Glycobiology Advance Access originally published online on March 27, 2007
Glycobiology 2007 17(7):725-734; doi:10.1093/glycob/cwm034
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Trypanosome trans-sialidase mediates neuroprotection against oxidative stress, serum/glucose deprivation, and hypoxia-induced neurite retraction in Trk-expressing PC12 cells
2 Department of Microbiology and Immunology, Queen's University, Kingston, Ontario, Canada K7L3N6
3 Fundamental and Applied Molecular Biology, Ghent University, Flanders Interuniversity Institute for Biotechnology (V.I.B.), Technologiepark 927, B-9052 Gent-Zwijnaarde, Belgium
4 Laboratory of Neural Signaling, Cell Biology Group, Robarts Research Institute, London, Ontario, Canada N6A 5K8
1 To whom correspondence should be addressed; Tel: +1-613-533-2457; Fax: +1-613-533-6796; e-mail: szewczuk{at}post.queensu.ca
Received on October 26, 2006; revised on February 20, 2007; accepted on March 16, 2007
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Trypanosome trans-sialidase (TS) is a sialic acid-transferring enzyme and a novel ligand of tyrosine kinase (TrkA) receptors but not of neurotrophin receptor p75NTR. Here, we show that TS targets TrkB receptors on TrkB-expressing pheochromocytoma PC12 cells and colocalizes with TrkB receptor internalization and phosphorylation (pTrkB). Wild-type TS but not the catalytically inactive mutant TS
Asp98-Glu induces pTrkB and mediates cell survival responses against death caused by oxidative stress in TrkA- and TrkB-expressing cells like those seen with nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF). These same effects are not observed in Trk deficient PC12nnr5 cells, but are re-established in PC12nnr5 cells stably transfected with TrkA or TrkB, are partially blocked by inhibitors of tyrosine kinase (K-252a), mitogen-activated protein/mitogen-activated kinase (PD98059) and completely blocked by LY294002, an inhibitor of phosphatidylinositol 3-kinase (PI3K). Both TrkA- and TrkB-expressing cells pretreated with TS or their natural ligands are protected against cell death caused by serum/glucose deprivation or from hypoxia-induced neurite retraction. The cell survival effects of NGF and BDNF against oxidative stress are significantly inhibited by the neuraminidase inhibitor, Tamiflu. Together, these observations suggest that trypanosome TS mimics neurotrophic factors in cell survival responses against oxidative stress, hypoxia-induced neurite retraction and serum/glucose deprivation. |
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Nerve growth factor (NGF) is well known as a neuroprotective agent in a variety of in vitro neuronal insult models (Heaton et al. 1993
; Luo et al. 1997; Culmsee et al. 1999; Goins et al. 1999; Gouhier et al. 2000; Mitchell et al. 1999; Semkova and Krieglstein 1999; Ahlemeyer et al. 2000; Takman et al. 2004; Tabakman et al. 2005). In addition, brain-derived neurotrophic factor (BDNF) has been shown to be neuroprotective against neonatal hypoxic-ischemic brain injury in vivo (Han and Holtzman 2000). Likewise, estrogen is another potential neuroprotective agent against damage produced by acute and chronic injuries in the adult brain (Amantea et al. 2005). Insulin-like growth factors (IGF-I and -II) are also potent trophic factors for motor and sensory neurons and glial cells (Feldman et al. 1997). The actions of these growth factors are mediated through their binding to respective receptors. Growth factor binding to its receptor is necessary to activate distinct signaling cascades, which in turn mediate their trophic effects. There are intriguing observations suggesting that microbial invaders of the nervous system like Trypanosoma cruzi might utilize neuronal receptor(s) to reduce tissue damage for maximizing host–parasite equilibrium and produce relatively little brain or nerve damage in long-lasting infections (Chuenkova and Pereira 2000, 2001, 2003; Chuenkova et al. 2001). However, the neuronal receptor(s) targeted by T. cruzi parasites had not been identified until now.
Insight for the neuronal receptor(s) targeted by T. cruzi parasites came from our previous studies demonstrating for the first time that T. cruzi trans-sialidase (TS) binds NGF tyrosine kinase receptor-A (TrkA) receptors (Woronowicz et al. 2004). Consequently, TS hydrolyzes sialyl 
-2,3-linked ß-galactosyl residues of TrkA receptors sufficient to induce receptor internalization and activation, and to promote cell differentiation (neurite outgrowth) (Woronowicz et al. 2004). At the same time, Chuenkova and PereiraPerrin (2004) have also provided supporting evidence that T. cruzi parasite is able to bind TrkA via its neuraminidase in a NGF-inhibitable manner, which leads to TrkA autophosphorylation, and the activation of the phosphatidylinositol 3-kinase (PI3K), PI3K/Akt kinase pathway. This T. cruzi neuraminidase also leads to neuroprotection from apoptotic cell death caused by growth factor deprivation (Chuenkova and Pereira 2000). These survival responses involve signaling through the PI3K/Akt pathway (Chuenkova et al. 2001). For neurite outgrowth, T. cruzi neuraminidase-induced signaling is mediated via the mitagen-activated protein kinase (MAPK)/extracellular signal-regulated kinase ERK cascade (Chuenkova and Pereira 2001). They also showed that their T. cruzi neuraminidase does not react with the neurotrophin receptor p75NTR (Chuenkova and PereiraPerrin 2004). Together these findings demonstrate that T. cruzi TS is a novel and unique TrkA ligand, sufficient to promote cell differentiation (neurite outgrowth) and neuroprotection independent of NGF.
