Acute dyslipoproteinemia induced by interleukin-2: lecithin:cholesteryl acyltransferase, lipoprotein lipase, and hepatic lipase deficiencies.

Recombinant human interleukin-2 (rIL-2) is used to treat refractory cancers. During such treatment, patients develop severe hypocholesterolemia along with striking alterations in the concentration and composition of the circulating lipoproteins. The present study was undertaken to gather information about the pathogenesis of these abnormalities. Patients were studied before-, during- and after a 5-day course of high dose i.v. rIL-2. Whole plasma cholesterol was markedly reduced by rIL-2 administration (52%; P < 0.001), whereas the triglyceride concentration did not change. Thus, the lipoproteins became triglyceride enriched (P = 0.004). Low density lipoprotein cholesterol, apolipoprotein B (apoB), high density lipoprotein cholesterol, and apoA-I concentrations all decreased. Esterified cholesterol levels were markedly reduced. Total plasma apoE increased markedly, and two kinds of abnormal particles appeared: 1) beta-migrating, very low density lipoproteins; and 2) discoidal, apoE- and phospholipid-containing particles with abnormal density and electrophoretic mobility. The activities of two lipoprotein triglyceride hydrolases, lipoprotein lipase and hepatic lipase, fell significantly during treatment and returned promptly to pretreatment levels after rIL-2 was discontinued. Lecithin:cholesteryl acyltransferase (LCAT) activity also decreased significantly (64%) during treatment, but in contrast to the lipases, remained low for at least 5 days after the last dose of rIL-2 (P < 0.001). High dose i.v. rIL-2 induces severe dyslipidemia with deficiencies of both postheparin lipases and acute LCAT deficiency. Most, if not all, of the lipoprotein changes observed are explained by the LCAT deficiency that follows IL-2-induced hepatocellular injury and cholestasis.

Mutations in exon 3 of the lipoprotein lipase gene segregating in a family with hypertriglyceridemia, pancreatitis, and non-insulin-dependent diabetes.

A proband with chylomicronemia, pancreatitis, and non-insulin-dependent diabetes (NIDDM) bears two different mutations in exon 3 of the lipoprotein lipase (LPL) gene: a missense mutation, 75Arg-->Ser, inherited through the paternal line and a truncation, 73Tyr-->Ter, through the maternal line. NIDDM appeared to be independently segregating. The R75S mutant was studied in extracts and media from transfected COS-1 cells. Detectable amounts of catalytically competent R75S LPL suggested destabilization of the active homodimer as with exon 5 mutants (Hata et al. 1992. J. Biol. Chem. 267:20132-20139). Hydrolysis of a short-chain fatty acid ester indicated that R75S does not directly affect activation of LPL by apoC-II. Subjects with NIDDM and wild-type LPL, and nondiabetic middle-aged carriers of the 73Tyr-->Ter truncation had moderate hypertriglyceridemia (260-521 mg/dl) and reduced high density lipoprotein cholesterol. A maternal aunt with NIDDM carried the truncation. Her phenotype (triglycerides of 5,300 mg/dl, eruptive xanthomatosis, and recurrent pancreatitis) was as severe as that in homozygotes or compound heterozygotes. We conclude: (a) diabetic carriers of dysfunctional LPL alleles are at risk for severe lipemia; and (b) the physiologic defects in NIDDM may be additive or synergistic with heterozygous LPL deficiency.

Binding of lipoprotein lipase to heparin. Identification of five critical residues in two distinct segments of the amino-terminal domain.

Binding to heparan sulfate governs many aspects of the physiological action and regulation of the lipolytic enzyme, lipoprotein lipase (LPL). In an attempt to identify the structural determinants which mediate this interaction, basic residues in three segments of the primary sequence of human LPL (residues 147-151, 279-282, and 292-304) were replaced with alanine, either singly or in various combinations, and variant proteins were subjected to affinity chromatography on heparin-Superose. Five basic residues in two distinct segments of the primary sequence were critical determinants of the high affinity for heparin manifested by the active enzyme (R279, K280, R282, K296, R297). By contrast, no such evidence could be detected for basic residues in the first cluster (K147, K148) or for other basic residues in the third cluster (K292, R294, K304), while the evidence for K300 was unresolved. The conformation of this heparin-binding domain can be inferred by reference to the three-dimensional structure of the homologous enzyme, pancreatic lipase (Winkler, F. K., D'Arcy, A., and Hunziker, W. (1990) Nature 343, 771-774). Affinity of the active enzyme for heparin could not be reduced below a threshold, suggesting that other heparin-binding determinants exist elsewhere in the molecule, as supported by recently published evidence (Davis, R. C., Wong, H., Nikazy, J., Wang, K., Han, Q., and Schotz, M. C. (1992) J. Biol. Chem. 267, 21499-21504).

Missense mutations in exon 5 of the human lipoprotein lipase gene. Inactivation correlates with loss of dimerization.

Most missense mutations of the lipoprotein lipase (LPL) gene identified among LPL-deficient subjects cluster in a segment of the sequence that encodes the catalytic triad as well as functional elements involved in the activation of the lipase at lipid-water interfaces. Consequently, loss of activity may result either from direct alterations of such functional elements or from less specific effects on protein folding and stability. This issue was addressed by examining biochemical properties of four such variants (A176T, G188E, G195E, and S244T) in a heterologous expression system (COS-1 cells). Variant G195E (GGA----GAA) was previously unreported. In all instances, inactive enzyme was recovered in medium, albeit at reduced levels. Cellular synthesis and extracellular degradation were similar to those for wild type, suggesting that reduced secretion resulted from increased intracellular degradation. When cell extracts were subjected to heparin-Superose affinity chromatography followed by elution on a linear salt gradient, all variants exhibited a single, inactive, low affinity immunoreactive peak. By contrast, wild-type enzyme presented an additional, high affinity, active species, which we interpret as homodimeric enzyme. Substitution of the active-site serine (S132A) led to loss of activity but maintenance of the high affinity species. When large amounts of the G188E variant were applied to the column, small but significant amounts of high affinity, active enzyme were recovered. Systematic substitutions at residue 188 showed that only glycine could accommodate structural constraints at this position. We conclude that the mutations examined did not impart lipase deficiency by affecting specific functional elements of the enzyme. Rather, they appear to affect protein folding and stability, and thereby formation and maintenance of subunit assembly.

An improved flow cytometric assay for somatic mutations at the glycophorin A locus in humans.

An improved method has been developed for the glycophorin A assay for somatic cell mutations in humans. The new assay, named the "BR6" assay, can be performed on a commercially available, single-beam flow cytometer, in contrast to the previously described 1W1 assay that required a dual-beam flow sorter. A modified cell labeling method developed for the BR6 assay results in improved separation of normal and mutant phenotype cells, as well as eliminating some cellular artifacts that affected the 1W1 assay. Parallel measurements on samples from 17 normal donors showed that the BR6 assay yields comparable variant cell frequencies and improved measurement precision compared with the 1W1 assay. A detailed analysis of three individuals who showed large differences in background variant frequency with the 1W1 assay confirmed that these differences could also be detected with the BR6 assay. A dramatically elevated variant cell frequency was seen with the BR6 assay of an individual exposed to a high level of ionizing radiation in an accident at Goiania, Brazil.

The physiological state as a modifier of radiation-induced cytotoxicity in heterogeneous murine tumor cells growing in vitro.

The oxic radiation response (cytotoxicity) of two heterogeneous murine tumor-cell lines cultured in vitro was studied as a function of the cell's physiological state at the time of X-irradiation. The proliferating (P) 66 and 67 cells displayed equal radiosensitivities; however, the quiescent (Q) cells were considerably more radiosensitive than the P cells, and the 66Q cells were even more radiosensitive than the 67Q cells. Also, the 66Q cells continued to proliferate slowly with about 85 per cent in the G1 phase and 10 per cent in the S phase, while the 67 Q cells displayed a more complete G1 arrest (92-95 per cent). A detailed analysis of the metabolic status vs cell-cycle age (i.e. G1 vs S phase) indicated that the cell-cycle age was the predominant factor influencing radiation-induced cytotoxicity in 67 cells. The data also showed that in the plateau phase Q-cell cultures, pH and cell contact were not influencing factors and that the increased radiosensitivity of the Q cells could not be explained on the basis of energy deprivation. Moreover, the 66Q, but not the 67Q cells displayed an increased sensitivity in addition to that caused by the predominant cell-cycle age shift. This extra increase in radiosensitivity is of unknown metabolic origin, but could be related to cellular membrane fragility in the stressed 66Q cells since this extra component of Q-cell radiosensitivity was reduced both by refeeding (metabolic activation) 4 h before X-irradiation and by delayed plating while incubating the cells in Q medium at 37 degrees C after X-irradiation.

Endogenous thiol levels in heterogeneous murine tumor cells as a function of the physiological state and the response to X-irradiation.

The endogenous thiols (PSH, protein sulfhydryls; NPSH, nonprotein sulfhydryls; and GSH, glutathione) were measured in the 66 and 67 murine carcinoma cells growing under different physiological conditions in vitro (e.g., proliferation, P; nutrient-deprived quiescence QI; and QI cells stimulated by refeeding the monolayer in situ and assayed 4 (St4) and 14 (St14) h later). The aerobic radiation response was also studied as a function of the physiological state and thiol concentration. The changes in PSH levels suggest that the proportion of thiol-containing proteins changed whenever the cells were in transition between different physiological states (e.g., when QI cells were stimulated by refeeding, the proportion of PSH was elevated dramatically over either QI or P cells). The NPSH and GSH levels were both down significantly in the QI vs. P cells as was the total thiol level (PSH plus NPSH). Fourteen h but not 4 h after stimulation, the NPSH and GSH levels had returned to or exceeded the P-cell levels. Also, the proportion of GSH in the NPSH fraction varied as a function of the physiological state. The 66 and 67 QI cells were both more radiosensitive than the respective P cells. Also, the 66 cell radiation-induced cytotoxicity had returned to the P response by about 4 h after refeeding but the stimulated 67 cells had not. However, no overall correlation was apparent between the various aerobic radiation responses and the pool sizes of either the total thiols or of the various subsets of thiols. The depressed total thiol level and the increased radiosensitivity of the QI cells could represent a cause-and-effect relationship or these parameters could be independent phenomena only related indirectly through the reduced metabolic activity of the quiescent cells.

A calmodulin antagonist has no effect on the repair of X-ray-induced damage in a murine mammary carcinoma cell line.

The effects of the calmodulin antagonist W13 were determined on potentially lethal damage repair, sublethal damage repair, and X-ray-induced DNA damage repair following X irradiation of 67 murine mammary carcinoma cells in the proliferative and quiescent states. Studies with W13 (20 micrograms/ml) on proliferating cells showed that the cells rounded up within 2 h but stayed attached to the dishes and there was a slight transient G2 block by 6 h. Also, the proportion of S-phase cells at 12 h was reduced to 65% of control with the concurrent [3H]thymidine incorporation reduced to 62% of control. There was no detectable effect from this pharmacological dose of W13 either on PLDR in proliferating cells at 400 and 800 rad or on quiescent cells at 200 and 400 rad. Likewise, there was no measurable effect on SLDR in either proliferating or quiescent cells at equally split doses of 800 and 600 rad, respectively. In addition, for control vs W13-treated proliferating cells, no difference was detected either in the induction of DNA damage by X irradiation or in the initial rate of repair (T 1/2 approximately equal to 7 min), as measured by the alkaline filter elution assay. In contrast to uv and bleomycin-induced damage, these data suggest that calmodulin may have no major role in either the molecular or cellular recovery from X-ray-induced damage in mammalian cells.

Toxic effects of extended glutathione depletion by buthionine sulfoximine on murine mammary carcinoma cells.

Extended depletion of glutathione to approximately equal to 5% of control in the murine mammary carcinoma cell line 66 was achieved with a concentration of 0.05 mM buthionine sulfoximine. At 24 hours, there was no evidence for cellular toxicity from the BSO treatment per se; however, by 48 hours, there was inhibition of protein and DNA synthesis and cell growth and cell kinetic data was suggestive of both a G1 and a G2 block. Glutathione depletion to this extent (i.e., 0.13 mM vs. 2.24 mM in control) did not modify the aerobic radiation response for cells in the physiological states of proliferation, quiescence, or stimulated quiescent cells. This degree of cellular toxicity may well be cell-type dependent, but the results do suggest that caution is in order if one should attempt long-term GSH depletion in vivo.

Heterogeneity of X-ray cytotoxicity in proliferating and quiescent murine mammary carcinoma cells.

A highly enriched (greater than or equal to 97%) quiescent (Q) tumor cell population can be induced in both the 66 and 67 murine mammary carcinoma lines in vitro by nutrient deprivation (7-day, unfed plateau cultures), while exponential cultures (2-day cultures) of this line are composed of greater than 98% proliferating (P) cells. We have used these two cell lines to determine how the radiation sensitivity varies as a function of genetic heterogeneity (two cell lines derived from the same tumor) and proliferative status (physiological state). The 67 Q cells were significantly more sensitive than were the P cells to single doses of X-rays, with Dos of 52 and 90 rads and Dqs of 188 and 250 rads, respectively. Cells from transition cultures (cells that have essentially stopped proliferation but are not in the biochemical state of Q cells) have a radiation sensitivity similar to that of P cells. When exponentially growing 67 cells were induced into a Q state by reducing the serum concentration (0.5 versus 15%), they, too, were more sensitive to X-rays than were their proliferating counterparts. This sensitivity of the Q cells was decreased by placing them 30 min prior to irradiation in either fresh medium, a balanced salt solution, or a balanced salt solution with 24 mM glucose. However, the Q cells in these conditions were still an order of magnitude more sensitive than the P cells after a 523-rad dose. Therefore, the increased sensitivity of the well-oxygenated 67 Q cells appears to be primarily related to physiological alterations accompanying the transition from P to Q. The radiation sensitivity of 66 cells has also been measured in P and Q states. These cells are significantly more radioresistant than are the 67 cells and, again, the 66 Q cells were more sensitive than were the 66 P cells, with Dos of 90 and 109 rads and Dqs of 150 and 368 rads, respectively. Furthermore, the heterogeneous radiation response of the 66 and 67 cells continues to be expressed under various physiological states, albeit in qualitatively different ways; i.e., in 66 Q versus P cells, the shift in sensitivity is primarily due to a markedly reduced Dq while, in the 67 Q versus P cells, the lowered radiosensitivity is due to a marked reduction in both Do and Dq. At least in these cell lines, it is unlikely that Q cells will determine the response of the tumor to radiation.

DNA damage repair in quiescent murine mammary carcinoma cells in culture.

Murine mammary carcinoma cells (line 67) were grown in unfed cultures for up to 9 days. In cultures (day 2-3) in which cells were proliferatively active and in day 3-5 (transition) cells, a large fraction of nuclear DNA was retained on polycarbonate filters when assayed by the alkaline filter elution technique. In contrast, the fraction of DNA retained on filters was significantly reduced for nonproliferating (Q, quiescent) cells from unfed 7-9 day cultures. The increase in endogenous DNA breaks followed both the decrease in proliferative state and clonogenicity in these cells. When day 7 Q cells were refed these endogenous DNA breaks were removed with a half-time of about 2.5 h. When the cells were exposed to X-irradiation and the integrity of their nuclear DNA measured by the alkaline filter elution assay, as much as a 2-fold greater frequency of radiation-induced DNA breaks was produced in Q versus P cells. DNA breaks were also removed from irradiated Q cells at a rate which was 0.23 that observed in P cells. We suggest that the depressed capacity for DNA damage removal in Q cells is responsible for their greater radiosensitivity, and the impaired DNA damage repair is probably due to a reduced level of energy sources in these unfed Q cell cultures.

Protamine precipitation of two reovirus particle types from polluted waters.

Two forms of virus particle are released from reovirus-infected cell cultures, infectious reovirus and potentially infectious reovirus (PIV). PIV particle forms have a complete outer coat and are not infectious until the outer coat is altered or removed. The PIV concentration in polluted waters, however, has not been determined. Protamine sulfate precipitation, using 0.25% fetal bovine serum and 0.005% protamine sulfate for the first precipitation of the sample and 0.0025% for the second, was employed to concentrate infectious reovirus and PIV from water and sewage. Infectious reovirus and PIV particles were concentrated over 500-fold from river water inoculated with virus, and virus recoveries of between 80 and 100% were achieved. Virus precipitates stored at -20 degrees C as a protamine-virus concentrate showed a 5% loss of PIV after 14 days. Virus preparations were assayed, before and after treatment, with 200 micrograms of chymotrypsin per ml, using a fluorescent-antibody procedure. Protamine sulfate precipitation and fluorescent-antibody detection are effective ways to recover and assay reoviruses present in raw sewage.

Evaluation of cell lines and immunofluorescence and plaque assay procedures for quantifying reoviruses in sewage.

Twelve continuous cell lines were tested to determine their sensitivity to reovirus types 1, 2, and 3 isolated from sewage. Madin-Darby bovine kidney (MDBK), rhesus monkey kidney (LLC-MK2), and human embryonic intestinal (intestinal 407) cells were most sensitive, respectively. In a similar study, MDBK cells were more sensitive than LLC-MK2 and Buffalo green monkey kidney (BGM) cells to sewage-isolated, protamine-precipitated reoviruses which had not been serotyped and had no previous cell contact. Sewage-isolated, protamine-precipitated reoviruses were also used in conjunction with MDBK cells in a comparative evaluation of immunofluorescent cell count and plaque assay procedures. The immunofluorescence assay is more sensitive and more rapid than the plaque assay. Reoviruses in excess of 10(4)/liter of raw sewage were detected by the immunofluorescent cell count assay.