D I M E T H Y L S U L F O X I D E ACTION ON FAST AXOPLASMIC TRANSPORT AND U L T R A S T R U C T U R E OF VAGAL AXONS
JOSE ALEJANDRO DONOSO, JUAN-PABLO ILLANES and FRED SAMSON
Ralph L. Smith Research Center, University of Kansas Medical Center, Kansas City, Kans. 66103 (U.S.A.)
(Accepted May 17th, 1976)
SUMMARY
The axonal microtubules (MT) are believed to be involved in fast axonal trans- port (FAXT). Dimethylsulfoxide (DMSO) has a strong stabilizing action on MT in vitro which may account for some of its reported biological effects. DMSO at con- centrations of 5 ~ disrupts the FAXT in a high percentage of axons emanating from the nodosum ganglion in the cat vagus nerve. Whereas 5 ~ DMSO does not affect the FAXT in all axons, 10 ~ DMSO blocks all the FAXT. The blockage is substantial- ly, but not completely, reversed by washing the vagus for 2 h. DMSO at 2 ~/o caused no discernible change in either the FAXT or the axonal morphology, but some swelling of glial cells occurred. Ultrastructurally, 10 ~ DMSO caused some axons to swell and others to shrink. The MT appeared normal and their total number per axon did not change. The spatial relationship of the axonal constituents is clearly altered by the DMSO and this may have contributed to the failure of the transport. It is suggested that the DMSO, through strengthening the forces involved in polymerization, renders them non-functional for FAXT.
INTRODUCTION
Proteins and other macromolecules synthesized in the cell bodies of neurons are continuously transported along the axons. The proteins move at two rather distinct rates, slow (1-3 mm/day) and fast (100-500 mm/day)6,1z,1a,19,z3,29, 3°. These trans- ported materials probably are related to the maintenance of the structural and func- tional integrity of the axons, nerve endings and often the innervated cells. The molec- ular mechanisms for the specificity and energy coupling of transport(s) are still poorly understood. However, there is evidence that implicates the neurofibrillary structures within the axoplasm, especially the microtubules. The evidence for an in- volvement of microtubules in the fast axoplasmic transport is based primarily on the
288
blocking effects of antimitotic agents (colchicine, vincristine, vinblastine) which bind somewhat specifically to the microtubular proteinS,a°,lL
Dimethylsulfoxide (DMSO) has a strong stabilizing effect on microtubule (MT) structure in vitro. That is, DMSO prevents the depolymerization of MT under certain conditions of cooling, pH and ionic strength 11. The question raised was: what in- fluence does DMSO have on fast axoplasmic transport (FAXT) in light of its stabilizing effect on MT? In this study the effects of DMSO on FAXT and on the ultrastructure of the vagus nerve were examined. It was found that DMSO has an inhibitory effect on FAXT, and causes complex changes in the size of the unmyelinated axonal com- ponents of the vagus.
METHOD
Adult cats were anesthetized with sodium pentobarbital (36 mg/kg, Napental) administered by intraperitoneal injection. The nodosum ganglion was exposed and local injections were made into the ganglion through a glass micropipette (5/~m). The injection consisted of 20-25 #Ci of [aH]leucine in 2 #l of modified saline solution, stained with bromophenol blue. The vagus nerve was ligatedjust distal to the ganglion at different times after the [aH]leucine was injected, with the purpose of preventing the continuous transport out of the ganglion of radioactive protein. Experiments without a ligature on the vagus also were done to show that this ligature did not affect the character of the transport itself. Three to six hours after the injection, the peripheral vagus nerve was exposed, dissected out and frozen on a piece of dry ice. It was sliced immediately into 2 mm sections, solubilized in NCS (Nuclear Chicago Co.) and the tritium activity was determined in a scintillation counter. The transported materials were analyzed in a few experiments by polyacrylamide disc gel electrophoresis. An average of 98 To of leucine-labeled molecules was recovered from the gels, indicating that the bulk of transported label is incorporated into macromolecules rather than present as free precursors.
