university of groningen bacillus subtilis at near-zero ... · pdf filebacillus subtilis at...

University of Groningen

Bacillus subtilis at near-zero specific growth ratesOverkamp, Wout

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Overkamp, W. (2015). Bacillus subtilis at near-zero specific growth rates: Adaptations to extreme caloricrestriction [Groningen]: University of Groningen

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 10-05-2018

https://www.rug.nl/research/portal/en/publications/bacillus-subtilis-at-nearzero-specific-growth-rates(7e8bcf52-d71d-4f2e-adfd-f34c6e118a99).html

4 1

Physiological and cell 2

morphology adaptation of 3

Bacillus subtilis at near-zero 4

specific growth rates 5

6

7

8

9

10 11 12

13

14

Part of this chapter was published in: 15

Wout Overkamp, Onur Ercan, Martijn Herber, Antonius J. A. van Maris, Michiel 16 Kleerebezem and Oscar P. Kuipers. Physiological and cell morphology adaptation of 17 Bacillus subtilis at near-zero specific growth rates: a transcriptome analysis. 18 Environmental Microbiology 17, Issue 2, pages 346–363 (2015). 19

Chapter 4

88

20

Bacillus subtilis physiology at near-zero growth rates

89

Abstract 21

Nutrient scarcity is a common condition in nature and the resulting extremely low 22 growth rates (below 0.025 h-1) are an unexplored area in B. subtilis. To understand 23 microbial life in natural environments, studying the adaptation of B. subtilis to near- 24 zero growth conditions is relevant. To this end, a chemostat modified for culturing an 25 asporogenous B. subtilis sigF mutant strain at extremely low growth rates (also named 26 a retentostat) was set up and biomass accumulation, culture viability, metabolite 27 production and morphology were analysed. During retentostat culturing the specific 28 growth rate decreased to a minimum of 0.00006 h-1, corresponding to a doubling 29 time of 470 days. The energy distribution between growth- and maintenance-related 30 processes showed that a state of near-zero growth was reached. Remarkably, an 31 aberrant filamentous morphology emerged, suggesting that cell separation is 32 impaired under near-zero growth conditions. 33

Introduction 34

Microbial growth is determined by available nutrient concentrations and in nature, 35 nutrient availability varies over time and per environment (Koch, 1971; Demoling et 36 al., 2007). Regardless of whether the environment facilitates a feast and famine 37 lifestyle (Koch, 1971) or an oligotrophic existence (Egli, 2010), the concentrations of 38 assimilable substrates are frequently low but not absolutely zero (Ferenci, 2001). 39 Consequently, the specific growth rate of most microorganisms is far below their 40 maximum during the vast majority of time. High microbial growth rates as achieved 41 in laboratory batch cultures are probably rare in nature (Brock, 1971; Koch, 1997; 42 Ferenci, 2001). Therefore, to better understand microbial life in natural 43 environments, knowledge about physiology at near-zero growth rates is very relevant. 44

Inside the cell of a microorganism numerous biological processes take place that 45 require energy. These cellular processes can be categorized into growth- and non- 46 growth related processes. The former involve reactions contributing to production of 47 new cell constituents as well as synthesis of the machineries required for cell division. 48 The latter are also referred to as maintenance processes, and involve reactions 49

Chapter 4

90

associated with functions other than biomass formation (e.g. maintenance of chemi- 50 osmotic potential and turnover of macromolecular compounds; van Bodegom, 2007). 51 When nutrients are present in excess and environmental conditions are favourable, 52 microorganisms invest the majority of the available metabolic energy in growth. In 53 nutrient-limiting conditions however, a relatively large fraction of the energy sources 54 has to be used for maintenance-related processes (Pirt, 1965; Nyström, 2004a; van 55 Bodegom, 2007). When the metabolic energy distribution between growth and 56 maintenance has come to a point where available energy is spent exclusively on non- 57 growth associated processes, zero-growth is the result. 58

Zero-growth is a metabolically active, non-growing state of a microorganism 59 and is fundamentally different from starvation encountered during stationary phase, 60 which involves deterioration of physiological processes. The energetic investment for 61 an organism to remain in a viable and active state without growth is quantified by the 62 so-called non-growth associated maintenance energy coefficient (Pirt, 1965). Resting 63 states as typified by spores, which have little or no metabolic activity, are not 64 considered to be in zero-growth state. 65

Even in well-studied organisms such as Bacillus subtilis, extremely slow growth 66 has not been investigated extensively, mostly due to a lack of experimental 67 accessibility to achieve near-zero growth rates (Pirt, 1987). For example, in batch 68 cultivations, where cells proliferate at their maximum growth rates until nutrients are 69 depleted, the transition from exponential to stationary phase is rapid and involves 70 continuously changing environmental conditions. 71

Chemostat cultivation allows growth under controlled environmental conditions 72 and at submaximal specific growth rates. The growth rate can be manipulated by 73 varying the dilution rate, making the chemostat a proven setup for studying the effect 74 of growth rate on physiology (Novick and Szilard, 1950; Herbert et al., 1956; Sauer 75 et al., 1996; Dauner et al., 2001). However, due to homogeneity problems at low 76 dilution rates caused by the drop wise feeding of medium, specific growth rates lower 77 than 0.025 h-1 cannot be set-up reliably in chemostats (Herbert et al., 1956; Daran- 78 Lapujade et al., 2009). 79

To be able to study microbial physiology at extremely low specific growth rates, 80 retentostat cultivation has been developed (Herbert, 1961; van Verseveld et al., 81 1986). A retentostat, or recycling fermentor, is a chemostat in which all the biomass is 82


91

retained in the fermenter vessel by a filter in the effluent line. Growing a culture at a 83 fixed dilution rate on an energy-limited medium leads to accumulation of biomass 84 and a progressive decrease of energy substrate availability per cell. Consequently, 85 energy available for growth becomes more and more limiting, ultimately leading to 86 zero-growth rates when the substrate consumption rate equals the substrate 87 requirements for maintenance processes. Meanwhile, starvation is prevented in this 88 setup because the substrate supply continues (Herbert, 1961; Chesbro et al., 1979; 89 van Verseveld et al., 1986; Tappe et al., 1996; Boender et al., 2009; Goffin et al., 90 2010; Ercan et al., 2013). 91

From previously conducted long-term retentostat cultivation studies (22 - 45 92 days) it was concluded that Saccharomyces cerevisiae (Boender et al., 2009), Lactobacillus 93 plantarum (Goffin et al., 2010) and Lactococcus lactis (Ercan et al., 2013) exhibit energy 94 distribution- and maintenance characteristics that can be predicted from higher 95 growth rate chemostats. This is in contrast with studies on Escherichia coli (Chesbro et 96 al., 1979), Bacillus polymyxa (Arbige and Chesbro, 1982), Paracoccus denitrificans and 97 Bacillus licheniformis (van Verseveld et al., 1986), which applied short-term setups (3 98 days) not truly allowing zero-growth conditions to establish. 99

