Hypertrophic Cardiomyopathy-Associated CRYABR123W Activates Calcineurin, Reduces Calcium Sequestration, Alters the CRYAB Interactome and the Proteomic Response to Pathological Hypertrophy

1. Introduction

Hypertrophic cardiomyopathy (HCM) is a pathologic condition defined by thickening of the ventricular wall, decreased ventricular chamber volume and diastolic dysfunction in the absence of predisposing factors, and remains the most common inherited cardiovascular condition affecting around 1:500 individuals world-wide [1]. It is inherited in an autosomal dominant pattern with a penetrance around 50-60% [2,3]. Linkage analyses of large families with HCM have identified mutations in multiple cardiac sarcomere genes, including β-myosin heavy chain (MYH7), myosin binding protein C3 (MYBPC3) and cardiac troponin T (TNNT2) [4,5,6,7,8,9]. Although the most common mutations involved in HCM are in the cardiac sarcomere genes, they only account for around 30% of mutations found in patients while a staggering 40% of patients do not carry any known HCM mutation [10,11]. Therefore, identifying and studying additional elusive causative genes outside of the sarcomere is essential to fully understand the pathogenic mechanisms behind hypertrophic cardiomyopathy [12].

Mutant alpha-crystallin B (CRYAB) was first implicated in cardiomyopathy decades ago from a genetic linkage study of a French family [13]. CRYAB is a member of the small heat shock protein family, which are molecular chaperones that are involved in a wide range of cellular functions including regulation of calcium signaling [14], protein folding and aggregation [15,16,17], autophagy [18,19] and apoptosis [20,21]. Under normal physiologic conditions CRYAB forms dimers and oligomers which mask the N- and C-terminal unstructured regions of CRYAB and thereby inhibit its function [22,23]. Following cellular stress, CRYAB de-oligomerizes exposing its N- and C-terminal domains leading to phosphorylation at different serine residues and its activation as a molecular chaperone [24,25]. CRYAB oligomers are stabilized by ionic bridges at aspartic acid residue 109 (D109) and arginine residue 120 (R120) between interacting CRYAB proteins [22,23]. Proper maintenance of CRYAB oligomers is vital to normal functioning of the protein and prevention of aberrant protein interactions, as such mutations at D109 and R120 have been implicated in various familial cardiomyopathies [26]. Several mutations have been noted at D109, but the most well studied is CRYAB^D109G which causes restrictive cardiomyopathy secondary to abnormal desmin aggregation [27]. Similarly disruptive mutations at R120 have been implicated in desmin-related cardiomyopathy with the most common mutation being CRYAB^R120G [28]. CRYAB^R120G acts in a dominant negative manner disrupting the normal functioning of wild type CRYAB and promotes protein aggregation as was noted in cryoelectron microscopy of CRYAB^R120G mutant cells with quaternary CRYAB^R120G structures double in size compared to wild type CRYAB cells [29,30]. A recently discovered novel mutation CRYAB^R123W was found to cause concordant development of HCM in monozygotic twins suggesting its pathogenic nature [31]. Mouse studies of Cryab^R123W have demonstrated a spontaneous development of diastolic dysfunction with aging and pathological hypertrophy with systolic dysfunction secondary to TAC-induced pressure overload [32]. CRYAB has also been shown to bind to calcineurin in co-immunoprecipitation experiments and CRYAB^R123W is implicated in pathologic calcium signaling [32]. Abnormal CRYAB structure secondary to the R123W mutation has been shown to have an increased B-sheet content and large structural changes adjacent to its mini-chaperone peptide sequence implicated in reduced chaperone activity while at the same time also increasing the propensity of the mutant protein to form aggregates [33].

In this study we demonstrate that CRYAB^R123W leads to increased calcineurin activity in H9C2 cells and impaired calcium uptake in mutant mouse cardiomyocytes. Structural analysis of CRYAB interaction with calcineurin using Alphafold suggests that CRYAB^R123W binds to the auto-inhibitory domain of calcineurin, supporting a possible mechanism of autoinhibition release for the increased calcineurin activity. Co-immunoprecipitation and global proteomic and phosphoproteomic mass spectrometry analysis revealed differential protein binding and regulation secondary to CRYAB^R123W indicating distinct cellular changes compared to other studied CRYAB mutations.

2. Results

CRYAB^R123W Increases Calcineurin Activity

To investigate the impact that CRYAB protein mutations have on calcineurin activity, we generated stable H9C2 cell lines overexpressing FLAG-tagged wild type, CRYAB^D109G, CRYAB^R120G, and CRYAB^R123W via lentivirus transduction. We measured calcineurin activity directly in cell lysates generated from these cell lines and an H9C2 negative control expressing normal levels of wild type CRYAB using a colorimetric calcineurin activity assay. We found a significant increase in calcineurin activity in cells overexpressing CRYAB^R123W compared to those expressing the other two mutant proteins or wild type protein (Figure 1). Overexpression of CRYAB, CRYAB^D109G or CRYAB^R120G did not promote an increase in calcineurin activity when compared to the negative control. These results indicate that the R123W variant of CRYAB is uniquely able to increase in Calcineurin activity when compared to wild type or the D109G or R120G variants.

