Calcium/calmodulin–dependent protein kinase IV (CaMKIV) activation contributes to the pathogenesis of experimental colitis via inhibition of intestinal epithelial cell proliferation

Abstract

The incidence and prevalence of inflammatory bowel disease (IBD) are increasing worldwide. IBD is known to be multifactorial, but inflammatory signaling within the intestinal epithelium and a subsequent failure of the intestinal epithelial barrier have been shown to play essential roles in disease pathogenesis. CaMKIV is a multifunctional protein kinase associated with inflammation and cell cycle regulation. CaMKIV has been extensively studied in autoimmune diseases, but a role in idiopathic intestinal inflammation has not been described. In this study, active CaMKIV was highly expressed within the intestinal epithelium of humans with ulcerative colitis and wild-type (WT) mice with experimental induced colitis. Clinical disease severity directly correlates with CaMKIV activation, as does expression of proinflammatory cytokines and histologic features of colitis. In WT mice, CaMKIV activation is associated with increases in expression of 2 cell cycle proarrest signals: p53 and p21. Cell cycle arrest inhibits proliferation of the intestinal epithelium and ultimately results in compromised intestinal epithelial barrier integrity, further perpetuating intestinal inflammation during experimental colitis. Using a CaMKIV null mutant mouse, we demonstrate that a loss of CaMKIV protects against murine DSS colitis. Small molecules targeting CaMKIV activation may provide therapeutic benefit for patients with IBD.—Cunningham, K. E., Novak, E. A., Vincent, G., Siow, V. S., Griffith, B. D., Ranganathan, S., Rosengart, M. R., Piganelli, J. D., Mollen, K. P. Calcium/calmodulin-dependent protein kinase IV (CaMKIV) activation contributes to the pathogenesis of experimental colitis via inhibition of intestinal epithelial cell proliferation.

Inflammatory bowel disease (IBD) represents a spectrum of chronic, relapsing, idiopathic inflammatory conditions of the gastrointestinal tract, including Crohn’s disease and ulcerative colitis (1, 2). In the United States, an estimated 1.6 million people have IBD, with both the incidence and prevalence of IBD increasing worldwide (2, 3). The annual direct U.S. health care costs approximate to 28 billion dollars (3), yet treatment failure rates remain unacceptably high; 50% of patients with Crohn’s disease and 16% of those with ulcerative colitis progress to advanced, complicated disease requiring surgical intervention (4). Although the cause of IBD remains elusive, a key feature of disease is ongoing inflammation within the intestinal epithelium progressing to ulceration, epithelial loss, and poor mucosal healing.

The intestinal epithelium is a dynamic single layer of cells and an indispensable barrier between the intestinal luminal contents, including the microbiota, and the mucosal immune system (5, 6). IBD is considered a pathologic sequela of interactions occurring among the gut microbiota, host immune system, host genetics, and the environment (7), but the intestinal epithelium constitutes the interface between these factors and likely plays a central role. For example, mice with genetic polymorphisms associated with increased levels of intestinal ATP exhibit heightened enterocyte proliferation and protection during experimental colitis (8). Additional studies have demonstrated cell cycle arrest in intestinal epithelial cells (IECs) during DSS-induced colitis (9–11). Thus, increased renewal of the intestinal epithelium is a fundamental component of normal recovery after exposure to toxic cellular injury and intestinal inflammation. However, the underlying mechanisms regulating cellular fate within the intestinal epithelium during inflammation remain to be fully characterized.

Calcium signaling plays a regulatory role in the events of cellular proliferation. The calcium/calmodulin-dependent protein kinases (CaMKs) are a family of multifunctional serine/threonine protein kinases that have been demonstrated to transduce intracellular calcium signals in mediating a variety of phenotypic functions (12). One member of this family, CaMKIV, has been extensively studied in diseases of the CNS and autoimmunity. However, research regarding CaMKIV in the intestine is extremely limited, and its role in regulating cellular fate within the intestinal epithelium is unknown (12, 13). The activation of CaMKIV is a tightly controlled process that requires the completion of 2 events (14). First, Ca^2+/CaM complex must bind to CaMKIV, inducing its basal activity. Second, CaMK kinase must phosphorylate CaMKIV to generate a highly active, Ca²⁺/CaM-independent CaMKIV enzyme (14). The active, phosphorylated form of CaMKIV has been shown to translocate to the nucleus where it regulates the transcriptional activity of several targets that play crucial roles in the immune response, inflammation, and cellular proliferation (12, 15). Arnould et al. (15) demonstrated that calcium flux as a result of mitochondrial dysfunction activates CaMKIV, which inhibits cell proliferation. Our laboratory has previously demonstrated significant mitochondrial dysfunction within the intestinal epithelium during human and experimental colitis (16).

In this study, CaMKIV activation is increased in the diseased intestinal epithelium of patients with IBD and mice with induced experimental colitis. Further, compared to wild-type (WT) mice, global CaMKIV-null mutant mice exhibit protection from experimental colitis with decreased weight loss, preservation of the intestinal epithelial architecture, and minimal expression of several proinflammatory cytokines. In addition, during intestinal inflammation, calcium-dependent activation of CaMKIV induced activation of cyclic AMP response element-binding protein (CREB) (17, 18), which resulted in the up-regulation of a signaling pathway that inhibits the proliferation of the IECs (15). Based on these data, we now hypothesize that CaMKIV plays an essential role in the pathogenesis of colitis.

MATERIALS AND METHODS

Materials and reagents

All oligonucleotides were purchased from Integrated DNA Technologies (Integrated DNA Technologies, Coralville, IA, USA). Primer sequences are listed in Table 1. Proteins were purchased from the following vendors: anti-total CaMKIV and anti-8-hydroxyguanosine (8OHdG) (Abcam, Cambridge, MA, USA); anti-phosphorylated CaMKIV (Thermo Fisher Scientific, Waltham, MA, USA); anti-total CREB, anti-phosphorylated CREB, anti-p53, and anti-β-actin (Cell Signaling Technologies, Danvers, MA, USA); and anti-p21 (Santa Cruz Biotechnology, Dallas, TX, USA).

TABLE 1.

qPCR primers

		Primer sequence, 5′–3′
Gene	Species	Forward	Reverse
Rplo	Mouse/Human	GGCGACCTGGAAGTCCAACT	CCATCAGCACCACAGCCTTC
CAMKIV	Human	AGAAGCTCCAAGAATTCAATGC	GATCAGATCTTGCTGTGGAAC
CamKIV	Mouse	ACCCAGAAGCCCTATGCTCT	CCCTTCTCCACAATCCTGTCA
CaMKIVWT	Mouse	TCAATTGAAAAGTTCACTGAAAGA	TTTGAGAGCATAGGGCTTCTG
CaMKIVKO	Mouse	CTTGGGTGGAGAGGCTATTC	AGGTGAGATGACAGGAGATC
Creb	Mouse	GAGAAGCGGAGTGTTGGTGA	ACTCTGCTGGTTGTCTGCTC
p53	Mouse	GGCTCACTCCAGCCTCCAGCCT	GGCGGGAAGTAGACTGGCCCTTC
p21	Mouse	CCTGGTGATGTCCGACCTG	CCATGAGCGCATCGCAATC
iNOS	Mouse	AATGAGTCCCCGCAGCCCCT	AGTCATCCCGCTGCCCCAGT
TNFa	Mouse	TTCCGAATTCACTGGAGCCTCGAA	TGCACCTCAGGGAAGAATCTGGAA
IL-6	Mouse	CCAATTTCCAATGCTCTCCT	ACCACAGTGAGGAATGTCCA
IL-1b	Mouse	AGTGTGGATCCCAAGCAATACCCA	TGTCCTGACCACTGTTGTTTCCCA
Muc2	Mouse	TAGTGGAGATTGTGCCGCTGAAGT	AGAGCCCATCGAAGGTGACAAAGT

All primers were designed for this study except RPLO (16), CaMKIVWT (103), and CaMKIVKO (103).

Study approval

All animal experimentation was approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh and was in accordance with both federal guidelines for the care and use of laboratory animals and with guidelines set forth by the Animal Research and Care Committee at the Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center (UPMC). Human tissue samples were obtained from the Children’s Hospital of Pittsburgh of UPMC after Institutional Review Board approval at the University of Pittsburgh.

Human tissue

Human intestinal specimens were obtained from patients ≤21 yr of age who underwent colonoscopy or bowel resection surgery for either IBD or another diagnosis (which served as age-matched, nondiseased controls) at the Children’s Hospital of Pittsburgh of UPMC. Patients were excluded from this study only if they met any of the following criteria: 1) lack of parental consent, 2) congenital anomalies, and 3) pregnant females. All patient samples were coded with a system known only to the research coordinator and were handled in a blinded manner in the laboratory. The samples were processed and analyzed independent of the subjects’ private information. Final pathology reports allowed for grouping according to IBD disease subset.

DSS-induced colitis

WT C57BL/6 and Camk4^tm1Tch/Camk4^tm1Tch [CaMKIV knockout (CaMKIV KO) mice (19); Jax stock 004994] were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) and housed in the Association for Assessment and Accreditation of Laboratory Animal Care International–accredited animal facility at the Children’s Hospital of Pittsburgh Rangos Research Building in accordance with the University of Pittsburgh animal care guidelines. The CaMKIV KO mouse strain is a commercially available transgenic mouse on a C57BL/6 background that globally expresses a catalytically inactive form of CaMKIV (19). The mutation involves replacement of exon 3 with a neomycin-resistance gene, disrupting the phosphorylation-binding site in the catalytic domain, rendering any transcribed CaMKIV catalytically inactive and prone to degradation (19). A 2% solution of dextran sodium salt (DSS; 36,000–50,000 kDa; MP Biomedicals, Solon, OH, USA) was administered in the drinking water for 7 d ad libitum (DSS-induced group). Control groups received standard animal vivarium house water. Clinical signs of colitis were recorded daily (as described in refs. 16 and 20), and the disease activity index (DAI) score was calculated for each day. The DAI score includes weight loss (0–4), stool consistency (0–4), and blood in the stool (0–4). Weight loss was calculated as the change in weight (in grams) relative to the weight before initiation of the experiment. Bromodeoxyuridine (BrdU; Thermo Fisher Scientific) (10 μg/g) was injected intraperitoneally 4 h before euthanasia.

Measurement of intestinal epithelial permeability via FITC-dextran

Mice were deprived of water (12–16 h) before fluorescein isothiocyanate-dextran (FITC-dextran; MilliporeSigma, Burlington, MA, USA) administration. FITC-dextran was prepared fresh in PBS and protected from ambient light. Mice were orally gavaged with FITC-dextran (44 mg/100 g body weight), given standard vivarium house water for 4 h, and then euthanized. Whole blood was collected in a Microtainer SST Tube (Thermo Fisher Scientific) and processed to separate the serum according to the manufacturer’s instructions. The collected serum was diluted with an equal volume of PBS, and 100 μl of the diluted serum was added in triplicate to a 96-well plate. The concentration of FITC in the serum was determined by fluorescence spectroscopy with a Spectramax M2 Microplate Reader equipped with Softmax Pro Data Acquisition and Analysis Software (Molecular Devices, Sunnyvale, CA, USA) with an excitation wavelength of 485 nm and an emission wavelength of 528 nm. Serum from mice not administrated FITC-dextran was used to determine the background fluorescence. The concentration of FITC in murine serum was then extrapolated from a standard curve using serially diluted FITC-dextran.

RNA isolation and cDNA synthesis

Total RNA was isolated from either murine colonic tissue or IEC-6 cell cultures with the RNeasy Kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s instructions. The concentration and purity of each RNA sample were measured via spectrophotometry (ND-2000 spectrophotometer; NanoDrop Technologies, Inc., Wilmington, DE, USA). First-strand cDNA (0.5 μg of RNA) was prepared by using a QuantiTect Reverse Transcription Kit (Qiagen) according to the manufacturer’s instructions. cDNA was diluted to 0.5 ng/μl and stored at −20°C until further use.

Real-time quantitative PCR and analysis

Gene expression was measured relative to the housekeeping gene 50s ribosomal subunit protein L15 (RPLO) (Table 1). cDNA was amplified using a Bio-Rad CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA), with a final reaction volume of 10 μl containing 2.5 ng of cDNA, primers (500 nM final concentration), and 1× Excella Sybr Mastermix, Rox (WorldWide Medical Products, Bristol, PA, USA). The amplification conditions for the quantitative PCR (qPCR) reactions were as follows: 1 cycle of initial denaturation at 95°C for 4 min followed by 40 cycles of denaturation at 95°C for 15 s, annealing at 56°C for 30 s, and elongation at 72°C for 30 s. All reactions were performed in triplicate. Primer efficiencies were validated to be similar, which allowed the qPCR data to be analyzed by the comparative threshold (ΔΔC_t) method. The derived ΔΔC_t values were converted into fold-difference values, and the range of the fold difference (shown by bars on the graph) was derived by the incorporation of the sd of the ΔΔC_t value into the fold-difference calculation.

