- Top of page
- Materials and methods
In β-amyloid (Aβ)-induced neurotoxicity, activation of the NMDA receptor, increased Ca2+ and oxidative stress are intimately associated with neuronal cell death as normally seen in NMDA-induced neurotoxicity. We have recently shown selective sparing of somatostatin (SST)-positive neurons and increased SST expression in NMDA agonist-induced neurotoxicity. Accordingly, the present study was undertaken to determine the effect of Aβ25–35-induced neurotoxicity on the expression of SST in cultured cortical neurons. Cultured cortical cells were exposed to Aβ25–35 and processed to determine the cellular content and release of SST into medium by radioimmunoassay and SST mRNA by RT-PCR. Aβ25–35 induces neuronal cell death in a concentration- and time-dependent fashion, increases SST mRNA synthesis and induces an augmentation in the cellular content of SST. No significant changes were seen on SST release at any concentration of Aβ25–35 after 24 h of treatment. However, Aβ25–35 induces a significant increase of SST release into medium only after 12 h in comparison with other time points. Most significantly, SST-positive neurons are selectively spared in the presence of a lower concentration of Aβ25–35, whereas, in the presence of higher concentrations of Aβ25–35 for extended time periods, SST-positive neurons decrease gradually. Furthermore, Aβ25–35 induces apoptosis at lower concentrations (5 and 10 μmol/L) and necrosis at higher concentrations (20 and 40 μmol/L). Consistent with the increased accumulation of SST, these data suggest that Aβ25–35 impairs cell membrane permeability. Selective sparing of SST-positive neurons at lower concentrations of Aβ25–35 at early time points directly correlates with the pathophysiology of Alzheimer’s disease.
Somatostatin (SST), a multifunctional peptide, was first described as a neurotransmitter and neuromodulator (Reichlin 1983). Increasing evidence suggests that disturbances of the somatostatinergic system are involved in several human neurodegenerative diseases including Alzheimer’s disease (AD), Parkinson’s disease, Huntington’s disease (HD), hypoxic–ischemic neuronal injury, neoplasia, prion disease and acquired immune-deficient syndrome encephalopathy (Agren and Lundqvist 1984; Beal et al. 1985; Vecsei et al. 1990; Banki et al. 1992; Krantic et al. 1992). In AD brain tissue, the elimination of SST in the cortex and hippocampus is directly linked to poor cognitive function and memory impairment (Davies et al. 1980; Grouselle et al. 1998). Furthermore, reduced concentration of SST in AD is frequently observed in cortical and cerebrospinal fluid (Davies et al. 1980; Beal et al. 1988; Bisette et al. 1998; Davis et al. 1999; Nemeroff 1999). In addition, SST also decreases in a age-dependent manner (Hayashi et al. 1997; Lu et al. 2004). Unlike in AD, SST levels in HD are increased in striatal and basal ganglia and intimately associated with motor coordination disability in Huntington’s chorea (Beal et al. 1986). Also subsets of SST-producing neurons are selectively resistant in NMDA agonist-induced neurodegeneration (Dawbarn et al. 1985; Ferrante et al. 1987; Kumar 2004). In parallel, it has also been shown that SST exerts a neuroprotective effect against NMDA-mediated toxicity (Forloni et al. 1997).
Previous studies have shown that β-amyloid (Aβ) increases the vulnerability of cultured cortical neurons to excitotoxic damage through glutamatinergic pathways. The increased accumulation of Aβ42, a cleavage product of β-amyloid precursor protein that accumulates into fibrils in AD brains, is neurotoxic and contributes to dementia (Yankner et al. 1990). Most in vitro data suggest that the amyloid fibril is the neurotoxic form of Aβ, whereas soluble Aβ and non-fibrillar aggregates are non-toxic (Lorenzo and Yankner 1994). In fact, it has been shown that, in culture, the Aβ protein is neurotrophic to undifferentiated hippocampal neurons at lower concentrations but is neurotoxic to mature neurons at higher concentrations (Yankner et al. 1990). Several previous studies investigating AD pathology have directly shown that SST-like immunoreactivity (SSTLI) is reduced in the cerebral cortex and cerebrospinal fluid as well as decreased in SST-binding sites in the frontal cortex in AD cases (Davies et al. 1980; Wood et al. 1982; Beal et al. 1985). Consistent with these data, we have recently shown the loss of SST-immunopositive neurons as well as somatostatin receptor-specific changes in AD cortex in comparison with age-matched normal brains (Kumar 2005). In AD cases, a lower level of SST mRNA expression, per somatostatinergic cell in the hippocampus, has been shown (Dournaud et al. 1994). However, it is currently unknown whether the changes in SST seen in in vivo can also be attributed in Aβ-induced neurotoxicity.
