Comprehensive Physiology offers downloadable PowerPoint presentations of figures for non-profit, educational use, provided the content is not modified and full credit is given to the author and publication.

Download a PowerPoint presentation of all images

	Figure 1. Embryological origin of the muscles of breathing in vertebrates. Panel A shows the development of cranial and trunk mesoderm in a conceptualized embryo. The embryo is shown in a superior view with the amnion cut away and the overlying ectoderm removed, except for the neural tube in the right panel. The primordial myoblasts are shown in black (cranial and pharyngeal mesoderm) and red (trunk mesoderm), which collectively contribute to the head, neck, and trunk musculature during the third and fourth week of development. This panel also shows relative positions of the prechordal plate and associated mesoderm and the cardiogenic mesoderm, as well as structures involved in the events of gastrulation, such as the primitive streak. The thick arrow pointing caudally (right side of Panel A) depicts the regression of the primitive streak as somites are being added and the notochord is lengthening toward the tail end. Note that the neural tube is forming simultaneously and lies over the notochord dorsoventrally. The notochord extends rostrally to approximately the prechordal plate. The otic placodes (yellow oval) contribute to the developing ear as a thickening of the ectoderm and are gross anatomical characters separating the head mesoderm (a.k.a. somitomeres) from the rostral somites. The pharyngeal mesoderm is shown in gray as it is forming pharyngeal arches from migratory myoblasts coursing ventrally from the cranial mesoderm (black arrows) and neural crest cells (not shown). (B) Lateral view of developing muscles from unsegmented cranial and segmented trunk mesoderm near embryonic week 4. Arrows indicate movements of myoblasts and cranial nerve associations from pharyngeal mesoderm and neural crest cell‐derived ectomesenchyme. (C) Cross section of the postgastrulation trilaminal disc near embryonic week 4. The approximate rostrocaudal plane is shown with broken black lines near the second somite in A. At this stage, the neural tube has begun to fold and will meet in the sagittal plane while the mesoderm is differentiating into region‐specific structures laterally. The dorsal aspect of the somite or dermomyotome (d, m) will produce the connective tissue of the dermis (d) and skeletal muscles (m), and the ventral somite or sclerotome (s) will produce most of the axial skeleton. The derivatives of the muscles and connective tissues of the head are a combination of cranial mesoderm and ectomesenchyme that create the pharyngeal arches (gray overlapping circles), as well as occipital somites (not shown). Human embryos have approximately 33 somite pairs. N = notochord; S = somite; NCC = neural crest cells; im = intermediate mesoderm; d = dermomyotome; m = myotome; s = sclerotome. Modified, with permission, from (72). See text for more detail.
	Figure 2. Example recordings of prenatal and perinatal breathing‐related neural activity from a developing zebra finch. Motor outflow is shown as rectified and moving time averaged (time constant = 200 ms) cranial nerve XI (spinal accessory) activity. Cranial nerve XI motor outflow is considered inspiratory as it innervates the cucullaris muscle in birds, which is thought to be homologous to the sternocleidomastoid and trapezius in humans. The top panel shows the first observable neural activity from cranial nerve XI at stage 22 or embryonic day 4 (E4) (310). Cranial nerve activity at this age is considered to be a generic large‐scale depolarization wave that spreads throughout the neuraxis. At this stage, most spinal and cranial nerves display synchronous discharges and a similar waveform. The middle panel shows a transitional period where the underlying chemical communication and morphological connectivity in the developing breathing network is being modified, although the mechanisms for these changes are unknown. The lower panel shows a cranial nerve XI recording from a hatchling bird (i.e., internally and externally pipped through the eggshell). This breathing‐related motor outflow achieves a high frequency and is critically dependent on AMPAergic neurotransmission (459).
	Figure 3. Muscles of the external nose. (A) The numbers on the muscles in the schematic correspond to the like‐numbered EMG recordings shown in in Panel B. The EMG recordings were obtained during voluntary nasal flaring. Panel C shows nasal muscle activity during quiet nasal breathing and immediately after exercise (“20 deep knee bends”), at which time subjects were instructed to breathe through the nose. EMG activities are expressed as a percentage of maximal activity, which was evoked by a maximal voluntary “grimace.” All figures adapted, with permission, from (60).
	Figure 4. Muscles of the head and neck. Panel A shows the location of the human masseter muscle and masseter EMG activity recorded in a human subject breathing against a large inspiratory resistance (see pharyngeal pressure trace). The dilator muscles of the nose were recorded simultaneously. Adapted, with permission, from (211). Panel B shows the locations of the sternocleidomastoid (SCM) and trapezius muscles, motor unit recordings from the SCM, and superimposed recordings of airway opening pressure and airflow (6). Trapezius muscle EMG recording and associated chest wall motion (inspiration up) are displayed from a healthy human subject who was asked to take deep breaths, from Daimon and Yamaguchi (93). Panel C is a schematic drawing of the lateral and medial pterygoid muscles. The associated recordings show EMG activity of the medial pterygoid in an awake human subject instructed to take deep inspirations; note the marked recruitment during inspiration, from Sauerland et al. (389). Panel D shows the location of the anterior, middle, and posterior scalene muscles and recordings of motor unit discharge during tidal breathing in a healthy male subject. Which of the three scalene muscles used for this recording was not indicated, but we presume the anterior head. Tracings show (from top to bottom): raw and integrated EMG, instantaneous motor unit firing rate, and chest wall motion (inspiration up). The inset to the right of the motor unit recording shows overlaid motor unit potentials from one of the identified motor units. Adapted, with permission, from Saboisky et al. (379).
