Abstract
Chromobacterium haemolyticum is a fatal Gram-negative bacterium which could infect human beings. Our investigation found severe piglet diarrhea in one farm; the morbidity rate of piglets was 30.65%. Then, we isolated a nonpigmented, β-hemolytic Gram-negative bacillus from the clinical samples of this farm, which was designated GDHYZ30 strain. The 16S rRNA gene sequence indicated it was most closely related to Chromobacterium haemolyticum. Similar clinical symptoms were successfully reproduced in experimental piglets with this isolate, and the isolate was also subjected to whole genome-sequencing. It is worth noting that this Chromobacterium haemolyticum has been isolated from several other pig farms with diarrhea of unknown causes. To our knowledge, this is the first report that discusses Chromobacterium haemolyticum as a new pathogen causing diarrhea and death in piglets and transmitting through water sources, and it provides a reference for the prevention and control of human/animal infections as well as food safety.
1. Introduction
The genus Chromobacterium, consisting of seven recognized species, is a Gram-negative, rod-shaped bacterium, and its most famous specie is Chromobacterium violaceum with a red pigment [1]. Chromobacterium is mainly distributed in tropical and subtropical regions, and it is frequently isolated in water sources and humid areas [2]. Chromobacterium haemolyticum (C. haemolyticum) was considered as a separate species, lacking violet pigmentation but demonstrating strong hemolysis on sheep blood agar plates [3]. C. haemolyticum has been isolated from a variety of water environments, such as rice roots, lakes, rivers, and sewage effluent [4–6]. There is evidence that C. haemolyticum is mainly transmitted through water sources, and some reports suggest that C. haemolyticum can cause pneumonia and sepsis in humans, and the patients in these cases have a history of wound exposure or contact with water sources, indicating that C. haemolyticum maybe an opportunistic pathogen [7, 8]. There also have been reports of diarrhea cases in humans that were related to C. haemolyticum, but C. haemolyticum causing diarrhea or even death in other animals has not been reported yet [9].
Piglets diarrhea is a serious disease that causes great harm to the swine industry. Due to the fact that the intestinal immunity of piglets has not fully developed yet, many pathogens are susceptible to piglets, causing acute diarrhea, vomiting, dehydration, and even death [10, 11]. Diarrhea in piglets can be devastating to pig farms if the main pathogen cannot be identified and treated accordingly. Some farms use mountain springs or river water as water sources without disinfection, which often leads to infection of some bacterial diseases. C. haemolyticum is pathogenic and usually detected in water sources; however, it has been paid less attention compared to its peers. Thus, it is meaningful to conduct in-depth research on the biological and genetic characteristics of this bacteria.
Our study focused on an unexplained case of piglet diarrhea that occurred in many pig farms in south China. After excluding some regular pathogens, C. haemolyticum was found both in the water source and in piglet fecal samples with diarrhea. The challenge of piglets with the isolate GDHYZ30 can cause diarrhea symptoms similar to those of the field cases. And the diarrhea of piglets had been cured after changing the drinking water source, indicating that the GDHYZ30 strain contaminating the drinking water is the agent for the piglet diarrhea and death.
2. Materials and Methods
2.1. Sample Collection and Pathogen Identification
The anal swabs from diarrheic piglets and water sources from pig farms were collected. Nucleic acids were extracted to detect common diarrhea-associated viruses; primers were referred to Supplemental file 1, and bacteria were isolated using sheep blood agar plates.
2.2. Antibiotic Susceptibility Assay
Antimicrobial susceptibility profiles were determined according to the CLSI (Clinical and Laboratory Standards Institute) [12]. The inhibition zone diameter was measured after 24 h incubation at 37°C.
2.3. Transmission Electron Microscopy
GDHYZ30 were grown for 24 h in LB medium, centrifuged (5000 g, 5 min), and resuspended in distilled water. Samples were placed on a carbon-coated grid (200 mesh) and stained with 1% phosphotungstic acid (PTA). The grid was air dried and examined in a Talos L120 C microscope (Thermo Fisher, USA).
2.4. Experimental Infection Trial
Nine 30-day-old healthy piglets were randomly divided into three groups. 2 mL of the 3.5 × 10^9 colony-forming unit (CFU) of GDHYZ30 were administered orally or intramuscularly injected, respectively, and 2 mL of PBS were administered orally to the control group.
2.5. Histological Examination
Fresh piglet intestinal tissues were collected from piglets fixed in 10% neutral buffered formalin, stained with hematoxylin and eosin, and examined under a light microscope.
2.6. Animal Ethics and Welfare
Animal studies were carried out in strict accordance with the recommendations made in the Guide for the Animal Care & Health and Supervisory Committee. Protocols were approved by the Committee on the Ethics of Animal Experiments of the South China Agricultural University.
