Discovery of long-chain polyamines embedded in the biosilica on the Bacillus cereus spore coat

Biosilicification is the process by which organisms incorporate soluble, monomeric silicic acid, Si(OH)4, in the form of polymerized insoluble silica, SiO2. Although the mechanisms underlying eukaryotic biosilicification have been intensively investigated, prokaryotic biosilicification has only recently begun to be studied. We previously reported that biosilicification occurs in the gram-positive, spore-forming bacterium Bacillus cereus, and that silica is intracellularly deposited on the spore coat as a protective coating against acids, although the underlying mechanism is not yet fully understood. In eukaryotic biosilicifying organisms, such as diatoms and siliceous sponges, several relevant biomolecules are embedded in biogenic silica (biosilica). These biomolecules include peptides, proteins, and long-chain polyamines. In this study, we isolated organic compounds embedded in B. cereus biosilica to investigate the biomolecules involved in the prokaryotic biosilicification process and identified long-chain polyamines with a chemical structure of H2N-(CH2)4-[NH-(CH2)3]n-NH2 (n: up to 55). Our results demonstrate the common presence of long-chain polyamines in different evolutionary lineages of biosilicifying organisms, i.e., diatoms, siliceous sponges, and B. cereus, suggesting a common mechanism underlying eukaryotic and prokaryotic biosilicification.

The C-terminal flexible region of branched-chain polyamine synthase facilitates substrate specificity and catalysis

Branched-chain polyamine synthase (BpsA) catalyzes sequential aminopropyl transfer from the donor, decarboxylated S-adenosylmethionine (dcSAM), to the acceptor, linear-chain polyamine, resulting in the production of a quaternary-branched polyamine via tertiary branched polyamine intermediates. Here, we analyzed the catalytic properties and X-ray crystal structure of Tth-BpsA from Thermus thermophilus and compared them with those of Tk-BpsA from Thermococcus kodakarensis, which revealed differences in acceptor substrate specificity and C-terminal structure between these two enzymes. To investigate the role of the C-terminal flexible region in acceptor recognition, a region (QDEEATTY) in Tth-BpsA was replaced with that in Tk-BpsA (YDDEESSTT) to create chimeric Tth-BpsA C9, which showed a severe reduction in catalytic efficiency toward N⁴ -aminopropylnorspermidine, but not toward N⁴ -aminopropylspermidine, mimicking Tk-BpsA substrate specificity. Tth-BpsA C9 Tyr³⁴⁶ and Thr³⁵⁴ contributed to discrimination between tertiary branched-chain polyamine substrates, suggesting that the C-terminal region of BpsA recognizes acceptor substrates. Liquid chromatography-tandem mass spectrometry analysis on a Tk-BpsA reaction mixture with dcSAM revealed two aminopropyl groups bound to two of five aspartate/glutamate residues (Glu³³⁹ , Asp³⁴² , Asp³⁴³ , Glu³⁴⁴ , and Glu³⁴⁵ ) in the C-terminal flexible region. Mutating each of these five amino acid residues to asparagine/glutamine resulted in a slight decrease in activity. The quadruple mutant D342N/D343N/E344Q/E345Q exhibited a severe reduction in catalytic efficiency, suggesting that these aspartate/glutamate residues function to receive aminopropyl chains. In addition, the X-ray crystal structure of the Tk-BpsA ternary complex bound to N⁴ -bis(aminopropyl)spermidine revealed that Asp¹²⁶ and Glu²⁵⁹ interacted with the aminopropyl moiety in N⁴ -aminopropylspermidine.