The Role of ACT-Like Subdomain in Bacterial Threonine Dehydratases

In bacteria, threonine dehydratases could convert L-threonine to 2-ketobutyrate. Some threonine dehydratases contain only a catalytic domain, while others contain an N-terminal catalytic domain and a C-terminal regulatory domain composed of one or two ACT-like subdomains. However, the role of the ACT-like subdomain in threonine dehydratases is not clear. Here, nine different bacterial threonine dehydratases were studied. Three of the nine contain no ACT-like subdomain, four of them contain a single ACT-like subdomain, and two of them contain two ACT-like subdomains. The nine genes encoding these threonine dehydratases were individually overexpressed in E. coli BL21(DE3), and the enzymes were purified to homogeneity. Activities of the purified enzymes were analyzed after incubation at different temperatures and different pHs. The results showed that threonine dehydratases with a single ACT-like subdomain are more stable at higher temperatures and a broad range of pH than those without ACT-like subdomain or with two ACT-like subdomains. Furthermore, the specific activity of threonine dehydratases increases with the increase of the number of ACT-like subdomains they contain. The results suggest that the ACT-like subdomain plays an important role in bacterial threonine dehydratases.


Introduction
Bacterial threonine dehydratase (TD) converts L-threonine to 2ketobutyrate. There are usually two types of TD: biosynthetic threonine dehydratase (BTD) and catabolic threonine dehydratase (CTD) (Fig. 1). BTD functions in the biosynthetic pathway of Lisoleucine when bacteria grow under aerobic conditions, and usually contains an N-terminal catalytic domain and a C-terminal regulatory domain. CTD contains only the catalytic domain, and plays a role in the degradation of L-threonine to propionate when bacteria grow under anaerobic conditions [1], its activity is activated by AMP and CMP [2].
ACT (named after the first letters of three proteins, Aspartate kinase, Chorismate mutase and TyrA) is a structural motif in proteins that function in the control of metabolism, solute transport, and signal transduction [3]. BTD encoded by the gene ilvA in Escherichia coli contains two ACT-like subdomains in its Cterminal regulatory domain. The activity of the enzyme is inhibited by the end product L-isoleucine but activated by Lvaline, the product of a competing biosynthetic pathway [4]. BTD encoded by ilvA in Bacillus subtilis, however, contains only one ACT-like subdomain in its C-terminal regulatory domain, and is inhibited by L-isoleucine or by high concentrations of L-valine [5]. The interaction between different domains may facilitate enzyme foldingand increase stability [6]. Since BTD is a key enzyme for improving L-isoleucine production in bacterial fermentation [7][8][9], it is necessary to investigate the role of the ACT-like subdomain in the enzyme.
In this study, nine TDs containing different numbers of ACTlike subdomains were chosen from six different bacterial strains. The nine genes encoding these TDs were individually overexpressed in E. coli, and activities of the purified enzymes were analyzed after they were incubated at different temperatures and different pHs. The results indicate that the ACT-like subdomain plays an important role in the activity of TDs in bacteria.

