Salmonella Typhimurium employ spermidine to exert protection against ROS-mediated cytotoxicity and rewires host polyamine metabolism to ameliorate its survival in macrophages

Salmonella Typhimurium employ spermidine to exert protection against ROS-

mediated cytotoxicity and rewires host polyamine metabolism to ameliorate its

survival in macrophages

Abhilash Vijay Nair

, Anmol Singh

, R. S. Rajamani

, Dipshikha Chakravortty

a,c

Department of Microbiology and Cell Biology, Division of Biological Sciences, Indian Institute

of Science, Bengaluru, India

Molecular Biophysiscs Unit, Indian Institute of Science, Bangalore, India

Adjunct Faculty, School of Biology, Indian Institute of Science Education and Research,

Thiruvananthapuram

Running Head: Salmonella mounts an antioxidative response using de-novo and host-acquired

spermidine

#Address correspondence to Dipshikha Chakravortty, [email protected], Tel: 0091 80 2293 2842,

Fax: 0091 80 2360 269

Keywords: Spermidine, Macrophages, Antioxidative response, Glutathionyl-spermidine

synthetase, D,L-α-difluoromethylornithine

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Abstract

Salmonella infection involves a cascade of attacks and defence measures. After breaching the

intestinal epithelial barrier, Salmonella is phagocytosed by the macrophages, inside which, the

bacteria face multiple stresses and, consequently, employ appropriate countermeasures. We show

that, in Salmonella, the polyamine spermidine activates a stress response mechanism by regulating

critical antioxidant genes. Salmonella Typhimurium mutants for spermidine transport and

synthesis cannot mount an antioxidative response, resulting in high intracellular ROS levels. These

mutants are also compromised in their ability to be phagocytosed by macrophages. Furthermore,

it regulates a novel enzyme in Salmonella, Glutathionyl-spermidine synthetase (GspSA), which is

known to prevent the oxidation of proteins in E.coli. Moreover, the spermidine mutants and the

GspSA mutant show significantly reduced survival in the presence of hydrogen peroxide in vitro,

and lesser organ burden in the mouse model of Salmonella infection. Conversely, in macrophages

isolated from gp91phox

-/-

mice, we observed a rescue in the attenuated fold proliferation previously

observed upon infection. Interestingly, Salmonella upregulates polyamine biosynthesis in the host

through its effectors from SPI-1 and SPI-2, which also solves the mystery of the attenuated

proliferation observed in spermidine transport mutants. Thus, inhibition of this pathway in the host

abrogates the proliferation of Salmonella Typhimurium in macrophages. From a therapeutic

perspective, inhibiting host polyamine biosynthesis using an FDA-approved chemopreventive

drug, D,L-α-difluoromethylornithine (DFMO), reduces Salmonella colonization and tissue

damage in the mouse model of infection, while enhancing the survival of infected mice. Therefore,

our work provides a mechanistic insight into the critical role of spermidine in stress resistance of

Salmonella. It also reveals a strategy of the bacteria in modulating host metabolism to promote

their intracellular survival and shows the potential of DFMO to curb Salmonella infection.

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Introduction

A condition with ample nutrients, and desired temperature, pH, oxygen concentration, and

osmolarity is usually considered optimal for the growth of microbes. An imbalance or any

alteration in these parameters impedes the growth and survival of the microbes, considered as

stress conditions. However, these parameters keep fluctuating in nature [1]. Hence, to persist and

survive in the natural environment, with unforeseen stressful or lethal conditions, they must be

adept in sensing, responding and adapting to them [2]. In the case of food-borne pathogens, they

face an array of various stresses in multiple environments in which they dwell, such as from the

natural habitat to commercial settings and inside the host system [3-5]. Salmonella is a food-borne

pathogen that enters the host system through contaminated food and water. In the host system, as

it traverses to the intestine, Salmonella faces multiple stressful conditions such as low pH, nutrient

deprivation, bile salt stress, competition with the resident microbes of the gut and

immunoglobulins, etc . Once it breaches the epithelial barrier, it is taken up by the macrophages at

the lamina propria, through which it disseminates throughout the host system. Macrophages are

phagocytic immune cells where Salmonella encounters a very hostile environment. Entry of the

pathogen into the cell cytoplasm leads to a burst of reactive oxygen species (ROS) and reactive

nitrogen species (RNS) [6, 7]. In macrophages, Salmonella resides in a specialised niche called

the Salmonella-containing vacuole (SCV), which presents multiple other stresses to the bacteria,

such as acidification, nutrient limitation and attack by the antimicrobial peptides. However,

Salmonella employs numerous weapons from its arsenal to counteract the stresses it faces within

the host macrophages.

Polyamines are a group of primordial stress response molecules in prokaryotes and eukaryotes [8].

Multiple research groups have elucidated the link between polyamines and bacterial stress

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

response. In Shigella spp. the polyamine spermidine accumulates during its infection into

macrophages, which increases the expression of katG and helps in bacterial antioxidant response

[9]. Moreover, spermidine localises to the surface of Pseudomonas aeruginosa and bind to the

lipopolysaccharide (LPS) to protect the cells from oxidative damage [10]. In the Gram-positive

bacteria, Streptococcus pyogenes, extracellular spermidine enhance the survival of the bacteria by

upregulating oxidative response genes [11]. A research group has shown that polyamines are vital

in resistance against nitrosative stress in Salmonella Typhimurium. Further, the group showed that

spermidine is required for the systemic infection of Salmonella Typhimurium in mice [12]. Thus,

it is conceivable that Salmonella Typhimurium utilises polyamines, such as spermidine, as a stress

response molecule; however, the mechanism remains elusive.

Here, we show that spermidine biosynthesis and transport mutants of Salmonella Typhimurium

exhibit reduced survival upon infection in RAW264.7 cells. This diminished proliferation is also

observed in mice models of Salmonella infection, which is rescued in gp91phox

-/-

mice. We

demonstrate that spermidine orchestrates the various arms of antioxidative response and aids in

tightly regulating intracellular ROS levels. We further identify a novel antioxidative enzyme, Gsp,

in Salmonella Typhimurium, which is regulated by spermidine. The fascinating question that arises

is why the transporter mutant shows reduced survival. To this, we find that Salmonella

Typhimurium harnesses the host polyamine biosynthesis for its survival. Furthermore, for the first

time, we show that an FDA-approved chemopreventive and anti-African Human Trypanosomiasis

drug that inhibit the polyamine biosynthesis in the host, is able to curb Salmonella infection in

mice models.

Material and methods

Bacterial strains and growth condition

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Salmonella enterica serovars Typhimurium (STM WT) wild type strain ATCC 14028s was used in

all experiments which was a kind gift from Prof. Michael Hensel, Abteilung Mikrobiologie,

Universität Osnabrück, 273 Osnabrück, Germany. The bacterial strain was cultured in Luria broth

(LB-Himedia) with constant shaking (175rpm) at 37°C orbital-shaker. Kanamycin,

Chloramphenicol and Ampicillin antibiotics (Kanamycin-50μg/ml, Cholramphenicol-20μg/ml and

Ampicillin-50μg/ml) were used wherever required. Strains were transformed with pFPV-m-cherry

plasmid for immunofluorescence assays. (Bacterial strain list in Supplementary table S-Table 1)

Bacterial gene knock-out and strain generation

The generation of gene knock-out in bacteria was done using the One-step chromosomal gene

inactivation protocol [13]. Briefly, the Kanamycin resistance gene and the Chloramphenicol

resistance gene amplified products were purified using chloroform-ethanol precipitation. Followed

100

by electroporation into the STM WT cells (expressing PKD-46 plasmid which provides the λ-Red

101

recombinase system) by a single pulse of 2.25 kV separately for the Kan

and Chlm

. The

102

transformant colonies were selected and patched on fresh plates and confirmed for knock-out using

103

PCR with primers designed corresponding to the ~100bp upstream and downstream of the genes

104

(knocked out) for the knock-out strains to observe a difference in PCR product size upon STM

105

ΔpotCD and STM ΔspeED knockout generation. For the generation of the double knock-out strain

106

(STM ΔpotCDΔspeED), the STM ΔpotCD (resistant to Kanamycin) was first transformed with the

107

plasmid pKD46 which provides the λ-Red recombinase system. To this transformed strain, the

108

purified PCR product to knock-out speED was electroporated to generate the STM

109

ΔpotCDΔspeED (resistant to Kanamycin and Chloramphenicol). For the generation of STM Δgsp,

110

Kanamycin resistance gene was amplified from pKD4 plasmid, and a similar protocol was used,

111

followed by selection on Kanamycin containing LB agar plates. For the generation of double gsp

112

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

spermidine mutants, STM Δgsp was electroporated with purified PCR product to knock-out speED

113

and potCD (both Chloramphenicol resistance cassette). (Knockout generation Primer list in

114

Supplementary table S-Table 2)

115

Cell culture and maintenance

116

RAW264.7 cells (murine macrophage cell line) were cultured in DMEM - Dulbecco’s Modified

117

Eagle Medium (Lonza) supplemented with 10% FBS (Gibco) and 1% Penicillin-streptomycin

118

(Sigma- Aldrich) at 37°C humidified chamber (Panasonic) with 5% CO

. For each experiment, the

119

cells were seeded onto the appropriate treated cell culture well plate at a confluency of 80% for

120

intracellular survival assay, and expression studies.

121

Gentamicin protection assay

122

The cells were infected with STM WT, STM ΔpotCD, STM ΔspeED, STM ΔpotCDΔspeED, STM

123

Δgsp and STM ΔkatG at MOI of 10 (for intracellular survival assay) and MOI 25 (for qRT-PCR).

124

After infecting the cell line with STM WT and the mutants, the plate was centrifuged at 700-900

125

rpm for 10 minutes to facilitate the proper adhesion. The plate was then incubated for 25 minutes

126

at 37°C humidified chamber and 5% CO

. Then the media was removed from the wells and washed

127

with 1X PBS. Fresh media containing 100 µg/mL gentamicin was added and again incubated for

128

60 minutes at 37°C and 5% CO

. The media was then removed, cells were washed with 1X PBS

129

twice, and fresh media containing 25µg/mL gentamicin was added. The plate was incubated at

130

37°C and 5% CO

till the appropriate time. For the intracellular survival assay, two time points

131

were considered 2 hours and 16 hours, and for qRT-PCR three time points were considered 2

132

hours, 6 hours and 16 hours. For phagocytosis assay upon opsonisation, all the strains were washed

133

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

with 1XPBS and incubated at 37°C with mouse complement sera for 1 hour before gentamicin

134

protection assay

135

Intracellular survival assay and phagocytosis assay

136

At the appropriate time post-infection, the cells were lysed using 0.1% Triton X followed by

137

addition of more 1X PBS and samples were collected. The collected samples were plated at the

138

required dilutions on LB agar plates and kept at 37°C. 12 hours post incubation the Colony forming

139

units (CFU) were enumerated for each plate.

140

The fold proliferation and invasion were determined as follows

141

Fold Proliferation = (CFU at 16 hours post-infection)/(CFU at 2 hours post-infection)

142

Percentage Phagocytosis = [(CFU at 2 hours post-infection)/(CFU of the Pre-inoculum)]×100

143

RNA isolation and qRT-PCR

144

RNA isolation was performed from infected cells after appropriate hours of infection with STM

145

WT, STM ΔpotCD, STM ΔspeED by RNA isolation was performed using TRIzol (Takara) reagent

146

according to manufactures’ protocol RNA was quantified using Thermo-fischer scientific Nano

147

Drop followed by running on 2% agarose gel for checking the quality. For cDNA synthesis, first

148

DNase I treatment with 3μg of isolated RNA was done at 37℃ for 60 minutes, which was then

149

stopped by heating at 65℃ for 10 minutes. Then RNA (free of DNA) was subjected to Reverse

150

transcription using Random hexamer, 5X RT buffer, RT enzyme, dNTPs and DEPC treated water

151

at 37°C for 15 minutes, followed by heating at 85℃ for 15 seconds. Quantitative real-time PCR

152

was done using SYBR green RT-PCR kit in BioRad qRT-PCR system. A 384 well plate with three

153

replicates for each sample was used. The expression levels of the gene of interest were measured

154

using specific RT primers. Gene expression levels were normalised to 16SrDNA primers of S.

155

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Typhimurium. Gene expression levels of eukaryotic gene of interest were normalised to beta-actin

156

of mouse/human as required. For expression studies in bacteria grown in LB media, the bacterial

157

samples were harvested at 3 hours, 6 hours, 9 hours and 12 hours post subculture in fresh LB media

158

in 1:100 ratio and 1mM H

was added to the broth, to study the gene expression in the presence

159

of oxidative stress. Then similar protocol was used to isolate total RNA using TRIzol (Takara)

160

reagent according to manufactures’ protocol (Expression Primer list in Supplementary table S-

161

Table 2)

162

Primary macrophages isolation and infection

163

Primary macrophages were isolated from C57BL/6 mice (male, 5-6 weeks old). The mice were

164

intraperitoneally injected with 8% Brewer’s Thioglycolate (HiMedia). Five days post injection the

165

primary macrophages were aseptically isolated by injecting ice-cold 1xPBS into the peritoneal

166

cavity the peritoneal lavage was collected. Any residual erythrocytes were lysed using RBC lysis

167

buffer ( Sigma- R7757) , and the isolated cells were maintained in complete RPMI 1640 media for

168

further experiments.

169

Intracellular Reactive oxygen species determination using H

DCFDA staining

170

Overnight cultures were sub-cultured in fresh LB media. Once the cultures reached OD 0.1 then

171

CFU/ml of each strain was incubated with 10µM of 2',7'-dichlorodihydrofluorescein diacetate

172

(H2DCFDA) (Sigma) in 1xPBS (pH 7.2) at 37°C for 30 minutes. The bacterial cells were

173

centrifuged and the cells were resuspended in 1xPBS (pH 7.2) with Hydrogen peroxide of different

174

concentrations (0mM- 10mM) , and incubated 37°C (orbital shaker) for 2 hours. The samples were

175

transferred to a 96 well ELISA plate and fluorescence was determined in Tecan-ELISA plate reader

176

Infinite series 200 ( Ex- 490nm/ Em- 520nm).

177

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Intracellular redox status determination

178

The STM WT, STM ΔpotCD, STM ΔspeED and STM Δgsp were transformed with pQE60-Grx1-

179

roGFP2 (a gift from Prof. Amit Singh, CIDR, IISc). Overnight cultures were sub-cultured in fresh

180

LB media. Once the cultures reached OD 0.1 then 10

CFU/ml of each strain was incubated with

181

Hydrogen peroxide (0mM-5mM) of different concentrations in 1xPBS (pH 7.2) with, and

182

incubated 37°C (orbital shaker) for 2 hours. The samples were centrifuged and resuspended in

183

fresh 1xPBS (pH 7.2). And tubes were analysed for GFP fluorescence at 408nm and 488nm

184

respectively using BD-FACS Verse flow cytometer (total 10,000 events for each sample). For

185

determination of intracellular redox status upon infection into RAW264.7 cells, 10

RAW264.7

186

cells seeded in a 24 well tissue culture plate, were infected with each of the strains harboring the

187

roGFP2 plasmid using gentamicin protection assay. After 16 hours post infection, the macrophages

188

were washed with 1X PBS and scrapped off using cell scraper and analysed for GFP fluorescence

189

at 408nm and 488nm respectively using BD-FACS Verse flow cytometer (total 10,000 events for

190

each sample).

191

In vitro sensitivity assays

192

Overnight cultures were sub-cultured in fresh LB media. Once the cultures reached OD 0.1 then

193

CFU/ml of each strain was incubated with Hydrogen peroxide or sodium nitrite of different

194

concentrations (0mM- 10mM) in 1xPBS(pH 7.2) (1xPBS of pH 5.4 was used for nitrite sensitivity

195

assay) with, and incubated 37°C (orbital shaker) for 2 hours. The samples were plated on SS agar

196

to enumerate the CFU and the percentage survival was determined as:

197

Percentage Survival : [ [CFU/ml for treated with H

/ NaNO

]/ [CFU/ml for untreated] ]×100

198

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Resasurin assay was used to determine the viable cells was performed in 96 well plate (triplicate

199

for each sample and concentration. Briefly, after incubation for 2 hours as previously performed,

200

Resasurin (0.2mg/ml in 1X PBS) was added (1:10 ratio) to each well of 96 well plate. The plate

201

was incubated at 37°C shaker incubator for 2 hours. The fluorescence was measure using Tecan

202

plate reader infinite series 200, with Ex-520nm and Em- 590nm. The values obtained as Relative

203

fluorescence units (RFU) and the percentage survival was determined as

204

Percentage survival = [RFU of sample with added hydrogen peroxide]/ [RFU of sample without

205

of hydrogen peroxide]

206

Immunoblotting

207

The bacterial strains were grown in LB media with added 1mM hydrogen peroxide until the log

208

phase of growth. The cells were centrifuged to remove the media, and the cells were resuspended

209

in the lysis buffer (Sodium chloride, Tris, EDTA, 10% protease inhibitor cocktail) after washing

210

with 1XPBS. The cells were lysed using sonication and centrifuged at 4°C to collect the cell lysate,

211

followed by estimation of total protein using the Bradford protein estimation method. 50µg of

212

protein was loaded onto a Polyacrylamide Gel Electrophoresis (PAGE) without β-mercaptoethanol

213

(prevent di-sulphde bond breakage as glutathionyl-spermidine modifies Cysteine residues through

214

a disulphide bond), then transferred onto 0.45μm PVDF membrane (GE Healthcare). 5% skimmed

215

milk (Hi-Media) in TTBS was used to block for 1hour at room temperature and then probed with

216

Anti-Spermidine primary (Novus Biologicals) and the secondary HRP-conjugated antibodies.

217

ECL (Biorad) was used for developing the blot, and images were captured using Chemi-

218

Doc(Biorad). All densitometric analysis was performed using the Image J. The normalisation was

219

done with respect to Ponceau S stained blot.

220

Transfection for knockdown

221

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

RAW 264.7 cells were seeded at a confluency of 50-60% 12 hours prior to transfecting using either

222

PEI (1:2 -DNA: PEI). We used two different constructs for knock-down of Odc1, A7 and F8 and

223

both as a mixed construct (A7:A8 at 1:1 ratio) and similarly for Srm , E9 and F1 and both as a

224

mixed construct (E9:F1 at 1:1 ratio) from the Sigma Mission shRNA library. 400ng of plasmid

225

DNA/well (ratio 260/280 ~1.8- 1.9) was used for transfection in a 24well plate. Cells were then

226

incubated for 8 hours at 37℃ in a humidified incubator with 5% CO2; after that, the media

227

containing transfecting DNA and reagents were removed, and cells were further incubated for 48

228

hours in complete media DMEM +10% FBS. Cells were harvested for further analysis or infected

229

with the required MOI using the gentamicin protection assay. (shRNA sequence list in

230

Supplementary table S-Table 3)

231

Immunofluorescence

232

After the appropriate incubation time after gentamicin protection assay, the media was removed,

233

and the cells were washed with 1X PBS and fixed with 3.5% Paraformaldehyde for 10 minutes.

234

The cells were then washed with 1X PBS, followed by incubation with the required primary

235

antibodies (anti-mouseLAMP1 and anti-Spermidine) in a buffer containing 0.01% saponin and 2%

236

BSA, and incubated at room temperature for 45-60 minutes. After washing with 1X PBS, the

237

secondary antibody tagged to a fluorochrome was added and incubated (anti-rat-Alexafluor488 for

238

LAMP1, anti-rabbit-Alexafluor647 for spermidine). The coverslips were then washed with PBS

239

and mounted on a clean glass slide using mounting media containing an anti-fade reagent and

240

observed under the confocal microscope (Zeiss 710 microscope, at 63X oil immersion, 2x319 3x

241

zoom, and 100X zoom for studying only bacterial samples, Zeiss 880 microscope, at 63X oil

242

immersion, 2x319 3x zoom).

243

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

The histopathological sections were deparaffinized and then incubated with the required primary

244

antibody (anti- Salmonella LPS) in a buffer containing 0.01% saponin and 2% BSA and incubated

245

at room temperature for 45-60 minutes. The primary antibody was then removed by washing with

246

1X PBS and then incubated with the appropriate secondary antibody tagged to a fluorochrome.

247

The sections were then washed with PBS and cover with coverslip after using mounting media

248

containing an anti-fade reagent. The coverslips were sealed with clear nail polish and observed

249

under the confocal microscope. For studying histopathology samples Zeiss 880 microscope 40X

250

oil immersion 2x319 3x zoom was used.

251

Intracellular RNS determination

252

For determination of intracellular RNS, a cell permeable nitric-oxide probe, 4, 5-

253

diaminofluorescein diacetate (DAF

A) (Sigma) was used. The protocol has been followed as Roy

254

Chowdhury A et. al. [14]. Briefly, After 16 hours of infection of RAW264.7 cells (transiently

255

knock down of Odc1) at MOI 10 with STM WT, the cells were incubated with fresh DMEM

256

containing 5µM of DAF

DA. The cells were incubated at 37℃ in a humidified incubator with 5%

257

for 30 minutes. The media containing dye was removed and the cells were washed with

258

1XPBS and the cells were acquired immediately for flow-cytometry (BD FACS Verse) (Ex-

259

491nm/Em- 513nm).

260

Intracellular Glutathione determination

261

The intracellular reduced Glutathione (GSH) concentration was determined by modification of a

262

pubmished protocol [15]. Briefly, a standard curve with known concentration of GSH (Sigma) was

263

prepared. Reaction mixture for each contained 600µL of phosphate buffer (0.1M, pH7), 40µL of

264

0.4% w/v 5,5-dithiobis(2-nitrobenzoic acid) (DTNB, from Sigma), 100 µL of the standard

265

solutions of GSH (0mM-1mM range) and autoclaved MilliQ water to make up the volume to

266

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

1000µL.The mixture was incubated at room temperature for 5 minutes and absorbance was

267

measured at 412nm using tecan plate reader. The bacterial strains were subcultured in fresh LB

268

media and grown till OD 0.1 (exponential phase) and washed with buffer (Tris, Sodium chloride

269

and EDTA) and lysed using sonication. The supernatant after sonication was used as the sample

270

for GSH detection. As previously described for standard curve. From the standard curve the

271

intracellular concentrations were interpolated.

272

In silico analysis

273

The in silico protein structure determination was performed using the SWISS-MODEL software

274

(https://swissmodel.expasy.org/), where we supplied the protein sequence of GspSA of

275

Salmonella Typhimurium from UniProt. We analysed the model with highest sequence identity

276

and maximum coverage with the Gsp from E.coli. The structure depicts a Homo-dimer with a

277

GMQE of 0.93 and QMEANisCo Global of 0.88 ± 0.05.

278

In vivo animal experiment

279

5-6weeks old C57BL/6 mice were infected by orally gavaging 10

CFU of STM WT, STM

280

ΔpotCD, STM ΔspeED, STM ΔpotCDΔspeED, STM Δgsp and STM ΔkatG. To study the

281

colonisation in organs, the intestine (Peyer’s patches), MLN, spleen and liver were isolated

282

aseptically, 5 days post-infection, and the CFU was enumerated on differential and selective SS

283

agar by serial dilution. For intraperitoneal infection 5-6weeks old C57BL/6 mice were infected

284

by intraperitoneally injecting 10

CFU of STM WT, STM ΔpotCD, STM ΔspeED, STM

285

ΔpotCDΔspeED, STM Δgsp and STM ΔkatG. To study the colonisation in organs, spleen and liver

286

were isolated aseptically 3 days post-infection. Blood was isolated by heart puncture 3 days post-

287

infection. The CFU was enumerated on differential and selective SS agar by serial dilution. Organs

288

were stored in 3.5%PFA before histopathogical sample preparation.

289

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

For inhibitor treatment, 5-6weeks old C57BL/6 mice were infected by orally gavaging 10

CFU

290

of STM WT. The inhibitor DFMO (Sigma D193) was intraperitoneally injected every alternate

291

day from day 1 at two doses 2mg/kg and 1mg/kg of body weight of mice. To study the colonisation

292

in organs, the intestine (Peyer’s patches), MLN, spleen and liver were isolated aseptically, 5 days

293

post-infection, and the CFU was enumerated on differential and selective SS agar by serial dilution.

294

For survival assay of mice 5-6weeks old C57BL/6 mice were infected by orally gavaging 10

CFU

295

of STM WT. The inhibitor DFMO was intraperitoneally injected every alternate day from day 1 at

296

two doses 2mg/kg and 1mg/kg of body weight of mice. The survival was monitored till 10 days.

297

Likewise, organs were stored in 3.5% PFA before histopathogical sample preparation.

298

Mass Spectrometry to determine intracellular GS-sp levels

299

The bacterial strains were grown in LB media with added 1mM hydrogen peroxide until the log

300

phase of growth. The cells were centrifuged to remove the media, and the cells were resuspended

301

in the lysis buffer (Sodium chloride, Tris, EDTA, 10% protease inhibitor cocktail) after washing

302

with 1XPBS. The cells were lysed using sonication and centrifuged at 4°C to collect the cell lysate.

303

Protein was precipitated using ice-cold acetone (Sigma, MS grade), 4 times the volume of the cell

304

lysate and by incubating at -20°C overnight. The precipitated proteins were removed by

305

centrifugation and the supernatatnt was used for analysis. Samples were analysed by ESI MS Q-

306

TOF, impact HD ( Bruker Daltonics Germany) connected to Agilent HPLC 1260. Samples were

307

passed through Agilent C18, 4.6X150mm column. Mobile phase used was water and Acetonitrile

308

with 0.1% formic acid. Linear gradient was used with flow rate 0.2ml/min. Data was analysed

309

using Bruker Daltonics software Data analysis 4.1. Mass of GS-sp (Glutathionyl spermidine) is

310

434g/mol, and (GS-sp)

(oxidized form, Di-glutathionyl spermidine) is 866g/mol.

311

312

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Statistical Analysis

313

Statistical analyses were performed with GraphPad Prism software. The Student’s t-test

314

(parametric, two-tailed, unpaired) was performed as indicated. For animal experiments Non-

315

parametric Mann-whitney (two-tailed) test was performed. The results are expressed as mean ±

316

SD or mean ± SEM. Group sizes, experiment number, and p values for each experiment are

317

described in figure legends.

318

Results

319

Loss of spermidine transporter and biosynthesis genes in Salmonella Typhimurium

320

compromises its ability to be phagocytosed by the macrophages

321

The pathoadaptation in Salmonella involves multiple players, which counteracts the stressful

322

condition it encounters in the host macrophages. Polyamines being a group of well-studied stress

323

response molecules, we were interested in determining the expression of the spermidine transporter

324

and biosynthesis genes in Salmonella Typhimurium. Our previous study shows that Salmonella

325

upregulates the spermidine transporter genes (potA, potB, potC and potD) and the biosynthesis

326

genes (speE and speD) during the log phase of growth in vitro [16]. Here we assessed the mRNA

327

levels of potA, potB, potC and potD in STM WT upon infection into the RAW264.7 macrophage

328

cell line. We noted that all the genes showed a 1.5-2 fold upregulation post 6 hours of infection

329

into macrophages till 16 hours (Fig 1 A). Further, our results showed that the spermidine

330

biosynthesis enzymes speE and speD were upregulated 1.5 to 2 folds post 6 hours to 16 hours post-

331

infection into macrophages (Fig 1 B). These results indicate that Salmonella Typhimurium

332

enhances its intracellular spermidine biosynthesis and imports from the extracellular milieu. It

333

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

might be a strategy of the pathogen to increase the intracellular pool of stress response molecules

334

like the polyamine spermidine, as it encounters the hostile environment of host macrophages.

335

We then determined how the spermidine transporter and biosynthesis mutants (STM ∆potCD, STM

336

∆speED and STM ∆potCD∆speED) behave upon infection into RAW264.7 macrophages. We

337

infected STM WT, STM ∆potCD, STM ∆speED and STM ∆potCD∆speED into RAW264.7 cells

338

and observed that STM ∆potCD, STM ∆speED and STM ∆potCD∆speED showed reduced

339

phagocytosis by the macrophages compared to the wild type (Fig 1 C). Interestingly, STM

340

∆potCD, STM ∆speED and STM ∆potCD∆speED showed a compromised ability to proliferate

341

intracellularly in the RAW264.7 cells further validating a previous study [12](Fig S1 A). To further

342

validate our results, we assessed the behaviour of the spermidine transporter and biosynthesis gene

343

mutants in primary macrophages isolated from the peritoneal lavage of C57BL/6 mice. Likewise,

344

the mutants showed attenuated proliferation and uptake by phagocytosis into the peritoneal

345

macrophages (Fig 1 D and S1 B). Collectively these results suggest that spermidine is a critical

346

molecule in Salmonella Typhimurium to infect and survive in macrophages. To further investigate

347

the reason behind the reduced ability to be taken up by macrophages upon loss of spermidine

348

biosynthesis and transport, we treated STM WT, STM ∆potCD, STM ∆speED and STM

349

∆potCD∆speED with mouse complement-sera. Mouse complement-sera acts as an opsonin and

350

thus potentiates the interaction of the bacteria with the macrophages. Upon pre-treatment with

351

mouse complement-sera, we noted a rescue in the reduced uptake of the mutants by peritoneal

352

macrophages isolated from C57BL/6 mice (Fig 1 E). A study on Salmonella Typhimurium

353

revealed that the aflagellate and non-motile Salmonella collide less frequently with macrophages

354

and gets merest time to maintain contact with the macrophages, thereby showing decreased

355

phagocytosis [17]. Our group previously showed that loss of spermidine production and import in

356

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Salmonella Typhimurium results in the loss of flagella formation on the bacterial cell surface [16].

357

Thus, we determined the percentage phagocytosis for a flagellin-deficient strain of Salmonella

358

Typhimurium, STM ∆fliC and observed that STM ∆potCD, STM ∆speED, STM ∆potCD∆speE

359

and STM ∆fliC exhibited a significantly decreased ability to be taken up by the RAW264.7

360

macrophages (Fig 1 F). Furthermore, we observed a rescue of the attenuated percentage

361

phagocytosis and fold proliferation of only STM ∆speED upon supplementation of spermidine

362

(100µM) during the growth of bacteria prior to infection (Fig 1 G and S1C). We generated single

363

gene mutants for abrogating the spermidine transport (STM ∆potA) and spermidine biosynthesis

364

(STM ∆speE) function and further complemented the genes through a vector (STM ∆potA:potA

365

and STM ∆speE:speE ), we observed a recovery of the fold proliferation and percentage

366

phagocytosis nearly to STM WT in the complemented strains (Fig S1 D-E) . Hence, the plausible

367

explanation is that the reduced ability to form flagella in the spermidine mutants causes less

368

frequent interaction with the macrophages and provides minimal contact time for infection, leading

369

to reduced phagocytosis by the macrophages.

370

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

371

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 1: Loss of spermidine transporter and biosynthesis genes in Salmonella Typhimurium

372

compromises its ability to be phagocytosed by the macrophages

373

A.The mRNA expression of spermidine transporter genes potA, potB, potC and potD in STM WT

374

upon infection into RAW264.7 cells, B. The mRNA expression of spermidine biosynthesis genes

375

speE and speD in STM WT upon infection into RAW264.7 cells, C. The percentage phagocytosis

376

of the spermidine mutants in RAW264.7 cells, D. The percentage phagocytosis in primary

377

macrophages isolated from peritoneal lavage of C57BL/6 mice, E. The percentage phagocytosis

378

upon pre treatment with mouse-complement sera which act as an opsonin, F. The percentage

379

phagocytosis in RAW264.7 cells with flagellin mutant (STM ∆fliC), G. The percentage

380

phagocytosis in RAW264.7 cells with the spermidine mutants grown in media supplemented with

381

100µM spermidine (SPD). Student’s t-test was used to analyze the data; p values ****<0.0001,

382

***<0.001, **<0.01, *<0.05. Two-way Anova was used to analyze the grouped data; p values

383

****<0.0001, ***<0.001, **<0.01, *<0.05.

384

385

Spermidine provides stress resistance in Salmonella Typhimurium by regulation of the

386

expression of numerous antioxidative enzymes

387

The loss of spermidine transport and biosynthesis function in Salmonella Typhimurium renders it

388

incapable of proliferation and survival in macrophages. In the host macrophages, the bacteria

389

encounter numerous threats, of which the foremost is the rapid oxidative burst mediated by the

390

NOX2. The reactive oxygen species superoxide radical can easily diffuse through the bacterial

391

membrane and pose a major threat to the pathogen. ROS acts on multiple molecules such as nucleic

392

acids, proteins and lipids, thus damaging the cell membranes, DNA and proteins within the bacteria

393

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

[18]. Spermidine has been linked to stress response against oxidative stress and protects bacteria

394

in E. coli and Pseudomonas aeruginosa. Thus, we were intrigued to understand the role of

395

spermidine in antioxidative response in Salmonella Typhimurium. We examined the survival of

396

STM WT, STM ∆potCD, STM ∆speED, STM ∆potCD∆speE upon exposure to the oxidative agent

397

hydrogen peroxide in vitro. We noticed that at high concentrations of hydrogen peroxide, 5mM

398

and 10mM, the STM ∆potCD, STM ∆speED and STM ∆potCD∆speE showed significantly lesser

399

survival than STM WT(Fig 2 A). Moreover, the complemented strains for spermidine transport

400

and synthesis mutants showed a rescue in attenuated survival in the presence of high

401

concentrations of hydrogen peroxide in vitro (Fig S2 A). We further determined the expression of

402

potA, potB, potC, potD, speE and speD in STM WT upon exposure to 1mM hydrogen peroxide.

403

There was a 2-3 fold upregulation in the mRNA expression of the transporter genes over the

404

untreated during the early log phase of growth (6 hours) and the late log phase of growth (12 hours)

405

in vitro (Fig S2 B-E). Similarly, the biosynthesis genes speE and speD were 4-6 fold upregulated

406

in their corresponding mRNA expressions during their early log phase of growth (6 hours) and the

407

late log phase of growth (12 hours) in vitro (Fig S2 F and G). Our results show that Salmonella

408

Typhimurium upregulates spermidine transport and biosynthesis upon oxidative stress suggesting

409

that spermidine mounts a protective function in such a stressful condition to aid bacterial survival.

410

Bacteria sense the environmental changes and cues to respond and adapt to the altered

411

environment. They use the two-component systems, transcriptional activators and repressors to

412

alter gene expression in response to a stimulus [19]. Polyamines in E. coli regulate multiple genes

413

at the transcription and translation together, referred to as the “Polyamine modulon”. These involve

414

the numerous mRNAs, tRNAs, sigma factors, translational factors and two-component systems

415

during the bacterial growth as well as in stress conditions [20-22]. Salmonella harbors multiple

416

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

antioxidative enzymes to detoxify the ROS intracellularly [23]. Our study so far show that STM

417

∆potCD, STM ∆speED and STM ∆potCD∆speE is attenuated in survival under in vitro oxidative

418

stress than STM WT. To gain mechanistic insight into the attenuated survival of Salmonella

419

Typhimurium we determined the mRNA expression of the critical transcription factor rpoS, which

420

activates the expression of the catalase enzymes (katG and katE) in order to detoxify hydrogen

421

peroxide to water in the bacteria enzymatically [24]. In both STM ∆potCD and STM ∆speED the

422

mRNA expression of rpoS is significantly down regulated from 6 hours post infection in

423

RAW264.7 cells (Fig 2 B and C, S3 A). Further, its downstream target katG was also down

424

regulated in STM ∆potCD and STM ∆speED from 6 hours post infection into RAW264.7 cells

425

(Fig 2 D and E, S3 B). We further assessed the mRNA expression of the transcription factor soxR,

426

which regulates the expression of superoxide dismutases (sodA and sodB) [25]. Upon infection in

427

RAW264.7 cells, expression of soxR was significantly downregulated in STM ∆potCD and STM

428

∆speED with respect to STM WT (Fig 2 F and G, S3 C). Superoxide dismutases act on superoxide

429

radicals, the potent ROS encountered in macrophages, converting to hydrogen peroxide. The

430

mRNA expression of both sodA and sodB were likewise downregulated in STM ∆potCD and STM

431

∆speED upon infection into RAW264.7 macrophages (Fig 2 H-K, S3 D and E). A major

432

antioxidant in most living organisms is glutathione (GSH), which directly acts as a quencher of

433

ROS [26]. GSH is synthesized by Glutathione synthase (GshA) which in turn is regulated by EmrR

434

transcription factor. We observed that the mRNA expression of emrR is downregulated in STM

435

∆potCD and STM ∆speED upon infection into RAW264.7 macrophages (Fig 2 L and M, S3 F).

436

Similarly, gshA transcript expression is downregulated as well (Fig 2 N and O, S3 G). To further

437

validate the down regulation of the glutathione synthesis arm in spermidine transport and

438

biosynthesis mutants, we determined the intracellular GSH levels and noted the in STM ∆potCD,

439

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

STM ∆speED and STM ∆potCD∆speED the levels were significantly less (Fig S3H and I). Taken

440

together these results indicate that spermidine regulates the transcription of multiple transcription

441

factors involved in oxidative stress response in Salmonella Typhimurium. Importantly, we found

442

a mechanism of oxidative stress resistance in Salmonella Typhimurium regulated by the

443

spermidine, potentiating the survival of the bacteria.

444

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

445

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 2: Spermidine provides stress resistance in Salmonella Typhimurium by regulation of

446

the expression of numerous antioxidative enzymes

447

A.The in vitro hydrogen peroxide sensitivity assay with the spermidine transport and biosunthesis

448

mutants, B. The mRNA expression of stress responsive transcription factor rpoS in STM ∆potCD

449

upon infection into RAW264.7 cells, C. The mRNA expression of stress responsive transcription

450

factor rpoS in STM ∆speED upon infection into RAW264.7 cells, D. The mRNA expression of

451

katG in STM ∆potCD upon infection into RAW264.7 cells, E. The mRNA expression of katG in

452

STM ∆speED upon infection into RAW264.7 cells, F. The mRNA expression of oxidative stress

453

responsive transcription factor soxR in STM ∆potCD upon infection into RAW264.7 cells, G. The

454

mRNA expression of oxidative stress responsive transcription factor soxR in STM ∆speED upon

455

infection into RAW264.7 cells, H. The mRNA expression of sodA in STM ∆potCD upon infection

456

into RAW264.7 cells, I. The mRNA expression of sodA in STM ∆speED upon infection into

457

RAW264.7 cells, J. The mRNA expression of sodB in STM ∆potCD upon infection into

458

RAW264.7 cells, K. The mRNA expression of sodB in STM ∆speED upon infection into

459

RAW264.7 cells, L. The mRNA expression of glutathione synthetase specific transcription factor

460

emrR in STM ∆potCD upon infection into RAW264.7 cells, M. The mRNA expression of

461

glutathione synthetase specific transcription factor emrR in STM ∆speED upon infection into

462

RAW264.7 cells, N. The mRNA expression of gshA in STM ∆potCD upon infection into

463

RAW264.7 cells, O. The mRNA expression of gshA in STM ∆speED upon infection into

464

RAW264.7 cells.

465

466

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Spermidine controls a novel enzyme Glutathionyl-spermidine synthetase in Salmonella

467

Typhimurium, and together mount an intracellular antioxidative response

468

The spermidine synthesized from putrescine has two fates. It is either acetylated by the enzyme

469

SpeG or it is covalently conjugated to GSH to form Glutathionyl spermidine (GS-sp) catalysed by

470

the enzyme Glutathionyl-spermidine synthetase (Gss). Tabor (1974) first discovered the existence

471

of this enzyme in E. coli [27]. In E. coli GS-sp is generated at a higher level in the stationary phase

472

and very less in the late exponential phase. It also interacts and modifies the thiol containing

473

proteins under high H

containing media leading to the formation of Gsp-thiolated proteins (PS-

474

Gsp). Certain in vitro experiments with dehydro-ascorbate suggested that GS-sp might have higher

475

antioxidant properties than GSH [28] and may be more effective in protecting against DNA

476

damage by free radicals [29]. However, in E. coli Gss could not be linked to its pathogenicity [30].

477

Among Enterobacteriaceae, Salmonella was found to possess this unique enzyme. Our in silico

478

analysis suggested that the enzyme in Salmonella Typhimurium (GspSA, encoded by gsp) has 90%

479

identity with E. coli gss and the SWISS MODEL predicts it to be a homo-dimeric protein (Fig S4

480

A-C). Also, the spermidine synthesised by SpeE in Salmonella Typhimurium is directly fed into

481

the pathway to synthesise GS-sp. We thus, investigated biological role of this novel enzyme in

482

Salmonella Typhimurium. We noted that the mRNA expression of gsp is significantly upregulated

483

at the late-log phase of growth of STM WT in LB media in the presence of H

(Fig S4 D). Upon

484

infection into RAW264.7 cells, STM WT upregulates the mRNA expression of gsp at 6 hours and

485

16 hours post-infection into RAW264.7 cells (Fig 3 A). Further, the Salmonella Typhimurium

486

mutant of gsp (STM ∆gsp) showed attenuated proliferation in RAW264.7 cells similar to STM

487

∆katG, which has reduced capability to detoxify ROS (Fig 3 B). Thus, our results suggest that gsp

488

is important in Salmonella Typhimurium to survive and cope with the oxidative stress and hostile

489

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

environment of macrophages. Interestingly, the mRNA expression of gsp was found to be

490

downregulated in both STM ∆potCD and STM ∆speED at 16 hours post-infection into RAW264.7

491

macrophages (Fig 3 C and D). These findings show that spermidine maintains its intracellular

492

levels by regulating the flux of spermidine into other pathways, such as the GspSA pathway. Also,

493

it further potentiates the ability of Salmonella Typhimurium to mount an antioxidative response by

494

regulating gsp expression.

495

GspSA enzyme in Trypanosomes was named Trypanothione synthetase, while the conjugate is

496

called Trypanothione (TSH). In trypanosomes TSH is essential, and these organisms entirely rely

497

on TSH and possess no GSH reductase. Also, they have evolved to use TSH reductase instead of

498

GSH reductase, Glutaredoxins or Thioredoxins [31]. In E. coli GS-sp forms bonds with the

499

Cysteine thiol groups of numerous proteins and protect them from oxidation under oxidative stress.

500

Cysteine thiol groups are highly prone to attack by ROS and get oxidized to sulphinic and

501

sulphonic acids. In Salmonella Typhimurium we observe that loss of gsp results in attenuated

502

proliferation and survival in macrophages. Thus, we investigated the survival of STM ∆gsp upon

503

exposure to hydrogen peroxide in vitro. STM ∆gsp exhibited reduced survival in the presence of

504

high 5mM and 10mM concentrations of H

(Fig S4 E). Likewise, upon exposure to agents of

505

oxidative stress and nitrosative stress H

and NaNO

together, we observe that STM ∆gsp

506

exhibited reduced survival at higher concentrations such as 5mM and 10mM (Fig 4 E). Thus, gsp

507

is critical in Salmonella Typhimurium to shield the bacteria from the action of ROS and RNS. As

508

we observed that spermidine transporter and biosynthesis mutants and gsp mutant of Salmonella

509

Typhimurium are compromised in their survival under oxidative stress and in macrophages thus,

510

we were interested to assess the intracellular ROS detoxification abilities of the strains. We

511

determined the intracellular ROS in STM WT, STM ∆potCD, STM ∆speED, STM ∆potCD∆speE,

512

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

STM ∆gsp and STM ∆katG. Interestingly, STM ∆potCD, STM ∆speED, STM ∆potCD∆speE,

513

STM ∆gsp and STM ∆katG showed significantly higher intracellular ROS levels when they were

514

exposed to 5mM and 10mM H

(Fig 3 F). Further the complemented strains for spermidine

515

transport and synthesis mutants showed a reduced intracellular ROS in the presence of high

516

concentrations of hydrogen peroxide in vitro (Fig S4 F). These results suggest that lower

517

intracellular levels of spermidine correlate to higher intracellular ROS levels in Salmonella

518

Typhimurium. To validate our observed results, we utilised a genetically engineered tool to sense

519

the redox status of the bacterial cytosol, roGFP2 is a genetically modified form of GFP, and we

520

use pQE60-Grx1-roGFP2 plasmid. The glutaredoxin (Grx1) fused to the roGFP2, reversibly

521

transfers electrons between the cytosolic pool of GSH/GSSG and the thiol group of roGFP2, and

522

the ratio of fluorescence ratio at 408nm and 488nm determine the redox status of bacterial

523

cytoplasm [14]. We observed that STM ∆potCD, STM ∆speED harbouring the pQE60-Grx1-

524

roGFP2 showed higher ratio of 405nm/488nm compared to STM WT in the presence of 5mM

525

in vitro and also upon infection into RAW 264.7 macrophages (Fig 3 G and H). However,

526

STM ∆gsp did not show a significantly higher ratio of 405nm/488nm. Moreover upon

527

supplementation of the growth media with 100 µM spermidine, only in STM ∆speED we observed

528

a lower intracellular ROS and lesser 405nm/488nm in higher concentration of H

(Fig S4 G

529

and H). Thus, our results indicate that spermidine is critical in mounting an antioxidative response

530

to detoxify the intracellular ROS, by regulating multiple antioxidant genes in Salmonella

531

Typhimurium. To validate our observed results we determined the intracellular levels of

532

glutathionyl-spermidine in STM WT, STM ∆potCD, STM ∆speED and STM ∆gsp using mass

533

spectrometry. Our study qualitatively shows that the synthesis of GS-sp and (GS-sp)

(oxidized

534

form, di-glutathionylspermidine) in STM WT upon exposure to 1mM hydrogen peroxide, which

535

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

is absent in the spermidine mutants and STM ∆gsp (Fig S5 A and B). We further determined the

536

presence of PS-Gsp and show that STM WT shows modifications of proteins by GS-sp (detected

537

by anti-spermidine antibody), which is reduced upon treatment with beta-mercaptoethanol. The

538

spermidine mutants show very less modification, while it is almost negligible in STM ∆gsp (Fig

539

S5 C-E).

540

We observe that STM ∆potCD, STM ∆speED show a higher intracellular ROS than STM ∆gsp.

541

Thus to understand does STM ∆gsp phenocopy STM ∆potCD, STM ∆speED, we generated double

542

mutants STM ∆gsp∆potCD and STM ∆gsp∆speED. We find that, the double mutants show a

543

similar reduced fold proliferation and percentage phagocytosis in RAW 264.7 cells (Fig S6 A-B).

544

However, they show an enhanced intracellular ROS upon exposure to hydrogen peroxide (Fig S6

545

C). Upon co-infection of the double mutants with STM ∆gsp in C57BL/6 mice, we see that STM

546

∆gsp out competes the double mutants in colonizing the liver (Fig S6 D-E). To further dissect the

547

role of spermidine in protection of Salmonella Typhimurium from oxidative stress, we infected

548

STM WT, STM ∆potCD, STM ∆speED, STM ∆potCD∆speE, STM ∆gsp and STM ∆katG in

549

primary macrophages isolated from the peritoneal lavage of gp91phox -/- mice. Gp91Phox is the

550

major subunit of the NOX2 complex, that aids in the catalysis of oxygen to superoxide radical.

551

Interestingly we observe a rescue in the attenuated fold proliferation of STM ∆potCD, STM

552

∆speED, STM ∆potCD∆speE, STM ∆gsp and STM ∆katG in peritoneal macrophages isolated

553

from gp91phox -/- mice (Fig 3 I). Our results demonstrate the vital role of spermidine in oxidative

554

stress resistance in Salmonella Typhimurium.

555

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

556

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 3: Spermidine controls a novel enzyme Glutathionyl-spermidine synthetase in

557

Salmonella Typhimurium, and together mount an intracellular antioxidative response

558

A.The mRNA expression of gsp in STM WT upon infection into RAW264.7 cells, B The fold

559

proliferation of STM WT, STM ∆gsp and STM ∆gsp:gsp in RAW264.7 cells, C. The mRNA

560

expression of gsp in STM ∆potCD upon infection into RAW264.7 cells, D. The mRNA expression

561

of gsp in STM ∆speED upon infection into RAW264.7 cells, E The in vitro Hydrogen peroxide

562

and nitic oxide sensitivity assay, F. Intracellular reactive oxygen species determination using the

563

cell permeable H

DCFDA dye, G. The intracellular degree of oxidation in spermidine and gsp

564

mutants using pQE60-grx1-roGFP2 construct in vitro, H. The intracellular degree of oxidation

565

using pQE60-grx1-roGFP2 construct in spermidine and gsp mutants upon infection into RAW

566

264.7 cells , I. The fold proliferation in primary macrophages isolated from wild type C57Bl/6

567

mice and gp91phox-/- mice, Student’s t-test was used to analyze the data; p values ****<0.0001,

568

***<0.001, **<0.01, *<0.05. Two-way Anova was used to analyze the grouped data; p values

569

****<0.0001, ***<0.001, **<0.01, *<0.05.

570

571

Spermidine is critical for Salmonella Typhimurium to colonise the primary and secondary

572

sites of infection in mice

573

Salmonella infects the host and breaching the epithelial cells at the Peyer’s patches in the distal

574

ileum, it disseminates to the secondary sites of infection namely the Mesentric Lymph node

575

(MLN), spleen and liver. From the basolateral surface of the epithelial cells at the lamina propria,

576

Salmonella is taken by the macrophages and polymorphonuclear cells (PMN). We observed that

577

the spermidine transporter and biosynthesis mutants show attenuated survival in macrophages and

578

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

under oxidative stress in vitro. We were intrigued to study the behaviour of the mutants during in

579

vivo colonisation in the mouse model of Salmonella infection. We infected C57BL/6 mice by orally

580

gavaging with STM WT, STM ∆potCD, STM ∆speED, STM ∆potCD∆speE, STM ∆gsp and STM

581

∆katG at a CFU of 10

per mice. We noted that STM ∆potCD, STM ∆speED, STM ∆potCD∆speE,

582

STM ∆gsp and STM ∆katG showed a significantly lower organ burden in Peyer’s patches, MLN,

583

spleen and liver upon oral gavage (Fig 4 A-E). Our previous study showed that the spermidine

584

transporter and biosynthesis mutants exhibit a poor invasiveness into IECs, which can explain our

585

in vivo colonisation upon oral gavage [16]. Oral gavage mimics the physiological route of

586

Salmonella infection into its host. Hence, it requires to be able to breach the intestinal barrier

587

successfully. Incapability to invade the IECs in STM ∆potCD, STM ∆speED, STM ∆potCD∆speE

588

and STM ∆gsp explains the diminished colonisation in the organs. To dissect the role of spermidine

589

in in vivo colonisation we infected C57Bl/6 mice intraperitoneally, by bypassing the entry by

590

breaching epithelial barrier. We observed that upon infecting intraperitoneally, STM ∆potCD, STM

591

∆speED, STM ∆potCD∆speE, STM ∆gsp and STM ∆katG exhibited reduced colonisation in

592

spleen and liver and less dissemination in blood compared to STM WT (Fig 4 F-I). Also, the

593

histopathological sections show significantly less liver tissue damage with STM ∆potCD, STM

594

∆speED, STM ∆potCD∆speE, STM ∆gsp and STM ∆katG , which is validated by disease scoring

595

of the same(Fig 4 J and K). Our results thus show that spermidine aids in the in vivo pathogenesis

596

and virulence of Salmonella Typhimurium.

597

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

598

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 4: Spermidine is critical for Salmonella Typhimurium to colonise the primary and

599

secondary sites of infection in mice

600

A.The experimental protocol for organ burden in C57BL/6 mice by orally gavaging 10

CFU per

601

mice, B. The organ burden post 5 days of oral gavage in intestine, C. in Mesentric lymph node

602

(MLN), D. in spleen, E. in liver, F. The experimental protocol for organ burden in C57BL/6 mice

603

upon intraperitoneal (I.P) infection with 10

CFU per mice, G. The organ burden 3 days post I.P

604

infection in spleen, H. in liver, I. dissemination in blood, J. The hematoxylin and eosin staining of

605

the sections of liver 3 days post I.P infection of C57Bl/6 mice, K. Disease score from the

606

histopathogical sections of liver, Here, Small necrotic areas (N), Congestion and damage to the

607

endothelial lining of the central vein (C), Congestion of the hepatic portal vein and poor hepatic

608

architecture (HPV), Inflammatory immune cells (IC). The disease score is as :0 for normal

609

pathology, 1 for mild/ minor pathology, 2 for moderate pathology, and 3 for severe pathological

610

changes. Mann Whitney test was used to analyse organ burden in mice; p values ****<0.0001,

611

***<0.001, **<0.01, *<0.05. Student’s t-test was used to analyze the data; p values ****<0.0001,

612

***<0.001, **<0.01, *<0.05.

613

614

Salmonella rewires host polyamine metabolism to potentiate its survival within host

615

macrophages

616

Most of the intracellular pathogens establish their persistence in the phagocytic cells and are often

617

found to be associated with different populations of the macrophages. Like Brucella abortus

618

preferentially resides in the Alternatively activated macrophages (AAM), where it survives and

619

replicates by exploiting the host polyamines. A research group has shown that the metabolism of

620

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

AAM is shifted to increase polyamine biosynthesis by Brucella abortus and thereby promote the

621

bacterial survival [32]. Similarly, Salmonella also resides in macrophages, a way to lead chronic

622

infections. We were interested to know whether Salmonella exploits the host polyamines and leads

623

to a rewiring of host cell metabolism. To understand the host-pathogen relationship in depth, we

624

assessed the mRNA expression of Ornithine decarboxylase 1 (mOdc, mouse Ornithine

625

decarboxylase) which catalyses the rate-determining step of polyamine biosynthesis and

626

Spermidine synthase (mSrm, mouse Spermidine synthase) that synthesises spermidine from

627

putrescine by transferring the aminopropyl group from decarboxylated-S-adenosyl methionine in

628

RAW264.7 cells upon infection with STM WT. We observed that mRNA expression of mOdc1

629

and mSrm were upregulated at 6 and 16 hours post-infection (Fig 5 A and B). Further, the mRNA

630

expression of mOdc1 and mSrm were enhanced in the spleen and liver of C57BL/6 mice 5 days

631

post oral gavage with STM WT (Fig C- F). Our results show that Salmonella Typhimurium upon

632

infection into the host enhances the expression of host polyamine biosynthesis genes. Moreover,

633

we have previously observed that the Salmonella Typhimurium that cannot import spermidine

634

cannot survive and proliferate as much as STM WT in macrophages. To further delve into the role

635

of host acquired polyamines, we knocked down Odc1 in RAW264.7 cells (Fig S6 A). Upon

636

knockdown of mOdc1, Salmonella Typhimurium showed significantly attenuated proliferation in

637

RAW264.7 cells (Fig 5 G). However, the knockdown of mOdc1 did not alter the percentage of

638

phagocytosis by the macrophages (Fig S7 C). Similarly, we knocked down Srm in RAW264.7 cells

639

and observed that knock-down of spermidine synthase in the host compromises the ability of STM

640

WT to proliferate and get phagocytosed by the macrophages (Fig S7 B, D and E).

641

The question that arises is how does Salmonella regulate the host polyamine metabolic pathways?

642

Salmonella utilizes Salmonella pathogenicity island 1 and 2 encoded effectors for its entry and

643

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

survival in the specialized host niches [33, 34]. Although SPI-1 is well studied in the initial process

644

of invasion, recent studies suggests that both SPI-1 and SPI-2 effectors are required for SCV

645

maturation and Salmonella survival in the host cells [35-37]. Thus to investigate how Salmonella

646

modulates host cell polyamine metabolism, we infected RAW264.7 cells with SPI-1 mutant (STM

647

∆invC) and a SPI-2 mutant (STM ∆ssaV), both of which lack the ability to import effectors of SPI-

648

1 and SPI-2 respectively. We observed that the mRNA expression of mOdc1 and mSrm, were

649

significantly downregulated in macrophages infected with STM ∆invC and STM ∆ssaV compared

650

to in STM WT (all normalized to Unifected) (Fig 5H and I). We further determined the

651

intracellular spermidine by immunofluorescence and found that macrophages infected with STM

652

∆invC and STM ∆ssaV showed reduced spermidine production compared to in STM WT and

653

uninfected cells (Fig 5J and K). Thus, our data suggests that Salmonella utilizes effectors from

654

SPI-1 and SPI-2 to modulate the host cell polyamine metabolic pathways.

655

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

656

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 5: Salmonella rewires host polyamine metabolism to potentiate its survival within host

657

macrophages

658

A.The mRNA expression of mOdc1 (mouse ornithine decarboxylase) in RAW264.7 cells upon

659

infection with STM WT, B. The mRNA expression of mSrm (mouse spermidine synthase) in

660

RAW264.7 cells upon infection with STM WT, C. The mRNA expression of mOdc1 (mouse

661

ornithine decarboxylase) in liver of C57BL/6 mice 5 days post infection with STM WT by oral

662

gavage, D. The mRNA expression of mSrm (mouse spermidine synthase) in liver of C57BL/6 mice

663

5 days post infection with STM WT by oral gavage,E. The mRNA expression of mOdc1 (mouse

664

ornithine decarboxylase) in spleen of C57BL/6 mice 5 days post infection with STM WT by oral

665

gavage, F. The mRNA expression of mSrm (mouse spermidine synthase) in spleen of C57BL/6

666

mice 5 days post infection with STM WT by oral gavage, G. The fold proliferation of STM WT in

667

RAW264.7 cells upon transient knockdown of mOdc1, H. The percentage phagocytosis of STM

668

WT in RAW264.7 cells upon transient knockdown of mOdc1. Here SCR is Scrambled (no target

669

for knock-down), two different targeted shRNA were used for knock-down purposes, Sh1is

670

shRNA-1 for knock-down, Sh2 is shRNA-2 for knock-down, and Sh1+Sh2 indicates where both

671

the shRNAs were used to obtain the knock-down, H. The mRNA expression of mOdc1 (mouse

672

ornithine decarboxylase) in RAW264.7 cells upon infection with STM WT, STM ∆ssaV, and STM

673

∆invC, normalized to the expression in uninfected macrophages, I. The mRNA expression of mSrm

674

(mouse spermidine synthase) in RAW264.7 cells upon infection with STM WT, STM ∆ssaV, and

675

STM ∆invC, normalized to the expression in uninfected macrophages, J. Immunofluorescence

676

imaging to study spermidine in RAW 264.7 cells upon infection with STM WT, STM ∆invC, STM

677

∆ssaV, here green is Anti-mouse LAMP1(Alexa fluor 488), Red is pFPV-M-cherry expressing

678

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Salmonella strains, magenta is anti-Spermidine (Alexa fluor 647) and UI- uninfected. Student’s

679

t-test was used to analyze the data; p values ****<0.0001, ***<0.001, **<0.01, *<0.05.

680

681

The chemopreventive drug DFMO, reduces Salmonella Typhimurium burden in the host by

682

enhancing nitric oxide production

683

Polyamines are essential molecules in eukarotes, with multiple roles in differentiation,

684

proliferation and development. Many studies have shown that polyamine levels are upregulated in

685

cancer cells and elevated levels of polyamines are associated with breast cancer, neuroblastoma,

686

hepatocellular carcinoma, prostate cancer, lung cancer, colorectal cancer and lukemia [38-42].

687

D,L-α-difluoromethylornithine (DFMO) , an inhibitor of ODC was developed as a potent drug to

688

treat cancer in the year 1970 [43]. DFMO irreversibly binds to the active site of ODC and acts as

689

a suicide inhibitor, thereby reducing polyamine levels and having a cytostatic effect. As a single

690

therapeutic agent it was found to be effective only in neuroblastoma , and cinical trial for other

691

cancer type was unsatisfactory [44, 45]. However, DFMO has been successfully developed as a

692

chemopreventive drug, with FDA approval for treatment of Human African Trypanosomiasis

693

(HAT) [46, 47]. To test whether DFMO can be used as a therapeutic drug against Salmonella

694

infection, we treated RAW264.7cells with DFMO during the infection with Salmonella

695

Typhimurium, and observed a significant attenuation in the fold proliferation of STM WT (Fig 6

696

A). Studies have shown that DFMO binds to ODC to prevent further production of putrescine from

697

ornithine and also acts on Arginase1 and reduces the available pool of ornithine for polyamine

698

biosynthesis [48, 49]. This, ensures the flux of arginine to be fed into the nitric oxide synthase

699

(NOS2) pathway and leads to elevated levels of nitric oxide in the cell, which in turn negatively

700

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

acts on ODC to further block polyamine synthesis [50].We assessed the mRNA expression of

701

mNos2 upon knockdown of mOdc1 in RAW264.7 cells followed by infection with STM. We

702

observed an upregulation of mOdc1 mRNA levels at 6 hours and 16 hours post infection in

703

RAW264.7 cells (Fig 6 B). Further we determined the levels of nitric oxide using a cell permeable

704

dye DAF-2DA, and noted that upon treatment with DFMO there was higher production of nitric

705

oxide (Fig 6 C and D). Taken together our results demonstrate the role of DFMO in diminishing

706

the survival and proliferation of Salmonella Typhimurium by blocking Odc1 and enhanced

707

production of nitric oxide in murine macrophages.

708

To further address whether DFMO also act as potent anti-Salmonella drug in mice model of

709

infection, we infected C57BL/6 mice with 10

CFU per mice by oral gavage. Every alternate day

710

the mice were injected with DFMO in two different doses for two different cohorts (2mg/kg of

711

body weight of mice and 1mg/kg of body weight of mice). Five days post oral gavage we observed

712

that DFMO treatment of 2mg/kg body weight of mice significantly reduced the colonisation of

713

STM WT in intestine, MLN, spleen and liver than in the untreated mice (Fig 6 E-I). Further, using

714

immunofluorescence we observed DFMO significantly reduced the colonisation of STM and and

715

the levels of spermidine in mouse ileum (Fig S8 A-D). Moreover, treatment of mice with DFMO

716

at a dose of 2mg/kg body weight of mice, increased the survival of mice upon infection with STM

717

WT (Fig 6 J and K). Also, the weight reduction in mice treated with with DFMO at a dose of

718

2mg/kg body weight, was less upon infection with STM WT (Fig S8 E). Next, we evaluated the

719

tissue damage upon infection of STM in mice liver. The results show that DFMO treatment of

720

mice at a dose of 2mg/kg of body weight significantly lowered the disease score, suggesting lesser

721

tissue damage compared to untreated (Fig 6 L and M ). Thus, DFMO serves as a potential drug to

722

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

treat Salmonella infection in mice by reducing the bacterial burden and tissue damage in mice and

723

enhancing the survival of mice upon infection with STM.

724

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

725

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Figure 6: The chemopreventive drug DFMO, reduces Salmonella Typhimurium burden in

726

the host by enhancing nitric oxide production

727

A. The intracellular fold proliferation of STM WT in RAW264.7 cells upon treatment with

728

Difluoromethyl ornithine (DFMO) (5µM), quantified by Gentamicin protection assay, B. The

729

mRNA expression of mNos2(mouse Nitric oxide synthase) in RAW264.7 cells, upon transient

730

knockdown of Odc1 followed by infection with STM WT, the expression is normalized to beta-

731

actin as an internal control, C. The representative scatter plot for DAF2DA dye positive population

732

of RAW264.7 cells, upon transient knockdown of Odc1 followed by infection with STM WT, the

733

data is determined by flow-cytometry post staining the infected macrophages with the dye, D. The

734

quantification of (C), E. The experimental procedure to study the organ burden of STM WT in

735

C57BL/6 mice upon treatment with DFMO, F-I. The organ burden of STM WT in Intestine, MLN,

736

Spleen and Liver of C57BL/6 mice upon intraperitoneal treatment of DFMO (2mg/kg and 1mg/kg

737

of body weight) as mentioned in (E), J. The Experimental procedure to study the survival of

738

C57BL/6 mice upon infection with STM WT and treatment with DFMO, K.The survival of

739

C57BL/6 mice upon infection with STM WT and upon intraperitoneal treatment of DFMO

740

(2mg/kg and 1mg/kg of body weight) as mentioned in (J), L. Hematoxylin and eosin staining of

741

histopathological sections of liver upon DFMO treatment to C57BL/6 mice. Here, Small necrotic

742

areas (N), Congestion and damage to the endothelial lining of the central vein (C), Congestion of

743

the hepatic portal vein and poor hepatic architecture (HPV), Inflammatory immune cells (IC), M.

744

The disease score index for liver tissue damage upon STM WT infection in C57BL/6 mice with

745

DFMO treatment. The disease score is as :0 for normal pathology, 1 for mild/ minor pathology, 2

746

for moderate pathology, and 3 for severe pathological changes, Student’s t-test was used to analyze

747

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

the data; p values ****<0.0001, ***<0.001, **<0.01, *<0.05. Two-way Anova was used to

748

analyze the grouped data; p values ****<0.0001, ***<0.001, **<0.01, *<0.05.

749

750

Discussion

751

A significant part of the life cycle of Salmonella during its pathogenesis involves its stay inside

752

the macrophages. The Gram-negative bacteria come across multiple host environmental stress

753

conditions inside the macrophages [51]. A key mechanism by which the host macrophages try to

754

limit the invading pathogen is by the rapid oxidative burst and ROS production. ROS includes

755

superoxide radicals, hydroxyl radicals, peroxy-nitrites, peroxy-chlorides and hydrogen peroxide.

756

ROS can pass through the bacterial cell wall and act on lipids, proteins and nucleic acids by

757

oxidising them and leading to cellular damage. Salmonella is often referred to as a smart pathogen.

758

Over the years, it has evolved to possess multiple strategies to combat the host-derived stresses

759

[52, 53]. To combat the oxidative burst generated NOX2 in macrophages, Salmonella possesses

760

multiple antioxidant enzymes such as the catalases KatE and KatG, the superoxide dismutases

761

SodA and SodB, the Alkyl hydroperoxide reductase, the glutaredoxins and thioredoxins and Hrg

762

transcriptional regulator [25]. Polyamines assist in Salmonella virulence and aid in stress

763

resistance. However, the mechanism behind the role of polyamines in Salmonella stress resistance

764

and virulence remains less appreciated.

765

Our study identifies spermidine as a stress-responsive regulatory molecule in Salmonella

766

Typhimurium. We show spermidine is critical for the survival and proliferation of STM in

767

macrophages and in the presence of oxidative stress in vitro. The spermidine transporter and

768

biosynthesis mutants show significant reduced capability to be phagocytosed by the macrophages.

769

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Findings from our previous study showed that the intracellular level of spermidine is significantly

770

less in both the spermidine transport and biosynthesis mutants, which further explains the

771

attenuated proliferation of both the mutants in macrophages [16]. The previous findings also

772

showed that in absence of the transport genes in Salmonella Typhimurium the synthesis genes are

773

down regulated and vice versa. The absence of spermidine transport and biosynthesis diminishes

774

the mRNA expression of multiple arms of oxidative stress response in Salmonella. Multiple studies

775

show the role of polyamines in the regulation of transcription of multiple genes by interacting with

776

DNA in eukaryotes. They bind to the DNA and change conformation as in C-MYC transcription,

777

and in other cases, enhance DNA-protein binding affinities like for Estrogen-response elements

778

[54, 55]. Thus, in Salmonella Typhimurium rpoS, soxR and emrR fall under the “Polyamine

779

modulon”.

780

We further characterise a novel enzyme GspSA in Salmonella Typhimurium, which synthesises a

781

conjugated product of glutathione (GSH) and spermidine called GS-sp. GspSA is critical for

782

Salmonella Typhimurium to survive and proliferate in macrophages. Our study shows that the

783

absence of GspSA attenuates the survival of STM under oxidative stress conditions in vitro,

784

suggesting a vital role of GspSA in protecting STM from oxidative damage. GS-sp in E. coli carries

785

out its function by modifying protein thiol groups under oxidative stress to protect the proteins

786

from getting oxidised and damaged. In Salmonella Typhimurium, we further show that the

787

spermidine regulates the gsp expression and the subsequent production of GS-sp. We expect that

788

GS-sp generated under oxidative stress conditions and inside the macrophages, likewise interact

789

with cysteine-thiol groups in proteins to shield them from oxidative damage. MK Chattopadhyay

790

(2013) showed that this enzyme (GspSA) is specifically present in two groups of organisms namely

791

bacteria and kineto-plastids respectively and completely absent in other organisms such as human,

792

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

rats, drosophila etc., among the bacteria group in all enterobacteria showed 27-100% homology

793

and >65% identity in around 50% of the species, with E. coli [30]. Thus, the absence of GspSA in

794

eukaryotes makes it a potent drug target for treatment of Salmonella infections. Thus, spermidine

795

exerts pleiotropic effects in Salmonella, by orchestrating the multiple arms of antioxidative

796

response. It strengthens the bacteria to cope with hostile host environments. Our studies in the in

797

vivo model of Salmonella Typhimurium infection further dissect the role of spermidine in assisting

798

in the pathogenesis by enhancing its virulence properties. Moreover, it is at the nexus of multiple

799

oxidative stress response arms in Salmonella, thereby assisting in mounting an antioxidative

800

response to promote its survival in macrophages.

801

Studies over the past years have given insights into the function of polyamines in the pathogenesis

802

of multiple virulent bacteria. Many human pathogenic bacteria have developed ways to exploit,

803

interfere and manipulate the polyamine metabolism of the host to enhance their fitness within the

804

niche. As in Shigella and Vibrio spp. polyamines produced by the bacteria are critical in

805

determining virulence [56, 57]. While bacteria such as Legionella spp., which does not synthesise

806

polyamines, depend on the host-acquired polyamines for their pathogenesis [58]. Another unique

807

mechanism is observed in H. pylori which activates polyamine oxidation, thereby dysregulation

808

the innate-immune response [59]. Also, in response to H. pylori infection, the host macrophages

809

increase the arginase activity and ornithine decarboxylase activity to produce polyamines [60]. A

810

recent study showed that SARS-Cov-2 hijacks the host polyamines for its reproduction and

811

infection [61].

812

Our findings also reveal that Salmonella Typhimurium enhances the polyamine production in the

813

host upon infection using its specialized pathogenicity island encoded effectors from SPI-1 and

814

SPI-2, which might be a strategy to hijack the host polyamines for its survival. This further explains

815

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

another reason for the reduced proliferation observed for the spermidine transport mutant in

816

macrophages. Also, the knockdown of host ornithine decarboxylase attenuates Salmonella

817

proliferation in macrophages. The upregulation of Odc1 activity manages to feed the amino acid,

818

L-arginine, into the polyamine biosynthesis and prevent nitric oxide production, which otherwise

819

would be detrimental for the pathogen. Polyamines have been a well-studied as a drug target for

820

treatment of multiple cancers. The oncogene MYC is upregulated in 70 percent of the cancer types.

821

Ornithine decarboxylase (Odc1) rate limiting enzyme of polyamine biosynthesis is

822

transcriptionally activated, directly by MYC, thereby increasing Odc1 levels in cancer [62]. RAS

823

is another essential factor in cell growth and cancer development, and it directly acts on Odc1 to

824

translationally activate Odc1 in cancer cells [63]. Polyamines biosynthesis is also associated with

825

multiple other oncogenes such as AKT1 and mTORC1 [64, 65]. We further show that upon using

826

Odc1 suicide inhibitor, DFMO, Salmonella proliferation could be diminished, and it reduces the

827

bacterial burden in mice. The FDA-approved chemopreventive drug, DFMO, for Human African

828

Trypanosomiasis treatment serves to be a potential drug to cure Salmonella infection. In the

829

devastating age of increasing antibiotic resistance, such a drug promises to effectively combat

830

deadly pathogens like Salmonella.

831

Conclusion

832

A substantial duration of the infection cycle of Salmonella involves the macrophages, which

833

present a very hostile environment to the bacteria. However, Salmonella is able to survive and

834

proliferate within host macrophages and utilizes it to disseminate to secondary sites of infection.

835

Our study identifies a novel strategy employed by Salmonella Typhimurium to counteract

836

oxidative and nitrosative stress within the host. We demonstrate that spermidine is a critical

837

regulatory molecule in Salmonella that regulates multiple antioxidative pathways along with a

838

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

novel antioxidative enzyme (GspSA) in Salmonella, to prevent oxidative damage and assist in its

839

virulence in mice. It further rewires host polyamine metabolism in a SPI-1 and SPI-2 dependent

840

manner to prevent nitric oxide production and enhance its survival. In the era of anti-microbial

841

resistance, this study further recognizes an FDA -approved chemopreventive drug, DFMO, which

842

inhibits the host-polyamine metabolism, as a prospective antidote to treat Salmonella infection.

843

844

845

Availability of data and materials

846

All data generated and analyzed during this study, including the supplementary information files,

847

are incorporated in this article. The data is available from the corresponding author on request.

848

Ethics statement

849

All the animal experiments were approved by the Institutional Animal Ethics Committee, and the

850

Guidelines provided by National Animal Care were strictly followed during all experiments.

851

(Registration No: 435 48/1999/CPCSEA).

852

Acknowledgement

853

Prof. G. Subba Rao from MCB, IISc is duly acknowledged for providing the for knockdown

854

generation. Divisional Mass Spectrometry facility IISc and Mrs. Sunita Joshi for the MS analysis.

855

Departmental Confocal Facility. Departmental Real-Time PCR Facility, Divisional Flowcytometry

856

facility and Central Animal Facility at IISc are duly acknowledged. Mr Sumith and Ms Navya are

857

acknowledged for their help in image acquisition. Dr. Ritika Chatterjee, Mr. Amartya Mukherjee,

858

Mr. Sushovan Bhattacharyya and Ms Bhavya Joshi are also acknowledged for technical help.

859

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

Author contribution statement

860

AVN and DC conceived the study. AVN and DC designed experiments. AVN and AS performed

861

experiments. RSR analysed for the disease score of tissue samples. AVN, analyzed the data,

862

prepared the figures and wrote the manuscript draft. AVN, AS and DC reviewed and edited the

863

manuscript. DC supervised the work. All the authors read and approved the manuscript.

864

Funding

865

This work is supported by the Department of Biotechnology (DBT), Ministry of Science and

866

Technology, the Department of Science and Technology (DST), Ministry of Science and

867

Technology. DC acknowledges DAE-SRC ((DAE00195) outstanding investigator award and funds

868

and ASTRA Chair Professorship funds. The authors jointly acknowledge the DBT-IISc partnership

869

program. Infrastructure support from ICMR (Center for Advanced Study in Molecular Medicine),

870

DST (FIST), UGC-CAS (special assistance), and TATA fellowship is duly acknowledged. AVN

871

acknowledges the IISc-MHRD for financial assistance. AS acknowledges UGC for the financial

872

assistance. RSR acknowledges IISc for the financial help.

873

Conflict of Interest

874

The authors declare no conflict of interest.

875

Reference

876

877

1. Hernandez, D.J., et al., Environmental stress destabilizes microbial networks. The ISME Journal,

878

2021. 15(6): p. 1722-1734.

879

2. Gray, D.A., et al., Extreme slow growth as alternave strategy to survive deep starvaon in

880

bacteria. Nature communicaons, 2019. 10(1): p. 890.

881

3. Wineld, M.D. and E.A. Groisman, Role of nonhost environments in the lifestyles of Salmonella

882

and Escherichia coli. Applied and environmental microbiology, 2003. 69(7): p. 3687-3694.

883

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

4. Dodd, C.E., P.J. Richards, and T.G. Aldsworth, Suicide through stress: a bacterial response to sub-

884

lethal injury in the food environment. Internaonal journal of food microbiology, 2007. 120(1-2):

885

p. 46-50.

886

5. Marmion, M., et al., Survive and thrive: Control mechanisms that facilitate bacterial adaptaon

887

to survive manufacturing-related stress. Internaonal Journal of Food Microbiology, 2022. 368:

888

p. 109612.

889

6. Vazquez-Torres, A. and F.C. Fang, Salmonella evasion of the NADPH phagocyte oxidase. Microbes

890

and infecon, 2001. 3(14-15): p. 1313-1320.

891

7. Chakravory, D., I. Hansen-Wester, and M. Hensel, Salmonella pathogenicity island 2 mediates

892

protecon of intracellular Salmonella from reacve nitrogen intermediates. The Journal of

893

experimental medicine, 2002. 195(9): p. 1155-1166.

894

8. Rhee, H., E.J. Kim, and J. Lee, Physiological polyamines: simple primordial stress molecules.

895

Journal of cellular and molecular medicine, 2007. 11(4): p. 685-703.

896

9. Barbagallo, M., et al., A new piece of the Shigella pathogenicity puzzle: spermidine

897

accumulaonby silencing of the speG gene. PloS one, 2011. 6(11): p. e27226.

898

10. Johnson, L., et al., Surface-localized spermidine protects the Pseudomonas aeruginosa outer

899

membrane from anbioc treatment and oxidave stress. Journal of bacteriology, 2012. 194(4):

900

p. 813-826.

901

11. Banerji, R., P. Iyer, and S.D. Saroj, Spermidine enhances the survival of Streptococcus pyogenes

902

M3 under oxidave stress. Molecular Oral Microbiology, 2022. 37(2): p. 53-62.

903

12. Espinel, I.C., P.R. Guerra, and L. Jelsbak, Mulple roles of putrescine and spermidine in stress

904

resistance and virulence of Salmonella enterica serovar Typhimurium. Microbial pathogenesis,

905

2016. 95: p. 117-123.

906

13. Datsenko, K.A. and B.L. Wanner, One-step inacvaon of chromosomal genes in Escherichia coli

907

K-12 using PCR products. Proceedings of the Naonal Academy of Sciences, 2000. 97(12): p.

908

6640-6645.

909

14. Roy Chowdhury, A., et al., Salmonella Typhimurium outer membrane protein A (OmpA) renders

910

protecon from nitrosave stress of macrophages by maintaining the stability of bacterial outer

911

membrane. PLoS Pathogens, 2022. 18(8): p. e1010708.

912

15. Salbitani, G., C. Boone, and S. Carfagna, Determinaon of reduced and total glutathione

913

content in extremophilic microalga Galdieria phlegrea. Bio-protocol, 2017. 7(13): p. e2372-

914

e2372.

915

16. Nair, A.V., et al., Spermidine facilitates the adhesion and subsequent invasion of Salmonella

916

Typhimurium into epithelial cells via the regulaon of surface adhesive structures and the SPI-1.

917

bioRxiv, 2023: p. 2023.06. 03.543567.

918

17. Achouri, S., et al., The frequency and duraon of Salmonella–macrophage adhesion events

919

determines infecon eciency. Philosophical Transacons of the Royal Society B: Biological

920

Sciences, 2015. 370(1661): p. 20140033.

921

18. Kim, J.-S., et al., Oxidave stress acvates transcripon of Salmonella pathogenicity island-2

922

genes in macrophages. Journal of Biological Chemistry, 2022. 298(7).

923

19. Miller, J.F., J.J. Mekalanos, and S. Falkow, Coordinate regulaon and sensory transducon in the

924

control of bacterial virulence. Science, 1989. 243(4893): p. 916-922.

925

20. Chaopadhyay, M.K., et al., Polyamines smulate the level of the σ38 subunit (RpoS) of

926

Escherichia coli RNA polymerase, resulng in the inducon of the glutamate decarboxylase-

927

dependent acid response system via the gadE regulon. Journal of Biological Chemistry, 2015.

928

290(29): p. 17809-17821.

929

21. Kashiwagi, K., et al., Idencaon of the polyamine-induced protein as a periplasmic oligopepde

930

binding protein. Journal of Biological Chemistry, 1990. 265(15): p. 8387-8391.

931

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

22. Sakamoto, A., et al., Three members of polyamine modulon under oxidave stress condions:

932

two transcripon factors (SoxR and EmrR) and a glutathione synthec enzyme (GshA). PLoS One,

933

2015. 10(4): p. e0124883.

934

23. Farr, S.B. and T. Kogoma, Oxidave stress responses in Escherichia coli and Salmonella

935

typhimurium. Microbiological reviews, 1991. 55(4): p. 561-585.

936

24. Storz, G., et al., Bacterial defenses against oxidave stress. Trends in Genecs, 1990. 6: p. 363-

937

368.

938

25. Rhen, M., Salmonella and reacve oxygen species: a love-hate relaonship. Journal of innate

939

immunity, 2019. 11(3): p. 216-226.

940

26. Song, M., et al., Low-molecular-weight thiol-dependent anoxidant and annitrosave defences

941

in S almonella pathogenesis. Molecular microbiology, 2013. 87(3): p. 609-622.

942

27. Bollinger, J.M., et al., Glutathionylspermidine Metabolism in Escherichia coli.: PURIFICATION,

943

CLONING, OVERPRODUCTION, AND CHARACTERIZATION OF A BIFUNCTIONAL

944

GLUTATHIONYLSPERMIDINE SYNTHETASE/AMIDASE

∗

. Journal of Biological Chemistry, 1995.

945

270(23): p. 14031-14041.

946

28. Krauth-Siegel, R.L. and H. Lüdemann, Reducon of dehydroascorbate by trypanothione.

947

Molecular and biochemical parasitology, 1996. 80(2): p. 203-208.

948

29. Ariyanayagam, M.R. and A.H. Fairlamb, Ovothiol and trypanothione as anoxidants in

949

trypanosomads. Molecular and biochemical parasitology, 2001. 115(2): p. 189-198.

950

30. Chaopadhyay, M.K., W. Chen, and H. Tabor, Escherichia coli glutathionylspermidine

951

synthetase/amidase: phylogeny and eect on regulaon of gene expression. FEMS microbiology

952

leers, 2013. 338(2): p. 132-140.

953

31. Wyllie, S., et al., Dissecng the essenality of the bifunconal trypanothione synthetase-amidase

954

in Trypanosoma brucei using chemical and genec methods. Molecular microbiology, 2009.

955

74(3): p. 529-540.

956

32. Kerrinnes, T., et al., Ulizaon of host polyamines in alternavely acvated macrophages

957

promotes chronic infecon by Brucella abortus. Infecon and immunity, 2018. 86(3): p.

958

10.1128/iai. 00458-17.

959

33. Forest, C.G., et al., Intracellular survival of Salmonella enterica serovar Typhi in human

960

macrophages is independent of Salmonella pathogenicity island (SPI)-2. Microbiology, 2010.

961

156(12): p. 3689-3698.

962

34. Ellermeier, J.R. and J.M. Slauch, Adaptaon to the host environment: regulaon of the SPI1 type

963

III secreon system in Salmonella enterica serovar Typhimurium. Current opinion in microbiology,

964

2007. 10(1): p. 24-29.

965

35. Brawn, L.C., R.D. Hayward, and V. Koronakis, Salmonella SPI1 eector SipA persists aer entry

966

and cooperates with a SPI2 eector to regulate phagosome maturaon and intracellular

967

replicaon. Cell host & microbe, 2007. 1(1): p. 63-75.

968

36. Giacomodonato, M.N., et al., SipA, SopA, SopB, SopD and SopE2 eector proteins of Salmonella

969

enterica serovar Typhimurium are synthesized at late stages of infecon in mice. Microbiology,

970

2007. 153(4): p. 1221-1228.

971

37. Brown, N.F., et al., Salmonella pathogenicity island 2 is expressed prior to penetrang the

972

intesne. PLoS pathogens, 2005. 1(3): p. e32.

973

38. Kingsnorth, A., A.B. Lumsden, and H.M. Wallace, Polyamines in colorectal cancer. Journal of

974

Brish Surgery, 1984. 71(10): p. 791-794.

975

39. Evageliou, N.F. and M.D. Hogarty, Disrupng polyamine homeostasis as a therapeuc strategy for

976

neuroblastoma. Clinical Cancer Research, 2009. 15(19): p. 5956-5961.

977

40. Tsoi, T.-H., et al., Urinary polyamines: a pilot study on their roles as prostate cancer detecon

978

biomarkers. PLoS One, 2016. 11(9): p. e0162217.

979

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

41. Choi, Y., et al., Targeng ODC1 inhibits tumor growth through reducon of lipid metabolism in

980

human hepatocellular carcinoma. Biochemical and biophysical research communicaons, 2016.

981

478(4): p. 1674-1681.

982

42. Soda, K., The mechanisms by which polyamines accelerate tumor spread. Journal of Experimental

983

& Clinical Cancer Research, 2011. 30: p. 1-9.

984

43. Weeks, C., et al., alpha-Diuoromethylornithine, an irreversible inhibitor of ornithine

985

decarboxylase, inhibits tumor promoter-induced polyamine accumulaon and carcinogenesis in

986

mouse skin. Proceedings of the Naonal Academy of Sciences, 1982. 79(19): p. 6028-6032.

987

44. Horn, Y., P.J. Schechter, and L.J. Marton, Phase I–II clinical trial with alpha-

988

diuoromethylornithine—an inhibitor of polyamine biosynthesis. European Journal of Cancer and

989

Clinical Oncology, 1987. 23(8): p. 1103-1107.

990

45. Lewis, E.C., et al., A subset analysis of a phase II trial evaluang the use of DFMO as maintenance

991

therapy for high-risk neuroblastoma. Internaonal journal of cancer, 2020. 147(11): p. 3152-

992

3159.

993

46. Roberts, S. and B. Ullman, Parasite polyamines as pharmaceucal targets. Current

994

Pharmaceucal Design, 2017. 23(23): p. 3325-3341.

995

47. Eperon, G., et al., Treatment opons for second-stage gambiense human African

996

trypanosomiasis. Expert review of an-infecve therapy, 2014. 12(11): p. 1407-1417.

997

48. Li, H., et al., Acvies of arginase I and II are liming for endothelial cell proliferaon. American

998

Journal of Physiology-Regulatory, Integrave and Comparave Physiology, 2002. 282(1): p. R64-

999

R69.

1000

49. Bojarska, J., et al., The rst insight into the supramolecular system of D, L-α-

1001

diuoromethylornithine: A new anviral perspecve. Froners in Chemistry, 2021. 9: p. 679776.

1002

50. Bauer, P.M., et al., Nitric Oxide Inhibits Ornithine Decarboxylase viaS-Nitrosylaon of Cysteine

1003

360 in the Acve Site of the Enzyme. Journal of Biological Chemistry, 2001. 276(37): p. 34458-

1004

34464.

1005

51. HaragaA, O.M., MillerSI. Salmonellaeinterplaywith hostcells, 2008. 6(1): p. 53G66.

1006

52. Buchmeier, N.A. and F. Heron, Inducon of Salmonella stress proteins upon infecon of

1007

macrophages. Science, 1990. 248(4956): p. 730-732.

1008

53. Shen, S. and F.C. Fang, Integrated stress responses in Salmonella. Internaonal journal of food

1009

microbiology, 2012. 152(3): p. 75-81.

1010

54. Kumar, N., R. Basundra, and S. Mai, Elevated polyamines induce c-MYC overexpression by

1011

perturbing quadruplex–WC duplex equilibrium. Nucleic acids research, 2009. 37(10): p. 3321-

1012

3331.

1013

55. Thomas, T., et al., Polyamine-mediated conformaonal perturbaons in DNA alter the binding of

1014

estrogen receptor to poly (dG-m5dC). poly (dG-m5dC) and a plasmid containing the estrogen

1015

response element. The Journal of Steroid Biochemistry and Molecular Biology, 1995. 54(3-4): p.

1016

89-99.

1017

56. Prosseda, G., et al., Shedding of genes that interfere with the pathogenic lifestyle: the Shigella

1018

model. Research in microbiology, 2012. 163(6-7): p. 399-406.

1019

57. Lee, J., et al., An alternave polyamine biosynthec pathway is widespread in bacteria and

1020

essenal for biolm formaon in Vibrio cholerae. Journal of Biological Chemistry, 2009. 284(15):

1021

p. 9899-9907.

1022

58. Nasrallah, G.K., et al., Legionella pneumophila requires polyamines for opmal intracellular

1023

growth. Journal of bacteriology, 2011. 193(17): p. 4346-4360.

1024

59. Chaturvedi, R., et al., Spermine oxidase mediates the gastric cancer risk associated with

1025

Helicobacter pylori CagA. Gastroenterology, 2011. 141(5): p. 1696-1708. e2.

1026

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint

60. Chaturvedi, R., et al., Inducon of polyamine oxidase 1 by Helicobacter pylori causes macrophage

1027

apoptosis by hydrogen peroxide release and mitochondrial membrane depolarizaon. Journal of

1028

Biological Chemistry, 2004. 279(38): p. 40161-40173.

1029

61. Makhoba, X.H. and S. Makumire, The capture of host cell’s resources: The role of heat shock

1030

proteins and polyamines in SARS-COV-2 (COVID-19) pathway to viral infecon. Biomolecular

1031

Concepts, 2022. 13(1): p. 220-229.

1032

62. Bachmann, A.S. and D. Geerts, Polyamine synthesis as a target of MYC oncogenes. Journal of

1033

Biological Chemistry, 2018. 293(48): p. 18757-18769.

1034

63. Origan, S. and L.M. Shantz, Ras transformaon of RIE-1 cells acvates cap-independent

1035

translaon of ornithine decarboxylase: regulaon by the Raf/MEK/ERK and phosphadylinositol

1036

3-kinase pathways. Cancer research, 2007. 67(10): p. 4834-4842.

1037

64. Gomes, A.P., T. Schild, and J. Blenis, Adding polyamine metabolism to the mTORC1 toolkit in cell

1038

growth and cancer. Developmental cell, 2017. 42(2): p. 112-114.

1039

65. Dai, F., et al., Extracellular polyamines-induced proliferaon and migraon of cancer cells by

1040

ODC, SSAT, and Akt1-mediated pathway. An-Cancer Drugs, 2017. 28(4): p. 457-464.

1041

1042

1043

1044

1045

1046

1047

1048

1049

1050

1051

1052

1053

1054

1055

1056

1057

The copyright holder for this preprintthis version posted November 10, 2023. ; https://doi.org/10.1101/2023.09.29.560257doi: bioRxiv preprint