|Home | About | Journals | Submit | Contact Us | Français|
Mast cell-deficient KitW/KitW-v mice are an important resource for studying mast cell functions in vivo. However, because they are compound heterozygotes in a mixed genetic background and are infertile, they cannot be crossed easily with other mice.
To overcome this limitation, we explored the use of KitW-sh/KitW-sh mice for studying mast cell biology in vivo.
These mice are in a C57BL/6 background, are fertile and can be bred directly with other genetically modified mice. Ten-week-old KitW-sh/KitW-sh are profoundly mast cell-deficient. No mast cells are detected in any major organ, including the lung. Gene microarrays detect differential expression of just seven of 16 463 genes in lungs of KitW-sh/KitW-sh mice compared with wild-type mice, indicating that resting mast cells regulate expression of a small set of genes in the normal lung. Injecting 107 bone marrow-derived mast cells (BMMC) into tail veins of KitW-sh/KitW-sh mice reconstitutes mast cell populations in lung, stomach, liver, inguinal lymph nodes, and spleen, but not in the tongue, trachea or skin. Injection of BMMC into ear dermis or peritoneum reconstitutes mast cells locally in these tissues. When splenectomized KitW-sh/KitW-sh mice are intravenously injected with BMMC, mast cells circulate longer and are found more often in the liver and inguinal lymph nodes, indicating that the spleen acts as a reservoir for mast cells following injection and limits migration to some tissues.
In summary, these findings show that mast cell-deficient KitW-sh/KitW-sh mice possess unique attributes that favour their use for studying mast cell functions in vivo.
Mast cells are immune effector cells whose progenitors originate in the bone marrow . They reside in many tissues including the skin, lung, heart, stomach, intestine, spleen, lymph nodes, and peritoneum . Although mast cells are widely distributed in normal and diseased tissues , their roles in disease are frequently debated. A number of experimental approaches have been used to explore the importance of mast cells. Increasingly, mast cell-deficient mice are used in models of human disease .
Mutations of the white spotting (W) locus in mice were identified initially because of effects on haematopoiesis, germ cell development and pigment formation . In 1988, mutations of the Kit gene were found to explain the phenotypic changes associated with this locus . Of the mutations at this locus , KitW and KitW-v have been studied in greatest detail. The KitW defect is a point mutation that causes exon skipping and produces a truncated receptor . Consequently, KitW/KitW mice lack the kit receptor and with rare exceptions, are not viable [3, 7]. The KitW-v defect is a point mutation in the tyrosine kinase domain of the receptor . KitW-v/KitW-v mice have diminished Kit activity and a modest reduction in mast cell numbers. KitW/KitW-v compound heterozygous mice, which combine the severe KitW mutation with the milder KitW-v mutation, have markedly reduced Kit receptor activity and are severely mast cell-deficient (~1% of normal) . Because of this deficiency, KitW/KitW-v mice are used to study mast cell biology in vivo. By comparing outcomes in KitW/KitW-v and mast cell-reconstituted KitW/KitW-v mice in various disease models, mast cells have been implicated in the pathogenesis of asthma , arthritis , experimental allergic encephalitis , experimental bullous pemphigoid , cancer progression , and defence against bacterial infections [14, 15]. However, KitW/KitW-v mice have been difficult or impossible to use for studying roles of mast cells in disease models involving genetically modified mice because they are sterile and of a mixed genetic background and cannot be bred with other genetically modified mice without resorting to complex breeding and genotyping schemes . Furthermore, the combined inheritance of KitW and KitW-v is tolerated poorly in certain backgrounds of mice, such as FVB/n .
KitW-sh/KitW-sh ‘sash’ mice are another strain of mast cell-deficient mouse that was first identified when a single female with a white sash around her midsection was found in an inbred litter (C3H/HeH × 101/H) . Genetic and complementation studies revealed that this coat colour phenotype is because of heterozygous inheritance of a mutation at the W locus. Thus, this ‘sash’ mutation was given the symbol W-sh. Subsequent studies determined that the W-sh mutation is an inversion of a segment of chromosome 5, ~ 70 kb upstream of the Kit gene [17, 18]. This inversion disturbs regulatory elements and markedly reduces mast cell expression of Kit in animals that are homozygous for the inversion . Compared with KitW/KitW-v mice, the noteworthy features of KitW-sh/KitW-sh mice are fertility , lack of anaemia , and availability in a uniform C57BL/6 background into which they have been backcrossed . Because these features are advantageous for studies of mast cell function in vivo, in the present study we explore the breeding characteristics, mast cell deficiency, lung gene expression, and kinetics of mast cell reconstitution in these animals.
C57BL/6 KitW-sh/KitW-sh mice were provided by Peter Besmer (Memorial Sloan-Kettering Institute, New York, NY, USA) . C57BL/6-TgN(ACTbEGFP) mice were purchased from Jackson Labs (Bar Harbor, ME, USA). These transgenic mice use a chicken actin promoter to drive expression of enhanced green fluorescent protein (GFP)3 in all cells . The UCSF Institutional Animal Care and Use Committee approved all experimental procedures.
For staining of mast cells in whole mounts of trachea by chloroacetate esterase (CAE) histochemistry, anaesthetized mice were first perfused through the ascending aorta for 2 min with 1% paraformaldehyde. Tracheas were removed, cut open lengthwise along the dorsal midline, stained overnight as described , and then whole-mounted in 100% glycerol. For other procedures, harvested tissues were washed in phosphate-buffered saline (PBS) and then fixed for 6–18 h in PBS containing 4% paraformaldehyde. Tissues were then embedded in paraffin and sectioned (5 μm). Prior to use, sections were de-paraffinized in xylene, hydrated through graded alcohols, and equilibrated in PBS. To visualize mast cells, tissue sections were incubated in 0.1% methylene blue for 10 s, rinsed with water, dehydrated in 100% ethanol, and mounted under coverslips with cytoseal 60 (Richard-Allan Scientific, Kalamazoo, MI, USA).
Right and left lungs were harvested, washed in PBS, snap-frozen in liquid nitrogen, and then pulverized using a mortar and pestle. Total RNA was extracted from the resulting powder using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and subsequently purified with the Rneasy mini kit and DNAse digestion (Qiagen, Valencia, CA, USA). The purified RNA was then used as a template for single-stranded cDNA synthesis using oligo (dT) primers and Superscript II reverse transcriptase (Invitrogen). Freshly synthesized cDNA was precipitated in sodium acetate (3m; 1/10 vol) and EtOH (2.5 vol), and subsequently coupled to N-hydroxysuccinimidyl esters of Cy5 dyes (CyScribe, Amersham Biosciences, Bucks, UK). For analysis purposes, all hybridizations were performed with a common reference pool, which was constructed using equal mixtures of total lung RNA derived from three strains of mice. Newly synthesized cDNA samples generated from the lung reference pool were similarly coupled to N-hydroxysuccinimidyl esters of Cy3 dyes (Amersham Biosciences). The labelled cDNA samples were individually purified using Qiaquick columns (Qiagen), and percent incorporation of fluorophores quantified via spectrophotometry. Equal amounts of labelled cDNA were then combined and competitive two-colour hybridizations were performed with spotted oligonucleotide (70-mer) arrays (Functional Genomics Core, University of California, San Francisco, CA, USA) using an automated hybridization system (GeneMachines, Genomic Solutions, Ann Arbor, MI, USA). After hybridization, each array was washed and imaged using an Axon GenePix 4000B scanner and GenePix 5.0 software (Union City, CA, USA). Differential gene expression was measured by comparing the signal intensity for each experimental sample labelled with Cy5 (red) to the standard lung reference pool labelled with Cy3 (green). Three independent replicates were performed within each group (i.e., Kit+/Kit+ and KitW-sh/KitW-sh). Microarray data were normalized and lists of differentially expressed genes were generated using the limmaGUI software package (http://bioinf.wehi.edu.au/limmaGUI/).
Bone marrow was harvested from GFP-transgenic C57BL/6-TgN (ACTbEGFP) mice and cultured in the presence of 10 ng/mL recombinant mouse IL-3 as described . Cells were cultured for 5 weeks, when cell populations consisted of >95% mast cells as assessed by the presence of metachromatic granules in toluidine blue-stained cells. Because these bone marrow-derived mast cells (BMMC) express cytosolic GFP, they can be detected by fluorescence microscopy. Ten million of these BMMC in 200 μL of PBS were injected via the tail vein into 5–7-week-old KitW-sh/KitW-sh mice. Mice were killed at intervals of between 1 and 28 weeks after reconstitution and tissues were harvested for analysis. In some experiments, BMMC were injected into KitW-sh/KitW-sh mice 2 weeks after splenectomy. Ears and peritoneums were locally reconstituted with mast cells by direct injection of BMMC (5 × 105 and 5 × 106, respectively) into these tissues.
Ten microlitres of heparinized blood, harvested from mice reconstituted with GFP+ BMMC, was applied to a slide and coverslipped. Fluorescent microscopy was then used to scan the blood for the presence of fluorescent cells.
Mast cells were counted in 5 μm sections of the left lung and the total numbers averaged for 4–7 mice/time-point. Each lung was processed in the same way and sectioned in the same orientation (i.e., longitudinally and medially). Numbers of mast cells in the spleen (n = 6 mice/time-point) were determined by randomly photographing 5 μm sections of spleen using the × 10 objective. Then the total numbers of mast cells were counted for each image and the area of tissue was determined using NIH ImageJ software (http://rsb.info.nih.gov/ij/download.html).
The line of KitW-sh mutation we used is in a C57BL/6 background. Homozygous KitW-sh/KitW-sh mice are fertile with 6–8 mice/litter. KitW-sh/KitW-sh mice are readily genotyped by coat colour because C57BL/6 Kit+/+ mice have a black coat (Fig. 1a), Kit+/KitW-sh mice have a white ‘sash’ around their midsection (Fig. 1b) and KitW-sh/KitW-sh mice have a white coat and black eyes (Fig. 1c). We also crossed KitW-sh/KitW-sh mice into the 129 background and found that 129 KitW-sh/Kit+ mice are brown with a white ‘sash’. Thus, the KitW-sh mutation can be crossed into backgrounds of mice with dark coat colours to generate strains of mast cell-deficient mice in backgrounds that differ from the mixed WB/B6 background of the commonly used KitW/KitW-v mice. This allows testing of the role of mast cells in disease models requiring specific genetic backgrounds of mice (e.g., 129).
The first step in determining whether KitW-sh/KitW-sh mice are useful for in vivo studies of mast cell function was to establish the extent of their mast cell deficiency. To this end, we used metachromatic (methylene blue) and enzyme histochemical (chloroacetate esterase) stains to identify mast cells in tissues of 10-week-old KitW-sh/KitW-sh mice. We found that they are devoid of mast cells in all tissues examined, including the skin (ear), tongue, trachea, lung, stomach, spleen, small intestine, mesentary, peritoneum, and inguinal lymph nodes (Fig. 2, and data not shown). However, our techniques readily detect mast cells in Kit+/Kit+ mice at all of these locations (Fig. 2 and data not shown). These findings indicate that KitW-sh/KitW-sh mice have a more severe mast cell deficiency than KitW/KitW-v mice, which retain small numbers (< 1% normal) of mast cells . Further, because KitW-sh/KitW-sh mice are profoundly mast cell-deficient, they can be used to test whether mast cells regulate specific biological processes by comparing phenotypes of KitW-sh/KitW-sh mice to those of Kit+/Kit+ mice, which contain normal numbers of mast cells.
Because Kit is expressed in several types of cells, we were concerned that reduced Kit expression in KitW-sh/KitW-sh mice might alter basal gene expression to a degree that they would not be useful for the in vivo study of mast cells. For this reason, and because we are interested in lung mast cells, we next sought to determine whether Kit (and by extension mast cells) regulate gene expression in normal mouse lungs. This possibility was tested using oligonucleotide gene arrays to compare relative levels of expression for 16 463 genes in KitW-sh/KitW-sh lungs vs. Kit+/Kit+ lungs. Interestingly, mast cell protease expression was undetectable on these arrays, suggesting that mast cells account for a small proportion of total RNA isolated from the lung. For each gene measured, we calculated moderated t-statistics  and P-values (adjusted using the Holm method). Using this approach, only one gene (Kit itself) reached statistical significance (adjusted P < 0.05). Using a less stringent cut-off (unadjusted P < 0.001 and at least a twofold change), only seven genes were differentially expressed in KitW-sh/KitW-sh lungs (Table 1). Of these seven genes, one was expressed at a higher level, and six at lower levels in KitW-sh/KitW-sh lungs. As expected , Kit expression was low and, indeed, manifested the greatest relative decrease in expression of all genes studied. These findings indicate that the reduction of Kit gene transcripts has relatively little effect on global gene expression in the lung.
Because the absence of mast cells in KitW-sh/KitW-sh mice is because of reduced Kit expression [18, 19], differences between KitW-sh/KitW-sh and Kit+/Kit+ mice may be explained either by reduced Kit expression or by the absence of mast cells . To differentiate between these two possibilities, the phenotype of KitW-sh/KitW-sh mice with and without reconstitution with mast cells must be compared because reconstitution can repair the mast cell deficiency but not reduced Kit expression in cells other than mast cells. Because KitW-sh/KitW-sh mice are much more useful for the in vivo study of mast cell biology if the mast cell deficiency can be corrected selectively, we next tested whether KitW-sh/KitW-sh mice can be reconstituted with mast cells by injection of in vitro-differentiated BMMC into tail veins or directly into a tissue of interest.
GFP+ and BMMC were used for reconstitution studies because they allow us to identify their presence in the blood of reconstituted mice. In pilot studies, we found that tail vein injection of 1.5−2 × 107 BMMC killed ~20% of the mice within 1 h, whereas injection of 107 BMMC was well tolerated. Why mice died is unknown. However, dying mice were noted to have laboured breathing, suggesting that they suffered from acute respiratory failure, which possibly was triggered by injection of histamine, serotonin, or other mast cell products. Thus, 107 BMMC is close to the upper limit of mast cells that can be safely injected into KitW-sh/KitW-sh mice. This number of cells was used in all subsequent studies. After intravenous injection of KitW-sh/KitW-sh mice with GFP–BMMC, we found that GFP+ cells circulate in the blood of most injected mice for 1 week after injection but disappear by 4 weeks (Table 2). Blood from control mice have no fluorescent cells. Mast cells begin to appear in most tissues by 4 weeks (Table 2). However, of the organs surveyed, the spleen is the only site to contain mast cells in 100% of injected animals at 4 weeks. Of the remaining organs that could be reconstituted by tail vein injection of BMMC, the lung and stomach are reconstituted in all mice by 12 weeks and liver is reconstituted by 28 weeks (Table 2 and Fig. 3). By contrast, reconstitution of mast cells in inguinal lymph nodes was more variable and we never achieved reconstitution of 100% of animals at any time-point.
Total numbers of mast cells found in reconstituted tissues increased as the time of reconstitution progressed. For example, spleens harvested 4 weeks after injection of BMMC averaged 8 ± 2 mast cells/mm2, at 12 weeks 85 ± 12 mast cells/mm2 and at 28 weeks 258 ± 33 mast cells/mm2 (× 13, × 141, and × 430 levels in Kit+/Kit+ spleen, respectively). Similarly, lungs harvested 4, 12, and 28 weeks after injection had 2 ± 1, 39 ± 10, and 245 ± 34 mast cells/lung section (× 0.19, × 3.6, and × 23.3 levels in Kit+/Kit+ lungs, respectively). These findings indicate that during the course of reconstitution, there is a critical time when tissues are reconstituted with numbers of mast cells similar to those found in Kit+/Kit+ mice. For the lung, this is 12 weeks after injection of BMMC because at this time all of the lungs are reconstituted (Table 1) and the mast cell density in reconstituted KitW-sh/KitW-sh mice is most similar to that of Kit+/Kit+ mice (10.5 ± 4 mast cells/lung section).
