Ran's C-terminal, basic patch, and nucleotide exchange mechanisms in light of a canonical structure for Rab, Rho, Ras, and Ran GTPases

Abstract

Proteins comprising the core of the eukaryotic cellular machinery are often highly conserved, presumably due to selective constraints maintaining important structural features. We have developed statistical procedures to decompose these constraints into distinct categories and to pinpoint critical structural features within each category. When applied to P-loop GTPases, this revealed within Rab, Rho, Ras, and Ran a canonical network of molecular interactions centered on bound nucleotide. This network presumably performs a crucial structural and/or mechanistic role considering that it has persisted for more than a billion years after the divergence of these families. We call these 'FY-pivot' GTPases after their most distinguishing feature, a phenylalanine or tyrosine that functions as a pivot within this network. Specific families deviate somewhat from canonical features in interesting ways, presumably reflecting their functional specialization during evolution. We illustrate this here for Ran GTPases, within which two highly conserved histidines, His30 and His139, strikingly diverge from their canonical counterparts. These, along with other residues specifically conserved in Ran, such as Tyr98, Lys99, and Phe138, appear to work in conjunction with FY-pivot canonical residues to facilitate alternative conformations in which these histidines are strategically positioned to couple Ran's basic patch and C-terminal switch to nucleotide exchange and effector binding. Other core components of the cellular machinery are likewise amenable to this approach, which we term Contrast Hierarchical Alignment and Interaction Network (CHAIN) analysis.

Figure 1.

Ran family alignments generated by CHAIN analysis procedures. (A) Conventional multiple alignment. The leftmostcolumn specifies each sequence's phylum; these are colored by major eukaryotic taxa as follows: metazoans, red; fungi, dark yellow; plants, green; protozoans, cyan. The top sequence is the query. The NCBI sequence identifiers are: 5453555, 17553976, 3113905, 6323324, 11067497, 1710007, 13812290, 4881271, 585782, 1172840, 15691764, 606985, 8593487, 14089387, 585780, and 14581093. (B–D) Contrast hierarchical alignment. CHAIN analysis applies three different sequence highlighting schemes to the Ran family alignment in A to reveal the selective constraints most characteristic of each of three hierarchical categories, which here correspond to P-loop GTPases in B; FY-pivot GTPases in C; and the Ran family in D. Organism descriptions (leftmost column) are colored by category as specified in Figure 2; sequences obtained from ESTs are indicated. Note that for the Basidiomycota protein, which was predicted from an EST, the replacement of histidine (H) at position 30 by aspartate (D) is likely due to a sequencing error. Chemically similar highlighted residues are colored similarly. Histograms above the alignments display the relative strength of the inferred selective constraint acting at each position within that category (quasi-logarithmic scaling is used; see Methods). This and other aspects of this representational scheme are explained in Figure 2. Dots below the histograms (and directly above the alignments) indicate those residues specifically assigned to each category. Gray dots in B and C indicate positions for which Ran deviates from the canonical residues for that category. Note that the conventional alignment in A helps identify residues associated with intermediate categories, which correspond to conserved positions in Ran that are inconsistently conserved within the three categories of this hierarchy. A few residue positions (such as T42^Ran) are misclassified in this analysis due to alignment errors; these were detected and addressed in our analysis through structural studies and CHAIN analysis of related GTPases.

Figure 1.

Ran family alignments generated by CHAIN analysis procedures. (A) Conventional multiple alignment. The leftmostcolumn specifies each sequence's phylum; these are colored by major eukaryotic taxa as follows: metazoans, red; fungi, dark yellow; plants, green; protozoans, cyan. The top sequence is the query. The NCBI sequence identifiers are: 5453555, 17553976, 3113905, 6323324, 11067497, 1710007, 13812290, 4881271, 585782, 1172840, 15691764, 606985, 8593487, 14089387, 585780, and 14581093. (B–D) Contrast hierarchical alignment. CHAIN analysis applies three different sequence highlighting schemes to the Ran family alignment in A to reveal the selective constraints most characteristic of each of three hierarchical categories, which here correspond to P-loop GTPases in B; FY-pivot GTPases in C; and the Ran family in D. Organism descriptions (leftmost column) are colored by category as specified in Figure 2; sequences obtained from ESTs are indicated. Note that for the Basidiomycota protein, which was predicted from an EST, the replacement of histidine (H) at position 30 by aspartate (D) is likely due to a sequencing error. Chemically similar highlighted residues are colored similarly. Histograms above the alignments display the relative strength of the inferred selective constraint acting at each position within that category (quasi-logarithmic scaling is used; see Methods). This and other aspects of this representational scheme are explained in Figure 2. Dots below the histograms (and directly above the alignments) indicate those residues specifically assigned to each category. Gray dots in B and C indicate positions for which Ran deviates from the canonical residues for that category. Note that the conventional alignment in A helps identify residues associated with intermediate categories, which correspond to conserved positions in Ran that are inconsistently conserved within the three categories of this hierarchy. A few residue positions (such as T42^Ran) are misclassified in this analysis due to alignment errors; these were detected and addressed in our analysis through structural studies and CHAIN analysis of related GTPases.

