Amino acid substitutions in the FXYD motif enhance phospholemman-induced modulation of cardiac L-type calcium channels

Kai Guo, Xianming Wang, Guofeng Gao, Congxin Huang, Keith S. Elmslie, Blaise Z. Peterson


We have found that phospholemman (PLM) associates with and modulates the gating of cardiac L-type calcium channels (Wang et al., Biophys J 98: 1149–1159, 2010). The short 17 amino acid extracellular NH2-terminal domain of PLM contains a highly conserved PFTYD sequence that defines it as a member of the FXYD family of ion transport regulators. Although we have learned a great deal about PLM-dependent changes in calcium channel gating, little is known regarding the molecular mechanisms underlying the observed changes. Therefore, we investigated the role of the PFTYD segment in the modulation of cardiac calcium channels by individually replacing Pro-8, Phe-9, Thr-10, Tyr-11, and Asp-12 with alanine (P8A, F9A, T10A, Y11A, D12A). In addition, Asp-12 was changed to lysine (D12K) and cysteine (D12C). As expected, wild-type PLM significantly slows channel activation and deactivation and enhances voltage-dependent inactivation (VDI). We were surprised to find that amino acid substitutions at Thr-10 and Asp-12 significantly enhanced the ability of PLM to modulate CaV1.2 gating. T10A exhibited a twofold enhancement of PLM-induced slowing of activation, whereas D12K and D12C dramatically enhanced PLM-induced increase of VDI. The PLM-induced slowing of channel closing was abrogated by D12A and D12C, whereas D12K and T10A failed to impact this effect. These studies demonstrate that the PFXYD motif is not necessary for the association of PLM with CaV1.2. Instead, since altering the chemical and/or physical properties of the PFXYD segment alters the relative magnitudes of opposing PLM-induced effects on CaV1.2 channel gating, PLM appears to play an important role in fine tuning the gating kinetics of cardiac calcium channels and likely plays an important role in shaping the cardiac action potential and regulating Ca2+ dynamics in the heart.

  • FXYD1
  • Cav1.2
  • activation
  • deactivation
  • inactivation

the fxyd family of ion transport regulators were first defined by Sweadner and Rael (29) based on an invariant peptide sequence known as the FXYD motif. The FXYD motif is localized on the NH2-termini of these single membrane spanning proteins and consists of five amino acids (Pro-Phe-X-Tyr-Asp). Seven members of the FXYD family (FXYD1 through FXYD7) have been identified in mammals and are widely distributed in tissues that perform fluid and solute transport (kidney, colon, breast/mammary gland, pancreas, prostate, liver, lung, and placenta) and in electrically excitable tissues (nervous system and muscle). One role of the FXYD family of ion transport regulators is to act as tissue-specific auxiliary subunits of Na-K-ATPase (NKA), which is the principal enzyme regulating the distribution of Na+ and K+ concentration across cell membranes (6, 8, 12, 29). FXYD1, also known as phospholemman (PLM), consists of a single transmembrane segment flanked by 17 amino acid extracellular NH2-terminal and 36 amino acid intracellular COOH-terminal domains (21). PLM is abundantly expressed in the cardiac sarcolemma, where it has been shown to modulate NKA (7, 10, 15, 27, 37) and the Na+/Ca2+ exchanger (NCX) (1, 19, 28, 31, 32, 36, 38).

We recently demonstrated that PLM associates with and alters the gating behavior of L-type calcium channels (33). Ca2+ entry through L-type calcium channels is an important first step that leads to myocardial contraction. The dysregulation of this process can lead to severe cardiac pathologies such as Long QT syndrome (35). The modulation of L-type calcium channels by signaling pathways and associated proteins is important for many physiological and pathophysiological processes (5, 20, 25). Their importance in maintaining normal heart function is further demonstrated by their value as drug targets for the treatment of heart disease (14, 24). The cardiac L-type calcium channel is a heteromultimeric complex consisting of α1-, β-, and α2δ-subunits (18) and the ubiquitous Ca2+-sensing protein calmodulin (22). The α1-subunit (CaV1.2) contains the major pore-forming and gating domains required for its function, the β- and α2δ-subunits are involved in trafficking the channel to the plasma membrane and modulation of certain gating functions (5, 18), and calmodulin is important for promoting Ca2+-dependent changes in gating such as Ca2+-dependent facilitation and Ca2+-dependent inactivation (22). Thus CaV1.2-associated proteins such as PLM represent an important mechanism by which the cardiac action potential and Ca2+ influx during the cardiac action potential and cardiac function can be regulated.

