Effect of vHDNP on paw withdrawal latencies (PWL). PWL is presented as ratios of the average time taken for withdrawal of the right vs. left foot for animals in each group. (A) The PWL ratios in uninfected (PBS injected) control rats and vHG-infected rats were approximately 1, while vHDNP-infected rats displayed a small but significant increase in PWL ratio to 1.25 (*P < 0.05, n = 8). (B) Effect of vHDNP on CAPS-induced heat hyperalgesia. vHDNP-inoculated rats had a 2.5-fold increase in PWL as compared with PBS- or vHG-infected rats (*P < 0.001).

Cobalt uptake in response to CAPS was used to determine the number of CAPS-responsive neurons in DRG cultures infected with vHDNP construct, with vHG construct and in noninfected neurons. (A) The infection parameters were set to give nearly 100% efficiency. (B) Representative micrographs of a TRPV1-negative neuron with no cobalt uptake and of noninfected, vHG-infected and vHDNP-infected CAPS-responsive neurons that turn various shades of gray to black after cobalt uptake. (C) The percentage of CAPS-sensitive neurons revealed by cobalt uptake is shown for uninfected, vHG- and vHDNP-infected neurons. The numbers of neurons (N) are shown in parentheses. Approximately 44% of noninfected neurons were CAPS-responsive, while 33% of vHG-infected neurons demonstrated CAPS sensitivity. By contrast, only 12% of the vHDNP-infected neurons were CAPS-sensitive. This was a significant reduction when compared with the number of CAPS-sensitive vHG-infected neurons (P < 0.001). Scale bars, 20 µm in all panels on A and B.

DNPKCε inhibits endogenous PKCε translocation in rat DRG neurons. GFP was used to identify neurons infected with either a vHG construct (A–F) or a vHDNP construct (G–L). A red fluorescent Cy3 secondary antibody was used to visualize endogenous PKCε. vHG-transduced neurons (A–F) displayed a diffuse pattern of GFP localization which was unaffected by stimulation with PDBu (D–F). Endogenous PKCε translocated to the plasma membrane when stimulated with PDBu (B and E). The merged images do not display co-localization of GFP and endogenous PKCε (C and F). (G–L) After PDBu treatment in vHDNP-infected neurons, DNPKCε translocated to the plasma membrane to a greater extent than endogenous PKCε (H and K). (L) The merged images show that after PDBu, the DNPKCε and PKCε products were also co-localized at plasma membrane to a lesser extent than the co-localization of the two proteins in the cytoplasm in the absence of PDBu, when they co-localized extensively (I). Scale bars as shown in C, F, I and L.

The responsiveness of CAPS-activated currents to the facilitatory effect of PDBu decays during the desensitization of CAPS-evoked currents. CAPS currents were normalized to the peak and averaged in nine vHG (A) cells and eight vHDNP (B) cells (cross hairs, x). Dotted lines were fitted to the average normalized currents by a sum of one rising and two decaying exponentials as shown above traces. (C) Graphs showing the decline in the responsiveness to PDBu measured as a decrease in the peak PDBu-evoked current (normalized to peak CAPS response) when PDBu was applied at various times after the start of CAPS application and during the desensitization of CAPS currents in the continuous presence of CAPS, as illustrated in Fig. 5. Data are fitted by a single exponential curve as shown above graphs using nonlinear regression analysis available in Sigma Plot. The time constants for decay were 4.5 ± 1.5 min for vHG cells and 0.85 ± 0.34 min for vHDNP neurons. (D) Percentage increase in CAPS current by PDBu relative to the amplitude of CAPS-evoked current at the time of PDBu application plotted against peak time of response for the initial three blocks of pooled data shown as symbols in A and B. Drug concentrations (in µm): CAPS, 0.5; PDBu, 0.5.

Construction and characterization of dominant-negative PKCε-expressing vector. (A) Full-length PKCε structure and functional domains. (B) DNPKCε retains the pseudo-substrate region (PS) and the C1a domain that contains the actin binding site and the variable region (V3); to this GFP was fused to allow identification of DNPKCε-expressing cells. (C) The DNPKCε gene was inserted into the UL41 locus of the replication defective HSV vector to produce the recombinant vHDNP and a reporter gene vector vHG expressing GFP from the same promoter that was used as a control for all experiments. (D)Western blot using a GFP-specific antibody that detected a 47-kDa protein in vHDNP-infected U2OS cells (multiplicity of infection of 3 at 16 hpi) and a 27-kDa protein in vHG-infected U2OS cells. (E) In vHDNP-infected cells, the DNPKCε product rapidly translocated to the plasma membrane from the perinuclear region in response to phorbolmyristate acetate (PMA). Control vector (vHG)-infected cells displayed a uniform expression of GFP in the cytoplasm and its localization was unaffected by PMA treatment. Scale bars as shown in E.

Comparison of CAPS-evoked currents and enhancement of CAPS currents by PDBu in vHG- and vHDNP-infected DRG neurons. (A) Infection of DRG neurons with vHDNP-construct led to a CAPS response that reached maximum faster, was considerably smaller in amplitude and desensitized faster than response to CAPS in vHG cells. PDBu enhanced the partially desensitized CAPS currents in both types of cells but in DNPKCε-expressing cells, the lag of the response to PDBu (inset) was larger and the time to peak considerably slower than in vHG-infected cells. TRPV1 antagonist (5 µm) blocked the PDBu-enhanced CAPS currents in both cell types. Solid line above recordings indicates that CAPS was present throughout the recordings. (B) BIM, a broad-range PKC inhibitor, suppressed the PDBu effect in both vHG and vHDNP cells. However, Gö6976, a PKC inhibitor selective for the conventional PKC subtypes but with little effect on PKCε, did not alter the amplitude or the rate of desensitization of the PDBu-enhanced CAPS currents in all the cell types tested. Current trace in vHDNP cells shows lack of Gö6976 effect. Solid line above recordings indicates that CAPS was present throughout the records. Drug concentrations (in µm): CAPS, 0.5, PDBu, 0.5; BIM, 0.5; Gö6976, 0.5; TRPV1Ant, 5. (C) PDBu administered in the presence of CAPS, 1–16 min after CAPS, markedly enhanced the currents in uninfected (168 ± 84% increase, n = 11) and vHG-infected (144 ± 84% increase, n = 12) neurons. However, vHDNP infection significantly diminished the effect of PDBu on CAPS-evoked currents (97 ± 79% increase, n = 23). (D) Infection with vHG or vHDNP had no measurable effect on the time to peak of CAPS-evoked currents. However, vHDNP slowed the time to peak of PDBu-induced currents in the presence of CAPS. (E) Compared with uninfected (con) cells, infection with vHG construct slowed the rapid desensitization (τ_fast) and accelerated the slow desensitization (τ_slow) of CAPS currents. Infection with vHDNP construct accelerated both the rapid and slow desensitization of CAPS-evoked currents. Infection with vHG had no effect on the desensitization of PDBu-induced currents in the presence of CAPS; however, vHDNP significantly increased the time constant of desensitization of PDBu-induced currents in the presence of CAPS. Data in D and E are averages ± SEM, for the number of cells shown in parentheses. Comparison was considered statistically significant if P ≤ 0.05 (*); a one-way analysis of variance was first carried out followed by post-hoc comparisons between the different groups using the Holm-Sidak test as described in the Methods (for P < 0.05, a two-tailed t-test, unequal variance was used, **P < 0.001).

Infection of DRG neurons with a virus containing the DNPKCε construct (vHDNP) reduces the amplitude and increases the rate of desensitization of CAPS-induced currents but does not alter depolarization-evoked firing. (A) Firing in response to short (5 ms, first action potential in the sequence) or long (600 ms, second action potential in the sequence) depolarizing pulses is not altered by infection with either vHG or vHDNP. (B) Time course of average CAPS (0.5 µm) currents expressed as current densities (pA/pF) in 53 uninfected (control), 23 vHG-infected and 30 vHDNP-infected neurons. CAPS application indicated by the horizontal line above the records was present throughout the duration of the experiment. Dotted lines are best fits of average current densities with a sum of one rising and two decaying exponentials as shown above control curve (see Methods). Note that in vHG cells currents desensitize more slowly than in control (con) cells while in vHDNP neurons the currents are significantly reduced in amplitude and desensitize more rapidly. Number of cells in each group shown in parentheses. (C) Area of current densities during the first 10 min of records in B in control, VHG and vHDNP cells. Note that the area under the vHG curve increases despite the apparent reduction of peak amplitudes, whereas the area under the vHDNP curve is significantly decreased (threefold reduction). (D) Summary of fitted time course of CAPS currents shown in B, on an expanded time scale and scaled to nearly match the amplitude of control current densities for ease of comparison. Solid line above recordings indicates that CAPS was present throughout the records. Note the slow desensitization in vHG cells and more rapid desensitization in vHDNP neurons. (E) Average peak current density in uninfected cells (control) and vHG-infected cells was reduced in vHDNP-infected cells. (F) The percentage of CAPS-responsive cells was not altered by infection with either vHG or vHDNP. Data in E and F are averages ± SEM, for the number of cells shown in parentheses. Comparison was considered statistically significant if P ≤ 0.05 (*); a one-way analysis of variance was first carried out followed by a post-hoc comparisons between the different groups using the Holm-Sidak test as described in the Methods (for P < 0.05, a two-tailed t-test, unequal variance was used, P < 0.001).