CHAPTER ONE: INTRODUCTION
1.1 BACKGROUND OF THE STUDY
Vibrio cholerae is the causative agent of Cholera disease which has spread globally in seven pandemic waves since 1817 (Tappero & Tauxe, 2011). Vibrios are highly motile, Gram-negative, comma-shaped rods with a single polar flagellum. Though few V. cholerae strains are clinically significant to humans, V. cholerae type O group 1 is the most important as a cause of epidemic Cholera disease. Cholera causes severe dehydrating diarrhea which can lead to death in untreated patients, especially in developing countries thus making cholera an issue of major public health importance in Kenya. This life threatening diarrhoea is attributed to massive luminal secretion of water and electrolytes from enterocytes induced by the cholera toxin (Salim, 2005).
Cholera was first discovered in Kenya in 1971 and since then several outbreaks have been reported thus rating it among the 35 priority diseases in Kenya. Fifteen discrete outbreaks have been reported from 1971 to 2010. These recurrent outbreaks mostly affected Nairobi, Nyanza and Coast regions. In addition, refugee camps, remote arid and semi-arid regions were also affected. The largest outbreak which resulted in 26,901 cases and 1,362 deaths was reported between 1977 and 1999 (Mutonga, et al., 2013). Earlier (1970s to early1990s) V.cholerae strains belonged to serotype Ogawa which showed resistance to a wide range of antibiotics like chloramphenicol, streptomycin, tetracycline, sulfamethaxazole/trimethoprim (cotrimoxazole) and ampicillin mediated by plasmid ca 100mD of incompatibility group C (Finch , et al., 1988). However, in the recent past, serotype Inaba emerged as the main cause of cholera disease with isolates showing reduced sensitivity to chloramphenicol and cotrimoxazole. Isolates collected between 1994-2007, contained strB ,Sull 2, dfrA1 and floR genes coding for resistance to streptomycin, sulfamethoxazole, trimethoprim and chloramphenicol respectively. These isolates were also negative for class 1, 2 and 3 integrons (Materu, et al.,1997 Kiiru, et al.,2009). It appears that the genetic relationship among past vibrio isolates is based on the type of technique used. In a past study for example, isolates studied using the Multilocus-Variable Tandem Repeat Analysis (MLVA), revealed that multiple genetic lineages of V.cholerae were simultaneously infecting persons in Kenya while those studied using pulsed field gel electrophoresis revealed that all isolates were genetically related (Mohamed, et al., 2012).
The standard way of treating severe cases of cholera currently is through administering rehydration fluids. However, the use of antibiotics reduces the duration of diarrhea by 50 – 56 %, reduces stool output by 8 – 92 % and reduces shedding of bacteria by 26 – 83 %. In general, antibiotics reduce the duration of illness by up to 50 % (Sack, et al., 2001).
Despite the advantages associated with the use of antibiotics, it has been compromised by the evolution and spread of strains conferring resistance to multiple antibiotics including those which have been recommended by WHO that include doxycycline, furazolidone, sulfamethoxazole, trimethoprim, chloramphenical and ciprofloxacin. An erratic shift in antibiotic resistance among the V. cholerae strains has also been noted in isolates collected in different regions over the same period of time.
The most common treatment used against bacterial pathogens is the administration of β-lactam antibiotics such as penicillins, monobactams and cephalosporins but the emergence of β-lactamases that mediate resistance to these antibiotics is on the rise (Shaikh, et al., 2015). β-lactamases are rare in Vibrio but recently, the prevalence of these enzymes has been increasing in this Genus. β-lactamases are frequently found in Gram negative bacteria which destroy the β-lactam ring of these antibiotics through hydrolysis thus resulting in ineffective compounds (Pitout, et al., 2005). The β-lactamases are classified based on two schemes namely, the Bush-Jacoby Medeiros functional classification scheme which is based on enzyme properties such as substrate and inhibitor profiles and the Ambler molecular classification scheme which is based on amino acid sequence of enzymes where class B enzymes are metallo-β lactamases which require zinc ions for substrate hydrolysis (Bush & Jacoby, 2010) and class A,C,D which utilize serine for β-lactam hydrolysis (Shaikh, et al., 2015 Bush & Jacoby, 2010). These enzymes have evolved over time through mutations and continuous overproduction which can be attributed to constant exposure of bacteria to a multitude of β-lactams (Shaikh, et al., 2015). This has then given rise to enzymes exhibiting increased activity against newly developed β-lactam antibiotics and these are the extended spectrum β-lactamases. Most β-lactamases are encoded in plasmids that are easily transferred from one bacterium to another (Shaikh, et al., 2015).
Evolution of transposons and integrons has greatly influenced flexibility in the genetic response to a variety of antibiotics (Wozniak, et al., 2009). This has however not been extensively studied in Kenya. Integrons are commonly found in many bacterial genomes. They are flexible gene acquisition systems which have the ability to acquire, express and disperse antibiotic resistant genes thus posing a major threat to antibiotic therapy. They are not capable of self-transposition hence they associate with insertion sequences, transposons and conjugative plasmids which serve as vehicles for transmission (Gillins, 2014). They may also have smaller integrons embedded in mobile genetic elements such as transposons and conjugative plasmids which can disseminate horizontally (Burrus.,et al,2006).Vibrio isolates harbor large chromosomal integrons giving them the capacity to rapidly transfer gene cassettes containing antibiotic resistant genes. Vibrio isolates have also been associated with class 1 integrons in other parts of the world but to date none have been isolated in Kenya. Integrative conjugative elements (ICEs) integrate and replicate within the host chromosome and can excise themselves and transfer between bacteria by conjugation (Wozniak et al., 2009). The ICEs commonly carry several antimicrobial drug resistance genes and play a major role in the spread of antimicrobial drug resistance in V. cholerae, a good example is the SXT element which harbors resistance to trimethoprim and sulphamethaxazole (Burrus et al., 2006). This element belonging to SXT/ R391 family was present in Kenyan isolates collected between 1994 and 2007 (Kiiru, et al., 2009).
Bacteriophages which are abundant in aquatic ecosystem also play an important role in the evolution of bacterial genomes. To date, no phage encoded resistant genes are known, however, they may assist in mobilizing plasmids during transduction (Davinson, 1999).
Resistance to antibiotics poses a global threat to the fight against infections and has been shown to vary with time in V.cholerae. This can be attributed to evolution of mobile genetic elements which has been poorly studied in Kenya. Therefore, it becomes a necessity to study evolution of antibiotic resistance over a period of time. This knowledge will then assist in making decisions with regards to antibiotic use and also in coming up with strategies of controlling the antibiotic resistant strains.
1.1.1 PROBLEM STATEMENT AND JUSTIFICATION
Despite advances made in understanding the cholera disease, it still continues to be a major public health problem in Kenya with the recent outbreak reported in the month of February 2015 (relief web, 2015).
Several studies conducted have shown increased multiple resistance to some of the antibiotics recommended for use in cholera by WHO and these include; tetracycline, doxycycline, furazolidone, sulfamethaxazole, trimethoprim, chloramphenical and ciprofloxacin. In addition, differences have been reported in antibiotic resistant patterns in studies carried among V.cholerae isolates during a given period of time in different regions. For instance, V.cholerae O1 strains isolated between 1994 and 1996 from five countries in the East African region (Kenya, Sudan, Somali, Tanzania and Rwanda) exhibited no uniformity in the antibiotic resistance pattern. However isolates collected in Kenya between 1992 and 2007 had similar resistance patterns (Kiiru.et al., 2009). Erratic shifts in antibiotic resistance among V. cholerae isolates have also been noted and yet the genetic basis of these changes has not been investigated in Kenya.
Over the years, antibiotic pressure in the bacteria’s environment increases the need for them to acquire a resistant gene through mobile genetic elements. However, only a small proportion of these elements have been screened in Kenyan isolates and temporal changes in carriage of these elements is still unknown.
Therefore, in order to understand changes in antimicrobials, it becomes necessary to carry out this study in isolates accumulated over a long period of time. This will guide the health policy makers in formulating policies with regards to antibiotic use and also in coming up with strategies of controlling the antibiotic resistant strains.
1.1.2 HYPOTHESIS
There is no difference in the resistance profile and the genetic determinants of resistance among isolates recovered between 2006 and 2015.
1.1.3 OBJECTIVES
1.1.3.1 GENERAL OBJECTIVE
To determine genetic basis for antibiotic resistance and phylogenetic relationship in V.cholerae clinical isolates collected between 2006 and 2015 in Kenya.
1.1.3.2 SPECIFIC OBJECTIVES
1. To determine shifts in antibiotic resistance profiles of V.cholerae isolates recovered over a period of ten years.
2. To determine changes in carriage of genetic determinants among isolates recovered over a ten year period.
3. To determine change in diversity of selected mobile genetic elements (plasmids, Integrons, Integrative Conjugative elements) among these isolates over a period of ten years.
4. To establish changes in phylogenetic relationship among the antibiotic resistant strains obtained between 2006 to 2015
2 CHAPTER TWO: LITERATURE REVIEW
2.1 Introduction
V. cholerae, a member of the family Vibrionaceae, is a comma shaped Gram negative flagellated bacillus. It was first discovered by Fillipo Paccini in 1854 as the causative agent of cholera disease, a disease which has spread globally in seven pandemic waves (Banerjee et al., 2014). The bacterium is serologically classified based on the differences in the sugar composition of the heat stable surface somatic “O” antigen. There are 206 “O” serogroups whereby only serogroups O1 and O139 are known to cause epidemic cholera (Shimada et al., 1994). The O1 serogroup exists as two biotypes namely E1 Tor and classical.
The first six pandemics which occurred between 1817 and 1923 were caused by the classical biotype which spread from the Indian subcontinent to most parts of the world (Tappero et al., 2011). EI Tor was discovered in 1935 when it caused a major epidemic in Indonesia, Asia. It later caused the seventh pandemic in 1961 where it spread to different parts of Asia and Africa gradually replacing the classical biotype cholera strain (Banerjee et al., 2014). The strains are further differentiated into three serotypes; inaba, ogawa, hijokama based on antigenic factors. Inaba and Ogawa are the main serotypes while Hijokama is rare (Raychoudhuri, et al., 2008).
Cholera causes significant morbidity and mortality worldwide especially in developing countries which experience poor sanitation (Kiiru et al., 2009). Half of the cases and deaths in developing countries occur in children under 5 years (Ali, et al., 2012).
2.2 Antibiotic resistance
Multiple antibiotic resistances among V. cholerae strains have emerged as a major problem worldwide (Faruque et al., 2007). For instance between 1977 and 1980, clinical V.cholerae isolates resistant to most commonly used antibiotics like tetracycline, streptomycin, kanamycin, ampicillin and spectinomycin emerged in Tanzania (Mhalu, Mmari, & Ijumba, 1979) and Bangladesh (Glass, et al., 1980). This resistance is attributed to target modifications or acquisition of resistance genes from mobile genetic elements. The major source of antibiotic resistance in cholera pathogens among the various mobile genetic elements are the Intergrative conjugative elements (Banerjee et al., 2014).
In 1982, strains resistant to ampicillin, trimethoprim-sulfamethoxazole and tetracycline were isolated in Kenya and this resistance was found to be mediated by a single plasmid which differed from those found in other regions (Morris, et al., 1988). Further studies of the isolates from 1982 to 1985, demonstrated the persistence of the resistant strain which could have been attributed to the mass tetracycline prophylaxis campaigns which were carried out between the year 1981 to 1988 (Sack et al., 2001).
Cholera outbreaks which occurred in Kenya between 1970s and 1980s were caused by V.cholerae 01 serotype Ogawa which showed resistance to several antibiotics like chloramphenicol, tetracycline, ampicillin, streptomycin and sulfonamides. However, recent studies have shown serotype Inaba as the main cause of current outbreaks in Kenya. This serotype has shown resistance to chloramphenicol, streptomycin, sulfamethoxazole and trimethoprim (Kiiru, et al., 2009).
Antibiotic resistance has also been shown to vary in studies conducted during the same period of time at different regions. For instance, V.cholerae strains isolated between 1994 and 1996 from five countries in the East Africa region (Kenya, Sudan, Somali, Tanzania and Rwanda) exhibited no uniformity in the resistant pattern. The main observation was that only strains from Kenya and Sudan were susceptible to tetracycline (Materu et al., 1997). In isolates collected in the 2005 outbreak, 97% were not susceptible to tetracycline (Mugoya et al., 2008). In other countries like India, reemergence of tetracycline resistant strains (Jesudasonm, 2006) has been reported.
2.2.1 Mutations
Target modifications as a result of spontaneous mutations in the bacterial chromosome can result in resistance to antimicrobial compounds. This was emphasized through a study conducted in Tanzania in 1980 that revealed that genes belonging to V.cholerae undergo higher mutation rates than E.coli genes thus facilitating resistance to various antibiotics (Banerjee et al., 2014)
2.2.2 Intergrative Mobile genetic element
Evolution of various antibiotic resistance elements in different natural and clinical environments probably involves a variety of integrated genetic processes (Davies & Davies, 2010). Mobile genetic elements are key players for evolution of antibiotic resistance through acquisition, rearrangement and expression of antibiotic resistant genes normally contained in gene cassettes (Mutreja et al., 2011). They are transferred from distant or closely related organisms through transduction, transformation or conjugation and integrated in either or both chromosomes by site specific or homologous recombination. These mobile genetic elements include integrative and conjugative elements (ICE) which possess the ability to encode many properties. The SXT family of ICEs consists of 30 different elements which share the same intergrating site, Int, in the 5’ end of prfC of the host chromosome. They have the ability to intergrate into the bacteria’s genome, rearrange resistance genes and be transferred horizontally through conjugation a function that is mediated by a conserved set of 24 genes contained in the element. SXT a 99.5 kb ICE was first discovered in the chromosome of V.cholerae 0139 M010 from India and its name given as a result of its role in sulfamethoxazole and trimethoprim resistance. In 2001, another SXTET was isolated from V.cholerae 01 strain. This element also has the ability to mobilize plasmids and chromosomal DNA from one strain to another and has been noted to contain genes coding for resistance to chloramphenical and streptomycin (Banerjee, et al., 2014).
Resistance to antibiotics has been described in integrons which were first identified and characterized in 1987 by Stokes and Hall as unusual gene acquisition elements (Davies & Davies, 2010). These functions are enhanced by three genes; the gene for integron integrase (intl) which is an enzyme that catalyses recombination between incoming gene cassettes, integron associated recombination site (attl) gene and the integron associated promoter (pc) gene that is involved in expressing the gene cassettes. Integrase recognizes a specific sequence in certain resistance genes, 59-be which is captured by recombination at attl site. The intl-attl fragment is known as 59-CS and is highly conserved in all integrons. Classification of integrons is based on the sequence of their integrase. For instance, class 1 integrons have the 59-CS region, then a variable region containing gene cassettes which are diverse and may include aadA coding for resistance against streptomycin/spectinomycin or aac (6’) – 1b – cr coding for resistance against the quinolones, and finally 39-CS which is a conserved region containing two genes namely the sulphonamide resistance gene (Sul 1) and the quaternary ammonium resistance gene (qacEDI) (M.Sabate & G.pratz 2002)
Class 1 integron has been extensively studied among the Gram negative organisms as compared to class 2 integrons which are mostly linked to tns genes of transposons Tn7 which carry gene cassettes like dfra1, sat1 and aadA1 conferring resistance to their respective antibiotics at the variable region (Roe, Vegok, & Pillai, 2003). In the recent years, novel rearrangements of gene cassettes in the class 2 integrons has been noted and as a result of this evolution, it then becomes necessary to screen for class 2 integrons among V. cholerae clinical isolates and establish which genes they could be associated with.
Conjugative plasmids have auxiliary genes conferring resistance to antibiotics and heavy metals. These genes are often on other mobile elements like composite transposons, simple transposons and integrons. In many instances, transposition has commonly been associated with antibiotic resistance determinants and hence they play a vital role in the emergence of resistance to antibiotics (He, et al., 2015).
DNA elements coding for their own mobility are known as insertion sequences. Composite transposons are formed when the insertion sequences insert on either side of a short DNA region Fig 1-1. These transposons often carry one or more genes conferring resistance to antimicrobials. Examples of composite transposons include Tn9 which encodes chloramphenicol resistance, Tn5 encoding for kanamycin and streptomycin resistance and Tn10 encoding tetracycline resistance (Mayers, 2009). Insertion sequences IS26, has often been associated with antibiotic resistance among Enterobactericeae and studies have revealed that it plays a role reorganization of plasmids through replicative transposition among clinically isolated multidrug-resistant bacteria (He, et al., 2015). However, little is known about it in V.cholerae hence making it necessary to screen for it among clinical isolates in this study.
Figure 2.1: Position of insertion sequences on composite transposons (Jacek FH, 2007)
Tn 7 which is a site specific transposons carries short inverted repeat at its end, tns A-E transposition genes, class 2 integrons, dfrA1 genes coding trimethoprim resistance, sat genes coding streptothiricin resistance and aadA1 genes coding streptomycin/spectinomycin resistance (Mayers, 2009). Strains collected in Kenya between 1994 and 2007 were shown to contain Tn 7 though they were not able to link it with resistance to antibiotics (Kiiru, et al., 2009).
V.cholerae is reported to have a distinct class of integrons which allows the organism to acquire open reading frames and later convert these open reading frames to new functional genes (Akoachere, et al., 2013). Between 1996 and 1997, a strain of V.cholerae 01 containing a 150-kb conjugative plasmid with class 1integron was isolated in Guinea-Bissau and was shown to contain gene cassettes encoding resistance to aminoglycosides (aadA1) and trimethoprim (dfrA12) (Dalsgaard, et al., 2000). In 1998, integrons containing an aadA2 gene cassette for resistance to spectinomycin and the SXT element were isolated during a cholera outbreak which occurred in South Africa (Dalsgaard, et al.,2001). However, class 1 integrons harboring gene cassettes encoding for antibiotic resistance have not been isolated in Kenya.
2.2.3 Extended spectrum β lactamases
They were first discovered in 1983 and since then, have been detected in a range of Enterobacteriaceae. They have the ability to hydrolyse monobactams and broad spectrum cephalosporins like cefotaxime, ceftriaxone and ceftazidime which have an oxmino side chain. However, this resistance can be inhibited by sulbactam, clavulate and tazobactam which are serine-type β-lactamase inhibitors (Zhao & Hu, 2013). The ESBLs pose a serious threat to treatment therapy (Pitout, et al., 2005). ESBLs are classified in the functional group 2 serine β-lactamases (Bush & Jacoby, 2010)) or in Ambler molecular class A (Petroni, et al., 2002). The first ESBL to be discovered in a clinical isolate in 1984 was SHV-2 and since then other families have been documented. These include CTX-M, TEM, PER, VEB, BES, GES, TLA, SFO and OXA (Pitout, et al., 2005).
The plasmid-mediated acquired cefotaximases, CTX-Ms whose progenitors were discovered from Kyluvera species in late 1980s (Shaikh, et al., 2015) have been shown to cause significant clinical impact and rapid growth among the β-lactamases (Zhao & Hu, 2013). CTX-M β lactamases were acquired through horizontal gene transfer from other bacteria through mobile genetic elements (Shaikh, et al., 2015). These enzymes have shown preference in hydrolyzing cefotaxime and are better inhibited by tazobactam as compared to clavulate and sulbactam (Shaikh, et al., 2015). They are divided into five groups based on their amino acid sequence namely CTX-M group 1,2,8,9 and 25 (Shaikh, et al., 2015). To date, more than 100 CTX-M variants have been identified and sequenced ranging from CTX-M-1to CTX-M-124 with CTX-M-14 and CTX-M-16 been most prevalent in clinical pathogens worldwide (Zhao & Hu, 2013). Majority are highly resistant to cefotaxime and ceftriaxone except for CTX-M-14 and CTX-M-16 which exhibit high resistance also to ceftazidime (Zhao & Hu, 2013). These enzymes have been identified in 26 bacterial species including V. cholerae (Zhao & Hu, 2013). Insertion sequence ISEcp1 is one of the most important genetic platforms associated with CTX-M and is involved in mobilizing bla CTX-M genes from the chromosomes to plasmid. Usually, plasmids are classified based on their incompatibility (Inc) which is the inability of two plasmids to be incorporated stably in the same bacterial strain. There are 29 inc groups among plasmids of enteric bacteria with IncP, IncA/C and IncQ having a broad host range. The transfer of CTX-M genes to new hosts is mediated by the conjugative plasmids (Zhao & Hu, 2013). Production of CTX-M in an isolate can be detected phenotypically or genotypically through PCR amplification of bla CTX-M genes.
In 1989, the first TEM variant to display ESBL properties phenotypically was discovered and this was TEM-3 which was generated by amino acid substitution of its parent enzyme (Shaikh, et al., 2015). To date there are over 100 derivatives of TEM with most of them as ESBLs except TEM-1, TEM-2 and TEM-13 (Paterson & Bonomo, 2005). In order to conclude that a certain ESBL present in an isolate is related to TEM, it requires PCR amplification of bla TEM genes with oligonucleotide primers followed by sequencing in order to differentiate from the non-ESBL parent enzymes (Shaikh, et al., 2015). CTX-M–type, TEM-63 and PER-2 have been detected in V. cholerae O1 isolates from South Africa and Argentina (Petroni, et al., 2002) (Wu, et al., 2015). PER-1 ESBL has been detected in non O1 and non O139 V.cholerae strain in China, in 2005, where bla PER-1 gene was carried on Inc A/C plasmid (Wu, et al., 2015).
2.3 Phylogenetic relationship
Phylogenetic relationship among the isolates was determined by enterobacterial repetitive intergenic consensuses (ERIC) which are short sequences consisting of 127-bp, imperfect palindromes occurring in multiple copies in the Enterobactericeae genome (Wilson & Sharp, 2006). It is used in generating information of genetic relatedness or the uniqueness of bacterial strains (Meacham, et al., 2003). Other techniques like pulsed-field gel electrophoresis and multilocus sequencing can be used. However, ERIC-PCR was used to determine genetic diversity among the 2006-2015 V. cholerae clinical isolates in this study as it is faster and more cost effective.
3 CHAPTER THREE; MATERIALS AND METHODS
3.1 Study area
The study was conducted at Center for Microbiology Research, a unit of the Kenya Medical Research Institute where 130 strains of V. cholerae are stored at -80 0 C. These strains were rectal swabs collected from health facilities in Busia, Malindi, Nyanza, Nairobi and Northern parts of Kenya following cholera outbreaks which occurred between 2006 and 2015. The number of rectal swabs per year were as follows 2006-23 strains (2006), 2007-13 strains (2007), 2009-11 strains (2009), 2010-14 strains (2010), 2012-19strains (2012), 2015-50 strains (2015).
3.2 Study design
This was an experimental study design whereby the strains had been previously isolated from rectal swabs using standard bacteriological methods (CDC, 1999). These rectal swabs had been obtained from patients presenting with passage of three or more watery diarrhea during suspected cholera outbreaks.
3.3 Sample size
Sample size was a total of all the 130 isolates collected between 2006 and 2015 that represented all isolates obtained in this ten year period.
3.4 Identification and confirmations of the V.cholerae isolates
The archived vials containing V.cholerae isolates were removed from the -800 C freezer and allowed to thaw at room temperature. A loopful of the isolate was inoculated using the streak plate method on Thiosulphate Citrate bile salts Sucrose agar (selective media for V. cholerae) and incubated at 37 0C for 18 to 24 hr. The presence of V. cholerae was identified by the appearance of large (2 to 4 mm in diameter) shiny, yellow, opaque colonies. This was then subcultured on Muller Hinton to get pure colonies and incubated at 370 C for 18 to 24 hr. A Known V.cholerae strain was used as positive control and E.coli strain was used as negative control.
3.4.1 Serological identification
Pure colonies were used for serological identification. This was done using the slide agglutination technique whereby polyvalent antisera for V.cholerae type 01 and monovalents for V. cholerae type 01 serotypes Inaba and Ogawa were used. A well isolated colony was picked aseptically and mixed in a drop of sterile normal saline on a glass slide to make a milky suspension. A drop of the antisera to be tested was mixed with the milky suspension and any agglutination observed in 4 minutes confirmed the presence of V.cholerae. The agglutination also confirmed the serotype and subtype of the strains. Positive and negative controls were included.
3.5 Antimicrobial susceptibility testing
Kirby-Bauer disk diffusion method was used for susceptibility testing. One colony was picked from Mueller Hinton agar and suspended in 3 mls of normal saline to obtain a 0.5 McFarland opacity equivalent. Using a sterile cotton swab, a thin smear of this suspension was spread evenly on MHA plate to obtain uniform growth upon incubation.
Antimicrobial susceptibility testing was performed using commercial discs following manufacturer’s instructions. Two plates were used. On one plate using a sterile dispenser, the following drugs were applied ampicillin (10µg), cefdoxime (10 µg), ceftazidime (30 µg), cefotaxime (30 µg) then with the use of a sterile cool forceps, amoxicillin- clavulanic acid (10/ 100 µg ratio) was placed at the center of the plate. On the second plate the following drugs were applied nalidixic acid (30 µg), tetracycline (30 µg), ciprofloxacin (10 µg), trimethoprim (5.2 µg), streptomycin (25 µg), sulphamethoxazole (30 µg), gentamicin (10 µg) and chloramphenical (30 µg). A Control strain of E.coli ATCC 25922 was subjected to the same set of discs as it is virtually susceptible to all antimicrobials and therefore used to ascertain potency of the discs and quality of the media.
3.6 Molecular characterization
3.6.1 Polymerase chain reaction for detection of Antibiotic resistant genes and mobile genetic elements
Plasmid DNA and chromosomal DNA were extracted from the isolates through boiling. Two hundred microlitre (200μl) tubes containing PuReTaq ready-to-go PCR beads were used to set the polymerase chain reaction. Each reaction contained 0.125µl of Taq DNA polymerase, 1µl MgCl2, 2.5 µl dATP, dCTP, dGTP and dTTP, stabilizers, 2.5µl Q soluion,14.875µl of PCR water/buffer, 1μl each of forward and reverse primers (6 pmol/μl), 4μl of template DNA, and 2µl of coral load was added to a final volume of 30μl. The PCR amplifications of the target regions were carried out in a thermal cycler (Bio-Rad).
Polymerase chain reaction (PCR) was performed to screen for genes coding for resistance against β lactamases (blaTEM, blaCTX-M), genes coding for antibiotic resistance to streptomycin (strB), sulfamethoxazole (sul ll), trimethoprim (dfra1) and screening for mobile genetic elements class 2 integrons (int 2), insertion sequence IS 26 and plasmid Inc A/C.
PCR amplification was done using primers specific for the genes been screened for as shown in table 1. The process involved initial denaturation at 950c for 5 minutes, followed by 35 cycles at 94oc each cycle going for 30 seconds. Annealing temperatures were selected based on the genes of interest. This was followed by extension at 720c for 1 minute 30 seconds. The final extension step was set at 720c for 10 minutes.
The presence of amplicons was analysed using 1.2 % agarose gel electrophoresis stained with ethidium bromide. This were visualized under ultra violet light and the amplicon sizes determined by the help of a generuler 100 bp DNA ladder marker. DNA containing the genes of interest were used as positive controls.
3.6.2 Phylogenetic relationships among the isolates
The ERIC-PCR typing technique was used to determine genetic diversity among the V.cholerae strains. The primers which were used include ERIC 1 (5’- ATG TAA GCT CCT GGG GAT TCA C-3’) and ERIC2 (5’-AAG TAA GTG ACT GGG GTG AGC C-3’) (C & H, 2001). The process involved initial denaturation at 950C followed by 900C for 30 secs, annealing at 49.7 0C for one min , extension at 650C for 5 min and 680C for 16 min. The process was carried out in 30 cycles. The sequences generated from PCR (Appendix 1) were then incorporated in the Gelcompar II software to generate the phylogenetic tree.
3.7 Data analysis and management
Results on antibiotic susceptibility data and presence or absence of a gene by PCR were reported in form of tables and entered in the Microsoft Excel Spreadsheet. The hypothesis was tested using chi-square, while phylogenetic relationship was analysed using Gell compar II software. A p value less than 0.05 was considered statistically significant.
PRIMER PRIMER SEQUENCE Amplicon
size Annealing
Temp
int 2 F 5’-AAATCTTTAACCCGCAA-3’
R 5’ATGTCTAACAGTCCATTTT-3’ 440bp 480C
blaCTX-M
Consensus F5’ATGTGCAGYACGAGTAARGTKATGGC-3’
R5’TGGGTRAARTARGTSACCAGAAYCAG-3’ 593bp 600C
blaTEM
consensus F5’ATGAGTATTCAACATTTCCG-3’
R5’CCAATGCTTAATCAGTGAGG-3’ 840bp 53.50C
su 2 F5’-AGGGGGCAGATGTGATCCC-3’
R5’TGTGCGGATGAAGTCAGCTCC3’ 625bp 640C
dfra1 F5’CGAAGAATGGAGTTATCGGG-3’
R5’TGCTGGGGATTTCAGGAAAG-3’ 372bp 540C
st B F5’GGCACCCATAAGCGTACGCC-3’
R5’TGCCGAGCACGGCGACTACC-3’ 470bp 640C
INC A/C F5’GAGAACCAAAGACAAAGACCTGGA-3’
R5’ACGACAAACCTGAATTGCCTCCTT-3’ 465bp 640C
IS26 F5’-AGCGGTAAATCGTGGAGTGA-3’
R5’-AGGCCGGCATTTTCAGCGTG-3’ 120bp 600C
Table 3.1: List of primer
4 CHAPTER FOUR: RESULTS
4.1 Antimicrobial susceptibility Trends (2006-2015)
Antimicrobial susceptibility patterns among V.cholerae isolates keep changing due to emergence of strains conferring resistance to multiple antibiotics.
Presence of resistance to each antimicrobial drug among V. cholerae isolates was noted in different years, between 2006 and 2015 (Fig 4-2). Apart from tetracycline drug whose susceptibility was noted among all isolates, all other antimicrobials reported some level of resistance as illustrated.
Figure 4.2: Percentage of isolates resistant to β lactam antibiotics between 2006 and 2015
Resistance to ampicillin was noted in 2006 in very few isolates (4%). However, isolates obtained in 2007 and 2009 did not exhibit resistance to this antimicrobial. Then, few isolates (7%) resistant to this drug reappeared in 2010. The prevalence of such strains continued to rise by 77 % in 2012 and a further 2 % in 2015. Nevertheless, the reason behind the high number of isolates resistant to ampicillin in 2012 and 2015 (p: <0.001 OR: 0.0057, CI: 0.0012-0.0274) as compared to other years is still not clear.
Resistance to amoxicillin/clavulanic acid (amoxyclav) varied over the years. For instance, in 2006, 74% of isolates were resistant to this antimicrobial but this declined by 28% in 2007 with the numbers declining even further by 36% in 2009 (p: 0.045, OR: 17.69, CI: 2.02-154.2). However, in 2010, the number of isolates resistant to this antimicrobial increased by 6% and this increased by 65% in 2012 with 2015 reporting the highest number of isolates (90%) resistant to this drug Figure 4.2.
Isolates obtained in 2006, 2007 and 2009 did not exhibit resistance to third generation cephalosporins (cefotaxime, cefpodoxime, ceftazidime) (Fig 4.2). The first isolates resistant to cefpodoxime and ceftazidime emerged in 2010 and these comprised the 7% of isolates obtained from this year. In the same year, 14% of isolates were resistant to cefotaxime. Resistance to these third generation cephalosporin continued in 2012 where it increased by 70% in cefotaxime and cefpodoxime. An increase of 3% in resistance to ceftazidime, from 7% in 2010 to 10.5% in 2012 was also noted in 2012 but the reason behind the slight increase as compared to other cephalosporins remains unknown. Resistance to ceftazidime continued to rise by 28 % from 10.5% in 2012 to 38% in 2015. This was however not the case for the other two third generation cephalosporins (cefotaxime, cefpodoxime) where a decline in resistance of up to 62% was noted for cefotaxime and a decline of 66% noted for cefpodoxime from 2012 to 2015. This decline could be a result of absence of isolates producing ESBLs phenotypically though this remains an assumption.
Resistance among the fluoroquinolones ,that is, ciprofloxacin and nalidixic acid varied among the isolates (Fig 4.3).
Figure 4.3: Percentage of isolates resistant to fluoroquinolones over the years
Isolates collected in 2006 ,2007,2009,2010 and 2012 were susceptible to ciprofloxacin which is a broad spectrum fluoroquinolone. However a few isolates (34%) which were intermediate resistant to this drug emerged in 2015. On the other hand, all isolates in 2006 and 2007 were resistant to nalidixic acid which was followed by a decrease of 18% from 100% in 2007 to 82% in 2009. This difference was however not statistically significant (p: < 0.199, CI: 0.9-16) All isolates in 2010 were resistant to this drug. More than 90% of isolates in 2012 and 2015 were resistant to this drug. All isolates in the ten year period reported resistance of not less than 80% to nalidixic acid indicating that V.cholerae isolates are highly resistant to nalidixic acid and this resistant has persisted over time.
Resistance patterns among the aminoglycosides, gentamycin and streptomycin varied over the ten year period (Fig 4.4).
Figure 4.4: Percentage of isolates resistant to Aminoglycosides over the years
None of the isolates obtained between 2006 and 2009 were resistant to gentamycin. However, 7% of isolates obtained in 2010 were resistant to this antimicrobial. This resistance increased by 77% in 2012. The reason behind this sharp increase is still not clear though this was the same year where isolates producing ESBLs were identified. This was however different for streptomycin whereby all isolates in 2006,2007,2009,2010 and 2012 were resistant to this drug. Even though isolates in 2008 and 2011 were not represented in this study, this resistance could still have persisted in these years. There was however a decline in resistance in 2015 by 52% (p<0.001).
Isolates in 2006 and 2007 were all resistant to SXT (Fig 4.5).
Figure 4.5, Percentage of isolates resistant to SXT between 2006 and 2015
It is still not clear how the resistance pattern progressed in 2008 but in 2009, a decrease to 91% from 100% in resistance was noted. This resistance then increased to 100% in 2010 and persisted in 2012 whereby all isolates were resistant to SXT. However, a significant increase in susceptibility from 100% in 2012 to 8% in 2015 was noted (p: <0.001, OR: 184.3, CI: 23.4-1450).
Thirty (30 %) of isolates in 2006 were resistant to chloramphenicol (Fig 4.6). This resistance increased steadily in 2007 by 8% through to 2009 by 7%.
Figure 4.6, Percentage of isolates resistant to chloramphenicol between 2006 and 2015
The number of isolates resistant to this antimicrobial increased from 30% in 2006 to 38.5% in 2007 then 40% in 2009 and 2010. A sharp increase from 43% in 2010 to 89% in 2012 thus reporting the highest number of isolates resistant to chloramphenicol over the ten year period (p: <0.001, OR: 0.0325, CI: 0.07-0.1504). This was the same year when isolates producing ESBLs were identified but it is still not clear if resistance to chloramphenicol is linked to the ESBL phenotype. The year 2015 reported the least number of isolates resistant to chloramphenicol where a decrease from 89.5% in 2012 to 2 % in 2015 was noted (p: <0.001, OR: 49, CI: 6.44 – 372).
.
4.1.1 Measurement of residual zones of inhibition to antimicrobial drugs by stored strains
Resistance to antimicrobial drugs was determined by measuring zone of inhibition in millimeters. A large zone diameter around an antimicrobial disc indicated increase in susceptibility while a smaller one indicated resistance. There are standardized zones of inhibition for each antimicrobial drug.
Isolates that were sensitive to ampicillin had a zone diameter of more than 17mm while intermediate isolates had a zone diameter of between 14mm-16mm and the resistance ones had a zone diameter of less than 13mm (Fig 4.7).
Figure 4.7, measurements of zones of inhibition to Ampicillin among V. cholerae isolates over the years
Majority of isolates in 2006 were clustered in the susceptible region with zone diameters between 17mm to 21mm with only one isolate in the intermediate zone. A similar pattern was observed with the 2007 isolates. However, all isolates in 2009 and 2010 were clustered in the susceptible region with zone diameters ranging between 17mm to 22mm. A completely different pattern was observed in 2012 isolates where majority (16) of isolates were in the highly resistant region with zone diameters of 6mm and only a small number was in the susceptible region with zone diameters ranging between 20mm to 22mm. isolates in 2015 were distributed among the three regions having zone diameters ranging from 7mm to 21 mm with the majority in the intermediate zone Fig 4.7.
Sensitive isolates to Amoxyclav were those that had a zone diameter of more than 18mm while those in intermediate zone had a zone diameter of between 14mm-17mm and the resistance strains had a zone diameter of less than 13mm (Fig 4.8).
Figure 4.8, measurements of zones of inhibition to Amoxyclav among V.cholerae isolates over the years