Figure 2.

CHAIN analysis representational schemes. The examples shown in A–D correspond to the hierarchical alignment in Figure 1. (A) Venn diagram representing hierarchical relationships between aligned sequence sets. The dotted oval corresponds to a hypothetical intermediate category consisting of conserved residues in Ran that fall outside the categories of this particular hierarchy. (B) Notation used within hierarchical alignments. Position 139 of Ran is shown. The main and superfamily sets, which contain too many sequences to display directly, are represented in the alignment as position-specific conserved patterns. (The total number of sequences for these two categories is shown in parentheses.) The corresponding residue frequencies (‘res_freq’) are given in integer tenths below conserved residues. For example, a ‘5’ in integer tenths indicates that the corresponding residue directly above it occurs in 50%–60% of the (weighted) sequences. Insertion and deletion frequencies are similarly given in integer tenths (black; range 10%–100%) or hundredths (gray; range 1%–9%) as indicated. Ran family aligned residues are displayed directly. Histogram bar heights are approximately logarithmically proportional to the measure of selective constraint (see Methods), as defined by the following urn model. (C) Urn model for measuring the selective constraint acting on a specific position. The residues observed in the main set at this position are modeled as distinctly colored balls in an urn. Some of the colors are similar (representing biochemically similar amino acids). The selective constraint is then defined as the difficulty of drawing by chance at least as many of the same- or similarly-colored balls from the urn as are observed in the subalignments (in this case, alanine for the FY-pivot superfamily or histidine for the Ran family). Note that our analysis of the Ran family uses the main set as the ‘superalignment’ urn (see Methods); alternatively, the FY-pivot GTPases may also be used as the superalignment urn, though the sparser data set would yield less accurate background frequency estimates. Note that the alignments in Figures 1A and 1B measure sequence constraints using a standard background model (see Methods). (D) Color scheme used for residue side-chains in Figures 4–9. (E) Color scheme for structural regions described in the text and figures. The structure of Sec4p (pdb code: 1G17) is shown in complex with a GTP analog (cyan) and magnesium (dark green).

Figure 3.

Characteristic conserved residues within distinct FY-pivot GTPases of known structure. FY-pivot GTPases from distinct families or subfamilies were structurally aligned against Sec4p using the CE program (Shindyalov and Bourne 2001); alignment errors were corrected based on direct structural observations. This alignment was then used as the ‘family set’ alignment for CHAIN analysis (see Methods) to obtain this two-tier hierarchical alignment. The histograms and highlighted residues correspond to category-specific selective constraints, as described in Figure 2B,C. (A) Conserved residues characteristic of P-loop GTPases. The structural regions shown in Figure 2E and in Figures 4–9 are indicated at the top. (B) Conserved residues characteristic of FY-pivot GTPases. Key residues shown in Figure 4 and in Figures 5–9 for Sec4p and Ran, respectively, are indicated at the bottom.

Figure 4.

Sec4p as a structural prototype of FY-pivot GTPases. The structure of Sec4p is shown in complex with GDP (pdb code: 1G16). The corresponding hierarchical alignment is given in Figure 3. Hydrogen bonds are depicted as dotted lines, and aromatic-aromatic and van der Waals interactions as dot clouds. Dotted lines into clouds depict CH-π or NH-π interactions (Weiss et al. 2001). Color scheme: GDP (cyan); main-chain traces and residue designations (colored by regions as indicated in Fig. 2E); residue side-chains and canonical glycine main-chains (color scheme of Fig. 2D); oxygen, nitrogen, and hydrogen atoms establishing hydrogen bonds (red, blue, and white, respectively); hydrogen bonding carbons (colored as their corresponding side-chains). Figures were generated using RasMol (Sayle and Milner-White 1995). (A) Canonical interactions between the LV.D and NK.D regions. These include a perpendicular aromatic-aromatic interaction (F108-Y100), four CH-π interactions (F108-Y100, G132-Y100, V142-F108, and I102-R140), and main-chain hydrogen bonds to two side-chains (T107 and R140). Y100 is the FY-pivot residue. The inset highlights interactions between the LV.D and P-loop regions. (B) Canonical interactions within and between the NK.D and SA regions. These include packing of a phenylalanine or tyrosine (F158) against two small residues (G147 and A151) in the helix following the NK.D loop, a salt bridge (R140 and E160), and several main-chain hydrogen bonds. Residues within the LV.D region are shown for comparison with A. (C) Canonical interactions between the SA and Switch I regions. Note that, unlike Ran (Figs. 7D, 8), there are no major structural differences between the GDP and GTP-forms of Sec4p in these regions. Also shown is a previously noted (Hall 2000) canonical aromatic-aromatic interaction between a phenylalanine (F45) and bound guanine. Residues homologous to R39 (mainly arginine, glutamine, or serine), though inconsistently conserved at the sequence level, conserve hydrogen bonding interactions with the SA region.

