Apo2Ltraildr5 structures

Overview

The crystal structure of Apo2L/TRAIL bound to the ECD of one of its signaling receptors, DR5, has recently been

Figure 1 Superposition of monomers from independent crystal structure determinations of Apo2L/TRAIL taken from complex structures with DR5, showing the variations in loop conformation. Monomer A (Hymowitz et al. [7]) is rendered in light gray; monomer B (Mongkolsapaya et al. [10]), in medium gray; and monomer D (Cha et al. [11]), in dark gray.

Figure 1 Superposition of monomers from independent crystal structure determinations of Apo2L/TRAIL taken from complex structures with DR5, showing the variations in loop conformation. Monomer A (Hymowitz et al. [7]) is rendered in light gray; monomer B (Mongkolsapaya et al. [10]), in medium gray; and monomer D (Cha et al. [11]), in dark gray.

determined by three different groups [7,10,11], These three independent structures are in good overall agreement, especially around the receptor binding site. As in the LT-TNF-R1 complex, the three symmetrical binding sites on the ligand are nestled between the monomer-monomer interfaces; each site forms two distinct patches, one centered around Tyr 216 and the other around Gln 205 (Fig. 2). Both of these residues have been shown to be critically important for Apo2L/TRAIL activity [6]. Two loops from DR5, the 50s loop in CRD2 and the 90s loop in CRD3, make almost all the contacts with the ligand. The conformation of the 90s loop is very different between DR5 and TNF-R1, and its sequence in other TNF-Rs is highly variable. This loop was proposed to be important for defining the specificity and cross-reactivity among TNF-R family members [11]. In contrast, the 50s loop has a backbone conformation very similar to the corresponding loop in TNF-R1 (the Ca atoms of DR5 residues 50 to 62 superimpose to within 0.4 A on the Ca atoms of TNF-R1 residues 60 to 72). Moreover, the sequence of this loop is well conserved among TNF-R family members, suggesting that its conformation and binding characteristics may be conserved as well (Fig. 3). Most of the differences in the independent structures of Apo2L/TRAIL-DR5 occur at loops on the lig-and away from the receptor binding site and involve poorly ordered parts of the molecule. In exception to this, the most

Figure 2 Crystal structure of Apo2L/TRAIL (light gray) bound to the ECD of DR5 (dark gray and black) (Hymowitz et al, [7]). Residues 132 to 143 of Apo2L/TRAIL are disordered and shown as small balls. The internal zinc ion is shown as a small dark sphere at the tip of the Apo2L/TRAIL trimer. The side chains of Apo2L/TRAIL residues Tyr 216 and Gln 205 are shown in close-packed rendering for one protomer (in gray).

Figure 2 Crystal structure of Apo2L/TRAIL (light gray) bound to the ECD of DR5 (dark gray and black) (Hymowitz et al, [7]). Residues 132 to 143 of Apo2L/TRAIL are disordered and shown as small balls. The internal zinc ion is shown as a small dark sphere at the tip of the Apo2L/TRAIL trimer. The side chains of Apo2L/TRAIL residues Tyr 216 and Gln 205 are shown in close-packed rendering for one protomer (in gray).

dramatic and functionally relevant difference occurs in the EF loop of Apo2L/TRAIL in the structure by Mongkolsapaya et al. [10], for which the lack of zinc resulted in a different, less ordered conformation than that seen in the other two structures. As indicated above, zinc is required for maintaining the structural integrity and activity of Apo2L/TRAIL, and the conformation observed in the structures by Hymowitz et al. [6,7] and Cha et al. [11] likely represents the native conformation.

Variation in the Orientation of CRD3 of DR5

Overall, the conformation of the receptors, and particularly of the loops that contact Apo2L/TRAIL, is well conserved among the three independent structures. One exception is revealed by a comparison of all independent copies of DR5 (Fig. 3). This comparison shows that the orientation of CRD3 with respect to the other domains is variable even among receptors within the same complex, resulting in a displacement of structurally equivalent residues by up to 5 Â. Similar variation is seen when individual DR5 structures are compared to TNF-R1 as well as in comparisons between different structures of TNF-R1 alone or bound to LT (Fig. 3). Therefore, this variability is likely to be an intrinsic property of the structure of TNF-Rs with multiple CRDs rather than a source for specificity among family members as was suggested on the basis of a single unique copy of DR5 in the structure of Mongkolsapaya et al. [10]. A structure of Apo2L/TRAIL bound to the extracellular

Figure 3 Superposition of the Ca traces of ten independent DR5 structures. The three chains from Hymowitz et al. [7] are shown in light gray; one chain from Mongkolsapaya et al. [10], in medium gray; six chains from Cha et al. [11], in dark gray; and TNF-R1 from Banner et al. [3], in black. The DR5 chains were superimposed using the Ca atoms of residues 22 to 101 (the ordered portion of CRD1, all of CRD2, and the first loop of CRD3). TNF-R1 was superimposed on DR5 using the Ca atoms of the structurally equivalent residues in CRD2.

Figure 3 Superposition of the Ca traces of ten independent DR5 structures. The three chains from Hymowitz et al. [7] are shown in light gray; one chain from Mongkolsapaya et al. [10], in medium gray; six chains from Cha et al. [11], in dark gray; and TNF-R1 from Banner et al. [3], in black. The DR5 chains were superimposed using the Ca atoms of residues 22 to 101 (the ordered portion of CRD1, all of CRD2, and the first loop of CRD3). TNF-R1 was superimposed on DR5 using the Ca atoms of the structurally equivalent residues in CRD2.

domain (ECD) of DR4 or to one of its decoy receptors would reveal if variation in CRD3 orientation is a common feature of other Apo2L/TRAIL receptors.

The AA' Loop of Apo2L/TRAIL

Much interest has centered on a possible role for the unusually long AA' loop of Apo2L/TRAIL (residues 130 to 150). In the structures by Hymowitz et al. [7] and Mongkolsapaya et al. [10], this loop turns away from the receptor at residue 131 and becomes poorly ordered. In contrast, in the structure by Cha et al. [11] residues 130 to 135, while still poorly ordered, are in a different conformation, allowing for additional contacts to DR5. However, in this model there appears to be insufficient space between DR5 and Apo2L/TRAIL for the disordered part of the chain (residues 136 to 145) to fold back across the protein, suggesting that the conformation of the last few marginally ordered residues in this model may not be biologically relevant.

In all three structures, the rest of the AA' loop uses residues 130 and 145 to 149 to make minor contacts with the receptor; however, none of these contacts involves burying significant accessible surface area. These observations are consistent with the lack of effect on binding (to DR4, DR5, or DcR2) or on the activity of single alanine substitutions at residues 130, 134, 136, 138, 140-143, and 149 within this loop. The reduction or elimination of receptor binding activity observed upon deletion (A137-152 [5]), shortening (A132-135 [11]), or multiple mutation [10] of this loop may therefore be an indirect effect on the conformation of other binding determinants. In light of the point mutation results and the high degree of conformational heterogeneity seen in this loop among the three crystal structures, we conclude that these peripheral interactions are likely to vary in other death receptor complexes and that, even when present, they are unlikely to contribute as much binding energy as contacts made by residues analogous to Apo2L/TRAIL Tyr 216 and Gln 205.

Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment