Phylogenetic, molecular, and microscopic investigation of Linguatula serrata infection in stray and road-killed dogs in Northwest Iran

Abstract

Background

Linguatula serrata is a cosmopolitan zoonotic parasite. Canids serve as definitive hosts, while herbivores acts as intermediate hosts. Adult L. serrata are found in the nasal and respiratory passages, whereas the immature stages are located in the mesenteric lymph nodes, liver, spleen, lungs, and, occasionally in other organs. Humans can serve as intermediate hosts, with visceral infections, or as definitive hosts with nasopharyngeal infections. This study aimed to investigate the infection rate of stray and road-killed dogs with L. serrata and to explore its phylogeny.

Material and methods

A total of 150 stray dogs and 100 road-killed dogs were examined for L. serrata eggs in feces and nasal secretions using microscopy and molecular tests, and for adult L. serrata through necropsy.

Results

The results showed that 174 of 250 (69.6%) studied dogs were positive for L. serrata by at least one of the examination methods. The prevalence in road-killed and stray dogs was determined to be 72% and 68%, respectively. The L. serrata infection rate based on fecal microscopy in the road-killed and stray dogs was 29% and 46.7%, respectively, and by fecal PCR was 59% and 48.7%, respectively. The rate of infection with L. serrata based on nasal secretions in road-killed and stray dogs was 14% and 26%, respectively, using microscopy, and 44% and 43.3%, respectively, using PCR. In the road-killed dogs, based on necropsy, 46% were positive. The results of this study show that PCR is more effective than microscopy for detecting L. serrata infections in dogs. PCR amplicon of the expected size for Linguatula of approximately 595 bp for 18S rRNA were generated from the L. serrata isolates using described specific primers.

Conclusion

It can be concluded that the rate of infection in dogs and possibly other carnivores, herbivores, and man, is high in this locality; hence, strict control measures should be conducted to overcome the risk of infection with this zoonotic disease.

Introduction

Belonging to the class Pentastomida, Linguatula serrata Fröhlich, 1789 is a zoonotic aberrant arthropod [1] that primarily infects dogs, and occasionally, foxes, cats, and other carnivores as the final hosts and herbivores such as sheep, goats, cattle, and camels as the intermediate hosts. The adult parasite can be detected in the upper respiratory system, nasal airways, and frontal sinuses of the final hosts. Nasopharyngeal discharges from the final hosts spread the eggs into the environment. Larvae emerge from the eggs, after being ingested by herbivores, and spread throughout the internal organs of the intermediate hosts reaching the mesenteric lymph nodes (MLNs), liver, lungs and, spleen, where they develop into infective nymphs [2, 3]. The final hosts are infected following ingestion of the viscera of intermediate hosts that contain infective nymphs [2]. Humans can serve as final or intermediate hosts. Consumption of raw or undercooked infected internal organs of intermediate hosts can result in nasopharyngeal disease with symptoms ranging from mild inflammation of the upper respiratory tract to temporary conductive deafness [4,5,6]. Consumption of L. serrata eggs (typically via water or vegetables) results in the visceral form in humans. Symptoms depend on the organs involved and can include abdominal pain, chronic cough, or night sweats [7]. In severe infections, secondary septicemia, pneumonia, or severe enterocolitis can occur [8, 9].

Recently, cases of linguatulosis in humans have been reported in various provinces of Iran, with an increasing incidence observed [9,10,11,12,13,14,15]. There are also reports indicating the endemicity of linguatulosis in goats, sheep, cattle, and camels in various regions of Iran and other countries [2, 16, 17]. Based on the above reports and given that dogs are the major source of linguatulosis of herbivorous animals and man, it is speculated that the prevalence of linguatulosis may be very high in stray dogs from different regions in Iran.

In Urmia, Iran, the prevalence has been reported as 1.9% in road-killed dogs [18] and 81.0% in stray dogs [19], with both studies using microscopic examination of the nasal cavity and brain. In other studies within Iran, mostly using necropsy, prevalence has ranged from 27.8% to 76.5% in stray dogs, showing a wide variation depending on the location [20,21,22,23]. In other parts of the world where L. serrata occurs, reported infection rates range from as low as 0.4% to over 67% [24,25,26]. Wild and stray dogs often show higher infection rates, with 67.6% of dogs in south-eastern Australia reported to be positive [27]. However, only 3.5% of stray dogs in the Rabat region, Morocco were positive for infection [28]. The infection rate in owned dogs in Thessaloniki, Greece was 0.4% [29] but was over 37.4% in Jalingo, Nigeria [30].

Given regional variations and the lack of information on the prevalence of linguatulosis in stray dogs in the southern areas of Iran, determining the frequency of infestation of this disease in stray dogs from a public health perspective is of great significance. However, it is a time-consuming, expensive, and hazardous task for research personnel due to the risk of direct contact with infected animals or their tissues, which can lead to zoonotic transmission of L. serrata. Adequate information on the prevalence of linguatulosis of stray dogs in this populated area is not yet available. Therefore, the present study was undertaken to examine the prevalence of this disease in stray dogs and road-killed dogs of the Tabriz area of the East Azerbaijan province in Iran using microscopy and PCR (feces and nasal secretions) and necropsy.

Materials and methods

Study area

With a population of over 1.5 million and geographic coordinates of 38° 4′ N and 46° 18′ E, Tabriz County, East Azerbaijan Province, in northwest Iran, was the location of the study. This city has a humid continental climate with distinct seasons, and it is situated between the Eynali and Sahand mountains. About 320 mm of rainfall is recived annually. During the warmer summer months, the climate is temperate; while the long, harsh winters bring temperatures as low as −10 °C.

Study animals

This study was conducted between March 2016 and March 2021 in Tabriz, East Azerbaijan Province, and involved 150 stray dogs and 100 road-killed dogs. All procedures were conducted in compliancewith ethical guidelines and were approved by the institutional animal care and use committee (approval number: 2312). Fecal samples from stray dogs were collected non-invasively from naturally defecated material, ensuring no harm to the animals. Road-killed dogs were obtained through local authorities or standard municipal road-cleaning operations, ensuring that no animals were euthanized specifically for research purposes. Necropsy procedures were carried out with strict adherence to ethical and biosafety standards. Using the following formula [31], the number of samples needed was determined, considering a 50–70% prevalence of L. serrata (derived from research in the region's herbivores), a 95% confidence interval, and a 5% accuracy.

$${N}_{\kappa }=\frac{4\left(1-\kappa \right)}{{W}_{\kappa }^{2}}\left[\left(1-\kappa \right)\left(<span class='convertEndash'><span class='convertEndash'>1-2</span></span>\kappa \right)+\frac{\kappa \left(2-\kappa \right)}{2\pi \left(1-\pi \right)}\right]{z}_{1-\alpha /2}^{2}$$

where K is the expected kappa coefficient of 0.7, α is the error rate, d is the accuracy, π is the predicted prevalence, W_k is the kappa width, and α is the 5% accuracy,

The Bureau of Veterinary Medicine and the Tabriz Environment Protection Organization granted authorization for the samples to be collected (Letter Number 2312). The Pardis Animal Shelter provided the stray dogs, who were housed there both during and after the study. As part of the trial, none of the dogs had anesthesia or were put to sleep.

As previous studies suggest [32], dogs were grouped into four age groups based on their teeth condition: under six months, six to less than 24 months, two to four years, and more than four years.

Detection of eggs by microscopy

For fecal and nasal egg detection, all dogs (250) were used. Fecal samples from road-killed dogs were collected directly from the rectum, while naturally defecated samples were used for stray dogs. Feces were examined using direct smear and flotation, using a saturated sugar solution, described by [33], with some modifications. Initially, 2 g of feces were mixed with 10 ml water, sieved (a 100 µm mesh), and centrifuged at 1800 g for 5 min. After pouring off the supernatant, the sediment was mixed with the saturated sugar solution and centrifuged with a coverslip on the tube at 1000 g for 5 min. The coverslip was subsequently examined under magnifications of 40 × and 100 × to detect L. serrata eggs. A portion of the sediment was stored at −20 °C for DNA extraction and PCR analysis.

To evaluate the presence of L. serrata eggs in nasal secretions, 2 ml of PBS (phosphate-buffered saline, pH = 7) was injected into the sinuses, and then 1–2 drops of the secretions were placed on a slide, covered with lamel and examined at a 40 × magnification. Additionally, a sample of the secretions was stored at −20 °C for DNA extraction. Morphological identification of the eggs was performed following the criteria outlined by Soulsby (1982).

Detection of L. serrata adult by necropsy

The heads of the 100 road-killed dogs were removed from the bodies and transferred to the necropsy laboratory. After collection of the nasal secretions, the skin was separated from the underlying bones, and the skull was sagittally cleaved on either side of the midline using a motorized thin-bladed saw to access both the frontal sinuses and nasal cavities. The nasopharynx area, nasal turbinates, sinuses, eustachian tubes, trachea, and brain were visually examined for the presence of L. serrata (Fig. 1) with any parasites seen carefully removed. Each section of the skull was submerged in warm physiological saline overnight. The next day, the head was discarded, the supernatant fluid decanted, and the sediment examined for parasites. The number of L. serrata in each infected dog was recorded. The collected adult parasites were preserved in 10% buffered formalin for species identification or stored at −20 °C for DNA extraction and PCR analysis. For morphological identification, the parasites were washed under running water, mounted on slides, cleared with lactophenol, and stained using azocarmine. The length and width of the parasites were measured, and the morphology and sex were studied using a light microscope.

Fig. 1

Adult form of L. serrata (female) harvested from infected dog by necroscopy (A and B); Eggs of L. serrata harvested from infected dog by microscopic method (C)

DNA extraction from feces, nasal secretions, and adult parasites

DNA was extracted from feces containing egg deposits, nasal secretions, and adult parasites using a DNA isolation kit (MBST, Iran) according to the manufacturer's protocol. In brief, individual parasites, 50 ml of feces containing egg deposits, or nasal secretions were lysed in 180 μl lysis buffer and 20 μl proteinase K (10 mg/ml) for 20 min to 2 h at 55 °C. Subsequently, 360 μl binding buffer was added, and the mixture was incubated for 10 min at 70 °C. Next, 270 μl ethanol (96–100%) was added to the solution, vortexed, and transferred to the MBST column. The column was centrifuged at 8000 × g, washed twice with 500 μl washing buffer, and then centrifuged at 12,000 × g. DNA was eluted from the column using 100 μl elution buffer. The extracted DNA was visualized on 0.8% agarose gel stained with ethidium bromide and observed under a UV transilluminator. The DNA concentration was quantified by measuring absorbance at 260 nm (A260). DNA was either used immediately in PCR or stored at –20 ºC for future use.

Polymerase chain reaction (PCR), sequencing and phylogenetic analyses

To investigate Linguatula serrata, specific primers targeting the 18S small subunit ribosomal RNA (18S rRNA) gene were employed. The primers, 18S-F (5’-CCATGGTTGTCACGGGTGACG-3’) and 18S-R (5’-CTTGCGACGATCCAAGAATTT-3’), were designed based on complete 18S rRNA sequences available in GenBank (accession no. JX088397).

PCR amplification was performed in 25 µL reaction volumes using an I-Cycler thermal cycler (Bio-Rad). The reaction mixture consisted of: 1 µL of genomic DNA (5 ng), 25 µM of each primer, 1.5 mM MgCl², 1250 µM of each dNTP, 1 × GoTaq® Green Flexi Reaction Buffer, and 1 U of GoTaq DNA polymerase (Promega, USA). The thermal cycling protocol included an initial denaturation at 94 °C for 1 min, followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 10 s. A final extension step at 72 °C for 1 min completed the protocol.

The amplified PCR products were resolved by 2% agarose gel electrophoresis, stained with DNA-safe stain (Yekta Tajhiz Azma, Iran; Cat no. YT0001), and visualized using a Gel Documentation System (Axygen®, Germany). PCR amplicons were purified with the QIAquick® PCR Purification Kit (QIAGEN) according to the manufacturer’s instructions. The purified PCR products were submitted to Bioneer Co. (Korea) for sequencing. Sequence analysis and phylogenetic studies were conducted using MEGA software (version 11.0.13). The neighbor-joining method was employed to construct phylogenetic trees, with bootstrap analysis performed using 1,000 replicates to ensure robustness.

L. serrata sequences were aligned with reference strains retrieved from GenBank to confirm identity and infer evolutionary relationships. The final sequence data were submitted to the NCBI GenBank database under the accession number OR125935.1.

Calculation of the agreement of two tests

The following formula was used to calculate the agreement between the two tests [31, 34, 35]. If the Kappa value was less than 0.2, 0.2–0.4, 0.4–0.6, 0.61–0.8, and more than 0.8, the agreement between the two tests is considered weak, moderate, or almost good, good and very good, respectively [36, 37].

$${p}_{Exp}=\frac{\left({n}_{00}+{n}_{01}\right)\left({n}_{00}+{n}_{10}\right)}{{n}^{2}}+\frac{\left({n}_{11}+{n}_{10}\right)\left({n}_{11}+{n}_{01}\right)}{{n}^{2}};{p}_{Agree}=\frac{{n}_{00}+{n}_{11}}{n}$$

$$\kappa =\frac{{p}_{Agree}-{p}_{Exp}}{1-{p}_{Exp}}$$

where P _Exp: Chance agreement; P _Exp: True agreement, n₀₀: The number of animals that were not infected in both tests; n₁₁: Number of animals infected in both tests; n₁₀ and n₀₁: The number of animals that were infected by one test and uninfected by another test and K: agreement of two tests or Kappa.

Statistical analysis

Associations between host factors (age and sex) and prevalence of infection were evaluated using the chi-square test with SPSS software version 21. Differences were considered significant at a P-value < 0.05.

Results

Prevalence of L. serrata in road-killed and stray dogs

Of the 250 dogs studied, 174 (69.6%) tested positive for L. serrata based on at least one of the detection method. The prevalence rates in road-killed and stray dogs were 72.0% and 68.0%, respectively (Table 1). In road-killed dogs, the highest infection rate was observed with fecal PCR (59.0%), while the lowest was detected using nasal secretion microscopy (14.0%). Similarly, in the stray dogs, the highest infection rate was found with fecal PCR (48.7%), and the lowest with nasal secretion microscopy (26.0%). Infection rates in both road-killed and stray dogs, as determined by all diagnostic methods except for nasal secretion microscopy in road-killed dogs showed a statistically significant correlation with age and sex, with infection rates increasing with age and being higher in males than in females.

Table 1 Rate of infection with L. serrata in road-killed and stray dogs based on age and sex using different diagnostic methods

Parasite distribution and necropsy findings

The number of parasites recovered from each road-killed dog ranged from 1 to 4, with an average of 2 L. serrata per infected dog. In 43 of the 150 road-killed dogs, the parasite was found on the left side of the skull, compared to 20 on the right side and 19 with parasites on both sides of the skull. No parasites were detected in the eustachian tubes, maxillary sinuses, and brain cavities.

Results of diagnostic method comparison

Good agreement was observed between the PCR of nasal secretions and necropsy with a kappa value of 0.6. In contrast, the kappa value in the two tests microscopic examination of nasal secretions versus necropsy and microscopic examination of feces versus necropsy were −0.09. and −0.16, respectively (Table 2).

Table 2 The results of Kappa in two tests

Findings of molecular identification and phylogenetic analysis

PCR amplicon of the expected size for Linguatula was approximately 595 bp for 18S rRNA gene. One positive sample was confirmed by sequencing. Phylogenetic analysis based on the 18S rRNA nucleotide sequence revealed no significant difference between Iranian isolates (Fig. 2). Tabriz-HG (accession number: OR125935.1) exhibited the highest nucleotide sequence similarity to IR-18 s-Cattle-Kerman (accession number: KU234183.1) and IR-4-Sheep-Tabriz (accession number: KF830130.1) from Iran (Tables 3 and 4).

Fig. 2

Phylogenetic relationship of L. serrata isolates based on partial sequence of 18S rRNA gene. The numbers at nodes represent bootstrap values based on 500 replicates

Table 3 Data for some Linguatula serrata isolates used in this study

Table 4 Percent identity of partial nucleotide sequences of the 18S rRNA gene between 22 L. serrata isolates

Discussion

The parasitic disease linguatulosis, which is brought on by L. serrata, affects both animal and human populations' public health. Although most cases of the infection are subclinical, reports of severe rhinitis, bloody nasal discharge, and sporadic coughing and sneezing have been made [33]. Depending on whether the animal species is the definitive or intermediate host, different diagnostic techniques are used to identify L. serrata infections. The majority of studies that determine the infection rate in definitive hosts, dogs, use necropsy or identify eggs in feces or nasal swabs [21]. Our results indicate that PCR may be a more successful technique for identifying infection in living dogs, especially when applied to nasal secretions. By using microscopy for direct egg screening, the risk of exposure to humans may be underestimated due to the underestimation of dog prevalence.

According to the data that is being given today, stray dogs have a relatively high infection rate. The close contact of dogs and domestic animals is the main cause of the high occurrence of linguatulosis in dogs. The main source of infection for domestic dogs is the offal of ruminant domestics; in addition, eating raw sheep, goats, and other herbivorous animal viscera can cause a high rate of pentastomosis in stray dogs. The parasite spreads more easily to humans and other animals because there are a lot of stray dogs that live as scavengers and have access to offal from slaughterhouses or the carcasses of infected herbivores. The practice of feeding dogs the raw viscera of home-slaught animals and the high frequency of domestic or occupational interaction between dogs and livestock provide ideal conditions for the continuation of the parasite life cycle [18, 22]. Additionally, the care of domestic animals may encourage the pollution of feed and water meant for animals with L. serrata eggs by dog feces.

Our research revealed significant variations in infection rates among dogs of varying ages and sexes. The infection rate was modest in puppies under six months old, but it rose as they aged. The highest occurrence rate was reported in canines older than four years. These findings contradict those of another study, which found no significant variation in the rate of infection among stray dogs of different sexes and ages in Shiraz, Iran [22]. The absence of linguatulosis in newborn puppies has been linked to their long maturation time (> 6 months) [38] and feeding on mother's milk, which eliminates the need to give them infected ruminant offal [23].

Female dogs exhibited a lower infection rate in the study presented here than male dogs, which was similar to the study conducted by [23]. However, there was no significant difference between the sexes among the age groups. As a result, the difference could be attributed to the inclusion of more male canines in the study.

In the present study, we molecularly evaluated samples to confirm the presence of L. serrata, and a phylogenetic tree was constructed to investigate the relationship between isolates of Iranian origin. Our isolate, Tabriz-HG (accession number: OR125935.1), was found to be 99.86% to 99.87% identical to other isolates from Iran, including those from Yazd (KJ009333.1), IR-4-Sheep-Tabriz (KF830130.1), and clone-1 (MG913249.1).

Additionally, the Tabriz-HG isolate exhibited high nucleotide similarity (99.87%) not only to Iranian isolates but also to isolates from other countries, such as Australian isolates (MN889436.1 and MT196141.1) and Romanian isolates (MZ677005.1 and MZ314332.1) [39, 40]. These isolates have been derived from different animal species, such as cattle, sheep, and other ruminants, further supporting the hypothesis of genetic continuity and limited genetic diversity of L. serrata across different host species. This suggests that L. serrata may exhibit relatively stable genetic characteristics regardless of the host species or geographic location.

In our study, the genetic differences observed between isolates within Iran and those from other countries are minimal, which could be attributed to the high degree of host specificity of the parasite and its relatively low mutation rate over geographic distances. The minimal genetic variation in these isolates may also reflect shared environmental and ecological factors that influence the parasite's transmission dynamics across different regions.

These results align with previous studies on L. serrata from other countries, which have found similar high nucleotide identities across isolates from various animal species. For instance, studies on L. serrata isolates from ruminants in Europe, Asia, and Australia have shown identical or near-identical nucleotide sequences for certain genes, suggesting that the parasite’s genetic makeup remains largely conserved, regardless of host species or geographic location [41, 42].

Based on the analysis of these sequences, the genetic differences among L. serrata isolates, both within Iran and internationally, appear to be relatively minor, indicating a degree of genetic stability and a lack of significant genetic differentiation among isolates across different regions and host species.

Conclusion

The high prevalence of L. serrata found in dogs in this study indicates that the human population of Tabriz is at a potentially high risk of L. serrata exposure. Although linguatulosis is difficult to diagnose in dogs, with PCR being the preferred method based on the study results presented herein, it should not be ignored. A strict control program conducted by the city authorities and education of the population on the epidemiology of the parasite could reduce the infection risk. Any measures taken in this direction may also help to decrease other zoonoses in which dogs play a role. Regular molecular monitoring of parasitic isolates circulating within a region should be done to continuously have accurate and updated information on the status of pathogens. Examining more samples collected from different hosts and regions will help to better understand the possible distribution of isolates and their evolution.