Improved Media Formulations for Primary Cell Cultures Derived from a Colonial Urochordate

The cultivation of marine invertebrate cells in vitro has garnered significant attention due to the availability of diverse cell types and cellular potentialities in comparison to vertebrates and particularly in response to the demand for a multitude of applications. While cells in the colonial urochordate Botryllus schlosseri have a very high potential for omnipotent differentiation, no proliferating cell line has been established in Botryllus, with results indicating that cell divisions cease 24–72 h post initiation. This research assessed how various Botryllus blood cell types respond to in vitro conditions by utilizing five different refinements of cell culture media (TGM1–TGM5). During the initial week of culture, there was a noticeable medium-dependent increase in the proliferation and viability of distinct blood cell types. Within less than one month from initiation, we developed medium-specific primary cultures, a discovery that supports larger efforts to develop cell type-specific cultures. Specific cell types were easily distinguished and classified based on their natural fluorescence properties using confocal microscopy. These results are in agreement with recent advances in marine invertebrate cell cultures, demonstrating the significance of optimized nutrient media for cell culture development and for cell selection.


Introduction
Countless attempts have been made to develop cell cultures from marine invertebrates. Since the 1960s, over 500 peer-reviewed publications have been published on this topic, with the main focus being on six phyla, which are Porifera, Cnidaria, Crustacea, Mollusca, Echinodermata, and Urochordata [1][2][3][4][5][6][7]. However, until recently [8], not a single continuous cell line was established for any aquatic invertebrate taxon, as all efforts to obtain lasting proliferating cultures from marine invertebrates have inexplicably failed [1,2,9], despite the increasing demand of these cultures for a wide range of applications [5][6][7][8]10,11]. Further, it has been repeatedly shown that primary cultures of marine invertebrate cells cease dividing 24-72 h from onset [1,2]. It is worth noting that a substantial number of unsuccessful attempts to establish primary cell cultures from marine invertebrates are not reported in the scientific literature, resulting in incomplete or fragmented knowledge of effective in vitro protocols [7,12], and researchers often end up revisiting unsuccessful methods and futile experimental protocols [1,2]. Considering the above discussion, a recent advancement in sponge cell culture has emphasized the significance of developing optimized nutrient media for the development of primary cell cultures [4][5][6][7][8][9][10][11][12][13], a research approach that has led to the succeeded establishment of long-lasting cultures from several sponge species [8,13].
with fresh medium. The criterion for experiment termination was ≤ 50% viability of total recovered cells.

Blood Cell Characterization
We used 12 colonies in blastogenic stages A-D (A = 3, B = 3, C = 3, and D = 3). Prior to cell extraction, each colony was photographed under a fluorescent microscope (Leica M205 FCA equipped with a camera Leica DFC9000 GT) using four channels (blue: Ex 325-378 nm, Em 438-485 nm; green: Ex 450-490 nm, Em 500-550 nm; red: Ex 540-580 nm, Em 593-667 nm; and far red: Ex 590-650 nm, Em 663-737 nm). Images were obtained using Leica software (LAS X version: 3.6.0.20104). Post cell extractions, the blood cells in ASW were imaged in 35 mm glass-bottom dishes (Ibidi, Cat. No. 81218) using a laser scanning confocal microscope (Zeiss LSM710) with four channels (blue: Ex 405 nm, Em 453 nm; green: Ex 488 nm, Em 536 nm; red: Ex 561 nm, Em 607 nm; and far red: Ex 633 nm, Em 697 nm), and images were obtained using Zen 2.3. The intensity of the fluorescent signal per cell was calculated using Fiji software (http://imagej.nih.gov/ij (accessed on 2 May 2023)). Each image was inverted to a grey scale followed by the measurement of the light signal emitted by cells in relative fluorescent units (RFU). For cell identification in cultures, primary cultures were sampled, and cells were stained with fluorescent dye Hoechst 33342 (Thermo Fisher, Cat. No. H3570). Then, cell pellets were washed with ASW (X3), followed by centrifugation at 1000× g for 10 min, and stained with DiD (Vybrant, V22887) for membrane labeling. Cell pellets were washed (three times) with ASW and centrifuged (1000× g, 10 min). The cells were then seeded on glass coverslips (coated with 0.01% Poly L Lysine solution, Sigma-Aldrich P4707, according to the manufacturer's instructions), and inserted at the bottoms of 12 well plates, left to adhere for approximately 4 h. Then, each well containing stained cells was fixed for 30 min with 4% paraformaldehyde (PFA; Thermo Fisher Scientific, Cat. No. 043368.9M) at room temperature, coverslips were mounted on cover glass slides, and cells were observed and photographed under an epifluorescence microscope (Zeiss AxioImager2, Zeiss, Oberkochen, Germany). Images were obtained using Zen 2.3.

Hematoxylin and Eosin (H&E) Staining
The H&E staining protocol was performed on blood cell samples that were seeded on glass coverslips (coated with 0.01% Poly L Lysine solution) at the bottoms of 12 well plates and left to adhere. Then, each well was fixed with 4% PFA. Mayer's Hematoxylin solution (Sigma; Cat. No. MHS32) and 0.5% Eosin Y-solution (Sigma, Cat. No. 1.09844.1000) were employed according to the manufacturer's instructions. Coverslips were mounted on cover glass slides, and cells were observed and photographed under an epifluorescence microscope (Zeiss AxioImager2) with bright field. Images were obtained using Zen 2.3.

Immunofluorescence Staining
Blood cell samples were seeded on glass coverslips that were inserted at the bottom of 12 well plates and coated with 0.01% Poly L Lysine solution and left to adhere. Each well was then fixed with 4% PFA. Then, samples were washed two times with PBS, were permeabilized in 0.1% Triton x-100 in PBS for 10 min at 4 • C, and were re-washed two times with 0.02% Tween-20 in PBS. Nonspecific binding sites were blocked by incubation in 5% bovine serum albumin diluted in 0.02% Tween-20 in PBS for 60 min at 4 • C. The samples were incubated overnight at 4 • C with primary anti PCNA (Proliferating Cell Nuclear Antigen) antibody developed against rabbit (Abcam, ab18197, 1:1000), washed with 0.02% Tween-20 in PBS (2 × 10 min), and incubated with secondary antibodies (Invitrogen, Alexa fluor 568 donkey anti rabbit, Cat. No. 10042, 1:1000) for 120 min at 4 • C. The samples were then washed twice with PBS, and the coverslips were mounted on slides using Fluoromount-G with DAPI (Invitrogen, Cat. No. 00-4959-52, Waltham, MA, USA). Negative controls for each experiment were established on coverslips that were incubated with blocking solution, lacking primary antibodies, and were exposed only to secondary antibodies. Cover slides were photographed under an epifluorescence microscope (Zeiss AxioImager2). Images were obtained using Zen 2.3. The counting of total stained cells was conducted using Fiji software (http://imagej.nih.gov/ij (accessed on 2 May 2023)).

Statistics
Statistical analyses were conducted on the cell types using SPSS V16. A one-way ANOVA test using a post hoc comparison (Bonferroni and Tukey HSD) was applied on the cell types, including PCNA + cells, at four time points (day 0, 24 h, days 3, 8).
Isolated haemocytes of B. schlosseri (from blastogenic A-D colonies) and thraustochytrids were subsequently identified using histological (H&E), nuclear (Hoechst), and membrane (DiD) staining ( . The haemoblasts (4-6 µm; Figure 3a,b) are small, brown, spherical blood cells. They contain a circular nucleus positioned in the middle of the cell and a nucleolus (often non-recognizable) stained ( Figure 4a) in blue while surrounded by a thin layer of basophilic cytoplasm (Figure 5a-d). The macrophage-like cells (10-20 µm) are large ameboid phagocytes, marked (using H&E) as orange, red, pink, and blue colored cells, with an elliptical nucleus at the periphery of the cell and recognizable brown, yellow, and black vacuoles (Figure 3c,d) that occupy most of the cells' volumes ( Figures 4b and 5e-h). Morula cells (7-11 µm) are spherical in shape, with a barely recognizable small nucleus (2 µm) situated at the periphery of the cell, possessing diverse brown, yellow, and black vacuoles (Figure 3e,f). H&E stained these cells in orange, pink, and blue colors, with recognizable vacuoles that occupy most of the cell volume (Figures 4c and 5i-l). Granular amoebocyte cells (7-15 µm) are oval shaped, containing a small nucleus (2 µm) and micro/macro granules, with vacuoles stained with orange, pink, red, and blue colors (Figures 4d and 5m-p). The pigment cells (7-20 µm) are spherical and granular in appearance and contain brown granules of varying sizes, with colors diverged from dark blue or brown to black that exhibit Brownian motion (Figure 3g,h). These cells were stained (H&E) as pink, red, and blue colors, and the round nucleus appeared in different positions within the cell (Figures 4e and 5q-t). The nephrocytes (7-20 µm) share similar staining character-  We further observed two distinct life stages of thraustochytrid cells, which may appear as cell clusters after nine days in vitro. The mononucleated cells (4-10 µm) are spherical and smooth, containing a small, hardly recognizable nucleus (1-2 µm). These cells appeared in clusters (up to 200 µm) composed of at least 3 cells, each. H&E staining marked the cell cytoplasm as pink, and the membranes surrounding the cell were stained with a thin blue line (Figure 4g). The sporangia cells (10-200 µm) are dark brown, spherical cells with a grainy texture and a barely recognizable nucleus. Stained (H&E) cells appeared pink with patterns of blue (Figure 4i).

