Search

The occurrence of anemia in older adults has been associated with adverse outcomes including functional decline, disability, morbidity, and mortality. It is not clear to what extent these outcomes are the result of the anemia or concurrent illness. We performed a cross-sectional, observational study to determine whether lower hemoglobin concentrations in older adults are associated with reduced health-related quality of life, functional status, depression, disability, and physical strength, independent of chronic disease. Three sites participated in this research: an academic geriatric practice, a hospital-based geriatric outpatient unit, and a community-based multispecialty internal medicine group. Health-related quality of life and functional status were measured using the Short Form-36 Health Survey (SF-36) and the Functional Assessment of Chronic Illness Therapy-Anemia (FACIT-An). Disability and depression were assessed using the Instrumental Activities of Daily Living (IADL) and the Geriatric Depression Scale (GDS) questionnaires, respectively. Handgrip strength was used as a physical performance measure. Anemia was defined as hemoglobin

Clonality assays, based on X-chromosome inactivation, discriminate active from inactive alleles. Skewing of X-chromosome allelic usage, based on preferential methylation of one of the HUMARA alleles, was reported as evidence of clonal hematopoiesis in approximately 30% of elderly women. Using a quantitative, transcriptionally based clonality assay, we reported X-chromosome–transcribed allelic ratio in blood cells of healthy women consistent with random X-inactivation of 8 embryonic hematopoietic stem cells. Furthermore, we did not detect clonal hematopoiesis in more than 200 healthy nonelderly women. In view of the susceptibility of aging hematopoietic stem cells to epigenetic dysregulation, we reinvestigated the issue of clonality in elderly women. Forty healthy women (ages 65-92 years; mean, 81.3 years) were tested by a novel, quantitative polymerase chain reaction (qPCR) transcriptional clonality assay. We did not detect clonal hematopoiesis in any of the tested subjects. We also tested DNA from the same granulocyte samples using the methylation-based HUMARA assay, and confirmed previous reports of approximately 30% extensively skewed or monoallelic methylation, in agreement with likely age-related deregulated methylation of the HUMARA gene locus. We conclude that the transcriptionally based X-chromosome clonality assays are suitable for evaluation of clonal hematopoiesis in elderly women.

Impaired hematopoiesis and dysregulated cytokine expression have important implications for cancer in the elderly. In aged people, hematopoiesis is dysregulated and becomes paradoxically down-modulated under periods of increased hematopoietic demand. This down-modulation may explain, at least in part, the increased incidence of anemia in the elderly, although the cause of anemia can usually be identified in these patients and frequently reversed with targeted therapy. An age-associated decrease in the expression of interleukin-2 may contribute to impaired cellular immunity. Additionally, the increased interleukin-6 production frequently found in the elderly may participate in promoting the survival and proliferation of malignant myeloma and in inducing resistance by myeloma cells to therapies that act through apoptosis. Dysregulation of the expression of these and other cytokines may be a mechanism contributing to age-related impairment of the hematopoietic response, the genesis and therapeutic resistance of specific malignancies, and cancer cachexia.

In seven patients with bronchopulmonary dysplasia and anemia, we evaluated the mechanisms causing the anemia. All had a normocytic, normochromic, hyporegenerative anemia (mean hematocrit 26%; range 21% to 30%). The low hematocrit values seemed physiologically significant because mean (+/- SD), heart rates fell after transfusion (162 +/- 7 to 149 +/- 9 beats/min; p less than 0.005), as did blood lactate concentrations (1.2 +/- 0.3 mumol/gm blood before vs 0.5 +/- 0.3 after transfusion; p less than 0.05). Anemia could not be explained by blood withdrawal or deficiency of vitamin E, folate, or iron. No dyserthropoietic or megaloblastic changes were observed. No erythroid regenerative response was seen in the marrow; however, when recombinant erythropoietic growth factors were added to marrow cells in tissue culture, erythroid cell growth in vitro was normal. In contrast to patients with the "anemia of chronic disorders," these patients had a normal or increased number of marrow sideroblasts and increased serum transferrin saturation. Serum concentrations of erythropoietin were low for patients with anemia (range 11.4 to 47.1 mU/ml); yet the in vitro sensitivity of bone marrow erythroid progenitors (colony-forming units--erythroid) to recombinant erythropoietin was increased (p less than 0.001). We conclude that the anemia in these patients was the result of deficient production of erythropoietin, and we speculate that administration of recombinant erythropoietin would correct the anemia.

A transient hyporegenerative neutropenia has been reported in neonates, but not in older children or adults, undergoing treatment with recombinant erythropoietin (epo). Monocytopenia has not been reported. We postulated that epo might selectively reduce the responsiveness of neonatal progenitors to Granulocyte Colony-Stimulating Factor (G-CSF), while not similarly affecting their responsiveness to Macrophage Colony-Stimulating Factor (M-CSF). To test this hypothesis two types of experiments were performed. First, progenitors of adult or fetal origin were pre-incubated with epo (or control), then washed, and their responsiveness to G-CSF and M-CSF evaluated in clonogenic culture assays. Second, clonogenic maturation was initiated using either G-CSF or M-CSF, after which the effect of a late addition of epo to the developing clones was evaluated. Indeed, pre-incubation with epo resulted in production of fewer neutrophils from fetal progenitors grown in G-CSF (P less than 0.001), but it did not reduce the number of macrophages generated from progenitors grown in M-CSF. Adding epo to the already-developing G-CSF-responsive and M-CSF-responsive adult and fetal clones did not alter colony development. Thus, epo appears to have an action on G-CSF-responsive, but not-M-CSF-responsive fetal progenitors, resulting in reduced production of neutrophils. This effect is no longer apparent, however, when progenitors have matured to the 8-cell clone stage.

