Brain-wide Electrical Spatiotemporal Dynamics Encode Depression Vulnerability.

Brain-wide fluctuations in local field potential oscillations reflect emergent network-level signals that mediate behavior. Cracking the code whereby these oscillations coordinate in time and space (spatiotemporal dynamics) to represent complex behaviors would provide fundamental insights into how the brain signals emotional pathology. Using machine learning, we discover a spatiotemporal dynamic network that predicts the emergence of major depressive disorder (MDD)-related behavioral dysfunction in mice subjected to chronic social defeat stress. Activity patterns in this network originate in prefrontal cortex and ventral striatum, relay through amygdala and ventral tegmental area, and converge in ventral hippocampus. This network is increased by acute threat, and it is also enhanced in three independent models of MDD vulnerability. Finally, we demonstrate that this vulnerability network is biologically distinct from the networks that encode dysfunction after stress. Thus, these findings reveal a convergent mechanism through which MDD vulnerability is mediated in the brain.

Geometric deep learning enables 3D kinematic profiling across species and environments.

Comprehensive descriptions of animal behavior require precise three-dimensional (3D) measurements of whole-body movements. Although two-dimensional approaches can track visible landmarks in restrictive environments, performance drops in freely moving animals, due to occlusions and appearance changes. Therefore, we designed DANNCE to robustly track anatomical landmarks in 3D across species and behaviors. DANNCE uses projective geometry to construct inputs to a convolutional neural network that leverages learned 3D geometric reasoning. We trained and benchmarked DANNCE using a dataset of nearly seven million frames that relates color videos and rodent 3D poses. In rats and mice, DANNCE robustly tracked dozens of landmarks on the head, trunk, and limbs of freely moving animals in naturalistic settings. We extended DANNCE to datasets from rat pups, marmosets, and chickadees, and demonstrate quantitative profiling of behavioral lineage during development.

Fatty acid binding protein 5 inhibition attenuates pronociceptive cytokine/chemokine expression and suppresses osteoarthritis pain: A comparative human and rat study.

OBJECTIVE: Osteoarthritis (OA) is often accompanied by debilitating pain that is refractory to available analgesics due in part to the complexity of signaling molecules that drive OA pain and our inability to target these in parallel. Fatty acid binding protein 5 (FABP5) is a lipid chaperone that regulates inflammatory pain; however, its contribution to OA pain has not been characterized. DESIGN: This combined clinical and pre-clinical study utilized synovial tissues obtained from subjects with end-stage OA and rats with monoiodoacetate-induced OA. Cytokine and chemokine release from human synovia incubated with a selective FABP5 inhibitor was profiled with cytokine arrays and ELISA. Immunohistochemical analyses were conducted for FABP5 in human and rat synovium. The efficacy of FABP5 inhibitors on pain was assessed in OA rats using incapacitance as an outcome. RNA-seq was then performed to characterize the transcriptomic landscape of synovial gene expression in OA rats treated with FABP5 inhibitor or vehicle. RESULTS: FABP5 was expressed in human synovium and FABP5 inhibition reduced the secretion of pronociceptive cytokines (interleukin-6 [IL6], IL8) and chemokines (CCL2, CXCL1). In rats, FABP5 was upregulated in the OA synovium and its inhibition alleviated incapacitance. The transcriptome of the rat OA synovium exhibited >6000 differentially expressed genes, including the upregulation of numerous pronociceptive cytokines and chemokines. FABP5 inhibition blunted the upregulation of the majority of these pronociceptive mediators. CONCLUSIONS: FABP5 is expressed in the OA synovium and its inhibition suppresses pronociceptive signaling and pain, indicating that FABP5 inhibitors may constitute a novel class of analgesics to treat OA.

Threshold and Laser Conversion in Nanostructured-Resonator Parametric Oscillators.

We explore optical parametric oscillation (OPO) in nanophotonic resonators, enabling arbitrary, nonlinear phase matching and nearly lossless control of energy conversion. Such pristine OPO laser converters are determined by nonlinear light-matter interactions, making them both technologically flexible and broadly reconfigurable. We utilize a nanostructured inner-wall modulation in the resonator to achieve universal phase matching for OPO-laser conversion, but coherent backscattering also induces a counterpropagating pump laser. This depletes the intraresonator optical power in either direction, increasing the OPO threshold power and limiting laser-conversion efficiency, the ratio of optical power in target signal and idler frequencies to the pump. We develop an analytical model of this system that emphasizes an understanding of optimal laser-conversion and threshold behaviors, and we use the model to guide experiments with nanostructured-resonator OPO laser-conversion circuits, fully integrated on chip and unlimited by group-velocity dispersion. Our Letter demonstrates the fundamental connection between OPO laser-conversion efficiency and the resonator coupling rate, subject to the relative phase and power of counterpropagating pump fields. We achieve (40±4) mW of on-chip power, corresponding to (41±4)% conversion efficiency, and discover a path toward near-unity OPO laser-conversion efficiency.

Death-associated protein kinase 3 regulates the myogenic reactivity of cerebral arteries.

NEW FINDINGS: What is the central question of this study? DAPK3 contributes to the Ca2+ -sensitization of vascular smooth muscle contraction: does this protein kinase participate in the myogenic response of cerebral arteries? What is the main finding and its importance? Small molecule inhibitors of DAPK3 effectively block the myogenic responses of cerebral arteries. HS38-dependent changes to vessel constriction occur independent of LC20 phosphorylation, and therefore DAPK3 appears to operate via the actin cytoskeleton. A role for DAPK3 in the myogenic response was not previously reported, and the results support a potential new therapeutic target in the cerebrovascular system. ABSTRACT: The vascular smooth muscle (VSM) of resistance blood vessels is a target of intrinsic autoregulatory responses to increased intraluminal pressure, the myogenic response. In the brain, the myogenic reactivity of cerebral arteries is critical to homeostatic blood flow regulation. Here we provide the first evidence to link the death-associated protein kinase 3 (DAPK3) to the myogenic response of rat and human cerebral arteries. DAPK3 is a Ser/Thr kinase involved in Ca2+ -sensitization mechanisms of smooth muscle contraction. Ex vivo administration of a specific DAPK3 inhibitor (i.e., HS38) could attenuate vessel constrictions invoked by serotonin as well as intraluminal pressure elevation. The HS38-dependent dilatation was not associated with any change in myosin light chain (LC20) phosphorylation. The results suggest that DAPK3 does not regulate Ca2+ sensitization pathways during the myogenic response of cerebral vessels but rather operates to control the actin cytoskeleton. A slow return of myogenic tone was observed during the sustained ex vivo exposure of cerebral arteries to HS38. Recovery of tone was associated with greater LC20 phosphorylation that suggests intrinsic signalling compensation in response to attenuation of DAPK3 activity. Additional experiments with VSM cells revealed HS38- and siDAPK-dependent effects on the actin cytoskeleton and focal adhesion kinase phosphorylation status. The translational importance of DAPK3 to the human cerebral vasculature was noted, with robust expression of the protein kinase and significant HS38-dependent attenuation of myogenic reactivity found for human pial vessels.

Supervising the Decoder of Variational Autoencoders to Improve Scientific Utility.

Probabilistic generative models are attractive for scientific modeling because their inferred parameters can be used to generate hypotheses and design experiments. This requires that the learned model provides an accurate representation of the input data and yields a latent space that effectively predicts outcomes relevant to the scientific question. Supervised Variational Autoencoders (SVAEs) have previously been used for this purpose, as a carefully designed decoder can be used as an interpretable generative model of the data, while the supervised objective ensures a predictive latent representation. Unfortunately, the supervised objective forces the encoder to learn a biased approximation to the generative posterior distribution, which renders the generative parameters unreliable when used in scientific models. This issue has remained undetected as reconstruction losses commonly used to evaluate model performance do not detect bias in the encoder. We address this previously-unreported issue by developing a second-order supervision framework (SOS-VAE) that updates the decoder parameters, rather than the encoder, to induce a predictive latent representation. This ensures that the encoder maintains a reliable posterior approximation and the decoder parameters can be effectively interpreted. We extend this technique to allow the user to trade-off the bias in the generative parameters for improved predictive performance, acting as an intermediate option between SVAEs and our new SOS-VAE. We also use this methodology to address missing data issues that often arise when combining recordings from multiple scientific experiments. We demonstrate the effectiveness of these developments using synthetic data and electrophysiological recordings with an emphasis on how our learned representations can be used to design scientific experiments.

A highly selective inhibitor of interleukin-1 receptor-associated kinases 1/4 (IRAK-1/4) delineates the distinct signaling roles of IRAK-1/4 and the TAK1 kinase.

Interleukin-1 receptor-associated kinase-1 (IRAK-1) and IRAK-4, as well as transforming growth factor ß-activated kinase 1 (TAK1), are protein kinases essential for transducing inflammatory signals from interleukin receptors. IRAK family proteins and TAK1 have high sequence identity within the ATP-binding pocket, limiting the development of highly selective IRAK-1/4 or TAK1 inhibitors. Beyond kinase activity, IRAKs and TAK1 act as molecular scaffolds along with other signaling proteins, complicating the interpretation of experiments involving knockin or knockout approaches. In contrast, pharmacological manipulation offers the promise of targeting catalysis-mediated signaling without grossly disrupting the cellular architecture. Recently, we reported the discovery of takinib, a potent and highly selective TAK1 inhibitor that has only marginal activity against IRAK-4. On the basis of the TAK1-takinib complex structure and the structure of IRAK-1/4, here we defined critical contact sites of the takinib scaffold within the nucleotide-binding sites of each respective kinase. Kinase activity testing of takinib analogs against IRAK-4 identified a highly potent IRAK-4 inhibitor (HS-243). In a kinome-wide screen of 468 protein kinases, HS-243 had exquisite selectivity toward both IRAK-1 (IC50 = 24 nm) and IRAK-4 (IC50 = 20 nm), with only minimal TAK1-inhibiting activity (IC50 = 0.5 µm). Using HS-243 and takinib, we evaluated the consequences of cytokine/chemokine responses after selective inhibition of IRAK-1/4 or TAK1 in response to lipopolysaccharide challenge in human rheumatoid arthritis fibroblast-like synoviocytes. Our results indicate that HS-243 specifically inhibits intracellular IRAKs without TAK1 inhibition and that these kinases have distinct, nonredundant signaling roles.

Missed Opportunities for Human Immunodeficiency Virus (HIV) Testing During Injection Drug Use-Related Healthcare Encounters Among a Cohort of Persons Who Inject Drugs With HIV Diagnosed During an Outbreak-Cincinnati/Northern Kentucky, 2017-2018.

BACKGROUND: Persons who inject drugs (PWID) have frequent healthcare encounters related to their injection drug use (IDU) but are often not tested for human immunodeficiency virus (HIV). We sought to quantify missed opportunities for HIV testing during an HIV outbreak among PWID. METHODS: PWID with HIV diagnosed in 5 Cincinnati/Northern Kentucky counties during January 2017-September 2018 who had ≥1 encounter 12 months prior to HIV diagnosis in 1 of 2 Cincinnati/Northern Kentucky area healthcare systems were included in the analysis. HIV testing and encounter data were abstracted from electronic health records. A missed opportunity for HIV testing was defined as an encounter for an IDU-related condition where an HIV test was not performed and had not been performed in the prior 12 months. RESULTS: Among 109 PWID with HIV diagnosed who had ≥1 healthcare encounter, 75 (68.8%) had ≥1 IDU-related encounters in the 12 months before HIV diagnosis. These 75 PWID had 169 IDU-related encounters of which 86 (50.9%) were missed opportunities for HIV testing and occurred among 46 (42.2%) PWID. Most IDU-related encounters occurred in the emergency department (118/169; 69.8%). Using multivariable generalized estimating equations, HIV testing was more likely in inpatient compared with emergency department encounters (adjusted relative risk [RR], 2.72; 95% confidence interval [CI], 1.70-4.33) and at the healthcare system receiving funding for emergency department HIV testing (adjusted RR, 1.76; 95% CI, 1.10-2.82). CONCLUSIONS: PWID have frequent IDU-related encounters in emergency departments. Enhanced HIV screening of PWID in these settings can facilitate earlier diagnosis and improve outbreak response.

Group-velocity-dispersion engineering of tantala integrated photonics.

Designing integrated photonics, especially to leverage Kerr-nonlinear optics, requires accurate and precise knowledge of the refractive index across the visible to infrared spectral ranges. Tantala (Ta2O5) is an emerging material platform for integrated photonics and nanophotonics that offers broadband ultralow loss, moderately high nonlinearity, and advantages for scalable and heterogeneous integration. We present refractive index measurements on a thin film of tantala, and we explore the efficacy of this data for group-velocity-dispersion (GVD) engineering with waveguide and ring-resonator devices. In particular, the observed spectral extent of supercontinuum generation in fabricated waveguides and the wavelength dependence of free spectral range (FSR) in optical resonators provide a sensitive test of our integrated photonics design process. Our work opens up new design possibilities with tantala, including with octave-spanning soliton microcombs.

Broadband, electro-optic, dual-comb spectrometer for linear and nonlinear measurements.

We demonstrate a dual-comb spectrometer based on electro-optic modulation of a continuous-wave laser at 10 GHz. The system simultaneously offers fast acquisition speed and ultrabroad spectral coverage, spanning 120 THz across the near infrared. Our spectrometer is highly adaptable, and we demonstrate absorption spectroscopy of atmospheric gases and a dual-comb configuration that captures nonlinear Raman spectra of semiconductor materials via coherent anti-Stokes Raman scattering. The ability to rapidly and simultaneously acquire broadband spectra with high frequency resolution and high sensitivity points to new possibilities for hyperspectral sensing in fields such as remote sensing, biological detection and imaging, and machine vision.

Nanophotonic tantala waveguides for supercontinuum generation pumped at 1560 nm.

We experimentally demonstrate efficient and broadband supercontinuum generation in nonlinear tantala (Ta2O5) waveguides using a 1560 nm femtosecond seed laser. With incident pulse energies as low as 100 pJ, we create spectra spanning up to 1.6 octaves across the visible and infrared. Fabricated devices feature propagation losses as low as 10 dB/m, and they can be dispersion engineered through lithographic patterning for specific applications. We show a waveguide design suitable for low-power self-referencing of a fiber frequency comb that produces dispersive-wave radiation directly at the second-harmonic wavelength of the seed laser. A fiber-connectorized, hermetically sealed module with 2 dB per facet insertion loss and watt-level average-power handling is also described. Highly efficient and fully packaged tantala waveguides may open new possibilities for the integration of nonlinear nanophotonics into systems for precision timing, quantum science, biological imaging, and remote sensing.

Mid-infrared frequency combs at 10 GHz.

We demonstrate mid-infrared (MIR) frequency combs at 10 GHz repetition rate via intra-pulse difference-frequency generation (DFG) in quasi-phase-matched nonlinear media. Few-cycle pump pulses (≲15fs, 100 pJ) from a near-infrared electro-optic frequency comb are provided via nonlinear soliton-like compression in photonic-chip silicon-nitride waveguides. Subsequent intra-pulse DFG in periodically poled lithium niobate waveguides yields MIR frequency combs in the 3.1-4.8 µm region, while orientation-patterned gallium phosphide provides coverage across 7-11 µm. Cascaded second-order nonlinearities simultaneously provide access to the carrier-envelope-offset frequency of the pump source via in-line f-2f nonlinear interferometry. The high-repetition rate MIR frequency combs introduced here can be used for condensed phase spectroscopy and applications such as laser heterodyne radiometry.

Machine learning does not improve upon traditional regression in predicting outcomes in atrial fibrillation: an analysis of the ORBIT-AF and GARFIELD-AF registries.

AIMS: Prediction models for outcomes in atrial fibrillation (AF) are used to guide treatment. While regression models have been the analytic standard for prediction modelling, machine learning (ML) has been promoted as a potentially superior methodology. We compared the performance of ML and regression models in predicting outcomes in AF patients. METHODS AND RESULTS: The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) and Global Anticoagulant Registry in the FIELD (GARFIELD-AF) are population-based registries that include 74 792 AF patients. Models were generated from potential predictors using stepwise logistic regression (STEP), random forests (RF), gradient boosting (GB), and two neural networks (NNs). Discriminatory power was highest for death [STEP area under the curve (AUC) = 0.80 in ORBIT-AF, 0.75 in GARFIELD-AF] and lowest for stroke in all models (STEP AUC = 0.67 in ORBIT-AF, 0.66 in GARFIELD-AF). The discriminatory power of the ML models was similar or lower than the STEP models for most outcomes. The GB model had a higher AUC than STEP for death in GARFIELD-AF (0.76 vs. 0.75), but only nominally, and both performed similarly in ORBIT-AF. The multilayer NN had the lowest discriminatory power for all outcomes. The calibration of the STEP modelswere more aligned with the observed events for all outcomes. In the cross-registry models, the discriminatory power of the ML models was similar or lower than the STEP for most cases. CONCLUSION: When developed from two large, community-based AF registries, ML techniques did not improve prediction modelling of death, major bleeding, or stroke.

An FDA/CDER perspective on nonclinical testing strategies: Classical toxicology approaches and new approach methodologies (NAMs).

Nonclinical testing of human pharmaceuticals is conducted to assess the safety of compounds to be studied in human clinical trials and for marketing of new drugs. Although there is no exact number and type of nonclinical studies required for safety assessments, as there is inherent flexibility for each new compound, the traditional approach is outlined in various FDA and ICH guidance documents and involves a combination of in vitro assays and whole animal testing methods. Recent advances in science have led to the emergence of numerous new approach methodologies (NAMs) for nonclinical testing that are currently being used in various aspects of drug development. Traditional nonclinical testing methods can predict clinical outcomes, although improvements in these methods that can increase predictivity of clinical outcomes are encouraged and needed. This paper discusses FDA/CDER's view on the opportunities and challenges of using NAMs in drug development especially for regulatory purposes, and also includes examples where NAMs are currently being used in nonclinical safety assessments and where they may supplement and/or enhance current testing methods. FDA/CDER also encourages communication with stakeholders regarding NAMs and is committed to exploring the use of NAMs to improve regulatory efficiency and potentially expedite drug development.

A Shared Vision for Machine Learning in Neuroscience.

With ever-increasing advancements in technology, neuroscientists are able to collect data in greater volumes and with finer resolution. The bottleneck in understanding how the brain works is consequently shifting away from the amount and type of data we can collect and toward what we actually do with the data. There has been a growing interest in leveraging this vast volume of data across levels of analysis, measurement techniques, and experimental paradigms to gain more insight into brain function. Such efforts are visible at an international scale, with the emergence of big data neuroscience initiatives, such as the BRAIN initiative (Bargmann et al., 2014), the Human Brain Project, the Human Connectome Project, and the National Institute of Mental Health's Research Domain Criteria initiative. With these large-scale projects, much thought has been given to data-sharing across groups (Poldrack and Gorgolewski, 2014; Sejnowski et al., 2014); however, even with such data-sharing initiatives, funding mechanisms, and infrastructure, there still exists the challenge of how to cohesively integrate all the data. At multiple stages and levels of neuroscience investigation, machine learning holds great promise as an addition to the arsenal of analysis tools for discovering how the brain works.

Generating few-cycle pulses with integrated nonlinear photonics.

Ultrashort laser pulses that last only a few optical cycles have been transformative tools for studying and manipulating light-matter interactions. Few-cycle pulses are typically produced from high-peak-power lasers, either directly from a laser oscillator or through nonlinear effects in bulk or fiber materials. Now, an opportunity exists to explore the few-cycle regime with the emergence of fully integrated nonlinear photonics. Here, we experimentally and numerically demonstrate how lithographically patterned waveguides can be used to generate few-cycle laser pulses from an input seed pulse. Moreover, our work explores a design principle in which lithographically varying the group-velocity dispersion in a waveguide enables the creation of highly constant-intensity supercontinuum spectra across an octave of bandwidth. An integrated source of few-cycle pulses could broaden the range of applications for ultrafast light sources, including supporting new lab-on-a-chip systems in a scalable form factor.

Dual-comb spectroscopy with tailored spectral broadening in Si3N4 nanophotonics.

Si3N4 waveguides, pumped at 1550 nm, can provide spectrally smooth, broadband light for gas spectroscopy in the important 2 µm to 2.5 µm atmospheric water window, which is only partially accessible with silica-fiber based systems. By combining Er+ fiber frequency combs and supercontinuum generation in tailored Si3N4 waveguides, high signal-to-noise dual-comb spectroscopy spanning 2 µm to 2.5 µm is demonstrated. Acquired broadband dual-comb spectra of CO and CO2 agree well with database line shape models and have a spectral-signal-to-noise as high as 48/√s, showing that the high coherence between the two combs is retained in the Si3N4 supercontinuum generation. The dual-comb spectroscopy figure of merit is 6 × 106/√s, equivalent to that of all-fiber dual-comb spectroscopy systems in the 1.6 µm band. based on these results, future dual-comb spectroscopy can combine fiber comb technology with Si3N4 waveguides to access new spectral windows in a robust non-laboratory platform.

Application of immobilized ATP to the study of NLRP inflammasomes.

The NLRP proteins are a subfamily of the NOD-like receptor (NLR) innate immune sensors that possess an ATP-binding NACHT domain. As the most well studied member, NLRP3 can initiate the assembly process of a multiprotein complex, termed the inflammasome, upon detection of a wide range of microbial products and endogenous danger signals and results in the activation of pro-caspase-1, a cysteine protease that regulates multiple host defense pathways including cytokine maturation. Dysregulated NLRP3 activation contributes to inflammation and the pathogenesis of several chronic diseases, and the ATP-binding properties of NLRPs are thought to be critical for inflammasome activation. In light of this, we examined the utility of immobilized ATP matrices in the study of NLRP inflammasomes. Using NLRP3 as the prototypical member of the family, P-linked ATP Sepharose was determined to be a highly-effective capture agent. In subsequent examinations, P-linked ATP Sepharose was used as an enrichment tool to enable the effective profiling of NLRP3-biomarker signatures with selected reaction monitoring-mass spectrometry (SRM-MS). Finally, ATP Sepharose was used in combination with a fluorescence-linked enzyme chemoproteomic strategy (FLECS) screen to identify potential competitive inhibitors of NLRP3. The identification of a novel benzo[d]imidazol-2-one inhibitor that specifically targets the ATP-binding and hydrolysis properties of the NLRP3 protein implies that ATP Sepharose and FLECS could be applied other NLRPs as well.

Extended duration of dilator use beyond 1 year may reduce vaginal stenosis after intravaginal high-dose-rate brachytherapy.

BACKGROUND: Vaginal dilators (VD) are recommended following vaginal or pelvic radiotherapy for patients with endometrial carcinoma (EC) to prevent vaginal stenosis (VS). The time course of VS is not fully understood and the optimal duration of VD use is unknown. METHODS: We reviewed 243 stage IA-II EC patients who received adjuvant brachytherapy (BT) at an academic tertiary referral center. Patients were instructed to use their VD three times per week for at least 1-year duration. The primary outcome was development of grade ≥ 1 VS using CTCAEv4 criteria during the follow-up period. The log-rank test and multivariable Cox proportional hazards modeling were used to evaluate the effect of VD use (noncompliance vs. standard compliance [up to 1 year] vs. extended compliance [over 1 year]) on VS. RESULTS: The median follow-up was 15.2 months over the 5-year study period. At 15 months, the incidence of VS was 38.8% for noncompliant patients, 33.5% for those with standard compliance, and 21.4% for those with extended compliance (median time to grade ≥ 1 VS was 17.5 months, 26.7 months, and not yet reached for these groups, respectively). On multivariable Cox regression analysis, extended compliance remained a significant predictor of reduced VS risk when compared to both noncompliance (HR 0.38, 95% CI 0.18-0.80, p = 0.012) and standard compliance (HR 0.43, 95% CI 0.20-0.89, p = 0.023). CONCLUSIONS: The risk of VS persists beyond 1 year after BT. Extended VD compliance beyond 1 year may mitigate this risk.

Synergistic role of HSP90α and HSP90ß to promote myofibroblast persistence in lung fibrosis.

Idiopathic pulmonary fibrosis (IPF) is a progressive disease of the lung parenchyma, causing significant morbidity through worsening dyspnoea and overall functional decline. IPF is characterised by apoptosis-resistant myofibroblasts, which are a major source for the excessive production of extracellular matrix (ECM) overtaking normal lung tissue. We sought to study the role of heat shock protein (HSP) isoforms HSP90α and HSP90ß, whose distinct roles in lung fibrogenesis remain elusive.We determined the level of circulating HSP90α in IPF patients (n=31) and age-matched healthy controls (n=9) by ELISA. The release of HSP90α and HSP90ß was evaluated in vitro in primary IPF and control lung fibroblasts and ex vivo after mechanical stretch on fibrotic lung slices from rats receiving adenovector-mediated transforming growth factor-ß1.We demonstrate that circulating HSP90α is upregulated in IPF patients in correlation with disease severity. The release of HSP90α is enhanced by the increase in mechanical stress of the fibrotic ECM. This increase in extracellular HSP90α signals through low-density lipoprotein receptor-related protein 1 (LRP1) to promote myofibroblast differentiation and persistence. In parallel, we demonstrate that the intracellular form of HSP90ß stabilises LRP1, thus amplifying HSP90α extracellular action.We believe that the specific inhibition of extracellular HSP90α is a promising therapeutic strategy to reduce pro-fibrotic signalling in IPF.