Local membrane deformation and micro-injury lead to qualitatively different responses in osteoblasts

Micro-damage of bone tissue is known to regulate bone turnover. However, it is unknown if individual bone cells can differentiate between membrane deformation and micro-injury. We generated osteoblasts from mouse bone marrow or bone morphogenetic protein 2-transfected C2C12 cells. Single cells were mechanically stimulated by indentation with the atomic force microscopy probe with variable force load either resulting in membrane deformation only, or leading to membrane penetration and micro-injury. Changes in the cytosolic free calcium concentration ([Ca 2+] i) in fluo4-AM loaded cells were analyzed. When deformation only was induced, it resulted in an immediate elevation of [Ca 2+] i which was localized to the probe periphery. Multiple consecutive local Ca 2+ responses were induced by sequential application of low level forces, with characteristic recovery time of ~2 s. The duration of [Ca 2+] i elevations was directly proportional to the tip-cell contact time. In contrast, cell micro-injury resulted in transient global elevations of [Ca 2+] i, the magnitude of which was independent of the tip-cell contact time. Sequential micro-injury of the same cell did not induce Ca 2+ response within 30 s of the first stimulation. Both local and global Ca 2+elevations were blocked in Ca 2+-free media or in the presence of stretch-activated channel blocker Gd 3+. In addition, amount of Ca 2+ released during global responses was significantly reduced in the presence of PLC inhibitor Et-18-OCH 3. Thus, we found qualitative differences in calcium responses to mechanical forces inducing only membrane deformation or deformation leading to micro-injury.


Introduction
Mechanical stimulation of bone is well-known to regulate bone volume, structure and composition 1,2 . It was recently suggested that, in addition to deformation forces, microdamage plays an important role in regulating bone turnover and strength 3 . Bone is restructured through the coordinated action of bone cells, osteoblasts responsible for bone formation and osteoclasts responsible for bone resorption. Cells of osteoblastic origin, including osteoblasts, osteocytes and bone lining cells are believed to act as mechanosensors in bone tissue 1,4 . Understanding how bone cells sense and react to mechanical forces is important for gaining insight into the mechanisms of bone adaptation to its mechanical environment.
Because of the complexity of the bone environment in vivo, several models have been developed to understand the effects of mechanical stimulation on bone cells in vitro 5 . These models include application of hydrostatic pressure, longitudinal substrate stretch and fluid shear. These studies have established that osteoblastic cells can sense mechanical stimulation through plasma membrane receptors, stretch activated ion channels, as well as focal adhesion sites 6 . Ca 2+ signaling was shown to be the prominent first response of osteoblastic cells to any type of mechanical stimulation [7][8][9] . Ca 2+ signaling induced by mechanical stimulation in turn influences numerous bone cell functions such as cytoskeletal reorganization 7 , gene expression 10 , proliferation and differentiation 6 . However, these studies also identified significant complex signaling interactions between multiple cells 6 , making it difficult to de-convolute the responses of single cells to mechanical stimulation. Moreover, these techniques do not allow exact control of forces applied to individual cells, nor do they report single cell micro-injury.
Local indentation techniques allow characterization of the response to mechanical stimulation at the single-cell level. Pipette microinjection 11 , pipette suction [11][12][13] , and atomic force microscopy (AFM) 14,15 have been used to study responses of individual osteoblasts to mechanical stimulation. From these techniques, only AFM allows application of precisely controlled forces in the nano-Newton scale, as well as provides readout of the extent of membrane deformation 16 . Moreover, AFM can be used with cantilever tips of different geometries, which allow additional control of the distribution of the force. Spherical tip allows creating high range of membrane strains 17 , while pyramidal tip allows reversible membrane penetration, which does not result in long-term cell damage 18,19 .
The goal of this study was to examine how a single osteoblastic cell responds to forces inducing cell membrane deformation only, or membrane deformation resulting in micro-injury. We used either primary bone marrow cells cultured with ascorbic acid, or C2C12 cells stably transfected with bone morphogenic protein (BMP) 2. C2C12 cells have been shown to undergo osteoblastic differentiation when treated with BMP-2 20 . Mechanical forces of different magnitude were applied using AFM. To monitor cell responsiveness to mechanical forces, changes in cytosolic free Ca 2+ concentration ([Ca 2+ ] i ) were assessed.  22 . The fluorescence signal in cells not exposed to mechanical stimulation did not demonstrate associated changes and was used as a reference for signal correction for bleaching and cantilever reflection artifacts using the protocol for data processing for systematic signal recovery that we established and described previously 22 .

