High-iron consumption decreases copper accumulation and colon length, and alters serum lipids

Abstract

In this study, we aimed to demonstrate that a significant increase in dietary iron intake disrupts the regulation of copper availability, ultimately leading to systemic copper deficiency. To investigate this, we conducted experiments using five-week-old male weanling Sprague–Dawley rats fed diets based on AIN-93G with some modifications. These diets featured varying iron content, offering choices of adequate iron (~ 120 μg/g, near the upper limit [UL]) or high iron (~ 7544 μg/g), along with low (~ 0.3 μg/g), adequate (~ 6 μg/g), or high (~ 153 μg/g) levels of dietary copper over a 5-week period. Rats consuming the high-iron diets displayed anemia, reduced copper levels in their organs and feces, and shortened colon lengths. Increased dietary iron intake resulted in an overall reduction in copper distribution within the body, likely leading to severe copper deficiency-related disorders in the experimental rats. However, the physiological disturbances caused by a high-iron diet were prevented when additional copper was included in the rodent diet. Furthermore, high iron intake led to copper deprivation, and high iron consumption resulted in elevated serum cholesterol levels. However, increasing dietary copper consumption led to a decrease in overall serum cholesterol levels. Additionally, serum alkaline phosphate and aspartate aminotransferase levels were increased by high-iron feeding, regardless of dietary copper concentration, while alanine aminotransferase levels decreased.

Introduction

Iron plays significant physiological roles in cellular oxygen delivery, DNA synthesis, and electron acquisition and donation. Iron deficiency and overload can lead to pathological disturbances, owing to the importance of iron in biology. Prolonged iron deficiency can trigger conditions such as anemia, fatigue, weakness, and pica [1,2,3]. The primary causes of iron deficiency include blood loss, inadequate dietary iron intake, and intestinal disorders [4,5,6]. Iron overload, known as hemochromatosis, is a genetic disorder caused by abnormalities in the hereditary hemochromatosis (HFE) gene [7]. Iron overload leads to the accumulation of excess iron in tissues and can result in symptoms such as joint pain [8] and insulin resistance [9] likely due to elevated oxidative stress [10]. Iron has a distinctive feature in that its intestinal absorption is crucial and tightly regulated. Therefore, almost all the required iron is biologically recycled, but a portion must be obtained through the diet. Intestinal iron absorption significantly increases in patients with low iron levels. However, in the presence of excess iron, hepatic hepcidin levels increase, further inhibiting intestinal iron absorption and/or macrophage iron recycling by suppressing the expression of ferroportin 1, an iron exporter [11, 12].

Copper also plays a vital role in promoting growth, maintaining pulmonary elasticity, regulating cardiac function, and supporting vascular health [13, 14]. Given its biological importance, copper deficiency and overload can both result in significant pathological responses. Genetic disorders such as Menkes disease (associated with copper deficiency) and Wilson's disease (linked to copper overload) are caused by the dysregulation of copper-transporting ATPases in the cellular trans-Golgi network [15,16,17]. Menkes disease, which results from an intestinal ATP7A mutation, is an X-linked genetic disorder that leads to growth retardation and severe neurodegenerative conditions in young children, often accompanied by abnormal iron absorption [18]. Wilson’s disease is an autosomal recessive hereditary disorder caused by mutations in the ATP7B gene [19]. Wilson’s disease can lead to liver cirrhosis and neuronal degeneration, which are primarily associated with copper metabolism [20]. Inappropriate dietary supplementation may also contribute to copper deficiency and its associated symptoms, such as anemia, irregular heartbeats, and loss of pigmentation in the integumentary system [21].

Biological interactions between iron and copper likely stem from their positive charges, involvement in oxidation–reduction reactions, and shared metabolic regulation [22]. Both dietary iron and copper are absorbed in the small intestine, within the brush border of small intestinal epithelial cells, and both minerals must undergo a reduction process [23]. After acquiring divalent forms, they must be oxidized before being exported into interstitial fluids and bound to their respective transport proteins [24]. Additionally, both divalent minerals act as cofactors in numerous enzymatic reactions and participate in electron donation and withdrawal [25]. There is a reciprocal relationship between biological iron and copper. Under iron deficiency conditions, copper accumulates in the liver to assist in iron utilization [26], whereas in cases of copper deficiency, iron accumulates in the liver [27]. Recent studies have focused on how excess dietary iron can lead to copper deficiency in experimental animals [28,29,30]. Excessive dietary iron intake results in growth retardation [21], increased erythropoietic demand (with decreased hemoglobin levels and elevated erythropoietin levels) [31], and copper deficiency [32], marked by reduced tissue copper levels and lower serum ceruloplasmin activity [33]. Importantly, all pathological disturbances induced by high iron intake can be significantly and gradually mitigated when additional dietary copper is provided [34].

