1DEPARTMENT OF PURE AND INDUSTRIAL CHEMISTRY,

Decolourisation of azo-based food colourants (Carmoisine and Ponceau 4R) by Escherichia coli isolated from human feaces was investigated. Decolourisation was assessed by changes in optical density of culture supernatant and total viable count in media containing the colourants as sole source of carbon and energy. Decrease in optical density was accompanied by increase in total viable count during incubation. Percentage decolourisation of carmoisine was 80% and 75% under aerobic and anaerobic cultivation respectively. The values in Ponceau 4R containing medium were 62.5% and 60% under aerobic and anaerobic cultivation respectively. The potential health risk associated with decolourisation intermediates (aromatic amines) is discussed.

Key words: Food Colourants, Decolourisation, Escherichia coli.

Azo dyes are the largest class of synthetic dyes with the greatest variety of colours. At least 3,000 different varieties of azo dyes are extensively used in the textile, paper, food, cosmetics and pharmaceutical industries (Chen et al., 1999). Azo dyes are largely non-toxic, however, decolourisation of azo dyes by reductive cleavage of the azo bond (chromophore) produces aromatic amines as intermediates. Aromatic amines are toxic, mutagenic and carcinogenic to humans and other animals (Cartwright, 1983; IARC, 1982; Brown and Devito, 1993; Levine, 1991; Ganesh, 1992).

Cleavage of azo bond is mediated by azoreductases. Azoreductases are synthesized by microorganisms (Chen et al., 1999; Gary et al., 1994; Chinwetkitvanich et al., 2000; Rasso-Flores et al., 1997; Brown and Labourer, 1983). Degradation of azo dyes by human intestinal microflora has been reported (Fatemeh et al., 1990 and King-Thom et al., 1978). These studies were carried out in complex media involving consortium of bacteria. The specific organism of the intestinal macroflora participating in azo dye reduction is poorly understood.

Most azo dyes are used as food colourants. The aims of this study were to isolate, identify some intestinal bacteria and assess the decolourisation potential of azo-based food colourants commonly used in Nigeria by the isolates. This study focused on Escherichia coli and food colourants (Carmoisine and Ponceau 4R).

One gram of freshly voided human feaces was sterially diluted (10^-1 – 10^-3) in physiological saline A loopful of each dilution was inoculated by streaking onto duplicate set of MacConkey agar plate. The plates were incubated at 37^oC for 24 – 48h.

Pink colonies, which developed were picked and inoculated into 10ml MacConkey broth and incubated at 37^oC for 24h. Duplicate set of Eosine

CARMOISINE

Methylene blue agar (EMB) plates were each inoculated with a loopful of the culture. Incubation was at 37^oC for 24 – 48h. Blue/black colonies surrounded by metallic sheen were picked and purified by repeated subculture and Gram stained. Colonies which were distinctly Gram negative and rod-shaped were used for various identification tests.

Various tests (motility, presence/absence of spore, urease, citrate utilization, hydrogen sulphide production, MRVP, catalase, oxidase, indole, starch hydrolysis and fermentation of various sugars) were carried out according to the methods of Cruickshank et al. (1980) and Holts et al. (1994). Based on the results of these various tests, and with reference to Bergey’s Manual of Determinative Bacteriology, the isolate was identified as Escherichia coli (E. coli). Stock cultures were maintained on EMB slants in a refrigerator at 4^oC.

Colonies were picked from the slants and inoculated into 10ml MacConkey broth. Incubation was at 37^oC for 24h.

BROTH CULTURE:

Duplicate set of 250ml Erlenmeyer flasks each containing 25ml of either MSC or MSP were inoculated with 1ml of a 24h MacConkey broth culture. Two sets of controls were also set up: positive control consisted of duplicate flasks inoculated with 1ml of heat-treated cells and negative controls consisted of uninoculated flasks.

The flasks were incubated at 37^oC with shaking and observed daily for decolourisation of medium (evidence of degradation) and increase in turbidity (evidence of growth). Increased turbidity and decolourisation of medium were observed in all the flasks which were inoculated with viable cells from the second day of incubation. These changes reached their peak on the fifth day. There was neither decolourisation nor turbidity in the control flasks.

A loopful of the standard inoculum was streaked onto duplicate set of either MSC or MSP agar plates. The plates were incubated at 37^oC and observed daily for appearance of zones of clearing (decrease in colour intensity) around the colonies. Controls consisting of duplicate set of uninoculated MSC or MSP agar plates were also set up. Zones of clearing were observed in all the inoculated plates after four days of incubation. The intensity of colour of the medium remained the same in the control plates. The results obtained in the screening tests showed that the organism degraded the dyes.

CARMOISINE

Triplicate set of 250ml Erlenmeyer flasks each containing 95ml of MSC broth was inoculated with 5ml of standard inoculum (ca. 10⁶ cfu ml^-1). The flasks were incubated at 37^oC with shaking. Controls consisted of triplicate set of uninoculated medium. Growth and decolourisation were monitored by determination of total viable count (TVC) and decrease in optical density (O.D.) of the culture supernatant respectively. Changes in pH were also monitored.

The set up was same as for Carmoisine except that the medium was MSP broth.

One millimeter was aseptically withdrawn at regular intervals from each flask and serially diluted (10^-1 to 10^-6) in 9ml physiological saline. One millimeter of appropriate dilution was inoculated into duplicate set of nutrient agar plates by the pour plate method. Plates were incubated at 37^oC for 24 – 48h and the number of colonies which developed counted. Results are expressed as colony-forming-unit per ml (CFU ml^-1).

Three loopfuls of standard inoculum (ca.1.2 x 10⁶ CFU ml^-1) was inoculated into either 10ml MSC of MSP medium contained in triplicate set of test tubes. The tubes were immediately overlaid with sterile vaseline. Incubation was at 37^oC for 7days in an Anaerobic jar.

After incubation the broth was centrifuged at 6,000 rpm for 30 min. The optical density of the supernatant was determined spectrophotometrically (Jenway 3100). The pH was also determined.

The pH of the above supernatant was determined using pH meter (Mettler Delta 3400).

Tables 1 and 2 shows the results obtained on decolourising potential of E. coli. Decolourisation (decrease in colour intensity of the medium) was noticeabe from the second day of incubation in broth culture. On the sixth day decolourisation was intense in medium which contained Carmoisine and moderate in Ponceau 4R medium.

On solid medium decolourisation was slower compared with the broth becoming noticeable on the third day. Decolourisation of azo dyes is attributed to reductive cleavage of azo bond (– N = N –). No decolourisation was observed in the control flasks and plates. The results show that decolourisation was due to metabolic activity of viable and growing cells

Table 1. Screening of E. coli for decolourisation of dyes in broth culture.

Bacterial have been reported to decolourise azo dyes: Bacillus subtilis (Ziss & Lyberotos, 1998); Proteus mirabilis (Chen et al., 1999); Pseudomonas sp. (Oranusi & Ogugbue, 2002) and Cellulomonas sp. (Oranusi & Ogugbue, 2003).

Figs. 2 & 3 depict the growth profile and changes in optical density (OD) of culture supernatant (decolourisation) when E. coli was cultured on MSC and MSP media respectively. In both medium increase in total viable count (TVC) was accompanied by concomitant decrease in optical density of culture supernatant over time.

In MSC (Fig. 2) TVC was 1.50 x 10⁶ CFU ml^-1 on first day and increased steadily to a maximum population density of 6.5 x 10⁹ CFU ml^-1 on the seventh day. The culture then remained fairly constant. The OD of culture supernatant decreased from initial value of 0.5 absorbancy unit to 0.10 absorbancy unit (80% decolourisation) on the seventh day.

In MSP medium (Fig. 3) the population density increased from 1.7 x 10⁶ CFU ml^-1 on first day to a maximum of 8.70 x 10⁸ CFU ml^-1 on the sixth day. The OD increased from 0.40 absorbancy unit on first day to 0.15 units (62.5% decolourisation) on the sixth day. Higher percentage decolourisation (80%) on MSC medium compared to 62.5% on MSP medium may be explained on the number of azo linkage. Carmoisine is a monoazo while, Ponceau 4R is a diazo. The cleavage of single bond in Carmoisine was faster than the double bond in Ponceau 4R.

The dye molecules were the sole sources of carbon and energy in MSC (Fig. 2) and MSP (Fig. 3) media. Aerobic metabolism of azo dyes involves the

Initial reductive cleavage of azo bond followed by ring opening of aromatic moiety of the aromatic amines. The carbon atoms are then utilized as carbon source. The increase in total viable counts (Figs. 2 & 3) show that the dye molecules served as sole sources of carbon and energy. Previous studies (Oranusi and Ogugbue, 2001; 2003, and Chen et al., 1999) reported bacterial utilization of azo dyes as sole sources of carbon and energy under aerobic conditions.

The pH increased from initial pH 7.0 to final pH 7.40 on medium which contained Carmoisine and pH 7.2 on medium which contained Ponceau 4R in aerobic cultures. Under anaerobic cultivation, the pH was 7.30 in MSC medium and 7.10 in MSP medium. The increase in pH to the alkaline range may be due to accumulation of intermediate by-products (aromatic amines) and/or other metabolic products.

Decolourisation was observed when the bacterium was cultured under anaerobic condition. Percentage dye decolourisation was 75% in Camoisine – containing medium (MSC) and 60% in medium which contained Ponceau 4R (MSP). No decolourisation was observed in the control flask. Escherichia coli is a facultative anaerobe. The results show that decolourisation could occur in the anaerobic environment of the gastrointestinal tract.

After cleavage of azo bond the component aromatic amines are absorbed in the intestine and may be excreted in urine and/or enter the general circulation (Brown and Devito, 1993). According to Brown and Devito (1993), azo dyes might be toxic only after reductive cleavage of azo bond producing aromatic amines. Aromatic amines have been implicated in cases of bladder cancer in humans, human methemoglobinemia, induced tumors in liver, intestine and urinary bladder in experimental animals (Medvedez et al., 1988 and Kech et al., 1997).

Food colourants are used for aesthetic reasons rather than for their nutritional qualities. The latency of various types of bladder cancer is 20 years. However, cases of bladder cancer have just been reported after few months of ingestion (Cartwright, 1983). Cases of intestinal cancer is more common in industrialized countries (Berg and Howell, 1974; Wynder, 1974) with increasing industrialization, urbanization and increased consumption of foods containing azo dye-based food colourants the potential health risk is a reality. In industrialized countries, the use of food colourants is under strict regulatory control (Houk et al., 1991). In developing economies cases of various forms of human cancer are more common among affluent population (no available data).

This study has demonstrated that E. coli decolourised the azo-based food colourants (Carmoisine and Ponceau 4R) under anaerobic and aerobic conditions. The azo reductase are extracellular and decolourised the dyes. Studies are continuing on other gastrointestinal and urinary bladder microflora which have the potential to decolourise these and other food colourants. The properties of the azo reductases and identification of the aromatic amines are also being investigated.

This research was funded by Senate Research Grant, University of Port Harcourt, Port Harcourt, Nigeria. We are grateful to C. J. Ogugbue and A. Sede for technical assistance.