Cell Locomotion

Cells move for a variety of reasons.  Bacteria swim by beating their flagella in order to exploit newly created micro-environments and amoeba crawl to gather bacteria to feed on.  The cells of vertebrates must move to heal wounds, fend off invaders and the eukaryotic flagella is used to propel sperm cells toward eggs.

Table 1 Speed record of different types of cell locomotion.

The Prokaryotic Flagellum

Bacteria invented the wheel! The bacterial flagellum is a helical structure that drives the cell through the media like a propeller.  The structure is rigid and turned by a rotatory motor at the base where it connects to the bacteria's body.  The rotary motor consist of several wheel-like discs one of which the M-ring (and/or possibly the S-ring) interact with the C-ring and studs to rotate the whole structure. The rotary motor is very like a stepping motor! The flagella is composed of a protein called flagellin which is synthesized in the cell body and transported through the narrow lumen of the growing flagella itself to polymerise at the tip as it is about to exit the bacteria!  This system has evolved into a syringe –like mechanism to inject toxins into the cells of vertebrates during infection (this is called "type 3 secretion").  There are two main type of prokaryotic flagella, those belonging to gram positive (one membrane) and gram negative (two membranes) bacteria (Figure 1).  The bacterial flagellum is driven by a proton motive force resulting from a gradient of protons.  Bacterial chemotaxis is brought about by alterations in the direction that the motor rotates in, this in turn is controlled by phosphorylation.

The Eukaryotic Flagellum
Although at first sight the flagella of eukaryotes is similar to the flagella of prokaryotes, our flagella are completely dissimilar in structure, function and in the genes that encode their components.  The principle component of the eukaryotic flagella is the microtubule, a tubular array of proteins of the tubulin family.  Instead of rotating as the prokaryotic flagella does, the eukaryotic flagella produces contortions in shape that travel around the structure like a Mexican wave.  The term cilia is generally used to describe small grouped structures less than 10mm, and flagella tend to be single structures about 40mm.  Our cilia are considered to be a cellular organelle and are almost certain to be derived from a primitive protist cell in the distant past.  In the human body they are used in mucus membranes to driven mucus around (out of the lungs), to drive sperm cells, but bizarrely, in development a single cilium is responsible for setting up the asymmetry of our internal organs (heart slightly to the left etc.) rare mutations in the genes encoding this structure cause situs inversus.  Like many small things, the eukaryotic flagella was first seen by Anton van Leuwenhoek.  The outer doublets are composed of microtubules and the outer and inner arms are dynein.  Dynein is a motor protein that works with microtubules much like myosin works on actin so that the whole flagellum is sent into spiral motions as each set of arms (dynein) walks up the microtubules.

Crawling locomotion of cells.

Figure 2
It has been suggested that the primary driving force behind the evolution of the actin cytoskeleton was to permit the equal division of components at cytokinesis.  If this is so, then a close second must have been the ability to move!  The subject of cell locomotion has been a very long one.  As far back as primitive microscopes were available scientists have been studying the so called "Giant amoeba" (Amoeba proteus & Chaos carolinensis).  At 1mm in length, these cells are enormous and so were ideal subject for the early studies.  The fact that these cells moved in "real time" (3-10mM/sec) meant that no time lapse photography was necessary either.  The large size of these amoebae made it possible to perform "Micrurgy", such as enucleation and tactile stimulation.  It was concluded (Jennings, 1906; Goldacre 1952) from such tactile studies that prodding the front of the cell caused the cell to change direction, whereas a gentle prod in the rear made the cell accelerate transiently (as it would any sensible creature!).  It was found  (Goldacre 1953) that the uroid was contractile and that the cell could move even after the nucleus was removed with fine glass needle. The finding that the uroid was contractile was the subject of about a hundred years of controversy as others (Allen, 1961) suggested that it was the front end of the cell that was contractile and active. 

Figure 3B

Observations on the amoeba revealed that there were two convertible states of cytoplasm, endoplasm and ectoplasm.  Endoplasm (seen at the cell centre, thus the name) was fluid, while ectoplasm under the cell membrane is gellated and comparatively static.  During active locomotion, endoplasm flows forwards faster than the speed of the cell.  As the fluid endoplasm reaches the "hyaloplasm", a special optically clear form of ectoplasm, the flow diverted toward the membrane whereupon the endoplasm gelled to form ectoplasm.  Vesicles, crystals, and other visible cytoplasmic inclusions are seen to become suddenly immobile having previously been seen to vibrate in Browning motion.

