Generation of a Tenascin-C-CreER2 Knockin Mouse Line for Conditional DNA Recombination in Renal Medullary Interstitial Cells

Renal medullary interstitial cells (RMIC) are specialized fibroblast-like cells that exert important functions in maintaining body fluid homeostasis and systemic blood pressure. Here, we generated a RMIC specific tenascin-C promoter driven inducible CreER2 knockin mouse line with an EGFP reporter. Similar as endogenous tenascin-C expression, the reporter EGFP expression in the tenascin-C-CreER2+/− mice was observed in the inner medulla of the kidney, and co-localized with COX2 but not with AQP2 or AQP1, suggesting selective expression in RMICs. After recombination (tenascin-C-CreER2+/−/ROSA26-lacZ+/− mice + tamoxifen), β-gal activity was restricted to the cells in the inner medulla of the kidney, and didn't co-localize with AQP2, consistent with selective Cre recombinase activity in RMICs. Cre activity was not obvious in other major organs or without tamoxifen treatment. This inducible RMIC specific Cre mouse line should therefore provide a novel tool to manipulate genes of interest in RMICs.


Introduction
The renal medullary interstitial cells (RMIC) are a population of specialized stroma-like cells in renal medulla. These cells, characterized by abundant cytoplasmic lipid droplets, are arranged in rows with their long axis perpendicular to adjacent tubules and vessels [1,2]. In addition to supporting the renal structure, RMICs have been demonstrated to play important roles in the maintenance of body fluid homeostasis and normal systemic blood pressure. Animal studies show that chemical ablation of RMICs with BEA leads to systemic hypertension [3]. RMIC cyclooxygenase-2 (COX2) expression is also suggested to play important roles in renal response to stress, such as sodium loading and water deprivation [4][5][6][7]. To better understand the molecular basis of the physiological roles of RMICs, a Cre-recombinase/ LoxP-based RMIC-specific gene deletion could be a powerful approach to investigate the significance of specific genes in RMICs. Here, we report an inducible RMIC-specific Crerecombinase line under the control of endogenous tenascin-C promoter.

Animals
Ethics statement: Mice used in the present study were maintained in the animal facility of Vanderbilt University Medical Center, where they were housed in a constant temperature room with a 12-hour dark/12-hour light circle, and allowed free access to standard rodent chow and water. All animal studies were approved by the Institutional Animal Care and Use Committees of Vanderbilt University. C57Bl/6J wild type mice were obtained from Jackson Laboratories. ROSA26-lacZ reporter mice and genotyping methods were previously reported [8].

Construction of the targeting vector
The homogenous recombination arms were derived from a BAC library (RPCI-22 mouse BAC library, Invitrogen). The targeting vector was assembled in pBlue and contains a 4 kb 59 arm, an inducible CreER2, an IRES-EGFP (Clontech), a FRT flanked PGK-neo selection cassette and a 2 kb 39 arm. The CreER2 was made from CreER version 1 (kindly provided by Dr. Andrew P. McMahon) via site directed mutagenesis according to literature [9].

Screen ES cells by Southern Blot
Run digested DNA in 1% agarose gel. Take picture of agarose gel to be blotted with phosphorescent ruler lined up along side it, such that the ruler is lined up with the top of the wells. Depurinate the DNA in the gel by rocking it in 0.25 M HCl for exactly 10 min, and alkaline denature the gel in 0.4 M NaOH 3615 min. While shake the gel in 20XSSC for 5 min, set up the blot from bottom to top: 1) A large dish filled with 206SSC with glass plate on top of it to rest the gel. 2) Two pieces of wick-blotting paper cut to the width of the gel and length such that the wick is in contact with the bottom of the dish. Wet the wick with 206SSC and smooth out the bubbles gently with a glass pipette. 3) Agarose gel, turned upside down, with a nick in the bottom right hand corner for orientation. Smooth out bubbles. Place plastic wrap to cover the entire gel and cut out the wrap around the gel such that the blot will not short-circuit. 4) Hybond N+ nylon membrane cut to the exact size of the gel, with a nick in the corner for orientation. Wet membrane with dH 2 O, place on top of gel and smooth it out. 5) Four pieces of blotting paper cut to size of the gel. Wet the first blotting paper with 206SSC, put on top of blotting paper, and smooth out. Put other three on top. 6) Glass plate and additional weight to keep blot in place. Transfer overnight. The next day, take apart blot being careful not to remove the membrane from the gel. Take off the gel and membrane together, and flip. Use a pencil to mark the wells. Auto X-link membrane with Stratalinker. Prehybridize the membrane at 65uC for at least 1 h, add probes and hybridize at 65uC overnight. Wash the membrane 3610 min at 65uC and then expose at 280uC for 4-7 days. Primers used for synthesizing 59 probe were: 59-TAGAGCAGGTGGTCCCA-AACAT-39 and 59-CCAGGAGCCAGGAAATAGCCTTA-39. Primers used for synthesizing 39 probe were: 59-GATGACGAC-TACACTGGGGAA-39 and 59-ACTGGGGCACCTTTGCT-CTT-39.

