| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Département d’Anatomie et Biologie Cellulaire, Faculté de Médecine et des Sciences de la Santé, Université de Sherbrooke, CIHR Team on Digestive Epithelium, Sherbrooke, Québec, Canada; and
Department of Medicine, GI Division, University of Pennsylvania, Philadelphia, Pennsylvania, USA
1Correspondence: Département d’Anatomie et de Biologie Cellulaire, Faculté de Médecine et des Sciences de la Santé, 3001 12e ave Nord, Fleurimont, QC, Canada, J1H 5N4. E-mail: francois.boudreau{at}usherbrooke.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: tight junction • polarization • intestinal epithelium
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The mature intestinal epithelium is composed of absorptive cells and secretory cell lineages (Paneth, goblet, and enteroendocrine) organized in a compartmentalized cellular monolayer, with cellular proliferation limited to the crypt region and terminally differentiated cells present on the villi (1)
. This epithelium is maintained by a constant renewal program initiated by crypt stem cell division, rapid proliferation of progenitor cells, differentiation, and apoptosis. The integrity of the intestinal epithelial barrier is achieved by macromolecular assembly of the tight junction (TJ) complexes between cells and contributes to the normal function of the intestinal epithelium (2)
. Claudins are crucial members of the transmembrane tetraspan family that constitute TJ (3)
. More than 14 different claudin members are detectable in the mouse adult intestine, some of which display differential expression patterns along the horizontal and vertical axes (4)
. Claudin-1, -3, and -4 are up-regulated in colorectal cancer (5)
and claudin-1 was recently shown to influence intestinal epithelial cell transformation and colon cancer metastasis (6)
.
Differential molecular mechanisms account for the intricate control of proliferation vs. differentiation of the epithelium. One of the best-characterized signaling pathways controlling proliferation and differentiation along the crypt-villus axis is the canonical Wnt/β-catenin pathway (7)
. Activation of canonical Wnt signaling leads to stabilization of β-catenin, which accumulates in the cytoplasm, translocates to the nucleus and associates with the transcription factor T cell factor/lymphoid-enhancing factor (TCF/LEF) (8)
. This signaling imposes a crypt progenitor phenotype (9)
and controls crypt Paneth cell positioning (10)
. Nuclear β-catenin delocalization is a prerequisite for intestinal epithelial cell differentiation (11)
, and sustained Wnt signaling significantly contributes to the initiation of colorectal cancer (12
13
14)
. The β-catenin/Tcf transcriptional complex has also been suggested to positively regulate transcription of the claudin-1 gene in colorectal cell lines (6
, 15)
.
In an effort to gain a better understanding of the molecular pathways involved in the control of intestinal epithelial cell homeostasis, we investigated target genes modulated during the induction of cell polarization and differentiation by gene expression analysis. As a model, we used the intestinal epithelial cell line IEC-6, which conditionally expresses Cdx2, a transcription factor that initiates intestinal epithelial differentiation (16
, 17)
. This analysis identified cathepsin L as a component induced during the course of this process.
Cathepsin L belongs to the cysteine protease class of the papain superfamily that is involved in intracellular and extracellular protein degradation (18)
. The intracellular role of cathepsin L has long been regarded as related exclusively to terminal degradation of proteins in the lysosomal compartment given its relatively high abundance in lysosomes and its ubiquitous distribution (19)
. Cathepsin L is synthesized as a preproenzyme and targeted to the endoplasmic reticulum for subsequent processing into its active, mature enzymatic form in the acidic environment of late endosomes or lysosomes (20)
. However, alternative routing has also been reported to account for the specific targeting of extracellular procathepsin L (21
, 22)
. A growing amount of evidence supports a direct link between secreted procathepsin L and progression of pathological conditions such as colorectal cancer (23
24
25
26)
. It is now well accepted that cysteine proteases such as cathepsin L promote the degradation of basement membranes and extracellular matrix, thereby facilitating invasion and metastasis (27)
. Recent reports have also demonstrated a crucial role for cathepsin L in hair follicle morphogenesis (28)
as well as in differentiation of spermatocytes in mice (29)
. Recent evidence has challenged the intracellular lysosome-restricted role of this cysteine protease with the demonstration that cathepsin L functions in the regulation of cell cycle progression through its presence in the nucleus and its ability to proteolytically process the CDP/Cux transcription factor (30)
. These observations suggest that normal cathepsin L expression could be crucial to the maintenance of cellular homeostasis and may play a role in the constant paradigm between proliferation and differentiation.
Here we demonstrate that inhibition of cathepsin L activity interferes with the molecular mechanisms that instruct epithelial cells to polarize and undertake a differentiation program. Inhibition of cathepsin L activity induces claudin-1 expression both in vitro and in vivo. Mice deficient for cathepsin L enzymatic activity have an enhanced tumor multiplicity in the intestinal neoplasia ApcMin mouse model. These observations open up a new field of investigation for the implication of intracellular cysteine proteases in the control of normal intestinal epithelial homeostasis as well as in the initiation of intestinal epithelial disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
was maintained at 37°C in a 5% CO2 atmosphere in Dulbecco’s modified Eagle’s medium supplemented with 4.5 g/liter D-glucose, 25 mM HEPES, 5% fetal bovine serum, and 0.1 U/ml insulin. The Caco-2/15 cell line was maintained under identical conditions except for 10% fetal bovine serum. Cells were subcultured for 5 to 10 passages and prevented from reaching confluence. For pharmacological treatments, cells were supplemented with the cell membrane-permeable Z-Phe-Tyr(tBu)-diazomethylketone (Z-Phe) (Bachem Bioscience, St. Helens, UK) or the cell membrane nonpermeable Z-Phe-Tyr-CHO (cathepsin L inhibitor 2, Inh 2) Calbiochem, San Diego, CA, USA) at varying concentrations, as indicated. Cell proliferation was monitored by the measurement of [methyl-3H]-thymidine incorporation exactly as reported elsewhere (31)
.
Mice and macroadenoma counts
C57BL/6J-ApcMin and C57BL/6J-furless mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). Homozygous mice and control littermates (wild-type) were obtained by breeding with C57BL/6J. For macroadenomas studies, age-matched ApcMin (control) and ApcMin (furless) mice on a C57BL/6J background were sacrificed at 90 to 100 days of age. Mice were genotyped for ApcMin alleles as reported previously (32)
. The furless point mutation was detected by real-time PCR using fluorogenic probes as described previously (33)
. The 5' amplification primer was 5'-agtgtgcatgtataaaggc-3' and the 3' amplification primer was 5'-ccttgagcgtgagaac-3'. The 5' hybridization probe (anchor probe) was 5'-cccgatgcgctaaacgcc-3' and the 3' hybridization probe (sensor probe) was 5'-aggaacatctgtccttctaggc-3'. Real-time PCR and melting point analysis were performed using the LightCycler apparatus and software 4.0 (Roche Diagnostics, QC, Nutley, NJ, USA). The melting curve analysis was obtained as follows: 95°C, 0 s, 40°C, 2 min increasing to 70°C at 0.2°C/s with a continuous read. To evaluate macroadenomas in adult mice, the gastrointestinal tract from stomach to rectum was removed. The stomach and small intestine was cut longitudinally and washed gently with PBS to remove fecal material. Tissues were placed in a Petri dish and the mucosa was overlaid with 1% methylene blue. A dissection microscope was used to count the polyps as described previously (34)
.
Microarray analysis
Microarray analysis was performed using the Atlas Mouse 588 cDNA Expression Array membrane (Clontech, Palo Alto, CA, USA). Total RNA was obtained from IEC6 and IEC6-Cdx2L1 cells that had been grown in the absence or presence of 4 mM isopropyl β-D-thiogalactoside (IPTG) for 30 h. Preparation of probes and hybridization were performed according to the user manual. Briefly, cDNA was labeled at 50°C for 25 min in the reverse transcription (RT)/amplification reaction containing 5 µg of DNase I-treated total RNA, 1 µl primer mix (provided by Clontech), 50 µM dNTPs, 0.5 µl [
-32P]dATP (3,000 Ci/mmol, 10 µCi/µl), 5 mM dithiothreitol, and 5 x reaction buffer with 200 U of M-MLV reverse transcriptase. The labeled cDNA was purified with a NucleoSpin Extraction Spin column (Clontech) to remove unincorporated nucleotides. Arrays were hybridized at 68°C overnight with the labeled probes, washed, and exposed for autoradiography. Signals on hybridized filters were recorded by a phosphor-imager (Storm PhosphorImager; Molecular Dynamics Inc., Sunnyvale, CA, USA). The result of each gene in each subject was recorded as the intensity of hybridization signal. The digitized data were analyzed using image analysis software associated with the cDNA arrays (Atlas Image, ver. 1.5; Clontech Laboratories). The signal intensity of each element on the array was determined and corrected by subtracting background. After signals in each array were normalized to the housekeeping (glyceraldehyde-3-phosphate dehydrogenase) G3PDH gene, each gene was compared from experiment to experiment for differential expression.
Plasmid construction and lentivirus production
The lentiviral shRNA expression vector (pLenti6-U6) was constructed by cloning the U6 promoter from pSilencer 2.0-U6 (Ambion, Austin, TX, USA) into pLenti6/V5-D-TOPO (Invitrogen, Carlsbad, CA, USA). Briefly, the U6 promoter was amplified by PCR from pSilencer 2.0-U6 using the forward primer 5'-ccatcgatcatgattacgaattgcaacg-3' (insertion of a ClaI site) and the reverse primer 5'-ggccagtgccaagcttg-3'. The PCR product was digested with BamHI and ClaI and cloned into recircularized pLenti6/V5-D-TOPO between the BamHI and ClaI sites replacing the cytomegalovirus promoter. shRNA oligonucleotides were designed according to Ambion guidelines (technical bulletin #506) using the siRNA sequences ggatggatcttgtaaatac (#1) or ggcagatagtgaatggcta (#2) with ttcaagaga as the loop sequence (see
Fig. 6A
). The oligonucleotide-annealed product was subcloned into plenti6-U6 between the BamHI and XhoI sites, giving rise to pLenti6-shCTSL. An irrelevant (irr) pLenti-sheGFP negative control was similarly generated with the siRNA sequence gccacaacgtctatatcatgg. Lentiviruses were produced and used for cell infection according to Invitrogen recommendations (ViraPower Lentiviral Expression System, Instruction Manual).
|
|
|
|
|
|
Cell fractionation along the crypt-villus axis
Segments of mouse CD-1 small intestine were inverted onto polyethylene tubing, ligated at both extremities, and washed extensively with KRP buffer, pH 7.5, as described previously (35)
. Segments were then incubated under agitation with ice-cold isolation buffer (2.5 mM EDTA and 0.25 mM NaCl) for 2 min. After each interval, cell suspensions were centrifuged at 400 g for 5 min. Pellets were then washed with ice-cold KRB buffer and lysed in chilled Triton lysis buffer [150 mM NaCl, 1 mM EDTA, 40 mM Tris, pH 7.6, 1% Triton X-100, 1% protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA), and 0.1 mM orthovanadate]. The crypt or villus origin of the various cell fractions was determined by evaluating sucrase-isomaltase and trehalase brush border enzyme activities as described previously (35)
.
Enzymatic assays
Total protein extracts were isolated in acetate buffer (0.1M sodium acetate, pH 5.2, 1 mM EDTA, 0.01% Triton X-100) by three freeze (liquid nitrogen) and thaw (37°C) cycles. Protocols and specific biochemicals for the assay of cathepsin L activity were provided by Calbiochem. Briefly, 10 µl of fresh cell lysate were diluted in 0.1% BRIJ 35 to a total of 500 µl. The enzyme was activated for 1 min at 30°C in the presence of 250 µl of assay buffer (340 mM sodium acetate, 60 mM acetic acid, 4 mM EDTA, pH 5.5; 8 mM DTT added just before use) after which 250 µl of the substrate solution (20 µM Z-Phe-Arg-7-amido-4-methylcoumarin) was added. After a 10 min incubation at 30°C, the enzymatic reaction was stopped by the addition of 1 ml of stop solution (100 mM sodium monochloroacetate, 70 mM acetic acid, 30 mM sodium acetate, pH 4.3). The 7-amido-4-methylcoumarin (AMC) fluorescent product was quantified using a spectrofluorometer with an excitation wavelength of 
370 nm and an emission wavelength of 460 nm. Standard curves were performed with dilutions of AMC stock solution using a 1:1 mixture of assay buffer and stop solution. One unit was defined as the amount of enzyme hydrolyzing 1 mmol of Z-Phe-Arg-AMC at 30°C, pH 5.5. To ensure specificity of cathepsin L activity detection, the specific cathepsin B substrate Z-Arg-Arg-amido-4-methylcoumarin (Calbiochem) was used in parallel. No significant cathepsin B activity was detected under these conditions.
In vitro transcription/translation and claudin-1 processing assay
In vitro transcription/translation was performed with rabbit reticulocyte lysate-TNT kit from Promega and vectors supporting expression for either Cux/CDP, human β-catenin or human claudin-1; 0.1 µl of synthesized protein was further used for each assay. Activated cathepsin L enzyme isolated from human liver (Calbiochem) was incubated with the in vitro synthesized products in reaction buffer (50 mM sodium acetate, 0.04 mM CaCl2, 5 mM DTT at pH 5.5) at 37°C for 15 min in a total volume of 20 µl as described elsewhere (36)
. The reactions were then analyzed by Western blot.
Western blot analysis
Total protein extracts were isolated with lysis buffer (150 mM NaCl, 50 mM Tris-Cl, pH 7.5, 1% Nonidet P-40, 0.5% Na-deoxycholate) containing protease and phosphatase inhibitors (37)
. Differential extraction of cellular proteins from the cytosol, membrane and organelles, nucleus and cytoskeleton was performed with the ProteoExtract Subcellular Proteome Extraction Kit according to the manufacturer’s instructions (Calbiochem, EMD Biosciences, San Diego, CA, USA). Twenty µg of protein extract or 20 µl of concentrated extracellular medium was analyzed by 4–12% BisTris NuPAGE (Invitrogen) and transferred onto a PVDF blotting membrane (Roche Diagnostics, QC). Western blot was then performed as described (37)
. The following affinity-purified antibodies from Santa Cruz Biotechnology (Santa Cruz, CA, USA) were used: cathepsin L (C-18) goat polyclonal antibody raised against a peptide mapping the C terminus of human cathepsin L (1/250), histone-H1 (FL-219) rabbit polyclonal antibody raised against the full-length peptide sequence of human histone-H1 (1/250), CDP affinity-purified goat polyclonal antibody mapping the C terminus of human CDP (1/250), and actin goat polyclonal antibody raised against a peptide mapping the C terminus of human actin (1/10000). Both claudin-1 (1 µg/ml) and claudin-2 (1 µg/ml) rabbit polyclonal antibody from Zymed Laboratories (Invitrogen) as well as a β-catenin rabbit polyclonal antibody raised against a peptide mapping the C terminus of human β-catenin (1/3000) from Cell Signaling Technology (Danvers, MA, USA) were used.
Electron microscopy
Cell cultures and portions of mouse intestinal segments were rinsed with PBS, prefixed for 15 min with a 1:1 mixture of culture medium (Dulbecco’s modified Eagle’s medium) and freshly prepared 2.8% glutaraldehyde in cacodylate buffer (0.1 M cacodylate and 7.5% sucrose), then fixed for 30 min with 2.8% glutaraldehyde at room temperature. After two rinses, specimens were postfixed for 1 h with 2% osmium tetroxide in cacodylate buffer. The tissues were then dehydrated using graded ethanol concentrations (40, 70, 90, 95, and 100%, three times each) and coated twice for 3 h with a thin layer of Araldite 502 resin (for ethanol substitution). Finally the resin was allowed to polymerize at 60°C for 48 h. The specimens were detached from the plastic vessels, inverted in embedding molds, immersed in Araldite 502, and polymerized at 60°C for 48 h. Ultramicrotome-prepared thin sections were contrasted with lead citrate and uranyl acetate, then observed on a Jeol 100 CX transmission electron microscope. All reagents were purchased from Electron Microscopy Sciences (Cedarlane, Hornby, ON, Canada).
RNA analysis
Total RNA was isolated from cultured cells and subjected to DNase treatment according to the manufacturer’s instructions (T¯oTALLY RNA kit, Ambion, TX, USA). Reverse transcription reactions were carried out at 42°C for 1 h in the presence of 1 µg RNA, 40 mU of polyoligo(dT)12–18 (Amersham Biosciences) and 40 U of reverse transcriptase (Roche Diagnostics, QC). Typical PCR reactions were performed in a total volume of 50 µl in the presence of 1 µl of RT reaction, 1 U of HotMaster TaqDNA polymerase (Eppendorf), and 100 ng of each specific oligonucleotide. Real-time PCR was performed using a LightCycler apparatus (Roche Diagnostics, QC). Experiments were run and analyzed with the LightCycler software 4.0 (relative quantification monocolor) according to the manufacturer’s instructions (Roche Diagnostics, QC). Synthesis of double-stranded DNA during the various PCR cycles was monitored using SYBR Green I (QuantiTect SYBR Green PCR Kit; Qiagen, Valencia, CA, USA) and PCR programs designed as detailed in the QuantiTect SYBR Green PCR Handbook (Qiagen). A standard calibration curve was prepared for each gene using serial dilutions of the calibrator sample (IEC6/Cdx2 cells); crossing point values were then plotted vs. the log of the relative concentration of each dilution. This standard curve was used to correct for differences in the efficiency of the PCR reactions varying from 1.82 to 2.00 depending on the amplified target. Primer sequences are available on request.
Indirect immunofluorescence on intestine or cultured cell preparations
Indirect immunofluorescence was performed as described before (38)
except that tissues and cell samples were fixed in methanol at –20°C. A claudin-1 rabbit polyclonal antibody from Zymed Laboratories (Invitrogen) was diluted 1/80 in BSA 2%. Images were captured on a Leica DMLB2 microscope using a Leica DC300 camera.
Statistical analysis
Data are expressed as mean ± SE. Statistical analysis was performed using the Student’s t test. Differences were considered significant with a P value of <0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 17)
, was used. Total RNA was isolated from 30 h IPTG-induced or uninduced cells and processed for microarray analysis. The cathepsin L gene product, predicted to be significantly up-regulated, was selected for further investigation (data not illustrated). The cathepsin L profile during IEC-6/Cdx2L1 differentiation was then established at the protein level. Western blots were performed, with total extracts obtained at various intervals from IEC-6/Cdx2L1 cells induced to differentiate. A strong increase in the 29 kDa catalytically active processed form of cathepsin L (CTSL) was observed (Fig. 1
A). Differential extraction of IEC-6/Cdx2L1 cellular proteins from the cytosol, membrane/organelles, and nuclear compartments demonstrated that the 29 kDa catalytically active form of cathepsin L was specifically up-regulated in the membrane/organelles (2-fold±0.08, P<0.001) and nuclear cellular compartments (1.8-fold±0.009, P<0.001) (Fig. 1B
). Since procathepsin L has been reported to be secreted by intestinal epithelial cells in culture (39)
, we next verified whether the increase in intracellular cathepsin L could be associated with the accumulation of extracellular procathepsin L. Cell culture media were concentrated from IEC-6/Cdx2L1 cells at different intervals after induction of cellular differentiation and subjected to Western blot analysis. Procathepsin L secreted protein was detected in the culture medium 1 day after IPTG supplementation and increased with time (Fig. 1C
). These observations indicated that the initiation of IEC-6/Cdx2L1 differentiation is associated with an intracellular accumulation of catalytically active cathepsin L as well as extracellular secretion of procathepsin L. A specific increase in catalytically active cathepsin L expression was also observed in postconfluent Caco-2/15 cells, which spontaneously differentiate into enterocytes upon reaching confluence (Fig. 1D
). To verify that catalytically active cathepsin L could be detected in the nucleus of enterocytes in vivo, epithelial cells were isolated from the mouse adult intestine. Differential extraction of cellular proteins confirmed the presence of the 29 kDa catalytically active form of cathepsin L into the nucleus (Fig. 1E
). To further reinforce the relationship between the induction of cathepsin L expression and intestinal epithelial differentiation, the expression profile of cathepsin L was investigated along the small intestinal epithelium in vivo. The expression profile of cathepsin L was assessed in crypt and villus cell populations sequentially isolated from adult mouse jejunum according to a modified Weiser procedure (35)
. Expression of the catalytically active form of cathepsin L was strongly increased in cell populations associated with differentiation (Fig. 1F
). Overall, these observations confirmed that catalytically active cathepsin L protein expression is up-regulated in differentiated intestinal epithelial cells.
Intracellular inhibition of catalytically active cathepsin L impairs intestinal epithelial cell polarization and differentiation
To further investigate the functional relationship between the increase in cathepsin L expression and intestinal epithelial cell differentiation, experiments were undertaken in which intracellular and extracellular cathepsin L activity was inhibited during cell differentiation. The effect of two highly specific and nonreversible cathepsin L pharmacological inhibitors (Z-Phe-tyr-(tBu)-diazomethylketone and Z-Phe-Tyr-CHO) was first evaluated on enzyme intracellular activity during IEC-6/Cdx2L1 differentiation (Fig. 2
A). Addition of 10 µM of Z-Phe-tyr-(tBu)-diazomethylketone (Z-Phe) led to a >95% reduction in cell cathepsin L activity (Fig. 2A
). As expected, the cell nonpermeable Z-Phe-Tyr-CHO inhibitor (Inh 2) failed to significantly influence intracellular cathepsin L activity under the same concentration range (Fig. 2A
). The proliferation rate of IEC-6/Cdx2L1 cells treated with 10 µM Z-Phe was not significantly altered when compared with nontreated cells (Fig. 2B
). IEC-6/Cdx2L1 cells were then supplemented with either Z-Phe or Inh 2 (10 µM), followed by induction of cell differentiation. IEC-6/Cdx2L1 cells revealed the formation of typical multicellular proliferating structures with the initiation of epithelial polarization on the surface (Fig. 2C, E
) (17)
. Electron microscopy analysis confirmed the initiation of enterocyte polarization on the surface of these multicellular structures (Fig. 2E
). IEC-6/Cdx2L1 cells exposed to various concentrations of the nonpermeable Inh 2 showed similar structures (data not illustrated). In contrast, IEC-6/Cdx2L1 cells treated with Z-Phe exhibited a strong disorganization of the multicellular structures with an accumulation of dense cellular clusters (Fig. 2D
). These clusters were composed of several anchored cells that had lost the characteristics associated with polarized enterocytes (Fig. 2F
) and which consistently detached into the culture medium. Long-term inhibition of cathepsin L activity resulted in an increase of caspase-3 cleavage that was not associated with irreversible cell death, since detached cells were not stained by the Trypan blue exclusion test; when they were subcultured into new vessels in the absence of cathepsin L inhibitor, they readhered and continued to proliferate (data not illustrated). Since it was obvious that cellular polarization was impaired in cathepsin L inhibited cells (Fig. 2)
, the mRNA profile of several molecules known to maintain adherent and tight junction integrity was next investigated. RT-PCR analysis failed to detect any significant modification of ZO-1, occludin, and E-cadherin expression levels during inhibition of cathepsin L activity (data not illustrated). However, claudin-1 gene expression was identified to be affected at the transcript level when IEC-6/Cdx2L1 cell cultures were supplemented with Z-Phe (Fig. 3
A). Indeed, the relative level of claudin-1 mRNA expression was decreased over time in IEC-6/Cdx2L1 cultured cells whereas the addition of Z-Phe significantly prevented this reduction (Fig. 3A
). Claudin-2, -3, and -4 mRNA expression was not affected under these conditions (data not illustrated). A Western blot analysis showed that claudin-1 protein was already up-regulated 24 h after pharmacological inhibition of intracellular cathepsin L and remained elevated during the entire treatment (Fig. 3B
). Since the mechanism for up-regulation of claudin-1 by the loss of cathepsin L activity appeared to be rapidly mediated at the protein stability, we hypothesized that claudin-1 protein could be a direct target of cathepsin L protease activity. In vitro synthesized proteins were incubated with active cathepsin L enzyme under defined conditions (36)
, followed by Western blot. CDP/Cux, used as a positive control, was rapidly cleaved by cathepsin L to produce the p110 isoform as previously reported (Fig. 3C
) (30)
. Incubation of recombinant claudin-1 protein with cathepsin L led to a rapid degradation of the protein with no apparent residual cleaved forms of the protein (Fig. 3C
). Recombinant β-catenin, a suspected regulator of claudin-1 transcription (6
, 15)
, was not altered under these experimental conditions (Fig. 3C
). Differential extraction of IEC-6/Cdx2L1 cellular proteins from the cytosol, membrane and organelles, nuclear and cytoskeletal compartments confirmed claudin-1 up-regulation after pharmacological inhibition of cathepsin L, the nuclear cellular compartment being the most affected (Fig. 4
A). The level of β-catenin protein remained unchanged during the pharmacological treatment (Fig. 4A
). An immunofluorescence experiment confirmed the induction of claudin-1 protein in the nucleus of IEC-6/Cdx2L1 cells incubated with Z-Phe (Fig. 4B
).
The next step consisted of confirming whether specific genes associated with intestinal differentiation could also be influenced under the same conditions. The SI gene was chosen because it is specific to the enterocyte lineage and is expressed only in the differentiated compartment of the intestinal crypt-villus axis (40
41
42)
. Moreover, this gene was the only one previously reported to be detectable at the mRNA level in differentiated IEC-6/Cdx2L1 cells (17)
. IEC-6/Cdx2L1 cells were supplemented with either 10 µM Z-Phe, 10 µM Inh 2, or no inhibitor and induced to differentiate with the addition of IPTG. Total RNA was isolated after 30 days of culture, when SI becomes significantly detectable in this culture model (17)
, and processed for RT-PCR analysis. Inhibition of intracellular cathepsin L activity with Z-Phe resulted in a reduction of SI mRNA expression compared with untreated differentiated cells (Fig. 5
). Addition of the nonpermeable Inh 2 inhibitor did not affect expression of the SI gene over the course of IEC-6/Cdx2L1 differentiation (Fig. 5)
.
An experiment to directly neutralize cathepsin L activity was next designed by a knockdown approach. Lentivirus constructs containing shRNA sequences under the control of the U6 promoter and predicted to target rat cathepsin L mRNA were generated (Fig. 6
A). Stable infection of a lentivirus-eGFP into IEC-6/Cdx2L1 cells consistently resulted in >90% of positive eGFP cells. The efficiency of two different shRNA CTSL lentiviruses was subsequently evaluated for the down-regulation of cathepsin L synthesis. Western blot analysis was performed, with total extracts obtained from stably infected IEC-6/Cdx2L1 populations. The cathepsin L protein level was decreased in IEC-6/Cdx2L1 infected with either of the CTSL shRNA lentiviruses compared with cells infected with an eGFP lentivirus or an irrelevant (irr) shRNA lentivirus (Fig. 6B
). The CTSL shRNA lentivirus #1 was further used for its potent effect in cathepsin L knockdown (Fig. 6B
). Stable IEC-6/Cdx2L1 populations were generated with either CTSL shRNA or irrelevant shRNA lentiviruses and induced to differentiate with the addition of IPTG. Total RNA was isolated and the level of cathepsin L mRNA expression evaluated by real-time RT-PCR. The expression of cathepsin L was reduced on average by >70% in CTSL shRNA populations compared with control populations (Fig. 6C
). The efficiency in targeting cathepsin L mRNA expression was still present 30 days after induction of cellular differentiation (Fig. 6C
). Cathepsin L intracellular activity was reduced by 50% in CTSL shRNA IEC-6/Cdx2L1 cells compared with controls as determined by an enzymatic assay. Total protein was then isolated at different days of culture and processed for Western blot analysis. A modest but reproducible up-modulation of the claudin-1 protein level was observed in the IEC-6/Cdx2L1 population (CTSL shRNA) that sustained 50% of residual intracellular cathepsin L activity (Fig. 6D
). Total RNA was also isolated from both populations of cells and processed for RT-PCR analysis. Reduction in intracellular cathepsin L activity resulted in the decline of SI mRNA expression in CTSL shRNA populations compared with control cells (Fig. 6E
). Overall, these observations confirm that fully sustained cathepsin L intracellular activity is essential to restrain claudin-1 expression as well as to sustain SI mRNA expression in differentiated enterocytes in culture.
Mice defective for cathepsin L enzymatic activity display enhanced claudin-1 expression and are prone to increased intestinal neoplasia
The molecular interaction between cathepsin L activity and claudin-1 expression was next monitored in furless mice, which express catalytically inactive cathepsin L (28)
. Total RNA was isolated from the small intestine of control and mutant mice and processed for real-time PCR analysis. Claudin-1 mRNA expression was significantly more elevated in furless mice than in control littermate mice (Fig. 7
A). Western blot analysis confirmed that the claudin-1 protein level was up-regulated in the small intestine of furless mice (Fig. 7B
). The cell distribution of claudin-1 protein in the small intestine was comparable between control and mutant mice, a typical signal being detected mostly at the apical junctions of enterocytes, as reported before (Fig. 7C
) (43)
. Ultrastructures of small intestinal lower crypt cells were then visualized by electron microscopy. Epithelial cells from control mice displayed typical TJ on the apical side of the cells whereas the TJ in furless mice systematically appeared less well defined, although it was technically difficult to quantitatively evaluate this difference (Fig. 7D
). TJ of villus epithelial cells were comparable between control and furless mice (data not illustrated). A comparative analysis of epithelial cell proliferation as well as expression of differentiated epithelial markers failed to demonstrate any significant modification between furless and control animals (data not illustrated). Because claudin-2 was reported to bring leakiness properties into the tight junction structural organization (44
45
46)
, its expression was next monitored in furless mice. Claudin-2 mRNA expression was significantly up-regulated in furless mice compared with control littermate mice (Fig. 7E
), a tendency also observed at the protein level (Fig. 7F
).
|
To further evaluate whether cathepsin L could interfere with intestinal tumorigenesis, we next crossed ApcMin mice with furless mice of the defined, inbred C57Bl6 background. The integrity of cathepsin L epithelial activity had an important effect on tumor load in the small intestine. While ApcMin control littermate mice had an average of 18.7 intestinal polyps (Fig. 8
), ApcMinfurless mice developed an average of 41 polyps per animal, a 2.2-fold increase over ApcMin control mice (Fig. 8)
. A similar tendency was observed in the colon, but the number of polyps in these mice was less important (0.6±0.13 vs. 1.8±0.37; P<0.01). In addition, furless mice did not develop intestinal polyps even at 1 year of age. These results suggest that cathepsin L enzymatic activity and Apc act cooperatively to regulate tumor multiplicity in the intestinal epithelium.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. More recently, experiments conducted with mice deficient in either cathepsin L or D have exemplified their involvement in the control of cellular proliferation and differentiation (28
, 29
, 47
, 48)
. Our study has identified cathepsin L as a component that is up-regulated in differentiated intestinal epithelial cells. Strategies used here to down-modulate cathepsin L activity have distinctly demonstrated the intracellular importance of this enzyme in enterocyte polarization and maturation in vitro. An important component of the tight junction, claudin-1, has been identified as a downstream target for cathepsin L activity both in vitro and in vivo. This has an impact on tumor multiplicity in ApcMin mice. These findings indicate for the first time that a proteolytic enzyme normally associated with lysosomal function is a crucial component of the molecular cascade that leads to normal enterocyte polarization as well as intestinal neoplasia.
The involvement of cathepsin L activity in intestinal epithelial cell polarization and differentiation represents an uncharacteristic finding for this protease. Cathepsin L has been proposed to positively influence tumor invasion and metastasis in different models (49
50
51)
. Cathepsin L plays a crucial role in the invasive activity of endothelial cells (48)
as well as in proliferative disorders and tumor growth in pancreatic islet cell carcinogenesis (36)
. Extracellular secretion of the preproenzyme has been extensively linked to progression of many different cancers (23)
and correlative relationships have been proposed for cysteine protease activities and their role in colorectal cancer progression (25
, 26
, 52)
. However, these studies did not address the specific subcellular targeting of cathepsin L action during cancer progression. Our analysis is the first to clearly demonstrate that intracellular cathepsin L expression is induced with differentiation of intestinal epithelial cells. Intracellular enzymatic inhibition of cathepsin L led to claudin-1 misexpression and lack of cellular polarization, whereas specific inhibition of the secreted extracellular preproenzyme with a nonpermeable inhibitor did not influenced differentiation of cultured intestinal epithelial cells. Other systems have suggested a role for cathepsin L in keratinocyte and spermatogenic normal cell maintenance (28
, 29)
. Indeed, furless mice present an alteration in hair follicle morphogenesis caused by hyperproliferation of hair follicle epithelial cells (28)
as well as defects in spermatocyte differentiation (29)
. As a simplistic explanation, we propose that modulation and cellular distribution of this protease represent an important variable for its biological role in the function of cell origin, initiation of disorder, and progression through carcinoma and metastasis. For example, cathepsin L was undetectable in normal cells of the pancreatic islet but became progressively induced as pancreatic endocrine neoplasms became more aggressive (36)
. This contrasts with our observation of enhanced expression in normal differentiated intestinal epithelial cells. Future functional experiments in vivo could determine the molecular relevance of increased intracellular vs. extracellular cathepsin L synthesis in the context of initiation and progression of intestinal cancer.
The spontaneous mouse mutant model furless, which exhibits inactivated cathepsin L proteinase activity (28)
, failed to demonstrate defects in intestinal epithelial proliferation-to-differentiation homeostasis (data not illustrated). Although inactivation of cathepsin L enzymatic activity may not be crucial in order to maintain the intestinal epithelium in this particular model, it remains plausible that these mice develop compensatory mechanisms during early development to overcome deficiency in cathepsin L activity. Lung tissue extracts of cathepsin L-deficient mice express increased amounts of cathepsin D (53)
. Mice deficient for cathepsin D display abnormalities in intestinal mucosal cell turnover (47)
. Strategies to investigate a putative cross-talk between these different proteases as well as other molecular determinants known to be involved in the establishment and maintenance of the intestinal epithelium represent a future challenge.
The increase in cathepsin L protein level was much more evident in the accumulation of the enzymatically intracellular active mature form than the unprocessed form during intestinal differentiation. Two nonexclusive possibilities could explain this specific pattern in normal intestinal epithelial cells. 1) Cathepsin L synthesis could be augmented via the classic endoplasmic reticulum route and lead to extracellular secretion of the procathepsin L form and preferential accumulation of the mature protein form within specialized intracellular vesicles. This is supported by the observation that secreted procathepsin L is up-regulated during IEC-6/Cdx2L1 differentiation. 2) A mechanism favoring the use of a downstream AUG site, thus escaping the signal peptide sequence of the traditional enzyme, could be activated. Indeed, fibroblasts can synthesize a cathepsin L isoform downstream of the traditional AUG site (30)
. This leads to a protease devoid of a signal peptide with the same molecular weight as the catalytically active form capable of translocating to the nucleus and proteolytically processing the CDP/Cux transcription factor (30)
. This second possibility is supported by strong induction of the mature form within the cells as well as the presence of a catalytically active isoform of cathepsin L observed here in the nucleus of intestinal epithelial cells. Although there was no evidence of CDP/Cux processing in intestinal epithelial cells with nuclear cathepsin L (data not illustrated), other uncharacterized nuclear targets of this protease in the context of differentiated epithelial cells remain to be identified.
How could the loss of cathepsin L activity promote initiation of intestinal tumorigenesis? As mentioned before, increases in secreted cathepsin L enzymatic activity have been well documented to be associated with tumor progression and metastasis. However, no clues were provided in this particular context for the specific roles of intracellular cathepsin L. Our study has identified claudin-1 as being rapidly induced in the absence of cathepsin L intracellular activity. In vitro experiments that tested whether recombinant claudin-1 was a direct target substrate for cathepsin L supported the notion that this enzyme might prevent claudin-1 protein accumulation. However, it is still unclear whether this protease can directly target claudin-1 under any specific subcellular region in vivo. The induction of claudin-1 is well documented in cancer, a process that has been correlated with decreased TJ integrity (54)
. Overexpression of claudin-1 in human colorectal cell lines resulted in epithelial-mesenchymal transition with increased growth of xenografted tumors, a process that was reversed when siRNA specific for claudin-1 was used (6)
. Another study showed that inhibition of claudin-1 expression with siRNA largely suppressed the invasion properties of oral squamous carcinoma cells (55)
. Claudin-1 knockout mice confirmed an important role in maintaining barrier integrity in the epidermis, as mice die within 1 day of birth with defects in the epidermal barrier (56)
. One could argue that increases in claudin-1 should improve tight junction formation, polarization, and maturation. Crypt epithelial cells of furless mice appeared less well defined than control mice, an observation that is probably linked to the induction of claudin-2 expression (44
45
46)
. Changes in claudin-1 localization, particularly within the cytosol and nucleus, are currently proposed to account for the effect of claudin-1 on cellular transformation; however, the molecular nature of this mechanism remains unknown (6
, 54)
. Inhibition of cathepsin L enzymatic activity resulted in an induction of nuclear claudin-1 in IEC-6/Cdx2L1 cells, an effect not seen in furless mice. This indicates that nuclear translocation of claudin-1 is not solely dependent on cathepsin L but other pathways must be required. IEC-6 cells are immortalized and could have acquired signaling pathways essential for claudin-1 nuclear translocation. Claudin-1 transcriptional regulation has been suggested to depend on the β-catenin/Tcf/Lef signaling pathway (6
, 15)
. A transient transfection assay for measuring transcriptional activation due to activated β-catenin signaling failed to demonstrate a direct interference of cathepsin L with this signaling pathway. In addition, the level of nuclear β-catenin was not affected in IEC-6/Cdx2L1 cells deficient in cathepsin L activity or in furless mice. This finding is consistent since furless mice do not present defects in intestinal epithelial proliferation nor develop spontaneous adenomas. ApcMin mice already display nuclear claudin-1 expression (6)
. We thus suggest that the increase in claudin-1 expression by the furless mice and the potential for claudin-1 to nuclear translocate in ApcMin mice (6)
act synergistically to initiate intestinal tumorigenesis.
In conclusion, our findings support an unsuspected role for intracellular cathepsin L in maintaining the normal intestinal epithelium. Future identification of cathepsin L-specific substrates during intestinal epithelial cell differentiation will allow a better understanding of how this protease protects intestinal epithelium integrity, an essential feature to limiting the progression of intestine-related diseases.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 30, 2007. Accepted for publication May 31, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-P. Babeu, M. Darsigny, C. R. Lussier, and F. Boudreau Hepatocyte nuclear factor 4{alpha} contributes to an intestinal epithelial phenotype in vitro and plays a partial role in mouse intestinal epithelium differentiation Am J Physiol Gastrointest Liver Physiol, July 1, 2009; 297(1): G124 - G134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Darido, M. Buchert, J. Pannequin, P. Bastide, H. Zalzali, T. Mantamadiotis, J.-F. Bourgaux, V. Garambois, P. Jay, P. Blache, et al. Defective Claudin-7 Regulation by Tcf-4 and Sox-9 Disrupts the Polarity and Increases the Tumorigenicity of Colorectal Cancer Cells Cancer Res., June 1, 2008; 68(11): 4258 - 4268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. R. Lussier, J.-P. Babeu, B. A. Auclair, N. Perreault, and F. Boudreau Hepatocyte nuclear factor-4{alpha} promotes differentiation of intestinal epithelial cells in a coculture system Am J Physiol Gastrointest Liver Physiol, February 1, 2008; 294(2): G418 - G428. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-tubulin, and calpain 1 antibodies. C) Western blot analysis was performed using a cathepsin L polyclonal antibody on concentrated culture medium harvested on various days from IEC-6/Cdx2L1 cells cultured in the presence of 4 mM IPTG. D) Total lysates of Caco-2/15 cells (SC, subconfluent; C, confluent; +8, 8 days after confluency; +20, 20 days after confluency) were harvested and a Western blot was performed with cathepsin L or actin antibodies. E) Intestinal epithelial cells were isolated from the jejunum of adult mice and subcellular fractions were prepared and blotted as described in panel B. F) Western blot analysis was performed with 20 µg of lysates from isolated mouse crypt and villus epithelial cell populations (fractions 1–7; see Materials and Methods).
