Aquaculture, 42 (1984) 225-239 225

Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands

DIGESTIVE PROTEASES OF PENAEUS VANNAMEI BOONE:

RELATIONSHIP BETWEEN ENZYME ACTIVITY, SIZE AND DIET

PHILLIP G. LEE, LINDA L. SMITH and ADDISON L. LAWRENCE

Texas Agricultural Experiment Station, Departments of Animal Sciences and Wildlife and
Fisheries Sciences, Texas A&M University, P.O. Drawer Q, Port Aransas, TX 78373
(U.S.A.)

(Accepted 29 June 1984)

ABSTRACT

Lee, P.G., Smith, L.L. and Lawrence, A.L., 1984. Digestive proteases of Penaeus van-
namei Boone: relationship between enzyme activity, size and diet. Aquaculture, 42:
225-239.

The relationships between protein level, protein source, size, and digestive protease
enzyme activities of the marine shrimp, Penaeus vannamei Boone, were investigated
during three 30-day growth experiments. Three sizes of shrimp (4.0, 9.8, 20.8 g) were
maintained in 2650-I indoor tanks and fed ad libitum with six isocaloric diets. The
protein sources were varied by changing the animal to plant protein ratio (a/p ratio),
2:l and 1 :l, while each of these two series was composed of three diets having protein
levels of 22, 30 and 38%. The following digestive enzyme activities were detected in
crude extracts prepared from the digestive system of the shrimp; trypsin, carboxy-
peptidase A, carboxypeptidase B, acid protease and general protease. The extracts lacked
chymotryptic and peptic activities. The level of protein in the diet had a greater effect on
the enzyme total activities (activity per g wet tissue) in the large shrimp (17-30 g) than
in small shrimp (< 10 g). When the specific activities (activity per mg of protein in the
extract) of the fed shrimp were evaluated, small shrimp (< 10 g) fed the 1:l a/p ratio
diets displayed lower activities than those fed the 2:l ratio diet. The protein level in-
fluenced the enzyme activities in shrimp of all sizes while the protein source had a greater
effect on the enzyme activities in small shrimp (< 10 g). This differing proteolytic re-
sponse to protein level and source as a function of size supports the formulation of
specific diets for shrimp of different sizes, taking into consideration the changes in
digestive physiology as the shrimp grow.

INTRODUCTION

Nutritional studies conducted with shrimp have classically been confined

to empirically designed dietary trials, while investigations of the bioener-
getics and digestive physiology of the organisms have received less emphasis.
The feed digestibilities, nutrient balance and bioenergetics of marine and
freshwater shrimp are currently receiving more attention (New, 1976,198O).
All of these studies have proven useful in spite of the inherent limitations

0044-8486/84/$03.00 o 1984 Elsevier Science Publishers B.V.

226

of each. One of the major limitations is the inability to assess the specific
digestive capabilities (ability to hydrolyze protein, lipid, fiber, etc.) of an
organism, i.e., the true potential and not the expressed efficiency which
results from the overall suitability or unsuitability of the diet. If those
potentials can be assessed utilizing digestive enzyme analysis, a more optimal
diet can be designed based upon a better understanding of the physiological
abilities of the organism.
The importance of digestive enzyme analysis as a tool in the study of
nutrition lies in the description of patterns in an animal’s dietary regimen,
such as ability to specifically hydrolyze individual materials in the diet,
response to differing nutrient sources and levels, bacterial contribution to
digestion, cyclic secretion and changes in these as the animal grows and
matures. The importance of the hydrolytic activity in the digestive system
of an animal is subordinate only to the composition and form of the food
and degree of mechanical trituration. The chemical breakdown of the food
ultimately determines the types of nutrients which are available for absorp-
tion.
Until recently most investigations concerning the digestive enzymes of
crustaceans have been qualitative and focused on the comparative aspects
of digestion. As a result, several reviews (Van Weel, 1970; Gibson and
Barker, 1979; Dal1 and Moriarty, 1983) dealing with the comparative physi-
ology of digestion in Crustacea have been published, but have contained
very little quantitative information. Since crustaceans are now being evalu-
ated for commercial culture, the changes in enzyme activities during the
life cycle and adaptation to new diets are being examined quantitatively
(Trellu and Ceccaldi, 1977; Laubier-Bonichon et al., 1977; Hood and Mey-
ers, 1977; Van Wormhoudt et al., 1972, 1980; Cuzon et al., 1980; Fair et
al., 1980; Lee et al., 1980; Maugle et al., 1982a, b).
The principal objective of this research was to obtain information con-
cerning the proteolytic enzymes present and changes in activities that
occur in the digestive tract of the marine shrimp, Penaeus uannamei, in
response to diet during their life cycle. In addition, the data are compared
with other nutritional information, growth and feed digestibility, which
were simultaneously measured for the same population of shrimp (Smith
et al., 1984). The secondary objective was to use this new information for
evaluating the formulation of prepared diets used in crustacean mariculture.

METHODS

Growth experiment

Shrimp used in the three growth experiments were obtained from two
different sources. Medium shrimp (9.8 f 1.8 g) were obtained from a O.l-ha
pond of the Texas A&M Mariculture Facility located east of Baytown,
Texas. These shrimp were originally derived from Guatemalan stock and
 227

had been over-wintered in heated effluent. Small shrimp (4.0 + 0.9 g) and
large shrimp (20.8 + 1.0 g) were obtained from two O.l-ha ponds at the
Texas A&M Shrimp Mariculture Project in Corpus Christi. These two groups
of shrimp were derived from Costa Rican stock. All three groups of shrimp
were maintained on a commercial shrimp ration (approximately 24% pro-
tein, dry pellet) prior to these experiments.
Three 30day growth trials were conducted indoors using 2650-1, circular
tanks. Small and medium shrimp were stocked at a density of 6.42 shrimp/
m2 while large shrimp were stocked at 4.25 shrimp/m2, or approximately
64 200 and 42 500 shrimp/ha, respectively. All three groups were fed
twice daily (ad libitum) with the prepared diets. In addition to six prepared
diets, a control group for each experiment was included. Controls con-
sisted of shrimp from the same initial population which had been starved
for 2 weeks, a period during which digestive enzyme activities reach a
baseline value (Cuzon et al., 1980).

TABLE I
Feed ingredients included in diets used in growth experiments. Values represent percent
of total diet

Feed Diet composition (%)

ingredient 2:l Series 1:l Series
38%8 30% 22% 38% 30% 22%

Shrimp meal’ 36.00 30.90 20.60 29.40 21.40 13.40

Menhaden meal2 3.50 3.50 3.50 3.20 3.20 3.20
Squid meal’ 2.00 2.00 2.00 1.50 1.50 1.50
Fish solubles2 2.00 2.00 2.00 2.00 2.00 2.00
Rice branc 22.75 35.00 36.00 35.00 35.50 34.50
Corn starch4 12.25 14.40 25.00 7.20 17.50 29.20
Alpha soy4 3.00 3.00 0.25 6.50 4.00 2.20
Wheat gluten4 6.00 0.00 0.00 6.25 4.25 2.00
Casein (vitamin-fre,e) 3.30 0.00 0.00 0.00 0.00 0.00
Alpha cellulose“ 1.00 0.90 1.85 1.00 1.75 2.70
Vitamin fortification mixture4 2.00 2.00 2.00 2.00 2.00 2.00
AIN mineral mixture 764 1.00 1.00 1.00 1.00 1.00 1.00
Cod liver oil4 0.70 0.30 0.80 0.45 0.90 1.30
Lecithin4 1.00 1.00 1.00 1.00 1.00 1.00
Cholesterol4 0.50 0.50 0.50 0.50 0.50 0.50
Sodium hexametaphosphate6 1.00 1.00 1.00 1.00 1.00 1.00
Kelco HV alginate’ 2.00 2.50 2.50 2.00 2.50 2.50

I Blum and Bergeron, Houma, LA.

2 Sea and Sound Trout Co., Beaufort, NC.
3 Oven dried (55”C, 10 h), laboratory prepared.
4 ICN Nutritional Biochemicals, Cleveland, OH.
5 Riviana Foods, Inc., Houston, TX.
6 Calgon Water Softener, Pittsburgh, PA.
’ Kelco, San Diego, CA.
R38%, 30% and 22% are the designed percent protein levels for the two (2:l and 1: 1)
a/p ratio series.
228

Diet formulations

The six experimental isocaloric diets were prepared with practical and
technical grade feedstuffs and bound with alginate (Meyers, 1980). All
ingredients were obtained from commercial sources (Table I, footnote)
with the exception of the squid meal. Dried squid meal was prepared from
frozen squid which was first boiled, the eyes and pens removed and then
ovendried at 55°C. When necessary, the feeds were reground to pass through
a l.O-mm screen, thereby reducing preferential selection of particles by
the shrimp.
Two series of three diets (Table I) were designed with the sources of
protein varying so that one series had a 1:l animal protein to plant protein
ratio (a/p ratio) and the second a 2:l a/p ratio. Each series included three
protein levels, 38%, 30% and 22%. The proximate composition (Table II)
of the diets varied slightly from the designed composition. The levels of
lipid, fiber and gross energy were not significantly (P > 0.05) different
between the six diets, whereas the levels of ash and moisture did vary sig-
nificantly (P < 0.05). The biochemical composition of the diets was deter-
mined as follows: protein using the micro-Kjeldahl procedure (Barnes,
1959); total lipid after Bligh and Dyer (1959); and fiber, ash, moisture and
gross energy according to A.O.A.C. (1975).

TABLE II

Proximate composition’ of diets used in growth experiments

Diet2 Protein Lipid Fiber Ash Moisture Gross

(%) (%) (%) (%) (%) energy
(kcal/g)

1:1, 38 36.90 5.93 10.05 14.90 8.87 3.93

53.12 kO.49 to.11 + 0.06 +0.14 f 0.08
l:l, 30 29.50 6.00 10.03 12.80 1.71 3.98
* 1.80 f 0.00 t-o.13 to.00 to.10 + 0.02
l:l, 22 21.32 6.13 9.57 10.50 7.97 3.98
t 0.80 r0.25 to.14 f 0.06 to.15 + 0.00
2:1, 38 38.34 6.10 9.85 15.40 8.19 3.94
k2.15 to.50 kO.24 kO.74 +O.Ol to.07
2:1, 30 31.84 6.23 10.50 15.80 6.80 3.87
* 1.80 to.38 kO.37 to.10 + 0.05 f 0.01
2:1, 22 24.03 5.90 10.01 12.80 7.85 3.88
t1.16 t-O.36 * 0.05 kO.06 kO.01 to.00

Mean 6.05 10.09 13.74 8.00 3.93

to.33 to.31 + 1.96 to.72 f 0.04

’ Values represent means i- standard

deviation of four replicates.
* Two series of three diets were designed so that one series had a 1:l animal protein to
plant protein ratio (a/p ratio) and the second series a 2:l a/p ratio. Each series included
three designed protein levels of 38%, 30% and 22%.
 229

Enzyme analyses

Upon termination of the growth experiments, the digestive tracts of the

shrimp were assayed for trypsin, chymotrypsin, carboxypeptidase A, car-
boxypeptidase B, pepsin, and general and acid proteases. For these assays,
the digestive gland, stomach and midgut of each shrimp were homogenized
individually (n = 6 for small and medium shrimp and n = 5 for large shrimp)
in 10 ml of buffer using a high speed tissue homogenizer. A 0.05 M Tris
buffer, pH 7.8, containing 0.02 M CaCl, was used for the tryptic, chymo-
tryptic and carboxypeptidase A and B assays. The general protease assay
used a 0.1 M phosphate buffer, pH 7.0, with 0.01 M NaCl. Peptic and acid
protease assays used a 0.1 M acetate buffer, pH 5.4. The resulting homo-
genate was centrifuged at 4800 X g and O--5% for 60 min and the super-
natent was then used in the assays. The total protein content of the extract
was determined according to the method of Lowry et al. (1951). Data
recorded on each shrimp analyzed included general physical condition,
sex, molt stage, length, wet weight, digestive gland wet weight and the
color of the digestive gland.
Enzyme activities were measured kinetically utilizing chromogenic sub-
strates or color-producing end products. Tryptic (Enzyme Commission
Number, E.C. 3.4.4.4.) activity was assayed using the benzoyl-DL-arginine-
p-nitroanilide (BAPNA) method of Erlanger et al. (1969). The method of
Erlanger and Fidel (1964) which utilized the substrate glutaryl-l-phenyl-
alanine-p-nitroanilide (GPANA) was used in determining the activity of
chymotrypsin (E.C. 3.4.4.5.). Carboxypeptidase A (E.C. 3.4.12.2.) was
assayed with hippuryl-l-phenylalanine (BzGPA) by the method of Folk
and Schrimer (1963). The method of Folk et al. (1960) employing the
substrate hippuryl-l-arginine (BzGA) was used to determine the activity
of carboxypeptidase B (E.C. 3.4.12.3.). Peptic (E.C. 3.4.23.1.) activity was
determined utilizing the substrate N-acetyl-L-phenylalanine-L-3,5-di-iodo-
tyrosine (APAIT) as described by Rick and Fritsch (1974). The hemoglobin
method of Anson (1938) was used to assay general protease (pH 7.0) and
acid protease (pH 2.0) activities (medium shrimp were not assayed for acid
protease activity).
All enzyme activities are expressed as total activities (amount of product
producedjmin per g of wet tissue or enzyme units per g of wet tissue) and
specific activities (amount of product produced/min per mg protein or
enzyme units per mg protein). An enzyme unit equals that amount of
enzyme which causes an increase in absorbance of 1.0 optical density (OD)
units per minute in the reaction mixture. The activity of each enzyme is
reported as a population mean f the standard error for each experimental
treatment. Possible relationships between the activities and diet, size,
growth, sex, molt stage and species were analyzed statistically using multiple
regression and correlation analysis. In addition, to examine the relation-
ship of the level and source of protein on enzyme activities, a Duncan’s
Multiple Range Test (P < 0.05) was performed on each experiment.
TABLE III M
0
General protease activities

Diet’ General protease total activity2 General protease specific activity’

(PM tyrosine produced/min per g wet tissue) (PM tyrosine produced/min per mg protein)

Small Medium Large Small Medium Large

l:l, 38 145.9 ? 10.Ba 102.3 t 3.4a 106.3 t 6.Baqb 0.908 f 0.116a 1.185 + O.lOla 1.457 f 0.040a
l:l, 30 29.8 + 4.0b 79.3 * 5.5b 145.5 + 5.2aqb 0.241 + 0.057b 1.061 + 0.062a 1.524 2 O.lOga
l:l, 22 122.3 * 27.2a 86.0 ? 6.7aqb 156.4 f 20.3a 0.876 + 0.220a 0.948 + 0.064a 1.280 f 0.106a
2:1, 38 149.3 * 6.gx 101.6 * 6.9’ 118.1 ? 8.4x 1.442 f 0.08gx 1.391 Z? o.09gx 1.251 + 0.023’
2:1,30 137.6 * 17.3x 127.1 + 8.Bx 122.6 k 5.1x 1.300 2 o.154x 1.401 f o.o35x 1.306 f 0.03Bx”
2:1, 22 140.2 + 6.6x 86.9 * 4.9’ 115.0 * 6.0x 1.453 * O.O46x 1.308 + O.O85x 1.397 * O.O26x

’ Two series of three diets were designed so that one series had a 1:l animal protein to plant protein ratio (a/p ratio) and the second
series a 2:l a/p ratio. Each series included three designed protein levels of 38%, 30% and 22%.
‘The activities are reported as the population mean + standard error (n = 6). Each size and the 1:l a/p ratio and 2:l a/p ratio diets
were separated for statistical analysis. Activities with different superscripts are statistically different (P < 0.05).

TABLE IV
Trypsin activities

Diet’ Trypsin total activity’ Trypsin specific activity’

(PM p-nitroanilide produced/min per g wet tissue) (PM p-nitroanilide produced/min per mg protein)

Small Medium Large Small Medium Large

l:l, 38 9.88 2 0.61a 11.59 k 0.70a 11.02 k o.94a 0.072 + 0.005a 0.135 t 0.006a 0.102 f 0.006a
l:l, 30 8.02 + 1.34a*b 9.05 * l.OBa 10.26 f 0.63’ 0.063 t 0.005a 0.129 t O.OOBa 0.113 + 0.006a
l:l, 22 6.52 t 0.66b 8.68 * 1.12a 7.29 ? 0.52b 0.067 f 0.004a 0.171 + 0.026a 0.108 i 0.004a
2:1, 38 8.02 + 0.60’ 9.03 f 0.94v 8.46 + 1.05’ 0.097 * 0.003v 0.110 k 0.004y 0.097 * o.olox
2:1,30 9.35 * 0.91x*’ 12.75 * 1.21x 11.62 r O.66x 0.115 + 0.008x’y 0.119 + O.OIOy 0.108 -t O.OO4x
2:1, 22 10.74 r o.37x 9.88 r 1.14x’y 7.42 + 0.59’ 0.135 f O.OOgx 0.182 * 0.020x 0.109 + O.O1lx

’ Two series of three diets were designed so that one series had a 1:l animal protein to plant protein ratio (a/p ratio) and the second
series a 2:l a/p ratio. Each series included three designed protein levels of 38%, 30% and 22%.
’ The activities are reported as the population mean t standard error (n = 6). Each size and the 1 :l a/p ratio and 2:l a/p ratio diets
were separated for statistical analysis. Activities with different superscripts are statistically different (P < 0.05).
 231

RESULTS

Qualitatively, all three sizes, large (24.20 g final wt), medium (17.55 g
final wt) and small (9.88 g final wt), of the shrimp Penaeus vannamei, ex-
hibited activities on the trypsin, carboxypeptidase A and carboxypeptidase B
specific substrates. In addition, using denatured hemoglobin as a substrate,
protease activity was demonstrated at a pH of 7.0 (general protease) and a
pH of 2.0 (acid protease). Neither chymotryptic nor peptic activities could
be detected using the specific substrates. Trypsin, carboxypeptidase A
and B and general protease were measured quantitatively (Tables III-VII).
Due to significant (P < 0.05) differences in the response of the digestive
enzyme activities to size and a/p ratio, the three sizes and two ratios are
treated separately in the tables.
Total and specific activities of the general protease assay are displayed
in Table III. Small shrimp exhibited higher (P < 0.001) total activities on
the 2:l a/p ratio diets and the 38% protein, 1:l a/p ratio diet when com-
pared with medium and large shrimp. Conversely, large shrimp exhibited
significant (P < 0.001) peaks in total activity on the 30% and 22% protein,
1:l a/p ratio diets. Specific activities in small shrimp were higher (P < 0.001)
on the 2:l a/p ratio diets than on the 1:l a/p ratio diets. The total activities
were also higher but significantly higher only on the 30% 2:l a/p ratio
diet. In addition, small shrimp exhibited the lowest (P < 0.001) specific
and total activities on the 30% protein, 1:l a/p ratio diet. Medium shrimp
displayed peaks for both specific and total activities on the 30% protein,
2: 1 ratio diet and 38% protein, 1 :l a/p ratio diet although these diets were
not significantly (P > 0.05) different from the other diets.
Total activity of trypsin for the three experiments is displayed in Table
IV. The highest total activity for the large and medium shrimp fed in the
2:l a/p ratio series was for shrimp fed the 30% protein diet, while those
fed the 38% protein, 1:l a/p ratio diet exhibited the highest activity for
that series. The highest activities for small shrimp were found with the
22% protein and 38% protein diets for the 2:l and 1:l a/p ratios, respec-
tively. This difference in the activity peaks when coupled with the varied
response to the protein level of the 2:l and 1:l a/p ratio diets, resulted in
significantly different responses for size (P < 0.0001) and the interaction
of a/p ratio and protein level (P < 0.003). In addition, small shrimp ex-
hibited lower activities on all diets with the exception of the 22% protein,
2:l a/p ratio diet than did large and medium shrimp, with medium shrimp
exhibiting slightly higher activities than large shrimp.
There were no significant differences in specific activities of trypsin
(Table IV) as a function of a/p ratio or protein levels in large or medium
shrimp but small shrimp exhibited lower (P < 0.001) activities on the 1:l
a/p ratio diets. In general, shrimp also displayed higher activities on the
lower protein (22%) diets in direct contrast to the results of the total activ-
ity of trypsin.
TABLE V
Carboxypeptidase A activities

Diet’ Carboxypeptidase A total activity* Carboxypeptidase A specific activity’

(enzyme units”/g wet tissue) (enzyme units/mg protein)
Small Medium Large Small Medium Large

l:l, 38 84.92 1.7a 96.6 + 3.2a 81.4 f 6.5a 0.614 + 0.052a 1.133 * 0.062* 0.748 * O.OIOa
l:l, 30 71.9 * 12.1a*b 58.8 ? 8.2” 68.3 t 6.0’ 0.551 + 0.055* 0.826 f 0.050= 0.747 * 0.046a
l:l, 22 51.5 + 3.2b 46.4 * 8.9h 44.6 * 7.8b 0.536 + 0.032a 0.895 f 0.134a*b 0.660 + 0.099=
2:1, 38 65.7 k 6.5x 69.1 f 8.Ox*y 44.6 t 5.8” 0.799 * 0.054x 0.839 + 0.047x 0.519 * 0.074v
2:1, 30 79.0 * 6.1x 91.0 ?. 8.1x 73.7 + 4.6x 0.999 + O.O85x 0.854 + 0.080x 0.685 * O.O35x
2:1, 22 68.0 + 3.3x 59.2 * 8.9Y 31.0 * 2.ov 0.851 + O.O53x 1.084 k 0.150x 0.539 t o.031x*v

’ Two series of three diets were designed so that one series had a 1 :l animal protein to plant protein ratio (a/p ratio) and the second
series a 2:l a/p ratio. Each series included three designed protein levels of 38%, 30% and 22%.
‘The activities are reported as the population mean ? standard error (n = 6). Each size and the 1:l a/p ratio and 2:l a/p ratio diets
were separated for statistical analysis. Activities with different superscripts are statistically different (P < 0.05).
‘An enzyme unit equals an increase of 1.0 in absorbancy per minute.

TABLE VI
Carboxypeptidase B activities

Diet1 Carboxypeptidase B total activity* Carboxypeptidase B specific activity*

(enzyme units’/g wet tissue) (enzyme units/mg protein)
Small Medium Large Small Medium Large

l:l, 38 68.3 t 4.4a 72.3 k 2.5= 121.9 t 12.0a 0.500 ? 0.044a 0.845 + 0.037b 1.122 + 0.085s
1:1, 30 61.6 + 7.5a 36.2 2 4.5b 93.4 t 5.4b 0.419 ? 0.044a 0.539 * 0.08gb 1.036 i 0.084’
1:1, 22 53.2 + 4.9’ 67.2 f 6.7a 67.6 * 6.7b 0.553 + 0.02ga 1.378 t 0.266a 0.999 * 0.054a
2:1, 38 65.3 + 6.2x 73.2 t 4.8’ 88.7 + 8.6x’y 0.789 ? 0.03gx 0.904 ? 0.036’ 1.017 + o.o75x
2:1, 30 71.0 f 1.3x 101.7 k 4.0x 110.9 i 7.gx 0.874 f 0.05EX 0.961 r 0.063’ 1.038 + 0.090x
2:1, 22 71.3 * 3.4x 70.4 ? 5.6’ 70.9 t 6.4’ 0.891 + O.O47x 1.291 * o.o77x 1.025 ? O.O73x

’ Two series of three diets were designed so that one series had a 1:l animal protein to plant protein ratio (a/p ratio) and the second
series a 2:l a/p ratio. Each series included three designed protein levels of 38%. 30% and 22%.
‘The activities are reported as the population mean ? standard error (n = 6). Each size and the 1:l a/p ratio and 2:l a/p ratio diets
were separated for statistical analysis. Activities with different superscripts are statistically different (P < 0.05).
3An enzyme unit equals an increase of 1.0 in absorbancy per minute.
 233

All three sizes of shrimp exhibited the highest carboxypeptidase A total

activities (Table V) on the 38% protein, 1:l a/p ratio diet and the 30%
protein, 2:l a/p ratio diet. There was a significant (P < 0.0001) relation-
ship between activity, size, protein level and the interaction between protein
level and a/p ratio. With respect to the specific activity (Table V), the
peaks still occurred at the 30% protein, 2:l a/p ratio diet and 38% protein,
1:l a/p ratio diet with the exception of the medium shrimp which exhibited
the peak on the 22% protein, 2:l a/p ratio diet. In addition, small shrimp
exhibited lower (P < 0.001) specific activities on the 1:l a/p ratio diets
while large shrimp exhibited lower (P < 0.001) total and specific activities
on the 2:l a/p ratio diets.
Large shrimp exhibited higher total activities (P < 0.001) for carboxy-
peptidase B (Table VI) than did small shrimp. There were significant (P
< 0.0002) relationships between total activity, size, a/p ratio and the inter-
action of the a/p ratio and protein level. Peaks in activity once again oc-
curred on the 30% protein, 2:l a/p ratio diet and 38% protein, 1:l a/p ratio
diet, with the exception of the small shrimp which exhibited a peak on the
22% protein, 2:l diet. Small and medium shrimp displayed less change in
total activity in response to diet than did the large shrimp. Examining the
specific activities of carboxypeptidase B (Table VI), a size relationship
(P < 0.0003) was established since large and medium shrimp exhibited
higher specific activities on all diets than did small shrimp. In addition, the
a/p ratio was found to be related (P < 0.0005) to specific activity for small
shrimp since the 2:l a/p ratio diets produced higher specific activities than
did the 1:l a/p ratio diets.
The use of multiple regression and correlation analysis to measure the
relationships between enzyme activities and molt stage, sex, digestive gland

TABLE VII
Enzyme activities for starved shrimp’

Enzyme Small Medium Large

General protease Total activity’ 142.5 f 19.3a 86.9 * 4.lb 100.1 f 12.3a*b
Specific activity3 1.799 f o.3o7x 1.448 f 0.141x 1.048 k 0.126’
Acid protease Total activity’ 7.22 t 3.6’ NA 2.90 * 0.4a
Specific activity3 0.110 + O.O55x NA 0.030 f 0.004y
Trypsin Total activity4 3.79 k 0.63b 6.99 + 1.05a*b 9.54 f 2.0P
Specific activity’ 0.063 t 0.004y 0.109 + 0.013* 0.100 i 0.022x.y
Carboxypeptidase A Total activity” 40.07 ? 4.6b 51.86 + 4.6a,b 63.43 f 0.4a
Specific activity’ 0.745 i 0.1 7x 0.826 * 0.051x 0.665 i 0.014x
if 44.44 f 10.8b 83.90 2 6.5a
Carboxypeptidase B Total activity6 32.44 + 6.7
Specific activity7 0.528 k 0.077x 0.738 ? 0.20ax 0.881 t o.oaox

’ The activities are reported as the population mean t standard error (n = 6). Total and specific activ-
ities were separated for statistical analysis. Activities with different superscripts are statistically dif-
ferent (P < 0.05).
2 Activity expressed as pM tyrosine produced/min per g wet tissue.
’ Activity expressed as pM tyrosine produced/min per mg protein.
‘Activity expressed as pM p-nitroanilide produced/min per g wet tissue.
’ Activity expressed as HIMp-nitroanilide produced/min per mg protein.
‘Activity expressed as enzyme units/g wet tissue.
’ Activity expressed as enzyme unitsjmg protein.
234

wet weight, percent weight gain of shrimp, feed digestibility and protein
digestibility produced no statistically significant relationships. The three
starved control groups exhibited different enzyme activity patterns (Table
VII). The small starved shrimp displayed lower total enzyme activities
(P < 0.05) for trypsin, carboxypeptidase A and B compared with large
shrimp. Therefore, starvation affected the three individual protease activ-
ities more strongly in small shrimp without affecting total protease activity.
Total enzyme activities for medium shrimp fell between those for small
and large shrimp except for carboxypeptidase B total activity where the
medium shrimp shared the same level with small shrimp. The differences
in specific enzyme activities among the fed and starved shrimp were usually
not significant (P > 0.05).

DISCUSSION

Penaeus uannamei has previously been shown to exhibit tryptic, carboxy-

peptidase A and B, leucine aminopeptidase, acid protease, general protease,
amylase, chitinase, esterase, and lipase activities (Lee and Lawrence, 1982).
However, peptic and chymotryptic activities were not demonstrated. These
same enzyme activities were also reported for Penueus setifeiw and Penaeus
stylirostris (Lee and Lawrence, 1982). Gates and Travis (1969, 1973) puri-
fied and characterized the trypsin and carboxypeptidase A and B enzymes
from Penueus setiferus. They also failed to detect chymotryptic activity.
Both tryptic and chymotryptic activities, as well as general protease activ-
ity, were reported for Penueus keruthurus (Van Wormhoudt et al., 1972).
Trellu (1978) determined the activities of trypsin, chymotrypsin, and
leucine, valine and cystine aminopeptidases in Penueus juponicus. The
detection of chymotryptic activity in P. keruthurus and P. juponicus could
be a result of the use of less specific substrates (benzoyl-L-tyrosine ethyl
ester and N-benzoyl-DL-phenylalanine-2naphthylamide, respectively) than
the substrate (glutaryl-L-phenylalanine-p-nitroanilide) used in the present
study. The latter substrate is less likely to be hydrolyzed by non-specific
esterases and collagenase than the other two. The tryptic activity of Penueus
juponicus has also been examined in other studies (Laubier-Bonichon et al.,
1977; Cuzon et al., 1980). Therefore, penaeid shrimp are capable of effec-
tively hydrolyzing proteins utilizing endopeptidases, trypsin and exopep-
tidases, carboxypeptidase A and B and aminopeptidases. Furthermore,
protease activity is displayed at a pH of 2.0, indicating a residual activity of
the extracellular tryptic enzyme or an intracellular cathepsin.
The size of the shrimp and the source and level of protein in the diet all
affected to some degree the proteolytic activity in P. uannumei. Shrimp
fed the 30% peotein, 2:l a/p ratio diet displayed the highest enzyme activ-
ities in these analyses yet grew less and exhibited lower protein digestibil-
ities than shrimp fed the other 2:l diets (Smith et al., 1984). Shrimp fed
1:l a/p ratio diets exhibited the highest enzyme activities on the 38% pro-
 235

tein diet while they also grew and exhibited the highest protein digestibility
for that diet. No direct relationship could be drawn to describe this situa-
tion, but the 30% protein, 2:l a/p ratio diet contained less purified protein,
soy, casein and wheat gluten than did any other diet except the 22% protein,
2:l a/p ratio diet. Smith et al. (1984) were able to draw an inverse correla-
tion between the level of purified protein in the 2:l a/p ratio diets and
growth for small and medium shrimp. In addition, these effects of protein
level on enzyme activities were variable and depended on the way in which
the activity was expressed. Total enzyme activities clearly reflected dif-
ferences associated with different protein levels while the differences in
specific activities were usually small. Therefore, the concentration of en-
zymes in the digestive tract changed in relation to the mass (wet weight) of
the digestive tract (total activity) but these changes were not as great with
respect to the soluble protein in the digestive tract (specific activity).
The response of the three sizes of shrimp to the two animal protein to
plant protein ratio (a/p ratio) series also differed. In terms of total enzyme
activities, there was an overlapping of the activities for the two series. How-
ever, small shrimp (9-11 g) exhibited higher specific activities on the 2:l
a/p ratio diets, indicating an adaptation to the higher quality animal protein
diets. Large shrimp (15-30 g) exhibited similar specific activities on the 2:l
and 1 :l a/p ratio diets which suggests that they may be less able to utilize
the higher quality animal protein diets than are small shrimp. This change
in response to protein quality may indicate a possible switch in the natural
dietary regimen of this species of shrimp when it reaches the lo-20 g size
range. However, no such change in food preference has been reported for
natural populations of P. uannamei. Van Wormhoudt et al. (1980), using
Pulaemon serratus, also found an effect of source of protein on general
protease activities. In their study the highest activities were exhibited in
shrimp fed a diet containing 70% “Atlanta meal” with a protein level of
45%. Shrimp fed other diets possessing similar or even higher protein levels,
and composed primarily of casein, soy, spirulina or fish protein concentrate,
exhibited lower activities. However, these authors were unable to demon-
strate any correlation between growth and enzyme activity.
Maugle et al. (1982b) examined the enzyme activities and growth rate
of Penaeus jqonicus fed practical, live clam and freeze-dried clam diets.
The growth rates of the shrimp fed the live clam and practical diets were
similar, and were higher than those of shrimp fed the freeze-dried clam diet.
The general protease activities of the shrimp fed the live clam diet were twice
those of the shrimp fed the practical diet, yet the growth rates were similar.
The higher activity in the shrimp fed the live clam diet was probably due to
the protease activity present in the clam tissue. However, this increased
protease activity provided no benefit in terms of growth rate to the shrimp
fed the live clam diet. The lack of correlation between growth and protease
activities in these studies suggests that protease activity is affected by dietary
variables other than protein. In studies such as these which utilize practical
236

