Duplicate plating of immune cell products: analysis of globulin class secretion by single cells.

Antibodies secreted by individual immune cells were collected focally in very thin "original" and "imprint" layers of agar containing the target antigen, sheep erythrocytes. Identical treatment of both layers led to mirror image patterns of hemolytic plaques. Development of one layer for immunoglobulin M hemolysins and the other for immunoglobulin G hemolysins produced unrelated plaque patterns indicating that few, if any, cells simultaneously release substantial amounts of both gammaM and gammaG antibodies.

Effect of whole-body hyperthermia and cyclophosphamide on natural killer cell activity in murine erythroleukemia.

Mice infected with the polycythemia-inducing strain of Friend virus complex (FVC-P) develop a fatal erythroid disease similar in some respects to leukemia. Six- to eight-week-old DBA/2 female mice were injected i.v. with 0.5 ml of a virus suspension containing approximately 5 X 10(4) plaque-forming units and 5 X 10(3) spleen focus-forming units. Four treatment regimens were begun 3 days postinjection: (a) no treatment; (b) whole-body hyperthermia (WBH) alone; (c) cyclophosphamide (CY) alone; (d) WBH combined with CY. WBH treatment utilized a microwave generator operating at 2450 MHz. The i.p. temperature of the mice receiving WBH was maintained at 39.5-40 degrees C for 30 min. The CY was given i.p. at a dosage of 20 mg/kg of body weight. The various treatments, CY, WBH, CY + WBH were given once a week for 2 weeks. Natural killer cell activity was examined in all four groups of mice and was found to be significantly higher in the animals treated with WBH or CY. Our results show that WBH, either alone or in combination with CY, can prolong the lifespan of mice infected with lethal dosages of the FVC-P, possibly via a mechanism involving natural killer cells.

Curative effect of split low dosage total-body irradiation on mice infected with the polycythemia-inducing strain of the Friend virus complex.

Split low dose total-body irradiation (TBI) with 150 cGy was assessed for its efficacy in modifying the disease induced in DBA/2 mice by the polycythemia-inducing strain of the Friend virus complex (FVC-P, composed of a Friend murine leukemia helper virus and a spleen focus-forming virus). All FVC-P injected mice were dead within 40 days; however, infected mice receiving TBI on days 5 and 12 exhibited long-term survival. FVC-P-injected mice receiving TBI treatment on days 5 and 12 had normal leukocyte counts, normal spleen weights, and no detectable spleen focus-forming virus. Although the FVC-P-infected mice had decreased proportions of L3T4+ cells and increased proportions of Lyt-2+ cells, these were returned to normal following TBI treatment. Apparently the time sequence of TBI treatments is important since one treatment with TBI on day 5, or two treatments with TBI on days 12 and 18, was not as efficacious. The inability of in vitro irradiation doses of up to 1000 cGy to inactivate FVC-P which was subsequently injected into murine hosts suggests that the effectiveness of the TBI treatment in vivo is not due to a direct radiation effect on the virus. These results indicate a possible relationship between L3T4+ and Lyt-2+ numbers or their ratio in the curative efficacy of TBI in FVC-P-infected mice.

Sulfhydryl reactive phenylarsine oxide inhibits signal transduction in NK and LAK cells: effect on zeta-chain phosphorylation and phosphatidylinositol level.

The specific role of sulfhydryl groups in cell-mediated cytotoxicity (CMC) is still unknown. Here we demonstrate that natural killer cells and lymphokine-activated killer cells, when incubated with phenylarsine oxide (PAO), an organoarsenic compound showed a dose- and time-dependent inhibition of CMC and antibody-dependent cellular cytotoxicity (ADCC). PAO interacted directly with the effector cells (EC) without affecting the target cells (TC) or EC:TC conjugate formation. The loss of cytotoxicity was not due to lack of degranulation or to inhibition of serine esterases in PAO-treated cells. However, PAO inhibited the target-induced down regulation of phosphatidylinositol (PI) level in NK cells indicating that PAO blocked the cytolytic cascade at an early stage, upstream of PI. In addition, PAO also did not affect the disulfide link of the zeta-chain dimers, implicated in signal transduction in cytotoxic lymphocytes but did cause the rapid phosphorylation of the zeta chain. Finally, the effect of PAO on CMC was competitively blocked by dithiothreitol, a dithiol, but not by beta-mercaptoethanol, a mono-thiol. Taken together, these results indicate for the first time how sulfyhydryl groups may regulate CMC and ADCC.

Electron spin resonance spectroscopy does not reveal hydroxyl radical production in activated natural killer lymphocytes.

We investigated hydroxyl radical (OH) production by human natural killer (NK) cells, using electron spin resonance (ESR) spectroscopy and 5.5 dimethyl-1-pyrroline-N-oxide (DMPO), a spin trap specific for OH. production. We confirmed that hydroxyl radical scavengers, n-propyl gallate and catechin, inhibited NK cell-mediated cytotoxicity (NK-CMC) in a dose-dependent manner and demonstrated that DMPO also inhibited NK-CMC. Polymorphonuclear leukocytes (PMNL) activated by opsonized zymosan (2.4 mg/ml) and mixed with DMPO (0.14 M) showed an early increase in hydroxyl radical production, leading to a net production of free radical of almost 400 pMol/10(6) cells. We then mixed NK cells with K562, an NK-sensitive tumor cell, at a 1:1 ratio and added DMPO (0.14 M). We pelleted the cells to increase EC to TC binding before taking the sample readings. Activated NK cells showed no increase in OH. production, leading to a net production of free radicals less than 1% that of activated PMNL. These data strongly suggest that hydroxyl radical production does not play a role in the early events of NK cell activation; they indicate a need to reevaluate the mechanism of inhibition of NK-CMC by OH. scavengers.

Target cell-directed degradation of perforin mRNA in CTL: lack of correlation with loss of protein and lytic ability.

We have previously shown that CTL and NK cells rapidly down regulate perforin mRNA and become functionally inactive within 4-6 hr after exposure to sensitive target cells (TC). We report here for the first time that CTL also down regulate perforin mRNA upon exposure to resistant, but binding, TC. When three separate human MHC-restricted CTL lines were exposed to resistant TC, perforin mRNA was rapidly degraded. Removal of both extracellular Ca++ and Mg++ prevented perforin message down regulation, whereas removal of Ca++ alone did not, indicating that CTL:TC binding was required. Unlike the response of CTL exposed to sensitive TC, resistant TC did not trigger serine esterase (SE) release, suggesting distinct signalling pathways for perforin mRNA down regulation and granule exocytosis. Moreover, using western analysis, we showed that there was limited (< 10%) perforin protein release after CTL:TC interaction, suggesting that CTL loss of lytic activity after exposure to sensitive TC is not due to massive depletion of perforin. Treatment of CTL with mAb to CD2, CD3, CD2 + CD3, CD8, Class I and LFA-1 did not induce perforin mRNA down regulation. Furthermore, mAb to CD2, CD3, CD8, Class I, Class II, CD54 and LFA-1 did not block TC-mediated perforin mRNA down regulation although lysis of TC was inhibited.

Fas-mediated cytotoxicity induces degradation of vesicular stomatitis virus RNA transcripts and reduces viral titer.

Several investigators have recently examined the effect of Fas (CD95)-mediated apoptotic cell death on target cells (TC). The effect of Fas-mediated death on viral RNA within the TC, however, has not been explored. In this study, we investigated the ability of the Fas pathway to mediate pre-lytic degradation of vesicular stomatitis virus (VSV) RNA and TC RNA. We show that engagement of Fas antigen on VSV-infected Jurkat cells induces pre-lytic degradation of VSV RNA transcripts, whereas full-length VSV genome RNA, known to be tightly associated with viral proteins, is not degraded. Cellular RNA, including beta-actin and glyceraldehyde-3-phosphate-dehydrogenase mRNAs, is also degraded by Fas-mediated cytotoxicity. In addition, Fas-mediated cytotoxicity reduced the yield of VSV plaque-forming units (PFU) from Jurkat by an average of 82.0%. An anti-Fas blocking Ab inhibited the RNA degradation and restored the number of VSV PFU to near control levels. These data indicate that the Fas lytic pathway could play a role in the elimination of viruses through degradation of intracellular viral RNA. reserved

Target cell-induced perforin mRNA turnover in NK3.3 cells is mediated by multiple elements within the mRNA coding region.

