Sensory system-predominant distribution of leukotriene A4 hydrolase and its colocalization with calretinin in the mouse nervous system

Neuroscience 141 (2006) 917–927

SENSORY SYSTEM-PREDOMINANT DISTRIBUTION OF LEUKOTRIENE A4 HYDROLASE AND ITS COLOCALIZATION WITH CALRETININ IN THE MOUSE NERVOUS SYSTEM Y. CHIBA,a A. SHIMADA,a,c* M. SATOH,a,c Y. SAITOH,a N. KAWAMURA,a A. HANAI,a H. KEINO,a Y. IDE,b T. SHIMIZUb,c AND M. HOSOKAWAa

Key words: leukotriene B4, sensory nervous system, immunohistochemistry.

a Department of Pathology, Institute for Developmental Research, Aichi Human Service Center, 713-8 Kamiya-cho, Kasugai, Aichi 480-0392, Japan

Leukotrienes (LTs) are a group of biologically highly active lipid mediators, mainly involved in immediate hypersensitivity and inflammatory reactions (Samuelsson, 1983; Funk, 2001). They take part in certain pathological processes in the CNS, such as aging (McCann et al., 1998; Uz et al., 1998), viral infection (Chen et al., 2001), experimental allergic encephalomyelitis (Gladue et al., 1996), ischemia and reperfusion injury (Barone et al., 1992), and traumatic spinal cord injury (Xu et al., 1990). There are several lines of evidence suggesting that LTs are essential for normal brain function in addition to the pathological roles. LTs and their synthesizing enzymes are constitutively present in the normal brain (Shimizu and Wolfe, 1990). Furthermore, deficiency in LTC4 synthesis leads to a fatal neuro-developmental syndrome in humans (Mayatepek, 2000), suggesting a critical role for cysteinyl LTs in the development of the CNS. However, the functions of LTs in the brain are not fully understood. We have recently revealed that LTC4 synthase, which transforms LTA4 to LTC4 (Shimizu et al., 1990), selectively localizes in the hypothalamic and extrahypothalamic vasopressin systems, indicating their involvement in the neuroendocrine system and vasopressinergic neural functions (Shimada et al., 2005). LTB4 is a mediator of various biological responses, such as leukocyte chemotaxis and aggregation, host-defense against infections (Bailie et al., 1996), platelet-activating-factor-induced lethal shock (Byrum et al., 1999), and vascular inflammation and arteriosclerosis (Jala and Haribabu, 2004). LTB4 is synthesized from LTA4, the same substrate for LTC4 synthase, through enzymatic hydrolysis by LTA4 hydrolase (Radmark et al., 1984). Enzymatic activity and the mRNA of LTA4 hydrolase are detected in the CNS (Izumi et al., 1986; Minami et al., 1995). Furthermore, LTB4 is synthesized from endogenous arachidonic acid in guinea-pig brain slices when stimulated with a calcium ionophore (Shimizu et al., 1987). High activity of LTA4 hydrolase is detected in the olfactory bulb, pituitary gland, hypothalamus, and cerebral cortex (Shimizu et al., 1987). These previous studies strongly suggest that LTB4, as well as LTC4, plays an important role in the CNS. However, the detailed cellular localization of LTB4 biosynthesis and its receptors remain to be elucidated. In this work, we examined the detailed distribution of LTA4 hydrolase in the mouse nervous system by immuno-

b Department of Biochemistry and Molecular Biology, Faculty of Medicine, The University of Tokyo, Tokyo 113-0033, Japan c

Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology Corporation, Japan

Abstract—Leukotriene B4 is a potent lipid mediator, which has been identified as a potent proinflammatory and immunomodulatory compound. Although there has been robust evidence indicating that leukotriene B4 is synthesized in the normal brain, detailed distribution and its functions in the nervous system have been unclear. To obtain insight into the possible neural function of leukotriene B4, we examined the immunohistochemical distribution of leukotriene A4 hydrolase, an enzyme catalyzing the final and committed step in leukotriene B4 biosynthesis, in the mouse nervous system. Immunoreactivity for leukotriene A4 hydrolase showed widespread distribution with preference to the sensory-associated structures; i.e. neurons in the olfactory epithelium and vomeronasal organ, olfactory glomeruli, possibly amacrine cells, neurons in the ganglion cell layer and three bands in the inner plexiform layer of the retina, axons in the optic nerve and tract up to the superior colliculus, inner and outer hair cells and the spiral ganglion cells in the cochlea, vestibulocochlear nerve bundle, spinal trigeminal tract, and lamina II of the spinal cord. Double immunofluorescence staining demonstrated that most of the leukotriene A4-hydrolase-immunopositive neurons coexpressed calretinin, a calcium-binding protein in neurons. The ubiquitous distribution of leukotriene A4 hydrolase was in sharp contrast with the distribution of leukotriene C4 synthase [Shimada A, Satoh M, Chiba Y, Saitoh Y, Kawamura N, Keino H, Hosokawa M, Shimizu T (2005) Highly selective localization of leukotriene C4 synthase in hypothalamic and extrahypothalamic vasopressin systems of mouse brain. Neuroscience 131:683– 689] which was confined to the hypothalamic and extrahypothalamic vasopressinergic neurons. These results suggest that leukotriene B4 may exert some neuromodulatory function mainly in the sensory nervous system, in concert with calretinin. © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. *Correspondence to: A. Shimada, Department of Pathology, Institute for Developmental Research, Aichi Human Service Center, 713-8 Kamiya-cho, Kasugai, Aichi 480-0392, Japan. Tel: ⫹81-568-88-0811; fax: ⫹81-568-88-0829. E-mail address: [email protected] (A. Shimada). Abbreviations: DAB, diaminobenzidine; EDTA, ethylenediaminetetraacetic acid; IHC, inner hair cell; INL, inner nuclear layer of the retina; LT, leukotriene; OHC, outer hair cell; PBS, phosphate-buffered saline; VNO, vomeronasal organ.

0306-4522/06$30.00⫹0.00 © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2006.04.012

917

918

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

histochemistry. We found that the distribution pattern of LTA4 hydrolase was quite similar to that of calretinin, a calcium binding protein in neurons, and investigated whether these two proteins colocalized in the same neurons using the double immunohistochemical labeling technique.

EXPERIMENTAL PROCEDURES Animals and tissue preparation Five male C57BL/6NCrj mice at the age of 8 weeks were anesthetized by an i.p. injection of sodium pentobarbital, perfused with phosphate-buffered saline (PBS) at a rate of 12.1 ml/min for 5 min and then Bouin’s fixative at the same rate for 10 min. After the perfusion, the mice were decapitated and brains, trigeminal nerves, pituitary glands, spinal cords with vertebrae and organs other than the nervous system (lung, pancreas, etc.) were removed and postfixed in the same fixative for 7 days at 4 °C. Inner ears were removed from temporal bone, postfixed in Bouin’s fixative for 3 days at 4 °C, and decalcified in 10% EDTA in PBS (pH 7.2) for 4 days at room temperature. The remainder of the heads, containing eyes, vomeronasal organs (VNOs) and olfactory epithelia, were postfixed in Bouin’s fixative for 7 days at 4 °C. After postfixation, tissues were cut into small blocks with a razor blade as follows. The brains were bisected along the midline: one hemibrain was cut into coronal slices at the levels of the olfactory

bulb, accessory olfactory bulb, anterior olfactory nucleus, nucleus accumbens, anterior commissure, ventral hippocampal commissure, habenular nucleus, substantia nigra, superior colliculus and cerebellar flocculus, and the other hemibrain was cut into two parasagittal slices. Spinal cords were cut with vertebrae into axial slices at seven to eight levels. Eyes were cut with the surrounding orbital bony wall along the plane parallel to the optic nerve and perpendicular to the skull base. VNOs and olfactory epithelia were cut with the surrounding tissues into coronal slices. These slices and other tissues (inner ears, pituitary glands and trigeminal nerve bundles) were embedded in paraffin and cut into 8-␮m-thick sections with a sliding microtome (Yamato Kohki Industrial Co., Ltd., Saitama, Japan). We took special care to minimize the number of animals used and their suffering. All procedures were approved by the committee of our institute and animals were handled in accordance with the Guidelines for Animal Experiments at our institute, which complies with the Guide for the Care and Use of Laboratory Animals (National Academy Press, Washington, D.C., 1996).

Antibodies and peptides Antibody against LTA4 hydrolase (rabbit polyclonal) was prepared as described previously, and the specificity of the antibody was confirmed by immunoblot analysis and the preabsorption test (Ohishi et al., 1990). Antibody against calretinin (rabbit polyclonal, RB-9002) and recombinant mouse calretinin protein (RP-9308) was purchased from Lab Vision Corporation (Fremont, CA, USA).

Fig. 1. Validation of antibodies. Myelogenous cells in the bone marrow of the vertebral spine are specifically stained with the anti-LTA4 hydrolase antibody (a). Anti-calretinin antibody does not highlight any of the bone marrow cells (b). Serial sections of VNO (c– e) indicate that pre-absorption of anti-calretinin antibody with calretinin antigen completely abolishes the immunohistochemical staining (d), compared with staining using anti-calretinin antibody without pre-absorption (c). Pre-absorption of anti-LTA4 hydrolase antibody with calretinin antigen does not affect the immunoreactivity (e). Scale bars⫽50 ␮m (a, b), 200 ␮m (c– e). Sections were counterstained with hematoxylin in (a) and (b).

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

919

Immunohistochemistry

Double immunolabeling experiment

The sections on slides were deparaffinized and soaked in 0.2% Triton X-100 at room temperature for 10 min. For staining with anti-calretinin antibody, sections were heated to 95 °C in 10 mM citrate buffer, pH 6.0, for 10 min prior to treatment with Triton X-100 for antigen retrieval. Non-specific binding was blocked with 10% normal goat serum, and incubated with primary antibodies against LTA4 hydrolase (1:600) or calretinin (1:200) at 4 °C overnight. All antibodies were diluted in 10% normal goat serum. The sections were incubated with biotinylated secondary antibodies (Vector Laboratories, Inc., Burlingame, CA, USA) at 4 °C for 30 min and then incubated for 30 min at room temperature in avidin peroxidase and visualized using diaminobenzidine (DAB). Immunohistochemical controls to rule out the possibility of cross-reaction were obtained using the immunoabsorption test. The antibodies for LTA4 hydrolase (2.89⫻10⫺13 mol) or for calretinin (2.67⫻10⫺13 mol) were incubated with 10-fold molar excess of calretinin antigen (3.50⫻10⫺12 mol) at 37 °C for 2 h. The reaction mixture was centrifuged at 16,000⫻g for 30 min and the sections were incubated with the supernatant.

Double immunostaining for LTA4 hydrolase and calretinin was performed using fluorescence-conjugated secondary antibodies. Both of these antibodies were polyclonal of rabbit origin, and thus we followed the protocol as described elsewhere (Suzuki et al., 2002). The sections were deparaffinized and pretreated in the same way as described above. They were incubated with the primary antibodies against LTA4 hydrolase (1:600) at 4 °C overnight followed by the secondary antibody, Alexa FluorTM 488 goat anti-rabbit IgG (Molecular Probes, Eugene, OR, USA; 1:400) at room temperature for 30 min. The sections were coverslipped using a SlowFadeTM Antifade Kit (Molecular Probes) and green fluorescent images were obtained using confocal laser scanning microscopy (Fluoview, Olympus, Tokyo, Japan). Thereafter, the coverslips were removed and the sections were heated to 95 °C in 10 mM citrate buffer, pH 6.0, for 10 min to abolish any antigenicity of the primary antibody necessary to bind with the secondary antibody, as well as to retrieve calretinin antigen. After this heating step, the sections were incubated with the primary antibody against calretinin (1:200) followed by Alexa FluorTM 568 goat

Table 1. Tissue distribution of LTA4 hydrolase and calretinin in the mouse CNS and peripheral nervous system

Olfactory system Neurons in olfactory epithelium Neuroepithelial cells in the vomeronasal organ Olfactory glomerulus Tufted cells of the olfactory bulb Neurons in the granule layer of the olfactory bulb Accessory olfactory bulb Visual system Neurons in the inner margin of the INL Three bands in the inner plexiform layer of the retina Neurons in the ganglion cell layer of the retina Axons in the optic nerve fiber layer of the retina Optic nerve and tract Superior colliculus Vestibulocochlear system IHC of the cochlea OHC of the cochlea Spiral ganglion cells of the cochlea Cochlear nerve fibers Cochlear nucleus Sensory cells in the vestibular end-organs Vestibular nucleus Somato-sensory system Trigeminal fibers and ganglion cells Spinal trigeminal tract Lamina II of the spinal cord Others Cerebral cortex Triangular septal nucleus Medial habenular nucleus Molecular layer of the dentate gyrus Basolateral amygdaloid nucleus Thalamus Neurons and glomeruli in the granular layer of the lobules 9 and 10 of the cerebellar vermis Granule cells of the cerebellum other than lobules 9 and 10 of the vermis Brainstem reticular formation White matter of the spinal cord Neurons around the central canal of the spinal cord

LTA4 hydrolase

Calretinin

⫹ ⫹ ⫹, Periglomerular cells, neurites ⫹ ⫹ ⫹

⫹ ⫹ ⫹, Periglomerular cells, neurites ⫹ ⫹ ⫹

⫹ ⫹ ⫹ ⫹ ⫹ ⫹, Fibers

⫹ ⫹ ⫹ ⫹ ⫹ ⫹, Fibers

⫹ ⫹ Partially ⫹ ⫹ ⫹, Fibers Partially ⫹ ⫹, Fibers

⫹ ⫺ Partially ⫹ ⫹ ⫹, Fibers Partially ⫹ ⫹, Fibers

Rarely ⫹ ⫹ ⫹, Cell bodies and fibers

Rarely ⫹ ⫹ ⫹, Cell bodies and fibers

⫹, Scattered neurons ⫹ ⫹, Neuropil ⫹ ⫺ Partially ⫹ ⫹

⫹, Scattered neurons ⫹ ⫹, Neuropil ⫹ ⫹ Partially ⫹ ⫹

⫺

⫹

Partially ⫹ ⫹, Scattered fibers Partially ⫹

Partially ⫹ ⫹, Scattered fibers Partially ⫹

920

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

anti-rabbit IgG (Molecular Probes; 1:50) at room temperature for 30 min. Red fluorescent images of exactly the same regions stained for LTA4 hydrolase were obtained using confocal laser scanning microscopy. The images of LTA4 hydrolase immunostain and those of calretinin immunostain from the same field were merged using Photoshop software (version 6.0, Adobe Systems, San Jose, CA, USA). To quantify the degree of colocalization of LTA4 hydrolase and calretinin, we examined the proportion of cells that were double-immunopositive for LTA4 hydrolase and calretinin to cells that were immunopositive for LTA4 hydrolase in several selected areas. We cut two sections per area with being about 80 ␮m apart from each other from a representative mouse. Using coronal sections at the level of the ventral hippocampal commissure, nuclear profiles of neurons in the entire neocortical region (between the interhemispheric fissure and the rhinal fissure) that were immunopositive for LTA4 hydrolase, calretinin, or both were counted separately. In midsagittal sections of the cerebellar vermis, nuclear profiles of neurons that were immunopositive for LTA4 hydrolase, calretinin, or both were counted separately. In other regions, rectangular regions of interest (100⫻200 ␮m) were set and nuclear profiles of neurons that were immunopositive for LTA4 hydrolase, calretinin, or both were counted separately. Cell counts from two sections were summed and the percentages of

double-positive cells to LTA4 hydrolase-positive cells were calculated.

RESULTS Validation of antibodies and immunohistochemical staining Cells of the granulocyte-macrophage lineage in the vertebral bone marrow were specifically stained by the antiLTA4 hydrolase antibody (Fig. 1a), whereas anti-calretinin antibody did not stain any bone marrow cells (Fig. 1b). Pre-absorption of anti-calretinin antibody with calretinin antigen abolished the immunohistochemical staining (Fig. 1c and d), but pre-absorption of anti-LTA4 hydrolase antibody with calretinin antigen did not affect the immunoreactivity (Fig. 1e). Differences in the immunostaining pattern between anti-LTA4 hydrolase and anti-calretinin antibodies were also observed in bronchial epithelial cells and pancreatic islet cells. Bronchial epithelia were immunopositive for LTA4 hydrolase, as previously reported (Ohishi et al., 1990), but negative for calretinin (data not shown). In

Fig. 2. LTA4 hydrolase immunoreactivity in the olfactory system. (a) Cell bodies and apical dendritic processes of neurons in the olfactory epithelium are immunohistochemically stained by anti-LTA4 hydrolase antibody. Nerve bundles in the lamina propria (arrows) are also immunopositive for LTA4 hydrolase. (b) Neuroepithelium in the medial wall of the VNO is strongly immunolabeled with anti-LTA4 hydrolase antibody. Apical dendritic processes are clearly labeled (arrow). In contrast, ciliated columnar epithelium lining the lateral wall of VNO is LTA4 hydrolase-negative (asterisk). (c) In the olfactory bulb, olfactory glomeruli, in which olfactory nerve afferents make synaptic contacts with apical dendrites of the mitral and tufted cells, are immunopositive for LTA4 hydrolase. Many periglomerular cells, interneurons surrounding the glomeruli, are also immunopositive. (d) A portion of the tufted cells and neurons in the granule layer of the olfactory bulb is immunostained, although mitral cells are not stained by anti-LTA4 hydrolase antibody (arrows). (e) Afferent axons of the vomeronasal nerve in the accessory olfactory bulb are strongly immunopositive for LTA4 hydrolase. AOB, accessory olfactory bulb; vn, vomeronasal nerve. Scale bars⫽50 ␮m (a– d), 200 ␮m (e). No section was counterstained.

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

contrast, pancreatic islet cells were immunopositive for calretinin, consistent with a previous report (Redecker and Cetin, 1997), whereas they were negative for LTA4 hydrolase (data not shown). Distribution of LTA4 hydrolase-expressing neurons in the central and peripheral nervous system Immunohistochemical staining with anti-LTA4 hydrolase antibody of the mouse nervous tissue showed a widespread labeling of the neurons and nerve fibers, with preference to sensory-associated structures (Table 1). In the olfactory system, neuronal cell bodies and their apical dendritic processes intercalated in the olfactory epithelium were immunohistochemically stained by anti-LTA4 hydrolase antibody (Fig. 2a). Nerve bundles in the lamina propria were also immunopositive (Fig. 2a, arrows). Neuroepithelial cells lining the medial side of the VNO were strongly immunolabeled with anti-LTA4 hydrolase antibody

921

(Fig. 2b). In contrast, ciliated columnar cells lining the lateral wall of VNO did not show any immunoreactivity for LTA4 hydrolase (Fig. 2b, asterisk). In the olfactory bulb, periglomerular cells surrounding olfactory glomeruli, neurites in glomeruli, tufted cells in the mitral cell layer, and neurons in the granule layer were immunopositive for LTA4 hydrolase (Fig. 2c and d). Mitral cells were devoid of LTA4 hydrolase immunoreactivity (Fig. 2d, arrows). In the accessory olfactory bulb, afferent axons of the vomeronasal nerve from the VNO were strongly immunostained with anti-LTA4 hydrolase antibody (Fig. 2e). LTA4 hydrolase immunostaining of the retina showed specific labeling of neurons in the inner margin of the inner nuclear layer (INL), which most likely represent amacrine cells, and those in the ganglion cell layer, which include both displaced amacrine and ganglion cells (Perry, 1981) (Fig. 3a). Furthermore, three distinct bands in the inner plexiform layer, to which

Fig. 3. LTA4 hydrolase immunoreactivity in the visual system. (a) In the retina, neurons in the inner margin of the INL and the ganglion cell layer are immunopositive for LTA4 hydrolase. The former are most likely amacrine cells and the latter presumably include both ganglion cells and displaced amacrine cells. Three distinct horizontal bands or strata (arrows) were clearly immunolabeled in the inner plexiform layer, to which amacrine cells extend dendrites (see also Fig. 7e). The innermost optic nerve fiber layer, which consists of axons of ganglion cells, is also immunostained. (b– d) Nerve fibers in the optic nerve (b) and optic tract (c) are strongly immunostained by anti-LTA4 hydrolase antibody. These fibers mainly terminate in the superficial layer of the superior colliculus (d). GCL, ganglion cell layer of the retina; IPL, inner plexiform layer of the retina; ONL, outer nuclear layer of the retina; OPL, outer plexiform layer of the retina; P, pigment epithelium of the retina; R&C, layer of rods and cones of the retina; 2n, optic nerve. Scale bars⫽50 ␮m (a, c, d), 500 ␮m (b). Sections were counterstained with hematoxylin in (a), (b) and (d).

922

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

neurons in the innermost INL extend dendrites, were clearly immunopositive for LTA4 hydrolase (Fig. 3a, arrows). Axons of the retinal ganglion cells were also positive for LTA4 hydrolase all the way from the optic nerve fiber layer of the retina (Fig. 3a) via the optic nerve (Fig. 3b) and optic tract (Fig. 3c) to the superficial layer of the superior colliculus (Fig. 3d). In the cochlea, the inner hair cells (IHCs) were strongly immunostained by anti-LTA4 hydrolase antibody (Fig. 4a). The outer hair cells (OHCs) were also immunopositive for LTA4 hydrolase, but showed much weaker expression compared with the IHCs (Fig. 4a). Dendrites of the spiral ganglion cells that run in the lamina spiralis ossea and terminate in the IHCs were also stained (Fig. 4a, arrow). In the spiral ganglion, some ganglion cells were strongly immunopositive, some weakly positive, and others negative for LTA4 hydrolase (Fig. 4b). Cochlear nerve bundles in the cochlea were immunostained (Fig. 4b). In the vestibular organ, several sensory cells in the macula statica and ampulla ductus semicircularis (Fig. 4c, arrowheads) as

well as the nerve fibers in the underlying connective tissue (Fig. 4c, arrows) showed immunoreactivity for LTA4 hydrolase. In the cochlear and vestibular nuclei in the brainstem, afferent axons were selectively immunostained by antiLTA4 hydrolase antibody, whereas neurons in the nuclei showed only rare immunoreactivity (Fig. 4d and e). Only few nerve fibers in the trigeminal nerve bundles and a small portion of the trigeminal ganglion cells were immunohistochemically stained by anti-LTA4 hydrolase antibody (Fig. 5a). Descending axons in the spinal trigeminal tract were immunopositive for LTA4 hydrolase (Fig. 5b). Axons in the white matter of the spinal cord showed scattered immunopositivity for LTA4 hydrolase, while staining was relatively scarce in the posterior column (Fig. 5c). Most prominent immunolabeling in the spinal cord was observed in the posterior horn. Neurons and surrounding neuropil in the lamina II strongly expressed LTA4 hydrolase (Fig. 5d). Scattered neurons around the central canal were also immunopositive (Fig. 5e). A few neurons in the anterior horn were immunostained by anti-LTA4 hydrolase

Fig. 4. LTA4 hydrolase immunoreactivity in the vestibulocochlear system. (a) An IHC is strongly immunostained by anti-LTA4 hydrolase antibody. Nerve fibers that arise from spiral ganglion neurons and terminate at the IHC are also stained (arrow). OHCs are also weakly immunopositive for LTA4 hydrolase. (b) Bipolar neurons in the spiral ganglion of the cochlea and cochlear nerve bundles are immunolabeled by anti-LTA4 hydrolase antibody. Among the spiral ganglion neurons, some are strongly positive, others are weakly positive, and still others are negative for LTA4 hydrolase. (c) Some of the sensory cells (arrowheads) in the crista ampullaris and nerve fibers arising from vestibular ganglion cells in the underlying connective tissue (arrows) are immunopositive for LTA4 hydrolase. (d) Cochlear nerve fibers in the posteroventral cochlear nucleus are positive for LTA4 hydrolase. Nerve fibers in the vestibular root of vestibulocochlear nerve, which pass along the medial side of the cochlear nucleus, are also immunostained. (e) Vestibular nerve fibers in the medial vestibular nucleus are immunostained by anti-LTA4 hydrolase antibody. SG, spiral cochlear ganglion; VCP, posterior ventral cochlear nucleus; 8n, cochlear nerve; 8vn, vestibular root of vestibulocochlear nerve. Scale bars⫽50 ␮m. Sections were counterstained with hematoxylin in (a), (b), (c) and (e).

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

923

Fig. 5. LTA4 hydrolase immunoreactivity in the somatosensory-associated structures. (a) A small portion of the trigeminal nerve fibers and a few ganglion cells are immunostained by anti-LTA4 hydrolase antibody. (b) Descending fibers of the spinal trigeminal tract in the pons are immunolabeled. (c) Strong immunoreactivity for LTA4 hydrolase is demonstrated in the posterior horn of the spinal cord. Longitudinal fibers in the white matter, as well as a few neurons in the anterior horn and central gray matter of the spinal cord, are also immunopositive. Posterior column contains relatively scarce immunopositive fibers for LTA4 hydrolase. (d) High magnification of the inset in (c). Neurons and surrounding neuropil in the lamina II are prominently immunostained by anti-LTA4 hydrolase antibody. (e) High magnification of the central part of the spinal cord (inset in (c)). Several neurons around the central canal are immunopositive for LTA4 hydrolase. CC, central canal; sp5, spinal trigeminal tract. Scale bars⫽100 ␮m (a– c), 50 ␮m (d, e). Sections were counterstained with hematoxylin in (a), (c), (d) and (e).

antibody, but most of the large anterior horn cells were immunonegative. Bipolar neurons in the dorsal root ganglia exhibited very rare immunoreactivity for LTA4 hydrolase (data not shown). Among the structures not directly associated with the sensory system, the following showed immunoreactivity for LTA4 hydrolase: scattered neurons in the entire cerebral cortex (Fig. 6a), neurons in the septal nuclei (Fig. 6b), a portion of the thalamic neurons, the neuropil in the medial habenular nucleus (Fig. 6c), molecular layer of the dentate gyrus (Fig. 6d), neurons in the granular layer and cerebellar glomeruli in the lobules 9 (the uvula) and 10 (the nodule) of the cerebellar vermis (Fig. 6e), and some of the neurons in the brainstem reticular formation. In the cerebellum, LTA4 hydrolase immunoreactivity in the granular layer was confined to the lobules of the vermis that have close connection to the vestibular nuclei. In contrast, hypothalamic neurons and the pituitary gland did not show any significant immunoreactivity for LTA4 hydrolase (data not shown), sites where LTC4 synthase was selectively expressed (Shimada et al., 2005).

Colocalization of LTA4 hydrolase and calretinin in the nervous system Since the distribution of the LTA4 hydrolase-immunopositive structures described above was quite similar to the reported distribution of calretinin, we then studied whether these two proteins colocalize in the same neurons, using the double immunofluorescence labeling technique. For verification of elimination of the first labeling, we confirmed that the green fluorescence of Alexa FluorTM 488 was abolished after 10-minutes of heating in citrate buffer (data not shown). Application of Alexa FluorTM 568-conjugated secondary antibody without any primary antibody after the heating step yielded no red fluorescence, verifying that the heating step completely abolished the antigenicity of the primary antibody (data not shown). As shown in Fig. 7, most of the LTA4 hydrolasepositive structures described above also showed immunoreactivity for calretinin, and the distribution of these two proteins considerably overlapped in the various central and peripheral nervous tissues (Tables 1, 2). However,

924

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

Fig. 6. LTA4 hydrolase immunoreactivity in the regions not associated with the sensory system. (a) There are scattered LTA4-hydrolase-positive neurons in the somatosensory cortex. (b) Neurons in the triangular septal nucleus are immunostained by anti-LTA4 hydrolase antibody. (c) Neuropil in the medial habenular nucleus is immunopositive for LTA4 hydrolase. Neurons in the nucleus do not show immunoreactivity. (d) Molecular layer of the dentate gyrus is faintly immunolabeled by anti-LTA4 hydrolase antibody. (e) Immunohistochemical staining of a sagittal section of the cerebellar vermis shows LTA4-hydrolase-immunoreactivity in several granule cells and cerebellar glomeruli (arrows) of the lobule 9 (the uvula). Purkinje cells are completely devoid of immunoreactivity (arrowheads). D3V, dorsal third ventricle; MHb, medial habenular nucleus; SFO, subfornical organ; TS, triangular septal nucleus; vhc, ventral hippocampal commissure. Scale bars⫽100 ␮m (a– d), 50 ␮m (e). Sections were counterstained with hematoxylin in (c), (d) and (e).

there were some differences in the intensity of immunoreactivity for these two antibodies in neurites in the olfactory glomeruli in that these neurites showed more intense immunoreactivity for LTA4 hydrolase than for calretinin (Fig. 7c). The distribution within the cells of each protein seemed to be different in several parts. LTA4 hydrolase was mainly expressed in the cytoplasm, whereas calretinin was expressed in the entire cell (Fig. 7d, g). Double staining of the cochlear sections was not successful because of the detachment of sections from slide glasses during the heating step. Instead, immunostaining of the adjacent sections showed that LTA4 hydrolase and calretinin have similar distribution patterns in the cochlea, except for the negative staining of the OHCs with anti-calretinin antibody (Table 1). Immunostaining of adjacent sections of the brain exhibited that neurons in the basolateral amygdaloid nucleus and most of the granule cells of the cerebellum were immunopositive for calretinin, whereas they were negative for LTA4 hydrolase (Table 1).

DISCUSSION In the present study, we demonstrated that LTA4 hydrolase is constitutively expressed in the mouse nervous system, and preferentially distributed in the sensory-associated structures. This result is in sharp contrast with the specific localization of LTC4 synthase in the vasopressinergic neurons that we recently reported (Shimada et al., 2005). This difference in distribution between LTA4 hydrolase and

LTC4 synthase suggests that these two enzymes have different functions in the nervous system, although they share LTA4 as a common substrate. Many studies have demonstrated that eicosanoids are present in the CNS, and have suggested their function in the modulation of synaptic transmission (Shimizu and Wolfe, 1990). Feinmark et al. (2003) have recently demonstrated that 12(S)-hydroperoxyeicosa-5Z,8Z,10E,14Ztetraenoic acid, a 12-lipoxygenase metabolite, can mediate long-term depression at the hippocampal CA3–CA1 synapses. Arachidonic acid metabolites are membranepermeable and can act as autocrine and paracrine messengers, which are released from the cells and bind to the membrane receptor or intracellular receptor of neighboring cells (Shimizu and Wolfe, 1990). This biochemical characteristic of eicosanoids makes them an attractive candidate for the messenger that mediates transsynaptic modulation of neuronal activity. LTB4 may also act as a modulator of neurotransmission, like other eicosanoids. Two types of receptors have been identified for LTB4, BLT1 (Yokomizo et al., 1997) and BLT2 (Yokomizo et al., 2000). BLT1 is a high-affinity receptor specific for LTB4 and expressed primarily in leukocytes, but much lower expression in the brain (Yokomizo et al., 1997; Tager and Luster, 2003; Iizuka et al., 2005). BLT2 is a low-affinity receptor that also binds other eicosanoids and shows more ubiquitous expression (Yokomizo et al., 2000). BLT1 and BLT2 are both G protein-coupled receptors and activation of them by

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

925

Fig. 7. Representative confocal laser scanning micrographs showing colocalization of LTA4 hydrolase with calretinin in the nervous system. Sections of mouse central and peripheral nervous tissues were sequentially labeled with anti-LTA4 hydrolase and anti-calretinin antibodies and fluorescence photomicrographs were taken using a confocal laser scanning microscope. Merged images clearly show considerable overlap of these two proteins. (a– c) Periglomerular cells and neurites in the olfactory glomeruli. See Fig. 2c for corresponding immunostaining visualized with peroxidase–DAB system. Note more intense labeling of nerve fibers in the olfactory glomerulus with anti-LTA4 hydrolase antibody, compared with anti-calretinin antibody. (d) Neuroepithelium of the VNO. LTA4 hydrolase is expressed mainly in the cytoplasm, whereas calretinin is expressed both in the nuclei and the cytoplasm. See Fig. 2b for corresponding immunostaining visualized with peroxidase–DAB system. (e) Retina. There is a neuron in the inner margin of the INL extending dendrites to the three distinct bands in the inner plexiform layer (arrow). See Fig. 3a for corresponding immunostaining visualized with peroxidase–DAB system. (f) Spinal trigeminal tract. See Fig. 5b for corresponding immunostaining visualized with peroxidase–DAB system. (g) Posterior horn of the spinal cord. LTA4 hydrolase is expressed mainly in the cytoplasm, whereas calretinin is expressed in the entire cell. See Fig. 5d for corresponding immunostaining visualized with peroxidase–DAB system. (h) Somatosensory cortex. See Fig. 6a for corresponding immunostaining visualized with peroxidase–DAB system. (i) Lobule 10 of the cerebellar vermis. An arrowhead indicates double-labeled cerebellar glomerulus corresponding to the structure indicated by arrows in Fig. 6e. Scale bars⫽50 ␮m.

LTB4 results in calcium mobilization and inhibition of adenylyl cyclase (Yokomizo et al., 1997, 2000). The distribution of these LTB4 receptors in the nervous system merits investigation. In addition to its specific activity to convert LTA4 to LTB4, the enzyme possesses aminopeptidase activity

(Haeggström et al., 1990; Minami et al., 1990). In vitro studies searching for naturally occurring substrates for aminopeptidase activity of LTA4 hydrolase demonstrated that the opioid peptide enkephalin (Griffin et al., 1992) and dynorphin (Nissen et al., 1995) can be cleaved by this enzyme. It is noteworthy that LTA4 hydrolase localizes in

Table 2. Number of cells immunopositive for LTA4 hydrolase, calretinin, or both in selected regions of the CNS and peripheral nervous system Region

LTA4 hydrolase-positive cells

Calretinin-positive cells

Double-positive cells

Double-positive cells/LTA4 hydrolase-positive cells (%)

Cerebral cortex. frontal Periglomerular cells, olfactory bulb Triangular septal nucleus Neurons in the granular layer, lobules 9 and 10 of the cerebellar vermis Retina Neurons in lamina II. spinal cord Sensory neuron, olfactory epithelium Neuroepithelial cells, VNO

130 66 112 53

125 76 112 52

113 66 112 52

86.9 100 100 98.1

58 38 32 67

58 47 32 67

58 38 32 67

100 100 100 100

926

Y. Chiba et al. / Neuroscience 141 (2006) 917–927

lamina II of the spinal cord and spinal trigeminal tract, both of which are involved in nociception. Furthermore, the posterior horn of the spinal cord is rich in enkephalinimmunopositive nerve fibers (Hunt et al., 1981). However, lack of evidence that the aminopeptidase activity of LTA4 hydrolase plays some role in vivo argues against the hypothesis that this second activity might be the main function of LTA4 hydrolase in the nervous system. As reported here, LTA4 hydrolase was widely distributed in the nervous system. At first we were unfamiliar with the significance of the pattern of distribution. To find a clue to understand the significance, we searched literature for proteins that might show similar distribution in the nervous system. Surprisingly, the reported distribution of calretinin in the olfactory chemoreceptor neurons (Kishimoto et al., 1993), retina (Haverkamp and Wassle, 2000), inner ear (Dechesne et al., 1994) and spinal cord (Ren et al., 1993) resembles that of LTA4 hydrolase. Calretinin is a member of the family of calcium-binding proteins and is mainly expressed in the nervous system (Rogers, 1987). We demonstrated by double immunofluorescence staining that the distribution of these two proteins shows considerable overlap in the central and peripheral nervous tissues. However, the intracellular distribution of LTA4 hydrolase and calretinin was not completely overlapped in several parts (Fig. 7d and g): LTA4 hydrolase mainly resided in the cytoplasm, whereas calretinin was expressed both in the cytoplasm and nuclei. This difference in the intracellular distribution is consistent with the previous reports (Arai et al., 1993; Haeggström, 2004). The distribution of LTA4 hydrolase and calretinin in tissues other than the nervous system did not overlap, indicating that the colocalization of them is restricted to the nervous system. This suggests that LTA4 hydrolase and calretinin probably have some neuron-specific functional correlation. Several lines of evidence suggest a neuroprotective function for calretinin in neurodegenerative diseases and aging, through its capacity to buffer intracellular calcium (Lukas and Jones, 1994; Fahandejsaadi et al., 2004; Idrizbegovic et al., 2004). Another proposed function of calretinin includes the regulation of protein phosphorylation in the brain (Yamaguchi et al., 1991). How LTA4 hydrolase and calretinin functionally link remains to be elucidated. One possible explanation is that LTB4 modulates neural activity through intracellular calcium recruitment via activation of BLT1 or BLT2, while calretinin protects neurons against calcium-induced dysfunction and cell death by buffering increased calcium concentration. Another possibility is that calcium mobilized by LTB4 binds calretinin to transduce signals through regulation of protein phosphorylation. A previous report showed that LTA4 hydrolase activity was detected in the pituitary gland and hypothalamus, as well as olfactory bulb and cerebral cortex in the guinea-pig brain (Shimizu et al., 1987). An in vitro study also demonstrated that LTB4 was produced from cultured rat anterior pituitary cells by stimulation with gonadotropin-releasing hormone and LTB4 stimulated luteinizing hormone release from these cells (Kiesel et al., 1991). These reports suggest that LTB4, as well as LTC4, may have a role in the

neuroendocrine system. However, we did not find any significant immunoreactivity in the hypothalamus and pituitary gland. The previous report hypothesized LTB4 production from microvessels and neighboring mast cells in the brain (Shimizu et al., 1987), both of which were immunonegative in the present study. The reason for this discrepancy between the previous study and our finding is unclear, but species difference might be one explanation. Alternatively, it is possible that LTA4 hydrolase may be induced by certain stimuli in neurons or vascular endothelial cells. Analysis of neurological phenotypes of LTA4 hydrolase-deficient animals and human cases would be useful for the elucidation of the neural functions of LTA4 hydrolase and LTB4. However, there has been no case report of congenital LTA4 hydrolase deficiency in humans. Acquired reduction in LTB4 level was reported in malnutrition states such as essential fatty acid deficiency (Cleland et al., 1994) and kwashiorkor (Mayatepek et al., 1993), but their neurological manifestation appeared non-specific and multifactorial. Although the biological and pathological roles of LTB4 in the brain remain to be fully elucidated, it is noteworthy that disturbance of LTB4 metabolism induces mental retardation and spasticity in addition to the various peripheral phenotypes (Sjögren-Larsson syndrome) (Willemsen et al., 2001). The present findings provide a pivotal clue to understand the role of LTA4 hydrolase/aminopeptidase, and LTB4 in the pathophysiology of the brain. Acknowledgments—This study was supported by Grants-in-Aid from the Ministry of Education, Culture, Science, Sports and Technology of Japan.

REFERENCES Arai R, Jacobowitz DM, Deura S (1993) Ultrastructural localization of calretinin immunoreactivity in lobule V of the rat cerebellum. Brain Res 613:300 –304. Bailie MB, Standiford TJ, Laichalk LL, Coffey MJ, Strieter R, PetersGolden M (1996) Leukotriene-deficient mice manifest enhanced lethality from Klebsiella pneumonia in association with decreased alveolar macrophage phagocytic and bactericidal activities. J Immunol 157:5221–5224. Barone FC, Schmidt DB, Hillegass LM, Price WJ, White RF, Feuerstein GZ, Clark RK, Lee EV, Griswold DE, Sarau HM (1992) Reperfusion increases neutrophils and leukotriene B4 receptor binding in rat focal ischemia. Stroke 23:1337–1347; discussion 1347–1348. Byrum RS, Goulet JL, Snouwaert JN, Griffiths RJ, Koller BH (1999) Determination of the contribution of cysteinyl leukotrienes and leukotriene B4 in acute inflammatory responses using 5-lipoxygenase- and leukotriene A4 hydrolase-deficient mice. J Immunol 163:6810 – 6819. Chen N, Restivo A, Reiss CS (2001) Leukotrienes play protective roles early during experimental VSV encephalitis. J Neuroimmunol 120: 94 –102. Cleland LG, James MJ, Proudman SM, Neumann MA, Gibson RA (1994) Inhibition of human neutrophil leukotriene B4 synthesis in essential fatty acid deficiency: role of leukotriene A hydrolase. Lipids 29:151–155. Dechesne CJ, Rabejac D, Desmadryl G (1994) Development of calretinin immunoreactivity in the mouse inner ear. J Comp Neurol 346:517–529.

Y. Chiba et al. / Neuroscience 141 (2006) 917–927 Fahandejsaadi A, Leung E, Rahaii R, Bu J, Geula C (2004) CalbindinD28K, parvalbumin and calretinin in primate lower motor neurons. Neuroreport 15:443– 448. Feinmark SJ, Begum R, Tsvetkov E, Goussakov I, Funk CD, Siegelbaum SA, Bolshakov VY (2003) 12-Lipoxygenase metabolites of arachidonic acid mediate metabotropic glutamate receptordependent long-term depression at hippocampal CA3-CA1 synapses. J Neurosci 23:11427–11435. Funk CD (2001) Prostaglandins and leukotrienes: advances in eicosanoid biology. Science 294:1871–1875. Gladue RP, Carroll LA, Milici AJ, Scampoli DN, Stukenbrok HA, Pettipher ER, Salter ED, Contillo L, Showell HJ (1996) Inhibition of leukotriene B4-receptor interaction suppresses eosinophil infiltration and disease pathology in a murine model of experimental allergic encephalomyelitis. J Exp Med 183:1893–1898. Griffin KJ, Gierse J, Krivi G, Fitzpatrick FA (1992) Opioid peptides are substrates for the bifunctional enzyme LTA4 hydrolase/aminopeptidase. Prostaglandins 44:251–257. Haeggström JZ (2004) Leukotriene A4 hydrolase/aminopeptidase, the gatekeeper of chemotactic leukotriene B4 biosynthesis. J Biol Chem 279:50639 –50642. Haeggström JZ, Wetterholm A, Vallee BL, Samuelsson B (1990) Leukotriene A4 hydrolase: an epoxide hydrolase with peptidase activity. Biochem Biophys Res Commun 173:431– 437. Haverkamp S, Wassle H (2000) Immunocytochemical analysis of the mouse retina. J Comp Neurol 424:1–23. Hunt SP, Kelly JS, Emson PC, Kimmel JR, Miller RJ, Wu JY (1981) An immunohistochemical study of neuronal populations containing neuropeptides or gamma-aminobutyrate within the superficial layers of the rat dorsal horn. Neuroscience 6:1883–1898. Idrizbegovic E, Bogdanovic N, Willott JF, Canlon B (2004) Age-related increases in calcium-binding protein immunoreactivity in the cochlear nucleus of hearing impaired C57BL/6J mice. Neurobiol Aging 25:1085–1093. Iizuka Y, Yokomizo T, Terawaki K, Komine M, Tamaki K, Shimizu T (2005) Characterization of a mouse second leukotriene B4 receptor, mBLT2: BLT2-dependent ERK activation and cell migration of primary mouse keratinocytes. J Biol Chem 280:24816 –24823. Izumi T, Shimizu T, Seyama Y, Ohishi N, Takaku F (1986) Tissue distribution of leukotriene A4 hydrolase activity in guinea pig. Biochem Biophys Res Commun 135:139 –145. Jala VR, Haribabu B (2004) Leukotrienes and atherosclerosis: new roles for old mediators. Trends Immunol 25:315–322. Kiesel L, Przylipiak AF, Habenicht AJ, Przylipiak MS, Runnebaum B (1991) Production of leukotrienes in gonadotropin-releasing hormone-stimulated pituitary cells: potential role in luteinizing hormone release. Proc Natl Acad Sci U S A 88:8801– 8805. Kishimoto J, Keverne EB, Emson PC (1993) Calretinin, calbindin-D28k and parvalbumin-like immunoreactivity in mouse chemoreceptor neurons. Brain Res 610:325–329. Lukas W, Jones KA (1994) Cortical neurons containing calretinin are selectively resistant to calcium overload and excitotoxicity in vitro. Neuroscience 61:307–316. Mayatepek E (2000) Leukotriene C4 synthesis deficiency: a member of a probably underdiagnosed new group of neurometabolic diseases. Eur J Pediatr 159:811– 818. Mayatepek E, Becker K, Gana L, Hoffmann GF, Leichsenring M (1993) Leukotrienes in the pathophysiology of kwashiorkor. Lancet 342: 958 –960. McCann SM, Licinio J, Wong ML, Yu WH, Karanth S, Rettorri V (1998) The nitric oxide hypothesis of aging. Exp Gerontol 33:813– 826. Minami M, Mutoh H, Ohishi N, Honda Z, Bito H, Shimizu T (1995) Amino-acid sequence and tissue distribution of guinea-pig leukotriene A4 hydrolase. Gene 161:249 –251.

927

Minami M, Ohishi N, Mutoh H, Izumi T, Bito H, Wada H, Seyama Y, Toh H, Shimizu T (1990) Leukotriene A4 hydrolase is a zinccontaining aminopeptidase. Biochem Biophys Res Commun 173: 620 – 626. Nissen JB, Iversen L, Kragballe K (1995) Characterization of the aminopeptidase activity of epidermal leukotriene A4 hydrolase against the opioid dynorphin fragment 1–7. Br J Dermatol 133: 742–749. Ohishi N, Minami M, Kobayashi J, Seyama Y, Hata J, Yotsumoto H, Takaku F, Shimizu T (1990) Immunological quantitation and immunohistochemical localization of leukotriene A4 hydrolase in guinea pig tissues. J Biol Chem 265:7520 –7525. Perry VH (1981) Evidence for an amacrine cell system in the ganglion cell layer of the rat retina. Neuroscience 6:931–944. Radmark O, Shimizu T, Jornvall H, Samuelsson B (1984) Leukotriene A4 hydrolase in human leukocytes. Purification and properties. J Biol Chem 259:12339 –12345. Redecker P, Cetin Y (1997) Rodent pancreatic islet cells contain the calcium-binding proteins calcineurin and calretinin. Histochem Cell Biol 108:133–139. Ren K, Ruda MA, Jacobowitz DM (1993) Immunohistochemical localization of calretinin in the dorsal root ganglion and spinal cord of the rat. Brain Res Bull 31:13–22. Rogers JH (1987) Calretinin: a gene for a novel calcium-binding protein expressed principally in neurons. J Cell Biol 105:1343–1353. Samuelsson B (1983) Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation. Science 220:568 –575. Shimada A, Satoh M, Chiba Y, Saitoh Y, Kawamura N, Keino H, Hosokawa M, Shimizu T (2005) Highly selective localization of leukotriene C4 synthase in hypothalamic and extrahypothalamic vasopressin systems of mouse brain. Neuroscience 131:683– 689. Shimizu T, Izumi T, Honda Z, Seyama Y, Kurachi Y, Sugimoto T (1990) Biosynthesis and functions of leukotriene C4. Adv Prostaglandin Thromboxane Leukot Res 20:46 –53. Shimizu T, Takusagawa Y, Izumi T, Ohishi N, Seyama Y (1987) Enzymic synthesis of leukotriene B4 in guinea pig brain. J Neurochem 48:1541–1546. Shimizu T, Wolfe LS (1990) Arachidonic acid cascade and signal transduction. J Neurochem 55:1–15. Suzuki T, Kato K, Ohara S, Noguchi K, Sekine H, Nagura H, Shimosegawa T (2002) Localization of antigen-presenting cells in Helicobacter pylori-infected gastric mucosa. Pathol Int 52:265–271. Tager AM, Luster AD (2003) BLT1 and BLT2: the leukotriene B4 receptors. Prostaglandins Leukot Essent Fatty Acids 69:123–134. Uz T, Pesold C, Longone P, Manev H (1998) Aging-associated upregulation of neuronal 5-lipoxygenase expression: putative role in neuronal vulnerability. FASEB J 12:439 – 449. Willemsen MA, IJlst L, Steijlen PM, Rotteveel JJ, de Jong JG, van Domburg PH, Mayatepek E, Gabreels FJ, Wanders RJ (2001) Clinical, biochemical and molecular genetic characteristics of 19 patients with the Sjögren-Larsson syndrome. Brain 124:1426–1437. Xu JA, Hsu CY, Liu TH, Hogan EL, Perot PL Jr, Tai HH (1990) Leukotriene B4 release and polymorphonuclear cell infiltration in spinal cord injury. J Neurochem 55:907–912. Yamaguchi T, Winsky L, Jacobowitz DM (1991) Calretinin, a neuronal calcium binding protein, inhibits phosphorylation of a 39 kDa synaptic membrane protein from rat brain cerebral cortex. Neurosci Lett 131:79 – 82. Yokomizo T, Izumi T, Chang K, Takuwa Y, Shimizu T (1997) A G-protein-coupled receptor for leukotriene B4 that mediates chemotaxis. Nature 387:620 – 624. Yokomizo T, Kato K, Terawaki K, Izumi T, Shimizu T (2000) A second leukotriene B4 receptor, BLT2. A new therapeutic target in inflammation and immunological disorders. J Exp Med 192:421– 432.

(Accepted 6 April 2006) (Available online 22 May 2006)