Abstract
N-terminal phosphorylation at residues T3 and S13 is believed to have important beneficial implications for the biological and pathological properties of mutant huntingtin, where IKBKB was identified as a candidate regulator of huntingtin N-terminal phosphorylation. The paucity of mechanistic information on IKK pathways, together with the lack of sensitive methods to quantify endogenous huntingtin phosphorylation, prevented detailed study of the role of IKBKB in Huntington’s disease. Using novel ultrasensitive assays, we demonstrate that IKBKB can regulate endogenous S13 huntingtin phosphorylation in a manner dependent on its kinase activity and known regulators. We found that the ability of IKBKB to phosphorylate endogenous huntingtin S13 is mediated through a non-canonical IRF3-mediated IKK-pathway, distinct from the established involvement of IKBKB in mutant huntingtin’s pathological mechanisms mediated via the canonical pathway. Furthermore, increased huntingtin S13 phosphorylation by IKBKB resulted in decreased aggregation of mutant huntingtin in cells, again dependent on its kinase activity. These findings point to a non-canonical IKK-pathway linking S13 huntingtin phosphorylation to the pathological properties of mutant huntingtin aggregation, thought to be significant to Huntington’s disease.
Introduction
In Huntington’s disease (HD) an expansion of a CAG repeat within exon 1 of the huntingtin (HTT) gene, which produces a huntingtin (HTT) protein with an expanded polyglutamine (polyQ) repeat, leads to a progressive and fatal neurodegenerative pathology (Bates, Dorsey et al., 2015, McColgan & Tabrizi, 2018, Ross & Tabrizi, 2011). Mutant HTT fragments comprising the polyQ repeat, produced by alternative splicing and/or proteolytic cleavage, are widely believed to contribute significantly to HD through the propensity of such fragments to misfold and form aggregates (Bates et al., 2015, Landles, Sathasivam et al., 2010, Mangiarini, Sathasivam et al., 1996, Sathasivam, Neueder et al., 2013).
HD onset and progression are known to be influenced by a number of modifiers (Flower, Lomeikaite et al., 2019, Genetic Modifiers of Huntington’s Disease Consortium. Electronic address & Genetic Modifiers of Huntington’s Disease, 2019, Goold, Flower et al., 2019, Holmans, Massey et al., 2017, Lee, Chao et al., 2017), some of which may influence pathology by facilitating CAG-repeat expansion in somatic cells. Other evidence indicates that disease modification may be attainable through mechanisms directly affecting mutant HTT. Hence, protein sequences flanking the polyQ repeat—such as the first 17 residues of the HTT protein (N17 domain) and the proline-rich domain at the N- and C-termini of the polyQ repeat, respectively—has been shown to profoundly alter the biological and biophysical properties of these mutant HTT fragments in isolated proteins, cells and of the full length mutant HTT protein in vivo (Chatterjee, Steffan et al., 2021, Saudou & Humbert, 2016). Since the N17 domain is the target of several post-translational modifications (PTMs; (Ehrnhoefer, Sutton et al., 2011, Saudou & Humbert, 2016)), significant attention has focussed on residues associated with PTMs that may be modulated pharmacologically by targeting relevant enzymes, with the aim of altering the pathological properties of mutant HTT. Indeed, studies with phosphor-mimetic or phosphor-abrogative substitutions of residues T3 and S13/S16 have revealed a key role for these residues in regulating mutant HTT properties in isolated proteins (Crick, Ruff et al., 2013, Kelley, Huang et al., 2009, Mishra, Hoop et al., 2012), cells (Aiken, Steffan et al., 2009, Branco-Santos, Herrera et al., 2017, Maiuri, Woloshansky et al., 2013, Thompson, Aiken et al., 2009, Zheng, Li et al., 2013), and mice (Gu, Cantle et al., 2015), where phosphor-mimetic mutants invariably mitigate the effects of polyQ expansion. More recently, the availability of HTT proteins bearing bona-fide phosphorylation at these residues has confirmed the key role for T3 and S13/S16 in modulating mutant HTT phenotypes in vitro (Cariulo, Azzollini et al., 2017, Chiki A., 2017, DeGuire, Ruggeri et al., 2018). Parallel advances in developing assays that can specifically detect and quantify HTT phosphorylation have provided the required tools to study these PTMs in HD models, identifying potential modulators of endogenous HTT phosphorylation (Cariulo et al., 2017) (Cariulo, Verani et al., 2019). Amongst such modulators are kinases capable of increasing phosphorylation at HTT T3 or S13/S16 (Bowie, Maiuri et al., 2018, Bustamante, Ansaloni et al., 2015, Chiki, Ricci et al., 2021, Ochaba, Fote et al., 2019, Thompson et al., 2009) as well as pharmacological tools for proof-of-concept studies (Alpaugh, Galleguillos et al., 2017, Atwal, Desmond et al., 2011, Bowie et al., 2018, Cariulo et al., 2019). Efforts to identify kinases and/or phosphatases associated with HTT phosphorylation at T3 and S13/S16 have so far led to the identification of IKBKB (Bustamante et al., 2015, Ochaba et al., 2019, Thompson et al., 2009), TBK1 (Hegde, Chiki et al., 2020) and PP1 (Branco-Santos et al., 2017) as tool enzymes for T3 (IKBKB, PP1) and S13/S16 (IKBKB, TBK1) regulation. Of these, IKBKB and/or IKK signalling have been mechanistically associated with mutant HTT toxicity (Khoshnan, Ko et al., 2009, Khoshnan, Ko et al., 2004, Khoshnan & Patterson, 2011, Khoshnan, Sabbaugh et al., 2017, Thompson et al., 2009), HTT transcriptional regulation (Becanovic, Norremolle et al., 2015) and modulation of HTT S13 phosphorylation in vivo (Ochaba et al., 2019). This evidence, together with the known amenability to pharmacological modulation, makes IKK pathway kinases (and particularly IKBKB) strong tool candidates that can elucidate the mechanisms involved in regulating mutant HTT S13/S16 phosphorylation (Atwal et al., 2011, Ochaba et al., 2019).
