The effect of starvation on refeeding, digestive enzyme activity, oxygen consumption, and ammonia excretion in juvenile white shrimp Litopenaeus vannamei.

ABSTRACT Juveniles of the white shrimp Litopenaeus vannamei were kept without food for between 0 to 15 days to evaluate the impact of starvation on physiologic state (oxygen consumption, poststarvation refeeding index, nitrogen excretion, and O:N ratio) and digestive enzymes activity. Physiologic changes were found after 6 days of fasting, and refeeding ability declined as a result. Nevertheless, the shrimp were able to survive 16 days without food. Starvation caused metabolism to drop progressively toward a basal level (21 J x [h.sup.-1] x [g.sup.-1]) and a decrease in the rate of ammonia excretion, because of the catabolism of amino acids from soluble protein in the hepatopancreas. This decrease led to an increase in digestive enzymes specific activity (U/mg protein). But, expressed as total U, all digestive enzyme activities decreased in the absence of substrate from 0.016 to 0.007 U/hepatopancreas (HP) for [alpha]-amylase and 2.58 to 0.63 U/HP for total trypsin. L. vannamei juveniles showed a true physiologic adaptation mechanism to food deprivation: no changes in body weight but loss in hepatosomatic index, no exuviations, including the utilization of HP soluble proteins (a drop from 269 to 53 mg/mL). After 10 days, a neoglycogenic pathway and the corresponding tissue enzymes activities seemed enhanced, and the animals derived all energetic substrates mainly from protein (O:N ratio of 17) to cover their metabolic costs. Estimates of basal metabolism (Hem) from the routine respiration rate per day (from 361 to 725 J x g [ww.sup.-1] x [day.sup.-1] through the 15-day starvation period), and loss of nonfecal energy (HxE) from the nitrogen excretion rate (varying from 39 to 57 J x g [ww.sup.-1] x [day.sup.-1] during the same period) were used in a bioenergetic partition model of a fasting juvenile, which indicated that the energetic requirement to survive without feeding was in the range of 418 and 771 J x g [ww.sup.-1] x [day.sup.-1] during the 15-day period of starvation.

KEY WORDS: digestive enzymes, Litopenaeus vannamei, metabolic rate, shrimp, starvation

INTRODUCTION

The Pacific while shrimp Litopenaeus vannamei is cultured in extensive, intensive, and semi-intensive systems and is, with Litopenaeus stylirostris, the most popular shrimp for aquaculture in Mexico and Central and South America. The state of Sinaloa on the northwest coast of Mexico, considered the most important agricultural producer in the country, has observed in recent years an important increase of aquaculture farms, from approximately 100 hectares of ponds in 1984 to 1850 hectares in 1998, with a growth rate of about 150 hectare/year. Nowadays, Sinaloa has more than 200 shrimp farms (about 75% of the national total) and produces around 10,000 tons yearly (65% of the national total) (Hernandez Cornejo & Ruiz Luna 2000).

The digestive gland is generally regarded as a major storage organ in decapods crustaceans (Allen 1971, Huggins & Munday 1968). The study of the digestive gland is of considerable interest because of its role in the accumulation and cyclic mobilization of reserves during the molting process, its contribution of nutrients to the ovary during vitellogenesis, the mobilization of its reserves during starvation, and its role in digestion and absorption. The level of the digestive enzymes in decapod crustaceans does not remain constant during the developmental cycles (Van Wormhoudt 1974) as a result of both external and internal factors. Among the external factors, the quantitative and qualitative variability of food is poignant.

Physiologic and biochemical effects of starvation have been studied in several decapod species (Anger 1986, Cuzon et al. 1980. Cuzon & Ceccaldi 1973, Dall & Smith 1986, Wehrtmann 1991), but little is known about the effects of prolonged food deprivation. Changes in biochemical composition during starvation have been reported in Mursupenaeus japonicus where there was a progressive suppression of metabolism compared with normally fed shrimp (Cuzon et al. 1980). The study found that the shrimp primarily used carbohydrates, then lipids to meet their energy requirement; proteins were significantly used only during the fourth week of starvation (Cuzon et al. 1980). In L. vannamei postlarvae, triacylglycerol provided energy during short periods of starvation whereas protein was used during prolonged starvation (Stuck et al. 1996).

Modifications in digestive enzyme activity have been found in several penaeid shrimp and related to the amount and quality of food (Lemos & Rodriguez 1998, Le Moullac et al. 1996, Le-Vay et al. 1993, Rodriguez et al. 1994). Rosas et al. (1995) found that the type and concentration of food influenced the ingestion rate of larval L. setiferus, which, in turn, affected metabolic rate. Changes in digestive enzyme activity were found under starvation conditions in juveniles of Marsupenaeus japonicus (Cuzon et al. 1980) and Metapenaeus ensis (Leung et al. 1990). Both studies found a decrease in digestive enzyme activity compared with fed shrimp.

