Calcineurin B homologous protein 3 (CHP3) is an EF-hand Ca2+-binding protein involved in regulation of cancerogenesis, cardiac hypertrophy, and neuronal development through interactions with sodium/proton exchangers (NHEs) and signalling proteins. While the importance of Ca2+ binding and myristoylation for CHP3 function has been recognized, the underlying molecular mechanism remained elusive. In this study, we demonstrate that Ca2+ binding and myristoylation independently affect the conformation and functions of human CHP3. Ca2+ binding increased local flexibility and hydrophobicity of CHP3 indicative of an open conformation. The Ca2+-bound CHP3 exhibited a higher affinity for NHE1 and associated stronger with lipid membranes compared to the Mg2+-bound CHP3, which adopted a closed conformation. Myristoylation enhanced the local flexibility of CHP3 and decreased its affinity to NHE1 independently of the bound ion, but did not affect its binding to lipid membranes. The data exclude the proposed Ca2+-myristoyl switch for CHP3. Instead, a Ca2+-independent exposure of the myristoyl moiety is induced by binding of the target peptide to CHP3 enhancing its association to lipid membranes. We name this novel regulatory mechanism ‘target-myristoyl switch’. Collectively, the interplay of Ca2+ binding, myristoylation, and target binding allows for a context-specific regulation of CHP3 functions.

Calcineurin B homologous protein 3 (CHP3) is an EF-hand Ca2+-binding protein involved in regulation of cancerogenesis, cardiac hypertrophy, and neuronal development through interactions with sodium/proton exchangers (NHEs) and signalling proteins. While the importance of Ca2+ binding and myristoylation for CHP3 function has been recognized, the underlying molecular mechanism remained elusive. In this study, we demonstrate that Ca2+ binding and myristoylation independently affect the conformation and functions of human CHP3. Ca2+ binding increased local flexibility and hydrophobicity of CHP3 indicative of an open conformation. The Ca2+-bound CHP3 exhibited a higher affinity for NHE1 and associated stronger with lipid membranes compared to the Mg2+-bound CHP3, which adopted a closed conformation. Myristoylation enhanced the local flexibility of CHP3 and decreased its affinity to NHE1 independently of the bound ion, but did not affect its binding to lipid membranes. The data exclude the proposed Ca2+-myristoyl switch for CHP3. Instead, a Ca2+-independent exposure of the myristoyl moiety is induced by binding of the target peptide to CHP3 enhancing its association to lipid membranes. We name this novel regulatory mechanism ‘target-myristoyl switch’. Collectively, the interplay of Ca2+ binding, myristoylation, and target binding allows for a context-specific regulation of CHP3 functions.

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Calcineurin B homologous protein 3 (CHP3) is an EF-hand Ca2+-binding protein involved in regulation of cancerogenesis, cardiac hypertrophy, and neuronal development through interactions with sodium/proton exchangers (NHEs) and signalling proteins. While the importance of Ca2+ binding and myristoylation for CHP3 function has been recognized, the underlying molecular mechanism remained elusive. In this study, we demonstrate that Ca2+ binding and myristoylation independently affect the conformation and functions of human CHP3. Ca2+ binding increased local flexibility and hydrophobicity of CHP3 indicative of an open conformation. The Ca2+-bound CHP3 exhibited a higher affinity for NHE1 and associated stronger with lipid membranes compared to the Mg2+-bound CHP3, which adopted a closed conformation. Myristoylation enhanced the local flexibility of CHP3 and decreased its affinity to NHE1 independently of the bound ion, but did not affect its binding to lipid membranes. The data exclude the proposed Ca2+-myristoyl switch for CHP3. Instead, a Ca2+-independent exposure of the myristoyl moiety is induced by binding of the target peptide to CHP3 enhancing its association to lipid membranes. We name this novel regulatory mechanism 'target-myristoyl switch'. Collectively, the interplay of Ca2+ binding, myristoylation, and target binding allows for a context-specific regulation of CHP3 functions. Editor's evaluation In this work, the authors provide important mechanistic insights into how the intracellular effector protein Calcineurin B homologous protein 3 (CHP3) can be regulated in a calcium-independent manner to expose its lipid membrane binding site. Compelling evidence demonstrates a binding partner protein (NHE1) triggers a conformation change and exposure of the myristoyl group in CHP3 resulting in membrane association. This provides mechanistic insight into the signalling mechanisms achieved by CHP3 in a target-binding dependent manner, which