Monothiol and dithiol glutaredoxin-1 from Clostridium oremlandii: identification of domain-swapped structures by NMR, X-ray crystallography and HDX mass spectrometry

Abstract

Protein dimerization or oligomerization resulting from swapping part of the protein between neighboring polypeptide chains is known to play a key role in the regulation of protein function and in the formation of protein aggregates. Glutaredoxin-1 from Clostridium oremlandii (cGrx1) was used as a model to explore the formation of multiple domain-swapped conformations, which were made possible by modulating several hinge-loop residues that can form a pivot for domain swapping. Specifically, two alternative domain-swapped structures were generated and analyzed using nuclear magnetic resonance (NMR), X-ray crystallography, circular-dichroism spectroscopy and hydrogen/deuterium-exchange (HDX) mass spectrometry. The first domain-swapped structure (β3-swap) was formed by the hexameric cGrx1-cMsrA complex. The second domain-swapped structure (β1-swap) was formed by monothiol cGrx1 (C16S) alone. In summary, the first domain-swapped structure of an oxidoreductase in a hetero-oligomeric complex is presented. In particular, a single point mutation of a key cysteine residue to serine led to the formation of an intramolecular disulfide bond, as opposed to an intermolecular disulfide bond, and resulted in modulation of the underlying free-energy landscape of protein oligomerization.

Keywords: disulfide bonds; domain swapping; glutaredoxin; oxidoreductases.

Figure 1

Structures of d-cGrx1, m-cGrx1 and the m-cGrx1–cMsrA complex. (a) d-cGrx1 is shown in cyan (left) and the domain-swapped structures of m-cGrx are shown in magenta (β3-swap; middle) and orange (β1-swap; right). (b) The heterohexameric state contains two cMsrA molecules and four cGrx1 molecules: cMsrA is in green, a domain-swapped dimer of cGrx1 is in blue and purple, and a disulfide dimer of cGrx1 is in beige. The catalytic pocket (CP) of cMsrA is indicated by a yellow circle. (c) Diagram of the heterohexamer of the monothiol cGrx1–cMsrA complex. The interfaces between cGrx1 and cMsrA are labeled in translucent boxes and disulfide bridges are shown as red lines. (d) The monomeric d-cGrx1 consists of three α-helices (α1, α2 and α3) and four β-sheets (β1, β2, β3 and β4). The N- and C-termini of d-cGrx1 are labeled N and C, respectively. Cys13 and Cys16 form a disulfide bond, which is represented as a stick.

Figure 2

β3 domain-swapped structure of m-cGrx1. (a) Overall tetrameric structure of m-cGrx1. The disulfide dimers of m-cGrx1 are shown in beige (subunits A and C) and the domain-swapped dimers of m-cGrx1 are shown in purple and blue (subunits B and D). The red arrow indicates an intermolecular disulfide bridge between subunits A and B. (b) The domain-swapped m-cGrx1 monomers exchange β3, β4 and α3 with one another and are linked by a hinge loop. (c) The β3 domain-swapped m-cGrx1 structure (blue) reveals that the tertiary structure is unwound compared with that of d-cGrx1 (beige) and that the hinge loop is located between α2 and β3. (d) Comparison of the catalytic cysteine (Cys13) between domain-swapped m-cGrx1 and d-cGrx1. The superimposed domain-swapped m-cGrx1 (blue) and d-cGrx1 (beige) show that the catalytic cysteines (Cys13) are oriented differently. A disulfide dimeric m-cGrx1 is shown in gray. A 2F _o − F _c electron-density map is shown at the disulfide bridge. The domain-swapped m-cGrx1 (blue) and disulfide dimeric m-cGrx1 form a disulfide bridge. The 2F _o − F _c electron-density map is shown at 2.0σ.

Figure 3

2D [¹H,¹⁵N]-HSQC spectra of d-cGrx1 and m-cGrx1 recorded at 298 K on a Bruker Avance 800 MHz NMR spectrometer. (a) d-cGrx1. (b) m-cGrx1. (c, d) 2D [¹H,¹⁵N]-HSQC spectra of cGrx1 with and without DTT. (c) d-cGrx1 in the absence of DTT is shown in blue and d-cGrx1 in the presence of DTT (5 mM) is shown in red.(d) m-cGrx1 in the absence of DTT is shown in blue and m-cGrx1 in the presence of DTT (5 mM) is shown in red.

Figure 4

β1 domain-swapped structure of m-cGrx1. (a) The domain-swapped m-cGrx1 molecules (subunits B and C) are shown in red and orange and the disulfide dimeric m-cGrx1 molecules (subunits A and D) are shown in gray and black. The red arrow indicates the intermolecular disulfide bridge between subunits A and C. (b) The hinge loop is located between β1 and α1. (c) Structural comparison of d-cGrx1 and β1 domain-swapped m-cGrx1. The β1 domain-swapped m-cGrx1 (orange) reveals that the tertiary structure is unwound compared with that of d-cGrx1 (beige) and the hinge loop is located between α1 and β1. The α1 helix is partially unfolded.

Figure 5

Deuterium uptake of d-cGrx1 and m-cGrx1. (a) The monomeric structures of d-cGrx1 and m-cGrx1 colored according to the differences in relative deuteration levels. Shades of red or blue reflect higher or lower deuterium uptake in the presence and absence of TCEP, respectively. (b, c) Uptake curves of selected peptides of d-cGrx1 and m-cGrx1. The observed relative deuterium uptake for each peptide, time point and condition were calculated and plotted against the labeling time. Error bars represent the average standard deviation observed across time points and replicates.