Acta Physiol Scand 1991, 142, 181-189

ADONIS 000167729100108K

A calmodulin inhibitor w i t h high specificity, compound 48/80, inhibits axonal transport in frog nerves without disruption of axonal microtubules P. A. R. E K S T R O M , M. WALLIN", M. K A N J E and A. E D S T R O M Department of Animal Physiology, University of Lund, Sweden

EKSTROM,P. A. R. WALLIN,M.*, KANJE,M. & EDSTROM, A. 1991. A calmodulin inhibitor with high specificity, compound 48/80, inhibits axonal transport in frog nerves without disruption of axonal microtubules. Acta Physiol Scund 142, 181-189. Received 2 April 1990, accepted 29 January 1991. ISSN 0001-6772. Department of Animal Physiology, University of Lund and Department of Zoophysiology, University of Goteborg, Sweden. The calmodulin inhibitor compound 48/80 has previously been shown to arrest axonal transport in vitro in the regenerating frog sciatic nerve. The inhibition was limited to the outgrowth region of nerves, which had been allowed to regenerate in vivo for 6 days after a crush lesion, before they were incubated with or without drugs in vitro overnight. The effects of compound 48/80 on the regenerating nerve were further investigated. A concentration of compound 48/80 (50 pg ml-'), which effectively inhibits axonal transport, did not cause observable changes of the microtubules of regenerating axons in the outgrowth region as judged by electron microscopy. Furthermore, it was shown that also a lower concentration (25 pg ml-') inhibited axonal transport. As a measure of possible metabolic effects, the level of ATP was assessed in the regenerating nerve after exposure to compound 48/80. Compound 48/80 at 25 pg ml-' did not change the level of ATP in the nerve. The assembly of bovine brain microtubule proteins in a cell-free system was unaffected by 25 pg ml-' of compound 48/80 and slightly inhibited by 50 pg ml-'. At higher concentrations ( > 100 pg ml-') assembly of microtubules appeared stimulated, and microtubule spirals as well as closely aligned microtubules could be seen. These effects appeared to be unrelated to the transport effects. The present results indicate that compound 48/80 arrests axonal transport via mechanisms other than destruction of axonal microtubules or interference with the energy metabolism. It is possible that these mechanisms involve inhibition of calmodulinregulated events essential to the transport. Key words : axonal transport, calmodulin, compound 48/80, frog sciatic nerve, microtubules, microtubule proteins.

T h e Ca2+-binding protein calmodulin (CaM) is involved in the regulation of a number of biological functions (Manalan & Klee 1984) and has been shown to be present in axons of central as well as peripheral neurons (Erickson et al.

* Department of Zoophysiology, University of Goteborg, Sweden. Correspondence : Dr Per Ekstrom, Department of Animal Physiology, University of Lund, Helgonavagen 3B, S-223 62 Lund, Sweden.

1980, Brady et al. 1981, Caceres et al. 1983, Mata & Fink 1988). I t has been suggested that C a M has a role also in fast axonal transport (AXT) (Ochs 1981, Ekstrom et al. 1982, Lavoie & Tiberi 1986). T h e latter is supported by the ability of a variety ofCaM-inhibitors to block A X T (Edstrom et al. 1973, Ekstrom et al. 1982, Kanje et al. 1982, Lavoie & Tiberi 1986, Ekstrom et al. 1987). Recently, this suggestion was further strengthened when one of the most potent and specific CaM-inhibitors as yet described, com-

181

182

P.A . R. Ekstriim et al.

pound 48/80 (48/80, a condensation product of formaldehyde) (Gietzen 1983, Gietzen et al. 1983), was shown to block .WT in z'itro in the outgrowth region of t h e regenerating frog sciatic nerve (Edstrom et uI. 1987). Other C a M inhibitors, such as the phenothiazines trifluoperazine ( T F P ) a n d chlorpromazine (CPZ), have been shown to block AXT but also to reduce the n u m b e r of axonal M T s (Edstrom et 01. 1973, Kanje e/ rrl. 1982). TFP and CPZ show less Ca3.l specificity than 48/80 (Gietzen et 01. 1983) and exhibit metabolic or other effects unrelated to inhibition of Cab1 (Ruben & Rasmussen 1981, L u t h r a 1982, Vale et a / . 1983). Although it is likely that TFP and CPZ have exerted CaM inhibitory effects in the nerve cells in these experiments (Edstrom et al. 19i.3, Ekstrom et ul, 1987), it is still possible that the effect on the MTs is unrelated t o C a M inhibition. I f the latter is true, it is difficult t o suggest C a M involvement in AXT from these results since AXT is dependent on MTs. 'To learn more of the relation between C a b 1 and .WT it is t h u s important to establish whether or not effects on $ITS are obligatory when A X T is arrested by a highly specific C a M inhibitor such as 48/80. O u r experiments with 48/80 on the regenerating frog sciatic nerve were thcrefore in the present study extended to include investigations of effects on axonal MTs. We also measured the level of ATP in different nerve regions as a test of possible metabolic interactions of 48/80. Finally, we studied t h e effects of 48/80 on the assembly of bovine brain MT proteins in a cell-free s y t e m .

>I . l T E R I A L S ,J. ND Sf E T H 0 D S The nerae preparatron. Frogs (Rana temporaria) were kept at +20 "C. Regeneration was initiated in animals anaesthetized with MS222. Both sciatic nerves were exposed and subjected to a crush lesion at the level of the thigh with a pair of watchmaker's forceps. After regeneration had been allom-ed to progress for six days in z'iao the animals were killed and both sciatic ncrves, including the spinal ganglia (nos. 8 and 9). uere isolated. .tletabolic labelling .f' phosphoproteins and elertrophoresis. To test if 48/80 gained entrance to the cells of the preparation its effect on protein phosphorylation was studied. Two pairs of crushed and regenerating sciatic nerI-es were incubated for 18 h at 18 "C in frog Ringer solution with or n-ithout drugs. The frog Ringer solution was of the following composition

(mbi): NaCI 111.1, KCI 1.9, CaCI, 1.1, MgSO, 1.6, NaHPO,, 0.45, Na,HPO, 2.6 and glucose 5.5. T h e solution was adjusted to p H 7.4 and gassed with oxygen before use. T h e preparations were next rinsed briefly in Hepes-buffered Ringer (as frog Ringer but with 1 0 m ~of Hepes instead of NaHPO, and Na,I-IPO,) and incubated for 1 h at room temperature in Hepes-buffered Ringer, with or without additions as indicated, containing 400 pCi ml-' [32P]orthophosphate (Amersham, U.K.) to allow metabolic labelling of phosphoproteins. At the end of incubation a 6 mm segment immediately distal to the crush was taken from each nerve. The segments were disintegrated in electrophoresis sample buffer (0.0625 M Tris-HCI, p H 6.8, 1 sodium-dodecyl-sulphate [SDS], 2"" 2/%mercaptoethanol. O.Olo/;, hromophenol blue, 10°i, glycerol) and the samples were thereafter subjected to electrophoresis in 5.-ljO;, linear gradient, discontinuous SDS-polyacrylamide gels. The fixed and dried gels were put on Hyperfilm /1'-max (Amersham, U.K.) for 1-3 days before development. Electron microscop.y. The nerve preparations were, after 6 days of regeneration in ciao, incubated overnight in frog Ringer solution with or without 50 j i g ml-' 48/80 after which they were fixed according to McDonald (1984) with small modifications. The nerves were slightly stretched and fixed at room temperature in 2.5 O 0 glutaraldehylde and 4(?" paraformaldehyde in 0.1 M cacodylate buffer. After 30 s when the nerves had become stiff, pieces were cut and put into small glass bottles with fresh fixative for another 2 h. After washing ( 3 x 10 min) in cacodylate buffer the pieces were postfixed in 0.5 '+& OsO, 0.8 nb K,,Fe(CN), in cacodj-late buffer for 1 h on ice. Subsequent to another wash the preparations were treated with 0 . 1 5 O , tannic acid for 5 min, washed in buffer and then in distilled water prior to incubation in 2"" aqueous uranyl acetate for 2 h. After another wash in distilled water the nerve pieces were dehydrated in a graded series of acetone and embedded in Epon-Araldite. Ultrathin sections were cut on an ultramicrotome (LKB, Sweden), mounted on formrar-coated copper grids and stained with lead and uranyl acetate and examined with a Zeiss EM10 electron microscope. ConntinK uf MTs. T h e interest was focused on regenerating axons approximately 2 mm distal to the crush lesion, i.e. in the outgrowth region where AXT in an earlier stud) was shown to be almost completely blocked by 50 ,ug ml-' 48/80 (Edstrom et a/. 1987). I n this region M T s were counted and the axonal cross section area measured. Regenerating axons were identified by their small diameter and localization between the Schwann cell basal lamina and the myelin sheath of degenerating axons. One hundred and twenty experimental axons and 109 control axons were studied. Each group was composed of axons

+

Compound 48/80 and a x o n a l transport from two different nerves. Cross-sectional areas of axons were determined on photocopies of the electron micrographs by cutting and weighing. Axonal transport. AXT was measured as previously described (Edstrom & Mattsson 1972, Edstrom et al. 1987). Sciatic nerves were isolated and mounted in a chamber, where the ganglia were separated from the rest of the nerve by a silicone grease barrier. ['HILeucine (140 Ci mmol-', Amersham, U.K.) was added to the ganglionic compartment (25 pCi ml-' in frog Ringer solution) and frog Ringer solution with or without drug to the nerve compartment. The preparations were incubated for 18 h at 18 "C. Subsequently, the nerves were cut into consecutive 2-mm segments and the TCA insoluble radioactivity of each segment, representing fast axonally transported proteins, was measured by liquid scintillation counting. 48/80 releases histamine from mast cells (Koibuchi et al. 1985). T o rule out this as the cause for the transport effects, AXT was measured in the presence of histamine. Furthermore, the amount of histamine in nerves incubated overnight with and without 48/80 was measured with a fluorimetric method by Professor H. Bergstrand, AB DRACO, Lund, Sweden. A TP determinations. Regenerating nerve preparations were incubated with or without 48/80 for 18 h after which pieces of interest were cut from the nerve. The pieces were one 6 m m nerve segment with its distal end about 20 mm proximal to the nerve crush, and one 6 mm segment immediately distal to the crush, which thus covered the outgrowth region. The pieces were subjected to nucleotide extraction in 0.2 ml of ice-cold 10% trichloroacetic acid (TCA) for 10 min. The TCA was removed with two volumes of a mixture of trioctylamine :freon (1 : 2). After a 10 min low-speed centrifugation the nucleotide containing upper phase was removed and stored at -24 "C until analysis. The thawed extracts were analysed by HPLC ion exchange chromatography as previously described (Ekstrom et al. 1987). The nerve pieces remaining after TCA-extraction were dissolved in 1 M NaOH and assessed for protein by the Bio-Rad protein assay. MT preparation. M T proteins were prepared from fresh bovine brain cortex in the absence of glycerol by two cycles of assembly-disassembly (Borisy et al. 1974, Deinum et al. 1982). The assembly buffer contained 0.1 M Pipes, 0.5 mM MgSO, and 1 mM G T P adjusted to pH 6.8 with NaOH. After the last disassembly the M T solution was clarified by centrifugation at 200,000 g for 30 min at 4 "C and dropfrozen in liquid nitrogen. Protein concentration of M T proteins. M T protein concentration was determined according to Lowry et al. (1951) with bovine serum albumin as a standard. Assembly nf MT proteins. Assembly of MT proteins diluted to 2 mg ml-' in assembly buffer was started by increasing the temperature from 4 to 37 "C. The

183

assembly was monitored by measuring the turbiditj as the apparent increase in absorbance at 350nm, A A,,,, in a spectrophotometer. 48/80 was dissolved in assembly buffer. Electron microscopy of M T proteins. Negativelystained specimens for electron microscopy were prepared after a 20-fold dilution of assembled MTs in assembly buffer containing 50% sucrose (w.v). A 5 pl protein sample was applied to a grid, dried after 20 s with filter paper and stained with 1yo uranyl acetate. The specimens were viewed in a Zeiss 109 electron microscope. Chemicals. 48/80, trifluoperazine and histamine were obtained from Sigma, MS222 from Sandoz and calmidazolium from Boehringer-Mannheim.

