Different posttranscriptional controls for the human neurofilament light and heavy genes in transgenic mice

Molecular Brain Research, 18 (1993) 23-31

23

Elsevier Science Publishers B.V. BRESM 70575

Different posttranscriptional controls for the human neurofilament light and heavy genes in transgenic mice Lucille Beaudet, Francine C6t6, Daniel Houle and Jean-Pierre Julien Centre for Research in Neuroscience, McGill Uniuersity, The Montreal General Hospital Research Institute, Montreal, Que. (Canada)

(Accepted 27 October 1992)

Key words: Neurofilament; Gene regulation; Translational control; Intermediate filament; Transgenic mouse; Neuronal expression

To investigate the mechanisms regulating neurofilament gene expression, we generated transgenic mice with high copy number of the intact human neurofilament light (NF-L) and heavy (NF-H) genes. Overexpression in transgenic mice of NF-L mRNA from 3- to 5-fold in different regions of the central nervous system (CNS) resulted only in a mild increase of 10-50% in the levels of NF-L proteins. The failure to enhance NF-L protein content was not due to interspecies differences in posttranscriptional NF-L regulation. For instance, based on specific immunodetection, it is estimated that human NF-L proteins composed 80% of total NF-L content in the spinal cord of transgenics. In contrast to the situation with NF-L, the CNS of transgenic mice bearing multiple copies of the human NF-H gene showed comparable increases in the levels of NF-H mRNA and proteins. These results suggest that the NF-L and NF-H genes are subject to different posttranscriptional regulation in the CNS. In vivo labeling of newly synthesized proteins by injection of [35S]methionine in the spinal cords of normal and transgenic mice provided evidence that the posttranscriptional regulation of NF-L expression in the CNS must occur, at least in part, at the level of translation.

INTRODUCTION A t least seven d i f f e r e n t i n t e r m e d i a t e f i l a m e n t ( I F ) p r o t e i n s p a r t i c i p a t e in t h e f o r m a t i o n o f the I F n e t w o r k in n e u r o n s : the n e u r o f i l a m e n t ( N F ) t r i p l e t p r o t e i n s , N F - L , N F - M a n d N F - H , t h a t a r e the most a b u n d a n t cytoskeletal elements of central and peripheral neurons; t h e v i m e n t i n a n d n e s t i n t h a t a r e e x p r e s s e d t r a n siently in n e u r o e p i t h e l i a l s t e m cells; t h e p e r i p h e r i n , which is m a i n l y f o u n d in p e r i p h e r a l n e u r o n s ; a n d t h e a l p h a - i n t e r n e x i n , p r e s e n t in m o s t n e u r o n s o f t h e central n e r v o u s system (CNS) 33"45'46. N F s a r e 8 - 1 0 n m s t r u c t u r e s r u n n i n g l o n g i t u d i n a l l y a n d p a r a l l e l with e a c h o t h e r inside o f t h e axons t4. T h e y a r e c o m p o s e d o f t h r e e subunits that have a p p a r e n t m o l e c u l a r weights on S D S gels o f 68,000 ( N F - L ) , 145,000 ( N F - M ) a n d 200,000 ( N F - H ) 16'35'2°. T h e s e N F subunits are e n c o d e d by t h r e e d i f f e r e n t genes 32'4°'23'25'29'31. N F s a r e cross-link e d by n u m e r o u s s i d e - a r m s p r o j e c t i o n s a n d a p p e a r as a t h r e e - d i m e n s i o n a l I F matrix. D e c o r a t i o n e x p e r i m e n t s using a n t i b o d i e s ~5 a n d p a r t i a l p r o t e o l y s i s e x p e r i m e n t s 2~ r e v e a l e d t h a t N F p r o j e c t i o n s a r e c o m p o s e d o f the

N F - M a n d N F - H proteins. It has b e e n s u g g e s t e d that the highly c h a r g e d extensions o f t h e s e two p r o t e i n s 21 can c o n t r i b u t e to N F spacing a n d p e r h a p s to t h e control of t h e axonal c a l i b e r 9"t7'19'28. T h e n e u r o n - s p e c i f i c e x p r e s s i o n of N F p r o t e i n s is m a i n l y a t t r i b u t a b l e to t r a n s c r i p t i o n a l activation. T r a n scriptional r e g u l a t i o n of N F g e n e e x p r e s s i o n was first i n v e s t i g a t e d by t r a n s f e c t i o n e x p e r i m e n t s of N F g e n e s driven by t h e i r own p r o m o t e r 23'36'43 a n d o f hybrid c o n s t r u c t s c o n t a i n i n g t h e b a c t e r i a l /3-galactosidase ( l a c Z ) o r t h e luciferase r e p o r t e r g e n e s u n d e r the control o f N F p r o m o t e r s 41'49. In all o f t h e s e cell c u l t u r e e x p e r i m e n t s , e l e m e n t s for tissue-specific a n d high-level e x p r e s s i o n could not b e identified. By contrast, intact m o u s e or h u m a n N F - L g e n e s 24'37 w e r e correctly exp r e s s e d in t h e n e r v o u s system w h e n t e s t e d in transgenic mice. A d e l e t i o n - m u t a n t analysis o f the h u m a n N F - L g e n e in t r a n s g e n i c mice f u r t h e r i n d i c a t e d t h e p r e s e n c e of i n t r a g e n i c r e g u l a t o r y e l e m e n t s involved in t h e n e u r o n a l specificity 2. A n u m b e r of o b s e r v a t i o n s also suggest t h a t postt r a n s c r i p t i o n a l r e g u l a t i o n c o u l d be p a r t of the regula-

Correspondence: J.-P. Julien, Centre for Research in Neuroscience, McGill University, The Montreal General Hospital Research Institute, 1650

Cedar Avenue, Montr6al, Que., Canada H3G 1A4. Fax: (1) (514) 937-3532.

