Cellular Slime Moulds - Social Amoebae

Above:the life-cycle of the social amoeba Dictyostelium. Each amoeba is only about 10 micrometers long (0.01 mm).

Below: a large single-celled amoeba (about 0.1 mm in diameter) from pond water as seen under a light microscope. Amoebae are single-celled organisms. This one lacks a cell wall and is able to use its cytoskeleton to rapidly change shape freely and crawl along at considerable speed (amoeboid locomotion). It feeds by absorbing and digesting prey organisms (such as bacteria and algae) which it internalizes enclosed inside a bubble-like fluid-filled vesicle (many vesicles and other organelles can be seen within this translucent cell).

Amoebae that form slime molds - social amoebae

Most amoebae live as single-celled organisms in the water and soil, such as the one photographed above, but some amoebae can also form multicellular slime structures. Amoebae are not bacteria, rather they are micro-organisms belonging to a group called the Protoctista. Amoebal cells are typically 10-100 times the diameter of bacterial cells and have the structure typical of animal cells, but they are not animals because animals always form complex multicellular bodies. Some amoebae (called myxamoebae), such as Dictyostelium, will live in the soil as single cells that feed and reproduce for many generations, but if these cells start to run out of food in their neighborhood, then they send chemical signals to one another and the amoebae respond by streaming in long conveys to a common rendezvous. When they arrive, these amoebae do something very strange, they form a multicellular mound or aggregate that piles on new coming cells, getting taller and taller. Eventually the cells at the tip of the mound form a nipple-like protuberance and this takes charge as it is designed to become the 'head' of our new organism. All this happens on a small scale, these mounds are only a few millimeters in diameter.

Above: model amoebae - single-celled shape-shifting organisms.

Above: aggregating amoebae. Thousands of the amoebae move, in pulsatile fashion, toward an aggregation center, forming concentric rings, spirals or streamers (as in this case) depending on cell density. Gradually, a mound forms in the center.

Above: a slime mould grex (rendered with Pov Ray) crawling across a glass slide, leaving a trail of slime behind it.

Above a mound of assembling amoebae on a glass slide, left, and a later stage with a tip, right (Pov-Ray model)

Eventually, this tipped mound falls over and starts crawling around like a slug, with the tip raised up like a snout behind which the rest of the body follows, leaving a trail of slime behind it as it does so. So our single celled creatures have all come together to form a temporary multicellular body! The reason is, that this way they are bigger and so can move faster and further. The job of the snout is to find a suitable place high up in the light and air, from which to release spores into the wind or running rain-water. This slug-like creature is called a grex (or pseudoplasmodium or slug, not to be confused with garden slugs!), and is one to a few millimeters long, which is not bad for something that started out as amoebae one hundredth of a millimeter in diameter! Each grex contains about 100 000 separate amoebae, all encased in slime and working together as a single unit! The grex of Dictyostelium discoideum is white and translucent, but experimenters frequently add coloring agents to make the different cells in the grex apparent.

Above:: when a grex finally finds a suitable place (or runs out of time) it will stop moving, then form a mound which elongates into a relatively long stalk (several millimeters long) with a rounded structure at the tip (the colour and form varies tremendously depending upon species). This structure, called a sporangium (spore capsule) will dry and break open, releasing amoebae in the form of spores, into the wind or rain water to be carried off to new habitats. This lead me to consider the issue that the sporangia are not really tall enough to break through a typical boundary layer of still air, to reach the turbulent layer for
optimum wind dispersal: when we think of toadstools they are typically several cm in height in order to achieve this. However, John Bonner (2009, The social Amoebae: The Biology of Cellular slime Molds, Princeton University Press) notes how the spores remain glued together and apparently do not dry out and separate well for efficient wind dispersal, but they may however be dispersed by rain splash or perhaps by flowing rainwater. (If my calculations are correct, they should be high enough to pass a typical boundary layer in water to reach the mainstream for efficient spore dispersal). Bonner also notes that often they are dispersed by insects; indeed the typical height of 2 mm for a Dictyostelium stalk would be ideally placed to brush against passing insects and other arthropods to which a packet of sticky spores can readily attach.

The spores are dormant cells with tough walls to resist drying out. Hopefully some of the spores will find a suitable place and germinate into single-celled amoebae and live and reproduce happily, until they run out of food that is ... then the cycle will start all over again!It should be noted that the fruiting bodies of different species of cellular slime molds can be strikingly different (there are about 100 known species of cellular slime mold). Some have branches, which may occur as whorls along the main
axis, with each branch ending in a small sporangium. They are very beautiful structures!

In Dictyostelium discoideum about 100 to 100 000 free-living amoeboid cells come together, in times of harsh conditions like starvation, to form one of the most simple animal-like multicellular organisms. A more detailed account of Dictyostelium discoideum is given below.

Cell differentiation and cell types

In Dictyostelium discoideum the cells in the mound, grex and fruiting body, can be distinguished into different types on the basis of morphology and biochemical markers – different cell types develop with different development pathways and end fates. This is differentiation and division of labour, characteristics of multicellular organisms.

Two principal classes of cells develop from the aggregate cells – prespore cells (psp) destined to become spores and prestalk cells (pst) destined to form the stalk. About 80% of the cells in the aggregate will eventually become spores, the remaining 20% become the stalk and attachment/basal disc, sacrificing their lineage in an act of sacrifice to increase the odds of survival of their kin, an example of kin selection, since the amoebae will be more-or-less related to one-another.