	Figure 5. Muscles of the tongue. Panel A contains schematic diagrams of the human tongue muscles from an impressive anatomic evaluation by Sanders and Mu (383). In brief, the authors used images from the Visible Human Project to create three‐dimensional reconstructions, and also used various sagittal images to provide a detailed description of the muscles of the human tongue. The upper row shows the intrinsic muscles superior longitudinalis (SL), inferior ongitudinalis (IL), and the transversus and verticalis (T/V) muscles, whose fibers are tightly intermingled (see text). The bottom row shows the extrinsic muscles genioglossus (GG), styloglossus (SG), and hyoglossus (HG). In each diagram, the indicated muscle or muscles is darker than the whole tongue (gray). Two important features emphasized by the authors that have significant bearing on the interpretation of tongue muscle EMG recordings include: (i) the superior longitudinal (SL) muscle is the only unpaired muscle, and it also spans the length of the entire tongue; (ii) although the muscles are presented as separate structures, the fibers of individual muscles intermingle extensively with those from adjacent muscles. A particularly important example is that the genioglossus (GG) becomes the verticalis (V), and see blue highlighted region in Panel C) muscle in approximately the middle third of the tongue. Panel B is a midline saggital image of a human tongue, highlighting the superior longitudinalis (SL), the transversus/verticalis bundle (TV), the geniohyoid (GH), and two bellies of the genioglossus, including the oblique GG (oGG) and the horizontal GG belly (hGG). Note that the fiber bundles of the hGG are longer than those of the oGG (white lines) that the fiber density of all muscles is greatest anteriorally and is reduced at the tongue base. It is clear that the human tongue has considerable curvature, as opposed to much more linear tongues in most lower mammals, especially rodents. Panel C shows extrinsic and intrinsic fibers of the human tongue using in vivo imaging with tractography, and nicely highlights the unique fiber orientation of the verticalis (blue) (158). The transversus muscle is not easily seen in sagittal section, so we have included a transverse section that nicely delineates the muscle (158). Panel D shows the human palatoglossus muscle (yellow arrow), which can also be seen in Panel C (orange fibers). The red and blue arrows in panel D identify the soft palate and hard palate, respectively. Adapted, with permission, from (257). Panel E is a representative record of genioglossus (“protrudor”) and hyoglossus (“retractor”) EMG recordings during 4 min of combined hypercapnia and hypoxia in a healthy human subject (288). Note that for both muscles the unprocessed EMG and the rectified and integrated EMG (iEMG) are shown. The upper tracing is expired airflow, and close examination shows that the tongue muscles discharge during inspiration. Panel F shows EMG recordings of genioglossus (GG) and styloglossus (SG) muscles in response to peripheral chemoreceptor stimulation with close arterial injection of NaCN in an anesthetized dog (483). (G) Recordings from human palatoglossal muscle moving time‐averaged EMG (top trace) and inspiratory flow (433). (H) The tracings represent intrinsic superior longitudinalis EMG activity and esophageal pressure (Pes) in a supine, urethane‐anesthetized rat (20). The left‐hand traces were obtained during quiet breathing and those on the right during hypercapnia. Clear inspiration‐related discharge can be seen in both instances. Thus, inspiration‐related activity has been observed in every mammalian tongue muscle that has been studied. See text for detailed discussion.
	Figure 6. Muscles of the tongue. Panel A contains schematic diagrams of the human tongue muscles from an impressive anatomic evaluation by Sanders and Mu (383). In brief, the authors used images from the Visible Human Project to create three‐dimensional reconstructions, and also used various sagittal images to provide a detailed description of the muscles of the human tongue. The upper row shows the intrinsic muscles superior longitudinalis (SL), inferior ongitudinalis (IL), and the transversus and verticalis (T/V) muscles, whose fibers are tightly intermingled (see text). The bottom row shows the extrinsic muscles genioglossus (GG), styloglossus (SG), and hyoglossus (HG). In each diagram, the indicated muscle or muscles is darker than the whole tongue (gray). Two important features emphasized by the authors that have significant bearing on the interpretation of tongue muscle EMG recordings include: (i) the superior longitudinal (SL) muscle is the only unpaired muscle, and it also spans the length of the entire tongue; (ii) although the muscles are presented as separate structures, the fibers of individual muscles intermingle extensively with those from adjacent muscles. A particularly important example is that the genioglossus (GG) becomes the verticalis (V), and see blue highlighted region in Panel C) muscle in approximately the middle third of the tongue. Panel B is a midline saggital image of a human tongue, highlighting the superior longitudinalis (SL), the transversus/verticalis bundle (TV), the geniohyoid (GH), and two bellies of the genioglossus, including the oblique GG (oGG) and the horizontal GG belly (hGG). Note that the fiber bundles of the hGG are longer than those of the oGG (white lines) that the fiber density of all muscles is greatest anteriorally and is reduced at the tongue base. It is clear that the human tongue has considerable curvature, as opposed to much more linear tongues in most lower mammals, especially rodents. Panel C shows extrinsic and intrinsic fibers of the human tongue using in vivo imaging with tractography, and nicely highlights the unique fiber orientation of the verticalis (blue) (158). The transversus muscle is not easily seen in sagittal section, so we have included a transverse section that nicely delineates the muscle (158). Panel D shows the human palatoglossus muscle (yellow arrow), which can also be seen in Panel C (orange fibers). The red and blue arrows in panel D identify the soft palate and hard palate, respectively. Adapted, with permission, from (257). Panel E is a representative record of genioglossus (“protrudor”) and hyoglossus (“retractor”) EMG recordings during 4 min of combined hypercapnia and hypoxia in a healthy human subject (288). Note that for both muscles the unprocessed EMG and the rectified and integrated EMG (iEMG) are shown. The upper tracing is expired airflow, and close examination shows that the tongue muscles discharge during inspiration. Panel F shows EMG recordings of genioglossus (GG) and styloglossus (SG) muscles in response to peripheral chemoreceptor stimulation with close arterial injection of NaCN in an anesthetized dog (483). (G) Recordings from human palatoglossal muscle moving time‐averaged EMG (top trace) and inspiratory flow (433). (H) The tracings represent intrinsic superior longitudinalis EMG activity and esophageal pressure (Pes) in a supine, urethane‐anesthetized rat (20). The left‐hand traces were obtained during quiet breathing and those on the right during hypercapnia. Clear inspiration‐related discharge can be seen in both instances. Thus, inspiration‐related activity has been observed in every mammalian tongue muscle that has been studied. See text for detailed discussion.
