One thousand three hundred and eighty three (1383) samples of food, feed, water and compost samples yielded 60 Aeromonas isolates through enrichment technique. Ampicillin dextrose agar and Aeromonas 'Ryan' medium were used successfully without Ampicillin. The isolates were identified to the species level using the Aerokey-II. This speciation yielded twelve A. hydrophila, sixteen A. veronii bv. sobria, twenty-six A. caviae, three A. trota, two A. schubertii and one A. jandaei strain. Out of the sixty strains isolated, five were sensitive to 10 ug ampicillin, which included four A. caviae and one A. trota. Some strains turned out to be slightly inhibited by 150 ug 2,4-diamino-6,7-diisopropyl-pteridine phosphate (0/129). These included one A. hydrophila strain, two each of A. veronii bv. sobria, A. caviae, A. !rota, and A. schubertii.
The use of Kaper's Aeromonas hydrophila (AB) multi-test medium proved to be very valuable in screening the suspected colonies. However, this medium did not prove to be reliable for testing of indole production, hydrogen sulfide (H2S) production from cysteine and for gas production.
Wild game birds had the highest incidence of the aeromonads (100%) followed in decreasing order by freshly slaughtered chicken (74%), fish / seafood (25%), compost / fertilizer (10%), milk (fresh/pasteurized), frozen chicken and fruits / vegetables (4-5%) and water and baby foods (-2%).Yoghurt, cheese, soft drinks and juices, animal feed and the dried / evaporated / reconstituted milk did not yield any aeromonads. Aeromonas caviae was the predominant species followed by A. veronii bv. sobria and A. hydrophila. A. caviae dominated over the other species among all the food types except fresh milk, where no A. caviae could be detected. All the three A. trota strains were isolated from fresh chicken. One A. schubertii was isolated from well-water and the other from fresh chicken. The only A. jandaei was from stored tank water.
The aeromonad colonies attained a size of >4 mm in 48 hours. They were round, flat, raised with entire margins. Most of the strains produced soft and smooth colonies while the others produced mucoid type. All grew well on MacConkey's agar producing both lactose negative and lactose positive colonies depending on their ability to ferment lactose. The lactose positive colonies were not too dark.
Most of the A. caviae and A. hydrophila strains produced a dark brown pigment on plate count agar (PCA). One strain each of A. trota and A. jandaei and a few of A. veronii by. sobria were also positive for this characteristic. All of the isolates of A. hydrophila, A. veronii by. sobria and A. trota but 18/22 of the A. caviae and 1/2 of A. schubertii exhibited β-hemolysis with large zones, whereas the rest of the strains showed α-hemolysis. On Kligler iron agar the isolates gave typical acid / alkaline reaction but many false negative results for gas from glucose and H2S production were revealed.
The growth of the aeromonads in purple broth (with glucose, glycerol or gluconate) gave characteristic features like acid production, gas production, turbidity, sediment and pellicle formation. Pellicle formation in gluconate broth could be equivalent to the determination of adhesiveness in the cell lines. The growth in Litmus milk medium did not provide any significant difference among the species. A. veronii by. sobria strains survived well (75%) in this culture medium for nine months at room temperature as compared to A. caviae and A. hydrophila (-40%). In trypticase soy broth at 37°C, A. hydrophila strain produced highest density (2 x 1010 cfu/ml), followed by A. veronii by. sobria (3.4 x 108 cfu/ml) andA. caviae (2.5 x 107 cfu/ml).
Seven types of API (Analytab Product Incorporation, France) strips (API 20 E, API 20 NE, Rapid ID 32 E, API 50 CH, ID 32 ON, ID 32 C, Rapid ID 32 A) were used for the biochemical characterization of the isolates. These strips provided an extensive biochemical profile of the isolates. API 20 E gave reliable results. The H2S tubule of this strip was added with gelatin-cysteine-thiosulfate medium and got excellent results for H2S production from cysteine. Esculin formulation in ID 32 C was proved most reliable as no false positive or false negative results were given by it. Rapid ID 32 E profile for the isolates were not reliable as many false positive and false negative results were given by it. All other strips provided test profiles and some of the characteristics in them appeared as significant in differentiation of the various species.