Materials and Methods
Bacterial strains, plasmids and DNA manipulation All bacterial strains and plasmids used in this study are listed in Table 1. E. coli JM109 was used for constructing plasmids, while E. coli BL21 (DE3) was used for expressing TD enzymes. Plasmid pET28a was selected as the vector for protein expression. All E. coli cells were grown at 37uC in LB media (1% peptone, 0.5% yeast extract, 0.5% NaCl).
All the primers used for the PCR amplifications are listed Table 2. CgtdcB was amplified from the genomic DNA of C. glutamicum, using the primer pair of CgCTD-F and CgCTD-R; EctdcB was amplified from the genomic DNA of E. coli, using the primer pair of EcCTD-F and EcCTD-R; SatdcB was amplified from the genomic DNA of S. aureus, using the primer pair of SaCTD-F and SaCTD-R; CgilvA was amplified from the genomic DNA of C. glutamicum, using the primer pair of CgBTD1-F and CgBTD1-R; SailvA was amplified from the genomic DNA of S. aureus, using the primer pair of SaBTD1-F and SaBTD1-R; SgilvA was amplified from the genomic DNA of S. griseus, using the primer pair of SgBTD1-F and SgBTD1-R; BsilvA was amplified from the genomic DNA of B. subtilis, using the primer pair of BsBTD1-F and BsBTD1-R; EcilvA was amplified from the genomic DNA of E. coli, using the primer pair of EcBTD2-F and EcBTD2-R; PpilvA was amplified from the genomic DNA of P. putida, using the primer pair of PpBTD2-F and PpBTD2-R. All PCR amplifications were carried out in a 50 mL reaction mixture containing 5 mL 106Extaq buffer, 4 mL dNTP mixture (2.5 mM each), 1 mL template DNA, 1 mL forward primer (20 mM), 1 mL reverse primer (20 mM) and 0.5 mL Extaq. The PCR product was digested by restriction enzymes, ligated into pET28a which was similarly digested, transformed into E. coli JM109 and selected on LB plates containing 50 mg/mL kanamycin. The correct recombinant plasmids were then purified and transformed into E. coli BL21(DE3), resulting in nine E. coli strains: BL21(DE3)/pET28a-CgtdcB, BL21(DE3)/pET28a-EctdcB, BL21(DE3)/pET28a-SatdcB, BL21(DE3)/pET28a-CgilvA, BL21(DE3)/ pET28a-SailvA, BL21(DE3)/pET28a-SgilvA, BL21(DE3)/pET28a-BsilvA, BL21(DE3)/pET28a-EcilvA and BL21(DE3)/pET28a-PpilvA.
respectively. Each strain was grown in LB media at 25uC. Isopropyl-b-D-thiogalactopyranoside was added to a final concentration of 0.5 mM when OD 600 of the cells reached 0.6. The cells were further grown for 10 h and harvested by centrifugation at 37506g for 20 min. The cell pellets were resuspended in the binding buffer (20 mM HEPES, 500 mM NaCl, 20 mM imidazole, pH 8.0), and disrupted by an ultrasonic disintegrator in an icy bath for 15 min. Cell debris was removed by centrifugation at 130006g for 20 min, and the crude extracts were used for TD purification with a Ni-NTA Sepharose column (QIAGEN). The column was firstly washed with 10 volumes of the binding buffer, applied the crude extracts, washed again with 10 volumes of the binding buffer, and then the TD enzyme was eluted with 5 volumes of the elution buffer (20 mM HEPES, 500 mM NaCl,

Activity assays of TD enzymes at different temperatures and pH
Enzyme activities of the purified TD enzymes were measured according to the published method [11]. The 1 mL reaction mixture contains 100 mM potassium phosphate buffer, pH 8.0, 40 mM threonine, 1 mM pyridoxal phosphate and 2 mg enzymes. The reaction was started by adding L-threonine, incubated at 22uC for 15 min, and then terminated by adding 1 mL stopping solution (1% semicarbazide and 0.9% sodium acetate). The amount of 2-ketobutyrate in the reaction mixture was determined by measuring the absorption at 254 nm (e = 0.52 mmol/cm) using a UV/Vis spectrophotometer. One unit activity was defined as the enzyme amount required for the formation of 1 mole 2ketobutyrate per minute at 22uC, and the specific activity was defined as the total activity per mg protein.
To measure the activity at different temperatures, the purified enzymes were incubated at the specific temperature for 10 min, and then cooled on ice. The T 50 values were measured as described in the published method [11].
To analyze the enzyme activity at different pH, the purified enzyme was first diluted 10 times in acetic acid/sodium acetate buffer (pH 5), phosphate buffer (pH 6-8), or glycine sodium hydroxide buffer (pH 9-11), and incubated on ice for 30 min, before assaying the enzymes activity in 100 mM potassium phosphate buffer (pH 8.0).

