SEARCH

SEARCH BY CITATION

Keywords:

  • cellular prion;
  • development;
  • glia;
  • neuron;
  • stress-inducible protein-1 and laminin

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The functions of cellular prion protein (PrPC) are under intense debate and PrPC loss of function has been implicated in the pathology of prion diseases. Neuronal PrPC engagement with stress-inducible protein-1 and laminin (LN) plays a key role in cell survival and differentiation. The present study evaluated whether PrPC expression in astrocytes modulates neuron-glia cross-talk that underlies neuronal survival and differentiation. Astrocytes from wild-type mice promoted a higher level neuritogenesis than astrocytes obtained from PrPC-null animals. Remarkably, neuritogenesis was greatly diminished in co-cultures combining PrPC-null astrocytes and neurons. LN secreted and deposited at the extracellular matrix by wild-type astrocytes presented a fibrillary pattern and was permissive for neuritogenesis. Conversely, LN coming from PrPC-null astrocytes displayed a punctate distribution, and did not support neuronal differentiation. Additionally, secreted soluble factors from PrPC-null astrocytes promoted lower levels of neuronal survival than those secreted by wild-type astrocytes. PrPC and stress-inducible protein-1 were characterized as soluble molecules secreted by astrocytes which participate in neuronal survival. Taken together, these data indicate that PrPC expression in astrocytes is critical for sustaining cell-to-cell interactions, the organization of the extracellular matrix, and the secretion of soluble factors, all of which are essential events for neuronal differentiation and survival.

Abbreviations used
CM

conditioned media

DMEM

Dulbecco’s minimum essential medium

E17

embryonic day 17

ECM

extracellular matrix

FCS

fetal calf serum

GFAP

glial fibrillary acid protein

LN

laminin

NCAM

neural cell adhesion molecule

NF200

neurofilament 200

PrPC

cellular prion protein

PrPSc

prion scrapie

recSTI1

recombinant STI1

STI1

stress-inducible protein-1

Prions are responsible for transmissible spongiform encephalopathies. The mechanism of disease propagation involves the interaction of prion scrapie (PrPSc) with its cellular isoform, cellular prion protein (PrPC), and the subsequent abnormal structural conversion of the latter (Prusiner et al. 1998). The role of PrPC in transmissible spongiform encephalopathies pathology is viewed as a gain-of-function transformation because of its conversion to the neurotoxic isoform PrPSc (Cohen 1999). Nonetheless, PrPC has been shown to be involved in several cellular functions indicating that its loss-of-function may be an important component for the pathogenesis of prion diseases (Samaia and Brentani 1998; Hetz et al. 2003; Westergard et al. 2007).

Cellular prion protein’s strong binding affinity for Cu2+ and its frequent localization in synaptosomal fractions suggest that it may be involved in copper metabolism (Brown et al. 1997a), protection against oxidative stress (Brown et al. 1997b), and that it may serve as a copper buffer in the synaptic terminal (Kretzschmar et al. 2000). PrPC association with several cellular proteins further suggests that it may be part of a multiprotein complex (Martins et al. 2002; Lee et al. 2003). We previously demonstrated that PrPC acts as a specific receptor for the C-terminal domain of the laminin (LN) γ-1 chain, which mediates neuronal adhesion and neuritogenesis (Graner et al. 2000a,b). PrPC has also been shown to interact with the 66 kDa protein stress-inducible protein 1 (STI1) (Martins et al. 1997; Zanata et al. 2002). The interaction between PrPC and STI1-induces activation of PKA and ERK1/2, with the former interaction promoting neuroprotection and rescuing undifferentiated post-mitotic retinal cells from apoptosis (Chiarini et al. 2002; Zanata et al. 2002). PrPC–STI1 engagement was recently shown to induce neuroprotection and neuritogenesis in hippocampal neurons through PKA and ERK1/2 pathways, respectively (Lopes et al. 2005). Neural cell adhesion molecule (NCAM) has also been shown to be associated with PrPC at the cell surface (Schmitt-Ulms et al. 2001), and this NCAM–PrPC interaction has been demonstrated to be involved in neurite outgrowth (Santuccione et al. 2005).

Both neurons and astrocytes can sustain prion propagation (Cronier et al. 2004), but the role of infected astrocytes in neurodegeneration remains controversial (Mallucci et al. 2003; Jeffrey et al. 2004). Astrocytes play a critical role in normal brain development, neuronal plasticity, axonal conduction, synaptic transmission throughout adult life, and neuronal survival particularly following injury. These functions are mediated by astrocyte–neuron homophilic and/or heterophilic interactions, as well as by the extracellular matrix (ECM) and secreted factors produced by astrocytes (Araque et al. 2001; Gomes et al. 2001; Sykova 2001; Fields and Stevens-Graham 2002). Therefore, it is possible that PrPC loss-of-function in astrocytes may also affect neuronal networking and survival.

To date experiments evaluating PrPC function have been conducted only on neurons, while the importance of PrPC expression in astrocytes has been largely ignored. Therefore, in the present study we examined PrPC function using chimeric neuron–astrocyte cultures and astrocyte secreted factors. Our study is relevant for elucidating the role of astrocyte PrPC expression in regulation of cell-to-cell interactions, secretion of soluble factors, and organization of the ECM processes that are necessary for neuronal survival and differentiation.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

All culture media components were obtained from Invitrogen-Life Technologies (Carlsbad, CA, USA). Recombinant STI1 (recSTI1), His6–STI1 and recombinant PrPC, His6–PrPC were purified as previously described (Zanata et al. 2002). Polyclonal anti-STI1 antibody raised in rabbits (Zanata et al. 2002) was produced by Bethyl (Montgomery, TX, USA). Monoclonal antibody anti-PrPC, 6H4 was acquired from Prionics (Schlieren, Zurich, Switzerland) and a polyclonal antibody against recombinant PrPC was produced in Prnp0/0 mice (Chiarini et al. 2002). Polyclonal antibodies against LN 1, purified from Engelbreth-Holm-Swarm sarcoma (EHS), were raised in rabbits as previously described (Lopes et al. 1985). Polyclonal anti-neurofilament 200 (NF200), anti-actin, secondary antibody conjugated with FITC (goat anti-rabbit), poly-l-lysine, bovine serum albumin, 3,3′-diaminobenzidine tetrahydrochloride, 4′-6-diamino-2-phenylindole, aprotinin, and leupeptin were purchased from Sigma (Saint Louis, MO, USA). Mouse anti-βIII-tubulin antibody was acquired from Chemicon (Temecula, CA, USA), and rabbit anti-CDK-4 antibody from Santa Cruz (Santa Cruz, CA, USA). Rabbit anti-glial fibrillary acid protein (GFAP) antibody, Secondary antibodies conjugated with peroxidase and Faramount Aqueous Mounting Medium were purchased from Dako (Carpinteria, CA, USA). Fluoromount was purchased from Southern Biotech (Birmingham, AL, USA) and Protein A Sepharose from Amersham Biosciences (Little Chalfont, Buckinghamshire, UK).

Animals

The ‘Principles of laboratory animal care’ (NIH publication No. 85–23, revised 1996) were strictly followed in all experiments. PrPC-null mice (Prnp0/0) were provided by Dr Charles Weissmann (Scripps Florida, FL, USA). The animals are descendants from the ZrchI line (Bueler et al. 1992). Wild-type animals (Prnp+/+) were generated by continuously crossing descendants from an initial 129/Sv and C57BL/6J mating. A second line of PrPC-null mice (Prnp−/−) provided by Drs Bruce Chesebro and Richard Race (Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, MT, USA) descended from the Edbg line (Manson et al. 1994). The Prnp−/− mice were backcrossed to C57BL/10 mice for at least four generations at Dr Bruce Chese-Bro’s laboratory and for four more generations in our laboratory. Heterozygous animals were mated and homozygous F1 descendants from the same littermate were crossed to generate PrPC-null (Prnp−/−) embryos or their respective wild-type controls, herein designated Prnpwt/wt. All of the adult animals used to generate Prnp−/− and Prnpwt/wt embryos were genotyped by PCR. DNA extracted from tail tips was amplified in two multiplex reactions. The first one amplified the neomycin gene using primers 5′-TTGAGCCTGGCGAACAGTTC-3′ and 5′-GATGGATTGCACGCAGGTTC-3′ and the retinoblastoma gene using primers 5′-AATAGAGGCACTCCCTTCAC-3′ and 5′-GGTAAGCCCTTGACCTAAAA-3′. The second reaction amplified the PrP gene using primers 5′-AACCGTTACCCACCTCAGGGT-3′ and 5′-GCGCTCCATCATCTTCACA-3′ and the retinoblastoma gene with the primers described above. The cycling conditions employed to amplify the neomycin gene were 94°C for 5 min, followed by 35 cycles of 94°C for 1 min, 57°C for 1 min, and 72°C for 45 s, followed by a final extension of 72°C for 5 min. The cycling conditions to amplify the PrP gene were 35 cycles of 94°C for 1 min and 60°C for 1 min and 72°C for 45 s followed by a final 5-min extension step at 72°C.