24 tory mechanisms governing the levels of NF proteins in neurons. First, after axotomy of the sciatic nerve, the ratios of the three NF-L proteins remain unchanged in the nerve ~8 even though there is a greater decrease of NF-L mRNA compared to NF-H and NF-M in motor neurons 39. Second, in undifferentiated embryonic carcinoma (EC) cells, significant levels of NF-L mRNAs are detectable albeit NF-L proteins are at exceedingly low levels 22. To study the regulation of NF gene expression, we have generated transgenic mice bearing multiple copies of intact NF genes, the human NF-L and NF-H genes. We report here, for the first time, evidence of different posttranscriptional controls regulating the levels of NF-L and NF-H expression in the CNS. Our results indicate that overexpression of NF-L proteins in the CNS is impeded by posttranscriptional controls occurring, at least in part, at the level of translation. In contrast, high level expression of human NF-H proteins in the nervous system of transgenic mice was not impeded by such posttranscriptional regulation. MATERIALS A N D METHODS Transgenic mice Microinjection of D N A fragments was performed as described by Brinster et al. 5. Generation and characterization of line 29 mice have been reported previously 24. Line 2 mice were obtained by microinjecting into fertilized mouse eggs a NotI linearized cosmid containing the h u m a n NF-H gene. This cosmid was isolated from a chromosome 22-enriched library. Extensive characterization of line 2 mice will be published elsewhere (C6t6, F., Lee, V.M.-Y., Houle, D. and Julien, J.-P., manuscript in preparation). Line 29 mice are homozygous for the h u m a n NF-L gene while line 2 mice are hemizygous for the h u m a n NF-H transgene. A cross between line 29 and line 2 mice generated doubly transgenic animals hemizygous for both genes. Normal mice are age-matched C3H mice (Charles River).

Northern blot analysis Total R N A s were prepared from nervous tissues of line 29, line 2 and normal mice. R N A extractions were performed according to Chomczynski and Sacchi 6. A quantity of 10 /zg of total R N A was loaded in each track of a 1% agarose-formaldehyde gel 8. D N A probes were labelled with [32PldATP (Amersham, 3000 C i / m m o l ) by random priming using deoxynucleotides and random hexanucleotides from Pharmacia. Blots were prehybridized for 2 h at 60°C in a solution containing 5 x S S C (1 x S S C is 150 m M NaCI and 15 m M sodium citrate, pH 7.0), 1% SDS, 10 m M Tris HCI pH 7.5, 5 x Denhardt's solution (1 x Denhardt's solution is 0.02% bovine serum albumin, 0.02% Ficoll and 0.02% polyvinylpyrrolidone), 5% dextran sulfate and 100 / z g / m l of denatured salmon sperm DNA. The radioactive D N A probes were added directly to the prehybridization solution. After an overnight incubation at 60°C, the filters were washed for 1 h in 0 . 3 x S S C and 0.5% SDS at 60°C. Filters were exposed on Kodak X-Omat A R films using intensifying screen.

Primer extension analysis Primer extension was performed on 20 /zg of total R N A from brain, thalamus, striatum, and spinal cord of transgenic mice and on brain of a normal control mouse. A 30-mer oligonucleotide homologous to both h u m a n and mouse NF-L was used as primer. This

oligonuceotide is located 189 and 185 nucleotides from the principal capping site of the mouse and h u m a n NF-L m R N A respectively. A quantity of 100 ng of primer was kinased in a buffer containing 50 m M Tris-HCI pH 7.4, 10 mM MgCI~, 5 m M DTT, 50 p~Ci of [32p]ATP (4500 C i / m m o l , A m e r s h a m ) and 11) U of T4 polynucleotide kinase (Pharmacia). After a 30 rain incubation at 37°C, the reaction mixture was passed through a G-50 column. The specific activity of the primer was of 2 x 108 c p m / / z g . R N A s and primers ( 6 x 105 cpm) were mixed together, heated at 85°C for 5 rain and cooled on ice. They were then precipitated and resuspended in 30/zl of annealing buffer containing 80% formamide, 400 mM NaCI, 5/) m M PIPES pH 6.6, l m M E D T A and 30 U of RNasin ribonuclease inhibitor (Promega). After an overnight incubation at 30°C, the annealing reactions were precipitated and the pellets were resuspended in 25 #1 of reverse transcription buffer containing 50 m M Tris-HC1 pH 8.3, 50 mM KCI, 5 m M MgCI 2, 5 m M DTT, 4 mM of dNTP mix (Pharmacia), 40 U of RNasin ribonuclease inhibitor (Promega) and 40 U of A M V reverse transcriptase (Pharmacia). After a 90 rain incubation at 42°C, the samples were heated at 75°C for 10 min, extracted with phenol-chloroform, precipitated and loaded on a 5% polyacrylamide sequencing gel. A sequencing reaction was run in parallel to determine the size of the bands. The dried gel was exposed for 7 days on a Kodak X-Omat A R film using intensifying screens.

S D S - PAGE analysis Total protein extracts were prepared by homogenizing directly the spinal cords in SDS sample buffer. A quantity of 40 mg of tissue was homogenized in 1 ml of buffer containing 15% glycerol, 2% SDS, 80 m M Tris-HCl pH 6.8, 5% /3-mercaptoethanol, and 0.01% bromophenol blue. After boiling the samples for 5 min, a volume of 5~zl of each sample was loaded on gel. Neurofilament (NF)-enriched preparation from mouse spinal cord and h u m a n brain were prepared as described previously 24. A quantity of 1 /~g of the NF-enriched preparation was loaded on gel. Insoluble extracts from brain and cerebellum were prepared by homogenization of the tissues in 1% Triton X-100, 10 m M sodium phosphate (pH 6.5), 100 m M NaCI and t m M EDTA. The insoluble fractions were obtained by centrifugation for 1 h at 27,00(lx g through 0.85 M sucrose containing the same buffer. A quantity of 5 /zg of the insoluble proteins was loaded on gel. Acrylamide protein gels of 7.5% were run according to Laemmli 27 using the mini-Protean System (Bio-Rad). Immune blotting Proteins fractionated on gels were transferred electrophoretically to nitrocellulose m e m b r a n e s that were then incubated either with a monoclonal antibody that recognizes both h u m a n and mouse NF-L protein (RPN.1105, 1:500 dilution, A m e r s h a m ) or with a monoclonal antibody recognizing the h u m a n but not the mouse NF-L protein (clone DP5-112, 1:2000 dilution, from Denise Paulin, Institut Pasteur, Paris). The primary antibodies were recognized by a peroxydase-linked anti-mouse Ig antibody (NXA 931, 1:2000 dilution, A m e r s h a m ) and the bands were visualized using the A m e r s h a m ECL detection kit. In uiuo labeling o] mouse spinal cords Anesthetized mice were injected in the spinal cord with 3 /zl of phosphate-buffered saline containing 250 p~Ci of [3SS]methionine ( > 800 C i / m m o l , A m e r s h a m ) and traces of Fast Green dye. After 5, 15 and 60 min, the spinal cords were dissected and 5 m m sections including the injection site were removed. Homogenization was performed in a volume of 250 p,l of SDS sample buffer containig 10 m M cold methionine per 10 mg of tissue. To evaluate the incorporated radioactivity, 10 #1 of the extracts were TCA-precipitated and counted. A quantity of 10 000 cpm was loaded on gel. After migration, the gel was stained with Coomassie blue, fixed in 30% methanol and 10% acetic acid and treated with E n H a n c e (New England Nuclear). The dried gel was then exposed on Kodak X-Omat A R autoradiography films.

