Generation and Analysis of Serine Protease Inhibitor Kazal Type 3-Cre Driver Mice

Abstract

Serine protease inhibitor Kazal type 1 (SPINK1; mouse homologue Spink3) was initially discovered as a trypsin-specific inhibitor in the pancreas. However, previous studies have suggested that SPINK1/Spink3 is expressed in a wide range of normal tissues and tumors, although precise characterization of its gene expression has not been described in adulthood. To further analyze Spink3 expression, we generated two mouse lines in which either lacZ or Cre recombinase genes were inserted into the Spink3 locus by Cre-loxP technology. In Spink3^lacZ mice, β-galactosidase activity was found in acinar cells of the pancreas and kidney, as well as epithelial cells of the bronchus in the lung, but not in the gastrointestinal tract or liver. Spink3^cre knock-in mice were crossed with Rosa26 reporter (R26R) mice to monitor Spink3 promoter activity. In Spink3^cre;R26R mice, β-galactosidase activity was found in acinar cells of the pancreas, kidney, lung, and a small proportion of cells in the gastrointestinal tract and liver. These data suggest that Spink3 is widely expressed in endoderm-derived tissues, and that Spink3^cre knock-in mice are a useful tool for establishment of a conditional knockout mice to analyze Spink3 function not only in normal tissues, but also in tumors that express SPINK1/Spink3.

Introduction

Serine protease inhibitor Kazal type 1 (SPINK1), also known as pancreatic secretory trypsin inhibitor (PSTI), was originally isolated from the pancreas [5]. In mice, the homologous gene is designated as Spink3(serine protease inhibitor Kazal type 3) [10]. SPINK1 is produced in acinar cells of the exocrine pancreas, and is packaged with digestive enzymes into granules that are secreted into the pancreatic duct [11]. SPINK1/Spink3 covalently binds to erroneously activated trypsin in the pancreas to form an inactive and stable complex to prevent acute pancreatitis, a major inflammatory disorder of the pancreas [11]. The SPINK1 gene is a candidate gene of hereditary pancreatitis, although its pathogenesis is unknown [18]. In addition to the pancreas, SPINK1 has been subsequently identified in mucus-producing cells throughout the gastrointestinal tract [4] and in a range of other tissues including the lung, liver, kidney, ovary, breast, and the collecting tubules and transitional epithelium of the renal pelvis in humans [7]. Interestingly, SPINK1 was isolated from urine of ovarian cancer patients, and reported as a tumor-associated trypsin inhibitor (TATI) [16]. Increased expression of SPINK1 protein has been reported in various cancers such as lung, colon, liver, and prostate cancers, and is associated with poor survival of patients [11]. However, the role of SPINK1/TATI in normal tissues and malignant tumors is unknown. We previously showed that excessive autophagy is induced in Spink3 knockout mice, and that Spink3 is essential to maintain the exocrine integrity of the pancreas and possibly acts as a growth factor for pancreatic acinar cells [10]. Spink3 knockout mice die within 2 weeks after birth, making it difficult to monitor the long-term effects of Spink3 deficiency.

To gain an insight into its function, we previously analyzed Spink3 expression profiles by in situ hybridization and lacZ expression during embryonic development of a Spink3 knock-in (Spink3^lacZ) mouse, in which the lacZ gene had been inserted into the Spink3 locus [17]. The lacZ gene was first observed in the foregut, midgut, and hindgut at 9.5 days post-coitus (dpc). In the pancreas, Spink3 mRNA was detected at 11.5 dpc prior to formation of the typical shape of the exocrine structure of the pancreas. After differentiation of the intestinal tract, lacZ expression was observed in the large intestine at 11.5 dpc, followed by expression in the small intestine at 13.5 dpc before the appearance of digestive enzymes in the intestine. Spink3 was also expressed in other tissues including mesonephric tubules and the urogenital ridge, the genital swelling, ductus epididymis, and seminal vesicles. These data suggest that Spink3 plays important roles in proliferation and/or differentiation of various cell types during development [17].

As the Spink3 gene is expressed in several tissues, the Cre driver mouse in which the Cre gene is expressed under the control of Spink3 promoter/enhancer will be a useful tool to produce conditional knockout mice. In this study, we generated the Spink3^cre mouse, in which the Cre gene had been inserted into the mouse Spink3 locus using the exchangeable gene targeting method [1]. Then we compared the expression pattern between Spink3^lacZ and Spink3^cre;R26R adult mice by X-gal staining. We showed that Spink3^cre mice are useful and unique tool for generating various conditional knockout mice.

Methods

Mice

In all experiments, 8-week-old mice were used. C57BL/6J mice were purchased from CLEA Japan (Tokyo, Japan). All mouse experiments were performed in accordance with the Declaration of Helsinki and were approved by the Kumamoto University Ethics Committee for Animal Experiments.

Northern and Western blot analyses

Total RNA was isolated from each mouse organ with Sepasol (Nacalai Tesque, Kyoto, Japan). For Northern blot analysis, 10 µg of RNA was fractionated by 1.4% agarose gel electrophoresis. Filter-bound RNA was sequentially hybridized with a digoxigenin-labeled RNA probe (full-length Spink3 cDNA). For Western blot analysis, each organ was homogenized in lysate buffer (50 mmol/L HEPES, pH 7.4, 150 mmol/L NaCl, 0.1% Triton X-100, 10% glycerol, 1 mmol/L NaF, 2 mmol/L sodium orthovanadate, 1 mmol/L ethylenediaminetetraacetic acid, and protease inhibitor cocktail [1:100 dilution; Sigma-Aldrich, Tokyo, Japan]). Protein extracts (15 µg per lane) were applied to 5–20% gradient polyacrylamide gels (E-T/R/D520L, ATTO, Tokyo, Japan) for electrophoresis, and then transferred to an Immobilon polyvinylidene difluoride filter (Millipore, Darmstadt, Germany). Primary rabbit antibodies against the following antigens were used at the indicated dilutions: Spink3 (1:1000; Transgenic Inc., Kobe, Japan) and actin (1:1000; Sigma-Aldrich). An anti-rabbit immunoglobulin G antibody conjugated to horseradish peroxidase (1:2500 dilution; Amersham, Buckinghamshire, UK) was used for detection.

Generation of the Spink3^lacZ mouse strains

Generation of Spink3^lacZ mice has been described previously [17].

Construction of the replacement vector loxJTZ17-NLS-Cre-lox2272-MC1-DT-A

The loxJTZ17 [3] fragment was inserted at the 5′ end of nuclear localization signal (NLS)-Cre-poly (A) (pA) fragment. A phosphoglycerate kinase-1 (PGK) promoter-driven puromycin resistance gene (PAC)-pA cassette was inserted at the 3′ end. The lox2272 sequence [1] was introduced into the multiple cloning site at the 3′ end of the PGK-PAC-pA cassette. Finally, a polyoma enhancer/herpes simplex virus thymidine kinase promoter (MC1)-driven diphtheria toxin A chain gene (DT-A)-pA cassette, which is used for negative selection, was cloned to the 3′ end of the lox2272 fragment, resulting in construction of the loxJTZ17-NLS-Cre-lox2272-MC1-DT-A plasmid [2]. Thus, this plasmid contained loxJTZ17, Cre fused to a SV40 large T antigen NLS and an intron, which include splice donor (SD) and splice acceptor (SA), and a pA signal derived from SV40, lox2272, and MC1-DT-A-pA.

Establishment of the Spink3^cre embryonic stem (ES) cell line

The Spink3^+/− ES cell line was obtained by knocking-in the PGK-neomycin resistance gene fragment flanked by mutant lox sites, lox71 and lox2272 into the exon 1 of Spink3 gene (Fig. 2A) and maintained as described previously [10]. The Cre expression vector, pCAGGS-Cre, was constructed by inserting the Cre fragment from pBS185 (Life Technologies, Tokyo, Japan) into pCAGGS [8].

Fig. 2.

Targeting strategy for lacZ and Cre knock-in mice. (A)LacZ and Cre recombinase genes were inserted into the Spink3 locus by Cre-loxP technology. ES cells carrying Spink3 targeted allele were first created by knocked-in the PGK-neo expression cassette into the exon 1 of Spink3 gene. These ES cells were then transfected with replacement vector for lacZ or Cre together with Cre expression vector. Closed triangles are PCR primers (S1 and A1) for detecting the Cre replaced clones. Closed box indicate Cre probe for Southern blot analysis. PstI-digested genomic DNA from ES cells was used. Open triangles are FLP recombination target (FRT) sites. SD; splice donor, and SA; splice acceptor. (B) PCR (left panel) and Southern blot (right panel) analysis of DNA from Cre replaced clones. R; replaced clone, N; not replaced clone, M; size marker.

The Spink3^+/− ES cells (1 × 10⁷) were co-electroporated with both pCAGGS-Cre and loxJTZ17-NLS-Cre-lox2272-MC1-DT-A vetcors (each 20 µg) using a Bio-Rad Gene Pulser (Bio-Rad, Tokyo, Japan), and then cultured in medium supplemented with 2 µg/ml puromycin (Sigma-Aldrich) for 48 h. Selection was performed for 5 days, and then colonies were picked, transferred to 48-well plates, and expanded for cryopreservation. The puromycin-resistant colonies were analyzed by Southern blot and polymerase chain reaction (PCR) to select ES cell lines with successful integration of cre. DNA from the puromycin-resistant colonies was digested with PstI and was used to detect replaced alleles by Cre probe. PCR analysis to detect the replaced alleles, was performed with the following primers: S1,5′-agttcttctggcttttgcaccc-3′, and A1, 5′atagtcgacttcttctttggttttcggg-3′’. This primer set is expected to give PCR products with a size of 1.2kbp. PCR consisted of an initial denaturation cycle at 94°C for 5 min, followed by the 30 cycles consisted of denaturation at 94°C for 20 s, annealing at 55°C for 30 s, and elongation at 68°C for 2 min. Five replaced ES cell clones were obtained among 30 puromycin-resistant clones analyzed (Fig. 2B). Positive clones were aggregated with ICR (Sea:ICR, purchased from Kyudo Co., Ltd., Tosu, Japan) morula according to a protocol described previously [10]. Germ-line transmission was observed in two mice after mating with C57BL/6J mice.

X-gal staining for β-galactosidase (β-gal) activity

We dissected tissues from Spink3^+/+;R26R, as a control, Spink3^lacZ, and Spink3^cre; R26R mice at 8 weeks of age. For whole organ staining, the method has been described previously [17]. For section staining, optimum cutting temperature (OCT) (Sakura, Tokyo, Japan)-embedded tissues were cut using a cryostat to prepare 60 µm-thick sections. The sections were fixed with 1% formaldehyde, 0.2% glutaraldehyde and 0.02% NP-40 for 30–60 min at 4°C. Fixed specimens were rinsed with phosphate buffered saline (PBS) three times, and then 0.1% Triton X-100 in PBS three times. The specimens were then immediately incubated at 37°C for 12–16 h, depending on the rate of the color reaction, in an X-gal staining solution containing 5 mM potassium hexacyanoferrate II, 50 mM potassium hexacyanoferrate III, 2 mM MgCl₂, and 0.5% (w/v) X-gal (4-chloro-5-bromo-3-indolyl-β-D-galactopyranoside) (Nacalai Tesque) in PBS [17]. Finally, the samples were washed twice with PBS and post-fixed in 15% formalin.

Results and Discussion

We first analyzed Spink3 expression at both mRNA and protein levels in 8-week-old C57BL/6J male mice. In Northern blot analysis, Spink3 mRNA was strongly expressed in the pancreas, small intestine, and kidney, but not in the brain, heart, lung, liver, or spleen (Fig. 1). In Western blot analysis, Spink3 was expressed strongly in the pancreas, and slightly in the small intestine, large intestine, and kidney. In the brain, heart, lung, liver, and spleen, there was no Spink3 expression (Fig. 1).

Fig. 1.

Expression pattern of Spink3 in the adult mouse. Northern (upper panel) and Western (lower panel) blot analysis of Spink3 in an 8-week-old C57BL/6J mouse. 18S; the small ribosomal subunit.

To study cell type-specific expression of Spink3, we generated knock-in mice expressing lacZ (Spink3^lacZ) or the Cre recombinase gene (Spink3^cre) driven by the endogenous Spink3 promoter by Cre-loxP recombination technology (Fig. 2A). These two heterozygous mouse lines appeared to be normal and were fertile. Spink3^cre knock-in mice were crossed with Rosa26 reporter (R26R) mice [15] to produce Spink3^cre;R26R mice with which Spink3 promoter activity can be monitored. We detected the activity of β-gal by X-gal staining. Whole mount staining of Spink3^lacZ mice showed β-gal activity only in the pancreas, but not in the liver (Fig. 3A). In contrast, Spink3^cre; R26R mice showed β-gal activity not only in the pancreas, but also in some liver cells in a mosaic pattern. Pancreatic acinar cells (exocrine pancreas), which are the primary source of digestive enzymes, account for nearly 90% of the pancreatic mass. Consistent with the known distribution of endogenous Spink3 expression, X-gal staining in Spink3^lacZ mice at 8-week-old was exclusively localized to pancreatic acinar cells, and not in duct or endocrine islets cells (Fig. 3B). In the pancreas of Spink3^cre; R26R mice, X-gal staining was localized to pancreatic acinar cells, but not in duct cells. Interestingly, there were some X-gal positive endocrine cells in Spink3^cre; R26R mice (Fig. 3B). This observation and our previous data [17] indicated that Spink3 was expressed not only in immature and mature acinar cells, but also in some immature endocrine cells. In the liver, there were some X-gal-positive liver cells in Spink3^cre; R26R mice, but not in Spink3^lacZ mice. Although Spink3 was not detected in the liver under physiological conditions, Kobayashi et al. reported that the concentration of human SPINK1 protein is higher in the livers of adult-onset type II citrullinemia patients, which is caused by a deficiency of argininosuccinate synthetase protein in the liver, compared with that in the normal control [6]. Ohmachi et al. reported that SPINK1 mRNA is highly expressed in hepatic cancer [9]. In addition, SPINK1 has been suggested to be an acute-phase reactant in humans, which is induced by inflammatory cytokines such as interleukin (IL)-1 and IL-6 [19]. Thus, Spink3 expression may be induced in the liver by inflammation or liver disease. In addition, we previously showed that Spink3 was widely expressed in digestive organs during development [17]. Taken together, it was possible that Spink3 was expressed in some hepatic cells during embryonic development.

Fig. 3.

β-gal activity of the pancreas and liver. Whole mount (A) and tissue section (B) X-gal staining of Spink3^+/+;R26R, Spink3^lacZ, and Spink3^cre;R26R mice. Dashed line indicates the pancreas (Pa). Li; liver, St; stomach, Du; Duodenum, Sp; spleen.

Previous reports have shown that SPINK1 protein and mRNA are expressed in mucus-producing cells of the normal gastrointestinal tract [4, 7, 13]. SPINK1 may act as a protease inhibitor to protect epithelial cells from some proteases. Our previous data also showed that lacZ is strongly expressed in epithelial cells of both the small and large intestines in Spink3^lacZ mice from 13.5 to 17.5 dpc [17]. However, in this study, β-gal activity was not detected in the gastrointestinal tracts of adult Spink3^lacZ mice (Figs. 4A and B). On the other hand, in adult Spink3^cre; R26R mice, spotty β-gal activity was detected in the digestive tract from the stomach to the large intestine (Figs. 4A and B). β-gal expression in adulthood can be explained as follows. In Spink3^cre; R26R mice, Cre is expressed during embryonic development, leading to β-gal expression. Once the recombination occurs, β-gal gene expression continues under the R26 promoter.

Fig. 4.

β-gal activity in the gastrointestinal tract. β-gal activity was not detected in the gastrointestinal tracts of Spink3^+/+;R26R (control) or Spink3^lacZ mice (whole staining: A, section staining: B). However, in Spink3^cre; R26R mice, β-gal activity was detected in both crypt and tip compartments of the small intestine.

In the kidney, Spink3 mRNA is expressed mainly within the medulla area and in a portion of the cortex during development [17]. In adult mice, β-gal activity is detected in the outer zone of the medulla. In addition, we detected β-gal activity in the cortex of the kidney in Spink3^cre; R26R mice (Fig. 5). Although Spink3 is expressed from the embryonic stage to adulthood in the kidney, the role of Spink3 is unknown. More precise histological analysis of Spink3 expression needs to be performed using specific molecular markers to analyze the role of SPINK1/Spink3 in the kidney.

Fig. 5.

β-gal activity in the kidney. In the kidney, β-gal activity was detected in the outer zone of the medulla in Spink3^lacZ mice. β-gal activity was also detected in the cortex of the kidney in Spink3^cre;R26R mice.

The expression and role of SPINK1/Spink3 have never been examined in the lung. We were unable to detect discernible lacZ expression or Spink3 mRNA in the lung. Northern and Western blot analyses revealed that Spink3 mRNA or protein was not expressed in the adult mouse lung (see Fig. 1). However, a low level of β-gal activity was detected only in the epithelium of the trachea, but not in the lung acini of Spink3^lacZ mice (Figs. 6A and B). In Spink3^cre; R26R mice, β-gal activity was detected not only in epithelial cells of the trachea, but also in lung acini.

Fig. 6.

β-gal activity of the lung. In the lung, β-gal activity was detected only in the epithelium of the trachea, but not in the lung acini of Spink3^lacZ mice. In Spink3^Cre;R26R mice, β-gal activity was detected not only in the epithelial cells of the trachea, but also in lung acini (whole staining: A; section staining: B). Arrows and arrow heads indicate trachea and lung acini, respectively.

It has been reported that SPINK1 is expressed not only in normal tissues/organs, such as the alimentary tract and kidney in addition to the pancreas, but also in malignant tumors of the digestive tract as well as in gynecological malignancies. However, the role of SPINK1 in these tissues is unclear. It is interesting that SPINK1 is structurally similar to epidermal growth factor (EGF) [14] in terms of the extracellular structure of the EGF receptor. In addition, SPINK1 can bind to the EGF receptor to activate its downstream targets [12]. These results suggest the possibility that SPINK1 participates in specialization and proliferation of cells in addition to its role as a trypsin inhibitor. Spink3^cre mice will be a unique tool for production of conditional knockout mice to analyze Spink3 function not only in normal tissues/organs but also in tumors that express Spink3.