Introduction

The peripheral nervous system comprises the cranial and spinal nerves, which function in relaying sensory and motor signals between the central nervous system and somatic tissues. The cranial peripheral nerves are derived from the neural crest cells (NCCs) and from specialized regional thickenings of the cranial ectoderm called placodes (D'Amico-Martel and Noden, 1983; Hamburger, 1961; Koontz et al., 2023; Trainor, 2014). During cranial nerve neurogenesis, both neural crest cells and ectodermal placode cells delaminate from their respective epithelia, interact, and ultimately lead to the formation of the peripheral nerves (Hamburger, 1961; Kurosaka et al., 2015). While peripheral nerve development has been a subject of study for many decades, the process is not fully understood.

The cranial nerves (CNI-CNXII) are a subset of peripheral nerves originating from the supraspinal elements of the central nervous system. In mice, the cranial nerves emerge at approximately embryonic day (E) 9.5, and axonal projection is apparent by E10.5 (Kurosaka et al., 2015; Sudiwala and Knox, 2019). Each of the twelve cranial nerves provides either sensory, motor, or mixed sensory/motor function to the craniofacial and cervical regions, as well as various abdominothoracic viscera in the case of parasympathetic innervation via the vagus nerve (CNX) (Cordes, 2001). The variety and importance of functions regulated by the cranial nerves are highlighted by the various disorders associated with cranial nerve dysinnervation, collectively known as cranial nerve dysinnervation disorders. In the literature, special attention has been paid to disorders characterized by dysinnervation of the oculomotor (CNIII), trochlear (CNIV), trigeminal (CNV), abducens (CNVI) nerves due to their importance in ocular and facial motor function (Kaeser and Brodsky, 2013; Traboulsi, 2004). Multiple genes are implicated in the pathogenesis of these disorders, including tubulin (Tubb3), kinesin (Kif21a), signal transduction proteins, and various transcription factors (Sall4, Phox2a, Hoxa1, Hoxb1) (Gutowski and Chilton, 2015). A further understanding of cranial nerve development is crucial to understand the pathophysiology of cranial nerve dysinnervation disorders.

The trigeminal ganglion (Vg) is a large sensory and motor nucleus that innervates the vertebrate face and jaw and consists of three main branches, the ophthalmic, maxillary, and mandibular nerves. Their axons originate from principal sensory neurons from within the brain stem, which connect to mesencephalic and spinal tracts (Covell and Noden, 1989). The development of the trigeminal nerve is a long-studied and complex process, with much of the early work utilizing the chicken embryo (Moody and Heaton, 1983a; Moody and Heaton, 1983b; Moody and Heaton, 1983c). Moody and Heaton (Moody and Heaton, 1983a; Moody and Heaton, 1983b) demonstrated that the Vg is required for organization of the trigeminal motor nucleus and subsequent outgrowth of motor axons. Interestingly, when membranes were placed between the trigeminal motor nucleus and the Vg, motor axons would circumvent the obstruction to interact with cell bodies in the Vg (Moody and Heaton, 1983c). In these experiments, only the motor axons that could intercalate with the Vg were able to innervate their motor targets, demonstrating that formation and interactions involving the Vg are required for trigeminal motor development. The branchiomotor neurons that innervate the masseter arise from rhombomere 2 of the hindbrain and chart along the V3 branch of the trigeminal nerve and are involved in jaw elevation and protrusion. These brachiomeric motor pools are specified by the paired homeobox transcription factor Phox2b, which is expressed in ventro-medial regions of the rostral hindbrain (Kang et al., 2007; Pattyn et al., 2000).

Cytoskeletal formation and remodeling are critical processes in both neurogenesis and neuritogenesis (i.e., the formation of axons and dendrites), enabling cell migration, differentiation, and neurite growth (da Silva and Dotti, 2002). We previously characterized Myeloid/lymphoid or mixed-lineage leukemia; translocated to chromosome 11 or ALL1 fused from chromosome 1q or T cell factor 7 cofactor (Mllt11/AF1q/Tcf7c; thereafter referred to as Mllt11) as a vertebrate-specific cytoskeletal-associating protein with restricted expression in the developing central nervous system by E9.5 and in the peripheral ganglia from E10.5 onwards (Yamada et al., 2014). More recently, we showed that Mllt11 is required for neuritogenesis of cortical neurons both in vitro and in vivo, with Mllt11 mouse null mutants exhibiting markedly reduced cortical neurite length and complexity, which affected the formation of cortical white matter tracts and commissural projections (Stanton-Turcotte et al., 2022). In the same study, we also showed that Mllt11 loss-of-function resulted in reduced migration of neurons towards the cortical plate. Mllt11 has also been implicated in cell migration and morphogenesis during retinogenesis. Specifically, knockout of Mllt11 led to failed migration of retinal ganglion cells and amacrine cells during retinal lamination, and these cells were unable to elongate and assume a bipolar morphology consistent with migrating neuroblasts, and instead adopted a more globular morphology associated with reduced migratory ability (Blommers et al., 2023). While we previously showed that Mllt11 regulates neuronal migration and neuritogenesis in the central nervous system, its role in the peripheral nervous system has yet to be explored. Using a Wnt1^Cre2/+ approach to inactivate a floxed Mllt11 allele in NCC derivatives, we now report a requirement for Mllt11 in the formation of the trigeminal nerve, the largest cranial nerve, and innervation of the masseter muscle. That was due, at least in part, to a reduction of Phox2b⁺ branchiomotor neurons in rhombomere 2, which exhibited unexpected Wnt1^Cre2/+ recombination activity in the embryonic hindbrain. Our findings are consistent with a role for Mllt11 in the development of branchiomotor neurons, including neuritogenesis and formation of neuromuscular connections to the muscles of the head.

