Fisheries science JSFS , tập 78, số 2, 2012 3

Fish Sci (2012) 78:209–219 DOI 10.1007/s12562-011-0458-0 ORIGINAL ARTICLE Fisheries Bottom shrimp trawling impacts on species distribution and fishery dynamics; Ungwana Bay fishery Kenya before and after the 2006 trawl ban Cosmas Munga • Stephen Ndegwa • Bernerd Fulanda • Julius Manyala • Edward Kimani • Jun Ohtomi • Ann Vanreusel Received: 10 September 2011 / Accepted: 14 December 2011 / Published online: February 2012 Ó The Japanese Society of Fisheries Science 2012 Abstract The Malindi–Ungwana Bay fishery Kenya is one of the most important marine fisheries of the Western Indian Ocean There are two fishing grounds: Formosa and Malindi, with a designated 5-nM no-trawl zone offshore However, the fishery was faced with numerous resource use conflicts and a decline in catches, culminating in a trawl ban in 2006 This study analyses catches and fishery dynamics before and after the 2006 trawl ban Results show that artisanal landings declined before the ban, but rapidly recovered within years after the ban was imposed However, shrimp landings in the artisanal fishery remain low Commercial shrimp landings gradually declined before the ban: *550 t in 2001 to 250 t in 2006, and the shrimp: fish bycatch ratio was 1:1.5 compared C Munga Á A Vanreusel Marine Biology Section, Ghent University, Krijgslaan 281, S8, 9000 Ghent, Belgium C Munga Á B Fulanda (&) Á E Kimani Kenya Marine and Fisheries Research Institute, P.O Box 81651, Mombasa 80100, Kenya e-mail: bernfulanda@yahoo.com S Ndegwa Fisheries Department-Kenya, P.O Box 90423, Mombasa 80100, Kenya B Fulanda The United Graduate School of Agricultural Sciences, Kagoshima University, Korimoto 1-21-24, Kagoshima 890-0065, Japan J Manyala Department of Fisheries and Aquatic Sciences, MOI University, P.O Box 1125, Eldoret 30100, Kenya J Ohtomi Faculty of Fisheries, Kagoshima University, Shimoarata 4-50-20, Kagoshima 890-0056, Japan to early reports of 1:7 in 1999 SIMPER analyses shows that and 16 families (groups) accounted for 91.0 and 90.2% of the similarity in catch within the Formosa and Malindi fishing grounds, respectively Formosa was important for Claridae, Cichlidae and Protopteridae, while Malindi recorded Carangidae, Siganidae, Carcharhinidae and Lethrinidae as the main families Future studies should therefore embark on analyses of the factors driving the spatio-temporal distributions of the species and assess the impacts of bottom trawling on fishery dynamics before the trawl ban can be lifted Keywords Malindi–Ungwana Bay Kenya Á Bottom trawl Á Artisanal fishery Á Catch per unit effort Á SIMPER analysis Introduction The Malindi–Ungwana Bay complex, Kenya, comprises the larger Ungwana Bay extending from Ras-Shaka in the north of Kipini to Ras Ngomeni in the south, and the smaller Malindi Bay, which straddles the mouth of the Athi River at Sabaki off the Eastern coast of Africa (Fig 1) The complex, commonly referred to as the Ungwana Bay, is part of the wider Western Indian Ocean (WIO) Ecoregion The continental shelf is narrow, extending up to only 60 km offshore, and the fishing grounds are shallow, averaging 12–18 m deep at 1.5 and 6.0 nM offshore [1] However, the waters provide rich fishing grounds both inshore and offshore, and are home to a commercial bottom trawl fishery as well as resident and migrant artisanal fishery sectors Two main rivers, the Athi and the Tana, drain into the Malindi and Ungwana bays, respectively, and thus enrich the waters of this complex and the associated fisheries The Malindi–Ungwana Bay commercial bottom trawl fishery is restricted to the 5–200 nM waters, while the 123 210 Fig Map of the Eastern Coast of Africa showing the location of the study site: the entire Malindi–Ungwana Bay Kenya and demarcation of the Formosa and Malindi fishing grounds of the commercial bottom trawlers resident-migrant artisanal fishery exploits the 0–5 nM Therefore, the 0–5 nM waters are designated as a trawl exclusion zone (TEZ) [2], setting an arbitrary area-based resource-use guide for the various fisheries of this important bay The fishing grounds are some of the most productive and extensive shrimping areas on the East African coast [3, 4] Consequently, fisheries remain an important source of livelihood for the coastal fisher communities of East Africa [1, 5] The commercial bottom trawl fishery dates back to the early 1970s and is Kenya’s only marine commercial shrimp fishery [6–9] The fishery targets five main penaeid species: Fenneropenaeus indicus H Milne Edwards, P monodon Fabricius, Metapenaeus monoceros Fabricius, P semisulcatus De Haan and Marsupenaeus japonicus Bate The fishing fleet is mainly comprised of industrial trawlers that range in size from 25 to 40 m long and 115–1,500 horsepower engines equipped with blast freezers and freezing holds with 30–350 t storage capacity [3] The trawlers employ double-rigged, stern or outrigger trawling as the predominant method of fishing, with funnel-shaped otter trawl gear mostly towed behind the vessels [1] The nets are made of polypropylene with 50–55 and \40 mm diamond mesh sizes at the body and cod end, respectively 123 Fish Sci (2012) 78:209–219 On the other hand, the resident-migrant artisanal fishery has been in existence for several hundreds of years and is closely associated with trade dhows dating back to the sixteenth century Arab invasion of the East Africa Coast [5, 10] The vessels used in the artisanal fishery are mainly traditional crafts including mtumbwi, hori, ngalawa and dau, which account for more than 40% of the vessels in the fishery The mtumbwi are dug-out canoes measuring about m long with curved bottoms On the other hand, the hori and ngalawa are canoes made of plankwood, but differ in that the ngalawa are fitted with outriggers [1] The dau is a flat bottom, plankwood vessel propelled by small sails Other fishing crafts such as mashua and jahazi employ dual modes of propulsion, including inbuilt engines and lateen sails, and account for \20% of the fishing vessels [5] The mashua are mainly used for out-of-reef fishing and employ sails as the main mode of propulsion, while the jahazi are the preferred fishing crafts for open-sea fishing and transportation of cargo [1] The artisanal fishery is mainly based on traditional fishing gear comprising homemade basket (malema) traps, intertidal fixed weir (uzio) traps made of sticks, spear guns (bunduki) made from wood and some rubber bands, and wooden spears (ngovya) for octopus and crab fishing [1] Modern gear in the fishery is limited to gillnets, drift nets, beach seines, handlines and longlines Sardine nets (kimia) with \5 cm mesh sizes are used to target the small-sized sardine species [1, 11] Worldwide, resource use conflicts between artisanal and commercial fisheries date back several centuries As early as the late fourteenth century, Jones [12] identified historical complaints about bottom trawling by artisanal fishermen, including indiscriminate harvesting of undersized and nontarget species in a deteriorating artisanal fishery in New Zealand In the Malindi–Ungwana Bay fishery, conflicts over resource use and partitioning between the artisanal and the commercial bottom trawl fisheries escalated before the trawl ban in 2006, augmented by undefined harvest strategies and an increase in the use of deleterious fishing practices over the years [1, 5, 13] These problems threaten the very livelihoods of the coastal fisher communities depending on these fisheries Moreover, the impacts of bottom trawling on target and non-target species, and the damage to habitats and the associated benthic biota among others cannot be ignored [12, 14– 18] To date, this long-established technique of bottom trawling continues to attract increasing criticism for both the perceived damage to the environment and to the fisher livelihoods it may cause, especially with conflicts over the partitioning of fishing grounds Many governments have devised harvest strategies incorporating seasonal bans and restricted fishing grounds, while others have banned bottom trawling altogether Such management strategies have helped the recovery of the affected fisheries and associated marine resources For example, while assessing the effects of a 1978 Fish Sci (2012) 78:209–219 sustained ban on trawling in an Indonesian shrimp fishery, Chong et al [19] reported that the over-fished stocks showed recovery within a 7-year period With this background, the Kenyan government suspended bottom trawling in the Malindi–Ungwana Bay in 2006 when resource use conflicts with artisanal fishers over perceived declining catches, habitat impacts and destruction of artisanal fishing gear by the trawlers escalated because of the continuous encroachment on the artisanal fishing grounds by the commercial vessels This encroachment is partly attributed to the higher abundance of the target shrimp species in the 3–5 nM waters [1] However, information on the status of the stocks and the biology of the species, including growth, reproductive cycles and feeding ecology, was still lacking, leading to an indefinite trawl ban in 2006 Consequently, conducting extensive research was necessary in order to provide the much needed data and information on the species for definition of sustainable resource exploitation strategies Therefore, a number of scientific trawl surveys were conducted, including the 2002 study by the Kenya Marine and Fisheries Research Institute (KMFRI-2002); the 2003 study by the Department of Fisheries and Aquatic Sciences, MOI University (DFASMOI, 2003); the 2009 trawl surveys under the Kenya Coastal Development Project (KCDP-2009); and the ongoing South West Indian Ocean Fisheries (SWIOF, 2010–2011) project These studies have gone a long way in ensuring protection, management and development of the marine and coastal ecosystems of the Eastern Africa Region as outlined in the UNEP Nairobi Convention, 2010 This study investigated the situation in the Malindi– Ungwana Bay fishery, looking at the trends in both the artisanal and commercial bottom trawl fisheries before and after the trawl ban in 2006 The study examined shrimp landings and retained fish bycatch in the commercial bottom trawl fishery during 2001–2006, and trends in landings from the artisanal fishery during the 2001–2006 pre-trawl ban period and the 2006–2008 no-trawl years Specifically, the study compared landings in both the commercial and the artisanal fisheries to investigate the temporal and seasonal and, bathymetric and spatial variations in the commercial bottom trawl fishery shrimp CPUE, and to model the spatial and temporal differences in composition of the artisanal landings The results of the study provide a baseline for future scientific assessments of the impacts of bottom trawling in the Malindi–Ungwana Bay fishery Materials and methods Study area The Malindi–Ungwana Bay complex extends along a 210-km coastal stretch running from Malindi town in the 211 south to Ras-Shaka in the north of Kipini (Fig 1) The bay straddles 2° 300 S and 3° 300 S, and longitudes 40° 000 E and 41° 000 E The fishery is resource-partitioned into a 0–5 nM TEZ artisanal fishery and a commercial bottom trawl fishery exploiting the 5–200 nM exclusive economic zone (EEZ) (Government of Kenya, 2008) The commercial trawling grounds are amorphously divided into three main areas: (a) Malindi shallow, lying off the Malindi Bay, (b) Ngomeni, running from Ras-Ngomeni to the waters off Mto-Tana, and (c) Kipini, covering the shrimping grounds off Mto-Tana to the waters off Ras-Shaka [1, 4] The fishing grounds cover an estimated 35,300 km2, but the coastline is characterized by fringing reefs with occasional outcrops, thus limiting the effective trawlable grounds to about 20,000 km2 [1, 4] Most of the trawling is conducted in waters shallower than 70 m [3, 7] The Tana and Athi Rivers drain into the bay, adding terrigenous sediments [20, 21] Like the rest of the East African coast, the bay experiences a tropical humid climate with two distinct seasons: the dry Northeast monsoon (NEM) season (October– March) and the wet Southeast monsoon (SEM) season (April–September) [5, 22] These seasons greatly influence the productivity of the marine and coastal fisheries as well as the fishing patterns along the coast [1, 22] Data collection In the commercial bottom trawl fishery, sampling surveys were conducted during 2001 through 2006 During this period, increased resource use conflicts led the government to impose stiffer legislations on the commercial bottom trawl fishery, including the need to utilize discarded bycatch and reduction of fishing effort by imposition of a ban on night trawling Further, the vessels were installed with mandatory turtle excluder devices (TEDs) and vessel monitoring systems (VMS) equipment A monitoring program was also initiated to assess the fishing activities of commercial bottom trawlers using onboard data collectors from the Fisheries Department (FD-Kenya) and KMFRI Data collected included coordinates of the fished areas, water depths, catch of target shrimp species and retained bycatch, tow and haul durations, and the number of hours fished each day Further, the quantity of discarded bycatch including debris was estimated by sampling each haul and extrapolating to the overall haul size Due the nature of the fishing activities of the commercial trawlers augmented by limited storage onboard the vessels, only a few discardedbycatch hauls could be selected for analysis of species composition This analysis would provide a quick assessment of its potential impacts and a baseline for future assessment of bycatch discards in the commercial bottom trawl fishery 123 212 In the artisanal fishery, data were collected using FDKenya data collectors at designated fish landing sites and villages (Government of Kenya, unpublished data, 1991) in 2001-2008 Moreover, the precision of data reporting in these designated sites has been enhanced by the recently initiated beach management units (BMUs) at the landing sites and villages under a community-based fisheries management program of the FD-Kenya Data collected included daily catch by species and fishing grounds, sizes and types of gear and vessels, number of fishers per vessel, and age and experience (duration of years fished) of the fishers However, wide variations in the fishing vessel design and size, gear types and numbers per vessel and demographic factors, including age and experience of the fishers, were evident in this fishery Consequently, precise data needed for standardization of the fishing effort in the artisanal fishery would require extensive manpower Available data showed very wide disparities for similar vessels, and the number of fishers per vessel varied daily even for the same vessels in addition to variations in vessel age, materials, and mode of construction and propulsion A similar observation was noted on the number and types of fishing gear onboard the vessels, and the age of the fishers in each vessel and hours fished each day These variations present numerous challenges for standardization of the fishing effort in the artisanal fishery, and therefore standardization of the fishing effort and analysis of CPUE in the artisanal fishery was considered of secondary importance in the present study Fish Sci (2012) 78:209–219 bycatch species by fishing area and depth were analyzed by zoning the Formosa and Malindi fishing grounds into ‘‘shallow’’ (\25 m) and ‘‘deep’’ ([25 m) Further, the spatial-bathymetric distribution of the shrimp stocks in the bay were assessed by analysis of the shrimp CPUE of the commercial bottom trawl fishery for the Formosa and Malindi fishing grounds, and by bathymetric zones comparing the shallow and the deeper fishing grounds All tests were considered significant at a probability level of p \ 0.05 (95% confidence) To assess the impacts of the bottom trawling, the artisanal fishery 2001–2008 catch data were analysed for differences in spatial and temporal composition in taxa or fishery groups and abundance using the non-metric multidimensional scaling (MDS) technique in CAP4 software Further, two-way analysis of similarity (ANOSIM) [23] was used to test for differences across years and fishing areas, while two-way similarity percentage (SIMPER) analysis [23] was used to identify the dominant taxa or taxon group contributing to similarity and dissimilarity within and between the fishing grounds during 2001–2008 Both the ANOSIM and SIMPER use the Bray–Curtis [24] measure of similarity The SIMPER analysis breaks down the contribution of each taxon to the observed similarity (or dissimilarity) between samples and allows identification of taxa that are most important in creating the observed pattern of similarity Results Data analysis Trends in fisheries landings Data analysis was conducted using MS Excel and Community Analysis Package 4.0 (CAP4-Pisces Conservation) software In the commercial bottom trawl fishery, analysis was conducted assuming a variant species system targeting shrimps only, and fish were considered only as bycatch The fishing effort in this fishery was expressed as the average hours fished within a 24-h day and the catch per unit effort (CPUE) expressed as kg/h Further, the ratio of catch of the target species against bycatch was calculated using the total retained catch for each haul In the fishery, discarding low value fish is common However, in the present study, the discarded bycatch quantities were not included in the analysis because of the quality of the estimations of this portion of the catch onboard the commercial vessels Prior to analysis, all data were tested for homogeneity (Levene test), and where necessary, they were normalized using the fourth-root transformation Spatio-temporal variations in CPUE were analysed using two-way ANOVA to test for significance differences between years and seasons, and between fishing areas Variations in spatial distribution of the target shrimp and 123 In the artisanal fishery, the annual landings of both fish and shrimp generally oscillated, with no discernible trends during the study period In this fishery, the annual landing of shrimps ranged from 71.5–187.1 t during 2001–2008, with the highest landings recorded during 2004 (Fig 2) The annual fish landings averaged at 885.4–1540 t and showed an increase from 2006, reaching a peak in 2008 The combined fish and shrimp landings in the artisanal fishery averaged 1,013.7–1,653.2 t during 2001–2008 Unlike the artisanal fishery, the commercial bottom trawl fishery showed a clear downward trend, and shrimp catches declined by more than 50% during 2001–2006: from 553.7 t in 2001 to 257.3 t in 2006 During the same period, the retained bycatch was 432.0 t in 2001, increasing to 602.3 t in 2004, but declined to 315.6 t in 2006 before the trawl ban The combined fish and shrimp landings during 2001–2006 averaged at 572.9–985.6 t, which is far lower than the artisanal fishery landings The mean ratio of the target shrimp catch to the retained bycatch was 1:1.5 The mean shrimp CPUE ranged between 42.95 ± 4.6 kg/h Fish Sci (2012) 78:209–219 213 17.3 ± 1.6 kg/h in the 2003 SEM seasons However, Tukey’s HSD post hoc tests only revealed significant differences in CPUEs between 2001 versus 2003, 2004 and 2005 NEM and SEMs, and 2001 NEM versus 2006 SEM; NEM and SEM in 2002 versus 2004 and 2005 NEM; 2002 SEM versus 2003 NEM; 2004 NEM versus both NEM and SEM in 2006; and the NEM seasons of 2005 versus 2006 (Fig 3) Spatial-bathymetric distribution of shrimp in the bay Fig Annual landings of shrimp and fish in the artisanal and commercial bottom trawl fisheries of the Malindi–Ungwana Bay Kenya The overall mean CPUEs varied by fishing area The Formosa ‘‘shallow‘‘ and ‘‘deep’’ recorded 31.2 ± 0.4 and 24.3 ± 1.8 kg/h compared to 21.8 ± 0.9 and 23.5 ± 0.7 kg/h in Malindi ‘‘shallow’’ and ‘‘deep’’, respectively Results of twoway ANOVA for the spatial-bathymetric distribution of the shrimp stocks showed significant differences by fishing area, and the Formosa grounds recorded higher CPUEs than the Malindi fishing grounds (p \ 0.05) (Fig 4) There were no significant differences in mean CPUE between the shallow and deep bathymetric zones (p = 0.17), although the Malindi fishing grounds recorded generally higher shrimp CPUE in the deeper bathymetric zones compared to the shallow zones On the contrary, the shallow bathymetric zones in the Formosa fishing grounds recorded higher CPUEs than the deep zones (Fig 4) Variations in species composition in the artisanal fishery Fig Annual and seasonal trends (fourth-root transformed) in shrimp CPUE (kg/h) in the commercial bottom trawl fishery Malindi–Ungwana Bay from 2001 to 2006 when the ban on bottom trawling was effected in the fishery recorded in 2001 and 15.76 ± 5.1 kg/h recorded in 2004 Two-way ANOVA tests for differences in the shrimp CPUE of the commercial bottom trawl fishery during 2001–2006 showed a highly significant difference between the years (p \ 0.05) Tukey’s HSD post hoc test showed significant differences between all years (p \ 0.05) except 2001 versus 2002, 2002 versus 2006, 2003 versus 2004 and 2005, and between 2004 versus 2005 Seasonal variations in CPUE between the NEM and SEM seasons within years were not significantly different However, highly significant differences were evident between seasons across the years (p \ 0.05) The mean CPUEs by season ranged between 47.6 ± 5.6 kg/h recorded in the 2001 to 10.7 ± 5.8 kg/h in the 2004 NEM seasons compared to 38.3 ± 3.4 kg/h in the 2001 and In 2001–2008, a total of 29 fish families and two ecological groups—‘‘mixed pelagic’’ and ‘‘mixed demersal’’ comprising small-sized pelagic and demersal species or species of low commercial/food value, respectively were identified and used for ordination analysis of the artisanal fishery Fig Spatial and bathymetric variation in shrimp CPUE (95% confidence interval) in the Malindi–Ungwana Bay commercial bottom trawl fishery The error bars show the mean (circle/square) ± standard deviation (SD) 123 214 Fish Sci (2012) 78:209–219 The mixed pelagic and mixed demersal groups are often landed by the artisanal fishers for food fish Results of the non-metric MDS on the composition of annual landings in the fishing grounds of the bay showed distinct differences across years and fishing areas (Fig 5) The non-metric Fig Non-metric multidimensional scaling (NMDS) catch composition data (by family) [Log10(X ? 1) transformed data] in catches of the artisanal fishery in the Formosa and Malindi fishing areas during 2001–2008 data (stress level 0.0095) Table SIMPER analysis of the artisanal fish landings across 2001–2008 showing the fish groups/families contributing to about 91.2% similarity within the (a) Formosa and (b) Malindi fishing grounds of Ungwana Bay, Kenya Group MDS showed similarity in the species composition between the 2001–2008 years, with the Malindi fishing grounds showing higher similarity over the years compared to the Formosa fishing grounds (Fig 5) Two-way ANOSIM analysis for differences across years and fishing areas showed significant differences across 2001–2008 in the Formosa and Malindi fishing areas (p \ 0.05) Further, two-way SIMPER analysis showed 78.6 and 77.5% average similarity where 91 and 90.2% within-area similarity was attributed to and 16 species in the Formosa and Malindi fishing grounds, respectively The variations in taxonomic composition of the artisanal landings were attributed to a higher abundance of brackish water families, including Claridae, Cichlidae and Protopteridae, in the Formosa fishing areas, while the Malindi grounds were dominated by the two ecological groups, mixed demersal and mixed pelagic species, and the families Carangidae, Siganidae, Carcharhinidae and Lethrinidae (Table 1a, b) The six taxonomic/ecological groups also accounted for higher contribution to the artisanal landings in the whole fishery Penaeid shrimps accounted for\1.5% of the combined artisanal fishery landings Furthermore, Aver abundance Aver similarity % Contribution Claridae 167.9 32.8 41.7 Cichlidae 95.9 17.7 22.5 Protopteridae 64.2 13.1 16.7 (a) Formosa area Penaeidae 19.9 3.1 4.0 Carcharhinidae 13.9 2.5 3.2 Mixed demersals 15.7 2.4 3.0 91.0 (b) Malindi area The average similarity across 2001–2008 was 78.6 and 77.5% for the Formosa and Malindi fishing grounds, respectively 123 Mixed demersals 151.2 13.5 17.5 Mixed pelagics Carangidae 123.9 91.8 10.7 7.1 13.8 9.2 Siganidae 53.1 4.8 6.1 Carcharhinidae 61.8 4.4 5.7 Lethrinidae 51.4 4.4 5.7 Penaeidae 36.2 3.3 4.3 Istiophoridae 36.6 3.2 4.1 Lutjanidae 37.3 3.1 4.1 Mugilidae 36.4 3.1 4.0 Scombridae 37.0 2.5 3.2 Acanthuridae 29.1 2.0 2.6 Serranidae 20.6 1.7 2.2 Octopodiformes 19.7 1.6 2.1 Scaridae 19.2 1.6 2.0 Clupeidae 17.6 1.5 1.9 Palinuridae 16.3 1.4 1.8 90.2 Fish Sci (2012) 78:209–219 215 Table SIMPER analysis of the artisanal fish landings during 2001–2008 showing the fish groups/families contributing to about 90.8% dissimilarity between the Formosa and Malindi fishing grounds of Ungwana Bay, Kenya Family/group Formosa Average Claridae 167.9 15.7 Mixed pelagics 0.8 123.9 Cichlidae 95.9 0.0 7.0 8.2 2.9 91.8 6.5 7.7 Mixed demersals Carangidae Protopteridae Siganidae Lethrinidae Malindi Abundance Average dissimilarity % contribution 0.0 12.2 14.4 151.2 9.7 11.4 8.8 10.4 64.2 0.0 4.7 5.5 2.5 53.1 3.6 4.3 2.2 51.4 3.5 4.1 13.9 61.8 3.5 4.1 Istiophoridae 1.3 36.6 2.5 3.0 Mugilidae 3.8 36.4 2.4 2.8 Scombridae 7.7 37.0 2.3 2.7 Lutjanidae 5.7 37.3 2.3 2.7 Acanthuridae 0.4 29.1 2.0 2.4 Chanidae 0.8 20.6 1.4 1.6 Serranidae 1.9 20.6 1.3 1.6 Sphyraenidae Scaridae 1.5 1.5 20.0 19.2 1.3 1.3 1.5 1.5 Clupeidae 0.0 17.6 1.3 1.5 Carcharhinidae 91.3 The average dissimilarity between the composition of the artisanal catch landings of the Formosa and Malindi fishing grounds during 2001–2008 was 84.8% SIMPER analysis for species composition in the Formosa and Malindi fishing grounds revealed that 19 species accounted for 91.3% dissimilarity between the two fishing grounds, and the target penaeid shrimp species were absent from this group, contributing only \1.5% to the dissimilarity between fishing grounds The overall average dissimilarity between the Formosa and Malindi fishing grounds was estimated at 84.8% (Table 2) Discussion Results of this study show that the combined fish and shrimp landings were higher in the artisanal fishery than in the commercial bottom trawl fishery However, it should be noted that the commercial bottom trawl fishery was also characterized by undisclosed amounts of discards of low value fish, juveniles and other discards Notwithstanding, the higher landings in the artisanal fishery clearly confirm the importance of the Malindi–Ungwana Bay to the fisher communities along this coast Moreover, the artisanal fisheries within the Malindi–Ungwana Bay account for about 61% of the total marine fish and shrimp landings from the bay Further, despite the small-scale nature of the artisanal fishery, the sector is a primary source of livelihood for thousands of households along the entire coast Current estimates show that the artisanal fishery of the Malindi–Ungwana Bay directly employs over 2,000 fishermen [25] who use environmentally sound fishing gear ranging from traditional traps, hand lines, long lines, cast nets and gill and seine nets In 2001–2006, wide fluctuations in landings were observed in the artisanal fishery, whereas the commercial bottom trawl fishery recorded a downward trend throughout the period before its ban in 2006 The fluctuations in artisanal landings may be attributed to variations in trawling activities related to the number of operational vessels during this period and fluctuations in fishing effort within the artisanal fishery The impacts of the extreme weather conditions associated with the 1997–1998 El Nin˜o may also partly explain the fluctuations due to long-term effects of these conditions especially on the ecosystem The El Nin˜o phenomenon may lead to tropicalization of the ecosystem, disruption of the normal food