Normal hematopoiesis constitutes the process of producing diverse, differentiated blood cell types in a manner related to physiological requirement. During aging, modulation of hematopoiesis becomes disordered, impairing the ability of older people to respond appropriately to the physiological demand for blood cell replacement triggered by stimuli such as blood loss or cytoreductive chemotherapy. This may contribute to the increase in the prevalence of anemia that is observed during aging. In addition, various age-related events, such as genomic mutations secondary to oxidative stress and impaired regulation of cytokine production, may contribute to or cause the emergence of abnormal clones of hematopoietic cells. Therefore, normal hematopoiesis is disrupted, and the hematopoietic system is populated with cells that are quantitatively and functionally deficient and are also subject to leukemic transformation. These defects in the production and maturation of the various differentiated blood cells are referred to as myelodysplastic syndromes. These syndromes are so tightly associated with aging that they are considered to be geriatric disorders; they can lead to anemia, neutropenia, and thrombocytopenia and to the development of acute nonlymphoblastic leukemia. Dysregulation of mechanisms controlling hematopoiesis is therefore an important characteristic of the hematopoietic system in the elderly, but the response of progenitor cells to humoral stimulators is preserved and accounts for the effectiveness of recombinant hematopoietic growth factors used as emerging treatment modalities for hematopoietic disorders in the elderly.

Aged humans and aged experimental animals exhibit a diminished ability to upregulate hematopoiesis, but the mechanism responsible for this is unknown. The purpose of the studies was to test the hypothesis that this disregulation might be attributable to altered expression of hematopoietic growth factors. We studied the in vitro ability of cells from aged humans or mice to release bioactive colony stimulating activity (CSA) and to accumulate mRNA for defined growth factors. Human light density leukocytes from the aged released less CSA than cells from the young. GM-CSF accounted for 72–100% of the CSA released by young humans, as judged by antibody inhibition studies. In contrast, GM-CSF accounted for 0–42% of CSA from aged humans. The observation of diminished expression of GM-CSF by cells from aged humans was further accompanied by a reduced accumulation of mRNA for GM-CSF. In other in vitro studies, splenic cells from aged mice released less CSA than cells from young mice, and in vivo experiments showed that splenic cells from mice challenged with E. coli accumulated less mRNA for M-CSF than young mice. These data demonstrate a similar defect in the expression of hematopoietic growth factors in aged man and mice. We propose that this diminished production of bioactive and identifiable hematopoietic growth factors may be mechanistically important in explaining the disordered hematopoiesis of aging.

Developmental delays, which impair antibacterial host defense, are present in the neutrophil system of human preterm neonates. We hypothesized that diminished production of interleukin-8 (IL-8), a neutrophil chemotactic peptide, might in part be responsible for these defects. To test this, monocytes from the blood of preterm neonates, term neonates, and adults were isolated immunologically by “negative panning” and subsequently stimulated with interleukin-1α (IL-lα), tumor necrosis factor a (TNF-α), or lipopolysaccharide (LPS), after which IL-8 levels in the supernatants were measured by ELISA. Total cellular RNA was extracted and IL-8 mRNA was assessed by Northern blotting and by competitive polymerase chain reaction (PCR) analyses. After stimulation with IL-lα, IL-8 accumulation was lowest in supernatants of monocytes from preterm neonates, intermediate in supernatants of monocytes from term neonates and greatest from monocytes of adults. Similarly, when stimulated with TNF-α or LPS, monocytes from preterm neonates produced less IL-8 than cells from term neonates and adults. These differences in IL-8 concentrations paralleled differences in IL-8 mRNA expression. J. Leukoc. Biol. 53: 399–403; 1993.

Several deficiencies in antibacterial defense have been described in neonates. Among those best characterized are delayed maturation of B cells into antibody producing cells, deficient T-cell maturation, and delayed cycling of hematopoietic progenitor cells after an infectious challenge. No unifying theory has been forwarded, however, to explain the concomitance of these three developmental deficiencies. IL-6, a cytokine produced primarily by monocytes and macrophages in response to stimulation by IL-1, is involved in the regulation of these three processes. Thus, we postulated that defective production of IL-6 could be a mechanism underlying these immune deficiencies of neonates. Indeed, we observed that a peak production, cells of five term neonates produced only one half as much IL-6 (14 120 +/- 2590 pg IL-6/10(6) monocytes) as those of five adults (28 940 +/- 1680 pg, p less than 0.001). Peak production was lower still by monocytes of six preterm neonates (7190 +/- 1400 pg, p less than 0.001 versus term). Production of IL-6 protein was inhibited by actinomycin D and the IL-6 mRNA content of monocytes from neonates, as assessed by competitive polymerase chain reaction, was less than that of adult monocytes. We speculate that defective IL-6 transcription might underlie some of the defects in immune regulation observed in neonates.

Congenital hypoplastic (Diamond-Blackfan) anemia is a rare macrocytic anemia, generally presenting during infancy or childhood. The condition usually occurs sporadically or in a pattern consistent with autosomal recessive inheritance, although autosomal dominant transmission has been proposed in some kindreds. We report the largest known kindred of congenital hypoplastic anemia, with at least 7 affected individuals over 3 generations, and propose that studies of this kindred may be useful for identifying the mechanism by which their genetic abnormality results in congenital hypoplastic anemia. Erythropoietic investigations on relatives show no inhibitors of erythropoiesis in serum, T-lymphocytes, or macrophages. Their erythroid progenitor cells (CFU-E and BFU-E) were generally quantitatively normal, and were capable of rapid proliferation, as judged by cell-cycle shortening. However, their erythroid progenitors displayed a relative insensitivity to recombinant erythropoietin, and produced relatively few normoblasts per erythroid progenitor cell. We propose that these and subsequent studies may be helpful in selecting candidate genes responsible for the molecular defect in this kindred.

