Flood waters can be devastating, especially if proactive measures are not adequately taken ahead of time to mitigate the effects of the flood. In addition to the direct impact of flood water is the transmission of waterborne, foodborne, and airborne infection sequelae. Most of these infections are caused by pathogenic and opportunistic bacteria carried in the water from one location to another and include salmonellosis, leptospirosis, shigellosis, staphylococcus infections, burkholderiosis, vibriosis, and other infections [1, 2]. Different bacteria have been described in water from different sources worldwide, but there is paucity of data on bacteria in flood water during massive flood sessions.
Unexpected massive flood waters, that have defied meteorological forecasts, the worse in the history of Malaysia, hit the east coast of Peninsula Malaysia from 15th December 2014 to 3rd January, 2015 with Kelantan being the worst affected state. This great flood was estimated to have destroyed public property worth MYR (Malaysian Ringgit) 2.85 billion (about 814,285,714 USD); caused 25 deaths; affected 541,896 victims; with 2,076 houses destroyed, and a further 6,698 houses damaged; and 168 government healthcare facilities affected with an estimated MYR 380 million (108,571,429 USD) damage, and water levels rose 5-10 m above floodplain [3].
A study in Pahang state of Malaysia identified
This study was conducted to elucidate the bacterial biodiversity of flood water, describe the antibiogram of some bacteria found in flood water, and postulate the possible public health impact of flood water using water samples taken from the Kelantan flood disaster in Malaysia.
Kota Bharu is the capital city of Kelantan state of Malaysia located on the east coast of Peninsula Malaysia at 6°8′N 102° 15′E and close to the Thailand border with a population of about 491,237.
During the unexpected massive floods that hit the city, water samples were taken from 6 locations in the city as follows:
Taman Bendahara (Universiti Malaysia Kelantan city campus hostel area) with Global Positioning System (GPS) coordinates N06°09.809′E102°17.070′ and elevation of 12 m above sea level.
KampungTok Sadang (along Airport Road) with GPS coordinates N06°10.560′E102°17.115′ and elevation of 10 m above sea level.
Jalan Gajah Mati (Clock Roundabout) with GPS coordinates N06°07.508′E102°14.222′ and elevation of 23 m above sea level.
Kota Bharu mall surroundings with GPS coordinates N06°07.116′E102° 14.396′ and elevation of 25 m above sea level.
Tesco Bus stop area with GPS coordinates N06°06.789′E102° 13.757′ and elevation of 21 m above sea level.
Jalan Kuala Krai with GPS coordinates N06°06.285′E102° 14.433′ and elevation of 26 m above sea level.
The GPS coordinates and elevation above sea level were measured using a Nuvi 255 WT receiver (Garmin, Lenexa, KS, USA). We collected 31 water samples in 50 mL sterile containers from the 6 different locations (5 samples from each of Taman Bendahara, Kampung Tok Sedang, Kota Bharu clock roundabout, Kota Bharu mall area, Tesco mall area, and 6 samples from Jalan Kuala Krai) for bacteriological analysis. Each water sample was inoculated into blood agar and nutrient agar using sterile swabs. The inoculated media were put in an incubator at 37°C for 24–48 hours. The various bacterial colonies were selected based on colony characteristics such as shape, color, hemolysis, and size. The selected colonies were subcultured on nutrient agar to obtain pure colonies. The pure colonies were subjected to the following biochemical tests: catalase, oxidase, triple sugar iron agar (TSI), citrate, urease, motility, indole, methyl red (MR), and Voges–Proskauer (VP) tests. Because biochemical tests were inadequate and not exhaustive, genomic DNA was further isolated from the pure colonies using a commercial genomic DNA extraction and purification kit (Vivantis, Selangor Darul Ehsan, Malaysia and Oceanside, CA, USA) following the manufacturer’s instructions. Suitable oligonucleotide universal primers were used in a conventional polymerase chain reactions (PCR) to target 16S rRNA sequences of the unknown bacteria isolates.