Atomic force microscopy
The experiments were conducted using an MFP-3D-BIO AFM (Asylum Research, Santa Barbara CA) mounted on an Olympus IX-71 inverted optical microscope. The sample placed in the closed fluid cell was left undisturbed for 15 min to achieve thermal equilibrium at 37°C. A 60X oil immersion objective with 1.45 NA (Olympus) was put into contact with the coverslip allowing optical access from the bottom and AFM access on top of the sample. The region of interest was located and aligned with the cantilever tip using the bright field and fluorescence images.

AFM probe preparation
The cantilever tip (NCLAuD, Nanosensors), was etched down to 1 μm 2 contact area with a focused ion beam microscope (FEI DB235). When sharp tips were used, the membrane rupture force event was not distinguishable in the force curve. Using Hooke's law (F=k . d), the force exerted by the probe (F) was determined from the cantilever spring constant (k, 39 ± 7.8 N/m) and deflection (d). The net extension of the piezo element, the speed, the time of indentation, the time between indentations and the number of indentations were controlled using the Asylum MFP3D software in IgorPro 6.22 platform.

Statistical analysis
Data are presented as representative images, representative traces, means ± SE with n being the number of experiments analyzed. The normalized amplitude as a function of time, the amplitude decay rate as a function of recovery time and the response duration as a function of the contact time were analyzed using least squares regression and fitted a line or an exponential as appropriate. Categorical data were analyzed as described previously 23

Experimental setup
We employed two cell models to study osteoblast mechanosensitivity: i) mouse bone marrow cells cultured in the presence of ascorbic acid (50 mg/ml) for 4-6 days ( Figure 1A) or ii) C2C12 cells stably transfected with BMP-2 and cultured for 2-6 days ( Figure 1B). Both models represent osteoblastic cells at the early differentiation stage due to the limitation of AFM use in confluent and multilayered cultures. Osteoblastic phenotype was confirmed in fixed cultured by alkaline phosphatase staining. In parallel live cultures, the cells exhibiting osteoblastic morphology (strongly adhered cells with relatively large body and several filopodia) were chosen for mechanical testing. Mechanical stimulation was performed with AFM force spectroscopy indentation ( Figure 1C) using a 1 μm 2 area tip ( Figure 1D). Using AFM allows strict control of the amount of force applied to the cell ( Figure 1E, F). The membrane penetration event is easily identifiable on the force-distance curve as a decrease in the force required to continue to move the probe. The membrane penetration force was found to be similar for osteoblasts obtained in different cultures: 516 ± 200 nN for primary osteoblasts and 672 ± 100 nN for C2C12 osteoblasts. We analyzed two distinct modes of mechanical stimulation: i) the maximum force was set below the membrane penetration force (at ∼400 nN), leading only to membrane deformation (low-load, Figure 1E, F, blue); and ii) the maximum force was set above the membrane penetration force (∼2800 nN), inducing membrane rupture and penetration (high-load, Figure 1E, F, red).

