Detection of adenovirus DNA in peripheral blood mononuclear cells by polymerase chain reaction assay.

Adenovirus can establish persistent infections which may reactivate and cause disease in immunocompromised hosts. Lymphocytes have been postulated to serve as a site of adenoviral persistence based upon the ability to isolate adenovirus from tonsils and to detect adenovirus DNA by Southern blot hybridization in peripheral blood mononuclear cells (PBMC). To test this hypothesis, a more sensitive and specific polymerase chain reaction (PCR) assay was developed to detect adenovirus DNA. Two sets of nested primers were designed to conserved sequences in the adenovirus E1A and hexon genes. The E1A and hexon primers amplified DNA from representative adenoviral serotypes in all six adenoviral groups (A-F). Both primers detected a single copy of the adenovirus type 2 genome but were less sensitive for the group B type 35. None of 33 PBMC specimens from healthy adults and only one of 40 pediatric samples was positive (at a low level) for adenovirus DNA by nested PCR assay. In comparison, PBMC from two children with fatal adenoviral infection were both strongly positive for adenovirus DNA. It is concluded that, in contrast to a previous study, PBMC are not a common site of persistent group C adenoviral infection. In addition, assay of PBMC by the adenovirus-specific PCR may help detect early invasive disease and warrants further evaluation.

Human adenovirus-specific CD8+ T-cell responses are not inhibited by E3-19K in the presence of gamma interferon.

Adenovirus has considerable potential as a gene therapy vector, but recent animal data suggest that transduced cells are destroyed by adenovirus-specific cytotoxic T-lymphocyte (CTL) responses. Therefore, it will be important to develop strategies to evade adenovirus-specific CTL responses in humans. As a first step, an assay was developed to detect and characterize human CTLs directed against adenovirus. Adenovirus-specific CTL responses were demonstrated to be present in four of five healthy adults by in vitro stimulation of peripheral blood mononuclear cells with autologous fibroblasts infected with the adenovirus type 2 (Ad2) E3 deletion mutant Ad2+ND1. Killing by adenovirus-specific CTLs was major histocompatibility complex class I restricted and was documented to be mediated by CD8+ T cells. Wild-type-Ad2-infected cells were poor CTL targets compared with cells infected with the E3 deletion mutant because of the expression of E3-19K, an early viral glycoprotein which prevents transport of major histocompatibility complex class I antigens out of the endoplasmic reticulum to the cell surface. However, preincubation of targets with gamma interferon resulted in enhanced killing of wild-type-Ad2-infected cells, to levels comparable to those obtained with Ad2+ ND1-infected cells. Radioimmunoprecipitation analysis revealed that gamma interferon not only increased the synthesis of class I antigens but also allowed excess molecules to escape from the endoplasmic reticulum. It is concluded that E3-19K expression in adenovirus-infected cells inhibits human CTL recognition in vitro but that gamma interferon may help overcome the E3-19K effect during acute infection in vivo.

Spontaneous, persistent infection of a B-cell lymphoma with adenovirus.

An adenovirus culture-positive lymphoblastoid cell line was derived from a bone marrow transplant recipient with fatal B-cell lymphoproliferative disease and adenovirus pneumonia. At autopsy, focal areas of the lymphoma infiltrating the patient's lung were positive for adenovirus proteins by immunohistochemical staining. The Epstein-Barr virus-transformed B-cell line Mk, established from pleural fluid cells, contained adenovirus virions in both the nucleus and the cytoplasm by electron microscopy. The majority of Mk cells expressed adenovirus proteins and produced a high level of infectious adenovirus by plaque assay analysis. However, in contrast to the rapid cell death induced by adenovirus in other permissive cell lines, Mk was maintained stably in tissue culture for 6 months. These data indicate that adenoviral replication is not sufficient for cell lysis and confirm that adenovirus can cause persistent infection in human lymphoid cells in vivo.

Characterization of human proliferative T cell responses to adenovirus.

