SEARCH

SEARCH BY CITATION

Keywords:

  • IBD;
  • ulcerative colitis;
  • biomarker;
  • leucine-rich alpha-2 glycoprotein;
  • DSS

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Background:

Reliable biomarkers for monitoring disease activity have not been clinically established in ulcerative colitis (UC). This study aimed to investigate whether levels of serum leucine-rich alpha-2 glycoprotein (LRG), identified recently as a potential disease activity marker in Crohn's disease and rheumatoid arthritis, correlate with disease activity in UC.

Methods:

Serum LRG concentrations were determined by enzyme-linked immunosorbent assay (ELISA) in patients with UC and healthy controls (HC) and were evaluated for correlation with disease activity. Expression of LRG in inflamed colonic tissues from patients with UC was analyzed by western blotting and immunohistochemistry. Interleukin (IL)-6-independent induction of LRG was investigated using IL-6-deficient mice by lipopolysaccharide (LPS)-mediated acute inflammation and dextran sodium sulfate (DSS)-induced colitis.

Results:

Serum LRG concentrations were significantly elevated in active UC patients compared with patients in remission (P < 0.0001) and HC (P < 0.0001) and were correlated with disease activity in UC better than C-reactive protein (CRP). Expression of LRG was increased in inflamed colonic tissues in UC. Tumor necrosis factor alpha (TNF-α), IL-6, and IL-22, serum levels of which were elevated in patients with active UC, could induce LRG expression in COLO205 cells. Serum LRG levels were increased in IL-6-deficient mice with LPS-mediated acute inflammation and DSS-induced colitis.

Conclusions:

Serum LRG concentrations correlate well with disease activity in UC. LRG induction is robust in inflamed colons and is likely to involve an IL-6-independent pathway. Serum LRG is thus a novel serum biomarker for monitoring disease activity in UC and is a promising surrogate for CRP. (Inflamm Bowel Dis 2012;)

The chronic inflammatory bowel diseases (IBDs), Crohn's disease (CD) and ulcerative colitis (UC), are typically characterized by episodes of acute flares and remission.1, 2 Depending on disease location and extent, exacerbation leads to diarrhea, abdominal pain, and systemic symptoms such as fatigue and weight loss.3–5 Disease activity indices have been developed as outcome measures in clinical trials.6, 7 They may help to reproducibly and validly assess the patients' status and to support therapeutic decision-making.6 Variables of disease activity indices comprise frequency of bowel movements, severity of abdominal pain, general well-being, occurrence of extraintestinal manifestations, and laboratory parameters.8

One of the most important protein biomarkers increased during the inflammatory state is C-reactive protein (CRP). However, elevation of serum CRP levels is not observed in certain inflammatory diseases. While serum CRP levels are highly increased in CD and rheumatoid arthritis (RA) patients and widely used for monitoring disease activity, only modest to absent CRP responses are observed in systemic lupus erythematosus (SLE), dermatomyositis, Sjogren's syndrome, or UC, although active inflammation is present.9–11 In UC, endoscopic disease activity may predict future clinical symptoms,12 but direct endoscopic or radiologic visualization of the degree of inflammation is rarely performed in outpatients with inactive or mild disease. Therefore, alternative biomarkers, which can conveniently and precisely monitor disease activity during therapy in inflammatory diseases, are required for the determination of adequate treatment.

By using a quantitative proteomic approach, we have previously reported that serum levels of leucine-rich alpha-2 glycoprotein (LRG) were elevated in patients with active RA and serum LRG levels were correlated with disease activity of not only RA but also CD, suggesting that serum LRG is a serological biomarker for monitoring disease activity.13 LRG is an ≈50 kDa glycoprotein and contains repetitive sequences with a leucine-rich motif, first purified from human serum.14, 15 LRG has been reported to be expressed by the liver cells and neutrophils16, 17; however, its function remains unclear. To date, the relationship between serum LRG levels and disease activity in UC has not been assessed. In this study we investigated serum LRG expression levels in UC patients and evaluated their correlation with clinical disease activity. Serum LRG levels were significantly increased in the active UC patients. LRG expression was upregulated in the inflamed colonic mucosa of UC possibly through stimulation by various cytokines including tumor necrosis factor alpha (TNF-α), interleukin (IL)-6, and IL-22, the expression of which are increased in active UC. Moreover, we show that serum LRG correlates more strongly than CRP with disease activity in UC. Therefore, serum LRG may be a useful disease activity biomarker for UC.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Patients and Sera

