Expression of enzymes of covalent protein modification during regulated and dysregulated proliferation of mammary epithelial cells: PKA, PKC and NMT.

Three proteins are functionally interlinked in the targeting of protein phosphorylation catalyzed by the C-subunit of PKA: PKA itself, AKAPs and NMT. Furthermore, in a variety of biological contexts, mechanisms exist whereby PKA and PKC are able to modulate the activity of one another. We have investigated the expression and subcellular distribution of these proteins in two models of mammary cell proliferation and differentiation--the normal rat mammary gland during pregnancy and lactation and human breast tissue before and after malignant transformation. Modulation of PKA does not acutely affect activity or sub-cellular distribution of PKC in mammary acini, nor does modulation of PKC acutely affect PKA activity or subcellular distribution. Therefore, the co-ordinate expression of these two protein kinases in normal and cancerous mammary epithelial cells and the greater basal activation level of them both accompanying increased mitogenic activity, which we have reported, does not result from short-term cross-talk between them. Although basal and total levels of PKA diminish in rodent mammary epithelial cells during the transition from proliferative to secretory functional mode, the level of expression of AKAPs increases. The expression of two apparently mammary-specific and mostly membrane-associated AKAPs is tightly linked to the onset and maintenance of differentiated function in rat mammary tissue. Paradoxically, the probable analogues of these two AKAPs in human mammary tissue are hyperexpressed when normal epithelial cells transform to a cancer phenotype--conventionally regarded as a process involving a degree of dedifferentiation. Mammary AKAP hyperexpression in breast cancers is accompanied by increases in the levels of total and basal PKA. One mechanism whereby NMT is targeted to membranes, via interaction with ribosomal proteins, has recently been elucidated. Our data support the contention that the localization of NMT is an important variable in the regulation of cellular proliferation, but they do not characterize the mechanisms whereby the differential targeting of NMT is achieved. As yet we lack a full tool-kit with which to examine NMT either to draw firm conclusions regarding the identity of particular isoforms found in particular sub-cellular locations or to define the relationships between these different molecular variants. However, it is technically possible to transfect cells with inducible NMT expression constructs engineered in such a way that the recombinant, catalytically competent, NMT that they encode is targeted either to membranes or to cytosol: an exploration of the effects of such transfections on cellular proliferation would afford a critical test of the mechanistic involvement of NMT in the control of mitogenesis.

Elevation of protein kinase A and protein kinase C activities in malignant as compared with normal human breast tissue.

Because of their central role in the transduction of extracellular signals, protein kinases A (PKA) and C (PKC) are critical enzymes in the control of cellular proliferation and differentiation. We have measured the catalytic activity of PKA and PKC, as well as the regulatory subunit expression for PKA, in paired samples of normal and malignant breast tissue from 13 patients with breast cancer. Paired non-parametric (Wilcoxon) analysis revealed significantly higher values for both basal (P = 0.0002) and total (P = 0.0002) PKA catalytic activity in malignant compared with normal breast in all 13 paired tissue samples. Expression of both R1- and RII-PKA regulatory subunits were also higher in malignant tissue from 12 (P = 0.0005) and 9 (P = 0.01) of the 13 pairs, respectively. However, the degree of RI-subunit overexpression in malignant tissue was greater than that of the RII-subunit, as demonstrated by an increase in the RI/RII subunit ratio in 10 of the 13 paired samples (P = 0.017). Total PKC catalytic activity was elevated in 11 of the 13 malignant tissue specimens when compared with corresponding normal breast tissue (P = 0.01). This was accounted for by an increase in Ca(2+)-dependent PKC activity (P = 0.01), there being no significant increase in Ca(2+)-independent PKC activity. These data suggest that the activities of both PKA and PKC signalling pathways are intrinsically higher in malignant compared with normal breast tissue and these may therefore represent targets for interventive treatment of breast cancer.

Properties of a resiniferatoxin-stimulated, calcium inhibited but phosphatidylserine-dependent kinase, which is distinct from protein kinase C isotypes alpha, beta 1, gamma, delta, epsilon and eta.

