- Email: [email protected]
Preferential blockade of the tubulin transport by colchicine
Brain Research, 190 (1980) 505-516 © Elsevier/North-Holland Biomedical Press
505
P R E F E R E N T I A L B L O C K A D E OF THE T U B U L I N T R A N S P O R T BY COLCHICINE
YOSHIAKI KOMIYA and MASANORI KUROKAWA Department of Biochemistry, Institute of Brain Research, Tokyo UniversityFaculty of Medicine, Hongo 7-3-1, Bunkyo ku, Tokyo 113 (Japan)
(Accepted November 1st, 1979) Key words: slow axoplasmic transport -- neurofilament polypeptides -- tubulins -- actin -- col-
chicine -- vinblastine sulphate - - cytochalasin D
SUMMARY L-[asS]Methionine was injected into the dorsal root ganglion (Ls) of the adult rat, and migration of the neurofilament polypeptides (the triplet with molecular weights of 200,000, 160,000 and 68,000 daltons), a- and fl-tubulins and actin in the sciatic nerve and the dorsal root was quantitatively determined and also examined by fluorography. Colchicine (4/~g) injected into the ganglion 10 min before methionine preferentially blocked the tubulin transport, with little if any blockade of the triplet and actin. Colchicine at this dose had no effects on the incorporation of L-[14C]leucine into the total protein and also into tubulins. In contrast to colchicine, vinblastine sulphate (4 #g) injected into the ganglion in a similar way blocked the transport of all the triplet, tubulins and actin. Cytochalasin D (1 /~g) had no effect on the slow axoplasmic transport.
INTRODUCTION Since the initial demonstration by Dahlstr6m 4 and Kreutzberg 14, interruption of axoplasmic transport by antimitotic agents has been documented under a variety of experimental conditions (for review, see ref. 8). It has been reported that rapid transport, including retrograde transport2,15, is almost completely blocked by colchicine, while blockade of slow transport is only partiala, 11,2~. However, some workers have observed that the slow transport is more profoundly affected by colchicine compared with the rapid transport10,17.
506 It has now been established that the major portion (approx. 75'!i;) of the slow axoplasmic transport consists of the triplet polypeptides, ~- and l#tubulins and actin 9,16,~s,~9. The initial suggestion that the triplet polypeptides are associated with neurofilaments 9 has been corroborated by a recent demonstration that the neurofilaments isolated from the peripheral nerve consist almost exclusively of three polypeptides with molecular weights of 200,000, 160,000 and 68,000 daltons z0. In addition, it is now possible to quantitate radioactivities associated with the triplet, tubulins and actin that are migrating within the axons ~9. In this situation, it is thought pertinent to re-examine the effects of antimitotic agents against the slow transport. Fluorographic examinations and quantitative determinations of the slowly migrating polypeptides in the present study indicate that colchicine preferentially blocks the transport of tubulins, while vinblastine blocks the transport of all the major polypeptides. Slow axoplasmic transport is not influenced by cytochalasin D. MATERIALS AND METHODS
Treatment of" the animal and nerve for the transport study Male albino rats of the Wistar strain, 9-week-old and weighing about 250 g, were used throughout. Under ether anaesthesia, dorsal root ganglia (bilateral L~) were exposed by a partial laminectomy, and L-[3~S]methionine (200 #Ci in 0.4 ul) was injected via a glass capillary (20-50 /~m in tip width) into the ganglion. In the experiments indicated, colchicine (4 or 8 #g in 0.2/~1), vinblastine sulphate (4/~g in 0.25/~1) or cytochalasin D (1/~g in 0.2 #1) was injected 10 rain before the injection Of L[aSS]methionine. In all the experiments, the contralateral ganglion that received the solvent served as control. At days 4, 7, 9, 14 and 28 after the injection, animals were killed by decapitation and the dorsal root (central axons), L5 ganglion and the sciatic nerve (peripheral axons) were dissected out. The tissue was placed on an ice-cold plastic plate, cut into consecutive 3 mm segments, and kept frozen at --80 °C until treated as below. Each segment was ground in a small teflon-glass homogenizer in 200/A of the medium which consisted of sodium dodecyl sulphate (2.3 o~, w/v), fl-mercaptoethanol (5 raM), glycerol (10 ~,, w/v) and Tris.HCl (62.5 raM), pH 6.8. The homogenate was centrifuged at 1500 × g for 5 rain at 25 °C, and the clear supernatant was subjected to further analyses. SDS-polyacrylamide slab gel electrophoresis, radiometry and fluorography of polypeptides migrating within the axons The nerve extract was incubated at 37 °C for 15 min and then heated in a boiling water bath for 5 rain, and an aliquot (50 #1) was placed in a counting vial. After adding 0.2 ml of Soluene-350 (Packard Instrument Co., Downers Grove, Ill., U.S.A.), the vial was heated at 50 °C for 2 h, and the radioactivity was measured in a liquid scintillation spectrometer using toluene-based scintillation fluid 1,13,
507 Aliquots from the two adjacent segments (20 #1 each) were combined and subjected to electrophoresis and subsequent fluorography as described previously19. In order to determine radioactivities associated with the triplet polypeptides, aand fl-tubulins and actin separately, the tissue was taken at day 9 postlabelling. Aliquots of the nerve extract from segments C2-C10 were combined; aliquots were combined also from P2-Pzs (Cx and Px denote the x-th central and peripheral 3 mm segments, respectively). Samples C2-C10 (in duplicate), C1, ganglion, Pl and P~Pzs (in duplicate) were electrophoresed in a single slab gel (with 10 mm wide slots), and the gel was stained with Coomassie brilliant blue. Portions stained for the triplet, tubulins and actin were cut off, and respective radioactivities were determined as described previouslylL Amino acid incorporation studies Colchicine (4 #g in 0.2 #1) and L-[14C]leucine (0.8/~Ci in 0.2 #1) were injected into the L5 ganglion as described above. The contralateral ganglion that received the solvent and L-[14C]leucine served as control. At minutes 15, 30 and 45, and at hours 1, 1.5, 2, 4, 8 and 24 after the isotope injection, the animals were killed by decapitation and the ganglion was removed together with 3 mm long central and peripheral stumps of axons. The tissue was ground in a teflon-glass homogenizer in 400/zl of the above-mentioned medium, centrifuged at 1500 × g for 5 min, and the clear supernatant was used for the following analyses. An aliquot (20 #1) was used to determine the total radioactivity. Another aliquot (30 #1) was mixed with the equal volume of 20 ~ (w/v) trichloroacetic acid and centrifuged, and radioactivities in acidsoluble and acid-insoluble fractions were determined. The third aliquot (200 #1) was electrophoresed and radioactivities associated with the triplet polypeptides, tubulins and actin were separately determined as described above. Electron microscopy Colchicine (4 #g) was injected into the L5 ganglion and, 7 days later, the animal was perfused through the heart with phosphate-buffered saline, and prefixed by perfusion through the heart with 2 ~ formaldehyde-2.5~ glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4. The L5 ganglion and nerve were removed and further fixed overnight in the cold. The tissue was then washed with 10~ (w/v) sucrose in 0.1 M cacodylate buffer, pH 7.4, for 10 min, and postfixed with 1 ~ OsO4 in 0.1 M cacodylate buffer, pH 7.4, for 2 h in the cold. The tissue was washed with water for 10 min, stained en bloc with 0.5 ~ aqueous uranyl acetate for 2 h at room temperature, dehydrated with several changes of ethanol of gradient concentrations, and embedded in Epon 812. Thin sections were cut with a diamond knife, doubly stained with lead citrate and uranyl acetate, and examined in a Hitachi ll-DS electron microscope at 75 kV. Chemicals L-[~S]Methionine (800-1200 Ci/mmol) and L-[14C]leucine (300 Ci/mmol) were obtained from Radiochemical Centre, Amersham, U.K. The label was taken to
508 dryness in vacuo, and brought to a final concentration of 500 mCi/ml for L-[:~:~S]methionine and 4 mCi/ml for e-[14C]leucine by the addition of distilled water. Colchicine was dissolved in dimethyl sulphoxide to 1 M, and stored at --20 *'C in the dark. On the day of experiment, the stock solution was diluted about 20 times with physiological saline. Cytochalasin D was also dissolved in dimethyl sulphoxide to 10 mg/ml and stored. Vinblastine sulphate was dissolved in saline to 16 mg/ml immediately before the injection. Other chemicals were of analytical grade. RESULTS
Since all the animals that received 8/~g of colchicine died within several days after the injection, the results described below are those obtained in animals that received 4 #g of colchicine.
Effects of colchicine on amino acid incorporation into proteins The time course of incorporation of L-[14C]leucine into the acid-insoluble fraction and into tubulins is not affected by colchicine treatment (Fig. 1A, B). Tubulins/ triplet and tubulins/actin ratios do not differ between the control and colchicinetreated ganglia (Table i), indicating that colchicine does not alter the incorporation of L-[14C]leucine into any of these polypeptides. I00
B
A 80
=* 60 "5
"g 4 0
o
2!
o n,
/
./.
20
r
O
I
I
T
~
O
r
O
i
r
I
Time after isotope injection (hours)
Fig. 1. Effects of colchicine on the incorporation of L-[14C]leucine into the acid-insoluble fraction (A) and into tubulins (B). The L5 ganglion was labelled with L-[14C]leucine (0.8/~Ci) and, at the time points indicated, the ganglion together with the peripheral and central stumps of axons (3 m m each) was removed and processed for determination of radioactivities associated with acid-insoluble materials (A) and with tubulins (B). Colchicine (4/~g) was injected into the ganglion 10 min before the isotope. O ©, control; • • , colchicine-treated. Determinations made at 4, 8 and 24 h postlabelling gave values practically identical with those obtained at 2 h.
509 TABLE I
Effects of colchicine on the incorporation of L-[laC]leucine into the neurofilamentpolypeptides (triplet), tubulins and actin in the L5 ganglion The radioactivity measurements were made with samples obtained at minutes 15, 30 and 45, and at hours 1, 1.5, 2, 4, 8 and 24 postlabelling, and means of respective ratios ± S.E.M. are given (see text; cf. Fig. 1). Colchicine (4/~g) was injected into the ganglion 10 rain before the isotope.
Control Colchicine-treated
Tubulins/triplet
Tubulins/actin
0.88 ± 0.02 (9) 0.86 ± 0.06 (9)
1.43 ± 0.05 (9) 1.41 :k 0.03 (9)
Effects of colchicine, vinblastine and cytochalasin D on slow axoplasmic transport Flow profiles of labelled proteins in the control, colchicine-treated and vinblastine-treated nerves are shown in Fig. 2. When the inhibitory effect of the drug is expressed in [1--(counts in drug-treated nerve/counts in drug-treated ganglion)/ (counts in control nerve/counts in control ganglion)] × I00, colchicine proves to inhibit the transport by 30-50 % and vinblastine by approx. 70 %. Fluorographic analyses demonstrate that colchicine preferentially blocks the transport of tubulins (Fig. 3A, B); transports of the triplet and actin are apparently unaffected. This visual impression is further supported by quantitative determination showing that, in both the central and peripheral axons, the tubulins/triplet ratio is I00
;',?/ ill l~o
50
//
- 30
/
20
~to
..~
/ 2//
".~, u
#
J
s
\
F',/ ~.
3 Central i 30
I
I
~
I
/ I5
I
i
I
(~ I
Distance
I 0
I K " V ""IDx'~ D ~
Peripheral ' from
'
'
" "X "'A~ ~ ~ -A.-~ '
I 15
l
'
'
E 3 l0
I''
I
r
the ganglion (mm)
Fig. 2. Flow profiles of radioactivities in the dorsal root (central axons) and the sciatic nerve (peripheral axons) 7 days after labelling the L5 ganglion with 200/~Ci of L-[aSSlmethionine. Colchicine (4/~g) or vinblastine sulphate (4/~g) was injected into the ganglion 10 min before the isotope. (3(3, control ; O - - - O , colchicine-treated; A - - - A , vinblastine-treated. Vertical axis indicates [radioactivities in 3 mm segment/radioactivities in the ganglion] × 100.
510 significantly decreased by colchicine treatment, while the actin/triplet ratio remains virtually unaffected (Table II). In contrast to colchicine, vinblastine blocks the transport of the triplet, tubulins and actin without distinction (Fig. 3C). Cytochalasin D does not affect the slow transport as visualized by fluorography (Fig. 3D).
200
68 56 53 tO "10 I
0 ×
° ~
•
200-
"5
160-
o 0
68565343-
I
C
I
®
''
'
' ~ '
'
'
'6'o
P
Distance from the ganglion (mm/
0
~
~
~
--
~
-.
..~;
~'~
..~
e~..~
e
,~. o ~. =.
~4~ S" =
°"
"'~
0
~" ~ ~. ~ o~
o
(10,
eo,
o
0) 00
0
--
h)
Molecular weights( × 10 -3 daltons)
O) O) O0
512 TABLE II Quantitative determinations of radioactivities associated with the triplet, tubulins and actitl that are migrating in the central and peripheral axons of the L5 ganglion cells
The tissue was removed 9 days after labelling the ganglion with L-[35S]methionine,and processed for radioactivity measurement as described in the text. Colchicine was injected into the ganglion 10 min before the isotope. Means of 4 determinations ± S.E.M. are given. Central
Peripheral
Tubulins/triplet
Control Colchicine-treated
1.20 ± 0.02* 0.72 - 0.05
1.43 J: 0.03* 0.71 ± 0.07
Actin/triplet
Control Colchicine-treated
0.57 ± 0.05 0.67 ± 0.07
0.45 ± 0.01 0.43 ± 0.05
* The difference between the control and colchicine-treated animals with respect to the tubulin/triplet ratio is significant at P < 0.01 (Students t-test). Electron microscopic observations o f the colchicine-treated nerve
Electron microscopic examinations of the ganglion show that neurones and periaxonal satellite cells are well preserved 7 days after the treatment, though chromatolytic changes are apparent in some cells. At the same time point, examinations of the sciatic nerve 8 m m away from the ganglion show occasional intact myelinated axons (Fig. 4A), degenerating myelinated axons (Fig. 4B, C) and myelin fragments phagocytosed by Schwann cells. However, intact unmyelinated sensory axons are hardly visible. In contrast, motor axons (both myelinated and unmyelinated), which can be well demarcated from sensory axons at this site of the nerve, remain intact (Fig. 4D). DISCUSSION Our present results clearly indicate that colchicine (4 big) preferentially blocks the intra-axonal migration of tubulins, in contrast with vinblastine (4 #g), which indiscriminately blocks the migration of all the triplet, tubulins and actin. We were unable to examine the effects of the higher dose (8 b~g) of colchicine because all the animals that received this dose died within several days. Effects of lower doses of vinblastine have not been studied in detail. However, the non-selectivity ofvinblastine action shown in the present experiment is not unexpected in that vinblastine interacts not only with tubulins, but also with actin 24 and the neurofilament triplet 2°, in contrast to colchicine which interacts rather specifically with tubulin dimer 23. Cytochalasin D does not affect the slow axoplasmic transport, in agreement with previous observations 17. Reduction of the amount of migrating tubulins by colchicine is not a consequence of their reduced synthesis in neurone soma (Fig. 1B; Table I). On electron microscopy, neurones and periaxonal satellite cells are found well preserved 7 days after the application of colchicine, though chromatolytic changes are apparent in a
513
Fig. 4. Electron micrographs of the colchicine-treated sciatic nerve. Colchicine (4 #g) was injected into the L~ ganglion, and the ganglion and nerve were removed 7 days later after prefixation by perfusion. Sections were prepared from the nerve site 8 m m away from the ganglion. A: intact myelinated axon, x 18,000. B: partially disrupted myelinated axon, x 12,000. C: highly disrupted myelinated axon, x 6800. D : Myelinated and unmyelinated motor axons, x 12,000.
514 limited number of cells. Tubulins can be synthesized in both neurones and glial cells in the ganglion. However, because of a more rapid metabolic turnover of proteins in neurones than in glial cells 12, a large proportion of tubulins that are synthesized in an early period after labelling the ganglion is considered to be of neuronal origin (cf. ref. 19). In control nerves, the amounts oftubulins that leave the soma and are loaded on the axoplasmic transport are more than 4 times as much as those that stay in the ganglion. In colchicine-treated animals, de novo synthesis of tubulins remains unaltered, and the amount of tubulins that leave the soma is reduced by 30-501',i. Under these circumstances, an accumulation of tubulins should be detected within the ganglion. However, our quantitative determination of radioactivities associated with tubulins in the ganglion failed to detect any such accumulation. It thus seems likely that tubulin dimer molecules which bind colchicine, and accordingly become unable to undergo polymerization, are easily destroyed. Blockade of tubulin transport by colchicine is not complete under our present experimental conditions. Thus, even in the colchicine-treated nerve, a proportion of tubulins is found to migrate at a rate apparently identical with that in the normal nerve (Fig. 3A, B). At least 4 possibilities exist to explain this finding. First, colchicine in higher doses may cause a complete blockade, though we were unable to confirm this, being hampered by a high mortality of animals treated with 8 #g of colchicine. Second, an amount of tubulins may have been synthesized before colchicine starts to exert its inhibitory action; colchicine is reported to require a certain lag period for exhibiting its maximal effect 21, possibly because of a slow process of colchicine-tubulin interaction. Third, assembly and disassembly studies of tubulins and microtubuleassociated proteins in vitro have indicated that there exist at least two types of tubulin-microtubule equilibria, with different sensitivities to colchicine 7. Fourth, the myelinated and unmyelinated axons may differ in their sensitivity to colchicine. This possibility is inferred from the electron microscopic observations that unmyelinated sensory axons are practically absent under conditions where myelinated axons, though more or less disrupted, are largely preserved (Fig. 4A-C). The failure to detect unmyelinated sensory axons is not due to technical insufficiencies, because unmyelihated m o t o r axons are typically demonstrated in the same section of the colchicinetreated nerve (Fig. 4D). The rate and amount of migration of the triplet polypeptides are apparently unaffected by colchicine (Fig. 3A, B), and this is not surprising in view of the lack of interaction between the isolated neurofilaments and colchicine 20. However, there are at least two factors that complicate the interpretation of behaviours of the 10 nm filaments in vivo. First, the neurofilament/microtubule ratio is relatively higher in myelinated than in unmyelinated axons. If myelinated axons are in fact more resistant to colchicine as inferred above, then subtle changes in any of the triplet polypeptides may be overlooked or underestimated. Second, the number of the 10 nm filaments in axons has been reported to increase under the influence of colchicine 6. A glance at Fig. 4A and Fig. 4B gives an impression that 10 nm filaments are increased in degenerating axons. However, deliberate quantitative analyses are needed to conclude that l0 nm
515 filaments are in fact increased in number, even when the filament density in degenerating axons is corrected for the shrinkage o f cross-sectioned area. I f the transport o f tubulins is preferentially or differentially affected by colchicine, this suggests that, in some situations, the transport o f neurofilament triplet (and o f actin) does not require coordinate transport oftubulins. A n apparently reverse situation is f o u n d in the crayfish nerve cord, where the slow (and fast) transport is shown to proceed in the absence o f neurofilaments 5. However, this does not preclude the possible interaction a m o n g the triplet, tubulins and actin migrating within the normal m a m m a l i a n axons. ACKNOWLEDGEMENTS We thank Dr. H. Ishikawa and Mr. S. Tsukita for preparing the electron micrographs (Fig. 4) and for their valuable discussions. Thanks are also due to the Shionogi Co. for its generous gift o f vinblastine sulphate. This w o r k was supported in part by Research Grants from the Ministry o f Education, Japan, and by a grant from the Mitsubishi Foundation.
REFERENCES 1 Abe, T., Haga, T. and Kurokawa, M., Rapid transport of phosphatidylcholine occurring simultaneously with protein transport in the frog sciatic nerve, Biochem. J., 136 (1973) 731-740. 2 Abe, T., Haga, T. and Kurokawa, M., Retrograde axoplasmic transport: its continuation as anterograde transport, FEBS Lett., 47 (1974) 272-275. 3 Boesch, J., Marko, P. and Cu6nod, M., Effects of colchicine on axonal transport of proteins in the pigeon visual pathways, Neurobiology, 2 (1972) 123-132. 4 Dahlstr6m, A., Effect of colchicine on transport of amine storage granules in sympathetic nerves of rat, Europ. J. PharmacoL, 5 (1968) 111-113. 5 Fernandez, H. L., Burton, P. R. and Samson, F. E., Axoplasmic transport in the crayfish nerve cord. The role of fibrillar constituents of neurons, J. Cell BioL, 51 (1971) 176-192. 6 Fink, B. R., Byers, M. R. and Middaugh, M. E., Dynamics of colchicine effects on rapid axonal transport and axonal morphology, Brain Research, 56 (1973) 299-311. 7 Haga, T. and Kurokawa, M., Microtubule formation from two components separated by gel filtration of a tubulin preparation, Biochim. biophys. Acta (Amst.), 392 (1975) 335-345. 8 Hanson, M. and Edstrfim, A., Mitosis inhibitors and axonal transport, Int. Rev. CytoL, Suppl. 7 (1978) 373-402. 9 Hoffman, P. N. and Lasek, R. J., The slow component of axonal transport. Identification of major structural polypeptides of the axon and their generality among mammalian neurons, J. Cell BioL, 66 (1975) 351-366. 10 James, K. A. C., Bray, J. J., Morgan, I. G. and Austin, L., The effect of colchicine on the transport of axonal protein in the chicken, Biochem. J., 117 (1970) 767-771. 11 Karlsson, J.-O., Hansson, H.-A. and Sjtistrand, J., Effect of colchicine on axonal transport and morphology of retinal ganglion cells, Z. Zellforsch., 115 (1971) 265-283. 12 Koenig, H., An autoradiographic study of nucleic acid and protein turnover in the mammalian neuraxis, J. biophys, biochem. CytoL, 4 (1958) 785-792. 13 Komiya, Y. and Kurokawa, M., Asymmetry of protein transport in two branches of bifurcating axons, Brain Research, 139 (1978) 354-358. 14 Kreutzberg, G. W., Neuronal dynamics and axonal flow. IV. Blockage of intra-axonal enzyme transport by colchicine, Proc. nat. Acad. Sci.(Wash.), 62 (1969) 722-728. 15 Kristensson, K. and Sjtistrand, J., Retrograde transport of protein tracer in the rabbit hypoglossai nerve during regeneration, Brain Research, 45 (1972) 175-181. 16 Lasek, R. J. and Hoffman, P. N., The neuronal cytoskeleton, axonal transport and axonal growth.
516 In R. Goldman, T. Pollard and J. Rosenbaum (Eds.), Cell Motility, Cold Spr. Harb. Laboratory, New York, 1976, pp. 1021-1049. 17 McGregor, A. M., Komiya, Y., Kidman, A. D. and Austin, L., The blockage of axoplasmic flow of proteins by colchicine and cytochalasins A and B, J. Neurochem., 21 (1973) 1059-i 066. 18 Mori, H., Komiya, Y. and Kurokawa, M., Slow axoplasmic transport. Its asymmetry in two branches of bifurcating axons, Proc. Japan Acad., Set. B, 53 (1977) 252-256. 19 Mori, H., Komiya, Y. and Kurokawa, M., Slowly migrating axonal polypeptides. Inequalities in their rate and amount of transport between two branches of bifurcating axons, J. Cell Biol., 82 (1979) 174-184. 20 Mori, H. and Kurokawa, M., Purification of neurofilaments and their interaction with vinblastine sulphate, Cell Struct. Funct., 4 (1979) 163-167. 21 Paulson, J. C. and McClure, W. O., Microtubules and axoplasmic transport. Inhibitionof transport by podophyllotoxin: an interaction with microtubule protein, J. Cell BioL, 67 (1975) 461467. 22 Sj6strand, J. and Hansson, H.-A., Effect of colchicine on the transport ofaxonal protein in the retinal ganglion cells of the rat, Exp. eye Res., 12 (1971) 261-269. 23 Weisenberg, R. C., Borisy, G. G. and Taylor, E. W., The colchicine-binding protein of mammalian brain and its relation to microtubules, Biochemistry, 7 (1968) 4466-4479. 24 Wilson, L., Bryan, J., Ruby, A. and Mazia, D., Precipitation of proteins by vinblastineand calcium ions, Proc. nat. Acad. Sci (Wash.), 66 (1970) 807-814.
