Disease-causing Mutations in the Cystic Fibrosis Transmembrane Conductance Regulator Determine the Functional Responses of Alveolar Macrophages

Abstract

Alveolar macrophages (AMs) play a major role in host defense against microbial infections in the lung. To perform this function, these cells must ingest and destroy pathogens, generally in phagosomes, as well as secrete a number of products that signal other immune cells to respond. Recently, we demonstrated that murine alveolar macrophages employ the cystic fibrosis transmembrane conductance regulator (CFTR) Cl⁻ channel as a determinant in lysosomal acidification (Di, A., Brown, M. E., Deriy, L. V., Li, C., Szeto, F. L., Chen, Y., Huang, P., Tong, J., Naren, A. P., Bindokas, V., Palfrey, H. C., and Nelson, D. J. (2006) Nat. Cell Biol. 8, 933–944). Lysosomes and phagosomes in murine cftr^−/− AMs failed to acidify, and the cells were deficient in bacterial killing compared with wild type controls. Cystic fibrosis is caused by mutations in CFTR and is characterized by chronic lung infections. The information about relationships between the CFTR genotype and the disease phenotype is scarce both on the organismal and cellular level. The most common disease-causing mutation, ΔF508, is found in 70% of patients with cystic fibrosis. The mutant protein fails to fold properly and is targeted for proteosomal degradation. G551D, the second most common mutation, causes loss of function of the protein at the plasma membrane. In this study, we have investigated the impact of CFTR ΔF508 and G551D on a set of core intracellular functions, including organellar acidification, granule secretion, and microbicidal activity in the AM. Utilizing primary AMs from wild type, cftr^−/−, as well as mutant mice, we show a tight correlation between CFTR genotype and levels of lysosomal acidification, bacterial killing, and agonist-induced secretory responses, all of which would be expected to contribute to a significant impact on microbial clearance in the lung.

Introduction

Macrophages and neutrophils are key cells of the innate immune system. Blood monocytes infiltrate different tissues and then differentiate into tissue-specific macrophages that perform vital host defense functions, although neutrophils (PMNs)² are recruited from the blood to sites of infection. Mature macrophages from distinct sources exhibit significant variation in molecular and cellular properties as well as gene expression profiles specific for their host tissue, while maintaining a common set of core functions (2). One such cell type is the alveolar macrophage (AM) that resides in the terminal airway alveoli of the lung where they recruit PMNs that respond chemotactically to microbial insult. PMNs are generally considered the dominant component of innate immunity because of their sheer numerical superiority. Although less numerous than the mobile PMNs, AMs live longer and are a more potent source of the cytokines that orchestrate the immune response to bacterial pathogens. Ultimately, macrophages are also responsible for clearing apoptotic PMNs from infection sites, also by phagocytosis (3, 4). Delayed removal of the dying cells results in chronic inflammation with resultant tissue damage as seen in several chronic inflammatory lung diseases, including cystic fibrosis (CF) (for review see Henson et al. (5)).

Professional phagocytes have specialized pathways that ensure efficient killing of pathogens in phagosomes (6). A common element in these pathways is organellar acidification that facilitates the optimal functioning of various degradative enzymes, particularly in phagosomes (7). Indeed, low pH is required in several organelles for diverse functions in many cell types, e.g. maturation of secretory products and their final secretion by the exocytotic pathway and dissociation and recycling in the endosomal pathway. Generation of low organellar pH is primarily driven by the vesicular proton-ATPases, proton pumps that use cytoplasmic ATP to load H⁺ into the organelle (8, 9). Alongside the pumps are various channels that shunt the transmembrane potential generated by movement of protons; in different organelles, these include H⁺ channels, K⁺ channels, and Cl⁻ channels. Without these shunt pathways, acidification is limited, and organelle function is compromised (10).

Cl⁻ channels are central to the function of several intracellular organelles (11), and recently, we have shown definitively that the specialized Cl⁻ channel CFTR is important in lysosomal and phagosomal acidification in murine and human AMs (1). In parallel, others have shown that CFTR expressed on secretory granules may be involved in human neutrophil phagocytic function, being expressed on secretory granules (12). Painter et al. (12) showed that CFTR is also expressed on phagolysosomes in neutrophils where it may contribute Cl⁻ crucial to the chlorination reactions involved in bacterial killing. They demonstrated that neutrophils from CF patients exhibited a defect in chlorination of Pseudomonas aeruginosa proteins, presumably due to poor production of hypochlorous acid in phagolysosomes. Interestingly, clinical studies also suggest a phagocytic defect in pulmonary neutrophils from CF patients, although the mechanism is not known (13). Combined with our own observations, these results suggested that CFTR function may operate at several levels in regulating normal innate immunity.

In this study, we have used AMs from both cftr^+/+ and cftr^−/− mice as well as AMs from mutant mice expressing the two most common disease-causing CFTR mutations, ΔF508 and G551D. The ΔF508 is an in-frame deletion of phenylalanine at position 508 of exon 10 in the cftr gene. In most tissues, this mutated protein is rapidly degraded in the endoplasmic reticulum due to misfolding; however, if the channel reaches the plasma membrane (or perhaps the lysosomal/endosomal compartment), it is functional with a 7-fold decrement in protein kinase A-dependent activation over that observed for wild type channels (14). CFTR G551D is a missense mutation in exon 11, which is located in the signature sequence of the first nucleotide binding domain. The G551D mutation results in a channel that is appropriately localized to the plasma membrane but also has an extremely low open state probability in response to cAMP stimulation as compared with cells expressing the wild type protein (15, 16). Genetically engineered G551D mice display a defect in the innate immunologic response to inflammatory stimuli, with higher than normal levels of neutrophils in the lung as well as a hypersensitive macrophage phenotype following intravenous lipopolysaccharide challenge (17). Recently, McMorran et al. (18) have reported that these mice have an impaired clearance of Pseudomonas suggesting that this mouse might closely model CF lung disease. We compared the functional response of primary AMs for the four CFTR genotypes with respect to phagosomal and lysosomal acidification, bacterial killing, phagocytic index, and GTPγS-induced secretion.

