The importance of cellular environment to function of the CD44 matrix receptor.

Much has been learned recently by experimental manipulation of the structure of CD44 and assessment of the resulting functions. However, even greater structural variation is naturally introduced by CD44-bearing cells. A structural model is now available for the portion of CD44 that recognizes hyaluronan, but it is clear that all domains of the molecule influence CD44 functions.

Limitin: An interferon-like cytokine that preferentially influences B-lymphocyte precursors.

We have identified an interferon-like cytokine, limitin, on the basis of its ability to arrest the growth of or kill lympho-hematopoietic cells. Limitin strongly inhibited B lymphopoiesis in vitro and in vivo but had little influence on either myelopoiesis or erythropoiesis. Because limitin uses the interferon alpha/beta receptors and induces interferon regulatory factor-1, it may represent a previously unknown type I interferon prototype. However, preferential B-lineage growth inhibition and activation of Janus kinase 2 in a myelomonocytic leukemia line have not been described for previously known interferons.

Defective colony formation by B lymphocytes from CBA/N and C3H/HeJ mice.

B cells from CBA/N mice did not form colonies in semisolid agar cultures under circumstances where normal B-cell clonal proliferation is linear with respect to the number of functional cells cultured. This was no due to the unresponsiveness of CBA/N cells to mitogens, and under appropriate liquid culture conditions many CBA/N lymphocytes differentiated to plasma cells containing large amounts of IgM in response to LPS. On the other hand, the same cells proliferated and matured poorly in liquid cultures prepared at low-cell density. The frequency of granulocyte-macrophage progenitors and multipotential hemopoietic stem cells in bone marrow, ability of peritoneal macrophages to elaborate soluble enhancing factors, and levels of serum inhibitors were normal in CBA/N mice. Together with the results of cell-mixing experiments, these findings confirm the selective and intrinsic nature of the CBA/N deficiency. It is suggested that the B-cell cloning technique may be of value in selectively enumerating and assessing functional capability of thymus-independent B cells. C3H/HeJ mice which have previously only been known to be hyporesponsive to certain forms of lipopolysaccharide had a subnormal incidence of colony-forming B cells.

Precocious and enhanced functional maturation of B lineage cells in New Zealand Black mice during embryonic development.

Previous reports suggest that large numbers of immunoglobulin-secreting cells appear in tissues of NZB strain mice from the time of birth. In this study, we investigated the development of B lineage cells during embryonic life and found that they were present 2-3 d earlier and in higher numbers in NZB embryos than several other strains of mice. That is, liver cell suspensions from NZB embryos contained larger numbers of surface Ig (sIg)- cells that could form B cell colonies in mitogen-dependent semisolid agar culture. Sephadex G-10-adherent cell depletion diminished numbers of colonies and this was partially restored by addition of humoral factors. The latter were partially purified from serum of very young NZB mice. These findings document that abnormal changes take place in B lineage cells and possibly also in cells that regulate their maturation in NZB strain mice at a very early stage of development.

Suppression of B lymphopoiesis during normal pregnancy.

We describe a dramatic reduction in numbers and activity of committed B lymphocyte precursors in the bone marrow of normal pregnant mice. Changes in cells responsive to IL-7 were evident as early as 6.5 d of pregnancy and values were < 10% of normal at parturition. B lineage precursors, identified by display of CD45R and absence of surface IgM, were also substantially depressed, and subpopulations representing different stages in the B lineage were assessed by three-color flow cytometry. Early pro-B cells are medium to large in size and have been previously characterized by low expression of the heat-stable antigen (HSA). This category of cells was not reduced, and in fact may have been slightly elevated, during pregnancy. In contrast, all subsequent populations of B lineage precursors, defined by patterns of expression of heat-stable and CD43 antigens, were substantially depressed. The immediate precursors of B cells (small pre-B cells) were identified by small size, expression of CD45R, absence of CD43, and lack of surface IgM. These were the most reduced of any phenotypically defined population in bone marrow. Numbers of newly formed B cells, characterized by the presence of sIgM, but not sIgD, were also diminished. However, B cells with a mature phenotype (sIgM+, sIgD+) were present in normal to somewhat elevated numbers. Mitogen-responsive B cells clonable in a semisolid agar assay were not significantly affected. A bromodeoxyuridine (BrdU) labeling technique was used to evaluate mitotic activity, which revealed an increased proportion of long-lived lymphocytes in the bone marrow of pregnant mice. These observations indicate that B lymphopoiesis is markedly downregulated during pregnancy and that all precursor populations beyond the early pro-B cell stage are affected. The pregnancy-related changes in bone marrow were selective for B lineage precursors, as cells expressing myeloid and erythroid markers were not reduced. In spleen, evidence was obtained for partial depletion of one subset of B cells. These cells, which have been reported to be recent immigrants from marrow, are characterized as having high levels of sIgM and HSA. Changes in other major B lymphocyte subsets in the spleen were less remarkable. When considered with results from the BrdU labeling procedure, the findings indicate that both production and export of lymphocytes from marrow may be substantially decreased. Numbers of B cell precursors were higher in postpartum animals whose litters were removed at birth, suggesting that lactation may prolong regeneration of lymphocyte production.(ABSTRACT TRUNCATED AT 400 WORDS)

Growth of B-lymphocytes clones in semisolid culture is mitogen dependent.

A substance which was mitogenic for murine B lymphocytes in the presence of 2-mercaptoethanol was isolated from agar. Stimulating activity of this material was stable to proteolysis or protein denaturants but was destroyed by periodate treatment. Agar-derived mitogen stimulation was distinct from that obtained with dextran sulfate or lipopolysaccharide and may define different populations of B lymphocytes.

Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis.

Adhesion molecules are probably required for retention of maturing lymphocyte precursors in bone marrow, where they closely interact with and are dependent on stromal cells. Lymphomyeloid cell lines avidly adhere to cloned stromal cell lines in culture and screening pairs of these resulted in a selection strategy for a new monoclonal antibody to a leukocyte adhesion molecule. Immunoprecipitation analyses and comparison to a previously described antibody showed that it recognizes the alpha 4 chain of the integrin, VLA-4. This antibody totally inhibited lymphopoiesis and retarded myelopoiesis in long-term bone marrow cultures. A similar selection strategy resulted in two additional antibodies which define a single 100-kD species on stromal cells. This stromal cell adhesion molecule is a potential counter-receptor/ligand for VLA-4 on murine lympho-myeloid cells. Our findings suggest a new role for VLA-4 in lymphoid progenitor-microenvironment interactions. Recognition molecules that function in cell migration and inflammation in peripheral tissues may be important for steady-state lymphopoiesis within bone marrow.

Glycosylation of CD44 negatively regulates its recognition of hyaluronan.

Although CD44 is expressed on a wide variety of cell types, few of them use it to recognize the ligand hyaluronan (HA). A glycosylation-defective clone of Chinese hamster ovary cells (Lec 8) bound HA, demonstrating that complete processing of glycoproteins with addition of a full complement of sialic acid is not required. On the contrary, subsequent findings revealed that complex sugars on CD44 can actually inhibit ligand recognition. Two subclones of wild-type Chinese hamster ovary cells with similar amounts of surface CD44 were isolated on the basis of HA binding and found to differ with respect to CD44 size as well as staining with fluorescent lectins. Treatment of the nonbinding clone with tunicamycin reduced the size of the protein and allowed the cells to recognize HA via CD44. This function was also induced by treatment with deglycosylating enzymes (either a mixture of endoglycosidase F and N-glycosidase F or neuraminidase alone). A possible role for glycosylation in regulation of adhesion was then sought with a series of normal and transformed murine cells. Disruption of glycosylation or treatment with deglycosylating enzymes did not induce ligand binding in an interleukin 7-dependent pre-B cell line, and splenic B cells also appeared to be in an inactive state. Some normal B cells acquired the ability to recognize HA after stimulation with lipopolysaccharide or interleukin 5 and had distinctive surface characteristics (loss of immunoglobulin D and acquisition of CD43). An additional subset of activated cells might have been in a transitional state, because the cells bound ligand after neuraminidase treatment. The ligand-binding ability of a purified CD44-immunoglobulin fusion protein dramatically increased after neuraminidase treatment. Thus, differential glycosylation of this molecule is sufficient to influence its recognition function. Cell adhesion involving HA can be regulated by multiple mechanisms, one of which involves variable glycosylation of CD44.

Highly restricted expression of a stromal cell determinant in mouse bone marrow in vivo.

B lymphocyte precursor cells in mouse bone marrow develop in close association with stromal cells which provide essential growth signals. To identify molecules that may normally play a role in this interaction we have examined the in vivo binding of a new monoclonal antibody (mAb) (KMI6) that recognizes a determinant on a bone marrow stromal cell line (BMS2) in vitro. Flow cytometric and radioautographic evaluations revealed that the antigen recognized by KMI6 is represented on the surface of an extremely small number of cells in bone marrow cell suspensions from adult mice. An apparent molecular mass of 110 kD was obtained by surface labeling of a stromal cell clone and immunoprecipitation. Purified mAb KMI6 labeled with 125I was then given intravenously to young C3H/HeJ mice. Unbound mAb was washed out by cardiac perfusion and femoral bone marrow was examined by light and electron microscope radioautography. KMI6 labeling was heavy on the plasma membrane of many stromal cells, especially those located towards the outer subosteal region. The KMI6-labeled stromal cells were usually associated with cells of lymphoid morphology which they often completely surrounded. The labeling was restricted to areas of stromal cell plasma membranes in contact with lymphoid cells. The lymphoid cells themselves, as well as macrophages and other hemopoietic cells, failed to bind mAb KMI6 significantly. Stromal cells in bone marrow depleted of hemopoietic cells by gamma-irradiation (9,5 Gy) bound mAb KMI6 at reduced intensity. The results demonstrate that the KMI6 determinant, a 110-kD protein, is expressed on bone marrow stromal cells in vivo. Its restriction to areas of interaction with lymphoid cells suggests a role in forming microenvironmental niches of B lymphopoiesis. The surface membrane of individual stromal cells may thus be functionally polarized towards interacting B cell precursors and other hemopoietic cells.

The fate of fetal and adult B-cell progenitors grafted into immunodeficient CBA/N mice.

The relative ability of various precursors to generate functional B cells in vivo was assessed by transferring normal, chromosomally-marked CBA/H-T6T6 cells to irradiated or unirradiated immunodeficient CBA/N mice. Emergence of donor-derived B cells was monitored by means of a B-cell cloning assay (in which CBA/N cells are inactive), and by karyotpic analysis of lymphoid, myeloid, and stem cell metaphases. Grafts of lymph node, spleen, anti-mu surface immunoglobin suppressed bone marrow, sIg+ cell-depleted marrow, normal marrow, fetal liver, and yolk sac suggest: (a) there is little self-renewal of sIg+ B cells in these models; (b) pre-committed cells have extensive proliferative/differentiative potential and at least initially contribute most of the newly-formed B cells; (c) populations or pre-B cells obtained from various sources differ in their regenerative ability; (d) CBA/N mice are deficient in a category of pre-B cells which are found in fetal liver; and (e) selective B-cell chimerism results from grafting of unirradiated CBA/N mice.

Regulation of B-lymphocyte clonal proliferation by stimulatory and inhibitory macrophage-derived factors.

A functional subpopulation of murine B lymphocytes proliferate in semisolid agar culture in the presence of 2-mercaptoethanol to form colonies. The effects of diffusible macrophage-derived factors on this focal proliferation was investigated using a two-layer culture system which prevented macrophage-lymphocyte contact and permitted B-cell activation to be critically assessed under conditions of extremely low cell densities. Adherent peritoneal macrophages incorporated within underlayers of spleen or lymph node cell cultures potentiated both the number and size of developing B-cell colonies. These effects were most striking when low numbers of spleen or lymph node cells, or macrophage- depleted lymphoid cell suspensions were used. Thus, macrophage-depleted lymph node ceils gave rise to virtually no colonies, but colony-forming ability was restored by the presence of an optimal number of macrophages. When the number of macrophages exceeded that required for optimal stimulation, colony formation was suppressed; an effect which was largely prevented by indomethacin, an inhibitor of prostaglandin synthesis. Under these conditions, stimulation and inhibition of B-cell activation by macrophages could be dissociated, indicating that each signal is selectively controlled by individual molecules elaborated by the macrophage. With an appropriate number of macrophages required for B-cell activation, and sufficient indomethacin to inhibit the accumulation of macrophage-derived prostaglandin, B-lymphocyte clonal proliferation was a linear function of the number of B cells placed in culture. In the absence of macrophages, B-cell colony formation was potentiated by both lipopolysaccharide and intact sheep erythrocytes through a mechanism different from that of the macrophage-derived stimulatory factor. In addition to their direct stimulatory effect on B-cell proliferation, lipopolysaccharide and sheep erythrocytes were each capable of modulating the production and/or release of B-cell stimulatory and inhibitory factors by the macrophage. Parallel studies of conventional mitogen- stimulated lymphocyte cultures did not show a requirement for macrophages and confirm that the semisolid assay is uniquely suited to studies on the regulatory role of the macrophage in B-cell activation.

Hyaluronate can function as a cell adhesion molecule and CD44 participates in hyaluronate recognition.

A cell adhesion model was previously used to select a series of monoclonal antibodies (mAbs), which were subsequently found to recognize CD44/Pgp-1. Interest in these reagents increased with the finding that they totally inhibited production of lymphoid or myeloid cells in long-term bone marrow cultures. Further investigation has now revealed that hyaluronate is a potential ligand for CD44 and that hyaluronate recognition accounts for the adhesion between B lineage hybridoma and stromal cells. The hybridoma cells adhered to hyaluronate-coated plastic wells as well as to monolayers of stromal cells. The adhesion in both cases was inhibited by treatment with hyaluronidases, and did not require divalent cations. Addition of exogenous hyaluronate also diminished binding of lymphoid cells to stromal cells. One of several mAbs to Pgp-1/CD44 was particularly effective at blocking these interactions. Since hyaluronate and Pgp-1/CD44 were present on both cell types, experiments were done to determine the cellular location of interacting molecules required for the adhesion process. Treatment of lymphoid cells with an anti-Pgp-1/CD44 antibody was more inhibitory than antibody treatment of the stromal cells. Conversely, hyaluronidase treatment of stromal cells reduced subsequent binding more than treatment of the lymphoid cells. Adhesive interactions that involve hyaluronate and CD44 could contribute to a number of cell recognition processes, including ones required for normal lympho-hemopoiesis.

Precursors of murine B lymphocytes. Physical and functional characterization, and distinctions from myeloid stem cells.

The emergence of functional B cells was monitored in irradiated or unirradiated CBA/N recipients of either adult bone marrow or fetal liver from CBA/HT6T6 donors. The cells that are primarily responsible for the generation of B lymphocytes, at least during the first 6 wk, are rapidly sedimenting (4.5-6 mm/h), lack surface immunoglobulin, and are found in both the adult bone marrow and the fetal liver from day 12 onward. These pre-B cells are distinct from the colony-forming unit spleen (CFU-s) as demonstrated by the following criteria: (a) absence from yolk sac (19), (b) lack of correlation between CFU-s number and the ability to generate B cells in fetal liver populations of different ages of gestation, and (c) hybridoma antibodies that significantly inhibited B cell reconstitution but have no effect on CFU-s numbers. The antigen detected by this antiserum is present on both the fetal liver and bone marrow B cell progenitor, although its expression is not restricted to the B lineage. The pre-B cells that we monitor are not homogeneous, however, as both physical and functional differences are found. These observations reinforce our thesis that committed progenitor cells for the humoral immune system are formed early in development and thereafter constitute the major precursor pool for the generation of B lymphocytes.

Beta transforming growth factors are potential regulators of B lymphopoiesis.

Members of the transforming growth factor beta (TGF-beta) family of polypeptides were found to be potent in vitro inhibitors of kappa light chain expression on normal bone marrow-derived and transformed cloned pre-B cells, and of the maturation of these cells to mitogen responsiveness. The inhibition by TGF-beta was selective in that Ia expression was not blocked. Together with the observations that LPS, IL-1, NZB serum factors, IL-4, and IFN-gamma preferentially induced either kappa or Ia, or both, on a pre-B cell line, these results further suggest that acquisition of Ig and class II molecules is independently controlled by different antagonists as well as agonists. In addition, kappa chain induction by IFN-gamma does not appear to be as sensitive to TGF-beta downregulation as that stimulated by other factors tested, and this raises the possibility that activation of the same gene may result from different transmembrane signaling pathways. In contrast to the inhibitory effects of TGF-beta on kappa acquisition by pre-B cells and on kappa increase after exposure of mature B cells to LPS, as measured by kappa RNA levels and/or surface fluorescence, no inhibition was observed on unstimulated spleen B cells or on two cloned B cell lines that constitutively produce kappa. Thus, TGF-beta may function during specific stages of B cell differentiation by inhibiting initiation of, or increased transcription of Ig genes, and therefore, may be an important negative regulator of B lymphopoiesis. It is the first natural substance found to have this effect.