Results

Previous studies have identified the expression of Mllt11 in the cranial ganglia (Yamada et al., 2014). In order to verify Mllt11 activity in the developing trigeminal ganglia, we utilized a β-gal knock-in allele for the Mllt11 locus, as we previously described (Stanton-Turcotte et al., 2022). Visualization of β-gal expression at E12.5 suggests that the Mllt11 locus is active within the trigeminal ganglion during development (Fig. 1 A,B). We next sought to determine the role of Mllt11 in the formation of the trigeminal ganglion. To investigate this, we generated Wnt1^Cre2/+; Mllt11^flox/flox; R26r^TdTomato/+ conditional knockout (cKO) embryos that had undergone targeted deletion of Mllt11 driven by the Wnt1^Cre2 transgenic allele to target NCC derivatives (Dinsmore et al., 2022; Stanton-Turcotte et al., 2022). We focused on the development of the trigeminal ganglion as it is the largest cranial ganglion that is significantly contributed by NCCs (Kurosaka et al., 2015; Moody and Heaton, 1983b).

Fig. 1. Mllt11 is expressed in the trigeminal ganglion.

(A)Mllt11 locus with a β-gal knock-in cassette reveals expression in the trigeminal ganglion (Vg) at E12.5. (B) High magnification view of Vg depicted in (A). Blue stain represents X-gal expression resulting from Mllt11 locus activity. Vg, trigeminal ganglion; ey, eye.

Figure

Fig. 1. Mllt11 is expressed in the trigeminal ganglion.

(A)Mllt11 locus with a β-gal knock-in cassette reveals expression in the trigeminal ganglion (Vg) at E12.5. (B) High magnification view of Vg depicted in (A). Blue stain represents X-gal expression resulting from Mllt11 locus activity. Vg, trigeminal ganglion; ey, eye.

Close

Mllt11 is not required for the formation of the trigeminal ganglia

We first confirmed Wnt1^Cre2/+ transgene activity in the NCC neurogenic derivatives of the embryonic head, using Cre⁺ males to generate NCC-specific targeting, as recommended previously (Dinsmore et al., 2022), by profiling Ai9 TdTomato reporter activity. The resulting Tomato⁺ signal identified NCC fate-mapped populations through the craniofacial region, including the trigeminal nerve and its branches, which were visualized by β-tubulin III (β3Tub) whole mount immunostaining at E10.5. We observed no difference in overall trigeminal ganglion morphology (Fig. 2 A,B), suggesting that the initial formation of the structure is unaffected by Mllt11 loss. We next examined whether Mllt11 loss affected the cellular composition of the ganglia in the Wnt1^Cre2/+-driven Mllt11 cKO mice using markers for cells arising from the neural crest (Sox10) vs. embryonic ectodermal placodes (Isl1/2) (Figs. 3 and 4, respectively). At E12.5, there was no statistically significant difference in the mean number of cells within the ganglia (Fig. 3 A,D,G; Table 1; p=0.739, T-test). Additionally, there was no statistically significant difference in the proportion of Wnt1^Cre2/+-fate labeled Tomato⁺ NCCs (Fig. 3 B,E,H; Table 1; p=0.184, T-test) or Sox10⁺ cells (Fig. 3 C,F,I; Table 1; p=0.215, T-test) relative to DAPI⁺ cells populating the trigeminal ganglion. Interestingly, Mllt11 cKO mice displayed a high degree of variation in total cell density (Fig. 3G), as well as the proportion of Tomato⁺ NCCs (Wnt1⁺) and Sox10⁺ cells, relative to controls (Fig. 3 H,I). Moreover, Mllt11 cKOs displayed a significant reduction in the ratio of Wnt1⁺ NCCs to Sox10⁺ cells within the trigeminal ganglion, compared to controls (Fig. 3J; Table 1; p=0.013, T-test). We next probed for alterations in placodal populations of the trigeminal ganglion by comparing the number of Isl1/2-expressing cells between control (Fig. 4 A,B) vs. Mllt11 cKO mutants (Fig. 4 C,D), but found no difference in the number of Isl1/2⁺ cells (Fig. 4E; Table 1; p=0.329, n=4, T-test). This suggested that Wnt1^Cre2/+-driven Mllt11 cKOs specifically affected NCC-derived populations of the ganglion, such as Sox10⁺ glial cells, and not placodal cells, which are instead derived from ectoderm.

Fig. 2. Normal trigeminal (Vg) nerve morphology in Mllt11 mutants.

(A, B) Whole-mount immunohistochemistry for neuronal-specific β3Tubulin (β3Tub) at E10.5 illustrating normal Vg cranial nerve morphology in Mllt11 cKO mutants (B) relative to control wildtype mouse embryos (A, Cntl). Blue, β3Tub; red, Wnt1^Cre2/+; Rosa26r^tdTomato/+ fate map; lateral view. Arrowhead indicates trigeminal ganglion.

Figure

Fig. 2. Normal trigeminal (Vg) nerve morphology in Mllt11 mutants.

