Solution structure and assembly of β-amylase 2 from Arabidopsis thaliana

Abstract

Starch is a key energy-storage molecule in plants that requires controlled synthesis and breakdown for effective plant growth. β-Amylases (BAMs) hydrolyze starch into maltose to help to meet the metabolic needs of the plant. In the model plant Arabidopsis thaliana there are nine BAMs, which have apparently distinct functional and domain structures, although the functions of only a few of the BAMs are known and there are no 3D structures of BAMs from this organism. Recently, AtBAM2 was proposed to form a tetramer based on chromatography and activity assays of mutants; however, there was no direct observation of this tetramer. Here, small-angle X-ray scattering data were collected from AtBAM2 and its N-terminal truncations to describe the structure and assembly of the tetramer. Comparison of the scattering of the AtBAM2 tetramer with data collected from sweet potato (Ipomoea batatas) BAM5, which is also reported to form a tetramer, showed there were differences in the overall assembly. Analysis of the N-terminal truncations of AtBAM2 identified a loop sequence found only in BAM2 orthologs that appears to be critical for AtBAM2 tetramer assembly as well as for activity.

Keywords: Arabidopsis thaliana; amylases; small-angle X-ray scattering; tetramer; β-amylase 2.

Figure 1

Small-angle scattering data for AtBAM2. The inset shows the R _g values calculated from the Guinier region (gray dots) or from the P(r) fit (blue dots). Error bars show the error from the fit. Protein concentrations were 30 µM (lightest gray), 50 µM (gray) and 100 µM (black) AtBAM2.

Figure 2

Analysis of the structure of AtBAM2. (a) Kratky analysis of merged AtBAM2 scattering data. (b) P(r) plot of AtBAM2. D _max was 126 Å, R _g was 42.6 ± 0.03 Å and the elongation ratio (ER) was 0.75. The total estimate for the fit was 0.79 with an α of 39.8. (c) Comparison of the AtBAM2 P(r) plot with the theoretical P(r) plots for a tetramer (red or black), dimer (blue or magenta) or monomer (green). (d) Comparison of the AtBAM2 P(r) plot with that of IbBAM5 from SASBDB (entry SASDA62). (e) Model of the AtBAM2 tetramer

Figure 3

Prediction of the 3D shape of AtBAM2. (a) Fit of the model in Fig. 2 ▸(e) to AtBAM2 from the FoXS server. (b) DAMMIN, (c) DAMMIF and (d) GASBOR dummy-atom models based on AtBAM2 data aligned with a tetrameric model of AtBAM2 (red). The fits of the dummy-atom models to the data are shown in the right panels of (b), (c) and (d) along with the χ² of the fit. The normalized spatial discrepancy (NSD) of the 13 DAMMIF models was 1.299, with a standard deviation of 0.141.

Figure 4

SAXS analysis of AtBAM2 with the Ndel1 or Ndel2 truncation. (a) Size-exclusion chromatography of the wild type (black), Ndel1 (magenta) and Ndel2 (blue) collected before SAXS analysis. (b) Scattering data for Ndel1 (magenta) and Ndel2 (blue). (c) P(r) plot for Ndel1 and Ndel2. The wild-type AtBAM2 data are shown by a dashed black line. For Ndel1, D _max was 109.8 Å and R _g was 40.6 ± 0.05 Å. The total estimate for the fit was 0.84, with an α of 2125. For Ndel2, D _max was 225.5 Å and R _g was 65.4 ± 0.4 Å. The total estimate for the fit was 0.75, with an α of 6.1.

Figure 5

Activity of AtBAM2 with the Ndel1 or Ndel2 truncation. (a) AtBAM model showing the amino acids that are removed in the Ndel1 truncation (magenta) and the Ndel2 truncation (magenta and blue). (b) Activity of AtBAM2 and the Ndel1 and Ndel2 truncated enzymes on soluble starch with (black) and without KCl (gray). Each point is an independent measurement, with three trials shown for each enzyme.

Figure 6

Regions of a sequence alignment of BAM sequences from a charophyte green alga, various land plants and a bacterium, illustrating the conservation of residues identified as important for the catalytic activity of Arabidopsis BAM2 (Monroe et al., 2017, 2018 ▸). Above the horizontal lines are BAM2-like sequences from the charophyte green alga Klebsormidium flaccidum (kfl00081_0270), the nonvascular plants Physcomitrella patens (XP_024360526.1) and Marchantia polymorpha (OAE23062.1), the seedless vascular plant Selaginella moellendorffii (XP_024540821.1), the gymnosperm Araucaria cunninghamii (JAG96982.1) and the flowering plants Erythranthe guttata (XP_012843727.1), Ricinus communis (XP_002511858.1), Fragaria vesca (XP_004306786.1) and Arabidopsis thaliana (NP_191958.3). Below the horizontal lines are other catalytically active BAMs from Arabidopsis, including BAM1 (NP_189034.1), BAM3 (NP_197368.1), BAM6 (NP_180788.2) and BAM5 (NP_567460.1), as well as BAM5 from sweet potato (Ipomoea batatas; XP_019180769.1) and a BAM from Thermoanaerobacterium thermosulfurigenes (P19584). The full-length sequence alignment is shown in Supplementary Fig. S1.

Figure 7

Model of the AtBAM2 interface showing the proximity of the Ndel1 (magenta) and Ndel2 (magenta and blue) truncation sites to Trp456, Arg490 and Ser464 (shown in yellow).