Monoclonal antibodies to Pgp-1/CD44 block lympho-hemopoiesis in long-term bone marrow cultures.

A new panel of mAbs was prepared to a stromal cell line known to support lymphocytes in Whitlock-Witte type long-term bone marrow cultures. These antibodies were then screened with a cell adhesion assay and four were selected that inhibited the binding of B lineage cells to stromal cell monolayers. Immunofluorescent and biochemical analyses revealed that these new antibodies detected epitopes of the previously described Pgp-1/CD44 antigen complex. Addition of Pgp-1/CD44 antibodies to Dexter-type long-term bone marrow cultures completely prevented emergence of myeloid cells and they also blocked lymphocyte growth in Whitlock-Witte type cultures. mAbs MEL-14, LFA-1, and CD45R did not inhibit under the same conditions and there was no apparent relationship to Ig isotype. Adherent layers in treated cultures were not unusual in terms of morphology and the antibodies did not affect factor-dependent replication of lymphoid or myeloid progenitor cells. Therefore, the mechanism of inhibition may not involve direct toxicity to precursors or microenvironmental elements. Previous studies in humans and mice have implicated Pgp-1/CD44-related glycoproteins in the migration of peripheral lymphoid cells, as well as interactions of cells with the extracellular matrix. These findings suggest that they may also be critical for formation of lymphoid and myeloid cells within bone marrow.

Requirements for hyaluronic acid binding by CD44: a role for the cytoplasmic domain and activation by antibody.

The CD44-negative T lymphoma AKR1 (CD44.2 genotype) was transfected with a CD44.1 cDNA. The intact cDNA conferred on the transfected cells the ability to bind hyaluronic acid (HA) both from solution and immobilized on culture plates. It also conferred a CD44-dependent and hyaluronidase-sensitive increase in adhesion to a lymph node endothelial cell line. A mutant cDNA which codes for a CD44 molecule lacking most of the cytoplasmic domain of CD44 was also transfected into AKR1, and cell sorting was used to select transfectants expressing levels of cell surface CD44 expression comparable with the line transfected with the wild-type CD44 cDNA. The cells transfected with the mutant construct bound fluoresceinated HA from solution very poorly, but did adhere to immobilized HA, though less well than cells transfected with the wild-type construct. This result indicates that the cytoplasmic domain of CD44 is necessary for binding of HA from solution but is not required for binding to immobilized HA, although it may contribute to adhesion following ligand recognition. A monoclonal antibody (mAb), IRAWB 14, which reacts with CD44 on all CD44+ cells dramatically induced HA binding by some CD44+ cell lines that did not constitutively bind HA. The transfectant expressing a CD44 molecule with a truncated cytoplasmic domain could be induced by this antibody to bind fluoresceinated-HA from solution. Splenic T cells did not bind fluoresceinated HA constitutively. In the presence of the IRAWB 14 mAb, virtually all CD44+ splenic T cells bound HA. Induction was immediate and occurred equally well at room temperature and at 4 degrees C, indicating that the new HA-binding activity was due to preexistent CD44 molecules. These results are compatible with an antibody-induced activation of CD44 by either a conformational change in the CD44 molecule or a change in the distribution of CD44 molecules on the cell surface.

Age-dependent deficiency of B lymphocyte lineage precursors in NZB mice.

Adult NZB mice (greater than 15 wk old) have very few bone marrow cells that can give rise to sIg+ clonable B cells during liquid culture. This deficiency corresponds to extremely low numbers of cells with cytoplasmic but not surface mu chains of IgM and reduced numbers of cells bearing a high molecular weight B-lineage antigen. Depletion of Thy-1-bearing cells and appropriate mixing experiments did not provide evidence either that suppressor cells are responsible for this phenomenon or that accessory cells are defective in NZB mice. Nor did it seem that B cells were being produced in extramedullary sites. B cell precursors were detectable in very young NZB mice, exceeded control values at 4-5 wk of age, and then declined rapidly. In contrast, these persisted for greater than 1 yr in normal BALB/c, DBA/2, and CBA/H mice. It appears possible that intermediate stages in B-lineage differentiation become prematurely exhausted through an accelerated aging process in NZB mice. These chronological changes have implications for understanding the sequence of events that lead to B lymphocyte formation and the processes that normally regulate it.

Increased B lymphopoiesis in genetically sex steroid-deficient hypogonadal (hpg) mice.

Interleukin 7 (IL-7) responsive B lineage precursors were greatly expanded in genetically hypogonadal female (HPG/Bm-hpg/hpg) mice that have a secondary deficiency in gonadal steroidogenesis. Estrogen replacement in these mice resulted in a dose-dependent reduction in B cell precursors. More modest increases were documented in genetically normal mice that were surgically castrated. These findings complement other recent observations that B lymphopoiesis selectively declines in pregnant or estrogen-treated animals. Sex steroids have long been known to influence such disparate processes as bone physiology and tumor growth, in addition to their importance for reproductive function. We now show that these hormones are important negative regulators of B lymphopoiesis.

Identification of stromal cell products that interact with pre-B cells.

Our understanding of lympho-hematopoietic microenvironments is incomplete, and a new cloning strategy was developed to identify molecules that bind to B lineage lymphocyte precursors. A cell sorting procedure was used for initial enrichment of cDNAs from stromal cell mRNA that contained signal sequences and were therefore likely to encode transmembrane or secreted proteins. A second step involved expression of the library as soluble Ig fusion proteins. Finally, pools representing these proteins were screened for the ability to recognize pre-B cells. This approach resulted in the cloning of biglycan, syndecan 4, collagen type I, clusterin, matrix glycoprotein sc1, osteonectin, and one unknown molecule (designated SIM). The full-length cDNA of SIM revealed that it is a type I transmembrane protein, and its intracellular domain has weak homology with myosin heavy chain and related proteins. Staining of established cell lines and freshly isolated hematopoietic cells with the Ig fusion proteins revealed distinct patterns of reactivity and differential dependence on divalent cations. Biglycan-, sc1-, and SIM-Ig fusion proteins selectively increased interleukin 7-dependent proliferation of pre-B cells. Overexpression of the entire SIM protein affected the morphology of 293T cells, while expression of just the extracellular portion was without effect. Thus, a series of stromal cell surface molecules has been identified that interact with blood cell precursors. Three of them promoted the survival and/or proliferation of pre-B cells in culture, and all merit further study in relation to lympho-hematopoiesis.

A VCAM-like adhesion molecule on murine bone marrow stromal cells mediates binding of lymphocyte precursors in culture.

Two new mAbs (M/K-1 and M/K-2) define an adhesion molecule expressed on stromal cell clones derived from murine bone marrow. The protein is similar in size to a human endothelial cell adhesion molecule known as VCAM-1 or INCAM110. VCAM-1 is expressed on endothelial cells in inflammatory sites and recognized by the integrin VLA-4 expressed on lymphocytes and monocytes. The new stromal cell molecule is a candidate ligand for the VLA-4 expressed on immature B lineage lymphocytes and a possible homologue of human VCAM-1. We now report additional similarities in the distribution, structure, and function of these proteins. The M/K antibodies detected large cells in normal bone marrow, as well as rare cells in other tissues. The antigen was constitutively expressed and functioned as a cell adhesion molecule on cultured murine endothelial cells. It correlated with the presence of mRNA which hybridized to a human VCAM-1 cDNA probe. Partial NH2 terminal amino acid sequencing of the murine protein revealed similarities to VCAM-1 and attachment of human lymphoma cells to murine endothelial cell lines was inhibited by the M/K antibodies. All of these observations suggest that the murine and human cell adhesion proteins may be related. The antibodies selectively interfered with B lymphocyte formation when included in long term bone marrow cultures. Moreover, they caused rapid detachment of lymphocytes from the adherent layer when added to preestablished cultures. The VCAM-like cell adhesion molecule on stromal cells and VLA-4 on lymphocyte precursors may both be important for B lymphocyte formation.