(A, B) Whole-mount immunohistochemistry for neuronal-specific β3Tubulin (β3Tub) at E10.5 illustrating normal Vg cranial nerve morphology in Mllt11 cKO mutants (B) relative to control wildtype mouse embryos (A, Cntl). Blue, β3Tub; red, Wnt1^Cre2/+; Rosa26r^tdTomato/+ fate map; lateral view. Arrowhead indicates trigeminal ganglion.

Close

Fig. 3. Reduced numbers of NCC-derived Sox10⁺ cells in Mllt11 cKO mutants.

(A-F) Control and Mllt11 cKO E12.5 trigeminal ganglia positive for DAPI (A-D), NCCs (Wnt1^Cre2/+; Rosa26r^tdTomato/+; B, E), and Sox10 (C, F) in transverse sections. Ganglia are outlined in white. Boxes represent example ROIs for cell-count sampling. (G-J) plots comparing the total cell density (G), proportion of NCC (H) and Sox10⁺ (I) cells, and the ratio of NCCs⁺ to Sox10⁺ cells (J) in the trigeminal ganglia between control (black) and Mllt11 cKO embryos (red). Error bars indicate standard deviation; ns, not statistically significant; * p < 0.05. Abbreviations: NCC, neural crest cells; ROI, region of interest.

Figure

Fig. 3. Reduced numbers of NCC-derived Sox10⁺ cells in Mllt11 cKO mutants.

(A-F) Control and Mllt11 cKO E12.5 trigeminal ganglia positive for DAPI (A-D), NCCs (Wnt1^Cre2/+; Rosa26r^tdTomato/+; B, E), and Sox10 (C, F) in transverse sections. Ganglia are outlined in white. Boxes represent example ROIs for cell-count sampling. (G-J) plots comparing the total cell density (G), proportion of NCC (H) and Sox10⁺ (I) cells, and the ratio of NCCs⁺ to Sox10⁺ cells (J) in the trigeminal ganglia between control (black) and Mllt11 cKO embryos (red). Error bars indicate standard deviation; ns, not statistically significant; * p < 0.05. Abbreviations: NCC, neural crest cells; ROI, region of interest.

Close

Fig. 4. Placode-originating cells are unaffected in Mllt11 cKO mutants.

(A-D) Control and cKO E12.5 Vg stained with DAPI (A, C) and anti-Isl1/2 (B, D). Ganglia are outlined in white. Boxes represent example ROIs for cell-count sampling. (E) plot comparing ratio of Isl1/2⁺ and DAPI⁺ cells between controls (black) and Mllt11 cKO (red) placodes. Error bars indicate standard deviation; ns, not statistically significant. Abbreviations: ROI, regions of interest; Vg, trigeminal ganglia.

Figure

Fig. 4. Placode-originating cells are unaffected in Mllt11 cKO mutants.

(A-D) Control and cKO E12.5 Vg stained with DAPI (A, C) and anti-Isl1/2 (B, D). Ganglia are outlined in white. Boxes represent example ROIs for cell-count sampling. (E) plot comparing ratio of Isl1/2⁺ and DAPI⁺ cells between controls (black) and Mllt11 cKO (red) placodes. Error bars indicate standard deviation; ns, not statistically significant. Abbreviations: ROI, regions of interest; Vg, trigeminal ganglia.

Close

Given that Mllt11 loss subtly affected the NCC-derived cell numbers in the trigeminal ganglion, we evaluated if this was due to enhanced cell death by staining for Cleaved Caspase-3 (CC3) (Porter and Janicke, 1999). CC3⁺ cells were detected in E12.5 control (Fig. 5 A-C) and Mllt11 cKO (Fig. 5 D-F) trigeminal ganglia, but we found no evidence of enhanced apoptosis in cKO ganglia (Fig. 5G; Table 1; p=0.263, T-test).

Fig. 5. Trigeminal ganglion composition differences in Mllt11 cKO mutants is not due to cell death.

(A-F) Control (A-C) and Mllt11 cKO (D-F) E12.5 trigeminal ganglia stained with DAPI (A, D), NCC-fate mapped cells (red, B, E), and Cleaved Caspase 3 (CC3; white, C, F) in transverse sections. Ganglia are outlined in white. (G) plot comparing density of CC3⁺ cells in the E12.5 trigeminal ganglia in control (black)vs. Mllt11 cKO (red) embryos. Error bars indicate standard deviation; ns, not statistically significant.

Figure

Fig. 5. Trigeminal ganglion composition differences in Mllt11 cKO mutants is not due to cell death.

(A-F) Control (A-C) and Mllt11 cKO (D-F) E12.5 trigeminal ganglia stained with DAPI (A, D), NCC-fate mapped cells (red, B, E), and Cleaved Caspase 3 (CC3; white, C, F) in transverse sections. Ganglia are outlined in white. (G) plot comparing density of CC3⁺ cells in the E12.5 trigeminal ganglia in control (black)vs. Mllt11 cKO (red) embryos. Error bars indicate standard deviation; ns, not statistically significant.

