In Vitro Cell. Dev. Biol. 29A:912-913, December 1993 © 1993 Tissue Culture Association 0883-8364./93 $01.50+0.00 Letter to the Editor o-VALINE SELECTIVE MEDIUM DOES NOT INHIBIT HUMAN FIBROBLAST GROWTH IN VITRO Dear Editor: The purpose of this letter is to inform fellow scientists that the amino acid a-vahne (D-vat) may not be universally used for the inhibition of fibroblast proliferation in mixed human cell culture systems. Fibroblasts proliferate more rapidly than the epithelial compo- nents of explants and dispersed cell systems, resulting in rapid fibroblast overgrowth. To overcome this problem, various strategies have been employed including cell separation techniques (2), the use of agents such as genticin (5) and putrescine (8), and spe- cific monoclonal antibodies with complement-mediated fibroblast lysis (7). It has been reported that fibroblasts lack the enzyme D-amino acid oxidase, preventing the stereoconversion of o- to L-amino acids which are required for protein synthesis (3). The replacement of the essential amino acid L-val with its stereoisomer D-val in culture medium has been shown to inhibit fibroblast overgrowth from some explanted tissues in culture and has been applied to other culture systems (3,9). The failure of a commercial D-vat medium to inhibit fibroblast overgrowth in dispersed human anterior pituitary ade- noma cell culture led us to reevaluate the effectiveness of D-val fibroblast inhibition in the presence and absence of fetal bovine serum (FBS). Three human fibroblast populations derived from neonatal and adult foreskin and adult pituitary adenoma were grown to con- fluence in Eagle's minimal essential medium (MEM) (containing L-val) supplemented with 10% FBS (MEM + 10%, GIBCO, Glas- gow, UK). Culture purity was determined by immunostaining for epithelial cell (cytokeratin markers, epithehal membrane antigen) and endothelial cell markers (Factor VIII-related antigen) which TABLE 1 FINAL CELL COUNTS MEAN (SD) FOR PITUITARY FIBRO- BLASTS IN D OR L-val MEM SUPPLEMENTED WITH FCS OR CONDITIONED WITH PITUITARY ADENOMA CELLS Conditioned Medium Serum Free 1% FBS 10% FBS (serum free) (× l 0 s) (× 103 ) (× 103) (X t 03) I~Val MEM 4.67 (0.37) 6.32 (1.30) 25.4 (1.6) ° 12.4 (0.57) b L-Val MEM 6.49 (0.29) c 6.63 (2.7) 29.5 (4.4) b 15.8 (1.11) c *P < 0.01 vs, other concentrations and conditioned medium; b p < 0.01 vs. serum free or FBS supplemented; ~P < 0.01 vs. n-val. o-val vs. L-val P = NS at 1%, 10% and in conditioned medium. 912 were negative. Proliferation experiments were carried out as fol- lows: Confluent cultures of fibroblasts were passaged employing 0.25% trypsin with 0.01% EDTA (GIBCO) and plated in six-well plates (Costar, Northumbria Biologicals, Cramlington, UK) at initial concentrations of 5 × 103 to 5 X 104 per well depending on cell type. Cells were resuspended in MEM + 10%, left for 1 h to attach to the plastic surface and then washed with either D-val or L-Vat MEM, which had otherwise identical formulation (G1BCO). Experi- ments were performed in the presence or absence of unstripped, charcoal stripped (CS), or dialyzed 0 to 10% FBS. In the case of pituitary fibroblasts, a further experiment was performed using serum-free L-val and D-val MEM which had been conditioned for 24 h by dispersed cultures from a single human anterior pituitary go- nadotrophinoma secreting follicle stimulating hormone, luteinizing hormone, and a-subunit in vitro (1). All experiments were carried out in triplicate. After 4 or 8 days cells were stripped using 0.25% trypsin with 0.01% EDTA and counted using a Coulter model Zn particle counter (Coulter Electronics, Luton, UK). Although proliferation of pituitary-derived fibroblasts was inhib- ited in serum-free conditions, this effect was entirely negated by the addition of serum (Table 1). The use of dialyzed serum gave identi- cal results (data not shown). In serum-free pituitary gonadotrophin- oma-conditioned medium, fibroblast proliferation did not differ be- tween D-val and L-val MEM and there was significant growth com- pared to serum-free medium alone. There was no difference between L-val and D-vat MEM for any concentration of unstripped or CS FBS for either neonatal or adult foreskin fibroblasts, although the proliferation of neonatal cells was significantly less than with unstripped serum (Tables 2-4). D-Val selective inhibition of proliferation was ineffective for all but pituitary-derived fibroblasts in serum-free medium. The major factor influencing fibroblast proliferation was escalating FBS con- centrations, and prior stripping or dialysis of FBS to remove free L-val reduced overall proliferation equally for both D-val and L-val MEM. This suggests that serum-free conditions rather than D-val is TABLE 2 FINAL CELL COUNTS FOR MEAN (SD) NEONATAL FORESKIN FIBROBLASTS Serum Free 1% FBS 10% FBS (XI0') (×10 4 ) (×10 4 ) D-Val MEM 13.2 (0.56) 17.6 (1.21) a 39.9 (1.06) b L-Val MEM 12.1 (1.0) 16.5 (5.4) a 34.3 (5.4) ~ ° P < 0.01 vs. other concentration; b p < 0.01 vs. other concentra- tions, o-val vs. L-vat P = NS at each concentration.