Anatomy and physiology of skeletal, smooth, and cardiac muscle

This chapter is most relevant to Section L1(i) from the 2017 CICM Primary Syllabus, which expects the exam candidates to be able to "describe the anatomy and physiology of skeletal, smooth, and cardiac muscle". Trying to guess the examiners' minds from the way this syllabus item is positioned (in the "musculoskeletal" section), we can infer that the detailed anatomy and physiology of cardiac muscle is probably intended for the cardiovascular section, and what they wanted from us here is more of a comparison of the ultrastructural elements that distinguish the three muscle types. This certainly seems to be the spirit of Question 11 from the second paper of 2015, which asked for a comparison of the anatomy and physiology of skeletal and smooth muscle. "It was expected answers would describe in detail the role of troponin, tropomyosin and calmodulin in mediating muscle contraction", the examiners rejoined. Details of histology and mechanisms of relaxation were also expected. This short three-sentence comment has informed the content and structure of what follows, but a lot of that material belongs in the section on excitation-contraction coupling.

Of the freely available peer-reviewed resources appropriate for revision of this topic, none beat Sweeney & Hammers (2018), as this review covers all possible examinable topics and has sections comparing skeletal muscle to smooth and cardiac muscle. There's even a table of comparison. 

Muscle tissue

Even though it is occasionally inconvenient for the surgeon or intensivist, movement is generally one of the most fundamental characteristics that define living things, and is present at every level of biological scale and sophistication. Basically everything can move, even if the movement involved is some kind of quiet intracellular shuffling of vacuole content, or a viral particle slowly contracting its envelope glycoproteins to squeeze itself into a cell.  Most of the mechanisms of motion rely on the conformational change of some molecule (usually a protein), and so unsurprisingly there are a huge range of different mechanisms available, often developed independently of each other and therefore cardinally different from one another even where they look the same and do exactly the same thing. Dedicated muscle tissue in multicellular organisms tends to exhibit a comforting structural homology, to effect that one could recognise striated muscle in a jellyfish and feel a warm kind of brotherly Verbundenheit, but it evolved probably at least four separate times in separate clades rather than arising from some early eumetazoan ancestor.

For animals, contractile tissues are usually separated into three distinct types, mainly on the basis of their fine structure.  At the most basic level we can separate muscle tissue into smooth and striated, where the striations originate from a repeating pattern of regularly arranged proteins, whereas "smoothness" is conferred by an irregular arrangement without a repeating pattern. The former is said to be better suited for rapid precise movements and is typically under voluntary control, whereas the latter is "automatic" and better suited for generation of a sustained force (for example, holding the valves of your shell tightly shut). Additionally, anything that has a closed circulatory system and a pulsatile propulsive mechanism to push the blood around will also tend to have a distinct cardiac striated cell type; this seems to be a mutant version of smooth muscle which has retained its automatic electrical activity but also gained striations. As is typical of living things in general, none of these are hard rules, and organisms break them with no regard for the mental health of biologists (for example Drosophila operate all of their viscera using only striated muscle, whereas tardigrades have no striated muscle in them anywhere except their pharynx). It is fortunate that absolutely none of this information has any relevance whatsoever for the CICM exam candidate, nor any candidate of any exam anywhere, and can be viewed as some kind of ballast, i.e. heavy inert material added to give stability to the rest of the chapter.

Definitions

The discussion of muscle will by necessity use a series of definitions that the CICM trainee might even be expected to regurgitate at some stage. If some of these may sound as if they are entirely made up, and lacking in scientifically rigorous foundation, it is because they totally are. To borrow a handsome turn of phrase from Last's Anatomy,

"There is no reality in these terms, though the sanctity of long usage forces their continued use"

In case any reader wishes to escape the grinding monotony of exam preparation by reading about the inconsistencies of early anatomical naming systems, they are redirected to Sawai (2018). Otherwise:

• Muscle is an unit of muscle tissue defined by anatomical rather than functional boundaries, i.e. it tends to be collected within a single connective tissue sheath.
• Myocyte is a single muscle cell,  occasionally referred to as a muscle fiber (specifically when referring to the clearly fibrous skeletal muscle)
• Myofibril is a single contractile organelle within a muscle cell, otherwise sometimes referred to as a sarcostyle.
• Myofilaments are individual contractile protein units, typically made up of a single protein type, eg. thin myofilaments are mainly actin and thick myofilaments are mainly myosin. 
• Sarcomere is a functional unit of striated muscle, defined as the muscle fibres arranged between two Z-lines.
• Sarcolemma is the name given to the cell membrane of skeletal and cardiac muscle, which deserves a special name because it has unique structural and functional properties.
• Sarcoplasmic reticulum is the term used to distinguish this structure from more mundane endoplasmic reticulum; it is a specialised organelle that mainly plays a role in coordinating calcium ion traffic.

Skeletal muscle

You call it "skeletal" by convention, as to call it "striated" would also end up including cardiac muscle, which contains striations. This is of course not an entirely accurate turn of phrase, as not all of these muscle fibres are attached to the skeleton by tendons (for example, the striated muscle of the upper oesophagus), to say nothing of all the animals that clearly have striated muscle but no bones. Occasionally one might also see it referred to as "voluntary" muscle, and this is again inaccurate because not all of these muscles are under voluntary control, nor is it easy to call anything a cnidarian does "voluntary" by any standard definition. In short, our nomenclature currently fails us, but for most normal people this is not a crisis, and nobody seems to be confused.