Gas-liquid chromatography / mass spectrometry (GC/MS) analysis of cellular fatty acids reveals that all the tested isolates except those belonging to A. veronii bv. sobria were similar in having C-12 to C-18 fatty acids. A. veronii bv. sobria strains were unique in possessing heptadecanoic acid and lacking octadecanoic acid. The various fatty acids detected in the different aeromonad isolates belonged to four categories: straight chained, branched chained, unsaturated, and hydroxy fatty acids. The unsaturated fatty acids were mono unsaturated. There was no difference among the fatty acid profile of the isolates from different sources qualitatively. However, quantitative difference was noted. In A. hydrophila the major fatty acids were dodecanoic acid, 3-hydroxy-tetradecanoic acid, hexadecenoic acid, and hexadecanoic acid. In A. veronii bv. sobria, these were dodecanoic acid, 3-hydroxytetradecanoic acid, hexadecanoic acid, hexadecanoic acid, 3-hydroxy-hexadecanoic acid, and octadecanoic acid. In A. caviae, dodecanoic acid, 3-hydroxy-tetradecanoic acid, hexadecenoic acid, hexadecanoic acid, 3-hydroxy-hexadecanoic acid were the major fatty acids. In A. schubertii, the major fatty acids were dodecanoic, tetradecanoic acid, 3-hydroxy-tetradecanoic acid, hexadecenoic acid, hexadecanoic acid, 3-hydroxy-hexadecanoic acid and octadecenoic acid and A. jandaei had dodecanoic acid, 3-hydroxy-tetradecanoic acid, hexadecenoic acid and 3-hydroxy-hexadecanoic acid. The major fatty acids common to all the species were dodecanoic acid, 3-hydroxy-tetradecanoic acid, and hexadecenoic acid. A. veronii bv. sobria and A. caviae were almost similar in major fatty acid composition. However, A. veronii bv. sobria had octadecenoic acids as an additional major fatty acid. A. schubertii resembled A. veronii bv. sobria but the former also possessed tetradecanoic acid among the major fatty acids. Similarly, A. jandaei resembled A. hydrophila but the former had 3-hydroxyhexadecanoic acid instead of hexadecenoic acid among the major acids. These differences in the presence / absence of major / minor fatty acids have been successfully used in the construction of a flow-chart for the speciation of the aeromonad isolates.
The Rapid Automated Bacterial Impedance Technique (RABIT) analysis concluded that all the isolates except one A. caviae strain from a bakery item source are mesophilic strains growing better at 37°C than at 30°C. Thus the occurrence of psychrophilic strains is rare. Also, there was no significant difference in the growth rate of aeromonads isolated from either food or the general environment. On comparing the generation time of the aeromonad isolates with respect to their temperature requirement for better growth, it was observed that most of the strains of A. hydrophila, A. veronii bv. sobria and A. caviae strains behaved similarly at 30°C. Their doubling time was more than 21 minutes. For the A. schubertii, A. trota and A. jandaei strains, the doubling time was less than 21 minutes. Among these, the A. jandaei isolate (WA-l) had the lowest generation time of 13.9 minutes. In general, all of the species, had a similar pattern of generation times at 37°C, being less than the time at 30°C. The only exception is one strain of A. caviae, the BK-2, whose generation time increased. Another notable thing was that the generation time of A. jandaei strain (W A-I) was only 9.5 minutes. So, this was the fastest growing strain of Aeromonas species, which appeared in the isolates of the present study both at 30°C and 37°C. There was no significant difference in the generation times of the environmental and food isolates.
The RABIT analysis for the growth response of the isolates at different pH revealed that A. hydrophila and A. veronii bv. sobria grew better at pH 7.2 and that A. caviae had a wider range of optimum growth pH with almost similar growth rate at pH 6.5 - 8.5. Moreover, all the three tested species failed to grow at pH 4.5 and A. caviae failed to grow at pH 9.5 at 37°C as well. Also, the pattern of growth at the tested pH was similar for A. hydrophila and A. veronii bv. sobria whereas A. caviae differed. A. caviae strain was found less tolerant of acidic pH (5.5) than of alkaline pH at 30°C. At 37°C, the tolerance to acidic pH increased as indicated by reduced generation time.
A total of 30 randomly selected isolates were tested for cytotoxin, cytotonic toxin, hemolysin, cell invasion and adhesiveness. Seventy percent of the tested isolates were cytotoxin producers and 80% were hemolytic. Cytotoxin was produced by 6 of 7 A. hydrophila, 6 of 13 A. caviae and 6 of 7 A. veronii bv. sobria strains, mostly from food sources. A. schubertii, A. jandaei and A. trota also produced both cytotoxin and hemolysin. All of the 30 isolates tested adhered to the Henle 407 cells. But none of the isolates was able to have invasive ability as noted through the in vitro assay. No significant correlation of co-occurrence of the putative virulence factors among the food isolates of the aeromonads was found.
Most of the strains were resistant to penicillins (ticarcillin, mezlocillin, oxacillin, piperacillin), sulfamethoxazole, trimethoprim and macrolides (erythromycin, vancomycin, clindamycin) but sensitive to tetracycline, chloramphenicol, nitrofurantoin, aminoglycosides (amikacin, gentamicin, tobramycin), cephalosporins (cefuroxime, ceftrioxone, cefazolin, cephalexin, cephalothin, cefoxitin, cefotaxime), quinolone (ciprofloxacin), colistin sulphate and SXT (trimethoprim-sulfamethoxazole).