- Email: [email protected]
Cytomatrix protein residence times differ significantly between the tract and the terminal segments of optic axons
Brain Research, 517 (1990) 143-150 Elsevier
143
BRES 15482
Cytomatrix protein residence times differ significantly between the tract and the terminal segments of optic axons Paola Paggi 1'2, Raymond J. Lasek' and Michael J.
Katz 1
I Bio-architectonics Center, School of Medicine, Case Western Reserve University, Cleveland, OH 44106 (U.S.A.) and 2Dipartimento di Scienze e Tecnologie Biomediche e Biometria, Universita' de L'Aquila, L'Aquila (Italy)
(Accepted 24 October 1989) Key words: Axonal transport; Axonal terminal; Cytoskeleton; Transport kinetics; Slow component; Clathrin; Actin
The window method of radiolabeled protein analysis was used to study the transport kinetics of axonally transported cytomatrix proteins as they move through segments of mouse optic axons. Three slow component b (SCb) proteins - - actin, a 30 kDa protein, and clathrin - were radiolabeled in the eye and were followed for up to 119 days by quantitative one-dimensional gel electrophoresis. These proteins appeared first in the optic nerve, next in the tract, and last in the superior colliculus. All of the radiolabeled proteins had passed through the optic axons and had been effectively removed from the terminals by 119 days. Two different axonal segments ('windows') were examined in detail: a segment of the axon shaft region in the optic tract, and a segment of axon terminal region in the midbrain superior coUiculus. The median transit times of the 3 proteins were 53-100% longer in the colliculus than in the tract, and the pulse transients (the total area under the transport curve in each window) were 180-350% larger in the colliculus than in the tract. These results indicate that at least certain cytomatrix and cytoskeletal proteins have longer residence times in the terminal regions than in the axon proper.
INTRODUCTION
T h e optic system has b e e n used for a wide range of studies of b o t h the t r a n s p o r t of proteins within axons and the m e t a b o l i s m of axonal proteins in axon terminals 1' 5-8,10,16,21,22. In the optic system, the retinal ganglion cell axons are c o n t a i n e d within an e n c a p s u l a t e d bundle (the optic nerve and optic tract), and all these axons then t e r m i n a t e c o m p a c t l y in either m i d b r a i n or thalamic nuclei - - in rodents, at least three-quarters of the axons t e r m i n a t e in the m i d b r a i n superior colliculus (in mouse, it is g r e a t e r than 70%9). Studies of the m e t a b o l i c kinetics of axonally transp o r t e d p r o t e i n s in the axon shaft and in axon terminals have often used the window m e t h o d 1"6-8'1°'18'21'22. With this p a r a d i g m , the arrival and residence of pulse-labeled structures is s a m p l e d in selected segments o r 'windows' along a set of axons. Paggi and L a s e k 18 used this m e t h o d to analyze the t r a n s p o r t and residence times of specific cytomatrix proteins in the axon shaft and in the axon terminals of avian p a r a s y m p a t h e t i c o c u l o m o t o r neurons. T h e y found that the residence time (as m e a s u r e d by the pulse-transient) for cytomatrix proteins was much longer in the terminal regions than in the axon shaft. T h e y also f o u n d that the residence times differed in the terminals for different proteins (specifically, the neurofilament
proteins had relatively short residence times while slow c o m p o n e n t b (SCb) proteins had longer residence times). Based on these observations, Paggi and L a s e k 18 p r o p o s e d that these differences in the residence times are p r o d u c e d by the mechanisms that o p e r a t e selectively at the axon terminal to r e m o v e particular p r o t e i n s from the axon. Previously, we used the window m e t h o d to examine the transport kinetics of specific cytomatrix proteins conveyed with SCb in axons of the mouse optic nerve and optic tract 19. In the p r e s e n t p a p e r , we extend the use of the window m e t h o d to e x a m i n e the t r a n s p o r t times and the pulse transients of these same cytomatrix proteins as they move through the optic tract and into the axon terminal segments in the midbrain. MATERIALS AND METHODS Radiolabeling of the retinal ganglion cell proteins, dissection and preparation of tissue extracts The retinal ganglion cells were labeled as previously described19. In brief, 0.19 mCi of [3SS]methionine (spec. act. >800 mCi/mM; New England Nuclear, Boston, MA) was injected into the right vitreous humor of anesthetized male C57BI/6j mice, between 6 and 8 weeks old. Mice were then killed at 1-119 days after the labeling injection and the following segments were collected from the dissected optic system: (1) a 2 mm segment of the right optic nerve (N) before the optic chiasm extending 3-5 mm from the scleral surface of the eye, (2) a 2 mm segment of the left optic tract (T) beyond the optic chiasm extending 6--8 mm from the scleral surface
Correspondence: P. Paggi, Bio-architectonics Center, School of Medicine, C.W.R.U., Cleveland, OH 44106, U.S.A.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
144 of the eye, and (3) the left superior colliculus (SC), a 1.5-2 mm (major axis) segment of midbrain tissue beginning 11-11.5 mm from the scleral surface of the eye. The optic nerve (N) and tract (T) segments contain the axons of the retinal ganglion cells, and the superior colliculus segment (SC) contains the axons and the terminals of the retinal ganglion cells. Each of the segments was homogenized in 250 ~tl of BUST (0.1 M Tris containing 2% fl-mercaptoethanol, 8 M urea, 1% SDS, pH 6.8). The nerve and tract homogenates were centrifuged at 22,000 g (Sorvall SS-34 rotor) for 20 min; the superior colliculus homogenates were centrifuged at 130,000 g (Beckman Ty 65 rotor) for 1 h2. Measured aliquots of each supernatant were used to determine the total radioactivity and to separate the radiolabeled proteins by 1- and 2-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). To deal with the unavoidable variability in the amount of labeled amino acid incorporated into the proteins of the retinal ganglion cells of different animals, an a priori correction4 was introduced as previously described19.
SDS-PAGE and fluorography Radiolabeled proteins were separated19 on gradient slab gels (6-17.5% acrylamide, with a 4% stacking gel) that were run according to the procedure of Laemmli~. Two-dimensional isoelectrofocusing/SDS-PAGE was performed according to O'Farre117, employing the standard range of ampholines (pH 5-7; LKB Instruments). The radioactive polypeptides were visualized by fluorography19according to the method of Bonner and Laskey3 and Laskey and MillsTM.
Quantification of radioactivity in individual transported proteins Fluorographs were used to locate the position of the labeled polypeptides on regions of the 1-dimensional gels. These regions were then excised, solubilized, and their radioactivity was counted 19. Data for individual segments (T and SC) were plotted as a function of the interval between the labeling injection and the collection of the segments for analysis, according to the 'window' method of analysis 19.
