All eukaryotic cells share the same basic layout, in that they are surrounded by a membrane, and filled with cytoplasm within which there is a variety of membrane-bound organelles that perform specialized tasks. One major organelle is the nucleus which contains DNA. At roughly one-thousandth the volume of the eukaryote cell volume, the organization of a prokaryote is relatively straightforward by comparison, although bacteria possess very similar structural aspects and machinery to eukaryotic cells. Their cell membrane is based on the same lipid bilayer (which we will discuss next) and their molecular machinery such as ribosomes for protein assembly functions in much the same way as eukaryotes. Bacteria do have specialized cell structures for motility such as flagella, and may possess internal membranes as in the case of a gas vacuole that serves as a buoyancy aid. However, the DNA in a bacterial cell is a single circular molecule and there is no separate nuclear compartment.

The cell membrane

All cells are enclosed by a boundary structure, the plasma membrane, which provides a barrier to other cells and the external environment. Although the membrane serves to contain the cell contents, unicellular organisms (bacteria included) usually have extra material on the outside and plant cells are characterized by a rigid cell wall of cellulose ( Figure 4 ). Over a century of investigation has shown membranes to be an incredibly complicated and dynamic mixture of lipid (fat)molecules and proteins. Although the basic structure of the plasma membrane is extremely thin-just a couple of lipid molecules thick, forming a lipid bilayer-it is extremely tough and flexible, and also permeable to allow for the constant exchange of molecules between the cell and its surroundings.This is achieved via the water that constantly enters and exits in a controlled manner bringing soluble molecules like oxygen(needed as a fuel), and exporting waste products such as carbon dioxide. Large external material can be physically engulfed by the membrane, a process known as phagocytosis. The reverse process is exocytosis, in which a membrane-bound vacuole of material destined for export reaches the cell surface, at which point the membranes fuse and open to the outside, releasing the contents without breaching the overall integrity of the membrane. The dynamic nature of the cell membrane is such that the entire plasma membrane is ‘turned over’ (replaced) on an hourly basis.

Some of the earliest experiments involving the interactions of lipids and water to explain the properties of membranes were performed towards the end of the 19th century, on a kitchen table in Germany by Agnes Pockels. Agnes studied the behaviour of oil poured onto water in a flat dish, identifying the in fluence of impurities on the surface tension of fluids. She sent her results to Lord Rayleigh, who was sufficiently impressed to get them published in the scientific journal Nature in 1891. In 1932,Irving Langmuir, working in New York, won a Nobel Prize for showing that lipids spread on water produce a layer only one molecule thick and that all the molecules were orientated the same way. This happens because one end of the lipid molecule is attracted to water (it is hydrophilic), and the other end is repelled(it is hydrophobic). Each lipid molecule is shaped like an old-fashioned wooden clothes peg, with the top of the peg being the hydrophilic end, and the two legs of the peg representing the hydrophobic region. All the pegs float on the surface of water head down, legs uppermost, forming a monolayer. In 1925, Evert Gorter and James Grendel isolated membranes from red blood cells, finding that membrane was made up of two layers (a bilayer) of lipids, making a sandwich with the hydrophilic peg heads on the outsides, and the hydrophobic legs on the inside. As there is water both inside and outside the cell, this arrangement is maintained, keeping both hydrophobic surfaces together on the inside of the membrane bilayer. This arrangement was directly visualized decades later with the advent of electron microscopy, where thin sections at high magnification showed membranes as two dark lines separated by a light region between them (see, for example,　Figures　4 and 14c). In 1935, James Danielli and Hugh Davson suggested that this lipid bilayer was covered on both sides by a layer of proteins, a model that lasted until 1972 when Seymour Singer and Garth Nicholson suggested that the proteins could also be threaded through the lipid layers, and project from either side of the membrane. Proteins with this confi guration are termed transmembrane proteins, and a single protein might weave its way through the lipid bilayer several times. The lipid molecules in a membrane are highly mobile, constantly moving past one another and the membrane proteins, leading to the description of the ‘fluid mosaic’ membrane. This activity of the lipids was neatly demonstrated by Michael Edidin in 1970, when he labelled the membrane lipids of two different cell types with either green or red fluorescent chemicals. This gave a ‘patchwork’ of red and green labelling from individual cells grown together in a mixed cell culture. Edidin then added viruses to the culture which caused the membranes of adjacent cells to fuse to each other, thus mixing their membrane lipids and, within an hour, all red and green patches of fluorescence were replaced by an overall orange labelling, showing complete mixing of the individually labelled lipid molecules within the fused membranes.

4. Section through a plant cell, showing the main differences from animal cells: a cell wall outside the cell membrane, chloroplasts with starch grains (SG) inside them, and a large vacuole in the centre of the cell

Within this constant motion of the lipids, groups of membrane proteins float around freely in the lipid bilayer rather like ice floes in the polar seas. Sometimes a few lipid molecules will form a cluster for a few seconds creating specialized areas called membrane rafts. Membrane rafts were discovered only recently and their function is not yet completely understood, although it seems likely they are involved in signalling between cells. There are around 500 different membrane lipids which surround and anchor different proteins that can form channels through the membrane and control the continuous flow of molecules across it.

