QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.045526v1 74/1/92    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Birkus, G. Articles by Cihlar, T. Search for Related Content PubMed PubMed Citation Articles by Birkus, G. Articles by Cihlar, T.

Activation of 9-[(R)-2-[[(S)-[[(S)-1-(Isopropoxycarbonyl)ethyl]amino] phenoxyphosphinyl]-methoxy]propyl]adenine (GS-7340) and Other Tenofovir Phosphonoamidate Prodrugs by Human Proteases

Gilead Science, Foster City, California

Received January 24, 2008; accepted April 22, 2008

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
9-[(R)-2-[[(S)-[[(S)-1-(Isopropoxycarbonyl)ethyl]amino] phenoxyphosphinyl]-methoxy]propyl]adenine (GS-7340) is an isopropylalaninyl phenyl ester prodrug of the nucleotide HIV reverse transcriptase inhibitor tenofovir (TFV; 9-[(2-phosphonomethoxy)propyl]adenine) exhibiting potent anti-HIV activity and enhanced ability to deliver parent TFV into peripheral blood mononuclear cells (PBMCs) and other lymphatic tissues in vivo. The present study focuses on the intracellular metabolism of GS-7340 and its activation by a variety of cellular hydrolytic enzymes. Incubation of human PBMCs in the presence of GS-7340 indicates that the prodrug is hydrolyzed slightly faster to an intermediate TFV-alanine conjugate (TFV-Ala) in quiescent PBMCs compared with activated cells (0.21 versus 0.16 pmol/min/10⁶ cells). In contrast, the conversion of TFV-Ala to TFV and subsequent phosphorylation to TFV-diphosphate occur more rapidly in activated PBMCs. The activity of GS-7340 hydrolase producing TFV-Ala in PBMCs is primarily localized in lysosomes and is sensitive to inhibitors of serine hydrolases. Cathepsin A, a lysosomal serine protease has recently been identified as the primary enzyme activating GS-7340 in human PBMCs. Results from the present study indicate that in addition to cathepsin A, a variety of serine and cysteine proteases cleave GS-7340 and other phosphonoamidate prodrugs of TFV. The substrate preferences displayed by these enzymes toward TFV amidate prodrugs are nearly identical to their preferences displayed against oligopeptide substrates, indicating that GS-7340 and other phosphonoamidate derivatives of TFV should be considered peptidomimetic prodrugs of TFV.

GS-7340 (Fig. 1) is a prototype molecule representing a novel class of TFV mono-phosphonoamidate prodrugs. Unlike TDF, GS-7340 contains phenol and alanine isopropyl ester as the phosphonate masking groups (Fig. 1). Relative to parent TFV, GS-7340 exhibits 500- to 1000-fold enhanced activity against HIV-1 in T-cells, activated peripheral blood mononuclear lymphocytes (PBMCs), and macrophages (Lee et al., 2005). It is also a potent inhibitor of hepatitis B virus (Lee et al., 2005) and exhibits markedly enhanced stability in plasma compared with TDF (Eisenberg et al., 2001). As a consequence, in vivo oral administration of GS-7340 results in an increased accumulation of parent TFV and its active diphosphate metabolite in PBMCs relative to TDF (Lee et al., 2005). GS-7340 also increases the accumulation of TFV in lymphatic tissues, which is likely to favorably affect the therapeutic efficacy of this novel prodrug (Lee et al., 2005).

View larger version (13K): [in this window] [in a new window]   Fig. 1. Activation pathway of GS-7340.

Phosphonoamidate prodrugs of TFV share structural similarity with previously explored aryl phosphoramidate derivatives of antiviral nucleosides (Balzarini et al., 1996; Valette et al., 1996). After penetration into cells, the activation of phosphoramidate prodrugs is initiated by hydrolytic enzymes that cleave the carboxyester bond in the prodrug moiety. This hydrolysis releases a metastable metabolite, from which the phenol group is spontaneously eliminated via intramolecular cyclization and hydrolysis (Balzarini et al., 1996; Valette et al., 1996). In the case of GS-7340, the resulting metabolite is a conjugate of TFV and alanine (TFV-Ala) (Fig. 1). Because GS-7340 is a lipophilic cell-permeant compound, the formation of negatively charged TFV-Ala has important pharmacological implications. As TFV-Ala undergoes further conversion to parent TFV, its efficient formation is crucial for the intracellular accumulation and retention of TFV metabolites delivered via GS-7340. In vitro experiments have demonstrated that the antiviral activity of different TFV phosphonoamidates is greatly affected both by the ester and amino acid moiety present in the prodrug (Lee et al., 2005), which is probably a consequence of differences in the efficiency of hydrolysis of the prodrug carboxy ester bond.

