Longistatin, a plasminogen activator, is key to the availability of blood-meals for ixodid ticks

Abstract

Ixodid ticks are notorious blood-sucking ectoparasites and are completely dependent on blood-meals from hosts. In addition to the direct severe effects on health and productivity, ixodid ticks transmit various deadly diseases to humans and animals. Unlike rapidly feeding vessel-feeder hematophagous insects, the hard ticks feed on hosts for a long time (5-10 days or more), making a large blood pool beneath the skin. Tick's salivary glands produce a vast array of bio-molecules that modulate their complex and persistent feeding processes. However, the specific molecule that functions in the development and maintenance of a blood pool is yet to be identified. Recently, we have reported on longistatin, a 17.8-kDa protein with two functional EF-hand Ca(++)-binding domains, from the salivary glands of the disease vector, Haemaphysalis longicornis, that has been shown to be linked to blood-feeding processes. Here, we show that longistatin plays vital roles in the formation of a blood pool and in the acquisition of blood-meals. Data clearly revealed that post-transcriptional silencing of the longistatin-specific gene disrupted ticks' unique ability to create a blood pool, and they consequently failed to feed and replete on blood-meals from hosts. Longistatin completely hydrolyzed α, β and γ chains of fibrinogen and delayed fibrin clot formation. Longistatin was able to bind with fibrin meshwork, and activated fibrin clot-bound plasminogen into its active form plasmin, as comparable to that of tissue-type plasminogen activator (t-PA), and induced lysis of fibrin clot and platelet-rich thrombi. Plasminogen activation potentiality of longistatin was increased up to 4 times by soluble fibrin. Taken together, our results suggest that longistatin may exert potent functions both as a plasminogen activator and as an anticoagulant in the complex scenario of blood pool formation; the latter is critical to the feeding success and survival of ixodid ticks.

Figure 1. Post-transcriptional silencing of longistatin-specific gene in adult ticks by injecting dsRNA.

(A) Semiquantitative RT-PCR analysis. One microgram of longistatin dsRNA was injected into the hemolymph of ticks of the RNAi-treated group. Ticks of the control group were treated with 1 µg of malE dsRNA. Actin was used as an internal control. Eng, engorged. (B) Quantitative RT-PCR using total RNA and primers specific for longistatin as in A. Eng, engorged. (C) In situ detection of longistatin expression in ticks' salivary glands. Salivary glands from the ticks of control and RNAi-treated groups. Endogenous longistatin was reacted with mouse anti-longistatin sera (1∶100). (D) Effect of gene silencing on longistatin post-translation by Western blot analysis. Salivary gland extracts were electrophoresed and transferred onto nitrocellulose membrane. Endogenous longistatin was probed with mouse anti-longistatin (1: 100). Eng, engorged.

Figure 2. Effects of post-transcriptional silencing of longistatin gene on blood pool formation and blood feeding.

(A) Impact of longistatin-specific mRNA silencing on blood-meal feeding from hosts. Adult ticks were injected with longistatin dsRNA or malE dsRNA (1 µg/tick) and were allowed to feed on a tick-naïve rabbit. RNAi-treated ticks failed to replete. (B) Postengorgement weight was significantly (p<0.01) lower in the RNAi-treated group than that of the control group. Dotted lines indicate mean±SD of body weight of ticks. (C) Effects of post-transcriptional silencing of longistatin gene on blood pool formation. RNAi was performed and ticks were fed in the same manner as in A. RNAi-treated ticks failed to establish a prominent blood pool. Arrows indicate site of tick attachment. (D) Blood pools were significantly (p<0.01) smaller in the RNAi-treated group. (E) Histopathological changes were studied using EVGS. Longistatin was detected in the feeding lesions of ticks on the host's tissues using mouse anti-longistatin sera (1∶100).

Figure 3. Anti-coagulation and fibrinogenolytic activity of longistatin.

(A) Effects of longistatin on the formation of fibrin clot. Fibrinogen (7.5 mM) was pre-incubated in a buffer in the absence or presence of longistatin (0.1, 0.2, 0.4, 0.8 and 1.6 µM) or plasmin (1.6 µM) and then thrombin was added (0.10 NIH unit/µl) as described in Materials and Methods. Clot formation was detected visually and also by determining changes in turbidity at OD₄₅₀ using a spectrophotometer after 15 min. (B) Longistatin (1.6 µM) delayed fibrin clot formation up to 90 min. Fibrinogen was incubated in the absence or presence of longistatin (1.6 µM) following the same procedures as mentioned in A and then thrombin was added. OD₄₅₀ was measured at 15 min intervals. (C) Fibrinogenolytic effect of longistatin. Fibrinogen (7.5 mM) was incubated in the absence or presence of longistatin (0.4, 0.8 and 1.6 µM) or plasmin (1.6 µM). Samples were collected at the indicated time period and were subjected to 12.5% SDS–PAGE analysis under reducing conditions. A gradual degradation of the α, β and γ chains of fibrinogen was detectable with the concomitant deposition of degraded products. Asterisks (^*) indicate that the difference compared with the negative control group (buffer only) is significant as determined by Student's t-test with unequal variance (^*p<0.05, ^**p<0.01).

Figure 4. Plasminogen activation by longistatin.

(A) Longistatin (40, 80, 160, 320 and 640 nM) was incubated without or with plasminogen (0.24 units) adding fibrin CNBr fragments (4 µg) in a total volume of 200 µl of buffer (50 mM Tris–HCl, pH 7.5; 100 mM NaCl and 5 mM CaCl₂) at 25°C for 2 h. Then, plasmin-specific fluorogenic substrate (100 µM, final concentration) was added and substrate hydrolysis was monitored by measuring excitation and emission wavelengths of 360 nm and 460 nm, respectively, at 15 min intervals. Inset, initial rate of plasminogen activation at different concentrations of longistatin. (B) Effects of fibrin CNBr fragments on the activation of plasminogen by longistatin. Plasminogen (0.24 units) was incubated with longistatin (640 nM) in the absence or presence of fibrin CNBr fragments (0.25, 1 and 4 µg) as described in Materials and Methods. All assays were performed in triplicate.

Figure 5. Longistatin induced fibrinolysis by activating plasminogen.

(A) Fibrin clot was formed by incubating fibrinogen (7.5 mM) and thrombin (0.10 NIH unit/µl) and was incubated in the presence of plasminogen-t-PA/-longistatin (40, 80, 160, 320 and 640 nM) mixture or buffer only at 25°C for 24 h. Clot lysis was measured at OD₄₅₀. Plasminogen induced complete lysis of fibrin clot in the presence of 640 nM longistatin or 154 nM t-PA. (B) Time-dependent activation of plasminogen by longistatin with concomitant lysis of fibrin clot. (C) Cleavage of plasminogen into the heavy and light chains. Digested product of fibrin clot was electrophoresed by 12.5% SDS–PAGE. Asterisks (^*) indicate that the difference compared with the negative control group (buffer only) is significant as determined by Student's t-test with unequal variance (^*p<0.05, ^**p<0.01, ^***p<0.001).

Figure 6. Binding of longistatin with fibrin clot.

(A) Detection of longistatin bound on fibrin meshwork. Fibrinogen at different concentrations (3.75, 7.5, 15, 30 and 60 mM; final concentration) was mixed in the absence or presence of longistatin (10 µg) or an equal amount of t-PA or u-PA in a buffer (50 mM Tris–HCl, pH 7.5; 100 mM NaCl and 5 mM CaCl₂) and thrombin (0.10 NIH unit/µl) was added immediately and was incubated at 25°C for 1 h. The clot was treated with anti-longistatin (1∶100), anti-t-PA (1∶100), anti-u-PA (1∶20) or pre-immune sera (1∶100). Bound antibodies were detected using green fluorescent-labeled secondary antibody (Alexa Flour 488 goat anti-mouse IgG). (B) Supernatant was analyzed by 12.5% SDS–PAGE under reducing conditions. (C) The target protein was extracted from the supernatant and its concentration was determined using micro-BCA reagent as described in Materials and Methods. The results are expressed as percentage of longistatin/t-PA/u-PA bound to the fibrin clot. Data represent mean ± SD, n = 3.

Figure 7. Lysis of platelet-rich thrombi by longistatin in dog plasma.

(A) Longistatin hydrolyzes platelet-rich thrombi ex vivo in dog plasma. Platelet-rich clot was produced by incubating 0.2 ml of dog blood and the thrombi were treated with longistatin at various concentrations in 0.5 ml of dog plasma at 37°C for 12 h and weighed at the indicated period. (B) Comparison of thrombolytic activity of longistatin with that of t-PA. Thrombi were treated with longistatin (640 nM)/t-PA (154 nM) under the same ex vivo experimental conditions.

Figure 8. A schematic diagram showing roles of longistatin in blood coagulation and fibrinolysis events.

In the initial phase, the tick bites and lacerates tissues at the site of attachment and damages vascular beds, which results in hemorrhage leading to the development of a blood pool. Longistatin is synthesized in and secreted from the salivary glands and injected into the blood pool during feeding process . Longistatin degrades fibrinogen and activates plasminogen to its active form, plasmin. HMWK, high-molecular-weight kininogen; PKK, prekallikrein; TF, tissue factor. Yellow arrows, contact activation (intrinsic) pathway; olive-green arrows, tissue factor (extrinsic) pathway; green arrows, common pathway of coagulation cascade and white arrows, fibrinolytic pathway. Figure adapted from ref. , , .