The crystal structure of mouse nerve growth factor (NGF) complexed with lysophosphatidylinositol was solved and the structural determinants of NGF for LysoPI molecule recognition were identified. The influence of LysoPI in the interactions between NGF and its two receptors is modelled.
Keywords: nerve growth factor, lysophosphatidylinositol, LysoPI, mNGF–LysoPI complex, LysoPS
Abstract
Nerve growth factor (NGF) is an important protein that is involved in a variety of physiological processes in cell survival, differentiation, proliferation and maintenance. The previously reported crystal structure of mouse NGF (mNGF) in complex with lysophosphatidylserine (LysoPS) showed that mNGF can bind LysoPS at its dimeric interface. To expand the understanding of the structural basis for specific lipid recognition by NGF, the crystal structure of mNGF complexed with lysophosphatidylinositol (13:0 LysoPI) was solved. Interestingly, in addition to Lys88, which interacts with the head glycerol group and the phosphate group of LysoPI, as seen in the mNGF–LysoPS structure, two additional residues, Tyr52 and Arg50, were found to assist in lipid binding by forming hydrogen bonds to the inositol moiety of the LysoPI molecule. The results suggest a specific recognition mechanism of inositol group-containing lipids by NGF, which may help in the design of bioactive compounds that can be delivered by NGF.
1. Introduction
Nerve growth factor (NGF) belongs to the neurotrophin (NT) family, which also includes neurotrophin-3 (NT3), neurotrophin-4/5 (NT4/5) and brain-derived neurotrophic factor (BDNF) (Soligo et al., 2013 ▸; Davies, 1994 ▸; Park & Poo, 2013 ▸). Neurotrophin family members share ∼50% sequence identity as well as highly conserved three-dimensional structural features. Neurotrophins function as noncovalent homodimers that modulate nerve-cell survival, maintenance, development and death (Huang & Reichardt, 2001 ▸; Heumann, 1994 ▸) by binding and activating two types of cell-surface receptors (Niewiadomska & Małecki, 1999 ▸; Friedman & Greene, 1999 ▸; Kaplan & Miller, 1997 ▸): tyrosine kinase receptors (Trks; Barbacid, 1994 ▸; Klein et al., 1991 ▸; Kaplan et al., 1991 ▸), which are specific for given neurotrophins such as TrkA for NGF, and the common neurotrophin receptor p75 receptor, which belongs to the death-promoting tumour necrosis factor family (Dechant & Barde, 1997 ▸; Yamashita et al., 2005 ▸).
Although NGF is typically named for its effects on neuronal cell growth and differentiation (Vinores & Guroff, 1980 ▸; Furukawa & Furukawa, 1991 ▸), it also engages in physiological events beyond neuronal systems, such as in the immune system and endocrine system (Tometten et al., 2005 ▸; Vega et al., 2003 ▸). NGFs from mice (mNGF) and snakes (sNGF) show similar neurotrophic effects but different immunological activities (Angeletti, 1970 ▸; Banks et al., 1973 ▸). In vitro addition of both mNGF and LysoPS (Fig. 1 ▸ b) to rat mast cells results in histamine release, while cobra NGF (cNGF) has a prominent histamine-release effect on rat mast cells in the absence of LysoPS. Our previous crystallographic work on cNGF revealed that cNGF naturally binds a two-tailed diacylglycerol (DAG)-like lipid molecule at its dimerization interface, with the longer tail buried in an inner tunnel (the A pocket) and the shorter tail located in a surface groove (the B pocket) at the dimeric seam (Tong et al., 2012 ▸). We subsequently solved the crystal structure of mNGF in complex with LysoPS. The LysoPS tail resides in the A pocket and its head glycerol group interacts with Lys88. From these observations, together with functional analyses, it is suggested that the physical interaction of LysoPS and NGF is correlated with the immunological activity of NGF. We also found that some other bioactive lipids, such as PIP2 and phosphatidic acid (PA), bind to NGF with high affinity (Tong et al., 2012 ▸). Compared with LysoPS, the head group of PIP2 contains an additional inositol moiety, which is a key component of many essential lipid messengers.
Figure 1.
(a) The chemical structure of LysoPI. (b) The chemical structure of LysoPS. (c–f) SPR analysis of LysoPI binding to mNGF at pH 4.0 (c), 5.0 (d), 6.0 (e) and 7.0 (f). The equilibrium dissociation constant, K d, as measured by SPR is given.
To investigate how inositol moiety-containing lipids are recognized by NGF, we attempted to crystallize mNGF in complex with two kinds of inositol-containing lipid, PIP2 and lysophosphatidylinositol (LysoPI; Fig. 1 ▸ a), which is an important lysophospholipid mediator that is able to activate signalling cascades relevant to cell proliferation, migration, survival and tumorigenesis (Grzelczyk & Gendaszewska-Darmach, 2013 ▸; Piñeiro & Falasca, 2012 ▸). Finally, we successfully obtained better diffraction-quality crystals of the mNGF–LysoPI complex.
Here, we report the crystal structure of mNGF in complex with LysoPI at 2.6 Å resolution. Our structure reveals that in addition to Lys88, which interacts with the hydroxyl group of the head glycerol group, as observed in the mNGF–LysoPS structure, the inositol moiety of the LysoPI molecule interacts with two additional binding residues, Tyr52 and Arg50. We speculate that this could be a common NGF-interacting mode shared by inositol-containing lipids such as LysoPI and PIP2.
2. Materials and methods
2.1. Macromolecule production
Nerve growth factor purified from murine submaxillary gland (mNGF) was purchased from AbD Serotec, Oxford, England. The amino-acid sequence of mNGF is shown in Table 1 ▸.
Table 1. Macromolecule-production information.
| Source organism | Mus musculus |
| Complete amino-acid sequence | SSTHPVFHMGEFSVCDSVSVWVGDKTTATDIKGKEVTVLAEVNINNSVFRQYFFETKCRASNPVESGCRGIDSKHWNSYCTTTHTFVKALTTDEKQAAWRFIRIDTACVCVLSRKATRRG |
2.2. Crystallization
mNGF was dissolved in 50 mM sodium phosphate buffer pH 6.5 to a concentration of 8 mg ml−1 and the lipids LysoPI and PIP2 (purchased from Avanti Polar Lipids) were dissolved at 1 mg ml−1 in deionized water. Initial crystallization screening of the mNGF–lipid complex was performed by the sitting-drop vapour-diffusion method at 16°C, mixing equal volumes of the mNGF and lipid solutions with the reservoir solution in a 5:1 volume ratio. We successfully obtained crystals of the mNGF–LysoPI complex using a reservoir solution consisting of 0.1 M imidazole pH 6.5, 1.0 M sodium acetate within three weeks at 16°C. The diffraction power of the crystals was improved after optimization by using the hanging-drop vapour-diffusion method and changing the buffer pH and the sodium acetate concentration. All crystals were incubated in reservoir solution containing 1.6 M sodium acetate as the cryoprotectant and dehydration reagent for 10 min before flash-cooling in liquid nitrogen. The final optimized crystallization condition is presented in Table 2 ▸.
Table 2. Crystallization.
| Method | Hanging-drop vapour diffusion |
| Plate type | 16-well crystallization plates |
| Temperature (K) | 289 |
| Protein concentration (mgml1) | 8 |
| Buffer composition of protein solution | 50mM sodium phosphate pH 6.5 |
| Composition of reservoir solution | 0.1M imidazole pH 6.6, 0.9M sodium acetate |
| Volume and ratio of drop | 1l:1l |
| Volume of reservoir (l) | 500 |
2.3. Data collection and processing
The diffraction data were collected on the BL17U beamline at the Shanghai Synchrotron Radiation Facility (SSRF), Shanghai, People’s Republic of China and were indexed, integrated and scaled using the HKL-2000 software suite (Otwinowski & Minor, 1997 ▸). The data-collection and processing statistics are presented in Table 3 ▸.
Table 3. Data collection and processing.
Values in parentheses are for the outer shell.
| Wavelength () | 0.97876 |
| Temperature (K) | 100 |
| Detector | ADSC Quantum 315 CCD |
| Crystal-to-detector distance (mm) | 350 |
| Rotation range per image () | 1.0 |
| Total rotation range () | 240 |
| Exposure time per image (s) | 1.0 |
| Space group | R32:H |
| a, b, c () | 96.69, 96.69, 164.34 |
| , , () | 90.0, 90.0, 120.0 |
| Mosaicity () | 0.77 |
| Resolution range () | 50.02.6 (2.642.60) |
| Total No. of reflections | 50064 |
| No. of unique reflections | 8340 (363) |
| Completeness (%) | 99.7 (100.0) |
| Multiplicity | 5.4 (5.5) |
| I/(I) | 13.58 (4.9) |
| R merge | 0.075 (0.534) |
| Overall B factor from Wilson plot (2) | 35.05 |
2.4. Structure solution and refinement
The structure was solved by molecular replacement with PHENIX (Adams et al., 2010 ▸; Terwilliger et al., 2009 ▸) using one mNGF dimer from the mNGF–LysoPS complex structure (PDB entry 4eax; Tong et al., 2012 ▸) as the search model. The LysoPI molecule was modelled into the apparent nonprotein density at the dimerization interface of mNGF. Structure refinement was performed by alternate cycles of manual model rebuilding with Coot (Emsley et al., 2010 ▸) and automated refinement with PHENIX (Afonine et al., 2012 ▸). The stereochemistry was checked with PROCHECK (Laskowski et al., 1993 ▸) and the structure was deposited in the PDB as entry 4xpj. The figures showing the structure were produced using PyMOL (http://www.pymol.org). The structure-solution and refinement statistics are presented in Table 4 ▸.
Table 4. Structure solution and refinement.
Values in parentheses are for the outer shell.
| Resolution range () | 37.302.60 (2.702.60) |
| Completeness (%) | 89.7 |
| Cutoff | F > 1.35(F) |
| No. of reflections, working set | 7926 (2008) |
| No. of reflections, test set | 414 (106) |
| Final R cryst | 0.195 (0.254) |
| Final R free | 0.266 (0.288) |
| Cruickshank DPI (coordinate error estimate) | 0.32 |
| No. of non-H atoms | |
| Protein | 1655 |
| Ligand | 35 |
| Water | 60 |
| R.m.s. deviations | |
| Bonds () | 0.009 |
| Angles () | 1.085 |
| Average B factors (2) | |
| Protein | 50.60 |
| Ligand | 78.10 |
| Water | 53.50 |
| Ramachandran plot | |
| Most favoured (%) | 86.7 |
| Additional allowed (%) | 12.8 |
| Generously allowed (%) | 0.5 |
| Disallowed (%) | 0 |
2.5. Biacore analysis
Binding assays between lipids and mNGF were performed by the SPR (surface plasmon resonance) method using a Biacore 3000 instrument (GE Life Sciences). mNGF was immobilized on different CM5 sensor-chip channels (GE Healthcare) using an amine coupling kit, with the remaining uncoupled sites blocked with 1 M ethanolamine at pH 8.5. The lipids were dissolved in deionized water to give a stock solution and the lipids were then further diluted in running buffer in a series of concentrations in the micromolar range from 5 to 30 µM in 5 µM steps. The running buffer systems for lipid-binding sensorgrams were composed of 20 mM buffer salt solution and 50 mM sodium chloride with a pH ranging from 4 to 7. The lipid binding affinities for mNGF were analyzed with the BIAevaluation 4.1 software. All assays were performed at 25°C.
3. Results and discussion
3.1. Lipid-binding assay analyses
To explore the interaction features of the mNGF–phospholipid complex and to verify the binding affinity of LysoPI for mNGF, we performed surface plasmon resonance (SPR) experiments at different pH values to measure the binding affinities. The results indicated that LysoPI has micromolar/submicromolar binding affinity for mNGF in the pH range 4–7 (Figs. 1 ▸ c, 1 ▸ d, 1 ▸ e and 1 ▸ f).
3.2. Overall structure of mNGF complexed with LysoPI
The crystal structure of the mNGF–LysoPI complex (Figs. 2 ▸ a, 2 ▸ b and 2 ▸ c) was determined at 2.6 Å resolution by the molecular-replacement method. Each asymmetric crystal unit contained two mNGF molecules forming a compact dimer, a common feature of NGF structures. Each mNGF protomer in the complex is composed of four antiparallel twisted β-strands connected by four loops (L1, L2, L3 and L4). Two protomers in one asymmetric unit form a dimer by extensive hydrophobic interactions between β-strands, showing no substantial deviations from the previous NGF structures besides variations in the loop regions compared with the native mNGF structure (PDB entry 1btg; Holland et al., 1994 ▸). At the dimeric interface, one LysoPI molecule, with well defined electron density and a fully modelled alkyl tail (C1–C13), is clamped by the mNGF dimer via both hydrogen-bonding interactions between its head group and polar residues and extensive hydrophobic interactions between its alkyl tail and the pocket A hydrophobic residues (Fig. 2 ▸ d).
Figure 2.
The structure of mNGF–LysoPI. (a) Overall structure of the mNGF–LysoPI complex. The LysoPI electron density is represented by a 2F o − F c map (grey) contoured at 0.9σ. (b) The structure of mNGF–LysoPI as in (a) but rotated by 90° about the vertical axis. (c) The structure of mNGF–LysoPI as in (b) but rotated by 90° about the horizontal axis. (d) Four mNGF aromatic hydrophobic residues are responsible for anchoring the alkyl tail of LysoPI (surrounded by the 2F o − F c map at 0.9σ).
3.3. Structural comparisons of mNGF–LysoPI with cNGF, mNGF and mNGF–LysoPS
When comparing the mNGF–LysoPI complex structure with the native mNGF structure, we found a notable difference in the conformation of the hairpin loop L2 formed by residues 41–49, especially Asn45 (Fig. 3 ▸ a), similar to the case in the mNGF–LysoPS complex structure. In the native mNGF structure two L2 loops bend over and are close to each other, with Asn45 of each protomer forming two hydrogen bonds, whereas in the mNGF–LysoPI and other complex structures the two L2 loops point away from the dimer interface, with the Asn45 residues pointing straight up to form a cavity, which is possibly induced by lipid binding, similar to that found in the mNGF–LysoPS complex structure.
Figure 3.
Structural comparisons. (a) Comparison of mNGF and the mNGF–LysoPI complex structure. The details of the L2 loop of mNGF, LysoPI and the mNGF–LysoPI complex are shown in light magenta, yellow and cyan, respectively. (b) Conformational changes of the cNGF, mNGF–LysoPS and mNGF–LysoPI complex structures. The mNGF–LysoPI complex, the mNGF–LysoPS complex and cNGF are shown in cyan, light pink and green, respectively. (c) The interactions between the LysoPI polar head and mNGF shown as a stereoview. (d) The interactions between the LysoPS polar head and mNGF. (e) Superimposition of mNGF–LysoPI on the NT-3–p75NTR complex (PDB entry 3buk); mNGF, LysoPI and NT-3–p75NTR are shown in cyan, yellow and wheat, respectively. (f) Superimposition of mNGF–LysoPI on the NGF–TrkA complex (PDB entry 2ifg); mNGF, LysoPI and NGF–TrkA are shown in cyan, yellow and bright orange, respectively.
Concerning the lipid molecule, the overall LysoPI conformation differs from that of the lipid in cNGF but is similar to that of LysoPS in the mNGF–LysoPS complex structure. The head glycerol groups of LysoPI and LysoPS are in a nearly identical position (Fig. 3 ▸ b), while the inositol moiety of LysoPI is exposed to the solvent to a greater extent and the interaction patterns are different. The hydroxyl of the head glycerol group and the phosphate group of LysoPI form hydrogen bonds to Lys88 of protomer A (Fig. 3 ▸ c) at distances of 2.4 and 3.1 Å, respectively, while the equivalent bonds in mNGF–LysoPS are 2.6 and 2.9 Å, respectively (Fig. 3 ▸ d). The hydrogen bonds between the lipid and Asn45 from an adjacent NGF dimer, which are caused by crystal packing, are also slightly altered (Figs. 3 ▸ c and 3 ▸ d). The most important difference between these two structures is that mNGF–LysoPI has two more lipid-interacting residues, Tyr52 and Arg50, which form hydrogen bonds to the inositol moiety of LysoPI (Fig. 3 ▸ c). The side chain of Tyr52 is hydrogen-bonded to the 4-OH group of the LysoPI inositol moiety, and the main chain of Arg50 interacts with the 3-OH group of the LysoPI inositol moiety. Previous SPR studies indicated that PIP2 binds mNGF with a K d of 86 nM, which is higher than that for LysoPI. Given that the NGF lipid head-group binding site has space to accommodate lipids with a larger polar head, such as PIP2, the high affinity of PIP2 can be explained by the ability of PIP2 to form more or stronger hydrogen bonds to the polar residues around Tyr52 and Arg50.
3.4. Functional implications
LysoPI is a bioactive lipid that can be released into the medium of Ras-transformed fibroblasts and can function as an autocrine cell-growth modulator (Hu et al., 2011 ▸; Ruban et al., 2014 ▸). It has recently been reported that LysoPI can be transported out of the cell by the ATP-binding cassette (ABC) transporter C1 multidrug resistance protein 1 (MRP1). The extracellular LysoPI can be captured by GPCRs and then initiates intracellular signalling cascades (Piñeiro & Falasca, 2012 ▸; Yamashita et al., 2013 ▸).
Our results show that LysoPI can form a stable complex with NGF. Previous studies have shown that the physical interaction between LysoPS and NGF is correlated to the mast-cell activation activity of NGF via the TrkA pathway (Kawamoto et al., 2002 ▸). Thus, we proposed that extracellular LysoPS or LysoPI could also be sensed by NGF.
Superimpositions of mNGF–LysoPI with the NT-3–p75NTR complex (PDB entry 3buk; Gong et al., 2008 ▸) and the NGF–TrkA complex (PDB entry 2ifg; Wehrman et al., 2007 ▸) show that the LysoPI head group has a steric conflict with Asp41 in a NGF-interacting loop in p75NTR (Fig. 3 ▸ e), while the binding lipid does not interfere in the interaction of NGF with TrkA (Fig. 3 ▸ f). Although the NGF–LysoPS and NGF–LysoPI structures indicated that lipid binding does not apparently influence TrkA binding by NGF, the TrkA signalling pathway still could be impacted via the following putative mechanisms: (i) some binding lipids are key cell-membrane components, such as PIP2, so the lipid molecules may direct NGF localization and result in the clustering and specific membrane distribution of NGF and its receptors, and (ii) some lipids acting as secondary signalling messengers can be transported with NGF and TrkA via endocytosis and be released/presented intracellularly to perform their physiological functions.
