Chemical inhibition of the contaminant Lactobacillus fermentum from distilleries producing fuel bioethanol

ENGINEERING, TECHNOLOGY AND TECHNIQUES

^IDepartamento de Ciências Biológicas; Laboratório de Biotecnologia Industrial; Universidade Estadual Paulista; Assis - SP - Brasil

^IIFaculdade de Engenharia de Alimentos; Universidade Estadual de Campinas; Campina - SP - Brasil

ABSTRACT

The purpose of this study was to determine the Minimum Inhibitory Concentration (MIC) of pure or mixed chemicals for Saccharomyces cerevisiae and Lactobacillus fermentum in the samples isolated from distilleries with serious bacterial contamination problems. The biocides, which showed the best results were: 3,4,4' trichlorocarbanilide (TCC), tested at pH 4.0 (MIC = 3.12 mg/l), TCC with benzethonium chloride (CBe) at pH 6.0 (MIC = 3.12 mg/l) and TCC mixed with benzalkonium chloride (CBa) at pH 6.0 (MIC = 1.53 mg /l). If CBa was used in sugar cane milling in 1:1 ratio with TCC, a 8 times reduction of CBa was possible. This formulation also should be tested in fermentation steps since it was more difficult for the bacterium to develop resistance to biocide. There was no inhibition of S. cerevisiae and there were only antibiotics as an option to bacterial control of fuel ethanol fermentation by S. cerevisiae.

Key words: Ethanol, lactic acid bacteria, Saccharomyces cerevisiae, Lactobacillus fermentum

INTRODUCTION

Currently, bioethanol is considered an important recycled fuel and an alternative to fossil fuels. In Brazil, it is produced by fed-batch or continuous fermentation process from sugar cane using Saccharomyces cerevisiae in cell recycles. In this process, the microbial contaminants are also recycled. Bacterial contamination is an aggravating factor associated with several problems such as the consumption of sugar, alcohol and other medium components, reduction in ethanol yield (Oliva-Neto and Yokoya 1994, 1996); release of toxins and organic acids in the work, and decrease in the viability of yeast cells (Maiorella et al. 1983, Essia-Ngang et al. 1989, Dorta et al. 2007). It is well known that in the fermentation process, Gram-positive bacteria are the main agents of contamination, especially Lactobacillus spp. (Galo 1989; Oliva-Neto 1990). The antagonism between L. fermentum and S. cerevisiae is due to organic acids produced by the bacterial cells (Oliva-Neto and Yokoya 1994, 1996). Furthermore, L. fermentum is a contaminant agent responsible for yeast flocculation, which is associated with corrosion and obstruction of pumps and centrifuges and a decrease in antibiotic effectiveness and fermentation yield (Oliva-Neto and Yokoya 1991; Ludwig et al. 2001).

In the fermentation process, antibiotics are added to control bacterial contamination, but use of these compounds is limited. This method of control is limited to a few products. Currently, HJ Kamoran is widely used in Brazilian distilleries, but at a high cost. Thus, the discovery of new products with antimicrobial activity against L. fermentum is necessary. Inhibitory action against the growth of L. fermentum has been determined for acid penicillin V (MIC 0.10-0.20 mg/L), clindamycin (MIC 0.05-0.40 mg/L), sulphite (MIC 10-40 mg/L), thiocianate (1.2-5.0 mg/L), formaldehyde (11.5-23 mg/L), cephamandole(0.26-1.45 mg/L), bromophenate (9-18 mg/L), methyldithiocarbamate (2.5 mg/L), copper sulphate (75-300 mg/L) and N-alkyl dimethylbenzyl ammonium chloride or benzalkonium chloride (MIC 8.0 mg/L) (Oliva-Neto and Yokoya 2001). Polymixin B sulphate showed MIC of 64 mg/L against L. fermentum, which was more than 1024 mg/L for other species of Lactobacillus spp. (Flores et al. 2008). The MIC of gentamicin against L. plantarum was 128 mg/L, but this antibiotic did not perform well against other species of Lactobacillus spp. (Rojo-Bezares et al. 2006). According to Danielsen and Wind (2003), metronidazole was not effective in inhibiting Lactobacillus spp. (MIC> 200 mg/L). The surfactants sodium lauryl sulphate, benzethonium chloride and benzalkonium chloride were also evaluated for inhibiting the growth of L. fermentum and S. cerevisiae (Silva et al. 1997).The 3,4,4' trichlorocarbanilide demonstrated a selective inhibition against L. fermentum and Leuconostoc mesenteroides (Oliva-Neto and Yokoya 1998).

In this work, in vitro antimicrobial activity (MIC) against L. fermentum and S. cerevisiae was determined for several antimicrobial compounds aiming to find new alternatives for the control of bacterial contamination in fuel ethanol production.

MATERIALS AND METHODS

Microorganisms

The bacterial cultures used in this study were L. fermentum CCT 1396, isolated from an alcohol distillery by Oliva-Neto and Yokoya (1994), and L. fermentum CCT 0559, obtained from the Tropical Culture Collection (CCT), Campinas – Brazil. Both the strains were activated and maintained in Man Rogosa, Sharpe medium (Difco) at pH 4.5. The yeast strains used were S. cerevisiae CCT 4370, obtained from the Tropical Culture Collection, and S. cerevisiae FCLA M26, also isolated from an alcohol distillery (Oliva-Neto et al. 2004) and obtained from the Culture Collection of the Laboratory of Industrial Biotechnology, São Paulo State University (UNESP), Assis – Brazil. Yeast strains were activated and maintained in nutrient medium consisting of (%, w/v) 2 sucrose, 0.5 yeast extract, 0.1 (NH₄)₂SO₄, 0.114 K₂HPO₄.3H₂O, 0.0017 MnSO₄.7H₂O, 0.0028 ZnSO₄.7H₂O, 0.024 MgSO₄.7H₂O and distilled water. All the cultures were incubated at 30ºC for 24-48h.

Minimum inhibitory concentration (MIC)

The following chemicals were tested for MIC at 140 mg/l: nalidixic acid (Sanofi-Synthelabo Ltda.), pipemidic acid (Zambon Pharmaceutical Laboratories Ltda), phenazopyridine hydrochloride (Blanver Farmoquímica Ltda.), metronidazole (Aventis Pharma Ltda.), nitrofurantoin (Laboratory Teuto Brasileiro S/A), sulphasalazine (Apsen Farmacêutica S/A), sulphamethoxazole/ trimethoprim (Roche Chemicals and Pharmaceutical Products S.A), sulphadiazine silver (Pharma Nostra), gentamicin sulphate (Lab Duct Pharmaceutical Industry Ltda), sulfacetamide sodium (Henrifarma) and polymyxin B sulphate (American Farmasa Laboratory of pharmacotherapy SA). Pipemidic acid and nitrofurantoin were formulated with sodium dodecyl sulfate (SDS) (3:1 and 1:1, w/w) to evaluate the synergistic effect with this compound. The other tested chemicals were at 250– 4000 mg/L tartaric acid and sodium gluconate, and at 0.78-12.5 mg/L 3,4,4' trichlorocarbanilide (TCC), benzethonium chloride (Cbe), benzalkonium chloride (Cba), cetyl trimethyl ammonium chloride (CTA) and Hj Kamoran^® (the antibiotic Monensin – Química Real - Ribeirão Preto - SP - Brazil). The products Cbe, Cba, tartaric and gluconate acids were dissolved in distilled water. TCC was dissolved in acetone. All the chemicals were autoclaved (121ºC/15 min) or microfiltered (0.22 µm membrane Millipore - EUA).

The MIC of these chemicals was determined by adapted macrodilution broth method (Jones et al. 1985). Assays were performed in the tubes containing 6.0 mL medium. The inoculum was standardized according to MacFarland 0.5 standard in aseptic conditions. The cultures were incubated at 30ºC for 24 h in an incubator and bacterial growth was aseptically measured by absorbance at 600 nm wavelength (A₆₀₀) by using a spectrophotometer (Spectronic 1100 Pharmacia USA). The MIC was defined as the minimum concentration able to inhibit at least 90% the microbial growth and performed in triplicate. The growth reduction was calculated as [1 – [(A₆₀₀ 24 h with antimicrobial compound – A₆₀₀ 0h with antimicrobial compound)/(A₆₀₀ 24 h control – A₆₀₀ 0h control)]] x 100 and expressed in %.

Tartaric acid showed a MIC of 2000 mg/L for S. cerevisiae (CCT 2652) and 4000 mg/L for S. cerevisiae M26 (Table 1). For sodium gluconate, a MIC more than 2000 mg/l was found for all the strains tested. These values of MIC, therefore, were not recommended for bacterial control, because the industrial dosage of biocides and antibiotics used in Brazilian distilleries have been 10-20 mg/L, and 3-4 mg/L, respectively. Furthermore, these products inhibited the yeasts as well as the bacteria.

From the pure cationic surfactants (Table 1) available, only cetyl trimethyl ammonium chloride - CTAC (Fig. 1) inhibited L. fermentum with a MIC between 3.12-6.25 mg/L. However, these values were close for S. cerevisiae (MIC 6.25-12.5 mg/L). Autoclaved or microfiltered Cba or Cbe showed higher values (MIC > 6.25 mg/L) than CTAC for both L. fermentum and similar MIC for S. cerevisiae. These results showed that pure cationic surfactants (Cbe, Cba and CTAC) could not be not recommended for the control of bacterial contamination in the fermentation process of fuel ethanol by S. cerevisiae. However, Cba is currently used in cane milling steps of fuel ethanol production to bacterial control, mainly for the control of Leuconostoc mensenteroides and Lactobacillus sp at 10-20 mg/L. Ammonium quaternary (QA) compounds, dithiocarbamate and halogenated phenols have been used in Brazilian distilleries (Cereda et al. 1981). Sanitization with QA compounds has been performed and has saved as much as 60% of sugar loss from lactic acid bacteria contaminants in the sugar industry (Tilbury et al. 1977). N alkyl-di-methyl-benzyl ammonium chloride (Cba) showed a similar MIC (8.0 mg/L) for S. cerevisiae, L. fermentum and L. mesenteroides in fuel bioethanol distilleries, therefore, it would not be a practical additive for alcoholic fermentation process, since yeast cells could also be inhibited by this compound at the dosage necessary for bacterial growth inhibition (Oliva-Neto and Yokoya 1998).

Pure TCC presented an MIC of 6.25 mg/L for the tested bacterium at pH 6.0 and 3.12 mg/L at pH 4.0 (Table 1). The latter pH was similar to industrial conditions. This compound was effective against Gram positive bacteria. For example, an MIC against Staphylococcus aureus of 0.078 mg/L has been reported by Hamilton (1971) and TCC action against L. mesenteroides (MIC = 0.5 mg/L) and L. fermentum (0.5-2.0 mg/L) has been reported by Oliva-Neto and Yokoya (1988), which were similar to the levels reported in the present work. TCC did not inhibit S. cerevisiae growth in concentrations similar to L. fermentum (MIC= 12.5 mg/L). An MIC higher to 200 mg/L against S. cerevisiae has been reported by Oliva-Neto and Yokoya (1988).

The synergism between TCC and cationic surfactants CBe and CBa inhibited bacterial growth. A formulation with a mixture of these compounds was justified because TCC was not water soluble and pure ammonium quaternary compounds inhibited yeast growth. The combination of TCC and CBe (1:1 w/w) showed a decrease in the MIC of TCC against L. fermentum compared to pure TCC (Table 2). This microfiltered formulation was more effective (MIC 3.12 mg/l) than the autoclaved (MIC = 6.12 g/L). Nevertheless, the TCC/CBe formulation did not inhibit S. cerevisiae (MIC >12.5 mg/l) in the tested concentrations, which was an important condition for the application of this product on an industrial scale for fuel ethanol production.

The synergistic effect between TCC and CBa was also observed. The proportional increase of TCC in relation to CBe caused an increase in the MIC against L. fermentum (Table 2). The MIC observed with TCC/CBa at 5:1 (w/w) was 6.25 mg/L, but the best MIC was 1:1 (w/w) (1.56 mg/L) for L.fermentum and only 12.5 mg/L for S. cerevisae (Fig. 2). This result was probably observed because the surfactant led a better dilution of TCC in the broth, which was associated with the inhibitory action of these products. According these results, if CBa was used in sugar cane milling at 1:1 ratio with TCC, it resulted 8 times reduction of CBa since this value was the quotient of pure CBa and CBa:TCC (1:1) MIC. The MIC of autoclaved TCC/CBa (2.5:1, w/w) decreased four times in relation to this microfiltered formulation, probably because this formulation was not heat-resistant. However, TCC/CBa formulation showed a higher MIC (>12.5 mg/L) against S. cerevisiae than L. fermentum. In this work, a ratio lower to 1:1 TCC:CBa was not available, but it was possible that the MIC against L. fermentum could be even lower. The improvement in antibacterial action of TCC when combined with surfactants follows the same pattern of TCC/Cbe. The surfactant probably improved the TCC solubility in bacterial cells, since the MICs of the chemicals CBa, Cbe, and TCC against L. fermentum were higher than those of the combined products.

The MIC of autoclaved Hj Kamoran^® against L. fermentum was the lowest found in this study (0.078-0.156 mg/L) and higher to the microfiltered formulation (0.156-0.312 mg/L), probably due to its heat-resistance and loss in the microfiltration process (Fig. 3). Hj Kamoran^® (antibiotic monensin) is currently used in bioethanol distilleries in Brazil and is the most important product for the control of bacterial infections in industrial alcoholic fermentation. The usual dosage in distilleries was formerly 1.0-3.0 mg/L (Oliveira et al. 1996) but has now been increased to 3.0-4.0 mg/l. This was probably due to an increase in bacterial resistance to the antibiotic.

Autoclaved Hj Kamoran^® presented an average MIC against L. fermentum (0.117 mg/L) 13.3 times lower than formulated TCC/CBa (1:1) MIC. Nevertheless, despite the efficiency of Hj Kamoran^®, the formulation TCC/Cba could be improved. It would be worth considering that these types of biocides were usually cheaper than antibiotics. Furthermore, the biocide mechanism of bacterial growth inhibition is generally less specific than that of antibiotics, and this factor may make it more difficult for bacteria to become the resistant to TCC/CBa.

Table 3 shows the MICs obtained for chemotherapeutic agents. Nalidixic acid, pipemidic acid, phenazopyridine, sulphadiazine silver, sulphasalazine, gentamicin sulfate, sulfamethoxazol/ trimethoprim, sulfacetamide sodium and polymyxin B sulphate showed MICs higher than 40 mg/L against L. fermentum. Since the usual concentrations of antibiotics used in fuel ethanol fermentation were lower than 4.0 mg/L, these compounds could not be recommended to control lactic acid bacterial contaminants of alcoholic fermentation. As for gentamicin, some authors have reported a resistance of 57.7% in Lactobacillus spp. from the fermented products (Olukoya et al. 1993). Moreover, it has been shown that the MIC of gentamicin for L. plantarum was almost 128 mg/L and indeed this antibiotic did not perform well against other species of the Lactobacillus genus (Rojo-Bezares et al. 2006). The MIC of polymyxin B sulphate for L. fermentum has been reported as 64 mg/l, which was higher than 1024 mg/L for other species of this genus (Florez et al. 2008). An MIC >1024 mg/L for sulphamethoxazole against Lactobacillus spp. has also been reported. These results confirmed previous reports about the intrinsic resistance presented by the genus Lactobacillus. Sulfas are competitive inhibitors of enzymes that convert the substrate para-aminobenzoic acid (PABA), an essential nutrient used by many bacteria, for use in the synthesis of the coenzyme folic acid (Tortora et al. 1998). Lactic acid bacteria, however, do not produce folic acid, and so sulfonamides such as sulfamethoxazol and sulfacetamide are inactive against Lactobacillus spp. (Katila et al. 2001).