Nitrogen excretion and metabolic rates are influenced by many factors, such as molt stage, feeding conditions, and level of activity. In this sense, O:N ratio has widely been used as an index of used substrate for oxidative metabolism (Chu and Ovsianico-Koulikowsky 1994, Dall & Smith 1986, Regnault 1981, Rosas et al. 1995). The catabolism of pure protein produced theoretical values of O:N of between 3 to 16 whereas the catabolism of equal quantities of proteins and lipids yield O:N values of between 50 and 60. Greater values of O:N correspond to an increase in lipid and carbohydrate catabolism (Mayzaud & Conover 1988).

The aim of this study was to evaluate the effect of starvation on digestive enzyme activity, oxygen consumption, ammonia excretion, and O:N ratio in L. vannamei juveniles. Afterwards, the effect of refeeding of starved juveniles was analyzed to assess the capability of shrimp to recover after prolonged starvation.

This information may contribute to a better understanding of the physiology of this commercially important shrimp species and will have an application in assisting with management in farms.

MATERIALS AND METHODS

Experimental Conditions and Design

Juveniles of L. vannamei were provided by a commercial farm, located in Mazatlan, Sinaloa, Mexico, during March 2001. The organisms were transported to the laboratory culture system of the Centro de Investigacion en Alimentacion y Desarrollo (CIAD) and fed with a commercial food during the acclimatizing period in 1000-L tanks. They were then transported to the laboratory, and groups of 20 organisms (wet weight = 0.998 [+ or -] 0.213 g) were maintained for different starvation periods (0, control group; 3, 6, 9, 12, and 15 days) in 10-L glass aquaria. Experimental conditions were 34 g x [kg.sup.-1] salinity, 24[degrees]C water temperature, and 12D:12L. Water was completely renewed daily, and molts and dead organisms were removed. The natural seawater used in this study was pumped from a nearby beach (Cerritos beach), filtered through a cartridge filter of 5 [micro]m and ultraviolet (UV) (Aquaplus, Mexico), and kept with aeration in a 600-L dark tank until use. Samples for bacterial load were regularly taken to confirm the absence of total heterotrophs and potential vibrios in the water.

Oxygen Consumption, Ammonia Excretion, and Atomic Ratio (O:N)

After each starvation period, groups of five individuals were placed in individual 500-mL respirometer chambers in a flow-through system using a continuous pump flow (ISMATEC, 12 mL/min flow). All measurements were done at the same time of day to obtain comparable data and assuming that all experimental shrimp were in minimum locomotive activity. Organisms were acclimated for 2 h in the beakers, and a sample of water from each chamber was taken to determine the initial concentration of oxygen and ammonia; flasks were sealed for 30 min, after which new samples were taken to measure final concentrations. The flow through and the sealed periods were adjusted to avoid a depletion of oxygen concentration by more than 0.5 mg/L. Samples for ammonia excretion were filtered and fixed with [H.sub.2]S[O.sub.4] (pH 2) and then frozen until processing.

The concentration of oxygen was measured using a polarographic oxygen electrode (YSI 59), and ammonia was determined by the indophenol technique (Parsons et al. 1984). Consumed oxygen and excreted ammonia were taken as being the net difference between the start and end of the sealed period. One out of every six chambers was left without a shrimp and measured as a control. The atomic O:N ratio was estimated according Taboada et al. (1998) using the individual values of oxygen consumption and ammonia excretion transformed to [micro]g At x [g.sup.-1] x [h.sup.-1] as follows: oxygen values were multiplied by 62.5 (1000 to convert the milligrams into micrograms divided by the atomic weight of oxygen, 16), ammonia values were multiplied by 58.9 [obtained through dividing 1000 by the product of the atomic weight of nitrogen (14) and the fraction of nitrogen in N[H.sub.3] (0.824)].

Digestive Enzyme Activities

After physiologic measurements, shrimps were dissected and the hepatopancreas stored at -70[degrees]C in individual 1.5-mL microtubes until enzyme assays were done.

Frozen samples were homogenized in 1 mL ice-cold pure water. Homogenates were centrifuged (at 14,000 x g for 6 min at 4[degrees]C) and the aqueous supernatant, crude or diluted (1:10 v/v), was immediately used for enzyme analysis. The soluble-protein content was measured by the method of Bradford (1976), using a microplate reader at 495 nm. Duplicate assays for each sample were made. Trypsin activity was measured by the method of Erlanger et al. (1961) with N-[alpha]-benzoyl-DL-Arg-p-nitroanilide (BAPNA) as substrate. Chymotrypsin activity was assayed by the method of Delmer et al. (1979) using N-[alpha]-succinyl-L-alanyl-L-prolyl-L-phenyl-alanine-4-nitroanilide (SAPNA) as substrate. Hydrolysis for both enzyme activities was made in 0.1 M Tris-buffer, pH 8 at 25[degrees]C, and the absorbance was measured at 405 nm. One unit of enzyme activity was defined as 1 [micro]mol of p-nitroanilide liberated in 1 min at 25[degrees]C.

