Asymmetric chromosome segregation and cell division in DNA … · 4 Suchitha Raghunathan1,2, Afroze Chimthanawala1,3, Sandeep Krishna1,4, Anthony G. 5 Vecchiarelli5 and Anjana Badrinarayanan1*

1

Asymmetric chromosome segregation and cell division in DNA damage-1

induced bacterial filaments 2

3

Suchitha Raghunathan1,2, Afroze Chimthanawala1,3, Sandeep Krishna1,4, Anthony G. 4

Vecchiarelli5 and Anjana Badrinarayanan1* 5

6 1National Centre for Biological Sciences (Tata Institute of Fundamental Research), 7

Bangalore, India, 2Transdisciplinary University (TDU), Bangalore, India, 3SASTRA 8

University, Tanjore, India, 4Simons Centre for the Study of Living Machines, 1National 9

Centre for Biological Sciences (Tata Institute of Fundamental Research), Bangalore, India, 10 5Molecular, Cellular, and Developmental Biology Department, Biological Sciences 11

Building, University of Michigan, Ann Arbor, Michigan, USA 12

Correspondence to: *[email protected] 13

14

Keywords 15

Escherichia coli, DNA damage, filamentation, cell size maintenance, chromosome 16

segregation, cell division, time-lapse imaging, SOS response, bacteria. 17

18

Running title: Division dynamics and cell size maintenance in DNA damage-induced 19

bacterial filaments 20

21

.CC-BY 4.0 International license(which was not certified by peer review) is the author/funder. It is made available under aThe copyright holder for this preprintthis version posted March 16, 2020. . https://doi.org/10.1101/2020.03.16.993485doi: bioRxiv preprint

https://doi.org/10.1101/2020.03.16.993485

http://creativecommons.org/licenses/by/4.0/

2

Abstract 22

Faithful propagation of life requires coordination of DNA replication and 23

segregation with cell growth and division. In bacteria, this results in cell size homeostasis 24

and periodicity in replication and division. The situation is perturbed under stress such as 25

DNA damage, which induces filamentation as cell cycle progression is blocked to allow for 26

repair. Mechanisms that release this morphological state for re-entry into wild type 27

growth are unclear. Here we show that damage recovery is mediated via asymmetric 28

division of Escherichia coli filaments, producing short daughter cells with wild type size 29

and growth dynamics. Division restoration at this polar site is governed by coordinated 30

action of divisome positioning by the Min system and modulation of division licensing by 31

the terminus region of the chromosome, with MatP playing a central role in this process. 32

Collectively, our study highlights a key role for concurrency between chromosome (and 33

specifically terminus) segregation and cell division in daughter cell size maintenance 34

during filamentous divisions and suggests a central function for asymmetric division in 35

mediating cellular recovery from a stressed state. 36

37


https://doi.org/10.1101/2020.03.16.993485


3

Introduction 38

For successful cell division to occur, accurate DNA duplication and segregation 39

must be completed. In bacteria, chromosome replication initiates bi-directionally from an 40

‘origin’ and finishes opposite to this position, at the ‘terminus’ (Reyes-Lamothe and 41

Sherratt, 2019; Kleckner et al., 2014, 2018). Several factors ensure that cells divide only 42

upon completion of this process by regulating the multiprotein division machinery called 43

the ‘divisome’ (Galli and Gerdes, 2012; Dewachter et al., 2018; Kleckner et al., 2018; 44

Männik et al., 2016). For example, E. coli encodes negative regulators of the tubulin 45

homolog FtsZ, that is required to initiate the assembly of the divisome at the division 46

plane. Nucleoid occlusion by SlmA prevents the formation of the FtsZ-ring at locations 47

where chromosomal DNA is present and MinCDE oscillations direct the position of the Z-48

ring near mid-cell (Tonthat et al., 2011; Bernhardt and de Boer, 2005; Tsang and 49

Bernhardt, 2015). Recent studies have also suggested coordination of division with the 50

terminus via proteins such as MatP and ZapAB that act as a bridge between the DNA as 51

well as the divisome (Espéli et al., 2012; Mercier et al., 2008; Männik et al., 2016). 52

Together, in unperturbed laboratory conditions, this results in daughter cells that 53

replicate and divide in a periodic manner and that do not show much deviation in birth 54

and division cell sizes (Donachie, 1968; Taheri-Araghi et al., 2015; Wallden et al., 2016; 55

Micali et al., 2018; Campos et al., 2014; Si et al., 2019; Harris and Theriot, 2016). Such size 56

maintenance has been described in other bacteria as well as eukaryotic systems 57

(Chandler-Brown et al., 2017; Lambert et al., 2018; Soifer et al., 2016). 58

Recent investigations in bacteria have proposed that cells follow an ‘adder’-based 59

principle that governs size homeostasis either via regulation at the stage of birth 60

(replication initiation) or division or both (Taheri-Araghi et al., 2015; Wallden et al., 2016; 61

Campos et al., 2014; Si et al., 2019). In addition, some studies have also proposed a role 62

for cell shape or concurrency between processes of replication and division in size control 63

(Micali et al., 2018; Harris and Theriot, 2018). This homeostatic state is perturbed under 64

conditions of stress, a situation that can be often faced by bacterial cells in their 65

environment (Horvath et al., 2011; Justice et al., 2008; Yang et al., 2016; Heinrich et al., 66


https://doi.org/10.1101/2020.03.16.993485


4

2019). For example, under DNA damage, a cell cycle checkpoint blocks cell division until 67

damage has been repaired. The bacterial SOS response is activated upon the binding of 68

RecA to single-stranded DNA that is exposed as a consequence of DNA damage 69

(Mukherjee et al., 1998; Jonas, 2014; Kreuzer, 2013). As part of this response, a cell 70

division inhibitor (such as SulA in E. coli) blocks polymerization of FtsZ, resulting in cellular 71

elongation or filamentation (Kreuzer, 2013; Mukherjee et al., 1998). Cell division 72

inhibition during damage is a conserved process, even though the effectors may vary 73

across bacteria. SOS-induced division inhibition is carried out by SidA in Caulobacter and 74

YneA in Bacillus (Mukherjee et al., 1998; Jonas, 2014; Modell et al., 2011; Mo and 75

Burkholder, 2010). SOS-independent DNA damage-induced division inhibitors have also 76

been identified, suggesting that this is an important step in DNA repair (Modell et al., 77

2014). Along with blocking division, chromosome cohesion or de-condensation is also 78

initiated in these filaments. It is thought that this can aid recombination-based repair 79

(Vickridge et al., 2017; Odsbu and Skarstad, 2014). Indeed, cells have also been shown to 80

change shape and size under other forms of stress including host environments, heat 81

shock and osmotic fluctuations (Justice et al., 2008; Heinrich et al., 2019; Wehrens et al., 82

2018; Kysela et al., 2016; Caccamo and Brun, 2018; Bos et al., 2015; Yang et al., 2016). 83

Together, this highlights the plasticity with which bacteria such as E. coli sample a range 84

of cell sizes including filamentous and non-filamentous cell lengths. While the process by 85

which cell division is regulated to result in elongation under DNA damage has been well-86

characterized (Mukherjee et al., 1998; Suzuki et al., 1967; Jonas, 2014; Modell et al., 2014, 87

2011; Mo and Burkholder, 2010; Kantor and Deering, 1966; Adler and Hardigree, 1965), 88

how such a state is exited after repair to reinitiate wild type, periodic replication and 89

division remains unclear. 90

In this study, we probe the mechanism by which filamentous E. coli reinitiate 91

chromosome segregation and cell division after DNA repair. We use single-cell, time-92

resolved fluorescence microscopy to follow the kinetics of division restoration after cells 93

face a pulse of DNA damage and observe that size is maintained in daughter cells 94

generated from dividing filaments. Size homeostasis in daughters is not governed by the 95


https://doi.org/10.1101/2020.03.16.993485


5

growth laws (in the filament) and is instead dictated by cell division. We further find that 96

division restoration is controlled by two steps: determining where and when to divide. 97

This stepwise process, regulated by a combination of MinCDE oscillations, FtsZ levels and 98

terminus segregation, is accompanied by asymmetric partitioning of repaired 99

chromosomes, resulting in the production of daughter cells of the right size and devoid of 100

DNA damage, thus facilitating recovery from a stressed state. 101

102

Results 103

DNA-damage induced filaments maintain daughter cell size during division 104

To understand how damage-induced filamentous E. coli (Supplementary video 1) 105

re-initiate cell division and wild-type growth after DNA damage, we followed division 106

restoration in cells after treatment with a sub-inhibitory dose of the DNA damaging agent, 107

Mitomycin-C via time-lapse imaging (1μg/ ml; (Dapa et al., 2017)) (Fig. 1A). While 108

unperturbed wild type cells divided at ~mid-cell (Fig. 1B), we found that a significant 109

proportion of cells deviated from this division pattern as they increased in length (Fig. 1C-110

D). Damage-treated cells close to wild type length (5-10 μm) tended to divide in the 111

middle resulting in production of two daughter cells of similar sizes. In contrast, 112

filamentous cells divided polarly (or asymmetrically) to produce a short ‘daughter’ cell 113

(SD) of size similar to wild-type and a long cell that continued to filament (LD) (Fig. 1C and 114

1E; Supplementary video 2). The probability of a cell to undergo asymmetric division 115

increased with increasing cell length with 85% cells dividing asymmetrically at lengths 116

>12μm. Varying durations of damage exposure (30, 60 or 90 min) resulted in different 117

degrees of filamentation. However, in each case, a filamentous cell at division resulted in 118

production of an SD that was close to wild type size (Fig. 1F-G and Fig. S1A-B). This 119

asymmetry in division was observed in damage-induced filaments independent of growth 120

media (Fig. S1C). To further test whether the switch from mid-cell to asymmetric division 121

was a consequence of filamentation, we treated E. coli with the cell division inhibitor, 122

cephalexin, where genome integrity is maintained and chromosome replication and 123

segregation takes place as wild type. However, cells are unable to divide, resulting in 124


https://doi.org/10.1101/2020.03.16.993485


6

elongation in length (Rolinson, 1980; Chung et al., 2009). Cephalexin-treated filamentous 125

cells also generated daughter cells (SD) of fixed length via asymmetric division upon 126

removal of the inhibitor (Fig. S1D; Supplementary video 3). Together with previous 127

observations (Wehrens et al., 2018; Taschner et al., 1988; Begg and Doanachie, 1977), our 128

data suggest that filamentous cells, irrespective of types or durations of stress treatment, 129

engage in this form of division that produces daughter cells of wild type size. 130

To characterize the recovery process further in DNA damage-induced filaments, 131

we followed the fate of the filament (LD) and the SD over time. We observed that only 132

16±2% of filaments returned to wild type cell size during recovery in the three treatment 133

regimens (Fig. 1H). However, filamentous cells underwent multiple divisions in a 1 hr time 134

period, generating daughter cells (SD) of wild type size at each division (Fig. 1I). In the 135

same time, wild type cells undergo three divisions on average, suggesting that more 136

daughter cells are produced from a filament than a cell of same size as wild type during 137

recovery. In contrast to the filaments, daughter cells of wild type size (SD) displayed 138

growth and division dynamics similar to non-damage conditions (Fig. 1H and 1J). As an 139

example, in Fig. 1E, we followed the fate of an SD (4.6 μm) generated from a filament and 140

found that the daughter cell reinitiated wild type growth dynamics soon after division. 141

Time taken between divisions for SD was close to the distribution seen for wild type, while 142

LD continued to grow and divide as damage-treated cells (Fig. 1J). Consistently, the DNA 143

damage marker, RecA, formed multiple foci in LD at the time of division, while SD had RecA 144

localization as seen for wild type cells (Lesterlin et al., 2014; Rajendram et al., 2015; 145

Vickridge et al., 2017) or elongated cells with no DNA damage (cephalexin treated) (Fig. 146

S1E). In line with these observations, we found that viable cell count as well as cell length 147

distribution was restored to close to that of wild type in the three treatment regimes (Fig. 148

S1F-G). Thus, even though the filament may or may not recover from damage exposure, 149

recovery of the population may be mediated via several rounds of asymmetric divisions 150

from a single filament, resulting in daughter cells of wild type size and growth dynamics. 151

Daughter cell size is governed by cell division dynamics of filaments 152


https://doi.org/10.1101/2020.03.16.993485


7

How is daughter cell size maintained in dividing filaments? Recent investigations 153

have extensively characterized various mechanisms for chromosome segregation 154

checkpoints and adder-based cell size maintenance principles in wild type cells (Taheri-155

Araghi et al., 2015; Campos et al., 2014; Si et al., 2019; Harris and Theriot, 2016; Arjes et 156

al., 2014; Kleckner et al., 2014; Hill et al., 2012; Campos et al., 2018). To address this 157

question in the current context, we analysed the growth and division dynamics of 158

filaments. In damage-induced filaments, we found that length added between divisions 159

was not fixed (Fig. 2A) and did not correlate with length of the cell at birth, with several 160

instances of consecutive divisions without any elongation in between each division. To 161

test whether there is periodicity in division timing, we measured time between 162

consecutive division events. While wild type cells divided every ~20min, filamentous cells 163

displayed a large distribution of times between consecutive division events (Fig. S2A). 164

Although a lack of periodicity in division timing was observed, we found that 67±4% of 165

divisions took place at the opposite pole of the previous division, suggesting that there 166

may be no preference for old or new pole during division licensing (See Fig. S2D-G and 167

materials and methods for details on how divisions are computationally analysed). 168

After damage filament length increased exponentially with a characteristic rate, 169

as seen in wild type conditions (Fig. S2B-C), with some fluctuation which probably arises 170

because of stochasticity in damage and/ or repair. When length added or removed was 171

plotted as a function of birth length in filaments, no clear trend was observed (Fig. 2A). 172

However, when length added/ removed was plotted as a function of division times, we 173

observed that length removed (SD cell length) neither increased nor decreased with inter-174

division time, while length added increased as time between divisions increased (Fig. 2B). 175

This is in contrast to wild type cells where length added/ removed are constant (Fig. 2D-176

E). Overall, our observations suggest that the adder mechanism seen in steady state 177

conditions breaks down in filamentous cells undergoing asymmetric division (see 178

methods section for analysis). This could occur if growth rate changes, or if the 179

relationship between time to division and birth length deviates from what is expected 180

(eq. 3 in supplementary results for a model of this relationship), or both. In case of 181


https://doi.org/10.1101/2020.03.16.993485


8

elongated cells, we find that cells continue to grow exponentially at a characteristic rate 182

regardless of length, while the relation between time to division and birth length becomes 183

much more noisy. Indeed, when we plotted division time as a function of birth length, a 184

decreasing trend was most clearly observed only for wild type and less so (with more 185

noise) in the case of damage-treated cells (Fig. 2C and 2F). However, two properties of 186

filament division remain constant: a. length removed (SD cell size) is invariant and b. 187

division takes place poleward, asymmetrically. Taken together, this supports the idea that 188

daughter cell size is governed by regulation of cell division rather than filament growth 189

dynamics. 190

Cell division positioning and licensing have distinct regulatory mechanisms 191

Role of Min system in divisome positioning 192

As stated above, we noticed that cells transitioned from mid-cell to asymmetric 193

divisions as they grew longer in length (Fig. 1C). To test if the Min oscillation system 194

(Wehrens et al., 2018; de Boer et al., 1989; Raskin and de Boer, 1999; Tsang and 195

Bernhardt, 2015; Bisicchia et al., 2013; Dewachter et al., 2018) could determine division 196

licensing events in damage-induced filaments, we first asked whether Min oscillations are 197

maintained in these long cells. We imaged MinD-GFP and, consistent with Wehrens et al., 198

2018, found that it localized at several positions along the length of the cell and in an 199

equidistant manner, with a distance of 7 µm on average between each Min localization in 200

MMC or cephalexin-treated cells (Fig. S3A). As expected, we observed that cells had 1-2 201

Min localizations at lengths between 5-10µm and the number of Min localizations 202

increased with increasing cell length (Fig. S3B). Multiple Min localizations would suggest 203

that the number of divisome localizations would also proportionally increase. However, 204

when we imaged the localization of FtsZ (Supplementary video 4) in filaments, we found 205

that the number of FtsZ rings did not scale with increasing cell length and the position of 206

FtsZ shifted from away from mid-cell as cell length increased (Fig. S3C). Importantly, 207

deletion of the min operon resulted in loss of cell size maintenance with divisions that 208

could occur at mid-cell as well (Fig. 3A and S3E). Thus, as in wild type conditions, the Min 209

system may dictate daughter cell size maintenance in damage-induced filaments as well. 210


https://doi.org/10.1101/2020.03.16.993485


9

Even though daughter cell size maintenance was inaccurate in a min deletion, we 211

found that in both wild type and Dmin cells division occurred only one-site-at-a-time (Fig. 212

3D, S2D-G and S3E). Since we did not observe multiple FtsZ rings or multiple constrictions 213

in recovering filaments, we wondered whether nucleoid occlusion or negative regulation 214

of FtsZ polymerization (Kreuzer, 2013; Mukherjee et al., 1998; Tonthat et al., 2011; 215

Bernhardt and de Boer, 2005) may play a role in restricting the numbers and hence timing 216

of division, resulting in division occurring one-site-at-a-time. However, we found that 217

daughter cell length is maintained in sulA or slmA deletions and the number of 218

constrictions did not increase (Fig. 3B-C and S3D). We then wondered whether FtsZ levels 219

itself could be limiting in these cells (Bi and Lutkenhaus, 1990), thus permitting only one 220

division event at a time. To test this, we over-expressed FtsZ from an arabinose-inducible 221

promoter during damage or cephalexin recovery. In the case of damage-induced 222

filaments, we did not observe more than one constriction on average in both wild type 223

and FtsZ overexpression conditions, irrespective of the length of the cell (Fig. 3E and S3F). 224

Given that chromosome segregation is perturbed in these conditions, we reasoned that 225

the effect of FtsZ overexpression would be better observed in cephalexin-treated cells, 226

where 30% cells had more than one constriction site in wild type background (Fig. S3D). 227

We found that overexpression of FtsZ in cephalexin recovery resulted in the formation of 228

multiple constriction sites in long cells (Fig. S2G, S3D and S3F). However, division timing 229

did not become faster here or in the damage-recovery case (Fig. 3F) and division still 230

occurred one-site-at-a-time (Fig. S2D-G and S3F). It is possible that another divisome 231

component, such as FtsN (Coltharp et al., 2016), may be limiting in filaments, thus 232

resulting in single division events. While we cannot rule out regulation of the divisome at 233

the level of protein activity, transcriptional profiling of cells during recovery from MMC 234

treatment did not show down-regulation of key division components (Fig. 3G). Thus, 235

although the Min system governs division site locations, some other factor(s) determine 236

division site licensing/ timing. 237

Regulation of division licensing: impact of chromosome and terminus segregation 238


https://doi.org/10.1101/2020.03.16.993485


10

Given that chromosomes in DNA damage-induced filaments are no longer 239

segregated, it is possible chromosome dynamics in filaments could contribute to control 240

of division licensing. To ascertain the role of chromosome segregation in regulation of 241

division timing, we followed chromosomes after DNA damage via imaging the nucleoid-242

associated protein, HupA (Youngren et al., 2014; Marceau et al., 2011), tagged with 243

mCherry after MMC or cephalexin treatment (Fig. 4A; Supplementary video 5). Since 244

filamentous cells carry several copies of their chromosomes, yet division licensing occurs 245

towards a cell pole, we wondered whether there could be differences in chromosome 246

segregation dynamics across the length of the filament. Estimation of location of least 247

intensity of HupA fluorescence (as a proxy for nucleoid-free regions and thus 248

chromosome segregation) revealed that such regions were not near mid-cell in a majority 249

of damage or cephalexin-treated elongated cells (Fig. 4B). In contrast to wild type, where 250

median relative position of nucleoid-free gap from a cell pole was 0.46 (~mid-cell), the 251

location of this gap was broadly distributed between the cell pole and mid-cell with 252

median position of 0.33 and 0.34 from a cell pole for MMC and cephalexin-treated cells 253

respectively. Indeed, the location of the FtsZ-ring also coincided with where the 254

chromosomes were most segregated at the time of division in our damage-recovery 255

conditions (Fig. S4A-B). 256

Interestingly, we observed that early divisions in long filaments resulted in 15±6% 257

of daughter cells that were anucleated. These cells were distinct from mini-cells (Adler et 258

al., 1967) in that their cell size was close to that of wild type (3.5µm) (Fig. 4C). We found 259

that 38% of anucleate divisions occurred at the first division event during recovery. This 260

percentage reduced to <7% by the third division. Thus, we hypothesized that if 261

chromosome segregation has no effect on division dynamics then the time from divisome 262

assembly to cell division should be the same between anucleated and nucleated divisions. 263

However, if segregation had an effect, then the dynamics would vary between the two 264

types of divisions. In the case of divisions that resulted in production of anucleated cells, 265

we found that the average period between FtsZ assembly at the division site to cell 266

division was 7 min. In contrast, in case of nucleated divisions, we noticed that FtsZ 267


https://doi.org/10.1101/2020.03.16.993485


11

localized in the nucleoid-free regions for significantly longer, with an average of 12 min 268

prior to division (Fig. 4D). 269

We wondered what could be the rate limiting step in allowing division to occur, 270

thus resulting in persistence of FtsZ in nucleoid-free regions in the case of nucleated cell 271

divisions. Given that the terminus region of the chromosome is the last to be segregated 272

prior to division in wild type cells, we followed the dynamics of the terminus during 273

division in recovering filaments using the parS-ParB locus labelling system (Espéli et al., 274

2012; Nielsen et al., 2006) (Fig. S4C-D). While the bulk of the chromosome segregated 275

well-before division, and divisome components downstream of FtsZ, (ZapA and FtsN; 276

Supplementary video 6-7), localized 8 min and 6 min prior to division respectively, a single 277

terminus focus could be observed to persist in the chromosome-free region, where 278

constriction had begun to occur. The constriction completed to division just prior to or 279

concomitant with the termini splitting into two foci on either side of the division plane, 280

within 2 min on average (Fig. 4E, S4C-D). If terminus segregation is indeed key to division 281

licensing to generate nucleated, viable daughter cells, then perturbing this function 282

should affect division dynamics during recovery. A key player in modulating terminus 283

segregation specifically is MatP, which has been implicated in a. structuring the ter 284

macrodomain (via prevention of MukBEF interaction with the terminus) and b. connecting 285

the ter region with the divisome (Espéli et al., 2012; Lioy et al., 2018; Mercier et al., 2008; 286

Nolivos et al., 2016). We found that a matP deletion was compromised in damage 287

recovery (Fig. S4E). These cells had a significant increase in anucleate cell production, with 288

50±3% of divisions being anucleated. This is in contrast to wild type or deletions of slmA 289

or sulA that do not affect terminus segregation, but instead regulate divisome assembly 290

(Fig. 4F). In addition, nucleated daughter cell size maintenance was inaccurate in matP 291

deleted cells (Fig. 4G). These data are consistent with the idea that chromosome and 292

more specifically, terminus segregation acts as the licensor of nucleated divisions in 293

filaments. Taken together, this suggests that coordination between Min-mediated 294

divisome localization and MatP-mediated terminus positioning/ segregation regulates 295

daughter cell size maintenance, thus facilitating recovery from DNA damage (Fig. S5). 296


https://doi.org/10.1101/2020.03.16.993485


12

Discussion 297

In laboratory conditions, wild type E. coli maintains a distinct periodicity of cell 298

growth and division that appears to be coupled with chromosome replication and 299

segregation (Donachie, 1968; Taheri-Araghi et al., 2015; Wallden et al., 2016; Micali et al., 300

2018; Campos et al., 2014; Arjes et al., 2014; Kleckner et al., 2014; Hill et al., 2012). 301

However, it is becoming increasingly evident that bacteria can exist in diverse 302

morphological states, in part dictated by their environmental conditions (Yang et al., 303

2016; Justice et al., 2008; Heinrich et al., 2019; Jonas, 2014; Muraleedharan et al., 2018; 304

Kysela et al., 2016; Caccamo and Brun, 2018; MacCready and Vecchiarelli, 2018). Even E. 305

coli can become highly filamentous under conditions of stress such as during infection 306

(Horvath et al., 2011; Justice et al., 2004). Transitions into filamentous morphologies are 307

thought to confer several advantages such as avoiding phagocytosis via the host immune 308

response or providing a means to dilute the effects of any inhibitors present in the 309

surroundings (Yang et al., 2016; Justice et al., 2008; Horvath et al., 2011; Justice et al., 310

2004). Recent reports have also suggested that filamentation (at least in the case of E. coli 311

treated with ciprofloxacin) may be the first step towards the emergence of antibiotic 312

resistance as daughter cells carry mutations making them ciprofloxacin resistant (Bos et 313

al., 2015). Thus, it becomes important to understand how bacterial cells enter and exit 314

these filamentous states to ensure survival under stress. Indeed, several studies have 315

characterized mechanisms by which filamentation is induced under DNA damage 316

(Kreuzer, 2013; Mukherjee et al., 1998; Suzuki et al., 1967; Jonas, 2014; Modell et al., 317

2014, 2011; Adler and Hardigree, 1965; Uphoff, 2018). However, once DNA repair has 318

been completed, how cells restore wild type growth is unclear. 319

Here we show that cell cycle restoration involves asymmetric chromosome 320

segregation and cell division in filaments, resulting in the production of daughter cells of 321

wild type cell size and with undamaged DNA. While the filament itself may or may not 322

recover, several short daughters generated from a single filament go on to replicate and 323

divide as wild type cells. The concept of asymmetric partitioning of cellular components 324

during stress seems to have been co-opted by several bacterial systems (Schramm et al. 325


https://doi.org/10.1101/2020.03.16.993485


13

2020) including M. tuberculosis, where one daughter cell inherits the growing pole, while 326

the other has to assemble a growth pole de novo. This results in difference in susceptibility 327

to antibiotic treatment between the two cell types; the daughter cell with the growing 328

pole is more sensitive to cell wall synthesis inhibitors when compared to cells that 329

inherited the non-growing pole (Aakre and Laub, 2012; Aldridge et al., 2012). The 330

swarmer cells of P. mirabilis and V. parahaemolyticus (Muraleedharan et al., 2018; 331

MacCready and Vecchiarelli, 2018) use the Min-system to regulate asymmetric cell 332

division to allow for such division while still preserving the population of filamentous cells. 333

Dim-light stress induces filamentation in the photosynthetic cyanobacterium S. 334

elongatus, which then divides asymmetrically via positioning by the Min system (Liao and 335

Rust, 2018). Even in E. coli filaments generated in non-DNA damage conditions, studies 336

have reported poleward division events (Adler and Hardigree, 1965; Wehrens et al., 2018; 337

Taschner et al., 1988; Begg and Doanachie, 1977; Mileykovskaya et al., 1998). Taken 338

together, this suggests that switching from mid-cell to filamentation-based division may 339

be a universal method for cells under stress to ensure viable cell divisions. 340

Consistent with previous reports (Wehrens et al., 2018), we find that damage-341

treated cells switch from mid-cell to polar division in a size-dependent manner. Our 342

results suggest that the growth dynamics of filaments does not regulate size homeostasis 343

at division. Instead, size of the daughter cell is determined by a combination of Min 344

oscillations and chromosome segregation (Fig. S5). Cell growth may or may not occur to 345

accommodate both these events, consistent with the observation that consecutive 346

divisions can take place independent of addition of length in between, likely due to the 347

accumulation of already duplicated termini that undergo segregation. Previous studies in 348

wild type conditions have proposed that division timing could be modulated by factors 349

such as rates of constriction or concurrency between processes of chromosome 350

segregation and division (Micali et al., 2018; Lambert et al., 2018). In damage-induced 351

filaments, we find that cell division is composed of two distinct steps of division site 352

location and licensing. This results in a one-site-at-a-time cell division, which may serve 353

as a checkpoint to ensure accurate and complete chromosome segregation. As seen for 354


https://doi.org/10.1101/2020.03.16.993485


14

E. coli filaments under other stresses (Wehrens et al., 2018), the MinCDE system 355

contributes to determining the location of divisome assembly. As cell length changes, the 356

location of the Min oscillation nodes closest to the poles are likely to remain at a constant 357

relative position, whereas the relative position of oscillation nodes away from the poles 358

are likely to keep changing as the cell length changes. This would also explain why cells 359

switch from mid-cell to poleward division as they increase in length. Subsequent to this, 360

divisome assembly may be stabilized due to chromosome segregation away from mid-361

cell. Indeed, a recent study has shown that physical forces (such as molecular crowding) 362

squeeze or confine DNA towards mid-cell in filaments (Wu et al., 2019), which could 363

facilitate better or faster separation of chromosomes nearer the poles during segregation.364

Our data are consistent with the idea that once the division machinery has 365

assembled at a potential site of division, licensing occurs only when no DNA is 366

encountered by the divisome; thus even in the absence of chromosome segregation 367

anucleate cells can be produced (Mulder and Woldringh, 1989). However, upon initiation 368

of chromosome resegregation, it is likely that a division checkpoint is triggered by the 369

terminus region of the chromosome, via negatively regulating the ability of division to be 370

completed. Consistently, we find that chromosome segregation can impose a delay in 371

division, with the terminus being the last region of the chromosome to be segregated. 372

Specifically affecting terminus segregation via deletion of matP results in perturbation to 373

damage recovery, with a significant increase in anucleate cell production. These cells also 374

have a broad distribution of nucleated daughter cell sizes. Hence concurrency between 375

terminus segregation and cell division contributes to daughter size maintenance during 376

asymmetric cell division. Given that MatP has two independent functions of ter region 377

organization and ter anchoring with the divisome (Männik et al., 2016; Bailey et al., 2014; 378

Nolivos et al., 2016), it would be important to understand which of these activities is 379

specifically necessary in the recovery process. The system characterized in this study now 380

provides the ability to assess these rate-limiting steps of division licensing and further 381

probe the mechanisms of asymmetric chromosome segregation that preferentially results 382

in daughter cells devoid of damage. 383


https://doi.org/10.1101/2020.03.16.993485


15

In sum, our study highlights the requirement for coordination between two 384

independent mechanisms of divisome regulation for division accuracy in filaments. The 385

Min system facilitates localization of FtsZ in nucleoid-free regions. In parallel, modulation 386

of terminus region dynamics by MatP regulates the licensing of cell division. Concurrency 387

between terminus positioning and segregation with cell division ensures daughter size 388

maintenance in dividing filaments. Together, this results in the production of viable 389

daughter cells with wild type growth and division dynamics and contributes to the 390

restoration of cell size distribution, even when the filament itself may not recover. The 391

conservation of asymmetric division in a range of stress-induced bacterial filaments 392

underscores the importance of this mechanism in facilitating robust exit of cells from the 393

DNA damage checkpoint via the generation of fit, damage-free daughter cells. 394


https://doi.org/10.1101/2020.03.16.993485


16

Author contributions 395

SR: conception of project, experimental design, execution of experiments, data analysis 396

and writing of manuscript. AC: experimental design and execution of experiments and 397

analysis related to RecA. SK: data analysis and interpretation, writing of manuscript. AV: 398

reagents, tools and interpretation. AB: conception of project, experimental design, 399

writing of manuscript and funding. 400

401

Declaration of interests 402

The authors declare no competing interests. 403

404

Acknowledgements 405

The authors are grateful to Thomas Bernhardt, Steven Sandler, Suckjoon Jun, Bill 406

Soederstroem, Fred Boccard, Olivier Espeli, Christian Lesterlin, Ramanujam Srinivasan and 407

Manjula Reddy for generous sharing of strains and plasmids. The authors acknowledge 408

assistance from Ismath Sadhir in RNA extraction, Nitish Malhotra in RNA-seq analysis, 409

Aditya Jalin, Alex Sam Thomas and Aalok Varma in writing Matlab scripts as well as the 410

Central Imaging and Flow Facility (CIFF) and Next-generation genomics facility (NGGF). 411

The authors thank Fagwei Si, Piet de Boer, Raj Ladher, Michael Laub, Tung Le and 412

members of the AB lab for helpful discussions/ feedback on the manuscript. This work is 413

supported by funding from NCBS-TIFR (AB and SK), by an HFSP Career Development award 414

(AB) and funding from the Simons Foundation (SK). 415

416

Materials and Methods 417

Bacterial strains and growth conditions 418

Strains and plasmids used in the study are listed in Table 1. Transductions were conducted 419

with P1 phage following the protocol in (Thomason et al., 2007). For Escherichia coli, cells 420

were grown at 37°C in either rich media (LB: for 1 L dissolved 10 g tryptone, 5 g yeast 421

extract and 10 g NaCl in doubled distilled water) or minimal media (M9-Cas: for 1 L 422

dissolved 5 g glucose, 1 g casamino acids, 1 ml of 0.5% thiamine, 1 ml of 1M MgSO4 and 423


https://doi.org/10.1101/2020.03.16.993485


17

200 ml 5x M9 salts in double distilled water). DNA damage was induced with 1 μg/ml of 424

mitomycin C (MMC) for 30, 60 or 90 min (LB) or 90 min (M9-Cas) (unless otherwise 425

indicated). Cell division was inhibited using 5 μg/ml of cephalexin for 60 min (LB) or 90 426

min (M9-Cas). 427

Fluorescence Microscopy 428

Imaging was performed on a widefield, epifluorescence microscope (Eclipse Ti-2E, Nikon 429

with motorized xy-stage, Z-drift correction), 63X plan apochromat objective (NA1.41), 430

pE4000 light source (CoolLED), OkoLab incubation chamber and Hamamatsu Orca Flash 431

4.0 camera. Images were acquired using the NIS-elements software (version 5.1). 432

Microfluidics imaging was performed using the CellASIC-ONIX2 Microfluidic System, 433

Temperature Controlled CellASIC-ONIX2 Manifold XT and CellASIC ONIX Plate for Bacteria 434

Cells (B4A) (Merck). Details of the imaging and acquisition setting are described here 435

(Raghunathan and Badrinarayanan, 2019). 436

Time-course and Time-lapse imaging 437

For the recovery time-course, cells were grown overnight in LB or M9-Cas, back diluted 438

to ~OD600 0.01 and allowed to grow to OD600 ~0.1. Culture was then treated with MMC or 439

cephalexin for 60 min (LB) or 90 min (M9-Cas). Cells were pelleted, washed with fresh 440

media, resuspended to OD600 ~0.1 and then allowed to recover from damage treatment. 441

Cultures were maintained in log phase (OD600 ~0.1 - 0.4) throughout the experiment. For 442

time-course, samples for microscopy were collected at time points indicated. 1 ml of 443

culture was pelleted, resuspended in 100 μl and then spotted on 1% agarose pad and 444

imaged (Chimthanawala and Badrinarayanan, 2019). 445

For time-lapse experiments, damaged-induced cells were pelleted and washed with fresh 446

media and were either loaded into the microfluidics device or spotted on a 1.5% agarose 447

pad (made with appropriate growth media) and imaged. All time-lapse images were taken 448

every 30 sec or 2 min for ~3 hrs. For the FtsZ overexpression or FtsN imaging experiments, 449

cells were induced with either 0.3% arabinose or 0.1% rhamnose respectively for 1 hr 450

prior to imaging. For MinD-GFP imaging, cells were induced with 0.2mM IPTG for 90 min 451


https://doi.org/10.1101/2020.03.16.993485


18

prior to imaging. Inducers were maintained in the agarose pads or microfluidic plates. 452

Experiments were performed in LB unless otherwise indicated. 453

RNA sequencing 454

Protocol for RNA extraction described in (Badrinarayanan et al., 2017) was followed. 455

Briefly, cell pellets were collected during the recovery time-course at the indicated times. 456

RNA was extracted using the Direct-zol™ RNA MiniPrep (Zymo, Cat no. R2052A81) and 457

RNA Clean & Concentrator™-25 (Zymo, Cat no. R1018A82). RNA libraries were prepared 458

using the TruSeq Stranded mRNA Library Preparation kit at NCBS NGS facility. Libraries 459

were sequenced using Illumina MiSeq sequencing platform. Raw reads (single end; read 460

length = 50 bp) were obtained as .fastq files. The reference genome sequence (.fna) and 461

annotation (.gff) files for the same strain (accession number: NC_000913.3) were 462

downloaded from the ncbi ftp website ("ftp.ncbi.nlm.nih.gov"). The raw read quality was 463

checked using the FastQC software (version v0.11.5). BWA (version 0.7.12-r1039) 464

(Burroughs and Aravind, 2016; Li and Durbin, 2009) was used to index the reference 465

genome. Reads with raw read quality ≥20 were aligned using BWA aln -q option. 466

SAMTOOLS (version 0.1.19-96b5f2294a) (Li et al., 2009) was used to filter out the multiply 467

mapped reads. BEDTOOLS (version 2.25.0) (Quinlan and Hall, 2010) was used to calculate 468

the reads count per gene using the annotation file (.bed) in a strand specific manner. 469

Normalized counts per millions (cpmn) was obtained using TMM normalization of EdgeR 470

package. The cpmn across time was calculated relative to the no damage control. The log2 471

fold change in expression was plotted from this. 472

Image Analysis 473

Images acquired were visualized and processed using ImageJ (Schindelin et al., 2012, 474

2015). Segmentation was performed using Oufti (Paintdakhi et al., 2016). Foci tracking 475

was accomplished using the Spotfinder Z function of MicrobeTracker (Sliusarenko et al., 476

2011). For recovery time-course, custom Matlab scripts were used to extract cell length 477

and total fluorescence intensity information for each cell. Anucleate cells were identified 478

using a threshold of 0.02 a. u. for total intensity of HupA-GFP in the cells. For time-lapse 479

analysis, at each division, the longer cell was termed LD and the smaller cell given the 480


https://doi.org/10.1101/2020.03.16.993485


19

identity of SD. A custom Matlab script was used to the extract the lengths of the LD and SD 481

at each division, the length added between divisions and time between divisions. 482

Daughter cells were classified as ‘recovers’ if at their first division, their length was <10µm 483

and they divided symmetrically. Cells that were longer at their first division and divided 484

asymmetrically were classified as ‘filaments’. RecA foci numbers were obtained using 485

Spotfinder Z and combined with cell length information from Oufti. Number of FtsZ rings 486

in a cell were obtained using Spotfinder Z and combined with cell length information from 487

Oufti. For analysis of constrictions and divisions, we used Oufti, which allows sub-pixel 488

segmentation of phase contrast images of cells in a time-lapse. The method is illustrated 489

in figure S2 in the supplementary methods section. In (a), phase profile for a wild type cell 490

before and after division is plotted. While the profile is, on average, a flat line for most of 491

the time imaged, one can identify a constriction prior to division (marked with *). Similarly 492

in (b), the phase profile for a filamentous cell (DNA damage-induced) undergoing division 493

is plotted. If the phase profile deviates by 20-25% of the cell width, it is called a 494

constriction. If the phase profile is discontinuous (with a gap), then it is marked as a 495

division by the segmentation algorithm automatically. For nucleoid tracking, fluorescence 496

intensity profiles of HupA-mCherry, obtained from Oufti was used. Data was first 497

smoothened using a Savitzgy-Golay filter (Luo et al., 2005). Following this the fluorescence 498

profile was inverted so that the regions of lowest fluorescence now had the highest 499

values. Peaks were determined using the peak prominence function after setting a 500

threshold of half the maximum intensity for each frame. Peaks correspond to lowest 501

fluorescence intensity in the cell. Regions right next to the cell poles were excluded. This 502

information was combined with FtsZ foci tracked using Spotfinder Z to obtain relative 503

positions of the lowest fluorescence intensity in the cell and FtsZ ring. Formation of FtsZ, 504

ZapA and FtsN foci, and segregation of the single terminus focus (ParB-GFP) to 2 foci at 505

the site of division was scored manually along with time to division. 506

507


https://doi.org/10.1101/2020.03.16.993485


20

Main Figure legends 508

Figure 1: DNA-damage induced filaments maintain daughter cell size during division. a. 509

Representative time-lapse montage of filamentous cells during recovery. White asterisks 510

indicate divisions occurring towards a cell pole. Scale bar - 5 µm; time in min here and in 511

all other images. b. Cell length of two daughter cells generated from a single division in 512

wild type conditions. Each grey dot represents a single division event (n = 157). The red 513

line plots the expected values if all cells were dividing at their mid-point. c. Cell length of 514

long daughter (LD) and short daughter (SD) generated from a DNA damage-induced 515

filament during recovery. Cells are treated with mitomycin C (MMC) for 60 min. Each grey 516

dot represents a single division event (n = 531). The red line plots the expected values if 517

all cells were dividing at their mid-point. d. Location of division is plotted as a function of 518

cell length in filamentous E. coli during recovery from DNA damage treatment (60 min; n 519

= 531). e. Cell length of long daughter (LD) and short daughter (SD) is tracked over time 520

during damage recovery. Decrease in cell length is indicative of division. f-g. As (b-c) for 521

cells treated with MMC for 30 min. h. Fate of SD and LD during recovery. Cell is classified 522

as recovered if it undergoes mid-cell division and filamentous if it continues to filament 523

after division (n ≥ 56) i. Number divisions per cell in 1 hr for all durations of damage 524

treatment. As a control, number of divisions wild type cells undergo is also shown (n≥100 525

division). j. Distribution of time between divisions for wild type (no damage control), LD 526

and SD during recovery from MMC (n ≥ 104). 527

Figure 2: Daughter cell size is governed by cell division dynamics of filaments. a. Length 528

added to a cell between divisions (black dots) and length removed at the latter division 529

(i.e., the length of the daughter cell) (blue dots) as a function of the birth length of the 530

cell for MMC-treated cells. b. Length added to a cell between divisions (black dots) and 531

length removed at the division (i.e., the length of the daughter cell) (blue dots) as a 532

function of the time between divisions for MMC-treated cells. c. Time between divisions 533

as a function of cell length at birth. Red line shows the relation that would be necessary 534

for the system to be an adder (equation (3) in supplementary results), given that cells are 535

growing exponentially with the rates given in Fig. S2B-C. d-f. (As a-c) for wild type cells. 536


https://doi.org/10.1101/2020.03.16.993485


21

Figure 3: Cell division positioning and licensing have distinct regulatory mechanisms. 537

Determining where to divide: role of Min system. a-c. Cell length of long daughter (LD) 538

and short daughter (SD) generated from a DNA damage-induced filament during recovery 539

for ΔminCDE (n=186), ΔslmA (n=144) and ΔsulA (n=246) backgrounds respectively (grey 540

dots). As a reference, lengths for wild type during recovery are shown in orange. The red 541

line plots the expected values if all cells were dividing at their mid-point. d. Representative 542

time-lapse montage of division in wild type cells during damage recovery. e. Number of 543

constrictions per cell plotted as a function of cell length during DNA damage recovery for 544

wild type and cells over-expressing FtsZ (n ≥ 91). f. Time between divisions for no damage 545

(control) and for cells after treatment with MMC or cephalexin with (++FtsZ) or without 546

(wild type) FtsZ overexpression (n ≥ 93). g. Heat map of transcript levels (from RNA-seq) 547

of genes involved in cell division during a damage recovery time-course. As a control, 548

genes induced under the SOS response are also highlighted (bold). Log2-fold change 549

normalized to control without damage is plotted. 550

Figure 4: Cell division positioning and licensing have distinct regulatory mechanisms. 551

Regulation of division licensing: impact of chromosome and terminus segregation. a. 552

Representative time-lapse montage of division in cells during recovery. Grey – phase, red 553

– HupA-mCherry (chromosome); scale bar - 5 µm; time in min. b. Position of least intensity 554

of HupA fluorescence (gaps between chromosomes) plotted as a function of cell length 555

(from one pole to mid-cell) in recovering MMC or cephalexin-treated filaments. As 556

reference, these data are also shown for wild type cells with no damage treatment 557

(control) (n ≥ 150). c. Cell length distribution for wild type cells (no perturbation) is 558

plotted. Along with this, cell length distribution of daughter cells during DNA damage 559

recovery is plotted for nucleate cells and anucleate cells. To highlight the distinction 560

between these cell division events and minicell formation, cell length distribution of 561

minicells (from min deleted cells) is also shown (n ≥ 100) d. Time from FtsZ localization to 562

division completion is plotted for divisions that result in nucleated or anucleated SD cells. 563

(n ≥ 38) e. Distribution of time to division after FtsZ, ZapA and FtsN localization to division 564

site is plotted. Along with this, time to division after segregation of terminus (ter) during 565


https://doi.org/10.1101/2020.03.16.993485


22

recovery after MMC treatment is also shown (n ≥ 93). f. Percentage of SD that are 566

anucleate, recover and filament is plotted for wild type and deletions of matP, sulA or 567

slmA during DNA damage recovery. (n ≥ 103) g. Daughter cell length distribution for 568

nucleated divisions in the case of matP deletion and wild type is plotted. As a reference 569

cell length distribution for cells without damage treatment is also plotted for each case (n 570

= 145). 571


https://doi.org/10.1101/2020.03.16.993485


23

References: 572 Aakre, C.D., and M.T. Laub. 2012. Asymmetric cell division: a persistent issue? Dev Cell. 22:235–573

236. doi:10.1016/j.devcel.2012.01.016. 574

Adler, H.I., W.D. Fisher, A. Cohen, and A.A. Hardigree. 1967. MINIATURE escherichia coli CELLS 575 DEFICIENT IN DNA*. Proc Natl Acad Sci U S A. 57:321–326. 576

Adler, H.I., and A.A. Hardigree. 1965. Growth and Division of Filamentous Forms of Escherichia 577 coli. J Bacteriol. 90:223–226. 578

Aldridge, B.B., M. Fernandez-Suarez, D. Heller, V. Ambravaneswaran, D. Irimia, M. Toner, and 579 S.M. Fortune. 2012. Asymmetry and aging of mycobacterial cells lead to variable growth 580 and antibiotic susceptibility. Science. 335:100–104. doi:10.1126/science.1216166. 581

Arjes, H.A., A. Kriel, N.A. Sorto, J.T. Shaw, J.D. Wang, and P.A. Levin. 2014. Failsafe mechanisms 582 couple division and DNA replication in bacteria. Curr. Biol. 24:2149–2155. 583 doi:10.1016/j.cub.2014.07.055. 584

Badrinarayanan, A., T.B.K. Le, J.-H. Spille, I.I. Cisse, and M.T. Laub. 2017. Global analysis of 585 double-strand break processing reveals in vivo properties of the helicase-nuclease 586 complex AddAB. PLOS Genetics. 13:e1006783. doi:10.1371/journal.pgen.1006783. 587

Bailey, M.W., P. Bisicchia, B.T. Warren, D.J. Sherratt, and J. Männik. 2014. Evidence for Divisome 588 Localization Mechanisms Independent of the Min System and SlmA in Escherichia coli. 589 PLOS Genetics. 10:e1004504. doi:10.1371/journal.pgen.1004504. 590

Begg, K.J., and W.D. Doanachie. 1977. Growth of the Escherichia coli cell surface. J Bacteriol. 591 129:1524–1536. 592

Bernhardt, T.G., and P.A.J. de Boer. 2005. SlmA, a nucleoid-associated, FtsZ binding protein 593 required for blocking septal ring assembly over Chromosomes in E. coli. Mol. Cell. 594 18:555–564. doi:10.1016/j.molcel.2005.04.012. 595

Bi, E., and J. Lutkenhaus. 1990. FtsZ regulates frequency of cell division in Escherichia coli. J 596 Bacteriol. 172:2765–2768. 597

Bisicchia, P., S. Arumugam, P. Schwille, and D. Sherratt. 2013. MinC, MinD, and MinE Drive 598 Counter-oscillation of Early-Cell-Division Proteins Prior to Escherichia coli Septum 599 Formation. mBio. 4:e00856-13. doi:10.1128/mBio.00856-13. 600

de Boer, P.A., R.E. Crossley, and L.I. Rothfield. 1989. A division inhibitor and a topological 601 specificity factor coded for by the minicell locus determine proper placement of the 602 division septum in E. coli. Cell. 56:641–649. doi:10.1016/0092-8674(89)90586-2. 603

Bos, J., Q. Zhang, S. Vyawahare, E. Rogers, S.M. Rosenberg, and R.H. Austin. 2015. Emergence of 604 antibiotic resistance from multinucleated bacterial filaments. Proc. Natl. Acad. Sci. 605 U.S.A. 112:178–183. doi:10.1073/pnas.1420702111. 606


https://doi.org/10.1101/2020.03.16.993485


24

Burroughs, A.M., and L. Aravind. 2016. RNA damage in biological conflicts and the diversity of 607 responding RNA repair systems. Nucleic Acids Res. 44:8525–8555. 608 doi:10.1093/nar/gkw722. 609

Caccamo, P.D., and Y.V. Brun. 2018. The Molecular Basis of Noncanonical Bacterial Morphology. 610 Trends in Microbiology. 26:191–208. doi:10.1016/j.tim.2017.09.012. 611

Campos, M., S.K. Govers, I. Irnov, G.S. Dobihal, F. Cornet, and C. Jacobs-Wagner. 2018. 612 Genomewide phenotypic analysis of growth, cell morphogenesis, and cell cycle events in 613 Escherichia coli. Mol. Syst. Biol. 14:e7573. doi:10.15252/msb.20177573. 614

Campos, M., I.V. Surovtsev, S. Kato, A. Paintdakhi, B. Beltran, S.E. Ebmeier, and C. Jacobs-615 Wagner. 2014. A constant size extension drives bacterial cell size homeostasis. Cell. 616 159:1433–1446. doi:10.1016/j.cell.2014.11.022. 617

Chandler-Brown, D., K.M. Schmoller, Y. Winetraub, and J.M. Skotheim. 2017. The Adder 618 Phenomenon Emerges from Independent Control of Pre- and Post-Start Phases of the 619 Budding Yeast Cell Cycle. Curr. Biol. 27:2774-2783.e3. doi:10.1016/j.cub.2017.08.015. 620

Chimthanawala, A., and A. Badrinarayanan. 2019. Live-Cell Fluorescence Imaging of RecN in 621 Caulobacter crescentus Under DNA Damage. Methods Mol. Biol. 2004:239–250. 622 doi:10.1007/978-1-4939-9520-2_18. 623

Chung, H.S., Z. Yao, N.W. Goehring, R. Kishony, J. Beckwith, and D. Kahne. 2009. Rapid β-lactam-624 induced lysis requires successful assembly of the cell division machinery. PNAS. 625 106:21872–21877. doi:10.1073/pnas.0911674106. 626

Coltharp, C., J. Buss, T.M. Plumer, and J. Xiao. 2016. Defining the rate-limiting processes of 627 bacterial cytokinesis. Proceedings of the National Academy of Sciences. 113:E1044–628 E1053. doi:10.1073/pnas.1514296113. 629

Dapa, T., S. Fleurier, M.-F. Bredeche, and I. Matic. 2017. The SOS and RpoS Regulons Contribute 630 to Bacterial Cell Robustness to Genotoxic Stress by Synergistically Regulating DNA 631 Polymerase Pol II. Genetics. 206:1349–1360. doi:10.1534/genetics.116.199471. 632

Dewachter, L., N. Verstraeten, M. Fauvart, and J. Michiels. 2018. An integrative view of cell cycle 633 control in Escherichia coli. FEMS Microbiol. Rev. 42:116–136. 634 doi:10.1093/femsre/fuy005. 635

Donachie, W.D. 1968. Relationship between Cell Size and Time of Initiation of DNA Replication. 636 Nature. 219:1077–1079. doi:10.1038/2191077a0. 637

Espéli, O., R. Borne, P. Dupaigne, A. Thiel, E. Gigant, R. Mercier, and F. Boccard. 2012. A MatP–638 divisome interaction coordinates chromosome segregation with cell division in E. coli. 639 EMBO J. 31:3198–3211. doi:10.1038/emboj.2012.128. 640

Galli, E., and K. Gerdes. 2012. FtsZ-ZapA-ZapB Interactome of Escherichia coli. Journal of 641 Bacteriology. 194:292–302. doi:10.1128/JB.05821-11. 642


https://doi.org/10.1101/2020.03.16.993485


25

Harris, L.K., and J.A. Theriot. 2016. Relative Rates of Surface and Volume Synthesis Set Bacterial 643 Cell Size. Cell. 165:1479–1492. doi:10.1016/j.cell.2016.05.045. 644

Harris, L.K., and J.A. Theriot. 2018. Surface Area to Volume Ratio: A Natural Variable for Bacterial 645 Morphogenesis. Trends in Microbiology. 26:815–832. doi:10.1016/j.tim.2018.04.008. 646

Heinrich, K., D.J. Leslie, M. Morlock, S. Bertilsson, and K. Jonas. 2019. Molecular Basis and 647 Ecological Relevance of Caulobacter Cell Filamentation in Freshwater Habitats. MBio. 10. 648 doi:10.1128/mBio.01557-19. 649

Hill, N.S., R. Kadoya, D.K. Chattoraj, and P.A. Levin. 2012. Cell Size and the Initiation of DNA 650 Replication in Bacteria. PLOS Genetics. 8:e1002549. doi:10.1371/journal.pgen.1002549. 651

Horvath, D.J., B. Li, T. Casper, S. Partida-Sanchez, D.A. Hunstad, S.J. Hultgren, and S.S. Justice. 652 2011. Morphological plasticity promotes resistance to phagocyte killing of 653 uropathogenic Escherichia coli. Microbes and Infection. 13:426–437. 654 doi:10.1016/j.micinf.2010.12.004. 655

Jonas, K. 2014. To divide or not to divide: control of the bacterial cell cycle by environmental 656 cues. Curr. Opin. Microbiol. 18:54–60. doi:10.1016/j.mib.2014.02.006. 657

Justice, S.S., C. Hung, J.A. Theriot, D.A. Fletcher, G.G. Anderson, M.J. Footer, and S.J. Hultgren. 658 2004. Differentiation and developmental pathways of uropathogenic Escherichia coli in 659 urinary tract pathogenesis. PNAS. 101:1333–1338. doi:10.1073/pnas.0308125100. 660

Justice, S.S., D.A. Hunstad, L. Cegelski, and S.J. Hultgren. 2008. Morphological plasticity as a 661 bacterial survival strategy. Nat. Rev. Microbiol. 6:162–168. doi:10.1038/nrmicro1820. 662

Kantor, G.J., and R.A. Deering. 1966. Ultraviolet radiation studies of filamentous Escherichia coli 663 B. J. Bacteriol. 92:1062–1069. 664

Kleckner, N., J.K. Fisher, M. Stouf, M.A. White, D. Bates, and G. Witz. 2014. The bacterial 665 nucleoid: nature, dynamics and sister segregation. Current Opinion in Microbiology. 666 22:127–137. doi:10.1016/j.mib.2014.10.001. 667

Kleckner, N.E., K. Chatzi, M.A. White, J.K. Fisher, and M. Stouf. 2018. Coordination of Growth, 668 Chromosome Replication/Segregation, and Cell Division in E. coli. Front. Microbiol. 9. 669 doi:10.3389/fmicb.2018.01469. 670

Kreuzer, K.N. 2013. DNA Damage Responses in Prokaryotes: Regulating Gene Expression, 671 Modulating Growth Patterns, and Manipulating Replication Forks. Cold Spring Harbor 672 Perspectives in Biology. 5:a012674–a012674. doi:10.1101/cshperspect.a012674. 673

Kysela, D.T., A.M. Randich, P.D. Caccamo, and Y.V. Brun. 2016. Diversity Takes Shape: 674 Understanding the Mechanistic and Adaptive Basis of Bacterial Morphology. PLOS 675 Biology. 14:e1002565. doi:10.1371/journal.pbio.1002565. 676


https://doi.org/10.1101/2020.03.16.993485


26

Lambert, A., A. Vanhecke, A. Archetti, S. Holden, F. Schaber, Z. Pincus, M.T. Laub, E. Goley, and 677 S. Manley. 2018. Constriction Rate Modulation Can Drive Cell Size Control and 678 Homeostasis in C. crescentus. iScience. 4:180–189. doi:10.1016/j.isci.2018.05.020. 679

Lesterlin, C., G. Ball, L. Schermelleh, and D.J. Sherratt. 2014. RecA bundles mediate homology 680 pairing between distant sisters during DNA break repair. Nature. 506:249–253. 681 doi:10.1038/nature12868. 682

Li, H., and R. Durbin. 2009. Fast and accurate short read alignment with Burrows-Wheeler 683 transform. Bioinformatics. 25:1754–1760. doi:10.1093/bioinformatics/btp324. 684

Li, H., B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, G. Marth, G. Abecasis, R. Durbin, 685 and 1000 Genome Project Data Processing Subgroup. 2009. The Sequence 686 Alignment/Map format and SAMtools. Bioinformatics. 25:2078–2079. 687 doi:10.1093/bioinformatics/btp352. 688

Liao, Y., and M.J. Rust. 2018. The Min Oscillator Defines Sites of Asymmetric Cell Division in 689 Cyanobacteria during Stress Recovery. Cell Syst. 7:471-481.e6. 690 doi:10.1016/j.cels.2018.10.006. 691

Lioy, V.S., A. Cournac, M. Marbouty, S. Duigou, J. Mozziconacci, O. Espéli, F. Boccard, and R. 692 Koszul. 2018. Multiscale Structuring of the E. coli Chromosome by Nucleoid-Associated 693 and Condensin Proteins. Cell. 172:771-783.e18. doi:10.1016/j.cell.2017.12.027. 694

Luo, J., K. Ying, and J. Bai. 2005. Savitzky-Golay Smoothing and Differentiation Filter for Even 695 Number Data. Signal Process. 85:1429–1434. doi:10.1016/j.sigpro.2005.02.002. 696

MacCready, J.S., and A.G. Vecchiarelli. 2018. In long bacterial cells, the Min system can act off-697 center. Mol. Microbiol. 109:268–272. doi:10.1111/mmi.13995. 698

Männik, J., D.E. Castillo, D. Yang, G. Siopsis, and J. Männik. 2016. The role of MatP, ZapA and 699 ZapB in chromosomal organization and dynamics in Escherichia coli. Nucleic Acids Res. 700 44:1216–1226. doi:10.1093/nar/gkv1484. 701

Marceau, A.H., S. Bahng, S.C. Massoni, N.P. George, S.J. Sandler, K.J. Marians, and J.L. Keck. 702 2011. Structure of the SSB-DNA polymerase III interface and its role in DNA replication. 703 EMBO J. 30:4236–4247. doi:10.1038/emboj.2011.305. 704

Mercier, R., M.-A. Petit, S. Schbath, S. Robin, M. El Karoui, F. Boccard, and O. Espéli. 2008. The 705 MatP/matS site-specific system organizes the terminus region of the E. coli chromosome 706 into a macrodomain. Cell. 135:475–485. doi:10.1016/j.cell.2008.08.031. 707

Micali, G., J. Grilli, M. Osella, and M.C. Lagomarsino. 2018. Concurrent processes set E. coli cell 708 division. Science Advances. 4:eaau3324. doi:10.1126/sciadv.aau3324. 709

Mileykovskaya, E., Q. Sun, W. Margolin, and W. Dowhan. 1998. Localization and function of early 710 cell division proteins in filamentous Escherichia coli cells lacking 711 phosphatidylethanolamine. J. Bacteriol. 180:4252–4257. 712


https://doi.org/10.1101/2020.03.16.993485


27

Mo, A.H., and W.F. Burkholder. 2010. YneA, an SOS-Induced Inhibitor of Cell Division in Bacillus 713 subtilis, Is Regulated Posttranslationally and Requires the Transmembrane Region for 714 Activity. Journal of Bacteriology. 192:3159–3173. doi:10.1128/JB.00027-10. 715

Modell, J.W., A.C. Hopkins, and M.T. Laub. 2011. A DNA damage checkpoint in Caulobacter 716 crescentus inhibits cell division through a direct interaction with FtsW. Genes Dev. 717 25:1328–1343. doi:10.1101/gad.2038911. 718

Modell, J.W., T.K. Kambara, B.S. Perchuk, and M.T. Laub. 2014. A DNA damage-induced, SOS-719 independent checkpoint regulates cell division in Caulobacter crescentus. PLoS Biol. 720 12:e1001977. doi:10.1371/journal.pbio.1001977. 721

Mukherjee, A., C. Cao, and J. Lutkenhaus. 1998. Inhibition of FtsZ polymerization by SulA, an 722 inhibitor of septation in Escherichia coli. PNAS. 95:2885–2890. 723 doi:10.1073/pnas.95.6.2885. 724

Mulder, E., and C.L. Woldringh. 1989. Actively replicating nucleoids influence positioning of 725 division sites in Escherichia coli filaments forming cells lacking DNA. J Bacteriol. 726 171:4303–4314. 727

Muraleedharan, S., C. Freitas, P. Mann, T. Glatter, and S. Ringgaard. 2018. A cell length-728 dependent transition in MinD-dynamics promotes a switch in division-site placement 729 and preservation of proliferating elongated Vibrio parahaemolyticus swarmer cells. Mol. 730 Microbiol. 109:365–384. doi:10.1111/mmi.13996. 731

Nielsen, H.J., J.R. Ottesen, B. Youngren, S.J. Austin, and F.G. Hansen. 2006. The Escherichia coli 732 chromosome is organized with the left and right chromosome arms in separate cell 733 halves: E. coli chromosome segregation. Molecular Microbiology. 62:331–338. 734 doi:10.1111/j.1365-2958.2006.05346.x. 735

Nolivos, S., A.L. Upton, A. Badrinarayanan, J. Müller, K. Zawadzka, J. Wiktor, A. Gill, L. 736 Arciszewska, E. Nicolas, and D. Sherratt. 2016. MatP regulates the coordinated action of 737 topoisomerase IV and MukBEF in chromosome segregation. Nature Communications. 738 7:10466. doi:10.1038/ncomms10466. 739

Odsbu, I., and K. Skarstad. 2014. DNA compaction in the early part of the SOS response is 740 dependent on RecN and RecA. Microbiology. 160:872–882. doi:10.1099/mic.0.075051-0. 741

Paintdakhi, A., B. Parry, M. Campos, I. Irnov, J. Elf, I. Surovtsev, and C. Jacobs-Wagner. 2016. 742 Oufti: an integrated software package for high-accuracy, high-throughput quantitative 743 microscopy analysis. Mol. Microbiol. 99:767–777. doi:10.1111/mmi.13264. 744

Quinlan, A.R., and I.M. Hall. 2010. BEDTools: a flexible suite of utilities for comparing genomic 745 features. Bioinformatics. 26:841–842. doi:10.1093/bioinformatics/btq033. 746

Raghunathan, S., and A. Badrinarayanan. 2019. Tracking Bacterial Chromosome Dynamics with 747 Microfluidics-Based Live Cell Imaging. Methods Mol. Biol. 2004:223–238. 748 doi:10.1007/978-1-4939-9520-2_17. 749


https://doi.org/10.1101/2020.03.16.993485


28

Rajendram, M., L. Zhang, B.J. Reynolds, G.K. Auer, H.H. Tuson, K.V. Ngo, M.M. Cox, A. Yethiraj, Q. 750 Cui, and D.B. Weibel. 2015. Anionic Phospholipids Stabilize RecA Filament Bundles in 751 Escherichia coli. Molecular Cell. 60:374–384. doi:10.1016/j.molcel.2015.09.009. 752

Raskin, D.M., and P.A.J. de Boer. 1999. Rapid pole-to-pole oscillation of a protein required for 753 directing division to the middle of Escherichia coli. Proc Natl Acad Sci U S A. 96:4971–754 4976. 755

Reyes-Lamothe, R., and D.J. Sherratt. 2019. The bacterial cell cycle, chromosome inheritance 756 and cell growth. Nat. Rev. Microbiol. 17:467–478. doi:10.1038/s41579-019-0212-7. 757

Rolinson, G.N. 1980. Effect of β-Lactam Antibiotics on Bacterial Cell Growth Rate. Microbiology,. 758 120:317–323. doi:10.1099/00221287-120-2-317. 759

Schindelin, J., I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. 760 Rueden, S. Saalfeld, B. Schmid, J.-Y. Tinevez, D.J. White, V. Hartenstein, K. Eliceiri, P. 761 Tomancak, and A. Cardona. 2012. Fiji: an open-source platform for biological-image 762 analysis. Nat. Methods. 9:676–682. doi:10.1038/nmeth.2019. 763

Schindelin, J., C.T. Rueden, M.C. Hiner, and K.W. Eliceiri. 2015. The ImageJ ecosystem: An open 764 platform for biomedical image analysis. Mol. Reprod. Dev. 82:518–529. 765 doi:10.1002/mrd.22489. 766

Schramm, F.D., K. Schroeder, J. Alvelid, I. Testa, and K. Jonas. Growth-driven displacement of 767 protein aggregates along the cell length ensures partitioning to both daughter cells in 768 Caulobacter crescentus. Molecular Microbiology. 0. doi:10.1111/mmi.14228. 769

Si, F., G. Le Treut, J.T. Sauls, S. Vadia, P.A. Levin, and S. Jun. 2019. Mechanistic Origin of Cell-Size 770 Control and Homeostasis in Bacteria. Curr. Biol. 29:1760-1770.e7. 771 doi:10.1016/j.cub.2019.04.062. 772

Sliusarenko, O., J. Heinritz, T. Emonet, and C. Jacobs-Wagner. 2011. High-throughput, subpixel 773 precision analysis of bacterial morphogenesis and intracellular spatio-temporal 774 dynamics. Molecular Microbiology. 80:612–627. doi:10.1111/j.1365-2958.2011.07579.x. 775

Soifer, I., L. Robert, and A. Amir. 2016. Single-Cell Analysis of Growth in Budding Yeast and 776 Bacteria Reveals a Common Size Regulation Strategy. Current Biology. 26:356–361. 777 doi:10.1016/j.cub.2015.11.067. 778

Suzuki, H., J. Pangborn, and W.W. Kilgore. 1967. Filamentous Cells of Escherichia coli Formed in 779 the Presence of Mitomycin. J Bacteriol. 93:683–688. 780

Taheri-Araghi, S., S. Bradde, J.T. Sauls, N.S. Hill, P.A. Levin, J. Paulsson, M. Vergassola, and S. Jun. 781 2015. Cell-size control and homeostasis in bacteria. Curr. Biol. 25:385–391. 782 doi:10.1016/j.cub.2014.12.009. 783

Taschner, P.E., P.G. Huls, E. Pas, and C.L. Woldringh. 1988. Division behavior and shape changes 784 in isogenic ftsZ, ftsQ, ftsA, pbpB, and ftsE cell division mutants of Escherichia coli during 785


https://doi.org/10.1101/2020.03.16.993485


29

temperature shift experiments. J. Bacteriol. 170:1533–1540. doi:10.1128/jb.170.4.1533-786 1540.1988. 787

Thomason, L.C., N. Costantino, and D.L. Court. 2007. E. coli genome manipulation by P1 788 transduction. Curr Protoc Mol Biol. Chapter 1:Unit 1.17. 789 doi:10.1002/0471142727.mb0117s79. 790

Tonthat, N.K., S.T. Arold, B.F. Pickering, M.W. Van Dyke, S. Liang, Y. Lu, T.K. Beuria, W. Margolin, 791 and M.A. Schumacher. 2011. Molecular mechanism by which the nucleoid occlusion 792 factor, SlmA, keeps cytokinesis in check. EMBO J. 30:154–164. 793 doi:10.1038/emboj.2010.288. 794

Tsang, M.-J., and T.G. Bernhardt. 2015. Guiding divisome assembly and controlling its activity. 795 Curr. Opin. Microbiol. 24:60–65. doi:10.1016/j.mib.2015.01.002. 796

Uphoff, S. 2018. Real-time dynamics of mutagenesis reveal the chronology of DNA repair and 797 damage tolerance responses in single cells. Proceedings of the National Academy of 798 Sciences. 115:E6516–E6525. doi:10.1073/pnas.1801101115. 799

Vickridge, E., C. Planchenault, C. Cockram, I.G. Junceda, and O. Espéli. 2017. Management of E. 800 coli sister chromatid cohesion in response to genotoxic stress. Nature Communications. 801 8:14618. doi:10.1038/ncomms14618. 802

Wallden, M., D. Fange, E.G. Lundius, Ö. Baltekin, and J. Elf. 2016. The Synchronization of 803 Replication and Division Cycles in Individual E. coli Cells. Cell. 166:729–739. 804 doi:10.1016/j.cell.2016.06.052. 805

Wehrens, M., D. Ershov, R. Rozendaal, N. Walker, D. Schultz, R. Kishony, P.A. Levin, and S.J. Tans. 806 2018. Size Laws and Division Ring Dynamics in Filamentous Escherichia coli cells. Current 807 Biology. 28:972-979.e5. doi:10.1016/j.cub.2018.02.006. 808

Wu, F., P. Swain, L. Kuijpers, X. Zheng, K. Felter, M. Guurink, J. Solari, S. Jun, T.S. Shimizu, D. 809 Chaudhuri, B. Mulder, and C. Dekker. 2019. Cell Boundary Confinement Sets the Size 810 and Position of the E. coli Chromosome. Curr. Biol. 29:2131-2144.e4. 811 doi:10.1016/j.cub.2019.05.015. 812

Yang, D.C., K.M. Blair, and N.R. Salama. 2016. Staying in Shape: the Impact of Cell Shape on 813 Bacterial Survival in Diverse Environments. Microbiology and Molecular Biology Reviews. 814 80:187–203. doi:10.1128/MMBR.00031-15. 815

Youngren, B., H.J. Nielsen, S. Jun, and S. Austin. 2014. The multifork Escherichia coli 816 chromosome is a self-duplicating and self-segregating thermodynamic ring polymer. 817 Genes & Development. 28:71–84. doi:10.1101/gad.231050.113. 818

819


https://doi.org/10.1101/2020.03.16.993485


Figure 1

a.

b. c.

4

3.5

3

2.5

22 2.5 3 3.5 4

daug

hter

1 ce

ll le

ngth

(µm

)

daughter2 cell length (µm)

wild type

0 10 20 30 40 500.0

0.1

0.2

0.3

0.4

0.5

length of �lament at division

rela

tive

posi

tion

of d

ivis

ion

f.e. g.

4

8

10 20 30 40

12

16

0

mitomycin C

S D c

ell l

engt

h (µ

m)

LD cell length (µm)

60 min 60 min

5

10

10 20 30 40

15

20

0

S D c

ell l

engt

h (µ

m)


30 min

0 10 20 30 40 500.0

0.1

0.2

0.3

0.4

0.5

length of �lament at division

rela

tive

posi

tion

of d

ivis

ion

30 min

wt0

2

4

6

8

30 60 90duration of damage (min)

num

ber o

f div

isio

ns (L

D,1

hr)

14

*

14 18 26 30 34 38 4222

*

*

d.

j.i.

80

60

40

20

0

time

betw

een

divi

sion

s (m

in)

control LD SD

0

0.2

0.4

0.6

frac

tion

reco

vere

d SD

LD

30 60 90duration of damage (min)

h.

0 20 40 60 80

10

20

30

40

time (min)

cell

leng

th (μ

m)

LD

SD (4.6μm)


https://doi.org/10.1101/2020.03.16.993485


Figure 2

a. b. c.

15

10

20

5

10 20 30 40 500�lament length at birth (µm)

leng

th (µ

m)

0

mitomycin C

f.e.d.

leng

th (µ

m)

54.543.532.52

2

1

0

3

4

5

cell length at birth (µm)

removedadded

wild type

time between divisions (min) 100 20 30 40 50 60

leng

th (µ

m)

0

5

10

15

20

mitomycin C

time between divisions (min) 100 15 20 25 30 35

leng

th (µ

m)

0

1

2

3

5

4

wild type

10

20

30

40

50

00 5 10 15 20 25 30 35 40

�lament length at birth (µm)

divi

sion

tim

e (m

in)

mitomycin C

divi

sion

tim

e (m

in)

cell length at birth (µm)

5101520253035

02 2.5 3.5 4.53 4

wild type


https://doi.org/10.1101/2020.03.16.993485


*

**

***

14 20 26 32 38 44 50 56

Figure 3

e.

b.a. c.

f. g.

controltime

betw

een

divi

sion

s (m

in)

mitomycin Ccephalexin

wild type ++FtsZ

recovery

100

80

60

40

20

0 10 20 30 40 50

10

20

30

40

50


S D c

ell l

engt

h (µ

m) ΔminCDE

wild type

LD cell length (μm) 0 10 20 30

S D ce

ll le

ngth

(µm

)

10

20

30ΔslmA

wild type

0 10 20 30LD cell length (μm)

S D ce

ll le

ngth

(µm

)

10

20

30 ΔsulA

wild type

d.

0

10

30

50

0 2100

10

20

30

21number of constrictions

(DNA damage)

cell

leng

th (μ

m)

++FtsZ wild type

**

20 60 100 140 180 300sulArecAlexAftsAftsEftsIftsNftsQftsWftsXftsZzapAzapBzipA

recovery time (min)

normalized log fold change

-0.50

1

2

3

4


https://doi.org/10.1101/2020.03.16.993485


Figure 4

a.24 28 32 5236 40 44 48

phase

HupA-mCherry

0

10

15

20

5cell

leng

th (µ

m)

nucleatecontrol anucleaterecovery

minicellscephalexin

recovery wild typeMMCre

lativ

e po

sitio

n of

leas

tH

upA

�uo

resc

ence

0.6

0.4

0.2

0control

c.b. d.

f.e.

terFtsZ ZapA FtsNlocalization segregation

30

10

20

0time

to d

ivis

ion

(min

)

g.

wild type ΔslmA ΔsulA ΔmatP0

20

40

60

80

100anucleate recover�lament

% c

ells

daughter cell

nucleate anucleate

FtsZ localization

0

10

20

30

time

to d

ivis

ion

(min

) 0

10

20

30

cell

leng

th (µ

m)

control SD control SD

wild type ΔmatP


https://doi.org/10.1101/2020.03.16.993485


Documents

Asymmetric chromosome segregation and cell division in DNA … · 4 Suchitha Raghunathan1,2, Afroze Chimthanawala1,3, Sandeep Krishna1,4, Anthony G. 5 Vecchiarelli5 and Anjana Badrinarayanan1*