Chapter 61: Production, Distribution, and Fate of Neutrophils

Neutrophils are produced in the marrow, where they arise from progenitor and precursor cells by a process of cellular proliferation and maturation. They differentiate from the pluripotential stem cell^1,2 through a series of progressively more committed progenitor or colony-forming units, including the granulocyte-monocyte colony-forming unit and the granulocyte colony-forming unit, which give rise to neutrophils.^3,4 The early progenitor cells cannot be recognized under the microscope but can be identified retrospectively by the type of colony formed in culture of marrow cells (Chap. 18). The earliest morphologically recognizable neutrophil precursor is the myeloblast. From there, the formal sequence of precursor development is myeloblast → promyelocyte → myelocyte → metamyelocyte → band neutrophil → segmented neutrophil (Chap. 60). The term granulocyte often is loosely used to refer to neutrophils but strictly speaking includes eosinophils and basophils. Eosinophilic (Chap. 62) and basophilic granulocytes (Chap. 63) develop from progenitors in a manner analogous to the neutrophils, although commitment to neutrophilic, eosinophilic, or basophilic development probably is established at an early progenitor stage, and are dependent on the cytokines interleukin (IL)-5 and stem cell factor, respectively.

The normal human neutrophil production rate is 0.85 to 1.6 × 10⁹ cells/kg per day. Mature neutrophils are stored in the marrow before they are released into the blood. In the absence of an inflammatory focus, they leave the circulation randomly, with a half-disappearance time of approximately 7 hours. The cells then enter the tissues and probably function for 1 or 2 days before their death or loss into the gastrointestinal tract through mucosal surfaces.

The marrow has an impressive capacity to produce neutrophils, yet it is carefully regulated both at steady state and in times of heightened demand. This chapter outlines current concepts of neutrophil production, distribution, and survival. For detailed data and methods, the reader is referred to original articles and reviews on neutropoiesis and neutrophil kinetics.^{5,6,7,8,9,10,11,12,13,14,15,16,17}

Although the primary cellular manifestation of commitment is the expression of receptors for lineage-specific hematopoietic growth factors, the “decision” for a stem cell to self-renew is, at least partially, a stochastic event.^1,18 On the other hand, stromal elements, collectively referred to as the hematopoietic microenvironment, release short-range signals that regulate the process of commitment from multipotential stem cell pools. Although many details of hematopoietic stem cell regulation (Chap. 18) remain to be elucidated, much is known regarding the interaction of hematopoietic cytokines with their receptors and actions on the committed granulocyte progenitor cells and their mature progeny.^{19,20,21,22,23,24}

HUMORAL REGULATORS

The humoral regulators involved in granulopoiesis have been defined by in vitro culture systems.^20,21 Originally identified by their ability to stimulate colony formation from marrow progenitor cells, the hemopoietic cytokines were initially termed colony-stimulating factors (CSFs) based on this assay system.²⁵ With regard to neutrophil production, at least four human CSFs have been defined. Granulocyte-monocyte colony-stimulating factor (GM-CSF) is a 22,000 relative molecular mass (Mr) glycoprotein that stimulates the production of neutrophils, monocytes, and eosinophils. Granulocyte colony-stimulating factor (G-CSF) has a Mr of 20,000 and stimulates exclusively the production of neutrophils. IL-3, or multi-CSF, also has a Mr of 20,000 and acts relatively early in hematopoiesis, affecting pluripotential stem cells. Finally, stem cell factor (also known as c-kit ligand or steel factor), with a Mr of 28,000, acts in combination with IL-3 and/or GM-CSF to stimulate the proliferation of the early hematopoietic progenitor cells, basophils and mast cells.

In addition to their effects on neutrophil precursors, G-CSF and GM-CSF act directly on the neutrophil, enhancing its function. These cytokines regulate the production, survival, and functional activity of neutrophils.^21,22,26,27 In a murine model of severe bacterial infection, endothelial cells translate pathogen signals into G-CSF–driven marrow neutrophil production.²⁸ The mature neutrophil lacks IL-3 receptors and thus is not affected by IL-3. In fact, the genetic elimination of IL-3 obliterates delayed type hypersensitivity. However, IL-3 receptors are present on mature eosinophils and monocytes. IL-3 is produced by activated T lymphocytes and thus is expected to have a physiologic role in circumstances of cell-mediated immunity. GM-CSF also is produced by activated lymphocytes. However, like G-CSF, it also is elaborated by mononuclear phagocytes and endothelial and mesenchymal cells when these cell types are stimulated by certain cytokines, including IL-1 and tumor necrosis factor, or bacterial products, such as endotoxin.^29,30,31 Stem cell factor is secreted by a variety of cells, including marrow stromal cells,^32,33 and affects the development of several kinds of tissues.^32,34

The activities of exogenously administered biosynthetic (recombinant) human G-CSF and GM-CSF in humans are well documented.^{22,27,35,36,37} G-CSF administration rapidly induces neutrophilia, whereas GM-CSF causes an increase in neutrophils, eosinophils, and monocytes. GM-CSF cannot be detected easily in normal plasma; thus, its role as a day-to-day, long-range modulator of neutrophil production is uncertain. Mice in which the GM-CSF gene is “knocked out” have generally normal hematopoiesis, but show macrophage abnormalities, pulmonary alveolar proteinosis, and decreased resistance to microbial challenge.^38,39,40,41 However, G-CSF appears to be a critical regulator of neutrophil development, as giving an animal an antibody to G-CSF leads to profound neutropenia.⁴² The G-CSF knockout mouse shows severe neutropenia.⁴³ Neutropenia that results from a production disturbance, such as exposure to cytotoxic drugs, is associated with high circulating serum concentrations of G-CSF.⁴⁴

NEUTROPHIL KINETICS

Methods used to study granulocyte kinetics can be categorized as follows: (1) neutrophil depletion or destruction to determine the size and rate of mobilization of reserves and the level of compensatory neutrophil production; (2) use of radioactive tracers to study neutrophil distribution, production rates, and survival times; (3) mitotic indices of marrow granulocytic cells to assess proliferative activity and cell cycle times; and (4) induced inflammatory lesions to study cell movement into the tissues. Of these categories, the most popular has been the use of radioactive tracers.

Neutrophil production and neutrophil kinetics usually are analyzed by describing neutrophil movement through a number of interconnected compartments. These compartments can be arranged into three major groups: the marrow, the blood, and the tissue (Fig. 61–1). The complexities of analyzing these compartments are covered in several recent reviews.^45,46,47,48

The Marrow

Marrow neutrophils can be divided into the mitotic, or proliferative, compartment and the maturation storage compartment. Myeloblasts, promyelocytes, and myelocytes are capable of replication and constitute the mitotic compartment. Earlier progenitor cells are few in number, not morphologically identifiable, and usually neglected in kinetic studies. Metamyelocytes, bands, and mature neutrophils, none of which replicate, constitute the maturation storage compartment.

The average number of cell divisions from the myeloblast to the myelocyte stage in the proliferative compartment has been estimated between four and five.⁴⁹ Data obtained using radioactive diisopropyl fluorophosphate (DF³²P) suggest the existence of three divisions at the myelocyte stage, but the number of cell divisions at each step may not be constant. The major increase in neutrophil number probably occurs at the myelocyte level, because the myelocyte pool is at least four times the size of the promyelocyte pool. Because of the difficulties in measuring human intramarrow neutrophil kinetics, a precise model of the dynamics of the mitotic compartment is not available.

Table 61–1 lists the estimated sizes of the marrow neutrophil compartments and the transit times and cell-cycle stages of the cells in the various compartments. Precise studies have measured a postmitotic pool of (5.59 ± 0.9) × 10⁹ cells/kg and a mitotic pool (promyelocytes and myelocytes) of (2.11 ± 0.36) × 10⁹ cells/kg. These studies led to a calculated normal marrow neutrophil production of 0.85 × 10⁹ cells/kg per day. Radioautographic studies with [³H]thymidine support the concept of an orderly progression from metamyelocytes to mature neutrophils within the maturation storage compartment. These studies also suggest a “first in, first out” pattern for cells leaving this compartment and entering the blood. Several labeling techniques indicate the myelocyte-to-blood transit time is 5 to 7 days.^12,50 Previous studies with DF³²P reported a range from 8 to 14 days.^9,49 During infections, however, the myelocyte-to-blood transit time may be as short as 48 hours.⁵¹

Whether the production of neutrophils in the mitotic compartment exactly equals the neutrophil turnover rate (NTR) is not known with certainty. Studies in dogs suggest some immature neutrophils die in the marrow (“ineffective granulopoiesis”).⁵² Ineffective granulopoiesis has not been shown in normal humans,^14,53 although ineffective granulopoiesis occurs in some pathologic states, including the myelodysplastic syndromes,⁵⁴ myelofibrosis, and some of the idiopathic neutropenic disorders. At present, however, no convenient means of quantitating ineffective granulopoiesis is available.

On completion of maturation, the neutrophils are stored in the marrow and are referred to as the mature neutrophil reserve. The reserve contains many more cells than are normally circulating in the blood. Table 61–2 lists comparative data on the characteristics of the maturation storage compartment. Under stress, maturation time may be shortened, divisions may be skipped, and release into the blood may occur prematurely.

The Blood

Neutrophils leave the marrow storage compartment and enter the blood without significant reentry into the marrow. The total blood neutrophil pool (TBNP) consists of all the neutrophils in the vascular spaces. Some of these neutrophils are free in the circulation (the circulating pool), while others roll along the endothelium of small vessels or are temporarily sequestered in the alveolar capillaries of the lung (the marginated pool).^55,56 Cells in the two pools are freely exchangeable. When neutrophils labeled with DF³²P are injected into normal subjects, approximately half can be accounted for in the circulating pool; the remainder enters the marginated pool.^5,6,7 Neutrophils shift from the marginated to the circulating pool with exercise, epinephrine injection, or stress, but eventually the neutrophils leave the blood and enter the tissues. Once the neutrophils enter the tissues, they do not normally return to the blood. The flow of cells is unidirectional.

DF³²P-labeled neutrophils disappear from the circulation with a half-time (T_1/2) of 6.7 hours.^7,57,58 These data are supported by the finding that more than half of Pelger-Huët cells infused into a normal individual disappeared after 6 to 8 hours.⁵⁹ Data obtained with ⁵¹Cr-labeled neutrophils give substantially longer half-times.⁶⁰ The exponential disappearance of cells from the blood suggests the cells leave in a random manner. Thus, neutrophils newly released from the marrow are as likely to leave the blood as are neutrophils that have been circulating for several hours. Neutrophils also are eliminated by programmed cell death and disposed of by the macrophage system.^{51,61,62,63,64}

Direct observations of blood vessels have revealed some degree of leukocyte rolling along the endothelium (first observed many years ago by Atherton and Born⁶⁰). Although the observation has been clearly confirmed by numerous laboratories in different species of animals, the extent to which this phenomenon contributes to the marginated pool of neutrophils is uncertain.

A more compelling concept of the marginated pool is derived from investigations of the vascular bed of the lung. A distinctive characteristic of this tissue is the complex interconnecting network of short capillary segments where the path from arteriole to venule crosses several alveolar walls (often more than eight) and often contains more than 50 capillary segments.^{65,66,67,68,69} Compared to blood in the large vessels of most vascular beds, the blood in this complex network contains approximately 50-fold more neutrophils and even more lymphocytes and monocytes.⁷⁰ Videomicroscopic study of these vessels in animal models has revealed the transit of neutrophils through this network required a median time of 26 seconds and mean time of 6.1 seconds.^71,72 In contrast, the transit times of red blood cells ranges from 1.4 to 4.2 seconds. The increased transit time results primarily from the time neutrophils are stopped within this vascular network. The longer time required for the neutrophils to pass through this bed apparently accounts for their increased concentration.

Recruitment of neutrophils into the lungs through the alveolar capillary network contrasts with the recruitment of neutrophils through postcapillary venules at sites of inflammation in a number of important ways. The tethering mechanisms required to capture neutrophils from flowing blood in larger vessels apparently are not necessary in the alveolar capillary bed. The diameters of spherical neutrophils (6 to 8 µm) are larger than the diameters of many capillary segments (2 to 15 µm). Approximately 50 percent of the capillary segments would require neutrophils to change their shape in order to pass through.^72,73,74,75 Given the large number of capillary segments through which a neutrophil must pass (often more than 50), most neutrophils must change shape during transit from arteriole to venule. Morphometric analysis of neutrophils in the alveolar capillary beds has revealed significant deviation from spherical shape.^72,73 Computational models of the capillary bed describing flow, hematocrit, pressure gradients, and the effects of deformation on the capillary transit times of neutrophils support the concept that the structure of the capillary bed and the deformation of neutrophils are critical under normal conditions. Thus, the enormous lung vascular bed contains a substantial number of neutrophils that can be mobilized into the systemic circulation with stimuli such as epinephrine or exercise.

During inflammation, much of the sequestration and infiltration occur through vessels so narrow that physical tapping is sufficient to stop the flowing neutrophil.^68,72,76,77 Binding of mediators such as chemotactic factors (e.g., C5a, the chemotactic fragment of complement component C5) to neutrophil receptors induces a transient resistance of the cells to deformation.^{78,79,80,81,82,83} Because neutrophils must deform to pass through the capillary bed, leukocyte activation by inflammatory mediators could affect further concentration of neutrophils at the alveolar walls.^65,76 The role of mechanical factors in the initial sequestration of neutrophils in the alveolar capillaries is supported by evidence that neither L-selectin nor β₂-integrins are required.^76,84,85 In contrast, both selectins and β₂-integrins are required for localization of neutrophils in postcapillary venules at sites of inflammation.

The events following the initial sequestration of neutrophils within alveolar capillary beds are influenced by adhesion molecules. For example, simple systemic activation of neutrophils by intravenous injection of chemotactic factors (e.g., IL-8 or C5a) results in rapid (<1 minute) neutropenia with massive sequestration of neutrophils within alveolar capillaries. This event is not dependent on L-selectin or β₂-integrins, but the retention times within this capillary bed are influenced by these adhesion molecules.^76,85 Adhesion likely is an interaction of leukocyte adhesion molecules and endothelial adhesion molecules. Blockade of the adhesive mechanism (e.g., using blocking monoclonal antibodies) results in release of neutrophils from the lungs.^{76,84,86,87,88} Mediator-induced decreases in deformability are temporally correlated with upregulation of β₂-integrins (e.g., both occurring within approximately 1 minute of exposure to IL-8). This allows both physical trapping and sticking to the vascular wall within the alveolar capillary bed. A similar phenomenon occurs in the liver where sequestration is the result of physical trapping and liver injury is heavily dependent on adhesion of leukocytes through the β₂-integrins.⁸⁹

Assuming a random loss of neutrophils from the blood, NTR can be calculated from T_1/2 and TBNP: NTR = 0.693 × TBNP/T_1/2. In the steady state, NTR measures the rate of effective neutrophil production. Table 61–3 lists the definitions and calculations related to blood neutrophil kinetics. Table 61–4 lists data for normal human blood neutrophil kinetics. The high production rate of neutrophils under normal conditions is remarkable, especially given that the rate may increase several fold in response to inflammatory stimuli.

Glucocorticoids increase TBNP by increasing influx from the marrow and decreasing efflux from the circulation. Five hours after a pharmacologic dose of glucocorticoid, the neutrophil count increases by approximately 4000/µL because of release from the marrow, demargination, and prolongation of T_1/2 to approximately 10 hours.^90,91,92 Consistent with the increase in T_1/2, prednisone reduces the accumulation of neutrophils at induced sites of skin inflammation.⁷⁵ With alternate-day, single-dose prednisone, neutrophil counts and kinetics are normal 24 hours after administration and during the day off.⁹³ Endotoxin causes a prompt neutropenia as a result of cell margination and sequestration, followed in 2 to 4 hours by a rebound neutrophilia as a result of cell release from the marrow. The size of the neutrophilic response correlates with the functional marrow reserves.^94,95,96,97 After epinephrine administration, a peak leukocytosis occurs in 5 to 10 minutes and rarely lasts more than 20 minutes. This finding reflects a shift of cells from the marginated to the circulating pool.

MIGRATION OF NEUTROPHILS INTO TISSUES

The migration of neutrophils from blood into tissue at sites of inflammation involves a series of sequential adhesive steps proceeding from tethering (rolling adhesion) on endothelium under shear conditions in postcapillary venules.⁹⁸ This model has been investigated in a variety of vascular beds⁹⁹ and in vitro with monolayers of endothelial cells in parallel plate flow chambers.⁹⁸ The tethering event in this model depends on adhesion molecules in the selectin family, E-selectin and P-selectin on the endothelium, L-selectin on the neutrophil, and ligands for the selectins expressed on both cell types. These adhesion molecules are necessary to efficiently initiate the cascade of adhesive steps ultimately leading to firm attachment of the neutrophils to endothelium. The cascade appears to be necessary for neutrophils to move from blood to tissues because the unstimulated neutrophil is not adhesive to endothelium.^98,100 The integrins necessary for firm adhesion and cell locomotion require stimulation to promote sufficient increases in avidity or affinity to support these functions (Chap. 19).