	Figure 7. Muscles of the soft palate. Panel A shows the roof of the mouth, beginning at the hard palate anteriorly, and extending dorsally to show the soft palate, which has been transected in the schematic diagram. The main palatal muscles, the tensor and veli palatini and musculus uvulae, are shown. Panel B shows moving time averaged EMG activities of musculus uvulae (here referred to as the palitinus), levator and tensor palatini muscles, and the diaphragm in an anesthetized, tracheotomized dog are shown. When breathing through the tracheotomy (left of the vertical arrow), there was a low level of inspiration‐related activity in all palatal muscles, and both phasic and tonic activities were recruited after the transition to nasal breathing, suggesting that receptors responding to airway negative pressure activate palatal muscle motoneurons (448). Panel C shows airflow (inspiration downward) and moving time‐averaged EMG activities (MTA) of the genioglossus (GG), alae nasi (AN), and tensor veli palatini muscles (TVP) during quiet breathing in a healthy human subject; note that there is consistent palatal muscle activity under these conditions (467). (D) Representative unprocessed and moving time‐averaged EMG activity (MTA EMG) of the levator veli palatini (LVP) in a healthy, awake human subject. The bottom trace is the airflow recording, with inspiration shown as a downward deflection (270). Bursts of activity are observed during both inspiratory and expiratory phases of the respiratory cycle.
	Figure 8. The suprahyoid muscles. In Panel A, a schematic anatomical drawing illustrates the suprahyoid muscles, including both anterior and posterior bellies of the digastric muscles. In Panel B, respiratory flow, EMG of the diaphragm (diaEMG) and digastric muscles (dEMG) are shown during the transition from normoxia to anoxia in neonatal rats. Rectified and integrated versions (int) of the EMG activities are also shown. Note the recruitment of the digastric near the end of the apnea (defined as the absence of diaEMG and flow), and continued activation of the muscle during the autoresuscitation phase prior to a terminal apnea (380). Panel C shows the moving‐time averaged geniohyoid EMG activity, which discharged in phase with inspiratory airflow, in healthy human subjects during wakefulness (top two sets of tracings), and NREM and REM sleep (third and fourth sets of tracings). Muscle activity was increased during NREM sleep, but declined to waking levels or below during REM sleep. Also, note that the respiration‐related activity of the geniohyoid was always much less than that observed during a swallow, as indicated in the second set of tracings (472). Panel D shows EMG activities of the sternocleidomastoid and mylohyoid muscles, which discharge in phase with rib cage expansion during quiet breathing in high tetraplegics, consistent with roles for these muscle in generating inspiration (106). Panel E shows EMG activities from the middle pharyngeal constrictors, stylohyoid, and diaphragm, together with the inspired tidal volume recording in an awake goat. The left‐hand panel was recorded during quiet breathing, and the right‐hand panel was recorded during a hypercapnic challenge with 6% inspired CO2. Note the increase in stylohyoid activity during hypercapnia, and the pattern of expiratory discharges under both conditions (324). However, as discussed in text, stylohyoid activity is inspiratory in the anesthetized rabbit (373).
	Figure 9. The infrahyoid muscles. Panel A is a schematic diagram showing the infrahyoid muscles. Panel B shows inspiratory‐related EMG activities of the genioglossus and sternothyroid muscle in the anesthetized rabbit; note the marked recruitment of these muscles during a brief nasal occlusion (horizontal bar) (372). Since EMG activity increased before there was time for significant changes in blood gases or pH, the increased activity is likely due reflex modulation of muscle activity by negative upper airway pressure or the removal of lung volume feedback from pulmonary stretch receptors. In Panel C, the moving time averaged EMG activities of thyrohyoid, sternohyoid, and diaphragm muscles in an anesthetized dog are shown before and after bilateral vagotomy (447). The thyrohyoid activity was largely expiratory before vagotomy, but inspiratory after vagotomy; this pattern was not always observed—some animals had consistent inspiratory activation both before and after vagotomy. The stylohyoid muscle discharges during the inspiratory phase under both conditions, although the activity of this muscle was present in only two of six dogs before vagotomy, but in all dogs after vagotomy. Panel D shows tracheal pressure, airflow, and the moving time averaged EMG of the omohyoid muscle (Omo EMG) in an anesthetized monkey. There is minimal activity in quiet breathing, but tonic and phasic inspiratory activity is recruited during tracheal occlusion, as denoted by the absence of flow and the large swings in tracheal pressure (136).
	Figure 10. The pharyngeal muscles. In Panel A, the sagittal schematic shows all the pharyngeal muscles except the salpingopharyngeus, which is shown in the anterior view in Panel B. In Panel C, from the top down, representative traces of airflow (inspiration downward) and moving‐time averaged EMG activities (MTA EMG) of the superior (SPC), middle (MPC), and inferior (IPC) pharyngeal constrictor muscles are shown (263). All three muscles were active during expiration, though only at high respiratory drive evoked by raising the inspired CO2 levels to 8% and then 9%. Panel D shows inspiratory‐related activity of the palatopharyngeus muscle in an awake, upright, human subject with obstructive sleep apnea (307). The activity of the levator palatini and a chest wall motion recording (respitrace, inspiration upward) are also shown. The EMG activities are moving‐time averages. Panel E shows inspiratory airflow and moving time averaged EMG activities of thyropharyngeus (TP), stylopharyngeus (SP), and the diaphragm (Dia) in an awake goat. Note that stylopharyngeus activity occurs during inspiration in eupnea and is abolished during a spontaneous apnea, which is defined as the absence of airflow and diaphragm EMG activity (144).
	Figure 11. The laryngeal muscles. For anatomical orientation, the laryngeal cartilages are shown from anterior (left) and posterior vantage points in Panel A, a superior view in Panel B, and a dorsal view in Panel C. Panel D shows, from top down, tidal volume, airflow (inspiration down), and raw and moving‐time averaged EMG activities of the posterior cricoarytenoid muscle in healthy human subjects. As with the thyroarytenoid, posterior cricoarytenoid activity during quiet breathing is brisk, and there is marked recruitment during progressive hypercapnia, as evoked with 7% and 9% CO2 in the inspired gas (260). Panel E shows EMG recordings of the levator veli palatini (LVP), the lateral cricoarytenoid (LCA), and rib cage expansion in an anesthetized dog during the transition from eupnea to progressive hypercapnia. Note that the LCA was active during expiration in eupnea, but was markedly reduced in hypercapnia—the opposite of the response of the LVP to hypercapnia (251). Panel F shows the activity of the arytenoideus muscle in a healthy adult human subject that transitioned spontaneously from eupnea to rapid, shallow breathing. The unprocessed and moving time averaged EMG recordings are shown, and the upper two traces are airflow (expiration upward) and tidal volume, respectively. The muscle is active during eupnea, and discharges in phase with expiration (262). Drive to the muscle was reduced during the period of rapid shallow breathing, and expiratory glottic area increased, consistent with a drop in expiratory flow resistance. As explained in the text, it was not possible to determine whether the electrodes were located in the transverse or the oblique arytenoid, as the muscles are very small and closely apposed to one another. Panel G shows representative recordings of unprocessed and moving time averaged EMGs of the lateral cricoarytenoid (CT) muscle and tidal volume (inspiration upward) in a human subject studied during progressive hyperoxic hypercapnia. Segments of the experiment at baseline and at two levels of increased ventilatory drive are shown. The rate of pulmonary ventilation in the three segments was 8.0, 18.1, and 26.5 L/min. Note that the activity of the lateral cricoarytenoid occurs during inspiration in eupnea, and that both peak phasic and tonic expiratory activities increase in hypercapnia (469). Panel H shows unprocessed and moving time averaged thyroarytenoid muscle EMG recordings and airflow (inspiration down) in a young, healthy human subject breathing quietly (“nonloaded”) and against three successively larger inspiratory resistive loads (232). Note the consistent, inspiration‐related activity during quiet breathing and progressive recruitment as the resistive load increased.
	Figure 12. The laryngeal muscles. For anatomical orientation, the laryngeal cartilages are shown from anterior (left) and posterior vantage points in Panel A, a superior view in Panel B, and a dorsal view in Panel C. Panel D shows, from top down, tidal volume, airflow (inspiration down), and raw and moving‐time averaged EMG activities of the posterior cricoarytenoid muscle in healthy human subjects. As with the thyroarytenoid, posterior cricoarytenoid activity during quiet breathing is brisk, and there is marked recruitment during progressive hypercapnia, as evoked with 7% and 9% CO2 in the inspired gas (260). Panel E shows EMG recordings of the levator veli palatini (LVP), the lateral cricoarytenoid (LCA), and rib cage expansion in an anesthetized dog during the transition from eupnea to progressive hypercapnia. Note that the LCA was active during expiration in eupnea, but was markedly reduced in hypercapnia—the opposite of the response of the LVP to hypercapnia (251). Panel F shows the activity of the arytenoideus muscle in a healthy adult human subject that transitioned spontaneously from eupnea to rapid, shallow breathing. The unprocessed and moving time averaged EMG recordings are shown, and the upper two traces are airflow (expiration upward) and tidal volume, respectively. The muscle is active during eupnea, and discharges in phase with expiration (262). Drive to the muscle was reduced during the period of rapid shallow breathing, and expiratory glottic area increased, consistent with a drop in expiratory flow resistance. As explained in the text, it was not possible to determine whether the electrodes were located in the transverse or the oblique arytenoid, as the muscles are very small and closely apposed to one another. Panel G shows representative recordings of unprocessed and moving time averaged EMGs of the lateral cricoarytenoid (CT) muscle and tidal volume (inspiration upward) in a human subject studied during progressive hyperoxic hypercapnia. Segments of the experiment at baseline and at two levels of increased ventilatory drive are shown. The rate of pulmonary ventilation in the three segments was 8.0, 18.1, and 26.5 L/min. Note that the activity of the lateral cricoarytenoid occurs during inspiration in eupnea, and that both peak phasic and tonic expiratory activities increase in hypercapnia (469). Panel H shows unprocessed and moving time averaged thyroarytenoid muscle EMG recordings and airflow (inspiration down) in a young, healthy human subject breathing quietly (“nonloaded”) and against three successively larger inspiratory resistive loads (232). Note the consistent, inspiration‐related activity during quiet breathing and progressive recruitment as the resistive load increased.
	Figure 13. The laryngeal muscles. For anatomical orientation, the laryngeal cartilages are shown from anterior (left) and posterior vantage points in Panel A, a superior view in Panel B, and a dorsal view in Panel C. Panel D shows, from top down, tidal volume, airflow (inspiration down), and raw and moving‐time averaged EMG activities of the posterior cricoarytenoid muscle in healthy human subjects. As with the thyroarytenoid, posterior cricoarytenoid activity during quiet breathing is brisk, and there is marked recruitment during progressive hypercapnia, as evoked with 7% and 9% CO2 in the inspired gas (260). Panel E shows EMG recordings of the levator veli palatini (LVP), the lateral cricoarytenoid (LCA), and rib cage expansion in an anesthetized dog during the transition from eupnea to progressive hypercapnia. Note that the LCA was active during expiration in eupnea, but was markedly reduced in hypercapnia—the opposite of the response of the LVP to hypercapnia (251). Panel F shows the activity of the arytenoideus muscle in a healthy adult human subject that transitioned spontaneously from eupnea to rapid, shallow breathing. The unprocessed and moving time averaged EMG recordings are shown, and the upper two traces are airflow (expiration upward) and tidal volume, respectively. The muscle is active during eupnea, and discharges in phase with expiration (262). Drive to the muscle was reduced during the period of rapid