Seasonal Changes in the Sexual Size Dimorphisms of the Chelipeds and Pleons of the Porcellanid Crab Petrolisthes japonicus

© 2021 Academia Sinica, Taiwan Open Access Seasonal Changes in the Sexual Size Dimorphisms of the Chelipeds and Pleons of the Porcellanid Crab Petrolisthes japonicus Katsuyuki Hamasaki* and Shigeki Dan Department of Marine Biosciences, Tokyo University of Marine Science and Technology, Konan, Minato, Tokyo 108-8477, Japan. *Correspondence: E-mail: [email protected] (Hamasaki) E-mail: [email protected] (Dan) Received 5 August 2020 / Accepted 5 February 2021 / Published 18 April 2021 Communicated by Benny K.K. Chan Decapod crustaceans develop secondary sexual structures, such as large chelipeds for weaponry and/ or display in males and a wider pleon for incubating eggs in females. These structures should be costly to produce and maintain; hence, the crustacean reproductive strategy might have adapted to conserve energy for developing secondary sexual characteristics during the reproductive season. However, little is known about seasonal changes in the sexual size dimorphisms of the chelipeds and pleons of decapod crustaceans. Anomuran crabs from the family Porcellanidae exhibit a true crab-like body shape, with relatively large chelipeds for their body size and a well-developed pleon. They also show sexual size dimorphism (i.e., males with larger chelipeds and females with wider pleons); therefore, it was expected that seasonal changes in the sexual size dimorphisms of these organs would occur in porcellanid crabs. We tested this expectation using Petrolisthes japonicus, which is common on the intertidal cobble and boulder shores of the temperate and tropical regions of Japan. We employed allometric growth analyses of the cheliped and pleon dimensions of P. japonicus females and males collected over a year. Our analyses supported our expectation regarding the seasonal changes in the sexual size dimorphisms of the chelipeds and pleons in P. japonicus. Intersexual size dimorphism of the chelipeds and pleons was evident during the breeding season: females allocated more energy to pleon growth, while reducing cheliped growth, and investment in the pleon increased with increasing female body size, whereas males invested their energy in cheliped growth, and the investment in chelipeds increased with increasing male body size. During the nonbreeding season, females and males conserved energy by reducing the size of pleons and chelipeds relative to their respective body sizes. Our results highlighted the sex-specific and seasondependent resource allocation and reproductive strategies of porcellanid crabs. Key words: Allometric growth, Life history, Reproductive strategy, Secondary sexual traits, Sexual maturity. BACKGROUND Decapod crustaceans show sexual dimorphisms in size or morphological traits (Davie et al. 2015); for example, in brachyuran crabs, females have a broad pleon for incubating eggs, which are attached to the second to fifth pleopods on the ventral pleomeres, whereas males have a relatively narrow pleon that covers the first and second pleopods, which are modified copulatory organs (i.e., gonopods) (Davie et al. 2015). Some decapods have enlarged chelipeds that are used to manipulate food organisms and weaponry in predator avoidance and contest competition (Lee 1995; Davie et al. 2015; Lavalli and Spanier 2015). Although decapod Citation: Hamasaki K, Dan S. 2021. Seasonal changes in the sexual size dimorphisms of the chelipeds and pleons of the porcellanid crab Petrolisthes japonicus. Zool Stud 60:18. doi:10.6620/ZS.2021.60-18. Zoological Studies 60:18 (2021) doi:10.6620/ZS.2021.60-18 1 © 2021 Academia Sinica, Taiwan chelipeds principally exhibit similar functions in both sexes, the chelipeds of males are generally larger than those of females (Crane 1975; Lee 1995; Mariappan et al. 2000; Emlen 2008; Hamasaki et al. 2020a). Sexual dimorphism in decapod cheliped size is thought to be a consequence of sexual selection (Baeza and Asorey 2012; Cothran 2020): males win the competition for females by having larger chelipeds; for instance, in