Efficient CD4Cre-Mediated Conditional KRas Expression in Alveolar Macrophages and Alveolar Epithelial Cells Causes Fatal Hyperproliferative Pneumonitis

This article is featured in In This Issue, p.1091

T cells, as a critical component in adaptive immunity, are pivotal in host defense against microbial infection and immune surveillance against tumors. However, T cell function must be tightly controlled to prevent the development of autoimmune diseases. Ample evidence has demonstrated that signaling from the TCR plays important roles in T cell development, homeostasis, activation, and tolerance (1, 2). An important event during TCR signaling is the activation of Ras via two guanine nucleotide releasing factors, RasGRP1 and Sos (3–6). Activation of Ras isoforms, including KRas, is a critical step for T cell development and activation (3, 7, 8). Ras activity also has to be tightly controlled. It is negatively controlled by RasGAPs, such as p120GAP or neurofibromin, which enhance the intrinsic GTPase activity (9). Additionally, RasGRP1 is activated by diacylglycerol (7, 10), which is negatively controlled by diacylglycerol kinases (11, 12). These opposing mechanisms ensure tight regulation of Ras activity (13–15). Elevated Ras signaling has been observed in 30% of solid tumors, indicating that unrestrained Ras activity in parenchymal cells contributes to carcinogenesis (16). In contrast, how elevated Ras activity may impact the immune system remains poorly understood.

Proper function of the lung is essential for life, as its gas exchange supplies oxygen for distribution to all the tissues in the body and discharges carbon dioxide created throughout the body. Gas exchange function relies on the unique structure of the alveoli, whose walls are composed of a single-cell layer of alveolar epithelial cells (AECs). AECs are defined into two categories. Type I AECs (AT1) are squamous cells that compose the vast majority of surface area of the alveolar wall. Type II AECs (AT2) are cuboidal, metabolically active cells that secrete pulmonary surfactant to ensure pulmonary compliance and prevent collapse of the lung at the end of expiration (17, 18). Additionally, multiple types of immune cells reside in the lung for protection from microbial infection and for proper pulmonary function. Alveolar macrophages (AMFs) are the most abundant immune cells residing in the distal lung. They play critical roles for lung homeostasis and respiratory function (19, 20). AMFs efficiently clear dead cells, surfactant, and pathogens, orchestrate pulmonary immune responses and tolerance, modulate tissue damage, regulate lung fibrosis, and are involved in lung cancer progression (21–28).

Cre-mediated recombination has proven an invaluable tool for tissue-specific and temporal manipulation of target genes in biomedical research. CD4Cre transgenic mice (29), which carry a Cre transgene under the control of the CD4 promoter, have been widely used for T cell–specific targeting of numerous genes. We initially aimed to investigate how selective enhancement of KRas signaling would affect T cell homeostasis and tolerance by analyzing mice carrying a conditional constitutively active KRas (caKRas) allele and the CD4Cre transgene. Highly efficient Cre-mediated recombination allowed caKRas expression in T cells in these mice, which causes dysregulation of T cell activation and function. Unexpectedly, CD4Cre mice also manifested highly efficient Cre-mediated recombination in AMFs, AECs, and bronchial epithelial cells (BECs), causing caKRas expression in these cells and subsequent AMF accumulation, AEC and BEC hyperplasia, multiple adenomas, and ultimate early lethality of the mice because of respiratory compromise. Our results not only illustrate the importance for tight control of Ras activity in T cells, AMFs, BECs, and AECs, but also caution data interpretation using the CD4Cre transgenic mice for T cell–specific gene manipulation, particularly when studying immune cell–mediated responses and pathogenesis in the lung.

B6.129S4-Kras^tm4Tyj/J (caKRas or Kras^LSL-G12D) mice (30), B6.Cg-Gt(ROSA)26Sor^{tm6(CAG-ZsGreen1)Hze}/J (ZsGreen) mice (31), and C57BL/6-Tg(Cd4-TcraDN32D3)1Aben/J (iVα14^tg) mice (32) were purchased from the Jackson Laboratory. B6.Cg-Tg(CD4Cre)1Cwi N9 (CD4Cre) mice (29) were obtained from Taconic Biosciences. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All mice were used in adherence to protocols approved by the Duke University Animal Care and Use Committee.

