Comparison of beta-lactamase genes in clinical and food bacterial isolates in India

Background: The present study aimed to determine the occurrence of human disease-causing enteric bacteria on raw vegetables, fruits, meats, and milk products sold in Indian markets. The study further aimed to analyze antibiotic resistance rates and the presence of blaCTX-M, blaTEM, blaSHV, and blaAmpC. Methodology: Twenty-three food-borne and 23 clinical isolates were compared for antibiotic resistance rates and the presence of blaCTX-M, blaTEM, blaSHV, and blaAmpC. Swabs were taken from unwashed and washed food items, as well as from some chopped food specimens, and inoculated on appropriate culture medium. Bacterial isolates were identified, antibiotic susceptibility was performed, and bla genes were detected by PCR. Results: Thirty-eight bacterial isolates were obtained from the food specimens, of which 36 (94.7%) were Gram-negative and two (5.3%) were Gram-positive bacterial species. Klebsiella pneumoniae was the most prevalent (52.6%; 20/38) bacterial species isolated, followed by Citrobacter koseri (18.4%; 7/38). In food isolates, the majority of the isolates were resistant to gentamicin (33.3%) followed by amikacin (11.1%). Resistance to a third-generation cephalosporin was noticed in only 5.6% isolates. However, in clinical isolates, maximal resistance was noticed against third-generation cephalosporins followed by ofloxacin in 91.3% and 86.9% isolates, respectively, and resistance to gentamicin and amikacin was noticed in 78.3% and 52.2% isolates, respectively. The presence of blaCTX-M, blaTEM, blaSHV, and blaampC in clinical isolates was noticed in 52.2%, 60.9%, 21.7%, and 43.5%, respectively. None of the isolates from food showed the presence of any of the above-cited genes. Conclusions: Probably bla genes have not yet disseminated to raw-food vegetation in India.


Introduction
Raw vegetables are widely consumed as salads and in other dishes in many countries, including India.In most instances these vegetables are washed with water and not treated with any antimicrobial/disinfectant solution, such as a solution of potassium permanganate.This is particularly true for food-consuming habits in India.
There are reports from many parts of the world indicating bacterial contamination of raw vegetables [1].Furthermore, there are also reports that implicate food as a source of resistant bacteria, raising concerns of disease transmission when these food items are consumed raw [2,3].However, reports looking for the occurrence of bla genes in food-associated enteric bacteria are elementary, particularly in India.
This preliminary study was performed to look for the occurrence of human disease-causing enteric bacteria on raw vegetables and fruits sold in Indian markets by taking swabs from unwashed food items to analyze the bacterial occurrence and load before and after washing them with sterile water (mimicking a condition of consumption of salads).Moreover, we also attempted to look for the occurrence of bacterial contamination in chopped food items.The Gramnegative bacterial isolates obtained from food items were further studied for antibiotic resistance rates and the presence of bla CTX-M , bla TEM , bla SHV , and bla AmpC .In addition to food items of plant origin, other foods, including minced meat, chopped meat and ice cream (as a source of animal-origin food) and sweets (Jalebi), which are sold openly in unhygienic condition in Indian markets, were also analyzed in a similar manner.A small subset of pathogenic bacteria obtained from clinical specimens was also included in the study to compare the results with bacterial isolates of food origin.

Materials and Methods
The present prospective study was conducted in the department of Microbiology of Jawaharlal Nehru Medical College and Hospital, Aligarh Muslim University, UP, India, during August 2008 to October 2008.

Food-borne specimens
The detailed algorithm showing various foodborne specimens used in this study as well as their processing steps is shown in Figure 1.All the food items were collected in sterile polythene bags and transported to the laboratory immediately for further processing.

