Role of IGF-I and IDE in amyloid-β toxicity

Read Also

Insulin-like growth factor-I (IGF-I) and insulin are structurally related circulating hormones. Although the role of insulin as a critical metabolic hormone has been recognized for many years, the biological significance of IGF-I has not been defined so easily. In recent years, IGF-I has been implicated as a physiologically relevant neuroprotective factor with a wide spectrum of actions in the adult brain. Evidence supporting this assumption comes from the observation that many neurodegenerative conditions show altered levels of serum IGF-I, and that levels of serum and brain IGF-I decrease with age (Arvat et al., 2000; Busiguina et al., 2000). It has been argued that Aβ accumulation in AD possibly results from reduced clearance, which may thus represent an important etiopathogenic event in the development of AD or other risks associated with ageing. Accordingly, it has been hypothesized that IGF-I potentially increases BBB permeability for transport of Aβ carrier proteins into the brain and subsequently increases Aβ clearance in the brain (Carro et al., 2002). To determine whether serum IGF-I affects brain Aβ levels, Carro and colleagues used over 18-month old rats, which displayed increased brain levels of Aβ compared to 3-month old rats, and treated them with IGF-I using chronic subcutaneous infusion. Interestingly, levels of Aβ in the hippocampus and cortex of the old rats appeared to be reduced to those seen in young rats. In the same study, chronic subcutaneous infusion of IGF-I resulted in reduced CNS Aβ burden in APPsw (Tg2576) transgenic mice (Carro et al., 2002). Subsequent experiments showed that IGF-I increased CNS transthyretin and albumin levels, both of which co-immunoprecipitate with Aβ and may thus mediate Aβ clearance (Carro et al., 2002). In support, higher brain levels of transthyretin and albumin after IGF-I administration correlated with lower Aβ in the brain. Although the administered IGF-I doses may have resulted in hyperphysiological IGF-I serum levels in this study, these findings convincingly indicate that IGF-I is likely involved in physiological and pathological brain ageing. Nevertheless, in view of the numerous protective actions exerted by IGF-I in the brain, the authors cannot exclude albumin and transthyretin-independent effects of IGF-1 leading to increased Aβ clearance (Carro et al., 2002).

The therapeutic benefit of reducing Aβ burden is widely supported (Golde, 2003). Prevention of Aβ toxicity may thus be beneficial in the inhibition of AD pathogenesis. After determining that IGF-I may be involved in Aβ clearance by mediating passage of the carrier proteins albumin and transthyretin into the brain, Carro et al. next reported that blockade of IGF-I receptors in the choroid plexus of Wistar rats was associated not only with brain amyloidosis, but also with hyperphosphorylation of tau and cognitive deficits (Carro et al., 2006). By expression of a dominant negative IGF-I receptor in the choroid plexus specifically, signaling through the IGF-I receptor was completely blocked in this brain structure. As expected, reduced levels of the Aβ carrier proteins albumin and transthyretin were found in the brain. Interestingly and in addition to increased levels of Aβ found in the cortex and hippocampus, blockade of IGF-I receptor signaling also led to a significant increase in the levels of hyperphosphorylated tau, impaired learning acquisition and memory loss. Restoring IGF-I receptor function through expression of the wild type IGF-I receptor, 3 months after blocking it, resulted in a reversion of all AD-like disturbances except learning. Although Aβ levels were increased in the brain of these rats, they developed no Aβ plaques or NTFs. As the authors note, this may be because additional factors associated with age-related changes are involved and are necessary to develop brain plaques and tangles (Carro et al., 2006). Furthermore, it is important to note that during normal ageing, rodents do not develop either plaques nor tangles, in comparison to humans. A shorter life-span, or structural differences in APP may account for this inter-species difference (De Strooper et al., 1995). Nevertheless, these data imply that a loss of IGF-I signaling is potentially associated with late-onset AD.  

In addition to Aβ clearance mediated by IGF-I-regulated albumin and transthyretin levels, insulin-degrading enzyme (IDE) has also been implicated as a key component responsible for the degradation and clearance of Aβ in the brain (Cook et al., 2003). IDE is a metalloprotease with a molecular weight of 110 kd that is expressed in many tissue types, with high concentrations noted in the brain (Kuo et al., 1991; Wang et al., 2010a). The ability of insulin to inhibit IDE-mediated degradation of Aβ has particular relevance for reports claiming that insulin resistance and consequently hyperinsulemia may increase the risk for AD. Although contrasting reports have been published regarding IDE expression in AD, Cook and colleagues associated IDE expression levels with APOE-4 status in AD patients (Cook et al., 2003). They showed that hippocampal IDE levels were significantly lower in AD patients with the APOE-4 allele, the isoform known to be a risk factor for late onset AD, compared to AD patients without the APOE-4 allele and to normal subjects both with and without the APOE-4 allele. In addition, IDE mRNA levels were similarly reduced in the hippocampus. These findings and those of another study (Du et al., 2009), indicate that reduced IDE expression is associated with a significant risk factor of AD and suggest that IDE may interact with APOE status to affect Aβ clearance. Furthermore, as it has been reported that IDE has preferential affinity for insulin substrates such as Aβ (Qiu et al., 1998). Therefore, it is important to know how brain insulin levels are changed due to insulin resistance and how this could influence IDE activity towards Aβ. It is also unclear whether APOE-4 status plays a causal role in reduced IDE expression in AD or whether IDE levels are reduced through an independent process that interacts with the presence of the APOE-4 allele during the development of the disease. Assuming that insulin levels are reduced in the brain as a result of peripheral hyperinsulemia/insulin resistance (Craft, 2007), it is possible that IDE levels are reduced due to lower levels of substrate. Multiple mechanisms may thus affect IDE expression or activity and differentially determine IDE’s role in AD pathogenesis.