Inhibition of Forskolin-induced Neurite Outgrowth and Protein Phosphorylation by a Newly Synthesized Selective Inhibitor of Cyclic AMP-dependent Protein Kinase, N-[2-(p-Bromocinnamylamino)ethyl]-54soquinolinesulfonamide (H-89), of PC12D Pheochromocytoma Cells*

A newly synthesized isoquinolinesulfonamide, H-89 (N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinoline-sulfonamide), was shown to have a potent and selective inhibitory action against cyclic AMP-dependent protein kinase (protein kinase A), with an inhibition con- stant of 0.048 f 0.008 WM. H-89 exhibited weak inhib- itory action against other kinases and Ki values of the compound for these kinases, including cGMP-depend-ent protein kinase (protein kinase G), Ca2+/phospho- lipid-dependent protein kinase (protein kinase C), cas- ein kinase I and II, myosin light chain kinase, and Ca2’/ calmodulin-dependent


Inhibition of Forskolin-induced
Since the discovery of CAMP-dependent protein kinase (protein kinase A) in rabbit skeletal muscle by Krehs and colleagues (Walsh et aE., 1968), the mechanisms of action of a number of additional extracellular signals have been found to involve regulation of the formation of CAMP and thereby control of its target enzyme (Greengard, 1978). Extension of the neuronal process is fundamental to establishment of the intricate network of the nervous system. The rat PC12 pheochromocytoma cell line is a pertinent model system for the study of neuronal differentiation (Greene and Tischler, 1976). Nerve growth factor (NGF)' is a well characterized polypeptide (Levi-Montalcini and Angeletti, 1968) which promotes a variety of responses in target cells (Greene and Tischler, 1976;Greene and Shooter, 1980). PC12 responds to NGF with neurite outgrowth (Greene and Tischler, 1976), increases in the rate of acetylcholine synthesis (Greene and Tischler, 1976;Greene and Rein, 1977), alterations in the surface morphology (Connolly et al., 1979), activation of adenylate cyclase, and generation of intracellular CAMP (Schubert et al., 1978). Permeable analogues of CAMP or a CAMPenhancing reagent can elicit responses which mimic the action of NGF in PC12 cells (Greene and McGuire, 1978;Rieger et al., 1980;Halegoua and Patrick, 1980;Greene et al., 1984).
The role of CAMP in the action of NGF remains unclear and whether or not the enhancement of intracellular CAMP-induced function in PC12 cells is mediated by protein kinase A is also unknown. A subline of PC12 cells, PCl2D cells, extends neurites very quickly in response not only to NGF but also to CAMP-enhancing agents (Katoh-Semba et al., 1987). To explore further the role of protein kinase A on neurite outgrowth, we investigated the effect of H-89, a newly synthesized protein kinase A inhibitor, on forskolin-, dibutyryl CAMP-, and NGF-induced neurite outgrowth from PC12D cells. from ICN Radiochemicals. Protein kinase C from rabbit brain was purified as described (Hidaka and Tanaka, 1987), and CAMP-dependent protein kinase catalytic subunit was prepared from bovine heart by the method of Beavo et al. (1974).
cGMP-dependent protein kinase was prepared from pig lung by the method of Kuo and Greengard (1974). Casein kinases I and II from bovine testis were purified as described (Chijiwa et al., 1989). Ca*'/calmodulin-dependent protein kinase II was prepared from rat brain by the method of McGuiness et al. (1985). Myosin light chain was prepared from chicken gizzard by the method of Perrie and Perry (1970).
Myosin light chain kinase was purified from chicken gizzard by the method of Adelstein and Klee (1981 Hidaka et al. (1984). C asein kinases I and II activities were measured as described (Chijiwa et al. 1989). Ca'+/calmodulin kinase II activity was measured by the method described by McGuinness et al. (1985).
Kinase activities were assayed at 30 "C for 2-5 min by measuring the transfer of '*P from [-r-32P]ATP to substrates. The reaction was terminated by adding 1 ml of 20% trichloroacetic acid, following the addition of 100 rg of bovine serum albumin as a carrier protein.
