• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Mar 10, 2009; 106(10): 3728–3733.
Published online Feb 20, 2009. doi:  10.1073/pnas.0813190106
PMCID: PMC2656148

Multifunctionality and mechanism of ligand binding in a mosquito antiinflammatory protein


The mosquito D7 salivary proteins are encoded by a multigene family related to the arthropod odorant-binding protein (OBP) superfamily. Forms having either one or two OBP domains are found in mosquito saliva. Four single-domain and one two-domain D7 proteins from Anopheles gambiae and Aedes aegypti (AeD7), respectively, were shown to bind biogenic amines with high affinity and with a stoichiometry of one ligand per protein molecule. Sequence comparisons indicated that only the C-terminal domain of AeD7 is homologous to the single-domain proteins from A. gambiae, suggesting that the N-terminal domain may bind a different class of ligands. Here, we describe the 3D structure of AeD7 and examine the ligand-binding characteristics of the N- and C-terminal domains. Isothermal titration calorimetry and ligand complex crystal structures show that the N-terminal domain binds cysteinyl leukotrienes (cysLTs) with high affinities (50–60 nM) whereas the C-terminal domain binds biogenic amines. The lipid chain of the cysLT binds in a hydrophobic pocket of the N-terminal domain, whereas binding of norepinephrine leads to an ordering of the C-terminal portion of the C-terminal domain into an α-helix that, along with rotations of Arg-176 and Glu-268 side chains, acts to bury the bound ligand.

Keywords: biogenic amine, bloodfeeding, leukotriene, odorant-binding protein, saliva

Successful bloodfeeding by female mosquitoes is a key event in the transmission of a variety of viral and parasitic diseases, including malaria and dengue fever. The salivary secretion injected into the host at the time of feeding is rich in pharmacologically active substances that facilitate the uptake of blood by counteracting host hemostatic, inflammatory, and immunological defenses (1, 2). With the onset of feeding, biogenic amines are rapidly released into the blood and skin from platelets, mast cells, and sympathetic neurons, while eicosanoids are secreted by mast cells (36). These molecules mediate a variety of early responses to wounding including vasoconstriction, platelet activation, swelling, itching, and pain. Prevention of these processes is particularly important for avoiding host defensive behaviors that may interrupt feeding. One mechanism for the neutralization of small-molecule effectors of inflammation is to rapidly bind them with abundant proteins having high-affinity ligand-binding sites (3, 4, 6, 7). Bloodfeeding ticks and triatomine bugs produce proteins belonging to the lipocalin family that act in this manner (3, 6, 7). More recently, the D7 family of proteins from mosquitoes has been found to perform essentially the same function (4, 8, 9).

The salivary D7 proteins in Aedes spp. and the D7-related (D7R) proteins in Anopheles spp. show a distant relationship with the arthropod odorant-binding protein (OBP) superfamily (4, 1013). D7 variants having one or two OBP domains are among the most abundant proteins in the salivas of culicine (Aedes and Culex spp.) and anopheline mosquitoes (1416). Proteins having a biogenic amine-binding function in Aedes spp. contain two OBP domains, whereas the anopheline forms have a single OBP domain. Here, we describe the structure and function of AeD7, the two-domain D7 protein from Aedes aegypti saliva. We find that in accordance with its two-domain structure, the protein is bifunctional and acts to neutralize two different classes of inflammatory mediators. It is the C-terminal domain that exhibits the described biogenic amine-binding function, whereas the N-terminal domain is found to have a novel leukotriene-binding activity. The binding of biogenic amines with the C-terminal domain is shown to involve extensive conformational changes in the protein that act to stabilize the bound ligand while at the same time increasing the structural diversity of mediators that can be accommodated at the ligand-binding site.


Overall Structure of AeD7.

The structure of AeD7 was determined by using single anomalous dispersion methods with a crystal derivatized by briefly soaking in cryoprotectant containing potassium bromide [supporting information (SI) Table S1]. As suggested from sequence comparisons, AeD7 contains two OBP-like domains that are distantly similar to one another in amino acid sequence (15% overall identity). In the ligand-free structure, each domain is made up of seven α-helices with the N-terminal domain (residues 1–150) being stabilized by two disulfide bonds and the C-terminal domain (residues 151–303) by three disulfide bonds (Fig. 1). Disulfide bonds of the N-terminal domain link Cys-16 of helix B with Cys-53 of helix C, and Cys-49 of helix C with Cys-106 of helix F. In the C-terminal domain disulfide bonds link Cys-157 of helix A2 with Cys-192 of helix C2, Cys-232 of helix E2 with Cys-250 of helix F2, and Cys-173 of helix B2 with Cys-301 near the C terminus. Of the five disulfides, four are found in other members of the arthropod OBP protein family. The disulfide bond that links helix B2 with the C terminus is found only in the mosquito AeD7 group (4, 8).