The prestalk cells can be divided into several distinct cell types:

1) pstA cells

These occur at the tip of the mound and grex, occupying the anterior-most 1/3 of the prestalk region (accounting for 5-10% of the total cells). They will develop into pstAB cells during culmination. PstA cells have active promoters for the ecmA gene, ecm for extracellular matrix, since this gene encodes an extracellular matrix protein. These cells originally form in random positions in the very early mound but move to the tip during cell-sorting. PstA cells are capable of transdifferentiating into pstO cells.

2) pstO cells

These occur in the anterior of the mound and grex behind the pstA cells, accounting for 5-15% of the cells. They also form originally in random positions within the mound. They are capable of undergoing transdifferentiation into either pstA cells or ALC cells.

3) pstB cells

These initially form scattered around the mound but sort to the base and form an anteroventral band in the grex and form the inner basal disc. They express the ecmB extracellular matrix gene. The pstB cells of the mound will form the basal disc if the gex stage is skipped, otherwise they are removed from the grex and deposited in the slime trail to be replaced by cells from the anterior prespore region which will form the inner basal disc.

4) pstAB cells

These develop from pstA cells when these cells enter the stalk tube and form a core inside the tip of the mound and grex. These cells express both ecmA and ecmB extracellular matrix genes and are destined to become stalk cells.

Non-prestalk cell types:

5) psp cells (prespore cells)

These are destined to become spores, though they may transdifferentiate into ALC cells.

6) ALC (anterior-like cells)

These are intermingled, as small groups, with psp cells and cluster to form the rearguard cells in the grex and culminant. One
subtype of ALC cells are the pstO/ALC cells (ALC/pstO cells) which have active pstO-specific promoters and interchange with pstO cells in the prestalk. Other cell types include ALC/pstA, ALC/pstAB, ALC/pstB cells, according to the cells they most closely resemble biochemically. ALC cells also go on to form the lower cup in the sorus of the fruiting body, situated below the spore mass, and the upper cup above the spore mass. ALC cells also form the outer part of the basal disc. There is lack of clarity or uncertainty in the literature about how readily different cell types can interchange. It appears that cell type is determined by position within the mound or grex and that cells crossing region boundaries may transform into the same type as their neighbors.

Other cell types:

7) Tip-organiser cells

These are pstA cells at the very tip. They possibly act as pacemaker cells by setting the frequency and direction of the cAMP waves that direct the other cells in the mound and grex.

8) Sentinel cells (S cells)

These are cells that seem to specialise in phagocytosis and so may have an immune or nutritive function. Most cells stop feeding in the aggregate and show signs of starvation in the grex, including the formation of autophagic vacuoles as they start to break-down and digest their own non-essential organelle systems.

9) Peripheral layer cells (PLC)

These form a flattened proto-epithelium covering the grex. They don’t seem to be biochemically specialised or genetically differentiated, but rather any cell type finding itself on the surface of the grex will change morphology and form part of the epithelium, perhaps as a result of mechanical forces. These cells are closely joined to one-another by what appear to be junctional contacts, especially in their outermost apical regions where there are no intercellular spaces. The posterior prespore region is covered in flattened squamous cells, elongated with anterior-posterior polarity. The anterior prestalk region is covered by less flattened cells with apical-basal polarity (they have some sort of tight junctions in their apical regions and wider intercellular spaces of 10 nm width basally) and which put-out fine pseudopods (filopodia) that interdigitate between interior cells. It is possible that these filopodia are the sites of material exchange, perhaps taking up materials released by interior cells to fuel secretion of the slime coat, though this is my speculation. The proto-epithelium on the anterior and ventrum are formed from pst cells, those on the dorsum from psp cells.

10) Stalk cells

These develop from prestalk cells and form the stalk and part of the basal disc. They are about 8 micrometres in diameter and develop single large autophagic vacuoles and a single layer of thin cell wall material and then die on terminal differentiation. They are polygonal in outline (probably due to close packing, though they do have thin walls). The walls contain randomly arranged cellulose microfibrils and ecmA/B glycoproteins. Those that are outermost secrete the stalk tube – a cylinder that traverses the sorus at its anterior end and embeds in the basal disc posteriorly. This sheath consists of cellulose microfibrils, with the outermost fibrils being arranged parallel to the long axis of the stalk. Outside the stalk sheath is generally a thin layer of ALC-derived cells which extend to the basal disc to form the outer region of the disc.

Above: a culminant (a mound developing a stalk and becoming a fruiting body).

Stages in Development - The Dictyostelium life-cycle

Free amoebae

The amoebae live free in the soil, eating primarily bacteria by phagocytosis, which they locate by chemotaxis. The bacteria are phagocytosed into endosomal vesicles and processed in the endosome vesicle system and the waste exocytosed, with the whole cycle taking about 90 minutes. They have a contractile vacuole network (CVN) of cisternae (membranous sacs) and interconnecting ducts and large cisternae acting as bladders that periodically expel excess water to the outside by contracting and discharging water through a pore in the plasma membrane. The CVN appears to be a distinctly separate membrane system that does not exchange membrane material with either the plasma membrane or the endoplasmic reticulum or endosomal vesicle system. The CVN functions in osmoregulation, expelling excess water to prevent the cell bursting by osmosis, and also helps regulate calcium ion levels inside the cell. The amoebae of Acytostelium leptosomum feeds on yeast as well as bacteria.

