PROTEIN FINGERPRINT BIOMARKERS OF COLLAGEN REMODELING CAN EVALUATE FIBROGENESIS IN AN IN VITRO COLONIC SCAR-IN-A-JAR MODEL

UEG Poster

Standard Poster

Log in to access this content.

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Log In Create a free account

Not sure what you can access? Learn more about account types.

Introduction

Intestinal fibrosis affects most inflammatory bowel disease (IBD) patients, resulting in severe clinical complications and reduced treatment response. With no treatments approved for intestinal fibrosis, there is a need for preclinical models to investigate the pathobiology and novel treatments. Myofibroblasts are the main drivers of intestinal fibrosis and the excessive accumulation of extracellular matrix in IBD patients. We investigated a prolonged scar-in-a-jar model of colonic fibrogenesis to validate protein fingerprint (PF) biomarkers of collagen formation for clinical use and the in vitro model as a screening tool.

Aims & Methods

Human colonic fibroblasts (CCD-18co, ATCC) were grown in EMEM supplemented with 10% FBS and 1% P/S in culture flasks with 200 µg/cm³gelatin to reduce cell activation. At 80% confluency, the cells were seeded at 30,000 cells/well in a 48-well plate noted as day -2. The cells were either stimulated from day 0 and every fourth day, except day 15, with either 50 pM TGF-β or 0.1 % DMSO. Starting at day 4 certain wells were co-stimulated with TGF-β1 and an ALK5 inhibitor (ALK5i) (0.01, 0.1, 1.0, or 10.0 µM). The experiment was terminated on day 15 upon which wells were decellularized and the collagen matrix stained with Sirius Red for visual confirmation of collagen formation. Imaging was done using light microscopy at X20 magnification using similar settings for all images. Fibrogenesis was quantitatively determined by measuring the biomarkers PRO-C1, PRO-C3, and PRO-C6 in the supernatants. The biomarkers reflect type I, type III, and type VI collagen formation, respectively. We applied Šídák's multiple comparisons test for differences in the collagen formation biomarker levels calculated as ng/mL or percentage change from day 4 following ALK5i stimulation.

Results

Stimulation with TGF-β signifcantly increased type I, III, and VI collagen formation at days 4, 8, 12, and 15 compared to DMSO stimulated cells quantified with the PRO-C1, PRO-C3, and PRO-C6 biomarkers as ng/mL (p value <0.0001) (Table 1). On day 4 certain cells were co-stimulated with TGF-β1 and an ALK5i, resulting in a significant reduction of type I, III, and VI collagen formation at days 8, 12, and 15 (p value <0.0001) (Table 1). The reduction of the collagen formation biomarkers were largest when stimulating with 1.0 or 10.0 µM of ALK5i, reducing PRO-C1 up to 125.7% on day 12, PRO-C3 up to 560.1% on day 12, and PRO-C3 up to 68.2% on day 8. Collagen deposition was confirmed by Sirius Red (data not shown). To determine stimulation effects on the metabolic activity, Alamar blue was measured on days 0 and 15 demonstrating no significant difference between TGF-β1 and the highest concentration of ALK5i (10.0 µM) (Table 1).

Stimulation comparison	PRO-C1, Type I collagen formation (Day)	PRO-C3, Type III collagen formation (Day)	PRO-C6, Type VI collagen formation (Day)	p-value for comparison(s) (Days 0, 4, 8, 12, and 15)
TGF-β1 vs. DMSO	191.7 vs. 75.9 (Day 4); 282.5 vs. 94.9 (Day 8); 273.1 vs. 100.4 (Day 12); 209.2 vs. 99.3 (ng/mL)	1.5 vs. 1.4 (Day 4); 3.8 vs. 1.4 (Day 8); 9.0 vs. 2.0 (Day 12); 10.5 vs. 1.7 (ng/mL)	0.8 vs. 0.4 (Day 4); 0.7 vs. 0.4 (Day 8); 0.5 vs. 04 (Day 12); 0.4 vs 0.4 (ng/mL)	PRO-C1 (all: <0.0001); PRO-C3 (ns, <0.01, <0.0001, <0.0001); PRO-C6 (<0.01, <0.01, ns, ns)
TGF-β1 vs. 0.01 µM ALK5 inhibitor	50.2% vs. 21.5% (Day 8); 43.3% vs. 14.4% (Day 12); 10.0% vs. 4.6% (Day 15)	155.1% vs. 128.8% (Day 8); 511.7% vs. 419.8% (Day 12); 606.6% vs. 400.1% (Day 15)	14.3% vs. -30.1% (Day 8); -29.4% vs. -51.4% (Day 12); -35.2% vs. -52.3% (Day 15)	PRO-C1 (all: ns); PRO-C3 (ns, ns, ns, <0.01); PRO-C6 (all: ns)
TGF-β1 vs. 0.1 µM ALK5 inhibitor	50.2% vs. 19.0% (Day 8); 43.3% vs. 1.7% (Day 12); 10.0% vs. 2.3% (Day 15)	155.1% vs. 66.5% (Day 8); 511.7% vs. 473.2% (Day 12); 606.6% vs. 437.4% (Day 15)	14.3% vs. -9.6% (Day 8); -29.4% vs. -42.4% (Day 12); -35.2% vs. -40.3% (Day 15)	PRO-C1 (ns, ns, <0.05, ns); PRO-C3 (ns, ns, ns, <0.05); PRO-C6 (
TGF-β1 vs. 1.0 µM ALK5 inhibitor	50.2% vs. -31.9% (Day 8); 43.3% vs. -63.4% (Day 12); 10.0% vs. -62.3% (Day 15)	155.1% vs. 22.5% (Day 8); 511.7% vs. 84.5% (Day 12); 606.6% vs. 51.1% (Day 15)	14.3% vs. -53.9% (Day 8); -29.4% vs. -54.0% (Day 12); -35.2% vs. -54.0% (Day 15)	PRO-C1 (ns, <0.0001, <0.0001, <0.0001); PRO-C3 (ns, ns, <0.0001, <0.0001); PRO-C6 (
TGF-β1 vs. 10.0 µM ALK5 inhibitor	50.2% vs. -56.1% (Day 8); 43.3% vs. -82.4% (Day 12); 10.0% vs. -82.1% (Day 15)	155.1% vs. 6.1% (Day 8); 511.7% vs. 100.7% (Day 12); 606.6% vs. 46.5% (Day 15)	14.3% vs. -49.1% (Day 8); -29.4% vs. -54.3% (Day 12); -35.2% vs. -54.3% (Day 15)	PRO-C1 (ns, <0.0001, <0.0001, <0.0001); PRO-C3 (ns, <0.05, <0.0001, <0.0001); PRO-C6 (ns, <0.05, ns, ns)
	Metabolic Activity at Day 0	Metabolic Activity at Day 15
TGF-β1 vs. 10 µM ALK5i	11928 vs. 11352	12058 vs. 10368		Day 0 (ns) and Day 15 (ns)

Conclusion

PF biomarkers can be applied in vitro to quantify fibrogenesis objectively, demonstrating a significant increase following TGF-β stimulation of colonic fibroblasts. The model may be combined with the biomarkers to evaluate drug targets of intestinal fibrosis.

Disclosure

M. Pehrsson, M.A. Karsdal, and J.H Mortensen are employed at Nordic Bioscience A/S

PROTEIN FINGERPRINT BIOMARKERS OF COLLAGEN REMODELING CAN EVALUATE FIBROGENESIS IN AN IN VITRO COLONIC SCAR-IN-A-JAR MODEL

Martin Pehrsson 1, Marta Sorokina Alexdóttir 1, Morten Asser Karsdal 1, Joachim Høg Mortensen 1

1 Nordic Bioscience A/S, Herlev, Denmark

Conference

UEG Week Copenhagen 2023

Topics

IBD

Submission format

Abstract

Session

PP 05 IBD (Posters)

Citation

United European Gastroenterology Journal 2023; 11 (Supplement 8)

Published

2023

UEG Poster

Standard Poster

Log in to access this content.

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Log In Create a free account

Not sure what you can access? Learn more about account types.

Introduction

Chronic inflammation in inflammatory bowel disease can lead to the development of intestinal fibrosis, exacerbating the disease. The activation of immune cells results in the release of pro-inflammatory cytokines suspected to drive the activation of fibroblasts and excessive formation of extracellular matrix proteins. While anti-inflammatory therapies exist, their direct effects on fibrogenesis are limited, and model systems for evaluating inflammation-dependent fibrosis are warranted.

Aims & Methods

We aimed to evaluate the profibrotic effects of activated macrophages in a gastrointestinal scar-in-a-jar model by quantifying type I, III, and type VI collagen and fibronectin formation biomarkers.
Macrophages (Mφ) were isolated from buffy coats from healthy donors and polarized to TNF-α producing M1-like macrophages (M1-Mφ) using 25 ng/mL IFN-γ and M-CSF over two rounds from days 0 to 6 and from days 7 to 12. The M1-conditioned media was collected on days 6 and 12 (experimental groups: days 0-6 and days 7-12). Colonic fibroblasts (CCD18-Co) were cultured in a crowded pseudo-3D environment using Ficoll (75 mg/mL Ficoll 400 and 112.5 mg/mL Ficoll 70) and supplemented with 50 µg/mL ascorbic acid to support collagen synthesis. Starting on day 0, exchanging the media every fourth day, the cells were stimulated with 0.3, 3, and 30% of M1 conditioned media, 50 pM TGF-β, or 25 ng/mL IFN-γ and M-CSF for 12 days, using 3% FBS as the control. The biomarkers PRO-C1, PRO-C3, PRO-C6, and FBN-C, quantifying type I, III, and VI collagen formation and fibronectin, were measured in the supernatants on days 0, 4, 8, and 12. The metabolic activity of the cells was assessed on days 0 and 12 using an Alamar Blue assay. The statistical differences in biomarker levels were calculated using the area under the curve (AUC) compared with the unstimulated FBS control applying Brown-Forsythe and Welch or one-way ANOVA tests.

Results

Quantified using the PRO-C1, PRO-C3, PRO-C6, and FBN-C biomarkers stimulation with 30% M1-Mφ media from days 0-6 and 7-12 resulted in a switch in collagen formation compared to the FBS control, reducing type I collagen and fibronectin and increasing type III and VI collagen formation (Table 1). A similar response was observed using 3% of 0.3% M1-Mφ media, though to a lesser degree (Table 1). The data suggest increased pathological fibroblast activity and fibrotic deposition of type III and VI collagen formation driven by Mφ-derived cytokines. TGF-β, a known profibrotic cytokine, induced a significant formation of type I, III, VI, and fibronectin compared to the FBS control (Table 1), demonstrating a different fibrotic biomarker profile than Mφ induced fibrosis. IFN-γ+M-CSF stimulation, used for Mφ activation, inhibited the formation of type III collagen (Table 1).

Table 1: Overview of the statistical differences in the PRO-C1, PRO-C3, PRO-C6, and FBN-C biomarkers between the stimulated and unstimulated FBS control
Calculation of statistical differences was done between the specific stimulation and the unstimulated FBS control applying Brown-Forsythe and Welch or one-way ANOVA tests. For each biomarker a notation of "Increased or Decreased" is applied to indicate the numerical change compared to the FBS control. P-values < 0.05 was deemed a significant change. Abbreviations: ns, non-significant
Stimulation	PRO-C1 (p-value)	PRO-C3 (p-value)	PRO-C6 (p-value)	FBN-C (p-value)
30% M1-Μφ (days 0-6/days 7-12)	Decreased: p<0.05 / p<0.05	Increased: p<0.01 / p<0.0001	Increased: p<0.01 / p<0.0001	Increased: ns / p<0.01
3% M1-Μφ (days 0-6/days 7-12)	Decreased: p<0.05 / ns	Increased: ns / p<0.01	Increased: ns / p<0.001	Increased: ns / p<0.05
0.3% M1-Μφ (days 0-6/days 7-12)	ns	Increased: ns / p<0.05	ns	Increased: ns / p<0.05
TGF-β	Increased: p<0.0001	Increased: p<0.01	Increased: p<0.0001	Increased: p<0.0001
IFN-γ+M-CSF	ns	Decreased: p<0.0001	ns	ns

Conclusion

Stimulating colonic fibroblasts with conditioned media from activated macrophages induced a profibrotic response with increased type III and type VI collagen formation. The applied serological biomarkers enabled the objective quantification of collagen and fibronectin formation to assess fibrogenesis in the model and their potential use for investigating macrophage induced fibrosis.

Disclosure

Martin Pehrsson, Marta Alexdottir, Helene Hector, Morten Karsdal, and Joachim Mortensen are employed at Nordic Bioscience
Yi Luo and Eun Mi Hur are employed at Bristol-Meyers Squibb.
The remaining authors have no declaration of conflict.

ACTIVATED MACROPHAGE-CONDITIONED MEDIA INDUCES FIBROGENESIS IN A GASTROINTESTINAL SCAR-IN-A-JAR MODEL THAT IS QUANTIFIABLE WITH SEROLOGICAL BIOMARKERS OF COLLAGEN FORMATION

Martin Pehrsson 1, Frederik Gillesberg 2, Yi Luo 3, Thomai Tsapanou-Katranara 1, Eun Mi Hur 3, Anne-Christine Bay-Jensen 1, Signe Holm Nielsen 1, Sidrah Khan 2, Marta Sorokina Alexdóttir 1, Helene Sofie Hector 1, Joachim Høg Mortensen 1

1 Nordic Bioscience A/S, Herlev, Denmark

2 University of Copenhagen, Copenhagen, Denmark

3 BMS, Lawrenceville, United States

Conference

UEG Week Vienna 2024

Topics

IBD

Submission format

Abstract

Session

Identifying drug targets in IBD: Part 2 (Posters)

Citation

United European Gastroenterology Journal 2024; 12 (Supplement 8)

Published

2024

Biomarker	Timepoints						P-values
C3F	Baseline	POD	1 month	3 months	6 months	12 months	Baseline vs. POD	Baseline vs. 12 months	Stenotic vs Luminal at Baseline
Stenotic, median [IQR]	28.01 [22.21, 37.97]	34.17 [24.27, 41.24]	23.26 [20.14, 35.17]	25.76 [21.36, 36.83]	26.62 [20.99, 37.23]	31.79 [22.04, 41.07]	0.006	0.346	0.022
Luminal, median [IQR]	20.51 [15.53, 31.22]	-	25.20 [20.37, 36.75]	25.14 [20.61, 35.03]	25.46 [18.38, 30.23]	26.43 [19.86, 38.20]	-	0.016	0.022
Spearman's correlation	Markers and Demographics
C3F	Hemoglobin	Platelets	CRP	Fecal calprotectin	Length of specimen	BMI	Age at inclusion	Disease duration
Rho	-0.146	0.072	0.350	0.149	0.004	-0.088	-0.019	-0.257
p-value	0.264	0.588	0.006	0.336	0.974	0.506	0.885	0.048

C3-HNE levels	Healthy vs CD vs UC	HD	CD	UC	p-value	Healthy vs IBD	HD	IBD	p-value
	Commercial	30.85 [25.10, 34.93]	38.2 [32.94, 44.30]	35.86 [29.80, 47.16]	0.001	Cohort II	31.2 [24.48, 36.54]	36.64 [25.17, 50.71]	0.003
	Cohort I	31.2 [24.48, 36.54]	37.44 [27.07, 57.44]	39.8 [25.45, 57.25]	0.006	Cohort I	31.2 [24.48, 36.54]	37.51 [26.90, 57.06]	0.001
Technical Performance characteristics	Assay Parameter	Result	Assay Parameter	Recovery	Assay Parameter	Recovery	Assay Parameter	Recovery
	Measurement range (LLOQ-ULOQ)	15.62 – 965.59 ng/ml	Intra-assay variation (CV%)	4 – 9%	Dilution recovery in serum (1+3)	91 – 108%	Kit Stability 24 hours at 20^οC	99 – 119%
	Mean IC50	58.11 ng/ml	Inter-assay variation (CV%)	3 – 12%	Freeze-thaw stability (5 cycles)	87 – 107%	Interference hemoglobin, low/high	94 – 96%/ 87 – 98%
DSS-induced colitis model	Two-way ANOVA	Source of Variation	% of total variation	p-value	Uncorrected Fisher's LSD	Control - 5% DSS	Predicted (LS) mean diff.	95.00% CI of diff.	p-value
		Control vs. 5% DSS	6.495	0.018		Day 4	-18.57	-34.72 to -2.418	0.0248

	r	95% confidence interval	P (two-taield)
CPa9-HNE	0.30	0.15 to 0.44	<0.0001
VICM	0.33	0.18 to 0.46	<0.0001
C1M	0.37	0.23 to 0.50	<0.0001
C3M	0.29	0.14 to 0.42	<0.0001
C4M	0.23	0.08 to 0.34	0.002
C6M	0.43	0.29 to 0.54	<0.0001
ELP-3	0.30	0.16 to 0.44	<0.0001
PRO-C3	-0.20	-0.34 to -0.05	0.0071
PRO-C4	0.35	0.20 to 0.47	<0.0001

UEG's educational platform for gastroenterology education bringing recordings, courses, guidelines, and more together in one place.

Most popular content

Rate for better recommendations!