 – 31.598x2 +
276.17x – 755.31
R 2 = 0.9888

3000
2000
1000
0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)

Figure 4 Protease activity showing polynomial (fourth degree) relationship with ontogenic development of Cirrhinus mrigala. Each
value represents mean ± SE (n = 3).

Specific amylase activity showed a sixth degree polynomial
relationship with the increasing age of fish during early
development.

Protease activity of 4-day-old mrigal was recorded as
28.61 ± 8.90 mU mg protein)1 min)1. There was a significant (P < 0.05) increase (111%) in protease activity on day10 compared with day-8 (Fig. 4). On day-14, the enzyme
activity increased by 164% than day-12. The protease
showed 38 and 25% increase in activities on day-18 and day24, respectively, compared with their respective previous
days. The activity increased constantly up to day-28
(3897 ± 297 mU mg protein)1 min)1). There was a 2%
decrease in activity on day-30 (3822 ± 235 mU mg protein)1 min)1) compared with day-28. Highest protease
activity was recorded on day-34 (4337 ± 388 mU mg

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


protein)1 min)1). The protease activity showed a fourth
degree polynomial relationship with the increasing age of
mrigal.

activity on day-34 compared with the activity found on
day-26. Trypsin activity showed a sixth degree polynomial
trend during the development of mrigal.

Trypsin activity in day-4 mrigal was 32 ± 1 mU mg protein)1 min)1. There was a significant (P < 0.05) decrease
(59%) in the enzyme activity on day-10 (13 ± 2 mU mg
protein)1 min)1) compared with day-4. From day-12
onwards, trypsin activity increased with the age of fish and
reached the highest level on day-26 (738 ± 15 mU mg protein)1 min)1), which was 45-fold higher compared with day12 (Fig. 5). Then the activity decreased significantly
(P < 0.01) on day-28 compared with the previous day and
this trend was continued up to day-34 (515 ± 23 mU mg
protein)1 min)1). There was a 30% decrease in trypsin

Chymotrypsin activity was recorded (15 ± 3 mU mg protein)1 min)1) in first feeding mrigal (4-day-old). The enzyme
activity was 2.88-fold higher on day-12 (42 ± 5 mU mg
protein)1 min)1) compared with day-4 (Fig. 6). A 10-fold
increase in the chymotrypsin activity was recorded on day-14
(449 ± 64 mU mg protein)1 min)1) compared with day-12.
A further threefold increase in the activity was recorded
on day-16 compared with day-14. A significant increase
(P < 0.01) in activity was also observed on day-22
(3025 ± 53 mU mg protein)1 min)1) and day-24 (4185 ±

44 mU mg protein)1 min)1) compared with their respective
previous days. The activity increased up to day-26
(7140 ± 215 mU mg protein)1 min)1). Then the enzyme
activity showed a decreasing trend between day-28 and
day-34. A 21% decrease in enzyme activity was recorded on
day-34 compared with day-26. Chymotrypsin activity
showed a fifth degree polynomial relationship with the
increasing age of fish.

y = 0.009x6 – 0.4407x5 + 8.0662x
4 – 68.996x3 + 288.35x2
– 541.98x + 357.61
R 2 = 0.9744

700
600
500
400
300
200
100
0

4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)

Figure 5 Trypsin activity showing polynomial (sixth degree) relationship with ontogenic development of Cirrhinus mrigala. Each
value represents mean ± SE (n = 3).

Chymotrypsin activity

(mUnits/mg protein/min)

8000
7000

y = –0.0238x5 – 0.7063x4 + 32.413x3 –
264.07x2 + 751.83x – 587.47
R 2 = 0.9746

6000
5000
4000
3000
2000
1000

Lipase activity of 4-day-old larva was recorded as
61 ± 16 mU. The activity ranged from 48 to 53 mU between
days-6 and 10 (Fig. 7). The lipase activity was reduced by
21.7% on day-12 compared with day-4. There was no
significant (P > 0.05) difference in lipase activity between

10
9
Lipase activity (mUnits)

Trypsin activity
(mUnits/mg protein/min)

800

8

y = –3E – 05x6 + 0.0012x5 – 0.0158x4 + 0.0802x3–
0.134x2 + 0.328x + 1.9331
R 2 = 0.9396

7
6
5
4
3
2
1

0
4 6 8 10121416182022242628 30 32 34
Age of fish (DAH)

Figure 6 Chymotrypsin activity showing polynomial (fifth degree)
relationship with ontogenic development of Cirrhinus mrigala. Each
value represents mean ± SE (n = 3).

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd

0
4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
Age of fish (DAH)

Figure 7 Lipase activity showing polynomial (sixth degree) relationship with ontogenic development of Cirrhinus mrigala. Each
value represents mean ± SE (n = 3).


day-10 (50 ± 6 mU) and day-12. The activity was significantly (P < 0.05) higher on day-18 compared with day-12.
The lipase activity decreased by 25% on day-22 compared
with day-18. From day-22 onwards, the activity showed an
increasing trend and reached the highest level on day-30
(103 ± 3 mU). The activity on day-30 was significantly
(P < 0.01) higher than the activity found on day-22. The
lipase activity decreased by 2% on day-34 compared with
day-32. The enzyme activity showed a sixth degree polynomial relationship during ontogenesis.

The inhibition study of the enzyme extracts treated with
SBTI showed 79.72–97.21% inhibition in protease activity
(Table 1). Samples treated with PMSF, TLCK and TPCK
showed 65.55–94.83, 45.41–75.31 and 40.78–64.72% inhibition in protease activity, respectively, during the experimental period. Enzyme samples treated with EDTA showed only
5.56–22.78% inhibitions.

The SDS-PAGE displayed the pattern of proteins during
early development of mrigal. The molecular weight of
proteins during ontogenesis of mrigal ranged from 19.9 to
103.8 kDa (Fig. 8). Two bands with reduced resolution
(103.9 and 58.9 kDa) were visible between day-4 and day12. The number and intensities of protein bands increased
with the age of the fish. Four bands (89.0, 48.9, 44.1 and

36.4 kDa) appeared in 14-day-old mrigal. Two new bands
of protein (55.8 and 40.9 kDa) appeared on day-16. One
band of 30.7 kDa was visible on day-18. Further two

bands of 24.4 and 19.9 kDa were observed on day-22 and
day-24, respectively. Totally 11 bands with molecular
weights ranging from 19.9 to 103.9 kDa were visible
between days-24 and 34.

