Multitarget nociceptor sensitization by a promiscuous peptide from the venom of the King Baboon spider.

The King Baboon spider, Pelinobius muticus, is a burrowing African tarantula. Its impressive size and appealing coloration are tempered by reports describing severe localized pain, swelling, itchiness, and muscle cramping after accidental envenomation. Hyperalgesia is the most prominent symptom after bites from P. muticus, but the molecular basis by which the venom induces pain is unknown. Proteotranscriptomic analysis of P. muticus venom uncovered a cysteine-rich peptide, δ/κ-theraphotoxin-Pm1a (δ/κ-TRTX-Pm1a), that elicited nocifensive behavior when injected into mice. In small dorsal root ganglion neurons, synthetic δ/κ-TRTX-Pm1a (sPm1a) induced hyperexcitability by enhancing tetrodotoxin-resistant sodium currents, impairing repolarization and lowering the threshold of action potential firing, consistent with the severe pain associated with envenomation. The molecular mechanism of nociceptor sensitization by sPm1a involves multimodal actions over several ion channel targets, including NaV1.8, KV2.1, and tetrodotoxin-sensitive NaV channels. The promiscuous targeting of peptides like δ/κ-TRTX-Pm1a may be an evolutionary adaptation in pain-inducing defensive venoms.

The King Baboon spider, Pelinobius muticus (formerly Citharischius crawshayi) (1, 2), is a large, fawn-colored tarantula found in Kenya and Tanzania that feeds on beetles, cockroaches, and other spiders, but will occasionally hunt vertebrates, such as mice, lizards, snakes, and birds. Larger species of Old World tarantulas like P. muticus spend most of their time in burrows and, unless provoked, pose no immediate risk to humans. Little is known about P. muticus envenomations, but like other tarantulas, they have evolved formidable chelicerae along with a venom that delivers mechanically and chemically painful bites to defend against perceived threats (3).

The triggering of unpleasant sensory experiences using pain-producing toxins would likely discourage most potential aggressors (4). Pain development and discomfort are universal features of spider envenomation (5). Spider venoms contain neurotoxins that produce paralysis and toxicity to humans through their ability to modulate ion channels and receptors. For example, mature male Australian funnel-web spiders use alogogenic δ-hexatoxin peptides in their venom to deter vertebrate predators after they leave their burrows in search of females (6).

The study of spider-venom toxin pharmacology has uncovered a vast collection of molecular tools for exploring pain physiology. Examples of such venom-derived peptides include Hm1a from the Togo starburst tarantula Heteroscodra maculata, which was used to show for the first time that the voltage-gated sodium (NaV) channel, NaV1.1, is involved in the transduction of mechanical pain (7), whereas Tsp1a from a Peruvian Thrixopelma tarantula was used to identify a key role for NaV1.7 in the chronic visceral pain associated with irritable bowel syndrome (8).

Despite suggestive evidence that African tarantulas produce potent neurotoxic venoms, their effects and molecular composition have not been thoroughly explored (9). Understanding the physiological process of pain generation by venom components is of considerable importance and could underpin effective treatments (9).

In the present study, we analyzed the venom proteome of P. muticus and identified an algogenic inhibitory cystine knot (ICK) peptide, δ/κ-theraphotoxin-Pm1a (Pm1a). We synthesized sPm1a, determined its three-dimensional (3D) structure, and evaluated its effects in vivo, ex vivo, and on heterologously expressed ion channel targets. We show that the spontaneous and ongoing pain from P. muticus envenomation originates from ectopic firing of peripheral nociceptors as a result of Pm1a’s synergistic modulation of multiple voltage-gated ion channels.

Results

Identification and Synthesis of Pm1a.

P. muticus venom and constituent peptide toxins have previously been shown to modulate the activity of voltage-gated ion channels (1, 2). A subsequent proteotranscriptomic analysis of P. muticus venom identified a peptide with a monoisotopic mass of 4524.5 Da, named δ/κ-theraphotoxin-Pm1a (δ/κ-TRTX-Pm1a). The transcript encoding δ/κ-TRTX-Pm1a (Fig. 1) was the sixth most abundant contig in the venom-gland transcriptome, with 38,213 transcripts per million (Fig. 1), consistent with its identification as a major venom component during venom fractionation (Fig. 1). δ/κ-TRTX-Pm1a is encoded as an 85-residue prepropeptide comprised of a 19-residue signal peptide, followed by a 24-residue propeptide and the 42-residue mature δ/κ-TRTX-Pm1a toxin (Fig. 1). The mature toxin contains six cysteine residues that form three disulfide bridges, with a theoretical molecular mass of 4524.89 Da (M+H+ = 4525.89) (Fig. 1). δ/κ-TRTX-Pm1a is homologous to a family of tarantula-venom peptides bearing the highly conserved N-terminal “GVDK motif” (1).

Fig. 1.

Discovery and structural characterization of δ/κ-theraphotoxin-Pm1a. (A) P. muticus venom-gland transcript encoding δ/κ-TRTX-Pm1a with signal peptide in lowercase, propeptide in uppercase, and the mature peptide in bolded capitals. (B) Read coverage of the contig encoding δ/κ-TRTX-Pm1a, with an Inset showing King Baboon spider. Image credit: Shutterstock/Audrey Snider-Bell. (C) C18 RP-HPLC chromatogram showing fractionation of 0.5 mg crude P. muticus venom (Left) and purified δ/κ-TRTX-Pm1a (Right). (D) Chromatogram showing coelution of native δ/κ-TRTX-Pm1a and synthetic sPm1a. Absorbance was monitored at 214 nm. (Right) MALDI-TOF mass spectra, with the monoisotopic M+H+ shown for each peptide. (E) Mature peptide sequence highlighting disulfide connectivity and β-sheet regions. (F) Ensemble of sPm1a structures determined using NMR spectroscopy. Disulfide bonds are labeled and shown as red tubes while β-strands are shown in blue. (G) Structure of sPm1a showing hydrophilic and hydrophobic faces. (H) Surface representation showing distribution of sPm1a’s hydrophilic residues (teal and cyan: Upper) and hydrophobic residues (purple: Lower).

To enable full structural and functional characterization, δ/κ-TRTX-Pm1a was synthesized in two fragments using Fmoc solid-phase peptide synthesis. Reduced, full-length toxin was generated by native chemical ligation (details in ). After folding and purification, the obtained synthetic product (hereafter sPm1a) was indistinguishable by HPLC and mass spectrometry from δ/κ-TRTX-Pm1a isolated from venom (Fig. 1).

Three-Dimensional Structure of sPm1a.

The 3D structure of sPm1a was determined using NMR spectroscopy. The final ensemble of 20 structures (Fig. 1) revealed a highly flexible C-terminal region and a moderately dynamic β-hairpin loop, which together lead to a high overall backbone atom root-mean-square deviation (RMSD) of 0.78 ± 0.20 Å. The backbone RMSD is much lower (0.23 ± 0.03 Å) for the relatively rigid regions (residues 3 to 27 and 31 to 35), suggesting that the structure is well defined when the flexible regions are excluded. The ensemble of sPm1a structures is of high stereochemical quality as judged by a MolProbity score of 1.7 ± 0.1 ().

The disulfide-bond connectivity of sPm1a was unequivocally determined from preliminary structure calculations performed without disulfide-bond restraints (10). The dipolar connectivities in the NOESY spectrum are consistent with the pattern Cys7–Cys21, Cys14–Cys26, and Cys20–Cys34 (red tubes in Fig. 1). The three disulfide bonds form a cystine knot in which the Cys7–Cys21 and Cys14–Cys26 disulfides and the intervening section of the polypeptide chain form a closed loop that is pierced by the Cys20–Cys34 disulfide bond. Together with a canonical β-hairpin that is supported by two of the disulfide bonds, the overall structure of sPm1a conforms to the ICK motif commonly adopted by spider toxins known to modulate voltage-gated ion channels (11).

In contrast to the highly disordered C terminus (residues 36 to 42), the N terminus of sPm1a is relatively inflexible. A hydrogen bond is formed between the backbone amide of Val2 and the carbonyl oxygen of Cys20 (Fig. 1), stabilizing the N-terminal tail in a position that is in close proximity to the core of the peptide. Hydrophobic patches are commonly found on the surface of gating-modifier peptides (GMPs) from tarantulas, such as SGTx1 and GxTx-1E, and they are proposed to interact with both the hydrophobic membrane lipids and the voltage-sensor domain on target ion channels (12, 13). Accordingly, sPm1a has most of its hydrophobic residues (Tyr9, Leu-10, Phe11, Leu24, Tyr33, and Trp36) on one face of the peptide, and they form a distinctive hydrophobic patch on the opposite side to the N terminus (Fig. 1 ).

sPm1a Elicits Nocifensive Behaviors in Mice.

To determine whether sPm1a can recapitulate the pain experienced after a P. muticus bite, we delivered the synthetic peptide directly to primary afferent terminals in the hindpaw of adult mice in vivo and evaluated the resulting behavioral responses. Intraplantar injection of sPm1a resulted in immediate and dose-dependent emergence of spontaneous nocifensive behaviors, including paw licking and flinching (flinches per 15 min: Control [Ctr]: 0.7 ± 0.2; 0.06 nmol sPm1a = 27.8 ± 3.5, P < 0.05; 0.6 nmol sPm1a = 187.0 ± 16.5, P < 0.0001; n = 4 to 8 per condition) (Fig. 2) that persisted for ∼15 min (). These sPm1a-induced nocifensive behaviors included a considerable TTX-resistant component, with intraplantar administration of TTX partially reversing paw flinching (flinches per 15 min: 0.6 nmol sPm1a + 0.02 nmol TTX = 74.4 ± 5.4; P < 0.001) (Fig. 2 and ). Consistent with a major contribution of TTX-R channels to sPm1a-induced effects, coadministration of the selective NaV1.8 inhibitor A803467 (14) significantly reversed spontaneous pain behaviors (flinches per 15 min: 0.6 nmol sPm1a + 0.2 nmol A803467 = 93.8 ± 8.8; P < 0.001) (Fig. 2 and ). Substantial changes in mechanical paw withdrawal threshold (PWT; Ctr: 3.7 ± 0.1 g; 0.06 nmol sPm1a = 1.3 ± 0.2 g; n = 6; P < 0.001) (Fig. 2) and thermal thresholds (withdrawal temperature: Ctr: 49.8 ± 0.3 °C; 60 pmol sPm1a = 46.9 ± 0.4 °C; n = 6, P < 0.001) (Fig. 2) were also observed following sPm1a injection (Fig. 2), indicating mechanical and thermal hypersensitivity. Both NaV channel blockers, TTX and A803467, partially and significantly reduced mechanical allodynia (paw withdrawal force: 600 pmol sPm1a = 1.0 ± 0.1 g; 600 pmol sPm1a + 20 pmol TTX = 2.4 ± 0.3 g; 600 pmol sPm1a + 200 pmol A803467 = 2.5 ± 0.2 g; n = 5 to 8; P < 0.001) (Fig. 2), suggesting the TTX-R component is not limited to the spontaneous nocifensive behavior but also plays a role in mechanical allodynia induced by sPm1a.

Fig. 2.

sPm1a elicits nocifensive behavior in mice, induces hyperexcitability in small DRG neurons, and enhances the activity of TTX-R NaV channels. sPm1a injection induces (A) dose-dependent spontaneous pain behavior (paw flinches/time), (B) mechanical allodynia (decreased PWT in g), and (C) thermal hypersensitivity (decreased paw withdrawal temperature thresholds) (doses are expressed in nmol/paw) (see for details). TTX and A803467 reverse spontaneous pain (A) and mechanical allodynia induced by sPm1a (B). (B) Image credit: Shutterstock/Eric Isselee. One-way ANOVA with Tukey’s multiple comparison test: *P ≤ 0.05; ‡P ≤ 0.0001. (D) Spontaneous firing in a small DRG neuron (black, Upper) was enhanced by exposure to 1 μM sPm1a (red, Lower). The Inset shows the first derivative (dV/dt) of representative APs in the absence and presence of sPm1a (black and red, respectively). (E) sPm1a-evoked concentration-dependent increase in AP firing in a small DRG neuron. The ramp stimulation protocol (Inset, 1 s, 0 to 250 pA) applied in control conditions (black trace) did not elicit AP firing, whereas increasing concentrations of sPm1a caused a concentration-dependent increase in the number of APs fired upon the same stimulus (yellow: 0.1 μM; green: 0.3 μM; red: 1 μM sPm1a). The first derivative (dV/dt) of the first AP from the recordings on top is shown in the Center Inset. (F–H) Paired responses of mouse small DRG neurons before (Ctr) and after exposure to sPm1a (1 μM) and mean ± SEM (F) Individual and number of APs fired (n = 36), (G) first latency (n = 24; **P ≤ 0.01) and (H) membrane potential at the end of the ramp protocol (all neurons n = 36, P = 0.9563; responsive neurons n = 11, P < 0.0001). Solid symbols joined by solid lines and empty symbols joined by dotted lines correspond to sPm1a-responsive and nonresponsive DRG neurons, respectively. ns, not significant.

sPm1a Induces Hyperexcitability in Small Dorsal Root Ganglion Neurons.