These channels enable animal cells to maintain an internal concentration of sodium that is one-twentieth of the external concentration. This requires constant pumping to remove sodium which otherwise would raise the osmotic pressure in the cell which, in turn, would draw in water, with the potential to burst the cell. At the same time, potassium is maintained within the cell at a much higher concentration than the external levels, and the same membrane pump brings potassium in at the same rate as sodium is pumped out, an activity that takes about one-third of the total energy of the cell.

Protein molecules at the surface of the membrane also act as receptors for signalling molecules from outside the cell. These messages are then passed down a series of proteins within the cell to the nucleus to switch on genes, if required, in response to altered circumstances. Hormones such as insulin will interact directly with the membrane, allowing sugar to pass into the cell.Virtually everything that goes on in the cell either influences, or is inflflfl uenced by, the activity of the membrane with its 500 types of lipid molecules and up to 10,000 types of membrane proteins.Cells ‘feel’ their immediate surroundings with fififi ne extensions called microvilli (see Figure 3a–c). In some epithelial cells such as those responsible for nutrient uptake from the guts, the membrane becomes fashioned into a brush border, where tightly packed microvilli increase the surface area 30-fold (see Figure 14in Chapter 5). While it is impossible to prioritize various parts of the cell, as they are mutually interdependent, without membranes,independent life could not exist.

Membranes inside cells

Membranes are also crucially important inside cells for two reasons: first, to provide surfaces on which chemical reactions can proceed, and secondly to provide separate areas inside the cell, allowing chemical reactions to proceed which might otherwise interfere with each other. In bacteria, the inner surface of the plasma membrane defi nes the position of everything within the cell and provides attachment points for intracellular contents that need to be in specific positions. Using the analogy of the cell as a factory, the internal membranes provide the workbenches, floors, ceilings, and walls for all the different parts of cell production, with the nucleus centrally positioned as the offi ce in which the information is stored. In small cells, such as bacteria, which are usually rod shaped, the inside of the plasma membrane provides a large surface area in relation to the cellinterior, so that anything that needs a fixed position can be ‘hung’ on the inside, and consequently bacteria and other prokaryotes generally have little or no internal membrane. As mentioned earlier, the internal volume of eukaryote cells is a thousand times that of a bacterium, so that the eukaryotic cell requires a vast internal membrane system around a hundred times the area of the plasma membrane itself. This internal segregation of biochemical activities is crucial as there are hundreds of chemical reactions going on that can seriously interfere with each other. Prokaryotes, with no internal membranes (and eukaryotes to some extent), get round this problem by aggregating groups of specific enzymes into multiprotein complexes, which work as free entities inside the cell. In addition, eukaryotes confi ne different metabolic processes within membrane-bounded compartments. The major internal membrane system in eukaryotic cells is the endoplasmic reticulum (usually shortened to ER), which forms a network throughout the entire cell. This part of the cell is called the cytoplasm (everything inside the cell membrane excluding the nucleus; Figure 5a , b) and everything in it is surrounded by the cytosol, a complex mixture of substances dissolved in water, like a very crowded ‘molecular soup’.

5. Internal cell membranes and structures. (a) The nuclear envelope

(Ne) separates the nucleus from the cytoplasm, which contains other organelles including mitochondria (M) and endoplasmic reticulum (ER).(b) A mitochondrion surrounded by spiral polyribosomes (r) attached to the surface of the ER. (c) A nuclear pore complex, channel arrowed. (d) Golgi bodies (Go) comprised of stacks of membranes

Organelles

Two organelles in the cytoplasm—mitochondria and, in plants,chloroplasts—have double, rather than single membranes. This is most likely a hangover from when they were free-living forms early in cellular evolution. When they were incorporated into a larger cell, their own membrane became surrounded by the cell membrane of their host. Both mitochondria and chloroplasts contain DNA, further evidence that they were once free living.There are two theories as to how chloroplasts and mitochondria became part of eukaryotic cells: they could have ‘invaded’ the eukaryote cell or, alternatively, been engulfed by a larger cell,forming a relationship in which both partners benefit.

The eukaryotic cell provided a ‘safe’ environment, in which the mitochondria generated energy that could be harvested by the host cell, and, in plant cells, chloroplasts produced glucose by photosynthesis. In mitochondria, energy is produced from glucose by a process called oxidative phosphorylation, which occurs on the surface of internal membranes called cristae( Figure 5b ). In chloroplasts, glucose is produced by photosynthetic enzymes in stacks of membranes, called thylakoids ( Figure 4 ).

All other membrane-bound organelles (collectively known as vacuoles or vesicles) have a single bilayer membrane. A typical cell will have around 1000 of these vacuoles, and a similar number of mitochondria. Secretory vesicles contain chemical messengers such as hormones for release from the cell. Endosomes, lysosomes,and peroxisomes (see Figure 2) all contain various mixtures of enzymes, proteins that catalyse chemical reactions. Lysosomes can be likened to the stomach of the cell, as they contain hydrolytic enzymes that break down biological material into its constituent parts to provide food for the cell. Lysosomes can also fuse with phagocytic vacuoles (phagocytes) containing engulfed material such as bacteria, killing and digesting the invading organisms.The Belgian scientist Christian de Duve discovered lysosomes, for which he received the Nobel Prize in 1974. He also discovered peroxisomes, which replicate by division like mitochondria and chloroplasts, but do not have their own DNA. Peroxisomes are involved in a variety of biochemical pathways and contain at least 50 different enzymes. They are important in the breakdown (oxidation) of substances such as fats, providing a major source of metabolic energy in animal, yeast, and plant cells. Because one of the products of oxidation is hydrogen peroxide, which is harmful to the cell, peroxisomes also contain an enzyme called catalase, which breaks down the hydrogen peroxide to water.

