All subjects signed an informed consent and the study procedures were approved by the University of Illinois at Chicago Institutional Review Board.
Study Population and Characteristics
Study subjects were recruited from patients scheduled for colonoscopies due to abdominal pain, bloating, change in bowel movements or for CRC screening at the University of Illinois at Chicago and John H. Stroger Jr. Cook County Hospital between May 2011 and June 2012. "Cases" were classified as newly diagnosed CRC with adenocarcinoma based on pathology reports for their tumor biopsies; "controls" were selected from the subjects with healthy colonic mucosa (absence of adenomatous polyps or GI abnormalities). Cases and controls (n = 20/group) were matched to have a similar difference within each pair for age (within 5 years), body mass index (BMI) (within 4 units) and waist circumference (within 5 cm). Due to gender-specific variation in reference ranges for iron parameters, participation was restricted to males. Additional exclusions included medical conditions that could affect iron status such as gastrointestinal bleeding, hemochromatosis, history of inflammatory bowel disease or infection.
Following informed consent, questionnaires for basic demographic information, health history, medication and supplement use and alcohol consumption were administered by a research team member. The Block Brief 2000 food frequency questionnaire (FFQ) was used to assess usual dietary intake over the previous 12 months. Height was measured with a stadiometer to nearest 0.1 cm and weight using a balance beam scale to nearest 0.1 kilograms with subjects wearing a hospital gown. Waist circumference was measured with a flexible tape (AccuFitness, Greenwood Village, CO) at the midpoint between the ribs and iliac crest, to the nearest 0.1 cm. Body mass index was calculated as weight in kilograms divided by height in meters squared. CRC staging (0-IV) based on tumor size, lymph nodes affected and metastasis (TNM) was classified using the American Joint Committee on Cancer (AJCC) criteria.
All blood samples were collected after a minimum 12 hour fast following the endoscopy or prior to surgical intervention. A separate analysis did not reveal any differences in serum parameters by blood collection type. Colonic adenocarcinoma tissue was obtained from cases at the time of surgical resection. Healthy colonic mucosa from the descending colon in the controls was obtained using standard sized biopsy forceps during colonoscopy.
Systemic iron regulation was assessed by serum hepcidin and iron status by serum transferrin receptor (sTfR), which primarily reflects erythroid iron demand. Serum transferrin receptor is not influenced by acute or chronic inflammation and can help differentiate between iron-related disorders. Elevated sTfR is indicative of iron-deficient erythropoiesis. Anemia was examined by hemoglobin (Hb). Inflammation was evaluated via quantification of C-reactive protein (CRP), interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α).
Serum hepcidin was measured by competitive enzyme-linked immunosorbent assay (c-ELISA) (Intrinsic LifeSciences, La Jolla, CA). For healthy men with normal iron status, the range for this assay is 29–254 ng/mL. Serum transferrin receptor was measured using the Quantikine IVD Human sTfR Immunoassay ELISA (R&D Systems, Minneapolis, MN; normal reference range 8.7–28.1 nmol/L). Hemoglobin was obtained from the medical chart and anemia was defined as <12 g/dL. CRP, IL-6 and TNF-α were measured using R&D Systems Quantikine ELISA kits (R&D Systems, Minneapolis, MN).
A portion of colonic tissue was placed in formalin and paraffin-embedded for histological analysis. The remaining portion of the colonic tissue was placed in RNAlater (Ambion, Austin, TX) and stored at -80°C for gene expression (mRNA) analysis.
Real Time Polymerase Chain Reaction (RT-PCR)
Total RNA was extracted from colonic mucosa sections using the Maxwell 16 System (Promega, Fitchburg, WI). The complementary DNA was synthesized from the RNA using iScript™ cDNA Synthesis Kit (BioRad, Hercules, CA). Gene expression (mRNA) of divalent metal transporter-1 (DMT-1), ferroportin (FPN), hepcidin and IL-6 were measured quantitatively by RT-PCR (For primers used see Additional file 1) using SsoAdvanced SYBR Green Supermix (BioRad). Amplifications were performed at 57°C for 40 cycles using the C1000 Touch (BioRad) and data analyzed using the CFX Manager software (BioRad). Results were normalized to reference genes Glyceraldehyde 3-phosphate dehydrogenase (GADPH) and β-actin to obtain ΔCt values (ΔCt = Ct[reference] - Ct[target]). As both GADPH and β-actin showed equivalent results, we used the average of the two values for the final representation. Fold change in mRNA expression in cases compared to controls was calculated using 2ΔΔCt, where ΔΔCt = average ΔCt for cases - average ΔCt for controls. Statistical comparison was performed using raw ΔCt values (Additional file 2).
Additional File 2.
mRNA expression of iron transporters and inflammatory proteins in colonic tissue of colorectal cancer cases and controls using raw ΔCt values
*p-value of difference between cases and controls using student's paired t-test.DMT-1, divalent metal transporter-1; FPN, ferroportin; Hamp, hepcidin; IL-6, interleukin-6.
Tissue iron accumulation was measured qualitatively by Perls' Prussian blue staining. The tissue section was treated with dilute hydrochloric acid to release ferric ions from binding proteins. These ions then reacted with potassium ferrocyanide to produce an insoluble blue compound (the Prussian blue reaction). The grading of iron staining was performed by a pathologist blinded to the experimental group and reported dichotomously as presence or absence of iron accumulation (+/-).
Prior to analysis, all variables were assessed for normality and presence of outliers. Descriptive statistics included mean, standard deviation (SD), median and interquartile range (IQR) for continuous variables and frequencies for categorical variables. Difference between cases and controls was assessed by student's paired t-test or non-parametric Wilcoxon signed-rank test for continuous variables and McNemar's test for categorical variables. Non-parametric Spearman's Rank correlation coefficient was used to explore bivariate relationships. Multivariable linear regression was performed to adjust for confounders. All analyses were performed using SAS version 9.3 (Cary, North Carolina). P-values were two-sided and the statistical significance level was defined as p < 0.05.
Nutr Metab. 2014;11(21) © 2014 BioMed Central, Ltd.