reproductive biology of three hydrothermal vent ...

7 downloads 0 Views 520KB Size Report
Apr 15, 2010 - N. SUBNODA AND PELTOSPIRA OPERCULATA) ASSOCIATED WITH. POMPEII WORMS ON THE EAST PACIFIC RISE. M. MATABOS1,2,4 ...
REPRODUCTIVE BIOLOGY OF THREE HYDROTHERMAL VENT PELTOSPIRID GASTROPODS (NODOPELTA HEMINODA, N. SUBNODA AND PELTOSPIRA OPERCULATA) ASSOCIATED WITH POMPEII WORMS ON THE EAST PACIFIC RISE M. MATABOS 1,2,4 AND E. THIEBAUT 2,3 1

Muse´um National d’Histoire Naturelle, De´partement Milieux et Peuplements Aquatiques, UMR7208 BOREA (MNHN, UPMC, CNRS), CP53, 61 rue Buffon, F-75231 Paris cedex 05, France; 2 Universite´ Pierre et Marie Curie-Paris 6, Station Biologique de Roscoff, UMR 7144, BP 74, F-29682 Roscoff cedex, France; 3 CNRS, Station Biologique de Roscoff, UMR 7144, BP 74, F-29682 Roscoff cedex, France; and 4 Present address: School of Earth and Ocean Sciences, University of Victoria, PO Box 3065, STN CSC, Victoria, BC V8W 3V6, Canada Correspondence: M. Matabos; e-mail: [email protected] (Received 12 June 2009; accepted 18 January 2010)

ABSTRACT

INTRODUCTION Hydrothermal vents host dense but ephemeral communities that thrive on chemosynthetic microbial production. Although it is commonly postulated that energy availability is high and continuous, populations inhabiting these environments frequently experience high spatial and temporal variability in physico-chemical conditions related to variations in hydrothermal activity. At another scale, nonperiodic tectonic events and volcanic eruptions lead to extinctions and creation of new vent habitats, and vent species must be able to colonize new vents, separated by tens to hundreds of metres within a vent field to hundreds of kilometres between vent fields, in order to maintain local populations. In such a geographically disjunct, variable and unpredictable environment, vent faunal species would be assumed to have an r-type life-history strategy including fast growth, early reproduction and wide spread dispersal of larvae (Eckelbarger, 1994; Ramirez-Llodra, 2002; Young, 2003). Life-history traits of vent invertebrate species are still only poorly known, but are essential to understanding ecological processes that influence the establishment and persistence of populations and community structure, or the ability of organisms to adapt to their environment (Tyler & Young, 1999; Ramirez-Llodra, 2002). Logistic constraints related to research vessel scheduling and sampling with submersibles have severely limited time-series studies of vent species. Sampling is thus often

inadequate for determining variability at the different scales of environmental fluctuations. Reproductive ecology has often been studied from one-time analyses of oocyte size distributions, from different locations, sampled in different years (Zal et al., 1995; Ramirez-Llodra, Tyler & Copley, 2000; Tyler et al., 2008). Very few studies are based on the analysis of spatiotemporal variation in gametogenesis, and then mainly over short-term periods (e.g. Copley et al., 2003; Faure et al., 2007 but see Dixon et al., 2006). Most of our knowledge of recruitment and dispersal of vent invertebrates has been derived from population structure and population genetics (Vrijenhoek, Shank & Lutz, 1998; Tyler & Young, 1999) although there have been recent larval culture and in situ larval sampling efforts (see Pradillon et al., 2001; Mullineaux et al., 2005). For molluscs, larval shell morphology has also been used to infer larval development and dispersal capability, assuming that species with nonplanktotrophic larvae have limited dispersal (Lutz et al., 1986). Criteria used to infer modes of larval development (e.g. the size of the protoconch I) are based on observations of shallow-water species and the same criteria may not reflect early life history of deep-sea species (Gustafson & Lutz, 1994). Furthermore, those criteria do not permit discrimination between species with pelagic larvae that are able to disperse, or species that brood their young or hatch juveniles from egg capsules (Jablonski & Lutz, 1983). While most reproductive features (e.g. embryo survival, age at first maturity, quantity of eggs produced, time of spawning)

Journal of Molluscan Studies (2010) 76: 257–266. Advance Access Publication: 15 April 2010 # The Author 2010. Published by Oxford University Press on behalf of The Malacological Society of London, all rights reserved.

doi:10.1093/mollus/eyq008

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

The gonad morphology and gametogenesis of three peltospirid gastropods (Nodopelta heminoda, N. subnoda and Peltospira operculata) from hydrothermal vents in the East Pacific Rise were analysed using histological techniques. Samples were collected on different hydrothermal structures from three chemically distinct sites in vent fields along the north East Pacific Rise (138N) to assess spatial variation in oocyte size –frequency distributions. The three species revealed the same gonad morphology. The ovary, which occupies a large part of the body, is confined to the dorsal area of the visceral mass and is located between the dorsal gill lamellae and the digestive gland that appear progressively from the posterior part. The oogonia develop into previtellogenic oocytes, and when they grow to a size of 50–70 mm, vitellogenesis begins. Maximal egg size ranges between 129 mm for P. operculata and 150 mm for N. heminoda. The presence of oocytes at all stages of development, in all females, suggested a continuous gametogenesis within those species. Continuous reproduction appears to be a common trait in vent gastropods and may be adaptive in relation to constant energy supply and the ephemeral and unstable vent environment. Percentage of vitellogenic oocytes (i.e. gametogenic maturity) was independent of shell length. The slight variability observed between gametogenetic maturity of females from different localities is likely due to random individual variations in reproductive patterns rather than variation in environmental conditions or food availability.