The amoebae, which are usually haploid, divide by mitosis, undergoing asexual binary fission. When food begins to run out, the
amoebae aggregate, if enough aggregate they will form a motile multicellular structure called a grex, however, if only a few cells
aggregate (as occurs, for example, in water suspension) then two of the cells may fuse into a diploid cell, in a process of fertilisation, which phagocytoses the other cells, resulting in one cell with a giant nucleus which secretes a multilayered protective wall around itself and then divides by meiosis and becomes a dormant macrocyst which germinates into haploid amoebae when conditions are once again favourable. This is sexual reproduction. In Dictyostelium discoideum, two different compatible strains are required for fertilisation prior to macrocyst formation, but in Dictyostelium mucoroides, sexual fusion and macrocyst formation occurs between cells of the same strain. Alternatively, some strains form microcysts instead - single cells enclosed by protective cellulose cell walls.

Parasexual Process

Cellular slime mould amoebae may fuse to give binucleate cells, or occasionally diploid cells. These diploid cells may divide for a number of generations, giving rise to diploid cells. However, some genetic exchange can occur between homologous pairs of
chromosomes during this mitosis (similar to crossing-over in meiosis). At some point this diploid state become sunstable and one of each pair of chromosomes is slowly lost over a number of amoeboid generations until haploid cells are formed again, but these cells may have new recombinant chromosomes.

Aggregation

i) cAMP wave propagation and chemotaxis

In Dictyostelium, aggregation occurs by chemotaxis to periodic cyclic AMP (cAMP) signals released from the aggregation centre that propagate as waves. Cyclic AMP or cyclic adenosine monophosphate is a cyclic molecule derived from ATP (adenosine trisphosphate). Cells move chemotactically towards increasing cAMP concentrations leading to aggregation streams and multicellular aggregates.

1. The cAMP is detected by a high affinity receptor, cAR1
2. Which upon binding cAMP couples to a hetero-trimeric G protein
3. Which liberates the beta-gamma complex
4. Which attracts the cytosolic cAMP regulator (CRAC) to the membrane
5. which then activates adenylyl (adenyl or adenylate) cyclise
6. leading to cAMP synthesis
7. cAMP is secreted and binds to the receptor again – autocatlytic feedback
8. Binding of cAMP to the receptor leads to desensitisation
9. cAMP is degraded extracellularly by phosphodiesterases, resensitizing the receptor.

This sequence of events is illustrated in the signal-flow diagram below (this model is based on peer-reviewed literature and is not exhaustive):

Arrows represent excitatory or stimulatory signals, pistils (such as that connecting PKA to Gα) represent inhibitory signals.This signaling circuit represents a tiny fraction of the circuitry inside a cell: living cells are essentially 'robots' with machinery controlled by a nanobrain of interacting signaling molecules. The cAR1 receptor is shown in red and is a protein embedded in the cell-surface membrane (the plasmalemma or 'skin' of the cell, consisting of a bilayer of phospholipids represented in green) and is a member of the GPCR (G-protein coupled receptor) family of proteins, also called serpentine or heptahelical protein family because the polypeptide chain that makes up the protein has seven helices that cross the membrane as rods (transmembrane domains). Note that everything above the green membrane is outside the cell, everything beneath it is inside the cell. Specifically GPCRs are heptahelical proteins that couple to trimeric G-proteins Gαβγ, which consist of three polypeptide subunits, α, β and γ. These G-proteins are anchored to the cell membrane and when cAMP activates cAR1, cAR1 activates the G-protein, causing a GDP molecule attached to the alpha-subunit to be replaced by a GTP molecule, causing the alpha subunit to detach from the beta-gamma subunits. The alpha subunit then goes on to activate a protein called phospholipase C (PLC) which converts the membrane phospholipid PIP2 (phopsphoinositide biphosphate) into IP3 (inositol triphosphate) and DAG (diacylglycerol). The DAG remains anchored in the membrane but the small, water-soluble IP3 molecule can activate proteins called calcium channels that then open, allowing calcium to enter the cytosol (the fluid around the organelles in the cell's cytoplasm) increasing the concentration of calcium ions - [Ca²⁺] square brackets indicates concentration of - inside the cell. The calcium ions bind to and activate the calcium-sensing protein calmodulin (CaM) which then binds to and activates the protein adenylate cyclase (ACA) which converts molecules of ATP (adenosine triphosphate, the same molecule that acts as an energy store in cells) into the signaling molecule cAMP (cyclic adenosine monophosphate). This cAMP is both secreted by the cell (through a transporter protein indicated by the yellow cylinder), passing the signal to other cells in the neighborhood, and also activate the protein PKC (protein kinase C). Note that the beta-gamma component of the G-protein can also activate ACA.These signaling motifs, such as the GPCR / trimeric G-protein sensor / transducer, PIP2, calmodulin and ACA are common motifs found not just in Dictyostelium but mammals including humans too. In humans these and other signaling mechanisms are critical in health and disease.

When analysing signal-flow diagrams it is important to note the overall effects of the circuit.This circuit essentially amplifies the cAMP signal by causing the cell to make and secrete more cAMP, an example of a positive feedback loop. Conversely, a negative feedback loop in which PKA deactivates the Gα subunit acts to switch off the response. G-proteins are so-called because the alpha subunit in this case degrades GTP to GDP after a certain period of time. This would cause the alpha subunit to switch itself eventually, once the GTP is converted (hydrolyzed) to GDP the alpha subunit will recombine with a beta-gamma pair and the G-protein becomes inactive again. PKA accelerates this switching off of the response. What a signal-flow diagram does not show is the timings of these various signals. The response is first amplified and then switched off after a certain delay. The result is that cAMP signaling will oscillate. Such oscillations are a common result of alternating positive and negative feedbacks.