Although cellular immune responses are likely important for recovery from acute adenovirus infection, they have not been studied in humans. As a first step, a sensitive assay for the detection of adenovirus-specific proliferative T cell responses was developed. Peripheral blood mononuclear cells from 29 of 30 healthy adults exhibited specific proliferative responses to adenovirus type 2 antigen. Adenovirus-specific proliferation was found to be mediated by CD4+ T cells. Specific proliferative responses were detected to purified Ad2 virions and by Ad2-infected cell lysates, indicating that adenovirus structural proteins are important targets. In addition, specific proliferative responses to the uncommon adenovirus type 35 were found in persons without serologic evidence of prior Ad35 infection. These data suggest that adenovirus-specific CD4+ T cells recognize conserved antigens among different serotypes and that a majority of persons develop long-lived CD4+ T cell responses to adenovirus.

Immunological evaluation of pediatric cancer patients receiving recombinant interleukin-2 in a phase I trial.

Immunological evaluations were performed on 14 pediatric cancer patients who received human recombinant interleukin-2 (rIL-2) as a bolus intravenous infusion every 8 h for 5 consecutive days in a phase I trial. Three-to-four patients were treated at dose levels of 10, 30, 60, and 100 x 10(3) Cetus U/kg. Six of the patients had stage D neuroblastoma; the remainder had other solid tumors or leukemias. Infusion of rIL-2 was associated with a rapid margination of IL-2-responsive cells followed by demargination and heightened proliferative and cytotoxic activity after therapy was completed. The predominant phenotypic change in circulating peripheral blood mononuclear cells (PBMC) was an increase in CD2 expression by CD56+ natural killer (NK) cells. Appearance of CD2+ CD56+ cells in the circulation correlated with increased lymphokine-activated killer (LAK) cell activity as defined by the ability to kill NK-resistant Daudi tumor cells in vitro. Sustained LAK activity appeared to be dependent on the bioavailability of rIL-2 in vivo as well as in vitro. After rIL-2 therapy, PBMC that were highly responsive to rIL-2 (activated and "poised" LAK cells) persisted for at least 72 h. In the patients tested, increased lysis of autologous and/or allogeneic, histologically similar tumor cell lines was also observed after therapy. The immunoenhancing effects of rIL-2 occurred even at the lower doses used in this study. However, an objective tumor response was not observed in any of the patients.

Kinetic analysis of human IL-2 activated cytotoxic cells.

Kinetic analysis was used to define lytic events in peripheral blood mononuclear cells (PBMC) activated in lymphokine conditioned medium (LCM) and recombinant interleukin-2 (rIL-2). This analysis provided quantitative information on the functional properties of these lymphokine-activated killer (LAK) cells against NK-resistant and NK-sensitive tumor cell lines. The maximum rate of target cell lysis (Vmax) and Km (target cell number resulting in 1/2 Vmax) were determined. IL-2 activated effector cells that bound to target cells also lysed them (i.e., non-lytic bystander lymphocytes did not influence the determination of kinetic parameters) in contrast to lysis mediated by unactivated NK cells. The extent of LAK cell binding to tumor target cells was dependent upon the tumor type. LAK cell frequency determinations were calculated where Km approximated the concentration of LAK cells that were capable of killing a particular target. LAK cells generated in rIL-2 were lytically more efficient than those activated in LCM, and T-depletion resulted in a LAK population with the highest maximum rate of lysis. The use of kinetic analysis to evaluate LAK cell frequencies and quantitate lytic events will be useful in determining the effects of drugs, biological response modifiers and disease states on LAK cell function.

Ability of circular extrachromosomal DNA molecules to carry amplified MYCN proto-oncogenes in human neuroblastomas in vivo.

Amplification of the proto-oncogene MYCN (also known as N-myc) in neuroblastomas has been shown to correlate with both disease stage and prognosis, yet little is known about the DNA structures that carry amplified MYCN genes in neuroblastomas in vivo. We have used DNA irradiation and pulsed-field gel electrophoresis to analyze MYCN amplification structures in eight neuroblastomas from separate patients (four primary tumors and four metastatic lesions exhibiting MYCN amplification). Six of the eight neuroblastomas (three primary tumors and three metastatic lesions) exhibited MYCN DNA irradiation profiles consistent with the presence of circular extrachromosomal DNA amplification structures. Five neuroblastomas possessed amplification structures within the size range of double minute chromosomes, and one contained smaller DNA circles. Two neuroblastomas exhibited MYCN DNA irradiation patterns consistent with larger (presumably chromosomal) amplification structures. Multiple sizes of DNA circles were observed in the neuroblastomas of four different patients, implying in vivo multimerization of amplification structures. The presence of circular MYCN amplification structures in six of eight neuroblastomas examined suggests that circular DNA molecules are important structures in in vivo gene amplification.

Preservation of lymphokine-activated killer activity following T cell depletion of human bone marrow.

T cell depletion has decreased the incidence and severity of graft-versus-host disease following transplantation of allogeneic bone marrow. In the treatment of leukemia, decreased GVHD has often been associated with diminished antileukemia or graft-versus-leukemia (GVL) reactivity resulting in higher relapse rates. However, we have not seen a loss of the GVL effect following transplantation of marrow grafts depleted of CD3+ T cells. This suggests that non-T-cell effectors may play a role in preventing leukemic relapse. To study whether natural killer and lymphokine-activated killer (LAK) activity in BM was compromised by T cell depletion, the effect of T-cell-specific monoclonal antibodies against CD3 and CD6 determinants alone, or in combination, on the generation and expansion of NK/LAK cells was examined in vitro and compared to the effect of T depletion on mitogen-driven T cell proliferation. Limiting dilution analysis revealed that T depletion with CD3 and/or CD6 specific antibodies significantly reduced the number of proliferating T lymphocytes but did not significantly affect the frequency of cells able to expand and mediate LAK activity. Bone marrow, depleted of CD3+ or CD6+ T cells, generated levels of LAK activity equivalent to non-T-cell-depleted bone marrow following long-term culture in recombinant interleukin 2. CD3- NKH-1+ cells were the predominant population in rIL-2 expanded marrow cultures prior to transplant and in the peripheral blood of patients who had received a CD3-depleted marrow graft 21-65 days earlier. These studies show that it is possible to selectively reduce GVH-reactive T cells in allogeneic bone marrow while retaining non-T-effector cells with potential to mediate an antileukemia effect in vivo.

The autologous mixed lymphocyte response: isolation of mononuclear cell populations by counterflow centrifugal elutriation.

We have used counterflow centrifugal elutriation (CCE) to isolate cell populations for an autologous mixed lymphocyte response (AMLR). Enriched T lymphocyte and monocyte populations were obtained rapidly without the need for cells to first form rosettes with sheep RBC (T cells) or adhere to plastic (monocytes). In 16 separate experiments, peripheral blood (PB) was separated by CCE (9) and/or a plastic adherence and nylon wool column method (9). Viability by all methods was greater than 95% CCE yielded greater than 97% lymphocytes in F20 and greater than 90% lymphocytes in F26. Cells obtained after passage through nylon wool yielded greater than 90% lymphocytes. Both CCE (F34) and plastic adherence techniques yielded greater than 75% monocytes. Similar AMLR response was obtained using either cell separation method. We conclude that mononuclear cells isolated by CCE are responsive in the AMLR and yield comparable data to other more tedious techniques.

Contribution of electron microscopy of cell cultures to the classification of three round cell sarcomas in children.

We have studied three round cell sarcomas from pediatric patients in tissue culture to compare the electron microscopic morphology of cells in culture to cells from original biopsy specimens. None of the original tumors displayed distinctive features by light microscopy that would allow classification of a specific tumor type, and electron microscopy was not helpful in identifying specific morphologic features that would allow further classification of tumor types. However, electron microscopy of cells in culture from the three neoplasms revealed distinctive morphologic features that did allow further classification of all three tumors. Cells from an inguinal lymph node, which were cultured in soft agar tumor colony-forming assay, revealed Z-bands and actin and myosin filaments indicative of a rhabdomyosarcomatous nature for the tumor. Cells from 5-day, 10-day, and 4-month cultures of a bone marrow metastasis of a second tumor revealed features of skeletal muscle in the young cultures and neuroblasts in the older culture, suggesting a primitive neuroectodermal neoplasm. Cultured cells from the third tumor, a neoplasm of the calf in an infant, displayed large lakes of glycogen, typical of cells of Ewing's sarcoma, which were not present in the cells examined from the original lesion. Ultrastructural studies of cells in culture have the potential to add morphologic data that may be useful to further define and classify a neoplasm, as illustrated in the 3 cases reported here.

Distinguishing characteristics of a new neuroblastoma cell line.