25

/

Densitometric analysis The intensity of NF-L bands on Coomassie blue-stained gels were evaluated using the LKB Ultrascan densitometer. For the NF-enriched preparation of brain, cerebellum, spinal cord, and the total homogenates of spinal cord of both normal and transgenic mice, three mice were analysed and measurements of two dilutions of the samples were performed. The calculated ratio of transgenic to normal levels of NF-L protein was a mean value of the measurements. Autoradiograms of the Northern blotting and primer extension analysis were also subject to densitometric analysis. Many exposures of the results were analysed in order to avoid evaluating under or overexposed signals. The NF-L signals of ECL-revealed immunoblots were also evaluated by densitometry. Different dilutions of the samples were analysed in order to ensure the linearity of the detection.

T

N

T

./ N

T

N

3.8 kb -

2.4 kb -

-

3.5 kb

M

actin RESULTS

1 High level expression of human NF-L mRNAs in transgenic CNS W e g e n e r a t e d several lines o f t r a n s g e n i c mice which b e a r a 21.5 kb g e n o m i c f r a g m e n t e n c o m p a s s i n g the h u m a n N F - L g e n e f l a n k e d with 14 kb of 5' s e q u e n c e s a n d 3.2 kb o f 3' s e q u e n c e s a f t e r t h e first p o l y a d e n y l a tion site 23. I n a p r e v i o u s r e p o r t , we s h o w e d t h a t exp r e s s i o n o f this N F - L f r a g m e n t was r e s t r i c t e d to n e u rons a n d t h a t h u m a n N F - L p r o t e i n s w e r e a s s e m b l e d into N F s 24. T h e levels o f t r a n s g e n e m R N A s w e r e n o t p r o p o r t i o n a l to copy n u m b e r 24 b u t we o b t a i n e d a t r a n s g e n i c line (no. 29) in which levels o f h u m a n N F - L m R N A clearly e x c e e d e d t h o s e o f t h e e n d o g e n o u s N F - L m R N A in the C N S 24. F u r t h e r i n c r e a s e in t r a n s g e n e e x p r e s s i o n could b e a c h i e v e d in mice h o m o z y g o u s for t h e h u m a n N F - L t r a n s g e n e ( d a t a n o t shown). In t h e P N S o f mice f r o m line 29, h u m a n N F - L m R N A s w e r e a p p r o x i m a t e l y 30% o f t h o s e o f t h e e n d o g e n o u s N F - L m R N A s ( d a t a not shown). Two a p p r o a c h e s have b e e n c a r r i e d o u t to assess the levels of h u m a n a n d m o u s e N F - L m R N A s in t h e C N S o f line 29. First, t h e N F - L m R N A s w e r e a n a l y s e d on N o r t h e r n blots using 10 /.~g o f total R N A from brain, c e r e b e l l u m a n d spinal c o r d o f b o t h n o r m a l a n d transgenic mice (Fig. 1). A m o u s e PstI f r a g m e n t c o m p r i s e d within the first exon o f t h e m o u s e N F - L g e n e was u s e d as a p r o b e . H y b r i d i z a t i o n a n d w a s h e s w e r e p e r f o r m e d at low stringency to allow c r o s s - h y b r i d i z a t i o n of t h e m o u s e p r o b e with h u m a n N F - L m R N A s . B o t h m o u s e a n d h u m a n N F - L g e n e s give rise to two m R N A s of d i f f e r e n t lengths. T h e two t r a n s c r i p t s a r e o f 2.3 a n d 3.5 kb for t h e m o u s e 24'32 a n d o f 2.4 a n d 3.8 kb for h u m a n 7'23. In b o t h species, t h e l a r g e r t r a n s c r i p t arises from t h e selective use o f a d i f f e r e n t p o l y a d e n y l a t i o n site l o c a t e d 1.2 kb in m o u s e 41 a n d 1.4 kb in h u m a n 3 d o w n s t r e a m of t h e first p o l y a d e n y l a t i o n site. D e n s i t o m e t r i c analysis r e v e a l e d an i n c r e a s e o f 3.7 t i m e s o f total N F - L t r a n s c r i p t s in t h e b r a i n o f t r a n s g e n i c mice in c o m p a r i s o n with n o r m a l m o u s e (Fig. 1, l a n e s 1 a n d

2

3

4

5

6

Fig. 1. Detection by Northern blotting of NF-L mRNAs in nervous tissues of transgenic line 29. RNA samples (10 p.g) from brain (lanes 1 and 2), cerebellum (lanes 3 and 4) and spinal cord (lanes 5 and 6) were hybridized with a mouse PstI probe derived from the first exon of the mouse NF-L gene.

