HeLp, a Heme Lipoprotein from the Hemolymph of the Cattle Tick,Boophilus microplus *

The main protein of the hemolymph of the cattle tick Boophilus microplus has been isolated and shown to be a heme lipoprotein (HeLp). HeLp has an apparent molecular mass of 354,000 and contains two apoproteins (103 and 92 kDa) found in equal amounts. HeLp presents a pI of 5.8 and a density of 1.28 g/ml and contains 33% lipids, containing both neutral lipids and phospholipids, and 3% of sugars. A remarkable feature of HeLp is the abundance of cholesterol ester (35% of total lipids), a lipid not previously reported in invertebrate lipoproteins. Western blot analysis showed HeLp in hemolymph from adult females and males, but not in eggs. Although HeLp contains 2 heme molecules, it is capable of binding 6 additional molecules of heme. Boophilus feeds large amount of blood, and we recently showed that this tick is unable to performde novo synthesis of heme (Braz, G. R. C., Coelho, H. S. L., Masuda, H., and Oliveira, P. L. (1999)Curr. Biol. 9, 703–706). Injection of tick females with55Fe-labeled heme-HeLp indicated that this protein transports heme from hemolymph to tissues. HeLp is suggested to be an essential adaptation to the loss of the heme synthesis pathway.

Heme is an ubiquitous molecule that takes part in several fundamental biochemical reactions such as respiration, oxygen transport, photosynthesis, and lipid desaturation (1). In contrast to all these beneficial features, heme is a powerful catalyst of the formation of reactive oxygen species, mainly by means of Fenton type reaction (2,3). In fact, there are several reports of heme-induced oxidative damage to a broad range of biomolecules as lipoproteins, which are especially susceptible to lipid peroxidation driven by heme (4 -6). Due to its potential toxicity, heme is always found associated with proteins, such as hemopexin and albumin, and several antioxidant mechanisms have been developed to protect cells (7)(8)(9)(10)(11). As described in the literature, heme transport and recycling does not occur in eukaryotic cells where the heme delivered to cells is degraded by heme oxygenase (1,12).
Hematophagous arthropods usually ingest large amounts of vertebrate blood, comprising several times their own weight (13). The blood-sucking insect Rhodnius prolixus has a hemebinding protein in the hemolymph that protects the insect from heme induced lipid peroxidation (14,15). In the case of cattle tick, Boophilus microplus, blood is the sole food source for all developmental stages and, during the last 48 h before dropping from the host, adult females may ingest as much as 100 times their weight in blood (16). We recently showed that Boophilus is an exception to the generally accepted statement that all eukaryotic cells synthesize their own heme (17). This tick does not have a functional heme synthesis pathway and thus must rely exclusively on the heme obtained from digestion of the vertebrate blood in order to make its own heme-proteins (18). The heme biosynthetic pathway is also defective in some pathogenic bacteria that depend on host blood as a heme source, and proteins involved in heme transport have an essential role in the recovery of heme (19). The dependence of the tick on heme from its diet requires the development of mechanisms for heme absorption, transport, and recycling, which have not been described for any other multicellular organisms. In this article we describe the isolation and characterization of a major hemelipoprotein (HeLp), from the hemolymph of the cattle tick, which is capable of binding additional heme and transport it to tick tissues.

EXPERIMENTAL PROCEDURES
Animals and Hemolymph-Ticks were from Porto Alegre strain of B. microplus, free of Babesia spp., and were reared on calves obtained from a tick-free area. The colony is maintained at the Faculdade de Veteriná ria at the Universidade Federal do Rio Grande do Sul, Brazil. Engorged adult females were kept in Petri dishes at 28°C and 80% humidity until completion of oviposition and death, for about 12 days. Hemolymph was collected between the 5th and 8th days after dropping from the bovine with a glass microcapillary pipette by puncturing the cuticle with a needle and applying gentle pressure to the tick abdomen. Hemolymph was mixed with a collecting solution (1:1, v/v) containing 0.15 M NaCl, 5 mM EDTA, and protease inhibitors (0.05 mg/ml soybean trypsin inhibitor, 0.05 mg/ml leupeptin, 1 mM benzamidine, and 2.5 mM pepstatin. The solution was centrifuged at room temperature for 5 min at 13,000 ϫ g, and the pellet was discarded. The supernatant was stored at Ϫ70°C until HeLp isolation. Experiments in vivo were carried out with fully engorged females, at the 4th day after dropping. HeLp Purification-The hemolymph mixed with collecting solution (2 ml) was diluted 10 times in 10 mM Tris-HCl, pH 8.4, and applied to a DEAE-Toyopearl column (15 ϫ 2 cm), equilibrated with the same buffer. The column was eluted with a NaCl gradient (0 -300 mM), and  1 were pooled, diluted 10 times in phosphate-buffered saline (PBS; 0.15 M NaCl, 10 mM sodium phosphate, pH 7.4), and applied to an agarose-iminodiacetic acid metal ion affinity column (Sigma) loaded with CuSO 4 (IDA-Cu 2ϩ ). After elution with a glycine gradient (0 -300 mM), fractions (1.5 ml) containing HeLp were pooled and dialyzed against PBS. Protein concentrations were measured according to Lowry et al. (20) using bovine serum albumin as standard.
