*3.4. 3rd Mariculture Trial: Assessment of Sponge Culture Productivity Polluted vs. Pristine Site*

During the first week of the experiment, 69 (polluted) and 70 (pristine) out of 200 explants dropped off the PVC plates. Fortunately, the PP plates that were placed under the PVC plates were able to catch 55 (at polluted site) and nine (at pristine site) of these explants, which attached onto the PP plates and continued to increase their surface area. Because they could not be related anymore to their original size, explants that were attached on the PP plates were left out of the surface area increase analysis. However, they were included in calculation of survival rates, which were highly different between the pristine and polluted sites (39–86%, respectively—Table 1 (c)). A total of 61 explants survived on the vertical PVC plates (30 explants on 15 plates at the pristine site, 31 explants on 16 plates at the polluted site), which were used for surface area increase analysis. The average increase in surface area over time of these *C. reniformis* explants is presented in Figure 6. After being cultured for 13 months, the average surface area increase was 79.0 ± 37.4% at the pristine site and 170.4 ± 109.1% at the polluted site (Table 1 (c)). Both culture site and time had a significant main effect on sponge surface area increase rates (Table 2). At both sites, explant surface area increased significantly, but it slowed down after the first six months at both sites (two-way factorial ANOVA, F1,27 = 55.550; *p* < 0.001), and for the pristine site even stalled after six months. Surface increase was significantly higher at the polluted site as compared to the pristine site (two-way factorial ANOVA, F1,27 = 14.439; *p* = 0.001), irrespective of time.

For *C. reniformis*, a highly significant correlation between surface area and wet weight was found (Pearson correlation, *r* = 0.92, *n* = 20, *p* = 0.000, two-tailed, Supplementary Figure S2), as well as between surface area and volume (*r* = 0.92, *n* = 20, *p* = 0.000, two-tailed, Supplementary Figure S3). The relationships are size-independent, leading to fixed conversion factors of 1.2 g wet mass per cm2 of surface area and 1.1 cm3 sponge volume per cm<sup>2</sup> surface area, respectively.

**Table 2.** Two-way mixed factorial ANOVA, demonstrating main and interactive effects of culture site and time on *C. reniformis* growth rates (*n* = 15–16).


\*\* Indicates significant effect (*p* < 0.01).

**Figure 6.** 3rd mariculture trial. Annual growth rate as surface area increase for *C. reniformis* (*n* = 15–16 plates) explants in polluted and pristine sites.

#### **4. Discussion**

This study explored the feasibility to integrate fish culture with a biomedically promising Mediterranean sponge species, *C. reniformis*. The main aim of the study was to derive the best mariculture practices of *C. reniformis* from a series of subsequent culture trials.

#### *4.1. Explant Survival Rates*

Survival of explants can be compromised by detachment and by disease. In terms of initial survival, cable-ties and chicken wire were the most effective means of attaching explants onto PVC substrates, with glue giving a slightly lower survival. In the long term, however, the use of chicken wire (mesh culture) gave ambiguous results. Sandwiched mesh structures were designed to promote the explants to grow out of the pocket and to ease harvesting [8]. Mesh culture that is used in turbid waters might reduce water flow and subsequently decrease available food for the explants if mesh size is too small [10]. Although the mesh size used in the second trial was sufficiently large (5 × 5 cm) as recommended in [41], after some time the space between meshes and the PVC plate was covered by epibionts, and the mesh did not prevent some explants from moving or even dropping themselves off the plate. Despite these drawbacks, the survival rate at the polluted site after one year was 79%, which is higher than in the study by [8], who reported 55% survival after seven months and who lost entire *C. reniformis* explants with the sandwiched mesh method. However, the mesh method is labor intensive, especially when considering that increased cleaning of biofouling on the mesh is recommended. By attaching the explants with glue in the third trial, it was anticipated to reduce both handling time and fouling. Despite the predicted improvements regarding initiation time (May vs. June) and culture angle (all at 90◦), the third trial showed low survival for the pristine site. This was probably due to occasional strong currents that prevail at this site, which may make the explants more prone to dropping of the plates and physical removal from the site. During the whole month of September 2013, flow velocities above 20 cm/s were recorded at this site by analyzing the velocity of neutrally buoyant particles (video clips of laterally moving natural particles, data not shown). *Chondrosia reniformis* inhabits both nearly stagnant to occasional high flow waters (M. Gokalp; personal observation), however the attachment of explants to PVC plates is probably less firm than attachment to natural substrates, especially during the acclimatization time after wounding them to explant the parent sponges. At the polluted site, the use of glue instead of chicken wire did slightly improve long-term survival rate, which shows that gluing is a suitable method to attach explants of *C. reniformis*. It is also the fastest and easiest method. A future recommendation is to perform the initial acclimatization (of 7–10 days; see [42]) at a more secluded site, after which the attached explants are placed at the study site.

During culture Trial 2, initial sponge survival was compromised by disease-like phenomena. Bacterial infections, which were possibly due to late seeding of explants in Mid-June with relatively higher water temperatures, might have been responsible for the initial losses at both sites. High water temperatures in summer have been reported to be a risk for sponge mariculture in temperate and subtropical climates [11,39,41,43], as it makes cuttings more vulnerable to bacterial attack, although such increased vulnerability had not been observed in our earlier studies on this species in this area [27].

