Previous Page Table of Contents Next Page



H. J. Schafer
School of Food Technology and Marine Sciences
Technological Institute of Monterrey
Monterrey, Mexico


In order to estimate the utilization of storage materials by pink shrimp during molting, the changes in size of the hepatopancreas at different stages of starvation were studied. Variations in the storage materials in the gland show that starved pink shrimp utilize mainly fat reserves, followed by protein. Total glycogen reserves available would provide only 13 percent of daily energy requirements.



A l'objet d'évaluer le degré d'emploi des matériaux de réserve pendant la mue, on a étudié les variations de taille de l'hépatopancréas à diverses périodes de jeûne. Les modifications des réserves de la glande indiquent que les crevettes rosées privées de nourriture utilisent d'abord dans une large mesure les lipides, puis les protéines. Le glycogène total ne fournirait que 13 pour cent du besoin énergétique journalier de P. duorarum.



Con el fin de estimar el grado de utilización de materiales de reserva durante la muda, se estudiaron las variaciones en el tamaño del hepatopancreas en diversos períodos de ayuno. Las modificaciones en las reservas de la glándula indican que el camarón rosado privado de alimento utiliza principalmente los lípidos y luego las proteínas. El glicógeno total apenas suministraría el 13 por ciento de las necesidades diarias de energía.


The pink shrimp, Penaeus duorarum Burkenroad, constitutes an important fishery resource in South Florida and some areas of the Gulf of Mexico and the Caribbean. Although there are fairly complete data on its life history, migrations, larval development, growth rate, and storage qualities, little is known about its physiology. In most cases we can only draw parallels from knowledge about other Crustacea.

The substrates metabolized by the pink shrimp to maintain life have not been investigated before. The present work was undertaken to study what kinds of storage materials these animals utilize while food is denied to them.

It is anticipated that changes in the amounts of storage materials in starved animals may be obscured by changes in these storage products brought about by the normal processes accompanying the molting cycle. To this end it is proposed to utilize data obtained from experimental animals in the intermolt stage only (Schafer, in press), and compare the relative amounts of glycogen, fat, and total nitrogen in the hepatopancreas of animals starved for varying periods of time.


Live shrimp were obtained from a local bait fisherman. The animals were captured in the southern part of Biscayne Bay, Florida. A preliminary study conducted by Costello (1963), has shown that the shrimp catches in this area comprise two species of penaeid shrimp, Penaeus duorarum and P. brasiliensis Latreille, but the first predominates.

One sample of 100 animals was obtained on 3 October 1962, the second sample was obtained on 3 January 1963, and the third sample was obtained on 5 April 1963. Since a small percentage of these samples could be expected to be represented by P. brasiliensis, it was important to determine the species of each shrimp in the sample. The criteria described by Voss (1955) were used. Individuals were also compared with unpublished drawings by Boschi. Species identification was uncertain in the smaller males, therefore only the females in the samples were used.

Since the nutritional status of the animals was not known, they were kept in a cage in running sea water and fed chopped liver regularly for at least two weeks prior to the experiments.

After this feeding condition was standardized as described, the shrimp were sexed and the species identified. Female P. duorarum were placed in partitioned 10-gallon (45 1) aquaria with filtered sea water. The shrimp were starved for different periods of time and then removed. The starvation periods were 1, 2, 4, 6, and 11 days for the first run, 1, 2, 4, and 7 days for the second run and 1, 6, and 12 days for the third run. One sample was taken just prior to the beginning of the experiments to represent zero time. Immediately upon removal from the water the endopodite of one pleopod was excised, mounted in sea water on a glass slide, and examined under the microscope to determine the stage of the molting cycle (Schafer, in press). The shrimps were then vitrified at -72°C immediately after removal from the aquaria by immersion in a dry-ice-ethanol mixture and stored at -20°C in labeled, tightly closed screw cap glass vials until analyzed.

2.1 Analysis

The shrimp were weighed and the hepatopancreas removed by making a cross-sectional cut in the center of the cephalothorax with scissors, followed by a sagittal cut with a scalpel. The hepatopancreas was cut into four pieces which could be easily separated from the frozen tissues of the shrimp. The frozen digestive gland was placed in a tared glass container and weighed. The glands were stored in the deep freeze until analyzed. The glycogen was extracted in trichloracetic acid and precipitated by ethanol.

