Rescue from Pseudomonas aeruginosa Airway Infection via Stem Cell Transplantation

Abstract

Cystic fibrosis is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which lead to impaired ion transport in epithelial cells. Although lung failure due to chronic infection is the major comorbidity in individuals with cystic fibrosis, the role of CFTR in non-epithelial cells has not been definitively resolved. Given the important role of host defense cells, we evaluated the Cftr deficiency in pulmonary immune cells by hematopoietic stem cell transplantation in cystic fibrosis mice. We transplanted healthy bone marrow stem cells and could reveal a stable chimerism of wild-type cells in peripheral blood. The outcome of stem cell transplantation and the impact of healthy immune cells were evaluated in acute Pseudomonas aeruginosa airway infection. In this study, mice transplanted with wild-type cells displayed better survival, lower lung bacterial numbers, and a milder disease course. This improved physiology of infected mice correlated with successful intrapulmonary engraftment of graft-derived alveolar macrophages, as seen by immunofluorescence microscopy and flow cytometry of graft-specific leucocyte surface marker CD45 and macrophage marker CD68. Given the beneficial effect of hematopoietic stem cell transplantation and stable engraftment of monocyte-derived CD68-positive macrophages, we conclude that replacement of mutant Cftr macrophages attenuates airway infection in cystic fibrosis mice.

Graphical Abstract

Transplantation of healthy bone marrow stem cells into irradiated cystic fibrosis mice restored the pool of alveolar macrophages in murine lungs. Chimeric mice displayed a better outcome, when infected with Pseudomonas aeruginosa. Consequently, pulmonary macrophage transplantation may become an option for the future treatment of chronic lung diseases.

Introduction

Clinical manifestations of cystic fibrosis (CF), the most common life-limiting autosomal recessive disease of the white population, result from a defect in a multifunctional ion channel, known as CF transmembrane conductance regulator (CFTR) protein. CFTR is expressed in secretory epithelia, including the airways, sweat glands, pancreas, intestine, and the male genital tract.1, 2, 3 Within the pulmonary microenvironment, mutations of CFTR in epithelial cells result in the defective homeostasis of mucus clearance, which is associated with progressive colonization by various bacterial species. While the clinical phenotype of CF is complex, treatment of CF lung disease predominantly concentrates on the modulation of present CFTR and antibiotic treatment to eradicate chronic airway infections.4, 5, 6 In this line, CFTR potentiators and CFTR correctors have been shown to partially reverse the underlying defect,7, 8, 9 to improve lung function, anthropometry, and quality of life, and to reduce the frequency of pulmonary exacerbations.^10,11 However, chronic colonization by the major bacterial pathogens in CF lungs, i.e., Staphylococcus aureus and Pseudomonas aeruginosa, could only be intermittently suppressed, but not eradicated.^12,13 Chronic airway infection and inflammation represent the main causes of morbidity in CF patients, highlighting the utmost need for an improved understanding in the contribution of immune cells in the disease progression of CF.

In fact, previous studies suggested defective CFTR expressed on non-epithelial cells to be involved in the disease progression of CF.14, 15, 16, 17, 18, 19, 20 In this line, targeting CFTR in monocytes and thereof derived macrophages provided conflicting evidence of whether CFTR regulates phagosome acidification in macrophages and therefore alters bactericidal activity.^16,21, 22, 23, 24 Similarly, conditional knockout of Cftr led to aberrant immune response due to Cftr-deficient lymphocytes²⁵ and showed increased inflammation and prolonged infection in CF mice when challenged with P. aeruginosa,²⁶ further suggesting the impact of hematopoietic cells of the lymphoid and myeloid lineage in the disease progression of CF.

In order to identify the contribution of pulmonary immune cells in CF, we performed hematopoietic stem cell transplantation (HSCT) in Cftr-deficient mice and thereby were able to demonstrate a protective phenotype of transplanted mice in an acute airway infection with P. aeruginosa. Successful homing of donor-derived Cftr-competent stem cells and their differentiation into alveolar macrophages (AMs) reveal the crucial impact of these professional phagocytes in chronic CF lung infections.

