Digestive enzymes present in Pacific threadfin Polydactylus sexfilis (Bloch & Schneider 1801) and bluefin trevally Caranx melampygus (Cuvier 1833)
Abstract
Pacific threadfin Polydactylus sexfilis (Bloch & Schneider 1801) and bluefin trevally Caranx melampygus (Cuvier 1833) are warmwater marine finfish currently under development for aquaculture in the Pacific. Differences in specific activities of digestive enzymes extracted from the stomach and mid-gut were compared to gain insight into their feeding habits in the wild and to understand their nutritional needs. Adult fish were maintained in captivity and fed a commercial pelleted feed. Serine protease measured in all tissues was at least 20 times higher in threadfin than in trevally. Aspartic proteases were the major digestive enzymes found in trevally. There was a 34-fold increase in collagenase activity in the intestine of threadfin from the prefed to the fed state. Chitinase activity was found in the stomach, pylorus and intestine of both species. However, specific activity in pylorus and intestine of threadfin increased 2.75 and 4 times, respectively, but showed little change in trevally. Amlyases were found only in trevally. Increase in lipase specific activity in the gut of trevally was higher than that for threadfin. The results indicated that the two species have diverse digestive capabilities. This appears to be consistent with their feeding habits in nature. Threadfin are more adapted to a wider range of food protein sources than trevally, but appear to be less well adapted than trevally to using complex carbohydrates. These observations may provide a basis for practical diet formulations for these two species.
Introduction
Digestive enzymes have been investigated for many years as a means of understanding the nutritional needs of animals and the effect of dietary constituents on enzyme activity ( Schneider & Flatt 1975). Among aquatic animals, greater emphasis has been placed on larval nutrition studies, in which enzyme profiles have been used to formulate appropriate diets. Dabrowski (1986) reviewed the ontogenic aspects of the nutritional requirements of fish and argued that nutrient requirements have not so far used ontogenic perspectives. Sabapathy & Teo (1993) made a quantitative assessment of digestive enzymes in rabbitfish Siganus canaliculatus and sea bass Lates calcarifer, and found that digestive enzyme profiles correlated with their feeding habits. Kohla, Saint-Paul, Friebe, Wernicke, Hilge, Braun & Gropp (1992) found that enzyme activity in the gut depended on fasting and gut fullness in Cuvier Colossoma macropomum. Tryptic activity, in contrast to peptic activity, was correlated with diet protein content, and α-amylase activity was not correlated with diet composition in this species. Feeding habits affect the type and activity of enzymes ( Jonas, Ragyanszki, Olah & Laszlo 1983), and surveys of this type have helped to define the limits for dietary protein ( Twining, Alexander, Huibregste & Glick 1983) and carbohydrates ( Spannhof & Plantikow 1983). An investigation of the enzymes present in the dolphin Coryphaena hippurus led to the conclusion that this high-level carnivore could adapt to a variety of feed ingredients ( Divakaran & Ostrowski 1990).
The purpose of this study was to compare digestive enzyme activities of adult Pacific threadfin Polydactylus sexfilis (Bloch & Schneider 1801) and bluefin trevally Caranx melampygus (Cuvier 1833), two warmwater marine finfish species that are currently under development for aquaculture in the Pacific ( Ostrowski 1997; Ostrowski & Molnar 1998). Although both species are high-level carnivores, each inhabits different ecological zones and exhibits different feeding habits and food preferences. These differences may probably be reflected in relative enzyme activities and have important implications in terms of formulations for potential practical diets. In this study, comparisons were made between captive fish that were at the same stage of development and were fed the same diet to ensure the ontogenetic accuracy of results.
Materials and methods
The threadfin and trevally used in this study were selected from domesticated broodstock, which had been held in captivity for over 2 years. Groups of fish were maintained in two separate 16 000-L fibreglass tanks and fed a commercial diet (Marine Grower; Moore-Clark, Vancouver, British Columbia, Canada; proximate content 55% protein, 14% fat, 1% fibre and 12% ash) at a rate of 3–4% body weight daily. Water temperatures remained relatively stable throughout the period of captivity ranging from 24 °C to 25 °C during winter and 26–27 °C during summer. Two threadfin (body weight 1.65 ± 0.04 kg) and two trevally (body weight 0.7 ± 0.2 kg) were euthanized by immersion in an ice bath and dissected. The stomach, pyloric caeca and intestine [the latter two termed mid-gut by Smith (1989)] were separated. No special efforts were made to separate the adhering pancreatic tissue from the pyloric caeca. Two sets of samples were collected from each of the species, one before feeding (designated as prefed) and another 3 h after feeding (designated as fed). The pH of the lumen of the gut was noted at the prefed and fed stages using a pH paper, and sample preparation was completed within 1 h of collection.
Crude enzyme extracts obtained from each fraction were homogenized separately in cold glycerol saline consisting of 20% glycerol in 0.2 N NaCl ( Maugle, Deshimaru, Katayama & Simpson 1982). The tissues were then homogenized in a blender. The homogenized samples were centrifuged at 40 000 × g for 30 min in a Sorval RC 2B refrigerated centrifuge. The supernatant containing the enzymes (enzyme extract) was saved in several 2-mL vials and stored at −80 °C until used for enzyme assay.
Protein content of the enzyme extract was determined by the method of Lowry, Rosenberg, Farr & Randal (1951). Protease, collagenase, chitinase, amylase, cellulase and lipase, as well as alkaline and acid phosphatases, were measured for their specific activity (unit activity mg−1 of tissue protein) using specific chromogenic substrates. The absorbency of the liberated chromogen was measured and compared with an enzyme of known specific activity using a Beckman DU 70 spectrophotometer. The enzymes assayed, substrates used, pH at which the enzymes were assayed and bibliographic citation for each assay are shown in Table 1.
| Enzyme | Substrate used | pH | Reference |
|---|---|---|---|
| Protease | |||
| Serine | BAPNA * | 8.2 | Erlanger, Kolkowsky & Cohen (1961) |
| Aspartic | Haemoglobin | 4.7 | GCI, Indiana ME 400-10 † |
| Collagenase | DNP-peptide | 7.5 | Masui, Takemoto, Sakakibura, Hori & Nagai (1977) |
| Chitinase | Chitin azure | 4.4 | Hackman & Goldberg (1964) |
| Amylase | |||
| Alkaline | Amylose azure | 7.4 | Rinderknecht, Wilding & Haverback (1967) |
| Acidic | Amylose azure | 5.4 | Rinderknecht et al. (1967) |
| Cellulase | Cellulose azure | 4.8 | Fernley (1963) |
| Lipase | β-Naphthyl nonanoate | 7.4 | Snell & Snell (1971) |
| Phosphatase | |||
| Alkaline | p-Nitrophenyl phosphate | 10.3 | Sigma Diagnostics |
| Acidic | p-Nitrophenyl phosphate | 4.8 | Sigma Diagnostics |
- *BAPNA, N-benzoyl- dl-arginine-4-nitroanilide.
- † Genencor International Indiana, Inc., procedure no. ME 400-10.
Results
The comparison of digestive enzymes in threadfin and trevally presented in this study is appropriate because both species were provided with the same diet, thus eliminating any variation caused by enzyme/substrate specificity. Measurements of specific activity of enzymes extracted from threadfin and trevally stomach are shown in Table 2; enzymes extracted from pyloric caeca are shown in Table 3; and enzymes extracted from the intestine are shown in Table 4. Aspartic proteases seem to play a major role in protein digestion in both threadfin and trevally. Specific activity of aspartic proteases increased 3 h after feeding by nearly 4.5 times in the stomach, 1.2 times in the pylorus and by 5.6 times in the intestine of threadfin. Aspartic protease activity in the stomach of threadfin was also complemented by a shift in pH from 7 to 4 from prefed to fed status. In trevally, aspartic proteases increased in specific activity by 3.7 times in the stomach, twice in the pylorus and four times in the intestine 3 h after feeding. In fed trevally, with the exception of the stomach ,where the pH shifted from 7 to 6, the pH remained unchanged at 7 in the mid-gut.
| Pacific threadfin | Bluefin trevaly | |||
|---|---|---|---|---|
| Enzyme | Prefed (pH 7) | Fed (pH 4) | Prefed (pH 7) | Fed (pH 6) |
| Protease | ||||
| Serine | 23 106 ± 203 | 17 311 ± 143 | 418 ± 7 | 419 ± 2 |
| Aspartic | 11 482 ± 1001 | 51 746 ± 1399 | 14 195 ± 35 | 52 031 ± 294 |
| Collagenase | 84 ± 5 | 13 ± 2 | 57 ± 1 | 48 ± 4 |
| Chitinase | 46 ± 17 | 61 ± 5 | 8 ± 3 | 0 ± 0 |
| Acidic | 0 ± 0 | 0 ± 0 | 0 ± 0 | 0 ± 0 |
| Cellulase | 0 ± 0 | 0 ± 0 | 0 ± 0 | 0 ± 0 |
| Lipase | 4 ± 0 | 4 ± 0 | 0 ± 0 | 0 ± 0 |
| Phosphatase | ||||
| Alkaline | 624 ± 3 | 489 ± 4 | 124 ± 3 | 72 ± 1 |
| Acidic | 268 ± 2 | 227 ± 3 | 59 ± 0 | 46 ± 1 |
- Values are average of two assays (± SD). The pH measured under prefed and fed conditions is shown in parentheses. pH measurements given are pH values measured in the lumen of the stomach.
| Pacific threadfin | Bluefin trevally | |||
|---|---|---|---|---|
| Enzyme | Prefed (pH 8) | Fed (pH 7) | Prefed (pH 7) | Fed (pH 7) |
| Protease | ||||
| Serine | 238 526 ± 1079 | 462 418 ± 238 | 12 043 ± 132 | 19 012 ± 247 |
| Aspartic | 44 401 ± 1170 | 53 393 ± 383 | 28 094 ± 168 | 54 615 ± 1006 |
| Collagenase | 661 ± 39 | 3211 ± 75 | 463 ± 6 | 562 ± 5 |
| Chitinase | 2 ± 0 | 5 ± 2 | 3 ± 1 | 2 ± 1 |
| Amylase | ||||
| Alkaline | 0 ± 0 | 0 ± 0 | 6 ± 1 | 13 ± 1 |
| Acidic | 0 ± 0 | 0 ± 0 | 3 ± 0 | 7 ± 2 |
| Cellulase | 0 ± 0 | 0 ± 0 | 5 ± 0 | 9 ± 1 |
| Phosphatase | ||||
| Alkaline | 13 805 ± 4 | 32 535 ± 63 | 815 ± 4 | 681 ± 7 |
| Acidic | 1909 ± 19 | 1814 ± 53 | 173 ± 1 | 266 ± 7 |
- Values are average of two assays (± SD). The pH measured under prefed and fed conditions is shown in parentheses. pH measurements given are pH values measured in the lumen of the pyloric caeca.