will be of broad scientific interest. https://doi.org/10.7554/eLife.83868.sa0 Decision letter Reviews on Sciety eLife's review process Introduction The calcineurin B homologous protein 3 (CHP3, tescalcin) belongs to the EF-hand Ca2+-binding protein (EFCaBP) family (Kolobynina et al., 2016) and is closely related to calcineurin B homologous proteins CHP1 and CHP2 (Di Sole et al., 2012). CHPs interact with several isoforms of sodium/proton exchangers (NHEs), and this interaction is required for the localization and function of NHE transporters on the plasma membrane (Pedersen and Counillon, 2019). Besides NHEs, CHP3 interacts with calcineurin A (Gutierrez-Ford et al., 2003), subunit 4 of COP9 signalosome (Levay and Slepak, 2014) and glycogen synthase kinase 3 (GSK3) (Kobayashi et al., 2015). Recent studies revealed that the CHP3 expression level correlates with the progression, metastasis, and invasiveness of gastric, renal, and colorectal cancers (Kang et al., 2016; Kim et al., 2019; Lee et al., 2018; Luo et al., 2019). In addition, genome-wide association studies in combination with neuroimaging identified CHP3 as a key regulator of neurogenesis for hippocampal volume formation (Dannlowski et al., 2015; Horgusluoglu-Moloch et al., 2019; Stein et al., 2012). Via regulation of GSK3 and calcineurin activities, it seems to counteract cardiac hypertrophy (Kobayashi et al., 2015; Viereck et al., 2020). Thus, CHP3 is an emerging important player in cellular Ca2+ signalling networks involved in the regulation of cell proliferation and development in different tissues. The underlying molecular mechanisms are not well understood. All three CHPs were shown to undergo conformational changes upon Ca2+ binding (Gutierrez-Ford et al., 2003; Liang et al., 2020). Notably, CHP3 has a lower affinity for Ca2+ (0.8 µM) (Gutierrez-Ford et al., 2003) in comparison to CHP1 and CHP2 (KD values below 100 nM) (Li et al., 2011; Pang et al., 2004). Further, Ca2+-binding affinities of CHP1 and CHP2 strongly increased (45- and 42-fold, respectively) upon binding of the NHE1 target peptide (Li et al., 2011; Pang et al., 2004). This modulation is a common feature in EFCaBPs (Gifford et al., 2007) and should also apply to the CHP3 isoform. CHP3 may thus respond to Ca2+ signals with conformational changes when the intracellular Ca2+ concentration elevates from 100 nM to 1 µM or higher (Roderick and Cook, 2008). Ca2+-induced conformational changes are characteristic for Ca2+ sensor proteins such as calmodulin (CaM) and calcineurin B (Creamer, 2020; Nelson and Chazin, 1998). At the resting Ca2+ level, they adopt a closed conformation, in which the hydrophobic target-binding pocket is occluded. Upon a rise of the intracellular Ca2+ concentration, Ca2+ binding to EF-hand(s) causes an opening of this pocket providing the structural basis for the transmission of Ca2+ signals (Nelson and Chazin, 1998). This opening often triggers the binding of EFCaBPs to their target proteins (Burgoyne et al., 2019). In addition, Ca2+-induced conformational changes of EFCaBPs such as calcineurin B or guanylyl cyclase activating proteins (GCAPs) stably associated with target proteins can affect the function of the latter (Creamer, 2020; Lim et al., 2014). Recently, we demonstrated an increase of CHP3 hydrophobicity upon Ca2+ binding that most likely resulted from the opening of the hydrophobic target-binding pocket (Liang et al., 2020). CHP3 binds also Mg2+ with a low affinity (KD = 73 µM) in the absence of Ca2+. The presence of 1 mM Mg2+ reduces the affinity for Ca2+ from 0.8 to 3.5 µM, indicating a direct competition between Ca2+ and Mg2+ for the EF-3 binding site (Gutierrez-Ford et al., 2003). Thus, CHP3 should be present in the Mg2+-bound state in a cell at basal Ca2+ concentration, ready to respond to Ca2+ signals, yet, the exact mechanism is not fully described. In addition, CHP3 harbors an N-terminal myristoylation site similar to other EFCaBPs such as recoverin and calcineurin B (Gutierrez-Ford et al., 2003; Zaun et al., 2012). This modification often enhances protein binding to cellular membranes, but it can also stabilize the structure of a protein and/or regulate its function (Yuan et al., 2020; Jiang et al., 2018). Membrane binding via the myristoyl moiety is usually enforced with either a second lipidation site or clusters of positively charged and/or hydrophobic residues (Yuan et al., 2020; Jiang et al., 2018). The exposure of the myristoyl moiety from the modified protein and thereby its binding to lipid membranes can be regulated by various signals, for instance by exchange of GDP to GTP (GTP-myristoyl switch in ADP ribosylation factor 1 [ARF1] GTPase) (Goldberg, 1998), by phosphorylation (myristoyl/phosphoserine switch in myristoylated alanine-rich C kinase substrate (MARCKS) [Braun et