shallow breathing, and expiratory glottic area increased, consistent with a drop in expiratory flow resistance. As explained in the text, it was not possible to determine whether the electrodes were located in the transverse or the oblique arytenoid, as the muscles are very small and closely apposed to one another. Panel G shows representative recordings of unprocessed and moving time averaged EMGs of the lateral cricoarytenoid (CT) muscle and tidal volume (inspiration upward) in a human subject studied during progressive hyperoxic hypercapnia. Segments of the experiment at baseline and at two levels of increased ventilatory drive are shown. The rate of pulmonary ventilation in the three segments was 8.0, 18.1, and 26.5 L/min. Note that the activity of the lateral cricoarytenoid occurs during inspiration in eupnea, and that both peak phasic and tonic expiratory activities increase in hypercapnia (469). Panel H shows unprocessed and moving time averaged thyroarytenoid muscle EMG recordings and airflow (inspiration down) in a young, healthy human subject breathing quietly (“nonloaded”) and against three successively larger inspiratory resistive loads (232). Note the consistent, inspiration‐related activity during quiet breathing and progressive recruitment as the resistive load increased.
	Figure 14. The thoracic muscles. Panels A and B are schematic diagrams showing those thoracic muscles that have documented respiration‐related activity. Panel C shows airflow (inspiration upward), ribcage movement (inspiration up), and the unprocessed and moving time averaged EMG signals of the dorsal external intercostal muscles from the third intercostal space during quiet breathing in a healthy human subject. The activity is clearly in phase with inspiration (109). Panel D shows intramuscular EMG recordings from the third, right intercostal space. The upper tracings show inspiratory activity recorded from the superficial layer, and the bottom traces show expiratory activity in a deeper layer. The airflow and tidal volume recordings are superimposed on each set of EMG recordings (arrows). Inspiration is upward for both flow and volume. This experiment indicates that expiratory activity predominates in the deep layers even in the rostral intercostal spaces, Panel F (434). Panel E shows tidal volume (inspiration upward), the moving time average EMG of an external intercostal muscle from the sixth interspace, and the moving time average and unprocessed EMG of the internal intercostal muscle from the ninth intercostal space, also on the left‐side, in a healthy human subject. Note that forced exhalation of the expiratory reserve volume (ERV) causes marked recruitment of the internal intercostal muscle. While the activity of both muscles was absent or minimal in eupnea, when the subject held his breath at end‐expiratory lung volume while rotating to the left, tonic EMG activity was evoked in the internal intercostal muscle and phasic expiratory activity was accentuated. This suggests that increasing synaptic input to the internal intercostal motoneuron pool by twisting brought the neurons closer to firing threshold. Adapted, with permission, from (367). Panel F is a schematic of the right side of the chest showing where intercostal activity was most consistently found during quiet breathing (dotted area, inspiratory; hatched area, expiratory), with the number of observations made at each location indicated. The two points marked “D” indicate the site of electrode placement for diaphragm recordings. In general, the upper parasternal intercostals were active only during inspiration, while the caudal and more lateral spaces were active during expiration (434). This observation, made in 1960, was largely confirmed in studies by De Troyer and colleagues, which are summarized in Panel G. These data show the relation between airway opening pressure and the activities of external, internal, and parasternal intercostal muscles recorded from four different intercostal spaces (110). Note that parasternal muscles discharge only in inspiration; external intercostals in in the rostral spaces are active during inspiration, but in the caudal spaces (T8 and below) they discharge during the expiratory phase. Internal intercostal muscles are consistently active in the expiratory phase. The recordings in Panel H show the anteroposterior diameter of the abdomen (increase upward, indicating inspiration) and EMG recordings of the triangularis sterni (or transversus thoracis), the external oblique abdominal muscle, and the “deeper abdominal muscle layer” from a healthy 66‐year‐old subject studied in the standing position are displayed. In this subject, all three muscles were phasically active during expiration (143). Panel I shows (from top down) representative recordings of mouth pressure, inspiratory flow, tidal volume, and the moving‐time averaged EMG of the pectoralis major muscle in a young healthy subject. In the left‐hand panels, the subject was asked to make an inspiratory effort at 80% vital capacity against an occluded airway, while in the right‐hand panels the subject breathed against a large inspiratory resistance (66) (see text for more details). Panel J shows representative surface EMG tracings from the right sixth intercostal space and the right serratus anterior muscles in a young, healthy, human subject instructed to take deep breaths so that the inspired volume approached the inspiratory capacity. Inspiration is represented by a downward deflection in the airflow signal and an upward deflection in the tidal volume trace (356).
	Figure 15. The thoracic muscles. Panels A and B are schematic diagrams showing those thoracic muscles that have documented respiration‐related activity. Panel C shows airflow (inspiration upward), ribcage movement (inspiration up), and the unprocessed and moving time averaged EMG signals of the dorsal external intercostal muscles from the third intercostal space during quiet breathing in a healthy human subject. The activity is clearly in phase with inspiration (109). Panel D shows intramuscular EMG recordings from the third, right intercostal space. The upper tracings show inspiratory activity recorded from the superficial layer, and the bottom traces show expiratory activity in a deeper layer. The airflow and tidal volume recordings are superimposed on each set of EMG recordings (arrows). Inspiration is upward for both flow and volume. This experiment indicates that expiratory activity predominates in the deep layers even in the rostral intercostal spaces, Panel F (434). Panel E shows tidal volume (inspiration upward), the moving time average EMG of an external intercostal muscle from the sixth interspace, and the moving time average and unprocessed EMG of the internal intercostal