In this study, we used adequate (AdFe) or high iron (HFe) based on previous in vivo studies [28, 29] with some modifications. We supplemented AdFe at levels close to the UL for rodents (approximately 120 μg/g), which, when extrapolated to clinical applications, surpasses the recommended iron levels (39 μg/g). Hence, we used higher AdFe levels compared to previous studies, which were set at approximately 80 μg/g [29, 35]. Our aim was to investigate how high dietary iron consumption affects responses in comparison to consuming iron levels near the UL. While previous studies have suggested that high dietary iron levels appear to influence various pathophysiological responses, potentially due to copper deprivation, their experimental designs focused primarily on iron- and copper-related metabolism. In this study, we also examined the effects of high iron consumption on glucose homeostasis, lipid profiles, and colon length.

Materials and methods

Animal experiments

All animal experimental protocols were carefully reviewed and approved by the Institutional Animal Care and Use Committee (IACUC; protocol number DKU-21–043) of Dankook University. Five-week-old, male, Sprague–Dawley (SD) rats (Doo Yeol Biotech, Inc., Seoul, Korea) were housed in overhanging wired mesh-bottomed stainless-steel cages for a 5-week dietary intervention period. The introduction age and dietary intervention periods were determined based on previous in vivo experiments with SD rats [16, 33, 36, 37], which demanded extra iron and copper due to rapid growth.

The experimental SD rats had ad libitum access to the designated diet, which was filtered through distilled water. The experimental diets were formulated based on the AIN-93G formulation with slight modifications [38, 39] (Doo Yeol Biotech, Inc.), and contained either AdFe or HFe in combination with low (LCu), adequate (AdCu), or high copper (HCu) (Tables 1, 2 and 3). In this formulation, we raised the iron content in the AdFe diet from 39 ppm (the recommended iron content in the AIN-93G diet) to approximately 120 ppm to promote normal growth and bring the dietary iron level closer to the UL for extrapolation to the human dietary reference intake (the UL of iron is approximately three times higher than the daily recommended dietary intake of iron) for these weanling rats. Additionally, the compositions of the three different HFe diets have been noted in previous peer-reviewed studies.

Table 1 Iron and copper concentration in experimental diets

Table 2 Constant ingredients in the 6 experimental diets

Table 3 Variable ingredients in the 6 experimental diets

The AdCu level was set at approximately 6 ppm following the recommendation for the AIN-93G diet, whereas HCu contained approximately 25 times more copper than the adequate level. This amount exceeded the gap between adequate intake and the UL for humans, which was approximately 11 times difference. We included HCu in our experimental design because we expected that the copper level in HCu would enhance intestinal copper absorption without surpassing a toxic threshold [28]. Previous reports have also indicated that elevated dietary copper levels can prevent iron-induced pathologies [28, 29].

Since carbonyl iron is generally not palatable, 100 g/kg sucrose was added to all formulated diets to enhance their palatability. It is worth noting that all diets containing HFe had slightly less energy (3726 kcal/kg) than AdFe-containing diets (3762 kcal/kg), since carbonyl iron was substituted for a small portion of the energy yield from corn starch. Nevertheless, the AdFe and HFe diets were considered almost isocaloric because the energy difference between the two formulations was less than 1.0%. In addition, the gross body weight of the animals was measured weekly, and food consumption was estimated by measuring the amount of food provided to each rat cage. The experimental animals were then humanely euthanized through thoracotomy after being subjected to carbon dioxide narcosis. Cardiac puncture was performed to obtain whole blood. Subsequently, the collected whole blood was centrifuged at 3000 × g and 4 °C for 15 min to separate the serum. The liver, spleen, heart, small intestine, kidney, and feces were collected, weighed, and stored at − 80 °C until further analysis. Colon length was measured using a ruler. The following are the equations employed in this section.

$$Delta\, body \,weight\, = \,Final\, body\, weight \,after \,dietary\, feeding\, -\, Initial\, body \,weight$$

$$Daily\, delta \,body \,weight \,= \,(Final \,body\, weight\, after\, dietary \,feeding \,- \,Initial \,body \,weight) / 35\, days$$

$$Food \,efficiency\, ratio \,= \,Final \,body\, weight\, after \,dietary\, feeding / Food\, intake$$

Hematological indices

After necrosis, fasting serum glucose and insulin were measured using commercial kits (Thermo Fisher Scientific, Waltham, MA, USA) and the homeostatic index of insulin resistance (HOMA-IR) was calculated by the prescribed equation [40]: HOMA-IR = serum insulin (μg/mL) · serum glucose (mg/dL)/405. Lipid panels including triglycerides (TG), total cholesterol (TC), and high-density lipoprotein cholesterol (HDL-C) levels were measured using commercially available kits (Embiel, Gunpo, Korea). Subsequently, non-HDL-C levels and the cardiac risk factor (CRF) was determined by previously reported methods [41, 42]. Non-HDL-C was calculated by subtracting HDL-C from TC and CRF was calculated by dividing TC by HDL-C. To determine hepatotoxic responses the enzymatic activities of aspartate aminotransferase (AST), alanine transaminase (ALT), and alkaline phosphatase (ALP) from serum were measured using their respective commercial kits (Embiel).

Iron and copper status

Hemoglobin (Hb) and hematocrit (Hct) levels were measured using a Cera-Chek Hb Plus reader (Green Cross, Yongin, Korea). Serum non-heme iron levels and total iron-binding capacity (TIBC) were quantified using a colorimetric analytical method [43, 44]. Subsequently, transferrin (Tf) saturation was calculated using the following equation: serum non-heme iron/TIBC × 100. The levels of iron and copper in the tissues were determined by inductively coupled plasma mass spectrometry (ICP-MS). In brief, a specimen weighing approximately 0.1 to 0.5 g was placed in a digestion vessel, and 5 mL of nitric acid (Daejung, Siheung, Korea) was added. Subsequently, the samples were processed using Titan MPS (PerkinElmer, Inc., Waltham, MA, USA). All prepared samples were then analyzed using an ICP-MS system (NexION 350X; PerkinElmer, Inc.) with appropriate dilutions using a standard solution, ICP-MS Calibration Standard 2 (AccuStandard, Inc., New Haven, CT, USA), and nitric acid, as required.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from the liver, and cDNA synthesis and relative qRT-PCR analyses were conducted according to methods described in previous studies [42, 45, 46]. Primers used for qRT-PCR are from previous reports for hepcidin [28, 35] and GAPDH [47].

Statistical analysis

The results are presented as means and standard deviations, or as box-and-whisker plots. All figures were created using the GraphPad Prism 5 software (GraphPad Software Inc., San Diego, CA, USA). Data were statistically analyzed using two-way ANOVA with Tukey’s post hoc test performed using XLSTAT 2012 (Addinsoft, Inc., Paris, France) to examine the effects of iron and copper, as well as their interactions. Statistical significance was set at P < 0.05. A summary of the two-way ANOVA results for both the main effects and interactions is provided in Table 4. Furthermore, the Pearson product-moment correction coefficient (r) was calculated to clarify the relationship between intestinal and fecal mineral content using XLSTAT 2012.

Table 4 Summary of statistical analysis by two-way analysis variance (ANOVA) for main effects and interactions

Results and discussion

Diet analysis

The iron concentrations in three different AdFe diets ranged from 117.0 to 122.2 μg/g, as determined by ICP-MS analysis. In HFe diets, the iron concentrations ranged from 6740 to 7544 μg/g. When we extrapolated our dietary iron concentration to the dietary reference intake for humans, it appeared that the dietary iron concentration may reach the UL, as it was approximately three times higher than the recommended iron intake for rapidly growing murine models.

The dietary copper levels in LCu were 0.31 and 0.37 μg/g in AdFe/LCu and HFe/LCu, respectively, as determined by ICP-MS analysis. In AdCu, the dietary copper levels were 6.05 and 6.39 μg/g in AdFe/AdCu and HFe/AdCu, respectively, and in HCu, the dietary copper levels were 153.5 and 154.8 μg/g in AdFe/AdCu and HFe/AdCu, respectively, according to ICP-MS analysis. The LCu diets contained approximately 20-fold less dietary copper, whereas the HCu diets had approximately 25-fold higher copper levels than the AdCu diets.

Body weight and food intake

As mentioned earlier, we expected that high iron consumption would result in a reduction in final body weight, delta body weight, and daily delta body weight due to copper depletion in experimental animals [28, 29]. However, in this experimental setup, high iron consumption had no effect on final body weight, delta body weight, or daily delta body weight (Fig. 1A–C). To investigate whether variations in dietary iron and copper levels affected energy intake, we assessed daily food intake, daily energy intake, and the food efficiency ratio. However, no significant differences were observed (Fig. 1D–E).

Fig. 1

Effects of high-iron feeding on the body weight, food intake, energy intake, and food efficiency ratio of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Final body weight (A), delta body weight (the final body weight after dietary feeding─the initial body weight) (B), daily delta body weight (C), food intake (D), energy intake (E), and food efficiency ratio (F) were measured. The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. L low; Ad adequate; H high

In this experiment, our failure to detect significant growth retardation from HFe feeding may be attributed to two dietary factors: higher AdFe and lower HFe levels compared with previous studies. AdFe levels were close to the recommended UL for humans, and copper deficiency was noted at this concentration. Therefore, the growth rate in the AdFe group may have been lower than those reported in previous studies. Furthermore, the overall iron concentration in HFe in this study (7249 μg/g) was 18% lower than the iron concentration in HFe in the two previous reports (8820 μg/g) [28, 29], and relatively lower iron levels in HFe may not be sufficient to trigger copper-depleted growth retardation.

Organ weights

HFe-induced growth retardation was not observed in our experimental setup. However, to investigate whether HFe consumption altered the previously observed pathological responses [28, 29], we measured the relative weights of the heart, liver, kidney, spleen, pancreas, and epididymal white adipose tissue (Fig. 2A–F). As noted previously, kidney and spleen weights did not vary among the six dietary interventions, and the weight of the epididymal white adipose tissue (EAT) remained unchanged [28, 29].

Fig. 2

Effects of high-iron feeding on relative organ weights of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Relative organ weights of the heart (A), liver (B), kidney (C), spleen (D), pancreas (E), and EAT (F) were measured. The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high; BW body weight

Notably, the relative weights of the heart and liver confirmed those reported in our previous animal experimental protocols [28, 48]. Two distinct features were observed in the relative heart weight, indicating cardiac hypertrophy. Regardless of the dietary iron concentration, LCu consumption significantly increased the heart weight. Furthermore, consumption significantly triggered cardiac hypertrophy. However, a gradual increase in dietary copper concentration normalized heart weight compared with AdFe/AdCu, as observed in peer-reviewed reports [28, 29].

Regarding the relative liver weight, high iron consumption induced significant hepatomegaly. However, as previously observed, HFe-induced hepatomegaly was normalized by additional copper consumption. Additionally, the pancreatic weight responded dynamically to various amounts of iron and copper. Lower dietary copper intake reduced pancreatic weight, and this decline was more significant when high iron intake was combined with low copper intake. In HFe/AdCu, the weight of the pancreas was restored; however, HFe/HCu decreased its weight compared to HFe/AdCu, but maintained a weight similar to that seen in the AdFe groups.

Serum glucose, insulin, and HOMA-IR

Uncontrolled iron levels, such as iron deficiency anemia [49] and iron overload are related to insulin resistance [50]. Iron overload, such as hereditary hemochromatosis [51] and β-thalassemia [52] triggers insulin resistance by destruction of pancreatic β-cells [50] probably due to extra iron accumulation. Due to the dynamic changes in relative pancreatic weight, we logically assessed glucose metabolism-related parameters in the fasting serum, including glucose, insulin, and HOMA-IR. Serum glucose levels increased following HFe consumption (Fig. 3A). The serum insulin level was the lowest in the HFe/LCu group (Fig. 3B), and the decrease in systemic insulin in the HFe/LCu group may be attributed to the reduction in pancreatic weight (Fig. 2E). In a genetic iron overload model, patients with hemochromatosis often experienced insulin resistance, likely due to elevated oxidative stress resulting from increased iron accumulation in the pancreas. Our findings suggest that a combination of high iron accumulation and low copper concentrations can exacerbate insulin resistance. However, HOMA-IR remained consistent across the six dietary interventions, with varying iron and copper levels (Fig. 3C).

Fig. 3

Effects of high-iron feeding on serum glucose and insulin levels of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Serum glucose (A), insulin (B), and HOMA-IR levels (C). The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high; HOMA-IR homeostasis model assessment-estimated insulin resistance

Lipid panels

Evidence suggests that both iron and copper are significant factors in the pathological modulation of atherosclerosis and cardiovascular diseases. However, the specific interactions that detail how dietary iron and copper affect lipid profiles have not yet been reported. In our study, we analyzed the blood lipid and cholesterol levels in the experimental animals. Various combinations of dietary iron and copper did not significantly alter the serum triglyceride (TG) levels (Fig. 4A). However, serum cholesterol levels responded dynamically to our dietary intervention (Fig. 4B). Serum TC levels were altered in a copper-sensitive manner (P < 0.05). In the AdFe group, extra copper consumption elevated TC levels. In the HFe groups, in combination with LCu, TC levels were the highest; however, a gradual increase in dietary copper decreased TC levels. AdFe/AdCu exhibited the lowest HDL-C levels among the various dietary interventions (Fig. 4C). In combination with LCu, HCu, or AdFe, HDL-C levels were significantly elevated. In HFe, as seen in the TC levels, HFe/LCu showed the highest HDL-C levels; however, the gradual induction of dietary copper decreased the TC levels in conjunction with HFe. In the AdFe group, non-HDL-C levels tended to increase with increasing dietary copper concentration (Fig. 4D). Therefore, AdFe/LCu exhibited the lowest non-HDL-C level, whereas AdFe/HCu exhibited the highest. However, in the HFe groups, increasing the dietary copper concentration tended to diminish the non-HDL-C levels (Fig. 4D). Therefore, HFe/LCu exhibited the highest non-HDL-C levels, whereas HFe/AdCu and HFe/HCu exhibited lower non-HDL-C levels than HFe/AdCu. It is likely that HFe consumption triggers dyslipidemia, which is generally observed in genetically iron-overloaded populations, and alterations in cholesterol levels may be a potential risk factor for the induction of cardiovascular diseases [53, 54]. Intriguingly, the gradual elevation of dietary copper significantly diminished the overall cholesterol levels, regardless of the cholesterol type. Exposure to additional divalent minerals may suppress de novo cholesterol synthesis [55]. Cardiac risk factors significantly decreased in all dietary treatments, except for AdFe/HCu, compared to AdFe/AdCu (Fig. 4E).

Fig. 4

Effects of high-iron feeding on serum lipid panels of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Serum triglyceride (A), total cholesterol (B), HDL-cholesterol (C), non-HDL-cholesterol (D), and cardiac risk factor (E). The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high; HDL high-density lipoprotein

Enigmatically, there is a report that excess iron (i.e., non-transferrin-bound iron) exposure to primary human umbilical vein endothelial cells induces de novo cholesterol synthesis, probably via regulation of sterol regulatory element-binding protein 2 [56]. Furthermore, excess iron exposure also induces cellular apoptosis, likely through the elevation of tumor necrosis factor alpha (TNFα) [56]. These findings were verified using a cholesterol chelator (hydroxypropyl-β-cyclodextrin) since cholesterol chelation reduces excess-iron-triggered TNFα-inducible apoptosis [56].

In our study, excess iron exposure decreased overall cholesterol levels regardless of the type of cholesterol. Probably, at the cellular level, initial cholesterol synthesis may be inducible, as reported previously [56]; however, due to long-term dietary exposure, both iron and copper may also trigger cellular apoptosis, ultimately leading to a decrease in cholesterol levels. Moreover, extra copper supplementation with extra iron rather decreased overall cholesterol levels; however, based on the current gathered information, we could not conclude whether diminished overall cholesterol levels were due to excess cellular apoptosis triggered by extra copper or had preventive effects described as other biological outcomes in this study. Therefore, further intensive and specific in vivo experiments are required to understand how dietary iron and copper affect overall cholesterol metabolism.

Hepatic functional biological indices

Both iron and copper are transient divalent metals; therefore, exposure to excess iron and copper may lead to changes in hepatic enzymatic functional biological markers such as ALP, AST, and ALT activity levels in the serum. To determine whether the six different dietary interventions could alter hepatic enzymatic functional biological markers, we assessed the ALP, AST, and ALT activity levels in the serum.

Serum ALP activity levels were significantly influenced by dietary iron (P < 0.0001), with HFe resulting in significantly lower ALP activity than AdFe (Fig. 5A). It is noteworthy that individual with genetic conditions such as Menkes disease (copper deficiency [18]) and Wilson's disease (copper overload [57]) exhibit lower ALP levels than normal subjects. Therefore, reduced serum ALP activity may be linked to HFe-induced copper deficiency [58]. Also, serum AST levels were strongly affected by dietary iron (P < 0.0001), and HFe resulted in significantly lower AST activity than AdFe (Fig. 5B).

Fig. 5

Effects of high-iron feeding on the enzyme levels related to hepatic function of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. ALP (A), AST (B), and ALT (C) activities. The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high; ALP alkaline phosphate; AST aspartate amino transferase; ALT alanine amino transferase

Interestingly, serum ALT activity was also significantly influenced by dietary iron (P < 0.0001); serum ALT activity was elevated in the HFe groups compared to that in the AdFe groups. Serum ALT activity can vary dynamically, either increasing or decreasing in conditions such as Wilson’s disease [59]. Therefore, serum ALT levels may serve as sensitive markers for copper alterations, and in our experimental setting, HFe-induced copper depletion may have contributed to the reduction in serum ALT activity.

Hematological parameters

All hematological parameters have been previously validated in peer-reviewed publications [28, 60]. In the AdFe group, AdFe/LCu significantly reduced Hb levels, which could be attributed to copper deficiency (Fig. 6A). In the HFe group, HFe/LCu reduced Hb levels to a greater extent than AdFe/LCu. However, with an increase in dietary copper levels in HFe, Hb levels gradually normalized to levels similar to those in the AdFe/AdCu group. Hct levels exhibited trends similar to those of Hb as previously reported (Fig. 6B) [28, 29]. AdFe/LCu showed a decreasing trend in Hct (although not statistically significant), and HFe significantly reduced the proportion of red blood cells. However, additional copper supplementation prevented the copper-deprivation effect induced by HFe. Serum non-heme iron levels were significantly elevated in HFe/HCu mice due to a significant dietary copper interaction (Fig. 6C; P < 0.01). The TIBC remained consistent across all six dietary interventions (Fig. 6D). Considering the serum non-heme iron and TIBC levels, HFe/HCu exhibited the highest Tf saturation among all dietary treatments (Fig. 6E).

Fig. 6

Effects of high-iron feeding on hematological parameters of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Hb (A), HCT (B), serum non-heme Fe (C), TIBC (D), and Tf saturation (E). The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high; Hb hemoglobin; Hct hematocrit; TIBC totaliron-binding capacity; Tf  transferrin

In a previous communication, we reported that the expression of erythropoietin (Epo) mRNA in the kidneys of rats accurately reflected the levels of the circulating hormone [28]. Furthermore, all hematological indices, including diminished Hb and Hct, were validated according to previous experimental animal protocols [28, 29, 35, 48]. In this experimental setting, when HFe conditions were combined with LCu, erythropoietic demands were expected to be higher than in other scenarios. However, these hematological disturbances were gradually prevented when extra dietary copper content was introduced [28, 29, 35, 48]. We logically postulated that the additional copper, when combined with high iron, should lead to a decrease in erythropoietic demands (i.e. Epo expression and red blood cell (RBC) contents in whole blood).

Tissue iron loading and hepatic hepcidin mRNA expression

Intestinal iron absorption is tightly regulated, and the distribution and metabolism of iron in multiple tissues is controlled [61, 62]. To understand how the six different dietary interventions affected the distribution of iron in the systemic and peripheral tissues, iron content was analyzed using ICP-MS. Hepatic iron loading was influenced by both dietary iron (P < 0.0001) and copper (P < 0.01). HFe treatment resulted in a significantly higher hepatic iron loading. Interestingly, regardless of the dietary iron concentration, extra copper decreased hepatic copper loading. This slight but significant reduction in hepatic copper may be due to the innate ferroxidase activity of biological copper, which enhances iron utilization.

Hepatic hepcidin mRNA expression is intricately linked to hepatic iron loading, therefore, we measured hepatic hepcidin mRNA expression using qRT-PCR (Table 5). However, only HFe/HCu increased hepatic hepcidin mRNA expression. Our findings which also showed the highest hepatic hepcidin mRNA expression among all HFe groups were not consistent with those of previous reports [28]. In previous studies, adequate iron levels likely targeted the recommended intake, whereas in our study adequate dietary iron was aimed at the UL. Therefore, high dietary iron levels may increase hepatic hepcidin mRNA expression. Differences in dietary settings may explain why this study did not replicate previous findings [28, 35]. However, a clear and evident finding is that high iron levels, when combined with high copper levels, can maximize hepatic hepcidin expression.

Table 5 Tissue iron and liver hepcidin levels

Splenic iron accumulation was influenced by both dietary iron (P < 0.0001) and copper (P < 0.0001). No significant differences were observed when AdFe was combined with either LCu or AdCu. However, AdFe/HCu significantly elevated splenic iron loading, likely due to the extra copper enhancing its utilization. Furthermore, in the HFe groups, HFe/LCu was loaded with similar levels of iron as AdFe/LCu and AdFe/AdCu. However, HFe/AdCu and HFe/HFe showed the highest splenic iron content among all groups. Cardiac iron loading was influenced not by dietary iron, but by dietary copper (P < 0.05), and only HFe/LCu loaded significantly less cardiac iron than the other dietary interventions. Renal iron accumulation was affected by both dietary iron (P < 0.01) and copper (P < 0.0001) levels. All the AdFe and HFe/LCu groups had statistically identical renal iron levels. However, in the HFe group, increasing dietary copper levels significantly elevated renal iron loading, which is consistent with previous reports [28, 35]. Serum iron levels were not influenced by dietary copper but were influenced by iron levels (P < 0.0001). The HFe group exhibited significantly higher serum iron levels than the AdFe group, regardless of the dietary copper concentration (Table 5).

Table 6 Tissue copper levels

Splenic iron acquisition plays a crucial role in the overall regulation of the body's iron metabolism [63, 64]. During iron deprivation, renal Epo expression is induced due to increased erythropoietic demands. Consequently, renal Epo upregulates splenic erythroferrone expression, which, in turn, may counteract hepatic hepcidin expression to facilitate additional iron acquisition [65]. In β-thalassemia patients, the specific role of splenic erythroferrone is emphasized, with erythroferrone exacerbating iron loading [66]. A previous study highlighted the sensitivity of splenic erythroferrone expression to iron-deprived conditions but not in conditions of HFe [28]. Additionally, in HFe groups, splenic erythroferrone expression showed no significant variation with different dietary copper contents [28]. Although, erythroferrone serves as a highly responsive regulatory hormone in iron metabolism, its role may be minimal in the context of high dietary iron consumption, potentially being more closely associated with splenic iron content.

Tissue copper loading

To understand how various dietary iron and copper levels influence copper loading in multiple tissues (such as the liver, spleen, heart, and kidney) and circulating biological copper content, we analyzed them using ICP-MS systems. Interestingly, copper levels in selected tissues and serum exhibited a similar pattern of response, influenced by dietary iron (P < 0.0001), copper (P < 0.0001), and the interaction between dietary iron and copper (P < 0.0001). All LCu groups loaded less copper than the other groups. However, when LCu was combined with HFe, the hepatic and renal copper loadings were further reduced. In general, HFe results in lower copper levels in the tissues and serum. Nevertheless, the HFe-induced copper depletion was gradually prevented by additional dietary copper consumption in our experimental SD rats, as previously reported [28, 35]. The observed copper depletion in the heart owing to HFe feeding may be a significant potential cause of cardiac hypertrophy (Fig. 2A, Table 6).

Colon length, iron and copper content in small intestine and feces

Following our dietary formula, both HFe and HCu supply iron and copper in amounts significantly higher than the biological requirements, and the excess should reach the colon. Thus, unabsorbed dietary iron and copper can alter the colonic environment and physiology. To investigate our research hypothesis, we first measured and summarized colon length (Fig. 7A, B). Colon lengths exhibited different patterns in the AdFe and HFe groups. In the AdFe groups, colon length was significantly reduced only when extra copper was added. However, in the HFe/LCu group, the colon length was the shortest among all groups. Nevertheless, a gradual increase in dietary copper concentration prevented shortening of the colon length.

Fig. 7

Effects of high-iron feeding on colon length and iron and copper concentration in feces and the small intestine of rats. Five-week-old, male, SD rats were fed one of six diets, which varied only in their iron and copper content for 5 weeks ad libitum. Colon length (A), image of colon length (B), iron concentration in feces (C), and the small intestine (D), copper concentration in feces (E), and the small intestine (F). The results are presented as means ± SDs; n = 10 per individual group. Data were analyzed by two-way ANOVA with Tukey’s post hoc test for further comparisons. Different letters indicate significant differences (P < 0.05) among the labeled means. L low; Ad adequate; H high

In our previous studies [35, 48], we consistently observed that HFe triggered copper depletion, and that HFe-induced systemic copper depletion was prevented by extra dietary copper supplementation. Therefore, the normalization of colon length in HFe when supplying extra copper, may be closely related to copper depletion and repletion. However, further in-depth-studies are required to understand the exact underlying molecular mechanisms. Intriguingly, in AdFe, extra copper decreases colon length and may elevate oxidative stress in the colon. However, further studies are required to fully elucidate these phenomena.

To assess the absorption and excretion of iron and copper in our experimental settings, we measured the iron and copper contents in the feces and small intestine. Fecal iron levels were influenced by dietary iron (P < 0.0001) and copper (P < 0.01) and were significantly higher in all HFe groups than in the AdFe groups (Fig. 7C). Interestingly, the highest fecal iron levels were observed in HFe/AdCu, with HFe/AdCu having the second-highest levels and HFe/LCu exhibiting the lowest fecal iron levels among the HFe groups. The rate of intestinal iron absorption is tightly regulated [67], which may explain the levels of fecal iron excretion. Notably, there was a positive correlation between intestinal and fecal iron content, as determined by Pearson’s test (P < 0.0001; r = 0.817; Fig. 7C, D).

Copper absorption and excretion are influenced by both dietary iron and copper content. The copper content in the feces was affected by dietary iron (P < 0.05) and copper (P < 0.0001) (Fig. 7E). Again, there was a positive correlation between intestinal copper content and fecal copper content, as indicated by the Pearson’s test (P < 0.0001; r = 0.922). Furthermore, the intestinal copper content was affected by dietary iron (P < 0.0001) and copper (P < 0.0001) (Fig. 7F). Although, HFe leads to Cu deficiency, intestinal Cu absorption and excretion are influenced by dietary iron and copper concentrations.

The colonial microbiome is influenced by various dietary factors, such as energy nutrients and vitamins [68, 69]. Recently, attention has been given to the influence of the microbiome on minerals, including iron [70]. In iron-deprived conditions, certain probiotics enhance intestinal iron absorption to prevent iron deprivation. For example, exposure to specific probiotics, such as Lactobacillus plantarum [71], L. plantarum 299v [72], Streptococcus thermophiles [73], L. fermentum [74], and L. acidophilus [75], significantly elevates iron utilization in vivo by increasing iron availability [70].

In addition to the consumption of probiotics to aid iron utilization, iron supplementation is an easy and practical means to prevent iron deficiency. However, excess iron absorption negatively alters the colonic environment due to unabsorbed iron [76], resulting in a decrease in short-chain fatty acid production and an increase in Enterobacteriaceae [76]. Indeed, the majority of colonic bacteria require iron to maintain normal physiology [76]. However, due to the competitiveness between beneficial and harmful colonic bacteria, harmful bacteria prefer to utilize iron over beneficial bacteria. Therefore, exposure to excess iron reduces beneficial Bifidobacterium and Lactobacillus spp., but elevates harmful Enterobacteriaceae spp. [76,77,78,79].

In this experiment, we found that extra iron decreased colon length, which may be closely related to colonial inflammation and/or oxidative stress. However, when extra copper was combined with HFe, colon length was normalized. This dynamic alteration in colon length is probably related to changes in the microbiome. While we expected HFe to elevate harmful bacterial population, the combination with extra copper may decrease the relative harmful bacterial population. Therefore, to answer our extended research question, we may characterize the fecal microbiome in a follow-up study.