Fig. 1.
Stereo ribbon diagram of the A. aegypti D7 without ligands. The N-terminal domain is colored red with the α-helices labeled A–G. The C-terminal domain is colored blue with the α-helices labeled A2–G2. Disulfide bonds are ...

The two domains of AeD7 are connected by a nine-residue coil and contact each other via an interface comprising parts of helix A from the N-terminal domain and portions of helices C2, D2, and E2 from the C-terminal domain (Fig. 1). The interface buries 1,770 Å2 of the molecular surface and is stabilized by numerous salt bridge and cation–π interactions (Fig. S1). Superposition of helices A–F (rmsd 2.3 Å for 57 Cα positions) in the N- and C-terminal domains shows that a kink in helix F of the N-terminal domain corresponds to a tight turn connecting helices F2 and G2 in the C-terminal domain. The C-terminal domain is similar in sequence and structure to the single domain biogenic amine-binding D7R proteins from Anopheles spp., suggesting that it is the biogenic amine-binding domain. The putative ligand-binding pocket of the N-terminal domain is substantially different in structure, suggesting a distinct ligand-binding function.

Binding Pocket Structure of the N-Terminal Domain.

The N-terminal domain of AeD7 has a centrally located, solvent-accessible ligand-binding pocket bounded by helices A–C and F that leads from the surface of the protein to the interior (Fig. 2). The channel is lined primarily, but not exclusively, with hydrophobic residues including Phe-13, Trp-37, Val-54, and Leu-55. The apparent entry point to the channel is located at the center of a positively charged surface of the N-terminal domain bounded by helices A, B, and G (Figs. 1 and and3).3). Based on the architecture and relatively hydrophobic nature of this putative ligand-binding site, we hypothesized that it may accommodate bioactive lipid mediators involved in host inflammatory responses to mosquito feeding.

Fig. 2.
Structure of the N-terminal domain of D7. (A) Ribbon diagram of the N-terminal domain (red) showing a bound molecule of 5(S)-hydroxy-(E,E,Z,Z)-7,9,11,14-eicosatetraenoic acid (cyan). (B) Bound ligand (cyan, with oxygen atoms colored in red) covered by ...
Fig. 3.
Electrostatic potential surfaces of AeD7 and D7R4 calculated using APBS (36). (A) Ligand-free AeD7 oriented in a manner similar to Fig. 1 Upper, with the N-terminal domain labeled as NT and the C-terminal domain labeled as CT. Surfaces having negative ...

Binding Specificity of the N-Terminal Domain.

A panel of proinflammatory and prohemostatic lipid compounds was tested in binding assays using isothermal titration calorimetry (ITC). These included a number of leukotrienes (LTs; Fig. 4), and the platelet activators arachidonic acid (Fig. 4), platelet-activating factor (PAF), and the thromboxane A2 analog U46619.

Fig. 4.
Structures of eicosanoid compounds tested for binding with D7. (A) CysLTs with the variable peptide substituent indicated as R. (B) Structures of the peptide portions of the cysLTs. R1, LTC4; R2, LTD4; R3, LTE4. (C) Structure of 14,15-LTC4. (D) Structure ...

The highest binding affinities were observed for LTs, particularly the cysteinyl LTs (cysLTs) LTC4, LTD4, and LTE4, which are released by mast cells in the skin where they serve as important mediators of inflammation and vascular tone [Table 1 and Figs. S2 and S3 (17)]. These compounds consist of an eicosenoid fatty acid conjugated with glutathione, or a glutathione fragment, via a thioether linkage at C6 of the lipid chain (Fig. 4). LTC4 is secreted from cells and then metabolized to LTD4 and LTE4 by successive cleavages of the glutathione group (Fig. 4). The similar binding affinities for all three cysLTs suggest that the lipid chain is the primary contact point with the protein and that with the possible exception of the cysteinyl group, the peptide portion of the molecule plays only a minor role in binding. The binding affinity for LTB4 was 2- to 3-fold lower than for the cysLTs (Table 1). This compound is structurally related to the cysLTs but has no conjugated peptide, is hydroxylated at both C5 and C12 of the lipid chain, and has a 6Z, 8E, 10E, 14Z pattern of unsaturations rather than the 7E, 9E, 11Z, 14Z pattern seen with the cysLTs (Fig. 4). As would be expected from a reaction mechanism dominated by the hydrophobic effect, the temperature dependence of ΔH for cysLT binding was strongly negative, producing values for the change in heat capacity (ΔCp) of ≈1 kcal mol−1 K−1 (Table 1 and Fig. S3).

