Detailed discussion of proteomic findings
Online resource 5
Proteomic findings upon cellular SIL1 increase are in accordance with a protective role in neurodegenerative and neuromuscular disorders
Although the alterations in membrane protein levels upon SIL1 increase at first glance might indicate altered BiP-substrate affinity. However, results of our surface protein studies do not indicate underrepresentation of membrane proteins at the cell surface, electron microscopic studies do show the build-up of electron-dense material corresponding to protein accumulation, immunoblot analysis of UPR-modulating proteins did not reveal accumulation of misfolded proteins as a results of a disturbed BiP-substrate affinity and the results of DNA-fragmentation analysis and cell proliferation did not show reduced cellular viability, a known side-effect of protein-misfolding and aggregation.
Several of the regulated proteins are important for the maintenance of the nervous system, i.e. decreased Scrib1, acting as a crucial regulator of brain development and spine morphology [54]. Some of these proteins even play major roles in the pathology of AD for which a protective function of SIL1 has recently been described [23]: TMEM59 affects the amyloid precursor protein (APP) shedding by reducing access of APP to the cellular compartments and avoiding AD amyloid beta peptide (Aβ) aggregation [55]. In this context decrease of TMEM59 in the membrane fractions of SIL1 overexpressing HEK293 cells indicated a detrimental effect of elevated SIL1 in the pathogenesis of neurodegenerative disorders such as AD. The initial indication of a detrimental effect of SIL1 elevation is supported by decrease of COQ6, a monooxygenase involved in the biosynthesis of the coenzyme Q10 (CoQ10), which is an electron carrier in the mitochondrial respiratory chain and a lipid-soluble antioxidant implicated in protecting cells from damage by reactive oxygen species. Q10 is a powerful endogenous antioxidant displaying therapeutic benefits against AD pathology and cognitive impairment in multiple AD mouse models [56].
However, SIL1 overexpression also caused an increase of ERAL1 within the membrane fractions. Membranous increase of ERAL1 most likely antagonizes mitochondrial dysfunction and reduction of oxidative stress burden[57], two major pathophysiological conditions in neurodegenerative disorders [58].
At first sight, additional decreased membrane levels of BRI2 disagree with a neuroprotective function of SIL1. BRI2 (ITM2B) in its mutated form is associated with British and Danish dementia. BRI2 can regulate critical processes involved in AD pathogenesis. These processes not only include the metabolism of APP and formation of Aβ, but also the levels of secreted insulin degrading enzyme (IDE), a main enzyme involved in Aβ clearance. Consequently, decreased BRI2 abundance in the membranes of “long term” SIL1 overexpressing HEK293 cells would suggest a promotion of the AD-pathophysiology rather than amelioration of the phenotype. However, recent studies observed increased levels of BRI2 in Aβ plaques in human AD hippocampus, which most likely affect BRI2 functional activity and a relationship between BRI2 protein changes, IDE activity and Aβ levels [59]. Thus, BRI2 accumulation (and formation of high of molecular weight BRI2 forms) observed in AD may impair IDE functioning and consequently lead to impaired Aβ clearance and to the accumulation of Aβ [59]. Consequently, BRI2 decrease in our in vitro model can also be considered as an antagonizing strategy supporting the neuroprotective function of SIL1. Interestingly, decrease of other membrane proteins upon increase of SIL1 support this hypothesis: Nectin-2 acts as a receptor for herpes simplex virus 1 which has been proposed as potential cause of AD because of its ability to form Aβ plaques and neurofibrillary tangles due to tau hyperphosphorylation and action of beta and gamma secretase on APP. Remarkably, different genetic association studies identified polymorphisms in nectin-2 and thus linked the virus with AD [60]. Therefore, down-regulation of the receptor in the enriched membrane fractions of SIL1-overexpressing cells is in accordance with the beneficial role of SIL1 in the pathogenesis of AD, especially regarding the described function in tau hyperphosphorylation[23]. Downregulation of the vesicular protein RAB9A which plays an essential role in the proper release of viral particles [61] supports this finding. Because pathological synaptic changes in AD include surface AMPAR loss and this loss correlates with increased AMPAR ubiquitination via NEDD4 [62], decrease of NDFIP2 as a NEDD4 supporting protein [63] fits with a neuroprotective effect of SIL1 increase. Strikingly, reducing Nedd4-1 levels in an AD in vitro model prevented surface AMPAR loss and synaptic weakening [62]. BASP1 is a brain abundant protein localized at the inner surface of the presynaptic plasma membrane[80]. Emerging evidence suggests that BASP1 is critically involved in various cellular processes such as the accumulation of phosphatidylinositol-4,5-diphosphate in lipid rafts. As it is known that BASP1 forms heterogeneously-sized oligomers and higher aggregates with an outward similarity to oligomers and protofibrils of amyloid proteins, down-regulation of this protein in the enriched membrane fractions of SIL1 overexpressing cells is also in accordance with a beneficial role of SIL1 and most likely acts towards the prevention of additional stress burden. TSPAN3 is a new modulatory co-receptor for the Nogo-A [64] and recent research shows that through binding to Nogo-A receptor, Nogo-A plays an important role in Alzheimer's disease (AD) pathogenesis. Particularly, Nogo-A receptors modulate the generation of Aβ [65]. Thus, decreased level of TSPAN3 are in accordance with the pathological benefit of SIL1 increase in the pathology of AD as well.
Apart from proteins downregulated in the membrane fractions of SIL1 overexpressing cells, proteins with increased abundances agree with a beneficial role of elevated SIL1 in AD pathology: SDF4 regulates calcium-dependent activities in the ER lumen or post-ER compartment. Interestingly, SDF-1α can decrease Aβ burden in vitro and in vivo, suggesting that SDF proteins could provide a novel and promising target for lowering Aβ pathology in AD [66]. In this manner, increase of SDF4 in the enriched membrane fractions of SIL1 overexpressing cells is in accordance with a beneficial role of SIL1 in AD pathology. FDFT1 was increased in the proteomic profiling of enriched membrane proteins in “long term” SIL1 overexpressing cells. Decrease of the same protein in the general proteome profiling (Fig. 7) could therefore be explained by forced ER-membrane translocation and underrepresentation of membrane proteins in the global approach. Notably, results of a microarray study of gene expression profiles in primary mouse cortical neurons in response to oligomeric Aβ revealed down-regulation of genes involved in the biosynthesis of cholesterol and other steroids and lipids including Fdft1[67]. Moreover, stimulation of the core metabolic reactions of cholesterol in the hippocampus, an approach which can replicate the trends of relative cholesterol levels in AD also uncovers Fdft1 as a key factor involved in this process [68]. Thus, translocation of the protein to the membrane might have an antagonizing effect.
Although our combined findings suggest that cellular SIL1 increase has no major impact on the BiP-substrate affinity and rather indicates that SIL1 might act in a BiP-independent manner, alteration of membrane abundance of the proteins discussed above could in principle be the result of selective altered BiP-substrate affinity upon SIL1 increase to promote the ameliorative effect of co-chaperone elevation.
Our proteomic findings suggest that apart from membrane proteins, proteins of other sub-cellular localizations with neuroprotective functions are affected by increased SIL1 expression: i.e. upregulated ACPH acts a hydrolase which degrades monomeric and oligomeric Aβ peptides[81]. The decrease of p53 also supports the hypothesis of a neuroprotective function of SIL1 in the genesis of AD because it’s up-regulation supports apoptosis in β-amyloid-induced neuron death [69]. THOC1, also decreased in our in vitro model, participates in another (p53 unrelated) apoptotic pathway which is characterized by the activation of caspase-6 [70].
In addition, other proteins with altered abundances upon cellular SIL1 elevation support a more general impact in protection of neurodegenerative and neuromuscular disorders. SIL1 elevation also results in an increase of the ubiquitously expressed HN1 protein which is known to downregulate Akt-mediated GSK3β signalling [71]. Notably, increased GSK3β signalling is a pathophysiological hallmark of myotonic dystrophy, a disorder affecting the lens, nervous system and skeletal muscle. ANXA1, upregulated upon cellular increase of SIL1 is a central player in the anti-inflammatory and neuroprotective role of microglia [72]. Transcriptome studies using FUS/TLS knockout mice which present with phenotypes possibly related to neuropsychiatric and neurodegenerative conditions, but distinct from an expected severe ALS and frontotemporal lobar degeneration phenotype, show increase of Taf15 transcript[82]. Remarkably, the corresponding protein is also increased in our in vitro model, the same as in GFPT1. Mutations in GFPT1 are causative for a CMS associated with a muscular dystrophy [73]and disturbances of the NMJ upon Sil1 mutation have been described [22]. MDH1, which is also increased upon SIL1 elevation, is important in transporting NADH equivalents across the mitochondrial membrane and its functions are primarily related to aerobic energy production for muscle contraction and neuronal signal transmission. This is in accordance with the strong MDH1 expression in brain and skeletal muscle [74]. Compensatory overexpression of this protein has already been described in dilated cardiomyopathy [74]. However, further protein alterations also support a potential beneficial role of SIL1 in muscle tissue: because myosin light chain phosphorylation is a primary biochemical mechanism for tension potentiation due to repetitive stimulation in fast-twitch skeletal muscle [75], increase of MYLK2 upon elevated SIL1 also suggests a beneficial role of this protein in muscular disorders. This assumption is supported by the concomitant increase of TBCD, a chaperone involved in de novo tubulin heterodimer assembly [76] as well as by the decrease of the intracellular level of secreted FBN2, a protein responsible for myopathy in respective null mice [77]. SIL1 overexpression also causes cellular increase of LMCD1, a Z-disc protein functioning as a DUX4 binding partner thus playing major roles during muscle differentiation [78]. PLIN3, which is upregulated in SIL1 overexpressing HEK293 cells, in known to be increased in skeletal muscle following exercise and is associated with fat oxidation [79]. However, altered proteomic signature upon SIL1 elevation is also indicative for a protective role of the co-chaperone in crystalline lens: loss of functional GALK is causative for recessive congenital cataracts [83], a clinical hallmark of MSS [14].
Labisch et al., „Tracking effects of SIL1 increase: taking a closer look beyond the consequences of elevated expression level”.