Introduction
Patellid limpets are among the most successful marine gastropods inhabiting rocky shores (Branch, Reference Branch1981) from the supratidal to the subtidal zones and as such, are subject to the most variable and unpredictable environmental conditions in nature (Harley et al., Reference Harley, Denny, Mach and Miller2009). The sharp environmental gradients of wave action, temperature variation and desiccation can prompt significant changes to the overall morpho-physiological characteristics of patellid limpets (Davies, Reference Davies1969). For instance, the reproductive biology and phenology of limpets are influenced by factors such as temperature and storminess (Branch, Reference Branch1974; Bowman & Lewis, Reference Bowman and Lewis1986), latitude and temperature (Lewis, Reference Lewis1986), climate change (Moore et al., Reference Moore, Thompson and Hawkins2011), parasitism (Firth et al., Reference Firth, Grant, Crowe, Ellis, Wiler, Convery and O'Connor2017) and anthropogenic pressures such as harvesting, pollution and habitat loss (Boukhicha et al., Reference Boukhicha, Kalthoum, Hassine and Tlig-Zouari2013; Henriques et al., Reference Henriques, Delgado, Sousa and Ray2017).
Hermaphroditism is regarded as an evolutionarily advantageous strategy supposed to increase the likelihood of successful fertilization for sedentary broadcast spawning species such as limpets, especially when occurring in low population densities (Branch, Reference Branch1981; Guallart et al., Reference Guallart, Calvo, Acevedo and Templado2013). Hermaphroditism may be simultaneous, with one individual releasing both types of gametes during the same season, or sequential, with one individual functioning first as one sex and then changing to another later in life (Orton, Reference Orton1928; Branch, Reference Branch1974; Ghiselin, Reference Ghiselin, Giese, Pearse and Pearse1987). Sequential hermaphroditism is designated as protandry when an individual is first a male and then changes to female (Branch, Reference Branch1981). Protogyny is when an individual first matures as female and then changes to male. In some limpet species the sex change can be reversed (Le Quesne & Hawkins, Reference Le Quesne and Hawkins2006; Rivera-Ingraham et al., Reference Rivera-Ingraham, Espinosa and García-Gómez2011), whereby males that became females can become males again. This could be linked to their reproductive fitness, since small females produce fewer oocytes and they would have a greater reproductive fitness as males (Guallart et al., Reference Guallart, Calvo, Acevedo and Templado2013). When sex change occurs repeatedly it is known as alternating sexuality (Hoagland, Reference Hoagland1984). Sex can also be determined by environmental circumstances where one individual can influence the sex of another (Heller, Reference Heller1993). Sex change in limpets has been shown to be density dependent by experimental manipulation in Lottia gigantea Gray in G. B. Sowerby I, 1834 (Fenberg & Roy, Reference Fenberg and Roy2008) and in Patella vulgata Linnaeus, 1758 (Borges et al., Reference Borges, Doncaster, Maclean and Hawkins2015). Earlier onset of switching to females has been shown in heavily exploited populations of Patella aspera Röding, 1798 in the Macaronesian Islands (Martins et al., Reference Martins, Borges, Vale, Ribeiro, Ferraz, Martins, Santos and Hawkins2017). In some species, sex change is triggered by specific environmental cues, in other species, sex change seems to be determined genetically, occurring mainly after the first reproductive season (Wright, Reference Wright1989; Fretter et al., Reference Fretter, Graham, Ponder, Lindberg, Beesley, Ross and Wells1998).
Hermaphroditism has been described in several families of limpets, for example Nacellidae (Mau et al., Reference Mau, Fox and Bingham2017), Siphonariidae (Pal & Hodgson, Reference Pal and Hodgson2005) and Patellidae (Orton, Reference Orton1928). This phenomenon is prevalent in some genera of the Patellidae, Cymbula (Branch, Reference Branch1981), Scutellastra (Branch, Reference Branch1974; Lindberg, Reference Lindberg2007), Helcion (Henninger & Hodgson, Reference Henninger and Hodgson2001), and especially in the genus Patella. Patella vulgata (Orton et al., Reference Orton, Southward and Dodd1956; Le Quesne & Hawkins, Reference Le Quesne and Hawkins2006), Patella ulyssiponensis Gmelin, 1791 (Thompson, Reference Thompson1979), Patella caerulea Linnaeus, 1758 (Montalenti, Reference Montalenti and Buzzati-Travesso1958), Patella aspera (Martins et al., Reference Martins, Borges, Vale, Ribeiro, Ferraz, Martins, Santos and Hawkins2017), Patella ferruginea Gmelin, 1791 (Espinosa et al., Reference Espinosa, Guerra-García, Fa and García-Gómez2009) and Patella rustica Linnaeus, 1758 (Prusina et al., Reference Prusina, Ezgeta-balic, Ljubimir, Dobroslavic and Glamuzina2014) are known to be protandrous hermaphrodites (Dodd, Reference Dodd1956; Le Quesne & Hawkins, Reference Le Quesne and Hawkins2006; Rivera-Ingraham et al., Reference Rivera-Ingraham, Espinosa and García-Gómez2011; Martins et al., Reference Martins, Borges, Vale, Ribeiro, Ferraz, Martins, Santos and Hawkins2017). Simultaneous hermaphroditism has been reported for the Azorean endemic limpet Patella candei gomesii (Cunha et al., Reference Cunha, Martins, Amaral and Rodrigues2007). Protandry does not seem to occur in populations of Patella depressa Pennant, 1777 in southern England (Orton & Southward, Reference Orton and Southward1961; Guerra & Gaudêncio, Reference Guerra and Gaudêncio1986), however, more recently some indications of protandry in populations from Portugal have been reported (Borges et al., Reference Borges, Doncaster, Maclean and Hawkins2015).
Hermaphroditism in patellid limpets is an important trait that needs to be taken into consideration in the conservation and management of heavily exploited populations (Espinosa et al., Reference Espinosa, Guerra-García, Fa and García-Gómez2006). It is known that size-selective harvesting negatively impacts the reproductive output threatening species survival (Fenberg & Roy, Reference Fenberg and Roy2008). In fact, larger specimens are more visible and valuable thus more prone to being harvested and at the same time contribute the most to the reproductive effort, since in these species the reproductive output is directly related to size, with larger individuals with larger gonads producing more gametes (Creese, Reference Creese1980). Removal of these individuals will therefore diminish the reproductive success of exploited populations (Espinosa et al., Reference Espinosa, Guerra-García, Fa and García-Gómez2009; Henriques et al., Reference Henriques, Delgado, Sousa and Ray2017). In hermaphroditic species, removal of larger individuals might also cause alterations in population dynamics, since the absence of the predominant sex at larger size is considered one of the cues that triggers sex change to occur earlier at smaller sizes (Rivera-Ingraham et al., Reference Rivera-Ingraham, Espinosa and García-Gómez2011).
Patella piperata Gould, 1846 is one of the 15 worldwide recognized species of the genus Patella Linnaeus, 1758 (WoRMS, 2018). This species is endemic to the Macaronesian archipelagos of Madeira, Canaries and Cape Verde (Christiaens, Reference Christiaens1973) and occurs together with P. aspera and Patella candei d'Orbigny, 1840 in the Madeira archipelago (Sousa et al., Reference Sousa, Delgado, Pinto and Henriques2017). It is closely related to P. rustica, a sister species occurring on continental Atlantic coasts and throughout the Mediterranean (Côrte-Real et al., Reference Côrte-Real, Hawkins and Thorpe1996; Sá-Pinto et al., Reference Sá-Pinto, Branco, Harris and Alexandrino2005). This important microphagous grazer controls algal biomass of the intertidal rocky shores (Bergasa et al., Reference Bergasa, Ramírez, Collado, Hernández, Gelado-Caballero, Haroun and Rodriguez2007) and is collected for human consumption by locals throughout the Canaries (Moro & Herrera, Reference Moro and Herrera2000; Bergasa et al., Reference Bergasa, Ramírez, Collado, Hernández, Gelado-Caballero, Haroun and Rodriguez2007) and occasionally in Madeira. A potentially hermaphroditic individual of P. piperata was macroscopically identified, while studying this species' reproductive biology. This find prompted the need to confirm whether hermaphroditism occurs in this species similar to its protandrous sister clade P. rustica (Prusina et al., Reference Prusina, Ezgeta-balic, Ljubimir, Dobroslavic and Glamuzina2014). Our specific objectives were: (i) to report the histological validation of sex change in P. piperata and (ii) to determine the size at which sex change occurs.
