Food Microbiology 42 (2014) 149e153
Contents lists available at ScienceDirect
Food Microbiology journal homepage: www.elsevier.com/locate/fm
Lag time for germination of Penicillium chrysogenum conidia is induced by temperature shifts Safaa Kalai, Maurice Bensoussan, Philippe Dantigny* Laboratoire des Procédés Alimentaires et Microbiologiques, UMR PAM A 02.102, Agro-Sup Dijon, Université de Bourgogne, 1 Esplanade Erasme, 21000 Dijon, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 November 2013 Received in revised form 12 March 2014 Accepted 22 March 2014 Available online 2 April 2014
In the environment, fungal conidia are subject to transient conditions. In particular, temperature is varying according to day/night periods. All predictive models for germination assume that fungal spores can adapt instantaneously to changes of temperature. The only study that supports this assumption (Gougouli and Koutsoumanis, 2012, Modelling germination of fungal spores at constant and fluctuating temperature conditions. International Journal of Food Microbiology, 152: 153e161) was carried out on Penicillium expansum and Aspergillus niger conidia that, in most cases, already produced germ tubes. In contrast, the present study focuses on temperature shifts applied during the first stages of germination (i.e., before the apparition of the germ tubes). Firstly, germination times were determined in steady state conditions at 10, 15, 20 and 25 C. Secondly, temperature shifts (e.g., up-shifts and down-shifts) were applied at 1/4, 1/2, and 3/4 of germination times, with 5, 10 and 15 C magnitudes. Experiments were carried out in triplicate on Penicillium chrysogenum conidia on Potato Dextrose Agar medium according to a full factorial design. Statistical analysis of the results clearly demonstrated that the assumption of instantaneous adaptation of the conidia should be rejected. Temperature shifts during germination led to an induced lag time or an extended germination time as compared to the experiments conducted ay steady state. The induced lag time was maximized when the amplitude of the shift was equal to 10 C. Interaction between the instant and the direction of the shift was highlighted. A negative lag time was observed for a 15 C down-shift applied at 1/4 of the germination time. This result suggested that at optimal temperature the rate of germination decreased with time, and that the variation of this rate with time depended on temperature. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Temperature Shift Germination Penicillium chrysogenum Predictive mycology
1. Introduction Most studies in predictive mycology have been concerned with the effect of environmental factors, principally water activity, temperature, pH, and modified atmospheres, on fungal germination, fungal growth, and mycotoxin production, at steady-state conditions (Dantigny and Bensoussan, 2013). Unfortunately, very little information on modeling these biological responses under fluctuating conditions is available. The lack of experimental devices allowing automatic monitoring of growth and germination, in addition to the use of solid media, may explain this shortage of experiments carried out under transient conditions (Dantigny and Nanguy, 2009). Steady-state is a very poor assumption in the environment where non constant conditions prevail. For example
* Corresponding author. Tel.: þ33 (0)3 80 77 40 71. E-mail addresses: [email protected], [email protected] (P. Dantigny). http://dx.doi.org/10.1016/j.fm.2014.03.016 0740-0020/Ó 2014 Elsevier Ltd. All rights reserved.
the effects of fluctuating moisture and temperature conditions on growth and viability of fungi in building materials were investigated (Pasanen et al., 2000). It has been reported that moisture conditions on the surface were critical for the development of fungal growth in a material, because fungi grow on the surface, but the medium can also serve as a reservoir of water. Therefore, it appeared that mass transfer phenomena should be taken into account for explaining the experimental results. Sedlbauer (2001) developed a biohygrothermal model to describe the effect of the humidity available at certain temperatures on the germination of fungal spores. The water content in the spore was recalculated every hour by the calculation of the ambient humidity in dependence of transient boundary conditions. When a critical water content was achieved inside the spore, germination was completed. Fluctuating conditions were also examined in the fields of plant pathology. The effect of temperature on the length of the incubation period of rose powdery mildew was studied (Xu, 1999). It was shown that models derived from constant temperature
150
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
experiments, used to predict the development under fluctuating temperatures, gave accurate predictions. The effects of transient temperatures on the growth (Gougouli and Koutsoumanis, 2010) and the germination (Gougouli and Koutsoumanis, 2012) of Aspergillus niger and Penicillium expansum were investigated. The implicit assumptions of the developed models were an instantaneous adaptation of the organisms to changes in temperature. These assumptions seemed correct for growth, although changes in the growth rate or induced lag time for growth are difficult to be detected by means of the available experimental devices. For germination, with the notable exception of only one experiment at 3 h temperature shift from 25 to 1 C. the disturbances were applied shortly before the emission of the germ tubes or when a significant percentage of spores had already germinated. Under these experimental conditions, the germination curves did not exhibit any additional lag time. It is uncertain whether these assumptions hold true when the same disturbances are applied during the other stages of the germination process. Spore germination is a key process common to all fungi. It can be divided into four stages: (i) breaking of spore dormancy/activation, (ii) isotropic swelling/growth, (iii) establishment of cell polarity, and (iv) formation of a germ tube and maintenance of polar growth (d’Enfert, 1997; Wendland, 2001). Oxygen is required for swelling and the process is energy dependent. During swelling, an isotropic increase in the diameter of the spore is observed. The polarization process consists of a delineation of a certain surface area of the isotropically growing cell and subsequent direction of several mechanisms of cell growth to this area including: i) membrane organization ii) cytoskeletal elements, iii) deposition of vesicles that contain cell wall building blocks (Dantigny et al., 2013). For Penicillium and Aspergillus species, under optimal conditions, the isotropic stage typically occurs after 2 h and the first bulging of the cell is observed between 5 and 6 h. In such a case, isotropic growth and polarized growth are somewhat equal. After 8 h most cells have formed a germ tube (Van Leeuwen et al., 2010). In the present work, temperature shifts were applied only prior to germ tube formation (i.e., during swelling or polarization). Firstly, germination times of Penicillium chrysogenum conidia were determined in steady state conditions at 10, 15, 20 and 25 C. Secondly, temperature shifts (e.g., up-shifts and down-shifts) were applied at 1/4, 1/2, and 3/4 of these respective germination times. The magnitude of the shifts was 5, 10 and 15 C. The present study compared the experimental germination times obtained under transients with the theoretical ones based on instantaneous adaptation of the conidia to temperature shifts.
2.3. Germination assessment After counting the conidia on a heamocytometer, the suspensions were standardized to 1 106 spores/ml. The experimental device used in this study was made from a Petri dish as described previously (Sautour et al., 2001). A sterile glass slide (1.8 1.8 cm2) was placed in the Petri dish on a cross bar, 0.5 cm height, to avoid flooding of the slide. A piece (1.5 1.5 0.2 cm3) of PDA medium (0.995 aw) was placed on the slide and inoculated with 10 ml of the standardized suspensions. The devices sealed with ParafilmÒ constituted the closed incubation chambers. Without opening the devices, at least 100 spores (20e25 per microscopic field) were examined through the Petri dish lid every hour (Lattab et al., 2012). Experiments were carried out in triplicate. The length of the germ tubes was measured by means of a Leica DMLB (200) (Leica, RueilMalmaison, France) connected to a IXC 800 (I2S, Pessac, France) camera. Pictures were analyzed using Matrox Inspector 2.2 (Matrox Electronics Systems Ltd, Dorval, Canada). Spores were considered germinated when the length of the germ tubes was greater to equal the greatest dimension of the swollen spore (Dantigny et al., 2006). The asymmetric model (Dantigny et al., 2011):
2
3 1 P ¼ Pmax 41 d 5 1 þ st
(1)
was used to describe the percentage of germinated spores P (%) as a function of time, t (h). Pmax (%) is the asymptotic P value at t / þf, the germination time s(h) is the point at which P ¼ Pmax / 2, and d () is a design parameter. In accordance with a previous workshop dedicated to fungal germination, Dantigny et al. (2006), at the germination time, half the viable spores had germinated. 2.4. Experimental design The germination time at steady-state was determined at 10, 15, 20, and 25 C. For transients, a full factorial design was used to assess the effects of the direction, (i.e., upward, and downward), the instant i, (i.e., 1/4, 1/2, and 3/4 of the germination time), and the magnitude (i.e., 5, 10, and 15 C) of the shifts, Table 1. The initial
Table 1 Full factorial design for assessing the effects of temperature shifts on germination time and induced lag time for Penicillium chrysogenum conidia on Potato Dextrose Agar. Temperature shifts
Germination time (h)
2. Material and methods
Direction
Instant Magnitude Observed ( C)
2.1. Mould and medium
Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift
1/4 1/4 1/2 1/2 3/4 3/4 1/4 1/4 1/2 1/2 3/4 3/4 1/4 1/4 1/2 1/2 3/4 3/4
P. chrysogenum 738 was isolated from baking products and maintained on potato dextrose agar (PDA) medium (bioMérieux, Marcy l’Etoile, France) at room temperature (18e25 C). The media for spore production and spore germination was PDA. The initial pH for all experiments was 5.7 0.1.
2.2. Production of the conidia The plates were spread with 1.5 ml of spore suspension (ca. 1 106 spores/ml) and incubated at 25 C for 7 d. Fresh conidia were harvested by flooding the surface of the plates with 4.5 ml of sterile saline solution (NaCl, 9 g/l of water) containing Tween 80 (0.05% vol/vol; Prolabo, Paris, France).
5 5 5 5 5 5 10 10 10 10 10 10 15 15 15 15 15 15
11.97 14.39 12.45 12.48 13.00 13.19 13.08 20.03 17.03 18.40 18.35 18.00 20.37 31.00 25.74 30.10 36.75 23.20
0.30 0.16 0.16 0.09 0.62 0.36 0.14 0.06 0.06 0.05 0.27 0.35 0.25 0.35 0.28 0.30 0.34 0.35
Theoretical 9.79 12.42 11.10 11.10 12.42 9.79 10.89 15.72 13.30 13.30 15.72 10.89 17.47 35.32 26.45 26.45 35.32 17.47
0.18 0.17 0.18 0.18 0.17 0.18 0.18 0.19 0.19 0.19 0.18 0.18 0.18 0.17 0.19 0.19 0.18 0.17
Induced lag (h)
2.18 0.48 1.97 0.33 1.35 0.34 1.38 0.27 0.58ns 0.79 3.40 0.54 2.19 0.32 4.31 0.25 3.73 0.25 5.10 0.24 2.63 0.45 7.11 0.53 2.90 0.43 4.32 0.52 0.71ns 0.47 3.65 0.49 1.43 0.52 5.73 0.52
Mean standard deviation. ns: non significantly different from 0 at p ¼ 0.05.
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
temperature for downshifts, and the final temperature for upshifts, was 25 C.
151
Table 2 Germination time, s (h), of Penicillium chrysogenum on Potato Dextrose Agar at steady state. Temperature ( C)
2.5. Calculation of the induced lag time
10
The theoretical germination time was calculated according to the hypothesis that conidia adapt instantaneously from an initial temperature Tini, to a final temperature Tfin. Assuming a shift applied at the instant, i (), the theoretical germination time was :
sg theoretical ¼ i$sðTini Þ þ ð1 iÞsðTfin Þ
(2)
where s(Tini) the germination time at steady state for Tini, and s(Tfin) the germination time at steady state for Tfin. The induced lag time, l (h) was calculated as the difference between the observed and the theoretical germination time :
l ¼ sg observed sg theoretical
s (h) 1/4 s (h) 1/2 s (h) 3/4 s (h)
44.3 11.1 22.2 33.2
15
0.20 0.05 0.10 0.15
The effects of the factors on the induced lag time were evaluated by analysis of variance (ANOVA) using Statistica for WindowsÒ (StatSoft France). Significance was evaluated at p ¼ 0.05. 3. Results 3.1. Steady state experiments Steady state experiments shown that all the conidia were viable because 100% germination was reached whatever the temperature, Fig. 1. The optimum temperature for germination was 25 C, then germination time was increased with decreasing temperature from the optimum one to 10 C. Between two consecutive temperatures, in the range 15e25 C, the germination time was increased by about 5 h, Table 2. In contrast, between 15 and 10 C, the germination time was increased by about 25 h. Therefore the impact of reduced temperature on germination time was clearly
Fig. 1. Influence of temperature, (6) 25 C, (B) 20 C, (7) 15 C, and (,) 10 C, on germination of Penicillium chrysogenum conidia on Potato Dextrose Agar, at steady state.
0.21 0.05 0.11 0.16
13.7 3.42 6.85 10.3
25
0.18 0.05 0.09 0.15
8.5 2.12 4.25 6.37
0.17 0.04 0.09 0.13
Mean standard deviation.
