Molecular characterization of a re-emergent Brugia malayi parasite in Sri Lanka, suggestive of a novel strain with close nucleotide homology to Brugia pahangi

Background Brugian lariasis has re-emerged in Sri Lanka after four decades of quiescence. As microscopy alone was insucient for ascertaining the species identity of the re-emerged sub-periodic Brugia spp. parasite, molecular speciation was performed. The transmission dynamics of the parasite was studied by entomological procedures. Methods Human blood samples positive for Brugia spp. microlariae (MF) (n=8) were collected and DNA extracted using ReliaPrep™ Blood DNA Miniprep System (modied). Polymerase chain reaction (PCR) was performed with pan-larial primers specic for the internal transcribed spacer region 2 (ITS2) of the ribosomal DNA (rDNA) of MF. Results Of those tested, seven (87.5%) yielded a band at 615bp establishing the species identity of the re-emerged larial parasite as B. malayi. Comparison of the ITS2 region gene sequences of B. malayi MF isolated from humans (n=2), dogs (n=3) and cats (n=6) with GenBank sequences revealed a higher sequence homology with B. pahangi than B. malayi, but phylogeny was closer to B. malayi. A total of 82 mosquitoes of genus Mansonia comprising of M. annulifera (65), M. uniformis (14) and M. indiana (3) were collected by cattle-baited traps. Mosquito dissections identied 17 infected mosquitoes: one M. uniformis (7.14%) and 16 M. annulifera (24.6%). The DNA extracts of all infected Mansonia mosquitoes elicited the 615bp band on pan-larial primer specic PCR. Conclusions re-emergent B. a genetic variant or a novel species closely related to B. malayi and B. pahangi. Mansonia spp. mosquitoes variant


Background
Lymphatic lariasis (LF), is a neglected tropical disease estimated to have affected 40 million people worldwide, with an at risk population of 893 million people residing in 49 endemic countries [1]. Although infection is not fatal, it is a leading cause of permanent disability and was targeted for elimination as a public health problem by 2020 [1].
Three species of larial worms, namely Wuchereria bancrofti, Brugia malayi (nocturnally periodic strain) and Brugia timori are known to cause classical lymphatic lariasis in humans. Of the three human larial parasites, W. bancrofti is the most prevalent, causing 90% of infections worldwide, while the rest are attributed to B. malayi (prevalent in Southeast Asia and in South-western parts of India) and to a lesser extent B. timori (limited to the islands of Timor-Leste in eastern Indonesia) [1]. A zoonotic strain of B. malayi (sub-periodic strain), which is a natural parasite of a variety of wild and domestic animals (cats, dogs, monkeys, slow-lorises) has been documented to cause accidental zoonotic infections in humans [2] [3] [4] [5]. Serological and molecular evidence of human infections by another zoonotic larial worm, B. pahangi (a natural parasite of felines) has been reported from Kuala Lumpur, with clinical manifestations consistent with lymphatic lariasis [6]. These zoonotic lariae (sub-periodic B. malayi and B. pahangi) also contribute to the human disease burden especially in the Northeastern United States, South America and Southeast Asian region [2] [3] [7]. Both W. bancrofti and B. malayi (nocturnally periodic strain) were prevalent in Sri Lanka in the past [8]. Vector control activities focussed on clearance of aquatic vegetation (a requirement for breeding of Mansonia spp. mosquitoes) resulted in clearance of B. malayi infections from the country in late nineteen sixties [8]. Subsequently ve rounds of single annual mass drug administration (MDA) was implemented from 2002-2006 in the three provinces (Western, Northwestern and Southern) endemic for bancroftian lariasis using diethylcarbamazine citrate (300 mg) combined with albendazole (400 mg). In 2016 Sri Lanka was categorized as a country that had eliminated LF as a public health problem, following ful lment of critical criteria stipulated by World Health Organization (WHO) for veri cation of elimination [9]. However, surveillance activities during the phases of post-MDA and post-elimination revealed the sporadic occurrence of brugian lariasis in all three LF endemic provinces [10] [11] [12] [18]. Periodicity studies revealed that the re-emergent Brugia species micro laria (MF) exhibited nocturnal sub-periodicity, implying a strain of different origin [10]. Zoonotic surveillance in areas affected by human brugian lariasis has revealed a high prevalence of B. malayi MF among dogs and cats [13]. The existence of another zoonotic larial species, Brugia ceylonensis among local canines has also been documented in the past [14].
Exact speciation of the re-emerged Brugia parasite was important as zoonotic B. malayi, B. pahangi and B. ceylonensis have all been implicated in accidental human infections [15] [4] [5] [6]. The morphological similarity of the MF of Brugia spp. necessitated a molecular based approach (PCR and sequencing) for species identi cation [15].
Studies on the genetic variability of B. malayi indicate that parasite isolates from different regions diverged signi cantly [16]. Gene sequences of zoonotic Brugia species are sparse in major data bases and with increasing case numbers there will be an increased need for genomic repositories [17]. Such information may support the genus or species clari cations based on consensus gene targets. Thus, sequencing of the Brugia spp. MF that were isolated from humans and animals of Sri Lanka were carried out to construct the phylogeny and assess their relationship to other Brugia species.
Knowledge of the transmission dynamics of the newly emerged parasite is important for control. The vector preferences of the re-emerged parasite may differ from the nocturnal periodic strain of B. malayi documented in the past; thus, entomological investigations were performed for clari cation of the vector species.
Against this background of events, a comprehensive analysis of the re-emerged Brugia spp. parasite of Sri Lanka, as to its species identity, the phylogeny and genetic relatedness to other Brugia species and its vector preferences were studied using conventional parasitological and molecular methods.
Human sample collection and DNA extraction Human blood samples positive for Brugia sp. MF by thick night blood smears (NBS) were collected from four districts belonging to the three endemic provinces, (Gampaha and Kalutara from the Western province n = 5; Puttalam from the North-western province n = 1 and Galle from the Southern province n = 1) and from a case detected from a non-endemic district in the North Central province (Anuradhapura). Blood samples positive for Brugia antibodies by the Brugia rapid test were also included in the molecular analysis. The samples were stored at -20° C in EDTA prior to DNA extraction. The MF in blood was concentrated by Nuclepore® membrane ltration and the lter membranes with trapped MF were used for DNA extraction. The DNA was extracted using the ReliaPrep™ Blood DNA Miniprep System (Catalogue number A5082) according to the manufacturer's instructions [18] with some modi cations.
Brie y, a volume of 180 µl of PBS was added to polycarbonate membranes with MF and incubated for 45 minutes while shaking (300 rpm). A volume of 20 µl Proteinase K and 200 µl cell lysis buffer solution was then added to the content and incubated at 56 °C for 10 minutes and to which 250 µl of binding buffer were added. The content was then transferred to ReliaPrep™ binding columns and centrifuged at 5000Xg. The binding columns were transferred to fresh collection tubes and 500 µl of column wash solution was added and centrifuged for three minutes at 5000Xg and this was then repeated twice. The binding columns were nally eluted with 50 µl Nuclease-Free water. The DNA concentrations was measured by uorometry according to the instructions (Manual #TM396: www.promega.com/protocols/) and stored at -20 °C until used.

DNA ampli cation
The procedure stated by Rishniw et al., 2006 was used with some modi cations using Promega reagents [24]. The pan-larial primers (DIDR-F1 5'-AGT GCG AAT TGC AGA CGC ATT GAG-3' and DIDR-R1 5'-AGC GGG TAA TCA CGA CTG AGT TGA-3') that spanned the ITS2 of the rDNA designed by Rishniw et al., 2006 was employed to amplify the target DNA region [24]. Known positive and negative controls were included in each PCR reaction. The PCR procedure consisted of an initial denaturing step at 94 °C for 2 min and 32 cycles of denaturing (30 s at 94 °C), annealing (30 s at 58 °C) and extension (30 s at 72 °C) and nal extension (7 min at 72 °C) in an Eppendorf Mastercycler thermal cycler.
Discrimination of the six species was based on the size of the ampli ed PCR products. DIDR-F1 and DIDR-R1 primers ampli ed 484 bp, 542 bp, 578 bp, 584 bp, 615 bp and 664 bp products from Diro laria repens, Diro laria immitis, Acanthocheilonema reconditum, Acanthocheilonema dracunculoides, B. malayi, and B. pahangi respectively [19]. Canine and feline sample processing The zoonotic surveillance for larial parasites were done at three locations in the districts of Gampaha (Weliweriya and Wattala) and Puttalam (Madampe) of Western and Northwestern provinces respectively and molecular speciation of the zoonotic Brugia parasite was performed by PCR using pan larial primers speci c for ITS2 region and species con rmed as B. malayi as detailed previously [13].

