A New Role for IQ Motif Proteins in Regulating Calmodulin Function*

IQ motifs are found in diverse families of calmodulin (CaM)-binding proteins. Some of these, like PEP-19 and RC3, are highly abundant in neuronal tissues, but being devoid of catalytic activity, their biological roles are not understood. We hypothesized that these IQ motif proteins might have unique effects on the Ca2+ binding properties of CaM, since they bind to CaM in the presence or absence of Ca2+. Here we show that PEP-19 accelerates by 40 to 50-fold both the slow association and dissociation of Ca2+ from the C-domain of free CaM, and we identify the sites of interaction between CaM and PEP-19 using NMR. Importantly, we demonstrate that PEP-19 can also increase the rate of dissociation of Ca2+ from CaM when bound to intact CaM-dependent protein kinase II. Thus, PEP-19, and presumably similar members of the IQ family of proteins, has the potential to alter the Ca2+-binding dynamics of free CaM and CaM that is bound to other target proteins. Since Ca2+ binding to the C-domain of CaM is the rate-limiting step for activation of CaM-dependent enzymes, the data reveal a new concept of importance in understanding the temporal dynamics of Ca2+-dependent cell signaling.

nals it is know to sample. For example, the kinetics of Ca 2ϩ binding to the C-domain of CaM are too slow (3)(4)(5)(6) to respond to rapid Ca 2ϩ transients such as those found in excitable cells (2,7,8). This has focused attention on mechanisms that could help tune the Ca 2ϩ binding properties of CaM to expand its potential to respond to specific Ca 2ϩ signals. Thus far, changes in the Ca 2ϩ binding properties of CaM have been observed only as an increase in affinity due to a decrease in the Ca 2ϩ dissociation rate upon binding Ca 2ϩ -CaM to target proteins and peptides (3)(4)(5)(6)9). The magnitude of this effect is dependent on a given target enzyme and can be so large as to promote constitutive association of CaM with targets even at basal Ca 2ϩ levels.
It would be of great functional significance if proteins were discovered that could increase, rather than decrease, the rate of dissociation of Ca 2ϩ from free or target-bound CaM or potentially modulate the rate of Ca 2ϩ association. This would be particularly important for the C-domain of CaM, which exhibits Ca 2ϩ binding kinetics that are too slow to respond to rapid Ca 2ϩ transients such as those found in excitable cells. Our search for such proteins led to analysis of the small, neuronal IQ motif proteins, or SNIQs, that include neuromodulin (Nm or GAP-43), neurogranin (Ng or RC3), and PEP-19 (for review see, Refs. 10 and 11). The SNIQs are highly abundant (up to 50 M) in neuronal tissues but have no known catalytic activity. Their ability to bind to apo-CaM supports the idea that SNIQs modulate effective levels and/or distribution of free CaM at basal Ca 2ϩ levels (12). However, Nm binds CaM equally well in the presence or absence of Ca 2ϩ at physiological salt concentrations (13). We hypothesized that these CaM binding properties could potentially affect the rates of both association and dissociation of Ca 2ϩ from CaM and thus provide an alternate function for the SNIQs that could have pervasive effects on CaM activity. We show here that PEP-19 accelerates the rates of association and dissociation of Ca 2ϩ from the C-domain of free CaM, and of CaM when bound to CaM-dependent protein kinase II ␣ (CKII␣).

EXPERIMENTAL PROCEDURES
Expression of PEP-19 -A human cDNA for PEP-19 was purchased from ResGen (IMAGE expressed sequence tag clone 4792589). The amino acid coding reading for PEP-19 was amplified using a 5Ј-primer (GTTGAGTTAGAGCCACCATGGCTGAGCGACAAGGTGC), the 3Јprimer (CGGAACTGCTAGCTTGGATCCATCAGGACTGAGACCCAG-CC), and subcloned into the NcoI and BamHI sites of the pET23d expression plasmid. The N terminus was modified from MSERQ to MAERQ to enhance expression levels of the recombinant protein.

Isolation of CaM and PEP-19 -Recombinant
CaM was expressed and isolated as described previously (14). CaM was labeled with 15 N as described previously for cardiac troponin C (15). PEP-19 was isolated as described in the Supplemental Material.
Equilibrium Ca 2ϩ Binding-Macroscopic equilibrium Ca 2ϩ binding constants were determined using the competitive binding assay described by Linse et al. (16).
Stopped-flow Measurements-Stopped-flow fluorescence experiments were acquired using an Applied Photophysics Ltd. (Leatherhead, UK) model SX.18 MV sequential stopped-flow spectrofluorimeter with a 150-watt xenon/mercury lamp and a dead time of 1.7 ms. We found that concentrations of 2 M CaM, 20 M Ca 2ϩ , and 300 M Quin-2 were sufficient to fully extract Ca 2ϩ from CaM complexes and the CaM⅐PEP-19 complex. All solutions contained a base buffer of 20 mM MOPS, pH 7.5, 100 mM KCl.
Fluorescence from Quin-2 was detected using excitation wavelengths of 334 or 334.5 nm and an Oriel emission cut-off filter 51282. Tyrosine fluorescence was detected using an excitation wavelength of 266 nm * This work was supported in part by National Institutes of Health Grants HL45724 and NS26086 and Robert A. Welch Foundation Grant AU1144. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. and Oriel filter 51662. All stopped-flow experiments were done at room temperature (23°C). CaM DANS was excited at 334 nm, and fluorescence was monitored at 485 nm.
NMR Methodology-Decalcified [ 15 N]CaM was lyophilized and resuspended to 0.5 mM in decalcified 100 mM KCl, 5% D 2 O. Calcium was added to 20 mM, and the pH of the sample was adjusted to 6.3. Titration of PEP-19 was done by additions from an 8 mM stock prepared in water.
All NMR experiments were collected on a DRX 600 MHz spectrometer instrument using a 5 mm TXI probe at 25°C for apo-[ 15

RESULTS
CaM has four Ca 2ϩ binding sites: two in the N-domain and two in the C-domain. Fig. 1A shows the rate of dissociation of Ca 2ϩ from CaM detected using the Ca 2ϩ -sensitive fluorescent dye Quin-2. Release of Ca 2ϩ from the N-domain is too rapid (Ͼ1000 s Ϫ1 ) to observe using conventional stopped-flow at room temperature (4). Thus, only the release of 2 mol of Ca 2ϩ is observed per mol of free CaM with a koff of 8.5 s Ϫ1 , which is identical to the value reported by Brown et al. (6). This observed rate of release of Ca 2ϩ was increased by a factor of 50 in the presence of PEP-19. Tyrosine fluorescence is a marker for Ca 2ϩ binding to the C-domain of CaM. The inset to Fig. 1A uses Tyr fluorescence to confirm that PEP-19, which has no Tyr residues, greatly accelerates the dissociation of Ca 2ϩ from the C-domain of CaM. Fig. 1B and Table I show that PEP-19 has little effect on equilibrium Ca 2ϩ binding to CaM. Given that PEP-19 increases the rate of dissociation of Ca 2ϩ from the C-domain of CaM, the rate of association of Ca 2ϩ must also be accelerated by PEP-19 to explain the unchanged equilibrium binding constant. We experimentally verified this relationship by measuring the rate of association of Ca 2ϩ with the C-domain of CaM using Tyr fluorescence. It is clear from the inset to Fig. 1B that the rate of association of Ca 2ϩ with the C-domain of CaM is significantly increased by PEP-19. Table I summarizes the respective Ca 2ϩ binding affinities and kinetic rate constants of N-and C-domains of CaM in the presence and absence of PEP-19. Values for the N-and C-domain of free CaM shown in Table I are consistent with those reported previously by others (3)(4)(5)(6). PEP-19 accelerates the rates of association and dissociation of Ca 2ϩ at the C-domain of CaM 40 -50-fold but has essentially no  The equilibrium calcium binding constant (K eq ) was derived from the data in Fig. 1B as described in the on-line Supplemental Material. The dissociation rate (k off ) was derived from the data in Fig. 1A. The association rate was calculated from K eq ϭ k on /k off .
38 Ϯ 4.4 430 Ϯ 50 170 effect on the equilibrium Ca 2ϩ binding constants for either the N-or C-domain. Based on the results in Fig. 1 and Table I Fig. 2A.
The histogram in Fig. 2A summarizes the effect of PEP-19 on CaM amide chemical shifts. All changes of greater than 1 S.D. are found in the C-domain and are clustered in two regions. The first cluster includes amino acids 105-117 spanning helix F and the linker between helix F and G. Binding of PEP-19 to this region on CaM could be responsible for affecting the Ca 2ϩ binding properties of sites III and IV. The second cluster includes amino acids 142-148 at the very C terminus of CaM. Fig. 2B illustrates the location of these clusters on the solvent accessible surface of the C-domain of Ca 2ϩ -CaM. The structure on the left emphasizes the hydrophobic surface of CaM shown in blue, which is now known to play a critical role in binding a variety of Ca 2ϩ -dependent CaM-binding proteins (18). Residues that are affected by PEP-19 are shown in red. This structure was rotated 180°to obtain the view on the right. Helix F is indicated in both views as a point of reference. Two regions on CaM are affected by PEP-19. Although these surfaces appear to be distinct, and to border the central portion of the hydrophobic pocket, this analysis does not account for potential conformational changes in CaM induced by PEP-19, which could present a contiguous binding surface. Nevertheless, it is clear from the two views of CaM that the most expansive surface that is affected by PEP-19 includes helix F and extends well away from the hydrophobic pocket. This suggests that PEP-19 could form ternary complexes with CaM and other binding proteins and potentially affect the Ca 2ϩ binding properties of the complex.
We used recombinant CaM-dependent protein kinases II ␣ (CKII␣) to determine the potential of PEP-19 to affect the Ca 2ϩ binding properties of CaM when bound to another target protein. CKII␣ binds CaM with high affinity (K d 10 Ϫ7 to 10 Ϫ8 M) that can be further enhanced by CaM-dependent autophosphorylation, but CaM/CKII␣ is fully active in the absence of autophosphorylation (19). Fig. 1C shows that 4 mol of Ca 2ϩ /mol of CaM are released when CaM is bound to CKII␣, with rate constants of 0.9 and 12 s Ϫ1 , respectively. Thus, association of CaM with CKII␣ must greatly slow the release of Ca 2ϩ from the N-domain, since this event is too fast to be observed with free CaM. We cannot currently assign these two rates to specific domains in CaM, however, the presence of PEP-19 greatly increased the fast rate from 12 to ϳ400 s Ϫ1 .
The results in Fig. 1C could be due to formation of a ternary complex between CaM, CKII␣, and PEP-19 or due to displacement of CaM from CKII␣ by PEP-19. We used IAEDANSlabeled CaM (CaM DANS ) to distinguish between these possibilities. Fluorescence from CaM DANS is increased upon binding to CKII␣ (14,20), but association with PEP-19 has little effect on fluorescence intensity. If PEP-19 displaced CaM DANS from CKII␣, then it should reverse the CKII␣-induced increase in fluorescence. The results of this experiment are shown in the inset to Fig. 1C. Addition of up to 40 M PEP-19 does not reverse the CKII␣-induced increase in fluorescence from CaM DANS , which supports the formation of a ternary complex between CaM, CKII␣, and PEP-19. Together, these data demonstrate that PEP-19 can affect the kinetics of Ca 2ϩ binding to CaM even when it is bound to another protein. DISCUSSION The rates of association and dissociation of Ca 2ϩ at the C-domain of CaM are up to 2 orders of magnitude slower than the N-domain (see Table I). Thus, Ca 2ϩ binding to the Cdomain is rate-limiting for activation of targets that require Ca 2ϩ -saturated CaM and will dictate its temporal response to Ca 2ϩ signals. The effect of PEP-19 on the C-domain of CaM, and localization of PEP-19 and other SNIQs to neurons, suggest that these proteins play a role in modulating the Ca 2ϩ binding properties of CaM so it can respond to challenging Ca 2ϩ signals, such as high frequency action potentials, which control Ca 2ϩ -dependent processes in pre and post-synaptic compartments. Fig. 3 simulates the effect of PEP-19 on Ca 2ϩ binding to the C-domain of CaM during a train of Ca 2ϩ pulses of 50 Hz and amplitude of 1.5 M. In the absence of PEP-19, intrinsic slow on and off Ca 2ϩ binding kinetics lead to a gradual increase in the percent saturation of the C-domain of CaM with Ca 2ϩ . In the presence of PEP-19, Ca 2ϩ binding to the C-domain of free CaM more closely parallels the rise and fall of free Ca 2ϩ . This illustrates that the C-domain of CaM could respond to Ca 2ϩ in distinct ways depending on the presence or absence of PEP-19. The effect of other SNIQs on CaM will likely be finetuned by differences in primary sequences. For example, Ng accelerates the dissociation of Ca 2ϩ from CaM but has little effect on the association rate, which results in decreasing the overall Ca 2ϩ binding affinity of the C-domain of CaM. 2 Posttranslational modification, such as phosphorylation of Ng and Nm that inhibits association with CaM (13,21,22), would provide another level of regulation. These data and simulations emphasize that the influence of PEP-19, and other proteins with similar activity toward CaM, should be incorporated into models of neuronal Ca 2ϩ dynamics and CaM activation that use stochastic approaches to account for CaM diffusion and spatial constraints.
Current structural models for the regulation of target proteins by CaM invoke distinct functions for its N-and C-domains. IQ CaM binding motifs found in diverse proteins may help facilitate these interactions. For example, Ca 2ϩ -dependent facilitation and inactivation of L-type and P/Q-type voltagedependent Ca 2ϩ channels (23)(24)(25)(26)(27) is due largely to the temporal and differential binding of the N-and C-domains of CaM to the IQ motif of these channel proteins. DeMaria et al. (27) proposed that the C-domain of CaM mediates channel facilitation by responding to millisecond spike-like elevations in local Ca 2ϩ that result from opening of individual channels. The data presented here suggest that the IQ motif in voltage-operated Ca 2ϩ channels could modulate the Ca 2ϩ binding properties of the C-domain of CaM such that it can appropriately respond to calcium spikes.
Calcium binding is the fundamental property of CaM that allows it to function as a central regulatory protein in a plethora of signal transduction pathways. From a general perspec-tive, the ability of CaM/protein interactions to modulate the Ca 2ϩ binding properties of CaM should be broadly recognized as an important facet of the overall mechanism of action of CaM. The modulator activity of CaM-binding proteins can achieve a tremendous dynamic range of Ca 2ϩ binding kinetics. For example, the rate of dissociation of Ca 2ϩ from the Cdomain of CaM bound to myosin light chain kinase (3) is at least 1000-fold slower than when CaM is bound to PEP-19. The ability of PEP-19 to affect the Ca 2ϩ binding properties of CaM when bound to other target proteins provides another dimension of functionality in that it has the potential to antagonize target-induced increases in the Ca 2ϩ binding affinity of CaM. The potential for PEP-19 to overcome rate-limiting binding of Ca 2ϩ to the C-domain of CaM would make it analogous to GTPase-activating proteins, regulators of G-proteins signaling, and guanine nucleotide exchange factors in that they all overcome rate-limiting steps to allow regulatory proteins to act in biologically meaningful time frames. Potential Ca 2ϩ -binding modulator activity that is targeted toward other Ca 2ϩ -dependent regulatory proteins may play important roles in a variety of Ca 2ϩ signaling pathways.
In summary, the data presented here provide evidence for a new biological role for PEP-19 as a modulator of the Ca 2ϩ binding properties of CaM. This activity could have broad effects on the activation of CaM-dependent proteins. PEP-19, and other homologous IQ motif CaM binding proteins with similar activity, would be important additions to the Ca 2ϩ -signaling toolkit (1) used to assemble Ca 2ϩ /CaM signaling systems with different spatial and temporal dynamics.