The results are periodic oscillations of cAMP. Cells undergoing chemotaxis are elongated. The waves may appear as expanding spirals or concentric ring waves (as predicted by mathematical models using reaction-diffusion equations). Accumulation of cells speeds up wave propagation. This locally distorts the wave front leading to the formation of bifurcating aggregation streams – that is branching streams of migrating cells that converge on a central position. Eventually a mound of cells develops in the center where the cells accumulate.

Migrating cells thus secrete cAMP, which is sensed by other migrating cells by binding to the cAR1 receptor (expressed during early development) which are thus stimulated to synthesise and secrete more cAMP to relay the signal to other cells. This sets-up periodic pulses or waves of cAMP, peeking in the nM range of concentrations, radiating away from the aggregation centre to which the cells are converging. (Such a center will be set-up once several cells move close together). The binding of cAMP to the receptor stimulates both adenylyl cyclase and guanylyl cyclase. Adenylyl cyclase is not essential for chemotaxis but is essential for aggregation. Guanylyl cyclase is essential for chemotaxis. There are a number of cAR receptors (cAR1 to cAR4) but cAR1 is the high-affinity receptor active early on.

Adenylyl cyclase is a membrane protein with 12 membrane spanning helices and several regions extending below the membrane into the cytosol, inclusing a catalytic site where ATP binds and is converted into cAMP. Adenylyl cyclase is regulated by G proteins, both stimulatory (Gs) and inhibitory (Gi). Guanylyl cyclase manufactures cyclic gunaine monophosphate, cGMP.

Above: the signaling pathway (simplified) activating cell movement in Dictyostelium (model based on peer-reviewed literature). Note that ACA, Ras, PIP2 and PIP3 are all anchored to the cell membrane, but that this is not always shown in signal-flow diagrams. If all the cAMP signal did was cause more cAMP to be made and secreted in waves, then it would be of little use! The oscillating cAMP wave causes the Dictyosteium amoebae to locomote by crawling towards the region of highest cAMP concentration, causing the cells to aggregate to form a mound and subsequent grex. The cAMP achieves this by activating the engine of the cell: its cytoskeleton which then moves the cell along. Again this is through the cAR GPCR with a trimeric G-protein acting as a signal transducer. This also results in activation of the proteins Ras and GC (guanylate cyclase). Guanylate cyclase converts GTP (guanosine triphosphate) into the signaling molecule cGMP (cyclic guanosine monophosphate). Ras is an example of a small G-protein. Small G-proteins are rather similar to the alpha subunit of the large or trimeric G-protein that we have already discussed. Small G-proteins (and the alpha subunit / beta-gamma-subunits of large G-proteins) function as binary switches, with two stable states we can often think of as ON and OFF.

The G-protein is in the ON (1) or active state generally when bound to GTP. They are activated when proteins called GEFs (Guanine-nucleotide Exchange Factors) remove the GDP and replace it with a GTP molecule. The G-protein eventually reverts to the off-state when it slowly hydrolyses its attached GTP to GDP + Pi (Pi = inorganic phosphate), a process that can be accelerated by proteins called GAPs (GTPase Activating Proteins) which activate the intrinsic GTPase activity of the G-protein, accelerating hydrolysis of the GTP to GDP. Small G-proteins occur in families (as do most proteins). protein families result from evolution by mutation of the gene encoding a protein. this is one way that Nature achieves the bewildering complexity seen in living organisms: by duplication followed by modification. For example, there are at least 36 Ras genes in humans and 14 in Dictyostelium, all of them differing in some important way. When a gene becomes duplicated, it can mutate more freely as long as the original maintains vital function, until a useful variation in function is obtained. As genomes consist of such families of duplicated and mutated genes, so organisms consist of duplicated body segments, duplicated limbs and duplicated cells, which also vary in form and function. Nature achieves complexity by this modular approach that facilitates evolution by mutation and natural selection. Cell signaling pathways do indeed form complicated networks, making up the nanobrain of the cell. In Dictyostelium, RasC and RasG are activated by cAMP.

There are several families of small G-proteins, for example the Ras, Rho, Rab, Arf and Ran families. Each tends to have different functions, though all act as binary switches. The Rho family (including Rho and Rac and a few others) is involved in controlling the cytoskeleton and hence movement of the cell. G-proteins act as signal transducers, converting the signal from the ligand binding to the receptor to a signal to other systems within the cell, such as the cytoskeleton.

Notice that our signals result in the activation of another key protein, PKB (Protein Kinase B) also called Akt (AKT). As a kinase this protein is an enzyme that phosphorylates downstream proteins (i.e. it adds an inorganic phosphate group to them, the phosphate coming from ATP). Adding a phosphate to a protein releases some energy as the bond is formed (as does adding a ligand like cAMP to the CAR receptor). It is this energy that brings about the conformational change in the target protein, causing it to move and adopt a new shape that can expose business regions of the protein: the protein switches to an active higher energy ON state.

