J. Phy8iol. (1972), 227, pp.

627-645 627
With 8 text-figure8
Printed in Great Britain

RATE OF FAST AXOPLASMIC TRANSPORT IN

MAMMALIAN NERVE FIBRES
By SIDNEY OCHS
From the Department of Physiology, Indiana University
Medical School, Indianapolis, Indiana, U.S.A.
(Received 3 January 1972)
SUMMARY
1. Fast axoplasmic transport in mammalian nerve fibres was deter-
mined by the presence of a crest of activity in the sciatic nerve after in-
jection of [3H]leucine into the L7 dorsal root ganglion or the L7 moto-
neurone region in the ventral horn region of the spinal cord. After incor-
poration into proteins by the cell bodies, a rate of transport close to 410
mm/day was found for cat sensory nerves. A closely similar rate was found
in the motor and sensory sciatic nerve fibres of the monkey, dog, rabbit,
goat and rat. In the longer nerves where longer downflow times were
possible, there was no decrement of rate with distance, or presence of later
appearing crests of activity indicative of multiple fast transport systems.
2. The rate of fast transport found in the long L7 dorsal roots of the
rhesus monkey was the same as that in the corresponding length of sciatic
nerve and the same fast rate was shown by the crest of activity ascending
in the dorsal columns of the spinal cord.
3. Labelled activity was found present inside myelinated nerve fibres
ranging in diameter from 3 to 23 ,tm in nerve segments taken at the for-
ward part of the crest suggesting that the rate of fast axoplasmic transport
is independent of fibre diameter.

]INTRODUCTION
The reality of both fast and slow components of axoplasmic transport
in nerve fibres is generally accepted (Lubinska, 1964; Barondes, 1967;
Grafstein, 1969; Ochs, 1969), but uncertainty exists about the rate of the
fast component. Rates given for fast axoplasmic transport range from
50 to over 2000 mm/day (Schmitt, 1968; Dahlstr6m, 1971). In part, this
spread is due to the lower rates found in nerves of the invertebrate and
lower vertebrate (Grafstein, 1969; Davison, 1970; Fernandez, Hunees &
Davison, 1970) as compared to the mammal. Variations are also due to
differences in the techniques used to determine rate.
23-2
628 SIDNEY OCHS
One method widely used is to measure the accumulation of components
carried to ligations made in the nerve or to nerve terminals (Lubinska,
1964; Barondes, 1969; Grafstein, 1969). Karlsson & Sjdstrand (1971) found
four peaks of accumulation of labelled components in the lateral geniculate
after injecting the eye with [3H]leucine with two fast rates reported at
150 and 40 mm/day. Hendrickson & Cowan (1971), on the other hand,
found only a single fast rate at 240 mm/day in the monkey optic system.
Using double ligated nerve to assess rate of accumulation of acetylcho-
linesterase (AChE), a rate of 260 mm/day in dog sciatic nerve was deter-
mined by Lubinska & Niemierko (1971). More recently, our study of AChE
transport gave a rate of 431 mm/day for cat sciatic nerve (Ranish & Ochs,
1972).
It is possible that species differences within the mammalian or differences
between various types of nerves could account for some of these divergent
rate determinations. To assess such factors a more direct method of
analysing the rate within the nerve fibres is required, one which has been
used in this laboratory for the last several years. An amino acid precursor
([3H]leucine) is injected either into the cat lumbar seventh (L7) dorsal root
ganglion or into the ventral horn of the L7 segment of the spinal cord, and
after a period of downflow a crest of labelled activity in the sciatic nerve
shows the presence of fast axoplasmic transport (Ochs, Sabri & Johnson,
1969; Ochs & Ranish, 1969). This crest contains protein incorporated
materials which move down the nerve in a linear and regular fashion at a
rate of about 400-410 mm/day. The present studies were undertaken to
determine if the rate of fast axoplasmic transport in the sciatic nerves
varies in different mammalian species. The rate of outflow in the motor
nerve fibres after the L7 spinal cord segment injection was also compared
with sensory fibre outflow. Earlier studies had indicated that the rate was
similar in the sensory and motor fibres of the cat (Ochs & Ranish, 1970)
and the present studies were extended to nerves of other species.
A difference in the rate of transport in the dorsal root as compared to
the sciatic nerve had been suggested (Barondes, 1969) by the steeper slope
of activity seen in the dorsal root as compared to the corresponding distal
length of sciatic nerve after injecting the L7 dorsal root ganglion with
[3H]leucine (Lasek, 1968). A direct comparison of the rates in the central
and peripheral processes of the L7 dorsal root ganglion neurones was
possible when using the much longer lengths of dorsal root present in the
larger rhesus monkey. The fibres of the dorsal root enter the spinal cord to
ascend the dorsal columns of the spinal cord and a crest of transported
activity in the columns was looked for in order to assess the rate of trans-
port in the fibres of this C.N.s. tract.
The regularity of the rate of fast axoplasmic transport in the various
 FAST TRANSPORT IN NERVE 629
nerves studied suggested that it is independent of nerve fibre diameter.
To investigate this point, radioautographic studies were made of fibres taken
at the front of the crest to determine if myelinated fibres over the whole
range of fibre diameters contain activity. As will be described herein, the
results indicated that the rate of fast axoplasmic transport is independent
of diameter and the similarity of its rate in the various nerve types so far
studied suggests a common underlying mechanism.