28 days after injection. These postinjection times represent the most significant aspects of the slow c o m p o n e n t b (SCb) of axonal transport through the mouse optic system. The SCb proteins were maximally radioactive in the tract between 2 and 7 days after the labeling injection and they were maximally radioactive in the superior colliculus between 7 and 28 days after injection (Fig. 3). A t 4 days postlabeling, the gel electrophoresis patterns of the superior colliculus contained a small amount of r a d i o l a b e l e d fast-transport proteins (Fig. 1). These particular proteins reach the superior colliculus within a few hours postlabeling, but by 4 days they have been largely r e m o v e d from the superior colliculus and only a small a m o u n t remains. A t 28 d a y s postlabeling, the slower moving r a d i o l a b e l e d SCb proteins were c o n c e n t r a t e d in the superior colliculus, and by then most of the radioactivity had passed through the optic nerve and tract (Fig. 1; see also reference19). Quantitative analysis of 3 SCb proteins - - actin, a 30 k D a protein, and clathrin - - was carried out using 1-dimensional gels. A s the 2-dimensional P A G E fluorographs show (Fig. 2), no o t h e r m a j o r r a d i o l a b e l e d proteins have molecular weights corresponding to the actin, the 30 k D a protein, or the clathrin bands, and this confirms previous observations that 1-dimensional gels
Determination of the pulse-transient The pulse-transient1s'19 is a mathematical summary of the transport curve through an axonal window: it is simply the total area under the curve. In a previous paper w, to compare the pulsetransients of the optic nerve segment with the optic tract segment pulse-transients, the values for the optic tract were increased by 2.5% to adjust for the 2.5% of the retinal ganglion cell axons that are uncrossed at the optic chiasm and continue into the ipsilateral optic tract. Thus, the pulse-transients for the optic tract in Paggi et al. 19 are 2.5% greater than in the present paper. In the present paper, we compare the pulse-transients of the optic tract with the superior colliculus. In our system, 26% of the optic tract axon radioactivity is delivered not to the superior colliculus nucleus in the midbrain but instead to the diencephalic optic terminals in the lateral geniculate nucleus (Paggi and Lasek, unpublished observation). Therefore, we increased the superior colliculus pulse-transient values by 26% for our final comparisons in Table II. RESULTS
SDS-PAGE analyses of individual slowly transported proteins: a 30 kDa protein, actin and clathrin We have e x a m i n e d 1- and 2-dimensional fluorographs of segments from m o r e than 100 optic systems of mice killed at 1, 2, 4, 7, 10, 15, 21, 28, 37, 42, 47, 52, 61, 75, 90 and 119 days after injection. Figs. 1 and 2 show representative fluorographs of radiolabeled proteins, s e p a r a t e d by 1- and 2-dimensional S D S - P A G E at 4 and
Fig. 1. Fluorographs showing polypeptides (1-dimensional SDSPAGE analyses) at 4 and 28 days after labeling the retinal ganglion cells. Aliquots of homogenates from the optic tract (T) and the superior colliculus (SC) (see Materials and Methods) were analyzed on gradient slab gels (6--17.5% acrylamide). The bars at the fight indicate the positions of the molecular mass standards (200, 94, 68, 57, 43, and 30 kDa, reading from top to bottom). The bands corresponding to clathrin (C), actin (A), and the 30 kDa protein (30) are indicated by labels.
145
T
4d
SC
4d
28d
SC
28d
m
vim
Q
m
T
Fig. 2. Fluorographs showing labeled polypeptides (2-dimensional SDS-PAGE analyses) at 4 and 28 days after labeling the retinal ganglion cells. Aliquots of homogenates from the optic tract (T) and the superior coUiculus (SC) were subjected to isoelectric focusing in the first dimension, then they were electrophoresed on 6-17.5% gradient slab polyacrylamide gels in the second dimension. Acidic and basic portions of the gels are indicated. The bars at the left indicate the positions of the molecular mass standards (200, 94, 68, 57, 43, and 30 kDa, reading from top to bottom). The spots corresponding to clathrin (C), actin (A), and the 30 kDa protein (30) are indicated by labels. are useful for specific analyses of these particular SCb proteins 18'19. Our data consist of more than 300 fluorographs made from preparations obtained between 1 and 119 days postlabeling. For visual comparison of the specific protein bands, we excised the actin, the 30 kDa
protein, and the clathrin bands from photographs of representative 1-dimensional P A G E fluorographs; then we assembled the excised bands in tabular form so that their densities in the nerve, tract, and superior colliculus could be compared at different postlabeling intervals
146 (Fig. 3). (These postlabeling time intervals effectively represent the larger number of intervals that were actually analyzed.) Fig. 3 shows the sequence of appearance of the 3 radiolabeled SCb proteins in the optic system: they arrive first in the optic nerve, next in the tract, and last in the superior colliculus. (For example, the 30 kDa protein, which is the most radioactive of these 3 proteins, was maximally labeled in the optic nerve at 4 days postlabeling, it was maximally labeled in the optic tract at 4-7 days postlabeling, and it was maximally labeled in the superior colliculus at 7-28 days postlabeling.) In addition, visual comparison of the radiolabeled protein bands in Fig. 3 shows that the maximum density of the bands in the superior colliculus was greater than the maximum density found in either the optic nerve or the optic tract. (For example, the density of the 30 kDa protein in the superior colliculus was greater at 10 days than it was at its maximum in the optic nerve or tract, i.e. at 4 days.) This difference in density is particularly dramatic for actin, and the same general pattern was also observed for clathrin. The full amount of radioactivity in these pulse-labeled waves is established at the nerve cell body: no radioactivity is added as the wave progresses along the axon. At the same time, however, the system is dynamic, with radioactivity entering, traversing, and exiting each segment of the axon. The greater density of label seen in the superior colliculus indicates a relative 'slowing' in the
12 (SC) 7(T) 4(N)
03 rr ILl I'- 12 (SC) w 7(T) '-i 4(N) _J :E
IO 1'5 2'837
! ! t
|I
I!
t
30K
J
[
ACTIN
J
52
!
t
,z(sc)
! I
7 (T)
| l
! I
[CLATHRINJ
4(N) i I I 1'2 4
-( IO 1~5 2837 5'2
DAYS Fig. 3.30 kDa (30 K) protein, actin, and clathrin bands excisedfrom photographs of representative 1-dimensionalPAGE fluorographsas in Fig. 1. The excised bands were assembled in tabular form to compare the densities of the protein bands from the nerve (N), the tract (T), and the superior colliculus (SC) at different postlabeling intervals.
dynamic processes locally in that region. Specifically, the pulse-labeled wave appears to exit more slowly from the superior colliculus than it does from the optic tract; in other words, the radiolabeled proteins reside longer in the superior colliculus than in the optic tract. To more rigorously compare the intensities of these radiolabeled bands, the amounts of radioactivity in the individual protein bands were quantified by liquid scintillation spectroscopy, and the results were then plotted as a function of time postlabeling (Fig. 4).
Transport kinetics of the three SCb proteins We found that all of the radiolabeled SCb proteins passed through the optic system between 1 and 119 days after injecting radioisotope into the eye of the mouse (Fig. 4) - - the radiolabeled proteins first entered the optic tract within 1-2 days postlabeling, and they were essentially all removed from the optic system in the superior colliculus by 119 days postlabeling. Our experimental paradigm includes two different windows for viewing the transit of the radiolabeled proteins through the optic system. T, the optic tract, is located 6-8 mm from the eye. SC, the superior colliculus, is located approximately 11-13 mm from the eye. As a transport wave passes through a window, it is shaped by the local features within that window. Because the optic axons course through the optic tract without deviation or interruption, we assume that the axon shaft segments in this window are fairly uniform. On the other hand, the axon segments in the superior colliculus window contain tract-like regions of axon shaft (as it enters the colliculus), preterminal regions of the axon, and finally axon terminals. The pulse-labeled waves in the optic tract show the behavior of slowly transported proteins in a relatively simple environment composed largely of the axon shaft, while the pulse-labeled waves in the superior colliculus reveal the effects of more complex axonal structures, as the radiolabeled proteins sequentially experience the tract-like regions of the axon shaft, the preterminal regions, and then finally the axon endings themselves. Fig. 4 shows that in both the tract and the superior colliculus, the transport curves are asymmetric, with a rapidly rising front. Also, both axonal windows show transport curves with an extended declining phase that reaches background levels of radioactivity between 90 and 119 days after the experimental injection of radioactive label. (When this declining phase reaches background, all the detectable radioactivity has exited or has been removed from that segment of axon.) The median transit times for the three SCb proteins Paggi et al. 19 have shown that as the pulse-labeled
147 wave advances along the axon shaft it spreads and diminishes in amplitude, and this spreading appears to be a natural consequence of the inherent variety of transport rates even among otherwise identical molecules. Fig. 4 shows that the transport curves for the 3 SCb proteins in the superior colliculus were much broader than those in the optic tract. In other words, the natural dynamics of the radiolabeled proteins included longer transit times in the superior colliculus than in the optic tract. The median transit time is a whole axon statistic specifically, it is the time during which 50% of the window-analysis wave has passed from the retina into and through a given segment. In the axon shaft, where radioactive SCb molecules move through the segment by the axonal transport mechanisms, the median transit time reflects the overall clearance of an axonai segment by these transport mechanisms. In the superior colliculus (the midbrain end of the optic system) the median transit time reflects the overall rates of clearance of radiolabeled proteins from the axon terminals by the indigenous
T 7mm
SC 12 mm
140 1
60
NI
_o
20 0 32"
iI
=
i
i
i
CLATHRIN
24" 168-
O-
-:~
'7
i,
,
,
,
,
; ACTIN
8060- ~ ~ , 4020 0 2 21 37 52 75 I,2
2~1 57' 52,
75,
II
DAYS AFTER INJECTION Fig. 4. Axonal transport kinetics of radiolabeled 30 kDa protein, actin, and clathrin after labeling the retinal ganglion cells. Fluorographs like those in Fig. 1 were used to excise the gel regions containing these proteins. The data for the optic tract (T) segment 6-8 mm from the retina and the superior colliculus (SC) 11.5-13 mm from the retina are plotted as a function of the interval between injecting the radioactive precursor and collecting the segments for analysis. The points are the average of 3-9 observations, for a total of 90 observations. In the superior colliculus, the transport curves for the three SCb proteins are much broader than in the optic tract.
TABLE I
Meidan transit times* for the 3 SCb proteins
Actin 30 kDa Clathrin
Optic tract (T)
Superior colliculus % Change (SC) [100 x (SC- T)/T]
16 15 12
30 23 24
87 53 100
* Time required for 50% of the population of the radiolabeled proteins to traverse the examined segments from their origin in the eye.
removal mechanisms, as well as the rates of transport into the terminals. As we had found for more proximal parts of the optic system x9, the median transit times continued to increase as the slow component proteins moved distally. Table I shows that the median transit times in the superior colliculus were 30 days for actin, 23 days for the 30 kDa protein, and 24 days for clathrin, and these median transit times were 53-100% longer than the comparable times for the same proteins more proximally, in the optic tract.
The pulse-transients of the three SCb proteins For all 3 SCb proteins, Fig. 4 shows that the amplitudes of the transport curves in the superior colliculus were greater than those in the optic tract. For example, the peak amplitude for the 30 kDa protein was 9.6 × 103 dpm in the tract while it was nearly two times greater (18 × 103 dpm) in the colliculus. To further characterize such changes in these window analysis curves, Paggi and Lasek 18 have introduced the pulse-transient. (See also reference19.) The pulse-transient - - the area under the window-analysis transport curve - - provides a summary of the transport waves as seen through a fixed window (i.e. a given, standard axonal segment). Table II compares the pulse-transients of the labeled SCb proteins in the optic tract and the superior colliculus. The pulsetransients for the 3 SCb proteins in the superior colliculus were 195%-376% higher than those in the optic tract. By
TABLE II
Pulse-transients of 3 SCb proteins in different segments of the optic system The pulse-transients are reported as dpm-days x 10-4 for 2 mm segment of the optic tract (T) and for 1 superior colliculus (SC).
T
SC
% Change
[loo × (sc- T)/T1 Actin 30 kDa Clathrin
1292 3767 116
4269 11118 552
+230% + 195% + 376%
148 contrast with this large increase in the pulse-transient between the optic tract and the superior colliculus, Paggi et al. 19 observed that when the pulse-labeled wave advanced only within the axon shaft (from the optic nerve to the tract), the pulse-transients either remained essentially constant or decreased moderately. All the radiolabeled proteins that reach the superior colliculus must traverse the optic tract; thus, all the radiolabeled SCb proteins in the superior colliculus window must have appeared previously in the optic tract window. For this reason, differences between the pulsetransients in the tract and the superior colliculus are produced by local factors, namely, differences in the way that the radiolabeled proteins traverse or exit these two segments. Our results indicate that the 3 radiolabeled proteins have longer local residence times in the superior colliculus than they have in the more proximal optic tract. DISCUSSION Proteins were radiolabeled as a pulse in the retinal ganglion cells of the mouse. Three specific cytomatrix proteins conveyed with the slow component b (SCb) complex of proteins were followed as they moved in waves down the optic axons, and the quantity of radioactivity of each protein was measured by onedimensional S D S D - P A G E in an optic tract window and in the superior colliculus window. The optic tract window provides a view of the proteins in a white matter region, which is largely composed of axon shafts; the superior colliculus window provides a view of the same proteins in a grey matter region, where the axons terminate synaptically on dendrites and cell bodies of other neurons. In any region of a neural system, the residence times of axonal proteins are determined by their rates of entry, transit, and exit. In regions filled largely with axon shafts, such as the nerve and tract of the optic system, the slow axonal transport mechanisms move the radiolabeled proteins from one axonal segment to the next. Here, the same transport mechanisms remove proteins from one segment as they simultaneously introduce the proteins into the next segment. Thus, the clearance of proteins from regions composed only of axon shafts, is a function of the transport mechanisms and other axonal conditions that affect the movement of transported elements within the axon - - for the cytomatrix proteins, the residence times are determined by the rate of the slow axonal transport system and by the intra-axonal environment through which these elements advance. P r e v i o u s l y , w e 19 studied the transit behaviors of the slowly transported cytomatrix proteins in the optic nerve window and optic tract window; these are two white matter regions of the optic system, which contain only the
shafts of retinal ganglion axons. In these earlier experiments, we found that the pulse transients of the 30 kDa protein and clathrin were relatively constant between these two regions. For clathrin and the 30 kDa protein, the residence times are similar in the optic nerve and tract, and the local transport environment remains relatively unchanged throughout the length of these purely axon shaft regions. In contrast to the two contiguous axon shaft regions, the present experiments show that these same SCb proteins have dramatically increased residence times in the axon terminal region (the superior colliculus) (Tables I and II). The environment that the axonally transported proteins experience in the superior colliculus is more complex than that found in the optic nerve and tract. Like the nerve and tract, the superior colliculus also contains axon shaft regions, where the optic axons course from the tract through the stratum opticum of the superior colliculus. In addition, the superior colliculus contains the specialized terminal regions of the optic axons, which end in the stratum griseum and in the stratum opticum 9. The terminal optic axons branch, and in certain places their course can become more tortuous, and thus longer, than in an equivalent region of the white matter where the axons are fairly straight. Such geometric changes in the course of the optic axons can increase the volume of axoplasm through which the transported elements must negotiate, and this increase lengthens the effective window width and decreases the calculated rate of transport (which in our paradigm is simply the linear distance traversed by radiolabeled molecules in a single dimension of the neural system). However, the most convoluted paths of the optic axons in the superior colliculus are limited to the final approximately 20 a m after they have branched. In much of the preterminal regions - - the stratum opticum - - the optic axons are fairly straight, and here it would appear that the residence times of the transported proteins should differ little from those in an equivalent length of the optic tract. Axon terminals Window analyses graph the effective residence times of radiolabeled proteins in standard, defined regions of particular neural systems. In both the avian oculomotor system TM and the mouse optic system (the combined results of the present experiments and Paggi e t al.19), slow component proteins were found to reside in and to pass through the terminal regions with a different time course than in the more proximal axonal segments. These differences varied from protein to protein - - they cannot be fully accounted for by a uniform slowing of the rate of transport of the three SCb proteins. It appears, there-
149 fore, that other, differential, and selective mechanisms must contribute to the observed effects. Within the axon tract, the time required for removal of the transported proteins depends solely on axonal transport mechanisms. On the other hand, the removal time for proteins from the superior colliculus includes additional clearance mechanisms, because in a mature axon, many of the transported proteins are eventually removed from the end of the axon by various degradative processes 12'~s'2°. In the axon shaft, clathrin and the 30 kDa protein move as part of the same overall SCb complex; nonetheless, when they arrive in the pretermihal-terminal regions, the two proteins exhibit different residence (and clearance) behaviors. For instance, between the optic tract and superior colliculus, the median transit time increased by 53% for the 30 kDa protein while it increased by 100% for clathrin, and the pulse transient increased 195% for the 30 kDa protein while it increased 376% for clathrin. Similarly, in our earlier study on the oculomotor system TM, we found that the increase in the pulse transient in the axon terminal segments for slow component cytomatrix proteins ranged from a 280% increase for neurofilament proteins, through a 470% increase for actin and a 645% increase for brain spectrin, to an 870% increase for fl-tubulin. In this way, results from the window analyses are consistent with the notion that the clearance and removal mechanisms at the terminals differ even for proteins that have similar transport kinetics within the axon. The effectiveness of the local degradative mechanisms can determine the relative proportions of various proteins in the cytoskeleton and cytomatrix of the axon terminals ~2'13'1s. For example, those proteins that are degraded rapidly (cleared quickly) after entering the terminals will have a diminished representation by comparison to those that are degraded slowly. Our results indicate that the clearance times differ among the different cytomatrix and cytoskeletal proteins, suggesting that the local degradative mechanisms are proteinselective. Garner 4a has shown clearly the effects of these differences in the degradative mechanisms on the con-
centration of radiolabeled cytoskeletal proteins in the axon terminals. Using differential centdfugation, she separated the optic axon terminals (synaptosomes) from the preterminal axons. She found that when the pulselabeled slowly transported proteins are traversing the terminal region, the neurofilament proteins are present in the preterminal axons but are undetectable in the terminals. Apparently, the neurofilaments are rapidly cleared from the terminals. By contrast, tubulin and many of the SCb proteins, including clathrin, were present in both the preterminals and the terminals, and in some cases these proteins were concentrated selectively in the terminals after they had cleared the preterminal axon. Like the cytomatrix and cytoskeletal proteins, membranous proteins also have different clearance times in the terminals - - Martz et al. 15 found that the anterograde-to-retrograde conversion mechanisms (critical for the removal of the rapidly transported membranous proteins from axon terminals) operate selectively and remove certain membraneous proteins faster than others. Thus, a host of local metabolic mechanisms appear to operate selectively on the cytological elements and their proteins at the axon terminal. The collective effect of these mechanisms may then produce the dramatic differences seen between the fairly uniform intracellular architecture of the axon shaft and the specialized and more heterogeneous architecture of the terminal. In summary, axonal transport mechanisms relentlessly convey a full complement of proteins from the axon shaft into the axon tip. After their entry into the mature axon terminal, metabolic mechanisms operate selectively on these proteins, degrading and restructuring them, and in this way the local machinery within the terminal can determine and maintain the intracellular features that characterize the specialized cyto-architecture of axonal endings. Acknowledgements. This research was supported by the National Institutes of Health (R.J.L. and M.J.K.) and Ministero della Pubblica Istruzione - - fondi 40 e 60% - - 1987, 1988 (RE). Shirley Ricketts and Diane Kofskey provided excellent technical assistance.
REFERENCES 1 Baitinger, C. and Willard, M., Axonal transport of Synapsin I-like proteins in rabbit retinal ganglion cells, J. Neurosci., 7 (1987) 3723-3735. 2 Black, M.M. and Lasek, R.J., Slow components of the axonal transport: two cytoskeletal networks, J. Cell Biol., 86 (1980) 616-623. 3 Bonner, W.M. and Laskey, R.A., A film detection method for tritium labelled proteins and nucleic acids in polyaerylamide gels, Eur. J. Biochem., 46 (1974) 83-88. 4 Fisher, R.A., The Design of Experiments, (8th edn.), Ch. IX, Sect. 56, Haffner, New York, 1966, pp. 168-171. 4a Garner, J.A., Differential turnover of tubulin and neurofila-
5 6
7 8 9
ment proteins in central nervous system neuron terminals, Brain Research, 458 (1988) 309-318. Garner, J.A. and Mahler, H.R., Biogenesis of presynaptic terminal proteins, J. Neurochem., 49 (1987) 905-915. Goodrum, J.E, Toews, A.D. and Morell, P., Axonal transport and metabolism of [3H]fucose- and [3~S]sulfate-labeled macromolecules in the rat visual system, Brain Research, 176 (1979) 255-272. Grafstein, B. and Forman, D.S., Intracellular transport in neurons, Physiol. Rev., 60 (1980) 1167-1283. Grafstein, B., Murray, M. and Ingoglia, N.A., Protein synthesis and axonal transport in retinal ganglion cells of mice lacking visual receptors, Brain Research, 44 (1972) 37-48. Hofbauer, A. and Drager, U., Depth segregation of retinal
150
10 11 12 13 14 15
ganglion cells projecting to mouse superior colliculus, J. Comp. Neurol., 234 (1985) 465-474. Karlsson, J.O. and Sjostrand, J., Synthesis, migration and turnover of protein in retinal ganglion cells, J. Neurochem., 18 (1971) 749-767. Laemmli, V.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (Lond.), 227 (1970) 680-685. Lasek, R.J. and Katz, M.J., Mechanisms at the axon tip regulate metabolic processes critical to axonal elongation, Prog. Brain Res., 71 (1987) 49-60. Lasek, R.J., Studying the intrinsic determinants of neuronal form. In R.J. Lasek and M.M. Black (Eds.), Intrinsic Determinants of Neuronal Form, A.R. Liss, New York, 1988. Laskey, R.J. and Mills, A.D., Quantitative film detection of 3H and 14C in polyacrylamide gels fluorography, Eur. J. Biochem., 56 (1975) 335-341. Martz, D., Garner, J.A. and Lasek, R.J., Protein changes during anterograde-to-retrograde conversion of axonally trans-
ported vesicles, Brain Research, 476 (1989) 199-203. 16 McEwen, B.S. and Grafstein, B., Fast and slow components in axonal transport of protein, J. Cell Biol., 38 (1968) 494-508. 17 O'Farrel, P.H., High resolution two dimensional electrophoresis of proteins, J. Biol. Chem., 250 (1975) 4007-4021. 18 Paggi, P. and Lasek, R.J., Axonal transport of cytoskeletal proteins in ocuiomotor axons and their residence times in the axon terminals, J. Neurosci., 7 (1987) 2397-2411. 19 Paggi, P., Lasek, R.J. and Katz, M.J., Slow component b protein kinetics in optic nerve and tract windows, Brain Research, in press. 20 Roots, B.I., Neurofilament accumulation induced in synapses by leupeptin, Science, 221 (1983) 971-972. 21 Specht, S. and Grafstein, B., Accumulation of radioactive protein in mouse cortex after injection of [3H]fucose into the eye, Exp. Neurol., 41 (1973) 705-722. 22 Specht, S.C. and Grafstein, B., Axonal transport and transneuronal transfer in mouse visual system following injection of [3H]fucose into the eye, Exp. Neurol., 54 (1977) 352-368.