CRYAB^R123W Causes Impaired Calcium Reuptake and Myocyte Relaxation

To examine changes in cardiomyocyte contraction and calcium handling, we isolated wild type CRYAB and CRYAB^R123W myocyte cells from mouse hearts and then performed IonOptix based assessment and analysis. No significant differences were seen in fractional shortening. We found that as compared to wild type myocytes, CRYAB^R123W myocytes had significantly impaired calcium reuptake and myocyte relaxation compared to wild type CRYAB as measured by the elevation in single exponential tau (Figure 2a). Interestingly, we also saw that CRYAB^R123W myocytes overall have a small but significant increase in sarcomere length compared to their wild type counterparts (Figure 2b). This provides evidence that the CRYAB R123W promotes dysregulation of calcium handling in mouse myocytes and is consistent with activation of calcineurin activity, NFAT reporter activity and diastolic dysfunction as shown in this study and reported previously [32].

Alphafold Predicts CRYAB^R123W Binds to the Autoinhibitory Domain of Calcineurin

To attempt to gain a better understanding of the protein structure of the CRYAB variants and their possible interaction with calcineurin, we used the Alphafold structural prediction software [34,35]. The sequences for calcineurin and wild type CRYAB were collected from the Alphafold Protein database and altered with the selected mutations [36,37]. First, we wanted to ensure that the addition of a triple FLAG tag to the structure was not predicted to have significant alterations in protein structure. A triple FLAG tag sequence was added at the N and C terminal of the wild type and CRYAB^R123W mutant sequences in Alphafold which was then visualized and compared in ChimeraX [38], with minimal alterations in protein structure following FLAG tag addition (Figure 3a-d). Second, we modeled the interaction of wild type CRYAB, CRYAB^D109G, CRYAB^R120G and CRYAB^R123W with calcineurin (Figure 4a-d). As has been previously published, wild type CRYAB binds at the NFAT binding site of calcineurin, blocking NFAT from binding to calcineurin [39] (Figure 4a). Both CRYAB^D109G and CRYAB^R120G are predicted to interact with the NFAT binding site but to a lesser extent compared to the wild type (Figure 4b,c), consistent with a loss of competition for NFAT binding model for promoting NFAT activation and cardiac hypertrophy. CRYAB^R123W, unlike the other proteins, was not predicted to bind at the NFAT binding site and thus would have complete loss of competition for NFAT binding and also was uniquely predicted to bind at the auto-inhibitory domain of calcineurin (Figure 4d). This suggests that unlike the other variants, CRYAB^R123W might have both a loss of competition for NFAT binding to calcineurin and an additional activating function by relieving the intrinsic autoinhibitory function of calcineurin to promote NFAT activation and cardiac hypertrophy. This offers a likely explanation for the significant increase in calcineurin activity seen in CRYAB^R123W compared to the other mutants.

The CRYAB^R123W Protein Interactome Diverges from CRYAB and CRYAB^R120G

To examine the impact that CRYAB mutations have on cellular protein interactions we analyzed FLAG antibody-based co-immunoprecipitation data from H9C2 cellular extract of FLAG tagged wild type CRYAB compared to FLAG tagged CRYAB^R120G and CRYAB^R123W with unbiased LC-MS proteomics. We found that despite their relative closeness in structure and mutation location CRYAB^R120G and CRYAB^R123W mutations have distinct patterns of changes in protein binding. When compared to wild type CRYAB, CRYAB^R120G seems to mostly have increased binding to proteins consistent with its documented propensity to form protein aggregates (Figure 5.1a) [29,30]. While CRYAB^R123W has both increased and decreased binding to proteins consistent with a documented loss of chaperone activity but also increase in aggregation (Figure 5.1b) [33]. When comparing CRYAB^R120G and CRYAB^R123W, we saw that the changes in protein binding between the two CRYAB mutants were distinct (Figure 5.2). CRYAB^R123W seems to have higher binding with cytoskeletal proteins, particularly proteins involved in actin binding such as Svil [40] and Lima1/EPLIN [41] (Figure 5.2). While CRYAB^R123W seems to have decreased binding to proteins involved in transcription including Brd9 [42] and Gcn1 [43], protein degradation such as Trim21 [44] and Nqo1[45] and the calcium binding ion channel Anxa2 [46] (Figure 5.2). Differential protein binding seen with CRYAB^R123W could be a driving force behind cellular pathway changes involved in the development of the HCM in contrast to the DCM phenotype seen with CRYAB^R120G or the restrictive CM phenotype seen with CRYAB^D109G.