The high degree of conservation of the PFXYD motif among all its family members in a wide range of species indicates that selective pressure has prevented this motif from undergoing spontaneous mutations. This rationale is supported by structural studies of the NKA α/β/FXYD10 complex that suggest aromatic residues in the FXYD motif may play a role in stabilizing the association between NKA α- and β-subunits (26). Together, these observations suggest that the PFXYD motif must possess important functional properties, yet no functional role for the PFXYD motif has been established for any member of the FXYD family of ion transport regulators. Here, we used site-directed mutagenesis and whole cell patch-clamp electrophysiology to demonstrate that two amino acids within the PFTYD segment of PLM are important for regulating PLM-induced changes in CaV1.2 channel kinetics. We were surprised to find that amino acid substitutions within the PFTYD segment enhance the ability of PLM to modulate CaV1.2 channel gating. We predict that the PFXYD segment provides a mechanism that enables PLM to fine tune the gating kinetics of cardiac calcium channels, thereby playing an important role in shaping the cardiac action potential.


Construction of PLM mutants.

cDNAs encoding wild-type canine PLM and the PLM mutants T10A and Y11A were gifts from Dr. Joseph Y. Cheung (Thomas Jefferson University). PLM P8A, F9A, and D12A mutants were constructed by site-directed mutagenesis using polymerase chain reaction (PCR). Briefly, primers were used to introduce the desired mutations into PLM cDNA at positions corresponding to amino acids 8, 9, and 12 of the mature PLM (reverse primer for P8A: 5′-CGA TCC GCA AGC TTT GGT AGT CGT AGG TGA ACG-3′, reverse primer for F9A: 5′-TCC GAT CCG CAA GCT TTG GTA GGC GTA GGT GAA CGG GTC GTG TTC CTG-3′, and reverse primer for D12A: 5′-TCC GAT CCG CAA GCT TTG GTA GGC GTA GGT GAA CGG GTC GTG TTC CTG-3′; the same forward primer was used for all three reactions: 5′-GTA CCG TCG ACG CGG CCG CTC GAG CCT AAG GTT GCT TGT TCT TTT TGC AG-3′). The amplified PCR products were purified and digested with XhoI/HindIII and subcloned into the XhoI/HindIII-digested pAdTrack-PLM expression vector. PLM D12K and D12C mutants were constructed by using QuikChange XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). Briefly, two synthetic oligonucleotide primers containing the desired mutations were used to introduce desired nucleotide substitutions into PLM cDNA by PCR (respective reverse primers for D12K and D12C: 5′-GAT CCG CAG GGA TTG GTA TTT GTA GGT GAA CGG GTC GTG-3′ and 5′-CGC AGG GAT TGG TAG CAG TAG GTG AAC GGG TC-3′; respective forward primers for D12K and D12C: 5′-CAC GAC CCG TTC ACC TAC AAA TAC CAA TCC CTG CGG ATC-3′ and 5′-GAC CCG TTC ACC TAC TGC TAC CAA TCC CTG CG-3′). The resulting PCR products were treated with Dpn I endonuclease to digest the parental DNA template and to select for mutation-containing synthesized DNA. Each PLM mutant was confirmed by qualitative restriction map analysis and directional DNA sequence analysis of the entire subcloned region. Functional expression of the mutant PLM was confirmed by Western blot analysis.

Cell culture and transfection.