Peroxisomes are also sites of synthesis of several enzymes, those in liver cells being responsible for the production of bile. As with most individual organelles, mutations in peroxisome formation have severe consequences, and any severe shortcoming will usually lead to a fertilized egg failing to develop past a few divisions.

The site of protein production for the contents of the various vacuoles such as lysosomes and peroxisomes is at the membranes of the ER. The ER was discovered by three pioneers of electron microscopy—Keith Porter and George Palade in New York and Fritiof Sjostrand in Sweden—in the early 1950s. Electron microscopy allows a 1000-fold increase in detail compared to conventional light microscopy, but imposes difficulties in the preparation of specimens in that an electron beam can only pass through extremely thin sections. In overcoming the difficulties of specimen preparation for electron microscopy, Porter and his colleagues opened up cell structure in a way that was previously unimaginable. At a stroke, indistinct and irregular shadowy lumps from light microscopy were viewed as sharply distinct organelles such as mitochondria ( Figure　5b ). In the words of Don Fawcett, a colleague of Porter and Palade, ‘for morphologists, the decade from 1950 to 1960 held the same anticipation and excitement that attends the opening of a new continent for exploration’. The Fawcett atlases of biological ultrastructure of human tissues are still classics to this day.

Power to the cell

Mitochondria were first isolated biochemically and analysed by Alfred Lehninger in 1949, confi rming the presence of the enzymes required for energy generation by oxidative phosphorylation, a highly efficient process in which nutrients are oxidized to produce adenosine triphosphate (ATP). The energy for doing work, building proteins and moving things around in cells is stored in a molecule of ATP. The energy in ATP is stored in ‘high energy’ phosphate bonds. To cut an involved biochemical process short, this energy comes from the release of electrons in the citric acid cycle within the mitochondrial membrane space, generating ATP synthase, an enzyme which then makes ATP. Energy is released from ATP when the phosphate bonds are hydrolysed (a process where the molecule is split into two parts by the addition of a molecule of water). With the release of energy, ATP is converted to ADP (adenosine diphosphate) which, in turn, is re-converted back to ATP, storing energy again, ready for the next release.

Only three years after Lehninger’s biochemical characterization of mitochondria, Palade’s electron micrographs showed their amazing membrane structure ( Figure　5b ). The order of these discoveries refiects an overall trend that the ‘grind and find’ workings of biochemistry have often generated seminal information on many cell parts in advance of their actual imaging in the electron microscope, although in the microscopist’s view, nothing can compare with seeing what the constituents of the cell look like. The different approaches of the biochemist and biologist can be illustrated as follows. Presume neither had ever seen a wristwatch, but was presented with one for investigation. A few days later the biochemist would report that the watch had been analysed by separating it (grinding it up) into its component parts. This analysis would show that the watch was made from various proportions of copper, brass, steel, and bronze, with maybe a few diamonds. The biologist would hand the watch back intact, having done no more than maybe remove the back, and report that it seemed to house a spring which powered a series of interlocking cog wheels, that drove two arms on the front of the watch that seemed to rotate at a constant speed. While massively oversimplified, this comparison gives an idea of the differences between the analytical approach of the biochemist and the observational approach of the biologist. Fortunately, used together, these have been fruitful indeed in cell biology.

Protein production

Returning to the ER, the majority of ER membranes are covered with ribosomes ( Figure　5a ) and are known as rough ER, whilst the remainder bear no ribosomes (smooth ER). The job of ribosomes is to make (synthesize) proteins from amino acids, holding and joining the amino acids to make peptides, then polypeptides and complete proteins. A series of RNA molecules are involved in protein synthesis. Inside the nucleus, the sequence of nucleotide bases forming the code for a particular protein is first copied from the template DNA in a process called transcription, producing a new molecule of messenger RNA (mRNA). Messenger RNA then passes out of the nucleus, undergoing modifi cation (called splicing) along the way. Once in the cytoplasm, ribosomes bind to messenger RNA, which then acts as a template for the linking together of amino acids into proteins, a process called translation.

The amino acids are brought to the ribosome by short RNA molecules known as transfer RNA. Proteins made on the ER enter the space (lumen) between the ER membranes ( Figure　5b ), where they are folded into a final confi guration before being passed on to other sites such as the Golgi bodies ( Figure　5d ). The Golgi body (or Golgi apparatus) is a stack of fiattened membrane vesicles, where new proteins are packed into vacuoles for distribution throughout the cell, and may also have sugars added in a process known as glycosylation. Newly synthesized proteins undergo stringent quality control and, should they be defective in any way, they are tagged by molecules of ubiquitin for swift degradation. Protein misfolding is very detrimental, leading to disorders such as cystic fibrosis and diabetes. Protein quality control mechanisms may become less effective as we get older, leading to Alzheimer’s disease and other age-related neurodegenerative conditions.

