In vitro proliferation of Mytilus edulis male germ cell progenitors

Our understanding of basic cellular processes has mostly been provided by mammalian cell culture, and by some non-mammalian vertebrate and few invertebrate cell culture models. Developing reliable culture conditions for non-model organisms is essential to allow investigation of more unusual cellular processes. Here, we investigate how cells isolated from different tissues of the marine mussel Mytilus edulis thrive and survive in vitro in the hope of establishing a suitable laboratory model for the investigation of cellular mechanisms specific to these bivalve mollusks. We found that cells dissociated from mantle tissue attached to the culture vessels and proliferated well in vitro, whereas cells isolated from gills, although remaining viable, did not maintain divisions over three to four weeks in culture. We used antibodies against the germ-line marker DEAD-box helicase 4 (DDX4), also known as VASA, and the epithelial cell marker cytokeratin to distinguish different cell types in culture. DDX4-positive cells were predominant in 25-day-old cultures from male mantles. Cells from other tissues remained in low numbers and did not seem to change in composition over time. Overall, the culture conditions described here allow an efficient selection of male germ cells that could be used to study specific cellular mechanisms in vitro.


Introduction
Biological studies of metazoans at the cellular level are mostly limited to vertebrates and a small number of invertebrate models.Yet, many fundamental cell mechanisms existed long before the divergence of Metazoa and may have evolved differently along divergent lineages.A good example of this is the complexity of apoptosis in mollusks that far exceeds the simple apoptotic network described in the model invertebrates Caenorhabditis elegans and Drosophila melanogaster [1].Apoptosis in mollusks is in fact similar in complexity to apoptosis in vertebrates but also has unique features related to their response to environmental changes and pathogens [1].
In recent years, in vitro, cultivated molluscan cells have contributed significantly to many disciplines, including neurobiology, immunology, toxicology, functional genomics, and others.
Most reports of successful attempts to culture molluscan cells indicated that primary cell cultures originating from various tissues, including hemolymph (hemocytes), heart, mantle, digestive gland and gill, can persist for extended periods in vitro, making them suitable systems for basic cell research (reviewed in [2]).Both short-and longer-term molluscan cell cultures have been exploited for advancing the understanding of complex biological phenomena and physiological systems [3].
One biological phenomenon specific to bivalve mollusks that could greatly benefit from cell culture experiments to be better understood is their unique mitochondrial transmission system.Mitochondria are usually transmitted strictly maternally in animal species, but many bivalve mollusks exhibit a very peculiar way of transmitting their mitochondria, as they have been shown to keep the mitochondria of paternal origin in the male germ-cell lineage [4][5][6].This phenomenon has been dubbed Doubly Uniparental Inheritance (DUI) and occurs in more than one hundred species distributed across twelve families of bivalve mollusks [7].The cellular basis allowing paternal mitochondria to be preserved specifically in the germ-cell lineage of male individuals is still largely unknown.A cellular model would be a valuable tool to advance our knowledge of the mechanisms underlying DUI and mitochondrial inheritance in general.A good model system among bivalve mollusks is the genus Mytilus.Mytilus spp.are largely distributed, widely studied as environmental sentinels, extensively cultured in commercial mariculture (or mytiliculture) and they exhibit DUI.Cell cultures from different Mytilus species have been used in both applied and basic research [3].For example, hemocytes and gill cells of Mytilus galloprovincialis have been used to assess cytotoxicity in several assays [8][9][10][11].Longer term cultures were established to look at viability and cryopreservation of M. edulis cells from different tissues [12][13][14][15].Apoptosis, cell differentiation and adhesion were studied in cultures of dissociated larvae cells from M. trossulus [16][17][18][19].Of these, only Daugavet and Blinova [14] used mantle dissociations, focussing on cells from the edge of the mantle therefore excluding gonads (gametogenesis takes place in this tissue in Mytilus spp.), and showed that these cells remained viable for extended periods, but did not significantly proliferate.
Very few studies examined mantle cell and gonadal cell cultures in other bivalve species, although earlier attempts demonstrated that mantle of the sea clam Paphia malabarica was a good source of viable cells in primary culture [20].More recently, viable cell cultures were established from gonad explants of the sea clam with DUI Ruditapes philipinarum [21].These cells proliferated for approximately 15 days before rapidly losing viability.Interestingly, some of these cells expressed the germ-line marker VASA [21].
Here, we determine suitable conditions to establish primary cell culture from Mytilus edulis, selecting cells that attach and proliferate in vitro.We focused on mantle (containing gonad tissues) and gills dissected from male and female individuals to ensure a variety of cell types.To our knowledge, it is the first time that the sex of the individuals from which the tissues are dissected is taken into account in such a study in Mytilidae.