Polyacrylamide Gel Electrophoresis--SDS-PAGE was carried out in 5-22.5% acrylamide gradient gels (21). Gels were run at 120 V at room temperature for approximately 90 min, and stained with Coomassie Brilliant Blue G, according to Neuhoff et al. (22). The molecular mass of polypeptides in SDS-PAGE was determined using the following proteins as standards: myosin (205 kDa), ␤-galactosidase (116 kDa), phosphorylase b (97 kDa), bovine serum albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), soybean trypsin inhibitor (20.1 kDa), ␣-lactalbumin (14.2 kDa). Pore-limiting native PAGE was performed as described by Blanche et al. (23), using 5-22.5% acrylamide gradient gels, and the electrophoresis was performed for 20 h. The native molecular mass of was determined using the following proteins as standards: tireoglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), and bovine serum albumin (67 kDa). NH 2 -terminal Sequence and Amino Acid Analysis-For NH 2 -terminal sequencing, purified HeLp (0.5 nmol) was submitted to SDS-PAGE (7.5%) and transferred to a polyvinylidene difluoride membrane (24). The HeLp bands and the higher molecular mass aggregate were cut and sequenced by automatic Edman degradation using a liquid phase sequencer (Porton PI 2020/2090) with a Hewlett Packard high performance liquid chromatograph and a reverse phase AminoQuant analytical column (2.1 ϫ 250 mm). Homology searches in protein data bases (GenBank, SWISS-PROT, and EMBL) were performed. The amino acid composition was determined for total HeLp protein moiety. Samples were hydrolyzed in constant boiling (150°C) in the presence of 6 N HCl, for 90 min under N 2 atmosphere. Samples were analyzed as described by Spackman et al. (25) on a Shimadzu automated instrument. Amino acid detection was made by post column o-phthalaldialdehyde derivatives.
Lipids-Purified HeLp (2-4 mg of protein) was dialyzed against water, dried in a vacuum centrifuge (Speed-Vac, Savant SC110) and weighted. Lipids were extracted with chloroform-methanol-water (2:1: 0.5; v/v) (26), and the organic phase containing lipids and the protein pellet were dried under a N 2 stream and weighed to determine the lipid content. Lipids were analyzed by high performance thin layer chromatography (HPTLC) on Silica gel 60 plates (Merck, Darmstadt, Germany). Plates were first developed in chloroform-methanol-acetone-acetic acid-water (40:13:15:12:8, by volume) until the solvent front reached the middle of the plate, for separation of phospholipids (27), and then in hexane-ethyl ether-acetic acid (60:40:1, by volume), for separation of neutral lipids (28). HPTLC plates were stained by spraying with sulfuric acid (30%) and heating at 120°C for 10 -15 min (29). Quantification of different lipids was performed by computer scanning and analysis of stained HPTLC plates with the Quantiscan software (Biosoft, Cambridge, United Kingdom). Quantification of different lipids was performed by computer scanning and analysis of stained HPTLC plates with the Quantiscan software (Biosoft).
Carbohydrate Content-Total reducing sugar content was determined according to the method of Du Bois (30). To analyze carbohydrate composition, 2 mg of HeLp was submitted to acid hydrolysis (6 N trifluoroacetic acid, 100°C for 5 h) and sugars were fractionated by paper chromatography using butanol/pyridine/water (3:2:1, v/v) as solvent (31).
Absorption Spectra and Heme Content-Light absorption spectra of native HeLp were obtained in PBS using a GBC-920 spectrophotometer (Victoria, Australia). Heme content was determined from the reduced minus oxidized spectra of the pyridine alkaline derivative (32).
Density Determination-Density of the HeLp particle was determined by KBr gradient (1.16 -1.35 g/ml, 10 ml) ultracentrifugation using a 50Ti Beckman rotor (20 h at 4°C, 45,000 rpm) using the conditions described by Shipman et al. (33). Refraction index and absorbance at 280 nm of the gradient fractions were determined, and samples were submitted to a SDS-PAGE.
Electrospray Mass Spectrometry (ES-MS)-Samples used for mass spectrometry analysis were total lipid extracts or lipids eluted from the HPTLC plate (with chloroform:methanol, 1:1). Lipids were dissolved in chloroform:methanol (1:1), containing 10 mM ammonium acetate and 0.1% formic acid. Spectra were obtained in a Finnigan LCQDuo ion trap mass spectrometer (Finnigan, ThermoQuest Inc., San Jose, CA). Samples were introduced into the electrospray source by injection through a 50 m fused silica capillary at a flow rate of 5 l/min. ESI capillary voltage was set to 36 -46 V, and temperature to 250°C. Spectra were acquired at 3 s/scan over a mass range of m/z 150 -2000. Collisioninduced (ES-MS/MS) fragmentation of parent ions was carried at relative collisional energy from 25 to 35 V. Source parameters were optimized using cholesterol, cholesteryl oleate, triacylglycerol (triolein), phosphatidylcholine, phosphatidylethanolamine, and phosphatidyl- 1 The abbreviations used are: PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; PC, phosphatidylcholine; ES, electrospray; MS, mass spectrometry; HPTLC, high performance thin layer chromatography; IDA-Cu 2ϩ column, agarose-iminodiacetic acid metal ion affinity column loaded with CuSO 4 . Antibodies and Immunological Assays-HeLp (1 mg) was emulsified with an equal volume of Freund's complete adjuvant and injected subcutaneously in the back of a rabbit. Boosters of 0.5-1.0 mg were injected at 2-month intervals, and blood was collected from an ear vein 3 months after the first injection. Western blots were performed according to Towbin et al. (34).
Heme Binding Assays-Binding of heme to HeLp (1 M, 1 ml) was followed by adding a solution of 0.5 mM hemin in 0.01 M NaOH and following the absorbance at 411 nm, the hemin isosbestic point (35). Hemin bound to HeLp has a higher molar extinction coefficient than hemin in aqueous buffer. Plotting the absorbance against the amount of added hemin reveals binding as a steeper initial part of the curve, and saturation of binding sites is determined by the intersection with a second component of the curve, which is parallel to hemin added to buffer. Alternatively, HeLp can be observed in a native PAGE gel without staining, due to its heme content. This allowed direct visualization of binding by adding 1 mM hemin in 0.1 M NaOH to purified HeLp (50 g) or crude hemolymph (1 l), applying the samples to a pore-limiting PAGE as described above and recording the gels with a computer scan.
Heme Transport-Radiolabeled 55 Fe-heme was synthesized as described by Galbraith et al. (36). HeLp was incubated with 55 Fe-heme at 1:1 molar ratio and passed through spin columns (37). 55 Fe-Heme-HeLp (5 l, 9000 CPM) was injected into the hemocoel of female tick, and animals were maintained at 28°C or 4°C. Hemolymph and ovaries were collected after different times, and radioactivity was determined in a scintillation counter.

RESULTS AND DISCUSSION
HeLp Purification-Isolation of HeLp was achieved by means of a DEAE-Toyopearl column, followed by chromatography in an IDA-Cu 2ϩ column (Fig. 1, A and B). The native molecular mass of HeLp, determined by pore-limit PAGE, was 342 kDa (Fig. 1C), a value close to the 365 kDa obtained by gel filtration fast protein liquid chromatography (data not shown). Therefore, an average molecular mass of 354 kDa is assumed hereafter. HeLp is composed by two apoproteins of 103 kDa (apoHeLp-A) and 92 kDa (apoHeLp-B), as judged by SDS-PAGE of the purified lipoprotein (Fig. 1D). The native PAGE profile showed that HeLp is the main hemolymphatic protein (Fig. 1C), and estimation of its concentration in the hemolymph by radial immunodiffusion showed titers around 50 mg/ml throughout adult development (data not shown). The apoproteins were found in a stoichiometry of 1:1, as judged by densitometry of the gel. Fig. 2 shows the NH 2 -terminal amino acid sequence of 103-and 92-kDa apoproteins, obtained by Edman degradation. Homology searches did not indicate significant similarity to known proteins. The higher molecular mass pro-tein bands in the SDS-PAGE gel (Fig. 1D) were shown to be aggregates of both polypeptides, as protein sequencing of its NH 2 -terminal portion resulted in a combination of both apoprotein sequences.
