Endocytosis of nerve growth factor by PC12 cells studied by quantitative ultrastructural autoradiography

Brain Research, 241 (1982) 145-156 Elsevier Biomedical Press

145

Endocytosis of Nerve Growth Factor by PC12 Cells Studied by Quantitative Ultrastructural Autoradiography RUTH HOGUE-ANGELETTI, ANNA STIEBER and NICHOLAS K. GONATAS Division of Neuropathology, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA 19104 (U.S.A.) (Accepted November 3rd, 1981) Key words: nerve growth factor - - PC12, pheochromocytoma - - endocytosis - - quantitative autoradiography

The endocytosis of [125I]nerve growth factor (NGF) by cultured rat pheochromocytoma cells, the PC12 line, was studied by ultrastructural quantitative autoradiography. Cells previously grown in the absence of NGF were incubated at 37 °C with P25I]NGF for periods of up to 26 h. Under these conditions, PC12 cells have not yet shown outgrowth of neurites. Heavy labeling of the plasma membrane was observed at 2 h. At 6 and 26 h, lower but significant levels of labeling of the plasma membrane were still noted. After 2, 6 and 26 h, endocytosis of [t25I]NGF was detected. Low breakdown of [leaI]NGF was observed only after 26 h. At 26 h, grain density distributions of [~25I]NGF showed significant labeling of lysosomes, while nuclei and rough endoplasmic reticulum showed the lowest levels of labeling. Significant apparent labeling of vesicles of smooth endoplasmic reticulum and of various cytoplasmic components, including cytoskeletal elements, was also observed. These findings indicate that [lzSI]NGF undergoes endocytosis quite slowly. During the initial phase of the interaction between NGF and PC12 cells, plasma membrane moieties are constantly labeled while lysosomes show progressively increasing uptake of NGF. The pathway of endocytosis of [I~5I]NGF included vesicles of the smooth endoplasmic reticulum but the Golgi apparatus was not unequivocally labeled. The findings do not support the hypothesis that NGF is transported to the nucleus for the initiation of transcriptional events. Our morphologic observations are consistent with the hypothesis that NGF constantly occupies sites on or adjacent to the plasma membrane, and that it slowly undergoes endocytosis into lysosomes. INTRODUCTION During the last several years there has been an increasing interest in the mechanism o f action of the nerve growth factor NGF3,13,19, 20. Available evidence suggests that N G F has multiple actions involving b o t h transcriptional and non-transcriptional events 3A3. Biochemical studies have shown binding of N G F to nuclear or perinuclear sites in addition to plasma membranes o f N G F responsive cellsZ,13,15, 18. However, electron microscopic autoradiographic studies o f [125I]NGF in dissociated sympathetic neurons a n d in animals have n o t shown localization o f N G F in nuclei, while s m o o t h endoplasmic reticulum, vesicular structures, lysosomes and multivesicular bodies were labeledT, 29. Similar observations were made with N G F coupled to the marker enzyme horseradish peroxidase ( H R P ) 20. In contrast to the above morphological studies performed in animals and with cultures of sympa0006-8993/82/0000-0000/$02.75 © Elsevier Biomedical Press

thetic neurons, Marchisio and coworkers have demonstrated nuclear and perinuclear localization o f [125I]NGF in cultured rat p h e o c h r o m o c y t o m a cells (line PC12) by qualitative light microscopic autoradiography 22. In view o f the apparent functional implications of the endocytosis and possible nuclear translocation o f N G F , we have examined by quantitative electron microscopic a u t o r a d i o g r a p h y the uptake of [lz5I]NGF by the PC12 line o f rat pheochromocytoma81, 32. PC12 cells have receptors to N G F and respond to it by neurite formation in a reversible fashion3L Furthermore, PC12 cells can be maintained in N G F - f r e e medium. Thus, m o r p h o logical studies o f [125I]NGF-PC12 interactions lend themselves to an analysis of the initial events of N G F on cells n o t previously exposed to the factor. In cultured sympathetic neurons or in animal experiments, endogenous N G F has already interacted with the target cells, prior to experimentation.

146 MATERIALS AND METHODS

Cell culture The PC12 pheochromocytoma cell line obtained from Lloyd Greene was grown in RPMI-1640 medium containing in 1 0 ~ horse serum (KC Biologicals, Lenexa, KA) and 5 ~ fetal calf serum (GIBCO) in the presence of 5 ~ CO2 at 37 °C 14. In preparation for the incubations with iodinated nerve growth factor, the cells were removed from the surface of the flasks by mechanical agitation and washed 3 times at 4 °C with 30 ml RPMI-1640 containing 1 mg/ml bovine serum albumin (Cohn fraction V, Sigma) which had been sterilized by filtration through a 0.2 # m filter (Falcon). The cells were replated in Falcon T-flasks and placed in the incubator for the duration of the experiment. Cells prepared in this manner adhered to the flask surface and maintained their viability throughout the course of the experiment as judged by Trypan blue exclusion. Agitation was not performed during the experimental incubations for it was found to disrupt a large proportion of the cells. At the termination of each experiment, the cells were harvested and centrifuged for 5 min at 600 g, resuspended in a minimal volume of medium and layered on a cushion (3 vols.) of RPMI-1640 containing 1 mg/ml albumin and 0.32 M sucrose in microcentrifuge tubes (1.5 ml BioRad) 15. After centrifugation for 2 min at 10,000 rpm in a Beckman Microfuge, the cell pellet was processed for electron microscopy. Preparation and iodination of nerve growth factor Mouse 2.5S nerve growth factor was prepared from adult male submandibular glands by the method of Bocchini and Angeletti 2. Iodination was performed using lactoperoxidase-glucose oxidase immobilized on Sepharose 6 enzymobeads, (BioRad). The incubation mixture contained 3 mCi of [1251]sodium iodide (New England Nuclear), 10/zg nerve growth factor and 25 #1 enzymobeads in a final volume of 175 #1 at pH 7.4. The reaction was allowed to proceed for 15 min at room temperature and was stopped by the addition of 200/~1 guanidine HC1. After centrifugation of the reaction mixture for 5 min at 1500 rpm in a Beckman TJ-6 tabletop centrifuge at 4 °C, the supernatant was applied to a Sephadex G-25 superfine column (1 × 20 cm) equili-

brated with RPMI-1640 containing 0.02 ~ gelatin as described by Costrini and Bradshaw 6. The first peak (96-99 ~ trichloracetic acid precipitable) was then stored at --70 °C until use. On the day of each experiment, an aliquot was thawed and applied to a BioGel P-100 column (1 × 25 cm) equilibrated in RPMI-1640 containing 1 mg/ml bovine serum albumin. The first peak eluted containing aggregated [125I]NGF was discarded, while the second peak, containing dimeric N G F , was used in the experiment of that day. Specific activities of 400-600 cpm/fmol were obtained such that experimental incubation mixtures with PC12 cells contained a final concentration of labeled N G F of about 1 nM. The final pool from the P-100 column was not sterilized before addition to the cultures, because the N G F absorbs to filters used for sterilization. Therefore, the RPMI-1640 plus albumin was sterilized by ultrafiltration (0.2/zm filter) and the columns were washed with 3 column volumes of solution before use. The labeled N G F so obtained could be completely competed off intact PC12 cells by 500-fold unlabeled N G F (Figs. 1 and 2). Binding experiments were performed as described by Herrup and Shooter 15 except that incubations were performed in RPMI-albumin medium. After incubations, determinations of TCA precipitability were performed both on the cell-free medium and on the cell pellets solubilized in 1 ~ sodium dodecyl sulfate and 8 M urea. All of the cell-associated radioactivity was solubilized in this manner. Fresb labeled N G F was found to possess unaltered TCA precipitability in the sodium dodecyl sulfate-urea solution. When digested with pronase, this same preparation had no significant TCA-precipitable material. Cell suspensions were fixed in 8 per cent formaldehyde in 0.2 M sodium cacodylate buffer, pH 7.35, at room temperature for 1 h. Cells were post-fixed in 1 ~ osmium tetroxide and 1.5 ~ potassium ferrocyanide, dehydrated in ethanol, embedded in Araldite and processed for autoradiography according to Salpeter s,11, 12,25,26. Briefly, 100 nm thick sections were mounted on collodion-coated slides, coated with carbon, and coated with a monolayer of Ilford L4 nuclear emulsion with a dipping machine (V. Aralaid, Toronto, Ontario) according to described methods s,11,12,25,26. Control cells were treated the same as test cells except they were not incubated with [lZaI]NGF. In

147 order to check the background of the emulsion and rule out positive chemography sections of control cells were mounted with sections from test cells on the same slide. Control cells had a few random grains (0.001-0.007 grains//zm 2) (Table II). In order to check for negative chemography, some slides from each experiment were exposed to light after coating with emulsion and then developed 5-6 weeks later; confluent grains were found over both section and slides. In order to randomize the data, sections from at least 5 different areas were used. Only one section from each area was photographed to avoid counting the same cell twice.