Culture angle directly affected explant survival, mainly in association with prevailing light levels. Lower light levels at the more turbid polluted site may, therefore, also explain the higher explant survival at the light-exposed angles at the polluted site. These results corroborate the findings of [35], which purport *C. reniformis* prefers shaded habitats.

#### *4.2. Explant Growth*

Since surface area of *C. reniformis* showed a size-independent relationship with wet weight and volume, surface area can be used as a proxy for growth. This enables a direct comparison of growth data obtained for this species using different methods.

Culture of *C. reniformis* has been considered to be difficult, to even unsuitable with the methods applied [5,8]. Wilkinson and Vacelet [35] reported moderate growth rates of 95% per year (55 weeks doubling time in volume, measured using volume displacement) when *C. reniformis* was cultured under shaded conditions. Ref. [27] obtained grow rates of 100 to 200% per year when growing *C. reniformis* on the bottom of metal wire cages under pristine conditions, but this study failed to achieve such results at a fish farm site as the explants cultured were smothered by effluents from the fish farm. Conversely, the current study demonstrates that if cultured using an appropriate method, *C. reniformis* will survive and grow (up to 170% in 13 months), even in a fish farm environment with a considerable particle load. These growth rates are considerably higher than those reported for naturally growing specimen. Garrabou and Zabala [36] reported an in situ growth rate of 2.3% per year (deduced from two-dimensional (2D) areal growth) for *C. reniformis*, which was an order of magnitude lower than the growth rate of three other Mediterranean sponge species in their study *Hemimycale Ccolumella* (Bowerbank)*, Oscarella lobularis,* and *Crambe Crambe* (Schmidt). They ascribed the slow growth rate of *C. reniformis* to a greater energy investment in tissue production per unit area as a result of its thick collagenous cortex. However, the data found by Osinga et al. [27] and those from the current study indicate that in aquaculture, *C. reniformis* exhibits growth rates that are nearly two orders of magnitude higher than the in situ rates reported by Garrabou and Zabala [35]. Under optimal circumstances, the production of collagen is apparently not hampered by energy input. The current results show a clear potential for collagen production through the aquaculture of *C. reniformis*. The highly variable growth of *C. reniformis* under different conditions and the high variability within treatments highlight the need for further optimization studies.

During Trial 3, *C. reniformis* surface area increase rates were significantly different between culture sites, with an approximate two-fold higher growth at the polluted site. This may relate to the higher food availability—i.e., higher TOC concentration as a result of fish farm activities—and, as mentioned earlier, correspondingly lower light levels at the polluted culture site. Hence, the combination with fish farming is potentially beneficial for culture success of this sponge species. The surface area increase of *C. reniformis* was clearly higher in the first six months after initiation of the cultures, regardless of culture site. Although this may partially be explained by seasonal effects (growth might cease in autumn and winter [11,44]), it is possible that the sponges exhibit lower specific growth rates when being in culture for a longer period [39]. This could be due to initial enhanced surface area increase due to explant cutting [45], which could hamper growth at later stages, due to high costs of wound healing and regeneration [42]. Fast initial surface area increase was also found in a side experiment where the explants were cultured starting in autumn 2013 (data not shown), pointing towards a wound healing and regeneration effect rather than a seasonal effect, but this observation needs to be further investigated.

#### *4.3. Culture of Chondrosia reniformis—Best Practices*

As stated in Schippers et al. [11], initial mariculture trials should span a complete annual cycle in order to perceive effects of seasonality, substrate preference, and growth physiology of the sponge, and possible external impacts to the culture site, such as the occurrence of fouling and specific sponge predators, boat traffic and anchoring, and the presence of fishermen and divers. Accordingly, in this study, valuable information was acquired regarding the preferences for attachment, survival, and growth of *C. reniformis* during the first two trials. *Chondrosia reniformis* explants attached to PVC plates tend to move on the plate, thus obfuscating multiple genotypic comparisons on one plate (our study was initially designed to investigate genotype effects, but this part of the study could not be completed due to random movement of the explants over the plates). In addition, fusion and fission of explants makes the proper assessment of survival difficult. Even though attachment to the PVC plates has succeeded, some *C. reniformis* still found ways to divide their body into several parts, moved around the PVC plate (possibly in pursuit of shaded areas), or dropped themselves to the ground possibly in search for better living conditions. Survival and growth is best at culture angles of 90◦ and above, where explants are not being exposed to direct sunlight, as *C. reniformis* performs better at low illumination levels.

In experiments 2 & 3, the initial losses and/or droppings of explants were slightly high and unpredictable, despite the variety of methods applied. Once attached for a longer time, the explants would remain attached. Therefore, initial losses and/or droppings are the main problem to be solved to secure better culture performance. Restraining bacterial attack on freshly cut explants by initiation of cultures early in the season (spring) and preventing exposure to high currents during the first months should be practiced together with the best performing methods regarding attachment.

Based upon the three mariculture trials described above, the following best practices have been deduced for culturing *C. reniformis* in sea-based aquaculture under turbid conditions:

