Subject characteristics are presented in Table 1. By design cases and controls were similar in age, BMI and waist circumference. Race/ethnicity, dietary iron intake (heme and non-heme food and supplemental sources), medication use and alcohol consumption were also similar between cases and controls.
Systemic Iron Regulation and Inflammation
Systemic iron status and inflammatory parameters are presented in Table 2. Cases had significantly lower Hb and higher sTfR compared to controls (p < 0.05), indicative of iron-restricted erythropoiesis. Serum hepcidin was mildly decreased in cases, likely as a response to the increased erythroid iron demand (as reflected by the elevated sTfR). However, considering that hepcidin concentrations were still in the normal range for cases, hepcidin may be elevated given their degree of iron restriction. Hepcidin is known to be low or undetectable in individuals with insufficient iron status. Cases also had significantly elevated CRP (p < 0.05) and a trend toward increased IL-6 concentrations compared to controls, indicating the presence of mild inflammation in these patients. This may stimulate hepcidin production and counterbalance the suppressive effect of iron-restricted erythropoiesis on hepcidin.
Associations between serum hepcidin and iron status and inflammatory parameters were explored with Spearman correlation coefficients. In controls, serum hepcidin was correlated with Hb (r = 0.54; p = 0.01) as expected, and no associations were found with sTfR or inflammatory markers (CRP, IL-6, TNF-α), which were in the normal range. In cases, serum hepcidin was not correlated with any of the parameters (Hb, sTfR or the inflammatory markers). Additionally, cancer staging did not change serum hepcidin concentrations (Stage IV: median 64.4 (IQR 249.1) versus Stage I: 72.8 (IQR 28.7) ng/mL p = 1.0). Since our lab previously demonstrated that obesity-induced inflammation is associated with elevated serum hepcidin levels, we stratified cases and controls by obesity status (obese ≥ 102 cm, lean < 102 cm).[26,27] Within cases, circulating markers of inflammation and serum hepcidin did not differ by obesity status (data not shown). Obese controls were more inflamed (CRP, IL-6) compared to lean; however, serum hepcidin concentrations were similar between the groups.
Iron Transporters in Colonic Mucosa
Gene expression (mRNA) of hepcidin, iron transporters (DMT-1 and FPN) and the pro-inflammatory protein IL-6 in colonic mucosa are presented in Table 3. Expression of DMT-1 and FPN were similar between cases and controls. Although cases had a 2.9-fold lower expression of hepcidin compared to controls (p < 0.05), overall both groups expressed very low hepcidin mRNA in the colonic mucosa (close to the limit of detection by qPCR). Cases had a 9.4-fold higher expression of IL-6 compared to controls (p < 0.05), confirming the inflammatory nature of the tumor. Correlation analyses did not reveal significant associations between expression of the serum and colonic markers of iron regulation, iron transport or inflammation in cases or controls (data not shown).
Iron Accumulation in Colonic Mucosa
To determine if CRC was associated with greater tissue iron accumulation, adenocarcinoma from cases and healthy colonic mucosa from controls was examined using Perls' Prussian blue stain. Iron accumulation was present in more cases than controls (n = 6/20; 30%; n = 1/20; 5%, χ2 = 5.00; p < 0.05). Illustrative Perls' stains are shown in Figure 1 (cases A-B, controls C-D). When cases were dichotomized by presence/absence of iron accumulation (+/-), cases with iron accumulation had higher serum hepcidin compared to the cases without iron accumulation (p < 0.05) (Table 4). However, after adjusting for Hb, differences in serum hepcidin between these subgroups became non-significant. This suggests that systemic iron sufficiency may have contributed to the higher serum hepcidin observed in the iron accumulation (+) group given that hepcidin is increased when iron stores are adequate. In support, there was a trend for lower sTfR in the iron (+) group, demonstrating iron bioavailability. Differences in systemic inflammatory markers (CRP, IL-6 or TNF-α) between cases with different iron accumulation (+/-) were non-significant. Additionally, no differences were observed for tissue level (mRNA expression for DMT-1, FPN, hepcidin, IL-6) parameters or with cancer staging (data not shown).
Iron staining in colonic tissue between cases and controls. Sections of colonic tissue (adenocarcinoma for cases, healthy mucosa for controls) were tested for presence of iron accumulation using the Perls' Prussian blue staining. Cases with presence (A) and absence (B) of iron accumulation were compared to controls with presence (C) and absence of iron accumulation (D) and presented as 20X magnification. There was more iron accumulation detected in cases than controls (n = 6/20; 30%; n = 1/20; 5%; McNemar's test: χ 2 = 5.0; p < 0.05).
Nutr Metab. 2014;11(21) © 2014 BioMed Central, Ltd.