Quantitative analysis of autoradiograms A half distance (HD) of 100 nm was used and grain density distributions were calculated according to Salpeter s,H,12,25,26. In order to expedite the morphometric studies a programmed minicomputer (MINC 11, Digital Equipment, Maynard, MA) interfaced with an 'Orthoplex Coordinate Sensor' (digitizer), by Ladd Research Industries, Burlington, and with a Decwriter IV, also by Digital Equipment was used. Software in use is RT 11 and F O R T R A N . Detailed account of the developed program for the calculation of normalized grain density histograms has been published elsewhere 23. The computerized analysis of grain density distributions has given results comparable to those obtained by the point method of analysis 2z. In order to further confirm the degree of coincidence of data obtained by the point and computerized methods, in one experiment we performed the morphometric analys;s with both methods with similar results (Figs. 7 and 8). For comparison of experiments performed at different times, and in which different exposures of sections to nuclear emulsion were used, grain densities at one H D inside the plasma membrane (PM) were normalized and grain densities at other HDs were expressed on the basis of the normalized value 25. For the analysis of the distribution of [125I]NGF over the various cellular organelles or compartments, the probability circle method was used ~-6. The center of a circle, 1.7 H D in diameter, was superimposed over the center of the autoradiographic grain and the grain was ascribed to the compartment included within the circle. If two compartments fell within the circle, both were credited with one-half a grain. The

following compartments or organelles were studied: nuclei, mitochondria, lysosomes, plasma membrane, rough endoplasmic reticulum, Golgi apparatus, smooth endoplasmic reticulum including vesicles near the region of the Golgi apparatus and 'other'. 'Other' included ribosomes or polysomes, microtubules, and unidentified areas of the cell where due to tangential sectioning definite identification of an organelle was not possible. Relative surface areas of various compartments were evaluated by the point method. The printed electron micrograph was perforated by a grid of points, 2 cm apart, and the probability circle method was applied on points as on grains. Statistical significance of data was evaluated by z2-test.

A cid phosphatase This reaction was carried out with disodium salt of cytidylic acid as substrate (CMP, Sigma Chemicals)10, 24. Briefly, cells were fixed for 4-15 min at 4 °C with 2 . 5 ~ glutaraldehyde and 2 ~ paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.4, containing CaC12. Cells were washed with 0.1 M cacodylate buffer containing 5 ~ sucrose for 2 h at room temperature. Subsequently, cells were washed with 0.05 M acetate buffer, pH 5, plus 5 ~ sucrose and incubated in the medium containing CMP and lead nitrate for 2 h at 37 °C. Sections were post-fixed in 1 ~ OsOa in 1.5 ~ potassium ferrocyanide for 50 min at room temperature and processed for electron microscopy. In controls, CMP was omitted, and the reaction was negative. RESULTS As illustrated in Figs. 1 and 2, there was specific binding of [125I]NGF on PC12 cells, since simultaneous incubation of [12aI]NGF with 500-fold unlabeled N G F resulted in the reduction of the number of observed grains to background levels. In order to ascertain the natttre of the radioactivity associated with the PC12 cells in the autoradiographic experiments described below, the time course of binding to the intact PC12 cells was performed and the amount of TCA-precipitable radioactivity determined. As seen in Fig. 3A, there is a rapid rise in binding at 37 °C followed by a gradual decrease in cell-associated label. Then a steady level of counts is

148

Fig. 1. PC12 cells incubated with [125I]NGFfor 2 h at 37 °C washed and fixed for autoradiography. One /~m thick sections of Araldite-embedded cells. Note that majority of grains are over cells while grains over plastic (background) are few. × 1400. Fig. 2. Same as in Fig. 1 except that [125IlNGFwas incubated with 500-foldexcess of unlabeled NGF. Compare with Fig. 1. Note low number of grains over cells and plastic. × 1400. maintained. Fig. 3B shows the amount of TCAprecipitable material associated with the cells and with the medium in the same experiment. There is a gradual but steady drop in the TCA-precipitable counts, with the medium always possessing the larger amount of apparently degraded material. Ultrastructural studies of PC12 cells have been reported before ~2. In a detailed cytochemical study of the adsorptive endocytosis of a lectin by PC12 cells, we have described the extensive network of the Golgi apparatus found in these cells, as well as the presence of lysosomes characterized by an acid phosphatase reaction 4. In Fig. 4 of a PC12 cell incubated for acid phosphatase, precipitates of lead phosphate are detected in vesicles and cisternae of the Golgi apparatus, as well as in all membranebound dense bodies (secondary lysosomes or residual bodies). In Table I, the 4 experiments are summarized and the numbers of cells, grains, and/~m 2 counted are presented. In cells incubated with [lzSI]NGF for 2 h at 4 °C most of the grains were observed at the plasma membrane and grain density distributions were consistent with predominantly plasma membrane labeling (Figs. 5 and 6) 25. The sharp decline of the grain densities at 2-6 H D is consistent with scatter from the adjacent heavily labeled plasma membrane2L The overall pattern of grain density