diets, the variation of one dietary component necessitates the variation of

other components. Regarding digestive enzymes, amino acid profiles and
metal ion levels are certainly as important as crude protein level. However,
the current importance of these studies lies with the description of digestive
enzyme adaptation for shrimp. Digestive enzyme adaptation has to this
point been an untestable hypothesis for invertebrates since studies were
limited to natural populations and natural diets.
Starvation produced a general depression in total activity of the individual
protease enzymes but had less effect on specific activities. Cuzon et al.
(1980) reported similar results in tryptic activity for 4-g Penaeus juponicus
in which one group was fed live Mytilus and another group was starved for
4 weeks. Although the tryptic activity in the digestive gland of starved
shrimp dropped by 50% to a rather constant level in 14 days compared
with that of fed shrimp, the specific activity in starved shrimp remained
higher than that of the fed group until the middle of the third week. At that
time, the specific activity of starved shrimp declined to levels significantly
lower than in fed shrimp. The authors suggested that bacterial flora as-
sociated with sand in the tank was ingested, thereby maintaining tryptic
activity. However, tanks in the present study contained no substrate, and the
intestines of starved shrimp were empty when dissected. More likely, the
amount of soluble protein in the digestive gland (which composes the
largest tissue mass in the digestive tract) drops during the initial phase of
starvation while the concentration of protease enzymes drops slightly or
remains relatively unchanged, thereby producing the apparent increase in
specific activity. Later, enzyme secretion might be adversely affected by
starvation and the specific activity would fall below that of fed shrimp as
described by Cuzon et al. (1980).
The described differences in proteolytic activity influence the formula-
tion of shrimp feeds in several ways. The source of protein in the diet had
a greater effect on enzyme activities of the small shrimp (9-11 g) than did
the level of protein. As the shrimp grew the source of dietary protein became
less important and shrimp fed diets with a greater proportion of plant pro-
tein (1 :l a/p ratio) exhibited activities equal to those of shrimp fed high
animal protein diets (2:l a/p ratio). As the catalytic abilities differ, post-
larvae and juvenile shrimp might be fed feeds similar to the starter feeds
used in other animal production systems (swine, poultry, etc.) and attention
would be focused on the sources of dietary protein, as well as on maximiz-
ing the level of protein. Shrimp weighing more than 15 g could be fed feeds
containing higher amounts of plant proteins, thereby reducing feed costs in
the later stages of intensive culture.

SUMMARY

Digestive enzyme analysis is a useful technique for the assessment of the

catalytic abilities of marine shrimp. The enzyme activities seem to respond
 231

to dietary differences over a shorter time period than is required to express

these differences in growth rate. However, the changes in enzyme activity
are not always directly related to growth rate, suggesting that enzyme activ-
ity may be affected by some nutritional variables not directly related to
growth. The levels of general protease activity are higher in the small shrimp
(9-11 g), decreasing and stabilizing at a lower level as growth continues
(15-30 g). The protein level has-an effect on the enzyme activities of shrimp
of all sizes whereas the protein source is of greater importance in small
shrimp. While specific activities of the protease enzymes in the digestive
tract of starved shrimp were comparable to those of fed shrimp, total activ-
ities did display a general depression. Therefore, as the total protease activ-
ities in the digestive tract of starved shrimp decreased, the amount of soluble
protein in the digestive tract decreased proportionally. When formulating
feeds for tank-reared marine shrimp, the source of protein should be as
carefully chosen as the protein level of the diet. This may mean designing
more than one diet for a given species since the larger shrimp (> 15 g) are
less able to chemically digest a high quality protein diet than are smaller
shrimp.

ACKNOWLEDGMENTS

This work is a result of a research program sponsored in part by grants

from the Caesar Kleberg Foundation for Wildlife Conservation to Texas
A&M University, Addison L. Lawrence, principal investigator; Texas A&M
University Sea Grant College Program, supported by the National Oceanic
and Atmospheric Administration, Office of Sea Grant, Department of Com-
merce under Grant #4-7-158-44105. The senior author was supported in
part by a Sea Grant College Marine Fellowship. The authors wish to thank
Judy Lee and Ginny Mitchell for their assistance in the statistical analysis
and preparation of this manuscript. We would also like to thank Dr. Frank
Castille, Karen Hall, Rhonda Barrera and Cheryl Fernandez for assisting with
the proximate analysis of the experimental diets and Tony Rubino for his
aid in preparing the diets.

REFERENCES

Anson, M.L., 1938. The estimation of pepsin, trypsin, papain, and cathepsin with hemo-
globin. J. Gen. Physiol., 22: 79-89.
Association of Official Analytical Chemists, 1975. Official Methods of Analysis. A.O.A.C.,
Washington, DC, 1094 pp.
Barnes, H., 1959. Apparatus and Methods of Oceanography. George Allen and Unwin
Ltd, New York, NY, 339 pp.
Bligh, E.G. and Dyer, W.J., 1959. A rapid method of total lipid extraction and purifica-
tion. Can. J. Biochem. Physiol., 37: 911-917.
Cuzon, G., Caha, C., Aldrin, J.F., Messager, J.L., Stephan, G. and Mevel, M., 1980.
Starvation effect on metabolism of Penaeus juponicus. Proc. World Maricult. Sot., 11:
410-423.
238