Studies using phosphor-mimetic mutations or bona-fide HTT phosphorylation have however suggested that increased phosphorylation at T3 or S13/S16 would need to be achieved (Atwal, Xia et al., 2007, Cariulo et al., 2017, Chiki A., 2017, DeGuire S.M., 2017, DeGuire et al., 2018, Gu, Greiner et al., 2009, Zheng et al., 2013), thereby pointing at the need to develop ways to increase (rather than inhibit) the activity of relevant kinases. The paradoxical observations that IKBKB can directly phosphorylate HTT at S13/S16 (Thompson et al., 2009) and that IKBKB inhibitors can achieve the same effect (Atwal et al., 2011) suggest a level of complexity which requires further investigation, examining the role of the distinct IKK signalling pathways downstream of IKBKB in the regulation of HTT N17 phosphorylation. IKK signalling is amongst the better understood kinase signalling pathways, and includes signalling by the IKK complex (IKBKB homodimers or IKBKB/IKBKA heterodimers in complex with IKBKG) regulating the activity of the transcription factor NF-kb (Hacker & Karin, 2006, Israel, 2010, Perkins, 2007, Yu, Lin et al., 2020), as well as non-IKK complex mediated signalling pathways, typically mediated by the non-canonical IKK kinases IKBKE and TBK1, the best understood of which acts via the nuclear factors IRF3/IRF7 (Balka, Louis et al., 2020, Clark, Peggie et al., 2011, Fitzgerald, McWhirter et al., 2003, Hacker & Karin, 2006, Sharma, tenOever et al., 2003, Yum, Li et al., 2021). While published evidence implicates IKBKB/IKBKA and canonical (NF-kb) mediated signalling in contributing to mutant HTT-mediated pathology and in HTT transcriptional regulation (Becanovic et al., 2015, Khoshnan & Patterson, 2011), little is known of the signalling downstream of IKBKB leading to HTT N17 phosphorylation.
Another level of complexity is provided by the presence of phosphorylation at T3 and S13/S16, which appear to functionally cross-talk at least in vitro (DeGuire et al., 2018). As IKBKB has been described capable of phosphorylating both these HTT phosphor-epitopes (Bustamante et al., 2015, Thompson et al., 2009), it is important to investigate the effects of T3 and S13/S16 phosphorylation by IKBKB from the perspective of cross-talk, exploring how T3 modification affects S13/S16 phosphorylation and vice-versa.
Here, we have addressed these open questions using novel ultrasensitive assays for measuring endogenous levels of S13 HTT phosphorylation (pS13 HTT). We confirm, for the first time on endogenous HTT, that IKBKB can quantitatively increase pS13 HTT levels in human cells and does so in a manner dependent on its catalytic activity, as determined by mutations inactivating its kinase domain and selective pharmacological inhibitors. We further uncover a role for phosphatases and in particular for PP2A in regulating the ability of IKBKB to modulate pS13 HTT levels. Interestingly, we found that the ability of IKBKB to increase pS13 HTT levels in human cells is shared with IKBKE but not with the other canonical IKK kinase, IKBKA, and used IKBKB mutants to define that monomeric as well as Nemo binding-incompetent IKBKB remain capable of increasing pS13 HTT levels, thus excluding a role for the IKK complex and canonical IKK signalling. Coherently, pS13 HTT modulation by IKBKB was found to be associated with activation of a non-canonical IKK pathway involving IRF3 activation rather than through IKBA (canonical IKK signalling) activation. Functionally, the modulation of pS13 HTT levels by IKBKB is associated with a robust decrease of aggregation in a commonly employed surrogate cell model of mutant HTT aggregation. Finally, we explored the role of residue cross-talk, providing the first evidence that T3 mutation affects basal and IKBKB induced pS13 HTT levels, an effect which translates functionally also in mutant HTT aggregation in cells. The identification of non-canonical IKK signalling as a regulator of HTT N17 phosphorylation and the finding that T3 and S13 phosphorylation may be subject to cross-talk represent major advances towards the understanding of approaches aiming at reducing mutant HTT pathology through increased N17 phosphorylation, and provide novel insights towards the mechanistic understanding of N-terminal phosphorylation, regulation and downstream pathophysiological implications.
Results
IKBKB modulation of endogenous pS13 HTT levels is regulated by okadaic acid-sensitive phosphatases
IKBKB can phosphorylate HTT N-terminal protein fragments in vitro (Thompson et al., 2009) and its overexpression can increase pS13 HTT levels in cells expressing HTT N-terminal fragments (Bustamante et al., 2015, Thompson et al., 2009). More recently, endogenous IKBKB was shown capable of regulating pS13 HTT levels in mice as determined by semiquantitative methods (Ochaba et al., 2019). Paradoxically, other studies showed that pharmacological inhibition of IKBKB leads to the same results in cells (Atwal et al., 2011). We sought to investigate this discrepancy by interrogating the role of phosphatases in regulating the effects of IKBKB on pS13 HTT levels. Phosphatases may affect pS13 HTT levels in different ways, including direct dephosphorylation of pS13 as well as regulation of the catalytic activity, localization and stability of S13 kinases and phosphatases. Of the different pharmacological tools available to probe phosphatase activity, okadaic acid (OA), a somewhat selective inhibitor of PP1 and PP2A (Dounay & Forsyth, 2002), has been previously used to demonstrate an increase in nuclear accumulation of N-terminal HTT fragments in cells associated with increased S16 HTT phosphorylation (Havel, Wang et al., 2011). Interestingly, both PP1 and PP2A have been associated with a role in regulating N-terminal HTT aggregation and phosphorylation (Branco-Santos et al., 2017, Metzler, Gan et al., 2010), while PP2A is a known regulator of IKBKB catalytic activity and downstream signalling (DiDonato, Hayakawa et al., 1997). We therefore used OA to investigate the role of PP1/PP2A in regulating basal and IKBKB-induced pS13 HTT levels. For these mechanistic studies, we chose the human cell line HEK293T, being widely employed as a reductionistic model to study HTT PTMs and their biology (Bustamante et al., 2015, Cong, Held et al., 2011, Daldin, Fodale et al., 2017, Fodale, Kegulian et al., 2014, Huang, Lucas et al., 2015, Schilling, Gafni et al., 2006, Zheng et al., 2013).This cell line expresses HTT protein endogenously at significant levels (Cariulo et al., 2017, Ratovitski, O’Meally et al., 2017) and the HTT protein is phosphorylated at T3 and S13/S16 under both endogenous and overexpression conditions (Cariulo et al., 2017, Cariulo et al., 2019), indicating the presence of the relevant signalling pathways required to phosphorylate HTT at these residues. The ability of overexpressed IKBKB or an IKBKB carrying a catalytically inactivating mutation K44M (Mercurio, Zhu et al., 1997) to increase pS13 HTT levels in HEK293T cells overexpressing an N571 mutant HTT fragment was then assessed by Western blotting and the Singulex immunoassay (SMC assay) (Cariulo et al., 2019) in the presence or absence of OA at relevant concentrations (Havel et al., 2011) (Fig. 1, A and B respectively). We assessed pS13 HTT levels by Western immunoblotting using the previously characterized rabbit polyclonal antibody, specific for pS13 HTT (Cariulo et al., 2019). As shown in Fig. 1A, all HEK293T samples expressed N571 Q55 HTT protein at comparable levels. However, in the absence of OA little if any increase in pS13 HTT levels was observed in cells transfected with IKBKB as compared to cells transfected with the empty expression plasmid, in spite of robust expression of IKBKB. However, this IKBKB protein was slightly inactive as judged by the level of phosphorylation at residues S177/S181, which are auto-phosphorylation sites commonly employed as a readout of IKBKB activity (Israel, 2010). We reasoned that the absence of robust pS13 HTT modulation by IKBKB was a result of low IKBKB activity, potentially due to the activity of endogenous phosphatases in HEK293T cells. In the presence of OA, however, a robust increase in pS13 HTT levels was observed in samples transfected with IKBKB but not in those transfected with an IKBKB kinase dead construct (Fig. 1A), paralleled with a strong increase in pS177/pS181 IKBKB levels, indicative of robust kinase activity. Interestingly, basal levels of pS13 HTT, in the absence of transfected IKBKB, were not affected by OA treatment, suggesting that control of pS13 HTT levels by OA-sensitive endogenous phosphatases is not the predominant regulatory mechanism. When samples were analysed by quantitative SMC assays, the capacity of OA to reveal a robust (ca. 10-fold), catalytic activity-dependent induction of pS13 HTT levels by IKBKB was confirmed (Fig. 1B). We concluded that while basal levels of pS13 HTT in cells expressing N571 Q55 HTT fragments are not significantly altered by OA-mediated phosphatase inhibition, the capacity of IKBKB to increase pS13 HTT levels is exquisitely controlled by OA-sensitive phosphatases, due to regulation of IKBKB activity, as evidenced by IKBKB pS177/pS181 levels. We next sought to determine if we could confirm the catalytic activity-dependence of IKBKB in regulating pS13 HTT levels and the effect of OA using pharmacological rather than genetic means, on either overexpressed N571 Q55 HTT (Fig. 1, C and D) or on endogenously expressed HTT in HEK293T cells (Fig. 1, E and F). As shown in Fig. 1C, N571 Q55 pS13 HTT levels were again not significantly modulated by IKBKB overexpression in the absence of OA, and addition of Bay-65-1942, a selective IKBKB inhibitor (Atwal et al., 2011), was similarly unable to significantly modulate these levels. In the presence of OA, the capacity of IKBKB to increase pS13 HTT levels was revealed and, importantly, it was sensitive to selective pharmacological inhibition by Bay-65-1942 in a concentration-dependent manner (Fig. 1D). Substantially comparable results were obtained when the response of pS13 HTT levels expressed from the endogenous HTT locus to IKBKB and OA was investigated under the same conditions (Fig. 1, E and F), and the more quantitative SMC assay analysis on these samples confirmed the findings obtained by Western blotting (no OA: Fig. 1G; OA treatment: Fig. 1H). Therefore, we concluded that pS13 HTT levels, whether from the endogenous HTT locus or from an overexpressed HTT N-terminal fragment, could be robustly increased in response to IKBKB in a catalytic activity-dependent and phosphatase-sensitive manner. As OA is described as an inhibitor of PP1 and PP2A (Dounay & Forsyth, 2002) but exhibits greater selectivity towards the latter (Swingle, Ni et al., 2007), the role of PP2A on OA’s ability to regulate IKBKB activity (S177/S181 auto-phosphorylation) and its effects on pS13 HTT levels was further evaluated using a specific siRNA to silence endogenous PP2A in HEK293T cells. As shown in Fig. 2A, the PP2A silencing produced an increase in IKBKB auto-phosphorylation, indicative of increased IKBKB activity, concomitantly with increased pS13 HTT levels. The magnitude of the effects of PP2A silencing on pS13 HTT modulation by IKBKB was not comparable to that attained with OA, likely due to incomplete PP2A silencing obtained with RNAi and the difficulty of phenocopying pharmacological inhibition of a catalytic enzyme with RNAi (Weiss, Taylor et al., 2007). Coherent with the observations made with OA, basal pS13 HTT levels (absence of IKBKB overexpression) were not affected by PP2A RNAi. Quantitative analysis of pS13 HTT levels of overexpressed N571 HTT bearing 55 CAG repeats (Q55) by SMC assay essentially confirmed the results obtained by Western blotting (Fig. 2B). Consistent with a role for PP2A in regulating IKBKB activity, we were able to detect an interaction between overexpressed IKBKB and endogenously expressed PP2A in HEK293T cells by both immunoprecipitation (Fig. 2C) and an ELISA interaction assay (Fig. 2D). We next sought to determine if overexpression of PPP2CA, the catalytic subunit of PP2A, could affect the ability of IKBKB to influence pS13 HTT levels, an experiment which needed to be performed in the absence of OA and therefore against the background of endogenous inhibitory effects of PP2A on IKBKB. Using overexpression of N571 Q55 HTT and shorter transfection times (24 hrs) than those employed in Fig. 1, we were able to detect a reproducible, modest induction of pS13 HTT levels upon co-transfection with IKBKB, concomitant with significant S177/S181 IKBKB phosphorylation as a measure of IKBKB catalytic activity (Fig. 2E and F). Under these conditions, co-transfection of PPP2CA effectively abolished IKBKB induced phosphorylation at S177/S181 and pS13 HTT levels, consistent with the observations made using RNA knock-down. Collectively, the data suggest that the capacity of IKBKB to influence pS13 HTT levels in HEK293T cells is dependent on its catalytic activity, that the catalytic activity of IKBKB is regulated by OA-sensitive phosphatases and that PP2A contributes to IKBKB inducible, but not basal, pS13 HTT levels. These findings are fully consistent with the known role of PP2A in regulating IKBKB activity (Barisic, Strozyk et al., 2008, Tsuchiya, Osaki et al., 2017).
IKBKB modulation of endogenous pS13 HTT levels involves a non-canonical IKK pathway associated with IRF3 signalling
IKK kinases have been extensively studied owing to their involvement in multiple areas of human disease, including inflammation, immunity and cancer (Clark et al., 2011, Hacker & Karin, 2006, Israel, 2010, Mercurio et al., 1997, Perkins, 2007, Sharma et al., 2003, Zandi, Rothwarf et al., 1997). Several functional mutants have been extensively characterized to probe the involvement of IKBKB in IKK pathways, which broadly can be grouped into IKK complex-mediated (canonical) and non-canonical. Canonical IKK signalling involves the formation of a functional complex involving IKBKB, IKBKA and IKBKG/Nemo, and results in IKBA phosphorylation and degradation, release of NF-kb from its inhibitor IKBA, and translocation of NF-kb to the nucleus. NF-kb executes its role as a transcription factor (Israel, 2010, Yu et al., 2020), regulating the expression of a number of genes involved in, amongst others, inflammation but also HTT itself (Becanovic et al., 2015). In contrast, non-canonical IKK signalling is largely IKBKG independent and includes pathways of which the best understood are those triggered by the non-canonical IKK kinases, TBK1 and IKBKE, which involve IRF factors (IRF3/7) rather than NF-kb as transcriptional effectors (Balka et al., 2020, Fitzgerald et al., 2003, Hacker & Karin, 2006, Yum et al., 2021). A range of IKBKB mutants have been generated and characterized which can be leveraged to genetically investigate the mechanism of IKBKB regulation of pS13 HTT levels. The HEK293T cell model, with its simplicity and amenability to manipulation, and the observation that OA unmasks the ability of IKBKB to modulate pS13 HTT by inhibiting endogenous phosphatases including PP2A, provides an appropriate environment for these mechanistic investigations. Further validation and hypothesis building, with regard to relevance for HD needs to be subsequently tested in other, more complex contexts. Equipped with IKBKB and a set of IKBKB mutants (Fig. 3A) including a catalytically inactive (IKBKB KD) variant (Mercurio et al., 1997), an IKBKG/Nemo binding-incompetent (IKBKB NBD) mutant (May, Marienfeld et al., 2002), a variant with decreased kinase activity through phosphor-mimetic mutations at residues S177/S181 (IKBKB S177E/S181E) (Huynh, Boddupalli et al., 2000, Kishore, Huynh et al., 2002, Liu, Misquitta et al., 2013, Mercurio et al., 1997, Polley, Huang et al., 2013) and an IKBKB mutant incapable of dimerization (IKBKB LZ) (Hauenstein, Rogers et al., 2014), we set out to dissect the mechanisms through which IKBKB increases pS13 HTT levels in the HEK293T model. All mutants were C-terminally FLAG-tagged to facilitate comparative analysis of expression levels. We co-transfected IKBKB and the indicated mutants in HEK293T cells together with an expression plasmid encoding N571 Q55 HTT, in the presence or absence of OA, and determined their activity on pS13 HTT levels. As shown in Fig. 3B, in the absence of OA treatment, IKBKB did not produce a robust increase in pS13 HTT levels, as previously observed, as did none of the IKBKB mutants in spite of varying expression levels of these latter (see anti-FLAG blotting). Interestingly in the presence of OA, IKBKB and IKBKB mutant incapable of forming a complex with IKBKG/Nemo (IKBKB NBD) or the IKBKB mutant incapable of dimerization (IKBKB LZ) were still capable of increasing pS13 HTT levels, with a concomitant increase in S177/S181 phosphorylation, consistent with an IKK complex (IKBKG/IKBKA)-independent mechanism. This was evident despite the disparity in expression levels of the different constructs, which became evident with the kinase dead IKBKB (IKBKB KD) and the IKBKG/Nemo binding-incompetent IKBKB (IKBKB NBD). The latter construct still produced a substantial increase in pS13 HTT levels, in spite of significantly lower expression levels with respect to for instance IKBKB. These samples were also analysed using SMC assays, producing essentially comparable results to those obtained by Western blotting (Fig. 3C). The data indicated that, in HEK293T cells, IKBKB regulates pS13 HTT levels through a non-canonical, IKK complex-independent mechanism, dependent on the catalytic activity of IKBKB which itself is regulated by protein phosphatases including PP2A. We sought to confirm and further extend these findings by analyzing the effects of overexpression of IKBKB and its mutants on endogenous pS13 HTT levels in comparison with overexpression of a canonical (IKBKA) or a non-canonical (IKBKE) IKK kinase, and of the non-catalytic scaffolding component of the canonical IKK complex, IKBKG/Nemo. As shown in Fig. 4A, in the absence of OA overexpressed IKBKB, IKBKB NBD and IKBKB LZ were able to modestly increase endogenous pS13 HTT levels, while IKBKB KD and the IKBKB S/E mutant (decreased kinase activity) were not, as observed in cells expressing the N571 Q55 HTT protein (Fig. 3, B and C). Interestingly, overexpression of the non-canonical kinase IKBKE but not of the canonical kinase IKBKA or of IKBKG, the non-catalytic scaffolding component of the canonical IKK complex, was also able to increase endogenous pS13 HTT levels (Fig. 4A and B). In the presence of OA these effects were greatly enhanced, as expected (Fig. 4C and D). These results confirmed that non-canonical IKK signalling, but not signalling mediated by the canonical IKK complex, was responsible for regulating pS13 HTT levels. To further support this finding, we examined the phosphorylation of IRF3, a characterized effector of non-canonical IKK signalling, which upon phosphorylation by non-canonical IKK kinases is activated and acts as the transcriptional mediator of the pathway (Fitzgerald et al., 2003, Hacker & Karin, 2006). Indeed, we observed complete correlation between pS13 HTT phosphorylation and IRF3 phosphorylation (Fig. 4A and 4C). Collectively, the effects of IKBKB on pS13 HTT levels expressed from the endogenous HTT locus or from overexpressed HTT fragments are dependent on its catalytic activity, which is regulated by phosphatases including PP2A, can be phenocopied by non-canonical IKK kinases such as IKBKE but not by canonical IKK kinase IKBKA and, consistently, are mediated by a non-canonical IKK pathway associated with IRF3 activation.
IKBKB is known to directly phosphorylate purified, isolated HTT exon 1 protein in vitro (Thompson et al., 2009) but no evidence of a direct interaction between IKBKB and HTT has been demonstrated in cellular or animal models. We therefore asked whether the kinase exerted its activity on S13 in a direct way, through interaction with HTT in a cellular context, by applying a quantitative protein-protein interaction assays (TR-FRET and ELISA) to HEK293T cells co-expressing N571 Q55 HTT construct with or without S13A/S16A phosphor-abrogative mutations and IKBKB. As shown in Fig. 5 (A, B and C), IKBKB was able to interact with HTT N571 Q55 protein and this interaction was independent of the phosphorylation status of S13, as the abrogation of S13 phosphorylation (S13A/S16A mutant) did not impact a direct interaction. On the other hand the interaction was dependent on the catalytic activity of IKBKB, as the catalytically inactive mutant (KD) interacted with the HTT protein to a lesser extent (Fig. 5D). Based on these data we conclude that IKBKB can interact with HTT in this cell context, and that this interaction is dependent on the integrity of the catalytic domain of IKBKB but independent of IKBKB’s HTT substrate phosphorylation.
IKBKB influences mutant HTT aggregation through increased S13 HTT phosphorylation and cross-talk between T3 and S13
One of the key functional consequences of polyQ expansion believed to contribute to HD pathology is misfolding and aggregation of short (exon 1 encoded) N-terminal HTT fragments, produced either from misplicing and/or proteolysis (Bates, 2003, Bates et al., 2015, Franich, Hickey et al., 2019, Gipson, Neueder et al., 2013, Landles et al., 2010, Neueder, Landles et al., 2017, Sathasivam et al., 2013, Saudou & Humbert, 2016, Weiss, Trager et al., 2012). Although mutant HTT aggregation is invariably (but to different extents) detectable in HD mouse models (Menalled & Brunner, 2014, William Yang & Gray, 2011), this process is not easily reproduced in cellular models of full length mutant HTT. To address this issue, expression of the exon 1 fragment of mutant HTT in cells, often tagged with fluorescent proteins, is generally employed as a surrogate mutant HTT aggregation model (Atwal et al., 2007, Bates et al., 2015, Branco-Santos et al., 2017, Cooper, Schilling et al., 1998, Li, Liu et al., 2016, Li & Li, 1998, Olshina, Angley et al., 2010, Ross & Tabrizi, 2011, Zheng et al., 2013). Indeed, mutant HTT aggregation and toxicity in cellular and animal models of HD is robustly affected by phosphor-mimetic mutations at S13/S16 as well as by other mutations within the N17 domain (Atwal et al., 2007, Branco-Santos et al., 2017, Greiner & Yang, 2011, Gu et al., 2009, Zheng et al., 2013). Although some pharmacological tools demonstrated to increase pS13/pS16 HTT levels have shown benefit in animal and cellular models of HD, modifying different phenotypes including mutant HTT aggregation (Alpaugh et al., 2017, Bowie et al., 2018, Di Pardo, Maglione et al., 2012), the lack of suitable direct genetic modulators of pS13/pS16 HTT levels have so far limited the capacity to determine if effects of phosphor-mimetic mutations are indeed a true phenocopy of increased phosphorylation. Together with the data presented here, the identification of IKBKB as a kinase capable of increasing S13/S16 HTT phosphorylation (Thompson et al., 2009) and the evidence supporting its role as a physiologically relevant modulator of N-terminal HTT phosphorylation (Ochaba et al., 2019) points to IKBKB as a suitable tool to probe the role of increased phosphorylation on mutant HTT aggregation. In order to do this, we first implemented a surrogate cellular model of mutant HTT aggregation, based on overexpression of HTT exon 1 bearing an expanded polyQ (Q72) repeat, C-terminally fused to GFP, in HEK293T cells (Fodale et al., 2014). To validate the model, we sought to confirm the previously reported effects of N17 mutations on mutant Exon 1 HTT (mut Ex1 HTT) aggregation, where phosphor-mimetic mutations at S13/S16 resulted in robustly decreased aggregation (Atwal et al., 2007, Branco-Santos et al., 2017, Greiner & Yang, 2011), using previously validated and characterized constructs (Cariulo et al., 2017, Fodale, Boggio et al., 2017). As illustrated in Fig. 6A and B, consistent with published data (Branco-Santos et al., 2017, Gu et al., 2009) S13A/S16A and S13D/S16D double mutations produced dramatic and opposite effects on mut Ex1 HTT aggregation (negative controls are shown in Fig. S1). These data are in agreement with the proposed role for phosphomimesis in decreasing mut Ex1 HTT aggregation (Greiner & Yang, 2011). Conversely, T3A and T3D mutations also decreased aggregation (Fig. 6), but to the same extent indicative that the effect cannot be reconciled with phosphomimesis, contrary to what observed for S13/S16. In agreement with the results obtained using the independent T3D and S13D/S16D mutations, the triple T3D/S13D/S16D mutant produced a robust decrease in aggregates. Interestingly, a triple T3A/S13A/S16A mutation resulted in increased aggregation, comparable to that attained by the double S13A/S16A mutation (Fig. 6), suggesting that the beneficial effect of the T3A mutation on aggregation requires the presence of intact S13/S16 residues. One possibility is that mutation of T3 results in increased phosphorylation of S13/S16, and that the decreased aggregation observed with T3 mutation at least in part is mediated by increased pS13/pS16 levels. To investigate this aspect, we measured basal and IKBKB-induced pS13 HTT levels by densitometry analysis of Western blotting in cells transfected with HTT Ex1 Q72-EGFP and with T3A or S13A/S16A mutations (Fig. 7, A and B). As previously observed, a modest effect of IKBKB co-transfection on pS13 HTT levels was observed in cells transfected with the HTT Ex1 Q72-EGFP construct, owing to the absence of OA treatment. Indeed, the presence of a T3A mutation significantly increased basal pS13 HTT levels, by about 5-fold and correspondingly potentiated the effects of IKBKB on pS13 HTT levels relatively to the protein expressed from the unmodified construct (Fig. 7, A and B). We also measured pT3 HTT levels to determine if S13/S16 modification would affect pT3 HTT levels, which was not the case (Fig. 7, A and C), however confirming the ability of IKBKB to increase pT3 HTT levels (Bustamante et al., 2015). Therefore, in this reductionistic, surrogate model of mutant HTT aggregation we were able to confirm previous reports of the capacity of phosphor-mimetic S13/S16 mutations to reduce aggregation in cells, pointing to the role for T3 in regulating basal- and IKBKB-induced pS13 HTT levels. This also provides an explanation for the observed effects of T3A mutations to reduce aggregation in an S13/S16 dependent manner (Fig. 6). We next proceeded with examining the role of IKBKB-regulated pS13 HTT levels in the aggregation process. As previously shown, IKBKA and IKBKB S/E co-expression did not modulate pS13 HTT levels and as reported in Fig. 8A and B, also failed to modulate aggregation while IKBKB robustly reduced HTT Ex1 Q72-EGFP aggregation in a catalytic dependent manner, demonstrated by lack of effects of IKBKB KD. As this effect could be mediated directly through N17-phosphorylation dependent or independent mechanisms, we interrogated the relative importance of S13/S16 and T3 for the IKBKB mediated reduction in mutant HTT aggregation using the available mutations. As illustrated in Fig. 8C and D, the effects of IKBKB on HTT Ex1 Q72-EGFP aggregation were exquisitely dependent on the presence of intact S13/S16 residues, with T3 mutation further enhancing the capacity of IKBKB to influence the aggregation process, although T3A HTT Ex1 Q72-EGFP showed an already reduced propensity to aggregate (Fig. 6). We conclude that, in this surrogate cellular model of mutant HTT aggregation, IKBKB can reduce a pathologically relevant phenotype in a manner dependent on its catalytic activity and on the presence of intact S13/S16 residues, which directly links the ability of IKBKB to regulate pS13 HTT levels with its capacity to reduce aggregation. To our knowledge this is the first demonstration that increasing S13 phosphorylation phenocopies the effects of S13/S16 phosphor-mimetic mutations.
Discussion
The role of HTT PTMs in ameliorating pathological phenotypes in HD mouse models is well supported by the findings that phosphor-mimetic mutations at residues associated with phosphorylation (e.g. S13/S16 and S421) can significantly affect mutant HTT biology (Gu et al., 2009, Kratter, Zahed et al., 2016, Xu, Ng et al., 2020). Recent discoveries of pharmacological modulators of kinase activity (Ferguson & Gray, 2018, Simpson, Hughes et al., 2009), open the possibility of developing candidate tools to better increase our understanding of the mechanisms involved in modulating HTT phosphorylation and their contribution to HD pathophysiology.