Sera were obtained from patients with UC (n = 82), appendicitis (n = 13), and diverticulitis (n = 4) and surgical or biopsy samples were obtained from patients with UC (n = 10) from Osaka University Hospital (Osaka, Japan) and the Department of Surgery, Osaka Rosai Hospital, respectively. Sera from healthy controls (HCs) (n = 50), age/sex-matched with UC patients, were used. Diagnosis of UC was based on conventional clinical, radiological, endoscopic, and histopathological criteria. Clinical activities were determined using the Clinical Activity Index (CAI) for UC.18 Clinical remission was defined as CAI <6.19 In addition to CAI, the endoscopic findings were also graded according to Matts' criteria.20 Endoscopic remission was defined as Matts' score ≤2. Detailed patient characteristics are presented in Table 1. For Caucasian patients with UC, sera (n = 30) were obtained from the Department of Medicine, University of North Carolina Hospital (Chapel Hill, NC). Sera from HCs (n = 19), age/sex-matched with UC patients, were used. Detailed patient characteristics are presented in Table 2, while data of disease activity of UC is not available.

Table 1. Characteristics of Patients with Ulcerative Colitis (UC)
CharacteristicsPatients with UCPatients with Appendicitis and Diverticulitis
Number (male:female)82 (41:41)17 (8:9)
Age, yr, mean (SD)40.1 (15.7)33.1 (13.7)
Age at diagnosis, yr, mean (SD)34.7 (15.6)33.1 (13.7)
Bowel surgery (including appendectomy), N (%)7 (8.54) 
Treatment  
 Salazosulfapyridine or mesalazine, N (%)66 (80.5) 
 Steroids, N (%)16 (19.5) 
 Immunomodulators, N (%)3 (3.7) 
Disease location (N)  
 Extensive colitis/left-sided colitis/proctitis37/30/15 
CRP, mg/dL, mean (SD)0.884 (1.967)8.47 (7.69)
WBC cells/μl, mean (SD)6716 (2317)12307 (3603)
CAI, mean (SD)4.71 (4.89) 
Matts's score, mean (SD)2.27 (0.89) 
Table 2. Characteristics of Patients with UC in a Caucasian Cohort
CharacteristicsPatients with UC
Number (male:female)30 (18:12)
Age, yr, mean (SD)42.9 (17.9)
Age at diagnosis, yr, mean (SD)33.2 (15.7)
Treatment 
 Salazosulfapyridine or mesalazine, N (%)14 (46.7)
 Steroids, N (%)9 (30.0)
 Immunomodulators, N (%)11 (36.7)
 Anti-TNF therapy3 (10.0)
Disease location (N) 
 Extensive colitis/left-sided colitis/proctitis16/11/3

Quantification of Serum LRG and Cytokines

Human serum LRG and mouse serum LRG were quantitated by human LRG assay kit (IBL, Fujioka, Japan) and mouse LRG assay kit (IBL, Fujioka, Japan). These enzyme-linked immunosorbent assay (ELISA) assays were performed in duplicate. The intraassay coefficients of variations for human LRG and mouse LRG were ≤7.98% and ≤8.93%, respectively. For the quantification of IL-6, TNF-α, and IL-22 in human serum samples, the human IL-6 Ultra Sensitive ELISA (Biosource International, Camarillo, CA), human TNF-α UltraSensitive ELISA kit (Invitrogen, Carlsbad, CA), and human IL-22 Quantikine ELISA Kit (R&D Systems, Minneapolis, MN) were used following the manufacturer's guidelines.