Cell cultures

Astrocyte primary cultures were prepared as previously described (Lima et al. 1997) from the cerebral hemispheres of embryonic day 17 (E17) wild-type, and PrPC-null mice. Briefly, single cell suspensions were obtained by dissociating cells of cerebral hemispheres in Dulbecco’s minimum essential medium (DMEM) supplemented with glucose (33 mmol/L), glutamine (2 mmol/L), penicillin/streptomycin (100 IU/100 μg/mL), and sodium bicarbonate (3 mmol/L). 3 × 105 cells were plated onto coverslips (13 mm) pre-coated with poly-l-lysine (5 μg/mL) and grown in DMEM enriched with 10% fetal calf serum (FCS). Cultures were incubated at 37°C in a humidified 5% CO2–95% air atmosphere with medium changed every 2 days until confluence. The astrocyte primary cultures were assessed by immunocytochemistry using anti-GFAP antibody and presented more than 95% purity.

Neuron primary cultures were prepared from the cerebral hemispheres of E17 wild-type and PrPC-null mice, as previously described (Gomes et al. 1999). Briefly, single cell suspensions were obtained by dissociating cells of cerebral hemispheres in DMEM with the same supplementation as described for astrocytes without FCS. Aliquots of 5 × 104cells were plated onto coverslips (13 mm) previously coated with poly-l-lysine (5 μg/mL) or over astrocyte isolated ECM. Neurons were maintained in DMEM without FCS or astrocyte conditioned media (CM) for 24 h at 37°C in a humidified 5% CO2–95% air atmosphere. In some experiments, 0.1 μmol/L recSTI1 or recombinant PrPC (0.1 or 1 μmol/L) were added to the culture medium. The neuronal primary cultures were assessed by immunocytochemistry using anti-NF200 or anti-βIII-tubulin antibodies and presented purity higher than 95%.

The ECM was obtained from astrocytes grown for 2 days to confluence. Cells were washed five times with phosphate-buffered saline at 37°C, submitted to an osmotic chock with ZnCl2 (1 mmol/L) for 30 min at 37°C and gently removed by washing five times with water (Niederreiter et al. 1994; modified). Neurons were plated over astrocyte ECM-containing coverslips. To obtain CM, astrocytes were growth in DMEM + 10% FCS until confluence, washed three times with DMEM and maintained in FCS-free DMEM for 48 h (Lima et al. 1997). Following this, astrocyte CM was collected, filtered and used in neuronal cultures.

Neuron–astrocyte co-cultures were prepared from cerebral hemispheres (Garcia-Abreu et al. 1995) of E17 wild-type and PrPC-null mice. Astrocytes grown to confluence were maintained in DMEM without serum for 48 h. Then 5× 104 neurons were plated onto an astrocyte monolayer and co-cultures maintained for 24 h at 37°C in a humidified 5% CO2–95% air atmosphere. In some experiments, monoclonal antibody anti-PrPC, 6H4, (8 μg/mL) or non-immune mouse IgG (8 μg/mL) were added to the culture medium.

Survival and neuritogenesis analysis of neuronal cells

Neurons cultured in the different conditions described above and stained for either NF200 or βIII-tubulin were examined with an inverted Olympus optical microscope (magnification 200×). The number of surviving cells was estimated in each coverslip and morphological analysis was used to assess cell integrity. Neuritogenesis was defined as the percentage of cells bearing at least one neurite three times longer than the diameter of the cell body. The number of surviving and neurite-bearing cells was quantified in 16 microscopic fields covering the diameter of the coverslip surface (33% of the available area). The counting was procedure only in neurons where the entire cell body was inside of the field. When neurons were touching the sides of the rectangular eyepiece only two sides (top and left) were counted. For neuritogenesis analysis only neurites in which the cell body was in the field or touching the pre-established side of the rectangular eyepiece were counted. Figure values represent the mean value of at least four independent experiments (with three determinants in sister wells per experiment). The bars represent SD. Statistical analyses were performed using anova followed by the Tukey–Kramer multiple comparisons test. A value of p < 0.05 was considered significant in all cases.

Immunocytochemistry and immunofluorescence

Cultured cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% triton X-100 for 3 min at 24°C (Lima et al. 2001). Cells were blocked with 5% bovine serum albumin in phosphate-buffered saline for 1 h and subsequently incubated with rabbit anti-NF200 (0.25 mg/mL), mouse anti-βIII-tubulin (1 : 400) or rabbit anti-GFAP (1 : 100) diluted in blocking solution, overnight at 4°C. Cells were then washed and incubated with secondary antibodies (goat anti-rabbit or anti-mouse; Dako Envision Labeled Polymer) for 1 h at 24°C, and then visualized using 0.3 μg/mL 3,3′-diaminobenzidine tetrahydrochloride. After staining, cells were counter-stained with hematoxylin and coverslips were mounted directly in Faramount Aqueous Mounting Medium and visualized using an inverted Olympus IMT2-NIC microscope. Negative controls were performed with non-immune rabbit or mouse IgG.

Immunofluorescence for LN matrix was performed without cell permeabilization using rabbit anti-LN 1 antibody (2 μg/mL) (Lopes et al. 1985) or rabbit anti-LN 1 antibody from Sigma (1 : 50) followed by secondary antibody conjugated to FITC (goat anti-rabbit IgG, 1 : 250). After immunolabeling cells were permeabilized and nuclei were counter-stained with 4′-6-diamino-2-phenylindole. Coverslips were mounted directly in Fluoromount and cells were imaged with a Bio-Rad Radiance 2100 Laser Scanning Confocal System (Bio-Rad, Hercules, CA, USA) coupled to a Nikon microscope (TE2000-U) running the software Laser Sharp 3.0. Image processing was performed with Photoshop (Adobe System).

Immunoblot analyses

Cell extract proteins were resolved on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes according to standard procedures. For LN expression analyses, cells were lysed with 10 mmol/L Tris–HCl, pH 7.4, 1% Triton X-100, 1% sodium deoxycholate, 1 mmol/L phenylmethylsulfonyl fluoride, 2 μg/mL aprotinin, and 2 μg/mL leupeptin. The cell extracts were centrifuged for 10 min at 4°C. Soluble proteins (supernatant) represent the intracellular pool, whereas insoluble material (pellet) contains essentially ECM-associated proteins (Parry et al. 1985; Chammas et al. 1994). Astrocyte CM was filtered to remove cell debris and concentrated 200-fold in an Amicon apparatus (Amicon, Houston, TX, USA) before electrophoresis. Western blotting assays were conducted using rabbit anti-LN 1 (0.4 μg/mL) (Lopes et al. 1985), rabbit anti-actin (2 μg/mL), rabbit anti-STI1 (0.5 μg/mL), mouse anti-PrPC (1 : 1000) (Chiarini et al. 2002), or rabbit anti-CDK-4 (1 : 1000) antibodies. Rabbit non-immune purified IgG or mouse pre-immune serum were used as negative controls.