Substrate SDS-PAGE of protease activity showed presence
of four zymograms (78.9, 68.9, 55.8 and 44.1 kDa) between
days-4 and 8. Three new activity bands (40.9, 36.4 and
21.7 kDa) appeared on day-10 (Fig. 9). On day-14, another
activity band of 26.8 kDa was visible. Further, two activity
bands of 31.7 and 19.9 kDa appeared on day-20 and day22, respectively. Totally 10 activity bands corresponding to
different proteases were visible from day-22 onwards
(Table 2).
In the inhibition assay of proteases with SBTI, only two
zymograms of 78.9 and 68.9 kDa were visible (Fig. 10). The
remaining eight bands that appeared in activity gel without
inhibition were absent. In PMSF-treated gel, four activity
bands (44.1, 36.4, 31.7 and 21.7 kDa) were inhibited
(Fig. 11). TLCK-treated samples inhibited three activity
bands of 40.9, 36.4 and 26.8 kDa (Fig. 12). TPCK inhibited
two chymotrypsin-like bands in the gel with molecular
weight of 36.4 and 21.7 kDa (Fig. 13). There was no inhibition of activity bands of enzyme extracts treated with
EDTA.

Table 1 Percentage of inhibition of protease activity by different inhibitors during ontogenesis of Cirrhinus mrigala (n = 3)
Percentage of Inhibition
SBTI

PMSF

TLCK

TPCK

EDTA

Age of fish (DAH)

Mean

SE

Mean

SE

Mean

SE

Mean

SE

Mean

SE

4
6

8
10
12
14
16
18
20
22
24
26
28
30
32
34

83.33
87.78
79.72
87.26
91.96
96.85
95.23
96.25
94.43
97.21
95.28
96.52
95.94
93.85
94.12

93.95

3.33
4.01
3.59
1.24
4.18
1.39
1.18
1.26
1.46
0.37
0.49
0.89
0.46
0.32
1.00
0.24

65.56
69.44
88.81
84.94
87.50
91.85
94.83
90.72
92.19
92.44
93.08

93.46
75.53
89.42
85.80
89.15

8.68
2.78
2.10
5.09
2.06
2.90
1.78
3.21
1.11
0.97
0.91
0.89
15.76
1.75
2.51
0.52

55.56
61.11
59.44
56.99
55.71
71.74
56.05

75.31
69.16
49.28
47.06
49.06
52.04
45.41
63.93
49.05

5.56
5.56
7.18
5.51
5.83
4.98
1.73
4.04
2.68
3.05
0.55
8.27
2.34
3.40
4.77
1.07

63.89
48.89
62.47

51.58
40.78
54.35
48.86
64.72
45.47
55.60
52.35
46.15
44.74
43.97
56.95
47.64

4.55
1.11
4.28
8.38
10.15
4.15
5.34
5.12
1.14
2.23
7.61
7.04
2.20
1.27
3.40
5.80

18.89
22.78
11.66
11.37
5.56
11.47
9.94
10.28
10.28
11.06
14.37
7.81
10.42
6.59
12.54
11.03

1.11
3.09
3.29
3.42
1.88
2.53
4.03
1.57
2.45
1.97
2.39
0.41

1.53
1.43
1.45
3.10

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


Figure 8 SDS-PAGE of enzyme extract
of Cirrhinus mrigala during ontogenesis.
Lane M corresponds to molecular
weight marker (SDS-PAGE low-range
standard Bio-Rad). Twenty microgram
proteins of 4- to 34-day-old larvae were
loaded into each well. Gel was stained
with Coomassie brilliant blue and
washed. Arrows indicate the appearance
of bands during development.

Figure 9 Substrate SDS-PAGE of digestive enzyme extract of Cirrhinus mrigala
during ontogenesis. Lane M corresponds to molecular weight marker
(SDS-PAGE low-range standard BioRad). Five milliunits activity of different
days was loaded in to each well. Arrows
indicate the appearance of bands during
development.

Table 2 Molecular weights of protease activity bands in substrate SDS-PAGE and their inhibition with SBTI, PMSF, TLCK, TPCK and
EDTA in Cirrhinus mrigala during ontogenic development (n = 3)

Molecular Weight of Proteases (kDa)
Age of fish (DAH)
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34

78.9
+
+
+
+
+
+
+
+
+
+

+
+
+
+
+
+

68.9
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

55.8
+
+
+
+

+
+
+
+
+
+
+
+
+
+
+
+

s

I
Is
Is
Is
Is
Is
Is
Is
Is
Is
Is
Is
Is
Is
Is

Is

44.1
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

40.9

36.4

31.7

26.8

21.7

19.9

s,p

I
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p
Is,p

+
+
+
+
+
+
+
+
+

+
+
+
+

Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l

+
+
+
+
+
+
+
+
+
+
+

+
+

I,s,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c
Is,p,l,c

+
+
+
+
+
+
+
+

Is,p
Is,p
Is,p
Is,p

Is,p
Is,p
Is,p
Is,p

+
+
+
+
+
+
+
+
+
+
+

Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l
Is,l

+

+
+
+
+
+
+
+
+
+
+
+
+

Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c
Is,p,c

+
+
+

+
+
+
+

Is
Is
Is
Is
Is
Is
Is

There was no visible inhibition with EDTA.
+, Activity band of protease; Is, band inhibited with SBTI; Ip, band inhibited with PMSF; Il, band inhibited with TLCK; Ic, band inhibited with
TPCK.

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


Figure 10 Zymograms of enzyme samples of digestive extract of Cirrhinus
mrigala treated with SBTI during
ontogenesis. Lane M corresponds to
molecular weight marker (SDS-PAGE
low-range standard Bio-Rad).

Figure 11 Zymograms of enzyme samples of digestive extract of Cirrhinus
mrigala treated with PMSF during

ontogenesis. Lane M corresponds to
molecular weight marker (SDS-PAGE
low-range standard Bio-Rad).

At first feeding, the digestive tract in most species is equipped
with required enzymes for protein, lipid and carbohydrate
metabolism (Kolkovski 2001). However, the activities of
these enzymes are relatively lower in the larval stage than
that of adults (Cousin et al. 1987). In the present study,
amylase, protease, trypsin, chymotrypsin and lipase activities
were observed at the first feeding and there were variations in
the activities with the age of the fish. Qualitative alterations
of the digestive and metabolic system occurred at the transition feeding and at the transition of maturation from larva
to juvenile in turbot development (Segner et al. 1995). Each
enzyme exhibits independent appearance with temperature,
feeding habits and species-specific changes during ontogenesis (Kolkovski 2001).
Presence of a noticeable amount of amylase in several
marine larvae has been reported earlier (Cahu & Zambonino
Infante 1994; Moyano et al. 1996). In the present study, the
specific amylase activity decreased up to day-18 in detritivore
mrigal. Kawai (1972) had reported similar results on maltase
activity among carnivorous fish red sea bream Pagrus

major; black sea bream Acanthopagrus schlegelii and aku
Katsuwonus pelamis. An initial decrease in carbohydrate
metabolism was reported in rainbow trout Oncorhynchus
mykiss (Kitakimado & Tachino 1960) and in freshwater
teleosts like pike Esox lucius L; perch Perca fluviatilis L;
bream Abramis brama L; roach Rutilus rutilus L; Senegal sole
Solea senegalensis; red drum Sciaenops ocellatus (KuzÕmina

1996; Ribeiro et al. 1999; Buchet et al. 2000), which agreed
with the present findings. Tanaka (1973) suggested that dietary carbohydrates may fill the energy gap between endogenous and exogenous protein demand of fish. The decrease in
the specific activity may be related to the fixing up of the
enzyme synthesis of mrigal for its suitable feeding habits.
Kim et al. (2001) found an increase in amylase activity followed by sharp decrease in unfed larval threadfins (Polydactylus sexfilis and Caranx melampygus) just before die-off
and suggested synthesis of amylase in early ontogenesis even
in absence of food, and the carbohydrates were actively
catalyzed during the time of organogenesis. The specific
amylase in the present study increased from day-20 onwards.
Rathore et al. (2005a) stated an increase in amylase activity
from 22-days in other Indian major carp Catla catla

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


Figure 12 Zymograms of enzyme samples of digestive extract of Cirrhinus
mrigala treated with TLCK during
ontogenesis. Lane M corresponds to
molecular weight marker (SDS-PAGE
low-range standard Bio-Rad).

Figure 13 Zymograms of enzyme samples of digestive extract of Cirrhinus
mrigala treated with TPCK during
ontogenesis. Lane M corresponds to
molecular weight marker (SDS-PAGE
low-range standard Bio-Rad).

(commonly known as catla). Silver carp, big head carp

and grass carp were reported to increase in their amylase
activities from day-21 when they actively consume large
phytoplankton and zooplankton (Volkova 1999). The
increase in specific amylase activity of mrigal after initial
decrease suggests adaptation of its larvae for carbohydrate
utilization. Cahu et al. (2004) opined that regulation of
amylase is post-transcriptional in early larval stages of sea
bass Dicentrarchus labrax (till day-25) and become transcriptional towards the end of larval period. A polynomial
relationship was found between amylase activity and the age
of mrigal. This suggests the modulation of enzyme activities
during the transition phase of gut maturation and physiological adaptations to the shifting of feeding.
The results of protease activity in mrigal agreed with
Kawai & Ikeda (1973) and IlÕina & Turetskiy (1987), who
reported poor availability of proteolytic enzymes in larvae of
Sparidae family at an early developmental stage. A considerable acceleration in the protease activity was observed on
day-14 in the present study. Our earlier findings with rohu,
catla and hybrid carp showed considerable acceleration in the
protease activity during the second week after hatching

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd

(Rathore et al. 2005a; Chakrabarti et al. 2006a,b). This fact
may be attributed for adaptation to digest greater protein
content in the food as these fishes were fed with artificial diet
at the second week of age. By this time, the fish were believed
to attain structural transition of the gastro-intestinal tract,
irrespective of the food composition and was believed to
affect the protease activity in fish (DementÕeva 1976). Sharma

& Chakrabarti (1999) found a direct relationship between the
length of the digestive tract and proteolyic enzyme activity of
common carp, Cyprinus carpio. Similar observation on protein hydrolysis of predatory fish was reported by Ugolev &
KuzÕmina (1993). Ershova et al. (2004) suggested higher
amount of proteases from 12 to 18 days after hatching when
compared with carbohydrates in salmonid larvae. Mrigal
showed an increasing trend in activity with the age. This
confirmed the earlier findings with common carp (Rathore
et al. 2005b).
The process of digestive system maturation in many species
is occurred by attaining the functioning of pancreatic secretion and at the onset of brush border membrane enzymes in
intestine (Ma et al. 2005). Different trypsinogen expressions
were reported in 5-day-old winter flounder Pleuronectes


americanus (Murray et al. 2004). The appearance of chymotrypsin activity at first feeding was reported in Senegal
sole (Alliot et al. 1980). Mrigal showed a decreasing trend in
trypsin activity after day-4. This can be related to major
metabolic changes, which was earlier reported in different
marine species (Pederson 1993; Ribeiro et al. 1999). Trypsin
and chymotrypsin showed a remarkable increase in the
activity after the incorporation of artificial diets in all fish.
Cahu et al. (2004) suggested that chain length of protein in
the food modulates trypsin transcription in sea bass larvae.
The increase in the trypsin activity may be attributed to the
post-transcriptional modulation for better protein digestion.
Applebaum & Holt (2003) suggested that chymotrypsin and
trypsin activities were very much dependent on the nutritional conditions of the fish and reported an increase in
chymotrypsin activity with age in properly fed red drum
larvae. Chymotrypsin was more pronounced serine protease

when compared with trypsin in the present study. However,
the chymotrypsin activity was very low till the first week after
hatching. Applebaum & Holt (2003) assumed that activity of
chymotrypsin varies according to nutritional conditions in
red drum larvae. Kumar et al. (2005) reported that the
amount of chymotrypsin was more than two-fold higher than
the activity of trypsin in the live food Daphnia carinata. Live
food might have influenced the pattern of digestive enzymes
in the present study. Blier et al. (2002) illustrated that the
trypsin/chymotrypsin could be related to the growth-limiting
indexes in coho salmon. Both trypsin and chymotrypsin
decreased after day-26. The explanation for the decline in
enzyme activity is not because of a diminution in enzyme
synthesis but is a consequence of growth, development of
new organs and tissues, and an increase in tissue proteins.
This particular enzymatic profile is specific to early stages,
and characterizes post-hatching changes in the activity of
digestive enzymes in fish larvae (Wang et al. 2006).
Lipase was detected in the first feeding mrigal in the
present study. Major lipase in fishes appears to be nonspecific and bile salt-dependant (Gjellesvik et al. 1992).
Different workers argued over presence and absence of lipase
at the first feeding. Presence of two types of lipase is reported
in Theragra chalcogramma, one is related to yolk sac
absorption and the other is related to digestion of exogenous
lipids (Oozeki & Bailey 1995). Cousin et al. (1987) did not
find any lipase activity till day-20 in turbout Scophthalmus
maximus. However, Martinez et al. (1999) observed maximal
lipase activity on day-10, probably related to the development of exocrine pancreas. Several workers cited the utilization of lipid as the main source of energy in some fishes
(Tocher & Sargent 1984; Sargent et al. 1989).

The lipase activity appeared to have a significant increase
at fourth week after hatching in mrigal. Better growth in
larval pollock was obtained by feeding the larva with
lipid-enriched diet (Davis & Olla 1992). Mrigal showed
polynomial trend in the lipase activity during ontogenesis.
This suggested an adaptive change in carps for better utilization of dietary lipids and possible physiological changes for
lipid metabolism and preparing the gut towards adult-type
nutrition. Similar results were reported by Davis & Olla
(1992) for larval Pollock.
The SDS-PAGE analysis of proteins of digestive systems
showed the appearance of diversity of proteins during early
stages. High molecular weight bands appeared during the
early development followed by lower molecular weight
bands appearing in the later part. Similar results have been
reported in other carps (Rathore et al. 2005a; b;
Chakrabarti et al. 2006a,b). The increasing number of bands
with increasing age of fish indicated the appearance of new
proteins during ontogenic development. The appearance of
digestive enzymes at intervals indicated the functional importance of these enzymes in digestion and absorption of food
nutrients during first feeding. The intensity of bands increased
with age. This suggested the increased protein synthesis with
the age of the fish. The low protease activity during first feeding
in all the carps in test tube analysis can be well substantiated
from the intensity of zymograms of proteases in substrate
SDS-PAGE. The appearance of new activity bands at intervals
was attributed to the fate of protease activities as well. Like
proteinograms, higher molecular weight protease bands also
appeared at the early stages than low molecular weight
proteases. Perez-Casanova et al. (2006) have reported the
appearance of protease bands ranging from 21.3 to 81 kDa

and from 21.4 to 93 kDa in haddock and Atlantic cod,
respectively. The authors also evidenced the appearance of low
molecular weight proteases later during development.
The abundance of serine proteases in the gut extract of
mrigal was confirmed by inhibition of the activity in test tube
analysis. SBTI showed more inhibition for serine proteases
compared with PMSF. This fact was clear in the substrate
SDS-PAGE analysis of the inhibitors. Earlier works on
intestinal extracts of juvenile piracanjuba, Brycon orbignyanus demonstrated inhibition of serine proteases by SBTI
and PMSF (Garcia-Carreno et al. 2002). The presence of
activity bands in the SBTI- and PMSF-treated samples
showed the presence of some non-serine proteases in the
enzyme extract. This suggests the importance of serine
proteases in detritivore mrigal during early development.
Perez-Casanova et al. (2006) demonstrated the importance of
serine proteases, as it has contributed a major part in the

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


general protease of haddock and Atlantic cod larva.
According to Moyano et al. (1996), serine proteases play an
important role in protein digestion contributing up to
75–80% of the total protease activity in carnivorous gilthead
seabream, Sparus aurata. Inhibition by specific TLCK and
TPCK confirmed the presence of three and two isoforms of
trypsin-like and chymotrypsin-like enzymes, respectively, in
the gut of the mrigal during early development. More than

one isoform of trypsin and chymotrypsin in many fishes
including carps have been reported (Cohen et al. 1981;
Simpson & Haard 1984; Martinez et al. 1988; Kristiansson &
Neilsen 1992; Pivnenko 2004).
The inhibitory effect of EDTA on intestinal proteases
was reported by several workers (Alarcon et al. 1998;
Chong et al. 2002). Very less inhibition of the metal ionspecific activity was detected in the test tube assay of
EDTA-treated samples compared with other inhibitors
in the present study. Perez-Casanova et al. (2006) had
reported the presence of two metallo-proteases in haddock
and one in Atlantic cod during development. However,
there was no visible inhibition by EDTA on gut extracts of
mrigal in gel. Alarcon et al. (1998) studied the protease
inhibition with EDTA in gilthead seabream, and common
dentex, Dentex dentex and they found a lower susceptibility
of dentex proteases to different inhibitors existing in raw
materials utilized as feed ingredients. In the present study,
the low inhibition of protease activities by EDTA suggested that the presence of metal proteinases might be due
to the microbial flora in the gut of fishes. The presence of
both Gram-positive and Gram-negative microbial floral
association in the gut of different fishes had been reported
earlier (Hoshino et al. 1997; Suyanandana et al. 1998;
Irianto & Austin 2002). Micro-organisms like Bacillus sp.
play a significant role in digestion of carps C. catla and
Labeo rohita (Ghosh et al. 2002).

Amylase, protease trypsin, chymotrypsin and lipase activities
were detected at the first feeding of mrigal. Pronounced
amylase activity during the first feeding suggested the
importance of carbohydrate during the early days of

development. Protease activity increased with the age of
mrigal. Trypsin and chymotrypsin activities showed decreasing trends after day-26. Lipase activity increased considerably
after the third week of hatching. The inhibition of enzymes
with SBTI and PMSF suggested the abundance of serine
proteases during the early development. The inhibition
studies also confirmed the presence of three and two isoforms

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd

of trypsin and chymotrypsin, respectively. All these information will contribute in formulation of stage-specific diets
for carps.

The authors are thankful to Department of Science and
Technology (DST), Govt. of India for providing financial
support.

Alarcon, F.J., Diaz, M., Moyano, F.J. & Abellan, E. (1998) Characterization and functional properties of digestive proteases in two
sparids; gilthead seabream (Sparaus aurata) and common dentex
(Dentex dentex). Fish Physiol. Biochem., 19, 257–267.
Albertini-Berhaut, J. (1978) Influence du regime alimentoire surles
activities enzymotiques digestives chez Mugil capito (Teleostean,
Mugilidae). Etude. Exper. Thethys, 8, 357–363.
Alliot, E., Pastouraed, A. & Trellu, J. (1980) Evolution des activities
enzymatiques dans le tratus digestif au cours de la vie larvaire de la
sole. Variation des proteinogrammes et des zymogrammes.
Biochem. Syst. Ecol., 8, 441–445.
Applebaum, S.L. & Holt, G.J. (2003) The digestive protease, chymotrypsin, as an indicator of nutritional condition in larval red
drum (Sciaenops ocellatus). Mar. Biol., 142, 1159–1167.

Bernfeld, P. (1955). Enzymes of carbohydrate metabolism. In:
Methods in Enzymology (Colowick, S.P. & Kaplan, N.O. eds), Vol.
1, pp. 149–541. Academic Press, New York.
Blier, P.U., Lemieux, H. & Devlin, R.H. (2002) Is the growth rate of
fish set by digestive enzymes or metabolic capacity of the tissues?
Insight from transgenic coho salmon Aquaculture, 209, 397–384.
Bradford, M.M. (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle
of protein dye binding. Anal. Biochem., 72, 248–254.
Buchet, V., Zambonino Infante, J.L. & Cahu, C.L. (2000) Effect of
lipid level in a compound diet on the development of red drum
Sciaenops ocellatus larvae. Aquaculture, 184, 339–347.
Buddington, R.K. (1985) Digestive secretion of lake sturgeon, Acipenser fulvescens, during early development. J. Fish Biol., 26, 715–
723.
Cahu, C.L. & Zambonino Infante, J.L. (1994) Early weaning of sea
bass (Dicetrarchus labrax) larvae with a compound diet: effect on
digestive enzymes. Comp. Biochem. Physiol., 109, 213–222.
Cahu, C., Ronnestad, I., Grangier, V. & Zambonino Infante, J.L.
(2004) Expression and activities of pancreatic enzymes in developing sea bass larvae (Dicentrarchus labrax) in relation to intact
and hydrolyzed dietary protein; involvement of cholecystokinin.
Aquaculture, 238, 295–308.
Chakrabarti, R., Rathore, R.M. & Kumar, S. (2006a) Study of
digestive enzyme activities and partial characterization of digestive
proteases in a freshwater teleost, Labeo rohita, during early
ontogeny. Aquac. Nutr., 12, 35–43.
Chakrabarti, R., Rathore, R.M., Mittal, P. & Kumar, S. (2006b)
Functional changes in digestive enzymes and characterization of
proteases of silver carp (#) and bighead carp ($) hybrid, during
early ontogeny. Aquaculture, 253, 694–702.
Chong, A.S.C., Hasim, R., Lee, C.Y. & Ali, A.B. (2002) Partial
characterization and activities of proteases from digestive tract of

discuss fish (Symphysodon aequifasciata). Aquaculture, 203, 321–
333.
Cohen, T., Gertle, A. & Birk, J. (1981) Pancreatic proteolytic
enzymes from carp Cyprinus carpio. Purification and physical
properties of trypsin, chymotrypsin, elastase and carboxypeptidase. Comp. Biochem. Physiol., 69, 639–646.
Cousin, J.C.B., Baudin-Laurecin, F. & Gabaudan, J. (1987) Ontogeny of enzymatic activities in fed and fasting turbout, Scophthalmus maximus L. J. Fish Biol., 30, 15–33.
Davis, M.W. & Olla, B.L. (1992) Comparison of growth, behavior
and lipid concentrations of walleye Pollock Theraga chalcogramma
larvae fed lipid-enriched, lipid-deficient field-collected prey. Mar.
Ecol. Prog. Ser., 90, 23–30.
DementÕeva, M.A. (1976) Anatomical-histological features of digestive tract in the rainbow trout at the early phases of ontogenesis.
Izv. GosNIORKh., 22, 876–884.
Erlanger, B.F., Kokowski, N. & Cohen, W. (1961) The preparation
and properties of two chromogenic substances of trypsin. Arch.
Biochem. Biophys., 95, 271–278.
Ershova, T.S., Volkova, I.V. & Zaitsev, V.F. (2004) Specific features
of digestive function development in larvae of some salmonid fish.
Physiology and Development. Russian J. Dev. Biol., 33, 233–237.
Garcia-Carreno, F.L. (1992) The digestive proteases of langostilla
(Pleuroncodes planipes, Decapoda): Their partial characterization
and the effect of food on their composition. Comp. Biochem.
Physiol., 103, 575–578.
Garcia-Carreno, F.L. & Haard, N.F. (1993) Characterization of
proteinase classes in langostilla (Pleuroncodes planipes) extracts.
J. Food Biochem., 17, 97–113.
Garcia-Carreno, F.L., Dimes, L.E. & Haard, N.F. (1993) Substrategel electrophoresis for composition and molecular weight of proteinases or proteinaceous proteinase inhibitors. Anal. Biochem.,
214, 65–69.
Garcia-Carreno, F.L., Albuquerque-Cavalcanti, C., Navarrete del

Toro, M. & Zaniboni-Filho, E. (2002) Digestive proteinases of
Brycon orbinyanus (Characidae, Teleostei): characteristics and
effects of protein quality. Comp. Biochem. Phsiol., 132, 343–352.
Ghosh, K., Sen, S.K. & Ray, A.K. (2002) Characterization of bacilli
isolated from the gut of rohu, Labeo rohita, fingerlings and its
significance in digestion. J. Appl. Aquac., 12, 33–42.
Gjellesvik, D.R., Lombardo, D. & Walther, B.T. (1992) Pancreatic
bile salt dependant lipase from cod (Gadus mohrua): purification
and properties. Biochem. Biophys. Acta., 1124, 123–134.
Hofer, R. & Nasiruddin, A.N. (1985) Digestive processes during the
development of the roach, Rutilus rutilus. J. Fish Biol., 26, 683–689.
Hoshino, T., Ishizaki, K., Sakamoto, T., Kumeta, H., Yumoto, I.,
Matsuyama, H. & Ohgiya, S. (1997) Isolation of a Pseudomonas
species from fish intestine that produces a protease active at low
temperature. Lett. Appl. Microbiol., 25, 70–72.
IlÕina, I.D. & Turetskiy, V.I. (1987) Development of digestive function in fishes. Vopr Ikhtiol., 27, 835–845. (in Russian).
Irianto, A. & Austin, B. (2002) Probiotics in aquaculture. Review.
J. Fish Dis., 25, 633–642.
Kawai, S.I. (1972) Studies on the Digestive Enzymes with Special
Reference to Carbohydrases. Ph.D. Dissertation, Kyoto University,
Kyoto.
Kawai, S.I. & Ikeda, S. (1973) Studies on the digestive enzymes of
fishes. IV. Development of the digestive enzymes of carp and black
sea bream after hatching. Bull. Jpn. Soc. Sci. Fish., 39, 877–891.
Kim, B.G., Divakaran, S., Brown, C.L. & Ostrowski, A. (2001)
Comparative digestive enzyme ontogeny in two marine larval
fishes: Pacific threadfin (Polydactylus sexfilis) and bluefin trevally
(Caranx melampygus). Fish Physiol. Biochem., 24, 225–241.

Kitakimado, M. & Tachino, S. (1960) Digestive enzymes of rainbow

trout I: carbohydrases. Nippon Suisan Gakkai Shi, 26, 679–684.
[Chem. Abstr. (1961), 55, 5789].
Kolkovski, S. (2001) Digestive enzyme in fish larvae and juvenilesimplications and applications to formulated diets. Aquaculture,
200, 181–201.
Kristiansson, M. & Neilsen, H.H. (1992) Purification and characterization of chymotrypsin-like proteinase from pyloric caeca of
Oncorhynchus mykiss. Comp. Biochem. Physiol., 101, 247–253.
Kumar, S. & Chakrabarti, R. (1998) Ontogenic development of
amylase activity in three species of Indian major carps, Catla catla,
Labeo rohita and Cirrhinus mrigala, in relation to natural diet.
Asian Fish. Sci., 10, 259–263.
Kumar, S., Srivastva, A. & Chakrabarti, R. (2005) Study of digestive
proteinases and proteinase inhibitors of Daphnia carinata. Aquaculture, 243, 367–372.
KuzÕmina, V.V. (1996) Influence of age on digestive enzyme activity
in some freshwater teleosts. Aquaculture, 148, 25–37.
Laemmli, U.K. (1970) Cleavage of structural proteins during
assembly of head of the bacteriophage T4. Nature, 277, 680–685.
Ma, H., Cahu, C., Zambonino, J., Yu, H., Duan, Q., Le Gall, M. &
Mai, K. (2005) Activities of selected digestive enzymes during
larval development of large yellow croaker (Pseudosciaena crocea).
Aquaculture, 245, 239–248.
Martinez, B., Ohlson, R.L. & Serra, J.L. (1988) Purification and
characterization of two trypsin-like enzymes from Anchovy.
Comp. Biochem. Physiol., 91, 202–214.
Martinez, I., Moyano, F.J., Fernandez-Diaz, C. & Yufera, M.
(1999) Digestive enzyme activity during larval development of
the Senegal sole (Solea senegalensis). Fish Physiol. Biochem., 21,
317–323.
Moyano, F.J., Diaz, M., Alarcon, F.J. & Sarasquete, M.C. (1996)
Characterization of digestive enzyme activity during larval development of gilthead seabream (Sparus aurata). Fish Physiol. Biochem., 15, 121–130.
Murray, H.M., Perez-Casanova, J.C., Galant, J.W., Johnson, S.C. &

Douglas, S.E. (2004) Trypsinogen expression during the development of the exocrine pancreas in winter flounder (Pleuronectes
americanus). Comp. Biochem. Physiol., 138, 53–59.
Oozeki, Y. & Bailey, K.M. (1995) Ontogenic development of digestive enzyme activities in larval walleye Pollock, Theragra chalcogramma. Mar. Biol., 122, 122–186.
Pederson, B.H. (1993) Growth and mortality in young larval herring
(Clupea harengus); effects of repetitive changes in food availability.
Mar. Biol., 117, 547–550.
Perez-Casanova, J.C., Murray, H.M., Gallant, J.W., Ross, N.W.,
Douglas, S.E. & Johnson, S.C. (2006) Development of the
digestive capacity in larvae of haddock (Melanogrammus aeglefinus) and Atlantic cod (Gadus morhua). Aquaculture, 251, 377–
401.
Pivnenko, T.N. (2004) Activity, composition, and distribution of
pancreatic serine proteinases in various taxonomic groups of
hydrobionts. J. Evol. Biochem. Physiol., 40, 18–27.
Rathore, R.M., Kumar, S. & Chakrabarti, R. (2005a) Digestive
enzyme patterns and evaluation of protease classes in Catla catla
(Family: Cyprinidae) during early developmental stages. Comp.
Biochem. Physiol., 142, 98–106.
Rathore, R.M., Kumar, S. & Chakrabarti, R. (2005b) Digestive
enzyme profile of Cyprinus carpio during ontogenic development.
World Aquac., 36, 37–41.
Ribeiro, L., Zambonino Infante, J.L., Cahu, C. & Dinis, M.T. (1999)
Development of digestive enzymes in larvae of Solea senegalensis,
Kaup 1858. Aquaculture, 179, 465–473.

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd


Sargent, J.R., Henderson, R.J. & Tocher, D.R. (1989) The lipids. In:

Fish Nutrition (Havler, J.R. ed), pp. 153–218. Academic Press,
New York.
Segner, H., Rosch, R., Schmidht, H. & Von Poeppinghausen, K.J.
(1989) Digestive enzyme in larval Coregonus lavaretus L. J. Fish
Biol., 35, 249–263.
Segner, H., Stroch, V., Reinecke, M., Kloas, W. & Hanke, W. (1995)
A tabular overview of organogenesis in larval turbot, (Scophthalmus maximus). ICES J. Mar. Sci. Symp., 201, 35–39.
Sharma, J.G. & Chakrabarti, R. (1999) Larval rearing of common carp
Cyprinus carpio: a comparison between artificial and natural diets
under three stocking densities. J. World Aquac. Soc., 30, 214–219.
Simpson, B.K. & Haard, N.F. (1984) Trypsin from Greenland cod
(Gadus ogac). Isolation and comparative properties. Comp. Biochem. Physiol., 79, 613–622.
Stroband, H.W.J. & Kroon, A.G. (1981) The development of the
stomach in Clarias lazera and the intestinal absorption of protein
macromolecules. Cell Tissue Res., 215, 397–415.
Suyanandana, P., Budhaka, P., Sassanarakkit, S., Saman, P.,
Disayaboot, P., Cai, Y. & Benno, Y. (1998). New probiotic
lactobacilli and enterococci from fish intestine and their effect on
fish production. In: Proceedings of International Conference on
Asian Network on Microbial Researches, pp. 23–25, February
1998. Yogyakarta.
Tanaka, M. (1971) Studies on the structure and function of the
digestive system in teleosts larvae. III. Development of the
digestive system during post larval stage. Jpn. J. Icthyol., 18,
164–174.
Tanaka, M. (1973) Studies on the Structure and Function of the
Digestive System of Teleost Larvae. Ph.D. Dissertation, Kyoto
University, Kyoto.

..............................................................................................

Aquaculture Nutrition 16; 569–581 Ó 2009 Blackwell Publishing Ltd

Tengjaroenkul, B., Smith, B.J., Smith, S.A. & Chatreewongsin, U.
(2002) Ontogenic development of the intestinal enzymes of cultured
Nile tilapia, Oreochromis niloticus L. Aquaculture, 211, 241–251.
Tocher, D.R. & Sargent, J.R. (1984) Analyses of lipids and fatty
acids in ripe roes of some northwest European marine fish. Lipids,
19, 492–499.
Ugolev, A.M. & KuzÕmina, V.V. (1993) PishchevaritelÕnye protsessy i
adatatsii u ryb (Digestive processes and adaptation in fish).
Gidrometeoizat, St Petersburg.
Verreth, J. & Segner, H. (1995) The impact of development on
larval nutrition. In: LarviÕ95, Fish and Shellfish Larviculture
Symposium (Lavens, P., Jasper, E. & Roelants, I. eds), pp. 229.
European Aquaculture Society Special Publication No. 24.,
Gent.
Volkova, I.V. (1999) Activities of Digestive Enzymes in Plant-Eating
Fish at Early Phases of Ontogenesis. Canad. Sci. (Biol.), Dissertation, Astrakhan.
Wang, C., Xie, S., Zhu, X., Lei, W., Yang, Y. & Liu, J. (2006) Effects
of age and dietary protein level on digestive enzyme activity and
gene expression of Pelteobagrus fulvidraco larvae. Aquaculture,
254, 554–562.
Winkler, U.K. & Stuckman, M. (1979) Glycogen, hyaluronate, and
some other polysaccharides greatly enhance the formation of exolipase by Serratia marcescens. J. Bacteriol., 138, 663–670.
Yetty, N., Roshada, H., Ahyaudin, A. & Chong, A. (2004) Characterization of digestive enzymes in a carnivorous ornamental fish,
the Asian bony tongue Scleropages formosus (Osteoglossidae).
Aquaculture, 233, 305–320.
Zambonino Infante, J.L. & Cahu, C.L. (1994) Influence of diet on
pepsin and some pancreatic enzymes in sea bass (Dicentrarchus

labrax) larvae. Comp. Biochem. Physiol., 109, 209–212.


Aquaculture Nutrition
doi: 10.1111/j.1365-2095.2009.00695.x

2010 16; 582–589

..............................................................................................

The Key Laboratory of Mariculture (Education Ministry of China), Ocean University of China, Qingdao, P. R. China

A 10-week feeding trial was conducted to determine the
optimal requirement of cobia (Rachycentron canadum
Linneaus) for dietary ascorbic acid (AA). Graded levels of
L-ascorbyl-2-polyphosphate (LAPP) were supplemented in
basal diet to formulate six semi-purified diets containing 2.70
(the control diet), 8.47, 28.3, 80.6, 241 and 733 mg AA
equivalent kg)1 diet, respectively. Each diet was randomly
fed to triplicate groups of fish in flow-through plastic tanks
(300 L), and each tank was stocked with 25 fish with average
initial weight of 4.59 ± 0.36 g. Observed deficiency signs
included poor growth, higher mortality and lower feeding
rate (FR) in the fish of the control group. Fish fed the control
diet had significantly lower weight gain (WG), lower feed
efficiency ratio (FER) and lower tissue AA concentrations in
fish liver and muscle. With the increase of dietary AA, the
survival, WG, FER, hepatic and muscular AA concentrations of cobia significantly increased and then levelled off.
The dietary AA requirement of cobia was estimated to be
44.7 mg kg)1 based on WG, 53.9 mg kg)1 or 104 mg kg)1

based on either hepatic or muscular AA concentration,
respectively.
KEY WORDS: cobia, dietary ascorbic acid, feeding, nutrition,
Rachycentron canadum Linneaus, requirement

Received 14 December 2008, accepted 13 May 2009
Correspondence: K.S. Mai, The Key Laboratory of Mariculture (Education Ministry of China), Ocean University of China, Qingdao 266003, P. R.
China. E-mail:

Ascorbic acid (vitamin C, AA), which plays an important
role in the growth of most animals including fish (Al-Amoudi
et al. 1992; Lin & Shiau 2005a), has numerous biological
functions. It is essential in the synthesis of collagen, helps to

maintain various enzymes in their reduced forms, facilitates
iron absorption and participates in the biosynthesis of carnitine and norepinephrine. Most teleosts are unable to synthesize AA due to lack of L-gulonolactone oxidase that is
responsible for synthesis of AA (Wilson 1973; Dabrowski
et al. 1990; Fracalossi et al. 2001). Absence of AA leads to
many deficiency signs in fish such as poor growth, anorexia,
anaemia, spine deformity and immunity depression (Halver
et al. 1969; Lim & Lovell 1978; Dabrowski et al. 1990;
Al-Amoudi et al. 1992; Verlhac et al. 1996; GouillouCoustans et al. 1998; Ai et al. 2004). So it is usually necessary
to supplement an exogenous source of AA in fish diets.
However, a free form of AA is unstable and most of its
activity in diets will be lost during processing and storage due
to exposure to high temperature, oxygen and light, and
during feeding in water (Hilton et al. 1977; Lovell & Lim
1978; Soliman et al. 1987). Therefore, several more stable
forms of AA derivatives have been used in recent years for
more accurate assessments of the AA requirements in fish,

such as ascorbate-2-sulfate (AAS) (Halver et al. 1975;
Sandnes et al. 1990; Lin & Shiau 2005b), L-ascorbyl-2monophosphate (AMP) (Shiau & Hsu 1999; Wang et al.
2003; Lin & Shiau 2004) and L-ascorbyl-2-polyphosphate
(LAPP) (Henrique et al. 1998; Ai et al. 2004, 2006b; Lin &
Shiau 2005b). The optimal dietary AA requirements of some
fish have been determined (Al-Amoudi et al. 1992; Mustin
and Lovell 1992; Gouillou-Coustans et al. 1998; Wang et al.
2003; Ai et al. 2004, 2006b; Lin & Shiau 2005a,b). However,
due to the differences in the fish species, size, diet formulation
and experimental conditions, the dietary AA requirements
vary to some extent.
Cobia, Rachycentron canadum Linneaus, is a carnivorous
fish species and widely distributes in tropical and subtropical
waters (Ditty & Shaw 1992). Due to its quick growth and
good flesh quality (Chen 2001; Liao et al. 2004), its potential
for aquaculture has been recognized in recent years. In the
southern area of China, cobia is usually cultured in sea cages
either for domestic consumption or for export. Currently, the

..............................................................................................

Ó 2009 Blackwell Publishing Ltd


studies of cobia nutrition are mainly on the requirements of
dietary protein, lipid and amino acids, and the substitution of
fishmeal in diets (Chou et al. 2001, 2004; Wang et al. 2005;
Craig et al. 2006; Lunger et al. 2006, 2007a, b; Zhou et al.
2006, 2007). There have been three reports on the dietary
vitamin requirements of cobia (Wang et al. 2006; Zhao et al.

2008; Mai et al., 2009) so far. The first two were on AA
requirements (Wang et al. 2006; Zhao et al. 2008). However,
the reported quantitative requirements were greatly different
from each other, from 70 mg AA kg)1 diet (Zhao et al. 2008)
to 750 mg AA kg)1 diet (Wang et al. 2006). Wang et al.
(2006) used an orthogonal experimental design that
employed too high levels of AA and a wide interval (500, 750
and 1000 mg AA kg)1 diet). Both studies used either AMP or
LAPP as a vitamin C source; however, the actual AA concentrations in diets were not measured. The theoretically
calculated values based on the supplemented levels were
directly considered as the actual AA concentrations. Furthermore, the AA level from the practical dietary ingredients
was not considered either. Obviously, the AA requirements
of cobia obtained by these two studies need to be further
refined. The present study was designed to refine the optimal
AA requirement of cobia, using semi-purified diets, smaller
juveniles and longer feeding duration.

The basal semi-purified diet, using casein (vitamin-free) and
gelatin as protein sources and fish oil and soybean lecithin as
lipid sources, was formulated to contain about 470 g kg)1
crude protein and 140 g kg)1 lipid (Table 1), which was
satisfied to the growth of cobia (Chou et al. 2001). Graded
levels (0, 10, 30, 90, 270 and 810 mg AA equivalent kg)1 diet)
of LAPP (35% AA equivalent, Sunpu Biochem Company
Ltd, Beijing, China) were supplemented separately to the
basal diet at the expense of microcrystalline cellulose. The
actual levels of dietary AA, analysed by reverse-phase highperformance liquid chromatography (HPLC, HP 1100, Palo
Alto, CA, USA), were 2.70, 8.47, 28.3, 80.6, 241 and
733 mg kg)1, respectively (Table 1).
Ingredients were ground into fine powder through 246-lm

mesh. All the ingredients were thoroughly mixed with
menhaden fish oil, and water was added to produce stiff
dough. The dough was then pelleted with an experimental feed
mill and dried for 24 h in a ventilated oven at 38 °C. After
that, the diets were broken up and sieved into proper pellet size
(2.5 · 5.0 mm), and were stored at )15 °C until used.

..............................................................................................

Aquaculture Nutrition 16; 582–589 Ó 2009 Blackwell Publishing Ltd

The cobia juveniles were purchased from a commercial farm
in Sanya, Hainan, China. Prior to the feeding trial, the fish
were reared in flow-through plastic tanks (1000 L) and fed
the basal diet for one week to acclimate to the experimental
diet and conditions.
At the initiation of the feeding trial, the fish were fasted
for 24 h and weighed after being anaesthetized with eugenol (1 : 10 000, purity 99%, Shanghai Reagent, Shanghai,
China). Fish juveniles with similar size (4.59 ± 0.36 g)
were randomly distributed into 18 flow-through plastic
tanks (300 L), and each tank was stocked with 25 fish.
Each diet was randomly assigned to triplicate tanks. Fish
were hand-fed to apparent satiation twice daily (8:00 and
17:00) for 10 weeks. During the trial period, the water
temperature ranged from 28.5 to 32 °C, salinity from 24 to
26 g L)1, dissolved oxygen content was approximately
7 mg L)1 and a diurnal light: dark cycle of 12 h light : 12 h
dark provided a standardized photoperiod. At the end of
the trial, all the fish were fasted for 24 h before harvest.
Total number and mean body weight of fish in each tank

were measured.

At the termination of the feeding trial, five fish were
pooled from each tank for proximate composition analysis.
Fish whole-body and diet composition were performed by
standard methods of AOAC (2000). Dry matter was
measured by drying at 105 °C for 24 h, crude protein by
the Kjeldahl method, crude lipid after extraction with
ether by the Soxhlet method and ash by combustion at
550 °C.
Liver and muscle were dissected from five fish per tank
and pooled for analysis of AA concentration as described
by Ai et al. (2004). Briefly, a weighed portion of fish liver or
muscle (about 2 g) was suspended in cold 5% metaphosphoric acid (Shanghai Reagent) in the presence of dithiothreitol (DTT, Shanghai Reagent) and homogenized for
2 min in an ice bath, then centrifuged at 2739 g for 6 min.
For diet analysis, approximately 3–5 g grounded feed was
treated with 25 mL chloroform and 100 mL distilled water
to shake for 25 min, and then kept still for 25 min. The
upper water solution of 30 mL was used to centrifuge at
2739 g for 5 min. The above supernatants (1 mL) plus with
4 mL of 0.2 M acetic acid buffer (pH = 0.48), containing
0.2% DTT, and with 5 mg acid phosphatase was kept in
37 °C bath for 2.0 h, and then centrifuged at 2739 g for