Isolation of vapA-positive Rhodococcus equi from soil and fecal samples in Mongolia

Abstract

In 2024, 90 soil samples and 11 fecal samples were collected from nine Mongolian provinces. Using NANAT selective agar, R. equi was successfully isolated from 23 soil samples (25.6%) across five provinces and from three fecal samples (27.3%) collected in two provinces. A total of 122 isolates were identified as R. equi via choE-targeted polymerase chain reaction (PCR) and subsequently screened for virulence-associated genes (vapA, vapB, and vapN) by PCR. Of these, 17 isolates tested positive for the vapA gene, while the remaining 105 isolates were negative for both vapB and vapN. Plasmid profiling of the vapA-positive isolates revealed the presence of an 85-kb type I virulence plasmid, which is common in isolates from Europe and North America. This is the first documented detection of vapA-positive R. equi in Mongolia.

Rhodococcus equi is a well-established bacterial pathogen known to cause pyogranulomatous pneumonia and pulmonary abscesses in foals younger than 6 months of age [8], and the foal rhodococcosis has been reported in Europe, North and South America, Australia, and other countries around the world [23]. This organism is widely distributed in the environment, particularly in soil, and colonize the intestines of grazing animals and omnivores. Besides horses, R. equi also infects various mammals, including pigs, cattle, goats, cats, dogs, and humans [4, 14].

The pathogenicity of R. equi is closely linked to the presence of virulence plasmids encoding a family of virulence-associated proteins (Vaps) [12]. Two primary circular virulence plasmid types have been identified: pVAPA, which encodes VapA and is typically isolated from virulent equine strains [19], and pVAPB, which encodes VapB and is associated with intermediate virulence in mice. The latter has been recovered from submaxillary lymph nodes of slaughtered pigs and clinical human specimens [13]. More recently, a third plasmid, pVAPN, was discovered. This linear plasmid encodes VapN and has been isolated from bovine hosts and human clinical samples [18, 26]. Evidence from epidemiological and experimental studies suggests that these plasmids exhibit host specificity: pVAPA is associated with equine isolates, pVAPB with swine, and pVAPN with ruminants such as cattle and goats [2, 14].

To date, plasmid diversity within the pVAPA group has been assessed using restriction enzyme digestion. Fourteen closely related plasmid types have been identified in virulent R. equi strains, including 52-kb, 85-kb types I–V, 87-kb types I–III, and 90-kb types I–V [23]. Molecular epidemiological studies employing this method have revealed distinct geographical distribution patterns of R. equi infections in foals at horse breeding farms worldwide [5,6,7, 9,10,11, 15,16,17, 21,22,23, 27].

Japan is home to eight native horse breeds—Hokkaido, Kiso, Noma, Misaki, Tokara, Taishu, Miyako, and Yonaguni—all of which are believed to have originated from Mongolian horses [25]. This historical link led us to hypothesize that virulent R. equi strains carrying different plasmid types may have been introduced into Japan through the historical migration of horses from the Mongolian grasslands. In 2004, we collected soil and fecal samples from campsites of 26 nomadic families around Ulaanbaatar, Mongolia. No R. equi was isolated from those samples [20]. Additionally, R. equi was isolated from soils in native horse farms in Hulun Buir (eastern Mongolia), Xilin Goler (southern Mongolia), and Tongliao City (Inner Mongolia, China), but all isolates were negative for the vapA gene [24].

In 2024, we revisited this question by examining foal feces and soil samples collected from Mongolian horse-grazed areas in nine Mongolian provinces. For the first time, we detected vapA-positive R. equi in both soil and fecal samples from three provinces, marking the first report of virulent R. equi isolation in Mongolia.

Between July and August 2024, 90 soil samples were collected from areas surrounding horse watering sites in nine Mongolian provinces: Bayankhongor, Bayan-Ulgii, Bulgan, Darkhan, Dornod, Khovd, Khuvsgul, Selenge, and Uvurkhangai. In addition, 11 dried horse fecal samples were obtained from two of these provinces, Bulgan and Darkhan, at the same locations where soil samples were collected (Fig. 1). Using a sterile mini shovel, surface soil was scraped and placed into sterile sampling bags.

Fig. 1.

Map of Mongolia showing the provinces where soil and fecal samples were collected. Provinces are numbered according to their sampling locations. Shading indicates the Rhodococcusequi isolation rates in each province. Provinces where vapA-positive strains were identified are marked with an asterisk (*). B and K are shown the two area Bayantumen and Khulunbuir in Dornod.

At the laboratory of the School of Veterinary Medicine, Mongolian University of Life Sciences, each soil or fecal sample (1 g) was serially diluted in sterile saline. Aliquots from each dilution were plated in duplicate on nalidixic acid which is effective against Gram-positive bacteria–novobiocin which has a penicillin-like range of activity but does not affect R. equi–actidione (cycloheximide) which inhibits protein synthesis in eukaryotic cells–potassium tellurite which has been used in selective media for Corynebacteria (NANAT) selective agar, following the method described by Woolcock et al. [28]. Plates were incubated at 30°C for two or three days. Colonies with morphological characteristics consistent with R. equi were counted, and colony-forming units per gram (CFU/g) of sample were calculated. Up to ten colonies per specimen were subcultured and identified as below.

For species confirmation, a choE-based polymerase chain reaction (PCR) assay was employed, as described previously [1]. Isolates positive for choE were subsequently screened for the presence of the virulence genes vapA, vapB, and vapN using established PCR protocols [18]. Plasmid profiles of R. equi isolates were analyzed according to the method described by Takai et al. [3].

