Elimination of cytokine production in stored platelet concentrate aliquots by photochemical treatment with psoralen plus ultraviolet A light.

BACKGROUND: Cytokines generated in platelet concentrates (PCs) during storage have been implicated as possible mediators of febrile nonhemolytic transfusion reactions. Two potential methods of white cell inactivation were compared for their ability to reduce cytokine synthesis in pooled random-donor PC aliquots: treatment with gamma-radiation and photochemical treatment (PCT) using psoralens and ultraviolet A light. STUDY DESIGN AND METHODS: ABO-matched PC aliquots were pooled and divided into separate aliquots. Aliquots (20 mL) were taken from each pool to serve as an untreated control and to undergo gamma-radiation. Aliquots were treated by using either gamma-radiation (2500 or 5000 cGy) or virucidal PCT. PCT with the psoralens 8-methoxypsoralen (8-MOP), aminomethyltrimethyl psoralen (AMT), and S-59 was investigated. PC aliquots were stored for 7 days and analyzed for levels of interleukin 8 by use of an enzyme-linked immunosorbent assay. Levels of DNA adduct formation were determined by using 3H-labeled psoralens. RESULTS: Levels of interleukin 8 in the untreated random-donor PC aliquots increased with increasing white cell counts, but they were not affected by pooling. The untreated control aliquots and the aliquots treated with gamma-radiation had significant increases in levels of interleukin 8 after 5 to 7 days of storage (p<0.05). PCT with S-59 resulted in a significant reduction in cytokine synthesis (p<0.05). Day 5 to 7 levels of interleukin 8 did not differ significantly from Day 0 levels. Inhibition of interleukin 8 production by PCT increased with increasing levels of DNA modification (S-59 > AMT > 8-MOP). CONCLUSION: PCT that utilizes S-59 has been developed to inactivate potential viral and bacterial pathogens in PC aliquots while maintaining in vitro platelet function. These data demonstrate that PCT of aliquots of pooled PC aliquots before storage also prevents white cell cytokine synthesis during storage. PCT may therefore offer the potential for reducing cytokine-associated febrile nonhemolytic transfusion reactions.

Inactivation of leukocytes in platelet concentrates by photochemical treatment with psoralen plus UVA.

A photochemical treatment (PCT) process using a novel psoralen and long wavelength ultraviolet light (UVA, 320-400 nm) has been developed to inactivate bacteria and viruses in platelet concentrates. This study evaluated the efficacy of PCT for inactivation of leukocytes that contaminate platelet preparations. Three psoralens, 8-methoxypsoralen (8-MOP), 4'-aminomethyl 4,5', 8-trimethylpsoralen (AMT), and the novel psoralen S-59, were compared using the following four independent but complementary biological and molecular assays. (1) T-cell viability: Treatment with 150 mumol/L S-59 and 1.0 to 3.0 Joules/cm2 UVA inactivated >5.4 +/- 0.3 log10 of T cells in full-sized single-donor plateletpheresis units. Using 1.0 Joule/cm2 UVA, the lowest dose of S-59, AMT and 8-MOP required to reduce the number of T cells to the limit of detection was 0.05 micromol/L, 1.0 micromol/L, and 10.0 micromol/L, respectively. (2) Cytokine synthesis: Treatment with 1.9 Joules/cm2 UVA and 150 micromol/L S-59 or AMT completely inhibited synthesis of the cytokine IL-8 by contaminating leukocytes during 5 days of platelet storage. After treatment with 75 micromol/L 8-MOP and 1.9 Joules/cm2 UVA, only low levels of IL-8 were detected. (3) Psoralen-DNA adduct formation: The combination of 1.9 Joules/cm2 UVA and 150 micromol/L S-59, AMT, or 8-MOP induced 12.0 +/- 3.0, 6.0 +/- 0. 9, and 0.7 psoralen adducts per 1,000 bp DNA, respectively. (4) Replication competence: Polymerase chain reaction (PCR) amplification of small genomic DNA sequences (242-439 bp) after PCT was inhibited. The degree of PCR amplification inhibition correlated with the level of adduct formation (S-59 > AMT > 8-MOP). In contrast, 2,500 cGy gamma radiation, a dose that inactivates >5 log10 of T cells in blood products, had minimal effect on cytokine synthesis and did not induce sufficient DNA strand breaks to inhibit PCR amplification of the same small DNA sequences. These results demonstrate that leukocytes are sensitive to PCT with psoralens and among the psoralens tested S-59 is the most effective. Therefore, PCT has the potential to reduce the incidence of leukocyte-mediated adverse immune reactions associated with platelet transfusion.

Removal of cytokines from HAS-containing solutions by adsorption onto silica.

Septic shock syndrome is a potentially fatal medical condition that is associated with elevated blood levels of low molecular weight proteins known as cytokines. Adsorption was investigated as a potential method for removing cytokines from blood. Saline with 50 mg/mL human serum albumin (HAS) spiked with pathological concentrations (ng-pg/mL) of radiolabeled cytokine was used to study cytokine adsorption. Adsorption isotherms were linear in the pathological concentration range, with adsorption constants ranging from 33.0 mL/g to 173 mL/g for tumor necrosis factor (TNF-alpha), interleukin-8 (IL-8),interleukin-6 (IL-6), and C3a. Adsorption constants were also determined for interleukin-1alpha (IL-1alpha), IL-1beta, and interferon-gamma (IFN-gamma). The adsorption of cytokines by several different silica adsorbents was investigated. Increased concentrations of NaCl reduced cytokine adsorption, but did not completely eliminate adsorption even at high concentrations, suggesting that adsorption wads not entirely electrostatic in nature. Possible mechanisms of cytokine adsorption are discussed. Data for batch adsorption for TNF-alpha was used to estimate the minimum amount of silica required to treat septic shock. It was concluded that a silica adsorbent has a sufficiently high capacity to be used for hemoperfusion. Adsorption of myoglobin and cytochrome c was also investigated as possible marker proteins for future dynamic adsorption studies in hemoperfusion devices.

Pressure stabilization of proteins from extreme thermophiles.

We describe the stabilization by pressure of enzymes, including a hydrogenase from Methanococcus jannaschii, an extremely thermophilic deep-sea methanogen. This is the first published report of proteins from thermophiles being stabilized by pressure. Inactivation studies of partially purified hydrogenases from an extreme thermophile (Methanococcus igneus), a moderate thermophile (Methanococcus thermolithotrophicus), and a mesophile (Methanococcus maripaludis), all from shallow marine sites, show that pressure stabilization is not unique to enzymes isolated from high-pressure environments. These studies suggest that pressure stabilization of an enzyme may be related to its thermophilicity. Further experiments comparing the effects of increased pressure on the stability of alpha-glucosidases from the hyperthermophile Pyrococcus furiosus and Saccharomyces cerevisiae support this possibility. We have also examined pressure effects on several highly homologous glyceraldehyde-3-phosphate dehydrogenases from mesophilic and thermophilic sources and a rubredoxin from P. furiosus. The results suggest that hydrophobic interactions, which have been implicated in the stabilization of many thermophilic proteins, contribute to the pressure stabilization of enzymes from thermophiles.

Estimation of melting curves from enzymatic activity-temperature profiles.

Measuring the reversible thermal unfolding of enzymes is valuable for quantifying the effects of environmental factors on the thermodynamic stability of proteins. The thermal unfolding behavior of enzymes is typically studied using calorimetry or optical techniques such as circular dichroism, fluorescence, or light scattering. These techniques often have practical limitations and usually require the protein to be electrophoretically pure. An alternative technique for analyzing the thermodynamic stability of enzymes is to estimate the melting curve from temperature-activity data. This technique does not require electrophoretically pure enzyme, provided the sample does not have competing enzymatic activities or proteins which can affect enzyme stability (e.g., proteases). Moreover, small amounts of contaminant proteins should not affect the results as long as enzymatic assays are performed at low protein concentrations where nonspecific protein-protein interactions are negligible. To illustrate this technique, the melting curve for beta-galactosidase from Escherichia coli in the presence of 1 mM EDTA, and the shift caused by adding 1 mM Mg(+2), were calculated from activity-temperature data. Melting temperatures predicted from activity-temperature data compared closely with those obtained using other techniques. Application of this analysis to multisubstrate enzymes is illustrated by estimating the melting profiles for partially purified hydrogenases from several thermophilic Methanococcii. Limitations and important considerations for estimating melting profiles from activity-temperature data are discussed.