distributions between cells incubated with [125I]N G F for 2 h at 37 °C is similar (compare Figs. 6 and 7). However, in cells treated with [125I]NGF for 2 h, the grain density at 1 H D inside the plasma membrane is about 4-6 times higher than grain densities at 4 °C or at 37 °C after 6 and 26 h. Also, more grains are found within cells at 5-10 H D or greater. This observation suggests that [12aI]NGF is initially associated with the plasma membrane and that it undergoes endocytosis quite slowly, unlike the epidermal growth factor which undergoes endocytosis quite rapidly in its target cells 5. Clustering of [125I]N G F in surface patches was not seen (Figs. 5 and 9). In that regard, it may be stated that convincing patching and capping of radiolabeled ligand has been observed by ultrastructural autoradiography (as in Fig. 5 of ref. 11). Therefore, the absence of patching or capping of [125I]NGF is real, and is not due to the limitations in methodology. Grain densities over 6 H D inside cells incubated for 2 h at 37 °C were higher than grain densities in cells incubated for 2 h at 4 °C (Figs. 6 and 7). This observation suggests that in cells incubated with [lzSI]NGF for 2 h at 37 °C, a significant percentage of the total radioactivity is attributed to definite intracytoplasmic localization. Grain density distributions from autoradiograms analyzed by the point and computerized methods gave similar results (compare Figs. 7 and

149 I

I

I

30,000

,.-I, u

,o 0

l

20,000

I0,Ooo

n

I

I

I

I

I

I

5

I0

15

20

3. I~1

I

1

I

I

!

I

I

I

ioo

,.o, ~t

50 t.)

F-

ttl O.

I

I

I

5

I

15 TIME OF INCUBATION (hr.) IO

2O

3b Fig. 3. Time course of association of [126I]NGF to PC12 cells. Incubations were performed in a final volume of 200 #1 of RPMI1640 containing 1 mg/ml bovine serum albumin. 105 cpm of freshly prepared dimeric [125I]NGF and 104 cells/tube were added. At the end of each incubation the mixture was layered on 3 vols. of the same medium containing 0.32 M sucrose and centrifuged. The supernatants were assayed for TCA-precipitability and the pellets were counted, solubilized and also assayed for TCA-precipitability. CPM denoted are those which can be competed off cells with unlabeled NGF. A: cell-associated [125I]NGF. B: TCAprecipitable counts in cells (Fq) and supernatant (©).

TABLE I

Experiment no.

Temp. of incubation Time (h) of [I~5] NGF (°C)

Number of grains counted

Number o f # 2 counted

Number of cells counted

2808A 2791 2808B 2823

4 37 37 37

898 1935 1639 2513

8920 6713 8680 8546

91 100 75 134

2 2 6 26

Fig. 4. PC12 cells processed for acid phosphatase. Note heavy labeling of lysosomes and of several cisternae and vesicles of the Golgi apparatus (GERL). × 38,000. G, Golgi apparatus; LY, lysosome.

Fig. 5. Ultrastructural autoradiography. PC12 cells incubated with [125I]NGF for 2 h at 37 °C. × 10,000.

151 PM

PM

2 hr.

4*

C

1.000

2 hr. 3 7 ' C

.900 ~80C

IO0

.coo

,\\-~ ~ \\\\ :-,-,)

x~o

"~

zoo

.~o

~.\~ ,\\'~

.=o~

Jl~ , 3 OUTSIDE

4

G

6

7

8

9

I0

II

/2

13

14

18 OVER

3 2 I ~ I 2 3 OUTSIDE J INS/DE

INSIDE

4

0

6

7

8

9

10

II

12

13

14

15 OVI~

7

Fig. 6. Normalized grain density distributions of [r'51]NGF in PC12 cells incubated with [125I]NGF for 2 h at 4 °C. PM, plasma membrane. Inside = grains inside cell. Outside = grains outside cell. Grain density at half distance inside cell were normalized to 1. Ordinate : normalized values of grain densities. Abscissa: half distances (Exp. 2808, see Table II). Data obtained and processed by computer. 8). There were no significant differences a t the 0.5 p r o b a b i l i t y level b e t w e e n d a t a derived f r o m the c o m p u t e r a n d by the m a n u a l p o i n t m e t h o d (Figs. 7 a n d 8; Z 2 value 13.56 with 7 degrees o f freedom). A u t o r a d i o g r a m s o b t a i n e d after i n c u b a t i o n o f cells with [125I]NGF for 6 h at 37 °C showed distinct endocytosis o f grains diffusely or in clusters. H o w ever, grains were still present at the p l a s m a m e m brane, while nuclei were n o t labeled (Figs. 9, 10 a n d Table II). G r a i n densities over nuclei v a r i e d f r o m 4 to 10 times b a c k g r o u n d levels; in other c o m p a r t ments grain densities were 154-840 times the b a c k g r o u n d levels. H i s t o g r a m s o f g r a i n distributions f r o m cells i n c u b a t e d with [125I]NGF for 6 h at 37 °C have confirmed the endocytosis o f significant

Fig. 7. Normalized grain density distribution of [z25I]NGF in PC12 cells incubated for 2 h at 37 °C (Experiment 2791A, see Table II). a m o u n t s o f [12~I]NGF, while equally significant n u m b e r s o f grains are detected at the p l a s m a m e m b r a n e (Fig. 10). This o b s e r v a t i o n is consistent with PM I,oou.goo~/ty777-y"7~'7//~

2 hr. 37 * e

200

•

7f

~

.100

I INSIDE

Fig. 8. Same as Fig. 7, but data obtained and processed by the point method, 'by hand'.