In the present study, the rat pheochromocytoma cell line PC12 and its Trk derivatives (Meakin et al. 1997, 1999; Meakin and MacDonald 1998; MacDonald et al. 2000) were used to further study the neuronal receptor involved in TS-mediated Trk activation and cell survival responses. We demonstrate that TS mimics BDNF, the natural ligand for TrkB receptors. We find that TS targets TrkB receptors and colocalizes with TrkB internalization and activation, which subsequently leads to survival responses against cell death caused by oxidative stress, serum/glucose deprivation or from hypoxia-induced neurite retraction. In addition, the survival responses induced by the neurotrophic effects of NGF and BDNF may be dependent on cellular sialidase(s) with similar properties like T. cruzi TS.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
T. cruzi TS binds to TrkB-expressing PC12 cells and colocalizes with phosphorylated TrkB receptors (pTyr490)
If T. cruzi TS is a novel ligand of TrkA receptors (Chuenkova and PereiraPerrin 2004; Woronowicz et al. 2004
) but not a ligand of neurotrophin receptor p75NTR (Chuenkova and PereiraPerrin 2004), we asked whether TS targets TrkB receptors as well. It is known that TrkB receptors exist as heavily glycosylated 145 and 95 kDa forms, and are distributed at the plasma membrane as type I membrane proteins with the N-terminus located extracellularly (Watson et al. 1999). Here, TrkB-nnr5 cells were treated with TS for 5, 15, 30, and 60 min time intervals or were left untreated as controls. Cells were fixed, permeabilized, and immunostained simultaneously with monoclonal mouse anti-E tag antibody specific for the E-tag (pCAGGS-TSE) sequence linked to the C-terminus domain of TS and polyclonal rabbit anti-pTyr490 antibody specific for pTrk followed by Alexa Fluor 488 anti-mouse immunoglobulin (Ig) G and Alexa Fluor 568 anti-rabbit IgG. Cells were visualized by a confocal inverted microscope (Leica TCS SP2 MP inverted Confocal Microscope) using a 100x objective (oil). The findings clearly show that TS binds to TrkB receptors and colocalizes with pTrkB 5 min after treatment. The percentage of overlay increased from 68 to 81% over the 60 min incubation period (Figure 1A).
|
To confirm TS-induced TrkB phosphorylation (pTrkB), TrkB-nnr5 cells pretreated with wild-type(wt) TS, catalytically inactive mutant TS
Asp98-Glu or BDNF in the presence or absence of Tamiflu were lyzed, and the lysates processed in a Western blot using goat anti-pTyr490 Trk antibody. The data clearly show that the wt TS but not the mutant TSAsp98-Glu induces pTrkB (Figure 1B). In addition, Tamiflu significantly inhibited BDNF-induced pTrkB compared to BDNF treated cells (Figure 1B). The blot of lysate from untreated control cells showed little pTrkB. Together, these results indicate that TS targets TrkB receptors and colocalizes with TrkB internalization and phosphorylation (pTrkB). It is noteworthy that during the 60 min TS treatment, the catalytic sialidase activity of TS specific for -2,3-linked sialic acids is the process involved in the activation of the receptor, and not the physical binding of TS to these cells (Woronowicz et al. 2004). We have also shown that BDNF binding to TrkB induces membrane sialidase activity in TrkB-expressing cells and cortical neurons, and this activity is blocked by the neuraminidase inhibitor, Tamiflu (Woronowicz et al. 2007). This process of activating membrane sialidase following neurotrophic growth factor stimulation requires Trk receptor. Using lectin blots, confocal microscopy colocalization, and lectin inhibition of NGF-induced TrkA activation, we identified a specific sialyl -2,3-linked ß-galactosyl residue of TrkA, which is rapidly targeted and hydrolyzed by the cellular sialidase(s) (Woronowicz et al. 2007).
NGF, BDNF, and T. cruzi TS mediate survival responses against oxidative stress and serum/glucose deprivation
Since TS treatment of TrkB-nnr5 (nnr5-TrkB) cells leads to Trk phosphorylation, we determined whether this TS-induced pTrkB mediates cell survival from death caused by oxidative stress. Here, TrkA- and TrkB-expressing cells were treated with either TS, NGF or BDNF for 24 h prior to exposure to 0.1 mM of hydrogen peroxide (H2O2) for 30 min. NGF, BDNF, and TS pretreatment of these cells mediated survival responses against cell death caused by oxidative stress (Figure 2A and C).
|
and 
) represent significant differences at 95% confidence using Bonferroni's multiple comparison test. G versus C, P < 0.01 for K-252a, P < 0.001 for PD98059, and P < 0.001 for LY294002; H versus D, P < 0.01 for K-252a, P < 0.01 for PD98059, and P < 0.001 for LY294002; G (PD98059) versus G (PD98059/LY294002), P < 0.001; H (PD98059) versus H (PD98059/LY294002), P > 0.05; G (K-252a) versus G (K-252a/LY294002), P < 0.01; and H (K-252a) versus H (K-252a /LY294002), P > 0.05.When TrkA-expressing PC12 cells and Trk-deficient PC12nnr5 cells were treated with wt TS, mutant TS
Asp98-Glu or NGF for 24 h prior to exposure to 0.1 mM of H2O2 for 30–60 min, the wt TS and NGF but not the catalytically inactive mutant TSAsp98-Glu-mediated cell survival responses against death caused by oxidative stress for each of the cell lines expressing TrkA receptors (Figure 2A and B). TS treatment of TrkA-deficient PC12nnr5 cells did not mediate neuroprotection against oxidative stress (Figure 2A and B). Neuroprotection against oxidative stress in TrkA-deficient PC12nnr5 cells was neither observed following treatment with NGF, wt TS or the mutant TSAsp98-Glu (Figure 2A and B).
To confirm that TrkA receptors may be involved in TS-induced neuroprotection against oxidative stress, cell viabilities were examined in TS-treated PC12nnr5 cells stably transfected with rat TrkA receptors (B2 and B5 cells). B5 cells overly express TrkA receptors by 100-fold compared to PC12 cells while B2 cells overly express TrkA receptors by 4-fold (Meakin et al. 1997; Meakin and MacDonald 1998). TS-treated B2 and B5 cells were protected against cell death caused by oxidative stress similar to those observed for NGF-treated cells (Figure 2A and B) and for TS-treated TrkB-nnr5 cells (Figure 2C). These findings suggest that TS-mediated neuroprotection in TrkA- and TrkB-expressing cells against cell death caused by oxidative stress requires Trk receptors.
When TrkA-PC12 cells were pretreated with K-252a, an inhibitor of tyrosine kinase, or with PD98059, an inhibitor of MAP/MEK protein kinases, the effects of both NGF- or TS-mediated cell survival responses against death caused by oxidative stress were significantly reduced but not completely blocked when compared to NGF or TS treatments without inhibitors (Figure 2D). These data suggest that TS-mediated neuroprotection from cell death due to oxidative stress partially involves the activation of tyrosine kinase and MAP/MEK protein kinase. When TrkA-PC12 cells were pretreated with LY294002, a specific inhibitor of PI3K, alone or in combination with K-252a or PD98059, the effects of both NGF- or TS-mediated neuronal cell survival responses due to oxidative stress were completely blocked when compared to NGF or TS treatments without inhibitors (Figure 2D). Taken together, these results suggest that TS-mediated neuroprotection against cell death due to oxidative stress of Trk-expressing cells partially involves Trk tyrosine kinase and MAP/MEK protein kinase pathways, but the major survival signaling pathway involves the PI3K.
If TS mediates cell survival responses from death caused by oxidative stress, we asked whether these same responses could be seen in cells exposed to serum/glucose deprivation. To test this, TrkA- and TrkB-expressing PC12nnr5 cells were pretreated with either wt TS, NGF, and BDNF or no pretreatment for 48 h. The cells were either maintained in medium containing 2.5% horse serum and 2.5% fetal calf serum (serum–glucose) as controls, or switched to medium without serum (no serum), or switched to medium without glucose (no glucose) for 24 h. Cell viability was assessed by acridine orange–ethidium bromide staining. For each of the cell lines expressing TrkA or TrkB receptors, the wt TS-mediated cell survival responses against death due to serum–glucose deprivation (Figure 3A and B). Furthermore, TS treatment of Trk-deficient PC12nnr5 cells did not promote neuroprotection against cell death due to serum–glucose deprivation (Figure 3C).
|
NGF and T. cruzi TS mediate neuroprotection against hypoxia-induced neurite retraction
Neurite-bearing TrkB-nnr5 (nnr5-TrkB) and nnr5-TrkA (B5) cells were first treated with the respective neurotrophic growth factors, wt TS or were left untreated for 4–7–days until extensive neurite networks were established. The cells were then exposed to 1% hypoxia for 6 h (transient hypoxic exposure) in a ProOx chamber. No statistically significant differences in neurite length were observed among neurite-bearing nnr5-TrkB or nnr5-TrkA cells supplemented with either respective growth factors, TS or maintained at standard atmospheric conditions (Figure 4). Likewise, no statistically significant changes in neurite length were observed among neurite-bearing nnr5-TrkB or B5 cells treated with either TS or growth factor following 1% hypoxia compared to those cells maintained at standard atmospheric condition (positive controls) (Figure 4). In contrast, neurite-bearing nnr5-TrkB and B5 cells supplemented with media alone prior to 1% hypoxic insult displayed significant (P < 0.001) neurite retraction in comparison with positive controls or with those cells treated with either TS or growth factor prior to hypoxic exposure (Figure 4). When PC12 cells were exposed to 1% hypoxia and glucose deprivation for 4 h, they showed no signs of cell death but a 24 h exposure revealed cell death in untreated cells and in Trk-deficient PC12nnr5 cells compared to TS- or NGF-pretreated groups (data not shown). Together, these results suggest that the neuroprotective effects of NGF, BDNF, and TS against hypoxia-induced neurite retraction and cell death require Trk receptor.
|
Tamiflu significantly inhibits NGF and BDNF-mediated neuroprotection against oxidative stress
When TrkA-PC12 and TrkB-nnr5 cells were pretreated with Tamiflu for 30 min, the effects of both NGF- or BDNF-mediated cell survival responses against death caused by oxidative stress were significantly reduced when compared to NGF or BDNF treatments without inhibitors (Figure 5). These results suggest that NGF and BDNF-mediated neuroprotection against oxidative stress may involve cellular sialidase(s).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Previous observations suggested that T. cruzi TS is a novel ligand of TrkA receptors (Chuenkova and PereiraPerrin 2004; Woronowicz et al. 2004
). Surprisingly, T. cruzi TS does not react with the neurotrophin receptor p75NTR (Chuenkova and PereiraPerrin 2004). These latter findings indicate that TS might be a selective ligand for TrkA receptors. Whether T. cruzi TS targets other Trk family of receptors such as TrkB has not been previously determined. In the present studies, we demonstrate that T. cruzi TS targets TrkB receptors on TrkB-expressing cells and colocalizes with TrkB internalization and phosphorylation (pTrkB). Furthermore, the catalytically inactive mutant TSAsp98-Glu is unable to induce pTrkB. It is noteworthy that the catalytic sialidase activity of TS specific for sialyl 2,3-linked ß-galactosyl residues is the process involved in the activation of the receptor, and not the physical binding of TS to these cells (Woronowicz et al. 2004). Together, these findings suggest that TS is not only a novel ligand for TrkA and TrkB receptors, but its sialidase activity plays an important role in Trk receptor internalization and activation.
Using the PC12 cell line and its Trk derivatives, we demonstrate that T. cruzi TS is a neuroprotective agent capable of mimicking the activity of NGF and BDNF in mediating cell survival responses. When TrkA-PC12 cells were pretreated with K-252a, an inhibitor of tyrosine kinase, or with PD98059, an inhibitor of MAP/MEK protein kinases, both NGF- or TS-mediated cell survival responses against oxidative stress were significantly reduced but not completely blocked (see Figure 2D). These data suggest that TS-mediated neuroprotection from cell death due to oxidative stress partially involves the activation of tyrosine kinase and MAP/MEK protein kinase. When TrkA-PC12 cells were pretreated with LY294002, a specific inhibitor of PI3K, alone or in combination with K-252a or PD98059, the effects of both NGF- or TS-mediated neuronal cell survival responses were completely blocked. Together, these observations suggest that TS-mediated neuroprotection against oxidative stress in TrkA-expressing PC12 cells partially involves TrkA tyrosine kinase and MAP/MEK protein kinase pathways, but the major cell survival signaling pathway involves the PI3K. In addition, the wt TS but not the catalytically inactive mutant TSAsp98-Glu is able to mediate cell survival responses against death caused by oxidative stress in TrkA- and TrkB-expressing cells. These same effects are not observed in Trk-deficient PC12nnr5 cells, but are re-established in PC12nnr5 cells stably transfected with TrkA or TrkB. If TS mediates cell survival responses from death caused by oxidative stress, we asked whether cell survival responses could be seen in TrkA- and TrkB-expressing PC12 cells exposed to serum/glucose deprivation. Indeed, TS mediates cell survival responses against death caused by serum/glucose deprivation in Trk-expressing PC12 cells. In contrast, these effects are not observed in TrkA-deficient PC12nnr5 cells.