Western blot using antiserum against purified HeLp showed that HeLp was not sex-specific (Fig. 3), being found in hemolymph from males (lane 1), fully engorged females (lane 2), and partially engorged adult females (collected before dropping from cattle) (lane 3). Plasma lipoproteins from both insects and vertebrates frequently are incorporated into yolk of developing oocytes, as it is the case of avian low density lipoprotein (38) or lipophorin, the main lipoprotein of insect hemolymph (39).   However, HeLp was not found in mature Boophilus eggs by Western blot (Fig. 3, lane 4), despite the high titer present in the hemolymph. Physical Properties and Chemical Composition-The amino acid composition of HeLp is very similar to an eukaryotic average protein (40), except for a rather low isoleucine content (Table I). HeLp isoelectric point was pH 5.8. HeLp carbohydrate content was 3% (w/w), and mannose was the most abundant component (Ͼ90%; data not shown), a feature shared with most arthropod glycoproteins, where high mannose type oligosaccharides with a low degree of trimming are the general pattern (41). Purified HeLp has approximately 33% of lipids (55% neutral lipids and 44% phospholipids) (Table II and (Fig. 4B). Based on previous data (42)(43)(44) and on the collision-induced dissociation daughter-ion spectra (data not shown) of the pseudomolecular ion species above, we were able to confirm the proposed assignments. The pseudomolecular ion species at m/z 663.4 most probably is an oxidized form of cholesterol oleate, which we were unable to assign as a precisely defined oxysterol ester. A true sample of cholesterol oleate (from Sigma, M r ϭ 651.1; m/z 668.5 ϭ [M ϩ NH 4 ] ϩ ) was partially converted to m/z 663.4 ion species when submitted to an air stream for 2 h (data not shown). The pseudomolecular ion species at mz 663.5 was also found in lipid extracts from crude hemolymph, and its relative intensity increased upon handling of samples during protein isolation and lipid extraction (data not shown). Therefore, the fraction isolated from HPTLC spot migrating as cholesterol ester showed only the m/z 663.4 ion species (Fig. 4B,  inset). In Fig. 4C (inset), we show that the pseudomolecular ion species at m/z 668.  at m/z 668.3 and 650.3 (Fig. 4C). To our knowledge, this is the first report of a cholesterol ester-rich lipoprotein in invertebrates. A density of 1.29 g/ml was determined by KBr gradient ultracentrifugation, a value that is higher than those reported for the more dense forms of high density lipoprotein, the mammalian cholesterol ester-rich lipoprotein, which are about 1.21 g/ml and have a lipid content of about 50% (45). This observed high density would fall into the very high density lipoprotein class, according to the classification of mammalian lipoproteins.
Analysis of HeLp secondary structure by circular dichroism (data not shown) showed a spectrum that suggests a reduced amount of ␣-helix when compared with mammalian high density lipoprotein apoproteins (46). The HeLp CD spectrum resembles those of insect lipophorins, which are characterized by a high proportion of ␤-sheet (47,48). However, despite this common feature, other aspects suggest that HeLp is a distinct particle, such as the absence of cholesterol ester in lipophorins, where diacylglycerol is the most abundant neutral lipid (49). Furthermore, lipophorins from all insect species have a very conserved quaternary structure, with one apoprotein of 220 -240 kDa and another with 70 -80 kDa (49), distinct from HeLp-A and HeLp-B. Nevertheless, a more detailed structural characterization of HeLp, including complete apoprotein sequence, will be necessary to exclude similarity to insect or vertebrate lipoproteins.
