Increased expression and localization of the RNA-binding protein HuD ...

Available online at www.sciencedirect.com R

Experimental Neurology 183 (2003) 100 –108

www.elsevier.com/locate/yexnr

Increased expression and localization of the RNA-binding protein HuD and GAP-43 mRNA to cytoplasmic granules in DRG neurons during nerve regeneration夞 K.D. Anderson,1 M.A. Merhege, M. Morin, F. Bolognani, and N.I. Perrone-Bizzozero* Department of Neurosciences, University of New Mexico School of Medicine, Albuquerque, NM 87131-5223, USA Received 19 July 2002; revised 13 December 2002; accepted 8 January 2003

Abstract The neuronal-specific RNA-binding protein, HuD, binds to a U-rich regulatory element of the 3⬘ untranslated region (3⬘ UTR) of the GAP-43 mRNA and delays the onset of its degradation. We have recently shown that overexpression of HuD in embryonic rat cortical cells accelerated the time course of normal neurite outgrowth and resulted in a twofold increase in GAP-43 mRNA levels. Given this evidence, we sought to investigate the involvement of HuD during nerve regeneration. It is known that HuD protein and GAP-43 mRNA are expressed in the dorsal root ganglia (DRG) of adult rat and that GAP-43 is upregulated in DRG neurons during regeneration. In this study, we examined the expression patterns and levels of HuD and GAP-43 mRNA in DRG neurons following sciatic nerve injury using a combination of in situ hybridization, immunocytochemistry, and quantitative RT-PCR. GAP-43 and HuD expression increased in the ipsilateral DRG during the first 3 weeks of regeneration, with peak values seen at 7 days postcrush. At this time point, the levels of HuD and GAP-43 mRNAs in the ipsilateral DRG increased by twofold and sixfold, respectively, relative to the contralateral DRG. Not only were the temporal patterns of expression of HuD protein and GAP-43 mRNA similar, but also they were found to colocalize in the cytoplasm of DRG neurons. Moreover, both molecules were distributed in cytoplasmic granules containing ribosomal RNA. In conclusion, our results suggest that HuD is involved in the upregulation of GAP-43 expression observed at early stages of peripheral nerve regeneration. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Posttranscriptional regulation; ELAV; Hu proteins; RNA-binding protein

Introduction The role of GAP-43 during peripheral nervous system (PNS) regeneration has been studied extensively, yet the mechanisms controlling its expression remain unclear. The GAP-43 mRNA is localized to the soma of dorsal root ganglia (DRG) neurons, while GAP-43 protein undergoes fast axonal transport to the sciatic nerve (Tetzlaff et al., 1989, 1991; Tsai et al., 1997; Vander Zee et al., 1989). The expression of GAP-43 mRNA increases right after axo夞 Supplementary data associated with this article can be found at doi: 10.1016/S0014-4886(03)00103-1 * Corresponding author. Fax: ⫹1-505-272-8082. E-mail address: [email protected] (N.I. Perrone-Bizzozero). 1 Current address: Reeve-Irvine Research Center, University of California at Irvine, Irvine, CA 92697, USA.

tomy, peaks within the first few days postcrush, and slowly returns to normal levels after 4 weeks. GAP-43 protein levels also increase in the sciatic nerve after a crush injury, but lags behind the increase in its mRNA, with the peak being between 6 and 14 days postinjury (Vander Zee et al., 1989). GAP-43 mRNA is first increased in small-diameter (⬍30 ␮m) unmyelinated neurons in L5 DRG, and then later in larger-diameter (⬎50 ␮m) myelinated neurons (Sommervaille et al., 1991). Injury to the peripheral branch of the DRG results in upregulation of GAP-43 (Schreyer and Skene, 1993). However, the distance of a peripheral nerve injury from the DRG has no effect on the magnitude of GAP-43 expression in the cell bodies (Liabotis and Schreyer, 1995). Axotomy of the central branch of the L5 DRG does not induce an increase in GAP-43 mRNA in the DRG (Schreyer and Skene, 1993), but does induce an in-

0014-4886/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0014-4886(03)00103-1

K.D. Anderson et al. / Experimental Neurology 183 (2003) 100 –108

101

ton et al., 1998), but its function before and after nerve injury has not been studied. Given the known molecular interactions between the GAP-43 mRNA and HuD protein, in this study we sought to investigate the role of this RNA-binding protein in GAP-43 expression in vivo during nerve regeneration. In situ hybridization, RT-PCR, and immunocytochemistry were used to examine GAP-43 mRNA and HuD protein expression patterns in the L5 DRG after a crush injury. Our results indicate that the temporal and spatial pattern of HuD expression during sciatic nerve regeneration correlates with that of GAP-43, suggesting that HuD contributes to the posttranscriptional control of GAP-43 expression in vivo.

