Antibiogram, Haemolysis and Biofilm Properties of Escherichia coli from Bovine Mastitis

Mastitis is the most economically significant diseases of dairy animals. It is widespread in dairy herds and is associated with a significant reduction in milk yield, increased costs of production and deteriorated milk quality. Mastitis triggered by E. coli is usually sporadic and clinical signs vary from very severe or even fatal forms to mild mastitis in which cows have only local signs in the udder (Sikrodia et al., 2020). While the severity of the disease depends on host immune response and genetic makeup, virulence of the bacterial strains involved may also play a role (Fernandes et al., 2011). The point sources of coliform bacteria that cause infections include bedding materials, soil, manure and other organic matter in the environment of cows. The portal of entry into the mammary gland for Gram-negative bacteria is the teat canal (Hogan and Larry, 2003). Incidence of coliform mastitis increases during climatic periods that maximize populations in the environment.

A high incidence of clinical and sub-clinical mastitis in cows due to Gram positive cocci was reported from the Malwa region of India (Ghose et al., 2001, 2003). However no studies had been done to study coliform mastitis in this area. Hence, the present study was undertaken.

A total of 300 milk samples were collected aseptically from cows from various organised and unorganised herds. All samples were processed and tested in the Department of Veterinary Microbiology, College of Veterinary Science and A.H., Mhow. The pooled milk samples were subjected to california mastitis test (CMT) (Schalm et al., 1971) for the diagnosis of subclinical mastitis (SCM), whereas clinical mastitis was detected by the examination of udder. The samples found positive for clinical and subclinical mastitis were included for further studies.
 
Isolation and identification of E. coli
 
Milk samples from clinical cases and those showing +2 or +3 reactions in CMT, were cultured bacteriologically to isolate and identify E. coli bacteria. Each sample was inoculated in Brain heart infusion (BHI) broth (Hi Media) and incubated aerobically at 37°C for 18-24 hrs. Subsequently a loopful of bacterial growth from BHI broth was on Eosin Methylene Blue (EMB) agar (Hi Media) and incubated aerobically at 37°C for 24 hrs. A single, well isolated colony with metallic sheen on EMB agar after confirming as Gram negative rods was transferred on nutrient agar (NA) slant. The NA slants were incubated aerobically at 37°C for 24 hrs and thereafter stored at 4°C for the preservation of isolates.

The presumptive identification of bacterial isolates as E. coli was accomplished by colonial and bacterial morphology and confirmed by a battery of biochemical tests, viz- ONPG, catalase, oxidase, MR, VP, urease, nitrate reduction, indole, citrate utilization, carbohydrate fermentation and growth on triple sugar iron agar as per the procedure described by Barrow and Feltham (1993).

The bacterial isolates identified as E. coli on the basis of cultural, morphological and biochemical characteristics were sent to National Salmonella and Escherichia Centre, Central Research Institute, Kasauli (H.P.) for serotyping of somatic (O) antigen.
 
Antibiotic sensitivity test of E. coli
 
In vitro antibiotic sensitivity test (AST) of E. coli isolates was conducted as per the method of Bauer et al., (1966) using antibiotic discs (Hi Media) of amikacin, amoxicillin, ampicillin, cefixime, cefoperazone, ceftriaxone, cefotaxime/clavulanic acid, cefotaxime, chloramphenicol, ciprofloxacin, gentamicin, kanamycin, nitrofurantoin, tetracycline and trimethoprim. Diameters of the clear zone of inhibition around antibiotic discs were measured in mm. The interpretation of the result was made in accordance with the instructions of the manufacturer. 
 
Virulence factors of E. coli
 
Haemolysis
 
The haemolytic activity of E. coli was studied on blood agar plate as per Agarwal et al., (2003). Zone of complete haemolysis (β hemolysis) around the bacterial colonies after 24 hrs incubation was noted as strongly positive and the zone which appeared after incubation of two days was recorded as weakly positive. Alpha haemolysis was characterized by a hazy zone of partial haemolysis (green discoloration), containing a proportion of unlysed cells and less clearly demarcated from the surrounding medium than β haemolysin activity.
  
