Comparison of different ChR2 variants for Ca2+ conductance in oocytes
To choose an ideal tool for Ca2+ modulation in primary cells, we first compared several high Ca2+-permeable ChR variants, especially the recently published CapChR212,13 and also ChR2 H134R4, since this ChR variant is widely used and a transgenic mouse line is available18. To compare their Ca2+ conductances, we measured the photocurrents in the bath solution containing 80 mM CaCl2. Before that, a final concentration of 10 mM Ca2+ chelator BAPTA was injected into the Xenopus oocyte to avoid the activation of Ca2+-activated Cl− channels (CaCCs). The ChR2 H134R4 showed minor Ca2+ conductance in comparison to ChR2 XXM10, CapChR212,13 and ChR2 XXM2.0 (Fig. 1a, b). Notably, ChR2 XXM2.0 had the most significant Ca2+ current compared with ChR2 H134R4, XXM10 and the recently published CapChR212,13 (Fig. 1a, b). Instead of Ca2+, further recordings were also performed with Ba2+, which is the most similar cation to Ca2+ but cannot activate the endogenous CaCCs. Similarly, ChR2 XXM2.0 showed the highest current with Ba2+ (Fig. 1c). In comparison to ChR2 H134R4, all three high Ca2+-permeable ChR variants mutants showed prolonged off kinetics (Fig. 1d), indicating potential higher light-sensitivities10.
Functional expression of ChR2 XXM2.0 in mouse MKs
In blood platelets, the elevation of intracellular Ca2+ contributes to various steps of cellular activation, but its role in MK differentiation and platelet production is less well defined. In order to enable spatially and temporally controlled manipulation of Ca2+ signaling in MKs, we first characterized the expression and localization of ChR2 XXM2.0 in bone marrow-derived MKs after viral transduction. ChR2 XXM2.0 with an EYFP tag co-localized with glycoprotein (GP) IX, demonstrating that ChR2 XXM2.0 is expressed in the plasma membrane and internal demarcation membrane system (DMS) of MKs (Fig. 1e). To assess whether ChR2 XXM2.0 is functional in MKs, we applied whole-cell patch-clamp and measured the light-induced current in MKs. Upon illumination, a strong inward current indicating cation influx was detectable in ChR2 XXM2.0 MKs, whereas no photocurrent was detected in non-transduced MKs (Fig. 1f). We also analyzed MKs of the widely used optogenetic knock-in mouse line ChR2-EYFP with the mutation H134R18 expressing Cre recombinase under the control of the platelet factor 4 (Pf4) promoter20. ChR2 H134R showed a similar expression pattern in MK membranes as compared to ChR2 XXM2.0 MKs (Supplementary Fig. 1a). However, the photocurrent in ChR2 H134R MKs was much weaker and was only ~ 8% of the ChR2 XXM2.0 MKs (Fig. 1f, Supplementary Fig. 1b). To confirm the Ca2+ influx, we used Cal 590™ to detect intracellular Ca2+ dynamics in MKs during 3 min global illumination. Increased intracellular Ca2+ levels were observable in ChR2 XXM2.0-positive MKs after illumination (Fig. 1g, h; Supplementary Video 1). Furthermore, spread ChR2 XXM2.0 MKs exhibited stress fiber formation and a further increase in spreading after illumination (Supplementary Fig. 2a-c). These data suggest that ChR2 XXM2.0 with enhanced Ca2+ permeability is functional in bone marrow-derived MKs to trigger Ca2+ influx.
Local activation of ChR2 XXM2.0 induces MK polarization towards the illumination side
Mature bone marrow MKs are located next to sinusoidal blood vessels and extend long cytoplasmic protrusions, designated proplatelets, into the vessel lumen, from which platelets are released. MKs need to polarize as a prerequisite for directional proplatelet release into sinusoidal vessels. Ca2+ has been reported to regulate cell polarity in plant cells and mammalian cells21–23. Therefore, we performed local illumination of ChR2 XXM2.0-positive MKs spread on fibrinogen (Fig. 2a). After 3 min local illumination in a peripheral region of MKs with blue light, ChR2 XXM2.0 MKs showed the capability of directional movement comparing to control MKs as determined by the distance between the center of mass of MKs before local illumination and at the end of the observation period (Fig. 2b-d; Supplementary Video 2). The majority of ChR2 XXM2.0 MKs showed a polarization behavior towards the illumination side as evidenced by the polarization trajectories and the rose diagram (Fig. 2e, f). In contrast, ChR2 H134R MKs did not show polarized movement in response to local blue light (Supplementary Fig. 1c), suggesting a potential role of Ca2+ in this process. To assess the role of Ca2+ in light-induced polarization of ChR2 XXM2.0 MKs, spread MKs were preincubated with the cell-permeant Ca2+ chelator BAPTA-AM or DMSO as control before local illumination. In the presence of 100 µM BAPTA-AM, the ChR2 XXM2.0 MKs showed impaired polarization compared with the DMSO-treated group (Fig. 2g, Supplementary Video 2). Taken together, these data demonstrated that local Ca2+-influx is involved in polarized MK movement.