Figure 5.

Canonical interactions between the LV.D and NK.D regions within the Ras, Rho, and Ran families. The perspective is as shown in Figure 4A. Representations and coloring are as described in Figure 4. (A) Human Ras-GTP (pdb code: 1QRA). Note that a canonical tryptophan (W114^Sec4) that typically forms CH-π and NH-π hydrogen bonds with P-loop main-chain atoms is displaced by a tyrosine (Y96^Ras; or, within other Ras proteins, a phenylalanine), suggesting that this residue performs a Ras-specific function. (B) The Rho family human Rac1 GTPase in complex with a GTP analog (pdb code: 1MH1). (C) Canine Ran-GDP (pdb code: 1BYU).

Figure 6.

Canonical interactions between the NK.D and SA regions within the Ras, Rho, and Ran families. The perspective is as shown in Figure 4B. Representations and coloring are as described in Figure 4. (A) Human Ras-GTP (pdb code: 1QRA). (B) The Rho family human Rac1 in complex with a GTP analog (pdb code: 1MH1). (C) Canine Ran-GDP (pdb code: 1BYU).

Figure 7.

Canonical interactions between the SA and Switch I regions within the Ras, Rho, and Ran families. Perspective is as shown in Figure 4C. Representations and coloring are as described in Figure 4. (A) Human Ras-GTP (pdb code: 1QRA). Note that at position 23 the canonical Phe or Tyr is replaced by Leu, which is highly conserved in the Ras subfamily. (B) The Ras family Rap2a GTPase from human in complex with GTP (pdb code: 2RAP). In the Rap2 subfamily, the canonical residue occurs at position 23. (C) The Rho family Rac1 GTPase from human (pdb code: 1MH1). (D) Human Ran in complex with importin-β and a GTP analog (pdb code: 1IBR). Note that H30 of Ran is noncanonical and that the canonical hydrogen bond between the previous residue and the SA region involves a N154 side-chain atom rather than a main-chain atom. These deviant features appear to be important to Ran's C-terminal switching mechanism (see Fig. 8).

Figure 8.

Interactions involving H30^Ran and Ran's C-terminal switching mechanism. See text for details. (A) Ran-GTP-importin-β. Note that H30^Ran forms an aromatic-aromatic interaction with F161^Ran that is similar to the canonical interaction (Fig. 4C). Rather than simply interacting with canonical F161^Ran, however, H30^Ran inserts itself into an aromatic pocket formed by F161^Ran and two other aromatic residues specifically conserved in Ran, F26^Ran and F157^Ran (data not shown). (B) Ran-GDP. H30^Ran appears to be critical for establishing this conformation, in which Ran's C-terminal linker displaces the F35^Ran interaction with guanine. The inset provides a different perspective that shows main-chain to main-chain hydrogen bonds between the C-terminal and H30 regions.

Figure 9.

Interactions involving H139^Ran, Y98^Ran, and Ran's basic patch switching mechanism. See text for details. This figure focuses on three structural forms: (A,D) canine Ran-GDP (pdb code: 1BYU), (B,E) human Ran in complex with RCC1 (pdb code: 1I2M), and (C) human Ran in complex with importin-β and a GTP analog (pdb code: 1IBR). The perspective for A–C is as that for Sec4p in Figure 4B; the perspective for D,E is rotated approximately 180° relative to that for A,B. The reaction pathway for these three forms of Ran is shown between the upper and lower figures. Note that in C the invariant histidine at position 139 in Ran forms both an aromatic-aromatic interaction and an NH-π (or perhaps a CH-π) interaction with Trp342 of importin-β. (F) Proposed mechanism for Ran nucleotide exchange. The conformation of Canine Ran-GDP is shown with arrows indicating the direction of movement of key residues and main-chain regions upon binding to RCC1. A movie demonstrating the conformational changes involved in this mechanism is available as supplementary information at www.genome.org.