Nociceptors are sensory neurons with crucial roles in the generation and conduction of pain signals and sensitization following injury or inflammation underlying hyperalgesia. The cell bodies of most sensory neurons reside in the dorsal root ganglion (DRG) in which nociceptors are typically recognized by their small cell size relative to other cell types (15). DRG neuron subtypes display different action potential (AP) waveforms and electrophysiological signatures that are common to major subclasses (16). For practical reasons, the overriding DRG neuron distinction used here was based on cell size estimated by membrane capacitance measurements, which is directly proportional to cell membrane area. Relatively broad APs with a “hump” in the falling phase made evident by two local minima in the AP first derivative (dV/dt) (see Insets in Fig. 2 ) are considered a distinctive feature of nociceptors. We performed whole-cell patch-clamp recordings from dissociated small DRG neurons (C ≤ 25 pF, equivalent to cell diameter ≤ 30 μm) from adult mice in the presence and absence of sPm1a. Sensitization, manifested by increased neuroexcitability, is characteristic of nociceptors (17), whereas sPm1a exposure enhances the excitability of small DRG neurons. The current-clamp recording shown in Fig. 2 was obtained from a DRG neuron that fired spontaneously in control conditions (Fig. 2 , Upper, black), while the addition of 1 μM sPm1a (Fig. 2 , Lower, red) resulted in a two- to fourfold increase in firing frequency, AP peak amplitude, and AP half-width (quantification for this cell is in included in ). Note that the nociceptor AP hump evidenced by the first derivative of representative voltage responses (Fig. 2 , Inset, dV/dt) in control conditions (Fig. 2, black) displays the stereotypical two local minima that spread further apart in the presence of 1 μM sPm1a (Fig. 2, red), suggestive of a repolarization delay in the presence of the venom peptide.

To compare excitability changes in small DRG neurons in the absence and presence of sPm1a, we used ramp current injections of 1 s duration (from 0 to 250 pA) (i.e., Fig. 2, Inset), which allows simultaneous assessment of the number of APs and firing latency. The recordings shown in Fig. 2 display the firing responses of a 15 pF neuron stimulated by a subthreshold somatic ramp current injection (0 to 250 pA) in which exposure to sPm1a caused concentration-dependent increases in excitability. Under control conditions (Fig. 2, black trace), no APs were observed, yet application of 0.1 μM sPm1a caused this cell to fire two APs (Fig. 2, yellow), with progressively higher sPm1a concentrations increasing the number of APs fired (0.3 μM, Fig. 2, green: 6 APs; 1 μM, Fig. 2, red: 9 APs). Accordingly, the first AP fired appeared earlier (first latency) during the stimulation as the sPm1a concentration bathing this neuron was increased. Superimposed first derivative traces for the first AP fired in each condition are displayed in Fig. 2, Center Inset, showing a concentration-dependent rightwards shift in the second local minima.

Sensitization of nociceptors develops as a consequence of tissue insult or inflammation and it is manifested as a reduction in the firing threshold or an increased response to noxious stimulation (17). Under control conditions, the firing characteristics of dissociated DRG neurons are notoriously heterogeneous. Paired responses in the absence (Ctr) and presence of 1 μM sPm1a from 36 small DRG neurons (obtained from seven independent isolations) are summarized in Fig. 2 . First, we observed that sPm1a caused a significant increase in the number of APs fired (Fig. 2) in 83% (30 of 36; number of APsCtr 1.3 ± 0.3 vs. number of APssPm1a 7.1 ± 1.2; n = 36, paired t test P < 0.0001). Second, sPm1a application significantly decreased the time to fire the first AP (Fig. 2) in 89% (24 of 27) of the neurons in which a first latency could be measured in control conditions (during 1 s ramps) (Ctr: 257.1 ± 53.1 ms vs. sPm1a: 195.5 ± 34.8 ms; n = 27, paired t test P = 0.0093). Third, sPm1a had variable effects on the resting membrane potential, with depolarization observed in some neurons as well as either hyperpolarization or no change in others (n = 36, paired t test P = 0.9563) (Fig. 2), likely reflecting the heterogeneity of the small DRG neurons tested.

From these observations, we infer that the function of δ/κ-TRTX-Pm1a within P. muticus venom involves sensitization of afferent neurons to maximize nociceptive signal generation upon the mechanical and chemical insults caused by bite puncture wounds and venom injection.

sPm1a-Enhanced Firing Involves Neuronal TTX-R NaV and Delayed Rectifier KV Conductances.

AP firing in sensory neurons is largely determined by the voltage-gated sodium conductance carried by TTX-S and TTX-R NaV channels. Small DRG neurons bathed in TTX at a concentration sufficient to completely block TTX-S NaV channels were exposed to sPm1a. The firing responses of a small DRG neuron to ramp current injection (0 to 250 pA) bathed in TTX (0.5 μM) (Fig. 3 , Left, dark gray), combined TTX and sPm1a (1 μM) (Fig. 3 , Center, red) and after removal of sPm1a (Fig. 3 , Right, light gray) are shown in Fig. 3. The additive exposure of sPm1a (Fig. 3, red) decreased latency to discharge, increased discharge activity, and impaired repolarization (right-shift in dV/dt local minima) (Fig. 3 , Inset TTX: gray, sPm1a/TTX: red). Removal of sPm1a from the bath (washout: wo) immediately reversed DRG neuron firing behavior to that recorded prior to toxin exposure (Fig. 3) (n = 4, ANOVA with Tukey’s multiple comparisons test, TTX vs. TTX/sPm1a P = 0.001; TTX vs. wo P > 0.9999; TTX/sPm1a vs. wo P = 0.0010). As the contribution of TTX-S NaV channels was pharmacologically abrogated in this small DRG neuron, the sensitizing effect of sPm1a on firing behavior appears to be supported by enhanced activity of TTX-R NaV channels.

Fig. 3.

sPm1a-enhanced firing is reversible and involves neuronal TTX-R NaV and delayed rectifier KV conductances. (A) Current-clamp recordings of a small DRG neuron stimulated with a ramp current injection (Right Inset) in the presence of TTX (0.5 μM; Left, dark gray), in a combination of TTX and sPm1a (1 μM; Center, red) and after washout of sPm1a (wo, Right, light gray). Center Inset: dV/dt of first AP before (dark gray) and after exposure to sPm1a (red, respectively). (B) Paired number of APs from neurons in (black circle: TTX), in sPm1a (red circle: TTX/sPm1a), and after washout (gray circle: wo). One-way ANOVA with Tukey's multiple comparisons test: **P ≤ 0.01. (C) Firing responses of an adapting small DRG neuron. Control (Left, black); exposure to sPm1a (0.3 μM; Center, green) and complete inhibition of firing upon exposure to A803467 (1 μM; Right, blue). Depolarized membrane potential at the end of the ramp in the A803467-inhibited condition is marked by the red arrow. Center Inset: dV/dt plot of first AP in absence and presence sPm1a (black and green, respectively). Right Inset: stimulation protocol. (D) Paired number of APs from neurons in control (black circle), in sPm1a (1 μM, red circle), and after addition of A803467 (1 μM, blue circle). One-way ANOVA with Tukey’s multiple comparisons test: **P ≤ 0.01; †P ≤ 0.001. (E) (Left) Voltage-clamp recording of small DRG neuron TTX-R Na+ current elicited by a depolarizing pulse (Inset, 25 ms, −10 mV, Vh −80 mV) in the presence of TTX (0.5 μM, dark gray). Exposure to sPm1a (1 μM, red) and A803467 (1 μM, blue). (Right) Diary plot showing reversible TTX-R enhancement by sPm1a and complete inhibition of the sPm1a-sensitive TTX-R component by A803467. (F, Left) DRG KV currents in the absence (black, KV total) and presence of 1 μM sPm1a (gray dotted, sPm1a-R). sPm1a sensitive KV currents are evidenced by point-to-point subtraction (red, sPm1a-S). (Right) Small DRG neuron IKDR measured after 5-s prepulses to progressively depolarized potentials () in the absence and presence of 1 μM sPm1a (n = 5; paired t test for control vs. sPm1a at each prepulse potential, *P < 0.05).

