Epidermal growth factor receptor kinase activity is required for gap junction closure and for part of the decrease in ovarian follicle cGMP in response to LH

Abstract

The meiotic cell cycle in mouse oocytes is arrested in prophase, and then restarted when LH acts on the surrounding granulosa cells. The granulosa cells keep meiosis arrested by providing a source of cGMP that diffuses into the oocyte through gap junctions, and LH restarts the cell cycle by closing the junctions and by decreasing granulosa cell cGMP, thus lowering oocyte cGMP. Epidermal growth factor receptor (EGFR) activation is an essential step in triggering LH-induced meiotic resumption, but its relationship to the cGMP decrease in the follicle is incompletely understood, and its possible function in causing gap junction closure has not been investigated. Here, we use EGFR agonists (epiregulin and amphiregulin) and an EGFR kinase inhibitor (AG1478) to study the function of the EGFR in the signaling pathways leading to the release of oocytes from prophase arrest. Our results indicate that the EGFR kinase contributes to LH-induced meiotic resumption in two different ways. First, it is required for gap junction closure. Second, it is required for an essential component of the decrease in follicle cGMP. Our data show that the EGFR kinase-dependent component of the cGMP decrease is required for LH-induced meiotic resumption, but they also indicate that an as yet unidentified pathway accounts for a large part of the cGMP decrease.

Figure 1

cGMP in mouse ovarian follicles at various times after applying LH. Bars indicate mean ± s.e.m., and numbers in parentheses indicate the number of samples.

Figure 2

Observation of the time course of nuclear envelope breakdown in intact follicle-enclosed oocytes in response to an EGFR agonist (epiregulin), or to LH in the presence or absence of an EGFR kinase inhibitor (AG1478). A. A mouse follicle on a Millicell membrane; the prophase-arrested nucleus and 2 nucleoli are visible. B. Nuclear envelope breakdown at various times after treatment of follicles with LH (350 nM), epiregulin (100 nM), or AG1478 (500 nM) for 1 hour followed by AG1478 + LH. Numbers in parentheses indicate the number of follicles.

Figure 3

cGMP in follicles under experimental conditions affecting EGFR kinase or MAP kinase activity. A. Measurement of follicle cGMP under the following conditions: no treatment (basal), 1 hour LH (350 nM), 1 hour epiregulin (100 nM), 1 hour AG1478 (500 nM) followed by 1 hour AG1478 + LH, 1 hour U0126 (10 μM) followed by 1 hour U0126 + LH. Bars indicate mean ± s.e.m., and numbers in parentheses indicate the number of samples. The cGMP concentration after 100 nM epiregulin treatment is significantly different from that for basal and LH conditions (p < .0001). Likewise, the cGMP concentration after AG1478 + LH treatment is significantly different from that for basal (p < .0001) and LH (p = .01) conditions. The cGMP concentration after U0126 + LH treatment is not significantly different from that after LH alone (p = .7), but is significantly different from the basal concentration (p = .0004). B. Demonstration that in follicles stimulated with LH, treatment with 500 nM AG1478 as described above reduces EGFR phosphorylation on tyrosine 1068 to a level that is indistinguishable from the basal level without LH. 20 μg of follicle lysates, and 2 μg of EGF-treated AG431 cells, were used for immunoblotting with an antibody specific for pY1068 EGFR and then with an antibody specific for total EGFR. The graph shows the ratio of the intensities of the ~170 kDa band measured with the 2 antibodies, normalized to the no treatment condition (mean ± S.D. for 2 independent experiments). Since the follicles were cultured for 24 hours in the absence of EGFR ligands or serum prior to preparing the samples, the basal signal seen with the pY1068 antibody is probably not due to EGFR kinase activity; it could result from phosphorylation of the EGFR by other kinases or to cross-reactivity of the antibody with non-phosphorylated EGFR.

Figure 4

Gap junction communication in follicles under experimental conditions affecting EGFR kinase activity. Gap junction communication was assessed by injecting the oocyte with Alexa Fluor 350 and imaging the follicle 20 minutes later. A. Images of follicles that were treated as follows prior to injection of the fluorescent tracer: no treatment (basal), 1 hour LH (350 nM), 1 hour epiregulin (100 nM), 1 hour epiregulin (1 μM) + amphiregulin (1 μM), 1 hour AG1478 (500 nM) followed by 1 hour AG1478 + LH. B. Ratios of fluorescence in the mural granulosa cells over that in the inner cumulus cells (MG/IC), at 20 minutes after injection, for each of the conditions indicated in A. Bars indicate mean ± s.e.m., and numbers in parentheses indicate the number of follicles. The MG/IC ratio after treatment with 100 nM epiregulin is significantly different from that after LH (p = .05), and from the basal ratio (p = .005). Likewise, the MG/IC ratio after treatment with 1 μM epiregulin + 1 μM amphiregulin is significantly different from that after LH (p = .0003); the difference from the basal ratio is marginally significant (p = .06). The MG/IC ratio for AG1478 + LH is significantly different from that for LH (p < .0001), but not from the basal ratio (p = 1.0).

Figure 5

Some of the known and unknown components of the signaling network linking LH receptor and EGFR receptor activation in the granulosa cells to the decrease in cGMP in the oocyte that leads to resumption of meiosis.