Edited by G.M. Gadd
Debaryomyces hansenii strains with different cell sizes and surface physicochemical properties adhere differently to a solid agarose surface
Article first published online: 9 JAN 2006
FEMS Microbiology Letters
Volume 249, Issue 1, pages 165–170, August 2005
How to Cite
Mortensen, H. D., Gori, K., Jespersen, L. and Arneborg, N. (2005), Debaryomyces hansenii strains with different cell sizes and surface physicochemical properties adhere differently to a solid agarose surface. FEMS Microbiology Letters, 249: 165–170. doi: 10.1016/j.femsle.2005.06.009
- Issue published online: 9 JAN 2006
- Article first published online: 9 JAN 2006
- Received 10 February 2005, Accepted 7 June 2005
- Debaryomyces hansenii;
- Initial adhesion rate;
- Agarose surface;
- Electron donation;
- Cell surface hydrophobicity and cell size
The initial adhesion of four Debaryomyces hansenii strains to a solid agarose surface was investigated and correlated with their cell size and some cell surface physicochemical properties, i.e. (i) hydrophobicity and (ii) electron donor/acceptor ability. One strain adhered very poorly, whereas the three other strains were more adhesive. The former strain had a very hydrophilic cell surface, whereas the latter strains had more hydrophobic cell surfaces. In addition, the strain with the lowest adhesion among the adhesive strains had a more hydrophobic cell surface than the two most adhesive strains. Finally, the more adhesive the strain was, the larger it was, and the better it was to donate electrons from its cell surface. These results show a clear relationship between the cell size, the cell surface physicochemical properties, and the initial adhesion of D. hansenii. A possible explanation of this relationship is discussed.
Adhesion of the yeast Debaryomyces hansenii to a solid surface is a prerequisite for initiation of growth to occur. Growth of D. hansenii on solid surfaces is an important co-factor in, e.g. surface ripening of several meat products and cheese [1,2]. Previous studies on yeast adhesion to solid surfaces have mainly focused on the pathogenic species Candida albicans[3,4] and to some degree Saccharomyces cerevisiae[5,6]. To our knowledge, no studies of the adhesion of D. hansenii to solid surfaces have been reported.
In this work, agarose is chosen as a model solid surface due to its transparent gel structure and strong gelating characteristics . Agarose is a linear galactan (d-galactose and 3,6-anhydro-l-galactose) repeating disaccharide structure . Among the methods used to study microbial adhesion to a solid substrate, flow chamber systems with in situ observations are considered superior to e.g. static systems, such as beads and slide methods. In static systems, a microbial suspension remains stationary with respect to a solid surface, and long-term adhesion is often studied. A flow chamber system offers the possibility of measuring initial microbial adhesion to solid surfaces in a controlled and quantitative way, rendering the data comparable on a widespread, absolute scale .
The “microbial adhesion to solvents” (MATS) method is a biphasic separation hydrophobicity assay which can be used for determining the ability of a cell surface to: (i) participate in hydrophobic interactions and to (ii) donate or accept electrons . As to the former application, the method is based on determining the cell affinity to the non-polar solvents, i.e. a high affinity indicates high cell surface hydrophobicity and a low affinity indicates low cell surface hydrophobicity. As to the latter application, the method is based on the comparison between cell affinity to two pairs of solvents, each pair comprising a polar solvent and a non-polar solvent . The polar solvent may be an electron acceptor (e.g. chloroform) or an electron donor (e.g. ethyl acetate). Thus, by comparing cell affinity to the solvent pair, comprising chloroform, indicates the ability of a cell surface to donate electrons. Conversely, by comparing cell affinity to the solvent pair, comprising ethyl acetate, indicates the ability of a cell surface to accept electrons. MATS experiments have shown that cell surfaces of both the bacteria Streptococcus thermophilus and Leuconostoc mesenteroides and the yeast S. cerevisiae are strong electron donors and very weak electron acceptors. In the latter study, the cell surface physicochemical properties of S. cerevisiae strains are correlated with adhesion to a glass surface. A relationship between cell surface physicochemical properties of D. hansenii and adhesion has, as yet, not been reported.
When working with cells on a horizontal surface in a flow chamber it is necessary to take into account that gravity can affect the examined cells . The yeast cell size renders the diameter well beyond the one where gravity forces start to control the sedimentation velocities of cells in flow systems , and it may thus be proposed to influence the adhesion of a yeast cell. Whether the cell size plays a role in the adhesion of D. hansenii in a flow system, is to our knowledge, not known.
In this work, we investigate the adhesion of four D. hansenii strains to a solid agarose surface using a flow chamber system. The adhesion analysis is correlated with cell surface physicochemical properties obtained from MATS experiments. Furthermore, the relationship between cell size and adhesion is examined.
2Materials and methods
2.1Yeast strains, growth media and growth conditions
In this study, the following D. hansenii strains were used: CBS 767, H 2 (Danisco Cultor Innovation, Copenhagen, Denmark), D 18335 (dairy isolate ), and MD 02 (Arla Innovation, Århus, Denmark). Prior to adhesion, MATS, and cell size experiments, a yeast freeze culture was transferred to 25 ml YNB media (containing 6.7 g/l yeast nitrogen base without amino acids, 10 g/l glucose, 7.1 g/l Na2HPO4· 2H2O and 63.5 g/l NaH2PO4· H2O, pH 5.3) in 100 ml shake flasks and incubated with agitation (120 rpm) at 25°C for 24 h. Subsequently, these cells were transferred to 100 ml YNB media in 250 ml shake flasks to an initial concentration of 1.0 × 106 cells/ml and incubated with agitation (120 rpm) at 25°C for 90 h (late stationary growth phase).
2.2Preparation of agarose plates
Agarose (Sigma no. A-0169 with low electroendosmosis (EEO)) was transferred to a glass tube containing 14 ml of 50 mM succinic acid–NaOH buffer (containing 5.91 g/l succinic acid and 2.16 g/l NaOH, pH 5) to a final concentration of 3% (w/w). The glass tube was placed in a water bath at 100°C for 20 min and transferred to a water bath at 60°C for 10 min. The agarose was subsequently poured into a preheated (60°C) glass Petri dish and left to solidify at 20°C for 15 min.
A suspension of 1.0 × 106 cells/ml in 50 ml of 50 mM succinic acid–NaOH buffer (pH 5) was placed on a stirrer and connected to a peristaltic pump (Alitea XV) calibrated to a pumping velocity of 0.250 ml/min. The Petri dish containing the agarose was mounted on a microscope (Zeiss Axioskop) with an attached CCD-camera (CoolSnap). A perfusion chamber (Sigma–Aldrich, type no. Z379115) was placed on the agarose surface. To prevent leakage, grease was carefully applied to the surface contact area of the chamber. The tubing from the pump was fixed in the inlet hole of the chamber and another peristaltic pump (Watson Marlow 101F) ensured removal of waste buffer from the outlet hole. The agarose surface was focused through the transparent top layer of the chamber with a 10 × objective (Plan NeoFluar). In order to assure that the measurements were performed under conditions with a stable flow rate, the area of observation (540 × 380 μm) was always in the center (width and length) of the perfusion channel . After activation of the pumps, two consecutive images were acquired every 2 min for a 20-min period. Where indicated, cell concentrations of 0.5 × 106 and 2.0 × 106 cells/ml were used. All adhesion experiments were performed in duplicate.
2.4Determination of initial adhesion rate (IAR) and image analysis
To determine the number of deposited cells, image analysis was performed with IGOR Pro 5.0 (WaveMetrics). All images were converted to 8 bit gray scale and the two images acquired every 2 min were averaged to eliminate noise from in-focus flowing cells. The images were binarized with a proper threshold and counted automatically. Initial adhesion rate (IAR) was determined using linear regression in IGOR Pro 5.0.
2.5MATS (Microbial adhesion to solvents) analysis
The method was carried out as described previously  with minor modifications. Briefly, yeast cells were washed twice in succinic acid–NaOH buffer (pH 5) and resuspended in buffer to an initial OD400 (Shimadzu UV-1201 spectrophotometer) of approx. 0.8. Subsequently, 2.4 ml of the cell suspension was transferred to a tube, overlaid with 0.4 ml of solvent, and vortexed for 1 min. The following pairs of solvents were used: chloroform, a polar electron acceptor solvent, and hexadecane, a non-polar solvent, as well as ethyl acetate, a polar electron donor solvent, and decane, a non-polar solvent. After 15 min, the phases were separated and 1 ml of the aqueous phase was carefully transferred to a cuvette and OD400 was measured. The solvent affinity was determined as the proportion of cells suspended in the solvents compared to the initial cell concentration, using the equation:
where A0 and A are the OD400 values of the yeast suspension before and after mixing, respectively. Each solvent affinity measurement was carried out in triplicate.
2.6Measurement of cell diameter and calculation of sedimentation velocities
Cells were immobilized on an object glass with concanavalin A (Sigma) and an image was acquired with a 100 × objective (Zeiss Fluar). Cell diameters were determined using MetaMorph 4.5 (Universal Imaging) and were calibrated to μm using a micrometer (Leica). In each experiment, a minimum of 35 cells were analyzed. The sedimentation velocity of the cells was determined using Stokes law:
where D is the cell diameter, g (gravity) is 9.807 m/s2, Δρ (cell density minus water density) is 100 kg/m3 and η (viscosity of water) is 9.04 × 10−4 kg/m s. A cell density of 1100 kg/m3 is adopted from experiments with S. cerevisiae.
Statistical analysis of adhesion, MATS, and cell size experiments was carried out using one-way ANOVA, and multiple comparisons between groups were made using LSD (least significant difference) test in the SAS software (SAS Institute, Inc.). Probabilities less than 0.05 were considered significant.
3.1Initial adhesion rate
In this work, we intended to study the initial adhesion of D. hansenii to a solid agarose surface by determining the number of cells per surface area per time unit in a flow chamber system. In literature, this parameter is defined as the deposition rate and it accounts for the difference between the adsorption and desorption rate of cells at a certain time . In this study, no desorption of cells was observed (data not shown), i.e. cells detected as adhered early in the experiments were also detected on the surface after 20 min. Therefore, we define the number of adsorbed cells per surface area per time unit as the adhesion rate.
The adhesion of D. hansenii (CBS 767) to an agarose surface was linear throughout the 20 min duration of the experiments (Fig. 1). The slope values of three different cell concentrations of D. hansenii (CBS 767) were significantly different (Fig. 1) and correlated in a linear fashion (data not shown). This demonstrates that there is no lack of adhesion sites in our experiments. Thus, the slope value represents the initial adhesion rate (IAR).
3.2Adhesion of D. hansenii strains to agarose
Linear adhesion was observed for all four D. hansenii strains used in this study (data not shown), rendering the IAR comparable between different yeast strains. The four examined D. hansenii strains had significantly different IAR to agarose (Table 1). The IAR (7325 cells/cm2 min) of the most adhesive strain (D 18335) was almost 10 times higher than the IAR (790 cells/cm2 min) of the strain (MD 02) with the poorest adhesion (Table 1). The two remaining strains (H 2 and CBS 767) showed moderate adhesion (5725 and 2740 cells/cm2 min, respectively) (Table 1).
|D. hansenii strains||Adhesion||MATSc||Sedimentation|
|IARb (cells/cm2min)||Chloroform–Hexadecane (%)||Ethyl acetate–Decane (%)||Average cell diameterd (μm)||ESVe (μm/s)|
|D 18335||7325 (±750)A||53.9A||−7.8A||4.44 (±0.12)A||1.19|
|H 2||5725 (±680)B||32.3B||4.9B||3.89 (±0.09)B||0.91|
|CBS 767||2740 (±65)C||13.5C||2.1B||3.22 (±0.09)C||0.62|
|MD 02||790 (±460)D||33.0B||3.7B||3.41 (±0.09)C||0.70|
3.3MATS analysis of the D. hansenii strains
The results from the MATS analysis (Fig. 2) showed large differences of the cell surface hydrophobicity of the four different D. hansenii strains. The type strain (CBS 767) was very hydrophobic, having affinities of approx. 80% to the two non-polar solvents hexadecane and decane (Fig. 2). Contrarily, MD 02 was very hydrophilic, showing affinities below 1% to the non-polar solvents (Fig. 2). The two remaining strains (D 18335 and H 2) had moderate affinities (approx. 40%) to hexadecane and decane (Fig. 2).
Furthermore, all four D. hansenii strains had a much larger tendency to donate an electron from their cell surfaces than to accept an electron, i.e. the (chloroform–hexadecane) values were much higher than the (ethyl acetate–decane) values (Table 1). D 18335 had the highest tendency to donate electrons and was significantly different from H 2 and MD 02 with moderate electron donating capacities (Table 1). H 2 and MD 02 could not be significantly separated from each other, but were both significantly higher than CBS 767 that showed the lowest capacity to donate electrons (Table 1).
3.4Cell size and estimation of sedimentation velocity of the D. hansenii strains
D. hansenii D 18335 had an average cell diameter of 4.44 μm and was significantly larger than the three other strains (Table 1). H 2 had an average cell diameter of 3.89 μm and was significantly larger than the cell diameters of MD 02 (3.41 μm) and CBS 767 (3.22 μm), which could not be significantly separated (Table 1).
Stokes law was used to calculate the estimated sedimentation velocity (ESV) of the four different strains using the average cell diameter (Table 1). The largest strain (D 18335) had an ESV of 1.19 μm/s, whereas the smallest strain (CBS 767) had an ESV of 0.62 μm/s (Table 1). The two remaining strains (H 2 and MD 02) had ESV of 0.91 and 0.70 μm/s, respectively (Table 1).
The four different D. hansenii strains show significantly different IAR when flushing the same cell concentrations over the agarose surface. One strain adheres very poorly, whereas the three other strains are more adhesive. This should render caution towards generalization within species, if adhesion of different yeast species is to be examined and compared in future studies.
The poor adhesion of MD 02 to the agarose surface could be due to its very hydrophilic cell surface, resulting in a pronounced hydration of the cell surface by water molecules in the surrounding buffer, and making it almost impossible for the yeast to bind to the agarose surface. This hypothesis concurs with the fact that hydrophobic groups on bacteria surfaces are believed to exhibit a dehydrating capacity, enabling the removal of the water film between the cell surface and the substratum .
All the examined D. hansenii strains show a much larger ability to donate electrons from their cell surfaces than to accept electrons. This finding agrees with earlier observations in the literature on S. cerevisiae. Apart from the poorly adhesive and hydrophilic strain MD 02, it further is significant that the better a D. hansenii strain is to donate an electron from its surface, the better it adheres to the agarose surface. These results suggest that the ability of a non-hydrophilic D. hansenii cell to donate electrons from its cell surface determines its adhesion to the agarose surface.
An interaction between an electron donor and an electron acceptor is defined as a Lewis acid–base (AB) interaction . AB interactions occur when electronegative atoms bonded to more electropositive atoms draw electrons closer. This results in an electron acceptor, having affinity for an electronegative group with a pair of available electrons, an electron donor . The most common AB forces in aqueous systems are hydrogen-donor/hydrogen-acceptor interactions, i.e. hydrogen bonds . Thus, the interactions mentioned above could be hydrogen bonds between electronegative atoms on the yeast cell surface and hydrogen atoms present in the hydroxyl-groups of agarose. However, this hypothesis needs further investigation.
The moderate adhesion of CBS 767 to the agarose surface could be due to its very hydrophobic (or non-polar) cell surface, resulting in too low an ability to donate electrons from the cell surface. The two remaining strains (D 18335 and H 2) adhere the best and their cell surfaces are moderately hydrophobic. Thus, optimum adhesion of a D. hansenii strain seems to occur when its cell surface is hydrophobic enough not to be almost completely hydrated in the surrounding water and hydrophilic enough to have some electron donor capacity.
Experiments in the literature have shown rather large differences in the amount of protein (10.9% and 15.9%), mannoproteins (28.9% and 32.3%), alkali-soluble glucan (14.5% and 37.4%) and alkali-insoluble glucan (48.4% and 20.9%) present in the cell wall in two different D. hansenii strains . Differences in cell wall composition and especially the ratio of hydrophobic and hydrophilic proteins/mannoproteins on the surface are probably the main explanations for the very different cell surface physicochemical properties and adhesion abilities of the D. hansenii strains in this study. This, however, remains to be elucidated.
Interactions involving electron donation are short-ranged, and a short distance between the surfaces (less than 5 nm) is therefore required before these interactions become operative . Gravity forces control the deposition of particles with a diameter > 0.2 μm  in flow systems with shear rates more than 10 times higher (app. 300 s−1) than those in the present study (app. 30 s−1) . Since the diameter of D. hansenii cells in this study is in the range of 3–5 μm, we would expect the cell size to play a role in depositing the yeast cells on the agarose surface, and thus to create a short distance between the cells and the agarose surface. In fact, there seems to be a significant correlation between the IAR of the three adhesive D. hansenii strains (D 18335, H 2 and CBS 767) and their cell size, i.e. the larger the cell diameter, the larger the ESV, and the higher the IAR. These results confirm the above mentioned hypothesis.
In conclusion, our results show that adhesion of D. hansenii to a solid agarose surface is strain-dependent. This strain dependency may be explained in the following way: gravity seems to initiate contact between the cell and the agarose surface, i.e. a strain consisting of large cells will reach the agarose surface faster, and thus in a higher number, than a smaller strain. Next, the cell surface hydrophobicity seems to determine the adhesive ability of a strain, i.e. strains with non-hydrophilic cell surfaces will adhere much better than strains with hydrophilic cell surfaces. Furthermore, an adhesive strain consisting of cells with moderate surface hydrophobicity and high surface electron donating ability will adhere better than an adhesive strain consisting of cells with high surface hydrophobicity and a low surface electron donating ability.
We thank Dr. Peter Nissen for assisting with image analysis in IGOR Pro, and Søren K. Lillevang (Arla Innovation) for providing yeast strains. This project is a part of the FØTEK programme funded by the Danish Dairy Research Foundation (Danish Dairy Board) and the Danish government.
- 1995) The yeast flora of surface-ripened cheeses. Milchwissenschaft 50, 458–462., (