Macroscopic organisation of skeletal muscle

It is at this stage that a textbook usually pulls out a diagram of a cut muscle fibre, with a myofibril extruded out of it like a string of cheese. There are plenty of examples available within the first screen of a Google search and every reader will have already seen one during their exposure to Anatomy and Physiology 1.01 in whatever undergraduate hell they spawned from. Instead on reproducing those images again and diluting the search results of future Googlers, it is probably better to infringe on the copyright of ancient and forgotten anatomy textbooks. So, here's a lightly modified cross-section from the Atlas d'anatomie Humaine by Toldt & Lucien (1912):

Single skeletal muscle

This is one fundamental upper-level unit of muscle tissue, typically attached to some bones by tendons, with the meaty middle part usually referred to as the "belly" even where it is not distinctly swollen-looking. This structure is generally surrounded by an epimysium (sheath of connective tissue) which would ordinarily be referred to as a fascia, except these are too numerous and insufficiently special -  quoting Last's Anatomy,  "it is seldom of such a nature as to warrant special description as a named fascia".  Even more nameless are the sheaths of perimysium which envelop groups of fibres (fascicles) within a single muscle. 

Muscle fascicles

A fascicle is a collection of muscle cells all wrapped together in a perimysium sheath. Each such structure might contain 20-60 myocytes. In case you need a visual aid, here is a grainy photograph of a single human fascicle obtained by Infantolino et al (2012) through the blunt dissection a donated body known only as "Cadaver 2: female, 56 years old, 174 cm in height, cause of death – hypothermia":

Each muscle may have tens hundreds or thousands of these, depending on its size. Infantolino et al (2012) dissected and counted the fascicles only from the first dorsal interosseous, and there were between sixty to one hundred fascicles in this small unsexy muscle.

Motor unit of muscle

A motor unit is the smallest regional franchise of the neuromuscular business, and consists of a large anterior horn cell, its motor axon, and the skeletal muscle fibers
innervated by that axon. The number of motor units involved in a muscle and the number of fibres innervated per axon depends on how much precision one requires from the movement of that muscle. For example, the motor units of gluteal muscles innervate whole hundreds of muscle fibres, permitting only the crudest of movements, which is fine because most normal people don't expect to do anything especially precise with their gluteal muscles. In contrast, each motor unit of eye muscles only innervates 5 myocytes, permitting the reader's gaze to move very carefully from one offensive word to the next. In case it is of any interest, here is an often-repeated table from Buchthal & Schmalbruch (1980) listing the number of motor units per average muscle, and the average number of muscle fibres per average unit in those muscles:

It's probably important to mention that these muscle fibres from a single motor unit don't necessarily need to be all in the same fascicle. They are often scattered randomly across multiple fascicles, and each fascicle may contain the fibres of several motor units. A single motor unit might innervate some random fibres in a mass of muscle over a territory about 5-10 mm in diameter and 30-40mm in length, according to a series of experiments by Buchtal et al (1959).

Innervation and automaticity

It need not be said that skeletal muscle should never have any initiative of its own, lest we find ourselves flung all around the room by random spasms. To be sure, some skeletal muscle functions occur below the surface of normal consciousness, but they are all controlled by descending motor neurons. In contrast, some smooth muscle and most cardiac muscle has the capacity to 

Skeletal muscle fibres 

Muscle fibres in skeletal muscle are single myocytes. These are tubular cells about 100 μm in diameter and potentially up to several centimetres in length, making them some of the largest cells in the body. For the biology nerd it would probably be more accurate to call them syncytia because they are multinucleated and formed by the fusion of multiple mononucleated precursor cells. In terms of length these are second only to the neuron; some reach extreme dimensions, as it appears that larger fibres are cheaper to maintain. Length is also important because it appears to be a major determinant of total excursion, i.e. how much shorter the muscle can become when it contracts. There is some considerable variation within even just the human lower limb, looking at this table from Ward et al (2009):

 

Ultrastructure of skeletal muscle fibres

Skeletal muscle cells have several unique structural characteristics which the CICM trainee would be expected to list in an answer to any question that asks for some kind of comparison between skeletal and smooth muscle:

• They are multinucleated.
• The contractile elements are arranged into myofibrils.
• Within myofibrils,  the myofilaments (basic elements of contractile protein) are arranged into sarcomeres, regular linear arrays
• The cell membrane has unique properties, and gets to be called sarcolemma to distinguish it from less interesting membranes
• This sarcolemma invaginates into the cell in the form of long tubular channels (T-tubules) which bring extracellular fluid and excitable membrane closer to the sarcoplasmic reticulum.
• The sarcoplasmic reticulum is also sufficiently distinct from normal endoplasmic reticulum; its main job is to sequester and release calcium, whereas the rest of the reticulum handles more mundane tasks like protein synthesis.

Multinucleation of skeletal muscle cells

Multinucleation is a characteristic feature of skeletal muscle, as smooth muscle and cardiac muscle are usually mononuclear. Myoblasts or "satellite cells", the precursor stem cells of myocytes, also start out mononucleated, but then fuse into "myotubes" which are long fusiform cells with all the myoblast nuclei lined up in a central row. The end product is a muscle fibre with numerous nuclei (hundreds or thousands) arranged relatively regularly around the cell. Here is an example, a false colour image of a human vastus lateralis fiber from Van der Meer et al (2011):

Myocyte stem cells amalgamate like this under the influence of surface fusogen proteins Myomaker and Myomerger. We know how they do this but the current understanding of why myocytes need perhaps a thousand nuclei remains in the realm of speculation.  The prevalent idea is that multinucleation is the necessary adaptation to being huge. A single nucleus really could not produce enough protein to constantly supply the needs of a cell that is several centimetres long. As the result, a massive myocyte has many nuclei, each responsible for supplying protein products to a small regional volume of cytoplasm, its "myonuclear domain". According to Hansson et al (2020) these domains are probably around 20-25 picoliters in volume (ten times larger than the volume of most mammalian cells).

Skeletal muscle myofibrils 

Myofibrils are the contractile organelles of skeletal muscle, and as contraction is the main purpose of skeletal muscle, each cell is completely packed with these organelles much in the same way that a red blood cell is packed with haemoglobin. A skeletal myocyte might have about 90% of its volume occupied by myofibrils, with the remaining 10% shared between mitochondria (5%), sarcoplasmic reticulum, fat droplets, and nuclear material. Eisenberg et al (1974) described these structures as cylindrical, with an irregular polyhedral cross-section, of approximately 1μm diameter, and extremely long - these threads are said to run the entire length of the muscle fiber (Vye, 1976). That's quite long even for the little hamster soleus, let alone for human muscles, making these some of the longest organelles around. It would obviously be rather hard to isolate and carefully preserve a long fragile stringy structure like that, so no images of a whole intact myofibril are available, but here's a close-up of a relatively short myofibril fragment stretched between two microneedles, from Rassier (2017):

Structurally, each myofibril is a stack of sarcomeres, which means one can make longer and longer myofibrils by simply stacking more and more sarcomeres (and this is indeed what happens to muscle cells as they elongate in the growing organism). In case it helps to visualise things, here is a 3D rendering of some electron-microscope-scanned myofibrils from Drosophila flight muscles by Ajayi et al (2022), clearly showing their fine structure:

Skeletal muscle sarcomeres

Sarcomeres are the basic contractile unit of a myofibril, an array of regularly arranged proteins from which the striated muscle tissue gets its stripes.  It is a cylinder about 1μm diameter and 2-3μm in length, at rest. One usually does not see them in this form, because it would be preposterously difficult to identify and isolate one single sarcomere unit. Rather, the main structural features are usually demonstrated using a low power transmission electron micrograph of a longitudinal slice. Virtually every textbook has a diagram like this somewhere in it, and the specific micrograph used for the one below is so ubiquitous that its original source has been lost through massive repetition.  

Like any simple diagram with lots of labels, this one is very attractive to the exam question writer. One can envision an SAQ asking trainees to draw and label a sarcomere, describing the functional relevance of the labelled features. In this case the expected mark-scoring elements would be the lettered regions of the sarcomere. In case the reader is wondering where the names came from, they are mostly capitalisation of German words, and date back to work describing Muskelfibrille und sarcoplasma by Gustaf Retzius in the 1880s and 1890s.

• I bands, consisting of thin actin filaments, and named because they are isotropic (referring to the behaviour of polarised light as it passes through them - i.e. their regular structure permits the passage of polarised light in one direction, but not in another)
• A bands, consisting of thick myosin filaments, and so named because they are anisotropic (the protein molecules here are thicker and polarised light passing through this structure is refracted in a bunch of different directions, irrespective of how you shine it)
• Z lines, sometimes referred to as "Z discs", so named because they are seen between sarcomeres (and used to define the boundaries of the sarcomere), which makes them zwischen (between). These act as anchoring points for titin, a long protein that acts as a sort of elastic scaffold for the rest of the sarcomere structure
• M lines, which are in the middle (mittel) of the A band, and which contain creatine kinase and myomesin - this is the region where thick myosin filaments are crosslinked.
• H zones, home to the creatine kinase proteins, so named because they are heller (brighter) than the surrounding A bands. The decreased density here is attributed to there being no actin filaments (whereas the rest of the A band is an overlap of actin and myosin filaments). This area is probably important to know about, as its length corresponds to the plateau of the length-tension relationship, i.e. the sarcomere contracts best where there is  still maximum overlap between the thick and thin filaments, and the tension decreases where the overlap is decreased (eg. where the sarcomere is stretched and the H zones are larger, or where the sarcomere is compressed and the H zones are diminished).

Most normal people would never create an exam question asking what the Z in Z line stands for, and in all honesty the ability to label a diagram of a sarcomere would surely never be the pass-or-fail decider in a written answer. As such, a deep dive into to the fine details of sarcomere protein structure would probably be a waste of everybody's time. Instead the reader is referred to this 39-page review by Craig & Padrón (2004), and the amazing microscopy by Wang et al (2021). 