male fiddler crabs, the cheliped greatly enlarges and functions as both a weapon to fight competitive males and as an ornament to court receptive females through claw waving display (Crane 1975; Christy 1982 1983; Pope 2000). The males of some brachyuran crabs have larger chelipeds, which they use to protect receptive females from other males during preand postcopulatory mate-guarding behaviour, and embrace females from behind with their walking legs (van den Brink and McLay 2010; Soundarapandian et al. 2013; Waiho et al. 2015). The males of some pagurid hermit crabs exhibit precopulatory guarding behaviour by grasping the aperture of the gastropod shell occupied by sexually mature females with their minor (i.e., left) cheliped, while guarding males frequently use their major (i.e., right) cheliped to defend females against intruders (Hazlett 1968 1972; Yasuda et al. 2011 2012). The selected sexual structures of decapod crustaceans should be costly to produce and maintain (Allen and Levinton 2007; Wilson et al. 2009; Doake et al. 2010); hence, their reproductive strategy might have adapted to conserve energy for developing secondary sexual characteristics during a reproductive season, and thus sexual dimorphism may change seasonally in these organisms. Secondary sexual characteristics become evident after pubertal moult in decapod crustacean species (Hartnoll 1974 1978); therefore, the growth of secondary sexual characteristics, such as chelipeds and pleons, relative to a reference body dimension was measured to elucidate their size at the onset of sexual maturity in numerous decapod crustaceans (McLay 2015), and these analyses highlighted the sexual size dimorphisms of the chelipeds and pleons. Only a few studies have documented seasonal changes in the sexual dimorphism of decapod crustaceans: adult females reduced pleon growth after the breeding season in the shore crab Pachygrapsus transversus Gibbes 1850 (Flores and Negreiros-Fransozo 1999), and adult males increased cheliped growth during the reproductive season in the anomuran crab Aegla franca Schmitt 1942 (Bueno and Shimizu 2009) and hermit crabs Diogenes nitidimanus Terao 1913 (Koga et al. 2010) and Pagurus minutus Hess 1865 (Yasuda et al. 2017). Anomuran crabs, which belong to the family Porcellanidae Haworth 1825, exhibit a brachyuran crablike body shape; broad and dorsoventrally flattened, relatively large chelipeds for their body size; and a well-developed, symmetrical pleon that is carried bent under the cephalothorax and held against the thorax (Jones 1977; Barría and González 2008; Osawa and McLaughlin 2010). Sexual size dimorphisms of the chelipeds and pleons have been documented in some porcellanid crab species (i.e., males with larger chelipeds and females with wider pleons) (Miranda and Mantelatto 2010; Baeza and Asorey 2012; Wassick et al. 2017); it is therefore likely that the relatively large chelipeds and well-developed pleons of porcellanid crabs are costly to produce and maintain, and that seasonal changes in sexual size dimorphism occur in their chelipeds and pleons. In the present study, we aimed to test this expectation of seasonal changes in the sexual size dimorphisms of the chelipeds and pleons of porcellanid crabs using Petrolisthes japonicus (De Haan 1849). This crab is a free-living species, common on the intertidal cobble and boulder shores of the temperate and tropical regions of Japan (Miyake 1998). We examined the sexual size dimorphisms of the chelipeds and pleons and their seasonal changes using allometric growth analyses of P. japonicus specimens collected over one year on an intertidal cobble and boulder shore on the Boso Peninsula, Japan. MATERIALS AND METHODS Measurement of crabs We collected 1,259 P. japonicus specimens through monthly sampling, conducted from April 2008 to March 2009 in an intertidal cobble and boulder zone (34°58'N, 139°46'E) on the Boso Peninsula, Japan (Table 1). These crab specimens were used in our previous study to investigate the reproductive traits and population structure of a P. japonicus population (Hamasaki et al. 2020b). The carapace