Fisheries Science

, Volume 78, Issue 6, pp 1309–1314

Potent cellulase activity in the hepatopancreas of mangrove crabs


    • Laboratory of Aquatic Product Utilization, Graduate School of AgricultureKochi University
  • Kentaro Toriyama
    • Laboratory of Aquatic Product Utilization, Graduate School of AgricultureKochi University
  • Tamaki Azekura
    • Laboratory of Aquatic Product Utilization, Graduate School of AgricultureKochi University
  • Katsuji Morioka
    • Laboratory of Aquatic Product Utilization, Graduate School of AgricultureKochi University
  • Prasert Tongnunui
    • Rajamangala University of Technology Srivijaya
  • Kou Ikejima
    • Laboratory of Aquatic Product Utilization, Graduate School of AgricultureKochi University
Original Article Chemistry and Biochemistry

DOI: 10.1007/s12562-012-0547-8

Cite this article as:
Adachi, K., Toriyama, K., Azekura, T. et al. Fish Sci (2012) 78: 1309. doi:10.1007/s12562-012-0547-8


Mangrove crabs play a crucial role in the carbon cycle in forests by consuming large amounts of mangrove litter, which is mainly composed of cellulose. However, the detailed mechanism of cellulose digestion remains to be elucidated. We tested endogenous hepatopancreatic cellulase activity in eight species of crabs, including three mangrove sesarmid crabs (Episesarma versicolor, Perisesarma indiarum, and Episesarma palawanense) native to Thailand. Endo-β-1,4-glucanase activity was significantly higher in the enzyme extract from mangrove crabs than in that from Japanese marsh crabs. A β-glucosidase assay revealed particularly high endo-β-1,4-glucanase activity for E. versicolor, whereas little activity was observed for the Japanese marsh crabs. In a zymogram analysis for endo-β-1,4-glucanase activity, endo-β-1,4-glucanase had a similar molecular mass (30.7–33.1 kDa) among the mangrove crabs, whereas various sizes (44.3–84.8 kDa) were found in Japanese crabs depending on the species. These results suggest that mangrove crabs efficiently digest cellulose endogenously.


MangroveCarbon cycleMangrove crabCellulaseEndo-β-1,4 glucanaseβ-1,4-Glucosidase


The mangrove ecosystem dominates most tropical coastlines. Recent studies on mangroves have confirmed the important role played by sesarmid crabs in the structure and function of these ecosystems. The feeding activities of the crabs enable organic matter, including mangrove litter, to be recycled within the forest [13], with various studies showing that 30–90 % of the annual litterfall in mangroves is consumed and buried by sesarmid crabs, regardless of the environment [46]. The sesarmid crabs pull the litter into their burrows for later degradation by microbes, and approximately half of the leaves are consumed immediately. The assimilation efficiency for carbon in litter when consumed by sesarmid crabs is estimated to be approximately 40–70 % [79]. Mangrove litter is a rich carbon source but is unattractive to other consumers in the environment because of its high carbon to nitrogen (C/N) ratio (often >100) [10]. Mangrove crabs physically shred the litter when feeding to increase the surface area of the fecal material, which facilitates microbial colonization for further degradation [7]. Furthermore, the C/N ratio of the feces is lower than that in the leaf litter, which indicates that it is converted into a more palatable microbial biomass upon consumption by the crabs [79]. Thus, the consumption of litter by the crabs plays a crucial role in mangrove ecosystems, especially in the carbon cycle. However, the digestive mechanism of mangrove leaves in these crabs has yet to be elucidated [1, 2].

The main component of litter is cellulose, which is a linear homopolymer that consists of glucose units linked by β-1,4 bonds. Cellulase is a general term for cellulolytic enzymes, which can be classified into three types on the basis of the mode of enzymatic action and substrate specificity: (1) endoglucanases (EC cleave amorphous sites of cellulose chains at random; (2) exoglucanases (EC and act on the nonreducing/reducing end of cellulose molecules to release cellobiose or glucose; (3) β-glucosidases (EC hydrolyze cellobiose or cello-oligomers to release glucose from the nonreducing ends [11]. It is generally accepted that cellulases are exclusively present in microorganisms, such as bacteria and fungi, that degrade plant litter. Although cellulase activity has been observed in herbivorous animals, including ruminants or termites, it has been typically thought to derive from symbiotic organisms [11]. However, recent data mainly obtained by genomic analysis has revealed that invertebrates contain endogenous cellulolytic enzymes. Endogenous cellulase genes were initially identified in the termite [12] and nematodes [13], but subsequent studies identified cellulase genes in a wide range of invertebrates, such as insects [14, 15], crustaceans [16] and echinoidea [17], and mollusks [18, 19].

Additional studies have reported endogenous cellulase activity in gecarcinid land crabs [2022], but to date little information is available on the endogenous cellulolytic activity of sesarmid crabs in mangroves [8, 23, 24], particularly in terms of the comparative biochemical properties of the cellulase, such as specific activity and molecular mass. Sesarmid crabs are also present in large numbers in temperate estuaries, where a salt marsh environment develops and plant leaves are available, but not throughout the year. Several sesarmids living in temperate regions also feed on plant materials, but these mostly consume a wider range of other food items, including detritus, algae, and animals (i.e., detritivores, omnivores, and carnivores) [2429]. This difference provides a framework for comparing the mechanisms of related species for digesting plant materials. We hypothesized that sesarmids from mangroves would show higher cellulase activity than those from temperate regions and that the cellulose would be produced by the sesarmids endogenously.

In the study reported here, we prepared a crude enzyme solution from various crabs, including sesarmid crabs from tropical mangroves (mangrove crabs) and a temperate estuary (Japanese marsh crabs), determined their cellulase activity, and performed a zymogram analysis to investigate the digestive mechanism of sesarmid crabs in mangroves.

Materials and methods


Individual crabs of the sesarmid crab species Episesarma versicolor and Perisesarma indiarum were collected in the mangrove forest located on the campus of Rajamangala University of Technology Srivijaya (Trang, Thailand). Episesarma palawanense crabs were purchased from a retailer in Samut Songkhram, Thailand. They had also been collected from the mangrove forest and surrounding brackish water ponds and were maintained in a cage for 24–48 h before sacrifice. Japanese marsh crabs were caught in the estuary of the Kodono river in Kochi, Japan and included three species of sesarmid marsh crabs (Chiromantes dehanni, C. haematocheir, and Parasesarma pictum) and two species of varunid crab (Helice tridens tridens and Hemigrapsus penicillatus). C. dehaani, C. haematocheir, and H. tridens tridens mainly inhabit the mud fringe of the upper-tidal zone with reed grass, whereas P. pictum, P. bidens, and H. penicillatus inhabit the inter-tidal mud and rock in the same estuary. These crabs were also maintained in a cage for 24 h before sacrifice. The data and the eating habits of the crabs are summarized in Table 1.
Table 1

Summary for crabs sampled in this study

Scientific name


Shell width (mm)

Body weight (g)

Sampling site

Feeding mode/habit



Episesarma versicolor


23 ± 0.8

9 ± 1.4

Mangrove forest in Trang, Thailand


Thongtham and Kristensen [8]


Perisesarma indiarum


31 ± 3.6

25 ± 7.9

Mangrove forest in Trang, Thailand


Boon et al. [23]


Episesarma palawanensea




Mangrove forest in Samut Songkhram, Thailand


Carpenter and Niem [24]

Helice tridens tridens


28 ± 3.6

15 ± 6

Kodono river in Kochi, Japan


Doi et al. [24]


Hemigrapsus penicillatus


23 ± 2.2

5 ± 1.2

Kodono river in Kochi, Japan

Deposit feeder/omnivore

Okkamoto and Kurihara [25]


Chiromantes dehaani


25 ± 2.2

11 ± 3.6

Kodono river in Kochi, Japan

Deposit feeder

Doi et al. [24]


