Radiation induces diffusible feeder cell factor(s) that cooperate with ROCK inhibitor to conditionally reprogram and immortalize epithelial cells

Abstract

Both feeder cells and Rho kinase inhibition are required for the conditional reprogramming and immortalization of human epithelial cells. In the present study, we demonstrated that the Rho kinase inhibitor Y-27632, significantly suppresses keratinocyte differentiation and extends life span in serum-containing medium but does not lead to immortalization in the absence of feeder cells. Using Transwell culture plates, we further demonstrated that physical contact between the feeder cells and keratinocytes is not required for inducing immortalization and, more importantly, that irradiation of the feeder cells is required for this induction. Consistent with these experiments, conditioned medium was shown to induce and maintain conditionally immortalized cells, which was accompanied by increased telomerase expression. The activity of conditioned medium directly correlated with radiation-induced apoptosis of the feeder cells. Thus, the induction of conditionally reprogrammed cells is mediated by a combination of Y-27632 and a diffusible factor (or factors) released by apoptotic feeder cells.

Figure 1

Y-27632 inhibits calcium- and serum-induced keratinocyte differentiation. A: Y-27632 stimulates keratinocyte proliferation in medium containing calcium and serum. HFKs were seeded at 5.0 × 10³ per well in six-well tissue culture plates under four culture conditions: KGM, F medium (F), F medium with Y-27632 (F+Y), and F medium with Y-27632 in coculture with a layer of lethally irradiated J2 feeder cells (F+J2+Y). The cells were harvested and counted 8 days later. The experiment was performed four times independently; results from one representative experiment are shown. B: Y-27632 inhibits expression of differentiation-associated genes in the presence of calcium and serum. 5.0 × 10⁶ HFKs were plated in 10-cm tissue culture dishes under three culture conditions. The cells were harvested 2 days later, and RT-qPCR was used to measure levels of mRNAs characteristic of keratinocyte differentiation. Data are expressed as means ± SD.

Figure 2

Feeder cells are required for conditional immortalization of keratinocytes. HFKs were passaged for up to 100 days under three culture conditions: KGM, F+Y, and F+Y+J2. Y-27632 alone did not induce conditional immortalization. Two independent experiments with two different HFK strains are shown (p2, top; p3, bottom).

Figure 3

Conditional immortalization of keratinocytes does not require physical contact with feeder cells. A: In the indirect coculture system, HFKs were plated on the bottom of the companion dish, and irradiated feeder cells were seeded on the Transwell polycarbonate membrane suspended 1 mm from the bottom. Cells shared the same medium, but were not in direct contact. B: HFKs were cultured in direct contact with irradiated feeder cells or in the indirect coculture system without contact for more than 90 days in the presence of Y-27632 in two independent experiments (top and bottom panels). HFKs showed similar rates of proliferation and conditional immortalization in both culture systems. HFKs grown in KGM served as a negative control for immortalization.

Figure 4

Irradiation of feeder cells induces diffusible factors that conditionally immortalize keratinocytes. HFKs were cocultured with lethally irradiated or nonirradiated J2 feeder cells for more than 70 days in the Transwell indirect coculture system in the presence of Y-27632. Only irradiated feeder cells supported conditional immortalization of the HFKs. The experiment was performed in duplicate, using HFKs at p5 (A) or p9 (B). Conditional immortalization was more rapidly observed with the late-passage HFKs.

Figure 5

Y-27632 is not required for the production of diffusible factors that stimulate keratinocyte proliferation. A: Conditioned F medium was collected from irradiated feeder cells in the absence or presence of Y-27632 or with Y-27632 added to the conditioned medium immediately before use. HFKs (5.0 × 10³) were plated in these various media in six-well tissue culture plates and were grown for 6 days. Conditioned medium without Y-27632 did not stimulate proliferation of the HFKs to the extent of that produced with Y-27632 or with Y-27632 added after collection. B: For quantification of the proliferation assay, the fixed and stained cells in A were solubilized, and absorbance at 564 nm was measured. CM, conditioned F medium; OD, optical density.

Figure 6

Conditioned medium supports conditional immortalization. A: Conditioned medium rescues late-passage HFKs undergoing senescence. Late-passage HFKs (p10, top; p11, bottom) cultured in F medium containing Y-27632 (as in Figure 2) were plated in F medium using three different conditions: in the presence of Y-27632 (F+Y), in the presence of irradiated feeder cells and Y-27632 (Feeders + Y), and in medium conditioned by irradiated feeder cells containing Y-27632 (Conditioned medium + Y). Y-27632 alone was unable to rescue the HFKs from rapidly undergoing senescence. In the presence of Y-27632, both irradiated feeder cells and conditioned medium were equally able to rescue HFKs from senescence. The experiment was performed in duplicate with two different strains of HFKs (top and bottom panels). B: Conditioned medium conditionally immortalizes keratinocytes. Early-passage HFKs (p2) were cultured for more than 80 days (60 population doublings) in F medium containing Y-27632 in direct contact with irradiated feeder cells or in conditioned F medium containing Y-27632. Under both conditions, the HFKs were efficiently immortalized and exhibited similar rates of proliferation.

Figure 7

Conditioned medium induces telomerase. A: Conditioned medium induces telomerase expression. Levels of hTERT mRNA were measured in HFKs cultured in serum-free medium (KGM), cocultured with irradiated feeder cells in F medium, or cultured in conditioned F medium with and without Y-27632. hTERT was induced equally well by feeder cells and by conditioned medium, irrespective of Y-27632. B: Levels of hTERT mRNA were measured in F medium or conditioned medium, both in the presence of Y-27632. F medium did not fully induce telomerase expression.

Figure 8

Activity of conditioned medium correlates with apoptosis of irradiated feeder cells. A and B: Conditioned medium and unconditioned F medium (control) were used to culture 5.0 × 10³ HFKs in six-well tissue culture plates in the presence of Y-27632. Cells were counted after 6 days to assess proliferation. A: Activity of conditioned medium dramatically increased 24 hours after irradiation. Irradiated J2 feeder cells (1.0 × 10⁷ to 1.5 × 10⁷) were plated with 30 mL of F medium in 175-cm² tissue culture flasks. Conditioned medium was collected after 24, 48, 72, or 96 hours. B: Conditioned medium was collected from a single flask of irradiated J2 cells every 24 hours (through 96 hours), with replacement of fresh F medium after each collection. C: Apoptosis of feeder cells dramatically increased 24 hours after irradiation. Irradiated J2 feeder cells (2.0 × 10⁴) were plated in each well of a 96-well tissue culture plate. Activity of caspases 3 and 7 was measured at 0, 24, 48, and 72 hours after irradiation. RLU, relative luminescence units.