**2.1 Function of DHDPS**

*Dihydrodipicolinate synthase* (DHDPS, EC 4.2.1.52) was first purified in 1965 from *E. coli* extracts (Yugari & Gilvarg, 1965). The enzyme is the product of the *dap*A gene, which has been shown to be essential in several bacterial species (Dogovski et al., 2009; Hutton et al., 2007). The *dap*A product, DHDPS, catalyses the condensation of pyruvate (PYR) and aspartate semialdehyde (ASA) to form 4-hydroxy-2,3,4,5-tetrahydro-*L,L*-dipicolinic acid (HTPA) (Fig. 1). It was first suggested that the product released by DHDPS was dihydrodipicolinate (DHDP), but studies using 13C-labelled pyruvate support the view that the product is the unstable heterocycle HTPA (Blickling et al. 1997a). Rapid decomposition of the 13C-NMR signals of HTPA following its production indicate that formation of DHDP occurs via a nonenzymatic step.

In all cases examined, the DHDPS-catalysed reaction proceeds via a ping-pong kinetic mechanism in which pyruvate binds the active site, resulting in the release of a protonated water molecule. ASA then binds and is condensed with pyruvate to form the heterocyclic product, HTPA (Blickling et al. 1997a).

In the first step of the mechanism, the active site lysine, (Lys161 in *E. coli* DHDPS) forms a Schiff base with pyruvate (Laber et al., 1992) (Fig 2). Formation of the Schiff base proceeds through a tetrahedral intermediate. It is proposed that a catalytic triad of three residues - Tyr133, Thr44 and Tyr107 (*E. coli* numbering) - act as a proton relay to transfer protons to and from the active site via a water-filled channel leading to bulk solvent (Dobson et al., 2004a). The Schiff base (imine) is converted to its enamine form, which then adds to the aldehyde group of ASA (Blickling et al., 1997a; Dobson et al., 2008). In aqueous solution, ASA is known to exist in the hydrated form rather than the aldehyde, but the biologicallyrelevant form of the substrate remains to be determined. HTPA is then formed by nucleophilic attack of the amino group of ASA onto the intermediate imine, leading to cyclisation and detachment of the product from the enzyme, with release of the active site lysine residue (Fig 2).

Fig. 2. The catalytic mechanism of DHDPS.

#### **2.1.1 Regulation of DHDPS activity**

In some organisms, the activity of DHDPS is regulated allosterically by lysine via a classical feedback inhibition process. Lysine feedback inhibition of DHDPS has been investigated in several plant, Gram-negative and Gram-positive bacterial species to date. Studies involving *Daucus carota sativa* (Matthews et al., 1979), *Pivus sativum* (Dereppe et al., 1992), *Spinacia aloeracea* (Wallsgrove et al., 1980), *Triticum aestivium* (Kumpaisal et al., 1987), and *Zea mays* (Frisch et al., 1991) show that DHDPS from plant species are generally strongly inhibited by

through a tetrahedral intermediate. It is proposed that a catalytic triad of three residues - Tyr133, Thr44 and Tyr107 (*E. coli* numbering) - act as a proton relay to transfer protons to and from the active site via a water-filled channel leading to bulk solvent (Dobson et al., 2004a). The Schiff base (imine) is converted to its enamine form, which then adds to the aldehyde group of ASA (Blickling et al., 1997a; Dobson et al., 2008). In aqueous solution, ASA is known to exist in the hydrated form rather than the aldehyde, but the biologicallyrelevant form of the substrate remains to be determined. HTPA is then formed by nucleophilic attack of the amino group of ASA onto the intermediate imine, leading to cyclisation and detachment of the product from the enzyme, with release of the active site

In some organisms, the activity of DHDPS is regulated allosterically by lysine via a classical feedback inhibition process. Lysine feedback inhibition of DHDPS has been investigated in several plant, Gram-negative and Gram-positive bacterial species to date. Studies involving *Daucus carota sativa* (Matthews et al., 1979), *Pivus sativum* (Dereppe et al., 1992), *Spinacia aloeracea* (Wallsgrove et al., 1980), *Triticum aestivium* (Kumpaisal et al., 1987), and *Zea mays* (Frisch et al., 1991) show that DHDPS from plant species are generally strongly inhibited by

lysine residue (Fig 2).

Fig. 2. The catalytic mechanism of DHDPS.

**2.1.1 Regulation of DHDPS activity** 

lysine (IC50 = 0.01-0.05 mM). In contrast, DHDPS from bacteria are significantly less sensitive to lysine inhibition than their plant counterparts. For example, DHDPS from Gram-negative bacteria, such as *E. coli* (Dobson et al., 2005a; Yugari and Gilvarg, 1965), *Niesseria meningitidis* (Devenish et al., 2009), and *Sinorhizobium. meliloti* (Phenix & Palmer, 2008), display IC50 values that range from 0.25 mM to 1.0 mM. Whereas, the enzyme from Gram-positive bacteria such as *Bacillus anthracis* (Domigan et al., 2009), *Bacillus cereus* (Hoganson & Stahly, 1975), *Corynebacterium glutamicum* (Cremer et al., 1988), *Lactobacillus plantarum* (Cahyanto et al., 2006) and *Staphylococcus aureus* (Burgess et al., 2008) show little or no inhibition by lysine.

The crystal structure of DHDPS in complex with lysine from *E. coli* shows that the lysine allosteric binding site is situated in a crevice at the interface of the tight dimer, distal from the active site, but connected to the active site via a water channel (Blickling et al., 1997a). Two inhibitory lysine molecules are bound in close proximity within van der Waals contact to each other. Seven residues located within the allosteric site bind lysine, namely Ala49, His53, His56, Gly78, Asp80, Glu84, and Tyr106 (Blickling et al., 1997a).

Studies show that lysine inhibition is cooperative with the second lysine molecule binding 105 times more tightly than the first (Blickling et al., 1997a). The mechanism by which lysine exerts regulatory control over bacterial DHDPS is not well understood, although kinetic and structural studies suggest that it is an allosteric inhibitor, causing partial inhibition (approximately 90%) at saturating concentrations (Blickling et al., 1997a). It has recently been suggested that lysine exerts some effect on the first half reaction by attenuating proton-relay and also the function of Arg138, thought to be crucial for ASA binding (Dobson et al., 2004b). The crystal structure of the *E. coli* DHDPS-lysine complex was solved in the absence of substrate; however, thermodynamic studies have illustrated that the substrate pyruvate has a substantial effect on the nature of enzyme-inhibitor association (Blickling et al., 1997a).