TABLE II [r~51] NGF grain densities ( G/~m 2) Experiment

Background

At 1 HD inside the plasma membrane

Over nuclei

Nuclear grains as a percentage of cytoplasmic grains

2808A* 2 h, 4 °C 2791A** 2 h, 37 °C 2808B*** 6 h 2823§ 26 h; 37 °C

0.007 0.001 0.007 0.0002

0.157 0.840 0.154 0.198

0.009 0.004 0.014 0.020

4.6 3.5 4.2 3.8

* Exposed for ** Exposed for *** Exposed for §i Exposed for

18 weeks. 3 weeks. 18 weeks. 8 weeks.

152

Fig. 9. Ultrastructural autoradiography. PC12 cells incubated with [125I]NGF for 6 h at 37 °C. × 12,000.

the notion that either plasma membrane N G F receptors are constantly becoming available for binding or that the initially bound N G F undergoes endocytosis quite slowly. The intracytoplasmic grains from [~25I]NGF in cells incubated with the ligand for 6 h at 37 °C did not show any preferential PM IO00

6 hr

37" C

9OO 800 .700

.500-

,,',,~" ,~

.400.

~,

:,x

2~02.00- \ .IO0-

t OUTSIDE

:" INSIDE

;"

e

9

Io

,I

a2

13

14

,5 o ~ R

t0

Fig. 10. Normalized grain density distribution of [125I]NGF in PC12 cells incubated for 6 h at 37 °C. (Experiment 2808B, Table II). Data obtained and processed by computer.

localization in lysosomes, the Golgi apparatus or other organelles. In contrast, in ceils incubated with [I~5I]NGF for 26 h, intracytoplasmic grains appeared in clusters over dense round bodies, representing secondary lysosomes (Fig. 11). In these cells, nuclei were virtually unlabeled (Fig. 11 and Table II). Quantitative analysis of grain density distributions in cells incubated with [125I]NGF for 26 h at 37 °C showed a pattern essentially similar to that observed in cells incubated with [x25I]NGF for 6 h (compare Figs. 10 and 12). A significant amount of radioactivity was noted at the plasma membrane as well as Jn submembranous areas 2-8 H D or 200-800 nm, inside the plasma membrane. This observation is consistent with the conclusions that N G F occupies constantly sites in or near the plasma membrane and that the ligand does not undergo a rapid endocytosis in a focal cytoplasmic or nuclear area. Statistical analyses of grain density distributions in experiments conducted at 4 °C and at 37 °C at different times showed highly significant differences (g 2

153

Fig. 11. Ultrastructural autoradiography. PC12 cells incubated with [I25I]NGF for 26 h at 37 °C. G, Golgi apparatus. >: 19,000 108.48 with 18 degrees of freedom for experiments 2808A and 2791, indicating signficiant differences at 0.1 probability level; Z2 = 249.39 for experiments 2808A and 2808B, and X~ = / 2 4 9 . 3 0 for experiments 2808A and 2823; see Table I). In order to gain more precise informatiort oi1 the compartmentalization of N G F , we analyzed grain density distributions of [125I]NGF in cells incubated for 6 and 26 h at 37 °C (Tables III, IV and Fig. 13). After 26 h, there was a convincing endocytosis of [125I]NGF in lysosomes, while the number of grains

over the plasma membrane, the Golgi apparatus and the smooth endoplasmic reticulum, including vesicles, did not change significantly between 6 and 26 h. After 26 h, some change was seen in the compartment summarily described as 'other'; this change may represent a real change or it may represent scatter from adjacent heavily labeled sources, especially the plasma membrane, the labeling of which decreases in cells treated for 26 h with [1251]NGF.

TABLE Ill Distribution o f [le51] NGF in nuclei and various organelles o f PC12 cells incubated with the factor for 6 h at 37 °C

Nuc, nuclei; Mit, mitochondria; Lys, lysosomes; PM, plasma membrane; Rer, rough endoplasmic reticulum; Ser, smooth endoplasmic reticulum, Other, polysomes, cytoskeleton, unidentifiable.