The characteristics of a new neuroblastoma cell line (MC-NB-1) established from the bone marrow of a 2-year-old male are described. Morphologically, the cells appear as flattened and epithelial-like or as small and spherical. Electron microscopy demonstrated microtubules and dense core secretory granules. The doubling time was approximately 35 hr. Isoenzyme patterns and catecholamine secretion indicated a human line of neuronal origin. The soft agar tumor colony forming system demonstrated drug resistance in vitro comparable to in vivo nonresponsiveness. The stemline karyotype of MC-NB-1 is 44,Y,del(1) (p22:), -4, -7, +del(7)(q22:), -16, +t(7;16)(16pter leads to 16q24::7q22 leads to 7q32), -17. Additionally, double-minute bodies were observed. However, no evidence of homogeneous staining regions (HSRs) were detected.

In vitro enhancement of peripheral blood mononuclear cell natural killer activity following short term incubation with fetal calf serum.

This study demonstrates that short term incubation of blood mononuclear cells (MC) in heterologous sera enhances their natural killer (NK) but not their antibody dependent killer (K) activity, and confirms that NK stimulation is not related to blastogenesis. MC were obtained from healthy donors and incubated with RPMI 1640 supplemented with various serum sources. NK and K activity of incubated vs. fresh MC against SK-N-SH, Chang or Raji cell line targets were compared in a 4-hr 51Cr release assay. A significant (p less 0.01) increase in NK cytotoxicity was detected when MC were incubated with fetal calf serum (FCS) or human AB serum (ABS) at 37 degrees C. When a more sensitive NK target cell (K-562) was used only the cells incubated in FCS demonstrated increased NK cytotoxicity. Augmentation of NK cell activity by FCS was not related to blastogenesis, mitosis, natural antibodies against lymphocytes or target cells, immunoglobulin complexes or alterations in MC OKT4 and OKT8 subpopulations. In contrast to NK activity, K cytotoxicity was not increased after incubation at 37 degrees C in FCS or ABS, and it was depressed in IPT (p less than 0.05). Thus FCS is capable of stimulating NK cell activity against human tumor cell lines in less in less than 24 hr.

Association between HLA-D region antigens and disease-free survival in childhood non-T, non-B acute lymphocytic leukemia.

The frequency of three serologically defined HLA-D region antigens--DR, MB, and MT--was determined in a group of 74 children with non-T, non-B acute lymphocytic leukemia (ALL). Statistically, there were no significant differences in the frequency of any antigen in these ALL patients as compared with a panel of 85 normal controls. However, significant differences in HLA-DR frequencies were observed between patients who relapsed or who remained disease-free during a 30-mo period of chemotherapy. An increased incidence of relapse was associated with DR5, while disease-free remission during chemotherapy was associated with DR7. Life table analysis also demonstrated that DR5 was significantly associated with a decrease in disease-free survival in these patients. These data suggest that HLA-associated genetic factors may influence the responses of ALL patients to chemotherapy.

Bone marrow and extramedullary variations of cell membrane antigen expression in childhood lymphoid neoplasias at relapse.

To increase our knowledge of the pathogenesis of tumour recurrence, cell membrane phenotypes were determined on bone marrow and extramedullary tumour cells at diagnosis and at relapse in 24 children with lymphoid malignancies. There were 19 patients with acute lymphoblastic leukemia and five with non-Hodgkin's lymphoma. Membrane characteristics used for classification were E-rosetting, T antigen, surface immunoglobulin (sIg) and Ia antigen. Twelve patients were phenotyped as non-T, non-B, five as T-like and seven as B-like leukemia/lymphoma. Eighty-eight tissue samples were assayed for cell surface markers at the time of relapse(s). Lymphoblasts from 18 children (75%) demonstrated no variation in membrane marker expression. Six patients (25%) showed alteration of antigen expression on lymphoblasts obtained during relapses. This was manifested by loss of Ia antigen in two patients, loss of E-rosetting in one patient and loss of sIgD in one patient. Lymphoblasts from two patients, initially non-reactive with the four membrane markers utilized, expressed Ia antigen on subsequent relapses. Despite these variations no patient was categorized differently (i.e. T-like becoming non-T, non-B). Simultaneous lymphoblast phenotype determination from multiple body sites in 13 children showed no variation in marker analysis. A lymphoblast's phenotype remains stable throughout repeated relapses and is not influenced by relapse site in most children with lymphoid neoplasias. This information may be helpful in designing protocols where cytotoxic monoclonal antibodies are used as a treatment modality in patients with recurrent disease.