Western Blot Analysis

Frozen colon tissue samples were lysed in RIPA buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.1% sodium deoxycholate, 0.1% SDS, 1 × protease inhibitor cocktail; Nacalai Tesque, Kyoto, Japan) and 1 × phosphatase inhibitor cocktail (Nacalai Tesque) followed by centrifugation (13,200 rpm, 4°C, 15 minutes), after which the supernatants were stored at −80°C until use. Extracted proteins were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) as described previously.21 Samples transferred onto PVDF membranes were treated with a rabbit antihuman LRG polyclonal antibody (Proteintech Group, Chicago, IL) or a rabbit anti-GAPDH polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was used as described previously.21

Immunohistochemistry

Immunohistochemical analyses were performed according to a method described in our previous report.22 Briefly, rabbit antihuman LRG polyclonal antibodies were used as the primary antibody. After incubation with the primary antibodies, the sections were treated with biotin-conjugated goat antirabbit IgG (Vector Laboratories, Burlingame, CA) and avidin-biotin-peroxidase complexes (Vector Laboratories). Immunoreactive cells were visualized with a diaminobenzidine substrate (Merck, Darmstadt, Germany) and were counterstained with hematoxylin.

Mice

C57BL/6 mice were purchased from Clea Japan (Tokyo, Japan). C57BL/6-background IL-6-deficient mice were kindly provided by Professor Yoichiro Iwakura (Laboratory of Molecular Pathogenesis, Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan). Mice were maintained under specific pathogen-free conditions. C57BL/6 and IL-6-deficient mice were used at 7–9 weeks of age. All experiments were conducted according to the institutional ethical guidelines for animal experimentation.

LPS-mediated Acute Inflammation

To induce acute inflammation, wildtype (WT) mice and IL-6-deficient mice were injected intraperitoneally with 0 or 10 mg/kg LPS (Escherichia coli LPS, Sigma, St. Louis, MO) dissolved in 500 μL phosphate-buffered saline (PBS). Blood was collected at before and 24 hours after LPS injection and the serum was separated by centrifugation and stored at −30°C until used for ELISA analysis.

Induction of Colitis

For induction of colitis, WT mice and IL-6-deficient mice were given 3% dextran sodium sulfate (DSS) (m/w 36,000–50,000; MP Biomedicals, Solon, OH) dissolved in drinking water provided ad libitum for 5 days, followed by provision of ordinary water for 20 days.

Assessment of Severity of DSS-induced Colitis

WT mice were weighed daily from day 0 to day 25. Changes in body weight were calculated as follows: body weight change (%) = [(weight on a given day (days 0–13) − weight on day 0)/weight on day 0] * 100. Blood was collected from WT mice on days 5, 7, 10, 15, and 25 after DSS administration or day 0 by cardiac puncture under anesthesia and on days 0 and 10 from IL-6-deficient mice. The serum was separated by centrifugation and stored at −30°C until used for ELISA analysis.

Cell Culture

The human colonic adenocarcinoma COLO205 cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA). Cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) (HyClone Laboratories, Logan, UT) and 1% penicillin–streptomycin (Nacalai Tesque) at 37°C under a humidified atmosphere of 5% CO2.

For the analysis of LRG protein induction, COLO205 cells were stimulated with various concentrations of cytokines for 24 hours and culture supernatant were concentrated using Amicon Ultra-4 10K MWCO (Millipore, Bedford, MA). Concentrated supernatants were used for western blot analysis. Full-length human LRG cDNA was inserted into pcDNA3.1/V5-His-TOPO vector (Invitrogen) and designated pcDNA3.1-LRG-V5-His. pcDNA3.1-LRG-V5-His vector was transfected into COS7 cells using Lipofectamine 2000 reagent (Invitrogen) and culture medium were used for the positive control.

Quantitative Real-time Reverse-transcription Polymerase Chain Reaction (RT-PCR) Analysis

For the quantification of mRNA levels of LRG, various mouse organs were analyzed by real-time RT-PCR as described previously.23 Levels of mouse LRG and mouse hypoxanthine phosphoribosyltransferase (HPRT) levels were determined by the 7900HT Real-time PCR system (Applied Biosystems, Foster City, CA) using specific primers: murine LRG forward 5′-ATCAAGGAAGCCTCCAGGAT-3′; reverse 5′-CAGCTGCGTCAGGTTGG-3′ and murine hypoxanitine phosphoribosyltransferase (HPRT) forward 5′-TCAGTCAACGGGGGACATAAA-3′; reverse 5′-GGGGCTGTACTGCTTAACCAG-3′.

Statistics

The Mann–Whitney U-test or one-way analysis of variance (ANOVA) followed by a Scheffe's test were used for statistical analyses. Two-tailed Student's t-test was used for significant differences in LRG expression between identical patients with UC in active and remission disease stage. One-way ANOVA followed by a Dunnett's test was used for multiple comparison of the difference of serum LRG levels at various timepoints after DSS treatment in mice. Pearson's test was used to analyze the relationship between LRG and CRP, IL-6, or CAI. For drawing of receiver operating characteristic (ROC) curves and estimation of the area under the ROC curve (AUC) statistics, the software Excel Statistics 2010 (Social Survey Research Information, Tokyo, Japan) was used to quantify the ability to differentiate between remission and active by CAI. P < 0.05 was considered significant.

Ethical Considerations

Informed consent was obtained from all donors and all studies involving human subjects were approved by the Institutional Review Boards of the National Institute of Biomedical Innovation, Osaka University Hospital, the Department of Surgery, Osaka Rosai Hospital, and the University of North Carolina.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Serum LRG Levels Are Increased in Active UC Patients

We quantified serum LRG concentrations by ELISA using sera from patients with UC. Serum LRG concentrations were significantly elevated in the active UC patients (CAI ≥6) (14.24 ± 8.08 μg/mL) compared with HC (3.07 ± 1.42 μg/mL; P < 0.0001) (Fig. 1A). There was also a significant difference between LRG serum levels in patients with active UC (CAI ≥6) (14.24 ± 8.08 μg/mL) compared with UC in remission (CAI <6) (5.34 ± 2.60 μg/mL; P < 0.0001) (Fig. 1A). To determine whether serum LRG levels are increased in non-IBD disease controls, we quantified serum LRG levels in patients with appendicitis and diverticulitis. Elevated serum LRG levels were also observed in appendicitis and diverticulitis (16.83 ± 6.50 μg/mL) compared with HC (3.07 ± 1.42 μg/mL; P < 0.0001) (Fig. 1A), suggesting that serum LRG levels are also increased in acute intestinal inflammation.

thumbnail image

Figure 1. Serum LRG levels are increased in patients with active UC. (A) Serum levels of LRG were determined in 82 patients with UC (57 patients in remission [CAI <6], 25 patients in active [CAI ≥6] stage), appendicitis (n = 13), diverticulitis (n = 4) and 50 healthy controls (HC). ***P < 0.0001 by one-way ANOVA followed by Scheffe's post-hoc test. (B) Disease extension in UC was grouped into three categories: In UC patients in remission, extensive colitis (n = 19), left-sided colitis (n = 24), and proctitis (n = 14); in active patients, extensive colitis (n = 18), left-sided colitis (n = 6), and proctitis (n = 1) and HC (n = 50). **P < 0.005, ***P < 0.0001 by one-way ANOVA followed by Scheffe's post-hoc test. (C) Serum levels of LRG were determined in patients with UC (n = 30) and HC (n = 19) in a Caucasian cohort. **P < 0.005 by Mann–Whitney U-test. (D) In a Caucasian cohort, disease extension in UC was grouped into three categories: extensive colitis (n = 16), left-sided colitis (n = 11), and proctitis (n = 3) and HC (n = 19). *P < 0.05 by one-way ANOVA followed by Scheffe's post-hoc test.

Download figure to PowerPoint

When UC were classified into three categories based on disease extent, significantly higher serum LRG concentrations were observed in active patients with extensive colitis (14.34 ± 7.89 μg/mL) compared with in remission (4.96 ± 2.68 μg/mL; P < 0.0001) and HC (3.07 ±1.42 μg/mL; P < 0.0001) and active patients with left-sided colitis (15.41 ± 9.16 μg/mL) compared with in remission (5.91 ± 2.41 μg/mL; P = 0.0003) and HC (3.07 ±1.42 μg/mL; P = 0.001) (Fig. 1B). Nonetheless, there was no clear difference between active patients with proctitis and HC, possibly due to the low number of patients in this group. In patients with UC in remission, serum LRG levels in all of three disease extent categories were comparable with HC (Fig. 1B). Significantly elevated serum LRG levels were also detected in a Caucasian UC cohort (9.46 ± 8.44 μg/mL) compared with HC (4.42 ± 1.91 μg/mL; P < 0.005) (Fig. 1C). In this Caucasian UC cohort, serum LRG levels were also significantly elevated in patients with extensive colitis (9.54 ± 8.05 μg/mL) compared with HC (4.42 ± 1.91 μg/mL; P < 0.05) and left-sided colitis (10.90 ± 9.16 μg/mL) compared with HC (4.42 ± 1.91 μg/mL; P < 0.02) (Fig. 1D). However, a clear difference was not observed between patients with proctitis and HC (Fig. 1D). These results suggest that serum LRG levels were elevated in active UC.

Serum LRG Levels Are Correlated with Disease Activity in UC Patients

We investigated the correlation between serum LRG levels and disease activity (CAI) in UC patients. A positive correlation was observed between LRG and CAI (r = 0.731, P < 0.00001) (Fig. 2A). This correlation was stronger than that observed between CRP and CAI (r = 0.654, P < 0.00001) (Fig. 2A). When patients with UC were classified into active and remission according to the endoscopic findings, significantly elevated serum LRG levels and CRP levels were observed in patients with active UC compared with patients in remission (P < 0.005, respectively) (Supporting Fig. 1A). While serum LRG levels were significantly correlated with CRP levels in patients with UC (r = 0.850, P < 0.00001, n = 82) (Supporting Fig. 2A), such a correlation was not found when a CRP-negative subgroup (CRP <0.2, n = 51) was analyzed (r = 0.101, P = 0.481) (Supporting Fig. 2B). In this CRP-negative group, serum LRG levels were significantly correlated with CAI (r = 0.416, P = 0.00241) (Supporting Fig. 2C); however, significant correlation was not found between CRP and CAI (r = −0.0896, P = 0.532) (Supporting Fig. 2D). Additionally, in the CRP-negative group elevated serum LRG levels were detected in patients with endoscopically active UC compared with patients with UC in remission (P = 0.0442) (Supporting Fig. 1B). These findings in patients with low CRP may explain a better correlation of CAI with LRG than that with CRP.

thumbnail image

Figure 2. Serum LRG levels are correlated with disease activity better than CRP in patients with UC. (A) Serum levels of LRG correlated with CAI (n = 82; P < 0.000001; r = 0.731) better than CRP (n = 82; P < 0.000001; r = 0.654) in patients with UC. (B) Serum levels of LRG correlated with disease activity in extensive colitis (n = 37; P < 0.000001; r = 0.690) and left-sided colitis (n = 30; P < 0.000001; r = 0.840) better than CRP in extensive colitis (n = 37; P = 0.000168; r = 0.580) and left-sided colitis (n = 30; P < 0.000001; r = 0.759), while neither LRG (n = 15; P = 0.649; r = −0.128) nor CRP levels (n = 15; P = 0.360; r = −0.255) were correlated with disease activity in proctitis. (C) Compared with 10 identical active patients with UC, serum levels of LRG were decreased in remission. *P < 0.002 by Student's t-test. (D) ROC curves for LRG and CRP for differentiation between UC patients with remission (n = 57) and active (n = 25) by CAI.

Download figure to PowerPoint

When UC was classified by disease extent, a significantly higher positive correlation was detected between LRG and CAI than CRP and CAI both in extensive colitis (r = 0.690, P < 0.000001 and r = 0.580, P = 0.000168) and left-sided colitis (r = 0.840, P < 0.000001 and r = 0.759, P < 0.000001), but not in proctitis (Fig. 2B). Importantly, by analyzing sera obtained at active (CAI ≥6) and remission (CAI <6) disease stages from 10 identical UC patients, a significant decrease in serum LRG levels in remission was detected (Fig. 2C).

By generating an ROC curve, the sensitivity and specificity of serum LRG for remission and active by CAI were determined (Fig. 2D). The AUC for serum LRG levels was 0.901, whereas the AUC for CRP levels was 0.845. The cutoff value of serum LRG levels was 7.21 μg/mL (sensitivity = 84.0%, specificity = 82.5%). In contrast, when the cutoff value of CRP levels was set to 0.20, a maximum CRP value of normal range, the sensitivity was 80.0% and the specificity was 80.7%. These results emphasize the usefulness of monitoring serum LRG levels for the evaluation of the disease activity of UC.

