The serological and genetic diversity of the Leptospira interrogans Icterohaemorrhagiae serogroup circulating in the UK

Leptospirosis is an emerging major public health problem in many countries and is one of the most widespread zoonoses in the world. It is an excellent example of the rightness of the “One Health” approach, where the relationship between humans, animals and ecosystems needs be considered in order to better understand and manage a disease (20). The disease is caused by infection with pathogenic spirochetes of the Leptospira genus, which can chronically infect domestic and wild animal species and in particular small rodents – being harboured in their renal tubules before being shed into the environment via their urine.

The Leptospira genus is currently subdivided into 68 genomically distinct Leptospira species (5). Besides this taxonomical division, members of the genus are also divided according to the serological similarity of surface-exposed epitopes as determined by cross-agglutinin absorption tests (CAAT); over 300 pathogenic serovars have been identified. Of all the pathogenic genospecies, Leptospira interrogans is the most frequently reported, with the Icterohaemorrhagiae serogroup representing more than half of the leptospires encountered in human infections (33, 38).

In Europe, the terms “Weil’s disease” and “Icterohaemorrhagiae infection” have become synonymous with the acute icteric form of leptospirosis transmitted by the brown rat, Rattus norvegicus. This has arisen since Weil originally described the disease in humans and the first isolations of the causative organism from clinical cases were shown to be what is now recognised as Leptospira interrogans serovar Icterohaemorrhagiae. In 1938 Borg-Peterson (29) described two antigenic subtypes of this serovar represented by the RGA and M20 strains. Each of these subtypes has since been given the status of serovar: RGA has remained Icterohaemorrhagiae and M20 has been named Copenhageni. The majority of strains isolated from acute cases of leptospirosis in high risk areas such as the favelas of large South American cities have been shown to be serovar Copenhageni (21).

The Icterohaemorrhagiae serogroup is currently composed of at least 16 serovars. They are primarily found in rats, which act as a maintenance host for them (6, 7, 27). Serovar identification is rarely performed because the serological methods used to distinguish between serovars is highly time-consuming, and consequently the causative organism of rat-transmitted leptospirosis is loosely referred to as “Icterohaemorrhagiae infection” or “Copenhageni infection”. Rat-borne strains also cause acute disease in animal species other than humans (11, 13).

Pathogenic Leptospira species, the aetiological agents of leptospirosis, infect a broad spectrum of hosts, annually causing 1 million human cases and 60,000 deaths globally (10). In the United Kingdom, leptospirosis and the particularly severe cases presenting as Weil’s disease are uncommon, but almost half such cases are attributed to the Icterohaemorrhagiae serogroup (15) after serogroup-specific serological tests. The situation is similar throughout Europe. Little is known about the strains that are circulating. Isolation and identification of the causative organism are rarely performed because of the fastidious nature of the bacteria and the difficulty of its isolation from cultures. Modern bacterial taxonomy is based on genomic analysis of the organisms. However, mainly for epidemiological reasons, it is still necessary to identify the serological status of Leptospira strains. It helps in understanding the immune response and is important for routine diagnostic purposes and vaccine-specific immunisation. The adaptation of strains to animal species is also at the serovar or sub-serovar level. Therefore, both genetic and serological typing are still of great practical importance. Whole-genome sequencing (WGS) requires sophisticated tools – both to carry out the WGS itself and among the resources needed to link the WGS data to typing and taxonomy (16). For this reason, there is an ongoing requirement for methods that can be routinely used in research laboratories to identify strains and show phylogenetic differences between them. Therefore, this study aimed to characterise 19 Leptospira isolates identified as belonging to the Icterohaemorrhagiae serogroup. The studied bacteria were retrieved from humans and wild and domestic animals in the United Kingdom using both serological and molecular typing methods in order to identify the dominating serovar and show genetic differences between serovars.

This study described serological and molecular characterisation of strains of the Icterohaemorrhagiae serogroup isolated in the United Kingdom. Classification to serogroup level based on cross-agglutination tests using group specific hyperimmune sera showed strong cross-reactivity between the isolates and the Icterohaemorrhagiae serovar but little cross-reactivity with the other 15 serogroups tested, indicating that all 19 strains belonged to the Icterohaemorrhagiae serogroup. It is known that Icterohaemorrhagiae serogroup strains can belong to different Leptospira species (38), but in this study species identification by PCR analysis of the ompL1 gene confirmed that all UK isolates belonged to Leptospira interrogans.