Transport in vitro was also studied, and in these experiments a ligature was tied close to the ganglion 2 h after the [aH]leucine injection. The nerve distal to the ganglion was dissected out and incubated for 1-4 h in a modified Locke's solution under 95 ~o O~, 5 ~ CO2 atmosphere, pH 7.4 at 37 °C. The ganglion also was prepared for scin- tillation counting.
In the experimental group, the nerve was incubated in the presence of DMSO (Sigma Co., grade 1), 2 ~ , 5 ~ and l0 ~ (v/v) for 2-3 h. To study the reversibility, the nerve was removed after 2 h exposure to l0 ~ DMSO, and then washed and incubated for another 2 h in normal Locke's solution in similar conditions to the previous group.
Samples of the control vagus nerve and nerve treated with DMSO at different concentrations were studied under the electron microscope. The samples for electron microscopy were fixed for 2-3 h in 3 ~ glutaraldehyde in 0.1 M sodium phosphate buffer at pH 7.3 at room temperature. The tissue pieces (1 mm) were washed in buffer solution for 1-2 h at room temperature and post-fixed for 1 h in 2.5 ~ osmium tetroxide in 0.1 M phosphate buffer. Tissue pieces were rinsed in distilled water briefly and left
289
for 1 h in 2 ~ uranylacetate, dehydrated in ethanol and embedded in Epon 812. Silver and gray sections were stained with uranyl acetate and lead citrate and examined in a Philips EM 300 electron microscope at 60 kV. The axonal areas were measured by planimetry and the microtubules within the same axons counted in sections from different regions of the cervical vagus nerves.
RESULTS
(I) Axoplasmic transport in the vagus nerve
( A ) Rate and distribution of labeled materials transported The distribution of radioactive protein in 2 mm segments of vagus nerve was
followed in vivo from 1 to 5 h after the injection of [3H]leucine. The amount of radio- activity (disint./min/2 mm segment) was plotted as a function of distance from the site of the precursor injection. High disint./min (dpm) values are observed in the ganglionic area and, depending on the period of time allowed for axoplasmic transport (AT), a 'plateau' and a wavefront of labeled material was observed along the nerve. The
distribution pattern of radioactive materials was observed in 6 experiments of which a representative set is illustrated in Fig. 1.
When the vagus nerve is ligated at a point where the fiber leaves the ganglion 2 h after [3H]leucine injection, the distribution of radioactivity now shows a 'saddle area', a peak, and a wavefront. With this technique a better analysis of drug effects on the axoplasmic transport can be made as seen in Fig. 2. The same pattern of distri- bution of labeled molecules was observed whether the nerve was left in the animal or incubated in vitro for 2-4 h.
The velocity of the transport was approximated from the distance the wavefront moved per unit time. To obtain the distance travelled, a line was extended from the slope of the wavefront and another from the level of background radioactivity. The starting point (injection site) was identified as the 2 mm piece of the ganglion with the highest radioactivity. The distance between where these lines intersected and the start- ing point was taken as the transport distance for the particular incubation time.
As can be noted, the rate of transport in the in vivo and in vitro preparations is similar. In fact, the regression coefficients calculated from in vitro and in vivo experiments indicate that the transport velocities are not significantly different. The
ld-
10-~
: e ~e . . . e . I . = ~ ' ~ - e e • •
DPM ~ 1
6 ' 2b ' 4b ' 6b ' 8'0 " , .
Fig. 2. Distribution of radioactivity in normal cervical vagus nerve in vivo after ligature of the nerve close to the ganglion. If the nerve is tied 1 h after pH]leucine injection, no pulse of radioactive material is detected (0) . The highest dpm values correspond to the ganglion. The wave of radioactive materials can be characterized by a 'saddle area', crest (peak), and the wavefront. The 'saddle area' is between the ganglionic peak and the wave of materials transported, the size of this area depends on the period of time allowed for axoplasmic transport. The wave front for a period of 4 h (ml) is 57 mm away from the ganglion, and 77 mm for 5 h (O).