2.2 Trichloracetic acid extraction

The sample to be extracted was weighed and placed with 25 ml of cold 5 percent trichloracetic acid (TCA) in a Servall high-speed homogenizer vessel. The vessel was immersed in an ice bath and the sample was homogenized for 4 min at 8,000 rev/min. The homogenate was transferred to a polyethylene centrifuge tube and centrifuged at 1150 xG for 15 min. The supernatant was decanted into an Erlenmeyer flask and the precipitate was resuspended in 20 ml 5 percent TCA, and re-homogenized under the same conditions. The homogenate was again centrifuged and the supernatant was combined with the others. The precipitate from the hepatopancreas samples was weighed and used to determine the total nitrogen of the sample. Five volumes of 95 percent ethanol were added with stirring to the combined supernatants in the Erlenmeyer flask. The flasks were placed overnight in the refrigerator. The next day the mixture was transferred after shaking into 250 ml polyethylene centrifuge bottles and centrifuged at 13,200 ×G for 25 minutes at 4°C. The supernatant was separated by aspiration and saved. The precipitate containing the ethanol-insoluble carbohydrates was dissolved in 30 ml distilled water and saved for glycogen determinations.

2.3 Glycogen determination

Glycogen was determined by the glucose it released on hydrolysis with concentrated sulfuric acid. Two ml of the TCA water-soluble ethanol-insoluble fraction were placed in a 19 × 150 mm test tube. Then 0.5 ml anthrone solution was pipetted into the test tube, followed by 5 ml concentrated sulfuric acid layered carefully into the tube. The tube was rotated to mix the layers. Careful mixing was continued until the flocculent anthrone precipitate disappeared. After this, complete mixing was insured by inverting the tube once. After the color was developed for at least 10 min, the optical density was measured at 620 mμ against a distilled water blank. Glucose standards containing from 0 to 80 μg of glucose in 2 ml of distilled water were run at the same time. The glycogen content of the unknown was estimated by multiplying the glucose equivalents by 0.9.

2.4 Nitrogen determination

The precipitate from the TCA extractions described earlier were placed on weighed lens tissue squares and dried in an oven at 100°C overnight, then weighed and placed in Kjeldahl flasks with 2 ml concentrated sulfuric acid and one “Hengar” granule. Standard micro Kjeldahl methods were followed.

2.5 Fat determination

The last batch of shrimp was used to determine total fat in the hepatopancreas of the animals. The group was subdivided into three samples, corresponding to starvation for 1, 6, and 12 days respectively. The animals were treated as described earlier. The glands were removed while frozen, weighed and stored in the deep freeze until analyzed. For analysis, the glands were placed one at a time in a mortar with 10 g sodium sulfate per 100 mg sample to absorb the water, and ground. The powder was placed on filter paper in a glass funnel and a measured amount of petroleum ether was poured over the powder to extract the fat. The amount of ether used was 20 ml ether/0.1g sample. The extract was collected in a weighed glass container and the ether was evaporated to dryness. The container was weighed again and the difference represented the total fat in the sample. The percentage of fat in the samples was calculated.

2.6 Treatment of data

  1. The weights of the hepatopancreas, expressed as mg hepatopancreas per body weight, were used as the Y values, against the number of days starved as the X values. The regression line for the values was calcultated and plotted. The slope of the line was tested at the 95 percent confidence level using Ho (slope = 0).

  2. The glycogen content of the hepatopancreas samples was expressed as mg per g of dry tissue. These values were taken as the Y values, against the days of starvation as the X values. The values were treated as before.

  3. The nitrogen content of the hepatopancreas samples, expressed as percent nitrogen, were used as the Y values, against the number of days starved as the X values. The values were treated as in (a).

  4. The fat content of the hepatopancreas samples, expressed as percent fat, were taken as the Y values, against the number of days starved as the X values. The values were treated as before.

  5. The contents are expressed as the mean values ± twice the standard errors of the mean.


3.1 Glycogen determinations

The linear regression for the data from intermolt shrimp was calculated and its slope found to be negative and significantly different from zero (Fig. 1). The data variance was 0.402.

3.2 Nitrogen determinations

The percentage of nitrogen present in hepatopancreas samples was determined by standard micro-Kjeldahl techniques (Horwitz, 1955). The linear regressions for the samples were calculated and the slope showed no significant difference from zero at the 95 percent confidence level. Total nitrogen in hepatopancreas samples had a mean value of 6.61 percent ± 0.168, equivalent to an estimated protein content of 41 percent.