2). F o r the c e r e b e l l u m , a 4.3 time i n c r e a s e is o b s e r v e d (Fig. 1, lanes 3 a n d 4) while in spinal c o r d (Fig. 1, lanes 5 a n d 6), N F - L R N A i n c r e a s e s by a factor o f 2.4. A s a m o u s e N F - L p r o b e was u s e d for hybridization, t h e signals o b t a i n e d m u s t r e p r e s e n t u n d e r e s t i m a t e s o f t h e actual values. N o c h a n g e s in t h e levels o f the e n d o g e nous N F - L m R N A levels o c c u r r e d as a c o n s e q u e n c e of o v e r e x p r e s s i o n o f the h u m a n N F - L t r a n s g e n e as illust r a t e d by the d e t e c t i o n of the specific 3.5 kb m o u s e N F - L b a n d in t r a n s g e n i c a n d n o r m a l R N A samples. T h e s e c o n d a p p r o a c h to m o n i t o r t h e N F - L m R N A s was b a s e d on a p r i m e r e x t e n s i o n analysis using total R N A from n e r v o u s tissues of n o r m a l a n d t r a n s g e n i c mice (Fig. 2). T h e p r i m e r is a 3 0 - m e r o l i g o n u c l e o t i d e c o n s e r v e d in b o t h m o u s e a n d h u m a n N F - L genes. T h e s e q u e n c e is at 189 a n d 185 n u c l e o t i d e s , from the prin-

mouse -human

1

2

3

4

5

Fig. 2. Detection of mouse human and human mRNAs by primer extension assay. Total RNA (20 ~g) from control brain (lane 1) and from the brain (lane 2), thalamus (lane 3), striatum (lane 4) and spinal cord (lane 5) of a line 29 transgenic were annealed to a 30 mer oligonucleotide conserved in both human and mouse NF-L mRNA sequences. Primer extension was performed as described in Materials ans Methods. The band corresponding to the mouse transcript is of 189 nucleotides while the human band is of 185 nucleotides, as determined from a sequencing reaction run in parallel.

26 cipal capping site of the mouse and human genes respectively. Ratios of the human signal over the mouse signal were determined by densitometric analysis. In transgenic brain (Fig. 2, lane 2) there is a 5-fold increase in the levels of total NF-L mRNA. The increase is higher in the thalamus (Fig. 2, lane 3) with 9-fold more NF-L transcripts while in the striatum (Fig. 2, lane 4), a 4-fold increase is observed. For the spinal cord, the levels of NF-L mRNAs in transgenic animals exceeded by 3-fold those found in normal mice.

F----] RNA

¢/) "13

Proteins

!

!

- 4- Ii d. z

,~3- I --~ 2O LL

The let~els of NF-L protein do not correlate with mRNA lerels We first examined the levels of NF-L protein in Triton X-100 insoluble extracts from the brain, cerebellum and spinal cord of the normal and three line 29 mice. Both human and mouse protein comigrate on SDS polyacrylamide gel electrophoresis (Fig. 5A). Fig. 3 shows a S D S - P A G E analysis performed on tissues from normal and transgenic mice. For each tissue, the same amount of protein was loaded on gel. After electrophoretic migration, the gels were stained with Coomassie blue. As illustrated in Figs. 3 and 4, the levels of total NF-L proteins detected in the filamentenriched preparations from transgenic mice did not match the levels of NF-L mRNAs. Thus, in transgenic brain and cerebellum, densitometric analysis revealed

T

N

rlll~lllllA

Cerebellum

Spinal cord

Fig. 4. Quantitative analysis of NF-L mRNA and protein in CNS regions of transgenic line 29. Values for mRNAs and proteins were taken from the densitometric measurements of the signals detected by Northern blotting (cerebellum) or primer extension (brain and spinal cord) analysis and from the Coomassie blue-stained gels of the filament-enriched fractions.

that the increase in NF-L protein is approximately 10% for a 4- to 5-fold increase in mRNA. In the transgenic spinal cord, it is of 50% protein increase for a 3-fold increase in mRNA. Translation of extra NF-L mRNAs could have triggered the formation of a pool of unassembled NF-L proteins not detected in Triton X-100 insoluble protein

/

/

J T

~11111111111~

Brain

N

T

N --

H

-

M

i~i~!~i~~! ~ L

L-

L-

1

2

3

4

5

6

Fig. 3. S D S - P A G E analysis of filament-enriched preparations from various CNS regions. Transgenic (lanes 1, 3 and 5) and normal mouse (lanes 2, 4 and 6) samples (5 #g) from brain (lanes 1 and 2), cerebellum (lanes 3 and 4) and spinal cord (lanes 5 and 6) were loaded on a 7.5% polyacrylamide gel. After migration, the gel was stained with Coomassie blue.

27 fractions but present in total protein homogenates. Thus, spinal cords from normal and line 29 mice were h o m o g e n i z e d directly in loading buffer and the resulting total protein extracts were migrated on gel (Fig. 5B). A duplicate gel was transferred to a filter and reacted with a m o n o c l o n a l antibody (RPN.1105, A m e r sham) that recognizes both h u m a n and mouse N F - L proteins (Fig. 5A). Densitometric m e a s u r e m e n t s of the Coomassie blue-stained gel and of the immunoblot reacted with the antibody RPN.1105 reveal an increase of 50% in the levels of N F - L proteins in transgenic mice (Fig. 5B). This value is similar to the one obtained with filament-enriched preparations of the spinal cord, arguing against the accumulation of N F - L proteins in a pool of unassembled N F - L subunits. Discrepancies in the levels of N F - L m R N A and protein in transgenic mice could have resulted from interspecies differences b e t w e e n h u m a n and mouse N F - L m R N A translatability. A p o o r translatability of h u m a n N F - L m R N A s in transgenic mice would be reflected by a low ratio of h u m a n to mouse N F - L protein in nervous tissues. Because h u m a n and mouse

.~

A

N F - L are undistinguishable on gel (Fig. 5A), we used a monoclonal antibody ( D P 5 - 1 1 2 from Denise PaulinLevasseur, Institut Pasteur, Paris) specific for the hum a n N F - L protein (Fig. 5A) to estimate the proportion of h u m a n and mouse N F - L proteins in the spinal cord of transgenic mice. Total h o m o g e n a t e of transgenic spinal cord was fractionated on SDS gel along with NFs purified from h u m a n brain. Triplicate gels were run in parallel. T h e first gel was stained with Coomassie blue and the two others were i m m u n o b l o t t e d with either the RPN.1105 antibody or with the anti-human N F - L D P 5 - 1 1 2 antibody. Both samples contained the same a m o u n t of N F - L protein based on Coomassie blue staining and immunoblotting with the RPN.1105 antibody (Fig. 5C). Using the a n t i - h u m a n N F - L antibody, densitometric m e a s u r e m e n t s revealed a N F - L signal in the transgenic spinal cord sample corresponding to 80% of the signal detected in the h u m a n N F p r e p a r a t i o n (Fig. 5C). Thus, approximately 80% of total N F - L proteins in the spinal cord must be derived from the translation of h u m a n N F - L m R N A . This p r o p o r t i o n is compatible with the ratio of h u m a n and

.R~

B

T

C

N

~ii !i!iii! i~i+:ii~

H

M

-

H

~

! ii))

i~

-

H

M -

M i:i;iiiiiiiiii ¸¸

!i!i~ii ........