291
! 70-
60-
50-
4O-
3O-
2O-
10- m m
• in vivo • in vitro
hrs.
Fig. 3. The distance from the [SH]leucine injection site to the wavefront of labeled materials for differ- ent experimental times. Combined data from in vivo and in vitro experiments were analyzed and the regression line for these results is shown. Regression coefficients were computed for in vivo (14.7 mm/h) and in vitro (16.4 mm/h) data separately and were not significantly different. The regression line does not pass through the origin, indicating a delay of about 20 min between injection and the starting of axoplasmic transport. The slope of the regression line determined the velocity of transport to be 15.25 mm/h (366 ram/day).
slope of the regression line represents a velocity of AT of 366 ± 20 (S.E.) mm/day (Fig. 3) and is close to the velocity reported by others in comparable avian and mam- malian nerve preparations3,S,n, TM. Also, there is a lag (20 min) in the detection of AT. This lag is probably due to the time required for the uptake and incorporation of the leucine, and possibly a linkage of synthesized molecules to the transport system.
(B) DMSO effects on fast axoplasmic transport When the incubation medium contains DMSO at concentrations of 5 ~ or more,
the axoplasmic transport is markedly perturbed. Incubation of 2-2.5 h in 10 ~ DMSO resulted in a wavefront at a much shorter distance from the ganglion than in the con- trol. Furthermore, the wavefront was irregular and the 'saddle' region behind the front was higher than in the control (Fig. 4B). This action of DMSO appears to be reversible in many axons when the nerve is removed from the DMSO medium and incubated in the control medium, for if the time of the DMSO exposure is subtracted from the total incubation time, the velocity of the wavefront is normal. However, the recovery is not complete for all axons as indicated by the presence of extra peaks behind the wavefront (Fig. 5).
Fig. 4. The distribution of radioactive materials along the vagus nerve in different experimental con- ditions is illustrated. A: the pattern of radioactivity along the nerve after 2 h of incubation in normal Locke's solution. Typical example from 15 experiments. B: the effect of 10~ DMSO, no 'saddle area' is observed. The wavefront farthest from the ganglion is 43 mm from the injection site. The distance expected for a period of 4.3 h is 62 mm. Typical example of 7 experiments. C: exposure of the nerve for 2 h at 5 ~ DMSO shows two extra peaks in the 'saddle area', never present in the control. The wavefront is 58 mm away from the ganglion and the distance expected for 4 h allowed for transport is 57 mm. Typical example from 6 experiments. D: 2 ~ DMSO for 2 h shows a pattern of radioactivity transported similar to the controls. Typical example from 6 experiments.
293
1¢
10 -~
10 -4 c
10 3-
10 =-
D ~ ~...~_. I
6 ' "~o ' , /o ' d o ' 8'o '
Fig. 5. The vagus nerve was left in the cat for 2 h after [3H]leucine injection, removed and incubated for 2 h in 10% DMSO and then the DMSO washed out by further incubation in normal Locke's medium for 2 h. The pattern of the radioactive protein distribution indicates that transport had resumed in some axons. (Front A is located where it would be expected if the 2 h of DMSO exposure is subtracted from the total incubation time.) The accumulations (B and C) indicate that the recovery is not complete.
The effect of 5 ~ DMSO had a somewhat different pattern than 10~o. That is, the wavefront velocity was the same as in the control but extra peaks (not seen in the controls) appear behind the wavefront indicating that the transport was blocked or slowed in some of the axons (Fig. 4C).
Treatment with 2 ~o DMSO for 2-2.5 h had no discernible effect on the velocity or the wave shape (Fig. 4D).
( C) Ultrastruetural studies in the vagus nerve
The fast axoplasmic transport of radioactive materials in the cervical vagus nerve, when the precursor is injected into the nodosum ganglion, is essentially carried out by unmyelinated axons. A high percentage of unmyelinated fibers in the cervical vagus have their cell bodies in the nodosum ganglionl,12,1L For this reason, the elec- tron microscopic studies were oriented to the unmyelinated axons.
(1) Control vagus' nerve. A cross-section view of the control vagus nerve reveals the presence of microtubules (MT), neurofilaments (NF), mitochondria, and smooth endoplasmic reticulum in the axoplasm of unmyelinated fibers. The number of mito- chondria per axon range from 1 to 2, and the smooth endoplasmic reticulum is not very prominent, at least with this method (Plate 1).
294
Plate 1. Transverse section of the control vagus nerve showing: (Ax) unmyelinated axons, (MT) microtubules, (M) mitochondria, (NF) neurofilaments, (G) cytoplasm of glial cells and nucleus of Schwann cells. The nerve was incubated for 3 h in a modified Locke’s solution with a gaseous phase of 95 % 02-5 % CO2 at 37 “C.
Plate 2. Transverse section through the vagus nerve treated with 10 ~ DMSO for 2 h. It shows changes in the axonal area (Ax), microtubule (MT) distribution, mitochondria (M) 'swelling' and swelling of the glial cell (G). Plate 3. Transverse section of the vagus nerve treated with 10 ~ DMSO for 2 h, washed, and incubated for an additional 2 h period. A complete recovery of the nerve structure is observed.
Plate 4. Transverse section through the vagus nerve treateo w~u~ ~/o ,-- . . . . . . . . . . . . . . . in the axonal area (Ax) and swelling of the glial cells (G). Plate 5. Transverse section through the vagus nerve treated with 2~o DMSO for 2.5 h. Note a slight 'swelling" only in some glial cells (G). The unmyelinated axons look normal.
297
(2) Vagus nerve treated with D M S O in vitro. The vagus nerve was incubated in 2 ~ , 5 ~ or 10 ~ DMSO for 2.5 h. With 10 ~ DMSO, there was an extensive 'swelling' of the glial cells, but the individual axons behaved very differently. Some axons were swollen and others were dramatically shrunken. In the latter, numerous well-formed microtubules occupied almost all of the intraaxonal space. In the axons with an in- creased diameter, the individual MTs appeared normal, but their distribution within the axon was somewhat more dispersed than the normal. Some mitochondria are swollen and others look normal. It was clear in all cases that the spatial relationships between the axonal organelles were altered (Plate 2).
In 5 ~ DMSO, the ultrastructure was changed in a similar manner to I 0 ~ DMSO, but the effect was less pronounced and fewer axons were affected (Plate 4). In 2 ~ DMSO there was no detectable change in the ultrastructure of unmyelinated axons but there was some 'swelling' of the glial cells (Plate 5). The myelinated axons appeared to be practically unchanged ultrastructurally by the DMSO even at 10 ~ . However, the accompanying glial cells were swollen and there was some disruption of the myelin.
Nerves exposed to 10~ DMSO for 2.5 h and then reincubated in Locke's solution for the same period of time showed a complete recovery in their structure, although the FAXT is not completely recovered (Plate 3).
In the nerves treated with 5 ~ and 10~ DMSO, the number of MTs per axon on first glance appeared to increase in some axons and decrease in others, so the question was: do the number of MTs per axon remain constant and the apparent changes in number result only from the alterations in axonal diameter? To answer
Fig. 6. Frequency distribution of the axons counted and grouped according to their diameter. There is a redistribution of the axons treated with DMSO resulting in an increased number of large and small axons.
298
260-
240-
220-
200-
~3 180- t~
16EF
o
120-
o . 100-
d 80- z
60"
40"
2 0 -
MT~4~2
t - • C o n t r o l - -
:. ~r D M S O 1 0 % . . . . .
• R e v e r s i b i l i t y - - -
.~ ".*-. .,
~" *' k . . . * " * .~.
