Updating the current knowledge on the relationships between Haplosyllis chamaeleon Laubier, 1960 (Annelida, Syllidae) and Paramuricea clavata (Risso, 1826) (Cnidaria, Plexauridae) in the Mediterranean Sea

In the Mediterranean Sea, the symbiosis between the gorgonian Paramuricea clavata (Risso, 1826) and the polychaete Haplosyllis chamaeleon Laubier, 1960 (Annelida, Syllidae, Syllinae) has only been documented from the western basin. Our findings extend its geographic distribution to the north-central basin and represent the first record of H. chamaeleon in Italy and Croatia. Periodic observations from the Ligurian Sea allowed establishing that the symbiont occurs on P. clavata almost throughout the year, showing a reproductive period longer than previously reported. Morphometric comparisons of three Mediterranean populations, from Portofino Promontory (Ligurian Sea), Cape of Creus (Catalan Sea) and Chafarinas Archipelago (Alboran Sea) proved that there were no significant differences in body measurements, whilst the observed differences in dorsal cirri length pattern could be consider intra-specific. Our behavioural observations confirm that the species had (i) a kleptoparasitic behaviour, (ii) did not cause injuries to the host and (iii) did not induce the host to generate any malformation.

The EU project Marine Ecosystem Restoration in Changing European Seas (MERCES) seeks to support decision-making aimed at conserving and restoring key European habitats and particularly emphasised the importance of life history in species affected by restoration activities, including the ecosystem engineer P. clavata. Within this frame, our study aimed (i) to review and update the geographic distribution of the symbiotic populations of H. chamaeleon, (ii) to describe the host/symbiont interactions based on populations sampled in the Ligurian Sea, (iii) to describe reproductive traits of the symbiont and (iv) to compare the main morphometric traits of the symbiont from the Ligurian Sea, the Cap de Creus (Catalan Sea) and the Chafarinas Archipelago (Alboran Sea).

Materials and methods
To analyse the infestation characteristics, we collected apical branches of both of yellow and red chromotypes of P. clavata (about 15 cm long) by scuba diving (Table 1). For the temporal monitoring, we collected 192 apical branches of P. clavata at about 40 m depth along the Portofino Promontory (Ligurian Sea, Italy) in September and November 2016, from May to October 2017, in January and February 2018 and in May and June 2018 (Table 2). In all sites, we randomly cut the branches with sharp scissors and enclosed them in individual plastic zip-bags to prevent faunal loss. Detailed collection information for the specimens collected in the Cap of Creus and the Chafarinas Archipelago can be found in Martin et al. (2002). The Italian polychaetes were identified according to Laubier (1960), López et al. (1996), Martin et al. (2002), and Lattig and Martin (2009). All records of H. chamaeleon and the related information were included in a datasheet following Di Camillo et al. (2018a) (Electronic Supplementary Material 1) to be analysed.
In all sampling sites, the frequency of the association was estimated as the number of branches hosting polychaetes versus the total number of collected branches (expressed as percentage). Infestation rates were estimated as the total number of polychaetes found in a given branch divided by the total linear branch length (in centimetre and measured from pictures using the "ImageJ 1.46r" software) and extrapolated to the number of symbionts per host linear metre. The possible changes in infestation through time were assessed on nontransformed data by the non-parametric Kruskal-Wallis test (Sokal and Rohlf 1981) as data were not normally distributed (Shapiro-Wilk's test). Analyses were performed using PAST for Windows version 4.0 (Hammer et al. 2001).
To describe the location and behaviour of the symbionts, gorgonian fragments were placed on Petri dishes with a base covered by graph paper, and examined in vivo under a stereomicroscope. In addition to direct observations, we obtained photographs and videos using a Canon PowerShot G16. After keeping samples at 4°C overnight, to relax them before fixation, the samples were preserved (separately, the gorgonians and the syllids) in 70% ethanol, except those used for observations in vivo, which were fixed either in 96% ethanol for light microscope observations or rinsed in distilled water, dehydrated through a graded ethanol series (20%, 50%, 70%, 90%, 100%), dried with hexamethyldisilazane 98%, and after complete evaporation, attached to stubs, coated with goldplatinum in a Balzer Union evaporator, and examined under a Philips XL20 scanning electron microscope (SEM). The specimens collected in July 2017 and May and June 2018 were set aside for a parallel chemical analysis (not included in the present paper).
According to Lattig and Martin (2009), the main morphological traits were measured on well-preserved, complete polychaete specimens (from Portofino, Cap of Creus and the Chafarinas Archipelago); 36 specimens for body measurements and six specimens for dorsal cirri length pattern (as number of antennae and cirri articles) were placed on slides with glycerine and measured using a micrometric scale. The inter-population morphometric differences were analysed for size-independent data (by dividing all individual measurements by their respective body width) according to Martin et al. (2017) and Meca et al. (2019). All data were normalised prior to the analyses. Data ordinations were performed by principal component analysis (PCA) and the significance of the obtained clusters was assessed by one-way analyses of similarity (ANOSIM) based on Euclidean distance resemblance matrices, both for the main body measurements and for the dorsal cirri length pattern. The morphometric characters most contributing to the average inter-population differences were estimated and shown as percentages according to Martin et al. (2017) and Meca et al. (2019). PCA and ANOSIM were executed with PRIMER, version 6.1.11, copyright by PRIMER-E Ltd. 2008 (Clarke and Warwick 2001;Clarke and Gorley 2006).
To describe the reproductive characteristics, we estimated the number of chaetigers in all still attached and already detached stolons (except in those from May and June 2018, kept apart for the chemical study, as indicated above). Sex was determined by the type of gametes present, and, when these were not detectable, we considered the number of stolon chaetigers (excluding the single stolon collected in February 2018) according to Martin et al. (2002). Some stolons were prepared for SEM observations. One 27-chaetiger stolon from September 2016 and the posterior portion of a ripe adult from November 2016 were previously fixed in glutaraldehyde (2.5%) for histological analyses, then dehydrated through a graded ethanol series (20%, 50%, 70%, 90%, 100%), embedded in resin (Technovit 8100), and attached to plastic stubs, cut in 5 μm in thick slices with a Histo-line MRS3500 microtome, stained with toluidine blue for 30 s, dried and mounted in Eukitt.

