Study design, site and source of bacteria isolates
This was a cross sectional-laboratory-based study conducted at the Microbiology Laboratory and Molecular Biology Laboratory, College of Veterinary Medicine Animal Resources and Biosecurity (CoVAB) Makerere University. The study involved use of archived MDR Escherichia coli samples isolated between January and December, 2019 from clinical specimens in the Microbiology Laboratories of Mulago National Referral Hospital (MNRH), Mbale Regional Referral Hospital (MRRH), Mbarara Regional Referral Hospital (MBRRH) and Kampala International University Teaching Hospital (KIU-TH). The samples were transported in peptone water to the Microbiology Laboratory, CoVAB. Overnight cultures of E. coli were prepared by pipetting 1ml of peptone water containing each isolate into 49ml of Luria-Bertani (LB) broth. Glycerol stocks of each different isolate were made by adding 500μl of the overnight LB culture to 500μL of 50% glycerol in a 2ml screw top tube and mixed gently mix. The screw tubes were stored at -800C until further use.
Biochemical assays to confirm the identity of E. coli
To confirm the identity of each isolate, Microgen (Micro-biology International) kits for biochemical assays were employed using procedures described by the manufacturer (www.microgenbioproducts.com).
Screening for carbapenem susceptibility
This was achieved using the Kirby Bauer Disk Diffusion method and the results obtained were interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines . Ampicillin (AMP) 25 𝜇g, Amoxicillin/clavulanic acid (AMO) 20/10 𝜇g, trimethoprim-sulfamethoxazole (TMP/SMX) 1.25/23.75 𝜇g, Ciprofloxacin (CIP) 5 𝜇g, Cefuroxime (CXM) 30 μg, Temocillin (TEM) 30 μg, Piperacillin-tazobactum (TPZ) 110 μg, Cefoxitin (FOX) 30 μg, Cefipime (FEP) 30μg, Ceftriaxone (CRO) 30 μg, Ceftazidime (CAZ) 30 μg, Cefotaxime (CTX) 30 μg, Ertapenem (ERT) 10 µg, Meropenem (MEM) 10 μg and Imipenem (IMI)10 μg (Oxoid United Kingdom) carbapenem antibiotics disks were used. The turbidity of overnight Escherichia coli broth was adjusted using peptone water to a standard uniform concentration of 0.5 McFarland. Each E. coli isolate was inoculated on Mueller Hinton agar (Oxoid, United Kingdom) plates. Three antibiotic discs were placed about 2.0 cm apart and from the edge of plates, then incubated at 37°C for 24 hours. The diameter zones of growth inhibition were scored in millimeters. For quality control, E. coli ATCC 25922 was used as a susceptible strain and Klebsiella pneumoniae ATCC BAA-1705 as a positive control.
Pure colonies of E. coli from different samples were selected and each sub-cultured in 5 ml of Luria-Bertani broth using sterile inoculating loop. The bacterial suspension was incubated in shaker incubator at 37ºC for 24hrs. Then, 1ml of bacterial suspension was transferred into a 1.5 ml eppendorf tube, centrifuged at 10,000 rpm for 10 minutes. The supernatant was discarded and the pellet was re-suspended in 200μl of Gram-negative bacteria lysis buffer provided in the Qiagen DNA extraction. Bacterial total genomic DNA was extracted following the Qiagen DNA extraction protocol and stored at -20oC until further use.
Molecular characterization of virulent genes and carbapenem resistance determinants
Molecular identification of carbapenem resistance determinants and virulent genes in Escherichia coli was carried out using multiplex PCR. Primers used for molecular characterization were obtained from Eurofins Genomics AT GmbH and PCR amplification was performed in a Bio-Rad PTC-200 Thermal Cycler (Bio-Rad, Hercules, CA, USA)
Multiplex PCR amplification of carbapenem resistance genes
The existence of carbapenem resistance genetic determinants was determined using primers targeting b𝑙𝑎VIM, 𝑏𝑙𝑎IMP, 𝑏𝑙𝑎KPC, 𝑏𝑙𝑎OXA-48, and 𝑏𝑙𝑎NDM that carbapenemase encoding genes, Table 1. For co-amplification of target genes, multiplex PCR was conducted by adapting methods used by Dallenne et al., . Briefly, 2.5 µl of template DNA (100 ng/ µl) was added to 47.5 µl PCR mix containing 200 µM dNTPs (Biomatik, USA), 0.5 µM of each primer pair and 1X PCR Buffer (1.5 mM MgCl2, 10 mM Tris–HCl, pH 8.3/50 mM KCl) (Biomatik USA) and 1.2 μl of 1U Taq DNA Polymerase. Amplification was performed as follows; preliminary denaturation at 95°C for 5 minutes; then denaturation at 95°C for 30 seconds; annealing at 56°C for 30 seconds and elongation at 72°C for 1 minute; and a final elongation at 72°C for 10 minutes. For quality assurance positive and negative control isolates were obtained as a kind donation from the Microbiology Laboratory, College of Health Science, Makerere University. Antibiotics susceptible DSMZ 9377 Klebsiella pneumoniae was used as a negative control for all genes. Klebsiella pneumonia Nr.8 for NDM-1, Klebsiella pneumoniae 714 for OXA-48, Klebsiella pneumoniae 211 (T) for KPC, P. aeruginosa for IMP (Positive control strains from the Institute of Microbiology, Giessen, Germany) and E. coli for the VIM gene, obtained from RESET research collaboration  were used as positive controls.
Multiplex PCR components and conditions for E. coli pathotyping
Virulent genes eae for EPEC; stx for STEC/EHEC; est for TS-ETEC; elt for TL-ETEC; aggR for EAEC; ipaH for EIEC were amplified by multiplex PCR to characterize the different pathogenic bacteria using primers outlined in Table 2 . Five E. coli strains INCQS 00181 (CDC 055 – EPEC), INCQS 00171 (CDC EDL – 933 – EHEC) and INCQS 00170 (CDC EDL – 1284 – EIEC) from Centre for disease control and prevention belonging to the five categories of pathogenic E. coli were used as control . Multiplex PCR reaction was performed using  modified method to enable the concurrent amplification of all target genes. A final PCR volume of 50μl containing 5μl of 100ng DNA sample, 25μl of 1X PCR Buffer mixed with MgCl2 (1.5 mM), 1.2 μl of 1U Taq DNA Polymerase and dNTPs (200 µM) plus 0.5 µM each primer pair for DEC pathotypes. Sterile distilled deionized water was used to top up to 50μl. The PCR was performed under the following condition; An initial denaturation at 95°C for 5 minutes then 30 amplification cycles at 95°C for 30 second, 50 °C for 30 second, 72 °C for 1 minute, and a final extension at 72 °C for 30 minute
PCR amplification of PAI Markers
Seven different PAI markers designated as PAI I536, II536, IV536, ICFT073, IICFT073, IJ96 and IIJ96 have been previously characterized in UPEC , Table 3. Thus, the multiplex PCR used in the detection of PAI Markers, contained 2.5 µl of template DNA (100 ng/µl), 1U Taq DNA polymerase (Biomatik) in 1X PCR buffer (Biomatik), 200 µM of each dNTP, 2.5mM MgCl2, and 0.5𝜇M of each primer, Table 3. The program consisted of initial denaturation at 94°C for 5min, followed by 30 cycles of 94°C for 1 minute, 55°C for 1 minute and 72°C for 1minute, with a final extension step at 72°C for 10 minutes . The positive control used in the PCR was J96 O4:K6.
Phylogenetic classification exhibited that the E. coli strains belonged to four groups (A, B1, B2, or D) based on the presence of the chuA and yjaA genes and the DNA fragment (TSPE4.C2). Thus, a multiplex PCR was run to determine the phylogenetic classes of the E. coli strains using primers targeting chuA, yjaA and TSPE4.C2 DNA sequences, Table 4. The PCR amplification was conducted by adapting  methods. Briefly, the PCR contained 2.5 µl of template DNA, 1U Taq DNA polymerase (Biomatik, USA) in 1x PCR buffer (Biomatik), 200 µM dNTP, 2.5mM MgCl2, and 0.8 𝜇M of each primer, Table 1. Amplification was conducted using the following PCR conditions; initial denaturation at 94°C for 5 minutes, then 30 cycles performed at 94°C for 5 seconds, 54°C for 10 seconds, 72°C for 30 second with a final extension step at 72°C for 5 minutes. Phylogenic groups and subgroups were assigned depending on chuA, yjaA, and TspE4.C2 gene combinations [13, 14], Table 5.
Data analysis was done using the SPSS version 25 (SPSS Inc., Chicago, IL). Statistical differences were computed by chi-square and Spearman’s correlation. A p value < 0.05 indicated substantial statistical difference.