If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence Address correspondence to: Huimin Yu, MD, PhD, Johns Hopkins University School of Medicine, Division of Gastroenterology and Hepatology, 720 Rutland Avenue, Ross Research Building, Room 933, Baltimore, Maryland 21205.
Division of Gastroenterology and Hepatology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MarylandDepartment of Physiology, Johns Hopkins University School of Medicine, Baltimore, Maryland
The adipokine CTRP3 has anti-inflammatory effects in several nonintestinal disorders. Although serum CTRP3 is reduced in patients with inflammatory bowel disease (IBD), its function in IBD has not been established. Here, we elucidate the function of CTRP3 in intestinal inflammation.
CTRP3 knockout (KO) and overexpressing transgenic (Tg) mice, along with their corresponding wild-type littermates, were treated with dextran sulfate sodium for 6–10 days. Colitis phenotypes and histologic data were analyzed. CTRP3-mediated signaling was examined in murine and human intestinal mucosa and mouse intestinal organoids derived from CTRP3 KO and Tg mice.
CTRP3 KO mice developed more severe colitis, whereas CTRP3 Tg mice developed less severe colitis than wild-type littermates. The deletion of CTRP3 correlated with decreased levels of sirtuin-1 (SIRT1), a histone deacetylase, and increased levels of phosphorylated/acetylated NF-κB subunit p65 and proinflammatory cytokines tumor necrosis factor-α and interleukin-6. Results from CTRP3 Tg mice were inverse to those from CTRP3 KO mice. The addition of SIRT1 activator resveratrol to KO intestinal organoids and SIRT1 inhibitor Ex-527 to Tg intestinal organoids suggest that SIRT1 is a downstream effector of CTRP3-related inflammatory changes. In patients with IBD, a similar CTRP3/SIRT1/NF-κB relationship was observed.
CTRP3 expression levels correlate negatively with intestinal inflammation in acute mouse colitis models and patients with IBD. CTRP3 may attenuate intestinal inflammation via SIRT1/NF-κB signaling. The manipulation of CTRP3 signaling, including through the use of SIRT1 activators, may offer translational potential in the treatment of IBD.
Here, we report that CTRP3 exerts its anti-inflammatory effect on intestinal inflammation through SIRT1/NF-κB signaling by using in vivo complementary KO and Tg mouse colitis models and ex vivo intestinal organoid cultures. In patients with IBD, reduced CTRP3 expression in the intestinal mucosa correlated with reduced SIRT1 and increased NF-κB activation and proinflammatory cytokines.
Inflammatory bowel disease (IBD), including Crohn's disease (CD) and ulcerative colitis (UC), is a chronic, debilitating inflammatory disorder of the gastrointestinal tract that affects 3 million people in the United States.
There is a pressing need to develop evidence-based, novel strategies to control refractory inflammation in IBD.
The complex interplay between mesenteric fat and intestinal inflammation in IBD has drawn interest since Drs Crohn and Otani first recognized the presence of hyperplastic mesenteric fat wrapped around the inflamed intestine as an indicator of severe disease.
Molecular, biochemical and functional characterizations of C1q/TNF family members: adipose-tissue-selective expression patterns, regulation by PPAR-gamma agonist, cysteine-mediated oligomerizations, combinatorial associations and metabolic functions.
In obesity, type 2 diabetes, and metabolic syndrome, reduced CTRP3 level has been found to correlate negatively with levels of several proinflammatory cytokines, including TNF-α, interleukin (IL)-6, IL-8, and C-reactive protein.
Recent murine studies using an adenovirus-mediated delivery system to overexpress CTRP3 suggest that CTRP3 may activate histone deacetylase sirtuin-1 (SIRT1) to exert protective effects in acute pancreatitis,
SIRT1 is a known key repressor of intestinal inflammation in IBD. Specifically, it is thought to suppress phosphorylation and acetylation of the nuclear factor (NF)-κB p65 subunit and thereby reduce expression of the proinflammatory cytokines TNF-α and IL-6.
Thus, we hypothesized that CTRP3 suppresses intestinal inflammation via SIRT1/NF-κB signaling.