In an attempt to elucidate the role of the somatostatinergic system in neurodegenerative diseases such as AD, the mechanism of glutamate toxicity was initially investigated. Additionally, quinolinic acid (QUIN), NMDA receptor (NMDAR) agonist, mimicked a model of HD and appeared to increase SST gene expression and peptide production (Patel et al. 1991). In Aβ-induced neurotoxicity, NMDAR activation and increase in Ca2+ are frequently seen. This is similar to what has been reported for NMDA-induced excitotoxicity. High levels of Aβ deposit alongside decreased levels of SST in AD brain tissue further incited us to elucidate the possible role of SST in Aβ-induced neurotoxicity. Our preliminary data support the concept that Aβ influences SST release and its accumulation in cortical neurons (Geci et al. 2000). Recently, Sato et al. (2004) showed that SST regulates the metabolism of Aβ through modulatory proteolytic degradation catalyzed by neprilysin. Accordingly, in the present study, employing cortical cell culture as an in vitro model, we tested the hypothesis that Aβ25–35 induces SST gene transcription, protein production, and secretion. Our data clearly demonstrate an increase in SST accumulation in cortical neurons and release of SST into culture medium in a time- and concentration-dependent manner in response to Aβ25–35-induced neurotoxicity.
- Top of page
- Materials and methods
The results presented here demonstrate an induction of SST mRNA alongside increased SST accumulation and release in cultured cortical neurons exposed to Aβ25–35-induced model of neurotoxicity. The increase in SST mRNA synthesis and SST accumulation is similar to what has been previously reported for NMDA- and QUIN-induced neurotoxicity (Patel et al. 1991). To our knowledge, the present study represents the first description of changes in expression of SST in cultured cortical neurons upon treatment with Aβ25–35. The neuronal culture used in the present study was 98% neuronal after treatment with deoxyuridine for 4 days. Accordingly, the data presented here exclude any possible interference of glial cell. SST expressions and its release in response to other neurotoxins, such as NMDA, have been characterized in depth (Tapia-Arancibia and Astier 1989; Ham et al. 1991, 1993; Patel et al. 1991, 1995; Kumar 2004). Furthermore, this effect of Aβ is not only selective to cultured cortical neurons, but also observed in hypothalamic and striatal cultures (Williams et al. 1991; Rage et al. 1994). The findings pertaining to Aβ-induced SST production recorded in this study are parallel to those described for NMDAR agonist-treated cultures, except for the constant increase in SST release induced by QUIN from 0 to 20 h (Patel et al. 1991).
The number of viable cells decreases significantly when cultures were exposed to synthetic Aβ25–35. In fact, cell viability decreased by 40–60% upon 24-h treatment with 40 μmol/L Aβ as measured by the MTT assay. This level of neuronal death is consistent with several previous studies. It is interesting to note that SST-positive neurons are selectively spared in Aβ25–35-induced toxicity. However, SST-positive neurons become susceptible to Aβ25–35 when exposed to a higher concentration of Aβ25–35 for a more extended period of time. This parallels the pathology of AD, whereby a decrease in SSTLI neurons directly correlates with the severity of the disease (Davies et al. 1980; Wood et al. 1982; Beal et al. 1988; Bisette et al. 1998; Davis et al. 1999; Nemeroff 1999). It should also be noted that the pattern of SST release from cultured cortical neurons treated with increasing concentrations of Aβ25–35 does not mirror the SST accumulation profile occurring simultaneously within the neuron. It would be expected that an increased SST production would translate into higher levels of SST secretion, as SST is primarily a secreted product, acting on neighboring cells as a neurotransmitter or a neuromodulator. Here, although SST production showed fourfold accumulation in cellular content, but the secretion of SST into culture medium remained relatively stable with only a slight enhancement in release was seen after treatment with 20 μmol/L of Aβ25–35. Time course studies of SST induction indicated an acute SST release, four times higher than basal levels, at 12 h with a subsequent return to baseline levels by 24 h. Such an acute response is not uncharacteristic of stimulated secretory cells. Indeed, SST release has been shown to be stimulated two- to fivefold by 10 mmol/L QUIN, over a 2- to 20-h period, with a fourfold increase at 12 h (Patel et al. 1991). Our observations using Aβ25–35 as neurotoxins suggest that Aβ25–35 itself impedes vesicular trafficking inside the cell, consequently affecting SST release. As Aβ25–35 toxicity is believed to be caused in part by the disruption of cholesterol uptake and metabolism, ultimately resulting in abnormal trafficking of membrane proteins which are crucial for normal neuronal function (Lynch and Mobley 2000).