Chiromantes haematocheir


25 ± 4

11 ± 5.9

Kodono river in Kochi, Japan

Deposit feeder

Kitami and Yoshimaru [26]


Perisesarma bidens


21 ± 0.9

7 ± 2.4

Kodono river in Kochi, Japan

Deposit feeder/herbivore

Mchenga and Tsuchiya [27], Poon et al. [28]


Parasesarma pictum


22 ± 2.6

6 ± 2

Kodono river in Kochi, Japan

Deposit feeder/herbivore

Mfiling and tsuchiya [29]


aThe sample solution for E. palawanense was a mixture of several hepatopancreases from individuals of this species. Therefore, we did not perform a statistical analysis on this sample

Preparation of crude enzyme solution

Experimental animals were euthanized by immersion in ice water, and the hepatopancreases were harvested and frozen at −80 °C until use. The frozen tissue was homogenized in 10 mM phosphate buffer (pH 7.0) and centrifuged (13,000 g) at 4 °C for 10 min. The obtained supernatant was used as crude enzyme solution.

Protein determination

The protein concentration of the crude enzyme solution was measured using the Bradford method; bovine serum albumin served as standard [30].

Endo-β-1,4-glucanase activity

Endo-β-1,4-glucanase activity was determined as the rate of production of glucose from carboxymethyl cellulose (CMC) as the substrate. A mixture of 5 μL of each enzyme solution and 50 μL of CMC solution (2 %) was prepared in 10 mM phosphate buffer (pH 7.0). After incubation at room temperature for 10 min, the reducing sugar that was generated was measured as glucose equivalents by the tetrazolium blue method of Jue and Lipke [31] using glucose as the standard. Absorption of the samples, standards, and blanks was read at 660 nm. The enzyme activity was evaluated as the amount of released sugar (glucose equivalents) (μmol)/reaction time (min)/amount of protein in the mixture (μg).

β-Glucosidase activity

β-Glucosidase was determined according to the rate of production of reducing sugars from cellobiose as the substrate. A mixture of 25 μL of enzyme solution, 50 μL of cellobiose (50 mM), and 25 μL of phosphate buffer (10 mM; pH 7.0) was incubated at 40 °C for 30 min. The reaction was stopped by the addition of 25 μL of 0.3 M trichloroacetic acid. Excess acid was neutralized with 5 μL of 2.5 M K2CO3. The released glucose was measured using the Wako Glucose Test (Wako, Osaka, Japan) according to the manufacturer’s protocol. Enzyme activity was evaluated as the amount of released glucose (μmol)/reaction time (min)/amount of protein in the mixture (μg).

Total cellulase activity

The apparent total cellulase activity was determined as the rate of production of glucose from microcrystalline cellulose. A mixture of 5 μL of enzyme solution and 50 μL of cellulose (4.4 μΜ) was incubated at room temperature for 10 min. The released glucose was determined using the Wako Glucose Test (Wako). Enzyme activity was evaluated as the amount of released glucose (μmol)/reaction time (min)/amount of protein in the mixture (μg).

Statistical analysis

Statistical analyses were performed using one-way analysis of variance (ANOVA) and post hoc Bonferroni multiple comparison tests.


The samples were mixed with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer without a reducing agent and separated at 4 °C by SDS-PAGE in a gel containing 0.2 % CMC. The gel was then washed for 30 min with 10 mM acetate buffer (pH 5.5) containing 0.1 % Triton X-100. After this process was repeated, the buffer was subsequently replaced with normal 10 mM acetate buffer (pH 5.5), and the gel was incubated for 30 min and then further incubated for 3 h at 37 °C for the digestion of CMC in the gel. The gel was alkalized by the addition of 0.1 N NaOH, incubated at room temperature for 30 min, and then stained with 0.1 % aqueous Congo red (Sigma-Aldrich, St. Louis, MO). The gel was then washed for 1 day with 1 M NaCl until cellulase activity was detected as clear zones in the gel. The molecular mass of the bands was calibrated by using TEFCO’s wide-range marker (TEFCO, Tokyo, Japan) as a standard.


Cellulase activity

Endo-β-1,4-glucanase activity

We detected endo-β-1,4-glucanase activity in all samples tested (Fig. 1a). The activity of the E. versicolor and P. indiarum samples was significantly higher than that of the Japanese crab samples. A similar value was obtained for E. palawanense, although the value was obtained from a mixture of samples (i.e., no statistical analysis could be performed.). The average specific endo-β-1,4-glucanase activity for E. versicolor, P. indiarum, and E. palawanense was 2.07, 2.45, and 2.86 reducing sugar (μg)/protein (μg), respectively.
Fig. 1

Endo-β-1,4-glucanase activity (a), β-glucosidase activity (b), and total cellulase activity (c) of a crude enzyme solution extracted from each crab species. The number below each panel represents the sample crab shown in Table 1. Values are the mean ± standard error (SE). Different letters above the columns indicate significant differences according to one-way analysis of variance (ANOVA) and post hoc Bonferroni multiple comparison tests (p < 0.01). The results for Episesarma palawanense were obtained from a mixture of hepatopancreases from several crabs; therefore, no statistical analysis was performed

β-glucosidase activity

Notable β-glucosidase activity was detected in the enzyme solutions from mangrove sesarmid crabs, especially E. versicolor, whereas the enzyme solutions from Japanese marsh crabs showed low enzyme activity (Fig. 1b). The specific activity of the E. versicolor enzyme solution (0.03 μg glucose released/μg protein) was at least sixfold higher than that of Japanese crabs (8 × 10−5–5 × 10−3 μg glucose released/μg protein). However, no significant difference was found in the comparison of P. indiarum and E. palawanense with Japanese crabs.

Total cellulase activity

Notable total celllulase activity was found in the enzyme solutions from mangrove crabs, especially P. indiarum, whereas negligible activity was found in the enzyme solutions from Japanese crabs and that from E. palawanense (Fig. 1c). A significant difference was found in the comparison of P. indiarum with other species, including mangrove crabs. The specific activity of cellulase in the P. indiarum enzyme solution was 5 × 10−3 μg glucose released/μg protein, which was significantly higher than the values observed in the other samples.

Zymogram analysis for endo-β-1,4-glucanase activity

The zymogram analysis of the enzyme solution of Japanese marsh crabs revealed clear bands at various positions, with the molecular mass ranging from 44.3 to 84.8 kDa depending on the species (Fig. 2a). Some minor bands were also found at a low molecular mass (<40 kDa). In the mangrove crab samples, similar band sizes were detected, including those at 33.1, 30.7, and 30.7 kDa for E. versicolor, P. indiarum, and E. palawanense, respectively (Fig. 2b). In the E. versicolor sample, a minor band (25.3 kDa) was also detected.
Fig. 2

Zymography for endo-β-1,4-glucanase activity in a crude enzyme solution obtained from each crab. a Crabs from a mangrove in Thailand, b crabs from a river in Japan. The number below the gel represents the sample crab shown in Table 1. The figure below the bands indicates the molecular mass calibrated by a protein marker. Each analysis was run under distinct conditions in terms of the amount of protein applied and the gel concentration: a 0.01 μg protein, 15 % gel; b 5 μg protein, 7.5 % gel


In this study, endo-β-1,4-glucanase activity was detected in all crabs tested, and the crude enzyme extract from the mangrove crabs E. versicolor, P. indiarum, and E. palawanense showed a significantly higher activity than that from Japanese marsh crabs. In addition, E. versicolor demonstrated significantly higher β-glucosidase activity than the other crabs. No significant difference was observed for the other two mangrove crabs, but they appeared to have higher β-glucosidase activity than the Japanese crabs. These data were supported by the finding of significantly higher total cellulase activity in P. indiarum. These lines of evidence indicate that mangrove crabs, but not Japanese crabs, can endogenously digest cellulose to produce glucose. Previous studies that quantified the ratio of leaf litter consumption to fecal production reported that mangrove crabs assimilate 40–70 % of the carbon present in litter [79]. In redclaw crayfish Cherax quadricarinatus, exposure to antibiotics resulted in a dramatic decline in the bacterial population in the gastric contents (>90 %), but only a 40 % decline in cellulase activity, implying that approximately half of the cellulose digestion in this crayfish depends on symbiotic bacteria [32]. Although we did not perform an enzymatic test for potential symbiotic bacteria, which may also partly contribute to cellulose digestion in crabs, our findings suggest that mangrove crabs endogenously digest cellulose to glucose in order to assimilate the latter.

From an ecological point of view, the partial digestion of litter with a high C/N ratio to provide a lower C/N ratio provides the basis for the detritus food chain in mangrove ecosystems [79]. Our findings provide evidence that the endogenous digestion of cellulose by sesarmid crabs participates in this process. Further studies are necessary to elucidate endogenous and symbiotic digestion in the mangrove crabs.

Parasesarma bidens and P. pictum are found in subtropical mangrove and consume mangrove litter [2729], but in our study they showed lower enzymatic activity than the mangrove crabs from Thailand (Fig. 1). Subtropical areas are subjected to pronounced seasonal changes; this results in less leaf litter being produced in subtropical areas than in tropical areas, especially in winter, and consequently in a relatively lower availability of mangrove litter [2729]. Thus, these crabs probably depend less exclusively on mangrove litter for food. This notion is supported by the findings of Mchenga and Tsuchiya [27] who showed that male P. bidens consume both mangroves litter and algae but prefer algae over litter. In addition, the Japanese marsh crabs in our study were sampled in an environment where less cellulose-related compounds are available compared with subtropical areas, as marsh plants grow only seasonally. Thus, our results reflect the likelihood that endogenous cellulose digestive systems may differ according to the dependence of the species on plant litter within the same genera/family of crabs. The Japanese marsh crabs showed a certain level of well-discernable cellulase activities, including endo-β-1,4-glucanase and β-glucosidase activities; however, they did not hydrolyze microcrystalline cellulose. These results suggest that these enzymes hydrolyze the polysaccharides of the cells walls of protozoans and algae that these crabs consume with deposit feeding. The relationship between enzyme activity and actual level and types of plant matter consumption remains to be studied.

Based on the zymogram analysis for endo-β-1,4-glucanase activity, the molecular mass of endo-β-1,4-glucanase was similar among the mangrove crabs (30.7–33.1 kDa), whereas various sizes (44.3–84.8 kDa) were observed in Japanese crabs, depending on the species. In crustaceans, the molecular size of endo-β-1,4-glucanase has been reported to be 47,887 and 50,295 Da for the redclaw crayfish C. quadricarinatus using time-of-flight-mass spectroscopy [33, 34], and as 31 and 27 kDa for the gecarcinid land crab Gecarcoidea natalis using zymogram analysis [22]. Thus, even in the same subphylum, the size of this molecule seems to vary; however, its size in mangrove crabs appears to converge (30.7–33.1 kDa). The high specific activity observed in our study thus suggests that the endo-β-1,4-glucanase in mangrove crabs may have convergently evolved to acquire its function. On the contrary, the molecular mass of this enzyme, and therefore the molecular structure, varied in Japanese marsh crabs. The origin and phylogeny of this enzyme remain unclear, but molecular genetic techniques will clarify how crabs obtained and evolved this enzyme in the mangrove forest, where organic matter is largely recycled via litter consumption.

In conclusion, our results reveal that mangrove sesarmid crabs in Thailand endogenously digest cellulose efficiently to glucose and produce feces with a high C/N ratio that can be decomposed more efficiently by consumers. The results from the zymogram analysis suggest the possible convergent evolution of endo-β-1,4-glucanase in mangrove crabs, but not in Japanese marsh crabs.

Copyright information

© The Japanese Society of Fisheries Science 2012