Rapid and efficient nonviral gene delivery of CD154 to primary chronic lymphocytic leukemia cells.

Interactions between CD40 and CD40 ligand (CD154) are essential in the regulation of both humoral and cellular immune responses. Forced expression of human CD154 in B chronic lymphocytic leukemia (B-CLL) cells can upregulate costimulatory and adhesion molecules and restore antigen-presenting capacity. Unfortunately, B-CLL cells are resistant to direct gene manipulation with most currently available gene transfer systems. In this report, we describe the use of a nonviral, clinical-grade, electroporation-based gene delivery system and a standard plasmid carrying CD154 cDNA, which achieved efficient (64+/-15%) and rapid (within 3 h) transfection of primary B-CLL cells. Consistent results were obtained from multiple human donors. Transfection of CD154 was functional in that it led to upregulated expression of CD80, CD86, ICAM-I and MHC class II (HLA-DR) on the B-CLL cells and induction of allogeneic immune responses in MLR assays. Furthermore, sustained transgene expression was demonstrated in long-term cryopreserved transfected cells. This simple and rapid gene delivery technology has been validated under the current Good Manufacturing Practice conditions, and multiple doses of CD154-expressing cells were prepared for CLL patients from one DNA transfection. Vaccination strategies using autologous tumor cells manipulated ex vivo for patients with B-CLL and perhaps with other hematopoietic malignancies could be practically implemented using this rapid and efficient nonviral gene delivery system.

A non-viral gene delivery system designed for clinical use.

Gene delivery can be accomplished using non-viral systems, and these have received increased attention These include ex vivo transfection of cells using an electric field to induce transient cell-membrane permeability (electroporation). This approach has the distinct advantage of not requiring the inclusion of a secondary agent (e.g. a lipid, viral package or carrier protein) any of which can be immunogenic or toxic. Available electroporation systems utilize a low volume (<1 mL) processing chamber and are open systems. The MaxCyte system employs a continuous flow design and can very rapidly process volumes ranging from 0.02 mL to >1 L. Transgenes for markers (eGFP) and functional proteins (e.g., cytokines, angiogenic factors) have been loaded in plasmids up to 14 kB in size. With appropriate application of pre- and post-processing cell manipulations, very satisfactory loading efficiencies and cell viability have been obtained. Cells can be processed with multiple plasmids, resulting in expression of the corresponding number of gene products. This capability has been considered for therapeutic and bioprocessing applications. The MaxCyte system was designed specifically for ex vivo clinical applications. The electrodes are manufactured of special materials and under precise conditions, in order to eliminate potential risks from electrolytic effects. The processing chamber and associated containers can be provided as disposable, sterile, closed (or functionally closed) systems-quite similar to the disposable harnesses used with cell separators. This system is thus suitable for integration into a current good manufacturing practice environment.

Red blood cell substitutes: evolution of approaches to demonstrating efficacy.

There has been a striking advancement in our understanding of red cell substitutes over the past decade. Throughout this period, regulatory oversight influenced many aspects of product development. The approach to demonstrating efficacy was an important example. I will review some events of the decade so that we may consider the question, "If we had it to do over again, could we do any better?" At the start of the decade, there were sufficient data on the mechanism of hemoglobin-induced renal toxicity to reach consensus among investigators that the content of alpha beta-dimers in any hemoglobin product must be strictly limited. An array of additional, unexpected reactions were seen in early clinical studies, reactions that led to several conclusions: hemoglobin is a pharmacologically active material; more basic studies are needed; and it would all be so much easier if communications between investigators were open and unhindered. Meaningful studies of efficacy could not be approached until the safety issues had been solved or until the community was convinced that they were not serious. The initial approach was based on the meager experience gained with previous products, e.g., Fluosol. That experience led to policies which required that, for example, if the product was to be used as a red cell substitute, it must be compared with red cells, and if it was to be used as an oxygen carrier, it must be shown to support organ function. Endpoints of clinical trials were based on delivery of a clear clinical benefit to the patient--surrogate endpoints could be used only if their relationship to clinical benefit had been demonstrated.

Effects of polymerization on the hypertensive action of diaspirin cross-linked hemoglobin in rats.