Peritoneal mast cells were rarely reconstituted (data not shown), and ear, tongue, and trachea mast cells were never reconstituted after tail vein injection of BMMC. However, local populations of mast cells could be reconstituted within 5 weeks with normal numbers of mast cells by injecting 5 × 105 BMMC directly into the ear or 5 × 106 BMMC into the peritoneum  (Fig. 3 and data not shown). These findings demonstrate that the microenvironment within the ear and peritoneum of KitW-sh/KitW-sh mice can support mast cells and suggest that failure of intravenous reconstitution is because of the absence of chemoattractants that recruit mast cells to these tissues or to the inability of circulating BMMC to penetrate the vasculature serving these sites.
As noted above, the spleen is not only the first organ to be reconstituted but it hosts greater numbers of mast cells than Kit+/Kit+ controls (Figs. 2g and and3c).3c). These observations suggested the possibility that mast cells mature in the spleen prior to migrating into tissues or that the spleen acts as a reservoir for mast cells following tail vein injection. To test these possibilities, we studied organ-specific reconstitution of splenectomized KitW-sh/KitW-sh mice. Eight weeks after tail vein injection of 107 GFP+ BMMC, 50% of splenectomized mice, compared with none of the animals with intact spleens had GFP+ cells in the blood (Table 2). Furthermore, a greater percentage of splenectomized KitW-sh/KitW-sh mice had mast cells in liver (100% vs. 63%) and inguinal lymph nodes (80% vs. 33%) than mice with intact spleens (Table 2). The remaining tissues were reconstituted to similar degrees in both groups 8 weeks after injection. These findings indicate that the spleen is not required for mast cell maturation after intravenous reconstitution but acts as a reservoir for injected mast cells and potentially slows their migration into liver and inguinal lymph nodes.
This study characterizes KitW-sh/KitW-sh mice, which may be useful for studying mast cell biology in vivo. In this regard, we found that KitW-sh/KitW-sh mice have a number of favourable attributes: (1) they are devoid of mast cells in all tissues, (2) there is little difference in basal gene expression in lungs of KitW-sh/KitW-sh vs. Kit+/Kit+ mice, consistent with the KitW-sh/KitW-sh chromosomal translocation affecting Kit expression in mast cells more so than other cell types, (3) KitW-sh/KitW-sh mice can be reconstituted with mast cells by intravenous or local injection of BMMC, and (4) KitW-sh/KitW-sh mice can be easily bred with mice of other backgrounds to generate mast cell-deficient mice for study of mast cells in mammalian biology.
Two requirements for using a particular strain of mice to study mast cell biology in vivo are that the mice lack mast cells and that the mast cell deficiency can be selectively repaired. KitW-sh/KitW-sh mice meet both these requirements. Previously, newborn KitW-sh/KitW-sh mice were demonstrated to have some skin mast cells, which disappear by 8 weeks . Herein, we show that 10-week-old KitW-sh/KitW-sh mice lack mast cells in all tissues. Because our methods only detect mast cells containing proteoglycans (methylene blue stain)  or active chymase (CAE stain) , it remains possible that immature mast cells or mast cell precursors, lacking metachromatic granules and chymase, are present in tissues of KitW-sh/KitW-sh mice. However, this is unlikely because if mast cells were present they would interfere with reconstitution, as has been shown in prior studies where injection of mast cells into the peritoneum of KitW/KitW-v mice suppressed subsequent recruitment of mast cell precursors . These results suggest that normal tissues can support only a limited number of mast cells. Further, even if immature mast cell precursors are present in KitW-sh/KitW-sh mice, it is unlikely that they function like the granulated mature mast cells of reconstituted KitW-sh/KitW-sh mice.
Prior work has shown that the lung has one of the highest levels of expression of stem cell factor (the ligand for Kit) in mice . While this may suggest that activation of the Kit receptor is important for normal lung homeostasis, the present study indicates that a marked reduction of Kit expression in lungs of KitW-sh/KitW-sh mice has little impact on the expression of other lung genes. This indicates that Kit has little direct influence on gene expression in normal lungs, that other genes compensate for Kit in its absence, or that low levels of Kit expression are sufficient to prevent major changes of other genes. The lack of change in gene expression in KitW-sh/KitW-sh lungs also indicates that mast cells have little influence on gene expression in normal lungs. This finding supports the concept that the main function of tissue mast cells is to identify and co-ordinate a response to changes in their tissue microenvironment, such as tissue invasion by parasites and other pathogens, rather than to influence the composition of their microenvironment during normal conditions.
Mast cells are innate immune cells that are widely distributed in most tissues, where they are early sensors of change or infection [2, 29]. When changes are detected, mast cells are activated to release mediators that modify the behaviour of adjacent cells  or recruit inflammatory cells to the site of perceived invasion or injury [12, 15, 30]. Because mast cell-activated cells, which are more abundant than mast cells themselves, amplify mast cell signals by releasing other soluble mediators or directly mediate the pathologic change, comparatively few mast cells may be required to co-ordinate pathologic responses. Furthermore, if a strain of mouse is deficient rather than devoid of mast cells (as is true of KitW/KitW-v mice) , the residual mast cells may mount a sufficiently strong response to a pathologic process such that the mast cells role may go unrecognized. Because KitW-sh/KitW-sh mice have no mast cells, they are potentially valuable to use in disease models where a role for mast cells is suspected, including models where no role for mast cells has been identified using KitW/KitW-v mice.
Although both KitW-sh/KitW-sh and KitW/KitW-v mice can be reconstituted with mast cells, the kinetics of reconstitution in KitW-sh/KitW-sh mice favour their use for in vivo studies. First, KitW-sh/KitW-sh mice have more complete reconstitution of tissues with mast cells (most notably the reconstitution of inguinal lymph nodes). Second, reconstitution is more durable in KitW-sh/KitW-sh than KitW/KitW-v mice . This is best demonstrated by comparing reconstitution kinetics of lungs of KitW/KitW-v mice, which is transient , and of KitW-sh/KitW-sh mice, which is more durable. Similarly, splenic mast cell density decreases as reconstitution progresses in KitW/KitW-v mice  but not in KitW-sh/KitW-sh mice. By exploiting favourable reconstitution kinetics, the KitW-sh/KitW-sh mice can be used to test the roles of mast cells and mast cell products in mouse models of human disease.
KitW-sh/KitW-sh mice can be used for a number of novel approaches, including some that cannot be used using KitW/KitW-v mice, for the in vivo study of mast cells and mast cell products. For example, KitW-sh/KitW-sh mice can be used in the classic approach to test whether mast cells play a role in a specific disease process . This is achieved by comparing responses of Kit+/Kit+ mice to those of KitW-sh/KitW-sh mice and of KitW-sh/KitW-sh mice reconstituted with Kit+/Kit+ BMMC (Fig. 4d). They can also be used to test the role of a specific mast cell product by generating mast cell-specific knockout mice. Such mice are created by reconstituting KitW-sh/KitW-sh mice with BMMC cultured from mice having the gene of interest deleted (Fig. 4b) . Because KitW-sh/KitW-sh mice have normal expression of this gene, the reconstituted mice (Fig. 4e) have normal expression of the gene in all cells except mast cells, where it is absent. As controls, KitW-sh/KitW-sh mice can be reconstituted with wild-type BMMC (Fig. 4d). These mice are phenotypically similar, but not identical, to Kit+/Kit+ mice (Fig. 4a). By comparing responses of these two groups in a disease model of interest, one can identify mast cell-specific roles for a gene expressed in multiple cell types. Finally, KitW-sh/KitW-sh mice can be used to generate knockout mice lacking mast cells by crossing KitW-sh/KitW-sh mice with the knockout mouse of interest (Fig. 4). This mouse can then be reconstituted with wild-type BMMC to create a mouse expressing the gene of interest exclusively in mast cells (Fig. 4f). These mice can then be used to test whether sole expression of the gene of interest in mast cells is sufficient to explain the observed biological functions of the gene, or to identify mast cell-specific roles for gene products that may be compensated for by other cells when absent in mast cells.
In summary, these data show that KitW-sh/KitW-sh mice are profoundly mast cell-deficient, have normal expression of lung genes, and that mast cell populations can be restored in most tissues in a time-dependent manner by systemic or local injection of wild-type BMMC. These findings distinguish them as a useful model for examining mast cell biology in vivo.
This work was supported in part by grants HL-72301, HL-04055, HL-075026 and HL-024136 from the National Institutes of Health, the American Lung Association of California, and the UCSF Sandler Center for Basic Research in Asthma.