Preparation of substrate
Since we wished to select adherent cells for further manipulations, we treated all culture dishes and coverslips with synthetic amino acids to provide suitable substrate for cell attachment.Initial attempts using Poly-L-Lysine failed to facilitate cell attachment.We thus selected Poly-D-Lysine (Gibco A3890401) and proceeded as follow.Borosilicate coverslips (Fisher Scientific 1254580) were sterilized by incubation in 95% ethanol for 1 hour.They were then washed with sterile ddH 2 O and transferred into 24 well plates.When dried, coverslips were covered with 50 µg/ml Poly-D-Lysine (PDL) solution overnight, washed with sterile ddH 2 O and allowed to air dry.All manipulations were carried out in a class II biosafety cabinet to ensure aseptic conditions.

Sample collection and tissue preparation
Live Mytilus edulis, farmed by Prince Edward Aqua Farm, Canada, were purchased from the market no more than 24 hours before use.Individuals were cleaned of epibionts and washed with fresh water.They were wiped with 70% ethanol, briefly dried and opened with a sterile scalpel.

Tissue dissociation and culture conditions
Basic culture conditions were optimized using mantle tissue dissociations from 10 male individuals.Tested media were composed of either ASW or Leibovitz 15 (L-15) dissolved in ASW, supplemented or not with bovine calf serum (BCS) alone or BCS and yeast extract (YE) (Table 1).Fragmented mantle tissues were incubated in 0.25 % trypsin/EDTA supplemented with 0.03 g/mL of Instant Ocean Sea Salt and rotated gently for 15 minutes to form a cloudy suspension.
To stop trypsinization, L-15 containing BSC and antibiotics was added to the suspension.The suspension was then passed through two cell strainers with 100 μm and 40 μm mesh size, respectively to form a more homogenous suspension and collected into a 50 mL falcon tube.
Fragments on mesh were mashed/pressed with the bottom of a syringe gently and washed with ASW containing antibiotics to allow more cells to pass through the cell strainer.The suspension was then centrifuged for 7 minutes at 300 g at room temperature.The supernatant was discarded, and the cell pellet was washed three times with ASW plus antibiotics.Finally, cells were seeded at approximately 2×10 4 cell/well in Poly-D-Lysine-coated 24 well-plates in triplicate in each culture medium.Plates were incubated at 15±2°C, and 25±2°C.Media was completely renewed every 4-5 days to remove debris and unattached cells.To prevent contamination, Penicillin (100 U/mL), Streptomycin (100 μg/mL), and Amphotericin B (1.5 μg/mL) were added to all medias [23].

Cell viability and proliferation
Cell number was assessed every five days by automatically counting stained nuclei.Nuclei were stained with 1.25 μg/mL Hoechst 33342 (Thermo scientific 62249) and counted from five randomly selected micrographs for each slide every 5 days from day 1 to day 30 using 20X magnification on a fluorescent microscope (EVOS-M5000).Cell number for each well is the average number from the five micrographs.

Sex and tissue comparison
After selecting the best culture medium and dissociation method, we compared viability and proliferation of cells dissociated from male and female mantles and gills.Mantles and gills from 10 males and 10 females were obtained as above and cultured for 25 days at 25°C in L-15 enriched with BCS and YE.Cells were counted as above.

Immunofluorescence
To identify different cell types in culture, immunocytochemistry analyses were conducted on cells cultured from mantle and gill tissues of 3 males and 3 females after 1, 5, and 25 days in culture.
Cells attached to coverslips were washed with 1X phosphate buffered saline (PBS PH 7.

Statistical analysis
Data were expressed as means ± standard deviation.Statistical analysis was done using two-way ANOVA, where needed, using IBM SPSS Statistics 28.0.Post Hoc Test assessed differences between the experimental groups.Differences were considered significant at the p value < 0.05.