The Frontal Contraction theory
Forwarded by Bob Allen (Allen, 1961) on the basis of the behaviour of cytoplasm released (accidentally in the first instance) into media of various formulations.  Allen found the liberated cytoplasm had some very peculiar properties, the cytoplasm was contractile and isolated pools were seen to writhe producing squirting motions.  Allen compared these movements with the reversible gel to sol conversions that were clearly visible in the living cell.  He proposed that as the endoplasm to ectoplasm transformation took place, the forming ectoplasm contracted pulling the cell forward (Figure 3A).

The generalised contraction theory
The generalised contraction theory eventually won the day and is more in favour of the rear contraction hypothesis rather than the frontal contraction hypothesis.   This hypothesis is only really applicable in its strictest sense to the giant amoeba as they differ from other cell types in that the cortex is not attached to the plasma-membrane as it is in smaller amoebae (Dictyostelium, Acanthamoeba, Naegleria) and vertebrate cells.  However, a loose interpretation of the generalised contraction theory is appropriate for some vertebrate cells at least, as there is a strong hydraulic component to the locomotion of many cell types.  The generalised contraction theory suggests that contraction occurs as the ectoplasm transforms to endoplasm or in the more modern parlance, as gel disassembled to sol.  Molecular mechanism have been suggested for this transformation and a body of experimental evidence supports this (Janson & Taylor 1993).

Two types of crawling, lamelapodal spread and hydraulic contraction.
Many cell types in vertebrates adopt a very broad veil-like lamella or pseudopod at the cell front as they advance forward by crawling.  This morphology is best illustrated by the keratinocyte where most of the projected surface area is occupied by the lamella (figure 4) (see Lamella expansion).  

Figure 4

The Myosins – Key motor proteins in Cell Motility and Locomotion.

Figure 6A

Figure 6 A.  sequential aggregation of non-muscle myosins into mini-filament (Sinard et al, 1989a)

B.  Myosin light chain kinase phosphorylates the light chain on threonine 18 and serine 19 inducing a conformational change from a folded shape, to an extended active shape that is able to form minifilaments.  Phosphorylation at serine 9 by protein kinase C inactivates the myosin even if it was previously activated by MLCK, furthermore PKC phosphorylation reduces the rate that the myosin can be phosphorylated by MLCK.

Figure 6B

Several myosin heavy chain kinases have been identified (table 4).  In Dictyostelium there are three related enzymes MHCK A, B and C (Liang et al, 2002).  The latter, MHCK C seems to be involved primarily in cytokinesis while forms A and B were localised with myosin II at the rear cortical region of moving cells, moreover their distribution here was independent of myosin II as the same pattern was observed in the absence of myosin II.  Dictyostelium also expresses another MHCK that is related to protein kinase C (Rabin & Ravid, 2002), and like PKC in vertebrate cells this kinase (MHCK-PKC) inhibits the formation of myosin filament assembly.  MHCK-PKC is localised at the anterior of the locomoting amoebae rather than the rear where many of the other kinases are.

Myosin squeezes cytoplasm forward pulls the lamella forward, and rips off redundant adhesion at the rear.
Myosin II seems to perform three simultaneous functions in cells, the relative contribution of myosin to these three areas dependent on the particular locomotory morphology of the cell.  In the lamella, myosin II becomes progressively organised in mini-filaments arranged perpendicular to the direction of locomotion, toward the rear of the lamella the highest concentration of mini-filaments often corresponds to the boundary where the lamella meets the cell body.  A series of experiments has been reported where cells are settled onto deformable surfaces onto which small beads have been stuck to act as position markers.  As cells translocate on these surfaces the pattern of force generation can be seen.  The forces generated are highly co relatable with the myosin II mini-filament distribution seen in the various cell types (except in fibroblasts whose primary function is to contract and not to locomote).  It has been possible now to make substrates that are so deformable that even the very weak forces that Dictyostelium exerts can be detected and studied (Uchida et al, 2002).  The contribution of myosin II to these forces has been assessed by comparing wild-type Dictyostelium with a strain that lacks myosin II.  The anterior region produced a pushing force as the cell retracted the posterior and a pulling forces was detected in the rear.  In line with the generalised contraction hypothesis the anterior pushing force was interpreted as being generated by myosin II squeezing the cell hydraulically.