Mosquito vector analysis
Sample collection, DNA extraction and ampli cation The mosquito surveillance was carried in the district of Gampaha in the same areas where zoonotic surveillance was done, using one cattle-baited trap in each location. Mosquitoes of Mansonia sp. were identi ed using morphological keys. Heads and thoraces of all Mansonia sp. mosquitoes were dissected to identify larial larvae.
DNA of larial larvae was extracted using the MightyPrep reagent (Takara Bio Inc, Japan) according to manufacturer's guide with some modi cations as follows: the mosquito heads and thoraces that were positive for larial larvae were crushed and mounted temporarily on glass slides and ushed with 200µL of MightlyPrep reagent into a microcentrifuge tube. The lysate was homogenized by hard vortexing for 10 seconds. New pipette tips were used each time to prevent cross contamination of samples. The lysates were incubated at 95 °C for 10 minutes. Subsequently the sample lysates were allowed to cool down to room temperature. The cooled lysates were hard-vortexed for another 10 seconds. Finally, the samples were centrifuged at 12,000 rpm for 10 minutes and stored at − 20 °C until used for PCR. DNA ampli cation was done using the same set of primers and procedure used for human and animals' samples.

Homology comparison
The DNA sequence homology was analysed using BLASTN. Mega 07 (https://rom yhe studywww.ncbi.nlm.nih.gov/pubmed/27004904) tool and the analysed data was used to generate the phylogenetic tree. Based on NCBI (https://www.ncbi.nlm.nih.gov/) blast, closely related sequences to Brugia sp. genes were retrieved.

PCR
Seven of the eight human blood samples positive for Brugia sp. MF by NBS that were tested by PCR, produced a band of 615 bp speci c for B. malayi (Fig. 1). The sample that failed to elicit a PCR band had very low micro laraemia (1MF/slide). All human blood samples that were only positive for anti-Brugia antibodies by BRT (n = 9) failed to elicit a PCR band. The zoonotic samples also elicited bands of 615 bp speci c for B. malayi (Fig. 2).

Phylogenetic analysis
A homology search carried out on GenBank using the BLASTN of NCBI and the ampli ed ITS2 sequences of the B. malayi MF of humans, dogs' and cats' in this study showed the sequences to have higher homology with B. pahangi than B. malayi (Fig. 3).
Phylogenetic analyses of the sequenced ITS2 gene of B. malayi MF from human, canine and feline blood samples clustered between B. pahangi and B. malayi. All were in a monophyletic group originating from the same ancestor except for one sample (No 5), which fell to a more distant location. The sequences obtained in the current study formed a separate cluster, with three sub-clusters within it (Fig. 3). Phylogenetic tree based on the nucleotide sequences of the ITS2 gene.