Global Proteomics and Phosphoproteomics Reveal Cytoskeletal, Metabolic, Immune and Cardiac Dysfunction in CRYAB^R123W Mouse Hearts

To investigate the changes in cellular protein composition and pathways we compared wild type CRYAB and CRYAB^R123W global mouse heart extracts with and without pressure overload in the form of TAC. For the mice without TAC we found 6971 unique proteins, 983 unique phosphorylated proteins and 1726 proteins common to both (Figure 6a) For mice with TAC we found 4432 unique proteins, 422 unique phosphorylated proteins and 1620 proteins common to both (Figure 6d). Without additional pressure overload the Cryab^R123W mutation does not induce an HCM phenotype in our mouse model [32], a common condition of HCM mouse models [47] as few develop spontaneous HCM [48,49,50,51]. Unsurprisingly, without TAC we observed few significant changes in protein expression or phosphorylation (Figure 6b,c). However, following TAC and the emergence of a pathological hypertrophic phenotype, we observed a significant change in expression of proteins and phosphorylated proteins (Figure 6e,f), notably down regulation of Grp39, Lgals3 and Sytl3(P) while Anax6, Ppp1r1b, Slc25a15, Hrc(P) and Atxn2l(P) were upregulated (Figure 6e,f). Given that the HCM phenotype does not develop unless there is additional stress on the heart, we designated the pathology of mouse hearts that have not undergone TAC as being a model of early-stage disease (Figure 6a-c) and those with TAC as models of late-stage disease (Figure 6d-f). As previously described in our lab, the CRYAB^R123W mouse model developed systolic dysfunction following TAC [32], a poor prognostic factor of patients with HCM [52,53] illustrating how our mouse model mimics late and severe stage of HCM pathology.

Pathway analysis of the mass spectrometry proteins using GO term analysis in R [54] showed significant associations with cytoskeletal structure and actin filament pathways (Figure 7).

However, to gain a better understanding if there are other possible pathways involved, we analyzed the mass spectrometry data using QIAGEN Ingenuity Pathway Analysis (QIAGEN Inc., https://digitalinsights.qiagen.com/IPA) (IPA) [55] software comparing the early to late mouse hearts via heat maps. IPA analysis revealed significant changes in pathways involved in metabolism, cardiovascular function and immune regulation with more congruent clustering using heatmap analysis of the global mass spectrometry data. The most significant pathways that showed increased clustering comparing early to late stages of the condition included oxidative phosphorylation (Figure 8.1a,b), mitochondrial dysfunction (Figure 8.2a,b) and mTOR signaling (Figure 8.3a,b) for metabolic pathways, NFAT involvement in HCM (Figure 8.4a,b) and cardiac conduction (Figure 8.5a,b) for cardiovascular pathways and Granzyme (Figure 8.6a,b) and neutrophil trapping (Figure 8.7a,b) for immune pathways. Finally, we used IPA to analyze upstream regulators predicted to be involved in the changes seen in the protein expression of the late disease mass spectrometry data which showed significant involvement of enzymes, kinases, G-protein coupled receptors (GPCR), phosphatases and ion channels (Table 1). Our analysis demonstrates how the stress of pressure overload reveals pathological hypertrophy pathways secondary to CRYAB^R123W mutation. Specifically, there are distinct changes secondary to CRYAB^R123W induced pathological hypertrophy in the cytoskeletal, metabolic, cardiac and immune pathways. Finally distinct upstream regulators are predicted to be involved in the changes seen with pathological hypertrophy secondary to CRYAB^R123W following pressure overload by TAC could be targets for future therapeutic intervention.

3. Discussion

In this study, we used both cell lines and mouse models to examine the cellular effects of mutations in the CRYAB protein. Previous studies have demonstrated that wild type CRYAB is a protein interactor of calcineurin and suppresses the activity of calcineurin in the setting of pressure overload, but did not explain how heterozygosity for CRYAB^R123W conferred susceptibility to pathological hypertrophy [32]. In a simple competition for NFAT binding model, wild type CRYAB present in CRYAB^R123W heterozygotes would still inhibit the interaction of NFAT with calcineurin and therefore limit calcineurin dephosphorylation of NFAT and activation of hypertrophic signaling. Our results show that calcineurin activity was significantly elevated in cells overexpressing mutant CRYAB^R123W, despite the continued presence of wild type CRYAB, indicating a unique link between the mutation and aberrant calcineurin activity that is not observed with CRYAB^D109G or CRYAB^R120G. Our structural prediction using Alphafold indicates a possible interaction between CRYAB^R123W and the autoinhibitory domain of calcineurin, a key intrinsic regulatory motif [56]. This predicted interaction likely provides the mechanism behind the observed significant and unique increase in calcineurin activity.

Hypertrophic cardiomyopathy has been characterized by dysregulated fibrosis, immune regulation, inflammation and metabolism [57]. Our global proteomics analysis demonstrates that HCM, secondary to the expression of mutant CRYAB^R123W leads to dysregulated oxidative phosphorylation and mitochondrial function pathways. Additionally, hearts from mice carrying the CRYAB^R123W mutation exhibit a significant dysregulation of immune function and cytoskeletal pathways likely secondary to the impaired chaperone activity of CRYAB^R123W [33]. Wild type CRYAB is significantly expressed in cardiac tissue [58] and has been shown to readily bind to various protein interactors [59,60] including proinflammatory proteins [61] acting in an anti-inflammatory manner [62,63,64]. Therefore, it is not surprising that mutant CRYAB proteins are involved in dysregulated inflammatory and immune responses. The role of immune and inflammatory dysregulation on the pathogenesis of HCM secondary to CRYAB^R123W raises the interesting question of whether there is a role for anti-inflammatory therapeutics in the treatment of HCM.

HCM pathology has been characterized by enhanced calcium signaling regardless of underlying pathologic mutation or cause [65,66]. Wild type CRYAB plays a crucial role in the regulation of pathologic calcium signaling in the setting of pressure overload through modulation of NFAT and calcineurin activity [32,67]. We observed a significantly impaired calcium reuptake in CRYAB^R123W mutant myocytes indicating that the mutant has a role in the dysregulation of cellular calcium handling and an altered calcineurin interaction. Further investigation into the role calcineurin and calcium play in the development of HCM in general and in the setting of CRYAB^R123W is required as is the viability of calcineurin inhibitors as a therapeutic intervention for HCM patients. Our work supports the notion that precision intracellular targeting could serve as a crucial future development for the treatment of HCM patients.

CRYAB has been well documented as a molecular chaperone in a variety of different tissues across the body. Previously documented mutations in CRYAB have been linked to the development of various cardiomyopathies via the disruption of ionic bridges critically involved in the molecular chaperone function of CRYAB [22,23]. CRYAB^R123W, however, is unique in its development of HCM [31,32]. Our comparative analysis of CRYAB^R120G and CRYAB^R123W co-immunoprecipitation data suggests a distinct mechanism behind the cellular changes seen with CRYAB^R123W of which further study could illuminate novel mechanism for the development of HCM. Recent structural studies of CRYAB^R123W demonstrated reduced chaperone activity while at the same time increased aggregation to cytoskeletal proteins [33]. Similarly, our co-immunoprecipitation experiments showed significant overall loss of protein binding with CRYAB^R123W but also showed increased binding to cytoskeletal proteins. Further validation of changes in interaction partners secondary to the CRYAB^R123W mutation could illuminate possible targets for therapeutic intervention.

4. Materials and Methods

AlphaFold Protein Structure Modeling – Native amino acid sequences for CRYAB and Calcineurin were downloaded from their respective Alphafold Protein Structure Database entries [36,37]. CRYAB sequences were edited into their mutated form by replacing the corresponding amino acids respectively, D109G, R120G and R123W. All sequences, individual and multiple, were run in AlphaFold2 using the multimer setting [34,35]. Highest fidelity predicted structures based on AlphaFold2 ranked analysis were considered. Images of structures were viewed and labeled using ChimeraX developed by UCSF [38].

CRYAB DNA Insert and Plasmid Preparation – Plasmid backbone was based on the pLJM1-Empty plasmid construct (Addgene plasmid #91980 ; http://n2t.net/addgene:91980 ; RRID:Addgene_91980) [68]. CRYAB gene inserts were created using IDT gene block fragment technology(https://www.idtdna.com/page). Gene insert sequences were created with a 40 BP Gibson homology sequence from the pLJM1 plasmid at the 5’ and 3’ ends and then internally flanked by restriction enzyme sites for NheI at the 5’ end and EcoRI at the 3’ end. Between the 5’ and 3’ ends the main sequences consisted of a Kozak site, start codon, triple FLAG sequence and CRYAB sequences or CRYAB sequences and triple FLAG sequence for both N and C terminal labeling and then finally a stop codon. Wild type, CRYAB D109G, R120G and CRYAB R123W DNA sequences were used for the inserts and were retrieved from Uniprot (https://www.uniprot.org/). Linearized pLJM1-Empty plasmid was prepared using FastDigest Green Buffer (ThermoFisher #B72) and the NheI (ThermoFisher #ER0971) and EcoRI (ThermoFisher #ER0271) restriction enzymes. The solution was then incubated at 37C for 30 minutes then at 80 ^oC for 5 minutes to inactivate the restriction enzymes. CRYAB gene inserts were incorporated into the linearized pLJM1 plasmid using Gibson assembly per protocol (ThermoFisher #A46633) [69].

Bacterial Transformation – Stable competent E. coli cells (NEB #C3040) were transformed using the heat shock method as previously described using 100ng of the Gibson reaction solution [70]. Transformed E. coli cells were then grown on LB plates with ampicillin selection for 12 hours before individual colonies were selected and cultured in LB media with ampicillin selection prior to plasmid isolation.

Plasmid Isolation and Sequencing – Stable E. coli cultures were pelleted, and plasmids were isolated using Qiagen miniprep (Qiagen #27104) or Takarabio Midiprep (Takara # 740422.50) per protocol. Prior to further usage plasmids were sequenced using Sanger sequencing to confirm proper insert sequence at the Tufts University Sequencing Core.

Cell Lines - H9C2 cells purchased from ATCC (CRL-1446) were maintained on DMEM media supplemented with 10% fetal bovine serum. H9C2 cell lines expressing triple FLAG tagged wild-type human CRYAB, D109G, R120G or R123W variants were generated using DNA sequences encoding the variants that were inserted to pLenti-puro vector (#39481, Addgene). H9c2 cells were transduced with 3rd generation CRYAB-expressing lentivirus, packaged as previously described [71]. Transduced cells were selected and maintained on 10 µg/ml puromycin.

Stable CRYAB H9C2 Cell Line Preparation – Using autoclaved forceps and sterile filter paper dipped in Trypsin individual colonies of transfected H9C2 cells were selected and grown in DMEM media supplemented with 10% fetal bovine serum and puromycin to make stable cell lines. Stable H9C2 CRYAB cell lines were checked for CRYAB expression using Western blotting with a primary monoclonal mouse anti-FLAG antibody (SigmaAldrich #F1804) and secondary immunofluorescent AlexaFluor488 goat anti-mouse light chain specific antibody (Jackson ImmunoResearch #115-545-174). Wild type and mutant CRYAB cell lines with approximately equal expression were selected for use in future experiments and maintained in DMEM media supplemented with 10% fetal bovine serum and puromycin.

Calcineurin Activity Assay – CRYAB wild type and mutant H9C2 stable cells were grown to 75% confluence on 10 cm plates before being harvested and lysed with 1 mL of ice-cold lysis buffer (50 mM HEPES buffer, 150 mM NaCl, 0.1% NP-40, Halt protease and phosphatase inhibitors (ThermoFisher # 78442)). Plates with lysis buffer were left on ice for 5 minutes before cells were detached with a cell scraper and collected in an Eppendorf tube. Cells were centrifuged at 12000g for 10 minutes at 4 ^oC and then the lysate supernatant was collected. Protein concentration was determined by BCA assay and equal protein amounts were used between samples, approximately 1 mg/mL concentration. Calcineurin activity was measured using a colorimetric calcineurin assay kit (SigmaAldrich #207007) with each sample being prepared in triplicate and as specified by the manufacturer. Samples were read at 620 nm absorbance and calcineurin activity was calculated based on the following equation: (Total – Background) – (EGTA – Background).

Coimmunoprecipitation (Co-IP) – CRYAB wild type and mutant H9C2 stable cells were grown to 75% confluence on 10 cm plates before being harvested and lysed with 1 mL of ice-cold lysis buffer (50 mM HEPES buffer, 150 mM NaCl, 0.1% NP-40, Halt protease and phosphatase inhibitors (ThermoFisher # 78442)). Plates with lysis buffer were left on ice for 5 minutes before cells were detached with a cell scraper and collected in an Eppendorf tube. Cells were centrifuged at 12000g for 10 minutes at 4 ^oC and then the lysate supernatant was collected. Bradford assays were performed on all lysate samples and 1 mg of each lysate was incubated with 2 ug monoclonal mouse anti-FLAG antibody (SigmaAldrich #F1804) and incubated on a rotator at 4 ^oC for 2 hours. Then 30 µL of 50% Protein G Sepharose 4 Fast Flow beads (SigmaAldrich #GE17-0618) were added to the lysate and rotated at 4 ^oC for an additional 45 minutes. After incubation, tubes were centrifuged at 10000g for 30 seconds at 4 ^oC and the supernatant was removed. The remaining beads were washed with 200 µL cold PBS and centrifuged at 10000g for 30 seconds 3 times. Co-immunoprecipitation samples were eluted off the beads using a 5% Formic acid solution and then centrifuged at 12000g for 30 seconds. The supernatant was collected and stored in Eppendorf tubes. Co-immunoprecipitation samples were lyophilized prior to shipment to the Harvard mass spectrometry core for preparation as per their protocols below.