METHODS
Pentobarbitone anaesthesia was used for surgical preparation of the animals. A
laminectomy was performed over the L7 dorsal root ganglia, and in cases where the
spinal cord was to be injected, over the L7-S1 segmental regions. The precursors
[3H]leucine or [3H]lysine obtained from New England Nuclear Co. had a specific
activity of 35 c/mmole. It was supplied as a sterile 0-85 % saline solution at a con-
centration of 5 mc/ml. and of a [3H]leucine purity of at least 90 %. The precursor
was injected into the L7 dorsal root ganglion for uptake by the sensory somata or into
the L7 ventral horn region for uptake by the motoneurone cell bodies as has been
previously described (Ochs & Burger, 1958; Ochs & Ranish, 1969). A volume of
approximately 15 #1. [3H]leucine (or in some cases of [3H]lysine) was injected using
glass micropipettes drawn down to tips approximately 30-40 ,tm in diameter. In
these experiments no difference was found in the shape or the rate of transport when
comparing the crests when [3H]leucine or [3H]lysine was used as the precursor (Ochs &
Ranish, 1970). In earlier studies [3H]leucine and [14C]leucine had also been compared
and the pattern of subcellular labelling was found to be the same (Ochs, Johnson &
Ng, 1967). The similarity of subcellular distribution was maintained over periods of
days and weeks indicating that there is little loss or transfer of the tritium label.
For the injection of the precursor into the L7 motoneurone region of the spinal
cord, small slits in the dura were made and volumes of 5-10 1A. injected into the
ventral horn region at three and five sites over the length of the L7 cord segment
(Ochs, Dalrymple & Richards, 1962). The exposed region was then covered with
cotton and warm Ringer solution poured over it to bring the temperature to 380 C
which was then maintained with a thermostatically controlled heat lamp.
To determine the pattern of outflow of labelled activity in the sciatic nerve, the
animal was sacrificed and the nerve ganglion and dorsal root were removed, cleaned
and then laid down on a strip of Bristol board. The tissue was allowed to partially
dry and it adhered to the board. This prevented longitudinal shrinkage when it was
further heat-dried to a point where the nerve could easily be cut into 3, 5, or in some
cases into 6 mm segments. Each segment was placed into a scintillation counting
vial to which 1-0 ml. Soluene (Packard) was added to solubilize the tissue (2.0 ml.
in the case of 6 mm segments). This generally took 1-2 days and in addition it was
usually necessary to heat the vials at 500 C for several hours. After solubilization,
10 ml. of a scintillation solution consisting of PPO (2,5-diphenylazol) and POPOP
(1,4-bis-2-5-phenylazol)-benzine in toluene was added to each vial and each vial
counted for either 2, 5 or 10 min in a Packard Model 3310 spectrometer. The base-
line level of activity was 5- 10 x the background count of 15 counts per minute
(cpm) and represents some degree of generalized leakage into the circulation from the
injected site with an uptake by the Schwann cells (cf. Fig. 5, Ochs & Ranish, 1969).
This base-line level is low enough in all our experiments to be readily distinguished
from the level of activity found in the crest. No correction for quenching was neces-
630 SIDNEY OCHS
sary to show the relative amplitude of the crest and counting error is well below 5 %
at the lowest levels found. In some cases deviations were seen, but these are known to
represent activity accumulated at the ends of some uncut branches of nerve which
were missed in cleaning the nerve (Ochs, 1971 a). Occasionally a chemiluminescence
was found to increase the base-line activity by as much as 100-150 cpm. This was
obviated in most cases by delaying or repeating the counting procedure after
several days.
To determine the pattern of outflow of labelled activity in the dorsal columns,
they were gently separated with a probe from the rest of the spinal cord. The columns
could then be removed and laid out flat on a strip of Bristol board. Because of the
possibility that some stretching could occur, two points on the columns were marked
with dye in situ. The same length between the points was then arrived at when the
tissue was laid out on the Bristol board. The dorsal columns were heat-dried without
any longitudinal shrinkage and then cut into sections as usual for determination of
the distribution of activity.
The bleeding due to the greater degree of vascularity in and around the muscles
and bony structures of the lower lumbar region of the rabbit made it difficult to
expose the L7 region of the cord for injection. Elevation of the hind limbs and a
somewhat higher level of pentobarbital anaesthesia were of some help. In the larger
animals a chronic preparation was employed when downflow times longer than 10
hr were used. After injection, the skin over the defect was closed with sutures and
covered with collodion or a commercial rubber-based preparation (Skin-Hesive) and
200,000 ,u. of procaine-penicillin was then injected I.M.
To study the pattern of outflow in the motor fibres of the rat, the lumbar enlarge-
ment of the spinal cord comparable to the L7 and SI segments in the cat was identi-
fied and the ventral horn region injected with [3H]leucine. Volumes of approxi-
mately 5 #le. were injected into three or four sites spaced along the length of the
L7-S1 segments on each side of the cord. The dura in the rat is thin and the pipette
could be passed down through it without slitting it beforehand. In lightly anaesthe-
tized animals some neuronal excitation was caused by the injection as indicated by
muscle movements in the hind leg. In animals too lightly anaesthetized the injection
could cause limb and trunk movements with a resulting damage to the cord and a
poor pattern of downflow. More leakage was seen in the fibres near the cord than
usual and in later experiments the dura was slit before injecting the cord in the
L7-S1 region, thus reducing the amount of leakage around the roots.
For radioautographic studies cat sciatic nerves were prepared by freeze-substitu-
tion (Ochs, 1965). Segments of the nerve were taken at the front of the estimated
crest position 5 or 6 hr after L7 dorsal root ganglion injection. The small pieces of
nerve were placed on thin strips of brass shim stock and then dipped into rapidly
stirred Freon-12 which had been cooled down towards its freezing point (- 160° C)
with liquid nitrogen. The frozen nerves were then placed into vials containing 1 %
osmium tetroxide in acetone at a temperature of approximately -50° C. Substitu-
tion was carried out for a period of 10-14 days at a temperature of approximately
-25° C. During this time the osmium tetroxide and acetone gradually replaced the
ice phase in the tissue. The substitution-fixed nerve tissue was then warmed up to
room temperature in fresh cold acetone over a period of several hours, cleared and
embedded in paraffin. Sections were cut at 6 j#m, placed on microscope slides and
coated in the dark with Ilford L4 photographic emulsion (Kopriwa & Leblond, 1962;
Ochs, 1966; Ochs & Johnson, 1969). During exposure periods ranging from 3 to 9
months, slides were taken at intervals and developed. These were coverslipped for
viewing and photography.
U
EM 10'

