Physiological and agonistic behavioural response of Procambarus clarkii to an acoustic stimulus

More than 500 recognised species of crayfish are distributed in aquatic habitats of all substrata types across all continents except Antarctica and Africa (Taylor, 2002). Shelters range from natural assemblages of rocks to constructed burrows in mud or sand. The red swamp crayfish, Procambarus clarkii (Girard 1852), is an invasive freshwater species that originated in the south-central United States and currently shows a cosmopolitan distribution. This species has been imported to Italy for farming purposes since 1987. Escaped crayfish have invaded natural habitats and become stabilised in many ponds, lakes and streams across Italy in recent years (Gherardi et al., 1999). Although this crayfish is an aquatic species, it is highly resistant to air exposure and is able to survive for several days outside the water (McMahon and Stuart, 1999). Several eco-ethological features of P. clarkii explain its rapid spread in the wild. The species’ biological cycle reflects the hydrogeological cycle and water temperature changes in the invaded areas (Gutiérrez-Yurrita et al., 1999). This crayfish is also highly resistant to environmental stress, including extreme temperatures (Gherardi and Holdich, 1999; Paglianti and Gherardi, 2004), the absence of water, high salinity, low oxygen concentrations and the presence of pollutants (Gherardi et al., 2002). The success of P. clarkii as an invader is further supported by its generalist feeding habits as well as its competitive superiority over native species due to its larger claws with high grip strength as well as highly aggressive behaviour (Gherardi and Cioni, 2004).

Individuals of both sexes and assorted sizes usually live together, and social activity is particularly notable from spring to autumn. During this period, adults commonly leave their burrows after sunset and move around the populated area. Crayfish exhibit agonistic behaviour when competing for habitat, shelter, mates and food (Bergman and Moore, 2003). The primary result of such agonistic interactions is the establishment of a dominance relationship that can alter each individual’s access to resources. Aggressive encounters between individuals (agonistic behaviour) are very common (Bergman and Moore, 2003; Buscaino et al., 2012). The crayfish touch each other and assume stereotyped postures aimed at threatening the opponent (Graham and Herberholz, 2009). Crayfish have been used as a behavioural model system to study aggression (Dingle, 1983; Hyatt, 1983) because of their very efficient (large dimensions and high grip strength) chelipeds (Garvey and Stein, 1993; Schroeder and Huber, 2001) and the ritualised nature of their agonistic fights (Bruski and Dunham, 1987). In particular, because of the high frequency of agonistic behaviour in P. clarkii (Bergman and Moore, 2003; Buscaino et al., 2012), observation of the agonistic event and motility (as a factor that could further the agonistic encounters) could evidence alteration in the baseline behaviour due to an external factor, such as an acoustic stimulus. Moreover, as the red swamp crayfish is characterised by a high resistance to environmental stress, this species could serve as a good model with which to examine the impacts of acoustic stimuli on behavioural dynamics and the physiological parameters that reflect stress conditions. In addition, because P. clarkii emits acoustic signals both in air and underwater (Favaro et al., 2011; Buscaino et al., 2012), it is possible that in this species sound (pressure variation and/or particle movements) plays an ecological role (e.g. perception of predator or conspecific movements) so as to make these animals sensitive to an acoustic stimulus. However, some studies have evaluated the effects of stimuli with very high sound pressure levels (e.g. air guns used for seismic surveys) on marine crustacean behavioural and biochemical parameters, such as haemocytes, serum proteins and enzymes, without significant effects (Christian et al., 2003; Andriguetto-Filho et al., 2005). For example, Payne et al. (Payne et al., 2007) found that lobster exposed to very high as well as low sound levels had experienced no effect on delayed mortality or damage to the mechanosensory system associated with animal equilibrium and posture. However, sub-lethal effects were observed with respect to feeding and serum biochemistry, with the effect sometimes being observed weeks to months after exposure.

Crustaceans might experience pain and stress in ways that are analogous to the experience in vertebrates (Elwood et al., 2009). Potentially painful stimuli applied to vertebrates typically produce physiological responses (behavioural changes such as an avoidance reaction), and changes in blood flow, respiratory patterns, and biochemical and endocrine processes (Elwood et al., 2009). There has, however, been limited examination of similar responses in crustaceans. Behavioural observations, in combination with physiological assessment, could provide a more complete understanding of the impact of an external stimulus on an organism, population or species. In particular, this combination of methods could have significant relevance in crustaceans, where the behavioural patterns in response to stress condition are not yet well known. For example, behavioural, physiological and biochemical adaptations have been identified in cave crayfish, such as a decrease in locomotion and oxygen consumption, as well as a decrease in metabolic rates after exposure to environmental stress (Caine, 1978). One possible avenue for evaluation of the impact of an external stimulus is through the cardiac and respiratory systems. It is well known that autonomic control of the respiratory and cardiovascular systems can regulate oxygen availability and nutrients to specific target tissues needed for an impending behavioural response. For example, Schapker et al. (Schapker et al., 2002) showed that crayfish rapidly alter heart rate (f_H) and ventilatory rate (f_V) in response to changes in the environment and that f_H and f_V as indicators were far more sensitive than behavioural data alone. Moreover, Bierbower (Bierbower, 2010) used the tail flip response in crayfish in combination with f_H and f_V as bioindices of the whole animal status to CO₂ exposure as environmental stressor. Specifically, Bierbower observed a repellence/avoidance behaviour that could be the result of avoiding the paralysis resulting from CO₂ exposure, and a decrease until cessation of f_H and f_V in correlation with increasing CO₂ levels. In fish, Buscaino et al. (Buscaino et al., 2010) showed the relationship between behaviour and haematological parameters in relation to noise exposure. In particular, this short-term noise experiment showed an increase in motility and glucidic metabolism of sea bream and sea bass. Hyperglycaemia is a typical response of many aquatic animals exposed to an external stress stimulus. In particular, in crustaceans, increased circulating crustacean hyperglycaemic hormone (CHH) titres and hyperglycaemia are reported to occur following exposure to several environmental stressors (Durand et al., 2000; Lorenzon et al., 2002). Moreover, environmental stress seems to be an important factor for determining the reduction of immunocompetence with increasing prevalence of disease in crustaceans (Sindermann, 1979). Several immune mechanisms in Crustacea, largely based on the activity of the haemolymph cells, have been described.

Haemolymph cells play a central role in the immune mechanisms in Crustacea (Söderhäll and Smith, 1983; Hose and Martin, 1989; Hose et al., 1992; Smith and Chisholm, 1992; Clare and Lumb, 1994; Destoumieux et al., 1997). Three cell types, hyaline, semigranular and granular, are commonly recognised in crustaceans (Bauchau, 1981; Tsing et al., 1989; Hose et al., 1990), and are involved in coagulation, phagocytosis and the production of melanin by the prophenoloxidase (proPO) system. Haemocytes are activated by microorganisms (Vargas-Albores, 1995; Vargas-Albores, et al., 1997) and are involved in the elimination of foreign particles (Hose and Martin, 1989; Bachère et al., 1995). The immune responses include the release of peroxinectin from the blood cells. This protein is involved in cell adhesion, degranulation, and opsonic and peroxidase activity (Johansson et al., 1995). Others proteins involved in defense mechanisms are the stress proteins, also known as heat shock proteins (Hsps), a highly conserved class of proteins that show elevated expression during periods of stress in organisms as phylogenetically divergent as bacteria and humans. Hsp70 is present at low levels in many cells but is highly induced by stress, regardless of the stage of the cell cycle (Hang and Fox, 1996). In decapod crustacean larvae, the elevation in Hsp70 expression was prolonged depending on the day of pesticide exposure (Snyder and Mulder, 2001). This effect was directly related to the observed increase in mortality (Snyder and Mulder, 2001). Liberge and Barthélémy (Liberge and Barthélémy, 2007) showed that heat stress induced the expression of Hsp70 and superoxide dismutase in the shell glands (structures involved in reproduction) of Hemidiaptomus roubaui (Copepoda, Crustacea), and particularly during the formation of the diapause egg envelope. The modulation of certain immunological and/or physiological parameters in response to stressful conditions may serve as an important indicator of health status (Perazzolo et al., 2002).

In this context, our understanding of crustacean immune mechanisms and the signals that trigger haemolymph cells (Jiravanichpaisal et al., 2006) together with behavioural observations could provide a more complete analysis of the effects of stress factors. In this study, we measured changes in agonistic behaviour and haemolymph parameters in red swamp crayfish exposed to 30 min of an acoustic stimulus. Specifically, we measured the motility, number of tail flips and number of fights, and analysed total and differential haemocyte counts (THC and DHC, respectively), glycaemic serum levels, total serum protein concentration and Hsp70 protein expression levels.

Thirty adult red swamp crayfish (P. clarkii) (17 males and 13 females) weighing 26.1±9.3 g (mean ± s.d.) and measuring 9.4±1.0 cm in total length and 4.7±0.6 cm in carapace length were used for this study. The crayfish were captured at the Preola and Gorghi Tondi Natural Reserve (NW Sicily) and acclimated for 1 month at the Capo Granitola/CNR laboratory (SW Sicily) in two shaded, outdoor PVC circular tanks (3.0 m in diameter, 1.0 m in depth) supplied with a thin layer of sand (1 cm deep). The temperature and salinity levels were monitored using a multiparametric probe (556 MPS, YSI Incorporated, Yellow Springs, OH, USA) and kept constant at 24.02±0.38°C (mean ± s.d.) and 0.9±0.01 ppt (mean ± s.d.), respectively, with a constant flow of water at a rate of 25±3.7 l min^–1 (mean ± s.d.). The animals were fed pellets and frozen fish ad libitum. The P. clarkii specimens were deprived of food for 2 days before the start of the experimental trials. All animals were kept under natural photoperiods.

The crayfish were randomly collected from the holding tanks in groups of five individuals, assigned to the control or test group and used in one experiment only. In total, six experimental trials (three controls and three tests) were performed in two experimental tanks (control and test tanks) that lacked shelter.

The animals, five control specimens and five test specimens, were simultaneously released into the control and test tanks, respectively (Fig. 1). After a 1 h habituation period, we monitored and video-recorded the behaviour of the crayfish for 2 h (1 h=pre-experimental phase; 30 min=during-experimental phase; 30 min=post-experimental phase). In the during-experimental phase, individuals in the test groups were exposed to an acoustic stimulus for 30 min. Members of the control group were not exposed to any stimuli. At the end of the post-experimental phase, both control and stimulated animals were captured with a net and placed on crushed ice for 30 min to induce torpor or ‘cold anaesthesia’ to allow sampling of the haemolymph. The samples were immediately collected from five control and five experimental animals, and the crayfish then were transferred into a small tank and released after recovery. This experimental procedure was repeated three times.

Although the ability of the red swamp crayfish to perceive acoustic signals is unknown, this species is able to generate wide-band pulses in air (Favaro et al., 2011) and in water (Buscaino et al., 2012). Based on the idea that animals that produce acoustic signals may be able to perceive said signals (e.g. for conspecific movement perception or communication), we decided to use a stimulus with frequencies contained in both of the signals (air and aquatic environment) produced by P. clarkii. The acoustic stimulus was therefore set to emit at a frequency band of 0.1–25 kHz. Moreover, in the natural environmental, this band frequency is mainly produced by vessel traffic (Sarà et al., 2007).

A 10 s linear sweep with a peak amplitude of 148 dB re. 1 μPa rms at 12 kHz was used to cover the selected frequency band (see Fig. 2). The linear sweep was repeated for 30 min without pause. The signals were generated by a waveform generator (model 33220A, Agilent Technologies, Santa Clara, CA, USA) connected to an underwater moving coil loudspeaker (model UW30, Lubell, Columbus, OH, USA) with a 100 Hz–10 kHz-rated frequency response.

The acoustic stimulus was recorded using a calibrated hydrophone (model 8104, Brüel & Kjær, Nærum, Denmark) with a sensitivity of –205.6±4.0 dB re. 1V μPa^–1 in the 0.1 Hz–80 kHz frequency band. The hydrophone was connected to a digital acquisition card (USGH416HB, Avisoft Bioacoustics, Berlin, Germany; set with a 40 dB gain) managed by dedicated Avisoft Recorder USGH software.

The signals were acquired at 300 kilosamples per second at 16 bits and analysed by the Avisoft-SASLab Pro software. The digital acquisition card was calibrated with pure tone sine waves at different frequencies (1 and 20 kHz) and different intensities (peak-to-peak 0.1 and 0.5 V) produced by a signal generator (Agilent 33220) using the SASLab Pro software.

To avoid disturbing the animals, we placed the equipment required for video monitoring and recording in a laboratory located 5 m away from the tank. The video monitoring was carried out using a low-light camera (model CCD colour camera 1090/205, Urmet Domus, Torino, Italy) placed above the centre of the tanks for an overall view of the experimental space (Fig. 1). The signals from the cameras were digitised and stored using a DAQ card (model DV-RT4 Real Time, D-Vision, Torino, Italy) managed by custom-written software (Model DSE, D-Vision).

The video data were analysed in continuous mode. We identified the agonistic behavioural events reported in other decapods (Buscaino et al., 2011a) and other Procambarus species (Bergman and Moore, 2003; Buscaino et al., 2012): fights and tail flips. Moreover, we considered an event to be an ‘encounter’ when a specimen approached another one without any threat display (Bergman and Moore, 2003; Buscaino et al., 2012).

A fight was considered the approach between two or more specimens that continued in series of agonistic activities including: (1) contact with chelae and progressing to pushing with closed chelae, (2) opened chelae used to grab an opponent and (3) the most intense interaction, in which an individual appears to attempt to injure or injure an opponent by grasping at chelae, legs or antennae (Bergman and Moore, 2003). The achievement of one or more of these behavioural stages in continuous progression was considered a single fight event.

The tail flip is a typical avoidance behaviour event consisting of a rapid abdominal flexion resulting in a new position away from the opponent. In crustaceans, the tail flip is highly associated with sound production (Buscaino et al., 2011a; Buscaino et al., 2011b; Buscaino et al., 2012).

The total number of events (encounter, fight and tail flip) was counted every 6 min. Moreover, because the number of encounters/fights could be influenced by the motility of the crayfish, at the end of the 6 min interval the number of specimens in movement (walking) or stopped (resting) was counted. The observers that analysed the video did not know whether they were observing the pre-, during- or post experimental phase, or the control or acoustic treatment. These events are behavioural indices that are useful for assessing intraspecific interactions in decapods (Bergman and Moore, 2003). Variations in these events represent alterations of the baseline activities.

One millilitre of haemolymph was drawn from the ventral sinus between the first and second abdominal segments using a 2 ml syringe fitted with a 23 gauge needle. To delay or prevent coagulation, the syringe was filled with an equal volume of anticoagulant. After the cellular counts the samples were centrifuged at 800 g for 10 min at 4°C to obtain plasma serum and pellets, which were stored at –20°C for further use.

Haemolymph samples mixed with anticoagulant were dropped directly onto a coverslip pretreated with 0.1% poly-l-lysine. The adherent monolayer was fixed in cacodylate buffer (0.1 mol l^–1, pH 7.3) containing 2.5% glutaraldehyde, post-fixed in osmium tetroxide (1%), dehydrated in a graded alcohol series and dried at the critical point. The samples were mounted on stubs, gold-coated in a sputter coater and observed by scanning electron microscopy (SEM) (LEO 420, LEO Electron Microscopy, Cambridge, UK).

Glucose levels were measured with the Accutrend GC kit (Boehringer, Mannheim, Germany). Osmolarity was estimated with an osmometer (Roebling, Messtechnik, Berlin, Germany). The total protein concentration of the crayfish haemolymph serum plasma was estimated using the Bradford method (Bradford, 1976). Bovine serum albumin (BSA) was used as the protein standard.

The haemagglutinating activity (HA) of twofold diluted samples was assayed in a 96-well microtitre U-plate containing a suspension of 1% rabbit red blood cells (RRBC) or sheep red blood cells (SRBC) in phosphate buffered saline (PBS-E: 6 mmol l^–1 KH₂PO₄, 0.11 mmol l^–1 Na₂HPO₄, 30 mmol l^–1 NaCl, pH 7.4). Erythrocytes were supplied by the Istituto Zooprofilattico della Sicilia (Palermo, Italy) and maintained in sterile Alsever’s solution (27 mmol l^–1 sodium citrate, 115 mmol l^–1 d-glucose, 18 mmol l^–1 EDTA and 336 mmol l^–1 NaCl in distilled water, pH 7.2). Tris-buffered saline (TBS; see below) enriched with 1% RRBC and SRBC with 0.1% (w/v) gelatin was used as the reaction medium. Twenty-five microlitres of plasma were mixed with an equal volume of RRBC or SRBC suspension and incubated at 37°C for 1 h. Divalent cation requirements were estimated by adding CaCl₂ or MgCl₂ to the reaction medium, up to a final concentration of 5–10 mmol l^–1. The titre of the haemagglutinating activity (HT) was expressed as the reciprocal of the highest dilution showing a positive score for agglutination.

TBS was used in place of plasma for the negative controls. Each assay was performed in duplicate using serum samples from different specimen preparations. The HA titre was expressed as the average of the recorded values.

Cells were crushed on ice for 1 h in 1 ml of lysis buffer [RIPA: 0.5% sodium deoxycholate (minimum 97%); 1% NP40; 0.1% SDS with PBS-T (1 mol l^–1 Na₂HPO₄, 1 mol l^–1 NaH₂PO₄, 1.5 mol l^–1 NaCl and 0.1% Tween-20, pH 7.5, supplemented with a cocktail of protease inhibitors: 2 μg μl^–1 antipain, leupeptin and bestatin, 1 μg μl^–1 aprotinin and pepstatin, 1 mmol l^–1 benzamidine and 0.1 mmol l^–1 AEBSF)]. The samples were then centrifuged at 15,000 g for 30 min at 4°C. The supernatants were collected and dialysed against 50 mmol l^–1 Trizma base (Tris [hydroxymethyl] aminomethane), pH 7.5, and the protein contents were estimated.

The equivalent of 25 μg of total lysates for each sample was separated on 7.5% SDS-PAGE under reducing conditions according to the Laemmli method (Laemmli, 1970). SDS-polyacrylamide minigels were transferred to nitrocellulose membranes using a semidry transfer apparatus (Bio-Rad Laboratories, Hercules, CA, USA) and were blocked with 5% BSA in TBS-T [20 mmol l^–1 Trizma base, pH 7.5, 300 mmol l^–1 NaCl, 0.1% (v/v) Tween-20 with 0.02% sodium azide] for 1 h at room temperature. According to Celi et al. (Celi et al., 2012), the membrane was incubated overnight at 4°C with the primary antibody (monoclonal anti-heat shock protein 70 antibody produced in mouse, Sigma-Aldrich, St Louis, MO, USA; 1:800 dilution), washed with TBS-T (three times for 5 min each) and incubated with alkaline phosphatase-conjugated goat anti-mouse IgG (1:7500 for 1 h at room temperature). After washing with TBS-T (three times for 5 min each), the membranes were incubated with the 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) liquid substrate system. AlphaDigiDoc RT and AlphaImager systems as well as AlphaEase FC software (Alpha Innotech, Santa Clara, CA, USA) were used for densitometric analysis of the immunoblotted bands.

Five specimens from each experimental group (control and test groups, 30 samples in total) were examined, and each test was repeated in triplicate.

Because the behavioural data were not normally distributed, the Mann–Whitney U-test was used to compare the encounter, fight and tail flip events between control and test groups as well as among pre-, during- and post-experimental phases. An unpaired t-test was used to determine significant differences in plasma glucose, total protein, THC, DHC and Hsp70 expression levels.

A total of 836 behavioural events were recorded during the six experimental trials, of which 379 were encounters, 380 were fights and 77 were tail flips. No significant differences in behavioural responses were observed between control and test groups in the pre- or post-experimental phases (Fig. 3). Conversely, in the during-experimental phase, significant differences in the numbers of encounters, fights and tail flip events were observed between the control and test groups (P<0.05; Fig. 3). In particular, during the acoustic stimulus, the test crayfish exhibited a lower number of encounters, fights and tail flips than the control animals. No significant differences were found in motility in the during-experimental phase between the control and test groups, nor in the ratio of during-/pre-/post-experimental phase motility for the control and test groups (P>0.05).

The number of circulating haemocytes (THC) was ∼4.7×10⁶±0.4×10⁵ cells ml^–1. Hyalinocytes represented ∼20±2.4% of the total circulating haemocytes, whereas semigranulocytes accounted for 22.5±3.3%, and granulocytes 57.5±2.2%.

Acoustic stimuli significantly affected both THC and DHC. Following the 30 min acoustic stimulus, the THC of the stressed crayfish decreased by ∼50% relative to the initial count (P<0.001; Table 1). A different pattern was observed for the DHC. In tested crayfish, a significant increase in hyaline cell number (from 20% to 58%, P<0.001) was accompanied by significant decreases in the relative proportions of granular and semigranular cells (P<0.001 and P<0.01, respectively) relative to the values determined for the control group (Table 1).

The haemocyte monolayers were comprised of flattened and well-spread cells and contained three morphologically distinct cell types that could be differentiated by the presence and size of granules. Three types of circulating haemocytes were identified by light microscopy and SEM. Granulocytes are ovoid or fusiform and contain large acidophilic granules (Fig. 4A) that give a rough texture to the cell surface (Fig. 4B). Semigranulocytes can be fusiform and generally display a smooth shape (Fig. 4D). These cells contain small cytoplasmic granules that are typically eosinophilic (Fig. 4C). Hyaline cells appear ovoid or fusiform in shape and are characterised by the absence of cytoplasmic granules (Fig. 4E) and a smooth surface (Fig. 4F).

Glucose levels were significantly higher (575±34 mg dl^–1, P<0.01) in the test group than in the control group (Table 2). However, no differences in osmolarity or total protein content were observed between groups (Table 2).

The serum samples from the control crayfish agglutinated both RRBC and SRBC. The highest titre was found with sheep SRBC (7.2±1.66), whereas the RRBC yielded a score of 6.3±2.1. The HA of the serum from stressed specimens was decreased by ∼50% (P<0.01; Table 2).

As shown in Fig. 5A, the anti-mouse Hsp70 mAb cross-reacted with a 70 kDa band in circulating haemocytes from both the control and test groups. A densitometric analysis of Hsp70 protein levels (Fig. 5B) revealed a significant increase in expression in the haemocytes collected from stressed animals. Hsp70 expression peaked (threefold higher than the untreated samples) after the period of acoustic stimuli.

This study showed that an acoustic stimulus can reduce the agonistic behaviour of the crayfish P. clarkii, as demonstrated by the significantly reduced numbers of both fights and tail flip events. According to Bergman and Moore (Bergman and Moore, 2003), P. clarkii engage in agonistic interactions with high frequency to establish dominance relationships that regulate access to resources. Similarly, Capelli and Hamilton (Capelli and Hamilton, 1984) have shown that in a laboratory environment, food and shelter affect the agonistic behaviour of the crayfish Orconectes rusticus. In particular, aggressive activity decreases with the increased availability of both shelter and food. To avoid these effects, we observed crayfish held in tanks without shelter and deprived of food for 2 days before the experimental trials.

In the during-experimental phase, although the motility of specimens in the test group was lower, we did not observe a significant reduction in comparison to the specimens of the control group. However, a significantly lower number of encounters in the test group were observed in the during-experimental phase. Similarly, the acoustic stimulus induced a decrease in the natural aggressive activity (number of fights and tail flip events) of the crayfish. Accordingly, when the acoustic stimulus was interrupted (post-experimental phase), an increase in aggressive agonistic behaviour was observed.

Our results indicate that P. clarkii could perceive all or part of the acoustic stimuli used in this study (0.1–25 kHz bandwidth) within the wider bandwidth of their underwater acoustic emissions (Buscaino et al., 2012). However, no data on the anatomical–functional structures with which they detect acoustic energy (such as variation in pressure) are currently available. The sensitivity of aquatic decapods to particle displacement and hydrodynamic stimulation is poor compared with other aquatic organism such as fishes (Breithaupt and Tautz, 1990; Goodall et al., 1990; Popper et al., 2001). From our study is not possible to discern the quantitative and qualitative receptors involved in the perception of the variation of pressure or particle water movement or determine whether the sound produced physical effects on the entire animal body without stimulating a specific receptor. Previous studies have shown that lobster and crayfish primarily respond to hydrodynamic stimulation (behavioural study) rather than pressure (Goodall et al., 1990; Popper et al., 2001). However, these studies focused on acoustic frequencies lower (20–180 Hz) than those used in the present study (0.1–25 kHz).

In crustaceans, mechanoreceptors are located in cuticular extensions of the exoskeleton and are called sensilla (Ali, 1987). Decapod mechanoreceptors include setae (hair-like cells), chordotonal organs and internal statocysts (Popper et al., 2001). In macruran decapods (crayfish), the hairs are sensitive to vibration (Breithaupt and Tautz, 1990) and can respond to frequencies up to 100 Hz. In P. clarkii, the antennules (lateral plus medial flagella) possess both chemosensory and mechanosensory setae. The latter respond to hydrodynamic stimuli up to 100 Hz (Breithaupt and Tautz, 1990). In particular, the medial flagellum functions as a hydrodynamic receptor (Horner et al., 2008; Monteclaro et al., 2010). In our study, the animals did not show any preference in the choice of the tank side to occupy (near to or far from the underwater loudspeaker) in the during-experimental phase. However, it is possible that the reverberation and/or reflection effects of the sound inside the tank made the animals unable to detect the sound source, inhibiting them from performing the avoidance behaviour. Moreover, the agonistic behaviours were reduced during the stimulation, probably as a consequence of the impact of an external stress condition, which may inhibit the aggressive state as a result of a preservation instinct. Crayfishes could prioritize the external stimuli (considered to be a stress source, and confirmed by the increase of glycaemic and Hsp70 levels and decrease of THC) with respect to other baseline agonistic behaviours.

In marine shrimps and crabs (Christian et al., 2003; Andriguetto-Filho et al., 2005), exposure to stronger acoustic stimuli (air guns) produced no obvious effects on behaviour or biochemical parameters (serum proteins, serum enzymes, calcium and haemocyte types). However, Payne et al. (Payne et al., 2007) observed a sub-lethal effect as well as a decreasing feeding rate in lobsters exposed to the air gun stimulus. The variant findings of these prior studies and the present investigation suggest that the effects of acoustic stimuli are perceived differently under different environmental conditions (e.g. tank or natural environment, distance from the acoustic source) and acoustic typologies (e.g. source level, frequency, duration) or even between different crustacean species. In fish, Santulli et al. (Santulli et al., 1999) reported that exposure to air gun blasts affected biochemical parameters (cortisol, glucose, lactate, AMP, ADP, ATP and cAMP) in sea bass, and Buscaino et al. (Buscaino et al., 2010) showed that the exposure of sea bass and gilthead sea bream to a 0.1–1 kHz linear sweep (150 dB_rms re. 1 μPa) caused a significant increase in motility that in turn influenced haematological parameters.

In aquatic crustaceans, various types of stress, including hypoxia (Le Moullac et al., 1997; Le Moullac et al., 1998), low salinity (Perazzolo et al., 2002), viral infection and administration of an immunostimulant (Hennig et al., 1998; Sritunyalucksana et al., 1999), affect haematological parameters. In aquatic animals, hyperglycaemia is a typical stress response to harmful physical and chemical environmental changes, including hypoxia and exposure to air during commercial transport (Spicer et al., 1990; Zou et al., 1996; Kuo and Yang, 1999; Morris and Olivier, 1999; Durand et al., 2000; Speed et al., 2001). Hyperglycaemia has been associated with increased circulating CHH titres (Lorenzon et al., 1997; Lorenzon et al., 2002; Durand et al., 2000; Santos et al., 2001), and has been used as an index to assess CHH activity and environmental stress (Webster, 1996; Bergmann et al., 2001; Toullec et al., 2002). Accordingly, in P. clarkii in the present study, acoustic stress led to a significant increase (P<0.01) in haemolymph glucose levels.

However, exposure of P. clarkii to an acoustic stimulus did not result in any significant effects on internal osmoregulatory capacity or total serum protein concentration. In aquatic crustaceans and particularly in decapods, the organs of the branchial chambers are the primary source of osmotic and ionic regulation (Péqueux, 1995). In terms of osmoregulation, exposure to environmental stressors and pathological agents usually results in a decrease of Na⁺ and Cl^– regulation in crustaceans. The partial or complete loss of osmoregulatory and ionoregulatory capacity is generally linked to distruptions of osmotic and ionic regulation. Studies on crustacean responses to various environmental stressors have revealed that the effect of stress upon osmotic and ionic metabolism is time- and dose-dependent (Charmantier et al., 1989; Charmantier and Soyez, 1994; Lignot et al., 2000).

Changes in the protein composition of haemolymph has been used like a stress indicator to monitor shrimp health status, and exposure to environmental stress (Chen et al., 1994; Chen and Cheng, 1995) seems to depend on certain physiological and environmental variables (Bursey and Lane, 1971; Chen and Cheng, 1993; Chen et al., 1994; Chen and Cheng, 1995), sex and animal size (Chen and Cheng, 1993).

Naturally occurring agglutinins, including those with erythrocyte targets (haemagglutinins), are involved in innate immunity in invertebrates (reviewed by Marques and Barracco, 2000). In our study, acoustic stress induced a significant decrease in the agglutinating titre of P. clarkii serum, as assayed with sheep and rabbit erythrocytes. Further analyses using sugar inhibition might elucidate whether the serum haemagglutinins are lectins (Sharon, 2007).

In crustaceans, haemocytes are involved in organismal homeostasis and manage several immune functions, including coagulation, phagocytosis, degranulation, opsonisation and production of melanin by the proPO system (Vargas-Albores, 1995; Vargas-Albores et al., 1997). According to Lanz et al. (Lanz et al., 1993), P. clarkii granular and semigranular haemocytes may participate in the proPO system as well as in phagocytic or cytotoxic functions.

