Phosphodiesterase Inhibitors: Factors That Influence Potency, Selectivity, and Action - Phosphodiesterases as Drug Targets - page 57

2.1 Comparison of Native and Recombinant Enzymes
A recurrent finding when investigating the properties of various PDEs is that the
kinetic behavior of a recombinant, ectopically expressed enzyme does not entirely
overlap with that of its endogenous native counterpart (Bolger et al. 1993 ; Livi et al.
1990 ; McPhee et al. 1999 ; Obernolte et al. 1993 ; Salanova et al. 1998 ; Saldou et al.
1998 ; Sullivan et al. 1994 ; Wang et al. 1997 ). The differences include affinity for
substrate, catalytic efficiency, and inhibitor potencies (Obernolte et al. 1993 ;
Salanova et al. 1998 ; Wang et al. 1997 ). Certainly, some differences are due to
the fact that native PDE preparations are often heterogeneous in terms of enzyme
composition (Obernolte et al. 1993 ). For many PDEs, the physicochemical proper-
ties of the various isoforms are not sufficiently different to allow for complete
separation with even the most current chromatographic methods. More frequently,
differences are associated with the heterologous expression system used (i.e.,
bacteria, yeast, insect cells, or mammalian cell lines) and the posttranslational
modification of the recombinant proteins (Hoffmann et al. 1998 ; Saldou et al.
1998 ). Aberrant phosphorylations, different states of phosphorylation, or other
modifications such as ubiquitination or proteolytic cleavage of overexpressed
proteins have all been reported for heterologously expressed PDEs. Heterogeneity
in the conformation of some recombinant proteins is an additional major variable.
Overexpression of recombinant PDEs often produces proteins that are improperly
folded, aggregated, or catalytically inactive (Rocque et al. 1997 ; Scapin et al. 2004 ;
Wang et al. 1997 ). The presence of catalytic domains in different conformations
due to posttranslational modification and interaction with partner proteins can
confound the kinetics of inhibition of enzyme preparations with potencies reported
for rolipram inhibition of PDE4 isoforms ranging from 0.01 to 10 m M (Hoffmann
et al. 1998 ; Houslay and Adams 2003 ; Huston et al. 1996 ; Laliberte et al. 2000 ;
McPhee et al. 1999 ; Rocque et al. 1997 ; Salanova et al. 1998 ; Tian et al. 1998 ).
Proteolysis of overexpressed enzymes may also alter interaction with inhibitors
without affecting catalytic activity (see below).
2.2 Truncated Versus Full-Length PDEs
An additional property to consider is whether truncated forms of PDEs recapitulate
the properties of the full-length enzymes. Two examples involving PDEs 4 and 5
are provided to expand upon this concept. Rolipram, as well as some other
compounds, inhibits isolated catalytic domains of PDE4s with potencies that are
often 10- to 100-fold lower than those obtained with full-length enzyme (Jacobitz
et al. 1996 ; Richter and Conti 2004 ; Tian et al. 1998 ). Progressive truncation at the
amino terminus of PDE4B and PDE4A causes a parallel decrease in the affinity for
rolipram (Jacobitz et al. 1996 ; Richter and Conti 2004 ; Rocque et al. 1997 ; Saldou
et al. 1998 ). The potencies of three inhibitors (RS-25344, rolipram, and TVX 2706)
for inhibition of the isolated catalytic domain of PDE4D are much weaker than for