IL-13 Induces Loss of CFTR in Ionocytes and Reduces Airway Epithelial Fluid Absorption

The airway surface liquid (ASL) plays a crucial role in lung defense mechanisms, and its composition and volume are regulated by the airway epithelium. The cystic fibrosis transmembrane conductance regulator (CFTR) is abundantly expressed in a rare airway epithelial cell type called an ionocyte. Recently, we demonstrated that ionocytes can increase liquid absorption through apical CFTR and basolateral barttin/chloride channels, while airway secretory cells mediate liquid secretion through apical CFTR channels and basolateral NKCC1 transporters. Th2-driven (IL-4/IL-13) airway diseases, such as asthma, cause goblet cell metaplasia, accompanied by increased mucus production and airway secretions. In this study, we investigate the effect of IL-13 on chloride and liquid transport performed by ionocytes. IL-13 treatment of human airway epithelia was associated with reduced epithelial liquid absorption rates and increased ASL volume. Additionally, IL-13 treatment reduced the abundance of CFTR-positive ionocytes and increased the abundance of CFTR-positive secretory cells. Increasing ionocyte abundance attenuated liquid secretion caused by IL-13. Finally, CFTR-positive ionocytes were less common in asthma and COPD and associated with airflow obstruction. Our findings suggest that loss of CFTR in ionocytes contributes to the liquid secretion observed in IL-13-mediated airway diseases.


INTRODUCTION
The airway surface liquid (ASL) which overlies the respiratory tract epithelium is an important component of host defense.The composition of this thin layer of liquid is tightly regulated and maintains ciliary function, mucus biophysical properties, and mucociliary transport (MCT) (1)(2)(3)(4).ASL also contains antimicrobial proteins and peptides that are important in the lung's innate defense system (5,6).Changes in ASL pH and volume have been implicated in airway diseases including cystic fibrosis (CF), asthma, and chronic obstructive pulmonary disease (COPD) (2,5,(7)(8)(9)(10).
Despite accounting for approximately 1% of the cells in the airway, ionocytes express very high levels of cystic fibrosis transmembrane conductance regulator (CFTR) channels compared to secretory cells (29)(30)(31)(32).We recently reported that whereas CFTR channels in human airway epithelial secretory cells increase transepithelial liquid secretion, CFTR channels in ionocytes increase liquid absorption across healthy airways (33), suggesting that cellular segregation of CFTR channels allows for control of ASL volume.Th2 inflammatory responses associated with subsets of asthma and COPD increase both Cl -and mucus secretion (9,13,34,35).How Th2 inflammatory responses, which increase Cl -and mucus secretion through secretory cells, affect ionocytes is unknown.In this study, we investigated how IL-13 impacts airway liquid handling and whether IL-13 treatment affects ionocyte number and/or function.

IL-13 exposure increases goblet cell abundance and Cl -secretion
Treating airway epithelia with IL-13 increased the number of MUC5AC-positive goblet cells (Figure 1, A and B).We evaluated the effect of IL-13 on CFTR-mediated Cl -secretion using the short-circuit current (Isc) technique (Figure 1C).Adding the epithelial Na + channel (ENaC) inhibitor amiloride induced a larger fall in Isc for control epithelia but only a minor decrease for IL-13 conditions (Figure 1, C and D).The change in Isc (DIsc) with forskolin and IBMX was greater for IL-13-treated epithelia compared to controls (Figure 1, C and D).
Similarly, TMEM16A inhibition, with Ani9, and CFTR inhibition, with CFTRinh-172, caused a greater DIsc for IL-13 treated, compared to untreated epithelia (Figure 1 The post-amiloride increase in Isc with IL-13 depended on CFTR because IL-13 failed to increase post-amiloride Isc values when studying CF epithelia (Supplemental Figure 1, A and B).Furthermore, in these experiments and additional experiments performed on non-CF epithelia bathed in nominally Cl --free solutions, IL-13 did not affect amiloride-sensitive Isc values suggesting that the large increase in Cl -channel function produced the difference in amiloridesensitive Isc observed after treating non-CF epithelia with IL-13 (Supplemental Figure 1, A-D).