102

10, lI
1~ ~ ~ o6-L7right
FAST TRANSPORT IN NERVE

Root

75 60 45 30 15
e-
G
RESULTS
Fast transport in sensory and motor fibres
The pattern indicative of fast axoplasmic transport was seen in the
sciatic nerve of the monkey after injection of [3H]leucine into the L7 dorsal
root ganglia (Fig. 1). It is the same as that previously reported for the cat
(Ochs et al. 1969; Ochs & Ranish, 1969). A high level of radioactivity
remains in the ganglion descending distally to a plateau in the sciatic
nerve. Still more distally in the nerve there is a rise to a crest of activity
before a rapid fall to base-line levels. The rapidly falling front or face of
107

105

Nerve

0 15 30 45 60 75 90 105 120 135 150 165

mm
Fig. 1. Downflow pattern in monkey sciatic nerve. After injection of the
631

L7 dorsal root ganglion on each side with [3H]leucine and an incorporation

and downflow time of 6 hr, the sciatic nerves, ganglia and dorsal roots each
side was sectioned into 3 mm segments. The activity in each segment is
given in cpm for each nerve (o, *) at the indicated distances from the
centre of the ganglion (G) taken as zero. Arrow 1 indicates the intersection
of the slope of the advancing crest of activity with respect to the base-line
level of activity in the nerve. The base-line represents a leakage of radio-
activity from the site of injection into the blood supply and local incorpora-
tion. Ordinate, cpm on a logarithmic scale. Abscissa, activity in the segments
at the distance in mm shown from the centre of the injected L7 dorsal root
ganglia. Subsequent Figures are similar.

the crest represents the earliest outflow of labelled materials leaving the
cell bodies to be carried down the axons by the fast transport system. The
distance from the intersection of the forward face of the crest with the
base-line level to the high level remaining in the cell bodies in the ganglion
and the time which had elapsed after injection of the ganglia, is used to
632 SIDNEY OCHS
determine the rate of fast axoplasmic transport. This averaged 416 + 30
(S.D.) mm/day in the sensory fibres of fourteen nerves (Table 1). This is to
be compared to a group of cat sciatic nerves where a rate of 409 + 50 (S.D.)
was found (Table 1).
TABLE 1. Rates of fast axoplasmic transport
No. of
Nerve type nerves Species Rate
Motor (sciatic) 18 Rat 411 +50
5 Monkey 400+ 35
Sensory (sciatic) 14 Monkey 416+30
26 Cat 409 + 50
2 Rabbit 394 (360, 415)
2 Goat 389 (382, 396)
4 Dog 423+15
Dorsal column spinal cord 3 Monkey 397+57
4 Cat 391+ 59
Rate values given as means and S.D. Sensory nerves, after L7 dorsal root ganglia
injection with [3H]leucine, motor nerves after [3H]leucine uptake by L7 motoneurone.
To compare the rates of fast axoplasmic transport in sensory and motor
fibres of the monkey, the precursor was injected into the L7 ganglion on
one side and into the L7 segment of the cord near the motoneurone cell
bodies on the other side. Downflow in the latter gives a measure of the rate
in the motor fibres. The displacement of the crests in the sciatic nerves on
the two sides (Fig. 2) is accounted for by the anterior-posterior displace-
ment of the sensory and motor cell body groups (Ochs & Ranish, 1969).
The longer length of dorsal root in the monkey gives rise to a greater
anterior-posterior displacement of the sensory and motor L7 cell body
groups as compared to the cat, one close to the 85 mm distance between
the crests seen in the example of Fig. 2. The small decreasing slope of
activity in the ventral root near the cord on the L7 motoneurone injected
side is likely to be due to leakage of precursor from the cord, with a possible
contribution from slow axoplasmic flow (Ochs et al. 1967; Kidwai & Ochs,
1969).
In the larger Rhesus monkeys the dorsal roots are even longer, and here
also the displacement of the crests of activity in the sciatic nerves on the
L7 ganglion and the L7 motoneurone injected sides was found to corre-
spond to the distance between the two cell body groups (Fig. 3). The crest
displacement of 110 mm in this example appeared to be a little greater
than the anatomical displacement of 100 mm, but this small discrepancy
may be due to an underestimation of the centre of the L7 motoneurone cell
body population in the cord by several mm.
 FAST TRANSPORT IN NERVE 633
107- ' ' ' '
I

10' L7 dorsal root

o \ * ~L7 motoneuron
8 hr monkey
105 -
EC-
U 104

10~~~~~~~~~~ I~~~~~~
102 Root -i- Nerve * *
251
, I, , II ,, , ,, , , ,, , , ,
80 40. 20 0 20 40 60 80 100 120 140 160 180
60
70 30 10 10 30 50 70 90 110 130 150 170
50
mm
Fig. 2. Comparison of downflow in sensory fibres and in motor fibres. The
L7 dorsal root ganglion of the monkey was injected on the right side (o)
and the L7 motoneurones in the ventral horn of the spinal cord on the left
side (0) injected with [3H]leucine. Lumbar 6th and spinal roots were first
ligated on the left side. Eight hours later the animal was sacrificed. The
pattern of outflow on dorsal root ganglion injected side is similar to that of
Fig. 1 with a more distally placed crest resulting from the longer downflow
time (arrow 2). The outflow pattern on the motoneurone injected side showed
a slope of activity dropping off in the ventral root, rising to a crest at a
more anterior position in its sciatic nerve (arrow 1). Distance between
arrow 1, 2 was 85 mm, close to the antero-distal displacement of the gan-
glion (G) at zero and the motoneurones in the cord where the end of the
ventral root is shown at the left of the curve.
107.
106 L7 ganglion
{\ * L7 motoneuron
10.5 - x15 hr monkey
10~~~~~~~~~~~~~~~~~~~
E14
U