Grains Grains Points Points Grains Points

Nuc

Mit

Lys

PM

Rer

Golgi

Set

Other

Total

30 5.2 528 22.4

22 3.8 333 14.1

24 4.2 94 4.0

106 18.5 328 13.9

1 0.1 54 2.3

6 1.0 47 2.0

93 16.4 407 17.3

287 50.4 564 23.9

569

0.232

0.270

1.050

1.331

0.043

0.500

0.948

2.109

2355

154 TABLE IV

Distribution o f [1251]NGF in nuclei and various organelles o f PC12 cells incubated with the factor for 26 h at 37 °C

Grains Grains Points Points Grains Points

Nuc

Mit

Lys

PM

Rer

Golgi

Ser

Other

Total

46 4.2 664 22.0

126 11.8 467 15.5

238 22.2 227 7.5

117 10.9 308 10.2

1 0.1 96 3.2

18 1.6 73 2.4

67 6.2 239 7.9

457 42.7 933 31.0

1070

0.190

0.761

2.96

1.06

0.031

0.666

0.784

1.37

DISCUSSION

A number of studies have established that nerve growth factor has both acute and long-term effects which include stimulation of transport and anabolic reactions, rapid alterations of the morphology of cell surfaces, neurite outgrowth and increased synthesis of transmitter-synthesizing enzymesla. Bradshaw has proposed a stepwise mechanism of action of nerve growth factor which consists of: (1) binding of nerve growth factor with plasma membrane receptors resulting in increased uptake of metabolites; (2) endocytosis of the receptor-nerve growth factor complex; and (3) transport and fusion of internalized vesicles containing the receptor-factor complexes with lysosomes for their degradation and transport to the nucleus for the initiation of the transcriptional events 3. A careful ultrastructural analysis of the pathway of NGF internalization is of importance in understanding its mechanism of action. However, several PM I

2 6 hr.

3007

experimental problems arise which can obscure significant events; for example, high levels of nonspecific binding and proteolytic degradation. Two approaches have been used in these experiments in order to minimize the non-specific binding. Most critical was the removal of aggregated [125I]NGF by gel filtration on the day of each experiment6. Together with the use of the gradient system of Herrup and Shooter 16 to remove unbound label, we were able to consistently achieve levels of non-specific binding of 2 ~ or less. This is shown visually in the morphological analysis of Figs. 1 and 2, where cells pretreated with excess unlabeled NGF show only random background silver grains. Although degradation of NGF could be important to biological processes, it could also mask the undegraded NGF which could be more significant biologically. In order to evaluate the extent of artefact due to degraded NGF, an aliquot of cells processed for electron microscopy was tested for the amount of TCA-soluble material. It was found that most of the cell-associated TCA-soluble material remaining at the end of an incubation was removed in the early

37 ° C

I .,~.

\ N

.IIGO

3 2 I I 2 3 OUTSIDE ~ INSIDE

4

8

e

7

o

9

JO

II

12

13

14

15 OVER NUC

12 Fig. 12. Normalized grain density distributions of [lz5I]NGF in PC12 cells incubated for 26 h at 37 °C. (Experiment 2823, Table II).

MIT

LY$

PM

RER

GOLGI

SER

OTHER

13

Fig. 13. Illustration of distribution of [125I]NGF in nuclei and various organelles of PC12 cells incubated with the factor for 6 and 26 h. Based on data in Tables III and IV.

155 stages of processing. Secondly, the experiment shown in Fig. 3B demonstrates that the level of degradation of internalized N G F is much lower than reported for other systems such as epidermal growth factor 6. These two properties of our experiment system, that is, low non-specific binding and low levels of degradation, have permitted us to examine the internalization process by quantitative electron microscopic autoradiography, and should enhance the success of subsequent experiments outlined below. The noradrenergic clonal line of rat adrenal phecchromocytoma cell (PC12) has been useful for investigations of the mechanism of action of N G F because the cells respond morphologically and biochemically like the normal target cells, yet they are naive to N G F 14,2°,21,z°,31. It has been shown by biochemical methods that these cells possess high affinity receptors for N G F both on the plasma membrane and the nuclear membrane 33. Ultrastructural studies on PC12 cells have revealed a prominent Golgi apparatus and secretory granules 14, while PC12 cells treated with N G F for 24 h or longer periods show neurites filled with microtubulesZL Our studies were conducted with PC 12 not previously exposed to N G F which were subsequently incubated with [~25I]NGF for initial periods of up to 26 h, during which outgrowth of neurites was not observed. Under the conditions of these experiments the PC12 cells bound significant amounts of NGF, most of which was not degraded. The autoradiographic studies showed increased labeling of lysosomes by [lzaI]NGF after 26 h (Fig. 13). However, at this time, only about 25 ~ of total grains were found over lysosomes (Table IV). This observation is consistent with the low level of degradation of [lz~I]NGF shown in Fig. 3B. As shown in Fig. 3A, the binding rises rapidly with time, and then drops slowly to a stable plateau, possibly due to a dimiREFERENCES 1 Antoine, J. S., Avrameas, S., Gonatas, N. K., Stieber, A. and Gonatas, J. O., Plasma membranes and internalized immunoglobulins of lymph node cells studied with conjugates of antibody or its Fab fragments with horseradish peroxidase. J. cell. Biol., 63 (1974) 12-23. 2 Bocchini, V. and Angeletti, P. U., The nerve growth factor: purification of a 30,000 molecular weight protein, Proc. nat. Acad. Sci. U.S.A., 64 (1969) 787-787.