[alpha]-amylase activity was assayed according to Bernfeld (1955) with 1.5% oyster glycogen as substrate in 10 mM phosphate buffer, pH 7. Absorbance measurements were made at 520 nm. For this method, one unit of enzymatic activity was defined as 1 mg of maltose liberated in 1 min at 37[degrees]C.

General protease activity was estimated in homogenates using azocoll as substrate in phosphate buffer, pH 7.5 (Todd 1949). Absorbance was measured in a spectrophotometer at 520 nm. For this method, one unit was defined as the amount of enzyme that catalyzes the release of azo dye causing a [DELTA]A/[DELTA]t = 0.001 min (Walter 1988).

[alpha]-glycosidase activity was estimated using p-nitrophenyl-[alpha]-D-glycopyranoside as substrate in 50 mM phosphate buffer, pH 6. Absorbance was measured at 410 nm. One unit of enzymatic activity was defined as the amount of enzyme that hydrolyzes 1 [micro]m of substrate per minute.

Post Starvation Refeeding Index

Other subgroups of organisms for each treatment (n = 6) were kept in individual chambers and exposed to a known amount of food previously lyophilized. The proximate composition of the food was 35% protein, 3.5% oil, 30% fiber, 16% ash, 12% humidity; average weight of each lyophilized pellet was 15.14 [+ or -] 2.28 mg; calorific content, 5.5 cal/mg of food. After 2 h 30 min, the nonconsumed food was removed and lyophilized again. The consumed food was calculated to measure the poststarvation refeeding index, defined as PSRFI (food consumed weight/body weight).

Bioenergetics Model

From a bioenergetic point of view, to integrate a model of energy partition for starved shrimp, we followed an equation (Bureau et al. 2000):

DE requirement = [RE + Hem + HiE + HxE [urine excretion energy (UE) + gill excretion energy (ZE)] + SE]

This model originally included parameters where DE is digestible energy requirement, RE is energy gain, Hem is maintenance of energy requirement, HiE is heat increment of feeding, HxE is nonfecal energy losses, and SE is surface loss or exuviae. With this model, a general energy balance model for starvation conditions in shrimp can be calculated.

However, in starved shrimps, the following parameters can be considered as 0:

DE = 0 (no feeding)

HiE = 0 (no heat increment of feeding)

SE = 0 (no exuvia)

Hence, a model for starving shrimp could be

RE = -[Hem + HxE(UE + ZE)]

in which RE can be understood as endogenous energy needed to survive (Rosas, pers. comm.) and it is negative because in absence of energy input it is not possible to have energy gain sensu stricto. It is called also the scope for growth (SFG). Using this model, we calculated the endogenous energy (RE) for starving shrimp.

Statistical Analysis

To determine significant differences among starvation periods, one-way analysis of variance (ANOVA) and Tukey range test were used when the data were normal. Those results that were not normally distributed were evaluated through the Kruskal-Wallis test and Dunn's multiple comparisons (Daniels 1978). For both analyses, P was set at 0.05.

RESULTS

Weight and Survival

No significant differences were observed (P > 0.05) in relation to individual wet weight among treatments. However, survival decreased the longer the period of starvation (Table 1). The percentage of total molts were approximately 10% for 3- and 6-day treatments. No molting occurred after 6 days of starvation.

Oxygen Consumption, Ammonia Excretion, and O:N Ratio

Oxygen consumption increased significantly at 3, 6, and 9 days in starved animals (mean value 1.75 mg [O.sub.2] [h.sup.-1] x [g.sup.-1]) compared with the control. At 12 and 15 days, values were not significantly different from the control group (mean value 1.13 mg [O.sub.2] [h.sup.-1] x [g.sup.-1]) (H = 14.1; P = 0.015) (Fig. 1).

[FIGURE 1 OMITTED]

There were no significant differences in the amount of ammonia excreted between starved and control animals (mean value 0.08 mg N-N[H.sub.3] [h.sup.-1] x [g.sup.-1], H = 1.74, P = 0.88) (Fig. 2). This may be because high standard deviations were observed, especially in the 6- and 12-day starvation groups (0.03 and 0.067 mg N-N[H.sub.3] [h.sup.-1] x [g.sup.-1] respectively).

