Shear-Stress Initiates Signal Two of NLRP3 Inflammasome Activation in LPS-Primed Macrophages through Piezo1.

To elucidate the ability of shear stress to initiate NLRP3 inflammasome activation, LPS-primed iBMDMs were treated with three different shear stresses ranging from 1.0 to 50 dyn/cm 2 by manipulating the syringe pump flow rate. Several orders of magnitude of shear stresses are found throughout the body, yet we focused on shear stresses correlating with laminar flow since they are more widely used and appropriate to the focus of pancreatitis in this study. 19 , 24 , 33 This type of shear stress interaction with macrophages is especially relevant in pressure-induced pancreatitis where Piezo1 and TLR4 have been demonstrated to be involved in disease progression. 23 , 25 , 31 However, studies that demonstrated these phenomena lacked a physiologically relevant 3D in vitro method to determine how shear stress induces a calcium influx and subsequently activates the NLRP3 inflammasome. In this study, a microfluidic device was used to investigate tissue-resident macrophages and their ability to induce NLRP3 inflammasome activation through shear stress. This allowed us to gain a deeper understanding of how physiologically relevant external stimuli can influence inflammatory disease progression as they appear in the body. To recapitulate tissue-resident macrophages or those accumulated at an inflammatory site, iBMDMs were uniformly seeded in a 2.5 mg/mL type I collagen gel in the center channel of a microfluidic device as described by Fish and Kulkarni. 28 A syringe pump flowed media through channels on either side of the collagen gel to apply fluid shear stress to cells as described by Fish and Kulkarni 28 ( Figure S1 ). The microfluidic system used for this study was the idenTx 3 (AIM BIOTECH), and the manufacturer provided correlations to equate shear stresses applied to cells to flow rates set on the syringe pump. This microfluidic system was developed to mimic tissue-resident macrophages under various physiologically relevant flow conditions as indicated by the shear stresses selected, ranging from 1.0 to 50 dyn/cm 2 . We primed iBMDMs with 100 ng/mL LPS for 4 h, followed by shear stress treatments for 30 min. This process initiates the activation of the NLRP3 inflammasome and the ASC adaptor protein, which oligomerizes, forming an ASC speck ( Figure 1A ). After shear stress exposure, the cells were stained to be imaged with confocal microscopy and analyzed ( Figure 1B ). The iBMDMs utilized in this study contain a cyan fluorescent protein (CFP) tag on ASC adaptor protein so that ASC–CFP speck presence, a well-studied indicator of NLRP3 inflammasome activation, could be quantified under various shear stress conditions. 21 , 22 , 27 Additionally, static controls were performed for untreated and LPS-primed iBMDMs to assess the baseline amount of NLRP3 inflammasome activation in comparison to shear stress conditions. When LPS-primed iBMDMs were treated with shear stresses ranging from 1.0 to 50 dyn/cm 2 , the formation of ASC–CFP specks significantly increased in comparison to the LPS control with no significant differences between shear stress magnitudes ( Figure 1C ). This indicates that shear stresses in this range provide a significant signal to initiate NLRP3 inflammasome activation. Cell death, marked by propidium iodide (PI), was significantly increased in LPS-primed iBMDMs after 50 dyn/cm 2 shear stress exposure in comparison to 1.0, 10 dyn/cm 2 , and LPS controls ( Figure 1D ). This demonstrates that shear stress at 50 dyn/cm 2 while initiating ASC–CFP speck formation comparable to lower shear stresses, the cells have a higher percentage of cell death. This trend has also been observed in a study that investigated the priming capabilities of shear stress on iBMDMs for NLRP3 inflammasome activation, where higher shear stresses resulted in lower cell viability. 28 These experiments suggested that while 30 min shear stress exposure ranging from 1.0 to 50 dyn/cm 2 all induced significantly higher NLRP3 inflammasome activation, 50 dyn/cm 2 shear stress resulted in significantly higher cell death than the others. In LPS-primed iBMDMs, we observed activation of the NLRP3 inflammasome in response to shear stress, however, we wanted to confirm that additional inflammasomes, such as AIM2, NLRP1, NLRP6, and NLRC4, were not being activated and contributing to ASC–CFP speck formation. AIM2 inflammasome activation is controlled largely by cytoplasmic double-stranded DNA, so LPS priming and shear stress exposure should not induce activation and speck formation in this signaling pathway. 34 – 36 NLRP6 inflammasomes are also primed by double-stranded DNA, but they can also be primed by lipoteichoic acid, which is found on Gram-positive bacteria. 37 , 38 In this study we used LPS, which is found on Gram-negative bacteria, so we do not expect the NLRP6 inflammasome to contribute to shear-stress-induced inflammasome activation. While the NLRC4 inflammasome can become primed in response to LPS exposure, as it is a pattern seen on Gram-negative bacteria, activation of the protein requires type III secretion system rod components. 39 – 42 Lastly, NLRP1 inflammasome priming and activation is triggered through various viral proteins, such as Bacillus anthracis , to protect the body from anthrax lethal toxin, so NLRP1 inflammasome activation should not be occurring in this scenario. 43 – 45 Nonetheless, to confirm that shear-stress-induced ASC–CFP speck formation is solely NLRP3-inflammasome-mediated, we utilized NLRP3 and caspase-1 KO iBMDMs containing the same modification on the ASC adaptor protein to express a CFP tag. We primed KO iBMDMs with LPS for 4 h, followed by a 30 min shear stress exposure with three magnitudes before imaging with confocal microscopy ( Figure 2A , B ). Untreated and LPS-primed controls were performed to assess the portion of ASC–CFP speck formation and PI-positive cells under these conditions. Similar to WT iBMDMs, having an ASC adaptor protein tagged with CFP made quantification of activation with confocal microscopy much easier. These results indicated that regardless of the shear stress magnitude or knockout cell line used, there were no significant changes in the presence of ASC–CFP specks in comparison to untreated groups ( Figure 2C , D ). However, we observed that the KO iBMDMs had significantly less ASC–CFP specks compared to the WT iBMDMs in every group ( Figure 2E ). This indicated that both caspase-1 and NLRP3 are heavily involved in shear-stress-induced NLRP3 inflammasome activation. The cell death in all shear stress groups was significantly lower than the LPS control in both KO iBMDMs, indicating the absence of NLRP3 or caspase-1 causes cell death to be significantly lower than the LPS control ( Figure 2F , G ). In comparison to WT iBMDMs, the absence of caspase-1 induces significantly higher cell death in the control groups while demonstrating lower cell death in the group treated with shear stress. NLRP3 knocked-out cell death is significantly lower at shear stress 10 and 50 dyn/cm 2 and remains statistically similar in the control groups ( Figure 2H ). These microscopy experiments clearly demonstrated the importance of both NLRP3 and caspase-1 for shear-stress-induced inflammasome activation in LPS-primed iBMDMs since there are significantly more ASC–CFP specks in WT iBMDMs for all shear stresses in comparison to the KO iBMDMs under the same conditions. NLRP3 and caspase-1 are not the direct mechanisms through which this activation occurs, but utilizing these KO cell lines allowed us to determine the involvement of the NLRP3 inflammasome in this process in LPS-primed iBMDMs under flow conditions. 17 , 46 As previously reported in literature, NLRP3 and caspase-1 are not the direct mechanisms that shear-stress-induced NLRP3 inflammasome activation occurs. It has been widely reported that mechanosensitive ion channel Piezo1 is responsible for inducing a calcium influx under shear stress conditions, which is a well-known signal to initiate activation (Signal 2) of the NLRP3 inflammasome. 17 , 22 , 26 This type of NLRP3 inflammasome activation is especially important in pancreatitis, where Piezo1 and the subsequent calcium influx have been shown to play an important role in disease progression in pressure-induced pancreatitis. 25 , 33 To determine how LPS-primed iBMDMs generate calcium influx in response to varying magnitudes of shear stress, we seeded cells into a microfluidic device as previously described and treated with shear stress for 30 min, followed by staining with Fluo-4 AM and NucBlue before being imaged with confocal microscopy ( Figure 3A ). LPS and LPS + nigericin control groups established a baseline and provided a positive control, respectively. For the LPS + nigericin control, iBMDMs were treated with 100 ng/mL for 4 h, followed by a 1 h treatment of 10 μ M nigericin. Calcium influx was quantified by taking the sum intensity of Fluo-4 AM and normalizing by the number of total cells ( Figure 3B ). As the magnitude of shear stress increased, the calcium influx per cell also increased. We expected this result since an activator of the Piezo1 receptor, Yoda1, demonstrated this trend in a concentration-dependent manner. 20 We also performed ELISA on the effluent media to quantify the release of IL-1 β in LPS-primed iBMDMs after 30 min of 50 dyn/cm 2 shear stress exposure ( Figure 3C ). For the shear stress groups, the media effluent was concentrated using centrifugal filter units to ensure the volumes were of similar magnitude across all groups. We demonstrated that in LPS-primed macrophages, shear stress treatment initiated a significantly higher release of IL-1 β compared to the untreated and LPS controls. The changes in IL-1 β release were not significantly different than those seen in iBMDMs treated with LPS + nigericin. Next, we wanted to look at the presence of gasdermin-D protein and how shear stress treatments could affect its cleavage. Since gasdermin-D is cleaved by active caspase-1 and forms the pores for IL-1 β export and pyroptotic death, we expected that this protein would be present in its active form in LPS-primed iBMDMs treated with either shear stress or nigericin. 21 , 27 , 47 We quantified gasdermin-D activity by performing capillary western using the SimpleWestern WES machine ( Figure S2A ). Inactive gasdermin-D (53 kDa) and active gasdermin-D (31 kDa) were quantified relative to the housekeeping gene vinculin (110 kDa) and normalized to the untreated groups ( Figure S2B , C ). The ratio of active to inactive gasdermin-D was determined relative to the untreated group, where this ratio was significantly higher in LPS-primed iBMDMs treated with either shear stress or nigericin compared to both untreated and LPS-primed iBMDMs ( Figure 3D ). We also observed no significant differences between the untreated and LPS groups. These results indicated that shear-stress-mitigated NLRP3 inflammasome activation in LPS-primed iBMDMs acts through calcium influx and initiates cleavage of active gasdermin-D and subsequent IL-1 β release downstream, demonstrating a direct correlation between shear stress and NLRP3 inflammasome activation hallmarks. Next, we wanted to elucidate a more detailed mechanism through which the calcium influx generated by shear stress exposure affects NLRP3 inflammasome activation. Recent studies have shown that CCL2 has been shown to be upregulated in the presence of Piezo1 and affects the progression of pancreatitis. 48 – 50 Additionally, the CXCR2/CXCL2 signaling pathway has been shown to be associated with NLRP3 inflammasome activation and upregulated as pressure-induced pancreatitis progresses. 17 , 51 – 54 To determine the mechanism through which LPS-primed iBMDMs undergo shear-stress-induced inflammasome activation, we performed qPCR after 50 dyn/cm 2 shear stress exposure. Untreated and LPS-primed controls assessed the relative gene expression in groups primed with LPS and activated with shear stress exposure. We assessed the expression of NLRP3 ( Figure 4A ), IL-1 β ( Figure 4B ), caspase-1 ( Figure 4C ), gasdermin-D ( Figure 4D ), Piezo1 ( Figure 4E ), CCR2 ( Figure 4F ), and CXCR2 ( Figure 4G ) normalized to β -actin controls. Cells treated with LPS for 4 h and 50 dyn/cm 2 shear stress for 30 min (LPS +50 dyn/cm 2 ) exhibited significant upregulation of NLRP3 ( Figure 4A ), IL-1 β ( Figure 4B ), gasdermin-D ( Figure 4D ), and Piezo1 ( Figure 4E ) compared to the untreated control group. NLRP3 and Piezo1 expression also significantly increased for the LPS +50 dyn/cm 2 with respect to the LPS control group. We saw no statistically significant differences in the expression of caspase-1 ( Figure 4C ), CCR2 ( Figure 4F ), and CXCR2 ( Figure 4G ) in the LPS +50 dyn/cm 2 group compared to the untreated control. We also wanted to observe the correlation between macrophage shear-stress-activation and CCL2, CXCL1, and CXCL2 release into the extracellular space, as their upregulation has also been connected to inflammasome activation. 52 , 55 For this, we performed ELISA on the supernatants of cells exposed to LPS, LPS + nigericin, or LPS +50 dyn/cm 2 shear stress to assess the relative secretion of CCL2 ( Figure 4H ), CXCL2 ( Figure 4I ), and CXCL1 ( Figure 4J ). We determined that cells treated with shear stress displayed higher secretion of CXCL1, CXCL2, and CCL2 compared to the untreated controls, while only CCL2 levels were significantly higher than the LPS control group. In the LPS + nigericin group, CCL2, CXCL1, and CXCL2 secretion significantly increased compared to the untreated group, while only CCL2 and CXCL2 were significantly higher than the LPS groups. In LPS-primed iBMDMs treated with shear stress, we demonstrated the upregulation of NLRP3, IL-1 β , and Piezo1 relative to the untreated controls and an increased secretion of CCL2 compared to LPS-primed iBMDMs. To further determine the involvement of Piezo1 in shear-stress-mediated NLRP3 inflammasome activation, we treated LPS-primed iBMDMs with a well-studied and selective Piezo1 inhibitor, Dooku1. 28 Cells were treated with the inhibitor for 2 h at concentrations of 10 μ M or 50 μ M before 50 dyn/cm 2 shear stress exposure for 30 min. To assess the involvement of Piezo1 in this pathway we investigated how the inhibitor affected ASC–CFP speck formation by performing confocal microscopy experiments ( Figure 5A ). We demonstrated a dose-dependent response in the presence of ASC–CFP specks where 10 μ M Dooku1 reduced it 2-fold and 50 μ M Dooku1 reduced it 4-fold, statistically similar to levels observed in the untreated groups ( Figure 5B ). Both concentrations of Dooku1 also greatly reduced the cell death with 50 dyn/cm 2 shear stress treatments to levels that were statistically similar to the LPS control ( Figure 5C ). This can be observed in the further enhanced images for shear-stress-treated iBMDMs with and without Dooku1 treatment ( Figure 5D , E ). Additionally, we looked at how 50 μ M Dooku1 treatments affected the gene expression of NLRP3 and Piezo1, genes that were greatly upregulated in LPS-primed macrophages treated with shear stress. Dooku1 treatment reduced the expression of both NLRP3 and Piezo1 to levels statistically similar to their respective untreated controls, greatly supporting their involvement in this pathway ( Figure 5F , G ). Utilizing this inhibitor allowed us to mechanistically determine that Piezo1 mediates the ability of shear stress to induce NLRP3 inflammasome activation, and inhibiting this part of the pathway completely ameliorates the formation of proteins and the gene expression hallmarks associated with this type of inflammasome activation ( Figure 6 ). Shear stress is a ubiquitous signal throughout the body that has profound effects on intracellular signaling mechanisms, but there is a limited understanding of how this can play a role in chronic inflammation. The role of shear stress in chronic inflammation is context dependent where it has the ability to act as Signal 1 or Signal 2 depending on the presence of LPS; lipids with TLR4-agonist properties demonstrated similar trends in other studies. 28 , 56 We aimed to take a deeper look into how shear stress mediates NLRP3 inflammasome activation within the context of calcium-signaling-dependent chronic inflammatory diseases such as pancreatitis. We also provided a further in-depth mechanistic analysis using ELISA, qPCR, capillary western, and a Piezo1 inhibitor study to determine that Piezo1 plays a critical role in Signal 2 of NLRP3 inflammasome activation. To uncover the mechanism of how shear stress induces NLRP3 inflammasome activation, qPCR and WES were utilized to observe the mRNA and protein-level expression, respectively, of the key inflammasome components. In qPCR studies, normalized NLRP3 expression significantly increased 6-fold in the LPS +50 dyn/cm 2 group, while NLRP3 expression in the LPS group only elevated 4-fold. Shear stress treatment significantly enhanced NLRP3 expression, showing a significant increase in comparison to the LPS group. Expression of mechanosensitive ion channel Piezo1 was also elevated in the LPS +50 dyn/cm 2 group compared to both the untreated and LPS groups. LPS treatment did not result in an increase in Piezo1 expression since LPS is a TLR4 agonist and does not interact with Piezo1. 28 We also proved the involvement of Piezo1 in NLRP3 inflammasome activation through Dooku1 inhibitor studies where ASC–CFP speck formation, cell death, and expression of NLRP3 and Piezo1 were all reduced to basal levels when pretreated with the inhibitor. In qPCR, IL-1 β expression was also elevated in both the LPS and LPS +50 dyn/cm 2 group compared to the untreated control; however, there was no statistically significant difference between these two groups. We also conducted IL-1 β ELISA and revealed that shear stress or nigericin treatment in LPS-primed iBMDMs both resulted in a significant increase in secretion of this cytokine over the LPS group. While these changes were not detected in the mRNA expression, they were revealed in the cytokine release profile, which makes mechanistic sense, as IL-1 β is secreted as macrophages undergo NLRP3 inflammasome activation and pyroptotic cell death, which is further supported by the PI imaging data. 21 , 47 Caspase-1 and gasdermin-D are also proteins whose expression is well associated with NLRP3 inflammasome activation and the downstream pore formation for inflammatory cytokine secretion. In qPCR, the mRNA expression of caspase-1 in the LPS +50 dyn/cm 2 group did not show any significant changes compared to the untreated groups, while gasdermin-D showed similar levels to LPS in the LPS +50 dyn/cm 2 group, both of which were significantly higher than the untreated group. While at first this may seem confounding, since both of these proteins are associated with NLRP3 inflammasome activation, our WES study demonstrated that the ratio of active to inactive gasdermin-D expression was significantly elevated in the LPS +50 dyn/cm 2 and LPS + nigericin groups with statistically insignificant differences between the two. An increase in this ratio also determines that caspase-1 cleavage is also happening since the active caspase-1 enzyme cleaves gasdermin-D into its active form. In addition to IL-1 β , we investigated how the secretion of CCL2, CXCL1, and CXCL2 was affected by shear stress since these proteins have proved important in the progression of pancreatitis and NLRP3 inflammasome activation in vitro and in vivo. As previously reported, we observed significant increases in all of these cytokines in the LPS + nigericin group compared to the untreated controls. 53 , 57 We also observed a stepwise increase in CCL2 secretion as cells were treated with LPS, followed by nigericin treatment. However, when shear stress was treated instead of nigericin, a significant increase in CCL2 secretion was observed, while other cytokines remained at levels similar to LPS-primed iBMDMs. This is due to the CXCL1/2 axis being regulated by MyD88 signaling, and while CCL2 can also be regulated by MyD88, it is also regulated by Piezo1 signaling, a MyD88-independent mechanism. 57 – 59 This explains why the LPS +50 dyn/cm 2 levels of CXCL1 and CXCL2 were not significantly different from groups treated with LPS only, and CCL2 secretion was significantly elevated. The release of these cytokines in vivo will contribute to the accumulation of other cell types, such as monocytes, macrophages, dendritic cells, and neutrophils, at the site of inflammation. 60 – 63 CXCL1 and CXCL2 are both well-known neutrophil chemotaxis that macrophages secrete to recruit neutrophils to the site of inflammation. 62 , 64 – 66 CCL2 is a monocyte chemoattractant protein that also has the ability to affect the localization of neutrophils to inflammatory sites. 61 , 63 , 67 , 68 This means that depending on the physiological environment and disease where a pathological shear stress increase is observed, differing populations of immune cells will localize to the site of inflammation. Specifically, in pancreatitis, the release of IL-1 β and CCL2 causes the localization of inflammatory monocytes in the pancreas, which are the major players that regulate fibrosis, increasing the probability of the formation of pancreatic ductal adenocarcinoma. 25 , 33 , 69 An increase in fibrosis and release of these cytokines has been demonstrated in studies where a Piezo1 agonist or high pancreatic pressure was used to study fibrosis in pancreatitis. 25 , 33 The upregulation of Piezo1, NLRP3, calcium signaling, active gasdermin-D, IL-1 β , and CCL2 secretion proved their involvement in shear-stress-mediated activation of the NLRP3 inflammasome as well as the ability to potentially induce fibrosis under disease conditions.

All experiment reagents were acquired from commercial suppliers. Ultrapure lipopolysaccharide (LPS) was bought from InvivoGen and nigericin was bought from Sigma-Aldrich. Materials required to culture the iBMDMs were all acquired from Gibco Life Technologies, such as DMEM (Dulbecco’s Modified Eagle Medium), heat-inactivated fetal bovine serum (FBS), the penicillin/streptomycin antibiotic cocktail, and trypsin–EDTA. AIM BIOTECH supplied the microfluidic devices and their accompanying luer-lock connectors; Fisher Scientific supplied sterile luer-lock syringes. Auxiliary materials for shear stress application (i.e., tubing and luer-lock to barb end adaptors) were acquired from Cole Parmer. NucRed Live 647 ReadyProbes Reagent, NucBlue, Fluoro-4 AM, and propidium iodide, were purchased from ThermoFisher Scientific. Amicon centrifugal filter units were purchased from Millipore Sigma. Dooku1 was acquired from MedChemExpress. Gasdermin-D and vinculin western antibodies were acquired from ABCAM and Cell Signaling Technologies, respectively. Disposables to run WES were purchased from BioTechne. Immortalized bone marrow-derived macrophages (iBMDMs) used for in vitro experiments contained a modified ASC protein tagged with a CFP residue. All iBMDMs (wild-type and knock outs) were gifted to us from Dr. Kate Fitzgerald from the University of Massachusetts Chan Medical School. All iBMDMs were cultured as previously described by Fish and Kulkarni. 28 Shear stress treatment and setting up the microfluidic system for these experiments were both done according to the procedures described by Fish and Kulkarni. 28 DMEM was loaded into the syringes for all experiments with the exception of ELISA and capillary western experiments where basal DMEM was used. iBMDMs seeded in a microfluidic device were incubated with 100 ng mL −1 LPS for 4 h, followed by 30 min shear stress exposure. NucRed (2 drops per mL) and propidium iodide (2 μ g mL −1 ) were incubated at 37 °C for 30 min to stain the cells. LPS only and shear stress controls were performed according to the previously described time points. Imaging was performed at 20× magnification on an A1R-TIRF Confocal Microscope. iBMDMs were seeded in a microfluidic device as previously described. The cells were pretreated with 100 ng mL −1 LPS for 4 h, followed by shear stress treatment for 30 min. After shear stress exposure, the cells were stained with 2.5 μ M Fluo-4AM dye in CPBS (1X PBS supplemented with CaCl 2 and MgCl 2 ) to stain for intracellular calcium for 10 min, followed by NucBlue (2 drops per mL) in media for 20 min. An LPS and LPS + 1 h Nigericin treatment groups were performed as controls. The cells were imaged at 20X on a CREST v2 TIRF Spinning Disc Confocal Microscope. Media samples were extracted from microfluidic devices or collection vessels and treated with HALT Protease Inhibitor Cocktail (ThermoFisher). Media from shear stress samples were concentrated using Amicon Ultra-15 10 kDa MWCO centrifugation filter units before the addition of the HALT Protease Inhibitor Cocktail. IL-1 β ELISA (Invitrogen), CCL2 ELISA (BioLegend), CXCL1 ELISA (BioLegend), and CXCL2 ELISA (ThermoFisher) quantified protein concentration in microfluidic supernatants according to manufaturer’s protocols. Microfluidic devices were seeded with iBMDMs at a density of 50 million cells/mL (500,000 cells per microfluidic device). The cells were pretreated for 4 h with 100 ng mL −1 LPS, followed by 50 dyn/cm 2 treatment for 30 min for the shear-stress-mitigated NLRP3 inflammasome activation group. Untreated, LPS, and LPS + 1 h nigericin treatment groups were performed as controls. Final treatment groups for these experiments were done in basal DMEM. The collagen gel of microfluidic devices was digested using Collagenase D (Roche) in Basal DMEM at 1 mg/mL for 15 min at 37 °C and 5% CO 2 . Cells were pelleted before processing of samples. Post treatment cells were collected and washed with 1X PBS. Washed cells were then resuspended in 20 μ L of 1X RIPA (ThermoFisher) with 1X HALT Protease Inhibitor. Cells were sonicated and supernatant was collected for BCA. Using a standard curve protein concentration was determined and 2.8 μ g of protein per group was loaded to each well. Following dilutions of targets, antibodies were loaded in the WES plate (1:50 Gasdermin-D, 1:50 Vinculin). Automated Western was performed using the WES machine, vinculin was used as housekeeping protein for Gasdermin-D. The collagen gel in the microfluidic devices were digested as previously described before performing RNA using TRIzol Reagent (Invitrogen) in accordance with the manufacturer’s protocol. Nanodrop determined sample concentration and purity. High-Capacity cDNA Reverse Transcriptase Kit (Applied Biosystems) reverse transcribed total RNA according to the manufacturer’s protocol. TaqMan Gene Expression Assay primers (Applied Biosystems) and TaqMan Fast Advanced Master Mix (Applied Biosystems) primers were used to perform qPCR according to the manufacturer’s protocols. The relative expression of the gene of interest was normalized by that of β -actin mRNA. Table S1 contains information about the qPCR primers purchased, including their assay ID numbers. In order to assess the presence of ASC–CFP specks in response to Piezo1 inhibition using microscopy, LPS-primed iBMDMs were treated with Piezo1 inhibitor Dooku1 at a concentration of 50 μ M for 2 h before 50 dyn/cm 2 shear stress treatment for 30 min and confocal microscopy staining and imaging as previously illustrated. For quantification of qPCR experiments, LPS-primed iBMDMs were incubated with 50 μ M Dooku1 for 2 h and exposed to 50 dyn/cm 2 shear stress for 30 min before isolating RNA through RNA extraction and quantifying gene expression through qPCR. GraphPad Prism 8 software was utilized to plot all graphs and perform statistical analyses. An unpaired two-tailed t -test was used to comparatively analyze two. Ordinary one-way or two-way ANOVA followed by Tukey post-test was used for multiple-group comparative analyses. All the data are displayed as mean ± s.e.m. (standard error of the mean). p value <0.05 was considered as significant.

In summary, we provided a mechanistic understanding of how shear stress can selectively influence the activation of the NLRP3 inflammasome through mechanosensitive ion channel Piezo1. Utilizing KO iBMDMs, we determined the involvement of NLRP3 and caspase-1 in the formation of ASC–CFP specks, calcium influx, gasdermin-D cleavage, and IL-1 β secretion, determining that shear stress selectively activates the NLRP3 inflammasome through the canonical pathway. We determined that shear-stress-mediated inflammasome activation induces calcium signaling through the mechanosensitive ion channel Piezo1. Through qPCR, we demonstrated upregulation of Piezo1 and NLRP3 expression when LPS-primed iBMDMs were treated with shear stress. ELISA of CXCL1, CXCL2, and CCL2 showed that LPS + shear stress upregulates CCL2 in macrophages and has no effect on their receptors CCR2 and CXCR2. This means that shear stress could have the ability to increase the presence of inflammatory monocytes at the site of inflammation through secretion of IL-1 β and CCL2, which have been shown to control fibrosis in pancreatitis disease conditions and increase the probability of developing worsening pancreatic conditions, such as pancreatic cancer. This work provides a foundation for understanding how changes in physiological shear stress can regulate chronic inflammation, fibrosis, and the progression of disease states. Additionally, understanding how the body’s physiological environment regulates inflammatory conditions can allow us to build better in vitro organ- and disease-on-a-chip models to screen next-generation therapeutics for improving patient outcomes.

One of the greatest threats to human health is chronic inflammation which can present in many forms throughout the body. One example of this is inflammation of the pancreas, pancreatitis, which is a leading cause of gastrointestinal-related hospitalizations in the United States and is one of the main risk factors for pancreatic cancer. 1 – 4 One of the primary regulators of chronic inflammatory diseases in innate immune cells is a multiprotein complex called the NLRP3 inflammasome, which responds to microbial invasion or cellular stress and danger signals. 5 – 8 Poor regulation of the NLRP3 inflammasome has been shown to contribute to the pathogenesis and progression of pancreatitis into worsening conditions. 9 – 11 Current two-dimensional macroscale assays used in vitro to study these diseases are lacking physiological relevance to their manifestation and progression in the body. 12 Therefore, there is a strong need for more physiologically relevant in vitro assays that can more precisely correlate 2D in vitro methods, animal models, and clinical trials since the majority of costs in therapeutic research and development is due to inadequate cell culture, animal, or computational models, which result in failure. 13 – 15 Microfluidic devices have been widely utilized to bridge this gap in therapeutic development in chronic inflammatory diseases due to their ability to accurately recapitulate the in vivo physiology in an in vitro setting. 12 , 16 One biophysical parameter that is crucial for studying cells in a physiologically relevant in vitro environment is the application of shear stress to cells. Shear stress has been widely demonstrated to affect gene expression, cytokine secretion, and ion transport; however, its ability to mount an inflammatory response or progress chronic inflammatory diseases in immune cells has not been widely studied. 17 , 18 The influx of extracellular calcium ions through the cell membrane, a well-established activator of the NLRP3 inflammasome, has been demonstrated to be regulated in numerous cell types by mechanosensitive ion channel Piezo1 at physiological shear stress. 19 – 22 Increases in cytosolic calcium through calcium influxes have also been demonstrated to be associated with the progression of pancreatitis. 10 , 23 – 25 Canonical NLRP3 inflammasome activation is process that requires two steps: a priming signal (Signal 1) and an activation signal (Signal 2). 21 , 26 Signal 1 is initiated by toll-like-receptor 4 (TLR4) when it recognizes well-conserved pathogen-associated molecular patterns (PAMPs) on the surface of Gram-negative bacteria, such as lipopolysaccharide (LPS), which initiates downstream signaling and activation of transcription factor NF-kB. 21 , 27 As a result of activation, NF-kB dimers travel to the nucleus, where they bind to DNA to initiate the transcription of inactive proteins, to be used later to activate the NLRP3 inflammasome; these proteins include NLRP3 (NOD-, LRR-, and pyrin-domain containing protein 3), pro-IL-18, and pro-IL-1 β . 6 , 7 , 26 The transcription of inactive gasdermin-D, pro-caspase-1, and apoptosis-associated speck-like protein containing CARD (ASC) also occurs, however, their production is independent of NF-kB activation. 21 , 28 Signal 2 can be transduced through any one or several combinations of the following methods: potassium efflux, calcium influx, mitochondrial reactive oxygen species (mROS) formation, lysosomal rupture, or other types of damage-associated molecular patterns (DAMPs). The initiation of Signal 2 causes the inactive proteins produced during Signal 1 to oligomerize, forming the NLRP3 inflammasome complex and the cleavage of caspase-1 into its active form. 5 , 22 Active caspase-1 enzyme next cleaves pore-forming protein gasdermin-D and pro-inflammatory cytokines IL-18 and IL-1 β into their active forms, allowing pro-inflammatory cytokines to be exported from the cell, initiating pyroptotic cell death. 5 , 21 , 26 In the pathogenesis of pancreatitis, TLR4 has been shown to be upregulated, play a role in tissue damage which is responsive to LPS treatments, and initiate downstream activation of NF-kB resulting in the production of inactive proteins for NLRP3 activation. 29 , 30 Stimulation of TLR4 primes the inflammasome for activation, where exposure to physiological shear stress has the potential to initiate activation and further progress pancreatitis. Despite numerous studies demonstrating the prevalence of Piezo1 and calcium dysregulation in pancreatitis, extensive mechanistic studies to determine how physiological shear stress can activate the NLRP3 inflammasome has not been explored. 24 , 25 , 31 , 32 In this study, we investigated how NLRP3 inflammasome activation is affected in LPS-primed immortalized bone marrow-derived macrophages (iBMDMs) treated with three physiologically relevant magnitudes of shear stress in a microfluidic device. We investigated how the presence of ASC–CFP specks was modulated by shear stress exposure magnitude in wild-type (WT) as well as NLRP3 and caspase-1 knockout (KO) iBMDMs to verify that shear stress only affects NLRP3 inflammasome activation. We also studied how shear stress affected the calcium influx, gasdermin-D cleavage, and the release of IL-1 β initiated in iBMDMs to confirm the mechanism through which NLRP3 inflammasome activation was occurring. To discern a more detailed mechanism of how shear stress exposure affects NLRP3 inflammasome activation, quantitative polymerase chain reaction (qPCR) was performed on LPS-primed iBMDMs exposed to shear stress, measuring the relative expression of NLRP3, Piezo1, IL-1 β , gasdermin-D, caspase-1, CCR2 (chemokine (C–C motif) receptor 2), and CXCR2 (Chemokine (C-X-C motif) receptor 2) normalized to β -actin control. We also performed ELISA on IL-1 β , CCL2 (chemokine (C–C motif) ligand 2), CXCL1 (chemokine (C-X-C) ligand 1), and CXCL2 (chemokine (C-X-C) ligand 2) to further discern how shear-stress activation of the NLRP3 inflammasome affected cytokine secretion in macrophages and how this could affect other immune cell trafficking to the site of inflammation. An inhibitor study was also performed using Dooku1, a well-studied Piezo1 inhibitor, to determine the necessity of Piezo1 in this signaling pathway further. In our results, we discovered a strong correlation between shear stress magnitude, calcium signaling, gasdermin-d cleavage, and ASC speck formation, indicating the inflammasome-activating capabilities of mechanical forces, such as shear stress. These were confirmed through multiple methods including confocal microscopy, qPCR, capillary western, and ELISA. We also demonstrated that shear stress modulates cytokine secretion, having the potential to modulate the trafficking of specific immune cells to the site of inflammation and affect fibrosis in chronic inflammatory diseases. Understanding the effect of shear stress on a physiologically relevant inflammasome activation model can provide new insights into how the progression of chronic inflammation is affected by the pathogenesis of other chronic inflammatory diseases.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.4c18845 . Catalog numbers with the associated qPCR primers, schematic of the microfluidic device used to study shear stress in iBMDMs, Western blots for gasdermin-d and vinculin using capillary western (WES). Table S1 . qPCR primers the primers used for qPCR were purchased from Applied Biosystems. Figure S1 . idenTx 3 AIM Biotech Schematic Figure S2 . Western Blots (A) representative blot of inactive (53 kDa) and active (35 kDa) gasdermin-d as well as housekeeping gene vinculin (110 kDa). (B,C) Quantification of active (B) and inactive (C) gasdermin-d relative to vinculin and normalized with the untreated group ( PDF )