However interesting conceptually, detailed studies assessing their potential therapeutic relevance has been hampered by the lack of specific and sensitive technologies to adequately determine the biophysical properties and biologically dynamic nature of bona fide PTMs. Where pharmacological modulators have been identified (Alpaugh et al., 2017, Bowie et al., 2018, Marion, Urs et al., 2014), their efficacy, potency, selectivity, precise mechanism of action and amenability as tools for further proof-of-concept studies in vivo remain to be demonstrated. As for investigational tools, several enzymes have been reported as modulators, and specifically IKBKB, TBK1 and CK2 for T3 and S13/S16 phosphorylation (Atwal et al., 2011, Bowie et al., 2018, Bustamante et al., 2015, Hegde et al., 2020, Ochaba et al., 2019, Thompson et al., 2009), PP1 for T3 phosphorylation (Branco-Santos et al., 2017), and AKT and SGK for S421 phosphorylation (Humbert, Bryson et al., 2002, Rangone, Poizat et al., 2004).
In general, however, these enzymes have not been demonstrated to potently, selectively and quantitatively influence the relevant endogenous HTT PTM, largely due to the lack of suitable readouts capable of detecting a dynamic, labile protein modification which likely affects only a portion of the total cellular protein pool. The availability and characterization of specific antibodies capable of sensitively detect HTT phosphorylation events (Aiken et al., 2009, Ansaloni, Wang et al., 2014, Atwal et al., 2011, Cariulo et al., 2017, Thompson et al., 2009) and their use in the recent development of novel ultrasensitive immunoassays (Cariulo et al., 2017, Cariulo et al., 2019) represents an appropriate platform for a more in depth validation of these current candidate modulators to provide deeper insights in the role of PTMs in HD. Of the candidate tools to increase N-terminal HTT phosphorylation, IKBKB represents one of the better validated enzymes reported to be involved in regulating pS13 or pS13/pS16 levels in cellular and animal models of HD (Atwal et al., 2011, Bustamante et al., 2015, Ochaba et al., 2019, Thompson et al., 2009), including in an endogenous HTT context in mice (Ochaba et al., 2019). Canonical (IKK complex-mediated) IKK signalling appears to be a contributor to HD toxicity mechanisms (Khoshnan & Patterson, 2011) and may control HTT transcription (Becanovic et al., 2015). Moreover IKBKB inhibitors increase pS13 HTT levels and decrease toxicity in cellular models of HD (Atwal et al., 2011). Although these findings suggest that inhibition of canonical IKK signalling lead to decreased toxicity and lower expression of mutant HTT, conditional IKBKB knock-out worsens the phenotype in a mouse model of HD (Ochaba et al., 2019). These paradoxical findings suggest a degree of complexity which requires further investigation (Greiner & Yang, 2011), as indeed very little is known of the downstream signalling pathways leveraged by IKBKB to increase phosphorylation of the HTT N17 domain. IKBKB is a pleiotropic kinase capable of signalling via diverse pathways (Chau, Gioia et al., 2008, Hacker & Karin, 2006, Israel, 2010, Perkins, 2007, Schrofelbauer, Polley et al., 2012) and may therefore contribute both positively and negatively to HD pathology, depending on cellular context and intracellular signalling homeostasis. Indeed, the illustration of a specific IKK pathway involved in the regulation of pS13 HTT levels adds an important aspect to the understanding of biological mechanisms involved in HD. Using a recently developed anti-pS13 antibody and ultrasensitive immunoassays (Cariulo et al., 2019), we have quantitatively examined the capacity of IKBKB to increase pS13 HTT levels expressed from the endogenous HTT locus in a human cell line widely employed for HTT PTM studies. We thereby identified a mechanism regulating the activity of IKBKB on the pS13 HTT substrate and propose a molecular pathway whereby IKBKB phosphorylates pS13 HTT. Importantly, we found that the ability of IKBKB to modulate pS13 HTT levels was dependent on its catalytic activity as determined by both genetic (kinase domain inactivating mutation) and pharmacological (selective IKBKB inhibitor) means. As might be expected, the effects of IKBKB on pS13 HTT levels were regulated by phosphatases, such as PP2A, likely acting at the level of IKBKB rather than by dephosphorylating HTT S13, since IKBKB-induced but not basal pS13 HTT levels were unaffected by the presence of OA, PPP2CA RNAi or PPP2CA overexpression. These observations are consistent with the described inhibitory effect of PP2A on IKBKB activity (DiDonato et al., 1997), and provide an additional level of complexity to data interpretation. Importantly, these findings offer a plausible explanation as to how pS13 HTT levels may be increased through inhibiting regulatory components of the kinase responsible for increasing pS13 HTT levels. To our knowledge, this is the first report which demonstrates a quantitative, catalytic activity-dependent, regulated modulation of endogenous HTT pS13 levels by IKBKB. The present work differentiates the IKBKB activity on pS13 HTT levels which we genetically demonstrate acts via an IKK-complex independent pathway, from IKBKB/IKK involvement in mutant HTT toxicity and HTT transcription, believed to act via a canonical, IKK complex-mediated pathway (Becanovic et al., 2015, Khoshnan & Patterson, 2011). The characterisation of the non-canonical, IRF-mediated IKK pathway as relevant for increasing pS13 HTT levels represents a significant advancement in the field and asks for further development of selective tools of non-canonical IKK pathway modulators to allow further research to elucidate mechanisms governing increases in pS13 HTT levels without worsening mutant HTT toxicity (Khoshnan et al., 2004) or increase its expression (Becanovic et al., 2015). We further elucidated the mechanism by which IKBKB phosphorylates HTT, providing the first evidence that IKBKB interacts directly with HTT, and that this interaction is dependent on the catalytic activity of IKBKB but independent of IKBKB’s HTT substrate phosphorylation. Finally, as a first piece of evidence that genetic modulation of pS13 HTT levels can reverse a pathologically relevant phenotype, we demonstrate, again for the first time, that increased pS13 HTT levels reduce mutant HTT aggregation in cells, fully phenocopying the effects of phosphor-mimetic mutations.
In conclusion, we believe these findings provide a significant advance in the field in several ways. From a technical perspective, this is the first demonstration that it is possible to quantitatively measure and modulate endogenous levels of a pathologically relevant PTM in HTT, phenocopying the effects of phosphor-mimetic mutations on a disease relevant readout. Conceptually, our findings advance the understanding of the relevance of IKBKB in HD biology in several ways. First, through the identification of a regulatory, phosphatase-dependent, mechanism acting via IKBKB and resulting in increased pS13 HTT levels. Second, through the identification of non-canonical, IRF-mediated IKK signalling pathway, resulting in increased pS13 HTT levels distinct from the canonical, IKK complex-mediated pathway, whose activity contributes to HD and to regulation of HTT expression. These data provide a rationale for identifying selective, non-canonical IKK signalling modulators and associated pharmacodynamic readouts, to further elucidate candidate pharmacological tools and pathways involved in HD pathophysiology.
