Cary Hsu

Paraoxonase Active Site Required for Protection Against LDL Oxidation Involves Its Free Sulfhydryl Group and Is Different From That Required for Its Arylesterase/Paraoxonase Activities Selective Action of Human Paraoxonase Allozymes Q and R

Human serum paraoxonase (PON 1) exists in 2 major polymorphic forms (Q and R), which differ in the amino acid at position 191 (glutamine and arginine, respectively). These PON allozymes hydrolyze organophosphates and aromatic esters, and both also protect LDL from copper ion-induced oxidation. We have compared purified serum PONs of both forms and evaluated their effects on LDL oxidation, in respect to their arylesterase/paraoxonase activities. Copper ion-induced LDL oxidation, measured by the production of peroxides and aldehydes after 4 hours of incubation, were reduced up to 61% and 58%, respectively, by PON Q, but only up to 46% and 38%, respectively, by an equivalent concentration of PON R. These phenomena were PON-concentration dependent. Recombinant PON Q and PON R demonstrated similar patterns to that shown for the purified serum allozymes. PON Q and PON R differences in protection of LDL against oxidation were further evaluated in the presence of glutathione peroxidase (GPx). GPx (0.1 U/mL) alone reduced copper ion-induced LDL oxidation by 20% after 4 hours of incubation. The addition of PON R to the above system resulted in an additive inhibitory effect on LDL oxidation, whereas PON Q had no such additive effect. The 2 PON allozymes also differed by their ability to inhibit initiation, as well as propagation, of LDL oxidation. PON Q was more efficient in blocking LDL oxidation if added when oxidation was initiated, whereas PON R was more potent when added 1 hour after the initiation of LDL oxidation. These data suggest that the 2 allozymes act on different substrates. Both PON allozymes were also able to reduce the oxidation of phospholipids and cholesteryl ester. PON Q arylesterase activity was reduced after 4 hours of LDL oxidation by only 28%, whereas the arylesterase activity of PON R was reduced by up to 55%. Inactivation of the calcium-dependent PON arylesterase activity by using the metal chelator EDTA, or by calcium ion removal on a Chelex column, did not alter PONs ability to inhibit LDL oxidation. However, blockage of the PON free sulfhydryl group at position 283 with p-hydroxymercuribenzoate inhibited both its arylesterase activity and its protection of LDL from oxidation. Recombinant PON mutants in which the PON free sulfhydryl group was replaced by either alanine or serine were no longer able to protect against LDL oxidation, even though they retained paraoxonase and arylesterase activities. Overall, these studies demonstrate that PONs arylesterase/paraoxonase activities and the protection against LDL oxidation do not involve the active site on the enzyme in exactly the same way, and PONs ability to protect LDL from oxidation requires the cysteine residue at position 283.

Human Serum Paraoxonase/Arylesterase’s Retained Hydrophobic N-Terminal Leader Sequence Associates With HDLs by Binding Phospholipids Apolipoprotein A-I Stabilizes Activity

In serum, human paraoxonase/arylesterase (PON1) is found exclusively associated with high density lipoprotein (HDL) and contributes to its antiatherogenic properties by inhibiting low density lipoprotein (LDL) oxidation. Difficulties in purifying PON1 from apolipoprotein A-I (apoA-I) suggested that PON1s association with HDL may occur through a direct binding between these 2 proteins. An unusual property of PON1 is that the mature protein retains its hydrophobic N-terminal signal sequence. By expressing in vitro a mutant PON1 with a cleavable N-terminus, we demonstrate that PON1 associates with lipoproteins through its N-terminus by binding phospholipids directly rather than binding apoA-I. Nonetheless, apoA-I stabilized arylesterase activity more than did phospholipid alone, apoA-II, or apoE. Consequently, we studied the role of apoA-I in PON1 expression and HDL association in mice genetically deficient in apoA-I. Though present in HDL fractions at decreased levels, PON1 arylesterase activity was less stable than in control mice. Furthermore, PON1 could be competitively removed from HDL by phospholipids, suggesting that PON1s retained N-terminal peptide allows transfer of the enzyme between phospholipid surfaces. Thus, our data suggest that PON1 is stabilized by apoA-I, and its binding to HDL and physiological distribution are dependent on the direct binding of the retained hydrophobic N-terminus to phospholipids optimally presented in association with apoA-I.

Primary Human T Lymphocytes Engineered with a Codon-Optimized IL-15 Gene Resist Cytokine Withdrawal-Induced Apoptosis and Persist Long-Term in the Absence of Exogenous Cytokine

IL-15 is a common γ-chain cytokine that has been shown to be more active than IL-2 in several murine cancer immunotherapy models. Although T lymphocytes do not produce IL-15, murine lymphocytes carrying an IL-15 transgene demonstrated superior antitumor activity in the immunotherapy of B16 melanoma. Thus, we sought to investigate the biological impact of constitutive IL-15 expression by human lymphocytes. In this report we describe the generation of a retroviral vector encoding a codon-optimized IL-15 gene. Alternate codon usage significantly enhanced the translational efficiency of this tightly regulated gene in retroviral vector-transduced cells. Activated human CD4+ and CD8+ human lymphocytes expressed IL-15Rα and produced high levels of cytokine upon retroviral transduction with the IL-15 vector. IL-15-transduced lymphocytes remained viable for up to 180 days in the absence of exogenous cytokine. IL-15 vector-transduced T cells showed continued proliferation after cytokine withdrawal and resistance to apoptosis while retaining specific Ag recognition. In the setting of adoptive cell transfer, IL-15-transduced lymphocytes may prolong lymphocyte survival in vivo and could potentially enhance antitumor activity.

Efficient nonviral Sleeping Beauty transposon-based TCR gene transfer to peripheral blood lymphocytes confers antigen-specific antitumor reactivity

Genetically engineered lymphocytes hold promise for the treatment of genetic disease, viral infections and cancer. However, current methods for genetic transduction of peripheral blood lymphocytes rely on viral vectors, which are hindered by production and safety-related problems. In this study, we demonstrated an efficient novel nonviral platform for gene transfer to lymphocytes. The Sleeping Beauty transposon-mediated approach allowed for long-term stable expression of transgenes at ∼50% efficiency. Utilizing transposon constructs expressing tumor antigen-specific T-cell receptor genes targeting p53 and MART-1, we demonstrated sustained expression and functional reactivity of transposon-engineered lymphocytes on encountering target antigen presented on tumor cells. We found that transposon- and retroviral-modified lymphocytes had comparable transgene expression and phenotypic function. These results demonstrate the promise of nonviral ex vivo genetic modification of autologous lymphocytes for the treatment of cancer and immunologic disease.

T-Cell Receptor Gene Therapy of Established Tumors in a Murine Melanoma Model

Adoptive cell transfer therapy using tumor-infiltrating lymphocytes for patients with metastatic melanoma has demonstrated significant objective response rates. One major limitation of these current therapies is the frequent inability to isolate tumor-reactive lymphocytes for treatment. Genetic engineering of peripheral blood lymphocytes with retroviral vectors encoding tumor antigen-specific T-cell receptors (TCRs) bypasses this restriction. To evaluate the efficacy of TCR gene therapy, a murine treatment model was developed. A retroviral vector was constructed encoding the pmel-1 TCR genes targeting the B16 melanoma antigen, gp100. Transduction of C57BL/6 lymphocytes resulted in efficient pmel-1 TCR expression. Lymphocytes transduced with this retrovirus specifically recognized gp100-pulsed target cells as measured by interferon-γ secretion assays. Upon transfer into B16 tumor-bearing mice, the genetically engineered lymphocytes significantly slowed tumor development. The effectiveness of tumor treatment was directly correlated with the number of TCR-engineered T cells administered. These results demonstrated that TCR gene therapy targeting a native tumor antigen significantly delayed the growth of established tumors. When C57BL/6 lymphocytes were added to antigen-reactive pmel-1 T cells, a reduction in the ability of pmel-1 T cell to treat B16 melanomas was seen, suggesting that untransduced cells may be deleterious to TCR gene therapy. This model may be a powerful tool for evaluating future TCR gene transfer-based strategies.