Chuenkova and Pereira (2000) have also shown that TS-mediated neuronal cell survival responses to growth factor deprivation are triggered by signaling through the PI3K/Akt pathway (Chuenkova et al. 2001) where TS activates Akt in a PI3K-dependent fashion. For neurite outgrowth, TS activates the MAPK/ERK cascade (Chuenkova and Pereira 2001). Previously, we have shown that TS targets TrkA receptors and hydrolyzes sialyl -2,3-linked ß-galactosyl residues (Woronowicz et al. 2004). This process facilitates receptor dimerization and subsequent phosphorylation of Trk leading to downstream activation of effector proteins such as MAPK. In support of these observations, using inhibitors of tyrosine kinase and MAP/MEK protein kinase, both NGF- and TS-induced neurite outgrowth in TrkA-expressing PC12 cells were completely blocked (Woronowicz et al. 2004).
The rat PC12 cells are a common dopaminergic neuronal model for in vitro studies of cell death (Fujita et al. 1989). PC12 cells were originally cloned from a solid pheochromocytoma tumor passaged subcutaneously in New England Deaconess Hospital Strain white rats (Greene and Tischler 1976). They have served as cellular models of serum starvation (Rukenstein et al. 1991), NGF deprivation (Park et al. 1998), excitotoxicity (Menei et al. 2000), cytotoxicity (Fischer et al. 2001), Parkinson's disease (Shimoke and Chiba, 2001), Alzheimer's disease (Troy et al. 2001), and ischemia (Pan et al. 1997; Tabakman et al. 2002). Tabakman et al. (2005) using an oxygen-glucose-deprivation device and PC12 cells exposed to ischemic insult provided support for the concept that MAPKs are activated during ischemia, stress and inflammation, and they may be the crucial cross points in the neurotoxicity/neuroprotection process. It may be that the MAPK signaling pathways used by neurotrophins during retrograde signaling differ from those used following direct stimulation of the cell (Watson et al. 2001). During retrograde signaling, endocytosed Trk receptors have been shown to activate the Erk5 pathway, leading to nuclear translocation of Erk5, phosphorylation of cAMP-responsive element binding protein, and enhanced neuronal survival (Watson et al. 2001). In contrast, Erk1/2 which mediates nuclear responses following direct cell body stimulation does not transmit a retrograde signal.
Neurite-bearing TrkA- and TrkB-expressing PC12 cells exhibit extensive neurite retraction and considerable thinning of neuronal connections when these cells are subjected to transient hypoxic insult (1% oxygen) (see Figure 4). In addition, hypoxia-induced cell death of naïve TrkA-PC12 cells can occur after 24 h exposure to this hypoxic insult (data not shown). In another study, Glass et al. (2002) examined the effects of graded hypoxia on mixed neuronal and astrocytic cultures established from fetal rat brains. They found that neuronal processes were not completely lost until after 24 h of hypoxic insult (1.8% oxygen), while neurite retraction in the present study occurred after only 6 h of hypoxia exposure. Differences in the hypoxic-sensitivities of the neuronal cells used may be the likely the cause of this discrepancy. Overall, our findings suggest that TS can protect neurite-bearing TrkA- and TrkB-expressing cells from hypoxia-induced neurite retraction similarly like that for NGF and BDNF.
A direct link between receptor glycosylation and activation following natural ligand interaction has not been previously observed until now. We discovered a membrane sialidase controlling mechanism that depends on ligand binding to its receptor to induce enzyme activity which targets and desialylates the receptor and, consequently, causes the induction of receptor dimerization and activation (Woronowicz et al. 2007). Also, we demonstrated that both Trk-expressing cells and primary cortical neurons following stimulation with specific neurotrophic growth factors express a vigorous membrane sialidase activity which is blocked by neuraminidase inhibitors, Tamiflu, BCX1812, and BCX1827 (Woronowicz et al. 2007). In the present studies, we asked if Tamiflu would have an inhibitory effect on NGF- or BDNF-mediated cell survival responses against death caused by oxidative stress. In fact, Tamiflu significantly inhibited NGF- and BDNF-mediated cell survival responses against oxidative stress. These observations suggest that NGF- and BDNF-mediated neuroprotection may involve cellular sialidase(s) as described previously (Woronowicz et al. 2007). The exact nature of the cellular sialidase(s) induced by these neurotrophic factors binding to their respective Trk receptors remains unknown.
There is considerable evidence indicating Trk and monosialoganglioside-1 (GM1) interaction. It has been shown that GM1 ganglioside interaction with TrkA receptor is required for receptor dimerization, phosphorylation, and signal transduction pathway activation (Mutoh et al. 1995; Rabin and Mocchetti 1995; Kimura et al. 2001; Duchemin et al. 2002). Actually, GM1 was found to interact specifically with glycosylated Trk and not unglycosylated Trk (Mutoh et al. 2000). A similar association was not observed with the low-affinity, highly glycosylated NGF-receptor p75NGR and epidermal growth factor receptor (Mutoh et al. 1995). When the ceramide analog, D-threo-1-phenyl-2-decanoylamin-3-morpholino-propanol (D-PDMP), an inhibitor of glucosylceramide synthase leading to extensive depletion of glycosphingolipids derived from glucosyl ceramide is applied to PC12 cells, NGF-induced autophosphorylation of Trk and neurite outgrowth was inhibited (Mutoh et al. 1998). These effects of D-PDMP were reversed by the addition of exogenous GM1 and appeared to be specific for the Trk receptor. These findings suggest that Trk requires endogenous GM1 gangliosides for its normal function in mediating the neurotrophic activity of NGF (Mutoh et al. 1998). Using NG-CR72 cells that are deficient in endogenous GM1 due to a mutation of GM1 synthase gene, Trk was expressed in these NG-CR72 cells but its location appeared not to be on the plasma membrane (Mutoh et al. 2002). As predicted, NGF could not induce Trk phosphorylation in these GM1 deficient NG-CR72 cells (Mutoh et al. 2002). These findings suggest that GM1 may be necessary for the normal expression of the Trk function and for normal targeting of the Trk receptor to the plasma membrane (Mutoh et al. 2002).
GM1 is known to be a major constituent of caveola or glycosphingolipid-enriched microdomain of the plasma membrane. GM1 is also the product of the plasma membrane ganglioside sialidase Neu3 (PMGS) (Miyagi and Tsuiki 1986; Kopitz et al. 1998). Overexpression of PMGS in primary neurons produces a major increase in phosphorylated TrkA (pTrkA), which is further enhanced by NGF (da Silva et al. 2005). When PMGS is overexpressed in neurons, it interacts and can be precipitated with pTrkA, especially after NGF stimulation (da Silva et al. 2005). These results indicate that PMGS activation could lead to localized activity of Trk receptors but their downstream effects may be directed only to pTrkA. A recent report indicated that Neu3 short hairpin RNA silenced Neuro2a murine neuroblastoma cells exhibited an increase in GM2 by 54%, no change in GM3 content, and a decrease in GM1 and GD1a by 66 and 50%, respectively (Valaperta et al. 2007). Surprisingly, a reduction of 70% of the plasma membrane-associated sialidase Neu3 activity, due to the Neu3 short hairpin RNA silencing, caused neurite elongation in these Neuro2a cells. Furthermore, da Silva et al. (2005) have demonstrated that PMGS-targeting iRNA interfered with axon generation in that 95.2% of the cells did not present tau polarization; however, the PMGS iRNA transfection did not affect the length of the minor neurites. Furthermore, neurons pretreated for 48 h with the specific PMGS inhibitor, 2-deoxy-2,3-dehydro-N-acetylneuraminic acid (NeuAc2en or DANA), remained morphologically unpolarized. The overall observations by da Silva et al. (2005) suggest that a spatially restricted threshold level of PMGS activity to the tip of one neurite specifies a single neurite to become the axon. Moreover, Rodriguez et al. (2001) were able to inhibit the activity of PMGS in cultured hippocampal neurons with the specific inhibitor NeuAc2en. Their NeuAc2en treatment of cells reduced the number of axons from 31 (untreated) to 7%, whereas the number of short neurites was enhanced from 41 (control) to 72%. These observed effects of NeuAc2en were caused by a true reduction of surface GM1 as evidenced by cholera toxin subunit B (ChTx-B) binding and immunofluorescence (Rodriguez et al. 2001). In fact, PMGS overexpression in these neurons resulted in the appearance of numerous neurons with one or two very long axon-like neurites (>40 µm) from 7 (untransfected) to 37%, and the number of neurites of intermediate length was strongly decreased (26% for overexpressing cells and 52% for controls). Together, these findings suggest the important involvement of PMGS in axonal growth.
The findings in the present studies suggest that T. cruzi TS is able to bind Trk family of tyrosine kinase receptors, which leads to Trk autophosphorylation and promotes neuroprotective signaling through the PI3K, PI3K/Akt kinase pathway. This T. cruzi TS acts as a neuroprotective agent mimicking NGF and BDNF trophic effects in a variety of in vitro neuronal insult models such as oxidative stress, serum–glucose deprivation, and hypoxia-induced neurite retraction.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Recombinant T. cruzi TS
The recombinant T. cruzi TS and catalytically inactive mutant TS
Asp98-Glu used in these experiments were isolated using the methylotrophic yeast Pichia pastoris system, and purified using an affinity anti-E-tag antibody column (Laroy and Contreras 2000). Both the wt TS and mutant TSAsp98-Glu have the E-tag (pCAGGS-TSE) sequence linked to the C-terminus domain (Laroy and Contreras 2000). The mutant TSAsp98-Glu has a point mutation in the catalytic domain of TS reducing the sialidase and sialyltransferase activities to approximately 3.6 and 4.5%, respectively, of the wt TS (Woronowicz et al. 2004). The optimal dose of 200 ng TS/mL has been predetermined for Trk activation in TrkA-PC12 cells (Woronowicz et al. 2004).
NGF and inhibitors
NGF (50 ng/mL) and BDNF (100 ng/mL) (Sigma, St. Louis, MO) were used at predetermined optimal dosage. K-252a (Calbiochem), inhibitor of Trk tyrosine kinase, PD98059 (Calbiochem), inhibitor of MAP/MEK protein kinase and LY294002 (Sigma), specific inhibitor of PI3K, were used at predetermined optimal dosage of 20 µM for PD98059, 200 nM for K-252a, and 100 µM for LY294002. Tamiflu (Pure oseltamivir phosphate, Hoffmann-La Roche Ltd., Mississauga, Ontario, Lot # BS00060168) was used at indicated concentrations.
Cell lines
The NGF responsive PC12 rat pheochromocytoma cell line, PC12nnr5 (Trk-deficient mutant, NGF nonresponsive) (Kaplan and Miller 1997, 2000), TrkA-PC12 cell line (overly expressing human TrkA receptors) (Hempstead et al. 1992), and the PC12nnr5 cell lines stably transfected with rat TrkA receptors (B5 and B2) (Meakin and MacDonald 1998) were used in these studies as previously described (Woronowicz et al. 2004). The nnr5-TrkB cells are PC12nnr5 cells stably transfected to express rat TrkB receptors (Baskey et al. 2002). In the presence of BDNF, these cells respond by stopping division and extending neurites which are similar to the activity of NGF on PC12 and B5 cells. All cell lines are grown at 37 oC in 5% CO2 in culture media containing Dulbeceo's modified Eagle's medium (Gibco, Rockville, MD) supplemented with 5% horse serum and 3% fetal bovine serum (Gibco).
Cell viability
Cells were treated with 200 ng of TS/mL, 50 ng of NGF/mL or 100 ng of BDNF/mL for 15 min or 24 h prior to exposure to 0.1 mM H2O2 for 30–60 min or serum–glucose deprivation for 24 h. Cell viability was assessed by acridine orange–ethidium bromide staining and epi-fluorescence microscopy. The percentage of viable cells was determined for total cell counts of 
200.
Scoring of neurite outgrowth
Cells (
5 x 106/glass slide) were grown in 24-well tissue culture plates on 12 mm circular glass slides coated with poly-D-lysine (Sigma) for 24 h prior to experimental treatment. After 2–3 days of treatment, the cells were fixed in 3.7% paraformaldehyde for 30 min, permeabilized with 0.2% Triton X-100 in phosphate-buffered saline (PBS) on ice for 5 min followed with a blocking solution of 3% bovine serum albumin (BSA) on ice for 20 min. The cells were then stained with phalloidin-rhodamine (Sigma) for 1 h at 37 °C. Cell staining was analyzed qualitatively using epi-fluorescence microscopy and quantitatively by counting the number of cells with multiple neurite extensions (200 of the total cells; one or more extensions >2 cell bodies in length).
TS colocalization with pTrkB receptors
The TrkB-nnr5 cells were treated with 200 ng/mL of wt TS for 5, 15, 30, and 60 min or left untreated as controls. Cells were fixed, permeabilized, and immunostained simultaneously with monoclonal mouse anti-E tag antibody specific for E-tag (pCAGGS-TSE) sequence linked to the C-terminus domain of TS and with polyclonal rabbit anti-phospho-Tyr490 antibody (antibody specific for phosphorylated Tyr490 of Trk (ab1445; Abcam Limited, Cambridgeshire, UK) followed by anti-mouse Alexa Fluor 488 (green) and anti-rabbit Alexa Fluor 568 (red). Cells were visualized using a confocal inverted microscope (Leica TCS SP2 MP inverted Confocal Microscope, Mannheim, Germany) using a 100x objective (oil). Images are captured using a z-stage of 8–10 images per cell at 0.5 µm steps, and are processed and merged using LCS Lite software.
Western blot
TrkB-nnr5 cells were grown in 25 cm2 flasks at 90% confluence. Cells were treated with 500 µM Tamiflu in serum free medium for 24 h or left untreated. Cells were stimulated with either 100 ng of BDNF/mL for 15 min and immediately lyzed in 500 µL of 1x Sodium dodecyl sulfate (SDS) sample buffer. In addition, cells were stimulated with 200 ng/mL of wt TS or 200 ng/mL of mutant TSAsp98-Glu for 60 min and immediately lyzed in 500 of 1x SDS sample buffer. To each of the cell lysates was added 3% 2-mercaptoethanol and boiled for 10 min. Cell lysates were resolved by 8% SDS–polyacrylamide gel electrophoresis (PAGE) gel, and the blots probed with either polyclonal goat anti-pY490 Trk antibody or polyclonal goat anti-pan Trk antibody followed by horseradish-peroxide (HRP) conjugated secondary anti-goat IgG antibody, and Western Lightning Chemiluminescence Reagent Plus. Sample concentration for gel loading was determined by Bradford reagent. Quantitative analysis was done by assessing the density of 145 kDa band corrected for background in each lane using Corel Photo Paint 8.0 software. Each bar in the figures represents the mean corrected density of 145 kDa band ± SEM for 6–8 replicate measurements (n) within each lane. Significant differences at 95% confidence using Bonferroni's multiple comparison test.
ProOx hypoxia chamber and neurite retraction
Cells were grown on poly-D-lysine-coated circular glass cover slides in 24-well tissue culture plates for 24 h. They were then stimulated with respective neurotrophin factors for 4–7 days to establish neurite outgrowth, or were left untreated. For nnr5-TrkB cells, they were stimulated with BDNF (100 ng/mL) while nnr5-TrkA (B5) cells with NGF (50 ng/mL). Both nonstimulated and stimulated nnr5-TrkA and nnr5-TrkB cells were supplemented with either the respective growth factors, TS (200 ng/mL) or media for 24 h prior to transient hypoxic insult or maintained in normoxic incubators as controls. Hypoxic conditions were established for 6 h in humidified, ProOx chambers set to 1% O2 (with 5% CO2 and the balance N2) at 37 °C. After 6 h hypoxic treatment, cells were fixed, permeabilized with 0.2% TritonX-100, blocked with 3% BSA in PBS, and followed by staining with phalloidin-rhodamine for 1 hr at 37 °C. Cell staining was analyzed qualitatively using fluorescence microscopy (Aviovert 100A DIC imaging) with a 40x objective. Analysis of neurite retraction was accomplished using the ruler function of the Aviovert 100A DIC Imaging software and neurite length was measured in microns.
Statistics
Comparisons between two groups were made by one-way ANOVA at 95% confidence using Bonferroni's multiple comparison test, or Dunnett's multiple comparison test for comparisons among more than two groups.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
The authors declare that they have no competing financial interests.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
The authors thank Dr Emilia F. Kazmierczak of the Department of Biochemistry and Jeff Mewburn of the Cancer Biology and Genetics Confocal Microscopy, Queen's University, for their expert technical assistance. Special thanks to Dr Katrina Gee of this department for an extensive review of the manuscript. A.W. is a recipient of a National Science and Engineering Research Council of Canada (NSERC) Scholarship. S.R.A. is a recipient of the Queen's University Research Award and the Robert J. Wilson Fellowship. P.J. is a recipient of the Queen's Graduate Award. These studies are partially supported by grants from NSERC, the Harry Botterell Foundation for Neuroscience Research, ARC, and Garfield Kelly Cardiovascular Research and Development Fund. The research in Ghent, Belgium is supported by GOA and FWO Flanders.