We have previously shown that in cytolytic cells exposed to sensitive targets the mRNA of the cytolytic protein perforin undergoes rapid downregulation. We now demonstrate that perforin message undergoes accelerated turnover in NK3.3 cells exposed to sensitive TC. This inducible mRNA decay phenomenon is specific for cytolytic protein messages, as levels of the constitutive message beta-actin are unchanged. This TC-induced perforin mRNA turnover cannot be attributed to a blockage of RNA synthesis, or to a rapid half-life (t 1/2). To determine the region(s) within the perforin transcript responsible for governing this TC-mediated turnover event, various segments of the perforin cDNA were cloned and inserted into the 3' UTR of rabbit beta globin (RG). Constructs containing perforin coding region cDNA, but not 3' UTR cDNA, mediated TC-induced mRNA turnover. These data indicate that multiple elements governing perforin mRNA stability reside within the coding region, a novel type of mRNA regulation not previously described.

Low dose total body irradiation: a potent anti-retroviral agent in vivo.

Immunosuppression characterizes many human diseases including leukemia and AIDS. Friend virus (FV)-induced murine leukemia is a useful model for studying both malignancy and immunosuppression. In a previous series of experiments, we have demonstrated that untreated FV-infected mice died within 40 days post-infection, whereas infected mice given 150 cGy total body irradiation (TBI) on days 5 and 12 exhibited long-term survival. In this report, we show that no leukemic cells or type C virus particles are found in the spleens of mice treated with TBI. In addition, both NK activity as well as bone marrow cell's proliferative responses to PHA and Con A were fully restored. This treatment produces long term control of FV-induced murine leukemia, and thus might have relevance for the treatment of a number of immunosuppressive diseases including AIDS.

Caspase-3-like proteases are activated by infection but are not required for replication of vesicular stomatitis virus.

Infection with vesicular stomatitis virus (VSV), the prototype rhabdovirus, causes apoptotic DNA fragmentation, but the role of apoptosis in the VSV-host interaction remains unclear. Apoptosis is the gene-regulated mechanism triggered by a wide variety of stimuli that lead to cell death in a choreographed manner. In the present study, infection of the Jurkat T cell line with VSV led to activation of caspase-3 and caspase-7, with subsequent apoptotic events involving poly (ADP ribose) polymerase (PARP) cleavage, DNA fragmentation, and membrane damage. Caspase activation was correlated with viral protein expression suggesting a link between viral replication and apoptosis. We hypothesized that VSV replication might depend on apoptosis and that the inhibition of apoptosis would lead to significant decreases in viral titers. When various inhibitors of apoptosis in VSV-infected cells were used, PARP cleavage and DNA fragmentation were inhibited but the production of infectious progeny was not affected. In addition, we demonstrated that the activation of caspase-3-like proteases is required for VSV-induced apoptosis but not in vitro viral replication. Apoptosis following VSV infection is likely to be either a host-cell attempt to control viral replication or may be a ploy used by the virus to facilitate its in vivo replication and spread.

Target cell-directed rapid degradation of TNF-alpha messenger RNA in human cytotoxic T cells.

Previously we have shown that upon exposure to sensitive target cells (TC), human cytotoxic T cells (CTL) and natural killer cells (NK) undergo functional inactivation and specifically degrade mRNA-encoding lytic proteins stored in cytoplasmic granules. These lytic proteins include perforin (PFP) and the serine proteases (granzymes). In this study we show that tumor necrosis factor-alpha (TNF-alpha) mRNA undergoes rapid down-regulation after interaction with TC, in a manner identical to PFP and serine protease mRNA. Both PFP and TNF-alpha messages are down-regulated and maintained at low baseline levels while effector cells (EC) are in contact with TC. Moreover, treatment of CTL with antibodies against CD2, CD3, CD8, and class I could not reproduce TC-directed mRNA down-regulation of TNF-alpha and PFP even though lytic ability was inhibited. Anti-CD2 and anti-CD3, however, markedly induced the expression of TNF-alpha mRNA within 15 min. The increase of TNF-alpha mRNA by anti-CD2 and anti-CD3 was offset by the presence of TC, suggesting that TC supply a negative signal to the CTL resulting in degradation of the TNF-alpha mRNA. Furthermore, treatment of CTL with cycloheximide (CHX) did not prevent PFP message loss and did not allow its recovery, suggesting that de novo protein synthesis is not required for mRNA degradation. In contrast, whereas CHX did not prevent TNF-alpha message loss, CHX allowed recovery of TNF-alpha mRNA within 120 min after TC-directed degradation. Taken as a whole, these data suggest that TC provides a negative regulatory signal that triggers the degradation of TNF-alpha message.

Regulation of naïve fetal T-cell migration by the chemokines Exodus-2 and Exodus-3.

We and other workers have recently isolated three novel CC chemokines termed Exodus-1/LARC/Mip-3alpha, Exodus-2/6Ckine/SLC/TCA4, and Exodus-3/Mip-3beta/CKbeta11/ELC. These chemokines share an amino terminal Asp-Cys-Cys-Leu sequence, unique among all chemokines. They also selectively regulate migration of adult T cells. Indeed, there is evidence that Exodus-2 and -3 are critical for adult T-cell adhesion to high endothelial venules in lymph nodes, a rate-limiting step for T-cell trafficking through nodal tissue. Less is known of the factors controlling migration of naïve human fetal T cells. We tested whether these chemokines could regulate chemotaxis in cord blood T-cell populations, and compared that efficacy with normal peripheral blood adult T cells. The findings indicated that naive CD45RA+ cord blood T-cell migration is stimulated by Exodus-2 and -3, and CD4+ cord blood T cells are attracted preferentially by Exodus-2 or -3 as compared with CD8+. Exodus-2 and -3 are likely to be critical in regulating the flux of naive CD4 + fetal T-cell population of secondary lymphoid tissue.

Nature of natural killer cell hyporesponsiveness in the Chediak-Higashi syndrome.

Two unrelated Chediak-Higashi syndrome (CHS) patients in the prelymphomatous phase and their families were tissue typed and tested for their immune functions including natural killer cell (NK) activity. NK activity was assessed by the conventional 4 hr chromium release assay and by a newly developed single cell liquid cytotoxic assay (SLA). SLA permits the simultaneous study of the overall binding of peripheral blood mononuclear cells (PMNC) to the targets and the number of lytic conjugates. As expected, both CHS patients had low NK activity although their family members, including an HLA-identical sibling, were well within normal range. When tested with SLA, the percent total binding of the patient's PMNC was normal, but the number of lytic conjugates was lower than the normal controls. None of the patients had the A3,B7 haplotype which had been previously implicated in NK-hyporesponsiveness. T (including T helper and T suppressor cells) and B lymphocyte functions were normal (as assessed by mitogen stimulation, mixed lymphocyte reaction, and immunoglobulin synthesis). Our results suggest that CHS: (1) is not due to impaired target binding, and (2) is mainly due to low lytic efficiency of mature NK cells.

Evidence that Fas-induced apoptosis leads to S phase arrest.