Survival and proliferation of M. edulis cells in culture
To determine the best dissociation method to obtain viable cells, mechanical, enzymatic, and a combination of both methods were compared.We used mantle tissue of male mussels, which is abundant and heterogeneous as it contains the gonads in the form of many ducts where gametogenesis occurs.Although the number of cells obtained with mechanical dissociation seemed at first higher, most cultures became rapidly contaminated, and this method was rapidly discarded.Using trypsin to dissociate the cells, with or without mechanical disruption, sporadically yielded contaminated cultures, but in most cases, cultures could be kept free of contamination for several weeks.This is possibly attributable to the enzyme itself, as it has been shown to disrupt biofilms and improve antimicrobial agent efficiency [24].In the end, a combination of both mechanical and enzymatic treatments was preferred.
To assess the best supplementation and culture conditions, six different culture media (Table 1) were tested at different temperatures (15, 18, 20 and 25°C).In all media tested, we observed a steady decline of cell populations when the cultures were kept at 15°C (Fig. 1B).After empirical testing of the different temperatures, it appeared that mantle cells proliferated best at higher temperatures (Fig. 1A and C).At 25°C, the presence of nutrient provided by L-15 was required to maintain cell viability, as all media based on artificial see water alone could not support cell populations.When the media was supplemented with BCS, the number of attached cells increased with time, showing that cells were dividing (p < 0.001; Fig. 1A and C).The addition of both BCS and YE seemed preferable as considerably less variability was obtained in this medium.
In these conditions, cell numbers doubled in 4 to 5 days and reached a plateau after 15 days in culture (Fig. 1A and C).At this point, cells had reached confluency, and only marginal growth was observed.As an alternative source of growth factor supplements, mussel serum was prepared from M. edulis mantle tissue following the method described by Vandepas et al. [23].However, this supplementation was not successful as even a low quantity of mussel serum added to the media resulted in the formation of a high viscosity film on the culture dish, preventing cell attachment and growth.

Proliferation of cells from different tissues
Having selected L-15 enriched with BCS and YE at 25 ˚C, we compared the proliferation of cells dissociated from male and female mantles and gills.Ten samples were obtained for each tissue and sex.Of these, only cells dissociated from male mantles could attach to the surface and establish cultures containing an appreciable number of cells, but viable cell cultures were obtained from all tissues, albeit in low abundance.Both male and female gill cells provided similar results, with a low number of cells attaching to the dish and proliferating poorly (Fig. 2A, B).The proliferation was more variable for female mantle cells, with some cultures thriving, but others showing very low proliferation rates.On average, these cells doubled in approximately 6 to 7 days (from interpolation) and plateaued after 15 days without reaching confluency.Cells dissociated from male and female gills showed only modest proliferation but remained present in culture for 30 days.Figure 2A shows that the number of cells obtained from male mantles was several orders of magnitudes higher than the other tissues.Figure 2B shows the relative growth of attached cells for the different tissues, again showing that male mantle tissue cells thrived better in culture than cells from female mantles or male and female gills.