Results

Generation of Chimeric Mice by Hematopoietic Stem Cell Transplantation

To dissect the impact of Cftr in professional phagocytes on the murine response to acute airway infections with P. aeruginosa, we compared the infections in chimeric mice. Subsequent to myeloablative irradiation, congenic wild-type (WT) or CF mice27, 28, 29 received wild-type or CF hematopoietic bone marrow stem cells. Of 183 irradiated mice (110 CF and 73 wild-type mice), 23 CF mice and 1 wild-type mouse died within the first 6 weeks after transplantation while 72 wild-type and 87 CF mice were successfully reconstituted. The 6-week mortality rate was significantly different between wild-type and CF recipients (p = 0.000032, Fisher’s exact test), indicating that Cftr deficiency reduced the fitness of the animals to cope with irradiation.^30,31

Genotyping of Chimeric Mice after Hematopoietic Stem Cell Transplantation

To determine the success of the rescue of defective Cftr in the hematological compartment, genomic DNA was genotyped in the Cftr locus by markers informative for wild-type and CF descent (see Figure S1 and Table S1). Genomic DNA extracted from peripheral blood of the chimeric mice (CF^B6, CF^CF, B6^CF, and B6^B6) was used for stepwise genotyping of the Cftr locus by two single-nucleotide polymorphisms (SNPs), two informative microsatellites, and two markers detecting the neomycin cassette of the mutated Cftr gene. First, all mice were typed at rs509334146 that differentiates the CF and wild-type B6 alleles by restriction digestion with HinfI (Figure S1A). Allele-specific PCR of the B6 allele of SNP rs49681000 was used to detect B6 DNA in CF^B6 and B6^CF chimeras (Figure S1B). Microsatellite Mit236 preferentially resolves the wild-type B6 (Figure S1C), and D6NC2 conversely distinguishes the mutated Cftr allele (Figure S1D). The genetic markers Neo1 and Neo2 detect the neomycin cassette in the disrupted Cftr gene and are therefore specific for the remaining CF DNA in CF^B6 chimeras (Figures S1E and S1F). Since we used markers with different sensitivity in the detection of the wild-type and the mutated Cftr allele, respectively, a combined assessment of markers allowed distinct quantification of Cftr genotype conversion up to a few percent (Table S1). Most chimeric recipients carried the donor Cftr genotype in more than 98% of their peripheral blood cells. B6 mice that had received CF bone marrow retained on average 2% of the endogenous Cftr genotype (range, 0%–8%), whereas nearly all CF mice were carrying their endogenous mutant Cftr in even lower amounts of 0%–2%.

Tracking the Success of Transplantation

Next, the engraftment and the cell type composition of transplanted cells in the murine recipients were investigated 8–10 weeks after transplantation. We applied flow cytometry (Figures 1A–1C; Figures S2A–S2C) and immunocytochemistry (Figure 1D; Figure S2D) by analyzing the expression of CD45.1 as a marker for transplantation. Recipient CF mice express the CD45.2 allele of the leukocyte surface marker; therefore, transplanted cells were identified by expression of the CD45.1 allele specific for the wild-type B6.CD45.1 donor strain. Since pulmonary macrophages represent the first line of defense in the airways,³² CD68 was chosen as a marker for macrophage-like cells.

Figure 1.

Transplanted Cells Engrafted Successfully in Chimeric CF Mice

(A–C) Cells were tracked via flow cytometry in (A) bone marrows, (B) lung lysates, and (C) bronchoalveolar lavage fluids (BALFs) of uninfected CF^B6.CD45.1 (dark gray symbols) and CF^CF mice (white symbols). Most cells harvested from CF^B6.CD45.1 chimeric mice express CD45.1, whereas CF^CF mice solely express CD45.1⁻ cells. Macrophage-like cells (CD68⁺, circles) can be differentiated into donor-derived (CD68⁺CD45.1⁺) or recipient-derived cells (CD68⁺CD45.1⁻). Scatter dot plots are shown as mean ± SEM. (D) Representative immunofluorescence micrographs from lungs (bronchial and alveolar regions) of CF^B6.CD45.1 and CF^CF chimeric mice. Wild-type B6.CD45.1 and untreated CF mice served as controls. CD45.1⁺CD68⁺ cells (arrows) could be detected in the lungs of wild-type B6.CD45.1 and CF^{B6 CD45.1} chimeras. The overlay of red and purple fluorescence signals resulted in a pink labeling of the double-positive cells (seen in the uppermost and third row), whereas the CD45.1⁻CD68⁺ CF cells indicated by asterisks show pure red fluorescence signals (seen in the second and fourth row). Antibodies: CD45.1 (1:100, allophycocyanin [APC], purple), CD68 (1:100, phycoerythrin [PE]-Cy7, red), EpCam (1:200, fluorescein isothiocyanate [FITC], green), nuclei counterstained with DAPI (blue). Original magnification, ×40; scale bars, 25 μm. Long-term engraftment of transplanted cells in chimeric CF mice is displayed in Figure S2.