IL-13 treatment reduces barttin-positive ionocyte abundance
In healthy human airway epithelia, Lei et al. found that CFTR channels on secretory cells participate in Cl -secretion while ionocytes absorb Cl -through apical CFTR channels and ionocyte-specific basolateral barttin/Cl -channels (33).To evaluate how these barttin-containing ionocytes respond to IL-13, we optimized a multicolor flow cytometry approach to determine the cellular composition of airway epithelia (Figure 2, A-C and Supplemental Figure 2).IL-13 treatment increased the number of goblet cells and reduced ciliated cell abundance (Figure 2, D   and E).Surprisingly, we detected about half as many barttin-positive ionocytes in IL-13-treated epithelia compared to controls (Figure 2E).At the transcript level, IL-13 increased epithelial CFTR expression, yet decreased epithelial BSND, which encodes ionocyte-specific barttin (Figure 2, F and G).These data suggest that IL-13 increases CFTR in secretory cells while simultaneously decreasing the absorption function of ionocytes.

IL-13 increases the number of CFTR-positive goblet cells, but decreases the number of CFTRpositive ionocytes
Finding an increase in CFTR transcripts and a decrease in BSND transcripts and barttinpositive ionocytes suggested that secretory cells drove the increase in CFTR.To test this hypothesis, we performed immunofluorescent confocal microscopy after a time course of IL-13 exposures to investigate which cell types express CFTR.IL-13 exposure increased the total number of cells and cells with CFTR within 6 days (Figure 3, A and B).CFTR was nearly always observed on goblet cells, and IL-13 increased the number of goblet cells as early as day 4 (Figure 3, C and D).Accompanying the increase in goblet cells was a decrease in ciliated cells (Figure 3E).We rarely detected CFTR in ciliated cells (Figure 3, B and E).IL-13 treatment decreased the number of barttin-positive ionocytes and CFTR detection in remaining ionocytes as early as 4 days, and CFTR was rarely detected in ionocytes by day 12 (Figure 3, B, F, and G).
Using fluorescent in-situ hybridization (FISH) we found that IL-13 treatment reduced CFTR mRNA expression on ionocytes (Supplemental Figure 3).These results and our flow cytometry data indicate that an increase in CFTR-expressing goblet cells drives the increase in CFTR during IL-13 exposure, and any remaining barttin-positive ionocytes lack CFTR.

IL-13 treatment reduces transepithelial liquid absorption
Since IL-13 decreased CFTR and barttin in ionocytes, we hypothesized that IL-13 would decrease liquid absorption.We tested this hypothesis using two approaches.First, we measured ASL volume and found that IL-13 treatment significantly increased ASL volume in cultured epithelium, a finding consistent with reduced transepithelial liquid absorption (Figure 4A).Second, we directly assayed the epithelial liquid absorption rate by adding saline to the apical epithelial surface and 4 hours later measuring the recovered liquid on the apical surface (33,(36)(37)(38).Under control conditions, human airway epithelia are absorptive (36)(37)(38).Therefore, we predicted that IL-13 treatment would reduce liquid absorption.IL-13-treated epithelia had a decreased liquid absorption rate, and we occasionally observed liquid secretion (Figure 4B).This shift towards a liquid-secretion phenotype, as indicated by a decrease in absorption, began as early as 3 days (Figure 4B).Thus, IL-13 created a less-absorptive epithelium, consistent with our observation that after IL-13 exposure most epithelial CFTR channels exist in secretory cells where CFTR mediates Cl -secretion.
Then we bathed the basolateral surface of epithelia in low [Cl -] solutions to drive Cl -movement from the apical to the basolateral chamber.Finally, we revealed CFTR-mediated apical-tobasolateral Cl -flow by treating epithelia with forskolin/IBMX and the CFTR inhibitor CFTRinh-172 (Figure 6, A-C) (33).Compared to control epithelia, CFTRinh-172 blocked less transepithelial current (It) in IL-13 treated epithelia (Figure 6, A-D), consistent with our earlier findings that IL-13 treatment reduced ionocyte number and ionocyte CFTR expression.In epithelia overexpressing FOXI1, we saw a significant increase in the CFTR-dependent It in the absence or presence of IL-13 (Figure 6D).Finally, increasing the number of ionocytes, both in control and IL-13-treated epithelia, significantly decreased ASL volume and increased the liquid absorption rate (Figure 6, E and F).These results indicate a decrease in ionocyte-mediated ASL absorption contributes to the IL-13 mediated increase in ASL volume.

Ionocyte CFTR expression is reduced in IL-13-associated human airway diseases
We found that IL-13 treatment caused significant changes in ionocyte number and CFTR expression.Thus, we hypothesized that ionocytes in airway human diseases with elevated IL-13 would also exhibit altered CFTR detection.Asthma and COPD are both heterogenous diseases, with their phenotypes associated with IL-13 (11,12,40,41).To test our hypothesis, we performed immunofluorescent microscopy on formalin-fixed lung sections from deceased donors.We obtained lung sections from one control without lung disease, two COPD donors, a donor with asthma, and a donor who succumbed to an asthma exacerbation (status asthmaticus).
We stained serial sections for MUC5AC and tubulin, to look for evidence of goblet cell metaplasia, or for barttin and CFTR.In the control donor, we consistently observed strong CFTR labeling at the apical surface of barttin-positive ionocytes (Figure 7A).In COPD and asthma samples, we found an increased number of MUC5AC-positive cells (goblet cell metaplasia), consistent with elevated exposure to IL-13 (Supplemental Figure 4).However, we did not detect apical CFTR on most ionocytes but detected CFTR staining on other cells (Figure 7, B and C; Supplemental Figure 5).These findings are consistent with our results from IL-13 treated human airway epithelia.
To further investigate ionocyte CFTR expression in asthma, we analyzed previously published scRNA-seq data obtained from lower airway biopsies of asthma patients or controls (42) (Figure 7, D-F).We preserved the cluster classifications designated by the original authors.
The clustering revealed more goblet cells in people with asthma compared to controls, however, ionocytes occurred at a similar frequency (Figure 7, E-G).We next determined the mean expression of CFTR per ionocyte for each participant.We restricted our analysis to samples with more than 3 ionocytes, which excluded two controls and one person with asthma.Nearly all control ionocytes contained CFTR (Figure 7H).In contrast, ionocytes from people with asthma appeared as two populations: one population expressed CFTR mRNA at levels comparable to controls, while the other population of ionocytes lacked CFTR mRNA (Figure 7H).Overall, the percentage of ionocytes with CFTR was reduced in asthmatic samples (Figure 7I).Interestingly, when we compared the number of ionocytes with detectable CFTR in a participant with their spirometry, we found that the CFTR-positive ionocyte number fell with worsening airflow obstruction (Figure 7J).These data suggest that in asthma and other IL-13-driven diseases the absorption property of ionocytes involving CFTR and barttin/Cl -channels could be affected, while the total number of ionocytes might remain unchanged.

DISCUSSION
We found that IL-13 has divergent effects on CFTR expression/function in different types of airway epithelial cells, secretory cells versus ionocytes.We and others have shown that IL-13-and/or IL-4-mediated Th2 airway inflammation increases goblet cell number, CFTR expression, Cl -secretion, and is accompanied by increased ASL volume (19-21, 25, 27, 43-49).
The ionocyte's role in this response has been unknown.Unexpectedly, we discovered that IL-13 decreases the number of CFTR-rich barttin-positive ionocytes and eliminates apical CFTR detection in the remaining ionocytes.Thus, IL-13 increases ASL volume by both increasing Cl - secretion, through secretory cells, and inhibiting Cl -absorption, through ionocytes.
The discovery of ionocytes has prompted multiple groups to consider how CFTR is differentially regulated in different cells (33,39,50,51).These studies underlie the idea that CFTR has opposing functions depending on the cell type that expresses it.This current study furthers the idea that CFTR performs Cl -secretion in secretory cells and contributes to fluid absorption in ionocytes.By challenging epithelia with IL-13, we demonstrated that the same agonist has opposing effects on CFTR regulation in a cell-type specific manner.IL-13 increases the contribution from secretory cells to epithelial CFTR function, while decreasing the contribution from ionocytes to epithelial CFTR function.These changes were associated with an increase in ASL volume.How specific cell types differentially regulate CFTR expression remains to be determined, but could have therapeutic implications.The mechanism(s) underlying reduced ionocyte CFTR expression following IL-13 treatment are unknown.Sonic hedgehog (SHH) signaling was recently discovered to play a key role in ionocyte specification (52).Inhibition of SHH signaling reduced both the total number of barttin-positive ionocytes and the fraction of barttin-positive cells that were CFTR-positive; these results are similar to what we found with IL-13 treatment.However, several studies found that IL-4/IL-13 enhances SHH signaling which would not fit with our findings.Thus, whether changes in SHH signaling account for our results is unknown and could depend upon differences in cell culture models, in vitro versus in vivo conditions, and/or acute vs. chronic treatments.
While we focused our study on the effect of IL-13 on CFTR in ionocytes, it will be important in the future to investigate the effect of IL-13 on other ionocyte-specific proteins, including ATP6v0d2, and the transcription factors FOXI1 and ASCL3 (29,30,52).IL-13 is increased in patients with airway disease, including asthma and COPD (11,12,40,41).To determine the relevance of our in vitro findings for human disease, we used two approaches.First, we obtained lung histological samples from patients with COPD, asthma, and status asthmaticus.We hypothesized that ionocytes in Th2 diseased lungs would have reduced CFTR levels.In healthy control lungs, we consistently found that barttin-positive ionocytes had strong CFTR immunofluorescence at the apical surface.In contrast, in all four donors with airway disease barttin-positive ionocytes lacked detectable CFTR immunofluorescence.This was associated with an increase in goblet cell abundance in all donor samples.Second, to further investigate how IL-13 might alter ionocyte morphology in the context of disease, we explored published scRNA-seq databases.We found that asthmatics had higher goblet cell proportions, and that ionocytes from asthmatic donors were more likely to lack CFTR than ionocytes from 13 healthy, controls.How losing CFTR expression on ionocytes alters airway physiology or contributes to disease is unknown.However, we found an association between the proportion of ionocytes with CFTR in asthmatics and airflow obstruction.Whether the loss of ionocyte CFTR is contributing to worsening pulmonary function or simply reflects a more advanced disease state is unknown and requires further study.
The degree to which ionocytes contribute to ASL volume regulation could vary by region and upper versus lower airway disease may differentially impact ionocyte function.Scudieri et al. found a "proximal-to-distal" gradient of ionocytes with more ionocytes present in nasal versus bronchial samples (53).In the nasal mucosa of children with chronic rhinosinusitis, Han et al. found that the number of CFTR-positive ionocytes was reduced (54).These data suggest that the changes we observed could occur in both upper and lower airways.
The therapeutic implications of our findings are unknown.However, this work raises several questions.First, in the setting of increased mucus secretion, is it beneficial or harmful to switch from an absorptive to secretory epithelium?Earlier work from Galietta and colleagues demonstrated that bicarbonate secretion is required for effective mucin release in IL-4 treated airway epithelial cells (26) suggesting that a secretory phenotype is likely beneficial in the setting of goblet cell hyperplasia.Second, do biologics that target the IL-4/IL-13 pathway, in asthma or COPD, affect the number of CFTR-rich barttin-positive ionocytes (55,56)?And if so, are there therapeutic effects?Finally, does this work have implications for CF?In the absence of functional CFTR, other strategies to enhance liquid secretion could be beneficial such as targeting other chloride channels (57,58).
Our study has strengths and limitations.Strengths include: (1) We optimized a multicolor flow-cytometry based approach to study epithelial populations.Other groups have published techniques for utilizing flow cytometry on airway epithelial cells (59,60).We were able to capture the diverse range of cells in airway epithelia.Our technique has the distinct advantages that a) it utilizes markers commonly used in airway biology, b) results in discrete, mutually exclusive populations, and c) is scalable for high-throughput experimentation.( 2 This study also has important limitations: (1) We limited our experiments to human airway epithelia.Effects of IL-13 on ionocyte biology could diverge among species.(2) While our flow cytometry analyses indicated a decrease in barttin-positive ionocytes after IL-13 treatment, it is interesting to note that scRNA-seq studies observe similar numbers of ionocytes in Th2-driven diseases (42,61).These differences may arise from using barttin as a marker for ionocytes instead of the ionocyte transcriptome.In addition to regulating CFTR, barttin may also be specifically reduced.If CFTR-negative, barttin-negative ionocytes are closer transcriptionally to "healthy" ionocytes than other epithelial cells, the tSNE analysis will still assign them to the ionocyte cluster.(3) Vieira Braga et al.only published normalized expression values, so we were unable to determine the sequencing depth for each cell.If ionocytes from asthmatic donors had lower sequencing depths, it would increase the likelihood that CFTRnegative ionocytes are false-negatives.