Cell death by apoptosis is an efficient mechanism of eliminating unwanted or aberrant cells. Triggering of Fas, a member of the tumor necrosis factor (TNF) receptor superfamily, by anti-Fas antibodies or by the Fas ligand (FasL), has been shown to cause cell death by apoptosis. A recent study from our laboratory has demonstrated that Fas crosslinking leads to the dephosphorylation of the tumor suppressor retinoblastoma protein (Rb) and that this dephosphorylation is inhibited by calyculin A, a serine/threonine phosphatase inhibitor. In this investigation, we compared the effect of Fas crosslinking by CH11, an anti-Fas mAb, with two cyclin-dependent kinase (CDK) inhibitors, a peptide that specifically inhibits CDK2 (cdk2 inh) and roscovitine, which inhibits CDK2, CDC2, and CDK5. We illustrate that roscovitine induced DNA fragmentation, whereas cdk2 inh did not. In contrast to Fas-induced apoptosis, roscovitine-induced apoptosis was resistant to calyculin A. Both cdk2 inh and roscovitine induced cleavage of poly (ADP-ribose) polymerase (PARP) within 2 h. Roscovitine, however, led to the degradation of Rb, whereas cdk2 inh did not. Furthermore, both CH11 and roscovitine caused cell cycle arrest in S phase. In contrast, cdk2 inh did not have any effect on Jurkat cell cycle progression. Taken together, our results strongly suggest that the maintenance of Rb in its hyperphosphorylated form during S phase may be necessary for cell survival and that Rb dephosphorylation during S phase may constitute a crucial step in Fas-induced apoptosis.

Fas-mediated apoptosis in T cells involves the dephosphorylation of the retinoblastoma protein by type 1 protein phosphatases.

Apoptosis is triggered by a number of different stimuli including the activation of Fas antigen, a member of the TNF family, by the Fas ligand. The signal transduction events implicated in apoptosis are complex and remain only partially understood. In this study, we used calyculin A, a potent inhibitor of serine/threonine (ser/thr) phosphatases types 1 and 2A, to investigate the role of ser/thr phosphatases in Fas-induced apoptosis. We showed that calyculin A inhibited Fas-induced DNA fragmentation and cytolysis in Jurkat cells and that this inhibition was not due to the modulation of Fas. Okadaic acid also inhibited Fas-induced apoptosis of Jurkat cells, but at much higher concentrations (microM level), thus implicating that type 1 phosphatases rather than type 2A are inhibited at nM concentrations. Cross-linking Fas led to the dephosphorylation of the retinoblastoma gene product (Rb) within 5 min, and to PARP cleavage within 2 h. Both events were inhibited by calyculin A indicating that apoptotic death triggered by Fas cross-linking involves the activation of type 1 ser/thr phosphatases.

Relevance of soluble cytotoxic factors generated by mycoplasma-contaminated targets to natural killer cell-mediated killing.

The possible involvement of soluble cytotoxic factors (SCF) in the lytic mechanism of natural killer (NK) cell-mediated cytotoxicity (CMC) was investigated. Tumor target cells (TC) were examined for their ability to stimulate the release of SCF from human peripheral blood lymphocytes (PBL). SCF were detected when effector cells (EC) were stimulated with mycoplasma-contaminated (MC)-TC and with concanavalin A (Con A), but mycoplasma-free (MF)-TC were unable to induce SCF. Pretreatment of EC with beta-interferon augmented the production of SCF with MC-K562 and Con A, but not with MF-K562. However, MF-K562 once infected with the culture supernatant from MC-K562, as well as mycoplasmas concentrated from culture supernatant, were able to induce SCF. Only Con A was capable of generating cytotoxic supernatants from EC that had been inactivated with MF-K562 and from EC depleted of Leu 11b positive cells. Moreover, when EC were evaluated for NK activity after coculture with MF-K562, their ability to kill fresh MF-K562 in the standard NK-CMC assay was greatly reduced. This loss of lytic potential was not accompanied by SCF release. Collectively, these data suggest that mycoplasma organisms play a crucial role in the induction of human SCF, and the implication that SCF, including NK cytotoxic factors (NKCF), are the lytic mediators in NK-CMC needs to be re-evaluated.

Over-expression of multidrug resistance P-glycoprotein inhibits NK granule-mediated lytic ability without affecting the Fas lytic pathway.

Multidrug resistance (MDR) in tumor cells is commonly associated wich the over-expression of P-glycoprotein (Pgp), the product of the MDR1 gene. In this study, we investigated whether over-expression of Pgp in natural killer (NK) cells would influence their granule- as well as fas-mediated cytolytic activities. YT-INDY, a human NK-like cell line, was transfected with the MDR 1 gene, then tested for Pgp activity the presence of various concentrations of R-verapamil, a potent Pgp inhibitor. We showed that, unlike control YT-INDY, the Pgp activity of the transfectants (YT-mdr(+)) was only partially inhibited by R-verapamil. We also showed that Fas lytic activity was unaltered and that the loss of granule-mediated cytotoxicity was not due to reduced LFA-1 expression or to a decrease in target cell (TC) binding. Our data indicate that Pgp may be involved in the release of cytotoxic molecules.

A comparative study on the specificity of Androctonus australis (Saharan scorpion) and Limulus polyphemus (horseshoe crab) agglutinins.

The specificities of Limulus polyphemus and Androctonus australis sera were compared by agglutination and agglutination-inhibition micromethods with human untreated and enzyme-treated peripheral blood cells (erythrocytes, normal lymphocytes and chronic lymphocytic leukemia lymphocytes). Androctonus sera give higher titers than Limulus sera with erythrocytes and lymphocytes. Both sera show more avidity for leukemic than for normal lymphocytes. Limulus sera fail to agglutinate neuraminidase-treated cells but Androctonus sera react with untreated and neuraminidase-treated erythrocytes at the same degree. Sialyl-lactose is the best inhibitor for Androctonus sera, followed by N-acetylneuraminic and N-glycolylneuraminic acids, glucuronic and galacturonic acids, N-acetyl-D-glucosamine and N-acetyl-D-galactosamine. The amounts of inhibitor required depend upon the particular cell and enzyme treatment: agglutination of neuraminidase-treated erythrocytes by Androctonus is inhibited by lower concentrations of the same compounds that inhibit untreated erythrocytes agglutination. Results suggest that although Limulus and Androctonus have specificity for sialic acid containing compounds, on the cell surface, the Limulus receptors might be different from Androctonus receptors.

NK cells recover early and mediate cytotoxicity via perforin/granzyme and Fas/FasL pathways in umbilical cord blood recipients.

Umbilical cord blood (UCB) is now widely accepted as a source of stem cells in patients with malignant hematologic and genetic disorders. We have recently reported that in a series of 30 pediatric UCB transplant recipients comparable outcome to that anticipated with other unrelated stem cell sources. In our series, however, the probability of GVHD for grade III-IV was 9% and no UCB recipient developed chronic GVHD. The reason for the low incidence of GVHD after UCB transplantation is not fully understood. Because functional NK cells are among the first population of lymphocytes to be detected in UCB transplant recipients, 2 months post-transplant on average, we wanted to establish whether NK cells could be implicated in reducing the risk of GVHD. Here, we confirm that early NK cells detected in UCB transplant recipients activate the granzyme/perforin lytic pathway and, in addition, they can mediate Fas/Fas ligand (FasL) activity, a finding not previously reported. Both pathways develop simultaneously and are detectable months before the other lymphocytes, notably CD8 are fully functional. Our contention, therefore, is that the low GVHD observed in UCB recipients may be partially due to early NK cells.

A comparison of lung metastases and natural killer cell activity in daily fractions and weekly fractions of radiation therapy on murine B16a melanoma.

C57BL/6J male mice were inoculated with 5 X 10(5) B16a melanoma cells. Seven days post-inoculation, when the tumor had grown to 8.0-10.0 mm in diameter, 120 tumor-bearing mice were randomly divided into three groups: (1) sham-irradiated controls, (2) mice receiving 200 cGy five times a week for 6 weeks, and (3) mice receiving 800 cGy once a week for 4 weeks. Thirty mice in each group were sacrificed 47 days postinoculation. Ten mice in each group were observed for the survival time data. The primary tumor was significantly smaller and the number of lung metastases were significantly fewer in mice treated with 800 cGy once a week compared to mice treated with 200 cGy five times a week. When natural killer (NK) cell activity was assessed against YAC-1 tumor targets, it was found to be significantly higher in mice treated with a single large weekly dose of irradiation. These results show that B16a melanoma responds more favorably to a single large dose of irradiation administered once a week compared to the smaller conventional fraction administered five times a week. This beneficial effect correlates with an increase in NK activity, indicating that there may be a causal relationship.