RESULTS

Electophoresis of metabolically labelled phosphoproteins It is shown in Figure 1 that incubation with 48/80 a t 25 pg ml-' affected the phosphorylation of proteins in the outgrowth region in a selective manner. M o r e specifically, 48/80 inhibited the phosphorylation of a 35 kDa protein. Incubation

Fig. 1. Effects of the CaM inhibitors 48/80 and trifluoperazine (TFP) on metabolic labelling of phosphoproteins in frog sciatic nerves, regenerating for 6 days in vivo. Such nerves were incubated for 18 h in buffer with or without drugs, followed by 1 h in buffer containing radiolabelled orthophosphate, again with or without drugs (see text). Nerve phosphoproteins from the outgrowth region were then analysed with gradient SDS-PAGE and autoradiography. Bars indicate the position of molecular weight standards, from top: 69 kDa, 46 kDa and 30 kDa. C, control nerve without drugs; 48/80, 25 pg ml-' 48/80; TFP, 0.05 mM trifluoperazine. Arrowheads indicate a phosphoprotein of approximately 35 kDa that disappears when either of the drugs is used.

184

P.A. R.Ekstrom et al.

Fig. 2. Electron micrograph of a 48/80 treated frog sciatic nerve showing regenerating axons. This cross-section is taken about 2 mm distal to the crush lesion on a nerve, that had been regenerating for 6 days in rive before incubation overnight at 18 "C in frog Ringer solution with 50 pg ml-* of 48/80. MT, microtubule; NF, neurofilament ; BL, basal lamina; MS, myelin sheath. with TFP, a phenothiazine derivative that inhibits CaM (see Introduction), at 0.05 miM did also result in inhibition of phosphorylation of this protein (Fig. 1). The labelled orthophosphate enters the intracellular ATP-pool, which serves as the donor of phosphate groups to proteins during the phosphorylation reactions. The phosphorylation we have studied in this system is thus an intracellular event. It is still possible, though, that 48/80 could affect intracellular processes via membrane effects due to its chemical nature. However, T F P affected the same protein phosphorylation as 48/80. This suggests that both drugs, which have CaM inhibition in common, have entered the cells and inhibited some CaM dependent protein kinase; it seems highly unlikely that two compounds as structurally different as 48/80 and T F P should affect the same, specific intracellular event via different, unspecific actions. Electron microscopy. M T-counting. Figures 2 and 3 show cross sections of axons from the

outgrowth region, approximately 2 mm distal to the crush, of regenerating frog sciatic nerves that had been incubated overnight with (Fig. 2 ) or without (Fig. 3) 48/80 at 50,ugml-l, a concentration known to almost completely inhibit AXT in this system (Edstrom et al. 1987). No differences in the microtubular morphology or in the intra-axonal ultrastructure could be observed when 48/80-treated nerves were compared with control nerves. T h e average number of M T s was not different in 48/80-treated axons and control axons respectively (Table 1). Nor were there any significant differences when selected groups of axons (i.e. axons with a cross-sectional area less or more than 0.1 pm2 etc.) were compared. No changes in the morphology of M T s or other axonal structures could be observed in longitudinal sections from the outgrowth region or in cross-sections taken proximal to the crush. Neurofilaments were not counted but did not appear to be affected by treatment with 48/80 in any of the segments studied.

Compound 48/80 and axonal transport

185

Fig. 3. Electron micrograph of a control frog sciatic nerve showing regenerating axons. This crosssection is taken about 2 mm distal to the crush lesion on a nerve, that had been regenerating for 6 days in vivo before incubation overnight at 18 "C in normal frog Ringer solution. MT, microtubule; NF, neurofilament ; BL, basal lamina ; MS, myelin sheath.

Table 1. Effects of 48/80 on ATP levels and microtubules in the regenerating frog sciatic nerve ATP, nmol/mg protein

48/80 Control Experiment of control (yo)

20 mm proxima1 to crush

Outgrowth region

MTs/axon

1.33k0.32 (6) 1.67 rf: 0.72 (6) 80"

1.43k0.11 (6) 1.33 rf: 0.23 (6) 108"

9.02f0.74 (120) 7.54f0.68 (109) 119"

= 48/80 not significantly different from control (student's r-test) ATP-levels and number of axonal microtubules in regenerating frog sciatic nerves. Segments used for ATP measurements were 6 mm long. Microtubules were counted approximately 2 mm distal ro the crush of nerves that had been allowed to regenerate for 6 days in vzvo prior to incubation for 18 h at 18 "C in frog Ringer with or without 48/80. ATP experiments = 25 pg ml-' 48/80, M T experiments = 50pg ml-' 48/80. Values are means k SEM. a

Separate experiments indicated that MTs remained unaffected also in nerves treated with a higher concentration (100 p g ml-l) of 48/80 overnight.

lransport Preparations of regenerating frog sciatic nerves were incubated for 18 h a t 18 "C with [3H]leucine