To test this hypothesis, we first generated acute DSS-induced colitis models in complementary CTRP3 knockout (KO) and overexpressing transgenic (Tg) mice and demonstrated that CTRP3 expression levels correlate negatively with intestinal inflammation. Second, we found that CTRP3 levels correlate positively with SIRT1 and negatively with phosphorylation and acetylation of NF-κB p65 and proinflammatory cytokines TNF-α and IL-6 in acute colitis models of CTRP3 KO and Tg mice. Third, we added SIRT1 activator resveratrol to KO intestinal organoids and SIRT1 inhibitor Ex-527 to Tg intestinal organoids and established that the SIRT1 accounted for the downstream, anti-inflammatory effects of CTRP3. Fourth, we found a similar CTRP3/SIRT1/NF-κB relationship in the intestinal mucosa of patients with UC and CD. Using in vivo complementary CTRP3 KO and Tg mouse colitis models, ex vivo mouse intestinal organoid cultures, and intestinal mucosa from patients with IBD, we link CTRP3 to a reduction in intestinal inflammation via SIRT1/NF-κB signaling and raise the possibility that the CTRP3 protein and/or SIRT1 activators may represent a novel drug target for IBD. More broadly, our findings add to the growing understanding of the role of adipokines in IBD pathogenesis, and may assist in the development of new, adipokine-based treatments for IBD.
Expression of CTRP3 in Intestinal Tissue and Attached Mesenteric Fat
We conducted immunofluorescence staining of full-thickness sections of wild-type (WT) mouse colonic tissue and found that CTRP3 protein was expressed in intestinal epithelial cells, intestinal smooth muscle cells, and attached mesenteric fat (Figure 1A). Single-molecule RNA fluorescence in situ hybridization of WT mouse colonic tissue confirmed CTRP3 mRNA expression in all 3 cell types (Figure 1B). Colonic tissue from CTRP3 KO mice was used as a negative control in both experiments (data not shown). Notably, CTRP3 mRNA and protein expression levels were highest in the colon, intermediate in the terminal ileum, and lowest in the mesenteric fat (Figure 1C and D). In our studies described here, we focused on the role of CTRP3 in intestinal inflammation.
Deletion of CTRP3 Aggravates Intestinal Inflammation in Dextran Sulfate Sodium–Induced Acute Colitis Models
To elucidate the function of CTRP3 in intestinal inflammation in vivo, we established a DSS-induced acute colitis model in CTRP3 KO mice and their corresponding WT littermates. The CTRP3 KO mice were born at the expected mendelian ratio, and were viable, fertile, and developed normally with no gross abnormal phenotype under basal conditions.
Four different concentrations of DSS (1.2%, 1.5%, 2.0%, and 2.5%) in drinking water were administered to CTRP3 KO and WT mice for 6–10 consecutive days, with 1.2% DSS treatment resulting in the most distinguishable colitis phenotypes between CTRP3 KO and WT mice. DSS of 1.2% was therefore used in subsequent experiments (Figure 2A). After 7 days of 1.2% DSS treatment, CTRP3 KO and WT mice developed obvious colitis, whereas the corresponding water-treated CTRP3 KO and WT mice showed no signs of colitis. Compared with the DSS-treated WT mice, DSS-treated CTRP3 KO mice developed more severe colitis, as evidenced by more pronounced weight loss (Figure 2B), higher colitis disease scores based on the evaluation of body weight change, stool consistency, and rectal bleeding
(Figure 2C), and a lower survival rate (Figure 2D). Histologic analysis of the proximal and distal colon revealed more tissue damage and inflammatory cell infiltrates in DSS-treated CTRP3 KO mice than in DSS-treated WT mice (Figure 2F and G). In addition, the colons of DSS-treated KO mice were significantly shorter than those of DSS-treated WT mice (Figure 2E; 5.45 ± 0.16 cm vs 6.56 ± 0.28 cm; P < .05).
Under basal conditions (water treatment), CTRP3 KO and WT mice showed no difference in clinical colitis phenotypes (Figure 2B and C), and histologic analysis likewise revealed no meaningful differences (Figure 2F and G). However, the colons of water-treated CTRP3 KO mice were significantly shorter than those of WT littermates (Figure 2E; 7.48 ± 0.40 cm vs 8.93 ± 0.24 cm; P < .01), a phenotype not previously reported. The clinical colitis phenotypes of female CTRP3 KO mice were similar to those of male CTRP3 KO mice.
Overexpression of CTRP3 Attenuates Intestinal Inflammation in Dextran Sulfate Sodium–Induced Acute Colitis Models
To further confirm the anti-inflammatory function of CTRP3 in intestinal inflammation, we developed a DSS-induced acute colitis model in CTRP3 overexpressing Tg mice and their corresponding WT littermates. The CTRP3 Tg mice were viable, fertile, and did not display any gross phenotype under baseline conditions.