The experimental evidence for the roles of myosin in cell locomotion.
Since the discovery of myosin in non-muscle cells , most hypothesis on the problem of cell locomotion have been based on the premise that somehow myosin and actin contractility was involved.  It was therefore quite a shock when the single myosin II gene of Dictyostelium was knocked out and that this did not completely prevent cells from locomoting (De Lozanne & Spudich, 1987), it did however significantly slow them down and prevented the cells displaying a chemotactic response.  The explanation for this is that Dictyostelium only adheres weakly to the substrate and it has many other myosins through which it can generate contraction (Dai et al, 1999).  Inhibitory antibodies to myosin II have been microinjected into adherent Acanthamoeba (Sinard & Pollard 1989b), like the situation with Dictyostelium, the removal of myosin II function slowed but did not stop the cells from moving.  Myosin II function has been reduced too in vertebrate cells but this produced a more radical pattern of disruption (Honer et al, 1988).  The locomotory rate was reduced to almost zero and the cells seemed incapable of removing rear adhesions resulting in bizarre morphologies.  It seems that these fibroblasts were more affected that either Dictyostelium or Acanthamoeba since these cells make much tighter adhesions.  Neutrophils cannot be easily microinjected but instead the contribution of myosin II to locomotion in these cells has been studied by the treatment of neutrophils with a myosin II inhibitor drug (BDM) (Eddy et al, 2000), but the action of this drug is uncertain (see BDM).

Cortical tube models for cell locomotion

If contraction/solation of the cortical tube takes place only at the uroid, then the shape of the cell would be approximately cylindrical.  In this most simple of cases a 10% reduction in the volume of the uroid would result in a 10% increase in volume at the anterior, since the diameter at each end is equal the width of the cell does not effect this calculation.  However, most cells are not cylindrical but are variously tapered toward the uroid.  The shape of many hydraulic cells ectoplasmic cortex is approximately a semi-ellipsoid (Figure 34a).  Sequential contraction of the cortex by myosin II coupled with solation results in the formation of sol which can then be recycled to new ectoplasmic cortex at the front of the cell.  Obviously there is a relationship between contraction and new cortex formation and this determines the speed of locomotion.  If we compute the length of cortical portions created by the same contraction of cortexes with arbitrary units (Figure 34b) it can be seen that the width of the cell is not important in determining the speed but rather it is the length of the cell cortex.  This in fact is supported by the literature on amoeba of various lengths and thicknesses.

                                 Cell length

Radius                    100          150          200

20                            0.67         1              1.3
40                            0.67         1              1.3
80                            0.67         1              1.3

B.  In the above example (Figure 34b) the relationship between the number of cortical portions produced by each 10% reduction by contraction of the cortex depends solely on the length and not the cell width.  For a length of 150 units, 1 unit of new cortex is produced by each 10% reduction in volume

Signal Transduction and Cell Locomotion

As one might expect, the signal pathways that regulate cell locomotion overlap to a large degree with those activated during the chemotactic response.    Activated Rac binds and activates Ca²⁺ and calmodulin dependent kinase which can then phosphorylated Myosin II tails causing disassembly.  Stimulation of both the PAK1 and PI(3)kinase pathways results in AKT/PKB phosphorylation and activation through p38MAPK and directly by PI(3)kinase.  AKT/PKB binds myosin when activates (Tanaka et al, 1999) but the consequence of this is not yet clear. Activation of MK2 results in the phosphorylation of myosin light chain and activation of the ATPase.  The simultaneous release of the heat shock protein HSP27 may also affect actin polymerization since this protein (under poorly understood conditions) binds actin.  A possible negative feed back mechanism may operate through PAK1 phosphorylation of MLCK which inhibits its activity.  MLCK is activated by the G-protein Rho and Rho also activated Rho-kinase which activates myosin by direct phosphorylation of the light chains.  The details of how these interconnected pathways manage to work together to determine when and where myosin filaments are formed and when and where they are activated are not at all clear, but in the neutrophil at least calcium seems to have a major effect in myosin activation (Eddy et al, 2000). 

Figure 35 Connecting myosin II activation to the major signalling pathways in neutrophils and Dictyostelium.

References

Bosgraaf, L., Russcher, H., Smith, J. L., Wessels, D., Soll, D. R. & Van Haastert, P. J. M. (2002) A novel cGMP signalling pathway mediating myosin phosphorylation and chemotaxis in Dictyostelium. EMBO J. 21, 4560-4570.