Our study suggests that the hyper-secretory phenotype seen in Th2-driven airway disease is mediated, in part, by opposing effects on CFTR.On one hand, IL-13 increases CFTR on secretory cells to increase Cl -secretion.At the same time, IL-13 reduces CFTR on barttinpositive ionocytes and reduces their abundance, decreasing Cl -absorption.Together, both mechanisms contribute to liquid secretion.Our findings suggest that precise targeting of CFTR in specific cells could have important therapeutic implications.

METHODS
Sex as a biological variable.Human airway epithelial cultures were randomized and studied based upon availability.Epithelia from both sexes were used for these datasets.
Airway epithelia culture.Mature cultures were established and maintained as previously described (62).Mature epithelia were dissociated with Accutase for 30-60 minutes and resuspended in 400 µL of Pneumacult Ex + buffer per epithelial insert.200 µL (approximately ~5 x 10 4 cells, a 1:2 seeding ratio) were added to 0.33 cm 2 Corning transwells that had been previously coated with Type IV collagen with 400 µL Pneumacult Ex + basolaterally.Cultures were incubated at 37°C and 5% CO2 for 24-36 hours, at which point apical media was removed by gently pipetting the apical surface to remove non-adherent cells from the filter and replaced with fresh Pneumacult Ex + ; basolateral media was replaced with USG.24-36 hours later, apical media was removed and cells were cultured at ALI. Basolateral USG was changed every 2-3 days.IL-13 was reconstituted in PBS lacking divalent cations (PBS -/-) and used at 20 ng/mL.Electrophysiology.After 3 weeks of culturing at the air-liquid interface, we studied epithelia using voltage-clamp technique and Ussing chambers.Each epithelium was bathed in a 37°C, pH 7.40 Krebs solution containing (in mM): 135 NaCl, 2.4 K2HPO4, 0.6 KH2PO4, 1.2 CaCl2, 1.2 MgCl2, 5 HEPES, and 5 glucose.Transepithelial voltage-sensing electrodes and current-passing electrodes were constructed with 3 M KCl-agar bridges and offset using a mounted culture insert without cells prior to mounting an epithelium.The empty culture insert was then replaced with an epithelial culture.Once the transepithelial voltage across the epithelial culture stabilized, the transepithelial voltage was held at 0 mV with an operational amplifier (VCC-MC8, Physiologic Instruments) and short-circuit current (Isc) was recorded using Acquire and Analyze software (Physiologic Instruments).Transepithelial conductance was periodically monitored by transiently holding the transepithelial voltage at -5 mV then +5 mV and recording the resultant changes in current.After the electrophysiology experiment, the voltage clamp was released, and each chamber was washed with 20 mL of fresh 37°C Krebs solution then dismounted for analysis by flow cytometry.
Flow cytometry.Epithelial cultures were washed using PBS -/-, stained with a Fixable Near IR or Fixable Yellow Viability dyes (ThermoFisher) prior to dissociation, and then incubated with 400 µL of Accutase (200 µL apical, 200 µL basolateral) for 30-60 minutes at 37°C.Accutase was diluted by adding 400 µL PBS -/-basolaterally.The single cell suspension was generated by pipetting the apical liquid vigorously through a P200 tip, then transferred to a 1.5 ml microcentrifuge tube.The basolateral liquid was subsequently used to wash the apical membrane in ~150 µL volumes, totaling 3-4 washes.Microcentrifuge tube was then spun for 1 minute at 1000x RCF to pellet the single-cell suspension.Single-cell suspensions were fixed using the FOXP3 Fixation and Permeabilization Kit (eBioscience/ThermoFisher) following manufacturer's protocols.Cells were stored in Flow Buffer (Permeabilization Buffer supplemented with 10% Superblock (ThermoFisher) and 1% normal goat serum (NGS; Jackson Labs).Cells from individual epithelia were stored in a volume of 500 µL.100-250 µL singlecell suspension stored in Flow Buffer was transferred into a 96-well V bottom plate.Cells were spun down and resuspended in Flow Buffer supplemented with the desired conjugated antibodies and mixed 2-3 times using a P200 multichannel pipette.The plate was incubated at 37°C for 20-30 minutes, and then washed 3 times and resuspended with Flow Buffer.Samples were analyzed on a 2019 4-laser Attune NxT Flow Cytometer (ThermoFisher) using an autosampler in the 96well V bottom plate used for sample preparation.The Attune is an injection-based system which allows for the calculation of cell counts in a given volume.Single-color controls were generated using AbC Total Compensation beads (ThermoFisher).FSC files were exported, compensated, and analyzed using FlowJo.v10.Antibodies are listed in Supplemental Table 1.
