Hedgehogs like it sweet, too
In this issue of PNAS, Leahy and coworkers (1) present the first crystal structure of a complex between the N-terminal fragment of Hh (HhN) and an Ihog fragment encompassing its two extracellular fibronectin domains (IhogFn1–2). This work is part of the group's ongoing effort to elucidate the structural basis for hedgehog (Hh) signal transmission at the cell surface (2). Based on this structure and supporting biochemical data, the authors put forth a new model in which heparan sulfate glycosaminoglycans (HSGAGs) play a direct role in promoting Hh signaling.
Hh is a major morphogen during embryogenesis and is also involved in the maintenance of adult stem cells (3). The Hh pathway subtly controls transcription through a positive regulator, CiA, and a negative regulator, CiR (reviewed in ref. 4). Depending on the balance of those two transcription factors, Hh target genes may be activated, repressed, or responsive to regulation by other signaling pathways. This intricate system of transcriptional control enables the establishment of gradients at critical junctures in development, such as during the patterning of the anterior–posterior axis in fruit flies. Deregulated Hh signaling has been implicated in basal-cell carcinoma and medulloblastoma cancers (reviewed in ref. 5). To signal, Hh binds to Patched (Ptc) and releases Ptc's inhibition of Smoothened (Smo) (reviewed in ref. 6), allowing Smo to derepress transcription. Ihog is a recently discovered component of the cell surface Hh signaling apparatus that is necessary for high-affinity binding between Ptc and Hh (7). Additionally, Hh requires the presence of HSGAGs that are thought to regulate the movement of Hh through tissues (reviewed in ref. 8).
Although the cocrystallized heparin (a highly sulfated HSGAG) molecules are disordered in the structure, Leahy and coworkers (1) infer that HS plays a direct role in the assembly of the Hh signaling apparatus. Regions of positive electrostatic potential on Hh and Ihog are juxtaposed in the structure and form a continuous basic strip where HSGAG is proposed to bind and enhance Hh–Ihog affinity (Fig. 1). The disordering of heparin is attributed to the high concentration of phosphate and sulfate ions in the crystallization buffer. The authors provide several lines of biochemical evidence that this basic strip represents the site of heparin binding. Mutation of the residues of this basic strip reduces IhogFn1–2 binding to a heparin-affinity column; heparinase treatment diminishes binding of HhN to Ihog-overexpressing cells; and heparin stabilizes the formation of Hh–Ihog complexes in solution. The authors also show that Ihog homodimerizes in vitro in the presence of heparin. Interestingly, a sulfated monosaccharide is trapped at the dimer interface, leading the authors to suggest that HS might stabilize Ihog homodimerization.
Hedgehog is a major morphogen and is involved in the maintenance of adult stem cells.
A schematic of the proposed role for HS in Hh signaling in one unit of the dimer. The blue shaded box represents the putative heparin-binding site formed at the region where Hh binds IhogFn1. Heparin, depicted by a black curved line, binds to the putative heparin-binding site to bridge the Hh–IhogFn1–2 interface.
The authors' data raise the possibility that, in addition to controlling Hh movement through the extracellular space, HSGAGs may also modulate Hh signaling by influencing the nature of the signaling apparatus itself. Given the fact that the sulfation and carbohydrate backbone composition of HSGAGs change dynamically during embryogenesis (9), the differential incorporation of HSGAGs into the Hh signaling apparatus may prove to be another means of regulating the Hh pathway. Consequently, the crystal structure solved by McLellan et al.(1) provides a stepping stone for further inquiry into the role of HSGAGs in the formation of the Hh cell surface receptor complex and its signal transduction.
HSGAGs facilitate the activity of several signal transduction systems, most prominently that of fibroblast growth factors (FGF), where its obligatory role is structurally well understood (reviewed in ref. 10). The crystal structure of a 2:2:2 FGF2–FGFR1c-heparin dimer shows that HSGAGs promote FGF signaling by stabilizing the formation of a symmetric 2:2 FGF–FGFR signaling unit. Both the ligand and receptor from one FGF–FGFR complex interact with the receptor in the adjoining FGF–FGFR complex, forming the dimer interface. The heparin-binding sites of the two receptors and the two ligands in the dimer merge to form one large basic canyon into which two HSGAG molecules bind in symmetric fashion. Heparin binding promotes protein–protein contacts between ligand and receptor within the 1:1 FGF–FGFR complex as well between the two complexes in the 2:2 FGF–FGFR dimer.
Although both FGF–FGFR and Hh–Ihog complexes use HSGAGs for signaling, there are two key differences between the modes of dimerization used by these two systems: (i) each Hh interacts with only one Ihog, and the dimer interface is exclusively mediated by Ihog–Ihog contacts, whereas FGF interacts with both FGFRs, and the dimer interface consists of both FGF–FGFR and FGFR–FGFR contacts; and (ii) the two heparin-binding basic strips are separated in the 2:2 Hh–Ihog dimer, and the putative heparin-binding site near the Ihog–Ihog dimer interface is likewise separated from the basic strip bridging Hh–IhogFn1. In stark contrast, all four heparin-binding sites in the FGF–FGFR dimer merge to form a basic canyon.
The coagulation cascade is another system where the role of HSGAGs in promoting protein–protein interactions can be appreciated (11). The ability of antithrombin (AT) to inactivate Factor Xa and thrombin is mediated throughHSGAGs that cause a 10,000-fold increase in the rate of inhibition of those coagulation factors (reviewed in ref. 12). The structure of the AT–thrombin–heparin ternary complex has been solved by two different groups (13, 14) and shows how a heparin dodecasaccharide bridges the heparin-binding sites of AT and thrombin. Upon heparin binding, AT undergoes a conformational change, which increases AT's affinity for heparin. An allosteric mechanism can be ruled out for the Hh–Ihog signaling system, however, because McLellan et al.(1) did not notice any conformational change in HhN or Ihog subsequent to binding.
The established literature regarding the role of HSGAGs in FGF signaling and AT anticoagulatory activity can serve as a framework for further inquiry into HSGAGs' role in Hh signaling. The molecular basis by which HSGAGs promote Ihog–Ihog dimerization is among the remaining questions to be answered in the Hh field. Although it is plausible that the basic strip observed at the Hh–IhogFn1 interface represents a genuine heparin-binding site, it is less likely that a sulfated monosaccharide situated between the Fn1 and Fn2 domains of Ihog could alone be sufficient for Ihog homodimerization. Figure 7 in ref. 1 shows that the monosaccharide's interactions are confined to just one copy of Ihog in the dimer pair. In all probability, some elusive electron density remains to be resolved that would further explain Ihog homodimerization. Biochemically determining the minimal HS epitope required for Hh signaling and Ihog homodimerization would complement these structural studies. Even after answering these questions, however, the definitive explanation of the role of HSGAGs awaits the solution of an Hh–Ihog crystal structure containing a resolved heparin oligosaccharide. Future challenges for structural biologists in the Hh field include delineating how the Hh–Ihog dimer interacts with Ptc and other components of the Hh signaling apparatus at the cell surface.
Footnotes
- *To whom correspondence should be addressed. E-mail: mohammad{at}saturn.med.nyu.edu
-
Author contributions: A.B. and M.M. wrote the paper.
-
The authors declare no conflict of interest.
-
See companion article on page 17208.
- © 2006 by The National Academy of Sciences of the USA