Dal], W. and Moriarty, D.J.W., 1983. Functional aspects of nutrition and digestion. In:
L.H. Mantel (Editor), Biology of Crustacea, Vol. 5, Internal Anatomy and Physiolog-
ical Regulation. Academic Press, New York, NY, pp. 215-261.
Erlanger, B.F. and Fidel, F., 1964. The utilization of a chromogenic inactivator in an
“all or none” assay for chymotrypsin. Biochemistry, 3: 346-348.
Erlanger, B.F., Kokowsky, N. and Cohen, W., 1969. The preparation and properties of
two chromogenic substrates of trypsin. Arch. Biochem. Biophys., 95: 211-278.
Fair, P.H., Fortner, A.R., Millikin, M.R. and Sick, L.V., 1980. Effects of dietary fiber on
growth, assimilation, and cellulase activity of the prawn, Macrobrachium rosenbergii.
Proc. World Maricult. Sot., 11: 369-381.
Folk, J.E. and Schrimer, E.W., 1963. The porcine pancreatic carboxypeptidase A system.
J. Biol. Chem., 238: 3884-3894.
Folk, J.E., Piez, K.A., Carroll, W.R. and Gladner, J.A., 1960. Carboxypeptidase B. IV.
Purification and characterization of the porcine enzyme. J. Biol. Chem., 235: 2272-
2279.
Gates, B.J. and Travis, J., 1969. Isolation and comparative properties of shrimp trypsin.
Biochemistry, 8 : 4483-4489,
Gates, B.J. and Travis, J., 1973. Purification and characterization of carboxypeptidase
A and B from the white shrimp (Penaeus setiferus). Biochemistry, 12: 1867-1874.
Gibson, R. and Barker, P.L., 1979. The decapod hepatopancreas. Oceanogr. Mar. Biol.
Annu. Rev., 17: 235-346.
Hood, M.A. and Meyers, S.P., 1977. Rates of chitin degradation in an estuarine environ-
ment. J. Oceanogr. Sot. Jpn., 33: 328-334.
Laubier-Bonichon, A., Van Wormhoudt, A. and Sellos, D., 1977. Croissance larvaire
controlee de Penaeus juponicus Bate enzymes digestives et changements de regimes
alimentaires. Actes Coll. C.N.E.X.O., 4: 131-145.
Lee, P.G. and Lawrence, A.L., 1982. A quantitative analysis of digestive enzymes in
penaeid shrimp; influences of diet, age and species. Physiologist, 25: 241.
Lee, P.G., Blake, N.J. and Rodrick, G.E., 1980. A quantitative analysis of digestive en-
zymes for the freshwater prawn, Macrobrachium rosenbergii. Proc. World Maricult.
sot., 11: 392-402.
Lowry, O.H., Roserough, N.J., Farr, A.L. and Randall, R.J., 1951. Protein measurements
with folin phenol reagent. J. Biol. Chem., 193: 265-275.
Maugle, P.D., Deshimaru, O., Katayama, T. and Simpson, K.L., 1982a. Characteristics
of amylase and protease of the shrimp Penaeus japonicus. Bull. Jpn. Sot. Sci. Fish.,
48: 1753-1757.
Maugle, P.D., Deshimaru, O., Katayama, T. and Simpson, K.L., 1982b. Effect of short-
necked clam diets on shrimp growth and digestive enzyme activities. Bull. Jpn. Sot.
Sci. Fish., 48: 1759-1764.
Meyers, S.P., 1980. Water stable extruded diets and feeding of invertebrates. J. Aquari-
cult., 1: 41-46.
New, M.B., 1976. A review of dietary studies with shrimp and prawns. Aquaculture, 9:
101-144.
New, M.B., 1980. A bibliography of shrimp and prawn nutrition. Aquaculture, 21:
101-128.
Rick, W. and Fritsch, W.P., 1974. Pepsin. In: H.U. Bergmeyer (Editor), Methods of
Enzymatic Analysis. Academic Press, New York, NY, pp. 247-257.
Smith, L.L., Lee, P.G., Lawrence, A.L. and Strawn, K., 1984. Growth and digestibility
by three sizes of Penaeus vannamei Boone: effect of dietary protein level and source.
Aquaculture (submitted for publication).
Trellu, J., 1978. Etude de diverses enzymes de Crustaces marins suivant ie &gime alimen-
taire, le dkeloppement, les chocs thermiques et l’blairement: application & I’aqua-
culture et aux effets des rejets thermiques en mer. Thi%e de Docteur Ingenier, Uni-
versitk d’Aux-Marseille, Marseille, France, 216 pp.
 239

Trellu, J. and Ceccaldi, H.J., 1977. Circadian variations of some enzymatic activities in
Pdaemon squilla Linne (Crustacea, Decapoda). J. Interdiscip. Cycle Res., 8: 357-
359.
Van Weel, P.B., 1970. Digestion in crustacea. In: M. Florkin and B.T. Scheer (Editors),
Chemical Zoology, Vol. V, Part A. Academic Press, New York, NY, pp. 97-115.
Van Wormhoudt, A., Ceccaldi, H.J. and LeGal, Y., 1972. Activite des proteases et amy-
lase chez Penaeus kerathurus: existence d’un rythme circadien. C.R. Acad. Sci.,
Paris, Ser. D, 274: 1200-1211.
Van Wormhoudt, A., Ceccaldi, H.J. and Martin, B.J., 1980. Adaptation de la teneur en
enzymes digestives de I’hepatopancreas de Palaemon serratus (Crustacea, Decapoda)
a la composition d’aliments experimentaux. Aquaculture, 21: 63-78.