143
BRES 15482
Cytomatrix protein residence times differ significantly between the tract and the terminal segments of optic axons Paola Paggi 1'2, Raymond J. Lasek' and Michael J.
Katz 1
I Bio-architectonics Center, School of Medicine, Case Western Reserve University, Cleveland, OH 44106 (U.S.A.) and 2Dipartimento di Scienze e Tecnologie Biomediche e Biometria, Universita' de L'Aquila, L'Aquila (Italy)
(Accepted 24 October 1989) Key words: Axonal transport; Axonal terminal; Cytoskeleton; Transport kinetics; Slow component; Clathrin; Actin
The window method of radiolabeled protein analysis was used to study the transport kinetics of axonally transported cytomatrix proteins as they move through segments of mouse optic axons. Three slow component b (SCb) proteins - - actin, a 30 kDa protein, and clathrin - were radiolabeled in the eye and were followed for up to 119 days by quantitative one-dimensional gel electrophoresis. These proteins appeared first in the optic nerve, next in the tract, and last in the superior colliculus. All of the radiolabeled proteins had passed through the optic axons and had been effectively removed from the terminals by 119 days. Two different axonal segments ('windows') were examined in detail: a segment of the axon shaft region in the optic tract, and a segment of axon terminal region in the midbrain superior coUiculus. The median transit times of the 3 proteins were 53-100% longer in the colliculus than in the tract, and the pulse transients (the total area under the transport curve in each window) were 180-350% larger in the colliculus than in the tract. These results indicate that at least certain cytomatrix and cytoskeletal proteins have longer residence times in the terminal regions than in the axon proper.
INTRODUCTION
T h e optic system has b e e n used for a wide range of studies of b o t h the t r a n s p o r t of proteins within axons and the m e t a b o l i s m of axonal proteins in axon terminals 1' 5-8,10,16,21,22. In the optic system, the retinal ganglion cell axons are c o n t a i n e d within an e n c a p s u l a t e d bundle (the optic nerve and optic tract), and all these axons then t e r m i n a t e c o m p a c t l y in either m i d b r a i n or thalamic nuclei - - in rodents, at least three-quarters of the axons t e r m i n a t e in the m i d b r a i n superior colliculus (in mouse, it is g r e a t e r than 70%9). Studies of the m e t a b o l i c kinetics of axonally transp o r t e d p r o t e i n s in the axon shaft and in axon terminals have often used the window m e t h o d 1"6-8'1°'18'21'22. With this p a r a d i g m , the arrival and residence of pulse-labeled structures is s a m p l e d in selected segments o r 'windows' along a set of axons. Paggi and L a s e k 18 used this m e t h o d to analyze the t r a n s p o r t and residence times of specific cytomatrix proteins in the axon shaft and in the axon terminals of avian p a r a s y m p a t h e t i c o c u l o m o t o r neurons. T h e y found that the residence time (as m e a s u r e d by the pulse-transient) for cytomatrix proteins was much longer in the terminal regions than in the axon shaft. T h e y also f o u n d that the residence times differed in the terminals for different proteins (specifically, the neurofilament
proteins had relatively short residence times while slow c o m p o n e n t b (SCb) proteins had longer residence times). Based on these observations, Paggi and L a s e k 18 p r o p o s e d that these differences in the residence times are p r o d u c e d by the mechanisms that o p e r a t e selectively at the axon terminal to r e m o v e particular p r o t e i n s from the axon. Previously, we used the window m e t h o d to examine the transport kinetics of specific cytomatrix proteins conveyed with SCb in axons of the mouse optic nerve and optic tract 19. In the p r e s e n t p a p e r , we extend the use of the window m e t h o d to e x a m i n e the t r a n s p o r t times and the pulse transients of these same cytomatrix proteins as they move through the optic tract and into the axon terminal segments in the midbrain. MATERIALS AND METHODS Radiolabeling of the retinal ganglion cell proteins, dissection and preparation of tissue extracts The retinal ganglion cells were labeled as previously described19. In brief, 0.19 mCi of [3SS]methionine (spec. act. >800 mCi/mM; New England Nuclear, Boston, MA) was injected into the right vitreous humor of anesthetized male C57BI/6j mice, between 6 and 8 weeks old. Mice were then killed at 1-119 days after the labeling injection and the following segments were collected from the dissected optic system: (1) a 2 mm segment of the right optic nerve (N) before the optic chiasm extending 3-5 mm from the scleral surface of the eye, (2) a 2 mm segment of the left optic tract (T) beyond the optic chiasm extending 6--8 mm from the scleral surface
Correspondence: P. Paggi, Bio-architectonics Center, School of Medicine, C.W.R.U., Cleveland, OH 44106, U.S.A.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
144 of the eye, and (3) the left superior colliculus (SC), a 1.5-2 mm (major axis) segment of midbrain tissue beginning 11-11.5 mm from the scleral surface of the eye. The optic nerve (N) and tract (T) segments contain the axons of the retinal ganglion cells, and the superior colliculus segment (SC) contains the axons and the terminals of the retinal ganglion cells. Each of the segments was homogenized in 250 ~tl of BUST (0.1 M Tris containing 2% fl-mercaptoethanol, 8 M urea, 1% SDS, pH 6.8). The nerve and tract homogenates were centrifuged at 22,000 g (Sorvall SS-34 rotor) for 20 min; the superior colliculus homogenates were centrifuged at 130,000 g (Beckman Ty 65 rotor) for 1 h2. Measured aliquots of each supernatant were used to determine the total radioactivity and to separate the radiolabeled proteins by 1- and 2-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). To deal with the unavoidable variability in the amount of labeled amino acid incorporated into the proteins of the retinal ganglion cells of different animals, an a priori correction4 was introduced as previously described19.
SDS-PAGE and fluorography Radiolabeled proteins were separated19 on gradient slab gels (6-17.5% acrylamide, with a 4% stacking gel) that were run according to the procedure of Laemmli~. Two-dimensional isoelectrofocusing/SDS-PAGE was performed according to O'Farre117, employing the standard range of ampholines (pH 5-7; LKB Instruments). The radioactive polypeptides were visualized by fluorography19according to the method of Bonner and Laskey3 and Laskey and MillsTM.
Quantification of radioactivity in individual transported proteins Fluorographs were used to locate the position of the labeled polypeptides on regions of the 1-dimensional gels. These regions were then excised, solubilized, and their radioactivity was counted 19. Data for individual segments (T and SC) were plotted as a function of the interval between the labeling injection and the collection of the segments for analysis, according to the 'window' method of analysis 19.
28 days after injection. These postinjection times represent the most significant aspects of the slow c o m p o n e n t b (SCb) of axonal transport through the mouse optic system. The SCb proteins were maximally radioactive in the tract between 2 and 7 days after the labeling injection and they were maximally radioactive in the superior colliculus between 7 and 28 days after injection (Fig. 3). A t 4 days postlabeling, the gel electrophoresis patterns of the superior colliculus contained a small amount of r a d i o l a b e l e d fast-transport proteins (Fig. 1). These particular proteins reach the superior colliculus within a few hours postlabeling, but by 4 days they have been largely r e m o v e d from the superior colliculus and only a small a m o u n t remains. A t 28 d a y s postlabeling, the slower moving r a d i o l a b e l e d SCb proteins were c o n c e n t r a t e d in the superior colliculus, and by then most of the radioactivity had passed through the optic nerve and tract (Fig. 1; see also reference19). Quantitative analysis of 3 SCb proteins - - actin, a 30 k D a protein, and clathrin - - was carried out using 1-dimensional gels. A s the 2-dimensional P A G E fluorographs show (Fig. 2), no o t h e r m a j o r r a d i o l a b e l e d proteins have molecular weights corresponding to the actin, the 30 k D a protein, or the clathrin bands, and this confirms previous observations that 1-dimensional gels
Determination of the pulse-transient The pulse-transient1s'19 is a mathematical summary of the transport curve through an axonal window: it is simply the total area under the curve. In a previous paper w, to compare the pulsetransients of the optic nerve segment with the optic tract segment pulse-transients, the values for the optic tract were increased by 2.5% to adjust for the 2.5% of the retinal ganglion cell axons that are uncrossed at the optic chiasm and continue into the ipsilateral optic tract. Thus, the pulse-transients for the optic tract in Paggi et al. 19 are 2.5% greater than in the present paper. In the present paper, we compare the pulse-transients of the optic tract with the superior colliculus. In our system, 26% of the optic tract axon radioactivity is delivered not to the superior colliculus nucleus in the midbrain but instead to the diencephalic optic terminals in the lateral geniculate nucleus (Paggi and Lasek, unpublished observation). Therefore, we increased the superior colliculus pulse-transient values by 26% for our final comparisons in Table II. RESULTS
SDS-PAGE analyses of individual slowly transported proteins: a 30 kDa protein, actin and clathrin We have e x a m i n e d 1- and 2-dimensional fluorographs of segments from m o r e than 100 optic systems of mice killed at 1, 2, 4, 7, 10, 15, 21, 28, 37, 42, 47, 52, 61, 75, 90 and 119 days after injection. Figs. 1 and 2 show representative fluorographs of radiolabeled proteins, s e p a r a t e d by 1- and 2-dimensional S D S - P A G E at 4 and
Fig. 1. Fluorographs showing polypeptides (1-dimensional SDSPAGE analyses) at 4 and 28 days after labeling the retinal ganglion cells. Aliquots of homogenates from the optic tract (T) and the superior colliculus (SC) (see Materials and Methods) were analyzed on gradient slab gels (6--17.5% acrylamide). The bars at the fight indicate the positions of the molecular mass standards (200, 94, 68, 57, 43, and 30 kDa, reading from top to bottom). The bands corresponding to clathrin (C), actin (A), and the 30 kDa protein (30) are indicated by labels.
145
T
4d
SC
4d
28d
SC
28d
m
vim
Q
m
T
Fig. 2. Fluorographs showing labeled polypeptides (2-dimensional SDS-PAGE analyses) at 4 and 28 days after labeling the retinal ganglion cells. Aliquots of homogenates from the optic tract (T) and the superior coUiculus (SC) were subjected to isoelectric focusing in the first dimension, then they were electrophoresed on 6-17.5% gradient slab polyacrylamide gels in the second dimension. Acidic and basic portions of the gels are indicated. The bars at the left indicate the positions of the molecular mass standards (200, 94, 68, 57, 43, and 30 kDa, reading from top to bottom). The spots corresponding to clathrin (C), actin (A), and the 30 kDa protein (30) are indicated by labels. are useful for specific analyses of these particular SCb proteins 18'19. Our data consist of more than 300 fluorographs made from preparations obtained between 1 and 119 days postlabeling. For visual comparison of the specific protein bands, we excised the actin, the 30 kDa
protein, and the clathrin bands from photographs of representative 1-dimensional P A G E fluorographs; then we assembled the excised bands in tabular form so that their densities in the nerve, tract, and superior colliculus could be compared at different postlabeling intervals
146 (Fig. 3). (These postlabeling time intervals effectively represent the larger number of intervals that were actually analyzed.) Fig. 3 shows the sequence of appearance of the 3 radiolabeled SCb proteins in the optic system: they arrive first in the optic nerve, next in the tract, and last in the superior colliculus. (For example, the 30 kDa protein, which is the most radioactive of these 3 proteins, was maximally labeled in the optic nerve at 4 days postlabeling, it was maximally labeled in the optic tract at 4-7 days postlabeling, and it was maximally labeled in the superior colliculus at 7-28 days postlabeling.) In addition, visual comparison of the radiolabeled protein bands in Fig. 3 shows that the maximum density of the bands in the superior colliculus was greater than the maximum density found in either the optic nerve or the optic tract. (For example, the density of the 30 kDa protein in the superior colliculus was greater at 10 days than it was at its maximum in the optic nerve or tract, i.e. at 4 days.) This difference in density is particularly dramatic for actin, and the same general pattern was also observed for clathrin. The full amount of radioactivity in these pulse-labeled waves is established at the nerve cell body: no radioactivity is added as the wave progresses along the axon. At the same time, however, the system is dynamic, with radioactivity entering, traversing, and exiting each segment of the axon. The greater density of label seen in the superior colliculus indicates a relative 'slowing' in the
12 (SC) 7(T) 4(N)
03 rr ILl I'- 12 (SC) w 7(T) '-i 4(N) _J :E
IO 1'5 2'837
! ! t
|I
I!
t
30K
J
[
ACTIN
J
52
!
t
,z(sc)
! I
7 (T)
| l
! I
[CLATHRINJ
4(N) i I I 1'2 4
-( IO 1~5 2837 5'2
DAYS Fig. 3.30 kDa (30 K) protein, actin, and clathrin bands excisedfrom photographs of representative 1-dimensionalPAGE fluorographsas in Fig. 1. The excised bands were assembled in tabular form to compare the densities of the protein bands from the nerve (N), the tract (T), and the superior colliculus (SC) at different postlabeling intervals.
dynamic processes locally in that region. Specifically, the pulse-labeled wave appears to exit more slowly from the superior colliculus than it does from the optic tract; in other words, the radiolabeled proteins reside longer in the superior colliculus than in the optic tract. To more rigorously compare the intensities of these radiolabeled bands, the amounts of radioactivity in the individual protein bands were quantified by liquid scintillation spectroscopy, and the results were then plotted as a function of time postlabeling (Fig. 4).
Transport kinetics of the three SCb proteins We found that all of the radiolabeled SCb proteins passed through the optic system between 1 and 119 days after injecting radioisotope into the eye of the mouse (Fig. 4) - - the radiolabeled proteins first entered the optic tract within 1-2 days postlabeling, and they were essentially all removed from the optic system in the superior colliculus by 119 days postlabeling. Our experimental paradigm includes two different windows for viewing the transit of the radiolabeled proteins through the optic system. T, the optic tract, is located 6-8 mm from the eye. SC, the superior colliculus, is located approximately 11-13 mm from the eye. As a transport wave passes through a window, it is shaped by the local features within that window. Because the optic axons course through the optic tract without deviation or interruption, we assume that the axon shaft segments in this window are fairly uniform. On the other hand, the axon segments in the superior colliculus window contain tract-like regions of axon shaft (as it enters the colliculus), preterminal regions of the axon, and finally axon terminals. The pulse-labeled waves in the optic tract show the behavior of slowly transported proteins in a relatively simple environment composed largely of the axon shaft, while the pulse-labeled waves in the superior colliculus reveal the effects of more complex axonal structures, as the radiolabeled proteins sequentially experience the tract-like regions of the axon shaft, the preterminal regions, and then finally the axon endings themselves. Fig. 4 shows that in both the tract and the superior colliculus, the transport curves are asymmetric, with a rapidly rising front. Also, both axonal windows show transport curves with an extended declining phase that reaches background levels of radioactivity between 90 and 119 days after the experimental injection of radioactive label. (When this declining phase reaches background, all the detectable radioactivity has exited or has been removed from that segment of axon.) The median transit times for the three SCb proteins Paggi et al. 19 have shown that as the pulse-labeled
147 wave advances along the axon shaft it spreads and diminishes in amplitude, and this spreading appears to be a natural consequence of the inherent variety of transport rates even among otherwise identical molecules. Fig. 4 shows that the transport curves for the 3 SCb proteins in the superior colliculus were much broader than those in the optic tract. In other words, the natural dynamics of the radiolabeled proteins included longer transit times in the superior colliculus than in the optic tract. The median transit time is a whole axon statistic specifically, it is the time during which 50% of the window-analysis wave has passed from the retina into and through a given segment. In the axon shaft, where radioactive SCb molecules move through the segment by the axonal transport mechanisms, the median transit time reflects the overall clearance of an axonai segment by these transport mechanisms. In the superior colliculus (the midbrain end of the optic system) the median transit time reflects the overall rates of clearance of radiolabeled proteins from the axon terminals by the indigenous
T 7mm
SC 12 mm
140 1
60
NI
_o
20 0 32"
iI
=
i
i
i
CLATHRIN
24" 168-
O-
-:~
'7
i,
,
,
,
,
; ACTIN
8060- ~ ~ , 4020 0 2 21 37 52 75 I,2
2~1 57' 52,
75,
II
DAYS AFTER INJECTION Fig. 4. Axonal transport kinetics of radiolabeled 30 kDa protein, actin, and clathrin after labeling the retinal ganglion cells. Fluorographs like those in Fig. 1 were used to excise the gel regions containing these proteins. The data for the optic tract (T) segment 6-8 mm from the retina and the superior colliculus (SC) 11.5-13 mm from the retina are plotted as a function of the interval between injecting the radioactive precursor and collecting the segments for analysis. The points are the average of 3-9 observations, for a total of 90 observations. In the superior colliculus, the transport curves for the three SCb proteins are much broader than in the optic tract.
TABLE I
Meidan transit times* for the 3 SCb proteins
Actin 30 kDa Clathrin
Optic tract (T)
Superior colliculus % Change (SC) [100 x (SC- T)/T]
16 15 12
30 23 24
87 53 100
* Time required for 50% of the population of the radiolabeled proteins to traverse the examined segments from their origin in the eye.
removal mechanisms, as well as the rates of transport into the terminals. As we had found for more proximal parts of the optic system x9, the median transit times continued to increase as the slow component proteins moved distally. Table I shows that the median transit times in the superior colliculus were 30 days for actin, 23 days for the 30 kDa protein, and 24 days for clathrin, and these median transit times were 53-100% longer than the comparable times for the same proteins more proximally, in the optic tract.
The pulse-transients of the three SCb proteins For all 3 SCb proteins, Fig. 4 shows that the amplitudes of the transport curves in the superior colliculus were greater than those in the optic tract. For example, the peak amplitude for the 30 kDa protein was 9.6 × 103 dpm in the tract while it was nearly two times greater (18 × 103 dpm) in the colliculus. To further characterize such changes in these window analysis curves, Paggi and Lasek 18 have introduced the pulse-transient. (See also reference19.) The pulse-transient - - the area under the window-analysis transport curve - - provides a summary of the transport waves as seen through a fixed window (i.e. a given, standard axonal segment). Table II compares the pulse-transients of the labeled SCb proteins in the optic tract and the superior colliculus. The pulsetransients for the 3 SCb proteins in the superior colliculus were 195%-376% higher than those in the optic tract. By
TABLE II
Pulse-transients of 3 SCb proteins in different segments of the optic system The pulse-transients are reported as dpm-days x 10-4 for 2 mm segment of the optic tract (T) and for 1 superior colliculus (SC).
T
SC
% Change
[loo × (sc- T)/T1 Actin 30 kDa Clathrin
1292 3767 116
4269 11118 552
+230% + 195% + 376%
148 contrast with this large increase in the pulse-transient between the optic tract and the superior colliculus, Paggi et al. 19 observed that when the pulse-labeled wave advanced only within the axon shaft (from the optic nerve to the tract), the pulse-transients either remained essentially constant or decreased moderately. All the radiolabeled proteins that reach the superior colliculus must traverse the optic tract; thus, all the radiolabeled SCb proteins in the superior colliculus window must have appeared previously in the optic tract window. For this reason, differences between the pulsetransients in the tract and the superior colliculus are produced by local factors, namely, differences in the way that the radiolabeled proteins traverse or exit these two segments. Our results indicate that the 3 radiolabeled proteins have longer local residence times in the superior colliculus than they have in the more proximal optic tract. DISCUSSION Proteins were radiolabeled as a pulse in the retinal ganglion cells of the mouse. Three specific cytomatrix proteins conveyed with the slow component b (SCb) complex of proteins were followed as they moved in waves down the optic axons, and the quantity of radioactivity of each protein was measured by onedimensional S D S D - P A G E in an optic tract window and in the superior colliculus window. The optic tract window provides a view of the proteins in a white matter region, which is largely composed of axon shafts; the superior colliculus window provides a view of the same proteins in a grey matter region, where the axons terminate synaptically on dendrites and cell bodies of other neurons. In any region of a neural system, the residence times of axonal proteins are determined by their rates of entry, transit, and exit. In regions filled largely with axon shafts, such as the nerve and tract of the optic system, the slow axonal transport mechanisms move the radiolabeled proteins from one axonal segment to the next. Here, the same transport mechanisms remove proteins from one segment as they simultaneously introduce the proteins into the next segment. Thus, the clearance of proteins from regions composed only of axon shafts, is a function of the transport mechanisms and other axonal conditions that affect the movement of transported elements within the axon - - for the cytomatrix proteins, the residence times are determined by the rate of the slow axonal transport system and by the intra-axonal environment through which these elements advance. P r e v i o u s l y , w e 19 studied the transit behaviors of the slowly transported cytomatrix proteins in the optic nerve window and optic tract window; these are two white matter regions of the optic system, which contain only the
shafts of retinal ganglion axons. In these earlier experiments, we found that the pulse transients of the 30 kDa protein and clathrin were relatively constant between these two regions. For clathrin and the 30 kDa protein, the residence times are similar in the optic nerve and tract, and the local transport environment remains relatively unchanged throughout the length of these purely axon shaft regions. In contrast to the two contiguous axon shaft regions, the present experiments show that these same SCb proteins have dramatically increased residence times in the axon terminal region (the superior colliculus) (Tables I and II). The environment that the axonally transported proteins experience in the superior colliculus is more complex than that found in the optic nerve and tract. Like the nerve and tract, the superior colliculus also contains axon shaft regions, where the optic axons course from the tract through the stratum opticum of the superior colliculus. In addition, the superior colliculus contains the specialized terminal regions of the optic axons, which end in the stratum griseum and in the stratum opticum 9. The terminal optic axons branch, and in certain places their course can become more tortuous, and thus longer, than in an equivalent region of the white matter where the axons are fairly straight. Such geometric changes in the course of the optic axons can increase the volume of axoplasm through which the transported elements must negotiate, and this increase lengthens the effective window width and decreases the calculated rate of transport (which in our paradigm is simply the linear distance traversed by radiolabeled molecules in a single dimension of the neural system). However, the most convoluted paths of the optic axons in the superior colliculus are limited to the final approximately 20 a m after they have branched. In much of the preterminal regions - - the stratum opticum - - the optic axons are fairly straight, and here it would appear that the residence times of the transported proteins should differ little from those in an equivalent length of the optic tract. Axon terminals Window analyses graph the effective residence times of radiolabeled proteins in standard, defined regions of particular neural systems. In both the avian oculomotor system TM and the mouse optic system (the combined results of the present experiments and Paggi e t al.19), slow component proteins were found to reside in and to pass through the terminal regions with a different time course than in the more proximal axonal segments. These differences varied from protein to protein - - they cannot be fully accounted for by a uniform slowing of the rate of transport of the three SCb proteins. It appears, there-
149 fore, that other, differential, and selective mechanisms must contribute to the observed effects. Within the axon tract, the time required for removal of the transported proteins depends solely on axonal transport mechanisms. On the other hand, the removal time for proteins from the superior colliculus includes additional clearance mechanisms, because in a mature axon, many of the transported proteins are eventually removed from the end of the axon by various degradative processes 12'~s'2°. In the axon shaft, clathrin and the 30 kDa protein move as part of the same overall SCb complex; nonetheless, when they arrive in the pretermihal-terminal regions, the two proteins exhibit different residence (and clearance) behaviors. For instance, between the optic tract and superior colliculus, the median transit time increased by 53% for the 30 kDa protein while it increased by 100% for clathrin, and the pulse transient increased 195% for the 30 kDa protein while it increased 376% for clathrin. Similarly, in our earlier study on the oculomotor system TM, we found that the increase in the pulse transient in the axon terminal segments for slow component cytomatrix proteins ranged from a 280% increase for neurofilament proteins, through a 470% increase for actin and a 645% increase for brain spectrin, to an 870% increase for fl-tubulin. In this way, results from the window analyses are consistent with the notion that the clearance and removal mechanisms at the terminals differ even for proteins that have similar transport kinetics within the axon. The effectiveness of the local degradative mechanisms can determine the relative proportions of various proteins in the cytoskeleton and cytomatrix of the axon terminals ~2'13'1s. For example, those proteins that are degraded rapidly (cleared quickly) after entering the terminals will have a diminished representation by comparison to those that are degraded slowly. Our results indicate that the clearance times differ among the different cytomatrix and cytoskeletal proteins, suggesting that the local degradative mechanisms are proteinselective. Garner 4a has shown clearly the effects of these differences in the degradative mechanisms on the con-
centration of radiolabeled cytoskeletal proteins in the axon terminals. Using differential centdfugation, she separated the optic axon terminals (synaptosomes) from the preterminal axons. She found that when the pulselabeled slowly transported proteins are traversing the terminal region, the neurofilament proteins are present in the preterminal axons but are undetectable in the terminals. Apparently, the neurofilaments are rapidly cleared from the terminals. By contrast, tubulin and many of the SCb proteins, including clathrin, were present in both the preterminals and the terminals, and in some cases these proteins were concentrated selectively in the terminals after they had cleared the preterminal axon. Like the cytomatrix and cytoskeletal proteins, membranous proteins also have different clearance times in the terminals - - Martz et al. 15 found that the anterograde-to-retrograde conversion mechanisms (critical for the removal of the rapidly transported membranous proteins from axon terminals) operate selectively and remove certain membraneous proteins faster than others. Thus, a host of local metabolic mechanisms appear to operate selectively on the cytological elements and their proteins at the axon terminal. The collective effect of these mechanisms may then produce the dramatic differences seen between the fairly uniform intracellular architecture of the axon shaft and the specialized and more heterogeneous architecture of the terminal. In summary, axonal transport mechanisms relentlessly convey a full complement of proteins from the axon shaft into the axon tip. After their entry into the mature axon terminal, metabolic mechanisms operate selectively on these proteins, degrading and restructuring them, and in this way the local machinery within the terminal can determine and maintain the intracellular features that characterize the specialized cyto-architecture of axonal endings. Acknowledgements. This research was supported by the National Institutes of Health (R.J.L. and M.J.K.) and Ministero della Pubblica Istruzione - - fondi 40 e 60% - - 1987, 1988 (RE). Shirley Ricketts and Diane Kofskey provided excellent technical assistance.
REFERENCES 1 Baitinger, C. and Willard, M., Axonal transport of Synapsin I-like proteins in rabbit retinal ganglion cells, J. Neurosci., 7 (1987) 3723-3735. 2 Black, M.M. and Lasek, R.J., Slow components of the axonal transport: two cytoskeletal networks, J. Cell Biol., 86 (1980) 616-623. 3 Bonner, W.M. and Laskey, R.A., A film detection method for tritium labelled proteins and nucleic acids in polyaerylamide gels, Eur. J. Biochem., 46 (1974) 83-88. 4 Fisher, R.A., The Design of Experiments, (8th edn.), Ch. IX, Sect. 56, Haffner, New York, 1966, pp. 168-171. 4a Garner, J.A., Differential turnover of tubulin and neurofila-
5 6
7 8 9
ment proteins in central nervous system neuron terminals, Brain Research, 458 (1988) 309-318. Garner, J.A. and Mahler, H.R., Biogenesis of presynaptic terminal proteins, J. Neurochem., 49 (1987) 905-915. Goodrum, J.E, Toews, A.D. and Morell, P., Axonal transport and metabolism of [3H]fucose- and [3~S]sulfate-labeled macromolecules in the rat visual system, Brain Research, 176 (1979) 255-272. Grafstein, B. and Forman, D.S., Intracellular transport in neurons, Physiol. Rev., 60 (1980) 1167-1283. Grafstein, B., Murray, M. and Ingoglia, N.A., Protein synthesis and axonal transport in retinal ganglion cells of mice lacking visual receptors, Brain Research, 44 (1972) 37-48. Hofbauer, A. and Drager, U., Depth segregation of retinal
150
10 11 12 13 14 15
ganglion cells projecting to mouse superior colliculus, J. Comp. Neurol., 234 (1985) 465-474. Karlsson, J.O. and Sjostrand, J., Synthesis, migration and turnover of protein in retinal ganglion cells, J. Neurochem., 18 (1971) 749-767. Laemmli, V.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (Lond.), 227 (1970) 680-685. Lasek, R.J. and Katz, M.J., Mechanisms at the axon tip regulate metabolic processes critical to axonal elongation, Prog. Brain Res., 71 (1987) 49-60. Lasek, R.J., Studying the intrinsic determinants of neuronal form. In R.J. Lasek and M.M. Black (Eds.), Intrinsic Determinants of Neuronal Form, A.R. Liss, New York, 1988. Laskey, R.J. and Mills, A.D., Quantitative film detection of 3H and 14C in polyacrylamide gels fluorography, Eur. J. Biochem., 56 (1975) 335-341. Martz, D., Garner, J.A. and Lasek, R.J., Protein changes during anterograde-to-retrograde conversion of axonally trans-
ported vesicles, Brain Research, 476 (1989) 199-203. 16 McEwen, B.S. and Grafstein, B., Fast and slow components in axonal transport of protein, J. Cell Biol., 38 (1968) 494-508. 17 O'Farrel, P.H., High resolution two dimensional electrophoresis of proteins, J. Biol. Chem., 250 (1975) 4007-4021. 18 Paggi, P. and Lasek, R.J., Axonal transport of cytoskeletal proteins in ocuiomotor axons and their residence times in the axon terminals, J. Neurosci., 7 (1987) 2397-2411. 19 Paggi, P., Lasek, R.J. and Katz, M.J., Slow component b protein kinetics in optic nerve and tract windows, Brain Research, in press. 20 Roots, B.I., Neurofilament accumulation induced in synapses by leupeptin, Science, 221 (1983) 971-972. 21 Specht, S. and Grafstein, B., Accumulation of radioactive protein in mouse cortex after injection of [3H]fucose into the eye, Exp. Neurol., 41 (1973) 705-722. 22 Specht, S.C. and Grafstein, B., Axonal transport and transneuronal transfer in mouse visual system following injection of [3H]fucose into the eye, Exp. Neurol., 54 (1977) 352-368.