Feeding Habits of Amphipods (Crustacea: Malacostraca) from Shallow soft Bottom Communities: Comparison between Marine Caves and Open Habitats

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Elsevier in: JOURNAL OF SEA RESEARCH on 2013, available at: https://doi.org/10.1016/j.seares.2012.12.011” Feeding habits of amphipods (Crustacea: Malacostraca) from shallow 1 soft bottom communities; comparison between marine caves and open 2 habitats 3 4 5 Carlos Navarro-Barranco1,3, José Manuel Tierno-de-Figueroa2,3, José Manuel Guerra- 6 García1,3, Luis Sánchez-Tocino2, José Carlos García-Gómez1 7 8 1 Laboratorio de Biología Marina, Departamento de Zoología, Facultad de Biología, 9 Universidad de Sevilla, Avda. Reina Mercedes 6, 41012, Sevilla, Spain. 10 2 Departamento de Zoología, Facultad de Ciencias, Universidad de Granada, Campus 11 Fuentenueva s/n, 18071,Granada, Spain. 12 3 Jun Zoological Research Center, C/ Los Jazmines nº 15, 18213, Jun, Granada, Spain. 13 14 Corresponding author: Carlos Navarro Barranco. Laboratorio de Biología Marina, 15 Departamento de Zoología, Facultad de Biología, Universidad de Sevilla, Avda. Reina 16 Mercedes 6, 41012, Sevilla, Spain. E-mail: carlosnavarr[email protected]. Tel: +34 954556229 17 18 Abstract 19 20 Marine caves are environments of great interest since the organisms that inhabit them 21 are forced to develop specific adaptations to high constraint conditions. Because of 22 some of these particular conditions, such as light absence or oligotrophy, it can be 23 expected that feeding strategies into caves differ from that present outside them. 24 Nevertheless, no studies have been done to compare the trophic structure of marine 25 caves and open habitats, at least for amphipod communities, considering their 26 importance both inside and outside of the caves. In this study, the diet of the dominant 27 amphipod species living on shallow sediments, both inside and outside of six marine 28 caves in western Mediterranean, was characterized. Thereby, the gut content of 17 29 amphipod species was studied, being this study the first attempt to establish the feeding 30 habit of most of these species. Analysis of digestive contents of the species showed that 31 amphipod diet is less diverse in sediments than in other environments, such as algae and 32 seagrasses. No herbivorous species were found in the sediment and carnivorous 33 amphipods showed a little variety of prey, feeding mainly on crustaceans. Differences in 34 the trophic structure were also found between marine caves and open habitats 35 sediments: while outside the caves detritivorous was the dominant group (both in 36 number of species and number of individuals), amphipods mainly play the role of 37 carnivorous inside the caves. No detritivorous species were found into the caves, where 38 carnivorous represents almost 60% of amphipods species and more than 80% of 39 amphipods individuals. This pattern obtained in amphipods differ from the general trend 40 observed in marine caves organisms, for which a generalist diet, such as omnivory, 41 usually is an advantage in these oligotrophic conditions. The possible causes of this 42 pattern are discussed. 43 44 Key words: 45 Feeding habits, marine caves, soft-bottom, amphipods, gut contents, trophic dynamics 46 47 Introduction 48 49 Submarine caves constitute interesting ecosystems characterized by distinctive 50 biological and ecological peculiarities (Gili et al., 1986). Despite the increased interest 51 of the last decades in the study of these habitats (Benedetti-Cechi et al., 1996), many 52 aspects remain almost unknown. Marine caves can be inhabited by specialized taxa, 53 many of them restricted to live in caves (sometimes exhibiting extreme adaptations), or 54 by more generalist taxa than can find temporal or permanent refuge there (e.g. Bussotti 55 and Guidetti, 2009; Harmelin et al., 1985; Vacelet et al., 1994). In any case, from the 56 feeding point of view, caves are usually poor environments that greatly condition their 57 trophic webs, at a global scale, and the particular diet of the species present, at a low 58 scale. Thus, the absence of light in marine caves (common fact to freshwater and 59 terrestrial caves) avoids the presence of primary producers (except in some cases 60 chemosynthetic bacteria) (e.g. Airoldi and Cinelli, 1996; Pohlman et al., 1997). 61 Parravicini et al. (2010) pointed out that, because of the differences in water 62 confinement and trophic depletion among different sectors inside marine caves, 63 consistent variations in the trophic guild of sessile species can be found. Additionally 64 for soft bottom communities, the existence of a lower hydrodynamism influences their 65 habitat structure, usually offering a finer sediment substrate than the surrounding seabed 66 (Bamber et al., 2008). Iliffe and Bishop (2007) notes, at community levels, that the 67 scarcity of food in anchialine caves drives organisms toward a generalist diet, being 68 expected that detritus feeders, together with omnivorous, dominate in these 69 environments. Nevertheless, the role of a particular taxon needs to be evaluated to really 70 categorize it inside the trophic web. In fact, it is amazing how little is known about the 71 ecology of caves in general, particularly when it comes to their trophic structure 72 (Romero, 2009). 