The sequence of primers used was as follows: forward 5′-GGTGGAGCATGTGGTTTA-3′, reverse 5′-CCATTGTAGCACGTGTGT-3′ [7]. The expected product size was 287 bp. The PCR product was purified and sequenced for identification of isolated bacteria using DNA Sanger sequencing. Decipher software was used to check for any suspected chimeric sequences [8]. These sequences were compared with highly similar sequences at NCBI BLAST and SepsiTest BLAST for identification at up to species level. The threshold for identification was set at >97% for species identification. Species were not reported for any sequences below the threshold. Sequences were deposited at the GenBank, NCBI, USA, and accession numbers were obtained (isolates and sequences for the rest of this article are referred to by their initial identity without the characters preceding this e.g. SUB882316 UMK1a1, will be referred to as simply 1a1). Two isolates (2d1 and 3d1) of interest to the authors because of their characteristic violet-to-black pigmented colonies were also confirmed using species specific PCR primers (
All 16s rRNA sequencing data were subjected to the following preprocessing procedure:
The quality control of the sequences was conducted by analyzing the trace files using SeqScanner version 1.0 (Applied Biosystems, Foster City, CA, USA). The leading vector, tailing and poor-quality (trace score >20) sequences were removed accordingly (file available on request).
The remaining sequences were trimmed at the 3′ or 5′ ends to remove low quality ends of the sequences because of the noise introduced by low quality regions (in Geneious version R8.1 (Biomatters,
For each sample, the paired reads (forward and reverse) were assembled through assembly in Geneious as shown following:
Species classification and relative abundance measurement using high throughput 16S rRNA amplicon sequencing data from environmental samples were performed using the cloud-based 16S rRNA biodiversity tool (Geneious version R8.1, (Biomatters,
Charts, Tables, and Figures can be requested by email from:
Bacteria Identified to species level from the flood and deposited at the GenBank
S/No. | Submission ID | Accession Number | Bacteria Identified |
---|---|---|---|
1 | SUB882316 UMK1a1 | KR027927 | |
2 | SUB882316 UMK1a2 | KR027928 | |
3 | SUB882316 UMK1b1 | KR027929 | |
4 | SUB882316 UMK1b2 | KR027930 | |
5 | SUB882316 UMK1b3 | KR027931 | |
6 | SUB882316 UMK1c2 | KR027932 | |
7 | SUB882316 UMK1d1 | KR027933 | |
8 | SUB882316 UMK1d2 | KR027934 | |
9 | SUB1092331 UMK1e1 | KT731961 | |
10 | SUB1092331 UMK1e2 | KT731962 | |
11 | SUB882316 UMK2a1 | KR027935 | |
12 | SUB1092331 UMK2a2 | KT731963 | |
13 | SUB882316 UMK2a3 | KR027936 | |
14 | SUB1092331 UMK2d1 | KT731964 | |
15 | SUB882316 UMK2e1 | KR027937 | |
16 | SUB882316 UMK2e2 | KR027938 | |
17 | SUB882316 UMK3a1 | KR027939 | |
18 | SUB882316 UMK3a2 | KR027940 | |
19 | SUB882316 UMK3a3 | KR027941 | |
20 | SUB882316 UMK3b1 | KR027942 | |
21 | SUB1092331 UMK3b2 | KT731965 | |
22 | SUB882316 UMK3b3 | KR027943 | |
23 | SUB1092331 UMK3c3 | KT731966 | |
24 | SUB1092331 UMK3d2 | KT731967 | |
25 | SUB882316 UMK3d3 | KR027944 | |
26 | SUB1092331 UMK3d4 | KT731968 | |
27 | SUB1092331 UMK3d5 | KT731969 | |
28 | SUB882316 UMK3e1 | KR027945 | |
29 | SUB882316 UMK3e2 | KR027946 | |
30 | SUB883111 UMK4a1w | KR048048 | |
31 | SUB883111 UMK4a1y | KR048049 | |
32 | SUB883111 UMK4a2 | KR048050 | |
33 | SUB882316 UMK4b1 | KR027947 | |
34 | SUB882316 UMK4b2 | KR027948 | |