In Leptospira classification, the serovar is a basic taxon, and identification of isolates to serovar level is essential for epidemiological studies and the development of appropriate preventive measures. The classical method for identifying Leptospira serovars is the CAAT (12); however, it is prohibitively time-consuming and costly for the rapid examination of large numbers of isolates and is used mainly on new strains representing new serovars or problematic isolates (30). Since the 1980s, REA has been widely used as a supplementary or alternative method to classify Leptospira cultures. Although there is a poor correlation between the serological and most genotypic classification systems, REA was proved to be a sensitive, reproducible technique for the identification of isolates to serovar or even subserovar level (1–4, 9, 36). In our study 16 Icterohaemorrhagiae serogroup reference strains were used to identify the digesting enzyme and discriminate between serovars. The results revealed that most of the serovars had a unique restriction fragment length polymorphism pattern when digested with the EcoRI and HhaI enzymes, and only four serovars, namely Copenhageni, Icterohaemorrhagiae, Monymusk and Budapest had patterns with strong mutual resemblance. Further REA typing of the field strains used in the study showed the same EcoRI and HhaI restriction patterns for all tested isolates, and they were compatible with the DNA profiles of those four very closely related serovars. These results correlated with those of some previous studies, revealing difficulties in genetic discrimination between those strains (8, 17–19). Moreover, using monoclonal antibody typing, Kobayashi et al. (22, 23) also showed a very close antigenic similarity of those strains, which led the authors to put those four serovars in one subgroup. Studies using a panel of monoclonal antibodies prepared by Korver et al. (26) found that the F12C3 mAb was able to reduce the group of highly similar serovars from four to two, namely Icterohaemorrhagiae and Copenhageni. Further typing using F70C24 and F70C14, each specific only for one serovar, revealed that 16 of the isolates were identified as serovar Copenhageni and only 3 as serovar Icterohaemorrhagiae. No association between serovar and specific animal species was observed.

Recent whole-genome sequencing analysis of 67 isolates belonging to the L. interrogans Copenhageni and Icterohaemorrhagiae serovars identified 1,072 SNPs and 258 indels (34). The phylogenetic analysis during this study revealed that both serovars are closely related but showed distinct spatial clustering. In the genetic era of bacterial identification, many molecular tools have been employed as alternatives to classical methods for the identification and classification of Leptospira or to evaluate genetic diversity among strains, such as multiple-locus variable-number tandem repeat analysis and multilocus sequence typing (MLST). In our studies we employed MLVA, which was shown to have a higher discriminatory power and was more concordant with serotyping than MLST for L. interrogans (24, 25). Analysis of 13 MLVA loci revealed genetic diversity between the studied isolates and their clustering into five different types. Strains belonging to the dominant groups A and B corresponded to the profile of both reference strains. Group E was constituted only by rat isolates, which were typed as serovar Copenhageni and had a unique VNTR profile consistent at loci V9 and V29. One of the rat isolates (S199/2) belonged to the Icterohaemorrhagiae serovar; however, its genetic profile was the same as that of strain M20, which is the reference strain for the Copenhageni serovar. When MLVA results were compared with those of the monoclonal antibody typing, this diversity did not correspond to the serovar status of the isolates. This is in agreement with a recent study conducted in China, where 128 serogroup Icterohaemorrhagiae strains were analysed using another genetic technique – MLST – and it was found that some isolates belonged to the same sequence type (ST) but different serovars. Conversely, some serovars were usually associated with more than one ST (38). This further substantiates the non-correlation of such genetic diversity to serological status and the unsuitability of serovars as indicators of genetic relatedness. The diversity of serovars is most likely to be due to horizontal gene transfer, even within these closely related examples. The studies referred to earlier of 67 different strains of L. interrogans of the Copenhageni and Icterohaemorrhagiae serovars, which showed 1,072 SNPs and 258 indels, only found one insertion of the lic12008 gene within the Icterohaemorrhagiae serotype, which may be important in differentiating these two serovars (34). This was also confirmed in our study.

However, the biological significance of the other genetic variations found in our studies is unclear. They could be useful in epidemiological tracking analysis and have potential application in the understanding of leptospiral molecular epidemiology. Clonality studies in outbreaks and rapid determination of the source of infection is critical in limiting the spread of infection and important to the development of appropriate prevention strategies. Zhang et al. (38) showed some clustering of serogroup Icterohaemorrhagiae strains from specific geographical regions in China and that the genetic diversity of Chinese strains was generally different to that of other countries. When disease outcome is considered rather than geography, there is no apparent difference between the Icterohaemorrhagiae and Copenhageni serovars. The genetic diversity revealed between strains nevertheless undoubtedly affords a better understanding of the diversity of those which are pathogenic and could also help identify the genotypes responsible for severe, often lethal infections (11).

In summary, the results demonstrate that monoclonal antibody typing is a useful method for discriminating between these two closely related serovars and is more cost effective than other methods used. Restriction endonuclease analysis which is based on the whole Leptospira genome does not reveal any differences in the DNA profiles of either reference strains or studied field isolates. In contrast to REA, MLVA revealed the diversity of VNTR loci among the strains, but this was not concordant with the serotyping results. Additionally, differences were brought to light in the lic12008 gene sequence which may be helpful in serovar identification. The results of the study indicate that the Icterohaemorrhagiae and Copenhageni serovars have both been isolated in UK. Most of those tested in the study (84.2%) were identified as the Copenhageni serovar. These data may help identify better options for infection prevention measures and contribute to the epidemiological characterisation of common strains circulating among infected animals and humans.