Immunodepletion assay

Astrocyte–CM was incubated with rabbit anti-STI1 antibody (IgG, 4 μg/mL) or mouse anti-PrPC (IgG, 0.8 μg/mL) (Chiarini et al. 2002) overnight at 4°C, mixed with Protein A Sepharose for 2 h at 4°C and then, centrifuged. Pellets (washed three times) and supernatants were analyzed for the presence of STI1 and PrPC. CM depleted of STI1 or PrPC was filtered and used to culture neurons.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

It is generally accepted that differing genetic backgrounds could account for many spurious results when one is dealing with transgenic mice. To address this issue, most of the experiments conducted here were performed in two independent lines of PrPC-null mice: ZrchI (Bueler et al. 1992) and Edbg (Manson et al. 1994) animals designated herein as Prnp0/0 and Prnp−/−, respectively. These animals were obtained by different gene deletion approaches and have different genetic backgrounds. Therefore, this experimental design strongly ensures that alterations observed in PrPC-null mice cells when compared with their respective wild-type controls, Prnp+/+ for ZrchI and Prnpwt/wt for Edbg.

PrPC expression in neurons and astrocytes is important for neuronal differentiation

Neurons from E17 Prnp+/+ or Prnp0/0 mice were plated over confluent monolayers of E17 Prnp+/+ or Prnp0/0 astrocytes and maintained in serum-free medium. Under these conditions, cell–cell interactions, secreted trophic factors, and the ECM support neuronal development. We observed that 24 h after co-culture, neuronal survival (average number of neurons/per field) was similar for all experimental conditions and thus, independent of the PrPC status in both neurons and astrocytes (Fig. 1a). The same results were obtained when co-cultures from Prnp−/− embryos and their respective wild-type controls Prnpwt/wt were performed (Fig. 1b). Furthermore, neuronal survival was not affected by addition of anti-PrPC (α-PrPC) antibody 6H4 to these co-cultures (Fig. 1b).

image

Figure 1.  Cellular prion protein (PrPC) expression in neurons and astrocytes in co-culture is important for neuronal differentiation but does not affect neuronal survival. Neuronal survival and differentiation were evaluated in neuron-astrocyte co-cultures from wild-type and PrPC-null mice. Cells were immunolabeled for βIII-tubulin followed by immunoperoxidase staining and hematoxylin counter-staining. (a and b) Neuronal survival was quantified as the average number of viable neurons/field 24 h after plating of neurons (5 × 104). (c) Co-cultures of both wild-type neurons and astrocytes or (d) Prnp0/0 neurons and astrocytes. (e and f) Neuritogenesis was quantified as the number of cells with neurites longer than three cell body lengths 24 h after plating of neurons (5 × 104). (a and e) Prnp+/+ (inline image) or Prnp0/0 (□) neurons plated in co-culture with Prnp+/+ (aPrnp+/+) or Prnp0/0 (aPrnp+/+) astrocytes. (b and f) Prnpwt/wt (bsl00001) or Prnp−/− (inline image) neurons plated in co-culture with Prnpwt/wt (aPrnpwt/wt) or Prnp−/− (aPrnp−/−) astrocytes. Anti-PrPC antibody, 6H4 (α-PrPC) or non-immune IgG (IgG) were used to treat co-cultures of Prnpwt/wt neurons plated on Prnpwt/wt astrocytes (b and f). Mean data values ± SD from at least four independent experiments are presented. *p < 0.05.

Download figure to PowerPoint

On the other hand, Prnp+/+ neurons plated over Prnp+/+ astrocytes developed long neurites (Fig. 1c) while only short or even no neurites were observed in Prnp0/0 neurons seeded over Prnp0/0 astrocytes (Fig. 1d). Quantitative analysis (Fig. 1e) demonstrated that when both cell types are Prnp+/+, 42% of the neurons grew neurites longer than three cell bodies in length, while the absence of PrPC either in neurons or astrocytes significantly reduced neuritogenesis. When neurons and astrocytes are Prnp0/0, the number of neurite-bearing cells decreases to about 16%. These experimental conditions were reproduced using co-cultures from Prnpwt/wt or Prnp−/− embryos (Fig. 1f). Furthermore, in the presence of the anti-PrPC (α-PrPC) antibody 6H4, wild-type neuron and astrocyte co-cultures showed neuritogenesis that was similar to that shown by Prnp−/− neurons seeded over Prnp−/− astrocytes (Fig. 1f). No effect was observed when a pre-immune IgG was used in Prnpwt/wt astrocyte–neuron co-cultures.

These results indicate that when neurons are maintained in co-culture with astrocytes PrPC expression does not affect neuronal survival. Conversely, the presence of PrPC in both astrocytes and neurons is important for neuritogenesis. Thus, PrPC may modulate neuronal–astrocytic interactions as well as the astrocytic secretion of soluble factors and ECM proteins that maintain neuronal networking.

ECM from PrPC-null astrocytes is less permissive to neuritogenesis than ECM from wild-type astrocytes

When Prnp+/+ or Prnp0/0 neurons were cultured over ECM produced by Prnp0/0 astrocytes for 24 h, the percentage of neurite-bearing cells decreased by 40% relative to those plated on ECM from Prnp+/+ astrocytes (Fig. 2a). Prnpwt/wt or Prnp−/− neurons cultured over ECM from Prnp−/− astrocytes also exhibited decreased neuritogenesis relative to those plated on ECM from Prnpwt/wt astrocytes (Fig. 2b). These observations indicate that the ECM contains important factors for neuronal differentiation, and that PrPC expression in astrocytes may regulate their production, secretion, and/or organization.

image

Figure 2.  Cellular prion protein (PrPC)-null astrocytes produce extracellular matrix (ECM) less permissive to neuritogenesis than wild-type astrocytes. Neuronal survival and differentiation were evaluated in neurons cultured on ECM from wild-type and PrPC-null astrocytes. Cells were immunolabeled for βIII-tubulin followed by immunoperoxidase staining and hematoxylin counter-staining. (a) Neuritogenesis (number of cells with neurites longer than three cell bodies) and (c) survival (average number of viable neurons/field 24 h after plating) were quantified in Prnp+/+ (inline image) or Prnp0/0 (□) neurons plated in ECM from Prnp+/+ (ECM aPrnp+/+) or Prnp0/0 (ECM aPrnp0/0) astrocytes. (b) Neuritogenesis and (d) neuronal survival were also evaluated in Prnpwt/wt (bsl00001) or Prnp−/− (inline image) neurons plated in ECM from Prnpwt/wt (aPrnpwt/wt) or Prnp−/− (aPrnp−/−) astrocytes. Mean data values ± SD from at least four independent experiments are presented. *p < 0.05.

Download figure to PowerPoint

In the presence of ECM from Prnpwt/wt astrocytes, Prnp−/− neurons presented a lower neuritogenesis than their wild-type controls (Fig. 2b). Under the same conditions, no differences were found when Prnp0/0 neurons were compared with their respective controls (Fig. 2a). Nonetheless, as the percentage of cells with neurites over ECM from Prnp+/+ astrocytes (Fig. 2a) was lower than that observed in cells with ECM from Prnpwt/wt astrocytes (Fig. 2b), experimental variability could limit the detection of neuritogenesis differences between Prnp+/+ and Prnp0/0 neurons cultured with Prnp+/+ ECM. Indeed, PrPC expression in neurons can also contribute to neuronal differentiation on ECM.

Nonetheless, the present observations suggest that neuronal survival under these conditions may be independent of PrPC expression in either neurons or astrocytes (Fig. 2c and d). Additionally, overall neuronal survival and differentiation in ECM (Fig. 2) were lower than those presented in co-culture conditions (Fig. 1), likely because of increased deprivation of cellular contacts and nutrients.

Laminin deposition pattern differs between wild-type and PrPC-null astrocytes and is related to cells’ ability to sustain neuritogenesis

As LN deposition pattern is important for neuronal differentiation (Garcia-Abreu et al. 1995) and PrPC interacts with the γ1-chain of LN-inducing neuritogenesis (Graner et al. 2000a), we evaluated its expression and organization in ECM produced by wild-type and PrPC-null astrocytes. Fluorescent immunocytochemistry with anti-LN 1 antibodies revealed that LN secreted by wild-type astrocytes, Prnp+/+ (Fig. 3a) or Prnpwt/wt (Fig. 3c), presented a fibrillary pattern while that deposited by PrPC-null astrocytes, Prnp0/0 (Fig. 3b) or Prnp−/− (Fig. 3d), showed a punctate distribution in which short fibers predominated. When astrocytes were removed leaving isolated ECM, the distribution patterns of LN were similar to those described above (data not shown).

image

Figure 3.  Wild-type astrocytes organize a fibrillar laminin (LN) at the extracellular matrix (ECM) while that from cellular prion protein (PrPC)-null astrocytes has a punctate pattern. Immunocytochemical detection of LN secreted and deposited by wild-type Prnp+/+ (a) and Prnpwt/wt (c) or PrPC-null Prnp0/0 (b) and Prnp−/− (d) astrocytes. Cells were cultured until confluence before fixation. The labeling of deposited LN was performed with a primary antibody against LN 1 followed by a secondary antibody conjugated to FITC. Cells were then permeabilized and nuclei counter-stained with 4′-6-diamino-2-phenylindole (DAPI). Bar, 50 μm. (e) Western blot assays from cultured Prnp+/+ or Prnp0/0 astrocytes were performed with antibodies against LN 1 (upper panel) or anti-actin (lower panel). Intracellular pool (soluble) and matrix-assembled proteins (insoluble) were analyzed. Size standards (in kDa) are shown on the left. EHS LN chains (α, β, and γ) are shown on the right. Blots probed with anti-actin (45 kDa) antibody served as loading controls.

Download figure to PowerPoint

Semi-quantitative immunoblot analyses revealed no difference in LN levels between Prnp0/0 and Prnp+/+ astrocytes, both in intracellular (soluble) and in extracellular (insoluble) fractions (Fig. 3e). A band of approximately 250 kDa corresponding to the co-migration of γ1 and β1 chains, as seen for purified LN 1, was identified in the samples (Fig. 3e). The LN α-chain (450 kDa) was not detected as the antibody used was against LN 1, which contains an α1 chain, the expression of which is very low in the brain (Indyk et al. 2003). LN was undetectable, via immunoblotting, in the soluble portion from astrocyte CM (data not shown). We observed no difference in fibronectin organization between ECM from wild-type and PrPC-null (data not shown).

It has been described that fibrillary distributed, but not punctate distributed LN is permissive for neuronal differentiation (Garcia-Abreu et al. 1995). Thus our observation of reduced neuritogenesis with ECM from PrPC-null astrocytes (Fig. 2a and b) is consistent with the punctate LN distribution in the ECM in the respective astrocytes (Fig. 3b and d). However, it does no exclude the participation of other components of the ECM on the neuronal differentiation. These results suggest that PrPC expression in astrocytes may be involved with pattern organization of LN in the ECM.

PrPC expression in astrocytes modulates secreted factors responsible for neuronal survival

We next evaluated whether the presence of PrPC in astrocytes modulates the secretion of soluble factors responsible for neuronal development. Prnp−/− (Fig. 4a) and Prnp0/0 (Fig. S1a) neurons maintained in fresh medium without complements (DMEM) or in any other treatment performed in this experiment presented a lower survival than their corresponding wild-type control neurons. These findings indicate that under these experimental conditions survival is critically compromised when neuronal PrPC is depleted.

image

Figure 4.  Stress-inducible protein-1 (STI1) is secreted by wild-type and cellular prion protein (PrPC)-null astrocytes and mediates PrPC-dependent neuronal survival (a) Neuronal survival was evaluated in wild-type (Prnpwt/wtbsl00001) or PrPC-null (Prnp−/−inline image) primary cultures plated in the presence of astrocyte conditioned medium (CM) from Prnpwt/wt (aPrnpwt/wt) or Prnp−/− (aPrnp−/−) astrocytes. Recombinant STI1 (recSTI1; 0.1 μmol/L) was added to the CM from PrPC-null (Prnp−/−) astrocytes or to serum-free Dulbecco’s minimum essential medium (DMEM). Cells were immunolabeled for βIII-tubulin. Neuronal survival was quantified as the average number of viable neurons/field 24 h after plating neurons (5 × 104) for all conditions. Mean data values ± SD from at least four independent experiments are presented. *p < 0.05. (b) PrPC, STI1, and CDK-4 detection by western blot assays in extracts (lanes 1 and 2) and CM (lanes 3 and 4) from Prnpwt/wt (wt/wt) and Prnp−/− (−/−) astrocytes. Immunoprecipitation assays in CM from Prnp−/− astrocytes (lanes: 5–8) were conducted using anti-STI1 IgG (lanes 6 and 8) or pre-immune IgG (lanes 5 and 7). The supernatants (lanes 5 and 6) or pellets (lanes 7 and 8) from these immunoprecipitations were assayed by western blot using anti-STI1, anti-PrPC, and anti-CDK-4 antibodies. Bands corresponding to the IgG heavy and light chains are labeled on the right. (c) STI1 was immunoprecipitated from Prnp+/+ astrocyte-CM or Prnp0/0 astrocyte-CM with a purified IgG anti-STI1 and the supernatant depleted of STI1 (+) used to treat Prnp+/+ (inline image) or Prnp0/0 (□) neuronal cultures. A control experiment using CM whose immunoprecipitation was conducted in the presence of pre-immune IgG was performed (depleted IgG). Prnp+/+ or Prnp0/0 neuronal cultures were also treated with CM depleted of STI1 plus 0.1 μmol/L of recombinant STI1 (+recSTI1) or 1 μmol/L recombinant PrPC (+recPrPC). Neurons were immunolabeled for βIII-tubulin and survival (average number of viable neurons/field 24 h after plating 5 × 104 cells) was quantified. Mean data values ± SD from four independent experiments are presented. *p < 0.05.

Download figure to PowerPoint

Meanwhile, the wild-type or PrPC-null neurons treated with CM from wild-type astrocytes (CM aPrnpwt/wt) showed a higher survival than corresponding cells maintained in DMEM. Conversely, CM from PrPC-null astrocytes (CM aPrnp−/−) was unable to support neuronal survival better than DMEM (Fig. 4a). These findings indicate that PrPC expression in astrocytes may regulate the expression and/or secretion of factors responsible for neuronal survival.

We have previously demonstrated that the PrPC ligand STI1-induces neuronal survival (Chiarini et al. 2002; Zanata et al. 2002). In the present study we observed that addition of recSTI1 to Prnpwt/wt neurons maintained in DMEM or in CM from Prnp−/− astrocytes increased cell survival. Interestingly, neuronal protection reached a level similar to that obtained when wild-type neurons were cultured in CM from wild-type astrocytes. In contrast, there was no evidence of survival rescue of PrPC-null neurons by the STI1 treatment (Fig. 4a). Similar results to those observed with Edbg animal cells were obtained when cultures from Prnp+/+ and Prnp0/0 ZrchI animals were used (Fig. S1a). Thus, these data suggest that STI1 may be one of the factors secreted by astrocytes that mediate PrPC-dependent neuronal survival.

STI1 is a neuronal survival trophic factor secreted by astrocytes

As shown in Fig. 4b, western blot analysis using a specific antibody (Zanata et al. 2002) demonstrated that STI1 was equally expressed in extracts from both Prnpwt/wt (lane 1) and Prnp−/− astrocytes (lane 2), as well as in CM from the two cell types (lanes 3 and 4). The absence of cell lysis was confirmed by the lack of CDK-4 (a cell cycle cyclin-dependent kinase expressed in cytoplasm and nucleus) in the CM. Similar results were obtained when CM from Prnp+/+ and Prnp0/0 astrocytes were evaluated (Fig. S1b). Note that PrPC was also present in the CM from wild-type astrocytes (lane 3). Immunoprecipitation assays were performed to deplete secreted STI1 from CM. We observed that STI1 immunoprecipitated in the presence of a specific antibody (lane 8), but STI1 was not detected when a pre-immune IgG (lane 7) was used (Fig. 4b). The supernatant from this immunoprecipitation was completely depleted of STI1 (lane 6). It is important to note that PrPC co-immunoprecipitates with STI1 (lane 8), confirming our previously reported results demonstrating that PrPC binds to STI1 (Zanata et al. 2002).

The effect of STI1 immunodepletion from CM upon neuronal survival was measured (Fig. 4c). Wild-type or Prnp0/0 neurons cultured in wild-type astrocyte-CM depleted of STI1 (+) presented a reduction in survival relative to cells treated with wild-type astrocyte–CM or cells treated with CM immunoprecipitated with a non-immune IgG (depleted IgG). Additionally, CM from Prnp0/0 astrocytes imunodepleted of STI1 promoted a lower survival in wild-type neurons when compared with non-depleted Prnp0/0-CM. Neuronal survival was rescued in wild-type (Fig. 4c) but not in Prnp0/0 neurons (data not shown) when recSTI1 was added to STI1-depleted CM. Taken together these results strongly suggest that STI1 secreted by astrocytes mediates PrPC-dependent neuronal survival.

Stress-inducible protein-1 secretion by wild-type and PrPC-null astrocytes was similar (Fig. 4b, lanes 3 and 4). This however, does not explain why CM from wild-type astrocytes induces greater neuronal survival than CM from PrPC-null astrocytes (Fig. 4a and c). Furthermore, although STI1 mediates survival only in wild-type neurons (Fig. 4a and Lopes et al. 2005; Zanata et al. 2002), CM from wild-type astrocytes increases survival in both types of neurons when compared with either fresh medium (Fig. 4a) or to CM from Prnp0/0 astrocytes (Fig. 4c). Therefore, other secreted factors, besides STI1, present in CM from wild-type astrocytes and lacking in CM from PrPC-null astrocytes may also regulate neuronal survival.

As the depletion of STI1 from CM also withdrew PrPC (Fig. 4b), we added recombinant PrPC to STI1 depleted CM. This treatment was able to rescue neuronal survival of both wild-type and Prnp0/0 neurons (Fig. 4c). Indicating that soluble PrPC may also present a neuroprotective function.

PrPC is secreted by astrocytes and induces neuronal survival

The secretion of PrPC has been previously documented (Vincent et al. 2000, 2001; Fevrier et al. 2004), and this protein is an obvious candidate to be found in CM from wild-type astrocytes, but not in that from PrPC-null cells. Indeed secreted PrPC detected in CM from wild-type astrocytes presented a similar size and had a similar glycosylation pattern to that detected in cellular extracts (Fig. 4b lanes 1 and 3 and Fig. S1b). This pattern is more consistent with its secretion in exosomes (Fevrier et al. 2004) than by proteolytic cleavage at the hydrophobic domain (Vincent et al. 2000, 2001). As expected, no PrPC was present in PrPC-null astrocyte extracts or CM (Fig. 4b, lanes 2 and 4).

An immunoprecipitation assay was performed to deplete secreted PrPC from wild-type astrocyte CM. Figure 5a shows that PrPC was immunoprecipitated in the presence of a specific antibody (lane 4), but was not detected when a pre-immune IgG was used for the precipitation (lane 3). The supernatant from this immunoprecipitation was completely depleted of PrPC (lane 2). STI1 was partially removed from the CM after PrPC immunoprecipitation (lane 1 vs. lane 2). STI1 co-immunoprecipitation with PrPC is clearly demonstrated in the immunoprecipitation pellets (lane 3 vs. lane 4). These results confirm that PrPC and STI1 bind one another but in this case, part of the secreted STI1 is still present in the CM after the complete withdrawal of PrPC.

image

Figure 5.  Cellular prion protein (PrPC) secreted by wild-type astrocytes induces neuronal survival in wild-type and PrPC-null neurons. (a) Immunoprecipitation assays in conditioned medium (CM) from Prnp+/+ astrocytes were conducted using anti-PrPC IgG (lanes 2 and 4) or non-immune IgG (lanes 1 and 3). The supernatants (lanes 1 and 2) or pellets (lanes 3 and 4) from these immunoprecipitations were assayed by western blot using anti-PrPC and anti-stress-inducible protein-1 (STI1) serum. Bands corresponding to the IgG heavy and light chains are labeled on the right. (b) Survival analysis of Prnp+/+ (inline image) and Prnp0/0 (□) neurons, or (c) Prnpwt/wt (bsl00001) and Prnp−/− (inline image) neurons cultured in Dulbecco’s minimum essential medium (DMEM) with or without recombinant (rec) PrPC (0.1 or 1 μmol/L). (d) Survival analysis in neurons cultured in CM from Prnp+/+ astrocytes (CM) before and after PrPC immunodepletion. CM was immunoprecipitated with IgG anti-PrPC or with non-immune IgG as control. Supernatants depleted of PrPC (depleted PrPC) or from control (depleted IgG) or supernatants depleted of PrPC plus 1 μmol/L recombinant PrPC (depleted PrPC + rec PrPC) were used to maintain Prnp+/+ (inline image) and Prnp0/0 (□) neurons. (e) Survival analysis in Prnp+/+ (inline image) and Prnp0/0 (□) neurons cultured in CM from Prnp0/0 astrocytes (CM aPrnp0/0) in the absence or presence of 1 μmol/L recombinant PrPC (+rec PrPC). (b–e) Cells were immunolabeled for βIII-tubulin and survival (average number of viable neurons/field 24 h after plating 5 × 104 cells) was quantified. Mean data values ± SD from four independent experiments are presented. In (d and e) p < 0.05 for all treatment conditions of Prnp+/+ neurons compared with Prnp0/0 neurons (*p < 0.05).

Download figure to PowerPoint

To better evaluate the role of soluble PrPC on neuronal survival, recombinant PrPC was added to fresh media of Prnp+/+ and Prnp0/0 neuronal cultures. The addition of soluble PrPC protein stimulated neuronal survival in a dose-dependent manner in both wild-type and Prnp0/0 cells (Fig. 5b). This pro-survival effect was replicated in parallel experiments in which Prnpwt/wt and Prnp−/− neurons were treated with recombinant PrPC (Fig. 5c).

Prnp+/+ and Prnp0/0 neurons maintained in PrPC-immunodepleted CM from wild-type astrocytes (depleted PrPC) presented a 40% lower survival rate than Prnp+/+ and Prnp0/0 neurons treated with non-depleted CM (Fig. 5d). Survival was rescued in both Prnp+/+ and Prnp0/0 neurons when PrPC-depleted CM was supplemented with recombinant PrPC. In addition, the supplementation of CM from Prnp0/0 astrocytes with recombinant PrPC also increased survival of wild-type and PrPC-null neurons (Fig. 5e). Although STI1 is partially removed after PrPC immunoprecipitation, a substantial fraction of the protein is still present in the CM (Fig. 5a), so decreased survival cannot be attributed to STI1 withdrawal.

These data indicate that PrPC secreted by astrocytes acts as a neuronal trophic factor. Neuronal survival in CM is similar to that in ECM and lower than the overall survival in co-cultures, indicating that an association of secreted factors, ECM and cell-to-cell contact are necessary to maintain neuronal viability.

PrPC and STI1 are secreted in CM (Fig. 4b) and their roles in neuronal differentiation have been recently demonstrated (Chen et al. 2003; Lopes et al. 2005; Santuccione et al. 2005). Our experimental conditions differed from previous experiments in that the culture media in our study was not supplemented with any factors, whereas the media in previous was supplemented with B-27 (Lopes et al. 2005) or LN (Santuccione et al. 2005). Although STI1 and PrPC were present in the astrocyte CM, neuritogenesis was not observed when neurons were plated under our non-supplemented condition. Thus, under our experimental conditions the CM secreted by astrocytes is a poor substrate for neuritogenesis, and other factors present in B-27 supplement or LN might be necessary for this effect.

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Here we have demonstrated that PrPC expression in astrocytes modulates neuron–astrocyte interaction, LN organization in the ECM, and the secretion of soluble and ECM factors responsible for neuronal survival and differentiation. Our experiments were conducted with neurons and astrocytes derived from two independent lines of PrPC-null mice (Bueler et al. 1992; Manson et al. 1994) and their respective wild-type controls to ensure that effects observed were not a consequence of any spurious differences between wild-type and PrPC-null mice.

Astrocyte–neuron cross-talk and the importance of PrPC, STI1, and laminin interactions in neuronal differentiation

Astrocyte–neuron co-cultures mimic a milieu closer to in vivo conditions than isolated cultures. Our experiments showed that neuritogenesis decreased when either neurons or astrocytes in neuron–astrocyte co-cultures did not express PrPC, and was decreased even further when both cells were PrPC-depleted. Homo- and heterophilic interactions, as well as soluble and insoluble factors, produced by astrocytes have been shown to modulate neuronal survival and differentiation (Fields and Stevens-Graham 2002). Regarding cell–cell interaction at least two PrPC ligands, STI1 (which is also detected at the cell surface, Zanata et al. 2002; Lopes et al. 2005) and NCAM, may participate in heterophilic interactions triggering neuronal differentiation (Lopes et al. 2005; Santuccione et al. 2005). STI1–PrPC binding triggers signaling through PrPC (Chiarini et al. 2002; Zanata et al. 2002), while PrPC–NCAM activates signaling through NCAM (Santuccione et al. 2005). Thus, STI1 expressed in wild-type and PrPC-null astrocytes can induce differentiation in neurons expressing PrPC, while PrPC present in wild-type astrocytes triggers neuritogenesis through an interaction with NCAM. This may explain why neuritogenesis is higher when both astrocytes and neurons express PrPC, is lower when PrPC is expressed in either astrocytes or neurons alone, and is even lower when both cells are PrPC-null. Nevertheless, there may be other, as yet unidentified, cell surface proteins involved in cell-to-cell interaction and signaling whose expression or activation may be modulated by PrPC. A schematic view of these points is shown in Fig. 6a.

image

Figure 6.  Astrocyte–neuron interactions and secreted factors mediating neuronal survival and differentiation. (a) Proposed mechanism involving heterophilic astrocyte–neuron interactions mediating cellular prion protein (PrPC)-dependent neuronal differentiation. Stress-inducible protein-1 (STI1) expressed in wild-type or PrPC-null astrocytes interacts with neurons expressing PrPC while PrPC expressed in wild-type astrocytes interacts with neural cell adhesion molecule (NCAM) (Santuccione et al. 2005) in neurons. Neuronal differentiation is greater when both astrocytes and neurons express PrPC. Intermediate differentiation levels are present when either neurons or astrocytes express PrPC and negligible differentiation is observed when PrPC-null neurons and astrocytes are co-cultured. (b) PrPC, STI1, and laminin (LN) secretion by astrocytes mediate neuronal survival and differentiation. Wild-type astrocytes deposit a fibrillary LN at the extracellular matrix (ECM) which induces neuronal differentiation (neuritogenesis) in wild-type and PrPC-null neurons, probably using either PrPC or integrins. On the other hand, PrPC-null astrocytes produce a punctate LN unable to support neuronal differentiation. STI1 is secreted by wild-type astrocytes and induces neuronal survival in wild-type neurons but not in PrPC-null neurons. PrPC is also secreted by wild-type astrocytes and induces survival in both wild-type and PrPC-null neurons probably using a cell surface receptor. This ligand was not characterized here but we hypothesize that it could be NCAM (Santuccione et al. 2005). PrPC-null astrocytes also secreted STI1 and its effect is detected in wild-type neurons however other trophic factors may be also missing in CM from these astrocytes. Recombinant STI1 added to CM from PrPC-null astrocytes increases the survival of wild-type neurons but is incapable of supporting survival of PrPC-null neurons.

Download figure to PowerPoint

The ECM protein LN plays a crucial role during nervous system development (Liesi and Silver, 1988; Luckenbill-Edds 1997). This molecule is organized in the ECM of the developing brain as aggregates of distinct morphologies (Zhou 1990). Similarly, the midbrain astrocyte cultures deposit LN matrices organized in different patterns, fibrillar, and punctate, with the former being able to support neurite outgrowth (Garcia-Abreu et al. 1995). Here, we demonstrated that LN produced by PrPC-null astrocytes presented a punctate rather than a fibrillary distribution, which is consistent with its inability to sustain neuritogenesis in wild-type and in PrPC-null neurons. Thus, our data are in agreement with what has been described previously regarding punctate LN and neuritogenesis (Garcia-Abreu et al. 1995).

In neutral buffer, LN matrices form aggregates of heterogeneous morphology, while in acidic buffer they form homogenous hexagonal sheet-like structures. Indeed, the neurite outgrowth is greater in matrices formed under acidic conditions (Freire et al. 2002). Thus, it is possible that different LN assembly modes can occur depending on the type of molecules (more or less acidic) expressed at the surface of astrocytes. PrPC contains two N-linked glycosylation sites and at least 52 different sugars. Additionally, PrPC’s glycosyl phosphatidylinositol anchor is modified with a sialic acid (Rudd et al. 2001), which could give a more acidic profile to the milieu of PrPC-expressing membranes and may thereby lead to a fibrillary LN organization.

Both integrins and PrPC act as LN receptors (Colognato and Yurchenco 2000; Graner et al. 2000a,b), promoting neuronal adhesion and neuritogenesis. As both wild-type and PrPC-null neurons exhibited impaired neuritogenesis on the punctate LN, we suggest that this type of assembly is not an ideal substrate for PrPC, integrins, or other LN receptors. Our findings provide strong evidence indicating that PrPC expressed on the astrocyte cell surface is involved in LN organization, and promotes LN assembly that is more permissive to neuritogenesis. Elucidating the mechanisms underlining the influence of PrPC on LN organization was not the objective of present study, but the merits are further investigated.

Astrocyte–neuron cross-talk and the importance of PrPC and STI1 in neuronal survival

Combinations of wild-type and PrPC-null neuron–astrocyte co-cultures demonstrated that under these conditions neuronal survival was not affected by PrPC expression. However, when neuron–astrocyte contact was abolished and PrPC-null neurons were cultured in the presence of DMEM or CM, survival decreased and the importance of PrPC expression in neurons was clearly demonstrated.

Some years ago, Kuwahara et al. (1999) demonstrated that neurons from a PrPC-null mouse strain were more sensitive to serum deprivation than cells from wild-type mice. This observation was attributed to the elevated levels of the neurotoxic protein Doppel, rather than an absence of PrPC (Moore et al. 1999). However, our data demonstrated that neurons from two other PrPC-null mice strains (Bueler et al. 1992; Manson et al. 1994), which do not express Doppel (Moore et al. 1999), also presented a lower survival in the absence of serum, indicating that in isolated cultures PrPC-null neurons are more sensitive to cell death. Together our findings demonstrate that signals triggered by homophilic and/or heterophilic interactions between neurons and astrocytes strongly contribute to neuronal survival, while isolated neurons seem to be more dependent on PrPC signaling per se. As many molecules may be involved in cell–cell contact, the relevance of PrPC in neuronal survival may be masked under co-culture conditions.

PrPC has been associated with survival signals (Chen et al. 2003), and rescues retina and hippocampal neurons from induced programmed cell death upon its interaction with STI1 (Chiarini et al. 2002; Zanata et al. 2002). However, the basal levels of apoptotic neurons in retina explants from wild-type or Prnp0/0 mice are not altered (Chiarini et al. 2002), and PrPC-null mice do not present abnormalities in brain anatomy or behavior (Bueler et al. 1992). Nonetheless, these mice present electrophysiological changes (Colling et al. 1996), sleep disturbances (Tobler et al. 1996) and are more vulnerable than wild-type mice to stress conditions such as kainic acid treatment or ischemia (Walz et al. 1999; Spudich et al. 2005).

Neurons, expressing PrPC or not, displayed the same survival when plated over ECM from wild-type or PrPC-null astrocytes, although the latter produced a LN whose organization differed from the former. Thus, proteins such as fibronectin at the ECM and integrins at the cell surface (Bozzo et al. 2004) may have a more prominent survival role than LN and PrPC under these experimental conditions. Neuronal survival in the presence of astrocyte CM is dependent on PrPC expression in both cells. This dual dependency suggests that secreted factors may play the most important role in PrPC-dependent neuronal survival.

We characterized STI1 as a ligand for PrPC and putative mediator of survival signals (Chiarini et al. 2002; Zanata et al. 2002). STI1 protein is abundantly expressed in the cytoplasm (Lassle et al. 1997) and on the cell surface (Zanata et al. 2002; Lopes et al. 2005), despite the absence of any signal peptide for this localization. Herein, we demonstrated that STI1 is also secreted from astrocytes, similarly to its human homolog, Hop, which is secreted by a fibrosarcoma cell line (Eustace and Jay 2004). HSP90α and its associated proteins have also been found in the extracellular space (Liao et al. 2000; Kakimura et al. 2002), despite having no signal peptide. The mechanism associated with the secretion of STI1, which is presently under evaluation, is unknown. However, STI1 appears not to be affected by PrPC expression as both wild-type and PrPC-null astrocytes express and secrete the same levels of STI1. Remarkably, HSP70 has been recently described as secreted by lumbar spinal astrocytes, mediating motor neuron survival (Robinson et al. 2005).