NaV1.8 activity largely underlies DRG TTX-R Na+ currents and these channels are abundantly expressed in nociceptive primary afferents where they are fundamental in mediating persistent pain (18, 19). The sPm1a-mediated increase in firing of DRG neurons was reversed by the NaV1.8-blocker A803467, as shown in Fig. 3 . Under control conditions, the cell shown fired six APs and adapted during the stimulation protocol (0 to 150 pA ramp). For this cell, sPm1a (0.3 μM) (Fig. 3, green) significantly delayed adaptation resulting in an approximately threefold increase in the number of APs. The AP waveform exhibited a clear rising phase inflection in the first derivative (Fig. 3C, Inset) with a diverging pattern of activity for APs evoked in the presence of 0.3 μM sPm1a and those recorded in the absence of peptide (Fig. 3, green and black dV/dt, respectively). Consistent with previous reports (14), the application of A803467 (1 µM) abolished sPm1a-induced neuronal firing (Fig. 3 , blue) (n = 6, one-way ANOVA with Tukey’s multiple comparisons test: Ctr vs. sPm1a P = 0.005; Ctr vs. A803467 P = 0.5989; sPm1a vs. A803467/sPm1a P = 0.0007), supporting the involvement of NaV1.8 channels in the sensitizing effects of this spider-venom peptide. Although AP firing was abolished by inhibition of NaV1.8, the membrane potential at the end of the stimulus ramp remained depolarized (red arrow in Fig. 3 , Right) compared to control conditions (Fig. 2 , cells represented as solid symbols joined by solid lines), suggesting additional effects of sPm1a over other small DRG neuron conductances.

When assessed under voltage clamp, whole-cell NaV currents from small sensory neurons in the presence of TTX (0.5 μM) were augmented significantly by sPm1a as can be observed from the recordings shown in Fig. 3 (stimulus: 25 ms, −10 mV, Vh −80 mV, 0.1 Hz). The dark gray current traces in Fig. 3 correspond to the total TTX-R component of the DRG neuron Na+ current. In the presence of TTX (0.5 μM), exposure to sPm1a (1 μM) resulted in a large increase in inward current (Fig. 3, red) that was reversed by removing sPm1a from the bath containing TTX. This can be readily seen in the diary plot of the accumulated charge during the stimulus estimated by integrating the area under the current trace (Fig. 3 , Right). In small DRG neurons, application of sPm1a (1 µM) (Fig. 3, red) caused an average 6.1 ± 2.8-fold increase in charge mediated by TTX-R NaV channels (n = 6). Removal of sPm1a from the bath solution allowed TTX-R Na+ currents to return to control magnitude whereas application of A803467 (1 μM) reduced the sPm1a-sensitive TTX-R NaV component of small mouse DRG neurons by 87.3 ± 11.1% (n = 5).

The suppression of sPm1a-enhanced firing and TTX-R Na+ current reduction by the NaV1.8-selective inhibitor A803467 suggests that the increase in AP firing is largely driven by enhancement of NaV1.8-mediated Na+ currents. However, TTX-R NaV1.9 channels are also expressed in DRG neurons. The use of a pulse protocol favoring NaV1.9 activation did not reveal a substantial contribution of this isoform in the population analyzed here (), hampering our ability to assess the effects of sPm1a on NaV1.9 mediated currents.

The first derivative of the voltage response to current injection in small DRGs exposed to sPm1a was typically right-shifted in the time axis indicating impairment in AP repolarization (Insets in Figs. 2 and 3 ). Pharmacological inhibition of TTX-R currents suppressed AP firing but the membrane potential remained depolarized in about a third of the small DRG neurons recorded (Fig. 2, solid circles joined by solid lines; i.e., arrow in Fig. 3). Therefore, we assessed whether the KV conductance of mouse small DRG neurons was modulated by sPm1a. Fig. 3 portrays a representative family of outward K+ current traces obtained from a standard current–voltage (I-V) protocol in control (Fig. 3, black, KV total) and after the addition of 1 μM sPm1a, thereby observing a sPm1a-resistant KV current (sPm1a-R) (Fig. 3, gray dotted traces). Point-to-point subtraction revealed the sPm1a-sensitive (sPm1a-S) KV component as a slow-activating outward current (Fig. 3, red), with voltage dependence of activation (V0.5: 5.8 ± 2.2 mV, n = 9) () similar to that of recombinant KV2.1 channels (V0.5: 6.0 ± 0.8 mV, n = 7) (Fig. 4). We also used a double-pulse protocol in which 5-s duration depolarizing prepulses (from −90 to +40 mV) maximally inactivated A-type small DRG KV channels to better observe the delayed rectifier (KDR) component of neuronal KV (). Application of sPm1a (1 µM) under these conditions caused a significant 30 to 40% reduction of KDR at all prepulse potentials tested (n = 5; Ctr vs. sPm1a at each prepulse potential paired t test P < 0.05) (Fig. 3 and ).

Fig. 4.

sPm1a modulates recombinant NaV1.8 and KV2.1 channels. (A) Confocal images of adult mouse DRG neurons showing the overlap between fluorescent signals from NaV1.8 (red) and KV2.1 (green) epitopes (Inset on far-right panel). Counterstain is DAPI (blue) and WGA (cyan). (B) Recombinant NaV1.8 whole-cell currents recorded in HEK293 cells activated by 25-ms pulses to –10 mV (Vh −80 mV, Inset) exposed to progressively higher concentrations of sPm1a (in μM) are shown in different colors. (C) Concentration–response relationship (CRC; EC50 = 1.1 ± 0.1 µM; nH = 1.5 ± 0.1, n = 3 to 9 per point). (D) Families of NaV1.8 currents in the absence (black) and presence of sPm1a (1 μM, red). (Inset) Stimulation protocol (100 ms, –80 mV to 50 mV, Δ5 mV and Vh –80 mV). Traces are shown every 10 mV for clarity. (E) Voltage-dependent activation and SSI of NaV1.8. Normalized macroscopic conductance (G/Gmax) in the absence (black open circle, V0.5: −6.2 ± 1.1 mV, n = 7) and presence of sPm1a (1 μM, red open circle, V0.5: −14.5 ± 0.8 mV, n = 16; unpaired t test P < 0.0001) calculated from peak currents from the protocol in D. The dashed trace corresponds to GsPm1a/GCtr to highlight the effect of sPm1a on the macroscopic conductance of NaV1.8. The voltage dependence of SSI was estimated from a double-pulse protocol (500-ms prepulse from −80 to 10 mV, Δ5 mV, 25 ms pulse to −10 mV, Vh −80 mV). Control (black circle, SSI V0.5: −38.7 ± 1.3 mV, n = 8) and sPm1a (red circle, −29.9 ± 1.3 mV, n = 8; unpaired t test P = 0.0003). (F) Whole-cell KV2.1 currents expressed in TSA201 cells activated by 250 ms pulses to 20 mV (Vh −80 mV, Inset) exposed to progressively higher concentrations of sPm1a (in μM) are shown in different colors. (G) Concentration–response curve for inhibition of recombinant KV2.1 by sPm1a (IC50 = 0.40 ± 0.01 µM; nH = 1.4, n = 7 per point). (H) KV2.1 activation and SSI. Normalized KV2.1 tail current (Itail/Itailmax) in the absence (black open circle, V0.5: 6.05 ± 0.8 mV, n = 7) and presence of sPm1a (red open circle, V0.5: 32.9 ± 0.4 mV, n = 7; unpaired t test P = 0.0001). The dashed trace corresponds to ItailsPm1a/ItailCtr. Voltage dependence of KV2.1 SSI was estimated from a double-pulse protocol (5-s prepulse from −90 to 60 mV, Δ10 mV, 250 ms pulse to 60 mV, Vh −80 mV) (). Control (black circle, SSI V0.5: −24.6 ± 3.5 mV, n = 5) and sPm1a (1 μM; red circle, −18.2 ± 4.9 mV, n = 5; unpaired t test P = 0.3189).

