1. Developmental Biology
Download icon

Unique morphogenetic signatures define mammalian neck muscles and associated connective tissues

  1. Eglantine Heude
  2. Marketa Tesarova
  3. Elizabeth M Sefton
  4. Estelle Jullian
  5. Noritaka Adachi
  6. Alexandre Grimaldi
  7. Tomas Zikmund
  8. Jozef Kaiser
  9. Gabrielle Kardon
  10. Robert G Kelly
  11. Shahragim Tajbakhsh  Is a corresponding author
  1. Institut Pasteur, France
  2. CNRS UMR 3738, France
  3. Brno University of Technology, Czech Republic
  4. University of Utah, United States
  5. Aix-Marseille Université, CNRS UMR 7288, IBDM, France
Research Article
  • Cited 1
  • Views 1,036
  • Annotations
Cite this article as: eLife 2018;7:e40179 doi: 10.7554/eLife.40179

Abstract

In vertebrates, head and trunk muscles develop from different mesodermal populations and are regulated by distinct genetic networks. Neck muscles at the head-trunk interface remain poorly defined due to their complex morphogenesis and dual mesodermal origins. Here, we use genetically modified mice to establish a 3D model that integrates regulatory genes, cell populations and morphogenetic events that define this transition zone. We show that the evolutionary conserved cucullaris-derived muscles originate from posterior cardiopharyngeal mesoderm, not lateral plate mesoderm, and we define new boundaries for neural crest and mesodermal contributions to neck connective tissue. Furthermore, lineage studies and functional analysis of Tbx1- and Pax3-null mice reveal a unique developmental program for somitic neck muscles that is distinct from that of somitic trunk muscles. Our findings unveil the embryological and developmental requirements underlying tetrapod neck myogenesis and provide a blueprint to investigate how muscle subsets are selectively affected in some human myopathies.

https://doi.org/10.7554/eLife.40179.001

Introduction

The neck is composed of approximately 80 skeletal muscles in humans that allow head mobility, respiration, swallowing and vocalization processes, containing essential elements such as the trachea, esophagus, larynx, and cervical vertebrae. These processes are ensured by a robust network of muscles at the head-trunk interface, a transition zone subjected to a spectrum of human muscle diseases such as dropped head syndrome, oculopharyngeal myopathy, myotonic dystrophy, Duchenne-type dystrophy and congenital muscular disorders (Emery, 2002; Martin et al., 2011; Randolph and Pavlath, 2015). Defining the embryology of these distinct muscle groups is critical to understand the mechanisms underlying the susceptibility of specific muscles to muscular dystrophies. While myogenesis at the cranial and trunk levels has been studied extensively, the developmental mechanisms at the basis of neck muscle formation are poorly documented and often debated (Ericsson et al., 2013).

In vertebrates, head and trunk muscles arise from different mesodermal origins and their development depends on distinct myogenic programs. At the cranial level, the cardiopharyngeal mesoderm (CPM) resides in pharyngeal arches and gives rise to branchiomeric muscles and the second heart field. CPM specification is initiated by the activation of genes such as Mesp1, Islet1 and Tbx1, while Pax7 subsequently marks muscle stem cells (Diogo et al., 2015; Kelly et al., 2004; Nathan et al., 2008; Saga et al., 1996; Sambasivan et al., 2009). In contrast, Pax3 and Pax7 are expressed in somitic mesoderm giving rise to trunk and limb muscles, with Pax3 then being downregulated in most muscles during fetal stages, while Pax7 maintains the stem cell pool (Kassar-Duchossoy et al., 2005; Relaix et al., 2005; Tajbakhsh et al., 1997). After the differential specification of cranial and trunk progenitors, the bHLH myogenic regulatory factors (MRFs) Myf5, Mrf4, Myod and Myog regulate myogenic cell fate and differentiation (reviewed in [Comai and Tajbakhsh, 2014; Noden and Francis-West, 2006]).

In early embryos, Tbx1 is required for robust activation of MRF genes and proper branchiomeric muscle formation (Grifone et al., 2008; Kelly et al., 2004; Kong et al., 2014; Sambasivan et al., 2009). In Tbx1 mutant embryos, the first pharyngeal arch is hypoplastic and posterior pharyngeal arches do not form resulting in variably penetrant defects of masticatory muscles and absence of muscles derived from more posterior arches (Kelly et al., 2004; Lescroart et al., 2015; Theis et al., 2010). In humans, TBX1 is a major gene involved in 22q11.2 deletion syndrome (DiGeorge/velo-cardio-facial syndrome), a congenital disease characterized by cardiovascular defects and craniofacial malformations (Papangeli and Scambler, 2013). In contrast, Pax3 acts upstream of MRF genes in somites and Pax3 mutants have defects of epaxial and hypaxial muscle formation while double Pax3/Pax7-null embryos lack trunk/limb muscles (Brown et al., 2005; Relaix et al., 2005; Tajbakhsh et al., 1997; Tremblay et al., 1998).

The neck constitutes a transition zone characterizing land vertebrates (tetrapods). The major muscle groups in the neck consist of: epaxial back muscles; ventral hypaxial musculature; pharyngeal, laryngeal and esophagus striated muscles located medioventrally; and cucullaris-derived muscles. The cucullaris is a generic term defining putative homologous muscles that are evolutionarily conserved and connect the head and trunk in jawed vertebrates (gnathostomes). In amniotes, the cucullaris represents the embryonic anlage that gives rise to trapezius and sternocleidomastoid muscles which are innervated by the accessory nerve XI (Diogo, 2010; Edgeworth, 1935; Ericsson et al., 2013; Kuratani, 2008; Kuratani et al., 2018; Lubosch, 1938; Tada and Kuratani, 2015).

While the somitic origin of epaxial/hypaxial neck muscles and CPM origin of pharyngeal, laryngeal and esophagus striated muscles are well defined (Gopalakrishnan et al., 2015; Noden, 1983; Tabler et al., 2017), the embryological origin of cucullaris-derived muscles has remained a subject of controversy (Couly et al., 1993; Edgeworth, 1935; Greil, 1913; Huang et al., 1997; Huang et al., 2000; Matsuoka et al., 2005; Noden, 1983; Piatt, 1938; Piekarski and Olsson, 2007). This muscle group was reported to originate either from lateral plate mesoderm (LPM) or CPM populations adjacent to the first three somites in chick and axolotl (Nagashima et al., 2016; Sefton et al., 2016; Theis et al., 2010). However, retrospective lineage analysis indicated that the murine trapezius and sternocleidomastoid muscles are clonally related to second heart-field-derived myocardium and laryngeal muscles, consistent with a CPM origin (Lescroart et al., 2015). Moreover, cucullaris development follows a branchiomeric program and cucullaris-derived muscles were shown to be absent in Tbx1-null mice (Kelly et al., 2004; Lescroart et al., 2015; Sefton et al., 2016; Theis et al., 2010). Nevertheless, the source of the cucullaris is still equivocal due to the location of its embryonic anlagen at the interface of cranial, somitic and lateral plate mesodermal populations.

Skeletal elements and muscle-associated connective tissue (MCT) also have distinct embryological origins along the rostro-caudal axis. The connective tissue of branchiomeric and tongue muscles originate from neural crest cells (NCCs) of cranial origin (Evans and Noden, 2006; Köntges and Lumsden, 1996; Noden, 1983; Noden, 1988; Ziermann et al., 2018b). Cranial NCCs also give rise to skeletal components and tendons in the head. In contrast, the skeleton and connective tissue originate from somitic mesoderm in the trunk and from LPM in limbs (Nassari et al., 2017). The neck and shoulder girdle contain skeletal elements and connective tissues of distinct NCC, LPM or somitic origins (Durland et al., 2008; Matsuoka et al., 2005; McGonnell et al., 2001; Nagashima et al., 2016; Tabler et al., 2017; Valasek et al., 2010). It has been suggested that NCCs form both connective tissue and endochondral cells at the attachment sites of neck muscles to shoulders in mouse (Matsuoka et al., 2005). However, studies in non-mammalian animals have contested a NCC contribution to the pectoral girdle (Epperlein et al., 2012; Kague et al., 2012; Ponomartsev et al., 2017).

Therefore, the neck region consists of muscle, skeletal and connective tissue elements of mixed cellular origins, underscoring the difficulty in assigning embryonic identities for these structures. In addition, the genetic requirements for the formation of non-somitic and somitic neck muscles remain to be defined. To resolve these issues, we used genetic lineage and loss-of-function mice combined with histology, µCT and 3D reconstructions to map the embryological origins of all neck muscles and associated connective tissues. In doing so, we show that cucullaris-derived muscles originate from a posterior CPM population and are differentially affected in Tbx1-null mice. Moreover, we identify a unique genetic network involving both Mesp1 and Pax3 genes for somite-derived neck muscles and we define a new limit of neural crest contribution to neck connective tissue and shoulder components.

Results

Distinct myogenic programs define neck muscle morphogenesis

To investigate the embryological origin of neck muscles in the mouse, we mapped CPM- and somite-derived myogenic cells using lineage-specific Cre drivers including Mef2c-AHF, Islet1, Mesp1 and Pax3 (Figure 1). The Mef2c-AHF (anterior heart field) enhancer is activated in the second heart field and myogenic progenitors of CPM origin (Lescroart et al., 2010; Verzi et al., 2005). Islet1 and Mesp1 genes are both expressed in early CPM and are essential for cardiac development. The Mesp1 lineage also marks some anterior somitic derivatives (Cai et al., 2003; Harel et al., 2009; Saga et al., 2000; Saga et al., 1999). In contrast, Pax3 is activated in all somitic progenitors and is a key actor during trunk and limb muscle formation (Relaix et al., 2005; Tajbakhsh et al., 1997; Tremblay et al., 1998). Given that the majority of Mef2c-AHF derivatives are myogenic cells (Lescroart et al., 2015; Lescroart et al., 2010; Verzi et al., 2005), we analyzed this lineage using Rosa26R-lacZ/+ (R26R) reporter mice. Islet1, Mesp1 and Pax3 genes are also expressed in cells contributing to skeletal components, connective tissues or neurons. To focus on the myogenic lineage, we used Pax7nGFP-stop/nlacZ (Pax7GPL) reporter mice, which mark cells with nuclear β-galactosidase (β-gal) activity following Cre recombination (Sambasivan et al., 2013).

Figure 1 with 3 supplements see all
Genetic lineage tracing of neck muscle progenitors.

