Terbium as a Luminescent Probe of Metal-binding Sites in Protein Kinase C*

In the present report, we demonstrate that Tb3’ binds to protein kinase C and serves as a luminescent re- porter of certain cationic metal-binding sites. Tb3’ titration of 50 nM protein kinase C results in a 20-fold enhancement of Tb3’ luminescence which is half-max- imal at 12 j&M Tb3’. A Kd of -145 nM was determined for Tb3’ binding to the enzyme. The excitation spectrum of bound Tb3+ exhibits a peak at 280 nm charac- teristic of energy transfer from protein tryptophan or tyrosine residues. The luminescence of this complex can be markedly decreased by other metals, including and chelation 2 EGTA. Tb3’ binding to protein kinase is correlated

In the present report, we demonstrate that Tb3' binds to protein kinase C and serves as a luminescent reporter of certain cationic metal-binding sites. Protein kinase C plays a major role in cell regulation through its involvement in signal transduction across the plasma membrane (1,2). Its activation requires phospholipid, diacylglycerol/phorbol esters, and Ca'+, which interact with a regulatory domain in the amino-terminal half of the enzyme. The enzyme occurs as a family of isoenzymes which all contain cysteine-rich regions and putative metal-binding sites in their regulatory domain (2)(3)(4)(5). In its regulatory domain is a tandem repeat of a cysteine-rich sequence characteristic of certain metalloproteins and DNA-binding proteins (6). This region does not appear to be involved in Ca2+ binding, but is essential for phorbol ester binding (7). Between the cysteine repeats and the catalytic domain of type 111 protein kinase C is a sequence (residues 292-303) which resembles the F helix of the Ca2+-binding E-F hand structure found in calmodulin and other Ca'+-binding proteins (4 sequence (including residues 279-287) may also function as Ca"'-binding sites, but are divergent from the E-F hand model (3,4).
Several previous studies of heavy metal effects on protein kinase C have suggested that Zn'+ and CW' can stimulate and inhibit protein kinase C activity in a biphasic manner (9-11). These studies utilized assay systems containing Ca'+-EGTA. ' Speizer et al. (8)  (15). In the presence of 5 pg/ml phosphatidylserine, its activity was stimulated at least 5-fold by 200 pM Ca".
The Bradford method (16) was used for protein determinations. Protein kinase C activity was assayed essentially as described by Kikkawa et al. (17). The reaction was initiated by the addition of enzyme, incubated at 30 "C for 2 min, and terminated by the addition of 30% trichloroacetic acid. Phosphoproteins were isolated by filtration (Millipore HAWP) and quantitated by liquid scintillation counting. A modification of the method of Tanaka et ~2. (18) was used to assay phorbol ester binding. The assay mix was incubated at 30 'C for 30 min, and terminated by the addition of 3 ml of cold 0.5% dimethyl sulfoxide. The mixture was filtered through Whatman GF/ C filters soaked with 0.3% polyethylenimine, washed twice with two 3-ml volumes of 0.5% dimethyl sulfoxide, and dried for scintillation counting. Nonspecific [3H]PDBu binding (assayed in the presence of 30 FM cold PDBu) was found to be slightly less than the value obtained in the absence of protein kinase C. The latter value was subtracted from the measured total bound [3H]PDBu to derive the specific binding data.
Prior to titration with Tb3+, protein kinase C aliquots were desalted with Spectra/Gel HW-40F (Spectrum) to remove EDTA and EGTA. Luminescence studies were conducted with a Perkin-Elmer LS-5B luminescence spectrometer, using gate and delay times of 1 ms. Excitation and emission wavelengths were 280 and 543 nm, respectively. Titrations were conducted in buffers made from high purity water containing 10% glycerol (Boehringer Mannheim enzyme grade) to stabilize the dilute enzyme. The reported luminescence values were corrected for minor background luminescence attributable to interaction of terbium with the buffer.

AND DISCUSSION
Tb3' titrations of 50 nM protein kinase produced an increase in Tb3' luminescence which was half-maximal near 12 FM and maximal at 20 pM Tb3' (Fig. 1) C. Protein kinase C (50 nM) was titrated in a buffer of 10 mM MOPS (pH 6.8) containing 90 mM KC1 and 10% glycerol. Luminescence was monitored at excitation and emission wavelengths of 280 and 543 nm, respectively, and is shown as a function of total Tb3+ added.
Inset, luminescence excitation spectra of 50 nM protein kinase C (spectrum I), 20 pM Tb3+ (spectrum 2), and 50 nM protein kinase C with 20 pM Tb3+ (spectrum 3) all obtained in the same buffer. At 20 @M Tb3+, the luminescence in the presence of protein kinase C was 20-fold greater than the luminescence of Tb3' alone. Addition of 2 mM EGTA to Tb3+ and protein kinase C partially reversed the luminescence increase (spectrun 4). For all excitation spectra, the emission wavelength was fixed at 543 nm. aromatic amino acid residues of the enzyme to bound Tb3'. The luminescence increase could be 67% reversed by chelation of Tb3' by 2 mM EGTA ( Fig. 1, inset, spectrum 4). The luminescence of Tb3+ was enhanced 20-fold by the presence of protein kinase C (compare spectra 2 and 3). In the absence of Tb3+, protein kinase C did not contribute significantly to luminescence (spectrum I). The increase in Tb3+ luminescence produced by protein kinase C was not observed after heat denaturation of the enzyme (100 "C for 5 min). Further, proteins which do not have metal binding sites (including bovine serum albumin and cY-chymotrypsin) produced no increase in Tb3+ luminescence under the same conditions used to study protein kinase C. This is consistent with Tb3+ binding to a specific metal binding site which exists on the native conformation of protein kinase C.
Diacylglycerol or phorbol esters increase the affinity of protein kinase C for Ca*', permitting full activation at resting cytosolic Ca2+ concentrations (19). We conducted Tb3' titrations of protein kinase C in the presence of PMA and phosphatidylserine to determine whether these regulatory ligands affect Tb3+ binding. The titration curve was unchanged by the addition of 150 nM PMA, but was shifted to the right (half-maximal -30 pM, maximal -70 pM) by addition of 20 pg/ml phosphatidylserine (data not shown). Titrations conducted in the presence of both 150 nM PMA and 20 wg/ml phosphatidylserine were not significantly different from those conducted in the presence of phosphatidylserine alone. Thus, whether in the presence or absence of phosphatidylserine, PMA did not appear to alter the affinity of protein kinase C for Tb3+. Phosphatidylserine alone did, however, decrease the apparent affinity of the enzyme for Tb3'.
We examined the ability of Pb'+, He, Zn'+, La3' Ca*+, and MgZ+ to displace Tb3+ from its binding site in protein kinase C and reduce its luminescence. Fig. 2 shows the effects of these metals on Tb3+-protein kinase C luminescence. Lead was most effective in displacing Tb3', causing a 50% luminescence decrease at 25 pM and an 84% decrease at 200 pM. La3+, Mg2C, and Ca2+ were progressively less effective, causing 50% decreases at 50 PM, 300 ELM, and 6 mM, respectively. Zn*+ decreased luminescence by only 35% at 10 mM, and Mg2+ (l-20 mM) had minimal ability to displace Tb3'. Thus, Pb*+, La3+, Hg*+, and, to a lesser extent, Ca2+ and Zn'+, appeared to compete with Tb3+ for binding sites on the enzyme. The enzyme's M$+-binding site was clearly distinct from the site which binds Tb3'. Pb*+ was the only metal we examined that appears to bind with similar or higher affinity to the lanthanide-binding site of protein kinase C than Tb3+ and La3+ themselves.
We examined the effects of Tb3+ on protein kinase C Protein kinase C (50 nM) and 20 pM Tb3+ were titrated with each metal in the buffer described in Fig. 1. Luminescence is shown as a function of total metal added. Excitation was at 280 nm, and Tb3+ luminescence was monitored at 543 nm. activity and [3H]PDBu binding to determine whether alterations in these parameters are correlated with Tb3' binding to the enzyme (as monitored by luminescence).
In addition, we examined other metals in this context to permit comparison with Tb3+. Fig. 3 shows the effect of Tb3' and other metals on protein kinase C activity in the presence of 20 PM Ca'+. Each of the metals tested inhibited protein kinase C activity as follows: Tb3+ and La3+ with ICsc = 8 PM, Hge+ with ICsc < 1 pM, Pb2+ with I& = 15 pM, and Zn2+ with I&c = 20 PM. In agreement with the report of Speizer et al. (a), we saw no evidence of heavy metal stimulation of protein kinase C activity. Interestingly, the concentration dependence of Tb3+ inhibition of enzyme activity displayed a strong positive correlation (r = 0.99) with Tb3+ luminescence enhancement caused by its binding to the enzyme over the range of l-20 pM. The inhibition of protein kinase C activity by 20 pM Tb3+ appeared to be partially reversible, since 200 pM EGTA was capable of restoring activity by 28%. This is consistent with the observation that EGTA chelation of Tb3' partially reversed its binding to the enzyme, as monitored by Tb3+ luminescence (Fig. 1, inset). Tb3+ binding and inhibition of protein kinase C activity could not be fully reversed even by 2 mM EGTA, consistent with its high affinity for the enzyme. Inhibition of protein kinase C by H$+ occurred well below its I& for Tb3+ displacement from protein kinase C (300 PM), suggesting that H$+ inhibition was caused by interaction with sites distinct from those which bind Tb3'. The cysteinerich regions of protein kinase C have been shown to be particularly susceptible to interaction with H$+ (8). In the assay system utilized, maximal protein kinase C activity was produced by 200 pM added Ca", while 20 PM Ca2+ was sufficient to produce half-maximal activation. We examined metal inhibition of protein kinase C activity in the presence of both 20 pM and 200 ELM Ca2+. The pattern of inhibition by all the metals was similar at 20 pM and 200 pM Ca'+, suggesting that inhibition was not due to metal competition with Ca2+ at the Ca'+-binding site which controls activity.
Inhibition of [3H]PDB~ binding by H$+, La3+, and Tb3' (Fig. 4) was closely related to their inhibitory effects on protein kinase C activity. The ICsO values for Hg2+ (2 PM), Tb3' (15 FM), and La3+ (15 FM) inhibition of [3H]PDBu binding were reasonably close to their respective ICss values for inhibition of protein kinase activity. Tb3+ binding to protein kinase C (as monitored by luminescence over the range of l-20 PM) was positively correlated with its inhibition  Inhibition of protein kinase C activity by Tb3+, La3+, HP, Pb*+, and Zn2+ could not be reversed by a lo-fold increase in Ca2', suggesting that they did not compete for Ca'+-binding sites on the enzyme. We therefore examined the possibility that Tb3+ binding might be competitive with the binding of phosphatidylserine or phorbol esters. Speizer et al. (8) have shown that the phospholipid dependence of protein kinase C activation is altered by certain heavy metals, including Cu*+. As shown in Fig. 5, Tb3+ had a similar effect. Increasing concentrations of phosphatidylserine were able to overcome Tb3+ inhibition of protein kinase C activity. In the absence of Tb3+, 4 rg/ml phosphatidylserine caused half-maximal activation of protein kinase C. In the presence of 10 &M and 20 pM Tb3+, half-maximal activation required 10 pg/ml and 20 rg/ml phosphatidylserine, respectively. Lineweaver-Burk analysis (shown in the inset to Fig. 5) suggested that Tb3+ is apparently a competitive inhibitor with respect to phosphatidylserine (K; = 3 wM).~ Thus, Tb3+ binds to protein kinase * The methods used to study the mechanism of inhibition are most appropriate for use with soluble tigands which exist as monomers in solution. It should be pointed out that the sonically dispersed phosphatidylserine used in our assays was not monomeric, but probably existed as a mixture of unilamellar and multilamellar vesicles. Assays were conducted as described in Fig. 4 in the presence of 0.25-32 nM [3H]PDBu.