M. MATABOS AND E. THIEBAUT The Peltospiridae are a gastropod family endemic to vents along Juan de Fuca Ridge (one species), the East Pacific Rise (13 species), the Mid-Atlantic Ridge (two species) and the Indian Ocean (one species) (Desbruye`res, Segonzac & Bright, 2006). Along the East Pacific Rise, some species are a major component of total fauna in mussel clumps, vestimentiferan tubes, active black smokers and alvinellid colonies. In this last environment, three peltospirid species have been commonly reported: Nodopelta heminoda McLean, 1989 from 218N to 138N, N. subnoda McLean, 1989 from 218N to 138N, and Peltospira operculata McLean, 1989 from 218N to 178S (Jollivet, 1996; Matabos et al., 2008). Although Alvinella colonies are commonly defined as the warmest and most toxic hydrothermal environment, organisms reported in this habitat undergo strong variations in physico-chemical conditions over small spatial (centimetres) and temporal (seconds) scales (Le Bris, Zbinden & Gaill, 2005; Matabos et al., 2008). The anatomy of peltospirids including the reproductive organs has been extensively studied by Fretter (1989), providing useful information on fertilization mode. In contrast, reproductive biology has been examined in detail for only one species, Rhynchopelta concentrica, from samples collected from different vent fields at different seasons (Tyler et al., 2008). In this study, we investigated female reproductive morphology and analysed, for the first time, oogenesis in three peltospirid gastropods (i.e. Nodopelta heminoda, N. subnoda and Peltospira operculata) sampled from different physico-chemical environments at 138N (East Pacific Rise). Results are discussed with respect to the life-history traits of gastropods inhabiting the hydrothermal environment, especially the harsh conditions encountered in the alvinellid colonies habitat, and in terms of implications for dispersal capability.

MATERIAL AND METHODS All individuals were collected in Alvinella pompejana colonies at different vent sites during the 2002 PHARE cruise to the 138N vent field along the northern East Pacific Rise (NEPR). Sampling was performed using the hydraulic arm of the ROV Victor 6000 (Ifremer), occasionally completed with the ROV suction device. A hierarchical sampling was performed: samples were collected at three different sites (Genesis, Parigo and Elsa) spaced by 1000 s of metres and, within each site, at one or two active sulphide structures (here referred to as ‘vents’) separated by 10’s of metres (Table 1). For each species, two physico-chemical habitat types were sampled: one with lower temperatures, near neutral pH and moderate sulphide concentrations (Parigo PP-Ph05 and Elsa PP-Ph01), and one with higher temperatures, slightly acidic pH and higher sulphide concentrations (Genesis PP12 and Elsa Hot3). The mean values of measured temperatures, pH and sulphide concentrations for individual biological sampling locations at each vent are given in Table 1 (see Matabos et al., 2008 for detailed description of temperature measurements and chemical analysis). Samples were washed in a 1-mm mesh sieve and fixed in 10% buffered formalin in seawater on the ship deck. In the

Table 1. Location of sampling sites of the three species of peltospirid gastropods along the East Pacific Rise at 138N. For each vent, the mean values of measured temperature, and estimated pH and sulphide concentrations are given for different locations at the surface of alvinellid colonies. Vent site

Vent

Latitude

Longitude ′



Dive N8

Temperature (8C)

pH

Sulphide (mmol/l)

Genesis

PP12

12848.632 N

103856.426 W

164-16

25.4 – 32.3

6.0 –6.1

306 –394

Elsa

Hot3

12848.145′ N

103856.266′ W

168-20

20.4 – 23.8

6.5 –6.6

339 –406

PP-Ph01

12848.150′ N

103856.267′ W

168-20

6.2

7.4

PP-Ph05

12848.585′ N

103856.400′ W

162-14

7.1– 16.0

7.4 –7.8

Parigo

258

53 164 –404

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

can vary in response to environmental conditions (i.e. food supply, habitat), vitellogenic mechanisms appear to be phylogenetically constrained (Gustafson & Lutz, 1994) and to greatly influence reproductive biology (Eckelbarger, 1994). For gastropods, which are among the most numerous and diverse invertebrates in hydrothermal vent habitats, knowledge of reproductive strategies remains largely fragmentary (Gustafson & Lutz, 1994). Hydrothermal gastropods are generally gonochoristic species. Only pyropeltid limpets (subclass Cocculiniformia) are known to be hermaphroditic, with separate testis and ovary. For most of the species of Cocculiniformia, Vetigastropoda, Neritimorpha and Caenogastropoda, the reproductive anatomy and/or the presence of secondary sexual characters (e.g. penis, prostate gland and seminal receptacle) indicate that fertilization probably occurs internally. Internal fertilization is generally associated with copulation except among the peltospirid limpets (subclass Vetigastropoda). For this family, males lack a copulatory organ and a well-developed prostate, although sperm have been observed in female seminal receptacles (Fretter, 1989). Sperm with elongated heads and long tails are probably liberated in the immediate vicinity of the female and enter the female mantle cavity and the receptacular ducts after a brief swimming period. In the scissurellid slit limpets (subclass Vetigastropoda), males also lack a penis, but given the sperm morphology, fertilization may occur in the female mantle cavity. External fertilization has been only reported for neolepetopsid limpets (subclass Patellogastropoda). Internal fertilization could be a strategy adopted by most hydrothermal gastropods to ensure that gametes are not directly exposed to the harsh environmental conditions (Fretter, 1989). From observations on larval shell (i.e. size and sculpture) and egg sizes, nonplanktotrophic development has been suggested for numerous hydrothermal gastropods (Berg, 1985; Lutz et al., 1986; Gustafson, Littlewood & Lutz, 1991). Planktotrophic development has been proposed for only two species of turrid gastropods (subclass Caenogastropoda) (Gustafson & Lutz, 1994). For lepetodrilid gastropods, protoconch morphologies and dimensions are consistent with nonplanktotrophic development (Lutz, Jablonski & Turner, 1984) although their maximum oocyte size might lead one to infer a planktotrophic larval development (Tyler et al., 2008). Recently, Tyler et al. (2008) described the gametogenic biology of seven species of vent gastropod, including Eulepetopsis vitrea (Patellogastropoda: Neolepetopsidae), Rhynchopelta concentrica (Vetigastropoda: Peltos-piridae), Cyathermia naticoides (Vetigastropoda: Neomphalidae) and four species of Lepetodrilus (Vetigastropoda: Lepetodrilidae). These same authors highlighted a lack of periodicity in oocyte production, suggesting a quasi-continuous and asynchronous reproduction. For lepetodrilid limpets, they found no evidence of spatial variation in the oocyte size–frequency among hydrothermal vent fields. Kelly & Metaxas (2007) also reported that Lepetodrilus fucensis undergoes continuous gametogenesis and asynchronous reproduction. Conversely, at vent scales, Kelly & Metaxas (2007) showed that gametogenic maturity and fecundity vary between actively venting and senescent habitats, highlighting the influence of habitat selection on the reproductive output of this species in relation to food availability.