nished but signficiant number of plasma membrane receptors after prolonged incubation with NGF. This conslusion is supported by the quantitative autoradiographic data (compare Fig. 7 with Figs. 10 and 12). The quantitative ultrastructural autoradiographic studies did not reveal significant endocytosis of N G F in the Golgi apparatus-GERL in perinuclear sites or in the nucleus itself 1,9,1°,12,15,1s. Only lysosomes were definitely labeled after incubation of PC12 cells with [125I]NGF for 26 h at 37 °C (Figs. 11 and 13). The grain density distributions from [lzsI]N G F incubated with PC12 cells for 2-26 h at 37 °C showed heavy labeling of the plasma membrane and of a peripheral cytoplasmic region beneath the plasma membrane about 1.6/~m wide (Figs. 7, 10 and 12). The pathway of endocytosis of membranebound [125I]NGF may involve the Golgi apparatus or G E R L (Fig. 13); while significant radioactivity is noted over lysosomes, the nucleus and perinuclear areas remain virtually unlabeled (Fig. 13). Of particular interest is the slow rate of internalization of N G F in PC12 cells. This is in contrast to the rapid internalization of E G F in multivesicular bodies (lysosomes) and of lectins and cholera toxin into neuronal G E R L and lysosomes 5,9-12,17,18. Experiments now in progress in our laboratory, will define by quantitative ultrastructural autoradiography the localization of [125I]NGF in PC12 cells treated with the factor for longer periods of time and showing neurite outgrowth. These experiments may reveal whether significant labeling by [125I]NGF of cytoplasmic organelles other than lysosomes occurs. ACKNOWLEDGEMENTS Supported by Grants NS-13201, NS-00201, and NS-05572 of the National Institutes of Health, USPHS, and by a grant of the Amyotrophic Lateral Sclerosis Society of America. 3 Bradshaw, R. A., Nerve growth factor, Ann. Rev. Biochem., 47 (1978) 191-216. 4 Burchanowski, B. J., Angeletti, R. H., Stieber, A., Gonatas, J. and Gonatas, N. K., The ultrastructure and cytochemistry of PC12 cells with special reference to the Golgi apparatus and GERL, J. NeurocytoL, (1980) in press. 5 Carpenter, G. and Cohen, S., Human epidermal growth factor and the proliferation of fibroblasts, J. Cell Physiol., 88 (1976) 227-238.

156 6 Costrini, V. V. and Bradshaw, R. A., Binding characteristics and apparent molecular size of detergent solubilized nerve growth factor receptor of sympathetic ganglia, Proc. nat. Acad. Sci. U.S.A., 76 (1979) 3242-3245. 7 Claude, P., Dunis, D. A. and Hawrot, E., Binding and uptake of 125I-Nerve Growth Factor by dissociated sympathetic neurons in culture: Localization by electron microscopic autoradiography, J. Cell. Biol., 83 (1979) 632. 8 Fertuck, H. C. and Salpeter, M. M., Quantitation of the junctional and extrajunctional acetylcholine receptors by electron microscope autoradiography after lZSIa-bungarotoxin binding at mouse neuromuscular junctions, J. Cell Biol., 69 (1976) 144-158. 9 Gonatas, N. K., Kim, S. U., Stieber, A., Graham, D. I., and Avrameas, S., Internalization of neuronal plasma membrane ricin receptors into the Golgi apparatus, Exp. Cell Res., 94 (1975) 426-531. 10 Gonatas, N. K., Kim, S. U., Stieber, A. and Avrameas, S., Internalization of lectins in neuronal GERL, J. Cell Biol., 73 (1977) 1-13. 11 Gonatas, N. K., Gonatas, J. O., Stieber, A., Antoine, J. C. and Avrameas, S., Quantitative ultrastructural autoradiographic studies of iodinated plasma membranes of lymphocytes during segregation and internalization of surface immunoglobulins, J. Cell Biol., 70 (1976) 477-497. 12 Gonatas, J., Stieber, A., Olsnes, S. and Gonatas, N. K., Pathways involved in fluid phase and adsorptive endocytosis in neuroblastoma, J. Cell Biol., 87 (1980) 579-588. 13 Greene, L. A. and Shooter, E. M., The nerve growth factor: biochemistry, synthesis and mechanism of action, Ann. Rev. Neurosei., 3 (1980) 353402. 14 Greene, L. A. and Tischler, A. S., Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells, Proc. nat. Acad. Sci., U.S.A., 72 (1976) 2424-2428. 15 Herrup, L. and Shooter, E. M., Properties of the N G F receptor of avian dorsal root ganglia, Proe. nat. Aead. Sci. U.S.A. 70 (1973) 3384-3388. 16 Johnson, E. M., Jr., Andres, R. Y. and Bradshaw, R. A., Characterization of the retrograde transport of nerve growth factor (NGF) using high specific activity [125I]NGF, Brain Research, 150 (1978) 319-331. 17 Joseph, K. C., Kim, S. U., Stieber, A. and Gonatas, N. K., Endocytosis of cholera toxin into neuronal GERL, Proe. nat. Acad. Sci. U.S.A., 75 (1978) 2815-2819. 18 Joseph, K. C., Stieber, A. and Gonatas, N. K., Endocytosis of cholera toxin in GERL-like structures of murine neuroblastoma cells pretreated with GM1 ganglioside, J. Cell Biol., 81 (1979) 543-554. 19 Levi-Montalcini, R., The nerve growth factor: its role in growth, differentiation and function of the sympathetic adrenergic neurons. In M. A. Corner and D. F. Swaab (Eds.), Perspectives in Brain Research, Progress in Brain Research, Vol. 45, Elsevier, Amsterdam, 1976, pp. 235-256. 20 Levi, A., Schechter, Y., Neufeld, E. J. and Schlessinger,

