Medvedev, R;Turner, D;Gorelik, J;Alvarado, F;Bondarenko, V;Glukhov, A;
| DOI: 10.1016/j.bpj.2022.11.1392
Atrial fibrillation (AF) is commonly observed in patients with hypertension and is associated with pathologically elevated cardiomyocyte stretch. AF triggers have been linked to subcellular Ca2+ abnormalities, while their association with stretch remains elusive. Caveolae are mechanosensitive membrane structures, that play a role in both Ca2+ and cyclic adenosine monophosphate (cAMP) signaling. Therefore, caveolae could provide a mechanistic connection between cardiomyocyte stretch, Ca2+ mishandling, and AF. In isolated mouse atrial myocytes, cell stretch was mimicked by hypotonic swelling, which increased cell width (by ∼30%, p
Bogdanov, V;Soltisz, A;Beard, C;Hernandez Orengo, B;Sakuta, G;Veeraraghavan, R;Davis, J;Gyorke, S;
| DOI: 10.1016/j.bpj.2022.11.1389
Aberrant Ca-CaM signaling has been implicated in various congenital and acquired cardiac pathologies, including arrhythmia, hypertrophy, and HF. We examined the impact of HF induced by trans-aortic constriction (TAC) on the distribution of the three CaM mRNAs (Calm 1,2 and 3) and their key protein target mRNAs (Ryr2, Scn5a, Camk2d, NOS1 and Cacna1c) in cardiomyocytes, using fluorescence in situ hybridization (RNAScope™). HF resulted in specific changes in the pattern of localization of Calms, manifested in redistribution of Calm3 from the cell periphery towards the perinuclear area and enhanced Calm2 attraction to the perinuclear area compared to sham myocytes. Additionally, HF resulted in redistribution of mRNAs for certain CaM target mRNAs. Particularly, NOS1 localization shifted from the cell periphery towards the perinuclear area, Cacna1c, Camk2d and Scn5a abundance increased at the perinuclear area, and Ryr2 attracted even closer to the cell periphery in HF myocytes compared to sham myocytes. The strength of non-random attraction/repulsion was measured as the maximal deviation between the observed distribution of nearest neighbor distances from the distribution predicted under complete spatial randomness. Consistent with the observed alterations in abundance and distribution of CaM and CaM target mRNAs, HF resulted in increased attraction between Calm1 and Scn5a, Ryr2 and Camk2d, between Calm2 and Ryr2 and Camk2d; and between Calm3 and NOS1 and Scn5a. In contrast, the attraction between Calm3 and Ryr2 decreased in HF myocytes compared to sham. Collectively, these results suggest distribution of Calms and their association with key target protein mRNAs undergo substantial alterations in heart failure. These results have new important implications for organization of Ca signaling in normal and diseased heart.
Rodriguez, M;Tsai, C;Tsai, M;
| DOI: 10.1016/j.bpj.2022.11.1391
The mitochondrial calcium uniporter is a multi-subunit calcium channel that imports Ca2+ into mitochondria. Its MICU subunits (MICU1, MICU2, and the neuron-specific MICU3) gate the channel by blocking the pore in low Ca2+. Upon local Ca2+ elevation, Ca2+ binds to MICUs to cause MICU unblock, thus opening the pore so Ca2+ can permeate. Previous work using cell lines suggests that the uniporter in mammalian cells is exclusively regulated by a MICU1-MICU2 heterodimer. However, we show here that multiple types of electrically excitable cells, including skeletal muscle and cardiac tissues, can also possess a MICU1-MICU1 homodimer or virtually no MICUs. Kinetic analyses demonstrate that MICU1 has a higher Ca2+ affinity than MICU2, and that without MICUs the uniporter is constitutively open. As a result, uniporters with the MICU1-1 homodimer or no MICUs exhibit higher transport activities, leading to mitochondria accumulating much higher levels of matrix Ca2+. Using a Seahorse assay, we show that cells with MICU1-1 or no MICUs have impaired basal oxidative phosphorylation, likely due to increased ROS and damaged respiratory-complex proteins, including NDUFS3 and COX2. These cells, moreover, are highly susceptible to apoptosis. The disadvantage of employing MICU1-1 or omitting MICUs, however, accompanies strong physiological benefits. We show that in response to intracellular Ca2+ signals, these mitochondria import more Ca2+ and consequently produce more ATP, thus better supplying the energy required for the cellular processes initiated by the Ca2+ signals. In conclusion, this work reveals that tissues can manipulate their mitochondrial calcium uptake properties to suit their unique physiological needs by customizing their MICU regulation of the mitochondrial calcium uniporter.
Ramlow, L;Falcke, M;Lindner, B;
| DOI: 10.1016/j.bpj.2022.11.1390
Stochastic spiking is a prominent feature of Ca2+ signaling. The main noise source at the cellular level are puffs from inositol-trisphosphate receptor (IP3R) channel clusters in the membrane of the endoplasmic reticulum (ER). While the random cluster activity has been known for decades, a stringent method to derive the puff noise term acting on the cytosolic Ca2+ concentration is still lacking. We adopt a popular description of neural spike generation from neuroscience, the stochastic integrate-and-fire (IF) model, to describe Ca2+ spiking. Our model consists of two components describing i) activity of IP3R clusters and ii) dynamics of the global Ca2+ concentrations in the cytosol and in the ER. Cluster activity is modeled by a Markov chain, capturing the puff. The global Ca2+ concentrations are described by a two-variable IF model driven by the puff current. For the Markov chain we derive expressions for the statistics of interpuff interval, single-puff strength, and puff current assuming constant cytosolic Ca2+, an assumption often well met because the Ca2+ concentrations vary much slower than the cluster activity does. The latter assumption also allows to approximate the driving Ca2+ dependent puff current by a white Gaussian noise. This approximation results in an IF model with nonlinear drift and multiplicative noise. We consider this reduced model in a renewal version and in a version with cumulative refractoriness. Neglecting ER depletion, the stochastic IF model has only one variable and generates a renewal spike train, a point process with statistically independent interspike intervals (ISI). We derive analytical expressions for the mean and coefficient of variation of the ISI and suggest approximations for the ISI density and spike-train power spectrum. Taking into account ER depletion, the two-variable IF model displays cumulative refractoriness as seen in experimental data.