muscle from the ninth intercostal space, also on the left‐side, in a healthy human subject. Note that forced exhalation of the expiratory reserve volume (ERV) causes marked recruitment of the internal intercostal muscle. While the activity of both muscles was absent or minimal in eupnea, when the subject held his breath at end‐expiratory lung volume while rotating to the left, tonic EMG activity was evoked in the internal intercostal muscle and phasic expiratory activity was accentuated. This suggests that increasing synaptic input to the internal intercostal motoneuron pool by twisting brought the neurons closer to firing threshold. Adapted, with permission, from (367). Panel F is a schematic of the right side of the chest showing where intercostal activity was most consistently found during quiet breathing (dotted area, inspiratory; hatched area, expiratory), with the number of observations made at each location indicated. The two points marked “D” indicate the site of electrode placement for diaphragm recordings. In general, the upper parasternal intercostals were active only during inspiration, while the caudal and more lateral spaces were active during expiration (434). This observation, made in 1960, was largely confirmed in studies by De Troyer and colleagues, which are summarized in Panel G. These data show the relation between airway opening pressure and the activities of external, internal, and parasternal intercostal muscles recorded from four different intercostal spaces (110). Note that parasternal muscles discharge only in inspiration; external intercostals in in the rostral spaces are active during inspiration, but in the caudal spaces (T8 and below) they discharge during the expiratory phase. Internal intercostal muscles are consistently active in the expiratory phase. The recordings in Panel H show the anteroposterior diameter of the abdomen (increase upward, indicating inspiration) and EMG recordings of the triangularis sterni (or transversus thoracis), the external oblique abdominal muscle, and the “deeper abdominal muscle layer” from a healthy 66‐year‐old subject studied in the standing position are displayed. In this subject, all three muscles were phasically active during expiration (143). Panel I shows (from top down) representative recordings of mouth pressure, inspiratory flow, tidal volume, and the moving‐time averaged EMG of the pectoralis major muscle in a young healthy subject. In the left‐hand panels, the subject was asked to make an inspiratory effort at 80% vital capacity against an occluded airway, while in the right‐hand panels the subject breathed against a large inspiratory resistance (66) (see text for more details). Panel J shows representative surface EMG tracings from the right sixth intercostal space and the right serratus anterior muscles in a young, healthy, human subject instructed to take deep breaths so that the inspired volume approached the inspiratory capacity. Inspiration is represented by a downward deflection in the airflow signal and an upward deflection in the tidal volume trace (356).
	Figure 16. The abdominal muscles. Panels A and C illustrate the diaphragm and abdominal muscles, respectively. Panel B shows the diaphragm EMG (top trace), the moving time averaged EMG, and a raster plot show the discharge frequency of the large motor unit that is visible in the EMG recording. The bottom trace is tidal volume and inspiration is upward. Note that the commonly observed inspiratory, augmenting discharge pattern of the diaphragm EMG is the result of a rapid rise in discharge frequency at inspiratory onset, followed by a plateau near the middle and end of the inspiratory period (379). Panel D shows three sets of recordings, and in each set (from top down), lung volume (inspiration upward), the anterior‐posterior dimensions of the abdomen, and EMG recordings from rectus abdominis (RA), external oblique (EO), and transversus abdominis (TA) muscles. Recordings were made in a young healthy male subject studied while sitting upright. The upper, middle, and lower sets of tracings were recorded during quiet breathing (end‐tidal CO2 = 42 mmHg), and two levels of hypercapnia (end‐tidal CO2 = 46 and 50 mmHg). Note that only the transversus was active during quiet breathing, and the other two muscles were not recruited until the end‐tidal CO2 was 50 mmHg (105). Panel E shows EMG recordings from each of the four anterior abdominal muscles, during severe hypercapnia (end‐tidal CO2 = 63 mmHg) in a young, healthy supine subject. Inspiration is represented by upward deflections in both flow and volume recordings (2). Panel F shows intramuscular EMG recordings from the rectus abdominis (RA) and external oblique abdominal muscles, and mouth pressure at rest and during progressive intensity, cycle ergometer exercise in a young, fit cyclist. In this study, the EO was recruited in light exercise, but the RA was recruited only during very heavy exercise. The EMG activities obtained during a maximal voluntary expiration against an occluded airway are shown in the bottom of the panel (3). Panel E shows representative EMG recordings of the quadratus lumborum (upper trace) and the diaphragm (lower trace) from an anesthetized rabbit during quiet breathing. The numbers represent different phases of the respiratory cycle: from left to right, expiration (5), preexpiration (4), end of inspiration (3), inspiration (2), and preinspiration (1). Activity is almost exclusively inspiratory, indicating that the quadratus lumborum is an accessory inspiratory muscle.
	Figure 17. The back muscles. Panels A and B illustrate the superficial (A) and deep back muscles (B), as described in the text. Panel C shows lung volume and intercostal muscle EMG recordings in an anesthetized, supine dog breathing quietly. Note that the levator costae, as well as the more commonly studied intercostal muscles, are phasically active during inspiration; adapted, with permission, from (108). Panel D shows (from top down) representative recordings of mouth pressure, inspiratory flow, tidal volume, and the moving‐time averaged EMG of an erector spinae muscle, the latissimus dorsi, and the trapezius in a young, healthy subject. In the left‐hand panels, the subject was asked to make an inspiratory effort at 80% vital capacity against an occluded airway, while in the right‐hand panels the subject breathed against a large inspiratory resistance (66). Activity was inspiratory in all three muscles. Panel E shows simultaneous recordings from intercostal and iliocostalis lumborum muscles in a male subject with respiratory myoclonus. Note that both muscles were activated during inspiration. The small, high‐frequency bursts observed in the expiratory period (arrows) confirm the myoclonus (237).