Once synthesized and folded, new proteins need to reach their final destination within the cell, amongst the other billions of protein molecules, constantly being synthesized and degraded. Some proteins may need to pass through one or two membrane barriers before reaching the site where they fulfll their function. In 1971, Günter Blobel and David Sabatini from the Rockerfeller Institute in New York suggested a ‘signal hypothesis’, in which proteins were given a luggage label, or zip code, to ensure they finished up in the right destination. Labelling takes the form of short sequences of amino acids, known as topogenic signals, which then attach to receptor proteins to allow them through membrane barriers to reach the correct destination. In 1999, Günter Blobel received the Nobel Prize for this work, which has explained the molecular mechanisms behind several diseases. Both cystic fibrosis and primary hyperoxaluria (a condition causing kidney stones at an early age) are caused by proteins failing to reach their correct destination. Blobel donated the million dollar prize money to the post-war reconstruction in Dresden, Germany, the country of his birth.

Lipid production

Besides its role in protein synthesis, the ER is a versatile organelle which can both receive and transmit signals and act as a cellular store for calcium, and is also responsible for the synthesis of lipids.Within individual cells, fat is produced at the surface of the ER as tiny individual droplets (lipogenesis). Although fat that we are familiar with around the edge of our steaks and often around our waistlines seems to be in solid homogenous lumps, it all exists within membrane-bound fat droplets in individual cells termed adipocytes (see Figure 3d). Given a continuous intake of nutrients,adipocytes will accumulate more and more lipid droplets, which coalesce with their neighbours to become larger and larger, accounting for the vast majority of the cell volume, which can reach over 100 times ‘normal’ size. Obesity is consequently a disorder of energy balance that results from the continued accumulation of lipid droplets within the adipocytes. At this point we might regret the efficiency of the ER, besides providing for fat storage, the ER also synthesizes enzymes on the smooth ER to break down fat by a process termed intracellular lipid hydrolysis or lipolysis. Thus, a major factor in body weight is the balance between the synthesis and breakdown of lipids in the ER. Considering the health consequences of being overweight, it is surprising that fat at the cellular level has received relatively little attention, with lipid droplets thought of as no more than simple storage depots. However, new studies are showing them to be remarkable organelles, and anything but ‘lumps of fat’. All eukaryotic cells have the ability to make lipids, which produce all the naturally occurring oils and fats—from rapeseed and olive oil in plant cells to milk fats, lanolin, and lard in animal cells. Lipid molecules are concentrated at the surface of the ER, then pinch off forming a droplet (uniquely surrounded by a single lipid monolayer membrane) and remain adjacent to the ER, where the enzymes that catalyse lipid synthesis are located. Mitochondria are closely associated with the sites of lipid production, providing the energy for fat formation. These mitochondria are actually tethered to the surface of the ER by a group of membrane proteins. As more lipid is accumulated, individual droplets fuse with their neighbours, during which the separate membranes merge, sequentially producing ever larger droplets (Figure 3d). During fat breakdown, this process is reversed. Big droplets are fragmented into smaller ones, and enzymes from smooth ER break down the lipid molecules that protrude through the membrane, reducing the size of the droplet from the outside inwards.

Deposition of lipids can also occur within the cells found in the lining of blood vessels, particularly those forming the walls of major arteries. Here accumulation of lipids leads to formation of fatty plaques, resulting in atherosclerosis (hardening of the arteries), which limits blood flow and thus can lead to heart attacks and strokes. At other sites, interruption of blood flow can also produce kidney failure or gangrene. Excessive accumulation of lipids is also a major factor in type two diabetes and hepatic steatosis (fatty liver). Excess consumption of alcohol can cause changes in the way that the liver breaks down and stores fats, leading to more severe conditions such as cirrhosis. Fortunately, the fat droplets can still be broken down in the cell, so the condition is reversible with reduced consumption of alcohol. All this bad news leads to the reasonable question of whether we might not have been better off without fat cells, but they function in response to evolutionary pressures, allowing food storage that may well have helped us survive in times of shortage, and also allowing many other mammals to survive severe winters by hibernation.

Brown fat cells

There is another type of fat cell, termed brown fat cells, in which fat is broken down to generate heat by a process termed thermogenesis. In humans, babies have the most brown fat cells, usually in the shoulder areas. It was thought that brown fat was lost by adulthood in most humans, but in small mammals such as rats and mice, where heat loss is greater (from the increased surface area/volume ratio), brown fat cells are retained throughout life. Overfeeding mice and rats shows that they can burn off excess food intake as heat. Although hibernating animals build up massive stores of white fat to maintain themselves for months without food, their brown fat cells only get switched on at the time of waking to raise body temperature. When a technique called PET (positron emission tomography) scanning became a standard medical imaging technique a few years ago, some patients (wearing gowns only and therefore cold) showed mystery patches of high metabolic activity around the shoulders and back that disappeared in warm conditions. These patches were brown fat deposits, triggered into activity by the cold. Adults do in fact retain brown fat, and some individuals have significantly more than others. A likely reason why some individuals can eat as much as they like without gaining weight is that they have more brown fat. In theory, if we could switch our white fat cells into brown fat cells we could eat as much as we wanted with the end point of being hotter rather than fatter. Brown fat cells themselves differ from white fat cells by having many more mitochondria, with the iron in these mitochondria producing the brown colouration. The normal mitochondrial metabolism which generates energy stored as ATP is altered to produce heat by a process called proton leakage, produced by uncoupling proteins called thermogenins. Workers continuously exposed to cold conditions, such as deep sea divers, appear to accumulate much higher than normal amounts of brown fat deposits, indicating that brown fat can be regenerated in adults. If white fat cells could be converted to brown fat cells (as has been achieved in tissue culture), then we could have a useful tool in the battle against obesity, literally ‘burning off’ excess fat.

