Regulation of cardiac fibroblast collagen synthesis by adenosine: roles for Epac and PI3K

Abstract

Rat cardiac fibroblasts (CF) express multiple adenosine (ADO) receptors. Pharmacological evidence suggests that activation of A(2) receptors may inhibit collagen synthesis via adenylyl cyclase-induced elevation of cellular cAMP. We have characterized the signaling pathways involved in ADO-mediated inhibition of collagen synthesis in primary cultures of adult rat CF. ANG II stimulates collagen production in these cells. Coincubation with agents that elevate cellular cAMP [the ADO agonist, 5'-N-ethylcarboxamidoadensoine (NECA), and forskolin] inhibited the stimulatory effects of ANG II. However, direct stimulators and inhibitors of protein kinase A (PKA) did not alter ANG II-induced collagen synthesis, indicating that PKA does not mediate the inhibitory effects of NECA. Inhibitors of AMP-kinase (AMPK) and extracellular signal-regulated kinase 1/2 (ERK1/2) do not alter NECA-inhibited collagen synthesis. However, activation of exchange factor directly activated by cAMP (Epac) mimicked the effects of NECA on ANG II-stimulated collagen synthesis. Inhibition of phosphoinositol-3 kinase (PI3K) reduced the inhibitory effects of NECA on ANG II-induced collagen synthesis, suggesting that NECA acts via PI3K. Furthermore, inhibition of PI3K also relieved the inhibitory effect of Epac activation on ANG II-stimulated collagen synthesis. Thus it appears that ADO activates the A(2)R-G(s)-adenylyl cyclase pathway and that the resultant cAMP reduces collagen synthesis via a PKA-independent, Epac-dependent pathway that feeds through PI3K.

Fig. 1.

A: 5′-N-ethylcarboxamidoadensoine (NECA) significantly inhibits the ANG II-mediated increase in [³H]proline incorporation. Cardiac fibroblast(s) (CF) were labeled with [³H]proline and treated with 100 nM ANG II, 10 μM NECA, or both for 14 h (n ≥ 40; #P < 0.001 vs. control; ***P < 0.001 vs. ANG II). B: NECA effects are insensitive to pertussis toxin (PTX) pretreatment. CF were pretreated with PTX (0.1 μg/ml) for 16 h and then labeled with [³H]proline and treated with 100 nM ANG II, 10 μM NECA, or both for 14 h. Pretreatment with PTX did not relieve the inhibition by NECA (n = 6; #P < 0.001 vs. PTX; ***P < 0.001 vs. PTX + ANG II).

Fig. 2.

A: forskolin (FSK) inhibits ANG II-mediated increases in [³H]proline incorporation. CF were labeled with [³H]proline and treated with 100 nM ANG II, 10 μM FSK, or both for 14 h (n ≥ 6; #P < 0.001 vs. control; ***P < 0.001 vs. ANG II). B: 8-bromoadenosine 3′,5′-cyclic monophosphothioate, Sp-isomer (Sp-8-Br-cAMPS) had no effect on the ANG II-mediated increase in [³H]proline incorporation. CF were labeled with [³H]proline and treated with 100 nM ANG II, 30 μM Sp-8-Br-cAMPS, or both for 14 h (n = 6; #P < 0.001 vs. control). C and D: protein kinase A (PKA) inhibitors do not relieve NECA-stimulated inhibition of the ANG II-mediated increase in [³H]proline incorporation. CF were pretreated with 10 μM H89 (C) or 100 μM Rp-8-Br-cAMPS (RP) (D) for 20 min and then concurrently labeled with [³H]proline and treated with 100 nM ANG II, 10 μM NECA, or both for 14 h (n = 6; #P < 0.001 vs. Rp-8Br-cAMPS. $P < 0.05 vs. H89; ***P < 0.001 vs. RP + ANG II; *P < 0.05 vs. H89 + ANG II).

Fig. 3.

A: Epac activator 8-CPT-2′-O-Me-cAMP (CPT) inhibits the ANG II-mediated increase in [³H]proline incorporation in CF. CF were labeled with [³H]proline and treated with 100 nM ANG II, 100 μM CPT, or both for 14 h (n ≥ 6; #P < 0.001 vs. control; ***P < 0.01 vs. ANG II). B: CPT increases CF extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation in a time-dependent manner. CF were treated with 100 μM CPT for the times indicated (mean ± range; n = 2; *P < 0.05 vs. control; **P < 0.01 vs. control; ***P < 0.001 vs. control).

Fig. 4.

A: inhibition of phosphoinositol-3 kinase (PI3K) relieves the NECA-stimulated inhibition of the ANG II-mediated increase in [³H]proline incorporation. CF were pretreated with 10 μM LY-294002 (LY) for 20 min and then concurrently labeled with [³H]proline and treated with 100 nM ANG II, 10 μM NECA, or both for 14 h. In the presence of LY, ANG II increased [³H]proline incorporation in the presence or absence of NECA; i.e., LY relieved the inhibitory effect of NECA (n = 6; #P < 0.001 vs. corresponding control). B: inhibition of mitogen-activated protein kinase kinase (MEK) (ERK1/2) does not relieve NECA-stimulated inhibition of the ANG II-mediated increase in [³H]proline incorporation. CF were pretreated with 10 μM PD-98059 (PD) for 20 min and then labeled with [³H]proline and treated with 100 nM ANG II, 10 μM NECA, or both for 14 h (n = 6; #P < 0.001 vs. PD-98059, **P < 0.01 vs. PD + ANG II).

Fig. 5.

Inhibition of PI3K relieves the CPT-stimulated inhibition of the ANG II-mediated increase in [³H]proline incorporation. CF were pretreated with 10 μM LY for 20 min and then labeled with [³H]proline and treated with 100 nM ANG II, 100 μM CPT, or both for 14 h (n = 6; #P < 0.001 for effects of ANG II vs. corresponding control).

Fig. 6.

NECA increases AKT (protein kinase B, PKB) phosphorylation in a time-dependent manner. CF were treated with 10 μM NECA for the times indicated. A: representative phospho- and total-Akt immunoblots. B: quantitation of AKT phosphorylation at 15- and 30-min time points (n ≥ 10; **P < 0.01; ***P < 0.001).

Fig. 7.

Proposed signaling pathway for adenosine-mediated inhibition of collagen production. See text for more information.