Teaching Material

J. Q. Pilarski, J. C. Leiter, R. F. Fregosi. Muscles of Breathing: Development, Function, and Patterns of Activation. Compr Physiol 9: 2019, 1023-1078.

Didactic Synopsis

Major Teaching Points:

1. Before brainstem respiratory neurons make contact with developing respiratory muscle motor neurons, the CNS generates rhythmic spontaneous neural activity (rSNA), which may underpin development of the circuits responsible for rhythmic activation of the motor neurons, and hence the breathing muscles.
2. In fully developed humans, there are approximately 63 muscles (most paired bilaterally) with a documented role in breathing, representing roughly 20% of the 320 pairs of human skeletal muscles.
3. Contraction of breathing muscles generates force, leading to changes in the size of the rib cage, lungs and upper airway.
4. Though breathing occurs with little awareness, it involves highly complex and poorly understood neural circuits that are largely located in the medulla oblongata.
5. The recruitment, timing, and intensity of breathing muscle contractions are adjusted automatically in activities ranging from sleep to maximal exercise.
6. Studying the development, function and neural control of the breathing muscles forms the basis for understanding breathing disorders.

Didactic Legends

The figures—in a freely downloadable PowerPoint format—can be found on the Images tab along with the formal legends published in the article. The following legends to the same figures are written to be useful for teaching.

Figure 1 This figure illustrates the embryonic development of vertebrate respiratory muscles. The primitive myoblasts are shown in black and red (Panel A), and they generate muscles of the head and neck, and trunk, respectively. Panel B is a lateral (or sagittal) view of developing muscles arising from unsegmented cranial and segmented trunk mesoderm around embryonic week 4. Arrows indicated movements of myoblasts and cranial nerves. Panel C shows the post-gastrulation trilaminal disc in cross section, which appears at about embryonic week 4; the name comes from our understanding that the embryo exists as three different germ layers - the ectoderm, the mesoderm and the endoderm, with each layer stacked on top of the other (hence, three-layered, or “trilaminar”). At this stage of development, it is clear that the embryo can be divided into right and left halves, as well as cranial and caudal ends.

Figure 2 This figure shows breathing related nerve activity at different stages of embryonic development in the zebra finch. All recordings represent motor activity recorded from cranial nerve XI, the spinal accessory nerve. In the finch, this nerve innervates the cucullaris muscle, which is thought to stiffen the neck musculature similar to the sternocleidomastoid and trapezius in humans. The top panel shows that activity is first observed on day 4 (E4). It is believed that the activity at this stage represents a widespread depolarization that flows throughout the nervous system. The middle panel shows a transitional period where the neurons underlying automatic breathing are beginning to make the appropriate connections. The bottom panel shows recordings from a hatchling bird (i.e., internally and externally pipped through the eggshell), which is developmentally similar to a newborn mammal. Here we can see an immature but functional breathing-related motor output that discharges at high frequency and is critically dependent on the excitatory neurotransmitter glutamate.

Figure 3 Most people do not think about nasal muscles, but they play an important role in facial expression, and breathing. Mammals, including humans, breathe through the nose at rest and even during low intensity exercise. Combined nose/mouth breathing commences when the pulmonary ventilation rate is about 30 L/min. Since the external nares (i.e., the “nostrils”) are composed of soft tissue they are collapsible. As a result, with intense contraction of the diaphragm and other chest wall inspiratory muscles, the negative airway pressure can suck the nasal airway closed. But activation of the nasal muscles stiffens the external nares and thus guards against collapse. This figure shows that most of the nasal muscles are active after exercise; many other studies show that the muscles are active during exercise, as discussed in the text.

Figure 4 Although there are many muscles that envelop the head and neck, those that have a documented role in breathing are shown in this figure. Muscles of the head, such as the masseter and pterygoid muscles, are activated only when the respiratory drive is very high, such as in gasping where the mouth is opened widely to facilitate airflow. In healthy subjects, the neck muscles, including the trapezius, scalenes and sternocleidomastoids are recruited during heavy exercise or with severe hypoxia as occurs at very high altitude. However, these muscles may be active even during quiet breathing in people with lung disease.