Geographical, temporal, infestation and reproductive variability
The two host chromotypes occurred separately in Costacuti Shoals (yellow) and Portofino, Santa Teresa di Gallura and Molunat (purple) ( Fig. 1   The record with the * has to be considered doubtful symbiont chromotypes yellow, with purple dorsal marks on the yellow hosts and transparent, with a strong purple pigmentation in purple hosts (Fig. 2a, c). All polychaetes (i) showed the typical simple unidentate chaetae of the species, with those in anterior parapodia with slightly bidentate chaetae and those from posterior parapodia being more hooded and more clearly bidentate (Fig. 2b), (ii) were present in all studied months, except January 2018, and (iii) occurred in 65 out of the 192 (i.e. 34%) apical branches of P. clavata collected in Italy and Croatia (Table 1, Electronic Supplementary Material 1). The temporal trends of the infestation in the Portofino Promontory were characterised by average infections of 0.84 ± 0.27 ind/m in spring, 1.7 ± 0.37 ind/m in summer, 0.25 ± 0.03 ind/m autumn and 0.03 ± 0.03 ind/m in winter (Fig. 3). However, only summer infestations were significantly different from those in winter (Kruskal-Wallis, H = 9.08; a b c d e  (Table 3). In turn, the specimens with the longest pharynx (on average) occurred in august 2017 and those with the longest proventriculum in February 2018. Mature adults showed intra-coelomic gametes and had attached growing stolons with 22-24 chaetigers, whilst free stolons showed more than 17 to 30 chaetigers, except one with 14 chaetigers from February 2018 (Tables 2 and 4). The spermatic cysts from a free stolon measured 211.5 ± 44.34 μm in diameter, whereas the oocytes from a developing stolon of a mature polychaete measured 50.75 ± 4.94 μm in diameter (Fig. 2d, e). Considering the number of the chaetigers of the stolons, nearly all ripe specimens were identified as males (Table 4).

Morphometry
The PCA based on morphometric data of the three populations of H. chamaeleon revealed a different trend when analysing the body measurements and the dorsal cirri length pattern (Fig. 4). Despite the population from Portofino were more numerous and thus showed a more disperse distribution in terms of body measurements, they were partially intermixed with those from Cape of Creus and Chafarinas (Fig. 4a), leading to the absence of significant differences amongst them (ANOSIM, Global R = 0.066; p = 0.175). Conversely, the dorsal cirri length patterns (in number of articles) were clearly different (Fig. 4b), thus showing overall significant differences (ANOSIM, Global R = −0.458; p = 0.001). In this case, however, the pairwise tests revealed that the only significant differences occurred between the Portofino population and those of Cape of Creus and Chafarinas Archipelago (Table 5). A detailed analysis of the morphometric measurements revealed that those from the Portofino Promontory population were always smaller than those from Cape of Creus and Chafarinas Archipelago; posterior dorsal cirri of Portofino specimens are shorter than those of the other populations, according to Table 6.

Behaviour
Most polychaetes were paced directly on the surface of the host gorgonian, usually (but not always) amongst the polyps. However, juveniles polychaetes were occasionally observed to directly enter the coelenteron through the polyp pharynx (Fig. 5) and disappear completely inside, without inducing any host reactions or visible tissue injuries.

Discussion
The association of H. chamaeleon with P. clavata was chiefly recorded along the Mediterranean coast of Spain (Martin et al. 2002;Musco and Giangrande 2005), as well as in the Alboran Sea (Baratech and San Martín 1987;López et al. 1996;Martin et al. 2002). Two additional reports, from the coasts of Egypt (Abdelnaby 2014) and from the Arafura Sea (Australia) (Wilson 2006), have to be considered doubtful. Particularly, the latter, likely corresponds to Haplosyllis depressa Augener, 1913, currently accepted as Trypanobia depressa (Augener, 1913) (Aguado et al. 2015). Therefore, our findings represent the first record for Italian and Croatian waters and extend the geographic range of H. chamaeleon to the Tyrrhenian and Adriatic Seas (Castelli et al. 2008;Micac and Musco 2010;Micac 2015). Except for the finding of the polychaete in association with P. grayi in the Iberian Atlantic coasts, our results contribute to confirm the probable overlapping distribution of H. chamaeleon with those of the purple and yellow host chromotypes (Di Camillo et al. 2018b;Pica et al. 2018), which are widely but exclusively distributed in the Mediterranean Sea (Ponti et al. 2019). In Portofino, however, only part of the studied colonies of P. clavata seemed to host H. chamaeleon (although we cannot discard that, in a given individual, the worms could be present in unsampled branches) and the polychaetes were present all year round, with a particularly high infestation in summer. The infestation rates were overall very low, which agrees with previously published data on other known populations (Martin et al. 2002;Lattig and Martin 2009). Morphometrically, the Portofino population were nonsignificantly distinguishable from those of Cape of Creus and the Chafarinas Archipelago in terms of body measurements, whilst they had significantly smaller dorsal cirri. Nevertheless, the explained percentage of difference (expressed by the ANOSIM Global R), is lower than 50%, so that we do not consider this as significant at the species level. More likely this seemed to be caused by the different population characteristics, with that from Portofino being collected during several periods of the year, whilst Iberian ones were collected during more restricted time-ranges (or even in a single season). Therefore, we suggest that the observed interpopulations differences must be considered as intra-specific,    Characteristically, all studied populations showed the presence of ripe adults with attached stolons, with the Portofino population allowing us to confirm for the first time its presence in different periods of the year. Conversely, free stolons were more rarely found on the gorgonian hosts, probably because they leave the host towards the water column for reproductive purposes. Some syllids may have short reproductive periods (Giangrande 1989;Franke 1999), whilst Syllis prolifera (Krohn, 1852) reproduces monthly under an endogenous control and according to a semi-lunar cycle (Franke 1986(Franke , 1999, but shows a longer stolonisation period during the coldest months of the year. In H. chamaeleon, male and female gametes develop within the anterior part (parent stock), as it happens in Haplosyllis spongicola (Wissocq, 1966), and are subsequently relocated to the stolons, where they are stocked in different stage of maturation and for variable time before their release. The fact that the adults started to regenerate the posterior segments after releasing the stolon confirmed that the species had life-spans longer than a year, although the exact duration of their life had not yet been stablished (Martin et al. 2002).
Haplosyllis chamaeleon is chiefly found on the apical portions of the gorgonian branches, where the high density of polyps probably maximises feeding opportunities (Laubier 1960;Martin et al. 2002). As previously reported in Martin et al. (2002), the syllids were observed to directly enter the coelenteron through the polyp pharynx (Fig. 5), likely to steal food, without inducing modifications of the host morphology; moreover, juveniles of our population from Portofino, were also observed completely to disappear inside the polyp, suggesting that they may enter also the stem canals (Fig. 5). This appeared to be a typical kleptoparasitic behaviour (i.e. stealing food captured by the host polyp) (Martin et al. 2002). However, the coincidence in colour morphs between host and symbiont observed in their different populations, including those collected in this study in Italian and Croatian waters, does not allow to discard the possibility of the