Materials and methods Surgery

Fig. 1. Expression of the GAP-43 mRNA in DRG neurons during sciatic nerve regeneration. DRGs from control animals (no crush) and from rats at 7 and 14 day postcrush were fixed with 4% paraformaldehyde (PFA) and sectioned at 20 ␮m on a cryostat. In situ hybridization was performed using an in vitro-transcribed digoxigenin-labeled antisense GAP-43 cRNA probe. Scale bar represents 50 ␮m.

crease in GAP-43 mRNA in the spinal motoneurons in the L5 spinal segment (Fernandes et al., 1999). It is thought that induction of GAP-43 after a sciatic nerve injury results from denervation from the peripheral target tissues, which normally provide a retrograde repressor (Benowitz and Perrone-Bizzozero, 1991; Karimi-Abdolrezaee and Schreyer, 2002). This process was shown to involve both transcriptional and posttranscriptional events (Namgung and Routtenberg, 2000; Vanselow et al., 1994). Yet, the precise mechanism by which GAP-43 induction is controlled during axonal regeneration is currently unknown. HuD is a neuronal-specific RNA-binding protein that is a member of the human ELAV-like/Hu family of RNA-binding proteins first discovered in Drosophila (Campos et al., 1985; Homyk et al., 1985; Robinow et al., 1988). Our laboratory has previously shown that HuD binds to a U-rich regulatory element in the 3⬘untranslated region (3⬘UTR) of the GAP-43 mRNA and increases the half-life of the transcript (Chung et al., 1997; Tsai et al., 1997). Overexpression of HuD in PC12 cells results in the stabilization of GAP-43 mRNA, followed by increased GAP-43 protein expression and the spontaneous formation of multiple neurites (Anderson et al., 2000; Mobarak et al., 2000). Furthermore, we have recently demonstrated that the overexpression of HuD in E19 rat cortical cells accelerated the time course of normal neurite outgrowth and resulted in a twofold increase in GAP-43 mRNA levels (Anderson et al., 2001). It is known that HuD is expressed in adult DRG neurons (Clay-

Male Sprague–Dawley rats weighing between 90 and 120 g were used for all experiments. Three animals were used per condition per time point. All surgeries were performed using standard sterile techniques and all protocols were approved by the UNM Animal Care and Use committee and conformed to NIH guidelines. Rats received a subcutaneous injection of Xylazine (10 mg/kg) followed by an intraperitoneal injection of Ketamine (60 –100 mg/kg). Surgery was performed according to De Koning et al. (1986). The skin was opened along a 1-cm length of the proximal half of the line between the trochanter major and the knee joint. The sciatic nerve was exposed where it emerged from beneath the m. gluteus maximus by blunt separation of the m. vastus lateralis and m. biceps femoris. An ink mark was made on the nerve at the point of emergence from the three muscles. Graefe iris forceps with a nontapering 1-mm-wide waffle-shaped mouth (Harvard Apparatus, Holliston, MA) were used to perform the crush. The exposed nerve was crushed maximally at the ink mark for 30 s. The muscle was pushed back together and the skin was closed using sterile wound clips. Upon completion of the procedure the animal was given a single dose of the analgesic buprenorphine (0.5 mg/kg). Animals were allowed to recover for 7, 14, or 21 days after the surgery. Animal behavior was observed everyday to verify the initial paralysis resulting from the crush injury and subsequent recovery. After surgery, animals were found to drag their injured hindlimb, but were able to move around their cages functionally. By 10 –14 days postcrush, animals could move their injured hindlimb and toes normally. Sham-operated animals did not exhibit any impairment after surgery. The success of the surgery and subsequent nerve regeneration was also verified microscopically, through staining of longitudinal sections of the nerve surrounding the lesion with neurofilament antibodies. As previously reported (Fernandes et al., 1999; Tetzlaff et al., 1991), we found that at 7 days postcrush NF-M protein levels were decreased distal

102


Fig. 2. Expression patterns of HuD protein in control and regenerating DRG neurons. No crush and 7 and 14 day postcrush DRG tissues were fixed with 4% PFA, frozen, and sectioned (20 ␮m). Fluorescent immunocytochemistry was performed with the ␣HuD antibody and a rhodamineconjugated secondary antibody. Scale bar represents 50 ␮m.

to the crush site; they began to increase by Day 14 and returned to control levels by Day 21 (data not shown).

sections were thawed briefly and postfixed for 10 min in 4% PFA in 0.1 M sodium phosphate buffer (pH 7.4). Prehybridization was for 1 h at 45°C in buffer containing 2⫻ SSC, 1% Denhardt’s reagent, 10% dextran sulfate, 0.5 mg/ml heparin, 0.5 mg/ml yeast tRNA, 0.25 mg/ml salmon sperm DNA, and 50% formamide. For hybridization, 200 – 500 ng of DiG-labeled probe in hybridization buffer was added to each slide and then the slide sealed with a HybriWell (Grace Bio-Labs, Bend, OR). Sections were incubated overnight at 55°C. Digoxigenin-labeled probes were detected by using the Genius kit according to the manufacturer’s protocol (Roche Molecular Biochemical, Indianapolis, IN) To visualize the levels and distribution of HuD protein in the cell, fluorescent immunocytochemistry was performed on adjacent sections. Sections were blocked for 15 min at room temperature in 2% BSA–PBS containing 0.3% Triton X-100 (PBS-T). HuD was detected with our polyclonal antibody ␣HuD (1:25 in PBS-T) overnight at 4°C (Anderson et al., 2000) and a goat anti-rabbit-TRITC (1:250 in PBS-T) secondary antibody for 30 min at 37°C and 30 min at room temperature. Sections were covered with glass slides using the PermaFluor aqueous mounting media (Immunon, Pittsburgh, PA) and dried overnight at 4°C in the dark. Images were taken on an Olympus (BX60) microscope at 200⫻ magnification using the Magnafire (IEEE 1394 Firewire) digital megapixel imaging camera (Optronics, Goleta, CA). Colocalization of HuD protein with the GAP-43 mRNA and ribosomal RNA (rRNA) by confocal microscopy

Perfusion After the appropriate survival times, animals to be used for in situ hybridization and immunocytochemistry studies were anesthetized with an overdose of pentobarbital (120 mg/kg) and transcardially perfused with ice-cold Procaine HCl (1 mg/ml) in saline followed by ice-cold 4% paraformaldehyde (PFA) in 0.1 M sodium phosphate buffer, pH 7.4. The sciatic nerve from each side of the body was carefully dissected out, and then a dorsal laminectomy was performed to remove both L5 DRGs. The tissues were postfixed for 4 h at 4°C in 4% PFA, rinsed for 1 h in 0.1 M sodium phosphate buffer, and then stored overnight at 4°C in 20% sucrose buffer. Tissues were frozen using 2-methylbutane cooled in liquid nitrogen and 20-␮m-thick sections were cut with a cryostat and thaw-mounted onto Vectabondcoated (Vector Labs, Burlingame, CA) glass slides. The sections were stored at ⫺80°C until further use. In situ hybridization and fluorescent immunocytochemistry For nonradioisotopic in situ hybridization, a GAP-43 cRNA probe was synthesized using digoxigenin (DiG)labeled UTP and in situ hybridization was performed as described by Paradies and Steward (1997). Frozen tissue

For colocalization studies of the GAP-43 mRNA and HuD protein, we used a combination of fluorescent in situ hybridization followed by fluorescent immunocytochemistry (FISH/FICC). RNA probes were synthesized in vitro in the presence of fluorescein-12-UTP (Roche Molecular Biochemicals, Indianapolis, IN) and used for in situ hybridization as described above. Briefly, 15 ng of probe was added to each slide and slides were incubated at 42°C overnight. Sections were then washed in 2⫻ SSC and 1⫻ SSC and unbound RNA probe was digested with 20 ␮g/ml RNase A. Finally, sections were washed with 1⫻ SCC for 30 min at 37°C. Following in situ hybridization, sections were processed for fluorescent immunocytochemistry using our polyclonal antibody ␣HuD as described above. Confocal images were captured on a Zeiss 510 laser scanning microscope (Carl Zeiss, Inc., Thornwood, NY) using a helium neon laser to excite the TRITC (543-nm excitation/560-nm emission) and an argon laser to excite the FITC (488-nm excitation/530-nm emission). Images were analyzed using the Zeiss LSM Image Browser software (Carl Zeiss, Jena, Germany). To examine whether HuD-containing cytoplasmic granules also contain rRNA, we performed double fluorescent immunocytochemistry using ␣HuD (1:25 in PBS-T) and the


103

Fig. 3. Increased expression of GAP-43 and HuD mRNA during PNS regeneration. RNAs were isolated from DRGs of control and Day 7 postcrush animals and used for semiquantitative RT-PCR as described under Materials and Methods. (A) Representative gels of RT-PCR products from control and 7-day postcrush tissues. All samples were analyzed in duplicate using 70 and 140 ng of RNA for GAP-43 and HuD and 3 and 6 ng of RNA for G3PD, respectively. (B) Plot of the relative levels of GAP-43 and HuD mRNA normalized to G3PD (*P ⬍ 0.05).

monoclonal antibody Y10B (1:250 in PBS-T) as described above. Y10B (Lerner et al., 1981) specifically recognizes a sequence in the 28S rRNA (Garden et al., 1995; Zheng et al., 2001). To visualize HuD and rRNA, cells were simultaneously incubated for 2 h at room temperature with both secondary antibodies, a goat anti-rabbit-TRITC and a goat anti-mouse-FITC antibody (both at 1:150 in PBS-T). DRG sections were scanned using a Zeiss 510 laser scanning confocal microscope as described above. RNA isolation and RT-PCR For RT-PCR studies, animals were sacrificed and the left and right L5 dorsal root ganglia were removed via a dorsal laminectomy and RNA was extracted from the tissue using the RNAqueous kit (Ambion, Austin, TX), according to manufacturer’s protocols. RNA samples were then treated with DNase I (1000 U/ml, Promega, Madison, WI) for 30 min at 37°C followed by phenol–

chloroform extraction and ethanol precipitation. RT-PCR was performed using the Titan One Tube RT-PCR system (Roche Molecular Biochemical, Indianapolis, IN) in the presence of 10 ␮Ci of [␣-32P]dCTP (NEN Life Science Products) as previously described (Mobarak et al., 2000). The following primers were used for GAP-43 (180 ng, 5⬘ GGAATAAGGATCCGAGGAGGAAA-GGAG 3⬘, 5⬘-CTTAAAGTTCAGGCATGTTCTTGGT 3⬘), HuD (180 ng, 5⬘TTGCTTAATATGGCCTATGGCGT 3⬘, 5⬘-ATAAGTAAGGGTGAGAAATTCAGG-3⬘), and glyceraldehyde-3-phosphate dehydrogenase (G3PD) (180 ng, 5⬘CCCACGGCAAGTTCAACGGCA-3⬘, 5⬘-TGGCAGGTTTCTCCAGGCGGC-3⬘). Reactions were performed in a Peltier Thermal cycler (PTC-200, MJ Research) using 70 and 140 ng of RNA for GAP-43 and HuD and 3 and 6 ng of RNA for G3PD and the following temperature profile: 45 min at 45°C, 2 min at 94°C, 24 cycles of 30 s at 94°C, 1 min at 58°C, and 2 min at 68°C, followed by a prolonged elongation for 7 min at 68°C. Aliquots containing 10 ␮l of each reaction were

104


the sciatic nerve was exposed but not crushed. At different time points, the expression patterns of GAP-43 mRNA and HuD mRNA and protein were examined by in situ hybridization and fluorescent immunocytochemistry. As shown in Fig. 1, low levels of the GAP-43 mRNA could be detected in control DRG neurons. GAP-43 expression increased in the ipsilateral DRG at 7 days postcrush and gradually decreased between 14 and 21 days after injury. In agreement with previous studies (Sommervaille et al., 1991), we found that GAP-43 mRNA expression first increased in smalldiameter neurons, then in larger diameter neurons. Similarly, low levels of HuD protein were observed in control DRGs (Fig. 2). During regeneration, the levels of this RNAbinding protein increased in the ipsilateral DRG at 7 days postcrush, remained elevated at 14 days postcrush, and decreased to control levels by Day 21 (data not shown). In the contralateral DRG, we found a small increase in the levels of HuD protein at 7 days after the injury (Fig. 2). Since similar levels of expression were observed in shamoperated animals (data not shown), it is likely that the increases in HuD protein seen in the contralateral DRG are the result of a systemic effect. Increased expression of GAP-43 and HuD mRNAs in regenerating DRG neurons

Fig. 4. Subcellular localization of the HuD protein and the GAP-43 and HuD mRNAs. Tissues from Day 7 postcrush DRG neurons were fixed in 4% PFA and processed for immunocytochemistry (A) and in situ hybridization (B). (A) Fluorescent immunocytochemistry was performed with the ␣HuD antibody and a rhodamine-conjugated secondary antibody. HuD protein was localized to fine granules throughout the cytoplasm. (B) In situ hybridization was performed separately for GAP-43 mRNA and HuD mRNA. Note the fine granular distribution of GAP-43 mRNA compared to the diffuse labeling of HuD mRNA. Scale bar represents 50 ␮m.

run on 5% polyacrylamide gels (Bio-Rad), and then stained for 30 min with ethidium bromide (0.5 ␮g/␮l in 0.18 M Tris base, 0.18 M boric acid, 4 mM EDTA, pH 8.3) to visualize the DNA ladder. Gels were dried for 45 min using a Bio-Rad Model 583 gel dryer, exposed to a phosphorimager screen (Molecular Dynamics Storm 860) for 24 h, and analyzed using the ImageQuant software package (Molecular Dynamics, Sunnyvale, CA).

Results Expression patterns of GAP-43 mRNA and HuD protein during sciatic nerve regeneration Young adult rats were subjected to a unilateral sciatic nerve crush and animals were allowed to recover for 7, 14, or 21 days. Sham operations were also performed in which

The levels of HuD and GAP-43 mRNA at different times postcrush were measured using RT-PCR in the presence of [32P]dCTP (Mobarak et al., 2000). RNAs isolated from L5 DRG pooled from three animals per condition and time point were used for cDNA synthesis and subsequent PCR. Relative levels of the RT-PCR products were obtained by normalizing the GAP-43 or HuD product to that of the housekeeping gene G3PD. Fig. 3A shows representative gels of RT-PCR products from control and 7 days postcrush. In the ipsilateral DRG, GAP-43 mRNA levels were increased 6-fold over baseline levels at 7 days postcrush (Fig. 3B). Also, we found an increase in GAP-43 mRNA levels in the contralateral DRG at 7 days postinjury, although not as pronounced as on the ipsilateral side (⬃2.5-fold). At 7 days after a crush injury, there was a 2-fold increase in HuD mRNA levels in the ipsilateral DRG. As seen with HuD protein (Fig. 2), we also found a small increase in the levels of HuD mRNA in the contralateral DRG. However, these increases were not specific for regeneration since they were also observed in sham-operated animals (data not shown). Statistical analysis using a two-tailed Student’s t test revealed that both HuD and GAP-43 mRNA levels were significantly increased in the ipsilateral DRG 7 days postcrush (P ⬍ 0.05). Colocalization of GAP-43 mRNA and HuD protein to rRNA-containing cytoplasmic granules At higher magnification, HuD was found to localize to granular material in the soma of regenerating DRG neurons


105

particularly in perinuclear granules, shown in yellow in the merged image (Fig. 6). There were additional areas of the cell in which these markers did not colocalize. For example, in agreement with the distribution of rRNA in the cell, the nucleolus was stained only with Y10B (see arrow in middle panel) and there were a few areas of the cell containing either green or red fluorescence. Additional images showing different confocal planes of the same neuron are shown online (see supplementary materials).

Discussion The RNA-binding protein HuD binds to a highly conserved regulatory element in the 3⬘ UTR of the GAP-43 mRNA. Previous work from our laboratory demonstrated that in neural cell lines and primary neuronal cultures overexpression of HuD results in the stabilization of the GAP-43 mRNA and concomitant increase in gene expression (for a review see (Perrone-Bizzozero and Bolognani, 2002)). The main goal of this study was to investigate whether similar mechanisms operate in vivo during peripheral nerve regeneration. The data presented here indicate that HuD gene Fig. 5. Colocalization of HuD and GAP-43 mRNA in cytoplasmic granules in regenerating DRG neurons. Sections from the ipsilateral DRG at 7 days postcrush were processed sequentially for FISH for GAP-43 mRNA and FICC for HuD protein as described under Materials and Methods. The distribution of both molecules was analyzed using confocal microscopy. As shown in the merged image, there is a significant codistribution of the GAP-43 mRNA and the HuD protein in the cell. Scale bar represents 50 ␮m.

(Fig. 4A). Analysis of multiple sections revealed that HuDcontaining granules were often concentrated near one pole of the cell and were absent from the nucleus. Similar to HuD protein, the GAP-43 mRNA appeared to localize in granules in the cytosol of regenerating DRG neurons (Fig. 4B). In contrast, the HuD mRNA was present diffusely throughout the cytoplasm. The results of Fig. 4 suggested that the GAP-43 mRNA and the HuD protein were colocalized in the cell. To confirm this idea, we used fluorescent in situ hybridization (FISH) and immunocytochemistry (FICC) in the same sections. As shown in Fig. 5, the GAP-43 mRNA and HuD protein showed a similar distribution in regenerating DRG neurons, with both molecules concentrated at one pole of the cell, presumably corresponding to the axon hillock. Analysis of the subcellular localization of other ELAVlike proteins showed that two of these proteins, Hel-N1 (Gao and Keene, 1996) and HuR (Gallouzi et al., 2000), are associated with polysomes. To study whether HuD-containing cytoplasmic granules also contained ribosomal RNA (rRNA), we used the Y10B monoclonal antibody that recognizes a specific sequence in the 28S rRNA (Lerner et al., 1981). A significant proportion of HuD protein staining in regenerating DRG neurons overlapped with that for Y10B,

Fig. 6. HuD-containing granules also contain ribosomal RNA. Additional sections from ipsilateral DRG neurons at 7 days postcrush were simultaneously stained for HuD and rRNA (Y10B epitope) using double FICC as described under Materials and Methods. Arrowhead in the middle panel shows the localization of the nucleolus. Bottom panel shows the colocalization of HuD and rRNA in perinuclear granules. Scale bar represents 50 ␮m. Additional images of different confocal planes from the same section are shown online (see movie in supplementary materials).

106


expression becomes elevated in the dorsal root ganglion after a sciatic nerve crush injury. This increase parallels the well-documented rise in GAP-43 expression during regeneration. In addition, both the HuD protein and the GAP-43 mRNA were found to localize to ribosome-containing granules throughout the cell body as regeneration occurred. The studies presented here are the first to demonstrate that the levels of HuD increase during regeneration in a temporal and spatial manner that is consistent with the role of this ELAV-like protein in the posttranscriptional control of GAP-43 expression. Using the contralateral DRG and the ganglia from shamoperated animals, we found a contralateral/systemic effect on GAP-43 and HuD expression at early times after crush injury. This finding is in agreement with previous studies on peripheral nerve regeneration. For example, Wells et al. (1994) observed a contralateral effect on nerve growth factor (NGF) expression in rat DRGs after a unilateral sciatic nerve crush. Between Days 1 and 4 postinjury there was a bilateral increase in NGF protein in lumbar and cervical ganglia, as well as in sham-operated animals. It was suggested that the contralateral increase in NGF expression was caused by systemic factors induced by stress or surgical trauma. Likewise, Lindå et al. (1992) found an increase in GAP-43 mRNA in contralateral motoneurons after a sciatic nerve injury. The authors postulated that the contralateral increase could be due to altered afferent input or to new functional demands of the spared leg. A similar contralateral effect has been documented for calcitonin gene-related peptide after PNS injury (Piehl et al., 1991). HuD is a member of the Hu family of RNA-binding proteins, which were first identified as the targets of autoantibodies from patients with paraneoplastic encephalomyelitis (Dalmau et al., 1990; Szabo et al., 1991). The proteins are homologous to Drosophila ELAV, an RNA-binding protein whose deletion results in an Embryonic Lethal Abnormal Vision phenotype in flies (Robinow et al., 1988). There are four mammalian ELAV/Hu proteins: HuR (also known as HuA), Hel-N1 (a.k.a, HuB), HuC, and HuD. In higher vertebrates HuB, HuC, and HuD are neuronal-specific, while HuR is expressed in other tissues. Hu proteins bind preferentially to AU-rich elements (ARE) found in the 3⬘UTR of specific mRNAs (Levine et al., 1993). Recent studies indicate that Hu proteins are involved in various aspects of mRNA regulation, from mRNA processing and stability to transport and translation (for a review see Perrone-Bizzozero and Bolognani, 2002). Using cell culture systems, different members of the ELAV-like/Hu protein family were shown to have a distinct subcellular distribution. For example, while HuR and HuC were observed primarily in the nucleus, HuB and HuD appear to be mostly cytosolic (Kasashima et al., 1999). Furthermore, the subcellular distribution of these proteins is rather dynamic, with HuR shuttling between the nucleus and the cytosol (Fan and Steitz, 1998) and the neuronal HuB, HuC, and HuD proteins being transported to axonal and dendritic processes during

differentiation (Antic and Keene, 1998; Aranda-Abreu et al., 1999; Aronov et al., 2002; Gao and Keene, 1996). Our results indicate that in DRG neurons HuD is primarily localized to the cytoplasm, where it is distributed in fine granular material. Furthermore, we found not only that HuD levels increased during regeneration, but also that the protein underwent a translocation to polysome-containing granules. Likewise, Hel-N1 was shown to localize to granular structures in the cytoplasm of embryonic cortical cells in culture and medulloblastoma cells (Antic and Keene, 1998; Gao and Keene, 1996). In these cells the protein was associated with messenger ribonucleoprotein particles (mRNPs), which are complexes composed of RNA and protein that serve to protect the RNA and enable a variety of posttranscriptional regulatory mechanisms to occur. Antic and Keene (1998) also demonstrated that Hel-N1-containing mRNPs could interact with polysomes via their association with cytoskeletal elements. The ubiquitously expressed ELAV-like protein HuR was also found to localize to distinct cytoplasmic granules, particularly after cells were exposed to a heat shock (Gallouzi et al., 2000). RNA granules similar to these were recently isolated from neuronal cell cultures (Krichevsky and Kosik, 2001). These granules also contain ribosomes and the RNA-binding protein Staufen and are present in dendrites, where they are thought to contribute to the control of local protein synthesis. Whether these dendritic and cytosolic granules constitute the same macromolecular entity or variations of similar mRNP particles is currently unknown. Nonetheless, it is becoming apparent that these granular structures could play an important role in posttranscriptional regulation of a number of neural genes. In this study, we have shown that during regeneration HuD is upregulated and translocated to fine granules where it may act to stabilize the GAP-43 mRNA. In support of this idea, we found that the GAP-43 mRNA, but not the HuD mRNA, was also distributed throughout the cytoplasm of regenerating DRG neurons within focal points of increased density. The finding that HuD protein and GAP-43 mRNA were localized in fine granules in the cell bodies in the DRG suggests that they may interact in vivo in a manner similar to which they interact in vitro in cultured cells (Anderson et al., 2000, 2001; Mobarak et al., 2000). The concomitant increase in GAP-43 and HuD expression by 7 days postinjury further supports the possibility for their interaction in vivo. It was already known that GAP-43 mRNA expression transiently increases in the DRG after a peripheral branch injury (Schreyer and Skene, 1993), presumably by the absence of a target-derived retrograde factor that represses GAP-43 expression (Andersen et al., 2000; Karimi-Abdolrezaee, 2002). Furthermore, both transcriptional and posttranscriptional events were shown to contribute to the induction of GAP-43 during peripheral nerve regeneration (Namgung and Routtenberg, 2000; Vanselow et al., 1994). In this regard, it is interesting to note that during regeneration of the goldfish optic nerve and the rodent facial nerve,


posttranscriptional mechanisms account for most of the accumulation of the mRNA (Perrone-Bizzozero et al., 1991; Namgung and Routtenberg, 2000). In view of these findings, our results suggest that HuD is involved in the posttranscriptional upregulation of GAP-43 expression within initial stages of peripheral nerve regeneration.

Acknowledgments This work was supported by the NIH (NS-30255) to N.P.B. The authors thank Dr. Jeffrey Twiss for providing the Y10B antibody used in these studies and Dr. Rebecca Lee for help with the use of the confocal microscope. The UNM-HSC Confocal Microscopy Facility was established with support from the NCRR (S10 RR14668 and P20 RR11830), NCI (R24 CA88339), and UNM Health Sciences Center.

References Andersen, P.L., Webber, C.A., Kimura, K.A., Schreyer, D.J., 2000. Cyclic AMP prevents an increase in GAP-43 but promotes neurite growth in cultured adult rat dorsal root ganglion neurons. Exp. Neurol. 166, 153–165. Anderson, K.D., Morin, M., Beckel-Mitchener, A., Mobarak, C., Neve, R.L., Furneaux, H.M., Burry, R., Perrone-Bizzozero, N.I., 2000. Overexpression of HuD, but not of its truncated form HuD I⫹II, promotes GAP-43 gene expression and neurite outgrowth in PC12 cells in the absence of NGF. J. Neurochem. 75, 1103–1114. Anderson, K.D., Sengupta, J., Morin, M., Neve, R.L., Valenzuela, C.F., Perrone-Bizzozero, N.I., 2001. Overexpression of HuD shortens the time course of neurite outgrowth and increases GAP-43 mRNA expression in embryonic rat cortical cells and embryonic stem cells in vitro. Exp. Neurol. 168, 250 –258. Antic, D., Keene, J.D., 1998. Messenger ribonucleoprotein complexes containing human ELAV proteins: interactions with cytoskeleton and translational apparatus. J. Cell Sci. 111, 183–197. Aranda-Abreu, G.E., Behar, L., Chung, S., Furneaux, H., Ginzburg, I., 1999. Embryonic lethal abnormal vision-like RNA-binding proteins regulate neurite outgrowth and tau expression in PC12 cells. J. Neurosci. 19, 6907– 6917. Aronov, S., Aranda, G., Behar, L., Ginzburg, I., 2002. Visualization of translated tau protein in the axons of neuronal P19 cells and characterization of tau RNP granules. J. Cell Sci. 115, 3817–3827. Benowitz, L.I., Perrone-Bizzozero, N.I., 1991. The expression of GAP-43 in relationship to neuronal growth and plasticity: when where, how and why? Prog. Brain Res. 89, 69 – 87. Campos, A.R., Grossman, D., White, K., 1985. Mutant alleles at the locus ELAV in Drosophila melanogaster lead to nervous system defects. A developmental-genetic analysis. J. Neurogen. 2, 197–218. Chung, S.M., Eckrich, M., Perrone-Bizzozero, N.I., Kohn, D.T., Furneaux, H.M., 1997. The ELAV-like proteins bind to a conserved regulatory element in the 3⬘ UTR of GAP-43 mRNA. J. Biol. Chem. 272, 6593– 6598. Clayton, G.H., Perez, G.M., Smith, R.L., Owens, G.C., 1998. Expression of mRNA for the ELAV-like neural-specific RNA binding protein, HuD, during nervous system development. Dev. Brain Res. 109, 271–280. Dalmau, J., Furneaux, H.M., Gralla, R.J., Kris, M.G., Posner, J.B., 1990. Detection of the anti-Hu antibody in the serum of patients with small lung cell cancer—a quantitative western blot analysis. Ann. Neurol. 27, 544 –552.

107

De Koning, P., Brakkee, J.H., Gispen, W.H., 1986. Methods for producing a reproducible crush in the sciatic and tibial nerve of the rat and rapid and precise testing of return of sensory function. J. Neurol. Sci. 74, 237–246. Fan, X.C., Steitz, J.A., 1998. Overexpression of HuR, a nuclear-cytoplasmic shuttling protein, increases the in vivo stability of ARE-containing mRNAs. EMBO J. 17, 3448 –3460. Fernandes, K.J., Fan, D.P., Tsui, B.J., Cassar, S.L., Tetzlaff, W., 1999. Influence of the axotomy to cell body distance in rat rubrospinal and spinal motoneurons: differential regulation of GAP-43, tubulins, and neurofilament-M. J. Comp. Neurol. 414, 495–510. Gao, F.B., Keene, J.D., 1996. Hel-N1/Hel-N2 proteins are bound to poly(A⫹) mRNA in granular RNP structures and are implicated in neuronal differentiation. J. Cell Sci. 109, 579 –589. Gallouzi, I.E., Brennan, C.M., Stenberg, M.G., Swanson, M.S., Eversole, A., Maizels, N., Steitz, J.A., 2000. HuR binding to cytoplasmic mRNA is perturbed by heat shock. Proc. Natl. Acad. Sci. USA 97, 3073–3078. Garden, G.A., Hartlage-Rubsamen, M., Rubel, E.W., Bothwell, M.A., 1995. Protein masking of a ribosomal RNA epitope is an early event in afferent deprivation-induced neuronal death. Mol. Cell. Neurosci. 6, 293–310. Homyk Jr., T., Isono, K., Pak, W.L., 1985. Developmental and physiological analysis of a conditional mutation affecting photoreceptor and optic lobe development in Drosophila melanogaster. J. Neurogen. 2, 309 –324. Karimi-Abdolrezaee, S., Schreyer, D.J., 2002. Retrograde repression of growth-associated protein-43 mRNA expression in rat cortical neurons. J. Neurosci. 22, 1816 –1822. Kasashima, K., Terashima, K., Yamamoto, K., Sakashita, E., Sakamoto, H., 1999. Cytoplasmic localization is required for the mammalian ELAV-like protein HuD to induce neuronal differentiation. Genes Cells 11, 667– 683. Krichevsky, A.M., Kosik, K.S., 2001. Neuronal RNA granules: a link between RNA localization and stimulation-dependent translation. Neuron 32, 12906 –12914. Lerner, E.A., Lerner, M.R., Janeway Jr., C.A., Steitz, J.A., 1981. Monoclonal antibodies to nucleic acid-containing cellular constituents: probes for molecular biology and autoimmune disease. Proc. Natl. Acad. Sci. USA 78, 2737–2741. Levine, T.D., Gao, F.-B., King, P.H., Andrews, L.G., Keene, J.D., 1993. Hel-N1: an autoimmune RNA-binding protein with specificity for 3⬘ uridylate-rich untranslated regions of growth factor mRNAs. Mol. Cell. Biol. 13, 3494 –3504. Liabotis, S., Schreyer, D.J., 1995. Magnitude of GAP-43 induction following peripheral axotomy of rat DRG neurons is independent of lesion distance. Exp. Neurol. 135, 28 –35. Lindå, H., Piehl, F., Dagerlind, Å., Verge, V.M.K., Arvidsson, U., Cullheim, S., Risling, M., Ulfhake, B., Ho¨ kfelt, T., 1992. Expression of GAP-43 mRNA in the adult mammalian spinal cord under normal conditions and after different types of lesions, with special reference to motorneurons. Exp. Brain Res. 91, 284 –295. Mobarak, C.D., Anderson, K.D., Morin, M., Beckel-Mitchener, A., Rogers, S.L., Furneaux, H., Perrone-Bizzozero, N.I., 2000. The RNA-binding protein HuD is required for GAP-43 mRNA stability, GAP-43 gene expression, and PKC-dependent neurite outgrowth in PC12 cells. Mol. Biol. Cell. 11, 3191–3203. Namgung, U., Routtenberg, A., 2000. Transcriptional and post-transcriptional regulation of a brain growth protein: regional differentiation and regeneration induction of GAP-43. Eur. J. Neurosci. 12, 3124 –3136. Paradies, M.A., Steward, O., 1997. Multiple subcellular mRNA distribution patterns in neurons: a nonisotopic in situ hybridization analysis. J. Neurobiol. 33, 473– 493. Perrone-Bizzozero, N.I., Bolognani, F., 2002. Role of HuD and other RNA-binding proteins in neural development and plasticity. J. Neurosci. Res. 68, 121–126. Perrone-Bizzozero, N.I., Neve, R.L., Irwin, N., Lewis, S., Fischer, I., Benowitz, L.I., 1991. Post-transcriptional regulation of GAP-43

