Multidrug-Resistant and Extended-Spectrum β-Lactamase (ESBL) - Producing Enterobacterales Isolated from Carriage Samples among HIV Infected Women in Yaoundé, Cameroon

The exacerbation of antimicrobial resistance (AMR) is a major public health threat worldwide. In sub-Saharan Africa, there is a scarcity of data regarding multidrug-resistant (resistance to at least one antibiotic of three or more families of antibiotics) as well as extended spectrum β-lactamase-producing Enterobacterales (ESBL-PE), isolated among clinical and asymptomatically healthy patients, especially in women living with HIV (WLHIV) despite their immunocompromised status. The overarching aim of this study was set to determine the prevalence and characterize genotypically multi-drug resistant Enterobacterales (MDR-E) and ESBL- PE isolated from vaginal swabs of WLHIV attending the Yaoundé Central Hospital, Yaoundé, Cameroon. A cross-sectional study was conducted among WLHIV during a four-month periods from 1 February to 31 May 2021. A total of 175 WLHIV, of childbearing age and under antiretroviral treatment were contacted. One hundred and twenty participants (120) were recruited and vaginal swabs were collected from them. After culture on Eosine-Methylen Blue (EMB) agar, the identification of Enterobacterales was performed using API 20E kit. A double-screening of ESBL-PE was performed using a combined disc diffusion method and ROSCO Diagnostica kits. An antibiotic susceptibility test was carried out by disc diffusion as per the Kirby-Bauer method and the β-lactamase resistance genes, blaCTX-M, blaCTX-M-group1-2-9, blaTEM were molecularly characterized using a conventional Polymerase Chain Reaction (PCR). Overall, 30.83% (37/120) of the included WLHIV were colonized with Enterobacterales and the prevalence of vaginal carriage of MDR Enterobacterales among them was 62.16% (23/37). Among MDR-E isolates, the most prevalent species were E. coli (56.0%; 14/25) and K. pneumoniae (20.0%; 5/25). High rates of resistance to trimethoprim-sulfamethoxazole (96.0%; 24/25), amoxicillin-clavulanic acid (88.0%; 22/25) and gentamicin (72%; 18/25) were observed. The resistance mechanisms detected among these isolates were ESBL (48.0%; 12/25), ESBL+ porin loss (8.0%; 2/25), ESBL+AmpC (24%; 6/25), with blaCTX-M, blaCTX-M-group-1,2,9 being identified at 48.0% (12/25) for each of them and blaTEM at 72.0% (18/25). Our findings confirm the high-prevalence of MDR as well as ESBL-PE isolated in WLHIV, and suggest that a real time monitoring system of antimicrobial resistant bacteria coupled with the reinforcement of infection prevention control (IPC) strategies are needed to sustainably contain these life-threatening pathogens especially in the most vulnerable populations.


Introduction
Antimicrobial resistance (AMR) remains a public health threat worldwide, especially in low-and middle-income countries (LMICs) where the limited hygiene and sanitation measures associated with numerous risk factors exacerbate this concern [1] (p. 10).
Enterobacterales have been described as the main causative agents in numerous infectious diseases in healthcare settings and have been identified in the carriage and clinical samples of patients in communities, especially those with numerous co-morbidities [2] (p. 1874). Several mechanisms to escape the activities of antibiotics have been described among Enterobacterales and the most common is the production of an extended spectrum ß-lactamase as well as carbapenemases.
The World Health Organization (WHO) has ranked extended spectrum β-lactamaseproducing Enterobacterales (ESBL-PE) as pathogens of critical priority in the research and development of new drugs [3] (p. 12). The emergence and global spread of multidrug resistant Enterobacterales (MDR-E), in hospital and community settings limits the use of therapeutic options, especially for people with co-morbidity conditions such as cancers and HIV [4] (p. 53-54). Several studies have highlighted the high prevalence of MDR-E and ESBL-PE in carriage samples of patients living with HIV (PLHIV) in Tanzania and Nigeria [5][6][7] (p. 611, 1542, 3). Furthermore, these studies postulated that HIV positivepatients are an important reservoir of ESBL-PE in Africa.
Moreover, there is a dearth of knowledge regarding the burden of MDR-E and ESBL-PE in people living with HIV, particularly in sub-Saharan Africa. A better understanding of the current epidemiology of these resistant pathogens in sub-Saharan Africa, the epicentre of the HIV pandemic (with around 67.0% of the world's population living with HIV) and a region accounting for around 63.0% of new infections from women of childbearing age yearly will considerably improve health outcomes of this vulnerable population [8] (p. 3). In Cameroon, financial constraints preclude the active surveillance of AMR in general, MDR-E and ESBL-PE in particular. Neither the prevalence of MDR-E nor that of ESBL-PE colonizing or infecting women living with HIV (WLHIV) in communities and hospitals is known in the country. Such data could help to understand if antibiotic resistant Enterobacterales might be involved in infections in WLHIV, understand their transmission pathways and can help to determine their burden in this vulnerable population, thus serve as evidence to implement adequate prevention and containment measures of AMR in the country. This study therefore aims to phenotypically and genotypically characterize MDR and ESBL-PE isolated from vaginal swabs among WLHIV in Yaoundé, Cameroon.

Sociodemographic and Clinical Characteristics
A total of 185 WLHIV were contacted during the sample collection period. Out of these, 120 (64.86%) agreed to participate and provided vaginal samples. The mean age of included women was 36.85 ± 7.22 years and median age was 37 years (min: 20 max: 50). Among the 120 included WLHIV, 37 (30.83%) were colonized by Enterobacterales, with 23 62.16% (23/37) being colonized by MDR-E. Of these women, 47.82% (11/23) also presented ESBL-PE. Among the 37 WLHIV colonized by an Enterobacterales, women between the ages of 35-39 (27.03%) were principally colonized followed by those between the ages of 40-44 (21.62%) and >45 (21.62%) years. Likewise, over 60.0% of WLHIV colonized by Enterobacterales had been to high school or had a high school degree (Table 1). Farmer (35.14%) and civil servant (21.62%) were the leading profession of the WLHIV colonized by an Enterobacterales isolate. Only five out of the 37 WLHIV were pregnant while all had a previous history of antibiotic treatment and an undetectable viral load of HIV. Among the WLHIV colonized by Enterobacterales, the majority of women aged between 40 and 44 years and those having been educated to a high school level were mainly colonized by MDR-E at respective percentages of 87.50% (7/8) and 66.67% (16/24). According to clinical parameters, 60% (3/5) of pregnant women were colonized by MDR-E and 20% (1/5) by ESBL-PE as shown in Table 1.
Overall n (%) 25 12 (48.0) 6 (24.0) 2 (8.0) 5 (20.0) 11 (44.0) 1 (4.0) Table 6. Distribution of β-lactamase genes between MDR-Enterobacterales.  Figure 2 shows the 1.5% agarose gel electrolysis of conventional PCR amplification products of the bla CTX-M , bla CTX-M-group-1,2,9 , bla TEM genes detected from nine MDR-E isolates. E. coli (56.0%; 14/25) and K. pneumoniae (20.0%; 5/25) were the most prevalent specie among the MDR-E isolates. These rates are similar to those obtained in an Ethiopian study assessing Enterobacterales in clinical samples where E. coli accounted for 51.54% (150/291 of MDR-E isolates [11] (p. [4][5]. Extended spectrum β-lactamase producing Enterobacterale (ESBL-PE) isolates were all MDR in our study, representing 48% (n = 12/25) of MDR-E isolates. We surmise that this might be due to the co-existence of genes encoding for these ß-lactamases along with other resistance genes encoding for resistance to other antibioti families, which has been regularly reported among ESBL-PE in the literature as they are known to be associated with mobile genetic elements such as plasmids [12] (p. 1175). E coli (41.16%) and K. pneumoniae (33.33%) were the leading ESBL-PE species in our study These results are comparable to those obtained in a study conducted in Togo, where E coli (63.64%) and K. pneumoniae (27.27%) were the most common ESBL-PE species isolated among vaginal swabs in clinical samples [12] (p. 1168). The presence of ESBL-PE in the vaginal microbiome could be associated with (i) the weakness of the immune system o WLHIV, (ii) the development of resistance mechanisms due to prolonged ARV treatmen or prophylactic antibiotic use, or (iii) limited hygienic measures [13] (p. 10).

MDR-E Species
Antibiotic susceptibility testing showed that the all MDR-E isolates displayed high resistance to trimethoprim-sulphamethoxazole (96.0%), piperacillin-tazobactam (88.0%) amoxicillin-clavulanic acid (80.0%) and chloramphenicol (68.0%). The high resistance rate to these antibiotics could be explained by the fact that they are readily available over-the

Discussion
The aim of this study was to characterize the phenotype and genotype of MDR-E and MDR-ESBL-Producing Enterobacterales E. coli (56.0%; 14/25) and K. pneumoniae (20.0%; 5/25) were the most prevalent species among the MDR-E isolates. These rates are similar to those obtained in an Ethiopian study assessing Enterobacterales in clinical samples where E. coli accounted for 51.54% (150/291) of MDR-E isolates [11] (p. [4][5]. Extended spectrum β-lactamase producing Enterobacterales (ESBL-PE) isolates were all MDR in our study, representing 48% (n = 12/25) of MDR-E isolates. We surmise that this might be due to the co-existence of genes encoding for these ß-lactamases along with other resistance genes encoding for resistance to other antibiotic families, which has been regularly reported among ESBL-PE in the literature as they are known to be associated with mobile genetic elements such as plasmids [12] (p. 1175). E. coli (41.16%) and K. pneumoniae (33.33%) were the leading ESBL-PE species in our study. These results are comparable to those obtained in a study conducted in Togo, where E. coli (63.64%) and K. pneumoniae (27.27%) were the most common ESBL-PE species isolated among vaginal swabs in clinical samples [12] (p. 1168). The presence of ESBL-PE in the vaginal microbiome could be associated with (i) the weakness of the immune system of WLHIV, (ii) the development of resistance mechanisms due to prolonged ARV treatment or prophylactic antibiotic use, or (iii) limited hygienic measures [13] (p. 10).
Antibiotic susceptibility testing showed that the all MDR-E isolates displayed high resistance to trimethoprim-sulphamethoxazole (96.0%), piperacillin-tazobactam (88.0%), amoxicillin-clavulanic acid (80.0%) and chloramphenicol (68.0%). The high resistance rate to these antibiotics could be explained by the fact that they are readily available over-thecounter in Cameroon and are thus used for self-medication. In addition, trimethoprimsulphamethoxazole is recommended to HIV-positive adults for prophylaxis against oppor-tunistic infections [13] (p. 9-10). This apparent unwarranted consumption of antibiotics has likely triggered the emergence of MDR in the human microbiome in Cameroon. The elevated rate of multi-drug resistance within E. coli and K. pneumoniae indicates that the effectiveness of antibiotics commonly used for bacterial infections among WLHIV is therefore threatened. This is even more worrying given that five of the WLHIV colonized by Enterobacterales were pregnant and three were colonized by MDR-E or MDR-ESBL-PE. Consequently, it is plausible that these pregnant women could not only suffer from a difficult-to-threat infection, but could also transmit the MDR-E or MDR-ESBL-PE isolates to their neonates either in utero or during delivery, thereby threatening the health of their progenies.
Our data confirm high rates of concomitant resistance to several antibiotics especially among MDR-ESBL positive compared to MDR-ESBL negative-Enterobacterales. This indicates that MDR-ESBL-PE are significantly more resistant to various antibiotic classes than those MDR-ESBL negative Enterobacterales. Interestingly, none of the MDR-ESBL positive and MDR-ESBL negative Enterobacterales were resistant to imipenem and up to twelve Enterobacterales were wild type with resistance to a maximum of two antibiotics. This result is highly relevant as it reveals that it is the time to act to preserve the efficacy of antibiotics for future generations. It further suggests that appropriate prevention and containment measures are urgently needed to sustainably curb AMR.
Interestingly, despite the high susceptibility of imipenem, bla OXA-48 (44%; 11/25) and bla KPC (4%; 1/25) have been phenotypically detected among the MDR-E isolates. These enzymes are among the main carbapenemase found in Enterobacterales. These data are worrying because ESBL-PE infections are treated as a last resort with carbapenems. Infections caused by carbapenemase-producing Enterobacterales (CPE) are generally associated with a very poor prognosis and high mortality, especially in immunocompromised patients [14] (p. 2). AmpC production has also been detected between MDR-E isolates at a rate of 20% (5/25). This enzyme is responsible for resistance to cephalosporin, penicillin and beta-lactamase inhibitors [15] (p. 61).
The bla CTX-M were the leading genes among MDR-ESBL-PE isolates with a 91.66% (11/12) prevalence. This could be due to the irrational consumption of ß-lactam antibiotics, especially cefotaxime, among WLHIV in Cameroon. These genes were also predominant among ESBL-PE isolated from HIV patients in Zimbabwe [5] (p. 611) and in Tanzania [6] (p. 1542). This is clearly in line with several reports which showed that CTX-M is the leading ESBL gene across the world. Following the detection of CTX-Mgroup-1, 2, 9 among MDR-E, we postulate the presence of bla CTX-M-15 among these isolates. Indeed, bla CTX-M-15 is part of group 1 of CTX-M, which is the most widely distributed across the world, and some studies have reported their presence in Cameroon [18,19] (p. 38, 5347). The presence of CTX-M among WLHIV is a danger. Not only because it encodes for ESBL, but also because it is carried by plasmids and could therefore horizontally spread within and between species as well as across hosts [16] (p. 12). Furthermore, as these women are of childbearing age, during an eventual pregnancy, vertical transmission of these MDR-E carrying bla CTX-M genes could occur, likely complicating the therapeutic choice in case of an infection of their newborns.

Study Settings
This study was conducted over a period of four months from 1 February to 31 May 2021 at one of biggest tertiary health care centre in Yaoundé, the Yaoundé Central Hospital (YCH). This healthcare structure provides patients with medical and paramedical teams specialised in various medical fields. It has a total of 650 beds, 70 doctors, 408 paramedical staff and 114 administrative staff and state agents. The day hospital, which is a fully fledged service of the YCH, is a pilot program for the fight against HIV/AIDS. It coordinates its activities with approximately 12,000 people living with HIV and is designated as an aggregated centre with a therapeutic committee, an ethics committee, a laboratory, and a pharmacy to ensure the distribution of anti-retroviral (ARVs).

Study Population
Our study included WLHIV, of childbearing age and under ARVs treatment at the YCH. All eligible women attending the day hospital were asked to participate in the study. Those who provided their informed consent to participate in the study were recruited. Sociodemographic and clinical information was recorded from pre-designed questionnaires on EPI INFO (version 7.7.3). Following this step, information was codified in order to ensure confidentiality.

Sample Collection
Vaginal swabs were collected aseptically in an appropriate sampling room. After an explanation of the sampling procedure, the patient was placed on a gynaecological bed and after proper speculum insertion the samples were collected with a sterile cotton swab.

Laboratory Analyses
The samples were immediately sent to the microbiology unit of the YCH laboratory for microbiological analysis and thereafter the isolates were molecularly characterized in the reproductive, maternal, neonatal and child health research unit of the Centre of Expertise and Biological Diagnostic of Cameroon (CEDBCAM).

Culture and Identification
Vaginal swab samples were directly plated on to Eosin Methylene Blue (EMB) agar. Culture media were thereafter incubated overnight at 37 • C for 18-24 h. Growing colonies were primarily identified based on their morphological characteristics, then subjected to Gramstaining and an oxidase test which was conducted to ascertain the Enterobacteriaceae to the family level. The API 20E kit (BioMérieux, Marcy l'Etoile, France) was used to biochemically characterize Enterobacterales species according to the manufacturer's instructions.

Antimicrobial Susceptibility Testing (AST)
The AST of isolated Enterobacterales was assessed using the Kirby-Bauer disc diffusion method. Briefly, Mueller-Hinton agar was swabbed with 0.

Screening and Confirmation for ESBL-and Carbapenemase -Producing Enterobacterales
Phenotypic screening of ESBLs production among our isolates was performed by looking for champagne cork or funnel shaped synergies according to the CA-SFM recom-mendation [20] (p. 40). This first test was conducted using a combination of amoxicillinclavulanic acid and cefotaxime, ceftriaxone and aztreonam to screen for ESBL production and amoxicillin-clavulanic acid and imipenem to screen for carbapenemase production. A second screening was performed using ROSCO Diagnostica kit according to the manufacturer's instructions to ascertain ESBLs, ESBL-porin loss, AmpC, KPC, MBL and OXA-48 in our isolates.

Genomic Extraction
The genomic DNA of the MDR-E isolates was obtained using a modified extraction by the boiling method as described previously [21] (p. 5347). Briefly, one to two pure MDR-E colonies were suspended in 400 µL of Tris-EDTA (10 mM Tris, 0.1 mM EDTA) and then vortexed for five seconds. The suspension was then incubated for 25 min at 95 • C in a dry bath digital (MIULab DKT200-1, Lasec International Ltd., Johannesburg, South Africa). After incubation, the suspension was centrifuged for 5 min at 9500 rpm and then, 150 µL of the supernatant containing DNA was subsequently transferred to a new Eppendorf tube then stored at -40 • C.

Conventional Polymerase Chain Reaction
The detection of the bla TEM , bla CTX-M and bla CTX-M-group-1,2,9 , genes Among the MDR-E isolates was performed by conventional PCR in thermal cycler BIO-RAD T100 (Bio-Rad Laboratories, Marnes-la-Coquette, France). The reaction occurred in a 25 µL reaction mix consisting of 9 µL of 2x Taq green polymerase master mix (New England Biolabs, Ipswich, MA, USA), 12.5 µL of nuclease-free water, 0.25 µL of each reverse and forward primer [10 µM] and 3 µL of DNA. The amplification steps were as follows: initial denaturation (94 • C for 30 s), 30 cycles of denaturation at 94 • C for 4 s annealing for 40 s (TEM) and 45 s (CTX m universal and CTX-M-group-1,2,9), elongation at 72 • C for 50 s (CTX m universal and CTX-M-group-1,2,9) and 60 s (TEM) and final elongation at 72 • C for 5 min (TEM) and 6 min (CTX m universal and CTX-M-group-1,2,9). Annealing temperatures and primer sequences are showed in Table 8.

DNA Electrophoresis and Visualization
After amplification, DNA electrophoresis was performed on agarose gel of 1.5% (v/v) that was run at 90 V for 45 min along with a 100 bp molecular ladder (New England Biolabs, MA, USA). The gel was then stained in ethidium bromide solution (0.5 µg/mL) and PCR products were visualised under UV light using a gel documentation system G-BOX chemi XL (Syngene, Cambridge, UK).

Data Analysis
Data analysis was performed using R software (version 4.1.0) and RStudio (version 2021.09.0). Proportions were compared using the Fischer exact test and two-sample t-test as appropriate. A participant was considered positive for MDR-E or ESBL-PE when at least one colony displaying the multidrug resistant or ESBL phenotype was detected. Results were considered statistically significant at a p-value < 0.05.

Conclusions
Our study revealed a high prevalence of MDR-E and MDR-ESBL-producing-Enterobacterales in the vaginal flora of WLHIV at Yaoundé Central Hospital. E. coli and K. pneumoniae were the leading resistant bacterial species while bla TEM and bla CTX-M were detected at elevated rates. These MDR-E, colonizing WLHIV, threaten not only WLHIV because they could be responsible for serious difficult-to-treat bacterial infections among these already vulnerable women, but also could compromise their reproductive health and health of their future progenies in case of intrapartum or neonatal (early or late onset) infections. These results evidence the need to implement stringent infection prevention and control measures, and to develop adequate strategies to monitor and limit the dissemination of MDR-Enterobacterales in this population. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Written informed consent was obtained from the patient(s) to publish this paper where applicable. Data Availability Statement: Not applicable here.