It is believed that the hypertensive effect of diaspirin crosslinked hemoglobin, a viable blood substitute, can be resolved by polymerization, which reduces the diffusion of this derivative into the interstitial space between nitric oxide-producing endothelium and the target vascular smooth muscle. We studied the systemic and renal responses to infusion of three cell-free human hemoglobins in anesthetized isovolemic rats: unmodified (HbA0), crosslinked (alpha-DBBF), and polymerized crosslinked (poly alpha-DBBF). HbA0 produced a significant increase in mean arterial blood pressure (MAP) throughout the 60-minute infusion. alpha-DBBF, on the other hand, produced a more marked and prolonged increase in MAP over 120 minutes. Only a moderate increase in MAP was observed in rats after a 30-minute infusion with poly alpha-DBBF. The extent of renal insufficiency produced by these proteins, as determined by the glomerular filtration rate, was in the following order: HbA0 > poly alpha-DBBF > alpha-DBBF. Infusion of poly alpha-DBBF, under hypovolemic but not isovolemic conditions in rats, produced an increase in heart rate, cardiac output, and stroke volume and a decrease in total peripheral resistance after 60 minutes. Chemical polymerization to increase the size of alpha-DBBF does not appear to improve its hemodynamic properties in rats, especially under partial exchange transfusion, a more clinically relevant indication for a hemoglobin-based blood substitute.

Endogenous ADP prevents PGE1-induced tyrosine dephosphorylation of focal adhesion kinase in thrombin-activated platelets.

Prostaglandin E1(PGE1) inhibits tyrosine phosphorylation induced by low thrombin concentration (0.05 U/ml), but this is overcome by a high thrombin (2.0 U/ml) concentration. Thromboxane A2 and ADP are endogenous platelet agonists released during platelet activation which potentiate platelet responses. We investigated how these endogenous agonists influenced the effects of PGE1 on thrombin (2.0 U/ml)-induced tyrosine phosphorylation by removing released ADP with apyrase (2.0 U/ml) and by inhibiting thromboxane A2 synthesis with indomethacin (1 microM). Adding PGE1 (1 microM) before thrombin in apyrase/indomethacin(A/I)-treated platelets selectively prevented thrombin-induced tyrosine phosphorylation of a 117 kDa protein while other substrates were not affected. This selective effect was evident only in the presence of apyrase and was not dependent on indomethacin. Addition of PGE1 to A/I-treated platelets after thrombin also caused selective tyrosine dephosphorylation of the 117 kDa protein. Conditions which prevented thrombin-induced 117 kDa protein tyrosine phosphorylation also decreased fibrinogen binding to platelets. The 117 kDa protein was identified as the focal adhesion kinase (FAK) by immunoprecipitation with a monoclonal antibody to FAK and by absence of its tyrosine phosphorylation in the presence of RGDS peptide which inhibits fibrinogen binding and platelet aggregation. Thus, released endogenous ADP selectively prevents PGE1-mediated tyrosine dephosphorylation of platelet FAK most likely by stabilizing fibrinogen binding to platelets.

Considerations of pool size in the manufacture of plasma derivatives.

BACKGROUND: The pooling of human plasma from many donors for the purpose of manufacturing therapeutic proteins increases the risk of exposing recipients of these proteins to pathogens that may contaminate 1 or a few units included in the pool. STUDY DESIGN AND METHODS: This risk is estimated for a range of manufacturing scales that would derive material from a varied number of donors and for a number of hypothetical infectious agents that may exist in the donor population over a wide range of prevalence. Risk is also calculated both for recipients of single doses of a plasma protein and for those who depend on long-term treatment with plasma derivatives. RESULTS: Risk of exposure increases with pool size and the prevalence of the agent in question and accumulates with repeated treatments with material manufactured from different pools. CONCLUSION: Reducing pool size would at best decrease this risk in proportion to the reduction in manufacturing scale. However, for individuals requiring repeated or continuous treatments, the risk of exposure to all but the rarest infectious agents would be only minimally affected, even by large reductions in manufacturing scale.

Fibrin sealant: summary of a conference on characteristics and clinical uses.