REPRODUCTION OF PELTOSPIRIDS ovary (Fig. 1B), growing to c. 10–20 mm before developing into previtellogenic oocytes that could be distinguished by their dark green cytoplasm and their smooth aspect (Fig. 1B, D). Oogonia and small previtellogenic oocytes had a large nucleus in comparison with the cytoplasm. Our observations suggest that vitellogenesis begins when previtellogenetic oocytes reach a size of about 50 mm in Peltospira operculata and 60 mm in Nodopelta species. Vitellogenic oocytes had a voluminous cytoplasm, granular in appearance due to the presence of yolk granules, and a large nucleus (i.e. germinal vesicle) containing a darkly stained nucleolus (Fig. 1B, D, F). For all three species, all females contained the three stages of oocytes: oogonia, previtellogenic oocytes and vitellogenic oocytes. Because of the low proportion or absence of small individuals in the samples, it was impossible to determine the minimal animal size at first maturity. However, the females of Nodopelta species appear able to reproduce at a size of about one half of their maximum size. Vesicles containing sperm were observed within the gonads in all three species and were found in a particularly large numbers in Nodopelta subnoda (Fig. 1D, F).

Oocyte size – frequency distributions The three species displayed the same pattern of oocyte size – frequency distribution, with two peaks: one major peak of previtellogenic oocytes and another minor peak of vitellogenic oocytes (Fig. 2). All oocyte developmental stages were present within all females so that there was a wide range of oocyte sizes within each individual. Maximum oocyte size was 129 mm for P. operculata, 156 mm for Nodopelta heminoda and 138 mm for N. subnoda (Table 2). For the three species, the Kruskal– Wallis multisample test indicated significant differences in gamete size–frequency distributions between females within a sample, except between females of P. operculata from the Genesis PP12 sample (Table 3; Fig. 2A). However, only one or two females within any one sample contributed to these differences, and intrasample synchrony between females ranged from 60% to 87.5% (Table 3; Fig. 2A). For all three species, comparison of oocyte size– frequency distributions between samples showed significant differences between sites. For P. operculata, oocytes in individuals from Genesis PP12 were significantly larger than those from the two Elsa vents (H ¼ 20.00, df ¼ 2, P , 0.0001). For N. heminoda, oocytes were significantly larger at Elsa Hot3 than at Genesis PP12 and Parigo PP-Ph05 (H ¼ 11.45, df ¼ 2, P ¼ 0.003). For N. subnoda, oocytes were significantly smaller in individuals from Elsa Hot3 than in individuals from Elsa PP-Ph01 (H ¼ 12.46, df ¼ 2, P ¼ 0.002). The percentage of vitellogenic oocytes in reproductively mature females ranged between 27% and 61% for P. operculata, 28% and 64% for N. heminoda and 30% and 59% for N. subnoda, and appeared independent of animal size (R 2 ¼ 0.09, P ¼ 0.33; R 2 ¼ 0.16, P ¼ 0.15; and R 2 ¼ 0.01, P ¼ 0.68, respectively; Fig. 3). For P. operculata, gametogenetic maturity varied among the three different localities with a larger proportion of vitellogenic oocytes at Elsa Hot3 in comparison with Elsa PP-Ph01 (H ¼ 6.00, df ¼ 2, P ¼ 0.049). No gametogenetic maturity differences were observed for N. heminoda or N. subnoda among the different localities (H ¼ 5.79, df ¼ 2, P ¼ 0.055 and H ¼ 0.24, df ¼ 2, P ¼ 0.89, respectively).

RESULTS General patterns of ovary morphology Ovary morphologies and gametogenesis were quite similar for all three peltospirid species (Fig. 1). The ovary occupied a large volume of the animal (around 20% of the visceral mass). It extended anteriorly on the right side and on the left side into the pallial roof, beside the dorsal gill lamellae and the digestive gland. Ventrally, the gonad was surrounded, on the left side, by the stomach and the digestive gland, gradually filling the entire space in the posterior part, and by the shell muscles on the right side (Fig. 1A, C, E). Oogonia appeared to develop from connective tissues found throughout the entire

DISCUSSION The ovaries of Nodopelta heminoda, N. subnoda and Peltospira operculata were morphologically similar. The reproductive system is located dorsally, confined to the roof of the mantle cavity on the right side. Ovaries are surrounded by shell muscle, the dorsal gill lamellae on the left, and the digestive gland that 259