Materials and methods
Monthly samples of P. piperata were collected from the coast of Madeira, North-eastern Atlantic (32°00′–33°30′N 15°30′–18°00′W), at two different sites (São Vicente and Santa Cruz), between 2017 and 2018, as part of a study focussing on growth and reproduction of this species. Specimens were randomly collected from the supra-tidal rocky shore, removing all the specimens found without considering their size, for a period of 30 min. Shell length (L) was measured using a Vernier calliper to the nearest 0.01 mm, and total wet weight (W) to the nearest 0.01 g was measured using an electronic balance. Specimens were dissected for macroscopic observation of the gonads and sexed according to Orton et al. (Reference Orton, Southward and Dodd1956) considering gonad pigmentation, with males exhibiting pale white to pink gonads and females brown to red gonads.
Gonads were removed, damp-dried, weighed to 0.01 g accuracy and preserved in a 10% buffered formaldehyde solution. All measurements were taken from fresh samples.
Histological analysis was performed on dissected gonads in five stages: (I) fixation; (II) dehydration (alcohol with different concentrations: 70, 90 and 95%); (III) embedding in resin; (IV) preparation of trimming blocks and (V) sectioning. A portion of about 1 cm3 from the middle of each gonad was embedded in Technovit 7100 resin, following standard protocols. Histological sections (3–4 µm thickness) were cut and stained with methylene blue solution. The slides of each specimen were sealed with a drop of neo-mount resin, and image acquisition using the visual image analysis system composed of the Leica DMLB microscope coupled to the Leica EC 3 camera and the software LAS 4.5. (Vasconcelos, Reference Vasconcelos2017).
An analysis of variance (ANOVA) was employed to compare limpet size among males, females and hermaphrodites. The size cumulative frequency was analysed using the non-parametric Kolmogorov–Smirnov test and a Pearson χ2 was employed to compare the sexual proportion by size classes between areas. All statistical analyses were performed using SPSS v.24.0 (IBM Corp., Armonk, NY). For all tests, statistical significance was accepted when P < 0.05.
The size at sex-change (L50) corresponds to the size at which 50% of mature individuals are the second sex (male for protogynous, female for protandrous) and was determined using a logistic regression (Allsop & West, Reference Allsop and West2003).
Results
A total of 1164 specimens of P. piperata were sampled between 2017 and 2018. Among the sampled individuals, 53 were macroscopically identified as potential hermaphrodites (Figure 1), representing 5% of the total sample. 65% were females and 30% were males. Hermaphrodites were only found from March to June.
Fig. 1. Hermaphrodite gonad of Patella piperata, where the ovary is represented by the central dark area and the testis is represented by the lighter areas.
The shell length of the hermaphrodites varied between 21.72 and 37.81 mm (
$\bar{x} = \; $ 31.54 ± 4.98 mm), female shell length varied between 11.17 and 37.34 mm (
$\bar{x} = \; $ 25.11 ± 4.56 mm) and male shell length varied between 11.65 and 37.52 mm (
$\bar{x} = \; $ 22.99 ± 5.34 mm). Differences in mean shell length among sexes were statistically significant (F = 94.067, P < 0.001).
The overall sex ratio was 2.16:1 (female:male) favouring females. The distribution of the cumulative frequency of males, females and hermaphrodites per size is represented in Figure 2. The observed differences were statistically significant (Z = −4.037, P < 0.001).
Fig. 2. Cumulative size frequency of males, females and hermaphrodites of Patella piperata.
The Pearson's χ2 showed that sexual proportion between size classes did not differ between sampling sites (χ2 = 2.006, P = 0.157) and that females were predominant between 17.00 and 35.00 mm (χ2 = 33.999, P = 0.005), indicating that the size structure of the populations was similar between localities and that males were more abundant between 11.00 and 16.00 mm and females between 17.00 and 37.00 mm. The size at sex change (L50) was estimated at a shell length of 36.01 mm, and at 29.22 mm 25% of limpets had started to change sex.
The histological validation of the potential hermaphrodites showed that defined sections of ovary and testis occurred simultaneously in specimens of P. piperata between March and June, suggesting that these individuals represent a transitional stage of sequential sex-change, probably occurring in the late stages of the spawning season (Figure 3). Testicular tissue, with some distinguishable seminiferous tubules but without sperm was observed, implying that male structures during transition aren't functional. Ovarian tissue with vitellogenic oocytes and some atresia was also observed, suggesting post-spawning degeneration of female structures (Figure 3A, B).
Fig. 3. Histological section of a hermaphrodite specimen of Patella piperata gonads. (A) General view highlighting the coexistence of spermatogenic and oogenic stages. ST, seminiferous tubules; VO, vitellogenic oocytes; POF, post-ovulatory follicles; A, atresia. (B) View detailing the spermatogenic and oogenic stages. ST, seminiferous tubules; VO, vitellogenic.
Discussion
Hermaphroditism is thought to represent an advantageous strategy in increasing the reproductive success of organisms, based on the assumption that an individual changing sex will increase its reproductive efficiency (Charnov, Reference Charnov1979; Munday et al., Reference Munday, Buston and Warner2006). When male and female reproductive values are closely related to size or age, natural selection will favour genes coding for sex change (Warner, Reference Warner1988). The direction of sex change is thought to be determined by the relative fitness returns for the two sexes over the course of the limpet's lifetime and not by size (Allsop & West, Reference Allsop and West2003; Guallart et al., Reference Guallart, Calvo, Acevedo and Templado2013). If males have lower reproductive success than females in smaller sizes and greater in larger sizes, sex change will probably be protogynous to maximize the reproductive success of males later on life when their fitness accelerates above that of the females (Guallart et al., Reference Guallart, Calvo, Acevedo and Templado2013). The inverse holds for protandrous sex change (Cunha et al., Reference Cunha, Martins, Amaral and Rodrigues2007). This sex changing allocation mechanism ensures that individuals have the optimal sex for the predominant circumstances, thus maximizing the organism's genetic contribution to the next generation (Warner et al., Reference Warner, Fitch and Standish1996).
Sequential hermaphroditism assumes the existence of sexual dimorphism in size classes (Orton, Reference Orton1919; Branch, Reference Branch1974; Thompson, Reference Thompson1979; Creese et al., Reference Creese, Schiel and Kingsford1990), with the occurrence of a small number of simultaneous hermaphrodites in intermediate size classes (Bacci, Reference Bacci1947; Branch, Reference Branch1974; Creese et al., Reference Creese, Schiel and Kingsford1990). In fact, direct observations have confirmed this pattern for the protandrous hermaphrodite P. vulgata (Le Quesne & Hawkins, Reference Le Quesne and Hawkins2006). Two forms of simultaneous hermaphroditism can occur, the very rare ‘mosaic’ form, in which patches of male and female gametes appear interspersed, usually considered aberrations since they mostly occur in gonochoristic species (Dodd, Reference Dodd1956) and the ‘transitional’ form, in which defined sections of male and female gamete have been observed to occur simultaneously and is thought to represent a gonad in a transitional state of sex change (Branch, Reference Branch1974). Both these forms have been observed in the order Patellogastropoda (Orton, Reference Orton1928; Branch, Reference Branch1974), the mosaic form in P. vulgata and the transitional form in P. caerulea and Cymbula oculus (Born, 1778) (Dodd, Reference Dodd1956; Branch, Reference Branch1974).