108


mRNA levels during neuronal differentiation and nerve regeneration. Mol. Cell. Neurosci. 2, 402– 409. Piehl, F., Arvidsson, U., Johnson, H., Cullheim, S., Villar, M., Dagerlind, Å., Terenius, L., Ho¨ kfelt, T., Ulfhake, B., 1991. Calcitonin-generelated peptide (CGRP)-like immunoreactivity and CGRP mRNA in rat spinal cord motorneurons after different types of lesions. Eur. J. Neurosci. 3, 737–757. Robinow, S., Campos, A.R., Yao, K.-M., White, K., 1988. The ELAV gene product of Drosophila, required in neurons, has three RNP consensus motifs. Science 242, 1570 –1572. Schreyer, D.J., Skene, J.H.P., 1993. Injury-associated induction of GAP-43 expression displays axon branch specificity in rat dorsal root ganglion neurons. J. Neurobiol. 24, 959 –970. Sommervaille, T., Reynolds, M.L., Woolf, C.J., 1991. Time-dependent differences in the increase in GAP-43 expression in dorsal root ganglion cells after peripheral axotomy. Neuroscience 45, 213–220. Szabo, A., Dalmau, J., Manley, G., Rosenfeld, M., Wong, E., Henson, J., Posner, J.B., Furneaux, H.M., 1991. HuD, a paraneoplastic encephalomyelitis antigen, contains RNA-binding domains and is homologous to Elav and Sex-lethal. Cell 67, 325–333. Tetzlaff, W., Alexander, S.W., Miller, F.D., Bisby, M.A., 1991. Response of facial and rubrospinal neurons to axotomy: changes in mRNA expression for cytoskeletal proteins and GAP-43. J. Neurosci. 11, 2528 –2544.

Tetzlaff, W., Zwiers, H., Lederis, K., Cassar, L., Bisby, M.A., 1989. Axonal transport and localization of B-50/GAP-43 like immunoreactivity in regenerating sciatic and facial nerves of the rat. J. Neurosci. 9, 1303–1313. Tsai, K.-C., Cansino, V.C., Kohn, D.T., Neve, R.L., Perrone-Bizzozero, N.I., 1997. Post-transcriptional regulation of the GAP-43 gene by specific sequences in the 3⬘ UTR of the mRNA. J. Neurosci. 17, 1950 –1958. Van der Zee, C.E.E.M., Nielander, H.B., Vos, J.P., Lopes de Silva, S., Verhaagen, J., Oestreicher, A.B., Schrama, L.H., Schotman, P., Gispen, W.H., 1989. Expression of growth-associated protein B-50 (GAP43) in dorsal root ganglia and sciatic nerve during regenerative sprouting. J. Neurosci. 9, 3505–3512. Vanselow, J., Grabczyk, E., Ping, J., Baetscher, M., Teng, S., Fishman, M., 1994. GAP-43 transgenic mice: dispersed genomic sequences confer GAP-43-like expression pattern during development and regeneration. J. Neurosci. 14, 499 –510. Wells, M.R., Vaidya, U., Schwartz, J.P., 1994. Bilateral phasic increases in dorsal root ganglia nerve growth factor synthesis after unilateral sciatic nerve crush. Exp. Brain Res. 101, 53–58. Zheng, J.Q., Kelly, T.K., Chang, B., Ryazantsev, S., Rajasekaran, A.K., Martin, K.C., Twiss, J.L., 2001. A functional role for intra-axonal protein synthesis during axonal regeneration from adult sensory neurons. J. Neurosci. 21, 9291–9303.