HEK293 cells were grown at 37°C and 6% CO2 in DMEM-F12 medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. HEK293 cells were transiently transfected by calcium phosphate precipitation as described previously (34). All cDNAs were cloned into cytomegalovirus promoters to maximize expression levels and to minimize unwanted effects from endogenous proteins. Briefly, cDNAs encoding CaV1.2 (α1 subunit of the cardiac L-type calcium channel), α2δ, and β1b (all subcloned into pcDNA3.1, Invitrogen, Carlsbad, CA) were cotransfected with either empty pAdTrack expression vector [(−)PLM; Stratagene, La Jolla, CA], wild-type [(+)PLM], or mutant (P8A, F9A, T10A, Y11A, D12A, D12K, and D12C) PLM/pAdTrack at equal molar ratios. In addition, cDNA encoding green fluorescent protein (pEGFP-N3; Clonetech Laboratories, Mountain View, CA) was transfected at a molar ratio of 0.2 to aid in the identification of successfully transfected cells.

Western blot analysis.

Cells were transfected as described above except that Flag-tagged CaV1.2 was used to enable the assessment of CaV1.2 expressions levels using the mouse monoclonal anti-Flag M2 antibody (cat. no. F3165; Sigma-Aldrich, St. Louis, MO). The anti-PLM (C2) rabbit polyclonal antibody was a generous gift from Dr. J. Y. Cheung (Thomas Jefferson University). HEK293 cells were harvested 28 h after transfection in lysis buffer (20 mM Tris, pH 7.5, 1% Triton X-100, 10% glycerol, 137 mM NaCl, 2 mM EDTA, 25 mM β-glycerophosphate, 1 mM Na3VO4, and EDTA-free complete protease inhibitor cocktail from Roche). Cell lysates were centrifuged at 14,000 g at 4°C for 15 min, and protein concentrations in cleared lysates were determined by BCA protein assay reagents (prod. no. 23228 and 23224, Thermo Fisher Scientific, Waltham, MA). Protein (30 μg) in 1× loading buffer was incubated at 75°C for 20 min, centrifuged 30 s, and subjected to NuPAGE (Invitrogen, Carlsbad, CA) electrophoresis and Western blot analysis.


Whole cell patch-clamp recordings were performed as described previously (34). Briefly, whole cell currents were recorded at room temperature within 24–48 h posttransfection. Pipettes were pulled from borosilicate glass (1B150F-3; World Precision Instruments, Sarasota, FL) using a Sutter P-97 Flaming/Brown micropipette puller (Sutter Instruments, Novato, CA) and fire polished on a MF200 microforge (World Precision Instruments, Sarasota, FL). Pipettes had resistances of 2.5–3.5 MΩ when filled with internal solution. Ionic currents were recorded in a bath solution containing (in mM) 130 N-methyl-d-glutamine (NMG)-aspartate, 10 HEPES, 10 4 aminopyridine, 10 glucose, 1 MgCl2, and 10 BaCl2. The internal solution contained (mM) 140 NMG-MeSO3, 10 EGTA, 1 MgCl2, 4 Mg-ATP, and 10 HEPES. Osmolarity was adjusted to 300 mmol/kg with dextrose, and the pH was adjusted to 7.35 with 5 mM NaOH. The use of Na+-, K+- and Ca2+-free solutions enable the recording of islolated CaV1.2 currents by eliminating possible contamination by currents originating from the NKA and NCX. Changes in the association of mutant PLM with the NKA or NCX may occur but are unlikely to be detected in these studies because PLM expression, driven by a strong cytomegalovirus promoter, is quite high compared with the levels of endogenous NKA and NCX. In addition, PLM-induced regulation of CaV1.2 currents occurs in the presence and absence of α2δ and regardless of the β-subunit isoform used (B. J. Peterson and X. Wang, unpublished data), suggesting that PLM binding determinants are located on the calcium channel α1-subunit, CaV1.2. Data were acquired using a HEKA EPC 10 amplifier and PULSE/PULSEFIT software. Ionic currents were sampled at 10 kHz and filtered at 3 kHz. Series resistance was typically <8 MΩ and was compensated by 70%. Leak and capacitive transients were corrected by P/4 leak subtraction.

Data analysis.

Data were analyzed using FitMaster (HEKA Instruments, Bellmore, NY) and Origin (OriginLab, Northampton, MA). One-way ANOVA was used to evaluate the statistical significance. All data are means ± SE, and statistical significance is set at P < 0.05. Error bars smaller than symbols do not appear in figures. Data significantly different from (−)PLM are indicated with an asterisk and data significantly different from (+)PLM are indicated with a hatch mark.


