Structure and Catalytic Mechanism of Mammalian Phospholipase

No mammahan PLD has been crystallized and its three-dimensional structure determined. However, the structure of a PLD from Streptomyces has been determined and that of the Nuc endonuclease from E. coli, which is a low molecular mass (16 kDa) member of the PLD superfamily [13, 14]. The crystal structure of another member of the PLD superfamily, tyrosyl-DNA phosphodiesterase, has 

The Streptomyces PLD consists of two highly interacting components of similar topology [13]. Overall, the structure is very similar to die Nuc dimer. Each component consists of a /1-sheet of 8-9 strands surrounded by nine a-helices. The catalytic center is also very similar to that of the Nuc dimer. Crystallization with a phosphate buffer indicates that the phosphate head group of the substrate lies in contact with His, Lys and Asn residues contributed from both HKD domains [13]. 

The tyrosyl-DNA phosphodiesterase is also composed of two similar domains related by a 2-fold axis of symmetry [15]. Conserved His, Lys and Asn residues are contributed by each domain to form the active site. Thus the structure and catalytic mechanism of this enzyme are very similar to those of other PLD superfamily members. A derived structure for the catalytic core of mammalian PLDl is shown 

The structure of fhe catalytic center of mammalian PLDs and also fhe catalytic mechanism are probably very similar to those reported above. In fact, there is evidence that fhe HDK motifs in conserved sequences II and IV must dimerize to form the catalytic center [17, 18] (Fig. 4.2). However, mutational studies indicate additional roles for conserved sequence III and also fhe extreme C-terminus [19, 20]. What roles these sequences play in catalysis are presently unknown. Interestingly, deletion of the PH and/or PX domains does not reduce fhe catalytic activity of mammalian PLDs [17, 21-23]. 

The cellular locations of the mammalian PLD isozymes remain controversial. This is because immunological detection of the endogenous isozymes in cells has largely been unsuccessful, and most studies have utilized overexpression of epitope- or GFP-tagged PLDs. Consequently, the possibility of mislocalization of the overexpressed enzymes is a key issue in these studies. Subcellular fractionation studies have indicated that both PLDl and PLD2 are membrane bound, except when the expression level is high [17, 24, 25]. Phosphorylation and palmitoylation of fhe enzymes can alter their membrane association [17, 24—26]. These post-translational modifications will be considered in fhe next section. 

---