P.- 4 . R. Ekstrom

186

et al. 0Compound 48/80 0.6

0 Normal Ringer

04 0

ln

c)

: 02 Proximal Distance (mm) from crush lesion

Fig. 4. The effect of 48/80 on AXT in regenerating frog sciatic nerves, that had been regenerating for 6 da!-s i'iz'o prior to incubation in z'itru. Radioactive protiles of [ "H1leucine labelled proteins were examined after incubation for 18 h a t 18 "C in frog Ringer with or without (0) 25 ,ug m1-l of48/80. Radioactivity per 2 mni of ncrve is expressed as a percentage of the total radioactivity in the nerve between points -4 and B. :!berage of 8 nerveskSE3I. Student's t-test for paired samples was used for statistical e\-aluation of difference5 hemeen 48/80 treated and control nerve piccc.;. * = significant difference, P < 0.05.

(u)

present in the ganglionic compartment and 48/80 o r normal Ringer in the nerve compartment. Figurc 4 shows the distribution of radioactivity, measured by scintillation counting, in 2 m m segments proximal and distal to the crush lesion. A concentration of 2.5 p g mlF' of 48/80 significantll- reduced A X T in the outgrowth region. I he frog sciatic nerve contained approximatel!- 3 ng histamine per 12 m m segment both after incubation for 18 h in normal frog Ringer and in Ringer containing 50 j i g ml-' 18/80. It is therefore not likely that AXT was inhibited b!48/80 due to release of histamine. I n support of this A X T was also unaffected in nerve preparahi histamine tions directly exposed to 3 x (not shown). This concentration is at least 100 times higher than that which should be attained ifall sciatic nerve histamine (l(k-12 ng per whole nerie) was released into the incubation medium. r .

4 TI' rlrtemnitnatrntir 'Ihc leicl of ATP was measured in different 6 mrn segments of the regenerating frog sciatic

0

5

10

Time (min)

Fig. 5. Effect of 48/80 on the assembly of M T proteins ( 2 mg ml-') in 100 mM Pipes, 0.5 mM MgSO,

and 1 mhi G T P at pH 6.8. Assembly was initiated by increasing the temperature from 4 to 37 "C. The assembll- was monitored by the increase in turbidity, A in the absence (a), or in the presence of .50 p g mi-' (b), 100 y g ml ( c ) , 150 p g m l ~ ' (d), 200 pg nil ~'(e) and 250 p g ml (9 of 48/80. nerie after treatment with 48/80. At 25 p g nil-' of 48/80 A T P levels were not reduced in either segment (Table 1).

if T [issrmbl)~

48/80 at 25 p g ml-' had no effect on the assembly of hlTs. A slight inhibition was observed at 50 ltg ml-' (Fig. 5 ) . This concentration had no effects on the morphology of MTs. Higher concentrations (10&250 p g m1-l) stimulated the assembly of MTs and induced formation of spiralized M T sheets as well as MTs that were closely aligned (Fig. 6 a x ) . T h e formation of these structures could in part be responsible for the stimulation of assembly as judged by turbidimetry. To test if these effects were due to interactions with residual CaM in the preparation another CaM inhibitor of similar potency, calmidazolium (Gietzen 1983), was used. Calmidazolium at equivalent concentrations did not affect the assembly of MTs as determined by turbidimetry.

?ompound 48/80 and axonal transport

187

This suggested that interactions between 48/80 and M T proteins were unrelated to CaM.