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

laboratory, all gastropod specimens were sorted, identified to species level and then transferred to 70% ethanol for storage. For the three peltospirid gastropods, shell lengths were measured from images captured with a camera mounted on a dissecting microscope using the ‘UTHSCSA Image tool’ free software version 3.0 (University of Texas, Health Science Center, San Antonio, web site: http://ddsdx.uthscsa.edu/dig/ itdesc.html). For Nodopelta heminoda and N. subnoda, the curvilinear shell length, defined as the maximal length from the anterior edge of the shell to the lip of the protoconch, was used as this parameter has been reported as a sensitive indicator of growth in hydrothermal limpets (i.e. Lepetodrilus elevatus; Sadosky et al., 2002). For Peltospira operculata, characterized by a coiled teleoconch, the total body length was measured. For histological observations of female sexual maturity, the shells of the limpets were gently removed using forceps. Males and females were first distinguished by examination of the external appearance of their gonads under a dissecting microscope; females were identified by the presence of round-shaped oocytes, visible through the transparent gonad wall. Because of the small sample sizes, only five to eight females per vent were analysed for each species. The bodies of females stored in 70% ethanol were dehydrated in 100% ethanol for at least 6 h, cleared in xylene for 6 h and embedded in paraffin wax in a 708C oven for c. 12 h. Individuals were then set in wax blocks and serial transverse sections of 7 mm thickness in a horizontal plane were prepared with a microtome. Sections were then stained using the haematoxylin and picro indigo carmine method, which stains the nucleus brown and the cytoplasm green (Gabe, 1968). Only oocytes in which the nucleus was visible were measured from images acquired with a digital microscope camera. Where possible, c. 100 oocytes were measured from two to three slides separated by 70 mm in order to (1) include variability of oocyte development among different sections of the ovary and (2) make sure that the same oocyte was not measured twice. As packing of the oocytes can severely distort the oocyte shape, the feret diameter based on the measurement of the cross-sectional area of each oocyte (Tyler et al., 2008) was calculated using the Lucia software application (Laboratory Imaging Ltd). Oocyte size– frequency distributions were established for each female at each vent. Differences in these frequency distributions among females within a vent or among vents were assessed using a nonparametric Kruskal –Wallis multisample test. To determine which sample or female contributed to the observed significant differences, multiple comparisons using the Nemenyi and Dunn test were computed (Zar, 1999). Gametogenic maturity was expressed as the percentage of vitellogenic oocytes on the total number of oocytes measured. Differences in gametogenic maturity among vents were also investigated using a Kruskal– Wallis multisample test and the Nemenyi and Dunn test was used to identify which locality contributed to the observed differences.

M. MATABOS AND E. THIEBAUT

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

Figure 1. Ovarian structure of the three studied peltospirid gastropods. A. Dorsal view of Peltospira operculata ovary. B. Detailed view of Peltospira operculata ovary of adult female. C. General view of Nodopelta heminoda visceral mass. D. Detailed view of Nodopelta heminoda ovary. E. General view of Nodopelta subnoda ovary. F. Detailed view of Nodopelta subnoda ovary. Abbreviations: ct, connective tissue; dl, dorsal gill lamellae; dg, digestive gland; n, nucleus; o, ovary; oo, oogonia; pvo, previtellogenic oocytes; sm, shell muscle; st, stomach; v, vesicle; vo, vitellogenic oocytes. (This figure appears in colour in the online version of Journal of Molluscan Studies.)

progressively fills the entire space below the gonad in the posterior portion of the animal. Examination of histological sections revealed the same gametogenic pattern among the three species, with the presence of oogonia developing in previtellogenic oocytes that begin vitellogenesis at c. 50–60 mm. Ovaries

contained the three stages of oocyte development within a single female although all oocyte size–frequency distributions tended to be bimodal. This pattern suggests a quasi-continuous gametogenesis within females for those three species. Furthermore, except for P. operculata at Genesis PP12 vent, intrasample 260

REPRODUCTION OF PELTOSPIRIDS

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

Figure 2. Oocyte size– frequency distributions of the three studied peltospirid gastropods for the different sampled sites. A. Box-plots showing oocyte size–frequency distributions for each individual female. Each box-plot displays the 1st, 2nd and 3rd quartile and the range values of measurements. The mean oocyte size is represented by the grey line. B. Mean oocyte size– frequency histograms (mean + SE) for each site. Abbreviations: N, number of individuals; n, number of oocytes measured.

261

M. MATABOS AND E. THIEBAUT Table 2. Shell length and oocyte size range of the three species of peltospirid gastropods collected along the East Pacific at 138N. The number of sampled females at each vent is given. Species Peltospira operculata

Nodopelta heminoda

Nodopelta subnoda

Vent site

Vent

Number of females

Shell length (mm)

Oocyte size range (mm)

Genesis

PP12

5

10.7 –12.4

12.5 – 129.2

Elsa

PP-Ph01

7

10.5 –12.7

10.7 – 127.1

Hot3

6

8.8 –11.6

11.0 – 125.2

Genesis

PP12

5

13.9 –20.4

16.4 – 156.3

Parigo

PP-Ph05

5

17.7 –23.8

16.7 – 149.5

Elsa

Hot3

8

15.0 –16.1

12.8 – 139.3

Genesis

PP12

5

8.1 –10.2

17.7 – 137.8

Elsa

PP-Ph01

5

7.7 –10.7

18.1 – 136.9

Hot3

5

6.3 –11.2

13.5 – 137.6

Table 3. Results of Kruskal –Wallis multisample tests and the Nemenyi and Dunn multiple range tests comparing oocyte size distributions between females of the three peltospirid gastropods within each sample. Species

Nodopelta heminoda

Nodopelta subnoda

Kruskal – Wallis multisample test

Nemenyi and Dunn multiple range test

Intrasample synchrony (%)

Elsa PP-Ph01

H ¼ 23.31, P ¼ 0.000 (6 df )

1/7 different

85.71

Elsa Hot3

H ¼ 21.02, P ¼ 0.000 (5 df )

1/6 different

83.33

Genesis PP12

H ¼ 5.85, P . 0.05 (4 df )





Parigo PP-Ph05

H ¼ 23.46, P ¼ 0.000 (4 df )

1/5 different

80

Elsa Hot3

H ¼ 28.48, P ¼ 0.000 (7 df )

1/8 different

87.5

Genesis PP12

H ¼ 18.51, P ¼ 0.000 (4 df )

1/5 different

80

Elsa PP-Ph01

H ¼ 11,84, P ¼ 0.000 (4 df )

1/5 different

80

Elsa Hot3

H ¼ 14.49, P ¼ 0.000 (4 df )

1/5 different

80

Genesis PP12

H ¼ 33.41, P ¼ 0.000 (4 df )

2/5 different

60

synchrony varied between 60% and 87%, possibly indicating asynchronous reproduction at the population level. A similar pattern of gametogenesis has previously been observed for another peltospirid species, Rhynchopelta concentrica, from samples collected at four different dates on the EPR (Tyler et al., 2008). This seems to be a common trait in hydrothermal gastropods. The same pattern has been reported in different families of Vetigastropoda including Lepetodrilidae (e.g. Lepetodrilus elevatus, L. ovalis, L. pustulosus, L. fuscensis and L. atlanticus), Neomphalidae (Neomphalus fretterae and Melanodrymia aurantiaca), Peltospiridae and Sutilizonidae (Sutilizona theca) (Fretter, 1989; Gustafson & Lutz, 1994; Kelly & Metaxas, 2007; Tyler et al., 2008). To our knowledge, discontinuous reproduction has never been described for vent gastropods. Vitellogenesis (i.e. yolk synthesis), influencing the rate at which eggs are produced, is thought to be phylogenetically constrained in marine invertebrates (Eckelbarger, 1994; Gustafson & Lutz, 1994). However, based on the high variability in the modes of reproduction observed in gastropods (Webber, 1977), continuous reproduction seems to be a common adaptation to hydrothermal vents; alternatively only species with continuous reproduction can maintain viable populations in such an environment. Indeed, the specific environmental conditions at hydrothermal vents may have favoured particular sets of life-history characteristics (Eckelbarger & Walting, 1995). This latter hypothesis is more likely as hydrothermal gastropods do not constitute a monophyletic group, at least in vetigastropod phylogenies (Geiger & Thacker, 2006; Kano, 2008). Continuous gametogenesis has also been noted in a taxonomically diverse range of other vent species. Most vent polychaetes from families like the Ampharetidae (Amphisamytha galapagensis), the Alvinellidae (Alvinella pompejana or Paralvinella pandorae) or the Siboglinidae (Riftia pachyptila) are thought to reproduce continuously or

semi-continuously (McHugh, 1989; Gardiner, Shrader & Jones, 1992; McHugh & Tunnicliffe, 1994; Faure et al., 2007). The presence of all oocyte development stages in females of three caridean shrimps from the Mid-Atlantic Ridge (Rimicaris exoculata, Chorocaris chacei and Mirocaris fortunata) suggested continuous gametogenesis and the absence of synchrony between females from the same locality (Ramirez-Llodra et al., 2000). Among bivalves, Calyptogena magnifica and Bathymodiolus thermophilus from the East Pacific Rise apparently reproduce continuously over a long period of time (Berg, 1985). In such an environment, where there is a continuous energy supply, a species would likely reproduce as long as the source is active in order to optimize the number of offspring released. The role of energy supply in the control of the gametogenesis has been recently highlighted for the bivalve Bathymodiolus azoricus at a shallow and a deep-vent site on the Mid-Atlantic Ridge (Dixon et al., 2006). For this species, seasonal reproduction has been suggested to coincide with a phytoplankton bloom in surface waters and the consequent input of organic matter to the benthos. The three peltospirid species studied in the present work appear to have internal fertilization (Fretter, 1989). This is further confirmed by the observation of sperm within seminal vesicles in the ovaries of all three species. It has been suggested that internal fertilization is an adaptation that allows gametes to avoid direct exposure to a potentially toxic surrounding environment (Fretter, 1989; Ware´n & Bouchet, 1989). This is likely to be the case in vent species since intertidal vetigastropods generally undergo external fertilization (Webber, 1977). However, despite a high proportion of mature oocytes reported in the gonad examined for the three different species examined here, no embryos were observed in the mantle cavity, suggesting that fertilized eggs are directly or rapidly released into the environment. 262

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

Peltospira operculata

Sample

REPRODUCTION OF PELTOSPIRIDS

Figure 3. Relationships between the animal size and the percentage of vitellogenic oocytes for each female of Peltospira operculata A, Nodopelta heminoda B and Nodopelta subnoda C. The proportion of vitellogenic oocytes was determined as the percentage of vitellogenic oocytes on the total number of oocytes measured. Abbreviations: N, number of individuals; n, number of oocytes measured; Ltot, total length; Lcurv, curvilinear length.

Egg sizes for the three species analysed in this study were similar. Maximum oocyte diameter was c. 129 mm for P. operculata, c. 156 mm for N. heminoda and c. 138 mm for N. subnoda. These values are in the same range as those reported for 263

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

another peltospirid, Rhynchopelta concentrica, by Berg (1985) (i.e. 151.7 + 15.9 mm by 131.7 + 14.5 mm) whereas Fretter (1989) found ova in the ovarian duct of this species to measure approximately ‘10% greater’ than those measured by Berg. Tyler et al. (2008) reported a maximum egg size of 184 mm for R. concentrica, with the next oocyte closest in size being 149 mm. Differences in measurement of egg size can be partly due to histological processes that lead to shrinkage of the tissues (Tyler et al., 2008). In addition, the size of irregular-shaped oocytes with noncentral nuclei can be misjudged when the plane of sectioning is not parallel to the major axis of the oocyte (Copley & Young, 2006). The maximum egg sizes observed here are intermediate among those reported for hydrothermal vent gastropods. While maximum egg size for Lepetodrilus species are around 90 –100 mm (Kelly & Metaxas, 2007; Tyler et al., 2008), it varies around 150 mm in different Neomphalidae (Neomphalus fretterae and Pachydermia laevis) and Seguenzioidea (Bathymargarites symplector), and can reach 250– 300 mm in Neolepetopsidae (Eulepetopsis vitrea and Neolepetopsis occulta) (Berg, 1985; Gustafson & Lutz, 1994). In marine invertebrates, egg size is a crucial life-history trait, because it is often correlated with the mode of development and the rate of dispersal. However, for gastropods, a compilation of data on the reproductive characteristics is lacking and most information is taxon-specific. For 62 species of Indo-Pacific Conus (Caenogastropoda), Kohn & Perron (1994) showed a significant correlation between egg size and mode of development: egg size of planktotrophic species ranged between 125 and 425 mm and egg size of direct developers ranged between 470 and 1,000 mm. Only one species exhibits a lecithotrophic larva, with an egg size of 390 mm. On the other hand, in the caenogastropod family Calyptraeidae, egg size of 78 species is not significantly different between modes of development: between 140 and 1,200 mm (mean: 336 mm) for direct developers, between 230 and 350 mm for lecithotrophic development (mean: 302 mm) and between 120 and 350 mm (mean: 189 mm) for planktotrophic developer (Collin, 2003). Thus, for some gastropods, egg size is a poor indicator of mode of development (Tyler et al., 2008). Although egg sizes of hydrothermal gastropods are small and often below 250 mm, observations made on larval shells suggest a lecithotrophic development for these species (Lutz et al., 1986). Interpretation of protoconch morphology can be misleading as a result of mechanical deformation during larval development. Inferences from protoconch observations may therefore not be reliable given the range of variation in larval development (Hickman, 1992). The question of how vent larvae disperse along distances that account for their large distribution area, despite a supposedly limited larval life span as a consequence of a lecithotrophic development, has been of great interest to biologists. Several explanations have been proposed during the past decade. First, the emergence of genetic studies has revealed that species with lecithotrophic larvae may display a genetic homogeneity along thousands of kilometres suggesting high rates of dispersal. The ‘type of larval development/dispersal distance’ relationship that would predict that lecithotrophic larvae have limited dispersal and that planktotrophic larvae have high dispersal capabilities may not be valid. As an example, the mussel Bathymodiolus thermophilus, which has a planktotrophic larva, shows greater genetic variation between populations and a smaller spatial distribution than the clam Calyptogena magnifica, which has large lecithotrophic larvae (Craddock et al., 1995; Karl et al., 1996). The average rate of gene flow estimated for C. magnifica is 1.5 times higher than that of B. thermophilus. Second, several authors have suggested that larvae dispersing in cold abyssal water may be able to reduce their metabolism and so delay their development

M. MATABOS AND E. THIEBAUT

Figure 4. Size– frequency distribution of the curvilinear shell length of samples collected at Genesis PP12 vent site. A. Nodopelta heminoda. B. N. subnoda. Abbreviations: n, number of measured individuals; Lcurv, curvilinear length.

(Gustafson & Lutz, 1994; Mullineaux, Mills & Goldman, 1998; Young, 2003). As an example, Pradillon et al. (2001) suggested that embryos of Alvinella pompejana, characterized by a lecithotrophic development can arrest their development when dispersing at 28C between vents until warm water is encountered. Experimental studies are needed in order to determine whether similar behaviour occurs in other vent species. Peltospirid juveniles were scarce in our samples although reproductive patterns observed in this study suggest that there is a continuous release of either fertilized eggs or embryos. When the sample size was large enough, animal size – frequency distributions were investigated. Medium- and largesized individuals with mean sizes of 6.9 + 1 mm for P. operculata, 17.5 + 3 mm for N. heminoda and 7.5 + 1.8 mm for N. subnoda dominated the distributions (Fig. 4). There are several possible explanations for these observations. First, given the small size of the protoconch (,250 mm) (McLean, 1988), juveniles may not be retained on a 1-mm sieve. However, small nonmature individuals (i.e. below 10 or 5 mm for N. heminoda and N. subnoda, respectively) were found, albeit in very low numbers. Second, different microhabitats contrasting in terms of physico-chemical environment have been described within a single alvinellid colony (Le Bris et al., 2005); however, spatial variation at the colony scale has never been investigated for gastropods at hydrothermal vents. Our sampling involved the collection of a large volume of the colonies and no juvenile peltospirids were collected from other hydrothermal habitats, such as Riftia clumps, during the same cruise (M.M. & E.T., personal observations). Thus, if present, juveniles may have been living in other cryptic, rarely sampled microhabitats, explaining their absence in the samples. Third, the occurrence of discontinuous recruitment could explain the pattern observed depending on the fate of the released larval pool in response to fluctuations in local hydrodynamic conditions (Tyler et al., 2008). Currents are major factors in larval dispersal and recruitment of vent invertebrates, the larval pool being either exported to neighbouring vent field or retained locally (Mullineaux et al., 2005). A continuous release of embryos

ACKNOWLEDGEMENTS We thank the captain and the crew of the NO L’Atalante, and the ROV Victor 6000 team for their helpful collaboration at sea. We are particularly grateful to N. Le Bris, chief scientist of the Phare’02 cruise who greatly supported our sampling program. We thank Madeleine Martin (MNHN, Paris) who provided valuable help and assistance with histological processes, and Ann Andersen (Station Biologique de Roscoff ) for their helpful 264

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

enhances retention by increasing the variation in the currents encountered by larvae (Byers & Pringle, 2006), thus facilitating the maintenance of local populations. The adults’ unimodal size –frequency distribution could then reflect a single recruitment event. This would suggest that the continuous reproduction of hydrothermal invertebrates is an adaptation to increase local retention of larvae in a highly variable hydrodynamic environment. Finally, Nodopelta species and lepetodrilid females are able to reproduce at half (this study) and at less than one-third of their maximum size, respectively, suggesting early maturity and potentially rapid growth for those species (Tyler et al., 2008). A continuous but low larval supply combined with a rapid growth of juveniles could account for the virtual absence of juveniles in the samples and the dominance of adults, if there is a decrease in animal growth rate over time. Such recruitment dynamics have been proposed for the polychaete Alvinella pompejana, which also exhibits a semicontinuous gametogenesis (Faure et al., 2007). Variability in oocyte size– frequency distributions was detected among the three localities for each species. Some differences in the gametogenetic maturity were also observed in P. operculata: females from Elsa PP-Ph01 exhibited a lower maturity (i.e. lower proportion of vitellogenic oocytes) than females from Elsa Hot3. However, no consistent pattern was observed among species in relation to the physico-chemical characteristics of the vent sites and the variability observed was more likely due to random individual variations in reproductive characteristics (e.g. time since last spawning event, females’ condition). Variations in reproductive output (e.g. fecundity, egg size, age at maturity) are related to energy allocated for growth, environmental conditions and food availability (Ramirez-Llodra, 2002). In the hydrothermal environment, variations in fluid flow should influence sulphide supply and primary production by chemoautotrophic bacteria (Sarrazin et al., 1999). Several hypotheses can be proposed to explain the lack of a clear relationship between oocyte size – frequency distribution and the chemical environment. The little spatial variability observed in this study may result in a trade-off between somatic growth and gametogenesis depending on environmental conditions and biological interactions (Kelly & Metaxas, 2007). From the same samples, Matabos et al. (2008) suggested that one of the main gastropods encountered along the EPR, Lepetodrilus elevatus, was able to outcompete peltospirid gastropods in its range of environmental conditions. While Kelly & Metaxas (2007) reported significant differences in the gametogenic maturity of Lepetodrilus fuscensis between active and senescent hydrothermal vents in the Northeast Pacific, they detected no differences between active sites that differed in terms of temperature and hydrothermal fluid flow vigour. On the other hand, given the high diversity of microbial metabolism and geochemical energy sources in the hydrothermal vents, sulphide may not be the only prime electron donor and could be a poor proxy for chemolithoautotrophic primary production (Schmidt et al., 2008). Finally, in the present study, gametogenetic maturity was only assessed through oocyte size and the proportion of vitellogenic oocytes while variations in energy can affect other reproductive parameters (e.g. fecundity).

REPRODUCTION OF PELTOSPIRIDS

REFERENCES BERG, C.J.J. 1985. Reproductive strategies of mollusks from abyssal hydrothermal vent communities. Bulletin of the Biological Society of Washington, 6: 185 –197. BYERS, J.E. & PRINGLE, J.M. 2006. Going against the flow: retention, range limits and invasions in advective environments. Marine Ecology Progress Series, 313: 27– 41. COLLIN, R. 2003. Worldwide patterns in mode of development in calyptraeid gastropods. Marine Ecology Progress Series, 247: 103–122. COPLEY, J.T.P., TYLER, P.A., VAN DOVER, C.L. & PHILIP, S.J. 2003. Spatial variation in the reproductive biology of Paralvinella palmiformis (Polychaeta: Alvinellidae) from a vent field on the Juan de Fuca Ridge. Marine Ecology Progress Series, 255: 171–181. COPLEY, J.T.P. & YOUNG, C.M. 2006. Seasonality and zonation in the reproductive biology and population structure of the shrimp Alvinocaris stactophila (Caridea: Alvinocarididae) at a Louisiana Slope cold seep. Marine Ecology Progress Series, 315: 199– 209. CRADDOCK, C., HOEH, W.R., LUTZ, R.A. & VRIJENHOEK, R.C. 1995. Extensive gene flow among mytilid (Bathymodiolus thermophilus) populations from hydrothermal vents of the eastern Pacific. Marine Biology, 124: 137–146. DESBRUYE`RES, D., SEGONZAC, M. & BRIGHT, M. 2006. Handbook of hydrothermal vent fauna. Denisia 18. Biologiezentrum, Linz, Austria. DIXON, D.R., LOWE, D.M., MILLER, P.I., VILLEMIN, G.R., COLACO, A., SERRA˜O-SANTOS, R. & DIXON, L.R.J. 2006. Evidence of seasonal reproduction in the Atlantic vent mussel Bathymodiolus azoricus, and an apparent link with the timing of photosynthetic primary production. Journal of the Marine Biological Association of the United Kingdom, 86: 1363–1371. ECKELBARGER, K.J. 1994. Diversity of metazoan ovaries and vitellogenic mechanisms: implication for life history theory. Proceedings of the Biological Society of Washington, 107: 193– 218. ECKELBARGER, K.J. & WALTING, L. 1995. Role of phylogenetic constraints in determining reproductive patterns in deep-sea invertebrates. Invertebrate Biology, 114: 256– 269. FAURE, B., CHEVALDONNE, P., PRADILLON, F., THIEBAUT, E. & JOLLIVET, D. 2007. Spatial and temporal dynamics of reproduction and settlement in the Pompei worm Alvinella pompejana (Polychaeta: Alvinellidae). Marine Ecology Progress Series, 348: 197– 211. FRETTER, V. 1989. The anatomy of some archaegastropod limpets (Superfamily Peltospiracea) from hydrothermal vents. Journal of Zoology (London), 218: 123–169. GABE, M. 1968. Techniques histologiques. Masson, Paris. GARDINER, S.L., SHRADER, S.E. & JONES, M.L. 1992. Preliminary observations on oogeneis in the tubeworm Riftia pachyptila Jones (Vestimentifera). American Zoologist, 32: 124A. GEIGER, D.L. & THACKER, C.E. 2006. Molecular phylogeny of basal gastropods (Vetigastropoda) shows stochastic colonization of chemosynthetic habitats at least from the mid Triassic. Cahiers de Biologie Marine, 47: 343– 346. GUSTAFSON, C., LITTLEWOOD, D.T.J. & LUTZ, R.A. 1991. Gastropod egg capsules and their contents from deep-sea hydrothermal vent environments. Biological Bulletin, 180: 34–55. GUSTAFSON, C. & LUTZ, R.A. 1994. Molluscan life history traits at deep-sea hydrothermal vents and cold methane/sulfide seeps. In: Reproduction, larval biology, and recruitment of the deep-sea benthos (C.M. Young & K.J. Eckelbarger, eds), pp. 76–97. Columbia University Press, New York. HICKMAN, C.S. 1992. Reproduction and development of trochacean gastropods. Veliger, 35: 245– 272.

265

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

JABLONSKI, D. & LUTZ, R.A. 1983. Larval ecology of marine invertebrates: Paleobiologic implications. Biological Review, 58: 21–89. JOLLIVET, D. 1996. Specific and genetic diversity at deep-sea hydrothermal vents: an overview. Biodiversity and Conservation, 5: 1619– 1653. KANO, Y. 2008. Vetigastropod phylogeny and a new concept of Seguenzioidea: independent evolution of copulatory organs in the deep-sea habitats. Zoologica Scripta, 37: 1– 21. KARL, S.A., SCHUTZ, S., DESBRUYERES, D., LUTZ, R. & VRIJENHOEK, R.C. 1996. Molecular analysis of gene flow in the hydrothermal vent clam (Calyptogena magnifica). Molecular Marine Biology and Biotechnology, 5: 193–202. KELLY, N.E. & METAXAS, A. 2007. Influence of habitat on the reproductive biology of the deep-sea hydrothermal vent limpet Lepetodrilus fucensis (Vetigastropoda: Mollusca) from the Northeast Pacific. Marine Biology, 151: 649–662. KOHN, A.J. & PERRON, F.E. 1994. Life history and biogeography patterns in Conus. Oxford University Press, New York. LE BRIS, N., ZBINDEN, M. & GAILL, F. 2005. Processes controlling the physico-chemical micro-environments associated with Pompeii worms. Deep-Sea Research Part I, 52: 1085–1092. LUTZ, R.A., BOUCHET, P., JABLONSKI, D., TURNER, R.D. & WARE´N, A. 1986. Larval ecology of mollusks at deep-sea hydrothermal vents. American Malacological Bulletin, 4: 49– 54. LUTZ, R.A., JABLONSKI, D.J. & TURNER, R.D. 1984. Larval development and dispersal at deep-sea hydrothermal vents. Science, 226: 1451–1454. McHUGH, D. 1989. Population structure and reproductive biology of two sympatric hydrothermal vent polychaetes, Paralvinella pandorae and P. palmiformis. Marine Biology, 103: 95–106. McHUGH, D. & TUNNICLIFFE, V. 1994. Ecology and reproductive biology of the hydrothermal vent polychaete Amphisamytha galapagensis (Ampharetidae). Marine Ecology Progress Series, 106: 111–120. McLEAN, J.H. 1988. New archaeogastropod limpets from hydrothermal vents: new family of Peltospiridae, new superfamily Peltospiracea. Zoologica Scripta, 18: 49–66. MATABOS, M., LE BRIS, N., PENDLEBURY, S. & THIE´BAUT, E. 2008. Role of physico-chemical environment on gastropod assemblages at hydrothermal vents on the East Pacific Rise (138N/ EPR). Journal of the Marine Biological Association of the United Kingdom, 88: 995–1008. MULLINEAUX, L.S., MILLS, S.W. & GOLDMAN, E. 1998. Recruitment variation during a pilot colonization study of hydrothermal vents (98500 N, East Pacific Rise). Deep-Sea Research Part II, 45: 441–464. MULLINEAUX, L.S., MILLS, S.W., SWEETMAN, A.K., BEAUDREAU, A.H., METAXAS, A. & HUNT, H.L. 2005. Vertical, lateral and temporal structure in larval distributions at hydrothermal vents. Marine Ecology Progress Series, 293: 1 –16. PRADILLON, F., SHILLITO, B., YOUNG, C.M. & GAILL, F. 2001. Development arrest in vent worm embryos. Nature, 413: 698–699. RAMIREZ-LLODRA, E. 2002. Fecundity and life-history strategies in marine invertebrates. Advances in Marine Biology, 43: 87–170. RAMIREZ-LLODRA, E., TYLER, P.A. & COPLEY, J.T.P. 2000. Reproductive biology of three caridean shrimps, Rimicaris exoculata, Chorocaris chacei and Mirocaris fortunata (Caridea: Decapoda), from hydrothermal vents. Journal of the Marine Biological Association of the United Kingdom, 80: 473 –484. SADOSKY, F., THIE´BAUT, E., JOLLIVET, D. & SHILLITO, B. 2002. Recruitment and population structure of the vetigastropod Lepetodrilus elevatus at 138N hydrothermal vent sites on East Pacific Rise. Cahiers de Biologie Marine, 43: 399 –402. SARRAZIN, J., JUNIPER, S.K., MASSOTH, G. & LEGENDRE, P. 1999. Physical and chemical factors influencing species distributions on hydrothermal sulphide edifices of the Juan de Fuca Ridge, Northeast Pacific. Marine Ecology Progress Series, 190: 89– 112. SCHMIDT, C., VUILLEMIN, R., LE GALL, C., GAILL, F. & LE BRIS, N. 2008. Geochemical energy sources for microbial primary

comments on the first draft of the manuscript. We are also grateful to K. Juniper for proofreading the English and to two anonymous referees for their valuable comments on an earlier version of this manuscript. This work was financially supported by the GDR Ecchis (Ifremer, CNRS) and is a contribution to the ANR ‘Deep Oases’ project (ANR-06-BDIV-005).

M. MATABOS AND E. THIEBAUT production in the environment of hydrothermal vent shrimps. Marine Chemistry, 108: 18–31. TYLER, P.A., PENDLEBURY, S., MILLS, S.W., MULLINEAUX, L., ECKELBARGER, K.J., BAKER, M. & YOUNG, C.M. 2008. Reproduction of gastropods from vents on the East Pacific Rise and the Mid-Atlantic Ridge. Journal of Shellfish Research, 27: 107–118. TYLER, P.A. & YOUNG, C.M. 1999. Reproduction and dispersal at vents and cold seeps. Journal of the Marine Biological Association of the United Kingdom, 79: 193 –208. VRIJENHOEK, R.C., SHANK, T.M. & LUTZ, R.A. 1998. Gene flow and dispersal in deep-sea hydrothermal vent animals. Cahiers de Biologie Marine, 39: 363–366. WARE´N, A. & BOUCHET, P. 1989. New gastropods from East Pacific hydrothermal vents. Zoologica Scripta, 18: 67– 102.

WEBBER, H.H. 1977. Gastropoda: Prosobranchia. In: Reproduction of marine invertebrates. Vol. IV (A.C. Giese & J.S. Pearse, eds), pp. 1 –97. Academic Press, New York. YOUNG, C.M. 2003. Reproduction, development and life-history traits. In: Ecosystems of the world, Ecosystems of the deep oceans. Vol. A (P.A. Tyler, ed.), pp. 381–426. Elsevier Science B.V., Amsterdam. ZAL, F., JOLLIVET, D., CHEVALDONNE´, P. & DESBRUYE`RES, D. 1995. Reproductive biology and population structure of the deep-sea hydrothermal vent worm Paralvinella grasslei (Polychaeta: Alvinellidae) at 138N on the East Pacific Rise. Marine Biology, 122: 637–648. ZAR, J.H. 1999. Biostatistical analysis. Edn 4. Prentice Hall, Upper Saddle River, NJ.

Downloaded from mollus.oxfordjournals.org at University of Victoria, McPherson Library Serials on June 3, 2011

266