Whole-mount X-gal stainings of Mef2c-AHFCre;R26R, Islet1Cre;Pax7GPL, Mesp1Cre;Pax7GPL and Pax3Cre;Pax7GPL mice at E10.5 (A–D), E11.75 (E–H) and E18.5 (I–L’) (n = 3 for each condition). See associated Figure 1—supplements 13. (A–H) Note labeling of mesodermal core of pharyngeal arches (PAs) and cucullaris anlage (ccl) by Mef2c-AHF, Islet1 and Mesp1 lineage reporters; β-gal+ cells in anterior somites of Mesp1Cre embryos and in the clp anlagen of Islet1Cre embryos. Pax3 lineage marked somitic mesoderm. (I–L’) Mef2c-AHF, Islet1 and Mesp1 lineages marked branchiomeric (mas, tpr, dg) and cucullaris muscles (stm, atp and stp). Pax3Cre and Mesp1Cre labeled somitic epaxial neck muscles (epm). atp, acromiotrapezius; ccl, cucullaris anlage; clp, cutaneous maximus/latissimus dorsi precursor; dg, digastric; epm, epaxial musculature; h, heart; hc, hypoglossal cord; lbm, limb muscle anlagen and limb muscles; ltd, latissimus dorsi; mas, masseter; nc, nasal capsule; nt, neural tube; PA1-2, pharyngeal arches 1–2; S3, somite 3; stm, sternocleidomastoid; stp, spinotrapezius; tpr; temporal. Scale bars: in D for A-D and in H for E-H, 1000 µm; in L for I-L’, 2000 µm.

https://doi.org/10.7554/eLife.40179.002

We first examined embryos after myogenic specification (E10.5 and E11.75), and fetuses when muscles are patterned (E18.5). In Mef2c-AHFCre;R26R embryos, β-gal-positive cells were observed in the mesodermal core of pharyngeal arches at the origin of branchiomeric muscles, in second heart field derivatives, and in the cucullaris anlage (Figure 1A,E). A spatiotemporal analysis of the cucullaris using Myf5Cre;Pax7GPL and Myf5Cre;R26mTmG embryos (Figure 1—figure supplement 1) showed that Myf5-derived muscle progenitors located at the level of the posterior pharyngeal arches, and adjacent to somites S1-S3 (Figure 1—figure supplement 1A’), were innervated by the accessory nerve XI (Figure 1—figure supplement 1G–G”). These cells gave rise to the trapezius and sternocleidomastoid muscles (Figure 1—figure supplement 1A–F’) thus confirming the identity of the cucullaris anlage in mouse (Tada and Kuratani, 2015).

In Islet1Cre;Pax7GPL and Mesp1Cre;Pax7GPL embryos, labeling was also observed in pharyngeal arch derivatives and the cucullaris (Figure 1B–C,F–G), the latter showing less contribution from the Islet1 lineage. On sections, a subset of the Myod-positive cells in the cucullaris originated from Islet1-derived cells (Figure 1—figure supplement 2A). Surprisingly, Pax3Cre;Pax7GPL embryos also showed lacZ expression in the cucullaris at E11.75, although no expression was detected at E10.5 (Figure 1D,H). Given that Pax3 and Pax7 are also expressed in neural crest cells (Relaix et al., 2004), and that these Pax3/Pax7-derived cells were excluded from the Myod-positive myogenic population at E12.5 after muscle specification (Figure 1—figure supplement 2B), they are likely to be of NCC origin. As expected, Pax3 lineage tracing also labeled the somite-derived myotomes, hypaxial migrating progenitors that form the hypoglossal cord (origin of tongue and infrahyoid muscles), and limb muscle progenitors. Furthermore, the hypaxial anlage, which is located at the proximal limb bud and gives rise to the cutaneous maximus and latissimus dorsi muscles, was Pax3-derived (Figure 1D,H; Figure 1—figure supplement 1D’) (Prunotto et al., 2004; Tremblay et al., 1998). Unexpectedly, this anlage and the latissimus dorsi muscle were also labeled in Islet1Cre;Pax7GPL but not in Mesp1Cre;Pax7GPL embryos (Figure 1F–G,J–K). On sections at E12.5, Islet1 expression was observed in Pax3-derived cells after the emergence of myogenic cells from the proximal limb bud (Figure 1—figure supplement 2C). In addition, the Mesp1 lineage contributed to anterior somitic derivatives during early embryonic development as previously reported (Loebel et al., 2012; Saga et al., 1999); strong lacZ expression was observed in the hypoglossal cord and somites S1-S6. Labeling decreased in more posterior myotomes and in forelimb muscle progenitors compared to Pax3Cre;Pax7GPL embryos (Figure 1C–D,G–H).

Lineage tracings with Mef2c-AHFCre, Islet1Cre and Mesp1Cre marked branchiomeric (temporal, masseter, digastric, mylohyoid and pharyngeal) and cucullaris-derived neck muscles (acromiotrapezius, spinotrapezius and sternocleidomastoid), all of which were excluded from the Pax3 lineage (Figure 1I–L, Figure 1—figure supplement 2D–G’). These findings support previous studies showing that cucullaris muscle development is controlled by a branchiomeric myogenic program (Kelly et al., 2004; Lescroart et al., 2015; Sefton et al., 2016; Theis et al., 2010). In addition, both Mesp1 and Pax3 lineages labeled somitic neck muscles (Figure 1K–L’, Figure 1—figure supplement 2F–G’).

Analysis of different somite-derived neck muscles on sections showed that Mesp1 and Pax3 lineages gave rise to the great majority of the Pax7-positive myogenic population (Figure 1—figure supplement 3), demonstrating the high recombination efficiency obtained with the Cre lines. The results indicate that neck somitic muscles originate from myogenic cells that have expressed both Mesp1 and Pax3 genes.

To further investigate the contributions of Mesp1 and Pax3 lineages to neck muscles, we examined sections using the R26tdTomato reporter co-immunostained with the myofibre marker Tnnt3 at three representative levels (A, B and C levels in Figure 1; see also Figure 2—figure supplement 1). At anterior levels, while Pax3 lineage contribution was limited to somite-derived neck muscles, the Mesp1 lineage marked branchiomeric muscles (mylohyoid, pharyngeal, laryngeal, esophagus), cucullaris-derived muscles (acromiotrapezius and sternocleidomastoid) and somite-derived neck muscles (Figure 2A–H, Figure 1—figure supplement 2F–G’, Figure 2—figure supplement 2A–H’). The epaxial and hypaxial neck muscles showed equivalent Tomato expression in both Mesp1Cre;R26tdTomato and Pax3Cre;R26tdTomato mice. These observations further indicate that Mesp1 and Pax3 lineages contribute equivalently to neck muscles derived from anterior somites.

Figure 2 with 2 supplements see all
Differential contributions of Mesp1 and Pax3 lineages to neck and shoulders.

Immunostainings on coronal cryosections of E18.5 Mesp1Cre;R26tdTomato and Pax3Cre;R26tdTomato mice for the myofibre Tnnt3 and Tomato markers at levels indicated in Figure 1. Higher magnifications of selected areas in (A–J) are shown in Figure 2—figure supplement 2; (n = 2 for each condition). See also the atlas of neck musculature in Figure 2—figure supplement 1. (A–H) Mesp1Cre labeled all neck muscles including branchiomeric (myh, esm, phm and ilm), cucullaris (stm, atp), somitic epaxial (epm) and hypaxial (tg, lcp, lcl, ifh) muscles. Pax3Cre marked somitic muscles. (I–J) At shoulder level, Mesp1-derived cells did not contribute to posterior somitic myofibres including scapular muscles (scp) compared to that observed in Pax3Cre embryos. ac, arytenoid cartilage; acp, scapular acromion process; atp, acromiotrapezius; cc, cricoid cartilage; epm, epaxial musculature; esm, esophagus striated muscle; hh, humeral head; ifh, infrahyoid muscles; ilm, intrinsic laryngeal muscles; lcl, longus colli; lcp, longus capitis; myh, mylohyoid; ob, occipital bone; oc, otic capsule; phm, pharyngeal muscles; stm, sternocleidomastoid; scp, scapular musculature; tc, thyroid cartilage; tg, tongue. Scale bars: in J for A-B 200 µm, for C-J 400 µm.

https://doi.org/10.7554/eLife.40179.006

At the shoulder level, we observed less Mesp1 contribution to more posterior somitic muscles (Figure 2I–J). In contrast to that observed at anterior levels, little or no Tomato expression was detected in myofibres of scapular muscles in Mesp1Cre;R26tdTomato mice (Figure 2—figure supplement 2I–J’). Therefore, Mesp1 lineage contribution was restricted to epaxial and hypaxial neck muscles, in contrast to pectoral and trunk muscles that originate from the Pax3 lineage (Figures 12) (Table 1). These observations lead us to propose that three distinct myogenic programs are involved in the formation of neck and pectoral musculature at the head-trunk interface. The branchiomeric and cucullaris-derived muscles depend on a common myogenic program involving Mef2c-AHF, Islet1 and Mesp1 lineages; the somitic neck muscles that originate from anterior somites derive from both Mesp1 and Pax3 lineages; the pectoral muscles derived from more posterior somites depend on the activation of Pax3 only (Table 1).

Table 1
Contribution of Mef2c-AHF, Islet1, Mesp1 and Pax3 lineages to neck and pectoral musculature.
https://doi.org/10.7554/eLife.40179.009
Mef2c/Islet1/Mesp1-derived musclesMesp1/Pax3-derived musclesPax3- derived muscles
Mylohyoid
Digastric muscles 


Pharyngeal muscles
Intrinsic laryngeal muscles
Esophagus striated muscle 

Sternocleidomastoid 
Acromiotrapezius 
Spinotrapezius 
Epaxial neck muscles
(splenius, semispinalis,
levator scapula,
rhomboid occipitalis,
suboccipital and
postvertebral muscles)

Hypaxial neck muscles
(tongue muscles*,
infrahyoid muscles,
longus capitis, longus colli)
Scapular muscles
(supraspinatus, Infraspinatus,
subscapularis)

Pectoralis

Latissimus dorsi
Cutaneous maximus
Branchiomeric myogenic programAnterior-most somite myogenic programMore posterior somite myogenic program
  1. *Including intrinsic and extrinsic tongue muscles of somitic origin

    Also derived from an Islet1 lineage

Dual neural crest and mesodermal origins of neck connective tissues

To define the cellular origin of neck muscle-associated connective tissue (MCT), we traced the contribution of different embryonic populations using Mesp1Cre;R26tdTomato and Pax3Cre;R26tdTomato mice as well as Wnt1Cre and Prx1Cre reporters that label NCC and postcranial LPM derivatives, respectively (Burke and Nowicki, 2003; Danielian et al., 1998; Durland et al., 2008). Both NCC and LPM populations were reported to contribute to trapezius MCT (Durland et al., 2008; Matsuoka et al., 2005). Moreover, it was suggested that the postcranial LPM is a source for cucullaris-derived muscles (Theis et al., 2010). A direct comparison of NCC and LPM derivatives allowed us to clarify the contribution of these two populations to cucullaris formation (Figures 34).