CRYAB Mouse Model Procedures and Global Proteomics Sample Preparation – CRYAB mouse models for both wild type and CRYAB^R123W were based on the model previously established in our lab [32]. All mice underwent echocardiography once between 10-12 weeks of age and then 5 weeks later as per protocol previously described [32]. 10–12-week-old wild type and CRYAB^R123W mice were respectively split into two groups, one without intervention and one to undergo severe (27G needle) transverse aortic constriction to induce pressure overload as previously described [72]. Following the procedure mice were allowed to recover for 5 weeks. After the second echocardiography all mice were anesthetized with isoflurane before being euthanized with cervical dislocation. The chest was cleaned with alcohol before being cut open and the heart was harvested by transecting the great vessels just above their exit from the heart. Hearts were washed and chambers emptied using ice cold PBS before being placed in an Eppendorf tube and flash frozen in liquid nitrogen. Hearts were stored at -80 ^oC until all the requisite hearts were ready for mass spectrometry preparation. Prior to mass spectrometry analysis hearts were homogenized in a lysis buffer (50 mM HEPES buffer, 150 mM NaCl, 0.1% NP-40, Halt protease and phosphatase inhibitors (ThermoFisher # 78442)) using an ultrasonic cell disrupter then centrifuged and the lysate supernatant was collected. Heart lysate samples were lyophilized prior to shipment to the Harvard mass spectrometry core for preparation as per their protocols below. All mice were handled in accordance with US National Institutes of Health standards, and all procedures were approved by the Tufts University Institutional Animal Care and Use Committee.

Co-IP Mass Spectrometry Sample Preparation - Dried IP elutions were solubilized in a buffer 200 mM EPPS pH 8.5, 8M urea, and protease inhibitors. A buffer exchange was carried out using a modified SP3 protocol [73]. Briefly, ~250 µg of Cytiva SpeedBead Magnetic Carboxylate Modified Particles (65152105050250 and 4515210505250), mixed at a 1:1 ratio, were added to each sample. 100% ethanol was added to each sample to achieve a final ethanol concentration of at least 50%. Samples were incubated with gentle shaking for 15 minutes. Samples were washed three times with 80% ethanol. Protein was eluted from SP3 beads using 200 mM EPPS pH 8.5 containing Lys-C (Wako, 129-02541). Samples were digested overnight at room temperature with vigorous shaking. The next morning trypsin was added to each sample and further incubated for 6 hours at 37º C. Acetonitrile was added to each sample to achieve a final concentration of ~33%. Each sample was labelled, in the presence of SP3 beads, with ~62.5 µg of TMTPro reagents (ThermoFisher Scientific). Following confirmation of satisfactory labelling (>97%), excess TMT was quenched by addition of hydroxylamine to a final concentration of 0.3%. The full volume from each sample was pooled and acetonitrile was removed by vacuum centrifugation. The samples were acidified with formic acid and desalted by StageTip eluted into autosampler inserts (Thermo Scientific), dried in a speedvac and reconstituted with 5% Acetonitrile, 5% formic acid for LC-MS/MS analysis.

Co-IP Liquid Chromatography and Tandem Mass Spectrometry - Mass spectrometric data were collected on an Orbitrap Eclipse mass spectrometer coupled to a Proxeon NanoLC-1000 UHPLC (Thermo Fisher Scientific). The 100 µm capillary column was packed in-house with 35 cm of Accucore 150 resin (2.6 μm, 150Å; ThermoFisher Scientific). Data were acquired for 180 min per run. A FAIMS device was enabled during data collection and compensation voltages were set at -40V, -60V, and -80V [74]. MS1 scans were collected in the Orbitrap (resolution – 60,000; scan range – 400-1600 Th; automatic gain control (AGC) – standard; maximum ion injection time – automatic). MS2 scans were collected in the Orbitrap following higher-energy collision dissociation (HCD; resolution – 50,000; AGC – 250%; normalized collision energy – 36; isolation window – 0.5 Th; maximum ion injection time – 100 ms.

Co-IP Mass Spectrometry Data Analysis - Database searching included all entries from the rat UniProt Database (downloaded in June 2022). The database was concatenated with one composed of all protein sequences for that database in the reversed order [75]. Raw files were converted to mzXML, and monoisotopic peaks were re-assigned using Monocle [76]. Searches were performed with Comet [77] using a 50-ppm precursor ion tolerance and fragment bin tolerance of 0.02. TMTPro labels on lysine residues and peptide N-termini (+304.207 Da), as well as carbamidomethylation of cysteine residues (+57.021 Da) were set as static modifications, while oxidation of methionine residues (+15.995 Da) was set as a variable modification. Peptide-spectrum matches (PSMs) were adjusted to a 1% false discovery rate (FDR) using a linear discriminant after which proteins were assembled further to a final protein-level FDR of 1% analysis [78]. TMT reporter ion intensities were measured using a 0.003 Da window around the theoretical m/z for each reporter ion. Proteins were quantified by summing reporter ion counts across all matching PSMs. More specifically, reporter ion intensities were adjusted to correct for the isotopic impurities of the different TMTPro reagents according to manufacturer specifications. Peptides were filtered to exclude those with a summed signal-to-noise (SN) < 180 across all TMT channels and < 0.5 precursor isolation specificity. The signal-to-noise (S/N) measurements of peptides assigned to each protein were summed (for a given protein). Data analysis based on the CRYAB CoIP raw spectrum file found in the MassIVE databased accession number MSV000097115.