THCs and DHCs have been used to assess crustacean health and the effects of stressful conditions (Jussila et al., 1997). Decreases in the THC under stressful conditions have been reported for several marine crustacean species (Smith et al., 1995; Hennig et al., 1998; Le Moullac et al., 1998; Sánchez et al., 2001). Similarly, under acoustic stimuli, the THCs of our P. clarkii specimens were reduced, suggesting the possibility of immune depletion as well as an increased risk of infection. Moreover, acoustic stimuli also altered the DHC (relative proportions of granulocytes, semigranulocytes and hyaline in the THC). Although the response of the DHC to different stressors is not well understood, it has been used as a stress indicator in crustaceans (Jussila et al., 1997; Johansson et al., 2000). Acoustic stimuli resulted in an increase in the relative number of hyaline cells and decreases in semigranulocytes and granulocytes. These results are similar to those reported in the literature (Jussila et al., 1997; Fotedar et al., 2001; Fotedar et al., 2006), where the proportions of granulocytes and semigranulocytes were lower in moribund lobsters than in healthy individuals (Bauchau, 1981; Sequeira et al., 1995).

Stress proteins (Hsps) are a highly conserved class of proteins that show elevated transcription during periods of stress in a wide range of organisms. These proteins have been shown to play numerous important roles in maintaining organismal health, e.g. in the host responses to environmental pollutants and food toxins as well as the development of inflammation. In shrimp, Hsps are involved in the specific and non-specific immune responses to bacterial and viral infections (Roberts et al., 2010). In particular, Hsp70 acts to repair damage to proteins following acute stress and thus plays a key role in cytoprotection (Feder and Hofmann, 1999).

In crustaceans, Hsp70 expression serves as a good bioindicator of stressful conditions, including pesticide exposure and heat stress (Snyder and Mulder, 2001; Chang, 2005; Liberge and Barthélémy, 2007). Little is known about the effects of noise on Hsp expression. Wu et al. (Wu et al., 2001) showed that Hsp70 expression increased after exposure to a stressful noise in humans, and Hoekstra et al. (Hoekstra et al., 1998) reported that the expression of Hsp70 (but not Hsp30, Hsp60 or Hsp90) is increased in birds after exposure to a loud noise. We have also recently detected increased Hsp70 expression in the fish Chromis chromis after exposure to sounds similar to those resulting from human activities (M.V., M.C., D. Arizza, G. Calandra, G.B., D.P., V. Ferrantelli, C. Bracciali and G. Sarà, unpublished). In the present study, we show for the first time that acoustic stimuli induce Hsp70 overexpression in P. clarkii haemocytes as an expression of stress status.

In conclusion, exposure to acoustic stimuli altered certain aggressive behavioural patterns and components of the haemato-immunological system of P. clarkii. Among the assessed haemato-immunological parameters, serum glucose concentration, protein concentration, agglutinating activity, THC, DHC and Hsp70 expression are the most promising parameters reflecting stress status in crayfish. The haemato-immunological responses to the stressful conditions occur in conjunction with behavioural changes.

In most natural aquatic environments, the soundscape has been permanently altered as a result of anthropogenic activities (e.g. traffic vessels, wind turbines, electroacoustic instruments for exploration and navigation), and the impact on aquatic organisms should be investigated over brief, medium and long-term exposure periods (Payne et al., 2007). However, in accordance with Goodall et al. (Goodall et al., 1990), further studies should be also performed in an open, controlled, natural environment (where the acoustical field is not influenced by walls such as in the small tanks) to increase the information about the effects of noise on the behavioural and physiological response.

 
• BSA

  bovine serum albumin

 
• CHH

  crustacean hyperglycaemic hormone

 
• DHC

  differential haemocyte count

 
• f_H

  heart rate

 
• f_V

  ventilatory rate

 
• HA

  haemagglutinating activity

 
• Hsp

  heat shock protein

 
• HT

  titre of the haemagglutinating activity

 
• PBS

  phosphate buffered saline

 
• proPO

  prophenoloxidase

 
• RRBC

  rabbit red blood cells

 
• SEM

  scanning electron microscopy

 
• SRBC

  sheep red blood cells

 
• TBS

  Tris-buffered saline

 
• THC

  total haemocyte count