The physiology of B. subtilis at various specific growth rates has been the subject 100 of numerous studies (Sauer et al., 1996; Dauner et al., 2001; Tännler et al., 2008; 101 Blom et al., 2011). However, retentostat studies with specific growth rates that 102 approximate zero are an unexplored area in the B. subtilis field. B. subtilis is found in a 103 large variety of niches, like soil and water, in which zero-growth is probably a 104 common state. The differentiation strategies that B. subtilis has evolved, such as 105 formation of highly resistant spores, development of natural competence and 106 motility, secretion of exoproteases and biofilm formation (Dubnau, 1991; Msadek, 107 1999; Branda et al., 2001; Errington, 2003; Kearns and Losick, 2005; Veening et al., 108 2008) are probably a reflection of its habitat flexibility. They facilitate survival under 109 a large variety of (dynamic) environmental conditions, making it an intriguing subject 110 of study. 111

Here, we describe the cultivation of an asporogenous B. subtilis sigF mutant strain 112 in an aerobic, glucose-limited retentostat with the aim to investigate the physiological 113 responses to near-zero specific growth rates. The sigF mutant strain is used to prevent 114 undesirable sporulation/germination processes, which would interfere with reaching 115

Chapter 4

92

near-zero growth rates. During these long-term retentostat cultivations, extremely 116 low growth rates were reached and remarkable cell morphology changes emerged. 117

Results 118

Implementation of aerobic retentostat cultivation for B. subtilis 119

Defining the experimental set-up for retentostat cultivation of B. subtilis required a 120 number of modifications relative to other zero-growth studies (Boender et al., 2009; 121 Ercan et al., 2013). Achieving a zero growth state in a retentostat with a sporulating 122 strain would be impossible. The endospores are dormant and thus for every spore 123 formed, new vegetative cells could grow. Therefore, growth rates will never approach 124 zero in this situation. Additionally, accumulation of spores will continue which makes 125 estimation of culture parameters like maintenance coefficient very difficult as is 126 described in a retentostat study of A. niger (Jørgensen et al., 2010). Because 127 sporulation will interfere with reaching a zero-growth state and therefore is 128 undesired, a sporulation-deficient sigF mutant was used in this study. This strain can 129 initiate sporulation, but is unable to express the genes necessary for continuation of 130 the process at stage II in the sporulation cascade (Piggot and Coote, 1976; Setlow et 131 al., 1991; Dworkin and Losick, 2005). The use of a knockout strain may impose a 132 limitation in terms of being able to interpret the natural response of the organism. 133 However, this particular mutant is still able to express genes for sporulation initiation 134 (Fawcett et al., 2000; Steil et al., 2005; Wang et al., 2006) and thereby still allows us 135 to see if sporulation is one of the responses B. subtilis applies. For retentostat 136 cultivation, bioreactors were equipped with a cross-flow filter (Fig. 1). These filters 137 were chosen for their particular large filtration surface and tangential flow, 138 minimizing the risk of clogging during cultivation (Koros et al., 1996). Complete 139 biomass retention was confirmed by regular plating of effluent. Continuous filtration 140 for up to 42 days was possible without noticeable clogging or observation of growth 141 in the effluent. 142

Culture purity can be an issue with such long-term cultivations since the risk of 143 contamination by other bacteria is high. To easily identify contamination by 144 microscopic examination, a fusion of green fluorescent protein (GFP) with the 145 promoter of the constitutively expressed ribosomal ribonucleic acid (RNA) operon 146


93

rrnB (Krásný and Gourse, 2004) was introduced in the strain cultivated. 147 Consequently, cells not expressing GFP could be identified as contamination. The 148 retentostat cultures described here did not experience any contamination. 149

Air

Base

Med

ium

Anti-

foam

Sam

ple

port

Cell-

free

efflue

nt

Figure 1. Schematic illustration of a retentostat. A constant flow of air and fresh medium is provided to the culture. Anti-foam is added with timed intervals to prevent foaming. Biomass is retained in the reactor by a cross-flow filter on the effluent line. By means of a pump the culture is circulated over the filter loop to prevent clogging and de-oxygenation in the filter. Not displayed in the figure are the DO-, pH-, temperature- and level sensors used for monitoring of the culture. The pH sensor is coupled to the base feed to maintain a constant pH. Upon contact of the culture with the level sensor, cell-free effluent is pumped out of the reactor to keep a constant culture volume. Samples are withdrawn directly from the culture using an aseptic sampler system. (Figure by Sietse Koenders).

Chapter 4

94

Growth of B. subtilis in medium-supply tubing was found to be a potential 150 problem during prolonged (> 7 days) cultivation. The slow feed rate of 35 ml h-1 151 allowed the bacteria to grow against the current when aerosols or droplets from the 152 culture came in contact with the medium inlet on the fermentor. To prevent growth 153 in the tubing, a dropper was installed between the medium inlet and the medium- 154 supply tubing, preventing direct contact. A sterile airflow in the dropper pushed the 155 droplets to the reactor and prevented cells from entering the dropper. 156

Vigorous mixing and sparging with air to keep the culture oxygenated in a 157 bioreactor will lead to formation of foam, which can cause problems. Formation of 158 foam needs to be prevented and during trial experiments it became clear that not 159 every anti-foaming strategy was adequate. When anti-foam was premixed with the 160 supplied medium it precipitated in the silicone tubing, partly blocking the inflow of 161 the medium. Not all anti-foam reached the reactor and as a consequence foaming 162 was not prevented properly. Foaming was prevented most effectively by supplying 163 anti-foam on regular intervals. In our setup this was achieved by automatic addition 164 of 5 ml of a 5% (wt.wt-1) antifoam solution every 13 minutes via a computer- 165 controlled pump. 166

Growth and viability in retentostat cultures 167

B. subtilis 168 trpC2 sigF::spec amyE::PrrnB-GFP was grown under aerobic retentostat 168 conditions in chemically defined M9 medium with glucose as the growth limiting 169 substrate. Two independent retentostat cultivations were successfully performed for 170 42 and 40 days (retentostat 1 and 2, respectively) to study the adaptation of B. subtilis 171 to near-zero specific growth rates. 172

Biomass (measured in grams dry weight (gdw)) accumulated in the bioreactor and 173 asymptotically levelled off to a near zero growth state during a total of approximately 174 20 days (Fig. 2A). During this period, the calculated specific growth rate (µ) decreased 175 from 0.025 h-1 to 0.0004 h-1 (Fig. 2A). At the end of the cultivation, after 176 approximately 40 days, a minimum was reached of 0.00006 h-1. The specific growth 177 rate reached at the end of the cultivation corresponded to a doubling time of 178 approximately 470 days. 179


95

The biomass accumulation in the retentostat cultivations can be modelled with 180 the van Verseveld equation (see materials & methods, equation 5) if the maintenance 181 substrate requirement (ms) and the maximum yield of biomass on the substrate 182 (Ysxmax) is independent of the specific growth rate (van Verseveld et al., 1986). 183 However, as Ysxmax seems to be varying with the growth rate (see below), modelling 184 with the van Verseveld equation will yield values different from measured values, 185 which is indeed the case in the initial phase of retentostat culturing (Fig. 2A). 186

Figure 2. Growth and viability of B. subtilis in retentostat cultures. Steady-state aerobic chemostat cultures (D = 0.025 h-1) were switched to retentostat mode at time-point zero. Displayed are data from retentostat cultivation 1 (■) and 2 (□). (A) Measured biomass concentration (gdw l-1). Data points represent mean + standard deviation of duplicate samples. Additionally, the biomass calculated with the fitted van Verseveld equation for retentostat 1 (---) and 2 (…) is shown, as well as the corresponding calculated specific growth rates ((● ) and (○), respectively). (B) Cultivability estimated by colony forming units (CFU ml-1) on solid medium. Data points represent mean + standard deviation of triplicate samples.

Chapter 4

96

Next to the assumption of a constant ms and Ysxmax, two requirements are to be 187 met for the van Verseveld equation: all biomass remains inside the bioreactor, and all 188 biomass is viable and metabolically active (van Verseveld et al., 1986). Both 189 requirements are experimentally found to be met: Regular plating of effluent showed 190 that no biomass escaped from the bioreactor and viability of the cultures was shown 191 to be approximately 99% as assessed with fluorescence microscopy. 192

Colony-forming units (CFU), determined by plating of the culture, followed a 193 trend during the first 20 days in correspondence with the biomass accumulation (Fig. 194 2B). Surprisingly, after approximately 20 days (between time points 2 and 3) the 195 colony-forming unit count decreased, whilst both the culture dry weight (Fig. 2A) and 196 proportion of live cells (as detected by fluorescence analysis) remained stable. This 197 was a decrease of approximately 30% for retentostat 1 and 70% for retentostat 2. 198

Cell morphology 199

Cell morphology was monitored using phase contrast microscopy (Fig. 3A). In 200 general, the cell morphology appeared to be heterogeneous. During the initial days of 201 retentostat cultivation, cells were shaped like the typical B. subtilis rod with a few cells 202 being shorter than normal. After roughly 20 days, when biomass accumulation 203 reached a plateau and specific growth rates decreased to 0.0004 h-1, a different cell 204 morphology emerged. Cells that were longer than the typical rod started to appear, 205 some reaching lengths of more than 10 times that of the usual vegetative single cell at 206 the start of retentostat culturing. Additionally, most of these cells were bent or 207 curved. The length of the cells and frequency of this aberrant morphology seemed to 208 be proportional to the duration of retentostat culturing. In both retentostats these 209 morphologies appeared, but to a greater extent in retentostat 2. The appearance of 210 the elongated cells coincided with the observed decrease in CFU, with retentostat 2 211 having the most pronounced decrease in CFU and largest number of long cells. 212 When a sample from the retentostat culture was grown overnight as batch in LB 213 medium, microscopic examination showed that the elongated cells were no longer 214 present and that solely the typical B. subtilis rod-shape could be observed. 215


97

. 216

Figure 3A. Morphological changes of B. subtilis during retentostat cultivation. During a period of 42 and 40 days an elongated morphology emerged in retentostat 1 and 2, respectively. First appearance of this morphology was after 18 and 20 days, respectively, coinciding with the biomass accumulation reaching a plateau. Scale bar indicates 5 μm.

Figure 3B. Membrane visualization in aberrant B. subtilis cells developed during retentostat cultivation. Staining of membranes with FM5-95 dye reveals that long cells are composed of multiple compartments. Shown are phase contrast- and fluorescent microscopy images (left and right, respectively). Scale bar indicates 5 μm.

Chapter 4

98

In order to determine whether the elongated cells were single cells or composed 217 of multiple undissociated cells, the cell membrane was visualized by incubating with 218 the FM5-95 membrane dye (Invitrogen, UK). From microscopy images (Fig. 3B) it is 219 clear that the filamentous cells are actually septated. Vigorous mixing did not lead to 220 separation of the multiple compartments, confirming that the cells remained attached 221 to each other. Additionally, the individual cells in such a chain appear to be 222 elongated. When estimations of average cell number per chain in retentostat 2 were 223 used to calculate the amount of CFU in a single cell- and chained cell scenario 224 (results not shown), a decrease of 67% with chained cells was the outcome when 225 compared to single cells. This calculated decrease is similar to the observed decrease 226 in retentostat 2. 227

Maintenance coefficient and maximum growth yield 228

In a retentostat culture, progressive biomass accumulation leads to decreasing 229 substrate availability per amount of biomass. When the substrate consumption rate 230 (qs) is plotted against time (Fig. 4A), it is visible that qs approaches an asymptote: the 231 value known as the maintenance substrate requirement (ms; Pirt, 1965). Ultimately, 232 when the point is reached where the qs equals ms, growth ceases. As is described in 233 material and methods, qs can be calculated from culture parameters. 234

If the maintenance substrate requirement and the yield of biomass on the 235 substrate (Ysxmax) are independent of the specific growth rate, they can be determined 236 from the linear dependence of the glucose consumption rate over the whole range of 237 specific growth rates achieved in the retentostat (Pirt, 1982; Boender et al., 2009) 238 (Fig. 4B). A linear trendline with an R2 of 0.9605 and 0.9661 could be drawn through 239 the data points of both retentostats. At a µ of zero, qs equals ms, which is the value at 240 the Y-axis intercept. In the present study, ms was found to be 0.24 mmol glucose gdw-1 241 h-1. The Ysxmax, represented by 1/slope of the plot, turned out to be variable among 242 different ranges of specific growth rates. The qs from our steady-state chemostats at 243 0.025 h-1 was in agreement with data from higher growth rate chemostats (Dauner et 244 al., 2001), yielding a Ysxmax of 64.56 mgdw mmol glucose-1. The average Ysxmax over 245 the range of growth rates in the current study (0.025 h-1 to 0.00006 h-1), was 246 determined to be 47.57 mgdw mmol glucose-1. At the lower growth rates, between 247 0.00275 h-1 and 0.00006 h-1 a Ysxmax of 16.74 mgdw mmol glucose-1 was found. 248


99

Although there was a linear dependence of qs over this lower range of growth rates, 249 the variation found for the different ranges of growth rates indicate a variable Ysxmax. 250

0 5 10 15 20 25 30 35 40 450.0

0.2

0.4

0.6

0.8

1.0

Time (d)Spec

ific

subs

trate

con

sum

ptio

n ra

te

(mm

olgl

ucos

e gdw

-1 h

-1)

0.000 0.005 0.010 0.015 0.020 0.0250.0

0.2

0.4

0.6

0.8

1.0

Specific growth rate (h-1)Spec

ific

subs

trate

con

sum

ptio

n ra

te

(mm

olgl

ucos

e gdw

-1 h

-1)

A

B

Figure 4. Substrate consumption kinetics in retentostat cultures of B. subtilis. Steady-state aerobic chemostat cultures (D = 0.025 h-1) were switched to retentostat mode at time-point zero. Displayed are data from retentostat cultivation 1 (■) and 2 (□). (A) Specific substrate consumption rate (qs) asymptotically approaches the maintenance substrate requirement (ms) during cultivation. (B) Relationship between specific substrate consumption rate (qs) and specific growth rate (µ). Y-axis intersect of the plotted data determines the maintenance substrate requirement (ms).

Chapter 4

100

To be able to calculate the energy produced from glucose during retentostat 251 culturing, residual glucose concentrations and catabolic end products were 252 determined by HPLC analysis. The glucose concentration in the spent medium was 253 below the detection limit. During retentostat cultivation no metabolites were detected 254 in the culture supernatant, indicating complete respiratory dissimilation of glucose. 255 The amount of adenosine triphosphate (ATP) formed with dissimilation of glucose 256 depends on the coupling efficiency of the electron transport chain to ATP synthesis. 257 B. subtilis possesses a three-branched electron transfer chain, of which the cytochrome 258 c branch is inactive in glucose grown cells (Winstedt and von Wachenfeldt, 2000) and 259 the bd oxidase branch is usually active under low oxygen conditions (Winstedt et al., 260 1998). Most relevant for our cultures is the aa3 oxidase branch, which is predominant 261 in B. subtilis under oxygen-rich conditions (Winstedt and von Wachenfeldt, 2000). 262 Transfer via this chain is expected to result in formation of 15 ATP per glucose 263 molecule (Dauner et al., 2001). Assuming that complete oxidation of a glucose 264 molecule to carbon dioxide yields 15 ATP, the maintenance requirement for ATP 265 (matp) is equivalent to 3.6 mmol ATP gdw-1 h-1. 266

The values for the maintenance energy coefficient ms calculated above showed 267 that during roughly 40 days of retentostat cultivation, the amount of substrate used 268 for maintenance increased from approximately 30% at a growth rate of 0.025 h-1 to 269 approximately 100% of the total substrate consumed at the end of the retentostat 270 cultivation (See Fig. 5). The two independent retentostat cultures displayed highly 271 similar growth kinetics and the distribution of energy between maintenance- and 272 growth-related processes above shows that a near-zero growth state is reached during 273 cultivation. 274

Discussion 275

A setup for culturing B. subtilis at extremely low growth rates was implemented and 276 the adaptation of B. subtilis to near-zero growth conditions was studied by analysis of 277 biomass accumulation, culture viability, metabolite production and morphology. 278

The extremely low specific growth rate (0.0006 h-1) reached at the end of the 279 retentostat cultivation shows that cell retention is an effective way of reaching near- 280 zero growth rates, while maintaining a steady supply of nutrients. At this stage, all 281


101

substrate consumed by the cells is used for maintenance purposes, which is estimated 282 to be 0.24 mmol glucose gdw-1 h-1. The ms was comparable to that found in studies on 283 B. subtilis at higher growth rates (Sauer et al., 1996; Dauner and Sauer, 2001; 284 Tännler et al., 2008). In these chemostat studies the ms was found to be ranging from 285 0.21 to 0.49 mmol glucose gdw-1 h-1. 286

A

B

0 7 12 22 28 35 420

50

100

Time (d)

% o

f glu

cose

con

sum

ptio

n ra

te

0 6 13 20 32 38 400

50

100

Time (d)

% o

f glu

cose

con

sum

ptio

n ra

te

Figure 4. Percentage of energy directed towards maintenance versus growth. Black bars represent energy directed towards maintenance. White bars represent energy directed towards growth. (A) Retentostat 1. (B) Retentostat 2.

Chapter 4

102

Both Ysxmax values found for steady-state chemostat (47.57 mgdw mmol glucose-1) 287 and zero-growth retentostat cultures (16.74 mgdw mmol glucose-1) were lower than 288 values determined in other chemostat studies that employed higher growth rates 289 (between 67 and 82 + 2 mgdw mmol glucose-1) (Sauer et al., 1996; Dauner et al., 290 2001; Tännler et al., 2008). The observed variability of the Ysxmax value implies that 291 the conversion of substrate to net biomass does not have a constant efficiency in B. 292 subtilis. Various explanations may underlie these observations, including partial lysis 293 of cells, or any other factor that results in higher biomass formation than is measured, 294 or specific growth-related processes that require elevated substrate consumption to 295 generate the same biomass amount at low growth rates. 296

B. subtilis is a typical soil inhabitant and while many soils provide limited 297 amounts of carbon sources (Aldén et al., 2001; Ekblad and Nordgren, 2002; Ilstedt 298 and Singh, 2005; Demoling et al., 2007), the maintenance metabolism does not seem 299 to be specifically evolved for near-zero growth conditions under carbon limitation, 300 because the maintenance coefficient calculated from retentostat cultivation and from 301 extrapolated chemostat cultures at higher growth rates are virtually identical. Instead, 302 B. subtilis employs other strategies for survival under challenging conditions, such as 303 the formation of endospores. The asporogenous sigF mutant used in this study is only 304 able to express genes for sporulation initiation and thereby still allows us to see if 305 sporulation is one of the responses B. subtilis applies. Whether these genes are 306 differentially expressed under retentostat conditions needs to be elaborated by 307 transcriptome analysis. 308

The occurrence of the cell chains and elongation of cells (most prominent in 309 retentostat 2) is likely correlated with the coinciding decline of CFU enumerations. 310 Since the viability remained high and the culture dry weight did not decrease, it is 311 unlikely that the decrease in colony-forming units is caused either by cell death or 312 lysis. An alternative explanation for the observed decrease of the CFU count would 313 be a change in the morphology of the cells, resulting in an increased weight per cell 314 or chains of cells that form a single colony when plated, while actually representing 315 multiple cells. Taking into account the average number of cells per chain enabled the 316 accurate explanation of the reduced CFU numbers, indicating that chain formation 317 is the predominant explanation for the observed CFU decline. The retentostat- 318 adapted cell chains rapidly reverted to the typical B. subtilis rod shape when 319


103

inoculated in normal batch culture conditions, illustrating that this condition- 320 dependent morphology is a response to retentostat cultivation. Chained cell 321 morphology has been observed previously for B. subtilis (Cozy and Kearns, 2010; 322 Kawai et al., 2003; Ohnishi et al., 1999; Salzberg and Helmann, 2008) and causes 323 were reported to be related with cell division, cell separation and membrane 324 composition. Kawai et al. (2003) showed that expression of the YneA protein resulted 325 in cell elongation and reduced localization of the cell division protein FtsZ to the cell 326 division site. YneA was found to be responsible for cell division suppression during 327 the SOS response in B. subtilis. Cozy and Kearns (2010) showed that a mutant of 328 LytABCDF leads to a phenotype where conjoined daughter cells are not separated 329 after division by binary fission. The LytABCDF operon codes for cell wall-remodelling 330 enzymes called autolysins, which facilitate cell separation (Blackman et al., 1998; 331 Chen et al., 2009; Margot et al., 1999; Vollmer et al., 2008). RNA polymerase with 332 the alternative sigma factor σD directs expression of the autolysin enzymes (Kearns 333 and Losick, 2005). Similar findings are reported by Ohnishi et al. (1999) in a study 334 where lytF and sigD mutants form filamentous cells. In a study by Salzberg and 335 Helmann (2008) it was found that mutants with altered membrane composition 336 formed aberrant cell morphologies. Cells with inactivation of three genes mprF, pssA 337 and ywnE involved in biosynthesis of complex lipids, were shown to be filamentous. 338 In this case the SigD regulon possibly plays a role too, since its expression was 339 negatively affected by these mutations. The studies mentioned above indicate that a 340 wide variety of causes for the reported aberrant cell morphology exist. Transcriptome 341 analysis of the B. subtilis retentostat culture will provide more insight. 342

As a consequence of cells experiencing caloric restriction in the retentostat 343 cultivation, the specific growth rate decreased over time and the substrate 344 consumption rate for maintenance-associated processes relatively increased. A state 345 of near-zero growth was reached, yet enough energy was available to keep the cells 346 alive and metabolically active. This work shows that retentostat culturing is (apart 347 from some pitfalls mentioned above) a controlled and stable cultivation condition, 348 suitable for the study of extremely slow growing cultures of B. subtilis, which allocate 349 the vast majority of substrate derived energy to maintenance. Future studies will now 350 focus on dissecting the adaptation of B. subtilis to near-zero growth rates in retentostat 351 cultures by means of transcriptome analysis. 352

Chapter 4

104

Material and methods 353

Strain, growth conditions and media 354

B. subtilis 168 trpC2 sigF::spec amyE::PrrnB-gfp+ was used for the retentostat 355 experiments in this study. This strain carries a GFP fusion to the promoter of the 356 constitutively expressed ribosomal RNA operon rrnB (Krásný and Gourse, 2004; 357 Veening et al., 2009), and is defective in sporulation, caused by a disruption in the 358 sigF gene. Precultures for chemostat and retentostat cultivations were prepared by 359 inoculating a single colony from an lysogeny broth (LB) agar plate into 10 ml LB 360 medium (Sambrook et al., 1989). This culture was grown at 37˚C until an optical 361 density at 600 nm (OD600) of 0.3 was reached. Subsequently 1000x dilutions were 362 made in 60 ml M9 medium (Miller, 1972) supplemented with 27.75mM glucose and 363 0.1mM Tryptophan. The M9 minimal medium contained, per liter of deionized 364 water, 8.5 g of Na2HPO4  ·  2H2O, 3.0 g of KH2PO4, 1 g of NH4Cl, and 0.5 g of 365 NaCl. The following components were sterilized separately and added per liter: 1 ml 366 of 0.1 M CaCl2, 1 ml of 1 M MgSO4, 1 ml of 50 mM FeCl3, and 10 ml of M9 trace 367 salts solution. The M9 trace salts solution contained (per liter) 0.1 g MnCl2  ·  4H2O, 368 0.17 g of ZnCl2, 0.043 CuCl2  ·  2H2O, 0.06 CoCl2  ·  6H2O, 0.06 Na2MoO4  ·  2H2O. 369 The cultures were grown overnight and used for inoculation of the bioreactors. 370 Chemostat- and retentostat media were acidified to pH 5 by addition of H2SO4 (95 371 to 97%) to avoid precipitation of medium components. During the cultivation in the 372 bioreactors the pH was maintained at 7.0 by automatic addition of NaOH 5M. 373

Chemostat cultivation 374

Duplicate chemostat cultures were performed at a dilution rate, D (defined as the 375 ratio of the medium feed rate (L h-1) and culture volume (L)) of 0.025 h-1. 2.0 L 376 bioreactors (Infors Benelux BV, the Netherlands) with 1.4 L working volume were 377 inoculated with an exponentially growing preculture to start the chemostats. The 378 bioreactors were operated at 37˚C under aerobic conditions. An airflow of 0.1 l.min-1 379 and a stirring speed of 800 r.p.m. was set to keep oxygen levels above 50% of air- 380 saturation. 381

The working volume was kept constant by means of a conductivity sensor placed 382 at the surface of the culture, activating a peristaltic pump that removed effluent. To 383


105

prevent foam formation, 5 ml of a 5% (wt.wt-1) solution of the antifoaming agent 384 Struktol J673 (Schill and Seilacher AG, Hamburg, Germany) was added per 24 385 hours, automatically spread over intervals of 13 minutes. Steady state was defined as 386 the condition in which culture parameters were constant for at least 5 volume 387 changes and when optical density at 600 nm (OD600) and cell dry weight (CDW) had 388 remained constant (<5% and <10% variation, respectively) for at least two volume 389 changes. Culture purity was routinely checked by phase-contrast- and fluorescence 390 microscopy. The PrrnB-GFP fusion allowed for identification of fluorescent cells as 391 being the inoculated B. subtilis. Additionally, cells were plated on LB agar plates to 392 check for possible contaminations. 393

Retentostat cultivation 394

A 2.0 L bioreactor (Infors Benelux BV, the Netherlands) was equipped with an 395 autoclavable polyethersulfone cross-flow filter with a pore size of 0.22 μm (Spectrum 396 Laboratories, CA, USA) to retain biomass in the reactor. The filter was connected to 397 the bioreactor via an external loop, through which culture was circulated. 398

Two individual retentostat experiments were initiated from chemostat cultures 399 at dilution rates of 0.025 h-1. After reaching steady state in the chemostat, the 400 bioreactors were switched to retentstat mode by withdrawing the effluent through the 401 filter instead of through the standard effluent tube. The retentostat cultivations were 402 operated under the same conditions (temperature, pH, medium flow rate, 403 oxygenation, stirring rate, anti-foam addition) as the chemostats. Since withdrawal of 404 biomass from the culture influences the kinetics of biomass accumulation, sampling 405 volumes and -frequency were kept to a minimum. The super safe sampler ports 406 (Infors Benelux BV, the Netherlands) that were used for fast and aseptic sample 407 withdrawal, allowed for accurate control of the sample volume. 408

Determination of biomass, substrate and metabolites 409

During chemostat- and retentostat cultivation, samples were withdrawn from the 410 bioreactor to determine biomass-, glucose- and organic acid concentrations. Cell dry 411 weight was determined in duplicate by cooled centrifugation of 5 mL of culture in 412 pre-weighted tubes, washing with 0.9% NaCl and drying at 105˚C for 24 h to 413 constant weight. Additionally, the optical density of the culture was determined by 414 measuring the optical density at 600 nm. Glucose and organic acid concentrations in 415

Chapter 4

106

culture supernatants were determined by high-performance liquid chromatography 416 (Shimadzu Scientific Instruments, MD, USA) using LC Solutions SP1 software from 417 Shimadzu (Kyoto, Japan). Culture supernatants were obtained by centrifugation 418 (10,000g for 10 min at 4˚C), filter sterilized and stored at -20˚C until HPLC analysis. 419 Samples were separated using an Aminex HPX-87H anion-exchange column (Bio- 420 rad Laboratories Inc., Richmond, CA) with sulphuric acid (5 mM; 0.6 ml · min-1) as 421 mobile phase at 55˚C. Detection was performed with a refractive index detector and 422 UV wavelength absorbance detector (Shimadzu Scientific Instruments, MD, USA). 423

Assessment of cell viability 424

To estimate the fraction of dead cells in culture samples, the red fluorescent 425 compound propidium iodide (PI) was used. PI binds deoxyribonucleic acid (DNA) 426 and can only do so in cells with permeabilized membranes. A 10 µL culture sample 427 was washed and resuspended in 990 µL 0.85% NaCl and 1 µL of PI (20 mM 428 dissolved in dimethyl sulfoxide (DMSO)) was added and incubated for 10 min. The 429 PrrnB-GFP fusion this strain carries causes metabolically active cells to produce GFP 430 and these could thus be detected as alive. Enumeration of live and dead cell fractions 431 was performed using these PI and GFP markers, in a Deltavision (Applied Precision) 432 IX7 1DV Microscope (Olympus) which is described in more detail below. 433 Additionally, colony forming unit (CFU) quantification was performed by plating 434 tenfold dilution series of the cultures (5 to 7 dilutions in LB broth) in triplicate on LB 435 agar (1.5% wt/vol) plates. After 24 hours of incubation at 37˚C, colonies were 436 counted. 437

Microscopy and analysis of cell morphology 438

Cell morphology was analysed by phase-contrast- and fluorescence microscopy. In 439 order to visualize the cell membrane, cells were incubated for 1 minute with an ice- 440 cold 5 µg/mL FM5-95 membrane dye solution (Invitrogen, UK) prior to microscopy 441 analysis. Images were taken with Deltavision (Applied Precision) IX71Microscope 442 (Olympus) using a CoolSNAP HQ2 camera (Princeton Instruments) with a 100× 443 phase-contrast objective. Fluorescence filter sets used to visualize GFP (excitation at 444 450/90 nm; emission at 500/50 nm) and red dyes (excitation at 572/35 nm, emission 445 632/60 nm) were from Chroma Technology Corporation (Bellows Falls, USA). 446 Exposure time was between 0.2 and 1 s with 32% transmission xenon light (300 W). 447


107

Exposure time for phase-contrast images was 0.05 s. Softworx 3.6.0 (Applied 448 Precision) software was used for image capturing. Time-lapse microscopy was 449 performed as described before (de Jong et al., 2011). 450

Calculation of retentostat growth kinetics 451

The following mass balance equations are used to calculate growth kinetics in 452 retentostat cultivations. Equation 1 is for biomass, assuming complete cell retention 453 and constant volume. Equation 2 is for substrate. Here, Cx is the biomass 454 concentration (g · liter-1), µ is the specific growth rate (h-1), Cs is the residual glucose 455 concentration (g · liter-1), D is the dilution rate (h-1), Cs,in is the glucose concentration 456 in the feed (g · liter-1), and qs is the specific glucose consumption rate (g · g-1 · h-1). 457

!"!!"= 𝜇 ∙ 𝐶! (1) 458

!"!!"= 𝐷 𝐶!,!" − 𝐶! − 𝑞! ∙ 𝐶! (2) 459

With the assumptions that variation of the glucose concentration in the 460 bioreactor is negligible in comparison to the amount of glucose supplied with fresh 461 medium (dCs/dt << D ∙ Cs,in) and that the residual glucose concentration in the broth 462 (consistently below the detection limit) is much smaller than the glucose 463 concentration in the medium, qs can be calculated with equation 3. 464

𝑞! =!∙!!,!"!!

(3) 465

The Herbert-Pirt (Pirt, 1982) equation (equation 4) describes the relation 466 between specific substrate consumption rate (qs), specific growth rate (µ), maximum 467 biomass yield on substrate (Ysxmax) and maintenance energy coefficient (ms). 468

𝑞! = !

!!"!"# +𝑚! (4) 469

Chapter 4

108

Equation 4 was used to determine ms and Ysxmax. This was done by plotting the values 470 of qs against µ. The intercept of a linear regression line with the y-axis determines ms. 471 Ysxmax was calculated by taking the slope-1 of the regression line. 472

The biomass accumulation during retentostat cultivation can be described by 473 the van Verseveld equation (van Verseveld et al., 1986) (equation 5). 474

𝐶! 𝑡 = 𝐶!,! −! !!,!"!!!

!!∙ 𝑒!!!∙!!"!"#∙! + !(!!,!"!!!)

!! (5) 475

This equation assumes an ideal situation with no loss of viability and growth rate 476 independent maintenance-energy requirements. 477

The specific growth rate of the retentostat cultivations is calculated with 478 equation 6: 479

𝜇 = !!!,!"!#$/!"!!,!"#$%&

(6) 480

In order to determine the derivative of the biomass accumulation data (dCx,total/dt), the 481 measured total biomass concentrations (viable and non-viable cells) were fitted with 482 the equation Cx = A · eB · t + C, which is of the same shape as equation 5. This was 483 done using GraphPad Prism 6 (GraphPad Software Inc., USA), minimizing the sum 484 of squares of errors by varying A, B and C. With A, B and C known, the derivative 485 (dCx,total/dt) could be determined. Because only viable biomass can replicate, this is 486 incorporated in the equation. The doubling time of the culture is calculated by 487 dividing the natural logarithm of 2 by the specific growth rate. 488

Acknowledgements 489

We thank Bert van der Bunt, Marjo Starrenburg and Erik de Hulster for valuable 490 help with the bioreactors; Mark Bisschops for valuable help with calculations; Agata 491 Pudlik and Tim Vos for valuable help with HPLC analysis; members of the joint 492 zero-growth project group (Kluyver Centre, the Netherlands) for support and 493 valuable discussions; and Sietse Koenders for the illustration in Figure 1. 494


109

This work was carried out within the research programme of the Kluyver 495 Centre for Genomics of Industrial Fermentation which is part of the Netherlands 496 Genomics Initiative / Netherlands Organization for Scientific Research. 497

References 498

Arbige, M., and Chesbro, W.R. (1982). Very slow growth of Bacillus polymyxa: 499 Stringent response and maintenance energy. Arch. Microbiol. 132, 338–344. 500

Blackman, S.A., Smith, T.J., and Foster, S.J. (1998). The role of autolysins during 501 vegetative growth of Bacillus subtilis 168. Microbiol. Read. Engl. 144 ( Pt 1), 502 73–82. 503

van Bodegom, P. (2007). Microbial Maintenance: A Critical Review on Its 504 Quantification. Microb. Ecol. 53, 513–523. 505

Boender, L.G.M., de Hulster, E.A.F., van Maris, A.J.A., Daran-Lapujade, P.A.S., 506 and Pronk, J.T. (2009). Quantitative physiology of Saccharomyces cerevisiae at 507 near-zero specific growth rates. Appl. Environ. Microbiol. 75, 5607–5614. 508

Branda, S.S., González-Pastor, J.E., Ben-Yehuda, S., Losick, R., and Kolter, R. 509 (2001). Fruiting body formation by Bacillus subtilis. Proc. Natl. Acad. Sci. 98, 510 11621–11626. 511

Brock, T.D. (1971). Microbial growth rates in nature. Bacteriol. Rev. 35, 39–58. 512

Čáslavská, J., Kodešová, J., and Horáková, I. (1972). Effect of prolonged cultivation 513 in a chemostat under nitrogen limitation on the morphology and 514 ultrastructure of Bacillus subtilis. Folia Microbiol. (Praha) 17, 126–131. 515

Chen, R., Guttenplan, S.B., Blair, K.M., and Kearns, D.B. (2009). Role of the 516 sigmaD-dependent autolysins in Bacillus subtilis population heterogeneity. J. 517 Bacteriol. 191, 5775–5784. 518

Chesbro, W., Evans, T., and Eifert, R. (1979). Very slow growth of Escherichia coli. J. 519 Bacteriol. 139, 625–638. 520

Cozy, L.M., and Kearns, D.B. (2010). Gene position in a long operon governs 521 motility development in Bacillus subtilis. Mol. Microbiol. 76, 273–285. 522

Chapter 4

110

Daran-Lapujade, P., Daran, J.-M., van Maris, A.J.A., de Winde, J.H., and Pronk, 523 J.T. (2009). Chemostat-based micro-array analysis in baker’s yeast. Adv. 524 Microb. Physiol. 54, 257–311. 525

Dauner, M., Storni, T., and Sauer, U. (2001). Bacillus subtilis Metabolism and 526 Energetics in Carbon-Limited and Excess-Carbon Chemostat Culture. J. 527 Bacteriol. 183, 7308–7317. 528

Dubnau, D. (1991). Genetic competence in Bacillus subtilis. Microbiol. Rev. 55, 395– 529 424. 530

Dworkin, J., and Losick, R. (2005). Developmental Commitment in a Bacterium. 531 Cell 121, 401–409. 532

Ercan, O., Smid, E.J., and Kleerebezem, M. (2013). Quantitative physiology of 533 Lactococcus lactis at extreme low-growth rates. Environ. Microbiol. 15, 2319– 534 2332. 535

Errington, J. (2003). Regulation of endospore formation in Bacillus subtilis. Nat. Rev. 536 Microbiol. 1, 117–126. 537

Eymann, C., Homuth, G., Scharf, C., and Hecker, M. (2002). Bacillus subtilis 538 functional genomics: global characterization of the stringent response by 539 proteome and transcriptome analysis. J. Bacteriol. 184, 2500–2520. 540

Ferenci, T. (2001). Hungry bacteria--definition and properties of a nutritional state. 541 Environ. Microbiol. 3, 605–611. 542

Goffin, P., van de Bunt, B., Giovane, M., Leveau, J.H.J., Höppener-Ogawa, S., 543 Teusink, B., and Hugenholtz, J. (2010). Understanding the physiology of 544 Lactobacillus plantarum at zero growth. Mol. Syst. Biol. 6, 413. 545

Herbert, D. (1961). A theoretical analysis of continuous culture systems. In 546 Continuous Culture of Micro-organisms, (London: Society of chemical 547 industry), pp. 21–53. 548

Herbert, D., Elsworth, R., and Telling, R.C. (1956). The continuous culture of 549 bacteria; a theoretical and experimental study. J. Gen. Microbiol. 14, 601– 550 622. 551

de Jong, I.G., Beilharz, K., Kuipers, O.P., and Veening, J.-W. (2011). Live Cell 552 Imaging of Bacillus subtilis and Streptococcus pneumoniae using Automated Time- 553 lapse Microscopy. J. Vis. Exp. JoVE. 554


111

Jørgensen, T.R., Nitsche, B.M., Lamers, G.E., Arentshorst, M., van den Hondel, 555 C.A., and Ram, A.F. (2010). Transcriptomic insights into the physiology of 556 Aspergillus niger approaching a specific growth rate of zero. Appl. Environ. 557 Microbiol. 76, 5344–5355. 558

Kawai, Y., Moriya, S., and Ogasawara, N. (2003). Identification of a protein, YneA, 559 responsible for cell division suppression during the SOS response in Bacillus 560 subtilis. Mol. Microbiol. 47, 1113–1122. 561

Kearns, D.B., and Losick, R. (2005). Cell population heterogeneity during growth of 562 Bacillus subtilis. Genes Dev. 19, 3083–3094. 563

Koch, A.L. (1997). Microbial physiology and ecology of slow growth. Microbiol. 564 Mol. Biol. Rev. 61, 305–318. 565

Koros, W.J., Ma, Y.H., and Shimidzu, T. (1996). Terminology for membranes and 566 membrane processes (IUPAC Recommendation 1996). J. Membr. Sci. 120, 567 149–159. 568

Krásný, L., and Gourse, R.L. (2004). An alternative strategy for bacterial ribosome 569 synthesis: Bacillus subtilis rRNA transcription regulation. EMBO J. 23, 4473– 570 4483. 571

Margot, P., Pagni, M., and Karamata, D. (1999). Bacillus subtilis 168 gene lytF 572 encodes a gamma-D-glutamate-meso-diaminopimelate muropeptidase 573 expressed by the alternative vegetative sigma factor, sigmaD. Microbiol. 574 Read. Engl. 145 ( Pt 1), 57–65. 575

Miller, J. (1972). Experiments in Molecular Genetics (Cold Spring Harbor, NY: Cold 576 Spring Harbor Laboratory). 577

Msadek, T. (1999). When the going gets tough: survival strategies and environmental 578 signaling networks in Bacillus subtilis. Trends Microbiol. 7, 201–207. 579

Nishino, T., Gallant, J., Shalit, P., Palmer, L., and Wehr, T. (1979). Regulatory 580 nucleotides involved in the Rel function of Bacillus subtilis. J. Bacteriol. 140, 581 671–679. 582

Novick, A., and Szilard, L. (1950). Description of the chemostat. Science 112, 715– 583 716. 584

Chapter 4

112

Nyström, T. (2004). Growth versus maintenance: a trade-off dictated by RNA 585 polymerase availability and sigma factor competition? Mol. Microbiol. 54, 586 855–862. 587

Ohnishi, R., Ishikawa, S., and Sekiguchi, J. (1999). Peptidoglycan Hydrolase LytF 588 Plays a Role in Cell Separation with CwlF during Vegetative Growth of 589 Bacillus subtilis. J. Bacteriol. 181, 3178 –3184. 590

Pazlarova, J. (2010). Chemostat-induced uneven division of Bacillus subtilis. J. Ind. 591 Microbiol. Biotechnol. 37, 1257–1261. 592

Pirt, S.J. (1965). The maintenance energy of bacteria in growing cultures. Proc. R. 593 Soc. Lond. Ser. B Contain. Pap. Biol. Character R. Soc. Gt. Br. 163, 224– 594 231. 595

Pirt, S.J. (1982). Maintenance energy: a general model for energy-limited and energy- 596 sufficient growth. Arch. Microbiol. 133, 300–302. 597

Pirt, S.J. (1987). The energetics of microbes at slow growth rates: Maintenance 598 energies and dormant organisms. J. Ferment. Technol. 65, 173–177. 599

Salzberg, L.I., and Helmann, J.D. (2008). Phenotypic and transcriptomic 600 characterization of Bacillus subtilis mutants with grossly altered membrane 601 composition. J. Bacteriol. 190, 7797–7807. 602

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular cloning: a laboratory 603 manual (Cold Spring Harbor Laboratory). 604

Sauer, U., Hatzimanikatis, V., Hohmann, H., Manneberg, M., van Loon, A., and 605 Bailey, J. (1996). Physiology and metabolic fluxes of wild-type and riboflavin- 606 producing Bacillus subtilis. Appl Env. Microbiol 62, 3687–3696. 607

Setlow, B., Magill, N., Febbroriello, P., Nakhimovsky, L., Koppel, D.E., and Setlow, 608 P. (1991). Condensation of the forespore nucleoid early in sporulation of 609 Bacillus species. J. Bacteriol. 173, 6270–6278. 610

Strahl, H., and Hamoen, L.W. (2010). Membrane potential is important for bacterial 611 cell division. Proc. Natl. Acad. Sci. 107, 12281 –12286. 612

Tännler, S., Decasper, S., and Sauer, U. (2008). Maintenance metabolism and 613 carbon fluxes in Bacillus species. Microb. Cell Factories 7, 19. 614


113

Tappe, W., Tomaschewski, C., Rittershaus, S., and Groeneweg, J. (1996). 615 Cultivation of nitrifying bacteria in the retentostat, a simple fermenter with 616 internal biomass retention. FEMS Microbiol. Ecol. 19, 47–52. 617

Tappe, W., Laverman, A., Bohland, M., Braster, M., Rittershaus, S., Groeneweg, J., 618 and Verseveld, H.W. van (1999). Maintenance Energy Demand and 619 Starvation Recovery Dynamics of Nitrosomonas europaea and Nitrobacter 620 winogradskyi Cultivated in a Retentostat with Complete Biomass Retention. 621 Appl. Environ. Microbiol. 65, 2471–2477. 622

Veening, J.-W., Smits, W.K., and Kuipers, O.P. (2008). Bistability, epigenetics, and 623 bet-hedging in bacteria. Annu. Rev. Microbiol. 62, 193–210. 624

Veening, J.-W., Murray, H., and Errington, J. (2009). A mechanism for cell cycle 625 regulation of sporulation initiation in Bacillus subtilis. Genes Dev. 23, 1959– 626 1970. 627

van Verseveld, H.W., Arbige, M., and R. Chesbro, W. (1984). Continuous culture of 628 bacteria with biomass retention. Trends Biotechnol. 2, 8–12. 629

van Verseveld, H.W., de Hollander, J.A., Frankena, J., Braster, M., Leeuwerik, F.J., 630 and Stouthamer, A.H. (1986). Modeling of microbial substrate conversion, 631 growth and product formation in a recycling fermentor. Antonie Van 632 Leeuwenhoek 52, 325–342. 633

Vollmer, W., Joris, B., Charlier, P., and Foster, S. (2008). Bacterial peptidoglycan 634 (murein) hydrolases. FEMS Microbiol. Rev. 32, 259–286. 635

Winstedt, L., and von Wachenfeldt, C. (2000). Terminal oxidases of Bacillus subtilis 636 strain 168: one quinol oxidase, cytochrome aa3 or cytochrome bd, is required 637 for aerobic growth. J. Bacteriol. 182, 6557–6564. 638

Winstedt, L., Yoshida, K., Fujita, Y., and von Wachenfeldt, C. (1998). Cytochrome 639 bd biosynthesis in Bacillus subtilis: characterization of the cydABCD operon. J. 640 Bacteriol. 180, 6571–6580. 641

642

643

Chapter 4

114

644

university of groningen bacillus subtilis at near-zero ... · pdf filebacillus subtilis at...

Documents