21

22

23

24

25

26

27

28

29

30

31

32

33

J., Mobility, clustering and transport of nerve growth factor in embryonal sensory cells and in a sympathetic neuronal cell line, Proe. nat. Acad. Sei. U.S.A., 77 (1980) 3469-3473. Luckenbill-Edds., R., Van Horn, C. and Greene, H. A., Fine structure of initial outgrowth of processes induced in a pheochromocytoma cell line (PC12) by nerve growth factor, J. Neuroeytol., 8 (1979) 493-511. Marchisio, P. C., Naltini, L. and Calissano, P., Intracellular distribution of nerve growth factor in rat pheochromocytoma PC12 cells: evidence for a perinuclear and intranuclear location, Proe. nat. Acad. Sci. U.S.A., 77 (1980) 1656-1660. Merten, C., Stieber, A., Grand, E., Gonatas, J. and Gonatas, N. K., The use of a minicomputer for quantitative ultrastructural autoradiography, J. Histoehem. Cytochem., (1981) in press. Novikoff, A. B. and Novikoff, P. M., Cytochemical contributions to differentiating G E R L from the Golgi apparatus, Histoehem. J., 9 (1977) 525-551. Salpeter, M. M., Bach.mann, L. and Salpeter, E. E., Resolution in electron microscope radioautography, J. Cell Biol., 41 (1969) 1-20. Salpeter, M. M. and Backmann, L., Electron microscope autoradiography. In M. A. Hayat (Ed.), Principles and Techniques o f Electron Microscopy. Biological Applications Vol. 2, Van Nostran Reinhold, New York, 1972, pp. 221-278. Schwab, M. and Thoenen, H., Selective transsynaptic migration of tetanus toxin after retrograde axonal transport in peripheral sympathetic nerves: a comparison with nerve growth factor, Brain Research, 122 (1977) 459-474. Schwab, M. E., Ultrastructural localization of a nerve growth factor-horseradish peroxidase (NGF-HRP) coupling product after retrograde axonal transport in adrenergic neurons, Brain Research, 130: 190-196. Stockel, K., Guroff, G., Schwab, M. and Thoenen, H., The significance of retrograde axonal transport for the accumulation of systemically administered nerve growth factor (NGF) in the rat superior cervical ganglion, Brain Research, 109 (1976) 261-284. Sutter, A., Riopelle, R. J., Harris-Warrick, R. M. and Shooter, E. M., Nerve growth factor receptors: characterization of two distinct classes of binding sites on chick embryo sensory ganglia cells, J. Biol. Chem., 254 (1979) 5972-5982. Tischler, A. S. and Greene, L. A., Nerve growth factorinduced process formation by cultured rat pheochromocytoma cell, Nature (Lond.), 258 (1978) 341-342. Tischler, A. S. and Greene, L. A., Morphological and cytochemical properties of a clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor, Lab. Invest., 39 (1978) 77-89. Yanker, B. A. and Shooter, E. M., Nerve growth factor in the nucleus: Interaction with receptors on the nuclear membrane, Proe. nat. Aead. Sei. U.S.A., 76 (1977) 1269-1273.