Cell morphology and markers expression
We used immunofluorescence to detect cells expressing DDX4 (VASA) and cytokeratin to identify what cell types were selected in these culture conditions.DDX4 is an RNA-dependent helicase expressed in animal germ cells and has been widely used as a marker of germ-cell progenitors in a wide variety of species [25].We used a pan-cytokeratin antibody as a marker of epithelial cells.Invertebrate intermediate filament protein expression is not as complex as vertebrates, but epithelial cells have been shown to express proteins sharing significant homology with the conserved regions of vertebrates' type A and B cytokeratin and to react with antibodies against these proteins [26].
We labelled cells cultured from male and female mantles and gills after different times in culture.To assess cell morphology, phase contrast images were also captured.Most cells present in male mantle cultures were small, rounded cells of approximately 2 µm diameter (Fig. 3 A'-F').
Larger round cells ranging from 4 to 6 µm in diameter were also frequently observed.More rarely, elongated fibroblast-like cells attached to the coverslip.In cultures obtained from female mantle samples, small cells were also present, although they were more ovoid and slightly larger (4 µm) than those present in male mantles (Fig. 4 A',   B').Elongated cells like those observed in male samples could also be found in cell cultures of female mantles (Fig. 4 C').Cells obtained from male and female gills were similar to those from the female mantle (not shown).More than half of the cells from male and female gills were large, round cells with 4 to 7 µm diameter.The second most common gill cells were small, round cells up to 3 µm diameter.
The presence of ovoid and elongated cells in male and female gill culture was rare and ranged from 4 to 7.5 µm in diameter.
Cells were identified as type A when positive for both DDX4 and cytokeratin, type B if positive only for DDX4, type C if positive only for cytokeratin or type D when neither marker was present.Tentatively, type A would represent germ cell precursors at early differentiation stages [27], type B more differentiated germ cells [21], type C would represent epithelial cells and type D any other cell type that could be present in the culture.
Most of the small, rounded cells present in male mantle cultures tended to be positive for both DDX4 and pan-cytokeratin in young cultures, although some cells were positive for only one or the other.Larger round cells were also often positive for both markers at early stages.Some elongated cells could be seen by autofluorescence and trans-illumination but expressed neither DDX4 nor cytokeratin.Figure 4 shows that cells obtained from female mantles positive for DDX4 and cytokeratin tended to be larger and more ovoid than male mantle cells (about 4 µm, Fig. 4A   and B).The presence of unlabelled elongated cells was also noted in these cultures (Fig. 4C).
Cells were monitored after 2, 5 and 25 days in culture.Using the type A to D nomenclature, we compared the evolution of cell populations with time in culture.The proportion of each cell types changed with time and the tissue of origin.Type A cells tended to be the most abundant cells attaching to the coverslips after two days in culture, representing between 35 and 75% of all cells present (Fig. 5).In male gills, type A cells were slightly more abundant (35%) than type D, with no staining (30%).After 5 days, type A cells remained the most abundant in cultures of male and female mantle cells, representing 70% and 63% of cells, respectively (Fig. 5A, B).In gill cells from both sexes, types C and D were more abundant by Day 5, representing 25 and 48% of cells in samples from males and 40 and 26% of cells in samples from females (Fig. 5C, D).After 25 days in culture, the proportions of type B cells expressing only DDX4 and type A cells expressing both markers had shifted in cultures originating from male mantle cells, with 26% type A and 66% type B (Fig. 5A).This shift did not occur in cultures originating from female mantle cells nor from gill cells of both sexes, the proportions of each cell type remaining more or less the same (Fig. 5B-D).To assess the proliferation of each cell type, we expressed the same data relative to the number of cells of each type present at the first sampling time, that is Day 2 (Fig. 6).As time in culture continued, cells only positive for DDX4 showed an important fold increase compared to the cells expressing cytokeratin and cells expressing neither marker at days 5 and 25 in cultures from male mantle tissues, which suggests that the culture conditions used favoured the proliferation of these cells (Fig. 6A).In cells cultured from male and female gill tissues, all cell types seemed to proliferate at a similar rate and showed a modest increase (3 to 6-fold, Fig. 6 B-D).This proliferation rate is reminiscent of the overall proliferation observed when counting all cells (Fig. 2).We also compared the morphology of the most abundant cells expressing only DDX4 in older male mantle cultures (i.e.type B more differentiated germ cells) to mature sperm cells.
Specifically, we stained cells dissociated from male mantle that had been cultured for 44 days and spermatozoa collected from mature males with DDX4 and ATP5, a mitochondrial marker (

