Prevalence and patterns of ectoparasites infesting Pallas’s squirrels (Callosciurus erythraeus) in Kanagawa Prefecture, Japan

Abstract

The Pallas’s squirrel (Callosciurus erythraeus) has invaded fragmented woodlands in urban areas of Kanagawa Prefecture, where frequent human contact occurs. We examined 538 squirrels in Hayama-machi, Kanagawa Prefecture, for ectoparasite infestations. A total of 1,164 lice, 877 fleas, and 231 ticks were retrieved from 297 (55.2%), 338 (62.8%), and 135 (25.1%) squirrels, respectively. The identified ectoparasite species were Neohaematopinus callosciuri, Ceratophyllus anisus, Haemaphysalis flava, Haemaphysalis hystricis, and Ixodes turdus. The prevalence of N. callosciuri and C. anisus was significantly higher in adult males (N. callosciuri 63.6%, C. anisus 70.4%) than that in adult females (N. callosciuri 36.1%, C. anisus 49.0%; P<0.0001). Such information on the dynamics of host-ectoparasite relations is crucial for evaluating the risk to public health.

Pallas’s squirrel (Callosciurus erythraeus), also known as the red-bellied squirrel, was introduced to Japan from Taiwan in the 1930s [38]. Since then, this species has established itself in multiple locations in the country, causing damage to forest plantations, agricultural crops, and housing [33]. The Japanese Ministry of the Environment has designated this species as a “regulated organism” under the Invasive Alien Species Act (https://www.env.go.jp/en/nature/as.html), and several municipal authorities, including Kanagawa Prefecture, have introduced control measures to eradicate or reduce their population.

Ectoparasites, including Haemaphysalis flava (Acari: Ixodidae), Neohaematopinus callosciuri (Anoplura: Haematopinidae), Ceratophyllus anisus (Siphonaptera: Ceratophyllidae), and Dermanyssus mites (Acari: Dermanyssidae), have been identified in Pallas’s squirrels in Kanagawa Prefecture, Japan [19, 29]. Katahira et al. [10] recovered Ceratophyllus indages indages and Leptotrombidium mites (Acari: Trombiculidae) from squirrels in the Yokosuka district of Kanagawa Prefecture and noted a higher abundance of ectoparasites than previously reported. Of these ectoparasites, H. flava infests humans [18, 41, 42], harbors pathogens that cause Japanese spotted fever and severe fever with thrombocytopenia syndrome [18, 28]. In Lithuania, Red squirrels (Sciurus vulgaris) and their flea Ceratophyllus sciurorum, a congener of C. anisus, are reservoirs and vectors, respectively, of the zoonotic bacterium Bartonella washoensis [16].

Since Pallas’s squirrels in Kanagawa Prefecture inhabits fragmented woodlands located between residential and industrial areas, where close contact with humans and domestic animals is frequent [8, 33], the associated public health risks warrant attention. In this study, we investigated the occurrence of ectoparasites in Pallas’s squirrels on the Miura Peninsula, Kanagawa Prefecture, where the squirrel population is concentrated and is sighted regularly [8]. We explored the role of sex and maturity of the host and seasonal factors in ectoparasite distribution in Pallas’s squirrel. Such information on the dynamics of host-ectoparasite relations is crucial for evaluating public health risks and contributes to the establishment of effective control strategies where necessary.

A total of 538 Pallas’s squirrels were captured using single-capture live traps, set overnight in three locations within a 3 km radius in Hayama-machi on the Miura Peninsula, Kanagawa Prefecture. The trapping procedure was approved by Hayama-machi and was carried out by EGO Co., Ltd. (Chigasaki, Japan). Trapping was conducted from April to June 2019; January, February, and May 2020; February to June 2021; and January to June 2022. Traps were baited with peanuts and sesame oils and operated for three to five consecutive days monthly. Captured squirrels were euthanized at the trapping site by sevoflurane inhalation and kept individually in a plastic bag at −30°C until further use. Maturity was determined based on scrotal and nipple pigmentation in males and females, respectively. Of the 538 squirrels caught, 321 were mature males, 155 were mature females, and 62 were juveniles. The mean body weights of the adult males, adult females, and juveniles were 327.12 g (S.D. 29.7), 336.6 g (S.D. 36.2), and 279.0 g (S.D. 38.4), respectively.

The carcasses were placed on a tray, and their entire bodies were brushed using a comb while visually inspecting for ectoparasites. This procedure was performed by two people, with the first person combing for exactly 2 min and the second person combing for 3 min, for a total of 5 min for each squirrel. Plastic bags containing the carcasses were also visually examined. Ectoparasites, including Ixodidae ticks, fleas, and lice that fell on the tray were retrieved using forceps immediately after detection and stored in 70% ethanol. All ectoparasites were counted and identified to the species level, where possible, under direct microscopic observation using the CellSens^® Standard software (Olympus, Tokyo, Japan). Ectoparasites were cleared with Gater’s solution where necessary. Species identification was conducted according to previously published morphological descriptions of Acari species by Sasa [27], Yamaguti et al. [40], and Takada et al. [31]; Anoplura species by Johnson [6] and Kim [11]; and Siphonaptera species by Sakaguchi [25] and Sakaguti and Jameson [26]. The developmental stages of ticks and lice were also recorded. Tick specimens were identified by analyzing the DNA barcode of the 16S rDNA region when morphological identification was not possible. We followed the protocol of Takano et al. [32] for DNA extraction and PCR was conducted using the primers mt-rrs1: 5ʹ-CTGCTCAATGATTTTTTAAATTGCTGTGG-3ʹ and mt-rrs2: 5ʹ-CCGGTCTGAACTCAGATCAAGTA-3ʹ [39]. PCR products were purified using the NucleoSpin Gel and PCR Clean-up kit (Macherey-Nagel, Düren, Germany) according to the manufacturer’s protocol and directly sequenced on both strands using the same primers used for the PCR in a sequencing facility (FASMAC Co., Ltd., Atsugi, Japan). Sequences were compared to homologous sequences available in GenBank (National Center for Biotechnology Information) using the Nucleotide Basic Local Alignment Search Tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

The prevalence, mean intensity, and Poulin’s discrepancy index, with the associated 95% confidence intervals (CI), for each ectoparasites according to maturity (adult vs. juvenile) and sex (adult female vs. adult male) were determined using Quantitative Parasitology (QPweb, version 1.0.15) [21, 23]. Prevalence and mean intensity were compared using an unconditional exact test [22] and a bootstrap two-sample t-test [4, 23], respectively. Poulin’s discrepancy indices were compared based on the CIs [23]. The average monthly temperature and precipitation for the Miura Peninsula were obtained from the Japan Meteorological Agency (https://www.data.jma.go.jp/obd/stats/etrn/index.php, in Japanese). The Japan Meteorological Agency defines January and February as winter, March to May as spring, and June as summer. Statistical differences with P value <0.05 were considered significant.

The overall prevalence of squirrels infested with at least one ectoparasite species was 87.2% (469/538), whereas 12.8% (69/538) of the squirrels were negative for ticks, fleas, or lice. Infestation with two or more ectoparasites was observed in 45.7% (246/538) of squirrels, with a combination of lice/flea species being the most common (23.0%, 124/538), followed by lice/fleas/ticks (10.0%, 54/538), fleas/ticks (7.6%, 41/538), and lice/ticks (5.2%, 28/538). The remaining squirrels were infested solely with lice, fleas, or ticks, with prevalence rates of 16.9% (91/538), 22.1% (119/538), and 2.2% (12/538), respectively.

A total of 1,164 lice, 877 fleas, and 231 ticks were retrieved from 297 (55.2%), 338 (62.8%), and 135 (25.1%) squirrels, respectively. Of the 1,164 lice, 1,160 were identified as N. callosciuri (392 adult males, 439 adult females, 329 nymphs), and of the 877 fleas, 870 were identified as C. anisus (458 adult males, 412 adult females). Four lice and seven fleas were not identified because the retrieved specimens were damaged. Of the 231 ticks, 210 were identified as H. flava (3 adult males, 3 adult females, 198 nymphs, and 6 larvae), 16 as Haemaphysalis hystricis nymphs, and one as adult Ixodes turdus. Additionally, four nymphs with damaged capitula were identified as H. flava (3) and H. hystricis (1) using molecular methods. The monthly numbers of ectoparasites retrieved from the 538 squirrels and their developmental stages are presented in Table 1.

Table 1. Monthly numbers of Neohaematopinus callosciuri, Ceratophyllus anisus, Haemaphysalis flava, and Haemaphysalis hystricis retrieved from 538 Pallas’s squirrels and their developmental stages.

Year	Month	Average temp. (°C)*	Average precipitation (mm)*	N. callosciuri					C. anisus				H. flava					H. hystricis
				No. of squirrels infested	Adult male	Adult female	Nymph	Total	No. of squirrels infested	Adult male	Adult female	Total	No. of squirrels infested	Adult	Nymph	Larva	Total	No. of squirrels infested	Nymph
2019	April	13.6	135.5	8/26	8	6	1	15	24/26	27	32	59	3/26	0	4	0	4	0/26	0
	May	19.2	154.5	5/12	9	1	1	11	7/12	20	6	26	1/12	0	1	0	1	0/12	0
	June	21.4	217.0	7/14	9	14	7	30	13/14	12	19	31	4/14	2	2	0	4	0/14	0
2020	Jan	8.0	113.0	29/47	53	55	34	142	16/47	21	24	45	15/47	0	28	4	32	0/47	0
	Feb	9.3	36.5	17/19	26	28	33	87	13/19	15	37	52	12/19	0	43	1	44	0/19	0
	May	19.1	87.0	34/65	29	45	23	97	42/65	72	54	126	17/65	0	29	1	30	2/65	2
2021	Feb	9.5	69.0	8/17	11	7	6	24	10/17	14	12	26	5/17	0	13	0	13	0/17	0
	Mar	13.0	128.0	13/37	10	20	10	40	26/37	38	40	78	6/37	1	7	0	8	0/37	0
	April	14.9	127.5	18/43	28	14	9	51	39/43	60	43	103	18/43	0	25	0	25	3/43	3
	May	18.9	122.5	17/29	29	31	15	75	19/29	18	16	34	2/29	0	2	0	2	1/29	1
	June	22.1	82.5	5/11	5	5	0	10	4/11	3	4	7	1/11	0	1	0	1	0/11	0
2022	Jan	5.8	16.5	43/51	73	84	108	265	16/51	13	12	25	12/51	0	12	0	12	1/51	1
	Feb	6.1	73.0	24/46	25	33	34	92	21/46	34	21	55	11/46	0	16	0	16	0/46	0
	Mar	11.4	76.0	3/7	1	4	1	6	6/7	1	11	12	4/7	0	5	0	5	1/7	1
	April	15.0	235.5	19/38	12	24	11	47	34/38	51	38	89	9/38	1	8	0	9	7/38	8
	May	18.6	226.5	15/38	19	17	11	47	30/38	42	31	73	7/38	1	5	0	6	1/38	1
	June	22.0	137.0	24/38	45	51	25	121	17/38	17	12	29	1/38	1	0	0	1	0/38	0
Overall				289/538	392	439	329	1,160	337/538	458	412	870	128/538	6	201	6	213	16/538	17

*The average temperature and precipitation for the Miura Peninsula were extracted from the Japan Meterological Agency.

The prevalence, mean infestation intensities, and Poulin’s discrepancy index with the associated 95% CI of the identified ectoparasite species are shown in Table 2. The highest overall prevalence was observed in C. anisus (62.6%, 95% CI: 58.4–66.7), whereas the highest mean intensity was observed in N. callosciuri (3.91, 95% CI: 3.48–4.51). No significant differences were observed in the prevalence, mean intensity, or Poulin’s discrepancy index between adults and juveniles for any of the examined ectoparasites (P>0.05). The prevalence of N. callosciuri and C. anisus was significantly higher in adult males (N. callosciuri 63.6%; C. anisus 70.4%) than that in adult females (N. callosciuri 36.1%; C. anisus 49.0%; P<0.0001; Table 2). The Poulin’s discrepancy index for N. callosciuri was significantly lower in adult males (0.661, 95% CI: 0.624–0.706) than that in adult females (0.811, 95% CI: 0.761–0.850; P<0.05; Table 2). Temporal prevalence in adult males and females is shown in Table 3. The prevalence of N. callosciuri was significantly higher in males than that in females in Spring and Summer: June 2019 (P=0.0391), March–June 2020 (P=0.0491, 0.0410, 0.0001, and 0.0068, respectively), and April (P=0.0025), and June 2022 (P=0.0091). The prevalence of C. anisus was significantly higher in males than that in females in March 2020 (P=0.0024), June 2020 (P=0.0264), and February 2022 (P=0.0215). No significant differences were observed in the prevalence of H. flava during any of the months examined. Infestation with H. hystricis first appeared in May 2020 and was thereafter detected in April and May 2021, January 2022, and March–May 2022.

Table 2. Prevalence, mean intensity, and Poulin’s discrepancy index with 95% confidence interval for Neohaematopinus callosciuri, Ceratophyllus anisus, Haemaphysalis flava, and Haemaphysalis hystricis.

Ectoparasite	Host		N	Prevalence % (95% CI)	Mean intensity (95% CI)	Poulin’s discrepancy index (95% CI)
Neohaematopinus callosciuri	Adult	Female	155	36.1 (28.6–44.2)**	3.38 (2.58–4.48)	0.811 (0.761–0.850)*
		Male	321	63.6 (58.0–68.8)**	3.95 (3.44–4.65)	0.661 (0.624–0.706)*
		Total	476	54.6 (50.0–59.2)	3.82 (3.37–4.43)	0.713 (0.682–0.746)
	Juvenile		62	59.7 (46.4–71.9)	4.49 (3.30–7.11)	0.695 (0.618–0.787)
	Total		538	55.2 (50.9–59.5)	3.91 (3.48–4.51)	0.713 (0.683–0.744)
Ceratophyllus anisus	Adult	Female	155	49.0 (40.9–57.2)**	2.22 (1.91–2.62)	0.676 (0.623–0.732)
		Male	321	70.4 (65.1–75.3)**	2.68 (2.41–3.00)	0.567 (0.527–0.608)
		Total	476	63.4 (58.9–67.8)	2.56 (2.35–2.81)	0.606 (0.575–0.642)
	Juvenile		62	56.5 (43.3–69.0)	2.74 (2.14–3.57)	0.634 (0.551–0.729)
	Total		538	62.6 (58.4–66.7)	2.58 (2.38–2.81)	0.611 (0.583–0.643)
Haemaphysalis flava	Adult	Female	155	23.2 (16.8–30.7)	1.81 (1.39–2.83)	0.845 (0.792–0.897)
		Male	321	22.7 (18.3–27.7)	1.51 (1.33–1.73)	0.826 (0.789–0.862)
		Total	476	22.9 (19.2–26.9)	1.61 (1.41–1.92)	0.835 (0.803–0.868)
	Juvenile		62	24.2 (14.2–36.7)	2.53 (1.67–3.60)	0.835 (0.762–0.898)
	Total		538	23.0 (19.6–26.8)	1.72 (1.52–2.05)	0.841 (0.814–0.870)
Haemaphysalis hystricis	Adult	Female	155	4.5 (1.8–9.1)	1.14 (1.00–1.43)	0.954 (0.904–0.973)
		Male	321	2.8 (1.3–5.3)	1.00	0.969 (0.941–0.981)
		Total	476	3.4 (1.9–5.4)	1.06 (1.00–1.19)	0.966 (0.943–0.978)
	Juvenile		62	0.0 (0.0–5.8)	-	-
	Total		538	3.0 (1.7–4.8)	1.06 (1.00–1.19)	0.970 (0.953–0.981)

Adult female vs. adult male: *P<0.05, **P<0.0001. CI: confidence interval.

Table 3. Temporal prevalence in adult male and female hosts for Neohaematopinus callosciuri, Ceratophyllus anisus, Haemaphysalis flava, and Haemaphysalis hystricis.

Year	Month	N. callosciuri			C. anisus			H. flava			H. hystricis
		Male	Female	P value	Male	Female	P value	Male	Female	P value	Male	Female	P value
2019	April	40.0% (6/15)	28.6% (2/7)	NS	93.3% (14/15)	85.7% (6/7)	NS	20.0% (3/15)	0% (0/7)	NS	0% (0/15)	0% (0/7)	-
	May	55.6% (5/9)	0% (0/2)	NS	66.7% (6/9)	0% (0/2)	NS	11.1% (1/9)	0% (0/2)	NS	0% (0/9)	0% (0/2)	-
	June	75.0% (6/8)	0% (0/3)	0.0391	100% (8/8)	66.7% (2/3)	NS	25.0% (2/8)	0% (0/3)	NS	0% (0/8)	0% (0/3)	-
2020	Jan	81.5% (22/27)	76.9% (10/13)	NS	33.3% (9/27)	30.8% (4/13)	NS	33.3% (9/27)	38.5% (5/13)	NS	0% (0/27)	0% (0/13)	-
	Feb	83.3% (5/6)	100% (7/7)	NS	83.3% (5/6)	85.7% (6/7)	NS	33.3% (2/6)	71.4% (5/7)	NS	0% (0/6)	0% (0/7)	-
	May	58.1% (25/43)	27.3% (3/11)	NS	69.8% (30/43)	45.5% (5/11)	NS	30.2% (13/43)	18.2% (2/11)	NS	2.3% (1/43)	9.1% (1/11)	NS
2021	Feb	37.5% (3/8)	42.9% (3/7)	NS	50.0% (4/8)	57.1% (4/7)	NS	37.5% (3/8)	14.3% (1/7)	NS	0% (0/8)	0% (0/7)	-
	Mar	52.9% (9/17)	21.1% (4/19)	0.0491	94.1% (16/17)	47.4% (9/19)	0.0024	11.8% (2/17)	21.1% (4/19)	NS	0% (0/17)	0% (0/19)	-
	April	51.6% (16/31)	16.7% (2/12)	0.041	90.3% (28/31)	91.7% (11/12)	NS	38.7% (12/31)	50.0% (6/12)	NS	3.2% (1/31)	16.7% (2/12)	NS
	May	80.0% (16/20)	0% (0/7)	0.0001	75.0% (15/20)	42.9% (3/7)	NS	10.0% (2/20)	0% (0/7)	NS	0% (0/27)	7.1% (1/14)	NS
	June	83.3% (5/6)	0% (0/5)	0.0068	66.7% (4/6)	0% (0/5)	0.0264	16.7% (1/6)	0% (0/5)	NS	0% (0/17)	0% (0/19)	-
2022	Jan	81.5% (22/27)	85.7% (12/14)	NS	37.0 (10/27)	21.4% (3/14)	NS	22.2% (6/27)	28.6% (4/14)	NS	5.0% (1/20)	0% (0/7)	NS
	Feb	60.0% (15/25)	30.8% (4/13)	NS	64.0% (16/25)	23.1% (3/13)	0.0215	20.0% (5/25)	38.5% (5/13)	NS	0% (0/25)	0% (0/13)	-
	Mar	66.7% (2/3)	25.0% (1/4)	NS	100% (3/3)	75.0% (3/4)	NS	33.3% (1/3)	50.0% (2/4)	NS	0% (0/3)	25.0% (1/4)	NS
	April	65.4% (17/26)	10.0% (1/10)	0.0025	88.5% (23/26)	90.0% (9/10)	NS	23.1% (6/26)	10.0% (1/10)	NS	19.2% (5/26)	20.0% (2/10)	NS
	May	42.3% (11/26)	28.6% (2/7)	NS	84.6% (22/26)	57.1% (4/7)	NS	15.4% (4/26)	14.3% (1/7)	NS	3.8% (1/26)	0% (0/7)	NS
	June	79.2% (19/24)	35.7% (5/14)	0.0091	54.2% (13/24)	28.6% (4/14)	NS	4.2% (1/24)	0% (0/14)	NS	0% (0/24)	0% (0/14)	-
Overall		63.6% (204/321)	36.1% (56/155)	<0.0001	70.4% (226/321)	49.0% (76/155)	<0.0001	22.7% (73/321)	23.2% (36/155)	NS	2.8% (9/321)	4.5% (7/155)	NS

NS: not significant.

A survey conducted two decades ago in Kanagawa Prefecture identified the same species, H. flava, N. callosciuri, and C. anisus [29], suggesting that the life cycles of these ectoparasites are well established among the squirrel populations in the prefecture. The present study showed male-biased infestation of N. callosciuri and C. anisus. Sex-bias in parasitism has been commonly reported in small rodents in lice [5, 17, 30] and fleas [13, 14, 20]. One reason for this difference is that male hosts have a larger home range and higher mobility, thereby increasing their chances of encountering infective ectoparasites in the environment or other infested hosts [15]. Indeed, the mean home range for male Pallas’s squirrels in Japan is 1.25–3.83 ha and overlaps extensively with that of other male and female squirrels, whereas the female home range is only 0.48–0.72 ha with little to no overlapping with that of other squirrels [35, 37]. It is also reported that elevated androgen levels in male hosts exert immunosuppressive effects [12, 24]. Reproduction is observed throughout the year in Pallas’s squirrels in Japan [36, 37], and it is possible that male squirrels maintain higher testosterone levels irresspective of the season, resulting in a higher ectoparasite prevalence. These elevated testosterone levels, together with ranging behaviors for achieving reporoductive success in male hosts, are likely to result in high infestation intensities in both N. callosciuri and C. anisus. Although females mate with multiple males during estrus [36], which increases their chances of encountering infested male hosts, N. callosciuri and C. anisus seem to exploit immunosuppressed males rather than females.

In Kanagawa Prefecture, host-parasite relationships between raccoons (Procyon lotor) and H. flava have already been well established, and a high intensity of adult and nymphal H. flava was observed during winter and spring [1, 3]. It is reported that H. flava overwinters on its host [7], and our results suggest that nymphal H. flava utilizes squirrels in addition to larger mammals (e.g., raccoons) for overwintering. However, adult ticks and larvae were rare in squirrels, indicating that the host preference for H. flava might change depending on their life stage. H. hystricis is a tick species often reported in western Japan [31]. However, a tick survey conducted in the Kanto region from 2015 to 2020 have reported the expansion of this tick species to Kanagawa Prefecture [2], and raccoons from the Miura Peninsula were found to be infested with this tick [3]. It is not clear when H. hystricis was introduced to these areas; however, recent increases in wild boar (Sus scrofa leucomystax) populations, which are one of the major hosts of H. hystricis, may have played a role in the spread of the tick population. Wild boars in Kanagawa Prefecture have been subjected to population control, and the number of animals caught on the Miura Peninsula has increased significantly from 8 in 2016 to 39 in 2017, and then to 67 in 2021 [9]. This coincided with the period in which we began observing H. hystricis in Pallas’s squirrels. Since then, we have consistently collected this tick species from Pallas’s squirrels during the warm seasons, suggesting that Pallas’s squirrels are becoming important hosts for nymphal H. hystricis. One specimen of I. turdus was identified in this study. The adult stage of this tick is generally found in bird species [31]; hence, the infestation is suggested to be incidental.

In conclusion, this study showed that host-parasite relationships are well established between Pallas’s squirrels and the ectoparasites H. flava, N. callosciuri, and C. anisus in the surveyed area. Male-biased parasitism was observed in N. callosciuri and C. anisus, indicating that male hosts are responsible for the transmission of these ectoparasites. This study also suggests that H. hystricis utilizes Pallas’s squirrels to maintain its population. In Kanagawa Prefecture, efforts have been made to suppress the expansion and invasion of Pallas’s squirrels into natural forests inhabited by native squirrel species [34]. Effective measures, including distribution surveys and trapping, are being conducted by local citzens in fragmented woodlands located between residential areas [8, 34], where the chances of encountering squirrels infested with ectoparasites are high. Hence, a continuing survey is crucial to assess public health risks for those involved in such eradication efforts.

POTENTIAL CONFLICTS OF INTEREST

The authors declare no conflicts of interest associated with this publication.