|
Abbreviations |
|---|
B2 cell line, PC12nnr5 stably transfected with TrkA receptors and expresses 4-fold more TrkA receptors than that found in PC12 cells; B5 cell line, PC12nnr5 stably transfected with TrkA receptors and expresses 100-fold more; BDNF, brain-derived neurotrophic factor; ERK, extracellular signal-regulated kinase; GM, monosialoganglioside; HRP, horseradish-peroxidase; Ig, immunoglobulin; K252a, an inhibitor of tyrosine kinase; LY294002, specific inhibitor of PI3K; MAP, mitogen-activated protein; MEK, mitogen activated kinase; NGF, nerve growth factor; PBS, phosphate-buffered saline; PC12, rat pheochromocytoma cell line; PC12nnr5, a tyrosine kinase deficient mutant (NGF nonresponsive); PD98059, MAP/MEK protein kinase inhibitor; PI3K, phosphatidylinositol 3-kinase; PMSF, phenylmethanesulfonyl fluoride; SDS-PAGE, sodium dodecyl sulfate-polacrylamide gel eletrophoresis; TrkA, tyrosine kinase receptor A; Trka-PC12, PC12 cells overly expressing human TrkA receptors; TrkB-nnr5 or nnr5-TrkB cell line, PC12nnr5 stably transfected with TrkB receptors; MAPK, mitogen-activated protein kinase; TS, trans-sialidase; TS-As
98-Glu, mTS with glutamic acid residue at position 98 replacing aspartic acid; wt, wild-type. |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Ahlemeyer B, Huhne R, Krieglstein J. Retinoic acid potentiated the protective effect of NGF against staurosporine-induced apoptosis in cultured chick neurons by increasing the trkA protein expression. J Neurosci Res. (2000) 60:767–778.[CrossRef][ISI][Medline]
Amantea D, Russo R, Bagetta G, Corasaniti MT. From clinical evidence to molecular mechanisms underlying neuroprotection afforded by estrogens. Pharmacol Res. (2005) 52:119–132.[CrossRef][ISI][Medline]
Baskey JC, Kalisch BE, Davis WL, Meakin SO, Rylett RJ. PC12(nnr5) cells expressing TrkA receptors undergo morphological but not cholinergic phenotypic differentiation in response to nerve growth factor. J Neurochem. (2002) 80:501–511.[CrossRef][ISI][Medline]
Chuenkova MV, Furnari FB, Cavenee WK, Pereira MA. Trypanosoma cruzi trans-sialidase: a potent and specific survival factor for human Schwann cells by means of phosphatidylinositol 3-kinase/Akt signaling. Proc Natl Acad Sci USA (2001) 98:9936–9941.
Chuenkova MV, Pereira MA. A trypanosomal protein synergizes with the cytokines ciliary neurotrophic factor and leukemia inhibitory factor to prevent apoptosis of neuronal cells. Mol Biol Cell. (2000) 11:1487–1498.
Chuenkova MV, Pereira MA. The T. cruzi trans-sialidase induces PC12 cell differentiation via MAPK/ERK pathway. Neuroreport (2001) 12:3715–3718.[CrossRef][ISI][Medline]
Chuenkova MV, Pereira MA. PDNF, a human parasite-derived mimic of neurotrophic factors, prevents caspase activation, free radical formation, and death of dopaminergic cells exposed to the Parkinsonism-inducing neurotoxin MPP+. Brain Res Mol Brain Res. (2003) 119:50–61.[Medline]
Chuenkova MV, Pereira M. Chagas' disease parasite promotes neuron survival and differentiation through TrkA nerve growth factor receptor. J Neurochem. (2004) 91:385–394.[CrossRef][ISI][Medline]
Culmsee C, Semkova I, Krieglstein J. NGF mediates the neuroprotective effect of the beta2-adrenoceptor agonist clenbuterol in vitro and in vivo: evidence from an NGF-antisense study. Neurochem Int. (1999) 35:47–57.[CrossRef][ISI][Medline]
da Silva JS, Hasegawa T, Miyagi T, Dotti CG, bad-Rodriguez J. Asymmetric membrane ganglioside sialidase activity specifies axonal fate. Nat Neurosci. (2005) 8:606–615.[CrossRef][ISI][Medline]
Duchemin AM, Ren Q, Mo L, Neff NH, Hadjiconstantinou M. GM1 ganglioside induces phosphorylation and activation of Trk and Erk in brain. J Neurochem. (2002) 81:696–707.[CrossRef][ISI][Medline]
Feldman EL, Sullivan KA, Kim B, Russell JW. Insulin-like growth factors regulate neuronal differentiation and survival. Neurobiol Dis. (1997) 4:201–214.[CrossRef][ISI][Medline]
Fischer SJ, Podratz JL, Windebank AJ. Nerve growth factor rescue of cisplatin neurotoxicity is mediated through the high affinity receptor: studies in PC12 cells and p75 null mouse dorsal root ganglia. Neurosci Lett. (2001) 308:1–4.[CrossRef][ISI][Medline]
Fujita K, Lazarovici P, Guroff G. Regulation of the differentiation of PC12 pheochromocytoma cells. Environ Health Perspect. (1989) 80:127–142.[ISI][Medline]
Glass TF, Reeves B, Sharp FR. Modeling both the mechanical and hypoxic features of traumatic brain injury in vitro in rats. Neurosci Lett. (2002) 328:133–136.[CrossRef][ISI][Medline]
Goins WF, Lee KA, Cavalcoli JD, O'Malley ME, DeKosky ST, Fink DJ, Glorioso JC. Herpes simplex virus type 1 vector-mediated expression of nerve growth factor protects dorsal root ganglion neurons from peroxide toxicity. J Virol. (1999) 73:519–532.
Gouhier C, Chalon S, Venier-Julienne MC, Bodard S, Benoit J, Besnard J, Guilloteau D. Neuroprotection of nerve growth factor-loaded microspheres on the D2 dopaminergic receptor positive-striatal neurones in quinolinic acid-lesioned rats: a quantitative autoradiographic assessment with iodobenzamide. Neurosci Lett. (2000) 288:71–75.[CrossRef][ISI][Medline]
Greene LA, Tischler AS. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. PNAS (1976) 73:2424–2428.
Han BH, Holtzman DM. BDNF protects the neonatal brain from hypoxic-ischemic injury in vivo via the ERK pathway. J Neurosci. (2000) 20:5775–5781.
Heaton MB, Paiva M, Swanson DJ, Walker DW. Modulation of ethanol neurotoxicity by nerve growth factor. Brain Res. (1993) 620:78–85.[CrossRef][ISI][Medline]
Hempstead BL, Rabin SJ, Kaplan L, Reid S, Parada LF, Kaplan DR. Overexpression of the trk tyrosine kinase rapidly accelerates nerve growth factor-induced differentiation. Neuron (1992) 9:883–896.[CrossRef][ISI][Medline]
Kaplan DR, Miller FD. Signal transduction by the neurotrophin receptors. Curr Opin Cell Biol. (1997) 9:213–221.[CrossRef][ISI][Medline]
Kaplan DR, Miller FD. Neurotrophin signal transduction in the nervous system. Curr Opin Neurobiol. (2000) 10:381–391.[CrossRef][ISI][Medline]
Kimura M, Hidari KI, Suzuki T, Miyamoto D, Suzuki Y. Engagement of endogenous ganglioside GM1a induces tyrosine phosphorylation involved in neuron-like differentiation of PC12 cells. Glycobiology (2001) 11:335–343.