To understand the activation of GS-7340 and other phosphonoamidates of TFV, including the mechanism of prodrug hydrolysis, the metabolism of GS-7340 in quiescent and activated human PBMCs was investigated. In addition, subcellular fractionation was used to demonstrate that GS-7340 undergoes its enzymatic hydrolysis in lysosomes. Finally, the present study shows that in addition to cathepsin A, which was identified as a major hydrolase responsible for the activation of GS-7340 in PBMCs (Birkus et al., 2007) multiple specific serine proteases and hydrolases cleave GS-7340 and other phosphonoamidate prodrugs of TFV. The characterized proteases display substrate preference toward TFV amidate prodrugs that closely resembles their substrate specificity for oligopeptide substrates, indicating that the amidates can be considered peptidomimetic prodrugs of TFV.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Reagents. Human cathepsin A was isolated from PBMCs as described previously (Birkus et al., 2007), human leukocyte elastase, human cathepsins B, D, G, and H, bovine cathepsin C, mouse granzymes A and B, porcine liver carboxylesterase, Percoll, 4-methylum-belliferyl-2-actamide-2-deoxy-β-O-glucopyranoside, iodonitrotetrazolium chloride, and all common chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Human chymotrypsin and trypsin, human mast cell chymase, and porcine pancreatic elastase were supplied by Calbiochem (San Diego, CA). Human proteinase 3 was from EPC (Owensville, MO). The custom synthesis of [¹⁴C]GS-7340 was performed by Moravek Biochemicals (Brea, CA). Amplify Fluorographic Reagent, ECL Western blotting detection kit, and Hyperfilm MP were purchased from GE Healthcare (Chalfont St. Giles, Buckinghamshire, UK). EZ-Link Sulfo-NHS-Biotinylation Kit was from Pierce Biotechnology (Rockford, IL). [³H]Chloroquine was purchased from American Radiolabeled Chemicals (St. Louis, MO). The monoamidate prodrugs of tenofovir were prepared using a modified procedure taken from literature (Lee et al., 2005).

Metabolism of GS-7340 in Human PBMCs. PBMCs were prepared from human buffy coats obtained from Stanford Medical School Blood Center (Stanford, CA). One human buffy coat (30 ml) was diluted with 10 ml of PBS, and 13 ml of diluted buffy coat was layered over 13 ml of Ficoll Plus (GE Healthcare). Tubes were centrifuged at 500g for 15 min at room temperature. The leukocyte layer was removed and mixed with hypotonic red cell lysis buffer (155 mM NH₄Cl, 0.1 mM EDTA, and 10 mM KHCO₃). Cells were centrifuged at 500g for 5 min at room temperature and washed with red cell lysis buffer. Finally, cells were washed in 50 ml of PBS and pelleted. Quiescent PBMCs were resuspended in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum to a final concentration of 2 x 10⁶ cells per ml. Eight milliliters of cell culture was kept in 15 ml of conical tubes for 1 h at 37°C in a 5% CO₂ air atmosphere before 10 µM [¹⁴C] GS-7340 (0.53 µCi/ml) was added. A portion of purified PBMCs were activated for 2 days in RPMI 1640 medium containing 10 units/ml Interleukin-2, 50 ng/ml phorbol 12-myristate 13-acetate, and 1 µg/ml phytohemagglutinin. After the activation, cells were spun at 500g and resuspended in fresh RPMI 1640 medium to a final concentration of 2 x 10⁶ cells/ml. One hour later, cells were exposed to 10 µM [¹⁴C]GS-7340 for 15 to 180 min. Eight milliliters of the culture were spun (500g, 5 min, 4°C) at each time point, and the cell pellets were washed twice with 12 ml of ice-cold PBS. Finally, cells were lysed with 700 µl of ice-cold 70% methanol in 10 mM TrisCl, pH 7.6. The lysates were incubated for 30 min at -20°C and then spun at 13,000g for 30 min at 4°C. Supernatants were collected, evaporated, and reconstituted in 200 µl of water. The total amount of all GS-7340 metabolites per 10⁶ cells was determined by scintillation counting of an aliquot of the reconstituted cell lysate. A calibration curve of cpm versus pmol of [¹⁴C]GS-7340 was constructed to calculate the total intracellular concentration of metabolites.

HPLC Analysis of GS-7340 Metabolites in PBMCs. The analysis was performed using a modified HPLC protocol developed previously (Eisenberg et al., 2001). The metabolites were separated on a Prodigy ODS-3 column (5 mm, 150 x 4.6 mm; Phenomenex, Torrance, CA) using the HP 1090 (series II) HPLC system connected to a Radiomatic Flo-One Beta liquid scintillation detector (Packard Series A-500). A gradient elution from buffer A (5% AcCN in 25 mM phosphate buffer, pH 6, and 5 mM tetrabutylammonium bromide) to 25% buffer A/75% buffer B (60% AcCN in 25 mM phosphate buffer, pH 6, and 5 mM tetrabutylammonium bromide) for 15 min and a flow rate of 1.2 ml/min and 40°C was used to separate the metabolites. The metabolites were identified by the comparison of their retention time with that of standards for TFV (t_R = 3.5 min), TFV-Ala (t_R = 6.3 min), TFVp (t_R = 8.8 min), TFVpp (t_R = 11.4 min) and GS-7340 (t_R = 13.6 min).

Preparation of PBMC Extract. Freshly isolated human PBMCs (4 x 10⁸ cells) were washed twice with 15 ml of PBS and centrifuged (500g, 5 min). The pellet was extracted with 4 ml of 10 mM Tris-HCl buffer, pH 7.3, 150 mM NaCl, 2 mM CaCl₂, 1 mM DTT, and 1% Nonidet P-40 for 20 min on ice. The extract was cleared by centrifugation at 13,000g for 30 min and the supernatant was frozen at -70°C.

Subcellular Fractionation. MT-2 T cells and PBMCs were grown in RPMI 1640 medium supplemented with 100 mM HEPES and 10% fetal bovine serum at 37°C. To perform subcellular fractionation, cells were disrupted and fractionated using a modified version of a method described previously (Wex et al., 2001). In brief, 5 x 10⁷ cells were washed twice with ice-cold PBS and once with buffer A (0.25 mM sucrose, 3 mM imidazole, pH 7.4, and 1 mM EDTA). The pellet was resuspended in 500 µl of buffer A and passed 10 times (MT-2 cells) or 35 times (PBMC cells) through a 26 gauge needle. This resulted in a disruption of >90% cells as monitored by microscopic examination. Disrupted cells were brought to a volume of 4 ml and spun at 1000g for 10 min at 4°C. The pellet containing nuclei and unbroken cells was discarded. The supernatant was removed, diluted with buffer A to a volume of 4.33 ml, and combined with 1.67 ml of 90% Percoll. The sample was centrifuged at 75,000g for 30 min at 4°C using a 50TI rotor on an ultracentrifuge (L7; Beckman Coulter, Fullerton, CA). A gradient of Percoll with a density of 1.040 to 1.15 g/ml was formed during the centrifugation. Fractions (0.25 ml) were collected from the top of the tube. To remove Percoll, each fraction was spun in a TLA100 rotor on a TL-100 ultracentrifuge at 100,000g for 1 h at 4°C. This step was necessary to prevent the interference of Percoll with enzyme marker assays.

Organelle Marker Assays. Lysosomal enzyme marker β-hexosaminidase was assayed using 5 µM fluorescent substrate 4-methylumbelliferyl-2-acetamide-2-deoxy-β-O-glucopyranoside in 100 mM sodium acetate and 0.1% TX-100, pH 4.5 (Merion and Sly, 1983). In addition, chloroquine was used as a small molecule marker for lysosomes (Matsuzawa and Hostetler, 1980). Cells were incubated with 20 nM [³H]chloroquine for 1 h at 37°C followed by disintegration and gradient fractionation. The amount of chloroquine in each fraction was determined by scintillation counting. Mitochondrial succinate dehydrogenase was assayed using iodonitrotetrazolium chloride according to a previously published protocol (Pennington, 1961). NADPH cytochrome-c reductase was used as a marker for endoplasmic reticulum according to previously published methods (Ouar et al., 2003). The activity of GS-7340 hydrolase was evaluated using the same reaction condition as described for the hydrolase from PBMC extracts (see below).

In the experiment evaluating TFV metabolites distribution among the organelles, the cells were incubated with 10 µM[¹⁴C]GS-7340 for 4 h at 37°C followed by disintegration and gradient fractionation. The total amount of TFV metabolites in each fraction was determined by scintillation counting.

Enzyme Assays. TFV prodrugs (100 µM) were incubated with each enzyme for 10, 30, and 120 min at 37°C. The reactions with leukocyte elastase (4 µg/ml), proteinase 3 (4.4 µg/ml), porcine liver carboxyl esterase (40.5 µg/ml), granzyme A (13.88 µg/ml), and granzyme B (13.88 µg/ml) was performed in a reaction buffer containing 25 mM Mes-Na⁺, pH 6.5, 100 mM NaCl, and 1 mM DTT. The extract from PBMCs (134 µg/ml) was assayed under the same conditions except that Nonidet P-40 was present at a final concentration of 0.1%. To perform the inhibition experiments, PBMC extract was preincubated with 0, 1, 10, and 100 µM tested inhibitors for 10 min. GS-7340 was added to a final concentration of 100 µM and incubated with the treated extract for 5, 10, and 20 min. After sample analysis, the concentration of inhibitor reducing the hydrolysis of GS-7340 by 50% (IC₅₀) was determined. Mast cell chymase (1 µg/ml), cathepsin G (0.333 µg/ml), chymotrypsin (1.47 µg/ml), trypsin (16 µg/ml), and pancreatic elastase (7.8 µg/ml) were incubated with TFV prodrugs in a reaction buffer containing 25 mM Tris-HCl, pH 8.0, 100 mM NaCl, and 1 mM DTT. The conversion of TFV prodrugs by cysteine proteases cathepsin H (7.5 µg/ml), cathepsin L (7.35 µg/ml), cathepsin B (8.33 µg/ml), and cathepsin C (40 µg/ml) was performed in 25 mM Mes-Na⁺, pH 6.5, 100 mM NaCl, and 5 mM DTT. Aspartic protease cathepsin D (5.2 µg/ml) was incubated with prodrugs in 25 mM acetate-Na⁺ buffer, pH 4.0, containing 100 mM NaCl and 1 mM DTT.

After the incubation, reactions with all enzymes were quenched by adding ice-cold methanol (final concentration, 70%) and samples were processed as described above. Reaction products were quantified by HPLC analysis on C18 column (Vydac, Hesperia, CA) using a gradient of acetonitrile (5%-45%, 6 min; flow rate, 0.25 ml/min; 260 nm) in a buffer containing 25 mM KPO₄, pH 6.0, and 5 mM tetrabutylammonium bromide. The identity of corresponding TFV-amino acid conjugates was confirmed by LC-mass spectrometry analysis. Hydrolase specific activity was expressed as picomoles of metabolites produced per minute per microgram of total protein. For all tested enzymes, the prodrug hydrolysis rates were determined under the steady-state reaction conditions.

Covalent Labeling of Serine Hydrolases with [¹⁴C]GS-7340. Bovine serum albumin (BSA), proteinase 3 (Pr3), leukocyte elastase (LE), pancreatic elastase (PE), and porcine liver carboxylesterase (PLCE) were diluted in 200 µl of 25 mM Mes-Na⁺ buffer, pH 6.5, containing 100 mM NaCl and 1 mM DTT to a final concentration of 10 µg/ml and kept on ice. Half of each enzyme solution was denatured for 10 min at 80°C before the labeling was performed. Native and denatured PLCE was also preincubated with nonradioactive diisopropyl fluorophosphate (DFP) (final concentration, 100 µM) before labeling with GS-7340. The denatured and native samples were simultaneously preincubated for 2 min at 37°C and [¹⁴C]GS-7340 (final concentration, 1.5 mM; 0.08 mCi/ml) was added. After an additional 2-min incubation at 37°C, the reaction was quenched with 0.3% SDS solution and proteins were precipitated with ice-cold trichloroacetic acid (10% final concentration). The samples were kept on ice for 30 min, spun at 13,000g for 30 min at 4°C, and protein pellets were washed twice with 700 µl of ice-cold acetone. The protein pellets were air-dried, dissolved in Nu-PAGE loading buffer (Invitrogen, Carlsbad, CA), and denatured for 10 min at 70°C. After electrophoresis in 4-12% Nu-PAGE BisTris gel with Mes buffer (Invitrogen), the gels were fixed in isopropanol/water/acetic acid (25:65:10) solution for 30 min and then soaked in Amplify Fluorographic Reagent (GE Healthcare) for 30 min. The gels were dried under vacuum at 80°C and exposed for 2 weeks to preflashed X-ray films. Native and denatured PLCE was also labeled with 50 µM [³H]DFP (0.1 mCi/ml) for 30 min at 37°C. The labeled protein was then processed as described above.