The 2-day conference clearly outlined the formulations of products that are being developed or are commercially available in Europe. The major difference between products in the United States and those in Europe is that US manufacturers are preparing fibrin sealant that does not contain aprotinin, epsilon amino caproic acid, or any other type of antifibrinolytic agent, whereas antifibrinolytic agents are included in all such preparations used in Europe. The conference provided no clear consensus that such agents are essential to the efficacy of the product. Although many investigators believe in the clinical benefit of fibrin sealant, most of the studies to demonstrate efficacy have not been performed in a well-controlled fashion. However, fibrin sealant, if found in a controlled trial to have clinical efficacy, could be approved by the FDA for a narrow indication. Opportunities remain for greater exploration of different forms of the product, not only as a hemostatic agent, but as an adjunct to wound healing and as a matrix for delivery of drugs and proteins with other biologic activities.

Redox reactivity of modified hemoglobins with hydrogen peroxide and nitric oxide: toxicological implications.

The rapid unloading of oxygen to tissue and the prevention of subunit dissociation have been the main concerns in the search for an effective hemoglobin-based red cell substitute. The presence of redox active iron however, raises some questions about its potential to enter into reactions that mediate the formation of cytotoxic oxygen free radicals. We tested the propensity of modified hemoglobins to undergo oxidative damage by peroxide (H2O2). We found differences in their susceptibility to oxidative modification and in their ability to form the highly cytotoxic ferryl species. This protein-associated oxidant may be a physiologically important contributor to reperfusion injury. Another potential mechanism of toxicity involves the reaction of cell-free hemoglobin with endothelium derived nitric oxide (NO). Marked hypertensive responses in intact animals infused with some of these hemoglobins were reported. Cell-free hemoglobin has the potential to bind the endothelial generated NO yielding methemoglobin and nitrate, an extremely rapid reaction in vivo. We describe subsequent redox reactions between NO and methemoglobin which may further deplete NO as a biological transducer, leading to greater effects on the extent of endothelial-dependent responses. The consequences of a potential linkage between oxidative toxicity of cell-free hemoglobin and its interaction with NO is addressed.

Econazole inhibits thapsigargin-induced platelet calcium influx by mechanisms other than cytochrome P-450 inhibition.

Cytochrome P-450 has been suggested as a mediator of the signal between depleted platelet calcium stores and an increase in plasma membrane permeability to calcium which follows depletion of the stores. This hypothesis is based on the observations that inhibitors of cytochrome P-450, such as the imidazole antifungal agents, also inhibit influx of a calcium surrogate (manganese) into calcium-depleted platelets. We tested the effects of econazole and of a cytochrome P-450 inhibitor, carbon monoxide (CO), on thapsigargin (TG)-induced platelet 45Ca2+ influx. TG specifically depletes internal calcium stores and activates store-regulated calcium influx. Econazole blocked 45Ca2+ influx when it was added before TG (IC50 11 microM). Econazole at a concentration (20 microM) that inhibited 83% of TG-induced calcium influx was not inhibitory to TG-induced calcium efflux from 45Ca(2+)-loaded platelets, and did not affect calcium fluxes in resting platelets. This econazole concentration was also inhibitory to calcium influx even when it was added after the stores had been calcium-depleted by EGTA and TG for 15 min and the signal to increase calcium influx had already been generated. Inhibition of cytochrome P-450 with CO bubbled through platelet suspensions did not change calcium influx in resting cells and potentiated TG-induced calcium influx (160% of control calcium accumulation at 20 min). This effect appeared to be concentration-dependent, such that a 5 min exposure to CO produced a greater influx potentiation than a 3 min exposure. These observations indicate that (1) cytochrome P-450 does not mediate store-regulated calcium influx, and (2) econazole probably inhibits store-regulated calcium influx by an alternative mechanism, such as interaction with plasma membrane calcium channels.

Oxidation reactions of human, opossum (Didelphis virginiana) and spot (Leiostomus xanthurus) hemoglobins: a search for a correlation with some structural-functional properties.