Figure 5 This figure identifies human tongue muscles, as well as the breathing-related EMG activity of most of them. The tongue is capable of moving in many directions, and its weight means that gravity can also change tongue position. The dorsal surface of the tongue makes up part of the anterior pharyngeal wall. As a result, when one assumes the supine posture gravity can move the tongue backwards, which brings the “anterior wall” (i.e., the tongue) close to or in direct contact with the posterior pharyngeal wall. This leads to narrowing or closure of the pharyngeal airway; narrowing results in an increase in airflow resistance, whereas closure of the airway results in an obstructive apnea; this is what occurs in people with obstructive sleep apnea. To prevent the tongue from moving backwards and blocking the airway, the appropriate tongue muscles must be activated so that the tongue moves forward, downwards and becomes stiffer. All of the seven tongue muscles have been shown to have breathing-related activity, depending on prevailing conditions. However, because most of the muscles are difficult to access, the genioglossus has been studied extensively in humans, while the other muscles have not. As a result, most of what we know about the other muscles derives from animal models.

Figure 6 The soft palate plays a key role in “sealing” the nasopharynx during swallowing. But this action also allows the palate to direct more or less airflow through the nose or mouth. This leads to the concept that the palatal muscles control “oro-nasal airflow partitioning”. Breathing-related activation of all three palatal muscles has been observed in humans and animal models, though activation of the muscles it typically observed only when respiratory drive is high (e.g., during exercise).

Figure 7 The suprahyoid muscles originate above the hyoid bone, and insert into it. The main function of this muscle group is to raise the hyoid bone and widen the esophagus during swallowing. However, when the infrahyoids and other muscles stabilize the hyoid, these muscles (especially the digastrics) can result in wide mouth opening. Thus, when the need for pulmonary ventilation is high this muscle group plays a role in mouth breathing. Even so, the second set of traces in panel C clearly shows that the activity of the geniohyoid is much greater in swallowing than during inspiration.

Figure 8 The infrahyoid muscles originate on bony structures below the hyoid and insert onto it. Considering such a mechanical arrangement, it is no surprise that the main function of these muscles is to depress the hyoid during swallowing, though these muscles also play an important role in speech. The infrahyoid muscles are active during breathing under some conditions, especially during labored breathing. The breathing related role of the infrahyoid muscles is to regulate airway resistance. This may occur by: fixing the hyoid bone, which stiffens the upper airway; dilating the upper airway via depression of the hyoid bone; or by elevation of the upper thorax, which can lead to rib cage expansion.

Figure 9 As with other upper airway muscles, the pharyngeal muscles participate in swallowing, speech and breathing. Because of their critical role in swallowing, the muscles have a constrictor action. Although pharyngeal constriction narrows the upper airway, the breathing related activity is thought to offer some airway protection by stiffening it. The increased stiffness helps prevent pharyngeal collapse under conditions where the airway pressure is significantly below atmospheric pressure; this occurs with intense contraction of thoracic muscles, leading to thoracic expansion.

Figure 10 These figures show the anatomy of the larynx and laryngeal muscles, as well as EMG recordings of some of the muscles under various conditions. The laryngeal dilator (or abductor) muscles increase the size of the laryngeal opening (the glottis) (Panel B), and are active as breathing is increased, for example with hypercapnia (Panel C). These muscles are active during inspiration. Laryngeal constrictors (adductors) narrow the glottis, and are active on expiration. The importance of this is to act as a “brake” on expiratory airflow, preventing the lungs from emptying too quickly. In adult animals, the laryngeal adduction has little influence on gas exchange, but instead helps regulate breathing frequency by prolonging the expiratory period. In contrast, infants tend to have low oxygen stores so the retardation of expiratory airflow may provide a mechanism for improving gas exchange by elevation of end-expiratory lung volume.

Figure 11 The muscles of the thorax can either increase or decrease thoracic dimensions. Muscles that increase thoracic volume are categorized as inspiratory muscles, as they promote lung inflation. The expiratory muscles compress the thorax and lungs, promoting lung emptying. The expiratory muscles are quiet at rest, as the elastic properties of the lung are sufficient to drive airflow and empty the lungs, even in the absence of muscular effort. On exercise, breathing frequency increases so it essential that the lungs empty more quickly, and the expiratory muscles are thus activated during exercise. Thoracic expiratory muscles may be active even at rest in some persons with obstructive lung diseases, such as emphysema.

Figure 12 The abdominal muscle group includes the diaphragm and the muscles of the abdominal wall. The diaphragm can be properly defined as a thoracic or abdominal muscle, and we defined it as belonging to the latter group in this review. The diaphragm separates the thorax from the abdomen, and thus forms the flow of the thoracic cavity. It is an inspiratory muscle, and it is active continuously throughout life. For this reason, the diaphragm is considered to be the primary muscle of inspiration. The abdominal wall muscles are very important for posture and trunk rotation. However, during exercise the abdominal muscles contract during the expiratory period. By pushing the diaphragm upwards, and compressing the lower ribs, the abdominal muscles reduce thoracic volume and help to accelerate expiratory airflow, working as agonists of the internal intercostal muscles.

Figure 13 This figure shows the respiration-related activity of the back muscles that have been studied in the context of the control of breathing. This includes the levatores costarum muscles (or simply the “levator costae”), the latissimus dorsi, and the erector spinae muscles. These muscles have an inspiratory action on the thorax, but only under conditions of high respiratory drive, as with vigorous exercise or when inspiratory airflow resistance is elevated.