The influence of gastrointestinal pH on speciation of copper in simulated digestive juice

Abstract Speciation can provide knowledge about absorption, reactivity to binding sites, bioavailability, toxicity, and excretion of elements. In this study, the speciation of copper in different model solutions under the influence of gastrointestinal (GI) pH was studied by ion selective electrode (ISE) and inductively coupled plasma optical emission spectrometry (ICP OES). It was found that the electrode response (mV) against Cu2+ decreased with the increase in pH and dropped to the lowest point at pH 7.5 in all model solutions. When amino acids and organic acids were present, the ratio of filtered copper (0.45 μm, pH 7.5) was more than 90%. When casein was present, whey protein, pancreatin, and starch were added, and the ratio of filtered copper was 85.6 ± 0.3, 56.7 ± 8.8, 38.5 ± 5.1, and 1.0 ± 0.3%, respectively. When there is not enough organic ligand, excessive copper will form copper hydroxide precipitation with the increase in pH, but it got the highest electrode response (mV) against Cu2+. From this study, it can be concluded that the speciation of copper in GI tract is strongly influenced by the pH and the composition of food. When there are few ligands coexisting in the GI tract, the concentration of copper ion may be relatively high.


| INTRODUC TI ON
2003Copper is an essential trace metal with a crucial role in various biological systems. As a structural or catalytic cofactor, it was required for respiration, connective tissue formation, iron metabolism, and many other processes (Zehra et al., 2021). However, excessive copper intake or copper excretion disorders can lead to oxidative tissue damage through free radical-mediated pathways, resulting in toxicity, such as Indian childhood cirrhosis and Wilson's disease (Dai et al., 2020;Gaetke et al., 2014;Jomova & Valko, 2011;Pierson et al., 2019). The imbalance of copper also promotes the production of neurodegenerative diseases (Li & Kerman, 2021;Ma et al., 2020;Pal, 2014;Squitti et al., 2014). Therefore, the balance of copper is important to human health.
The major source of copper is food for humans, and the absorption and toxicity of copper in various food seem different. Organic copper has higher bioavailability and lower toxicity than inorganic copper (Liu et al., 2019;Vieira et al., 2020;Zhi et al., 2020). The protein source had a significant effect on copper bioavailability in rainbow trout; those fed diet with soy and corn proteins require lower copper than those fed menhaden and blood proteins (Read et al., 2014). The efficiency of apparent copper absorption from the lacto-ovo-vegetarian diet was less (33%) than that from the nonvegetarian diet (42%) (pooled SD: 9%; p < .05) (Hunt & Vanderpool, 2001). The food with different composition (breakfast, lunch, and dinner) leads to various copper bioaccessibility (8.48 ± 3.54, 27.04 ± 10.49, and 31.09 ± 18.08%, respectively) (Velasco-Reynold et al., 2007). In our previous study, the mice that intake copper with water have higher oxidative stress, serum-free copper, and brain's Aβ 1-42 when compared to mice intake copper with food (Wu et al., 2016).
The different absorption rate and toxicity of copper in various food are related to its speciation, such as the concentration of dissolved copper, copper ions, and binding strength of ligands with copper in the lumen of the GI tract. Despite the complex and variable environments during digestion, previous research has established that the pH gradient in the GI tract is one of the most important environmental factors for the speciation of copper (Dendougui & Schwedt, 2002). Few researches are focused on the changes in speciation of copper in the process of digestion. Mills add copper in an aqueous extract of herbage and found copper was in the form of complexes that appear to be stable above pH 2-5 (Mills, 1956).
Schwedt uses ion selective electrode (ISE) to determine the complexing of copper in food extracts and predicted that all the copper absorbed by intestines will be complexed, after neutralization no free ions can be found (Dendougui & Schwedt, 2002). Although there have been some achievements in the changes of copper during the pH gradient in the GI tract, it is also difficult to deduction the solubility and complexation of copper in the digestive tract, especially for different types of food.
To gain a better understanding of the copper speciation in humans' GI tract, copper and some food components/digestion enzymes were added in the simulated gastric fluid (SGF). The electrode response (mV) against Cu 2+ , which is linearly related to the logarithm of the Cu 2+ concentration under standard conditions (Momin & Pillai, 2015;Silanikove et al., 2003), was tested at a pH of 2.0-7.5 by ISEs. The ratio of dissolved copper, which was defined as that which passes a membrane filter with 0.45 μm pore size (Stella & Ganzerli-Valentini, 1980;Stiff, 1971), was tested by ICP-OES.

| Apparatus
The pH measurements were done using Titrando 902 with a pH electrode (Metrohm, Swiss) and software tiamo 2.4. The NaOH solution was added automatically by 800 Dosino (Metrohm, Swiss) equipped with a 20-ml burette. The potential was measured by SevenExcellence TM (Mettler Tolido, Swiss) with software LabX direct pH 2.4, which can automatically record the potential at an interval of 1 s. A copper ion selective electrode (Bante, U.S.A.) was used to measure potential variation. A magnetic stirrer (IKA, Germany) and a constant temperature water bath apparatus (Guohua Electronic Appliance Co., Ltd., China) were used.
Copper concentration was determined by ICP OES (SPS8000 Bjhaiguang Co., Ltd,) under standard operating conditions.

| Preparation of the model solutions
Before experiment, the composition of simulated gastric fluid (SGF, 1.25×concentrates) was determined as previously described (Minekus et al., 2014), and copper stock solution of accurate concentration was prepared. In 100 ml ultrapure water, 4.69 g CuSO 4 .5H 2 O was dissolved, and the accurate concentration was tested by ICP OES.
Divided the Food/digestion components are into three groups.
Blank solution: Added 64-ml SGF solutions (1.25×concentrates) and 1.00-ml copper stock solution, after adjusting the pH to 2.0 with HCl solution, to make up the volume to 80.0-ml ultrapure water. The concentration of copper in the model solution was around 150 mg/L. KSLG groups: Added the reserve solution according to the above method, and then 31.25 mmol/L food components that were 10 times of copper were added.
CLG groups: Added the reserve solution according to the above method, and then 0.80 g protein or starch was added.
DLG groups: The reserve solution was added according to the above method, and then 0.256 g of pepsin/pancreatin was added in the solution. The concentration was according to the in vitro digestive model previously described (3.2 g/L) (Liu et al., 2012(Liu et al., , 2013. Add 5.5ml fresh bile to make the bile salt 10 mmol/L in the solution (Minekus et al., 2014). The concentration of the bile salt was measured using commercial assay kits (Jiancheng Institute, Nanjing, China).

| Potential measurements
The pH electrode and ISE were immersed in the solution before titration. After mixing for 5 min, sodium hydroxide solution was slowly added by Titrando to raise the pH from 2.0 to 7.5. The data (pH and electrode responses against Cu 2+ ) were automatically recording by tiamo 2.4. and LabX direct pH 2.4. During the titration process, a 37°C water bath, a constant electrodes distance, and stirring speed were offered. pH and copper ion selective electrodes were checked every day before the measurement, and the membrane was polished with aluminum hydroxide when necessary.

| Percentage of dissolved copper
The added volume was record by Titrando 902 and the solution was

| Statistical analysis
The experiments were performed in triplicate, and experimental data are expressed as the mean ± standard deviation (SD). Statistical significance was determined by one-way ANOVA using IBM SPSS Statistics version 22 (IBM Corporation, New York, U.S.A.). A probability of p < .05 was considered statistically significant.

| Dissolved copper and electrode response (mV) against Cu2+ in KSLG
It can be observed that some flocculent precipitates were appeared along with the raise of pH in the blank, fructose, glucose, and sucrose solutions. However, it has always been clear in organic acid and amino acid solutions. The results of Figure 1 and Table 1 showed that organic acids and amino acids model solutions were almost completely dissolved, while blank, fructose, glucose, and sucrose solutions rarely dissolved (the proportion of copper below 3%). The visible precipitate and low solubility of copper indicate the production of copper hydroxide precipitation in the blank, fructose, glucose, and sucrose solutions.  was slow after pH 6.0 for oxalic acid solution (56.7 ± 0.8 Mv at pH 6 and 54.8 ± 1.0 Mv at pH 7.5), while the potential variation was still obviously decreased for the EDTA solution (−51.6 ± 1.6 Mv at pH 6 and −84.5 ± 1.1 Mv at pH 7.5) and other solutions. The initial potential change indicates that oxalic acid and EDTA begin to form complex with copper at pH 2, with the continued increase in pH, the complexation of oxalic acid with copper reached saturation, while the complexation between EDTA and copper did not (Figure 1a).
Little copper existed in dissolved form (1.0 ± 0.3%) in starch model solution.
The ΔE pH 2.0-7.5 and various solubility indicate copper complex with these ingredients, and the complex had different molecular size Trance metals in aqueous solutions can exist as free (hydrated) ions, complexes with organic or inorganic ligands. Considering the solubility product constant of cupric hydroxide (K sp = 2.2 × 10 -20 ) (Speight, 2005), most of the copper may bind to hydroxyl ions to form cupric hydroxide in the blank solution with the decrease in hydrogen ions, which is hard to dissolved in water, and the concentration of copper ions in pure water should be 0.014 mg/L at pH 7.5 under standard conditions. However, the results of the filtered copper concentration in blank solution were 1.0 ± 0.2 mg/L in our study, which is much higher than 0.014 mg/L. As there were only some inorganic ions existed in the blank solution, these ions may combine with copper to form copper complexes, which will affect the solubility of copper. The stability constant (lgK) was 4.31 with ammonia, 0.1 with chloride, and 6.79 with pyrophosphate (Speight, 2005). In another hand, some small precipitates may also pass through the membrane to increase the concentration of filtered copper. Same as the blank solution, low solubility and similar potential changes occurred in the fructose, glucose, and sucrose solutions (Figure 1c). Consider the protons are difficult to dissociate from these components, they are not good ligands for copper. Along with the rise of pH, flocculent precipitate was produced and the dissolved ratio decreased. From the study of these solution, it can be seen that the food without sufficient ligand may not provide enough binding sites for copper during digestion. Hence, with the decrease in pH, excess copper will be combined with hydroxyl to form copper hydroxide precipitation.

| D ISCUSS I ON
However, although the dissolved copper concentration is low in these solutions, the electrical potential reflects that the content of copper ion in GI tract is much higher than other solution.

F I G U R E 4
Electrode response (mV) in CLG (a) and DLG (b) model solutions between pH 2.0 and 7.5, n = 3

F I G U R E 5 Fraction of all glycine species (a) and all oxalic acid species (b)
By contrast, N am , O COO , and nitrogen atoms of imidazole rings (N im ) show remarkable affinities for copper (Carrera et al., 2004;Kang et al., 1985;Manceau & Matynia, 2010 with histidine (Martell & Smith, 1974, 1989Speight, 2005). Hence, organic acids and amino acids are good ligands for copper. With the raise of pH, they dissociate protons and offer binding sit for copper, copper complex with them to form stable and soluble complex, and lead to low copper potential and high dissolved ratio. Take L-histidine. However, its 18.1 (CuL 2 ) with histidine (Speight, 2005;Martell & Smith, 1974;Martell & Smith, 1989). The different stability constant (lgK) of copper complexes indicates, although peptide and protein may offer lots of binding cites for copper, the molecular steric hindrance may hinder the binding of copper to ligands. Therefore, the complexation of protein with copper is also related to the molecular structure of protein. Fresh bile is complex mixture containing bilirubin, amino acids, bile salts, and other components (Hofmann & Mysels, 1992). Another important finding was that the content of copper ion is different in various types of food. The concentration of copper ions may be extremely low in the presence of organic matter, and high when the ligand is insufficient. However, trace amount of Cu 2+ has been detected in food samples, living zebrafish (Zhou et al., 2021), and AD rat brains (Yu et al., 2017) and significant toxicity was found.
The toxicity of copper ions has been well established by a lot of articles. Ocean acidification, which increases the concentration of Cu 2+ , significantly increases the toxicity responses to mussels and purple sea urchins (Lewis et al., 2016). The correlation between the free copper concentration and bacterial growth seems to be better than the correlation involving the total concentration when the free copper concentration is seven to eight orders of magnitude lower than the total copper concentration (Hasman et al., 2009 As various types of food may offer different content of copper ions in the digestive tract, we should pay attention to the toxicity of copper in different food. Besides, common food and drinking water are also a source of copper for human. With the widely use of copper plumbing, the concentration of copper in water may be high. It was found that the concentration of copper is 0.4-2.0 mg/L in copper pipes (Donohue et al., 2005;Lehtola et al., 2004;Pettersson & Rasmussen, 1999). As there are no organic ligands in drinking water, the concentration of copper ions in digestive tract may be high. In a study performed on rabbits, Sparks et al. found that the addition of trace amounts of copper (0.12 ppm) in water for 10 weeks induced cholesterol-fed rabbits' amyloid-beta (Aβ) accumulation and significantly hindered the ability of rabbits to learn a difficult trace conditioning task (Sparks & Schreurs, 2003). For mice, a concentration of 0.13 mg/L Cu 2+ in water increased Aβ production and neuroinflammation, increasing the severity of Alzheimer's disease (Singh et al., 2013). As more studies found the toxicity of copper in drinking water, the trace content of copper ions in digestive tract and their different toxicity should be future assessment.

| CON CLUS ION
As a transition metal element, copper is easy to bind to ligands because of its strong and nonspecific association with nearly all ligands, the coexisting components in the digestive tract can be seen as different ligands; the pH influences the acid-base dissociation of complexation group, which affects the concentration of the ligands; and the stability constants of copper complex decide the complexation strength with copper. In addition, molecule size and molecular steric also affect the interaction of copper and various ligands. As the environment in the digestive tract is complicated and changeable, it is hard to theoretically calculate or measure the accurate speciation of copper in digestive tract. However, different types of food will lead to different forms of copper distribution. The food rich in soluble ligands may lead to high soluble copper and then high bioavailability. And the food without ligands (water) may have high copper ions in digestive tract, which may lead to toxicity. As the widely use of copper plumbing, the safety of copper in drinking water should be widely concerned in the future.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.