Substrate-induced assembly of Methanococcoides burtonii D-ribulose-1,5-bisphosphate carboxylase/oxygenase dimers into decamers

Abstract

Like many enzymes, the biogenesis of the multi-subunit CO(2)-fixing enzyme ribulose-1,5-bisphosphate (RuBP) carboxylase/oxygenase (Rubisco) in different organisms requires molecular chaperones. When expressed in Escherichia coli, the large (L) subunits of the Rubisco from the archaeabacterium Methanococcoides burtonii assemble into functional dimers (L(2)). However, further assembly into pentamers of L(2) (L(10)) occurs when expressed in tobacco chloroplasts or E. coli producing RuBP. In vitro analyses indicate that the sequential assembly of L(2) into L(10) (via detectable L(4) and L(6) intermediates) occurs without chaperone involvement and is stimulated by protein rearrangements associated with either the binding of substrate RuBP, the tight binding transition state analog carboxyarabinitol-1,5-bisphosphate, or inhibitory divalent metal ions within the active site. The catalytic properties of L(2) and L(10) M. burtonii Rubisco (MbR) were indistinguishable. At 25 degrees C they both shared a low specificity for CO(2) over O(2) (1.1 mol x mol(-1)) and RuBP carboxylation rates that were distinctively enhanced at low pH (approximately 4 s(-1) at pH 6, relative to 0.8 s(-1) at pH 8) with a temperature optimum of 55 degrees C. Like other archaeal Rubiscos, MbR also has a high O(2) affinity (K(m)(O(2)) = approximately 2.5 microM). The catalytic and structural similarities of MbR to other archaeal Rubiscos contrast with its closer sequence homology to bacterial L(2) Rubisco, complicating its classification within the Rubisco superfamily.

FIGURE 1.

Ligand induced sequential assembly of MbR L₂ units at room temperature into stable decamers. Nondenaturing PAGE (A, C, and E) and nanoESI-MS (B, D, and F) of Rubisco complexes with the corresponding masses derived from the m/z of the ions shown in brackets. A, separation of purified tobacco L₈S₈ Rubisco (t^wt), tobacco produced R. rubrum L₂ Rubisco from ^cmtrL1 (t^Rr) (24) and E. coli purified L₂ MbR (E^MbR) (supplemental Fig. 2). B, nanoESI-MS of the denatured E^MbR monomer (5% (v/v) formic acid) and native dimer (0.1 m ammonium acetate, pH 7). The major ion in the spectrum of the monomer is at m/z 3524.8 corresponding to [MbR+15H]¹⁵⁺, and for the dimer it is 5564.9 corresponding to [MbR+19H]¹⁹⁺. C–F, induced assembly of CO₂-Mg²⁺ activated MbR L₂ units (final concentration, 3.8 μm) into L₄, L₆, and L₁₀ complexes by 1 mm RuBP (C) and 1 mm CABP (E) over time and the corresponding nanoESI-MS of nonactivated MbR after 30 min of incubation with 100 μm RuBP (D) or CABP (F). Where RuBP or CABP are present (D and F), the peaks in nanoESI mass spectra are broad, making it difficult to determine precise binding stoichiometries. The data were, however, consistent with the binding of one RuBP or CABP/monomer. m, marker proteins (sizes shown).

FIGURE 2.

Influence of activation with Mg²⁺, Ca²⁺, or Co²⁺ on MbR activity, sugar-phosphate binding, and structure. A, recovery of RuBP-dependent ¹⁴CO₂ activity over time relative to the activity of the Mg²⁺ activated MbR after 8 min. B, recovery of bound [¹⁴C]CABP or [³H]RuBP after activating decarbamylated L₂ MbR for 15 min at 25 °C with 25 mm NaHCO₃ and either 20 mm MgCl₂ or 20 mm CaCl₂ before incubating with 2 μm of [¹⁴C]CABP (1.6 MBq·μmol⁻¹, white bars) or [³H]RuBP (16 MBq·μmol⁻¹, black bars) for 15 min and resolving the amount bound to MbR by Superdex G50 fine gel filtration (25). The values ± S.D. are the averages of duplicate (n = 2) samples. C, nondenaturing PAGE analysis of the influence of activation with Mg²⁺, Ca²⁺, or Co²⁺, with or without CABP, on the assembly of L₂ MbR into L₁₀ after 20 min at 25 °C. D, nanoESI-MS of MbR in 100 mm ammonium acetate, pH 7, incubated for 30 min at 2 °C with 100 μm of MgCl₂, CaCl₂, or CoCl₂.

FIGURE 3.

RuBP stimulates assembly of L₁₀ MbR in vivo. A, comparative organization of the plastomes of wild-type tobacco (t^wt), the tobacco master line ^cmtrL1 (24) (t^Rr), and transplastomic tobacco producing MbR (t^MbR). Plasmid pLevMbiiL was biolistically transformed into ^cmtrL1 where the flanking plastome sequence in pLevMbiiL (indicated by dotted lines with the numbering relative to the Nicotiana tabacum plastome sequence; GenBank^TM accession number Z00044) directed the homologous replacement of the R. rubrum L₂ Rubisco gene (^cmrbcM), psbA 3′-untranslated region (T), and loxP (triangle) sequences with a bicistronic gene encompassing rbcL^Mb, 216 bp of the tobacco rbcL 3′-untranslated region (T), a promoter-less aadA (marker) gene, and rps16 3′-untranslated region (t) sequence. B, SDS-PAGE and immunoblot analysis of 10 μg of leaf protein from the representative tobacco lines (noting the N terminus of MbR in t^MbR, MSPQTETKASVGF, varies from the E. coli purified MbR sequence, MSLIYEDLV). C, nondenaturing PAGE of the samples from B and the purified H₆Ub-MbR expressed in RuBP producing E. coli-PRK cells before (^UbL₁₀) and after (L₁₀) removal of the H₆Ub fusion (see supplemental Fig. 2). L₈S₈, Form I tobacco Rubisco; m, marker proteins (sizes shown).

FIGURE 4.

The L₂ to L₁₀ transition is reversible. A, nondenaturing PAGE analysis of purified L₁₀ MbR (final concentration, 7.5 pm) before (0 h) and after 24 h of dialysis in buffer (50 mm HEPES-NaOH, pH 7.0, 20 mm EDTA) at 4 or 22 °C. After 24 h at 22 °C, the dialyzed MbR was incubated with 20 mm MgCl₂ and 1 mm CABP for 30 min at 22 °C prior to electrophoresis. B, schematic proposing a reversible assembly process of L₂ MbR via possible L₄ and L₆ intermediates into a pentameric ring.

FIGURE 5.

Influence of pH and temperature on MbR activity. Comparative preference of L₁₀ MbR for low pH (A) and high temperature (B). The RuBP-carboxylase activity of L₂ (○) and L₁₀ (●) MbR at 25 and 55 °C are alike. All of the assays were performed anaerobically.