Having said all that, in Question 8 from the first paper of 2014 the examiners did note that "marks were gained for a brief outline of the structure of a sarcomere and how it facilitates shortening", though they did not mention how many marks, and a superficial reading of the question itself would not immediately prompt the candidate to write about sarcomere structure. So: just how does it facilitate shortening? What could they have possibly meant by this? Perhaps this is referring to the overlapping structure of the filaments, which is essential for their sliding along each other? In which case a point-form explanation would look like this:

• Contractile proteins are arranged into sarcomere units, composed of overlapping regular bands of fibres, and this structure facilitates contraction.
• Thin (actin) filaments overlap with thick (myosin) filaments in the A band region
• The I band represents a region where there is no overlap (only thin filaments)
• When the filaments slide along each other due to cross-bridge cycling, the I bands shrink in size as the area of overlap between actin and myosin, bringing Z bands closer together
• Because the thin filaments are tethered to the myocyte cytoskeleton at the Z band, the mechanical force generated by actin and myosin sliding along each other is transmitted to the surrounding myocyte structures.

Readers are cautioned that this interpretation is fraught, as the minds of the examiners are unknowable, but it seems plausible that somebody somewhere might one day be asked about this.

The sarcolemma

Sarcolemma or myolemma is the cell membrane of a striated muscle cell, apparently named differently mainly to sow confusion. Yes, it has unique features which make it suitable for its purpose, but so do many other cell membranes that don't have separate nomenclature. But to be fair, this thing does need to stand up to a lot of mechanical stress, transmit action potentials quickly and efficiently to all the sarcomeres, glue all the muscle fibres together into a functional unit, and still perform all the other routine housekeeping functions of a normal cell membrane. From Campbell & Stull (2003), the main characteristic structural and functional features of the sarcolemma are:

• Glycocalyx,  a 20-60nm layer of glycoprotein that links the sarcolemma directly to the basal lamina, acting as a mechanical coupling between adjacent myocytes
• Costameres, submembrane anchoring proteins that tether the sarcolemma to the Z-discs
• Caveolae, little pit-like structures that protect this membrane from mechanical stress (in a really simplistic sense, these things flatten to release more membrane material in response to stretch)
• Excitability, i.e. the capacity to propagate an action potential (no, this is not special, but it is an important feature)
• T-tubules,  long invaginations of the membrane which penetrate deep into the muscle cell to deliver exchangeable electrolytes and action potentials

You can see most of these features (except excitability) in this slightly doctored version of an excellent image from McNutt (1975):

T tubules of the sarcolemma

T tubes or T tubules, so called because they are transverse, are usually described in textbooks as invaginations of the sarcolemma. This word invagination appears quite a lot in important official works, and so could probably be borrowed by the reader to add authority to whatever they are writing (for example, "the pocket is an invagination of the trouser"). These are tunnels that open to the surface of the sarcolemma and lead deep into the thick (100 μm, we said) skeletal muscle fibre. They are also lined with the right sort of ion channels that the action potential can follow them down into the depths of the cell, activating all the sarcomeres throughout the myocyte. In that sense, these are an essential component of normal muscle function.

It might actually help to have a sense of the real scale and shape of these structures, particularly when one is accustomed looks at artists' impressions of them. These people are often handed a fairly vague explanation of what is required for the scientific paper, and can accidentally create convincingly professional-looking artwork that unfairly misrepresents the shape, size, or relationship of the cellular architecture. For example, here's a familiar image of a myofibril, representative of its kind - from Eisenberg et al (1974), by an artist not credited anywhere in the paper, who left us to squint at their indistinct signature along an edge of sarcolemma on the right (does that say "Gloege"?)

It's beautiful; those plump glistening mitochondria look almost edible; and it does represent the structures well enough that you could probably use this image to teach somebody about the ultrastructure of muscle; but regular readers of Φ would probably regard the use of such stylised images as a sort of blue pill. For those weirdos, here are some microscopy images of real T tubules to give a more realistic representation of what they look like. 

So: transverse tubules open to the surface of the cell like little pits, roughly at the same position as the intracellular Z lines of the sarcomeres. Here's an SEM image of a monkey myocyte from Kostin et al (1998), where the regularity of their spacing can be easily appreciated:

 

They penetrate the cell radially, converging on the centre. These neon-orange images are from Jayasinghe et al (2015) - this is light microscopy of ventricular myocytes from horses and humans, where the live cell was soaked in a dextran-fluorescein mixture that got sucked up deep into every tubule. 

Closer inspection using serial confocal microscopy can reveal this network in a lot more detail. Here is an image concatenated from works by Jayasinghe et al (2015) and Cully et al (2017), showing that some of the transverse tubules are actually connected to each other longitudinally (violating their name).

Each T-tubule typically has a diameter somewhere around 200 nm, but they can vary from 20 to 450 nM, and are fairly irregular in diameter, as seen in this crossection:

Down there in the depths of the muscle fibre these tubules interface with terminal cisternae of the sarcoplasmic reticulum to form the "triads" of the muscle cell, where a lot of the ionic magic happens. Whenever you see microphotographs of triads in textbooks, the images are often credited to Clara Franzini-Armstrong, to whose work we owe a lot of our understanding of muscle physiology (the one below was modified from her 1970 paper):

Sarcoplasmic reticulum

This smooth endoplasmic reticulum earns the prefix "sarco", even though it is still endo, because it is sufficiently distinct from normal endoplasmic reticulum at a molecular level.  It handles a different range of functions and contains a different selection of proteins, mainly focused on commanding calcium ions (Michalak & Opas, 2009). Structurally, it is a network of fine hollow tubular structures that surrounds the myofibrils, with the ends gathered into dilated cisternae at the ends (hence them being called "terminal cisernae"). Following a broad trend in muscle nomenclature, this name is also inaccurate, as these cisternae aren't terminal - in the triad the two cisterns are linked by multiple connections and the network is continuous from one end of a muscle fibre to the other. In other words, to call this sprawling labyrinthine complex a network does no justice, as it fails to adequately describe its vast extent and interconnectivity.  The only thing that could do this would be an image, such as this 3D reconstruction from Pinali et al (2013). Sarcoplasmic reticulum is red; the little greyish central structure in the white-squared focus area is a T tubule, cuddled by the embrace of terminal cisternae.