Modification of lipid production

In plant cells, genetic manipulation of the mechanism of formation of lipid droplets has staggering implications for seed crops. The biochemistry of production of oils in plant cells follows very similar routes to lipids in animal cells, and one group of enzymes (the diacyl glycerol transferases which catalyse triglyceride production) have already been genetically manipulated to approximately double the yield of oil and oleic acid in maize.

The cytoskeleton

A ‘skeleton’ instantly brings to mind the bony remnants of a long dead individual. The longevity of bone is due to the deposition of minerals such as calcium phosphate into the bone matrix by cells called osteoblasts, creating the structural rigidity. In life, the bones of the human skeleton are held together with tendons and ligaments, flexibly attached to each other as a system of levers, ready to generate movement which is driven by muscle contraction. There are structures with similar functions within cells, where microtubules play the role of levers and the ‘muscle’ activity is provided by actin microfilaments in association with myosin, which slide over each other to provide contraction much as they do in muscle itself. However, unlike the relative rigidity and permanence of our own musculo-skeletal system, the overriding characteristic of the cytoskeleton is one of extreme plasticity and dynamism, where components can be built (polymerized) from building blocks with remarkable speed and just as quickly removed by being broken down (depolymerization). The old-fashioned idea of a cell as essentially a balloon filled with jelly could not be more misleading; cell shape is controlled from within, modulated by signals received from the external environment, and capable of rapid response. Cells continually change their shape, change position relative to their neighbours, move through solid tissues, or take long journeys around the body by entering and exiting the bloodstream. Add to this the reorganization of the entire cytoskeleton required to separate chromosomes at division (as will be described in Chapter 4), and the dynamic nature of the cytoskeleton becomes its main characteristic.

An organized cytoskeleton is a property restricted to eukaryotes, although similar proteins do exist in a rudimentary form in some bacteria. The eukaryotic cytoskeleton is defined as a network of three types of large proteins: microtubules (formed from the smaller protein, tubulin); intermediate filaments (a group of fibrous proteins with similar properties); and microfilaments (formed from the smaller protein, actin) ( Figure　6 a, b). Each of these proteins has many associated proteins to help them fulfll their roles in just about every aspect of cellular function. Although each of the elements of the cytoskeleton provides specific parts of the overall function required to produce shape change and movement, the best way to think about the cytoskeleton is as an integrated system, involving all components together. As well as whole cell responses, the cytoskeleton also plays a crucial role in moving components within cells, where microtubules interact with motor proteins such as dynein, providing ‘railway lines’ for movement of vacuole-bound cargo or organelles throughout the cell.

6. Cytoskeleton components. (a), ( b) The network of fibres making up the cytoskeleton of intact cells, exposed by removal of cytoplasmic organelles leaving just the central nucleus. (c) Microtubules, that have been assembled in a test tube. (d) Section through a flagellum, showing the 9+2 arrangement of the axoneme. (e) Section through a suctorian tentacle, showing the microtubular arrays which surround the food canal

Whether or not fibrous proteins permeate the nucleus to create an equivalent structure to the cytoskeleton has been controversial. Because the cytoskeleton is so obviously crucial to cytoplasmic organization, it is surprising that there has been such a resistance to an equivalent structure in the nucleus. The protein actin, one of the main cytoskeletal components, was first isolated from muscle over 70 years ago. Now, ‘non-muscle’ actin is accepted as a routine component of cytoplasm, and is in fact the most common protein in the cell. More recently still, actin has also become accepted as a constituent of the nucleus, forming part of a ‘nucleoskeletal’ arrangement of long filamentous proteins along with intermediate filaments of the nuclear lamina which provide a fibrous scaffolding for the arrangement of nuclear contents, forming the ‘nucleoskeleton’ (see Figure 7c in Chapter 3).

Cilia and flagella

Whip-like ‘tails’ have been observed on single cells from the earliest days of light microscopy in the late 17th century. Usually one or two such tails (flagella) move the cell through an aqueous medium by propagation of a series of waves from base to tip. In the epithelial tissues lining some organs such as the lungs, cells are covered in numerous flagella (called cilia in large numbers) which move a surface layer of mucus. In the windpipe, cilia beat together in a to and fro motion, producing a constantly moving layer of mucus upwards towards the larynx, thus preventing any accumulation of potential infective agents in the respiratory tract.