al., 2000] and in the C-subunit of protein kinase A [Gaffarogullari et al., 2011]) or by pH change (pH-dependent myristoyl-histidine switch in hisactophilin [Hanakam et al., 1996]). In many Ca2+ sensor proteins (for instance recoverin, neurocalcin, visinin-like proteins), the myristoyl group becomes accessible to lipid membranes after Ca2+ binding (Lim et al., 2014). This mechanism is called Ca2+-myristoyl switch. However, not all myristoylated EFCaBPs have a Ca2+-myristoyl switch. In the neuronal calcium sensor-1 (NCS-1) and in KChIP1, the myristoyl moiety becomes exposed in the presence of lipid membranes even at low Ca2+ concentration (Lim et al., 2014; McFerran et al., 1999; O'Callaghan and Burgoyne, 2004; O'Callaghan et al., 2003); and it is constantly hidden within the protein core in GCAPs (Lim et al., 2014). Myristoylated CHP1 binds to microsomal membranes in a Ca2+-dependent manner indicating the presence of a Ca2+-myristoyl switch (Andrade et al., 2004), whereas binding of myristoylated CHP3 to lipid membranes has not been reported so far. Co-expression of NHE1 with CHP3 that lacks myristoylation and/or Ca2+-binding sites significantly reduced the half-life at the cell surface and the activity of this transporter (Zaun et al., 2012). Simultaneous myristoylation and Ca2+ binding was suggested to be important for NHE1 stabilization by CHP3 (Zaun et al., 2012). Based on these results, the presence of a Ca2+-myristoyl switch for CHP3 was proposed (Gutierrez-Ford et al., 2003; Zaun et al., 2012), though the exposure of the myristoyl group in response to Ca2+ binding has not been shown experimentally. The interaction of CHP3 with NHE1 is an ideal system to analyse the effects of myristoylation and Ca2+ binding on CHP3. We previously demonstrated that CHP3 binds at 1:1 ratio to the CHP-binding region of human NHE1 (CBD) with high affinity in the presence of Mg2+ (Fuchs et al., 2018). At the same time, it was shown by co-immunoprecipitation of CHP3-myc and NHE1-HA that addition of Ca2+ increased the amount of the complex formed (Zaun et al., 2012). Here, we show that Ca2+ and N-terminal myristoylation independently regulate the conformation of CHP3 and its interaction with NHE1 providing the molecular basis for regulation of CHP3 function. This excludes a Ca2+-myristoyl switch in CHP3, instead, surface exposure of the myristoyl moiety was triggered by target peptide binding as probed by interaction with liposomes. We named this novel mechanism 'target-myristoyl switch'. Our study provides fundamental mechanistic understanding of the regulation of CHP3 function. Results Pure fully myristoylated and non-myristoylated untagged CHP3 are functional in binding a single calcium ion In order to dissect the effects of Ca2+ and myristoylation on target and lipid binding of human CHP3, we aimed for pure untagged CHP3 and myristoylated CHP3 (mCHP3). Non-myristoylated CHPs were previously produced with affinity tags and purified by corresponding affinity chromatography (Gutierrez-Ford et al., 2003; Liang et al., 2020; Fuchs et al., 2018). Yet, the peptide tag used for affinity purification of rat CHP1 was shown to interact with the hydrophobic target-binding pocket (Naoe et al., 2005) and might interfere with conformational changes. Myristoylated untagged CHP1 had been produced with low yield by co-expression with yeast N-myristoyltransferase (Timm et al., 1999). Here, we produced untagged CHP3 and mCHP3, the latter by co-expression with human N-myristoyltransferase 1, and used Ca2+-dependent hydrophobic interaction chromatography (HIC) for purification. HIC has been previously used for purification of other EFCaBPs including CaM and calcineurin B (Tanaka et al., 1984; Wei and Lee, 1997). Combining HIC and gel filtration, we obtained pure untagged proteins, as shown by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) analysis (Figure 1A) with an average yield of 9 mg/l of expression culture. Single symmetrical peaks in the elution profiles of analytical gel filtration indicated monodisperse protein preparations of CHP3 and mCHP3. Both proteins form only monomers under reducing conditions in the presence of 2 mM TCEP (Figure 1—figure supplement 1). The slightly higher electrophoretic mobility of mCHP3 compared to CHP3 resolved in high-resolution SDS–PAGE analysis already indicated that the protein was covalently modified (Figure 1A). To check the degree of myristoylation, we analysed mCHP3 and CHP3 by native mass spectrometry (MS). The measured mass of mCHP3 was increased by ~211 Da (Figure 1B, bottom) in comparison to CHP3 (Figure 1B, top), which is in agreement with the covalent attachment of one myristoyl group (Mr = 210 Da). Native MS analysis of three independently produced and purified samples confirmed reproducible complete myristoylation of recombinant CHP3. Interestingly, the ion mobility arrival time was shorter for mCHP3 (Figure 1C) indicating a more compact shape of mCHP3 compared to CHP3. To reveal the stoichiometry of Ca2+ binding, we performed native MS analysis of CHP3 and mCHP3 after addition of Ca2+. The molecular masses of both proteins were shifted by ~40 Da (Figure 1B), which corresponds to the binding of a single Ca2+ ion and, thus, documents that both recombinant CHP3 and mCHP3 are functional in respect to Ca2+ binding and confirms the presence of one Ca2+-binding site (Gutierrez-Ford et al., 2003; Perera et al., 2001). Figure 1 with 2 supplements see all Download asset Open asset Quality control of the purified non-myristoylated CHP3 and myristoylated CHP3 (mCHP3). (A) Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) analysis (16%, tricine mini gels) showed a mobility shift of mCHP3 in comparison to CHP3; positions of co-migrated molecular mass standards are indicated in kDa on the left; Figure 1—source data 1: Full gel for (A). (B) Left, deconvoluted mass spectra of CHP3 (top) and mCHP3 (bottom) show a measured intact protein mass of 24.617 and 24.828 kDa (∆Mr = 211 Da), which is in line with the addition of the myristoyl group of 210 Da. For mCHP3, also low abundant dimeric species were observed by native MS analysis. Right, zoom-in spectra in the range of 24.8 ± 0.3 kDa that show differences in masses as indicated by dashed lines. In the presence of Ca2+, the measured intact protein mass of CHP3 and mCHP3 was increased by ~40 Da (mass accuracy ± 1 Da), respectively, indicating binding of one Ca2+ ion. (C) Ion mobility arrival time distributions for the +8 charge state of CHP3 (solid line) and mCHP3 (dashed line). A shoulder with shorter arrival times indicates the presence of a low abundant dimeric species for CHP3 and mCHP3. Arrival times of mCHP3 were reduced by 1.36 ms (±0.37 ms; n = 3; p = 0.024) compared to CHP3. See Figure 1—figure supplement 2 for non-deconvoluted mass spectra showing charge state distributions of CHP3 and mCHP3 measured in the positive ion mode by native MS. Figure 1—source data 1 Full gel for Figure 1A. https://cdn.elifesciences.org/articles/83868/elife-83868-fig1-data1-v2.zip Download elife-83868-fig1-data1-v2.zip Ca2+-induced conformational changes are similar in CHP3 and mCHP3 Next, we analysed the effect of Ca2+ binding in the presence of Mg2+ and of the N-terminal myristoylation on CHP3 conformation. For this purpose, we optimized the FPH (fluorescence probe hydrophobicity) assay, previously developed to monitor Ca2+-induced conformational changes of CHPs (Liang et al., 2020), by using the dye ProteOrange (Lumiprobe) at defined micromolar concentration (see Materials and methods). In this assay, the fluorescence of the dye strongly increases upon its binding to hydrophobic protein surfaces (Niesen et al., 2007). In the Mg2+-bound state, CHP3 and mCHP3 showed low fluorescence (Figure 2—figure supplement 1A). After Ca2+ addition, the fluorescence intensity increased by 30% and 40% for CHP3 and mCHP3, respectively, reflecting an increase in hydrophobicity (Figure 2A). Ca2+ removal by EGTA reduced the fluorescence to the initial level, indicating the reversibility of Ca2+ binding and of the respective conformational changes. Depletion of both Mg2+ and Ca2+ by EDTA turned the protein into the non-physiological apo-state, the fluorescence was in between that of the Mg2+- and Ca2+-bound states. The proteins appeared to be destabilized in the apo-state and could not be reverted into the functional form by Ca2+ addition (Figure 2A). Interestingly, the N-terminal myristoylation did not affect the conformational changes of CHP3. As a control, we probed the hydrophobicity of recoverin, the prototypic protein with a classical Ca2+-myristoyl switch (Ames et al., 1997; Figure 2—figure supplement 1B). Ca2+ induced similar changes of non-myristoylated recoverin as of CHP3 in the FPH assay, whereas hydrophobicity of the myristoylated protein increased much stronger (about 3.5-fold) in response to Ca2+ binding, which most likely resulted from the exposure of the myristic group. Figure 2 with 2 supplements see all Download asset Open asset Ca2+-induced conformational changes in CHP3 and mCHP3. (A) Kinetic fluorescence probe hydrophobicity (FPH) assay. Fluorescence of dye bound at hydrophobic protein surfaces was monitored at λem = 585 nm (excitation λex = 470 nm) and at 22°C. Protein (1.5 µM) was prepared in the Mg2+-bound state (2 mM MgCl2, 1 mM EGTA). CaCl2, EDTA, and EGTA were sequentially added as indicated. First, 2 mM CaCl2 was added and then chelated with addition of 3 mM EGTA. Next, 3 mM EDTA was added to remove both divalent ions (CHP3 in apo-state) followed by another addition of 4 mM CaCl2. (B) Intrinsic tryptophan fluorescence was monitored at λem = 330 nm (excitation λex = 280 nm) at 22°C. Protein (2.5 μM) in the Mg2+-bound state was used and the additions were performed as described above for the FPH assay. (C) AlphaFold2.0 model (Varadi et al., 2022) of CHP3 in surface presentation, with surface coloured for hydrophobicity (Eisenberg et al., 1984). The model most likely resembles the open or target-bound conformation. The single tryptophan residue highlighted in green (Trp191) is located in the hydrophobic target-binding pocket. EC50 values for binding of Ca2+ to CHP3 (D) and mCHP3 (E) determined with FPH assay. Fluorescence of samples with CHP3 or mCHP3 at different Ca2+ concentrations was measured at 590 nm in the presence of 2 mM MgCl2. Three biological replicates (shown in different colours) with three to four technical replicates each for CHP3 and mCHP3 were measured. The data were fitted with Hill equation using global non-linear regression. (F) 95% confidence intervals (Δχ2 of 3.84) of EC50 values for binding of Ca2+ calculated with profile likelihood method. EC50 values (asterisks) are shown with confidence intervals in square brackets below the graph. To prove that changes in the dye-mediated fluorescence observed in the FPH assay were indeed caused by conformational changes, we probed the intrinsic tryptophan fluorescence of CHP3 and mCHP3. According to the 3D model of CHP3 predicted with AlphaFold2.0 (Varadi et al., 2022), the single tryptophan residue (Trp191) is located in CHP3's hydrophobic target-binding pocket at the protein surface (Figure 2C) in line with the measured emission maxima of CHP3 and mCHP3 of ~330 nm (Figure 2—figure supplement 2). Intensity and emission maximum of intrinsic tryptophan fluorescence depend on the local environment of the residue (Vivian and Callis, 2001; Eftink, 2000). Ca2+ addition decreased the intrinsic fluorescence of CHP3 (Figure 2B) reflecting conformational changes accompanied with changes in the local environment of Trp191 in the hydrophobic pocket. This is in line with an increase of CHP3 hydrophobicity observed with the FPH assay. These conformational changes monitored by intrinsic fluorescence were also reverted by Ca2+ removal (EGTA addition) (Figure 2B). Only for mCHP3, the removal of both Mg2+ and Ca2+ caused irreversible changes in the intrinsic fluorescence. In order to evaluate whether myristoylation affects the Ca2+-binding affinity of CHP3, we determined the EC50 values for binding of Ca2+ to CHP3 and mCHP3 in the presence of Mg2+ using the FPH assay. Addition of Ca2+ at saturating concentration increased the fluorescence (Figure 2A). We now titrated Ca2+ concentration from 0.3 µM to 8 mM and measured fluorescence using biological and technical replicates (Figure 2D, E). We determined Ca2+ EC50 values performing a global fit for all data. The difference between EC50's obtained for CHP3 (161.1 [112.6; 233.8] μM; Figure 2D) and mCHP3 (152.6 [113.7; 209.7] µM; Figure 2E) is insignificant, indicating that myristoylation does not affect the Ca2+-binding properties of CHP3. Combining the data of the FPH assay and of the intrinsic tryptophan fluorescence, we conclude that CHP3 undergoes reversible Ca2+-induced conformational changes as typical for a Ca2+ sensor protein. The hydrophobic target-binding pocket of CHP3 is occluded in the Mg2+-bound state (closed conformation), and becomes exposed to the environment upon Ca2+ binding (open conformation). Ca2+ binding and myristoylation independently affect thermal stabilities of CHP3 and its complex with NHE1 target peptide To further dissect the effects of Ca2+ and N-terminal myristoylation on CHP3, we probed the thermal stability of CHP3 alone and in complex with the CHP-binding domain of NHE1 (CBD, NHE1 residues 525–545) using nano-differential scanning fluorimetry (nanoDSF). To obtain the CHP3:CBD complex used for nanoDSF, we co-expressed CHP3 and CBDHis using the pETDuet-1 system and purified the resulting complex. For the myristoylated complex (mCHP3:CBD), simultaneous co-expression of three proteins (CHP3, CBDHis, and human N-myristoyltransferase 1 (NMT1)) from a modified pETDuet-1 vector was performed, and the resulting complex was purified. Myristoylation of CHP3 in the complex was confirmed by electrospray ionization (ESI)-TOF mass spectrometric analysis. Free CHP3 had the highest thermal stability in the Mg2+-bound state (Tmapp 70.1 ± 0.3°C), and Ca2+ binding reduced the melting point to 66.9 ± 0.2°C (CHP3 in Figure 3). In the presence of both ions, the Ca2+ effect appeared to be dominant with a melting temperature (Tmapp) of 65.0 ± 0.9°C. In line with the results of the FPH assay, which indicated a destabilized apo-state, the latter showed strongly reduced thermal stability (Tmapp 56.1 ± 0.5°C). Notably, binding of the target peptide CBD strongly increased the thermal stability for the Ca2+-bound and apo-states (compare CHP3 and CHP3:CBD in Figure 3), with about 12°C in the presence of Ca2+, 13°C in the presence of Ca2+ plus Mg2+ and about 11°C for the apo-state. Thus, the CHP3:CBD complex showed the highest thermal stability when Ca2+ was bound. Surprisingly, CBD binding did not have any effect on the thermal stability of the Mg2+-bound state. Figure 3 with 1 supplement see all Download asset Open asset Ca2+ and target peptide binding affect the thermal stability of CHP3 and mCHP3. Thermal stabilities of free proteins and complexes with the target peptide CBD were measured with nanoDSF in the presence of either 10 mM Mg2+, Ca2+, Ca2+ + Mg2+, or EDTA. Temperatures of thermal unfolding (apparent Tm) are shown as mean ± standard deviation (SD) of three independent biological replicates. Raw nanoDSF traces are shown in Figure 3—figure supplement 1. The N-terminal myristoylation decreased the Tmapp of CHP3 in all states to the same degree (9.8°C for Mg2+-bound and 10.6°C for Ca2+-bound states, 10.9°C in the presence of both ions, 4.5°C for the already destabilized apo-state) (compare CHP3 and mCHP3 in Figure 3). A similar destabilizing effect of the myristoylation was observed for the mCHP3:CBD complex with a decrease of Tmapp in the range from 6.4°C for the Ca2+-bound state to 3.2°C for the apo-state (compare CHP3:CBD and mCHP3:CBD in Figure 3). Thus, the N-terminal myristoylation lowers the stability of CHP3 independently of the bound ion and of the target binding that means independent of the conformation. Ca2+ binding to CHP3 allows more effective cleavage within EF-2, whereas CBD binding drastically increases proteolytic stability In order to pinpoint flexible regions of CHP3 in a given conformation and to probe whether these regions are affected by myristoylation, ion and target binding, we performed limited trypsinolysis of CHP3, mCHP3, and their complexes with CBD in the presence of Ca2+, Mg2+, or EDTA. CHP3 was readily proteolysed by trypsin with nearly one half of the protein already degraded after 5 min and almost no full-length protein remained after 60-min incubation (Figure 4A, top, FL). Trypsin cleavage sites (Arg and Lys residues) are distributed all over the CHP3 amino acid sequence (Figure 4—figure supplement 1A), yet productive cleavage requires not only availability of the site but also flexibility of the peptide chain at the cleavage site to adapt to the active site of the protease (Teilum et al., 2009). During the reaction, we observed the appearance of two major fragments (labelled 1 and 2 in Figure 4A). We analysed those fragments with MS to determine exact masses and precisely locate the cleavage sites (Figure 4—figure supplement 1C). Fragment 1 has a mass of 17.92 kDa corresponding to residues 2–155 of CHP3 (UniProtID Q96BS2-1) lacking the C-terminal part (Figure 4C). Fragment 2 has a mass of 9.63 kDa and includes residues 74–155. Thus, it derived from a cleavage of fragment 1 within the predicted EF-2 (Figure 4C). Noteworthy, both fragments still contain the single functional EF-hand of CHP3 (EF-3). Target peptide binding nearly completely prevented the trypsin cleavage, and the intensity of the full-length CHP3 band was only slightly reduced even after 60-min incubation (Figure 4A, bottom). Next, we compared the proteolytic stability of CHP3 in different states. In the presence of only Ca2+, Mg2+, or both ions together, the degradation rate of the full-length protein (FL) was comparable, whereas the apo-state degraded much faster (Figure 4B, FL). Fragment 1 appeared already after 5 min and degraded further over time (Figure 4B, 1), its cleavage was more pronounced in the presence of Ca2+ and was accompanied with an increase of fragment 2 (Figure 4B, 2). This indicates that the cleavage site located in EF-2 becomes more available for the protease in the open conformation (Ca2+ bound). N-terminal myristoylation did not affect the proteolytic stability of the full-length protein, whereas it slightly reduced the stability of fragment 1 in all states. In addition, a second minor band below fragment 1 appeared for mCHP3 (Figure 4B). The small differences in the tryptic cleavage of CHP3 and mCHP3 indicate a local destabilizing effect of the myristoyl moiety, in line with the lower thermal stability of mCHP3 compared to CHP3. The complex formation with CBD drastically increased the proteolytic stability of both CHP3 and mCHP3, as nearly no degradation occurred under all analysed experimental conditions (Figure 4B, CHP3:CBD and Figure 4—figure supplement 1B). Figure 4 with 2 supplements see all Download asset Open asset Ca2+-binding and complex formation change the accessibility of trypsin cleavage sites in CHP3 and mCHP3. (A) Time-dependent (0–60 min) limited proteolysis (trypsin) of CHP3 (top) and the complex of CHP3:CBD (bottom) in the presence of both Mg2+ and Ca2+. Positions of full-length protein (FL) and two major proteolytic fragments (1 and 2) as well as CBD are indicated on the right of the Coomassie-stained SDS–PAGE gel, positions of co-separated molecular mass standards (mass in kDa) – on the left; the sample containing only trypsin was loaded on the first lane (T); Figure 4—source data 1: Full gels of (A). (B) Time-dependent limited proteolysis of CHP3 and mCHP3 in the presence of Mg2+, Ca2+, both ions or in the absence of them. Sections of the gel with bands corresponding to the full-length protein (FL) and two major proteolytic fragments (1 and 2) are shown. Nearly no degradation was observed for CHP3 and mCHP3 in the complex with CBD in all conditions (Mg2+ + Ca2+ condition is presented here, other gels are shown in Figure 4—figure supplement 1B and its source data). (C) Schematic representation of full-length CHP3 with indication of N- and C-lobes, four EF-hand motifs (active EF-3 is highlighted in green). Proteolytic fragments 1 and 2 and trypsin cleavage sites were identified by mass spectrometry. (D) Combined ribbon and surface presentation of the CHP3 AlphaFold2.0 model (Varadi et al., 2022) with N- and C-lobes shown in blue and orange, respectively, and the connecting CHP-loop in cyan; the two major trypsin cleavage sites are highlighted in yellow and Ca2+ ion as a green sphere. The Ca2+ position in EF-3 was modelled by superimposition of the CHP3 model with the CHP1 X-ray structure, pdb ID 2ct9 (Andrade et al., 2004). Figure 4—source data 1 Full gels for Figure 4A. https://cdn.elifesciences.org/articles/83868/elife-83868-fig4-data1-v2.zip Download elife-83868-fig4-data1-v2.zip We mapped both major cleavage sites on the 3D model of CHP3 predicted with AlphaFold2.0 (Varadi et al., 2022; Figure 4D). This model closely resembles the structures of CHP1 and CHP2 in complex with CBD (Ammar et al., 2006; Mishima et al., 2007), of CHP1 in complex with full-length NHE1 (Dong et al., 2021) and of CHP1 with an artificial C-terminal helix bound in the target-binding pocket (Kennedy et al., 1996), that is structures of target-bound CHPs. All these structures represent the Ca2+-bound state. The cleavage site R155/S156 is located in this model at the N-side of the incoming α-helix of EF-4 and though surface exposed the α-helical location may hamper the proteolytic digest (Fontana et al., 1997). The other site, R73/K74 is located in the middle of the EF-2 loop (non-functional Ca2+-binding loop) (Figure 4D). Though the site appears to be surface exposed in the model, one should note that CHP3 has a 9-amino acid long insertion in that position as compared to CHP1 and CHP2 (Figure 4—figure supplement 2). The structure of CHP3 may deviate from the model and information for other conformations is lacking. Clearly, limited proteolysis showed that both sites are accessible and sufficiently flexible for productive cleavage in the free CHP3, yet become protected against cleavage upon complex formation. Both major cleavage sites are highly unlikely to be covered by CBD in the complex, as they are facing away from the target-binding pocket (Figure 4D). Thus, the drastic changes in the proteolytic stability indicate reduced flexibility in EF-2 and EF-4, which could be caused by structural rearrangements of CH

BACKGROUND:Time-series forecasting models play a central role in guiding intensive care coronavirus disease 2019 (COVID-19) bed capacity in a pandemic. A key predictor of future intensive care unit (ICU) COVID-19 bed occupancy is the number of new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in the general population, which in turn is highly associated with week-to-week variability, reporting delays, regional differences, number of unknown cases, time-dependent infection rates, vaccinations, SARS-CoV‑2 virus variants, and nonpharmaceutical containment measures. Furthermore, current and also future COVID ICU occupancy is significantly influenced by ICU discharge and mortality rates. METHODS:Both the number of new SARS-CoV‑2 infections in the general population and intensive care COVID-19 bed occupancy rates are recorded in Germany. These data are statistically analyzed on a daily basis using epidemic SEIR (susceptible, exposed, infection, recovered) models using ordinary differential equations and multiple regression models. RESULTS:Forecast results of the immediate trend (20-day forecast) of ICU occupancy by COVID-19 patients are made available to decision makers at various levels throughout the country. CONCLUSION:The forecasts are compared with the development of available ICU bed capacities in order to identify capacity limitations at an early stage and to enable short-term solutions to be made, such as supraregional transfers.

The COVID-19 pandemic has led to a high interest in mathematical models describing and predicting the diverse aspects and implications of the virus outbreak. Model results represent an important part of the information base for the decision process on different administrative levels. The Robert-Koch-Institute (RKI) initiated a project whose main goal is to predict COVID-19-specific occupation of beds in intensive care units: Steuerungs-Prognose von Intensivmedizinischen COVID-19 Kapazitäten (SPoCK). The incidence of COVID-19 cases is a crucial predictor for this occupation. We developed a model based on ordinary differential equations for the COVID-19 spread with a time-dependent infection rate described by a spline. Furthermore, the model explicitly accounts for weekday-specific reporting and adjusts for reporting delay. The model is calibrated in a purely data-driven manner by a maximum likelihood approach. Uncertainties are evaluated using the profile likelihood method. The uncertainty about the appropriate modeling assumptions can be accounted for by including and merging results of different modelling approaches. The analysis uses data from Germany describing the COVID-19 spread from early 2020 until March 31st, 2021. The model is calibrated based on incident cases on a daily basis and provides daily predictions of incident COVID-19 cases for the upcoming three weeks including uncertainty estimates for Germany and its subregions. Derived quantities such as cumulative counts and 7-day incidences with corresponding uncertainties can be computed. The estimation of the time-dependent infection rate leads to an estimated reproduction factor that is oscillating around one. Data-driven estimation of the dark figure purely from incident cases is not feasible. We successfully implemented a procedure to forecast near future COVID-19 incidences for diverse subregions in Germany which are made available to various decision makers via an interactive web application. Results of the incidence modeling are also used as a predictor for forecasting the need of intensive care units.

Zusammenfassung Hintergrund Zeitdynamische Prognosemodelle spielen eine zentrale Rolle zur Steuerung von intensivmedizinischen COVID-19-Kapazitäten im Pandemiegeschehen. Ein wichtiger Vorhersagewert (Prädiktor) für die zukünftige intensivmedizinische (ITS-)COVID-19-Bettenbelegungen ist die Anzahl der SARS-CoV-2-Neuinfektionen in der Bevölkerung, die wiederum stark von Schwankungen im Wochenverlauf, Meldeverzug, regionalen Unterschieden, Dunkelziffer, zeitabhängiger Ansteckungsrate, Impfungen, SARS-CoV-2-Virusvarianten sowie von nichtpharmazeutischen Eindämmungsmaßnahmen abhängt. Darüber hinaus wird die aktuelle und auch zukünftige COVID-ITS-Belegung maßgeblich von den intensivmedizinischen Entlassungs- und Sterberaten beeinflusst. Methode Sowohl die Anzahl der SARS-CoV-2-Neuinfektionen in der Bevölkerung als auch die intensivmedizinischen COVID-19-Bettenbelegungen werden bundesweit flächendeckend erfasst. Diese Daten werden tagesaktuell mit epidemischen SEIR-Modellen aus gewöhnlichen Differenzialgleichungen und multiplen Regressionsmodellen statistisch analysiert. Ergebnisse Die Prognoseergebnisse der unmittelbaren Entwicklung (20-Tage-Vorhersage) der ITS-Belegung durch COVID-19-Patienten*innen werden Entscheidungsträgern auf verschiedenen überregionalen Ebenen zur Verfügung gestellt. Schlussfolgerung Die Prognosen werden der Entwicklung von betreibbaren intensivmedizinischen Bettenkapazitäten gegenübergestellt, um frühzeitig Kapazitätsengpässe zu erkennen und kurzfristig reaktive Handlungssteuerungen, wie etwa überregionale Verlegungen, zu ermöglichen.

Reproducibility and reusability of the results of data-based modeling studies are essential. Yet, there has been—so far—no broadly supported format for the specification of parameter estimation problems in systems biology. Here, we introduce PEtab, a format which facilitates the specification of parameter estimation problems using Systems Biology Markup Language (SBML) models and a set of tab-separated value files describing the observation model and experimental data as well as parameters to be estimated. We already implemented PEtab support into eight well-established model simulation and parameter estimation toolboxes with hundreds of users in total. We provide a Python library for validation and modification of a PEtab problem and currently 20 example parameter estimation problems based on recent studies.