The main role of this network is to release and then re-sequester calcium in precise coordination with the action potentials propagating through the tubule network. More detail regarding this process is offered in the chapter dealing with excitation-contraction coupling, and in this excellent review paper by Rossi et al (2008). 

Role of troponin, tropomyosin and calmodulin

This is probably needed here because comparing differences in contractility is a necessary part of comparing different types of muscle tissue, even though it infringes somewhat on the chapter about excitation-contraction coupling. In the briefest possible form:

• Skeletal muscle contains thick and thin filaments 
• Thick filaments contain heavy and light chains of myosin:
  • Heavy and light chains interact to form the "heads" of myosin
  • The heads of myosin interact with actin
• Thin filaments contain actin, tropomyosin and troponin
  • Actin contains binding sites for myosin, which are obscured by troponin and tropomyosin
  • This prevents actin and myosin from interacting at rest
  • One of the troponins (Troponin C) binds calcium to initiate muscle contraction
• Contraction:
  • Intracellular calcium attaches to troponin C
  • This results in a conformational change, revealing active sites
  • Actin binds myosin
  • Myosin then binds ATP and undergoes a conformational change which causes it to dissociate from actin and "cock" its head, moving down to the next binding site
  • The ATP bound to myosin becomes hydrolysed and myosin can then bind the next actin binding site, moving up the filament
  • This is the "cross-bridge cycle" which repeats to produce contraction
  • It occurs rapidly in response to calcium entry into the cell, and continues only while calcium is present
• Relaxation
  • When calcium concentration decreases, troponin and tropomyosin return to blocking actin binding sites, and the muscle relaxes
• Role of calmodulin is ...non central. Off centre?
  • Let's call it peripheral
  • Calmodulin does participate in skeletal muscle function, but mainly as a regulatory protein that coordinates the activity of other protein kinases involved in things like plasticity and adaptability of skeletal muscle (Walsh, 1983)

We will see this contrasted with other mechanisms of muscle contraction.

Types of skeletal muscle fibre

Just like people can be awkwardly classified into insultingly inaccurate stereotypes, so muscle fibres can be crudely divided into vague overlapping groups according to their responsiveness to neural stimuli, metabolic characteristics, speed of movement, and, apparently, colour. Not all fibres fit neatly into these groups, each muscle may be composed of multiple different fibre types, and trying to classify by different histological mechanical or physicochemical methods could place the same fibre into completely different categories depending on the method. Moreover, even within the same classification system not everyone can settle on the exact divisions - for instance some resources refer to 2b and 2x fibres as being the same thing, whereas others treat them separately, and others still refer to some mysterious 2d fibres or dabble in forbidden heretical systems that include seven subtypes in total.

This confusing mess of disagreeing classifications is discussed calmly and systematically by Scott et al (2001). It would be normal to feel helpless, if one were truly expected to learn meaningless classifications for a high-stakes exam, but fortunately the college have never asked CICM trainees to regurgitate these memorised categories, and if we are lucky they never will. Still, just in case, it would not hurt to arm oneself with some kind of quick and unofficial aide-mémoire. So, here's a table concocted from numerous such tables from Gohil et al (2013), Thamrin (2008), Maglischo (2015), and probably others, though realistically it does not matter because they are all largely the same.

Types of Muscle Fibers

Characteristic	Type I	Type IIa	Type IIb
Contraction speed	Slow	Fast	Very fast
Force produced	High	High	Low
Capillary density	High	Moderate	Low
Fibres per motor unit	<300	>300	>300
Mitochondrial density	High	Moderate	Low
Storage fuel	Triglycerides	Creatine phosphate, glycogen	Creatine phosphate, glycogen
Metabolism	Aerobic	Anaerobic	Anaerobic
Lactate removal rate	Low	Highest	High
Fatigue resistance	High	Moderate	Low
Myoglobin content	High	High	Low
Glycogen content	Low	Moderate	High
Colour	Red	..Reddish?	White

To this, one could also add some short discussion of extrafusal and intrafusal fibres. This classification is simple: extrafusal fibres are innervated by α motor neurons and do the bulk of the contractile work, whereas the other ones are innervated by γ motor neurons and perform the role of stretch receptors, participating in tendon reflexes and the maintenance of muscle tone. They retain some contractile properties but their main role is in proprioception and the regulation of contraction, so the discussion of their properties is probably best left to the neurology section.

Smooth muscle

You call it "smooth" because that's how it appears under light microscopy, because - to borrow the expressions of Giorgio Gabella - "There is no apparent lateral register between myofilaments in smooth muscle cells", and therefore no regular striations can be seen.

Anatomy of smooth muscle tissues

Unlike skeletal muscle which tends to get noticed by collecting into attractive subcutaneous lumps,  the beauty of smooth muscle is harder to appreciate because it is diffusely scattered through basically all tissues, and so it would be impossible to ogle its flexing explicitly, or discuss its anatomical macrostructure in a generic way that would be accurate for all of its occurrences in the body.  Still, a few common features of smooth muscle anatomy can be gathered from disparate resources. Bizarrely, nobody seems to write review articles consolidating this material, and the best resource was this 1997 book chapter by Giorgio Gabella. 

Smooth muscle, ladies and gentlemen:

• It typically occurs in layers or sheets
• Its overall shape usually has some circularity to it, i.e. it tends to encircle things, for example the lumens of hollow organs. This is of course not mandatory (linear stretches of smooth muscle do exist, for example the arrector pili muscle, a bundle of smooth muscle cells attached to each hair, or the long bands of taeniae coli).
• Where something hollow is encircled, often there are layers of cells arranged in perpendicular arrays, eg. in vascular and visceral smooth muscle there is an encircling inner layer and a longitudinal outer layer which allows peristaltic activity.
• It does not need to attach to a bone or tendon, and in fact that would be very unusual behaviour for it.
• Unlike skeletal muscle, which basically pulls things in a direction, smooth muscle has a diverse range of mechanical and functional roles, and is capable of greater excursion (i.e. it can shorten a lot more than skeletal muscle)
• There are no "fascicles" to speak of, and no endomysium or epimysium (though various hollow organs do tend to feature fascia-like connective tissue sheaths)

In case this is of any use to anybody, the total mass of smooth muscle in the body is vastly less than the total mass of skeletal muscle. Skeletal muscle may comprise 30-40% of the total body weight, whereas skeletal muscle only contributes 2-3%. For the numerologist, Gabella produces some values to describe the rough quantities of smooth muscle in a normal adult:

Smooth muscle cell morphology

In comparison to skeletal muscle cells, these are tiny. A smooth muscle cell may only be 2-4 μm in width, and measuring no more than 1000 μm in length - roughly the same proportion as an undersized striated myofibril. The cell volume is apparently about 2,500-3,500 cubic microns for a large one, which is the same as a decent-sized monocyte. As such these cells have no need of multiple nuclei, and typically only have one. Here is a representative image of an isolated visceral myocyte from a human colon, the source of which is unknown:

They are usually described as "spindle-shaped" or fusiform, and the contractile apparatus inserts along the entire length of the cell rather than just at the ends. These cells are also packed full of contractile proteins - 90% of their volume are myofilaments, according to  Wray & Burdyga, 2010.

Sarcolemma of the smooth muscle cell

The reader is warned that for some resources the term "sarcolemma" appears to be unofficially reserved for striated muscle membranes, whereas others refer to the "sarcolemma of the smooth muscle cell", and it is unclear where one's fortunes may fall when using this word in an exam. It is certainly distinct from the cell membrane of striated cells:

• It contains no T tubules. Those are a feature of skeletal muscle only.
• It has many more caveolae, presumably to facilitate the incredible stretchiness of smooth muscle
• It often features gap junctions, which skeletal muscle sarcolemma does not (allowing action potentials to propagate directly from cell to cell).

The presence or absence of gap junctions can also be used to subclassify smooth muscle into "multi-unit" and "single-unit", the latter joined into a single functional whole by being connected through gap junctions, and the former being a collage of individual cells with no connections. Singe-unit smooth muscle is more typical of hollow viscera, particularly those whose job is something peristaltic, whereas multi-unit smooth muscle is more typical of tissues that need to maintain some regional individuality, for example vascular and airway smooth muscle. 

Sarcoplasmic reticulum in smooth muscle

Like in skeletal muscle, the SR of smooth muscle serves a calcium-handling role, i.e in both cases this organelle amplifies calcium flux by means of calcium-sensitive calcium channels. On the basis of this boring sameness, it feels wrong to write very much about it here, except to borrow from Wray & Burdyga (2010):

"It seems to us that the pendulum swung from initially dismissing the SR’s role in modulating Ca signals and CICR in smooth muscle to viewing the SR as playing a very similar role to that occurring in striated muscles. We are perhaps now able to readjust the pendulum and see more clearly that smooth muscle SR is important to [excitation-contraction] coupling, but in its own distinctive way."

The authors go on to explain the subtle differences between skeletal and smooth SR over about sixty pages, which seems excessive for this brief overview. Only important differences will be highlighted:

• It tends to run in sheets longitudinally along the cell, and all smooth muscle tissue seems to share this common structural feature.
• Because there are no transverse tubules, there are no cisternae, and therefore no triads, but the SR of smooth muscle does have structural "junction regions" where it abuts membrane caveolae, and those seem to serve the same purpose.
• There is less of it (only about 5% of the cell volume is SR, instead of 10%)

Contractile protein composition of smooth muscle

There is a lot more actin in smooth muscle than there is in skeletal muscle, and a lot less myosin - only about 20% relative to skeletal muscle. The result is a thick:thin filament ratio of about 1:15, with myosin filaments distributed widely across the smooth muscle. Myosin is probably the most important protein here from the point of view of contractility: without going into too much detail about the excitation-contraction coupling of smooth muscle, it will suffice to say that contraction is mostly regulated by calmodulin-dependent phosphorylation of the myosin regulatory light chains. Of the actin, some participates in this, and some is mainly there for structural support (so-called "cytoskeletal actin", that does not make contact with any myosin filaments). 

