Axoplasmic transport of mitochondria in cultured dorsal root ganglion cells

Brain Reseamh, 528 (1990) 285-290 Elsevier

285

BRES 15879

Axoplasmic transport of mitochondria in cultured dorsal root ganglion cells Toshifumi Takenaka 1, Tadashi Kawakami 1, Naoshi Hikawa I and Hideki Gotoh 2 JDepartment of Physiology, Yokohama City University, Fukuura Kanazawaku, Yokohama (Japan) and 2Department of Physiology, Iwate Medical University, Morioka, Iwate (Japan) (Accepted 27 March 1990) Key words: Axonal transport; Mitochondrion; Axon; Video microscopy; OrganeUe motion; Rhodamine 123

The movements of individual mitochondria in cultured mouse dorsal root ganglion cells were directly observed by using fluorescent staining with rhodamine 123 in combination with video microscopic techniques. This gives greater spatial and temporal resolution and much higher specificity than possible by conventional methods. The instantaneous velocities were 0.55 + 0.11 gm/s anterograde and 0.60 + 0.10 #m/s retrograde. Movement of the mitochondria was in fits and starts, and some reversed direction. The number of mitochondria moving retrogradely was 1.5-1.9 times greater than the number moving anterogradely. The average length of mitochondria moving retrogradely was 2.8 gm and of mitochondria moving anterogradely was 4.1/~m. These results suggest that mitochondria increase their numbers by division in the nerve fiber terminal. INTRODUCTION The supply of mitochondria to the axon and axon terminal is critical for neural function. The axoplasmic transport of mitochondria has previously been studied by observing the accumulation of mitochondria at ligations, or the kinetics of the distribution of radioactive labels. These studies provided evidence for rapid bidirectional movement of mitochondria, but suggested that mitochondria move more slowly than the most rapidly transported particles. Jeffrey et al. 9 used 59Fe as a marker for mitochondrial cytochromes and reported that they are transported in the nerve fibers at a rate of 1-4 mm/day. Additional support for a slow net anterograde movement of the mitochondria was obtained using the pattern of accumulation of the mitochondrial enzyme monoamine oxidase 1°. The movements of mitochondria can also be directly observed by light microscopy in living axons in which mitochondria can be seen to display rapid bidirectional saltatory movements. Smith 13 observed particle movement using dark field optics and identified the particles as mitochondria on the basis of electron microscopical observations. Recently Forman et al. 7 studied mitochondria movement more quantitatively by using a video enhanced technique in lobster axons and found that they are

transported at much higher rates than previously supposed, i.e. 0.72 + 0.26/xm/s in the anterograde direction and 1.33 + 0.64/zm/s in the retrograde direction. They assumed from the size and shape that the elongated organelles seen by light microscopy were all mitochondria. When using light microscopy, Nomarski optics, and phase-contrast techniques, identification of mitochondria is sometimes ambiguous, and for thick cells it is virtually impossible. The addition of a vital dye such as Janus green, which stains the mitochondria fairly specifically, aids recognition of the mitochondria but causes distortion of the mitochondria and cell death within several hours. Electron microscopy provides the greatest clarity and reliability in viewing the mitochondria, but it is restricted to fixed cells. Recently Chen et al. 4 found that rhodamine 123 is a good vital stain for mitochondria. Rhodamine 123 is much less toxic and has an additional advantage in that it can be excited and visualized by a standard filter system used for fluorescein. We used rhodamine 123 to track mitochondria in axonal transport and directly observed the movement of individual mitochondria in both directions with greater special and temporal resolution and much higher specificity than has been hitherto possible with conventional methods.

Correspondence: T. Takenaka, Department of Physiology, School of Medicine, Yokohama City University, 3-9, Fukuura, Kanazawaku, Yokohama, Japan 236. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

286

A micro tube pump

/ / /

cover glass - - - - / /

Fig. 1. Diagram of the experimental chamber. See text for details.