Bresnick, A. R. (1999) Molecular mechanisms of non-muscle myosin-II regulation, Curr.Op.Cell Biol. 11, 26-33.

Dai, J., Ting-Beall, H. P., Hochmuth, R. M., Sheetz, H. P. & Titus, M. A. (1999) Myosin 1 contributes to the generation of resting cortical tension., J.Biophys. 77, 1168-1176.

De Lozanne, A. & Spudich, J. A. (1987) Disruption of the Dictyostelium myosin heavy chain gene by homologous recombination., Science. 236, 1086-1091.

Eddy, R. J., Pierini, L. M., Matsumura, F. & Maxfield, F. R. (2000) Ca²⁺-dependent myosin II activation is required for uropod retraction during neutrophil migration. J.Cell Sci. 113, 1287-1298.

Janson, L. W. & Taylor, D. L. (1993) In vitro models of tail contraction and cytoplasmic streaming in amoeboid cells.  J. Cell Biol. 123, 345-356.

Jennings, H. S. (1906) in Behaviour of the lower organisms  pp. 1-25, Indiana. USA.

Nagasaki, A., Itoh, G., Yumura, S. & Uyeda, T. Q. P. (2002) Novel Myosin Heavy Chain Kinase Involved in Disassembly of Myosin II Filaments and Efficient Cleavage in Mitotic Dictyostelium Cells. Mol. Biol. Cell. 13, 4333-4342.

Levi, S., Polyakov, M. V. & Egelhoff, T. T. (2002) Myosin II dynamics in Dictyostelium: Determinants for filament assembly and translocation to the cell cortex during chemoattractant responses. Cell Mot.Cytoskel. 53, 177-188.

Liang, W., Licate, L. S., Warrick, H. M., Spudich, J. A. & Egelhoff, T. T. (2002) Differential localization in cells of myosin II heavy chain kinases during cytokinesis and polarized migration. BMC Cell Biology. 3.

Nishikawa, M., Sellers, J. R., Adelstein, R. S. & Hidaka, H. (1984) Protein kinase C modulates in vitro phosphorylation of the smooth muscle heavy chain meromyosin by myosin light chain kinase. J.Biol.Chem. 259, 8808-8814.

Rubin, H. & Ravid, S. (2002) Polarization of myosin II heavy chain-protein kinase C in chemotaxing Dictyostelium cells., J.Biol.Chem. 277, 36005-36008.

Shu, S., Liu, X., Parent, C. A., Uyeda, T. Q. P. & Korn, E. D. (2002) Tail chimeras of Dictyostelium myosin II support cytokinesis and other myosin II activities but not full development. J Cell Sci. 115, 4237-4249.

Sinard, J. H., Stafford, W. F. & Pollard, T. D. (1989a) The mechanism of assembly of Acanthamoeba myosin-II minifilaments: minifilaments assembly by three successive dimerization steps. J. Cell Biol. 109, 1537-1547.

Sinard, J. H. & Pollard, T. D. (1989b) Microinjection into Acanthamoeba castellanii of monoclonal antibodies to myosin-II slows but does not stop cell locomotion. Cell Motility Cytoskeleton. 12, 42-52.

Svitkina, T. M., Verkovsky, A. B., McQuade, K. M. & Borisy, G. G. (1997) Analysis of the actin-myosin II system in fish epidermal keratocytes: Mechanism of cell body translocation. J.Cell Biol. 139, 397-415.

Tanaka, M., Konishi, H., Touhara, K., Sakane, F., Hirata, M., Ono, Y. & Kikkawa, U. (1999) Identification of myosin II as a binding protein to the PH domain of protein kinase B., Biochem Biophys Res Commun. 255, 169-174.

Uchida, K. S. K., Kitanishi-Yumura, T. & Yumura, S. (2003) Myosin II contributes to the posterior contraction and the anterior extension during the retraction phase in migrating Dictyostelium cells. J Cell Sci. 116, 51-60.

Yonemura, S. & Pollard, T. D. (1992) The localization of myosin I and myosin II in Acanthamoeba by fluorescence microscopy. J.Cell Sci. 102, 629-642.

Yumura, S. & Fukui, Y. (1985) Reverse cyclic AMP-dependent change in distribution of myosin thick filaments in Dictyostelium. Nature. 314. 194-196.