Distribution of contractile protein in smooth muscle

The defining feature that makes muscle "smooth" is that they do not contain myofibrils. Instead of being bunched into cylindrical lengths of regularly stacked sarcomeres, the myofilaments are spaced irregularly across the cell. They do not connect along the cell end-to-end; the contractile elements insert along the entire length of the myocyte, anchored to "dense bodes" (or "dark bodies") instead of Z lines. These structures are so called because they have high affinity for the electron-dense metal stains used in electron microscopy, and therefore end up looking darker or denser on the images. Here's a representative image from a Yale website:

They are rigid, as in they seem to resist bending and twisting, and rather large -  up to 1.2 μm long and 0.3 μm wide. Zhang et al (2010) is a long exploration of their role, which can be summarised as "anchoring". They function as attachment points for intermediate filament cables made of desmin or vimentin and for contractile actin filaments. Their main constituent is α-actinin, the same protein that you expect to find in the Z lines of a skeletal fibre, and given what we know about the ontogeny of muscle cells, it is believed that Z lines in skeletal muscles arose from dense bodies and represent a design revision focused on linear mechanical force. The overall structure is best described with a diagram such as this one from Sweeney & Hammers (2018), outlining the relationship of all these cytoskeletal elements:

Thus, contractile filaments connect to dense bodies diagonally across the muscle cell instead of taking a linear path parallel to its main axis. The myofibrils are still arranged vaguely longitudinally, i.e. parallel to the long axis of the cell, but their alignment is usually slightly off-centre and they insert into their membrane anchor points at acute angles (Gabella, 1984). It is rather difficult to find any clear discussion in the literature as to what the point of this arrangement might be. Sweeney & Hammers (2018) conjecture that this arrangement favours elasticity and increases the possible shortening of a smooth muscle cell, some of which can contract down to 20% of their original length. When they are unconstrained by peer pressure from neighbouring structures, isolated in vitro smooth myocytes can even assume a totally spherical shape when properly motivated. This is permitted by the scissor-lift-like organisation of their myofibrils, but would be impossible with the telescope-like sarcomere structure of striated muscle. 

Innervation and automaticity

Unlike skeletal muscle, which receives mostly descending innervation via α motor neurons from the motor cortex, smooth muscle is innervated mainly by either the autonomic nervous system, or frequently not at all. To be completely without innervation is an option for some fibres in single-unit smooth muscle tissue because the sheet is interconnected and a wave of excitation can be expected to propagate across it, removing the responsibility from individual fibres. This accounts for the observed ability of hollow organs to persist with their peristaltic movements even in the absence of neural or hormonal input. Interestingly, the literature describing smooth muscle innervation as a wider topic is hard to find  (i.e. discussions of how it compares to the innervation of other structures, rather than focusing on some obscure intracellular signalling pathway). McLaughlin et al (2006) is a notable exception, albeit unreferenced and abbreviated (but perhaps that is what you want?); an excellent academic alternative is Di Natale et al (2021). In short, the following observations can be made about the innervation of smooth muscle as a tissue type:

• It can be without innervation, and get its stimuli to contract from paracrine or hormonal signals, or it can contract autonomously to the metronome of a pacemaker current
• Of the innervated smooth muscle, the autonomic nervous system is the most important control apparatus
• Regional control of the enteric smooth muscle is also exerted by the myoenteric nervous system
• For single-unit smooth muscle tissues, the effect of nervous system input is a modulation of some baseline level of activity (eg. the rate of peristalsis, or the intensity of tonic contraction)
• Often enough smooth muscle is "asymmetrically" innervated, where a superficial layer of cells is directly in contact with nerve endings, and deeper cell strata depend on gap junctions and relayed signals for their innervation (for example, arterial smooth muscle). In contrast, each skeletal muscle fibre is reliably innervated at one neuromuscular junction.

There are no structured neuromuscular junctions with narrow gaps in smooth muscle, like in the case of skeletal muscle; instead, the autonomic nerve endings simply terminate somewhere in the vicinity of the smooth muscle cell, relying on diffusion to carry neurotransmitters vaguely in the direction of their receptors. The whole thing resembles a sprinkler system: axons travel along the smooth muscle tissue and occasionally their Schwann cell myelination is interrupted to form a little "varicosity" which is designed to spray neurotransmitters haphazardly into the general area. This is generously referred to as a "diffuse junction", though in reality there's no junction at all, and one could accurately call this "accidental innervation". 

A slightly more civilised version of this would be the "contact junctions", which are still unstructured but where the neurotransmitter at least benefits from a narrower gap between the varicosity surface and the surface of the muscle cell. The synaptic cleft here is close to what you would expect in skeletal muscle, i.e 30-50 nm. Diffusion is therefore faster and the smooth muscle is therefore more reactive. The term "contact junctions" does not appear to be particularly widely accepted, and a search for the term yields mainly several identical replicas of the same entry from Guyton & Hall ("These are called contact junctions, and they function in much the same way as the skeletal muscle neuromuscular junction", etc). Multi-unit smooth muscles are said to feature this sort of low-latency junction, according to other respectable online resources.

Types of smooth muscle

Like with skeletal muscle, smooth muscle can be classified into several imperfect categories, and the only way the exam candidate could be truly wrong here is if the classification schema they choose ends up being the one that the examiners have never heard of. The "single unit vs multi unit" classification system is probably the most common, and should be familiar to everyone; or at least one tends to see it frequently in textbooks. Of course the main reason that one tends to see it in textbooks is that it is old. It appears that some authors have modernised it by referring to single-unit smooth muscle as "phasic" and to multi-unit smooth muscle as  "tonic", on the basis of their behaviour: phasic smooth muscle tends to contract rhythmically, and tonic smooth muscle is constantly contracted and modulates its constant tone or force. Thus, visceral (eg. intestinal, ureteric) smooth muscle is single unit and phasic, while vascular smooth muscle is multi unit and tonic. If you look long enough under the hood of their metabolic and motor machinery, it is possible to subclassify smooth muscle even further; for example Boberg et al (2018) were able to identify four classes on the basis of differences in their expression of the inserted myosin heavy chain.

Mechanical  and metabolic properties of smooth muscle

Again trying not to repeat whole sections of the excitation-contraction coupling chapter, smooth muscle can be broadly described as a slow low-octane tractor tug.  Both striated and smooth muscle fibres can ultimately generate approximately the same magnitude of force, but even the fastest smooth muscle fibres (bladder) are orders of magnitude slower than even the slowest skeletal muscle (Fisher, 2010). On the other hand, even the hungriest smooth muscle consumes much less metabolic substrate than striated muscle. In particular, tonic smooth muscle can sustain a high force of contraction with minimal energy expenditure, something occasionally referred to as a "latch" phase, referring to the analogous function of adductor muscles that shut the valves of bivalve molluscs. 

Role of troponin, tropomyosin and calmodulin

To very briefly outline the main differences between smooth and skeletal muscle here:

• Smooth muscle contains thick and thin filaments, much like skeletal muscle, but:
  • Thin filaments:
    • Contain actin, but no troponin
    • much longer and more numerous
    • some of them are anchored to the membrane and the dense bodies
  • Thick filaments:
    • Fewer than in skeletal muscle
    • some of the smooth muscle myosin is depolymerised
• Contraction:
  • Intracellular calcium attaches to calmodulin (there's no troponin)
  • Calmodulin activates myosin light chain kinase
  • Myosin light chain kinase phosphorylates the head of myosin
  • Only phosphorylated myosin heads can participate in cross-bridge cycling, which means this step permits contraction
  • Contraction then occurs via actin-myosin cross bridge formation,  same in all muscle, be it striated or smooth; but in smooth muscle the ATPase activity is much lower, which leads to a much slower cycling speed. 
• Relaxation
  • When calcium concentration decreases, myosin light chain phosphatase dephosphorylates the myosin light chain kinase and puts an end to the contraction
  • This is a key difference: merely removing calcium from the cell will not produce a rapid relaxation; dephosphorylation is also required
  • Myosin light chain phosphatase is activated by cGMP-dependent protein kinase, and is therefore responsive to nitric oxide.
• Role of calmodulin is central:
  • It stands in for troponin C, and binds calcium
• Different source of calcium
  • Skeletal muscle mainly uses calcium in the capacious endoplasmic reticulum network, whereas smooth muscle is more reliant on extracellular sources

Cardiac muscle

Now that basically everything about smooth and skeletal muscle has been explained a little bit (or at least mentioned in passing), the task of writing about the cardiac muscle tissue becomes easier, as cardiac myocyte properties are a fusion of familiar characteristics. Cardiac myocytes are striated, and yet they are significantly different from skeletal muscle cells, occupying a sort of middle ground between smooth and skeletal muscle:

• Like skeletal muscle cells, they are striated, with myofibrils spanning the length of a roughly cylindrical cell.
• Unlike the huge skeletal muscle cells, they are much smaller, no more than 100–300 μm in length and 20 –30 μm in diameter, making them closer to smooth muscle cells in their dimensions
• They are binucleated or trinucleated, i.e. they occupy the middle ground between mononucleated smooth muscle cells and massively multinucleated skeletal muscle cells
• Like single-unit smooth muscle, they are attached to each other with gap junctions (intercalated disks), which permit the transmission of an action potential from cell to cell (whereas skeletal muscle cells are electrically isolated). The trainees that use the term "electrical syncytium" in their exam answers may appear unfairly cleverer than their peers, but that is the term used to describe this sort of structure.
• Like smooth muscle cells these tend to get their calcium from the extracellular fluid as well as from their own sarcoplasmic reticulum, whereas skeletal muscle depends almost entirely on the SR.
• The duration of the action potential of a cardiac myocyte (200-300 ms) is much slower  than that of skeletal muscle, but much faster  than the usual contraction of a smooth myocyte
• Like smooth muscle, cardiac myocytes are not under any direct voluntary control from the central nervous system, and are innervated only by autonomic nerves.
• Like smooth muscle, cardiac myocytes can adjust their contractility properties rapidly in response to stimuli, ranging from stretch (eg. the Frank-Starling relationship) to neuroendocrine stimuli (eg. catecholamines). For example, the cardiac-specific isoform of troponin (cTnI) is a phosphorylation target for protein kinase A, which is activated by the cyclic AMP released by β-receptor activation.