| Pacific threadfin | Bluefin trevally | |||
|---|---|---|---|---|
| Enzyme | Prefed (pH 7) | Fed (pH 7) | Prefed (pH 7) | Fed (pH 7) |
| Protease | ||||
| Serine | 189 018 ± 3961 | 596 789 ± 1571 | 1179 ± 1 | 2542 ± 58 |
| Aspartic | 18 116 ± 1626 | 101 723 ± 6105 | 5019 ± 48 | 15 167 ± 492 |
| Collagenase | 117 ± 5 | 3981 ± 351 | 90 ± 1 | 321 ± 14 |
| Chitinase | 1 ± 0 | 4 ± 1 | 1 ± 0 | 1 ± 1 |
| Amylase | ||||
| Alkaline | 0 ± 0 | 0 ± 0 | 2 ± 0 | 5 ± 2 |
| Acidic | 0 ± 0 | 0 ± 0 | 1 ± 0 | 2 ± 1 |
| Cellulase | 0 ± 0 | 0 ± 0 | 0 ± 0 | 0 ± 0 |
| Lipase | 40 ± 2 | 26 ± 0 | 2 ± 0 | 3 ± 0 |
| Phosphatase | ||||
| Alkaline | 10 544 ± 86 | 19 852 ± 162 | 641 ± 1 | 1485 ± 24 |
| Acidic | 2507 ± 9 | 1773 ± 19 | 134 ± 2 | 425 ± 1 |
- Values are average of two assays (± SD). The pH measured under prefed and fed conditions is shown in parentheses. pH measurements given are pH values measured in the lumen of the intestine.
The specific activity of serine proteases in threadfin far exceeded that found in trevally. For example, the specific activity of serine protease in the stomach of threadfin was nearly 500 times higher compared with trevally, although the pH condition (pH 4) in fed threadfin would not have been ideal for serine protease digestion, as manifested by a 25% reduction in its specific activity. The increase in specific activity in the mid-gut of fed threadfin was also consistently higher compared with trevally.
Collagenase was found in both threadfin and trevally. The increase in specific activity of collagenase from prefed to fed fish by nearly 4.8 times in the pylorus and 34 times in the intestine of threadfin indicates strong collagenolysis in the mid-gut region of this species. Differences in specific activity of collagenase between unfed and fed trevally remained unchanged in the stomach and increased only by factors of 1.2 and 3.6 in the pyloric caeca and intestine respectively.
Chitinase activity was found in the stomach, pylorus and intestine of both species. Little change in activity occurred in the stomach of threadfin from the fed to the unfed state, whereas an eightfold increase in activity occurred in the stomach in trevally. In contrast, specific activity in the pylorus and intestine of threadfin increased 2.75 and four times, respectively, but showed little change in trevally.
α-Amylase activity tested in the acid and alkaline pH was not detected in the digestive tract of threadfin. In trevally, specific activity for amylase tested in the alkaline and acid pH was absent in the stomach, highest in the pylorus and low in the intestine. Lipase specific activity was recorded in the stomach and mid-gut in threadfin, but the activity remained unchanged between prefed and fed in the stomach, and decreased in the mid-gut. In trevally, no lipase activity was detected in the stomach. Cellulase was not detected in the digestive tract of either fish. Other than the digestive enzymes, acid and alkaline phosphatases were found in much higher amounts in threadfin than in trevally.
Discussion
The differences in digestive enzyme capabilities for protein between threadfin and trevally appear to correlate well with observed feeding habits of these species in nature. Pacific threadfin are known to feed at all times during the day ( Kanayama 1973) and would probably exhibit digestive mechanisms that allow rapid breakdown and absorption of ingested protein sources. In contrast, bluefin trevally are crepuscular, feeding primarily at dawn and dusk ( Potts 1980), and would be more apt to have mechanisms that allow digestion to be spread out over time and ensure that nutrients are absorbed as efficiently as possible. Both aspartic and serine proteases of the threadfin were present in greater amounts and more widely distributed in the gut than those of the trevally, suggesting more efficient digestive enzyme capability in the former. Specific activities of these enzymes also increased dramatically in threadfin from prefed to fed status, indicating more rapid digestive action. A change in stomach pH from 7 in the prefed state to 4 in the fed state in the threadfin, followed by reduction in the intestine to pH 7, provides typical conditions for optimal peptic and tryptic digestion respectively. This pH pattern is consistent with that of most vertebrate digestive systems, in which hydrolysis by aspartic proteases (peptic hydrolysis) in the stomach is followed by serine proteases (tryptic hydrolysis) in the mid-gut region ( Stevens & Hume 1995). In trevally, pH decreased only slightly from 7 to 6 in the stomach and remained at 7 in the mid-gut from the prefed to the fed state, suggesting a dominant role of aspartic proteases in this species. Serine protease activity was noticeable only in the pyloric region of the trevally.
The distribution of other proteolytic enzymes and their changes in specific activity in fed threadfin emphasizes the fact that threadfin are adapted to a wider variety of prey organisms, differing in their chemical composition of ingested protein. In fact, diets of threadfin in Hawaiian waters consist of both fish and invertebrates, primarily penaeid and ciridian shrimps ( Kanayama 1973), while those of trevally consist almost exclusively of fish ( Sudekum 1984). Threadfin appears to digest collagen primarily in the intestine, as indicated by the 34-fold increase in specific activity of collagenase from the prefed to the fed state. Although collagenase was found in all regions of the gut, collagenase activity showed only a fourfold increase in the intestine. This also supports the more intensive and potentially rapid digestive capabilities of the threadfin compared with the trevally, including those for collagen. Chitinase is important for threadfin because crustaceans are an essential component of their diet. The observed pattern and distribution of chitinase activity in threadfin may optimize digestion of crustaceans.
Absence or failure to detect amylase in threadfin should be viewed with caution. Kohla et al. (1992) suggested that failure to detect amylase activity in several other freshwater and marine species may be related to the lack of appreciable carbohydrate sources in diets or foods fed. In this study, however, this does not appear to be the case, as both threadfin and trevally were fed the same diet. Carbohydrate-digesting enzymes other than amylase could be present in threadfin, which were not examined in this study. Cellulase was also not found in either species, a testimony to their exclusively carnivorous feeding habits. The specific activity of lipase in both pylorus and intestine of threadfin was much higher than that in trevally, although the increase from the prefed to the fed state was more pronounced in trevally. This may indicate an important role for fats as a non-protein energy source in the diet of trevally. In fact, an optimal pH environment for the digestion of lipids was prevalent throughout the entire digestive system of the trevally rather than of the threadfin.
Besides the digestive enzymes, specific activity of phosphatases, which are ubiquitous cellular enzymes, varied considerably between the two species. Alkaline and acid phosphatases have been found in the gut of many other species of fish, and their presence in several regions of the gut has been demonstrated histochemically in developing stages of fed and fasting turbot Scophthalmus maximus L. ( Cousin, Baudin-Laurencin & Gabaudan 1987). Specific activity for both alkaline and acid phosphatases was many times higher in threadfin than in trevally. Although phosphatases are not digestive enzymes, they remove phosphate groups selectively from amino acid side-chains that are covalently modified by the addition of phosphate. This reversible phosphorylation/dephosphorylation can greatly affect the biological activity of other enzymes ( Copeland 1996). A much higher level of phosphatases in the mid-gut of threadfin compared with trevally may also be associated with increased activity in the smooth muscles. Phosphatase presence in larger amount in threadfin could be related to the higher rate of turnover of food and more efficient detoxifying mechanisms ( Cousin et al. 1987 ) for removing undesirable feed components
In conclusion, the results of this study comparing the specific activities of digestive enzymes between prefed and fed adult Pacific threadfin and bluefin trevally indicate that these two species have different abilities to digest food ingredients. This appears to be consistent with adapted feeding habits and preferences in nature. Threadfin are more adapted to a wider range of food protein sources than the trevally, but appear to be less adapted than trevally to using the non-protein sources for metabolic purposes. Practical diet implications are that threadfin may require higher protein diets with lower levels of non-protein energy sources, although the digestive potential for simple sugars was not investigated. Trevally may require diets containing lower protein levels, with the ability to use dietary fats and complex carbohydrates to supply metabolic energy and spare protein for growth and production. Future studies on practical diet development for these species will confirm or deny the validity of these observations.
Acknowledgments
The research reported here was supported by a grant from the National Oceanic and Atmospheric Administration (NOAA), award no. NA76FV0539. The views expressed are those of the authors and do not necessarily reflect the view of NOAA or any of its subagencies.