sPm1a Enhances NaV1.8 and Inhibits KV2.1 Channels.

In nociceptors, NaV1.8 and the KDR are the dominating conductances of the AP (20). KV2.1 homo- and heteromeric channels mediate the majority of the delayed rectifier KV current in small DRG neurons (21), and because of their slow activation and inactivation kinetics the activity of KV2 channels is expected to contribute significantly to neuronal repolarization/after-hyperpolarization (22). We performed double immunostaining for NaV1.8 and KV2.1 channels in our primary cultures of dissociated small DRG neurons from adult mice. Fig. 4 shows confocal images of adult mouse DRG neurons in which NaV1.8 (Fig. 4, red) and KV2.1 (Fig. 4, green) are abundant and recognized within the same neurons as evidenced by overlap of individual fluorescence signals (Fig. 4 , Far Right, and ).

sPm1a action on human NaV1.8-mediated Na+ currents in HEK293 cells mirrored its effect on mouse small DRG neurons (Fig. 3), with NaV1.8 currents enhanced by sPm1a in a concentration-dependent manner (Fig. 4). Peak current amplitude, like other macroscopic current metrics, is a complex variable reflecting net changes in gating and conductance, therefore we tracked concentration-dependent effects of sPm1a by measuring the absolute charge (area under the curve) flowing during the stimulation pulse (25 ms, –10 mV, Vh –80 mV) at different concentrations of sPm1a. The concentration–response relationship shown in Fig. 4 was normalized to the absolute charge increase registered upon exposure to 10 µM sPm1a, giving an estimated EC50 of 1.1 ± 0.1 μM (nH = 1.5 ± 0.1, n = 3 to 9 per concentration) for enhancement of hNaV1.8 currents.

The primary and tertiary structure of sPm1a resembles those of venom-derived GMPs. In order to evaluate the effect of sPm1a on recombinant NaV1.8 channels, we compared Na+ currents in the absence and presence of 1 μM sPm1a (Fig. 4, black and red, respectively) that were elicited by 100 ms depolarizing pulses from –80 mV to 50 mV, in 5-mV steps, from a holding potential of –80 mV. (Fig. 4 , Inset). Comparison of Na+ current families (shown in 10-mV steps for clarity) in control (Fig. 4, black) and in the presence of sPm1a (Fig. 4, red) revealed that the gating of NaV1.8 is modified by this spider-venom peptide. sPm1a exposure caused a right-shift in NaV1.8 steady-state inactivation (SSI, V0.5: −38.7 ± 1.3 mV, n = 8; vs. −29.9 ± 1.3 mV, n = 8; P = 0.0003) (Fig. 4, closed symbols) and delayed open-state inactivation kinetics (τinact at –10 mV: Ctr: 7.0 ± 1.0 ms, n = 8; sPm1a: 26.2 ± 3.6 ms, n = 8; P = 0.0002), resulting in a 16.7 ± 5.4% (n = 16) increase in the persistent currents recorded at the end of a 100 ms pulse to −10 mV (Fig. 4 ). An ∼8 mV hyperpolarizing shift in the voltage dependence of activation (V0.5: −6.2 ± 1.1 mV, n = 7; vs. −14.5 ± 0.8 mV, n = 16; P < 0.0001) (Fig. 4, open symbols) and a delay in activation kinetics (τact at –10 mV: Ctr: 0.5 ± 0.3 ms, n = 5; sPm1a: 1.6 ± 0.1 ms, n = 5; P = 0.0083) were detected in the presence of sPm1a. Hence, our data provide compelling evidence that sPm1a is a gating-modifier peptide that strongly potentiates rodent and human NaV1.8-mediated currents.

The similarities in current kinetics, voltage dependence of activation, strong KV2.1 immunoreactivity in mouse DRG neuron primary cultures and its frequent targeting by spider NaV GMPs (23), prompted us to investigate the effects of sPm1a on human KV2.1 heterologously expressed in mammalian cells using automated patch-clamp electrophysiology. We found KV2.1 channels to be sensitive to sPm1a (Fig. 4), with an IC50 of 0.4 ± 0.01 μM (nH = 1.4, n = 7) (Fig. 4). Similar to other spider-venom peptides like hanatoxin, sPm1a caused a ∼27-mV depolarizing shift in the voltage dependence of KV2.1 activation (V0.5: 32.9 ± 0.4 mV, n = 7; P = 0.0001) without significant changes in SSI (Ctr V0.5: −24.6 ± 3.5 mV, n = 5, vs. sPm1a V0.5: −18.2 ± 4.9 mV, n = 5; unpaired t test P = 0.3189) (Fig. 4). Functional enhancement of repolarizing KV currents depresses neuronal excitability in primary sensory neurons; thus, inhibition of KDR by a large shift in the activation voltage as verified for KV2.1 channels suggests that sPm1a may also enhance excitability by altering the neuronal refractory period. These results are consistent with a role for sPm1a inhibition of KDR currents in neuronal sensitization and hyperexcitability leading to pain perception upon P. muticus envenomation.

sPm1a Modulates NaV Channels, But Not Nicotinic Acetylcholine Receptors.

Gating-modifier spider-venom peptides often target multiple NaV channels. Automated patch-clamp screening of human NaV1.1–NaV1.7 isoforms revealed that sPm1a nonselectively inhibits peak currents mediated by all these channels with IC50 values of 0.38 to 2.3 μM (). In the presence of 1 μM sPm1a, depolarization-triggered fast inactivation was substantially delayed three- to fourfold for currents elicited by the TTX-S channels NaV1.1, NaV1.3, NaV1.6, and NaV1.7 (τsPm1a/τCtr = 3.5 to 4.5) (), which are abundantly expressed in peripheral neurons. In contrast, the inactivation kinetics of NaV1.2 (brain), NaV1.4 (skeletal muscle), and NaV1.5 (cardiac muscle) were not significantly modified (τsPm1a/τCtr ∼1) ().

