Serum-specific stimulation of proliferation and mineralization of fish bone-derived cells

Fish culture

An homogenous group of 24 gilthead seabream were obtained from a local aquaculture facility (body weight 180–200 g) then reared for 12 months in 1000-L circular tanks containing gravel-filtered, aerated seawater (salinity: 35 ppt; temperature: 18–25°C; oxygen content: > 80% saturation) until they reached a mean body weight of 680 ± 4 g. During rearing period, fish were submitted to natural photoperiod and hand fed a commercial seabream diet (AquaGold, crude protein 45%, crude fat 16%, Sorgal) to apparent satiety in two daily meals. Fish were starved for 24 h prior to blood sampling.

Blood collection and serum preparation

Fish were submitted prior to blood collection to a lethal anesthesia (10 min in seawater containing 400 μl/L of AQUI-S, New Zealand Ltd.). Blood was collected into glass tubes by puncture of the caudal vein then clotted at room temperature for 2 h. Clot was contracted for 4 h at 4°C then samples were centrifuged at 1590 g (10 min, 4°C). Serum, i.e. supernatant fraction, was transferred to a new tube and stored at −80°C.

Cell culture and extracellular matrix mineralization

VSa13 and VSa16 cells, both developed from seabream vertebra, were routinely cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 1% penicillin-streptomycin, 1% fungizone, 2 mm l-glutamine and 10% FBS (fetal bovine serum) as described previously (Pombinho et al., 2004). Extracellular matrix (ECM) mineralization was induced in confluent cultures by supplementing culture medium with 50 μg/ml of l-ascorbic acid, 10 mmβ-glycerophosphate and 4 mm CaCl₂. Culture medium was renewed twice a week. At appropriate times, mineral deposition was revealed as described by von Kossa (1901) and quantified through densitometry analysis. Cell culture reagents and dishes were purchased from Invitrogen and Sarstedt, respectively.

Cell proliferation assay

Cell proliferation was determined from cultures seeded in 96-well plates at 1.5 × 10³ cells per well using the CellTiter 96 non-radioactive proliferation assay kit (Promega) according to manufacturer’s instructions. At appropriate times, cells were incubated for 1 h with 20 μl of reagent mixture then cell proliferation was determined from absorbance at 490 nm. In some cases, cultures were trypsinized (1 min at room temperature using 0.2% trypsin-EDTA solution) and cell number was determined using a Neubauer counting chamber.

RNA extraction and quantitative real-time PCR

Total RNA was extracted from cell cultures as described by Chomczynski and Sacchi (1987) and its quality and quantity assessed on agarose-formaldehyde gels and by UV spectrophotometry, respectively. Total RNA (1 μg) was reverse-transcribed at 37°C for 1 h using the Moloney-murine leukemia virus reverse transcriptase (Invitrogen), RNase Out (Invitrogen) and oligo-d(T) universal primer [5′-ACGCGTCGACCTCGAGATCGATG(T)₁₃-3′]. Quantitative real-time PCR (qPCR) assays were performed using the iCycler PCR system (Bio-Rad). Reaction mixture containing 1× iQ SYBR Green I mix (Bio-Rad), 0.2 μm of forward and reverse primers and 10 ng of reverse-transcribed RNA was submitted to the following PCR conditions: 4 min at 95°C, 50 ×  [30 s at 95°C] then 30 s at 68°C. Relative gene expression of MGP, COL1A1, OC1, TNAP, OP, ON and BMP2 genes was normalized using RPL27a (primer list in Table 1).

Statistical analysis of data

Data are presented as the average ± standard deviation of values from at least three separate experiments. Statistical significance of data was evaluated through the Student’s t-test or the analysis of variance (anova). Differences were considered to be significant for P < 0.05.