Several concentrations of DSS (1.5%, 2.0%, and 2.5%) in drinking water were administered for 6–10 days. We found that 2% DSS treatment best distinguished the colitis phenotypes between CTRP3 Tg and WT mice (Figure 3A). After 7 days of 2% DSS treatment, CTRP3 Tg mice developed significantly less severe colitis than their DSS-treated WT mice, experiencing less pronounced weight loss (Figure 3B), lower colitis disease scores (Figure 3C), and a higher survival rate (Figure 3D). Histologic analysis of the proximal and distal colon showed that DSS-treated CTRP3 Tg mice exhibited significantly less tissue damage and fewer inflammatory cell infiltrates than their WT littermates (Figure 3F and G). Furthermore, the colons of DSS-treated CTRP3 Tg mice were significantly longer than those of their DSS-treated WT littermates (Figure 3E; 8.42 ± 0.08 cm vs 7.71 ± 1.05 cm; P < .001).
The effects of CTRP3 overexpression under basal conditions were largely inverse to those of CTRP3 deletion. That is, water-treated CTRP3 Tg and WT mice showed no differences in clinical colitis phenotypes (Figure 3B and C), nor did histologic analysis reveal meaningful differences (Figure 3F and G), but the colons of water-treated CTRP3 Tg mice were significantly longer than those of WT littermates (Figure 3E; 9.72 ± 0.09 cm vs 9.06 ± 0.07 cm; P < .05). Similarly, the clinical colitis phenotypes of female CTRP3 Tg mice were similar to those of male CTRP3 Tg mice.
Thus, CTRP3 expression levels were positively correlated with mouse colon length in basal (water treatment) and colitis (DSS treatment) conditions in CTPR KO and Tg mice, compared with their corresponding WT littermates. Of note, DSS treatment did not result in additional colon length difference between WT and CTRP3 KO/Tg mice.
Deletion of CTRP3 Reduces Intestinal SIRT1 and Upregulates Phosphorylated/Acetylated NF-κB p65 and Proinflammatory Cytokines in Dextran Sulfate Sodium–Induced Acute Colitis Models
The SIRT1/NF-κB signaling complex is known to influence intestinal inflammation in IBD.
To evaluate whether CTRP3 attenuates intestinal inflammation through SIRT1/NF-κB signaling, we compared the colonic expression levels of CTRP3, SIRT1, phosphorylated/acetylated NF-κB p65, and proinflammatory cytokines (TNF-α and IL-6) in CTRP3 KO and WT mice after DSS and water treatment. As expected, colonic CTRP3 levels were undetectable in water- and DSS-treated CTRP3 KO mice (Figure 4A, E, and F). Seven days of DSS treatment resulted in a significant reduction of CTRP3 levels in WT mouse colons (Figure 4A, E, and F). Consistent with previous murine studies,
SIRT1 mRNA and protein levels were reduced in the colonic tissue of the DSS-treated WT mice compared with water-treated WT mice (Figure 4B, E, and F). Compared with WT mice, SIRT1 mRNA and protein levels were lower in CTRP3 KO mice in both water- and DSS-treated groups (Figure 4B, E, and F). NF-κB p65 levels were similar in the colonic tissue of all 4 groups; however, phosphorylated NF-κB p65 levels were significantly higher in CTRP3 KO mice relative to WT mice in water- and DSS-treated groups, respectively (Figure 4E and F). Acetylated NF-κB p65 levels were higher in CTRP3 KO mice relative to WT mice in the DSS, but not in water-treated groups (Figure 4E and F). As expected, colonic mRNA levels of TNF-α and IL-6 were upregulated after DSS treatment in WT and CTRP3 KO mice (Figure 4C and D). In water- and DSS-treated groups, CTRP3 KO mice had upregulated mRNA levels of TNF-α and IL-6 relative to WT mice in the colonic tissue (Figure 4C and D).
In sum, in an acute colitis model (DSS treatment), compared with WT mice, CTRP3 KO mice had decreased colonic levels of SIRT1 and increased levels of phosphorylated/acetylated NF-κB p65, along with increased TNF-α and IL-6. Under basal conditions (water treatment), CTRP3 KO mice exhibited a similar change compared with WT mice except that acetylated NF-κB p65 did not increase in CTRP3 KO mice (Figure 4E and F). These data are consistent with the hypothesis that CTRP3 exerts its anti-inflammatory effect in acute colitis mouse models through SIRT1/NF-κB signaling.