7111~¸~II~III!I~iiiiiiiiiill ~I ii¸ i!ii!iiiii~!iiii!i i!~ii!i!i!~i!i~i~i!i~!~i!~i iiiiii~iiiii!iiii~i iiii~iii~ii!!~ iii!~ili!i!iliili! ii~iiiilli i

L

-

q.,.~....,,

RPN.1105-

DP5-112

L

-

~

1

RPN.1105

-

-

L

~,,,,,,~

1

2

IIII~IIIU ii ¸

2

-

RPN.1105

-

~,,,~,mm~

DP5-112

-

,m~qBp

1

2

Fig. 5. Specific immunodetection of human NF-L protein in spinal cord homogenates of transgenic line 29. A: neurofilament-enriched preparation samples (1 p.g) from mouse brain (line 1) and human brain (lane 2) were loaded on 7.5% polyacrylamide triplicate gels. One gel was stained with Coomassie blue. The two others were immunoblotted with either the RPN-1105 monoclonal antibody (from Amersham) recognizing both human and mouse NF-L protein or with the human specific DP5-112 monoclonal antibody (from Denise Paulin, Institut Pasteur, Paris). B: total homogenates (5/xl) from line 29 transgenic (lane 1) and normal (lane 2) spinal cords were loaded on duplicate gels. The first gel was stained with Coomassie blue and the second one was immunoblotted with the RPN-1105 monoclonal antibody. C: a filament-enriched preparation from human brain (1 /xg, lane 1) and a total homogenate from line 29 transgenic spinal cord (5/zl, lane 2) were migrated on triplicate gels. The first gel was stained with Coomassie blue and the two others were immunoblotted either with the RPN-1105 monoclonal antibody or with the human-specific DP5-112 monoclonal antibody. Based on densitometric measurements, it is estimated that human NF-L contributes to approximately 80% of total NF-L in spinal cord of line 29.

28 mouse N F - L m R N A s indicating that b o t h m o u s e a n d h u m a n N F - L m R N A s are subject to similar control m e c h a n i s m s in the CNS.

High let,el NF-H expression is not impeded by posttranscriptional controls A transgenic m o u s e line b e a r i n g m u l t i p l e copies of the h u m a n N F - H gene (line 2) was g e n e r a t e d with a l i n e a r i z e d cosmid clone c o n t a i n i n g the c o m p l e t e human N F - H gene isolated from a c o s m i d library enriched for c h r o m o s o m e 22 sequences. T o t a l R N A from the spinal cords of t r a n s g e n i c and n o r m a l mice were hybridized on N o r t h e r n blots with a p r o b e s p a n n i n g the first exon of m o u s e N F - H . H y b r i d i z a t i o n a n d washes were p e r f o r m e d at low stringency in o r d e r to e v a l u a t e N F - H m R N A levels in the t r a n s g e n i c spinal cord. T h e h u m a n and mouse N F - H t r a n s c r i p t s are d i s t i n g u i s h a b l e on N o r t h e r n blots (Fig. 6A, lane 1). T h e c o m b i n e d N F - H m R N A signals in t r a n s g e n i c spinal cord exceed by 3 fold the n o r m a l level of N F - H m R N A . SDS p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s was perf o r m e d on total h o m o g e n a t e s of t r a n s g e n i c and n o r m a l spinal cords (Fig. 6B). D e n s i t o m e t r y scanning of C o o m a s s i e blue s t a i n e d gel r e v e a l e d that total N F - H p r o t e i n was o v e r e x p r e s s e d by 3-fold in t r a n s g e n i c mice as c o m p a r e d to n o r m a l mice (Fig. 6B). T h e h u m a n N F - H p r o t e i n is easily d i s t i n g u i s h a b l e on SDS gel with

A T

-

M

L _

-

L

1

2

3

Fig. 7. In vivo labeling of NF proteins after injection of [35S]methionine in spinal cords of normal, line 29, and doubly transgenic mice. Anesthetized animals were injected in the spinal cord with 250 #Ci of [~SS]methionine in 3 p-1 (Amersham, 800 Ci/mMol). After 15 rain of in vivo labeling, the mice were killed and a 5 mm portion of the spinal cord comprising the injection site was homogenized in sample buffer containing 10 mM cold methionine as described in Materials and Methods. Approximately 10 4 cpm were loaded in each lane. The position of NF proteins were determined from the Coomassie blue-stained gel.

N its e l e c r o p h o r e t i c mobility s u p e r i o r to e n d o g e n o u s m o u s e N F - H protein. A n excess of h u m a n N F - H p r o teins was also d e t e c t e d in brain, c e r e b e l l u m a n d b r a i n stem of these transgenics (C6t6, F., Collard, P., Houle, D. and Julien, J.-P., m a n u s c r i p t in p r e p a r a t i o n ) . Thus, in c o n t r a s t to the situation prevailing with the h u m a n N F - L t r a n s g e n e , the i n c r e a s e of N F - H m R N A s led to a c o r r e s p o n d i n g increase in proteins.

N -

mouse

H

__

-

M-

actin

M-

B T

human

H

-

Ecidence of a translational control specific for NF-L expression

L _

1

2

Fig. 6. High-level expression of both human NF-H mRNA and protein in transgenic mice. A: detection by Northern blotting of NF-H mRNA in total RNA (10 p,g) prepared from spinal cords of line 2 transgenic (lane 1) and normal mice (lane 2). The filters were hybridized with a mouse probe spanning the first exon of the NF-H gene at reduced stringency to detect the human NF-H mRNA. It is estimated that the levels of NF-H mRNAs are increased by at least 3-fold in the transgenic CNS. B: detection of NF-H proteins by SDS/PAGE in total homogenates of spinal cords (5 >1) from line 2 transgenic (lane 1) and normal mice (lane 2). The gel was stained with Coomassie blue.

W e sought to d e t e r m i n e the rate of N F - L synthesis in t r a n s g e n i c mice (line 29) o v e r e x p r e s s i n g N F - L m R N A in c o m p a r i s o n with n o r m a l mice. This was d o n e by m e a s u r i n g the i n c o r p o r a t i o n o f [35S]methionine injected into the spinal cord of a n e s t h e t i z e d mice. T h e mice were killed after e i t h e r 5, 15 or 60 min labeling p e r i o d a n d for e a c h mouse, a total h o m o g e n a t e of a 5 m m piece of the spinal cord c o m p r i s i n g the site of [35S]methionine injection was p r e p a r e d in s a m p l e buffer c o n t a i n i n g 10 m M cold m e t h i o n i n e . E q u a l a m o u n t of

29 counts were put on gel from each mouse sample. After migration, the gel was stained with Coomassie blue and treated for fluorography. The stained gel was used to mark the positions of the NF proteins on the fluorograms. After 15 min, no significant increase in NF-L synthesis was observed in line 29 mouse in comparison with the normal mouse (Fig. 7, lane 1 and 2), indicating comparable rate of NF-L synthesis in the two mice. Similar results were obtained with spinal cord samples labeled for 5 and 60 min with [35S]methionine (data not shown). An in vivo labeling experiment with [35S]methionine was also carried out with a doubly transgenic mouse bearing human NF-L and NF-H transgenes. This mouse was obtained by mating a mouse from line 29 with a mouse from line 2. After 15 min of [35S]methionine in vivo labeling, newly synthesized NF-H proteins appear as multiple phosphorylated variants with the upper band migrating as the heavily phosphorylated human NF-H protein (Fig. 7, lane 3). Clearly, the newly synthesized NF-H proteins in the doubly transgenic (lane 3) were more abundant than in samples from normal mice (lane 1) or in mice of line 29 (lane 2) while no significant change in NF-L protein was observed. These in vivo [35S]methionine labeling experiments suggest the existence of different translational controls for NF-L and NF-H expression in the CNS of transgenic mice. DISCUSSION The data reported in this paper demonstrate that the human NF-L and NF-H genes are subject to different posttranscriptional controls in the CNS of transgenic mice. Considerable increases in the levels of NF-L mRNAs, ranging from 3- to 9-fold, occurred in different regions of the CNS of transgenic mice bearing multiple copies of a human NF-L transgene. However, there was no corresponding increase in the NF-L protein content of enriched-filament preparations (Fig. 4) and in total homogenates of spinal cords (Fig. 5B). The discrepancies between the m R N A level and protein content cannot be explained by species differences in the efficiency of m R N A translation. Immunoblotting experiments with a specific anti-human NF-L antibody clearly showed that human NF-L mRNAs in the mouse spinal cord was translated to produce about 80% of the total NF-L protein (Fig. 5). Previous immunohistochemical results demonstrated expression of the human NF-L transgene in vast populations of CNS neurons in the mouse line 29 24. However, it is unclear yet to what extent the degree of overexpression of human NF-L transcripts differs from one cell type to another.

In contrast to the situation with NF-L, overexpression of NF-H was not impeded by such posttranscriptional controls. Thus, in transgenic mice bearing multiple copies of the human NF-H gene, high level expression of NF-H proteins could be detected in the spinal cord (Fig. 6B, lane 2), as well as in other regions of the CNS and PNS (data not shown). Moreover, in vivo [35S]methionine labeling of a doubly transgenic bearing the human NF-H and NF-L transgenes showed an elevated level of newly synthesized NF-H species while no such increases occurred for the NF-L protein (Fig. 7). These results provided evidence that the regulation of NF-L expression in the CNS must occur, at least in part, at the level of translation and they argue for the existence of different translational control mechanisms for NF-L and NF-H expression in CNS neurons. Those differences may be the reflect of distinct structural and functional properties of NF-L and NF-H subunits. NF-L forms the core of the filament structure while NF-H subunits form at the NF periphery highly charged side-arms projections. NF-L is the only subunit capable of self-assembly into 10 nm filaments in vitro 1e'34 while NF-H homopolymerisation generates only short rodlike structures 15. The existence of a specific translational control for NF-L expression could explain a number of previous observations. For example, in undifferentiated P19 EC ceils, significant levels of NF-L m R N A are expressed but NF-L proteins remain at exceedingly low levels 22. Another puzzling phenomenon is that after axotomy of the sciatic nerve, NF-L m R N A is more dramatically affected than the NF-M or the NF-H m R N A levels in motor neurons while the ratio of the three NF proteins remain unchanged. Our results with the intact human NF-L gene comply with the observation reported by Monteiro et al. 37 that over-expression of a mouse NF-L construct under the control of the murine sarcoma virus (MSV) long terminal repeat did not result in corresponding increase in total NF-L proteins in brain of transgenic mice. In this situation however, transgenic mice expressed fusion M S V / N F - L mRNAs in which about 70% of the 5' U T R of the mouse NF-L mRNA was swapped for MSV LTR. The pertinence of the results obtained with the M S V / L T R transgene has remained uncertain because an aberrant posttranscriptional regulation could have been triggered by the extensive modifications made in the 5' U T R of the NF-L mRNA. For instance, in this system, there was no tag in the coding sequences of the M S V / N F - L construct to distinguish the transgene products from the endogenous mouse NF-L gene. Our conclusion of posttranscriptional controls for NF-L expression does not appear to

30

be generalized to PNS neurons. In peripheral sensory neurons of transgenic mice bearing the M S V / N F - L construct 37, increased NF-L m R N A levels correlated with elevated NF-L content peripheral axons. At the present, one can only speculate about the nature of the control mechanisms involved in the regulation of NF-L translation. Indeed, mechanisms of translational control are still poorly understood (reviewed in Hershey l3). The best documented example of such mechanisms is probably the ferritin regulation by iron (reviewed in Klausner and Harford26; Theip7). All ferritin mRNAs have a conserved stem loop structure in their 5' UTR, the iron-responsive element, that is bound to a cytoplasmic translation repressor when iron is limiting 3°'44. Stem loop structures in mRNA have often been implicated in translational control (for examples, see Aziz and Munro l; Wulczyn and Kahmann4S). Computer analysis of both mouse and human NF-L m R N A secondary structure revealed conserved putative stem-loops downstream of the A U G that could perhaps play a role in the repression of NF-L m R N A translation. A well known example of translational control for a cytoskeletal protein is the autoregulation of /3-tubulin. Thus, it has been proposed that soluble tubulin subunits could destabilize translating tubulin mRNAs by interacting with the nascent proteins 11. However, there is no evidence of important soluble pools of NF proteins in neurons 38'4'42. Finally, one can imagine the existence of a positive regulator of NF-L m R N A translation that would be present in limiting amount in the cells. For example, in cells expressing low amounts of the general initiation factor eIF-4E, translation of ornithine aminotransferase m R N A is impaired, probably because of the low affinity of these mRNAs for the factor eIF-4E. Translation of mRNAs encoding the ornithine aminotransferase is facilitated by an increased concentration of the initiation factor eIF-4E following transfection of the cells with eIF4-E coding sequences l°. Experiments are now in progress to define the NF-L m R N A regions involved in the translational control of NF-L expression. The finding that NF-H expression is not subject to the same translational regulation than NF-L suggests the possibility to carry out swapping experiments between human NF-H and NF-L genes to define the m R N A regulatory sequences involved in the NF-L specific translational control.

Acknowledgements. This work was supported by the Medical Research Council of Canada (J.-P.J) and the Canadian Network of Center of Excellence for Neural Regeneration. L.B. was a recipient of a studentship from the MRC.

REFERENCES

1 Aziz, N. and Munro, H.N., Iron regulates ferritin mRNA translation through a segment of its 5' untranslated region, Proc. Natl. Acad. Sci. USA, 84 (1987) 8478-8482. 2 Beaudet, L., Charron, G., Houle, D., Tretjakoff, I., Peterson, A. and Julien, J.-P., Intragenic regulatory elements contribute to transcriptional control of the neurofilament light gene, Gene, 116 (1992) 205-214. 3 Beaudet, L. and Charron, G., Origin of the two mRNA species for the neurofilament light gene~ Biochem. Cell. Biol., (1992) 279-284. 4 Black, M.M., Keyser, P. and Sobel, l., Interval between the synthesis and assembly of cytoskeletal proteins in cultured neurons, J. Neurosci., 6 (1986) 1004-1012. 5 Brinster, R.L., Chen, H.Y., Trumbauer, M.E., Senear, A.W., Warren, R. and Palmiter, R.D., Somatic expression of herpes thymidine kinase in mice following injection of a fusion construct into eggs, Cell, 27 (1981) 223-231. 6 Chomczynski, P. and Sacchi, N., Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction, Anal. Biochem., 162 (19871 156-159. 7 Clark, A.W., Krekoski, C.A., Parhad, I.M., Liston, D., Julien, J.-P. and Hoar, D.I., Altered expression of genes for amyloid and cytoskeletal proteins in alzheimer cortex, Ann. Neurol., 25 (19891 331-339. 8 Davis, G.L., Dibner, M.D. and Battey, J.F., Formaldehyde Gel]br Electrophoretic Separation of RNA and Northern Blot, Elsevier, New York, 1986, pp. 143-146. 9 de Waegh, S.M., Lee, V.M.-Y. and Brady, S., Local modulation of neurofilament phosphory[ation, axonal caliber, and slow axonal transport by myelinating Schwann cells, Cell, 68 (1992) 451-468. 10 Fagan, R.J., Lazaris-Karatzas, A., Sonenberg, N. and Rozen, R., Translational control of ornithine aminotransferase. Modulation by initiation factor eIF-4E.J, Biol. Chem., 266 (1991) 16518-16523. 11 Gay, D.A., Sisodia, S.S. and Cleveland, D.W., Autoregulatory control of /3-tubulin mRNA is linked to translation elongation, Proc. Natl. Acad. Sci. USA, 86 (19891 5763-5767. 12 Geisler, N. and Weber, K., Self-assembly in vitro of the 68000 molecular weight component of the mammalian neurofilament triplet proteins into intermediate sized filaments, J. Mol. Biol., 151 (1981) 565-571. 13 Hershey, J.W.B., Translational control in mammalian cells, Annu. Ree. Biochem., 60 (1991) 717-755. 14 Hirokawa, N., The crosslinker system between neurofilaments, microtubules and membranous organelles in frog axons revealed by quick freeze, freeze-fracture, deep-etching method, Z Cell Biol., 94 (1982) 129-142. 15 Hisanaga, S. and Hirokawa, N., The structure of peripheral domains of the neurofilament revealed by low angle rotary shadowing, J. Mol. Biol., 202 (1988) 297-306. 16 Hoffman, P.N. and Lasek, R.J., The slow component of axonal transport. Identification of major structural polypeptides of the axon and their generality among mammalian neurons, J. Cell Biol., 66 (1975) 351-366. 17 Hoffman, P.N., Griffin, J.W. and Price, D.L., Control of axon caliber by neurofilament transport, J. Cell Biol., 99 (1984) 705714. 18 Hoffman, P.N., Thompson, G.W., Griffin, J.W. and Price, D.L., Changes in neurofilament transport coincide temporally with alterations in the caliber of axons in regenerating motor fibers, J. Cell. Biol., 101 (19851 1332-1334. 19 Hoffman, P.N., Cleveland, D.W., Griffin, J.W., Landes, P.W., Cowan, N.J. and Price, D.L., Neurofilament gene expression: A major determinant of axonal caliber, Proc. Natl. Acad. Sci. USA, 84 (1987) 3472-3476. 20 Julien, J. and Mushynski, W.E., Multiple phosphorylation sites in mammalian neurofilament polypeptides, J. Biol. Chem., 257 (1982) 10467-10470.