.i ~, 1.'O 1.'5 2'.0 ~ 2 Axonal Area
Fig. 7. Number of microtubules per sq.Fm plotted against axonal area. In the control the number of microtubules per unit area does not change with increasing axonal size. Nerves treated with 10% DMSO show an increase in the number of MT/sq./~m in axons with small areas and a decrease in the number of MT/sq./~m in axons with large areas. This effect was reversible by washing the nerve after treatment with DMSO.
this question, the cross-sectional area of the axons and the number of MTs were determined for control and DMSO-treated nerves.
A frequency distribution of axons according to their calculated diameter re-
vealed a redistribution of axon sizes when the nerve was treated with 10 % DMSO.
The number of very large and very small axons increases (Fig. 6). The number of MTs per unit area is independent of the axonal cross-sectional area in controls.
Nerves treated with 10% DMSO show an increase in the number of MTs per sq./~m in axons with a small cross-sectional area and a decrease in the number of MTs per
sq. #m in axons with a large area. However, the mean number of microtubules per
axon in normal and DMSO-treated nerves is about the same. Therefore, DMSO is primarily affecting the area of the axons and not the number of microtubules. The changes in the axonal area due to DMSO treatment was reversible by washing the nerve (Fig. 7).
DISCUSSION
The experimental work presented here describes the 'fast' axoplasmic transport of protein in vagus nerve axons emanating from the nodosum ganglion, and the effect of DMSO on this transport, along with the related ultrastructural changes.
299
The injection of labeled amino acid into the nodosum ganglion allows the uni- polar neurons resident there to incorporate [all]amino acid into proteins and these proteins are transported in both axonal branches; one goes to the central nervous system (solitary nucleus) and the other to the periphery (lungs, aortic bodies, etc.). Differences in the total amount of protein transported in each direction have been found, but the velocity of the fast transport is the same in both branches z4,2s. Although the vagus nerve is a mixed nerve containing both afferents and efferents, only those axons whose cell bodies are located in the nodosum would be transporting radioactive protein in the experiments reported here. The peripheral branch ('axon') of these unipolar neurons, which is carried by the vagus, is sensory in function.
It was found that a better resolution of the effect of drugs on the transport of radioactive materials along the vagal nerve fibers was achieved by tying the nerve just distal to the ganglion approximately 2 h after the injection of the radioactive leucine. The tie stops further flow of labeled proteins from the cell bodies into the axons, and if only a small percentage of the axons is affected by a drug, it can be detected by the presence of extra-peaks in the 'saddle area' behind the front. Otherwise the alterations in transport are obscured by a continual flow of labeled material from the ganglion.
DMSO is known or claimed to have a wide spectrum of biological actions 16, e.g., cryoprotection 21, radioprotection 2,2°, analgesia 4,25, percutaneous absorption 14,
etc. Of particular interest here is the dramatic action it has on microtubules which prevents them from depolymerizing under conditions which ordinarily cause depoly- merization, such as cooling and ionic concentration increase 11. Since the integrity of microtubules appears to be critical for fast axoplasmic transport, the effect of DMSO on FAXT and the axonal ultrastructure are of interest. Further, the unusual solvent properties of DMSO make it a useful solvent in many experimental studies so the actions of DMSO itself are relevant.
There appears to be a differential susceptibility of axons to DMSO in both the actions on FAXT and the ultrastructure. That is, there may be classes of axons in their sensitivity to DMSO. It is well documented that DMSO crosses cell membranes rapidly and can be easily washed out of cells or tissues by simple dilution 22 so the distribution of DMSO would be expected to be uniform. In the FAXT, the front ap- pears normal in 5 ~ DMSO but the region behind the front, the 'saddle' area, shows peaks of radioactivity that represent a blockade or extreme slowing of transport in a number of axons. When the concentration of the DMSO is increased to 10 ~ there is a complete blockage of transport, indicating that all the axons were affected; but the scattering observed in the front (Fig. 4B) suggests that the blockage is not simultaneous in all the fibers.