Overexpression of CTRP3 Increases Intestinal SIRT1 and Downregulates Phosphorylated/Acetylated NF-κB p65 and Proinflammatory Cytokines in Dextran Sulfate Sodium–Induced Acute Colitis Models
To further investigate whether CTRP3 may attenuate intestinal inflammation through SIRT1/NF-κB signaling, we performed similar experiments in water- and DSS-treated CTRP3 overexpressing Tg and WT mice. As anticipated, in water- and DSS-treated CTRP3 Tg mice, colonic CTRP3 levels were significantly higher relative to WT mice (Figure 5A, E, and F). Once again, 7 days of DSS treatment correlated with a significant reduction in CTRP3 levels in WT and CTRP3 Tg mouse colons (Figure 5A, E, and F). Compared with WT mice, SIRT1 mRNA and protein levels were higher in CTRP3 Tg mice in water- and DSS-treated groups (Figure 5B, E, and F). As in our CTRP3 KO mice experiments, NF-κB p65 levels had no show of statistical significance in the colonic tissue among 4 groups (Figure 5E and F), and phosphorylated/acetylated NF-κB p65 levels were upregulated after DSS treatment. In the DSS-treated groups, CTRP3 Tg mice had lower phosphorylated/acetylated NF-κB p65 levels compared with WT mice. As expected, colonic mRNA levels of TNF-α and IL-6 were upregulated after DSS treatment in WT and CTRP3 Tg mice (Figure 5C and D). In DSS-treated (but not water-treated) groups, CTRP3 Tg mice had significantly downregulated colonic mRNA levels of TNF-α and IL-6 relative to WT mice (Figure 5C and D).
The results of our mechanistic experiments in CTRP3 Tg mice were largely inverse to those conducted in their CTRP3 KO counterparts. In DSS-induced acute colitis models, CTRP3 overexpression correlated with increased SIRT1 and decreased phosphorylated/acetylated NF-κB p65, associated with decreased proinflammatory cytokines in the colons. However, under basal conditions (water treatment), CTRP3 overexpression correlated only with increased SIRT1 levels, but not phosphorylated/acetylated NF-κB p65 or proinflammatory cytokines in the colons. Taken together, our data from CTRP3 KO and Tg mice suggest that CTRP3 may attenuate intestinal inflammation through SIRT1/NF-κB signaling.
CTRP3 Regulates NF-κB p65 Signaling Via SIRT1 in Mouse Intestinal Organoids Derived from CTRP3 KO and Tg Mice
To investigate further whether SIRT1 is the downstream effector of CTRP3 in intestinal inflammation, we generated 3-dimensional intestinal organoids from the distal colons of CTRP3 KO and Tg mice. Two-dimensional intestinal monolayers on Transwell filters were then developed (150 μL media added in the apical chamber and 600 μL media added in the basolateral chamber). Confluent monolayers were differentiated for 24 hours before treatment as follows: CTRP3 KO organoids were treated with a SIRT1 activator (resveratrol, 100 μM) and CTRP3 Tg organoids were treated with a specific SIRT1 inhibitor (Ex-527, 50 μM). Dimethyl sulfoxide (DMSO) was used for vehicle control groups. The monolayers and apical and basolateral culture media were collected after 24 hours of treatment.
We first determined whether the addition of SIRT1 activator resveratrol to CTRP3 KO organoids could restore the anti-inflammatory effects of CTRP3. SIRT1 levels were lower in the KO-DMSO group compared with the WT-DMSO group. Consistent with our in vivo CTRP3 KO mice experiments, NF-κB p65 levels were similar among 4 groups, whereas phosphorylated/acetylated NF-κB p65 levels were upregulated in the KO-DMSO group compared with the WT-DMSO group (Figure 6A and B). After the addition of resveratrol, SIRT1 levels were increased in KO-resveratrol and WT-resveratrol groups, whereas phosphorylated/acetylated NF-κB p65 levels were significantly downregulated in the KO-resveratrol group, but not in the WT-resveratrol group (Figure 6A and B). In the apical media, TNF-α and IL-6 levels were similar across all 4 groups. In the basolateral media, TNF-α and IL-6 levels were significantly elevated in the KO-DMSO group compared with the WT-DMSO group, and reduced in the KO-resveratrol group, but not the WT-resveratrol (Figure 6C and D). These results suggest that the deletion of CTRP3 results in reduced SIRT1 levels and upregulated NF-κB p65 activity and proinflammatory cytokines; and that the addition of a SIRT1 activator can restore the anti-inflammatory effects of CTRP3 by suppressing NF-κB p65 activity and proinflammatory cytokines levels.
We then determined whether the addition of SIRT1 inhibitor Ex-527 to CTRP3 Tg organoids could reverse the anti-inflammatory effects of CTRP3. As expected, CTRP3 and SIRT1 protein levels were significantly elevated in the Tg-DMSO group relative to the WT-DMSO group. After the addition of Ex-527, CTRP3 levels were not significantly elevated in both Tg and WT groups, whereas SIRT1 levels were significantly reduced in Tg-Ex-527 and WT-Ex-527 groups compared with Tg-DMSO and WT-DMSO, respectively. The addition of Ex-527 upregulated phosphorylated/acetylated NF-κB p65 levels in the Tg groups (Tg-DMSO vs Tg-Ex-527), but not in the WT group (WT-DMSO vs WT-Ex-527) (Figure 6E and F). In the apical media, TNF-α and IL-6 levels were lower in the Tg groups relative to the WT groups (DMSO and Ex-527). In the basolateral media, TNF-α and IL-6 levels were similar between the Tg-DMSO and the WT-DMSO groups. Addition of Ex-527 increased TNF-α and IL-6 levels in the WT-Ex-527 group compared with WT-DMSO, but not in the Tg-Ex-527 group compared with Tg-DMSO (Figure 6G and H). These findings suggest that CTRP3 modulates downstream SIRT1, thereby suppressing NF-κB signaling and reducing inflammatory cytokine synthesis, ultimately attenuating intestinal inflammation.