Biofilm formation
 
A qualitative assessment of biofilm formation by tube method was determined as per the method described by Mathur et al., (2006). The culture was inoculated in 5 ml Tryptone soya broth (TSB) (Hi Media) and incubated for 12 hrs at 37°C after which 50% of spent media from each tube was gently aspirated and replaced by equal quantity of fresh TSB, with sucrose (0.25%). The tubes were reincubated at 37°C for 12 hrs. Subsequently, the broth was decanted and tubes were gently washed twice with PBS, kept inverted for drying and stained with 0.1% safranin solution. Excess stain was removed and the tube was gently washed with deionised water. Tubes were then dried in an inverted position and observed for biofilm formation.
 
Congo red dye binding assay
 
The test was performed as an indicative of invasiveness following the method of Ishiguro et al., (1985). The isolates were streaked on tryptone soya agar media (Hi Media) containing 0.03% Congo red dye and incubated for 48 hrs at both 37°C and 25°C. A positive reaction was indicated by the appearance of intense orange or brick red colonies. A negative result was evidenced by pale or white colonies.
 
Molecular characterization of fimH and pap genes                  
                 
The isolates were cultivated by inoculating in BHI broth (Hi Media) and incubating at 37°C for 12-18 hours (overnight grown culture). DNA was extracted using a kit (HipurA Bacterial genomic DNA purification kit, HiMedia) as per the manufacturer’s instructions. The PCR was performed for the amplification of fimH and pap of E. coli using suitable primers (Fernandes et al., 2011) (Table 1). It was carried out in final reaction volume of 12.6 µl using master-mix (Sigma) in 0.2 ml thin wall PCR tube (Table 2).



​

The DNA amplification reaction was performed in Thermocycler (Eppendorf Research, Germany) with a pre-heated lid. The cycling conditions for PCR included the steps shown in Table 3. PCR products were kept at -20°C until further analysis by agarose gel electrophoresis. The amplified PCR products (10 µl) were electrophoresed in a 1.5% agarose gel (in TBE buffer), stained with ethidium-bromide solution.

​

In the present study 121(40.33%) out of 300 milk samples from cows were found positive for mastitis. The incidence of mastitis was higher in unorganized farms (42.70%) as compared to organized farms (36.11%) (Fig 1). The overall prevalence reported in the present study is in close agreement with the results of Birhanu et al., (2017), but is lower than the findings of Zeryehun and Abera (2017), who reported a prevalence of more than 50%. The difference in the prevalence of mastitis could probably be due to differences in farm management practices, breed and age of animals, production status, stage of lactation, season etc.



Out of 121 mastitis positive samples, E.coli was isolated from only 30 (24.79%) samples (Fig 2). Sangeetha et al., (2020) also reported 25.7% incidence of E. coli from cases of mastitis, but Sharma et al., (2015) and Sudheer et al., (2019) reported a higher rate of isolation.

​

Amongst 30 E.coli isolates, 21 were typed into 8 different ‘O’ serogroups, while the remaining 9 were untypable. The most common serogroup was O83 (43.33%) followed by O157 (6.66%). The other ‘O’ serogroups isolated in the present study were O8, O20, O49, O119, O128 and O145 (3.33% each). The serotypes O8, O20, O49, O83, O119, O128 and O157 have been isolated from earlier also from bovine mastitis in different frequencies (Iguchi et al., 2015; Sharif et al., 2017). O157 serotype recovered in the present study is one of the most important STEC that causes severe diseases in humans. However, until its ‘H’ antigen is also characterized, it will be too early to implicate pathogenicity of present E. coli isolates of O157 type in human diseases.

The results of antibiogram studies (Fig 3) revealed that ciprofloxacin (93.33%) was most effective drug against E. coli, followed by gentamicin (80%), tetracycline and nitrofurantoin (76.66%) and chloramphenicol (63.33%), amikacin and cefixime (46.66%) and trimethoprim (43.33%). On contrary, they were resistant to amoxicillin (96.66%), kanamycin (93.33%), cefotaxime/clavulanic acid (73.33%), ampicillin/sulbactam (66.64%), cefixime and cefotaxime (53.33% each) and ceftriaxone (43.33%). Our results are in corroboration with reports of Perez et al., (2017) and Sikrodia et al., (2020) who also reported ciprofloxacin to be most effective against E. coli isolates from bovine mastitis. However, Puvarajan et al (2020) recorded low (38.0%) sensitivity for ciprofloxacin and a high sensitivity to ceftriaxone and cefotaxime in E. coli causing bovine mastitis, which is contradictory to the present report. Sudheer et al., (2019) observed high resistance to cephalosporins in E. coli causing bovine mastitis. Tetracycline resistance was found in 23.64% of isolates, which is similar to report of Marashifard et al., (2019), but contradictory to those of Chandrasekaran et al. (2015). The antimicrobial resistance pattern of the bacterial population in the cow’s environment can vary between herds, reflecting the quantitative and qualitative aspects of antimicrobial treatments. The use of antimicrobials may also select bacteria with virulence factors linked to antimicrobial resistance (Lehtolainen et al., 2003).

