Ping Wu

Identification and characterization of a new human type 9 cGMP-specific phosphodiesterase splice variant (PDE9A5): Differential tissue distribution and subcellular localization of PDE9A variants

Previously, four splice variants of human cGMP-specific phosphodiesterase (PDE) 9A (PDEs 9A1, 9A2, 9A3 and 9A4) have been identified. In this study, we have cloned a cDNA representing a new human PDE9A variant (PDE9A5). PDE9A5 encodes a protein of 492 amino acids, smaller than PDEs 9A1 and 9A2 but larger than PDEs 9A3 and 9A4. The exon structure of PDE9A5 is different from those of PDEs 9A1, 9A2, 9A3 and 9A4 in that, of the 20 exons of PDE9A gene, it lacks exons 2 and 5. PDE9A5 has been characterized in comparison with PDE9A1, the longest PDE9A variant. PDEs 9A5 and 9A1 have similar enzymatic properties. They both have a high affinity for cGMP with similar Km values (0.39 and 0.25 microM, respectively), although they have slightly different Vmax values (2.55 and 0.96 micromol/min/mg, respectively). They exhibit very similar divalent metal ion dependency and inhibitor sensitivity. Real-time quantitative PCR analysis shows that PDEs 9A5 and 9A1 exhibit differential tissue distribution. They are highly expressed in immune tissues (spleen, lymph node and thymus) and are more abundant in T cells than in B cells, neutrophils and monocytes. When transiently expressed in HEK293 cells, PDEs 9A5 and 9A1 proteins exhibit differential subcellular localization. PDE9A5 localizes exclusively in the cytoplasm, whereas PDE9A1 localizes in the nucleus only. The nuclear localization of PDE9A1 is dependent on a unique pat7 motif. By Western blot analysis, native PDE9A1 is detectable in the nucleus but not in the cytoplasm of T cells. Thus, to our knowledge, PDE9A1 is the only PDE isoform found to localize exclusively in the nucleus. We speculate that the physiological role of the PDE9A diversity may be imparting cGMP-metabolizing ability to specific cellular compartments in appropriate tissues.

Human phosphodiesterase 8A splice variants: cloning, gene organization, and tissue distribution.

We have cloned cDNAs representing five full-length human phosphodiesterase (PDE) 8A splice variants (PDE8As 1-5) from testis and T cells. PDE8A1 encodes a hydrophilic protein of 829 amino acids, containing an N-terminal REC domain, a PAS domain, and a C-terminal catalytic domain. PDE8A2 encodes a protein of 783 amino acids, identical to PDE8A1 but lacking the PAS domain. PDE8A3 encodes a shorter protein equivalent to the C-terminal 449 amino acids of PDE8A1, containing the catalytic but not the REC and PAS domains. PDE8A4 and PDE8A5, though different from each other at the nucleotide level, encode an identical protein equivalent to the C-terminal 582 amino acids of PDE8A1, including half of the PAS domain. The PDE8A gene is revealed to contain 23 exons, and its exon-intron boundaries have been defined. In addition, we have mapped a common transcription initiation site, and further determined the upstream 5-flanking sequence of 1740 bp containing the putative promoter. Compared to PDE8A1, PDE8As 2-5 appear to be expressed in much lower abundance. Among various tissues and organs, PDE8A1 and PDE8A2 are expressed at various levels.

Induction of lysosomal and plasma membrane-bound sialidases in human T-cells via T-cell receptor

Among the three isoenzymes of neuraminidase (Neu) or sialidase, Neu-1 has been suggested to be induced by cell activation and to be involved in IL (interleukin)-4 biosynthesis in murine T-cells. In the present study, we found that antigen-induced airway eosinophilia, a typical response dependent on Th2 (T-helper cell type 2) cytokines, as well as mRNA expression of Th2 cytokines, including IL-4, are suppressed in Neu-1-deficient mice, thereby demonstrating the in vivo role of murine Neu-1 in regulation of Th2 cytokines. To elucidate the roles of various sialidases in human T-cell activation, we investigated their tissue distribution, gene induction and function. Neu-1 is the predominant isoenzyme at the mRNA level in most tissues and cells in both mice and humans, including T-cells. T-cells also have significant levels of Neu-3 mRNAs, albeit much lower than those of Neu-1, whereas the levels of Neu-2 mRNAs are minimal. In human T-cells, both Neu-1 and Neu-3 mRNAs are significantly induced by T-cell-receptor stimulation, as is sialidase activity against 4-methylumbelliferyl- N -acetylneuramic acid (a substrate for both Neu-1 and Neu-3) and the ganglioside G(D1a) [NeuAcalpha2-3Galbeta1-3GalNAcbeta1-4(NeuAcalpha2-3)Galbeta1-4Glcbeta1-cer] (a substrate for Neu-3, but not for Neu-1). The expression of the two sialidase genes may be under differential regulation. Western blot analysis and enzymic comparison with recombinant sialidases have revealed that Neu-3 is induced as a major isoform in activated cells. The induction of Neu-1 and Neu-3 in T-cells is unique. In human monocytes and neutrophils stimulated with various agents, the only observation of sialidase induction has been by IL-1 in neutrophils. Functionally, a major difference has been observed in Jurkat T-cell lines over-expressing Neu-1- and Neu-3. Upon T-cell receptor stimulation, IL-2, interferon-gamma, IL-4 and IL-13 are induced in the Neu-1 line, whereas in the Neu-3 line the same cytokines are induced, with the exception of IL-4. Taken together, these results suggest an important immunoregulatory role for both Neu-1 and Neu-3 in humans.

Characterization of human, dog and rabbit corpus cavernosum type 5 phosphodiesterases

Human, dog and rabbit corpus cavernosum type 5 phosphodiesterases (PDE5) were isolated and their characteristics were compared. The three enzymes showed Km values of 0.8, 2.1 and 2.3 uM, respectively. They exhibited similar pH-dependence with optimal pH being 7.5. They required Mg++ for activity and the activity was suppressed by high concentrations of Zn++ (0.1-1 mM). Sildenafil potently inhibited the three enzymes with IC50 values of 3.6, 1.7 and 3.0 nM, respectively. Dipyridamole and IBMX (3-isobutyl-1-methylxanthine) each also inhibited the three enzymes with similar, albeit lower, potencies (IC50 about 1.1 and 5.7 uM, respectively). However, zaprinast exhibited a significantly higher potency against the rabbit enzyme (IC50 53 nM) than against the human and dog PDE5s (IC5s 332 and 217 nM, respectively). Thus, the corpus cavernosum PDE5s are very similar among the various species with the only significant difference being their sensitivity to zaprinast. Human platelet PDE5 was also characterized by comparison with the corpus cavernosum enzyme. The platelet enzyme exhibited a Km, pH-, Mg++- and Zn++-dependence, and sensitivity to sildenafil and zaprinast very similar to those of the corpus cavernosum PDE5. However, compared with corpus cavernosum PDE5, the platelet enzyme exhibited higher sensitivity to dipyridamole and IBMX (IC50 0.46 and 1.8 uM, respectively). This study shows that despite similar kinetics and enzymatic properties, corpus cavernosum PDE5s from different species, and corpus cavernosum and platelet PDE5s, can have differential sensitivity to pharmacological inhibitors.

Design and synthesis of xanthine analogues as potent and selective PDE5 inhibitors.

We have discovered potent and selective xanthine PDE5 inhibitors. Compound 25 (PDE5 IC(50)=0.6 nM, PDE6/PDE5=101) demonstrated similar functional efficacy and PK profile to Sildenafil (PDE5 IC(50)=3.5 nM, PDE6/PDE5=7).