Table 1.
Binding parameters for N-terminal domain ligands as measured by ITC

Of the other lipids tested, only the phosphatidylcholine derivative PAF showed detectable binding, but the affinity was ≈200-fold lower than observed for the cysLTs (Table 1), suggesting that the interaction is probably not physiologically relevant (Table 1). Arachidonic acid and U46619 showed no detectable binding, further indicating that the primary function of the N-terminal domain is to limit inflammatory responses via LT binding rather than inhibiting platelet activation by binding other mediators.

The lack of discrimination between the different cysLTs tested in binding experiments suggested that the lipid chain is inserted into the narrow hydrophobic channel of the N-terminal domain whereas the glutathione moiety of the cysLTs and the charged head group of PAF lie outside of the channel. A positional isomer of LTC4, 14,15-LTC4, has the glutathione conjugated at C14 of the lipid chain and a hydroxyl group at C15, rather than having conjugation at C6 and hydroxylation at C5 (Fig. 4). This compound showed no detectable binding, indicating that a bulky group near the ω-end of the lipid can interfere with ligand association and that it is this end of the molecule that inserts into the binding pocket. Furthermore, the importance of the 5-hydroxyl group was demonstrated by the diminished binding exhibited by (Z)-11-eicosenoic acid (Table 1). Although this compound otherwise resembles the lipid portion of the cysLTs in shape and polarity, it lacks a 5-hydroxyl and exhibits a binding affinity ≈15-fold lower than the cysLTs (Table 1).

Structure of the AeD7–LTE4 Complex.

The structure of the complex between AeD7 and LTE4 was determined after soaking a crystal in 250 μM LTE4 added to the crystallization drop for 2 days before freezing (Table S1). Continuous electron density corresponding to the lipid chain of the ligand (FoFc omit map contoured at 2.5 σ, with phases calculated from a ligand-free AeD7 model) was observed in the binding pocket of the N-terminal domain. A molecule of 5(S)-hydroxy-(E,E,Z,Z)-7,9,11,14-eicosatetraenoic acid was built into the electron density, and the model of the complex was refined (Fig. 2). It is clear from the refined structure that the fatty acyl portion of the molecule is inserted into the binding channel with its ω-end occupying the interior-most point. The lipid chain could be traced from its ω-end to the carboxyl group that forms hydrogen bonds with Lys-149 and Thr-135 (Fig. 2). The hydroxyl substituent at C5 is also visible, forming hydrogen bonds with the side chain of Trp-37 and the carbonyl oxygen of Gly-130, but the cysteinyl group is not well ordered, and only unconnected fragments of positive electron density are visible at its likely location in the 2FoFc map.

Structure of the C-Terminal Domain.

The C-terminal domain of AeD7 is similar in its sequence and function to the single-domain D7R group of proteins found in Anopheles gambiae (8). Full-length AeD7 has been shown to bind biogenic amines in a manner similar to the D7R proteins (4). The stoichiometries of both biogenic amine (4) and LT binding (Table 1), as measured by ITC, are near unity, indicating that a single binding site for each type of ligand is present in each AeD7 molecule. LTD4 was found to bind normally in the presence of a saturating (20 μM) concentration of serotonin (Table 1), confirming that cysLTs bind only to the N-terminal domain, biogenic amines only to the C-terminal domain, and that binding of biogenic amines does not affect the binding of LTs.

As expected from the similarity of their sequences, the structure of the C-terminal domain of ligand-free AeD7 is similar to the reported structure of D7R4 from A. gambiae (8). Surprisingly, the C-terminal eighth helix (H) seen in D7R4 is missing in AeD7 (Fig. 5). Instead, this part of the protein forms a coil extending from residue 280 to 293 in which residues 290–293 are poorly ordered, followed by an ordered region containing a type I′ β-turn (18), extending from Met-294 to Cys-301 (Fig. 5 and Fig. S4). The observed conformation exposes a large, negatively charged surface that corresponds to the interior of the ligand-binding pocket of D7R4 (Fig. 3).