Subexcitability is commonly attributed to inactivation of NaV channels, and in small DRG neurons the TTX-S channel NaV1.7 is important in setting the threshold of AP firing, ultimately regulating neuronal excitability (24–26). A closer examination of NaV1.7 modification by sPm1a (1 μM) by manual patch-clamp of HEK293 cells revealed that this peptide slows NaV1.7 fast inactivation by approximately threefold (τinact Ctr: 0.8 ± 0.2 ms, n = 6; vs. sPm1a: 2.7 ± 0.4 ms, n = 6; unpaired t test P = 0.0017) and shifts the voltage dependence of activation to more positive potentials (V0.5 Ctr: −25.4 ± 1.8 mV, n = 5; vs. sPm1a: −10.6 ± 1.7 mV, n = 6; P = 0.0002) without evident changes to SSI (SSI V0.5 Ctr: −75.5 ± 0.5 mV, n = 5; vs. sPm1a: −76.5 ± 0.4 mV, n = 7; P = 0.1458) ().

We also examined whether the symptoms induced by P. muticus bites may be due in part to modulation of the nicotinic acetylcholine receptors (nAChRs) at the neuromuscular junction. We used two-electrode voltage-clamp electrophysiology to examine the effect of sPm1a on human (h) neuronal and rodent (r) muscle nAChR subtypes expressed in Xenopus laevis oocytes. The amplitude of acetylcholine-evoked currents mediated by hα3β2, hα4β2, hα7, rα1β1δγ, and rα1β1δε nAChRs were unchanged in the presence of 1 μM sPm1a (), suggesting that cholinergic neurotransmission is not directly affected by sPm1a.

sPm1a modulation of TTX-S NaV channels in nociceptive neurons, such as NaV1.7, may contribute to the pain-sensitization evoked by sPm1a. However, the multimodal effects of sPm1a and compositional variability and amplitude of the TTX-S component in DRG neurons challenge our ability to rigorously elucidate the contribution of this conductance to modified excitability within the scope of this study. We have therefore attempted to address this question from a theoretical standpoint in the following section.

Modeling sPm1a-Induced Hyperexcitability of Small DRG Neurons.

sPm1a modulates TTX-R NaV1.8, delayed rectifier KV currents, and TTX-S NaV channels in small DRG neurons and their recombinant molecular counterparts in heterologous expression systems. The overall effects of sPm1a on a variety of small DRG neurons are consistent with sensitization through facilitation of repetitive firing, lower AP thresholds, and decreased firing latency. We, therefore, employed a mathematical model of nociceptor excitability to “dissect” the contribution of selected sPm1a-sensitive conductances in isolation and their concerted actions.

Hodgin–Huxley formulations of NaV1.8 and TTX-S NaV channels, based on those published by Verma et al. (27), were implemented in the Brain-Dynamics Toolbox (28) and NaV1.8 currents were simulated in response to standard voltage-clamp protocols. Model parameters describing voltage dependence and rates of channel activation and inactivation (details in ) (29) were adjusted to recapitulate NaV1.8 behavior both at baseline and in the presence of sPm1a (Fig. 5). This model of the NaV1.8/sPm1a molecular interaction was then incorporated into a single-compartment model of small DRG neuron soma (26) to predict emergent effects on neuronal electrophysiology. In addition to modification of NaV1.8 by sPm1a, the effects on KDR were modeled as a 30% decrease in conductance without changes to kinetic parameters as indicated by our experimental data (Fig. 3). The TTX-S NaV voltage dependence and rates parameters of activation and inactivation in control and in the presence of sPm1a are provided in . Simulated effects on neuronal excitability of individual molecular targets of sPm1a (NaV1.8/TTX-S/KDR) are presented in Fig. 5 and . Under control conditions, current injection generates a single AP (Fig. 5 , Top), consistent with our data from mouse DRG neurons. Inclusion of sPm1a-modified NaV1.8 (in the absence of any sPm1a-dependent changes in KDR or TTX-S) increased the number of APs fired in the modeled neuron (Fig. 5, red dotted trace, 2 APs), while inclusion of either reduced KDR conductance or altered TTX-S NaV gating/conductance alone did not cause an increase in firing (Fig. 5 , Bottom; dotted orange and blue traces, respectively, and Fig. 5 , Left) and likely counteract each other (Fig. 5 , Center, first latency for KDR and TTX-S). However, inclusion of the effects of all sPm1a targets (NaV1.8, KDR, TTX-S) in the simulated neuron resulted in substantially enhanced excitability (Fig. 5 ), consistent with a typical mouse DRG neuron stimulated with an equivalent current injection (Fig. 5), while the simulated number of APs and latency (Fig. 5 ) qualitatively reproduced our in vitro data (Fig. 2 ). Thus, the modifications that we implemented to the Mandge and Manchanda model (30) closely reflect the effects of sPm1a on small DRG neurons reported here.

Fig. 5.

Compounded effects of sPm1a on small DRG NaV and KV conductances contribute to neuronal sensitization. Modeling of sPm1a-induced changes in NaV1.8, KDR, and TTX-S NaV nociceptor excitability. (A) Experimentally recorded NaV1.8-mediated currents (Left) in the absence (black) and presence of sPm1a (1 μM) and corresponding Hodgkin and Huxley formalism fitted traces (Right). (B–D) The model nociceptor was stimulated with a 1 s step current injection of 140 pA (Bottom Inset in B). (B, Top) Model nociceptor firing under control conditions (black). (Middle) Trace resulting from implementation of sPm1a-modified NaV1.8 parameters from A () showing an increase in number of APs (red) and a shift in the second local minima in dV/dt (Inset). (Scale bar, 5 ms.) (Bottom) Traces from individually modeled sPm1a-like inhibition of KDR (orange dotted) and modification of TTX-S conductance (blue dotted). (C) Model nociceptor responses compounding sPm1a effects on NaV1.8 and KDR (Left, orange) or modification of NaV1.8, KDR, and TTX-S (Right, red). (D) Quantification of number of APs, relative first latency (first latX/first latCtr), and expanded first derivative dV/dt of the first AP number of modeled data from each condition. Legend is shown below. (E) Recorded membrane potential responses from a mouse small DRG neuron to a step stimulus (Bottom Inset) in the absence (Top, black) and presence of sPm1a (1 μM, red, Bottom). The first derivative analysis traces to the first AP in both conditions is shown in the Inset. (Scale bar, 5 ms.) (F) Proposed mechanism of sPm1a-induced nociceptor sensitization and its pharmacological reversal. Pm1a produces nociceptor sensitization by promoting repetitive firing (enhancement of NaV1.8 currents) and impairing repolarization (inhibition of delayed rectifier KV channels) with a potential contribution from the modulation of TTX-S NaV channels. Pm1a-induced hyperexcitability can be reversed by inhibition of NaV1.8 and TTX-S NaV currents. From left to right, image credits: Shutterstock/Audrey Snider-Bell; Shutterstock/Rudmer Zwerver; Shutterstock/Eric Isselee; Shutterstock/Egoreichenkov Evgenii.

A key motivation for computationally modeling biological observations is that it allows the study of variables in isolation that would otherwise not be possible, as it is the case of promiscuous (NaV1.8, KDR, NaV TTX-S, and so forth) and multimodally (gating and kinetics) active peptides like sPm1a. Of particular interest is how modification of TTX-NaV conductance (TTX-SC) and kinetics (TTX-SK) contribute to the effects of sPm1a on neuronal excitability. Modeled traces resulting from independent incorporation of changes to TTX-SC and TTX-SK into NaV1.8+KDR models () show that neither of these parameters in isolation (or combined) affects firing responses at the beginning of the current stimulus (). For the neuron in our reductionist model, inhibition of the TTX-S conductance progressively decreases the time between AP bursts which, in combination with an enhancement of NaV1.8 and impairment of KDR, decreases the time between burst for later time points (). Notwithstanding, our model suggests that a sPm1a-like delay in TTX-S NaV inactivation kinetics (TTX-SK) appears to dominate over its inhibitory effects on this conductance ().

It has been shown that TTX-S NaV1.7 activation amplifies small depolarizations that bring peripheral neurons within firing threshold, with the NaV1.8 conductance contributing the bulk of the inward current underlying the AP upstroke during repetitive firing of pain-sensing neurons (31). Experimentally, sPm1a inhibits peak currents and delays fast inactivation of TTX-S NaV channels (). Overall, our computational approach lends support to the notion that modification of the NaV1.8 current alone increases neuronal excitability, yet it is insufficient to fully recapitulate our experimental observations and therefore the additional inhibition of KDR is required for the enhancement of excitability observed in small DRG neurons. Furthermore, multimodal modification of TTX-S NaV channels may also contribute to sPm1a’s actions depending on the complement and relative contribution of TTX-S NaV channels present in individual DRG neurons.

Discussion

Spontaneous activity and increased firing frequency of nociceptors are the key sources of pain in the peripheral nervous system. Although many venom-derived peptides promiscuously modulate the activity of ion channels involved in nociception, their combined actions on neuronal excitability have not been thoroughly elucidated (32). In this study, we identified a peptide from P. muticus venom, δ/κ-TRTX-Pm1a, chemically synthesized sPm1a, determined its 3D structure, and functionally characterized its actions, in vivo, ex vivo, and heterologous expression systems. Mathematical modeling supported our functional findings whereby sPm1a enhances nociceptor excitability mainly by the simultaneous stimulation of repetitive firing (through TTX-R NaV1.8 current enhancement) and impairment of repolarization (by inhibiting KDR, KV2.1), with a potential contribution from TTX-S NaV modified excitability, leading to manifestations of pain behavior in mice and their reversal through inhibition of NaV1.8 and TTX-S NaV channels (Fig. 5).

sPm1a modulates the availability of the quintessentially nociceptive channels NaV1.8 and NaV1.7, as well as a major contributor to membrane repolarization, KV2.1, to produce severe pain. We observed overlapping expression of these NaV channels with KV2.1 in small DRG neurons (Fig. 4 and ), in agreement with previous analyses of unmyelinated C-fiber low-threshold mechanoreceptors, MrgD+ nonpeptidergic nociceptors, and CGRP+ peptidergic nociceptors (16). NaV channel activation mediates the positive feedback mechanisms necessary to initiate an AP, and therefore changes in the availability or distribution of these channels can have a profound effect on cell excitation. AP-firing thresholds are primarily determined by Na+ current density and the voltage dependence of activation and inactivation but are also fine-tuned by other neuronal conductances, in particular by delayed rectifier KV channels.

The TTX-R conductance in DRG neurons is enhanced through the action of sPm1a on the voltage dependence of NaV1.8 activation and inactivation, as well as its current amplitude and kinetics. In parallel, sPm1a induces a depolarizing shift in the voltage dependence of activation of the most abundant low-threshold TTX-S channel in nociceptors, NaV1.7, which results in an apparent loss of current availability at hyperpolarized potentials. Substantially delayed TTX-S NaV channel inactivation kinetics at depolarized potentials is also observed. Inactivation of the TTX-R Na+ current is known to limit the duration of small DRG neuron firing in response to maintained stimuli, hence inactivation of NaV1.8 is believed to underlie adaptation in capsaicin-sensitive small DRG neurons (33), whereas inhibition of inactivation results in sensitization to painful stimuli. Patients suffering from small fiber neuropathies carry gain-of-function mutations in NaV1.8 that lower the threshold for channel activation, accelerate recovery from inactivation, impair inactivation (34), and increase resurgent currents (35). In these patients, DRG neuronal hyperexcitability is characterized by lower AP generation thresholds, increased firing rates in response to suprathreshold stimulation, and spontaneous firing, thus providing a pathophysiological basis for the reported allodynia and hyperalgesia. This neurophysiological behavior is in agreement with the observed actions of sPm1a on endogenous and recombinant NaV1.8 currents, small DRG neuron sensitization, and pain behavior observed in mice.

Together with modification of NaV1.8, the multimodal effects of sPm1a on DRG neuron TTX-S NaV isoforms (NaV1.7, NaV1.1, and NaV1.6) can perhaps contribute to the increased excitability of DRG neurons. For example, the delay in inactivation caused by sPm1a mimics gain-of-function mutations of NaV1.7 that lead to increased excitability and spontaneous activity in nociceptive neurons in congenital chronic pain patients (36). Nevertheless, in any given nociceptor, the relative contribution of particular TTX-S isoforms and the extent to which their conductance and inactivation kinetics are modulated by sPm1a hinder accurate prediction.

The three main NaV isoforms expressed in nociceptive neurons are TTX-R NaV1.8 and NaV1.9 and TTX-S NaV1.7 (36). The major Na+ current during an AP in small DRG somata is mediated by TTX-R NaV1.8 and NaV1.9 channels (37). The pulse protocol used in this study to assess small DRG neuron TTX-R favored activation of NaV1.8 channels. Implementation of an alternative pulse protocol to enhance the observation of NaV1.9-mediated currents showed strong potentiation of NaV1.8-like currents in the presence of sPm1a but little to no NaV1.9-like inward Na+ currents in DRG neurons. Whether sPm1a also interacts with NaV1.9 and how it would affect neuronal excitability and contribute to the observed in vivo nocifensive behavior in mice remains an open question. The absence of NaV1.9-mediated currents in the population of mouse DRG neurons analyzed here, their small cell size, and abundance of NaV1.8-mediated currents suggests a population of C-fiber nociceptors (38). Notably, the suppression of sPm1a-enhanced firing by A803467 suggests that the increase in AP firing and sensitization are largely driven by enhancement of NaV1.8-mediated Na+ currents in nociceptive DRG neurons.