We have separated a resiniferatoxin-stimulated histone-kinase activity from human neutrophils, elicited mouse macrophages and murine alveolar macrophages by hydroxyapatite chromatography. The assay conditions for resiniferatoxin kinase were optimized as part of this study and in the presence of phosphatidylserine but absence of Ca2+ the Ka for histone IIIs phosphorylation by resiniferatoxin was calculated as 16 nM. Using a phosphate gradient of 20-500 mM, peaks of protein kinase C activity could be washed from the hydroxyapatite column in 300 nM phosphate and resiniferatoxin kinase recovered in 500 mM phosphate. At the optimum concentration of 160 nM, the ability of resiniferatoxin to induce enzyme activity was compared with a range of phorbol esters all at the same concentration. These related compounds failed to activate resiniferatoxin kinase although they have previously been shown to activate protein kinase C isotypes. Similarly sn-1,2,-dioleoylglycerol and the potent irritant capsaicin at 30 microM failed to activate the kinase. A Scatchard analysis of [3H] phorbol dibutyrate binding produced a linear plot (Kd 41.6 nM; Bmax 11.6 fmol unit-1) and binding was inhibited by resiniferatoxin and 12-O-tetradecanoylphorbol-13-acetate (TPA), with resiniferatoxin 700 times more potent than TPA in this respect. A radiolabelled resiniferatoxin binding assay was also used to demonstrate specific binding of [3H]resiniferatoxin which could be inhibited by unlabelled compound. Resiniferatoxin kinase activity was shown to be distinct from the protein kinase C isotypes alpha, beta 1, gamma, delta and epsilon by means of immunological analysis and from the eta isotype, because that isotype was not stimulated by resiniferatoxin but was stimulated by TPA when a pseudosubstrate was used. In addition the resiniferatoxin-stimulated activity was inhibited in-vitro by the addition of Ca2+ (Ki 0.1-0.5 nM free Ca2+). Further purification of resiniferatoxin kinase by Superose chromatography indicated a major activity fraction of about 70-90 kDa. Thus resiniferatoxin kinase, isolated from human and mouse inflammatory cells is distinct from the known isotypes of protein kinase C and is a major resiniferatoxin receptor.

HL-60 cell differentiation induced by phorbol- and 12-deoxyphorbol-esters.

The human promyelocytic leukaemia cell (HL-60) undergoes differentiation into a macrophage-like form when exposed to both tumour promoting- and non-promoting phorbol esters. We have investigated the effect of the two non-promoting phorbol esters, 12-deoxyphorbol-13-O-phenylacetate (Dopp) and 12-deoxyphorbol-13-O-phenylacetate-20-acetate (Doppa) on HL-60 cultures, and compared them with the tumour promoter 12-O-tetradecanoylphorbol-13-acetate (TPA). All phorbol esters tested were found to be able to stop HL-60 proliferation and induce cell adherence and morphological changes characteristic of differentiation. TPA, fully differentiating at 1 nM, was more potent than Dopp and Doppa, which required 100 nM for full differentiation effects within the 4 day study. Doppa initially appeared weaker than Dopp at inhibiting incorporation of thymidine, the earliest effect studied, but we were able to detect rapid C-20 deacylation of Doppa, converting it to Dopp, using an HPLC protocol presented here. A detailed study of this thymidine incorporation inhibition showed that both TPA (10 nM or greater) and Dopp (500 nM or greater) have very similar time courses, with 50% inhibition occurring at approximately 12 h, in contrast to Doppa which had a significantly delayed time course at all doses tested. Exposure tests indicated that Dopp and Doppa could be washed from the cells much more easily than TPA. The data presented here strongly support the notion that the metabolic conversion of Doppa to Dopp by HL-60 cells was necessary to mediate its differentiating effects. Since protein kinase C (PKC)-beta 1, present in HL-60 cells, has been found to be the only PKC isotype activated so far in vitro by Doppa, our results suggest that activation of this isotype is not sufficient to drive HL-60 differentiation in vivo.

12-Deoxyphorbol-13-O-phenylacetate-20-acetate is not protein kinase C-beta isozyme-selective in vivo.