C to exclude phosphatidylserine and does not bind to protein kinase C when phosphatidylserine is bound. This finding is consistent with our observation that phosphatidylserine decreases the apparent affinity of the enzyme for Tb3+. Fig. 6 shows Scatchard analysis of [3H]PDBu binding to protein kinase C in the presence and absence of 20 pM Tb3+. Tb3' inhibited phorbol ester binding to the enzyme by decreasing the maximal extent of binding (B,,,)  binding assay used in this study is almost absolutely dependent on the presence of phosphatidylserine (18). It is therefore likely that Tb3+ decreases the number of phorbol ester binding sites by competition with the phosphatidylserine binding sites which allosterically control phorbol ester binding.
Despite their physical similarity, our studies do not suggest that lanthanides and Ca*+ share the same binding site in protein kinase C. Tb3+ and La3+ were unable to substitute for Ca*+ and stimulate enzyme activity. Unlike Ca*', the apparent affinity of protein kinase C for Tb3+ was unchanged by phorbol esters. Tb3+ and La3+ produced a pattern of inhibition that cannot be significantly reversed by a lo-fold excess of Ca*+. In addition, Ca2+ was relatively ineffective in displacing Tb3+ from its binding site in protein kinase C. The last finding suggests that if Ca*+ binds to the Tb3+-binding site it is with relatively low affinity. Alternatively, Ca2+ may bind at a site that is capable of altering the conformation of the Tb3+binding domain. As previously mentioned, protein kinase C is thought to contain more than one potential Ca'+-or metalbinding site (4). The primary structure of the regulatory domain of type III protein kinase C contains two potential metal-binding sites which lie in close proximity, are nonidentical, and are distinct from the cysteine-rich region (3,4). Tb3' appears to interact with protein kinase C at the phospholipid-binding site and apparently acts as a competitive inhibitor with respect to phosphatidylserine. While it is possible that Tb3+ could interfere with activation of the en-zyme by altering the conformation of phospholipid vesicles, this would probably not occur in the 5-20 PM concentration range over which Tb3+ produces inhibition.
Further, Tb3' binding inhibits protein kinase C activity over the same range in which it bound to the enzyme in our luminescence studies, suggesting a Tb3+-protein kinase C interaction in the absence of phosphatidylserine.
Moreover, the ability of phosphatidylserine to decrease the apparent Tb3+-binding affinity of protein kinase C is consistent with the competitive model of inhibition suggested by Lineweaver-Burk analysis. The mechanism of protein kinase C inhibition by Tb3+ is apparently similar to that of certain heavy metals, including Cu*+. Similar to Tb3+, Cu*+ inhibits activity by altering the phosphatidylserine dependence of protein kinase C activation and noncompetitively inhibits [3H]PDBu binding (8). Pb2+ and He appear to be able to displace Tb3+ from its binding site, suggesting proximity or overlap between the heavy metaland lanthanide-binding sites of protein kinase C. Like La3+, Pb*+ inhibits protein kinase C activity over nearly the same range in which it displaces Tb3' from the enzyme, suggesting that Pb*+ inhibition could be mediated by interaction with the Tb3+-binding site. It is likely that H$+ inhibits protein kinase C activity by interacting with a different site (perhaps a cysteine-rich region), since its IC,,, for displacement of Tb3+ is 300-fold greater than its IGo for inhibition of enzyme activity, and Hg2+ is known to interact with cysteine residues on proteins.
In conclusion, inhibition of protein kinase C by Tb3+, La3+, and certain heavy metals (including the environmental pollutants Pb'+, Cd'+, and Cu"+) is mediated by a metal-binding site on the native enzyme that is distinct from the Ca*+binding site which regulates enzyme activity. Metals binding to this site compete with phosphatidylserine binding and noncompetitively inhibit phorbol ester binding to inactivate the enzyme. Our studies suggest that if these metals were to interact with protein kinase C in duo, they could impair its activation and translocation by inhibiting its binding to membrane phosphatidylserine.
Many Ca*+-and diacylglyceroldependent cellular processes could be dramatically altered by the disruption of this essential protein kinase.