Lead and cadmium biosorption from milk by Lactobacillus acidophilus ATCC 4356

Abstract The food and water contamination with heavy metals is increasing due to the environmental pollutions. Lead and cadmium are the toxic heavy metals for humans that can be found in air, soil, water, and even food. Lactic acid bacteria have the ability to remove and diminish the level of heavy metals. In this study, Lactobacillus acidophilus was used to remove lead and cadmium in milk and the capability of this valuable bacterium in biosorption of these metals low concentrations (µg/L or ppb) in milk was evaluated. First, the variables on lead and cadmium removal by this bacterium have been studied by Plackett–Burman design. Then, the bioremoval process was optimized and the three main factors, the bacterium concentration, contact time, and the initial heavy metal concentration were chosen by using a central composite design. The optimum lead and cadmium bioremoval yield of 80% and 75% were observed, respectively, at 1 × 1012 CFU of L. acidophilus in milk at the 4th day and the initial ion concentration of 100 µg/L. The 3D plots analysis showed the interaction effects on metal biosorption. This study showed that L. acidophilus is a natural effective biosorbent for lead and cadmium removal from milk.

Milk as the most common dairy product may be contaminated by various pollutants such as heavy metals, so it is one of the most serious problems for the public health in the world (Derakhshesh & Rahimi, 2012;Fischer, Schilter, Tritscher, & Stadler, 2015). As toxic metals' presence in milk is so important, thus their levels should be controlled continuously (Semaghiul, Simona, Gabriela, & Alina, 2008).
According to "Codex Standard for Contaminants and Toxins in Food," the permissible limits for Pb and Cd concentrations in milk should be <20 and 10 µg/L, respectively. Some studies have reported Pb and Cd contamination, respectively, more than these standard levels in milk around the world such as China 40 and 20 µg/L (Qin, Wang, Li, Tong, & Tong, 2009) There are some physiochemical ways for heavy metal detoxification such as ion exchange, chemical precipitation, and membrane technologies but they are not so useful for foodstuffs because they are very costly and not usable in foodstuffs (Kobielsk, Howarth, Farha, & Nayak, 2018;Zabochnicka-Åšwi & Krzywonos, 2014;Zhao et al., 2018).
Lactic acid bacteria have the ability to remove and diminish the level of heavy metals from aqueous solution (Halttunen, Finell, & Salminen, 2007;Zhai et al., 2015). They are in the list of GRAS (Chen & Narbad, 2018) and also able to bind heavy metals in low concentrations. The negative surface charge of the LABs helps them in binding to metal cations (Zoghi, Khosravi-Darani, & Sohrabvandi, 2014).
Potential of various LABs for the heavy metals bioremoval have been reported such as Pb, Cd removal by L. bulgaricus (Chang, Choi, & Kikuchi, 2012), and Cd by co-culture of L. plantarum and B. coagulans (Majlesi, Shekarforoush, Ghaisari, Nazifi, & Sajedianfard, 2017), as well as As removal by L. acidophilus (Singh & Sarma, 2010), Cd by L. plantarum (Hao, Reiske, & Wilson, 2000). Biosorption of Cd and Pb by lactic acid bacteria (Li et al., 2020) and Pb and Cd removal by lactic acid (Halttunen, Salminen, Jussi, Raija, & Kalle, 2008). The gap of research in all these reports seems to be the lack of considering heavy metals' removal in the range of very small amounts (ppb) in foodstuff and water.
In the previous projects of this experimental team, we used Saccharomyces cerevisiae for biosorption of heavy metals from drinking water (Hadiani, Khosravi-Darani, Rahimifard, & Younesi, 2018) and milk (Massoud, Hadiani, et al., 2019; and also the biosorption of mercury in milk by the help of L. acidophilus . In this study, Lactobacillus acidophilus ATCC 4356 was used to remove Pb and Cd in milk due to its popularity in dairy industry, the previous studies and the experiences of our research team, and the capability of this valuable bacterium in biosorption of these metals low concentrations (µg/L or ppb) in milk was evaluated. So, the 5 process variables (contact time, metal concentration, bacterial biomass, inoculation temperature, and shaking rate) were chosen for the Plackett-Burman design (PBD) and distinguishing of main factors. After that, the interaction of main variables (metal concentration, contact time, and biomass concentration) has been studied in 5 levels by response surface methodology (RSM) to reveal the optimum condition of bioremoval. Eventually, the biosorption isotherms (Langmuir and Freundlich models) were applied to clarify the ability of L. acidophilus (10 12 CFU/ml) for Pb and Cd bioremoval at different metal concentrations. Up to now, there is no published report about the biosorption of Pb and Cd by L. acidophilus in milk, our project would be the first step in using this profitable bacterium as the biosorbent in foodstuffs.

| Bacterial strain and chemicals
Lactobacillus acidophilus ATCC 4356, the commonly used LAB in dairy industry, was selected and purchased from Tak Gene Zist Company. The bacteria were inoculated in 10 ml of MRS broth and then incubated for 48 hr at 37°C. The bacteria's viability was determined by total plate counting and MRS agar and plate count agar for L. acidophilus counting (Vinderola & Reinheimer, 2000).
The standard solution of Pb and Cd (1,000 mg/L, Merck), MRS agar, MRS broth, and plate count agar were prepared from Liofilchem.

| Pb and Cd detection by ICP-MS
The inductively coupled plasma mass spectrometer (ICP-MS) (England) with a cross flow rate nebulizer and a Peltier-cooled quartz spray chamber was used in this study. The instrument was checked by the aqueous multielement before each experiment. First, the prepared samples were digested by the microwave with a MPR-600 rotor (at 35 bar and 260°C) (Khan et al., 2014;Li et al., 2018).

| Process variables
All possible process factors which may have an impact on the biosorption yield of Pb and Cd biosorption by L. acidophilus were selected based on literature reviews (Halttunen et al., 2008;Jadán-Piedra et al., 2017;Pakdel, Soleimanian-Zad, & Akbari-Alavijeh, 2018;Singh & Sarma, 2010) including the contact time, size of inoculation, metal concentration, temperature and shaking rate. Then, according to our pre-experiences, the process variables' interaction effects should be examined. The experimental planning by the help of a screening design was applied, and the significant factors are highlighted. Finally, the central composite design was used for the optimization design of the experiments (Coruh, Elevli, & Geyikçi, 2012).

| Plackett-Burman design (PBD)
According to the literature reviews and pre-experienced trials, the 5 independent variables in Pb and Cd biosorption by L. acidophilus were as follows: the contact time, Pb and Cd concentrations, shaking rate, biomass dosage, and inoculation temperature. Table 1 shows PBD for evaluation of these process variables in two experimental levels.
The aim of our study is to investigate the capability of L. acidophilus to eliminate the small amounts of Pb and Cd in milk. The heavy metal concentrations were chosen in very low levels as our teams' previous study on the drinking water. Then, the project was planned by Design Expert software to verify the maximum removal conditions.
For the designed experiment of Pb and Cd bioremoval in this study, the bacterial biomass (1 × 10 11 and 1 × 10 12 CFU) was added to sterile milk at 37°C; then, the samples were rested for 20 min and after that the metals (40 and 100 µg/L) were added and put at on the shaker. In the 1st and 4th day (as the defined contact time), the bacteria cells were centrifuged (at 8,000 × g for 20 min). Finally, the supernatants were analyzed for residual Pb and Cd concentrations by using ICP-MS. All the experiments were done in triplicates.
The ability of L. acidophilus to absorb Pb and Cd was evaluated as follows: where C 0 is the initial and C 1 is residual mercury concentration (Zhai et al., 2015).
The collected data were analyzed by Minitab statistical software (version 14). According to the results, 3 main variables were as follows: L. acidophilus concentration, the contact time, and metal concentration. For the experimental optimization of this study, the fractional factorial design was performed.

| Response surface methodology (RSM)
Response surface methodology is a mathematical technique that helps to analyze the effects of process variables on the responses (Aslan & Cebeci, 2007). The Plackett-Burman results showed that three process variables: metal concentrations, contact time, and L. acidophilus concentration had significant effects on Pb and Cd bioremoval. RSM was designed for optimizing the variable levels for Pb and Cd bioremoval and also the minimizing the number of tests.
In this project, CCD was applied to investigate the optimum biosorption conditions by the help of the experimental factors shown in Table 2.
The Design Expert 7.1.5 (Stat-Ease Inc.) software was applied for designing the run tests and also data analysis. The runs were designed in the 5 levels (Table 2) while the other factors were kept constant as the following: the shaking rate at 50 rpm and the inoculation temperature at 25°C.

| Determination of the binding metal capacity
The isotherm models like Langmuir and Freundlich would assist in predicting the maximum ability of binding heavy metals. These absorption models were applied to explain capability of the bacterial cells to absorb Cd and Pb at a predetermined time (Andreasen et al., 2018;Freundlich, 1906;Langmuir, 1918).

| Statistical analysis
The statistical analysis was analyzed by MINITAB statistical software (version 14), and then, the response surface plots were designed. The statistical data were prepared by analysis of variance. All

| Optimization of Pb and Cd bioremoval by RSM
The analysis of variance represented the effects of the process variables that designed by Plackett-Burman design. Applying RSM after the variance analysis showed that the Pb and Cd biosorption levels are the result of the 3 main variables as shown in

| Effect of contact time and Lactobacillus acidophilus concentration
In this study, the ability of L. acidophilus concentrations (range of 10 10 -10 13 CFU) on Pb and Cd bioremoval efficiency during the contact times (1-4 days) was evaluated. Figure 1a shows the Pb and Figure 1b Cd absorption by L. acidophilus at different contact times. As Figure 1a and b shows, the maximum binding rate of Pb and Cd occurred in the 4th day, respectively. The removal efficiency of these heavy metals first increased by rising the bacterial biomass and contact time and reached to the maximum level, then by the further increase of bacterial biomass, a light decrease in removal levels occurred.
In general, heavy metal biosorption mechanism is complicated and metal binding happens through the ion exchange in cell walls' peptidoglycan and teichoic acid, the precipitation or the ligand formation (Zoghi et al., 2014). The cell walls of L. acidophilus contain a thick layer of teichoic acid, peptidoglycan, and exopolysaccharides.
The functional groups such as hydroxyl, carboxyl, and phosphate create the negative charges. So, the bacteria would be able to absorb the heavy metals' cations (Halttunen et al., 2007;Wang et al., 2015;Wang, Wang, Cheng, Bian, & Guo, 2017). It has been reported that metal binding occurs on the bacteria cell surface with no energy consumption (Halttunen et al., 2008).
In this study, the highest Pb and Cd removal efficiency was 80% and 72%, respectively, in biomass of 1 × 10 12 CFU in the 3rd day.
The difference in sorption of these two metals could be described as their ionic radius is different (ionic radius of Cd < Pb) (Bekri- , Bayoudh, & Baklou, 2006). Similar studies reported that LAB showed the same results for increasing biosorption by passing the time (Halttunen et al., 2007;Rayes, 2012).

TA B L E 3 Analysis of variance of parameters for Pb (a) and Cd (b) bioremoval by Lactobacillus acidophilus
As shown in Figure 1 a

| Effect of metal concentration and Lactobacillus acidophilus concentration
The effect of initial metal concentration and L. acidophilus concentration on the bioremoval in the range of 10 10 -10 13 CFU and 40-100 µg/L was investigated (Figure 2a and b). The results represented that by rising the metal concentration, the absorption increased. As shown in Figure 2a and b, by increasing L. acidophilus concentration up to 1 × 10 12 CFU, the removal efficiency enhanced. The maximum Pb and Cd removal efficiency, 75% and 72%, respectively, was observed at the concentration of 100 µg/L and the biomass concentration of 1 × 10 12 CFU.
Lactobacillus acidophilus has been reported to have a high affiliation for biosorption of heavy metals (Bhakta et al., 2012;Kinoshita et al., 2013). Heavy metals' bioremoval is a surface process due to the binding of metal cations to the anionic functional groups and it also depends on the capacity of the bacteria strains and the metal electronegativity (Kang, Kwon, & So, 2016). LAB have some polymers like lipoteichoic acid that can be responsible for the interactions between the heavy metals and the negative charge of bacteria surface Zoghi et al., 2014).
In this study, it is observed that by rising the metal concentration, the absorption to the bacteria receptors would also increase, which results in the higher biosorption levels (Dobrowolski, Szcześ, Czemierska, & Jarosz-Wikołazka, 2017;Halttunen et al., 2007).
According to our findings, Pb and Cd biosorption efficiency enhanced by increasing the metal concentration (from 40 to 100 µg/L).
The important factors in bioremoval as shown in Figure 2a

| Isotherm studies
The ability of L. acidophilus (10 12 CFU/ml) for Pb and Cd bioremoval was determined at different metals initial concentrations One of the most common models that used in scientific studies is Langmuir model. The Langmuir equation is correct for monolayer absorption by the following equation (Langmuir, 1918): where C e (µg/L) is the equilibrium metal concentration in milk, Q e (µg/L) is the amount of metal in absorbing process, Q max (µg/L) is the maximum metal absorption in high C e level, and K L (L/µg) is the Langmuir constant. The C e /Q e versus C e indicates a straight line of slope 1/Q max and also intercept of 1/K L Q max .
The Freundlich equation is using the following equation (Freundlich, 1906): where the parameters n and K f are the Freundlich constants. The n and K f are defined from the linear plot of ln Q e versus ln C e . Freundlich equation varies with the materials heterogeneity.
As shown in Figures 3a,b and 4a

| CON CLUS ION
Nowadays, heavy metals' presence in food and water is a serious health concern all around the world. Pb and Cd are two toxic heavy metals for people. Milk is the most used dairy products that should be safe for consumption. In this project, Pb and Cd were removed from L. acidophilus successfully. The biosorption increased by increasing the bacteria and metal concentrations as well as the contact time. This study revealed that L. acidophilus is a natural effective biosorbent for Pb and Cd removal in very low concentration levels (ppb) from milk. Our findings are as the first step of investigating the ability of heavy metals binding by LABs in milk.
There is still a need for more studies around other LAB strains in milk and other foods to eliminate the harmful effects of the toxic heavy metals.

ACK N OWLED G M ENTS
We kindly thank Dr. Younesi for providing the L. acidophilus ATCC 4356.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

O RCI D
Kianoush Khosravi-Darani https://orcid. (2) C e ∕Q e = 1∕ K * Q max + C e ∕Q max (3) Ln Q e = Ln K f + 1∕n Ln C e F I G U R E 3 Langmuir absorption isotherm curve for Pb (a) and Cd (b)