Figure 3 with 4 supplements see all
Neural crest contribution to neck muscle-associated tissue.

Immunostainings on coronal cryosections of E18.5 Wnt1Cre;R26tdTomato mice at levels indicated in Figure 1. Tnnt3/Tomato immunostainings are shown in (A–D) and immunostainings for Tuj1/Tomato on selected areas of (A–D) are shown with higher magnifications in (A’–D’). See associated Figure 3—figure supplement 14; (n = 2). (A–A’) Note high Wnt1 contribution in the acromiotrapezius but not in epaxial muscles where Wnt1-derived cells marked neuronal cells. (B–C’) Wnt1-derived cells marked differentially the distinct muscles composing the sternocleidomastoid and laryngeal musculatures. (D–D’) At shoulder level, the Wnt1 cells did not contribute to attachment of acromiotrapezius to scapula. ac, arytenoid cartilage; acp, scapular acromion process; atp, acromiotrapezius; cc, cricoid cartilage; clm, cleidomastoid; clo, cleido-occipitalis; ct, cricothyroid; epm, epaxial musculature; hh, humeral head; ifh, infrahyoid muscles; lca, lateral cricoarytenoid; MCT, muscle-associated connective tissue; pca, posterior cricoarytenoid; phm, pharyngeal muscles; scp, scapular musculature; std, sternomastoid; tam, thyroarytenoid muscle; tc, thyroid cartilage; vm, vocal muscle. Scale bars: in D’ for A-D 400 µm for A’-D’ 200 µm.

https://doi.org/10.7554/eLife.40179.010
Figure 4 with 2 supplements see all
Prx1-LPM lineage contribution to neck and pectoral girdle.

See also Figure 4—figure supplement 1, 2. (A–D) X-gal stainings of Prx1Cre;R26R reporter mice at E9.5 (n = 3) (A) and E18.5 (n = 3) (C–D), and immunostaining for GFP and the Pax7/Myod/My32 myogenic markers in Prx1Cre;R26mTmG E12.5 embryo (n = 2) (B). Note Prx1-derived cells in postcranial LPM (A, black arrowheads) and Prx1-derived cells among, but not in, cucullaris myofibres (B–D). (E–F’’) Immunostaining for β-gal, Tnnt3 and Tcf4 on coronal cryosections of E18.5 Prx1Cre;R26R mice (n = 2) showed β-gal+ cells constituting the pectoral girdle (E, level C in Figure 1) and in MCT fibroblasts (F-F’’, white arrowheads), but not in trapezius myofibres. acp, scapular acromion process; atp, acromiotrapezius; ccl, cucullaris anlage; ccp, scapular coracoid process; cl, clavicle; epm, epaxial musculature; hh, humeral head; lb, forelimb bud; lbm, limb muscle anlagen; LPM, lateral plate mesoderm; ltd, latissimus dorsi; PA1-6, pharyngeal arches 1–6; S3, somite 3; scp, scapular muscles; stm, sternocleidomastoid; stp, spinotrapezius. Scale bars: in A for A, B 500 µm; in C for C-D 2000 µm, for E 500 µm; in F’’ for F-F’’ 20 µm.

https://doi.org/10.7554/eLife.40179.015

We first investigated the distribution of neck muscles and NCCs using Myf5nlacZ/+, Mef2c-AHFCre;R26R, Pax3Cre;R26R and Wnt1Cre;R26R embryos (Figure 3—figure supplement 1). At E10.5, the cucullaris anlage was positioned at the level of posterior pharyngeal arches where Wnt1-derived-positive cells were detectable (Figure 1A–C, Figure 1—figure supplement 1A’, Figure 3—figure supplement 1A–B). Subsequently, the cucullaris progenitors expanded caudally from E11.5 to E13.5. The posterior limit of the cranial NCC domain also extended posteriorly; however, the Wnt1-labeled cells did not cover the posterior portion of cucullaris-derived muscles (Figure 3—figure supplement 1C–H). At E14.5, the acromiotrapezius and spinotrapezius attained their definitive position in Myf5nlacZ/+ and Mef2c-AHFCre;R26R embryos (Figure 3—figure supplement 1I–J). Wnt1-derived cells were observed in the anterior acromiotrapezius muscle, but not in the spinotrapezius that was situated in a Pax3-derived domain (Figure 3—figure supplement 1K–L). Analysis of whole-mount embryos indicated that the cranial NCCs did not contribute to connective tissue of posterior cucullaris derivatives, in contrast to what was reported previously (Matsuoka et al., 2005).

To further analyze NCC contribution to the cervical region at the cellular level, we performed immunostainings on sections for Tomato and Tnnt3 in E18.5 Wnt1Cre;R26tdTomato fetuses (Figure 3, Figure 3—figure supplement 2). Given that the Wnt1 lineage is a source of both neuronal and connective tissue derivatives, we associated Tomato immunostaining with Tuj1 that marks neuronal cells and with Tcf4 that labels MCT fibroblasts (Figure 3, Figure 3—figure supplements 23). At the cranial level, the MCT of branchiomeric (masseter, mylohyoid), tongue and acromiotrapezius muscles was derived from Wnt1- and Pax3-lineages but not from the mesodermal Mesp1 lineage (Figure 3—figure supplement 2A–B’, Figure 3—figure supplement 3A,F, Figure 3—figure supplement 4A–D,G). The acromiotrapezius showed a high contribution from Wnt1-derived cells while the underlying epaxial muscles had considerably less labeled cells that were limited to the neuronal Tuj1-positive population (Figure 3A–A’). The Wnt1 lineage gave rise to Tcf4-positive fibroblasts in the acromiotrapezius, but not in epaxial neck muscles, where fibroblasts were derived from the Mesp1 lineage (Figure 3—figure supplements 3B–C and 4E). These observations are in accordance with a NCC origin of branchiomeric, anterior trapezius and tongue connective tissue as reported previously (Matsuoka et al., 2005).

However, the NCC contribution to connective tissue in the sternocleidomastoid subset of cucullaris-derived muscles appeared more heterogeneous than that observed in the acromiotrapezius. In rodents, the sternocleidomastoid is composed of three individual muscles (cleidomastoid, sternomastoid and cleido-occipitalis portions); a differential NCC contribution to MCT was observed in these muscles. While Wnt1-derived NCCs were widely present in the sternomastoid and cleido-occipitalis, the NCC contribution was limited in the cleidomastoid (Figure 3B–B’). Indeed, Tcf4-positive fibroblasts in the cleido-occipitalis originated from the Wnt1 lineage, whereas the majority of MCT fibroblasts in the cleidomastoid were derived from the Mesp1 lineage (Figure 3—figure supplements 3D–E and 4F).

A differential contribution of NCCs to connective tissue was also seen within the laryngeal and infrahyoid musculature. Extensive Wnt1 lineage contributions to MCT was observed in laryngeal muscles (thyroarytenoid and cricothyroid) that connect to the thyroid cartilage, which is of NCC origin (Figure 3C–C’) (Tabler et al., 2017). In contrast, the laryngeal muscles (cricoarytenoid and vocal muscles) that link mesoderm-derived laryngeal cartilages (cricoid, arytenoid and medio-caudal portion of the thyroid) did not contain NCC-derived connective tissue (Figures 2G–H and 3C–C’) (Tabler et al., 2017). In these muscles, the Wnt1-derived cells were neuronal, as observed in the esophagus, whereas the MCT fibroblasts were derived from the Mesp1 lineage (Figure 3C–C’, Figure 3—figure supplements 2D–D’ and 4H). As another example, Wnt1-derived cells contributed to a greater extent to MCT in infrahyoid muscles (thyrohyoid muscles) that connect the hyoid and thyroid cartilage that are of NCC origin, compared to infrahyoid muscles (omohyoid and sternohyoid muscles) that link posteriorly pectoral structures of mesodermal origin (Figure 3—figure supplement 2C,C’, H; Figure 3—figure supplement 3G–H). These observations suggest that MCT composition within laryngeal and infrahyoid muscles correlates in part with the embryonic origin of the skeletal components to which they attach (Figure 2G–H, Figure 3C–C’, Figure 3—figure supplement 2C–C’, H).

Given our findings that connective tissues of neck muscles have differential contributions of NCC and mesodermal populations, we analyzed the caudal connections of the cucullaris-derived muscles to the pectoral girdle (Figure 3D, Figure 3—figure supplement 2E–H). The acromiotrapezius attaches dorsally to the nuchal ligament and ventrally to the scapular acromion process in continuity with the scapular spine. While Wnt1-derived cells were present dorsally (Figure 3A, Figure 3—figure supplement 2E), this contribution diminished gradually and was undetectable at the insertion on the scapula (Figure 3D–D’, Figure 3—figure supplement 2F). Similarly, the sternocleidomastoid muscle showed limited NCC contribution to the attachment sites of the clavicle and sternum (Figure 3—figure supplement 2G–H). In contrast to what was previously described (Matsuoka et al., 2005), we did not observe NCC contribution to the shoulder endochondral tissue nor to the nuchal ligament (Figure 3—figure supplement 2E–H). Taken together, these observations define a novel boundary for neural crest contribution to neck/pectoral components. The posterior contribution limit of neural crest to branchiomeric MCT occurs at the level of laryngeal muscles that connect to NCC skeletal derivatives. Moreover, NCCs do not participate in connecting posterior cucullaris and infrahyoid muscles to their skeletal elements.

To assess the cellular origin of cucullaris connective tissue at posterior attachment sites, we next traced the contribution of lateral plate mesoderm (LPM) to the neck/shoulder region using Prx1Cre reporter mice (Durland et al., 2008; Logan et al., 2002) (Figure 4, Figure 4—figure supplements 12). Analysis of E9.5 embryos showed that Prx1-derived cells contribute to the forelimb bud and cells adjacent to the anterior-most somites, but not to pharyngeal arches (Figure 4A). At E12.5, the postcranial Prx1-derived domain clearly defined the lateral somitic frontier along the rostrocaudal axis (Durland et al., 2008) and did not include the cucullaris anlage (Figure 4—figure supplement 1, white arrowheads). Whole-mount immunostainings for the myogenic markers Pax7/Myod/My32 and for GFP in Prx1Cre;R26mTmG embryos showed that Prx1-derived cells were present in the dorsal part of the cucullaris but did not contribute to myofibres (Figure 4B, white arrowheads). At E18.5, the Prx1 lineage marked the limb, scapular and abdominal regions, whereas only a few Prx1-derived cells were detected in the cucullaris-derived sternocleidomastoid, acromiotrapezius and spinotrapezius muscles (Figure 4C–D). On sections, immunostaining for β-gal and Tnnt3 showed that Prx1-derived LPM contributed to limb/shoulder MCT and to skeletal components of the pectoral girdle (Figure 4E, Figure 4—figure supplement 2A–B). In contrast, fewer Prx1-derived cells were detected in the acromiotrapezius and little or no contribution was observed in the epaxial muscles (Figure 4E, Figure 4—figure supplement 2B–C). In addition, only a limited number of Prx1-derived cells gave rise to Tcf4-positive fibroblasts in the trapezius muscles, but they contributed more extensively to the fibroblast population in scapular muscles (Figure 4F–F’’, white arrowheads, Figure 4—figure supplement 2D–D”). Notably, β-gal expression for this lineage was not detected in trapezius myofibres thereby confirming the results obtained at E12.5 (Figure 4B–F”, Figure 4—figure supplements 12).