Materials and Methods
Antibodies
The MW1, 2B7 and 4C9 antibodies bind to the polyQ stretch, the N17 domain and the polyproline region of HTT respectively. They were obtained from the CHDI Foundation (New York, NY) and their use in Western blotting and SMC assays has been described previously (Cariulo et al., 2017). The anti-pS13 antibody is an affinity purified, rabbit polyclonal antibody specific for the phosphorylated S13 residue of human huntingtin raised against the peptide LMKAFE(pS)LKSFQ developed by the CHDI foundation and available from the Coriell Institute for Biomedical research, HD Community Biorepository (ID CH01115). The validation of this antibody in different applications has been formerly reported (Cariulo et al., 2019). The here used anti-pT3 antibody was previously described (Cariulo et al., 2017). Monoclonal antibodies mAb2166 (catalog #MAB2166; Merck) and anti-GAPDH (catalog #G9545; Sigma-Aldrich), anti-FLAG mouse antibody (catalog #F1804; Sigma-Aldrich), anti-FLAG rabbit antibody (catalog #F7425; Sigma Aldrich), anti-IKBKB antibody (catalog #ab32135; Abcam), anti-pS176/180-IKBKB antibody (catalog #2697; Cell Signaling), Anti-Glial Fibrillary Acidic Protein GFAP (catalog #G9269; Sigma-Aldrich), anti-PP2A antibody (catalog #SAB4200266; Sigma-Aldrich), anti-IRF3 antibody (catalog #AB76409; Abcam) and anti-pS386-IRF3 antibody (catalog #AB76493; Abcam) were all supplied by a commercial source. Secondary antibodies used for Western blotting were Goat-Anti-mouse IgG HRP conjugated (catalog #12-349; Merck) and Goat-Anti-rabbit IgG HRP conjugated (catalog #12-348; Merck). The D2-fluophore and terbium antibody labelling for TR-FRET assay were custom made at CisBio (Bagnols, France). The Alexa-647 labelling for detection used in SMC and in ELISA assays was performed using the Alexa Fluor-647 Monoclonal Antibody Labeling Kit from Thermo Fisher Scientific (catalog #A20186) following manufacturer’s instructions. MW1 antibody was conjugated to magnetic particles for SMC assays, following the manufacturer’s recommendations (catalog #03-0077-02; Merck).
Plasmid and constructs
All plasmids employed in this study have been ordered or custom synthetized at TEMA Ricerca or Genescript (Nanjing, China) respectively. The chosen vector for constructs IKBKB, IKBKB KD, IKBKB S/E, IKBKB NBD, IKBKB LZ is pCDNA3.1. IKBKA (#RC216718), IKBKG (#RC218044), IKBKE (#RC212481) and PPP2CA (#RC201334) are commercially sourced by TEMA Ricerca. These constructs are also FLAG-tagged at the C-terminus. cDNAs constructs encoding for N-terminal HTT fragments (exon 1 HTT 1-90, based on Q23 numbering; Table S2) bearing different polyQ lengths (Q16 or Q72) and mutations (T3A, S13A/S16A, T3D, S13D/S16D, T3A/S13A/S16A, T3D/S13D/S16D) codify also for a EGFP-Tag at the C-terminus and are inserted in a pCMV vector. N-terminal HTT fragment, HTT 1-571, based on Q23 numbering (Table S3), is encoded in an untagged pCDNA3.1 vector. Expression of all constructs in mammalian cells has been validated as previously reported (Cariulo et al., 2019, Fodale et al., 2014).
HEK293T cell culture and manipulation
For Western blotting and immunoassay analysis HEK293T cells were cultured, transfected and lysed as previously described (Cariulo et al., 2019). For immunofluorescence analysis, HEK293T cells were plated in Poly-L-lysine (100 μg/ml) coated 96-well black plates with μClear bottom (catalog #655090; Greiner) at 20.000 cells per well and transfected using Lipofectamine 2000 (Thermo Fisher Scientific) according to manufacturer’s protocols. Lysis was performed 24 hours from transfection in lysis buffer (TBS, 0.4% Triton X100) supplemented with 1X protease inhibitor cocktail (catalog #11697498001; Roche) and phosphatase inhibitor cocktail (catalog #04986837001; Roche). Okadaic Acid (catalog #5934S; Cell Signaling) was delivered to HEK293T cells at a concentration of 350 nM for 1 hour at 37°C before their lysis. The IKBKB inhibitor compound (Bay-65-1942, (Atwal et al., 2011)) was sourced by the CHDI foundation and HEK293T cells were treated for 24 hours at 37°C before lysis. The combination of the treatment with OA and Bay-65-1942 was performed as follows: HEK293T cells were transfected as previously reported; after 24 hours from transfection, Bay-65-1942 compound was delivered to cells at the reported concentration, medium was changed and treatment was carried out for 24 hours; 1 hour before lysis, cells were treated with OA as previously described. Silencing via RNA interference (RNAi) was performed using specific PP2A-Cα siRNA commercially supplied (catalog #sc-43509; Santa Cruz Biotechnology). The siRNA transfection was performed as reported above, employing the siRNAs at the final concentration of 100 nM. Lysis of cells was carried out after 48 hours from transfection.
Western blot
Samples were denatured at 95°C in 4X Loading Buffer (125 mM Tris-HCl pH 6.8, 6% SDS, 4 M urea, 4 mM EDTA, 30% Glycerol, 4% 2-Mercaptoethanol and Bromophenol Blue) and loaded on NuPAGE 4-12% Bis-Tris Gel (catalog #WG1402BOX; Thermo Fisher Scientific). Proteins were transferred on PVDF membrane (catalog #162-0177; Bio-Rad Laboratories) using wet blotting. After fixing in 0.4% paraformaldehyde solution and blocking with 5% non-fat milk in TBS/0.1% Tween-20, primary antibodies incubation was carried out overnight at 4°C and secondary antibody incubations for 1 hour at room temperature. Protein bands were detected using chemiluminescent substrate (Supersignal West Femto Maximum catalog #3406; Supersignal West Pico Maximum catalog #34087; Thermo Fisher Scientific) on Chemidoc XRS+ (Bio-Rad Laboratories). Densitometric analysis was performed using ImageJ software.
Immunoprecipitation
Immunoprecipitation was performed using Dynabeads® Protein G (catalog #10004D; Thermo Fisher Scientific) following the manufacturer’s instructions and using anti-FLAG (mouse) antibody or an unrelated antibody (Anti-Glial Fibrillary Acidic Protein GFAP) as previously described (Vezzoli, Caron et al., 2019). The pulled down material was loaded on a SDS-PAGE and the Western Blot was performed as described above.