Cell polarization

Cell polarization is a process whereby a roughly spherical cell changes shape to adopt a front end and a rear end. This is necessary if the amoeba is to crawl towards the sources of the cAMP waves.The formation of PIP3 is central to this: the PIP3 is produced in the region of the membrane near to where the original cAMP signal was received. Proteins that bind to PIP3 and activate have a PH-domain. Proteins are modular, being made up of structural features called motifs and functional regions called domains. A domain is a region of a protein that forms a particular function. Modules can, over the course of evolution, be added to or removed from proteins or adapted for other functions or swapped between proteins.

Once again, modularity allows Nature to build complexity through evolution in a much easier way. This is why arguments that attempt to calculate the likelihood of complex life evolving 'by chance' by assuming each component to be unique are fallacious. Instead Nature copies, reuses and modifies components at different levels of organization. This makes evolution very efficient.

PKB has a PH-domain and binds PIP3 at the cell membrane. Thus, formation of PIP3 recruits PKB to the cell membrane in the region where the cAMP signal was detected. Proteins that interact with PKB can subsequently be recruited to the membrane-bound and activated PKB to be phosphorylated by PKB's kinase domain. This results in a chain of signaling that ultimately activates the cytoskeleton, polarizing the cell by giving it a 'front' or leading edge that is powered by the cytoskeleton to form a locomotive protuberance called a pseudopod (specifically a lamellipod when on a flat surface) causing the amoeba to crawl towards the source of the cAMP. The amoeba can turn as it goes, or reabsorb the lamellipod and produce one in another direction should the bearing towards the cAMP signal change. Eventually the amoebae crawl together and form a mound of cells.

Above: PKB is activated when its PH domain binds to PIP3 on the interior surface of the cell membrane (combined with phosphorylation by other kinases) changing its conformation and exposing the business kinase domain.

The importance of these cell signaling networks

Dictyostelium is of course worth studying in its own right, as an organism that straddles the boundary between being unicellular and multicellular. However, Dictyostelium is also a useful model organism, allowing scientists to investigate cell signaling systems in a convenient way and learn more about the similar signaling systems that are so central to human cell biology and disease. The similarities between the systems described here and those in human cancer cells, for example, are striking. Mutations in Ras signal transducers are important in about 30% of human cancers. A chief characteristic of cancer cells is their mobility, they can crawl from one part of the body to another (they can crawl into the bloodstream and then out again at a remote location). This is the process of metastasis that causes cancers to spread around the body. This involves cell polarization and locomotion, but in the absence of an appropriate signal to do so.

Why an oscillating signal?

Each amoeba emits periodic pulses of cAMP in response to pulses detected from other cells. This tends to have a synchronizing effect and all the individual waves of cAMP, each produced by a single cell, become synchronised and summate by constructive superposition (a property of waves) into a single larger wave. It is true that summations of many smaller waves can give rise to a larger rhythm, however the synchronization need not be perfect which may result in more complex wave patterns which possibly explains observations that the wave pulses occasionally cease for a time and become continuous (as noted in Bonner, 2009).

Some have suggested that the periodic pulse acts as a pacemaker to synchronize cell activity. This is a possibility, if the cells are timing longer term changes, such as switching genes on for later fruiting body synthesis. Developing multicellular organisms do often use timed signals to synchronize their activity. There is growing evidence that some deep sea sponges may be using light signals to synchronise cell activity across the sponge body during growth and wound repair. However, I am not currently aware of any research showing the importance of a periodic pacemaker signal in Dictyostelium.

Nevertheless, the tendency, in Dictyostelium discoideum at least, is for a periodic signal to be generated. It has been pointed out that the amoebae could, in principle, be attracted to a continuously emitted signal set up as a continuous concentration gradient, growing in concentration towards the focus of aggregation, in much the same way as human neutrophils may chemotact along a gradient towards a group of bacteria. However, there are advantages to using a periodic signal: first of all it is less demanding on resources to emit a periodic pulse of cAMP than to secrete cAMP constantly. Additionally, it is possible that the cells reach maximum sensitivity at the anticipated time of the next pulse, perhaps whilst secretion of their own signal is minimum, and so are in an optimal 'listening' mode. By alternating signal emission and signal 'listening' the cells could be minimizing response to their own signal and maximizing their response to the signals of others. Signalling systems generally show adaptation or habituation. Consider walking into a room with a distinctive odor (perhaps somebody forgot to take the trash out) at first the aroma is obvious, but after a few moments you tend not to sense it, unless you walk out of the room and re-enter it after a few minutes at which point the smell hits you again! Your olfactory system adapts and ceases to respond to a constant 'background' stimulus and is tuned to detect changes in stimulus intensity. Some olfactory receptors are actually GPCRs like cAR1 and it is a general property of GPCR signalling that it adapts. To lure a cell in with a chemical 'bait' one would need either a noticeably increasing concentration of the chemical towards the source, or a periodic signal. A periodical signal allows the receptors to reset to their original sensitivity in between pulses and perhaps maximizes the sensitivity of the system.

ii) Signals in mounds

Strain specific patterns of periodic cAMP signals occur within the mound. Some mounds are organized by single concentric ring
pacemakers, several concentric ring pacemakers, or 1,2,3,5 or multi-armed spirals. These patterns may interchange. Initially the waves propagate fast at low frequency; later frequency increases while speed decreases. All these patterns eventually produce a tip, which protrudes from the top of the mound. Differences and changes in wave patterns may result from alterations in the speed of cAMP production and sensitivity to cAMP (e.g. a switch from cAR1 to the less sensitive cAR2 and cAR3 receptors).