505
P R E F E R E N T I A L B L O C K A D E OF THE T U B U L I N T R A N S P O R T BY COLCHICINE
YOSHIAKI KOMIYA and MASANORI KUROKAWA Department of Biochemistry, Institute of Brain Research, Tokyo UniversityFaculty of Medicine, Hongo 7-3-1, Bunkyo ku, Tokyo 113 (Japan)
(Accepted November 1st, 1979) Key words: slow axoplasmic transport -- neurofilament polypeptides -- tubulins -- actin -- col-
chicine -- vinblastine sulphate - - cytochalasin D
SUMMARY L-[asS]Methionine was injected into the dorsal root ganglion (Ls) of the adult rat, and migration of the neurofilament polypeptides (the triplet with molecular weights of 200,000, 160,000 and 68,000 daltons), a- and fl-tubulins and actin in the sciatic nerve and the dorsal root was quantitatively determined and also examined by fluorography. Colchicine (4/~g) injected into the ganglion 10 min before methionine preferentially blocked the tubulin transport, with little if any blockade of the triplet and actin. Colchicine at this dose had no effects on the incorporation of L-[14C]leucine into the total protein and also into tubulins. In contrast to colchicine, vinblastine sulphate (4 #g) injected into the ganglion in a similar way blocked the transport of all the triplet, tubulins and actin. Cytochalasin D (1 /~g) had no effect on the slow axoplasmic transport.
INTRODUCTION Since the initial demonstration by Dahlstr6m 4 and Kreutzberg 14, interruption of axoplasmic transport by antimitotic agents has been documented under a variety of experimental conditions (for review, see ref. 8). It has been reported that rapid transport, including retrograde transport2,15, is almost completely blocked by colchicine, while blockade of slow transport is only partiala, 11,2~. However, some workers have observed that the slow transport is more profoundly affected by colchicine compared with the rapid transport10,17.
506 It has now been established that the major portion (approx. 75'!i;) of the slow axoplasmic transport consists of the triplet polypeptides, ~- and l#tubulins and actin 9,16,~s,~9. The initial suggestion that the triplet polypeptides are associated with neurofilaments 9 has been corroborated by a recent demonstration that the neurofilaments isolated from the peripheral nerve consist almost exclusively of three polypeptides with molecular weights of 200,000, 160,000 and 68,000 daltons z0. In addition, it is now possible to quantitate radioactivities associated with the triplet, tubulins and actin that are migrating within the axons ~9. In this situation, it is thought pertinent to re-examine the effects of antimitotic agents against the slow transport. Fluorographic examinations and quantitative determinations of the slowly migrating polypeptides in the present study indicate that colchicine preferentially blocks the transport of tubulins, while vinblastine blocks the transport of all the major polypeptides. Slow axoplasmic transport is not influenced by cytochalasin D. MATERIALS AND METHODS
Treatment of" the animal and nerve for the transport study Male albino rats of the Wistar strain, 9-week-old and weighing about 250 g, were used throughout. Under ether anaesthesia, dorsal root ganglia (bilateral L~) were exposed by a partial laminectomy, and L-[3~S]methionine (200 #Ci in 0.4 ul) was injected via a glass capillary (20-50 /~m in tip width) into the ganglion. In the experiments indicated, colchicine (4 or 8 #g in 0.2/~1), vinblastine sulphate (4/~g in 0.25/~1) or cytochalasin D (1/~g in 0.2 #1) was injected 10 rain before the injection Of L[aSS]methionine. In all the experiments, the contralateral ganglion that received the solvent served as control. At days 4, 7, 9, 14 and 28 after the injection, animals were killed by decapitation and the dorsal root (central axons), L5 ganglion and the sciatic nerve (peripheral axons) were dissected out. The tissue was placed on an ice-cold plastic plate, cut into consecutive 3 mm segments, and kept frozen at --80 °C until treated as below. Each segment was ground in a small teflon-glass homogenizer in 200/A of the medium which consisted of sodium dodecyl sulphate (2.3 o~, w/v), fl-mercaptoethanol (5 raM), glycerol (10 ~,, w/v) and Tris.HCl (62.5 raM), pH 6.8. The homogenate was centrifuged at 1500 × g for 5 rain at 25 °C, and the clear supernatant was subjected to further analyses. SDS-polyacrylamide slab gel electrophoresis, radiometry and fluorography of polypeptides migrating within the axons The nerve extract was incubated at 37 °C for 15 min and then heated in a boiling water bath for 5 rain, and an aliquot (50 #1) was placed in a counting vial. After adding 0.2 ml of Soluene-350 (Packard Instrument Co., Downers Grove, Ill., U.S.A.), the vial was heated at 50 °C for 2 h, and the radioactivity was measured in a liquid scintillation spectrometer using toluene-based scintillation fluid 1,13,
507 Aliquots from the two adjacent segments (20 #1 each) were combined and subjected to electrophoresis and subsequent fluorography as described previously19. In order to determine radioactivities associated with the triplet polypeptides, aand fl-tubulins and actin separately, the tissue was taken at day 9 postlabelling. Aliquots of the nerve extract from segments C2-C10 were combined; aliquots were combined also from P2-Pzs (Cx and Px denote the x-th central and peripheral 3 mm segments, respectively). Samples C2-C10 (in duplicate), C1, ganglion, Pl and P~Pzs (in duplicate) were electrophoresed in a single slab gel (with 10 mm wide slots), and the gel was stained with Coomassie brilliant blue. Portions stained for the triplet, tubulins and actin were cut off, and respective radioactivities were determined as described previouslylL Amino acid incorporation studies Colchicine (4 #g in 0.2 #1) and L-[14C]leucine (0.8/~Ci in 0.2 #1) were injected into the L5 ganglion as described above. The contralateral ganglion that received the solvent and L-[14C]leucine served as control. At minutes 15, 30 and 45, and at hours 1, 1.5, 2, 4, 8 and 24 after the isotope injection, the animals were killed by decapitation and the ganglion was removed together with 3 mm long central and peripheral