Resistance to gentamicin is associated with membrane impermeability of the Lactobacillus spp. (Elkins and Mullis 2004). With respect to metronidazole, this antibiotic was not effective in inhibiting either bacterial or yeast growth, presenting an MIC > 200 mg/l for Lactobacillus concentrations than the antibiotics normally used spp. Furthermore, it has been reported that 128 in fuel ethanol fermentation. The effectiveness of mg/l metronidazole may actually stimulate nitrofurantoin against Lactobacillus spp. has been Lactobacillus growth (Choi et al. 2003). previously reported (MIC 8.0 mg/l). More Nitrofurantoin was the only chemoterapeutic agent specifically, for six strains of L. acidophilus, its that provided an inhibitory effect against L. MIC was reported as 2.0 mg/L (Danielsen and fermentum, (MIC= 15 mg/L), but it took higher Wind 2003).

CONCLUSION

There was a synergistic effect between 3,4,4' trichlorocarbanilide (TCC) and benzethonium chloride (CBe) or benzalkonium chloride (CBa), improving the inhibitory action against the growth of L. fermentum. These formulations also increased the selective action only on bacteria, not inhibiting the growth of S. cerevisiae. The best MIC (1.56 mg/L) of the combined chemicals against L. fermentum was obtained by the formulation TCC: CBa 1:1 (w/w). CBa could be tested in sugar cane milling in 1:1 ratio with TCC since a reduction of 8 times in the use of this product was possible. This formulation also should be tested in fermentation steps since it was more difficult for the bacteria to develop resistance to biocide. There was no inhibition of S. cerevisiae. Currently, there are only antibiotics as an option to bacterial control of fuel ethanol fermentation by S. cerevisiae. The present results demonstrated the importance of further studies about the TCC mixed with surfactants for the improvement of fuel bioethanol technology.

ACKNOWLEDMENTS

The authors thank the Fundação de Amparo à Pesquisa do Estado de São Paulo – FAPESP (Brazil) for financial support, Fundação Tropical André Tosello for the donations of the cultures.

Received: August 06, 2012;

Accepted: December 23, 2013.

