Two lines of evidence indicate that microtubule tips approach focal adhesions at a range close enough for the precise exhange of molecular signals. The first evidence comes from the use of evanescent wave microscopy (Toomre and Manstein, 2001).
If a beam of light illuminates a transparent, reflecting surface at the angle of total internal reflection, an evanescent wave penetrates above the surface to a depth of around 100 — 200nm. With living cells growing on a glass coverslip, the illumination is then restricted to the ventral surface, which includes the substrate adhesion sites.
Evanescent wave microscopy of cells labelled with GFP-tubulin shows that microtubules dip down into the evanescent wave as they grow towards the cell periphery:
Evanescent wave microscopy of a fibroblast expressing GFP-tubulin. By this technique, the sample is illuminated with exciting light in a layer only 100 — 200nm above the substrate. The appearance of microtubules only at the cell periphery indicates that microtubules dip down towards the ventral cell surface in these peripheral regions. Note that some microtubules follow identical tracks. (From Krylyshkina et al., 2003)
Double labeling of cells for an adhesion component (zyxin) and microtubules shows that the dipping down of microtubules correlates with their targeting of adhesion sites:
Evanescent wave microscopy of a fibroblast expressing GFP-tubulin and Ds-Red zyxin and imaged simultaneously in the green and red fluorescent channels. Both pseudo colour and a black and white superimposition of images are shown. Where microtubules dip down towards the ventral cell surface, they target adhesion sites. (From Krylyshkina et al., 2003)
The intensity of the evanescent wave decreases exponentially away from the substrate and the intensity of fluorescence of the microtubule tips gives a measure of their depth of penetration into the evanescent wave. Calculations show that the microtubules approach the ventral cell surface within a range of 50nm. We return later to this phenomenon in the context of the complex of proteins found at microtubule tips.
A second indication of an intimate cross-talk between microtubules and focal adhesions is indicated by the ability of adhesion sites to capture depolymerising microtubules and to temporarily stabilise them. This is illustrated in the following video: in this experiment, the cell was exposed during the sequence to nocodazole to depolymerise microtubules. Following addition of the drug (at the point of loss of focus), microtubules start to shorten. One of them (T) is subsequently captured during shortening at an adhesion site (V) and stops depolymerising there for several minutes (real time) before finally shrinking into the cell body (Kaverina et al., 1998):
Adhesion sites can capture microtubules. Video shows the peripheral region of a fibroblast that was co-injected with cy-3 tubulin and rhodamine-vinculin. During the video, nocodazole was added to the medium to initiate the depolymerisation of microtubules (addition at point of defocus). A shrinking microtubule (T) is captuterd at a focal adhesion (V) and temporarily stabilised at the adhesion against depolymerisation. (From Kaverina et al., 1998)
What is the consequence of the targeting of adhesion sites by microtubules? Different experiments suggest that microtubules retard the growth or stimulate the disassembly of focal adhesion sites.
A first example of the influence of microtubules on adhesion dynamics is shown in the movie below. Here the formation and turnover of adhesions were monitored in cells that were settling and spreading on a substrate. A control cell (left) is compared with a cell that was pretreated and maintained in nocodazole to depolymerise all microtubules. As the control cell spreads, the first adhesions formed at the cell perimeter (marked by arrowheads) are dissolved and replaced by new adhesions established at a larger radius. In the cell lacking microtubules (right) the early adhesion do not dissolve, but just elongate as the cell radius increases. This lack of turnover of adhesions in the absence of microtubules retards spreading.
Microtubules are required to promote adhesion site turnover during the spreading of cells on a substrate. Video pair shows two fish fibroblasts during the early stages of spreading on a substrate (both cells were transfected with GFP-zyxin to mark adhesion sites). A control cell (with a normal complement of microtubules) is compared with a cell without microtubules (obtained by pretreatment and incubation during spreading in nocodazole). Note the difference in turnover of the adhesion sites marked by arrowheads in the two situations. Early adhesion sites in the control cell turnover during spreading, to be replaced by new adhesions at a larger radius, whereas those in the nocodazole-treated cell persist and elongate.
A second example is shown below. In this experiment the destabilising effect of microtubules on adhesion is shown. The edge of a cell was exposed to nocodazole through a microneedle to depolymerise microtubules locally. Adhesion sites marked by vinculin populate the cell edge. The video shows the result of removing the nocodazole to allow repolymerisation of the microtubules. Note that the microtubules grow into the adhesion sites at the cell periphery and that this correlates with the release of the adhesions from the substrate.
Microtubule targeting promotes the release of adhesions from the substrate. In this experiment, a fish fibrobast was injected with cy-3 tubulin and rhodamine vinculin and then exposed to nocodazole through a micropipette to depolymerise microtubules locally. The video shows the events after removal of nocodazole. Note that the microtubules grow to the cell periphery and target the peripheral adhesion sites (arrowheads). Targeting is followed by the release of the adhesions and retraction of the cell edge. (From Kaverina et al., 1999)
A third example is shown for a motile cell in the next two figures: the overview movie shows a fish fibroblast protruding on a broad front and retracting at the rear and flanks. Retraction occurs in concert with the dissociation from the substrate of peripheral focal adhesions. Details of the retracting flank are shown in the inset. Note that the dissociation of the adhesion sites (marked by arrowheads) is preceded and accompanied by the multiple targeting of the adhesion sites by microtubules.
Microtubule-adhesion site targeting in a moving cell. Video shows a fish fibroblast that was injected with cy-3 tubulin and rhodamine vinculin. Note protrusion of the cell front and the retraction of the flanks and rear. Details in the inset are shown in the movie below.
Inset region from the previous movie, showing details of microtubule-adhesion site targeting. Selected adhesion sites are indicated by arrowheads. Note that the retraction of the adhesion sites is preceded and accompanied by multiple targeting events by microtubules. (From Kaverina et al., 1999)
A further example of the dissolution of adhesion foci following multiple targeting by microtubules is shown in the movie below:
Dissolution of adhesion sites following targeting by microtubules. Video shows a region of a cell that was transfected with GFP-tubulin and Ds-Red zyxin. The two adhesion sites circled disappear after multiple targeting events by microtubules. (Video produced by Irina Kaverina)
- Kaverina, I., Rottner, K., Small, J. V. (1998). Targeting, capture, and stabilization of microtubules at early focal adhesions. J. Cell Biol. 142, 181– 190.
- Kaverina, I., Krylyshkina, O., Small, J. V. (1999). Microtubule targeting of substrate contacts promotes their relaxation and dissociation. J. Cell Biol. 146, 1033 — 1044.
- Krylyshkina, O., Anderson, K. I., Kaverina, I., Upmann, I., Manstein, D. J., Small, J. V., Toomre, D. K. (2003). Nanometer targeting of microtubules to focal adhesions. J. Cell Biol. 161, 853 — 859.
- Toomre, D. and Manstein, D. J. (2001). Lighting up the cell surface with evanescent wave microscopy. Trends Cell. Biol. 11, 298 — 303.