Heme Content and Binding-The most remarkable characteristic of HeLp when compared with all other known lipoproteins is the presence of about 2 mol of heme/mol of HeLp (Table  II). As can be seen in Fig. 4A, there is some amount of heme that is recovered together with lipids, and is separated by the HPTLC system, but the majority remains at the aqueous phase. HeLp visible absorption spectrum was clearly that of a hemeprotein, showing a characteristic peak at 398 nm (Fig. 5). However, the HeLp Soret bandwidth (about 80 nm) was unusually large when compared with other hemeproteins such as cytochromes and hemoglobin, which are typically in the range of 40 -50 nm, thus suggesting the existence of multiple quantum states of the heme molecules in HeLp. Spectral profiles similar to that of HeLp were reported for albumin-bound heme, for heme bound to phospholipid membranes, and for heme in organic solvents such as Me 2 SO (50,51). Heme has been implicated in oxidative damage to membrane phospholipids and to lipoproteins such as vertebrate low density lipoprotein (4 -6). Lipophorin from the blood-sucking insect Rhodnius prolixus also was shown to be damaged by heme-promoted oxidation (10), and, in this case, one of the antioxidant mechanisms described was a heme-binding protein that renders the heme molecule redox inactive, thus avoiding the production of free radicals.
Besides having two heme molecules bound, purified HeLp was capable of binding additional heme. This can be directly observed by the intensification of the HeLp band in native page without staining when it was incubated with hemin (Fig. 6A). The same experiment was carried out with crude hemolymph, showing that the binding of heme by HeLp was not an artifact generated during isolation of the protein (Fig. 6A). Moreover, only HeLp showed up in the gel after addition of heme, indicating that it is the main heme-binding protein in the hemolymph of the tick. Spectrophotometric titration of the hemebinding sites of purified HeLp revealed that each molecule of HeLp can bind up to six additional molecules of heme (Fig. 6B). Binding of these extra heme molecules did not altered HeLp far UV circular dichroism spectrum (data not shown), similar to results found for binding of heme by vertebrate albumin (52). The interactions of HeLp heme with its lipid and protein moieties, as well as its possible antioxidant role, are currently under study in our laboratory.
Heme Transport-In mammals, the intracellular pool of heme is controlled in order to keep equilibrium between synthesis and degradation (1). However, Boophilus microplus is not able to synthesize its own heme; therefore, we suggest that an obligatory counterpart of tick dependence on heme from its host should be the development of efficient ways to absorb heme from the midgut and transport this molecule to tissues. In mammals, there are two proteins involved in extracellular heme transport: hemopexin and albumin (12). As HeLp is the main heme-protein in the hemolymph, it would be a candidate to act as a vehicle in transport of heme to tissues. Oogenesis is the main event occurring in B. microplus adult females and tick eggs contain a large amount of heme, which gives then their characteristic deep brown color (13). Therefore, HeLp labeled with 55 Fe-heme was injected into the hemocoel of female ticks, and radioactivity at hemolymph and ovaries were accompanied FIG. 6. Binding of heme to HeLp. A, purified HeLp (50 g) and hemolymph (Hemol, 100 g, of protein) were incubated in the presence and absence of hemin and applied to a pore limit native PAGE, and gel was scanned before staining. B, HeLp (1 M) was titrated with hemin by adding hemin to a cuvette and measuring the absorbance at 411 nm. Hemin was added to the HeLp in PBS solution (q) or to PBS as a control curve (E). Shown are results typical of four independent experiments. FIG. 7. Heme transport. 55 Fe-Heme-HeLp (9 ϫ 10 3 cpm, 5 l) was injected into tick hemocoel, and radioactivity in hemolymph samples (A) or in the ovaries (B) was measured. Ticks were maintained either at 28°C (panel A, q; panel B, as indicated) or 4°C (panel A, (E); panel B, as indicated). Data are mean Ϯ S.E. for four determinations. (Fig. 7). In panel A we observed that radioactivity quickly decreased in the hemolymph after injection, in parallel to incorporation by the ovaries (panel B). Ticks maintained at 4°C after injection did not show either ovarian uptake nor hemolymph clearance, suggesting dependence on active metabolism. Taken together, our data indicate that HeLp plays a key role in the transport and recycling of heme in the tick, featuring as a major biochemical adaptation to blood feeding. We are currently looking for intracellular proteins involved in the uptake and reutilization of heme by tick tissues. Besides elucidating an important aspect in the biology of ticks, heme reutilization mechanisms could be targets in the development of control methods specifically directed toward hematophagous animals.