Lung was cut into small pieces with scissors and placed in 1 ml complete IMDM media (IMDM-10, 10% FBS, penicillin/streptomycin, 50 μM 2-ME) containing 5 mg/ml Collagenase Type IV (LS004189; Worthington) and 0.5 mg/ml DNase (LS002147; Worthington). After incubation at 37°C for 1 h, lung remnants were meshed completely and passed through a filter mesh. Lung single-cell suspension was pelleted and treated with ACK buffer (0.15 M NH₄Cl, 0.1 mM KHCO₃, 0.1 mM EDTA) to get rid of RBCs. After two washes with 10 ml IMDM-10, cells were suspended in IMDM-10.

Fluorescence-conjugated anti-mouse Ly-6G (1A8), CD11c (N418), Gr-1 (RB6-8C5), CD45.2 (104), CD45 (30-F11), Ly-6C (HK1.4), F4/80 (BM8), CD11b (M1/70), EpCAM (G8.8), I-A/I-E (M5/114.15.2), podoplanin (8.1.1), TCRγδ (GL3), CD4 (GK1.5), CD8 (53-6.7), CD62L (MEL-14), CD44 (IM7), TCRβ (H57-579), CD3 (17A2), CD69 (H1.2F3), anti–IFN-γ (XMG1.2), anti–IL-17A (TC11-18H10.1), Cre recombinase (900903), IgG isotype control (HTK888), and streptavidin were purchased from BioLegend. Anti–SiglecF (E50-2440) was purchased from BD Biosciences. Anti-prosurfactant protein C (ProSPC, O34784) was purchased from Abcam. Rhodamine-conjugated donkey anti-rabbit IgG (sc-2095) was purchased from Santa Cruz Biotechnology.

Single-cell suspensions were stained with PBS containing 2% FBS and fluorescently conjugated Abs at 4°C for 30 min. Cells were washed at least two times with PBS containing 2% FBS and resuspended in PBS containing 2% paraformaldehyde. Intracellular staining for ProSPC and Cre recombinase (1:2000) was performed using the eBioscience Foxp3 Staining Buffer Set. All flow cytometry data were collected using a FACSCanto II (BD Biosciences) and analyzed with FlowJo software. For intracellular cytokine detection, 500,000 lymph node cells in 200 μl IMDM-10 were seeded in U-bottom 96-well plates and left unstimulated or stimulated with 50 ng/ml PMA and 0.5 μg/ml ionomycin in the presence of 4 μM monensin. Five hours after stimulation, cells were stained for CD4 and CD8, followed by intracellular staining with fluorescently labeled anti–IFN-γ and anti–IL-17A Abs.

T cells were labeled with CFSE according to a previously published protocol (12). To determine spontaneous and TCR-induced CD4⁺ and CD8⁺ T cell proliferation, CFSE-labeled splenocytes in IMDM-10 were seeded at the concentration of 1.5–2 × 10⁶ cells per well in 48-well plates and left unstimulated or stimulated with anti-CD3ε (2C11; BioLegend) in the indicated concentrations. After incubation at 37°C for 72 h, cells were stained for CD4 and CD8 and then analyzed by flow cytometry.

Lungs were inflated using PBS/OCT at 1:1 ratio and then frozen in OCT at −80°C. Lung cryosections (5 μm) on slides were fixed precooled (−20°C) acetone and methanol 1:1 mixture for 9 min. After air-drying and blocking with 3% BSA in PBS for 30 min at room temperature, samples were washed two times with 1 ml PBS and then stained with anti-ProSPC (1:50 dilution) or biotin-conjugated anti-CD45 (1:50) for 1 h. Rhodamine-conjugated donkey anti-rabbit IgG (1:400; Santa Cruz Biotechnology) or PE-conjugated streptavidin (1:100) was added and incubated for 1 h. After washing two times, samples were covered with VECTASHIELD mounting medium containing 1.5 μg/ml DAPI (Vector Laboratories). Images were acquired using a Zeiss Axio Imager fluorescence microscope. Photoshop (Adobe Systems) was used for postacquisition processing of brightness and contrast.

Lung, liver, and pancreas were processed and embedded in paraffin. Five-micron sections were cut, and slides were stained with H&E. For immunohistochemistry staining of surfactant protein C (SPC) and podoplanin, lung sections were deparaffinized in xylene and rehydrated through graded ethanol. Endogenous peroxidase was blocked using 3% H₂O₂. Enzymatic epitope retrieval was performed using Carezyme II: Pepsin (for SPC; Biocare Medical, Concord, CA) on the IQ Kinetic Slide Stainer for 3 min at 37°C. Nonspecific sites were blocked using Rodent Block M (for SPC; Biocare Medical) or 2.5% normal horse serum (for podoplanin; Vector Laboratories, Burlingame, CA) for 20 min at room temperature. The sections were then incubated with rabbit anti-ProSPC polyclonal Ab (1:2500 dilution; 1 h, catalog no. AB3786, lot no. 2117989; Millipore/Chemicon International, Billerica, MA) or goat polyclonal podoplanin Ab (1:750 dilution; 1 h, catalog no. AF3244, lot no. WUF0313081; R&D Systems, Minneapolis, MN). Normal Rabbit Serum (Jackson Immunoresearch, West Grove, PA) or normal goat Ig (lot no. B1262332; BioLegend, San Diego, CA) at an equivalent dilution was used for negative control slides. Further, the sections were incubated with Rabbit-on-Rodent HRP-Polymer (for SPC, 15 min; Biocare Medical) or ImmPRESS Reagent Anti-Goat Ig Peroxidase (for podoplanin, 30 min; Vector Laboratories). Biotinylated Abs were followed by incubation with VECTASTAIN Elite ABC reagent (Vector Laboratories) for 30 min at room temperature. The Ag–Ab complex was visualized by incubating the section with 3-diaminobenzidine chromogen (DakoCytomation, Carpenteria, CA) for 6 min at room temperature. Finally, the sections were counterstained with hematoxylin, dehydrated through graded ethanol, cleared in xylene, and coverslipped.

Data were analyzed with the unpaired Student t test or the Wilcoxon signed-rank test, when all control’s values were zero, using the GraphPad Prism 5 software. A p value <0.05 is considered statistically significant.

ZsGreen⁺ AMFs and AECs were sorted from KRas^LSL-G12D–ZsGreen–CD4Cre mice and ZsGreen–CD4Cre mice. Total RNAs isolated from these cells using TRIzol Reagent (Sigma-Aldrich) were reverse-transcribed into first strand cDNA. KRas cDNA was amplified by PCR using KRas-specific primers (mkras-182F: 5′-AGGCCTGCTGAAAATGACTG-3′ from exon 2 and mkras-358R: 5′-TTGACCTGCTGTGTCGAGAA-3′ from exon 3). Amplified PCR products were sequenced using the mkras-358R primer.

Data from Cohen et al. (33) were analyzed using Seurat 3.0.0 in R3.5.1 (34). Only the CD45⁻ fraction was considered for analysis. Raw gene counts were plotted in the t-distributed stochastic neighbor embedding space and across the lung developmental timeline identified in Cohen et al (33). Cell counts and Cd4⁺ (>0 raw count) cell counts were summarized in total (all cells) and Sftpc⁺ (>0 raw count) cells.

Data made available through the LGEA Web Portal (https://research.cchmc.org/pbge/lunggens/mainportal.html) was accessed on 4/16/19. Cd4 expression was queried in the Drop-Seq dataset from postnatal day 1 and visualized using tools made available by the LGEA Web Portal.

To investigate how overactivation of Ras may impact primary T cells, we bred the B6.129S4-KRas^tm4Tyj/J (KRas^LSL-G12D) mice (30) with the B6.Cg-Tg(CD4Cre)1Cwi N9 (CD4Cre) mice (29) to generate KRas^LSL-G12D–CD4Cre (caKRas) mice. KRas^LSL-G12D mice have a LoxP-Stop-LoxP (LSL) cassette upstream of the starting codon in the endogenous KRas locus containing a G12D point mutation and were maintained in heterozygosity throughout the study. KRas^LSL-G12D is only expressed after Cre-mediated deletion of the STOP cassette. Previous studies in KRas^LSL-G12D–CD4Cre mice revealed enhanced Erk1/2 and mTOR signaling in thymocytes and impaired iNKT cell terminal maturation (35, 36). We found that splenic T cells from the KRas^LSL-G12D–CD4Cre mice displayed hyperactive properties, indicated by elevated expression of activation markers CD69 (Fig. 1A), decreased naive T cells, increased CD44⁺CD62L⁻ and CD44⁺CD62L⁺ effector/memory cells (Fig. 1B), and a 2-fold increase of IL-17A–producing and at least a 10-fold increase of IFN-γ–producing CD4 T cells (Fig. 1C). In vitro, both CD4⁺ and CD8⁺ caKRas T cells proliferated more vigorously than controls following anti-CD3 stimulation (Fig. 1D). Thus, dysregulated KRas activity leads to increased activated T cells in vivo and enhanced T cell activation in vitro.

The KRas^LSL-G12D–CD4Cre mice displayed noticeable growth retardation and early lethality, with most of the mice dying within 1 mo of age (Fig. 2A). Although mild mononuclear cell infiltration could be detected in the liver and pancreas (Fig. 2B), the most striking pathological abnormality in these mice occurred in the lung. The KRas^LSL-G12D–CD4Cre mice displayed markedly enlarged and dense lungs (Fig. 2C) with alveolar spaces filled with mononuclear cells (Fig. 2D). The total lung cellularity recovered from the lung digest was increased, with increases in both CD45⁺ (hematopoietic) cells and CD45⁻ (nonhematopoietic) cells (Fig. 2E). The severe structural distortion and consolidation of the lung nearly packed alveolar spaces, and impairment of lung function was presumed to be the cause of the early lethality of the animals.

In the KRas^LSL-G12D–CD4Cre lung, B cell numbers were decreased, αβT cell and dendritic cell numbers displayed decreased and increased trends, respectively, but the differences were not statistically significant, and eosinophils and neutrophils were not obviously different in number from wild-type (WT) control (Fig. 3A). By contrast, F4/80⁺SiglecF⁺ AMFs were increased in percentage both within the CD45⁺ cell population (Fig. 3B, 3C) and within total lung cells (Fig. 3D). AMFs were also dramatically increased in absolute number (Fig. 3E) in KRas^LSL-G12D–CD4Cre mice. Interstitial macrophages (IMFs; CD45⁺F4/80⁺CD11c⁻SiglecF⁻) were, however, decreased (Fig. 3E). Thus, the KRas^LSL-G12D–CD4Cre lung had unexpected substantial accumulation of AMFs.

Although CD45⁺ cells were increased in the lung from KRas^LSL-G12D–CD4Cre mice, they could not account for the drastic increase of total lung cells. Indeed, CD45⁻ cells were also substantially increased in these mice and contributed more to the increase in total lung cellularity (Fig. 2E). Further analysis of the CD45⁻ cells revealed marked increases in both percentages and numbers of EpCAM⁺MHC class II (MHCII)^hi AECs (Fig. 4A, 4B). H&E staining revealed BEC and AEC hyperplasia and development of multiple pulmonary adenomas in these mice (Fig. 4C). AECs can be defined into AT1 that express podoplanin (T1-α) and aquaporin-5 and AT2 that express ProSPC (37–39). Immunohistochemistry of lung thin sections revealed obvious diffuse consolidative accumulation of ProSPC ⁺ AT2 cells in these mice (Fig. 4D). The accumulation of these cells was further confirmed by immunofluorescence staining for podoplanin and ProSPC (Fig. 4E). However, the accumulation of podoplanin⁺ AT1 cells was not obvious or was much less severe than AT2 cells (Fig. 4D, 4E). Thus, drastic accumulation of AT2 and adenomas developed in the lung of KRas^LSL-G12D–CD4Cre mice at a very young age.

Although T cells were hyperactive in KRas^LSL-G12D–CD4Cre mice, adoptive transfer of peripheral T cells from these mice into adult T cell–deficient mice failed to induce a similar lung pathological condition in the recipient mice (data not shown). To explore other cell lineages that contributed to the lung pathological condition in KRas^LSL-G12D–CD4Cre mice, we investigated whether Cre-mediated deletion could occur in lineages other than T cells in CD4Cre mice. To this end, we bred the CD4Cre mice with conditional ZsGreen reporter mice (B6.Cg-Gt(ROSA)26Sor^{tm6(CAG-ZsGreen1)Hze}/J) (31) to generate ZsGreen–CD4Cre mice and ZsGreen mice as controls. In ZsGreen reporter mice, ZsGreen with an upstream LSL cassette was knocked into the Rosa26 locus. ZsGreen is only expressed after Cre-mediated deletion of the STOP cassette. As expected, CD4Cre achieved virtually 100% deletion efficiency in the αβ T cell lineage, including CD4⁺CD8⁺ double-positive (DP) and CD4⁺ or CD8⁺ single-positive (SP) thymocytes, and peripheral αβ T cells (Fig. 5A–C). iNKT cells also showed∼90% Cre-mediated recombination (Fig. 5D). However, only∼50% of iNKT cells from the iVα14TCR transgenic mice (32) were ZsGreen⁺, suggesting an abnormal developmental path of at least a portion of these transgenic iNKT cells, which might be caused by premature expression of the iVαTCR transgene.

At the CD4⁻CD8⁻CD3⁻ double-negative (DN) stage, the percentages of ZsGreen⁺ cells were very low in CD24⁺cKit⁺CD44⁺CD25⁻ ETP and CD44⁺CD25⁺ DN-II cells but gradually increased in CD44⁻CD25⁺ DN-III and CD44⁻CD25⁻ DN-IV cells (Fig. 5E). CD4Cre mice manifested variegated deletion efficiency in γδ T cells ranging from 25 to 70% among different tissues, with the highest deletion occurring in the liver (Fig. 5F). At present, it is unclear whether such variegated patterns resulted from altered Cre expression in γδT cells in local tissues or from differences of ZsGreen⁺ and ZsGreen⁻ γδT cells in homing potential to different tissues.

Correlated with ZsGreen reporter data, Cre protein was detected at low levels in DN thymocytes, at high levels in DP and CD4SP thymocytes, and in a small portion of CD8SP thymocytes (Fig. 5G, 5H). Of note, Cre protein was only expressed in peripheral CD4 T cells but not CD8 T cells (Fig. 5I).

Because both AMFs and AECs accumulated in KRas^LSL-G12D–CD4Cre mice, we examined ZsGreen expression in AMFs and AECs in adult ZsGreen–CD4Cre mice. Unexpectedly, virtually all CD45⁺F4/80⁺CD11c⁺SiglecF⁺ AMFs from ZsGreen–CD4Cre mice were ZsGreen⁺ (Fig. 6), suggesting efficient deletion of the STOP cassette by Cre in AMFs or their precursor cells. About 30% of CD45⁺F4/80⁺CD11c⁻SiglecF⁻ IMFs in the lung were ZsGreen⁺. Some other tissue-resident macrophages, including red pulp macrophages in the spleen and Kupffer cells in the liver, displayed modest 40–50% deletion efficiencies. However, peritoneal macrophage and microglia cells manifested <10% deletion efficiency (data not shown). Additionally, bone marrow–derived macrophages also displayed close to 30% deletion efficiency.

Besides AMFs,∼80% CD45⁻EpCam⁺MHCII^hi AECs were also ZsGreen⁺ and majority of these ZsGreen⁺ AECs expressed intermediate levels of ZsGreen with the remaining expressed high levels of ZsGreen (Fig. 7A, 7B). These MHCII^hi AECs contain ProSPC⁻podoplanin⁺ AT1 and ProSPC⁺podoplanin⁻ AT2 cells. In lung single-cell preparation, most AECs are AT2 cells and very low percentages are AT1 cells because of low integrity through the collagenase digestion processes (Fig. 7C, top panel). Both ZsGreen^hi and ZsGreen^int populations could be detected in AT1 and AT2 (Fig. 7C bottom panels) cells, suggesting heterogeneity at least in Rosa26 promoter activity within these AEC subtypes. A small portion of AECs in adult mice are MHC-II negative (MHCII⁻). It has been reported that AT1 cells are relatively enriched in this population of AECs (40). We found that the majority of CD45⁻EpCam⁺MHCII⁻podoplanin⁺ AT1 cells were also ZsGreen⁺ (Fig. 7D). Immunofluorescence microscopy further confirmed ZsGreen⁺ ProSPC⁺ AT2 cells and ZsGreen⁺ ProSPC⁻ flat cells that were consistent with AT1 morphology in the lung of ZsGreen–CD4Cre mice (Fig. 7E). Additionally, strong ZsGreen expression in bronchial epithelium could also be detected in these mice (Fig. 7F), indicating ZsGreen expression in BECs in these mice. Of note, none of these AEC populations or AMFs expressed detectable Cre protein (data not shown), suggesting that Cre protein might be very transiently expressed in these cells or expressed only in their progenitor cells.

To determine whether low-level Cd4 expression in epithelial cells occurs during lung development, we reanalyzed published single cell RNA–sequencing datasets (33, 41). In CD45⁻ cells in the developing mouse lung (33), we found sparse Cd4 expression compared with AT1-associated genes (Aqp5, Pdpn) and the AT2-associated gene (Sftpc) (Supplemental Fig. 1A). During developmental stages, sporadic expression of Cd4 occurred but was not restricted to one developmental time point (Supplemental Fig. 1B). In annotated datasets from Du et al. (41), sparse Cd4 expression was observed in nonhematopoietic cells, including but not limited to AT1/AT2 hybrid and AT2 cells (Supplemental Fig. 1C). Cd4 expression was also noted in a subset of lung macrophages. Further, we summarized Cd4 expression in total and Sftpc-positive cell fractions and observed sparse Cd4 expression overall but also in Sftpc-positive cell fractions (Supplemental Fig. 1D). Overall, these data suggest that Cd4 gene expression, although sporadic and sparse, is plausible in a subset of epithelial cells of the developing mouse lung. These data were consistent with our inability to detect Cre protein in AECs and support the possibilities that either Cd4 was transiently expressed in AEC precursors or the Cd4Cre transgene might be very transiently expressed in some AECs but sufficient to mark the cells by deleting the STOP cassette in ZsGreen reporter mice.

In KRas^LSL-G12D­–CD4Cre mice carrying the ZsGreen reporter, the lung itself appeared greenish in color compared with ZsGreen–CD4Cre mice (Fig. 8A). Massive presence of CD45⁺ and CD45⁻ ZsGreen⁺ cells could be observed to fill the alveolar space in the lung (Fig. 8B). The expanded AMFs (Fig. 8C) and AECs (Fig. 8D) were not only ZsGreen⁺ but also expressed much higher levels of ZsGreen than their respective controls in ZsGreen–CD4Cre mice, likely because of increased transcription or translation of the reporter because of enhanced Ras activity, which could increase not only transcription factor AP1 but also mTOR activation (35). DNA sequencing of KRas cDNAs amplified from sorted ZsGreen⁺ AMFs or AECs from caKRas–ZsGreen–CD4Cre mice showed equal expression of WT and KRas^G12D mRNAs in these cells (Fig. 8E), suggesting deletion of the STOP cassette in these cells by Cre protein. Thus, highly efficient Cre-mediated recombination and subsequent expression of caKRas in AMFs and AECs may cause severe expansion and transformation of these cells, leading to loss of pulmonary function, tumorigenesis, and early lethality.

Using a ZsGreen reporter, we have traced Cre-mediated recombination in the CD4Cre transgenic model. In T cells, Cre expression and mediated recombination start at very low levels in DNII thymocytes and gradually increase as the cells mature. At the DP stage, most cells express Cre and have completed Cre-mediated recombination. Cre protein continues to be expressed by CD4 SP T cells but ceases to express in CD8 T cells. γδT cells display variegated Cre-mediated recombination efficiency between different tissues with the highest efficiency in hepatic γδT cells. iNKT cells contain highly efficient Cre recombination, which is consistent with the fact that they are derived from DP thymocytes. Different from iNKT cells generated in WT mice, iNKT cells generated in mice carrying the iVα14 transgene only demonstrate∼50% recombination efficiency, suggesting aberrant development of some of these transgenic iNKT cells. Thus, caution must be exercised when using this strain of iNKT cell transgenic mice in combination with the CD4Cre transgene for studying iNKT cells.

We have revealed highly efficient Cre-mediated recombination in BECs, AECs, and AMFs in the ZsGreen–CD4Cre mice. We have detected high levels of ZsGreen expression in these cells using both fluorescence microscopy and flow cytometry. Additionally, both AECs and AMFs in caKRas-CD4Cre–ZsGreen mice express equal levels of WT and mutant KRas mRNAs, further indicating deletion of the LSL cassette in these cells via Cre-mediated recombination. Different from DP thymocytes and CD4 T cells, AMFs and AECs do not express detectable Cre, implying that Cre-mediated recombination occurs in their progenitor cells. AMFs arise from fetal monocytes that seed the alveolar space in the early postnatal period. AMF differentiation and maintenance is dependent on GM-CSF derived from AECs and basophils (33, 42). Thus, future fate-mating studies using this strain of CD4Cre transgenic mice might be instrumental for defining AEC and AMF ontogeny.

Previous studies have shown that caKRas impairs iNTK cell terminal maturation (22), correlated with enhanced mTOR signaling (35). We show in this study that caKRas causes hyperactivation of T cells and early onset of cellular infiltration in the liver and pancreas, suggesting the importance of tight control of Ras activity in T cells. Ras functions via activating multiple effector molecules/pathways, such as the Raf-Mek1/2-Erk1/2-AP1 pathway, PI3K, Ral, and PKCε (43, 44). In caKRas T cells, Erk1/2, mTOR, and PI3K activities are elevated (35). Erk1/2, PI3K, and mTOR are known to mediate T cell activation and/or survival. Elevated Erk1/2, PI3K/Akt, or mTOR signaling cause dysregulated T cell activation, resistance to anergy, loss of tolerance or quiescence, and autoimmunity (45–48). In addition to αβT cells, the RasGRP1–Ras–Erk pathway regulates γδT cell activation (49–51). Further studies should determine how caKRas may impact on γδT cell development and function.

The most striking pathological abnormalities that occur in caKRas-CD4Cre mice are the early onset and fatal accumulation of AMFs, hyperplasia of AECs and BECs, and development of multiple adenomas in the lung within 1 mo of age. Hypermorphic KRas has been found in numerous solid tumors. In animal models, expression of caKRas in airway epithelial cells induces tumor generation. However, tumorigenesis in these models usually takes a much longer time than in caKRas-CD4Cre mice (52, 53), suggesting that expression of caKRas in AMFs, AECs, BECs, and possibly in T cells might alter the local environment that facilitates tumorigenesis.

In summary, our data indicate that enhanced Ras activity leads to hyperactivation of T cells and fatal accumulation of AMFs, hyperplasia of both ATIs and ATIIs, and multiple adenomas in the lung, highlighting the importance of tight control of Ras activity in both the immune and respiratory systems. caKRas-CD4Cre mice may provide a novel model to study proliferative pneumonia and lung carcinoma. The unexpected high efficiency of Cre-mediated recombination in AMFs, BECs, and AECs in CD4Cre mice suggests that caution is warranted when using this strain of mice for T cell–specific gene targeting and suggests the utility of this strain of mice for gene manipulations in multiple cell lineages in the lung and for investigation of BEC/AEC ontogeny. Given the important roles that AMFs, BECs, and AECs play in lung homeostasis and function, host defense, and immunopathogenesis of diseases, including asthma, we caution data interpretation when using the CD4Cre model for evaluation of T cell functions/roles in the respiratory system. Extrapolating further, we speculate that it is likely that analogous cases of unexpected Cre-mediated recombination are yet to be discovered with other conditional Cre drivers.

We thank Duke Cancer Center Flow Cytometry Core Facility for providing sorting services and the National Institutes of Health Tetramer Facility for providing the CD1d tetramer. We thank the histology and immunohistochemistry core laboratories at the National Institute of Environmental Health Sciences (Research Triangle Park, Durham, NC) for their help with histology and immunohistochemistry.

The authors have no financial conflicts of interest.