Close

Table 1

Comparisons of trigeminal ganglion cellular composition at E12.5

Comparison	Control Mean ± SD; n	cKO Mean ± SD; n	P-value
DAPI+ cells	172.5 ± 3.688; 3	169.9 ± 12.11; 5	0.739
NCC/ DAPI+ cells	0.481 ± 0.038; 3	0.435 ± 0.044; 5	0.184
Sox10+/ DAPI+ cells	0.428 ± 0.036; 3	0.512 ± 0.096; 4	0.215
NCC/ Sox10+ cells	1.132 ± 0.125; 3	0.864 ± 0.060; 4	0.013
Isl1/2+/DAPI+ cells	0.758 ± 0.013; 3	0.739 ± 0.028; 4	0.329
CC3+ cells/mm2	0.652 ± 0.093; 3	0.552 ± 0.046; 2	0.263

SD, standard deviation. NCCs were fate labeled with Tomato+. P-value derived from T-test.

Mllt11 cKO in NCCs disrupts trigeminal motor innervation of the masseter

The Vg is required for trigeminal motor innervation via interactions between motor axons and cell bodies within the Vg (Moody and Heaton, 1983a; Moody and Heaton, 1983b; Moody and Heaton, 1983c). As no major difference in trigeminal ganglion development was obvious at E12.5, we reasoned that Wnt1^Cre2/+- mediated ablation of Mllt11 in NCC derivatives, or possibly branchiomotor neurons (see below), may lead to defects in the ganglion that may impair its role in supporting trigeminal motor innervation. We focused on trigeminal motor innervation and examined the presence of nerve fibers and neuromuscular junctions (NMJs) in the masseter muscle. We used neurofilament (NF) immunostaining along with neuromuscular endplate labeling with fluorescent bungarotoxin (BTX) staining, which identifies nicotinic acetylcholine (nAChRs) receptors on muscle fibers. Interestingly, we observed a high degree of variation in NF density in the masseter of E18.5 Mllt11 cKOs compared to controls (Fig. 6 B,E,H,J,M; Table 2; p=0.066, T-test), with the majority of individuals exhibiting reduced NF⁺ fiber staining in the masseter relative to controls (Fig. 6M). In contrast to the disruption in NF bundle density, we did not observe a difference in the density of BTX⁺ staining of NMJs between control and cKO mice (Fig. 6 C,F,H,J,N; Table 2; p=0.120, T-test). Under higher magnification, we confirmed a striking reduction in the number of NMJs (NF⁺/BTX⁺) in E18.5 Mllt11 cKO mouse masseters compared to controls (Fig. 6 H,J,O; Table 2; p=0.034, T-test). Together, these results demonstrate that inactivation of Mllt11 in NCCs and their neurogenic derivatives leads to disruptions in trigeminal nerve motor innervation of the masseter.

Fig. 6. Masseter muscle innervation is disrupted in Mllt11 cKO fetuses.

(A-F) Transverse sections of control (Ctrl, A-C) and Mllt11 cKO (D-F) mouse E18.5 masseter muscles in displaying NCC-fate mapped cells (red; A, B), neurofilament (NF) identifying nerve fibers (A-E), and BTX stain identifying clustered nicotinic acetylcholine receptors (nAChR) characteristic of NMJs (purple; A, C, D, F). (A-F) Muscles are outlined in white. (G-J) Higher magnification views of the masseter NMJs consisting of NF⁺ (green) and BTX⁺ (purple) revealed close contact of nerve fibers and muscle in control (G, H; arrowheads in H), but Mllt11 cKO masseter NMJ lacked co-staining of intramuscular BTX⁺/NF⁺ signal (J). (K) Representative image of control (K) NMJs displaying interdigitation of NF⁺ fibers in BTX⁺ NMJs. (L) Mllt11 cKOs displayed disrupted NMJs, with a failure of NF⁺ fibers to enwrap BTX⁺ NMJs. (M-O) Plots comparing density of NF⁺ fibers (L), BTX⁺ points (M), and NMJs (N), normalized over muscle area, between control (black) and Mllt11 cKO (red) mice, which showed significantly reduced NMJs. Error bars indicate standard deviation; ns, no statistically significant difference; *, p < 0.05. Abbreviations: NMJ, neuromuscular junction, BTX, bungarotoxin toxin.

Figure

Fig. 6. Masseter muscle innervation is disrupted in Mllt11 cKO fetuses.

(A-F) Transverse sections of control (Ctrl, A-C) and Mllt11 cKO (D-F) mouse E18.5 masseter muscles in displaying NCC-fate mapped cells (red; A, B), neurofilament (NF) identifying nerve fibers (A-E), and BTX stain identifying clustered nicotinic acetylcholine receptors (nAChR) characteristic of NMJs (purple; A, C, D, F). (A-F) Muscles are outlined in white. (G-J) Higher magnification views of the masseter NMJs consisting of NF⁺ (green) and BTX⁺ (purple) revealed close contact of nerve fibers and muscle in control (G, H; arrowheads in H), but Mllt11 cKO masseter NMJ lacked co-staining of intramuscular BTX⁺/NF⁺ signal (J). (K) Representative image of control (K) NMJs displaying interdigitation of NF⁺ fibers in BTX⁺ NMJs. (L) Mllt11 cKOs displayed disrupted NMJs, with a failure of NF⁺ fibers to enwrap BTX⁺ NMJs. (M-O) Plots comparing density of NF⁺ fibers (L), BTX⁺ points (M), and NMJs (N), normalized over muscle area, between control (black) and Mllt11 cKO (red) mice, which showed significantly reduced NMJs. Error bars indicate standard deviation; ns, no statistically significant difference; *, p < 0.05. Abbreviations: NMJ, neuromuscular junction, BTX, bungarotoxin toxin.

Close

We next focused on the development of branchiomotor pools that send axons along the V3 branch of the Vg to innervate the masseter. To do this, we examined the distribution of neurons expressing the paired box homeodomain transcription factor Phox2b, which is required for the specification of brachiomeric-targeting motor neurons from the hindbrain, but not required for innervating somatic muscles (Pattyn et al., 2000). Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} E11.5 embryos showed extensive labeling of NCC derivatives, but surprisingly also showed reporter activity in the rostral hindbrain, in rhombomere 2 (Fig. 7A). We therefore examined rhombomere 2 motor neurons by staining for Phox2 and NF in longitudinal sections across the hindbrain of E11.5 embryos (section angle indicated in Fig. 7A). In Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} control embryos, Phox2⁺/ NF⁺ double positive cells were distributed in medial pools next to the mantle zone (mz, Fig. 7 B-D). Confirming the whole mount analysis, we observed Tomato⁺ recombination activity within the mz and Phox2b⁺ branchiomotor neurons in rhombomere 2. These are not NCC-derived tissues and reflect aberrant Wnt1^Cre2/+ activity within the ventromedial region of the rostral hindbrain. Interestingly, in Wnt1^Cre2/+; Mllt11^flox/flox; R26r^{TdTomato/TdTomato} cKOs we noted a decrease in Phox2b⁺/NF⁺ branchiomotor pools clustered medially from the mz (arrows, Fig. 7 E-G; N=4 cKOs, N=3 controls). This region overlapped with Tomato⁺ activity (Fig. 7E), demonstrating that Mllt11 loss in differentiating Phox2b⁺ cells within rhombomere 2 using the Wnt1^Cre2/+driver led to reductions of motor neurons that target the masseter.

Fig. 7. Wnt1^Cre2/+ activity in rhombomere 2 led to reduced Phox2b⁺ branchiomotor neurons in Mllt11 cKO mutants.

(A) Whole mount of E11.5 Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} control embryo. Tomato⁺ expression reflects recombination activity in NCC-derived tissues, including peripheral nerves. Aberrant Tomato⁺ activity was noted in R2. Line indicates longitudinal sectioning plane at the level of R2 depicted in B-G. (B-D) Merge of immunostaining of Phox2b (white, C), NF (green, D), and Tomato expression (fuchsia, B) in Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} control (Cntl) R2 region. Arrows in indicate Phox2b⁺ branchiomotor pool adjacent NF⁺ mz. (B) Tomato⁺ activity reflecting Wnt1^Cre2/+-mediated recombination in mz and adjacent Phox2b⁺ cells (arrows). (E-G) Merge of immunostaining of Phox2b (white, F), NF (green, G), and Tomato expression (fuchsia, E) in Wnt1^Cre2/+; Mllt11^flox/flox; R26r^{TdTomato/TdTomato} conditional knockout (cKO) R2 region. (F) Arrows indicate reduced Phox2b⁺ branchiomotor neurons in cKO. Scale bar: A, 500µm; B, 50µm. Abbreviations: Ma, mandible; Mb, midbrain; Mx, maxillary; mz, mantle zone, NF, neurofilament; R1, rhombomere 1; R2, rhombomere.

Figure

Fig. 7. Wnt1^Cre2/+ activity in rhombomere 2 led to reduced Phox2b⁺ branchiomotor neurons in Mllt11 cKO mutants.

(A) Whole mount of E11.5 Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} control embryo. Tomato⁺ expression reflects recombination activity in NCC-derived tissues, including peripheral nerves. Aberrant Tomato⁺ activity was noted in R2. Line indicates longitudinal sectioning plane at the level of R2 depicted in B-G. (B-D) Merge of immunostaining of Phox2b (white, C), NF (green, D), and Tomato expression (fuchsia, B) in Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} control (Cntl) R2 region. Arrows in indicate Phox2b⁺ branchiomotor pool adjacent NF⁺ mz. (B) Tomato⁺ activity reflecting Wnt1^Cre2/+-mediated recombination in mz and adjacent Phox2b⁺ cells (arrows). (E-G) Merge of immunostaining of Phox2b (white, F), NF (green, G), and Tomato expression (fuchsia, E) in Wnt1^Cre2/+; Mllt11^flox/flox; R26r^{TdTomato/TdTomato} conditional knockout (cKO) R2 region. (F) Arrows indicate reduced Phox2b⁺ branchiomotor neurons in cKO. Scale bar: A, 500µm; B, 50µm. Abbreviations: Ma, mandible; Mb, midbrain; Mx, maxillary; mz, mantle zone, NF, neurofilament; R1, rhombomere 1; R2, rhombomere.

Close

Table 2

Comparison of NMJ innervation of the masseter muscle at E18.5

Comparison	Control Mean ± SD; n	cKO Mean ± SD; n	P-value
NF+ fibers/mm2	280.5 ± 65.8; 4	138.8 ± 114.8; 5	0.066
BTX+ points/mm2	96.6 ± 20.88; 4	72.34 ± 20.10; 5	0.120
NMJs/mm2	79.11 ± 12.71; 4	38.89 ± 28.13; 5	0.034

SD, standard deviation. P-value derived from T-test.

