Prevalence Of Antimicrobial Resistance And Characterization Of Fecal Indicator Bacteria And Staphylococcus Aureus From Farm Animals, Wildlife, Pets And Humans In Tanzania

ABSTRACT

This thesis is prepared according to “Publishable manuscript” format of the Sokoine

University of Agriculture. It is divided into four sections; the first section presents

introduction, problem statement, objectives, rationale of the study and hypothesis while

the second section is a review of literature in relation to the work performed. The third

section contains four manuscripts which cover the research work done, whereas the last

section details overall conclusions of the study. The first manuscript is about the findings

on the awareness of risks associated with antimicrobials use and practices that contribute

to selection for antimicrobial resistance in livestock. The second and the third

Manuscripts present the prevalence of antimicrobial resistance in indicator bacteria

namely S. aureus from nostrils, E. coli and Enterococci spp from faecal samples obtained

from livestock, dogs and contact human beings, and the last manuscript presents findings

on antimicrobial resistance in faecal indicator organisms from buffalo, wildebeest, zebra

and cattle grazing with wild ungulates. The studies involved testing resistance using

antibiotic discs, inclusion of antibiotics in the media and evaluating bacterial DNA for

resistance genes against commonly used antibiotics in livestock. In response to the first

objective presented in manuscript 1, awareness of human health risks due to use of

antibiotics among livestock keepers and factors that contribute to selection of antibiotic

resistance bacteria within livestock in Tanzania were determined. Looking at awareness

on health risks among the livestock keepers surveyed, one hundred and sixty randomly

selected Tanzanian small-scale livestock keepers were involved in the questionnaire-based

survey. The results showed that 30% of the respondents were not aware of the concept of

antibiotic resistance whereas 52% were not aware of the types of diseases that can be

iii

treated with antibiotics, 22% did not know the essence of withdrawal period after

treatment of animals with antibiotics and 40% did not know if antibiotics used in livestock

pose risk to human health. Side effects of antibiotics to human beings known by the

respondents include drug resistance, allergy, poisoning, skin rashes and cancer. Further, it

was found that the most commonly used antibiotics were oxytetracyline (TE) (62.9 %)

followed by sulphadimidine/ trimethoprim / (STX) (23.2 %), penicillin-streptomycin (13.4

%) and gentamycin (CN) (0.5 %).Only 54% of the respondents obtained their antibiotics

through prescription by veterinarians. Potential risk factors were identified to be the

sources of development of antimicrobials resistance. These included livestock

management systems, antibiotics handling and types of antibiotics used in the study areas.

Points of selling antibiotics and their prescriptions were found to be handled by agriculture

and veterinary input shops, veterinarians, livestock markets and exhibition areas. The

study concludes that there is a gross lack of awareness on health threats associated with

antibiotics resistance among the farming communities thus posing serious public health

threats in Tanzania.

On the other hand, prevalence of nasal carriers, population structure and antimicrobial

resistance of Staphylococcus aureus and S. pseudintermedius in healthy humans, pigs and

dogs in Tanzania were investigated in the same study area as the second objective and

manuscript 2 of the study. Nasal swabs were taken from healthy humans, pigs and dogs

(n=100 per species) in Tanzania and examined for growth of S. aureus on the Oxacillin

(OX) Resistance Screening Agar Base (ORSAB) media. Presumptive S. aureus isolates

were confirmed by nuc PCR tested for the presence of mecA, mecC and were spa-typed

while antimicrobial susceptibility patterns were determined by disc diffusion, S. aureus

was isolated from 22% of humans, 11% of dogs nasal bacteria isolates and 4% of pigs. S.

pseudintermedius was isolated from 4% of dogs. A total of 21 spa types were identified:

13 from human isolates, 7 between dogs and one among pig isolates. Three spa types

iv

(t314, t223 and t084) were isolated from both humans and dogs. Disc diffusion test

revealed resistance to sulphamethoxazole/trimethoprim (STX) (3/37; 8.1%), rifampicin

(RD) (2/37; 5.4%), amoxycillin-clavulanic (AMC) acid (1/37; 2.8%), oxacillin (OX)

(4/37; 10.8%), chloramphenicol ( C) (3/37; 8.1%), gentamicin (CN) (3/37; 8.1%) and

penicillin (P) (37/37; 100 %). All the four isolates resistant to oxacillin were negative by

PCR for mecA but positive for mecC. mecC encoded Methicillin Resistant Staphylococcus

haemolyticus (MRSH) was detected in pigs, dogs and humans. The study presented the

first report on strains with this gene in Tanzania. Moderate levels of antimicrobial

resistance were observed irrespective of the host species from which the strains were

isolated.