Inflamed Intestinal Mucosa from Patients with Inflammatory Bowel Disease Express Reduced CTRP3 and SIRT1 and Upregulated Phosphorylated/Acetylated NF-κB p65
We next examined whether patients with IBD exhibit the CTRP3/SIRT1/NF-κB correlations observed in our murine studies. Colonic mucosa was obtained from the ascending colon and grouped based on histologic reports as follows: active UC (n = 5) and healthy control subjects (n = 8). CTRP3 and SIRT1 mRNA and protein levels were reduced in the active UC group relative to the control group (Figure 7A, B, E, and F). NF-κB p65 levels were similar in both groups; however, phosphorylated and acetylated NF-κB p65 levels were significantly higher in the active UC group relative to the control group (Figure 7E and F). As expected, colonic mRNA levels of TNF-α and IL-6 were upregulated in the active UC group (Figure 7C and D).
Similarly, we also obtained 3 types of terminal ileal mucosa from patients with CD who underwent endoscopic or surgical procedures and grouped them based on histologic reports as follows: (1) disease-affected mucosa with discernable ulcers/erosions from patients with CD (active; n = 8); (2) normal-appearing mucosa from the same patients with CD (inactive; n = 8); and (3) normal-appearing mucosa from control patients (control, normal histology; n = 9). The terminal ileum was chosen because it is the most commonly affected intestinal segment to develop creeping fat in CD.
Immunofluorescence staining of the terminal ileal mucosa showed that CTRP3 protein levels were substantially reduced in patients with active CD relative to control patients (Figure 8A-C). Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) analysis of terminal ileal mucosa showed that mRNA levels of CTRP3 and SIRT1 were reduced in the active and inactive CD groups compared with the control groups (Figure 8D). The difference in mRNA levels of CTRP3 and SIRT1 between the inactive and active CD groups was not statistically significant (Figure 8D). Protein levels of CTRP3 and SIRT1 in terminal ileal mucosa were likewise reduced in both inactive and active CD groups relative to the control group, again with no statistically significant difference found between the active and inactive groups (Figure 8E and F). These results are consistent with a bulk RNA-Seq study by Gettler et al
finding lower levels of CTRP3 and SIRT1 in the ileal tissue from the patients with active CD (n = 77) compared with the control subjects (n = 41). Similarly, NF-κB p65 and phosphorylated NF-κB p65 were significantly higher in the active CD groups relative to the control and inactive CD groups (Figure 8E and F). Again, no statistically significant difference was observed in NF-κB p65 or phosphorylated NF-κB p65 between either the control and inactive groups or the inactive and active CD groups (Figure 8E and F). Acetylated NF-κB p65 from the terminal ileal mucosa was highest in the active CD group, higher in the inactive CD group, and lowest in the control group (Figure 8E and F).
Taken together, these data indicate that, compared with control subjects, patients with UC and CD with histologically active inflammation have reduced CTRP3 and SIRT1 and increased phosphorylated/acetylated NF-κB p65 in the intestinal mucosa. These findings are consistent with our murine studies and suggest that CTRP3 may attenuate intestinal inflammation through SIRT1/NF-κB signaling in patients with IBD.
Mesenteric fat and its secreted adipokines are increasingly recognized as integral to IBD pathophysiology. Here we report the function of a novel adipokine, CTRP3, in intestinal inflammation. CTRP3 has known anti-inflammatory effects in several inflammatory states (eg, obesity and diabetes) and is found in reduced serum levels among patients with IBD. However, the function of CTRP3 in intestinal inflammation remains unknown. Here, we provide the first in vivo evidence that CTRP3 attenuates intestinal inflammation by modeling acute colitis in CTRP3 KO and overexpressing Tg mice. CTRP3 KO mice exhibited more severe colitis, whereas CTRP3 Tg mice exhibited less severe colitis, compared with their corresponding WT littermates, suggesting that CTRP3 is a protective adipokine that suppresses intestinal inflammation.
