Introduction

Wastewaters containing Cr(VI) are generated in many industrial processes, including leather-tanning, chrome-plating, stainless steel-welding, pigment production and nuclear weapons generation (Gonzalez et al. 2003). However, due to leakage, poor storage and improper disposal, Cr(VI) has become one of the most frequently detected contaminant at the waste sites (Thacker et al. 2006). Cr(VI) compounds are in general more toxic than Cr(III) compounds due to their stronger oxidizing power (Cervantes et al. 2001; Katz and Salem 1994). From a biological perspective, Cr(III) compounds have very low solubility, thus, restricting their spread and their biological availability (Gonzalez et al. 2003).

Detoxification of Cr(VI) via reduction of Cr(VI) to Cr(III) can be carried out with chemical or biological methods. Conventional methods for the treatment of chromate include chemical reduction using a reducing agent such as sodium sulfite and adsorption on ion exchange or chelating resins. However, these methods consume high amounts of energy and large quantities of chemical reagents which is not economically feasible. Furthermore, the resultant metal-containing chemical sludge can be a potential source of metal pollution. On the other hand, biological methods such as microbiological detoxification of polluted water are economical, safe, and sustainable (Eccles 1995; Shakoori et al. 2000).

A number of chromium resistant bacteria, i.e., Bacillus sp., Leucobacter sp., Exiguobacterium sp. (Sarangi and Krishnan 2008), Pseudomonas sp. (Desai et al. 2008), Brucella sp. (Thacker et al. 2007), Shewanella oneidensis (Daulton et al. 2007) and Achromobacter sp. (Zhu et al. 2008) have reported to reduce Cr(VI). Bacterial reduction of Cr(VI) can be considered as a mechanism of resistance to Cr(VI) (Ohtake et al. 1987). Cr(VI) resistance mechanisms include periplasmic biosorption, intracellular bioaccumulation and biotransformation either via direct enzymatic reaction or indirectly with metabolites (Camargo et al. 2005).

Recently, the potential of Acinetobacter sp. as a Cr(VI) remediation agent has been demonstrated (Srivastava and Thakur 2007; Zakaria et al. 2007). Srivastava and Thakur (2007) reported removal of more than 75% chromium by Acinetobacter sp. after 7 days incubation in a medium containing sodium acetate (0.2% w/v), sodium nitrate (0.1% w/v) and pH 7 (optimized carbon, nitrogen sources and pH). The findings by Zakaria et al. (2007) also gave early indication for a potential industrial application of A. haemolyticus. However, exploring the metal-microbe interaction is a prerequisite for designing effective and efficient remediation strategies and biotechnology treatment facilities (Boyanov et al. 2003).

The present work elucidates the mechanisms of Cr(VI) resistance and removal by Acinetobacter haemolyticus via instrumental analysis, hence its novelty. Preliminary studies were carried out to evaluate the Cr(VI) tolerance and reduction capacity of the isolate. FTIR was used to determine the functional groups involved in the interaction with chromium. FESEM-EDX and TEM analyses were carried out to determine the localization of chromium. The Cr(VI) reductase activity in a soluble crude cell-free extract (CFE) of the bacteria was also investigated. These results provide a greater understanding of Cr(VI) resistance mechanisms by Acinetobacter haemolyticus and aid in the further development of the microorganism for bioremediation of chromium from industrial wastewater.

Materials and methods

Bacterial strain

Acinetobacter haemolyticus (A. haemolyticus), GenBank Acc. No. EF369508, was isolated from Cr(VI)-containing textile dye effluent. This strain has been demonstrated as being able to reduce Cr(VI) by Zakaria et al. (2007).

Growth media

Three growth media were used in this study (in 1 l of distilled deionized water, DDI): LB broth [10 g of tryptone (Oxoid), 5 g of yeast extract (Merck) and 10 g of NaCl (Riedel-de-Haën)], LB agar [10 g of tryptone, 5 g of yeast extract, 10 g of NaCl and 13 g of agar (Merck)] and MS broth [0.03 g NH4C1 (Riedel-de-Haën), 0.03 g K2HPO4 (BDH), 0.05 g KH2PO4 (Univar), 0.01 g NaC1 and 0.01 g MgSO4 · 7H20 (Merck) (Wang and Xiao 1995)]. All media were autoclaved at 120°C for 15 min and stored at room temperature prior to use.

Determination of Cr(VI) tolerance

The Cr(VI) tolerance of A. haemolyticus was evaluated using repli-plate, agar and shake-flask techniques. LB broth was dispensed into the repli-plate dish (Sterilin, USA) followed by Cr(VI) from a stock solution (1,000 mg l−1) to give a Cr(VI) concentration between 5 and 150 mg l−1. A 5% (v/v) inoculum of mid-exponential phase cultures was then added to the repli-plate and incubated at 30°C for 7 days (Memmert, USA). Bacterial growth was measured at OD600 (Genesys 20 Thermospectronic, USA). LB broth with Cr(VI) but without bacterial cell served as control. Cr(VI) tolerance was also evaluated in LB agar where freshly prepared LB agar supplemented with 30–60 mg Cr(VI) l−1 was inoculated with 0.1 ml of a mid-exponential phase culture. Growth on LB agar plates without Cr(VI) served as the control. The cultures were incubated at 30°C for 24 h. Cr(VI) tolerance was measured as the number of colonies formed after the incubation period. For the shake-flask technique, overnight cultures of A. haemolyticus (10%, v/v) were inoculated into 2 l conical flask containing 200 ml LB broth added with 30 and 60 mg Cr(VI) l−1, respectively, from a stock solution. The samples were incubated at 30°C, 200 rpm for 24 h. Three millilitre aliquots were withdrawn at regular time intervals and analyzed for growth at OD 600 nm. A similar experimental setup without bacterial cells served as a control. Stock chromium solutions (1,000 mg l−1) were prepared as follows; 2.829 g K2Cr2O7 (294.18 g mol−1, BDH-GPR™) and 5.124 g CrCl3·6H2O (266.45 g mol−1, Fluka), respectively, were dissolved in 1 l of DDI water. The solutions were filter-sterilized using a 0.2 μm Whatman filter paper.

Total chromium analysis

Total chromium concentrations were determined using atomic absorption spectrophotometry (AAS, Perkin–Elmer AAnalyst 400) at λ = 357.9 nm with a detection limit of 0.1 mg l−1. The Cr(III) content in liquid solution was obtained by subtracting the content of Cr(VI) from that of total chromium.

Interaction between A. haemolyticus-Cr(VI) during Cr(VI) reduction

FESEM-EDX and TEM analysis

Cells of A. haemolyticus were incubated for 24 h on LB agar plates containing 30 mg Cr(VI) l−1. The bacterial colonies formed were then pooled and fixed in 2.5% (v/v) glutaraldehyde (Sigma, USA) in DDI water for 1–2 h at room temperature. Fixed cells were washed twice with DDI water, and post-fixed with 2% (v/v) osmium tetroxide (Fluka, Switzerland) for 1 h. The pellets were subsequently dehydrated using 10–100% (v/v) ethanol for 5 min each and left to dry overnight in a desiccator. The specimens were mounted onto the sample holder with carbon-conductive adhesive tapes and coated with platinum using a sputter coater (Auto fine coater JFC-1600, JEOL) prior to viewing using a field-emission scanning electron microscope (FESEM JSN-670 1F, JEOL) fitted with EDX analysis.

A similar experimental setup was used for TEM analysis followed by 2 min washes in 100% (v/v) acetone after ethanol washing. The cell pellets were then embedded in 50 and 100% (v/v) of epoxy resin (Ted-Pella, California) in acetone for 15 min each. Then, the cell pellets were infiltrated a second time with fresh 100% (v/v) epoxy resin and cured at 60°C overnight. Specimens of 90 nm thickness were sectioned from the embedded blocks using a Leica UltraCut UCT ultramicrotome and mounted on 200-mesh copper TEM grids. The specimens were stained with uranyl acetate (Ted-Pella, California) and post-stained with lead citrate (Ted-Pella, California) for 5 min each. The samples were viewed using a transmission electron microscope (Tecnai G2, Philips), operated at 80 kV.

FTIR

Infrared spectra for A. haemolyticus cells grown for 24 h in LB broth supplemented with (30, 60, and 100 mg Cr(VI) l−1) and without Cr(VI) for 24 h were obtained using a Fourier Transform Infrared Spectrometer (FTIR 1600, Perkin–Elmer). Bacterial cell suspensions were pelleted by centrifugation at 7,000 rpm, 4°C for 30 min. The supernatant obtained was discarded while the pellet was washed in subsequent changes of 0.85% (v/v) saline and DDI water prior to air-drying at 50°C for 8 h (Kamnev et al. 1997). The dried pellet was ground with 200 mg of KBr (spectroscopic grade) in a mortar before pressed into 10 mm diameter disks under 6 tones of pressure before FTIR analysis. The analysis conditions used were 16 scans at a resolution of 4 cm−1 measured between 400–4,000 cm−1.

Crude cell-free extract Cr(VI) reductase test

Preparation of crude CFE

The crude CFE was prepared based on the methods by Thacker et al. (2006, 2007). Briefly, a mid-exponential phase culture, grown in 200 ml LB broth, was harvested by centrifugation at 6,000 g for 20 min at 4°C, washed twice with 20 ml of 10 mM Tris–HCl buffer, pH 7.2 and resuspended in 30 ml of the same buffer. Cells were disrupted by sonication (VibraCell, Sonic and Materials Inc.) for 20 min in cold condition. The resultant homogenate was centrifuged at 8,000 g for 30 min at 4°C and the supernatant (CFE) was used for chromate reductase assays. Protein estimation was carried out using the method by Lowry et al. (1951). The supernatant obtained after harvesting the cells was also filter sterilized and used for chromate reductase assay.

Cr(VI) reductase assay

The methods by McLean and Beveridge (2001) and Pal and Paul (2004) were adopted for Cr(VI) reductase assays. Cr(VI) reduction assays using the crude CFE or supernatant (10 ml) were conducted at 30°C with agitation at 120 rpm in Erlenmeyer flasks containing 5 mg Cr(VI) l−1 under aerobic condition. LB broth without Cr(VI) and autoclaved crude CFE served as controls. Each treatment was done in duplicate and samples were collected at regular time interval for 8 h. Reactions were stopped by heating the samples in hot water bath at 80°C for 1 h immediately after collection. Cr(VI) was determined according to the method described by Thacker et al. (2006).

Results and discussion

Growth and Cr(VI) tolerance of A. haemolyticus

Growth of A. haemolyticus in LB broth was not affected at 30 and 60 mg Cr(VI) l−1, respectively. However, the growth rate decreased from 0.33 (in the absence of Cr(VI) to 0.18 h−1 after the addition of 60 mg Cr(VI) l−1. A. haemolyticus was able to tolerate up to 90 mg Cr(VI) l−1 in LB broth based on a minimum growth percentage of 80%. However, the bacterial growth percentage dropped to 48% at 110 mg Cr(VI) l−1 (Fig. 1) indicating the apparent Cr(VI) toxicity.

Fig. 1
figure 1

Effect of Cr(VI) on growth of A. haemolyticus in LB broth

Full size image

One possible mechanism for Cr(VI) toxicity is the alteration of genetic material and altered metabolic and physiological reactions of the bacteria (Losi et al. 1994). However, the use of the repli-plate technique may have contributed to the low Cr(VI) tolerance level due to limited O2 diffusion, hence would ultimately affect the metabolic activity of A. haemolyticus, which is an obligate aerobe (Zakaria et al. 2007). A. haemolyticus exhibited a significantly lower Cr(VI) tolerance in LB agar than in LB broth. A high bacterial count (4.6 × 107 CFU ml−1) was determined at concentrations up to 30 mg Cr(VI) l−1. However, growth was completely inhibited at 40 mg Cr(VI) l−1. A lower Cr(VI) tolerance for A. haemolyticus in solid media may be due to diffusion, complexation and availability of metals (Hassen et al. 1998). In a study by Megharaj et al. (2003), Bacillus sp. and Arthrobacter sp. grew in tryptic soy broth or minimal salts medium-glucose agar added with 100 mg Cr(VI) l−1. Although both bacterial strains exhibited similar resistance on agar media supplemented with Cr(VI), Bacillus sp. did not grow (bacteriostatic) in liquid medium supplemented with 100 mg Cr(VI) l−1. The use of nutrient-rich media may have affected the actual Cr(VI) toxicity via a ‘masking’ effect (Desai et al. 2008; Caravelli et al. 2008). Chelation to various components of the media, such as tryptone and yeast extract, together with the formation of complexes may cause a reduction in the availabilities of free Cr(VI) ions. Of particular importance is the sorption or chelation of metals to unspecified organic compounds found in most biological media (Angle and Chaney 1989).

FESEM-EDX

FESEM micrographs for A. haemolyticus grown with and without Cr(VI) on LB agar plates are presented in Fig. 2. A. haemolyticus showed the following morphology when grown in the absence of Cr(VI); coccobacilli, rough surface and an average cell diameter of 0.577 μm (n = 3) (Figs. 2a, b). However, upon cultivation in 30 mg Cr(VI) l−1 for 24 h, the cells lost their shape and became irregular with appearance of wrinkles on the surface (Figs. 2c, d). The average cell diameter also increased to 0.704 μm (n = 3). Enlargement or elongation of cells were also observed for other bacteria exposed to Cr(VI). Non-adapted E. coli K-12 cells exposed to 250 M potassium chromate (K2CrO4) exhibited extreme filamentous morphology within 3 h of Cr(VI) challenge (Ackerley et al. 2006). Interestingly, numerous pilus-like structures were also observed on A. haemolyticus upon exposure to Cr(VI). A similar observation was reported for Shewanella oneidensis MR-1 where the structures appeared during the transition of growth from 3 to 22°C. these structures were proposed to be sex pili, involved in genetic or nutrients exchange (Abboud et al. 2005).

Fig. 2
figure 2

FESEM micrographs of A. haemolyticus cells grown on LB agar in the absence of Cr(VI) a, b and presence of 30 mg Cr(VI) l−1; bar represents 1 μm

Full size image

A ‘wrinkled’ cell surface similar to that of A. haemolyticus upon growing on agar plates containing Cr(VI) has also been reported for another Gram negative bacterium, namely Arthrobacter K-2 (Lin et al. 2006). Acinetobacter sp. was also reported to undergo morphological changes upon contact with Cr(VI) such as the appearance of distinct ridges in the cell wall region (Srivastava and Thakur 2007). However, in this study, no microstructures resembling chromium precipitates were observed on the cell surface of A. haemolyticus implying that chromium was most likely adsorbed onto the surface of A. haemolyticus instead of forming precipitates. The adsorbed chromium is anticipated to be in the Cr(III) form due to the inability of chromate anions to bind with the electronegative functional groups commonly found on gram-negative envelopes (McLean and Beveridge 2001). From the EDX analysis (Fig. 3), small chromium peaks (0.19% wt. analysis) were observed for cells grown with Cr(VI) as opposed to those without Cr(VI). The main elements that co-localized with chromium were typically carbon, oxygen, nitrogen, phosphorus, and sulphur (Niftrik et al. 2008).

Fig. 3
figure 3

EDX analysis of A. haemolyticus cells grown on LB agar without Cr(VI) (a) and with 30 mg Cr(VI) l−1 (b); Inset FESEM of the samples analyzed with EDX probe

Full size image

TEM

To locate the intracellular accumulation of chromium, TEM analysis was carried out. A. haemolyticus grown on LB agar with Cr(VI) revealed the presence of electron opaque particles that were located mostly in the cytoplasmic region with minimum precipitation on the cell envelopes. This pattern was not found in thin sections of A. haemolyticus grown without Cr(VI) (Fig. 4). A similar finding was also reported for Acinetobacter sp. strain PCP3 by Srivastava and Thakur (2007).

Fig. 4
figure 4

TEM images of thin sections of A. haemolyticus cells grown without Cr(VI) (a, b) and with 30 mg Cr(VI) l−1 (c, d). Arrows in (c, d) indicates electron-opaque particles; the bar represents 1 μm (a, c) and 0.1 μm (b, d), respectively

Full size image

It is tempting to speculate that the electron-opaque particles are attributable to chromium precipitations. We propose that upon entering the cell, Cr(VI) will be reduced to Cr(III) due to the reducing environment and enzymes present inside the cell (Srivastava and Thakur 2007). Subsequently, Cr(III) formed would be free to bind to ionic sites and, once bound, will act as a template for further heterogeneous nucleation and crystal growth. However, long term accumulation of chromium by bacteria can lead to deleterious effect on growth as reported for Shewanella sp. strain MR-4 (Bencheikh-Latmani et al. 2007). Another factor affecting chromium accumulation by A. haemolyticus is the type of growth medium used. A higher chromium accumulation was observed when solid medium was used (as in this study) compared to the use of liquid medium such as nutrient broth. It was noted that in solid medium, the electron-opaque particles were denser (indicating higher chromium accumulation) as compared to the formation of fine particles when A. haemolyticus was grown in nutrient broth with Cr(VI) (Zakaria et al. 2007), which suggested a higher chromium uptake by cells in the solid medium.

FTIR

The FTIR spectra of A. haemolyticus grown in LB broth with and without Cr(VI is shown in Fig. 5.

Fig. 5
figure 5

FTIR spectra of A. haemolyticus cells grown a without Cr(VI) (control) and with b 30 c 60 d 100 mg Cr(VI) l−1

Full size image

FTIR spectra of A. haemolyticus grown in LB broth without Cr(VI) suggests the presence of amino, carboxyl, hydroxyl and sulphonate groups with the following band characteristics; 3,400–3,290 cm−1 (broad) represents the –OH and –NH stretching groups, most probably from glucose and proteins, 1,540–1,640 cm−1—primary and secondary amides from the peptide bonds which correspond to –NH bending, 1,100–1,000 cm−1 due to the C–O bond, which is the characteristic peak for polysaccharides, 800–850 cm−1—sulphonate group on the cell surface (Das and Guha 2007). Phosphate functional groups such as P–O, orthophosphate (PO4 3−), and P–OH have characteristic absorption peaks at 1,150, 1,100–1,030, and 1,040–910 cm−1, respectively. However, the peaks occur around the same position as C–N stretching which is between 1,350–1,000 cm−1 (Mungasavalli et al. 2007). When A. haemolyticus was grown in 30, 60, and 100 mg Cr(VI) l−1 (Fig. 5b–d), changes were observed in the region of 1,655–750 and 3,450–2,800 cm−1. These changes indicate the metal binding process taking place on the surface of the cells involving certain functional groups (Bueno et al. 2008) such as amino, carboxyl and hydroxyl groups (Table 1). Similar findings were reported for chromium-treated cyanobacteria where significant alterations were observed for the characteristic peaks of COOH− groups. This observation implies chromium complexation with protein molecules (Pandi et al. 2007; Mungasavalli et al. 2007).

Table 1 FTIR analysis of A. haemolyticus grown with and without Cr(VI)
Full size table

Crude Cr(VI) reductase test

The specific activity for crude CFE of A. haemolyticus was 0.52 μg Cr(VI) reduced per mg of protein an hour at pH 7.2 and 37°C. Percentage of Cr(VI) reduction for both the supernatants and crude CFE (relative to LB broth and autoclaved crude CFE) is shown in Table 2.

Table 2 Percentage reduction of 5 mg Cr(VI) l−1 in supernatant and crude CFE of A. haemolyticus
Full size table

The autoclaved crude CFE and the bacterial growth supernatant showed negligible Cr(VI) reductase activity, indicating that that Cr(VI) reductase activity is not associated extracellularly but with the soluble fraction of the cells. The enzyme responsible could be either cytoplasmic or periplasmic-based (Thacker et al. 2006; McLean and Beveridge 2001). The specific activity of the cell fractions in other Cr(VI)-reducing bacteria is shown in Table 3.

Table 3 Specific activity of cell fractions in Cr(VI)-reducing bacteria
Full size table

Aerobic Cr(VI) reduction is commonly associated with soluble chromate reductases that uses external electron donors such as NADH, NADPH or lactate (Cervantes and Campos-Garcia 2007). NADH, that can also act as cofactors, was also reported to increase the Cr(VI) reduction activity by CFE for a number of bacteria such as Providencia sp. (Thacker et al. 2006), Pseudomonas ambigua G-1 (Suzuki et al. 1992), and E. coli ATCC 33456 (Shen and Wang 1993). However, some strains such as Cellulomonas sp. ES6 showed Cr(VI) reduction activity without external electron donors, suggesting endogenous electron donors for Cr(VI) reduction (Viamajala et al. 2007). Since the CFE of A. haemolyticus also showed substantial Cr(VI) reduction without the addition of external electron donors, it is proposed that A. haemolyticus can also use endogenous electron donors for Cr(VI) reduction.

Conclusion

This study demonstrated that A. haemolyticus, which showed Cr(VI) tolerance and reducing ability, may be used as a potential agent to decontaminate Cr(VI) contamination in wastewater systems. This approach looks promising based on our analysis using FTIR, SEM-EDX, TEM, and Cr(VI) reductase assay where chromium species were demonstrated to interact with the bacterium during the Cr(VI) reduction assay. The Cr(VI) reductase activity was associated with the intracellular soluble fraction of the cells.