SMC assay
50 μL/well of dilution buffer (6% BSA, 0.8% Triton X-100, 750 mM NaCl, and protease inhibitor cocktail) were added to a 96-well plate (catalog #P-96-450V-C; Axygen). Samples to be tested were diluted in artificial cerebral spinal fluid (ACSF: 0.3 M NaCl; 6 mM KCl; 2.8 mM CaCl2-2H2O; 1.6 mM MgCl2-6H20; 1.6 mM Na2HPO4-7H2O; 0.4 mM NaH2PO4-H2O) supplemented with 1% Tween-20 and protease inhibitor cocktail, in a final volume of 150 μL/well. Finally, 100 μL/well of the MW1 antibody coupled with magnetic particles (appropriately diluted in Erenna Assay buffer catalog #02-0474-00; Merck), were added to the assay plate, and incubated for 1 hour at room temperature under orbital shaking for the capturing step. The beads were then washed with Erenna System buffer (catalog #02-0111-00; Merck) and resuspended using 20 μL/well of the specific detection antibody labeled with Alexa-647 fluorophore appropriately diluted in Erenna Assay buffer (anti-pS13 antibody for pS13 HTT levels readout and 2B7 for total HTT levels readout). The plate was incubated for 1 hour at room temperature under shaking. After washing, the beads were resuspended and transferred in a new 96-well plate. 10 μL/well of Erenna buffer B (catalog #02-0297-00; Merck) were added to the beads for elution and incubated for 5 minutes at room temperature under orbital shaking. The eluted complex was magnetically separated from the beads and transferred in a 384-well plate (Nunc catalog #264573; Sigma) where it was neutralized with 10 μL/well of Erenna buffer D (catalog #02-0368-00; Merck). Finally, the 384-well plate was heat-sealed and analyzed with the Erenna Immunoassay System.
TR-FRET assay
5 μL/well of samples and 1 μL/well of antibodies cocktail (2B7-Tb 1 ng/μL; FLAG-D2 10 ng/μL) diluted in lysis buffer (composition described before) were added to the 384-well assay plate (Low Volume-F Bottom catalog #784080; Greiner). Assay was performed according to (Fodale et al., 2014).
ELISA assays
Assay plates (Nunc™ MaxiSorp™ 96-well polystyrene plate catalog #449824; Thermo Fisher Scientific) were coated with 100 μL/well of the capture antibody (2B7 or anti-PP2A) diluted in TBS at the final concentration of 1 μg/mL and incubated without shaking overnight at 4°C. After washing the plate three times with wash buffer (TBS, 0.1% Tween20), the plate was blocked with 300 μL/well of BSA 1% in TBS (Blocking Buffer) for 30 minutes at room temperature and washed three times with wash buffer. 100 μL/well of the analyte diluted at the appropriate concentration in sample buffer (ACSF supplemented with 1% Tween-20 and protease inhibitor cocktail) was added to the wells and incubated for 1 hour at room temperature. After three washing steps, 50 μL/well of detection antibody (anti-FLAG-D2) diluted in Erenna Assay buffer (catalog #02-0474-00; Merck) at the final concentration of 1 μL/mL were added to the assay plate and incubated for 1 hour at room temperature. After five washing steps, 20 μL/well of Erenna buffer B (catalog #02-0297-00; Merck) were added to the assay plate and incubated for 5 minutes at room temperature under orbital shaking. 20 μL/well of the neutralization Erenna buffer D (catalog #02-0368-00; Merck) were added to the assay plate. After 5 minutes of incubation at room temperature under mild agitation, 35 μL/well of the neutralized samples were transferred in a final 384-well plate (Nunc catalog #264573; Sigma). The 384-well plate was heat-sealed and analyzed with the Erenna Immunoassay System.
High-content imaging and analysis
After 24 hours from transfection (described before), medium was removed and HEK293T cells were washed 3 times with TBS and fixed for 20 min at room-temperature in 4% paraformaldehyde/4% sucrose in TBS solution. After two washes in TBS, cells were incubated with blocking solution (1% Donkey Serum, 0.1% Triton-X, 3% BSA in TBS) for 1 hour at room temperature and then incubated with primary 2B7 and anti-FLAG (rabbit) antibodies (previously described) diluted 1:300 in staining solution (0.2% Triton-X, 3% BSA in TBS) overnight at 4 °C under constant mild agitation. The next day cells were washed 3 times in TBS and incubated for 1 hour at room temperature with donkey anti-mouse IgG antibody conjugated to Alexa-647 or donkey anti-rabbit IgG antibody conjugated to Alexa-555 (1:2000 in staining solution) provided by a commercial source (catalog #A31571 and catalog #A31572; Life Technologies) together with 1 μg/ml of Hoechst staining solution (catalog #H33342; Sigma-Aldrich). After 3 washes with TBS, images were acquired in an automated fashion employing the Incell-2000 high-content imaging system (GE-Healthcare) using a 20x objective and appropriate filters. Three independent experiments were performed. For each experiment every condition was evaluated in technical duplicates, 2 well/conditions and 5 fields per well were acquired. Imaging times settings were: 50 ms for H33342 (DAPI), 30 ms for GFP and 700 ms for Alexa-647 and Alexa-555 (2B7 and anti-FLAG). Images were analyzed using the Developer toolbox 1.9.1 software (GE-Healthcare). For counting of aggregates, a vesicle algorithm was employed identifying aggregates based on shape, size and intensity of small objects.
SMC data analysis and statistics
Normalized pS13 signal on total HTT levels measured by SMC assay was determined as the ratio between the fold increase obtained from MW1/pS13 SMC signal (pS13 HTT levels readout) and the fold increase calculated on the MW1/2B7 SMC signal (total HTT levels readout). This ratio was derived as follows: MW1/pS13SMC assay and MW1/2B7 SMC assay were performed in parallel on the same samples which were analyzed in a serial dilution curve (6 dilution points 1:3 plus blank, technical duplicates). For each readout, a curve fitting per sample (described by a four-parameter logistic curve fit) was calculated, and a fold increase among curves (which shared the top, the bottom and the slope) was assessed basing on EC50 parameter (fixing as a reference one of the analyzed samples). The statistical significance was assessed using a paired t-test (two-tailed) in the case of the analysis of n=2 samples, or using a One-way analysis of variance, applying Dunnett’s Test (which allows to compare all columns versus a fix control column) where n > 2 samples need to be compared. Graphs were generated and statistical analysis was performed using the software GraphPad Prism 6.
Author contributions
CC, PM and MV: Investigation, formal analysis and data curation
RL, LTS and CD: Funding acquisition, resources and supervision
CC, LP and AC: Conceptualization and writing - original draft
LP and AC: Project administration
Conflict of interest
All authors declare that they have no conflict of interest.
Figure legends
Acknowledgements
The work was funded by CHDI Foundation Inc., a non-profit biomedical research organization exclusively dedicated to collaboratively developing therapeutics that will substantially improve the lives of individuals with Huntington’s disease.