35 | SUB882316 UMK4b3 | KR027949 | |
36 | SUB882316 UMK4c2 | KR027950 | |
37 | SUB882316 UMK4c3 | KR027951 | |
38 | SUB882316 UMK4d2 | KR027952 | |
39 | SUB882316 UMK4d3 | KR027953 | |
40 | SUB882316 UMK4e1 | KR027954 | |
41 | SUB882316 UMK4e2 | KR027955 | |
42 | SUB882316 UMK5a1 | KR027956 | |
43 | SUB882316 UMK5a2 | KR027957 | |
44 | SUB882316 UMK5b1 | KR027958 | |
45 | SUB882316 UMK5d1 | KR027959 | |
46 | SUB883111 UMK6a1w | KR048051 | |
47 | SUB883111 UMK6a1y | KR048052 | |
48 | SUB1092331 UMK6a2 | KT731970 | |
49 | SUB1092331 UMK6b1 | KT731971 | |
50 | SUB1092331 UMK6b2 | KT731972 | |
51 | SUB1092331 UMK6c | KT731973 | |
52 | SUB882316 UMK6d | KR027960 | |
53 | SUB882316 UMK6e | KR027961 | |
54 | SUB1092331 UMK6x1 | KT731974 | |
55 | SUB1092331 UMK6x2 | KT731975 |
Antibiotic sensitivity test of some bacteria isolates from the flood
No. | Isolate (bacteria) | TE 30-tetracycline | AML 10– amoxycillin | S 25 – streptomycin | CN 10– gentamycin | E 15– erythromycin | AMP 10– ampicillin | P10– penicillin G | % Total resistance to all antibiotics |
---|---|---|---|---|---|---|---|---|---|
1 | 2dl ( | 31 (S) | R | 15 (I) | 19 (S) | 20 (I) | R | R | 43% |
2 | 3a2 ( | 25 (S) | R | 11 (R) | 19 (S) | 12 (R) | R | R | 71% |
3 | 3a3 ( | 22 (S) | 12 (R) | 15 (I) | 19 (S) | 11 (R) | R | R | 57% |
4 | 4b3 ( | 16 (I) | R | 20 (I) | 20 (S) | 24 (S) | R | R | 43% |
5 | 4d2 ( | 28 (S) | 13 (R) | 23 (S) | 25 (S) | 30 (S) | 12 (I) | R | 29% |
6 | 4d3 ( | 21 (S) | 27 (S) | 18 (I) | 22 (S) | 15 (I) | R | R | 29% |
7 | 4e2 ( | 12 (R) | 20 (S) | 21 (S) | 25 (S) | 23 (S) | R | 22 (S) | 29% |
8 | 5a2 ( | 14 (R) | 15 (I) | 12 (R) | 18 (S) | 12 (R) | 17 (S) | R | 57% |
9 | 5b 1 ( | 26 (S) | 40 (S) | 22 (S) | 25 (S) | 28 (S) | 44 (S) | 30 (S) | 0% |
10 | 5dl ( | 21 (S) | 22 (S) | 20 (I) | 25 (S) | 19 (I) | 20 (S) | 12 (R) | 14% |
11 | 6a lw ( | 20 (S) | 13 (R) | 22 (S) | 24 (S) | 15 (I) | 12 (R) | R | 43% |
12 | 6a ly ( | 22 (S) | R | 17 (I) | 18 (S) | 23 (S) | R | R | 43% |
13 | 6a2 ( | 25 (S) | 31 (S) | 25 (S) | 31 (S) | 28 (S) | 31 (S) | 28 (S) | 0% |
14 | 6b 1 ( | 23 (S) | R | 20 (I) | 33 (S) | 13 (R) | R | R | 57% |
15 | 6b2 ( | 11 (R) | 17 (I) | 18 (I) | 22 (S) | 25 (S) | R | 18 (S) | 29% |
16 | 6c ( | 12 (R) | 12 (R) | 18 (I) | 25 (S) | 26 (S) | 12 (I) | 7 (R) | 43% |
17 | 6e ( | 23 (S) | 32 (S) | 24 (S) | 25 (S) | 16 (I) | 32 (S) | 25 (S) | 0% |
Key: R = Resistant; I = Intennediate; S = Susceptible
Analysis of variance of antibiotics sensitivity (multiple comparisons) with Tukey honest significant difference post hoc test
(I) Antibiotic | (J) Antibiotics | Mean Difference (I–J) | Std. Error | 95% Confidence Interval | ||
---|---|---|---|---|---|---|
Lower | Upper | |||||
TE 30–tetracycline | AML 10–amoxycillin | 5.47 | 3.09 | 0.57 | –3.79 | 14.73 |
S 25–streptomycin | 1.82 | 3.09 | >0.99 | –7.44 | 11.09 | |
CN 10–gentamycin | –2.53 | 3.09 | 0.98 | –11.79 | 6.73 | |
E 15–erythromycin | 0.71 | 3.09 | >0.99 | –8.56 | 9.97 | |
AMP 10–ampicillin | 9.59 P<0.05; | 3.09 | 0.04 | 0.33 | 18.85 | |
P 10–penicilling | 11.77 P<0.01; | 3.09 | 0.004 | 2.50 | 21.03 | |
AML 10–amoxycillin | TE 30–tetracycline | –5.47 | 3.09 | 0.57 | –14.73 | 3.79 |
S 25–streptomycin | –3.65 | 3.09 | 0.90 | –12.91 | 5.62 | |
CN 10–gentamycin | –8.00 | 3.09 | 0.14 | –17.26 | 1.26 | |
E 15–erythromycin | –4.77 | 3.09 | 0.72 | –14.03 | 4.50 | |
AMP 10–ampicillin | 4.12 | 3.09 | 0.83 | –5.14 | 13.38 | |
P 10–penicilling | 6.29 | 3.09 | 0.40 | –2.97 | 15.56 | |
S 25–streptomycin | TE 30–tetracycline | –1.82 | 3.09 | >0.99 | –11.09 | 7.44 |
AML 10–amoxycillin | 3.65 | 3.09 | 0.90 | –5.62 | 12.91 | |
CN 10–gentamycin | –4.35 | 3.09 | 0.80 | –13.62 | 4.91 | |
E 15–erythromycin | –1.12 | 3.09 | >0.99 | –10.38 | 8.14 | |
AMP 10–ampicillin | 7.77 | 3.09 | 0.16 | –1.50 | 17.03 | |
P 10–penicilling | 9.94 P<0.05; | 3.09 | 0.03 | 0.68 | 19.20 | |
CN 10–gentamycin | TE 30–tetracycline | 2.53 | 3.09 | 0.98 | –6.73 | 11.79 |
AML 10–amoxycillin | 8.00 | 3.09 | 0.14 | –1.26 | 17.26 | |
S 25–streptomycin | 4.35 | 3.09 | 0.80 | –4.91 | 13.62 | |
E 15–erythromycin | 3.24 | 3.09 | 0.94 | –6.03 | 12.50 | |
AMP 10–ampicillin | 12.12 P<0.01; | 3.09 | 0.003 | 2.86 | 21.38 | |
P 10–penicilling | 14 29 P<0.001. | 3.09 | 0.001 | 5.03 | 23.56 | |
E 15–erythromycin | TE 30–tetracycline | –0.71 | 3.09 | 0.99 | –9.97 | 8.56 |
AML 10–amoxycillin | 4.76 | 3.09 | 0.72 | –4.50 | 14.03 | |
S 25–streptomycin | 1.12 | 3.09 | 0.99 | –8.14 | 10.38 | |
CN 10–gentamycin | –3.24 | 3.09 | 0.94 | –12.50 | 6.03 | |
AMP 10–ampicillin | 8.88 | 3.09 | 0.07 | 0.38 | 18.14 | |
P 10–penicilling | 11.06 P<0.05; | 3.09 | 0.009 | 1.80 | 20.32 | |
AMP 10–ampicillin | TE 30–tetracycline | –9.59 P<0.05; | 3.09 | 0.04 | –18.85 | –0.33 |
AML 10–amoxycillin | –4.12 | 3.09 | 0.83 | –13.38 | 5.14 | |
S 25–streptomycin | –7.76 | 3.09 | 0.16 | –17.03 | 1.50 | |
CN 10–gentamycin | –12.12 P<0.01; | 3.09 | 0.003 | –21.38 | –2.86 | |
E 15–erythromycin | –8.88 | 3.09 | 0.07 | –18.14 | 0.38 | |
P 10– penicillin G | 2.18 | 3.09 | >0.99 | –7.09 | 11.44 | |
P 10–penicillin G | TE 30–tetracycline | –11.76 P<0.01; | 3.09 | 0.004 | –21.03 | –2.50 |
AML 10–amoxycillin | –6.29 | 3.09 | 0.40 | –15.56 | 2.97 | |
S 25–streptomycin | –9.94 P<0.05; | 3.09 | 0.03 | –19.20 | –0.68 | |
CN 10–gentamycin | –14 29 P<0.001. | 3.09 | 0.001 | –23.56 | –5.03 | |
E 15–erythromycin | –11.06 P<0.01; | 3.09 | 0.009 | –20.32 | –1.80 | |
AMP 10–ampicillin | –2.18 | 3.09 | 0.99 | –11.44 | 7.09 |
The unexpected nature of this study, which was conducted during the course of a devastating flood, may have affected some parameters that could have improved this study. The use of only culturing of water samples and looking for only bacterial colonies that grew within 48 hours may have excluded other bacteria such as
After the flooding, the prevalence of bacterial infections usually increases triggering episodes of intestinal symptoms such as diarrhea, vomiting, stomach aches, and other gastrointestinal disease symptoms as contaminated water moves from one geographical location to another, carrying a “cocktail” of bacteria along with it [16-18]. The distribution of the bacteria isolated did not show any remarkable difference between the different locations from which water samples were taken. This is probably because there was no large difference in the elevation between the locations, and during flood there is a massive movement of water from one location to the other, which can carry and distribute bacteria almost uniformly from one flood water location to the other. Different water depths would be expected to produce a variety of bacteria species [19]. This has potential public health implications because it implies waterborne infections can easily be carried from one flood location to another.