symbiont feeding on host tissues, a behaviour that has also been reported in this species and in other symbiotic syllids (Pawlik 1983;Magnino and Gaino 1998;Martin et al. 2002;Lattig and Martin 2011). Kleptoparasitism has also been reported for different cnidarian associates. Amongst them, the caprellids Pseudoprotella phasma (Montagu, 1804), living on the hydroid Eudendrium glomeratum Picard, 1952(Bavestrello et al. 1996, and Paracaprella pusilla Mayer, 1890, living on Eudendrium racemosum (Cavolini, 1785) (Ros and Guerra-García 2012) In turn, the nudibranch Cratena peregrina (Gmelin, 1791), associated to E. racemosum, feeds preferentially on hydroids that have recently captured preys, thus combining kleptoparasitism with predation (Willis et al. 2017). In light of our data, we cannot Fig. 4 Graphical representation of the PCAs based on body measurements (a) and dorsal cirri length pattern as number of articles; (b). BL, body length; Phl, pharynx length; Prl, proventriculum length; NC, number of chaetigers; 1 to 10, segments of dorsal cirri. P, Portofino; CC, Cape of Creus; CH, Chafarinas Archipelago Fig. 5 Haplosyllis chamaeleon. Scheme representing the hypothetical behaviour. ax, axis; sc, stem canals; s, sclerites; question mark, hypothetical exploitation of stem canals to move from one polyp to another (likely for juveniles) discard such a combination for H. chamaeleon. It has also been suggested that the presence of the symbiont could generate some benefit for the host, particularly by cleaning its surface (Martin et al. 2002), a possibility that would move the association towards mutualism and that we cannot discard in light of our data. This possibility certainly merits further behavioural observations.
Even though H. chamaeleon probably lives its whole adult benthic stage on the gorgonian host, there was no evidence of branch damage or injury either when being sheltered on the host surface or whilst using the gorgonians to stole planktonic preys from inside their polyps. Also, they did not induce the hosts to generate any malformation (Martin et al. 2002, this work). Conversely, the congeneric H. anthogorgicola excavates galleries in the coenenchyme, inducing the formation of protuberances on the branches, whereas H. villogorgicola forms gall-like structures by bending several branches, which suggested they were in different steps of the evolution of their respective associations (Martin et al. 2002). Nevertheless, the behaviour of these two species was also considered as being kleptoparasitic (Martin et al. 2002). Kleptoparasitisms differ from parasitisms in that the symbionts did not cause direct damage to the host (Dales 1957;Margulis et al. 1982), being thus more closely related to commensalisms (Martin and Britayev 2018). The real status of the association between H. chamaeleon and its gorgonian hosts is still an open question that could possibly be answered by analysing the biochemical composition of both partners or by studying the composition of the faecal pellets produced by the symbionts, but these two interesting approaches were far from the objectives of the present paper, and certainly merit further studies.
Funding Open access funding provided by Università Politecnica delle Marche within the CRUI-CARE Agreement. This research has been performed within the frame of the European Union's Horizon 2020 research and innovation programme, as part of the project MERCES: Marine Ecosystem Restoration in Changing European Seas (grant agreement No. 689518) and is a contribution of Daniel Martin to the Consolidated Research Group on Marine Benthic Ecology of the Generalitat de Catalunya (2017SGR378) and to the Research Project PopCOmics (CTM2017-88080), funded by the 'Ministerio de Ciencia, Innovación y Universidades' of Spain (MICINU), the 'Agencia Española de Investigación' (AEI) and the European Funds for Regional Development (FEDER).

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Ethical approval All necessary permits for sampling and observational field studies have been obtained by the authors from the competent authorities and are mentioned in the acknowledgements, if applicable.
Sampling and field studies All necessary permits for sampling and observational field studies have been obtained by the authors from the competent authorities and are mentioned in the acknowledgements. The study is compliant with CBD and Nagoya protocols..
Data availability All data generated or analysed during this study are included in this published article [and its supplementary information files].
Author contribution CC and DM conceived and designed research. LP, BC, CGDC conducted experiments. CC, DP and DM performed the underwater sampling. LP, BC, CGCD and DM analyzed data. LP, BC and CGDC wrote the original draft. All authors read and approved the manuscript. CC covered the costs of field activities and lab analyses.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.