HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD.

Airway epithelial damage is a common feature in respiratory diseases such as COPD and has been suggested to drive inflammation and progression of disease. These features manifest as remodeling and destruction of lung epithelial characteristics including loss of small airways which contributes to chronic airway inflammation. Histone deacetylase 6 (HDAC6) has been shown to play a role in epithelial function and dysregulation, such as in cilia disassembly, epithelial to mesenchymal transition (EMT) and oxidative stress responses, and has been implicated in several diseases. We thus used ACY-1083, an inhibitor with high selectivity for HDAC6, and characterized its effects on epithelial function including epithelial disruption, cytokine production, remodeling, mucociliary clearance and cell characteristics. Primary lung epithelial air-liquid interface cultures from COPD patients were used and the impacts of TNF, TGF-β, cigarette smoke and bacterial challenges on epithelial function in the presence and absence of ACY-1083 were tested. Each challenge increased the permeability of the epithelial barrier whilst ACY-1083 blocked this effect and even decreased permeability in the absence of challenge. TNF was also shown to increase production of cytokines and mucins, with ACY-1083 reducing the effect. We observed that COPD-relevant stimulations created damage to the epithelium as seen on immunohistochemistry sections and that treatment with ACY-1083 maintained an intact cell layer and preserved mucociliary function. Interestingly, there was no direct effect on ciliary beat frequency or tight junction proteins indicating other mechanisms for the protected epithelium. In summary, ACY-1083 shows protection of the respiratory epithelium during COPD-relevant challenges which indicates a future potential to restore epithelial structure and function to halt disease progression in clinical practice.

Introduction

The human lungs are lined by a layer of epithelium which serves as a barrier to prevent direct access of inhaled luminal contents to the subepithelium. With each inhalation, the lung epithelium is in contact with the external environment, with the potential of exposure to harmful environmental particles and infectious organisms [1]. The respiratory epithelium constitutes a physical barrier between the outer and inner environments, and the epithelium dictates the initial responses to these stimuli. One mechanism by which the epithelium does this is by tightly regulating transport across the epithelial barrier by means of cell-cell junctions. These include tight junctions (TJs), adherens junctions (AJs), gap junctions, and desmosomes, which all allow for epithelial cells to be connected to their neighbors and are the primary components of the physical barrier formed by the epithelium [2-4]. In addition to regulating the ability of particles to cross through the epithelium, the epithelial junctions also act to segregate the basal compartment from the apical compartment to create epithelial polarization [5, 6].

In recent years the lung epithelium has gained more interest for respiratory diseases such as COPD, asthma and IPF. These diseases are characterized by remodeling and destruction of the lung, resulting in chronic inflammation, susceptibility to infection and loss of small airways. COPD is currently the 4th leading cause of death in the world and accounts for 6% of all deaths globally (as more than 3 million died of COPD in 2012) [7-9]. Even early in disease, patients have changes in their small airways that can be observed by different technologies [10, 11]. During disease progression, infiltration of immune cells and remodeling, such as fibrotic lesions and emphysema, occur. Increased levels of inflammatory cytokines eg TNF, IL-6 and IL-1β, and several chemokines and proteases, such as MMPs, are detected systemically and/or in the lung at diagnosis [12-15]. Despite efforts with anti-inflammatory therapies, there is a huge unmet medical need for new ways of treating respiratory patients and to understand early events leading to inflammation and remodeling [16-18]. Recent papers [11, 19] have discussed the importance of epithelial integrity in lung diseases and reviewed the literature for evidence about epithelial injury as a symptom of damaged epithelium. Strikingly, the epithelial changes often appear before the onset of clinical symptoms such as changes in FEV1 and can be measured in patients using new and more sensitive imaging technologies, eg computed tomography (CT) [20, 21]. Other technologies for assessment of small airway function are emerging, eg oscillometry, enable both an early diagnosis and a way to document efficacy of an intervention [22, 23]. Cigarette smoke, which is a risk factor for COPD, and other irritants have been reported to disrupt TJs and AJs proteins, leading to a dysfunctional epithelium and a disrupted barrier which affects important functions like cilia length and beat frequency, mucus production and the normal organization of the epithelium [4, 24–28]. Mechanisms for this are discussed by Aghapour et al., [19] and there are also studies showing evidence of loss of these epithelial structures in COPD patients [24, 25, 29]. In fact, remodeling of the lung often starts with loss of TJ proteins altering the phenotype of cells resulting in epithelial plasticity with significant change in both the structure and function of the cells [30]. This starts a cascade of events with further remodeling down the line and loss of normal lung functions as a result.

HDAC6 is a histone deacetylase that has been described to play a role in epithelial and endothelial destruction and remodeling. The exact mechanisms for the described effects are to date unknown but its reported functions involve regulation of the cilia cell cycle and autophagy, and has substrates and interacting proteins such as ERK, PKCα, β-catenin, NF-κB, EGFR and peroxiredoxins, that have been implicated in epithelial dysfunction [31, 32]. HDAC6 activity is increased by several stimuli including cigarette smoke and TNF, and is described to regulate cilia disassembly, epithelial to mesenchymal transition (EMT), AJ markers and oxidative stress, both in vitro and in vivo with similar results [33-40]. However, most studies performed to dissect the actions of HDAC6 use inhibitors such as Tubastatin A, Tubacin or Ricolinostat at high concentrations where the impact of other ‘off targets’ and/or other HDACs cannot be disregarded [41-44]. ACY-1083, a small molecule inhibitor highly potent for HDAC6 with a good selective profile, have been reported to have effect in injury-related in vivo models [45-49].

In this paper, we studied the effect of ACY-1083 on epithelial barrier function in primary COPD cells. We show protective features of this inhibitor in different in vitro injury models, where treated cell cultures display reduced damage and cytokine production, and sustained mucus movement resulting in a functioning and intact epithelium.

Material and methods

Characterization of inhibitors

Inhibitors

ACY-1083 (CAS1708113-43-2, Mw 348.35) is commercially available but was prepared according to a published procedure [50, 51]. Tubastatin A (HY-13271, Mw 371.86 HCl salt) and Ricolinostat (HY-16026, Mw 433.5) were obtained from MedChemExpress.

HDAC enzymatic activity

Assays were performed at Eurofins Pharma Discovery Services Catalog reference HDAC1 2491; HDAC2 2492; HDAC3 2083; HDAC4 2493; HDAC5 2494; HDAC6 2495; HDAC7 2610; HDAC8 2247; HDAC9 2611; HDAC10 2662 and HDAC11 2663.

Plasma protein binding

Plasma protein binding analysis was performed as reported previously using human blood plasma samples [52].

Pharmacokinetic analysis in cell media

To determine the stability and binding to plastic and cell media components, cell media was added to a test tube and spiked with ACY-1083 to a final concentration of 10 μM. Compound concentration was then analysed after 18hrs incubation at 37˚C by LC-MSMS (Method DMPK&BAC 001–04).

Secondary pharmacology

Second pharmacology screening was performed at Eurofins Pharma Discovery Services in a panel of 190 in vitro radioligand binding and enzyme assays covering a diverse set of enzymes, receptors, ion channels, and transporters. The cardiovascular panel (hERG, Nav1.5, Kv4.3 and IKs) were performed on the SyncroPatch 384PE (Nanion Technologies) high throughput patch clamp platform (Chinese hamster ovary K1 cell line) at room temperature in a 6 point cumulative assay.

Air-Liquid interface culturing of human bronchial epithelial cells

Patient characteristics from donor cells

Primary COPD human bronchial epithelial cells (HBEC) were purchased from Lonza (Basel, Switzerland). We have used cells from four COPD donors and the information on these donors is limited but include age (years), gender (F/M) and smoking history (yes/no): donor 1 (Cat#00195275, Batch#0000630389, 54, F, yes), donor 2 (Cat#00195275S, Batch 0000370751, 65, F, yes), donor 3 (Cat#00195275S, Batch 0000436083, 59, M, yes), donor 4 (Cat#00195275S, Batch 0000430905, 66, M, yes). Additional five donors of primary healthy HBECs were purchased from Lonza and Epithelix SàRL (Geneva, Switzerland): donor 1 (Cat# hAECB, Batch# AB068001, 71, F, no), donor 2 (CC-2540, Batch# 0000501935, 56, M, yes), donor 3 (CC-2540S, Batch# 0000619260, 65, F, yes), donor 4 (hAECB, Batch# AB079301, 62, M, no) and donor 5 (hAECB, Batch# #AB077201, 63, M, no).

Culturing conditions

Passage one cells were seeded into T75 cell culture flasks (250,000 cells/flask) and expanded in PneumaCult ExPlus medium (StemCell Technologies, Vancouver, Canada) at 37°C, 5% CO2. Once cells reached 70% confluency, they were dissociated with TrypLE Express dissociation media (Thermo Fisher Scientific, Waltham, USA) and frozen at -150°C in ExPlus with 10% DMSO. Passage two cells were thawed and seeded (20,000 cells/membrane) into 24-well HTS Transwell plates (Corning, Wiesbaden, Germany) for air-liquid interface (ALI) culturing. Cells were expanded in ExPlus medium until >50% confluent (four days). Apical medium was then removed, and basolateral medium was replaced with PneumaCult ALI medium (StemCell Technologies). Medium was changed three times per week during a four-week long differentiation phase.

For cilia beating frequency (CBF) and cell velocity measurements, donor cells of primary healthy HBECs were amplified on rat tail collagen I coated flasks and seeded on 12-mm-diameter Transwell inserts [4, 53, 54].

Assessment of epithelial integrity

To determine epithelial integrity of ALI cell cultures the trans-epithelial electrical resistance (TEER) of HBEC ALI cultures was measured using EVOM2 resistance meter (World Precision Instruments Inc, Sarasota, USA). For experiments conducted with healthy donors for the CBF and cell velocity the bronchial epithelial barrier function was evaluated by quantifying TEER using an EVOM epithelial voltohmmeter (World Precision Instruments, FL, USA) connected with the STX2 electrodes.

The paracellular permeability was determined by using fluorescein isothiocyanate labeled 4kD-dextran (Sigma-Aldrich, St. Louis, USA). 18 hours after addition of 0.2 mg FITC-dextran in 200 μL medium to the apical side of the ALI cultures, the fluorescence in the basolateral medium was measured using PHERAstar FSX (excitation 485 nm and emission 520 nm) (BMG LABTECH, Ortenberg, Germany). The time point was chosen for optimal window in the assays as well as practical reasons and has been kept consistent in between different stimulations throughout this work. Data was plotted as fold-change of 0.1% DMSO control.

Challenge models in ALI

All HBEC cultures had been in ALI for four weeks at the start of each experiment. All compounds and stimulus, except for whole cigarette smoke, were added on the basolateral side of the ALI cultures.

For experiments using TNF (R&D Systems, Minneapolis, Canada), TNF (ranging from 5–50 ng/ml) was added together with ACY-1083 or vehicle (0.1% DMSO) on the basolateral side at every medium change for up to ten days in four COPD donors. TEER and permeability were measured after seven days.

For TGF-β1 (R&D Systems) experiments, cells from two COPD donors were pre-treated with ACY-1083 or vehicle (0.1% DMSO) for three days before addition of TGF-β1 (0.4–10 ng/ml). Permeability was measured after 48 hours.

Cigarette smoke extract (CSE) experiments were performed using CSE containing media. In short, smoke from five filterless Kentucky research cigarettes, 3R4F (Kentucky Tobacco Research and Development Center, Lexington, USA), was bubbled through 25 ml of PBS at a speed of five minutes per cigarette. The CSE was filtered through a 0.2 μm sterile filter and stored at -80°C until addition to the basolateral side of HBEC ALI cultures from two COPD donors. Three days of ACY-1083/0.1% DMSO pre-treatment was followed by 48 hours CSE-challenge. Permeability was measured after 48 hours.

Whole cigarette smoke (WCS) experiments were performed using the Smoking Robot VC 10® S-TYPE (Vitrocell, Waldkirch, Germany). The Vitrocell smoke robot delivers either WCS or humidified air to the apical surface of the cell layers. Briefly, inserts with fully differentiated HBEC in ALI from one COPD donor were placed in the Vitrocell smoke chamber filled with DMEM medium (Thermo Fisher Scientific) and CS or humidified air was puffed onto the apical surface. ISO Puff Standard was used with each CS exposure consisting of two 3R4F Kentucky research cigarettes (one 35 ml puff every 60 seconds). Cells were exposed to WCS four times during two days with a minimum of two hours elapsed between smoke exposures. Treatment with ACY-1083 started after the first smoke exposure. Permeability was measured 20 hours after the last smoke session.

Bacterial infection was carried out on fully differentiated HBEC ALI cultures from one COPD donor at three different occasions. Cells were pretreated with 10 μM ACY-1083 or 0.1% DMSO for nine days. Haemophilus influenzae, Pittman 576, type b (Culture Collection University of Gothenburg, Gothenburg, Sweden) was grown over night in Brain-heart infusion media (VWR, Radnor, PA, USA) at 37°C. When optical density (620 nm) was above one, bacteria were centrifuged at 4000 rcf for ten minutes and then washed in cold PBS once. Serial dilutions of bacteria were done in PBS starting at 2e8 CFU/ml. 50 μL of bacterial dilutions were added to the apical side of the ALI cultures. After two hours incubation at 37°C, apical surfaces were washed three times with PBS and paracellular permeability was measured 18hrs later.

Alcian Blue/Periodic Acid-Schiff staining

Eight days post-treatment, HBEC ALI-cultures were fixed in 4% paraformaldehyde for fifteen minutes and washed three times in PBS. Dehydration in ethanol and xylene followed by paraffin (Merck, New Jersey, USA) infiltration was done with a short program for biopsies (1 hour 55 minutes) on a Microm STP 120 Spin Tissue Processor (Thermo Fisher Scientific). Cell layers were embedded in paraffin and 4 μm thick sections were made using a Leica RM2165 microtome (Leica, Wetzlar, Germany). Alcian Blue/Periodic Acid-Schiff (AB/PAS) with Haematoxylin staining were performed using standard protocols on a Leica ST5020. After staining, the sections were dehydrated in ethanol and xylene, mounted with Pertex mounting medium (Histolab, Göteborg, Sweden) and scanned with a Aperio Scanscope slide scanner (Leica Biosystems, Buffalo Grove, USA).

Quantification of cellular features from IHC sections

IHC images were analyzed in HALO v3.1 (Indica Labs). For all analyses, images were annotated manually to exclude out of focus and damaged areas, and a random forest classifier was applied to intact areas of the sections to detect epithelial areas. All following analyses were done in epithelial areas only. For goblet cell counting, algorithm CytoNuclear 2.0.9 was used to segment cells based on nuclear staining and categorize them as AB/PAS (mucin) positive and negative cells based on cytoplasmic AB staining. Goblet cell counting was represented as the percentage of AB positive cells of total cells.

To measure epithelial thickness, the apical and basal sides of the epithelium in the sections were manually delineated using a pen annotation tool. Distance between paired apical and basal annotations was measured every 10 μm along the epithelium, and recorded in sequence as epithelial thickness.

To measure intraepithelial and intercellular AB staining, algorithm Area Quantification v2.1.7 was used. Two thresholds were applied to capture AB staining: a lower one to detect all AB staining in both goblet cells and intercellular areas, and a higher one to detect the stronger AB staining inside goblet cells only. Intercellular AB area was calculated by subtracting goblet AB staining (higher threshold) from total AB staining (by lower threshold).

Occludin staining

Eight days post-treatment, HBEC ALI-cultures were fixed in 4% paraformaldehyde for fifteen minutes and washed three times in PBS. Cells were permeabilized with 0.2% Triton X-100 in PBS for two hours and then incubated in 10% goat serum in PBS. Blocking buffer was replaced after one hour with 7 μg/mL mouse anti-occludin antibody (cat# 33–1500, Invitrogen) in PBS with 5% goat serum. After overnight incubation at 4°C, cells were washed three times in PBS and incubated with 2 μg/mL goat anti-mouse IgG Alexa Fluor 594 (Invitrogen) for one hour at room temperature. Cells were washed in PBS and nuclei stained with 2 μM Hoechst 33342 (Thermo Scientific) for 30 minutes followed by three PBS washes. Membranes were then cut from the inserts with a scalpel and mounted on glass slides in ProLong Gold Antifade Mountant (Thermo Fisher Scientific). Stained cell cultures were imaged using a Zeiss LSM 880 confocal microscope.

Western blot of membrane protein from ALI cell cultures

For extraction of membrane and cytosol proteins the cells were lysed according to the Mem-PER Plus Membrane Protein Kit protocol (#89842, ThermoFisher Scientific). The supernatants were kept frozen in −80°C until total protein quantification. Quantification of total protein in the lysates was done using Pierce BCA Protein Assay Kit (#23227, Thermo Scientific) following the manufacturer´s instructions. Lysates containing 8 μg total protein were mixed with NuPAGE LDS Sample buffer 4x (NP007, Invitrogen), NuPAGE Sample Reducing Agent 10x (NP004, Invitrogen) and deionized water and then heated for ten minutes at 70°C. The samples were loaded onto NuPAGE 4–12% Bis-Tris Gels (NP0335, Invitrogen) and run at 120V for 100 minutes in NuPAGE MOPS SDS Running buffer (NP0001, Invitrogen). The proteins were transferred to nitrocellulose membranes (LC2001, Invitrogen) at 35 mA overnight using NuPAGE Transfer buffer (NP0006, Invitrogen) supplemented with 20% methanol. The membranes were stained with Revert Total Protein Stain (926–11011, LI-COR Biosciences) for five minutes, washed and then imaged in the 700 nm channel using Odyssey CLX imaging system (LI-COR Biosciences). To correct for variation in loading the total signal intensity of all proteins in each lane were used for normalization. An inter-membrane control sample was loaded to all gels to assure there were no differences due to methodology between the blots. The stain was removed with Revert Destaining Solution (926–11013, LI-COR Biosciences) and the membranes were blocked in Intercept (TBS) Blocking Buffer (927–60001, LI-COR Biosciences) for one hour on a shaker. After blocking the membranes were incubated cold overnight with primary antibodies diluted in Intercept (TBS) Blocking Buffer (E-cadherin mAb #14472 Cell Signaling Technology 1:1000, β-catenin mAb #8480 Cell Signaling Technology 1:1000, Occludin mAb #33–1500 Invitrogen 1:500). On the next day the membranes were washed three times for ten minutes in Tris-buffered saline + 0.05% Tween (#91414, Merck) followed by incubation with IRDye Goat anti-Mouse 800CW (#926–32210, LI-COR Biosciences) and Donkey anti-Rabbit 680RD (#926–68073, LI-COR Biosciences) secondary antibodies (1:10000 dilution) for one hour at room temperature. After a final wash, three times for ten minutes in TBS-T, the fluorescent signals were detected using the Odyssey CLX imaging system and the signals were analysed using Image Studio software (v4.0, LI-COR Biosciences).

Mucociliary clearance

In the TNF (20 ng/ml) challenged cultures, the apical surfaces were washed with pre-warmed PBS the day before addition of 50 μL of CountBright Absolute Counting beads (Invitrogen, Carlsbad, CA, USA) diluted 1:10 000 in PBS with Mg2+ and Ca2+. Inserts were transferred to a 12-well glass bottom plate (Cellvis, Mountain View, USA) and monitored in a Zeiss LSM 880 (Carl Zeiss AG, Oberkochen, Germany) at 37°C. Three regions in each well of each condition in three COPD donors were filmed and analyzed. Beads were tracked in the ImageJ software with the Fiji plugin Tracking [55] and the mean velocity of each bead was plotted.

Ciliary beat frequency

CBF was quantified for the differentiated healthy HBECs treated with vehicle, TNF, and TNF + ACY-1083. The plates containing the pseudostratified epithelia were incubated at 37 °C with 5% CO2 in the 3i Marianis/Yokogawa Spinning-Disk Confocal microscope (Leica Microsystems, Sugar land, USA) as reported in our previous publications [53, 56, 57]. High-speed time-lapse videos were taken at 32X air at 100 Hz with a total of 250 frames using a scientific Hamamatsu C11440-42U30 CMOS camera (Bridgewater, USA). Five areas were imaged per insert. A Matlab (R2020a) script (validated against SAVA) (previously described in [58] was used to determine average CBF per video, to generate a heat map indicating CBF.

Cell velocity

Cell migration was quantified by performing Particle Image Velocimetry (PIVlab) on Matlab using multi-pass cross-correlation analysis with decreasing interrogation window size on image pairs to obtain the spatial velocity field as described previously [59]. Using a phase contrast microscopy of 3i Marianis Spinning-Disk Confocal microscope (Leica Microsystems) at 32X air objective, time-lapse videos of the epithelial cell monolayer were captured for every five minutes for two hours following treatment with TNF, and/or ACY-1083, and the average velocity for the area was computed.

Cytokine release

To assess cytokines released, basolateral supernatants were collected on day seven and immediately frozen at -80°C until MSD analysis of IL-6, CCL2 and CXCL10. In short, U-PLEX MSD plates (Mesoscale Discovery, Rockville, Maryland, USA) were coated, washed and stored overnight. Standard curves were prepared and added in duplicates on each plate. Supernatants were thawed on ice and added to the plates, which were incubated on an orbital shaker for one hour before sulfo-tagged antibodies were added and incubated for another hour. Plates were washed, read buffer added and read on MESO SECTOR S 600 (Mesoscale Discovery, Rockville, Maryland, USA). Data was plotted as fold change to DMSO control.

RNA analysis

To study transcriptional changes, RNA purification was carried out using the RNeasy Plus 96 kit (74192) (Qiagen, Venlo, The Netherlands) according to manufacturer’s protocol. The liquid was collected and frozen at –80°C until further analysis. RNA was quantified and cDNA synthesis was performed according to standard procedures using High-capacity cDNA Reverse Transcription Kit (4368813) (ThermoFisher, Waltham, MA, US). Real-time polymerase chain reaction (qPCR) was done using validated TaqMan® Gene Expression Assays (ThermoFisher, Waltham, MA, US), IL8 (Hs00174103_m1), CCL2 (Hs00234140_m1), CXCL10 (Hs00171042_m1), MMP9 (Hs00957562_m1), IL1B (Hs01555410_m1), MUC5AC (Hs01365616_m1), MUC5B (Hs00861595_m1) with TaqMan Fast Advanced Master Mix (4444557) (Applied Biosystems, Foster City, CA, US). Samples were run in triplicates and analyzed on the QuantStudioTM 7 Flex Real-Time PCR 384-well System (ThermoFisher Waltham, MA, US). Data was normalized to the housekeeping genes GAPDH (Hs99999905_m1), and RPLP0 (Hs99999902_m1) and delta-delta CT values were plotted (fold change to DMSO).

Statistical evaluation

Statistical differences between samples were assessed with two-tailed paired t-test, one-way or two-way analyses of variance (ANOVA). Differences at p-values below 0.05 are considered significant. All statistical analyses were performed using Graphpad PRISM 8.4.2 (GraphPad Software, Inc., La Jolla, CA). All statistics comparing data within a treatment group was made using two-way ANOVA with Sidak’s post test, and all comparisons between groups were made using two-way ANOVA with Dunnet’s post-test. Tukey’s test of variance was used for the thickness variation data analysed with R [60].

Results

ACY-1083 is highly selective against other HDACs

When using inhibitors as tools to validate a target, it is important to understand the selectivity towards other structurally similar targets. In the HDAC assay panel, HDAC6 inhibitors that are termed to be selective, such as Ricolinostat and Tubastatin A, have selectivity margins to class I HDACs of as low as ten-fold and even lower towards Class II HDACs. Results from using these compounds in preclinical settings, claiming to be HDAC6-driven, should thus be interpreted with caution. ACY-1083 is a highly selective inhibitor of HDAC6 with a selectivity profile demonstrating >400-fold selectivity towards HDAC5 and HDAC7, >1000-fold selectivity over HDAC1, 4, 8, 9, 10 and 11, and >7000-fold selectivity over HDAC2 and 3 (Table 1). Structures can be found in S1 Fig. ACY-1083 is stable in human plasma with a free fraction of 23%. A simple spike test of adding 10 μM ACY-1083 to ALI cell media in a test tube showed an approximate 4 μM drug concentration after 18 hrs. This indicates that more than half of the compound is lost in binding to the tube/cell media proteins/components. In the preclinical models used in these experiments, this indicates that at concentration ranges up to 10 μM, ACY-1083 should target mainly HDAC6 and no other HDACs (but possibly 5 and 7 at below their respective IC50). ACY-1083 has a reported cell potency (IC50) of 30–100 nM, measured as the inhibition of deacetylation of α-tubulin [49]. ACY-1083 further displayed excellent selectivity in a panel of >190 in vitro radioligand binding and enzyme assays covering a diverse set of enzymes, receptors, ion channels, and transporters. The only activity identified below 30 μM was cyclooxygenase 2 (COX2, IC50 = 6.48 μM), dopamine active transporter (DAT, IC50 = 0.37 μM) and Sigma-1 receptor (IC50 = 17 μM). Additionally, ACY-1083 had no activity below 30 μM in a panel of cardiovascular ion channels.

Table 1

HDAC enzyme activity for ACY-1083, Ricolinostat and Tubastatin A.

Compound		ACY-1083*	Ricolinostat**	Tubastatin A**
		IC50 μM
Class I	HDAC1	17.2	0.515	1.91
	HDAC2	70.7	1.78	9.86
	HDAC3	>87.5	0.954	12.4
	HDAC8	15.8	0.826	0.345
Class IIA	HDAC4	12.5	7.13	0.910
	HDAC5	4.17	2.81	0.821
	HDAC7	6.51	2.32	0.229
	HDAC9	19.3	62.2	0.416
Class IIB	HDAC6	<0.01	0.0215	0.064
	HDAC10	18.5	0.901	3.22
Class IV	HDAC11	18.2	2.30	2.67

Demonstrating μM potencies for the different subtypes, color-coded for margins to HDAC6 (<100 fold = red, 100–1000 fold = yellow, >1000 fold = green),

*mean of two separate measurements

**one measurement

ACY-1083 reduces paracellular permeability

To study epithelial barrier dysfunction in COPD, we used a 3D model system with primary human bronchial epithelial cells isolated from COPD donors. Cells were cultured at ALI until a pseudostratified epithelium was developed, consisting of ciliated cells, goblet cells and basal cells. These cells form tight junctions in vitro which will give rise to resistance over the membrane. Paracellular permeability of small polysaccharide molecules, such as FITC-dextran, is a way of measuring barrier integrity and has been used to capture changes in the epithelium.

ACY-1083 was added basolaterally to fully differentiated HBEC ALI cultures from four COPD donors at 10 μM for eight days and permeability was measured to assess barrier integrity. Passage of FITC-dextran molecules were significantly reduced with ACY-1083 in all four donors (p = 0.0023) (Fig 1), which indicates that ACY-1083 strengthens the epithelial barrier per se since no challenge was added.

Fig 1

ACY-1083 reduces paracellular permeability in unchallenged fully differentiated COPD HBEC ALI cultures.

Graph shows the fluorescence intensity of FITC-labelled dextran (4kDa) passed through cell layers to the basolateral side. Box plot with dots representing the average of three individual replicates for each of 4 COPD donors and whiskers showing min and max. Statistical analysis was performed using two-tailed paired t-test.

ACY-1083 reduced epithelial injury after multiple challenges

To mimic the epithelial injury seen in COPD, we developed a TNF-challenge model where cells were treated with TNF in the basolateral media, until a noticeable drop in resistance, and an increase in paracellular permeability was observed. 18hrs were chosen as time point where we saw a clear window in permeability. HBEC ALI cultures from four COPD donors were treated with 10 μM of ACY-1083 for eight days at the same time as TNF was added at different concentrations. After seven days, TEER and permeability were measured and showed that compound-treated wells had higher TEER at all concentrations of TNF (0, 5, 20 and 50 ng/mL), as well as lower permeability at 5, 20 and 50 ng/mL TNF as compared to the DMSO control (Fig 2A and 2B). ACY-1083 hence rescued the cells from TNF-induced epithelial barrier disruption.

Fig 2

ACY-1083 improves the epithelial integrity in fully differentiated HBEC ALI cultures challenged with different barrier disruptors.

(A) Resistance measurements after 7 days of TNF-challenge (0–50 ng/ml), or (B) FITC-Dextran measurement after 8 days of TNF challenge (0–50 ng/ml), with or without 10 μM ACY-1083. Box plot with dots representing the average of three replicates for each of 4 COPD donors and whiskers showing min and max. (C) FITC-Dextran measurement after 2 days of TGF-β-challenge (0–10 ng/ml) on cultures pretreated for 3 day with 10 μM ACY-1083 or vehicle. Box plot with dots representing the average of three replicates for each of 2 COPD donors and whiskers showing min and max. (D) FITC-Dextran measurement after 48 hours CSE-challenge (0–6%) on cultures pretreated with 10 μM ACY-1083 or vehicle for 3 days. Box plot with dots representing the average of three replicates for each of 2 COPD donors and whiskers showing min and max. (E) 4 exposures to whole cigarette smoke during 2 days. Treatment with 10 μM ACY-1083 or vehicle started after the first WCS exposure and permeability was measured 20 hours post last exposure. Box plot with dots representing two replicates of 1 COPD donor and whiskers showing min and max. (F) 2 hour incubation of serial diluted Haemophilus influenzae followed by wash and permeability measurement at 18 hours. Cultures were pre-treated with 10 μM ACY-1083 or vehicle for 8 days. Representative graph of 1 COPD donor run at three different occasions. Box plot with dots representing three replicates of 1 COPD donor. Statistical analysis was performed using two-way ANOVA with Sidak´s multiple comparisons test.

In this experiment it was obvious that there is variation in donor susceptibility to TNF challenge but TEER and permeability results were always in accordance. Thus, we simplified the readouts to only measure permeability in the experiments with additional challenges.

Since COPD is a disease with multiple causes to epithelial injury, we also wanted to investigate other insults such as CSE, TGF-β, direct smoke and bacterial infection. In a similar fashion, CSE, TGF-β, whole smoke and H. Influenzae were added in different concentrations to COPD ALI cell cultures, and paracellular permeability was measured (S2 Fig). Treatment with TGF-β increased permeability and ACY-1083 decreased it at all concentrations (0.4, 2 and 10 ng/ml) (Fig 2C). The same pattern was seen for CSE where permeability concentration dependent and treatment with ACY-1083 revealed a reduction in permeability at all concentrations (1, 3 and 6%) (Fig 2D). For WCS, the window was even greater and ACY-1083 could at all concentrations used except for the highest (1, 0.5 and 0.2 l/min dilution) decrease permeability (Fig 2E). For the bacterial challenge we saw a similar pattern where ACY-1083 reduced paracellular permeability (Fig 2F). In addition, in a pilot experiment to test if ACY-1083 had a bactericidal effect by itself, bacteria suspended in media were incubated for one hour with ACY-1083 revealed no effect on bacterial counts. For all challenges used, ACY-1083 decreased paracellular permeability (Fig 2B–2F). For WCS and bacterial challenge work, only one COPD donor was used for each experiment, the data must thus be considered exploratory despite the clear effect with ACY-1083. These data made us conclude that ACY-1083 reduced the epithelial injury seen after multiple challenges, all relevant to COPD.

Inflammatory cytokines and mucins were reduced by ACY-1083

To help protecting the lungs from infectious agents and to maintain host defense, epithelial cells are major producers of cytokines and chemokines. To study the inflammatory responses after TNF challenge, several cytokines/chemokines were analyzed from the basolateral supernatants. IL-6, CCL2, and CXCL10 increased dose-dependently by TNF (S3 Fig). Induction of all three cytokines were reduced after simultaneous treatment with ACY-1083 at 5, 20 and 50 ng/ml TNF, indicating that ACY-1083 has a protective effect on the epithelium (Fig 3A–3C).

Fig 3

ACY-1083 reduces pro-inflammatory cytokines and mucins in TNF-challenged HBEC ALI cultures.

Protein concentrations in basolateral supernatants of (A) IL-6, (B) CCL2 and (C) CXCL10 in basolateral supernatants after 7 days of TNF-challenge and ACY-1083/vehicle treatment measured by MSD. Relative mRNA levels of (D) IL6, (E) CCL2, (F) CXCL10, (G) MMP9, (H) IL1B, (I) MUC5AC and (J) MUC5B were assessed by RT-PCR from cells lysed after 8 days of TNF-challenge and ACY-1083/vehicle treatment. Box plot with dots representing the average of three replicates for each of 4 COPD donors and whiskers showing min and max. Statistical analysis was performed using two-way ANOVA with Sidak´s multiple comparisons test.

Upon visual examination in a microscope, the wells treated with ACY-1083 looked more similar to the non-TNF control with a brighter apical layer. We therefore proceeded with RNA analysis of the cell cultures to further investigate mucins as well as different cytokines. TNF increased CCL2, CXCL10, MMP9, IL1B and reduced the mucin MUC5AC (S3 Fig). Treatment with ACY-1083 significantly reduced the transcripts of IL6, CXCL10, MMP9 and IL1B (Fig 3D and 3F–3H) at high concentrations of TNF. ACY-1083 also reduced baseline production of MUC5AC and MUC5B in unchallenged cells (Fig 3I and 3J) as well as at 5 ng/ml TNF but had no significant effect at higher concentrations of TNF.

ACY-1083 protects from morphological changes during epithelial injury

After having established that there was significant inflammation present after TNF-challenge, in addition to the increased permeability, we also wanted to investigate whether the damage of the epithelium had been substantial enough to give morphological changes. Cell cultures were fixed for immunohistology and sections were stained with hematoxylin as well as AB/PAS. Visual assessment of the morphology of no-TNF DMSO-treated cultures showed a pseudostratified epithelium with basal cells at the membrane and a mix of ciliated and goblet cells on the apical side (Fig 4A, left panel). TNF-treated cultures displayed in a dose-dependent fashion, a damaged epithelium with holes, mucus spread within the cell layers, an uneven apical side with patchy spots of ciliated cells and an unorganized basal cell layer (Fig 4A, left panel). Therefore, we concluded that the challenge was severe enough to cause visual epithelial injury. Examining cell cultures treated with ACY-1083 during TNF-challenge showed less damage, intact goblet cells, an even apical surface and a more organized epithelium (Fig 4A, right panel). Sections from all four donors are shown in S4 Fig.

Fig 4

ACY-1083 protects COPD HBEC ALI cultures from TNF-induced damage.

(A) AB/PAS staining of sections of ALI cultures show pseudostratified epithelium with goblet cells stained in strong blue. Quantification of (B) goblet cell numbers, (C) intercellular mucus within the epithelium and (D) thickness variation. Box plot with whiskers representing min and max for B and C, 10–90 percentile for D, for each of 4 COPD donors. Statistical analysis was performed using one-way ANOVA with Dunnett´s multiple comparisons test.

To be able to get an unbiased measurement of the morphological changes seen by eye, we developed a pipeline that was able to discriminate between goblet cells stained with AB, and other cells. After TNF challenge (20 ng/ml), the number of goblet cells in relation to the total number of cells in the sections were decreased as compared to the no-TNF DMSO control (p = 0.0111) (Fig 4B). Interestingly, AB staining was observed in between the cells in TNF-challenged epithelium, and quantification also showed significant increase in this intraepithelial and intercellular areas (p<0.0001) (Fig 4C). After treatment with ACY-1083, the percentage of goblet cells increased but did not differ significantly from the no-TNF DMSO control or the TNF DMSO culture (Fig 4B), and the intracellular AB area was reduced to the no-TNF level (p<0.0001) (Fig 4C). No difference in goblet cell percentage was detected between samples treated with only ACY-1083 or no-TNF DMSO (Fig 4B). Looking at individual donors, three out of four donors had a reduced percentage of goblet cells with TNF DMSO stimulation, which was reverted by ACY-1083 treatment, whereas one donor did not show this pattern (S5 Fig). We therefore concluded that treatment with ACY-1083 partly prevents the change in goblet cell count and protects cells from the intraepithelial and intracellular AB staining seen during TNF challenge, most likely due to a reduced damage of the epithelium.

The apical surface was another striking observation from the IHC sections, where ACY-1083-treated cell cultures appeared more even and with a maintained tight epithelium. In a similar fashion, we developed a pipeline that detected a drop in height at the apical surface and logged the difference in total height change. Comparing DMSO with TNF challenge in four donors, there was an increase in thickness variation (p<0.001). After treatment with ACY-1083 in TNF-challenged wells, the variance in height was smaller in the ACY-1083-treated cell cultures (p<0.001) indicating a rescue of the uneven epithelial surface caused by TNF challenge (Fig 4D). Despite being exploratory, this could also clearly be seen in HBEC ALI cultures stained with occludin and imaged with confocal microscopy. The epithelial cell layer looked more even and tighter after ACY-1083 treatment, but was not quantified so can only be considered as descriptive (S6 Fig).

Tight junction protein levels are unchanged in membrane fractions after ACY-1083 treatment

After establishing that there were quantifiable changes of the morphology, we analysed the levels of E-cadherin, β-catenin and Occludin in membrane protein extractions from four COPD donors from ALI cell cultures treated with TNF and ACY-1083. TNF treatment did not seem to affect the membrane levels noteworthy of these three proteins, and at 10 μM ACY-1083 the expression was not significantly higher even though a trend could be observed (Fig 5A–5C).

Fig 5

ACY-1083 does not increase tight-junction proteins in TNF-challenged HBEC ALI cultures.

Western blot analysis of membrane proteins normalized to DMSO control without TNF (dotted line), (A) E-cadherin (B) β-catenin (C) Occludin. Box plot with dots representing 4 COPD donors and whiskers showing min and max. Statistical analysis was performed using one-way ANOVA with Dunnett’s multiple comparisons test.

ACY-1083 protects mucociliary clearance and cellular velocity during TNF challenge

Since treatment with ACY-1083 protected against TNF-induced morphological changes, we next studied cilia function. At day ten, fluorescent beads were added on the apical side and tracked with confocal microscopy and the velocity of the beads were analysed and used as a measure of mucociliary clearance (Fig 6A). Quantification showed that TNF drastically reduced the bead movement as compared to the no-TNF control (Fig 6B). ACY-1083 treatment significantly protected from reduction in speed observed after TNF challenge (p = 0.0026) (Fig 6B).

Fig 6

ACY-1083 maintains mucociliary clearance during TNF-challenge.

Mucociliary transport rates were determined by tracking fluorescent microspheres placed on top of cell layer of unchallenged or TNF-challenged HBEC ALI cultures with or without ACY-1083 treatment. (A) Images of tracked beads in the ImageJ software with the Fiji plugin Tracking. Lines show movement of beads. (B) Microsphere speeds determined from 3 COPD donors and three fields per well. Box plot with dots representing 3 COPD donors and whiskers showing min and max. Data was normalized to the DMSO-control without TNF for each donor (set to 100%). Statistical analysis was performed using one-way ANOVA with Dunnett´s multiple comparisons test.

We further evaluated the effect of TNF on the airway epithelial plasticity phenotypes by measuring the barrier function, CBF, and cellular velocity. As previously seen, TNF decreased TEER (p = 0.0023) and ACY-1083 reversed this decrease (p = 0.0187) (Fig 7A). As for CBF, there might be a slight increase after TNF treatment but ACY-1083 had no effect on this measurement (Fig 7B). TNF increased cellular velocity (p = 0.0241) and interestingly ACY-1083 could counteract this increase (0.0411) bringing it down to unchallenged levels (Fig 7C).

Fig 7

ACY-1083 preserves epithelial plasticity phenotypes induced by TNF.

Epithelial plasticity phenotypes were determined by measuring (A) TEER, (B) CBF and (C) Cellular velocity of unchallenged and TNF-challenged healthy HBEC at ALI cultures with and without ACY-1083 treatment. Box plot with dots representing the average of triplicates for 3 healthy HBEC donors (unchallenged and TNF + 0.1% DMSO) and the average of triplicates for 2 donors (TNF + ACY-1083) and whiskers showing min and max. Statistical analysis was performed using one-way ANOVA with Dunnett´s multiple comparisons test.

Discussion

Airway epithelial barrier dysfunction, such as increased permeability and morphological alterations, has been demonstrated to be present early in COPD progression and their lungs show increasing destruction and inflammation. In this paper we have sought to investigate the ability to protect and restore the airway epithelial barrier function by using ACY-1083, an HDAC6 inhibitor with superior selectivity over other HDACs, in COPD preclinical model systems. We have used disease-relevant stimuli, such as TNF, TGF-β, CS and bacterial challenge on COPD primary epithelial cells, to establish in vitro challenge systems displaying signs of destruction of the epithelial barrier. ACY-1083 dramatically protects the airway epithelial structures and functions in vitro which indicate a potential to protect and restore the epithelial barrier, and thus disrupt disease progression, in COPD.

Tight junctions and adherence junctions are forming the barrier that is important for host defense and maintenance of normal epithelial functions. A first sign of an affected epithelium is that the barrier breaks and starts leaking with downstream effects being remodeling and inflammation. Epithelial injury has been discussed as a driver of disease progression and defects in repair mechanisms in lung diseases have been established [61]. Based on this, we set out to study barrier dysfunction in our cell models of COPD-relevant epithelial injury. By challenging ALI cell cultures derived from COPD BECs with TNF, we observed a dose-dependent increase in destruction as demonstrated by changes in cell culture morphology (IHC), increased permeability, inflammatory cytokines, and effects on ciliary function. In these models, ACY-1083 maintained the epithelial integrity at almost unchallenged levels which made us speculate that ACY-1083 treatment must have effects beyond permeability. When analyzing cytokine and protein responses there was a clear downregulation of inflammation as well as mucin production, with cell morphology being markedly protected from damage. To try to understand the exact mechanism by which ACY-1083 is having its protective effects, we analyzed membrane fractions of the ALI cell cultures to see if key molecules in the adherence and tight junctions were altered by ACY-1083 treatment. Surprisingly, the effects on E-cadherin, β-catenin and occludin were modest. Exploratory imaging of occludin staining reveals what looks like a more localized and sharp image as compared to the TNF-treated control, but due to method limitations no quantification was performed to establish this. There have however been reports that β-catenin and other tight junction markers have been increased at the membrane after HDAC6 inhibition [35, 37, 62]. Even though there might be a role for HDAC6 in tight junctions and epithelial permeability, and β-catenin being one of its substrates, we cannot conclusively determine the mechanism for the epithelial protective features of ACY-1083 that we are observing.

A consequence of epithelial injury can be remodeling and EMT, a feature that is characterized by loss of epithelial markers such as E-cadherin and upregulation of mesenchymal markers. This could lead to loss of ciliated cells which in turn would affect the clearance of mucus and pathogens. It is established that COPD patients have increased remodeling and production of mucins, shorter cilia and alterations in cilia-related genes [28]. The ciliated cells in the TNF-challenged ALI cell cultures looked unaffected, but when analysing bead movement as a model of mucociliary clearance capacity, ACY-1083 significantly increased the essentially eradicated ability to displace mucus/beads observed with TNF challenge. In line with these findings, a few papers have shown protective effects of HDAC6 inhibition on cilia destruction after challenges both in vitro and in vivo [33, 34]. However, we believe that the effects that we are observing are likely due to reduced damage and intact cells rather than a direct effect on the cilia, despite the described function of HDAC6 in regulating the cilia cell cycle [63]. This is also supported by the fact that CBF was not changed after TNF challenge and ACY-1083 did not alter this. Another outcome of injury and loss of epithelial features such as E-cadherin is that cells start moving and become less tightly attached to one another. When measuring cell velocity in our models, we observed that ACY-1083 completely normalized the movement triggered by the TNF challenge. If this is due to the fact that E-cadherin and other tight junction markers were trending towards an increase in the membrane fractions can only be speculated. Potentially the observed reduced cytokine secretion which could impact the overall inflammatory milieu in the media, is impacting this. In conclusion, we believe that in this model of epithelial injury, we had no effect on cilias per se and no potential effect by ACY-1083 could be observed, and rather propose that epithelial features and morphology were protected from injury after treatment with ACY-1083.

Establishing the role for a target by means of pharmacological evaluation is indeed sensitive as there may be unknown secondary target effects. We have evaluated the secondary pharmacology binding and enzymatic panels of ACY-1083 in both a panel of HDAC isoforms and in secondary pharmacology safety panels with only 4 hits below 10 μM (HDAC5, HDAC7, COX2 and DAT). Moreover, we have tried to make sure we do not use higher concentrations than needed with regards to full inhibition at HDAC6. A 10-fold IC50 (i.e. nearby full inhibition) from a cell assay would indicate full HDAC6 blockade at 0.3–1 μM [49], but more complicated cell systems including membranes and equilibration of two compartments (basolateral and apical) as in the ALI system, require approximately 10-fold higher concentrations from previous experience. Moreover, the ACY-1083 concentrations in this work (often 10 μM) are the theoretical final basolateral concentrations after addition of the drug to the basolateral compartment. However, we conducted a concentration evaluation of ACY-1083 and observed that approximately 40% is remaining after a simple spike test, ie compound does get lost in the complication of the system (to plastic and media components). Taken together, we are not expecting any secondary pharmacological effects at approximately 3–10 μM concentrations of ACY-1083 from the secondary pharmacology targets tested. Safety aspects of HDAC6 inhibition locally in the lung and literature on HDAC6 KO mice suggest this is a safe target [64, 65] and pharmacological intervention where the catalytic effects are inhibited should be no different.

As with all in vitro work, our study has limitations. We have a limited number of donors for our experiments, and some are also merely to be considered exploratory. Despite this fact, our results are very convincing and ACY-1083 succeed in protecting the epithelium from injury. The underlying mechanism for this is also something that we have tried to address by having a broad and diverse set of functional assays, but it is unclear to us exactly how ACY-1083 has this protecting effect. To better understand this we believe that more molecular approaches may be helpful, such as next generation sequencing, as well as proteomics, where pathways and acetylation of proteins can be investigated. Since some aspects of our results could not confirm published findings, eg the cilia involvement described for other HDAC6 inhibitors, we also suggest a more targeted approach eg direct knock-out of HDAC6 in relevant human primary cells to validate that this has the expected effect. Lastly, this study does not involve any in vivo work. As with other targets, mechanistic understanding is sometimes easier to establish in vivo, as are the effects of an inhibitor such as ACY-1083, and this could be addressed in future work.

In conclusion, our study suggests that ACY-1083 has the potential to protect the lung epithelium from damage caused by external toxins such as cigarette smoke and pollution, as well as reduce the induction of cytokines induced by such challenges. This implies promise for future treatment opportunities in respiratory disease, which may even be disease modifying, such as in early COPD. As the airway epithelium is “at the surface”, local therapeutic options by means of inhalation to restore airway epithelial integrity and function with minor systemic exposure further increase the attractiveness of intervening at the barrier.

Structures of compounds used.

(TIF)

Click here for additional data file.

Different barrier disruptors reduce the epithelial integrity in fully differentiated HBEC ALI cultures.

(A) Resistance measurements after 7 days of TNF-challenge (0–50 ng/ml). Box plot with dots representing the average of three replicates for each of 4 COPD donors and whiskers showing min and max. (B) FITC-Dextran measurement after 8 days of TNF challenge (0–50 ng/ml). Box plot with dots representing the average of three replicates for each of 4 COPD donors and whiskers showing min and max. (C) FITC-Dextran measurement after 2 days of TGF-β-challenge (0–10 ng/ml). Box plot with dots representing the average of three replicates for each of 2 COPD donors and whiskers showing min and max. (D) FITC-Dextran measurement after 48 hours CSE-challenge (0–6%). Box plot with dots representing the average of three replicates for each of 2 COPD donors and whiskers showing min and max. Statistical analysis was performed using one-way ANOVA with Dunnett´s multiple comparisons test.

(TIF)

Click here for additional data file.

TNF increases pro-inflammatory cytokines and decreases mucins in HBEC ALI cultures.

Protein concentrations in basolateral supernatants of (A) IL-6, (B) CCL2 and (C) CXCL10 in basolateral supernatants after 7 days of TNF-challenge measured by MSD. Relative mRNA levels of (D) IL6, (E) CCL2, (F) CXCL10, (G) MMP9, (H) IL1B, (I) MUC5AC and (J) MUC5B were assessed by RT-PCR from cells lysed after 8 days of TNF-challenge. Box plot with dots representing the average of three replicates for each of 4 COPD donors and whiskers showing min and max. Statistical analysis was performed using one-way ANOVA with Dunnett´s multiple comparisons test.

(TIF)

Click here for additional data file.

AB/PAS staining of sections of ALI cultures from 4 COPD donors challenged with different concentrations of TNF (0–50 ng/ml) with and without 10 μM ACY-1083.

(TIF)

Click here for additional data file.

Goblet cells are protected from TNF challenge by ACY-1083 in HBEC ALI cultures in three out of four donors.

Quantification of goblet cell numbers in HBEC ALI sections stained with AB/PAS from 4 COPD donors.

(TIF)

Click here for additional data file.

Occludin staining using confocal microscopy on ALI cell cultures treated with ACY-1083 during TNF challenge.

Occludin (in red) and nuclei (in blue) staining of HBEC ALI cultures from 1 COPD donor challenged with 20 ng/ml TNF. Images showing intersection of cell layer treated with A) vehicle or B) 10 μM ACY-1083. Occludin staining from the same sections as in A-B, C) showing vehicle and D) 10 μM ACY-1083.

(TIF)

Click here for additional data file.

(PDF)

Click here for additional data file.

(PDF)

Click here for additional data file.

2 Jun 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Yulia Komarova

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Competing Interests section:

I have read the journal's policy and the authors of this manuscript have the following competing interests: AstraZeneca employees may hold shares in the company.

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript by Horndahl and colleagues demonstrated that ACY-1083, a selective HDC6 inhibitor, showed protection of the respiratory epithelium treated with TNFa, TGFb, cigarette smoke and Haemophilus infection in air liquid interface cultured COPD bronchial epithelial cells. Thus, an HDAC6 inhibitor has potentials to restore structure or function impaired in COPD. This study is organised very well. The strengths of this study are 1)used air liquid interface cultured epithelium rather than submerged monolayer cells, 2)full characterization of ACY-1083 (primary and secondary pharmacology). 3)used several relevant stumulats 4)used many readout including TEER, clearance, cilia beat etc. Findings are interesting and convincing, and the quality of experiments is at high standard.

However, I have several concerns and questions, and hope that authors are able to address these issues.

1) Major concern is a lack of patient information. What kind of COPD patients are used? Please give more details of donors, including gender, age, COPD severity, lung function.

2) ALI culture should be treated as an individual, and I saw some experiments were conducted using only one donor. The data from one donor is not representative. Please increased number of donors, for example bacteria work.

3) I still feel 10µM is high for the work. I understand authored mentioned that more complicated cell systems including membranes and equilibration of two compartments (basolateral and apical) as in the ALI system, require approximately 10-fold higher concentrations from previous experience. What kind of experience? How can you generalise your finding (x10-fold")? Please check exposure level in ALI epithelium after treatment.

4) Authors used only COPD ALI, but to understand the finding is specific to COPD or not, a small study with healthy ALI (if possible asthma ALI) is suggested.

Reviewer #2: 1. Reviewer report

2022.05.11

Recommendation:

Accept with moderate changes

Comments to the author:

Manuscript number: PONE-D-22-07956

Title: HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

General comment and summary:

The manuscript presented by Horndahl et al addresses the role of HDAC6 in airway epithelial damage in the context of COPD. The authors motivate for the use of ACY-1083, an inhibitor with high selectivity for HDAC6, and characterized its effects on epithelial function including epithelial disruption, cytokine production, remodeling, mucociliary clearance and cell characteristics. I think the manuscript is well-presented but could be improved by some mild language editing and the moderate edits I suggest below.

Main comments:

1. General commentary, the figure quality needs to be improved as it was exceptionally difficult to assess the manuscript.

2. L144- ‘…five donors of primary healthy HBECs…’ why did you decide on 5 donors per group? Was any a priori sample calculation conducted? Please could you clarify the number of donors per experiment using n’s in each figure legend.

3. L249- You mention the antibodies that were used in these studies. However, concerningly, you do not mention or show any membrane control such as GAPDH. Was there a reason for this? Could you also explain the non-specific bands that appear on the blots (below 50 kDa)?

4. L343- Is there any information regarding the classification of these donor cells? What COPD GOLD stage were these donors?

5. Whilst the epithelial ex vivo assays utilized in this manuscript are well-conducted, what translational value do these data have given the absence of an in vivo model?

6. L375- Nicely planned experiments with relevant perturbations utilized in the probing of this model. How was the MOI of H. Influenzae chosen? Is this clinically relevant?

7. L402- ‘…permeability measurement at 18 hours.’ Why was 18h the time point chosen for permeability measurements. Please clarify this in the manuscript.

8. L480- You would certainly expect a change in these epithelial-associated proteins. I would recommend checking mRNA levels to ensure that the model is producing the desired molecular changes.

9. L495- ‘At day ten, fluorescent beads were added…’ I really enjoyed the thought process behind this assay, however, I have some questions about the translational utility. Were the beads similar to mucous size? Were the beads coated in anything to mimic mucus?

10. L533- ‘…we could mimic the inflammatory milieu…’ I don’t think that this is accurate. The inflammatory milieu consists of immune cells and a number of cytokines not included in these assays. I think this statement should be removed.

11. L564- ‘…cell cultures looked unaffected…’ Why did your cells not express decreased cilia length? This is obviously a crucial proponent of mucociliary clearance.

12. A more detailed description of the limitations and future directions of this study should be included.

Minor comments:

1. Include the molecular weight of the compounds in the figure S1

2. Figure S2. It appears that there are only two measurements in C and D? Is this correct?

3. Figure S5. No statistics have been conducted. Please update this. There also appears to be very little difference in donor 3-could you comment on this?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Jul 2022

I have now provided a file named "Response to Reviewers" where I have addressed all comments made by the editor and Reviewers.

These are also copied in below. Please let me know if you have any other inquiries. Best wishes and thank you, Mia Collins

ANSWERS TO REVIEWER’S COMMENTS

PONE-D-22-07956

HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

PLOS ONE

Editors’ summary & main concerns:

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The authors should provide following information to address the Reviewers’ concerns:

1. include additional details related to the donor sample including gender, age, COPD severity, lung function.

Thank you for this comment. This is relevant information but all our donors are purchased from Lonza and we have very limited information. We have added age, gender and smoking history on these donors in the M&M on line 134 , detailed answer given in the response to Reviewer #1’s first comment below.

2. provide details related to the power analysis used to determine the sample size, the number of donors per experiment, type of the statistical analyses

Thank you for these comments as they are indeed relevant. We believe that we have addressed this in a detailed fashion below (in response to Reviewer #2, second comment)

3. include the loading controls for the western blotting assays

This has now been clarified below (Reviewer #2, comment 3).

4. provide additional details on the methods

Thank you – We hope that with Reviewer #1 and Reviewer #2’s careful assessments of the manuscript and our responses below, that we have provided sufficient details in the M&M section.

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript by Horndahl and colleagues demonstrated that ACY-1083, a selective HDC6 inhibitor, showed protection of the respiratory epithelium treated with TNFa, TGFb, cigarette smoke and Haemophilus infection in air liquid interface cultured COPD bronchial epithelial cells. Thus, an HDAC6 inhibitor has potentials to restore structure or function impaired in COPD. This study is organised very well. The strengths of this study are 1)used air liquid interface cultured epithelium rather than submerged monolayer cells, 2)full characterization of ACY-1083 (primary and secondary pharmacology). 3)used several relevant stumulats 4)used many readout including TEER, clearance, cilia beat etc. Findings are interesting and convincing, and the quality of experiments is at high standard.

However, I have several concerns and questions, and hope that authors are able to address these issues.

more details of donors, including gender, age, COPD severity, lung function.

Thank you for this important remark. The donors used, both the COPD and healthy bronchial epithelial cells are originating from epithelial cells purchased from Lonza. We have very limited information about these but we have included age, gender and smoking history for these donors. We have made a new section starting on line 134, now reading:

Patient characteristics from donor cells

Primary COPD human bronchial epithelial cells (HBEC) were purchased from Lonza (Basel, Switzerland). We have used cells from four COPD donors and the information on these donors is limited but include age (years), gender (F/M) and smoking history (yes/no): donor 1 (Cat#00195275, Batch#0000630389, 54, F, yes), donor 2 (Cat#00195275S, Batch 0000370751, 65, F, yes), donor 3 (Cat#00195275S, Batch 0000436083, 59, M, yes), donor 4 (Cat#00195275S, Batch 0000430905, 66, M, yes). Additional five donors of primary healthy HBECs were purchased from Lonza and Epithelix SàRL (Geneva, Switzerland): donor 1 (Cat# hAECB, Batch# AB068001, 71, F, no), donor 2 ( CC-2540, Batch# 0000501935, 56, M, yes), donor 3 (CC-2540S, Batch# 0000619260, 65, F, yes), donor 4 (hAECB, Batch# AB079301, 62, M, no) and donor 5 (hAECB, Batch# #AB077201, 63, M, no).

In doing so, we have also changed figure S5 so that the donors have the same numbering as stated above. This means that it will be donor 1 that does not display the same pattern and response to TNF as the other donors, not donor 3 as in the answer below (Reviewer #2, minor comment 3).

2) ALI culture should be treated as an individual, and I saw some experiments were conducted using only one donor. The data from one donor is not representative. Please increased number of donors, for example bacteria work.

Thank you for this comment and this observation. It is true that in two experiments we only have one donor, especially with regards to the exploratory experimentations done to explain findings from the planned investigation.

The first one is whole cigarette smoke and the bacteria work. For the bacteria work we repeated the same donor three times and we found tight and convincing data. The whole cigarette smoke experiment has also been repeated several times but due to methodology, it is hard to get a smoke dose where you get sufficient damage to the epithelium, but the cells don’t die. However, with our long experience with ACY-1083, we found it interesting to show the data, despite lack of statistical analysis. Since the bacteria work is part of the test with several stimulations where we see the same effect of ACY-1083, we thought that this work was still worth including, and that it would strengthen the findings, despite being exploratory. We have highlighted “exploratory” in the manuscript, with the sentence at line 403.

For WCS and bacterial challenge work, only one COPD donor was used for each experiment, the data must thus be considered exploratory despite the clear effect with ACY-1083.

The second experiment with n=1 donor is the confocal imaging on the sections from the ALI cultures (S6 Fig). We state that we don’t have the methods to quantify this, but still thought the stainings were interesting to show, as representative for the more even and tight epithelium.

These two experiments are considered exploratory investigations but we are open to removing the data, if preferred by editor/reviewers. Unfortunately we don’t have the possibility to build on this data set and perform more experiments for proper statistical evaluation. We have also here clarified that these data are descriptive and exploratory at line 486:

Despite being exploratory, this could also clearly be seen in HBEC ALI cultures stained with occludin and imaged with confocal microscopy. The epithelial cell layer looked more even and tighter after ACY-1083 treatment, but was not quantified so can only be considered as descriptive (S6 Fig).

And on line 573 the word “exploratory” has been added and the sentence now reads:

Exploratory imaging of occludin staining reveals what looks like a more localized and sharp image as compared to the TNF-treated control, but due to method limitations no quantification was performed to establish this.

3) I still feel 10µM is high for the work. I understand authored mentioned that more complicated cell systems including membranes and equilibration of two compartments (basolateral and apical) as in the ALI system, require approximately 10-fold higher concentrations from previous experience. What kind of experience? How can you generalise your finding (x10-fold")? Please check exposure level in ALI epithelium after treatment.

This is a great comment and we agree with the reviewers’ concern. We did measure the concentration as stated in the manuscript, and our results indicate that in this case 60% of the compound was lost in the system (line 343). We have long experience with ALI and we have run multiple drug projects using this model. We believe that the ‘loss’ in expected/added drug concentration is due to binding of the drug to the tissue, the plastic in the plate, and proteins in the media that form a surface for the compound to bind to. This is dependent on the physiochemical properties of the drug. We have data indicating that at 10 µM, the only other HDACs that we could possibly be hitting are HDAC5 and 7. As with all pharmacological inhibition, the secondary pharmacology is hard to predict or outrule. ACY-1083 has physiochemical characteristics that should not enrich it inside the cell, but rather equilibrate over the cell membrane and that the measured 4µM after adding 10µM should be what it is also inside the cell. Something that has also been discussed is the use of challenge models in vitro. To demonstrate that you have a drug that can rescue cells from injury over 1-2 weeks, whereas in real life this kind of damage may take years to develop, may also call for higher concentrations of compound.

4) Authors used only COPD ALI, but to understand the finding is specific to COPD or not, a small study with healthy ALI (if possible asthma ALI) is suggested.

Thank you for this comment and it is a very good suggestion. When we first started using ACY-1083 in epithelial biology, we started with healthy bronchial epithelial cells and we are confident that these findings are not only specific to COPD ALI, but for all kinds of ALI, and cells that form tight junctions. Since this was part of a drug project and we wanted to use as “disease relevant” model systems as possible, we only used COPD ALI for our complete set of experiments. The reason for this is that we discovered that when titrating permeability with different stimulations to validate this model, there were differences in the response to eg TNF. COPD cells were more sensitive and the duration of treatment to achieve a good window for permeability was different versus healthy ALI. Also, our aim was to test the concept and ACY-1083 on COPD lungs, hence we found them more relevant to use for our purposes. But the response to ACY-1083 was always the same, it decreased permeability and the other features that we present. In figure 7 we show cilia characteristics and these experiments are performed on healthy ALI where ACY-1083 shows the expected effect.

Reviewer #2: 1. Reviewer report

2022.05.11

Recommendation:

Accept with moderate changes

Comments to the author:

Manuscript number: PONE-D-22-07956

Title: HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

General comment and summary:

The manuscript presented by Horndahl et al addresses the role of HDAC6 in airway epithelial damage in the context of COPD. The authors motivate for the use of ACY-1083, an inhibitor with high selectivity for HDAC6, and characterized its effects on epithelial function including epithelial disruption, cytokine production, remodeling, mucociliary clearance and cell characteristics. I think the manuscript is well-presented but could be improved by some mild language editing and the moderate edits I suggest below.

Main comments:

1. General commentary, the figure quality needs to be improved as it was exceptionally difficult to assess the manuscript.

Thank you for pointing this out. We have corrected this issue.

2. L144- ‘…five donors of primary healthy HBECs…’ why did you decide on 5 donors per group? Was any a priori sample calculation conducted? Please could you clarify the number of donors per experiment using n’s in each figure legend.

We have not done any statistical evaluation before the start of the experiments, but have selected sample size based upon our experience of variability in the model. We believe that clinically relevant differences should be able to be seen in a limited number of patients and wells per donor. Even though it would be preferable to have more donors, we see that the results are convincing and statistically significant at this sample size.

We have added numbers of patients consistently instead of writing with letter and hope that this will faciliate to easily read this.

3. L249- You mention the antibodies that were used in these studies. However, concerningly, you do not mention or show any membrane control such as GAPDH. Was there a reason for this? Could you also explain the non-specific bands that appear on the blots (below 50 kDa)?

Thank you for pointing this out, it is a mistake on our part that we don’t explain this further. We have used total protein for normalization instead of eg GAPDH. The reason for this is that that it was hard to find a protein that was present in the membrane fractions.

We have added an explanatory sentence on line 260:

To correct for variation in loading the total signal intensity of all proteins in each lane were used for normalization. An inter-membrane control sample was loaded to all gels to assure there were no differences due to methodology between the blots.

As for the unspecific bands, unfortunately we don’t know what these are.

We will also upload a separate PDF with the total protein staining.

4. L343- Is there any information regarding the classification of these donor cells? What COPD GOLD stage were these donors?

Thank you for this important remark. Reviewer 1 has asked the same thing and we write: The donors used, both the COPD and healthy bronchial epithelial cells are originating from cells purchased from Lonza. We have very limited information about these but we have included age, gender and smoking history for these donors on line 134.

5. Whilst the epithelial ex vivo assays utilized in this manuscript are well-conducted, what translational value do these data have given the absence of an in vivo model?

This is a very good remark. Since there is no preclinical aspects of this model present, eg immune cells of the lung, fibroblasts and extracellular matrix or blood flow, the translational value cannot be determined. But we have long experience with in vivo models and it is very hard to claim that we have animal models that can mimic human disease. Even with an in vivo model, it wouldn’t give us better confidence that it would work in clinical practice. Our concern with potential mouse models is that HDAC6 lacks the cytoplasm-anchoring motif that makes it enter the nucleus and there also de-acetylates histones.

6. L375- Nicely planned experiments with relevant perturbations utilized in the probing of this model. How was the MOI of H. Influenzae chosen? Is this clinically relevant?

Thank you for this question. We titrated the dose of bacteria to where we would have a reduction in permeability, as well as having an obvious epithelial injury. The MOI of bacteria was chosen broadly from 2e2 to 2e8, unfortunately we are not sure of the clinical relevance of the doses, but we do think that the assay per se is still relevant.

7. L402- ‘…permeability measurement at 18 hours.’ Why was 18h the time point chosen for permeability measurements. Please clarify this in the manuscript.

Thank you for this comment. We titrated this carefully and shorter time points gave too small window. 18h were chosen for optimal and practical reasons and we wanted to keep it consistent between assays with eg TNF. A sentence has been added in the M&M at line 170 reading:

The time point was chosen for optimal window in the assays as well as practical reasons and has been kept consistent in between different stimulations throughout this work.

to address this uncertainty. As well as a clarifying the time point on line 208, and in the results section at line 381:

18hrs were chosen as time point where we saw a clear window in permeability.

8. L480- You would certainly expect a change in these epithelial-associated proteins. I would recommend checking mRNA levels to ensure that the model is producing the desired molecular changes.

Thank you for this suggestion. We have checked many different genes, including tight junction genes, but we have found it very hard to catch meaningful and reproducible changes. We believe that it is due to the fact that we have a chronic model where the injury happens over time and TNF is added every other day where each donor is responding a bit differently. Over time this builds up to the changes that we present in the manuscript, but as stated, it is hard to determine the exact mechanism/-s and time course for the effects we are seeing with ACY-1083.

9. L495- ‘At day ten, fluorescent beads were added…’ I really enjoyed the thought process behind this assay, however, I have some questions about the translational utility. Were the beads similar to mucous size? Were the beads coated in anything to mimic mucus?

This is a great question. We picked 7µm beads for these experiments. We looked at size of not only mucous (polymeres approximatey 10µm), but also bacteria (1-10µm) and particulate matter (<10µm) which are also transported by cilia. The beads were not coated with anything.

10. L533- ‘…we could mimic the inflammatory milieu…’ I don’t think that this is accurate. The inflammatory milieu consists of immune cells and a number of cytokines not included in these assays. I think this statement should be removed.

This is correct and we have removed the statement. The sentence on line 554 now reads:

We have used disease-relevant stimuli, such as TNF, TGF-β, CS and bacterial challenge on COPD primary epithelial cells, to establish in vitro challenge systems displaying signs of destruction of the epithelial barrier.

11. L564- ‘…cell cultures looked unaffected…’ Why did your cells not express decreased cilia length? This is obviously a crucial proponent of mucociliary clearance.

We believe that with TNF as a stimulus, we don’t affect cilia length per se, but rather affect the level of damage to the epithelium by inflammatory cytokines etc. It seems that this is creating an uneven surface of the apical epithelium and that particles and mucous get stuck and are not moving properly. We speculate that this is the reason for CBF not to be affected by TNF, but possibly the synchronization of the beating is. This is the reason why it is hard to show that ACY-1083 has a direct effect on cilia length, but it should be mentioned that we lack the proper methods to measure this.

12. A more detailed description of the limitations and future directions of this study should be included.

Thank you for this suggestion, we have forgotten to raise this important discussion. A section has now been added to the Discussion on line 618.

As with all in vitro work, our study has limitations. We have a limited number of donors for our experiments, and some are also merely to be considered exploratory. Despite this fact, our results are very convincing and ACY-1083 succeed in protecting the epithelium from injury. The underlying mechanism for this is also something that we have tried to address by having a broad and diverse set of functional assays, but it is unclear to us exactly how ACY-1083 has this protecting effect. To better understand this we believe that more molecular approaches may be helpful, such as next generation sequencing, as well as proteomics, where pathways and acetylation of proteins can be investigated. Since some aspects of our results could not confirm published findings, eg the cilia involvement described for other HDAC6 inhibitors, we also suggest a more targeted approach eg direct knock-out of HDAC6 in relevant human primary cells to validate that this has the expected effect. Lastly, this study does not involve any in vivo work. As with other targets, mechanistic understanding is sometimes easier to establish in vivo, as are the effects of an inhibitor such as ACY-1083, and this could be addressed in future work.

Minor comments:

1. Include the molecular weight of the compounds in the figure S1

This has now been specified on line 111 which now reads:

ACY-1083 (CAS1708113-43-2, Mw 348.35) is commercially available but was prepared according to a published procedure (50, 51). Tubastatin A (HY-13271, Mw 371.86 HCl salt) and Ricolinostat (HY-16026, Mw 433.5) were obtained from MedChemExpress.

2. Figure S2. It appears that there are only two measurements in C and D? Is this correct?

This is correct, it is two donors, with three individual measurements.

3. Figure S5. No statistics have been conducted. Please update this. There also appears to be very little difference in donor 3-could you comment on this?

Thank you for this comment. In figure 4B, we have the pooled results from all four donors. To our surprise, there was no significant reduction of goblet cell percentage with ACY-1083 between the no-TNF and the TNF treatment. When looking at the individual donors as shown in S5 Fig, it is clear that donor 3 did not respond to TNF in the same fashion as the other 3 donors. This is likely why we do not reach statistical significance. We can however not calculate statistics on the individual donors in S5 Fig since we for data originating from IHC sections, only have one measurement per donor.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

________________________________________

Submitted filename: Response to reviewers.docx

Click here for additional data file.

26 Aug 2022

12 Sep 2022

Please also see the file named "Response to Reviewers" where the attached patient data can also be found.

ANSWERS TO REVIEWER’S COMMENTS

PONE-D-22-07956R1

HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

PLOS ONE

Editors’ summary & main concerns:

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The authors should contact Lonza to acquire additional information regarding the donors. This information is usually available upon request.

Thank you for this comment, we agree that it is very important information. We have contacted Lonza and we have the table of patient data that they have given us. Unfortunately there is no disease stage data. We have included the table further down in the response to Reviewer #2, first question.

The authors should remove the reference to cilia length as recommended by the Reviewer 2.

We have changed the manuscript accordingly, please see the response to Reviewer #2, second question for a detailed description of changes.

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

________________________________________

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

________________________________________

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: N/A

________________________________________

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Regarding to patient background, the number of donors, authors did not solve the problems fully, but described clearly (exploratory vs. main experiment) or provided best possible information. Also authors use very high concentration of compound for ALI, and from my experiences, the compound has problems in mucosal residency. This is an experimental drug, and all findings are for basic science rather than clinical trial mimic. Therefore, I am satisfied with all comments provided by authors. For future work, I wish authors conduct power calculation and use appropriate number of donors even for ALI work.

We thank Reviewer 1 for the careful assessment of the manuscript. We will in the future make sure we do power calculations.

Reviewer #2: The author have replied convincingly to the majority of my points. However, I still have a few remaining questions.

(1)Thank you for this important remark. Reviewer 1 has asked the same thing and we

write: The donors used, both the COPD and healthy bronchial epithelial cells are

originating from cells purchased from Lonza. We have very limited information about

these but we have included age, gender and smoking history for these donors on line

134.

I understand that the Lonza bought cells are difficult to get information on. However, the COPD stage is exceptionally important given the different responses you see amongst donors. I would still recommend that this is added.

We agree that this is important and the GOLD stage might as you are saying give us an explanation as to why the patients are responding differently to the treatments and stimulations. We have again contacted Lonza to ask for this information but they refer us to this table that I have also attached and say that they are not normally able to access information regarding GOLD stages. Had we been using our own patient cohorts or collaborations we would have been able to add this information but unfortunately this is not always the case for commercially available cells. We are sad to say that we have not been able to access the information asked for.

(2) We believe that with TNF as a stimulus, we don’t affect cilia length per se, but rather

affect the level of damage to the epithelium by inflammatory cytokines etc. It seems

that this is creating an uneven surface of the apical epithelium and that particles and

mucous get stuck and are not moving properly. We speculate that this is the reason

for CBF not to be affected by TNF, but possibly the synchronization of the beating is.

This is the reason why it is hard to show that ACY-1083 has a direct effect on cilia

length, but it should be mentioned that we lack the proper methods to measure this.

I would recommend removing any reference to cilia length in this case.

This is a good point. We have changed two sentences and thereby removed any specific mentioning of cilia length in the discussion.

“It is established that COPD patients have increased remodeling and production of mucins, shorter cilia and alterations in cilia-related genes” on line 580

and

“In line with these findings, a few papers have shown protective effects of HDAC6 inhibition on cilia destruction after challenges both in vitro and in vivo” on line 584

Submitted filename: Response to reviewers.docx

Click here for additional data file.

26 Sep 2022

HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

PONE-D-22-07956R2

Dear Dr. Collins ,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Yulia Komarova

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors have attempted to answer the question sufficiently. Whilst it is important to know what GOLD stage the patient samples are obtained from, this does not seem to be within the control of the authors as they state.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

**********

2 Oct 2022

PONE-D-22-07956R2

HDAC6 inhibitor ACY-1083 shows lung epithelial protective features in COPD

Dear Dr. Collins:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Yulia Komarova

Academic Editor

PLOS ONE