• Cereda MP, Serra GE, Cagliari AM, Meneguim MA. Microbial method for evaluating antiseptics used in alcohol industry. Brasil Açucareiro 1981; 97: 44-51.
• Choi SY, Chang CE, Kim SC, So J. Antimicrobial susceptibility and strain prevalence of Korean vaginal Lactobacillus spp. Anaerobe. 2003; 9: 277-280.
• Danielsen M, Wind A. Susceptibility of Lactobacillus spp. to antimicrobial agents. Int J Food Microbiol 2003; 82: 1-11.
• Dorta C, Oliva-Neto P, Abreu-Neto MS, Nicolau-Junior N, Nagashima AI. Synergism among lactic acid, sulfite, pH and ethanol in alcoholic fermentation of Saccharomyces cerevisiae (PE-2 and M-26). World J Microbiol Biotechnol 2006; 22: 177-182.
• Elkins CA, Mullis LB. Bile mediated amiloglycoside sensitivity in Lactobacillus species likely results from increased membrane permeability attributable to cholic acid. Appl. Env. Microb 2004; 70(12): 7200-7209.
• Essia-Ngang JJ, Letourneau F, Villa P. Alcoholic fermentation of beet molasses effects of lactic acid on yeast fermentation parameters. Appl Microb Biotechnol 1989; 31: 125-128.
• Florez AB, Ammor MS, Mayo B. Antimicrobial susceptibility profiles of 32 strains of Lactobacillus, Bifidobacterium, Lactococcus and Streptococcus spp. Intern J Antimicrob Agents 2008; 3: 484-504.
• Gallo CR. Determinação da microbiota bacteriana de mosto e de dornas de fermentação alcoólica. Campinas, Ph.D. Thesis. Faculdade de Engenharia de Alimentos UNICAMP. 1989; 338p.
• Hamilton WA. Membrane active antibacterial compounds. In: Hugo, W.B (ed.). Inhibition and destruction of the Microbial Cell. Academic Press: New York, 1971; 230-235.
• Jones RN, Barry AL, Gavan TL, Washington II JA. Susceptibility tests: Microdilution and Macrodilution broth procedures. In: Lennette EH, Balows A, Halousler WJJr., Shadom Y. (eds). Manual of Clinical Microbiology, American Society of Microbiology: Washington -DC, 1985; 972-977.
• Katla AK, Kruse H, Johsen G, Herikstad H. Antimicrobial susceptibility of starter culture bacteria used in Norwegian dairy products. Int J Food Microb 2001; 67: 147-152.
• Ludwig KM, Oliva-Neto P, Angelis DF. Quantification of Saccharomyces cerevisiae flocculation of contaminants in alcoholic fermentation by bacteria. Cienc Tecnol Aliment 2001; 21: 63-68.
• Maiorella B, Blanch HWC, Wilke CR. By Product inhibition effects on ethanolic fermentation by Saccharomyces cerevisiae Biotech Bioeng 1983; 25: 103-121.
• Oliva-Neto, P. Influence of contamination by lactic acid bacteria in alcoholic fermentation by fed batch process. Master Thesis, Faculdade de Engenharia de Alimentos UNICAMP. Campinas, 1990; 200p.
• Oliva-Neto P, Yokoya F. Evaluation of bacterial contamination in a fed-batch alcoholic fermentation process. World J Microbio Biotechnol 1994; 10: 697-699.
• Oliva-Neto P, Yokoya F. Influence of yeast extract on the stability of contaminated alcoholic fermentation by Lactobacillus fermentum Ciênc Tecnol Aliment. 1996; 16: 170-174.
• Oliva-Neto P, Yokoya F. Effect of 3,4,4'trichlorocarbanilide on growth of lactic acid bacteria contaminants in alcoholic fermentation. Bioresource Technol 1998; 63: 17-21.
• Oliva-Neto P, Yokoya F. Susceptibility of Saccharomyces cerevisiae and lactic acid bacteria from the alcohol industry to several antimicrobial compounds. Braz J Microbiol. 2001; 32: 10-14.
• Oliva-Neto P, Ferreira MA, Yokoya F. Screening for yeasts with antibacterial properties from an ethanol distillery. Bioresource Technol 2004; 92: 1-6.
• Oliveira AJ, Gallo CR, Alcarde VE, Godoy A, Amorim HV. Methods for microbiological control in the production of sugar and alcohol. Piracicaba: FERMENTEC - ESALQ, Piracicaba, Brazil. 1996; 89p.
• Olukoya, DK, Ebigwei, SI, Adebano, OO, Osiyemi, FO. Plasmid profiles and antibiotic susceptibility patterns of Lactobacillus isolated from fermented foods in Nigeria. Food Microb 1993; 10:279-285.
• Rojo-Bezares B, Saenz Y, Poeta P, Zararaga M, Ruitz-Larrea F, Torres C. Assessment of antibiotic susceptibility within lactic acid bacteria strains isolated from wine. Int J Food Microbiol 2006; 111: 234-240.
• Silva ED, Silva SA, Nogueira LC, Oliva-Neto P. Susceptibility of Saccharomyces cerevisiae and lactic acid bacteria on aldehydes and surface active agents on medium with sugar cane juice. II Simpósio de Ciência de Alimentos, Campinas, SP Brazil. 1997.
• Tilbury RH, Hollingsworth BS, Graham SD, Pottage P. Mill sanitization a fresh approach to biocide evaluation. Proceeding of I.S.S.C.T. XVI Congress, 1977; 2749-2768.
• Tortora GJ, Funk BR, Case CL. Microbial Metabolism Energy production. In: Microbiology -An Introduction. Addison Wesley Longman Inc. California - USA. 1998; p.118.
• Yokoya F, Oliva-Neto P. Characteristics of yeast flocculation by Lactobacillus fermentum Rev Microbiol 1991; 22: 21-27.

Chemical inhibition of the contaminant Lactobacillus fermentum from distilleries producing fuel bioethanol

Pedro de Oliva Neto^I,*; Fabíola Aliaga de Lima^I; Ketrin Cristina da Silva^I; Douglas Fernandes da Silva^I; Ana Flavia Azevedo Carvalho^II; Catarina dos Santos^I

Publication Dates

• Publication in this collection
  30 May 2014
• Date of issue
  June 2014

History

• Received
  06 Aug 2012
• Accepted
  23 Dec 2013