Fish serum enhances proliferation of fish bone-derived cell lines

Proliferative effects of serum were initially tested on cells cultured in a medium supplemented with 10% of either GSS or FBS (considered as control thereafter) and evaluated at appropriate times using CellTiter proliferation assay kit and/or cell counting. Both methods revealed that proliferation of seabream cell lines was stimulated by GSS after 13 days, although to a lesser extent in VSa13 cells (Fig. 1). Morphological changes were also observed, i.e. cells were typically bipolar and fibroblastic (sub-confluent cultures) then polygonal (confluent cultures) when grown in bovine serum and epithelial-like and sharply outlined when grown in seabream serum (Fig. 1). Effect of different FBS/GSS ratios (final serum concentration was maintained at 10%) were then tested and highest proliferative effect was attained using FBS/GSS ratios of 5/5 and 7.5/2.5 for VSa13 and VSa16 cells, respectively, suggesting that GSS proliferative effect is dose-dependent and cell type-specific (Fig. 2). Using these conditions, proliferation was stimulated approximately 3.5-fold in both cell lines compared with control (Fig. 2). Again, both methods used to assess cell proliferation (i.e. CellTiter proliferation assay and cell counting) indicated similar stimulatory effects although data from CellTiter proliferation assay has a tendency for value compression.

Altogether, these results indicate that GSS is more effective than FBS to promote the growth of seabream bone-derived cells and suggest the presence in seabream serum of growth factor(s), which are most probably more effective than their bovine homolog(s). Interestingly, reducing GSS proportion to 50% (VSa13) or 25% (VSa16) while maintaining final serum concentration at 10% by increasing FBS proportion enhanced seabream cell proliferation. Similar experiments using goldfish fin cells revealed that a mixture of carp and bovine sera, 50% each, was more effective in cell growth than carp serum alone and that early passages of these cells responded better to higher proportions of carp serum (Hashimoto et al., 1997). Proliferation data from goldfish and seabream cells suggest that fish cells grow better in the presence of fish serum and that fish cells previously maintained in mammalian serum still require its presence for an optimal growth, probably because of some adaptation versus weaning mechanisms. The clear modification of GSS-treated cell phenotype is in agreement with the previous observation by Hashimoto et al., 1997 who mentioned that cells cultured in medium supplemented with different sera show singular shape responses. This change of phenotype in GSS-treated cells could indicate the capacity of seabream cells to return to their original phenotype (a phenotype lost upon culture in FBS) and the presence of factor(s) in seabream serum capable of restoring innate characteristics of seabream cells. Factors present in seabream serum could be similar to cell adhesion factors identified in rainbow trout serum (Lee and Bols, 1991), and responsible for the stimulation of cell spreading and subsequent morphological changes (Hashimoto et al., 1997). In conclusion, our data indicate that seabream serum contains factors essential for cell growth and phenotype establishment, and support the idea that bovine serum is not totally adequate to culture fish cells.

Fish serum stimulates in vitro mineralization of fish bone-derived cell lines

Effects of serum on extracellular matrix mineralization were tested on cells cultured in mineralogenic medium supplemented with 10% of either GSS, FBS or a combination of both sera. While mineral deposition was only detected after 3.5 weeks in FBS-treated VSa13 cell cultures, it was clearly visible after only 2 weeks in GSS-treated cells and the levels of mineral deposition were much higher (approx. 3 times) after 3.5 weeks when compared to those obtained after an equal period in cells supplemented with FBS (Fig. 3). These results suggest the presence in seabream serum of mineralogenic factor(s), which modify onset and extent of mineralization in chondrocytic VSa13 cells in a dose-dependent manner (effect was noticeable with 5% GSS and higher with 10%). Similarly, mineral deposition was observed in GSS-treated VSa16 cell cultures after 2 weeks, while none was detected in FBS-treated cells until 3.5 weeks (Fig. 3). Unexpectedly, GSS exhibited a strong cytotoxic effect at 3.5 weeks as indicated by the reduced mineralization in cultures grown with 2.5% of GSS and absence of mineralization in cultures grown with 10% of GSS (Fig. 3). Despite being toxic for these cells, GSS also demonstrated mineralogenic effects in osteoblastic VSa16 cells, similar to those observed in VSa13 cells, also in a dose-dependent manner (Fig. 3). Recent studies in which demineralized rat tibia were incubated in different sera (including serum from vertebrates and invertebrates) then evaluated for recalcification, revealed the presence of factors essential for mineralization in vertebrate sera (e.g. rat, bony fish and shark) but not those from invertebrates (Price et al., 2004; Hamlin et al., 2006). Our data also suggest the presence of mineralogenic factors in vertebrate serum and confirm that fish serum is more appropriate than bovine serum to trigger an optimal mineralization of fish bone-derived cells, most probably because of the higher specificity of fish mineralogenic factors for fish cells. Toxicity in VSa16 cell cultures treated for 3.5 weeks with GSS suggest that proportion of fish serum is still not optimal and should be reduced for culturing these cells.

Regulation of mineralization-related gene expression by GSS

To better understand mechanisms underlying mineralogenic effect of GSS, expression of several bone-related genes was evaluated by qPCR after 1 week of mineralization in VSa13 cells (VSa16 cells were not tested due to the cytotoxic effect previously observed). Comparative analysis of gene expression in GSS- versus FBS-treated mineralizing VSa13 cells (Fig. 4) revealed a decrease of matrix Gla protein (MGP) expression (approx. 6-fold) and an increase of bone morphogenetic protein 2 (BMP-2) expression (approx. 10-fold). MGP is a vitamin K-dependent protein secreted by chondrocytes and vascular smooth muscle cells into the extracellular compartment where it functions as a physiological inhibitor of mineralization, delaying chondrocyte maturation and blocking endochondral ossification (Luo et al., 1995; Yagami et al., 1999; Murshed et al., 2004). Any mechanism decreasing MGP gene expression and protein production in bone-derived cell lines would remove or reduce its inhibitory action, therefore advancing onset and/or extent of ECM mineralization. BMP-2 is a member of the TGF-β family of growth factors and a potent inducer of bone formation by stimulating osteoblast differentiation (Ducy and Karsenty, 2000; Huang et al., 2002). Any mechanisms increasing BMP-2 gene expression and protein production would stimulate osteogenesis, therefore advancing onset and/or extent of ECM mineralization. We propose that mineralogenic effect of GSS could be attributed to serum factors regulating the transcription of MGP and BMP-2 genes and consequently ECM mineralization. Previous studies (Hamlin et al., 2006) had shown that serum from bony fish and shark contains a potent nucleator of collagen calcification capable of remineralizing a demineralised rat tibia. However, the nature of this serum nucleator and its mode of action are still poorly understood. Changes in expression of other bone-related genes did not provide clear evidence of their participation in the mineralogenic effect of GSS. Expression of osteocalcin 1 (OC1), a Gla-containing protein associated with tissue ossification (Ducy et al., 1996; Hunter et al., 1996) and a marker of osteoblast differentiation (Nieden et al., 2005; Li et al., 2009), was apparently significantly reduced in mineralizing GSS-treated VSa13 cells but was found at levels close to qPCR detection limits in all conditions. Previous expression data showing its absence in VSa13 cells (Pombinho et al., 2004) suggest that the observed decrease in osteocalcin expression in GSS-treated cells may be negligible. Similar conclusions applied to osteopontin (OP) while changes in the expression of osteonectin (ON) and tissue non-specific alkaline phosphatase (TNAP) genes were not significant. Expression of collagen type 1 (COL1A1), an early marker of osteoblast commitment and the major protein in bone extracellular matrix (Karsenty and Park, 1995), was increased in both mineralizing GSS- and FBS-treated cells, although this increase was more pronounced in cells cultured in FBS (approx. 1.5 times). Any mechanism increasing type 1 collagen gene expression should enhance extracellular matrix formation and probably mineralization. However, levels of mineralization were lower in GSS-treated cells. There is no obvious explanation for this result and future experiments (e.g. measurement of total collagen content) will be needed to understand the relationship between decreased COL1A1 expression and increased ECM mineralization. In general, our results suggest the presence of factor(s) in GSS serum capable of regulating specific bone-related gene expression in a specific manner towards VSa13 cell differentiation/mineralization.