Clonality studies can establish the single cell origin of tumors and differentiate nonmalignant from malignant states. Detection of clonal cells may be genotype-based relying on somatic mutations to mark the clonal population (e.g. 9q+:22q– translocation in CML), or phenotype-based, where the clonal population is identified by expression of surrogate genes which facilitate tracking the clonal process. Methods for determining phenotypic clonality rely on the principle of X chromosome inactivation (XCIP), unique to women, and are based on differentiating transcriptionally active from inactive X-chromosomal genes. Detection of the polymorphic state of genes subjected to inactivation may be done by either: discrimination of DNA methylation status, detection of mRNA transcripts, or polymorphic isozyme protein products. Extreme skewing of X-chromosome allelic usage by methylation-based clonality assay has been reported in ∼30% of healthy elderly females precluding clonality studies in this population. In contrast, by X-chromosome quantitative transcriptional clonality assay (TCA), we previously reported a normal skewing range based on our determination of random X-chromosome inactivation in 8 progenitors of pluripotent hematopoietic stem cells at the time of inactivation during the blastocyst stage of development (J Exp Med, 183:748, 1996). Moreover, we have not observed clonal XCIP in TCA studies involving over 200 healthy heterozygous females, indicating the rarity of this phenomenon. However, we did not systematically study females >65 years old. Furthermore, our TCA protocol was laborious, technically demanding and required significant amounts of highly radioactive isotopes. In addition, due to susceptibility of DNA methylation to environmental factors we decided to re-investigate the issue of clonality in older females using a novel quantitative real time PCR assay based on a unique primer design, we previously reported for the JAK2V617F mutation (Exp Hematol. 2007;35(1):32–8). Females >65 years of age with no history of malignant disorders, unexplained anemia, or autoimmune disorders were recruited for our IRB approved study. Genomic DNA and total RNA was isolated from peripheral blood granulocytes, reticulocytes, and platelets where applicable. Clonality studies were performed using BTK, FHL1, IDS, G6PD, and MPP1 exonic polymorphisms (∼95% females are informative for at least one marker; Blood101:3294–301, 2003). Genomic DNA was used for genotyping exonic polymorphisms by TaqMan based allelic-discrimination assays. Thirteen elderly females (age range 65–92, mean 75.5, median 75) and 5 younger females (age range 30–40, median 36, mean 35), heterozygous for one or more markers, were identified. TCA was performed using total RNA on markers found to be informative. Neither clonal XCIP, nor extreme skewing XCIP was noted in any of the study subjects. Based on reported data, ∼30% elderly women were found to have extremely skewed XCIP; hence, we expected to find 4/13 elderly women with clonal XCIP in our study group. Statistical analysis, using an exact binomial test, indicates a low probability of false positive results by our assay (p=0.014, exact 95% CI [0,0.22]). In conclusion, hematopoiesis is not clonal in healthy elderly females. #equal contribution.

The mechanisms responsible for the unusual susceptibility of multiple myeloma (MM) patients to infections are incompletely defined. Since MM is associated with decreased production of normal serum proteins, we investigated the possibility that the production of opsonins might also be impaired. The neutrophil chemiluminescence assay of opsonization was used to evaluate the ability of serum from patients with MM to opsonize zymosan. It was found that sera from 18 MM patients exerted only 50% +/- 2.5% (mean +/- SEM) of the opsonic activity found in 18 control sera (p less than 0.001). In mixture experiments, untreated normal serum completely restored the opsonic activity of MM serum, suggesting a deficiency of opsonic factors rather than an inhibitor. In other mixture experiments, heat-inactivated normal serum only partially corrected the opsonic defect in MM serum. Serum from three patients had low C3 levels, and treatment of particles with these resulted in a greater opsonic defect than the patient population as a whole (p less than 0.02). No correlation between the opsonic defect and infections was established over an 18-mo period. These data suggest that MM serum lacks both heat-stable and heat-labile opsonic activity, the direct clinical significance of which remains to be clarified. However, these studies support the concept that defective host resistance in MM may be multifactoral, combining opsonic abnormalities with other defects previously described.

Neonates are particularly susceptible to infection with Escherichia coli K1. To investigate the mechanisms which lead to this susceptibility, we examined: (a) the bactericidal activity of neutrophils; (b) opsonization, and (c) the bactericidal activity of serum in developing rats. Neutrophils from adult rats killed E. coli K1 more efficiently than did neutrophils from young animals. Opsonization of E. coli K1 by serum of prematurely delivered rats was poor. Serum from prematurely delivered and term rats promoted growth of E. coli K1, while serum from adult rats killed > 95% of the organisms within 90 min. However, the mixture of heat-inactivated serum from adult rats plus serum from prematurely delivered rats killed E. coli K1.

During a three-year period, 26 neonates with bacterial infection and neutropenia were studied. In order to assess the marrow neutrophil reserves, bone marrow aspirates were obtained from each of these patients. The neutrophil storage pool (percent polymorphonuclear + band neutrophils + metamyelocytes in 1,000 nucleated marrow cells) was significantly greater in those who survived their infection (X= 20.1%, range 3.2% to 60.8%) than in those who died (X = 1.9%, range 0.4% to 5.2%, P < .002). In an attempt to improve survival in this group, seven neutrophil-depleted patients with sepsis were given granulocyte transfusions and all survived. In contrast only one of nine nontransfused, neutrophil-depleted infants with sepsis survived (P

This paper describes two cases of nafcillin-induced platelet dysfunction, with positive rechallenge data for one patient. Nafcillin resulted in abnormal bleeding times in both patients and a clinically apparent bleeding episode in one of the cases. Platelet function tests were performed on one patient during the initial therapy and after rechallenge with nafcillin. Platelet aggregation showed abnormal responses to ADP, collagen, and epinephrine. Platelet count and morphology were normal. Nafcillin should be recognized as another antibiotic which causes platelet function abnormalities and clinical bleeding episodes.

A modified form of serum immunoglobulin G (pH 4.25) was tested for its effect on neutrophil kinetics and survival rates in neonatal rats with type III, group B streptococcal pneumonia and sepsis. Each of 30 animals received a transthoracic inoculation of 10(5) organisms/g of body weight; all died within 48 hr. When 100, 1,000, or 2,000 mg of immunoglobulin G/kg was administered intraperitoneally at the time of bacterial inoculation, survival rates rose to 20%, 90%, and 100%, respectively. Even when the immunoglobulin preparation was administered intraperitoneally 2 hr after transthoracic inoculation of bacteria, all 19 animals survived. Only seven of 15 animals survived when immunoglobulin administration was delayed for 22 hr. Immunoglobulin facilitated the neutrophil inflammatory response: when immunoglobulin (rather than an albumin control) was administered, neutrophils were released more rapidly from the storage pool and accumulated more quickly at the site of bacterial inoculation. Unlike infected control animals, immunoglobulin recipients did not develop neutropenia or depletion of the neutrophil storage pool.