Fig. 5.
Structure and conformational changes of the C-terminal domain. (A) Unliganded structure showing nonhelical C terminus and open conformation of Arg-176 and Glu-268. The poorly ordered portion of the C-terminal coil is shown in red, and the terminal disulfide ...

Conformational Change Stabilizes the Norepinephrine Complex.

The structure of the AeD7–norepinephrine complex was determined after soaking a crystal in cryopreservative containing 1 mM norepinephrine before freezing (Table S1). Upon binding of the ligand, the C-terminal portion of the biogenic amine-binding domain is ordered into an α-helix (H2) extending from Ser-285 to Cys-301 and corresponding in its position to the eighth α-helix (helix H) of D7R4 (Figs. 5 and and66A and S4). The formation of helix H2 is accompanied by rotations of the side chains of Arg-176 and Glu-268 that act as a gate, trapping the bound ligand and further closing the entry path (Fig. 5 and Fig. S5). In its new position, the side chain of Glu-268 forms a hydrogen bond with the amino group of norepinephrine (Fig. 7A and Fig. S6). Previous studies in which the corresponding residue in D7R4 (Glu-114) was mutated show that this electrostatic interaction is important in stabilizing the bound ligand (8). The ordering of helix H2 also brings the side chain of Asp-297 into position to hydrogen bond with the amino group of the bound ligand through an intervening water molecule (Fig. 7A). An analogous interaction, involving Asp-139, is also observed in the D7R4 complex with serotonin [Fig. 7C (8)]. When the ligand-bound AeD7 C-terminal domain and the D7R4–serotonin complex are superimposed, the binding pocket structures are found to be nearly identical (Fig. 7 A and C).

Fig. 6.
Comparison of ligand entry pathways in AeD7 and D7R4 from A. gambiae. (A) Stereo representation of the Cα trace for the C-terminal domain of AeD7 (blue) and D7R4 (red) showing the overall similarity of the ligand-bound structures. The eight α-helices ...
Fig. 7.
Comparison of ligand-binding modes in the C-terminal domain of A. aegypti D7 and A. gambiae D7R4. (A) Stereoview of the C-terminal domain ligand-binding pocket of D7 (cyan) complexed with norepinephrine (green). Hydrogen bonds are shown as broken red ...


Inflammatory Mediators and Mosquito Feeding.

In the short time needed for a female mosquito to obtain a blood meal, swelling, itching, and pain responses are initiated in the host. Presumably, the antiinflammatory agents contained in saliva are aimed at limiting these early responses. As a result of biting, biogenic amines are released by platelets and mast cells in the skin and blood where they elicit pain responses (1921), which in turn trigger defensive behaviors by the host. Additionally, norepinephrine and serotonin stimulate adrenergic and serotonin receptors in the vasculature, resulting in constriction of blood vessels and limitation of the flow of blood to the mouthparts. In a parallel fashion, bioactive lipids, most notably the cysLTs, are released by mast cells and induce swelling, erythema, pain, and itching that begin almost immediately (22, 23). In some tissues, these compounds also induce vasoconstriction (22). Again, limitation or delay of these effects may be critical to ensuring an uninterrupted period of feeding. Any disruption of feeding by a female mosquito would have the obvious consequence of reducing the efficiency of virus or parasite transmission, which underscores the importance of understanding how mosquitoes delay the host response.

AeD7 is tailored to bind a number of biologically important molecules while maintaining a high degree of selectivity. The N-terminal domain is specific for bioactive lipids, particularly the cysLTs, whereas the C-terminal domain binds only biogenic amines. The N-terminal domain binding pocket is a mainly hydrophobic channel leading to the protein interior (Fig. 2), whereas the C-terminal pocket contains polar and charged residues that form electrostatic and hydrogen-bonding interactions with the ligand (Fig. 7A and Fig. S6). The electrostatic surfaces presented at each binding site are consistent with the type of ligand bound (Fig. 3). The N-terminal binding pocket is surrounded by basic residues that would interact favorably with the carboxyl groups of the eicosenoid and peptide portions of the cysLTs (Fig. 3). Likewise, the binding site for the cationic biogenic amines is seen as a negatively charged patch on the surface of the C-terminal domain (Fig. 3).