We offer a putative model that helps to understand the molecular mechanisms through which CTRP3 exerts this anti-inflammatory effect on intestinal inflammation by using in vivo complementary KO and Tg mouse colitis models, ex vivo mouse intestinal organoid cultures, and human intestinal mucosa from patients with IBD. Our study correlates concurrent reductions of CTRP3 and SIRT1 with upregulated phosphorylated/acetylated NF-κB p65 and proinflammatory cytokines TNF-α and IL-6. Although the precise signaling relationship between CTRP3 and SIRT1 remains unknown, the effects of introducing a SIRT1 activator to KO intestinal organoids and a SIRT1 inhibitor to Tg intestinal organoids further suggest that CTRP3 exerts its anti-inflammatory effect in IBD by altering SIRT1 levels and thereby suppressing NF-κB p65 activity.
We also observed that intestinal CTRP3 levels (in addition to serum levels) were reduced in colitis mouse models and patients with IBD. We located the expression of CTRP3 in intestinal epithelial cells, intestinal smooth muscle cells, and mesenteric adipocytes. Although we have not yet established the relative involvement of these cellular sources of CTRP3 in regulating intestinal inflammation, a possible explanation for this finding is that intestinal inflammation causes structural and functional damage to local cellular sources of CTRP3 (or to the most relevant source), resulting in reduced CTRP3 levels, which further exacerbates intestinal inflammation, thus forming a negative feedback loop. Of possible relevance, we found that the difference in the reduction of CTRP3 in patients with active and inactive CD was not statistically significant. This may suggest that the reduced CTRP3 levels are likely caused by IBD-related pathophysiological changes other than inflammatory activities.
Our findings also showed that CTRP3 exerts an anti-inflammatory effect even under basal conditions. In the water-treated groups, although CTRP3 KO/Tg mice had similar clinical phenotypes and intestinal histologic scores as corresponding WT littermates, the anti-inflammatory effect of CTRP3 was discernable. First, CTRP3 levels correlated positively with SIRT1 levels and negatively with phosphorylated NF-κB p65 transcriptional activity and proinflammatory cytokines in CTRP3 KO mice (no differences observed in CTRP3 Tg mice). Second, CTRP3 levels correlated positively with colon length, which is typically shortened in inflammatory settings. These findings suggest that CTRP3 not only attenuates pathologic intestinal inflammation (DSS treatment in our colitis model), but also preserves an appropriate level of physiological intestinal inflammation (water treatment in our colitis model), that is, the innate immune response of intestinal mucosa to nonpathologic (or prepathologic) aggravation.
It is possible that reduced CTRP3 levels in patients with IBD are inadequate to enable the maintenance of homeostatic physiological inflammation. Future studies will investigate whether and how CTRP3 functions within the innate immune system (in IBD and basal settings). More generally, these findings add to the increasing evidence that adipokines are critical to the maintenance of homeostasis in the intestinal epithelium.
In conclusion, our findings indicate that CTRP3 expression levels correlate negatively with intestinal inflammation in an acute colitis mouse model and in patients with IBD. Furthermore, our findings suggest that CTRP3 attenuates intestinal inflammation through SIRT1/NF-κB signaling to suppress the production of proinflammatory cytokines. Future studies are needed to confirm the precise molecular mechanisms through which CTRP3 activates this signaling, the specific sources of CTRP3 that influence intestinal inflammation, and whether the administration of recombinant CTRP3 and/or SIRT1 activators offers translational potential as a new IBD treatment strategy.
Materials and Methods
CTRP3 Knockout and Overexpressing Transgenic Mice
All animal protocols were approved by the Institutional Animal Care and Use Committee of the Johns Hopkins University School of Medicine. CTRP3 KO (-/-) and overexpressing Tg mouse strains were previously generated.
CTRP3 KO mice and WT littermates were generated by intercrossing CTRP3 heterozygous (+/-) mice. Littermates were used throughout the study. Male and female mice were examined. All mice were cohoused to control for microbiome differences. On termination of the study, mice were fasted overnight and euthanized and tissues were collected and snap-frozen in liquid nitrogen. Tissues were kept at −80°C until analysis.
Human intestinal epithelial tissue and mesenteric fat were obtained from patients with CD and control patients undergoing endoscopic or surgical procedures through 2 research protocols approved by the Johns Hopkins Institutional Review Boards. Tissue was deidentified, but age, sex, and ethnicity information were noted. Both male and female patients were included. Terminal ileal epithelial tissue was obtained from the following patients: (1) disease-affected mucosa with discernable ulcers/erosions from patients with CD (active; n = 8), (2) normal-appearing mucosa from the same patients with CD (inactive; n = 8), and (3) normal-appearing mucosa from control patients (control, normal histology; n = 9). Ascending colonic epithelial tissue was obtained from patients with UC with active colitis (n = 5) and control patients (n = 8).