Cultivation of B. schlosseri Blood Cells
We used 40 B. schlosseri colonies (7-15 zooids per colony) in blastogenic stages A-D (Table 1). Blood cells were extracted from the marginal ampullae and washed with one of three types of washing solutions (WS1, WS2, WS3; Table 1), were subjected to specific antibiotic combinations, and were then cultured in suspensions of one of five medium variants (TGM1-TGM5). In total, 15 experiments were carried out with high cell viability observed at onset (87.08-94.65%; Table 1). Cultures were observed every second day for up to 26 days (Figure 6), and four time points (0, 24 h, days 3, 8) were specifically studied by counting cell types ( Figure 7) and assessing the proliferative status using PCNA (Figures 8 and S1-S4).   nucleus (2 μm) and micro/macro granules, with vacuoles stained with orange, pink, red, and blue colors (Figures 4d and 5m-p). The pigment cells (7-20 μm) are spherical and granular in appearance and contain brown granules of varying sizes, with colors diverged from dark blue or brown to black that exhibit Brownian motion (Figure 3g,h). These cells were stained (H&E) as pink, red, and blue colors, and the round nucleus appeared in different positions within the cell (Figures 4e and 5q-t). The nephrocytes (7-20 μm) share similar staining characteristics as pigment cells. These cells also contain brown and yellow granules ( Figure

Primary Cultures: TGM1 Medium
Cell cultures were studied for 18 days (Figure 6a) with minimal contamination events (just two [fungi] out of 17 plates, on days 12 and 14; Figure S5f). At the beginning of the study (day 0; Figure S5a), one-way ANOVA revealed significant differences (p < 0.001) in the distributions of cell types, which formed three distinct groups (morula; macrophagelike and pigment cells; granular amoebocyte cells; Figure 7a Figure S5d), only two distinct groups of PCNA + cells were recorded (p < 0.05; haemoblasts and nephrocyte cells, respectively; Figures 8p and S4a). Cell viability was reduced from >90% to 84% at days 4-6 ( Figure 6f), with cell populations dominated by macrophage-like cells, morula cells, and pigment cells. From day 10 to 18, cell numbers further decreased from 1.99 × 10 6 ± 0.14 to 0.83 × 10 6 ± 0.16 cells/mL −1 , respectively, and a sharp decline of viability was observed from 78.17% ± 1.5 to 48.4% ± 0.8, respectively (Figures 6a,f and S5e). We further observed two distinct life stages of thraustochytrid cells, which may appear as cell clusters after nine days in vitro. The mononucleated cells (4-10 μm) are spherical and smooth, containing a small, hardly recognizable nucleus (1-2 μm). These cells appeared in clusters (up to 200 μm) composed of at least 3 cells, each. H&E staining marked the cell cytoplasm as pink, and the membranes surrounding the cell were stained

Primary Cultures: TGM3 Medium
Cell cultures were studied for 12 days (Figure 6c) with no contamination. At day 0 ( Figure S7a), significant differences (p < 0.001; one-way ANOVA) were recorded in cell type distributions that formed three distinct groups (macrophage-like cells and morula cells; haemoblasts and pigment cells; nephrocytes and granular amoebocyte cells; Figure 7c Figures 8r and S4c). Along this period, the only observed change in cellular morphologies was of storage cells (pigment and nephrocyte cells) that transformed from oval to elongate structures, which is consistent with the results of Rinkevich and Rabinowitz [29]. Cell viability was reduced from >84% to 77% at days 4-6 ( Figure 6h), with cell populations dominated by haemoblasts, macrophage-like cells, morula cells, and pigment cells. From day 10 to 12, cell numbers further decreased from 0.6 × 10 6 ± 0.1 to 0.53 × 10 6 ± 0.1 cells/mL −1 , and a sharp decline of viability was observed from 55.3% ± 3.4 to 46.8% ± 2.7 (Figures 6c,h and S7e).