Somatostatin mRNA levels were also affected by Aβ25–35 treatment. It is known that the SST gene is transcriptionally regulated by cAMP-response element binding protein, the cAMP-dependent transcription factor (Gonzalez et al. 1989), but whether these effects are mediated directly or indirectly on SST cells, and whether involve cAMP-response element binding protein in Aβ-mediated neurotoxicity remain to be elucidated. As mentioned previously, steroid hormones, cytokines, growth hormone, and insulin may also affect SST mRNA expression, either directly or indirectly. In addition, it is interesting to note that changes in SST expression have not only been observed in the context of NMDAR or Aβ-mediated neurotoxicity, but also in models of HIV encephalitis (Koutkia et al. 2004). For example, although brain-derived neurotrophic factor (BDNF), an endogenous growth factor, leads to a significant increase in SST production, HIV-1 envelope glycoprotein gp120 and BDNF together lead to an even greater SST production in comparison with that of BDNF alone, in cultured human fetal cortex (Barnea et al. 1999). Therefore, it is possible that SST induction is potentiated by a neurotoxin such as NMDA or Aβ25–35 acting alone or by the synergistic action of an endogenous peptide and a neurotoxin. This implies that the magnitude of SST production and secretion might be under the direct influence of other neuropeptides through different mechanism depending on cell type and neurotoxins.
Although it is not known whether SST involves in apoptosis or not but the role of somatostatin receptor subtypes have been shown in apoptosis. Here, we present the data that lower concentration of Aβ25–35-induced apoptosis and followed by necrosis at the higher concentration by using Hoechst 33258 and PI staining to discriminate apoptotic and necrotic cells, respectively. It has previously been questioned that PI does not discriminate between apoptotic and necrotic cells. In our experimental conditions, we determined necrotic cells on the basis of cell morphology in addition to PI staining. Neuronal cell showing swollen cell bodies and strongly stained with PI was considered apoptotic. Secondly, there was no prominent staining of PI at lower concentration of Aβ25–35, whereas in contrast apoptotic cells were stained strongly with Hoechst 33258 dye but not with PI. Third, lactate dehydrogenase (LDH) release to the culture medium is an index of necrosis, which does not occur in the case of apoptosis. We did not observe any change in LDH release when apoptosis was prominent at lower concentration of Aβ25–35, whereas in the presence of higher concentration of Aβ25–35, LDH was released significantly as an index of dying cells (data not shown). LDH release is supplemented with MTT assay (Fig. 1) for the quantifications of neurons.
Previously, Yankner et al. (1990) demonstrated that Aβ is neurotrophic and that once aggregated, it exerts a neurotoxic effect. However, most significantly, a recent study shows that SST blocks the aggregation of Aβ through the regulation of neprilysin activity (Sato et al. 2004). Furthermore, SST also affects neprilysin expression and its synaptic localization (Sato et al. 2004). These data support our observations with regard to SST release at early time point in response to Aβ and confirm the fact that SST may have an ability to inhibit the aggregation of Aβ. Indeed, we have demonstrated that there is a significant increase in SST release in response to Aβ at 12 h. This increased release of SST could underlie the neurotrophic ability of Aβ by blocking the aggregation of Aβ. However, Aβ may be neurotoxic in the presence of higher concentration when SST release to the culture medium is impaired. As the biological effect of SST is mediated by five different receptor subtypes, it will be essential to determine the roles of individual receptors in Aβ-induced neurotoxicity. Further studies are progressing in this direction. Consistent with gradual reduction of SST in AD brain and aging, our data strongly support that SST can be used as true marker of AD.