B

MATERIALS AND METHODS

Tissue cultures Dorsal root ganglia (DRG) from 3-month-old C57BL mice were incubated in 0.25% collagenase in Ham's F12 medium at 37 °C for 1.5 h. The cells were washed 3 times in Ham's F12 containing 10% fetal calf serum (FCS) and seeded onto 40 x 50 mm polylysinecoated cover glasses. DRG neurons were identified as having round and phase-bright cell bodies under a phase contrast light microscope. The cells used in this experiment were 16-30/am in diameter, and corresponded to class B of Sommer et al.14. After 3 days culture the 40 x 50 mm cover glasses were attached with waterproof tape to the underside of a 0.5-mm-thick stainless-steel plate (50 x 80 mm) with a lozenge-shaped hole. The upper side of the steel plate was covered with another cover glass, leaving small openings on both sides to perfuse solutions through (Fig. 1). This plate was placed on an inverted Zeiss Axiomat microscope equipped with a halogen lamp and neurons were observed with an AVEC (Allen video enhanced contrast)-differential interference contrast video microscope, with an oil-immersed planapoehromat 100 × 1.30 objective and a 3.2x zoom lensI. Video images with enhanced contrast were obtained using a video camera (Hamamatsu-Photonics Chalnicon camera).

Measurement of fluorescence intensity After 3 days in culture the cells grown on the polylysine-coated cover glass were washed twice with F12 medium and incubated in F12 solution containing 10/tM of rhodamine 123 (10/~g/ml, Kodak) for 5 min at 37 °C. The cells were then washed 3 or 4 times in PBS solution and set in a temperature-controlled chamber filled with F12 medium on the stage of a Zeiss Axiomat microscope. Fluorescence was detected with a silicon-intensified target camera (HamamatsuPhotonics SIT camera). The video output was fed through an AVEC system C1966 to a time-lapse tape recorder (SONY BVU870). The viewing chamber was maintained at 37 °C by warm air regulated by a thermosenser feedback system.

C

D

Fig. 2. Sequential pictures of mitochondria movements. B, C and D were recorded 23, 40 and 70 s, respectively, after recording A. a, b and c indicate individual mitochondria. Bar indicates 10/~m.

65 s and suddenly m o v e d in the retrograde direction at the speed of 0.57 pm/s. M i t o c h o n d r i o n b also stayed in the same position for 50 s and m o v e d at the speed of 0.44 pm/s. M i t o c h o n d r i o n c moved very slowly. F r o m this figure it can be seen that the mitochondrial m o v e m e n t was in fits and starts.

RESULTS

Velocities of mitochondria M i t o c h o n d r i a were observed to move in both anterograde and retrograde directions in a jerky, stop-and-go fashion. Velocities were measured only when mitochondria were actually in motion. The i n s t a n t a n e o u s velocity was 0.55 + 0.11 pm/s in the anterograde direction and 0.60 _ 0.10 pm/s in the retrograde direction. Fig. 2 shows sequential pictures of their movements. The relationship b e t w e e n their displacement and time is shown in Fig. 3, as calculated from the video tape record. In this record B, C and D were recorded 23, 40 and 70 s, respectively, from A. M i t o c h o n d r i o n a stayed in the same position for

~m

25 20

A #

B i

C i

D #

./

o.a" []

15

................ ~ b/~/

10

/'"

5 .....

Q

[]

o ......

o .....

a

,,. o

0 I

0

I

I

20

I

I

40

I

6/ 0

l

I

80

1

sec

Fig. 3. Relationships between spatial displacement and time of mitochondrial movements (a, b and c). This shows that mitochondria move in fits and starts.

287

Velocities of mitochondria and other organeUes Particle movements in the axon were also observed by video-enhanced differential interference contrast microscopy. Fig. 4 at the left shows the velocity histogram of the retrograde component obtained by this method. The histogram has a maximum peak at approximately 1.1 /zm/s, which coincides with previous reports 6"7'15"16. The maximum peak probably consists of movements of prelysosomal organelles. The histogram is not symmetric, but has a broad shoulder in the slow region. The rising phase of this shoulder coincides with the velocity distribution of mitochondria (filled circles) determined by the present vital staining method. Fig. 4 at right shows the velocity histogram of the anterograde component, in which the filled circles also show the velocity distribution of mitochondria. There are also much faster particles, whose speed ranged up to 6.0/zm/s, but these particles are much smaller than the slower particles and are not included in this figure. In both anterograde and retrograde transport the mitochondria were distributed mainly in the slow region of the transported substances.

Trajectories of movements We studied differences between the anterograde and retrograde movements of the mitochondria. As shown in Fig. 5, we classified the trajectories of 49 mitochondria observed in a single axon for 3 min into type 1 (straight), type 2 (inwardly deflected), type 3 (outwardly deflected), and type 4 (reversed). Of the particles, 10% suddenly reversed their direction during the observation period from anterograde to retrograde, but there was no reversal of movement from retrograde to anterograde. About 50% of the mitochondria moved straight in both anterograde and retrograde movements. The other 40%

25

/ ~

/

~0 ./.

i

/

15

1o / s

/ 9 .

~

~r © c 0

•

14

\

12

\

10

~l/oooo

o

-

"

e~ ~r © c0 o

e~ ~-

66~

TYPE 11 ,

RETROGRADE

ANTEROGRADE ~

~

m

m

,YPE ,,, ~

BE

•

IV

TYPE

60

40

0

20

i 20

s 40

i

60

(%) NUMBER OF PARTICLES

Fig. 5. Trajectories of mitochondrial movements which were classified into 4 types, straight, inwardly deflected, outwardly deflected and reversed. were of the deflected type. The overall characteristics seemed to be similar in both directions except for the reversed type.

Routes of mitochondria A single axon was divided into 3 equal longitudinal bands, a central and two peripheral routes. There was no difference between the two peripheral routes in the number of passing particles, and the central route had a slightly larger number of passing particles than the two peripheral routes. The direction of movement had no correlation with the route. The mean velocity of the mitochondria was 0.65/tm/s in the central route and 0.44 /~m/s in the two peripheral routes. The movement was faster in the central route.

Pattern of the mitochondrial movement We compared the flow rate of mitochondria in the anterograde direction with that in the retrograde direction at 3 sites (A, B, and C in Fig. 6) in the axon, 32.8, 61.2 and 84.8/~m from the cell body. The number passing through lines A, B, and C during a period of 180 s was 64, 52, 44 for the anterograde direction and 100, 90 and 85 for the retrograde direction. The retrogradely moving

\ "\.

oooo

olZ****

c~ ~ ~ c~ ~: ~ . £ ~: ~ ¢~ ¢~ ~ i o i n o I o I I I n I

L

/

4 ~ / 6 . .

\ \

q**

*\._.

s/" ~.

~. o o ~ o o o

o ***

16

TYPE I

\

~,~,,j

\

~. ,~. ~. ~. o. ~. ~. ~. ~, o. ~. ~. 0 0 0 0 . . . . ~ ¢~ ~ ~ I I I i I * i u * a a n

~ ~ - : ~ ~ - : ~ al o o o o ~ - , VELOCITY IN MICRONS PER SECOND

~ ~

Fig. 4. Velocity histogram of particles and mitochondria movements. The solid curve shows the particle movements observed by video-enhanced differential interference contrast microscopy. Filled circles show the velocity distribution of mitochondria observed by rhodamine 123 staining. Left: retrograde axoplasmic transport, Right: anterograde axoplasmic transport,

"~ ~

~

~(/

44 ~'

~

~

.

~

~, 52 ~..~'~'~IB 4~lur

~ lpr~

CELL BODY

85 LINE A

LINE a

Fig. 6. Flow rate of mitochondria in the anterograde and retrograde direction.The number passing through lines A, B and C were counted. Lines A, B and C were 32.8, 61.2, 84.8/tin from the cell body.

288 (pm) 10 uJ

0 ,(

y=o.0815+0.879X

/

(r=0'996U 0 "04 ,,r

z2 ,,J

2

4

6

8

LENGTH OF FLUORESCENT

10 (pro) IMAGE

Fig. 7. The correlation between the length of the mitochondria measured by fluorescent image and Nomarski image.

component became dominant at sites proximal to the cell body. This indicates that the mitochondria had a tendency to reverse direction as they moved toward the terminal. We actually observed the sudden reversal of mitochondrial movement from the anterograde to the retrograde direction• The size of retrogradely moving mitochondria was also smaller than that of anterogradely moving mitochondria. The number of mitochondria passing through the axon was 1•5-1.9 times greater in the retrograde direction than in the anterograde direction. The same tendency was observed in central and peripheral branches of neurites. We measured the length of mitochondria in the

lOOi Anterograde Retrograde

- - e - •

(t) UJ ..J

:,, ,,

pnO. U.

50

, ,

0

\ \

nUJ

Z

0

•

0

0

0

0

0

0

0

~4

~4

4

=

~

~

=

LENGTH lum) Fig. 8. Relationship of the size of mitochondda between anterograde and retrograde movement. The abscissa shows the length and the ordinate shows the number of particles. The length was measured in the long axis.

fluorescent image, but this image does not show true size. In order to compare the length of mitochondria in the anterograde and retrograde directions, we first compared the fluorescent image and the Nomarski image. Fig. 7 shows the correlation between the two in 25 examples• This figure shows that there is a good correlation between them (r = 0•99)• Next we compared the length of mitochondria in the fluorescent image. In Fig. 8 the length of the image is plotted on the abscissa and the number of particles is plotted on the ordinate. This figure shows that the average size of the mitochondria was 2.8 /zm in the retrograde direction and 4.1 /~m in the anterograde direction. Therefore, the size of retrograde mitochondria was 0.68 times that of anterograde mitochondria. The number of retrogradely moving particles was larger than that of anterogradely moving particles. This is difficult to explain in the light of reversed movements. It could be explained by the division of the anterograde mitochondria after they accomplished their task as an energy source in the growth cones• DISCUSSION

Specificity of rhodamine 123 staining Walsh et al. ]7 reported that only mitochondria were stained when mixed isomers of rhodamine B isothiocyanate were used. In their subsequent studies, the major component responsible for mitochondrion-specific staining was characterized as rhodamine 3B. Later, rhodamine 123 was found to be the isomer of choice because it provides low-background high-resolution fluorescent images of mitochondria without apparent cytotoxic effects4. NO staining of the plasma membrane or nuclear envelope is detected with rhodamine 123 and the membranes of lysosomes, endoplasmic reticulum, and the Golgi complex also are not stained. In the present study, rhodamine 123 faintly stained the axolemas of the cultured D R G cells, but this fluorescence faded quickly when irradiated. The fluorescence from the mitochondria, on the other hand, persisted for at least 15 rain of irradiation, probably because rhodamine molecules were effectively taken up into the mitochondria. Comparing the fluorescent images with non-specific Nomarski images, we concluded that the rhodamine 123 stains mitochondria specifically. The question may be raised whether the staining mechanism of rhodamine 123 modifies the interactions of these organelles with microtubules by altering mediation through kinesin, dynein or MAPs, since there was a special affinity of the dye for the outer mitochondria. But this can be ruled out, because there was no observable difference in mitochondrial movements between stained and unstained preparations.

289

The velocity of mitochondria The instantaneous velocity of mitochondria was 0.55 + 0.11 pm/s in the anterograde direction and 0.60 + 0.10 /~m/s in the retrograde direction. Such similarity of velocities between the anterograde and retrograde transport was also reported by Smith 13. His conclusion was based on electron microscope studies. These observations are in disagreement with those of Forman et al. 7, who concluded that the retrograde velocity was significantly larger than the anterograde velocity on the basis of non-specific Nomarski optics. The discrepancy could be due to this non-specific microscopy. Some early reports described the mitochondria as moving at a much slower net velocity. The net velocity of an individual vesicle or groups of vesicles can be described by the formula Vn = Vin

tm tm+ts

where Vn is the net velocity, Vin the instantaneous velocity, tm the average duration of movement, and ts the average duration of a stationary state. Previous papers reported that mitochondrial movement was less frequent than spherical particle movement 5'7. Quantitative determination of the proportion of mitochondria in a stationary state was difficult and was not done in the present study. However, we actually observed that a considerable proportion of the mitochondria were stationary, and that some begun to move suddenly (Fig. 3). If we analyze the temporal course of these movements, we may get a net velocity that is close to the early data on the slow movement of mitochondria. Blaker et al. 3 reported an exceptionally fast transport of mitochondria. They labeled a mitochondria-specific lipid, diphosphatidyl glycerol, with [2-3H]glycerol in the rat visual system. On the basis of the time difference between the arrival of label in the optic tract and in the superior colliculus, they calculated that the lipid was transported at a rate of about 200 mm/day, which is 4 times greater than our results. This peculiar result might be due to the exceptional characteristics of the visual system, but its precise explanation is open to further studies.

Routes of mitochondria We found that the direction of the movement had no correlation with routes. The mean velocity of mitochonREFERENCES 1 Allen, R.D., Allen, N.S. and Travis, J.L., Video-enhanced contrast, differential interference contrast (AVEC-DIC) microscopy: a new method capable of analyzing microtubule-related motility in the reticulopodial network of Allogromia laticollaris,

dria appeared to be faster in the central route. The relation between the route and the direction of mitochondria has been studied by several workers. Allen et al. 2 reported that the movement of mitochondria was mostly anterograde in the periphery of squid giant axons. Forman et al. 6 reported that in the interior of lobster giant axons more than 80% of particles moved toward cell bodies, but immediately inside the plasma membrane (_-<1/~m) a majority of particles (almost 70%) moved in the anterograde direction. We did not observe such difference between routes, and this discrepancy might be because we used cultured vertebrate neurons, or because the observation apparatus and analysis protocol were different.

Difference between anterograde and retrograde movement of mitochondria In the present work, the number of mitochondria returning from an axon terminal was more than 1.5 times the number of those moving anterogradely (Fig. 6). The size of returning mitochondria was significantly smaller than that of anterogradely moving ones (Fig. 8). If mitochondria simply exited the cell body and then returned, the number should be the same in both directions, but our data shows a significant difference. This difference cannot be explained even if we consider the reversal of some mitochondrial movement from anterograde to retrograde in the middle part of the axon. This leads to the conclusion that the mitochondria divide themselves at their reversing site or at the axon end. Mitochondria are quite dynamic entities in the sense that they oscillate, bend, become distorted, inflate, and sometimes divide, depending on the circumstances 11. Stempak 1: made a serial section analysis of mitochondrial form in rat liver cells. He found many dumbbell-shaped mitochondria. Sometimes the mitochondria lost structural detail in a thin central area, but showed a haze, suggesting fission. Glezer 8 reported division of mitochondria in a synaptic terminal of rat cerebral cortex. It has been known for some time that mitochondria themselves have DNA and RNA, and it has also become clear that they proliferate by a form of division 1~. These data suggest the possibility of mitochondrial division in the nerve terminal, and this could be the reason for the different number and size of mitochondria moving anterogradely and retrogradely.

Cell Motil., 1 (1989) 291-302. 2 Allen, R.D., Metuzals, J., Tasaki, I., Brady, S.T. and Gilbelt, S.P., Fast axonal transport in squid giant axon, Science, 218 (1982) 1127-1129. 3 Blaker, W.D., Goodrum, J.F. and Morell, P., Axonal transport of the mitochondria-specificlipid, diphosphatidyl-glycerol,in the

290 rat visual system, J. Cell Biol., 89 (1980) 579-584. 4 Chen, L.B., Summerhayes, I.C., Johnson, L.V., Walsh, M.L., Bernal, J.D. and Lampids, T.J., Probing mitochondria in living cells with rhodamine 123, Cold Spring Harbor Symp. Quant. Biol., XLVI (1982) 141-155. 5 Cooper, P.D. and Smith, R.S., The movement of optically detectable organelles in myelinated axons of Xenopus Laevis, J. Physiol., 242 (1974) 77-97. 6 Forman, D.S., Brown, K.J. and Livengood, D.R., Fast axonal transport in permeabilized lobster giant axons is inhibited by vanadate, J. Neurosci., 3 (1983) 1279-1288. 7 Forman, D.S., Lynch, K.J. and Smith, R.S., Organella dynamics in lobster axons: anterograde, retrograde and stationary mitochondria, Brain Research, 412 (1987) 96-106. 8 Glezer, I.I., Ultrastructure of mitochondria in neocortical neurons of the white rat, Fed. Proc., 23 (1963) T719-T724. 9 Jeffrey, P.I., James, K.A.C., Kidman, A.D., Richard, A.M. and Austin, L., The flow of mitochondria in chicken sciatic nerve, J. Neurobiol., 3 (1972) 199-208. 10 Khan, M.A. and Ochs, S., Slow axoplasmic transport of mitochondria (MAO) and lactic dehydrogenase in mammalian nerve fibers, Brain Research, 96 (1975) 267-277. 11 Rabinowitz, M. and Swift, H.H., Mitochondrial nucleic acids

and their relation to the biogenesis of mitochondr~a, Physiol.

Rev., 50 (1970) 376-427. 12 Stempak, J., Serial section analysis of mitochondrial form and membrane relationships in the neonatal rat liver cell, J. Ultrastr. Res., 18 (1967) 619-636. 13 Smith, R.S., The short term accumulation of axonally transported organelles in the region of localized regions of single myelinated axons, J. Neurocytol., 9 (1980) 39-65. 14 Sommer, E.W., Kazimierczak, J. and Droz, B., Neuronal subpopulation in dorsal root ganglion of the mouse as characterized by combination of uitrastructural and cytochemical features, Brain Research, 346 (1985) 310-326. 15 Takenaka, T., Gotoh, H. and Horie, H., Particle movements and microtubules in axoplasmic transport. In Y. Tsukada (Ed.), Perceptives on Neuroscience from Molecule to Mind, Univ. of Tokyo Press, 1985, pp. 81-90. 16 Takenaka, T., Particle movements in axoplasmic transport. In Z. Iqubal (Ed.), Axoplasmic Transport, CRC Press, 1986, pp. 10%118. 17 Walsh, M.L., Jen, J. and Chen, L.B., Transport of serum components into structures similar to mitoehondria, Cold Spring Harbor Conf. Cell Proliferation, 6 (1979) 513.