PrPC was also secreted from wild-type astrocytes and induced neuronal survival in both wild-type and PrPC-null cells. These signals may be transduced by PrPC association with a protein at the neuronal surface. STI1 may not be a candidate for this process, as we have previously shown that the survival signal passes from STI1 through membrane anchored PrPC (Chiarini et al. 2002; Zanata et al. 2002). However NCAM, another PrPC ligand (Santuccione et al. 2005) expressed at the cell surface, could be a candidate and its role in PrPC-dependent neuronal survival deserves further study. Together, the present data add PrPC and STI1 to the class of trophic factors secreted by astrocytemediating neuronal survival. A schematic view of these points is shown in Fig. 6b.

Because of the importance of astrocytes in neuronal survival and differentiation, the evaluation of PrPC function in astrocyte–neuron cultures may provide important clues to the PrPC loss of function in prion diseases. Astrocytes are intimately associated with synapses and strategically allocated to regulate synaptic transmission (Araque et al. 2001). Therefore, these cells may have important roles in neuronal plasticity and cognition.

It has been demonstrated that PrPSc initially accumulates in astrocytes and later spreads to neurons (Diedrich et al. 1991; Raeber et al. 1997), however the importance of this event in prion diseases is controversial. PrPSc propagation by glia can lead to neuronal lesions independent of PrPC expression in neurons (Jeffrey et al. 2004). Conversely, spongiform degeneration in the hippocampus as well as behavioral abnormalities promoted by prion infection are reversed when PrPC is selectively depleted in neurons despite of the continuous prion replication in astrocytes (Mallucci et al. 2002, 2003, 2007).

