Bacterial, serum and cellular modulation of granulopoietic activity

Mechanisms of neutropenia involving myeloid maturation arrest in burn sepsis

OBJECTIVE

To determine the mechanisms that lead to the decrease in bone marrow production of neutrophils during burn sepsis.

SUMMARY BACKGROUND DATA

Impaired bone marrow granulopoiesis during burn sepsis often results in neutropenia despite elevated circulating levels of granulocyte colony-stimulating factor (G-CSF). To date, neither the specific stages of neutrophil maturation involved in the bone marrow suppression nor the mechanisms for the impairment have been determined.

METHODS

Peripheral blood absolute neutrophil count and G-CSF levels were determined in mice 3 days after randomization to control, burn alone, or burn plus a topical inoculation of Pseudomonas aeruginosa (1000 colony-forming units). Bone marrow aspirates were analyzed for their neutrophil differentiation patterns by Gr-1 antigen expression and their G-CSF receptor status. Histologic analysis of liver, lung, spleen, and wound site was performed.

RESULTS

In burn sepsis, absolute neutrophil count was reduced whereas plasma G-CSF levels were elevated, and myeloid differentiation was significantly shifted toward the immature mitotic myeloid cells. Bone marrow G-CSF receptor mRNA levels and G-CSF-stimulated proliferation were substantially decreased in burn sepsis. Histologic analysis revealed no significant neutrophil infiltration into the tissues.

CONCLUSIONS

In thermal injury with superimposed sepsis, neutropenia and myeloid maturation arrest, despite the elevated levels of G-CSF, correlate with the reduction in bone marrow G-CSF receptor expression. These observations may provide a potential mechanism for neutropenia in sepsis.

Postburn serum inhibits in vitro production of colony-stimulating factor by mononuclear peripheral blood cells

The effects of postburn serum (PBS) on the production of colony-stimulating factor (CSF) was evaluated in 13 burned patients by adding PBS to normal peripheral blood mononuclear cells (MNC) and assaying the MNC-conditioned media for CSF content. PBS inhibited CSF production by at least 50%. PBS from non-survivors significantly inhibited CSF production more than PBS from survivors. The addition of lithium chloride restored production of CSF in the presence of day 15 PBS but could not overcome the inhibitory effects of day 1 or day 8 PBS. The nature of the inhibitor(s) is uncertain, but correction of the CSF production defect by lithium chloride later in the course of thermal injury suggests that the defect may be reversible.

Release of growth promotors from mononuclear human blood cells by LPS, lectins and lithium, as tested in the mouse CFU-C assay

Medium (MCM, 20% human serum), conditioned for 24 h by mononuclear human blood cells, had no colony-forming ability when tested in the mouse CFU-C assay. However, when combined with L-CSF, which predominantly generates macrophage colonies, MCM increased the colony number and size. Even more significant was the increased cellularity with a striking shift towards granulocyte production. Some human sera induced GIF formation without additives, but mostly LPS was required. After heat inactivation, all sera became dependent on LPS to yield an active MCM. Lithium, PHA, Con A and PMW also stimulated GIF formation, which itself depended upon protein synthesis. Heat inactivation of LPS and serum did not reduce their ability, in combination, to yield an active MCM. However, when serum and LPS were mixed before heat treatment, the ability to induce GIF production was abolished, but could be restored by adding intact LPS. This may indicate that LPS exerts its effect by combining with a serum factor yielding a heat-sensitive complex. However, even in the absence of serum, LPS had a stimulatory effect when used in large concentrations, but still the MCM was less active than MCM with serum.

Effects of OKT3+, OKT4+ and OKT8+ T cell subsets on steady-state granulopoiesis in vitro

The present study was undertaken to elucidate the role of T cell subsets in regulating the in vitro growth of human granulopoietic progenitor cells (CFU-C). Prior to CFU-C assay, mononuclear cells obtained from bone marrow and cord blood were depleted of T cells or functionally distinct T cell subsets by the method of complement-mediated cytolysis with the use of monoclonal antibodies, OKT3, OKT4 and OKT8. The depletion of any T cell subset of OKT3+, OKT4+ or OKT8+ cells from bone marrow and cord blood cells showed no significant alterations in the generation of CFU-C. The supplementation to the in vitro culture system of OKT4+ or OKT8+ cells, which had been negatively selected by complement-mediated cytolysis using the mutually exclusive monoclonal OKT8 or OKT4 antibody, respectively, did not alter the growth of CFU-C-derived colonies without mitogenic stimulation. In contrast, the production of colony-stimulating activity (CSA) in peripheral blood mononuclear cells was significantly reduced by removing not only OKT3+ cells but also OKT4+ or OKT8+ cell subsets. There were no significant differences in the degree of reduction between the different procedures. These results suggested that T cell subsets played an important role in the regulation of steady-state granulopoiesis through the stimulation of CSA production. The presence of a specific subset of T cells in CSA production was not demonstrated.

Schistosomal stimulation of eosinophil production by human bone marrow in vitro

These studies were undertaken to determine if schistosomal and ragweed antigens incubated with human peripheral blood mononuclear cells (MNC) would stimulate the production of an eosinophil colony stimulating factor (CSF) which was active on human bone marrow. These studies have shown that schistosoma mansoni antigen in a concentration of 5 micrograms/ml incubated with MNC's from unsensitized individuals lead to production of a conditioned medium which did not stimulate an increased number of total colonies but did increase the number of eosinophil colonies (44 to 61 colonies). Ragweed antigen also stimulated increased eosinophil CSF (from 44 to 63 colonies). The granulocyte response to parasitic antigens almost invariably involves the eosinophil system. Eosinophils play a vital role in the control of these diseases in humans. Further understanding of the interaction of parasitic antigens with the granulopoietic system will hopefully give new and important clues as to how such parasites interact with man and how they are controlled.

Migration inhibitory action of bacterial lipopolysaccharide on progenitor cells of monocyte/macrophage lineage growing in culture in the presence of colony-stimulating factor (CSF-1)

Addition of lipopolysaccharide (LPS) to the culture of mouse myeloid stem cells (CFUc) increased the incidence of compact colonies and decreased that of dispersed ones in the presence of colony-stimulating factor (CSF-1) which had not such an effect by itself even in high concentrations. Although colony morphology was thus changed, nearly all colonies were composed of monocytes. The incidence of compact colonies increased with the increase of LPS concentration but plateaued at about 50%. Bone marrow cells of LPS-tolerant mice responded to LPS in vitro to a slightly decreased extent. The activity of LPS was decreased by alkaline or acid hydrolysis of the LPS molecule and inhibited by polymixin B, but not by indomethacin, alpha-L-fucose, nor by alpha-methyl-D-mannoside. Other immunopotentiating substances, such as OK-432, Lentinan, and Levamisole, had no effect on the colony morphology. Both muramyl dipeptide and poly(I).poly(C) were also ineffective. Furthermore, the action of LPS was not abolished by the use of heat-inactivated serum in the culture. LPS was no longer stimulative for the induction of lysosomal enzymes in the CSF-stimulated culture, although it greatly enhanced the enzyme induction in the unstimulated culture. These results indicate that the cells of monocyte/macrophage lineage develop the capacity for migration before they become responsive to LPS, and that the LPS-responding monocytic cells can proliferate even in a state of confluence induced by LPS.

Regulation of granulopoiesis and distribution of granulocytes in early phase of bacterial infection

Studies have been carried out to determine the effect of bacterial infection on CSF production, CFU-C activation, and bacterial clearance by mature granulocytes in mice infected with Escherichia coli. These studies have shown that immediately after bacterial infection (5 minutes), serum colony-stimulating factor (CSF) levels and bone marrow colony-forming units in culture (CFU-C) levels are elevated. This is followed by oscillatory rises in both of these parameters and the appearance of granulocytes in the infected site. With clearance of bacteria, CSF and CFU-C levels return to normal. These studies have indicated further that bacterial infection is a major stimulus for granulocyte production through the CSF-CFU-C system and that clearance of bacteria by mature granulocytes may serve as a negative feedback regulatory arm.

Autologous neutrophils inhibit production of colony stimulating activity by normal human lymphocytes

Experiments were designed to study the way in which normal human polymorphs (PMNs) inhibit the production or release of colony stimulating activity (CSA) by normal lymphocytes incubated alone or in conjunction with normal monocytes. PMNs were first incubated with lymphocytes for various periods at various concentrations. The PMNs were then removed and the 'conditioned' lymphocytes were used alone or after addition to adherent mononuclear cells (monocytes) for the production of conditioned medium. The samples of conditioned medium were then assayed for CSA in a standard system for culturing granulocyte/macrophage progenitor cells (CFU-C) in agar. We found that the capacity of PMNs to inhibit CSA production by lymphocytes or by lymphocytes plus monocytes was directly proportional to the number of PMNs originally incubated and maximal at relatively short incubation times (i.e. 2-4 h). Such inhibition could be counteracted by the introduction of known stimulators of CSA production by monocytes, e.g. phytohaemagglutinin or bacterial toxin. We conclude that normal lymphocytes 'conditioned' or 'programmed' by contact with PMNs may themselves have a reduced capacity to produce CSA and may also act to reduce CSA production by monocytes. Such programmed lymphocytes could thus be a component of the mechanism by which PMNs exert a physiological inhibition on CSA-dependent granulopoiesis in vivo.

Control factors of granulopoiesis in human serum

The role of serum factors in the modulation of production of colony-stimulating activity (CSA) has been investigated. A factor has been described, and partially characterized, in human serum that has the capacity to stimulate increased synthesis and release of CSA by human mononuclear cells (MNC). MNC RNA and protein synthesis are required to demonstrate this effect of serum, but DNA synthesis and mitotic division are not required. The factor in serum resulting in this effect is a heat-labile protein with a molecular weight slightly greater than that of CSA.

Bacterial stimulation and granulocyte inhibition of granulopoietic factor production

We attempted to determine the effect of live bacteria (Staphylococcus epidermidis) on granulocyte colony-stimulating-factor production by human peripheral blood mononuclear cells (monocytes and lymphocytes) in vitro. Addition of bacteria to mononuclear-cell cultures enhanced colony-stimulating-factor production by these cells, as assayed on both human and mouse bone marrow. Addition of peripheral blood granulocytes to parallel cultures eliminated this enhancement effect, presumably by bacterial removal or inactivation. These data suggest that micro-organisms may have a pivotal role in granulocyte production and maturation by serving as a stimulus to increase colony-stimulating-factor production and also as negative control through their removal by the newly formed granulocytes.