View larger version (15K): [in this window] [in a new window]   Fig. 2. Metabolism of GS-7340 in quiescent and activated PBMCs. 2 x 106 cells per ml were incubated with 10 µM [3H]GS-7340 (0.52 µCi/ml) for 15, 30, 60, 120, and 180 min. Cells were extracted with 70% methanol and GS-7340 (), TFV-Ala (), TFV (), and TFVpp () were analyzed using high-performance liquid chromatography-liquid scintillation detection as described under Materials and Methods.

	Results

Top Abstract Materials and Methods Results Discussion References

The initial rate of GS-7340 hydrolysis to TFV-Ala observed in quiescent and activated PBMCs is approximately 0.21 and 0.16 pmol/10⁶ cells/min, respectively (Table 1). The slightly higher rate of TFV-Ala production observed in quiescent cells results in approximately 1.3-fold higher total level of TFV metabolites compared with activated cells (Fig. 2). Although the initial rate of TFV formation (Table 1) is similar in both quiescent and activated PBMCs (0.11-0.13 pmols/10⁶ cells/min), the equilibrium level of TFV-Ala in quiescent cells (55µM) was 2.5-fold higher than the equilibrium concentration of TFV-Ala reached in activated cells (22µM). This observation suggests that the conversion of TFV-Ala to TFV is more efficient in activated PBMCs, possibly because of a higher expression of the putative phosphoamidase in these cells. Despite the presence of higher concentrations of TFV in quiescent cells, the rate of intracellular accumulation of TFVpp is independent of the cell activation state (0.05 pmols/10⁶ cells/min), suggesting that TFV is slightly more efficiently phosphorylated in activated cells compared with quiescent cells.

Characterization of GS-7340 Hydrolase Activity in PBMC Extract. The rapid hydrolysis of GS-7340 observed in PBMCs and cellular extracts (Table 1 and 2) indicates the presence of intracellular hydrolytic enzyme(s) capable of catalyzing the hydrolysis of the carboxyl ester bond of GS-7340. The structural similarity between the phosphonophenyl ester isopropyl alanyl amidate and the tripeptide Phe-Ala-Ala (Fig. 3) suggests that the phenol ring, alanine moiety, and the isopropyl ester of GS-7340 may mimic P2, P1, and P1' residues of a peptidic substrate. [Nomenclature for the substrate amino acid preference is P4, P3, P2, P1, P1', P2', P3', and P4'. Amide bond hydrolysis occurs between P1 and P1'. S4, S3, S2, S1, S1', S2', S3', and S4' denote the corresponding enzyme binding sites (Schechter and Berger, 1967).] Together with the known fact that proteases and peptidases can efficiently hydrolyze ester in addition to amide bonds, these structural characteristics of GS-7340 suggest the potential involvement of specific cellular proteases and/or peptidases in the activation of GS-7340. To further characterize this hydrolytic activity, GS-7340 was incubated with PBMC extract in the presence and absence of various protease inhibitors.

View larger version (16K): [in this window] [in a new window]   Fig. 3. Alignment of GS-7340 (in blue) and Phe-Ala-Ala tripeptide (green).

Various inhibitors of cysteine proteases [E64; L-trans-epoxysuccinyl-leucylamido(4-guanidino)butane] (Barrett et al., 1982), aspartic proteases (pepstatin A) (Rich et al., 1982), and metalloproteases (EDTA, bestatin) (Powers and Harper, 1986) at concentrations up to 100 µM failed to inhibit the hydrolysis of GS-7340 in PBMC extract. In contrast, GS-7340 hydrolysis was inhibited by both DFP and 3,4-dichloroisocoumarin, two potent inhibitors of serine proteases (Powers et al., 2002), with IC₅₀ values of 10.5 and 6.5 µM, respectively.

In addition, the hydrolytic activity present in PBMC extracts exhibited a substrate preference for TFV prodrugs containing linear aliphatic or aromatic residues (Ala, Gly, -ABA, or Phe) compared with those containing aliphatic branched amino acid moieties (Leu, Ile, or Val) (Table 2, Fig. 4). Together, these data suggest that the hydrolysis of the alanine isopropyl ester in GS-7340 as well as ester bonds in other amidate prodrugs of TFV may be mediated by specific proteases present in PBMC extract.

View larger version (12K): [in this window] [in a new window]   Fig. 4. Structures of tenofovir phoshonoamidate prodrugs.