Interestingly, humans have a higher ratio of glia to neurons in their brains than any other species. This observation together with the distinctive cognitive facilities of humans, suggest that glia are key regulatory elements to higher cortical functions (Araque et al. 2001). Accordingly, the expression of PrPC by glia, as well as by neurons, and its interaction with STI1 could play a major role in the brain and PrPC loss-of-function may be associated with the pathogenesis of prion diseases. Furthermore, these results suggest that astrocyte’s secreted factors mediating neuronal survival and differentiation may represent new targets for therapeutic intervention in neurodegenerative conditions such as prion diseases.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported by a grant from FAPESP (Fundação de Amparo a Pesquisa do Estado de São Paulo) (03-13189-2) and Vilma Regina Martins is supported by a grant from the Howard Hughes Medical Institute. FRSL, CPA, and AGM are fellows of the FAPESP. We thank Drs Charles Weissmann, Bruce Chesebro, Richard Race, and Jean Manson for PrPC-null mice lines and Dr Glaucia N. Hajj for critical reading of this manuscript.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Araque A., Carmignoto G. and Haydon P. G. (2001) Dynamic signaling between astrocytes and neurons. Annu. Rev. Physiol. 63, 795813.
  • Bozzo C., Lombardi G., Santoro C. and Canonico P. L. (2004) Involvement of beta(1) integrin in betaAP-induced apoptosis in human neuroblastoma cells. Mol. Cell. Neurosci. 25, 18.
  • Brown D. R., Qin K., Herms J. W. et al. (1997a) The cellular prion protein binds copper in vivo. Nature 390, 684687.
  • Brown D. R., Schulz-Schaeffer W. J., Schmidt B. and Kretzschmar H. A. (1997b) Prion protein-deficient cells show altered response to oxidative stress due to decreased SOD-1 activity. Exp. Neurol. 146, 104112.
  • Bueler H., Fischer M., Lang Y., Bluethmann H., Lipp H. P., DeArmond S. J., Prusiner S. B., Aguet M. and Weissmann C. (1992) Normal development and behaviour of mice lacking the neuronal cell-surface PrP protein. Nature 356, 577582.
  • Chammas R., Taverna D., Cella N., Santos C. and Hynes N. E. (1994) Laminin and tenascin assembly and expression regulate HC11 mouse mammary cell differentiation. J. Cell Sci. 107 (part 4), 10311040.
  • Chen S., Mange A., Dong L., Lehmann S. and Schachner M. (2003) Prion protein as trans-interacting partner for neurons is involved in neurite outgrowth and neuronal survival. Mol. Cell. Neurosci. 22, 227233.
  • Chiarini L. B., Freitas A. R., Zanata S. M., Brentani R. R., Martins V. R. and Linden R. (2002) Cellular prion protein transduces neuroprotective signals. EMBO J. 21, 33173326.
  • Cohen F. E. (1999) Protein misfolding and prion diseases. J. Mol. Biol. 293, 313320.
  • Colling S. B., Collinge J. and Jefferys J. G. (1996) Hippocampal slices from prion protein null mice: disrupted Ca(2+)-activated K+ currents. Neurosci. Lett. 209, 4952.
  • Colognato H. and Yurchenco P. D. (2000) Form and function: the laminin family of heterotrimers. Dev. Dyn. 218, 213234.
  • Cronier S., Laude H. and Peyrin J. M. (2004) Prions can infect primary cultured neurons and astrocytes and promote neuronal cell death. Proc. Natl Acad. Sci. USA 101, 1227112276.
  • Diedrich J. F., Bendheim P. E., Kim Y. S., Carp R. I. and Haase A. T. (1991) Scrapie-associated prion protein accumulates in astrocytes during scrapie infection. Proc. Natl Acad. Sci. USA 88, 375379.
  • Eustace B. K. and Jay D. G. (2004) Extracellular roles for the molecular chaperone, hsp90. Cell Cycle 3, 10981100.
  • Fevrier B., Vilette D., Archer F., Loew D., Faigle W., Vidal M., Laude H. and Raposo G. (2004) Cells release prions in association with exosomes. Proc. Natl Acad. Sci. USA 101, 96839688.
  • Fields R. D. and Stevens-Graham B. (2002) New insights into neuron-glia communication. Science 298, 556562.
  • Freire E., Gomes F. C., Linden R., Neto V. M. and Coelho-Sampaio T. (2002) Structure of laminin substrate modulates cellular signaling for neuritogenesis. J. Cell Sci. 115, 48674876.
  • Garcia-Abreu J., Cavalcante L. A. and Moura N. V. (1995) Differential patterns of laminin expression in lateral and medial midbrain glia. Neuroreport 6, 761764.
  • Gomes F. C., Garcia-Abreu J., Galou M., Paulin D. and Moura N. V. (1999) Neurons induce GFAP gene promoter of cultured astrocytes from transgenic mice. Glia 26, 97108.
  • Gomes F. C., Spohr T. C., Martinez R. and Moura N. V. (2001) Cross-talk between neurons and glia: highlights on soluble factors. Braz. J. Med. Biol. Res. 34, 611620.
  • Graner E., Mercadante A. F., Zanata S. M. et al. (2000a) Cellular prion protein binds laminin and mediates neuritogenesis. Brain Res. Mol. Brain Res. 76, 8592.
  • Graner E., Mercadante A. F., Zanata S. M., Martins V. R., Jay D. G. and Brentani R. R. (2000b) Laminin-induced PC-12 cell differentiation is inhibited following laser inactivation of cellular prion protein. FEBS Lett. 482, 257260.
  • Hetz C., Maundrell K. and Soto C. (2003) Is loss of function of the prion protein the cause of prion disorders? Trends Mol. Med. 9, 237243.
  • Indyk J. A., Chen Z. L., Tsirka S. E. and Strickland S. (2003) Laminin chain expression suggests that laminin-10 is a major isoform in the mouse hippocampus and is degraded by the tissue plasminogen activator/plasmin protease cascade during excitotoxic injury. Neuroscience 116, 359371.
  • Jeffrey M., Goodsir C. M., Race R. E. and Chesebro B. (2004) Scrapie-specific neuronal lesions are independent of neuronal PrP expression. Ann. Neurol. 55, 781792.
  • Kakimura J, Kitamura Y., Takata K. et al. (2002) Microglial activation and amyloid-beta clearance induced by exogenous heat-shock proteins. FASEB J. 16, 601603.
  • Kretzschmar H. A., Tings T., Madlung A., Giese A. and Herms J. (2000) Function of PrP(C) as a copper-binding protein at the synapse. Arch. Virol. Suppl. 16, 239249
  • Kuwahara C.., Takeuchi A. M., Nishimura T. et al. (1999) Prions prevent neuronal cell-line death. Nature 400, 225226.
  • Lassle M., Blatch G. L., Kundra V., Takatori T. and Zetter B. R. (1997) Stress-inducible, murine protein mSTI1. Characterization of binding domains for heat shock proteins and in vitro phosphorylation by different kinases. J. Biol. Chem. 272, 18761884.
  • Lee K. S., Linden R., Prado M. A., Brentani R. R. and Martins V. R. (2003) Towards cellular receptors for prions. Rev. Med. Virol. 13, 399408.
  • Liao D. F., Jin Z. G., Baas A. S., Daum G., Gygi S. P., Aebersold R. and Berk B. C. (2000) Purification and identification of secreted oxidative stress-induced factors from vascular smooth muscle cells. J. Biol. Chem. 275, 189196.
  • Liesi P. and Silver J. (1988) Is astrocyte laminin involved in axon guidance in the mammalian CNS? Dev. Biol. 130, 774785
  • Lima F. R.., Trentin A. G., Rosenthal D., Chagas C. and Moura N. V. (1997) Thyroid hormone induces protein secretion and morphological changes in astroglial cells with an increase in expression of glial fibrillary acidic protein. J. Endocrinol. 154, 167175.
  • Lima F. R., Gervais A., Colin C., Izembart M., Neto V. M. and Mallat M. (2001) Regulation of microglial development: a novel role for thyroid hormone. J. Neurosci. 21, 20282038.
  • Lopes J. D., Dos R. M. and Brentani R. R. (1985) Presence of laminin receptors in Staphylococcus aureus. Science 229, 275277.
  • Lopes M. H., Hajj G. M. N., Muras A. G., Mancini G. M., Castro R. M. P. S., Ribeiro K. C. B., Brentani R. R., Linden R. and Martins V. R. (2005) Interaction of cellular prion and stress inducible protein 1 promotes neuritogenesis and neuroprotection by distinct signaling pathways. J. Neurosci. 25, 1133011339.
  • Luckenbill-Edds L. (1997) Laminin and the mechanism of neuronal outgrowth. Brain Res. Brain Res. Rev. 23, 127.
  • Mallucci G. R., Ratte S., Asante E. A., Linehan J., Gowland I., Jefferys J. G. and Collinge J. (2002) Post-natal knockout of prion protein alters hippocampal CA1 properties, but does not result in neurodegeneration. EMBO J. 21, 202210.
  • Mallucci G., Dickinson A., Linehan J., Klohn P. C., Brandner S. and Collinge J. (2003) Depleting neuronal PrP in prion infection prevents disease and reverses spongiosis. Science 302, 871874.
  • Mallucci G. R., White M. D., Farmer M., Dickinson A., Khatun H., Powell A. D., Brandner S., Jefferys J. G. and Collinge J. (2007) Targeting cellular prion protein reverses early cognitive deficits and neurophysiological dysfunction in prion-infected mice. Neuron 53, 325335.
  • Manson J. C., Clarke A. R., Hooper M. L., Aitchison L., McConnell I. and Hope J. (1994) 129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal. Mol. Neurobiol. 8, 121127.
  • Martins V. R., Graner E., Garcia-Abreu J., De Souza S. J., Mercadante A. F., Veiga S. S., Zanata S. M., Neto V. M. and Brentani R. R. (1997) Complementary hydropathy identifies a cellular prion protein receptor. Nat. Med. 3, 13761382.
  • Martins V. R., Linden R., Prado M. A., Walz R., Sakamoto A. C., Izquierdo I. and Brentani R. R. (2002) Cellular prion protein: on the road for functions. FEBS Lett. 512, 2528.
  • Moore R. C., Lee I. Y., Silverman G. L. et al. (1999) Ataxia in prion protein (PrP)-deficient mice is associated with upregulation of the novel PrP-like protein doppel. J. Mol. Biol. 292, 797817.
  • Niederreiter M., Gimona M., Streichsbier F., Celis J. E. and Small J. V. (1994) Complex protein composition of isolated focal adhesions: a two-dimensional gel and database analysis. Electrophoresis 15, 511519.
  • Parry G., Lee E. Y., Farson D., Koval M. and Bissell M. J. (1985) Collagenous substrata regulate the nature and distribution of glycosaminoglycans produced by differentiated cultures of mouse mammary epithelial cells. Exp. Cell Res. 156, 487499.
  • Prusiner S. B., Scott M. R., DeArmond S. J. and Cohen F. E. (1998) Prion protein biology. Cell 93, 337348.
  • Raeber A. J., Race R. E., Brandner S. et al. (1997) Astrocyte-specific expression of hamster prion protein (PrP) renders PrP knockout mice susceptible to hamster scrapie. EMBO J. 16, 60576065.
  • Robinson M. B., Tidwell J. L., Gould T., Taylor A. R., Newbern J. M., Graves J., Tytell M. and Milligan C. E. (2005) Extracellular heat shock protein 70 a critical component for motoneuron survival. J. Neurosci. 25, 97359745.
  • Rudd P. M., Wormald M. R., Wing D. R., Prusiner S. B. and Dwek R. A. (2001) Prion glycoprotein: structure, dynamics, and roles for the sugars. Biochemistry 40, 37593766.
  • Samaia H. B. and Brentani R. R. (1998) Can loss-of-function prion-related diseases exist? Mol. Psychiatry 3, 196197.
  • Santuccione A., Sytnyk V., Leshchyns’ka I. and Schachner M. (2005) Prion protein recruits its neuronal receptor NCAM to lipid rafts to activate p59fyn and to enhance neurite outgrowth. J. Cell Biol. 169, 341354.
  • Schmitt-Ulms G., Legname G., Baldwin M. A. et al. (2001) Binding of neural cell adhesion molecules (N-CAMs) to the cellular prion protein. J. Mol. Biol. 314, 12091225.
  • Spudich A., Frigg R., Kilic E., Kilic U., Oesch B., Raeber A., Bassetti C. L. and Hermann D. M. (2005) Aggravation of ischemic brain injury by prion protein deficiency: Role of ERK-1/-2 and STAT-1. Neurobiol. Dis. 20, 442449.
  • Sykova E. (2001) Glial diffusion barriers during aging and pathological states. Prog. Brain Res. 132, 339363.
  • Tobler I., Gaus S. E., Deboer T., Achermann P., Fischer M., Rulicke T., Moser M., Oesch B., McBride P. A. and Manson J. C. (1996) Altered circadian activity rhythms and sleep in mice devoid of prion protein. Nature 380, 639642.
  • Vincent B., Paitel E., Frobert Y., Lehmann S., Grassi J. and Checler F. (2000) Phorbol ester-regulated cleavage of normal prion protein in HEK293 human cells and murine neurons. J. Biol. Chem. 275, 3561235616.
  • Vincent B., Paitel E., Saftig P., Frobert Y., Hartmann D., De Strooper B., Grassi J., Lopez-Perez E. and Checler F. (2001) The disintegrins ADAM10 and TACE contribute to the constitutive and phorbol ester-regulated normal cleavage of the cellular prion protein. J. Biol. Chem. 276, 3774337746.
  • Walz R., Amaral O. B., Rockenbach I. C., Roesler R., Izquierdo I., Cavalheiro E. A., Martins V. R. and Brentani R. R. (1999) Increased sensitivity to seizures in mice lacking cellular prion protein. Epilepsia 40, 16791682.
  • Westergard L., Christensen H. M. and Harris D. A. (2007) The cellular prion protein PrPC: its physiological function and role in disease. Biochim. Biophys. Acta 1772, 629644.
  • Zanata S. M., Lopes M. H., Mercadante A. F. et al. (2002) Stress-inducible protein 1 is a cell surface ligand for cellular prion that triggers neuroprotection. EMBO J. 21, 33073316.
  • Zhou F. C. (1990) Four patterns of laminin-immunoreactive structure in developing rat brain. Brain Res. Dev. Brain Res. 55, 191201.