3. Results
3.1. Clinical and Biological Properties
Our investigation focused on severe piglet diarrhea that occurred at one farm in south China, where the morbidity rate of piglets was 30.65% (149/486) and the mortality rate was 35.57% (53/149). To figure out the pathogen responsible for diarrhea in piglets, we collected fecal samples from piglets with diarrhea and tested a variety of diarrhea-causing viruses. The viruses tested includes porcine epidemic diarrhea virus (PEDV), porcine deltacoronavirus (PDCov), transmissible gastroenteritis virus (TGEV), swine acute diarrhea syndrome coronavirus (SADS-CoV), porcine rotavirus (PRoV), porcine circovirus type 3 (PCV3), porcine circovirus-like viruses (PCLV), and other common pathogens listed in the supplemental files. The viral presence had been monitored by PCR methods, and all the above pathogens had not been detected. Excitingly, a Gram-stained negative, rod-shaped (Figure 1(b)) β-hemolysis bacterium was isolated on sheep blood agar plates (Figure 1(a)). The following identification verified that the isolated bacterium is about 3 μm in diameter and with a single polar flagella (Figure 1(c)). The strain was purified by colonization and named the GDHYZ30 strain.
(a)
(b)
(c)
3.2. Genetic Evolution and Prevalence
In order to study the genetic characteristics of the bacterium, we carried out the whole gene-sequencing analysis. The 16S rRNA phylogenetic tree analysis showed that the bacterium belongs to Chromobacterium, which is the same branch as C. haemolyticum. We compared C. haemolyticum from different sources and found that GDHYZ30, GDMM22, and GDMMLY were closest to the KM2 strain, and GDLY and GDHY20 were closest to the PGS9 strain (Figure 2(a)). The climatic characteristics of Guangdong Province are more suitable for the reproduction of Chromobacterium, so we expanded the detection range. Consequently, C. haemolyticum was detected in water sources and an anal swab of piglets in other areas with unexplained diarrhea. The bacteria were specifically detected in five regions, including HEYUAN, MAOMING, ZHANJIANG, SHANTOU, and SHANWEI, while the clinical symptoms were more severe in the HEYUAN region (Figure 2(b)). These results suggest that the GDHYZ30 strain should be taken seriously in south China since this C. haemolyticum could be an important agent of diarrhea and death in piglets and a potential infection risk factor for human beings health in the pork supply as well.
(a)
(b)
3.3. Genome Annotation
Whole genome sequencing showed that the genome size of strain GDHYZ30 is 4,785,117 bp with 4,398 coded genes. The genome’s GC content is 62.67%, and the longest protein-coding gene is 11,694 bp. The average length of genes that encode a protein is 957 bp, and the proportion of genes encoding proteins was 88.04% (Figure 3(a) and Table 1). After comparing the reference sequences using BLAST x against the NR (NCBI nonredundant protein sequences), Swiss-Prot, COG (Clusters of Orthologous Groups of proteins), and KEGG (Kyoto Encyclopedia of Genes and Genomes) databases. We identified 4,220 unigenes that provided significant results in NR and annotated 3,179, 3,346, and 2,145 unigenes from the Swiss-Prot, COG, and KEGG databases, respectively (Figure 3(b)). By aligning with the NR library, it is possible to view the approximate transcript sequence of the species and the similar species, as well as the functional information of the homologous sequence. The GDHYZ30 strain has 3,059 sequences aligned with the C. haemolyticum strain (Figure 3(c)), the most homology among all species analyzed. Functional prediction and classification of unigenes were performed by comparing sequence data against the COG database. A total of 1,935 unigenes were annotated and grouped into 14 categories according to COG function classifications. Among them, the top 3 clusters for general function prediction were “amino acid transport and metabolism” (330 genes), “transcription” (262 genes), and “energy production and conversion” (191 genes) (Figure 3(d)). This whole-genome sequencing project has been deposited at GenBank under the accession number: PRJNA802706.
(a)
(b)
(c)
(d)
3.4. Antibiotic Susceptibility Assay
Antibiotic susceptibility assays show that the five C. haemolyticum strains are sensitive to ofloxacin, ciprofloxacin, gentamicin, enrofloxacin, spectinomycin amikacin, and in contrast, they are highly resistant to vancomycin, cefradine, sulfamethoxazole, ampicillin sodium, and sulbactam sodium (Table 2).
3.5. Virulence Genes and Drug Resistance
The analysis of drug resistant genes found that the GDHYZ30 strain had 57 drug resistant genes, granting it resistant to β-lactams, carbapenem hydrolyzing β-lactamases, etc (Supplemental file 2). GDHYZ30 was compared with the SETA and SETB databases, respectively, matching with 390 virulence factors in the SETA database and 432 virulence factors in the SETA database (Supplemental file 3). These results provide a reference for the treatment and pathogenesis of the disease.
3.6. Pathogenicity of the GDHYZ30
To determine the pathogenicity of the GDHYZ30, Koch postulates experiments were conducted in terms of the GDHYZ30. Nine 30-day-old healthy piglets were randomly divided into three groups. 2 mL of 3.5 × 10^9 CFU bacteria were administered orally or intramuscularly injected, respectively, and 2 mL of PBS was administered orally to the control group. At the second day, we found that all infected groups developed clinical symptoms similar to that of the field cases (Figure 4(b)). One piglet in the intramuscular group had diarrhea and two piglets had died (Figure 4(a)). On the other hand, piglets in the control group were asymptomatic. Thin intestinal walls and flatulence had been observed in piglets with diarrhea from infection groups (Figure 4(b)). In HE staining of intestinal tissues from various infectious groups, exfoliation of epithelial cells and infiltration of inflammatory cells, accompanied by severe ileal bleeding, can be observed. The colon in the irrigation group had more damage, such as inflammatory cell infiltration, than the colon in the intramuscular group (Figure 4(c)). After examination, the GDHYZ30 strain was isolated from the blood and heart samples from the dead piglets, but the GDHYZ30 strain only was isolated from intestinal samples from the oral group. The above results indicated that the GDHYZ30 strain was the pathogen that caused piglet diarrhea and acute death.
(a)
(b)
(c)
4. Discussion
Our investigation focused on the frequent outbreaks of diarrhea and death in piglets which drink the mountain spring water, and we isolated a novel hemolytic Gram-negative bacterium in the fecal and drinking water samples of infected piglets, naming it GDHYZ30. 16S rRNA analysis demonstrates that the bacteria were closely related to C. haemolyticum. Then we conducted the Koch postulate experiments to confirm that the GDHYZ30 strain was the causative agent of this kind of piglet diarrhea and death. Then the whole gene sequencing analysis of the GDHYZ30 strain was carried out. The phylogenetic tree based on 16S rRNA showed that the GDHYZ30 strain belongs to the same branch as C. haemolyticum. At the same time, we isolated Chromobacterium in other pig farms where unexplained diarrhea was present, and the homology of these Chromobacterium were minor different compared with GDHYZ30 strain. These Chromobacterium mainly cause diseases characterized by acute diarrhea and death. It has been reported that C. haemolyticum has antibacterial properties, and we speculate that infection with C. haemolyticum, except for the direct intestinal damage, may cause changes in the intestinal flora, reducing beneficial bacteria and leading to diarrhea consequently as well. [13]. At the same time, we found that infection with this C. haemolyticum can cause death, and the intramuscular injection of the GDHYZ30 strain can even cause acute death in piglets. Meanwhile, the GDHYZ30 strain could be isolated from the blood samples and hearts of dead pigs, while the gavage group only caused diarrhea. We also got similar results when mice were infected with C. haemolyticum by irrigation or intramuscular. These results suggest that C. haemolyticum can cause bacteremia leading to acute death. The main causative factor may be the hemolytic properties of C. haemolyticum. Challenge assays have shown that C. haemolyticum can infect piglets through water sources. Reports of human infection with C. haemolyticum have been linked to wound exposure to water sources; whether piglets can be infected C. haemolyticum through wounds remains to be confirmed [7, 14]. C. haemolyticum has different clinical manifestations compared with diarrhea-related pathogens: Escherichia coli mainly infects preweaning piglets, while C. haemolyticum mainly infects postweaning piglets, and C. haemolyticum infection does not cause dehydration in piglets compared with PEDV. The bacteria have 390 virulence genes and 57 drug resistance genes; whether these genes are pathogen-associated still needs further research. Clinical trials and drug resistance analysis show that C. haemolyticum is a potential threat to humans and animals. The antibiotic susceptibility assay shows the five isolated C. haemolyticum are resistant to most antibiotics, and their susceptibility to some antibiotics varies greatly. Therefore, the drug resistance of the isolates should be measured so that the effective drugs could be administrated.
The growth and development of C. haemolyticum require moist water sources. Consequently, the geographical location and climate of Guangdong Province provide an ideal habitat for the reproduction of C. haemolyticum [6, 15]. We also detected C. haemolyticum in the drinking water of other pig farms with unexplained diarrhea, and after changing the water source, the piglet diarrhea was controlled. Therefore, checking the sanitation of water sources can prevent the occurrence of the disease.
Data Availability
This whole-genome sequencing project has been deposited at GenBank under the accession number: PRJNA802706.
Conflicts of Interest
All authors declare that there are no conflicts of interest.
Authors’ Contributions
LYZ, PFZ, XZ, and CXS conceived and designed the experiments. GX provided the electron microscope photos. KSS is responsible for the English touch-up of the manuscript. TRL, JWP, STW, NJY, LY, YLL, XMZ, and PSL are responsible for experimental development, including animal testing and drug sensitivity testing. Leyi Zhang and Pengfei Zhang contributed equally to this work.
Acknowledgments
This study was supported by the National Key Research and Development Program of China (2018YFD0501102) and the Key-Area Research and Development Program of Guangdong Province (2019B020211003).
Supplementary Materials
The sequence of primers used in this article is listed in Supplemental file.1. We tested the GDHYZ30 strain for drug resistance genes, and the results are listed in Supplemental file.2. The GDHYZ30 stration virulence factor was compared with SETA and SETB databases, the results are listed in Supplemental file.3. Supplemental file.1: Pathogen detection primer sequences. Supplemental file.2: Drug resistance gene statistics. Supplemental file.3: Virulence factor statistics. (Supplementary Materials)