Discussion

The aim of this study was to determine if the novel cytoskeletal-interacting protein Mllt11 plays a role in the formation of the trigeminal ganglion. Previous studies have demonstrated its requirement for neurogenic processes in different systems throughout the mammalian nervous system, including migration and neuritogenesis in the lamination of the neocortex, retina, and cerebellum (Blommers et al., 2023; Blommers et al., 2024; Stanton-Turcotte et al., 2022). Here we show that Mllt11 is required for the proper innervation of the masseter. Using Wnt1^Cre2 to ablate Mllt11 within NCC-derived cranial nerves, we demonstrated that Mllt11 plays a role in maintaining the composition of the trigeminal ganglia and the innervation of a trigeminal motor target, the masseter. Because Mllt11 regulates neuritogenesis in the CNS, it is possible that its NCC-specific knockout leads to microarchitectural changes in the Vg, disrupting interactions with trigeminal motor axons. Surprisingly, during this study, we noted that Wnt1^Cre2 activity was present in rhombomere 2, specifically encompassing most of the branchiomotor pool that innervates the masseter. This is not a NCC-derived population and instead represents aberrant Cre activity within the CNS, and when mated to Mllt11^flox/flox mice, resulted in a reduction of Phox2b⁺ branchiomotor neurons. Our findings raise concerns about using the Wnt1^Cre2/+ mouse driver allele to evaluate gene function in sensory vs motor development in the craniofacial region.

In the trigeminal ganglion at E12.5, fate-mapped NCCs (Tomato⁺ cells) and Sox10⁺ cells represent a mix of both neurons and glia (Schwann cells), as both populations arise from the neural crest (Koontz et al., 2023; Le Douarin and Smith, 1988; Méndez-Maldonado et al., 2020). We observed a modest reduction in NCC-derived Sox10⁺, but not placodal (Isl1/2⁺), upon Wnt1^Cre2/+-mediated Mllt11 ablation. However, Mllt11 loss in Wnt1⁺ cells did not appear to overtly affect NCC migration, as evidenced by largely normal patterns of NCC fate-mapped Tomato⁺ cells populating trigeminal ganglia. Nor was the reduction of Sox10⁺ due to enhanced cell death. We, therefore, wondered if the deletion of Mllt11 in NCC derivatives had any secondary effects on the trigeminal system, specifically on its interaction with trigeminal motor inputs to the masseter. We found that Mllt11 loss in peripheral ganglia resulted in reduced motor innervation of the masseter, as evidenced by the reduction of NF density and NF⁺/BTX⁺ (i.e. NMJs) in the muscle. NMJs are critical sites for communication between motor neurons and muscle fibers, and their proper assembly is essential for muscle contraction and motor control.

Kummer et al., (Kummer et al., 2004) demonstrated the cell-autonomous ability of skeletal muscle cells to form the post-synaptic motor assembly in-vitro without a neuronal presence or signal. This is consistent with our in vivo observations of the maintenance of nAChRs in masseters of Wnt1^Cre/+; Mllt11^flox/flox mutants despite the loss of invading neurites. The main role of the masseter is mastication; however, it plays a minor role in facial expression and speech, by stabilizing the mandible, in humans (Kent, 2004). In mice, the role of the masseter is limited more to mastication and biting, providing the main muscle mass for these functions. Future research into the role of Mllt11 in craniofacial nervous system development should examine the behavioral manifestations of the knockout in post-natal mice.

Despite the reduction in innervation in the Mllt11 mutants, we did not observe any change in the density of BTX-rich regions, signifying that they maintained post-synaptic assembly of the nAChR-dense apparatus. This is not surprising given that Mllt11 expression is restricted to developing neurons (Yamada et al., 2014) and our genetic strategy using the Wnt1^Cre2/+ driver intended to ablate Mllt11 primarily in peripheral neurons. However, while previous reports did not indicate Wnt1^Cre activity in motor nuclei of the rostral hindbrain (Chen et al., 2017), a recent report demonstrated that the rederived Wnt1^Cre2/+ mouse allele does indeed show aberrant recombination in the ventral neural tube (Gandhi et al., 2024). Indeed, our investigations revealed unexpected Wnt1^Cre2/+; R26r^{TdTomato/TdTomato} recombination activity in the Phox2b⁺ branchiomotor pools in rhombomere 2 and the adjacent mantle zone at E11.5. This was associated with reductions in Phox2b⁺ rostral hindbrain motor neurons in Mllt11 cKOs, likely contributing to reduced NMJ formation in the masseter at late fetal stages.

Concerning possible molecular targets of Mllt11 action in nerve development, we previously identified α/β Tubulins, Actin, and atypical Myosins as potential Mllt11-interaction targets in the fetal brain (Stanton-Turcotte et al., 2022). The role of the cytoskeleton in NCC development is supported by work in chicken embryos showing that β3Tub is expressed in pre-neuronal cells of the neural plate and dorsal neural tube, that is, the pre-migratory neural crest (Chacon and Rogers, 2019). Interestingly, this expression coincides with the upregulation of NCC-specific markers such as Sox10. In contrast, Mllt11 does not appear to be expressed in the pre-migratory or migratory NCCs. Instead, it is expressed later in differentiated neurons throughout the central and peripheral nervous systems, co-expressing with robust β3Tub levels (Yamada et al., 2014). It is currently unclear whether Mllt11 promotes cytoskeletal dynamics in filopodia and growth cones to promote migration and axonogenesis, but its cytoskeletal association and neuritogenic phenotype in the cranial ganglia suggest that it may promote the formation of stabilized structures in the axon shaft (da Silva and Dotti, 2002; Kapitein and Hoogenraad, 2015). Similarly, in the CNS, Mllt11 is required for the formation of long-range axonal connections across the corpus callosum and the formation of complex dendritic arborizations in cortical pyramidal neurons (Stanton-Turcotte et al., 2022). In addition, Mllt11 promotes the migration of immature neurons in the cortex and retina, reflecting its expression in nascent neurons or neuroblasts (Blommers et al., 2023; Stanton-Turcotte et al., 2022). Given the restricted expression of Mllt11 in developing cranial ganglia and CNS neurons, but not in migrating crest cells, we reasoned that Mllt11 loss most acutely affected outgrowth of peripherally targeting axons in the craniofacial region; a process that depends on cytoskeletal dynamics. Our findings are consistent with Mllt11 being a microtubule-interacting protein functioning to stabilize the cytoskeleton (Stanton-Turcotte et al., 2022).