The DMSO action appears to be substantially reversible even with a 10~ DMSO treatment, since transport resumed in a high proportion of axons when the DMSO was washed out. If the period of exposure to the DMSO is subtracted from the total incubation time, then the wavefront is localized at the same place as the control preparation for that incubation time. The duration of time for the DMSO to produce its effect and for the reversal were not directly determined in these studies but can be inferred from the results to be l h for all the axons to be affected and about 1 h to recover after washing.
300
From an ultrastructural standpoint, DMSO primarily affects the cross-sectional area of the unmyelinated fibers and not the number of microtubules. The extensive changes in axonal area induced by the DMSO indicate that major reductions or ex- pansions of the axonal membrane have rapidly occurred. The mechanisms underlying these changes are not understood. However, it is clear that the blockade of the FAXT is not a causal factor because we have never seen such area changes when FAXT was completely inhibited in the same type of vagus nerve preparation by other agents such as vincristine sulfate (manuscript in preparation). These changes in axonal area might disrupt the spatial relationships among the axonal organelles, destroying the ultra- structural organization necessary for the mechanism of axonal transport.
Although DMSO actions on microtubules and FAXT are compatible with the idea that MTs are involved in FAXT, these experiments do not rule out the possible parti- cipation of other axonal organelles such as neurofilaments, smooth endoplasmic reticulum, microfilaments and the plasma membrane. The transport 'machinery' very likely is a coordinated system with many components. Indeed, bridges between MT 7, MT and organelles have been reported 27. If one of these elements is disrupted, the transport machinery would be expected to malfunction.
The ultrastructure of the microtubules themselves (in all concentrations of DMSO) is indistinguishable from the control. The normal appearance of the micro- tubules does not mean, of course, that they are functional. A maximal stabilizing effect of DMSO on microtubules in vitro occurs at a concentration of 10 ~ (ref. 11), a concentration which completely blocks FAXT. These results are compatible with the concept that the microtubules must be in a 'dynamic' condition to be biologically functional. Thus, by strengthening the polymerization forces, DMSO may make the microtubules structurally stronger but non-functional for FAXT.
ACKNOWLEDGEMENTS
We wish to acknowledge the technical assistance of Lorraine Hammer and discussions with Louis Green and Irene Bettinger.
This research was supported in part by U.S.P.H.S. Grant HD-02528 and a grant from the Kansas Division, American Cancer Society, to Dr. Irene Bettinger. J. Alejandro Donoso held a fellowship from the Bank of International Development through the Catholic University of Chile (1974).
REFERENCES
1 Agostini, E., Chinnock, J. E., DeBurgh Daly, M. and Murray, J. G., Functional and histological studies of the vagus nerve and its branches to the heart, lungs and abdominal viscera in the cat, J. Physiol. (Lond.), 135 (1957) 182-205.
2 Ashwood-Smith, M. J., Radioprotective and cryoprotective properties of dimethyl sulfoxide in cellular systems, Ann. N. Y..4cad. Sci., 141 (1967) 45~2.
3 Barondes, S. H. and Samson, F. E., Axoplasmic transport, Neurosci. Res. Prog. Bull., 5 (1967) 307-419.
4 Becker, D. P., Young, H. F., Nulsen, F. E. and Jane, J. A., Physiological effects of dimethyl sulfoxide on peripheral nerves: possible role in pain relief~ Exp. Neurol.~ 24 (1969) 272.-276.
301
5 Bray, J. J. and Austin, L., Flow of protein and ribonucleic acid in peripheral nerve, J. Neurochem., 15 (1968) 731-740.
6 Bray, J. J. and Austin, L., Axoplasmic transport of 14C protein at two rates in chicken sciatic nerve, Brain Research, 12 (1969) 230-233.
7 Burton, P. R. and Fernandez, H. L., Delineation by lanthanum staining of filamentous elements associated with the surfaces of axonal microtubules, J. Cell Sci., 12 (1973) 567-583.
8 Cu6nod, M., Sandri, C. and Akert, K., Enlarged synaptic vesicles in optic nerve terminals induced by intraocular injection of colchicine, Brain Research, 39 (1972) 285-296.
9 Dahlstr6m, A. and H~ggendal, J., Studies on the transport and life-span of amine storage granules in a peripheral adrenergic neuron system, Acta physiok scand., 67 (1966) 278-288.
10 Davison, P. F., Microtubules and neurofilaments: possible implications in axoplasmic transport, Advanc. Biochem. Psychopharmacol., 2 (1970) 289-302.
11 Dulak, L. and Crist, R. D., Microtubule properties in dimethylsulfoxide (DMSO), J. Cell Biol., 63 (1974) 900.
12 DuBois, F. S. and Foley, J. O., Quantitative studies of the vagus nerve in the cat. II. The ratio of jugular to nodose fibres, J. comp. Neurol., 67 (1937) 69-87.
13 Edstr6m, A. and Mattson, H., Fast axonal transport in the sciatic system of the frog, J. Neuro- chem., 19 (1972) 205-221.
14 Eltbaum, S. G. and Laden, K., The effect of dimethylsulfoxide on percutaneous absorption: a mechanistic study, part III. J. soc. cosmet. Chem., 19 (1968) 841-847.
15 Foley, J. O. and DuBois, F. S., Quantitative studies of the vagus nerve in the cat. I. The ratio of sensory to motor fibers, J. comp. Neurol., 67 (1937) 49-67.
16 Jacob, S. W. and Herschler, R., Biological actions of dimethyl sulfoxide, Ann. N.Y. Acad. Sci., 243 0975) 1-508.
17 Kreutzberg, G. W., Neural dynamics and axonal flow. IV. Blockage of intraaxonal enzyme trans- port by colchicine, Proc. nat. Acad. Sci. (Wash.), 62 (1969) 722-728.
18 Lasek, R. J., Protein transport in neurons, Int. Rev. Neurobiol., 13 (1970) 289-324. 19 Lasek, R. J., Axonal transport of proteins in dorsal root ganglion cells of the growing cat: a
comparison of growing and mature neurons, Brain Research, 20 (1970) 121-126. 20 Lappenbusch, W. L. and Willis, D. L., The effect of dimethyl sulphoxide on the radiation response
of the rough-skinned newt (Taricha tranulosa), Int. Radiat. Biok, 18 (1970) 217-233. 21 Malinin, T. I., Perry, V. P. and Doland, M. F., Dimethylsulfoxide toxicity in cell cultures,
Cryobiology, 2 (1965) 19. 22 Malinin, T. I. and Perry, V. P., Toxicity of dimethylsulfoxide on HeLa cells, Cryobiology, 4
(1967) 90-96. 23 McEwen, B. S. and Grafstein, B., Fast and slow components in axonal transport of protein, J.
Cell Biok, 38 (1968) 494-508. 24 Ochs, S., The dependence of fast transport in mammalian nerve fibers on metabolism, Acta neuro-
path. (Berk), Suppl., 5 (1971) 86-96. 25 Penrod, D. S., Bacharach, B., and Templeton, J. Y., Dimethyl sulfoxide for incisional pain after
thoracotomy: preliminary report, Ann. N. Y. Acad. Sci., 141 (1967) 493-495. 26 Slater, C. R., Time course of failure of neuromuscular transmission after motor nerve section,
Nature (Lond.), 209 (1966) 305-306. 27 Smith, B. H., Neuroplasmic transport in the nervous system of the cockroach Periplaneta ameri-
cana, J. Neurobiol., 2 (1971) 107-108. 28 Watson, D. F., Donoso, J. A., Illanes, J. P. and Samson, F. E., Comparison of transported
proteins in the central and peripheral processes of unipolar neurons, Trans. Amer. Soc. Neurochem., 6 (1975) 109.
29 Weiss, P., Neuronal dynamics, Neurosci. Res. Prog. Bull., 5 (1967) 371-401. 30 Weiss, P., Neuronal dynamics and neuroplasmatic axonal flow, In S. H. Barondes (Ed.) Cellular
Dynamics o f the Neuron, Vok 8, Academic Press, New York, 1969, pp. 3-34.
