**6.1 Function of diaminopimelate aminotransferase**

*Diaminopimelate aminotransferase* (*LL*-DAP-AT, EC 2.6.1.83) is a PLP-dependant enzyme that catalyses the conversion of *L*-2,3,4,5,-tetrahydrodipicolinate (THDP) to *L,L*-2,6 diaminopimelate (*LL*-DAP) (Fig. 1). This transamination reaction utilises glutamate as an amino donor to yield α-ketoglutarate. (Hudson et al., 2006, 2008; Liu et al., 2010; McCoy et al., 2006)

The enzyme was first isolated from plant and cyanobacterial species and thus demonstrated a new branch of the lysine biosynthesis pathway existed (Hudson et al., 2006). Although plants are known to synthesise lysine *de novo*, components of the pathway required for conversion of THDP to *meso*-DAP had not been identified previously despite years of investigation. Studies of crude cell extracts had shown that plants do not catalyse reactions specific to the succinylase, acetylase or dehydrogenase branches of the pathway. This was subsequently confirmed with the observation that annotated plant genomes, including that from *Arabidopsis thaliana*, lack some or all genes associated with the three classical branches (Chatterjee et al., 1994; Hudson et al., 2005). The identification and characterisation of *LL*-DAP-AT from *A. thaliana* demonstrated for the first time the means by which plant species catalyse the conversion of THDP to *meso*-DAP via the aminotransferase sub-pathway (Hudson et al., 2006).

More recently *LL*-DAP-AT has been identified in algal, archaeal and bacterial species including, *Chlamydia trachomatis* (McCoy et al., 2006)*, Chlamydomonas reihardtii* (Hudson et al., 2011)*, Methanocaldococcus jannaschii* (Liu et al., 2010), and *Protochlamydia amoebophila* (McCoy et al., 2006). Comparative genomic analyses shows that *LL*-DAP-AT is restricted to the eubacterial lineages, *Bacteroidetes*, *Chlamydiae*, *Chloroflexi*, *Cyanobacteria*, *Desulfuromonadales*, *Firmicutes*, and *Spirochaeta*; and the archaea, *Archaeoglobaceae* and *Methanobacteriaceae* (Hudson et al., 2008)*.* The phylogeny of *LL*-DAP-AT from these species has established the existence of two classes of *LL*-DAP-AT orthologues, namely, DapL1 and DapL2, which differ significantly in primary amino acid sequence. DapL1 and DapL2 are found predominantly in eubacteria and archaea, respectively (Hudson et al., 2008).

*LL*-DAP-AT enzymes are classified as members of the PLP-dependant protein superfamily of class I/II aminotransferases (Hudson et al., 2008; Jensen et al., 1996; Sung et al., 1991). Orthologues are in general 410 amino acids in length and can share as little as 29% sequence identity. Kinetic parameters for the *LL*-DAP-AT reaction have been determined for enzymes from a number of species, including *A. thaliana, C. trachomatis, Desulfitobacterium hafniense, Leptospira interrogans, Methanobacterium thermoautotrophicus, Morella thermoacetica,* and *P. amoebophila.* (Hudson et al., 2006, 2008; McCoy et al., 2006). In the human pathogen *C.*  *trachomatis*, the *K*M values for the substrates THDP and glutamate have been reported as 19 µM and 2.1 µM, respectively (Hudson et al., 2008).

### **6.2 Structure of** *LL***-DAP-AT**

At present, the PDB reports twelve *LL*-DAP-AT X-ray crystal structures from three species, namely, *A. thaliana, C. trachomatis and C. reihardtii* (Watanabe et al., 2007, 2008, 2011; Dobson et al., 2011)*.* The tertiary and quaternary structure of all three proteins are very similar with *LL*-DAP-AT existing as a homodimer (Fig. 14).

Fig. 14. Structure of dimeric *C. trachomatis LL*-DAP-AT. Monomers, indicated in blue and green, associate to form a functional dimer (PDB: 3ASA).

The subunit structure of *C. trachomatis LL*-DAP-AT is described as containing two domains, a large domain (LD) (residues 48-294) and a small domain (SD) (residues 1-47 and 295-394) (Watanabe et al., 2011; Watanabe & James, 2011). The LD is composed of α-β-α sandwich, whilst the SD assumes an α-β complex (Fig. 14). The LD is involved in binding PLP and also dimer formation, whereas the SD forms an N-terminal arm and also the C-terminal region. The active site is situated in a groove between the two domains of the monomer (Fig. 14). Importantly, the dimer structure is proposed to be essential for function as both subunits participate in substrate binding. Study of the structures of apo and ligand-bound forms of *C. trachomatis LL*-DAP-AT have revealed that the enzyme adopts an open and closed conformation (Watanabe et al., 2011). In the absence of ligand, the enzyme assumes an open state, whereby the active site is exposed to solvent. Upon PLP binding, the enzyme adopts a closed conformation. Within the active site, PLP is covalently linked to Lys236 via a Schiff base and is stabilised through an aromatic stacking interaction with Tyr128. PLP also forms a network of hydrogen bonding interactions with residues within the enzyme active site (Watanabe et al., 2011) (Fig 15).

*trachomatis*, the *K*M values for the substrates THDP and glutamate have been reported as 19

At present, the PDB reports twelve *LL*-DAP-AT X-ray crystal structures from three species, namely, *A. thaliana, C. trachomatis and C. reihardtii* (Watanabe et al., 2007, 2008, 2011; Dobson et al., 2011)*.* The tertiary and quaternary structure of all three proteins are very similar with

Fig. 14. Structure of dimeric *C. trachomatis LL*-DAP-AT. Monomers, indicated in blue and

The subunit structure of *C. trachomatis LL*-DAP-AT is described as containing two domains, a large domain (LD) (residues 48-294) and a small domain (SD) (residues 1-47 and 295-394) (Watanabe et al., 2011; Watanabe & James, 2011). The LD is composed of α-β-α sandwich, whilst the SD assumes an α-β complex (Fig. 14). The LD is involved in binding PLP and also dimer formation, whereas the SD forms an N-terminal arm and also the C-terminal region. The active site is situated in a groove between the two domains of the monomer (Fig. 14). Importantly, the dimer structure is proposed to be essential for function as both subunits participate in substrate binding. Study of the structures of apo and ligand-bound forms of *C. trachomatis LL*-DAP-AT have revealed that the enzyme adopts an open and closed conformation (Watanabe et al., 2011). In the absence of ligand, the enzyme assumes an open state, whereby the active site is exposed to solvent. Upon PLP binding, the enzyme adopts a closed conformation. Within the active site, PLP is covalently linked to Lys236 via a Schiff base and is stabilised through an aromatic stacking interaction with Tyr128. PLP also forms a network of hydrogen bonding interactions with residues within the enzyme active site

green, associate to form a functional dimer (PDB: 3ASA).

(Watanabe et al., 2011) (Fig 15).

µM and 2.1 µM, respectively (Hudson et al., 2008).

*LL*-DAP-AT existing as a homodimer (Fig. 14).

**6.2 Structure of** *LL***-DAP-AT** 

Fig. 15. Catalytic site of *LL*-DAP-AT from *C. trachomatis* (PDB: 3ASA). Ligand binding induces a closed conformation. PLP is covalently linked to Lys236 *via* a Schiff base.