PCR Genotyping
Mice were genotyped using genomic DNA isolated from tail biopsy. PCR primers used to amplify across the region where the 59 CreER2 was inserted were: sense: 59-GGGGGCAAGAAGG-CAAAAAT-39; antisense-1: 59-GTTCTGCGGGAAACCATTT-39; antisense-2: 59-TCTCGCTTGTGCCTGATGAT-39. Primer pair of sense and antisense-1 gave a band of ,430 bp for a targeted allele and no band for a wild type allele. Primer pair of sense and antisense-2 gave a band of ,300 bp for a wild type allele and no band for a targeted allele.

In Situ Hybridization
In situ hybridization was performed as previously described [5]. Briefly, prior to hybridization, tissue sections were deparaffinized, refixed in 4% paraformaldehyde, treated with proteinase K (20 mg/ml), washed with PBS, refixed in 4% paraformaldehyde,  and treated with triethanolamine plus acetic anhydride (0.25% vol/vol). Finally, sections were dehydrated with 100% ethanol. 35 S-labeled antisense and sense riboprobes from mouse tenascin-C were hybridized to the sections at 55uC for 18 h. After hybridization, the sections were washed at 65uC once in 5X SSC (1X SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0) plus 10 mM b-mercaptoethanol (BME), then once in 50% formamide, 2X SSC, and 100 mM BME for 30 min. After an additional two washes in 10 mM Tris, 5 mM EDTA, 500 mM NaCl (TEN) at 37uC, the sections were treated with RNase A (10 mg/ml) at 37uC for 30 min, followed by another wash in TEN at 37uC. Sections were then washed twice in 2X SSC and twice in 0.1X SSC at 65uC. Slides were dehydrated with graded ethanol containing 300 mM ammonium acetate. Photomicrographs were taken from slides dipped in K5 emulsion (Ilford Ltd., Knutsford, Cheshire, United Kingdom) diluted 1:1 with 2% glycerol/water and exposed for 7 days at 4uC. After development in Kodak D-19, slides were counterstained with hematoxylin. Photomicrographs were taken with a Zeiss Axioskop microscope.

Immunofluorescent Staining
The kidney tissues were first fixed in 4% paraformaldehyde and then incubated in 30% sucrose overnight. Cryostat sections (5 mm) were blocked with 3% normal donkey serum for 20 min, and then incubated with primary antibody for 60 min at room temperature. After washing in PBS, the sections were incubated in Cy2 or Cy3 conjugated anti-IgG secondary antibody (Jackson ImmunoResearch Laboratories, 1:200) for 30 min. Sections were washed again in PBS for 5 times and microscopy was performed with a Zeiss Axioskop and spot-cam digital camera (Diagnostic Instruments) or confocal microscope (Zeiss LSM510). The primary antibodies used for immunofluorescence studies were: antiaquaporin-1 (AQP1) antibody (Santa Cruz mouse monoclonal, 1:100), anti-aquaporin-2 (AQP2) antibody (Santa Cruz goat polyclonal, 1:400), anti-COX2 antibody (Cayman rabbit polyclonal, 1:500).

X-gal Staining
Frozen sections were freshly cut and fixed with cold formalin for 10 minutes at 4uC. After 3 changes of PBS wash for 5 minutes each, the slides were rinsed in distilled water and washed in b-gal wash buffer (0.1 M Phosphate buffer, 2 mM MgCl 2 , 5 mM EGTA, 0.01% Sodium Deoxycholate, 0.02% NP40, pH 7.3) for 10 minutes at RT. Then the slides were moved to X-gal stain solution (5 mM Potassium Ferrocyanide, 5 mM Potassium Ferricyanid, 1 mg/ml X-gal in b-gal wash solution) in a humidified chamber for 24 hours at 37uC. After 2 changes of PBS wash for 5 minutes each, the slides were rinsed with distilled water and mounted with aqueous mounting medium.

Results and Discussion
Our microarray study identified tenascin-C as one of RMIC specific gene products in the kidney (Table 1). Selective expression of tenascin-C mRNA in the renal medullary interstitial cells was further confirmed by in situ hybridization ( Figure 1). The data are consistent with published studies showing tenascin-C expression in the renal medullary stroma of the adult mice [10][11][12].
Tenascin-C is one of the four members of the tenascin family (tenascin-C, -X, -R, -W) that encode glycoproteins found in the extracellular matrix [13][14][15]. Tenascin-C is abundantly expressed in the mesenchyme surrounding developing epithelia in virtually every organ during development. In contrast, its expression in normal adult tissues is very low or undetectable and highly restricted in stromal cells.
Considering that the mouse tenascin-C gene has a 31 kb intron1 (Figure 2a) that may contain important transcriptional regulatory elements, and heterozygous tenascin-C knockout mice have normal phenotypes [16,17], a ''knock-in'' strategy was used to create a Crerecombinase mouse driven by endogenous tenascin-C promoter. The Cre transgene was engineered immediately behind the translation start site (ATG) of the tenascin-C gene to ensure that the expression pattern of Cre is similar to that of the endogenous tenascin-C gene. Mouse genomic tenascin-C gene was obtained by a BAC library screen. The targeting construct included a 4 kb 59 arm homogenous to 59 region upstream of the tenascin-C gene translation start codon in exon 2, an inducible CreER2, an IRES (internal ribosome entrance site)-EGFP (enhanced green fluorescence protein), a PGK-Neo selection cassette flanked by FRT, and a 2 kb 39 homogenous arm. The IRES-EGFP cDNA in the targeting construct allows for detecting site of Cre expression in mice carrying the transgene. The targeting construct was linearized by Ahd I and introduced into E14 129SvJ ES cells with established protocols [18] at Vanderbilt Mouse/ESC Shared Resource. Neo-resistant ES cell clones were screened by southern blot, and four correctly targeted clones were identified (Figure 2b). Two targeted clones (3D4 and 5C3) were picked for blastocyst injection. Germline transmission of the transgene was obtained from both of these two lines.
Genotyping was achieved through PCR of tail DNA (Figure 2c). Mice heterozygous for the tenascin-C-CreER2 allele were viable and of normal size without significant developmental or functional abnormalities.
To test if the ectopic genes are expressed in the same pattern as endogenous tenascin-C gene, EGFP expression of the mice heterozygous for the tenascin-C-CreER2 allele was examined. Abundant EGFP expression was observed in the inner renal medulla but not in the cortex or outer medulla (Figure 3a). The EGFP-positive cells are arranged in rows, a typical morphology of renal medullary interstitial cells. Co-labeling studies showed that the EGFP did not co-localize with renal collecting duct marker AQP2 or descending thin limb marker AQP1, but co-localized with renal medullary interstitial cell specific COX2 (Figure 3b). Figure 5. b-gal activity in major organs of tenascin-C-CreER2 +/2 /ROSA26-lacZ +/2 mice with tamoxifen injection. Pictures of X-gal staining shows no significant b-gal activity in the heart, spleen, liver, skin and cerebrum but in certain cells lining between different layers of cerebellum (indicated by yellow arrows) of tenascin-C-CreER2 +/2 /ROSA26-lacZ +/2 mice with tamoxifen injection. Scale bar, 100 mm. doi:10.1371/journal.pone.0079839.g005 This pattern is consistent with selective expression of EGFP in the renal medullary interstitial cells. No obvious EGFP was observed in other organs or tissues of adult mice examined including the heart, spleen, liver, skin and brain (data not shown).
ROSA26-lacZ reporter mice were used to monitor Cre activity by virtue of b-galactosidase (b-gal) expression that is dependent on Cre recombinasemediated release of a stop codon before b-gal cDNA. Tamoxifen (1.5 mg/25 g bw/day) was administered to tenascin-C-CreER +/2 /ROSA26-lacZ +/2 mice at the age of 8 wks for continuous two weeks. Three days after the last injection, b-gal activity in major organs was examined by X-gal staining. The recombination reporter b-gal activity was highly restricted to the inner medulla of the kidney but not in the renal cortex (Figure 4a). Co-labeling with renal collecting duct marker AQP2 showed that the b-gal positive cells were localized in the inner medullary interstitium between collecting (Figure 4b). No significant b-gal activity was observed in the heart, spleen, liver, skin and cerebrum of the brain ( Figure 5). However, there was detectable b-gal activity in some cells between different layers of cerebellum ( Figure 5), consistent with previous studies showing tenascin-C expression in the forebrain and the cerebellum of the adult mice [19]. In addition, no b-gal activity was detected without tamoxifen treatment ( Figure 6), suggesting no leak of the inducible Cre recombinase activity of the tenascin-C-CreER2 +/2 /ROSA26-lacZ +/2 mice.
In summary, we have generated a tenascin-C-CreER2-IRES-EGFP knockin mouse line, and demonstrated that this mouse line has EGFP expression and inducible Cre activity predominantly in the medullary interstitial cells of the kidney. The tenascin-C-CreER2 knockin mouse is therefore able to be utilized for introducing DNA recombination specifically in the renal medullary interstitial cells of the kidney. By inactivating genes of interest or driving ectopic gene expression specifically in renal medullary interstitial cells though inducing LoxP based recombination, this mouse line will greatly facilitate our exploration of the physiological roles of renal medullary interstitial cells. Furthermore, as elevated tenascin-C expression is reported to play important roles in pathological conditions such as inflammation, infection and cancer in multiple organs including the kidney [12,14,20], this tenascin-C-CreER2 knockin mouse may also provide a valuable tool.