3.3 Fat determinations

Total ether-extractable fat content of hepatopancreas was recorded and the linear regression for the data calculated. The slope of the line was negative and differed significantly from zero, indicating that the fat content of the hepatopancreas samples decreases as starvation continues. This decrease is shown in Fig. 2.

3.4 Relative weights of hepatopancreas

The percentage of the total weight represented by the hepatopancreas was recorded and the linear regression for the data calculated. The slope of the line was negative and differed significantly from zero. This suggests that the relative size of the glands decreases with time of starvation. The decrease in size is shown in Fig. 3.


4.1 General aspects of metabolism

To gain a better understanding of the energy stores available to P. duorarum (Table I), they were calculated for an average 10 g shrimp. All calculations were based on utilization of materials stored in the hepatopancreas. The hepatopancreas decreased in size progressively during starvation. From Fig. 3 it can be estimated that the hepatopancreas of a 10 g animal decreased 23 mg in weight during each day of starvation. This figure, coupled with the protein content of the gland and the daily decreases in glycogen and fat (Fig. 1,2) in the gland, allows an estimate to be made of the amount of the three different materials that would have been used by the 10 g shrimp in one day. Table II shows the number of calories contributed by each of these compounds, as well as the amount of oxygen required for their complete oxidation, and the carbon dioxide produced in metabolizing these substrates.

Changes in Hepatopancreas glycogen with starvationFat content of the Hepatopancreas during starvation (Dry weight)
Fig. 2
Fig. 1
Fig. 1Fig. 2

Table I contains a summary of the calculations of the material flux through a 10 g shrimp. Examination of the number of calories made available by the storage material suggests that during starvation P. duorarum derives most of its energy from fats (85.81%) and protein (14.07%).


Substrates utilized, caloric content, O2 and CO2 changes in the hepatopancreas of starved P. duorarum of 1-g live weight per day

SubstrateWeight (μg)Caloric content (kcal)O2
Consumed (ml)
Produced (ml)
Protein3292.7514.2073.182  2.548
Carbohydrate    29.10  0.1220.024  0.024
Total 100.993  21.703  15.546


Caloric content, O2 consumed and CO2 released by different substrates per day (1962 samples)

O2 Consumed
CO2 Produced
Fats9.3–9.4612019.3  1427.3  

Carbohydrate contributes only 0.12 percent to the energy pool. These results agree with those of Neiland and Scheer (1953), who found that fasting crabs use protein or lipids in preference to glycogen. The maximum glycogen content of the hepatopancreas of normal (i.e., fed) individuals was 8.63 mg/g. The hepatopancreas of a 10 g shrimp weighs only 383 mg (Fig. 3). This organ contains only 3.31 mg total glycogen. Metabolizing this amount of carbohydrate would provide at most 13.776 calories, which is only a little over 13 percent of the calculated daily caloric need of P. duorarum (Tables I and II).

This shift in reserves is common in starved animals. Determinations of respiratory quotients (R.Q.) have been employed to determine the relative proportions of types of foods being oxidized (Fruton and Simmonds, 1959; Prosser, 1952; Wolvekamp and Waterman, 1960). The interpretation of R.Q.s must be done with caution since in Crustacea interaction of CO2 with the exoskeleton may give rise to abnormal R.Q. values (Wolvekamp and Waterman, 1960). When only carbohydrates are metabolized, the R.Q. is 1, while for fats alone it is 0.707. When proteins are metabolized the R.Q. is variable and depends on how far they are metabolized (Fruton and Simmonds, 1959). Using the assumptions of Table I, the respiratory quotients of starved P. duorarum is 0.721, which agrees very well with those given by Wolvekamp and Waterman (1960) for starved Crustacea (Table III).

Table III also shows that when the animals are fed, the R.Q. shifts to a higher value, indicating a possible shift towards greater participation of carbohydrates in metabolism. The total oxygen consumption calculated in Table I may be expressed as 90.4 μl of oxygen per hour per gram wet weight. This value falls well within the range of values given by Wolvekamp and Waterman for the oxygen consumption by Crustacea and is comparable to the value obtained by Zein-Eldin (1961) for P. aztecus Ives of 130 ml/g/h.

Weight of Hepatopancreas during starvation

Fig. 3

Fig. 3


Glycogen expressed in g per 100 g dry weight

             FEEDING PERIOD

Renaud 1949


Only those values for oxygen consumption above 20°C were included in the Table, since all animals in the present study were maintained above 20°C during the experiments.

4.2 Changes in the hepatopancreas

P. duorarum showed a reduction in size, together with alterations in composition of the hepatopancreas, with induced starvation. The fat content of the gland decreased markedly with continued starvation. The glycogen content of the hepatopancreas declined with increasing starvation. The molting cycle is accompanied by changes in the composition and size of the hepatopancreas in Crustacea (Renaud, 1949; Passano, 1960). It may well be that these changes are partially due to the obligatory fasting period during the cycle and thus can be equated to the changes observed in P. duorarum during their experimentally induced fast. The dry mass varies from a minimum value at stage C1 of the molting cycle to a maximum of three times the minimum size at stage D1 (Passano, 1960). The mass then declines somewhat during D2 and remains nearly constant until A, declining after this to a minimum at C1. During the molting cycle, the major part of the hepatopancreas organic reserves accumulated during C4 consists of lipids (Passano, 1960). The fatty acid content of Cancer pagurus Linn. was tripled during the feeding period, C1 to D1, and most accumulation occurred during C4 (Renaud, 1949). By D3, hepatopancreas lipids showed a high turnover rate and almost disappeared by stage C (Passano, 1960). The decrease in lipids during starvation in samples in the present experiment paralleled the decrease in lipids shown during obligatory fasting from stages D2 to C1 in crustaceans. It has been shown (Bliss, 1953) that if both eyestalks of the land crab Gecarcinus are removed, the animals assume a state of proecdysis and the R.Q. falls from 0.77 to 0.69. The latter R.Q. agrees with the estimate here of an R.Q. of 0.721 for starved P. duorarum. This suggests that starved crustaceans, whether the condition is induced experimentally or naturally, metabolize fats to a greater extent than other reserves.


Respiratory Quotient (R.Q.) of crustaceans (Modified from Wolvekamp and Waterman, 1960)

SpeciesTemp. °C       R.Q.
Daphnia pulex20.01.03 average
1.13 fed
0.71 starved
Palaemon serratus15.8
0.9–0.95 fed
0.74 starved
Astacus astacus (F)14.00.91 fed
0.72 starved
Astacus astacus (M)11.50.93 fed
0.76 starved

F - females,
M - males

During the molting cycle of crustaceans, hepatopancreas glycogen increases through the intermolt period and continues to do so during the early portions of the proecdysal period (D) (Passano, 1960). After stage D2, the glycogen content decreases until it reaches a low at C1 (Fig. 4).

The glycogen content of the hepatopancreas of P. duorarum decreased during starvation. Calculations indicate that if all the glycogen in the hepatopancreas of these animals were metabolized it would provide only 13 percent of the caloric requirement for one day.

Passano (1960) states that “nearly 70 percent of the protein nitrogen of Canoer found at stage D has been calculated to be utilized in the succeeding ecdysis. Soluble nitrogen representing catabolic products of protein and purine breakdown, essentially parallels the changes in protein, although it is slightly higher from D2 through D4.”

Neiland and Scheer (1953) indicate that Hemigrapsus uses protein rather than carbohydrate and fat as primary energy source, during starvation. Present results show no significant changes in the nitrogen level of the hepatopancreas of P. duorarum with starvation, but since the gland becomes smaller as starvation proceeds, the absolute level of protein (nitrogen) diminishes with starvation (Fig. 3). Calculations indicate that about 14 percent of the caloric requirements of starved animals were met by protein metabolism. Needham (1956) has shown that in the isopod Asellus, nitrogen excretion (NH3 and amino nitrogen) increased just prior to exuviation and reached a maximum shortly after. This was followed by a decline in nitrogen excretion. The increase in nitrogen excretion suggests an increase in the catabolism of nitrogen-bearing substrates and may be due to increasing use of protein as a substrate. On the other hand there is a possibility that acetylglucosamine, from the breakdown and reabsorption of chitin by-products prior to molting, is being oxidized to CO2, as was demonstrated in rats by Kohn, Winzler and Hoffman (1962). Work by the authors cited, the results obtained on the changes in storage materials from the hepatopancreas of P. duorarum during induced fasting, and the chitin resorption of 70 percent indicated by Schafer (1967), suggest that glycogen and fat are stored in the hepatopancreas during intermolt. It appears that during the obligated fast after molting, fat and protein are utilized by the animals and chitin formation is greatest during this period. Prior to ecdysis, chitin is resorbed and its components may be utilized in the general metabolic processes.


When pink shrimp, P. duorarum, are starved they metabolize mainly fat reserves and protein. The size of the hepatopancreas decreases with increasing starvation and its composition is altered. Fat stores in the hepatopancreas decrease markedly during starvation, and glycogen stores also decrease.

Carbohydrate storage material seems to play a minor role in maintaining these animals alive during starvation, but it probably contributes more during the period in which they feed. A review of the literature indicates that carbohydrates are probably of great importance as intermediates for the interconversion of other reserves.

The changes observed in the size and composition of the hepatopancreas of starved animals seem to parallel those reported during the molting cycle of other Crustacea. It is suggested that these changes are brought about to some extent by the obligatory fast imposed on Crustacea by the molting cycle. It is indicated that during the post-molt period, chitin synthesis is most active. During the intermolt period, carbohydrates probably contribute considerably to the energy requirements of the animals. During stage D2 of the premolt period, the animals cease to feed and fat stores as well as protein reserves are metabolized. It is possible that at this time the breakdown products of chitin resorption may play a role in the metabolic processes.


Bliss, D.E., 1953 Endocrine control of metabolism in the land crab, Gercarcinus lateralis (Fréminville). 1. Differences in the respiratory metabolism of sinusglandless and eyestalkless crabs. Biol.Bull.mar.biol.Lab., Woods Hole, 104(3):275–96

Costello, T.J., 1963 Pink shrimp life history. Annual Report, Fishery Research Galveston Biological Laboratory, Circ.Fish Wildl.serv., Wash., (161):35–7

Fruton, J.S. and S. Simmonds, 1959 General biochemistry. New York, John Wiley and Sons, 2nd ed., 1077 p.

Horwitz, W. (Ed.), 1955 Official methods of analysis. Washington, D.C., Association of Official Agricultural Chemists, pp. 805-6

Kohn, P., R.J. Winzler and R.C. Hoffman, 1962 Metabolism of D-glucosamine and N-acetyl-D-glucosamine in the intact rat. J.biol.Chem., 237(2):304–8

Needham, A.E., 1956 Nitrogen output and ecdysis in crustacea. Nature, Lond., 178(4531): 495–6

Neiland, K.A. and B.T. Scheer, 1953 The hormonal regulation of metabolism in crustaceans. 5. The influence of fasting and of sinus gland removal on body composition of Hemigrapsus nudus. Physiologia comp. Oecol., 3:321–6

Passano, L.M., 1960 Molting and its control. In The physiology of Crustacea, edited by T. H. Waterman, New York, Academic Pres, vol.1, pp. 473–536

Prosser, C.L., 1952 Comparative animal physiology Philadelphia, Pa., W.B. Saunders Co., 888 p.

Renaud, L., 1949 Le cycle des réserves organiques chez les Crustacés Décapodes. Annls océanogr., Manaco, 24:259–357

Schafer, H.J., 1967 Chitin resorption by Penaeus duorarum. Contr.Fdn mar.Sci., Mexico, 1:1–8

Schafer, H.J., (in press) The determination of some stages of the molting cycle of Penaeus duorarum by microscopic examination of the setae of the endopodites of pleopods. Paper presented to FAO World Scientific Conference on the Biology and Culture of Shrimps and Prawns, Mexico City, 12–21 June, 1967

Voss, G.L., 1955 A key to the commercial and potentially commercial shrimp of the family Penaeidae of the Western North Atlantic and the Gulf of Mexico. Tech.Ser. Fla St.Bd Conserv., (14):1–23

Wolvekamp, H.P. and T.H. Waterman, 1960 Respiration. In The physiology of Crustacea, edited by T.H. Waterman, New York, Academic Press, vol.1, pp. 35–100

Zein-Eldin, Z.P., 1961 Shrimp physiology. Annual Report, Fishery Research Galveston Biological Laboratory. Circ.Fish Wildl.Serv.,Wash., (129):44–8

Previous Page Top of Page Next Page