Molecular Typing of Burkholderia mallei Isolates from Equids with Glanders, India

We collected 10 Burkholderia mallei isolates from equids in 9 districts in India during glanders outbreaks in 2013–2016. Multilocus variable-number tandem-repeat analysis showed 7 outbreak area–related genotypes. The study highlights the utility of this analysis for epidemiologically tracing of specific B. mallei isolates during outbreaks.


RESEARCH LETTERS
B urkholderia mallei is the etiologic agent of the contagious and fatal infection in equids known as glanders. It is one of the most ancient diseases and is distributed worldwide. B. mallei infections are frequently reported in South America, the Middle East, South Asia, and some countries in Africa. Equine glanders is a notifiable zoonotic disease; surveillance measures are enforced by the World Organisation for Animal Health (1).
Since 2006, equine glanders has been reported in India with consistently higher numbers from the Uttar Pradesh state (2,3). Regular glanders surveillance programs revealed presence of the disease in 14 states and, during 2015-2018, fresh B. mallei infections were reported in 6 states: Jammu and Kashmir, Gujarat, Rajasthan, Delhi, Madhya Pradesh, and Tamil Nadu (4). Epidemiologic investigations indicated that trading of equids from Uttar Pradesh to other states played a major role in spreading glanders (2). However, B. mallei isolates were not genotyped, which is necessary for understanding the epidemiologic association between glanders outbreaks across India.
Our study describes molecular typing of 10 B. mallei isolates recovered from horses (n = 4) and mules (n = 6) during 2013-2016 (Table; Appendix 1 Figure, https://wwwnc.cdc.gov/EID/article/27/6/20-3232-App1.pdf). All the affected equids were used for cart pulling and kept in small household stables. Five isolates (3324, 3478, 3701, 3711, and 3712), originating from 3 horses and 2 mules, were from adjoining districts of Uttar Pradesh state, which is regarded as a glanders hotspot zone (2). Three isolates (3076, 3081, 3595) from mules were located in 2 districts of Himachal Pradesh. Available information from the equine keeper suggested that these animals were traded from Uttar Pradesh and were responsible for the reported glanders incidence in this state. One isolate was recovered from a mule (3880) in Gujarat and 1 from a horse (3897) in Haryana state; both animals had no recent travel history.
We conducted MLVA by PCR amplification and sequencing of 23 loci using previously described primers (7). We determined sequence length and repeat number for each locus using Geneious software version 6.1.8 (https://www.geneious.com). A distance matrix giving the number of VNTR loci differing between isolates was used for analysis applying the minimum-evolution method implemented in MEGA X software version 10.0.5 (https://www.megasoftware.net).
MLVA assigned the 10 isolates to 7 genotypes, indicating considerable variability among B. mallei isolates in India (Figure, panel A). Identical MLVA patterns were observed for isolates 3076 and 3081 from Himachal Pradesh and isolate 3324 from Uttar We collected 10 Burkholderia mallei isolates from equids in 9 districts in India during glanders outbreaks in 2013-2016. Multilocus variable-number tandem-repeat analysis showed 7 outbreak area-related genotypes. The study highlights the utility of this analysis for epidemiologically tracing of specific B. mallei isolates during outbreaks. Pradesh. These findings correlate with epidemiologic investigations regarding the spread of a particular strain of B. mallei by equine movement, emphasizing the need to control equine trade between states. An identical pattern was also observed for B. mallei 3701 and 3712, which were isolated from Kasganj and Ghaziabad districts, 190 km apart in Uttar Pradesh state. The isolates 3897 and 3880 from Haryana and Gujarat differ clearly from the isolates from Himachal Pradesh-Uttar Pradesh cluster (Figure, panel A). However, isolates 3712, 3880, and 3897 were previously grouped into the L2B2sB2 branch by HRM-PCR analysis (8), which indicates superiority of MLVA for better epidemiologic resolution of glanders outbreaks.
Comparative MLVA between old and recent isolates from India revealed that most of the earlier isolates Mukteswar, BMQ, NCTC3708, NCTC3709, and India 86-567-2 are distantly related, whereas the isolate SAVP1 showed the highest similarity to the new isolates (Figure, panel A; Appendix 2 Table 3).
Further analysis of these B. mallei isolates plus 77 from other countries revealed that the 10 recent isolates of our study form a cluster that is most similar to isolates from Pakistan, followed by isolates from Turkey (Figure, panel B). This finding suggests that B. mallei strains prevalent in geographically close countries might have originated from an ancestral clone and gradually disseminated to different areas. Of interest, adoption of a strict regulatory movement policy at the beginning of the 19th century for control and eradication of glanders might have resulted in establishing specific B. mallei lineages at different ecologic settings. Our finding confirms previous observations regarding circulation of different B. mallei MLVA types in the Middle East (9,10).
In summary, MLVA proved useful as a genetic tool for classifying B. mallei isolates and tracing possible infection chains of glanders outbreaks in equids. VNTR information from more B. mallei isolates from India and other countries would be helpful to draw an epidemiologic conclusion between outbreaks.