width (CW; the widest measurement between the lateral margins of the carapace) of the crabs was measured using a digital calliper (CW > ~3 mm, accuracy 0.05 mm) or a stereomicroscope with the aid of a micrometre (CW < ~3 mm, accuracy 0.01 mm). The crabs with > 2.3 mm CW were sexed based on the position of the gonopores (i.e., at the coxae of the third pereiopods in females and at the coxae of the fifth pereiopods in males) (Osawa and McLaughlin 2010) because the minimum body size of females has been found to be 2.3 mm CW (Nakasone 1972; Hamasaki et al. 2020b). Each female was then classified according to the presence or absence of eggs attached to the pleopods (ovigerous or non-ovigerous females, respectively). page 2 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan Our crab specimens comprised 568 females (including 156 ovigerous females), 585 males and 106 unsexed juveniles (Table 1). Porcellanid crabs easily autotomize their first pereiopods (i.e., chelipeds) due to physical stress (Wasson et al. 2002). In our crab specimens, 112 females, 126 males (including three males with noticeably small regenerated chelipeds) and 35 unsexed juveniles had lost one or both chelipeds naturally and/or due to physical stress during the sampling procedure (Table 1). We therefore measured several body parts of intact specimens (456 females, 459 males and 71 unsexed juveniles; Table 1) using a digital calliper or a stereomicroscope as follows: propodus length (from the tip of the propodus to the carpal joint) and width (the widest part of the propodus) of the right and left chelipeds, and pleon width (the widest part of the fifth pleonite). Heterochely is known in some porcellanid crab species (Baeza and Asorey 2012), and one of the chelipeds was larger than the second in the pair of P. japonicus in the present study, as shown in figure S1; therefore, we selected the largest measurement between the right and left chelipeds as the cheliped dimension. Sexual size dimorphisms of chelipeds and pleons In the present study, the statistical analyses were performed with the statistical software R (R4.0.2; R Core Team 2020) at a 5% significance level. The sexual size dimorphisms of the chelipeds and pleons were evaluated based on allometric growth analyses. The growth of the body dimensions relative to a reference dimension (CW) was examined using the following allometric growth equation (Huxley 1932): y = axb, where x is CW, y is the measurement for another body part (cheliped propodus length and width, and pleon width), b is the allometric growth coefficient and a is the initial growth constant. The relative growth patterns were defined as follows: b > 1 indicated positive allometric growth, or faster growth of y than x; b = 1 indicated isometric growth, or the same growth rate for y and x; and b < 1 indicated negative allometric growth, or slower growth of y than x. The coefficients and standard errors were estimated by applying a general linear model to the log-transformed equation lny = lna + blnx using the lm function, and the statistical significance of the coefficient estimates and the linear model were evaluated with t-tests and F-tests, respectively. To statistically infer the intersexual dimorphisms as the differences between females and males in the allometric growth coefficients of the cheliped and pleon sizes, using overall specimens, we compared two models that included a continuous explanatory variable (CW) and a categorical explanatory variable (Sex: female or male) in the allometric growth analyses, as follows: Model 1, lny ~ lnCW + lnCW × Sex; and Model 2, lny ~ lnCW, where y is the cheliped or pleon size; we then selected the model with the lower AIC value (Akaike 1973). The models showed whether the allometric growth equation had the same intercept and different slopes (Model 1) or the same intercept and slope (Model 2) between the different sexes. When Model 1 was selected, intersexual dimorphism was detected. Table 1. Collection records for porcellanid crab Petrolisthes japonicus specimens, collected from April 2008 to March 2009 on an intertidal cobble and boulder shore on the Boso Peninsula, Japan Sampling date Total number of crabs Number of females Number of ovigerous females Number of lost cheliped females Number of intact females Number of males Number of lost cheliped males Number of intact males Number of unsexed juveniles Number of lost cheliped juveniles Number of intact juveniles April 21 86 40 0 4 36 46 6 40 0 0 0 May 19 190 86 6 19 67 104 25 79 0 0 0 June 19 142 74 59 11 63 68 15 53 0 0 0 July 18 120 61 52 15 46 57 7 50 2 0 2 August 19 107 41 29 10 31 32 14 18 34 16 18 September 16 125 47 9 4 43 44 2 42 34 9 25 October 16 73 15 0 6 9 27 8 19 31 10 21 November 26 25 12 0 5 7 10 4 6 3 0 3 December 14 104 39 0 9 30 64 15 49 1 0 1 January 17 111 60 0 10 50 50 8 42 1 0 1 February 14 91 49 1 14 35 42 9 33 0 0 0 March 16 85 44 0 5 39 41 13 28 0 0 0 Total 1259 568 156 112 456 585 126 459 106 35 71 page 3 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan To evaluate the monthly changes in intersexual size dimorphism, allometric growth Model 1 was applied using monthly data from April 2008 to March 2009. In this analysis, the data from October and November were pooled due to the small sample size in November (Table 1). Additionally, to detect the intrasexual dimorphism during the breeding and nonbreeding seasons, the allometric growth analyses were conducted separately for females and males based on the following model: lny ~ lnCW + lnCW × Season, where y is the cheliped or pleon size and Season is the categorical explanatory variable (breeding season or non-breeding season). In our sample population of P. japonicus, ovigerous females were found from May to September and in February, but only one was detected in February (Table 1); therefore, the breeding and nonbreeding seasons were defined as the period from May to September and the period from December to April, respectively. RESULTS The relationships among CW, cheliped propodus length and width, and pleon width are shown for the overall specimens of females, males and unsexed juveniles in figure 1. The data plots for unsexed juveniles were placed on the same line as those for smaller females and males. Of the two log-transformed allometric growth equation models applied to the cheliped and pleon dimensions of females and males, Model 1 was selected (Table 2), indicating the existence of intersexual size dimorphisms of the chelipeds and pleons of P. japonicus. The chelipeds had a larger Table 2. Intersexual size dimorphism evaluated using a log-transformed allometric growth equation model for some body parts of the porcellanid crab Petrolisthes japonicus. Two models were applied: Model 1, lny ~ lnCW + lnCW × Sex and Model 2, lny ~ lnCW, where CW is the carapace width, y is the measurement for another body part (cheliped propodus length and width, and pleon width) and Sex is female or male. The coefficient estimate with a standard error (SE) for each explanatory variable, including the Sex, was the output for males, representing changes in the response variable relative to the baseline category (female). The statistical significance of the coefficient estimates and each model was evaluated with t-tests and F-tests, respectively. Response variable Coefficients and statistics Model 1 Model 2 Estimate SE t value PEstimate SE t value P Chela propodus length Intercept 0.0292 0.0091 3.205 0.0014 0.0024 0.0113 0.212 0.8320 InCW 1.1966 0.0057 209.346 < 0.0001 1.2398 0.0068 183.356 < 0.0001 lnCW × Sex-Male 0.0522 0.0023 22.856 < 0.0001 AIC -2628 -2216 R20.9832 0.9736 F266670 33620 df 2, 912 1, 913 P< 0.0001 < 0.0001 Chela propodus width Intercept -0.8382 0.0105 -79.690 < 0.0001 -0.8589 0.0117 -73.480 < 0.0001 InCW 1.1794 0.0066 178.980 < 0.0001 1.2127 0.0070 174.010 < 0.0001 lnCW × Sex-Male 0.0403 0.0026 15.320 < 0.0001 AIC -2368 -2161 R20.9767 0.9707 F19130 30280 df 2, 912 1, 913 P< 0.0001 < 0.0001 Pleon width Intercept -0.4941 0.0089 -55.610 < 0.0001 -0.8589 0.0117 -73.480 < 0.0001 InCW 1.1666 0.0056 209.560 < 0.0001 1.2127 0.0070 174.010 < 0.0001 lnCW × Sex-Male -0.0921 0.0022 -41.410 < 0.0001 AIC -2677 -2161 R20.9800 0.9707 F22390 30280 df 2, 912 1, 913 P< 0.0001 < 0.0001 page 4 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan allometric growth coefficient estimate in males (propodus length, b = 1.25; propodus width, b = 1.22) than in females (propodus length, b = 1.20; propodus width, b = 1.18), whereas the pleon width had a larger allometric growth coefficient estimate in females (b = 1.17) than in males (b = 1.07; Table 2). Considering the relationships among CW, the cheliped propodus length and width, and the pleon width of females and males on a monthly basis (Figs. 2–4), the intersexual size dimorphisms of the chelipeds and pleons appeared to be distinct during the breeding season from May to August, regardless of the reproductive status of the females (non-ovigerous or ovigerous). The seasonal changes in intersexual size dimorphism could be evaluated quantitatively by calculating the monthly coefficient estimates for the explanatory variable, including the Sex (i.e., lnCW × Sex), which represented the degree of difference between the allometric growth coefficients of males and females (Fig. 5; Tables S1–S3). The coefficient Fig. 1. Growth of the cheliped propodus length (A), cheliped propodus width (B) and pleon width (C) relative to the carapace width in overall specimens of the porcellanid crab Petrolisthes japonicus, collected from April 2008 to March 2009. Data are shown for females, males and unsexed juveniles. 0 5 10 15 20 25 024 6 8 10 12 Propodus length (mm) Carapace width (mm) (A) Males Females Unsexed juveniles 0 2 4 6 8 10 02468 10 12 Propudus width (mm) Carapace width (mm) (B) 0 2 4 6 8 10 02468 10 12 Pleon width (mm) Carapace width (mm) (C) page 5 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan Fig. 2. Growth in the cheliped propodus length relative to the carapace width in specimens of the porcellanid crab Petrolisthes japonicus, collected monthly from April 2008 (A) to March 2009 (K). Data are shown for females, ovigerous females and males. The data for October and November were pooled because of the small sample size in November. Males Females Ovigerous females 0 5 10 15 20 25 0246 8 10 12 (G) October-November 0 5 10 15 20 25 0246 8 10 12 (H) December 0 5 10 15 20 25 0246 8 10 12 (I) January 0 5 10 15 20 25 0246 8 10 12 (J) February 0 5 10 15 20 25 0246 8 10 12 (K) March 0 5 10 15 20 25 0246 8 10 12 Chela length (mm) (A) April 0 5 10 15 20 25 0 5 10 15 Chela length (mm) (B) May 0 5 10 15 20 25 0246 8 10 12 Chela length (mm) (C) June 0 5 10 15 20 25 0246 8 10 12 Chela length (mm) (D) July 0 5 10 15 20 25 0246 8 10 12 Chela length (mm) (E) August 0 5 10 15 20 25 0246 8 10 12 Chela length (mm) (F) September Propodus length (mm) Carapace width (mm) page 6 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan Fig. 3. Growth of the cheliped propodus width relative to the carapace width in specimens of the porcellanid crab Petrolisthes japonicus, collected monthly from April 2008 (A) to March 2009 (K). Data are shown for females, ovigerous females and males. The data for October and November were pooled because of the small sample size in November. Males Females Ovigerous females 0 2 4 6 8 10 024 6 8 10 12 (G) October-November 0 2 4 6 8 10 024 6 8 10 12 (H) December 0 2 4 6 8 10 024 6 8 10 12 (I) January 0 2 4 6 8 10 024 6 8 10 12 Carapace width (mm) (J) February 0 2 4 6 8 10 024 6 8 10 12 Carapace width (mm) (K) March 0 2 4 6 8 10 02468 10 12 Chela width (mm) (A) April 0 2 4 6 8 10 02468 10 12 Chela width (mm) (B) May 0 2 4 6 8 10 02468 10 12 Chela width (mm) (C) June 0 2 4 6 8 10 02468 10 12 Chela width (mm) Carapace width (mm) (D) July 0 2 4 6 8 10 02468 10 12 Chela width (mm) (E) August 0 2 4 6 8 10 02468 10 12 Chela width (mm) (F) September Propodus width (mm) Carapace width (mm) page 7 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan Fig. 4. Growth in the pleon width relative to the carapace width in specimens of the porcellanid crab Petrolisthes japonicus, collected monthly from April 2008 (A) to March 2009 (K). Data are shown for females, ovigerous females and males. The data for October and November were pooled because of the small sample size in November. Males