Fluorescent in-situ hybridization.Primary human airway cultures were fixed with cold 4% paraformaldehyde (PFA) solution (ThermoFisher) for 24 hours.Airway cultures were washed with PBS and dehydrated with 30% sucrose overnight.Airway culture filters were embedded with Tissue-Plus™ O.C.T. Compound (ThermoFisher).Ten-micron thickness sections were cut and subjected to a single-molecule fluorescence in situ hybridization method named proximity ligation in situ hybridization (63) with slight modifications (64).In brief, tissue sections were fixed in PFA, incubated in 1X citric buffer with 0.05% lithium dodecyl sulfate (Sigma-Aldrich), and 0.1% triton-X-100 at 65 o C for 30 minutes, and digested with 0.05 mg/mL pepsin (Sigma-Aldrich) in 0.1 M HCl.Tissue sections were then sequentially incubated with synthesized oligo probes including gene specific hybridization probes and circularization probes (63) Immunofluorescence microscopy.Individual epithelia were washed with filtered PBS, fixed using 4% paraformaldehyde (PFA, Electron Microscopy Sciences) for 15 minutes, and permeabilized using 0.3% triton-X (Sigma-Aldrich) for 20 minutes.Cultured epithelia were blocked using Superblock (ThermoFisher) supplemented with 5% normal goat serum (Jackson ImmunoResearch) for at least 2 hours.Primary antibodies were resuspended in blocking buffer and added to epithelia for 2 hours before being washed.Conjugated Alexa Fluor antibodies (ThermoFisher) were resuspended in blocking buffer at 1:1000 and added to epithelia for 45 minutes before being washed off.Finally, epithelia were mounted onto slides using Vectashield Hardset mounting medium with DAPI (Vector Laboratories) and imaged using an Olympus confocal microscope (Olympus Fluoview Fv3000).Antibodies are listed in Supplemental Table 1.For human tissue sample studies, we performed immunofluorescent microscopy on formalinfixed lung sections from deceased donors obtained through the University of Iowa Comparative Pathology Lab.
Briefly, cultured epithelia were treated overnight with 10 µM forskolin and 100 µM IBMX and imaged the following day.Brightfield images were acquired.For analysis, the menisci intensity vs. distance were integrated and compared to a calibration curve.
, C and D).Inhibition of the basolateral NKCC1 transporter with bumetanide dissipated the remaining Isc (Figure 1, C and D).Therefore, IL-13 treatment increased transepithelial Cl -secretion.The large residual Isc after amiloride treatment produced by IL-13-treated epithelia (Figure 1, C D) indicated that IL-13 altered the driving force for Na + across ENaC.
) Flow cytometry is well suited to the study of ionocytes.The low abundance of ionocytes presents a challenge for studying the effect of interventions on ionocyte populations.Current scRNA-seq technology limits the number of cells that can reasonably be profiled.Available datasets usually contain only a handful of ionocytes per sample, amplifying noise from the stochastic selection of cells.On the other hand, microscopy-based analyses are slow and labor intensive.Flow cytometry allowed us to quickly measure the abundance of basal cells, ciliated cells, goblet cells, secretory cells, and ionocytes.Importantly, we were able to observe ~10 4 -10 5 cells per sample, allowing us to observe ~10 2 -10 3 ionocytes per replicate.This greatly increased the power of our study and allowed us to detect a two-fold reduction in ionocytes after IL-13.(3) The effect of IL-13 on ASL volume was measured leaving the ASL intact.(4) The effect of IL-13 on transepithelial liquid absorption, which includes both electrogenic and electroneutral processes, was measured.(5) We complemented our in vitro findings with scRNA-seq and spirometric data from human asthmatics.(6) We confirmed our flow cytometry findings by immunofluorescence confocal microscopy, detecting the reduction in barttin-positive ionocytes in IL-13 cultures, and highlighting how both flow cytometry and confocal microscopy can be complementary techniques for studying epithelial populations.