Although the mechanism of intracellular release or presentation for lipids bound to NGF is currently unknown, studies of other carrier/transporters for PI and its derivatives provide some hints. For example, the Sec14 family homologous proteins are major PI/PC transfer proteins that regulate lipid metabolism (Bankaitis et al., 2010 ▸, 2012 ▸). Recent studies indicated that two Sec14 family homologous proteins, Sfh1 and Sfh3, function as the nanoreactor that transports and presents PI to PI kinases on the plasma membrane for modifications such as phosphorylation (Yang et al., 2013 ▸). Structural comparison of the NGF–LysoPI complex structure with the Sfh3–PI complex reveals that they share protein–lipid interaction similarities; namely, the lipid resides in a hydrophobic cavity with the polar head group exposed in solution, facilitating its recognition by other proteins, which is in line with our speculation that NGF has a functional role in lipid presentation. Although it is still an open question whether the lipid bound to NGF can be readily released when it is transported via endocytosis to a low-pH environment (such as the lysosome), the relatively high NGF binding affinity of LysoPI measured by SPR at pH 4.0 (K d ≃ 0.73 µM; Fig. 1 ▸ c) implies that a low pH is unlikely to be able to cause the release of LysoPI from NGF without the participation of other cell components.
Our NGF–LysoPI complex structure provides new insights into NGF lipid-recognition specificity, which may help further lipid-involved NGF functional studies and in the development of bioactive compounds that can be delivered by NGF.
Supplementary Material
PDB reference: nerve growth factor–lysophosphatidylinositol complex, 4xpj
Acknowledgments
We thank the BL17U beamline staff of the Shanghai Synchrotron Radiation Facility (SSRF) in China for assistance with data collection. We thank M. Xu for assistance with structure determination and Y. Y. Chen for technical support in the Biacore assay. This work was financially supported by grants from the National Basic Research Program of China (2011CB910302), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB08010301) and the National Natural Science Foundation of China (31025009 and 31200558).
References
- Adams, P. D. et al. (2010). Acta Cryst. D66, 213–221.
- Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352–367. [DOI] [PMC free article] [PubMed]
- Angeletti, R. H. (1970). Proc. Natl Acad. Sci. USA, 65, 668–674.
- Bankaitis, V. A., Ile, K. E., Nile, A. H., Ren, J., Ghosh, R. & Schaaf, G. (2012). Adv. Biol. Regul. 52, 115–121. [DOI] [PMC free article] [PubMed]
- Bankaitis, V. A., Mousley, C. J. & Schaaf, G. (2010). Trends Biochem. Sci. 35, 150–160. [DOI] [PMC free article] [PubMed]
- Banks, B. E., Banthorpe, D. V., Charlwood, K. A., Pearce, F. L., Vernon, C. A. & Edwards, D. C. (1973). Nature (London), 246, 503–504. [DOI] [PubMed]
- Barbacid, M. (1994). J. Neurobiol. 25, 1386–1403. [DOI] [PubMed]
- Davies, A. M. (1994). Curr. Biol. 4, 273–276. [DOI] [PubMed]
- Dechant, G. & Barde, Y. A. (1997). Curr. Opin. Neurobiol. 7, 413–418. [DOI] [PubMed]
- Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501. [DOI] [PMC free article] [PubMed]
- Friedman, W. J. & Greene, L. A. (1999). Exp. Cell Res. 253, 131–142. [DOI] [PubMed]
- Furukawa, S. & Furukawa, Y. (1991). No To Shinkei, 43, 1101–1112. [PubMed]
- Gong, Y., Cao, P., Yu, H.-J. & Jiang, T. (2008). Nature (London), 454, 789–793. [DOI] [PubMed]
- Grzelczyk, A. & Gendaszewska-Darmach, E. (2013). Biochimie, 95, 667–679. [DOI] [PubMed]
- Heumann, R. (1994). Curr. Opin. Neurobiol. 4, 668–679. [DOI] [PubMed]
- Holland, D. R., Cousens, L. S., Meng, W. & Matthews, B. W. (1994). J. Mol. Biol. 239, 385–400. [DOI] [PubMed]
- Hu, G., Ren, G. & Shi, Y. (2011). Oncogene, 30, 139–141. [DOI] [PubMed]
- Huang, E. J. & Reichardt, L. F. (2001). Annu. Rev. Neurosci. 24, 677–736. [DOI] [PMC free article] [PubMed]
- Kaplan, D. R., Hempstead, B. L., Martin-Zanca, D., Chao, M. V. & Parada, L. F. (1991). Science, 252, 554–558. [DOI] [PubMed]
- Kaplan, D. R. & Miller, F. D. (1997). Curr. Opin. Cell Biol. 9, 213–221. [DOI] [PubMed]
- Kawamoto, K., Aoki, J., Tanaka, A., Itakura, A., Hosono, H., Arai, H., Kiso, Y. & Matsuda, H. (2002). J. Immunol. 168, 6412–6419. [DOI] [PubMed]
- Klein, R., Jing, S., Nanduri, V., O’Rourke, E. & Barbacid, M. (1991). Cell, 65, 189–197. [DOI] [PubMed]
- Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. (1993). J. Appl. Cryst. 26, 283–291.