Kopitz J, von RC, Burchert M, Cantz M, Gabius HJ. Galectin-1 is a major receptor for ganglioside GM1, a product of the growth-controlling activity of a cell surface ganglioside sialidase, on human neuroblastoma cells in culture. J Biol Chem. (1998) 273:11205–11211.
Laroy W, Contreras R. Cloning of Trypanosoma cruzi trans-sialidase and expression in Pichia pastoris. Protein Expr Purif. (2000) 20:389–393.[CrossRef][ISI][Medline]
Luo J, West JR, Pantazis NJ. Nerve growth factor and basic fibroblast growth factor protect rat cerebellar granule cells in culture against ethanol-induced cell death. Alcohol Clin Exp Res. (1997) 21:1108–1120.[CrossRef][ISI][Medline]
MacDonald JI, Gryz EA, Kubu CJ, Verdi JM, Meakin SO. Direct binding of the signaling adapter protein Grb2 to the activation loop tyrosines on the nerve growth factor receptor tyrosine kinase, TrkA. J Biol Chem. (2000) 275:18225–18233.
Meakin SO, Gryz EA, MacDonald JI. A kinase insert isoform of rat TrkA supports nerve growth factor-dependent cell survival but not neurite outgrowth. J Neurochem. (1997) 69:954–967.[ISI][Medline]
Meakin SO, MacDonald JI. A novel juxtamembrane deletion in rat TrkA blocks differentiative but not mitogenic cell signaling in response to nerve growth factor. J Neurochem. (1998) 71:1875–1888.[ISI][Medline]
Meakin SO, MacDonald JI, Gryz EA, Kubu CJ, Verdi JM. The signaling adapter FRS-2 competes with Shc for binding to the nerve growth factor receptor TrkA. A model for discriminating proliferation and differentiation. J Biol Chem. (1999) 274:9861–9870.
Menei P, Pean JM, Nerriere-Daguin V, Jollivet C, Brachet P, Benoit JP. Intracerebral implantation of NGF-releasing biodegradable microspheres protects striatum against excitotoxic damage. Exp Neurol. (2000) 161:259–272.[CrossRef][ISI][Medline]
Mitchell JJ, Paiva M, Walker DW, Heaton MB. BDNF and NGF afford in vitro neuroprotection against ethanol combined with acute ischemia and chronic hypoglycemia. Dev Neurosci. (1999) 21:68–75.[CrossRef][ISI][Medline]
Miyagi T, Tsuiki S. Evidence for sialidase hydrolyzing gangliosides GM2 and GM1 in rat liver plasma membrane. FEBS Lett. (1986) 206:223–228.[CrossRef][ISI][Medline]
Mutoh T, Hamano T, Tokuda A, Kuriyama M. Unglycosylated Trk protein does not co-localize nor associate with ganglioside GM1 in stable clone of PC12 cells overexpressing Trk (PCtrk cells). Glycoconj J. (2000) 17:233–237.[CrossRef][ISI][Medline]
Mutoh T, Hamano T, Yano S, Koga H, Yamamoto H, Furukawa K, Ledeen RW. Stable transfection of GM1 synthase gene into GM1-deficient NG108-15 cells, CR-72 cells, rescues the responsiveness of Trk-neurotrophin receptor to its ligand, NGF. Neurochem Res. (2002) 27:801–806.[CrossRef][ISI][Medline]
Mutoh T, Tokuda A, Inokuchi J, Kuriyama M. Glucosylceramide synthase inhibitor inhibits the action of nerve growth factor in PC12 cells. J Biol Chem. (1998) 273:26001–26007.
Mutoh T, Tokuda A, Miyadai T, Hamaguchi M, Fujiki N. Ganglioside GM1 binds to the Trk protein and regulates receptor function. Proc Natl Acad Sci USA (1995) 92:5087–5091.
Pan Z, Sampath D, Jackson G, Werrbach-Perez K, Perez-Polo R. Nerve growth factor and oxidative stress in the nervous system. Adv Exp Med Biol. (1997) 429:173–193.[ISI][Medline]
Park DS, Morris EJ, Stefanis L, Troy CM, Shelanski ML, Geller HM, Greene LA. Multiple pathways of neuronal death induced by DNA-damaging agents, NGF deprivation, and oxidative stress. J Neurosci. (1998) 18:830–840.
Rabin SJ, Mocchetti I. GM1 ganglioside activates the high-affinity nerve growth factor receptor trkA. J Neurochem. (1995) 65:347–354.[ISI][Medline]
Rodriguez JA, Piddini E, Hasegawa T, Miyagi T, Dotti CG. Plasma membrane ganglioside sialidase regulates axonal growth and regeneration in hippocampal neurons in culture. J Neurosci. (2001) 21:8387–8395.
Rukenstein A, Rydel RE, Greene LA. Multiple agents rescue PC12 cells from serum-free cell death by translation- and transcription-independent mechanisms. J Neurosci. (1991) 11:2552–2563.[Abstract]
Semkova I, Krieglstein J. Neuroprotection mediated via neurotrophic factors and induction of neurotrophic factors. Brain Res Brain Res Rev. (1999) 30:176–188.[CrossRef][Medline]
Shimoke K, Chiba H. Nerve growth factor prevents 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced cell death via the Akt pathway by suppressing caspase-3-like activity using PC12 cells: relevance to therapeutical application for Parkinson's disease. J Neurosci Res. (2001) 63:402–409.[CrossRef][ISI][Medline]
Tabakman R, Jiang H, Shahar I, Arien-Zakay H, Levine RA, Lazarovici P. Neuroprotection by NGF in the PC12 in vitro OGD model: involvement of mitogen-activated protein kinases and gene expression. Ann NY Acad Sci. (2005) 1053:84–96.
Tabakman R, Lazarovici P, Kohen R. Neuroprotective effects of carnosine and homocarnosine on pheochromocytoma PC12 cells exposed to ischemia. J Neurosci Res. (2002) 68:463–469.[CrossRef]