Global Sample Preparation for Mass Spectrometry – CRYAB mouse heart samples for protein analysis were prepared essentially as previously described [79,80]. Proteomes were extracted using a buffer containing 200 mM EPPS pH 8.5, 8M urea, 0.1% SDS and protease inhibitors. Following lysis, 150 µg of each proteome was reduced with 5 mM TCEP. Cysteine residues were alkylated using 10 mM iodoacetamide for 20 minutes at RT in the dark. Excess iodoacetamide was quenched with 10 mM DTT. A buffer exchange was carried out using a modified SP3 protocol [73]. Briefly, ~1500 µg of Cytiva SpeedBead Magnetic Carboxylate Modified Particles (65152105050250 and 4515210505250), mixed at a 1:1 ratio, were added to each sample. 100% ethanol was added to each sample to achieve a final ethanol concentration of at least 50%. Samples were incubated with gentle shaking for 15 minutes. Samples were washed three times with 80% ethanol. Protein was eluted from SP3 beads using 200 mM EPPS pH 8.5 containing Lys-C (Wako, 129-02541). Samples were digested overnight at room temperature with vigorous shaking. The next morning trypsin (ThermoFisher Scientific) was added to each sample and further incubated for 6 hours at 37º C. Following digestion, an equal volume of each sample was pooled to generate a pooled sample to be used as a bridge between two TMTPro 16-plex experiments. Acetonitrile was added to each sample to achieve a final concentration of ~33%. Each sample was labelled, in the presence of SP3 beads, with ~300 µg of TMTPro reagents (ThermoFisher Scientific). Following confirmation of satisfactory labelling (>97%), excess TMT was quenched by addition of hydroxylamine to a final concentration of 0.3%. The full volume from each sample was pooled and acetonitrile was removed by vacuum centrifugation for 1 hour. The pooled sample was acidified, and peptides were de-salted using a Sep-Pak 50mg tC18 cartridge (Waters). Peptides were eluted in 70% acetonitrile, 1% formic acid and dried by vacuum centrifugation.

Global Sample Phosphopeptide Enrichment - A phosphopeptide enrichment was performed using a High-Select Fe-NTA Phosphopeptide Enrichment Kit (ThermoFisher Scientific). Dried phosphopeptides were de-salted by Stage-Tip and re-dissolved in 5% formic acid/5% acetonitrile for LC-MS/MS. The flow through from the phosphopeptide enrichment was used for total proteome profiling.

Global Sample Basic pH Reversed-phase Separation (BPRP) - TMT labeled peptides were solubilized in 5% acetonitrile/10 mM ammonium bicarbonate, pH 8.0 and ~300 µg of TMT labeled peptides were separated by an Agilent 300 Extend C18 column (3.5 μm particles, 4.6 mm ID and 250 mm in length). An Agilent 1260 binary pump coupled with a photodiode array (PDA) detector (Thermo Scientific) was used to separate the peptides. A 45-minute linear gradient from 10% to 40% acetonitrile in 10 mM ammonium bicarbonate pH 8.0 (flow rate of 0.6 mL/min) separated the peptide mixtures into a total of 96 fractions (36 seconds). A total of 96 Fractions were consolidated into 24 samples in a checkerboard fashion and vacuum dried to completion. Each sample was desalted via Stage Tips and re-dissolved in 5% formic acid/ 5% acetonitrile for LC-MS3 analysis.

Global Sample Liquid Chromatography Separation and Tandem Mass Spectrometry (LC-MS2) for Proteomics - Proteome data were collected on an Orbitrap Fusion Lumos or Orbitrap Eclipse mass spectrometer (ThermoFisher Scientific) coupled to a Proxeon EASY-nLC 1000 LC pump (ThermoFisher Scientific). Fractionated peptides were separated using a 120 min gradient at 525-650 nL/min on a 35 cm column (i.d. 100 μm, Accucore, 2.6 μm, 150 Å) packed in-house. MS1 data were collected in the Orbitrap (Fusion Lumos - 120,000 resolution; maximum injection time 50 ms; AGC 6 × 10⁵; Eclipse - 60,000 resolution; maximum injection time 50 ms; AGC 4 × 10⁵). Charge states between 2 and 6 (Fusion Lumos) or 2 and 5 (Eclipse) were required for MS2 analysis, and a 120 s dynamic exclusion window was used. Top 10 MS2 scans were performed in the ion trap with CID fragmentation (Fusion Lumos - isolation window 0.5 Da; Rapid; NCE 35%; maximum injection time 50 ms; AGC 1.5 × 10⁴; Eclipse - isolation window 0.5 Da; Turbo; NCE 35%; maximum injection time 50 ms; AGC 1 × 10⁴). For data collected on the Fusion Lumos, an on-line real-time search algorithm (Orbiter) was used to trigger MS3 scans for quantification [81]. MS3 scans were collected in the Orbitrap using a resolution of 50,000, NCE of 55%, maximum injection time of 200 ms, and AGC of 3.0 × 10⁵. The close out was set at two peptides per protein per fraction [81]. For data collected on the Eclipse, real-time search was used to trigger MS3 scans for quantification [81]. MS3 scans were collected in the Orbitrap using a resolution of 50,000, NCE of 55%, maximum injection time of 250 ms, and AGC of 1.25 × 10⁵. The close out was set at two peptides per protein per fraction [81].

Global Sample Liquid Chromatography Separation and Tandem Mass Spectrometry (LC-MS2) for Phosphoproteomics - Phosphorylation data were collected on an Orbitrap Eclipse mass spectrometer (ThermoFisher Scientific) coupled to a Proxeon EASY-nLC 1000 LC pump (ThermoFisher Scientific). Fractionated peptides were separated using a 120 min gradient at 525 nL/min on a 35 cm column (i.d. 100 μm, Accucore, 2.6 μm, 150 Å) packed in-house. A FAIMS device enabled during data acquisition with compensation voltages set as −40, −60, and −80 V for the first shot and −45 and −65 V for the second shot [74]. MS1 data were collected in the Orbitrap (120,000 resolution; maximum injection time set to auto; AGC 4 × 10⁵). Charge states between 2 and 5 were required for MS2 analysis, and a 120 second dynamic exclusion window was used. The cycle time was set at 1 second. MS2 scans were performed in the Orbitrap with HCD fragmentation (isolation window 0.5 Da; 50,000 resolution; NCE 36%; maximum injection time 300 ms; AGC 2 × 10⁵).

Global Sample Data Analysis - Raw files were converted to mzXML, and monoisotopic peaks were re-assigned using Monocle [76]. Searches were performed using the Comet search algorithm against a mouse database downloaded from Uniprot in May 2021. We used a 50-ppm precursor ion tolerance, 1.0005 fragment ion tolerance, and 0.4 fragment bin offset for MS2 scans collected in the ion trap, and 0.02 fragment ion tolerance; 0.00 fragment bin offset for MS2 scans collected in the Orbitrap. TMTPro on lysine residues and peptide N-termini (+304.2071 Da) and carbamidomethylation of cysteine residues (+57.0215 Da) were set as static modifications, while oxidation of methionine residues (+15.9949 Da) was set as a variable modification. For phosphorylated peptide analysis, +79.9663 Da was set as a variable modification on serine, threonine, and tyrosine residues. Each run was filtered separately to 1% False Discovery Rate (FDR) on the peptide-spectrum match (PSM) level. Then proteins were filtered to the target 1% FDR level across the entire combined data set. Phosphorylation site localization was determined using the AScorePro algorithm [82]. For reporter ion quantification, a 0.003 Da window around the theoretical m/z of each reporter ion was scanned, and the most intense m/z was used. Reporter ion intensities were adjusted to correct for isotopic impurities of the different TMTPro reagents according to manufacturer specifications. Peptides were filtered to include only those with a summed signal-to-noise (SN) ≥ 170 or 180 across all TMT channels. An extra filter of an isolation specificity (“isolation purity”) of at least 0.5 in the MS1 isolation window was applied for the phosphorylated peptide analysis. For each protein or phosphorylation site, the filtered peptide TMTPro SN values were summed to generate protein or phosphorylation site quantification values. The signal-to-noise (S/N) measurements of peptides assigned to each protein were summed (for a given protein). These values were normalized so that the sum of the signal for all proteins in each channel was equivalent thereby accounting for equal protein loading. The resulting normalization factors were used to normalize the phosphorylation sites, again accounting for equal protein loading. Data analysis based on the CRYAB NonTAC Proteome 1-12 and Phospho 1-2 raw spectrum files and the CRYAB TAC Proteome 1-12 and Phospho 1-3 raw spectrum files found in the MassIVE database accession number MSV000097115.

Cardiomyocyte Isolation and IonOptix – Wild type Cryab and Cryab^R123W mouse hearts were extracted, and single ventricular cardiomyocytes were isolated using the O’Connell and Simpson modification method and apparatus as previously described [83,84,85,86]. Single cardiomyocyte calcium transient and contraction dynamics were captured and analyzed using IonOptix equipment and software as previously described [85].

Data and Statistical Analysis – All bar plots, volcano plots, GO term plots and heat maps were prepared in R [54] using the following packages: ggplot2, dplyr, reshape2, ggpubr, tidyverse, RColorBrewer, ggrepel, clusterProfiler, DOSE, org.Mm.eg.db, enrichplot, ggupset, and pheatmap [87,88,89,90,91,92,93,94,95,96,97,98,99]. Global proteomic and phosphoproteomic data was analyzed using QIAGEN Ingenuity Pathway Analysis software (IPA) [55]. All statistical measurements are shown as mean +/- standard deviation. All p-values were calculated using Student T Test for 2 group comparisons and by two-way ANOVA for multi-group comparisons.