stumps of axons. The tissue was ground in a teflon-glass homogenizer in 400/zl of the above-mentioned medium, centrifuged at 1500 × g for 5 min, and the clear supernatant was used for the following analyses. An aliquot (20 #1) was used to determine the total radioactivity. Another aliquot (30 #1) was mixed with the equal volume of 20 ~ (w/v) trichloroacetic acid and centrifuged, and radioactivities in acidsoluble and acid-insoluble fractions were determined. The third aliquot (200 #1) was electrophoresed and radioactivities associated with the triplet polypeptides, tubulins and actin were separately determined as described above. Electron microscopy Colchicine (4 #g) was injected into the L5 ganglion and, 7 days later, the animal was perfused through the heart with phosphate-buffered saline, and prefixed by perfusion through the heart with 2 ~ formaldehyde-2.5~ glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4. The L5 ganglion and nerve were removed and further fixed overnight in the cold. The tissue was then washed with 10~ (w/v) sucrose in 0.1 M cacodylate buffer, pH 7.4, for 10 min, and postfixed with 1 ~ OsO4 in 0.1 M cacodylate buffer, pH 7.4, for 2 h in the cold. The tissue was washed with water for 10 min, stained en bloc with 0.5 ~ aqueous uranyl acetate for 2 h at room temperature, dehydrated with several changes of ethanol of gradient concentrations, and embedded in Epon 812. Thin sections were cut with a diamond knife, doubly stained with lead citrate and uranyl acetate, and examined in a Hitachi ll-DS electron microscope at 75 kV. Chemicals L-[~S]Methionine (800-1200 Ci/mmol) and L-[14C]leucine (300 Ci/mmol) were obtained from Radiochemical Centre, Amersham, U.K. The label was taken to
508 dryness in vacuo, and brought to a final concentration of 500 mCi/ml for L-[:~:~S]methionine and 4 mCi/ml for e-[14C]leucine by the addition of distilled water. Colchicine was dissolved in dimethyl sulphoxide to 1 M, and stored at --20 *'C in the dark. On the day of experiment, the stock solution was diluted about 20 times with physiological saline. Cytochalasin D was also dissolved in dimethyl sulphoxide to 10 mg/ml and stored. Vinblastine sulphate was dissolved in saline to 16 mg/ml immediately before the injection. Other chemicals were of analytical grade. RESULTS
Since all the animals that received 8/~g of colchicine died within several days after the injection, the results described below are those obtained in animals that received 4 #g of colchicine.
Effects of colchicine on amino acid incorporation into proteins The time course of incorporation of L-[14C]leucine into the acid-insoluble fraction and into tubulins is not affected by colchicine treatment (Fig. 1A, B). Tubulins/ triplet and tubulins/actin ratios do not differ between the control and colchicinetreated ganglia (Table i), indicating that colchicine does not alter the incorporation of L-[14C]leucine into any of these polypeptides. I00
B
A 80
=* 60 "5
"g 4 0
o
2!
o n,
/
./.
20
r
O
I
I
T
~
O
r
O
i
r
I
Time after isotope injection (hours)
Fig. 1. Effects of colchicine on the incorporation of L-[14C]leucine into the acid-insoluble fraction (A) and into tubulins (B). The L5 ganglion was labelled with L-[14C]leucine (0.8/~Ci) and, at the time points indicated, the ganglion together with the peripheral and central stumps of axons (3 m m each) was removed and processed for determination of radioactivities associated with acid-insoluble materials (A) and with tubulins (B). Colchicine (4/~g) was injected into the ganglion 10 min before the isotope. O ©, control; • • , colchicine-treated. Determinations made at 4, 8 and 24 h postlabelling gave values practically identical with those obtained at 2 h.
509 TABLE I
Effects of colchicine on the incorporation of L-[laC]leucine into the neurofilamentpolypeptides (triplet), tubulins and actin in the L5 ganglion The radioactivity measurements were made with samples obtained at minutes 15, 30 and 45, and at hours 1, 1.5, 2, 4, 8 and 24 postlabelling, and means of respective ratios ± S.E.M. are given (see text; cf. Fig. 1). Colchicine (4/~g) was injected into the ganglion 10 rain before the isotope.
Control Colchicine-treated
Tubulins/triplet
Tubulins/actin
0.88 ± 0.02 (9) 0.86 ± 0.06 (9)
1.43 ± 0.05 (9) 1.41 :k 0.03 (9)
Effects of colchicine, vinblastine and cytochalasin D on slow axoplasmic transport Flow profiles of labelled proteins in the control, colchicine-treated and vinblastine-treated nerves are shown in Fig. 2. When the inhibitory effect of the drug is expressed in [1--(counts in drug-treated nerve/counts in drug-treated ganglion)/ (counts in control nerve/counts in control ganglion)] × I00, colchicine proves to inhibit the transport by 30-50 % and vinblastine by approx. 70 %. Fluorographic analyses demonstrate that colchicine preferentially blocks the transport of tubulins (Fig. 3A, B); transports of the triplet and actin are apparently unaffected. This visual impression is further supported by quantitative determination showing that, in both the central and peripheral axons, the tubulins/triplet ratio is I00
;',?/ ill l~o
50
//
- 30
/
20
~to
..~
/ 2//
".~, u
#
J
s
\
F',/ ~.
3 Central i 30
I
I
~
I
/ I5
I
i
I
(~ I
Distance
I 0
I K " V ""IDx'~ D ~
Peripheral ' from
'
'
" "X "'A~ ~ ~ -A.-~ '
I 15
l
'
'
E 3 l0
I''
I
r
the ganglion (mm)
Fig. 2. Flow profiles of radioactivities in the dorsal root (central axons) and the sciatic nerve (peripheral axons) 7 days after labelling the L5 ganglion with 200/~Ci of L-[aSSlmethionine. Colchicine (4/~g) or vinblastine sulphate (4/~g) was injected into the ganglion 10 min before the isotope. (3(3, control ; O - - - O , colchicine-treated; A - - - A , vinblastine-treated. Vertical axis indicates [radioactivities in 3 mm segment/radioactivities in the ganglion] × 100.
510 significantly decreased by colchicine treatment, while the actin/triplet ratio remains virtually unaffected (Table II). In contrast to colchicine, vinblastine blocks the transport of the triplet, tubulins and actin without distinction (Fig. 3C). Cytochalasin D does not affect the slow transport as visualized by fluorography (Fig. 3D).
200
68 56 53 tO "10 I
0 ×
° ~
•
200-
"5
160-
o 0
68565343-
I
C
I
®
''
'
' ~ '
'
'
'6'o
P
Distance from the ganglion (mm/
0
~
~
~
--
~
-.
..~;
~'~
..~
e~..~
e
,~. o ~. =.
~4~ S" =
°"
"'~
0
~" ~ ~. ~ o~
o
(10,
eo,
o
0) 00
0
--
h)
Molecular weights( × 10 -3 daltons)
O) O) O0
512 TABLE II Quantitative determinations of radioactivities associated with the triplet, tubulins and actitl that are migrating in the central and peripheral axons of the L5 ganglion cells
The tissue was removed 9 days after labelling the ganglion with L-[35S]methionine,and processed for radioactivity measurement as described in the text. Colchicine was injected into the ganglion 10 min before the isotope. Means of 4 determinations ± S.E.M. are given. Central
Peripheral
Tubulins/triplet
Control Colchicine-treated
1.20 ± 0.02* 0.72 - 0.05
1.43 J: 0.03* 0.71 ± 0.07
Actin/triplet
Control Colchicine-treated
0.57 ± 0.05 0.67 ± 0.07
0.45 ± 0.01 0.43 ± 0.05
* The difference between the control and colchicine-treated animals with respect to the tubulin/triplet ratio is significant at P < 0.01 (Students t-test). Electron microscopic observations o f the colchicine-treated nerve
Electron microscopic examinations of the ganglion show that neurones and periaxonal satellite cells are well preserved 7 days after the treatment, though chromatolytic changes are apparent in some cells. At the same time point, examinations of the sciatic nerve 8 m m away from the ganglion show occasional intact myelinated axons (Fig. 4A), degenerating myelinated axons (Fig. 4B, C) and myelin fragments phagocytosed by Schwann cells. However, intact unmyelinated sensory axons are hardly visible. In contrast, motor axons (both myelinated and unmyelinated), which can be well demarcated from sensory axons at this site of the nerve, remain intact (Fig. 4D). DISCUSSION Our present results clearly indicate that colchicine (4 big) preferentially blocks the intra-axonal migration of tubulins, in contrast with vinblastine (4 #g), which indiscriminately blocks the migration of all the triplet, tubulins and actin. We were unable to examine the effects of the higher dose (8 b~g) of colchicine because all the animals that received this dose died within several days. Effects of lower doses of vinblastine have not been studied in detail. However, the non-selectivity ofvinblastine action shown in the present experiment is not unexpected in that vinblastine interacts not only with tubulins, but also with actin 24 and the neurofilament triplet 2°, in contrast to colchicine which interacts rather specifically with tubulin dimer 23. Cytochalasin D does not affect the slow axoplasmic transport, in agreement with previous observations 17. Reduction of the amount of migrating tubulins by colchicine is not a consequence of their reduced synthesis in neurone soma (Fig. 1B; Table I). On electron microscopy, neurones and periaxonal satellite cells are found well preserved 7 days after the application of colchicine, though chromatolytic changes are apparent in a
513
Fig. 4. Electron micrographs of the colchicine-treated sciatic nerve. Colchicine (4 #g) was injected into the L~ ganglion, and the ganglion and nerve were removed 7 days later after prefixation by perfusion. Sections were prepared from the nerve site 8 m m away from the ganglion. A: intact myelinated axon, x 18,000. B: partially disrupted myelinated axon, x 12,000. C: highly disrupted myelinated axon, x 6800. D : Myelinated and unmyelinated motor axons, x 12,000.
514 limited number of cells. Tubulins can be synthesized in both neurones and glial cells in the ganglion. However, because of a more rapid metabolic turnover of proteins in neurones than in glial cells 12, a large proportion of tubulins that are synthesized in an early period after labelling the ganglion is considered to be of neuronal origin (cf. ref. 19). In control nerves, the amounts oftubulins that leave the soma and are loaded on the axoplasmic transport are more than 4 times as much as those that stay in the ganglion. In colchicine-treated animals, de novo synthesis of tubulins remains unaltered, and the amount of tubulins that leave the soma is reduced by 30-501',i. Under these circumstances, an accumulation of tubulins should be detected within the ganglion. However, our quantitative determination of radioactivities associated with tubulins in the ganglion failed to detect any such accumulation. It thus seems likely that tubulin dimer molecules which bind colchicine, and accordingly become unable to undergo polymerization, are easily destroyed. Blockade of tubulin transport by colchicine is not complete under our present experimental conditions. Thus, even in the colchicine-treated nerve, a proportion of tubulins is found to migrate at a rate apparently identical with that in the normal nerve (Fig. 3A, B). At least 4 possibilities exist to explain this finding. First, colchicine in higher doses may cause a complete blockade, though we were unable to confirm this, being hampered by a high mortality of animals treated with 8 #g of colchicine. Second, an amount of tubulins may have been synthesized before colchicine starts to exert its inhibitory action; colchicine is reported to require a certain lag period for exhibiting its maximal effect 21, possibly because of a slow process of colchicine-tubulin interaction. Third, assembly and disassembly studies of tubulins and microtubuleassociated proteins in vitro have indicated that there exist at least two types of tubulin-microtubule equilibria, with different sensitivities to colchicine 7. Fourth, the myelinated and unmyelinated axons may differ in their sensitivity to colchicine. This possibility is inferred from the electron microscopic observations that unmyelinated sensory axons are practically absent under conditions where myelinated axons, though more or less disrupted, are largely preserved (Fig. 4A-C). The failure to detect unmyelinated sensory axons is not due to technical insufficiencies, because unmyelihated m o t o r axons are typically demonstrated in the same section of the colchicinetreated nerve (Fig. 4D). The rate and amount of migration of the triplet polypeptides are apparently unaffected by colchicine (Fig. 3A, B), and this is not surprising in view of the lack of interaction between the isolated neurofilaments and colchicine 20. However, there are at least two factors that complicate the interpretation of behaviours of the 10 nm filaments in vivo. First, the neurofilament/microtubule ratio is relatively higher in myelinated than in unmyelinated axons. If myelinated axons are in fact more resistant to colchicine as inferred above, then subtle changes in any of the triplet polypeptides may be overlooked or underestimated. Second, the number of the 10 nm filaments in axons has been reported to increase under the influence of colchicine 6. A glance at Fig. 4A and Fig. 4B gives an impression that 10 nm filaments are increased in degenerating axons. However, deliberate quantitative analyses are needed to conclude that l0 nm
515 filaments are in fact increased in number, even when the filament density in degenerating axons is corrected for the shrinkage o f cross-sectioned area. I f the transport o f tubulins is preferentially or differentially affected by colchicine, this suggests that, in some situations, the transport o f neurofilament triplet (and o f actin) does not require coordinate transport oftubulins. A n apparently reverse situation is f o u n d in the crayfish nerve cord, where the slow (and fast) transport is shown to proceed in the absence o f neurofilaments 5. However, this does not preclude the possible interaction a m o n g the triplet, tubulins and actin migrating within the normal m a m m a l i a n axons. ACKNOWLEDGEMENTS We thank Dr. H. Ishikawa and Mr. S. Tsukita for preparing the electron micrographs (Fig. 4) and for their valuable discussions. Thanks are also due to the Shionogi Co. for its generous gift o f vinblastine sulphate. This w o r k was supported in part by Research Grants from the Ministry o f Education, Japan, and by a grant from the Mitsubishi Foundation.
REFERENCES 1 Abe, T., Haga, T. and Kurokawa, M., Rapid transport of phosphatidylcholine occurring simultaneously with protein transport in the frog sciatic nerve, Biochem. J., 136 (1973) 731-740. 2 Abe, T., Haga, T. and Kurokawa, M., Retrograde axoplasmic transport: its continuation as anterograde transport, FEBS Lett., 47 (1974) 272-275. 3 Boesch, J., Marko, P. and Cu6nod, M., Effects of colchicine on axonal transport of proteins in the pigeon visual pathways, Neurobiology, 2 (1972) 123-132. 4 Dahlstr6m, A., Effect of colchicine on transport of amine storage granules in sympathetic nerves of rat, Europ. J. PharmacoL, 5 (1968) 111-113. 5 Fernandez, H. L., Burton, P. R. and Samson, F. E., Axoplasmic transport in the crayfish nerve cord. The role of fibrillar constituents of neurons, J. Cell BioL, 51 (1971) 176-192. 6 Fink, B. R., Byers, M. R. and Middaugh, M. E., Dynamics of colchicine effects on rapid axonal transport and axonal morphology, Brain Research, 56 (1973) 299-311. 7 Haga, T. and Kurokawa, M., Microtubule formation from two components separated by gel filtration of a tubulin preparation, Biochim. biophys. Acta (Amst.), 392 (1975) 335-345. 8 Hanson, M. and Edstrfim, A., Mitosis inhibitors and axonal transport, Int. Rev. CytoL, Suppl. 7 (1978) 373-402. 9 Hoffman, P. N. and Lasek, R. J., The slow component of axonal transport. Identification of major structural polypeptides of the axon and their generality among mammalian neurons, J. Cell BioL, 66 (1975) 351-366. 10 James, K. A. C., Bray, J. J., Morgan, I. G. and Austin, L., The effect of colchicine on the transport of axonal protein in the chicken, Biochem. J., 117 (1970) 767-771. 11 Karlsson, J.-O., Hansson, H.-A. and Sjtistrand, J., Effect of colchicine on axonal transport and morphology of retinal ganglion cells, Z. Zellforsch., 115 (1971) 265-283. 12 Koenig, H., An autoradiographic study of nucleic acid and protein turnover in the mammalian neuraxis, J. biophys, biochem. CytoL, 4 (1958) 785-792. 13 Komiya, Y. and Kurokawa, M., Asymmetry of protein transport in two branches of bifurcating axons, Brain Research, 139 (1978) 354-358. 14 Kreutzberg, G. W., Neuronal dynamics and axonal flow. IV. Blockage of intra-axonal enzyme transport by colchicine, Proc. nat. Acad. Sci.(Wash.), 62 (1969) 722-728. 15 Kristensson, K. and Sjtistrand, J., Retrograde transport of protein tracer in the rabbit hypoglossai nerve during regeneration, Brain Research, 45 (1972) 175-181. 16 Lasek, R. J. and Hoffman, P. N., The neuronal cytoskeleton, axonal transport and axonal growth.
516 In R. Goldman, T. Pollard and J. Rosenbaum (Eds.), Cell Motility, Cold Spr. Harb. Laboratory, New York, 1976, pp. 1021-1049. 17 McGregor, A. M., Komiya, Y., Kidman, A. D. and Austin, L., The blockage of axoplasmic flow of proteins by colchicine and cytochalasins A and B, J. Neurochem., 21 (1973) 1059-i 066. 18 Mori, H., Komiya, Y. and Kurokawa, M., Slow axoplasmic transport. Its asymmetry in two branches of bifurcating axons, Proc. Japan Acad., Set. B, 53 (1977) 252-256. 19 Mori, H., Komiya, Y. and Kurokawa, M., Slowly migrating axonal polypeptides. Inequalities in their rate and amount of transport between two branches of bifurcating axons, J. Cell Biol., 82 (1979) 174-184. 20 Mori, H. and Kurokawa, M., Purification of neurofilaments and their interaction with vinblastine sulphate, Cell Struct. Funct., 4 (1979) 163-167. 21 Paulson, J. C. and McClure, W. O., Microtubules and axoplasmic transport. Inhibitionof transport by podophyllotoxin: an interaction with microtubule protein, J. Cell BioL, 67 (1975) 461467. 22 Sj6strand, J. and Hansson, H.-A., Effect of colchicine on the transport ofaxonal protein in the retinal ganglion cells of the rat, Exp. eye Res., 12 (1971) 261-269. 23 Weisenberg, R. C., Borisy, G. G. and Taylor, E. W., The colchicine-binding protein of mammalian brain and its relation to microtubules, Biochemistry, 7 (1968) 4466-4479. 24 Wilson, L., Bryan, J., Ruby, A. and Mazia, D., Precipitation of proteins by vinblastineand calcium ions, Proc. nat. Acad. Sci (Wash.), 66 (1970) 807-814.