Some bacteria also possess flagella, but they are relatively simple, consisting of a rigid helical tail, which acts like a propeller, rotated at its base by a molecular motor. Eukaryotic cilia and flagella are rooted within the cell by a structure called a basal body, and generate their whip-like movement within the length of the flagellum itself by a system of microtubules, organized into a structure known as an axoneme. To quote Don Fawcett, in his classic 1961 work　The Cell, ‘few cellular activities have proved more fascinating to cytologists than ciliary and flagellar motion’. In 1887, Jensen squashed sperm flagella between a microscope slide and cover glass, describing how the sperm tails were ‘frayed into a number of fibrils’, some 60 years before this was confi rmed by electron microscopy. Irene Manton, an English botanist who had managed to get an early electron microscope on the post-war ‘Lend Lease’ system from the USA, showed that there were 11 fibres in all plant flagella, matching those in animals, confi rming that flagellar structure has been spectacularly conserved throughout evolution. The ‘standard’ axoneme structure consists of a central pair of microtubules surrounded by nine peripheral tubules ( Figure　6d ). Within the cell, the basal body is formed by a short cylinder of nine triplet microtubules without a central pair.

What makes a microtubule?

Each microtubule is a hollow tube with the wall made out of a protein called tubulin. Two molecules of tubulin form a dimer, which resembles a shelled peanut. These dimers join end to end making a long filament (protofilament), and 13 protofilaments are joined lengthwise to form the wall of the hollow tube that makes the microtubule ( Figure　6c ). The whole structure is stabilized by associated proteins. In the axoneme of flagella and cilia, movement is produced by a motor protein called dynein which links adjacent microtubules and allows them to slide over each other in a synchronized manner to produce a bend that travels down the flagellum, creating the ‘whiplash’ movement. We now know that the links between the dynein arms and adjacent tubules, which were discovered in the 1950s by Bjorn Afzelius, are successively made and broken, rather like climbing a rope hand over hand.

Should flagellar dynein be mutated or absent, then the consequences are significant. Twenty-five years after his initial discovery of the dynein links, Afzelius looked at the sperm of four patients at an infertility clinic in Sweden, and found that the dynein arms were absent in the sperm tail axonemes so that the sperm were ‘non swimmers’, which, not surprisingly, led to the infertility. Half of the patients also suffered from a condition known as　situs inversus , where the major organs of the viscera (heart, spleen, and pancreas), which are normally on the left side of the body, become switched to the right. This turned out to result from a lack of functioning cilia early in embryo development, when the left–right body axis is established. This condition is called Kartagener’s syndrome, after Manes Kartagener who described the condition in the 1930s.

A particular type of cilium is found in every cell. These ‘primary cilia’ act as sensory structures—rather like a radio aerial for collecting information from the immediate surroundings. They cannot move independently, as they lack both the central pair of microtubules and dynein links between the peripheral nine tubules. Primary cilia are now known to have a whole host of functions, acting as receptors for both mechanical and chemical stimuli. In the lining of the nose, modified primary cilia connect the receptors in the specialized cells of the olfactory epithelium in the nose (dendritic knobs) where smell is perceived. In the eye, the specialized photoreceptors of the retina are attached to their cell bodies by a primary cilium. The primary cilium also plays a controlling role in cell division, and is almost certainly involved in cell locomotion.

Diseases caused by defective cilia are known as ciliopathies, and they include a wide range of symptoms, often recognized and categorized as separate syndromes long before the underlying common cellular cause was identified. Some symptoms may be common to all patients, while others are unique. Patients with oral-facial-digital syndrome suffer from polydactyly (extra fingers and toes) and kidney problems. Patients with Bardet–Biedl syndrome (first identified in the late 19th century) also have kidney problems but in addition suffer from retinal degeneration which can lead to blindness, along with obesity and diabetes—all as a result of ineffective cilia.

Intracellular microtubules

Microtubules were thought to be limited to axonemes until improvements in electron microscopy preparation the early 1960s resulted in their discovery throughout the cytoplasm. Because they always appeared as straight rods, they were initially thought of as rigid and persistent structures. However, Lewis Tilney and Keith Porter showed that by merely cooling a protozoan called Actinosphaerium to around 4 degrees centrigrade, all the microtubule-supported cell extensions collapsed as the microtubules broke apart, subsequently re-forming after a few minutes at room temperature. Not until the 1980s did it become apparent just how dynamic microtubules actually were, when Tim Mitchison showed that microtubules could essentially collapse and re-form in seconds, a process that he termed ‘dynamic instability’.

Microtubules also form the framework of the mitotic spindle, by which the chromosomes are distributed to daughter cells at division (see Chapter 4). By exposing dividing cells to a drug called colchicine (which binds to tubulin and stops it joining together to form filaments), the formation of the mitotic spindle can be inhibited, ‘freezing’ the process of division, and accumulating cells for chromosome analysis. Colchicine was the active ingredient in extracts from the autumn crocus　(Colchicum autumnale ) first used by the ancient Egyptians for arthritic conditions. Inhibition of mitotic spindle formation can also be achieved by drugs that stabilize cytoplasmic microtubules, preventing them from breaking down in order to re-form as spindle microtubules. One such drug is Taxol (extracted from the bark of the Pacific yew). Taxol became a potential blockbuster drug in cancer treatment, and because removal of the bark kills the tree, demand for the bark almost caused the loss of all Pacific yew trees in the USA. Fortunately, Taxol was subsequently chemically synthesized as paclitaxel. Due to the accelerated rate of division of cancer cells, almost any drugs that interfere with microtubules and spindle formation are potential cancer treatments.

