Biology Reference
In-Depth Information
microtubules destined for the axon are nucleated at the centrosome, released, and
then transported into the axon.
Electron microscopic analyses of different kinds of neurons at different
developmental stages vary with regard to the appearance of the centrosome, but most
studies reveal relatively few microtubules directly attached to the centrosome. In our
studies on cultured sympathetic neurons, generally fewer than ten and often no
microtubules were observed to be attached to the centrosome (Baas and Joshi
1992
;
Yu et al.
1993
). These observations raised the possibility that axonal microtubules
may not originate at the centrosome and that the neuronal centrosome may actually
be relatively inactive. Alternatively, however, the nucleation and release of
microtubules from the neuronal centrosome may be so rapid that there is insufficient
time for substantial numbers of attached microtubules to accumulate at the
centrosome before they are released. To address this issue, we tested the capacity of
the neuronal centrosome to nucleate large numbers of microtubules, using the same
drug-recovery regime that we used on the axon (Yu et al.
1993
). Within a few
minutes of drug removal, hundreds of microtubules reassembled in the region of the
centrosome, and most of these microtubules were clearly attached to it (Fig.
18.3
).
Some of the microtubules were not attached to the centrosome, but were aligned
side-by-side with the attached microtubules, suggesting that the unattached
microtubules had been released from the centrosome after their nucleation. In
addition, unattached microtubules were present in the cell body at decreasing levels
with increasing distance from the centrosome. By 30 min after removing the drug,
the microtubule array was indistinguishable from that of control neurons, suggesting
that the hundreds of microtubules nucleated from the centrosome were subsequently
released and translocated away from the centrosome.
We next tested whether microtubules derived from the centrosome are essential
for the initiation and growth of the axon. Our strategy was to microinject into
cultured sympathetic neurons a function-blocking antibody to gamma-tubulin pre-
viously shown to arrest microtubule nucleation at the centrosome when microin-
jected into other cell types (Ahmad et al.
1994
). We reasoned that if centrosomally
derived microtubules are required for the growth of the axon, we would expect
inhibition of centrosome function to compromise or inhibit axonal growth. These
experiments were tricky, however, because the cell body of the neuron is packed
with microtubules that had presumably (according to our hypothesis) already been
nucleated and released from the centrosome. Therefore, it was also necessary to
deplete the neuron experimentally of pre-existing microtubules. After depolymer-
izing existing microtubules with nocodazole, the antibody was microinjected into
neurons, and then the drug was rinsed from the cultures. Reassembly of microtubules
over the next two hours was severely diminished under these conditions, and axonal
growth was either compromised or completely abolished. These results, using an
admittedly complicated experimental regime, suggested that microtubules gener-
ated from the centrosome are important for axonal growth.
Finally, we set forth to test if the microtubules nucleated at the centrosome are the
same microtubules that ultimately arrive in the axon. To test this, we modified our
pharmacological experiments into a kind of ''pulse-chase'' regime that permitted us