​

Multiple drug resistance (MDR) was observed in 96.67% of the isolated E. coli strains. The percentage of MDR isolates recovered in the present study is in close association with the studies done by Srinivasan et al., (2007) and Jena et al., (2014) who reported 90.7 and 100 percent in MDR strains of E. coli, respectively.

Antibiotic-resistant bacteria pose a severe challenge to both clinicians and dairy animal producers because they have a negative impact on therapy. Development of resistance has been attributed to the extensive therapeutic use of antimicrobials (Abo-Shama 2014) as exemplified by high resistance towards beta-lactam antibiotics and low towards tetracyclines, in our study. The usage of antibiotics correlates with the emergence and maintenance of antibiotic resistant traits within pathogenic strains. These traits are coded by genes that may be carried on the bacterial chromosome, plasmids, transposons or on gene cassettes that are incorporated into integrons (Daka et al., 2012) and thus are easily transferred among isolates.

E. coli species is very diverse comprising of commensal as well as pathogenic strains clustered in different pathovars based on clinical data and specific virulence properties (Kempf et al., 2016). Pathogenicity of strains is conditioned by a specific repertoire of virulence factors located on the mobile genetic elements and transmitted by horizontal gene transfer (Baidy-Chudzik et al., 2015).

Among 30 isolates, 14 were found to be haemolytic. Haemolysins are identified as important virulence factors of enterohaemorrhagic E. coli (EHEC), which also produce verotoxin, verocytotoxin or shiga toxins affecting the cell membrane.

Biofilm production was demonstrated by 23.33% E. coli isolates. Biofilms are highly organised communities of microorganisms structured within an array of exopolysaccharides (EPS) and adhering to a living or inert surface with the function of protecting the microorganisms in stress environments. The biofilm potentially plays an important role in the development of persistent infections as well as recurrent clinical symptoms after antibiotic therapy despite quite good in vitro antimicrobial susceptibility of the agent. They are associated with antimicrobial treatment failure (Melchior et al., 2006).

During the present investigation, total of 12 (40.0%) isolates were found positive for the Congo red binding test. The results are in accordance with Lamey et al., (2013) who reported 38.1% E.coli isolates from bovine mastitis positive for Congo red binding activity. The ability to bind CR dye has been proposed as a marker for the invasive property of E. coli (Sharma et al., 2006).

The severity of mastitis in bovine and the pathogenicity of E. coli are greatly affected by the presence of genes coding virulence factors. The organism produces a large number of potential virulence factors, such as capsule, biofilm production and pili, which have important roles in the pathogenesis and colonization in mammary gland; Type 1 fimbriae are the most common adhesive organelles of E. coli, which mediate the adhesion of the organism to the host’s mannose-containing glycoproteins (Dubravka et al., 2015). In the present study fimH gene was demonstrated in nine isolates (Fig 4). The fimH is an important virulence associated gene associated with the expression of curli fimbriae that has an influence on biofilm formation (Dubravka et al., 2015).  Similar to this study the detection of fimH gene was also reported by Dogan et al., (2006) and Fernandes et al., (2011). However, pap gene was not found in any of the isolates as also reported earlier (Fernandes et al., 2011), but contrary to those of Kaipainen et al., (2002). A low prevalence of virulence genes in E.coli associated with bovine mastitis was also recorded by Marashifard et al., (2019). Thus the present results indicate that the pathogenicity of E. coli in bovine mastitis is not a consequence of specific virulence factors. Only isolates with successful combinations of virulence factors will be capable of causing disease.

​

The authors are thankful to the Dean, College of Veterinary Science, Mhow for providing facilities. First author is grateful to the Government of Madhya Pradesh for granter her study leave.