1. Relative to human HbA, opossum (Didelphis virginiana) hemoglobin was found to be more susceptible to autoxidation. While the initial rate of autoxidation of spot (Leiostomus xanthurus) hemoglobin is close to that of HbA, complete oxidation occurs in 50 hr. 2. Direct addition of hydrogen peroxide (H2O2) induced oxidation of hemoglobins in a definite order: spot Hb > HbA > opossum Hb. Excess H2O2 led to heme degradation and precipitation that occurred much faster for spot Hb than the case with other proteins. 3. Exposure of hemoglobins to a continuous flux of H2O2, generated by the glucose/glucose oxidase system, induced the formation of heterogeneous protein-associated oxidation products. 4. Differential reactivity among these hemoglobins under the same or different oxidative conditions, with respect to methemoglobin formation and stability of the ferric form, may reflect the differences in the local heme environment of these proteins.

Nitric oxide binding to human ferrihemoglobins cross-linked between either alpha or beta subunits.

We have examined the interactions between nitric oxide (NO) and oxidized human hemoglobin, comparing the behavior of unmodified HbA0 with that of two chemically modified hemoglobins. The latter are promising red cell substitute candidates due to their lower oxygen affinity and greater stability as tetramers. The modified forms examined were HbA-DBBF, cross-linked between the alpha chains with bis(3,5-dibromosalicyl) fumarate, and HbA-FMDA, modified between the beta chains with fumaryl monodibromoaspirin. NO binding to the oxidized forms of these hemoglobins is biphasic, due to the differing reactivities of alpha and beta chains. The structural modifications result in altered rate constants for NO binding to both alpha and beta chains. The affinity of the ferric hemes for NO is not correlated with their oxygen affinities in the ferrous state. In a much slower first-order process, the ferric hemes of HbA become reduced. Faster and more heterogeneous kinetics are observed for reduction of the modified hemoglobins. These results may have physiological relevance, since endogenously produced NO is now recognized to play an important role in the relaxation of vascular smooth muscles. If present in vivo, cell-free hemoglobins exposed to NO become rapidly oxidized. Our results show that subsequent interactions of NO with ferrihemoglobin can result in redox cycling. This has the potential of depleting NO and further altering vascular tone with rates dependent on structural parameters of the ferrihemoglobin that are not determined by oxygen affinity.

Consequences of chemical modifications on the free radical reactions of human hemoglobin.

Hemoglobin-based oxygen carriers (HBOCs) are candidates for use as blood substitutes and resuscitation fluids. We determined that HBOCs of specific types differ in their ability to generate or interact with free radicals. The differences do not correlate with oxygen affinity. Detailed comparisons with unmodified human hemoglobin, HbA0, were carried out with two cross-linked derivatives: HbA-FMDA, produced by the reaction of human oxyhemoglobin with fumaryl monodibromoaspirin, and HbA-DBBF, produced by the reaction of human deoxyhemoglobin with bis(3,5-dibromosalicyl) fumarate. Both derivatives had lower oxygen affinity than unmodified HbA0. As previously reported, exposure of oxyhemoglobin to H2O2 causes generation of free radicals capable of generating formaldehyde from dimethyl sulfoxide. Relative to the reaction catalyzed by 50 microM HbA (18.0 +/- 3.5 nmol/30 min/ml), the formaldehyde formation was roughly 70% for HbA-DBBF and 50% for HbA-FMDA under comparable conditions. More profound differences are exhibited at lower hemoglobin concentrations. Spectral changes of the HBOCs during the reaction differ qualitatively and occur at different rates. The HBOCs also differ in rates of hemoglobin-catalyzed NADPH oxidation and aniline hydroxylation, reactions mediated by reactive oxygen species. These results show that stereochemical differences brought about by chemical cross-linking alter the ability of HBOCs to generate radicals and to react with activated oxygen species. These studies also show that the ability of hemoglobin to produce activated species of oxygen can be enhanced or suppressed independently of oxygen affinity.

Hemoglobin-based oxygen carriers (HBOCs): structural alterations that affect free radical generation.