73 In soft-bottom marine substrates, amphipod crustaceans are one of the most 74 important and diverse component of the fauna (Dauvin et al., 1994; Fincham, 1974; 75 Lourido et al., 2008; Prato & Biandolino, 2005) structuring benthic assemblages (Duffy 76 and Hay, 2000). Amphipods are frequent both inside and outside caves (Bamber et al., 77 2008; Iliffe, 2005; Navarro-Barranco et al., 2012). In relation with all the previously 78 noted, the study of the feeding habits of amphipods can constitute an interesting tool to 79 understand the trophic webs in the benthos of marine caves and their surroundings. 80 Thus, the aim of the present study is to describe the feeding habits of the most 81 abundant species of amphipods inhabiting inside and outside of some marine caves in 82 western Mediterranean, to compare the obtained results and to test the hypothesis than 83 species and/or populations living inside the caves are more detritivorous or omnivorous 84 than those living outside of the caves. 85 86 87 Material and methods 88 Six karstic marine caves were selected for this study (Fig. 1 and Table 1). All the 89 caves presented similar length (10-25 m) and morphology, with a single submerged 90 entrance followed by a rectilinear blind-ending tunnel without air chambers. The 91 sampling was conducted during July and August of 2011. Two sampling stations were 92 selected in each cave: one in the exterior area (outside the cave) and another inside the 93 cave (each one approximately 10 m from the cave mouth). Samples were collected in 94 each station using a hand-held rectangular core of 0.025m2 to a depth of 10 cm by 95 SCUBA diving. Samples were washed using a 0.5 mm mesh sieve, fixed with ethanol 96 (70%) stained with rose Bengal. Amphipods were sorted under binocular microscope. 97 Additional samples were collected at each station for physicochemical analyses of the 98 sediment. All samples were immediately stored frozen until the laboratory analyses. 99 Granulometry was determined following the method proposed by Guitián and Carballas 100 (1976). Organic matter was analyzed by dichromate oxidation and titration with ferrous 101 ammonium sulphate (Walkley and Black, 1934). 102 To characterize the feeding habits of the amphipod community, we selected the 103 species which contributed with at least 1% to the total abundance. In total, we studied 104 137 specimens of 7 species of amphipods living inside caves and 351 specimens of 13 105 species living outside caves. The whole list of analyzed material and details of 106 provenance is included in the table 1, and the mean abundance of the amphipod species 107 inside and outside the caves is included in Figure 2. 108 For the diet study, individuals were analyzed following the methodology 109 proposed by Bello and Cabrera (1999) with slight variations. This method has been 110 successfully used to study the gut contents of different arthropod groups and other 111 animals, both aquatic and terrestrial forms and both ethanol and formalin preserved 112 samples, revealing that it as a very appropriate method for gut content analysis (e.g. Bo 113 et al. 2012; Fenoglio et al., 2008; Tierno de Figueroa et al., 2006). Particularly, this 114 method has been used previously in Amphipoda (Alarcón-Ortega et al., 2012; Guerra- 115 García and Tierno de Figueroa, 2009; Vázquez-Luis et al., 2012). Each individual was 116 introduced in a vial with Hertwig’s liquid (consisting on 270 g of chloral hydrate, 19 ml 117 of chloridric acid 1N, 150 ml of distillated water and 60 ml of glycerin) and heated in an 118 oven at 65 ºC for approximately 3 hours. After this, they were mounted on slides for its 119 study under the microscope. The percentage of the absolute gut content (at 100x), as the 120 total area occupied by the content in the whole digestive tract, and the relative gut 121 content (at 400x), as the area occupied for each component within the total gut content, 122 were estimated using the microscope equipped with an ocular micrometer. Mean and 123 standard error of the mean were calculated. Amphipod species were assigned to their 124 feeding group (detritivorous, carnivorous, omnivorous) according to the diet. When the 125 gut content included more than 50% of prey, we catalogued the species as a 126 carnivorous, considering that other material can appear as the prey gut content or 127 accidentally ingested when preying. 128 The affinities among species according to the dietary analysis were explored by 129 cluster analysis using UPGMA (unweighted pair group method using arithmetic 130 averages) and Bray Curtis similarity index. This multivariate analysis was carried out 131 using the PRIMER package (Clarke and Gorley, 2001). For the three species inhabiting 132 both inside and outside of the caves (Harpinia crenulata, Harpinia pectinata and 133 Perioculodes longimanus), we used an analysis of variance (ANOVA) with the 134 following factors: ‘Position’, a fixed factor with two levels: outside and inside the 135 caves; ‘Species’ a fixed factor and orthogonal, with three levels (the three species). The 136 number of specimens with detected digestive contents was different among species. 137 Consequently, to properly conduct a balanced ANOVA design, we chose the species 138 with the lowest values (n=8 for Perioculodes longimanus) and we selected randomly 8 139 specimens of the remaining species. Additional ANOVA was used to test whether the 140 organic matter and silt and clay percentage in the sediment changes between positions. 141 Prior to ANOVA, heterogeneity of variance was tested with Cochran’s C-test. 142 Univariate analyses were conducted with GMAV5 (Underwood et al., 2002). 143 144 145 Results 146 147 In total, we studied 488 specimens of 17 species of amphipods. Digestive 148 contents were found in 238 of the 351 specimens outside the caves (67.8%) and 82 of 149 the 137 specimens inside the caves (59.9%) (Table 2). The average of the total area 150 occupied by the content in the whole digestive tract ranged from 11.1 to 73.3% outside 151 the caves and from 9.4 to 58.2% inside the caves. Gut contents of the studied amphipod 152 species included mainly detritus and crustaceans, and secondly microalgae and 153 polychaetes. According to the diet, species were included in three different feeding 154 groups: detritivours (6 species), carnivorous (5 species) and omnivorous (6 species) 155 (Table 2). This classification was supported when the cluster analysis was conducted 156 (Figure 3). Carnivorous and detritivorous constituted well defined groups, while 157 omnivorous species were in an intermediate position, with most of species closer to 158 detritivorous, and one species, Hippomedon massiliensis, closer to carnivorous. 159 We found species belonging to the three feeding groups outside the caves, while 160 inside the caves the detritivorous were absent. Carnivorous dominated the caves, both in 161 number of species and abundances, while detritivorous were dominant outside the caves 162 (Figure 4). The three studied species which were found inside and outside the caves, 163 Harpinia crenulata, H. pectinata, and Perioculodes longimanus, were carnivorous. 164 There were no significant differences in digestive contents (measured as percentage of 165 detritus) among the three species. Furthermore, each species showed the same feeding 166 pattern inside and outside the caves (Table 3, Figure 5). 167 In connection with physicochemical parameters, organic matter was higher 168 inside the caves, although values were always low (<1%) both inside and outside the 169 caves (Figure 6). The studied caves showed a dominance of very fine and fine sands and 170 this pattern was obtained inside and outside. However, the percentage of silt and clay 171 was significantly higher inside the caves (Figure 6). 172 173 Discussion 174 175 The present work is the first comprehensive study on the feeding habits of 176 amphipod species inhabiting soft bottom communities. There are very few studies 177 dealing with the diet of amphipod species and this biological aspect is unknown for 178 many of them (Scipione 1989). Moreover, the knowledge about the feeding of many 179 species was based in assumptions from fecal pellets composition or morphological 180 characters, but these assumptions must be taken carefully since they are not always 181 fulfilled. For example, the feeding habit of caprellids have been traditionally inferred 182 from mouthparts or antennae morphology (Caine 1974, 1977); however, Guerra-García 183 and Tierno de Figueroa (2009) showed that there are not always clear relationships 184 among these characters and the digestive contents. 185 Only the diet of 2 of the 17 species considered had previously been studied 186 based in the gut contents. The data obtained in the present study for these two species, 187 Pariambus typicus and Phtisica marina, agree with those obtained by Guerra-García 188 and Tierno de Figueroa (2009). On the other hand, the knowledge about feeding habits 189 in amphipods is limited, in some cases, to family level. In this sense, our results usually 190 agree with the previous literature. According to our data, Siphonocetes sabatieri is a 191 strictly detritivorous species, which agree with the previously recorded for the 192 Corophidae family (Bellan-Santini and Ruffo, 1998). Studying the crop contents of 193 some phoxocephalid species, Oliver et al. (1982) concluded that this group acts as a key 194 taxon in soft bottom communities, playing an important functional role as carnivorous. 195 Other studies using stable isotope analyses also showed the carnivorous behavior of 196 Harpinia species (Fanelli et al., 2009). Our data support this assertion, since almost all 197 the phoxocephalid species studied (Harpinia pectinata, H. crenulata and Metaphoxus 198 fultoni) had a diet mainly composed by crustaceans. Nevertheless, in some cases, 199 species belonging to the same family or even genus showed different feeding habits. For 200 example, Harpinia antennaria had an omnivorous diet in the present study. Urothoe 201 elegans was omnivorous while Bathyporeia guilliamsoniana (both of them belonging to 202 Haustoridae family) fed exclusively on detritus. These results show that the feeding 203 habits of an unstudied species cannot be inferred from studies of congeners, as it has 204 been noted for other aquatic invertebrates (e. g. Stewart and Stark, 2002), and 205 consequently gut content analyses are needed to correctly assign each taxon to each 206 feeding group. 207 Within a same species, variations in the trophic preferences can also be found 208 when studying different environments. In a comparative work carried out in the 209 Mediterrranean to study the variations in the gut contents of some amphipod species 210 living in native and invasive seaweeds, Vazquez-Luis et al. (2012) found that some 211 species (such as Amphitoe ramondi or Dexamine spiniventris) had different feeding 212 modes depending on the habitat they live on. In the same way, Alarcón-Ortega et al. 213 (2012) reported variations in the feeding behavior of the caprellids Aciconula 214 acanthosoma and Caprella equilibra when they live in algae, hydroids or gorgonians. In 215 our study, the three species found both inside and outside the caves did not show 216 significant changes in their gut components between both habitats. The absence of 217 differences in the feeding modes for Harpinia crenulata, H. pectinata, and Perioculodes 218 longimanus (all of them carnivorous) between habitats could be due to the availability 219 of this food source inside and outside the caves. 220 Strong differences between habitats can also been found when all amphipod 221 assemblages are considered. Our data reflects that amphipod diet in sediments is less 222 diverse than in other environments such as algae, seagrasses (Guerra-García & Tierno 223 de Figueroa, 2009; Vázquez-Luis et al., 2012). The percentage of algae in the gut 224 contents was zero or very low in all cases. Obviously, that pattern was expected inside 225 the caves, where no primary producer is present, but it is remarkable that no herbivorous 226 species could be found in the external sediments. The variety of prey in carnivorous 227 species was also lower, composed mainly (and often exclusively) by crustaceans. 228 Diet diversity in soft bottom amphipod assemblage was even less diverse inside 229 the caves. No detritivirous species was present inside the caves and, while this feeding 230 group was the dominant group (both number of species and number of individuals) in 231 the external sediments, cave community was dominated by carnivorous species. This 232 pattern in amphipod species is highly unexpected since the general pattern in cave 233 communities drives organisms toward a generalist diet (with a dominance of 234 omnivorous and detritivorous species). However, as in such case, some particulars 235 groups can follow a different trend. Bianchi (1985) found a high percentage of 236 carnivorous species in polychaetes assemblages of marine cave. Accordingly to 237 Romero (2009), caves amphipods feeds mostly on detritus but our data show that they 238 occupy mainly the role of carnivorous. The reason of the absence of detritivorous 239 amphipods into these caves remains still uncertain. A potential factor could be the 240 limited food supply, since marine caves are usually considered as oligrotrophic habitats. 241 Experimental studies in vegetated habitats reveal that species richness and abundance of 242 amphipods was highly and positively related with the detritus content (Vazquez-Luis et 243 al., 2009). However, our chemical data showed significant higher concentrations of 244 organic matter inside the caves, although it has been suggested that the quality of 245 organic matter in marine caves may decrease towards the inner parts (Fichez, 1991). 246 The feeding behavior, not only the feeding group, should be also an important factor to 247 consider. Because of the low hidrodinamism inside caves, strong differences between 248 habitats can be expected in the way in which such organic matter is available to 249 organisms. Thereby, while inside the caves the organic matter percentage in the 250 sediment is higher, the suspended detritus is much less abundant. In that way, gatherer- 251 collectors find abundant settle detritus to feed inside caves while filter feeders should 252 have more problems to colonize cave environments, since they need more exposed 253 habitats, with a greater degree of turbulence. Nevertheless, little is known about the 254 feeding behavior of detritivorous species present inside our caves and about amphipods 255 in general. Moreover, Shillaker and Moore (1987) pointed out that some detritivorous 256 species can even change its feeding methods, using alternative techniques in poor 257 feeding conditions. So, we cannot relate assertively the absence of detritivorous species 258 inside the caves with their feeding behavior. The finer sediment inside the caves neither 259 seems to be the problem, since many of the detritivorous species present outside the 260 caves usually inhabit muddy sediments (Bellan-Santini et al., 1982, 1989, 1993). There 261 are many others physical gradients (light, salinity or oxygen) and the low capacity of the 262 larvae to colonize such habitats, among others, which can acts like limiting factor for 263 the presence of detritivorous. 264 Further studies about the ecological conditions in marine caves and the feeding habit 265 and behavior of all the soft-bottom community are necessary to understand the trophic 266 dynamic of such environments. 267 268 Acknowledgments 269 409 410 411