Specimen processing
Swabs were taken from unwashed and washed food items by rubbing and rolling a sterile swab over a rectangular section of 5.0 cm x 2.5 cm from various surfaces of food items.In the case of coriander and mint leaves, the swabs were collected from a surface of 2.5 cm x 2.5 cm from both the surfaces.These swabs were inoculated over blood agar (BA) and McConkey agar (McA) for bacterial culture and the plates were incubated overnight at 37 o C in aerobic conditions.The detailed descriptions of food items processed by taking the swabs are shown in Figure 1.After taking the swabs, the washed food items were chopped in sterile petriplates and 1 gm of chopped material was inoculated in 10 ml of sterile nutrient broth and incubated for six hours at 37 o C. A loopful of broth was then subcultured over BA and McA for bacterial identification.Colonies grown on the plates from unwashed and washed specimens were counted and analyzed for the elimination or reduction of bacterial load in washed specimens.
Minced meat, chopped meat, ice cream, and Jalebi were first inoculated in sterile broth and processed in the same manner described above.

Clinical Specimens
Clinical samples of pus and urine submitted to the Department of Microbiology of Jawaharlal Nehru Medical College and Hospital of Aligarh Muslim University, during the specified period of this study, were also included for comparative analyses of the occurrence of bla genes.Clinical specimens were subjected to direct microscopy and culture on Blood Agar and McConkey Agar, and culture plates were incubated aerobically for 18 hours at 37 o C. A small subset of bacteria (E. coli = 17, K. pneumoniae = 6) Algorithm showing various food-originated specimens used in this study as well as their processing steps.
obtained from urine and pus cultures were included for further study.

Bacterial identification
All the bacterial isolates (both Gram-negative and Gram-positive) were identified to the species level based on catalase test, oxidase test, motility, coagulase test, carbohydrate fermentation, and other specific biochemical tests according to the standard protocol [4].

Phenotypic detection of extended-spectrum betalactamases (ESBLs)
ESBL production was inferred phenotypically by combination disk method using cefoperazone (Cs) and cefoperazone/sulbactam (Cfs) disks.An increase in zone diameter of > 5 mm in Cfs in comparison to Cs alone was taken as potential ESBL production.

Preparation of DNA template
Template DNA from clinical and food-borne bacteria was prepared from freshly cultured bacterial isolates by suspending 3-5 colonies in 50 µl of molecular grade water, and then heating at 95°C for 5 minutes and immediately chilling at 4°C.Positive controls harboring bla CTX-M , bla TEM , bla SHV , and bla AmpC and negative control (E. coli ATCC 25922) were processed in the same way for DNA extraction.

Detection of bla genes by PCR
Molecular detection of bla CTX-M , bla TEM , bla SHV , and bla AmpC was performed in all the food and clinical isolates by using polymerase chain reactions (PCRs) according to the methods described previously with  minor modifications [6,7].The primers used for respective detection of these genes are shown in Table 1.The cycling conditions for detection of bla AmpC , bla TEM , bla SHV were as follows: initial denaturation at 95 C for 15 minutes; 35 cycles of 94 C for one minute; 58 C for 2 minutes; 72 C for 3 minutes; and a final elongation at 72 C for 10 minutes.However, the cycling conditions for bla CTX- M detection were as follows: initial denaturation at 94 C for 7 minutes; 35 cycles of 94 C for 50 seconds; 50 C for 40 seconds; 72 C for one minute; and a final elongation at 72 C for 5 minutes.PCR products were analyzed by gel electrophoresis after running 5 µl products in 2% agarose gels.Controls harbouring bla CTX-M , bla TEM , bla SHV , and bla AmpC were used as molecular weight markers.

Bacterial isolates obtained from food specimens
A total of 38 bacterial isolates were obtained from 23 food-borne specimens.Out of the 38 bacterial isolates, 36 (94.7%) were Gram-negative and 2 (5.3%) were Gram-positive species.Klebsiella pneumoniae was the most prevalent (52.6%; 20/38) bacterial species noticed, followed by Citrobacter koseri (18.4%; 7/38).S. aureus was isolated from chopped meat and Enterococcus faecalis from the sweet specimen.The details of the bacterial isolates obtained from various food specimens are shown in Table 2.

Antibiotics resistance
The details of the antibiotic resistance rates are shown in Table 3.Among the food-borne bacterial isolates, maximum resistance was noticed against gentamicin (33.3%), followed by amikacin (11.1%).Maximal resistance to a third-generation cephalosporin was noticed in only 5.6% isolates and all the isolates were found susceptible to piperacillin/tazobactam and imipenem.Among the clinical isolates, maximum resistance was noticed against third-generation cephalosporin followed by ofloxacin in 91.3% and 86.9% isolates respectively.Resistance to gentamicin and amikacin was noticed in 78.3% and 52.2% isolates respectively.

Phenotypic ESBL detection in food and clinical isolates
Based on combination disk tests using Cs and Cfs, ESBL could be inferred in 5.6% (2/36) and 91.3% (21/23) food-borne and clinical isolates, respectively.

Occurrence of bla genes
The frequency of the occurrence of various bla genes in clinical isolates is shown in Figure 2. TEM was found to be the most prevalent ESBL noticed in 60.9% (14/23) followed by CTX-M in 52.2% (12/23) isolates.AmpC and SHV beta-lactamases were noticed in 43.5% and 21.7% isolates, respectively.It Table 2. Showing bacterial isolates obtained from various food specimens was noteworthy to observe that none of the ESBL genes could be detected from food-borne bacterial isolates.

Discussion
bla CTX-M is the among the most prevalent and widely disseminated genes in the clinical bacterial population in India [6,8].In the present study, we found bla TEM as the most prevalent ESBL gene followed by bla CTX-M .The high prevalence of these resistant genes is attributed to the irrational use of antibiotics in India since antibiotics are freely available over the counter in this country [9,10].These resistance genes may freely disseminate to the vegetation-associated bacterial population during irrigation using sewage water, which is a common irrigation practice in India.Keeping these facts in mind, we planned this study to determine the occurrence of enteric pathogens on vegetation and to screen for the presence of bla genes.
We also included samples of animal-origin food and sweets that are openly sold in Indian markets, as well as a small subset of clinical bacteria to compare their resistance patterns and occurrence of resistance genes.
The most prevalent organism isolated in the present study was Klebsiella pneumoniae followed by Citrobacter koseri.The majority of the specimens provided mixed bacterial growth and there was a predominance of Gram-negative over Gram-positive bacteria.The strain of S. aureus isolated from chopped meat was identified as a methicillin-resistant strain based on the detection of mecA and femB genes (data not shown) using a protocol described earlier [11].It was alarming to note that, after washing the specimens, the bacteria could not be eliminated completely; however, the bacterial load was reduced markedly (data not shown).This observation emphasizes that the common practice of eating salad vegetables after only washing with water is not sufficient, and that these vegetables should be disinfected by using some antiseptic solutions such as potassium permanganate.
A marked difference was noticed in the susceptibility patterns of food-borne and clinical isolates.Food-borne isolates showed maximum resistance to gentamicin (33.3%) followed by amikacin (11.1%).Maximal resistance to any of the third-generation cephalosporins was noticed only in 5.6% isolates (see Table 3).On the other hand, the clinical isolates showed maximal resistance to cephalosporins (91.3%) followed by fluroquinolones, then ofloxacin (86.9%).Resistance to gentamicin, amikacin and piperacillin/tazobactam was noticed in 78.3%, 52.2%, and 52.2% isolates respectively.One K. pneumoniae isolate also demonstrated resistance to imipenem (Table 3).Similarly, a marked difference was noticed in the phenotypic demonstration of ESBL production in clinical versus food isolates.
It was interesting to observe that none of the foodborne isolates demonstrated the presence of any of the bla genes; however, bla CTX-M , bla TEM , bla SHV , and bla ampC were detected in 52.2%, 60.9%, 21.7%, and 43.5% respectively in clinical isolates.
Although these resistance genes have not yet disseminated to food vegetation, they remain a global threat.The findings from this study suggest that we must check and introduce strict irrigational and antibiotic policies to prevent the wide dissemination of these genes.Moreover, strict policies should be Table 3. Percent antibiotic* resistance rates in Gram-negative isolates obtained from food and clinical specimens * P. aeruginosa was also tested against piperacillin and found resistant to piperacillin instituted and over-the-counter availability of antibiotics should immediately be stopped in India.
To the best of our knowledge, this is the first report from India looking for these resistance genes in foodborne bacterial populations and also among the premier reports simultaneously looking for bla genes, including CTX-M, in Indian clinical isolates.

Table 1 .
Primers used and the expected amplicon size for dectection of bla CTX-m , bla TEM , bla SHV , and bla AmpC