Western blot analysis

Mouse colonic tissue or whole cell lysates were prepared with 1× RIPA buffer (Boston Bio, Ashland, MA, USA). Protein concentrations were determined with a bicinchoninic acid assay (MilliporeSigma). Protein lysates (10–30 μg/sample) were separated on an SDS gel (8–12%) by SDS-PAGE and then transferred onto a PDVF membrane. Membranes were washed in 1× 20 mM Tris, 150 mM NaCl, 0.1% Tween-20 (TBST) for 5 min and then blocked with 5% nonfat dry milk or 5% BSA in 1× TBST. Membranes were probed with primary antibody at 4°C overnight. The membrane was washed 3 times for 15 min and then probed with an horseradish peroxidase–conjugated secondary antibody for 1 h. The membrane was developed by incubating it with a chemiluminescent horseradish peroxidase substrate (Thermo Fisher Scientific) according to the manufacturer’s instructions. The image was then acquired with a Kodak X-Omat 2000 processor (Eastman Kodak Company, Rochester, NY, USA). Where necessary, relative intensity of protein bands was quantified and analyzed with ImageJ (National Institutes of Health, Bethesda, MD, USA; http://rsb.info.nih.gov/jj/). Relative intensity was calculated as the ratio of total target protein to total actin or as the ratio of phosphorylated target protein to total target protein (21–23).

Extraction of nuclear proteins from colonic tissue

Nuclear extracts were prepared as previously described by Blough et al. (24). In brief, ∼50 mg of murine colonic tissue was minced and thoroughly homogenized in 250 μl of ice-cold buffer 1 [10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES; pH 7.5), 10 mM MgCl₂, 5 mM KCl, 0.1 mM EDTA (pH 8.0), 0.1% Triton X-100, and 1 mM DTT, 0.1 mM PMSF, 2 µg/ml aprotinin, and 2 µg/ml leupeptin] (all from MilliporeSigma), which were added fresh before use]. An additional 750 μl of buffer 1 was added after the tissue was homogenized. The minced tissue was then centrifuged at 3000 g for 5 min at 4°C. The pellet (nuclear extract) was resuspended in 500 μl of ice-cold buffer 2 [20 mM HEPES (pH 7.9), 25% glycerol, 500 nM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA (pH 8.0), 0.5 mM DTT, 0.2 mM PMSF, 2 μg/ml aprotinin, and 2 μg/ml leupeptin, added fresh before use], and incubated on ice with intermittent mixing for 1 h. After the incubation, the nuclear pellet was centrifuged as before for 5 min at 3000 g. The supernatant was transferred to a new microcentrifuge tube. An equal volume of binding buffer [20 mM HEPES (pH 7.9), 40 mM KCl, 2 mM MgCl₂, 10% glycerol, 0.5 mM DTT, 0.2 mM PMSF, 2 µg/ml aprotinin, and 2 µg/ml leupeptin, which were added fresh before use] was added to the transferred supernatant. The solution was then loaded onto an Ultra Centrifugal Filter Unit (3000 nominal molecular weight limit; MilliporeSigma) and centrifuged according to the manufacturer’s instructions for 30 min at 4°C. Another 500 μl of binding buffer was added to the filter and centrifuged for an additional 30 min at 4°C. The remaining solution in the filter was transferred into a new tube. The concentration of the nuclear extracts was determined with the bicinchoninic acid assay. All nuclear extracts were stored at −80°C until further use.

EMSA

Nuclear proteins were extracted as previously described. The Gel Shift Assay System (Promega, Madison, WI, USA) was used to radiolabel the provided CREB-specific consensus oligonucleotide (5′-AGAGATTGCCTGACGTCAGAGAGCTAG-3′) with γ-[³²P]-ATP (3000 Ci/mmol at 10 mCi/ml), per the manufacturer’s instructions. Radiolabeled DNA was separated from unlabeled γ-[³²P]-ATP (PerkinElmer, Boston, MA, USA) by using G-25 Sephadex Columns (Roche, Indianapolis, IN, USA), according to the manufacturer’s protocol. The percentage incorporation was determined, and the radiolabeled probe was used only in an EMSA if incorporation was above 30%. Nuclear extracts (15 μg) were incubated with radiolabeled CREB DNA probes (0.29 ng or 0.02 pmol) for 30 min at room temperature in 1× binding buffer [10 mM Tris-HCl (pH 7.5), 50 mM KCl, 1 mM DTT, 6% glycerol, 0.5 mM EDTA, 50 μg/ml sheared salmon sperm DNA (ssDNA), and 50 μg/ml BSA; Thermo Fisher Scientific] for a total reaction volume of 20 μl. Nuclear extract from HeLa cells (provided with Promega kit) was used as a positive control. The negative control contained a reaction with the labeled DNA probe but no protein lysate. Samples were run in 0.5× Tris base, acetic acid, and EDTA (TAE) buffer [50× TAE: 242 g of Tris Base, 57.1 ml glacial acetic acid, 100 ml 0.5 M EDTA (pH 8.0), water added to 1 L] on a 6% nondenaturing polyacrylamide gel at 150 V. Gels were dried on a gel dryer (Savant SpeedGel System; Thermo Fisher Scientific) for 1 h at 80°C and then analyzed by autoradiography.

Histology

Murine colonic tissue or human intestinal tissues were fixed in 4% paraformaldehyde overnight, dehydrated, and embedded in paraffin. Tissue sections were cut at 5 μm, deparaffinized, and rehydrated through a gradient of xylene and ethanol baths. Tissue sections were then stained with hematoxylin and eosin (MilliporeSigma) for light microscopic examination. Tissues were evaluated by a blinded pathologist for signs of disease.

Immunofluorescence

Murine colonic tissue or human intestinal tissues were fixed in 4% paraformaldehyde overnight, dehydrated, and embedded in paraffin wax blocks. Tissue sections were cut at 5 μm, deparaffinized, and rehydrated through xylene and ethanol gradient baths. Heated citric acid was used as the antigen retrieval step. Sections were then blocked in 1% BSA and 5% donkey serum in 1× PBS. Sections were incubated with primary antibody overnight at 4°C and then probed with a secondary antibody for 1 h the following day. The sections were mounted with a coverslip and allowed to dry thoroughly before being imaged on a fluorescence microscope (Olympus, Center Valley, PA, USA). The level of fluorescence in a given region was determined by using the following equation: corrected total cell fluorescence = integrated density (area of selected cell × mean fluorescence of background readings). The corrected total cell fluorescence (CTCF) values, which are arbitrary numbers, were graphed, and standard deviations were calculated. The brightness in some images was increased to better show staining. Significance was determined between 2 groups was via Student’s t test. Values of P ≤ 0.05 were considered significant.

In vitro model of intestinal inflammation

All in vitro studies were completed using the small intestinal crypt-derived IEC-6 cell line (American Type Culture Collection, Manassas, VA, USA). Intestinal epithelial cell (IEC)-6 cells were grown in DMEM (Thermo Fisher Scientific) supplemented with 5% fetal bovine serum (Thermo Fisher Scientific) and 2 mM glutamine, 4 μg/ml insulin, and 1% penicillin-streptomycin (MilliporeSigma) (25). IEC-6 cells were treated with cytomix (10 μM hydrogen peroxide (MilliporeSigma) and a cytokine cocktail: 100 ng/ml TNF-α, 100 ng/ml IL-1β, and 5 μg/ml IFN-γ (all from Thermo Fisher Scientific) for 60 min at 37°C. In some experiments, cells were pretreated with 20 μM 1,2-bis(2-aminophenoxy)ethane-N,N,N’,N’-tetraacetic acid tetrakis(acetoxymethyl ester) (BAPTA-AM) (Thermo Fisher Scientific), a cell-permeant calcium chelator, for 3 h at 37°C before cytomix treatment.

Statistics

Results are expressed as means ± sd. Statistical analysis was performed using Prism software (GraphPad, La Jolla, CA, USA). ANOVA was used for comparisons in experiments involving more than 2 experimental groups. A 2-tailed Student’s t test was used for comparison in experiments consisting of 2 experiment groups. A nonparametric t test (Mann-Whitney) was used to analyze the human qPCR data because of the wide spectrum of human disease. In all cases, differences between groups were considered significant with values of P ≤ 0.05.

RESULTS

CaMKIV is activated in human IBD and murine experimental colitis

To characterize CaMKIV activation in the intestine during IBD, we obtained surgical samples from patients with ulcerative colitis or Crohn’s disease and from control patients. CaMKIV mRNA expression was significantly increased in intestinal specimens from patients with ulcerative colitis and Crohn’s disease, relative to those from control patients (Fig. 1A). Focusing then on patients with ulcerative colitis, we noted that by immunofluorescent staining, CaMKIV was clearly expressed at baseline in the intestinal epithelium of control tissue and that the staining intensity of CaMKIV appeared to be significantly increased in diseased tissue relative to control tissue (Fig. 1B, C). Because CaMKIV displayed higher gene expression and apparent staining intensity in the intestines of patients with ulcerative colitis, we analyzed the levels of phosphorylated (activated) CaMKIV protein in these patients. Protein levels of CaMKIV via Western blot correlated with mRNA and immunofluorescence data (Fig. 1D). The elevated CaMKIV protein in these patients was the active p-Thr^196–200 CaMKIV (pCaMKIV) form. We therefore hypothesized that CaMKIV would be similarly increased in the intestines of mice subjected to experimental colitis. WT C57BL/6 mice were subjected to 2% DSS–induced colitis for 7 d, and as expected, our results in murine intestinal tissue mirrored the results in human tissue. Immunofluorescent staining confirmed that CaMKIV was evident within the intestinal epithelium at baseline. Staining was increased in mice with DSS-induced colitis, and this increase in staining was located predominantly within the IECs (Fig. 2A, B). Based on this result, parallel findings using qPCR and Western blot on whole-tissue lysates likely reflect epithelial changes (Fig. 2C–E). Thus, these data implicate CaMKIV activation in the pathogenesis of human and murine colitis.

Figure 1.

CaMKIV is upregulated in human colitis and is present in the intestinal epithelium. A) Human intestinal tissue from patients with or without IBD (Crohn’s disease and ulcerative colitis) was analyzed by qPCR for CaMKIV gene expression. B) Immunofluorescence for CaMKIV (green), E-cadherin (red), and DAPI (blue) was performed in human control and IBD intestinal tissue samples. CaMKIV was present within the intestinal epithelium and was increased in diseased tissue. Image in the small solid outline on the left side of each representative image is a magnified view of the image within the small dotted outline. Scale bar, 50 μm. C) Intensity of CaMKIV immunofluorescence was determined by calculating the CTCF in control and diseased tissue sections. Calculations show an increase in CaMKIV intensity in ulcerative colitis intestinal tissue vs. control tissue. D) Human tissue from patients with ulcerative colitis and control patients were analyzed for phosphorylated and total CaMKIV protein levels via Western blot (n = 5–9/group). Values are means ± sd.

Figure 2.

Intestinal CaMKIV is upregulated in murine experimental colitis. C57BL/6 (WT) mice were subjected to 2% DSS induction of colitis for 7 d (n = 18/group). Murine intestinal tissue from control and DSS-induced mice was analyzed. A) Immunofluorescence for CaMKIV (green), E-cadherin (red), and DAPI (blue) was performed. At baseline, CaMKIV was expressed within the intestinal epithelium of WT control mice. After provocation with 2% DSS, the signal in the intestinal epithelium intensified. Image in the small solid outline on the left side of each representative image is a magnified view of the image within the small dotted outline. Images shown are a representative image. Scale bar, 50 μm. B) Intensity of CaMKIV immunofluorescence was determined by calculating the CTCF in colonic tissue from control and DSS-induced mice. CTCF values for CaMKIV are increased in WT mice subjected to DSS. C) CaMKIV mRNA expression was measured using qPCR in colonic tissue from control and DSS-induced mice. D) Phosphorylated CaMKIV protein levels were analyzed via Western blot. Values are means ± sd. E) Densitometry illustrates that there is a significant increase in the ratio of phosphorylated/total CaMKIV in intestinal tissue from control vs. patients with ulcerative colitis.

CaMKIV-deficient mice are protected from experimental colitis

We next explored whether CaMKIV was involved in the pathogenesis of colitis, although first we characterized the intestinal phenotype of the Camk4^tm1Tch/Camk4^tm1Tch null mutant (CaMKIV KO) mouse. We observed reduced CaMKIV expression in the intestines of CaMKIV KO mice compared to WT mice (Fig. 3A). However, there were no significant differences in the colon lengths (Fig. 3B), proliferative indices (Fig. 3C), number of goblet cells (Fig. 3D, E), or the integrity of the intestinal epithelium between strains at baseline (Fig. 3F). We therefore concluded that the CaMKIV KO mice were healthy and phenotypically normal at baseline.

Figure 3.

Characterization of the intestine at baseline in CaMKIV KO vs. WT mice. The intestinal epithelium of the CaMKIV KO mouse was characterized before any experimental procedures. A) CaMKIV mRNA expression was measured by qPCR in colonic tissue from WT control and CaMKIV KO control mice. B) There was no difference in colon lengths between WT control and CaMKIV KO control mice at baseline. C) A comparable proliferative index analyzed via BrdU incorporation was seen in both WT and CaMKIV KO control mice. D) Alcian blue staining revealed no significant difference in the number of goblet cells in the intestine between the strains. Images are representative. Original magnification, ×20. E) There is no significant difference in the gene expression of Muc2. F) FITC-dextran oral gavage showed there was no significant difference in the integrity of the intestinal epithelium between strains at baseline. Values are means ± sd. NS, not significant.

We next subjected CaMKIV KO and WT mice to 2% DSS colitis for 7 d. As expected, CaMKIV-deficient mice developed a significantly less severe clinical colitis than WT mice, as evidenced by decreased weight loss (Fig. 4A) and decreased DAI scores (Fig. 4B). Furthermore, CaMKIV KO mice displayed significant preservation of intestinal architecture with an intact epithelium and minimal immune cell infiltrate relative to WT control mice (Fig. 4C, D). We next analyzed the integrity of the intestinal epithelial barrier by using FITC-dextran via oral gavage on d 7. CaMKIV KO mice demonstrated significantly less FITC-dextran serum absorbance compared to WT mice (Fig. 4E), implying that CaMKIV adversely influences the permeability of the intestinal epithelium during inflammation. In addition, when we evaluated CaMKIV KO intestinal tissue by qPCR, we found minimal mRNA expression of proinflammatory cytokines known to be increased in patients with IBD and mice with experimentally induced colitis, including TNF-α, IL-1β, iNOS, and IL-6 (Fig. 4F). Thus, these data suggest that CaMKIV activation influences the development and progression of experimental colitis.

Figure 4.

CaMKIV KO mice are protected from experimental colitis. WT and CaMKIV KO mice were subjected to 2% DSS for 7 d. A, B) CaMKIV-deficient mice developed significantly less severe clinical colitis than WT mice, as evidenced by decreased weight loss (A) and decreased DAI scores (B), with a significant difference between WT DSS and CaMKIV KO DSS from d 5 to 7. C) Hematoxylin and eosin staining illustrates in representative images that CaMKIV KO mice displayed significant preservation of intestinal architecture with an intact epithelium and minimal immune cell infiltrate after provocation with 2% DSS vs. WT control mice. Original magnification, ×20. D) Average histology scores, as analyzed by a blinded pathologist. E) CaMKIV-deficient mice demonstrated significantly less FITC-dextran serum absorbance than WT mice. F) Intestinal tissue from WT control and DSS-induced mice and CaMKIV control and DSS-induced mice was analyzed by qPCR for the gene expression of proinflammatory cytokines known to be increased in patients with IBD and mice with induced experimental colitis. G) Levels of oxidized DNA damage was assessed by immunofluorescence with an antibody specific for 8OHdG—a marker for oxidized DNA. CaMKIV KO mice subjected to DSS showed similar levels of oxidized DNA damage compared to WT DSS-induced mice. Images shown are representative images. Scale bar, 50 μm. Values are means ± sd. NS, not significant.

We next sought to determine whether the CaMKIV KO mice were protected from the initial cellular damage inflicted by DSS. Given that DSS is known to cause oxidized DNA damage within IECs, we analyzed levels of 8OHdG, a marker for oxidized DNA (Fig. 4G). WT and CaMKIV KO mice displayed similar levels of oxidized DNA damage in the intestinal epithelium during DSS-induced colitis (Fig. 4G). These data suggest that although CaMKIV is contributory in the development of experimental colitis, it does not protect IECs from the initial genotoxic stress of DSS exposure.

Activated CaMKIV facilitates CREB activation during experimental colitis

Our data demonstrate that although the absence of CaMKIV resulted in protection against DSS-induced intestinal inflammation, the IECs were still susceptible to DNA damage. Therefore, we hypothesized that CaMKIV KO mice are protected from disease because the absence of CaMKIV disrupts a signaling system that alerts the cell to damage or promotes secondary healing within the intestinal mucosa. Activated CaMKIV is known to regulate the activity of cyclic AMP response element-binding protein (CREB), a nuclear transcription factor activated by various stimuli and that regulates the expression of many genes (13). We observed that the intestines of CaMKIV KO mice demonstrated reduced phosphorylated CREB:total CREB ratios as compared to those of WT mice (Fig. 5A, B). Furthermore, to illustrate that the increase in phosphorylated CREB correlates with an increase in transcriptionally activated CREB, we analyzed nuclear extracts from WT and CaMKIV KO intestinal lysates in an EMSA. The nuclear extracts of CaMKIV KO DSS-induced mice exhibited decreased levels of CREB-DNA complex formation relative to WT DSS-induced mice (Fig. 5C), which was consistent with the levels of phosphorylated CREB in WT and CaMKIV KO mice subjected to DSS (Fig. 5A). Furthermore, when intestinal tissues were evaluated by qPCR, CREB gene expression was increased in diseased WT mice compared to CaMKIV KO mice (Fig. 5D). Finally, we analyzed intestinal tissue by immunofluorescence to determine the location of CREB activation. The increase in phosphorylated CREB was seen almost entirely within the intestinal epithelium of WT mice subjected to DSS (Fig. 5E, F). Moreover, there was a significant decrease in the levels of phosphorylated CREB in the intestinal epithelium of CaMKIV mice subjected to DSS. Taken together, these data suggest that CaMKIV mediates CREB activation within the intestinal epithelium during DSS-induced colitis.

Figure 5.

Activation of CaMKIV during intestinal insult induces the activation of CREB. A) Intestinal protein lysates from WT and CaMKIV KO DSS-subjected mice demonstrated an increase in phosphorylated CREB in WT mice vs. CaMKIV KO mice. B) Densitometry illustrates that there was a significant increase in the ratio of phosphorylated/total CREB in WT control vs. WT DSS-induced mice. There is no significant increase in CaMKIV KO mice subjected to DSS. C) EMSAs were performed with oligonucleotides specific for CREB. Incubation of nuclear extracts from DSS-induced WT mice with labeled CREB oligonucleotide resulted in complex formation (as evidenced in the shift of labeled DNA probe). Incubation of nuclear extracts from DSS-induced CaMKIV KO mice with labeled CREB oligonucleotide did not result in complex formation. D) Intestinal tissues were evaluated by qPCR for CREB gene expression. E) Immunofluorescence for phosphorylated CREB (red), E-cadherin (green), and DAPI (blue) was performed in murine intestinal tissue samples. Phosphorylated CREB was present within the intestinal epithelium and was increased in WT mice subjected to DSS. Small solid box on left side of each image is a magnified view of the small dotted boxes. Images shown are a representative image. Scale bar, 50 μm. F) Intensity of phosphorylated CREB immunofluorescence was determined by calculating the CTCF in control and diseased tissue sections. Calculations show an increase in the intensity of phosphorylated CREB in the intestinal epithelium of WT mice subjected to DSS intensity. Values graphed are means ± sd.

CaMKIV-dependent activation of CREB upregulates cell cycle proarrest signals during experimental colitis

Our data suggest that during an intestinal insult, CaMKIV is activated, which in turn, activates CREB. Activated CREB is known to mediate a p53-dependent downregulation of proliferation (15). Indeed, our data suggest that one mechanism by which CaMKIV activation influences the development and progression of colitis is through a disruption of normal IEC proliferation (Fig. 4C). Thus, we analyzed intestinal tissue from WT control and WT DSS-induced mice by qPCR and found a significant increase in the transcript levels of p53 in the diseased mice, which was markedly attenuated in the intestines of CaMKIV KO mice (Fig. 6A). Previous research has shown that CREB-induced activation of p53 results in an increase in the transcriptional activity of p53 (15). One of the downstream targets of p53 is the cell cycle proarrest gene p21, which halts progression of cell division when there is the detection of cellular damage (26–28). We found a significant increase in the transcript levels of p21 in WT mice induced with DSS relative to WT control mice. Similarly, the protein expression of p53 and p21 were increased in WT mice during colitis, which was not observed in CaMKIV KO mice (Fig. 6B, C). This suggests that the absence of CaMKIV interrupts a signaling system intended to halt cellular proliferation in the setting of an insult. To further prove that CaMKIV activation causes an inhibition of cellular proliferation, we analyzed the proliferative index of the intestinal epithelium of WT and CaMKIV KO control and DSS-induced mice by BrdU staining (Fig. 6D, E). There was a significant decrease in the number of proliferating cells per crypt in WT mice with DSS- induced colitis relative to the WT control mice. The number of proliferating cells per crypt in CaMKIV KO mice was significantly higher when compared with WT DSS-induced mice (Fig. 6D, E). These data support the hypothesis that during intestinal injury, phosphorylated CaMKIV inhibits proliferation of the intestinal epithelium, at least in part, through CaMKIV-dependent CREB activation.

Figure 6.

Activation of CREB results in cell cycle arrest during intestinal inflammation. A) Intestinal tissues from WT control and DSS-induced mice and CaMKIV control and DSS-induced were evaluated by qPCR for p53 and p21 gene expression. B) Western blot analysis showed an increase in the protein levels of p53 and p21 in WT mice subjected to DSS. CaMKIV KO mice subjected to DSS did not show an increase in the protein levels of p53 and p21. C) Densitometry of immunoblots in B. D) There is a significant decrease in the number of proliferating cells per crypt in WT mice subjected to DSS vs. WT control mice by BrdU analysis. The number of proliferating cells per crypt in CaMKIV KO mice was significantly higher vs. WT DSS-induced mice. Images are representative. E) BrdU⁺ cells per crypt were counted for each strain (at least 2 crypts per slide and at least 12 different mice). F) Intestinal lysates from control and diseased human tissue samples were analyzed by Western blot for the protein levels of CREB, p53, and p21. G) Densitometry illustrates that there is a significant increase in the phosphorylated:total CREB ratio, p53, and p21 in intestinal tissue from control vs. patients with ulcerative colitis. Values are means ± sd. NS, not significant.

CaMKIV activation leads to upregulation of cell cycle proarrest pathway in ulcerative colitis

We show that activation of CaMKIV during experimental colitis results in the induction of cell cycle proarrest signals, leading to an inhibition of cellular proliferation. Because we initially demonstrated an increase in CaMKIV activation in the intestines of patients with ulcerative colitis, we wanted to further confirm the presence of this CaMKIV-induced pathway in human patients. We analyzed the levels of these proteins in intestinal lysates from control and diseased human tissue samples (Fig. 6F, G). We found that the levels of phosphorylated CREB were increased in patients with ulcerative colitis. We also found an increase in the expression of p53. Furthermore, we found a dramatic increase in the levels of p21. Together with previous results illustrating that CaMKIV is activated in human intestinal inflammation, these data suggest that CaMKIV-dependent inflammation induces a downstream pathway that inhibits cellular proliferation.

CaMKIV activation of cell cycle proarrest signals is calcium dependent

We have demonstrated that CaMKIV activation induces a signal transduction cascade that results in inhibition of cellular proliferation in both murine experimental colitis and human ulcerative colitis. CaMKIV activation is dependent on calcium (12, 29). Thus, we explored whether the activation of CaMKIV during intestinal inflammation is regulated by a calcium signal. In an in vitro model of intestinal inflammation, we first demonstrated that CaMKIV activation and upregulation of the downstream proarrest cell signaling pathway was indeed present within our mode (Fig. 7A, B). IEC-6 cells exposed to cytomix exhibited an increase in pCaMKIV and CREB protein levels. There was an associated increase in the protein levels of p53 and p21. Chelation of intracellular calcium with BAPTA-AM inhibited CaMKIV activation and decreased protein levels of both p53 and p21. p21 is a cyclin-dependent kinase (CDK) inhibitor that tightly regulates the activity of cyclin/CDK complexes, and thus progression of the cell cycle. We show that IEC-6 cells treated with cytomix have a marked reduction in the levels of cyclin-D1 and -D3, but no change in cyclin E. Pretreatment with BAPTA-AM, however, leads to a preservation of cyclin-D1 and -D3 protein levels. These data suggest that activation of CaMKIV and the subsequent CaMKIV-dependent induction of cell cycle proarrest signals during an intestinal epithelial insult are dependent upon a cellular calcium signal.

Figure 7.

Activation of CaMKIV and subsequent induction of cell cycle arrest in an in vitro model of intestinal inflammation is dependent upon calcium signals. A) Western blot analysis showed that pretreatment of IEC-6 cells exposed to cytomix with BAPTA-AM, a cell-permeant calcium chelator, inhibited activation of CaMKIV and the downstream cell arrest signaling pathway, including p53 and p21. B) Western blot analysis showed that the positive regulators of cell cycle G₁-S phase transition cyclin-D1 and -D3 were markedly decreased in IEC-6 cells treated with cytomix, but not in IEC-6 cells treated with BAPTA+cytomix.

DISCUSSION

IBD is known to be a multifaceted disease, involving the interplay of genetic predisposition, immune dysregulation, microbial dysbiosis, and environmental influence. In this study, we demonstrated that activation of the multifunctional serine/threonine protein kinase CaMKIV is clinically important in the pathogenesis of colitis in both humans with severe IBD and mice undergoing experimental colitis. Specifically, phosphorylated CaMKIV (active) enhanced the proinflammatory response largely through an inhibition of IEC proliferation, contributing to an increase in intestinal permeability during the pathogenesis of colitis (Fig. 8). CaMKIV has been shown to play important roles in various cellular functions, including the regulation of gene transcription, sperm motility, apoptosis, cell cycle progression, as well as T-cell activation, proliferation, differentiation, and effector function. CaMKIV is predominately expressed in neural tissue and the thymus and has been extensively studied in diseases of the CNS and autoimmune conditions. In this study, we showed the significance of CaMKIV expression in IECs during intestinal inflammation, including IBD. The CaMKIV-dependent mechanisms involve an activation of the downstream transcription factor CREB and an increase in both the transcription and protein levels of p53, which results in an arrest in cellular proliferation via p21, a prosurvival protein. This CaMKIV-dependent decrease in cellular proliferation contributes to increased intestinal permeability and clinical disease during murine experimental colitis. These data suggest that CaMKIV activates a key proarrest pathway during human and murine intestinal inflammation, which limits the proliferative ability of epithelial cells during an ongoing cellular intestinal insult.

Figure 8.

CaMKIV activation plays a critical role in the pathogenesis of IBD. This figure illustrates our current working hypothesis, which outlines a mechanism by which CaMKIV activation leads to intestinal inflammation. Our data support the presence of CaMKIV in IECs and suggest that during intestinal inflammation CaMKIV is activated by a calcium-dependent signal. Activation of CaMKIV induces the activation of the transcription factor CREB, which transactivates p53. p53 upregulates the expression of the proarrest signal p21, which inhibits proliferation of the intestinal epithelium. We hypothesize that this upregulated pathway prevents proliferation of the intestinal epithelium during intestinal inflammation, which ultimately results in a compromised intestinal epithelial barrier and further perpetuates intestinal inflammation.

An intact intestinal epithelial barrier is crucial to maintaining intestinal homeostasis (30). Defects in intestinal barrier function are associated with both Crohn’s disease and ulcerative colitis. Tight junctions, which seal the space between adjacent epithelial cells, are a key determinant of intestinal permeability in the absence of injury (i.e., ulceration) (30, 31). We demonstrated in this study that CaMKIV KO mice subjected to DSS demonstrate preservation of intestinal architecture. Furthermore, our FITC-dextran analysis suggests that CaMKIV adversely influences the permeability of the intestinal epithelium during inflammation. Although we attribute the barrier dysfunction of the intestinal epithelium in the WT mice to an inhibition of proliferation within the IECs, we cannot exclude the possibility that CaMKIV also regulates tight junction proteins during inflammation, either directly, or indirectly, by modulating the expression of proinflammatory cytokines. CaMKII has been demonstrated to influence barrier function through regulation of tight junctions (32). Shiomi et al. (32) showed that inhibition of CaMKII results in an increase of tight junction protein levels and a subsequent upregulation of epithelial barrier function. Further research is needed, however, to confirm whether CaMKIV is directly involved in modulating the function of tight junctions within the intestinal epithelium. The functional properties of tight junction proteins are also known to be influenced by proinflammatory cytokines, such as TNF-α (33–35), IL-1β (36–38), and IL-6 (39). CaMKIV KO mice subjected to DSS demonstrated significantly lower mRNA levels of proinflammatory cytokines in the intestinal epithelium, as compared to WT mice similarly treated. However, it is difficult to discern whether a decrease in proinflammatory cytokines is solely responsible for the observed protection from detrimental changes in intestinal permeability in CaMKIV KO mice. It is also interesting to note the importance of calcium in the regulation of tight junctions. Studies have shown that extracellular calcium is essential not only for the stabilization of mature tight junctions (40–42), but also the development of new junctions (43–47). Understanding the regulation and effects of calcium signaling during intestinal inflammation is a current direction of our laboratory.

We demonstrated from our in vitro data that activation of CaMKIV is dependent on intracellular calcium signals. However, the exact source of calcium responsible for activating CaMKIV during intestinal inflammation remains unknown. Arnould et al. (15) demonstrated that mitochondrial dysfunction can lead to an increase in cytosolic free calcium which, in turn, can activate CaMKIV and cause increased levels of phosphorylated CREB protein in several mammalian cell lines. Our laboratory has recently shown that there is an increase in mitochondrial dysfunction in the intestinal epithelium during human and murine experimental colitis (16). Mitochondrial stress is known to alter cellular calcium levels, activating various calcium-dependent signaling proteins, including CaMKIV (48, 49). Furthermore, it has been demonstrated that nuclear gene expression is altered in response to mitochondrial dysfunction (50, 51), with CaMKIV being one of several key proteins implicated in communicating this retrograde signal (15, 52, 53). However, further studies are needed to determine the exact role that CaMKIV plays in relaying calcium stress signals from the mitochondria to the nucleus during intestinal inflammation. Nonetheless, these data highlight how calcium-dependent signaling within the intestinal epithelium contributes to inflammation and likely interferes with epithelial healing during intestinal inflammation.

We cannot exclude the role that the endoplasmic reticulum (ER) stress plays in inducing CaMKIV and this proarrest pathway. ER stress, characterized by the build-up of unfolded and misfolded proteins in the ER lumen, leads to a disruption of cellular calcium homeostasis (54). Human genome-wide association studies have associated primary genetic abnormalities in several genes involved in ER protein folding and homeostasis with IBD (55, 56). Recent studies have also highlighted the importance of ER stress in the homeostasis of the intestinal epithelium and have demonstrated that ER stress is a major driver of intestinal inflammation (55, 57–64). The ER interacts directly with ∼20% of the mitochondrial surface at contact points called mitochondria-associated membranes (65, 66). These physical connections play a critical role in various cellular functions, including lipid transport, energy metabolism, and cell survival (67–71). Mitochondria-associated membranes also play a pivotal role in calcium signaling between these 2 organelles (71). Moreover, tight modulation of ER–mitochondria calcium transfer is an important governor of prosurvival and prodeath pathways. Nonetheless, the free cytosolic calcium needed to activate CaMKIV during intestinal inflammation may derive from the mitochondria, ER, or both.

Progression through the cell cycle is controlled by several checkpoints that coordinate the various phases of cell division and link the cell cycle with extracellular signals that regulate cell proliferation (72, 73). Cellular damage caused by extrinsic or intrinsic factors is capable of activating cell cycle checkpoints, resulting in arrest of the cell cycle, which allows the cell to repair any damage before continuing (72–74). The p53 protein is activated by an array of cellular stresses and has a central role in deciding the fate of the cell through inducing arrest or apoptosis (27). We show here that CaMKIV-induced CREB activation during experimental colitis is associated with increased levels of p53, which appears to result in up-regulation of p21 and subsequent cell cycle arrest. p21 is a cyclin-dependent kinase (CDK) inhibitor that tightly regulates the activity of cyclin/CDK complexes, and thus progression of the cell cycle. Cyclin-D and -E, however, drive G₁-S phase transition and oppose the activity of p21. In our in vitro model, we show that IEC-6 cells treated with cytomix demonstrate a marked reduction in protein levels of cyclin-D1 and cyclin D3. Cyclin-D1 dimerization with CDK-4 and -6 is the rate-limiting step of G₁-S phase transition (75). Further, studies have shown that degradation of cyclin-D1 is sufficient to induce cell cycle arrest (76). Our data suggest that cytomix induces G₁ phase cell cycle arrest. Furthermore, since the levels of cyclin-D1 are maintained in cells pretreated with BAPTA-AM before cytomix exposure, calcium chelation appears to prevent cytomix-induced cell cycle arrest. Taken together, these experiments lend weight to the hypothesis that CaMKIV activation via intracellular calcium signaling during inflammation induces cell cycle arrest.

DNA damage is a key activator of p53-induced cell cycle arrest (27); however, the exact cause of cell cycle arrest and inhibition of proliferation regulated by p53 during human and experimental colitis remains unknown. We show here that p53-induced cell cycle arrest during an in vitro model of intestinal inflammation is contingent upon calcium, apparently through activation of CaMKIV. Nonetheless, although cell cycle arrest and damage repair might be beneficial in some diseases such as cancer (77–79), our data suggest that this is not the case in acute intestinal inflammation. Damage to the intestinal epithelium enhances and perpetuates mucosal inflammation by allowing translocation of luminal antigens and microbes (5, 80). Thus, given that intestinal epithelial homeostasis is dependent on healthy IECs and intact intestinal stem cell proliferation (81), it is reasonable to expect that inhibition of cellular proliferation in the setting of an intestinal epithelial insult would exacerbate barrier dysfunction and perpetuate inflammation (5, 80). Indeed, our data suggest that during experimental colitis, activation of CaMKIV in the intestines of WT mice facilitates cell cycle arrest and subsequent inhibition of intestinal epithelial proliferation, which leads to a loss of integrity and increased permeability of the gut barrier. In contrast, mice deficient in CaMKIV maintained epithelial cell proliferation during experimental colitis and fared much better against the development of disease. However, further research is necessary to differentiate the role CaMKIV plays in healing vs. proliferation of the intestinal epithelium during inflammation.

It is important to note that other factors may influence the activity of CREB during intestinal inflammation. For example, existing evidence would suggest that eicosanoids might impact intestinal epithelial homoeostasis. Eicosanoids are enzymatically generated oxidation products of arachidonic acid that regulate a plethora of physiologic and pathologic responses (35). Arachidonic acid can be metabolized by cyclooxygenases (COXs) to produce prostaglandins and thromboxanes or by 5-lipoxygenase to produce 5-hydroxyeicosatetraeonoic acid, leukotriene A₄ and cysteinyl leukotrienes. Eicosanoids have been reported to play a role in IEC growth. Cabral et al. (82) showed that prostaglandin E₂ regulates epithelial cell growth through CREB phosphorylation. However, Paruchuri et al. (83) demonstrated that leukotriene D4 mediates IEC survival through phosphorylation of CREB. Nonetheless, eicosanoids have been shown to have both proinflammatory and anti-inflammatory effects in IBD. In addition, a recent study on the production of lipid mediators over a DSS time course in mice concluded that COX-derived eicosanoids are mainly produced during the induction phase of DSS colitis, whereas the lipoxygenase-derived eicosanoids were primarily synthesized during the recovery phase (84). One study has also shown that CaMKII regulates expression of COX-2 expression and the production of prostaglandin-E2 by activating CREB in rat peritoneal macrophages (85). However, further research is needed to fully understand the role of eicosanoids during intestinal inflammation and whether CaMKIV and CREB are regulated, in part, by eicosanoids.

We demonstrated the importance of CaMKIV activity in the intestinal epithelium during intestinal inflammation, but it is important to note that CaMKIV is strongly expressed in lymphocytes, specifically CD4⁺ T cells (86). Moreover, CaMKIV is known to regulate T-cell activation, proliferation, differentiation, and effector function (13, 87, 88). Because intestinal infiltration and activation of lymphocytes is a hallmark of IBD, it is possible that CaMKIV exerts important regulatory effects on effector T cells during IBD pathogenesis (89–91). This is an important area of ongoing investigation in our laboratory. However, it is clear from these studies that upregulation of CaMKIV activity in the intestine during human IBD and murine experimental colitis occurs primarily within the epithelial compartment. As such, these investigations highlight a novel mechanism by which CaMKIV promotes intestinal inflammation through an arrest of epithelial proliferation.

IBD is known to be a multifaceted disease, involving the interplay of genetic predisposition, immune dysregulation, microbial dysbiosis, and environmental influence. The intestinal epithelium comprises the interface between these factors, and as such, plays a critical role in governing this interplay. A predominant characteristic of IBD is recurrent damage to the intestinal epithelium concurrent with decreased intestinal barrier function (92–94). Our results presented here suggest that continued regeneration of the intestinal epithelium is a key factor in combating intestinal inflammation (95–97). Likewise, clinical evidence has illustrated that complete mucosal healing of the intestinal epithelium is associated with long-term remissions, decreased risk of hospitalizations and surgery, and improved quality of life in patients with IBD (98–102). Thus, new therapeutics targeting signaling pathways that enhance the health of the intestinal epithelium would prove beneficial in promoting mucosal healing during intestinal inflammation.

In summary, we demonstrated that CaMKIV activation in the intestinal epithelium may be a significant contributor to intestinal inflammation. Furthermore, CaMKIV-induced CREB activation appears to result in the arrest of cellular proliferation in the intestinal epithelium, which, when in the presence of an ongoing inflammatory insult, further exacerbates disease. These results not only offer insights into the molecular mechanisms underlying the development and progression of colitis, but also identify potential targets for this highly morbid disease.

Glossary

BAPTA-AM

1,2-bis(2-aminophenoxy)ethane-N,N,N’,N’-tetraacetic acid tetrakis(acetoxymethyl ester)

BrdU

bromodeoxyuridine

CaMK

CaM kinase

CaMKIV

calcium/calmodulin-dependent protein kinase IV

CaMKIV KO

CaMKIV knockout

CDK

cyclin-dependent kinase

COX

cyclooxygenase

CREB

cyclic AMP response element–binding protein

CTCF

corrected total cell fluorescence

DAI

disease activity index

DSS

dextran sodium sulfate

ER

endoplasmic reticulum

FITC-dextran

fluorescein isothiocyanate-dextran

HEPES

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

IBD

inflammatory bowel disease

IEC

intestinal epithelial cell

pCaMKIV

active phosphorylated Thr^196–200 CaMKIV

qPCR

quantitative PCR

RPLO

50s ribosomal subunit protein L15

ssDNA

sheared salmon sperm DNA

TAE

Tris base, acetic acid, and EDTA

UPMC

University of Pittsburgh Medical Center

WT

wild type

AUTHOR CONTRIBUTIONS

K. E. Cunningham and E. A. Novak designed and performed the research, analyzed the data, and wrote the paper; G. Vincent, V. S. Siow, and B. D. Griffith performed the research; S. Ranganathan analyzed the data; M. R. Rosengart, and J. D. Piganelli designed the research and contributed new reagents; and K. P. Mollen designed the research, analyzed data, and wrote the paper.