Cultured fibroblasts have been the cells of choice for the study of microtubule function. Fibroblasts are found in connective tissue such as joints, ligaments, and tendons. Fibroblasts grown in culture are long and fiattened, and move around the surface of the culture dish with a broad leading edge and a narrow trailing edge ( Figure　3c ). In contrast, cells from epithelia stay fiattened and many-sided in culture ( Figure　3b ). In all cultured cells, cytoplasmic microtubules radiate outwards in the cytoplasm from a structure close to the nucleus called the centrosome. Centrosomes act as a microtubule organizing centre, controlling the turnover and distribution of microtubules. They contain structures (centrioles) identical to the basal bodies found at the base of each flagellum or cilium. These centrioles occur in pairs, positioned at right angles to each other. Early in cell division they separate and migrate to opposite ends of the cell to organize the microtubules that make up the mitotic spindle.

It is a short technical step from growing cells in plastic flasks to providing a suitable environment (a chamber at 37 degrees centigrade), allowing the flask to be placed on a microscope stage for living cells to be observed as they go about their business. Because living cells are largely transparent, it is hard to see much detail without various optical systems such as phase contrast microscopy, which convert small differences in the refractile properties in the cell components into light and dark regions. For this crucial advance, Frits Zernicke received the Nobel Prize in 1953. Nowadays, virtually any protein can be ‘tagged’ to fluoresce (when illuminated with UV light) by combining its genes with those of a protein called green fluorescent protein (GFP, proper name aequorin), originally extracted from a ‘glow in the dark’ jellyfish. By changing its sequence of amino acids, GFP has since had its fluorescence properties altered and is available in blue, orange, yellow, and red fluorescent varieties, allowing several different proteins to be followed over time in the same living cell. Add to this the ability of low-light cameras to capture signals from just a few molecules per cell, together with laser illumination and computerized imaging and analysis, and observing living cells currently provides a wealth of information that was unimaginable only a few years ago. Nowadays it is feasible to watch a particular living cell process over time in the light microscope, then ‘flash freeze’ the cell of interest in milliseconds and prepare it for examination by electron microscopy. Tiny probes called quantum dots, which are both fluorescent for light microscopy and electron dense for electron microscopy, allow labelling of the same molecules for both techniques.

An initial brief look at most living cells under the phase contrast microscope might be disappointing to the uninitiated, as not a great deal appears to be going on. Unicellular organisms will zoom around in real time, driven by their flagella or cilia, and amoebae crawl at a speed easy to see in real time. For cells in culture, the cellular activity seen in popular science programmes will invariably be time-lapse footage, with individual images recorded at intervals of a few seconds and then played speeded up. In this way, the hour a cell takes to divide is recorded with one image every ten seconds. Playing the images back at 25 frames per second compresses the process into just under ten seconds, making it appear much more dynamic.

Around the mid 1970s, time-lapse microscopy showed a movement inside cells which seemed unusual. As well as the continuous and random Brownian motion of the cytoplasm, there was a distinct stop/go movement, where a particle suddenly moved several micrometres across the cell, often stopping and moving on again. This ‘saltatory’ (related to leaping) movement surprisingly took place in straight lines, as though on rails. Saltatory movement was disrupted in the presence of cold or colchicine, conditions which break down microtubules, but not in the presence of paclitaxel, which stabilizes microtubules. Clearly, intact microtubules were acting like guide rails for vacuoles to travel along. Just how the material was driven along the microtubules was not discovered until the mid 1990s, when the motor proteins responsible for the movement were characterized. One such motor protein, kinesin, is shaped like an inverted ‘Y’ so that it has the molecular equivalent of two legs, and literally ‘walks along’ the microtubule, rather like a tightrope walker holding a large balloon (the attached vacuole) above his head. The cytoplasmic form of dynein (which drives flagellar microtubules past each other) works in a very similar way. The energy for both molecules is provided by the conversion of ATP to ADP. Each step is 16 nanometres, requiring 62 steps per micrometre of travel, and several micrometres (halfway across the cell) can be covered in a few minutes. The molecular interactions between kinesin and microtubules have been determined by resolving molecular detail in the electron microscope, helped by technology that permits instantaneous freezing of the cell, retaining molecular arrangements exactly as they were in life. Interactions of kinesin and microtubules are beautifully illustrated by molecular animations available on the Web (see Further reading).

Intermediate filaments

The requirements of a motile lifestyle in animals have resulted in the creation of mechanical strength by completely different ways. Whole organisms make a skeleton, either as a shell, as in the exoskeletons of insects and crustaceans, or an internal skeleton as in fish, amphibians, reptiles, birds, and mammals. In all cases, the skeletal material is made of proteins secreted by cells, and in some cases mineralized for further rigidity. This material is known as the extracellular matrix. Within individual animal cells,mechanical strength is provided by a group of proteins called intermediate filaments, high-tensile yetflexible cables which permeate the entire cell. Cells in tissues are joined to their neighbours by strengthened membrane junctions called desmosomes, which are anchored by intermediate filaments, providing a tensile network throughout the tissue.