Prestalk (pst) and prespore (psp) cells relay the signal. Prestalk cells originally form at random positions within the mound. Prestalk cells turnover cAMP faster and have the low affinity cAR2 receptor, allowing them to relay the cAMP signal at high amplitude. The result is an accumulation of prestalk cells in the tip, which thus becomes the sole signal-generating center. Prespore cells express cAR3. During this transition, localised groups of cells may form other centers, but ultimately only the center in the tip survives. Periodic microinjection of cAMP into mounds counteracts the endogenous signal and disrupts mound formation. The details are more complex, with prestalk type A (pstA) cells accumulating in the tip and prestalk type B (pstB) cells accumulating in the base of the mound, both these types form initially at random positions in the mound. Type pstA cells express ecmA markers, pstB cells express ecmB. EcmA and EcmB are extracellular matrix proteins.

iii) Cell movement in mounds

Cell movement is directed antiparallel to the direction of wave propagation. The different patterns of cell migration are strain
specific. In spirals, cell movement is counter-rotational and cell movement is several times faster than in concentric ring patterns, where cells may be periodically stationary.

In strain AX3, the cell speed is 10 micrometers/min during aggregation, 50 micrometers/min during mound formation. Cell velocity increases slightly when cells enter the aggregation streams. At the aggregation center, movement slows down and becomes temporarily disordered. Cell movement then suddenly increases in speed and becomes highly ordered and strongly rotational. Movement slows again at the time of tip formation. Regulation of these processes could be due to changes in cAMP receptor expression, changes in the cytoskeleton, cell adhesion and cell-matrix interactions?

iv) Tip formation

The amoebae differentiate into two principal cell types: prestalk (pst) and prespore (psp) cells. The prestalk cells differentiate and sort out to form the tip on the top of the mound (pstA cells at the very tip, followed by pstO cells). The mound extends into the air and contracts at the base. Cell movement in the tip appears to be always rotational. The period suddenly increases from 2 minutes to 4 minutes (switch from cAR1 to cAR2?). Tip formation is cAMP dependent. Low affinity receptors may allow prestalk cells to further respond to cAMP gradients when prespore cells are adapted. Prestalk cells also move faster and are less adhesive, which may favour migration to the center of the cell mass.

v) Arrest in mound stage

Evidence indicates that movement up to the top of the mound requires a high motive force involving both actin and myosin. Mutants with actin and myosin defects are unable to pass the mound stage and culminate. Prestalk cells undergo rotational migration while the cells at the base of the mound undergo periodic upward movement. The tip contracts and the mound elongates to form a standing slug or grex. The grex eventually becomes unstable and topples over. In Acytostelium leptosomum several grexes usually form from a single aggregation center.

vi) Cell movement and signal propagation in slugs

The grex migrates at about one grex body length per hour, or 30 micrometres/min.

Most species produce an internal stalk (as a central rigid chord of cells) continuously during grex migration, but some, including
Dictyostelium discoideum form stalkless migrating slugs. In D. discoideum, prestalk cells (which are situated in the tip which is raised in the air) and especially pstO cells undergo rotational movement around the central axis of the grex at an angle to the direction of grex migration and at an average speed greater than that of the grex. The prespore cells move periodically forward in the direction of grex migration and with the same average velocity as the grex. These movements are probably coordinated by rotating and planar waves of cAMP – rotating (scroll) waves in the tip and planar waves in the main body of the grex. Models indicate that a transition from rotating to planar waves occurs if the prespore cells are less excitable than the prestalk cells. The tip is permanently raised above the substrate and hence prestalk cells do not provide traction, but are involved in coordinating the grex. The tip secretes a slime sheath, which surrounds the grex and may also be secreted into the interior. This slime sheath provides a substrate for the migrating cells. Cells at the top migrate at the same velocity as cells in the bottom of the grex.

PstA cells are formed in the anterior outer prestalk zone; pstO cells at the boundary between prestalk and prespore cells and pstAB cells occur in the central core of the prestalk zone. The core of the rotary wave in the prestalk zone is a region of low cAMP concentration, which activates genes in these cells. Computer simulations show that a reduction in the difference in excitability between prespore and prestalk cells causes the rotary wave to twist upon entering the prespore zone, extending the core throughout the whole grex, which may account for the continuous stalk formation seen in, for example Dictyostelium mucoroides, which deposits a horizontal stem as it crawls, before turning the tip of the stem vertical during culmination (without forming a basal disc). Analysis of cell movements supports this hypothesis. Cell movement is faster in D. mucoroides, but the cells follow spiral trajectories and overall grex migration is slower. The slugs often describe a spiral path. The cytoskeletal protein myosin is known to be important in grex migration.

The model of Odell and Bonner, 1986

Odell and Bonner developed a model of grex migration driven by pulses of cAMP travelling back from the tip. In this model cAMP directs individual cells though a second unidentified chemical signal is required to increase vigor of movement, with the
concentration of this second signal peeking in the inner core of the grex where cells flow fastest. In this model some cells contribute more to grex locomotion than others, though their positions may interchange. The outermost cells, near the surface, move slowly, but deeper layers within the grex involve cells crawling over their outer neighbours at speeds increasing with depth, creating a shear flow. The central cells thus move fastest in an interior fountain motion that contributes most of the thrust. If the grex is stalled, by say encountering an obstacle that prevents forwards movement, then the fountain in the central chord reverses direction. However, with low resistance the surface cells, the slow crawlers, move backwards. The movement is driven by a group of cAMP secreting pacemaker cells in the tip (tip-organiser cells). The flow is predicted to cause the pacemaker to occasionally be displaced backwards, at which point it loses dominance and thus is pushed forwards again, maintaining dominance by dynamic equilibrium. This could account for the observed pulses in the tip of migrating grexes. It would take about 15-45 minutes for cells to move the whole length of the grex from the back to the front.

