Fish Sci (2010) 76:1–11 DOI 10.1007/s12562-009-0186-x ORIGINAL ARTICLE Fisheries Classification of fish schools based on evaluation of acoustic descriptor characteristics Aymen Charef • Seiji Ohshimo • Ichiro Aoki Natheer Al Absi • Received: 27 May 2009 / Accepted: 15 October 2009 / Published online: December 2009 Ó The Japanese Society of Fisheries Science 2009 Abstract Acoustic surveys were conducted from 2002 to 2006 in the East China Sea off the Japanese coast in order to develop a quantitative classification typology of a pelagic fish community and other co-occurring fishes based on acoustic descriptors Acoustic data were postprocessed to detect and extract fish aggregations from echograms Based on the expert visual examination of the echograms, detected schools were divided into three broad fish groups according to their schooling characteristics and ethological properties Each fish school was described by a set of associated descriptors in order to objectively allocate each echo trace to its fish group Two methods of supervised classification were employed, the discriminant function analysis (DFA) and the artificial neural network technique (ANN) We evaluated and compared the performance of both methods, which showed encouraging and about equally highly correct classification rates (ANN 87.6%; DFA 85.1%) In both techniques, positional and then morphological parameters were most important in discriminating among fish schools Fish catch composition from midwater trawling validated the fish group classification through one representative example of each grouping Both methods provided the essential information A Charef (&) Á I Aoki Graduate School of Agriculture and Life Science, University of Tokyo, Bunkyo, Tokyo 113-8657, Japan e-mail: aymen_charef@yahoo.com S Ohshimo Seikai National Fisheries Research Institute, Fisheries Research Agency, Nagasaki 851-2213, Japan N Al Absi Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164-8639, Japan required for assessing fish stocks Similar techniques of fish classification might be applicable to marine ecosystems with high pelagic fish diversity Keywords Acoustic descriptor Á Artificial neural network Á Discriminant function analysis Á Fish classification Á Species identification Introduction The northern part of the East China Sea represents one of the main spawning and nursery areas of small pelagic fishes in the waters off of the Japanese coast It also constitutes an important fisheries ground for commercially valuable pelagic fishes During the last half decade, the average landing was estimated to be roughly 250,000 tons per year and was composed of Japanese anchovy Engraulis japonicus, round herring Etrumeus teres, jack mackerel Trachurus japonicus, chub mackerel Scomber japonicus and spotted chub mackerel Scomber australasicus (according to statistics from the Ministry of Agriculture, Forestry and Fisheries, Government of Japan) The fish stock size assessment is crucial for fisheries management in these waters Broadly, the main assessment techniques are based on the virtual population analysis (VPA) method This method makes use of commercial catches, which might bias the assessments and then generate very serious overfishing problems [1, 2] To eliminate such complications, reliable and fishery-independent data are needed Hydroacoustic methods are one of the few techniques used in order to provide fisheries independent quantitative estimates of fish stocks Fisheries acoustics have experienced dramatic development in technologies and data management Acoustic surveys using quantitative scientific 123 echo sounders commonly employed to determine the abundance and biomass of pelagic fish are becoming increasingly important for the management of pelagic fisheries [3] Owing to the common aggregative behavior, small pelagic species appear in echograms as a mixture of diverse fish assemblages [4] Echo integration is used to estimate fish quantity since the sampled volume contains overlapping target fish echoes [3] The obtained target strengths and the backscattering strength can be translated into biomass units if the proportions of different species and their length distribution and target strength on fish size are known In such a context, distinguishing among fish targets is greatly needed to deal with each target fish echo separately Therefore, identification of echo traces of fish schools is crucial in conjunction with accurate acoustic surveys to give reliable estimates of target strength and consequently improve the fish stock assessment The classification and subsequent identification of acoustic targets to taxa or species are still the great challenge of fisheries acoustics [5, 6] Species identification has been limited by the difficulty in objectively classifying backscattered energy of echo traces to species [6, 7] Echotrace classification defined as the detection and description of aggregations in acoustic data can be used to study behavioral and ecological processes in aquatic environments [8] It is generally agreed that besides integration of target species’ biomass, useful information, such as features from digitized echograms, can be extracted from the acoustic data Many studies have attempted to develop echo-trace classification in order to study shoaling behavior and predator-prey interactions, to characterize fish aggregations, their spatial distribution and their relationship to environmental variables; see Horne [9] for a review First attempts at fish identification introduced basically subjective and time-consuming methods These methods involved expert scrutiny of echograms combined with concurrent trawling data Visual scrutiny of acoustic data depends on human experience and is therefore subject to biases and difficult to be quantified This makes objective methods more efficient, timely, less or not dependent on subjective interpretation, and controlled by evaluating their accuracy [10] These automated methods require data processing and detection of acoustic features from echograms as a first step, and secondly, description of selected schools characteristics with a set of descriptors [11] They aim to train an algorithm on a set of identified, single species schools Then the algorithm is adopted to identify other schools [12, 13] Success of objective methods relies primarily on a suitable choice of acoustic descriptors concerning number and efficiency In the case of high diversity ecosystems, such as the East China Sea, where small schools are numerous, species classification highly depends on verification via trawl data 123 Fish Sci (2010) 76:1–11 In the East China Sea, some attempts to estimate the pelagic fish populations’ biomass were made with acoustic surveys These studies were restricted to subjective classification of fish species [14, 15], while limited to single species such as anchovy [14, 16] and sardine [17, 18] in other works In this work, we applied two objective tools of supervised echo-trace classification, discriminant function analysis (DFA) and artificial neural network (ANN) The aim of this paper is to describe and to evaluate the efficacy of the two methods, based on a set of acoustic descriptors, in objectively classifying fish schools of pelagic fish community and other co-occurring fishes such as pearlside and lantern fish Materials and methods Data collection Acoustic surveys were conducted annually in the late summer from 2002 to 2006 by the Japanese Fisheries Research Agency on board the RV Yoko Maru Surveys were carried out along 27 parallel transects spaced by 10 nautical miles (Fig 1) During surveys, vessel speed was approximately 10 knots and total length of transects ranged from 593 to 828 nautical miles (Table 1) Fig Study area and acoustic survey scheme Fish Sci (2010) 76:1–11 Table Year, beginning and end dates, total length of transects, number of detected schools and number of stations of CTD casts and midwater trawls during each acoustic survey Year Begin date End date Total length of transects (nautical miles) Number of detected schools Number of stations 2002 22 August 24 September 828 221 17 2003 27 August 25 September 828 187 21 2004 24 August 12 September 593 163 12 2005 24 August 10 September 805 168 18 2006 23 August September 791 91 20 Acoustic data were collected using a calibrated hullmounted SIMRAD EK505 scientific echo-sounder system operating at 38 kHz with a time-varied gain function set at 20 log R The echo-sounder pulse length was ms, its ping rate was 0.33 ping s-1, and its estimated sound speed 1500 ms-1, giving a target resolution of 0.001 s 1500 m s-1/2 = 0.75 m Acoustic measurements were logged continuously during all surveys and recorded only during daytime Small pelagic fish species may reduce the risk of daytime predation by schooling [19] The schooling behavior typically characterizes each fish school in daylight, which is essential for the fish identification However, during twilight and nighttime, fish schools scatter and overlap, which biases the fish identification in acoustic processing [4, 20] Acoustic data processing Acoustic data were postprocessed using Echoview Software version 4.50 [21] The seafloor was automatically detected using the ‘‘maximum Sv backstep’’ algorithm, where the backstep was set at m Data deeper than m above the selected bottom line were removed due to the false bottom detection Data shallower than 10 m were also removed from analyses to eliminate the transmit pulse and reduce backscatter by surface bubbles A background threshold of -67 dB was applied equivalently to all echograms The threshold was determined by analyzing a subset of data collected from each year and allowed accurate detection of all possible aggregations of target fishes Fish aggregations were detected and characterized using the ‘‘Schools detection’’ module implemented in Echoview Input parameters were set according to schools’ features observed in acoustic records The algorithm pattern required schools to be at least m long and m high Adjacent aggregations were linked to shape one school if the maximum horizontal linking distance was 15 m and maximum vertical connection distance m Then echograms were visually inspected, and doubtful and ‘false’ detections (scattering layer, acoustic interference) were removed Connected aggregations with dimensions smaller than the minimum school length and height parameters were discarded For each detected acoustic target, a set of five school descriptors was calculated and extracted, and they fell into three categories (Table 2): (1) morphological: school length, height and height mean; (2) energetic: mean volume backscattering strength (Sv); (3) positional: mean school altitude (Depth) Midwater trawl catch data Midwater trawling was used to identify acoustic targets and to establish their weight composition Midwater trawling was only performed at nighttime because of the high netavoidance rate of fish targets in the daytime, which makes it difficult to sample the observed fish schools in acoustic recordings [22] Visual inspection of echograms for several hours permitted the characterization of schooling behavior and swimming depth of target species The position of the Table Definitions and units of school descriptors used in both analysis methods Descriptor Unit Indication Length m Height m The horizontal distance along the transect from the first to last ping crossing the school The vertical distance separating the maximum and minimum depths of the rectangle bounding the school Height mean m The mean distance from the upper to lower limit along each ping crossing the fish school dB The mean energy produced by pixels shaping a fish school, which indicates its mean density m The distance from the sea surface to the geometric center of the fish school Morphological Energetic Mean volume backscattering strength (Sv) Positional Mean school depth (Depth) 123 Fish Sci (2010) 76:1–11 Fig Acoustic recordings showing typical schools of three different fish groups trawl stations was decided beforehand according to the location of peculiar fish concentrations detected during acoustic surveys in the daytime A total of 88 midwater trawls were conducted (Table 1) Towing speed was approximately knots for a towing time of 30 Towing depth was targeted to fish schools by adjusting the towing speed and warp length The mouth of the trawl net was approximately 20 m by 20 m, and the mesh sizes of the cod end and the inner bag were 60 and 20 mm, respectively The trawl catch was separated by species, and the total weight of each species was determined Other data Conductivity-temperature-depth (CTD) profiles were taken along the survey tracks at the beginning of each trawling operation The on-board data recording and entry system was deployed to record series of time (GMT), geographic position and the EK 500 vessel log Fish-group classification School images were selected and allocated to a species through visual expert examination of the echogram displays based on prior experience knowledge, in conjunction with the interpretation of echograms The identified target fishes were classified into three types of fish groups according to their schooling characteristics and ethological properties The verification of this typology also involved the results of the midwater trawl catch amount and composition The classification was partially based on the previous findings of Ohshimo [15] from acoustics surveys conducted following a similar survey scheme on the same study area The first type (G1) consisted of compactly aggregated schools, assumed to be Japanese anchovy and round herring, within the upper layer of the water column The 123 second group (G2) appeared in the midwater layers, mostly above the bottom rise structure, and it was thought to be composed of jack mackerel and chub mackerel The last group (G3), assumed to consist of lantern fish and pearlside, occurred in demersal layers mainly along slopes and formed horizontally elongated schools in contact with the seabed (Fig 2) Some detected fish schools that did not fall within this typology were neglected Statistical analysis Discriminant function analysis (DFA) is a well-known statistical procedure used to predict group membership based on a combination of the interval variable [23] The five school descriptors constituted the predictor variables for this discrimination analysis, whereas the dependent variable was fish group (G1, G2, G3) defined a priori on the basis of visual expert scrutiny and direct sampling results DFA was performed using SPSS (version 6.0) based on Mahalanobis distances (D) Mahalanobis distance is the distance between a case and the centroid for each fish group (of the dependent variable) in attribute space By this procedure, each school is allocated to the fish group for which D has the smallest value [24] Classification accuracy was estimated with leave-one-out cross-validation, in which the discriminant function is first derived from only n - schools and then used to classify the other school left out The procedure is repeated n times, each time omitting a different observation [25] DFA was applied for overall years data pooled together Artificial neural networks Artificial neural networks (ANN) were also used as the method of species classification and identification of fish schools from acoustic data They imitate human neuron functioning and solve problems by applying knowledge gained from past experience to new situations [26] Fish Sci (2010) 76:1–11 Results Classification using discriminant function analysis Fig Network architecture for the model used in this study A multiple layer perceptrons (MLPs) neural network was constructed and computed using Matlab 6.0 MLPs are the most commonly and the simplest network type used, primarily due to their speed and versatility [27] They consist of three feed-forward layers: input, hidden and output (Fig 3) The input layer was composed of five variables The number of nodes in the hidden layer was determined by testing the performance of the model using a range of node numbers The dependent variable fish groups represented the output layer The data set was split into a training set and validation set consisting of 70 and 30% of the identified schools, respectively, with the same proportion of each fish group Based on supervised learning, the neural network was trained by means of a backpropagation learning algorithm (BP) in order to develop the ability to correctly classify new fish schools from further acoustic data [28] The school fish’s classifications based on their relative descriptors occurred in two major phases First, during the learning phase, internal parameters within the network were adjusted iteratively The performance of the network, equivalent to classifying schools into fish groups accurately, was maximized; this stage continued until there was no further increase in network performance or classification success Although the aim of the training is to reduce the error as much as possible, reducing the error too much leads to the network learning the noise rather than underlying relationships Precautions were taken to avoid over-fitting (over-training) of the network’s model Finally, during the validation phase, which is the second phase, the optimal network was applied to test sets, along with crossvalidation Discriminant function analysis was computed using 830 detected schools and five acoustic descriptors (Tables 1, 3) Since the dependent variable, fish school, has three groups, two canonical discriminant functions were determined Both functions were significant, but nearly all of the variance in the model is captured by the first discriminant function The small Wilk’s lambda coefficients indicated also that only the first function is useful The eigenvalues confirmed the significant difference between both discriminant functions The standardized discriminant function coefficients were used to compare descriptors measured on different scales Coefficients with large absolute value correspond to variables with greater discriminating ability This implies that within the first function, for instance, depth contributed the most Thus, descriptors in rank order of efficacy in discriminating fish schools are depth, height, height mean and length, while mean volume backscattering strength Sv comes last The confusion matrix showed the results of the DFA using five acoustic descriptors for discriminating fish schools from survey data of years (Table 4) Emboldened values on the main diagonal of each confusion matrix represent the number of schools that were correctly identified within every fish group The overall correct classification was evaluated at 85.1% The correct recognition rates per group showed high scores for G1 schools Almost 95% were well assigned and distinguished from other groups G2 schools represent 57% of the total number of schools and were the least correctly classified with a relatively low rate of 80.3% The proportion of G3 schools is small, with only 13.25%, and had a correct classification score of 81.8% Classification using an artificial neural network Application of the trained network to years of pooled acoustic data resulted in predicted species compositions that corresponded well to those observed with an overall correct classification evaluated at 87.6% for the validation data set (Table 5) The model performed well for Table Results of discriminant analysis using five descriptors for overall years data Analyzed school group discriminant function Wilk’s k First function Second function % of variance Eigenvalue 0.269 94.6 2.291 0.885 5.4 0.130 Standardized canonical discriminant function coefficient Depth Sv Significance level Height Height mean Length 0.971 0.305 -0.304 0.158 0.043 0.000 -0.296 0.537 -0.424 0.835 -0.093 0.000 123 Fish Sci (2010) 76:1–11 Table Confusion matrix of DFA analysis Predicted group G1 G2 Total % Correct G3 Observed group G1 234 10 244 95.9 G2 62 382 32 476 80.3 G3 Overall 15 90 110 81.8 301 407 122 830 85.1 Number of schools from each group (true classification) distributed over predicted groups Values in bold denote correctly classified schools Fig Proportion of the contribution factor of each descriptor used as input into the artificial neural network Table Results of ANN classification from the two data sets Predicted group G1 G2 Total % Correct 97.7 G3 Observed group Training data set G1 169 173 G2 21 297 13 331 89.7 G3 69 76 90.8 Overall 191 Validation data set 306 83 580 92.2 G1 71 74 95.9 G2 15 120 143 83.9 G3 28 33 84.8 87 127 36 250 87.6 Overall Values in bold represent correct assignment classifying G1 schools with a correct classification rate of 95.94%, but less for G3 and G2 schools, with 84.84 and 83.91%, respectively The contribution factor of a variable is the sum of the absolute values of the weights generated from this particular variable It reveals the importance of input variables, descriptors, to classify fish schools The analysis showed similar ordering of descriptor categories to DFA results and indicated that the heaviest impact in classifying was assigned to positional, morphological and then energetic properties of a school However, the ascending order within the morphological descriptors category differs slightly, though depth was the most efficient descriptor (Fig 4) Validation with catch data Midwater trawling catch assisted in fish identification simultaneously with visual scrutiny of echograms The recorded acoustic data in daytime permitted to observe typical shapes of fish schools and then facilitated their 123 identification Examination of catch data over all 88 tows showed that the dominant target species was jack mackerel, which contributed 22% by weight of the total catch, followed by Japanese anchovy (18.4%) and lantern fishes (16%) (Table 6) Round herring was an exception in 2002 and was the most abundant species, reaching 20% of the total catch by weight in the same year The non-target species that did not fall in the three identified categories of major species were clustered into one group as ‘‘others’’ and represented around 28% of the total catch amount (Table 6) Catch composition was also valuable to verify the classification of target species into three groups of fish schools Table shows catch composition data from selected trawls hauled near the locations where schools of G1, G2 or G3 were observed in daytime Each group of species was assigned according to the most dominant species comprised in each trawl catch A summary of trawl hauls with fish schools matching with acoustically detected schools is shown in Table Looking at both tables simultaneously (Tables 7, 8) permitted examining the catch composition according to the amount and number of trawl hauls In the overall data for years, the number of detected schools evenly matched with the catch amount of target species The correspondence between detected and caught G1 schools was estimated to be 44% of the catch from 11 hauls, mainly made up of Japanese anchovy as it is the most abundant species in G1 The mismatch is primarily due to the high amount of catch of the G2 and G3 species Around 34% of the detected G2 schools were validated by catch data from 11 hauls Other co-occurring species, mainly represented by puffer fishes and squid, made up 43% of the total catch amount and were fairly abundant in 15 hauls; some of them were small catches (less than kg) In the case of G3 schools, nearly 41% of identified schools were validated by catch results Bycatch species that were caught during the same trawl hauls represented 33% of the total catch but belonged to one trawl haul Fish Sci (2010) 76:1–11 Table Catch amount (kg) by midwater trawling of abundant species assumed to compose acoustically detected fish schools Fish group Species 2002 2003 2004 Engraulis japonicus Japanese anchovy 44.2 (13.4) 7.2 (1.8) 55.4 (42.0) 36.9 (9.3) Etrumeus teres Round herring 67.7 (20.5) 10.8 (2.7) 0.9 (0.7) 5.5 (1.4) 23.0 (1.8) 107.8 (7.0) Sardinops melanostictus Decapterus macrosoma Japanese sardine 0.1 (0.03) 0.1 (0.1) 0.1 (0.03) 0.7 (0.2) 1.0 (0.1) Shortfin scad 7.3 (2.2) 6.4 (1.6) 3.4 (2.6) 0.9 (0.2) 3.0 (1.1) 21.0 (1.4) Decapterus maruadsi Round scad 0.2 (0.1) 28.1 (7.1) 0.3 (0.1) 28.6 (1.9) Scomber japonicus Chub mackerel 2.1 (0.6) 0.5 (0.1) 0.6 (0.5) 8.4 (2.1) 11.6 (0.8) Scomber austratasicus Spotted chub mackerel 0 2.8 (2.1) 8.4 (2.1) 5.6 (2.0) 16.8 (1.1) Trachurus japonicus Japanese jack mackerel 38.6 (11.7) 215.1 (54.2) 35.0 (26.5) 37.6 (9.5) 15.1 (5.3) 341.4 (22.2) Diaphus spp Lantern fishes 43.3 (10.9) 3.8 (2.9) 191.7 (48.5) 7.3 (2.6) 246.1 (16.0) Maurolicus japonicus Pearlside 0.8 (0.2) 0.1 (0.03) 0.1 (0.1) 50.2 (12.7) 51.3 (3.3) Arothron spp Puffer fishes 113.8 (34.4) 60.6 (15.3) 8.6 (6.5) 0.8 (0.2) 0.6 (0.2) 184.3 (12.0) Auxis rochei Bullet tuna 6.3 (1.9) 0 1.9 (0.5) 3.9 (1.4) 12.1 (0.8) Diodon hystrix Loglio edulis Porcupinefish Swordtip squid 0.9 (0.3) 5.9 (1.8) 0.2 (0.1) 8.7 (2.2) 8.3 (6.3) 16.6 (4.2) 32.0 (11.2) 14.4 (5.1) 33.1 (2.2) 53.8 (3.5) Psenopsis anomala Melon seed Scientific name G1 G2 G3 Others 2005 2006 Total Common name 139.7 (49.1) 283.4 (18.4) 0.1 (0.4) 0.2 (0.1) 1.7 (0.4) 5.9 (1.5) 4.7 (1.7) 12.4 (0.8) Todarodes pacificus Japanese common squid 6.7 (2.0) 2.6 (0.7) 0.7 (0.5) 2.0 (0.5) 1.6 (0.6) 13.5 (0.9) Others 36.4 (11.0) 39.3 (9.9) 12.4 (9.4) 0.8 (0.2) 32.7 (11.5) 121.5 (7.9) 330.9 396.8 131.9 395.4 284.5 1539.5 Total catch of all species Values between brackets represent percentage (%) Table Comparison of acoustically detected schools with trawl catch composition Fish Number Caught species group of hauls G1 Japanese Anchovy G2 Round Herring Japanese sardine G3 Scomber spp Trachurus spp Decapterus spp 1.2 (0.3%) 142.2 (31.3%) (0.4%) Lantern fishes Pearlside Others Detected schools G1 11 184.6 (40.7%) 16.5 (3.6%) 0.5 (0.1%) G2 28 58.6 (11.9%) 28.3 (7.3%) 0.05 (0.01%) 11.7 (2.4%) 138.7 (28.1%) 20.2 (4.1%) 22.8 (4.6%) G3 10.4 (5%) 41 (19.7%) 0.2 (0.1%) 2.1 (1%) 92.4 (20.4%) 0.1 (0.02%) 30.35 (6.7%) 0.1 (0.02%) 213.42 (43.2%) 34.2 (16.4%) 51 (24.5%) 69.45 (33.3%) Upper line of each row represents catch amount per kg Lower line indicates percentage of catch amount Values in bold represent fish species included in each group Discussion Comparison of classification techniques In this study, ANN and DFA models were optimized in order to classify fish schools Both techniques showed nearly similar recognition performance The overall classification rate was higher for ANN than DFA, but nevertheless was only slightly higher As for the three fish groups’ relative classifications, there were minor differences in classification success based on the two specified methods In particular, differences were trivial for G1 schools, whereas the successes of discrimination of G2 and G3 schools were significantly more important with ANN than DFA (Tables 4, 5) The particularly effective power of ANN to classify fish schools is attributed to its ability to 123 Fish Sci (2010) 76:1–11 Table Summary of acoustically detected schools with the most abundant caught species in number of trawl hauls Caught species G1 G2 Total G3 Others Detected schools a G1 2 11 G2 G3 11 15a 28 Five hauls scored less than kg of total catch per haul handle non-linear relationships between descriptors and dependent variables, through the presence of many intervening information-processing units, which each uses the binary logistic activation function [27] A further advantage of ANN is the small impact of extreme values on discrimination success and the absence of any specific assumptions on the distribution of the data In fact, ANN established functional relationships of the data by learning from the input training data set [29, 30] On the other hand, despite these advantages, a liability of its application is that it needed much more computing time than discriminant analysis, especially during optimization procedures such as weight analysis Similar performances of ANN and DFA in identifying fish schools that have been found in several studies corroborate our finding that ANN is more effective than DFA [12, 13, 31] Moreover, they reported better, sometimes far better, overall classification rates These mentioned case studies inferred that an increasing number of descriptors should lead to an improvement in discrimination effectiveness However, Scalabrin et al [32] found a lower rate when classifying only three species using nine school parameters Theoretically, the greater the number of parameters that can be included in the model, the more likely the analysis will assign a school image to the correct group [13] However, in our practical analysis, for both classification methods we were limited to five acoustic descriptors as input variables Parameters controlling fish-group classification The fish schools’ classification is defined as the discrimination of acoustic backscatters to the species, genus or group level, depending on the richness of fish diversity [10] In this work, the classification of fish echo traces into three fish groups was reliable due to the high fish diversity in the East China Sea The acoustic aggregations of the numerous target species were categorized based solely on their schooling characteristics The feasibility of this approach is justified by the existence of acoustic populations; groups of echo traces show a consistent pattern in 123 space and time at a regional scale [33] In tropical waters, Gerlotto succeeded in dividing highly multispecific fish communities into four fish acoustic populations [34] However, for some marine systems at high latitudes, such as the North Atlantic Ocean, species richness is relatively poor The low number of target fishes and the occurrence of monospecific schools permitted a lower level of discrimination and yielded a higher successful classification rate [32] The vertical distribution of fish schools in the water column gave evidence of the typology applied in this study G1 schools existed predominantly in the upper layer of the water column above the thermocline detected at approximately 50 m depth (Fig 5a) The G2 schools were observed within a deeper layer below the thermocline G3 schools were distributed in the bottom half of the water column below 150 m depth until the closest layer to the sea bottom The vertical distribution of G3 species is in agreement with the vertical range (160–200 m) reported by Fujino et al [35] in the case of pearlsides and below 200 m depth in the case of mesopelagic lantern fishes [36] The vertical distribution of fish schools exhibited a noticeable pattern that corresponds to the overlap of G1– G2 schools and G2–G3 schools in water layers at 60–80 and 160–180 m, respectively Fish schools co-occurring within these depths could not be discriminated properly on the basis of the positional descriptor Both methods (DFA and ANN) resulted in relatively weaker performance within these overlap layers; the correct classification rate did not exceed 88% Notwithstanding the fair limitation of fish-group classification within ‘overlap’ layers, the results of DFA and ANN revealed that the school’s altitude in the water column was the most effective acoustic descriptor in successfully discriminating schools into the three groups On the other hand, morphological acoustic descriptors and backscattered volume Sv contributed to distinguishing among species In fact, the G3 species pearlsides and lantern fishes formed generally large elongated aggregations that were fairly dense and characterized by relatively low Sv values The G2 species jack mackerel, spotted mackerel and chub mackerel aggregated in relatively smaller schools marked by higher Sv values (Fig 5b, c, d) Although the vertical distribution of the target fishes cannot be addressed in detail within the scope of this paper, it provided valuable information about the environmental and physiological properties of identified target species The occurrence of Japanese anchovy, Japanese sardine and round herring above the thermocline was most likely related to temperature gradient patterns Temperature at the sea surface varied between surveys from 26.5 to 28.8°C (Fig 6), and below 60 m depth, temperature profiles were Fish Sci (2010) 76:1–11 facilitated the identification of species confined to the upper layers [14] The availability of food as well as avoidance of predation could also be plausible key factors concerning the vertical distribution patterns [37] Small fish may have migrated to a depth level with a lower concentration of larger fish to avoid predation [36] Myctophids and pearlsides fishes feed on zooplankton They ascend from the sea bottom at night following food and prey patterns and are thought to compete for food with pelagic fish within the upper layer [14, 15] Fish identification improvement Fig Distribution of detected schools in relation to depth (a), height (b), length (c) and mean backscatter volume Sv (d) fairly homogenous The thermocline might have played the role of a barrier that restricted the migration of these species to deeper layers Thus, the thermal barrier implicitly Several works have been using multiple frequency echo sounding to allocate fish echoes to species by using the frequency difference in mean volume backscattering strength (MVBS) and target strength differencing [38–40] These methods have shown considerable promise and provided high rates of correct classification in restricted ecological situations, that is, none have provided a classifier that can be applied over broad ranges of time and space [41] In the East China Sea particularly, owing to the high fish diversity, the use of an extended number of narrowband acoustic frequencies may facilitate the identification of fish species More precisely, low frequencies might be the best aid to increase species discrimination, for instance, midwater layers of mesopelagic fish appear much stronger on 12 kHz than on 38 kHz [42–44] Simultaneously, with more accurate acoustic surveys, additional trawl data should facilitate the identification of fish species within each group In parallel, the increase in the amount of collected data enhances ANN training and thus its efficacy Taking the advantage of its fast performance and the speed of processing using modern computers, the application of ANNs in real-time classification would be advantageous in fisheries stock assessments In the same order of magnitude, further statistical analysis should be performed to evaluate the consistency of acoustic data and trawl data Ideally, the fish schools detected during daylight acoustic surveys will be caught using the midwater trawling conducted only at nighttime The horizontal migration of fish may bias the verification of identified fish schools using trawl data However, in this work, the time lag was neglected since midwater stations were meticulously chosen to correspond to locations of target fish schools observed previously in echograms Quantification of uncertainty of the match between both data sets (acoustic and trawl data) may lead to improving the objectivity of fish identification and classification In conclusion, this study demonstrated that the neural network can perform reasonably well in classifying fish schools and that it performs slightly better than DFA This 123 Fish Sci (2010) 76:155–159 Michel F, Rudloff V (1989) Isolation and characterization of the rainbow trout erythrocyte band-3 protein Eur J Biochem 181:181–187 Aoki T, Fukai M, Ueno R (1996) Glycoproteins in red cell membranes from carp and rainbow trout Fish Sci 62:498–499 Barret AJ (1980) Fluorimetric assays for cathepsin B and cathepsin H with methylcoumarylamide substrates J Biochem 187:909–912 Aoki T, Yokono M, Ueno R (2002) A cathepsin B-like enzyme from mackerel white muscle is a precursor of cathepsin B Comp Biochem Physiol 133B:307–316 Lowry OH, Rosbrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the folin phenol reagent J Biol Chem 193:265–275 10 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 227:680–685 11 Hirano H, Watanabe T (1990) Microsequencing of proteins electrotransferred onto immobilizing matrices from polyacrylamide gel electrophoresis: application to an insoluble protein Electrophoresis 11:573–580 12 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets; procedure and some applications Proc Natl Acad Sci USA 76:4350–4354 13 Dominingo A, Marco R (1989) Visualization under ultraviolet light enhances 100-fold the sensitivity of peroxidase-stained blots Anal Biochem 182:176–181 14 Yamashita M, Konagaya S (1991) Hydrolytic action of salmon cathepsins B and L to muscle structural proteins in respect of muscle softening Nippon Suisan Gakkaishi 57:1917–1922 159 15 Kirschke H, Kembhavi AA, Bohley P, Barrett AJ (1982) Action of rat liver cathepsin L on collagen and other substrates Biochem J 201:367–372 16 Yamashita M, Konagaya S (1990) Participation of cathepsin L into extensive softening of the muscle of chum salmon caught during spawning migration Nippon Suisan Gakkaishi 56: 1271–1277 17 Aoki T, Nakano T, Ueno R (1997) Purification and some properties of a latent form cathepsin L from mackerel white muscle Fish Sci 63:824–829 18 Yamashita M, Konagaya S (1990) Purification and characterization of cathepsin L from the white muscle of chum salmon, Oncorhynchus keta Comp Biochem Physiol 96B:247–252 19 Tsunemoto K, Osatomi K, Nozaki Y, Hara K, Ishihara T (2004) Molecular characterization of cathepsin L from hepatopancreas of the carp Cyprinus carpio Comp Biochem Physiol 137B: 107–114 20 Smith SM, Kane SE, Gal S, Mason RW, Gottesman MM (1989) Glycosylation of procathepsin L does not account for species molecular-mass differences and is not required for proteolytic activity Biochem J 262:931–938 21 Yamashita M, Konagaya S (1992) An enzyme-inhibitor complex of cathepsin L in the white muscle of chum salmon (Onchoryncus keta) in spawning migration Comp Biochem Physiol 103B: 1005–1010 22 Yamamoto K, Ueno E, Uemura H, Kato Y (1987) Biochemical and immunochemical similarity between erythrocyte membrane aspartic proteinase and cathepsin E Biochim Biophys Res Commun 148:267–272 123 Fish Sci (2010) 76:161–165 DOI 10.1007/s12562-009-0182-1 ORIGINAL ARTICLE Chemistry and Biochemistry Identification of enzyme genes in the liver of the Bleeker’s squid Loligo bleekeri by expressed sequence tag analysis Hidehiro Kondo • Takami Morita • Maki Ikeda • Chihiro Kurosaka • Aiko Shitara • Yuka Honda • Reiko Nozaki • Takashi Aoki • Ikuo Hirono Received: 30 July 2009 / Accepted: October 2009 / Published online: 25 November 2009 Ó The Japanese Society of Fisheries Science 2009 Abstract Expressed sequence tag (EST) analyses were performed with the aim of identifying enzyme genes in the liver of the Bleeker’s squid Loligo bleekeri Of the 768 ESTs identified and sequenced, 669 were grouped into 324 clusters Of these clusters, 123 comprising 245 ESTs were found to be homologous to genes reported to date Among these, 43 clusters were annotated as enzymes according to the Enzyme Commission (EC) numbering system Two EC groups, oxidoreductases and hydrolases, possessed a large number of ESTs A cluster homologous to the glutathione peroxidase, an enzyme in the oxidoreductase group, contained 16 ESTs, which accounted for 2.4% of the total ESTs sequenced There are three serine proteases, three cathepsins, two triacylgricerol lipases, and two chitinases among the clusters homologous to the enzymes in the hydrolase group Since the squid liver functions in the digestive process, these enzymes would be involved in food digestion Our data provide information on the various types of enzymes expressed in the squid liver and may provide a useful basis for further characterization of these enzymes Keywords Bleeker’s squid Á EC number Á Enzyme Á Expressed sequence tag H Kondo (&) Á M Ikeda Á C Kurosaka Á A Shitara Á Y Honda Á R Nozaki Á T Aoki Á I Hirono Laboratory of Genome Science, Tokyo University of Marine Science and Technology, Konan 4-5-7, Minato, Tokyo 108-8477, Japan e-mail: h-kondo@kaiyodai.ac.jp T Morita National Research Institute of Fisheries Science, Fukuura 2-12-4, Kanazawa-ku, Yokohama, Kanagawa 236-8648, Japan Introduction Squids are active carnivores and feed upon fishes and other squids [1] They are also important prey organisms for marine mammals and seabirds [2, 3] They consequently occupy a high trophic level and play an important role in the marine food web Most species of squids have a short life cycle of less than year, and their rapid growth rate makes them a candidate as a useful food source in the future In Japan, both the muscles and the livers of many squid species are commonly consumed in various cooked dishes and processed food products The livers, alternatively designated as the hepatopancreas, are quite large and function in digestive process by producing a variety of enzymes [4] Squid liver has been known to concentrate essential and non-essential elements and has, therefore, been used as a bioindicator to monitor for the presence of pollutants in the ocean [5, 6] In studies aimed at further developing the applications of the squid liver, a number of enzymes have been purified from the squid liver [7–13] Expressed sequence tag (EST) analysis is a powerful tool for identifying genes expressed in animals EST analyses have been applied in various aquatic animals, such as bivalves, fish, oyster, sea urchin, and shrimp [14–18] These analyses were performed not only to discover novel genes but also to identify genetic markers for breeding The ESTs from three species of squid are currently available in databases, and the EST analysis of the light organ tissues in the Hawaiian sepiolid squid Euprymna scolopes has been reported [19] We report here our sequencing of 669 ESTs isolated from the liver of the Bleeker’s squid Loligo bleekeri The Bleeker’s squid is distributed throughout the waters surrounding Japan, excluding the northern half of 123 162 Hokkaido This species is the most utilized loliginid squid in Japan [20] The ESTs were grouped into clusters and homology search analyses were performed Further characterization demonstrated that the 43 clusters were homologous to enzymes annotated with Enzyme Commission (EC) numbers Materials and methods Living Bleeker’s squid individuals (length approx 30 cm) were purchased from a local fish supplier In the laboratory, the squids were killed and the liver removed and cut into smaller pieces These pieces were then stored in RNAlater (Ambion, Austin, TX) at -20°C until use Total RNA was extracted using TRIZOL reagent (Gibco BRL Life Technologies, Rockville, MD) and subsequently used for mRNA purification using the Poly(A) Quick mRNA Isolation kit (Stratagene Cloning Systems, La Jolla, CA) The cDNA library was constructed using the Superscript Plasmid System with Gateway Technology for cDNA Synthesis and Cloning kit (Invitrogen Life Technologies, Rockville, MD) according to the manufacturer’s protocol The cloned cDNA fragments were amplified by PCR using ExTaq DNA polymerase (Takara, Otsu, Japan) with the vector-specific primer and cleaned up using ExoSap IT (GE Healthcare, Piscataway, NJ) The 50 -end single-pass sequencing of the PCR products was performed with the vector-specific primer using the BigDye Terminator Cycle Sequencing kit with the ABI3100 sequencer (Applied Biosystems, Foster City, CA) Sequencing data were trimmed by trace2dbEST (http:// www.nematodes.org/bioinformatics/trace2dbEST/ ‘‘Accessed 30 Oct 2009’’.) Sequences[100 bp were selected and then clustered using the Partigene program [21] A homology search of the clusters was performed using the BLAST program [22] against the UniProt database with an E-value cut-off of \10-8 to omit ambiguous annotations The clusters were further annotated using the Annot8r program [23] Amino acid sequences of serine proteases were aligned with the ClustalW program [24] Results and discussion A total of 768 ESTs were isolated and subsequently sequenced from the squid liver cDNA library After trimming, 669 ESTs (accession nos FS372549–FS373217) were grouped into 324 clusters, including 88 clusters containing multiple ESTs Based on a homology search against the UniProt database, we identified 123 clusters comprised of 245 ESTs that showed homology to the gene products reported to date, whereas 424 ESTs were not 123 Fish Sci (2010) 76:161–165 annotated by our homology search against the database The genes showing homology were categorized based on the GO slim term using the Annot8er program A large population of the genes were found to possess the biological process of the category ‘‘metabolic process’’ (Fig 1) Genes with a cellular component in the intracellular region and those with the molecular function of ‘‘catalytic activity’’ were also found to be abundant Based on the annotation according to the EC number—oxidoreductases (EC1), transferases (EC2), hydrolases (EC3), lyases (EC4), and isomerases (EC5)—43 clusters containing 121 ESTs showed homology to the enzymes (Table 1) Of these, 13 and 22 clusters containing 46 and 66 ESTs, respectively, belonged to oxidoreductase and hydrolase categories, respectively These ESTs account for about 6.9 and 9.9% of the total liver ESTs, respectively Therefore, the biochemical reactions catalyzed by these enzymes would actively take place in the liver The cluster homologous to glutation peroxidase (EC no 1.11.1.9) constains the largest number of ESTs in EC group 1, indicating that the gene is predominantly expressed in the liver (Table 1) Glutathione peroxidase is an enzyme involved in the protection of the organism from oxidative damage by reducing lipid hydroperoxides and hydrogen peroxide levels [25] There were two clusters homologous to ferroxidase in EC group 1, with seven and two ESTs, respectively The products expressed from these ESTs could be ferritin, which also protects the organism from oxidative damage by binding to free iron that produces free radicals from oxygen species [26] Squid liver is known to concentrate various elements, which may cause the production of free radicals Moreover, the enzymatic digestion of lipids and other energy sources, which also may produce reactive oxygen species, is likely to take place in the squid liver Consequently, the abundance of these transcripts suggests that the squid liver needs to prevent oxidative stress Three clusters homologous to serine proteases in the EC group (EC no 3.4.21) were abundant among the ESTs identified (Table 1) The full sequences of the serine proteases (LBC00043, LBC00130, and LBC00134) were determined and deposited into the databases as serine protease 1, 2, and with accession numbers AB512499, AB512500, and AB512501, respectively The serine protease catalytic triad of histidine, aspartic acid, and serine was conserved in the squid Serine proteases have important catalytic activities Interestingly, although these squid serine proteases showed homology to tryptase and limulus clotting enzyme, their amino acid identities to the serine proteases in the database were as low as approximately 30% On