However, in our investigations, we discovered that the widely used Wnt1^Cre2/+ allele for targeting NCC populations is also active in the rostral hindbrain mantle zone and at least some branchiomotor neurons in rhombomere 2. This confounded our interpretation of the tissue of origin of jaw muscle innervation defects we observed in the Mllt11 cKOs. Namely, we observed reductions in the Phox2b⁺ branchiomotor neuron pool in rhombomere 2, likely accounting at least in part for the reduced innervation of masseter. Furthermore, the neuritogenetic defects in Mllt11 cKO peripheral neurons also contributed to reduced neuromuscular connectivity in the masseter. Reduced branchiomotor development in rhombomere 2 of Wnt1^Cre2/+ driven Mllt11 cKOs may also explain the subtle disruptions in Vg cellular composition, as interactions between the Vg and V motor axons growing from the brain stem may affect the differentiation of cranial ganglia (Moody and Heaton, 1983a; Moody and Heaton, 1983b). While the Wnt1^Cre2/+; Mllt11 cKO phenotype was variable, in the most striking examples of masseter dysinnervation we observed a near total absence of NF⁺ fibers in the periphery near the muscle target. Future research should expand the analysis presented in this study to examine any possible effects on the innervation of other craniofacial muscles and cutaneous sensory structures.

Conclusions

This study demonstrates that the novel protein Mllt11 plays a role in the formation of the trigeminal ganglion and branchiomotor innervation of one of its target muscles, the masseter. We showed that Mllt11 deficiency using the Wnt1^Cre2/+ mouse driver allele led to alterations in the normal balance of cell types in the ganglion, reduced Phox2b⁺ branchiomotor pools in rhombomere 2, and disrupted the motor innervation from the trigeminal ganglion. The most likely explanation for the phenotype is a combination of reduced branchiomotor neurons and altered neurite outgrowth of cranial nerves following Mllt11 ablation, leading to reduced interactions between the trigeminal ganglion and trigeminal motor fibers, and failed axonal endplate integration in the masseter muscle. Cranial nerves, including the trigeminal nerve, play crucial roles in functions such as facial motor control and sensation. As such, these findings have important implications for our understanding of cranial nerve development and motor function. Our study provides a foundation for future research on cranial ganglion development and the investigation of mechanisms regulating peripheral nerve innervation.

Materials and Methods

Acknowledgements

We gratefully acknowledge funding from the National Science and Engineering Research Council of Canada (RGPIN 03925-20), the Dalhousie University Faculty of Medicine Gladys Osman Estate, and the Killam Foundation for fellowship support to EW. We thank Sarah Whitehead for assistance with animal husbandry and Dr. Victor Rafuse for BTX reagents.

Declarations

References

• Blommers M., Stanton‐Turcotte D., Iulianella A. (2023). Retinal neuroblast migration and ganglion cell layer organization require the cytoskeletal‐interacting protein Mllt11. Developmental Dynamics 252: 305-319.

  Article | PubMed

• Blommers M., Stanton‐Turcotte D., Witt E. A., Heidari M., Iulianella A. (2024). Cerebellar granule cell migration and folia development require Mllt11/Af1q/Tcf7c. Developmental Neurobiology 84: 74-92.

  Article | PubMed

• Chacon J., Rogers C. D. (2019). Early expression of Tubulin Beta-III in avian cranial neural crest cells. Gene Expression Patterns 34: 119067.

  Article | PubMed | PubMed Central

• Chen G., Ishan M., Yang J., Kishigami S., Fukuda T., Scott G., Ray M. K., Sun C., Chen S.Y., Komatsu Y., Mishina Y., Liu H.X. (2017). S pecific and spatial labeling of P0‐Cre versus Wnt1‐Cre in cranial neural crest in early mouse embryos. genesis 55: 10.1002/dvg.23034.

  Article | PubMed | PubMed Central

• Cordes S. P. (2001). Molecular genetics of cranial nerve development in mouse. Nature Reviews Neuroscience 2: 611-623.

  Article | PubMed

• Covell D. A., Noden D. M. (1989). Embryonic development of the chick primary trigeminal sensory‐motor complex. Journal of Comparative Neurology 286: 488-503.

  Article | PubMed

• D'amico‐Martel A., Noden D. M. (1983). Contributions of placodal and neural crest cells to avian cranial peripheral ganglia. American Journal of Anatomy 166: 445-468.

  Article | PubMed

• Da Silva J. S., Dotti C. G. (2002). Breaking the neuronal sphere: regulation of the actin cytoskeleton in neuritogenesis. Nature Reviews Neuroscience 3: 694-704.

  Article | PubMed

• Dinsmore C. J., Ke C.Y., Soriano P. (2022). The Wnt1‐Cre2 transgene is active in the male germline. genesis 60: e23468.

  Article | PubMed | PubMed Central

• Gandhi S., Du E.J., Pangilinan E.S., Harland R.M., (2024). The Wnt1-Cre2 transgene causes aberrant recombination in non-neural crest cell types. bioRxiv Preprint: 2024.11.06.622365.

  Article | PubMed | PubMed Central

• Gutowski N. J., Chilton J. K. (2015). The congenital cranial dysinnervation disorders. Archives of Disease in Childhood 100: 678-681.

  Article | PubMed

• Hamburger V. (1961). Experimental analysis of the dual origin of the trigeminal ganglion in the chick embryo. Journal of Experimental Zoology 148: 91-123.

  Article | PubMed

• Kaeser P.F., Brodsky M. C. (2013). Fourth Cranial Nerve Palsy and Brown Syndrome: Two Interrelated Congenital Cranial Dysinnervation Disorders?. Current Neurology and Neuroscience Reports 13: 352.

  Article | PubMed

• Kang B.J., Chang D.A., MacKay D.D., West G.H., Moreira T.S., Takakura A.C., Gwilt J.M., Guyenet P.G., Stornetta R.L. (2007). Central nervous system distribution of the transcription factor Phox2b in the adult rat. Journal of Comparative Neurology 503: 627-641.

  Article | PubMed

• Kapitein L. C., Hoogenraad C. C. (2015). Building the Neuronal Microtubule Cytoskeleton. Neuron 87: 492-506.

  Article | PubMed

• Kent R. D. (2004). The uniqueness of speech among motor systems. Clinical Linguistics & Phonetics 18: 495-505.

  Article | PubMed

• Koontz A., Urrutia H. A., Bronner M. E. (2023). Making a head: Neural crest and ectodermal placodes in cranial sensory development. Seminars in Cell & Developmental Biology 138: 15-27.

  Article | PubMed | PubMed Central

• Kummer T. T., Misgeld T., Lichtman J. W., Sanes J. R. (2004). Nerve-independent formation of a topologically complex postsynaptic apparatus. The Journal of Cell Biology 164: 1077-1087.

  Article | PubMed | PubMed Central

• Kurosaka H., Trainor P. A., Leroux-Berger M., Iulianella A. (2015). Cranial Nerve Development Requires Co-Ordinated Shh and Canonical Wnt Signaling. PLoS One 10: e0120821.

  Article | PubMed | PubMed Central

• Le Douarin N. M., Smith J. (1988). Development of the Peripheral Nervous System from the Neural Crest. Annual Review of Cell Biology 4: 375-404.

  Article | PubMed

• Méndez-Maldonado K., Vega-López G. A., Aybar M. J., Velasco I. (2020). Neurogenesis From Neural Crest Cells: Molecular Mechanisms in the Formation of Cranial Nerves and Ganglia. Frontiers in Cell and Developmental Biology 8: 635.

  Article | PubMed | PubMed Central

• Moody S. A., Heaton M. B. (1983a). Developmental relationships between trigeminal ganglia and trigeminal motoneurons in chick embryos. I. Ganglion development is necessary for motoneuron migration. Journal of Comparative Neurology 213: 327-343.

  Article | PubMed

• Moody S. A., Heaton M. B. (1983b). Developmental relationships between trigeminal ganglia and trigeminal motoneurons in chick embryos. II. Ganglion axon ingrowth guides motoneuron migration. Journal of Comparative Neurology 213: 344-349.

  Article | PubMed

• Moody S. A., Heaton M. B. (1983c). Developmental relationships between trigeminal ganglia and trigeminal motoneurons in chick embryos. III. Ganglion perikarya direct motor axon growth in the periphery. Journal of Comparative Neurology 213: 350-364.

  Article | PubMed

• Pattyn A., Hirsch M.R., Goridis C., Brunet J.F. (2000). Control of hindbrain motor neuron differentiation by the homeobox gene Phox2b. Development 127: 1349-1358.

  Article | PubMed

• Porter A. G., Jänicke R. U. (1999). Emerging roles of caspase-3 in apoptosis. Cell Death & Differentiation 6: 99-104.

  Article | PubMed

• Stanton-Turcotte D., Hsu K., Moore S. A., Yamada M., Fawcett J. P., Iulianella A. (2022). Mllt11 Regulates Migration and Neurite Outgrowth of Cortical Projection Neurons during Development. The Journal of Neuroscience 42: 3931-3948.

  Article | PubMed | PubMed Central

• Sudiwala S., Knox S. M. (2019). The emerging role of cranial nerves in shaping craniofacial development. genesis 57: e23282.

  Article | PubMed | PubMed Central

• Traboulsi E.I., (2004). Congenital abnormalities of cranial nerve development: overview, molecular mechanisms, and further evidence of heterogeneity and complexity of syndromes with congenital limitation of eye movements. Trans Am Ophthalmol Soc 102: 373-389.

  PubMed | PubMed Central

• Trainor P.A., (2014). Neural Crest Cells: Evolution, Development and Disease. Elsevier/Academic Press, Amsterdam.

  Book

• Yamada M., Clark J., Iulianella A. (2014). MLLT11/AF1q is differentially expressed in maturing neurons during development. Gene Expression Patterns 15: 80-87.

  Article | PubMed