Semi-artificial mouse skin membrane feeding technique for adult tick, Haemaphysalis longicornis

Naturally engorged ticks

The parthenogenetic Okayama strain of the tick H. longicornis has been maintained by feeding on rabbits in our laboratory since 1976 [7]. In this study, NEr ticks were prepared according to the usual method described previously [8]. Briefly, 20 adult ticks were placed on the ears of a rabbit to feed. At the beginning of the engorgement period, 9 ticks, which spontaneously detached from the rabbit after 5 days were collected. Of those, 6 randomly selected ticks were subjected to phenotypic analysis, and the remaining 3 ticks were used for transcription analysis. NEm ticks were also prepared according to the method described previously [9]. Briefly, 5 mice (BALB/c, 3 weeks old) were infested with 10 adult ticks (2 ticks per mouse). Six randomly selected NEm ticks were subjected to phenotypic analysis, and the remaining 4 ticks were used for transcription analysis. Ethical approvals of conducting all animal experiments were provided by the Animal Care and Use Committee, National Institute of Animal Health (NIAH, Approval nos. 441 and 578).

Artificially engorged ticks

To prepare the mouse skin membrane, 5 to 7 adult ticks were allowed to feed on the shaven back of each of four tick-naïve SPF mice (BALB/c, 3 weeks old) following the described method [9]. After 4 to 5 days (at the beginning of the expansion period [10]), a rectangular section (~9 cm²) of the mouse skin with the ticks attached was carefully removed from the mouse’s body. Figure 2A shows the Hematoxylin and Eosin (HE)-stained lesion of mouse skin membrane used for the feeding system. Hypodermal layers around the feeding lesion were carefully removed from the skin with sterile tweezers as much as possible. The body of the feeding system (Figure 2B) was constructed using the upper part of a Falcon tube (#352059, Becton, Dickinson and Company, Franklin Lakes, NJ) by cutting the tube at ~3 cm from the top. Then, the roof of the cap of the Falcon tube was removed and the skin membrane with ticks was placed on the mouth of the tube keeping the ticks outside and held tightly in place by applying the cap. Only two ticks were selected and allowed to feed within the area of the skin membrane in this system, and the ticks in excess of two were removed by tweezers and weighed. The mean body weight of the removed ticks (24.9 ± 5.2 mg) was subtracted from the weight of ticks after semi-artificial feeding to estimate weight gain. After construction, the inside of the membrane was washed with sterilized phosphate-buffered saline (PBS) supplemented with 100 units/ml penicillin and 100 μg/ml streptomycin (Life Technologies Corporation, Carlsbad, CA). Then, pre-warmed (30°C) rabbit blood containing 300 μl washed red blood cells (RBC) and 700 μl sterile serum (filtered with a syringe filter; #4652, 0.2 μm, Acrodisc Syringe Filters, Pall Co., Cornwall, UK.) was poured into the device (Figure 2C). To secure a sufficient volume of blood for the duration of tick feeding, we collected blood from a tick-naïve female SPF Japanese white rabbit (3- to 5-months-old). The open end of the tube was covered with a piece of parafilm. All procedures including system construction and blood exchange were performed inside a biosafety cabinet. The system was kept in a humidified chamber with >95% relative humidity at 30°C, and the rabbit blood was changed at every 12 h. When partially fed ticks of the expansion period (4–5 days post-infestation, DPI) were used, fully engorged ticks dropped off within 12 to 48 h of the onset of artificial feeding (Figure 2D-F). In contrast, ticks in the late-growth phase (3 DPI) required feeding on blood for more than 48 h to become engorged (data not shown). These feeding patterns were quite similar to the on-host feeding patterns of ticks described previously [10], suggesting that our in vitro feeding device effectively supports the expansion process of ticks and provides a sufficient amount of blood.

Phenotypic analysis

Blood feeding, reproductive parameters, and hatching rate of eggs were quantified to investigate the influences of artificial feeding on tick physiological phenotype. After repletion, the body weight gain of AE ticks was approximately 10 times that of their initial body weight. There were no significant differences in engorged body weight between AE and NEr or NEm ticks measured after spontaneous detachment of ticks following full engorgement (Figure 3A). After collection, the engorged ticks were placed separately in small sterile vials and incubated in a humidified chamber with >95% relative humidity at 25°C for egg production. Almost all of the ticks started laying eggs at 4 to 5 days post-engorgement, however, ticks were monitored until 10 days post-oviposition (DPO) at which time point each egg mass obtained from individual ticks was collected separately and weighed. To determine hatching rates, the collected egg mass was placed separately in a paper envelope and incubated under the same conditions described above for 40 days until the eggs hatched. After hatching, larvae were counted manually and hatchability was estimated as described previously [11]. There were no significant differences between AE and NEr or NEm in terms of egg mass weight (Figure 3B), egg conversion ratio (total egg weight/engorged body weight, Figure 3C), or hatching rate (Figure 3D). In addition, larvae derived from the AE, NEr, and NEm lineages of ticks were able to feed normally on rabbit (data not shown).

Transcription analysis

We evaluated the effects of artificial feeding on the transcription of eight selected genes related to proteolysis of tick blood digestion in the midgut [12] such as H. longicornis serine proteinase (HlSP) [GenBank:AB127388] [13], Longepsin [GenBank:AB218595] [14], Longipain [GenBank:AB255051] [15], H. longicornis serine carboxypeptidase 1 (HlSCP1) [GenBank:AB287330] [16], H. longicornis legumain (HlLgm) [GenBank:AB279705] [17], HlLgm2 [GenBank:AB353127] [18], H. longicornis leucine aminopeptidase (HlLAP) [GenBank:AB251945] [19], and HlLAP2 (Hatta and Tsuji, unpublished data). Additionally, we checked one defensin gene related to tick innate immunity in the midgut [20], namely, H. longicornis midgut defensin (Hlgut-defensin) [GenBank: EF432731] to judge the microbial contamination or infection in AE ticks since the transcription of this gene is drastically induced by lipopolysaccharide (LPS) [21]. To do this, the midguts of NEr, AE, and NEm ticks were collected in sterile PBS at 3 days after repletion. Immediately after collection, the midguts were submersed into 3 ml of Buffer RLT (RNeasy Mini Kit, Qiagen, Valencia, CA, USA) supplemented with 30 μl of 2-mercaptoethanol, and thoroughly homogenized by passing the tissue mass through a 20-gauge needle (NN-2038R, Terumo, Tokyo, Japan) fitted to a 5 ml syringe (SS-05LZ, Terumo) five times. Total RNA was isolated from the homogenates according to the manufacturer’s instructions and used to synthesize cDNA (RNA PCR Kit Ver 3.0, Takara Bio INC., Shiga, Japan). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) using SYBR Green I dye (LightCycler FastStart DNA Master SYBR Green I, Roche Diagnostics, Mannheim, Germany) and the primer sets listed in Table 1 was conducted to measure mRNA expression levels, as described previously [22]. Expression of each of the genes was normalized to that of β-actin [GenBank:AY254898] based on copy number (Figure 4). As expected, blood digestion-related gene expression patterns were quite similar between AE and NEr ticks, suggesting that the semi-artificial feeding technique does not affect midgut function parameters, such as blood uptake, expansion of midgut size, and blood-digestion. However, a slight difference in expression of two genes (HlSP and HlLgm2) was observed in ticks of the feeding groups AE and NEm or of NEr and NEm, which may be due to the feeding of blood from different hosts, corroborating the findings of others [23].

Even though a biosafety cabinet was used throughout the study during construction and manipulation of the feeding device, the possibility of microbial contamination in the blood cannot be completely neglected since we did not use antibiotics in the supplied blood. Nevertheless, similar expression patterns of the Hlgut-defensin gene among AE, NEr, and NEm ticks indicate that the feeding system was quite free from bacterial contamination. However, to increase the certainty of avoiding bacterial contamination, it would be better to apply antibiotics to this feeding system for other applications.