Six bacterial strains of S. aureus isolated from raw milk were used in this study. The isolates were cultured in two types of milk UHT (skimmed and whole) for 24 h at 37˚C. Then, the cells were retrieved and washed twice using sterile distilled water by centrifuged for 15 minutes at 8400 g and 37˚C, and suspended in a KNO₃ for an ionic force of 0.1 M. The optical densities of each bacterial suspension were adjusted between 0.7 and 0.8 to have almost 10⁸ FCU/ml.

The solid support selected for this study was the microscope slide glass (Knittel Glazer, Germany). The samples were divided into 1 cm², before being covered with milk. The coupons were soaked for 15 minutes in a 70% ethanol solution (Vol/Vol) washed in distilled water and then sterilized in an autoclave at 120˚C for 15 minutes. Glass was used as a model support.

The cleaned glass was placed in a Petri dish and 10 ml of two types of milk were added: ultra-high temperature (UHT) treated milk and skimmed UHT milk. The glass was put in contact with the milk for 3 hours at 25˚C. After the contact time, the coupons in the glass were rinsed three times with distilled water.

The hydrophobicity of glass coupons was determined by measuring the contact angle of water (θw). The lifshitz van der Waals (γ^LW) component and the electron donor (γ−) and electron acceptor (γ+) surface energy parameters of glass coupons were determined by measuring contact angles using the approach proposed by Van Oss [11]. In this approach, in which spreading pressure is ignored, the contact angle θ, measured with a pure liquid (L), is expressed as:

cos θ = − 1 + 2 γ S LW + γ S + γ L − + γ S − γ L + γ L (1)

Three liquids (one non-polar and two polar) are used. Thus, three equations are obtained which are solved to obtain γ^LW component and γ^−s and γ^+s parameters for the solid substrate. The three pure liquids used were water (Milli-Q plus), formamide and diiodomethane.

The Lifshitz-van der Waals (γ^LW) component and the electron donor (γ⁻) and electron acceptor (γ⁺) surface energy parameters of glass were found to be 31.97, 46.24 and 1.14 mJm⁻², respectively.

Surface hydrophobicity was assessed by contact angle measurements and the van Oss approach [11] [12]. In this model, the level of hydrophobicity of a material is defined as the free energy of interaction between two elements of the material when it is submerged in water (w): ∆G_iwi. If the interactions between the two elements are greater than the interactions of each element and water, the element is hydrophobic. (∆G_iwi < 0); and vis versa for a hydrophilic material, ∆G_iwi > 0. According to the following formula, ∆G_iwi is calculated through the surface tensions of the interacting elements:

Δ G i w i = − 2 γ i w = − 2 [ ( ( γ i LW ) 1 2 − ( γ w LW ) 1 2 ) 2 + 2 ( γ i + γ i − ) 1 2 + ( γ w + γ w − ) 1 2 − ( γ i + γ w − ) 1 2 − ( γ w + γ i − ) 1 2 ] (2)

Ten millimeters of bacterial suspension containing 108CFU ml1 was incubated in a Petri dish containing a glass coupon treated with milk (whole, skimmed) for 3 h at 25˚C. After incubation for 3 h at 25˚C, the coupons were then rinsed three times with sterilized distilled water to remove non-adherent bacteria.

The quantification of the adhesion is carried out using a program developed on Matlab v.7.1 software by the research team Laboratory of Bioprocess and Bio-interfaces at the faculty of Sciences and Techniques of Beni Mellal [13].

The obtained microscopic images are converted into images of 256 gray levels and are processed in order to establish a histogram representing the number of pixels according to their intensity. In such treatment, the intensity 0 is attributed to a black pixel, whereas a white pixel corresponds to the intensity 255. The zones corresponding to the colonization of the microorganisms are easily detected with respect to the original image.

The statistical analyses were performed using software STATISTICA version 6. The Newman-keuls test was used to compare the means log10UFC/cm² of each surface before and after treatment (P < 0.05). The means presented in the figures with the same letters are not significantly different and those with different letters are significantly different.

In the agri-food environment, materials surface is quickly covered with an organic conditioning film [7], this film can occur within seconds of exposure to an aqueous environment [14]. On dairy industry, this film could be composed of different milk components [14].

Table 1 presents the surface physicochemical characteristics of glass before treatment with the two types of milk. The contact angle with water can be used as a qualitative indication of the cell surface hydrophobicity [15]. Based on the approach of van Oss [16], it is possible to determine the absolute degree of hydrophobicity of any substance towards water (w), which can be accurately expressed in the internationally applicable system of units. The glass surface was

γ^LW: The Lifshitz-Van der Waals components of the surface tension. γ⁻: electron donor components of the surface tension. γ⁺: electron acceptor components of the surface tension. ΔG_iwi: the free energy of interaction between two entities of that material when immersed in water.

hydrophilic (∆G_iwi > 0) and predominantly highly electron donor with a low electrons acceptor character.

The surface tension values have been measured after treatment of the surface with two types of milk at 25˚C. The results are presented in Table 1. A large change of physicochemical properties was observed regardless. Important variations in electron donor were detected after contact of the glass with the two types of milk. Regardless of the treatment conditions, the electron donor for the glass decreases after treatment by more than 90%. The difference in character of the electron donor after treatment with the two types of milk gives the following results: the value of the donor character after treatment with UHT milk decreases, on the other hand, the value of the acceptor character increases. Then the level of these changes depends on the type of milk used for treatment.

The results also show that the glass surface changes its physicochemical properties. This could be due to the nature of the culture medium. The results also show that the glass surface quantitative hydrophobicity increase after treatment with a different milk. After treatment, the glass quantitative hydrophobicity increases with the augmented presence of the fat components. Finally, we can suggest that there is a parallelism between the presence of fat components of milk, electrons donor character and quantitative hydrophobicity of glass. The parameters increase in the presence of milk, particularly for skimmed milk where the ΔG_iwi pass from (42.74) to (−35.6). Several works have shown that the hydrophobicity measured by the contact angle, is directly correlated with the high ratio of N/C concentrations and inversely correlated with O/C concentrations [17] - [25].

We have evaluated the risk of biocontamination of two different situations of glass treated with skimmed and whole milk for 3 h. The results of adhesion of the six S. aureus strains cultured in milk on untreated and treated glass show differences in terms of the degree of bacteria attachment to glass supports.

The adhesion of six S. aureus cultivated in skimmed milk on the untreated glass shows a better adhesion of the bacteria for all strains except the number 2 (Figure 1 and Figure 2). The results show that six strains of S. aureus adhere easily to a clean (untreated) substrate. These results are in agreement with those obtained by other authors [13] [26] [27] [28]. The results of the untreated glass test showed that when strains 1, 3, 4, 5 and 6 (except the strain 2) were grown in skimmed milk, the adhesion was high, with values of the occupied surface respectively of 7.33%, 19.21%, 26.09%, 17.07% and 11.54%. However, bacteria grown in whole milk with a percentage of the occupied surface area of (4.59%), (2.77%), (8.5%), (4.95%) and (2.89%) for strains 1, 3, 4, 5 and 6 respectively (Figure 2). In addition, an ANOVA statistical analysis showed that the type of milk (skim or whole) had a large effect on the adhesion of S. aureus on untreated glass coupons (0.05 > P). However, when glass coupons were treated with milk (skim or whole); did not affect the adhesion of S. aureus (0.05 < P). Numerous studies have shown that S. aureus adheres in different to untreated surfaces such as stainless steel [13] [26] [28].

In the case of glass surfaces treated with milk, it should be noted that the bacteria do not adhere in the same way as the case of treated surface with skimmed milk. The percentage of surface occupied decreases for all strains except forstaph 1 and 2. In addition, for the surface treated with whole milk the percentage of surface occupied increases for staph 1, 2 and 3 and almost no change for staph 4, 5 and 6. The strains are not always the same level of attachment on the glass (Figure 3 and Figure 4) with those reported by Barnes, Lo [29] who observed that the preconditioning of stainless steel surface with milk or individual milk proteins reduced the adhesion of Staphylococcus aureus, Pseudomonas fragi, Escherichia coli, Listeria monocytogenes, and Serratia marcescens to stainless steel after 2 h of incubation.