The sample was centrifuged at 3000 rpm for 10 min, the pellet was resuspended in 5% trichloroacetic acid solution, the final pellet was dissolved in 1 ml of 1 N NaOH and the radioactivity was measured in a liquid scintillation counter. Cell Culture-Cultures of PCl2D cells were expanded in loo-mm tissue culture plastic dishes. The medium used was Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum, 10% horse serum, 5 pg/ml insulin, 5 pg/ml transferrin, 10m4 M putrescine, 12 X lo-* M progesterone, and 3 X 10-s M sodium selenite. We replanted the subconfluent cells on 6well tissue culture plastic plates at a density of 4 x 10" cells/well for analysis of neurite outgrowth and at a density of 4 X lo5 cells/well for analysis of protein phosphorylation and CAMP accumulation.

Effect of H-89 on Various Protein Kim-se Activities in Vi-
tro-The effects of this newly synthesized isoquinolinesulfonamide on CAMP-dependent protein kinase (protein kinase A), cGMP-dependent protein kinase (protein kinase G), Ca'+/ phospholipid-dependent protein kinase, (protein kinase C), myosin light chain kinase, Ca*+/calmodulin-dependent protein kinase II (Ca*'/calmodulin kinase II), and casein kinase I and II were investigated. K, values for various protein kinases are summarized in Table I. As can be seen, H-88 (N-(2-cinnamylaminoethyl)-5-isoquinolinesulfonamide) exhibited a potent inhibitory action on cyclic nucleotide-dependent protein kinases. H-89, a brominated derivative of H-88, proved to be the most potent and selective CAMP-dependent protein kinase inhibitor among the isoquinolinesulfonamide compounds tested, and the K, value for protein kinase A was 0.048 f 0.008 PM. The Ki value of this drug for protein kinase G was about 10 times higher (0.48 f 0.13 PM), suggesting that H-89 was a specific protein kinase A inhibitor (Table I). H-89 was shown to have a more selective inhibition for protein kinase A than for protein kinase G and much weaker effects for other kinases (at least 500-3000-fold less potent). The structures of these protein kinase A inhibitors are shown in Fig. 1. Inhibition patterns of protein kinase A by H-89 and  respectively; the CAMP level without forskolin, 28.6 f 4.2; mean f SE., n = 3). When examining the effect of H-89 on adenylate cyclase from bovine brain and cyclic 3'-5' nucleotide phosphodiesterase from rabbit brain, we found that this compound did not inhibit these enzymes up to a dose of 100 pM. Neither H-88 nor H-89 directly influenced activities of cyclic AMP synthesizing (adenylate cyclase) and metabolizing (phosphodiesterase) enzymes in uitro. Thus, H-89 may act directly on the protein kinase A in the PC12D cells.  15.9 f 5.4 CAMP (1 mM)-induced neurite outgrowth at various time points was investigated using phase contrast microscopy. Neurite outgrowth was maximal at 16 h after the addition of forskolin, and this outgrowth was maintained for 48 h ( effects of H-89 on NGF-induced neurite outgrowth from PCl2D cells. Exposure of NGF (50 rig/ml) produced a stimulation of neurite outgrowth in a time-dependent fashion, and the maximal effect was seen at 24 h (Fig. 4B). Pretreatment with H-89 (20 PM) for 30 min before the addition of NGF (50 rig/ml) exerted no inhibitory action on the NGF-induced neurite outgrowth of PCl2D cells. We also investigated the effect of H-89 on dibutyryl CAMP-induced neurite outgrowth of the PClPD cells (Table II). After pretreatment with H-89 for 30 min, PC12D cells were exposed for 24 h to culture medium which contained 1 mM dibutyryl CAMP. H-89 inhibited dibutyryl CAMP-induced neurite outgrowth dose-dependently. The finding that H-89 inhibited both forskolin-and dibutyryl CAMP-induced neurite outgrowth in PC12D cells suggests that the inhibitory action of H-89 is not mediated through the inhibition of CAMP synthesis or degradation. It would thus appear that protein kinase A acts as mediator of forskolin-or dibutyryl CAMP-induced neurite outgrowth but not of NGF-induced neurite outgrowth. Figs. 5 and 6 show typical photographs of forskolin-, NGF-, and dibutyryl CAMPinduced neurite outgrowth, with or without H-89.