Expression of LRG Was Increased in Inflamed UC Colons

Next, to investigate whether local inflammatory sites in patients with UC are a potential source of increased serum LRG we first looked at the expression of LRG in the colon by western blot analysis on inflamed and noninflamed sites of surgically resected full-thickness colon specimens from patients with UC. Western blot analysis showed that LRG expression in colon tissues was increased in inflamed sites of active UC patients compared with noninflamed colon tissues (Fig. 3A). Next, we tried to examine the localization of LRG. By immunohistochemistry, increased expression of LRG was detected in the cytoplasm of intestinal epithelial cells (IECs) in inflamed tissues (Fig. 3B–E). In contrast, expression of LRG was lower in noninflamed tissues (Fig. 3B–E). These data suggest that inflamed colon tissue is a potential source of increased serum LRG in patients with UC.

thumbnail image

Figure 3. Expression of LRG is increased in lesion sites of ulcerative colitis. (A) Representative western blot analysis of three separate experiments for LRG using paired surgically resected full-thickness colon specimens from noninflamed (N) and inflamed (I) sites in patients with UC. GAPDH was used as a control of the relative amounts of proteins in each sample. Full-thickness colon tissues from UC in inflamed and noninflamed sites were evaluated by immunohistochemical analysis for LRG expression (n = 10 per experimental group). (B) Noninflamed mucosa (×42). (C) Inflamed mucosa from active UC (×42). (D) Noninflamed mucosa (×400). (E) Inflamed mucosa from active UC (×400). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

LRG Is Induced by Stimulation with TNF-α, IL-6, or IL-22

It has been reported that IL-6 is an inducer of LRG expression.16 However, it is not clear whether LRG is induced by cytokines other than IL-6. At first we investigated the serum levels of IL-6, IL-22, and TNF-α, known to be increased at the inflamed tissue in active UC.24–26 Indeed, ELISA analysis using sera from 82 UC patients revealed that serum TNF-α, IL-6, and IL-22 levels were significantly elevated in active UC patients compared with those patients in remission (P = 0.0178, P = 0.00690, and P < 0.0001, respectively) (Fig. 4A). Next, to investigate which proinflammatory cytokines induce expression of LRG we stimulated human colonic adenocarcinoma COLO205 cells with TNF-α, IL-6, or IL-22 for 24 hours. After cytokine stimulation, secretion of LRG protein into the culture media was analyzed by western blotting. Interestingly, LRG was induced not only by stimulation with IL-6, but also by TNF-α and IL-22 in a dose-dependent manner (Fig. 4B). These results indicate that expression of LRG is induced by various proinflammatory cytokines including IL-6.

thumbnail image

Figure 4. Expression of LRG was induced by TNF-α, IL-6, and IL-22. (A) Serum levels of TNF-α, IL-6, and IL-22 were determined in patients with UC (57 patients in remission [CAI <6] and 25 patients in active [CAI ≥6] stage). Data are expressed as mean ± SEM. *P < 0.05, **P < 0.005, ***P < 0.0001 by Mann–Whitney U-test. (B) LRG was determined in supernatants of COLO205 cells left untreated or stimulated with TNF-α, IL-6, and IL-22 at 1.0, 10, 100 ng/mL for 24 hours and analyzed by western blotting. There was a dose-dependent increase in LRG levels after treatment with TNF-α, IL-6, and IL-22.

Download figure to PowerPoint

Expression of LRG Through an IL-6-independent Pathway Is Demonstrated in LPS-mediated Acute Inflammation and DSS-induced Colitis

CRP is one of the representative acute phase proteins in humans and CRP production is primarily dependent on liver by circulating IL-6. To examine the possible differences in induction mechanisms between LRG and CRP, particularly with regard to the involvement of IL-6, we took advantage of murine models. We first assessed whether LRG is induced in WT mice by injecting LPS, an inducer of proinflammatory cytokines from macrophages, because CRP is poorly induced in mice during acute inflammation. At 24 hours after intraperitoneal injection of LPS, serum samples were prepared and serum LRG levels were determined by ELISA. Compared with WT mice, significant elevation of serum LRG levels were detected in LPS-administered WT mice (Fig. 5A), suggesting that LRG is induced during acute inflammation in mice as in humans.