Significance of CysLT Binding by the N-Terminal Domain.

The similar affinities observed for LTC4, LTD4, and LTE4 suggest that removal of all of these lipid species is necessary, or at least beneficial, to a feeding mosquito. Although LTC4 is the primary mast cell product, the three cysLTs show similar activity in inducing erythema and wheal formation when injected into the skin (23). This being the case, a binding site that specifically recognizes the lipid features common to all cysLTs, without distinguishing differences in the peptide portion of the molecule, may have the most beneficial overall antiinflammatory effect.

Recognition of the lipid portion of the cysLT structure by AeD7 is based mainly on the presence of the carboxyl group, a hydroxyl group at C5, and a specific arrangement and geometry of the unsaturations. With its all-cis configuration, arachidonic acid does not take on the approximately linear conformation of the C1–C15 portion of the lipid chain that is observed in the structure of the AeD7–LTE4 complex. This, along with the lack of hydroxylation, would explain the absence of detectable binding by AeD7. LTB4 differs from the cysLTs in both hydroxylation and unsaturation pattern. It consequently binds less tightly to AeD7 than do the cysLTs and would be removed less efficiently from the area of the bite.

Conformational Change in the C-Terminal Domain and Ligand Selectivity of D7 Proteins.

The conformational change of AeD7 appears to act as a stabilizing mechanism for bound biogenic amines and may play a role in determining the binding specificity of the protein. D7R4, the only other member of the group to be structurally characterized, does not undergo a conformational change and remains in the closed conformation even in the absence of ligands. Nevertheless, it binds serotonin with an affinity equal to AeD7, whereas AeD7 binds norepinephrine as tightly as serotonin and ≈600-fold more tightly than does D7R4 (4). Conformational changes resulting in burial of the bound ligand may act to improve the binding of norepinephrine, which differs from serotonin in having a hydroxyl group at the Cβ position of the alkyl side chain. Ligand-free AeD7 presents a broad, negatively charged surface lying between helix A2, helix G2, and the C-terminal coil region that would be attractive to the cationic biogenic amines (Figs. 3A, ,5,5, and and66B). The structures described here suggest that Arg-176, Glu-268, and helix H2 fold around the ligand by an induced-fit type mechanism, possibly allowing the accommodation of a greater diversity of ligand structures and conformations. In D7R4, helix H is present in both the ligand-bound and -free forms and blocks the entry path seen in AeD7. In this case, the ligand appears to enter the more rigid structure through a channel at a negatively charged surface lying between helices B, G, and H (Figs. 3C, ,55B, and and66B).

The binding mode of norepinephrine in its AeD7 and D7R4 complexes is clearly different and consistent with the observed differences in binding affinity. In the D7R4 complex, the secondary hydroxyl group at Cβ of the ligand forms hydrogen bonds with the side chains of Asp-111 and Glu-114. In the AeD7–norepinephrine complex, the side chain of the ligand is rotated relative to its position in D7R4, allowing the amino group to form hydrogen bonds with Asp-265 and Glu-268 (equivalent to Asp-111 and Glu-114 in D7R4) and the secondary hydroxyl to hydrogen bond with the carbonyl group of Arg-176 (Fig. 7A). As situated in D7R4, norepinephrine is not in position to form hydrogen bonds between its phenolic hydroxyl groups and the protein (Fig. 7 A and B and Fig. S6). In the AeD7–norepinephrine complex, the phenolic hydroxyls of the ligand are in position to form hydrogen bonds with the side chains of Glu-158 and His-189 (Fig. 7 A and B). In the AeD7-binding mode, the ligand forms seven hydrogen bonds, whereas only four are observed in the D7R4–norepinephrine complex. The AeD7 arrangement more closely resembles the high-affinity binding mode seen in the D7R4 complex with serotonin (Fig. 7C) (4). It appears that AeD7 is able to stabilize bound norepinephrine by accommodating a different conformation of the bound ligand that allows the amino group to form salt bridges or hydrogen bonds with Asp-265 and Glu-268.

Relationships of D7 Proteins from Bloodfeeding Diptera.

Salivary proteins of bloodfeeding arthropods are commonly encoded by clusters of related genes having demonstrable functional differences. In Anopheles mosquitoes, the biogenic amine-binding function of AeD7 has been taken over by the D7R proteins that apparently resulted from duplication of the C-terminal domain of a two-domain protein similar to AeD7 (4). The two-domain D7 forms present in Anopheles (Fig. S7) show a high degree of conservation of the N-terminal domain, suggesting that they also bind LTs. The C-terminal domains of the Anopheles forms differ from AeD7 in a number of essential binding site residues, suggesting that they do not bind biogenic amines (Fig. S7) and may perform a different function.

Experimental Procedures


l-(−)-Norepinephrine-(+)-bitartrate salt monohydrate (>97% purity), (Z)-11-eicosenoic acid (99% purity), and PAF (>97% purity) were obtained from Sigma–Aldrich. LTs (>97% purity) B4, C4, D4, and E4 and arachidonic acid (> 97% purity) were obtained from Cayman Chemical. U46619 (100% purity) was from Calbiochem.

Protein Production.

AeD7 protein was produced in HEK-293 cells by using methods described in ref. 4. Processing of the signal peptide leaves an additional seven vector-encoded residues at the N terminus of the protein that were disordered in the crystal and not modeled.

Crystallization and Data Collection.

AeD7 was crystallized by using the hanging-drop vapor diffusion method from 16 to 20% PEG 6000, 100 mM Tris·HCl (pH 7.5). The crystals were frozen for data collection in the crystallization buffer described above, containing 15% glycerol. A potassium bromide derivative was obtained by soaking crystals for 30–60 s in cryoprotectant containing 1 M KBr (24). Norepinephrine was added to the crystal by soaking in cryoprotectant containing 1 mM ligand for 10–15 min. LTE4 dissolved in crystallization buffer solution was added to a crystallization drop to a concentration of 250 μM. After 2 days, the crystals were frozen by using the cryoprotectant described above.

Data collection was performed at beamlines 19-ID and 19-BM of the Structural Biology Center, Advanced Photon Source (APS), Argonne National Laboratory, or beamline 22-BM of the South East Regional Collaborative Access Team (SER-CAT), APS, Argonne National Laboratory. The crystals belong to the space group P21 and contain one monomer per asymmetric unit (Table S1). Four datasets were collected from a potassium bromide-soaked, unliganded crystal (1.70-Å resolution), an unliganded crystal (1.30-Å resolution), a norepinephrine-soaked crystal (1.75-Å resolution), and a LTE4-soaked crystal (1.80-Å resolution).

Structure Solution and Refinement.

The structure of AeD7 was determined by using single anomalous dispersion methods after data collection near the bromine absorption edge (24). The data were indexed, integrated, and scaled using HKL3000 (25). Initial phases were also obtained by using the HKL3000 package that combines SHELXD-E, MLPHARE, and DM for the location of bromine sites, solvent flattening, phasing, and density modification (2530). From the experimental map, a large part of the structure could be built by using ARP-WARP (31). The model was completed by manual building in Coot (32) and refined by using REFMAC (26). In some cases a TLS model was applied for refinement by using a single TLS group (26). The structures of the unliganded protein in the presence and absence of potassium bromide were nearly identical [rmsd for all Cα positions = 0.24 Å as measured by using LSQMAN (33)]. The initial model for 5(S)-hydroxy-(E,E,Z,Z)-7,9,11,14-eicosatetraenoic acid was built by using SYBYL (Tripos).


Binding measurements were made by using ITC on a Microcal VP-ITC microcalorimeter. Lipid ligand compounds dissolved in ethanol or chloroform were prepared by evaporating the solvent to dryness under a stream of nitrogen. The lipid ligand candidates were then dissolved in 20 mM Tris·HCl (pH 8.0), 0.15 M NaCl at concentrations beneath the critical micellar concentration (cmc) of 73 μM reported for arachidonic acid (34). LTB4, C4, D4, and E4 are reported to be soluble at concentrations above 100 μM by the producer, Cayman Chemical. In the case of PAF, the final concentration of ligand in the calorimeter cell was ≈35 μM, which is close to the reported cmc of 36 μM in buffer containing 85 mM salt (35). The observation of binding stoichiometries near 1:1 for all ligands in the ITC experiments suggests that no fraction of the lipid in solution was unavailable for binding. All lipid solutions were sonicated for 10 min before use to ensure dissolution. Experiments were run at 30 °C, and 10-μL injections of either lipid or biogenic amine were made for data collection. For heat capacity measurements of the cysLTs, data were also collected at three additional temperatures. After integration, the heats were converted to enthalpies by using the Microcal Origin software, and the heat of dilution, as estimated from postsaturation heats (Fig. S3), was subtracted from each point. Thermodynamic parameters were obtained by fitting the values to a single site binding model in the Microcal Origin software package.

Supplementary Material

Supporting Information:


We thank Drs. D. Garboczi, A. Gittis, and K. Singh for discussions and assistance with data collection. We also thank the staffs of the Structural Biology Center Collaborative Access Team and the Southeast Regional Collaborative Access Team, Advanced Photon Source, Argonne National Laboratory for assistance with X-ray data collection. This work was supported by the intramural research program of the National Institute of Allergy and Infectious Diseases, National Institutes of Health. Use of the Advanced Photon Source beamlines was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract W-31-109-Eng-38.


The authors declare no conflict of interest.

Data deposition: The atomic coordinates have been deposited in the Protein Data Bank, www.pdb.org [PDB codes 3DY9 (bromide derivative), 3DXL (unliganded structure), 3DZT (LTE4 complex), and 3DYE (norepinephrine complex)].

This article contains supporting information online at www.pnas.org/cgi/content/full/0813190106/DCSupplemental.


1. Ribeiro JMC. Blood-feeding arthropods: Live syringes or invertebrate pharmacologists. Infect Agents Dis. 1995;4:143–152. [PubMed]
2. Ribeiro JM, Francischetti IM. Role of arthropod saliva in blood feeding: Sialome and post-sialome perspectives. Annu Rev Entomol. 2003;48:73–88. [PubMed]
3. Mans BJ, Ribeiro JM, Andersen JF. Structure, function and evolution of biogenic amine-binding proteins in soft ticks. J Biol Chem. 2008;283:18721–18733. [PMC free article] [PubMed]
4. Calvo E, Mans BJ, Andersen JF, Ribeiro JM. Function and evolution of a mosquito salivary protein family. J Biol Chem. 2006;281:1935–1942. [PubMed]
5. Boyce JA. Mast cells and eicosanoid mediators: A system of reciprocal paracrine and autocrine regulation. Immunol Rev. 2007;217:168–185. [PubMed]
6. Andersen JF, Francischetti IM, Valenzuela JG, Schuck P, Ribeiro JM. Inhibition of hemostasis by a high affinity biogenic amine-binding protein from the saliva of a blood-feeding insect. J Biol Chem. 2003;278:4611–4617. [PubMed]
7. Paesen GC, Adams PL, Harlos K, Nuttall PA, Stuart DI. Tick histamine-binding proteins: Isolation, cloning, and three-dimensional structure. Mol Cell. 1999;3:661–671. [PubMed]
8. Mans BJ, Calvo E, Ribeiro JM, Andersen JF. The crystal structure of D7r4, a salivary biogenic amine-binding protein from the malaria mosquito Anopheles gambiae. J Biol Chem. 2007;282:36626–36633. [PubMed]
9. James AA, Blackmer K, Marinotti O, Ghosn CR, Racioppi JV. Isolation and characterization of the gene expressing the major salivary gland protein of the female mosquito, Aedes aegypti. Mol Biochem Parasitol. 1991;44:245–253. [PubMed]
10. Calvo E, deBianchi AG, James AA, Marinotti O. The major acid-soluble proteins of adult female Anopheles darlingi salivary glands include a member of the D7-related family of proteins. Insect Biochem Mol Biol. 2002;32:1419–1427. [PubMed]
11. Valenzuela JG, et al. The D7 family of salivary proteins in blood-sucking diptera. Insect Mol Biol. 2002;11:149–155. [PubMed]
12. Sandler BH, Nikonova L, Leal WS, Clardy J. Sexual attraction in the silkworm moth: Structure of the pheromone-binding protein–bombykol complex. Chem Biol. 2000;7:143–151. [PubMed]
13. Du G, Prestwich GD. Protein structure encodes the ligand-binding specificity in pheromone-binding proteins. Biochemistry. 1995;34:8726–8732. [PubMed]
14. Choumet V, et al. The salivary glands and saliva of Anopheles gambiae as an essential step in the Plasmodium life cycle: A global proteomic study. Proteomics. 2007;7:3384–3394. [PubMed]
15. Kalume DE, et al. A proteomic analysis of salivary glands of female Anopheles gambiae mosquito. Proteomics. 2005;5:3765–3777. [PubMed]
16. Valenzuela JG, Pham VM, Garfield MK, Francischetti IM, Ribeiro JM. Toward a description of the sialome of the adult female mosquito Aedes aegypti. Insect Biochem Mol Biol. 2002;32:1101–1122. [PubMed]
17. Khanapure SP, Garvey DS, Janero DR, Letts LG. Eicosanoids in inflammation: Biosynthesis, pharmacology, and therapeutic frontiers. Curr Top Med Chem. 2007;7:311–340. [PubMed]
18. Chou KC. Prediction of tight turns and their types in proteins. Anal Biochem. 2000;286:1–16. [PubMed]
19. Andoh T, Kuraishi Y. Intradermal LTB4, but not prostaglandin E2, induces itch-associated responses in mice. Eur J Pharmacol. 1998;353:93–96. [PubMed]
20. Oliveira MC, Pelegrini-da-Silva A, Parada CA, Tambeli CH. 5-HT acts on nociceptive primary afferents through an indirect mechanism to induce hyperalgesia in the subcutaneous tissue. Neuroscience. 2007;145:708–714. [PubMed]
21. Xanthos DN, Bennett GJ, Coderre TJ. Norepinephrine-induced nociception and vasoconstrictor hypersensitivity in rats with chronic post-ischemia pain. Pain. 2008;137:640–651. [PubMed]
22. Boyce JA. Eicosanoid mediators of mast cells: Receptors, regulation of synthesis, and pathobiologic implications. Chem Immunol Allergy. 2005;87:59–79. [PubMed]
23. Soter NA, Lewis RA, Corey EJ, Austen KF. Local effects of synthetic leukotrienes (LTC4, LTD4, LTE4, and LTB4) in human skin. J Invest Dermatol. 1983;80:115–119. [PubMed]
24. Dauter Z, Dauter M. Entering a new phase: Using solvent halide ions in protein structure determination. Structure. 2001;9:R21–R26. [PubMed]
25. Minor W, Cymborowski M, Otwinowski Z, Chruszcz M. HKL-3000: The integration of data reduction and structure solution: From diffraction images to an initial model in minutes. Acta Crystallogr D. 2006;62:859–866. [PubMed]
26. CCP4. The CCP4 suite: Programs for protein crystallography. Acta Crystallogr D. 1994;50:760–763. [PubMed]
27. Cowtan K, Main P. Miscellaneous algorithms for density modification. Acta Crystallogr D. 1998;54:487–493. [PubMed]
28. Otwinowski Z. In: Wolf W, Evans PR, Leslie AGW, editors. Isomorphous Replacement and Anomalous Scattering: Proceedings of the CCP4 Study Weekend; Warrington, UK: Daresbury Laboratory; 1991. pp. 80–86.
29. Schneider TR, Sheldrick GM. Substructure solution with SHELXD. Acta Crystallogr D. 2002;58:1772–1779. [PubMed]
30. Sheldrick GM. Macromolecular phasing with SHELXE. Z Kristallogr. 2002;217:644–650.
31. Perrakis A, Morris R, Lamzin VS. Automated protein model building combined with iterative structure refinement. Nat Struct Biol. 1999;6:458–463. [PubMed]
32. Emsley P, Cowtan K. Coot: Model-building tools for molecular graphics. Acta Crystallogr D. 2004;60:2126–2132. [PubMed]
33. Kleywegt GJ, Jones TA. Where freedom is given, liberties are taken. Structure. 1995;3:535–540. [PubMed]
34. Glick J, Santoyo G, Casey PJ. Arachidonate and related unsaturated fatty acids selectively inactivate the guanine nucleotide-binding regulatory protein, Gz. J Biol Chem. 1996;271:2949–2954. [PubMed]
35. Teruel JA, Soler F, Gomez-Fernandez JC. On the effect of lysophosphatidylcholine, platelet-activating factor and other surfactants on calcium permeability in sarcoplasmic reticulum vesicles. Chem Phys Lipids. 1991;59:1–7. [PubMed]
36. Holst M, Kozack RE, Saied F, Subramaniam S. Protein electrostatics: Rapid multigrid-based Newton algorithm for solution of the full nonlinear Poisson–Boltzmann equation. J Biomol Struct Dyn. 1994;11:1437–1445. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...