Integrated Histological and Molecular Analysis of Filarial Species and Associated Wolbachia Endosymbionts in Human Filariasis Cases Presenting Atypically in Thailand

ABSTRACT.

Atypical presentations of filariasis have posed diagnostic challenges due to the complexity of identifying the causative species and the difficulties in both diagnosis and treatment. In this study, we present the integrative histological and molecular analysis of seven atypical filariasis cases observed in regions of nonendemicity of Thailand. All filariasis cases were initially diagnosed based on histological findings. To confirm the causative species, molecular characterization based on both filarial mitochondrial (mt 12S rRNA and COI genes) and nuclear ITS1 markers was performed, together with the identification of associated Wolbachia bacterial endosymbionts. Among the cases studied, Brugia pahangi (N = 3), Brugia malayi (N = 1), Dirofilaria sp. “hongkongensis” (N = 2), and a suspected novel filarial species genetically related to Pelecitus copsychi (N = 1) were identified. By targeting the 16S rRNA gene, Wolbachia was also molecularly amplified in two cases of infection with Dirofilaria sp. “hongkongensis.” Phylogenetic analysis further revealed that the detected Wolbachia could be classified into supergroups C and F, indicating the high genetic diversity of this endosymbiont in Dirofilaria sp. “hongkongensis.” Furthermore, this study demonstrates the consistency between histological findings and species identification based on mitochondrial loci rather than on the nuclear ITS1. This suggests the utility of mitochondrial markers, particularly COI, as a highly sensitive and reliable diagnostic tool for the detection and differentiation of filarial species in clinical specimens. Precise identification of the causative species will facilitate accurate diagnosis and treatment and is also essential for the development of epidemiological and preventive strategies for filariasis.

INTRODUCTION

Filarial worms are nematode parasites of the superfamily Filarioidea that cause the disease called filariasis.¹ Transmission of filariae in this nematode superfamily occurs via blood-feeding arthropods such as mosquitoes, black flies, and deer flies.² Eight species of filariae have humans as their definitive host, whereas the remaining species require other animals as their definitive hosts.^1,3 Several species of filariae can cause infections in humans, mammals, birds, reptiles, and amphibians.^3,4

Human filariasis is categorized into three groups based on its location in the human body: lymphatic filariasis, serous cavity filariasis, and subcutaneous filariasis.^1,3 Human lymphatic filariasis is caused by three main species of filariae, including Wuchereria bancrofti, Brugia malayi, and Brugia timori.^1,3 Serous cavity filariasis is caused by Mansonella perstans and Mansonella ozzardi, with the adult stages of these worms residing in various human body cavities such as the peritoneal, pericardial, and pleural cavities.³ The adult stages of certain filariae species, such as Loa loa, Mansonella streptocerca, and Onchocerca volvulus, inhabit the subcutaneous tissue beneath the skin, leading to the manifestation of subcutaneous filariasis.³

Besides human filariasis, zoonotic filariae of animal origin can rarely infect humans and cause various clinical manifestations.⁴ In addition, animal filariae of the genus Brugia, Dirofilaria, Onchocerca, Dipetalonema, Loaina, and Meningonema can accidentally infect humans and undergo incomplete development in human tissues.⁴ When animal filariae infect humans, the infection may be asymptomatic or may elicit various degrees of immune response from the host.⁴ Animal filariae have been found in various human tissues, including subcutaneous tissues, heart, lungs, lymphatic system, eyes, and central nervous system.⁴ However, subcutaneous tissues are the most frequently reported recovery site.⁴ Microscopic examination of key morphological features can sometimes enable the identification of filarial species at the genus level.⁴ However, this can be challenging, particularly in cases where worms are degenerating or only partial parts are recovered. Therefore, molecular characterization for the identification of unknown filarial species appears to be essential. Rapid diagnosis of filarial species is crucial for making a treatment decision, conducting epidemiological surveillance, and controlling disease transmission, especially in areas of nonendemicity.

Although several atypical filariasis cases have been previously reported, not all studies have molecularly characterized the causative filarial species.^5–9 In Southeast Asia, there has been a growing number of reported filariasis cases where worms are incidentally discovered in various human tissues.^9–14 Therefore, it is crucial to identify the causative species at the molecular level. This is essential for monitoring the potential causes of emerging filariasis and for understanding the associated risk factors behind the increasing number of reported cases. In addition, Wolbachia has been known to be an endosymbiont of filarial species essential for several metabolic pathways of the worm, including reproduction and embryogenesis.¹⁵ The presence of Wolbachia may help identify the causative filarial species.¹⁶

In this study, we present findings on seven atypical filariasis cases in Thailand, wherein adult worms of filarial species were incidentally discovered in various sites within the human body. These cases’ clinical presentations and histological findings were described, and the causative species were further identified by histological and molecular characterization with phylogenetic analysis. Furthermore, we also identified and phylogenetically characterized the Wolbachia bacterial endosymbiont of filarial species, which may prove valuable for determining the causative filarial species in clinical samples.

MATERIALS AND METHODS

Patients, specimen processing, and ethical consideration.

This study was conducted at the Department of Parasitology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. The study protocol was approved by the Siriraj Institutional Review Board of Research involving human subjects (SIRB) (COA no. Si 397/2019). A waiver of consent was approved by the SIRB, as this research involved no more than minimal risk to the subjects. The waiver did not adversely affect the rights and welfare of the owners of the specimens. Seven cases of human filariasis that were incidentally diagnosed during physical examinations or through histological findings at Siriraj Hospital or other affiliate hospitals in Bangkok, Thailand from January 2015 to June 2022 were included in this study. The medical records were reviewed to obtain the patient’s medical history, including age, sex, symptoms, and the site from which the worms were recovered. Paraffin-embedded tissue sections containing a worm obtained directly from the patients were examined histologically using hematoxylin and eosin staining. Molecular identification of the causative species was also performed.

DNA extraction from the specimens.

DNA was then extracted from the paraffin-embedded sections (cases 1, 2, 3, 5, and 6) or worm samples (cases 4 and 7) and subjected to molecular identification of the causative filarial species and associated Wolbachia endosymbionts. DNA extraction was performed using a QIAmp DNA mini kit (Qiagen, Hilden, Germany) in accordance with the manufacturer’s protocol. The purity and concentration of DNA from each sample were measured using a NanoDrop 2000 UV-vis spectrophotometer (Thermo Fisher Scientific, Waltham, MA). DNA samples were used as templates for polymerase chain reaction (PCR) amplification.

PCR amplification and DNA sequencing for filarial and Wolbachia species identification.

Filarial species identification was performed using primers targeting filarial mitochondrial 12 S rRNA (mt 12 S rRNA), cytochrome c oxidase subunit 1 (COI), Brugia internal transcribed spacer 1 (ITS1), and Dirofilaria ITS1 genes with PCR cycling conditions as previously described.^17–19 Wolbachia identification was performed using primers targeting Wolbachia 16S rRNA, Wolbachia surface protein (wsp), and ftsZ genes as previously described.^20,21 All PCRs were performed using a Platinum^® Taq DNA polymerase, high-fidelity kit (Thermo Fisher Scientific). All PCRs were set up in a volume of 25 µL containing 2.5 µL of 10× high-fidelity PCR buffer, 0.5 µL of each 10 µM primer, 10 ng of DNA, 1 µL of 50 mM MgSO₄, 10 mM deoxynucleoside triphosphate, 0.1 µL of Platinum Taq DNA polymerase, and H₂O up to 25 µL. The primers and PCR cycling conditions are listed in Supplemental Table 1.

Because of the small size of the mt 12 s rRNA amplicon, PCR products were cloned into the pGEM^®-T easy vector systems (Promega Corp., Madison, WI) to obtain the full length of the amplicon and increase the accuracy of nucleotide reading. The recombinant plasmids were then transformed into Escherichia coli DH5α competent cells (Thermo Fisher Scientific) and plated onto Luria-Bertani (LB) agar plates containing ampicillin, X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside), and IPTG (isopropyl-β-D-thiogalactopyranoside) for blue/white colony selection. White colonies were screened for the presence of inserts by colony PCR. Three positive clones from each clinical sample were then grown in LB medium with ampicillin for plasmid propagation and isolation. Sanger plasmid sequencing was performed bidirectionally using the M13 universal forward and reverse primers.

The PCR products of the other genes, including COI, Brugia ITS1, Dirofilaria ITS1, Wolbachia 16S rRNA, wsp, and ftsZ genes, were verified on 2% agarose gel, stained with SafeView^TM nucleic acid stains (Applied Biological Materials, Richmond, British Columbia, Canada), and visualized using a Gel Doc™ EZ gel documentation system (Bio-Rad Laboratories, Hercules, CA). The purified amplicons of these genes were then subjected to direct Sanger sequencing using forward and reverse primers as previously used in the PCR step.

Phylogenetic tree analysis.

Nucleotide sequences of the partial mt 12 S rRNA, COI, and ITS1 genes were aligned against those references in the GenBank database using the BLASTn program (National Center for Biotechnology Information, Bethesda, MD) for homology search.²² The highest degree of sequence homology was based on the highest percent identity of the sequence with the highest query coverage. Phylogenetic trees were constructed using the maximum likelihood method with a bootstrap value of 1000 replicates. Evolutionary analyses were performed with the best substitution models using Molecular Evolutionary Genetic Analysis (MEGA) software version 11.0.²³

RESULTS

Clinical and histopathological data of atypical filariasis cases.

From 2015 to 2022, seven cases of filarial infection with atypical presentations were referred to or diagnosed at the Faculty of Medicine, Siriraj Hospital, Thailand. The mean age of the atypical filariasis cases was 64 ± 13.2 years (range, 40–79 years). The clinical characteristics of the cases were as follows; subcutaneous nodule (4/7, 57.1%), periorbital mass (1/7, 14.3%), breast mass (1/7, 14.3%), and intraocular lesion (1/7, 14.3%). Skin manifestations as a movable subcutaneous mass or nodule with a history of migratory lesions were the main presentations found in this study. Most of the patients were female (6/7, 85.7%). There was no significant history of infection. Most of the patients (4/7, 57.1%) lived in the central region of Thailand. The duration of clinical manifestation ranged from 2 weeks to 2 months. Demographic data, clinical characteristics, diagnosis, treatment, and outcome of each case are summarized in Table 1. Blood microfilaria and serum filarial IgG4 tests were conducted only for cases 3, 4, 5, and 7, and all were negative. The details of clinical characteristics, diagnosis, and treatment of cases 3, 4, and 5 have been previously published by our group.^11–13 Additional molecular characterization and phylogenetic analyses of the causative species in cases 3, 4, and 5 were performed in this study.

Table 1.

Clinical characteristics of seven filariasis cases with atypical presentations in Thailand from 2015 to 2022

Case	Age (years)	Sex	Year	Hometown (region)	Characteristic Manifestations	Site	Duration	Other Findings	Treatment	Outcome	GenBank Accession No. based on Molecular Characterization
											Filarial 12 S rRNA	Filarial COI	Filarial ITS1	Wolbachia 16S rRNA
1	79	F	2015	N/A	Movable subcutaneous nodule	Left thigh	NA	Eosinophilia	Excision	Complete recovery	MT135222.1	NA	OQ780598.1	NA
2	73	F	2019	Nonthaburi (Central)	Subcutaneous nodule with pus	Right arm	6 wk	None	Excision	Complete recovery	MT135223.1	OQ713625.1	OQ780599.1	NA
3	63	F	2019	Nonthaburi (Central)	Right breast mass observed by mammogram	Breast mass	3 mo	Left breast mass: invasive ductal carcinoma/no eosinophilia	Excision	Complete recovery	MT887286.1*	OQ713626.1*	MT732324.1	NA
4	70	F	2020	Nonthaburi (Central)	Periorbital mass	Adjacent to the medial canthus of the left orbit	2 wk	No eosinophilia	Excision	Complete recovery	MT584737.1†	OQ713627.1	OQ780600.1	NA
5	54	F	2020	Kanchanaburi (West)	Subcutaneous nodule	Left shoulder	1 mo	Eosinophilia	Excision	Complete recovery	MW051483.1‡	OQ713628.1	OQ780601.1	OQ780644.1
6	40	M	2021	Chachoengsao (Central)	Decreased visual acuity, marked conjunctival injection with a mobile parasitic-like organism in the anterior chamber with increased cellular infiltration	Left eye	3 wk	No eosinophilia	Injection of a viscoelastic substance in the anterior chamber of the left eye followed by surgical removal of the worm	Complete recovery	OK429322.1	OQ713629.1	OQ780602.1	NA
7	69	F	2022	Sa Kaeo (East)	1.5 cm moveable subcutaneous nodule adhered to the skin	Left paranasal	2 mo	Underlying rheumatoid arthritis and hypertension	Excision	Complete recovery	OQ732706.1	OQ713630.2	OQ780603.1	OQ780645.1

F = female; M = male; mo = month; N/A = not available; NA = no amplification; wk = week.

^*Previously published data (Nunthanid et al.¹¹).

^† Previously published data (Thongpiya et al.¹³).

^‡ Previously published data (Thongpiya et al.¹²).

For histopathological evaluation, sections from cases 1 and 4 showed the presence of a smooth and thin cuticle, prominent lateral chords, a few well-developed, low, broad coelomyarian muscle cells per quadrant, and uterine tubes with microfilariae or developing cells compatible with female Brugia species (Figure 1A–D). In cases 5 and 7, the histological sections show the presence of a multilayered cuticle with outer longitudinal ridges, prominent musculature, a pair of uterine tubes, and an intestine in the body cavity typical of female Dirofilaria species (Figure 1E, G, and H). Unfortunately, the morphological features of case 6 could not be determined because of the degeneration of the parasite specimen (Figure 1F).

Figure 1.

Histological findings of filarial nematodes in human cases in the present study. (A) Case 1. A smooth thin cuticle with a few well-developed, low, broad coelomyarian muscle cells per quadrant of the worm surrounded by lymphocytic infiltration was observed. A pair of uterine tubes (U) were filled with microfilariae. (B and C) Cases 2 and 3. Prominent lateral chords and two uterine tubes (U) with developing cells inside were also seen. The arrowhead indicates the intestine. (D) Case 4. A longitudinal section through the worm showed a smooth thin cuticle and two uterine tubes (U) with developing cells. (E) Case 5. Longitudinal cuticular ridges, prominent muscle cells, and two uterine tubes (U) with developing cells inside (arrowhead) were seen. The worm was surrounded by intense eosinophilic lymphoplasmacytic granulomatous reactions. (F) Case 6. A worm with a degenerated cuticle was observed. (G and H) Case 7. A whitish nematode with tapered ends, approximately 12 cm long, was recovered. Cuticular ridges and released egg cells were observed. The section showed a multilayered cuticle with external ridges, conspicuous musculature and lateral chords, and uterine tubes (U) filled with developing eggs inside.

Molecular characterization and phylogenetic analysis of filarial mt 12 S rRNA gene.

Identification of the causative filarial species is crucial for correct diagnosis, appropriate treatment, and epidemiologic disease surveillance. Therefore, we first identified the causative filarial species by PCR targeting of the conserved filarial mt 12 S rRNA gene. Sequences from all cases were aligned using the BLASTn program, phylogenetically analyzed, and submitted to the GenBank database. Based on the mt 12 S rRNA sequences, the causative species in case 1 and case 2 were 96.33% identical to B. pahangi (AP017680.1) and 100% identical to B. pahangi 46YT MNHN (KP760318.1), respectively. The filarial species found in case 3 was 100% identical to B. malayi L01 (OQ727407.1). The causative species in case 4 was 100% identical to B. malayi 8YT MNHN (KP760317.1). In case 5, the filarial species identified was Dirofilaria sp., with 99.12% identity to Dirofilaria sp. “hongkongensis” HKU1-HKU9 (KY750550.1). The causative filarial species in case 6 was 98.08% identical to Pelecitus copsychi 268-5 MNHN (OK480977.1). For case 7, the sequence was 100% identical to Dirofilaria sp. “hongkongensis” HKU1-HKU9 (KY750550.1). As demonstrated in Figure 2, phylogenetic analysis of the mt 12 S rRNA gene indicates that the causative filarial species in cases 1 and 2 are closely related to B. pahangi, whereas those in cases 3 and 4 are likely to be B. malayi. In cases 5 and 7, filariae of the genus Dirofilaria appeared to be the causative species. In case 6, P. copsychi was identified. However, the mt 12 S rRNA sequences of Dirofilaria spp., including Dirofilaria repens, Dirofilaria immitis, and Dirofilaria sp. “hongkongensis” are highly similar to each other and phylogenetically clustered in the same clade, making it difficult to differentiate these closely related species. Additional gene-based analysis is therefore required to confirm the causative species.

Figure 2.

Phylogenetic analysis of the filarial mt 12 S rRNA sequences among the causative filarial species reported in this study (in bold). The collection site, host, and country are listed next to the species. The maximum likelihood tree was constructed using the Tamura 3-parameter model with gamma distribution and invariant sites.

Molecular characterization and phylogenetic analysis of filarial COI.

Molecular identification and phylogenetic analysis of the mitochondrial COI gene were also performed to further confirm the causative filarial species. Because of the limited availability of tissue and DNA samples from case 1, COI sequence analysis could not be performed for this case. Based on COI sequence analysis, the causative species of case 2 was 99.54% identical to B. pahangi THAI62 (MT027204.1), previously reported in Thailand. This is consistent with data based on the mt 12 S rRNA locus. The causative species of cases 3 and 4 were 99.85% and 100% identical to B. malayi D14 (MK250720.1) and B. malayi (NC004298.1), respectively. As shown in Figure 3, phylogenetic analysis of the COI gene suggested that the causative species of case 2 was in the same clade as B. pahangi, while those of cases 3 and 4 were in the same clade as B. malayi. For cases 5 and 7, the causative species showed 99.85% and 99.83% similarity to the mitochondrial genome of Dirofilaria sp. “hongkongensis” (MN564742.1) and Dirofilaria sp. “hongkongensis” (NC_031365.1), respectively. The COI phylogenetic tree also supported this finding. Notably, the COI sequence of the causative species of case 6 showed 90.91% similarity to P. copsychi 1684 MNHN (OK480041.1) with 92% query coverage.

Figure 3.

Phylogenetic analysis of filarial COI sequences among the causative filarial species reported in this study (in bold). The maximum likelihood tree was constructed using the Tamura-Nei model with gamma distribution.

Molecular characterization and phylogenetic analysis of filarial ITS1.

Causative species identification was further performed using ITS1, a common nuclear marker. The results showed that the causative species of case 1 was 92.72% identical to B. pahangi C46Cat5 (EU419351.1) based on this locus. Causative species of cases 2 and 3 were 93.02% and 98.83% identical to B. pahangi C18Cat7 (EU373645.1) and B. pahangi C14Cat7 (EU373652.1), respectively. The species identification of case 4 was inconclusive, as the ITS1 sequence was only 91.80% identical to B. pahangi C46Cat5 (EU419351.1). The causative species of case 7 was 98.98% and 98.14% identical to Dirofilaria sp. CRS-2021 isolate 20115 (MZ736411.1) and Dirofilaria sp. ‘hongkongensis” HKU2 (JX290195.1), respectively. The ITS1 sequences of cases 5 and 6 could not be amplified using Brugia ITS1 primers, and thus another set of primers specific to Dirofilaria ITS1 was used for PCR amplification and sequencing. The sequences of cases 5 and 6 were 96% and 99% identical to D. repens (AB973229.1). Because of the difference in the target site of the ITS1 genes for species identification in cases 5 and 6, the sequences of these cases were not included in the phylogenetic analysis (Figure 4).

Figure 4.

Phylogenetic analysis of filarial ITS1 sequences among the causative filarial species reported in this study (in bold). The maximum likelihood tree was constructed using the Tamura 3-parameter model.

The GenBank accession numbers and BLASTn alignment results of the mt 12 S rRNA, COI, and ITS1 sequences obtained in all cases are detailed in Table 2. Using the data with significant similarity and conducting phylogenetic analyses of the mt 12 S rRNA, COI, and ITS1 sequences obtained from the cases, we propose the following: B. pahangi as the causative filarial species in cases 1 to 3, B. malayi in case 4, Dirofilaria sp. “hongkongensis” in cases 5 and 7, and a putative novel filarial species genetically related to P. copsychi in case 6.

Table 2.

Nucleotide BLASTn results of filarial and Wolbachia sequences in the present study

Case	Locus	Amplicon Size (bp)	GenBank Accession No.	BLASTn Analysis (closest sequence)	% Query Coverage	% Identity
1	12 S rRNA	109	MT135222.1	Brugia pahangi (AP017680.1)	100	96.33
	ITS1	492	OQ780598.1	Brugia pahangi C46Cat5 (EU419351.1)	100	92.72
2	12 S rRNA	109	MT135223.1	Brugia pahangi 46YT MNHN (KP760318.1)	100	100
	COI	657	OQ713625.1	Brugia pahangi THAI62 (MT027204.1)	100	99.54
	ITS1	502	OQ780599.1	Brugia pahangi C18Cat7 (EU373645.1)	100	93.02
3	12 S rRNA	109	MT887286.1	Brugia malayi L01 (OQ727407.1)	100	100
	COI	657	OQ713626.1	Brugia malayi D14 (MK250720.1)	100	99.85
	ITS1	511	MT732324.1	Brugia pahangi C14Cat7 (EU373652.1)	100	98.83
4	12 S rRNA	110	MT584737.1	Brugia malayi 8YT MNHN (KP760317.1)	100	100
	COI	657	OQ713627.1	Brugia malayi (NC004298.1)	100	100
	ITS1	497	OQ780600.1	Brugia pahangi C46Cat5 (EU419351.1)	100	91.80
5	12 S rRNA	113	MW051483.1	Dirofilaria sp. “hongkongensis” HKU1-HKU9 (KY750550.1)	100	99.12
	COI	657	OQ713628.1	Dirofilaria sp. “hongkongensis” (MN564742.1)	100	99.85
	ITS1 *	608	OQ780601.1	Dirofilaria repens (AB973229.1)	100	96.22
	Wolbachia 16S rRNA	415	OQ780644.1	Wolbachia sp. in Ctenocephalides felis wCfeF (CP116767.1)	99	99.27
6	12 S rRNA	105	OK429322.1	Pelecitus copsychi 268-5 MNHN (OK480977.1)	99	98.08
	COI	657	OQ713629.1	Pelecitus copsychi 1684 MNHN (OK480041.1)	92	90.91
	ITS1 *	577	OQ780602.1	Dirofilaria repens (AB973229.1)	70	98.76
				Mansonella perstans (MN432520.1)	69	98.02
7	12 S rRNA	113	OQ732706.1	Dirofilaria sp. “hongkongensis” HKU1-HKU9 (KY750550.1)	100	100
	COI	669	OQ713630.2	Dirofilaria sp. “hongkongensis” mitochondrion genome (NC_031365.1)	100	99.85
	ITS1	505	OQ780603.1	Dirofilaria sp. CRS-2021 isolate 20115 (MZ736411.1)	96	98.98
				Dirofilaria sp. “hongkongensis” HKU2 (JX290195.1)	74	98.14
	Wolbachia 16S rRNA	428	OQ780645.1	Wolbachia sp. in Dirofilaria repens (MK050782.1), Dirofilaria immitis (AF487892.1) and Onchocerca sp. (JX075217.1)	99–100	98.83

^*The ITS1 sequences of cases 5 and 6 were amplified using Dirofilaria ITS1-specific primers.

Molecular detection and phylogenetic analysis of Wolbachia endosymbionts.

Screening of Wolbachia endosymbionts was performed in all cases by using the nested PCR methods targeting the Wolbachia 16S rRNA, wsp, and ftsZ genes. Only the 16S rRNA gene was amplified in cases 5 and 7, whereas the PCRs for the wsp and ftsZ genes were all negative. The Wolbachia 16S rRNA sequences obtained from cases 5 and 7 were submitted to GenBank under accession numbers OQ780644.1 and OQ780645.1, respectively.

Interestingly, the Wolbachia 16S rRNA sequence from case 5 was found to be 99.27% identical to Wolbachia found in Ctenocephalides felis wCfeF isolate Perak from Malaysia (CP116767.1) with 99% query coverage. The sequence obtained in this case was also phylogenetically clustered with those found in Mansonella spp. in supergroup F. For case 7, the sequence showed 98.83% similarity to the Wolbachia endosymbionts of D. repens (MK050782.1), D. immitis (AF487892.1), and Onchocerca sp. (JX075217.1) and was also phylogenetically clustered with those of Dirofilaria spp. in supergroup C. The phylogenetic analysis of Wolbachia detected in these two cases is demonstrated in Figure 5.

Figure 5.

Phylogenetic analysis of Wolbachia 16S rRNA among the causative filarial species reported in this study (in bold). The maximum likelihood tree was constructed using the Kimura 2-parameter model with gamma distribution.

DISCUSSION

Atypical cases of filariasis in which adults or near-adult worms of filarial species are found incidentally at multiple sites in human tissues have been increasingly reported worldwide.⁶ Several filarial species have been reported to cause unusual presentations of filariasis in humans.^3,4,6 Most of these are mainly of zoonotic origin; however, human filarial nematodes that infect only humans can also cause these phenomena.^4,6 Most atypical cases of filariasis presented with a movable subcutaneous nodule or mass with a history of migratory lesions consistent with the cases reported in this study.⁴ For zoonotic filariae infecting humans, the worms are usually present in tissues that are not normal habitats in animals.^4,6

Wuchereria bancrofti and B. malayi are two main lymphatic filarial species that are endemic to the western and southern parts of Thailand, respectively.²⁴ Wuchereria bancrofti infects humans exclusively, whereas B. malayi has animal reservoirs with potential zoonotic transmission.^1–3 In addition to lymphatic filariae, zoonotic filariasis caused by B. pahangi, D. repens, and D. immitis has been reported in dogs and cats throughout Thailand along with a few human case reports.^{17,19,25–28}

Some recent publications have identified a new species of Dirofilaria, namely Dirofilaria sp. “hongkongensis,” in dogs, jackals, and humans on the Indian continent and Hong Kong; however, the existence of this new species is still controversial, as there is no suitable morphological description for this species and it is not currently registered with the ZooBank.^10,29,30

Pelecitus spp., avian filariae endemic to Africa and South America, have been sporadically reported in zoonotic cases in South America.³¹ The main transmission vectors of Pelecitus spp. are mosquitoes, chewing lice, and tabanids.³¹ Human infections with Pelecitus spp. are rare and typically present as intraocular filariasis.^32,33 Recently, the discovery of a new species, Pelecitus copsychi, in Copsychus malabaricus (white-rumped shama bird) in Pahang, Peninsular Malaysia, has raised the possibility of human infection with this species in the region.³⁴

In this study, we report seven cases with atypical presentations of filariasis in which young adult female worms were recovered from multiple sites in human organs or tissues. Most patients were middle-aged women whose filarial nematodes were incidentally found in various tissues, including subcutaneous tissues, breast mass, the periorbital area, and intraocularly. Peripheral eosinophilia was observed in two but not all of the patients. Atypical cases of filariasis presented in this study were caused by filarial species of the genus Brugia, Dirofilaria, and a suspected novel species genetically related to P. copsychi. None of the patients reported here were initially diagnosed with filariasis. The diagnosis was established incidentally after the tissues or worms were surgically removed and examined. Diagnosis of the causative species based on the morphology of the recovered tissue is a major challenge, especially when only partial portions of the worm are obtained or when the worms have degenerated. Most of the inflammatory cells found in histological tissues are neutrophils. In some cases, necrotic tissue and granulomatous reactions were found in the tissue surrounding the worm.

The morphologies of filarial worms belonging to the genera Wuchereria and Brugia are generally similar, except for the larger size of Wuchereria worms.^4,6,30 Both genera have a thin and smooth cuticle that is more prominent over the lateral chords, with three or four low, broad muscle cells per quadrant.^4,6,30 Differentiation of Brugia species based on only their morphology is limited, because all species share similar typical features, including a thin smooth cuticle and low, broad muscle cells.^4,6,30 For Dirofilaria, D. repens and D. immitis are zoonotic filarial species that occasionally infect humans.^4,30 Another species, Dirofilaria tenuis, for which raccoons are natural hosts, is another common species that can infect humans, but this species is restricted to North America.⁴ Dirofilaria repens and D. immitis have similar morphological features, including prominent muscle cells and internal lateral ridges, with the exception that D. immitis has a smooth cuticle, whereas D. repens has external longitudinal ridges.^6,8

The morphological and histological features of filaria observed in cases 1 to 4 closely resembled those of Brugia species. On the other hand, the causative species of cases 5 and 7 were compatible with Dirofilaria species. In case 6, morphological identification of the causative species is limited because of the degeneration of the parasitic tissue. Therefore, molecular characterization of conserved filarial species genes should be performed to confirm the causative species. Differentiating between B. malayi and B. pahangi based on only a single gene, such as the mt 12 S rRNA, COI, or ITS1 gene, is challenging because of their close sequence similarity.³⁵ Therefore, a comprehensive analysis of multiple genes is crucial for accurate species identification.

For cases 1 and 2, both histological data and molecular analysis of the causative species, based on mitochondrial and nuclear markers, are consistent. This evidence suggests that the filarial species infecting these cases is B. pahangi. However, inconsistencies between these markers and those in the other cases were observed in this study, as seen in cases 3 and 4, where results from mitochondrial genes (mt 12 S rRNA and COI) and a nuclear gene (ITS1) gave different results. It should be noted that the main challenge with barcode-based identification is the possibility of obtaining confusing results when a sample contains more than one filarial species, since the hypothesis being tested assumes only a single infecting species. It is also noteworthy that almost all the ITS1 references were obtained from infected animals in areas of filariasis endemicity in Thailand, where coinfections with multiple filarial species are common.³⁶ Therefore, it is likely that there may be some inconsistencies between identified reference species and associated DNA sequences in the database, complicating the BLASTn analysis. To rule out coinfection, a large number of recombinant clones need to be sequenced. Alternatively, a metabarcoding approach is highly appreciated for differentiating a population of amplicons from the samples with multiple infecting species. In addition, a high degree of polymorphic variation in the ITS1 locus of Brugia species has previously been observed across geographic locations, suggesting that the final alignment scores calculated by BLASTn and species determination may also be affected.^36,37

Despite the relatively short length of the 12 S rRNA amplicons, the primers specific to this region were originally designed for high-resolution melting analysis, targeting the sequence region with significant nucleotide polymorphism across filarial species.²⁶ In this study, the utility of two mitochondrial markers, i.e., 12 S rRNA and COI, provided consistent results for filarial species identification. Accordingly, we assert that the filarial species recovered from cases 3 and 4 are most likely B. malayi, supported by the 100% and 99.85% similarity of the mt 12 S rRNA and COI genes with B. malayi references.

The discrepancy in sequence identification of Dirofilaria species was also observed in case 5 of this study. Two mitochondrial markers in this case showed almost perfect similarity to Dirofilaria sp. “hongkongensis,” while the nuclear ITS1 was only 96.22% identical to that of the European isolate of D. repens (AB973229.1). Our phylogenetic result also supports a previous report that showed that the ITS1 sequence of D. repens shared the same cluster with Dirofilaria sp. “hongkongensis” whereas some ITS1 sequences of D. repens grouped with those of D. immitis isolates.³⁸ Similar to Brugia as mentioned above, these results support the genetic heterogeneity of the ITS1 sequences among Dirofilaria species.³⁸ Similar to case 5, Dirofilaria sp. “hongkongensis” was identified as the causative species in case 7, based on the sequence similarity of two mitochondrial markers, 12 S rRNA and COI. The ITS1 sequence similarity further supports the diagnosis of Dirofilaria sp. “hongkongensis,” although the query coverage of ITS1 with the reference sequence (JX290195.1) is only 74%. This may be due to the limited availability of ITS1 data for this species in the GenBank database. Only the complete mitochondrial genome of Dirofilaria sp. “hongkongensis” is available; the complete nuclear genome dataset remains inaccessible. Hence, whole-genome sequencing of Brugia and Dirofilaria species is needed to provide a more comprehensive understanding of genetic variation. This approach may prove instrumental in species identification and improvement of the information base for filarial species, especially for newly proposed species such as Dirofilaria sp. “hongkongensis.”

As such, it appears that the mitochondrial markers would be more helpful than the ITS1 marker for both Brugia and Dirofilaria species identification, enabling us to confirm the definite diagnosis of the cases in this study. According to our phylogenetic results, the COI sequences of the congeneric species were grouped in the same cluster but showed a clear divergence between the conspecific groups, strongly supporting that the mitochondrial COI gene has a significant discrimination power for the identification of multiple filarial species.^39,40 Also, COI-PCR is typically more sensitive than ITS1-PCR, because of the higher number of copies of mitochondrial DNA than nuclear DNA. Therefore, the strength of this research is to demonstrate the utility of mitochondrial barcoding, particularly of COI, as a highly sensitive, reliable diagnostic tool for the identification of filarial species in clinical specimens.

Surprisingly, the causative species responsible for intraocular filariasis in case 6 shows sequence similarity to P. copsychi, based on two mitochondrial markers, 12 S rRNA and COI. However, no ITS1 sequence of P. copsychi is currently available in the GenBank database. This new species was recently identified in the white-rumped shama bird in Malaysia.³⁴ Interestingly, the percent identity of this case to the P. copsychi COI reference is only 90.91%, which remains insufficient for species reaffirmation, whereas the ITS1 sequence has ∼98% identity over ∼70% of query coverage to that of D. repens (AB973229.1) or Mansonella perstans (MN432520.1). These similarity patterns suggest either a coinfection with multiple nematode species or a potentially novel species. From the clinical history of this case, only a single worm was removed from the anterior chamber of the left eye, supporting the latter speculation of a suspected novel species. Unfortunately, the microanatomical structure of the worm in this case was unclear. Therefore, we can only propose that the worm most likely belongs to the suspected novel filarial species genetically related to P. copsychi.

However, it remains challenging not only to assess coinfections but also to characterize emerging and/or novel filarial species using conventional PCR and Sanger sequencing on a single mitochondrial COI gene. Accordingly, the utility of high-throughput sequencing such as Oxford Nanopore technology can overcome these issues. This long-read metabarcoding approach is a promising diagnostic platform that can generate complete or long mitochondrial sequences with high taxonomic resolution, allowing us to rapidly and accurately identify and genotype multiple or emerging filarial species from various clinical samples in a single day.⁴¹

In this study, we also identified the presence of Wolbachia bacterial endosymbionts in filarial specimens recovered from the patients. It is widely known that these obligatory mutualists have essential roles in the embryonic development, reproduction, and long-term survival of filarial worms.⁴² In this study, Wolbachia was molecularly detected in only two cases, i.e., cases 5 and 7. In the other cases, we were unable to detect Wolbachia, which could be attributed to the limitations of sensitivity of our detection methods or the means by which the specimens were preserved. Interestingly, Wolbachia endosymbionts were molecularly detected in both cases in which Dirofilaria sp. “hongkongensis” infection was diagnosed. Moreover, Wolbachia detected in our cases were phylogenetically characterized into supergroups C and F. Wolbachia supergroups C, D, and J were previously known to be exclusively found in a mutualistic relationship with filarial nematodes.⁴³ Of these, Wolbachia supergroups C and D have recently been detected in blood samples from Thai dogs that tested positive for Dirofilaria sp. “hongkongensis” infection.⁴⁴ Wolbachia supergroup F is the only supergroup that has been found to atypically infect phylogenetically distant host groups, both arthropods and filarial nematodes.^45,46 Members of this unique supergroup have previously been detected in filarial parasites that cause cavity filariasis in humans, including Mansonella spp.^18,47,48 Thus, this study has revealed, for the first time, the occurrence of Wolbachia supergroup F in Dirofilaria sp. “hongkongensis,” demonstrating the significant genetic diversity of the bacterial endosymbionts in this novel filarial species.

In contrast to their arthropod counterparts, it was observed that all filarial Wolbachia in supergroups C, D, J, and F underwent convergent pseudogenization or gene loss of the bacterioferritin, a regulator important for heme homeostasis.⁴⁵ These findings indicate an obligatory relationship in which the absence of bacterioferritin in all filarial Wolbachia would render synthesized heme a supplement for filarial hosts lacking a heme biosynthetic pathway.⁴⁵ By studying the presence and role of Wolbachia in filarial infections, we can gain valuable insights into disease mechanisms, transmission dynamics, and potential avenues for targeted interventions.

Of concern, the number of cases of zoonotic filariasis has been continuously increasing, highlighting the need for accurate and prompt diagnosis, appropriate treatment, and effective vector control measures to minimize infection and transmission of this mosquito-borne disease. In Thailand, several Mansonia spp. have been reported to be the principal natural vectors of B. malayi,^49,50 while B. pahangi is known to be transmitted by Armigeres subalbatus.^51,52 Additionally, several mosquito species of the genera Aedes, Armigeres, Culex, and Mansonia are known to be involved in Dirofilaria transmission.^2,53 However, the vector species of the suspected zoonotic filarial nematode Pelecitus sp. in Thailand remains unknown. Essentially, conducting more intensive studies of the geographical distribution of competent vectors and animal reservoirs in areas affected by atypical filariasis is also required for a better understanding and effective management and control of this neglected tropical disease.

Note: Supplemental materials appear at www.ajtmh.org.