In addition, the Prevalence of antimicrobial resistance in indicator bacteria isolated from

faecal samples of livestock, poultry and humans in Tanzania was studied as the third

objective and presented in manuscript 3. In order to quantify the public health risks posed

by handling animals, their waste and animal products, a total of 587 faecal samples were

collected from humans, pigs, beef, dairy cattle and exotic and indigenous chickens to

investigate the relatedness of antibiotic resistant bacteria from humans and animals in

Tanzania. A sum of 485 E. coli isolates were obtained from faecal samples grown on

MacConkey agar plates. Also, tetracycline (TE), s sulphadimidine/ trimethoprim (STX),

ampicillin (AMP) and cefataxime (CTX) were included in the media to test the resistance

of E. coli isolates, the results showed higher resistance to TE (385, 79.4%), STX (374,

77.1 %) and AMP (362, 74.6%) and less resistance was observed in CTX (194, 40%).

Further, antibiotic discs were used to determine the resistance of 350 E. coli isolates,

whereas higher resistance was also observed to TE (241, 68.8%), STX (214, 61.1%), S

(195, 55.7%) and AMP (192, 54.8%). Moreover, resistance to CTX was observed in 143

isolates (40.9%) and ciproflaxin (CIP) 30.9 % (108) while less resistance was observed in

AMC (48, 13.7%) and CN (37, 10.6%). Besides E. coli, a total of 550 Enterococci spp

v

isolates were also speciated and tested for resistance. Out of the 164 enterococci isolates

tested, 151 (92.1%) were E. faecium, 5 (3.1%) E. faecalis and 8 (4.9%) E. hirae. Growth

of resistant indicator bacteria colonies was carried out in AMP and VA in which higher

resistance was observed in AMP (85, 15.5%) as compared to vancomycin (VA) (30,

5.5%). For 240 Enterococci spp tested with antibiotic discs, higher resistance was

observed to rifampicin (RD) (174, 75.2%), TE (33, 13.8%) and e rythromycin (E) (36,

15%) whereas; there was no resistance to CN (120μg) and C (5, 2.1 %). Other antibiotics

tested for this group included CIP (9, 3.8%), AMP (12, 5%) and VA (12, 5%). Also, sulII

and tetW resistance genes were found in all livestock and humans samples (n=22 per

species) with degree of variations in when calculating the proportion of resistance genes in

relation to number of bacteria estimated from “total” 16s RNA. For all the sample

percentages of tetW to the total of 16s genes were 0.06 % Humans, 11.88% Cattle, 2.4%

Pigs, 2.15% local chickens and 0.98% exotic chickens; whereas, sulII represented; 0.22%

Humans, 0.34 % Cattle, 0.2%Pigs, 0.3% local chickens and 0.02% exotic chickens. All

isolates (n=194) which showed resistance to CTX on MacConkey agar (59 of which were

non-lactose fermenters) revealed a band of the same size to that of a positive CTX-M

control strain in all isolates analyzed, indicating that they all contained an ESBL gene of

this class. Antibiotics that are commonly used in Tanzania had higher resistance as

estimated from both colony forming unit (CFU) of resistant bacteria and resistance

profiles determined by the use of antibiotic discs. Resistance genes were also

demonstrated to commonly used antibiotics (TE, STX) in the country. The findings of this

study indicate that prolonged use of TE and STX has caused indicator bacteria to develop

resistance against these antimicrobials and the resistance genes observed for the same

antibiotics suggest that the genes circulate between animals and humans sharing the same

environment.

vi

Furthermore, the forth objective presented in manuscript 4 was to determine the

prevalence of antimicrobial resistance in faecal indicator organisms from buffalo,

wildebeest, zebra and cattle grazing with wild ungulates in the study area. The study

investigated the antibiotic resistance in E. coli and Enterococci spp isolated from faecal

samples of wild ungulates and cattle grazing with them. Faecal samples were collected

from 35 buffalos, 40 wildebeest and 40 zebra from two study sites namely Mikumi

National Park (MNP) and Ngorongoro Conservation Area (NCA).Also, additional 20

faecal samples were obtained from cattle grazing together with the wild ungulates in NCA

in order to determine whether co-grazing with cattle constitutes a risk factor for carriage of

antibiotic resistant bacteria. AMP, STX, TE and CTX resistant E. coli were observed in

52.6%, 39.2 %, 31.9% and 17.8% of wildlife samples respectively. The mean of log10CFU

antibiotic resistant bacteria did not differ significantly between the three wildlife species.

Log10 CFU of ampicillin resistant E. coli was significantly higher among wildlife grazing

together with cattle, whereas, cefotaxime resistance was more common in samples from

MNP. In antibiotic discs test of antimicrobials in the samples, one hundred and twenty E.

coli colonies were purified and antibiotic resistance profiled. Wildlife isolates from MNP

(without cattle) were more frequently resistant to TE, CN and ENR than isolates from

NCA (with cattle). Notably, E. coli resistant to CTX and ENR were observed among the

randomly obtained isolates from all wildlife but not from cattle. Out of 120 enterococci

isolates speciated, 95 (79.2%) were E. faecium, 9 (7.5%) E. faecalis, 7 (5.85) E.

gallinarum and 9 (7.5%) E. hirae. On the other hand, AMP and VA resistant Enterococci

were detected in 7.41% of wildlife samples but not in cattle and one hundred and twenty

eight purified colonies of Enterococci spp were antibiotic resistance tested. Isolates for

wild animal grazing together with cattle showed higher levels of resistance to STX than

cattle without. While,the average ratio of sulII gene-copies to 16s rRNA copies in faecal

samples from wildlife did not differ between the two sampling sites, the ratio of tetW to

vii

16s rRNA was significantly higher in samples from MPN than from NCA. In conclusion,

grazing together with cattle was not found to be a risk factor for the presence of antibiotic

resistant bacteria in faecal samples of wild ungulates.

Overall findings document for the first time in Tanzania that livestock keepers are

unaware of antimicrobial resistance, uncontrolled antibiotic dispensing for animals

presents high risk for spread of antimicrobial resistance among domestic animals with

potential to contribute antimicrobial resistance to human beings and moderate levels of

antimicrobial resistance observed in S. aureus isolates irrespective of the host species

whose strains were isolated. Also different spa types of S. aureus circulating in Tanzania

and for the first time detection of mecC encoded MRSH from S. haemolyticus in pigs,

dogs and humans is reported in Tanzania. However, the use of antimicrobial resistant

bacteria from farm animals and human beings and similarities in the resistance by

phenotyping and genotyping were shown to tetracycline and sulphamethazole which are

commonly used antimicrobials in Tanzania, the genotyping showed the similarity in genes

coding for TE and STX in both human and farm animals isolates. Also the study in

wildlife species showed that wild ungulates are reservoir of antibiotic resistance bacteria

and resistance genes thus carrying risk of transfer of resistant genes to humans and

livestock in the livestock-wildlife interfaces of MNP and NCA and humans are at public

health risks by sharing the same environment with pets and farm animals. Further studies

in molecular typing of the isolates obtained from humans, dogs and domestic animals

sharing the same environment also cattle and wild ungulates grazing together in MNP

NCA are indicated.

With regard to the general objective of the study; “to isolate, characterize and investigate

the extent of antibiotic resistance among selected bacteria (Escherichia coli, Enterococci

spp and Staphylococcus aureus) in humans, livestock, dogs and selected wild animals”,

viii

the findings sufficiently and positively addressed the objective in terms of isolation,

characterization and extent of antibacterial resistance in targeted species.