Cell Types Changes in Primary Cultures-An Overview
The distribution of cell types varied among the five media and changed over time (Figure 7, Table 2). Initially (day 0), cell type distribution profiles exhibited a high degree of similarity, with macrophage-like cells and morula cells being the predominant cell types in all media. After 24 h, macrophage-like cells and morula cells remained the most prevalent cell types, but the presence of pigment cells (TGM3, TGM5) and haemoblasts (TGM5) was also observed. By day three, alterations in the distribution of cell types were noticed, with morula and pigment cells being more prominent in the TGM1 medium and macrophagelike cells and pigment cells being more prevalent in the TGM2 medium. However, the abundance profiles of TGM3, TGM4, and TGM5 media remained similar to that of the 24 h time point. By day 8, no further changes in cell type distribution were detected, and the abundance profiles of all five media were almost identical to those of day three. PCNA + activity varied among the five media and changed over time ( Figure 8, Table 2). Initially, haemoblasts and macrophage-like cells were the most actively proliferating cell types in all media, with an additional case of PCNA + activity in pigment cells in the TGM4 medium. After 24 h, we observed PCNA + cell activity in haemoblasts, macrophage-like cells, and morula cells in TGM1 medium; macrophage-like cells in TGM2; haemoblasts and macrophage-like cells in TGM3; haemoblasts, macrophage-like cells, and pigment cells in TGM4; and haemoblasts and macrophage-like cells in TGM5. On day three, the most actively proliferating cell types varied by medium: haemoblasts and macrophagelike cells in TGM1; pigment cells in TGM2; haemoblasts in TGM3; haemoblasts and pigment cells inTGM4; and haemoblasts and macrophage-like cells in TGM5. By day 8, haemoblasts were the most actively proliferating cell types in TGM1, TGM4, and TGM5 media, while morula cells and pigment cells were the most actively proliferating in TGM2 and TGM3, respectively.

Discussion
The world's oceans and seas host hundreds of thousands of animal species, primarily invertebrates, with many types of cells that exhibit a wide range of cellular potentialities [38] and offer endless applications. Yet, while numerous cell lines have been commonly derived from vertebrates and terrestrial invertebrate taxa, e.g., insects and arachnids, until just recently [8], all efforts to develop cell lines from marine invertebrates have failed, despite many attempts made on a wide range of species over the past decades [1][2][3]6,7,9]. However, to establish a sustainable pharmaceutical industry based on the "blue economy", the most viable approach is (a) to cultivate cells under controlled conditions by creating widespread "cell factories", still an unaccomplished objective, and (b) to scale up cell cultures from marine organisms in bioreactors. These evolving stages not only allow the production of large supplies for any needed bioactive material, but they further make available the manufacturing of a diverse range of novel industrial bioproducts. The recent advances in marine invertebrate cell culture methodology supported by the development of optimized nutrient media for primary sponge cell cultures [4,13] have led the authors to establish long-lasting cultures from several species of sponges [8,13], revealing the importance of fine-tuning the medium for successful outcomes.
Following the importance of B. schlosseri as a model system in a wide range of biological disciplines [14,15,[39][40][41] and the need to develop approved in vitro methodologies for its research (general approach in Rinkevich and Rabinowitz [36]), several research endeavors have centered around obtaining high cell yields; identifying appropriate factors for proper cell adhesion and attachment [30,32]; initiating primary cell cultures from embryos, epithelial cells, and circulating blood cells [10,17,18,29]; revealing impacts of media additives on primary cultures, such as growth factors [29]; establishing a defined medium for circulatory blood cells [29]; and evaluating the in vitro delayed stemness of extirpated colonial organs, including the emerged stemness signatures in epithelial monolayers [30,34,35]. All attempts to establish a proliferating cell line from Botryllus cells have been unsuccessful, with cells dividing for a brief period of 24-72 h post extraction [1,2,17,29]. To address these difficulties, the present study aimed to investigate the response of various types of B. schlosseri blood cells (haemoblasts, macrophage-like cells, granular amoebocytes, morula cells, pigment cells, and nephrocytes; [42,43]) under in vitro conditions. Specifically, we assessed cell type-specific responses using five versions of a basic medium during the initial few crucial weeks after initiation. To aid in cell identification, we utilized confocal microscopy and the differential autofluorescence of various cells, which provided new insights. For example, our findings indicated that haemoblasts were the only B. schlosseri cell type to exhibit a high signal in the far-red channel, which distinguishes them from thraustochytrid cells, and they lacked blue channel fluorescence, supporting Rinkevich and Rabinowitz's [29] previous findings.
The different media variations (TGM1-TGM5) had distinct impacts on primary cultures of B. schlosseri blood cell types, reflected as changes in cell proliferation, viability, and dominant cell types. With regards to cell proliferation, we showed that the medium may stimulate the proliferation of distinct circulating cell types at different time points (onset, 24 h, 3 days, and 8 days). During this period, abundant cell types (haemoblasts, macrophage-like cells, morula cells, and pigment cells) exhibited varying activity patterns in different media that were formulated with varying proportions of basal media and ASW. For instance, TGM1 and TGM2 contained DMEM F12/HAM and RPMI, respectively, without ASW, whereas TGM3 contained ASW without DMEM F12/HAM or RPMI. In this study, cell proliferation was observed for at least 5 days, exceeding the documented 3 days post isolation in vitro barrier [2,29,32,37].
Similar to cell proliferation, differences in the viability of the cells were observed among the five media versions, resulting in variations in culture longevities. When setting 50% survivorship as a cut-off value, the TGM1, TGM2, and TGM3 cultures lasted for 18, 16, and 12 days, respectively, while TGM4 and TGM5 allowed for longer cultivation periods of 26 and 24 days, respectively. These findings align with those of [44], who conducted a short-term in vitro study on coral cells and found that cell viability decreased from 70% to 30% within the first week, as well as with Rinkevich and Rabinowitz [29], who studied Botryllus blood cell cultivation and observed a decrease in viability within four weeks. Rabinowitz and Rinkevich [30] reported shorter viability (7-9, 5-13, and 6-8 days) for Botryllus epithelial monolayers cultured with DMEM, RPMI, and HAM F12 on coated collagen 1 substrate. Regarding cellular morphology, this culturing criterion remained largely unchanged, except for the storage cells (pigment and nephrocyte cells) cultured in TGM3 medium (containing 85% ASW), which transformed from oval to elongated structures. This observation aligns with Rinkevich and Rabinowitz's [29] findings that pigment cells undergo changes in shape when exposed to high salt concentrations in the culture medium.
The distribution of cell types varied among the five media and underwent alterations over time, similar to cell proliferation and viability. In the first 24 h, macrophage-like cells and morula cells were the most abundant cell types in all tested media. By day 3, while the abundance profiles of TGM3 and TGM4 media remained similar to that of 24 h, morula and pigment cells were more prominent in TGM1 medium. At day 8, no further changes in cell type distribution were observed, and the abundance profiles of all five media were almost identical to those of day three. As mentioned above, to the best of our knowledge, no study has yet revealed the distribution of Botryllus blood cell types under prolonged in vitro conditions. Our results show that, at onset, macrophage-like cells, granular amoebocyte cells, and morula cells varied between 23 and 28% each, while granular amoebocyte and morula cells varied between 40 and42% each. These results are in line with the literature [29,43] regarding the distributions of granular amoebocyte and morula cells but not for macrophage-like cells.
The findings of this study demonstrate that altering the basic culture medium can cause varying growth and proliferation rates among different types of cells, as observed in our study of Botryllus blood cells. These findings align with a recent study on the cultivation of sponge cells [4] that followed up with studies documenting the enhancement of the quantity and viability of sponge cells [8,13]. The results of the present study thus demonstrate that it is possible to culture Botryllus blood cells in vitro for up to one month in a consistent and reliable manner. During the initial week of culture, there was a noticeable medium-dependent increase in the proliferation of distinct blood cell types (could be further supported by increased mortality in other cell types), which eventually led, within less than one month from initiation, to the development of medium-specific primary cultures. This discovery may pave the way for the creation of various cell cultures, each consisting of distinct cell types. Further, the aforementioned outcomes were reinforced by the ease with which cell types could be identified and classified based on their natural fluorescence patterns using confocal microscopy, an additional tool, for improved cell type identification in the development of cell cultures from B. schlosseri circulating blood cells. Compared to mammalian and insect cultures, this approach is novel to marine invertebrate cell cultures.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/cells12131709/s1, Figure S1: Immunofluorescence staining of B. schlosseri primary cultures under five media conditions at onset; Figure S2: Immunofluorescence staining of B. schlosseri primary cultures under five media conditions at 24 h from initiation; Figure S3: Immunofluorescence staining of B. schlosseri primary cultures under five media conditions at 3 days from initiation; Figure S4: Immunofluorescence staining of B. schlosseri primary cultures under five media conditions at 8 days from initiation; Figure S5: B. schlosseri primary blood cell cultures in medium TGM1; Figure S6: B. schlosseri primary blood cell cultures in medium TGM2; Figure S7: B. schlosseri primary blood cell cultures in medium TGM3; Figure S8: B. schlosseri primary blood cell cultures in medium TGM4; Figure S9: B. schlosseri primary blood cell cultures in medium TGM5; Table S1: Comparisons of media TGM1, TGM2, TGM3, TGM4, and TGM5 compositions.