Previously, we found that PLM regulates cardiac CaV1.2 channels by slowing both activation and deactivation and by speeding voltage-dependent inactivation (VDI) (33). These effects were accompanied by an increased steepness of and leftward shift in activation versus voltage relationship (33). In the present study, we found that the effect of PLM on the activation-voltage relationship is dependent on voltage-step duration. Our previous work utilized 25-ms voltage steps to show the significant 10-mV left shift and steepening of the activation-voltage relationship. The present study uses 300-ms steps, which are sufficiently long to allow us to simultaneously evaluate the effect of PLM on the rates of activation and inactivation. We were surprised to find that we could no longer observe a PLM-induced shift in the activation-voltage relationship (Table 1). This is because PLM-enhanced VDI becomes more pronounced with longer voltage steps, and the amplitudes of tail currents measured preceded by longer voltage steps are smaller than if they were preceded by short 25-ms voltage steps. These findings indicate that longer voltage steps cannot be used to reliably assess the effects PLM has on the voltage dependence of activation. However, single Boltzmann fits through activation-voltage data indicate that amino acid substitutions in the PFXYD motif do not produce gross changes in the voltage dependence of activation of CaV1.2 channels compared with wild-type PLM (Table 1). The remainder of this paper focuses on PLM-induced slowed activation, slowed deactivation, and accelerated inactivation, which are robust and easily measured effects.

View this table:
Table 1.

CaV1.2 gating parameters in the absence and presence of wild-type and mutant PLM

To test the potential impact of the PFTYD motif on the PLM-induced gating changes, we individually exchanged each of the five amino acids with alanine to generate the PLM mutants P8A, F9A, T10A, Y11A, and D12A. Western blot analysis performed on lysates from cells expressing CaV1.2 in the absence and presence of wild-type and mutant PLM indicate that wild-type and mutant PLM do not alter CaV1.2 expression levels (Fig. 1). Likewise, expression levels of all mutant PLM constructs are comparable to that of wild-type PLM. These results demonstrate that changes in channel gating associated with amino acid substitutions in the PFTYD segment of PLM (described below) do not result from changes in expression levels of CaV1.2 or PLM mutants. Therefore, observed changes in CaV1.2 gating are the direct result of changes in the interactions between PLM and CaV1.2.

Fig. 1.

Amino acid substitutions within the PFTYD segment of phospholemman (PLM) do not alter CaV1.2 or PLM expression levels. Lysates from cells expressing CaV1.2 and wild-type and mutant PLM were subjected to polyacrylamide gel electrophoresis. Separated proteins were transferred to PVDF membranes that were probed with anti-CaV1.2 and anti-PLM antibodies. Films were scanned, and the intensity of each band was quantified by densitometry. Data were normalized to conditions depicted in lane 4. Top blot and bar graph: expression of CaV1.2 was not altered by the coexpression of wild-type or mutant PLM. Bottom blot and bar graph: expression levels of mutant PLM constructs are similar to that of wild-type PLM.

Importantly, each PLM mutant altered CaV1.2 channel gating, which indicates that each was functionally expressed (Table 1). However, the gating changes induced by the majority of mutant PLMs were not statistically different from wild-type PLM, suggesting that the amino acids substituted in these mutants do not mediate the effects of PLM on CaV1.2 channel gating. One mutation that did alter PLM-dependent changes in CaV1.2 gating was T10A, which abrogated the PLM-induced acceleration of VDI and greatly enhanced the PLM-induced slowing of activation. In addition, D12A had an intermediate effect on inactivation and abrogated PLM-induced slowed deactivation, which suggested that the loss of the negative charge could impact gating. We hypothesized that the replacement of the negative charge with a positively charged amino acid (lysine, K) would further impair PLM-dependent slowed activation and deactivation. In addition, we made a second neutral amino acid mutant PLM, D12C, to ensure that any changes induced by D12K were related to charge. As can be seen in Fig. 2, these two additional mutations had dramatic effects on CaV1.2 channel gating, but the above hypothesis was not supported.

Fig. 2.

Single amino acid substitutions in the PFTYD segment of PLM alter multiple gating parameters of CaV1.2 currents. A–D: whole cell CaV1.2 currents were recorded with (−)PLM (black line), (+)PLM (light gray line), PLM mutants (T10A, D12A, D12K, and D12C, dark gray line) were evoked by 300-ms depolarizing steps from a holding potential at −90 mV to the indicated voltages indicated to the right. Currents from 7 to 10 cells were normalized and averaged.

Substitution of alanine for Thr-10 enhances PLM-induced slowing of CaV1.2 activation.

Of the five amino acid positions tested, only substitutions at Thr-10 and Asp-12 were found to impact PLM-induced changes in CaV1.2 gating (Table 1). As previously demonstrated (33), activation of CaV1.2 currents in the presence of wild-type PLM (Fig. 2A, light gray lines) is much slower than in the absence of PLM (Fig. 2A, dark lines) at −20 mV, but this difference disappears with stronger depolarizing steps (Fig. 2A). The currents depicted in Fig. 2A also illustrate the increase in the speed and magnitude of VDI measured at depolarized voltages in cells expressing wild-type PLM. The CaV1.2 channels coexpressed with T10A (Fig 2A, dark gray lines) activate much more slowly than channels expressed with wild-type PLM.

These effects were quantified by measuring the time for currents to activate from 10% to 90% of the peak current (T10–90) (33). The T10–90 was plotted against step voltages for comparison among (−)PLM, wild-type PLM, T10A, D12A, D12K, and D12C (Fig. 3). This analysis demonstrates that T10A enhanced PLM-induced slowed activation. T10–90 values for T10A are increased as much as sevenfold over those determined for (−)PLM at voltages from −20 to 0 mV. In addition, T10–90 values determined for T10A were significantly different from those determined for wild-type PLM at all potentials tested (Fig. 3A). Thus T10 appears to be a critical amino acid involved in controlling the effect of PLM on CaV1.2 channel activation.

Fig. 3.

Substitution of alanine for Thr-10 enhances PLM-induced slowing of CaV1.2 activation. T10–90, the time required to activate from 10% to 90% of the peak currents, was measured from currents elicited during 300-ms steps ranging from −20 to +20 mV (10-mV increments). T10–90 for (+)PLM (black) is larger than (−)PLM at −20 and −10 mV, P < 0.05. For clarity, the significant differences between (−)PLM and (+)PLM are not indicated in the panels. A: T10–90 for T10A is larger than (−)PLM at −20, −10, 0 mV, P < 0.05 and also larger than (+)PLM at all indicated voltages, P < 0.05. B: T10–90 for D12A (gray) is larger than (−)PLM at −20 mV, P < 0.05. C: T10–90 for D12K (gray) is smaller than (−)PLM at 0, +10, +20 mV, P < 0.05, and smaller than (+)PLM at all indicated voltages, P < 0.05. D: T10–90 for D12C (gray) is larger than (−)PLM at −20, −10 mV, P < 0.05. *P < 0.05 vs. (−)PLM; #P < 0.05 vs. (+)PLM.

Substitutions at the D12 position had differential effects on activation dependent on the physical and/or chemical properties of the newly inserted amino acid. As with P8A, F9A, and Y11A (not shown), D12A and D12C had little or no effect on PLM-induced slowed activation (Fig. 2, B and D; Fig. 3, B and D). However, D12K significantly suppressed slowed CaV1.2 channel activation (Fig. 2C). T10–90 for D12K was significantly smaller than that for wild-type PLM at all test potentials and was even significantly smaller than that of (−)PLM measured at 0, +10, and +20 mV (Fig. 3C). D12 appears to be another amino acid that controls PLM-induced slowing of CaV1.2 channel activation, but this effect appears to require a positively charged amino acid at the 12 position. However, activation is difficult to measure independently of inactivation when both gating events have a similar speed (11). Thus the apparent faster activation induced by coexpression of the D12K mutant could result from the effect of the mutation on inactivation. However, this concern is mitigated by the fact that D12C has a similar effect on inactivation as D12K but does not appear to alter the PLM effect on activation. Thus we conclude that coexpression of the D12K PLM mutant significantly speeds CaV1.2 channel activation.

Neutralization of Asp-12 attenuates PLM-induced slowed deactivation.

Activation and deactivation are generally thought to be reversible gating processes so that effects on one process can affect the other. Coexpression of PLM slows CaV1.2 channel activation and also slows deactivation (33), which suggests that mutations that affect activation (e.g., T10A) could have a similar effect on deactivation. Therefore, we determined the effect of our PLM mutations on deactivation. Tail currents were measured at −50 mV following 100-ms steps to +80 mV to maximally activate CaV1.2 currents. Tail currents were normalized and superimposed to highlight the effect of PLM mutants on the time course of deactivation relative to wild-type and (−)PLM controls (Fig. 4). Unexpectedly, the two mutants that affected activation had no effect on deactivation. Tail current deactivation with T10A and D12K was similar to that of wild-type PLM (Fig. 4, A and C), suggesting that these mutations (like P8A, F9A, and Y11A) do not affect PLM-induced slowing of channel closing. However, deactivation of channels coexpressed with either D12A or D12C are similar to that for (−)PLM (Fig. 4, B and D), which shows that changes at the D12 position can abrogate PLM-induced slowed deactivation.

Fig. 4.

Neutralization of Asp-12 attenuates PLM-induced slowed deactivation. Sample traces for tail currents with (−)PLM (black), (+)PLM (light gray lines), and PLM mutants (dark gray lines). A–D: currents were evoked by repolarizing steps to −50 mV following 100-ms voltage steps to +80 mV. Tail currents were normalized to facilitate comparison of deactivation time courses.

These effects were quantified by measuring the relative tail current amplitude 1 ms following peak tail current (R1.0). These isochronic measurements of deactivation were determined for tail currents evoked from repolarizing steps to −50 mV following 100-ms depolarizing steps ranging from −20 to +80 mV and are illustrated by plots of R1.0 vs. step voltage (Fig. 5). As expected from the qualitative assessment of CaV1.2 tail currents (Fig. 4), the R1.0-voltage relationships for T10A and D12K overlap that determined for wild-type PLM (Fig. 5, A and C). In contrast, R1.0 values for the D12A are similar to those determined for (−)PLM and are significantly smaller than R1.0 values for (+)PLM at voltages ranging from +40 to +80 mV (Fig. 5B). D12C mutants also reduced the PLM effect on CaV1.2 channel deactivation so that there is no significant difference from (−)PLM, but the effect is incomplete so that the data are also not significantly different from wild-type PLM (Fig. 5D). It is interesting that changing the charge on the amino acid at position 12 from negative to positive had no effect on PLM-induced slowed deactivation, whereas removal of the charge abrogated that effect.

Fig. 5.

CaV1.2 deactivation is differentially altered by wild-type PLM and PLM mutants. R1.0 values, the fraction of current remaining 1 ms after the peak tail current, for (+)PLM, (−)PLM, and mutant PLM are plotted against step voltages. A: R1.0 verses voltage plot for T10A overlaps with that for (+)PLM and is larger than (−)PLM at all voltages measured, P < 0.05. B: R1.0 values for D12A are similar to that of (−)PLM and are significanly smaller than (+)PLM at voltages from +40 to +80 mV, P < 0.05. C: R1.0 curve for D12K is similar to that for (+)PLM and is larger than that of (−)PLM at all indicated voltages, P < 0.05 (C). The R1.0 values for D12C (gray diamonds) lie between those for (−)PLM and (+)PLM and are not significantly different from either (−)PLM or (+)PLM, P > 0.05 (D). *P < 0.05 vs. (−)PLM, #P < 0.05 vs. (+)PLM. Note that R1.0 for (+)PLM is significantly larger than (−)PLM at all indicated voltages.

Amino acid substitutions at Asp-12 enhance VDI.

PLM increases VDI of CaV1.2 currents (33). To assess the role the PLM PFTYD motif plays in determining inactivation gating, we measured the fraction of current remaining at the end of 300-ms steps (R300) ranging from −20 to +20 mV. The plot of R300 versus voltage from CaV1.2 currents expressed with T10A shows that this mutation abrogated the PLM-induced increase in VDI, so R300 values for T10A are significantly different from those determined for wild-type PLM at all voltages examined (Fig. 6A). As with the other measured PLM-dependent changes in gating, the R300 values measured with coexpression of P8A, F9A, and Y11A mutants were not significantly different from wild-type PLM (Table 1). D12A-induced enhancement of VDI was not significantly different from wild-type PLM (Fig. 6B). In contrast, R300 values for D12K and D12C were substantially smaller than those of either (+)PLM or (−)PLM at every voltage tested (Fig. 6, C and D), indicating that these amino acid changes greatly enhance the ability of PLM to increase VDI.

Fig. 6.

Amino acid substitutions at Asp-12 enhance voltage-dependent inactivation (VDI). R300 values, the fraction of peak current measured at the end of 300-ms voltage steps to the indicated voltages, are plotted verses voltage (representative currents are depicted in Fig. 2, A–D). The R300 plot for (+)PLM exhibits a steeper voltage dependence compared with (−)PLM, and R300 values are smaller than (−)PLM at +10, +20 mV, indicating that a larger fraction of the cannels inactivate during the 300-ms voltage steps, P < 0.05. For clarity, significant differences between (−)PLM and (+)PLM are not indicated. A: The R300 plot for T10A lies slightly above of that for (−)PLM, and R300 values for T10A are significantly larger than (+)PLM at all indicated potentials, P < 0.05. B: The R300 values for D12A are not significantly different from either (−)PLM or (+)PLM, P > 0.05. C and D: both of the R300 plots for D12K and D12C (gray diamonds) exhibit a much stronger voltage dependence than that for (−)PLM and R300 values for both mutants are substantially smaller than both (−)PLM and (+)PLM at all indicated potentials, P < 0.05. *P < 0.05 vs. (−)PLM; #P < 0.05 vs. (+)PLM.


All five PLM mutants described in these studies retained their ability to modulate CaV1.2 gating, indicating that the PFTYD segment is not essential for the association of PLM with CaV1.2. Instead, the PFTYD segment appears to play a role in coupling PLM binding to PLM-induced changes in CaV1.2 gating. We found that T10A greatly enhances PLM-induced slowing of activation, has no effect on PLM-induced slowed deactivation, and abrogates the PLM effect on VDI. D12A and D12C abrogates the PLM-induced slowed deactivation without altering the PLM-induced changes in activation, whereas the D12A mutant fails to affect VDI. Both D12K and D12C dramatically increase PLM-induced enhancement of VDI. This is in contrast to structural studies suggesting that F9 and Y11 (and Y13; our numbering) stabilize the association between NKA α- and β-subunits (26). Thus it appears that the interactions between the PFXYD motif of FXY10 and NKA differ from those of PLM and CaV1.2. The discovery that substitutions at Thr-10 and Asp-12 enhance PLM-induced changes in CaV1.2 gating far beyond that of wild-type PLM indicate that the PFTYD segment functions as a regulatory site on PLM and that changes in the chemical and/or physical properties of amino acids at this location can change the gating behavior of CaV1.2 channels in important and interesting ways. We conclude that the PFTYD motif in PLM is important for fine tuning the gating kinetics of cardiac calcium channels and likely plays an important role in regulating Ca2+ homeostasis in the heart.

PLM-induced enhancement of activation and VDI.

Only position “X” in the PFXYD segments from all the FXYD family of proteins is variable (29), suggesting that this position can tolerate a wider range of physical and/or chemical properties compared with other positions in this highly conserved segment. Replacing Thr-10 with alanine at position “X” (i.e., Thr-10) has both positive (enhancement) and negative (abrogation) effects on the degree to which PLM regulates CaV1.2 gating. This substitution enhances PLM-induced slowing of CaV1.2 activation, suggesting that the identity of the amino acid at position X in FXYD proteins is important for fine tuning FXYD protein actions on their target proteins. In support of this notion, this same mutation eliminates PLM-dependent enhanced VDI, but PLM-induced slowing of deactivation is insensitive to this mutation.

It is possible that enhanced slowing of CaV1.2 channel activation by T10A is an indirect consequence of the loss of PLM-induced VDI (2, 11). That is, T10A may not slow activation more than wild-type PLM but only appears to slow activation because VDI is attenuated. If true, one would expect PLM mutants that promote enhanced VDI (i.e., D12C and D12K) to abrogate PLM-induced slowed activation. This hypothesis is supported by the associated enhancement of VDI and speeding of activation by the D12K mutant. However, the impact on VDI by the D12C mutation is similar to D12K, yet the activation kinetics of CaV1.2 current with D12C is similar to that of wild-type PLM. These findings support the conclusion that T10A slows activation to a much greater extent than wild-type PLM, despite its reduction in VDI.

PLM-induced slowed deactivation.

Wang et al. (33) demonstrated that PLM-induced slowed deactivation is both time and voltage dependent and postulated that this reflects a PLM-induced shift from a low open probability (Po) gating mode (mode 1) into a high Po gating mode (mode 2) as described by Pietrobon and Hess (23). This idea was supported by the observation that the development of slowed deactivation was prominently increased by PLM at depolarized potentials (e.g., +20 mV) and longer voltage steps (e.g., 250 ms) (33), suggesting that effects would be most pronounced during the plateau phase of the human cardiac action potential (16).

Of the five alanine substitutions investigated in these studies, both D12A and D12C prevented PLM-induced slowed deactivation of CaV1.2 channels. This result led us to hypothesize that the negatively charged aspartate residue is crucial for this effect. Since the PFTYD segment is extracellularly located, we reasoned that the negative charge may interact with one of the S4 segments (the voltage sensors) (3, 4) to facilitate a transition from mode 1 to mode 2 gating (23). However, replacing the aspartic acid with a positively charged lysine had no impact on PLM-induced slowed deactivation, suggesting that the negative charge at position 12 is not important to PLM-induced slowed deactivation. In addition, there is little correlation between the size of the side chain at position 12 and its effects on CaV1.2 deactivation (e.g., the Van der Waals volumes of aspartate and cysteine are 91 and 86 Å3, respectively). It is noteworthy that alanine and cysteine residues are very hydrophobic and tend to have a relatively small portion of their surface areas exposed to the solvent compared with aspartate and lysine. It is possible that any charged residue (either negative or positive), through its extensive interactions with the solvent, stabilizes a functional conformation of PLM, such as the stabilization of helix-1 (9, 26, 30), and thereby promotes slowed deactivation.

Physiological relevance of the PFTYD amino acids in PLM.

We have recently reported that slowing of CaV1.2 channel deactivation plays a major role in increasing the duration of the cardiac action potential (35). The combination of increased Ca2+ entry during repolarization combined with increased action potential duration generates excessive Ca2+ influx and is predicted to drive arrhythmogenic processes such as early afterdepolarizations and Long-QT Syndrome (13, 17). We have found that PLM also slows CaV1.2 closing and enhances VDI, suggesting that PLM induces a balanced modulation of the channels such that Ca2+ influx increases via slowed channel closing to drive stronger contractions but also limits arrhythmogenesis by enhancing VDI. Our findings suggest that it may be possible to fine tune this delicate balance by altering the functional behavior of PLM. Here, we have demonstrated that changes in the chemical and/or physical properties of amino acids within the PFTYD segment of PLM can offset the balance of opposing actions of PLM-induced regulation of CaV1.2 gating.


This work was supported by grants from National Heart, Lung, and Blood Institute R01 HL-074143 (to B. Z. Peterson), China Scholarship Council, Chinese Scholarship Fund (to K. Guo), and the Pennsylvania Department of Health using Tobacco Settlement Funds (to B. Z. Peterson and K. S. Elmslie).


No conflicts of interest, financial or otherwise, are declared by the author(s). The Pennsylvania Department of Health specifically disclaims responsibility for analyses, interpretations and conclusions presented here.


We thank Yunhua Wang for great technical assistance.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
View Abstract