Therefore, these observations reveal a dual NCC/LPM origin of trapezius connective tissue, with a decrease of NCC contribution at posterior attachment sites. Moreover, our analysis shows that the postcranial LPM does not give rise to cucullaris myofibres in contrast to what was suggested previously (Theis et al., 2010), thus providing further evidence for a branchiomeric origin of the cucullaris.

Divergent functions of Tbx1 and Pax3 in neck development

Given the key role for Tbx1 and Pax3 genes in the specification of the CPM and somites respectively, we analyzed the effect of inactivation of these genes on neck muscle formation, compared to the muscle phenotypes observed at cranial and trunk levels.

Analysis has been performed by immunostainings on sections and 3D reconstructions of the neck and pectoral girdle using high-resolution micro-computed tomographic (µCT) scans of control, Tbx1-/- and Pax3-/- fetuses (Figures 56).

Neck muscle phenotypes in Tbx1 and Pax3 mutants.

(A–I) Immunostainings for Tnnt3 on coronal cryosections of control, Tbx1-null and Pax3-null fetuses at E18.5 (n = 3 each condition). Yellow asterisks indicate missing muscles. Note absence of branchiomeric laryngeal (ilm), esophagus (esm) and trapezius (atp) muscles and severe alteration of somitic infrahyoid muscles (ifh) in Tbx1 mutants. Scapular (scp) and pectoral (ptm) muscles are missing in Pax3 mutants. ac, arytenoid cartilage; atp, acromiotrapezius; cc, cricoid cartilage; cl, clavicle; epm, epaxial musculature; esm, esophagus striated muscle; ifh, infrahyoid muscles; ilm, intrinsic laryngeal muscles; lcp, longus capitis; ptm, pectoralis muscles; sc, scapula; scp, scapular muscles; st, sternum; tc, thyroid cartilage; tg, tongue. Scale bars: in A for A-I 500 µm.

https://doi.org/10.7554/eLife.40179.018
3D reconstructions of neck musculoskeletal system in Tbx1 and Pax3 mutants.

See interactive 3D PDFs in Supplementary file 13; control n = 1; mutants n = 2. (A–C) Branchiomeric and cucullaris-derived muscles marked by Mef2c-AHF/Islet1/Mesp1 lineages are indicated in pink. (D–F) Anterior somitic muscles (Mesp1, Pax3 lineages), in red. (G–I) Scapular muscles from more posterior somites (Pax3 lineage), in violet. atp, acromiotrapezius; cc, cricoid cartilage; dg, digastric muscles; epm, epaxial musculature; ifh, infrahyoid muscles; ilm, intrinsic laryngeal muscles; lcl, longus colli; lcp, longus capitis; myh, mylohyoid; sc, scapula; scp, scapular muscles; stm, sternocleidomastoid; tc, thyroid cartilage; tg, tongue.

https://doi.org/10.7554/eLife.40179.019

In the early embryo, Tbx1 is expressed in pharyngeal mesoderm and is required for proper branchiomeric muscle formation (Grifone et al., 2008; Kelly et al., 2004). While Tbx1 is expressed in other cranial populations including the pharyngeal ectoderm and endoderm (Arnold et al., 2006; Huynh et al., 2007), the gene is known to be required cell autonomously during CPM myogenesis (Kong et al., 2014; Zhang et al., 2006). Analysis of Tbx1 mutants revealed unexpected features in cucullaris and hypaxial neck muscle formation. As previously described (Gopalakrishnan et al., 2015; Kelly et al., 2004), anterior branchiomeric muscles (digastric and mylohyoid) showed phenotypic variations, whereas posterior branchiomeric muscles (esophagus and intrinsic laryngeal muscles) and the acromiotrapezius were severely affected or undetectable (Figure 5B,E,H; Figure 6B) (Table 2). However, detailed examination of the cucullaris-derived muscles revealed a heterogeneous dependence on Tbx1 function that was not reported previously (Lescroart et al., 2015; Theis et al., 2010). Unexpectedly, the sternocleidomastoid muscle was present bilaterally but smaller (Figure 6B); the different portions (cleido-occipitalis, cleidomastoid and sternomastoid) were unilaterally or bilaterally affected in a stochastic manner. Moreover, while the epaxial neck and scapular muscles were unaffected (Figure 5E, Figure 6E–H), the hypaxial neck muscles derived from anterior somites were altered. Indeed, the tongue and longus capitis were reduced and the infrahyoid and longus colli muscles were severely affected or undetectable (Figure 5B,H, Figure 6E,H; see interactive 3D PDFs in Supplementary file 12).

Table 2
Summary of the neck muscle phenotype observed in Tbx1- and Pax3-null fetuses.
https://doi.org/10.7554/eLife.40179.020
Tbx1-nullPax3-null
Branchiomeric muscles (Mef2c-AHF/Islet1/Mesp1 lineage)
Mylohyoid
Digastric muscles
Intrinsic laryngeal muscles
Esophagus striated muscle
Sternocleidomastoid
Acromiotrapezius

+/-
+/-


+/-

++
++
+
++
+
+
Anterior-most somite muscles (Mesp1/Pax3 lineage)
Epaxial musculature
Longus capitis
Longus colli
Infrahyoid muscles
Tongue muscles*

++
+/-


+

+
++
++
+/-
+
More posterior somite muscles (Pax3 lineage)
Scapular muscles
Pectoralis

++
++


  1. ++,normal; +, altered morphology; +/-, affected; -, severely affected or undetectable

    *Including intrinsic and extrinsic tongue muscles of somitic origin

Analysis of Pax3 mutants showed that the neck and pectoral muscles were differentially affected. As expected, branchiomeric and epaxial muscles developed normally but displayed morphological differences adapted to malformations noted in some skeletal components (Figure 5C,F; Figure 6C,I). However, whereas hypaxial trunk/limb muscles were severely affected or undetectable in Pax3 mutants (Figure 5F,I; Figure 6F,I) (Tajbakhsh et al., 1997; Tremblay et al., 1998), surprisingly the majority of hypaxial neck muscles derived from both Mesp1 and Pax3 lineages were present. Tongue muscles were reduced in size but patterned, the infrahyoid were hypoplastic, whereas the longus capitis and longus colli were unaffected (Figure 5C; Figure 6F,I; see interactive 3D PDF in Supplementary file 3). The phenotypes of the different muscle groups observed in Tbx1 and Pax3 mutants are summarized in Table 2 (see also Figure 7—figure supplement 1).

Taken together, these observations revealed that hypaxial muscles in the neck were less affected in Pax3 mutants than more posterior hypaxial muscles, pointing to distinct requirements for Pax3 function during neck and trunk muscle formation. In addition, Tbx1 mutants exhibited more severe phenotypes in hypaxial neck muscles, thus highlighting distinct roles for this gene in branchiomeric and hypaxial neck myogenesis.

Discussion

The embryological origins of neck muscles and connective tissues at the head-trunk interface have been poorly defined largely due to their localization at a transition zone that involves multiple embryonic populations. Using a combination of complementary genetically modified mice and 3D analysis that identifies muscles in the context of their bone attachments, we provide a detailed map of neck tissue morphogenesis and reveal some unexpected features regarding the muscle and connective tissue network.

Branchiomeric origin of cucullaris-derived muscles

The mammalian neck consists of somitic epaxial/hypaxial muscles, branchiomeric muscles and cucullaris-derived muscles (Table 1). The latter constitute a major innovation in vertebrate history, connecting the head to the pectoral girdle in gnathostomes and allowing head mobility in tetrapods (Ericsson et al., 2013). Recent studies in different organisms including shark, lungfish and amphibians suggest that the cucullaris develops in series with posterior branchial muscles and that its developmental origin and innervation is conserved among gnathostomes (Diogo, 2010; Ericsson et al., 2013; Naumann et al., 2017; Noda et al., 2017; Sefton et al., 2016; Tada and Kuratani, 2015; Ziermann et al., 2018a; Ziermann et al., 2017). However, multiple embryological origins including CPM, LPM and somites have been reported for the cucullaris, underscoring the difficulty in deciphering the morphogenesis of this and other muscles in the head-trunk transition zone (Huang et al., 2000; Nagashima et al., 2016; Sefton et al., 2016; Theis et al., 2010).

Our study shows that the cucullaris anlage is innervated by the accessory nerve XI and develops contiguously with the mesodermal core of posterior arches and anterior-most somites 1–3. Our lineage analysis reveals that cucullaris development depends on a branchiomeric myogenic program involving Mef2c-AHF, Islet1 and Mesp1 lineages in keeping with previous results (Table 1) (Lescroart et al., 2015; Sefton et al., 2016; Theis et al., 2010). However, our detailed functional analysis and 3D reconstructions lead us to modify the view of the genetic requirements of cucullaris-derived muscles (Lescroart et al., 2015; Theis et al., 2010). Notably, these muscles are differentially affected in Tbx1-null fetuses; the acromiotrapezius does not form while the sternocleidomastoid is present but reduced. Therefore, Tbx1 is differentially required for sternocleidomastoid and trapezius formation, suggesting that distinct subprograms regulate cucullaris development.

We also demonstrate that the cucullaris anlage is excluded from the postcranial Prx1-derived expression domain, which delineates the trunk LPM field (Figure 4). The Prx1 lineage instead gives rise to connective tissue, thereby excluding a contribution from LPM to cucullaris-derived myofibres. Thus, our results, combined with innervation studies, retrospective clonal analyses and grafting experiments in chick and axolotl (Lescroart et al., 2015; Nagashima et al., 2016; Sefton et al., 2016; Tada and Kuratani, 2015), suggest that the postcranial extension of the CPM lateral to the first three somites in tetrapod embryos is a source of cucullaris myogenic cells (Figure 7A). The discordance with previous studies regarding the origin of the cucullaris is likely due to its proximity to both anterior somites and LPM (Figure 7A–B), and consequently, to potential contamination of embryonic sources in grafting experiments (Couly et al., 1993; Huang et al., 1997; Huang et al., 2000; Noden, 1983; Piekarski and Olsson, 2007; Theis et al., 2010).

Figure 7 with 1 supplement see all
Model for musculoskeletal and connective tissue relationships during murine neck development.

See also Figure 7—figure supplement 1. (A, C) CPM (pink), anterior somites (red) and more posterior somites (violet) muscles are defined by three distinct myogenic programs. (B) Note that the cucullaris develops in a NC domain (blue dots), but is excluded from the postcranial LPM (yellow dots). (C) Dual NC/LPM origin of trapezius connective tissue is indicated in (a). NC contribution to connective tissue extends to tongue and anterior infrahyoid musculature (b). (D) Mixed origins of muscle connective tissues at the head-trunk-limb interface. Example of representative muscles: (a) masseter, (b) spinalis dorsi, (c) deltoid. atp, acromiotrapezius; ccl, cucullaris; CPM, cardiopharyngeal mesoderm; epm, epaxial neck musculature; hpm, hypaxial neck musculature; hy, hyoid bone; LPM, postcranial lateral plate mesoderm; NC, neural crest; PA1-2, pharyngeal arches 1–2; PM, paraxial mesoderm; stm, sternocleidomastoid; stp, spinotrapezius; tc, thyroid cartilage.

https://doi.org/10.7554/eLife.40179.021

A unique genetic program for somite-derived neck muscles

Our study also points to a unique Mesp1/Pax3 genetic program in anterior somites for epaxial/hypaxial neck muscle formation (Table 1). While it was shown that the Mesp1 lineage gives rise to tongue muscles (Harel et al., 2009), we demonstrate that it also contributes to all neck muscles. In chordates, Mesp genes appear to play a conserved role in cardiogenesis and axis segmentation. In mouse, Mesp1 inactivation causes early embryonic death from abnormal heart development, and Mesp1/Mesp2 double-knockout embryos lack non-axial mesoderm (Moreno et al., 2008; Saga, 1998; Saga et al., 2000; Satou et al., 2004; Sawada et al., 2000). During early murine development, Mesp1 shows two waves of activation; initially in the nascent mesoderm destined for extraembryonic, cranial and cardiac mesoderm at the onset of gastrulation; later during somitogenesis, transient Mesp1 expression is limited to anterior presomitic mesoderm (Saga, 1998; Saga et al., 1996; Saga et al., 2000; Saga et al., 1999). Our lineage analysis shows that Mesp1 extensively labels the anterior mesoderm, including the CPM and anterior somites 1–6, while contribution decreases in more posterior somites (Figure 1) (Loebel et al., 2012; Saga et al., 2000; Saga et al., 1999). Previous fate mapping experiments have shown that the mesoderm of late-streak stage embryos contributes to both CPM and anterior somites (Parameswaran and Tam, 1995). It appears that the first wave of Mesp1 expression defines not only the CPM field but also includes the mesoderm destined for anterior somites. In contrast, the Mesp1-labeled cells observed in more posterior somites using the sensitive Pax7GPL reporter may result from the transient wave of Mesp1 expression in the presomitic mesoderm during axis segmentation. Furthermore, we show that Mesp1-derived anterior somites give rise to all epaxial/hypaxial neck muscles in contrast to trunk/limb muscles originating from more posterior somites marked by Pax3. The boundary of Mesp1 lineage contribution to muscles corresponds to the neck/pectoral interface. Our findings indicate that the anterior somitic mesoderm employs a specific transition program for neck muscle formation involving both Mesp1 and Pax3 genes implicated in CPM and somitic myogenesis, respectively (Figure 7A–C).

Whereas little is known about the function of Mesp genes in chordates, there is evidence that Mesp1 might be differentially required during anterior versus posterior somitic formation. In mouse, different Mesp1 enhancer activities have been observed between CPM/anterior somites and posterior somites indicating that the regulation of Mesp1 expression might differ in the two embryonic compartments (Haraguchi et al., 2001). In zebrafish, quadruple mutants of Mesp genes (Mesp-aa/-ab/-ba/-bb) lack anterior somite segmentation while the positioning of posterior somite boundaries is unaffected, suggesting distinct requirements for Mesp genes in anterior and posterior somites (Yabe et al., 2016). Interestingly, during early ascidian development, Mesp is expressed in B7.5 founder cells that give rise to both CPM and anterior tail muscles (ATM) (Satou et al., 2004). In Ciona, the CPM precursors at the origin of heart and atrial siphon (pharyngeal) muscles depend on the ascidian homologs of Mesp1, Islet1 and Tbx1 (reviewed in [Diogo et al., 2015]), indicating that a conserved genetic network promotes chordate myogenesis in the anterior embryonic domain.

Our lineage analysis also reveals an unexpected contribution of Islet1-derived cells to the formation of cutaneous maximus and latissimus dorsi muscle progenitors (Table 1) (Prunotto et al., 2004; Tremblay et al., 1998). Islet1 is activated in a subset of CPM progenitors giving rise to branchiomeric muscles and second heart field myocardium (Cai et al., 2003; Harel et al., 2009; Nathan et al., 2008). At the trunk level, while Islet1 is widely expressed in the nervous system and in the LPM forming the hindlimb bud (Cai et al., 2003; Yang et al., 2006), to our knowledge its expression in somitic myogenic cells has not been reported. The cutaneous maximus and latissimus dorsi muscles are missing in both Pax3 and Met mutants (Prunotto et al., 2004; Tajbakhsh et al., 1997; Tremblay et al., 1998). Therefore, the formation of the latissimus dorsi and cutaneous maximus muscles depends on a specific developmental program implicating Pax3, Islet1 and Met genes. Given that the latissimus dorsi and cutaneous maximus participated in the gain in mobility of the forelimbs towards the shoulder girdle in tetrapods, our findings provide insights into their genetic and evolutionary origins.

Our detailed analysis of Tbx1- and Pax3-null mice on sections and in 3D reconstructions now provides a clarified view of neck muscle morphogenesis (Table 2). In both Tbx1 and Pax3 mutants, whereas the epaxial neck musculature is unaffected, the hypaxial muscles originating from anterior somites are perturbed with a more severe phenotype observed in Tbx1 mutants (Table 2). Whereas no Tbx1 expression has been reported in early myotomes in somites, Tbx1 transcripts appear in hypaxial limb and tongue precursors after myogenic specification (Grifone et al., 2008; Kelly et al., 2004; Zoupa et al., 2006). Tbx1-null embryos show normal myotomal and limb muscle morphology while the hypoglossal cord is hypoplastic, resulting in reduced tongue musculature (Table 2) (Grifone et al., 2008; Kelly et al., 2004). Therefore, we cannot exclude the possibility that Tbx1 is activated and plays a role after specification of neck hypaxial muscles (Okano et al., 2008; Zoupa et al., 2006). The hypaxial muscle defects might also be secondary to a failure of caudal pharyngeal outgrowth (Kelly et al., 2004). While Tbx1 acts cell autonomously in mesodermal progenitors (Kong et al., 2014; Zhang et al., 2006), its expression in pharyngeal endoderm might imply an indirect role in CPM myogenesis (Arnold et al., 2006). Defects in signaling from pharyngeal endoderm may explain the hypoglossal cord deficiency and the potential non-autonomous role for Tbx1 in neck hypaxial myogenesis. Detailed analysis of muscle formation in conditional Tbx1 mutants is needed to resolve the relative roles of Tbx1 in neck myogenesis.

It has been shown that hypaxial muscles are perturbed to a greater extent than epaxial muscles in Pax3 mutants (Tajbakhsh et al., 1997; Tremblay et al., 1998), suggesting a different requirement for Pax3 in these muscle groups, possibly through differential gene regulation (Brown et al., 2005). An unexpected outcome of our analysis was that hypaxial neck muscles (derived from Mesp1 and Pax3 lineages) are less perturbed in Pax3-null mutants than hypaxial trunk/limb muscles (Pax3 lineage only) that are severely altered or undetectable (Table 2). Our results indicate that Pax3 is not essential for the formation of neck muscles derived from anterior somites in contrast to hypaxial muscles originating from more posterior somites. These observations support our model that a distinct genetic program governs somitic neck muscles compared to more posterior trunk muscles.

Connectivity network of the neck and shoulders

Assessing the non-muscle contribution to the neck region is a major challenge due to the extensive participation of diverse cell types from different embryological origins. Previous studies in amphibians, chick and mouse reported that branchiomeric and hypobranchial connective tissue originates from NCCs (Hanken and Gross, 2005; Köntges and Lumsden, 1996; Matsuoka et al., 2005; Noden, 1983; Olsson et al., 2001; Ziermann et al., 2018b). It has been shown that the neural crest provides connective tissue for muscles that link the head and shoulders, whereas mesodermal cells give rise to connective tissue for muscles connecting the trunk and limbs (Matsuoka et al., 2005).

Our findings demonstrate that not all branchiomeric muscles are composed of neural crest-derived connective tissue, thereby redefining a new limit for NCC contribution to the neck and shoulders. Unexpectedly, we noted that the contribution of the neural crest lineage is limited in infrahyoid and posterior branchiomeric muscles that connect skeletal components of mesodermal origin. Indeed, it appears that the connective tissue of muscles that link exclusively mesodermal skeletal derivatives is of mesodermal origin. In contrast, the connective tissue of cucullaris-derived muscles is of a mixed origin, first developing in a cranial NCC domain at early stages, then expanding to incorporate connective tissue from both neural crest and LPM populations (Figure 7B). While NCCs are present in the anterior acromiotrapezius, sternocleidomastoid and infrahyoid muscles, contribution gradually decreases at posterior attachment sites and is undetectable at scapular level. In parallel, the LPM gives rise to shoulder skeletal components and to connective tissue at the attachment sites of associated musculature including trapezius muscles (Figure 7C). Therefore, the dual NCC/LPM origin of the trapezius connective tissue correlates with the embryonic origin of skeletal components to which it is connected.

Wnt1Cre and Sox10Cre NCC reporter mice were used to show that endochondral cells connecting the cucullaris-derived muscles on the scapula, clavicle and sternum share a common NCC origin with the connective tissue (Matsuoka et al., 2005). However, NCCs are not found in pectoral components of fish, axolotl and chick, while contribution to neurocranium is conserved, suggesting that NCC involvement in shoulder formation would be specific to mammals (Epperlein et al., 2012; Kague et al., 2012; Piekarski et al., 2014; Ponomartsev et al., 2017). In contrast to this view, our lineage analysis reveals that the neural crest lineage shows limited contribution to cucullaris connective tissue and does not form endochondral cells at the posterior attachment sites (Figure 7C). Differences in genetic lineage tracers and reagents might explain these discordant results (Matsuoka et al., 2005).

Taken together, our findings indicate that the gradient of neural crest and mesodermal contributions to neck connective tissue depends on the embryonic source of attachment sites. Therefore, it reveals that connective tissue composition in the neck region correlates with the cellular origin of associated skeletal components, independently of the myogenic source or ossification mode, forming a strong link between muscles and bones of the head, trunk and limb fields (Figure 7D).

