Crystal structure and biochemical analyses reveal that the Arabidopsis triphosphate tunnel metalloenzyme AtTTM3 is a tripolyphosphatase involved in root development

Abstract

SUMMARYThe Arabidopsis protein AtTTM3 belongs to the CYTH superfamily named after its two founding members, the CyaB adenylate cyclase from Aeromonas hydrophila and the mammalian thiamine triphosphatase. In this study we report the three-dimensional structure of a plant CYTH domain protein, AtTTM3, determined at 1.9 A resolution. The crystal structure revealed the characteristic tunnel architecture of CYTH proteins, which specialize in the binding of nucleotides and other organic phosphates and in phosphoryl transfer reactions. The b barrel is composed of eight antiparallel b strands with a cluster of conserved inwardly facing acidic and basic amino acid residues. Mutagenesis of these residues in the catalytic core led to an almost complete loss of enzymatic activity. We established that AtTTM3 is not an adenylate cyclase. Instead, the enzyme displayed weak NTP phosphatase as well as strong tripolyphosphatase activities similar to the triphosphate tunnel metalloenzyme proteins from Clostridium thermocellum (CthTTM) and Nitrosomonas europaea (NeuTTM). AtTTM3 is most highly expressed in the proximal meristematic zone of the plant root. Furthermore, an AtTTM3 T-DNA insertion knockout line displayed a delay in root growth as well as reduced length and number of lateral roots, suggesting a role for AtTTM3 in root development.

Triphosphate Tunnel Metalloenzyme Function in Senescence Highlights a Biological Diversification of This Protein Superfamily

“…We have previously characterized AtTTM3 as a tripolyphosphatase with a potential function in root development (Moeder et al, 2013) and AtTTM2 as a negative regulator of the SA amplification loop during pathogen resistance responses (Ung et al, 2014). Knockout lines of AtTTM1, which shows 65% identity and 92% similarity to AtTTM2 at the amino acid level, did not display enhanced disease resistance, but rather showed a delayed senescence phenotype, as was expected based on our analysis of coexpressed genes.…”

“…Both AtTTM1 and 2 displayed low activity on ATP, ADP, or PPP i substrates, while high affinity was observed for PP i with a K m of 17 mM, which is comparable to the K m of AtTTM3 or NeuTTM for PPP i (43 mM and 21 mM, respectively; Moeder et al, 2013;Delvaux et al, 2011). We tested several biologically relevant diphosphates such as thiamine diphosphate (Jordan, 2007), ADP ribose (Adams-Phillips et al, 2010) and NADH (Ishikawa et al, 2010), all of which were not hydrolyzed by AtTTM1 and AtTTM2 (Ung and Yoshioka, unpublished data).…”

“…However, all three recombinantly expressed TTM proteins from Arabidopsis did not display adenylate cyclase activity (Moeder et al, 2013;Ung et al, 2014). Rather, we previously demonstrated that AtTTM3 displays strong tripolyphosphatase activity with a weaker affinity for nucleotide triphosphates (Moeder et al, 2013).…”

“…Detection of free phosphates was performed as previously described (Bernal et al, 2005;Moeder et al, 2013).…”

“…S1). Previously, we demonstrated that AtTTM3 exhibits tripolyphosphatase activity and may play a role in root development (Moeder et al, 2013), whereas AtTTM2 displays pyrophosphatase (PPase) activity and is a negative regulator of pathogen defense responses (Ung et al, 2014). Interestingly, publicly available microarray data show significantly different expression patterns for AtTTM1 and AtTTM2.…”