This model also predicts that most of the traction is not generated by the ventrum of the grex which is in contact with the substratum, supported by observations in which a grex will move over a rough surface, with which it has very little of its surface in contact, with normal speed. However, the slime sheath may be providing traction. A grex can also climb objects and descend gaps within its slime sheath, without touching a solid surface.

Longer grexes move faster

Bonner (2006) points out that longer grexes locomote faster than shorter grexes of the same species. He suggests an analogy of several rows of oars on a boat: the more oars the faster the boat can travel, but says he has had little positive support from the physics community. If we imagine a muscle cell contracting: the muscle consists of a long serial chain of contractile units called sarcomeres. Each sarcomere contracts a small amount, say one micron (one thousandth of a mm) in a second. When connected in series, the distances moved are additive, so 1000 sarcomeres will contract by 1 mm, 50 000 by 5 cm, in one second. Thus the longer muscle cell will contract with the greatest speed. A similar argument applies to muscle lengthening by relaxation. In the case of the grex this kind of coupling can only work if the cells are somehow attached to one-another. We know that they are not rigidly attached, since they can swap position quite freely, however they are loosely cemented together by slime. Thus, I would indeed expect longer grexes to move faster. (A suitable physics analogy would be a series of springs connected in series).

I would also expect thicker grexes to move faster: since each layer of cells moves over the one beneath it, the fastest moving cells should be in the centre (much like fluid flowing through a pipe). With more layers the central cells which reach a higher peak velocity. Indeed, Dictyostelium polycephalum moves about a fifth of the speed of Dictyostelium discoideum for a given length and is a much thinner grex. It is thin and worm-like, which allows it to more easily move through narrow spaces and it crawls through soil much more easily than the grexes of other species. The 'fluid-flow' model in which amoebae in a migrating grex behave rather like particles of fluid flowing through a pipe, may account for this (at least in part).

Structure of the grex

The cell fate map of the grex has already been discussed (see diagram) and so has the structure of the proto-epithelium cells that cover its surface. The anterior of the grex secretes a slime sheath which has a similar composition to the stalk tube, containing EcmA and EcmB glycoproteins. The sheath is added to by other regions of the grex, such that it gets thicker further from the tip. This has been suggested as a mechanical guide to help ensure cells move forwards, since the cells tend to move in the direction of least mechanical resistance and may be constrained by the thicker and stronger sheath posteriorly. The sheath continuously streams off the posterior end of the grex, as a hollow tube which collapses to form the slime trail.

vii) Culmination

Slugs stop migrating in order to culminate, that is produce a vertical fruiting body. Translational movement in the tip ceases, but the rear of the grex continues moving until it is positioned beneath the tip which is pushed vertically upwards. Possibly the grex
effectively migrates up the stiffening rod of vacuolating stalk cells (which eventually form a rigid structure with the stalk tube). The vacuolation, formation of walls and the formation of the stalk sheath by the maturing stalk cells gives the stalk rigidity. The process is summarised as follows:

i.  Tip arrests and orients upwards
ii. The posterior moves forwards until it is beneath the prestalk tip
iii. The pst cells form the stalk tube (extracellular matrix)
iv. Cells at the apex (pst cells) move into the stalk tube anteriorly, depositing matrix, vacuolating and dying
v.  Stalk elongation continues until more-or-less all pst cells are incorporated into the stalk.

Cell movement in the tip is still rotational. The prestalk cells in the tip form the stalk. PstB cells form the inner basal disc and
rearguard cells, which develop from anterior-like cells (ACL) in the back (prespore region) of the slug form the outer basal disc and lower cup. A second signalling centre seems to appear at the base of the culminate. The nature of this secondary signal is unclear. At least some of the rearguard cells can move forward rapidly to form a pile of cells at the prespore-prestalk boundary, the upper cup, joined by cells from other locations. This aggregation seems to be under the control of a second signalling centre. During culmination prespore cells move over the pile, which ends up in the back of the slug. These cells then undergo rotational migration.

Most of the ALC cells will move up with the prespore mass to form the lower cup, others remain to form the outer part of the basal disc. In Acytostelium leptosomum the stalk is a narrow cellulose tube containing no cells at maturity and produces spherical spores. The fruiting body of Distyostelium discoideum is normally yellow in colour.

Sporulation

About 80% of the total cells form unicellular spores. The spores are elongated ovoids and have thick three-layered walls secreted by each cell. The inner layer is cellulose, the outer and inner layers are glycoproteins, with 10 major glycoproteins contributing. The outer glycoprotein layer is loose and easily removed, but the inner layer is covalently cross-linked. These glycoproteins are stored in prespore vesicles (PSVs) and the prespore cells mature more-or-less in synchrony, secreting their spore coats as the PSVs fuse with the cell membrane in exocytosis. The spores are resistant to desiccation and temperature extremes. The spores are released when the sorus dries out and ruptures. The whole point of the life-cycle is to release the spores above the boundary layer where they can be better dispersed. If the grex found a good place of elevation, then the height for the stalk is sufficient to clear the most stagnant region of the boundary layer.