EXPERIMENTAL PROCEDURES

Materials

The murine macrophage-like cell line, J774, was propagated as described previously (19). P. aeruginosa/EGFP (27853-EGFP) (20) was a kind gift from Dr. John C. Alverdy and Dr. Olga Zaborina (University of Chicago). Unlabeled zymosan, fluorescein (494/518)-labeled zymosan, Alexa Fluor-488-labeled zymosan, fluorescein tetramethylrhodamine (TMR)-tagged dextran (average M_r 70,000), nigericin, and valinomycin were obtained from Molecular Probes (Eugene, OR). pHrodo® Escherichia coli, a pH-sensitive rhodamine-based pHrodo® dye conjugated to E. coli BioParticles®, and 10-kDa dextran doubly conjugated to pHrodo® and Rhodamine green were received as a gift from Molecular Probes. The thiazolidinone CFTR inhibitor (CFTR_inh-172) and GTPγS were obtained from Sigma.

Animals

The studies detailed herein conform to the principles set forth by the Animal Welfare Act and the National Institutes of Health guidelines for the care and use of animals in biomedical research and were approved by the University of Chicago Institutional Animal Care and Use Committee. CFTR null mice (stock Cftr <tm1Unc>/TgN(FABPCFTR)#Jaw/Cwr) and ΔF508 mutant mice (C57BL/6 Cftr<tm1Kth>/TgN(FABPCFTR)#Jaw/Cwr) homozygous for ΔF508 (21) were purchased from the Cystic Fibrosis Animal Core at Case Western Reserve University. These mice express the human CFTR protein in the gut under the influence of the rat FABP promoter and are referred to as “gut corrected.” Mice carrying the CFTR mutation G551D (22) were a kind gift from Dr. Gerald B. Pier (Harvard Medical School, Boston). Animals were housed in a specific pathogen-free biohazard level 2 facility maintained by the University of Chicago Animal Resources Center (Chicago). Genotyping of G551D mice was performed by Embark Scientific (Austin, TX).

Bronchoalveolar lavage and peritoneal lavage fluids were collected from mice using standard methods. A total number of AMs recovered from bronchoalveolar lavage depended on the genotype and ranged from ∼0.6 × 10⁵ per mouse in WT mice to 1.1 × 10⁵ per mouse in ΔF508 mutant mice.

Analysis of Zymosan Uptake (Phagocytic Index)

Macrophages were incubated with fluorescein-conjugated zymosan at 0.5 mg/ml at 37 °C for 30 min; cells were then washed five times with phosphate-buffered saline. Cells with phagocytosed particles were imaged with a Leica SP2 AOBS inverted confocal microscope using a 488 nm argon laser, a 63× oil objective lens (NA = 1.4), and an emission bandwidth of 500–535 nm. The ingested zymosan particles were counted, and their intracellular localization confirmed by scanning z-series sections through the whole cell (one section about 1 μm). Phagocytic index was defined as the number of ingested particles per cell.

Analysis of Time Course of Phagosomal Acidification

Alveolar macrophages isolated from wild type mice cultured on glass coverslips were placed in HEPES-buffered DMEM, pH 7.4, prior to microscopy. Zymosan particles were labeled with Rhodamine Green (pH-insensitive) and pHrodo® (pH-sensitive) dyes (Molecular Probes/Invitrogen). Cells were visualized using a Leica SP5 confocal microscope (HCX PL APO CS 63× oil objective, NA = 1.4) on a 37 °C heated stage. Labeled zymosan particles suspended in DMEM were pipetted onto cells, and Z-stacks were taken at 1-μm steps through the entire depth of the cells every 30 s for 45 min. Excitation/emission spectra used for acquisition were as follows: 488 nm excitation, 500–540 nm emission (Rhodamine Green); 561 nm excitation, 566–700 nm emission (pHrodo®). Laser settings were adjusted to minimize photobleaching during image acquisition.

Image analysis was performed with ImageJ using custom macros for tracking individual particles over time (V. P. Bindokas, University of Chicago). Starting time point of particle uptake into the cell was determined visually by scanning through focal planes. To determine phagosomal pH, image stacks from each time point were flattened, and the fluorescent signals of all confocal slices were summed into a single image for each channel. Regions of interest (ROI) representing individual zymosan particles were defined using the green channel, and then the surface area, mean intensity of each channel, and fluorescence ratio were recorded from each ROI at each time point. Fluorescence ratios were normalized to start point value and fit to a smooth curve using Origin analysis software (Microcal Software, Inc., Northampton, MA).

Determination of Lysosomal Acidification

Alveolar and peritoneal macrophages and blood monocytes were isolated from WT and cftr^−/− mice as well as ΔF508 and G551D CFTR mutant mice and incubated with fluorescein-TMR-tagged dextran (average M_r 70,000; 5 mg·ml⁻¹) for 1 h at 37 °C. Labeled cells were washed several times with serum-free medium and chased in fresh medium for 2 h. Live cell imaging was carried out in medium buffered with 20 mm HEPES, pH 7.4. Cells were visualized with a Leica SP2 AOBS inverted confocal microscope using a 488-nm laser and a 594-nm laser, a 63× oil objective lens (NA = 1.4), and emission bandwidths of 500–535 and 607–686 nm, respectively. Precautions were taken to minimize the 488-nm laser intensity to prevent photobleaching of fluorescein (the laser was used at 9% power or less, and the images were taken within 10 s). No photobleaching was registered during this time interval in our pilot experiments (data not shown).

Area-integrated intensities for each fluorophore and their computed TMR/FITC intensity ratios were determined using a custom ImageJ macro (M. Cammer and V. P. Bindokas, University of Chicago). Corresponding pH values for each ratio were interpolated from the calibration curve. In experiments with doubly conjugated 10-kDa dextran to pHrodo® (pH-sensitive) and Rhodamine Green (pH-insensitive) (Molecular Probes, Eugene, OR), the same dye-loading protocol was used.

Lysosomal pH Calibration

Calibration curve for determining lysosomal pH was created by incubating cells for 15–20 min at room temperature in triple component calibration buffer at different pH values (1) and the ionophores nigericin (10 μm), monensin (1 μm), valinomycin (10 μm), and the H⁺-ATPase pump inhibitor, bafilomycin (0.1 μm). The pH was adjusted from 4.5 to 7.5 in 0.5 pH unit increments. Cells were imaged using a Leica SP2 AOBS inverted confocal microscope (see above). TMR/FITC ratios were plotted against pH values fitted to an exponential decay function. The ratio of pHrodo® (pH-sensitive) to Rhodamine Green (pH-insensitive) emission was calibrated in situ as described above, and the experimental values obtained from samples were determined by interpolation.

Determination of Phagosomal Acidification

We used two experimental techniques to determine acidification of phagosomes. In the first approach, J774A.1 murine macrophages were seeded at 70,000 cells per well in a 96-well assay plate and cultured overnight in DMEM + 10% fetal calf serum. The cells were pretreated in complete medium at 37 °C in the absence or presence of CFTR_inh-172 at 10 or 20 μm concentrations for 30 min before removing the medium and switching to 100 μl of uptake buffer (HBSS buffered with 20 mm HEPES, pH 7.4) containing 1 mg/ml pHrodo® E. coli bioparticles (Molecular Probes/Invitrogen) in the absence or presence of CFTR_inh-172 at the indicated concentrations. The assay plate was incubated for 30 min at 37 °C in a humidified chamber, and then read on a Flex Station® plate reader (Molecular Devices, Sunnyvale, CA). Net phagocytosis was calculated by subtracting the base-line fluorescence from wells containing 100 μl of 1 mg/ml pHrodo® E. coli bioparticles but no cells.

In the second approach, AMs isolated from mice with different CFTR genotypes were incubated with the fluorescein-conjugated zymosan (Molecular Probes), 0.5 mg/ml, at 37 °C for 30 min, washed five times, and incubated for another hour in complete DMEM at 37 °C. During data acquisition, the medium was buffered at pH 7.4 with 20 mm HEPES. In all experiments with CFTR_inh-172, the inhibitor at the indicated concentrations was present at all times during the zymosan loading, further incubation, and image acquisition. Images were collected on a Leica SP2 AOBS laser confocal microscope employing a DMIRE2 platform and a 63× (NA 1.4) oil objective. During microscopy, cells were maintained at 37 °C using a stage heater in the medium buffered at pH 7.4 with 20 mm HEPES. Samples were located with minimum light exposure. Image capture used excitation via the 488-nm line of the argon laser (5% acousto-optical tunable filter power) and was recorded in the 500–535-nm emission range. Single zymosan particles found in a single focal plane were taken as ROIs and represented data points.

Phagosomal pH Calibration

Phagosomal pH was determined using an in situ calibration. Cells that had ingested fluorophore-conjugated zymosan particles were equilibrated in the three-component buffer with ionophores as used in the lysosomal calibration (see above). Cells were incubated for 15–20 min at room temperature in calibration buffers, pH 4.5–7.5, containing ionophores as above. Data were analyzed using ImageJ software. ROIs were drawn around single zymosan particles found in a focal plane and represent data points. The calibration curve was fit with a dose-response function using Origin (Microcal Software, Inc., Northampton, MA). In further analysis, the ratio of fluorescent particle intensity, inside versus outside the cells, was used to compare fluorescent intensity in the absence and presence of 10 μm CFTR_inh-172.

Determination of Bactericidal Activity of Alveolar Macrophages

AMs from WT, cftr^−/−, ΔF508, or G551D mutant mice were incubated with P. aeruginosa-EGFP (27853/pUCP24-EGFP) (20) (multiplicity of infection <10) for 40 min at 37 °C with 5% CO₂. Noningested bacteria were carefully washed away with culture medium lacking fetal bovine serum. Cells were incubated with a ceftazidime/amikacin mixture at 1 mg/ml each for 20 min at room temperature to eliminate the remaining noningested bacteria adherent to the plastic. In our pilot experiments, 95% of P. aeruginosa was killed under these conditions (data not shown). After two washings, complete DMEM buffered with 20 mm HEPES, pH 7.4, was added to the cells; 1 ml of mineral oil was carefully layered on the surface of the medium in the dish to prevent evaporation. The dish was kept in a microincubator (Warner) at 37 °C during image acquisition. Time lapse series were recorded on Olympus IX81 microscope using minimal light exposure (0.25% lamp output). Z-stacks were taken every 2 h (0, 2, 4, and 6 h) with ∼1-μm step. During analysis, ROIs were drawn around the entire cell, and the fluorescence intensities in all slices were determined and subsequently summed. Increase in fluorescence intensity over time was indicative of bacteria proliferation inside the cell.

Live Cell Image Data Analysis

Images were analyzed using Fluoview software, ImageJ (National Institutes of Health), and MetaMorph (Universal Imaging Corp., Downingtown, PA). Data are expressed as means ± S.E. Significance between groups was determined using the Student's t test.

Electrophysiological Recording

Electrophysiological data were acquired using an Optiplex 755, Core 2 Quad PC. Capacitance recordings were made in the whole cell configuration using an EPC-9 amplifier with a built in data acquisition interface (ITC-16, Instrutech, Port Washington, NY) in combination with PULSE acquisition software (HEKA Electronik, Lambrecht, Germany) and Igor Pro (Wavemetrics, Lake Oswego, OR) for graphics and data analysis. Recordings were made in the “sine + dc” stimulus mode under the LOCK-IN module of PULSE. The temporal resolution of the capacitance data were 10 ms/point using a 1-kHz 30-mV sine wave. The holding potential was −10 mV. All experiments were conducted at room temperature (22–24 °C). Pipettes with resistances of 6–8 megohms were obtained using quartz glass (Sutter Instrument Co., Novato, CA) and a Sutter model P-2000 puller. Experiments were carried out using a standard pipette solution that contained (in mm) the following: 80 K-MES, 40 KCl, 2 MgCl₂, 1.1 EGTA, 0.2 CaCl₂, and 10 HEPES, pH 7.2. The bath solution contained (in mm) the following: 100 NaCl, 50 KCl, 2 MgCl₂, 2 CaCl₂, and 10 HEPES, pH 7.4. Solution osmolarities were monitored with a vapor pressure osmometer (model 5500; Wescor, Logan, UT), and measured to be 290 mosm.

Electrophysiological Data Analysis

Electrophysiological data were acquired and analyzed off-line using PC computer-based IGOR Pro (WaveMetrics, Lake Oswego, OR). The number and size of the step changes in the capacitance recordings were obtained using an automated step analysis detection routine in Igor written in the laboratory (23). Amplitude histograms were constructed and fitted using Origin (Microcal Software, Inc.) with a multiple Gaussian algorithm. Step changes in capacitance were not distinguishable from noise at levels below 10 fF; therefore, histograms were not fit below 10 fF. In addition, step changes with a rise time of greater than 0.3 fF/s were ignored. The histogram bin size was 0.20 fF.

Capacitance Change (ΔCm) and Delay (Δt) Measurement

In analyzing capacitance (Cm) changes with time, we constructed averaged Cm time curves by zeroing the initial Cm value and averaging all Cm time recordings for each CFTR genotype. We measured ΔCm at t = 0 and t = 450–500 s, where the Cm of WT CFTR reached a constant Cm value. A threshold value of ΔCm = 100 fF was used to determine the lag time before the initiation of an increase in Cm.

All average results are presented as mean ± S.E. with the number of experiments in parentheses followed by the number of mice used in each experimental data set. Student's t test as well as the Kolmorgorov-Smirnov test was used to analyze data. A probability level of <0.05 was considered significant. All experiments were conducted at room temperature with the exception of the phagocytosis experiments, which were conducted at 37 °C.

RESULTS

The central tenet of our previously published studies, namely that the Cl⁻ conductance introduced into the AM lysosomal and phagosomal membrane by CFTR is used for charge neutralization of the primary H⁺ pump, thus enabling these organelles to develop an internal pH of ∼5.6 (1), is the platform for the current investigation. A recent report from Haggie and Verkman (24) presented data that appeared to conflict with our previously published data on the involvement of CFTR in lysosomal acidification. In their studies, they used the inhibitor CFTR_inh-172 in a variety of cell types to re-examine whether or not CFTR was involved in phagolysosomal acidification. In their studies, the inhibitor was ineffective in producing a decrease in phagolysosomal acidification at the concentration of 10 μm. Their findings are in contrast to our published findings in which CFTR_inh-172 inhibited lysosomal acidification in cftr^+/+ mice (1), a conclusion further supported by data obtained from cftr^−/− AMs. To address the disparate results obtained by Haggie and Verkman (24), we carried out fluorescence ratio imaging experiments on phagosomal acidification in wild type murine alveolar macrophages in the presence of CFTR_inh-172. In agreement with our previous data, we observed a significant decrement in phagosomal acidification in the inhibitor-treated cells following engulfment of FITC-conjugated zymosan particles (Fig. 1). The concentration dependence of the CFTR_inh-172-induced defect in acidification was further verified in studies using cells from the murine macrophage cell line J774A.1 (Fig. 2A). In these studies, cells were fed pHrodo® E. coli bioparticles where a decrease in phagosomal pH is associated with an increase in fluorescence. Averaged fluorescence following bacterial uptake was read in a plate reader and compared in the presence of both 10 and 20 μm CFTR_inh-172. As can be seen in the summary data in Fig. 2A, we observed a significant dose-dependent decrease in fluorescence (corresponding to an alkalinized phagosomal compartment) in the presence of the inhibitor. The data in Fig. 2A suggested that perhaps handling or cold storage of the highly hydrophobic CFTR_inh-172 might shift the dose-response curve of the drug. Therefore, we investigated whether cold storage of the solubilized CFTR_inh-172 resulted in a shift in the dose required to inhibit CFTR in cftr^+/+ AMs as has been observed for respiratory epithelial cells (25). In these studies, we used dextran labeled with pHrodo® (as the pH sensor) and Rhodamine Green (pH-insensitive) as the compartment marker. The doubly conjugated dextran has the advantage that the pH marker pHrodo®, in contrast to fluorescein, increases in fluorescence with increasing acidification of the compartment where it is concentrated. A calibration curve and a summary of our data obtained with the pHrodo®-Rhodamine Green dextran double conjugate in murine AMs are given in Fig. 2, B and C. We show that there was a significant shift in the concentration of the inhibitor required to produce maximal CFTR inhibition and thus an alkalinizing shift in lysosomal pH, between the freshly solubilized inhibitor and that reconstituted from frozen DMSO stock solution. In the freshly solubilized solution, inhibition/alkalinization is seen at 10 μm, whereas 30 μm was needed to achieve a similar level of inhibition/alkalinization when the inhibitor was reconstituted from frozen stock regardless of storage time.

FIGURE 1.

CFTR_inh-172-induced inhibition of phagosomal acidification in AMs isolated from WT mice. A, fluorescein-conjugated zymosan was exposed to WT AMs in the absence (control) of inhibitor or to cells that had been pretreated with 10 μm CFTR_inh-172 for 30 min. Images were collected on a Leica SP2 AOBS laser confocal microscope (see supplemental material). Particles not ingested by the cells are marked with white arrows. DIC, differential interference contrast. B, calibration curve was performed on the same cells using a three-component buffer with ionophores. Data were analyzed using ImageJ software. Single zymosan particles found in focal plane were taken as ROIs and represent data points on the curve. Data are expressed as means ± S.E. Calibration curve was fitted using Origin and the dose-response model. C, ratio of green fluorescence between particles inside versus outside cells (white arrows in A) in control cells (upper panel) and in the presence of 10 μm CFTR_inh-172 (lower panel). For analysis, fluorescence intensity of each inside particle was divided by the mean intensity of all outside particles. Asterisk indicates significance. Bars represent the mean of these ratios ± S.E.

FIGURE 2.

Concentration dependence of the CFTR_inh-172-induced inhibition of acidification in phagosomes and lysosomes. A, J774A.1 murine macrophages were fed with 1 mg/ml E. coli bioparticles conjugated to a pH-sensitive dye pHrodo® (Molecular Probes/Invitrogen) in the absence or presence of CFTR inhibitor at the indicated concentrations. The assay was performed in a 96-well plate format. The fluorescence was read on a Flex Station plate reader (Molecular Devices). Mean values are relative fluorescence units (RFU), calculated from four wells per sample. Error bars show standard deviation. B, AMs from WT or cftr^−/− (KO) were incubated with dextran (10 kDa) doubly conjugated to a pH-sensitive dye pHrodo® and Rhodamine Green. The ratio of pHrodo® (pH-sensitive) to Rhodamine Green (pH-insensitive) emission was calibrated in situ using a triple buffer system, and the experimental values obtained from samples in C were determined by interpolation. C, data summary is presented as the mean ± S.E. where the number of the cells examined is indicated above each bar. Note a significant shift in the concentration of the inhibitor required to produce the maximum inhibition of acidification between freshly solubilized (“fresh”) and reconstituted from stored frozen DMSO stock (“stored”).

Time Course of Phagosomal Acidification

There appears to be a controversy in the literature as to the precise timing of the phagosomal acidification process (24). We developed a dynamic assay investigating the kinetics of individual phagosomes undergoing the acidification process. Phagosomal acidification in macrophages is intimately tied to the acquisition of the vesicular proton-ATPase conferred by endolysosomal fusion (1, 26, 27). The point at which the phagosome reaches the pH of the mature lysosome, approximately pH 5.0, is one of the markers that can be used to determine full phagosomal maturation (28). Once the lysosome fuses with the nascent phagosome, the time course over which phagolysosomal acidification occurs is reported to be ∼12–15 min (1, 28–30) and consistent with the time course of phagosomal acidification in experiments where particle uptake is synchronized (24, 31). Synchronized particle uptake uses a cooling cycle to allow particles to bind followed by a heating cycle in which it is assumed that, once warmed, all particles are taken up into cells as a “synchronous wave” (32), and the resultant acidification kinetics measured on a population of particles is assumed to reflect that of individual phagosomes.

