Na(v)1.6a is required for normal activation of motor circuits normally excited by tactile stimulation

Abstract

A screen for zebrafish motor mutants identified two noncomplementing alleles of a recessive mutation that were named non-active (nav(mi89) and nav(mi130)). nav embryos displayed diminished spontaneous and touch-evoked escape behaviors during the first 3 days of development. Genetic mapping identified the gene encoding Na(V)1.6a (scn8aa) as a potential candidate for nav. Subsequent cloning of scn8aa from the two alleles of nav uncovered two missense mutations in Na(V)1.6a that eliminated channel activity when assayed heterologously. Furthermore, the injection of RNA encoding wild-type scn8aa rescued the nav mutant phenotype indicating that scn8aa was the causative gene of nav. In-vivo electrophysiological analysis of the touch-evoked escape circuit indicated that voltage-dependent inward current was decreased in mechanosensory neurons in mutants, but they were able to fire action potentials. Furthermore, tactile stimulation of mutants activated some neurons downstream of mechanosensory neurons but failed to activate the swim locomotor circuit in accord with the behavioral response of initial escape contractions but no swimming. Thus, mutant mechanosensory neurons appeared to respond to tactile stimulation but failed to initiate swimming. Interestingly fictive swimming could be initiated pharmacologically suggesting that a swim circuit was present in mutants. These results suggested that Na(V)1.6a was required for touch-induced activation of the swim locomotor network.

Fig. 1

nav mutants exhibit abnormal spontaneous coiling amplitude, and diminished touch-evoked behaviors. (A) Top: a 22 hpf a wild type sibling exhibiting a single spontaneous coil. Bottom: an aged matched nav mutant embryo exhibiting a weaker spontaneous coil when compared to wild type sibs. (B) Frequency (left) and amplitude (right) of the spontaneous coils (angle of rotation of the tail) of wild type sib (n=33) were greater than that of nav mutant (n=12) embryos at 21 hpf (t test: p < 0.01 for frequency; p < .05 for amplitude). (C) Top: a 24 hpf wild type sibling touched on the head responds with multiple escape contractions. Bottom: an aged matched nav mutant embryo responds with a single contraction. (D) Percent of touch-evoked escape contractions consisting of no contractions, one contraction and greater than one contraction in wild type (n=30) and nav mutant (n=30) embryos at 24 hpf. All differences (asterisks) were significantly different (t test: p <0.05). (E) Top: a 48 hpf wild type sibling touched on the tail responds with an escape contraction followed by swimming. The embryo appears twice in some frames as the behavior was faster than the video capture rate. Bottom: an aged matched nav mutant responds with an escape contraction but no swimming. (F) Progression of the nav phenotype over the first few days of development compared to wild type. Values represent the average ± SEM escape response displayed by either wild type or mutant embryo groups (n = 3 groups each, 25 embryos each group).

Fig. 2

Tactile stimulation induces abbreviated bouts of touch-evoked activity in muscle and motor neurons but normal activation of M cells in nav mutants. (A) A schematic depicting the simplest neural circuit mediating escape contractions. Sensory input from the mechanosensitive Rohon-Beard (RB) neurons activate M cells, which make monosynaptic contacts with motor neurons (MN) that innervate axial skeletal muscle. (B) A Prolonged bout of touch-evoked fictive swimming is observed in skeletal muscle of wild type siblings (n = 5) while an arrhythmic abbreviated response is recorded in nav mutants (n = 5). Arrows here and in panel C indicate the approximate time of stimulus. (C) Prolonged bouts of touch-evoked bursting in primary motor neurons are observed in wild type sibling (n = 5), but not in nav mutant (n = 5) embryos. (D) Touch-evoked M cell spiking recorded extracellularly from wild type sibling (left, n = 5) and nav mutant (right, n = 5) embryos. (*) denotes M cell spiking followed by an electromyogram (EMG). Of note, the amplitude of extracellular activity varies with respect to the location of the recording electrode.

Fig. 3

nav RBs exhibit decreased voltage-gated sodium currents, but retain the ability to generate overshooting action potentials. (A) Whole-cell current responses recorded in wild type and nav mutant RBs (48-52 hpf) following membrane depolarizations. (B) Peak inward current plotted as a function of the membrane potential. Values represent the average ± SEM (n = 12 for wild type, and n = 12 for mutant). * denotes that difference between wild type and mutant was significant (t- test, p < 0.05). (C) Action potentials in wild type and nav RBs evoked by depolarizing current injections (2 ms) shown below.

Fig. 4

Abbreviated fictive swimming can be evoked by NMDA in nav mutants (48-52 hpf). (A) Top: intracellular voltage recordings showing several minutes of NMDA-evoked fictive swimming from a wild type muscle fiber. Bottom: a faster sweep of two episodes of fictive swimming. (B) Top: intracellular voltage recordings showing several minutes of NMDA-evoked fictive swimming from a nav mutant fiber. Bottom: a faster sweep of two episodes of fictive swimming. (C) Cumulative frequency plots (left) of episode periods from wild type siblings and nav mutant embryos (n = 5 for each) reveals no difference in how often episodes of fictive swimming are initiated. Cumulative frequency plots (right) of episode durations from wild type siblings and nav mutant embryos (n = 5 for each) reveals that mutants typically swim for a shorter duration. (D) Fictive swimming frequency in nav mutants is slower when compared to wild type sibling (* p < 0.05).

Fig. 5

Missense mutations found in Na_V1.6a from nav^mi130 and nav^mi89 abolish channel activity. (A) Top: Na_V1.6a membrane topology and location of nav missense mutations. Bottom: sequence alignment of Na_V1.6a from several different species with the conserved leucine 277 and methionine 1461 highlighted in gray. (B) Two electrode voltage-clamp recordings made from oocytes injected with either wild type or nav^mi89 RNA. Of note oocytes exhibited variable endogenous outward currents.

Fig. 6

scn8aa is widely expressed within the CNS and PNS and motor nerves develop normally in nav mutants. (A) Expression of scn8aa in a 24 hpf embryo by the posterior lateral line ganglion (arrowheads) and RB neurons. Scale bar, 200 μm. (B) enlarged image of region indicated in the top panel showing presumptive RB neurons (arrows highlight a few) expressing scn8aa. Scale bar, 50 μm. (C) Expression of scn8aa in a 48 hpf embryo is more widespread. Asterisk denotes the trigeminal ganglion which is shown at higher power in (D). Boxed area highlights RBs shown in (E). (D) Enlarged image of region containing the trigeminal ganglion that was denoted by an asterisk in (C). (E) Presumptive RBs (right, arrows highlight a few) expressing scn8aa. (F) Sideview of the mid-trunk focused on the dorsal branches of the motor nerves (arrows) in a wild type sibling. Motor nerves were labeled with MAb Zn5 at 66 hpf. (G) Sideview of the mid-trunk showing normal dorsal branches of the motor nerves (arrows) in a nav mutant at 66 hpf. Anterior is to the left and dorsal up in both panels.

Fig. 7

Riluzole preferentially blocks Na_V1.6a persistent current, and phenocopies the nav mutant response to touch in wild type embryos. (A) Two electrode voltage clamp recordings from oocytes co-expressing Na_V1.6a and β1 in the absence or presence of Riluzole (50 μM) demonstrating selective blockade of the persistent sodium current. (B) Concentration-response relationship of Riluzole effect on persistent and transient sodium currents. Values represent the average ± SEM (n = 10). Riluzole (10 μM) mimics the nav behavioral response to touch (C), and the abbreviated pattern of touch-evoked synaptic drive to nav axial skeletal muscle in wild type embryos (48 hpf) (D).