Bacillus cereus is a Gram positive rod-shaped aerobic, endospore-forming bacterium. Strains of B. cereus are widely distributed in the environment, mainly in soil, from where they easily spread to many types of foods, especially of vegetable origin, as well as meat, eggs, milk, and dairy products. This bacterium is one of the leading causes of food poisoning in the developed world. B. cereus causes two types of food-borne intoxications. One type is characterized by nausea and vomiting and abdominal cramps and has an incubation period of 1 to 6 hours. This is the "short-incubation" or emetic form of the disease. The second type is manifested primarily by abdominal cramps and diarrhea with an incubation period of 8 to 16 hours. This type is referred to as the "long-incubation" or diarrheal form of the disease [1, 2].

Different strategies may be employed to prevent B. cereus poisoning, like heating food above 75°C before use to kill vegetative cells. However, increasing trends for use of packed foods require new food preservation methods to increase the safety levels against B. cereus. One of the current approaches is the use of antimicrobial peptides (either alone or in combination with other hurdles) such as enterocin AS-48 and other bacteriocins [3–5].

Bacteriocins are small, ribosomally-synthesized antimicrobial peptides synthesized and used by one bacterium as to inhibit growth of similar or closely related bacterial strains [6]. Bacteriocins are categorized in several ways, e.g. on basis of the producing strain, common resistance mechanisms, and mechanism of killing. Enterocin AS-48 is a broad-spectrum antimicrobial peptide produced by Enterococcus faecalis S-48, belonging to Class III of enterococcal bacteriocins or enterocins [7]. Enterocin AS-48 is a 70-residue cyclic peptide with a molecular weight of 7.15 kDa [8]. The crystal structure of enterocin AS-48 has been resolved to 1.4 Ǻ resolution [9]. It is unique with respect to its natural cyclic structure in which N and C termini are linked by a peptide bond. It has been shown that enterocin AS-48 adopts different oligomeric structures according to physiochemical conditions: it exists in monomeric form at pH below 3 and in dimeric form in the pH range of 4.5 to 8.5. The molecules of AS-48 in the crystal are arranged in chains of pairs of molecules linked either by hydrophobic interactions (dimeric form I, abbreviated to DF-I), or by hydrophilic interactions (dimeric form II, abbreviated to DF-II). The molecules within the DF-I interact through the hydrophobic helices H1 and H2. On the other hand, the hydrophilic surfaces of helices H4 and H5 are interacting in DF-II.

The mode of action of enterocin AS-48 has been elucidated [10]. This bacteriocin makes pores of an approximate size of 0.7 nm in the bacterial cytoplasmic membrane thereby disrupting the proton motive force and causing cell death [10]. It also shows a secondary, bacteriolytic effect against some of the target bacteria. Based on its crystal structure, the proposed mechanism of action suggests that the two different stages of molecular association, DF-I and DF-II, are involved in changing from the water-soluble DF-I to the membrane-bound DF-II stage at the membrane surface. This transition implies a 90° rotation of each protomer within DF-I, in a way that the partially hidden hydrophobic helices H1 and H2 become solvent accessible [9]. This would permit AS-48 to insert into the bacterial membrane.

Although the mechanism of action of enterocin AS-48 has been studied extensively at physiological and physico-chemical levels, nothing is known about the responses of sensitive bacterial cells upon exposure to the bacteriocin. Previous experiment in our laboratory with AS-48 against Listeria monocytogenes showed that bacterial cells can be adapted to AS-48, thereby increasing resistance against AS-48 [11]. This adaptation can be achieved with subsequent inoculation in the presence of low, but still inhibitory concentrations of AS-48. However, the adaptation is gradually lost upon repeated subcultivation. Given the great interest of enterocin AS-48 as a food preservative, it is of high relevance to know how the target bacteria react to bacteriocin treatment. This may have direct implications on the elucidation of probable mechanisms for cell adaptation as well as the development of bacteriocin resistance mechanisms. Moreover, a better knowledge of the bacterial response to enterocin AS-48 may also allow identification of new targets that could be exploited to enhance bacteriocin activity. The purpose of the present study was to determine the genome-wide response of B. cereus cells exposed to enterocin AS-48 and to identify components that help the bacterium to survive bacteriocin treatments.

In the present study a sublethal AS-48 concentration was used to detect gene expression differences in B. cereus ATCC14579 that result from interaction of AS-48 with the cells, but not in response to cell death induced by AS-48. We aimed to determine which genes help B. cereus to survive confrontation with AS-48 and identify possible resistance mechanisms. While there was very mild change in the growth after 30 min. incubation with a sublethal bacteriocin concentration, at least 24 genes were affected significantly (Table 1). The observed changes in gene expression were mostly related to up-regulation of membrane associated or periplasmic proteins and downregulation of an operon involved in arginine/ornithine catabolism. Downregulation of argnine/ornithine metabolic genes might be related to the slight difference in growth upon AS-48 treatment that is not apparent using OD measurements. Also, this downregulation might cause a change in local pH at the cell wall in view of the decreased catabolic production of NH₃ and CO₂. Upregulated genes coded for hypothetical membrane proteins or putative transporters. The BC4206-BC4207 operon was most heavily upregulated in B. cereus upon AS-48 treatment. BC4206 is PadR type regulator, while BC4207 is a hypothetical membrane protein with 4 transmembrane segments. Members of the PadR family are known to have a function in regulating cellular pathways resulting in multidrug resistance, virulence or detoxification [20, 21]. These proteins involved in resistance mechanisms, are generally encoded in the vicinity of the padR genes. Overexpression of the BC4207 protein in both B. cereus and B. subtilis results in elevated resistance against AS-48. Upon overexpression of BC4207, we have found no other genes to be upregulated (data not shown), suggesting that increase in BC4207 expression alone raised the resistance of B. cereus against AS-48. Interestingly, enhanced resistance upon BC4207 overexpression was specific to enterocin AS-48 and not observed in the presence of bacitracin or nisin. Bacitracin and nisin both effect cell wall biosynthesis through blocking the lipid II cycle [16] and forming pores in the cell membrane during interaction with lipid II [17, 18], respectively. This is not the case of enterocin AS-48, since the primary action of this antimicrobial peptide, like most other bacteriocins, is the disruption of the cytoplasmic membrane.

In spite of recent advances on genome and transcriptome analysis, there are very few reports on the effects of antimicrobial substances on bacterial gene expression. Recently, Martínez et al. (2007) [22] reported that lactococcin 972 (Lcn972) significantly upregulated the expression of 26 genes in Lactococcus lactis, most of which encode membrane proteins of unknown function and the two component system (TCS) CesSR (formerly known as TCS-D) associated with cell-envelope stress. Lcn972 is a non pore-forming bacteriocin that inhibits the synthesis of peptidoglycan at the septum in Lactococcus lactis. Moreover, the response of a number of Gram-positive bacterial species towards cell wall active antibiotics has been studied recently by using genome-wide transcription analysis [19, 23–27]. Essentially, these reports describe a very complex system involving the concerted action of extracellular sigma factors and two-component systems (TCSs) [28]. LiaRS, the B. subtilis homologue of CesSR, was unable to activate liaI expression in B. subtilis in response to AS-48 treatment. Therefore, the effect of AS-48 on bacterial gene expression clearly differs from the mechanisms described earlier for B. subtilis [28].

The precise way in which BC4206 responds to the presence of AS-48 needs to be deciphered by further experimental work, including determining the target genes of BC4206 and the exact signal sensed by this PadR-type regulator. The structure and function of the BC4207 membrane protein and its role in the resistance mechanism against AS-48 is also particularly intriguing and target of our future research.

B. cereus cells, when treated with bacteriocin AS-48, increase the expression of the BC4207 gene coding for a putative membrane protein. Targeted inactivation of the BC4207 protein might be useful to increase the effect of AS-48 on food poisoning B. cereus cells.