DISCUSSION The present paper shows that fast axonal transport (AXT) in vitro is inhibited by 25 pg ml-I of the calmodulin (CaM) inhibitor compound 48/80 (48/80), a concentration only half of that previously reported to arrest AXT (Edstrom et al. 1987). The inhibition was limited to the outgrowth region of sciatic nerves, which had been allowed to regenerate in viva for 6 days after a crush lesion, before they were incubated in vitro over night. We were able to show that in contrast to the effects of other CaM inhibitors, such as trifluoperazine (TFP) and chlorpromazine (CPZ) (Edstrom et al. 1973, Ekstrom et al. 1987), incubation with 48/80 was not accompanied by disturbances of the axonal microtubules (MTs). These drugs are not as specific as 48/80 (Gietzen et al. 1983) and may have effects unrelated to CaM (Ruben & Rasmussen 1981, Luthra 1982, Vale et al. 1983). Furthermore, there is evidence that CaM activation causes disassembly of M T s (Marcum et al. 1978, Keith et al. 1986) and thus would an inhibition of CaM not be expected to result in disruption of MTs. In fact, CaM inhibitors have been shown to prevent breakdown of M T s by Ca2+/CaM (Schliwa et al. 1981). The effects on axonal M T s by CPZ and TFP (Edstrom et al. 1973, Ekstrom et al. 1987) may therefore have been unspecific to CaM inhibition. Metabolic effects by 48/80 leading to energy deprivation, related or unrelated to the inhibition of CaM, were most likely not the cause for the arrest of AXT, since 25 pug ml-' 48/80 did not reduce the level of ATP in the nerve. Furthermore, Dahl et al. (1982) have shown that energy deprivation leads to breakdown of axonal MTs, while we observed that M T s of regenerating axons remained intact after exposure to 50 and 100 pg ml-' 48/80. We also found 48/80 to interfere with Fig. 6. Effect of 48/80 on the morphology of assembled MTs. MT proteins were assembled in the absence (a) or in the presence of 200 pg ml-' (b and c ) of 48/80 and the samples were negatively stained. For experimental conditions see Fig. 4. Bar = 0.1 ,urn.

188

P. ,4. R. Ekstrom et

al.

and presence of microtubule-associated proteins. Biochim Biophys Acta 719, 370-376. A4,, EKSTROM, P., KANJE,M. & SJOBERG, J. EDSTROM, 1987. The use of the regenerating frog sciatic nerve for pharmacological studies of orthograde and retrograde axonal transport. Brain Res 401, 34-42. A,, HANSSON, H.-A. & NORSTROM, A. 1973. EDSTROM, Inhibition of axonal transport in vitro in frog sciatic nerves by chlorpromazine and lidocaine. A biochemical and ultrasuvctural study, Z Zellfrsch 143, 53-69. H. 1972. Fast axonal EDSTROM.A. & MATTSSON, transport in citro in the sciatic system of the frog. 3. j'Veurochern 19, 205-22 1. AXT. P., KANJE,M. & EDSTR~M, A. 1982. Effects In conclusion, the present results show that EKSTROM, of phenothiazines on axonal transport and micro48/80 arrests AXT via mechanisms other than tubule assembly in ritro. Acta Ph-ysiol Scand 116, destruction of intracellular MTs or interference 121-125. with the energy metabolism. T h e high specificity EKSTROM, M. 1987. The P., MCLEAN, W.G. & KANJE, of 48/80 towards C a M suggests that it has effects of trifluoperazine on fast and slow axonal interfered with AXT via an effect on intracellular transport in the rabbit vagus nerve. 3 Neurobiul 18, CaM-regulated events, for instance CaM-de283-294. pendent protein kinases, as indicated by experi- ERICKSON,P.F., SEAMON, K.B., MOORE, B.W., LASHER,R.S. & MINIER,L.N. 1980. Axonal ments concerning metabolic labelling of transport of the Ca2+-dependentprotein modulator phosphoproteins. T h e relation between the latter of 3'-5'-cyclic-AMP phosphodiesterase in the rabbit and AXT remains to be studied. visual system. 3 Neurochem 35, 242-248. GIETZEN, K. 1983. Comparison of the calmodulin The present work was supported by grants from antagonists compound 48/80 and calmidazolium. NFR, Crafoordska Stiftelsen, Lars Hiertas Mime and Biochem 3 216, 61 1-616. Magnus Bergvalls Stiftelse. P., WOTRICH, K., ADAMCZYK-ENGELMANN, We thank professor Hikan Bergstrand at AB GIETZEN, 4 . , KONSTANTINOVA, A. & BADER,H. 1983. DR.\CO, Lund, for his co-operation regarding the Compound 48/80 is a selective and powerful histamine measurements. Electron microscopy was inhibitor of calmodulin-regulated functions. performed at the Department of Zoology, University Biochim Bioph,ys Acta 736, 109-118. of Lund and the skilful help of Miss Peggy Petersen A. & EKSTROM, P. 1982. The KASJE,M., EDSTROM, is gratefully acknowledged. role of CaL- in rapid axonal transport. In: D.G. Weiss (ed.) Asoplasmic Transport, pp. 294-300. Springer Verlag, Berlin. A.S.,RATAN, R., MAXFIELD, F.R. KEITH,C.H., BAJER, R E FE R E N CE S & SHELANSKI, M.C. 1986. Calcium and calmodulin BORISY,G.G., OIAlSTED, J.R., MARCLM, J.%I. & in the regulation of the microtubular system. Ann AILEN, C. 1974. Microtubule assemblv in rirro. 1Yl' Arad Sci 466, 375-391. Fed. Pror. 33, 167-174. A., NAKAGAWA, M. & KOIBWHI,Y., ICHIKAWA, M.,HERIOT, K. & LASEK, R.J. BRADY,, S.T., TYTELI., TOMITA, K. 1985. Histamine release induced from 1981. Axonal transport of calmodulin : a ph!-siomast cells by active components of compound logical approach to identification of long-term 48/80. Eur 3 Pharmacol 115, 163-170. associations between proteins. 3. Cell Biol 89, LAVOIE, M. 1986. Inhibition of fast P.A. & TIBERI, 607-6 14. axonal transport in bullfrog nerves by dibenzazepine CACERES, .4., BENDER, P., SNAVELY, L., REBHUN, L.I. and dibenzocycloheptadiene calmodulin inhibitors. & STEWARD, 0. 1983. Distribution and subcellular 3 Yeurobiol 17, 681-695. localization of calmodulin in adult and developing LOWRY,O.H., ROSEBROUGH, N.J., FARR,A.L. & brain tissue. Neurosci 10, 449461. RANDALL, R.J. 1951. Protein measurement with the DAHL,N.A., LOONEY, G.ii. & BLACK,W.H. 1982. Folin phenol reagent. 3.Biol Chem 193, 265-275. Ultrastructure of non-myelinated neurons during LUTHRA,M.G. 1982. Trifluoperazine inhibition of energy deprivation. Acta Neuropathol (Bert) 57, calmodulin-sensitive Ca2+-ATPaseand calmodulin 111-120. insensitive (Na+-K+)- and Mg2+-ATPase activities L. & WALLIN, M. 1982. Effect DEINUM, J., SORSKOG, of human and rat red blood cells. Biochim Biophys of cibacron blue on tubulin assembly in the absence Acta 692, 271-277.