To determine whether, in fact, all particles taken up into AMs acidify at identical rates, we examined AMs exposed to pHrodo®-Rhodamine Green-conjugated zymosan in real time using video imaging. Cells were maintained at 37 °C while particles were dropped onto the cells. Particles reached the cells at different rates and were taken up into the cells with a variable time course. Not all zymosan particles that attached to the cell surface were taken up immediately; some particles resided bound to the cell surface for periods of minutes before entering the cell. The kinetics of acidification were determined off-line using particle tracking software. Time zero was determined visually by entry of the particle into the cell. Phagosomal acidification was initiated with a variable lag time once the particle entered the cell presumably by the acquisition of the lysosomal vesicular proton-ATPase (Fig. 3A) We compared rates of acidification in 15 phagosomes in a total of two cells. Surprisingly, once acidification started, the rate was fast and relatively constant from phagosome to phagosome as determined by exponential fits to individual acidification curves as in Fig. 3C. The average lag time between particle uptake and onset of acidification was 72 ± 13 s (range was 30–210 s). The average τ derived from single exponential fits to the acidification time course was 50.9 ± 10 s (summary statistics are given in Fig. 3C).

FIGURE 3.

Kinetics of phagosomal acidification. A, kinetics of phagosomal acidification was determined in experiments in which zymosan particles doubly conjugated to Rhodamine Green and pHrodo® were pipetted onto cells, and particle uptake was followed with live cell video microscopy. Movies of particle uptake were analyzed and particles followed with particle tracking software. B, fluorescence ratio imaging for the particle/phagosome outlined by boxes in A was calculated and plotted normalized to initial values obtained upon particle entry into the cell as a function of time. Fluorescence ratios (proportional to acidification) were determined every 30 s following particle uptake. The acidification time course of the particle/phagosome isolated by the boxes in A is depicted in the curve with the open boxes. The data were fitted with a smooth line through the time points. The two curves represent the acidification time course of two phagosomes that showed distinctly different time courses for acidification. The time course data for the phagosome depicted with open boxes is representative of a class that acidified rapidly upon entry. The data represented by the closed boxes is representative of a phagosome that acidified with a significant lag following particle uptake. C, initiation of acidification was determined as the time at which the pH/fluorescence ratio changed to ≤90% of its initial value. The rate of acidification (τ) was determined from exponential fits to the data with the initial time point taken at a value that was ≤70% of the total change in pH. These kinetic data are summarized in the box plots for a total of 15 phagosomes from cells isolated from two WT mice. The average lag time for the onset of acidification was 72.0 ± 13 s, and the average time constant (τ) was 50.9 ± 10 s.

Macrophage Tissue Source Defines Dependence of Lysosomal Acidification on CFTR Expression

To determine whether all macrophages, regardless of tissue source, showed lysosomal acidification to be sensitive to CFTR expression, we carried out a comparative analysis of lysosomal acidification using macrophages from three sources isolated from both WT and cftr^−/− mice. We examined AMs, peritoneal macrophages, and circulating monocytes, which all express CFTR (1). Interestingly, only AMs showed a dependence of lysosomal acidification upon CFTR expression (Fig. 4). In the WT AMs, fluorescein emission is quenched due to the acidic nature of the compartment (pH 5.16 ± 0.06, n = 58 cells (three mice)), although in CFTR KO (pH 6.91 ± 0.05, n = 161 (eight mice)) very little quenching is observed (Fig. 4, left panels). Lysosomal compartments in peritoneal macrophages (WT, pH 5.16 ± 0.06, n = 58 (four mice); CFTR KO, pH 5.27 ± 0.07, n = 58 (three mice)) and blood monocytes (WT, pH 5.61 ± 0.15, n = 12 (one mouse); CFTR KO, pH 5.65 ± 0.14, n = 14 (one mouse)) did not depend on CFTR expression. We were unable to observe a significant difference in lysosomal pH in AMs isolated from WT and clc-3^−/− animals (see supplemental Fig. 1S) demonstrating that a Cl⁻ channel important in intracellular acidification in primary hepatocytes (33) and neutrophils (34) did not play a significant role in AM acidification.

FIGURE 4.

Lysosomal acidification in AM, peritoneal macrophages, and blood monocytes as a function of CFTR expression. Lysosomal pH was compared between cftr^+/+ and cftr^−/− alveolar and peritoneal macrophages, and blood monocytes were loaded with dextran doubly conjugated with FITC and TMR. Representative differential interference contrast (DIC) and fluorescent images are given in A–C. Lower panel, data comparison between cell types using ratiometric pH determination. ROIs for quantification were drawn around the whole cell. Summary data are expressed as pH means ± S.E. of the mean where the number of cells examined under each condition is shown above each data point (the number of animals from which cells were taken for each experimental point is given in parentheses). Asterisk indicates significance. Inset, ratiometric data for each cell was calibrated in vivo using a multiple buffer system, and the pH values obtained from the samples were determined by interpolation.

Mutational Analysis of CFTR in the Functional Response of the Murine Alveolar Macrophage

Phenotype-genotype relationships as a function of CFTR expression and mutations were examined in murine AMs. We compared lysosomal acidification, bacterial killing, phagocytic index, and stimulus-evoked secretion in AMs derived from cftr^+/+ cftr^−/−, ΔF508, and G551D mice.

Lysosomal Acidification

We examined and compared lysosomal acidification in AMs as a function of CFTR genotype using live cell fluorescence ratio imaging. These data are summarized in Fig. 5. Our data show a clear phenotypic relationship between lysosomal acidification and CFTR genotype, with the most severe lysosomal alkalinization in the absence of CFTR expression in the AMs isolated from the cftr^−/− mice (pH 6.91 ± 0.05, n = 161 (eight mice)). The ΔF508 CFTR mutation also resulted in significantly impaired lysosomal AM acidification (pH 6.11 ± 0.06, n = 77 (three mice)). Acidification was the least affected in the G551D AMs (pH 5.91 ± 0.04, n = 73 (four mice)), where the protein is expressed at the plasma as well as intracellular membranes but is only marginally functional as a Cl⁻ conductance.

FIGURE 5.

Relationship between CFTR genotype and phenotype in the regulation of lysosomal acidification. A, murine AMs were obtained from homozygous ΔF508 and G551D animals and compared with data obtained from AM isolated from WT and cftr^−/− animals. Representative cell images from each of the cell types are shown. B, determination of intra-organelle pH in lysosome-like compartments. Fluorescence was averaged over the whole cell at a single focal plane and then background-subtracted. Summary data are expressed as pH means ± S.E. of the mean where the number of cells examined under each condition is shown above each data point (the number of animals from which cells were taken for each experimental point is given in parentheses). Asterisk means significance in all. C, ratio of TMR to fluorescein emission was calibrated in situ using a multiple buffer system, and the values obtained from the samples shown in A were determined by interpolation.

Bacterial Killing

The consequence of a decrease in steady-state lysosomal acidification in AMs is a decrease in bacterial killing (1). Thus, the next set of experiments was carried out to examine whether the lysosomal alkalinization seen in cells expressing mutant CFTR results in a similar decrement in their microbicidal capacity. Bacterial killing among the CFTR genotypes was examined using confocal video microscopy. Macrophages were allowed to ingest P. aeruginosa expressing EGFP over a 40-min period, washed with mixture of two antibiotics at 1 mg/ml each in culture medium without fetal bovine serum to kill remaining undigested bacteria adherent to the dish, and then observed microscopically for 6 h to assess intracellular bacterial proliferation as we have published previously (1). Although wild type cells exhibited a steady restriction of bacterial proliferation, cftr^−/− cells showed continued bacterial growth as assayed by increasing levels of intracellular fluorescence over time. These data are summarized in Fig. 6, A and B. Note the strong genotypic dependence on the extent of bacterial killing over time with the G551D cells showing the greatest microbicidal efficacy over time among the mutant cells. Experiments done to examine the effect of the CFTR_inh-172 on bacterial killing demonstrated that the inhibitor was itself an antibiotic in bacterial cultures at the micromolar concentrations that are used to inhibit channel function (data not shown).

FIGURE 6.

Capacity of macrophages to eliminate internalized bacteria is a function of AM genotype. A comparison of intracellular bacterial growth in single murine AMs was carried out using live cell microscopy. Cells were allowed to ingest EGFP-expressing P. aeruginosa for 30 min (multiplicity of infection <10). Adherent and noningested bacteria were then removed by washing and incubation with antibiotics, and live AMs were observed microscopically for ∼6 h. Representative AMs from ΔF508 CFTR mutant mouse with ingested bacteria at the initiation of the incubation period and after 6 h are seen in A. B, summary data from at least three separate experiments. C, comparison of phagocytic index across CFTR genotypes. AMs isolated from mice with different CFTR genotypes were incubated with fluorescein-conjugated zymosan A (Molecular Probes/Invitrogen) at the concentration of 0.5 mg/ml for 30 min at 37 °C with 5% CO₂. Noningested particles were removed by excessive washings. The cells were visualized by confocal microscopy that allowed scanning each cell through its depth to count all ingested particles and discriminate them from those attached but not ingested by the cell. The data are represented as the means ± S.E. The number of cells analyzed is given above each bar. The phagocytic index of the WT cells was significantly different from the ΔF508 cells (p < 0.001) using the Student's t test. Asterisk indicates significance.

Phagocytic Index

A defect in bacterial killing could be due to a decrease in the phagocytic index or due to a defect in the degradative process once bacterial uptake is complete. Therefore, we examined the phagocytic index as a function of CFTR genotype. As in our original findings (1) comparing wild type and cftr^−/− cells, the phagocytic response to zymosan uptake was identical in cells isolated from the two animals. Interestingly, the AMs isolated from the ΔF508 animals exhibited a significant decrease in their phagocytic index as summarized in Fig. 6C indicating that the capacity for bacterial proliferation in the killing studies would be underestimated in the mutant ΔF508 expressing AMs due to a significant decrement in bacterial uptake.

Secretion

Filling, priming, and subsequent exocytosis of synaptic vesicles and large dense core granules is dependent upon chloride channel expression as well as granule/vesicle acidification (35–37). Macrophages are secretory cells containing a panoply of vesicles with respect to both size and cargo (38). Although there are no data as to the acidification level of these granules, their release in macrophages is dependent upon a signaling process that generates both changes in small GTPases and intracellular Ca²⁺ levels (19, 23, 39) as is the case for all hematopoietic cells (19, 40–43). To determine whether the release of membrane bound vesicles from AMs was a function of CFTR expression perhaps linked to acidification, we conducted high resolution membrane capacitance recordings of the stimulus-induced secretory response in AMs derived from both WT and mutant mice. Capacitance measurements (Cm) track surface membrane area (in picofarads) in real time and can detect both exocytosis and endocytosis. For these studies, we used GTPγS as a secretory stimulus added to the electrophysiological pipette solution. The average GTPγS secretory response from both mutant and WT cells and cells exposed the vesicular H-ATPase inhibitor bafilomycin is given in Fig. 7A. When WT cells were exposed to bafilomycin prior to capacitance recording to induce vesicular alkalinization, GTPγS-driven exocytosis/secretion was completely abolished indicating that vesicle acidification is critical for release. Cells expressing mutant CFTR showed a smaller increase in cell size (as assayed as an increase in surface area or picofarads) or secretory response following stimulation with GTPγS as compared with the control cells (Fig. 7, A and B). The time to the initial increase in cell size thresholded at 100 fF also increased significantly in the mutants ΔF508 and G551D over that observed in WT cells (Fig. 7C). These data are consistent with a model in which vesicle acidification is necessary for release in AMs and CFTR appears to play a role in the acidification process.

FIGURE 7.

Mutations in CFTR alter the GTPγS-induced secretory response in AM. Secretion was stimulated by the intracellular introduction of 400 μm GTPγS through the patch clamp pipette. A, whole cell capacitance recordings were obtained with an EPC-9 computer-controlled patch clamp amplifier (HEKA Electronik, Lambrecht, Germany) running PULSE software (HEKA). The EPC-9 includes a built-in data acquisition interface (ITC-16, Instrutech, NY). The software package controlled the stimulus and data acquisition for the software lock-in amplifier in the sine + dc mode. The temporal resolution of the capacitance data were 40 ms/point with a 1-kHz, 20-mV sine wave. The holding potential in the capacitance experiments was −10 mV. All electrophysiological experiments were performed at room temperature. A comparison of averaged secretory responses in voltage-clamped AMs over time was performed. The gray shading over the smooth line represents the S.E. of the average membrane capacitance increase as a function of mutant CFTR genotype over time. Cell number (number of mice used in parenthesis) in the WT experiments was 6(4); ΔF508 was 12(6); G551D was 6(3); cftr^−/− KO was 3(2), and in the bafilomycin (BAF) experiments was 4(1). B, summary of the average change in membrane capacitance as well as percent change in membrane capacitance increase as a function of cell genotype. C, data summary of the time to initial secretory response thresholded at 100 fF.

