**4. Succinylase pathway**

#### **4.1 Tetrahydrodipicolinate N-succinyltransferase**

*Tetrahydrodipicolinate N-succinyltransferase* (THPC-NST, EC 2.3.1.117) is a succinyl-coenzyme A (SCoA) dependant enzyme that catalyses the conversion of cyclic *L*-2,3,4,5, tetrahydrodipicolinate (THDP) to acyclic N-succinyl-*L*-2-amino-6-ketopimelate (NSAKP) (Simms et al., 1984) (Fig. 1). The reaction occurs via a *L-*2-amino-6-ketopimelate (AKP) intermediate. The transfer of an acyl group functions to maintain a linear conformation of the product of the reaction (NSKAP) and exposes the 6-keto group for subsequent transamination (Beaman et al., 2002). Substrate and cofactor kinetic parameters for *E. coli* THPC-NST have been determined. Studies show that the *K*Mapp for THDP and succinyl-CoA are 20 µM and 15 µM, respectively (Berges et al., 1986b; Simms et al., 1984).

The *dap*D gene encoding THPC-NST is found in a large number of bacterial species including *E. coli* and *Mycobacterium* species (Beaman et al., 1997; Richaud et al., 1984; Schuldt et al., 2009). Expression of this gene in *E. coli* is weakly inhibited by lysine (Ou et al., 2008; Richaud et al., 1984). THPC-NST enzymes characterised to date are comprised of approximately 290 residues and show greater than 18% sequence identity (Beaman et al., 1997; Richaud et al., 1984; Schuldt et al., 2009).

The crystal structure of THPC-NST from *Mycobacterium bovis* (Fig. 11) shows that the enzyme forms a homotrimer. The monomer consists of three domains, namely, the (i) Nterminal, (ii) left handed parallel β-helix (LβH), and (iii) C-terminal domains (Beaman et al., 1997). The N-terminal domain is comprised of four α-helices and two hairpin loops. The LβH domain, comprising 50% of the subunit, contains the hexapeptide repeat motif ([LIV]- [GAED]-X2-[STAV]-X) within each turn of the β-helix. The LβH domain is interrupted by two loops, including a flexible loop (residues 166-175) that is involved in binding substrate. The C-terminal domain consists of a β-stranded structure. All three domains contribute to inter-subunit contacts. The structure of THPC-NST from other bacterial species have since been determined and show a high degree of similarity to that of *M. bovis* THPC-NST (Nguyen et al., 2008; Schuldt et al., 2009).

*Tetrahydrodipicolinate N-succinyltransferase* (THPC-NST, EC 2.3.1.117) is a succinyl-coenzyme A (SCoA) dependant enzyme that catalyses the conversion of cyclic *L*-2,3,4,5, tetrahydrodipicolinate (THDP) to acyclic N-succinyl-*L*-2-amino-6-ketopimelate (NSAKP) (Simms et al., 1984) (Fig. 1). The reaction occurs via a *L-*2-amino-6-ketopimelate (AKP) intermediate. The transfer of an acyl group functions to maintain a linear conformation of the product of the reaction (NSKAP) and exposes the 6-keto group for subsequent transamination (Beaman et al., 2002). Substrate and cofactor kinetic parameters for *E. coli* THPC-NST have been determined. Studies show that the *K*Mapp for THDP and succinyl-CoA

The *dap*D gene encoding THPC-NST is found in a large number of bacterial species including *E. coli* and *Mycobacterium* species (Beaman et al., 1997; Richaud et al., 1984; Schuldt et al., 2009). Expression of this gene in *E. coli* is weakly inhibited by lysine (Ou et al., 2008; Richaud et al., 1984). THPC-NST enzymes characterised to date are comprised of approximately 290 residues and show greater than 18% sequence identity (Beaman et al.,

The crystal structure of THPC-NST from *Mycobacterium bovis* (Fig. 11) shows that the enzyme forms a homotrimer. The monomer consists of three domains, namely, the (i) Nterminal, (ii) left handed parallel β-helix (LβH), and (iii) C-terminal domains (Beaman et al., 1997). The N-terminal domain is comprised of four α-helices and two hairpin loops. The LβH domain, comprising 50% of the subunit, contains the hexapeptide repeat motif ([LIV]- [GAED]-X2-[STAV]-X) within each turn of the β-helix. The LβH domain is interrupted by two loops, including a flexible loop (residues 166-175) that is involved in binding substrate. The C-terminal domain consists of a β-stranded structure. All three domains contribute to inter-subunit contacts. The structure of THPC-NST from other bacterial species have since been determined and show a high degree of similarity to that of *M. bovis* THPC-NST

are 20 µM and 15 µM, respectively (Berges et al., 1986b; Simms et al., 1984).

