Introduction

The key role of sex pheromones in mate recognition of many decapod crustaceans has been well documented (Dunham 1978, 1988 for reviews). In many groups, such as crabs (Ryan 1966; Gleeson 1980; Seifert 1982; Hardege et al. 2002), lobsters (Atema 1984 for a review), and crayfish (Ameyaw-Akumfi and Hazlett 1975; Tierney et al. 1984), the urine of females contains a soluble pheromone that acts over a distance to attract mating partners. However, as Burkenroad (1947) proposed, some shrimp may, in addition to distance cues, possess contact pheromones (substance coating female’s cuticle), allowing a male to recognize the receptive female (Kamiguchi 1972; Bauer 1979). Behavioral evidence suggests the existence of such contact pheromones in the caridean shrimps Palaemonetes pugio (Caskey and Bauer 2005), Lysmata ankeri (personal observation), Lysmata boggessi and Lysmata wurdemanni (Zhang and Lin 2006).

Shrimps in the genus Lysmata have attracted much attention because they have an unusual reproductive system, protandric simultaneous hermaphroditism (see Bauer 2000 for a review). To date, all these studies indicate that individuals in the genus first develop into a male-phase and then may change sex to a euhermaphrodite-phase (termed female-phase by Bauer and his colleagues, or simultaneous hermaphrodite by R. Calado) with both male and female functions (Bauer 2000 for a review). The intermolt euhermaphrodite-phase shrimp that functions as a male is able to mate with the newly molted euhermaphrodite-phase shrimp that plays the female role. Male-role shrimp (MP or intermolt EP) of L. ankeri, L. boggessi, and L. wurdemanni have an active pre-copulatory behavior that indicates these shrimp track and locate the receptive females by both distance and contact pheromones (Zhang and Lin 2006; personal observation).

Contact pheromones may play a series of important ecological roles, such as in mate recognition, kin recognition, sexual selection, and preventing gene exchange between sibling taxa (e.g. Higgie et al. 2000; Shine et al. 2002; Howard et al. 2003). Contact sex pheromones are well known and have been identified in a variety of insect species including Drosophila where they were identified as cuticular hydrocarbons (e.g. Linn and Roelofs 1995; Etges and Jackson 2001; Ginzel et al. 2003; Zhang et al. 2003; Sugeno et al. 2006), but they have not been well investigated in aquatic invertebrates, including crustaceans (Fletcher and Hardege 2009). In aquatic animals, it has been found that a diverse group of invertebrates use surface glycoproteins as contact chemical recognition signal. For example, sea urchin and bivalve gamete recognition depends on glycoproteins (Glabe and Clark 1991; Lopez et al. 1993; Focarelli and Rosati 1995). A marine rotifer (Snell et al. 1995) and some copepods (Snell and Carmona 1994; Kelly and Snell 1998; Ting et al. 2000) use glycoproteins as mate recognition signal. Copepods are the only group of crustaceans reported to use glycoproteins as contact cue for mate recognition (Snell and Carmona 1994; Kelly and Snell 1998; Ting et al. 2000). Results from the studies on copepods indicate that lectins bind to surface glycoproteins on females and obscure male mate-recognition (Lonsdale et al. 1996) and treating adult males of the marine harpacticoid Tigriopus japonicus with the Triticum vulgaris lectin significantly inhibited mate recognition and guarding behavior (Kelly and Snell 1998). Glycoproteins also exist on the cuticular surface of decapod crustaceans (Marlowe et al. 1994; Shafer et al. 1994), but whether decapod crustaceans also use cuticular glycoproteins as contact cues in mate recognition is unknown.

In this study, we tested whether three Lysmata species use cuticular glycoproteins as contact cues in mate recognition. We expected a similar result to copepods, i.e. lectin-binding would reduce response of MP shrimp to lectin-treated newly molted EP shrimp. If the hypothesis is rejected, newly molted EP shrimp would be treated with ethylenediamine tetraacetate (EDTA) to remove glycoproteins on the cuticle surface, and response of MP shrimp to EDTA-treated EP shrimp would be observed. EDTA has been commonly used to extract glycoproteins (e.g. Andersen 1991; Shafer et al. 1994; Snell and Stelzer 2005).

Materials and methods

Experimental animals

Lysmata boggessi shrimp were collected from Hernando Beach, Florida, U.S.A., L. wurdemanni shrimp were collected at Sebastian Inlet, Sebastian, Florida, U.S.A., and L. ankeri shrimp were originally collected in Haiti. The MP and EP shrimp, between 2.6 and 3.8 cm in total length (TL), were maintained in 20-l tanks with flow-through seawater of 35‰ salinity and 26–28°C temperature, and were fed frozen adult Artemia ad libitum once daily.

Lectin binding and mate recognition

Lectin-binding experiments following the procedures of Kelly and Snell (1998) were performed to determine whether shrimp of the three Lysmata species use glycoproteins as contact chemical cues in mate recognition. Twenty lectins that have binding affinity with glycoproteins on exocuticle of decapod crustaceans (Marlowe et al. 1994) were screened. They belong to four affinity groups: glucose/mannose group, including agglutinin from Pisum sativum (PSA), Lens culinaris (LCA), and concanavalin A (Con A); N-acetyl glucosamine group including lectins from Datura stramonium (DSL), Glycine max (SBA), Lycopersicon esculentum (LEL), Ricinus communis (RCA), Solanum tuberosum (STL), lectin of Vicia villosa (VVL), Triticum vulgaris (WGA), and succinylated wheat germ agglutinin (SWGA); N-acetyl galactosamine/galactose group including lectins of Bandeiraea (Griffonia) simplicifolia (BSL I), Dolichos biflorus (DBA), Erythrina corallodendron (ECL), Jacalin (Jac), peanut agglutinin (Arachis hypogaea; PNA), Sophora japonica (SJA); and L-fucose group including Ulex europaeus (ULEX I); and oligosaccharides group including leucoagglutinin and erythroagglutinin of Phaseolus vulgaris (PHA-L, E). All lectins (lyophilized powder and fluoroisothiocyanate labeled) were purchased from Sigma Chemical Company or Vector Laboratories.

Because contact cues only exist in newly molted EP shrimp (NMEP) (Zhang and Lin 2006), lectin binding intensity by NMEP and intermolt EP (IEP) shrimp should be different if Lysmata shrimp use glycoproteins as their contact cues. Both IEP and NMEP shrimp were tested and compared. Result from the copepods indicates that lectin binding also inhibits male’s performance in mate recognition (Kelly and Snell 1998). Hence, intermolt MP shrimp (IMP) was also treated with lectins to see whether lectin binding affect MP shrimp’s response to NMEP. Three concentrations, 10, 20, and 50 μg ml−1, were pre-tested to determine the efficient binding concentration for the bioassay. Each lectin labeled with fluoroisothiocyanate (FITC) was used. All concentrations were effective, so the intermediate concentration (20 μg ml−1) was used for the binding experiment. Three EP and MP shrimp were tested. For NMEP and IEP shrimp, exocuticle of each shrimp was dissected and cut into 21 pieces. Each was treated with one lectin of 20 μg ml−1 and seawater (control) for 60 min, then rinsed 5 times with seawater for about 3 min. (Because mating occurred within 2 min after NMEP shrimp were placed into the bucket containing IMP shrimp, see the Results). Previous result indicated that 30-min rinsing did not affect lectin binding (Marlowe et al. 1994). For IMP shrimp, only the antennae/antennules were tested, because IMP shrimp rely on two simple seta on antennae/antennules to detect the contact cues (Zhang and Lin 2006; Zhang et al. 2008). The treated exocuticles and the two simple setae on the antennae/antennules were observed under epifluorescent microscope. Microscope images were digitally captured, and the fluorescence can easily be quantified with the free program ImageJ. The binding affinities of the lectins were determined based on the difference in fluorescent readings between lectin-treated and control exocutiles or antennae/antennules. Lectin binding was graded based on relative staining intensities. Intense staining was noted as +++, moderate staining as ++, and weak staining as +. No mark indicates that the level of fluorescence was not different from that of the controls.

For bioassays, both MP and EP shrimp were treated with the lectins. Five to seven hours before EP shrimp molted (EP shrimp molt 12–24 h after larval hatching), they were moved from the maintenance tank and placed in a separate tank. The shrimp were checked every 30–60 min to determine the molting status. NMEP shrimp were exposed to 50 ml individual lectins at 20 μg ml−1 for 60 min. The experiments were performed with 10 MP and 10 EP shrimp of each species for each lectin. Control shrimp were exposed to seawater. The shrimp were re-used for different lectins test after at least one molting cycle (7–11 days).

Lectin-treated or non-lectin-treated EP shrimp were placed into 20-l buckets containing 10 l seawater with one lectin-incubated or with non-lectin-incubated MP shrimp to examine the response of MP shrimp to EP shrimp. MP shrimp (2.8 ± 0.2 cm, mean ± SD) were <0.5 cm smaller in total length than EP shrimp (3.1 ± 0.1 cm, mean ± SD), therefore would not affect mating success (Zhang and Lin 2005a). Environmental conditions (35‰ salinity and 26–28°C temperature) were the same as in the maintenance tanks. For control (non-lectin-treated MP and EP shrimp), MP shrimp would immediately grasp the EP shrimp and copulate with her after the MP shrimp had detected the EP shrimp with their antennae/antennules (Zhang and Lin 2005b, 2006). If glycoproteins were the contact cues, MP shrimp should not respond to lectin-treated EP shrimps or at least show a diminished response.

EDTA treatment and mate recognition

Because results from lectin binding experiment rejected the hypothesis that glycoproteins are contact cues of Lysmata shrimps (see Results), one of the three shrimp species, L. boggessi, was used to validate the lectin binding results. Newly molted EP shrimp were treated with 100 ml of 0.1 M EDTA at pH 7.5–8.0 for 30 min. EP shrimp were then rinsed with fresh EDTA solution, followed with seawater three times. Treated EP shrimp was placed into a 20-l bucket (containing 10 l seawater) containing 1 MP shrimp. Response of MP shrimp to EDTA-treated EP shrimp and control (non-EDTA treated newly molted EP shrimp) was observed and compared. Ten replicates were conducted.

Results

The patterns of binding intensity were the same among the three species (Table 1). All the lectins tested bound to the exocuticle of NMEP shrimp of all three species. The greatest binding was observed in the three lectins, Con A, Jac, and VVL. Most of these lectins bound only moderately or weakly to the exocuticle and two simple setae on antenna/antennules in all three species (Table 1).

Table 1 Lysmata ankeri, Lysmata boggessi, Lysmata wurdemanni
Full size table

Lectin treatment did not affect copulation success in the three Lysmata species (Fisher’s exact test, P = 1.000), as all of the 10 MP shrimp tested in each species copulated successfully with lectin-treated and with control EP shrimp in each of the three treatments (non-lectin-treated MP and lectin-treated EP shrimp, lectin-treated MP and lectin-treated EP shrimp, and lectin-treated MP and non-lectin-treated EP shrimp). Responses of the MP shrimp in the three treatments and the control to lectin-treated and control EP shrimp were also not different for each species. When NMEP shrimp were placed into the bucket containing MP shrimp, the MP shrimp grasped the EP shrimp after detecting them with antennae/antennules within 2 min, and copulated with them regardless of the lectin treatment. As such all the 10 MP shrimp copulated with all 20 different lectin-treated and control EP shrimp successfully in each species.

EDTA treatment did not affect copulation success in L. bogessi (Fisher’s exact test, P = 1.000). All 10 L. boggessi MP shrimp copulated with EDTA-treated newly molted shrimp and control newly molted shrimp immediately after they detected the EP shrimp with antennae/antennules. Interestingly, MP shrimp still copulated with six of the 10 EP shrimp killed by the EDTA.

Discussion

Since surface glycoproteins were discovered to function as mating pheromones in single-cell aquatic organisms such as algae (Wiese 1965) and ciliates (Miyake and Beyer 1974), these molecules have also been found in a rotifer (Snell et al. 1995) and in copepods to play a role in mate recognition (Snell and Carmona 1994; Kelly and Snell 1998; Ting et al. 2000). Although a number of aquatic organisms use surface glycoproteins as contact chemical recognition signals, lectin-binding experiments in the present study indicate that they are not contact pheromones for the Lysmata shrimps because MP shrimp of the three Lysmata species could recognize lectin-treated EP shrimp and copulated with them. EDTA treatment test further validated the results of lectin binding experiment.

In the harpacticoid copepod Coullana canadensis, mate guarding was reduced by binding lectins to female surface glycoproteins (Lonsdale et al. 1996). Glycoproteins also exist on cuticle surface of decapod crustaceans (Marlowe et al. 1994); however, binding lectins to surface glycoproteins of NMEP shrimp did not reduce responses of MP shrimp to the treated EP shrimp and did not affect copulation success. This suggests that glycoproteins are not likely involved in mate recognition in the Lysmata shrimps. On the other hand, lectins may also lower sensitivity of male copepods’ chemoreceptor, so that male mate-guarding is significantly decreased (Kelly and Snell 1998). Although lectins also bound on two simple setae on antennae/antennules, which are probably the detectors of contact cues in Lysmata shrimps, male’s responses to NMEP were not reduced.

EDTA has been used in extracting glycoproteins from cuticles of crustaceans (e.g. Andersen 1991; Shafer et al. 1994). Removal of glycoproteins in female rotifers with EDTA incubation significantly reduced mating responses from males (Snell and Stelzer 2005). No difference in copulation response from MP shrimp to EDTA-treated and control EP shrimp in the present study further validates that glycoproteins are not likely involved in mate recognition in the Lysmata shrimps.

Chemical components of sexual pheromone bouquets often exist in one sex and are absent in the other (e.g. Wabnitz et al. 2000; Shine et al. 2002; Zhang et al. 2003), most prominently shown in sex specific attractants in insects but also in crustaceans. For Lysmata shrimps, there maybe different contact cues for different molt stages because MP shrimp only mate with NMEP shrimp. Active components should exist on the exocuticle of the newly molted animals, and absent or in very low concentrations in the other stages of the molt cycle. Studies have shown that the surface glycoprotein composition of decapod crustaceans changes during the molt cycle (Marlowe et al. 1994; Shafer et al. 1994), but lectin binding results indicate that glycoprotein concentrations on the exocuticle of newly molted decapod crustaceans are not different from those in other molt stages (Marlowe et al. 1994). Our study shows that lectin binding intensities were similar between NMEP and IEP (Table 1), further suggesting that glycoproteins are not likely involved in mate recognition in the Lysmata shrimps.

The reason Lysmata shrimps do not use glycoproteins as contact chemical cues may be the insufficient variation in glycoprotein composition on the exocuticle at different molt stages. Contact pheromones have been intensively studied in insects and have been identified as cuticular hydrocarbons, not glycoproteins (e.g. Linn and Roelofs 1995; Ginzel et al. 2003; Howard et al. 2003; Zhang et al. 2003; Sugeno et al. 2006). Future studies on Lysmata shrimps will examine whether their contact pheromones are hydrocarbons and attempt to chemically characterize the contact sex pheromones.