Root - Nerve
II i II II I 1'
Ii1 I
100 80 60 40 20 0 20 40 60 80 100120140160180200 220240 260 280
mm
Fig. 3. Comparison of sensory and motor nerve downflow at a longer time.
The downflow time was 15 hr. All else is the same as in Fig. 2. Arrows 1 and
2 indicating the forward positions of the crests on the two sides are approxi-
mately 100 mm apart. The estimated dorsal root ganglion-motoneurone
distance in this larger monkey was approximately 110 mm.
634 SIDNEY OCHS
When longer downflow times were allowed, as is possible in the larger
monkeys with their longer lengths of sciatic nerve, the crests had the
position expected of a linear rate of fast axoplasmic transport (Fig. 3). The
front of the crest seen in this example has somewhat more of a slope but
this is not related to the longer downflow distance (cf. Fig. 7). Sloping
fronts of the crests are occasionally seen when shorter downflow times
have elapsed and the front of the crest after longer downflow times was
most often seen to have the usual form (Figs. 1, 5, 7). The factors which
determine the shape of the face of the crest are at present under study and
are likely to be related to conditions obtaining during the earliest outflow
of labelled components from the cell bodies into the axons. This variation
of the shape of the crest is to be distinguished from the decrement of fast
axoplasmic transport and failure which occurs after blocking glycolysis
(Ochs & Smith, 1971 a).
107

106 - pt 4 hr monkey
106S

10'
U

103

- Dorsal root---~--. Nerve

101 0 ' 260
110 90 70 50 30 10 10 30 50 70 90 110
100 80 60 40 20 0 20 40 60 80 100
mm
Fig. 4. Transport in the dorsal root compared to sciatic nerve. L7 dorsal
root ganglion injected with [3H]leucine and 4 hr later the monkey was
sacrificed and the distribution of activity in dorsal root and comparable
length of sciatic nerve compared. Arrow 1 shows furthest extent of a crest
of activity in the dorsal roots, arrow 2, crest descent in the sciatic nerve.

Fast transport in the dorsal root

The longer length of the dorsal root present in the larger Rhesus monkeys
permitted the whole of the crest of activity to be displayed within the
dorsal root (Fig. 4) and thus to determine the rate of fast axoplasmic
transport. In this example, 4 hr of downflow had been allowed after in-
 FAST TRANSPORT IN NERVE 635
jection of the L7 dorsal root ganglion with [H]leucine. The crest of
activity in the dorsal root had moved to approximately the same distance
as the crest in the sciatic nerve. The two crest positions in this particular
example are slightly different. An exact comparison of crest positions in
the dorsal root and sciatic nerve is difficult. The larger monkeys tend to
have somewhat larger sized ganglia and a spread of the precursor over a
longer distance sometimes makes the zero point somewhat arbitrary,
especially when greater volumes of precursor are injected. An example of
such a spread of outflow around the ganglion can be noted in the example
of Fig. 1 where a much broader spread of activity around one of the ganglia
can be seen. However, when comparing the crests in the dorsal root and in
the corresponding part of the sciatic nerve in a series of animals, the
differences in the derived rates are seen to be small (Table 2).
TABLE 2. Comparison of rates of fast transport in monkey
dorsal root and sciatic nerve
Corresponding
Dorsal nerve
Expt. no. Time (hr) root segment
12/18 (4) 3 424 424
12/18 (4) 3 432
12/18 (3) 3 416 416
12/18 (3) 3 384 400
667 3-5 337 398
667 3.5 398 398
263 3 440 448
263 3 448 448
260 4 468 390
260 4 498 450
282 3 416 424
282 3 424 424
254 4-25 496 496
254 4*25 496 423
267 3.5 398 391
267 3.5 398 398
12/17 3 400 400
12/17 3 416
Means arnd S.D. x = 428+44 420 + 28
Values for dorsal root and the corresponding part of the sciatic nerve distal to the
L7 dorsal root ganglion given as means and S.D. t test shows no significant difference
between means with P > 0 1.

One consistent finding was that the level of activity in the crest in the
sciatic nerve was greater than that present in the dorsal root, by as much
as 3-5 x (Fig. 4). This difference in the amount of material moved had
little obvious effect on the rate of transport.
636 SIDNEY OCHS

Fast transport in the dorsal columns of the spinal cord

Only a part of the dorsal root fibres entering the cord ascends the dorsal
columns of the spinal cord. In order to determine if a crest of activity could
be found in the dorsal columns with the same fast rate of transport, the
amount of labelled material ascending the dorsal columns was augmented
by injecting both the L7 dorsal root ganglia with [3H]leucine. After a

106

105

104 E
03.
U
103

I102
E
U 10'