- Niewiadomska, G. & Małecki, M. (1999). Postepy Biochem. 45, 21–31. [PubMed]
- Otwinowski, Z. & Minor, W. (1997). Methods Enzymol. 276, 307–326. [DOI] [PubMed]
- Park, H. & Poo, M.-M. (2013). Nature Rev. Neurosci. 14, 7–23. [DOI] [PubMed]
- Piñeiro, R. & Falasca, M. (2012). Biochim. Biophys. Acta, 1821, 694–705. [DOI] [PubMed]
- Ruban, E. L., Ferro, R., Arifin, S. A. & Falasca, M. (2014). Biochem. Soc. Trans. 42, 1372–1377. [DOI] [PubMed]
- Soligo, M., Nori, S. L., Protto, V., Florenzano, F. & Manni, L. (2013). Int. Rev. Neurobiol. 111, 91–124. [DOI] [PubMed]
- Terwilliger, T. C., Adams, P. D., Read, R. J., McCoy, A. J., Moriarty, N. W., Grosse-Kunstleve, R. W., Afonine, P. V., Zwart, P. H. & Hung, L.-W. (2009). Acta Cryst. D65, 582–601. [DOI] [PMC free article] [PubMed]
- Tometten, M., Blois, S. & Arck, P. C. (2005). Chem. Immunol. Allergy, 89, 135–148. [DOI] [PubMed]
- Tong, Q., Wang, F., Zhou, H.-Z., Sun, H. -L., Song, H., Shu, Y.-Y., Gong, Y., Zhang, W.-T., Cai, T.-X., Yang, F.-Q., Tang, J. & Jiang, T. (2012). FASEB J. 26, 3811–3821. [DOI] [PubMed]
- Vega, J. A., García-Suárez, O., Hannestad, J., Pérez-Pérez, M. & Germanà, A. (2003). J. Anat. 203, 1–19. [DOI] [PMC free article] [PubMed]
- Vinores, S. & Guroff, G. (1980). Annu. Rev. Biophys. Bioeng. 9, 223–257. [DOI] [PubMed]
- Wehrman, T., He, X., Raab, B., Dukipatti, A., Blau, H. & Garcia, K. C. (2007). Neuron, 53, 25–38. [DOI] [PubMed]
- Yamashita, T., Fujitani, M., Hata, K., Mimura, F. & Yamagishi, S. (2005). Anat. Sci. Int. 80, 37–41. [DOI] [PubMed]
- Yamashita, A., Oka, S., Tanikawa, T., Hayashi, Y., Nemoto-Sasaki, Y. & Sugiura, T. (2013). Prostaglandins Other Lipid Mediat. 107, 103–116. [DOI] [PubMed]
- Yang, H., Tong, J., Leonard, T. A. & Im, Y. J. (2013). FEBS Lett. 587, 1610–1616. [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
PDB reference: nerve growth factor–lysophosphatidylinositol complex, 4xpj