Discussion
Molecular characterization con rmed that the re-emerged sub-periodic Brugia parasite was the same as that circulating among dogs and cats in the region. Sequencing the ITS2 region of the rDNA of both human and zoonotic parasites indicated that the infecting B. malayi had a higher sequence homology to B. pahangi than B. malayi. However, phylogeny reconstruction based on phylogenetic tree indicated that the phylogeny of the parasite was closer to B. malayi.
The possibilities for this ambiguous outcome are two-fold. The rst is that the re-emergent species is a naturally occurring hybrid species of B. malayi and B. pahangi. The second possibility is that it is a hitherto undetected genetic variant or a novel species of B. malayi. Although interspecies hybrization of Brugia species has been reported under experimental conditions with production of viable offspring (micro lariae) [21], it has not been reported to occur under natural conditions. Under experimental hybridization only the female progeny was fertile, requiring mating with parental stock for maintenance of the hybrid strain [21]. Analysis of zoonotic blood specimens by PCR failed to detect any other brugian larial parasites other than B. malayi. Thus, the requirement for mating between hybrid females and males of parental stock to produce viable MF cannot be ful lled. This raises the likelihood that the reemergent B. malayi is a novel strain or genetic variant rather than an evolutionary mature hybrid of B. pahangi and B. malayi.
The presence of a novel zoonotic Brugia species closely related to B. malayi and B. pahangi has been previously queried on molecular characterization of the rDNA sequences of a larial nematode and comparison with known gene sequences in the GeneBank [22]. These larial nematodes were isolated from the inguinal lymph nodes of a patient based in the city of New York with an extensive travel history to Central America and the Caribbean [22].
A canine survey in Kerala, had reported the presence of B. malayi like MF that had elicited a histochemical staining pattern consistent with that of B. malayi. The Hha1 primer PCR product of this B. malayi like parasites were cloned and sequenced (2 clones, Accession numbers JN 601136 and JN 601137) and the phylogeny revealed that the B. malayi like parasite was genetically closer to B. pahangi suggesting the existence of a novel species / genetic variant closely related to B. malayi and B. pahangi in the natural environment [23].
The possibility of this variant strain of B. malayi being the species documented as B. ceylonensis requires due consideration. The adult larial nematodes of B. ceylonensis was rst described in 1962 in the lymphatic glands of dogs from Sri Lanka. It was documented as a novel species closely related to B. patei based on morphological characteristics [24]. The sheathed MF of B. ceylonensis had a close resemblance to MF of B. malayi (length of cephalic space) [24].
Most zoonotic brugian infections are associated with unapparent or mild symptoms and serologic testing for Bm14 IgG4 is not reliable, thus the recorded cases may be an underestimate of the true burden of infection. This variant B. malayi appears to have a limited capacity to infect humans as case numbers were relatively low compared to the heavy zoonotic reservoirs of infection in Sri Lanka [13]. This could be attributed to low transmissibility owing to vector characteristics (low a nity of Mansonia sp. mosquitoes for human blood) or the enhanced immune response generated by a poorly adapted zoonotic parasite conferring natural resistance to infection [10]. The rising number of cases in the recent past may be indicative of the parasite's potential to evolve and adapt to humans or may be a re ection of the research interest generated by the PELF.
Sentinel surveillance of animals and xenomonitoring may serve in de ning populations at risk of infection. Vector identi cation is important not only for implementing the appropriate vector control measures but also for mapping the distribution of infection. M. annulifera and M. uniformis are zooanthropophagic mosquito species in Sri Lanka that were implicated in the transmission of periodic B. malayi in the past [25] [26]. This study con rmed their capability of transmitting the variant B. malayi (subperiodic) in Sri Lanka.
The ITS regions 1 and 2 have been used by many investigators for studies on phylogenetic reconstruction, genetic variability and divergence of closely related taxa of a wide range of organisms [27] [28] [29]. The molecular characterization of the MF of this variant B. malayi isolated from humans was limited to two samples and analysis was focussed only on the ITS2 region of the rDNA are some of the limitations of this study. A more comprehensive genome wide analysis of rDNA as well as mitochondrial DNA of MF and adult stages; and chromosome number, gonad organization of this variant B. malayi may be required for clari cation of the taxonomy of the parasite. Further studies on B. malayi parasites isolated from humans and animals from different geographical locations in the country as well as comprehensive entomological surveys which cover a wide array of mosquito species are indicated in order to characterise the novel variant zoonotic B. malayi.

Conclusion
The re-emerged B. malayi parasite in Sri Lanka is a novel genetic variant which has a closer genetic homology to B. pahangi than B. malayi. Domestic canines and felines were identi ed as the zoonotic reservoirs and Mansonia sp. mosquitoes (M. annulifera and M. uniformis) were implicated as vectors. This re-emerged parasite could well pose a threat to the LF elimination status of the country because vectors and infectious zoonotic reservoirs are present in abundance. The application of molecular identi cation techniques will be invaluable for clari cation of taxonomy, epidemiology and ecology of this novel genetic variant B. malayi.

Declarations
Ethics approval and consent to participate Ethical clearance for the study was obtained from the Ethics Review Committees of the Faculty of Medicine University of Kelaniya (P/108/09/2016) and the Medical Research Institute (40/2016). Informed written consent was obtained from all adult participants and one of the parent or guardian for less than 18 years old participants. In additional to the consent, written assent was obtained from participants between 7 and 18 years of age. Informed written consent was obtained from owner of the cats and dogs. Consent for publication Figure 1 Gel electrophoresis of the PCR products of human samples; The gel image shows the results of the PCR ampli cation of samples 1 to 7 using Pan-larial primers (Lanes 1 to 5-human samples, Lane 6-positive control, Lane 7-negative control and Lane L-50bp DNA marker.)  Phylogenetic tree constructed from partial rDNA sequencing, (ITS2 region); The phylogenetic tree was constructed using the aligned sequences with Mega 07 tool. The lengths of the horizontal lines are proportional to the minimum number of nucleotide differences required to join nodes. Vertical lines are for spacing branches and labels. Numerical numbers in the nodes are bootstrap con dence intervals which were calculated by 1000 heuristic search replicates. The reference strain was given accession numbers (bold and Italic). {Cluster A -4 (Dog -Pubudugama) and 26 (Human -Pubudugama); Cluster B -8, 6 (Cat-