31 21 Julien, J. and Mushynski, W.E., The distribution of phosphorylation sites among identified proteolytic fragments of mammalian neurofilaments, J. Biol. Chem., 258 (1983) 4019-4025. 22 Julien, J.-P. and Grosveld, F., Structure and expression of neurofilament genes. In: The Neuronal Cytoskeleton, Wiley/Liss, 1991, pp. 215-231. 23 Julien, J.P., Grosveld, F., Yazdanbaksh, K., Flavell, D., Meijer, D. and Mushynski, W., The structure of a human neurofilament gene (NF-L): a unique exon-intron organization in the intermediate filament gene family, Biochim. Biophys. Acta, 909 (1987) 10-20. 24 Julien, J.-P., Tretjakoff, I., Beaudet, L. and Peterson, A., Expression and assembly of a human neurofilament protein in transgenic mice provide a novel neuronal marking system, Genes Dec., 1 (1987) 1085-1095. 25 Julien, J.-P., C6t6, F., Beaudet, L., Flavell, D., Grosveld, F. and Mushynski, W.E., Sequence and structure of the mouse gene coding for the largest neurofilament subunit, Gene, 68 (1988) 307-314. 26 Klausner, R.D. and Harford, J.B., Cis-trans models for post-transcriptional gene regulation, Science, 240 (1989) 870-872. 27 Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of bacterophage T4, Nature, 227 (1970) 680-685. 28 Lasek, R.J., Oblinger, M.M. and Drake, P.F., Molecular biology of neuronal geometry: expression of neurofilament genes influences axonal diameter, Cold Spring Harbor Symp. Quant. Biol., 48 (1983) 731-744. 29 Lees, J.F., Schneidman, P.S., Skuntz, S.F., Carden, M.J. and Lazzarini, R.A., The structure and organization of the human heavy neurofilament subunit (NF-H) and the gene encoding it, EMBO J., 7 (1988) 1947-1955. 30 Leibold, E.A. and Munro, H.N., Cytoplasmic protein binds in vitro to a highly conserved sequence in the 5' untranslated region of ferritin heavy- and light-subunit mRNAs, Proc. Natl. Acad. Sci. USA, 85 (1988) 2171-2175. 31 Levy, E., Liem, R.K.H., D'Eustachio, P. and Cowan, N.J., Stucture and evolutionary origin of the gene encoding mouse NF-M, the middle-molecular-mass neurofilament protein, Eur. J. Biochem., 166 (1987) 71-77. 32 Lewis, S.A. and Cowan, N.J., Anomalous placement of introns in a member of the intermediate filament multigene family: an evolutionary conundrum, Mol. Cell. Biol., 6 (1986) 1529-1534. 33 Liem, R.K.H., Neuronal intermediate filaments, Curr. Opin. Cell Biol., 2 (1990) 86-90. 34 Liem, R.K.H. and Hutchison, S.B., Purification of individual components of the neurofilament triplet: filament assembly from the 70000-Dalton subunit, Biochem., 21 (1982) 3221-3226. 35 Liem, R.K.H., Yen, S.H. and Shelanski, M.L., Intermediate filaments in nervous tissues, J. Cell BioL, 79 (1978) 637-645.

36 Monteiro, M.J. and Cleveland, D.W., Expression of NF-L and NF-M in fibroblasts reveals coassembly of neurofilament and vimentin subunits, J. Cell Biol., 108 (1989) 579-593. 37 Monteiro, M.J., Hoffman, P.N., Gearhart, J.D. and Cleveland, D.W., Expression of NF-L in both neuronal and nonneuronal cells of transgenic mice: increased neurofilament density in axons without affecting caliber, J. Cell Biol., 111 (1990) 1543-1557. 38 Morris, J.R. and Lasek, R.J., Stable polymers of the axonal cytoskeleton: the axoplasmic ghost, J. Cell Biol., 92 (1982) 192198. 39 Muma, N.A., Hoffman, P.N., Slunt, H., Applegate, M.D., Liederburg I. and Price, D.L., Alterations in levels of mRNAs coding for neurofilament protein subunits during regeneration, Exp. Neurol., 107 (1990) 230-235. 40 Myers, M.N., Lazzarini, R.A., Lee, V.M.Y., Schlaepfer, W.W. and Nelson, D.L., The human mid-size neurofilament subunit: a repeated protein sequence and the relationship of its gene to the intermediate filament family, EMBO J., 6 (1987) 1617-1626. 41 Nakahira, K., Ikenaka, K., Wada, K., Tamura, T., Furuichi, T. and Mikoshiba, K., Structure of the 68-kDa neurofilament gene and regulation of its expression, J. Biol. Chem., 265 (1990) 1978619791. 42 Nixon, R.A., Lewis, S.E., Dahl, D., Marotta, C.A. and Drager, C.U., Early postranslational modifications of the three neurofilament subunits in the mouse retinal cells: neuronal sites and time course in relation to subunit polymerization and axonal transport, Mol. Brain Res., 5 (1989) 93-108. 43 Pleasure, S.J., Lee, V.M.-Y. and Nelson, D.L., Site-specific phosphorylation of the middle molecular weight human neurofilament protein in transfected non-neuronal cells, J. Neurosci., 10 (1990) 2428-2437. 44 Rouault, T.A., Wright, M.W., Caughman, S., Harford, J.B. and Klausner, R.D., Binding of a cytosolic protein to the iron-responsive element of human ferritin messenger RNA, Science, 241 (1988) 1207-1210. 45 Shaw, G., Neurofilament proteins. In: The Neuronal Cytoskeleton, Wiley/Liss, 1991, pp. 185-214. 46 Steinert, P.M. and Liem, R.K.H., Intermediate filament dynamics, Cell, 60 (1990) 521-523. 47 Theil, E.C., Regulation of ferritin and transferrin receptors mRNAs, J. Biol. Chem., 265 (1990) 4771-4774. 48 Wulczyn, F.G. and Kahmann, R., Translational stimulation: RNA sequence and structure requirements for binding of Com protein, Cell, 65 (1991) 259-269. 49 Zopf, D., Dineva, B., Betz, H. and Gundelfinger, E.D., Isolation of the chicken middle-molecular weight neurofilament (NF-M) gene and characterization of its promoter, Nucl. Acids Res., 18 (1990) 521-529.