The size of the step changes in capacitance is proportional to the size of the vesicle fusion events at the plasma membrane. As detectable increases in capacitance were only observed in WT and ΔF508 cells, we performed a step size analysis between records obtained from the two genotypes (Fig. 8). Steps were detected using an automated program that accepted events based upon the slope and minimal size (100 fF) of the capacitance increase. Events were summed over cells from each genotype. Records in Fig. 8A demonstrate the variability in step size during stimulus-induced secretion. Summation of the detected events in all cells for a given genotype is displayed in histogram format in Fig. 8B, and fit with a single Gaussian distribution in the case of the WT AMs and a double Gaussian distribution in the case of the mutant ΔF508-expressing cells. The mean amplitude of the small vesicle population was between 105 and 108 fF in both cell types. The mutant cells secreted a second larger vesicle population of ∼382 fF, which was not seen in the WT cells.

FIGURE 8.

Analysis of step size during the GTPγS-induced secretory response from wild type and ΔF508 AMs. Time-dependent changes in membrane capacitance were obtained from cultured AMs isolated from WT and ΔF508 mice following intracellular introduction of 400 μm GTPγS as in Fig. 7. A, representative capacitance traces showing step changes due to small vesicle fusion (less than 0.2 pF) and large vesicle fusion (greater than 0.2 pF). B, histograms of the total number of step changes in capacitance recorded during experiments obtained from 6 (4 mice) WT and 12 (6 mice) ΔF508 cells. The number and amplitude of step changes in capacitance were obtained with an automated step analysis detection routine in IGOR (Wavemetrics, Lake Oswego, OR) written in the laboratory. Steps observed in the capacitance traces were summed for each genotype. Data were fit in the case of WT with a single Gaussian and in the events detected in the ΔF508 cells with a double Gaussian. The solid lines through the bars indicate the best Gaussian fit with the peak average above each peak. The x axis indicates both step size in picofarads and vesicle diameter in μm (in red). C, cumulative amplitude histogram of step changes in capacitance pooled from WT (solid black circles) and ΔF508 cells (solid green circles). The amplitude distribution for the ΔF508 cells is shifted to the right indicating larger vesicle release in the mutant cells (p < 0.01, Kolmogorov-Smirnov).

DISCUSSION

Intracellular acidification plays a key role in a broad spectrum of cellular functions, including maintaining proteolytic enzymes in the lysosomal compartment at optimal pH (44), processing prohormones in large dense core granules to the active state (45), determining neurotransmitter loading of synaptic vesicles (35), and contributing to pathogen killing in the phagolysosomal compartment (1, 28, 29). The vesicular ATPase drives the influx of protons into the various membrane-bound organelles; however, the magnitude of the proton gradient inside each organelle is determined by the co-expression and activation of chloride channels that act as charge shunt pathways in the same organelle. The number and identity of chloride channels subserving this role are the subject of active investigation and appear to include members of the ClC family (46, 47) of anion channels as well as CFTR (1, 25, 48–50). Perhaps one of the most interesting of the unsolved questions is how the diversity of chloride channels involved in these processes is targeted to a given organelle and whether the activity is regulated by the development of an intragranular proton gradient.

We have determined in this study that CFTR is involved in a number of processes integral to AM function that are driven by vesicular acidification and that the most common disease-causing mutations of CFTR, ΔF508 and G551D result in the following: 1) a secretion defect; 2) a decrease in the killing of internalized bacteria; 3) a decrease in the phagocytic response in the cells isolated from the ΔF508 mutant mice; and 4) a down-regulation in acidification of an endosome/lysosome-like compartment in resting cells and in the phagosomal compartment in phagocytosing cells.

Regulated Secretory Activity in Macrophages Is Influenced by CFTR

The AM contribution to innate immunity is first and foremost due to their phagocytic and microbicidal activity. Accompanying this critical function, they also recruit other inflammatory cells by secreting a variety of products, including enzymes, cytokines, and reactive oxygen intermediates. Hematopoietic cells, in general, and macrophages, in particular, contain several distinct secretory organelles that are selectively mobilized by different stimuli (38). Macrophages contain different types of dense core granules, in addition to phagosomes and phagolysosomes (38). It has been known for some time that macrophages possess a class of lysosomes that is specialized for secretion (the so-called “secretory lysosome”; for review see Ref. 51). Using electrophysiological techniques, we have examined the intracellular requirements for secretion of these organelles and others in a series of studies (19, 23, 39). In this study, we determined that CFTR plays an integral role in determining the secretory response in AMs. The response correlates with the expression of CFTR as well as CFTR mutations common in human disease. Although we have not measured the steady-state pH of the secretory granules themselves in the AMs expressing the mutant forms of CFTR, the response to the absence of CFTR expression is equivalent to that evoked by a complete inhibition of the vesicular proton-ATPases and therefore is strongly linked to secretory granule alkalinization. A comparison of capacitance step size during the exocytotic response revealed that both WT- and ΔF508-expressing macrophages secrete a population of vesicles that have a mean amplitude of ∼105 fF. Given a membrane capacitance of 10 fF/μm², transformation of vesicle capacitance predicts a spherical structure of ∼0.9 μm in diameter. Thus, most steps in Fig. 8A result from the regulated release of small vesicles. The ΔF508-expressing cells showed an additional capacitance peak at 382 fF corresponding to a larger vesicle population with an average diameter of 1.7 μm. Normalization of peak size to number of cells studied revealed that the number of small vesicles secreted was equivalent in both WT and ΔF508 cells (data not shown). Thus, the mutant cells appeared to secrete a larger vesicle population absent in the WT cells. Although it is tempting to hypothesize that this unique vesicle population might be responsible for the heightened level of inflammatory cytokines released by CF alveolar macrophages (52), it is difficult to establish a direct correlation between vesicle size and associated cargo.

CFTR Is Critical to Normal Phagosomal Function in AMs

Recently, we established that AMs express functional CFTR and that cells from CFTR null mice exhibit defective bactericidal activity (1). The cause of this deficiency is apparently a failure of lysosomes and phagosomes to acidify properly in the knock-out mouse. The phagocytic index in the cftr^−/− cells per se is not affected, and it does not appear that CFTR affects phagolysosomal fusion or reactive oxygen species production. In this study, we have extended those observations to show that mutations in CFTR also modulated in a graded manner both lysosomal acidification as well as bactericidal events in AMs. The defect in bacterial killing appears to correlate with mutant channel trafficking to plasma membrane sites. The ΔF508 mutation, which is rapidly degraded due to misfolding, produces a defect in killing equivalent to that seen in the cftr^−/− cells. The defect in bacterial killing in the ΔF508 cells is undoubtedly exacerbated by the fact that the phagocytic index in these cells is reduced over that seen in the cftr^−/−, and G551D-expressing cells.

Two recent reports (24, 31) present data comparing the kinetics of acidification in WT-, cftr^−/−-, and ΔF508-expressing cells. Both groups studied the acidification time course of multiple phagosomes taken up into macrophages using differing dyes and particles, e.g. zymosan or labeled bacteria. The time course of phagosomal acidification was highly variable between the two studies but appeared to show little difference as a function of CFTR expression. Two important points should be noted when comparing our results with theirs. Barriere et al. (31) did show in pH dissipation assays that in AMs a protein kinase A-activated counteranion permeability associated with acidification of what they termed immature phagosomes formed within 5 min after ingestion of fluorescein-conjugated P. aeruginosa. This counteranion conductance was undetectable in the cftr^−/− cells. In addition, they were also able to show that recycling endosomes in AMs as well as in peritoneal macrophages demonstrated a proton efflux that was assumed to represent alkalinization dependent upon CFTR expression and consistent with our earlier observation of a considerable CFTR conductance at the plasma membrane of AMs (1). This observation is not consistent with our study showing that peritoneal endolysosomal acidification is not dependent upon CFTR expression.

Quantification of acidification levels in intracellular organelles is highly dependent upon a number of factors that are variably agreed upon among investigators in the field. Of prime importance is the cellular loading with fluid phase markers, dyes that are conjugated to dextrans and that traffic along the endocytic pathway with a predicted time constant. In some investigations, the loading time can be as long as overnight (53) or as little as 30 min (24) with an equally variable chase time. The time chosen to load cells with the fluid phase marker can have a significant impact on the steady-state pH measured in a given compartment. As an example, loading overnight can shift the observed steady-state lysosomal pH to a more alkalinized level by over half of a pH unit than that observed after 1 h of incubation with equivalent chase times (see supplemental Fig. 2S). The alkalinized compartment seen with the longer loading time could be due to an overload with lysosomotropic agents that are all weak bases. The fluid phase markers can be a doubly conjugated dextran (24) or two single conjugated dextrans that are assumed to label the targeted organellar population with a single kinetic time constant (31), which may or may not be the case with resultant inaccuracies in normalizing pH changes to compartment size. Some investigations have used a single dye that has a pH-sensitive excitation range as well as a relatively pH-insensitive range (but not zero) that can be used as a ratiometric indicator to normalize for compartment loading. This technique has a disadvantage that small changes in the relatively pH-insensitive range can cause profound changes in the ratiometric data. Once cells are loaded, an in situ calibration curve is constructed for conversion of fluorescence ratios or intensity to pH. Analysis schemes can vary as well. One method is to utilize averaged whole cell fluorescence so as not to introduce a detection bias to regions within the cell that are brighter than others. Regions of interest within a cell can vary widely throughout the pH range and thereby introduce a bias that may affect the collective observed pH. Finally, not all cellular populations may acidify in the same way even within a given classification. As we have seen in this study, tissue macrophages differ in the Cl⁻ channel type that is expressed in a given vesicle compartment acting as a charge shunt pathway. Cell lines made from primary cell populations, as in the case of the bone marrow macrophages in the studies of Lamothe and Valvano (53), may also express a different population of intraorganellar ion channels. Thus, to compare acidification data observed between laboratories, one must compare across similar loading criteria, dye usage, and analysis procedures in similar cell types. The mice used in such studies could also be a source of inconsistencies as a result of breeding strategies and genetic drift/ectopic expression of region-specific promoter-driven human CFTR used in gut-corrected mice. We examined the possibility of significant transgene ectopic expression in the population of cftr^−/− mice that have been gut-corrected by expression of the human CFTR protein driven by the rat L-type FABP promoter. The animals available from Case Western Reserve University as well as animals bred in small colonies outside of Case Western have been bred as homozygotes. In colonies that are inbred for a significant period of time, the possibility exists that AMs may acquire aberrant transgene expression that is then passed on to future generations quite rapidly if the colony is not re-derived regularly and bred as heterozygotes. To test for this, we carried out a genotype-phenotype experiment with a cftr^−/− mouse that was purchased from Case Western within the past year (unlike mice used in our earlier experiments that were acquired about 2 years ago and were phenotypically dissimilar). In this study, we isolated AMs from a single animal, isolated RNA from a fraction of the cells, and cultured the remaining cells for fluorescence imaging. These data are shown in supplemental Fig. 3S. In supplemental Fig. S3A, PCR analysis of macrophage cDNA using primers specific to human CFTR revealed that indeed the AMs from the “inbred” cftr^−/− mouse expressed the human transgene CFTR transcript not present in the cftr^+/+ mouse used as a control. We probed only for the human gene as routine genotyping for the murine homolog revealed that it was absent in the mouse. Fluorescence imaging experiments of cells loaded with doubly conjugated FITC-TMR dextran revealed that, in fact, AMs isolated from the WT mouse and from the inbred cftr^−/− mouse showed identical levels of acidification. As a positive control, we compared the “inbred” cftr^−/− and cftr^+/+ lysosomal acidification data to data obtained in ΔF508 AMs, which are bred as heterozygotes and show a significant shift in lysosomal acidification levels (supplemental Fig. S3C).

In summary, we have detected CFTR protein in peritoneal macrophages and observed no acidification defect in peritoneal macrophages from cftr^−/− animals. We surmise that other Cl⁻ channels may play a similar role in phagosomal function in these and possibly other innate immune cells. We determined that mice null for ClC-3 express normal lysosomal acidification in AMs. However, mice null for ClC-3 are susceptible to sepsis, and Moreland et al. (34) suggest that ClC-3 is crucial for normal host defense by mechanisms that may involve phagolysosomal and secretory behavior in neutrophils. Our hypothesis is that Cl⁻ channel function is intimately connected to phagosomal acidification in innate immune cells and that the molecular species of Cl⁻ channel may vary between cell types.