Study design and participants
Study participants were recruited from 8 LF endemic communities (Busia, Butre, Achowa, Princess Town, Akatakyi, Ampatano, Dixcove and Asemkow) in the Ahanta West Municipal. Ahanta West Municipal remains one of the hotspots in Ghana with a microfilariae prevalence of 2.2% and has been part of clinical trials to test doxycycline (46,47) . This study followed all protocols regarding human participants and ethical clearance was given by the Committee of Human Research and Publications and Ethics, School of Medicinal Science, KNUST (CHRPE/AP/649/19). Participants who consented to the study included individuals with different stages of LF, who had lived in the endemic community for ten (10) years or more and were between the ages of eighteen (18) and sixty (60) years with wounds. These potential participants were enlisted by health extension workers prior to the study. Participants were given informed consent, which they consented to either by signing or thumb printing and countersigned by an independent witness.
Data on antibiotic usage
To determine the possible impact of antibiotic usage on antimicrobial resistance in LF patients, a simple structured questionnaire was used to get their response as to whether they had used antibiotics within the last three months for treatment and, if any, to name the antibiotic.
Collection of samples
Swabs were taken from LF wounds patients using Becton Dickson and Company (BBL) culture swab. Swabs were stored between 2 - 8oC on the field, transferred to a liquid nitrogen tank at the end of the day’s collection and transported to Kumasi Centre for Collaborative Research (KCCR) for culture and isolation of bacteria. The culture and isolation have been described previously (31) .
Tests for identification of isolated organisms
Biochemical identification of MRSA
To identify S. aureus, coagulase test and catalase test were done on the gram-positive isolates while observing for β-hemolysis on blood agar (BA). Organisms were cultured on BA or Mueller-Hinton agar (MHA) and incubated at 35+2oC for 18+2 h. In determining catalase production in the isolates, colonies of bacteria from a fresh culture were picked from Mueller-Hinton agar and placed in droplets of catalase reagent (3% H2O2) and observed for bubble formation (48). Bubble formation was indicative of a catalase positive isolate while no bubble formation indicated a negative test. The coagulase test was used to differentiate between S. aureus and coagulase-negative staphylococci. The coagulase test was performed by using fresh colonies from a culture. This was placed in a rabbit serum in a tube and incubated for 4 h, where the presence of clumping indicated a positive test, while no clumping indicated a negative test (49) . β-hemolysis was recorded as positive if there were formation of clear zones around the S. aureus isolates growth on BA that had been incubated at 35+2oC for 18+2 h (49) .
Biochemical identification of P. aeruginosa & E. coli
Gram negative organisms were subjected to oxidase, indole, citrate and triple sugar iron (TSI) tests to aid in the identification and characterization of the phenotypes of P. aeruginosa and E. coli isolates. Bacteria isolates were cultured on MacConkey agar (MAC) or MHA and incubated at 35+2oC for 18+2 h before all the tests were done. The oxidase test was performed to identify isolates that catalyze oxidase-reduction via the cytochrome C oxidase enzyme. To perform the oxidase test BBL Dryslide and Remel oxidase were used. Isolates grown on MHA were streaked on BBL Dryslide. For a positive test a bluish colouration was seen after 20 sec, while no colour change indicated negative test (50). This was confirmed by using the Remel oxidase reagent (51). Indole production was used to differentiate Enterobacteria that degrade tryptophanase. Indole production test was done using 4-5 colonies of isolate grown on non-selective media (MHA) placed on a sterile swab stick (52). The indole Remel reagent was then poured on the swab and within a period of 120 sec, bluish colouration of the colonies indicated indole production (positive test), while pink colouration indicated no indole production (negative test). In determining bacteria isolates that possess citrate permease, the citrate utilization test was performed. Citrate utilization test employed the use of 5-6 colonies streaked on the slant of citrate agar. This was later incubated (35+2oC for 18+2 h) and a bluish colouration was indicative of a positive test for citrate utilization (53). TSI was done to differentiate gram-negative enteric bacilli as previously described by (54).
MALDI-TOF for confirmation of bacteria
Isolates of bacteria were grown on BA and later prepared for inoculation on the MALDI Biotyper [Bruker Daltonic GmbH] (31). To confirm the isolates using ribosomal and housekeeping proteins, the MALDI-TOF was used. Generated spectra were compared with current libraries (MALDI Flex Control software system Server Version: 4.1.31 and 60 Hz Nitrogen Laser (337 nm wavelength) and log scores generated (Scores of the log were classified as: >2.0 = High confidence identification, 1.7-2.0 = Low confidence and <1.7.0 = Unreliable).
Sensitivity testing and identification of multi-drug resistant MRSA, EC and PA
Having confirmed the isolates, it was important to assess the antimicrobial resistance of S. aureus, P. aeruginosa and E. coli in LF patients with wounds. Colonies of bacteria were picked from MAC/BA and inoculated in 1 mL of normal saline to give 0.5 McFarland standard. Cotton swab sticks (sterile) were dipped in the inoculum and spread on MHA plates that had passed quality control (45). The plates were left to dry for 5 min and incubated at 35+2oC for 18+2 h after appropriate antibiotic disks had been placed on the plates. The following antibiotics were used for S. aureus: cefoxitin (30 µg), erythromycin (15 µg), chloramphenicol (30 µg), tetracycline (30 µg), ciprofloxacin (5 µg), sulfamethoxazole-trimethoprim (23.75 µg /1.25 µg), clindamycin (2 µg), gentamicin (10 µg) and penicillin (10 UI). The following antibiotics were used for P. aeruginosa: ciprofloxacin (5 µg), ampicillin (10 µg), aztreonam (30 µg), sulfamethoxazole-trimethoprim (23.75 µg /1.250 µg), amikacin (30 µg), ampicillin-sulbactam (10 µg /10 µg), ceftriaxone (30 µg), gentamicin (10 µg), cefuroxime (30 µg), meropenem (10 µg), ceftazidime (30 µg) and tetracycline (30 µg). For E. coli, all the antibiotics used for P. aeruginosa were used with the exception of aztreonam (30 µg) and amikacin (30 µg). Zones of inhibition were read after incubation and breakpoints were recorded (45). MDR S. aureus, P. aeruginosa and E. coli were identified as resistant to at least one antibiotic in three or more classes. The following antibiotic classes were used in this study: aminoglycosides, carbapenems, cephalosporins, fluoroquinolones, macrolides, monobactams, penicillins, sulfonamides and tetracyclines.
Minimum inhibitory concentration (MIC) assay
Preparation of inoculum
To gain an insight into how the isolates from LF patients differed from typed strains (ATTC) in terms of inhibitory concentrations of antibiotics, MIC was performed using a broad-spectrum antibiotic (ciprofloxacin). Distinct colonies (5 - 6) grown on MHA were picked from an 18+2 h culture and transferred to normal saline to give approximately 1.5 × 108 cfu/mL (0.5 McFarland standard). A volume of 0.1mL of 1.5 × 108 cfu/mL inoculum was transferred to 9.99 mL of broth to give 1 × 106 cfu/mL (55). This was homogenized on a roller. An equal volume (100 µL) of an antimicrobial agent was added to the same volume of broth containing 1 × 106 cfu/mL to give a final concentration of 5 × 105 cfu/mL (56). Tubes were inoculated within 30 min of standardizing the inoculum to maintain viable cell density.
Viable cell count and purity check
Viable cell counts were performed to ensure the final concentration of inoculum was 5 × 105 cfu/mL. This was performed by taking 3 µL of the stock of inoculum (5 × 105 cfu/mL) and diluting it in 3 mL of saline. A volume of 100 µL was taken and then spread on MHA plates, where an average of 50 colonies were counted from the final concentration. Purity plating was done by taking an inoculum and plating on an unselective medium (BA) to check if growth is uniform throughout the plate.
Determination of MIC of ciprofloxacin using microdilution assay
Labelled wells of microtitre plates (1 - 11) were filled with 100 µL of broth (1 × 106 cfu/mL), while well 12 was filled with broth without any bacteria inoculum. Test solutions of 100 µL were added to each well (1-8) in the microtitre plates to give concentrations in the range (128 mg/mL to 1 mg/mL). Wells 9 and 10 were filled with 100 µL of positive control (ciprofloxacin), while wells 11 and 12 were filled with normal saline (56) . This was then incubated at 35+2oC for 18+2 h. To determine MIC, 20 µL of 3-(4,5-dimethylthiazol-2-yl)-2-5-diphenyltetrazolium bromide [MTT] (1.25 × 103 mg/mL) was dispensed in each well and the sample re-incubated for 30 min (57). MIC for ciprofloxacin against test organisms were indicated by the presence of a violet colouration (due to the production of insoluble formazan from MTT by viable cells). MIC was taken as the immediate well that did not show a violet colouration after 30 min of addition of MTT.
Minimum bactericidal concentration
To determine the concentration at which ciprofloxacin was bactericidal, an inoculum was taken from the MIC wells prior to MIC determination. The inoculum was plated on MHA plates and incubated at 35+2oC for 18+2 h. The value of MBC was recorded after incubation.
Phenotypic identification of MRSA and extended β- lactamases (ESBL)
MRSA was identified as S. aureus with a zone of inhibition < 22 mm for cefoxitin [30 µg] (45,58). Screening of suspected extended-spectrum β- lactamases for E. coli and P. aeruginosa were done by plating the confirmed isolates on MacConkey plates with a concentration of 0.1 of ceftazidime or cefotaxime. Isolates that grew on the ceftazidime or cefotaxime MAC agar underwent the double-disk synergy test (59).
Extraction of DNA
ESBL-positive and MRSA isolates were examined for the presence of blaTEM, blaSHV, blaCTX-M and mecA using conventional PCR. DNA extractions were performed from overnight bacterial cultures using the boiling method (31,60). A loop full of bacteria was emulsified in 1 mL of phosphate-buffered saline and vortexed vigorously. The isolates were then centrifuged at 8000 rpm for 3 min and the supernatant discarded. A volume of 100 µL of nuclease-free water was added to the bacterial cells, vortexed vigorously and then incubated at 95oC for 5 min. After incubation, centrifugation was done at 14000 rpm for 5 min. The supernatant was pipetted into a well labelled 1.5 mL Eppendorf tube as DNA. The extracted DNA was stored at -20°C until use. The extracted DNA were used as templates for the detection of the genes of interest.
Amplification of genes
To determine possible genes potentiating resistance in the MDR isolates, PCR amplification was performed. Detection of blaTEM, blaSHV, blaCTX-M and mecA were performed using primers in Table 1. Amplification using the Veriti® thermal cycler was conducted using the following PCR conditions; an initial denaturation at 95.0oC for 15 min; 35 cycles at 94oC for 30 sec, 60oC for 1 min and 72.0oC for 1 min. With a final extension of 72.0oC for 7 min, the reaction was put on hold at 4oC until ready to be viewed for bands. Each reaction volume was 25 µL and was done with modifications from Lin and colleagues protocol (61). The following PCR conditions were used for mecA gene; an initial denaturation at 94.0oC for 30 sec; 30 cycles of 94oC for 15 sec, 45oC for 1 min and 68.0oC for 1 min. With a final extension of 68.0oC for 5 min, the reaction was put on hold at 4oC. Each reaction volume was 25 µL (62). PCR products were visualized by agarose gel electrophoresis using UV-transilluminator (Vilber Lourmat, Collegien, France) after staining with ethidium bromide. The stained gel was captured onto a desktop computer using the infinity® software.
Data analysis
Data were analyzed using Python (v 3.8.10), Matplotlib (v 3.3.3), Numpy (v 1.19.4) and Scipy (v 1.5.4).