The contractile protein non-muscle myosin IIA is involved in MK polarization
Non-muscle myosin IIA is an important regulator of adhesion and polarity in cell migration. Elevation of intracellular Ca2+ has been shown to control cell migration through activation of myosin IIa by generating contractile forces in fish epithelial keratocytes24. Furthermore, platelet migration was reported to be controlled by myosin IIA, which was activated by increased intracellular Ca2 + 25. Thus, we hypothesized that light-induced MK polarization might also depend on non-muscle myosin IIA activity, which was supported by the increased myosin light chain 2 (MLC2) phosphorylation in ChR2 XXM2.0 MKs after illumination (Fig. 3a). To further confirm this, spread MKs were preincubated with 100 µM blebbistatin to inhibit myosin IIA function. ChR2 XXM2.0 MKs showed impaired polarization in the presence of blebbistatin compared to DMSO-treated MKs (Fig. 3b, c), suggesting that myosin IIA-dependent force generation triggered by increased intracellular Ca2+ is involved in light-induced ChR2 XXM2.0 MK polarization.
Local activation of ChR2 XXM2.0 triggers localized binding of integrin αIIbβ3 to fibrinogen
As a major component of platelet signaling, intracellular Ca2+ rise leads to platelet responses, among others to activation of the integrin αIIbβ3, which is the dominant integrin on the platelet surface26. Thus, we sought to determine if the illumination of ChR2 XXM2.0 MKs results in activation of the fibrinogen receptor αIIbβ3. In order to observe integrin αIIbβ3 activation, fibrinogen labelled with Alexa Fluor 488 was added to the cell culture suspension, and a red fluorescence tag mKate2 was fused to ChR2 XXM2.0 instead of the EYFP. Subsequently, ChR2 XXM2.0-mKate2 MKs spread on fibrinogen were first globally illuminated (Fig. 4a). Illuminated MKs showed increased peripheral fibrinogen binding during the 10 min observation time, indicating activation of integrin αIIbβ3 (Fig. 4b, d). To assess the effect of polarized Ca2+ influx on integrin αIIbβ3 activation, ChR2 XXM2.0 expressing MKs were locally illuminated at its peripheral region (Fig. 4a). Fibrinogen binding on the plasma membrane of MKs was detectable only in the illuminated area (Fig. 4c, e), indicating polarized integrin αIIbβ3 activation. Next, we addressed whether the polarized fibrinogen binding was due to changes in the distribution of the integrin αIIbβ3, and incubated the ChR2 XXM2.0-mKate2 MKs with the anti-integrin αIIbβ3 JON6-Fab Alexa Fluor 488 antibody. However, local illumination did not change the distribution of integrin αIIbβ3 (Supplementary Fig. 3a, b), suggesting that polarized Ca2+ influx triggers local integrin activation rather than redistribution. To assess whether the binding between activated integrin αIIbβ3 and fibrinogen was necessary for light-induced MK polarization, spread MKs on fibrinogen were preincubated with 50 µg/ml JON/A-Fab antibody to block the activated integrin αIIbβ3 and prevent its binding to immobilized fibrinogen after local illumination. We found that blockade of the binding between integrin αIIbβ3 and fibrinogen inhibited light-induced polarized MK movement (Fig. 4f). These data demonstrated that binding of integrin αIIbβ3 to fibrinogen is important for light-induced MK polarization.
Generation of mice specifically expressing ChR2 XXM2.0 in MKs and platelets
So far, the viral approach has only allowed us to study light-induced effects on MK, not platelet function. Despite progress over the years, producing platelets under in vitro conditions remains very challenging. To establish optogenetics in platelets and manipulate platelet functions by light, we generated the optogenetic ChR2 XXM2.0 transgenic mouse line by CRISPR/Cas-mediated genome engineering. The CAG promoter-loxP-PGK-Neo-6*SV40 pA-loxP-Kozak-ChR2 XXM2.0-EYFP -rBG pA cassette was cloned into the ROSA26 locus to obtain conditional (loxP) ChR2 XXM2.0 mice, which were finally crossed with Pf4-Cre mice20 to remove the stop codon and express ChR2 XXM2.0-EYFP specifically in MKs and platelets (Fig. 5a). Similar to the viral approach, ChR2 XXM2.0 was expressed in MKs of the transgenic mice (Fig. 5b). It showed high photocurrent upon global illumination (Fig. 5c) and polarized movement upon focal illumination (Fig. 5d-f, Supplementary Video 3). Successful expression of ChR2 XXM2.0 was also confirmed in platelets of transgenic mice by immunoblotting and fluorescence microscopy (Supplementary Fig. 4a, b). Phosphatidylserine (PS) exposure of platelets is induced by high sustained cytosolic Ca2+ flux. This prompted us to test whether platelets from homozygous ChR2 XXM2.0-EYFPtg/tg, Pf4−cre mice express PS after blue light exposure. Illumination of half of the visual field resulted in a strong PS-positive signal followed by P-selectin exposure on the platelet membranes, while neighboring unilluminated platelets were unaffected (Fig. 6a-d, Supplementary Video 4). These results indicated that the ChR2 XXM2.0-EYFP platelets transition into a procoagulant state and highlighted the potential of optogenetics for manipulating the function of specific platelet subgroups within a population. Taken together, the newly generated conditional ChR2 XXM2.0 mouse line (loxP/Pf4-cre) enabled precise manipulation of cellular Ca2+ influx, thereby regulating MK and platelet function.