We used cells from marrow aspirations that had been performed on 10 infants with the "anemia of prematurity" and tested the responsiveness of their erythroid colony-forming units (CFU-E) to recombinant human erythropoietin. For comparison, we also tested marrow-derived CFU-E from five healthy adults, and circulating CFU-E from cord blood of five healthy neonates. CFU-E from the anemic infants had a 50% maximal response at 0.073 +/- 0.024 U erythropoietin per milliliter (mean +/- SD). They were therefore at least as responsive as were CFU-E from adults, which displayed a 50% maximal response at 0.118 +/- 0.076 U/ml, and as were circulating CFU-E of cord blood origin, which had a 50% maximal response at 0.109 +/- 0.047 U/ml. Because CFU-E from infants with the "anemia of prematurity" appeared highly sensitive to erythropoietin in vitro, we propose that its administration to these patients would likely result in a significant increase in erythrocyte production in vivo.

Blood and Marrow Neutrophils during Experimental Group B Streptococcal Infection: Quantification of the Stem Cell, Proliferative, Storage and Circulating Pools

We explored the effect of in vitro phagocytosis and in vivo inflammation on the MPO content of functioning neutrophils and on the ability of these cells to export active MPO into the extracellular environment. After ingestion of staphylococci, neutrophils retained 52% of their MPO and released 8% into the medium in active form; the remaining 40% of their MPO could no longer be detected. During bacterial infection induced by intradermally injecting staphylococci, neutrophils harvested from minced infected lesions contained 52% of the MPO of circulating neutrophils that had not reached the lesions. Extracellular fluid from the lesions contained active MPO secreted by the neutrophils, and concentrations of 10–45 U/ml were detected. These data demonstrate that functioning neutrophils can lose approximately half of their MPO. In vitro, 4%-8% of neutrophilic MPO appears in the extracellular space and 40% is inactivated. In vivo, the MPO content of inflammatory neutrophils also decreases, and MPO appears in the extracellular fluid in active form where it is available to participate in a variety of physiologic processes.

Monoclonal antibody(AB) may eventually have a role in the treatment of life-threatening infection. In our previous study, mortality fell from 90% to 0% when neonatal rats, inoculated with Group B Streptococci (GBS), were injected with AB. However, AB was administered simultaneously with GBS and at the same site, unlike the situation which would exist clinically. In the present study, we gave AB intraperitoneally 2 hrs after transthoracic GBS, & 76 of 76 lived. When AB was delayed for 4, 5 or 6 hrs, survival was 92% (11/12), 60% (15/25) and 29% (5/17). When delayed for 7, 8, or 9 hrs, none of 39 lived. Failure of AB to protect after 6 hrs coincided temporally with depletion of the animals' neutrophil storage pool (NSP=PMN+bands+metamylocytes in the marrow+liver+spleen, 12.0 1.2×106/gm control vs 1.5±2×106/gm infected, X±SD, p

Groups of newborn rats were transthoracically inoculated with 1 X 10(6) type III group B streptococci/g body wt, either alone or in combination with 1.5 microgram/g body wt of type-specific antibody derived from hybridoma cell lines. Ninety-four percent of the animals who received bacteria alone died. In contrast, none of those treated with antibody died (P less than 0.005). Kinetic studies suggested that antibody may have offered protection, In part, by facilitating the neutrophil response. Animals who received only bacteria exhibited a marked neutropenia (20 +/- 18/mm3, mean +/- S.E.M.) whereas infected animals treated with antibody did not (3800 +/- 30/mm3, P less than 0.001). Furthermore, within 2 h of inoculation, antibody-treated animals mobilized and stored neutrophils, whereas significant neutrophil mobilization did not occur in the animals which received bacteria alone until 6 h. In the animals receiving bacteria alone, exhaustion of the neutrophil supply quickly occurred (remaining storage neutrophils at 6 h, 0.2 +/- 0.1 X 10(6) cells). In contrast, animals, which received antibody, maintained an adequate supply of stored neutrophils (7.0 +/- 0.4 X 10(6) P less than 0.001). The migration of neutrophils to the site of inoculation was measured by assaying the lungs' content of myeloperoxidase, a marker enzyme for granulocytes. The right and left lungs of animals not receiving antibody accumulated the same quantity of neutrophils, with peak pulmonary neutrophil accumulation occurring 6 h after the infection. In antibody recipients, however, the inoculated lung accumulated significantly more neutrophils than the opposite lung and peak pulmonary neutrophil accumulation occurred at 2 rather than 6 h.

The association of eosinophilia and Löffler’s endomyocardial fibrosis with acute lymphoblastic leukemia in a 9-yr-old boy is reported. During the illness, there were three leukemic relapses, each preceded by eosinophilia which subsided upon induction of remission. It is suggested that the eosinophilia in this case was not "eosinophilic leukemia," but more likely a nonneoplastic response to an antigenic stimulus from the leukemic process.