Acute Dextran Sulfate Sodium Colitis Mouse Models
Age- and sex-matched CTRP3 KO and Tg mice, along with their WT littermates, were subjected to different concentrations of DSS (1.2%, 1.5%, 2.0%, and 2.5%) in drinking water for 6–10 consecutive days. Mouse body weight, stool consistency, and rectal bleeding were recorded and scored daily as shown in Table 1.
Surviving analysis was based on the clinical data collected over 10 days of DSS treatment. On termination of the study, mice were dissected for tissues of interest (mesenteric fat, terminal ileum, and colon). Tissue was fixed in 4% paraformaldehyde overnight followed by paraffin embedding. Paraffin sections 5–8 μm thick were stained with hematoxylin and eosin and examined in a blinded manner by 2 researchers (YF and HY). Tissue damage and inflammatory cell infiltrates were scored as shown in Table 2.
Single-Molecule RNA Fluorescence In Situ Hybridization for RNA Detection in Mouse Intestine
Single-molecule RNA fluorescence in situ hybridization uses a probe library consisting of short DNA oligonucleotides labeled with a fluorescent dye that hybridize to a target RNA in fixed cells, allowing RNA quantification and localization at the single-cell level and with single-molecule resolution.
All buffers were prepared following RNase precautions and using RNase-free water. Briefly, intestinal tissue was fixed in 4% paraformaldehyde overnight and washed 3 times with phosphate-buffered saline (PBS). The tissue was then embedded in the Optimal Cutting Temperature compound, cut into 5–10 μm frozen sections, and mounted onto collagen-coated coverslips. The sections were washed with 1X PBS and incubated in 70% alcohol overnight at 4°C. After removing 70% alcohol and submerging them in 10% formamide and 2X SSC, they were incubated in prehybridization buffer for 20 minutes. Then, sections were incubated in 50 μL of hybridization solution containing 80 nM DNA probe libraries for mouse CTRP3 (a mixture of 55 cy3-labeled oligonucleotides), mouse RNA polymerase II subunit A (a mixture of 98 cy5-labeled oligonucleotides, a positive control), and mouse glyceraldehyde 3-phosphate dehydrogenase (a mixture of 33 Atto 590–labeled oligonucleotides, a positive control) in a humidified chamber overnight at 37°C. After that, sections were rinsed with prewarmed hybridization buffer and incubated for 20 minutes at 37°C. Then, sections were incubated again in the hybridization buffer for 20 minutes at room temperature. Finally, sections were washed with 1X PBS 2 times and mounted on slides with ProLong Diamond Antifade Mountant (Thermo Fisher Scientific). Images were taken and analyzed using an automated inverted Nikon Ti-2 wide-field microscope equipped with ×60 1.4NA oil immersion objective lens (Nikon), Spectra X LED light engine (Lumencor), and ORCA-Flash 4.0 V2 sCMOS camera (Hamamatsu).
Immunostaining of Mouse and Human Intestine
Mouse intestinal tissue with attached mesenteric fat or human terminal ileal epithelial tissue was fixed overnight at 4% paraformaldehyde and then embedded in the Optimal Cutting Temperature compound. Tissue staining was performed according to standard immunofluorescence protocols with the following primary antibodies: mouse CTRP3 (generated and provided by GWW, rabbit antimouse), human CTRP3 (Invitrogen, PA5-115061, rabbit antihuman), α-SMA-Cy3 (Sigma, A2547), and Phospho-Ezrin (Cell Signaling, #3141). For detection, the following fluorescent-labeled secondary antibodies were used: Alexa Fluor 647 phalloidin (Molecular Probes, A22287), Alexa 488 donkey antirabbit IgG (Molecular Probes, A11008), and Alexa 594 goat antirabbit IgG (Molecular Probes, A11012). The tissue was counterstained using 4′,6-diamidino-2-phenylindole (Sigma, D8417). Images were taken and analyzed using an Olympus IX83 inverted microscope at the Johns Hopkins Ross Fluorescence Imaging Center.
mRNA gene expression of CTRP3, SIRT1, TNF-α, and IL-6 were measured by RT-qPCR. Total RNA of mouse and human intestinal tissue was isolated using the RNeasy Mini Kit (Qiagen) and total RNA of mouse mesenteric fat was isolated using the RNeasy Lipid Tissue Mini Kit (Qiagen). The isolated RNA was transcribed to complementary DNA via Superscript IV VILO Master Mix (Invitrogen). Quantitative PCR was performed with the PowerTrack SYBR Green Master Mix (Thermo Fisher Scientific). All qPCR primers used are listed in Table 3. Data were analyzed by the comparative cycle threshold (Ct) method as means of relative quantitation of gene expression, normalized to an endogenous reference control gene (mouse beta-2-microglobulin or human beta-actin) and relative to a calibrator (normalized Ct value obtained from WT mice or control patients), and expressed as 2-ΔΔCT.