LIFE SPAN OF NEUTROPHILS

After emigrating into tissue, the life span of neutrophils can be significantly prolonged (24 to 48 hours).¹⁰¹ Programmed cell death (apoptosis) accounts for significant removal of tissue neutrophils through phagocytosis by macrophages. The constitutive rate of apoptosis of neutrophils is altered by inflammatory cytokines and chemokines. For example, tumor necrosis factor-α (TNF-α) accelerates the rate, but endotoxin, G-CSF, GM-CSF, IL-15, and IL-3 inhibit the rate of apoptosis. The balance of these effects at specific inflammatory sites is poorly understood, but the functional life of neutrophils in tissue appears to be controlled by the rate of apoptosis. Apoptotic neutrophils lose the ability to release granular enzymes in response to external stimuli (see below), and marked changes in cell surface proteins occur (e.g., CD16, CD43, CD62L are greatly reduced). Although the loss of responsiveness may contribute to resolution of the inflammatory process, evidence indicates macrophages also are altered by the phagocytosis of apoptotic neutrophils. In contrast to the macrophage response to phagocytosis of microbes, where secretion of proinflammatory cytokines (e.g., IL-1β) and chemokines (e.g., IL-8) is stimulated, phagocytosis of apoptotic neutrophils fails to provoke secretion of proinflammatory factors; instead, phagocytosis stimulates release of factors that may suppress inflammatory responses (e.g., transforming growth factor-β and prostaglandin E₂). Macrophage recognition of apoptotic neutrophils is partially understood to involve the vitronectin receptor α_Vβ₃ and the thrombospondin receptor CD36 on the macrophage surface. In addition, phosphatidylserine residues on the neutrophil are involved.⁹⁸

Neutrophils are capable of phenotypic changes depending on the tissue and cytokine/chemokine milieu at the time of their migration into tissue (Chap. 60). Because our understanding of neutrophil physiology is relatively new, knowing the extent of this phenomenon on neutrophil life span in tissues is not possible at present.

WHITE CELL COUNT AND MARROW CELLULARITY

White cell and absolute neutrophil counts are the most widely used guides to the status of neutrophil production. They are useful in evaluating the effects of cytotoxic chemotherapy, although they do not provide quantitative information on the rate of neutrophil production or destruction, the status of marrow reserves, or the presence of abnormalities in cell distribution.

Gauging neutrophil production by the appearance of marrow films, clot sections, or biopsies suffers from the limitations of sampling error and relatively poor correlation with kinetics, as measured by other techniques.⁶⁸ For example, the morphologic findings in the marrow of a “maturation arrest,” with little neutrophil development beyond the promyelocyte or myelocyte stage, does not distinguish between a defect in precursor cell maturation and rapid mobilization of postmitotic cells from the marrow. Similarly, distinguishing by purely morphologic means neutropenic conditions resulting from ineffective neutrophil production from conditions caused by peripheral destruction of neutrophils often is difficult. However, despite these limitations, when the absolute neutrophil count and marrow cellularity are used together, they provide a useful guide in most clinical settings. If the absolute neutrophil count is less than 1.0 × 10⁹/L and multiple marrow aspirations and/or biopsies are hypocellular, the patient almost invariably has impaired production of marrow neutrophils. Very low neutrophil counts predispose to infections by bacteria and certain fungi (e.g., Candida and Aspergillus). Such infections become especially troublesome as the neutrophil count falls below 0.5 × 10⁹/L (Chap. 65). Unfortunately, the converse is not true. The finding of cellular marrow and neutrophil count >1.0 × 10⁹/L does not mean production is normal. Nevertheless, when marrow cellularity and absolute neutrophil count are considered together, they provide the most clinically useful assessment of neutrophil production.

FUNCTIONAL EVALUATION

Several agents that increase neutrophil numbers in circulation, including glucocorticoids, endotoxin, and etiocholanolone, have been used to evaluate neutrophil reserves in a clinical setting. These agents have been supplanted by recombinant human G-CSF, a remarkably nontoxic cytokine that, when given in therapeutic doses (5 to 8 mcg/kg), increases the blood neutrophil count by stimulating neutrophil production and accelerating neutrophil release from the marrow storage compartment (Chap. 65). The increase in neutrophil production results from a threefold increase in the number of cell divisions in the mitotic compartment and shortening of the maturation time from myelocyte to neutrophil from 4 to 5 days to less than 1 day.^102,103 Thus, as a byproduct of its therapeutic action, G-CSF administration directly tests an individual’s capacity to produce neutrophils. This effect of G-CSF makes obsolete most of the older methods for evaluating neutrophil compartments.

G-CSF does not test the distribution of neutrophils between the marginated and circulating pools. On the rare occasions when such information is desirable, epinephrine stimulation can be used to assess the distribution. For this purpose, epinephrine 0.1 mg infused intravenously over 5 minutes has been used, and blood for white counts is obtained before and 1, 3, and 5 minutes after completion of the epinephrine infusion. Normally the neutrophils increase by approximately 50 percent after epinephrine infusion.¹⁰⁴