Considerable information regarding the regulation of GH and prolactin (PRL) release has been generated using pituitary cell lines as model systems. Inasmuch as these cultures have been derived from single cells by clonal selection it has frequently been assumed that they are composed of homogeneous populations of hormone secretors. However, experience with GH3 cells clearly demonstrates that such is not the case, since this GH- and PRL-producing line is comprised of a mixture of cells that are bihormonal, secrete only GH, or secrete neither hormone. Interestingly, the relative amount of these phenotypic subpopulations is not fixed, but can be altered by treatment with established regulators of GH and PRL secretion. This potential for secretory heterogeneity and phenotypic plasticity prompted us to examine the cellular composition of other commonly used GH- and/or PRL-secreting cell lines under control and treatment conditions. First, GH4C1, GH1, GC, MMQ and P0 cells were maintained according to established media and culture protocols and the relative abundance of GH and PRL secretors was assessed by reverse haemolytic plaque assays. As shown previously for GH3 cells, two cell lines were found to be functionally heterogeneous. Specifically, GH4C1 and GH1 were comprised of mixed populations of GH (25·9±1·1% and 51·3±6·5% (s.e.m.) respectively) and PRL (44·8±3·7 and 66·1 ±4·1% respectively) secretors. However, MMQ and GC cell cultures were relatively homogeneous with respect to hormone secretion in that the MMQ cells released PRL (72·2 ± 4·9%) but not GH, while GC cells released GH (93·6 ±1·4%) but not PRL. Similarly, P0 cell cultures contained predominantly GH-secreting cells (96·4±1·2%), but a few cells were detected that released PRL (1·1 ±0·1%). We then tested whether cortisol or oestradiol-17β (OE2) could alter the secretory composition of cell lines identified as reasonably homogeneous (GC, MMQ and P0) by treating cultures with steroids at doses ranging from 0·001 to 1000 nmol/l for up to 144 h. The presence of steroids in GC and MMQ lines did not affect the percentage of GH and PRL secretors. However, treatment of P0 cells with 0·001 to 1000 nmol OE2/l for 24, 48 and 144 h resulted in a dramatic increase in PRL cells when compared with controls. These increases occurred without alterations in GH percentages and in the absence of significant cell proliferation. In fact, OE2 suppressed the proliferation of P0 cells to 20% of the control rate (P<0·05), suggesting that OE2 induced pre-existing GH-secreting cells to release both GH and PRL. Taken together, these results clearly indicate that clonal origin does not guarantee homogeneity of function, and alteration in the culture environment can have profound effects on the secretory cell composition of acidophilic cell lines. These considerations have important implications for the interpretation of results generated with GH- and/or PRL-secreting cell lines.
Journal of Endocrinology (1994) 140, 455–463