The wide AP waveforms associated with pain-sensing DRG neurons, such as those in MrgD+ nonpeptidergic nociceptors, allow for a large contribution of slow KV2 currents during the AP (16), while also influencing AP repolarization. In small DRG neurons, sPm1a decreased KDR and impaired repolarization consistent with the actions of other gating-modifier venom peptides, such as GxTX-1E (κ-TRTX-Pg1a) (39) and ScTx1 (κ-TRTX-Sc1a) (22). These peptides modify KV2.1 gating properties by shifting its activation to more depolarized voltages (functionally decreasing conductance) and causing neuronal hyperexcitability. Taking these data together, a key role for KV2 channels as carriers of a sizeable proportion of the K+ current during re- and hyperpolarization phases in sensory neurons that encode multiple pain modalities is highlighted (40).

Double immunostaining of NaV1.8 and KV2.1 showed 100% coincidence of these channels in small mouse sensory neurons, consistent with simultaneous detection of their encoding transcripts in >90% of these neurons using single-cell qRT-PCR (41). Yet, only 30% of the recorded DRG neurons in this study displayed a depolarized membrane potential at the end of the ramp stimulus in the presence of sPm1a. Such discrepancy may be related to the specific physiological functions of heteromeric KV channels within particular cellular contexts, as well as major gaps in their pharmacology profiling (42). A large fraction of mouse DRG neuron KDR current is carried by homotetrameric KV2 channels, as well as their heterotetramers with electrically “silent” α-subunits KV5, KV6, KV8, and KV9 (21, 41, 43) known to underlie subtle functional differences that fine-tune excitability patterns in diverse neuronal cell types.

Arachnid venom-derived GMPs have been shown to interact with voltage-gated calcium (CaV), NaV, and KV channels (12, 23, 44–47). We have shown here that sPm1a modulates NaV1.1–NaV1.8 and KV2.1 channels. Experiments assessing endogenous DRG neuron high voltage-activated CaV currents, as well as other recombinantly expressed CaV and KV channels (), failed to identify consistent modulatory effects of sPm1a. However, given the vast array of molecular players involved in neuronal excitability, other sPm1a targets may be identified in future studies.

Pain is a complex process that involves the detection of noxious stimuli in the peripheral nervous system and processing of the nociceptive input in the central nervous system (48). Spontaneous activity in nociceptors has been described for patients with painful polyneuropathy (49). Mechanically insensitive or silent nociceptors comprise ∼30% of all C-fibers in vertebrates, with activation thresholds that are typically very high and thus are thought not to contribute to acute pain signaling (50). However, inflammation and nerve injury may lower AP-firing thresholds, thereby rendering silent nociceptors active, which might be important in mechanical hyperalgesia and central sensitization (51). Increased Na+ influx per AP via TTX-R NaV1.8 is proposed to underlie the characteristically pronounced activity-dependent slowing of conduction in mechano-insensitive, “silent” nociceptors (52). The effects of sPm1a on each nociceptive neuron would rely predominantly on the relative expression of NaV1.8 and the abundance of (sensitive) delayed rectifier K channels, but also on other targets such as TTX-S NaV channels, including NaV1.7. High levels of NaV1.8 expression have been reported in mechano-insensitive CHRNA3+ murine nociceptors and in functionally silent porcine nociceptors (53). Unsilencing by exposure to δ/κ-TRTX-Pm1a/sPm1a upon envenomation would recruit this population of C-fibers to contribute to hyperalgesia. We propose that the lowered AP threshold and decreased repolarization in small DRG neurons observed in the presence of sPm1a enhance neuronal excitability and is likely a key contributor to the hyperalgesic symptoms reported after P. muticus envenomations.

Pain caused by bites from Australian funnel-web spiders (genera Atrax, Hadronyche, and Illawarra) and Brazilian armed spiders (genus Phoneutria) is due, at least in part, to enhancement of neuronal NaV channel activity by δ-hexatoxins and δ-phonetoxins, respectively (4, 54). Treatments for envenomation by these spiders typically involve antivenom (where available), muscle relaxants, narcotics, analgesics, and intravenous calcium (54). We showed that coadministration of A803467 (inhibiting NaV1.8), or TTX (blocking NaV1.1-NaV1.4, NaV1.6, and NaV1.7), reduces sPm1a-evoked nocifensive behavior in mice, while inhibition of NaV1.8 reversed sPm1a-induced hyperexcitability of DRG neurons. Hence, our results are consistent with the use of currently approved and widely available local anesthetics to relieve pain elicited by P. muticus envenomations.

Landmark structure–function studies have shed light on the molecular mechanism of actions of prototypical gating-modifier spider-venom peptides (55, 56). Amphipathic tarantula toxins partially partition into the membrane and interact with extracellular protein–lipid interfaces of voltage-sensor domains binding to the paddle motif (S3b helices, S3–S4 linkers, and S4 helices), and hence are also called voltage-sensor toxins (12, 46, 55–58). Pm1a’s sequence, structure, and functional modification of NaV and KV channel gating are consistent with the action of voltage-sensor toxins. Future research aimed at uncovering the molecular details of the interaction between Pm1a and the channel targets identified herein is warranted as it may contribute important insights into the molecular mechanism of action of GMPs.

In summary, we show here that a single venom peptide can modulate three major determinants of neuronal excitability. Target promiscuity, defined as the ability to modulate the function of more than one type of receptor, is a common feature of spider toxins (23, 59). Many gating-modifier toxins are highly promiscuous and target a wide range of ion channels and receptors (23, 59). For example, Protoxin-I (β/ω-TRTX-Tp1a) from venom of the tarantula Thrixopelma pruriens targets a wide range of NaV channels, CaV3.1, and TRPA1 (transient receptor potential ankyrin 1) (60). Hm1a was originally identified as a KV channel inhibitor (61) but later recognized to elicit robust pain behaviors in mice and hypersensitivity to mechanical stimuli via potent enhancement of NaV1.1 mediated currents in colonic afferents (7). Hence, Hm1a is another example of a toxin with cross-target promiscuity, ultimately resulting in the generation of pain such as we describe here for Pm1a. Recognition of common structures in different receptors has been proposed to contribute to target promiscuity (46, 56, 57, 62). Trading selectivity for functionality through the deployment of a single algogenic peptide-like δ/κ-TRTX-Pm1a to simultaneously modulate multiple receptors involved in AP generation in sensory neurons is an economic and effective tactic to evoke pain in predators as a defense strategy. Targeting a physiological function (excitability) rather than a specific molecular target (i.e., membrane protein) may represent an evolutionary adaptation of pain-producing defensive venoms. By extension, depressing neuronal excitability by targeting complementary players with “multimodal analgesics” may provide novel therapeutic alternatives for the treatment of chronic pain.

Materials and Methods

Animal experiments were conducted in accordance with the Australian National Health and Medical Research Council Code for the Care and Use of Animals for Scientific Purposes (63) and were approved by the local University of Queensland Animal Ethics Committee. Details of the P. muticus venom biochemical and bioinformatics analysis, δ/κ-TRTX-Pm1a purification, sPm1a synthesis, NMR spectroscopy, in vivo and in vitro functional characterization of sPm1a, and all materials used in this study are listed in detail in .