Culture conditions and cell proliferation
We tested different cell dissociation methods, as well as different media compositions at different temperatures using M. edulis male mantle samples to select the best dissociation method and conditions for primary cultures of mussel cells.The mantle is a sheet of dorsal epidermis extending along both shells in bivalve mollusks.It is thought to fulfill several functions, including the secretion of calcium carbonate to generate the shell and sensory functions.It also contains muscle cells [28].In Mytilus, the mantle comprises connective and gonadal tissues [29].It thus contains many different cell types, of which some were shown to proliferate in vitro [16].For example, mantle cells were maintained viable for up to 22 months in L-15 dissolved in artificial seawater supplemented with 2% bovine serum and kept in sealed plates at 10°C without changing the medium.These cells attached to the bottom of the dish after approximately 6 days and remained viable but did not significantly proliferate [14].The present study found that mantle cells obtained using a combination of both mechanical and enzymatic dissociation rapidly attach to the matrix.
The addition of BCS was essential to sustain proliferation, and yeast extract provided stability to the culture media, reducing the variability between cultures.Yeast extract was previously used in culture media as a source of vitamins and other supplements [30].It is thus reasonable to presume that its addition complements the L-15 formulation and provides unknown useful components for mussel cell maintenance [31].
It was somewhat surprising that the proliferation of mantle cells in culture could only be obtained at relatively high temperatures (room temperature up to 25°C).All cultures kept at 15°C, cells proliferating in male mantle cells.The culture system we put forward here thus clearly supports M. edulis germ cells' selection, proliferation, and possibly differentiation.
The arrangement of mitochondria on one pole of the cells and the mitochondrial localization of DDX4 suggests that cells proliferating in male mantle cultures could be at an intermediate/late stage of sperm differentiation.Indeed, in the earliest stages of spermatogenesis until early spermatids, the mitochondria in Mytilus spp.are rather uniformly distributed throughout the cytoplasm.It is only from the mid-spermatid to the late-spermatid stage that mitochondria are reduced to five in number, and restricted in position to one juxtanuclear region which will eventually become the sperm middle-piece [36].Moreover, previous studies showed that the mitochondria of maturing sperm cells in bivalve species uptake DDX4 (VASA).
That said, small, rounded cells positive for both DDX4 and pan-cytokeratin in young male mantle cultures also seemed to present this arrangement of mitochondria on one pole of the cells (Fig. 3), suggesting that Mytilus male germ cells may express keratin until the spermatid stage or during the spermatid stage.For example, certain forms of keratins are expressed during mammalian spermatogenesis, with one group present in spermatogonia, spermatocytes, and spermatids.In contrast, another form is expressed only during the elongation and condensation of the spermatid nucleus [37].With this in mind, one could speculate that most of the small, rounded cells positive for both DDX4 and pan-cytokeratin in young Mytilus male mantle cultures would be mid-spermatids while cells expressing only DDX4 that became largely dominant after a few weeks would be late-spermatids.It would be interesting to determine if they could further differentiate into spermatozoa, given adequate conditions.
Indeed, male germ cells in both vertebrates and invertebrates can differentiate in vitro to produce motile sperm under certain conditions [38][39][40][41].In vitro spermatogenesis has been achieved in several types of fish, and it was found that the addition of 11-ketotestostrone to the culture medium was essential to stimulate spermatogonial stem cell mitosis and complete spermatogenesis [42].Other studies showed that the cocultivation of isolated germ cells with feeder cells mimicking Sertoli cells also supported successful spermatogenesis [43][44][45].In mice, fetal bovine serum was indispensable to induce the differentiation of spermatogonia to haploid round spermatids but also suppressed the formation of elongating spermatids or spermatozoa.
Using neonatal mouse testis explants and serum-free culture media, spermatids and sperm were obtained that resulted in healthy and reproductively competent offspring through microinsemination [46].Presumably, germ cells obtained from M. edulis mantle could be brought to full sperm maturation by modifying the media supplementation to support spermatogenesis better.

Conclusion
We have tested a number of conditions to establish the primary culture of cells extracted from different tissues in Mytilus edulis.Cells from female and male mantles and gills showed a moderate level of proliferation and remained viable for several weeks.Different cell types were present, and in vitro conditions seemed to favour epidermal and unknown cell types based on cytokeratin expression, except in preparations obtained from male mantles.Cells extracted from male mantles provided the best proliferation rate.Based on DDX4 expression, they differentiated into germ cells and proliferated well in vitro.This culture system can thus provide viable and proliferative cells that may be used to explore spermatogenesis in bivalve mollusks and fundamental questions regarding the unique system of doubly uniparental inheritance of mitochondria in bivalves.

Figure 1 .
Figure 1.Male mantle cell viability and proliferation in different culture media.A) Example

Figure 2 .
Figure 2. Average number of cells per microscopic frame (0.3 mm 2 ) from dissociated male

Figure 3 .
Figure 3. Different cell morphologies and expression of DDX4 and cytokeratin in cells from

Figure 4 .
Figure 4. Different cell morphologies and expression of DDX4 and cytokeratin in cells

Figure 5 .
Figure 5. Change in markers expression over time in cultures originating from mantle and

Figure 6 .
Figure 6.Fold change in the number of cells expressing DDX4 and cytokeratin over time in Fig 7).Mature spermatozoa inMytilus harbor a characteristic group of five mitochondria arranged in circle at the base of the flagella.These can be seen in the micrograph of Fig 7 B and C, along with the flagella and acrosome of mature sperm cells.Whereas DDX4 positive cells in mantle culture show the same peculiar arrangement of mitochondria on one pole of the cells (Fig 7A), they are not fully differentiated sperm cells as they do not have a flagellum, nor acrosome.However, it is reasonable to suppose that they have differentiated into spermatids (see Discussion).

Figure 7 .
Figure 7. Mantle cells and spermatozoa stained with DDX4 and ATP5.Light micrographs and