Bone marrow cells of CF^B6.CD45.1 recipient mice displayed a similar expression of CD45.1 as did untreated B6.CD45.1 mice of about 80% positive cells. The CD45.2⁺ CF mice and CF^CF transplanted mice showed a false-positive cross-reactive signal of 1%–2%. As in B6.CD45.1 donor mice, around 10% of the transplanted cells expressed the macrophage marker CD68 in the bone marrow (Figure 1A).

Bronchoalveolar lavage fluid (BALF) and the myeloid compartment of processed lungs³³ contained around 95% and about 85% CD45.1⁺ donor cells, respectively, confirming the successful engraftment of the transplanted cells in the lungs of chimeric CF^B6.CD45.1 mice (Figures 1B and 1C). The macrophage marker CD68 was expressed in more than 95% of BALF cells and in around 10% of cells of the myeloid compartment. The residual pool of endogenous CD45.1⁻CD68⁺ BALF cells in the irradiated recipient varied between 0% and 8%.

Immunofluorescence microscopy of lungs of CF^B6.CD45.1 mice additionally confirmed the almost complete exchange of CD68⁺ cells by the applied stem cell transplantation protocol (Figure 1D).

To assess the stability of long-term engraftment of transplanted cells in the bone marrow and lungs of CF^B6.CD45.1 mice, analyses were extended up to 6 months after transplantation (Figure S2). Again, most transplanted cells could be found in the lungs (Figures S2 ), although the donor-derived cells apparently decreased over time and had been replenished by stem cells of the recipient (Figure S2A).

Acute Airway Infection with P. aeruginosa

The experiments described above demonstrated that our hematopoietic stem cell transplantation protocol successfully led to reconstitution of bone marrow and lung compartments of CF mice with Cftr proficient cells. Hence, we next asked whether the provision of wild-type Cftr to the pool of non-epithelial cells had increased the fitness of the chimeric mice to cope with an infection with the most common bacterial pathogen in CF airways, i.e., P. aeruginosa.

After a minimal recovery period of 6 weeks, groups of CF^B6/CF^CF or B6^B6/B6^CF irradiated chimeric mice were infected by intratracheal instillation of 1–2 × 10⁶ P. aeruginosa PAO1 bacteria. CF^B6 mice showed increased survival, less weight loss, and lower disease scores during 144 h of follow-up compared to the CF^CF controls (Figure 2, left panel). However, groups were indistinguishable in their temporal course of rectal temperature (Figure 2D) and lung function parameters (Figure 3, left panel; Figure S3, left panel), indicating that the provision of wild-type hematopoietic stem cells to Cftr-deficient recipients attenuated the outcome of infection without significantly affecting the acute symptoms during the first 24 h. Conversely, groups of B6^B6/B6^CF chimeric mice did not behave significantly different in the examined pathophysiological features of airway infection (Figures 2, 3, and S3, all right panels), suggesting that the supply of Cftr-deficient hematopoietic stem cells did not significantly influence the symptoms of infection in irradiated wild-type mice.

Figure 2.

Attenuation of Acute P. aeruginosa Airway Infection in CF Mice Transplanted with Wild-type Hematopoietic Stem Cells

(A–D) The graphs show (A) survival, (B) disease score, (C) body weight, and (D) rectal temperature of chimeric mice infected with 1–2 × 10⁶ CFU of P. aeruginosa PAO1. Left panel: CF chimeric mice (CF^B6: dark gray line, bars, and symbols; CF^CF: dotted line, white bars, and symbols). Right panel: wild-type chimeric mice (B6^B6: black line, bars, and symbols; B6^CF: light gray line, bars, and symbols). CF^CF, n = 15; CF^B6, n = 27; three independent experiments. B6^CF, n = 16; CF^B6, n = 16; two independent experiments. Symbols and bars display mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 3.

Lung Function of Chimeric Mice during Infection

(A–E) The graphs show (A) tidal volume, (B) minute volume, (C) time of inspiration plus expiration, (D) expiratory flow at 50% of expiration (EF50), and (E) breathing rate of chimeric mice infected with 1–2 × 10⁶ CFU of P. aeruginosa PAO1. Left panel: CF chimeric mice (CF^B6: dark gray circles; CF^CF: white circles). Right panel: wild-type chimeric mice (B6^B6: black squares; B6^CF: light gray squares). CF^CF, n = 15; CF^B6, n = 27; three independent experiments. B6^CF, n = 16; CF^B6, n = 16; two independent experiments (see also Figure S4). Symbols display mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Histology and Bacterial Load of Infected Lungs

Colony-forming unites (CFU) of P. aeruginosa in murine lungs were determined at 6 and 24 h post-infection (Figure 4). The bacterial load was significantly lower in CF^B6 than in CF^CF mice at 24 h, but not at 6 h. No differences in bacterial load were noted between B6^B6 and B6^CF chimeric mice (Figure 4A). Pathohistological analysis of lung sections revealed profound inflammation at the 24-h time point that was characterized by interstitial, perivascular, and peribronchiolar edema and cellular infiltration (Figure 4B). Consistent with the observation that the clinical symptoms of acute infection were indistinguishable during the first 24 h, no differences in lung histology were detected between groups of CF^B6, CF^CF, B6^B6, and B6^CF irradiated chimeric mice (Figure 4C). Bacterial counts and lung morphology indicate that the lung habitat of all four groups of chimeric mice responded to a matching dose of P. aeruginosa with a similar degree of influx of cells and inflammation. As indicated by reduced lung bacterial numbers, CF^B6 chimeric mice displayed a more proficient bacterial killing than did CF^CF mice, suggesting that Cftr is involved in the host’s response to infection in accordance with the hypothesis of a functional defect in CF phagocytes.34, 35, 36, 37

Figure 4.

Lung Bacterial Numbers and Histopathology of Chimeric Mice

CF^B6 (dark gray circles), CF^CF (white circles), B6^B6 (black squares), and B6^CF chimeric mice (light gray squares). (A) CFU in the lungs of bone marrow transplanted mice 24 h post-infection (p.i.) with 1–2 × 10⁶ CFU of P. aeruginosa PAO1. CF^B6 mice displayed significantly reduced bacterial load in their lungs compared to CF^CF controls, whereas no differences were seen between B6^B6 and B6^CF mice (mean ± SEM; ∗p < 0.05). Caudal and accessory lobes of the right lungs were taken for CFU determination; bacterial numbers were calculated per gram. (B) Representative micrographs of infected lungs. Upper row: overviews of infected lungs. Lower row: detailed views of marked areas. H&E stain. Original magnification, ×20; scale bars, 500 and 50 μm, respectively. (C) Quantitative evaluation of infected lungs. Inflammation was assessed separately for alveolar air space (alveoli) and lung tissue. The pathohistological signs of inflammation in the infected lungs manifested similarly in all four groups.

Cytometry of CD68⁺ Cells

CD68 is a marker protein of monocytes and tissue macrophages.³⁸ The population of these CD68⁺ cells from BALF was further differentiated by iterative chip cytometry (Figure 5A). Whereas the mean, albeit significant, differences in CD68, CD206, CD11b, and CD36 expression were only 2-fold or less between the four chimeric mice, an about 100-fold difference was noted between wild-type and CF recipient mice for the expression of CD44 (Figure 5B). CD44 is a major hyaluronan receptor on immune cells. During acute infection, the hyaluronan levels are known to rapidly increase in the lung.³⁹ Irrespective of the origin of transplanted cells, CF mice showed 100-fold higher levels of CD44 of the macrophage population, compared to B6 mice (Figure 5B). This finding might hint at a general higher inflammatory level in CF mice, which could be reduced in the presence of wild-type donor CD68 macrophages (Figure S4). The potential beneficial effect of healthy macrophages, however, did not reach levels of significance, and further assessment of the inflammatory response would need a comprehensive quantification of cell infiltration. When the expression data of the six markers were combined, the individual cells of the CD68⁺ population segregated by genotype, separating into B6^B6, CF^B6, B6^CF, and CF^CF groups (Figures 5G–5J). Within these groups the expression of CD68 decreased in the order B6^B6 > B6^CF > CF^CF > CF^B6. Although B6^B6 and B6^CF mice behaved indistinguishably when they were exposed to P. aeruginosa (Figures 2 and 3), the marker profile of the CD68⁺ cells was distinct, suggesting that the presence or absence of functional Cftr in the transplanted bone marrow cells led to divergent subsets of macrophages in the lung compartment. Any implications in macrophage activity, if they exist, were too subtle to be captured by the monitoring of macroscopic disease symptoms.

Figure 5.

Expression of Inflammatory and Activation Markers of CD68⁺ Cells in BALF of Chimeric Mice

Alveolar macrophages identified by CD68-high expression⁴⁰ were harvested from BALF of infected chimeric mice and analyzed via iterative chip-based cytometry (iCBC) regarding their expression of representative markers for inflammation and cell activation. The analysis is based on 85–100 cells/mouse, n = 3 per condition. (A–F) Relative expression of CD68 (A), CD44 (B), CD206 (C), CD11b (D), CD36 (E), and major histocompatibility complex class I (MHC-II) (F) of CF chimeric mice (upper panel: CF^B6, dark gray dots; CF^CF, white dots; lower panel: B6 chimeric mice, B6^B6, black dots; B6^CF, light gray dots; mean ± SEM) ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (G–J) Heatmaps (G and I) and constellation cluster analysis (H and J) of the expression of markers CD68, CD11b, CD206, CD44, CD36, and MHC-II in individual cells of CF^B6 and CF^CF (G and H) or of B6^B6 and B6^CF (I and J) chimeric mice.

Discussion

This study explored the impact of Cftr-proficient hematopoietic stem cell transfer on the outcome of severe pulmonary bacterial infection in a CF mouse model. Hematopoietic stem cell transplantation promoted a milder clinical course and higher survival from a potentially lethal infection with the typical CF pathogen P. aeruginosa. The transfer of Cftr-proficient cells did not attenuate the acute inflammatory response in the lungs of transplanted CF mice, but it improved their fitness to cope with and to resolve the infection. This scenario is similar to the course of acute airway infections in patients with CF who during the initial phase of infection display similar clinical symptoms and responses to a pathogen as do their non-CF peers, but due to their lack of functional CFTR, they subsequently do not proficiently clear the pathogen and resolve the infection.⁴¹

The study underlines the role of Cftr in non-epithelial host defense cells. Cftr-proficient CD68⁺ macrophage-like cells immigrated into the lungs and replaced the endogenous population of Cftr-deficient macrophages of the irradiated CF recipients, thereby increasing their fitness to cope with the bacterial infection. Interestingly, this role of Cftr-proficient donor cells was only evident in the context of the Cftr-deficient CF recipient. In the case of the congenic Cftr-proficient host, the clinical course and outcome of the infection with P. aeruginosa were indistinguishable, irrespective of whether wild-type or CF hematopoietic stem cells had been transferred. The population of resident Cftr-proficient airway cells could apparently compensate the functional deficits of the immigrated cell population of Cftr-deficient stem cell descendants. Going along with findings in myeloid Cftr-inactivated mice,^29,42 results from our murine model suggest that functional Cftr in non-epithelial cells makes a difference to contain an infection for an individual with CF, but not necessarily for the healthy immunocompetent host.

Cftr-deficient murine macrophages are known to be defective in the killing of internalized bacteria.^16,43 Within the human setting, the role of CFTR in professional phagocytes, such as neutrophils and macrophages, is still a subject of debate.^24,34,37 However, our growing clinical experience with CFTR modulation points to an important role of non-epithelial CFTR in host defense.^13,21,44 According to a recent study, modulator therapy can even partially restore CFTR function in platelets.⁴⁵ An increase of CFTR function in epithelial and non-epithelial cells, induced by systemic application of CFTR modulators in CF patients, has the capacity to overcome the lack of defense capabilities. Registry data have shown that chronic exposure to a CFTR modulator reduces the frequency of pulmonary exacerbations.^11,46,47 Pharmacological provision of some additional CFTR activity supports the CF patient’s capacity to cope with an infection.¹⁰ As in our CF mice that received Cftr-proficient host defense cells, some drug-mediated rescue of residual CFTR activity in phagocytes may be instrumental to promote an adequate host response during infection.

The study has limitations. We used CF mice generated by insertional mutagenesis²⁷ that, similar to class V CFTR mutations in humans,⁴⁸ produce low amounts of correctly spliced Cftr mRNA transcript in the presence of major amounts of aberrantly spliced isoforms.⁴⁹ The class II mutation p.Phe508del represents the most common CFTR mutation in humans,⁵⁰ and hence a p.Phe508del mouse could confer a higher translational impact for the molecular therapy of CF. However, although the three p.Phe508del mouse models in contrast to our CF mice are rather homogeneous in their Cftr mRNA phenotype, these three lines carrying a different genetic background are highly different in disease phenotype, highlighting the importance of genetic modifiers.^48,51, 52, 53

Our study could prove the effect of hematopoietic stem cell transplantation in an acute P. aeruginosa infection model, but a subsequent investigation of hematopoietic stem cell transplantation in the context of chronic infections, e.g., by the use of agar beads,⁵⁴ could shed light into the longevity of our approach. In contrast to the CF pig and CF ferret animal models, CF mice do not develop overt lung disease.⁵⁵ However, in the context of this work, an almost normal lung texture may be even advantageous because the role of deficient Cftr in host defense cells within the airways becomes even more accentuated. Of the descendants of the transferred hematopoietic stem cells, we selected CD68⁺ cells to trace the engraftment into the lung because macrophages are the resident cells of the first immune response, and pulmonary transplantation of stem cell-derived macrophages has been demonstrated to be sufficient to rescue mice from a lethal airway infection with P. aeruginosa.^56,57 Likewise, this decisive cell population, and not any other, was chosen for deep immune phenotyping by chip cytometry. However, we are aware that other immune cells, such as monocytes and neutrophils, also display alterations in the immune response in the CF lungs, which has been shown previously.58, 59, 60

As it is, our protocol of hematopoietic stem cell transfer into the irradiated host is too harmful for being translated into the clinical management of pulmonary infections in the human setting. However, having demonstrated that the transfer of host defense cells can rescue mice from a life-threatening bacterial infection, we consider pulmonary cell transfer to tackle human infections to be a promising strategy that is worth further pursuit. Macrophages derived from patient-specific induced pluripotent stem cells (iPSCs) are expanded ex vivo in bioreactors and are then locally supplied to the airways of the compromised host who is prone to recurrent and/or chronic bacterial airway infection.⁶¹ Pulmonary cell transfer may be recommended for patients with secondary immune deficiency or patients with chronic lung disease such as chronic obstructive pulmonary disease (COPD), bronchiectasis, or primary ciliary dyskinesia. In cases of monogenic diseases such as CF, the ex vivo expansion may be combined with gene correction. The generation of specific immune cells from iPSC sources in scalable bioreactors may extend the field of immunotherapy of bacterial airway infections in the foreseeable future.

Materials and Methods

Bacteria

P. aeruginosa strain PAO1⁶² was grown in lysogeny broth (LB) overnight at 37°C. The overnight culture was washed twice with the same volume of sterile phosphate-buffered saline (PBS; Thermo Fisher Scientific, Waltham, MA, USA) to remove cell debris and secreted exopolysaccharides. Then, the optical density of the bacterial suspension was determined and the intended number of CFU was extrapolated from a standard growth curve. Inocula with 1–2 × 10⁶ CFU in 30 μL were prepared by dilution with sterile PBS. This infection dose was able to produce a clinical infection without mortality.

Animal Procedures

Experiments were performed with CF mice (B6.129P2(CF/3)-Cftr^{TgH(neoim)Hgu}, named as CF) and wild-type mice of strain C57BL/6JHanZtm (named as B6) or B6.SJL-Ptprca-Pep3b/BoyJ (named as B6.CD45.1). The congenic CF mice at our site had been generated by inbreeding²⁸ from transgenic Cftr^{TgH(neoim)Hgu} mice.^26,27 Mice were regularly monitored for infection by typical pathogens according to the Federation of European Laboratory Animal Science Associations (FELASA) recommendations.⁶³

Generation of Bone Marrow Chimeras

Bone marrow chimeras were generated by myeloablative irradiation of mice followed by transplantation of freshly isolated bone marrow cells. CF^B6, CF^CF, B6^B6, and B6^CF chimeric mice (superscripts indicate the genotypes of donor cells) were utilized for infection experiments and genotyping, while CF^B6.CD45.1 were investigated by immunofluorescence and flow cytometry. To create CF^B6 or CF^B6.CD45.1 bone marrow chimeras, 12- to 14-week-old recipient congenic CF mice were conditioned by myeloablative irradiation (9.5 Gy) and within the next 24 h intravenously (i.v.) transplanted with bone marrow cells freshly isolated from femur, tibiae, and pelvis of wild-type donor (either from B6 or B6.CD45.1) mice. One wild-type donor provided cells for two CF recipients. The bones were flushed with PBS supplemented with 2% fetal calf serum (Thermo Fisher Scientific); harvested cells were then filtered through a 100-μm cell strainer (Greiner Bio-One, Kremsmünster, Austria) and after lysis of red blood cells stored on ice until i.v. application into the tail vein of mice (1–2 × 10⁷ cells/mouse). CF mice that received isogenic cells, generating CF^CF chimeras, served as controls. Mice were allowed to reconstitute for a minimum of 6 weeks, with the first 3 weeks receiving antimicrobial treatment with 0.1 mg/mL ciprofloxacin (Fresenius Kabi, Bad Homburg, Germany) via drinking water. Reconstitution of chimeric mice after irradiation was verified by analysis of blood cells. Blood was analyzed by automatic blood cell counting (blood analyzer Vet abc, scil animal care, Viernheim, Germany) using 50–100 μL of blood taken from the vena submandibularis of mice 6–8 weeks post-transplantation. Numbers of blood cells were compared to the relevant reference strains (see Table S1). Untreated CF and wild-type (B6.CD45.1) mice served as controls. Animals displaying continuous weight loss (up to 20%) or showing other abnormalities referred to X-ray irradiation disease were sacrificed.^30,31 Corresponding vice versa experiments were carried out in B6^CF chimeras and B6^B6 controls. Most transplantation experiments were done with wild-type strain B6. In experiments that were targeted toward the tracking of the transplanted cells, B6.CD45.1 mice were used.

Lung Infection Experiments

After 6–8 weeks of reconstitution, mice were infected intratracheally with 1 × 10⁶ CFU of P. aeruginosa reference strain PAO1. Comparisons were made between CF^B6 versus CF^CF and B6^B6 versus B6^CF always, since these groups were run simultaneously in experiments. Airway infection was carried out according to the previously established protocol of Munder et al.⁶⁴ To characterize the course of the bacterial infection, the overall health condition, weight, and rectal temperature of the mice were followed for up to 120 h. Furthermore, lung function was assessed by non-invasive head-out spirometry.⁶⁵ Subgroups of mice were sacrificed at 6 and 24 h for determination of lung CFU, histopathology, and cytokines, as well as for iterative chip-based cytometry (iCBC).⁶⁶

Histopathology of the Lungs

The right lungs were fixed via the trachea (4% [v/v] paraformaldehyde), embedded in paraffin, cut in 5-7 μm slices, and stained with hematoxylin and eosin. Quantitative histopathology was done on lung sections showing cuts from all three lobes of the right lung. Slides were digitalized at ×20 magnification using a microscopic slide scanner (Axio ScanZ.1, Carl Zeiss, Jena, Germany). For quantitative evaluation of lung inflammation, 36–42 micrographs were randomly selected from each tissue slide using Visiopharm software (Visiopharm, Hoersholm, Denmark) and analyzed with STEPanizer 1.0 software.⁶⁷ For the analysis, a point grid was laid over the images and points on selected structures (lung tissue and alveolar air space) were counted as well as on edema and inflammatory cells within these structural compartments i.e., erythrocytes, neutrophils, and macrophages. Finally, we set up a ratio for each inflammatory parameter within the different structural compartments.

Lung Bacterial Numbers

The left lungs of the euthanized mice were ligated, resected, and homogenized. Aliquots were plated and bacterial numbers of whole organs were calculated.

Cytokine Analysis

Levels of KC, interleukin (IL)-5, tumor necrosis factor (TNF)-α, macrophage inflammatory protein 1α (MIP-1α), IL-6, and monocyte chemoattractant protein-1 (MCP-1) in BALF of CF^CF- and CF^B6-infected mice were determined by a custom-designed bead-based multiplex assay according to the manufacturer’s instruction (R&D Systems, Minneapolis, MN, USA).

Genotyping

To verify the efficacy of stem cell transplantation, the Cftr locus of chimeric mice was genotyped by intragenic microsatellite (D6Mit236, D6NC2) and dimorphic markers (rs49681000, rs49144420, rs50934146) and by the neo marker gene of the targeting vector inserted into exon 10 of Cftr.²⁷ Analysis of the Cftr locus in blood samples of chimeric mice was done with a combination of genetic markers (see Table S2). CF and wild-type Cftr alleles were discriminated by restriction digestion of PCR products of three informative SNPs (http://www.informatics.jax.org) (Figures S1A and S1B). Furthermore, informative dinucleotide microsatellites spanning the murine Cftr locus²⁸ were selected from the AF162137 genome deposited in the NCBI database.⁶⁸ Microsatellites D6Mit236 and D6NC2, which preferentially resolve the CF and wild-type Cftr alleles, respectively, were visualized by direct blotting electrophoresis of PCR amplicons on polyacrylamide gels (Figures S1C and S1D).^28,69 Third, the residual amounts of the mutated Cftr allele in CF^B6 chimeras were quantified by PCR kinetics of the neo marker gene inserted by mutagenesis into the Cftr locus of the CF mice (Figures S1E and S1F).^58,70 Combining these markers allowed quantitative assessment of genotype conversion in the Cftr gene. A comprehensive table showing the readout of blood analyses for the genetic markers of all experimental animals is displayed in Table S3.

Tracking of Transplanted Cells

CF mice carrying the allele CD45.2 of the leukocyte surface marker CD45 were transplanted with freshly isolated bone marrow from B6.CD45.1 mice. For the tracking of transplanted cells in non-infected CF^B6.CD45.1 chimeric mice, left lungs were examined via immunofluorescence microscopy for CD45.1 and macrophage markers (CD68). Therefore, left lungs were filled with PBS and Tissue-Tek OCT compound (Sakura, Alphen aan den Rijn, the Netherlands), a specimen matrix of water-soluble glycols and resins for cryostat sectioning (3:1 [v/v]), shock-frozen on dry ice, and stored at −80°C until the preparation of the cryosections. In parallel, the right lungs were lavaged with 2 mL of PBS. Thereafter, lungs were treated according to the protocol of Gereke et al.,³³ optimized for the preparation of myeloid cells. The processed right lungs, the BALF solution, and the bone marrow of the chimeras were then analyzed by flow cytometry.

Flow Cytometry

For the detection of surface and intracellular proteins, BALFs (100,000 cells/sample), bone marrows, and lung lysates of murine chimeras (each 500,000 cells/sample) were stained with antibodies against CD45.1 and CD68 (Table S4). Antibodies were used in a dilution of 1:100 and incubated with the cells for 30 min at 4°C. Prior to the staining with the intracellular marker CD68, cells were permeabilized with the FOXP3 Fix/Perm buffer set (BioLegend, San Diego, CA, USA) according to the manufacturer’s instructions. Unstained samples and single-stained samples were used as controls. Fluorescence was measured using a FACSCanto flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) or a CytoFLEX (Beckman Coulter, Brea, CA, USA), respectively. Results were analyzed by FlowJo software version 10.5.1 (Becton Dickinson). The gating strategy is shown in the Figure S5.

Immunofluorescence Microscopy

Cryo-sections (5-6 μm) of left lungs were fixed with acetone and blocked for 30 min with 10% porcine serum. Staining was performed for 2 h at room temperature with antibody combinations described for flow cytometry (dilution 1:100) modified from the protocol of Mucci et al.⁷¹ Nuclei were counterstained with DAPI. Slides were covered with Fluorescence Mounting Medium (Agilent Technologies, Santa Clara, California) and observed with a DM6000B microscope (Leica, Wetzlar, Germany).

iCBC

Cells harvested from BALF were characterized via iCBC.⁶⁶ BALF cells were differentiated into AMs, interstitial macrophages (IMs), and neutrophils by the markers listed in Table S5. Iterative cycles of staining, imaging, and bleaching of dye-tagged probes allow comprehensive immune phenotyping of tissues and cell populations. The instrument ZellScannerONE (Zellkraftwerk, Leipzig, Germany) is made up of a Zeiss Axio Imager M2 automated microscope equipped with a Plan Apochromat ×20/0.8 objective and fitted with a Basler scA1400-17gm monochromatic camera. For iCBC, cells were allowed to attach inside of ZellSafe cell-adhesive microfluidic chips (for 45 min at 4°C). After staining, the cells bound within the microfluidic chip were scanned by the imaging system. To enable iterative staining of the cells, remaining signals were bleached using the same excitation filter as for imaging by extending illumination by 30 s per position. Bleaching was followed by acquisition of a second fluorescent light image from the same position for background correction. Afterward, the chip was removed from the microscope and stained for the next cycle. Precise re-positioning, which is a prerequisite for repetitive tracking of each individual cell, was accomplished by using the transmitted light image of the pattern formed by the cells at the respective position. Data acquisition and analysis of fluorescence intensities were done using the ZellExplorer application (Zellkraftwerk), and neutrophils, AMs, and IMs were identified as already described⁴⁰ and are shown in the gating strategy in the Figure S6). Heatmaps and cluster constellation graphs were created using Ward’s hierarchical clustering of single cells based on their expression of phenotypic markers using JMP software, version 12 (SAS Institute, Cary, NC, USA).

Statistical Analysis

Data were analyzed using GraphPad Prism 8 (GraphPad, San Diego, CA, USA). Values are shown as mean ± SEM. Either an unpaired Student’s t test or analysis of variance (ANOVA) was performed to assess statistical significance, which was determined using the Holm-Sidak method and set to ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, or not significant (ns).

Study Approval

Animal procedures were reviewed and approved by the local Animal Welfare Committee (reference nos. 33.14-42502-04-12/0901 and 33.9-42502-04-16/2072) and carried out according to the ARRIVE guidelines.⁷²