Results and Discussion
In order to investigate the role of ACT-like subdomains in TD enzymes of bacteria, nine genes encoding TD enzymes in different bacteria were cloned in pET28a and overexpressed in E. coli. According to the analysis of CDART (Conserved Domain Architecture Retrieval Tool) [12], CgCTD, EcCTD and SaCTD contain no ACT-like subdomain; CgBTD1, SaBTD1, SgBTD1 and BsBTD1 all contain a single ACT-like subdomain; while EcBTD2 and PpBTD2 both contain two ACT-like subdomains ( Table 3). The size of the protein increases with the increase of the number of ACT-like subdomains it contains. Nine TD enzymes were overexpressed in E. coli and purified to homogeneity. The purity of the proteins was confirmed by SDS-PAGE analysis (Fig. 2). The migrating rates of the nine enzymes in the gel are consistent well with their theoretical size ( Table 1).
The activity of each TD enzymes was analyzed after the enzyme was incubated for 10 min at 30, 40, 43, 45, 47, 50, 53, 55, 57 and 60uC, respectively (Fig. 3). The activities of EcBTD2 and PpBTD2 significantly decreased with the increase of temperature, and their activities were totally lost when incubating at 47uC for 10 min. The activity loss of CgCTD, EcCTD and SaCTD showed the similar level to that of EcBTD2 and PpBTD2; the activities of CgBTD1, SaBTD1, SgBTD1 and BsBTD1, however, decreased much slower than that of EcBTD2 and PpBTD2 with the increase of temperature. The T 50 value was around 43uC for CgCTD, EcCTD, SaCTD, EcBTD2 and PpBTD2, but ranged between 47-55uC for CgBTD1, SaBTD1, SgBTD1 and BsBTD1 (Table 4). These data reveal that at higher temperatures BTD1 is usually more stable than CTD and BTD2, suggesting that the TD enzyme containing a single ACT-like subdomain is more stable at higher temperatures.
The specific activity at 30uC for each TD enzyme was also determined ( Table 4). The highest specific activity for CTD, BTD1 and BTD2 was 9.9, 89 and 171 U/mg, respectively, suggesting that ACT-like subdomains could increase the specific activity of TD enzyme. Interestingly, though the BTD2 has one more ACT-like subdomain, its stability at higher temperatures is lower than that of BTD1 and equal to that of CTDs. Except for the higher specific activity, structural flexibility might be another benefit of TDs containing multiple ACT-like subdomains such as BTD2. The mechanical flexibility is critical for the function of proteins [13], and may be a promoter for generating BTD2 to adapt bacteria for the more complex environment [14,15].
The activities of the nine TD enzymes were also analyzed after incubation for 30 min in buffers with different pH. For all nine TDs, the remaining activity after incubating at pH 8.0 was maximal and that incubation at various other pH values prior to assaying at pH 8.0 affected the activities of the TDs in the manner shown in Fig. 4. After incubation at pH ranges from 5 to 11, the remaining activities of CgBTD1, SaBTD1, SgBTD1 and BsBTD1 were more than 60%, and were higher than those of CgCTD, EcCTD, SaCTD, EcBTD2 and PpBTD2. After incubation at pH range from 6 to 9, all the nine TD enzymes maintained more than 50% of their activities. These data reveal that BTD1 with a single ACT-like subdomain could function better at a broader range of pH than CTD and BTD2.
ACT family domain is widely conserved in bacteria and evolutionarily mobile. It is always combined with other domains to provide easily regulated enzymes [3,16]. Most ACT domaincontaining proteins have been discovered in Bacteria and Archaea, indicating that the domain appeared a long time ago [6,17,18]. CTD exists not only in bacteria, but also in plants and yeast [19,20], suggesting that the pathway of L-threonine degradation may exist in the ancestral cell before the divergence of the three kingdoms. In the primordial cell where organic compounds were rich, catabolic pathways may have been more prevalent than biosynthetic pathways, therefore, only CTD may have been required for gaining energy under anaerobic conditions [17]. With the increase of the number of primordial cells, the prebiotic supply of amino acids may have been exhausted, and thus 2-ketobutyrate produced by CTD was possibly used for L-isoleucine biosynthesis. Therefore, BTD were created by combining CTD and ACT-like subdomain to better adapt the environment [21]. The hypothesis that multiple ACT domains promote TD flexibility requires further study to confirm.