Table 3Primers Information for Reverse-Transcription Quantitative Polymerase Chain ReactionGene name
Mouse or human intestinal tissue was lysed in 100 μL ice-cold RIPA buffer (Thermo Fisher Scientific) with protease inhibitors (Thermo Fisher Scientific). Homogenates were incubated in a cold room for 2 hours and then centrifuged at 16,000 g for 20 minutes at 4°C. Proteins of the whole-cell extracts of 10–15 μg were separated on 10% sodium dodecyl sulfate polyacrylamide gels for 1 hour at 150 V. Proteins were then transferred to nitrocellulose membranes for 1.5 hours at 100 V. The membranes were blocked with PBS supplemented with 5% dry milk for 1 hour at room temperature, and probed with the following primary antibodies as indicated by each experiment: β-actin (Sigma, A1978), mCTRP3 (provided by GWW), hCTRP3 (Invitrogen, PA5-115061), SIRT1 (Cell Signaling, #2028 and #8469), NF-κB p65 (Santa Cruz, sc-372), p-NF-κB p65 (Cell Signaling, #3033), and Acetyl-NF-κB p65 (Cell Signaling, #3045) at 4°C overnight, and then washed in TBST for 30 minutes. The membranes were probed with corresponding secondary antibodies (IRDye 800CW Goat anti-Rabbit IgG, #LI-COR 926-32211 and IRDye 680RD Goat anti-Mouse IgG, #LI-COR 926-68070) at 1:10000 dilution for 1 hour at room temperature and the bands were visualized by a Li-Cor Odyssey Fc imaging system and Image Studio software. Signals were quantified using ImageJ (National Institutes of Health, Bethesda, MD) and normalized to β-actin.
Mouse Intestinal Organoid Culture
Mouse intestinal organoids were prepared from the distal colon of male CTRP3 KO and Tg mice (8–12 weeks) as previously described.
Briefly, mice were euthanized via isoflurane inhalation followed by cervical dislocation, and a portion of the colon ∼2 cm long was removed and bathed in ice-cold PBS with 200 U/mL pen/strep at 4°C for 45 minutes. Isolated crypts obtained after EGTA chelation for 1 hour were embedded in Matrigel (Corning) in 24-well plates (Corning) and culture media was changed every 2 days. Mouse organoids were cultured in 5% CO2 atmosphere at 37 degrees and passaged via mechanical trituration every 4 days.
Mouse Intestinal Monolayers
Mouse intestinal organoids from passage 5–20 were used to generate monolayers. Briefly, Transwell inserts (polyester membrane with 0.4 m pore, Corning, 3470) were coated with 200 μg/mL rat tail collagen I (Corning, 354236) and incubated at 37°C overnight. Filters were washed with base-conditioned media without growth factors before plating. Mouse organoids were freed from Matrigel and mechanically triturated before being plated in a precoated Transwell insert with 150 μL media in the apical chamber and 600 μL in the basolateral chamber. Monolayer progression and confluence were monitored by transepithelial electrical resistance (EVOM2 Voltohmmeter, World Precision Instruments). Once the monolayers were confluent as determined via transepithelial electrical resistance over 100 Ω·cm2, the monolayers were switched to differentiation media. After incubation for 24 hours, the culture media were replaced with differentiation media containing the following medication: 100 μM Resveratrol (Sigma, 34092) for CTRP3 KO groups and 50 μM Ex-527 (Sigma, E7034) for CTRP3 Tg groups. Both medications were prepared in DMSO (Sigma, D8418). Cells and culture media were collected for future analysis after 24 hours of treatment.
Cytokine Enzyme-Linked Immunosorbent Assay
Mouse DuoSet ELISA kits (R&D Systems) were used to measure levels of TNF-α (DY410) and IL-6 (DY406) in the apical and basolateral culture media collected from mouse intestinal organoids following the manufacturer’s instructions.
All data presented are representative of 3 or more independent experiments and are shown as mean ± SEM unless stated otherwise. The difference between the 2 groups was assessed by Student t test. The difference among 3 or more groups was compared using 1-way or 2-way analysis of variance with Tukey multiple comparisons test. Log-rank (Mantel-Cox) testing was used for survival analysis. Statistical analyses were performed using GraphPad Prism software. A statistically significant difference was considered when ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001.
Molecular, biochemical and functional characterizations of C1q/TNF family members: adipose-tissue-selective expression patterns, regulation by PPAR-gamma agonist, cysteine-mediated oligomerizations, combinatorial associations and metabolic functions.