thumbnail image

Figure 5. Induction of LRG has IL-6-independent pathway in LPS-mediated acute inflammation and active stage of DSS-induced colitis. (A) WT mice and IL-6-deficient mice were injected intraperitoneally with 0 or 10 mg/kg LPS dissolved in 500 μL PBS and serum LRG levels were measured after 24 hours. Data are expressed as mean ± SEM. **P < 0.005, ***P < 0.0001 by one-way ANOVA followed by Scheffe's post-hoc test. (B) Relative body weight changes of mice with DSS-induced colitis in this study. Data are expressed as mean ± SEM (n = 4). (C) Expression of LRG is upregulated in murine DSS-induced colitis. At the indicated time, serum LRG levels were determined by ELISA analysis. **P < 0.005, ***P < 0.0001 by one-way ANOVA followed by a by Dunnett's post-hoc test. (D) Nine days after control or DSS treatment, mice were euthanized and gene expression of LRG in the colon, liver, spleen, and kidney was determined by quantitative PCR analysis. Gene expression was calculated relative to HPRT. Data were expressed as mean ± SD (n = 5). *P < 0.05, **P < 0.005 by Student's t-test. (E) IL-6-deficient mice were used for DSS-induced colitis. Nine days after DSS administration, serum levels of mouse LRG was determined by ELISA analysis. *P < 0.05 by one-way ANOVA followed by Scheffe's post-hoc test.

Download figure to PowerPoint

We next used a murine IBD model to investigate induction mechanisms of LRG during colonic inflammation. DSS-induced colitis is often used as a murine model of UC.27 We induced colitis in WT mice by treating them with 3% DSS for 5 days and measured changes in relative body weight. Body weight began to decrease at day 5, showed greatest reduction at day 9, and recovered at 18 days after DSS treatment (Fig. 5B). We analyzed changes in serum LRG levels by ELISA before and 5, 7, 10, 15, and 25 days after DSS treatment. Consistent with body weight loss, serum LRG levels were significantly elevated at 5 days after DSS treatment (Fig. 5C). Serum LRG levels remained high until day 15, but decreased at day 25. Delayed normalization of serum LRG levels is likely due to the prolonged inflammation at inflamed tissue sites. Additionally, a long half-life of serum LRG might also be involved in this, since our preliminary data suggest that the half-life of serum human LRG levels are about two times longer than that of CRP (data not shown). To investigate which organs produce LRG in DSS-induced colitis, RNA was extracted from colon, liver, and spleen before and 9 days after DSS treatment. By quantitative PCR analysis (Fig. 5D), expression of LRG was significantly increased in liver (P = 0.00106) and spleen (P = 0.0376); however, the strongest induction was observed in colon (P = 0.000126).

To investigate whether LRG induction is dependent on IL-6 or not, we analyzed serum LRG levels in IL-6-deficient mice. Interestingly, basal LRG levels in IL-6-deficient mice were similar to those in WT mice and LRG was robustly induced by LPS administration in IL-6-deficient mice (Fig. 5A). Moreover, increased serum LRG levels were also detected in the active stage (day 9) of DSS-induced colitis in IL-6-deficient mice (Fig. 5E). Importantly, the increase of serum LRG in IL-6-deficient mice was similar to that in WT mice (Fig. 5A,E). These findings indicate that LRG expression can be induced in the absence of IL-6.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

In this study we first demonstrated that serum LRG levels were significantly increased in sera of active UC patients compared with patients in remission and HC. Serum LRG is likely elevated in diverse racial groups, because we detected increased serum LRG levels not only in Japanese patients (Fig. 1A)13 but also in Caucasian patients with UC (Fig. 1C,D) and CD (data not shown). In addition, levels of serum LRG were significantly correlated with disease activity in UC and the correlation was stronger than CRP. Moreover, by analyzing ROC curve and AUC, serum LRG levels showed higher AUC than CRP and serum LRG levels represented superior sensitivity and specificity to CRP for remission and active of UC by CAI (Fig. 2D), indicating that LRG is a useful marker to evaluate disease activity in UC. In the normal state, serum LRG is thought to be produced from liver and LRG is abundantly found in the sera of HC. In colonic inflammation, we found that the expression of LRG is increased in the inflamed mucosa of UC patients and mice with DSS colitis, suggesting that inflamed tissues can be a source for production of LRG (Fig. 3). The increased expression of LRG in inflamed tissue has previously been observed in appendix during acute appendicitis.28 Moreover, in acute inflammatory disorders, including appendicitis and diverticulitis, increased expression of serum LRG was observed (Fig. 1A). These results indicate that the elevated expression of LRG at inflamed sites and in sera occurs in various acute and chronic inflammatory disorders. Therefore, increased serum LRG levels are not suitable for use as a specific diagnostic marker of IBD.

CRP is the most common serum marker used to evaluate disease activity in inflammatory diseases. However, serum CRP is primarily dependent on liver production induced by circulating IL-6. Compared with CD and RA, only modest to absent CRP responses are observed in UC, despite active inflammation in colon.9 Indeed, our cohort of 82 UC patients, analyzed in this study, included five patients with normal value of CRP while having active disease (Fig. 2A). However, our study demonstrated that serum LRG levels were significantly increased in active UC patients' sera and correlated better with disease activity of UC than CRP levels (Figs. 1A, 2A). Particularly, in the group of patients with negative CRP (CRP <0.2), significant correlation was observed between serum LRG levels and CAI (Supporting Fig. 2C). Similarly, among CRP-negative patients serum LRG levels were significantly elevated in those with endoscopically active UC, compared with UC in remission (Supporting Fig. 1B). In addition, serum LRG levels were decreased after therapy (Fig. 2C), suggesting that LRG is a useful serological biomarker for evaluating disease activity and therapeutic effect in UC.

Better correlation of serum LRG levels with disease activity of UC than CRP might be explained in part by the differences in induction mechanisms between LRG and CRP. While the expression of CRP is essentially dependent on IL-6, several cytokines may compensate for the absence of elevated IL-6 in induction of LRG expression. Accordingly, expression of LRG in COLO205 cells was induced not only by IL-6 but also by TNF-α and IL-22 (Fig. 4B), all of which were increased in sera of UC patients (Fig. 4A). Expression of LRG was strongly induced by IL-22 in COLO205 cells, correlating with enhanced STAT3 (Tyr705) phosphorylation by IL-22 compared with IL-6 (data not shown). Thus, inflammatory cytokines such as TNF-α and IL-22 may mediate LRG expression in the absence of IL-6. Moreover, using DSS-induced colitis in IL-6-deficient mice we could demonstrate an IL-6-independent pathway for LRG induction (Fig. 5E). Because promoter regions of human and mouse LRG share high sequence homology and contain putative binding sites for transcription factors such as C/EBP, MZF1, and STAT,17 it is conceivable that the similar IL-6-independent mechanisms of LRG induction are also involved in humans. Future studies are required to fully elucidate the induction mechanisms of LRG in both humans and mice.

In the three disease categories of UC based on extent of disease, serum LRG levels tended to be low in proctitis compared with extensive colitis and left-sided colitis (Fig. 1B). In addition, correlation between serum LRG levels and disease activity did not reach significance in proctitis (Fig. 2B). Although the low number of patients with active proctitis may preclude the proper evaluation of LRG levels, limited inflamed area of proctitis may also be a reason for slight increases of serum LRG levels in these patients. Given the increased production of LRG in inflamed colonic mucosa, fecal LRG might be a more sensitive disease biomarker for UC including proctitis. Optimization for the measurement of fecal LRG is currently under way in our laboratory.

This study also highlights the potential usefulness of LRG in evaluating murine colitis. Our results indicate that serum LRG levels increase as the disease progresses in a DSS-induced colitis model (Fig. 5B,C). In addition, the LRG expression is significantly upregulated in the colon with DSS-induced colitis (Fig. 5D). Thus, LRG in mice can be an objective disease activity marker for colitis models and may be useful for preclinical studies of IBD.

In conclusion, serum LRG levels reflect disease activity of UC better than CRP, especially in patients with low CRP. In the inflammatory condition, LRG is expressed in the inflamed tissue and expression of LRG is regulated by mechanisms different from that of CRP. These findings suggest that serum LRG is a novel and potential serologic biomarker for evaluating disease activity of UC.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

We thank T. Mizushima for provision of appendicitis and diverticulitis patients' sera, Y. Kanazawa for secretarial assistance, and M. Urase and A. Morimoto for technical assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IBD_22936_sm_SuppInfo.doc20KSupporting Information
IBD_22936_sm_SuppFig1.tif98KSupporting Information Figure 1.
IBD_22936_sm_SuppFig2.tif593KSupporting Information Figure 2.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.