The results reported herein are the first observations of a transitional form of sequential hermaphroditism in P. piperata. This species size distribution ranges from 3 to 46 mm shell length (R. Sousa, personal communication) and the hermaphroditic individuals occurred in the size classes ranging from 20 to 38 mm representing ~5% of the examined individuals. The size at which sex change is estimated to occur was 36.08 mm. This corresponds to the size class at which 50% of the individuals in this population are of the second sex. Nonetheless, sex change in P. piperata starts to occur earlier, with 25% of limpets having changed sex at 29.22 mm shell length and the smallest hermaphrodite found had 21.72 mm shell length. The sex ratio was biased towards females in most of the sampled size range, with males being predominant in the smaller size classes and females in the larger size classes, corresponding to the size at which hermaphrodites become more common. The skewed sex ratio to females is uncommon in protandrous species where small males are often numerically dominant (Allsop & West, Reference Allsop and West2004). This could be a result of incomplete sampling with smaller recently matured males occurring lower on the shore or in cryptic habitats.
In gastropods, sex change is also thought to be determined genetically and to occur mainly after the first reproductive period (Fretter et al., Reference Fretter, Graham, Ponder, Lindberg, Beesley, Ross and Wells1998). In some species, sex change is triggered by specific environmental signals and this may be the case of P. piperata. However, the environmental cues that trigger sex change in this species are yet to be determined.
Hoagland (Reference Hoagland1978) stated that in protandrous gastropods the presence of females or other larger conspecifics delays or inhibits sex change in males, with abundance acting as a trigger for sex change. As such, sex change in P. piperata in Madeira may be density dependent, where a greater overall abundance of one sex may prompt a sex change to balance the sex ratio of these populations, as reported for P. aspera (Martins et al., Reference Martins, Borges, Vale, Ribeiro, Ferraz, Martins, Santos and Hawkins2017), P. vulgata (Borges et al., Reference Borges, Doncaster, Maclean and Hawkins2015) and P. ferruginea (Espinosa et al., Reference Espinosa, Guerra-García, Fa and García-Gómez2009; Rivera-Ingraham et al., Reference Rivera-Ingraham, Espinosa and García-Gómez2011).
The occurrence of 5% of simultaneous hermaphroditic individuals in the sampled populations in the size classes between 20–38 mm, warrants further studies to better understand the impact of this phenomenon on the reproductive biology and population dynamics of P. piperata in Madeira namely, whether the mature gametes produced in hermaphrodites are functional or not and provide evidence of the direction of sex change. Another aspect that requires additional research is the environmental cues that control this phenomenon in this species.
Acknowledgements
The authors are grateful to the Fisheries Research Service (DSI) from the Regional Directorate of Fisheries of the Autonomous Region of Madeira. We also acknowledge Jorge Lucas and Filipe Andrade for their help during this work, namely in biological sampling and harvesting field surveys.
Financial support
The first author (RS) was supported by a grant from ARDITI OOM/2016/010 (M1420-01-0145-FEDER-000001-Observatório Oceânico da Madeira-OOM) and the third author (JV) by a grant from Fundação para a Ciência e a Tecnologia – FCT (SFRH/BSAB/143056/2018) and UID/MAR/04292/2019 granted to MARE. The present study has also been supported by the UE European Regional Development Fund in the framework of the Projects MARISCOMAC (MAC/2.3d/097), MACAROFOOD (MAC/2.3d/015) and the Regional Government of Madeira.