Intermediate filaments are so called on account of their diameter (10 nanometres) which is intermediate between microfilaments (6 nanometres) and microtubules (25 nanometres). Although actin filaments and microtubules are the same in every cell (including plants), and intermediate filaments (lamins) in the nucleus are standard, cytoplasmic intermediate filaments are specialized according to their tissue type and embryonic origin. Cells in connective tissues are characterized by an intermediate filament called vimentin, whereas neurofilaments are found in nerve tissue, and desmin is characteristic of vertebrate muscle. Keratins are a large group of intermediate filaments found in epithelial cells, forming the structural protein of skin. Keratins secreted outside the cell form hair, wool, fingernails, horns, and hooves. These ‘hard’ keratins are stable extracellular secretions and consequently dead, although there are many other cytoplasmic keratins that are dynamic. Mutations in keratins can weaken skin, causing a condition known as epidermolysis bullosa, where even gentle friction can cause skin blistering so severe that it can be life threatening to new-born babies.

Microfilaments

The thinnest filamentous proteins in the cell are called microfilaments. They are around 6 nanometres in diameter, half that of intermediate filaments, and made from the protein actin. Actin is a globular protein (G-actin), but it takes a different form (F-actin) when assembled into filaments. Filamentous actin is then organized into bundles or networks by a host of actin-binding proteins, forming at least 15 different structures in non-muscle cells. The actin story goes back over 60 years, to the 1940s when Albert Szent-Gyorgi established the presence of both actin and myosin in striated muscle. Further work in the 1950s by Andrew Huxley and Hugh Huxley (unrelated) established that when muscle contracts, actin filaments slide over the myosin filaments, an interaction which shortens the muscle, producing force. The conversion of ATP to ADP releases the necessary energy for this molecular interaction to take place. Muscle has a highly organized geometrical molecular architecture, where every myosin molecule is surrounded by a ‘cylinder’ of six actin molecules allowing the molecules to slide over each other. This arrangement is only found in muscle cells, so for many years the possibility that actin and myosin could interact to produce contraction in non-muscle cells seemed unlikely. However, in 1973, Tom Pollard showed that there was more than one type of myosin in non-muscle cells. We now know that there are over 40 different myosins in mammals, and that (along with F-actin) they supply the motile force involved in cell division, cell movement, and the uptake of external material by cells (endocytosis). Actin also has a structural role in the cytoskeleton. Cores of actin filaments, bound together by the protein villin, support the finger-like projections of the cell membrane (filopodia and microvilli).

Cytoskeletal–nuclear interactions

From the interior of the nucleus to the surface of the cell, there are links between just about all of the filamentous proteins. The lamins within the nucleus are a class of intermediate filaments (see Chapter 3), which are joined to the cytoplasmic intermediate filaments by protein bridges that cross the nuclear envelope. All the cytoskeletal elements are joined to each other, with direct protein links (plakins) between intermediate filaments and microtubules, and also between intermediate filaments and the microfilaments that form the third major element of the cytoskeleton. This interconnected protein scaffold of microtubules, intermediate filaments, and microfilaments, all with different properties, functions together to maintain the structural and mechanical integrity of the animal cell and its ability to move (see Chapter 4).

Within a living cell, all three cytoskeletal components work in unison, as might be expected after four billion years of evolution. To explain the components of the cytoskeleton individually is like describing a piston, connecting rod, and crankshaft without mentioning a working engine. In both cases the whole is considerably more than the sum of the working parts.

Tensegrity

Thirty years ago, when Donald Ingber was an undergraduate at Yale University, he was convinced that the view of the cell as a ‘rubber bag filled with jelly’ was somewhat oversimplified. Ingber was intrigued by the revolutionary architecture of Buckminster Fuller in the 1940s, who created a series of robust structures called geodomes (including his own house). Geodomes are constructed from a shell of multiple small rigid triangles without any major supporting structures such as beams or columns. Fuller himself had been infi uenced by the sculptures of Kenneth Snelson, where rigid stainless steel rods appeared to float in thin air, but are actually supported by a system of cables, rather like the rigging on a sailboat, where the mast is kept in place by a balance of tension and compression. The mast itself is rigid in order to resist the compression produced by the tension in the rigging. The structure is robust, and will only fail if the mast buckles or the rigging breaks. This is the principle of tensile integrity or tensegrity, which offers the maximum amount of strength for the minimum expenditure of energy and materials. Ingber reasoned that tensegrity exists in every cell, mediated by rigid microtubules which resist the compression produced by actin and intermediate filaments. Tensegrity thus generates strength within all cell shapes, be they fiattened hexagonal cells found in epithelia, or extended nerve axon cells which may be a metre in length. Tensegrity is at work even when a cell changes its shape, as happens in division. In culture, fiattened or elongated cells will round up at the start of division, then pinch off leaving two spherical daughter cells, which then fiatten and spread. As the edges fiatten, triangles formed by actin fibres are clearly visible around the edge, with six neighbouring triangles forming a hexagon, exactly like the edge of a Buckminster Fuller geodome. The cell then fiattens further, changing shape to a typical extended fibroblast form, forming attachments between the membrane and the base layer termed focal adhesions, before migrating as a single cell. In contrast, cells grown in culture from epithelial tissues will attach to their neighbours and move around as a sheet. As in living tissues, the epithelial cells attach to each other with structures called desmosomes, which are tough plaque-like structures formed by local reinforcement of the cell membrane, and anchored within the cell by intermediate filaments. In skin, a tissue constantly bent or stretched, epidermal cells have multiple desmosomes, and their intermediate filaments are strengthened by numerous keratin filaments ( Figure　6a, b ). When repeated through every cell, this arrangement creates an extremely tough tissue.