Changes in [Ca 2+ ] i induced by different indentation regimes
Since [Ca 2+ ] i elevations are known to be the most common first responses to mechanical stimulation 4,9,11,24 , we analyzed the changes in [Ca 2+ ] i in response to different loading regimes in the osteoblasts loaded with [Ca 2+ ] i fluorescent indicator fluo4-AM. In the low load indentation regime, where the force exerted on the cell induced membrane deformation (Figure 2A), only local [Ca 2+ ] i transients were observed ( Figure 2B). To analyze the transients, the spatiallyaveraged intensity over a circle of 5 μm in diameter centered on the point of indentation (red circle in Figure 2B) was normalized to the initial baseline signal and plotted as a function of time ( Figure 2C). When larger circles were taken, the change in averaged intensity was smaller; if the whole cell was selected, the local response was not apparent. When consecutive low load indentations were performed at the same location on the same cell multiple local responses were induced (Figure 2A, C).
In the high load regime, which resulted in local membrane rupture and penetration of the cell ( Figure 2D), whole cell [Ca 2+ ] i elevations were observed ( Figure 2E). When the normalized fluorescence intensity was averaged over an ellipse surrounding the whole cell and plotted as a function of time ( Figure 2F), it was apparent that global [Ca 2+ ] i elevations were transient. When we focused on the membrane level to directly visualize the closure of the membrane micro-injury, we observed that it seals within 40-60 s after the stimulation ( Figure 2G). It is conceivable that high load indentation may affect the cell viability. However, in all our experiments, the cell previously exposed to high load indentation maintained basal calcium levels within normal range for the whole period of observation (5-30 min), and when directly tested using Trypan Blue exclusion test, six out of six cells previously indented were viable ∼10 minutes after high load indentation. Taken together, these data strongly suggest that the micro-injury induced by AFM probe is reversible and does not result in cell death.
correct the fluorescence traces for reflection artifacts, as well as photobleaching 22 . The traces were processed as follows: 1) the region of interest was selected in the digital image; 2) the average fluorescence intensity data were extracted and normalized to the initial basal reading to correct for differences in dye loading; 3) the signal was corrected for bleaching and reflection artifacts as described Bone marrow cells were cultured in the presence of ascorbic acid (50 μg/ml) for 4-6 days to induce osteoblast differentiation or C2C12 cells stably transfected with BMP-2 were cultured for 2-6 days to obtain osteoblastic phenotype. A, B) Representative images of bone marrow culture on day 5 (A) and C2C12 culture on day 3 (B) stained for osteoblastic marker, alkaline phosphatase (red). C) Schematic representation of the AFM operation under liquid: the sample and the tip are submerged and changes in the laser reflection signal due to deflection of the cantilever are monitored. D) Experiments were performed using a probe with 1 μm 2 tip area; the tip is etched using focused ion beam. E) Schematic representation of events observed using force-distance curve: a) as the AFM tip approaches the cell, the contact point is evident as increase in force required to move the probe; b) when the probe indents the cell membrane, the force increases until it reaches the membrane rupture force and the tip penetrates the cell (apparent by a decrease in the cell resistance); c) indentation is continued until a predetermined maximum force is reached, and the probe retraction is initiated. The retraction curve commonly deviates from the approach curve, reflecting that energy is required for deformation and penetration of the cell membrane. The maximum force was set to be either below (blue) or above (red) the membrane rupture force. F) The example force-distance curves from the experiments in which deformation only (blue) or deformation plus penetration (red) were induced.
Since application of AFM may potentially interfere with fluorescence measurements, we have previously performed in depth analysis of the system performance 22 . We have found that in experiments where local repetitive stimulations were performed, reflection artifacts correlated with cantilever motion represented a significant component of the fluorescent signal. We developed a protocol to qualitatively similar responses in the primary and C2C12-derived osteoblasts ( Figure 3A). However, when we compared the percentage of cells exhibiting local [Ca 2+ ] i transients in response to indentation performed at low and high speeds, we found that a higher percentage of C2C12-derived osteoblasts responded to the mechanical stimulation compared to primary osteoblasts ( Figure 3B). In addition, a higher percentage of C2C12-derived osteoblasts exhibited [Ca 2+ ] i transients when indented at lower speed compared to higher speed ( Figure 3B). In primary osteoblasts a similar trend was observed, however it did not reach statistical significance since the

Characterization of local [Ca 2+ ] i responses induced by membrane deformation
First, we compared the responses in C2C12-derived and primary osteoblasts. We have found that the low load indentation induced was plotted as a function of the corresponding contact time demonstrating that the average calcium response duration was directly proportional to the contact time, R 2 = 0.90 ( Figure 4D). We next assessed if calcium responses of different durations can be induced in a single cell by varying the contact time (as the piezo extension time) in consecutive indentations ( Figure 4E). We have found that a single cell exhibited [Ca 2+ ] i elevations of different durations ( Figure 4E, i-iv, black) when indentations with different contact times ( Figure 4E, i-iv, red) were performed.
Since we have found that repeated low level mechanical stimulations can elicit multiple consecutive responses in a single cell, we rate of response was low in these cells. We analyzed in depth the [Ca 2+ ] i responses induced in C2C12-derived osteoblasts by a single low-load indentation performed at different speed (2, 10 and 20 μm/s). The [Ca 2+ ] i elevations were further characterized by measuring i) the relative amplitude of the response (with respect to basal); ii) the duration of the response (the width of a transient at the half maximum amplitude); and iii) the amount of [Ca 2+ ] i (duration multiplied by amplitude). The speed of indentation did not significantly affect the amplitude of calcium responses ( Figure 3C). However, indentations performed at low speed induced calcium responses of longer duration ( Figure 3D) and thus resulted in release of greater amount of [Ca 2+ ] i ( Figure 3E) compared to indentations performed at high speed. To analyze if indenting at different speeds delivered different mechanical stimulation, the force curves were characterized by quantifying i) the extent of membrane deformation (the distance between the contact point and the maximum piezo extension) ( Figure 3F), and ii) the energy spent to deform the cell (the area between the approach and retraction curves) ( Figure 3F, right scale). No statistically significant difference was observed in these parameters when different indentation speeds were compared. We next assessed how the contact time (the net time the probe spent in contact with the cell during stimulation) affects calcium responses to low-load indentations performed at different speeds. We varied the contact time by increasing by 1 s the time that the tip spent in the maximally extended piezo position. The reflection of a cantilever on the calcium fluorescence recording allows achieving exact temporal overlay between the fluorescence intensity and the forcedistance curve (as described in 22 ). The [Ca 2+ ] i signal was initiated when the tip came into contact with the cell and was maximum at the maximum deflection ( Figure 4A, B). When the contact time was increased by 1 s, the comparable increase in calcium response duration was observed ( Figure 4C). The mean of the response duration

Characterization of global [Ca 2+ ] i responses induced by membrane deformation and micro-injury
When the maximum force was set at 2800 nN (three-fold above the membrane penetration force), the force-distance curves confirmed that membrane penetration occurred in all the cells tested. We have found that qualitatively similar responses were induced in primary and C2C12-derived osteoblasts ( Figure 6A, B), however the response rate was significantly lower for the primary osteoblasts ( Figure 6C). When the whole cell average fluorescence intensity and the forcedistance curves of individual cells were overlaid in time, we found that, in contrast to low-load indentation, micro-injury-induced global [Ca 2+ ] i elevations were delayed with respect to the maximum deflection of the probe (which indicates maximum cell penetration), and reached their amplitude after the tip was released from cell contact ( Figure 6A, B). We assessed [Ca 2+ ] i changes induced in C2C12-derived osteoblasts by a single high-load indentation performed at different speeds (2, 10 and 20 μm/s). There was no significant correlation between the indentation speed and the amplitude ( Figure 6D), duration ( Figure 6E) or amount of Ca 2+ released ( Figure 6F) during the responses. The force-distance curve analysis did not demonstrate statistically significant differences in the extent of membrane deformation prior to the membrane penetration ( Figure 6G), extent of membrane penetration ( Figure 6H), energy spent in deformation, micro-injury and penetration ( Figure 6I) or membrane rupture force ( Figure 6J). next examined how the exposure to mechanical stimulation affects subsequent responses of the same cell. Multiple low load indentations were performed on the same cell with different frequencies, and [Ca 2+ ] i changes in response were assessed ( Figure 5). We have found that when cells were stimulated with higher frequency, the amplitude of the response in consecutive stimulations decreased ( Figure 5A-C). Within each sequence of Ca 2+ responses to the consecutive indentations, we normalized the amplitudes with respect to the amplitude of Ca 2+ response to the first indentation and quantified the amplitude decay rate as a slope (α i ) of the best fit line for the responses. This slope is negative, representing a decrease in the amplitude, and when multiplied by 100, this slope is expressed as percentage decay. We have found that amplitude decay is significantly higher when cells are stimulated with higher frequencies (Figure 5D). At different frequencies two parameters are changing -the contact time, defined as the time the probe spends in contact with the cell, and recovery time, defined as the time off contact between stimulations. We next specifically varied the recovery time by introducing a delay of 1 s between consecutive stimulations, and plotted the average amplitude decay rates as a function of the recovery time ( Figure 5E). An exponential curve with a characteristic recovery time of 2.00 ± 0.08 s was found to be the best fit for the relationship between the amplitude decay rate and the recovery time. These data suggest that a refractory period of ~10 s is required for complete amplitude recovery after low load mechanical stimulation.
but not in vehicle-treated cells compared to control ( Figure 7I). The percentage of cells exhibiting global [Ca 2+ ] i transients in response to membrane penetration was not significantly affected by Et-18-OCH 3 or vehicle ( Figure 7J). However, the amount of Ca 2+ released during the response was significantly decreased in cells treated with Et-18-OCH 3 ( Figure 7K). These data indicate that mechanosensitive Ca 2+ channels mediate both local and global responses, while the intracellular Ca 2+ stores are important for global responses only.

Discussion
In this study, we have examined the ability of individual osteoblasts to respond to mechanical stimulation applied using AFM. Variable force load was applied resulting in either membrane deformation only or in membrane penetration and micro-injury, and changes in [Ca 2+ ] i in fluo-4 loaded cells were analyzed. Qualitatively different Ca 2+ responses were observed in different force loads. In response to membrane deformation only, immediate local [Ca 2+ ] i elevations limited to the indentation region were observed. Multiple stimulations of a single cell resulted in Ca 2+ responses of similar amplitude if a recovery time of more than 2 s between the stimulations was allowed. The duration of Ca 2+ responses to low-load indentation was proportional to the duration of deformation. In contrast, micro-injury induced global Ca 2+ elevation, which continued to develop after the removal of the probe and was independent of the duration of indentation.

Contribution of mechanosensitive channels and calcium stores to elevations in [Ca 2+ ] i induced by membrane deformation and micro-injury
Since the rise in [Ca 2+ ] i can be due to influx from the extracellular space as well as to release from intracellular stores, we next investigated the contribution of these pathways to membrane deformation and to micro-injury-induced [Ca 2+ ] i responses. We analyzed changes in [Ca 2+ ] i in response to low-load and high-load mechanical stimulations in C2C12 osteoblasts maintained in the control Ca 2+containing buffer or in Ca 2+ -free, EGTA (10 mM)-containing buffer ( Figure 7A-C). Both local [Ca 2+ ] i elevations ( Figure 7A, B) and global [Ca 2+ ] i responses ( Figure 7C) were prevented by the lack of calcium in the extracellular space. We next inhibited mechanosensitive membrane channels using Gd 3+ (50 μM) 17 . In the presence of Gd 3+ both local [Ca 2+ ] i transients in response to membrane deformation ( Figure 7D) and micro-injury-induced global [Ca 2+ ] i responses ( Figure 7E) were prevented. To inhibit the phospholipase C (PLC)inositol triphosphate (IP 3 ) pathway leading to calcium release from intracellular stores, we used PLC inhibitor Et-18-OCH 3 (5 μM) 17 .
In the low-load regime, local [Ca 2+ ] i transients ( Figure 7F) occurred at the same frequency as in control ( Figure 7G), and the amount of Ca 2+ released during the responses was not significantly affected ( Figure 7H). However, global [Ca 2+ ] i transients induced by the highload stimulation were noticeably smaller in Et-18-OCH 3 -treated, used AFM with spherical tip of 10-30 μm diameter, which would deform a 100-fold larger area compared to the pyramidal tip with 1 μm 2 contact area used in our study. Thus, the extent of horizontal membrane involvement in the deformation may be important for cell mechanosensitivity. While Xia and Ferrier 11 used patchclamp micropipette with similar dimensions to the pyramidal tip used in our study, it has been shown that pipette suction creates substantially larger strains compared to AFM micro-indentation 14 . Therefore, higher calcium responses observed in the Xia and Ferrier study compared to our study are in keeping with previously demonstrated critical role of vertical membrane deformation in cell mechanosensitivity 17 . We have now shown that the duration of Ca 2+ Data are means ± SE, n is the number of response from 10 trials: for 2 μm/s n = 9, for 10 μm/s n = 10, and for 20 μm/s n = 9, no significant difference. G-J) Force-distance curves were analyzed for the extent of membrane deformation (G), penetration depth (H), energy spent to deform and penetrate the membrane (I) and membrane rupture force (J). Data are means ± SE, n = 10, no significant difference.
These data demonstrate that Ca 2+ responses to local membrane deformation exhibit threshold properties when micro-injury is induced.
In our model, micro-injury was required in order to achieve global elevations of [Ca 2+ ] i in the majority of cells. Many techniques used to study cell mechanosensitivity, such as fluid flow, substrate strain and pipette indentation, do not allow sufficient resolution to detect membrane micro-injury. However, a number of studies, in which the absence of membrane penetration can be reliably confirmed, demonstrated that global Ca 2+ responses can be induced by membrane deformation only 11,17 . Several differences in the experimental setup could account for this discrepancy. Charras and Horton 17 responses is directly proportional to time the cell membrane spends in deformation. Therefore, it is likely that cell sensitivity to membrane deformation is related to a combined effect of: i) the extent of the horizontal involvement of the membrane, ii) the vertical deformation of the membrane and iii) the duration of the deformation. Mechanosensitive calcium channels were strongly implicated in generating Ca 2+ responses to mechanical stimulations in previous studies 11,17,25 and were confirmed to provide a critical contribution to both local and global Ca 2+ responses in our study. Therefore, it is possible that local intracellular Ca 2+ acts as an integrating signal that increases when open channels are more numerous or are activated longer, until it reaches the threshold Ca 2+ concentration necessary to induce global Ca 2+ response. Micro-injury then contributes to increasing local Ca 2+ levels to threshold concentrations, inducing global response. Once global calcium elevation is triggered, it proceeds independently of the level of mechanical stimulation, as suggested by the lack of correlation between the magnitude of global Ca 2+ response and mechanical stimulation in our study and that of Charras and Horton 17 . Our data demonstrate that calcium release from intracellular stores plays an important role in this process.
We identified the refractory periods during which the responses to subsequent mechanical stimulations were either absent or diminished in amplitude. This refractory period following membrane deformation only was relatively short, and the calcium response was fully reestablished after the recovery time of 10 s. In contrast, in experiments where micro-injury and global elevation of calcium were observed, the recovery period was longer than 30 s. Previously, the refractory periods for global Ca 2+ responses to fluid flow were reported to be in the order of 600-900 s 26 . The presence of refractory periods is of potential importance, since it was shown previously, that in order to induce potent osteogenic response, mechanical loading of bone should be performed as a series of repeated loading periods separated by the periods of rest 27,28 . Of interest, the rest period of 10 s was found to be sufficient to induce potent bone formation in response to low magnitude mechanical loading 28,29 .
Taken together, our study provides new insights into the complex dynamics of cellular responses to mechanical stimulations. In contrast to many previous studies of cell responses to mechanical forces, atomic force microscopy allows very precise control and monitoring of the physical parameters of the experiment, such as forces and deformations applied at a single cell level. Using a wellestablished readout of cellular response, calcium signaling, allowed us to identify novel correlates between mechanical stimulation and cell responses. Such knowledge is important for better understanding 1.

2.
distance has been largely described and commented by the group of Ning Wang presently working at University of Illinois at Urbana Champaign (e.g., ). Clearly a tensed Wang 2009 et al., cytoskeleton plays a key role in such a distant phenomena and using, e.g., cytochalasin D, would have been of interest to prove that the distant local effects are related to the tensed cytoskeleton. Authors should at least comment on this interesting result revealed by the local membrane deformation test. Such an distant effect is not so evident in case of membrane rupture suggesting (since mechanical effect become global) important modifications for the cytoskeleton between the two types of stress studied. These distant effects could advantageously be analyzed and quantified.
It is not clear whether the cells are tested in isolated conditions (like in Fig 2B) or with some degree of confluence (like in Fig 2E)... In the latter case, mechano-chemical interactions between cells could play a role in calcium signaling. This point should be clarified all along the paper.

Detailed comments
Figures 1A, 1B seem to show the culture at higher confluence than 50-70% as used for experiments.
The linear fits in Fig. 5 A, B, C don't represent the decay of peaks.
We have read this submission. We believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Overall, I think the authors have done a good at el. job in characterizing qualitatively the differences in Ca responses induced by local membrane deformation and micro-injury. I particularly liked the measurements involving cyclic membrane perturbation and inhibition of Ca channels and internal store. I think the latter provide a good understanding on the different origin of the Ca released when cells are probed or micro-injured. Nevertheless, i have a technical concern and some questions that i feel the authors could have discussed more in-depth in the manuscript: Figure 4c and 4e(iv) show large deviations in the deflection of the cantilever when cells are indented for longer time periods. Which kind of feedback was used to control the AFM tip position? It appears as if no deflection feedback was really used, but rather the piezo was kept at constant z. It seems the system is either drifting, or the cell is changing its mechanical properties (softening) or its height (shortening) due to the Ca release. Could the authors comment on that?
The fit obtained from fig 4d could be considered to have a slope of 1 (0.98), indicating an on-off response. More interestingly, the offset of 0.6s indicates that no matter the duration of the stimulation (and thus the total Ca released), the time required to recover baseline levels is constant. Would that be an expected behaviour, or is this the case only because the released Ca levels are small and very localized compared to the whole volume cell? 3.
levels are small and very localized compared to the whole volume cell?
In the discussion, the authors suggest that 'intracellular Ca acts as an integrating signal that increases until it reaches the threshold Ca concentration necessary to induce global Ca response'. If that is the case, have they observed any global Ca responses when using multiple cyclic indentations with very short recovery times? Taking into account their results on Ca released for 1 single indentation, could they predict the number (or duration) of small indentations required to integrate enough Ca signal to start a global Ca response?
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Competing Interests: