1 Non-Lactose Fermenting (NLF) Escherichia coli: Following In The Footsteps

2 of Lactose Fermenting E. coli High-risk Clones

3
4
5

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

1|Page
26 Abstract

27 Multiresistant pathogenic strains of non-lactose fermenting Escherichia coli (NLF E. coli) are
28 responsible for various intestinal and extraintestinal infections. Although a few studies have
29 characterised such strains to some extent using conventional methods, it has not been
30 comprehensively studied at the genomic level. To address this gap, we used whole-genome
31 sequencing (WGS) coupled with detailed microbiological and biochemical testing to investigate
32 17 NLF E. coli from a diagnostic centre (icddr,b) in Dhaka, Bangladesh. The prevalence of NLF
33 E. coli was 10%, of which 47% (8/17) exhibited MDR phenotypes. All 17 (100%) isolates
34 couldn't ferment lactose sugar and were confirmed as E. coli by API and WGS species
35 identification. WGS- based analysis revealed international high-risk clonal lineages. The most
36 prevalent sequence types (STs) were ST131 (23%), ST1193(18%), ST12 (18%), ST501(12%),
37 ST167(6%), ST 73 (6% and ST12 (6%). Phylogenetic analysis corroborated a striking clonal
38 population among the studied NLF E. coli isolates. The predominant phylogroup detected was
39 B2 (65%). The blaCTX-M-15 ESBL gene was present in 53% of isolates (9/17). A total of 11 out of
40 17 isolates were affiliated with pathogenic pathotypes. All ExPEC pathotypes (100%)
41 demonstrated β-hemolysis. Our study underscores the presence of critical pathogens and MDR
42 clones among NLF E. coli, similar to the lactose fermenting (LF) E. coli. We suggest that NLF
43 E. coli be considered equally capable as lactose fermenting E. coli in causing intestinal and
44 extraintestinal infections. Further, there is a need to undertake systematic, unbiased monitoring
45 of predominant lineages among NLF E. coli that would help in better treatment and prevention
46 strategies.

47

48

49

50

51

52

53

2|Page
54

55 Introduction

56 Escherichia coli is a versatile Gram-negative bacterium with huge genetic diversity (nia Santos
57 Braz et al.; Mazumder et al., 2021b). It is the most common organism responsible for
58 opportunistic infections. Its primary habitat includes the lower intestinal tract of humans and
59 animals (nia Santos Braz et al.). However, infections with variant strains of E. coli are
60 responsible for various clinical manifestations ranging from diarrhea, urinary tract infections,
61 and life-threatening septicemia, resulting in over two million deaths every year globally (Hussain
62 et al., 2012; Huang et al. Moreover, these infections are often associated with cephalosporin and
63 carbapenem-resistant strains, impacting the mortality of patients and imparting huge health care
64 costs (Logan and Weinstein, 2017)

65 E. coli are nonpathogenic facultative anaerobic flora of the intestinal tract in humans. The gram-
66 negative E. coli bacilli ferments lactose to produce hydrogen sulfide. However, up to 20% of E.
67 coli isolates from patients are reported to be atypical, which are slow or non-lactose fermenters
68 due to the deficiency in enzyme lactose permease encoded by the lacY gene. (Nicoletti et al.,
69 1988; Hossain, 2012; Chang et al., 2014; Yaratha et al., 2017; Johnson et al., 2019). The (NLF)
70 E. coli can be identified as colourless, transparent colonies on MacConkey agar and by negative
71 lactose (sugar) fermentation tests (Hossain, 2012; Siqueira et al., 2021). Various strains of E.
72 coli are equipped to cause different forms of enteric and extraintestinal infections in human hosts
73 with varying propensities (Kaper et al., 2004). The pathogenic strains of E. coli belong to both
74 lactose fermenting and non-lactose fermenting E. coli types. However, little attention is given to
75 these atypical strains of E. coli, particularly the non-lactose fermenting E. coli in routine
76 diagnostic testing laboratories.

77 Escherichia coli as an etiological agent of diarrhea is well established (Colonna et al., 1992;
78 Nataro and Kaper, 1998; Hossain, 2012). Diarrheagenic strains of E. coli fall into different
79 categories possessing distinct pathogenic mechanisms, the three important E. coli strains include
80 enteropathogenic E. coli (EPEC), enterohemorrhagic E. coli (EHEC), and enteroaggregative E.
81 coli (EAEC) (Nataro and Kaper, 1998). These E. coli categories cause a range of clinical
82 syndromes such as watery diarrhea in children, persistent diarrhea, traveller's diarrhea and
83 hemolytic uremic syndrome (Nataro and Kaper, 1998). A few studies have reported that NLF E.

3|Page
 84 coli strains were identified as typical pathogenic E. coli isolates, having a possible diarrheagenic
85 role (Nicoletti et al., 1988; Colonna et al., 1992; Hossain, 2012)

86 Urinary tract infections (UTIs) represent one of the most common bacterial infections. It is a
87 serious public health problem affecting around 150 million people worldwide (Flores-Mireles et
88 al., 2015). Escherichia coli is the most common pathogen responsible for complicated and
89 uncomplicated urinary tract infections (Flores-Mireles et al., 2015). Such strains harbour
90 multiple virulence factors which play their role in the pathophysiology of UTIs. These Virulence
91 factors affect bacteria colonisation and evasion of host defences (Hussain et al., 2014; Mazumder
92 et al., 2020). Recent reports suggest that clonal groups of multiresistant, multivirulent E. coli are
93 increasingly responsible for UTIs, presenting a challenge for treating UTIs (Riley, 2014;
94 Mazumder et al., 2021a). In recent years, a few studies have demonstrated that NLF E. coli have
95 been increasingly isolated from urine specimens in the microbiology laboratories and have
96 reported their clinical significance concerning antibiotic resistance, virulence and emerging
97 clones (Chang et al., 2014; Chakraborty et al., 2016; Wu et al., 2017)

98 Rapid identification of bacterial pathogens in microbiology laboratories is critical for initiating

99 successful infection treatment. Screening of gram-negative bacteria from urine and stool samples
100 is routinely performed on MacConkey agar, and the colourless transparent non-lactose
101 fermenting (NLF) E. coli variants are usually missed in the screening process as little attention is
102 given to these atypical strains of E. coli. Although reports of such isolates are limited worldwide,
103 a few studies have reported the occurrence and characterisation of NLF E. coli from clinical
104 specimens. They have characterised the strains using only conventional methods. This study
105 aims to decipher the biochemical profiles, population structure, and genomic characteristics of
106 NLF E. coli from stool and urine samples isolated at a referral diagnostic centre (icddr,b) in
107 Dhaka, Bangladesh. Studies such as these will inform us about the evolutionary trajectories of
108 atypical E. coli variants and shed light on their clinical and public health significance.

109
110 Materials and methods

111 Bacterial isolates

4|Page
112 E. coli (175 isolates) was isolated from urine (98 isolates) and stool (77 isolates) culture plates.
113 The isolates were collected randomly from the Clinical Microbiology Laboratory of the
114 International Center for Diarrheal Disease Research, Bangladesh (icddr,b), located in Dhaka,
115 Bangladesh. The isolates were sampled between (month) 2019 and (month)2020 as part of a
116 larger study on 1% AMR surveillance. From this collection of 98 urine and 77 stool E. coli
117 isolates, we identified 9 and 8 NLF E. coli isolates, respectively. These 17 NLF E. coli were
118 subjected to complete biochemical profiling and whole-genome sequencing. NLF E. coli were
119 presumptively identified by their inability to ferment lactose on the MacConkey agar. Further,
120 confirmation was made by standard biochemical methods, sugar fermentation tests and API 20E
121 (Table 1). Hemolytic properties of strains were evaluated using 5% sheep blood agar plates.
122 Clearing zones around the colonies were suggestive of beta-hemolysis.

123 Antimicrobial susceptibility testing

124 The Genome Centre at icddr,b reassessed 17 NLF E. coli isolates with an extensive panel of 18
125 antibiotics (Oxoid, US) by the Kirby-Bauer disk diffusion method. The results were interpreted
126 by referring to the 2019 CLSI guidelines. Upon AST the NLF E. coli isolates identified as
127 intermediate and resistant were considered as resistant. They were classified as MDR if they
128 were nonsusceptible to at least one agent belonging to three different antimicrobial classes.

129 Whole-genome sequencing

130 Total bacterial DNA was isolated and purified from the overnight grown bacterial cultures using
131 the QIAmp DNA Mini kit (Qiagen, Germany). DNA QC and quantification were performed
132 employing a Nanopore spectrophotometer (Thermo Fisher Scientific, United States) and
133 Qubit4.0 fluorometer (Life Technologies) (Baddam et al., 2020). DNA libraries for the short-
134 read paired-end sequencing were prepared using the Nextera DNA Flex library prep kit
135 (Illumina). Size selected and pooled libraries were sequenced at the icddr,b Genome Centre in a
136 single Illumina Miseq using a 150-base paired-end V3 chemistry kit (Illumina).

137 Sequence assembly and annotation

138 High-quality reads were obtained by filtering the paired-end data using fastp (Chen et al.). De
139 novo assemblies were generated using SPAdes v.3.11.11 (Bankevich et al., 2012). NCBI
140 Prokaryotic Genome Annotation Pipeline (PGAP) (Tatusova et al., 2016) was used for

5|Page
141 annotating the genome assemblies. The genome statistics from the resulting file were gleaned
142 using quast 5.2.0 (Gurevich et al., 2013)

143 In silico analysis

144 Species identification was confirmed using Kmer Finder version 3.2 (Hasman et al., 2014;
145 Larsen et al., 2014; Clausen et al., 2018). Phylogenetic groups of the genomes were inferred
146 using the Clermon Typing tool (Beghain et al., 2018). The sequence types (STs) were
147 determined by submitting the contigs to MLST version 2.0, scheme #1 (Larsen et al., 2012).
148 SerotypeFinder 2.0 (Joensen et al., 2015) and CH typer 1.0 (Roer et al., 2018) were used for
149 elucidating serotypes and clonotypes, respectively.

150 Resistance genes were screened by BLASTn analysis of the genome assemblies against the data
151 downloaded from the ResFinder database (Bortolaia et al., 2020). The 70% query coverage and
152 90% identity indicated a positive hit in the genome. Plasmid incompatibility groups were
153 determined by PlasmidFinder 2.1 (Carattoli et al., 2014). Mobile genetic elements associated
154 with acquired AMR genes were determined using Mobile Element finder v1.0.3 (Johansson et
155 al., 2021). Chromosomal point mutations conferring fluoroquinolone resistance were identified
156 by PointFinder (Zankari et al., 2017). Contigs with blaNDM-5 and blaCTX-M-15 were analysed by
157 BLASTn analysis against the database in NCBI to identify the chromosomal or plasmid origin of
158 these genes.

159 Virulence gene content of strains was determined by BLASTn analysis of 17 genomes against
160 the E. coli database selected in VFDB (Chen et al., 2005). The genes with 90% identity and 70%
161 query coverage were considered present. Isolates were classified as ExPEC based on Johnson's
162 criteria (Johnson and Stell, 2000)

163 The In-silico analysis described above was performed on default parameters unless otherwise
164 stated.

165 Phylogenomic and pangenomic analysis

166 The phylogenetic tree was inferred with the reference sequence alignment-based phylogeny
167 builder (REALPHY) (Bertels et al., 2014) using the default 50-read length parameter. E. coli K-
168 12 MG1655 genome (accession No.U00096) was used as the single reference sequence and

6|Page
169 PhyML (Guindon et al., 2010) was used as the tree builder. The phylogenetic tree was later
170 visualised using the Interactive Tree of Life (iTOL) (Letunic and Bork, 2016). Pangenomic
171 analysis was conducted using the Anvi'o 7.1 (Eren et al., 2015) software package following the
172 microbial pangenomics analysis workflow.

173 Data availability

174 The whole-genome sequence data of the 17 studied NLF E. coli isolates were deposited in NCBI
175 (GenBank) under the BioProject identifier (ID) PRJNA839767. The accession numbers of all 17
176 genomes are enlisted in Table 2.

177 Results

178 Bacterial and biochemical characteristics

179 The colourless non-lactose fermenting (NLF) colonies on the MacConkey agar plate appeared
180 like Shigella spp colonies. Colonies were colourless (NLF) even in 24 hr old MacConkey agar
181 cultures, removing the possibility of late lactose fermenters. Serology of all colonies revealed
182 negative reaction to the four serogroups of Shigella: S. flexneri, S. dysenteriae, S. boydii, and S.
183 sonnei. Further, streaking of colonies on CHROMagar Orientation media revealed small, pink-
184 red colonies typical of E. coli. The isolates were then tested by routine biochemical tests
185 described in table 1 and identified as E. coli. Notably, all isolates were ONPG positive and did
186 not ferment lactose sugar. Further confirmation was done by generating a biochemical profile for
187 each isolate using 20 reactions of the API 20E strip. The API results gave three distinct API
188 profiles: 7144532 (n=11), 5044552 (n=3) and 5144572 (n=3) and identified all isolates as E. coli
189 with 99.8% probability. Gram-staining confirmed gram-negative bacilli. The optimum growth
190 temperate of NLF E. coli isolates ranged from 26°-42°C.

191 Molecular analysis and phylogenomics of NLF E. coli isolates

192 WGS of the 17 NLF E. coli study isolates yielded an average genome size of 5,160,336 bp
193 (range 4,890,228 to 5,480,336 and the average GC content was 50.6% (range 50.3 to 50.7).
194 KmerFinder results confirmed all the isolates as E. coli. On average, the genome assemblies had
195 a genome coverage of 86X (range from 62 X to 124X) (Table 2).

7|Page
196 According to in silico MLST, isolates were classified into nine distinct sequence types (STs) and
197 they represented some of the international high-risk lineages as shown in Fig. 1. ST131 was the
198 predominant ST that accounted for 4 (23%) of the 17 NLF E. coli isolates. ST1193 and ST12
199 were present in equal proportions, 3 (18%) each, followed by ST501 2 (12%). Other significant
200 STs detected were ST167 (6%), ST73 (6%) and ST12 (6%). The predominant phylogroup
201 detected was B2 (65%) which comprised all isolates affiliated with ST131, ST1193, ST12 and
202 ST73. This was followed by group A (n=2: 1ST167; 1 ST2089); other phylogroups were
203 represented by single isolates (Fig. 1). By serogroup, the 4 ST131 isolates were either O16:H4
204 (n=2: fumC40:fimH41) or O25:H4 (n=2: fumC40:fimH30) (Fig. 1). All ST1193 isolates were
205 O75:H5 with fumC14:fimH64. All ST12 isolates were O4:H1and had fumC13. Other STs
206 exhibited diverse serotypes and CH types (Fig. 1)

207 Based on the alignment of detected core genome SNPs, a phylogenetic tree was built for 17 NLF
208 E. coli from our study with the MG1655 genome as the reference genome. There was relatively
209 low diversity of the studied NLF E. coli as the majority of strains clustered together, resulting in
210 clades and subclades. The analysis clustered the 17 isolates into two major clades, one tight
211 cluster corresponding to phylogroup B2 and the other loosely clustered clade included strains
212 from A, B1, D and F phylogroup. Compared to NLF E. coli from stool, the urine E. coli isolates
213 formed a tight clade with strains belonging to ST12, ST1193 and ST131. Three strains from stool
214 origin clustered closely with urine strains reinforcing the gut as a possible reservoir of UTI-
215 causing agents. The blaCTX-M-15, MDR phenotype and β-hemolysis did not correspond with the
216 clusters indicating their widespread presence across E. coli lineages. However, the sequence
217 types, serogroups, and CH-types corresponded closely with the phylogenetic clusters of E. coli
218 genomes. The ST131 strains with O16:H5 (fumC40:fimH41) formed a subcluster with ST131
219 strains having O25:H4 (fumc40:fimH30), indicating there are sublineages within ST131.

220 We also analysed the pangenomes of the 17 NLF E. coli genomes (Fig. 2). The number of core
221 genes identified was 3400, and the accessory genes were 30563. Contrary to the core-genome-
222 based phylogenetic tree, no two genomes shared the same gene content with respect to the
223 pangenome. A closer look into the pangenome dendrogram reflects a clustering pattern that
224 mimics the clades of the core genome-based phylogenetic tree (Fig. 2). These observations

8|Page
225 indicate that the 17 NLF E. coli genomes do not differ drastically, even with regard to their
226 accessory gene content. They all have a few unique gene families, as shown in figure 2.

227 Virulence genes

228 Our analysis identified 73 virulence-associated genes out of 335 genes analysed. The distribution
229 of 73 virulence genes among 17 NLF E. coli isolates is shown in Fig. 3. Overall, 65% (11/17) of
230 all NLF E. coli isolates were assigned to different pathotypes, as classified by the presence of
231 specific virulence markers. The majority of urine NLF E. coli isolates were assigned to the
232 EXPEC pathotype [67% (6/9)]. Suspected diarrheagenic variants comprised 62% (5/8) of the
233 stool NLF E. coli isolates. The most prevalent pathotype among diarrheagenic variants was
234 enteroaggregative E. coli (EAEC) (38%) which harboured both aggR and aatA genes, followed
235 by enteropathogenic E. coli (EPEC) (13%) that showed the presence of intimin gene (eae). One
236 isolate from stool origin was also classified as ExPEC. The remaining six isolates (6/17) that did
237 not belong to any pathotype included three strains from ST1193, two from ST131 and one from
238 ST501. Out of 4 ST131 studied strains, two O25:H4 serogroup (fimH30) strains qualified as
239 ExPEC and two strains with O16:H5 serogroup (fimH41) did not qualify as ExPEC. Among all
240 the pathotypes, the seven (100%) ExpEC strains produced β-hemolysis in 5% sheep blood agar.
241 This hemolytic phenotype was strongly associated with alpha-hemolysin genes (hlyA, hlyB and
242 hlyD). Although the three ST1193 isolates did not qualify as ExpEC pathotypes, they all carried
243 a similar set of virulence genes, including adhesins (fimH), toxins (sat, hlyE/clyA), siderophores
244 (chuA, fyuA, sitABCD, iutA) and others. The NLF E. coli belonging to urine origin had a higher
245 prevalence of virulence genes as the aggregate virulence score [median (range)] for urine isolates
246 [90 (72-114)] was higher than that of stool NLF E. coli isolates [73 (62-116)].

247 Antimicrobial resistance

248 The phenotypic and genotypic trends of antimicrobial resistance of the studied NLF E.
249 coli strains are shown in Fig. 4. Overall, the NLF E. coli strains from the stool and urine samples
250 showed high antimicrobial drug resistance rates to nalidixic acid (100%), ceftazidime (94%),
251 cefradine (94%), followed by ciprofloxacin (88%), ampicillin (82%), cefotaxime (65%) and
252 cefepime (65%). The resistance rates to cefuroxime (59%), ceftriaxone (53%), co-trimoxazole
253 (41%), imipenem (35%) were moderate and the resistance rates to gentamicin (18%), amikacin
254 (12%), doxycycline (6%), meropenem (6%) were low. In contrast, the resistance rate to

9|Page
255 chloramphenicol, fosfomycin and tigecycline was 0. Compared to the stool isolates, the urine
256 NLF E. coli isolates had relatively low resistance to amikacin, cefepime, ceftriaxone and
257 cefuroxime; consequently, there were 63% (5/8) MDR isolates in the stool category and 33%
258 (3/9) MDR isolates in urine category.

259 We detected 31 unique AMR gene alleles belonging to different classes, with most genomes
260 exhibiting at least 6 AMR genes (Fig. 4). By clinical source, the AMR genes had a relatively
261 similar distribution between the stool and urine NLF E. coli isolates, with strains from both the
262 groups having a minimum of 6 AMR genes per genome, albeit substantial variations among
263 isolates (range 3-15 for stool E. coli and 2-13 for urine E. coli isolates). No specific AMR profile
264 was associated with isolates from a particular clinical source.

265 A variety of ESBL gene variants encoded cephalosporin resistance. Amongst the identified
266 ESBL genes in NLF E. coli isolates, blaCTX-M-15, was predominant [9/17 (53%)], but we also
267 identified blaOXA-1, blaTEM1B, blaDHA-1 and blaTEM-1C genes in a number of isolates (Fig. 4). We
268 observed 100% concordance between the presence of blaCTX-M-15 gene and resistance phenotypes
269 to the following cephalosporin antibiotics, cefradine, cefotaxime, ceftazidime, ceftriaxone,
270 cefuroxime and cefepime. One blaCTX-M-15 positive stool NLF E. coli harboured a blaNDM-1 gene;
271 this strain (A4) demonstrated difficult-to-treat resistance (DTR) as it was resistant to 14 of 18
272 antibiotics tested, including all first-line agents. It was sensitive to chloramphenicol,
273 doxycycline, fosfomycin and tigecycline. The probable genome locus of blaCTX-M-15 gene was
274 detected to be a plasmid for 4 out of 9 NLF E. coli isolates, as shown in Table 3. The genetic
275 environment of blaCTX-M-15 gene in a majority of isolates [56% (5/9)] consisted of the insertion
276 element ISEcp1preceeding the gene (Table 3). The blaCTX-M-15 gene was mainly associated with
277 the epidemiological significant clonal E. coli lineages such as ST131, ST167, ST1193 and
278 ST6303. Interestingly, these strains were all non-lactose fermenting (NLF).

279 We screened for mutations in the quinolone-resistance-determining regions (QRDRs) of gyrA,

280 parC and parE. We identified amino acid substitutions at codon positions S83I (16/17 isolates),
281 D87N (9/17 isolates) in the amino acid sequence of gyrA and S80I (9/17 isolates), S57T (one
282 isolate), E84V (4/17 isolates), E84G (two isolates) and S80R (two isolates) in parC. We also
283 identified mutations at codon positions S45A, I529L and L416F in parE gene. The gyrA
284 mutation S83L was strongly associated with resistance to ciprofloxacin. The ESBL gene blaCTX-M-

10 | P a g e
285 15 and gyrA S83L, parC E84V were strongly linked in ST131 and ST167 lineage strains.
286 Additionally, the isolated also harboured acquired AMR genes encoding fluoroquinolone (FQR)
287 resistance, including Qnrs1 (one isolate), Qnrs13 (one isolate) and aac(6')-Ib-cr gene (3
288 isolates). All the three acquired FQR genes were associated with ciprofloxacin resistance.

289 Multidrug resistance, defined by resistance to a minimum of one agent belonging to three
290 different antimicrobial classes, was frequently observed in isolates with multiple AMR genes
291 identified in 8 (47%) isolates. Cephalosporin resistance genes and mutation in gyrA S83L were
292 strongly associated with the MDR phenotype.

293 PlasmidFinder identified seven unique plasmid replicon groups, with each isolate carrying an
294 average of 2.5 plasmid groups (Fig.3). FIB was the most commonly encountered plasmid group
295 (n=13/17), followed by FII (n=9/17), CoI (n=8/12) and FIA (n=7/17). FIA, FIB and CoI
296 plasmids were strongly associated with ST131 E. coli. The NDM-positive E. coli strain (A4)
297 concomitantly hosted three IncF-type replicons, FIA, FIB, and FII groups. We could not
298 associate AMR genes with known plasmid replicons in two strains (A6 & U8) as no replicons
299 were detected in them. Surprisingly, these strains belong to well-known ESBL-producing
300 lineages (ST501 and ST131, respectively).

301

302 Discussion

303 We found a high prevalence of epidemiologically significant clonal lineages among NLF E. coli
304 isolates from patients with suspected UTIs and enteric infections in a referral diagnostic centre
305 (icddr,b) in Dhaka, Bangladesh. None of the previous studies has focused on the genomic
306 epidemiology of NLF E. coli. In the current study, we carried out a comprehensive
307 characterisation of 17 NLF E. coli isolates to delineate their microbiological, biochemical and
308 molecular characterisation using WGS to identify their population composition and detect their
309 molecular markers of AMR and virulence. This first report forms a baseline study on the
310 genomic epidemiology of clinical NLF E. coli isolates. Knowledge about such strains' virulence
311 and resistance properties is important to help direct researchers, clinicians and laboratory
312 technologists screen, analyse and report atypical variants of E. coli recovered from clinical
313 samples.

11 | P a g e
314 In line with previous reports (Hossain, 2012; Yaratha et al., 2017), our study revealed that the
315 prevalence rate of NLF E. coli in patients in the community is around 10%. This makes it
316 important to identify NLF E. coli in clinical specimens, which would help initiate appropriate
317 antimicrobial therapy. Given the difficulty in differentiating between NLF E. coli and Shigella
318 spp, multiple tests need to be performed as described in this study, including serology,
319 biochemical identification (manual and API) and molecular methods like WGS. It is reported
320 that routine MALDI-TOF MS analysis cannot reliably differentiate between E. coli and Shigella
321 species (Ling et al., 2019). It can be noted here that the chromogenic media- CHROMagar
322 Orientation cannot distinguish between lactose fermenting and non-lactose fermenting E. coli
323 and can be used to isolate/identify both types of E. coli in one go.

324 The predominant sequence types detected in this study were epidemiologically significant high-
325 risk clones, including ST131, ST1193, ST12, ST73 and ST167. Our group previously reported
326 these clones from Bangladesh among ESBL-producing lactose fermenting E. coli (Mazumder et
327 al., 2021a). None of the studies has reported them from NLF E. coli except for ST1193 (Wu et
328 al., 2017). The dominance of clonal sequence types among this collection of NLF E. coli is
329 interesting. It warrants close attention because the inclusion of isolates in this study was not
330 based on any antimicrobial resistance phenotypes and genotypes. These STs are predominantly
331 reported to be associated with extraintestinal infections (Pitout, 2012; Riley, 2014; Manges et al.,
332 2019). Also, a review article on global ExPEC STs enlisted these STs to be responsible for the
333 enormous burden of extraintestinal human infections globally (Manges et al., 2019). These
334 highly successful clonal groups are a major mode of spreading antimicrobial resistance by the
335 mechanism of global expansion (Tchesnokova et al.; Shaik et al., 2017).

336 ST131 E. coli is an established pandemic extraintestinal multidrug-resistant pathogen (Rogers et

337 al., 2011; Nicolas-Chanoine et al., 2014). The 4 ST131 isolates identified in our collection were
338 of two O and CH types, two ST131 E. coli exhibited O25:H4 with fumC40:fimH30 and the other
339 two ST131 E. coli had O16:H5 serogroup and fumC40:fimH41 CH type. The two strains with
340 O25: H4 serogroup belonged to the subclade C2, known as the H30Rx sublineage, which is the
341 highly drug-resistant and successful sublineage of the ST131 clone. ST1193 is reported to be an
342 emerging multiresistant clonal group (Johnson et al., 2019). The three ST1193 isolates in this
343 study carried the same set of three nonsynonymous chromosomal mutations in gyrA (S83L),

12 | P a g e
344 gyrA (D87N) and parC (S80I) that confer fluoroquinolone resistance and had O75:H5 serogroup
345 with fumC14: fimH64 CH type, they were all ciprofloxacin-resistant. ST1193 prevalence is
346 increasing rapidly worldwide and is expected to replace the most successful clone, ST131, in the
347 near future (Pitout et al., 2022). The three ST12 isolates in this collection were qualified as
348 ExPEC pathogens and were ciprofloxacin-resistant. ST12 is reported to be one of the
349 predominant STs that cause bloodstream infections (Kallonen et al., 2017; Manges et al., 2019).
350 The ST73 isolate detected in this study harboured 104 of 335 virulence genes, including several
351 markers of ExPEC and sepsis-associated E. coli (SEPEC). In contrast, it carried a smaller
352 number of AMR genes. It demonstrated β-hemolysis on 5% sheep blood agar and had the
353 CHtype-fumC24:fimH30. THE ST167 strain we identified showed a difficult-to-treat phenotype
354 and carried blaNDM-5 gene. It qualified as an EAEC pathotype and this strain has been associated
355 with a global spread of ESBL-E. coli in humans (Ewers et al., 2012). ST167 associated with
356 blaNDM-5 gene was reported as an emerging carbapenem-resistant high-risk clone (Garcia-
357 Fernandez et al., 2020).

358 Overall, the NLF E. coli isolates carried the AMR genes and were moderately resistant to
359 antibiotics, but a higher number were assigned to different pathotypes. The isolates carried
360 extensive virulence genes that belonged to different categories, including adhesion, invasion,
361 colonisation and iron uptake ability. Nevertheless, the combination of resistance and virulence in
362 these NLF E. coli isolates could contribute to their increased fitness and potential to expand
363 globally. This study has multiple limitations. First, limited clinical and epidemiological data
364 were available. Second, a small sample size constrained the power to generalise our findings and
365 confirm statistical associations, particularly with the high-risk clones. The main gain of the study
366 is that emphasis on inclusion was not laid on MDR or ESBL phenotypes; this has led to a clear
367 understanding of the important ExPEC lineages within NLF E. coli. The WGS approach helped
368 to provide a greater resolution to the study observations.

369 In conclusion, the prevalence of NLF E. coli variants isolated at a referral diagnostic centre
370 (icddr,b) in Dhaka, Bangladesh, was moderate (10%). Our study revealed multiresistant strains
371 and high-risk clonal groups, including ST131, ST1193, ST12, ST73 and ST167, emerging within
372 NLF E. coli isolates. To our knowledge, this is the first report on such a phenomenon in NLF E.
373 coli; most studies focus on pathotypes encompassing single STs. These high-risk clones may

13 | P a g e
374 further evolve by acquiring carbapenem and colistin resistance genes to cause difficult-to-treat
375 infections. Therefore, strengthening microbiology laboratories to detect and report NLF E. coli
376 isolates is important to accomplish successful patient treatment and to feed data to AMR
377 surveillance programs. Further national and regional multicentre One Health studies are required
378 to ascertain the significance of NLF E. coli as significant pathogens impacting public health.

379 Acknowledgments: This research study was funded by core donors which provide unrestricted
380 support to icddr,b for its operations and research. Current donors providing unrestricted support
381 include: Government of the People's Republic of Bangladesh; Global Affairs Canada (GAC);
382 Swedish International Development Cooperation Agency (Sida) and the Department for
383 International Development (UK Aid). We gratefully acknowledge these donors for their support
384 and commitment to icddr,b's research efforts.
385
386 Transparency declarations: None to declare
387

388 References
389 Baddam, R., Sarker, N., Ahmed, D., Mazumder, R., Abdullah, A., Morshed, R., et al. (2020).
390 Genome Dynamics of Vibrio cholerae Isolates Linked to Seasonal Outbreaks of Cholera in
391 Dhaka, Bangladesh. MBio 11. doi:10.1128/mBio.03339-19.
392 Bankevich, A., Nurk, S., Antipov, D., Gurevich, A. A., Dvorkin, M., Kulikov, A. S., et al.
393 (2012). SPAdes: A New Genome Assembly Algorithm and Its Applications to Single-Cell
394 Sequencing. J. Comput. Biol. 19, 455–477. doi:10.1089/cmb.2012.0021.
395 Beghain, J., Bridier-Nahmias, A., Le Nagard, H., Denamur, E., and Clermont, O. (2018).
396 ClermonTyping: an easy-to-use and accurate in silico method for Escherichia genus strain
397 phylotyping. Microb. Genomics 4. doi:10.1099/mgen.0.000192.
398 Bertels, F., Silander, O. K., Pachkov, M., Rainey, P. B., and Van Nimwegen, E. (2014).
399 Automated Reconstruction of Whole-Genome Phylogenies from Short-Sequence Reads.
400 Mol. Biol. Evol. 31, 1077–1088. doi:10.1093/MOLBEV/MSU088.
401 Bortolaia, V., Kaas, R. S., Ruppe, E., Roberts, M. C., Schwarz, S., Cattoir, V., et al. (2020).
402 ResFinder 4.0 for predictions of phenotypes from genotypes. J. Antimicrob. Chemother.
403 doi:10.1093/jac/dkaa345.
404 Carattoli, A., Zankari, E., García-Fernández, A., Voldby Larsen, M., Lund, O., Villa, L., et al.
405 (2014). In Silico Detection and Typing of Plasmids using PlasmidFinder and Plasmid
406 Multilocus Sequence Typing. Antimicrob. Agents Chemother. 58, 3895–3903.
407 doi:10.1128/AAC.02412-14.

14 | P a g e
408 Chakraborty, A., Adhikari, P., Shenoy, S., and Saralaya, V. (2016). Virulence factor profiles,
409 phylogenetic background, and antimicrobial resistance pattern of lactose fermenting and
410 nonlactose fermenting Escherichia coli from extraintestinal sources. Indian J. Pathol.
411 Microbiol. 59, 180. doi:10.4103/0377-4929.182032.
412 Chang, J., Yu, J., Lee, H., Ryu, H., Park, K., and Park, Y. J. (2014). Prevalence and
413 characteristics of lactose non-fermenting Escherichia coli in urinary isolates. J. Infect.
414 Chemother. 20, 738–740. doi:10.1016/J.JIAC.2014.07.005.
415 Chen, L., Yang, J., Yu, J., Yao, Z., Sun, L., Shen, Y., et al. (2005). VFDB: A reference database
416 for bacterial virulence factors. Nucleic Acids Res. doi:10.1093/nar/gki008.
417 Chen, S., Zhou, Y., Chen, Y., and Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor.
418 doi:10.1093/bioinformatics/bty560.
419 Clausen, P. T. L. C., Aarestrup, F. M., and Lund, O. (2018). Rapid and precise alignment of raw
420 reads against redundant databases with KMA. BMC Bioinformatics 19, 1–8.
421 doi:10.1186/S12859-018-2336-6/TABLES/2.
422 Colonna, B., Ranucci, L., Fradiani, P. A., Casalino, M., Calconi, A., and Nicoletti, M. (1992).
423 Organization of aerobactin, hemolysin, and antibacterial resistance genes in lactose-
424 negative Escherichia coli strains of serotype O4 isolated from children with diarrhea. Infect.
425 Immun. 60, 5224–5231. doi:10.1128/IAI.60.12.5224-5231.1992.
426 Eren, A. M., Esen, O. C., Quince, C., Vineis, J. H., Morrison, H. G., Sogin, M. L., et al. (2015).
427 Anvi’o: An advanced analysis and visualization platformfor ’omics data. PeerJ 2015,
428 e1319. doi:10.7717/PEERJ.1319/SUPP-5.
429 Ewers, C., Bethe, A., Semmler, T., Guenther, S., and Wieler, L. H. (2012). Extended-spectrum β-
430 lactamase-producing and AmpC-producing Escherichia coli from livestock and companion
431 animals, and their putative impact on public health: a global perspective. Clin. Microbiol.
432 Infect. 18, 646–655. doi:10.1111/J.1469-0691.2012.03850.X.
433 Flores-Mireles, A. L., Walker, J. N., Caparon, M., and Hultgren, S. J. (2015). Urinary tract
434 infections: epidemiology, mechanisms of infection and treatment options. Nat. Rev.
435 Microbiol. 2015 135 13, 269–284. doi:10.1038/nrmicro3432.
436 Garcia-Fernandez, A., Villa, L., Bibbolino, G., Bressan, A., Trancassini, M., Pietropaolo, V., et
437 al. (2020). Novel Insights and Features of the NDM-5-Producing Escherichia coli Sequence
438 Type 167 High-Risk Clone. mSphere 5.
439 doi:10.1128/MSPHERE.00269-20/ASSET/6E7B35EC-775E-40A6-B00C-
440 FDACD968538C/ASSETS/GRAPHIC/MSPHERE.00269-20-F0003.JPEG.
441 Guindon, S., Dufayard, J. F., Lefort, V., Anisimova, M., Hordijk, W., and Gascuel, O. (2010).
442 New Algorithms and Methods to Estimate Maximum-Likelihood Phylogenies: Assessing
443 the Performance of PhyML 3.0. Syst. Biol. 59, 307–321. doi:10.1093/SYSBIO/SYQ010.
444 Gurevich, A., Saveliev, V., Vyahhi, N., and Tesler, G. (2013). QUAST: quality assessment tool
445 for genome assemblies. Bioinformatics 29, 1072–1075.
446 doi:10.1093/BIOINFORMATICS/BTT086.
447 Hasman, H., Saputra, D., Sicheritz-Ponten, T., Lund, O., Svendsen, C. A., Frimodt-Moller, N., et
15 | P a g e
448 al. (2014). Rapid whole-genome sequencing for detection and characterization of
449 microorganisms directly from clinical samples. J. Clin. Microbiol. 52, 139–146.
450 doi:10.1128/JCM.02452-13.
451 Hossain, A. (2012). Presence and pattern of virulence genes in non-lactose fermenting
452 Escherichia coli strains isolated from stools of children <5 years in rural and urban
453 Bangladesh. Int. J. Infect. Dis. 16, e395. doi:10.1016/J.IJID.2012.05.525.
454 Hussain, A., Ranjan, A., Nandanwar, N., Babbar, A., Jadhav, S., and Ahmed, N. (2014).
455 Genotypic and phenotypic profiles of Escherichia coli isolates belonging to clinical
456 sequence type 131 (ST131), clinical non-ST131, and fecal non-ST131 lineages from India.
457 Antimicrob. Agents Chemother. 58, 7240–9. doi:10.1128/AAC.03320-14.
458 Joensen, K. G., Tetzschner, A. M. M., Iguchi, A., Aarestrup, F. M., and Scheutz, F. (2015).
459 Rapid and easy in silico serotyping of Escherichia coli isolates by use of whole-genome
460 sequencing data. J. Clin. Microbiol. doi:10.1128/JCM.00008-15.
461 Johansson, M. H. K., Bortolaia, V., Tansirichaiya, S., Aarestrup, F. M., Roberts, A. P., and
462 Petersen, T. N. (2021). Detection of mobile genetic elements associated with antibiotic
463 resistance in Salmonella enterica using a newly developed web tool: MobileElementFinder.
464 J. Antimicrob. Chemother. 76, 101–109. doi:10.1093/JAC/DKAA390.
465 Johnson, J. R., and Stell, A. L. (2000). Extended virulence genotypes of Escherichia coli strains
466 from patients with urosepsis in relation to phylogeny and host compromise. J. Infect. Dis.
467 doi:10.1086/315217.
468 Johnson, T. J., Elnekave, E., Miller, E. A., Munoz-Aguayo, J., Figueroa, C. F., Johnston, B., et
469 al. (2019). Phylogenomic Analysis of Extraintestinal Pathogenic Escherichia coli Sequence
470 Type 1193, an Emerging Multidrug-Resistant Clonal Group. Antimicrob. Agents
471 Chemother. 63. doi:10.1128/AAC.01913-18.
472 Kallonen, T., Brodrick, H. J., Harris, S. R., Corander, J., Brown, N. M., Martin, V., et al. (2017).
473 Systematic longitudinal survey of invasive Escherichia coli in England demonstrates a
474 stable population structure only transiently disturbed by the emergence of ST131. Genome
475 Res. 27, 1437–1449. doi:10.1101/GR.216606.116/-/DC1.
476 Kaper, J. B., Nataro, J. P., and Mobley, H. L. T. (2004). Pathogenic Escherichia coli. Nat. Rev.
477 Microbiol. 2, 123–140. doi:10.1038/nrmicro818.
478 Larsen, M. V., Cosentino, S., Lukjancenko, O., Saputra, D., Rasmussen, S., Hasman, H., et al.
479 (2014). Benchmarking of methods for genomic taxonomy. J. Clin. Microbiol. 52, 1529–
480 1539. doi:10.1128/JCM.02981-13/SUPPL_FILE/ZJM999093347SO4.PDF.
481 Larsen, M. V., Cosentino, S., Rasmussen, S., Friis, C., Hasman, H., Marvig, R. L., et al. (2012).
482 Multilocus sequence typing of total-genome-sequenced bacteria. J. Clin. Microbiol.
483 doi:10.1128/JCM.06094-11.
484 Letunic, I., and Bork, P. (2016). Interactive tree of life (iTOL) v3: an online tool for the display
485 and annotation of phylogenetic and other trees. Nucleic Acids Res. 44, W242–W245.
486 doi:10.1093/nar/gkw290.
487 Ling, J., Wang, H., Li, G., Feng, Z., Song, Y., Wang, P., et al. (2019). A novel short-term high-
16 | P a g e
488 lactose culture approach combined with a matrix-assisted laser desorption ionization-time of
489 flight mass spectrometry assay for differentiating Escherichia coli and Shigella species
490 using artificial neural networks. PLoS One 14, e0222636.
491 doi:10.1371/JOURNAL.PONE.0222636.
492 Logan, L. K., and Weinstein, R. A. (2017). The Epidemiology of Carbapenem-Resistant
493 Enterobacteriaceae: The Impact and Evolution of a Global Menace. 215.
494 doi:10.1093/infdis/jiw282.
495 Manges, A. R., Geum, H. M., Guo, A., Edens, T. J., Fibke, C. D., and Pitout, J. D. D. (2019).
496 Global extraintestinal pathogenic escherichia coli (Expec) lineages. Clin. Microbiol. Rev.
497 doi:10.1128/CMR.00135-18.
498 Mazumder, R., Abdullah, A., Ahmed, D., and Hussain, A. (2020). High prevalence of blactx-m-
499 15 gene among extended-spectrum β-lactamase-producing escherichia coli isolates causing
500 extraintestinal infections in bangladesh. Antibiotics. doi:10.3390/antibiotics9110796.
501 Mazumder, R., Hussain, A., Abdullah, A., Islam, M. N., Sadique, M. T., Muniruzzaman, S. M.,
502 et al. (2021a). International High-Risk Clones Among Extended-Spectrum β-Lactamase-
503 Producing Escherichia coli in Dhaka, Bangladesh. Front. Microbiol. 12.
504 doi:10.3389/FMICB.2021.736464.
505 Mazumder, R., Hussain, A., Abdullah, A., Islam, M. N., Sadique, M. T., Muniruzzaman, S. M.,
506 et al. (2021b). International High-Risk Clones Among Extended-Spectrum β-Lactamase–
507 Producing Escherichia coli in Dhaka, Bangladesh. Front. Microbiol. 12, 2843.
508 doi:10.3389/FMICB.2021.736464/BIBTEX.
509 Nataro, J. P., and Kaper, J. B. (1998). Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11,
510 142–201. doi:10.1128/CMR.11.1.142/FORMAT/EPUB.
511 nia Santos Braz, V., Melchior, K., and Gallina Moreira, C. Escherichia coli as a Multifaceted
512 Pathogenic and Versatile Bacterium. doi:10.3389/fcimb.2020.548492.
513 Nicolas-Chanoine, M.-H., Bertrand, X., and Madec, J.-Y. (2014). Escherichia coli ST131, an
514 intriguing clonal group. Clin. Microbiol. Rev. 27, 543–74. doi:10.1128/CMR.00125-13.
515 Nicoletti, M., Superti, F., Conti, C., Calconi, A., and Zagaglia, C. (1988). Virulence factors of
516 lactose-negative Escherichia coli strains isolated from children with diarrhea in Somalia. J.
517 Clin. Microbiol. 26, 524–529. doi:10.1128/JCM.26.3.524-529.1988.
518 Pitout, J. D. D. (2012). Extraintestinal Pathogenic Escherichia coli: A Combination of Virulence
519 with Antibiotic Resistance. Front. Microbiol. 3, 9. doi:10.3389/fmicb.2012.00009.
520 Pitout, J. D. D., Peirano, G., Chen, L., DeVinney, R., and Matsumura, Y. (2022). Escherichia
521 coli ST1193: Following in the Footsteps of E. coli ST131. Antimicrob. Agents Chemother.
522 doi:10.1128/AAC.00511-22.
523 Riley, L. W. (2014). Pandemic lineages of extraintestinal pathogenic Escherichia coli. Clin.
524 Microbiol. Infect. 20, 380–390. doi:10.1111/1469-0691.12646.
525 Roer, L., Johannesen, T. B., Hansen, F., Stegger, M., Tchesnokova, V., Sokurenko, E., et al.
526 (2018). CHTyper, a Web Tool for Subtyping of Extraintestinal Pathogenic Escherichia coli

17 | P a g e
527 Based on the fumC and fimH Alleles. J. Clin. Microbiol. 56, 63–81.
528 doi:10.1128/JCM.00063-18.
529 Rogers, B. A., Sidjabat, H. E., and Paterson, D. L. (2011). Escherichia coli O25b-ST131: a
530 pandemic, multiresistant, community-associated strain. J. Antimicrob. Chemother. 66, 1–14.
531 doi:10.1093/JAC/DKQ415.
532 Shaik, S., Ranjan, A., Tiwari, S. K., Hussain, A., Nandanwar, N., Kumar, N., et al. (2017).
533 Comparative Genomic Analysis of Globally Dominant ST131 Clone with Other
534 Epidemiologically Successful Extraintestinal Pathogenic Escherichia coli (ExPEC)
535 Lineages. MBio 8, e01596-17. doi:10.1128/mBio.01596-17.
536 Siqueira, F. M., De Carli, S., Lopes, C. E., Machado, L., Vieira, T. R., Pöppl, G., et al. (2021).
537 Non-lactose-fermenting uropathogenic Escherichia coli from dogs: virulence profile
538 characterization. Lett. Appl. Microbiol. 72, 596–603. doi:10.1111/LAM.13454.
539 Tatusova, T., Dicuccio, M., Badretdin, A., Chetvernin, V., Nawrocki, E. P., Zaslavsky, L., et al.
540 (2016). NCBI prokaryotic genome annotation pipeline. Nucleic Acids Res. 44, 6614–6624.
541 doi:10.1093/NAR/GKW569.
542 Tchesnokova, V., Radey, M., Chattopadhyay, S., Larson, L., Lee Weaver, J., Kisiela, D., et al.
543 Pandemic fluoroquinolone resistant Escherichia coli clone ST1193 emerged via
544 simultaneous homologous recombinations in 11 gene loci. doi:10.1073/pnas.1903002116.
545 Wu, J., Lan, F., Lu, Y., He, Q., and Li, B. (2017). Molecular characteristics of ST1193 clone
546 among phylogenetic group B2 Non-ST131 fluoroquinolone-resistant Escherichia coli.
547 Front. Microbiol. 8, 2294. doi:10.3389/FMICB.2017.02294/BIBTEX.
548 Yaratha, G., Perloff, S., and Changala, K. (2017). Lactose vs Non-Lactose Fermenting E. coli:
549 Epidemiology, Clinical Outcomes, and Resistance. Open Forum Infect. Dis. 4, S589.
550 doi:10.1093/OFID/OFX163.1546.
551 Zankari, E., Allesøe, R., Joensen, K. G., Cavaco, L. M., Lund, O., and Aarestrup, F. M. (2017).
552 PointFinder: A novel web tool for WGS-based detection of antimicrobial resistance
553 associated with chromosomal point mutations in bacterial pathogens. J. Antimicrob.
554 Chemother. doi:10.1093/jac/dkx217.
555

556

557

558

559

560

561

562

18 | P a g e
563

564

565

566 Figure Legends

567 Figure 1. Phylogenetic relationships amongst sequenced NLF E. coli genomes. The
568 maximum likelihood phylogenetic tree is based on the alignment of detected core genome SNPs
569 of 17 NLF E. coli genomes with the MG1655 genome strain as a reference genome. The
570 phylogroups, clinical source, blaCTX-M-15 status, beta-hemolysis, MDR status, STs, serogroups, CH
571 types and blaNDM-5 status are shown to the right of the tree.
572

573 Figure 2.
574 Pangenome analysis of 17 NLF E. coli genomes from Anvi'o pangenomics suite. The circles
575 represent individual genomes arranged as per their pangenomic phylogenetic relationship
576 depicted by the outside dendrogram. In the circles, dark colour indicates the presence of a gene
577 group and light colour its absence.
578

579 Figure 3.
580 Heat map depicting the distribution of 73 virulence genes among 17 NLF E. coli genomes. Dark
581 brown represents the presence, and light brown blocks represent a virulence gene's absence.
582

583 Figure 4.
584 Heat map demonstrating the AST profile, plasmid replicon types and acquired AMR gene profile
585 of 17 NLF E. coli isolates/genomes. Dark green represents resistance/ presence and light green
586 blocks represent the sensitivity/ absence of a particular trait. Definition of abbreviated antibiotics
587 for AST: AMK, amikacin; Amp, ampicillin; FEP, cefepime; CTX, cefotaxime; CAZ,
588 ceftazidime; RAD, cephradine; CRO, ceftriaxone; CXM, cefuroxime; CHL, chloramphenicol;
589 CIP, ciprofloxacin; SXT, trimethoprim-sulfamethoxazole; DOX, doxycycline; FOF, fosfomycin;
590 GEN, gentamicin; IPM, imipenem; MEM, meropenem; NAL, nalidixic acid; TGC, tigecycline.
591

592

593

594

19 | P a g e
595

596

597

598 Table 1: Microbiological and biochemical characteristics of 17 NLF E. coli isolates

Sl. No. (%)

Biochemical tests
No.
Positive Negative
1 Catalase test 17 (100%) 0
2 Oxidase test 0 17 (100%)
3 TSI agar
a. Acid production in slant 17 (100%) 0
b. Acid production in butt 17 (100%) 0
c. Hydrogen sulfide production(H2S) 0 17 (100%)
d. Gas production 17 (100%) 0
4 Motility Indole Urea's Test (MIU)
a. Motility 17 (100%) 0
b. Indole Production 17 (100%) 0
c. Urea hydrolysis 0 17 (100%)
5 Simmons Citrate reaction test 0 17 (100%)
6 Acetate 17 (100%) 0
7 Sugar fermentation
a. Glucose 17 (100%) 0
b. Lactose 0 17 (100%)
c. Sucrose 11 (65%) 6(35%)
e. Mannose 17 (100%) 0
f. Arabinose 17 (100%) 0
g. Sorbitol 17 (100%) 0
h. Mannitol 17 (100%) 0
i. Inositol 0 17 (100%)
Ortho-Nitrophenyl-β-galactoside
8 (ONPG) 17 (100%) 0
9 Gelatin liquefaction 0 17 (100%)
10 Vogas-proskauer 0 17 (100%)
11 Lysine decarboxylase 17 (100%) 0
12 Ornithine decarboxylase 14(82%) 3 (18%)
13 Arginine Dihydrolase 11 (65%) 6(35%)
Microbiological Characteristics [17 (100%)]
Colorless and transparent (NLF
Growth on MacConkey Agar
1 colony)
Colorless and transparent (NLF
Growth on SS agar
2 colony)
3 Growth on CHROMagar™ Orientation Pink color colony
4 Growth Temperature (26-42°C)
5 Gram staining Negative rods
599

20 | P a g e
600

601

602 Table 2: Genomic features of the 17 whole-genome sequenced NLF E. coli isolates
Geno Contig
Isol Isolati
Sl Strain me No. Genome No. GC
ate Source on Accession No.
No. ID covera (>500 size (bp) CDS %
ID year
ge bp)
BDEc_N JAMKCS00000
1 A1 LF-A1 Stool 2019 91 106 5,305,815 5077 50.5 0000
BDEc_N JAMKCT00000
2 A3 LF-A3 Stool 2019 101 103 5,480,336 5335 50.6 0000
BDEc_N JAMKCU00000
3 A4 LF-A4 Stool 2019 96 96 5,019,286 4772 50.7 0000
BDEc_N JAMKCV00000
4 A5 LF-A5 Stool 2019 63 77 5,074,445 4830 50.5 0000
BDEc_N JAMKCW0000
5 A6 LF-A6 Stool 2019 121 84 4,890,228 4656 50.6 00000
BDEc_N JAMKCX00000
6 A7 LF-A7 Stool 2020 90 70 4,978,296 4710 50.7 0000
BDEc_N JAMKCY00000
7 A8 LF-A8 Stool 2020 124 151 5,033,902 4934 50.6 0000
BDEc_N JAMKCZ00000
8 A9 LF-A9 Stool 2020 91 43 5,068,010 4850 50.6 0000
BDEc_N JAMKDA00000
9 U-1 LF-U1 Urine 2019 71 44 5,073,930 4893 50.6 0000
BDEc_N JAMKDF00000
10 U-4 LF-U4 Urine 2019 62 53 5,251,806 4988 50.4 0000
BDEc_N JAMKDG00000
11 U-6 LF-U6 Urine 2019 79 44 5,054,491 4847 50.6 0000
BDEc_N JAMKDH00000
12 U-7 LF-U7 Urine 2019 88 109 5,302,191 5061 50.5 0000
BDEc_N JAMKDI00000
13 U-8 LF-U8 Urine 2020 71 56 5,186,272 5002 50.6 0000
U- BDEc_N JAMKDB00000
14 12 LF-U12 Urine 2020 83 57 5,021,789 4793 50.7 0000
U- BDEc_N JAMKDC00000
15 13 LF-U13 Urine 2020 84 69 5,262,816 5008 50.5 0000
U- BDEc_N JAMKDD00000
16 14 LF-U14 Urine 2020 70 145 5,304,507 5153 50.7 0000
U- BDEc_N JAMKDE00000
17 15 LF-U15 Urine 2020 75 79 5,417,595 5190 50.36 0000
603
604
605
606
607
608

21 | P a g e
609
610
611 Table 3: Characteristics of CTX-M-15 associated NLF E. coli isolates

Isolate Genome
ID Source ST Phylogroup locus MGEsa Plasmid replicon
A1 Stool 6303 D Plasmid ISEcp1 IncFIB, IncFII
A3 Stool 131 B2 Plasmid NDb Col, IncFIA, IncFIB, IncFII, IncY
A4 Stool 167 A Chromosome ISEcp1 IncFIA, IncFIB, IncFII
A5 Stool 501 D Chromosome ISEcp1 IncFIB, IncFII
A7 Stool 131 B2 Chromosome NDb Col, IncFIA, IncFIB, IncFII
A8 Stool 2089 A Plasmid ISKpn19 IncFIB
A9 Stool 1193 B2 Chromosome ISEcp1 Col, IncFIA
U-8 Urine 131 B2 Plasmid IS3 NDb
U-12 Urine 131 B2 Chromosome ISEcp1 Col, IncFIA, IncFIB, IncL
612
613 a
Mobile genetic elements
614 b
Not detected
615

22 | P a g e