Lentiviral Vector Design for Optimal T Cell Receptor Gene Expression in the Transduction of Peripheral Blood Lymphocytes and Tumor-Infiltrating Lymphocytes

Lentiviral vectors containing promoters of distinct origins, that is, strong viral promoters (cytomegalovirus [CMV] and murine stem cell virus [MSCV]), a cellular promoter (phosphoglycerate kinase [PGK]), and two composite promoters (CAG [a composite promoter sequence comprised of the CMV enhancer and portions of the chicken beta-actin promoter and the rabbit beta-globin gene] and SV40/CD43), were used to evaluate green fluorescent protein (GFP) reporter gene expression in human primary peripheral blood lymphocytes (PBLs) and tumor-infiltrating lymphocytes (TILs). In PBLs, vectors containing the MSCV promoter were found to be optimal for expression in both minimally stimulated and highly activated lymphocytes. The stability of gene expression was monitored for up to 7 weeks in culture and the MSCV promoter-containing vector was found to be comparable to the cellular PGK promoter-containing vector. The MSCV promoter-containing lentiviral vector was also the most active in transduced TILs and these cells retained biological activity as measured by antimelanoma antigen reactivity. Using the knowledge gained in comparing individual promoters, a series of two-gene-containing lentiviral vectors was constructed in an attempt to produce the alpha and beta chains of antitumor antigen T cell receptors (TCRs). Dual-promoter or internal ribosome entry site (IRES)-containing vector designs were evaluated and found to be unable to produce both chains of the TCR in amounts that led to significant biological activity. In contrast, if the alpha and beta chains were linked by a 2A ribosomal skip peptide, both proper TCR chain pairing and biologically activity were observed. This paper emphasizes the need to optimize both promoter function and protein synthesis in constructs that require stoichiometric production of multiple protein subunits.

Evidence that several conserved histidine residues are required for hydrolytic activity of human paraoxonase/arylesterase.

Recent evidence has been acquired that implicates an important role for several histidine residues in the hydrolytic mechanisms of human paraoxonase/arylesterase (PON1). Following titration with diethylpyrocarbonate (DEPC), both human serum and recombinant human type Q PON1 were inhibited in respect to their hydrolytic activity in a dose-responsive manner. Human PON1 treated with varying concentrations lost hydrolytic activity, and with each histidine modified, there was an exponential drop in hydrolytic activity. The reaction was followed spectrophotometrically at 244 nm. Recombinant wild-type and C283A PON1 enzymes inhibited with DEPC and subsequently treated with hydroxylamine had partial restoration of activity. The C283A mutant lacks a free sulfhydryl group, indicating that its inactivation is due to histidine specific modification. The dose response and time course of inactivation as well as the extent of reactivation by hydroxylamine were similar for both the wild-type and mutant recombinant enzymes. Mutants of PON1 containing an asparagine substituted for each of several conserved histidine residues lost hydrolytic activity for each single substitution. The mutants of PON1 constructed and assayed for arylesterase activity were H114N, H133N, and H284N. Each single aminoacid substitution rendered the enzyme catalytically inactive. These two pieces of evidence implicate an important role for several histidine residues in the hydrolytic mechanism of PON1. Although it is unusual for a calcium dependent enzyme to require histidines for its catalytic activity, acquired data suggest such a circumstance.

Properties of the retained N-terminal hydrophobic leader sequence in human serum paraoxonase/arylesterase

Human serum paraoxonase/arylesterase (PON1) is HDL-associated and appears to protect low density lipoproteins (LDL) from oxidation. Mature PON1 retains its N-terminal hydrophobic signal sequence, which may be needed for binding to HDL. By site-directed mutagenesis, we created a mutant PON1 (A19A20) with a cleavable N-terminus to determine if this peptide mediated binding to lipoproteins. As a model system, we studied binding of mutant and wild type PON1s to lipoproteins in fetal bovine serum-containing expression medium and found that the wild type recombinant enzyme associated with lipoproteins whereas the A19A20 mutant did not. These results show that the N-terminus is required for binding to either apolipoproteins or phospholipids. Furthermore, we showed that wild type enzyme can bind to phospholipids directly without apolipoproteins. To determine if lipid binding is a requirement for PON1s protection against LDL oxidation, we used a copper ion-induced oxidation system and found that the wild type enzyme and A19A20 mutant showed similar reductions in both peroxide and aldehyde formation. We conclude that PON1 depends upon its N-terminal hydrophobic peptide for its association with serum lipoproteins.

767. Genetic Engineering of an Immortal Human CD8+ T Cell Line

Immunotherapy using adoptively transferred autologous tumor specific T lymphocytes in patients has been shown to be effective in treating EBV-induced post-transplant lymphoproliferative diseases and melanoma. These treatments involve concurrent treatment with high dose interleukin-2 as an adjunct to promote the growth and maintenance of these cells in vivo. Persistence of these tumor-specific T-cells has been correlated with objective response. The requirement for the generation of autologous anti-tumor antigen specific T cells is a major limitation in widespread use of adoptive immunotherapy.