As shown in Table 1, R. equi was isolated from 23 (25.6%) of 90 soil samples from five of nine provinces and three (27.3%) of 11 fecal samples from two provinces. The number of R. equi in soil ranged from 50 to 26,500 CFU/g. The median R. equi counts per gram of soil were 25 in Selenge, 200 in Dornod Bayantumen, and 0 in the remaining provinces.

Table 1. Isolation of Rhodococcus equi from soil and fecal samples in Mongolia.

Province	Source	No. of samples	No. of R. equi-positive samples	Isolation rates(%)	Range of R. equi/g of sample	Median of R. equi/g of sample	No. of isolates	No. of vapA-positive(%)
Bayankhongor	Soil	15	0	0	-	0	-	-
Bayan-ulgii	Soil	3	0	0	-	0	-	-
Bulgan	Soil	14	3	21.4	100–3,500	0	16	4 (23.5)
Darkhan	Soil	3	0	0	-	0	-	-
Dornod B*	Soil	10	5	50	400–7,200	200	24	1 (4.2)
Dornod K**	Soil	22	9	40.1	5–26,500	0	37	9 (24.3)
Khovd	Soil	6	1	16. 7	500	0	1	0
Khuvsgul	Soil	6	1	16.7	50	0	1	0
Selenge	Soil	8	4	50	50–5,000	25	25	0
Uvurkhangai	Soil	3	0	0	-	0	-	-
Subtotal		90	23	25.6	-	0	104	14 (13.5)
Bulgan	Feces	5	1	20	500	0	1	0
Darkhan	Feces	6	2	33.3	800–3,500	0	17	3 (17.6)
Subtotal		11	3	27.3	-	0	18	3 (16.7)
Total		101	26	25.7	50–26,500	0	122	17 (13.9)

*Bayantumen, **Khulunbuir.

A total of 122 isolates from the 26 positive samples were examined for the presence of the choE, vapA, vapB, and vapN genes by PCR. All 122 isolates tested positive for choE; 17 were positive for vapA, while none were positive for vapB or vapN. Specifically, vapA was detected in four (25.0%) of 16 soil isolates from Bulgan, one (4.2%) of 24 soil isolates from Dornod Bayantumen, nine (24.3%) of 37 soil isolates from Dornod Khulunbuir, and three (17.6%) of 17 fecal isolates from Darkhan (Fig. 1). In Dornod Province, samples were collected from two separate areas—Bayantumen and Khulunbuir—with Khulunbuir showing higher contamination rates, similar to Bulgan (Table 1). Plasmid profiling of the 17 vapA-positive isolates revealed that all harbored an 85-kb type I plasmid.

This study provides the first evidence of vapA-positive R. equi in Mongolia. The bacterium was identified in three of the nine surveyed provinces, and all vapA-positive isolates carried the 85-kb type I virulence plasmid.

In our previous studies in Japan, avirulent R. equi are widespread in the horse-breeding environment on every farm with ranging from 10³ to 10⁴ CFU/g in soil, and the environment of the horse farms having endemic R. equi infections demonstrated heavy contamination (around 20% or more) with vapA-positive R. equi [8]. In contrast, the present study showed that although R. equi was detected in soil from five of nine Mongolian provinces, its concentration was markedly lower. This difference is likely due to the nomadic grazing practices in Mongolia compared with the fixed-location horse breeding systems in Japan. Additionally, Mongolia’s arid climate and low rainfall may reduce environmental R. equi survival, even in areas frequented by horses such as watering holes. Indeed, R. equi concentrations in Mongolian soils were estimated to be less than 100th of those observed in Japan [8]. Nevertheless, it is noteworthy that viable R. equi was also isolated from dry feces, demonstrating the organism’s ability to survive under such conditions.

Similar findings were reported in Inner Mongolia, where isolation rates of R. equi from soil in the Hulun Buir and Xilin Goler grasslands ranged from 25.9% to 30.0% [24].

Genetic studies have indicated that native Japanese horse breeds are descendants of Mongolian horses [25]. At least 14 subtypes of the pVAPA virulence plasmid have been identified globally, with geographical variation in subtype distribution [23]. In this study, all virulent Mongolian isolates harbored the 85-kb type I plasmid. This was unexpected, as previous isolates from Kiso native horses in Japan and Cheju native horses in Korea possessed the same 90-kb type II plasmid, and Japanese isolates contained 87-kb type II and 90-kb type I to V plasmids which were not found in isolates from the other continents [6, 11, 15]. Interestingly, the 85-kb type I plasmid is more commonly associated with isolates from Europe and North America [23]. The unexpected identity of the plasmid type with isolates from Western countries also suggests that further molecular epidemiological studies, including whole genome analysis and/or the phylogenetic analysis using appropriate markers, are needed to determine the plasmid diversity and distribution of virulent R. equi in Mongolia and to assess potential zoonotic risks.

Currently, contamination with virulent R. equi appears to be localized, with positive samples detected in Bulgan, Darkhan, and Dornod Provinces. Historically, R. equi infection in foals has not been recognized by veterinarians in Mongolia. However, with recent international horse imports into Dornod, the risk of introduction or dissemination of virulent strains warrants attention. Starting with the three provinces where the pVAPA-harboring strains were isolated in this study, we plan to conduct extensive monitoring of the Mongolian environment and horses to evaluate the spread of R. equi infection in terms of bacterial isolation and seroepidemiology.

CONFLICT OF INTEREST

The authors declare no potential conflicts of interest with respect to the research, authorship, or publication of this article.