Directional movement of migrating slugs - taxis

The cell aggregate undergoes tip formation and upwards extension to form standing ‘fingers’ that fall over and crawl away as slugs. Slug migration may last from one hour to nearly two weeks, depending on strain and conditions like humidity and osmolarity. Average grex speeds are 0.2 to 2.0 mm/h (about 30 micrometres/min). The slugs are phototactic (moving towards light), thermotactic, rheotactic (moving towards wind currents) and acidotactic (moving in response to pH gradients). These processes help the grex find its way out of the soil or other substrate and climb a suitably elevated object for spore dispersal.

In thermotaxis slugs migrate towards warmth with maximum accuracy at temperatures close to the optimum growth temperature. At higher and lower temperatures the accuracy of orientation declines, until, at several degrees either side of optimum, there is a transition to negative thermotaxis. Sign reversals in phototaxis cause slugs to migrate at an angle either side of the direction of the light source. Those whose direction of travel is at an angle less than the critical angle turn away from the light, while those travelling at a greater angle turn towards the light. In wildtype slugs under most conditions the critical angle is sufficiently small that bi-directional phototaxis is indistinguishable from unidirectional phototaxis.

Maree et al. devised a model of phototaxis, supported by empirical data, in 1999. In this model, the translucent cells of the grex act as a lens, focusing light falling onto the tip at one side on to the cells in the tip at the far-side. (This is supported by experiment). Light striking these cells causes a change in the cAMP waves generated by the pacemaker cells in the tip, with the waves tilting in one direction, causing the cells in the grex to turn in response to the new gradient. This contrasts with models in which individual cells sense the direction of the light. In this model, the light focused on the far cells causes them to produce more ammonia, which inhibits cAMP production by the cells on this side. This is supported by empirical evidence, in which an equal concentration of ammonia around the grex inhibits phototaxis and the tendency of the grex to avoid ammonia. Ammonia and light have no observable effect on cell speed, so the idea is that they affect the shape of the cAMP waves.

Key points:

• The slug tip controls slug behaviour.
  
• Photosensory and thermosensory signal transduction pathways converge early and share most components.
  
• Slug turning is mediated by transient lateral shifts in slug tip position. Light intensity and temperature gradients across the slug tip cause turning responses by altering the balance between tip activation and inhibition.
  
• Tip activation signals are carried by cAMP waves (scroll-shaped), while tip inhibition signals may be borne by one or more of the molecules: ammonia, adenosine and Slug Turning Factor (STF). Inhibitory signals may influence cAMP-wave pacemaker frequency.
  
• Signal transduction components: cAMP, inositol polyphosphates, cGMP, calcium ions, and cytoskeletal proteins. Mitochondria are also involved.
  

Bibliography & References

1. Rieu, J.-P., K. Tsuchiys, S. Sawai and Y. Maeda, 2003. Cell movements and traction forces during the migration of 2-dimensional Dictyostelium slugs. J. Biol. Phys. 29: SN1-SN4.
2. Dormann, D., B. Vasiev and C.J. Weijer, 2002. Becoming multicellular by aggregation; the morphogenesis of the social amoebae Dictyostelium discoideum. J. Biological Physics, 28:765-780.
3. Maree, A.F.M. and P. Hogeweg, 2001. How amoeboids self-organize into a fruiting body: Multicellular coordination in
4. Dictyostelium discoideum. PNAS, 98: 3879–3883.
5. Maree, A.F.M., A.V. Panfilov and P. Hogeweg, 1999. Phototaxis during the slug stage of dictyostelium discoideum: a model study. Proc. R. Soc. Lond. B, 266:1351-1360.
6. Fuchs, M., M.K. Jones and K.L. Williams, 1993. Characterisation of an epithelium-like layer of cells in the multicellular
7. Dictyostelium discoideum slug. J. Cell Sci. 105:243-253.
8. Odell, G.M. and J.T. Bonner, 1986. How the Dictyostelium discoideum grex crawls. Phil. Trans. R. Soc. Lond. B, 312: 487-525.
9. Ashworth, J.M. and J. Dee, 1975. The biology of slime moulds. The Institute of Biology's Studies in Biology no. 56. Edward Arnold (pub).
10. Garrod, D.R. 1974. The cellular basis of movement of the migrating grex of the slime mould Dictyostelium discoideum:
11. chemotactic and reaggregation behaviour of grex cells. Embryol. exp. Morph. Vol. 32: 57-6.
12. Garrod, D.R.  1969. The Cellular basis of movement of the migrating grex of the slime mould Dictyostelium discoideum. J. Cell Sci. 4: 781-798.

The following charming book is full of amazing facts about cellular slime molds and will appeal to anyone with an interest in cellular slime molds, from lay person to the professional biologist. It lacks many of the intricate details of cell signalling networks, but gives a fascinating and rich account of the behavior and general biology of cellular slime molds and is packed full of fascinating facts:

    Bonner, J.T. 2009. The Social Amoebae: The biology of cellular slime molds. Princeton University Press.

An excellent textbook on cell signaling in general is:

    Marks, F., Klingmüller, U. and Müller-Decker, K. 2009. Cellular Signal processing - An introduction to the molecular mechanisms of signal transduction. Garland Science, Taylor & Francis Group.

Article updated: 17 Sep 2017, 6 April 2022, 13 April 2022