[FIGURE 2 OMITTED]

No significant differences between treatments were observed with regard to O:N atomic ratios (H = 10.14, P = 0.071) (Fig. 3). However, there was a trend toward increased ratios between days 3 and 9, suggesting that lipid was being catabolized. By day 12, the drop in the ratio to 10-20 suggests that protein was once again the main source of energetic fuel.

[FIGURE 3 OMITTED]

Enzyme Activity

Significant differences were found in the hepatosomatic index after different periods of starvation (H = 25.6, P = 0.0001). The index was significantly lower after animals were starved beyond 9 days. Beyond the ninth day, index decreased a further 50% until reaching the end of the experiment (Table 1). Hepatopancreatic total soluble protein was significantly affected by the period of starvation (ANOVA, P < 0.05). The highest value (268.7 [+ or -] 52.13 mg/mL) was obtained in the control, followed by the 3- and 6-day starvation experiments (mean value 159.8 mg/mL). The lowest significant values were found in the 9- to 15-day experiments (53.23 mg/mL) (Fig. 4).

[FIGURE 4 OMITTED]

Digestive enzyme activity was, in general, significantly affected by different periods of starvation. Hepatopancreatic digestive carbohydrases results are shown in Figure 5. Total amylase activity was significantly decreased by starvation (ANOVA, P < 0 05) from the highest value in fed juveniles (control) (0.016 [+ or -] 0.002 total U), followed by the 3- and 6-day experiments (mean value 0.011 total U) and the lowest value in the 9-, 12-, and 15-day starved groups (mean value 0.007 total U) (Fig. 5a). However, the specific activity increased as the soluble protein content decreased (Fig. 5b).

[FIGURE 5 OMITTED]

Glycosidase activity was significantly decreased (mean value 1.47 total U) by all starvation periods as compared with control fed animals (3.6 [+ or -] 1.5 total U) (Fig. 5c). For this enzyme, the specific activity also increased as the soluble protein content in the hepatopancreas decreased (Fig. 5d).

Proteinase activity was significantly affected by the starvation periods (ANOVA, P < 0.05) (Fig. 6). The highest value (5.3 [+ or -] 2.4 total U) was observed in the control, which was not significantly different from 3- and 6-day experiments, whereas the lowest value of total proteinase activity (1 [+ or -] 0.12 total U) was observed in the shrimps starved for 15 days (Fig. 6a). As with the carbohydrases, the specific activity of total proteinases was highest in shrimps starved for 12 days (Fig. 6b).

[FIGURE 6 OMITTED]

Total trypsin activity was significantly lower from the ninth day to the end of starvation (mean value 0.63 total U). Fed shrimp and those deprived for 3 and 6 days yielded a mean trypsin activity value of 2.58 total U (Fig. 6c).

Total chymotrypsin activity decreased over the period of starvation. Significant differences were found among the fed shrimps (45.41 [+ or -] 11.89 total U), the 3- and 6-day starvation groups (mean value 25.02 total U), and the 9-, 12-, and 15-day starvation groups (mean value 7.87 total U) (Fig. 6c). The specific activities of these latter two endoproteinases were not significantly affected by starvation (Figs. 6d and 6f).

Poststarvation Refeeding Index

The PSRFI was not significantly different between the control and 3- and 6-day starvation experiments (PSRFI = 5.09%, 4.71%, and 4.75%, respectively) whereas the amount of food consumed decreased for the 9-, 12-, and 15-day starvation experiments (PSRFI = 3.72%, 3.3%, and 2.97%, respectively) (Fig. 7) (Kruskal-Wallis, H = 21.5; P = 0.006).

[FIGURE 7 OMITTED]

Bioenergetics Model

Estimation of RE values indicates that at the beginning as well as at the end of the starvation period, shrimp needed the same endogenous energy to survive, whereas between 3 and 9 days they needed more energy, used essentially in respiration to maintain basal metabolism (Table 2). However, the routine respiration rate measured was 61% higher at day 3 of starvation than in fed shrimp and remained at this high level until day 9 of starvation. After then, values decreased to the same value as in controls.

DISCUSSION

In the current study the wet weight did not change with starvation, but survival rate decreased, and the suppression of molt after 6 days of starvation was observed. Anger and Spindler (1987) have observed in Hyas araneus a delay in molt under starved conditions and associated it with the point of reserve saturation (PRS). The PRS is the critical point where sufficient energy and/or essential substances have been accumulated to allow autonomous (food-independent) development through the entire rest of the molt cycle. Stuck et al. (1996) have also observed the inhibition of molting in response to starvation in L. vannamei. Under starvation conditions, a first event commonly observed is a weight loss in relation to energy expenditure for basal metabolism; after a few days, shrimp in premolt stages will not evolve further and refrain from exuviation, saving around 1.4 kJ (Read & Caulton 1980), which is the energy expenditure at molt. This means that shrimp have the adaptation to tolerate starvation, saving energy from exuvia, including the energy challenged to mobilize reserves, chitin digestion, and exuviation. In this way, shrimp could compensate for weight loss, maintaining body weight without significant changes, as was observed in the current study.