Fig. 10. Inhibitors of DHDPR.

**4. Succinylase pathway** 

**4.1 Tetrahydrodipicolinate N-succinyltransferase** 

1997; Richaud et al., 1984; Schuldt et al., 2009).

(Nguyen et al., 2008; Schuldt et al., 2009).

Fig. 11. Structure of trimeric *M. bovis* THPC-NST in complex with *L*-2-aminopimelate and succinamide-CoA. The N-terminal (orange), LβH (blue) and C-terminal (green) domains are indicated. The substrate *L*-2-aminopimelate (yellow) and cofactor succinamide-CoA (yellow) are bound via the THPC-NST active site residues (pink) (PDB: 1KGQ).

Crystal structures of *M. bovis* THPC-NST in complex with substrate analogs and several forms of coenzyme A have resulted in a model describing substrate binding and catalysis (Beaman et al., 1998, 2002). Self-association of the monomer subunit results in a homotrimer complex containing three active sites. The AKP and SCoA binding sites are located at the LβH domain interfaces. Binding of SCoA and possibly AKP is thought to promote a large conformational change that encloses the bound substrate and cofactor within the active site. In this state, the 2-amino group of AKP is placed in close proximity to the SCoA thioester, allowing nucleophilic attack and transfer of the succinyl group (Beaman et al., 2002).

Studies have shown that *L*-2-aminopimelic acid, an analog of AKP, is an inhibitor of THPC-NST, although it does not display antibacterial activity (Berges et al., 1986a). However, peptide derivatives of 2-aminopimelic acid show significant antibacterial activity against a range of Gram-negative bacteria (Berges et al., 1986a).

### **4.2 N-succinyldiaminopimelate aminotransferase**

*N-succinyldiaminopimelate aminotransferase* (NSDAP-AT, EC 2.6.1.17) catalyses the conversion of NSKAP to N-succinyl-*L,L*-2,6,-diaminopimelate (NSDAP) (Fig. 1). The reaction begins by the formation of a Schiff base linkage between an active site lysine and the cofactor pyridoxal-5'-phosphate (PLP). An amino group, donated by glutamate, is transferred to PLP, to form pyridoxamine phosphate (PMP). The enzyme subsequently transfers the amino group from PMP to NSAKP to yield N-succinyl-*L,L*-2,6,-diaminopimelate (NSDAP) and αketoglutarate (Peterkofsky & Gilvarg., 1961; Ledwidge & Blanchard., 1999). Studies of *E. coli* NSDAP-AT report *K*M values for the substrates NSKAP and glutamate of 0.5 mM and 0.52 mM, respectively (Peterkofsky & Gilvarg., 1961).

The gene encoding NSDAP-AT (*dapC*), is found in a large number of bacterial species including *Bordetella pertussis* (Fuchs et al., 2000), *C. glutamicum* (Hartmann et al., 2003)*, E. coli,* (Peterkofsky & Gilvarg., 1961) and *M. tuberculosis* (Weyand et al., 2006). In *E. coli,* the gene encoding NSDAP-AT is annotated *argD* (Ledwidge & Blanchard., 1999). This enzyme also functions as a N-acetylornithine aminotransferase, a component of the arginine biosynthesis pathway. The *dapC* gene in *B. pertussis* (Fuchs et al., 2000)*, C. glutamicum,*  (Hartmann et al., 2003)*,* and *E. coli* (Bukari & Taylor., 1971) has been found to map in close proximity to the *dapD* gene on the chromosome. Sequence analyses have shown that NSDAP-AT consists of approximately 400 residues and shares greater than 26% identity across species (Fuchs et al., 2000; Hartmann et al., 2003; Peterkofsky & Gilvarg., 1961; Weyand et al., 2006). The NSDAP-AT sequence is characterised by the presence of the PLP binding sequence motif, SLSKXSNVXGXRAG, that includes an active site lysine residue (underlined) (Fuchs et al., 2000).

Structure studies of *M. tuberculosis* NSDAP-AT in complex with PLP shows that the enzyme forms a homodimer (Fig. 12). The structure is characteristic of the aminotransferase family of class I PLP-binding proteins (Weyand et al., 2007). The monomer subunit is comprised of (i) an α-helical N-terminal extension, (ii) a central domain comprising an 8-stranded β-sheet surrounded by 8 α-helices, and (iii) a Cterminal domain consisting of a four stranded β-sheet flanked by 4 α-helices. The active site of each subunit is located at the dimer interface with residues from both subunits contributing to the architecture of the active sites. PLP is bound to the active site Lys232, presumably via a Schiff base, and makes a number of noncovalent contacts with other residues within the active site via a hydrogen bond network.

A number of hydrazino-dipeptide analogs of NSDAP inhibit NSDAP-AT with *K*i values ranging from 22-556 nM and show significant antibacterial activity against *E. coli* (Cox et al., 1998).

Fig. 12. Structure of dimeric *M. tuberculosis* NSDAP-AT in complex with PLP. The α-helical Nterminal extension (orange), central (blue) and C-terminal (green) domains are indicated. The cofactor PLP (yellow) is bound by the NSDAP-AT active site residues (pink) (PDB: 2O0R).

The gene encoding NSDAP-AT (*dapC*), is found in a large number of bacterial species including *Bordetella pertussis* (Fuchs et al., 2000), *C. glutamicum* (Hartmann et al., 2003)*, E. coli,* (Peterkofsky & Gilvarg., 1961) and *M. tuberculosis* (Weyand et al., 2006). In *E. coli,* the gene encoding NSDAP-AT is annotated *argD* (Ledwidge & Blanchard., 1999). This enzyme also functions as a N-acetylornithine aminotransferase, a component of the arginine biosynthesis pathway. The *dapC* gene in *B. pertussis* (Fuchs et al., 2000)*, C. glutamicum,*  (Hartmann et al., 2003)*,* and *E. coli* (Bukari & Taylor., 1971) has been found to map in close proximity to the *dapD* gene on the chromosome. Sequence analyses have shown that NSDAP-AT consists of approximately 400 residues and shares greater than 26% identity across species (Fuchs et al., 2000; Hartmann et al., 2003; Peterkofsky & Gilvarg., 1961; Weyand et al., 2006). The NSDAP-AT sequence is characterised by the presence of the PLP binding sequence motif, SLSKXSNVXGXRAG, that includes an active site lysine residue

Structure studies of *M. tuberculosis* NSDAP-AT in complex with PLP shows that the enzyme forms a homodimer (Fig. 12). The structure is characteristic of the aminotransferase family of class I PLP-binding proteins (Weyand et al., 2007). The monomer subunit is comprised of (i) an α-helical N-terminal extension, (ii) a central domain comprising an 8-stranded β-sheet surrounded by 8 α-helices, and (iii) a Cterminal domain consisting of a four stranded β-sheet flanked by 4 α-helices. The active site of each subunit is located at the dimer interface with residues from both subunits contributing to the architecture of the active sites. PLP is bound to the active site Lys232, presumably via a Schiff base, and makes a number of noncovalent contacts with other

A number of hydrazino-dipeptide analogs of NSDAP inhibit NSDAP-AT with *K*i values ranging from 22-556 nM and show significant antibacterial activity against *E. coli* (Cox et

Fig. 12. Structure of dimeric *M. tuberculosis* NSDAP-AT in complex with PLP. The α-helical Nterminal extension (orange), central (blue) and C-terminal (green) domains are indicated. The cofactor PLP (yellow) is bound by the NSDAP-AT active site residues (pink) (PDB: 2O0R).

(underlined) (Fuchs et al., 2000).

al., 1998).

residues within the active site via a hydrogen bond network.