assembly of M T s in ritro. in that high concentrations ( > 100 pg m1-l) stimulated assembly, as judged by turbidimetry, and the formation of spirals and closely aligned MTs. However, no MT spirals, aligned or otherwise distorted MTs could be observed in regenerating axons that had been exposed to 50 or lOOpgml-' 48/80. I n addition, concentrations that arrest AXT, 25 and 50 pg ml-' (this paper, Edstrom et al. 1987), had n o or only slight effects on assembly. T h e effects of 48/80 on MT assembly, as observed in zitro, can therefore probably not explain the arrest of

Compound 48/80 and axonal transport MCDONALD, K. 1984. Osmium ferricyanide fixation improves microfilament preservation and membrane visualization in a variety of animal cell types. f Ultrastruct Res 86, 107-118. A.S. & KLEE,C.B. 1984. Calmodulin. Adv MANALAN, Cyclic Nucleotide Prot Phosph Res 18, 227-278. MARCUM, J.M., DEDMAN,J.R., BRINKLEY, B.R. & MEANS, A.R. 1978. Control of microtubule assembly-disassembly by calcium-dependent regulator protein. Proc Nat Acad Sci U.S.A. 75, 3771-3775. MATAM., &FINK,D.J. 1988. Calmodulin distribution in nerves: an EM immunocytochemical study. Brain Res 475, 297-304. OCHS,S. 1981. Characterization of fast orthograde

189

transport. Basic parameters. Neurosciences Res Prog Bull 20, 19-3 1. H. 1981. Phenothiazines and RUBEN,L. & RASMUSSEN, related compounds disrupt mitochondria1 energy production by a calmodulin-independent reaction. Biochzm Biophys Acta 637, 415-422. SCHLIWA, M., EUTENEUER, U., BULINSKI,J.C. & IZANT, J.G. 1981. Calcium lability of cytoplasmic microtubules and its modulation by microtubuleassociated proteins. Proc. Nat Acad Sci U.S.A.78, 1037-1041. VALE,M.G.P., MORENO,A.J.M. & CARVALHO, A.P. 1983. Effects of calmodulin antagonists on the active Ca2+ uptake by rat liver mitochondria. Biochem f 214, 929-935.