Litopenaeus vannamei juveniles, as many species of crustaceans, showed a biochemical adaptation response to an absence of food (decrease of digestive enzyme activities) using their own reserves (hepatopancreatic glycogen, protein, and probably lipids, estimated through O:N ratio variations) for homeostasis and to channel enough energy for basal metabolism in that period (Cuzon and Ceccaldi 1973, Dall and Smith 1986; Dall and Smith 1987, Leung et al. 1990).

Homeostasis in shrimp can change to help animals sustain severe food deprivation and survive; it is not instantaneous as the poststarvation and refeeding showed; and not all starved shrimps stayed alive (Table 1). It seems that under the experimental conditions tested, there was a kind of discrimination between shrimps that were capable of surviving starvation and those that were not. Some individuals may be more able than others to change some metabolic routes and resist food deprivation without significant morbidity. The previous life history of the animal before animals were sampled in the farm and used in experiments may also influence the variability between individuals to survive starvation. It must also be appreciated that these results will be biased due to the selection by farmers of the most robust and fastest growing individuals in the population.

According to these results, a high rate of mobilization of reserves was observed between days 3 and 9 of starvation, when the recovered energy (RE) was maximal. Shrimp reserves are mainly limited to lipids stored in the digestive gland. When shrimp are starving, they will use those reserves, increasing the energy debt and reducing the digestive gland weight and its components. Reduction of enzyme activity and soluble protein during starvation underlined the change. It means the shrimp are biochemically and energetically well adapted to fasting because they could mobilize their reserves to be used as energetic sources; at the same time, the enzyme activity in the digestive gland was maintained. This type of strategy has been observed in other shrimp species. Cuzon et al. (1980) showed that M. japonicus used protein to obtain energy though the catabolism of amino acids present in digestive gland cells during a prolonged starvation period. A similar response was observed in Penaeus esculentus by Smith and Dall (1991), evidencing that shrimp can mobilize their own energetic reserves through catabolizing lipids (after 3 to 6 days of starvation) or protein to sustain themselves through food deprivation. Otherwise, a negative correlation was observed in starvation condition between total hepatopancreatic soluble protein and total activity for all measured digestive enzymes. This correlates well with the lower values of poststarvation refeeding rate.

In general terms, digestive enzymes follow the presence or absence of food. Samain et al. (1983) found that amylase increases in case of food deprivation with a peak, and then the enzyme production decreases as an adaptation to low nutrition status and to save energy. Digestive enzymes of M. japonicus showed a similar trend that supports previous results (Cuzon et al. 1980). However, results obtained after refeeding indicate that the necessary time to recover the digestive gland integrity depends on the fasting period. These results show that although L. vannamei have an adaptation mechanism to tolerate fasting conditions, recovery cannot be achieved if starvation is long enough to produce physical damage in the digestive gland and loss of enzyme synthesis. A pattern of variation for digestive enzymes presented by L. vannamei during a shorter experimental period differed from the one of M. japonicus. At day 15, for example, specific enzymes activity increased as a sort of adaptation to absorb the minute amount of food, but as energy expenditure increased, the peak disappeared shortly. Both amylases and proteases exhibited the same trend.

Mayzaud and Conover (1988) described starvation condition regarding the use of energetic substrates estimated through O:N ratio. It can be assumed that acetyl CoA, which is the final product of [beta]-oxidation of fatty acids, could be challenged into biologic oxidation, requiring 46-52 oxygen atoms whereas, in normal condition, [beta]-oxidation of neutral lipids requires 14-16 oxygen atoms to produce acetyl CoA. Increase in [O.sub.2] consumption of shrimps between days 3 and 9 of starvation with no increase of nitrogen excretion resulted in a high O:N ratio. Then, [beta]-oxidation of hepatopancreatic lipid reserves could be occurring in this period. An explanation for a change can be found in the O:N ratio in which the use of lipid as energetic substrate was clear followed by protein as the main energy source. Ammonia excretion values provided indication on substrate oxidation; although nitrogen excretion was not significantly different from the control along the starvation period, O:N ratio values showed a trend to use lipids as energetic substrate between days 3 and 9 of starvation. Such a trend has also been evidenced in M. rosenbergii (Clifford & Brick 1983). After 12 starvation days, shrimp returned to the use of protein to derive energy. Likewise, as Mayzaud and Conover (1988) reported for planktonic crustaceans, a decrease in O:N ratio with time of starvation seemed to be common to all species with a predominantly protein-based metabolism. O:N ratio variation can be related to glucose homeostasis, through the regulation of glyconeogenesis (Cuzon et al. 2001) and glycolysis (Hochachka et al. 1988). These results are not in contradiction to those reported by Dall and Smith (1986) for Penaeus esculentus. These authors pointed out a reduction in metabolic rate with starvation providing a mechanism for prolonged survival. O:N ratios gave an explanation for successive fuel substrates as starvation increased in intensity (Table 3). Glycogen in the hepatopancreas is affected first, as in Crangon crangon (Regnault 1972), then neutral lipids are hydrolyzed, and toward the end of the starvation period, protein is used similarly as in M. japonicus juveniles (Cuzon et al. 1980).

This study has shown that L. vannamei is unable to sustain food deprivation for periods longer than 15 days. By comparison, the temperate species M. japonicus can survive 4 weeks of starvation. Both species show similar levels of digestive enzyme activity. However, basal metabolic rates in the farmer are higher because of its tropical habitat (24-27[degrees]C) as opposed to the temperate conditions (20[degrees]C) inhabited by the latter. They possess a similar level of digestive enzyme activities; then, there are good reasons, according to values of metabolism rates obtained in this study, to think that tropical species present a higher basal metabolism and activity that leaves them more dependent on regular food supply for their development.

The current work indicates L. vannamei is dependent on a regular food supply for its development, which has implications for its management in farms.

TABLE 1.
Wet weight (g), survival (%), and hepatosomatic index (%) of
the food-deprived shrimps L. vannamei.

                                    Starvation Days

                                  0                   3

Wet weight (g)            1.07 [+ or -] 0.1   0.86 [+ or -] 0.2
Survival (%)                     100                 90
Hepatosomatic index (%)    4.1 [+ or -] 0.7   3.3 [+ or -] 0.4

                                    Starvation Days

                                  6                    9

Wet weight (g)            0.97 [+ or -] 0.23   0.99 [+ or -] 0.3
Survival (%)                      90                  65
Hepatosomatic index (%)   2.90 [+ or -] 0.45   2.5 [+ or -] 0.4

                                    Starvation Days

                                  12                 15

Wet weight (g)            0.85 [+ or -] 0.1   1.04 [+ or -] 0.12
Survival (%)                      65                 55
Hepatosomatic index (%)    2.1 [+ or -] 0.3   1.98 [+ or -] 0.9

Data are mean [+ or -] SD.

TABLE 2.
Energetic balance of starved juveniles of L. vannamei at different
starvation days.

                        Hem                       HxE
Starvation       (J x [d.sup.-1] x         (J x [d.sup.-1] x
   Days           g [ww.sup.-1])             g [ww.sup.-1])

    0        449 [+ or -] [94.sup.b]     53 [+ or -] [5.sup.a]
    3        725 [+ or -] [68.sup.a]     46 [+ or -] [12.sup.a]
    6        664 [+ or -] [187.sup.ab]   39 [+ or -] [22.sup.a]
    9        598 [+ or -] [233.sup.ab]   43 [+ or -] [15.sup.a]
   12        361 [+ or -] [167.sup.bc]   57 [+ or -] [39.sup.a]
   15        428 [+ or -] [138.sup.b]    45 [+ or -] [16.sup.a]

                          RE
Starvation        (J x [d.sup.-1] x
   Days             g [ww.sup.-1])

    0          -473 [+ or -] [77.sup.b]
    3        -771.7 [+ or -] [65.sup.a]
    6          -696 [+ or -] [178.sup.ab]
    9          -655 [+ or -] [262.sup.ab]
   12          -418 [+ or -] [166.sup.b]
   15          -473 [+ or -] [152.sup.b]

Hem. maintenance of energy requirement; HxE, nonfecal energy losses;
RE, the endogenous energy: ww, wet weight. Data are mean [+ or -] SD.

a,b,c Values in the same column that share the same superscript
letter do not differ significantly (P > 0.05).

Table 3.
Theoretical limits of the O:N ratio according to Charmantier. *

Days of Fasting   [J.sub.1]   [J.sub.3]     [J.sub.6]      [J.sub.10]

   O:N             70-100                      50-60
   Substrate       Glycogen                Triglycerides

Days of Fasting   [J.sub.15]

   O:N               2-16
   Substrate      Amino acids

* Charmantier (pers. comm.).

ACKNOWLEDGMENTS

This study was funded by CONACYT Project No. 34952-b (A. Roque), UNAM-DGAPA-IN-231599 (G. Gaxiola), CONICET-Argentina (L.Comoglio and O. Amin), and IOC (O. Amin). The authors thank Dr. Carlos Rosas for critically reviewing the manuscript and Mrs. Ingrid Mascher for editorial assistance.

LITERATURE CITED

Allen, W. V. 1971. Amino acid and fatty acid composition of tissues of the Dungeness crab (Cancer magister). J. Fish. Res. Bd. Can. 28:1191-1195.

Anger, K. 1986. Changes of respiration and biomass of spider crab (Hyas araneus) larvae during starvation. Mar. Biol. 90:261-269.

Anger. K. & K. Spinder. 1987. Energetics, moult cycle, and ecdysteroid titers in spider crab (Hyas araneus) larvae starved after Do threshold. Marine Biology 94:367-375.

Bernfeld. P. 1955. Sur une methode de microdosage des amylases. In: N.O.K.S.P. Colowick, editor. Methods of enzymology. New York: Academic Press. pp. 149-157.

Bradford. M. M. 1976. A rapid sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.

Bureau. D. P., P. A. Azevedo, M. Tapia-Salazar & G. Cuzon. 2000. Pattern and cost growth and nutrient deposition in fish and shrimp: potential implications and applications. In: L. E. Cruz-Suarez, D. Rique-Marie, M. Tapia-Salazar, M. A. Olvera-Novoa & R. Civera-Cerecedo, editors. Avances en Nutricion Acuicola V. Memorias del V Simposio Internacional de Nutricion Acuicola. Merida, Uanl, Mexico. pp. 111-140.

Chu, K. & N. N. Ovsianico-Koulikowsky. 1994. Ontogenetic changes in metabolic activity and biochemical composition in the shrimp, Metapenaeus ensis. J. Exp. Mar. Biol. Ecol. 183:11-26.

Clifford, H. C. & R. W. Brick. 1983. Nutritional physiology of the freshwater shrimp Macrobrachium rosenbergii (de Man). 1. Substrate metabolism in fasting juvenile shrimp. Comp. Biochem. Physiol. 74A: 561-568.

Cuzon, G. & G. Ceccaldi. 1973. The effect of starvation on the metabolism of the shrimp Crangon crangon (L.). C.R. Seances Soc Biol. Fil. 167: 66-69.

Cuzon, G., C. Cahu, J. F. Aldrin, J. L. Messager, G. Stephan & M. Mevel. 1980. Starvation effect of metabolism of Penaeus japonicas. World Mariculture Society Proceedings 11: LSU, Baton Rouge, LA. pp. 410-430.

Cuzon, G., C. Rosas. G. Gaxiola, G. Taboada, & A. Van Wormhoudt. 2001. Effect of dietary carbohydrates on gluconeogenesis in premolt Litopenaeus stylirostris juveniles and pre adults. J. Shellfish Res. 20: 931-937.

Dall, W. & D. M. Smith. 1986. Oxygen consumption and ammonia excretion in fed and starved tiger prawns, Penaeus esculentus Haswell. Aquaculture 55:23-33.

Dall, W. & D. M. Smith. 1987. Changes in protein-bound and free amino acids in the muscle of the tiger prawn Penaeus esculentus during starvation. Mar. Biol. 95:509-520.

Daniels, W. W. 1978. Applied nonparametric statistics. Boston: Houghton Mifflin Company. 503 pp.

Delmer, E. G., C. Largman, V. W. Brodrick & M. C. Grokas. 1979. A sensitive new substrate for chymotrypsin. Anal. Biochem. 99:316-320.

Erlanger, B. F., N. Kokowsky & W. Cohen. 1961. The preparation and properties of two chromogenic substrates of trypsin. Arch. Biochem. Biophys. 95:271-278.

Hernandez Cornejo, R. & A. Ruiz Luna. 2000. Development of shrimp farming in the coastal zone of southern Sinaloa (Mexico): operating characteristics, environmental issues, and perspectives. Ocean and Coastal Management. 43:597-607.

Hochachka, P. W., B. Emmett & R. K. Suarez. 1988. Limits and constraints in the scaling of oxidative and glycolytic enzymes in homotherms. Can. J. Zool. 66:1128-1138.

Huggins, A. K. & K. A. Munday. 1968. Crustacean metabolism. In: O. Lowenstein, editor. Advances in Comparative Physiology and Biochemistry, vol. 3. New York: Academic Press. pp. 271-378.

Lemos, D. & A. Rodriguez. 1998. Nutritional effects on body composition, energy content and trypsin activity of Penaeus japonicus during early postlarval development. Aquaculture 160:103-116.

Le Moullac, G., G. B. Klein, D. Sellos & A. Van Wormhoudt. 1996. Adaptation of trypsin, chymotrypsin and [alpha]-amylase to casein level and protein source in Penaeus vannamei (Crustacea, Decapoda). J. Exp. Mar. Biol. Ecol. 208:107-125.

Leung, K. M., H. L. Chen & K. H. Chu. 1990. Effects of starvation on biochemical composition and digestive enzyme activities in the hepatopancreas of the shrimp Metapenaeus ensis. The second Asian fisheries forum. Proceedings of the Second Asian Fisheries Forum: Asian Fisheries Society, Tokyo, Japan. pp. 445-448.

Le-Vay, L., A. Rodriguez, M. S. Kamarudin & D. A. Jones. 1993. Influence of live and artificial diets on tissue composition and trypsin activity in Penaeus japonicas larvae. Aquaculture 118:287-297.

Mayzaud, P. & R. J. Conover. 1988. O:N atomic ratio as a tool to describe zooplankton metabolism. Mar. Ecol. Prog. Ser. 45:289-302.

Parsons, T. R., Y. Maita & C. M. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Oxford: Pergamon Press. 187 pp.

Read, G. H. L. & M. S. Caulton. 1980. Changes in mass and chemical composition during the moult cycle and ovarian development in immature and mature P. indicus Milne Edwards. Comp. Biochem. Physiol. 66A:431-437.

Regnault, M. 1972. Developpement de l'estomac chez les larves de Crangon septemspinosa Say (Crustacea, Decapoda, Crangonidae); son influence sur le mode de nutrition. Bull. Mus. Hist. Nat. Paris, France. 67. 53: 841-856.

Regnault, M. 1981. Respiration and ammonia excretion of sand shrimp Crangon crangon (L.). Metabolic responses to prolonged starvation. J. Comp. Physiol. 141:549-555.

Rodriguez, A., L. Le Vay, G. Mourente & D. A. Jones. 1994. Biochemical composition and digestive enzyme activity in larvae and postlarvae of Penaeus japonicus during herbivorous and carnivorous feeding. Mar. Biol. 118:45-51.

Rosas, C. S., E. Diaz, L. A. Soto, G. Gaxiola, R. Brito, M. Baez & R. Pedroza. 1995. Oxygen consumption and ammonia excretion of Penaeus setiferus, P. schmitti, P. duorarum and P. notialis postlarvae fed purified test diets: effect of protein levels on substrate metabolism. Aquat. Living Resour. 8:161-169.

Samain, J. F., J. Moal, J. R. Le Coz & J. Y. Daniel. 1983. Amylase et trypsine du zooplancton et surveillance du milieu pelagique In: L. Bretagne Cabioch, M. Glemarec & J. F. Samain, editors. 17th European Symposium on Marine Biology. Brest, France: Oceanol Acta. pp. 177-182.

Smith, D. M. & W. Dall. 1991. Metabolism of proline by the tiger prawn Penaeus esculentus. Mar. Biol. 110:85-91.

Stuck, K. C., S. A. Watts & S. Y. Wang. 1996. Biochemical responses during starvation and subsequent recovery in postlarval Pacific white shrimp, Penaeus vannamei. Mar. Biol. 125:33-45.

Taboada, G., G. Gaxiola, T. Garcia, R. Pedroza, A. Sanchez, L. A. Soto & C. Rosas. 1998. Oxygen consumption and ammonia-N excretion related to protein requirements for growth of white shrimp, Penaeus setiferus (L), juveniles. Aquaculture Research. 29:822-833.

Todd, E. W. 1949. Quantitative studies on the total plasmin and the trypsin inhibitor of human blood serum. J. Exp. Med. 89:295-308.

Van Wormhoudt, A. 1974. Variations of the level of the digestive enzymes during the intermolt cycle of Palaemon serratus: influence of the season and effect of the eyestalk ablation. Comp. Biochem. Physiol. 49: 707-715.

Walter, H. E. 1988. Proteinases (proteins as substrates) method with haemoglobin casein and azocoll as substrate. In: Methods of enzymatic analysis, vol. V. Weinheim: Chemie Verlag. pp. 270-277.

Wehrtmann, I. 1991. How important are starvation periods in early larval development for survival of Crangon septemspinosa larvae? Mar. Ecol. Prog. Ser. 73:183-190.

LAURA I. COMOGLIO, (1), * GABRIELA GAXIOLA, (2) ANA ROQUE, (3,4) GERARD CUZON (5) AND OSCAR AMIN (1)

(1) Centro Austral de Investigaciones Cientificas (CADIC-CONICET), Ushuaia, Tierra del Fuego, Argentina; (2) Facultad de Ciencias, UNAM, Mexico; (3) Centro de Investigacion en Alimentacion y Desarrollo, Unidad en Acuicultura y Manejo Ambiental, Mazatlan, Sinaloa, Mexico; (4) IRTA-Centre d' Aguiculture Sant Carlescle la Rapita, Spain; (5) Centre Oceanologique du Pacifique, Tahiti, IFREMER, Francia

* Corresponding author. E-mail: [e-mail]