Table 3 Analysis of variance of the effects of the direction, the instant and the magnitude of temperature shifts on the induced lag time for germination of Penicillium chrysogenum conidia on Potato Dextrose Agar. Effects
Sum of squares
ddf
Mean square
Total Direction Instant Magnitude Direction Instant Direction Magnitude Instant Magnitude Residuals
208.35 1.51 11.68 42.06 44.38 4.71 31.35 78.74
53 1 2 2 2 2 4 40
208.35 1.51 5.84 21.03 22.19 2.35 7.83 1.96
(3)
2.6. Statistical analysis of the results
18.3 4.52 9.05 13.6
20
F
p
0.771 2.97 10.7 11.3 1.20 3.98
0.385 0.0628 0.000191* 0.000131* 0.313 0.00821*
*Significant at p < 0.05.
demonstrated at 10 C. The use of the asymmetric model for germination allowed very accurate estimations of the germination time, the standard deviation was ca. 0.20 h, Table 2. It is pointed out that, for all the temperatures, none of the spores had already germinated at 3/4 of the germination time. For example, at 10 C, the percentage of germination was nil, Fig. 1, at 33.2 h, Table 2. 3.2. Transient experiments In 16 cases out of 18 experiments, a significant induced lag time was exhibited, Table 1. Therefore the hypothesis of an instantaneous adaptation of germination conidia of P. chrysogenum should be rejected. The lag time did not differ significantly from zero for two up-shifts experiments, but the instant and the magnitude of the shifts were different. The analysis of variance showed a significant effect of the magnitude of the shift, Table 3. The instant of the shift was also of paramount importance, due to its interactions with the two other factors. For example, downshifts experiments were characterized by the greater, l ¼ 7.11 h and the smaller, l ¼ 4.32 h, induced lag times for 3/4, 10 C amplitude, and 1/4, 15 C amplitude, respectively, Table 1. The interactions between the direction and the instant of the shifts were evaluated by merging the results obtained for the different magnitudes. Because magnitude was a significant factor, the standard deviations were important, Fig. 2. The induced lag time was almost constant at about 2 h for upshifts experiments. In contrast, there is a trend of an increase of the induced lag time with increasing the instant of the shift. Interactions between the magnitude and the instant of the shifts showed that the induced lag times were greater at 10 C than at 5 C, Fig. 3. At these temperatures, the effect of the instant of the shift was not marked. In contrast, a decrease in the induced lag time was observed for a 15 C magnitude with decreasing the instant of the shift. In particular, at 1/4, the induced lag time was negative, i.e., the observed germination time was less than the theoretical one. Germination can be defined as a succession of biochemical reactions that occur within the conidia during activation, swelling, polarization, and germ tube formation, thus eventually leading to
152
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
Fig. 2. Combined effects of the direction and the instant of temperature shifts on the induced lag time for germination of P. chrysogenum on Potato Dextrose Agar. error bars: standard deviation.
Fig. 3. Combined effects of the instant and the magnitude of temperature shifts on the induced lag time for germination of P. chrysogenum on Potato Dextrose Agar. Error bars: standard deviation.
Fig. 4. Schematic relationship between the rate of germination, and the germination time for linear (solid line) and non-linear (broken line) assumptions. The amount of germination is determined by the area between the curve and the x-axis. For a germination time equal to 1, i.e., area equal to 8 squares, germination is completed.
was applied at 1/4 at 25 C, 3.5 biochemical reactions were already done, thus 4.5 more biochemical reactions were needed to complete germination. At 10 C, assuming a linear rate, a little bit more than half the biochemical reactions (i.e. 4.5/8) at 10 C was required to finish germination. This remaining time, 24.9 h was less than the theoretical time, 33.3 h, based implicitly on linear rates of germination. However, the observed induced lag time, 4.3 h was greater than the expected one with this approach, 24.9e33.3 ¼ 8.4 h. In fact, during the shift from 25 to 10 C, an induced lag time would have occur, but was overcome by the time saved by moving from a non-linear curve to a linear one. From the physiological point of view, 2 h at 25 C were equivalent to about 20 h at 10 C. In contrast, no negative lag time was observed for downshift experiments, from 25 C, to 20 and 15 C. This can be explained by a smaller magnitude of the shift. For example the difference between the steady-state germination times at 25 and 15 C, was about 10 h only, as compared to a difference of 33.3 h between 25 and 10 C. In addition the profiles of the germination rate at 20, and 15 C remained unknown. Downshift experiments carried out from 20, and 15 C, to 10 C would provide useful information for future modeling of the induced lag time due to temperature shifts. 4. Discussion
germination. The rate of germination defined as the number of biochemical reactions per time throughout the germination process, should not be confused with the percentage of germination that increases from 0 to 100% (when all spores are viable) during the last stage of germination. The rate of germination, arbitrary units, was plotted against the germination time, Fig. 4. When the germination time is equal to 1, half the viable spores had completed germination, but of course this time depends on temperature. A linear and a non-linear rates of germination are shown in solid line, and dotted line, respectively (Fig. 4). The linear rate was characterized by “two” biochemical reactions per quarter of the germination time. The number of biochemical reactions is evaluated by integrating the rate of germination with respect to the germination time. Whatever the rate, linear or not, “eight” biochemical reactions (i.e. squares) are needed for germination to be completed. Let’s consider non-linear, and linear rates of germination at 25 C, and 10 C, respectively. In our experiment, when the downward shift
This study demonstrated the significant, either main or combined, effects of the direction, the instant, and the magnitude of temperature shifts on the induced lag time for germination. Experiments were carried out on synthetic medium at 0.995 aw. Therefore in these conditions the lag time (h) was negligible compared to the shelf-life of bakery products. But the lag time, expressed in percentage, was in the range 12% to 65% of the germination time. Therefore assuming the same range for the lag time for conidia contaminating bakery products (0.91e0.93 aw for example) the effect on the shelf-life can be significant. The effect of the direction of the shift was already mentioned for bacteria. Mellefont and Ross (2003), reported that temperature downshifts induced larger relative lag times, than equivalent upshifts. The same authors also reported the importance of the physiological state of the organisms when shifts are applied. This maybe explain why the effect of the instant of the shift was significant, although in