Females Ovigerous females 0 2 4 6 8 10 02468 10 12 (G) October-November 0 2 4 6 8 10 02468 10 12 (H) December 0 2 4 6 8 10 02468 10 12 (I) January 0 2 4 6 8 10 02468 10 12 (J) February 0 2 4 6 8 10 02468 10 12 (K) March 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (A) April 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (B) May 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (C) June 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (D) July 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (E) August 0 2 4 6 8 10 02468 10 12 Pleon width (mm) (F) September Pleon width (mm) Carapace width (mm) page 8 of 13Zoological Studies 60:18 (2021) © 2021 Academia Sinica, Taiwan estimates for the cheliped dimensions increased starting in April and peaked between June and July; they then declined, and the significant differences between the sexes in these coefficient estimates generally disappeared between December and March. The coefficient estimates of pleon width fluctuated similarly to those of the cheliped dimensions, but the differences between the sexes were still evident during the period from December to March. The intrasexual size dimorphisms of the chelipeds and pleons of females and males during the breeding (May to September) and non-breeding seasons (October to April) could be evaluated quantitatively by calculating the estimates of the explanatory variable, including the Season (i.e., lnCW × Season), which represented the degree of difference in the allometric growth coefficient between the breeding season and the non-breeding season (Fig. 6; Table S4). The coefficient Fig. 5. Monthly changes in the intersexual size dimorphisms of the chelipeds and pleons of the porcellanid crab Petrolisthes japonicus specimens, collected from April 2008 to March 2009. Intersexual size dimorphism was represented by the coefficient estimates for the explanatory variable, including the Sex (lnCW × Sex), in the log-transformed allometric growth equation model: lny ~ lnCW + lnCW × Sex, where CW is carapace width, y is the measurement for another body part (cheliped propodus length and width, and pleon width) and Sex is female or male. The coefficient estimate of lnCW × Sex was the output for males, representing changes in the response variable relative to the baseline category (female) (Tables S1–S3), and thus also representing the degree of difference in the allometric growth coefficient between males and females. Vertical lines indicate standard errors. The dotted bars indicates the estimates that were not significantly different between sexes. The data for October and November were pooled because of the small sample size in November. Fig. 6. Intrasexual size dimorphisms of the chelipeds and pleons during the breeding and non-breeding seasons of the porcellanid crab Petrolisthes japonicus, collected from April 2008 to March 2009. Intrasexual size dimorphism was represented by the coefficient estimate of the explanatory variable, including the Season (lnCW × Season), in the log-transformed allometric growth equation model: lny ~ lnCW + lnCW × Season, where CW is carapace width, y is the measurement for another body part (cheliped propodus length (PrL) and width (PrW), and pleon width (PlW)), and Season is the breeding season (May to September) or non-breeding season (October to April). The coefficient estimate of lnCW × Season was the output for the breeding season, representing the change in the response variable relative to the baseline category (non-breeding season) (Table S4), and thus also representing the degree of difference in the allometric growth coefficient between the breeding season and the non-breeding season. Vertical lines indicate standard errors. The dotted bar indicates an estimate that was not significantly different between the breeding season and non-breeding season. -0.15 -0.10 -0.05 0.00 0.05 0.10 Coefficient estimate Ppropodus length Propodus width Pleon width -0.04 -0.03 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 PrL PrW PlW PrL PrW PlW elaMelameF Coefficient estimate page 9 of 13 Zoological Studies 60:18 (2021)