The phorbol ester, 12-deoxyphorbol-13-O-phenylacetate-20-acetate (DOPPA) has been shown to activate specifically the protein kinase C (PKC)-beta 1 isozyme in vitro (1). We have investigated the potential of DOPPA as a PKC-beta 1/2 isozyme-specific agonist in intact cells, employing U937 cells, which express beta 1/2, epsilon and zeta PKC and in Swiss 3T3 cells which lack PKC-beta 1/2 but express alpha, delta, epsilon and zeta PKC. Immunoblot analysis with isozyme-specific antibodies indicated that DOPPA can mediate the subcellular redistribution and down-modulation of all endogenous PKC isozymes (except PKC-zeta) in both U937 and Swiss 3T3 cells. Prolonged treatment (> 6 h) of cultures in down-modulation studies is complicated by the metabolism of DOPPA to 12-deoxyphorbol-13-phenylacetate (DOPP), a compound which activates all PKC isozymes tested in vitro (Ryves, W. J., et al. (1991) FEBS Lett., 288, 5-9). Nevertheless, because DOPPA induced rapid and dose-dependent phosphorylation of p80 in cells which do not express PCK-beta, p80 phosphorylation in Swiss 3T3 cells indicates that DOPPA can activate a non-beta PKC in vivo. The data suggest that DOPPA cannot be used as a PKC-beta-selective agonist in intact cell studies.

Molecular modelling of tumour-promoting and non-promoting diterpene esters and molecular co-ordinates of the potent irritant resiniferatoxin.

The lowest energy conformer of seventeen diterpenes, representing five different phorbol and daphnane diterpene nuclei, has been generated. TPA possessed the highest minimum free energy of these compounds; all other compounds possessed a lower minimum free energy. Compounds based on the resiniferonol nucleus possessed the lowest minimum free energy (9.4-16.6% of that of TPA). The molecular co-ordinates of the non-promoting but potent irritant resiniferatoxin (Rx) are also reported. These studies may be important in elucidation of the biochemical mechanisms of action of diterpene esters, including an understanding of the interactions of diterpene esters with the phorbol ester binding domain of the protein kinase C isoform family.

The potent irritancy of the daphnane orthoester, resiniferatoxin, exhibits features of a mixed aetiology.

Resiniferatoxin-induced erythema of mouse ear was shown to possess characteristics of both a phorbol ester-mediated response and that induced by the neurogenic irritant, capsaicin. Whereas the response to the phorbol ester, sapintoxin D, was delayed and prolonged, and was augmented by capsaicin pretreatment, the response to resiniferatoxin was biphasic, with the early phase being antagonized by capsaicin desensitization. However, resiniferatoxin was most potent in inducing a delayed erythema which, unlike the capsaicin response, was sensitive to inhibition by low dose hydrocortisone treatment, but not to chronic capsaicin desensitization. It is concluded that the erythema response to resiniferatoxin has a mixed aetiology, which may explain the unique potency of this toxin.

Platelet protein phosphorylation and protein kinase C activation by phorbol esters with different biological activity and a novel synergistic response with Ca2+ ionophore.

Phorbol esters with different biological activities have been tested for their ability to induce the phosphorylation of human platelet proteins. We have shown that only the potent platelet aggregatory phorbol esters were able to stimulate the phosphorylation of proteins of 76, 68, 47, 30 and 20 kDa in intact platelets. The ability of these esters to stimulate phosphorylation of the 47-kDa protein ('p47') correlated with their ability to cause platelet aggregation. When a non-platelet aggregatory deoxyphorbol (12-deoxyphorbol 13-phenylacetate 20-acetate) was combined with a subthreshold dose of the Ca2+ ionophore, A23187, a large increase in phosphorylation of p47 and a fourfold decrease in Ka was observed. This was in contrast to a barely detectable stimulation of phosphorylation at micromolar levels of this phorbol ester in the absence of the ionophore. This synergism was not evident for the potent platelet aggregatory derivatives. The Ka for DOPPA with a mixture of total platelet protein kinase C was 530 nM in the absence of calcium decreasing to 120 nM in the presence of calcium. In the presence of calcium, 12-deoxyphorbol 13-phenylacetate 20-acetate was shown to stimulate preferentially one of the isoforms of protein kinase C.