Li-En Shao

Suppression of IL-6 biological activities by activin A and implications for inflammatory arthropathies

Activin A is a cytokine whose multiple functions have yet to be fully determined. In this study, the role of proinflammatory cytokines in regulatory control of activin A production was shown in synoviocytes and chondrocytes. Additional facets of functional inflammation‐related activities of activin A were also determined. Results showed that activin A concentrations in the synovial fluid of patients with rheumatoid arthritis and gout were elevated relative to those in patients with osteoarthritis. Further studies showed that production of activin A by synoviocytes and chondrocytes in culture was stimulated by cytokines such as IL‐1, transforming growth factor‐beta (TGF‐β), interferon‐gamma (IFN‐γ), and IL‐8, consistent with previous studies in regard to the control of activin A production in marrow stromal cells and monocytes by cytokines, glucocorticoids and retinoic acid. In addition, the relationship of activin A to IL‐6‐induced biological activities was investigated. Three major IL‐6 activities involved in inflammatory responses were found to be suppressed by activin A. In a dose‐dependent manner, activin A efficiently suppressed IL‐6‐induced proliferation of 7TD1 B lymphoid cells, phagocytic activity of monocytic M1 cells, and fibrinogen production in HepG2. Therefore, it is likely that activin A serves as a suppressor for IL‐6, dampening inflammatory responses, and has the potential to perform some previously unrecognized roles in inflammation.

Induced expression of the new cytokine, activin A, in human monocytes: inhibition by glucocorticoids and retinoic acid.

The capacity of recombinant human granulocyte–macrophage colony‐stimulating factor (GM‐CSF), glucocorticoids or all‐trans‐retinoic acid to modulate production of activin A by human monocytes was studied. It was shown that GM‐CSF stimulated monocytes to accumulate activin A RNA after as few as 4 hr of incubation, reaching a peak of stimulation at approximately 16 hr of incubation. The activin A transcripts accumulated in the monocytes after stimulation with only 5 U/ml of GM‐CSF and reached a maximum plateau level of expression between 25 and 50 U/ml of GM‐CSF. Biologically active activin A molecules were detected in the conditioned media by a bioassay, performed both in the absence and presence of a neutralizing antiserum for activin A. Accumulation of bioactive activin A in conditioned medium of monocyte cultures was detected after 24 hr of incubation with GM‐CSF and high levels of activin A were maintained for 72 hr. The production of the dimeric βAβA in these monocytes was further confirmed by a sandwich enzyme‐linked immunosorbent assay (ELISA) specific for activin A. In contrast to the stimulatory effect of GM‐CSF, hydrocortisone, dexamethasone or all‐trans‐retinoic acid at 1 × 10−7 to 1 × 10−5 m inhibited the constitutive expression of activin A and greatly suppressed the GM‐CSF‐stimulated production. Thus, the expression of activin A is modulated in monocytes by different agents. These observations may imply new roles for activin A at sites of inflammation where monocytes accumulate.

Detection of Functional and Dimeric Activin A in Human Marrow Microenvironment

Activin A, which was initially recognized as a gonadal protein, was implicated in the modulation of erythropoiesis through a paracrine control in the bone marrow microenvironment. Present studies demonstrate that, in contrast to T lymphocytes and cultured skin fibroblasts, human marrow stromal cells produce a functional and dimeric beta A beta A molecule (i.e., activin A). RT-PCR further indicates that both alpha and beta A mRNAs of inhibin A/activin A are produced in human stromal cells. The level of beta A subunit mRNAs, however, is in large excess over that of alpha subunit mRNAs, suggesting the predominant production of beta A beta A dimers, as well as some inhibin A (alpha beta A). It should be noted, however, that the beta A subunit can form dimeric proteins other than activin A, such as activin AB (beta A beta B) and inhibin A (alpha beta A). Hence, the presence of the beta A subunit may not necessarily indicate the production of the activin A molecule in any tissue. Therefore, a special quantitative sandwich ELISA assay specific for the dimeric beta A beta A molecule was developed for the measurement of activin A. With this assay, production of activin A in marrow stromal cells is found to be greatly enhanced by cytokines and inflammatory mediators such as TNF-alpha, IL-1 alpha, and lipopolysaccharide. These studies thus suggest that inflammatory cytokines are the inducers for activin A, probably serving a role of up-regulating activin A production locally in bone marrow microenvironment. At present, activin A is not known to play any role in inflammatory reaction; this study may thus raise the possibility that activin A performs more functions than are currently recognized. Alternatively, the enhanced production of this molecule in the bone marrow microenvironment may be regarded as a compensatory mechanism in host defenses, countering inflammatory mediators that are known to suppress erythropoiesis.

Engraftment of human T-cell acute lymphoblastic leukemia in immunodeficient NOD/SCID mice which have been preconditioned by injection of human cord blood.

Childhood T‐cell acute lymphoblastic leukemia (T‐ALL) is one of the most common childhood cancers. Study of leukemia biology, as well as preclinical testing of potential therapeutic regimens directed at T‐ALL, has been impeded by the lack of an efficient in vivo model of primary leukemia. We have reported elsewhere some observations that human cord blood conditioned medium enhances leukemia colony formation in vitro and preconditioning of sublethally irradiated nonobese diabetic/ severe combined immunodeficient (NOD/SCID) mice with cord blood mononuclear cells (MNCs) facilitates the subsequent engraftment of primary T‐ALL cells in these mice. Here we characterize in greater detail this in vivo xenograft model of human leukemia in NOD/SCID mice. Consistent with the thesis that cord blood facilitates engraftment, the engraftment of human leukemia can be shown to increase with increasing number of cord blood MNCs injected. In addition, we documented the expression of chemokine receptor CXCR4 by primary T‐ALL from patients and found that the presence of these receptors did not result in the transmigration of T‐ALL cells induced by stromal cell‐derived factor‐1α. Finally, we show that in this xenograft system T‐ALL cells recovered from engrafted bone marrow are characterized by upregulated expression of interleukin 2 receptor γ chain, suggesting that cord blood preconditioning may function in part to increase T‐ALL responsiveness to growth factor(s).

Analysis of activin A gene expression in human bone marrow stromal cells

Activin A, a member of the TGF‐β superfamily, plays roles in differentiation and development, including hematopoiesis. Our previous studies indicated that the expression of activin A by human bone marrow cells and monocytes is highly regulated by inflammatory cytokines and glucocorticoids. The present study was undertaken to investigate the regulation of activin A gene expression in the human bone marrow stromal cell lines L87/4 and HS‐5, as well as in primary stromal cells. Northern blots demonstrated that, like primary stromal cells, the cell lines expressed four activin A RNA transcripts (6.4, 4.0, 2.8, and 1.6 kb), although distribution of the RNA among the four sizes varied. The locations of the 5′ ends of the RNAs were investigated by Northern blots and RNase protection assays. The results identified a transcription start site at 212 nucleotides upstream of the translation start codon. In addition, luciferase expression assays of a series of deletion constructs were used to identify regulatory sequences upstream of the activin A gene. A 58 bp upstream sequence exhibits promoter activity. However, severalfold higher expression requires a positive element consisting of an additional 71 bp of the upstream region. Promoter activity was also identified between 2.5 and 3.6 kb upstream of the start codon. These findings suggest that expression of activin A at the transcriptional level follows complex patterns of regulation. J. Cell. Biochem. 70:8‐21, 1998.

Truncated activin type II receptor inhibits erythroid differentiation in K562 cells

Two receptor serine/threonine kinases (types I and II) have been identified as signaling transducing activin receptors. We studied the possibility of inhibiting activin A‐dependent differentiation in K562 cells, using a dominant negative mutant of type II receptor. A vector was constructed expressing activin type II truncated receptor (ActRIIa) that lacks the cytoplasmic kinase domain. Since activin type I and II receptors form heteromeric complexes for signaling, the mutant receptors compete for binding to endogenous receptors, hence acting in a dominant negative fashion. K562 cells were stably transfected with ActRIIa, and independent clones were expanded. The truncated cDNA was integrated into the genome of the transfectants, as shown by polymerase chain reaction; and the surface expression of truncated receptors was shown by affinity cross‐linking with 125I‐activin A. In wild‐type K562 cells, activin A induced erythroid differentiation and cells started to express hemoglobins. In transfected cells expressing ActRIIa, the induction of erythroid differentiation was abrogated and less than 10% of cells were hemoglobin‐containing cells after culture with activin A. Further transfection with wild‐type type II receptors rescued the mutant phenotype of these transfectants, indicating that the effect of ActRIIa is dominant negative. In addition, phosphorylation of the cytoplasmic kinase domain of the type II receptor in vitro confirms the autophosphorylation of this portion of the receptor. Therefore, induction of erythroid differentiation in vitro is mediated through the cell surface activin receptor, and interference with this receptor signaling inhibits this process of differentiation in K562 cells. J. Cell. Biochem. 78:24–33, 2000.

Human cord blood conditioned medium enhances leukemia colony formation in vitro

Childhood T-cell acute lymphoblastic leukemia (T-ALL) is one of the most common childhood cancers. Recently, we observed that pre-conditioning sub-lethally irradiated immunodeficient mice with human cord blood mononuclear cells facilitates in these mice high level engraftment of primary T-ALL cells obtained from patients. Here we report that human cord blood cells secrete a factor(s) which markedly enhances in vitro both colony number and burst size of the T-ALL clonogenic progenitors from patients. The enhancing activity does not correspond to IL-2, IL-15, nor to several other cytokines implicated in T cell proliferation/activation. Thus, it is possible cord blood may secrete an as yet unidentified factor(s) acting on leukemia clonogenic progenitors of T-ALL. Collectively, these studies should prove invaluable in addressing the growth properties of primary T-ALL cells from patients.

Phenotypic and functional characterization of a new human macrophage cell line K1m demonstrating immunophagocytic activity and signalling through HLA class II

A human macrophage line, designated K1m, has been established from peripheral blood. K1m expresses a number of lineage‐specific markers as well as a broad array of intercellular adhesion molecules. In particular, K1m expresses high levels of human leucocyte antigen (HLA) class I and class II. In response to ligation of HLA class II (HLA‐DR), but not in response to ligation of HLA class I, K1m forms tighter homotypic aggregates and develops a striking ‘stellate’ culture phenotype. K1m also expresses Fc receptors for immunoglobulin G (IgG) (CD64, CD32, and CD16) and can be shown to phagocytose polystyrene latex beads, as well as neuroblastoma cells in the presence of tumour‐specific monoclonal antibody (mAb). The K1m cell line should therefore prove useful for studying both signalling through macrophage HLA class II and immunophagocytosis.

Sensitivity of committed hematopoietic progenitor cells in vitro (BFU-E, CFU-E, CFU-GM) and two human carcinoma cell lines toward rhodamine-123 and phosphonium salt II-41.

Lipophilic cationic compounds accumulate more rapidly in the mitochondria of many carcinoma-derived cells than in non-transformed cells, thus leading to their pronounced cytotoxic effects on carcinoma cells. In this report, in order to measure tumoricidal effects vs cytotoxicity to normal hematopoietic progenitors, we studied the sensitivity of committed human hematopoietic cells in vitro and two human carcinoma cell lines (2008 ovary carcinoma cells and HT29 colon cells) toward two such compounds, rhodamine-123 and phosphonium salt II-41. Continuous exposure of human marrow cells to rhodamine-123 or phosphonium salt II-41 for 7 and 14 days produced dose-related inhibition of colony formation of erythroid burst-forming units (BFU-E), erythroid colony forming units (CFU-E), and CFU-granulocyte/macrophage (CFU-GM). The average values of IC50 for several different human bone marrows are approximately 0.9-1.1 microM for rhodamine-123 toward BFU-E, CFU-E and CFU-GM, and 31-38 microM for phosphonium salt II-41 toward the same hematopoietic progenitors. These IC50 values are similar for each type of hematopoietic progenitors. In each case, rhodamine-123 appears to be at least 30-fold more growth suppressive than phosphonium salt II-41 in these in vitro colony assays. In addition, the sensitivity of these hematopoietic progenitors toward these two compounds is comparable to the inhibition of colony formation for the two human carcinoma cell lines. The lack of differences in the sensitivity among the various hematopoietic progenitors in vitro may disagree with previous studies showing there are vast differences in the state of cell cycle for these hematopoietic progenitor cells. However, these observations about the cytotoxicity in vitro can be explained by assuming that the cytotoxicity of these compounds depends on other factors such as differentiation processes, which result in the appearance of many or very active mitochondria. Alternatively, the lack of differences in the sensitivity of the in vitro colony formation can also be attributed to a reported decrease in expression of P-glycoprotein, a multidrug efflux pump, in the differentiating hematopoietic progeny cells.