web, and induced changes in species composition and migrations of a large number of fish and invertebrate species populations, as noted in the South American Pacific Coast fishery after the 1982–1983 El Nin˜o [26] Schwing et al [27] noted that the factors of concern are those affecting the general biological productivity and availability of food, aggregation for schooling and reproduction, larval dispersal, barriers to migration, physiological effects of extreme conditions, and changes in species composition and interactions Furthermore, the El Nin˜o weather is often preceded and followed by La Nin˜a-type weather, and hence the impacts of the El Nin˜o are often long term [28, 29] In the Malindi–Ungwana Bay fishery, the main factors include the effects of freshwater flooding into the bay and input of terrigenous sediments/nutrients from the rivers draining into the bay Moreover, the adverse 1997–1998 El Nin˜o weather also orchestrated a reduction in fishing intensity in the bay, thus giving the fishery time to recover, especially for overfished species Consequently, the period after the El Nin˜o provided a great opportunity for recovery of the Malindi– Ungwana Bay fishery stocks Further, the period after these adverse weather conditions and the expected effects of change in exploitation patterns presented an opportunity for re-assessment of the fishery and species composition within the fishing grounds, although few or no studies were conducted to assess the El Nin˜o impacts The recorded steady increase in artisanal fishery landings during 2002–2004 coincides well with the decrease in fishing efforts in the commercial bottom shrimp fishery An increase in the trawling activities during 2004–2006 before 123 216 the trawl ban and the continued encroachment into the artisanal TEZ grounds may also explain the decline in the artisanal fishery landings during this period Moreover, increased conflicts and damage to fishing gear of the artisanal fishery by the trawlers due to TEZ encroachment by commercial vessels also disrupted the fishing activities within the artisanal fishery and may partly account for the decline in catches This is evidenced by the increase in artisanal landings during 2006–2008 after the ban on trawling activities in the bay Additional factors include the recovery of the benthic habitats and fish stocks, reduced pressure on the TEZ from the commercial bottom trawl fishery fleet and a likely increase in fishing activities in the artisanal fishery in the absence of El Nin˜o phenomenon within the WIO region after 1998 In the commercial bottom trawl fishery, a steady decline in catches during 2001–2003 was recorded and was attributable to the decrease in fishing effort due to an imposed 4-month closed-season regulation in 2001, running from November–February each year [30] This seasonal closure was meant to safeguard the breeding populations and allow for the recovery of the stocks based on earlier studies indicating that the November–February period was the main breeding season for the target penaeid stocks and other fish species [4, 9] However, despite the ban on commercial bottom trawling, the artisanal fishery continued to record low landings of the target shrimp species This may be attributed to the fact that the penaeid shrimps are not target species for artisanal fishery Moreover, this subsector may be poorly equipped to exploit the bottom shrimp stocks since the main gear used are inexpensive passive gillnets, spears and driftnets, which target only fish This may suggest that there were no conflicts between the artisanal and the commercial bottom trawl fisheries in terms of the target species However, conflicts in partitioning of the fishing grounds and the impacts of bottom trawls on the ecosystem cannot be ignored The impacts of bottom trawling on the Malindi–Ungwana Bay fisheries resources have been documented in earlier studies [3] In the Malindi–Ungwana Bay fishery, the trawling activities have been characterized by excessive discarding of low value bycatch at sea In this study the overall ratio of shrimp to retained fish bycatch was 1:1.5 compared to a ratio of 1:7 recorded by Fulanda [1] Moreover, the retained bycatch increased from 432.0 t in 2001 to 602.3 t in 2004 although a decline was recorded in 2006, indicating that the retained bycatch increased over the years before the trawl ban Mwatha [4] estimated the rate of combined bycatch discard in the commercial bottom trawl fishery at t/day (average of 340 kg/trawler/h), which is still substantial compared to fish and shrimp landing estimates of 4.2–6.9 t/day in the artisanal fishery Similarly, 123 Fish Sci (2012) 78:209–219 Mwatha [4] also noted that over 25% of the discarded bycatch consisted of juveniles of commercial fish species such as Otolithes ruber, Johnius sp (Sciaenidae) and Pomadysis sp (Haemulidae), which are target species for the artisanal fishery Moreover, even the low-value commercial species are edible food fish that present valuable bycatch for the artisanal fishers of this coast Consequently, policies for utilization of the discarded fish bycatch must be designed to ensure lower discards, and high food and protein sufficiency for the coastal communities whose livelihoods depend on these resources Furthermore, there was a continued TEZ encroachment by the trawlers especially during 2001–2003 before a ban on night trawling was imposed Therefore, a substantial part of the catch was obtained from the 3–5 nM TEZ area, and the discarded bycatch thus ultimately impacted the artisanal fishery landings in this bay [3] Consequently, the years preceding the 2006 trawl ban were characterised by severe conflicts between the artisanal and commercial bottom trawl fisheries sectors with regard to resource partitioning and the deleterious fishing methods of the commercial bottom trawl fishery These conflicts may further explain the variations in annual landings especially in the commercial bottom trawl fishery In the late 1990s the FD-Kenya recommended retention of all bycatch in the commercial bottom trawl fishery in an effort to secure fish food supply and at the same to engage in resolution of the conflicts associated with the commercial bottom trawl fishery [31] During 2001–2004, a regional remedial action on shrimp trawl bycatch management in the WIO region was initiated in Kenya to, among others, promote bycatch reduction and undertake measures to increase utilisation of bycatch in the commercial bottom trawl fisheries of the WIO [32] The reduction in bycatch discards is indicated by the higher amounts of retained bycatch and the increase in shrimp: retained bycatch ratio from 1:7 in 1999 [3] to 1:1.5 recorded in the present study is partly attributed to these initiatives by the FD-Kenya to ensure sustainable management of the Malindi–Ungwana Bay fishery The current estimated ratio of shrimps to discarded bycatch recorded in the Malindi–Ungwana Bay fishery appears within the 1:3–1:15 ranges reported in other bottom trawl fisheries in the tropics [17] The initiatives to reduce discarding in bycatch and promote the utilization of these edible species have greatly improved the conditions in the artisanal fishery, and the increase in annual landings during 2006–2008 may be suggestive of a recovering fishery and habitat The 2006 trawl ban also appears to have safeguarded habitat degradation associated with bottom trawling and the encroachment on the shallower TEZ grounds by the commercial bottom trawl fishery vessels The absence of a significant increase in artisanal landings may therefore be attributed to the continued use of technologically inferior Fish Sci (2012) 78:209–219 vessels and gear This contrasts earlier observations that landings in the artisanal fishery would be significantly higher due to increased fishing activities and access to wider fishing grounds after the trawl ban [11] Consequently, the 2006 ban on commercial bottom trawling provides for proliferation of the Malindi–Ungwana Bay fishery stocks and an opportunity to re-design strategic long-term resource exploitation patterns for sustainable management In this study, the landings of the target shrimp species of the commercial bottom trawl fishery were not significantly different between the NEM and SEM seasons within years This suggests that the fishing activities of the commercial bottom trawl fishery were not influenced by the seasons Moreover, the trawlers exploited the absence of the artisanal fishers in the TEZ grounds during the adverse SEM season to encroach on and exploit the fishing grounds within 3–5 nM TEZ However, juvenile penaeid shrimp abundance, catchability and size appeared to be slightly influenced by seasons and bathymetric factors Macia [33] observed that water depth, salinity, temperature and turbidity are key factors influencing the spatial distribution of juvenile shrimp species In a separate study in north Kuwait Bay, Bishop and Khan [34] found that some species of juvenile penaeid shrimps such as Metapenaeus affinis were more catchable at shallower waters, while the bigger sizes were more abundant at deeper fishing grounds The Malindi–Ungwana Bay fishery is predominantly a shallowwater shrimp resource [1], but wide variations in depth are evident between the Formosa and Malindi fishing grounds In the present study there was a significant difference in the spatial distribution of the target shrimp species between the shallow water fishing grounds of Formosa and Malindi based on commercial bottom trawl fishery CPUE However, the deeper fishing grounds of both the Formosa and Malindi areas showed no significant differences in CPUE, confirming that resource-use patterns in the bay were predominantly targeted on shallow fishing grounds These observations may be partly attributed to variability in fishing effort, trawlability of the fishing grounds and the spatio-temporal distribution of the species [3] Similar observations have been recorded in other fisheries [1, 34] The overall mean CPUEs of the target shrimp species in both the Formosa and Malindi fishing grounds are slightly lower than an estimate of 47 kg/h reported by Mwatha [4] for the entire fishery based on a single trawl survey Consequently, there is a need to continuously monitor the Malindi–Ungwana Bay fishery in an effort to maintain a rich data and information base for the sustainable management of the fishery The SIMPER analysis for spatial–temporal distribution of the species with water depth reveals that the Formosa fishing grounds are more important for the artisanal fishery 217 partly because of the high contribution of brackish water species The Malindi fishing grounds are equally important and significantly contribute to the fisheries catch of mixed pelagic and demersal species, and Carangidae Somers [35] noted that sediment type is an important factor in the distribution of prawn species, and the spatial distributions of individual shrimp species are often related to depth and/or sediment type Thus, sediment and nutrient discharge from rivers feeding the bay is important for the fishery In the Gulf of Carpentaria, Australia, Somers [36] observed that Fenneropenaeus merguiensis mainly occurred in waters shallower than 20 m, while P esculentus was dominant at \35 m water depths where the sediments were mainly sand or muddy sands In contrast, P semisulcatus preferred mud or sandy mud sediments, while Metapenaeus endeavouri also preferred sand or muddy sand sediments In the far northern Great Barrier Reef, Australia, the spatial distribution of commercially important penaeid shrimps has been coarsely differentiated by a combination of three factors: water depth, mud content of the sediment and seafloor rugosity [37] Earlier studies in the Malindi–Ungwana Bay indicate that fish and shrimp larvae are more abundant from the shore up to nautical miles off shore [38] Therefore, the distribution of the shrimp species in Malindi–Ungwana Bay appears to be influenced by a combination of several factors, including water depth, salinity, temperature and turbidity, sediment type and seafloor rugosity Furthermore, the shrimp species utilize the mangrove creeks and near-shore ecosystems as nursery grounds during their early life stages (Wakwabi EO, unpubl data, 1988) The near-shore ecosystems are therefore critically important as nursery grounds for both fish and shrimps Consequently, the restriction or total ban of trawling activities in the near shore ecosystems is crucial for the maintenance of the important ecological functions of these habitats More research is needed for an extensive assessment of the Malindi–Ungwana Bay fishery habitats, and to test the ecological and economical implications of any changes applied in the management of this important fishery including considerations of lifting of the trawl ban In 2001–2006, the Malindi–Ungwana Bay fishery was placed under increased surveillance and monitoring to curb the fishing patterns of the commercial bottom trawl fishery and reduce resource-use conflicts with the artisanal fishery Before the ban in 2006, annual landings in both the artisanal and commercial bottom trawl fisheries fluctuated widely, mainly because of variations in fishing efforts due to increasing conflicts between the two subsectors Consequently, the increase in landings in the artisanal fishery after 2006 clearly indicates a spill-over effect from the ban on commercial bottom trawling since there was no observed change in fishing activities, vessels and/or gear used in this artisanal fishery Furthermore, Fulanda et al 123 Fish Sci (2012) 78:471–483 a 475 b c d Fig Serial transverse sections of DO muscle (block 3) from the D group of PBT NADH-diaphorase staining (a), mATPase after alkali pre-incubation (b) and after acid pre-incubation at pH 4.3 (c) and at pH 5.0 (d) Arrows (a, b) indicate small fibers that are interspersed with large fibers (bars 100 lm) acid pre-incubation at pH 4.0 and 4.3 (Fig 2c) but was stable at pH 4.5 or 5.0 (Fig 2d) The reaction intensity was uniform in all muscle fibers of DO and LO muscles at pH 4.5 or 5.0 regardless of the diameter (Fig 2d) On the other hand, after acid pre-incubation at pH 4.5 or 5.0, three mATPase staining profiles—low, medium and high—were commonly observed in the LO muscle portion of LOD muscle blocks adjacent to dark muscle (Fig 3) in all PBT groups except the F group (Fig 4d) The low, medium and high staining profiles in the LO muscle portion of LOD were associated with large, intermediate and small fibers, respectively, and are considered a mosaic pattern Thus, the smaller the FD, the higher the stability of muscle fibers at pH 4.5 or 5.0 and large-diameter fibers are less activated This pattern in LO muscle (a portion of the LOD muscle) spread from the separation line (Fig 5, double-headed arrow) between dark and LO muscles towards the dorsal or ventral end in a very thin sheet-like region (Fig 1b, red coloring) This mosaic thin sheet-like area is parallel to the dark muscle from lateral to medial areas, separates the dark from white muscle and may be a transition zone This mosaic pattern in LO muscle (a portion of LOD muscle) was observed in A, B, C, D and E groups of PBT (data not shown for all group) However, the mosaic pattern of the muscle fibers in LO muscle was indistinguishable from the muscle fibers of DO muscles, which did not react with S-58 slow-muscle myosin antibody (Fig 5b, arrowheads) In dark muscle, all small and many large fibers showed high NADH-diaphorase activity, and the remaining large fibers showed moderate NADH-diaphorase activity (Fig 6a) Large-diameter fibers in dark muscle showed intermediate NADH-diaphorase reactivity compared with those in DO and LO (low NADH-diaphorase reaction), suggesting that these fibers in dark muscle have an intermediate oxidative potential Muscle fibers that were lightly stained by NADH-diaphorase (Fig 6a, filled arrowhead) were intensely stained with mATPase after alkali preincubations (Fig 6b, filled arrowhead) On the other hand, muscle fibers that were strongly stained by NADH-diaphorase (Fig 6a, open arrowhead) were lightly stained with mATPase after alkali pre-incubations (Fig 6b, open arrowhead) The intense or light mATPase staining after alkali pre-incubation suggests that these fibers also have glycolytic potential (high or intermediate) However, in dark muscle, no muscle fibers reacted with mATPase after pre-incubation at pH 4.0 and 4.3 (Fig 6c) Using immunohistochemistry, we observed that the muscle fibers that reacted intensely with NADH-diaphorase (Fig 5a, right pointing) also reacted positively with the S-58 slow-muscle myosin antibody (Fig 5b, right pointing), suggesting that these fibers are slow-twitch oxidative muscle fibers in dark muscle On the other hand, muscle fibers that were lightly stained with NADH-diaphorase (Fig 5a, left pointing) did not react with S-58 (Fig 5b, left pointing) These fibers have low oxidative but high glycolytic potential (positive mATPase activity after alkali pre-incubations) and can be classified as fast-twitch oxidoglycolytic (intermediate white) fibers In dark muscle, we observed that the fast-twitch oxido-glycolytic fibers transformed to slow-twitch oxidative muscle fibers with increasing body weight of the PBT (Fig 7) Muscle growth The mean and maximum diameter fibers of different muscles of different groups of PBT are shown in Table There was a tendency for the mean and maximum diameter of muscle fibers to increase with increasing PBT body weight However, the mean FD in the C group was significantly lower than that in the B group for LO muscle and that in the D group was lower than that in the C group for dark muscle The muscle FD was about two fold higher in DO and LO muscle compared with dark muscle fibers in all groups of PBT In LO and DO muscle, the diameter of ‘‘maximum diameter’’ fibers varied from 99 to 132 lm in different PBT groups The frequency of different FD classes in different PBT groups is shown in Table In DO, LO and dark muscle, the muscle fibers were larger in diameter with heavier body weight The large body weight PBT (F group) did not contain muscle fibers of the smallest diameter (B20 lm) In DO and LO muscles, the frequency of small-diameter 123 476 Fish Sci (2012) 78:471–483 fibers (up to 80 lm) decreased, and large-diameter fibers (80–160 lm) increased with increasing PBT body weight Also, in dark muscle, the frequency of small-diameter fibers (up to 40 lm) decreased, and larger diameter fibers (41–70 lm) increased with increasing PBT body weight The new fibers were produce by hyperplasia and converted to small diameter fibers Also, the small-diameter fibers converted to large-diameter fibers, and thus, the frequency of small-diameter fibers decreased with increasing PBT body weight Muscle fibers B20 lm in diameter represented 4.25% of the fibers in DO, 1.02% of the fibers in LO and 8.72% of the fibers in dark muscle at an average body weight of 3.03 kg (A group) These percentages fell to 0.06% in DO, 0.25% in LO and 0.88% in dark muscle at an average body weight of 38.4 kg (E group), and were absent in fish with an average body weight of 54.3 kg (F group) We also investigated muscle fiber recruitment sites in the whole myotome in the different muscle blocks (blocks 1–5) of DO, LO and dark muscles in PBT In all groups of M H L Fig Transverse section of LO part of LOD muscle (block 4) from C group of PBT Myosin ATPase activity after acid pre-incubation at pH 5.0 in the LO muscle adjacent to the dark muscle L, M and H indicate low, medium and high mATPase activity, respectively The bar in the upper left corner of the photograph indicates the separation line between LO part of LOD muscle and dark muscle by thick connective tissue (bar 100 lm) a b Fig Transverse section of LO part of LOD muscle (block 4) from the F group of PBT NADH-diaphorase staining (a), mATPase after alkali pre-incubations (b), mATPase after acid pre-incubation at pH a Fig Transverse serial sections from LOD muscle (block 1) of the B group of PBT NADH-diaphorase (a) and S-58 slow muscle myosin antibody (b) (the dark muscle portion is at left; LO portion is at right) Left pointing indicates that large muscle fibers that were lightly stained by NADH-diaphorase in the dark muscle portion were not stained with S-58 antibody Right pointing indicates that all small and some large fibers that were intensely stained with NADH-diaphorase 123 c d 4.3 (c) and at pH 5.0 (d) The thick connective tissue that separates LO part of LOD muscle and dark muscle situated parallel to lower line of the photographs indicated by arrows (bars 200 lm) b in the dark muscle portion were positively stained with S-58 antibody Arrowheads indicate that all muscle fibers that showed low activity with NADH-diaphorase (in the LO portion) were not stained with S-58 antibody Double-headed arrow indicates the fascia, a thick connective tissue sheet that separates the dark muscle from the LO muscle (bars 100 lm) Fish Sci (2012) 78:471–483 a 477 c b Fig Transverse serial sections from dark muscle (block 2) of the D group of PBT NADH-diaphorase staining (a), mATPase after alkali pre-incubations (b) after acid pre-incubation at pH 4.3 (c) and at pH 5.0 (d) Filled arrowheads indicate that muscle fibers that were intermediately stained with NADH-diaphorase (a) were intensely a d stained with mATPase after alkali pre-incubation (b) No fibers showed mATPase activity after acid pre-incubation at pH 4.3 (c) Open arrowheads indicate that muscle fibers that were intensely stained with NADH-diaphorase (a) were lightly stained with mATPase after alkali pre-incubation (b) (bars 100 lm) b c Fig Transverse sections of dark muscle immunohistochemically stained with S-58 slow muscle myosin antibody Slow-twitch oxidative fibers (arrowheads) were positive to S-58 antibody, but the large fast- twitch oxido-glycolytic (intermediate) fibers (arrows) were negative to S-58 antibody Dark muscle sections in a, b, and c are from block of dark muscle in PBT groups A, B, and F, respectively (bars 200 lm) Table Mean and maximum diameter of muscle fibers in dorsal ordinary (DO), lateral ordinary (LO) and dark of the PBT groups Muscle Diameter (lm) PBT group A B e C d D c F b 83.44 ± 0.59 87.49 ± 0.56a Mean fiber diameter 57.76 ± 0.53 59.94 ± 0.48 66.86 ± 0.45 Mean of maximum diameter fibersA 107.25 ± 0.84c 98.50 ± 0.95d 104.45 ± 0.99c 114.87 ± 1.09b 128.21 ± 2.03a 128.87 ± 0.76a LO Mean fiber diameter Mean of maximum diameter fibersA 59.63 ± 0.48e 103.80 ± 0.57e 69.51 ± 0.47c 99.28 ± 0.88f 67.89 ± 0.52d 107.08 ± 1.01d 69.60 ± 0.35c 109.61 ± 0.56c 76.44 ± 0.55b 114.69 ± 1.45b 90.32 ± 0.58a 132.07 ± 1.38a Dark Mean fiber diameter 30.40 ± 0.15f 31.26 ± 0.17e 38.36 ± 0.17b 34.86 ± 0.12d 37.89 ± 0.16b 42.96 ± 0.24a d d b c c 58.23 ± 1.32a DO Mean of maximum diameter fibersA 45.13 ± 0.25 45.47 ± 0.35 55.53 ± 0.25 67.48 ± 0.49 E c 49.95 ± 0.40 49.40 ± 0.49 Values are the mean ± SE Values within a row with different lowercase superscript letters are significantly different (p \ 0.05) The PBT groups A, B, C, D, E and F were explained in ‘‘Materials and methods’’ A The 20 largest (in diameter) muscle fibers were examined from each PBT for DO, LO and dark muscle to calculate the mean of maximum diameter fibers PBT, new muscle fiber recruitment occurred non-uniformly throughout the myotomal cross section of different muscle blocks in all muscles (data not shown) The blocking of DO, LO and dark muscle was done to determine the muscle fiber types and differences in muscle FDs in different blocks of each muscle The individual myomere thickness differed in different blocks of muscle in the whole myotome of PBT Also, the color of the DO or LO muscle differed with its position in the myotome of PBT We assumed that the color and thickness difference of individual myomeres is due to different muscle fiber types and diameters, respectively However, no difference was 123 478 Fish Sci (2012) 78:471–483 Table Frequency (%) of each diameter class muscle fibers in dorsal ordinary (DO), lateral ordinary (LO) and dark muscles of the PBT groups Muscle DO Muscle fiber diameter class (lm) B d B 20 B 4.25 ± 1.06a a C 2.48 ± 0.41ab b D E F 0.48 ± 0.13b 0.68 ± 0.25b 0.06 ± 0.04b c c 10.42 ± 0.48 2.64 ± 1.11d 1.33 ± 0.52d – 25.34 ± 1.37 41 B d B 60 24.39 ± 0.19a 29.71 ± 0.96a 29.74 ± 0.16a 26.82 ± 1.97a 12.17 ± 0.32b 9.68 ± 3.41b 61 B d B 80 27.62 ± 1.91 b 38.11 ± 0.66 a 37.65 ± 1.72 a b b 25.6 ± 0.51b 15.45 ± 0.72 c 13.46 ± 1.77 c 22.47 ± 0.73 b a 38.62 ± 2.64a 2.83 ± 0.89 b 0.71 ± 0.22 b 2.53 ± 1.05 b a 20.93 ± 1.73a a 101 B d B 120 a 0.11 ± 0.09 1.02 ± 0.26a 21 B d B 40 25.12 ± 1.40a 41 B d B 60 23.57 ± 2.08 a 32.89 ± 3.53 ab 15.45 ± 1.47 c 1.95 ± 0.31 c 121 B d B 140 B d B 20 61 B d B 80 81 B d B 100 101 B d B 120 Dark A 21 B d B 40 81 B d B 100 LO PBT group 121 B d B 140 – 141 B d B 160 – 15.53 ± 0.78 30.19 ± 0.35 b 25.64 ± 0.34 b 5.99 ± 1.92 29.76 ± 2.58 32.96 ± 3.46 16.87 ± 3.77 – 0.31 ± 0.22a a 0.06 ± 0.05 0.39 ± 0.16a a 0.25 ± 0.11 0.99 ± 0.37a 5.00 ± 3.37 0.25 ± 0.18a 3.84 ± 0.95a – 6.15 ± 0.96b 8.28 ± 0.60b 10.01 ± 0.92b 6.11 ± 0.61b 0.77 ± 0.0c 24.02 ± 0.23 ab 40.95 ± 2.64 a 28.27 ± 1.12 ab 0.31 ± 0.11 c – 9.19 ± 0.24 30.86 ± 3.45 a 32.98 ± 0.43 ab 23.46 ± 2.95 bc 4.02 ± 1.52 bc – – a 8.83 ± 1.16 b 31.21 ± 1.54 ab 29.19 ± 1.92 bc 4.34 ± 1.07 – b b 16.04 ± 1.25 26.28 ± 0.51b a 37.57 ± 3.60a b 9.43 ± 4.46 24.23 ± 0.15a 0.98 ± 0.69a 4.71 ± 1.50a 31.11 ± 1.25 36.09 ± 1.87 0.5 ± 0.5a – b b 8.72 ± 1.84 21 B d B 30 45.39 ± 1.25a 35.31 ± 0.27ab 15.14 ± 2.43 31 B d B 40 39.33 ± 0.42a 48.42 ± 0.76a 43.81 ± 3.03a 55.47 ± 2.64a 59.19 ± 8.66a 39.52 ± 14.39a 41 B d B 50 6.56 ± 1.76b 7.09 ± 0.73b 32.57 ± 3.65a 17.49 ± 2.07ab 29.34 ± 10.82a 35.91 ± 4.28a – – 61 B d B 70 – – 6.52 ± 1.89 – cd a 2.65 ± 0.39 5.92 ± 1.13c ab B d B 20 51 B d B 60 1.96 ± 0.29 24.27 ± 0.63 – – a ab 24.02 ± 4.35bc a 0.37 ± 0.17 – 0.88 ± 0.62 9.6 ± 2.25d a 1.00 ± 0.71 – – 5.34 ± 3.04d 16.55 ± 11.26a 2.70 ± 2.70a Values are the mean ± SE Values within a row with different letters are significantly different (p \ 0.05) ‘‘d’’ represents diameter of muscle fiber ‘‘–’’ symbol represents the absence of muscle fiber of individual diameter class in the PBT group observed in muscle fiber types in different blocks of DO (from dorsal to medial), LO or dark (from lateral to medial) muscle The muscle FD of different muscle blocks of each muscle did not show a consistent pattern in different groups of PBTs (Table 5) Discussion The muscle fibers in DO and LO muscles showed low NADH-diaphorase activity, high mATPase activity after alkali pre-incubations (pH 9.4, 10.3 and 10.5) and inactivity after acid pre-incubation at pH 4.3 Thus, modifications of the histochemical procedure showed that DO and LO muscles are composed of white fibers [42, 43] Therefore, the muscle fibers from DO and LO muscle blocks of PBT myotome possess only white muscle fibers, similar to other teleost white muscles (in Takifugu rubripes [44]; in Rutilus rutilus [45]; in Noemacheilus barbatus [46]; in Dicentrarchus labrax [47]) This histochemical profile also permits classification of the muscle fibers of 123 DO and LO muscle blocks of PBT as fast-twitch glycolytic fibers [48, 49] The low, medium and high mATPase staining profiles in the LO muscle portion of LOD were associated with large, intermediate and small fibers, respectively The mATPase activity that corresponds to the staining profile was expressed within a range of fiber sizes, indicating a gradual transition in mATPase isoforms from the high staining of small fibers to the low staining of the large white muscle fibers [50] Small white fibers with stable mATPase activity after acid pre-incubation (pH 4.5 or 5.0) progressively transform into large white muscle fibers, and medium-diameter white fibers with moderate mATPase activity represent a transitional form with a mixture of myosin isoforms [50] The variation in the mATPase staining intensities after acid pre-incubation (pH 4.5 or 5.0) in LO muscle may be related to the change or transition in the composition of the myosin heavy chain isoforms or co-expression of the different myosin isoforms [47, 51] These may be related to the phenotypic plasticity in muscle contractile functions as the fish grows [47, 52–54] The mosaic pattern of the fibers in the LO portion of LOD Fish Sci (2012) 78:471–483 479 Table Muscle fiber diameters in different blocks of dorsal ordinary (DO), lateral ordinary (LO) and dark muscles of the PBT groups Muscle PBT group Muscle fiber diameter (lm) Block DO A 52.31 ± 1.10b B 56.16 ± 0.80 d C D E F LO Block 51.30 ± 1.00b cd Block 63.62 ± 1.27a 61.26 ± 1.27a 62.19 ± 1.17a c b 71.95 ± 1.44a 72.84 ± 1.07a 64.25 ± 0.84b 56.72 ± 1.18 59.26 ± 0.90 66.31 ± 0.99b 65.24 ± 1.12b 65.96 ± 0.97b 64.04 ± 1.13 c b 72.1 ± 1.22 a 81.34 ± 0.94 b 92.83 ± 1.61 a 84.07 ± 1.05 c 89.23 ± 1.32 b 68.18 ± 1.21 c 76.13 ± 1.18 d 77.76 ± 1.16 Block 63.09 ± 1.26 66.13 ± 0.96 bc 91.56 ± 1.37 a 67.6 ± 0.94bc 78.86 ± 1.36bc 90.44 ± 1.23 b 97.43 ± 1.19a A B b 57.69 ± 1.06 62.47 ± 1.21c b 56.74 ± 1.05 70.8 ± 1.01b a 61.23 ± 1.08 76.44 ± 1.09a a 61.52 ± 0.99 68.42 ± 1.05b 61.02 ± 1.14a 68.74 ± 0.82b C 65.28 ± 1.06d 69.24 ± 1.28b 74.74 ± 1.25a 69.50 ± 1.08b 61.94 ± 1.05c D 65.94 ± 0.94 b a 71.29 ± 0.76 a 67.62 ± 0.58 b 71.55 ± 0.75a 74.67 ± 1.25 b 75.07 ± 1.17 b 78.84 ± 1.22 a 78.63 ± 1.34a 83.47 ± 1.06 c 99.59 ± 1.33 a 92.85 ± 1.25 b 93.5 ± 1.15b A 29.15 ± 0.31 c 28.57 ± 0.28 31.52 ± 0.38 ab 30.91 ± 0.32 ab B 30.16 ± 0.32c 31.45 ± 0.36ab 32.43 ± 0.37a 31.9 ± 0.49a C 39.00 ± 0.44 a ab 37.54 ± 0.36 b 38.87 ± 0.38 a 38.18 ± 0.35ab 34.67 ± 0.32 b 35.06 ± 0.25 b 33.39 ± 0.18 c 33.53 ± 0.21c 40.23 ± 0.36 a 37.98 ± 0.24 b 36.07 ± 0.39 c 38.19 ± 0.35b 42.04 ± 0.43 b 42.28 ± 0.50 ab 43.78 ± 0.70 a 43.65 ± 0.55a E F Dark Block D E F 73.78 ± 1.21 b 75.26 ± 1.15 c 81.6 ± 1.23 c 38.32 ± 0.34 a 36.99 ± 0.27 c 36.81 ± 0.37 ab 43.02 ± 0.50 32.08 ± 0.35a 30.43 ± 0.42b Values are the mean ± SE Values within a row with different lowercase superscript letters are significantly different (p \ 0.05) in different blocks of muscle in the same group of PBT For DO muscle, blocks 1–5 were chronologically numbered from dorsal to medial of the PBT myotome For LO and dark muscle, blocks 1–5 were chronologically numbered from lateral to medial of the PBT myotome The PBT groups A, B, C, D, E and F were explained in ‘‘Materials and methods’’ muscle that was indistinguishable from the fibers of DO muscles and did not react with S-58 slow-muscle myosin antibody A similar result was observed in the fast muscle fibers of Takifugu rubripes [37] with the S-58 slow-muscle myosin antibody The histochemical mATPase mosaic pattern is an indicator of myotomal maturity and a consequence of muscle hyperplasia [3, 47, 53, 55] The presence of the mosaic pattern in A, B, C, D and E groups of PBT suggests that PBTs at a body weight of 38.4 kg are still at a juvenile stage In dark muscle, small fibers showed high NADHdiaphorase activity and large fibers showed moderate NADH-diaphorase activity compared with the fibers in DO muscle Thus, small-diameter fibers may have high oxidative activity [44], and glycolytic fibers are largely inversely reactive compared with oxidative fibers [56] The intense or light mATPase staining of large fibers after alkali pre-incubation suggests that these fibers also have glycolytic potential (high or intermediate) The largediameter fibers in dark muscle that have alkali-stable mATPase activity characteristic of white muscle fibers are similar to the red muscle characteristics of trout [57] Thus, the mATPase activity of the small fibers in dark muscle may simply be less stable than that in the white fibers after alkali pre-incubation and inactive after acid at pH 4.0 and 4.3 The classical ‘‘acid reversal’’ (acid-stable and alkalilabile) mATPase activity was observed in red muscle fibers of different fish species (in Nile tilapia [42]; in mullet, goby and guppy [57]) But in PBT dark muscle, muscle fibers did not show ‘‘acid reversal’’ mATPase activity, and this loss of acid stability of mATPase was also observed in the red muscle of certain fish species (trout, carp, goldfish, eel and catfish) [57] Histochemical and immunohistochemical reaction suggests that the dark muscle of PBT is typically heterogeneous in its fiber composition, i.e., slow-twitch oxidative fibers are intermingled with fast-twitch oxido-glycolytic fibers Such types of muscle fibers are generally found in mammalian and avian muscles [58, 59] as well as salmonid red muscle [60] In some fish species (mullet and trout), red muscle consists of red (slow-twitch oxidative) fibers interspersed with fast-type muscle fibers [57] Tuna maintains a higher total metabolic rate and thus can maintain higher aerobic speeds than its prey [8] The intermingled muscle fiber types in dark muscle of PBT may be the reason for their high aerobic speed and metabolic rate 123 480 In dark muscle of PBT, the fast-twitch oxido-glycolytic muscle fibers are transformed to slow-twitch oxidative fibers with increasing body weight Muscle fibers may have a dynamic structure that exhibits high plasticity and undergoes a type shift following an obligatory pathway: type I (slow-twitch oxidative) $ IIA (fast-twitch oxidative) $ IIX (fast-twitch oxido-glycolytic) $ IIB (fasttwitch glycolytic) [61–66] Muscle growth The mean diameter of muscle fibers increases with increasing PBT body weight But this tendency was not followed in LO muscle between B and C group and in dark muscle between C and D group of PBT This might be that muscle growth may occur both by hypertrophy and hyperplasia at this stage The muscle FD was observed larger in DO and LO muscle compared with dark muscle in all PBT groups This is similar to reports regarding other fish species [33, 55], which reported that the diameters of white muscle fibers are largely comparable with their red muscle fiber counterparts The maximum diameter of white muscle fibers in temperate fish is around 150–200 lm, reflecting some constraint on the diffusion of metabolites [67, 68] Muscle fibers have a maximum diameter to maintain the muscular aerobic capacity that is necessary for diffusion of oxygen and metabolites [67, 69] However, a maximum of 300 lm diameter muscle fiber in Takifugu rubripes fast muscle was observed at a body weight of kg [37] Furthermore, Antarctic notothenioid muscle contains giant muscle fibers (219–400 lm) [34] The presence of giant muscle fibers in Antarctic notothenioid muscle reflects the relaxation of diffusion constraints at low temperature [70, 71] There is an optimal maximum FD in fish that reflects a trade-off between avoiding diffusion constraints and the need to minimize the costs of ion pumping, an idea known as the ‘‘optimal fiber size hypothesis’’ [39] Small-diameter fibers are expected to grow more quickly than large-diameter fibers because of their higher surface-to-volume ratio and greater capacity to assimilate nutrients [13, 72] In DO, LO and dark muscle, the muscle fibers were larger in diameter with heavier body weight The smalldiameter fibers converted to large-diameter fibers, and thus, the frequency of small-diameter fibers decreased with increasing PBT body weight Thus, hyperplastic growth may decrease as the fish grow [73] Smallest diameter fibers (B20 lm) represent newly recruited fibers by hyperplasia [57] The smallest diameter fibers gradually decreased from small to large body weight PBT and observed absence in F group PBT, consistent with the complete cessation of muscle fiber recruitment [34] Thus, the recruitment of new muscle fibers continues up to 123 Fish Sci (2012) 78:471–483 the weight of 38.4 kg (E group PBT) in all muscles and then ceases Thus, hyperplastic growth in PBT stops between a body weight of 38.4 and 54.3 kg, consistent with the notion that hyperplastic growth may cease completely after fish reach a certain size [74] Therefore, after cessation of hyperplasia in PBT, muscle growth mainly occurs by hypertrophy and elongation of the muscle fibers (because body length increases with muscle fiber growth) by adding newly formed sarcomeres to the ends of muscle fibers [75] The presence of smallest diameter fibers (20 lm) in the A, B, C, D and E groups of PBT indicated that both hyperplastic and hypertrophic growth occurred in these groups Teleosts may recruit muscle fibers throughout the juvenile phase and into adulthood until a certain body size is reached [13, 67, 71, 76, 77] Previous research has shown that fish species that experience long periods of hyperplasia reach a larger body size than small-sized fish, which grow primarily through hypertrophy [3, 74, 77] The findings of our current study are in agreement with this trend Hyperplasia persisted well into the adult phases of PBT, a species that attains a relatively large ultimate body size Therefore, new fiber recruitment may play a main role in determining the ultimate size of PBT In Harpagifer species hyperplasia completely stops at their adult stage and subsequently growth occurs only by hypertrophy [78] In this study, we observed that in PBT, new fiber formation ceased at only 9% of the maximum recorded body size (600 kg) [79] and 34% of the recorded body length (430 cm) [80] such that the major portion of growth during juvenile and adult stages occurred through hypertrophy and elongation of muscle fibers This observation suggests that the rapid growth stage of PBT may be the hypertrophy stage; rapid growth periods favor fiber hypertrophy, and slow growth periods favor fiber recruitment [67] The new muscle fiber recruitment ceases at around 44% of the maximum length in different species of North American freshwater fish, and subsequent muscle growth occurs only by hypertrophy of the fibers [13] In Antarctic notothenioid (Notothenia coriiceps) fish species, new fiber formation ceases at 21.3% of the maximum recorded body size [34] New muscle fibers recruitment occurred non-uniformly throughout the myotomal cross-section of different muscle blocks in all muscles of PBT Similar results were observed in myotomal cross-sections in juvenile fishes of the species E maclovinus, C gobio and P magellanica [34], suggesting that the recruitment of new muscle fibers in PBT myotomes does not occur uniformly The muscle FD of different muscle blocks of each muscle did not show a consistent pattern in different groups of PBT, indicating that the muscle FD is not related to the thickness of the individual myomeres The color differences in different blocks of DO and LO muscle in whole PBT myotome may therefore be related to differences in Fish Sci (2012) 78:471–483 myoglobin and lipid content PBTs can accumulate a large amount of myoglobin in their fast skeletal (white) muscle, and the redder color of deep DO and LO muscle blocks (white muscle) is mainly due to the presence of myoglobin [81] The higher lipid content in superficial muscle blocks from DO and LO muscle may be the cause of faded red color muscle in this region of PBT myotome It may be that the superficial muscle blocks (chu-toro) of PBT contain higher lipid than deep muscle blocks (akami) of DO or LO (white muscle) muscle [82] DO and LO muscles consist of only white muscle fibers, which reacted homogeneously in different histochemical reactions regardless of their diameter The presence of two types of muscle fibers in dark muscle of PBT that differ from the typical red muscle suggests that the functional activity of the dark muscle is more complex than generally believed The mosaic hyperplasia that was observed in the LO portion of the LOD muscle during the prolonged period of growth in PBT is very important, because a higher recruitment of fibers endows the fish with the potential to accomplish further growth by hypertrophy This observation is of practical interest as it is responsible for the fish reaching a desirable commercial size The prolonged hyperplasia and lack of a maximum fiber size in cultured PBT is likely to delay the onset of impaired metabolic exchange in this fish It is therefore reasonable to conclude that this prolonged hyperplastic growth is responsible for enabling PBT to grow to their impressive ultimate size As skeletal muscle constitutes the edible part of PBT, knowledge of muscle fiber types and their growth mechanisms is important for the development of PBT aquaculture The influence of seasonal, environmental and nutritional factors on muscle phenotypic plasticity and muscle growth in this species is worthy of further studies Acknowledgments This study was supported in part by the International Education and Research Center for Aquaculture Science of Bluefin Tuna and Other Cultured Fish, Kinki University Global COE Program for the Ministry of Education, Culture, Sports, Science and Technology of Japan The authors are grateful to Nantake Suisan, Osaka, Japan, for the assistance in the collection and supply of muscle samples 481 10 11 12 13 14 15 16 17 18 19 20 21 References 22 Hagen Ø, Solberg C, Johnston IA (2006) Sexual dimorphism of fast muscle fiber recruitment in farmed Atlantic halibut (Hippoglossus hippoglossus L.) Aquaculture 261:1222–1229 Sa¨nger AM, Stroiber W (2001) Muscle fiber diversity and plasticity In: Johnston IA (ed) Muscle growth and development Fish physiology, vol 18 Academic Press, San Diego, pp 187–250 Stickland NC (1983) Growth and development of muscle fibers in the rainbow trout (Salmo gairdneri) J Anat 137:323–333 Hoyle J, Gill HS, Weatherley AH (1986) Histochemical characterization of myotomal muscle in the grass pickerel, Esox 23 24 25 americanus vermiculatus (LeSeuer), and the muskellunge, E masquinongy (Mitchell) J Fish Biol 28:393–401 Luther PK, Munro PMG, Squire JM (1995) Muscle ultrastructure in fish Micron 26:431–459 Ellerby DJ, Altringham JD, Williams T, Block BA (2000) Slow muscle function of pacific bonito (Sarda chiliensis) during steady swimming J Exp Biol 203:2001–2013 Katz SL (2002) Design of heterothermic muscle in fish J Exp Biol 205:2251–2266 Katz SL, Shadwick RE, Syme D (2001) Enhanced power in yellowfin tuna Nature 410:770–771 Johnston IA, Davison W, Goldspink G (1977) Energy metabolism of carp swimming muscles J Comp Physiol 114:203–216 Johnston IA, Alderson R, Sandham C, Mitchell D, Selkirk C, Dingwall A, Nickell D, Baker R, Robertson B, White D, Springate J (2000) Patterns of muscle growth in early and late maturing populations of Atlantic salmon (Salmo salar L.) Aquaculture 189:307–333 Rowlerson A, Mascarello F, Radaelli G, Veggetti A (1995) Differentiation and growth of muscle in the fish Sparus aurata (L): II Hyperplastic and hypertrophic growth of lateral muscle from hatching to adult J Muscle Res Cell Motil 16:223–236 Veggetti A, Mascarello F, Scapolo PA, Rowlerson A (1990) Hyperplastic and hypertrophic growth of lateral muscle in Dicentrarchus labrax (L.) Anat Embryol 182:1–10 Weatherley A, Gill HS, Lobo A (1988) Recruitment and maximal diameter of axial muscle fibers in teleost and their relationship to somatic growth and ultimate size J Fish Biol 33:851–859 Hagen Ø, Vieira VLA, Solberg S, Johnston IA (2008) Myotube production in fast myotomal muscle is switched-off at shorter body lengths in male than female Atlantic halibut (Hippoglossus hippoglossus L.) resulting in a lower final fiber number J Fish Biol 73:139–152 Love RM (1958) Studies on the North Sea cod I Muscle cell dimensions J Sci Food Agric 9:195–198 Yada O, Tsuchimoto M, Wang Q, Apabraza PAG, Jabarsyah A, Tachibana K (2000) Differences of muscle fiber type and temporal change of K-value among parts toward depth of dorsal muscle in carp (cultured) Fish Sci 66:147–152 Bekhit AED, Faustman C (2005) Metmyoglobin reducing activity Meat Sci 71:407–439 Hood DE (1980) Factors affecting the rate of metmyoglobin accumulation in prepackaged beef Meat Sci 4:247–265 Ryu YC, Kim BC (2005) The relationship between muscle fiber characteristics, postmortem metabolic rate, and meat quality of pig Longissimus dorsi muscle Meat Sci 71:351–357 Candek-Potokar M, Lefaucheur L, Zlender B, Bonneau M (1999) Effect of slaughter weight and/or age on histological characteristics of pig Longissimus dorsi muscle as related to meat quality Meat Sci 52:95–203 Davie PS, Sparksman RI (1986) Burnt tuna: an ultrastructural study of post mortem changes in muscle of yellowfin tuna (Thunnus albacares) caught on rod and reel and southern bluefin tuna (Thunnus maccoyii) caught on hand line or long line J Food Sci 51:1122–1128 Watson CL, Morrow HA, Brill RW (1992) Proteolysis of skeletal muscle in yellowfin tuna (Thunnus albacares): evidence of calpain activation Comp Biochem Physiol 103B:881–887 Carpene` E, Veggetti A, Mascarello F (1982) Histochemical fiber types in the lateral muscle of fishes in fresh, brackish and salt water J Exp Biol 20:379–396 Kiessling A, Ruohonen K, Bjørnevik M (2006) Muscle fiber growth and quality in fish Archiv Tierzucht Dummerstorf 49:137–146 Catarci C (2005) The world tuna industry—an analysis of imports and prices, and of their combined impact on catches and tuna 123 482 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Fish Sci (2012) 78:471–483 fishing capacity Section 3-Tuna fishing industry In: Bayliff WH, de-Leiva Moreno JI, Majkowski J (eds) Management of tuna fishing capacity: conservation and socio-economics Second meeting of the technical advisory committee of the FAO project Food and Agriculture Organization of the United Nations, Madrid, p 243 Fauconneau B, Chmaitilly J, Andre S, Cardinal M, Cornet J, Vallet JL, Dumont JP, Laroche M (1993) Characteristics of rainbow trout flesh: I Chemical composition and cellularity of muscle and adipose tissues Sciences des Aliments 13:173–187 Hatae K, Yoshimatsu F, Matsumoto JJ (1990) Role of muscle fibers in contributing firmness of cooked fish J Food Sci 55:693–696 Hurling R, Rodell JB, Hunt HD (1996) Fiber diameter and fish texture J Texture Stud 27:679–685 Johnston IA (1999) Muscle development and growth: potential implications for flesh quality in fish Aquaculture 177:99–115 Hattori N, Miyashita S, Sawada Y, Kato K, Nasu T, Okada T, Murata O, Kumai H (2001) Lateral muscle development of the Pacific bluefin tuna, Thunnus thynnus orientalis, from juvenile to young adult stage under culture condition Suisanzoshoku 49:23–28 Sawada Y, Okada T, Miyashita S, Murata O, Kumai H (2005) Completion of the Pacific bluefin tuna Thunnus orientalis (Temminck et Schlegel) life cycle Aquac Res 36:413–421 Dubowitz V, Brooke MH (1973) Histological and histochemical reactions In: Muscle biopsy: a modern approach, vol Saunders, London, pp 20–73 Johnston IA, Patterson S, Ward P, Goldspink G (1974) The histochemical demonstration of myofibrillar adenosine triphosphatase activity in fish muscle Can J Zool 52:871–877 Fernandez DA, Clavo J, Franklin CE, Johnston IA (2000) Muscle fiber types and size distribution in sub-antarctic notothenioid fishes J Fish Biol 56:1295–1311 Crow MT, Stockdale FE (1986) Myosin expression and specialization amongst the earliest muscle fibers of the developing avian limb Dev Biol 113:238–254 Barresi MJF, Stickney HL, Devoto SH (2000) The zebrafish slow-muscle-omitted gene product is required for hedgehog signal transduction and the development of slow muscle identity Development 127:2189–2199 Fernandes JMO, Mackenzie MG, Elgar G, Suzuki Y, Watabe S, Kinghorn JM, Johnston IA (2005) A genomic approach to reveal novel genes associated with myotube formation in the model teleost, Takifugu rubripes Physiol Genomics 22:327–338 Hammill E, Wilson RS, Johnston IA (2004) Sustained swimming performance and muscle structure are altered by thermal acclimation in male mosquitofish J Therm Biol 29:251–257 Johnston IA, Abercromby M, Vieira VLA, Sigursteindo´ttir RJ, Kristja´nsson BK, Sibthorpe D, Sku´lason S (2004) Rapid evolution of muscle fiber number in postglacial population of Arctic charr Salvelinus alpinus J Exp Biol 207:4343–4360 Bjørnevik M, Karlsen O, Johnston IA, Kiessling A (2003) Effect of sustained exercise on white muscle structure and flesh quality in farmed cod (Gadus morhua L.) Aquac Res 34:55–64 Alami-Durante H, Me´dale F, Cluzeaud M, Kaushik SJ (2010) Skeletal muscle growth dynamics and expression of related genes in white and red muscles of rainbow trout fed diets with graded levels of a mixture of plant protein sources as substitutes for fishmeal Aquaculture 303:50–58 Dal Pai-Silva M, Carvalho RF, Pellizzon CH, Dal Pai V (2003) Muscle growth in Nile tilapia (Oreochromis niloticus): histochemical, ultrastructural and morphometric study Tissue Cell 35:179–187 Devincenti CV, Diaz AO, Goldemberg AL (2000) Characterization of the swimming muscle of the anchovy Engraulis anchiota (Hubbs and Martini 1935) Anat Histol Embryol 29:197–202 123 44 Akolkar DB, Kinoshita S, Yasmin L, Ono Y, Ikeda D, Yamaguchi H, Nakaya M, Erdogan O, Watabe S (2010) Fiber type-specific expression patterns of myosin heavy chain genes in adult torafugu (Takifugu rubripes) muscles J Exp Biol 213:137–145 45 Kilarski W (1990) Histochemical characterization of myotomal muscle in the roach Rutilus rutilus (L.) J Physiol 36:353–362 46 Kilarski W, Kozlowska M (1985) Histochemical and electromicroscopical analysis of muscle fiber in myotome of teleost fish (Noemacheilus barbatus, L.) Gegenbaurs Morphologis Jahrbuch 131:55–72 47 Scapolo PA, Veggetti A, Mascarello F, Romanello MG (1988) Developmental transitions of myosin isoforms and organization of the lateral muscle in the teleost Dicentrarchus labrax (L.) Anat Embryol 178:287–295 48 Rome LC, Funke RP, Alexander RM, Lutz G, Aldridge H, Scott F, Freadman M (1988) Why animals have different muscle fiber types Nature 335:824–827 49 Stoiber W, Haslett JR, Wenk R, Steinbacher P, Gollmann HP, Sa¨nger AM (2002) Cellularity changes in developing red and white fish muscle at different temperatures: simulating natural environmental conditions for a temperate freshwater cyprinid J Exp Biol 205:2349–2364 50 Lo´pez-Albors O, Ayala MD, Gil F, Garcı´a Alcazar A, Abella´n E, Latorre R, Ramı´rez Zarzosa G, Va´zquez JM (2003) Early temperature effects on muscle growth dynamics and histochemical profile of muscle fiber of sea bass Dicentrarchus labrax L., during larval and juvenile stages Aquaculture 220:385–406 51 Staron RS, Pette D (1986) Correlation between myofibrillar ATPase activity and myosin heavy chain composition in rabbit muscle fibers Histochemistry 86:19–23 52 Brooks S, Johnston IA (1993) Influence of development and rearing temperature on the distribution, ultrastructure and myosin sub-unit composition of myotomal muscle-fiber types in the plaice (Pleuronectes platessa) Mar Biol 117:501–513 53 Mascarello F, Rowlerson A, Radaelli G, Scapolo PA, Veggetti A (1995) Differentiation and growth of muscle in the fish Sparus aurata (L): I Myosin expression and organization of fiber types in lateral muscle from hatching to adult J Muscle Res Cell Motil 16:213–222 54 Veggetti A, Mascarell F, Scapolo PA, Rowlerson A, Carnevali C (1993) Muscle growth and myosin isoform transitions during development of a small teleost fish, Poecilia reticulata (Peters) (Atheriniformes, Poeciliidae): a histochemical, immunohistochemical, ultrastructural and morphometric study Anat Embryol 187:353–361 55 Carpene` E, Veggetti A (1981) Increase in muscle fibers in the lateralis muscle (white portion) of Mugilidae (Pisces, Teleostei) Experientia 37:191–193 56 Walesby NJ, Nicol CJM, Johnston A (1982) Metabolic differentiation of muscle fibers from a haemoglobin-less (Champsocephalus gunnari Lonnberg) and a red-blooded (Notothenia rossi Fischer) Antarctic fish Br Antarct Surv Bull 51:201–214 57 Rowlerson A, Scapolo A, Mascarello F, Veggetti A (1985) Comparative study of myosins present in the lateral muscle of some fish: species variations in myosin isoforms and their distribution in red, pink and white muscle J Muscle Res Cell Motil 6:601–640 58 Oshima I, Iwamoto H, Nakamura YN, Takayama K, Ono Y, Murakami T, Shiba N, Tabata S, Nishimura S (2009) Comparative study of the histochemical properties, collagen content and architecture of the skeletal muscles of wild boar crossbred pigs and commercial hybrid pigs Meat Sci 81:382–390 59 Roy BC, Oshima I, Miyachi H, Shiba N, Nishimura S, Tabata S, Iwamoto H (2007) The histochemical properties and collagen architecture of M iliotibialis lateralis and M puboischiofemoralis in Fish Sci (2012) 78:471–483 60 61 62 63 64 65 66 67 68 69 70 71 male broilers with different growth rates induced by feeding at different nutritional levels Br Poult Sci 48:312–322 Rome LC (2000) Fish as an experimental model for studying muscle function In: Ostrander GK (ed) The handbook of experimental animals—the laboratory fish Academic Press, London, pp 319–329 Agbulut O, Noirez P, Butler-Browne G, Jockusch H (2004) Specific isomyosin proportions in hyperexcitable and physiologically denervated mouse muscle FEBS Lett 561:191–194 Chin ER, Grange RW, Viau F, Simard AR, Humphries C, Shelton J, Bassel-Duby R, Williams RS, Michel RN (2003) Alterations in slow-twitch muscle phenotype in transgenic mice over expressing the Ca2? buffering protein parvalbumin J Physiol 547:649–663 Ishihara A, Kawano F, Ishioka N, Oishi H, Higashibata A, Shimazu T, Ohira Y (2004) Effects of running exercise during recovery from hindlimb unloading on soleus muscle fibers and their spinal motoneurons in rats Neurosci Res 48:119–127 Pette D, Staron RS (2000) Myosin isoforms, muscle fiber types, and transitions Microsc Res Tech 50:500–509 Pette D, Staron RS (2001) Transitions of muscle fiber phenotypic profiles Histochem Cell Biol 115:359–372 Ustunel I, Akkoyunlu G, Demir R (2003) The effect of testosterone on gastrocnemius muscle fibers in growing and adult male and female rats: a histochemical, morphometric and ultrastructural study Anat Histol Embryol 32:70–79 ˚ sga˚rd T (1991) Kiessling A, Kiessling KH, Storebakken T, A Changes in the structure and function of the epaxial muscle of rainbow trout (Oncorhynchus mykiss) in relation to ration and age I Growth dynamics Aquaculture 93:335–356 Weatherley AH, Gill HS (1985) Dynamics of increase in muscle fibers in fishes in relation to size and growth Experientia 41:353–354 Johnston IA, Manthri S, Bickerdike R, Dingwall A, Luijkx R, Campbell P, Nickell D, Alderson R (2004) Growth performance, muscle structure and flesh quality in out-of-season Atlantic salmon (Salmo salar) smolts reared under two different photoperiod regimes Aquaculture 237:281–300 Egginton S, Skilbeck C, Hoofd L, Calvo J, Johnston IA (2002) Peripheral oxygen transport in skeletal muscle of Antarctic and sub-Antarctic notothenioid fish J Exp Biol 205:769–779 Johnston IA, Fernandez DA, Calvo J, Vieira VLA, North AW, Abercromby M, Garland T Jr (2003) Reduction in muscle fiber 483 72 73 74 75 76 77 78 79 80 81 82 83 number during the adaptive radiation of notothenioid fishes: a phylogenetic perspective J Exp Biol 206:2595–2609 Valente LMP, Rocha E, Gomes EFS, Silva MW, Oliveira MH, Monteiro RAF, Fauconneau B (1999) Growth dynamics of white and red muscle fibers in fast and slow growing strains of rainbow trout J Fish Biol 55:675–691 Weatherley AH, Gill HS (1987) The biology of fish growth Academic Press, London, p 443 Kundu R, Mansuri A (1992) Growth of pectoral muscle fibers in relation to somatic growth in some marine fishes Neth J Zool 42:595–606 Tu˚mova E, Teimouri A (2009) Chicken muscle fibers characteristics and meat quality: a review Scientia Agriculturae Bohemica 40:253–258 Johnston IA, Manthri S, Alderson R, Smart A, Campbell P, Nickell D, Robertson B, Paxton CGM, Burt ML (2003) Freshwater environment affects growth rate and muscle fiber recruitment in seawater stages of Atlantic salmon (Salmo salar L.) J Exp Biol 206:1337–1351 Zimmerman AM, Lowery MS (1999) Hyperplastic development and hypertrophic growth of muscle fibers in the white seabass (Atractoscion nobilis) J Exp Biol 284:299–308 Johnston IA, Vieira VLA, Fernandez DA, Abercromby M, Brodeur JC, Peck L, Calvo J (2003) Muscle growth in Polar fish: a study of Harpagifer species with sub-Antarctic and Antarctic distributions Fish Sci 68:1023–1028 Masuma S, Takebe T, Sakakura Y (2011) A review of the broodstock management and larviculture of the Pacific northern bluefin tuna in Japan Aquaculture 315:2–8 Anonymous (2007) http://en.wikipedia.org/wiki/Northern_bluefin_ tuna Ochiai Y, Watanabe Y, Ozawa H, Ikegami S, Uchida N, Watabe S (2010) Thermal denaturation profiles of tuna myoglobin Biosci Biotechnol Biochem 74:1673–1679 Roy BC, Miyake Y, Ando M, Kawasaki KI, Tsukamasa Y (2010) Proximate and fatty acid compositions in different flesh cuts of cultured, cultured fasted, and wild Pacific bluefin tuna (Thunnus orientalis) J Aquat Food Prod Technol 19:284–297 Balshaw S, Edwards JW, Ross KE, Daughtry BJ (2008) Mercury distribution in the muscular tissue of farmed southern bluefin tuna (Thunnus maccoyii) is inversely related to the lipid content of tissues Food Chem 111:616–621 123 Fish Sci (2012) 78:485–490 DOI 10.1007/s12562-012-0472-x ORIGINAL ARTICLE Food Science and Technology Effects of ultrasonic treatment on collagen extraction from skins of the sea bass Lateolabrax japonicus Hyun Kyung Kim • Young Ho Kim • Yun Ji Kim • Hyun Jin Park • Nam Hyouck Lee Received: August 2011 / Accepted: 13 January 2012 / Published online: February 2012 Ó The Japanese Society of Fisheries Science 2012 Abstract In this study we investigated the effects of ultrasonic wave treatment on the extraction yield of acidsoluble collagen from sea bass skins Two extraction methods were compared: a 24 h acid treatment using 0.5 M acetic acid (1:200 sample/acid, w/v) and an extraction using ultrasonic treatment after the addition of a 0.5 M acetic acid solution The results indicated that the extraction yield of collagen increased with the ultrasonic treatment, with the extraction rate increasing rapidly at higher amplitudes of ultrasonic treatment The subunit compositions of the collagen extracted by ultrasonic treatment were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, which revealed that the a1(a3), a2, and b chains of collagen were present early in the ultrasonic treatment An unknown component, believed to be a product of collagen degradation induced by the ultrasonic treatment, was detected only after a longer treatment time The component extracted by the ultrasonic treatment was determined to be collagen based on the finding that there were no changes in the main components of collagen, specifically, the a1(a3), a2, and b chains, following pepsin treatment Keywords Sea bass skin Á Collagen Á Ultrasonic treatment Á Extraction H K Kim Á Y H Kim Á Y J Kim Á N H Lee (&) Korea Food Research Institute, 516 Baekhyundong, Bundanggu, Songnamsi Kyounggido 466-746, Korea e-mail: lnh@kfri.re.kr H J Park College of Life Science and Biotechnology, Korea University, Anamdong, Seongbukgu, Seoul 136-701, Korea Introduction Collagen is a fibrous protein present in almost every tissue, including the skin, blood vessels, tendons, and teeth It accounts for 30% of all protein in the body and is distributed throughout the entire body, making up 40% of the skin and 20% of the bones and cartilage, with additional collagen present in the blood vessels and intestines Collagen is generally extracted from livestock and is widely used in medical products, cosmetics, and foods [1] More recently, collagen has been used to improve skin elasticity and to produce functional foods intended for the treatment of obesity In addition, the interest in collagen peptide manufacturing technologies has increased due to rigorous research into the physiological features of the protein [2, 3] which has revealed that collagen has potential utility in the treatment of blood clots, the protection of the gastric mucous membrane, appetite suppression [4, 5], the inhibition of morphine analgesia, and memory improvement [6, 7] However, due to occurrences of bovine spongiform encephalopathy (BSE, also known as mad cow disease) and foot-and-mouth disease, the demand for livestock collagen has decreased This has resulted in increased interest in fish collagen as a replacement Fish skins contain abundant amounts of collagen The type I collagen present in fish skins has a lower melting point than that of livestock collagen, resulting in an excellent affinity for skin [8–12] However, there are disadvantages to fish collagen, including the protracted processing time required, as this collagen is obtained by extended periods of acid-soaking, and this approach is closely associated with the potential for environmental pollution and wastewater management issues Fish farms that raise red sea bream, sea bass, flatfish, and trout have been established in South Korea Among these fishes, approximately 2,385 tons of sea bass were produced 123 486 in 2009, which is more than one fourth of all fish produced in fish farms Because sea bass is mainly consumed as sashimi, a large quantity of sea bass skins, almost exclusively discarded, is produced as a byproduct Ultrasonic wave treatment generates a large amount of energy from the cavitation induced by the vibration This treatment increases the kinetic energy of the particles in the treated substance, due to a high local temperature, which provides sufficient energy for a reaction that may be effectively used in the extraction process, due to the shock effect of the ultrasonic energy [9] High-intensity ultrasonic technologies have recently been implemented in chemical synthesis, pharmaceutical preparations, and food processing [10] Studies of the applicability of these technologies in high-intensity biopolymers have increased [11], including the examination of the effects of ultrasonic treatment on the structural and functional properties of proteins [12], as well as on low-molecular-weight b-(1,6)branched b-(1,3)-D-glucans [13] and carrageenan [14] In addition, ultrasonic treatment has been used for the extraction of various bioactive materials [15–17] Therefore, the aim of this study was to examine collagen extraction methods using ultrasonic waves to achieve effective collagen extraction from sea bass skin Fish Sci (2012) 78:485–490 Collagen extraction through ultrasonic treatment An ultrasonic processor (VCX 750; Sonics & Materials, Newtown, CT) was used for the ultrasonic treatment The purified sample was soaked in 0.5 M acetic acid with a sample:solution ratio of 1:200 (w/v; based on the dry weight of the sample) The collagen was subsequently extracted by ultrasonic treatment under the conditions described in Table The control was extracted in 0.5 M acetic acid with a sample:solution ratio of 1:200 (w/v) To prevent an increase in temperature induced by the ultrasonic treatment, the sample jacket was cooled by a circulating water bath (NCB-2200; EYELA Co., Tokyo, Japan) maintained at 4°C Collagen yield The dissolved sample was centrifuged for 20 at 15,000 g The collagen yield was calculated as the ratio of supernatant protein to the total protein concentration The protein content was measured using the Biuret method [17] The formula used to calculate the collagen yield is: Collagen yield ð%Þ ¼ ðprotein concentration of supernatant = total protein concentrationÞ Â 100 Rate of collagen yield Materials and methods Materials The samples used were frozen sea bass skins Lateolabrax japonicas provided by the Oseong Fishery Co (Busan, Korea) The skins were cut into 1.0 1.0-cm segments after removing residual muscles and scales and washing in ice water to remove impurities A 0.5 M NaCl solution was added to the skins in a 1:20 skin/solution ratio (w/v), and the solution was stirred for 10 The precipitate was obtained via centrifugation for 10 at 6,000 g This process was repeated more than three times, and all processes were performed at below 4°C The precipitate was washed with cold distilled water, followed by 20 washes with cold ethanol to remove fat The sample that was purified through this fat removal process was stored below -20°C until use The collagen yield was increased by ultrasonic treatment The rate constant of the increase (ki) was calculated as follows: ki ¼ ðnt À n0 Þ=t where nt is the solubility after ultrasonic treatment for time t, n0 is the solubility before ultrasonic treatment, and t is the ultrasonic treatment time Sodium dodecyl sulfate-polyacrylamide gel electrophoresis Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed on 7.5% polyacrylamide gels containing 1% SDS [18] Samples for electrophoresis were Table Conditions of ultrasonic treatment Ultrasonic treatment Proximate analyses Proximate analyses were performed according to the methods of Association of Official Analytical Chemists [16] The analyses were performed in samples both before and after fat removal 123 Conditions Frequency 20 kHz Amplitude 20, 40, 60, and 80% Pulse on/off 20/20 s Time 0–24 h Temperature 4°C Fish Sci (2012) 78:485–490 487 Digestion of collagen Pepsin (Yakuri Pure Chemical Co., Tokyo, Japan) digestion was performed to confirm that the materials extracted from the ultrasonic treatment were indeed collagen In order to make this comparison, the gelatin obtained from collagen by heating for at 100°C also was treated with pepsin The concentrations of collagen and gelatin were adjusted to mg/ml, and 0.5% (w/w) pepsin was added to the collagen and gelatin and incubated for 0–30 at 10°C After this treatment, a solution containing M SDS, 2% mercaptoethanol, and 20 mM Tris–HCl (pH 8.0) was added to each sample, which was subsequently heated for at 100°C and analyzed by SDS-PAGE Commercially available acid soluble collagen (type I, from white rabbit skin; Sigma, St Louis, MO) was used as a standard collagen material Results Proximate analyses The proximate analyses of the samples both before and after the removal of fat by 100% ethanol are shown in Table The moisture contents of the samples after and prior to fat removal were 8.02 and 28.92%, respectively Conversely, a lower protein content was observed in the sample that had not been treated with ethanol (68.51%) than in the sample with the fat removed (90.39%) The ash contents of the samples following and prior to fat removal were 1.42 and 0.43%, respectively Collagen yield through ultrasonic treatment A 0.5 M acetic acid solution (1:200 sample/solution, w/v) was added to the fish skin to remove fat and impurities The results are shown in Fig The yield of collagen extracted from sea bass skin using the acid treatment for 24 h increased with reaction time, reaching approximately 20% after 24 h However, with the ultrasonic extraction method, collagen yield rapidly increased with increasing longer ultrasonic treatment time and higher amplitude, reaching yields of 45, 70, and 85% after 24 h at amplitudes of 20, 40, and 60%, respectively However, the yield of collagen at 80% amplitude was 80% after h, reaching 93% after 24 h following a gradual increase Effect of ultrasonic treatment on collagen extraction As shown in Fig 1, ultrasonic treatment in the presence of 0.5 M acetic acid resulted in a higher extraction efficiency of collagen from fish skin than that observed with the 0.5 M acetic acid solution without ultrasonic treatment In this study, the results depended on the ultrasonic amplitude, and the increased rate for each result (ki) from Fig was calculated to determine this dependency, as shown in Fig The ki value was 0.9 for the 0.5 M acetic acid (0% amplitude) solution without ultrasonic treatment However, the ki values were 2.0, 3.0, 3.8, and 9.3 for the ultrasonic treatments with amplitudes of 20, 40, 60, and 80%, respectively These results indicate that the ultrasonic treatments in the presence of the acid solution were 2.2-, 3.3-, 4.2-, and 10.3-fold faster than the acid treatments 100 Collagen yield (%) dissolved in Tris–HCl buffer (pH 8.0) containing M urea, 2% SDS, and 2% mercaptoethanol and were heated at 100°C for The protein components were stained with Coomassie Brilliant Blue R250 80 60 40 20 Table Proximate analyses of sea bass skin Constituents Contents (%) Before fat removal After fat removal Moisture 28.92 ± 0.45 8.02 ± 0.27 Crude protein 68.51 ± 2.40 90.39 ± 0.69 Crude fat 1.55 ± 0.03 Crude ash 0.43 ± 0.04 – 1.42 ± 0.07 Data are given as the average ± standard deviation (SD) from triplicate determinations 10 15 20 25 Ultrasonic treatment (h) Fig The yield of collagen extracted from sea bass skin by acid treatment and ultrasonic treatment at a constant frequency of 20 kHz and various amplitudes Acid-soluble collagen was extracted with 0.5 M acetic acid solution for 24 h (filled circles control) or sonicated in acid solution at different amplitudes: 20% (open circles), 40% (inverted filled triangles), 60% (inverted open triangles), and 80% (filled squares) 123 488 Fish Sci (2012) 78:485–490 SDS-PAGE analysis of collagen extracted from fish skin 10 ki 0 20 40 60 80 Amplitude (%) Fig The effect of ultrasonic treatment on the increased rate constants of collagen extraction from sea bass skin The increased rate constant for each result (ki) was calculated as: ki = (nt - n0)/t, where nt denotes solubility after ultrasonic treatment for ‘t’ time, and n0 denotes solubility before the ultrasonic treatment The subunit composition of the collagen extracted by the acid and ultrasonic treatments was analyzed by SDSPAGE, and the results of the extractions at ultrasonic amplitudes of 20, 40, 60, and 80% are shown in Fig The amounts of typical components of type I collagen, including the a1(a3), a2, and b chains, that were extracted by the acid treatment increased with increasing treatment period The collagen extracted by the ultrasonic treatment was studied for possible changes in the SDS-PAGE pattern The results revealed that the a1(a3), a2, and b chains appeared with longer treatment times and that they rapidly increased at higher amplitudes Certain unknown substances thought to be the products of collagen degradation induced by the ultrasonic treatment were also observed with longer treatment times However, there was no decrease in the main components of collagen as the concentration of these unknown substances increased Checking collagen extracted by ultrasonic treatment The pepsin digestion patterns of gelatin obtained by heating collagen for at 100°C were examined and compared to the SDS-PAGE patterns of collagen and gelatin extracted by the ultrasonic treatment (Fig 5) The collagen extracted by the ultrasonic treatment retained the main components of collagen, namely, the a1(a3), a2 and b chains, following pepsin treatment (Fig 5, lanes c, d, e) However, most of the components of the gelatin obtained by heating the collagen samples were found to be degraded upon pepsin treatment, whereas the proportion of lowmolecular-weight components increased (lanes g, h, i) Ultrasonic treatment (h) 30 25 20 15 10 0 20 40 60 80 Amplitude (%) Fig Minimum ultrasonic treatment time (h) required to reach a 20% collagen extraction yield with the ultrasonic treatment at various amplitudes alone, depending on the amplitude Approximately 20% of the collagen in the sea bass skin was extracted after 24 h with the 0.5 M acetic acid treatment The lengths of time required to obtain 20% extraction yields with the ultrasonic treatment at all tested amplitudes were measured, and the results are shown in Fig Treatment times of 10, 8, 6, and 1.5 h were required to achieve a 20% extraction yield of collagen at amplitudes of 20, 40, 60, and 80%, respectively These results indicated that the extraction time was reduced by 16-fold with the ultrasonic treatment at 80% amplitude 123 Discussion Collagen from fish skin is generally extracted by incubation with concentrated acid solutions for 2–3 days However, these processes are not effective because of the time required for extraction We investigated the effects of ultrasonic wave treatment on the efficacy of collagen extraction The results of the proximate analysis show that the constituents of the sea bass skin samples prior to fat removal differ from those of sea bass skin samples following fat removal (Table 2) It was considered that the sample was dehydrated by the fat removal process, resulting in the protein content of the skin samples following fat removal being relatively higher at the same weight As shown in Fig 1, the yield of collagen extracted from sea bass skin with a 0.5 M acetic acid solution increased Fish Sci (2012) 78:485–490 489 Fig Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis performed on 7.5% polyacrylamide gels of collagen extracted from sea bass skin by the acid treatment and ultrasonic treatment, respectively Collagen was extracted by ultrasonic treatment at 0% (a), 20% (b), 40% (c), 60% (d), and 80% (e) amplitude, respectively S Molecular weight marker, CS acidsoluble collagen (type I, from white rabbit skin), 3, 6, 9, 12, 24 duration of extraction (h) Fig SDS-PAGE patterns of collagen and gelatin during pepsin digestion at 10°C Analysis was performed on 7.5% polyacrylamide gels The gelatin was obtained from extracted collagen by heating for at 100°C S Molecular weight marker, a acid-soluble collagen (type I, from white rabbit skin), b collagen extraction by ultrasonic treatment, c, d, e collagen treated with 0.5% (w/w) pepsin for 10, 20 and 30 at 10°C, respectively, f gelatin, g, h, i gelatin treated with 0.5% (w/w) pepsin for 10, 20, and 30 min, respectively, at 10°C respectively (Fig 3) These results indicate that the extraction time was reduced by 16-fold using ultrasonic treatment at 80% amplitude The applications of ultrasonic waves in the food industry can be divided into two distinct categories, depending on whether they use low-intensity (typically under W/cm2) or high-intensity (typically in the range 10–1,000 W/cm2) ultrasonic waves [12] The high-intensity waves cause the physical disruption of the material or promote certain chemical reactions The main cause of those changes is the cavitation effect, a phenomenon of energy dissipation that propagates through a medium via the generation and subsequent collapse of vapor bubbles Our results, therefore, suggest that the ultrasonic treatment caused structural changes in the collagen fiber, which led to a more facile extraction of collagen when the ultrasonic treatment was applied compared with acid treatment alone The subunit composition (Fig 4) of collagen extracted by the acid and ultrasonic treatment, respectively, consisted of the a1(a3), a2, and b chains, but unknown substances thought to be sonication-induced collagen degradation products were also observed following longer treatment times This phenomenon prevented the extraction of collagen by the ultrasonic treatment, and the collagen obtained from the early stage of extraction was in a dissolved form These results suggest that the increase in collagen extraction yield induced by ultrasonic treatments was mainly due to changes in the collagen fiber structure and to the breakdown of the collagen structure In the ultrasonic treatment method, 60% amplitude was determined to be the most effective extraction condition for the production of high-quality collagen, resulting in a yield of approximately 40% Based on our data, we suggest that the ultrasonic treatment increases the yield of collagen and decreases the relatively long extraction time required when an acid solution is used for collagen extraction We also examined with the duration of incubation However, collagen yield following the ultrasonic treatment of sea bass skins soaked in 0.5 M acetic acid solution rapidly increased compared to that of the skins treated in acid without ultrasonic treatment Also, the yield of collagen extraction induced by ultrasonic treatment was affected by the amplitudes of the ultrasonic waves The rates of collagen extraction at 20, 40, 60, and 80% amplitude were approximately 2.2-, 3.3-, 4.2-, and 10.3-fold faster, respectively, than that of extraction with acid alone (Fig 2) Determination of the extraction times required to obtain a 20% collagen extraction yield with the ultrasonic treatment at the different amplitudes revealed that 10, 8, 6, and 1.5 h of extraction were required to reach a 20% yield at amplitudes of 20, 40, 60, and 80%, 123 490 whether the material extracted using the ultrasonic treatment was collagen or a component of another type, such as gelatin, due to sonication-induced structural changes It is known that collagen has a rigid structure and is dissolved by collagenase, but not by pepsin or trypsin However, collagen could potentially be dissolved by digestive enzymes, such as pepsin, if the material had undergone structural changes through the ultrasonic extraction process Therefore, we tested the integrity of the material by pepsin to determine whether the material extracted using ultrasonic treatment is a genuine collagen molecule or not Following the pepsin treatment, we found that the material had not dissolved, thereby confirming that the material extracted by the ultrasonic treatment was indeed collagen We suggest that more effective collagen extraction from fish skin is possible were acid and ultrasonic treatments to be used in tandem Acknowledgments This study was supported by the research fund from Korea Research Institute, Korea References Cavallaro JF, Kemp PD, Kraus KH (1994) Collagen fabrics as biomaterials Biotechnol Bioeng 43:781–791 Togashi SI, Takahashi N, Iwama M, Watanabe S, Tamagawa K, Tukui T (2002) Antioxidative collagen-derived peptides in human-placenta extract Placenta 23:497–502 Samonina G, Ashmarin I, Lyapina L (2002) Glyproline peptide family: review on bioactivity and possible origins Pathophysiology 8:229–234 Ashimarin IP, Karazeeva EP, Lyapina LA, Samonia GE (1988) The simplest proline-containing peptide PG, GP, PGP and GPGG: regulatory activity and possible sources of biosynthesis Biochemistry (Moscow) 63:119–124 Lyapina LA, Pastorava VE, Samonia GE, Ashimarin IP (2000) The effect of PGP peptide and PGP-rich substances on haemostatic parameters of rat blood Blood Coagul Fibrinolysis 11:1–6 123 Fish Sci (2012) 78:485–490 Samonia GE, Kopyliva GN, German SV, Umarova BA, Bakaeva ZV, Zeliaznik NJ Zhuykova SE, Sergeev VI, Lukjanzeva GV, Smirnova EA, Lelekova TV Endogenous peptides and gastric mucosal homeostasis In: Pathophysiology of organs and system Typical physiological processes (experimental and clinical aspects) Proc 2nd Russian Congress of Photophysiology Moscow, p 133 Gudasheva TA, Boyko SS, Akparov VKH, Ostroskaya RU, Scoldinov SP, Rozantsev GG, Votonia TA, Zherdev VP, Seredenin SV (1996) Identification of a novel endogenous memory facilitation cyclic dipeptide cyclo-proly in rat brain FEBS Lett 391:149–152 Ciarlo A, Paredi M, Fraga A (1997) Isolation of soluble collagen from hake skin (Merluccius hubbsi) J Aquat Food Prod Tech 6:65–77 Kruus P, O’Neill M, Robertson D (1990) Ultrasonic initiation of polymerization Ultrasonics 28:304–309 10 Ana CS, Mar V (2010) Effect of ultrasound on the technological properties and bioactivity of food: a review Trends Food Sci Technol 21:323–331 11 Kandos N, Lhche JL (2001) Sonochemistry of carbohydrate compounds Carbohydr Res 332:115–131 12 McClements D (1995) Advances in the application of ultrasound in food analysis and processing Trends Food Sci Technol 6:293–299 13 Shin HJ, Lee DC (2003) Study on the process to decrease the molecular weight of b-(1,6)-branched b-(1,3)-D-glucans Korean J Biotechnol Bioeng 18:352–355 14 Kim SM, Park SM, Choi HM, Lee KT (1999) Optimal processing parameters of low of low molecular weight carrageenan by ultrasound J Korean Fish Soc 32:495–500 15 Lee BS, Ko MS, Kim HJ, Kwak IS, Kim DH, Chung BW (2006) Separation of 6-gingerol from ginger and antioxidative activity Korean J Biotechnol Bioeng 21:484–488 16 Association of Official Analytical Chemists (AOAC) (1999) In: Helrich K (ed) Official methods analysis, 14th edn AOAC, Washington D.C 17 Gornall AG, Bardawill CJ, David MM (1949) Determination of serum proteins by means of the biuret reaction J Biol Chem 177:751–766 18 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 227:680–685 [...]... 30 5, 7, 8, 10–12 10 .3 32 .0 J–A 12 5, 9 36 .2–78.7 A 3, 5, 7–12 4.4–29.9 J–A 3, 5, 7, 9, 1 1, 12 4.8 34 .3 J–A Gobiidae Acentrogobius caninus Acentrogobius kranjiensis (Ak) 1,6 74 Acentrogobius malayanus (Am) 425 Acentrogobius viridipunctatus 16 5, 9, 1 1, 12 68.9–97.6 A Acentrogobius sp 1 11 5 4.4–9.7 J Amoya moloanus Amoya sp 1 2 15 7, 9 3, 5, 7, 9, 11 55.4–58.9 6 .3 15.7 A J 11 63. 6 A Arcygobius baliurus... sp 1 32 Drombus triangularis 4 5, 7, 10–12 5.5–7 .3 J 7, 9 14.0–27.5 A Eugnathogobius oligactis 4 11 6.9–7.6 J Favonigobius reichei (Fr) 38 3, 4, 7 12.0–20.8 J–A Glossogobius bicirrhosus 2 1 1, 12 29.2–58.1 A Glossogobius circumspectus Gobiopterus brachypterus 3 1 0, 11 21.5– 63. 1 A 24 10–12 13. 7–16.5 A Gobiopterus sp 1 (Gs) 71 5, 7, 8, 10–12 8.8– 13. 2 J Oxyurichthys microlepis (Om) 37 5, 7, 9, 1 1, 12 28.1–81.8... Res 52 :32 5 33 5 17 Yokoo T, Kanou K, Moteki M, Kohno H, Tongnunui P, Kurokura H (2006) Juvenile morphology and occurrence patterns of three Butis species (Gobioidei: Eleotridae) in a mangrove estuary, southern Thailand Ichthyol Res 53: 330 33 6 18 Tongnunui P, Ikejima K, Yamane T, Horinouchi M, Medej T, Sano M, Kurokura H, Taniuchi T (2002) Fish fauna of the Sikao Creek mangrove estuary, Trang, Thailand... Period analyzed 29.4 0.75 3 – RB2 33 .5 1.05 13 – RB3 RB4 33 .0 32 .0 0. 83 0.75 11 10 – – RB5 34 .5 0.94 22 – 36 .8 1 .3 – 0900 hours on 20 Oct 2010 to 1159 hours on 23 Oct 2010 RB7 35 .2 1 .3 – 0900 hours on 20 Oct 2010 to 02 53 hours on 23 Oct 2010 RB8 35 .8 1.4 – 2100 hours on 24 Oct 2010 to 1959 hours on 25 Oct 2010 RB6 BL body length, BM body mass BL (cm) Field Fish Sci (2012) 78:229– 235 231 DO meter Data logger... Institute, Fisheries Research Agency, 1551-8 Taira-machi, Nagasaki 851-221 3, Japan e-mail: ytohya@affrc.go.jp T Yasuda Graduate School of Agriculture, Kinki University, 33 27-204 Nakamachi, Nara 631 -850 5, Japan K Komeyama Faculty of Fisheries, Kagoshima University, 4-50-20 Shimoarata, Kagoshima 890-005 6, Japan e-mail: komeyama@fish.kagoshima-u.ac.jp K Kato Fisheries Laboratory of Kinki University, 31 53 Shirahama,... Center, Shimane University, 1060 Nishikawatsu, Matsue, Shimane 690-850 4, Japan e-mail: tyokoo@soc.shimane-u.ac.jp Introduction K Kanou Center of Water Environment Studies, Ibaraki University, 137 5 Ohu, Itako, Ibaraki 31 1-240 2, Japan e-mail: kkano@mx.ibaraki.ac.jp P Tongnunui Department of Marine Science, Faculty of Science and Fisheries Technology, Rajamangala University of Technology Srivijaya, Sikao,... Monogr 27 :32 5 34 9 Govt of Kenya (2008) Marine waters fisheries frame survey report, Fish Dept, Nairobi Arntz WE, Tarazona J (1990) Effects of El Nin˜o 1982– 83 on benthos, fish and fisheries off the South American Pacific Coast Elsevier Oceanogr Ser 52 :32 3 36 0 Schwing FB, Palacios DM, Bograd SJ (20 03) El Nin˜o impacts of the California current ecosystem NOAA Fisheries, SWFSC, CA Schwing FB, Murphree T, deWitt... purposes, vol 6 FAO, Rome, pp 35 78 36 03 27 Nelson JS (2006) Fishes of the world, 4th edn Wiley, New York 28 Koumans FP (19 53) Gobioidea In: Weber M, Beaufort LF (eds) The fishes of the Indo-Australian Archipelago X Brill, Leiden 29 Hoese DF (1986) Gobiidae In: Smith MM, Heemstra PC (eds) Smith’s sea fishes Springer, Berlin, pp 774–807 30 Kottelat MA, Whitten J, Kartikasari SN, Wirjoatmodjo S (19 93) Freshwater... Acentrogobius caninus, Acentrogobius viridipunctatus, Acentrogobius sp 1, Amoya moloana, Arcygobius baliurus, Brachygobius sp 1, Drombus triangularis, Eugnathogobius oligactis, Glossogobius bicirrhosus, Glossogobius circumspectus, Gobiopterus brachypterus, Oxyurichthys microlepis, Oxyurichthys sp 2, Pseudogobius javanicus, Pseudogobius sp 1– 4, Stigmatogobius sp 1–2 and Yongeichthys nebulosus; group III, no diagnostic... Morioka (&) Fisheries Division, Japan International Research Center for Agricultural Sciences, 1-1 Owashi, Tsukuba 30 5-868 6, Japan e-mail: moriokas@affrc.go.jp P Cacot International Cooperation Center of Research in Agronomy for Development, TA B-110/A, Campus International de Baillarguet, 34 398 Montpellier Cedex 5, France M Moteki Department of Ocean Sciences, Faculty of Marine Science, Tokyo University