Effect of H-89 on Various Protein Kinases in Cell Lysates of PCl2D Cells-To investigate the effect of pretreatment on H-89 on various protein kinase activities in PCl2D cells, cell lysates of PC12D cells were used. Forty ~1 each of all lysates with or without pretreatment of H-89 (30 PM) were used to measure protein kinases A, G, and C, Ca'+/calmodulin-dependent protein kinase, and Lu-casein phosphorylation activity. Kinase reactions were initiated by addition of 10 pM [y-"P]ATP and incubated for 5 min at 30 "C. Protein kinase A activities of cell lysates in PC12D cells with or without pretreatment of H-89 (30 PM) were 1.3 + 0.6 and 6.7 + 1.6 (picomoles of P/5 min), respectively (Table IV). H-89 significantly inhibited protein kinase A activity in cell lysates of PC12D cells, but H-89 did not affect other kinase activity such as protein kinases G and C, Ca*+/calmodulin-dependent protein kinases, and a-casein phosphorylation activity. We also investigated the effect of H-89 on forskolin-activated kinase activity of cell lysate in PC12D cells. Histone IIb phosphorylation activity of cell lysate was measured without CAMP. After pretreatment of H-89 (30 PM) for 1 h in PC12D cells, forskolin (10 PM) was added and incubated for 30 min, and then the cells were immediately washed two times with cold Krebs-Henseleit buffer. The cells were quick-frozen in liquid nitrogen. Aliquots of dissolved cell lysates were used to investigate histone IIb phosphorylation activity. Phosphorylation activity was measured by incubating at 30 "C for 5 min after addition of [Y-~~P]ATP. Forskolin (10 PM)-activated histone IIb (1 mg/ml) phosphorylation activities of cell lysates of PC12D cells with or without H-89 (30 PM) were 0.7 + 0.3 and 6.9 f 3.5 (picomoles of P/5 min), respectively. These results suggest that H-89 also selectively inhibited protein kinase A activity of PCl2D cells.

DISCUSSION
There is some discrepancy regarding the effective dose of H-89 between in vitro and in uiuo systems. This can be   cr-Casein phosphorylated 1.3 + 0.9 1.5 + 1.0 explained as follows: presumably it is related to the permeability of H-89 in cell membranes and the high concentration of ATP in the initial phase of the experiment, as observed in the case of H-8 (Tash et al., 1986). Results in Table IV support this assumption. Thus, pretreatment of PCIBD cells with 30 gM H-89 occurred to selectively inhibit protein kinase A activity without inhibition of other kinase activities in each lysate. Our present results provide the first evidence that forskolin-induced neurite outgrowth is mediated through protein kinase A. No direct evidence regarding the role of protein kinase A on neurite outgrowth of PC12 cells by forskolin has been presented, even though intracellular CAMP levels were reported to be elevated after the treatment of the cells with forskolin.
It has been argued whether CAMP mediates the action of NGF on neurite outgrowth.
Detailed biochemical mechanisms related to NGF-induced neurite outgrowth have yet to be determined. Some workers suggested that the CAMP level increased after the exposure of NGF in PC12 cells and that CAMP is a second messenger of NGF (Schubert and Whitlock, 1977;Schubert et al., 1978), while several groups reported that NGF has the ability to enhance CAMP levels but that CAMP is not the second messenger of NGF (Gunning et al., 1981;Greene et al., 1979). It was reported that NGF is unable to enhance CAMP levels in PC12 cells (Hatanaka et al., 1978). Our results suggest that even if NGF is able to enhance CAMP levels in PCl2D cells, NGF-induced neurite outgrowth is not mediated through protein kinase A. H-89 is one derivative of H-8, a specific inhibitor of cyclic nucleotidedependent protein kinase (Hidaka et al., 1984). The antiprotein kinase A activity of H-89 is 30 times more potent than that of H-8, whereas the anti-protein kinase G activity is 10 times less potent. These newly synthesized selective protein kinase A inhibitors H-88 and H-89 should serve as useful tools for clarifying the physiological roles of protein kinase A.