We examined how changes in oxygen affinity brought about by different chemical modifications of hemoglobins affect their oxidation-reduction reactions. The three modified hemoglobins studied were HbA-FMDA, HbBv-FMDA, produced by the reaction of human or bovine oxyHb with fumaryl mono-dibromoaspirin; and HbA-DBBF, produced by the reaction of human deoxyHb with bis(3,5-dibromosalicyl) fumarate. Exposure of oxyHb to H2O2 causes generation of free radicals capable of cleaving dimethylsulfoxide (Me2SO) to produce formaldehyde (HCHO). Relative to the reaction rate for HbAo (630 +/- 130 M/min) the rates of HCHO formation were roughly 70% for HbA-DBBF, 50% for HbA-FMDA and 16% for HbBv-FMDA. Exposure to H2O2 also caused spectral changes at varied rates for the HBOCs analyzed. Although these rates were not directly correlated with the rates of free radical formation, addition of mannitol or thiourea slowed both the rate of spectral changes and HCHO formation. The relative ability of the ferric derivatives of the HBOCs to participate in free radical reactions was monitored by assays of non-enzymatic NADPH oxidation and aniline hydroxylation. HbBv-FMDA showed significantly slower rates than the other HBOCs in both assays. The observed differences between HBOCs in these assays indicate differences in their ability to generate or interact with free radicals.

Effects of two viral inactivation methods on platelets: laser-UV radiation and merocyanine 540-mediated photoinactivation.

Two viral inactivation methods suggested for use with cellular blood products have been evaluated as to their effects on platelets. In the first study, it was proposed that pulsed laser-ultraviolet radiation (UVB) at 308 nm could favor photodamage to UVB-sensitive viral nucleic acid with minimal effects on blood platelets. A "window of efficacy" was observed with UVB doses of 10.5-21.5 J/cm2 at which 4-6 log10 poliovirus were inactivated while platelets were relatively tolerant. However, this "window" occurred only with low-intensity UVB radiation (less than or equal to 0.25 MW/cm2). Damage to platelet proteins, evident at high laser intensities, was probably due to multiple photon excitation of amino acids. In the second study, platelets and viruses were treated with the photosensitizer, merocyanine 540 (MC 540) (less than or equal to 24 micrograms/ml), and visible light (450-600 nm) (less than or equal to J/cm2). Activation of washed platelets by dye/light treatment resulted in a spontaneous release of serotonin, spontaneous aggregation, and marked morphological changes. Increasing concentrations of albumin in the suspension medium protected against dye-mediated photodamage to platelets, but also significantly reduced the antiviral activity of MC 540 and light. These results illustrate the relative sensitivities of platelets and viruses to two inactivation methods and the difficulty in optimizing inactivation of viruses and preservation of platelet function in a protein-rich medium.

Inhibition by albumin of merocyanine 540-mediated photosensitization of platelets and viruses.

The effect of the photosensitizer merocyanine 540 (MC 540) on platelets and on three marker viruses was examined to assess its potential in reducing virus transmission by blood products. The results demonstrated several deleterious effects of MC 540 (4-24 micrograms/mL) on platelet morphology and function in both the absence and presence of light (450-600 nm). Treatment of washed platelets with MC 540 in the dark resulted in a significant release of serotonin in the absence of added agonist, as well as a diminished response to thrombin as measured in vitro. In addition, photosensitization caused spontaneous platelet aggregation and release of 92 percent of the releasable serotonin without the addition of an agonist. Because photo-treatment of blood products is likely to be performed in a protein-rich medium, the influence of albumin on the phototoxic effects on platelets was assessed. Albumin added to the suspension medium at concentrations greater than or equal to 1.0 percent protected the platelets against the effects of MC 540 in the dark, whereas 5-percent albumin was required for protection against the phototoxic effects of MC 540 on the platelet response to thrombin. The antiviral activity of MC 540 and light was examined by using the lipid-containing viruses herpes simplex virus (HSV) and bacteriophages phi 6 and PM2. Of the lipid-enveloped viruses, HSV was 25 times more photosensitive to MC 540 than was phi 6 (15 micrograms/mL). PM2, which has an internal lipid layer, was almost 300 times less sensitive to MC 540 and light than was HSV.(ABSTRACT TRUNCATED AT 250 WORDS)