0 24 48
12 12 36 60 84 108 132 156
mm
Fig. 5. Transport into the dorsal columns of the spinal cord. The L7 ganglia
on the two sides were injected with [3H]leucine and 8 hr later the monkey
sacrificed. Distribution in the sciatic nerves (0, 0) shown in curves B with
arrows 1 and 2 at the forward face of the crests. The ganglion on the left
(@) was replotted and also its dorsal root, the dorsal columns taken from
the spinal cord in the upper curve (A). The distribution was displayed so
that it could be compared with the activity in the sciatic nerves. The
ascending crest of activity in the dorsal columns is shown (arrow 1) to be at
approximately the same position as downflow in the sciatic nerve (arroms
1, 2). Arrow 3 indicates the addition of some local collaterals at the site of
entry of the dorsal root into the cord.

period of outflow, the usual crest of activity was found present in the dorsal
columns (Fig. 5). The distribution of activity in the dorsal columns was
arranged in the figures so that its crest of activity could be directly com-
pared with the crests present in the sciatic nerves. The similarity of the
 FAST TRANSPORT IN NERVE 637
distances to which the crests had moved from the ganglia into the dorsal
columns and into the peripheral nerves shows that their rate of fast axo-
plasmic transport is similar (Table 1). The increased activity seen at the
entry of the dorsal roots is due to an admixture of local collaterals con-
taining labelled activity when taking the dorsal columns from this part
of the cord.
By varying the time when the dorsal columns are taken after L7
ganglion injection, the rate of fast axoplasmic transport was found to be
linear in the columns as it is in peripheral nerves. An example is shown for
a large cat where a longer outflow time of 14-5 hr was allowed (Fig. 6). By
10'- I I

106 . Dorsal column

105- 0f\1l Entry ¢~~~
Sciatic nerve
E104
C-_
103

102 G

10' 396
20 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280
mm
Fig. 6. Transport in dorsal columns over a longer period of time. Injection
of [3H]leucine into L7 dorsal ganglia of a large cat and dorsal column and
nerve patterns of outflow display of activity similar to that of Fig. 5. Both
peroneal and tibial nerves were taken down to the ankle allowing the full
pattern of activity and the crests to be seen over this longer distance at
14-5 hr. The entry zone for the dorsal column curve indicates some adventi-
tious addition of activity (see text).

taking the tibial and peroneal branches of the sciatic nerve almost down
to the ankle it was possible to see the full extent of the outflow pattern
in the nerves. Again, a similar rate of fast axoplasmic transport in nerve
and dorsal columns was found as can be seen from their outflow patterns
and the similarity of their crest positions.
In several animals the outflow from the ganglion was terminated several
hours after injecting it by heat coagulating the ganglion or placing a liga-
tion just below it (Ochs & Ranish, 1969). In such cases where this was done
3 hr after injection, and further downflow of 10 hr or so allowed, the level
of activity was decreased in the plateau region without much change in the
shape of the crest pattern or the position to which it had moved down the
638 SIDNEY OCHS
nerve. This shows that the material carried into the plateau region repre-
sents a later contribution from the neuronal cell bodies.
Fast axoplasmic transport had earlier been shown to take place inside
nerve fibres, by the procedures of freezing combined with radioautography
(Ochs, 1966; Ochs & Johnson, 1969; Ochs & Ranish, 1969). Freezing the
nerve at dry ice temperature for a fraction of a min causes the fibres to
become 'blinded-off' and this in turn produces a damming of the labelled
activity flowing down to that region. The cord was frozen for several
minutes with an aluminium rod cooled down to a temperature below - 300 C
to produce a similar block. The L7 dorsal root ganglia were then injected
with [3H]leucine and a period of 18- 24 hr allowed for transport. An in-
crease of activity was found in the dorsal columns just below the frozen
region with a fall of activity to base-line levels just above it. The pattern
was similar to the damming of fast transport found in ventral roots and in
peripheral nerves after a similar freezing procedure which was related to
an intra-axonal location of transported activity (Ochs & Johnson, 1969;
Ochs & Ranish, 1969).
Fast transport in various sciatic nerves
Fast axoplasmic transport in the sciatic nerves of the rabbit at 394 mm/
day was estimated from the position of the crest of activity after injection
of the dorsal root L7 ganglia with [3H]leucine (Table 1). The L7 ganglia in
the dog injected with [3H]leucine showed a fast rate of transport of 423
mm/day estimated from the front of the crest (Table 1).
The goat was of interest because of the long length of its sciatic nerve.
The sciatic nerve in this animal is thicker than that of the cat or dog,
mainly because its greater content of fibrous tissue. Although the actual
content of nerve fibres was not determined, its relative content may very
well be less than that of the sciatic of the cat or monkey in that the ungu-
late lacks digits and requires a less extensive nerve supply for the muscles
of the lower leg. The L7 dorsal root ganglia were rather hard to locate and
to assure a sufficient uptake of [PH]leucine, a greater than usual volume of
precursor was injected. After a downflow period of 22 hr, a typical crest
pattern indicative of fast axoplasmic transport was found (Fig. 7). The
average rate determined from two animals was 389 mm/day (Table 1).
The rat was chosen to show fast axoplasmic transport in a small-sized
animal. After injection of [3H]leucine into the L7-S1 segments of rat spinal
cords for uptake by their motoneurones, times of 2-6 hr were allowed for
transport into the motor fibres of the sciatic nerve. Crests were found at
the usual distance from the estimated position of the motoneurone cell
bodies in the cords with a rate of fast transport in a group of eighteen
nerves of 411 mm/day (Table 1).
 FAST TRANSPORT IN NERVE 639

Rate of transport in various sized myelinated fibres

The similarity of the shape of the crest found in the different nerves
studied raised the question as to whether the rate of fast axoplasmic trans-
port has any relation to the diameter of the nerve fibres. To investigate
this point, the content of activity present in fibres at the crest of down-
flowing activity was examined by radioautography to determine if fibres
107-
Goat
ot=22 hr
106- ¢\V= 382-
106 It ~~~~~~~~mm/day

E 104t
103

102 G
Root Nerve
30 30 60 90 120 150 180 210 240 270 300 330 360
0
mm
Fig. 7. Transport in goat sciatic nerve. Ganglion injected with [3H]leucine
and nerve removed 22 hr later. Six mm segments taken for distribution
determination. Arrow shows the downflow expected distance with the rate
of fast axoplasmic transport, at 382 mm/day in this example.

of all diameters contained radioactivity. For this purpose the L7 dorsal

root ganglia of cats were injected with [3H]leucine as usual, and after
allowing a downflow time of 5 or 6 hr, 1 cm lengths of sciatic nerve encom-
passing the crest region were removed and prepared by freeze-substitution
for radioautography (Methods). The time required for the appearance of
grains in the radioautographs took at least 3-4 months. In these prepara-
tions the osmium tetroxide dark stained myelin sheaths with diameters
ranging from 3 to 23 ,tm are seen as nearly circular profiles with their
sheath compact and well preserved (Fig. 8) (Ochs, 1965). These features
indicate that the technique of freeze-substitution holds the shape of the
fibre close to that of the normal nerve fibre and with less fixation distortion
(Baker, 1958). However, some fine alterations may occur as shown at the
electron microscope level. A few of the neurofilaments and microtubules
are clumped and ice crystallization occurs. This is seen as small irregularly
640 SIDNEY OCHS
shaped holes scattered throughout the preparations (Ochs, 1966). However,
crystallization may be sparse and in some areas the tissue can be free of
this disturbance altogether. In any case the irregularly shaped holes are
easy to eliminate as an artifact and do not affect inferences relating to the
shape and contents of the nerve fibres.
Nerve fibres of all sizes at the front of the crest region were seen to con-
tain grains of radioactivity within the axons (Fig. 8). These represent pro-

40 jca
Fig. 8. Radioautography at the crest position. Injection of the cat L7
ganglion with [3H]leucine and nerve removed 5 hr later. A short nerve seg-
ment taken at estimated crest position was prepared for histology by freeze-
substitution and sections of the nerve coated with Ilford L4 for radio-
autography. Myelinated nerve fibres shown as circular profiles with dark
osmium-stained compact myelin sheath. The fibres ranged in diameter from
3 to 23 #sm. Large nerve fibres labelled A show fairly uniform scattering
of labelled grains over the axon with the grains staining somewhat darker
than the myelin sheath. Smaller fibres B and C show a compact labelling of
grains, though some have a clear space in the centre. Still smaller fibres D
in the range of about 4 jum shows grains present inside them. An example
of small fibres E with a clear open space in the centre are given. Note, letter
labelling does not designate any special size class of fibre.
 FAST TRANSPORT IN NERVE 641
teins incorporating the labelled amino acid precursor (Droz & Leblond,
1962; Ochs et al. 1967). The few grains found scattered between the fibres
represents in a part a small amount of precursor which leaks out into the
circulation during the injection and is taken up and incorporated by
Schwann cells or bound to interneural elements. However, some regions
between the myelinated fibres show small patches of activity which sug-
gests the possibility that these grains are present in non-myelinated fibres.
Completion of electron microscope radioautography studies now in pro-
gress is, however, required to determine if such grains are in fact located
within non-myelinated axon profiles and thus if the rate of transport in
them is as fast as it is in myelinated fibres.
The grains inside the myelinated fibres were usually found dispersed
inside the axon without overlap on to the myelin sheath. Such results agree
with earlier studies indicating that the resolution possible with this
technique is close to and perhaps better than a micron (Ochs, 1966). In
some fibres the grains were positioned along the inside of the myelin
sheath with a central open area or an open area at one side of the axon.
This could possibly be due to an increased concentration of filamentous
material (microtubules and/or neurofilaments) in the clear areas. A similar
effect is seen in beaded nerves where the filamentous and fluid components
of the axoplasm are partitioned and a lower concentration of labelled
material was found present in the constricted regions where the filamentous
elements are concentrated (Ochs, 1965; Ochs, 1966). Such constrictions are
sometimes present in unstretched nerves. A more detailed study of the
effect of beading to partitioning of the labelled components carried down
the nerve fibres by fast axoplasmic transport with and without beading will
be presented elsewhere.
Another feature in the radioautographs is that a dense accumulation of
grains is found located over some fibres while others show very little or no
activity at all. It is obvious that there would be no activity in the motor
fibres after L7 dorsal root ganglion injection or in the sensory fibres when
the L7 ventral horn region was injected for a study of downflow of activity
in the motor fibres. However, when dorsal roots were examined after L7
dorsal root ganglion injection, or ventral roots after L7 ventral horn
injection, a number of the fibres were still found to have no activity or a
low level of activity. This variability in the content of labelled activity in
individual fibres is in part related to the local concentration of the injected
precursor relative to the cell-body population and possibly in part to
variations in the synthetic capabilities of individual neurones (Ochs, 1966;
Ochs & Johnson, 1969).
642 SIDNEY OCHS

DISCUSSION
The present study has shown that the same crest pattern of activity and
rate of fast axoplasmic transport is present in the sciatic nerves of mam-
malian species ranging in size from the goat down to the rat. The average
of all the rates of the larger groups of animals shown in Table 1 ranged
from 389 to 423 mm/day with a mean close to 410 mm/day. Statistical
comparison of the groups show that these means were taken from the same
population and even the relatively small variability in the rates between
the groups could have been due to experimental variation, chiefly of
temperature. The Q10 found for fast axoplasmic transport was 2-0-2-3
(Ochs & Smith, 1971b). For a 0 5' C change in temperature this would
amount to a 15 mm/day change in the rate of fast axoplasmic transport.
In these experiments the temperature in the lumbar region and upper
limbs was kept close to 380 C, but the temperature within the limbs was
not monitored and the actual temperature could have been somewhat
different. Temperature was better controlled in in vitro experiments where
a rate of 407 + 21 (S.D.) mm/hr at 380 C was determined (Ochs & Smith,
1971 a). However, several hours of downflow was required in the animal
before the nerves were removed and placed in the chambers so that some
uncertainty still remains as to the exact temperature dependence. How-
ever, that data obtained with the closest control of temperature is re-
assuringly close to the value of 409 + 50 (S.D.) mm/day found for the in vivo
fast transport rate for the cats in the present study. Also, it is close to the
over-all mean of values for all groups and we may therefore take 410 mm/
day as a reasonably close estimate of the rate of fast axoplasmic transport
either in vivo or in vitro of the labelled material investigated in these
studies. To be noted in this regard is that Kennedy, Fink & Byers (1972)
using techniques similar to ours recently reported an in vitro rate of
408 mm/day for the rabbit vagus nerve.
There was little evidence of a difference in rate between the sensory and
motor fibres either in the cat or in the monkey. A similar fast axoplasmic
transport was also shown for the motor fibres of the rat with a measured
rate of 411 + 50 (S.D.) mm/day in eighteen nerves (Table 1). This is of
interest with respect to the rate of 360 mm/day estimated by Miledi &
Slater (1970) from the difference in the time of failure of m.e.p.p. activity
in rat diaphragm muscles after cutting the phrenic nerve at low and at
high levels. This suggests that either transmitter or some material required
to maintain synaptic transmission is carried down the motor fibres by the
fast axoplasmic transport system.
If there are differences in the kinds of materials carried down into the
motor and sensory fibres by fast axoplasmic transport, this is not reflected
 FAST TRANSPORT IN NERVE 643
in their rate of transport. This is also seen to be the case for the materials
carried out into the two branches of the neurones of the dorsal root ganglia
(Fig. 4) where one branch of the neurone descends in the sciatic nerve to
terminate as a sensory receptor, the other ascends into the C.N.s. to end as
a presynaptic terminal for synaptic transmission.
In longer experiments no later appearing crests or peaks were ever seen
as might be expected from a second fast axoplasmic transport system even
as slow as 40 mm/day (cf. Figs. 3, 7). The possibility remains that a spread
of slower or intermediate rates exists, these constituting the plateau of
activity behind the crest. However, the alternative possibility that the
activity in the plateau represents a later egress of labelled components
from a 'compartment' in the cell body with some components left behind
the advancing front of the crest fits better with existing evidence. A com-
partmentalization was inferred from the insensitivity to the protein
blocking agents puromycin or cyclohexamide injected a short time after
the precursor [PH]leucine rather than beforehand (Ochs & Ranish, 1969).
The inference of a later outflow of fast transported labelled materials with
some amount remaining in the fibres behind the advancing crest of
activity was made from the level of activity in the plateau when the out-
flow from the soma was shut off by ligations below the ganglion made
several hours after its injection with [3H]leucine. A pattern very similar
to the decrease in plateau levels earlier reported for shorter downflow
times was found (Figs. 7-9, Ochs & Ranish, 1969 and cf. Fig. 1, Ochs &
Hollingsworth, 1971; Fig. 1, Kennedy et al. 1972). The level of the activity
present in the plateau after somal shut-off was found to redistribute itself
as expected of its fast transport with a portion of labelled activity re-
maining unchanged over long periods of time (so far up to 22 hr). This was
taken to indicate a fixity of labelled components in the axon or a rate of
transport slower than the low end of the range of fast rates reported.
Further experiments now in progress in this laboratory supporting this
conclusion will be presented elsewhere but the estimate so far arrived at is
that there is no 'slower' rate down to at least 40 mm/day for the type of
labelled material investigated in this study.
The maintenance of the shape of the crest over long distances, and its
linear outward movement with time indicates an active mechanism of
transport all along the axon rather than a process of diffusion or of peri-
stalsis. Direct evidence for an active transport, one depending on local
oxidative phosphorylation, has been obtained in in vitro studies (Ochs,
1972). In analogy to the sliding filament theory of muscle, an hypothesis
of a 'transport filament' was presented to account for the variety of labelled
components carried from the soma down inside the nerve fibres, all at the
same fast rate (Ochs, 1971 b, 1972). The close similarity of the rate of fast
644
644
SIDNEY
SDE
OCHS
01
axoplasmic transport seen in the nerves of different sized mammals, of
differing function and in myelinated fibres over the whole range of dia-
meters, suggests that the same underlying transport mechanism is generally
present in all mammalian nerve fibres. This possibility will, however,
require for its substantiation additional investigations to be made using
the techniques described here in many different species and nerve types
and also the investigation of transport of other materials.
The assistance of Mrs Mary Ann Neel, Mr Chris Velonis and in later experiments
Mr Tom Eckhoff in the course of these studies was very much appreciated. Thanks
are also due to Dr John Phillis who introduced us to the technique of anaesthetiza-
tion of the goat and to Mr James Glore and the Illustration Department for the
Figures. Supported by NIH 08706, NSF 28664X and the John A. Hartford Foun-
dation, Inc.
REFERENCES
BAKER, J. R. (1958). Principles of Biological Microtechnique. London: Methuen and
Co.
BARONDES, S. H. (1967). Axoplasmic transport. Neurosci. Re8. Prog. Bull. 5,
307-419.
BARONDES, S. H. (1969). Axoplasmic transport. In Handbook of Neurochemistry,
ed. LAJTHA, A., chap. 18, vol. 2. New York: Plenum Press.
DARLSTROM, A. (1971). Axoplasmic transport (with particular respect to adrenergic
neurons). Phil. Tranm. R. Soc. B 261, 325-358.
DAvISON, P. F. (1970). Axoplasmic transport: physical and chemical aspects. In
The Neurosciences: Second Study Program, ed. ScnTimTr, F. O., pp. 851-857. New
York: Rockefeller University Press.
DROZ, B. & LEBLOND, C. P. (1962). Migration of proteins along the axons of the
sciatic nerve. Science, N.Y. 137, 1047-1048.
FERNANDEZ, H. L., HUNEEUS, F. C. & DAvIsoN, P. C. (1970). Studies on the
mechanism of axoplasmic transport in the Crayfish cord. J. Neurobiol. 1, 395-409.
GRAFSTEIN, B. (1969). Axonal transport: Communication between soma and synapse.
In Advances in Biochemistry and Psychopharmacology, ed. COSTA, E. & GREEN-
GARD, P., vol. 1, pp. 11-25. New York: Raven Press.
HENDRICKSON, A. E. & CowAN, M. N. (1971). Changes in the rate of axoplasmic
transport during postnatal development of the rabbit's optic nerve and tract.
Expl Neurol. 30, 403-422.
KARLSsoN, J. & SJ6STRAND, 0. (1971). Synthesis, migration and turnover of proteins
in retinal ganglion cells. J. Neurochem. 18, 749-767.
KENNEDY, R. D., Fmnx, B. R. & BYERS, M. R. (1972). The effect of halothane on
rapid axonal transport in the rabbit vagus. Anesthesiology, 36, 433-443.
KIDWAI, A. M. & OcEs, S. (1969). Components of fast and slow phases of axoplasmic
flow. J. Neurochem. 16, 1105-1112.
KOPRIWA, R. M. & LEBLOND, C. P. (1962). Improvements in the coating technique
of radioautography. J. Histochem. Cytochem. 10, 264-284.
LASEK, R. J. (1968). Axoplasmic transport of labelled proteins in rat ventral
motoneurons. Expl Neurol. 21, 41-51.
LuIJsxA, L. (1964). Axoplasmic streaming in regenerating and in normal nerve
fibers in mechanisms of neural regeneration. In Progress in Brain Research,
ed. SINGER, M. & SCHAD, J. P., vol. 13, pp. 56-66. Elsevier.
 FAST TRANSPORT IN NERVE 645
LunI*sx, L. & NI mmE o, S. (1971). Velocity and intensity of bidirectional
migration of acetylcholinesterase in transacted nerves. Brain Res. 27, 329-342.
MILEDI, R. & SLATER, C. R. (1970). On the degeneration of rat neuromuscular
junctions after nerve sections. J. Physiol. 207, 507-528.
OCHS, S. (1965). Beading of myelinated nerve fibers. Expl Neurol. 12, 84-95.
OCHS, S. (1966). Axoplasmic flow in neurons. In Macromolecules and Behaviour, ed.
GAITO, J. New York: Appleton-Century-Crofts.
OCHS, S. (1969). Fast axoplasmic flow of protein and polypeptides in mammalian
nerve fibers. In Protein Metabolism of the Nervous System, ed. LAJTHA, A., pp.
291-304. New York: Plenum Press.
OCHS, S. (1971a). Local supply of energy to the fast axoplasmic transport. Proc.
natn. Acad. Sci. U.S.A. 69, 1279-1282.
OCHS, S. (1971b). The dependence of fast transport in mammalian nerve fibres on
metabolism. Acta Neuropath. suppl. v, 86-96.
OCHS, S. (1972). Fast transport of materials in manmmalian nerve fibres. Science
N.Y. 176, 252-260.
OCHS, S. & BURGER, E. (1958). Movement of substance proximodistally in nerve
axons as studied with spinal cord injections of radioactive phosphorus. Am. J.
Physiol. 194, 499-506.
OCHS, S., DALRYMPLE, D. & RICHARDS, G. (1962). Axoplasmic flow in ventral root
nerve fibers of the cat. Expl Neurol. 5, 349-363.
OCHS, S. & HoILINGSWORTH, D. J. (1971). Dependence of fast axoplasmic transport
in nerve on oxidative metabolism. J. Neurochem. 19, 107-114.
OCHS, S. & JOHNSON, J. (1969). Fast and slow phases of axoplasmic flow in ventral
root nerve fibers. J. Neurochem. 16, 845-853.
OCHS, S., JOHNSON, J. & NG. M.-H. (1967). Protein incorporation and axoplasmic
flow in motoneuron fibers following intra-cord injection of leucine. J. Neurochem.
14, 317-331.
OCHS, S. & RANISH, N. (1969). Characteristics of the fast transport system in mam-
malian nerve fibers. J. Neurobiol. 1, 247-261.
OCHS, S. & RANIsH, N. (1970). Metabolic dependence of fast axoplasmic transport
in nerve. Science, N.Y. 167, 878-879.
OCHS, S., SABRI, M. I. & JOHNSON, J. (1969). Fast transport system of materials in
mammalian nerve fibers. Science, N.Y. 163, 686-687.
OCHS, S. & SMITH, C. (1971a). Fast axoplasmic transport in mammalian nerve in
vitro after block of glycolysis with iodoacetic acid. J. Neurochem. 19, 833-844.
OCHS, S. & SMITH, C. (1971 b). Effect of temperature and rate of stimulation on fast
axoplasmic transport in mammalian nerve fibers. Fedn Proc. 30, 665.
RANISH, N. & OCHS, S. (1972). Fast axoplasmic transport of acetylcholinesterase in
mammalian nerve fibres. J. Neurochem. (in the Press).
SCHMITT, F. 0. (1968). Fibrous proteins-neuronal organelles. Proc. natn. Acad. Sci.
U.S.A. 60, 1092-1101.