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
combination with the other factors. In fact, the physiological state of the germinating conidia depends on the phase of germination. In the present study, microscopic observations have shown that the diameter of the conidia was greater than that of the native spores, and that no germ tube was formed. Therefore, shifts were applied when conidia were either in the swelling or in the polarization phase. However, a microscopic observation was not sufficient to determine precisely the stage of the conidia during germination. Other tools, such as molecular biology, should be used in future studies to assess precisely during which phase shifts are applied. During germination of A. niger conidia at optimum temperature, the number of genes up/down regulated differed depending on the phase of germination, i.e., 917/1986, 856/290, 476/297, and 790/179, for 2, 4, 6, and 8 h, respectively (Van Leeuwen et al., 2013). This result suggested that germination is probably not a linear process because many more genes were up/down regulated in the first hours of germination, than in the last ones. This assumption allows a possible explanation for the negative lag time observed in the present study for a downshift from 25 to 10 C, at 1/4. Gougouli and Koutsoumanis (2012) applied a temperature shift to P. expansum conidia at 3 h, from 25 to 1 C. A careful examination of their results showed that the predicted germination time (i.e., time at which 50% of the conidia had germinated) was about 120 h, whereas only 20% of the conidia had effectively germinated at that time. Therefore, the observed time at which 50% of the conidia had germinated was greater than 120 h. This is apparently contradictory to the assumption of a non-linear germination rate at optimum temperature, that would lead to an observed germination time less than the theoretical one. But the instant of the shift, 0.03 (3 h/120 h) as compared to 1/4, the magnitude of the shift, 24 C as compared to 15 C, and the species were different in the study of Gougouli and Koutsoumanis (2012) and in the present study. 5. Conclusions The present study demonstrated that temperature shifts applied before the apparition of the germ tubes induced lag time for germination. In contrast to previous studies that only compared the complete observed and predicted germination curves, the present study was based on the calculation of theoretical and observed germination times, thus may be explaining the different conclusions of the studies. Moreover, the effects of the shifts on the induced lag time depended greatly on the physiological state of the conidia, i.e., the phase of germination. Combined effects of the direction, the instant, and the magnitude of the shifts were highlighted. The results obtained suggested that under optimum temperature, the rate of germination is greater at the beginning than at the end of the process. It is also suggested that the relationship between the rate of germination and the time, is temperature dependent. Many assumptions were made as an attempt
153
to explain the experimental results. These assumptions should now be verified by molecular biology techniques, in particular to discriminate the different phases of germination, and to elucidate the temperature dependency of the germination rate with time. The application of this study would improve knowledge of germination kinetics during transients, and also optimize the production of pre-germinated starters. References Dantigny, P., Bensoussan, M., Vasseur, V., Lebrihi, A., Buchet, C., Ismaili-Alaoui, M., Devlieghere, F., Rousssos, S., 2006. Standardisation of methods for assessing mould germination: a workshop report. Int. J. Food Microbiol. 108, 286e291. Dantigny, P., Nanguy, S.P.-M., 2009. Significance of the physiological state of fungal spores. Int. J. Food Microbiol. 134, 16e20. Dantigny, P., Nanguy, S.,P.-M., Judet-Correia, D., Bensoussan, M., 2011. A new germination model for fungi. Int. J. Food Microbiol. 146, 176e181. Dantigny, P., Bensoussan, M., 2013. Introduction to predictive mycology. In: Dantigny, P., Panagou, E.Z. (Eds.), Predictive Mycology. Nova Science Publishers, New York, pp. 1e26. Dantigny, P., Kalai, S., Nanguy, S.P.-M., 2013. Primary models for germination of fungal spores. In: Dantigny, P., Panagou, E.Z. (Eds.), Predictive Mycology. Nova Science Publishers, New York, pp. 47e62. d’Enfert, C., 1997. Fungal spore germination: insights from the molecular genetics of Aspergillus nidulans and Neurospora crassa. Fungal Genet. Biol. 21, 163e172. Gougouli, M., Koutsoumanis, K.P., 2010. Modelling growth of Penicillium expansum and Aspergillus niger at constant and fluctuating temperature conditions. Int. J. Food Microbiol. 140, 254e262. Gougouli, M., Koutsoumanis, K.P., 2012. Modelling germination of Penicillium expansum and Aspergillus niger at constant and fluctuating temperature conditions. Int. J. Food Microbiol. 152, 153e161. Lattab, N., Kalai, S., Bensoussan, M., Dantigny, P., 2012. Effect of storage conditions (relative humidity, duration, and temperature) on the germination time of Aspergillus carbonarius and Penicillium chrysogenum. Int. J. Food Microbiol. 160, 80e84. Mellefont, L.A., Ross, T., 2003. The effect of abrupt shifts in temperature on the lag phase duration of Escherichia coli and Klebsiella oxytoca. Int. J. Food Microbiol. 83, 295e305. Pasanen, A.-L., Kasanen, J.-P., Rautiala, S., Ikäheimo, M., Rantamäki, J., Kääriäinen, H., Kalliokoski, P., 2000. Fungal growth and survival in building materials under fluctuating moisture and temperature conditions. Int. Biodeterior. Biodegrad. 46, 117e127. Sautour, M., Dantigny, P., Divies, C., Bensoussan, M., 2001. Application of Doehlert design to determine the combined effects of temperature, water activity and pH on conidial germination of Penicillium chrysogenum. J. Appl. Microbiol. 91, 900e 906. Sedlbauer, K., 2001. Prediction of Mould Fungus Formation on the Surface of and Inside Building Components. PhD dissertation. Department of Building Physics, University of Stuttgart, Stuttgart, Germany, 247 p. (accessed 05.11.10.) at http://www.ibp.fraunhofer.de/content/dam/ibp/en/documents/ ks_dissertation_etcm1021-30729.pdf. Van Leeuwen, M.R., Van Doorn, T.M., Golovina, E.A., Stark, J., Dijksterhuis, J., 2010. Water- and air-distributed conidia differ in sterol content and cytoplasmic microviscosity. Appl. Environ. Microbiol. 76, 366e369. Van Leeuwen, M.R., Krijgsheld, P., Wyatt, T.T., Golovina, E.A., Menke, H., Dekker, A., Stark, J., Stam, H., Bleichrodt, R., Wösten, H.A.B., Dijksterhuis, J., 2013. The effect of natamycin on the transcriptome of conidia of Aspergillus niger. Stud. Mycol. 74, 71e85. Wendland, J., 2001. Comparison of morphogenetic networks of filamentous fungi and yeast. Fungal Genet. Biol. 34, 63e82. Xu, X.-M., 1999. Effects of temperature on the length of the incubation period of rose powdery mildew (Spaerotheca pannosa var. rosae). Eur. J. Plant Pathol. 105, 13e21.
Contents lists available at ScienceDirect
Food Microbiology journal homepage: www.elsevier.com/locate/fm
Lag time for germination of Penicillium chrysogenum conidia is induced by temperature shifts Safaa Kalai, Maurice Bensoussan, Philippe Dantigny* Laboratoire des Procédés Alimentaires et Microbiologiques, UMR PAM A 02.102, Agro-Sup Dijon, Université de Bourgogne, 1 Esplanade Erasme, 21000 Dijon, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 November 2013 Received in revised form 12 March 2014 Accepted 22 March 2014 Available online 2 April 2014
In the environment, fungal conidia are subject to transient conditions. In particular, temperature is varying according to day/night periods. All predictive models for germination assume that fungal spores can adapt instantaneously to changes of temperature. The only study that supports this assumption (Gougouli and Koutsoumanis, 2012, Modelling germination of fungal spores at constant and fluctuating temperature conditions. International Journal of Food Microbiology, 152: 153e161) was carried out on Penicillium expansum and Aspergillus niger conidia that, in most cases, already produced germ tubes. In contrast, the present study focuses on temperature shifts applied during the first stages of germination (i.e., before the apparition of the germ tubes). Firstly, germination times were determined in steady state conditions at 10, 15, 20 and 25 C. Secondly, temperature shifts (e.g., up-shifts and down-shifts) were applied at 1/4, 1/2, and 3/4 of germination times, with 5, 10 and 15 C magnitudes. Experiments were carried out in triplicate on Penicillium chrysogenum conidia on Potato Dextrose Agar medium according to a full factorial design. Statistical analysis of the results clearly demonstrated that the assumption of instantaneous adaptation of the conidia should be rejected. Temperature shifts during germination led to an induced lag time or an extended germination time as compared to the experiments conducted ay steady state. The induced lag time was maximized when the amplitude of the shift was equal to 10 C. Interaction between the instant and the direction of the shift was highlighted. A negative lag time was observed for a 15 C down-shift applied at 1/4 of the germination time. This result suggested that at optimal temperature the rate of germination decreased with time, and that the variation of this rate with time depended on temperature. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Temperature Shift Germination Penicillium chrysogenum Predictive mycology
1. Introduction Most studies in predictive mycology have been concerned with the effect of environmental factors, principally water activity, temperature, pH, and modified atmospheres, on fungal germination, fungal growth, and mycotoxin production, at steady-state conditions (Dantigny and Bensoussan, 2013). Unfortunately, very little information on modeling these biological responses under fluctuating conditions is available. The lack of experimental devices allowing automatic monitoring of growth and germination, in addition to the use of solid media, may explain this shortage of experiments carried out under transient conditions (Dantigny and Nanguy, 2009). Steady-state is a very poor assumption in the environment where non constant conditions prevail. For example
* Corresponding author. Tel.: þ33 (0)3 80 77 40 71. E-mail addresses: [email protected], [email protected] (P. Dantigny). http://dx.doi.org/10.1016/j.fm.2014.03.016 0740-0020/Ó 2014 Elsevier Ltd. All rights reserved.
the effects of fluctuating moisture and temperature conditions on growth and viability of fungi in building materials were investigated (Pasanen et al., 2000). It has been reported that moisture conditions on the surface were critical for the development of fungal growth in a material, because fungi grow on the surface, but the medium can also serve as a reservoir of water. Therefore, it appeared that mass transfer phenomena should be taken into account for explaining the experimental results. Sedlbauer (2001) developed a biohygrothermal model to describe the effect of the humidity available at certain temperatures on the germination of fungal spores. The water content in the spore was recalculated every hour by the calculation of the ambient humidity in dependence of transient boundary conditions. When a critical water content was achieved inside the spore, germination was completed. Fluctuating conditions were also examined in the fields of plant pathology. The effect of temperature on the length of the incubation period of rose powdery mildew was studied (Xu, 1999). It was shown that models derived from constant temperature
150
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
experiments, used to predict the development under fluctuating temperatures, gave accurate predictions. The effects of transient temperatures on the growth (Gougouli and Koutsoumanis, 2010) and the germination (Gougouli and Koutsoumanis, 2012) of Aspergillus niger and Penicillium expansum were investigated. The implicit assumptions of the developed models were an instantaneous adaptation of the organisms to changes in temperature. These assumptions seemed correct for growth, although changes in the growth rate or induced lag time for growth are difficult to be detected by means of the available experimental devices. For germination, with the notable exception of only one experiment at 3 h temperature shift from 25 to 1 C. the disturbances were applied shortly before the emission of the germ tubes or when a significant percentage of spores had already germinated. Under these experimental conditions, the germination curves did not exhibit any additional lag time. It is uncertain whether these assumptions hold true when the same disturbances are applied during the other stages of the germination process. Spore germination is a key process common to all fungi. It can be divided into four stages: (i) breaking of spore dormancy/activation, (ii) isotropic swelling/growth, (iii) establishment of cell polarity, and (iv) formation of a germ tube and maintenance of polar growth (d’Enfert, 1997; Wendland, 2001). Oxygen is required for swelling and the process is energy dependent. During swelling, an isotropic increase in the diameter of the spore is observed. The polarization process consists of a delineation of a certain surface area of the isotropically growing cell and subsequent direction of several mechanisms of cell growth to this area including: i) membrane organization ii) cytoskeletal elements, iii) deposition of vesicles that contain cell wall building blocks (Dantigny et al., 2013). For Penicillium and Aspergillus species, under optimal conditions, the isotropic stage typically occurs after 2 h and the first bulging of the cell is observed between 5 and 6 h. In such a case, isotropic growth and polarized growth are somewhat equal. After 8 h most cells have formed a germ tube (Van Leeuwen et al., 2010). In the present work, temperature shifts were applied only prior to germ tube formation (i.e., during swelling or polarization). Firstly, germination times of Penicillium chrysogenum conidia were determined in steady state conditions at 10, 15, 20 and 25 C. Secondly, temperature shifts (e.g., up-shifts and down-shifts) were applied at 1/4, 1/2, and 3/4 of these respective germination times. The magnitude of the shifts was 5, 10 and 15 C. The present study compared the experimental germination times obtained under transients with the theoretical ones based on instantaneous adaptation of the conidia to temperature shifts.
2.3. Germination assessment After counting the conidia on a heamocytometer, the suspensions were standardized to 1 106 spores/ml. The experimental device used in this study was made from a Petri dish as described previously (Sautour et al., 2001). A sterile glass slide (1.8 1.8 cm2) was placed in the Petri dish on a cross bar, 0.5 cm height, to avoid flooding of the slide. A piece (1.5 1.5 0.2 cm3) of PDA medium (0.995 aw) was placed on the slide and inoculated with 10 ml of the standardized suspensions. The devices sealed with ParafilmÒ constituted the closed incubation chambers. Without opening the devices, at least 100 spores (20e25 per microscopic field) were examined through the Petri dish lid every hour (Lattab et al., 2012). Experiments were carried out in triplicate. The length of the germ tubes was measured by means of a Leica DMLB (200) (Leica, RueilMalmaison, France) connected to a IXC 800 (I2S, Pessac, France) camera. Pictures were analyzed using Matrox Inspector 2.2 (Matrox Electronics Systems Ltd, Dorval, Canada). Spores were considered germinated when the length of the germ tubes was greater to equal the greatest dimension of the swollen spore (Dantigny et al., 2006). The asymmetric model (Dantigny et al., 2011):
2
3 1 P ¼ Pmax 41 d 5 1 þ st
(1)
was used to describe the percentage of germinated spores P (%) as a function of time, t (h). Pmax (%) is the asymptotic P value at t / þf, the germination time s(h) is the point at which P ¼ Pmax / 2, and d () is a design parameter. In accordance with a previous workshop dedicated to fungal germination, Dantigny et al. (2006), at the germination time, half the viable spores had germinated. 2.4. Experimental design The germination time at steady-state was determined at 10, 15, 20, and 25 C. For transients, a full factorial design was used to assess the effects of the direction, (i.e., upward, and downward), the instant i, (i.e., 1/4, 1/2, and 3/4 of the germination time), and the magnitude (i.e., 5, 10, and 15 C) of the shifts, Table 1. The initial
Table 1 Full factorial design for assessing the effects of temperature shifts on germination time and induced lag time for Penicillium chrysogenum conidia on Potato Dextrose Agar. Temperature shifts
Germination time (h)
2. Material and methods
Direction
Instant Magnitude Observed ( C)
2.1. Mould and medium
Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift Upshift Downshift
1/4 1/4 1/2 1/2 3/4 3/4 1/4 1/4 1/2 1/2 3/4 3/4 1/4 1/4 1/2 1/2 3/4 3/4
P. chrysogenum 738 was isolated from baking products and maintained on potato dextrose agar (PDA) medium (bioMérieux, Marcy l’Etoile, France) at room temperature (18e25 C). The media for spore production and spore germination was PDA. The initial pH for all experiments was 5.7 0.1.
2.2. Production of the conidia The plates were spread with 1.5 ml of spore suspension (ca. 1 106 spores/ml) and incubated at 25 C for 7 d. Fresh conidia were harvested by flooding the surface of the plates with 4.5 ml of sterile saline solution (NaCl, 9 g/l of water) containing Tween 80 (0.05% vol/vol; Prolabo, Paris, France).
5 5 5 5 5 5 10 10 10 10 10 10 15 15 15 15 15 15
11.97 14.39 12.45 12.48 13.00 13.19 13.08 20.03 17.03 18.40 18.35 18.00 20.37 31.00 25.74 30.10 36.75 23.20
0.30 0.16 0.16 0.09 0.62 0.36 0.14 0.06 0.06 0.05 0.27 0.35 0.25 0.35 0.28 0.30 0.34 0.35
Theoretical 9.79 12.42 11.10 11.10 12.42 9.79 10.89 15.72 13.30 13.30 15.72 10.89 17.47 35.32 26.45 26.45 35.32 17.47
0.18 0.17 0.18 0.18 0.17 0.18 0.18 0.19 0.19 0.19 0.18 0.18 0.18 0.17 0.19 0.19 0.18 0.17
Induced lag (h)
2.18 0.48 1.97 0.33 1.35 0.34 1.38 0.27 0.58ns 0.79 3.40 0.54 2.19 0.32 4.31 0.25 3.73 0.25 5.10 0.24 2.63 0.45 7.11 0.53 2.90 0.43 4.32 0.52 0.71ns 0.47 3.65 0.49 1.43 0.52 5.73 0.52
Mean standard deviation. ns: non significantly different from 0 at p ¼ 0.05.
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
temperature for downshifts, and the final temperature for upshifts, was 25 C.
151
Table 2 Germination time, s (h), of Penicillium chrysogenum on Potato Dextrose Agar at steady state. Temperature ( C)
2.5. Calculation of the induced lag time
10
The theoretical germination time was calculated according to the hypothesis that conidia adapt instantaneously from an initial temperature Tini, to a final temperature Tfin. Assuming a shift applied at the instant, i (), the theoretical germination time was :
sg theoretical ¼ i$sðTini Þ þ ð1 iÞsðTfin Þ
(2)
where s(Tini) the germination time at steady state for Tini, and s(Tfin) the germination time at steady state for Tfin. The induced lag time, l (h) was calculated as the difference between the observed and the theoretical germination time :
l ¼ sg observed sg theoretical
s (h) 1/4 s (h) 1/2 s (h) 3/4 s (h)
44.3 11.1 22.2 33.2
15
0.20 0.05 0.10 0.15
The effects of the factors on the induced lag time were evaluated by analysis of variance (ANOVA) using Statistica for WindowsÒ (StatSoft France). Significance was evaluated at p ¼ 0.05. 3. Results 3.1. Steady state experiments Steady state experiments shown that all the conidia were viable because 100% germination was reached whatever the temperature, Fig. 1. The optimum temperature for germination was 25 C, then germination time was increased with decreasing temperature from the optimum one to 10 C. Between two consecutive temperatures, in the range 15e25 C, the germination time was increased by about 5 h, Table 2. In contrast, between 15 and 10 C, the germination time was increased by about 25 h. Therefore the impact of reduced temperature on germination time was clearly
Fig. 1. Influence of temperature, (6) 25 C, (B) 20 C, (7) 15 C, and (,) 10 C, on germination of Penicillium chrysogenum conidia on Potato Dextrose Agar, at steady state.
0.21 0.05 0.11 0.16
13.7 3.42 6.85 10.3
25
0.18 0.05 0.09 0.15
8.5 2.12 4.25 6.37
0.17 0.04 0.09 0.13
Mean standard deviation.
Table 3 Analysis of variance of the effects of the direction, the instant and the magnitude of temperature shifts on the induced lag time for germination of Penicillium chrysogenum conidia on Potato Dextrose Agar. Effects
Sum of squares
ddf
Mean square
Total Direction Instant Magnitude Direction Instant Direction Magnitude Instant Magnitude Residuals
208.35 1.51 11.68 42.06 44.38 4.71 31.35 78.74
53 1 2 2 2 2 4 40
208.35 1.51 5.84 21.03 22.19 2.35 7.83 1.96
(3)
2.6. Statistical analysis of the results
18.3 4.52 9.05 13.6
20
F
p
0.771 2.97 10.7 11.3 1.20 3.98
0.385 0.0628 0.000191* 0.000131* 0.313 0.00821*
*Significant at p < 0.05.
demonstrated at 10 C. The use of the asymmetric model for germination allowed very accurate estimations of the germination time, the standard deviation was ca. 0.20 h, Table 2. It is pointed out that, for all the temperatures, none of the spores had already germinated at 3/4 of the germination time. For example, at 10 C, the percentage of germination was nil, Fig. 1, at 33.2 h, Table 2. 3.2. Transient experiments In 16 cases out of 18 experiments, a significant induced lag time was exhibited, Table 1. Therefore the hypothesis of an instantaneous adaptation of germination conidia of P. chrysogenum should be rejected. The lag time did not differ significantly from zero for two up-shifts experiments, but the instant and the magnitude of the shifts were different. The analysis of variance showed a significant effect of the magnitude of the shift, Table 3. The instant of the shift was also of paramount importance, due to its interactions with the two other factors. For example, downshifts experiments were characterized by the greater, l ¼ 7.11 h and the smaller, l ¼ 4.32 h, induced lag times for 3/4, 10 C amplitude, and 1/4, 15 C amplitude, respectively, Table 1. The interactions between the direction and the instant of the shifts were evaluated by merging the results obtained for the different magnitudes. Because magnitude was a significant factor, the standard deviations were important, Fig. 2. The induced lag time was almost constant at about 2 h for upshifts experiments. In contrast, there is a trend of an increase of the induced lag time with increasing the instant of the shift. Interactions between the magnitude and the instant of the shifts showed that the induced lag times were greater at 10 C than at 5 C, Fig. 3. At these temperatures, the effect of the instant of the shift was not marked. In contrast, a decrease in the induced lag time was observed for a 15 C magnitude with decreasing the instant of the shift. In particular, at 1/4, the induced lag time was negative, i.e., the observed germination time was less than the theoretical one. Germination can be defined as a succession of biochemical reactions that occur within the conidia during activation, swelling, polarization, and germ tube formation, thus eventually leading to
152
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
Fig. 2. Combined effects of the direction and the instant of temperature shifts on the induced lag time for germination of P. chrysogenum on Potato Dextrose Agar. error bars: standard deviation.
Fig. 3. Combined effects of the instant and the magnitude of temperature shifts on the induced lag time for germination of P. chrysogenum on Potato Dextrose Agar. Error bars: standard deviation.
Fig. 4. Schematic relationship between the rate of germination, and the germination time for linear (solid line) and non-linear (broken line) assumptions. The amount of germination is determined by the area between the curve and the x-axis. For a germination time equal to 1, i.e., area equal to 8 squares, germination is completed.
was applied at 1/4 at 25 C, 3.5 biochemical reactions were already done, thus 4.5 more biochemical reactions were needed to complete germination. At 10 C, assuming a linear rate, a little bit more than half the biochemical reactions (i.e. 4.5/8) at 10 C was required to finish germination. This remaining time, 24.9 h was less than the theoretical time, 33.3 h, based implicitly on linear rates of germination. However, the observed induced lag time, 4.3 h was greater than the expected one with this approach, 24.9e33.3 ¼ 8.4 h. In fact, during the shift from 25 to 10 C, an induced lag time would have occur, but was overcome by the time saved by moving from a non-linear curve to a linear one. From the physiological point of view, 2 h at 25 C were equivalent to about 20 h at 10 C. In contrast, no negative lag time was observed for downshift experiments, from 25 C, to 20 and 15 C. This can be explained by a smaller magnitude of the shift. For example the difference between the steady-state germination times at 25 and 15 C, was about 10 h only, as compared to a difference of 33.3 h between 25 and 10 C. In addition the profiles of the germination rate at 20, and 15 C remained unknown. Downshift experiments carried out from 20, and 15 C, to 10 C would provide useful information for future modeling of the induced lag time due to temperature shifts. 4. Discussion
germination. The rate of germination defined as the number of biochemical reactions per time throughout the germination process, should not be confused with the percentage of germination that increases from 0 to 100% (when all spores are viable) during the last stage of germination. The rate of germination, arbitrary units, was plotted against the germination time, Fig. 4. When the germination time is equal to 1, half the viable spores had completed germination, but of course this time depends on temperature. A linear and a non-linear rates of germination are shown in solid line, and dotted line, respectively (Fig. 4). The linear rate was characterized by “two” biochemical reactions per quarter of the germination time. The number of biochemical reactions is evaluated by integrating the rate of germination with respect to the germination time. Whatever the rate, linear or not, “eight” biochemical reactions (i.e. squares) are needed for germination to be completed. Let’s consider non-linear, and linear rates of germination at 25 C, and 10 C, respectively. In our experiment, when the downward shift
This study demonstrated the significant, either main or combined, effects of the direction, the instant, and the magnitude of temperature shifts on the induced lag time for germination. Experiments were carried out on synthetic medium at 0.995 aw. Therefore in these conditions the lag time (h) was negligible compared to the shelf-life of bakery products. But the lag time, expressed in percentage, was in the range 12% to 65% of the germination time. Therefore assuming the same range for the lag time for conidia contaminating bakery products (0.91e0.93 aw for example) the effect on the shelf-life can be significant. The effect of the direction of the shift was already mentioned for bacteria. Mellefont and Ross (2003), reported that temperature downshifts induced larger relative lag times, than equivalent upshifts. The same authors also reported the importance of the physiological state of the organisms when shifts are applied. This maybe explain why the effect of the instant of the shift was significant, although in
S. Kalai et al. / Food Microbiology 42 (2014) 149e153
combination with the other factors. In fact, the physiological state of the germinating conidia depends on the phase of germination. In the present study, microscopic observations have shown that the diameter of the conidia was greater than that of the native spores, and that no germ tube was formed. Therefore, shifts were applied when conidia were either in the swelling or in the polarization phase. However, a microscopic observation was not sufficient to determine precisely the stage of the conidia during germination. Other tools, such as molecular biology, should be used in future studies to assess precisely during which phase shifts are applied. During germination of A. niger conidia at optimum temperature, the number of genes up/down regulated differed depending on the phase of germination, i.e., 917/1986, 856/290, 476/297, and 790/179, for 2, 4, 6, and 8 h, respectively (Van Leeuwen et al., 2013). This result suggested that germination is probably not a linear process because many more genes were up/down regulated in the first hours of germination, than in the last ones. This assumption allows a possible explanation for the negative lag time observed in the present study for a downshift from 25 to 10 C, at 1/4. Gougouli and Koutsoumanis (2012) applied a temperature shift to P. expansum conidia at 3 h, from 25 to 1 C. A careful examination of their results showed that the predicted germination time (i.e., time at which 50% of the conidia had germinated) was about 120 h, whereas only 20% of the conidia had effectively germinated at that time. Therefore, the observed time at which 50% of the conidia had germinated was greater than 120 h. This is apparently contradictory to the assumption of a non-linear germination rate at optimum temperature, that would lead to an observed germination time less than the theoretical one. But the instant of the shift, 0.03 (3 h/120 h) as compared to 1/4, the magnitude of the shift, 24 C as compared to 15 C, and the species were different in the study of Gougouli and Koutsoumanis (2012) and in the present study. 5. Conclusions The present study demonstrated that temperature shifts applied before the apparition of the germ tubes induced lag time for germination. In contrast to previous studies that only compared the complete observed and predicted germination curves, the present study was based on the calculation of theoretical and observed germination times, thus may be explaining the different conclusions of the studies. Moreover, the effects of the shifts on the induced lag time depended greatly on the physiological state of the conidia, i.e., the phase of germination. Combined effects of the direction, the instant, and the magnitude of the shifts were highlighted. The results obtained suggested that under optimum temperature, the rate of germination is greater at the beginning than at the end of the process. It is also suggested that the relationship between the rate of germination and the time, is temperature dependent. Many assumptions were made as an attempt
153
to explain the experimental results. These assumptions should now be verified by molecular biology techniques, in particular to discriminate the different phases of germination, and to elucidate the temperature dependency of the germination rate with time. The application of this study would improve knowledge of germination kinetics during transients, and also optimize the production of pre-germinated starters. References Dantigny, P., Bensoussan, M., Vasseur, V., Lebrihi, A., Buchet, C., Ismaili-Alaoui, M., Devlieghere, F., Rousssos, S., 2006. Standardisation of methods for assessing mould germination: a workshop report. Int. J. Food Microbiol. 108, 286e291. Dantigny, P., Nanguy, S.P.-M., 2009. Significance of the physiological state of fungal spores. Int. J. Food Microbiol. 134, 16e20. Dantigny, P., Nanguy, S.,P.-M., Judet-Correia, D., Bensoussan, M., 2011. A new germination model for fungi. Int. J. Food Microbiol. 146, 176e181. Dantigny, P., Bensoussan, M., 2013. Introduction to predictive mycology. In: Dantigny, P., Panagou, E.Z. (Eds.), Predictive Mycology. Nova Science Publishers, New York, pp. 1e26. Dantigny, P., Kalai, S., Nanguy, S.P.-M., 2013. Primary models for germination of fungal spores. In: Dantigny, P., Panagou, E.Z. (Eds.), Predictive Mycology. Nova Science Publishers, New York, pp. 47e62. d’Enfert, C., 1997. Fungal spore germination: insights from the molecular genetics of Aspergillus nidulans and Neurospora crassa. Fungal Genet. Biol. 21, 163e172. Gougouli, M., Koutsoumanis, K.P., 2010. Modelling growth of Penicillium expansum and Aspergillus niger at constant and fluctuating temperature conditions. Int. J. Food Microbiol. 140, 254e262. Gougouli, M., Koutsoumanis, K.P., 2012. Modelling germination of Penicillium expansum and Aspergillus niger at constant and fluctuating temperature conditions. Int. J. Food Microbiol. 152, 153e161. Lattab, N., Kalai, S., Bensoussan, M., Dantigny, P., 2012. Effect of storage conditions (relative humidity, duration, and temperature) on the germination time of Aspergillus carbonarius and Penicillium chrysogenum. Int. J. Food Microbiol. 160, 80e84. Mellefont, L.A., Ross, T., 2003. The effect of abrupt shifts in temperature on the lag phase duration of Escherichia coli and Klebsiella oxytoca. Int. J. Food Microbiol. 83, 295e305. Pasanen, A.-L., Kasanen, J.-P., Rautiala, S., Ikäheimo, M., Rantamäki, J., Kääriäinen, H., Kalliokoski, P., 2000. Fungal growth and survival in building materials under fluctuating moisture and temperature conditions. Int. Biodeterior. Biodegrad. 46, 117e127. Sautour, M., Dantigny, P., Divies, C., Bensoussan, M., 2001. Application of Doehlert design to determine the combined effects of temperature, water activity and pH on conidial germination of Penicillium chrysogenum. J. Appl. Microbiol. 91, 900e 906. Sedlbauer, K., 2001. Prediction of Mould Fungus Formation on the Surface of and Inside Building Components. PhD dissertation. Department of Building Physics, University of Stuttgart, Stuttgart, Germany, 247 p. (accessed 05.11.10.) at http://www.ibp.fraunhofer.de/content/dam/ibp/en/documents/ ks_dissertation_etcm1021-30729.pdf. Van Leeuwen, M.R., Van Doorn, T.M., Golovina, E.A., Stark, J., Dijksterhuis, J., 2010. Water- and air-distributed conidia differ in sterol content and cytoplasmic microviscosity. Appl. Environ. Microbiol. 76, 366e369. Van Leeuwen, M.R., Krijgsheld, P., Wyatt, T.T., Golovina, E.A., Menke, H., Dekker, A., Stark, J., Stam, H., Bleichrodt, R., Wösten, H.A.B., Dijksterhuis, J., 2013. The effect of natamycin on the transcriptome of conidia of Aspergillus niger. Stud. Mycol. 74, 71e85. Wendland, J., 2001. Comparison of morphogenetic networks of filamentous fungi and yeast. Fungal Genet. Biol. 34, 63e82. Xu, X.-M., 1999. Effects of temperature on the length of the incubation period of rose powdery mildew (Spaerotheca pannosa var. rosae). Eur. J. Plant Pathol. 105, 13e21.