Evolutionary and clinical perspectives

Our findings demonstrate that the hybrid origin of the skeletal, connective tissue and muscle components of the neck is defined during early embryogenesis. The close proximity of neural crest, CPM, LPM and somitic populations is unique along the body plan and underscores the difficulty in defining their relative contributions to structures in the neck (Figure 7A–B). Our results refine the relative contributions of the neural crest and mesodermal derivatives in mouse, thereby providing a coherent view of embryonic components at the head-trunk interface in gnathostomes. Our study highlights the limited NCC contribution to posterior branchiomeric and infrahyoid muscle connective tissue, that is instead of mesodermal origin. This reinforces recent notions suggesting that the cranial NCCs and the postcranial rearrangement of mesodermal populations at the head-trunk interface had been central for the establishment of the neck during gnathostome evolution (Adachi et al., 2018; Kuratani et al., 2018; Lours-Calet et al., 2014; Nagashima et al., 2016; Sefton et al., 2016). The contribution of anterior mesoderm in the origin of the neck needs to be elucidated in future studies of gnathostomes.

Our study reveals that neck muscles develop in a complex domain that is distinct from the head and trunk (Figure 7A–D), and that might be a contributing factor to pathologies that affect subsets of neck muscles in specific myopathies (Emery, 2002; Randolph and Pavlath, 2015). In human, TBX1 has been identified as a major candidate gene for 22q11.2 deletion syndrome (Papangeli and Scambler, 2013). Laryngeal malformations, esophageal dysmotility and shortened neck are frequent in patients. Moreover, the neck deficiencies might not be exclusively due to cervical spine abnormalities but also to neck muscle defects (Hamidi et al., 2014; Leopold et al., 2012; Marom et al., 2012). Therefore, our analysis of Tbx1-null mutants provides a better understanding of the etiology of the 22q11.2 deletion syndrome and has direct implications in establishing clinical diagnosis in cases where patients present failure in neck-associated functions.

Materials and methods

Key resources table
Reagent type
(species) or resource
DesignationSource or referenceIdentifiersAdditional information
Strain, strain
background
(Mus musculus)
B6D2F1/JRjJanvier
Genetic
reagent
(M. musculus)
Mef2c-AHFCrePMID:16188249MGI:3639735Dr. Brian L Black
(Cardiovascular Research
Institute, University of
California, USA)
Genetic
reagent
(M. musculus)
Islet1CrePMID:11299042MGI:2447758Dr. Thomas M Jessell
(Howard Hughes Medical
Institute, Columbia
University, USA)
Genetic
reagent
(M. musculus)
Mesp1CrePMID:10393122MGI:2176467Pr. Yumiko Saga
(National Institute of
Genetics, Japan)
Genetic
reagent
(M. musculus)
Pax3CrePMID:22394517MGI:3573783Dr. Jonathan A. Epstein
(Perelman Shool of Medicine,
University of Pennsylvania, USA)
Genetic
reagent
(M. musculus)
Myf5CrePMID:17418413MGI:3710099Dr. Mario R Capecchi
(Institute of Human
Genetics, University
of Utah, USA)
Genetic
reagent
(M. musculus)
Wnt1CrePMID:9843687MGI:J:69326Pr. Andrew P. McMahon
(Keck School of Medicine
of the University of
Southern California, USA)
Genetic
reagent
(M. musculus)
Prx1CrePMID:12112875MGI: J:77872Dr. Clifford J Tabin
(Department of genetics,
Harvard Medical School, USA)
Genetic
reagent
(M. musculus)
Pax7GPLPMID:19531352MGI:3850147Dr. Shahragim Tajbakhsh
(Department of Developmental
and Stem Cell Biology,
Institut Pasteur, France)
Genetic
reagent
(M. musculus)
Rosa26R-lacZPMID:9916792MGI:1861932Pr. Philippe Soriano
(Icahn School of Medicine
at Mt. Sinai, USA)
Genetic
reagent
(M. musculus)
R26mTmGPMID:17868096MGI:3716464Pr. Philippe Soriano
(Icahn School of Medicine
at Mt. Sinai, USA)
Genetic
reagent
(M. musculus)
R26tdTomatoPMID:20023653MGI:3809524Dr. Hongkui Zeng
(Allen Institute for Brain
Science, USA)
Genetic
reagent
(M. musculus)
Myf5nlacZ/+PMID:8918877MGI:1857973Dr. Shahragim Tajbakhsh
(Department of Developmental
and Stem Cell Biology,
Institut Pasteur, France)
Genetic
reagent
(M. musculus)
Tbx1-nullPMID:11242110MGI:2179190Dr. Virginia Papaioannou
(Department of Genetics
and Development,
Columbia University
Medical Center, USA)
AntibodyChicken polyclonal
anti-β-gal
AbcamCat. #: ab9361IF (1:1000)
AntibodyRabbit polyclonal
anti-β-gal
MP BiomedicalsCat. #: MP 559761IF (1:750)
AntibodyChicken polyclonal
anti-GFP
Aves LabsCat. #: 1020IF (1:500)
AntibodyChicken polyclonal
anti-GFP
AbcamCat. #: 13970IF (1:1000)
AntibodyMouse monoclonal
IgG1 anti-Islet1
DSHBCat. #: 40.2D6IF (1:1000)
AntibodyMouse monoclonal
IgG1 anti-My32
SigmaCat. #: M4276IF (1:400)
AntibodyMouse monoclonal
IgG1 anti-Myod
DakoCat. #: M3512IF (1:100)
AntibodyMouse monoclonal
IgG1 anti-Pax7
DSHBCat. #: AB_528428IF (1:20)
AntibodyRabbit polyclonal
anti-Tcf4
Cell SignallingCat. #: C48H11IF (1:150)
AntibodyMouse monoclonal
IgG1 anti-Tnnt3
SigmaCat. #: T6277IF (1:200)
AntibodyRabbit polyclonal
anti-Tomato
ClontechCat. #: 632496IF (1:500)
AntibodyMouse monoclonal
IgG2a anti-Pax7
OzymeCat. #: BLE801202IF (1:1000)
Software,
algorithm
GE phoenix
datos|x 2.0
GE Sensing and
Inspection Technologies
GmbH
Software,
algorithm
3D PDF makerSolidWorks
Corporation
Software,
algorithm
ZenZeiss
Chemical
compound, drug
X-galFisherCat. #: 10554973
Chemical
compound, drug
paraformaldehydeElectron Microscopy
Sciences
Cat. #: 15710
Chemical
compound, drug
Triton X-100SigmaCat. #: T8787
Chemical
compound, drug
Tween 20SigmaCat. #: P1379
Chemical
compound, drug
Histoclear IINational DiagnosticsCat. #: HS-202

Animals

Animals were handled as per European Community guidelines and the ethics committee of the Institut Pasteur (CTEA) approved protocols (APAFIS#6354–20160809 l2028839). Males carrying the Cre driver gene, Mef2c-AHFCre (Verzi et al., 2005), Islet1Cre (Srinivas et al., 2001), Mesp1Cre (Saga et al., 1999), Pax3Cre (Engleka et al., 2005), Myf5Cre (Haldar et al., 2007), Wnt1Cre (Danielian et al., 1998), Prx1Cre (Logan et al., 2002), were crossed to reporter females from previously described lines including Pax7GPL (Sambasivan et al., 2013), Rosa26R-lacZ (R26R) (Soriano, 1999), R26mTmG (Muzumdar et al., 2007) and R26tdTomato (Madisen et al., 2010). Myf5nlacZ/+ KI mice and mice carrying the Tbx1tm1pa allele (referred to as Tbx1-null) were previously described (Jerome and Papaioannou, 2001; Kelly et al., 2004; Tajbakhsh et al., 1996). To generate experimental Pax3-null fetuses, Pax3WT/Cre males and females were intercrossed (Engleka et al., 2005) (n = 5 Tbx1 and Pax3 mutants analysed including n = 2 by µCT scanning). Mice were crossed and maintained on a B6D2F1/JRj background and genotyped by PCR. Mouse embryos and fetuses were collected between E9.5 and E18.5, with noon on the day of the vaginal plug considered as E0.5.

X-gal and immunofluorescence stainings

Whole-mount samples were analysed for beta-galactosidase activity with X-gal (0.6 mg/ml) in 1X PBS buffer (D1408, Sigma, St. Louis, MO) containing 4 mM potassium ferricyanide, 4 mM potassium ferrocyanide, 0.02% NP-40 and 2 mM MgCl2 as previously described (Comai et al., 2014). For immunostaining on cryosections, foetuses were fixed 3 hr in 4% paraformaldehyde (PFA) (15710, Electron Microscopy Sciences, Hatfield, PA) 0.5% Triton X-100 (T8787, Sigma) at 4°C, washed overnight at 4°C in PBS 0.1% Tween 20 (P1379, Sigma), cryopreserved in 30% sucrose in PBS and embedded in OCT for 12–16 µm sectioning with a Leica cryostat (CM3050 S, Leica, Wetzlar, Germany). Cryosections were dried for 30 min and washed in PBS. For immunostaining on paraffin sections, samples were fixed overnight in 4% PFA, dehydrated in graded ethanol series and penetrated with Histoclear II (HS-202, National Diagnostics, Atlanta, GA), embedded in paraffin and oriented in blocks. Paraffin blocks were sectioned at 10–12 µm using a Leica microtome (Reichert-Jung 2035). Sections were then deparaffinized and rehydrated by successive immersions in Histoclear, ethanol and PBS. Samples were then subjected to antigen retrieval with 10 mM Citrate buffer (pH 6.0) using a 2100 Retriever (Aptum Biologics, Rownhams, UK).

Rehydrated sections were blocked for 1 hr in 10% normal goat serum, 3% BSA, 0.5% Triton X-100 in PBS. Primary antibodies were diluted in blocking solution and incubated overnight at 4°C. Primary antibodies included the following: β-gal (1/1000, chicken polyclonal, ab9361, Abcam, Cambridge, UK; 1/750, rabbit polyclonal, MP 559761, MP Biomedicals, Illkirch, France), GFP (1/500, chick polyclonal, 1020, Aves Labs, Tigard, OR; 1/1000, chick polyclonal, 13970, Abcam), Islet1 (1/1000, mouse monoclonal IgG1, 40.2D6, DSHB), My32 (1/400, mouse monoclonal IgG1, M4276, Sigma), Myod (1/100, mouse monoclonal IgG1, M3512, Dako, Santa Clara, CA), Pax7 (1/20, mouse monoclonal IgG1, AB_528428), Tcf4 (1/150, rabbit polyclonal, C48H11, Cell Signalling, Leiden, Netherlands), Tnnt3 (1/200, monoclonal mouse IgG1, T6277, Sigma), Tomato (1/500, rabbit polyclonal, 632496, Clontech, Shiga, Japan; 1/250, chick polyclonal, 600-901-379, Rockland, Pottstown, PA) and Tuj1 (1/1000, monoclonal mouse IgG2a, BLE801202, Ozyme, Montigny-le-Bretonneux, France). After 3 rounds of 15 min washes in PBS 0.1% Tween 20, secondary antibodies were incubated in blocking solution 2 hr at RT together with 1 μg/ml Hoechst 33342 to visualize nuclei. Secondary antibodies consisted of Alexa 488, 555 or 633 goat anti-rabbit, anti-chicken or anti-mouse isotype specific (1/500, Jackson Immunoresearch, Cambridgeshire, UK). After 3 rounds of 15 min washes in PBS 0.1% Tween 20, slides were mounted in 70% glycerol for analysis.

For whole-mount immunofluorescence staining, embryos were dissected in PBS, fixed in 4% PFA, washed in PBS and stored at −20°C in 100% methanol. After rehydration in PBS, whole mount immunostainings were performed incubating the primary and secondary antibodies for 3 days each. Samples were cleared using benzyl alcohol/benzyl benzoate (BABB) clarification method (Yokomizo et al., 2012).

µCT scan analysis

For µCT scan analysis, the fetuses were treated with the phosphotungstic acid (PTA) contrast agent to well reveal skeletal and muscle structures. After dissection of the cervical region (including the mandible and scapular components, see Figure 2—figure supplement 1), the fetuses were fixed in 4% PFA for 24 hr at 4°C. Samples were then additionally fixed and dehydrated by exchanging the fixative and washing solutions to incrementally increasing ethanol concentrations (30%, 50%, 70%) with 2 days in each concentration to minimize the shrinkage of tissues. To start the contrasting procedure, the embryos were firstly incubated in ethanol-methanol-water mixture (4:4:3) for 1 hr and then transferred for 1 hr into 80% and 90% methanol solution. The staining procedure was then performed for 10 days in 90% methanol 1.5% PTA solution (changed every day with fresh solution) to ensure optimal penetration of the contrast agent. Staining was followed by rehydration of the samples in methanol-grade series (90%, 80%, 70%, 50% and 30%) and stored in sterile distilled water. The samples were placed in polypropylene tubes and embedded in 1% agarose gel to avoid movement artefacts during measurements. μCT scanning was performed using laboratory system GE Phoenix v|tome|x L 240 (GE Sensing and Inspection Technologies GmbH, Hamburg, Germany), equipped with a nanofocus X-ray tube with maximum power of 180 kV/15 W and a flat panel detector DXR250 with 2048 × 2048 pixel2, 200 × 200 µm2 pixel size. The μCT scan was carried out at 60 kV acceleration voltage and 200 µA tube current with voxel size of 5.7 μm for all samples. The beam was filtered by a 0.2 mm aluminium filter. The 2200 projections were taken over 360° with exposure time of 900 ms. The tomographic reconstructions were done using the software GE phoenix datos|x 2.0 (GE Sensing and Inspection Technologies GmbH) and data segmentations and visualizations were performed by combination of software VG Studio MAX 2.2 (Volume GraphicsGmbH, Heidelberg, Germany) and Avizo 7.1 (Thermo Fisher Scientific, Waltham, MA) according to (Tesařová et al., 2016). The interactive 3D PDFs were set up using 3D PDF maker software.

Imaging

Images were acquired using the following systems: a Zeiss Axio-plan equipped with an Apotome, a Zeiss stereo zoom microscope V16 or a Zeiss LSM 700 laser-scanning confocal microscope with ZEN software (Carl Zeiss, Oberkochen, Germany). For whole-mount rendering, acquired Z-stacks were 3D reconstructed using Imaris software. All images were assembled in Adobe Photoshop (Adobe Systems, San Jose, CA).

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
    The triple origin of skull in higher vertebrates: a study in quail-chick chimeras
    1. GF Couly
    2. PM Coltey
    3. NM Le Douarin
    (1993)
    Development 117:409–429.
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
    The Cranial Muscles of Vertebrates
    1. FH Edgeworth
    (1935)
    Cambridge University Press.
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
    Entwicklungsgeschichte des Kopfes und des Blutgefässsystemes von Ceratodus forsteri. II. Die epigenetischen Erwerbungen während der Stadien 39–48. Denkschr Med-Naturwiss Ges Jena
    1. A Greil
    (1913)
    Denkschriften Der Medicinisch-NaturwissenschaftlichenGesellschaft Zu Jena 9:935–1492.
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
    Rhombencephalic neural crest segmentation is preserved throughout craniofacial ontogeny
    1. G Köntges
    2. A Lumsden
    (1996)
    Development 122:3229–3242.
  36. 36
  37. 37
    The neural crest and evolution of the head/trunk interface in vertebrates
    1. S Kuratani
    2. R Kusakabe
    3. T Hirasawa
    (2018)
    Developmental Biology, 10.1016/j.ydbio.2018.01.017, 29408469.
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
    Handbuch Der Vergleichenden Anatomie Der Wirbeltiere
    1. W Lubosch
    (1938)
    Muskeln des Kopfes: Viscerale Muskulatur, Handbuch Der Vergleichenden Anatomie Der Wirbeltiere, Berlin, Urban & Schwarzenberg.
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
    Interactions and fates of avian craniofacial mesenchyme
    1. DM Noden
    (1988)
    Development 103:121–140.
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
  70. 70
  71. 71
  72. 72
    MesP1: a novel basic helix-loop-helix protein expressed in the nascent mesodermal cells during mouse gastrulation
    1. Y Saga
    2. N Hata
    3. S Kobayashi
    4. T Magnuson
    5. MF Seldin
    6. MM Taketo
    (1996)
    Development 122:2769–2778.
  73. 73
  74. 74
    MesP1 is expressed in the heart precursor cells and required for the formation of a single heart tube
    1. Y Saga
    2. S Miyagawa-Tomita
    3. A Takagi
    4. S Kitajima
    5. J Miyazaki
    6. T Inoue
    (1999)
    Development 126:3437–3447.
  75. 75
  76. 76
  77. 77
  78. 78
  79. 79
    Zebrafish Mesp family genes, mesp-a and mesp-b are segmentally expressed in the presomitic mesoderm, and Mesp-b confers the anterior identity to the developing somites
    1. A Sawada
    2. A Fritz
    3. YJ Jiang
    4. A Yamamoto
    5. K Yamasu
    6. A Kuroiwa
    7. Y Saga
    8. H Takeda
    (2000)
    Development 127:1691–1702.
  80. 80
  81. 81
  82. 82
    Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus
    1. S Srinivas
    2. T Watanabe
    3. CS Lin
    4. CM William
    5. Y Tanabe
    6. TM Jessell
    7. F Costantini
    (2001)
    BMC Developmental Biology 1:4.
  83. 83
  84. 84
  85. 85
  86. 86
  87. 87
  88. 88
  89. 89
  90. 90
  91. 91
  92. 92
  93. 93
  94. 94
  95. 95
  96. 96
  97. 97
  98. 98
  99. 99

Decision letter

  1. Clare Blackburn
    Reviewing Editor; MRC Centre for Regenerative Medicine, University of Edinburgh, United Kingdom
  2. Didier YR Stainier
    Senior Editor; Max Planck Institute for Heart and Lung Research, Germany
  3. Valerie Wilson
    Reviewer; University of Edinburgh, United Kingdom

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "Unique morphogenetic signatures define mammalian neck muscles and associated connective tissues" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Didier Stainier as the Senior Editor. The following individual involved in review of your submission has agreed to reveal his identity: Robert Knight (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This manuscript examines the complex lineage and genetic regulation of the muscles in the lower head and neck. To characterize lineage, Heude et al. use mouse lines expressing Cre under the control of relatively well-characterised regulatory elements expressed in subsets of the musculature (Islet1, Mesp1, Pax3), the neural crest (Wnt1) and trunk lateral mesoderm (Prx1). They also examine the phenotype of these head and neck muscles in mice lacking Tbx1 and Pax3. They focus on the cucullaris muscles that connect head and neck skeletal elements. This is interesting from an evolutionary perspective as an essential adaptation of gnathostomes allowing movement of the head relative to the body, and also because certain genetic syndromes differentially affect muscles in the head and neck. The origin of this muscle group has so far been controversial. The main findings of this work are that differential expression of Mesp1, Islet1 and Pax3 distinguish the head, neck and shoulder muscles, and that the cucullaris originates from branchiomeric (head) paraxial mesoderm, not trunk lateral mesoderm as previously suggested. Tbx1 is essential for subsets of head-derived muscle while Pax3 is required for anterior somite-derived muscle.

This is an excellent study of some complex developmental anatomy, presented in a generally beautifully clear and informative way. The results are significant from an evolutionary perspective and relevant to mechanisms that may underlie various musculoskeletal abnormalities in humans. The three-dimensional micro-CT scans presented as interactive PDFs are particularly informative and an ideal way to display complex data of this type – the sections can display only a small fraction of this information.

Essential revisions:

The authors are requested to clarify the following issues:

1) Figure 7A displays a hypothetical early origin of the head and neck muscles that is not presented in the data. It crucially shows bilateral 'fingers' of cranial paraxial mesoderm sitting immediately lateral to somites 1-3, with lateral mesoderm situated more lateral to this. Is there any reason why there are no stained embryos of this stage (or even at E9.5) in the series presented in Figure 1? Without this, it could be argued that the controversial question of paraxial versus lateral plate mesoderm origin of the cucullaris muscle has not been completely resolved. For example Prx1 may not be expressed in a small subset of lateral plate mesoderm corresponding to future cucullaris, or conversely Mesp1/Islet1 may secondarily be expressed in a cucullaris anlage that shares a lineage origin with trunk lateral plate mesoderm. If the possibility of a lateral mesoderm origin of cucullaris muscle cannot be ruled out by the current approach, this should be acknowledged in the text.

2) Please indicate how expression of the constitutive reporters of Mesp1, a subset of Mef2c, Pax3 and Islet1 expression is affected in the Tbx1 and Pax3 mutant mice that are described later in the paper, as this could allow expression of the reporters to be correlated with muscle phenotypes and so make the association between expression and phenotypic outcome clearer. Please also comment on whether the expression of Islet1 in somitic myogenic cells is dependent on Tbx1 or Pax3 during early stages of muscle formation.

3) The use of Pax3 transgenic line for tracking somitic mesoderm may complicate analysis as it is also expressed in CNCC. The authors note this in the third paragraph of the subsection “Distinct myogenic programs define neck muscle morphogenesis” as they observe a contribution of Pax3-expressing cells to the cucullaris but do not address this point further. As a central tenet of the paper is that there is no somite-derived contribution to the cucullaris this should be addressed by comparing Pax3 reporter expression to NCC markers, since an alternative interpretation is that there is a contribution of somite-derived cells to this muscle that migrate to the anlage after E10.5.

https://doi.org/10.7554/eLife.40179.028

Author response

Essential revisions:

The authors are requested to clarify the following issues:

1) Figure 7A displays a hypothetical early origin of the head and neck muscles that is not presented in the data. It crucially shows bilateral 'fingers' of cranial paraxial mesoderm sitting immediately lateral to somites 1-3, with lateral mesoderm situated more lateral to this. Is there any reason why there are no stained embryos of this stage (or even at E9.5) in the series presented in Figure 1? Without this, it could be argued that the controversial question of paraxial versus lateral plate mesoderm origin of the cucullaris muscle has not been completely resolved. For example Prx1 may not be expressed in a small subset of lateral plate mesoderm corresponding to future cucullaris, or conversely Mesp1/Islet1 may secondarily be expressed in a cucullaris anlage that shares a lineage origin with trunk lateral plate mesoderm. If the possibility of a lateral mesoderm origin of cucullaris muscle cannot be ruled out by the current approach, this should be acknowledged in the text.

The hypothetical early origin of postcranial CPM extension has been proposed on the basis of our results and data reported previously. Unfortunately, there are no specific markers or reporters available to date to trace the early CPM presented in Figure 7A. Our genetic lineage analysis at later stages, in combination with previous work involving innervation studies, grafting experiments, and retrospective clonal analyses indicate that the mesoderm adjacent to the 3 first somites correspond to a posterior branchiomeric population in the early tetrapod embryo (see also Couly et al., Development, 1992). The analysis of the Prx1 lineage, that appears to define the trunk LPM field, corroborates that the cucullaris muscles are clonally related to the posterior branchiomeric muscles as shown previously through retrospective clonal analysis (Lescroart et al., 2015). We have modified the Discussion to clarify this point (subsection “Branchiomeric origin of cucullaris-derived muscles”, first paragraph). Moreover, Mesp1 does not seem to be secondarily expressed in the cucullaris anlagen. We propose that Mesp1 activation in the nascent mesoderm during gastrulation is more likely defining both CPM and anterior somites.

2) Please indicate how expression of the constitutive reporters of Mesp1, a subset of Mef2c, Pax3 and Islet1 expression is affected in the Tbx1 and Pax3 mutant mice that are described later in the paper, as this could allow expression of the reporters to be correlated with muscle phenotypes and so make the association between expression and phenotypic outcome clearer. Please also comment on whether the expression of Islet1 in somitic myogenic cells is dependent on Tbx1 or Pax3 during early stages of muscle formation.

Expression of the constitutive reporters in Tbx1 and Pax3 mutants has not been analysed in the present study. However, based on lineage analyses and anatomical descriptions, we have defined the genetic embryological origin of the different head/neck/pectoral muscles analysed in mutant foetuses (Figures 5-6, see also our new Figure 7—figure supplement 1). Please find the revised Table 2 that now highlights the myogenic lineages used for the development of the muscle groups, correlating lineage analysis and phenotypic outcomes. Please refer to the colour codes used for the different myogenic programs reported in the main Figures (1, 2, 3, 6, 7), in Figure 7—figure supplement 1 and in the 3D interactive PDFs (Supplementary files 1-3).

To address Islet1 contribution to the somitic mesoderm, we have modified Figure 1—figure supplement 2C. The results now show that Islet1 expression is observed in Pax3-derived cells in the precursor of the cutaneous maximus and latissimus dorsi somitic muscles (clp), completing information reported in Table 1 (footnote **). We have now modified the Results section accordingly (subsection “Distinct myogenic programs define neck muscle morphogenesis”, third paragraph). Moreover, we describe in the Discussion that these muscles depend on a specific myogenic program implicating Pax3 and Islet1 genes and that they are absent in Pax3 mutants (subsection “A unique genetic program for somite-derived neck muscles”, third paragraph). In contrast, Tbx1 expression has never been reported during early somitic muscle development, indicating that clp formation is dependent on Islet1 and Pax3 but not on Tbx1.

3) The use of Pax3 transgenic line for tracking somitic mesoderm may complicate analysis as it is also expressed in CNCC. The authors note this in the third paragraph of the subsection “Distinct myogenic programs define neck muscle morphogenesis” as they observe a contribution of Pax3-expressing cells to the cucullaris but do not address this point further. As a central tenet of the paper is that there is no somite-derived contribution to the cucullaris this should be addressed by comparing Pax3 reporter expression to NCC markers, since an alternative interpretation is that there is a contribution of somite-derived cells to this muscle that migrate to the anlage after E10.5.

To address this point, we have added an analysis of the Pax3 and Mesp1 contributions to the perinatal trapezius at high magnification (Figure 3—figure supplement 4C-D). This new result demonstrates that the Pax3-derived cells in the trapezius correspond to the Tcf4-positive fibroblast population associated with myofibres, similar to that observed in the cranial masseter and tongue muscles, that are composed of NCC-derived connective tissue (Figure 3—figure supplement 2A-B’, Figure 3—figure supplement 4A-B, G). We have modified the Results section accordingly (subsection “Dual neural crest and mesodermal origins of neck connective tissues”, third paragraph). Taken together, our results exclude the possibility of a Pax3 lineage contribution to cucullaris myofibres.

https://doi.org/10.7554/eLife.40179.029

Article and author information

Author details

  1. Eglantine Heude

    1. Department of Developmental and Stem Cell Biology, Institut Pasteur, Paris, France
    2. CNRS UMR 3738, Paris, France
    Contribution
    Conceptualization, Validation, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared
  2. Marketa Tesarova

    Central European Institute of Technology, Brno University of Technology, Brno, Czech Republic
    Contribution
    Resources, Data curation, Formal analysis, Investigation, Visualization, Methodology
    Competing interests
    No competing interests declared
  3. Elizabeth M Sefton

    Department of Human Genetics, University of Utah, Salt Lake City, United States
    Contribution
    Validation, Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6481-612X
  4. Estelle Jullian

    Aix-Marseille Université, CNRS UMR 7288, IBDM, Marseille, France
    Contribution
    Validation, Investigation
    Competing interests
    No competing interests declared
  5. Noritaka Adachi

    Aix-Marseille Université, CNRS UMR 7288, IBDM, Marseille, France
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9482-8436
  6. Alexandre Grimaldi

    1. Department of Developmental and Stem Cell Biology, Institut Pasteur, Paris, France
    2. CNRS UMR 3738, Paris, France
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
  7. Tomas Zikmund

    Central European Institute of Technology, Brno University of Technology, Brno, Czech Republic
    Contribution
    Resources, Supervision
    Competing interests
    No competing interests declared
  8. Jozef Kaiser

    Central European Institute of Technology, Brno University of Technology, Brno, Czech Republic
    Contribution
    Resources, Supervision, Funding acquisition, Project administration
    Competing interests
    No competing interests declared
  9. Gabrielle Kardon

    Department of Human Genetics, University of Utah, Salt Lake City, United States
    Contribution
    Resources, Supervision, Funding acquisition, Project administration, Writing—review and editing
    Competing interests
    No competing interests declared
  10. Robert G Kelly

    Aix-Marseille Université, CNRS UMR 7288, IBDM, Marseille, France
    Contribution
    Resources, Supervision, Funding acquisition, Project administration, Writing—review and editing
    Competing interests
    No competing interests declared
  11. Shahragim Tajbakhsh

    1. Department of Developmental and Stem Cell Biology, Institut Pasteur, Paris, France
    2. CNRS UMR 3738, Paris, France
    Contribution
    Conceptualization, Resources, Supervision, Funding acquisition, Validation, Visualization, Methodology, Project administration, Writing—review and editing
    For correspondence
    shahragim.tajbakhsh@pasteur.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1809-7202

Funding

Institut Pasteur

  • Eglantine Heude
  • Alexandre Grimaldi
  • Shahragim Tajbakhsh

Agence Nationale de la Recherche

  • Eglantine Heude
  • Alexandre Grimaldi
  • Shahragim Tajbakhsh

Centre National de la Recherche Scientifique

  • Eglantine Heude
  • Estelle Jullian
  • Noritaka Adachi
  • Alexandre Grimaldi
  • Robert G Kelly
  • Shahragim Tajbakhsh

French Muscular Dystrophy Association

  • Eglantine Heude
  • Alexandre Grimaldi
  • Robert G Kelly
  • Shahragim Tajbakhsh

Central European Institute of Technology

  • Marketa Tesarova
  • Tomas Zikmund
  • Jozef Kaiser

March of Dimes Foundation

  • Elizabeth M Sefton
  • Gabrielle Kardon

National Institutes of Health

  • Elizabeth M Sefton
  • Gabrielle Kardon

Fondation pour la Recherche Médicale

  • Estelle Jullian
  • Noritaka Adachi
  • Robert G Kelly

Fondation Leducq

  • Estelle Jullian
  • Noritaka Adachi
  • Robert G Kelly

Yamada Science Foundation

  • Noritaka Adachi
  • Robert G Kelly

Bourses du Gouvernement Français

  • Noritaka Adachi

project CEITEC 2020

  • Marketa Tesarova
  • Tomas Zikmund
  • Jozef Kaiser

The Ministry of Education, Youth and Sports of the Czech Republic under the National Sustainability Programme II

  • Marketa Tesarova
  • Tomas Zikmund
  • Jozef Kaiser

CEITEC Nano Research Infrastructure (MEYS CR, 2016–2019)

  • Marketa Tesarova
  • Tomas Zikmund
  • Jozef Kaiser

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Drs. Claudio Cortes and Françoise Helmbacher for providing transgenic mice and Mirialys Gallardo for technical assistance. We also thank Cécile Mazur for illustrations.

Ethics

Animal experimentation: Animals were handled as per European Community guidelines and the ethics committee of the Institut Pasteur (CTEA) approved protocols. (APAFIS#6354-20160809l2028839)

Senior Editor

  1. Didier YR Stainier, Max Planck Institute for Heart and Lung Research, Germany

Reviewing Editor

  1. Clare Blackburn, MRC Centre for Regenerative Medicine, University of Edinburgh, United Kingdom

Reviewer

  1. Valerie Wilson, University of Edinburgh, United Kingdom

Publication history

  1. Received: July 17, 2018
  2. Accepted: November 17, 2018
  3. Accepted Manuscript published: November 19, 2018 (version 1)
  4. Version of Record published: December 28, 2018 (version 2)

Copyright

© 2018, Heude et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,036
    Page views
  • 226
    Downloads
  • 1
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Cell Biology
    2. Developmental Biology
    Chet Huan Oon, Kenneth E Prehoda
    Research Article Updated
    1. Developmental Biology
    2. Neuroscience
    Ryan J Kast et al.
    Research Article