Clinical observations during infection suggest that in aged patients, the kinetic or proliferative responses of neutrophils to infection may be deranged. To test this hypothesis, the neutrophil responses of 6- month-old and 30-month-old mice were compared. After intrapulmonary injection of Escherichia coli, young mice exhibited neutrophilia and diminution of the neutrophil storage pool (NSP) by a mean of 6.4 x 10(6) neutrophils/two femurs. This was accompanied by an increase in the pool of CFU-GM from a control value of 1.1 x 10(5) cells/two femurs (range 0.7 to 1.4) to 1.5 x 10(5) (1.1 to 1.9) (P less than .05) and the thymidine suicide (relative proliferative rate) of CFU-GM rose from 27% (19 to 42) to 51% (31 to 61) (P less than .05). Furthermore, the CFU-GM of infected young mice displayed enhanced differentiation to the neutrophil series. In contrast, old mice exhibited a greater mean diminution of the NSP: 12.8 x 10(6) neutrophils. Also, old mice experienced a reduction in CFU-GM from 2.3 x 10(5) (1.0 to 3.9) (controls) to 1.3 x 10(5) (1.2 to 1.5)/two femurs (P less than .05), a reduction in the proliferation of CFU-GM and reduced differentiation of CFU-GM to neutrophils. These experiments establish that the neutrophil response of infected old mice is disordered, with exaggerated depletion of the NSP and lack of stimulus-driven granulocytopoiesis as reflected by a paradoxical reduction in the number and proliferative rate of precursors. This defect may be compounded by decreased differentiation of precursors to neutrophils.

In clonogenic assays of hematopoietic progenitors, high concentrations (4 U/mL) of erythropoietin (epo) reduced the formation of granulocyte- macrophage (GM) colonies and diminished the number of granulocytes formed per culture plate. Fetal progenitors were more sensitive to these effects of epo than were progenitors from adults, displaying these reductions at greater than or equal to 1 U epo/mL. The mechanism was investigated by growing fetal progenitors stimulated by recombinant GM-CSF, in the absence of epo, and when eight-cell clones first appeared, mapping their location, then adding epo, and assessing its effect on the subsequent differentiation of the clones. In the absence of epo, the clones developed exclusively into GM colonies. However, if developing clones were presented with epo, 85% matured into GM colonies, but 15% became multilineage or normoblast colonies. In addition, developing clones that were presented with epo produced colonies that contained fewer neutrophils. These effects of epo on neutrophil generation were observed with each of three varieties of recombinant epo, and also with purified human epo, but were not observed using epo that had been neutralized with rabbit anti-epo antiserum.

A high rate of mortality is observed in the subset of neonates with early-onset bacterial sepsis who develop profound neutropenia and depletion of the marrow neutrophil reserve. Several studies suggest that transfusion of apheresis-collected adult neutrophils into such subjects can be beneficial, yet inherent delays in preparation of neutrophil concentrates can limit their clinical utility. In six neonates with sepsis and profound neutropenia, we performed double-volume exchange transfusions using freshly drawn donor blood and determined that a net gain of 0.3-0.7 X 10(9) neutrophils/kg body weight had occurred. The blood neutrophil count of the patients increased from 220 to 1,660/mm3 before the transfusion to 1,650-5,140/mm3 immediately afterward. We next attempted to further increase the number of neutrophils delivered by exchange transfusion by subjecting blood donors to 2 min of physical exercise in order to induce neutrophilia. After such exercise, 30% (range 18-55%) more neutrophils were harvested. The metabolic burst activity and in vitro migration of neutrophils after exercise did not differ from paired preexercise values.

The hematopoietic growth factor CSF-1 has been considered relatively lineage specific for the production of macrophages, whereas GM-CSF elicits a predominance of neutrophils. It is likely that in vivo, individual clones are stimulated by the two CSFs, although the effect of dual stimulation on progenitors and their progeny has not been completely explored. We found that in cultures initiated with low concentrations of CSF-1 or GM-CSF, alone or in combination, production of macrophages predominated. Maximally stimulatory concentrations of CSF-1 elicited a predominance of macrophages, whereas maximal GM-CSF elicited many more neutrophil/macrophage colonies and pure neutrophil colonies. A combination of maximal CSF-1 and GM-CSF elicited the same differentiation as GM-CSF alone. Delayed addition of GM-CSF to cultures initiated with CSF-1 elicited colonies indistinguishable from GM-CSF alone, suggesting that neutrophil production had been switched on by GM- CSF. In mapping studies, colonies initiated by CSF-1 increased or switched on neutrophil production when GM-CSF was added as a second stimulus. These studies show that individual clones are responsive to both CSFs, and that the differentiating influence of GM-CSF predominates over that of CSF-1. In cultures to which only CSF-1 was added, a population of progenitors was sustained that produced neutrophils only after a GM-CSF stimulus. Thus, CSF-1 may participate in maintaining a reserve of progenitors for neutrophils during periods of increased neutrophil demand.

Changes in blood neutrophil (neut) number and in degree of left shift are used as diagnostic clues when neonatal sepsis is suspected. However, we report a pitfall in reliance upon CBC changes. We observed 4 cases of fatal group B streptococcal (GBS) sepsis with a normal neut count (mean 7730;range 6140-9610) and a normal immature|total(i/t) neut ratio (0.04;0.00-0.09). In 2 cases this was used as a reason to withhold antibiotics. In 2, the CBC was repeated prior to death (0 and 150|mm3). In an animal model of lethal GBS infection, we determined the length of the "latent period" between bacterial inoculation and CBC changes. Neonatal and adult rats received GBS after which blood and marrow were serially examined. The neonates died by 24 hours, but no change in CBC occurred until after 4 hours when an increase in i|t ratio (0.40 ± 0.05, control vs 0.58±0.3 infected, mean±SEM, p

Neutrophil release and migration in mice were studied over a 24-hr period after the sc implantation of a single polyvinyl sponge. The release of neutrophils from the marrow was evaluated by directly counting the residual neutrophils in the femoral marrow of animals with sponges. Sponge and tissue neutrophil content was determined by extraction and assay of myeloperoxidase (MPO), a marker enzyme for neutrophils. A maximum depletion of 48% of the mature neutrophils in the marrow was observed 5 hr after sponge implantation, in keeping with significant release of neutrophils for migration to the sponge. The released cells were not found in the circulating granulocyte pool, since neutropenia was noted. The accumulation of neutrophils in the sponge increased throughout the 24-hr period, whereas in the tissue adjacent to the sponge maximum accumulation of neutrophils occurred within 7 hr. In fact, neutrophils migrated to at least three sites--the sponge, the skin overlying the sponge, and the skin in which an incision had been made to insert the sponge. The sponge content of neutrophils represented 0.3-33% of the neutrophils migrating to the combined lesion (sponge and skin sites). Therefore, if the neutrophil response to foreign body implantation is to be measured in its entirety, it is necessary to quantify not only the neutrophils within the foreign body but also those in the tissues surrounding it. These studies describe an animal model for neutrophil release and migration to tissues following a standard stimulus. It is proposed that this model may be useful in exploring the factors which influence the release and migration of neutrophils in vivo.

In these studies, the DF32P leukokinetic technique has been modified so that neutrophil kinetics may be studied in the nonsteady state. In order to achieve this modification, an acid citrate solution was used in preparing leukocytes for radioactive assay. This results in a suspension free of cell agglutination so that a specific activity related only to neutrophils is obtained. This technique has been successfully employed in the study of the early phase of cortisol-induced granulocytosis, allowing measurement of cell flow rate in and out of the circulating granulocyte pool. This method is also useful in the study of subjects with pronounced neutropenia.

The kinetics of blood neutrophils was investigated by means of the in vitro radioactive diisopropyl fluorophosphate method in 35 patients with a chronic, steady-state neutropenia. There were 17 patients in whom the half disappearance time of neutrophils was normal. In 10 of these patients, the production of neutrophils was low and in 7, production was normal. In 18 patients the half disappearance time of neutrophilic granulocytes was shorter than normal. The production of neutrophilic granulocytes was low in five of these patients, normal in eight patients, and increased in five. An attempt was made to correlate other laboratory measurements with the kinetic picture, but no relationship was found; the marrow neutrophil reserve as measured by endotoxin or cortisol injection; marrow cellularity on aspiration or biopsy; in vitro-labeling index with 3HTdR; or serum lysozyme concentration proved of no value in identifying the various kinetic groups. The only finding that seemed to correlate with the kinetic picture was the presence or absence of splenomegaly. In 12 of the 18 patients with a short half disappearance time, splenomegaly was present whereas in 15 of 17 patients with a normal half disappearance time, there was no splenomegaly. Of 20 patients with greater than 1000 neutrophils per mm3, 17 were found to have a normal total-blood neutrophil pool. Thus these patients, with many of their cells marginated, agree to have a “shift neutropenia.” Myelocyte to blood transit time and myelocyte generation time, as measured in seven patients by in vivo labeling with diisopropy fluorophosphate, proved to be essentially normal. Thus, it appears that in chronic neutropenia, increased or decreased production of neutrophils is accomplished by increasing or decreasing early precursor input into the system.

These studies were designed to determine whether colony stimulating factor (CSF) and diffusible granulocytopoietic stimulator (DGS) represented a single substance which stimulates the production of granulocytes in vitro and in vivo. The effect of endogenous CSF upon DGS activity was studied after injection of 40 µg endotoxin. This resuited in both DGS activity and high serum CSF levels, but the time of appearance and duration of the two activities followed different patterns. In particular, CSF peaked and was declining 24 hr following endotoxin at which time DGS was not yet detectable. Then, as serum CSF fell, the DGS activity rose. These data support the concept that the DGS activity detectable with short term Millipore chamber cultures is not due to CSF and suggests the effects of CSF in vitro and DGS in vivo are due to different factors.

We examined the hypothesis that myeloperoxidase (MPO), a plentiful constituent of neutrophils, might serve as a marker for tissue neutrophil content. To completely extract MPO from either neutrophils or skin, hexadecyltrimethylammonium bromide (HTAB) was used to solubilize the enzyme. With this detergent treatment, 97.8 +/- 0.2% of total recoverable MPO was extracted from neutrophils with a single HTAB treatment; 93.1 +/- 1.0% was solubilized with a single treatment of skin. Neutrophil MPO was directly related to neutrophil number; with the dianisidine-H2O2 assay as few as 10(4) neutrophils could be detected. The background level of MPO within uninflamed tissue was 0.385 +/- 0.018 units per gram of tissue, equivalent to only 7.64 +/- 0.36 X 10(5) neutrophils. In experimental staphylococcal infection, skin specimens contained 34.8 +/- 3.8 units MPO per gram, equivalent to 8.55 +/- 0.93 X 10(7) neutrophils. These studies demonstrate that MPO can be used as a marker for skin neutrophil content: it is recoverable from skin in soluble form, and is directly related to neutrophil number. Further, normal skin possesses a low background of MPO compared to that of inflamed skin.

To assess release of neutrophils from the neutrophil storage pool (NSP) and subsequent neutrophil migration into tissue sites, polyvinyl sponge discs were placed into subcutaneous pockets in either newborn or 4-wk-old rats. After 20 hr, the sponges were removed, and the animals' NSP size and the number of neutrophils which had migrated into the sponges were measured. In response to the sponge stimulus, neonatal rats released 83.6 +/- 1.2% (mean +/- S.E.) of their NSP (106.3 +/- 15.3 X 10(5) neutrophils), whereas the older animals released 28.4 +/- 5.0% (628.5 +/- 124.4 X 10(5) neutrophils) of their NSP (P less than 0.001). Neonatal neutrophils displayed a decreased ability to successfully migrate from the NSP to the sponges. The proportion of released neutrophils accumulating in the sponges was 9.2 +/- 0.6% in the neonates, compared with 58.5 +/- 7,8% (P less than 0.001) in the 4-wk-old rats.

In contrast to the attitude prevalent a decade ago, clinicians entertaining the diagnosis of neonatal bacterial sepsis now often place considerable reliance on the blood neutrophil count and degree of left shift. In this report we present four cases which illustrate that in some patients, no derangement of the complete blood count (CBC) is present early in the course of bacterial sepsis. In order to determine the length of time between bacterial inoculation and the appearance of changes in the CBC, we used an animal model of early onset Group B streptococcal sepsis. In adult animals we observed characteristic changes in the CBC within 1 hour of bacterial inoculation, but in neonates this latent period was considerably longer, lasting 4 hours. Thus a normal CBC might actually be expected during the first several hours of early onset neonatal sepsis. This delay in appearance of CBC changes constitutes a previously uninvestigated feature of neonatal neutrophil kinetics, the "latent period."

Summary: The availibility of neutrophils is an important factor in host resistance to bacterial infection. Therefore, circulating and storage neutrophil quantification was carried out on groups of neonatal rats intranasally inoculated with type II group B streptococci. The dose of type II group B streptococci used produced 56% mortality in 48 neonatal rats with death being due to pneumonia and sepsis. Neutropenia (1300 ± 150/mm3 versus 2300 ± 170/mm3; mean ± S.D.; P < 0.01) and an elevation in band/polymorphonuclear ratio (0.98 ± 0.04 versus 0.30 ± 0.04: P < 0.01) were observed in infected neonatal rats 24 hr following inoculation. Femoral marrow as well as splenic and hepatic neutrophil storage compartment quantification revealed dimunition of postmitotic (polymorphonuclear, band, and metamyelocyte) neutrophils in the infected group (P < 0.01) with sparing of the proliferative neutrophils (myeloblasts, promyelocytes, and myelocytes). Repletion of the myeloid but not the splenic or hepatic neutrophil storage compartments with normalization of the neutrophil count and band/polymorphonuclear ratio occurred in animals surviving 72 hr. These studies establish that neutropenia and neutrophil storage pool depletion are prominent features of experimental type II group B streptococci infection in neonates. Speculation: Although neutrophil storage pool depletion is rare in infected adult animals or humans, its presence can be correlated with an increased mortality. It appears that neutrophil storage pool depletion is a prominent feature of experimental neonatal type II group B streptococci infection. The resulting deficiency of phagocytic cell supply probably represents diminution in the already compromised host defense mechanism of the neonate. Thus, it is postulated that neutrophil storage pool depletion represents a pathophysiologic factor contributing to the high mortality rate observed in neonatal type II group B streptococci infection.

When neutropenia due to exhaustion of the marrow neutrophil reserve, develops in a neonate with bacterial sepsis the likelihood of survival is very small. We report such a case who was treated with a double- volume exchange transfusion using fresh unstored whole blood. We were able to determine a net gain of 5 × 10(8) neutrophils per kg. Then, in neutropenic neonatal animals, neutrophil transfusion by double-volume exchange transfusion with unstored blood was investigated.

We measured the effect of lithium on the blood neutrophil mass and neutrophil production, using standard di-isopropylfluorophosphate (DF32P) methods. In 12 lithium-treated patients the total blood neutrophil pool was 105 (42 to 270) X 107 cells per kilogram (median and 95 per cent limits) as compared with 61 (27 to 138) X 107 in 71 controls (P

AN ADEQUATE SUPPLY of neutrophils is an important factor in host resistance to bacterial infection.' Indeed, when the supply is inadequate, the probability of surviving an infection diminishes." One sign of depleted neutrophil supply is blood neutropenia, a finding which is associated with increased mortality in infants with sepsis? 4 It is tempting to assume that the neutropenia of such infants is a reflection of inadequate neutrophil supply, but neutropenia can occur even when neutrophil supply is normal, as during the margination of circulating neutrophils primarily to post-capillary venular sites after endotoxin challenge? Thus, the finding of neutropenia itself does not establish that neutrophil supply is diminished, nor does it necessarily signify compromised host resistance. In order to determine whether neutropenia signaled neutrophil supply exhaustion in infected infants, we examined the marrow of nine living infants and six infants who had died from sepsis. In two infected neutropenic infants the neutrophil storage pool was found to be normal; both of these infants survived. Seven other neonates had diminished NSP, and those with the most profound depletion died. All infants who died of sepsis had NSP depletion. These observations suggest that neutropenia in infected infants frequently reflects NSP depletion. The identification of infants with NSP depletion may be of value as a prognostic sign and may aid in the consideration of therapeutic manuevers such as neutrophil transfusion.

Since lithium causes granulocytosis in some patients, its effect upon granulocyte production was investigated using mouse marrow in the agar culture system. When lithium was added to semisolid cultures of mouse marrow, there was no stimulation of colony formation in the absence of colony-stimulating activity (CSA). In addition, lithium did not potentiate the action of already formed CSA. However, lithium did stimulate the production of CSA by lung tissue. Lithium enhancement of CSA production was blocked by puromycin, indicating that lithium action required active new protein synthesis. It was concluded that lithium promoted enhanced granulocyte production in vitro by stimulating the synthesis of CSA.

In groups of adults, and term and preterm neonates, we determined: the blood concentration, the proliferative rate, and the variety of progeny of committed granulocyte-macrophage progenitor cells (CFU-GM). In five of eight term neonates and in all premature infants, a potentially significant limitation of neutrophil production was detected. Unlike the slowly proliferating CFU-GM present in the blood of healthy adult subjects (7% thymidine suicide, range 0% to 32%), the circulating CFU-GM in the premature subjects were proliferating at a near maximal rate (55%, range 40% to 75%, P less than 0.001). Because CFU-GM proliferation is nearly maximal in the baseline, noninfected state, neonates may have restricted ability to increase neutrophil production from CFU-GM during times of increased neutrophil need, such as during bacterial infection. Such inability may predispose neonates to exhaustion of the neutrophil supply during bacterial infection.

The leukocyte left shift is commonly used as an adjunct to the early diagnosis of bacterial infection. Many different methods have been employed for its quantification, such as the absolute band count, band/seg ratio, band/total neutrophil ratio, and immature/total neutrophil ratio. We examined blood and bone marrow samples in groups of noninfected and infected neonatal dogs and human beings in order to determine which method most clearly reflects an increased call upon marrow neutrophil reserves and which correlates best with the presence and severity of infection. We found that the neutrophil ratios were more frequently abnormal during neonatal sepsis than was the the absolute band count. All subjects, canine and human, in whom the immature/total neutrophil ratio exceeded 0.800 were found to have depletion of the marrow neutrophil reserves, and those with the most profound depletion died. This study supports the concept that an elevated immature/total neutrophil ratio can aid in the diagnosis of bacterial infection in the newborn infant and suggests that the degree of elevation may serve as a method for detecting subjects at high risk for depletion of the marrow neutrophil reserves and death from sepsis.

The process by which resting hemopoietic stem cells become activated is poorly understood, but it has been suggested that cyclic nucleotide levels in the cell may play an important role. In the present study, the effect of various, nucleotides and stimulators of nucleotide synthesis upon the formation of in vitro granulocyte colonies has been examined. It was found that physiologic concentrations of 3′, 5′-cyclic guanosine monophosphate enhanced the formation of granulocyte- macrophage colonies in the presence of colony-stimulating activity. The cells sensitive to cGMP activation were resistant to thymidine suicide and could not be activated by colony-stimulating activity alone. Therefore it was suggested that the cGMP sensitive stem cell was an ordinarily resting stem cell which was triggered into a proliferative state by cGMP.

Summary: Newborn dogs were inoculated intratracheally with 0.5–1.0 x 108 Staphylococcus aureus/g body weight. Neutropenia (490 ± 280 neutrophils/mm3 versus 8,390 ± 490 control, ± S.E., P< 0.001), and depletion of the marrow neutrophil storage pool (3 ± 1% versus 27 ± 2% storage neutrophils P < 0.001) occurred 5–6 h following the inoculation. All animals died at 6–10 h. Additional inoculated pups were selected at random to receive transfusions of either granulocytes, plasma or red blood cells. Granulocyte transfusions (3x109 neutrophils/kg) improved survival (P < 0.005), but plasma and red blood cells did not. Speculation: Previous reports have shown that certain neonates with bacterial sepsis utilize neutrophils more rapidly than they are replaced by the marrow mitotic neutrophil pool. The quantitative neutrophil deficiency thus produced, results in blood neutropenia, reduced neutrophil availability to affected tissues and high mortality. In this study, neonatal dogs were infected experimentally and found to develop neutropenia, depletion of neutrophil reserves, and death. Granulocyte transfusions improved the survival of these animals while plasma or packed red blood cell transfusions did not. Thus, granulocyte transfusion may present a useful therapeutic modality in those septic neonates who are found to have neutropenia and a depleted neutrophil supply.

In experimental animals, the quantity of myeloperoxidase (MPO) in a volume of whole blood, and the neutrophil concentration in that same volume, were determiner and the results expressed as units of MPO (10-7)/neutrophil. Two situations are repored in which the concentration of MPO/neutrophil was found to change; during the growth and development of animals and during bacterial infection. Premature and newborn rats had only 25% or the MPO/neutrophil found in adults. One to three wk olds had 50% of the adult enzyme concentration. During fatal bacterial infection, MPO/neutrophil fell rapidly, often to undetectable levels, but during sublethal infections, following a 24–h lag period in adults and a 48–h lag in neonates, the concentration increased to twice normal.

Bacterial sepsis is a relatively common problem in the neonatal period, particularly among prematurely delivered infants. The newborn rat has been widely used as a model for sepsis neonatorum, and in that model incomplete development of the neutrophil system has been postulated to be an important factor predisposing neonates to death from bacterial infection. In this study, that hypothesis was further tested by assessing neutrophil development in rats of various pre- and postnatal ages. Using standard soft agar colony techniques for detecting granulocyte-macrophage progenitor cells [CFU(c)], the number of CFU(c)/g of body weight was seen to increase from 0.5 + 0.1 X 10(3) at 19-20 days gestation to 10.5 +/- 0.2 X 10(3) at 4 weeks. The anatomic location of CFU(c) changed from totally hepatic at 16 days gestation to almost totally myeloid at 4 weeks. Lastly, the proportion of mature, stored neutrophils/CFU(c) decreased from 2440 +/- 40 at 19-20 days gestation to 430 +/- 75 at 4 weeks.

Fetal blood may be useful for the study of hematopoietic development in humans. However, the methods used to study blood cell colonies in adults may not be optimal for the study of colonies derived from fetal blood. Using cord blood from six healthy term pregnancies and marrow from six healthy adult volunteers, we compared the chronology of emergence, morphology, and differentiation of progenitor cell colonies from the two sources. In cultures of adult marrow, erythroid colony-forming unit colonies reached maximal concentrations after 8.5 +/- 0.3 days of culture (mean +/- SEM), but erythroid colony-forming unit colonies derived from fetal blood reached maximal concentrations sooner, after 6.5 +/- 0.2 days (p less than 0.01). Single-centered-erythroid burst-forming unit colonies from adult marrow (45 +/- 5/10(5) light-density, "accessory-cell"-depleted cells) reached a peak at 14 days, but from cord blood they were significantly greater at 9 days (200 +/- 15/10(5) cells) than at 14 days. When studying fetal blood erythroid colony-forming unit and "mature-erythroid burst-forming unit," colonies should be enumerated earlier than when studying adult marrow-derived progenitors. Otherwise, the concentrations of these progenitors will be significantly underestimated.

We observed a 24-month-old infant who developed anaemia, thrombocytopenia and neutropenia while recuperating from an extensive burn. In order to determine the mechanism(s) responsible for the pancytopenia, we quantified marrow-derived haematopoietic progenitor cells, assessed the relative proliferative rate of haematopoietic progenitor cells, and sought the presence of progenitor cell inhibitors. The concentration and relative proliferative rate of pluripotent progenitors (CFU—GEMM) were elevated. No inhibitors of progenitor cells were observed; in fact, the patient's serum contained very high levels of stimulatory activity for CFU—GEMM as well as for granulocytemacrophage progenitors (CFU—GM). However, the marrow concentration of erythroid progenitors (BFU—E and CFU—E) was diminished. We conclude that the anaemia in this patient was the result of either hypoproduction of differentiated erythroid progenitors or intramyeloid destruction of early erythroid cells. In contrast, the neutropenia was likely to be due to accelerated neutrophil consumption at a rate that exceeded the capacity for increasing neutrophil production.