the other hand, the amino acid identities among the squid serine proteases showed a homology ranging Fish Sci (2010) 76:161–165 Fig Gene classification based on gene ontology The expressed sequence clusters (EST) clusters showing homologies were grouped into different categories according to the GO slim terms using the Annot8or program a Biological process, b cellular component, c molecular function 163 a biological process 1.0 % Response to stimulus 1.0 % Amino acid and derivative metabolic process 5.9 % Cellular process 5.9 % Nucleobase, nucleoside, nucleotide and nucleic acid metabolic process 19.8 % Transport 66.3 % Metabolic process b cellular component 3.7 % Extracellular region 75.3 % Intracellular 21.0 % Cell c molecular function 5.2 % Transporter activity 0.7 % Enzyme regulator activity 6.7 % Nucleic acid binding 42.2 % Catalytic activity 18.5 % Structural molecule activity 26.7 % Binding from 44 to 52%, suggesting a paralogous relationship among them Three clusters were homologous to cathepsin B, L, and D (EC no 3.4.22.1, 3.4.22.15, and 3.4.23.5, respectively) in EC group Among these cathepsins, the cathepsin D of Japanese common squid Todarodes pacificus has been already purified from the liver and characterized [11] Cathepsin belongs to the aspartic proteinase group The aspartic proteinase in Japanese common squid accounts for as much as 10% of the total enzymatic content [11] Since the transcripts of these proteases were abundantly observed, they should be of physiological significance in the squid liver, probably playing a role in enzymatic digestion of protein from food The clusters homologous to triacylgricerol lipases (EC no 3.1.1.3) were found to have many ESTs in EC group (Table 1) These enzymes would be also involved in food digestion Intriguingly, they only have low similarities to those homologs in the databases, and their 50 -end sequences are likely to be incomplete (data not shown) Furthermore, partial amino acid sequences of hepatic lipase in Japanese common squid [13] show no homology to each sequence of the clusters homologous to triacylgricerol lipases Hence, further investigation is required to determine whether these transcripts from ESTs in these clusters are lipases There were two clusters homologous to chitinase (EC no 3.2.1.14) in EC group (Table 1) Chitinase is known 123 164 Fish Sci (2010) 76:161–165 Table Clusters of ESTs homologous to enzymes with EC numbers Cluster No of EST EC No Product name Accession No.a E-valuea LBC00152 1.1.1.211 Long-chain-3-hydroxyacyl-CoAdehydrogenase Q8BMS1 1E-43 LBC00071 1.2.1.12 Glyceraldehyde-3-phosphatede hydrogenase (phosphorylating) Q94469 3E-18 LBC00080 1.2.1.12 Glyceraldehyde-3-phosphatede hydrogenase (phosphorylating) Q05025 1E-72 LBC00102 1.6.5.3 NADH dehydrogenase (ubiquinone) O47478 2E-34 LBC00370 1.6.5.3 NADH dehydrogenase (ubiquinone) P33511 1E-10 LBC00390 1.6.5.5 NADPH:quinonereductase P42865 2E-38 LBC00061 1.9.3.1 Cytochrome-c oxidase P29865 7E-34 LBC00092 1.9.3.1 Cytochrome-c oxidase O47475 1E-113 LBC00239 1.9.3.1 Cytochrome-c oxidase Q34941 8E-86 LBC00058 16 1.11.1.9 Glutathione peroxidase Q00277 2E-42 LBC00034 LBC00044 1.14.99.36 1.16.3.1 Beta-carotene15,150 -monooxygenase Ferroxidase Q9JJS6 P42577 6E-15 2E-55 LBC00208 1.16.3.1 Ferroxidase P42577 2E-62 LBC00148 2.1.1.45 Thymidylate synthase P07607 2E-60 LBC00101 2.7.1.20 Adenosine kinase P55264 1E-72 LBC00288 2.8.2.3 Aminesulfo transferase O46640 6E-16 LBC00017 3.1.1.3 Triacylglycerol lipase P54315 9E-12 LBC00162 3.1.1.3 Triacylglycerol lipase P54318 7E-18 LBC00136 3.1.1.13 Sterolesterase Q64194 9E-24 LBC00268 3.1.11.2 Exodeoxyribonuclease III Q9R1A9 4E-20 LBC00036 3.2.1.14 Chitinase Q6RY07 4E-17 LBC00344 3.2.1.14 Chitinase Q13231 2E-30 LBC00039 3.4.13.20 Beta-Ala-Hisdipeptidase Q66HG3 2E-18 LBC00330 3.4.15.1 Peptidyl-dipeptidase A Q10751 4E-77 LBC00043 13 3.4.21.59 Tryptase Q9N2D1 1E-27 LBC00130 3.4.21.86 Limulus clotting enzyme P21902 2E-27 LBC00134 LBC00197 3.4.21.86 3.4.22.1 Limulus clotting enzyme Cathepsin B P21902 P07688 5E-25 6E-43 LBC00272 3.4.22.15 Cathepsin L Q26636 2E-33 LBC00145 3.4.23.5 Cathepsin D P18242 2E-69 LBC00111 3.4.25.1 Proteasome endopeptidase complex Q9PTW9 1E-106 LBC00334 3.5.1.28 N-acetylmuramoyl-L-alanineamidase Q8INK6 3E-33 LBC00352 3.5.1.28 N-acetylmuramoyl-L-alanineamidase Q8INK6 3E-33 LBC00236 3.5.99.6 Glucosamine-6-phosphatedeaminase Q9CRC9 2E-13 LBC00271 3.6.1.45 UDP-sugardiphosphatase Q05B60 3E-22 4E-18 LBC00065 3.6.3.14 H(?)-transporting two-sector ATPase Q03105 LBC00077 3.6.3.14 H(?)-transporting two-sector ATPase Q9ZEC1 1E-15 LBC00363 3.6.3.14 H(?)-transporting two-sector ATPase Q2N8Z5 1E-101 LBC00243 4.4.1.1 Cystathionine gamma-lyase Q58DW2 8E-40 LBC00252 5.2.1.8 Peptidylprolyl isomerase P0C1H7 1E-69 LBC00305 5.2.1.8 Peptidylprolyl isomerase P18253 8E-63 LBC00380 LBC00307 1 5.3.3.8 5.3.4.1 Dodecenoyl-CoA isomerase Protein disulfide-isomerase Q9WUR2 P38657 4E-13 4E-24 Forty-three clusters are listed in the order of the EC numbers of the homologs EST Expressed sequence tag, EC Enzyme Commission a Accession numbers of homologs and E-values of the homology search results are listed 123 Fish Sci (2010) 76:161–165 to have a wide range of physiological roles, such as digestion, defense, morphogenesis, and aggression [27] The products expressed from ESTs in two clusters homologous to chitinase were considered to be involved in food digestion Chitinase has also been purified from Japanese common squid liver [7] The chitinase has isozymes of different molecular mass, 38 and 42 kDa, but their amino acid sequences have not yet been determined [9] In conclusion, the enzyme homologs that are mainly involved in protection from oxidative stress and food digestion process have been identified by EST analysis A number of enzymes had been purified in Japanese common squid liver, but their structures have not yet been determined Our ESTs will provide the basis for further determination of the nucleotide sequences of such enzymes 165 13 14 15 16 17 References Packard A (1972) Cephalopods and fish: the limits of convergence Biol Rev 47:241–307 Ogi H (1980) The pelagic feeding ecology of Thick-Billed Murres Uria lomvia in the North Pacific (March to June) Bull Fac Fish Sci Hokkaido Univ 31:50–72 Marcovecchio JE, Gerpe MS, Bastida RO, Rodrı´guez DH, Moron SG (1994) Environmental contamination and marine mammals in coastal waters from Argentina: an overview Sci Total Environ 154:141–151 Nasu K, Okutani K, Ogura M (2002) Squid (in Japanese) Seizandoshoten, Tokyo Miramand P, Bentley D (1992) Concentration and distribution of heavy metals in tissues of two cephalopods, Eledone cirrhosa and Sepia officinalis, from the French coast of the English Channel Mar Biol 114:407–414 Umezu T (1994) Oceanic distribution of 108mAg estimated using squid liver—a result of squid watch Water Rep 4:9–11 Matsumiya M, Mochizuki A (1997) Purification and characterization of chitinase from liver of Japanese common squid Todarodes pacificus Fish Sci 63:409–413 Tamori J, Kanzawa N, Tajima T, Tamiya T, Tsuchiya T (1999) Purification and characterization of a novel isoform of myosinase from spear squid liver J Biochem 126:969–974 Matsumiya M, Miyauchi K, Mochizuki A (2002) Characterization of 38 kDa and 42 kDa chitinase isozymes from the liver of Japanese common squid Todarodes pacificus Fish Sci 68:603– 609 10 Yokozawa Y, Tamai H, Tatewaki S, Tajima T, Tsuchiya T, Kanzawa N (2002) Cloning and biochemical characterization of astacin-like squid metalloprotease J Biochem 132:751–758 11 Komai T, Kawabata C, Amano M, Lee BR, Ichishima E (2004) Todarepsin, a new cathepsin D from hepatopancreas of Japanese common squid (Todarodes pacificus) Comp Biochem Physiol 137B:373–382 12 Komai T, Kawabata C, Tojo H, Gocho S, Ichishima E (2007) Purification of serine carboxypeptidase from the hepatopancreas 18 19 20 21 22 23 24 25 26 27 of Japanese common squid Todarodes pacificus and its application for elimination of bitterness from bitter peptides Fish Sci 73:404–411 Park J, Cho S-Y, Choi S-J (2008) Purification and characterization of hepatic lipase from Todarodes pacificus BMB Rep 41:254–258 Tanguy A, Bierne N, Saavedra C, Pina B, Bache`re E, Kube M, Bazin E, Bonhomme F, Boudry P, Boulo V, Boutet I, Cancela L, Dossat C, Favrel P, Huvet A, Jarque S, Jollivet D, Klages S, Lape`gue S, Leite R, Moal J, Moraga D, Reinhardt R, Samain J-F, Zouros E, Canario A (2008) Increasing genomic information in bivalves through new EST collections in four species: development of new genetic markers for environmental studies and genome evolution Gene 408:27–36 Cossins AR, Crawford DL (2005) Fish as models for environmental genomics Nat Rev Genet 6:324–333 Quilang J, Wang S, Li P, Abernathy J, Peatman E, Wang Y, Wang L, Shi Y, Wallace R, Guo X, Liu Z (2007) Generation and analysis of ESTs from the eastern oyster, Crassostrea virginica Gmelin and identification of microsatellite and SNP markers BMC Genomics 8:157 Tassanakajon A, Klinbunga S, Paunglarp N, Rimphanitchayakit V, Udomkit A, Jitrapakdee S, Sritunyalucksana K, Phongdara A, Pongsomboon S, Supungul P, Tang S, Kuphanumart K, Pichyangkura R, Lursinsap C (2006) Penaeus monodon gene discovery project: the generation of an EST collection and establishment of a database Gene 384:104–112 Poustka AJ, Groth D, Hennig S, Thamm S, Cameron A, Beck A, Reinhardt R, Herwig R, Panopoulou G, Lehrach H (2003) Generation, annotation, evolutionary analysis, and database integration of 20, 000 unique sea urchin EST clusters Genome Res 13:2736–2746 Chun CK, Scheetz TE, Bonaldo M, Brown B, Clemens A, Crookes-Goodson WJ, Crouch K, Demartini T, Eyestone M, Goodson MS, Janssens B, Kimbell JL, Koropatnick TA, Kucaba T, Smith C, Stewart JJ, Tong D, Troll JV, Webster S, WinhallRice J, Yap C, Casavant TL, Mcfall-Ngai MJ, Soares MB (2006) An annotated cDNA library of juvenile Euprymna scolopes with and without colonization by the symbiont Vibrio fischeri BMC Genomics 7:154 Roper CFE, Sweeney MJ, Nauen CE (1984) FAO species catalogue, vol Cephalopods of the world An annotated and illustrated catalogue of species of interest to fisheries FAO Fish Synop 125:1–277 Parkinson J, Anthony A, Wasmuth J, Schmid R, Hedley A, Blaxter M (2004) PartiGene—constructing partial genomes Bioinformatics 20:1398–1404 McGinnis S, Madden TL (2004) BLAST: at the core of a powerful and diverse set of sequence analysis tools Nucleic Acids Res 32:W20–W25 Schmid R, Blaxter ML (2008) Annot8r: GO, EC and KEGG annotation of EST datasets BMC Bioinformatics 9:180 Thompson JD, Higgins DG, Gibson TJ (1994) ClustalW Nucleic Acids Res 22:4673–4680 Gilca M, Stoian I, Atanasiu V, Virgolici B (2007) The oxidative hypothesis of senescence J Postgrad Med 53:207–213 Arosio P, Levi S (2002) Ferritin, iron homeostasis, and oxidative damage Free Radic Biol Med 33:457–463 Gooday GW (1997) The many uses of chitinases in nature Chitin Chitosan Res 3:233–243 123 Fish Sci (2010) 76:167–175 DOI 10.1007/s12562-009-0193-y ORIGINAL ARTICLE Chemistry and Biochemistry Transcriptional activities of medaka Oryzias latipes peroxisome proliferator-activated receptors and their gene expression profiles at different temperatures Hidehiro Kondo • Ryohei Misaki • Shugo Watabe Received: 11 August 2009 / Accepted: 29 October 2009 / Published online: 22 December 2009 Ó The Japanese Society of Fisheries Science 2009 Abstract Medaka Oryzias latipes peroxisome proliferator-activated receptors (PPARs) a1, a2, b, and c were characterized for their sequence structures and transcriptional activities While the genes encoding PPARa2, b, and c were determined for their full length, the PPARa1 gene remained undetermined for its 50 -end region, which encodes the DNA-binding domain and a part of ligandbinding domain Transcriptional activities of medaka