The predominance of proteobacteria making up 67% of bacteria in this study is similar to a study conducted in Thailand, which reported that majority of bacteria from the 2011 Thailand flood were from the phylum proteobacteria, which made up 56.5% to 91.4% of bacteria in different water samples in Thailand [6]. However, the majority of families and genera of bacteria reported from Thailand differ from the ones in this study, demonstrating the heterogeneity of bacterial communities across different flooded environments.
Flood water is a conglomeration from different sources including overflowing seas, rivers, streams, springs, wells, and other water. Humankind′s activities during floods including bathing, swimming, washing, and excretion of waste into flood water affects its bacterial composition [28]. Some of the pollution may also have come from industrial waste, sewage water contamination, and admixtures of water from all manner of unhealthy sources during the course of the flooding [29, 30]. The study from Thailand showed pathogenic bacteria and high cross-contamination between flood water and other water sources [5]. The degree of pollution of soil surface and the metallic components of the soil also determine the richness and diversity of the bacteria present with presence of zinc decreasing both diversity and levels of species richness [31]. The rich bacterial diversity of our flood water showed bacteria from human and animal sources, and bacteria from the environment. A study in Brazil found that usual environmental water is a rich source of many species of bacteria with varying degrees of antibiotic resistance, showing some bacterial communities tolerating up to 600 times the clinical treatment levels of common antibiotics [32].
There appeared to be some strain variation in pathogenicity of some of the isolates as indicated by the same species showing different antibiotic sensitivities. Further studies that go beyond species identification to identification of different strains and genes coding for resistance are required to establish if indeed these are different strains of the same bacteria with varying pathogenicity.
There were 9 bacteria in the present study not previously reported from any source in Malaysia based on our literature search. All of them were bacteria recently reported and classified within the last 20 years.
During this massive flood session, there was a rich bacterial biodiversity including some species of potentially pathogenic bacteria that could endanger public health. This altered bacterial composition of normal water outside of flooding, and may explain why there are outbreaks of various infectious diseases during and after flood disasters. During the flood disaster period the only functional tertiary hospital (Hospital Universiti Sains Malaysia) in Kelantan handled 180 cases/day in the emergency department [38]. In adjacent Pahang state, 1,220 flood-related cases were handled within the first 6 days of the Kelantan flood disaster [39].
Most of the bacteria isolated from this study were resistant to one or more commonly used antibiotics. This is of interest to health practitioners and health policy makers because the presence of multidrug resistant bacteria should guide clinicians in the choice of antibiotics during flood disasters for effective treatment and control of waterborne infections. Gentamycin and tetracycline antibiotic classes appeared to be the best antibiotics to consider, but this may be an ever-changing picture. During flooding human and animal contact with flood water should be minimized, if not avoided completely, and adequate provisions should be made for provision of clean water to avoid outbreaks of waterborne diseases.