View larger version (23K): [in this window] [in a new window]   Fig. 5. Fractionation of MT-2 cells (A and B) and PBMCs (C and D) using Percoll gradient. Lysosomal markers hexosaminidase () and chloroquine (), mitochondrial marker succinate dehydrogenase (), endoplasmic reticulum marker NADPH cytochrome c reductase (), GS-7340 hydrolase activity (), and TFV metabolites ()

TFV prodrug containing Phe in the amidate moiety (i.e., in the putative P1 position) is cleaved most efficiently by chymotrypsin-like serine proteases (chymotrypsin, cathepsin G, and mast cell chymase) (Table 2). In contrast, a prodrug containing Leu was cleaved 7- to 13-fold less efficiently by these proteases. Chymotrypsin and mast cell chymase exhibited low activity against a prodrug with Gly and -aminobutyric acid (-ABA) in the P1 position. It is noteworthy that chymotrypsin-like proteases failed to hydrolyze GS-7340 with Ala as the P1-like residue. In contrast, elastase-like serine proteases (LE, Pr3, and PE I) exhibited a preference for TFV prodrugs containing Ala, Ile, Val, and -ABA moieties at this position (Table 2). PE displayed the broadest substrate specificity toward TFV phosphonoamidate prodrugs, the -ABA-containing prodrug being cleaved most efficiently. In contrast, Ala-containing GS-7340 was the best substrate for leukocyte elastase among the prodrugs tested.

It is noteworthy that none of the tested prodrugs were hydrolyzed by trypsin-like protease (trypsin, granzyme A) or aspase-like protease (granzyme B). Among cysteine proteases, cathepsins H and L were the only enzymes capable of hydrolyzing TFV prodrugs, albeit with a very low efficiency (Table 2). Cathepsin D, the only aspartic protease tested, hydrolyzed no TFV phosphonoamidate prodrugs. As expected, the nonselective liver carboxyl esterase exhibited a broad substrate specificity toward a wide range of tested TFV prodrugs (Table 2). However, the efficiency of prodrug hydrolysis was substantially lower compared with more specific proteases.

Cathepsin A exhibited the highest specific activity for the GS-7340 hydrolysis and was also able to effectively cleave TFV prodrugs containing Phe, -ABA, or Gly. It is noteworthy that chymotrypsin, cathepsin G, and chymase hydrolyzed the Phe-containing prodrug with efficiency comparable with that of cathepsin A. On the other hand, TFV prodrugs with Val, Leu, and Ile were not substrates for cathepsin A, but they were hydrolyzed by LE, Pr3, and PE I.

Covalent Labeling of Hydrolases with GS-7340 Substrate. The reaction mechanism of serine hydrolases involves the formation of a covalent bond between serine in the active site and the substrate acyl group (Satoh and Hosokawa, 1998). Denaturing the enzyme during the course of reaction should allow for capturing this intermediate. Consistent with this assumption, we have recently demonstrated that cathepsin A can be covalently labeled with [¹⁴C]GS-7340 (Birkus et al., 2007). Likewise, the other enzymes tested in this study and found capable of GS-7340 hydrolysis (PLCE, LE, Pr3, and PE) were successfully labeled with GS-7340 (Fig. 6). The specificity of this method was confirmed by the lack of labeling of heat-denatured enzymes or inert bovine serum albumin. In addition, the preincubation of enzymes with the inhibitor DFP prevents their labeling by [¹⁴C]GS-7340. These observations further confirm the interactions of GS-7340 with multiple proteases observed in enzyme kinetic experiments.

View larger version (36K): [in this window] [in a new window]   Fig. 6. Labeling of hydrolases with [14C]GS-7340. Labeling of PLCE, BSA, Pr3, LE, and PE with [14C]GS-7340 was performed as described under Materials and Methods. A, native PLCE labeled with: [14C]GS-7340 (lane 1), [3H]DFP (lane 2), preincubated with unlabeled DFP and labeled with [14C]GS-7340 (lane 3); denatured PLCE labeled with: [14C]GS-7340 (lane 4); [3H]DFP (lane 5); preincubated with unlabeled DFP and labeled with [14C]GS-7340 (lane 6); native (lane 7) and denatured (lane 8) BSA labeled with [14C]GS-7340. B, native Pr3 (lane 1), PE (lane 2), and LE (lane 3) labeled with [14C]GS-7340; denatured Pr3 (lane 4), PE (lane 5), and LE (lane 6) labeled with [14C]GS-7340

	Discussion

Top Abstract Materials and Methods Results Discussion References

The initial rate of GS-7340 hydrolysis was marginally faster in quiescent cells than in activated PBMCs; however, the rate of TFV formation in cells was independent of their activation status. TFV-Ala reached a steady-state concentration approximately 2.5-fold higher in quiescent cells than in activated cells, suggesting higher activity of a phosphoamidase in activated cells (Chou et al., 2007). Alternatively, enhanced acidification of activated cells may cause more efficient conversion of TFV-Ala to TFV as a result of a limited stability of TFV-Ala under acidic conditions present in lysosomes. TFV-Ala exhibits t of 72 and 28 min at pH 5.4 and 4.9, respectively (data not shown). Despite higher levels of TFV-Ala and TFV in quiescent cells, the overall rate of the intracellular accumulation of TFVpp was similar regardless of the PBMC activation state, suggesting more efficient phosphorylation of TFV in activated cells.

Hydrolysis of GS-7340 in PBMC extract was sensitive to serine protease inhibitors but not to inhibitors of cysteine, aspartic, or metalloproteases. In addition, subcellular fractionation of the MT-2 T-cells and PBMCs conclusively demonstrated that GS-7340 hydrolase is localized in the high-density lysosomal fraction (Fig. 4). Lysosomes and lysosome-related organelles are known to contain a wide variety of hydrolases, including serine proteases (Borregaard and Cowland, 1997; Blott and Griffiths, 2002). Because the prodrug moiety of GS-7340 can be viewed as a structural analog of Phe-Ala-Ala tripeptide (Fig. 3), these results suggested that lysosomal serine protease(s) are probably responsible for the hydrolysis of GS-7340. Consistent with this hypothesis, we have recently identified cathepsin A as an enzyme capable of effectively activating GS-7340 (Birkus et al., 2007). In this follow-up study, we identified multiple other lysosomal proteases that are able to cleave a set of structurally diverse phosphonoamidate prodrugs of TFV, including GS-7340.

Leukocyte elastase and proteinase 3 are elastase-like serine proteases and prefer peptide substrates containing small hydrophobic amino acids in P1 position (Blow, 1977; Harper et al., 1984; Kam et al., 1992). These proteases hydrolyzed TFV prodrugs containing Val, Ala, or -ABA at P1-like position but not those with bulkier hydrophobic moieties such as Phe or Leu. Likewise, the chymotrypsin-related serine proteases cathepsin G and chymase, which prefer large hydrophobic residues in P1 position (Powers et al., 1985; Polanowska et al., 1998), efficiently hydrolyzed TFV prodrugs with Phe or Leu but not those with smaller amino acids, including GS-7340. In contrast, none of the tested TFV prodrugs were hydrolyzed by trypsin or granzymes A/B (Kam et al., 2000), which are known to display a strong preference for substrates with charged residues Lys/Arg or Asp at P1, respectively. Likewise, the relative specific activity of digestive serine proteases chymotrypsin and pancreatic elastase toward various TFV amidate prodrugs matched their preference for their peptide substrates (Bergman and Fruton, 1943). The susceptibility of TFV prodrugs to cleavage by digestive serine proteases may affect their stability in the gastrointestinal tract and consequently influence their oral bioavailability. In contrast to serine proteases, the primary structural determinant for the substrate specificity of papain-like cysteine endopeptidases is the P2 amino acid residue. Cathepsins B, H, and L all prefer hydrophobic amino acids, including Phe in the P2 position (Kärgel et al., 1980; Koga et al., 1992; Rothe and Dodt, 1992). Although all tested prodrugs contain phenol in the P2-like position that may mimic Phe residue, only cathepsins H and L were able to hydrolyze TFV amidates. This suggests that the phenol residue, which is shorter than natural Phe by one atom, may shift the scissile bond further from the enzyme catalytic dyad, thereby reducing the catalytic efficiency of substrate hydrolysis.

Cysteine protease cathepsin C and aspartyl protease cathepsin D both exhibit unique substrate specificity. The former cleaves two residues from the N terminus of polypeptide substrates (Turk et al., 1997). Cathepsin D has an elongated binding site cleft requiring interaction with up to seven amino acid residues in a polypeptide chain; therefore, it is inactive against shorter synthetic peptides (Press et al., 1960; Conner, 2004). Consistent with their substrate specificity, neither of these two enzymes hydrolyzed any of the tested TFV phosphonoamidate prodrugs. Taken together, these data demonstrate that the specificity of serine proteases toward TFV phosphonoamidate prodrugs closely matches their natural substrate specificity, indicating that nucleotide amidate prodrugs should be considered mimics of oligopeptide substrates.

PLCE, a close homolog of human liver carboxylesterase 1 (Lange et al., 2001), was shown to hydrolyze carboxyl ester bonds in a wide variety of substrates, including phosphoramidate prodrugs of nucleoside analogs such as 2', 3'-didehydro 3'-deoxythymidine (d4T) and 3'-azido 3'-deoxythymidine (AZT) (Balzarini et al., 1996; Valette et al., 1996). The current data indicate that PLCE displays relatively low specific activity against both GS-7340 and the other TFV prodrugs. However, because of its abundance in liver, the carboxylesterases may contribute to the hepatic clearance of this class of TFV prodrugs.

The present study demonstrates that multiple intracellular proteolytic enzymes exist that are capable of hydrolyzing GS-7340 and other TFV amidate prodrugs. More than 553 different human proteases have been identified thus far (Puente et al., 2003), many of them being specifically expressed only in a narrow range of tissues or body compartments. Even though cathepsin A is present in a variety of tissues, its expression is elevated in cells of the immune system (data not shown). This makes it an attractive target for the activation of prodrugs for HIV therapy. In contrast, other proteases have much more restricted expression patterns. For example, leukocyte elastase, proteinase 3, and cathepsin G are present in neutrophils (Owen and Campbell, 1999), granzymes in natural killer cells and cytotoxic T lymphocytes (Kam et al., 2000), and tryptases and chymases in mast cells (Sayama et al., 1987). Tissue-specific expression of prodrug hydrolases may allow for the rational design of nucleotide phosphonoamidate prodrugs that can be preferentially activated in relevant target cells. Furthermore, many viruses and other infectious agents produce their own specific proteases in host cells. These enzymes can be targeted for the activation of specific prodrugs, enhancing the therapeutic selectivity toward infected cells or tissues. Detailed knowledge about the localization and substrate preference of the targeted prodrug-activating proteases as well as hydrolytic enzymes potentially affecting absorption and/or metabolic elimination (e.g., digestive proteases and liver esterases) should facilitate the rational design of novel prodrugs with improved pharmacokinetic and therapeutic properties.

	Footnotes

 
ABBREVIATIONS: TFV, tenofovir (9-[(2-phosphonomethoxy)propyl]adenine); HIV, human immunodeficiency virus; TFVpp, TFV diphosphate; TDF, tenofovir disoproxil fumarate; GS-7340, 9-[(R)-2-[[(S)-[[(S)-1-(isopropoxycarbonyl)ethyl]amino] phenoxyphosphinyl]-methoxy]propyl]adenine; PBMC, peripheral blood mononuclear lymphocyte; DTT, dithiothreitol; BSA, bovine serum albumin; Pr3, proteinase 3; LE, leukocyte elastase; PE, pancreatic elastase; PLCE, porcine liver carboxylesterase; DFP, diisopropyl fluorophosphate; -ABA, -aminobutyric acid; E64, trans-epoxysuccinyl-L-leucylamido(4-guanidino)butane.

Address correspondence to: Gabriel Birkus, Gilead Sciences, Inc., 362 Lakeside Drive, Foster City, CA 94404. E-mail: gabriel_birkus{at}gilead.com

	References

Top Abstract Materials and Methods Results Discussion References

 
Antoniou T, Park-Wyllie LY, and Tseng AL (2003) Tenofovir: a nucleotide analog for the management of human immunodeficiency virus infection. Pharmacotherapy 23: 29-43.[CrossRef][Medline]

Balzarini J, Karlsson A, Aquaro S, Perno CF, Cahard D, Naesens L, De Clercq E, and McGuigan C (1996) Mechanism of anti-HIV action of masked alaninyl d4T-MP derivatives. Proc Natl Acad Sci U S A 93: 7295-7299.[Abstract/Free Full Text]

Barrett AJ, Kembhavi AA, Brown MA, Kirschke H, Knight CG, Tamai M, and Hanada K (1982) L-trans-Epoxysuccinyl-leucylamido(4-guanidino)butane (E-64) and its analogues as inhibitors of cysteine proteinases including cathepsins B, H and L. Biochem J 201: 189-198.[Medline]

Bergman M and Fruton JS (1943) The specificity of proteinases. Adv Enzymol 1: 63-98.

Birkus G, Wang R, Liu X, Kutty N, MacArthur H, Cihlar T, Gibbs C, Swaminathan S, Lee W, and McDermott M (2007) Cathepsin A is the major hydrolase catalyzing the intracellular hydrolysis of the antiretroviral nucleotide phosphonoamidate prodrugs GS-7340 and GS-9131. Antimicrob Agents Chemother 51: 543-550.[Abstract/Free Full Text]

Blott EJ and Griffiths M (2002) Secretory lysosomes. Nat Rev Mol Cell Biol 3: 122-131.[CrossRef][Medline]

Blow AM (1977) Action of human lysosomal elastase on the oxidized B chain of insulin. Biochem J 161: 13-16.[Medline]

Borregaard N and Cowland JB (1997) Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 89: 3503-3521.[Free Full Text]

Chou TF, Baraniak J, Kaczmarek R, Zhou X, Cheng J, Ghosh B, and Wagner CR (2007) Phosphoramidate pronucleotides: a comparison of the phosphoramidase substrate specificity of human and Escherichia coli histidine triad nucleotide binding proteins. Mol Pharm 4: 208-217.[CrossRef][Medline]

Conner GE (2004) Cathepsin D, in Handbook of Proteolytic Enzymes (Barret AJ, Rawlings ND, and Woessner JF, eds) pp 43-52, Elsevier, Amsterdam.

De Clercq E (2003) Clinical potential of the acyclic nucleoside phosphonates cidofovir, adefovir, and tenofovir in treatment of DNA virus and retrovirus infections. Clin Microbiol Rev 16: 569-596.[Abstract/Free Full Text]

Eisenberg EJ, He GX, and Lee WA (2001) Metabolism of GS-7340, a novel phenyl monophosphoramidate intracellular prodrug of PMPA, in blood. Nucleosides Nucleotides Nucleic Acids 20: 1091-1098.[CrossRef][Medline]

Grim SA and Romanelli F (2003) Tenofovir disoproxil fumarate. Ann Pharmacother 37: 849-859.[Abstract/Free Full Text]

Harper JW, Cook RR, Roberts CJ, McLaughlin BJ, and Powers JC (1984) Active site mapping of the serine proteases human leukocyte elastase, cathepsin G, porcine pancreatic elastase, rat mast cell proteases I and II. Bovine chymotrypsin A alpha, and Staphylococcus aureus protease V-8 using tripeptide thiobenzyl ester substrates. Biochemistry 23: 2995-3002.[CrossRef][Medline]

Kam CM, Hudig D, and Powers JC (2000) Granzymes (lymphocyte serine proteases): characterization with natural and synthetic substrates and inhibitors. Biochim Biophys Acta 1477: 307-323.[CrossRef][Medline]

Kam CM, Kerrigan JE, Dolman KM, Goldschmeding R, Von dem Borne AE, and Powers JC (1992) Substrate and inhibitor studies on proteinase 3. FEBS Lett 297: 119-123.[CrossRef][Medline]

Kärgel HJ, Dettmer R, Etzold G, Kirschke H, Bohley P, and Langner J (1980) Action of cathepsin L on the oxidized B-chain of bovine insulin. FEBS Lett 114: 257-260.[CrossRef][Medline]

Koga H, Mori N, Yamada H, Nishimura Y, Tokuda K, Kato K, and Imoto T (1992) Endoand aminopeptidase activities of rat cathepsin H. Chem Pharm Bull 40: 965-970.[Medline]

Lange S, Musidlowska A, Schmidt-Dannert C, Schmitt J, and Bornscheuer UT (2001) Cloning, functional expression, and characterization of recombinant pig liver esterase. Chembiochem 2: 576-582.[CrossRef][Medline]

Lee WA, Gong-Xin H, Eisenberg E, Cihlar T, Swaminathan S, Mulato A, and Cundy K (2005) Selective intracellular activation of a novel prodrug of the human immunodeficiency virus reverse transcriptase inhibitor tenofovir leads to preferential distribution and accumulation in lymphatic tissue. Antimicrob Agents Chemother 49: 1898-1906.[Abstract/Free Full Text]

Matsuzawa Y and Hostetler KY (1980) Studies on drug-induced lipidosis: subcellular localization of phospholipid and cholesterol in the liver of rats treated with chloroquine or 4,4'-bis (diethylaminoethoxy)alpha, beta-diethyldiphenylethane. J Lipid Res 21: 202-214.[Abstract]

Merion M and Sly WS (1983) The role of intermediate vesicles in the adsorptive endocytosis and transport of ligand to lysosomes by human fibroblasts. J Cell Biol 96: 644-650.[Abstract/Free Full Text]

Ouar Z, Bens M, Vignes C, Paulais M, Pringel C, Fleury J, Cluzeaud F, Lacave R, and Vandewalle A (2003) Inhibitors of vacuolar H+-ATPase impair the preferential accumulation of daunomycin in lysosomes and reverse the resistance to anthracyclines in drug-resistant renal epithelial cells. Biochem J 370: 185-193.[CrossRef][Medline]

Owen CA and Campbell EJ (1999) The cell biology of leukocyte-mediated proteolysis. J Leukocyte Biol 65: 137-150.[Abstract]

Pennington RJ (1961) Biochemistry of dystrophic muscle Mitochondrial succinatetetrazolium reductase and adenosine triphosphatase. Biochem J 80: 649-654.[Medline]

Polanowska J, Krokoszynska I, Czapinska H, Watorek W, Dadlez M, and Otlewski J (1998) Specificity of human cathepsin G. Biochim Biophys Acta 1386: 189-198.[CrossRef][Medline]

Powers JC and Harper JW (1986) Inhibitors of metalloproteases, in Proteinase Inhibitors (Barrett AJ and Salvesen G eds) pp 219-298, Elsevier, Amsterdam.

Powers JC, Asgian JL, Ekici OD, and James KE (2002) Irreversible inhibitors of serine, cysteine, and threonine proteases. Chem Rev 102: 4639-4750.[CrossRef][Medline]

Powers JC, Tanaka JW, Harper Y, Minematsu Y, Barker L, Lincoln D, Crumley KV, Fraki JE, Schechter NM, Lazarus GL, et al. (1985) Mammalian chymotrypsin-like enzymes. Comparative reactivities of rat mast cell proteases, human and dog skin chymases, and human cathepsin G with peptide 4-nitroanilide substrates and with peptide chloromethyl ketone and sulfonyl fluoride inhibitors. Biochemistry 24: 2048-2058.[CrossRef][Medline]

Press EM, Porter RR, and Cebra J (1960) The isolation and properties of a proteolytic enzyme, cathepsin D, from bovine spleen. Biochem J 74: 501-514.[Medline]

Puente XS, Sánchez LM, Overall CM, and López-Otín C (2003) Human and mouse proteases: a comparative genomic approach. Nat Rev Genet 4: 544-558.[CrossRef][Medline]

Rich DH, Boparai AS, and Bernatowicz MS (1982) Synthesis of a 3-oxo-4(S)-amino acid analog of pepstatin. A new inhibitor of carboxyl (acid) proteases. Biochem Biophys Res Commun 104: 1127-1133.[CrossRef][Medline]

Robbins BL, Greenhaw J, Connelly MC, and Fridland A (1995) Metabolic pathways for activation of the antiviral agent 9-(2-phosphonylmethoxyethyl)adenine in human lymphoid cells. Antimicrob Agents Chemother 39: 2304-2308.[Abstract]

Rothe M and Dodt J (1992) Studies on the aminopeptidase activity of rat cathepsin H. Eur J Biochem 210: 759-764.[Medline]

Satoh T and Hosokawa M (1998) The mammalian carboxylesterases: from molecules to functions. Annu Rev Pharmacol Toxicol 38: 257-288.[CrossRef][Medline]

Sayama S, Iozzo RV, Lazarus GS, and Schechter NM (1987) Human skin chymotrypsin-like proteinase chymase: Subcellular localization to mast cell granules and interaction with heparin and other glycosaminoglycans. J Biol Chem 262: 6808-6815.[Abstract/Free Full Text]

Schechter I and Berger A (1967) On the active site of proteases 3. Mapping the active site of papain; specific peptide inhibitors of papain. Biochem Biophys Res Commun 27: 157-162.[CrossRef][Medline]

Suo Z and Johnson KA (1998) Selective inhibition of HIV-1 reverse transcriptase by an antiviral inhibitor, (R)-9-(2-phosphonylmethoxypropyl)adenine. J Biol Chem 273: 27250-27258.[Abstract/Free Full Text]

Turk B, Turk V, and Turk D (1997) Structural and functional aspects of papain-like cysteine proteinases and their protein inhibitors. Biol Chem 378: 141-150.[Medline]

Valette G, Pompon A, Girardet JL, Cappellacci L, Franchetti P, Grifantini M, La Colla P, Loi AG, Perigaud C, Gosselin G, et al. (1996) Decomposition pathways and in vitro HIV inhibitory effects of isoddA pronucleotides: toward a rational approach for intracellular delivery of nucleoside 5'-monophosphates. J Med Chem 39: 1981-1990.[CrossRef][Medline]

Wex T, Bühling F, Wex H, Günther D, Malfertheiner P, Weber E, and Brömme D (2001) Human cathepsin W, a cysteine protease predominantly expressed in NK cells, is mainly localized in the endoplasmic reticulum. J Immunol 167: 2172-2178.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.045526v1 74/1/92    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Birkus, G. Articles by Cihlar, T. Search for Related Content PubMed PubMed Citation Articles by Birkus, G. Articles by Cihlar, T.