PPARa2, b, and c were also examined by reporter assay using reporter plasmids containing three copies of acylcoenzyme A (CoA) oxidase PPAR response element The activity of PPARa2 was 1.5- to 2.0-fold upregulated by 5,8,11,14-eicosatetraynoic, arachidonic, eicosapentaenoic, and oleic acids The transcripts of medaka PPARs were expressed in brain, gut, heart, kidney, and liver The messenger RNA (mRNA) levels of PPARs in brain from medaka acclimated to 10°C were higher than those from fish acclimated to 30°C The mRNA levels of PPAR target genes encoding carnitine palmitoyltransferase-I, carnitine acylcarnitine translocase, and medium-chain acyl-CoA The nucleotide sequences reported in this paper have been registered into the DDBJ/EMBL/GenBank databases with accession numbers AB469411–AB469414 H Kondo Á R Misaki Á S Watabe (&) Department of Aquatic Bioscience, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo, Tokyo 113-8657, Japan e-mail: awatabe@mail.ecc.u-tokyo.ac.jp Present Address: H Kondo Laboratory of Genome Science, Graduate School of Marine Science and Technology, Tokyo University of Marine Science and Technology, Minato, Tokyo 108-8477, Japan dehydrogenase, which are involved in b-oxidation of fatty acids in other vertebrates, were also higher in various tissues from fish acclimated to 10°C than at 30°C Keywords b-Oxidation Á Ligand assay Á Medaka Á PPAR Á Temperature acclimation Introduction Temperature is one of the factors that has an enormous effect on the physiology and biochemistry of poikilotherms, especially eurythermal temperate fish inhabiting a wide range of temperature Such fish change expressions of various genes to maintain their activities at different temperatures For example, mitochondrial adenosine triphosphate (ATP) synthase b-subunit is expressed much higher in carp Cyprinus carpio acclimated to 10°C than fish acclimated to 30°C [1], and concomitantly its specific activity is increased in fish acclimated to 10°C [2] Medaka Oryzias latipes cultured cell lines such as OLHNI-e1, derived from the embryo of HNI strain medaka, showed different expression levels of various genes according to different temperatures, as revealed by reverse-transcription polymerase chain reaction (RT–PCR) [3] and complementary DNA (cDNA) microarray [4] These results suggest that energy metabolism is also maintained at a constant level in different temperatures by changing the expression levels of relevant genes in eurythermal fish species Fish prefer lipid to sugar as an energy source, and thus consideration of lipid metabolism is more important [5] b-Oxidation, an important step of fatty-acid catabolism that reproduces ATP via oxidative phosphorylation of adenosine diphosphate (ADP), is regulated by the rate-limiting 123 168 enzyme carnitine palmitoyltransferase-I (CPT-I) in mitochondria [6] It has been reported that CPT-I activity in striped bass Morone saxatilis inhabiting 5°C waters is higher than that in fish reared at 25°C [7] The expressions of such enzymes involved in b-oxidation are regulated by peroxisome proliferator-activated receptors (PPARs), members of the nuclear hormone receptor superfamily [8] PPARs consist of three isotypes, PPARa, b, and c, having two well-conserved regions of DNA- and ligand-binding domains PPARa and PPARb activate lipid catabolism by regulating transcription of target genes encoding enzymes that are involved in peroxisomal and mitochondrial b-oxidation [9, 10] On the other hand, PPARc is involved in lipid accumulation and regulation of adipocyte differentiation [11] All PPAR isotypes from mammals and amphibians are activated by various ligands, including polyunsaturated fatty acids (PUFAs) and eicosanoids [12] On the other hand, it has been claimed that hypolipidemic agents, including Wy-14643, specifically activate transcriptional activity of PPARa in mammals, whereas activation by 5,8,11,14eicosatetraynoic acid (ETYA), an arachidonic acid homolog, is specific to PPARb in amphibians [13] Meanwhile, thiazolidinediones such as rosiglitazone activate transcriptional activity of PPARc in mammals and amphibians [13] The full-length nucleotide sequences encoding PPARa, b, and c have been determined in zebrafish Danio rerio [14], torafugu Takifugu rubripes [15, 16], sea bass Dicentrarchus labrax [17], plaice Pleuronectes platessa, and sea bream Sparus aurata [18], along with partial nucleotide sequences in brown trout Salmo trutta [19] Furthermore, full-length nucleotide sequences have been reported for PPARa in thicklip grey mullet Chelon labrosus [20] and PPARb in goldfish Carassius auratus [21], along with partial nucleotide sequences of PPARc in Atlantic salmon Salmo salar [22, 23] and PPARa and b in rainbow trout Onchorhyncus mykiss [24] Interestingly, torafugu contains types of PPARa [15, 16], and zebrafish contains types of PPARb [14] Although, the functional significance of these PPAR isoforms of either PPARa or b from fish remains unknown, their existence is specific to fish, and these isoforms have not been found in genomes of tetrapods In the present study, four PPAR genes, encoding PPARa1, a2, b, and c, were cloned and examined for their transcript tissue distribution in medaka acclimated to 10 and 30°C In addition, accumulated mRNA levels of the PPAR target genes encoding CPT-I, carnitine acylcarnitine translocase (CAT), and medium-chain acylCoA dehydrogenase (MCAD) involved in b-oxidation were also investigated in fish acclimated to 10 and 30°C 123 Fish Sci (2010) 76:167–175 Materials and methods Fish HNI strain medaka [25] for cDNA cloning were acclimated at 25°C for weeks by feeding ad libitum Tissues were isolated from fish after dissection under anesthesia with 200 ppm 2-phenoxyethanol and were stored at -80°C after freezing in liquid nitrogen Fish acclimated to 10 and 30°C for more than weeks [26] were dissected for removal of brain, gut, heart, kidney, and liver and these tissues were subjected to real-time polymerase chain reaction (PCR) assay Three different series of experiments of real-time PCR assay were performed, each using the tissues collected from individuals cDNA cloning Nucleotide sequences encoding medaka PPARs were obtained by performing homology search using the basic local alignment search tool (BLAST) program on a medaka genome database (NIG DNA Sequencing Center: http://dolphin.lab.nig.ac.jp/medaka/, accessed November 2009) using the nucleotide sequences of the genes encoding teleost PPARs reported to date The exon–intron organization of the obtained genes was predicted using the GENSCAN program (GENESCAN at GeniusNet: http://genome dkfz-heidelberg.de/cgi-bin/GENSCAN/genscan.cgi, accessed November 2009) Primer sets for amplification of fulllength cDNAs were designed referring to the sequences in the 50 - and 30 -untranslated regions, obtained as described above (Table 1) Since the 50 -untranslated region of the PPARa1 gene was not available in the database, a forward primer was designed to amplify a partial cDNA fragment based on the available 50 -site sequence which showed homology with those from other vertebrates Total RNA was isolated using ISOGEN (Nippon Gene, Toyama, Japan) and treated with deoxyribonuclease I (Takara, Otsu, Japan) to eliminate contaminating DNA Total RNA was then subjected to cDNA synthesis by using Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol PCR was performed using ExTaq DNA polymerase (Takara) according to the manufacturer’s protocol PCR consisted of an initial step at 94°C for followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 60°C for 30 s, and polymerization at 72°C for min, with a final extension step at 72°C for Amplified cDNA fragments were subcloned into pGEM-T easy vector (Promega, Madison, WI, USA) The deduced amino acid sequences were aligned with those of the PPAR genes from torafugu (accession nos AB275885 to AB275888) by using the ClustalW program [27] Fish Sci (2010) 76:167–175 Table Primers used for amplifying cDNA fragments encoding a part of PPARa1 and the full length of PPARa2, b, and c 169 Gene Primer Sequence PPARa1 mPPARa-F1 50 -AGAGCACGTCCGTGGAGAC-30 mPPARa1-3UTR-R 50 -GAACCTCAAGCTCTTCCTGC-30 mPPARa2-F-BamHI 50 -ATGGATCCATGGCGGCGGAACTG-30 mPPARa2-R-HindIII 50 -ATAAGCTTTCAGTACATGTCTTTGT-30 mPPARb-F-BamHI 50 -ATGGATCCATGGACGGGTTTCAGC-30 mPPARb-R-EcoRI 50 -ATGAATTCTTAATACATGTCTTTGTAG-30 mPPARg-F-BamHI 50 -ATGGATCCATGGTGGACACCCAGCAG-30 mPPARg-R-EcoRI 50 -TAGAATTCTCAGTACAAGTCCTTCATG-30 PPARa2 PPARb The sequences of restriction sites for BamHI, HindIII, and EcoRI are underlined PPARc Transcriptional activity assay Full-length cDNAs encoding PPARs except PPARa1 were digested with restriction enzymes, which recognize the sequences inserted into the primers, and ligated into pBKCMV vector (Promega) digested with the same restriction enzymes The expression plasmids containing African clawed frog Xenopus laevis PPARs, pSG5-xPPARa, pSG5-xPPARb, and pSG5-xPPAR [13], were kindly provided by Dr Wahli of University of Lausanne The reporter plasmid PPRE3TK-Luc [28], which expresses firefly luciferase under regulation by three copies of acyl-CoA oxidase PPAR response element (PPRE) connected to the upstream of herpes virus thymidine kinase promoter, was a gift from Dr Evans of the Salk Institute The mb-pRL vector, which expresses Renilla reniformis luciferase under regulation by the promoter of the medaka b-actin gene [26], was used as an internal control Ligands examined were Wy-14643, ETYA, and PUFAs, including arachidonic, docosahexaenoic, eicosapentaenoic, and oleic acids purchased from Cayman Chemical (Ann Arbor, MI, USA) and rosiglitazone obtained from Alexis Biochemical (San Diego, CA, USA) OLHNI-e1 cell line [3] was cultured at 33°C in Leibovitz L15 medium supplemented with 20% fetal bovine serum (FBS), streptomycin (100 lg/ml), and penicillin (100 U/ml) Cells were harvested and plated into 96-well plates at 104 cells/well in 200 ll medium and cultured until cell density exceeded 90% These cells were washed with phosphate-buffered saline (PBS) and added with 50 ll L15 medium containing 1% FBS and 50 ll Opti-MEM medium (Invitrogen) containing 160 ng PPRE3-TK-Luc, ng mb-pRL, 40 ng medaka PPAR expression vector, and 0.5 ll lipofectamine 2000 (Invitrogen) After h, 100 ll medium containing ligands was added to the culture cells, which were subsequently incubated for another 18 h All ligands were dissolved in PBS containing 1% bovine serum albumin (BSA) and added into the medium at volume ratio of 1:10 Subsequently, Wy-14643, ETYA, and rosiglitazone were added at 10 lM and PUFAs at 50 lM, respectively The cells were harvested and lysed in 20 ll passive lysis buffer (Promega), and the luciferase activities of the lysates were determined by Dual-luciferase kit (Promega) using a TD20/20 luminometer (Promega) The luciferase activities were normalized to those of Renilla reniformis luciferase, and the relative values were calculated in comparison with those without ligands Data were analyzed with one-way analysis of variance (ANOVA), and differences shown in ANOVA were analyzed with Dunnett’s method Real-time PCR Primer sets for real-time PCR were designed using Primer Express software (Applied Biosystems, Foster City, CA, USA) (Table 2) Primer sets for PPARs were designed based on sequences obtained in the present study, whereas those for b-actin were cited from the DDBJ/EMBL/ GenBank databases (accession no S74868) Expression sequenced tags homologous to the genes encoding CAT, CPT-I, and MCAD were obtained by homology search using the BLAST program on the medaka genome database based on the nucleotide sequences of the homologous genes from other vertebrates, and used for designing primer sets First-strand cDNAs were synthesized using various tissues of fish acclimated to 10 and 30°C and diluted 100-fold with distilled water cDNA solution (3 ll) was mixed with 10 ll 29 SYBR premix ExTaq DNA polymerase (Takara), followed by 0.4 ll 509 ROX reference dye and 0.5 ll of each primer at 10 lM After the volume of the mixture was adjusted to 20 ll with distilled water, real-time PCR analysis was performed using 7500 real-time PCR system (Applied Biosystems) according to the manufacturer’s protocol Data obtained were analyzed using detection system software (v2.0 release 4, Applied Biosystems), and mRNA levels were calculated relative to values for b-actin Differences between two temperatures were evaluated using Student’s t test Results cDNA cloning and primary structures of medaka PPARs Full-length nucleotide sequences of the genes encoding medaka PPARa2, b, and c were obtained by in silico 123 170 Table Primers used for quantitative real-time PCR for transcripts of medaka PPARs and their target genes and b-actin Fish Sci (2010) 76:167–175 Gene Primer Sequence PPARa1 mPPARa1-F230 50 -ATGGATGTGGAGGAGCAGAGA-30 mPPARa1-R302 50 -GTTGCTGCCTCTGAAGTGGTT-30 mPPARa2-F27 50 -TGCAAGCAGGAGATTTGAGTGT-30 mPPARa2-R138 50 -TTTGAGAACTGCAAGTGGTTATTGT-30 mPPARb-F284 50 -CCACAACGGTCACTTAACAGGAA-30 mPPARb-R359 50 -GCTTTGCTATGTGGATGAAACAAG-30 mPPARg-F113 50 -CATCAGACTACGCTTCCATTTCC-30 mPPARg-R190 50 -TGGACGACATCGAAGACACAA-30 mCAT-RT-F215 50 -CGGAACCCTGGACTGCTTTA-30 mCAT-RT-R286 50 -GCCGCCATTCCTTTGTAGAG-30 mCPT1-RT-F143 50 -ACCCCAGAGGGCATTGATCT-30 mCPT1-RT-R214 50 -GGAGCGAACACCAGAGAGGTA-30 mMCAD-RT-F326 50 -TGACAGCTGCCTCATCACAGA-30 mMCAD-RT-R397 50 -GAATTGGCCTCCATAGCAGTCT-30 m-b-actin-F1594 m-b-actin-R1691 50 -TGCCCAGTGGTTGAGCATACT-30 50 -AGGACCAAGATTTAAGGCTGAAAG-30 PPARa2 Sequences cited from the medaka EST database for carnitine palmitoyltransferase-I (CPT-I) (Accession no MF01SSA026D08), carnitine acylcarnitine translocase (CAT) (Accession no MF01FSA038C14), and medium-chain acyl-CoA dehydrogenase (MCAD) (Accession no MF01SSA169H05); medaka bactin gene was cited from the DDBJ/EMBL/GenBank databases (S74868) PPARb PPARc CAT CPT-I MCAD b-Actin Fig Amino acid sequences of medaka PPARa2 (a2), b (b), and c (g) in comparison with those of torafugu Regions of DNA-binding and ligand-binding domains for medaka (m) and torafugu (f) PPARs are represented by open and shaded boxes, respectively Gaps, represented by dashes, were introduced to maximize alignment Dots represent amino acids identical to those of medaka (m) PPARa2 (a2) Filled triangles indicate amino acid residues essential to binding of PPARs to ligands analysis However, the 50 -untranslated region of the PPARa1 gene was not found in the medaka genome database nor amplified by 50 -rapid amplification of cDNA ends (RACE) PCR (data not shown) The full-length amino acid sequences of PPARa2, b, and c are shown in Fig DNA-binding domains of medaka PPARa2, b, and c showed 90%, 98%, and 98% identities to those of corresponding PPARs of torafugu, respectively On the other hand, the ligand-binding domain of medaka PPARa2, b, and c showed 79%, 96%, and 77% identities to their torafugu counterparts, respectively Medaka PPARa2 and c had additional sequences in the N-terminal region of the 123 Fish Sci (2010) 76:167–175 171 Relative luciferase activity 1.0 2.0 no ligand ETYA Wy14643 rosiglitazon 18:1n-9 20:4n-6 20:5n-3 22:6n-3 Fig Partial amino acid sequence of medaka PPARa1 in comparison with fish PPARas C-terminal part of ligand-binding domain of medaka (m) PPARa1 (a1) is aligned with corresponding regions of torafugu (f) PPARa1 (a1), and medaka (m) and torafugu (f) PPARa2 (a2) Dots represent amino acids identical to those of medaka (m) PPARa1 (a1) Filled triangles indicate amino acid residues essential to binding of PPARs to ligands ligand-binding domain compared with those reported for torafugu homologs [16] Four amino acid residues to form hydrogen bonds with the ligands [29] were conserved in medaka PPARs, except for the histidine residue of PPARa2 at position 327, which was tyrosine in torafugu The partial sequence of medaka PPARa1 is shown in Fig along with the corresponding region of fish PPARas (Fig 2) The sequence contained only ligand-binding domain and showed 91%, 80%, and 84% amino acid identities to torafugu PPARa1, medaka PPARa2, and torafugu PPARa2, respectively Transcriptional activities of medaka PPARs Transcriptional activities were measured using medaka OLHNI-e1 cell line transfected with PPAR expression vectors (Fig 3) Cells transfected with the expression vector containing African clawed frog PPAR genes as reference showed activities (data not shown) Medaka PPARa2 activity was enhanced 1.5- to 2.0-fold by ETYA at 10 lM and by arachidonic, eicosapentaenoic, and oleic acids at 50 lM, but not by Wy14643 or rosiglitazone at 10 lM or by docosahexaenoic acid at 50 lM On the other hand, no activation were observed with PPARb and c no ligand ETYA Wy14643 rosiglitazon 18:1n-9 20:4n-6 20:5n-3 22:6n-3 no PPAR * PPARα2 * * * no ligand ETYA Wy14643 rosiglitazon 18:1n-9 20:4n-6 20:5n-3 22:6n-3 PPARβ no ligand ETYA Wy14643 rosiglitazon 18:1n-9 20:4n-6 20:5n-3 22:6n-3 PPARγ Fig Effects of various ligands on transcriptional activities of medaka PPARs Medaka culture cells transfected with plasmids containing PPRE-TK3, medaka PPAR expression vector, and mb-pRL were treated at 33°C for 18 h each with ETYA, Wy14643, and rosiglitazone at 10 lM and each with oleic (18:1n-9), arachidonic (20:4n-6), eicosapentaenoic (20:5n-3), and docosahexaenoic acids (22:6n-3) at 50 lM Relative luciferase activity is represented as the mean value from three different experiments, taking activity without ligands as Error bars indicate standard deviations Asterisks indicate significant difference (P \ 0.01) in comparison with activity without ligands those acclimated to 30°C Although mRNA levels of PPARa1, a2, and c tended to be higher in fish acclimated to 10°C than in those acclimated to 30°C in all tissues examined, the differences were not significant statistically except in the cases described above Tissue distribution of medaka PPARs transcripts Accumulated mRNA levels of PPAR target genes in medaka Transcripts of medaka PPARs were detected in brain, gut, heart, liver, and muscle from medaka acclimated to both 10 and 30°C (Fig 4) Levels in medaka acclimated to 10°C were significantly higher than those in fish acclimated to 30°C for all PPARs from brain Similarly, the level of PPARb in gut was higher in fish acclimated to 10°C than in mRNA levels of PPAR target genes encoding CAT, CPT-I, and MCAD are shown in Fig Transcripts were detected in brain, gut, heart, liver, and muscle from medaka acclimated to both 10 and 30°C Medaka acclimated to 10°C showed significantly higher transcriptional levels of CAT in gut and muscle, CPTI in brain and 123 172 10 °C mPPARα1 0.012 Relative mRNA level Fig Relative mRNA levels of PPARs in brain, gut, heart, liver, and muscle from medaka acclimated to 10 and 30°C Relative levels were calculated using those of b-actin as the control; mean values are presented Error bars indicate standard deviations Asterisks indicate significant difference (P \ 0.05) between 10 and 30°C Fish Sci (2010) 76:167–175 30 °C * 0.008 0.006 0.010 0.008 0.012 0.012 0.010 0.010 0.010 0.008 0.008 0.008 0.006 0.006 0.006 0.004 0.004 0.004 0.002 0.002 0.002 0.004 0.006 0.004 0.002 0.002 mPPARα2 Relative mRNA level 0.08 Gut Brain * Heart 0.015 Muscle Liver 0.08 0.04 0.06 0.03 0.04 0.02 0.02 0.01 0.03 0.06 0.010 0.02 0.04 0.005 0.01 0.02 Gut Brain mPPARβ Relative mRNA level 0.012 0.03 * 0.005 * Heart Muscle Liver 0.03 0.02 0.004 0.02 0.02 0.02 0.003 0.01 0.002 0.01 0.01 0.01 0.001 Brain Gut Heart Liver Muscle mPPARγ * 0.02 0.0075 0.0075 Relative mRNA level 0.0015 0.010 0.01 0.005 0.0025 0.0025 0.0005 Brain liver, and MCAD in brain, gut, and heart than those acclimated to 30°C However, the differences observed in the mRNA levels of all other tissues were not statistically significant 123 0.0050 0.0050 0.0010 Gut Heart Liver Muscle Discussion Two types of PPARa have been identified in torafugu [16] In the present study, we found four PPAR genes in the Fish Sci (2010) 76:167–175 10 °C 30 °C CAT * 0.02 Relative mRNA level Fig Relative mRNA levels of CAT, CPT-I, and MCAT in brain, gut, heart, liver, and muscle from medaka acclimated to 10 and 30°C Relative levels were calculated using those of b-actin as the control; mean values are presented Error bars indicate standard deviations Asterisks indicate significant difference (P \ 0.05) between 10 and 30°C a The mean values are from two different series of experiments, instead of from three different series of experiments for other comparisons (see details in the text) 173 0.02 0.10 0.02 0.02 0.08 * 0.06 0.01 0.01 0.01 0.01 0.04 0.02 Brain CPT I Relative mRNA level 0.02 Gut Heart Liver Muscle * 0.004 0.008 0.02 0.003 * 0.015 a 0.006 0.010 0.01 0.01 0.002 0.004 0.005 0.002 0.001 Brain MCAD Gut Liver Muscle * * 0.050 Relative mRNA level Heart 0.10 * 0.10 0.10 0.5 0.4 0.3 0.05 0.025 0.05 0.05 0.2 0.1 Brain medaka genome database, two of PPARa and one each of PPARb and PPARc, as in the case of torafugu PPARs The 50 -untranslated region of medaka PPARa1 was not contained in the database nor determined by 50 -RACE However, transcripts of PPARa1 were actually detected, although their mRNA levels were lower than those of PPARa2, as revealed by real-time PCR (Fig 4) In order to determine functional differences between the two types of medaka PPARa, further investigation is needed The transcriptional activity of medaka PPARa2 was enhanced by ETYA, and arachidonic, eicosapentaenoic, and oleic acids, but not by Wy14643 or docosahexaenoic acid It has been reported that Wy14643 can enhance Gut Heart Liver Muscle transcriptional activities of plaice and seabream PPARa as in the case of ETYA [18] However, the enhancement of transcriptional activity of torafugu PPARa1 and a2 by Wy14643 has been reported to be less than that by ETYA [16] The differences in the ligand effect between different species may be reflected by amino acid variations in the ligand-binding domain It was noted that one of four amino acid residues essential to ligand binding of PPARa2 at position 327 was histidine in medaka (Fig 1) in place of tyrosine in torafugu PPARa2 [16] The transcriptional activities of medaka PPARb and c were not enhanced by various ligands examined in the present study, as in the case of torafugu counterparts [16] It has been claimed that 123 174 PPARs in mammals and amphibians form heterodimers with retinoid X receptor, which is activated by retinoic acid [8] Furthermore, ligand-dependent transcriptional activities of rainbow trout PPAR are enhanced in the presence of 9-cis-retinoic acid [24] However, further investigation is required to confirm whether medaka PPARb and c also require another type of receptor to exert their activities mRNA levels of PPARs were generally higher in medaka acclimated to 10°C than in those acclimated to 30°C in all tissues examined It has been reported that mRNA levels of PPARa in zebrafish reared at 18°C were lower than those in fish reared at 28°C, whereas no differences could be noticed in PPARb [30] Similar tendencies of PPARa and b gene expression were observed in goldfish [31] Furthermore, MCAD gene expression in white muscle and liver of goldfish were lower at 4°C than at 20 and 35°C, but not changed in red muscle and heart On the other hand, mRNA levels of MCAD in medaka tissues were higher at 10°C than at 30°C It is likely that upregulation of PPARs and their target gene expression is important for medaka to maintain lipid metabolism even at low temperatures Among a variety of PPARs target genes in vertebrates, those involved in lipid metabolism have been well studied In the present study, mRNA levels of CAT, CPT-I, and MCAD, which are known to be involved in b-oxidation, tended to be higher in medaka acclimated to 10°C than in those acclimated to 30°C, as in the case of PPARs It has been reported that expression of these genes is regulated by PPARs, mainly PPARa in vertebrates rather than in fish [10] Therefore, it is urgent to confirm and validate whether gene expression of CAT, CPT-I, and MCAD is regulated by PPARs in fish as well In conclusion, although medaka PPARa1, a2, b, and c were cloned, transcriptional activity was enhanced only for PPARa2 by ETYA and arachidonic, eicosapentaenoic, and oleic acids mRNA levels of medaka PPARs were higher in fish acclimated to 10°C than in those acclimated to 30°C, irrespective of isotype, and similar temperature dependency was observed for CAT, CPT-I, and MCAD, potent PPAR target genes Since PPARs are key regulators in lipid metabolism in vertebrates, different transcriptional levels of medaka PPARs between fish acclimated to 10°C and those acclimated to 30°C may reflect temperature compensation in the lipid metabolism of this fish Acknowledgments We thank Dr Wahli of Lausanne University for kindly providing the expression vectors for African clawed frog PPARa, b, and c and advice for our experiments We also thank Dr Evans of the Salk Institute for allowing us to use the reporter plasmid PPRE3-TK-Luc This work was partly funded by a Grantin-Aid from the Japan Society for the Promotion of Science 123 Fish Sci (2010) 76:167–175 References Kikuchi K, Itoi S, Watabe S (1999) Increased levels of mitochondrial ATP synthase b-subunit in fast skeletal muscle of carp acclimated to cold temperature Fish Sci 65:629–636 Itoi S, Kinoshita S, Kikuchi K, Watabe S (2002) Changes of carp FoF1-ATPase in association with temperature acclimation Am J Physiol 284:R153–R163 Hirayama M, Mitani H, Watabe S (2006) Temperature-dependent growth rates and gene expression patterns of various medaka Oryzias latipes cell lines derived from different populations J Comp Physiol B 176:311–320 Hirayama M, Ahsan M, Mitani H, Watabe S (2008) CYR61 is a novel gene associated with temperature-dependent changes in fish metabolism as revealed by cDNA microarray analysis on a medaka Oryzias latipes cell line J Cell Biochem 104:1297–1310 Watanabe T (1982) Lipid nutrition in fish Comp Biochem Physiol 73B:3–15 Ronnett GV, Kim EK, Landree LE, Tu Y (2005) Fatty acid metabolism as a target for obesity treatment Physiol Behav 85:25–35 Rodnick KJ, Sidell BD (1994) Cold acclimation increases carnitine palmitoyltransferase I activity in oxidative muscle of striped bass Am J Physiol 266:R405–R412 Desvergne B, Wahli W (1999) Peroxisome proliferator-activated receptors: nuclear control of metabolism Endocr Rev 20:649– 688 Wang YX, Lee CH, Tiep S, Yu RT, Ham J, Kang H, Evans RM (2003) Peroxisome proliferator-activated reseptor d activates fat metabolism to prevent obesity Cell 113:159–170 10 Mandard S, Mu¨ller M, Kersten S (2004) Peroxisome proliferatoractivated receptor a target genes Cell Mol Life Sci 61:393–416 11 Nedergaard J, Petrovic N, Lindgren EM, Jacobsson A, Cannon B (2005) PPARgamma in the control of brown adipocyte differentiation Biochim Biophys Acta 1740:293–304 12 Hihi AK, Michalik L, Wahli W (2002) PPARs: transcriptional effectors of fatty acids and their derivatives Cell Mol Life Sci 59:790–798 13 Krey G, Braissant O, L’Horset F, Kalkhoven E, Perroud M, Parker MG, Wahli W (1997) Fatty acids, eicosanoids, and hypolipidemic agents identified as ligands of peroxisome proliferator-activated receptors by coactivator-dependent receptor ligand assay Mol Endocrinol 11:779–791 14 Robinson-Rechavi M, Marchand O, Escriva H, Bardet PL, Zelus D, Hughes S, Laudet V (2001) Euteleost fish genomes are characterized by expansion of gene families Genome Res 11:781–788 15 Maglich JM, Caravella JA, Lambert MH, Willson TM, Moore JT, Ramamurthy L (2003) The first completed genome sequence from a teleost fish (Fugu rubripes) adds significant diversity to the nuclear receptor superfamily Nucleic Acids Res 31:4051– 4058 16 Kondo H, Misaki R, Gelman L, Watabe S (2007) Liganddependent transcriptional activities of four torafugu pufferfish Takifugu rubripes peroxisome proliferator-activated receptors Gen Comp Endocrinol 154:120–127 17 Boukouvala E, Antonopoulou E, Favre-Krey L, Diez A, Bautista JM, Leaver MJ, Tocher DR, Krey G (2004) Molecular characterization of three peroxisome proliferator-activated receptors from the sea bass (Dicentrarchus labrax) Lipids 39:1085–1092 18 Leaver MJ, Boukouvala E, Antonopoulou E, Diez A, Favre-Krey L, Ezaz MT, Bautista JM, Tocher DR, Krey G (2005) Three peroxisome proliferator-activated receptor isotypes from each of two species of marine fish Endocrinology 146:3150–3162 19 Batista-Pinto C, Rodrigues P, Rocha E, Lobo-da-Cunha A (2005) Identification and organ expression of peroxisome proliferator Fish Sci (2010) 76:167–175 20 21 22 23 24 25 activated receptors in brown trout (Salmo trutta f fario) Biochim Biophys Acta 1731:88–94 Raingeard D, Cancio I, Cajaraville MP (2006) Cloning and expression pattern of peroxisome proliferator-activated receptor alpha in the thicklip grey mullet Chelon labrosus Mar Environ Res 62:S113–S117 Mimeault C, Trudeau VL, Moon TW (2006) Waterborne gemfibrozil challenges the hepatic antioxidant defense system and down-regulates peroxisome proliferator-activated receptor b (PPARb) mRNA levels in male goldfish (Carassius auratus) Toxicology 228:140–150 ¨ , Gjøen T, Ruyter B, Andersen Ø, Dehli A, Farraants AKO Thomassen MS (1997) Peroxisome proliferator activated receptors in Atlantic salmon (Salmo salar): effects on PPAR transcription and acyl-CoA oxidase activity in hepatocytes by peroxisome proliferators and fatty acids Biochim Biophys Acta 1348:331–338 Andersen Ø, Eijsink VG, Thomassen MS (2000) Multiple variants of the peroxisome proliferator activated receptor (PPAR) gamma are expressed in the liver of Atlantic saimon (Salmo salar) Gene 255:411–418 Liu G, Moon TW, Metcalfe CD, Lee LE, Trudeau VL (2005) A teleost in vitro reporter gene assay to screen for agonists of the peroxisome proliferator-activated receptors Environ Toxicol Chem 24:2260–2266 Hyodo-Taguchi Y, Sakaizumi M (1993) List of inbred strains of the medaka, Oryzias latipes, maintained in the Division of 175 26 27 28 29 30 31 Biology, National Institute of Radiological Sciences Fish Biol J Medaka 5:29–30 Liang CS, Ikeda D, Kinoshita S, Shimizu A, Sasaki T, Asakawa S, Shimizu N, Watabe S (2008) Myocyte enhancer factor regulates expression of medaka Oryzias latipes fast skeletal myosin heavy chain genes in a temperature-dependent manner Gene 407:42–53 Thompson JD, Higgins DG, Gibson TJ (1994) Clustal W: improving sensitivity of progressive multiple sequence alignment through sequence weighting positions-specific gap penalties and weight matrix choice Nucleic Acids Res 22:4673–4680 Forman BM, Chen J, Evans RM (1997) Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are ligands for peroxisome proliferator-activated receptors a and d Proc Natl Acad Sci USA 94:4312–4317 Willson TM, Brown PJ, Sternbach DD, Henke BR (2000) The PPARs: from orphan receptors to drug discovery J Med Chem 43:527–550 McClelland GB, Craig PM, Dhekney K, Dipardo S (2006) Temperature- and exercise-induced gene expression and metabolic enzyme changes in skeletal muscle of adult zebrafish (Danio rerio) J Physiol 577:739–751 LeMoine CM, Genge CE, Moyes CD (2008) Role of the PGC-1 family in the metabolic adaptation of goldfish to diet and temperature J Exp Biol 211:1448–1455 123 [...]... from either the left or right lobes of the ovaries of 18 4 females The possibility of Method Capture date range Number Size range (mm; FL) Weight range (g) Gillnet 17 /06/04–30/05/05 13 6 577 1, 2 32 1, 7 58–6,3 31 Line 14 /03/04 12 /09/05 49 772 1, 4 70 4, 410 –38,570 Prawn/fish trawl Overall 29/ 01/ 03–06/ 01/ 05 29/ 01/ 03 12 /09/05 18 203 18 1–458 18 1 1, 4 70 31 6,3 31 31 38,570 Fig 2 Length–frequency distributions (in 50... CPUE (t/day/vessel) 0 .1 8.0 4 .17 0.58 Medium-sized CPUE (t/day/vessel) 2.6–20.4 8.73 1. 13 2.2–57.9 41. 2–89.0 32.85 65 .10 4 .12 3.67 Total CPUE (t/day/vessel) Large-sized proportion (%) Medium-sized proportion (%) a Range t Value (p value) Ratioa (%) Mean SE 18 .45 2.28 -3.08 (0.006) 71. 2 0.3–28.0 7.58 1. 73 -2. 51 (0.0 21) 55.0 2.6–20.5 10 .67 1. 29 -2.78 (0. 012 ) 81. 8 6.5–82.6 17 .2– 91. 1 37.93 60.82 4.72 4.57... test D = 0 .16 7; P = 0.0 51) (Fig 2) The sex ratio of females to males was 2 .18 :1 and was significantly different to the expected 1: 1 (chi-square df = 1; P \ 0.05) Results A total of 315 cobia were examined, ranging from 12 5 to 1, 6 33 mm FL and 0.06 to 55 kg (Table 1) No differences in external morphology between sexes were identified Males (n = 93) ranged from 12 5 to 1, 2 80 mm FL and from Table 1 Summary... gillnet and rod and line in northeastern Australia between January 2003 and September 2005 45 Males n = 93 40 35 30 25 20 15 Frequency 10 5 0 5 10 15 20 25 30 35 40 Females n = 203 45 10 0 200 300 400 500 600 700 800 900 10 00 11 00 12 00 13 00 14 00 15 00 16 00 Fork length (mm) 12 3 Fish Sci (2 010 ) 76:33–43 a difference in oocyte development depending on the location inside each lobe was examined by comparing sections... abundance is larger than the high-seas Acknowledgments This research was funded by the Fisheries Agency, Council of Agriculture, Executive Yuan, and the National 12 3 Fish Sci (2 010 ) 76: 21 31 Science Council, Taiwan (95AS -14 .1. 2-FA-F1(3), NSC 95-2 511 -S026-004, 96AS -15 .1. 2-FA-F2) We are very grateful to the Overseas Fisheries Development Council in Taiwan for providing logbook data from the Taiwan saury... the East China Sea in summer, based on acoustic surveys from 19 97 to 20 01 Fish Sci 70:389–400 16 Iversen SA, Zhu D, Johannessen A, Toresen R (19 93) Stock size, distribution and biology of anchovy in the Yellow Sea and East China Sea Fish Res 16 :14 7 16 3 Fish Sci (2 010 ) 76 :1 11 17 Takeshita K, Ogawa N, Mitani T, Hamada R, Inui E, Kubota K (19 88) Acoustic surveys of spawning sardine, Sardinops melanosticta,... for the Japanese fishing fleets from the Pacific saury fishing grounds to shore were \15 0 km in 19 71 19 91, except in 19 81 19 86 when distances were 17 0–330 km [7] The migration of the Pacific saury to the coastal waters, including inshore and offshore areas, have been well studied and documented since the 19 50s [1 3, 8 11 ] However, the Pacific saury migrating to the high-seas fishing grounds are exclusively... sea surface temperature Fish Sci 72 :11 53 11 65 29 Hofmann EE, Powell TM (19 98) Environmental variability effects on marine fisheries: four case histories Ecol Appl 8(suppl 1) :S23–S32 30 Saitoh S, Kosaka S, Iisaka J (19 86) Satellite infrared observations of Kuroshio warm core rings and their application to study of Pacific saury migration Deep Sea Res 33 :16 01 16 15 31 31 IOC, IHO, BODC (2003) Centenary... al [10 ]), and the cardiac response in the far field after deflating the swimbladder (present study) Signs indicate significant differences (P \ 0. 01) between thresholds obtained in the near field by Ishioka et al (asterisk), by Iwashita et al (plus), and far field 12 3 18 Fish Sci (2 010 ) 76 :13 –20 14 0 SPL (dB) ABR × 12 0 × × ECG in far-field 10 0 Vibration 80 noise ABR 60 noise far-field 40 10 0 10 00 10 000... western North Pacific Ocean, Tohoku sea area, Japan Bull Tohoku Natl Fish Res Inst 56 :11 5 16 3 (in Japanese with English abstract) Fish Sci (2 010 ) 76: 21 31 17 Sugisaki H, Kurita Y (2004) Daily rhythm and seasonal variation of feeding habit of Pacific saury (Cololabis saira) Fish Oceanogr 13 (suppl 1) :63–73 18 Kurita Y (20 01) Seasonal changes in spawning grounds and the abundance of egg-laying of Pacific