In invertebrates such as worms and sessile marine creatures, erythropoiesis occurs adjacent to peritoneal and endothelial cells. In premammalian species, the spleen is the primary site of erythropoiesis. With evolutionary advancement, the function gradually shifts to the liver and the sinusoidal cavities of bones. These observations suggest that sufficient oxygen, a stagnated flow of blood to avoid dispersion of factors produced locally, and extensive and redundant surfaces for cell–cell interactions are essential to supporting red cell production. Similar sites support erythropoiesis during human development. During both phylogeny and ontogeny, the liver and spleen are primarily erythropoietic organs; granulocytic cells dominate in the bone marrow. Within the bone marrow, hematopoiesis is restricted to the extravascular space, where compact collections of cells are interspersed among venous sinuses. These sinuses originate adjacent to the endosteal bone surface and empty into a central longitudinal vein. Studies in mice demonstrate that BFU-E follows a bimodal distribution with peaks adjacent to the periosteum and mid cavity, whereas CFU-E and later erythroid cells have a broad distribution with highest incidence toward the axis of the femur, adjacent to the central vein, thus suggesting that the local anatomy (specialized niches?) influences the maturation of erythroid cells.

The bone marrow microenvironment consists of three broad components: stromal cells (e.g., fibroblasts, endothelial cells, mesenchymal stem cells and their diverse descendant progeny), accessory cells (monocytes, macrophages, megakaryocytes, T cells), and extracellular matrix (a protein-carbohydrate scaffold). In the bone marrow, although early studies described two distinct hematopoietic niches for hematopoietic stem cells/progenitor cells (i.e., the “endosteal” niche and the “endothelial or vascular niche” within the medulla), more recent studies suggest that this dis tinction may be artificial, because both cellular structures can be intimately associated in trabecular bone. Thus mesenchymal stem cell–derived osteoprogenitor cells and stromal reticular cells (nes tin+ or leptinR+) are intimately associated with sinusoidal endothelium and seem to be pivotal organizers of the bone marrow niche. Reciprocal communication of hematopoietic stem cells with cells/ matrix in their bone marrow niche ensures both their quiescent state and self-renewal dynamics.

EPO, in addition to promoting erythropoiesis directly, enhances erythropoiesis indirectly by decreasing the interaction of hematopoietic stem cells with their niches, reducing the amount of trabecular bone and downregulating CXCR4 expression, the receptor for CXCL12/ SDF1 expressed by stromal cells, on hematopoietic stem cells.

Accessory cells are progeny of hematopoietic stem cells; hence after marrow transplantation these cells are of donor origin, whereas stromal cells remain mostly host derived. Extracellular matrix molecules are synthesized and secreted by microenvironmental cells and include collagens (types I, III, IV, and V), glycoproteins (fibronectin, laminins, thrombospondins, hemonectin, and tenascin), and glycosaminoglycans (hyaluronic acid, chondroitin, dermatan, and heparan sulfate). The production of extracellular matrix proteoglycans by mesenchymal stromal cells may be regulated by the Wnt pathway.

Besides providing structure to the marrow space and a surface for cell adhesion, the microenvironment is important for hematopoietic cell homing, engraftment, migration, and the response to physiologic stress and homeostasis. Although the functional consequences of the microenvironment ultimately must be defined by in vivo studies in mice, dissection of the cellular components of the microenvironment, definition of the cytokines that are produced by individual cells, and the nature of cell–cell interactions have been aided by in vitro models. Long-term bone marrow cultures provide an experimental approach for such studies. Under these in vitro conditions, murine hematopoiesis can be maintained for 8 to 10 months and human hematopoiesis for 2 to 3 months. An adherent layer consisting of fibroblasts, adipocytes, and macrophages is a crucial component of the culture system. Progenitor cells adherent to stroma are generally quiescent (dormant), whereas those in the non-adherent cell compartment are in active cell cycle.

In vitro studies have demonstrated that stromal cells, including endothelial cells and fibroblasts, elaborate cytokines such as GM-CSF, G-CSF, IL-1, IL-3, IL-6, IL-11, KL, Flt-3 ligand, activin A, and basic fibroblast growth factor, which influence, alone or in combination, the growth of adjacent marrow progenitors.91 In addition to positive regulators of replication and differentiation, stromal cells elaborate factors such as TGF-β, TPO, CXCL12, IFN-γ, TNF-α, and TRAIL, which exert a negative influence on proliferation and may help maintain a dormant (noncycling) state. These negative regulators are responsible, at least in part, for the anemia associated with chronic inflammatory states. The effects of TNF-α and TRAIL are mediated through induction of apoptosis at specific stages of erythroid matura tion. In the case of TRAIL, a complex system of signaling and decoy receptor isoforms determines the precise cell window susceptible to TRAIL-induced apoptosis.95 TRAIL probably induces apoptosis by competing with EPO for activation of Bruton tyrosine kinase. Its effects are counteracted by KL and PKCε signaling. TRAIL is also involved in the pathobiology of the anemia associated with multiple myeloma (TRAIL is overproduced by the malignant plasma cells of these patients) and myelodysplastic syndrome (MDS) (myelodysplastic erythroid progenitors overexpress the adaptor Fas-associated death domain of the TRAIL receptor). On the other hand, the negative effects of TGF-β are mainly achieved by accelerating cell differentiation, whereas data from mouse models indicate that chronic exposure of IFN-γ reduces the erythrocyte life span and inhibits erythropoiesis by promoting the expression of PU.1, a transcription factor that antagonizes GATA1, a master transcriptional regulator of erythropoiesis. Because some regulators inhibit differentiation along certain lineages but not others, there is an intriguing possibility that lineage specific regulation within the microenvironment can be achieved through negative, rather than positive, factors.

Among the factors elaborated by stromal cells, KL has the most profound effect on erythropoiesis as emphasized by genome-wide association studies (GWAS) that have identified several single nucleotide polymorphisms (SNPs) lying close to a prominent DNase hypersensitive region ∼115 kb upstream of the gene encoding its receptor, KIT, associated with variability in RBC counts, mean RBC volume, and mean RBC hemoglobin content observed in the normal population. Stromal cells produce a membrane bound form of KL that may be cleaved in a soluble form and released in the circulation. Mice with KIT mutations (W mutations), leading to absence of or compromised kinase activity of the intracellular domain of the receptor, and steel mice, with mutations of KL, have disproportion ate and severe reduction of the numbers of late erythroid progenitors, CFU-E, and differentiated erythroid precursors resulting in anemia. Steel (Sl) mice unable to synthesize KL die in utero because of severe anemia while steel-Dickie (Sld) mice that are unable to make the membrane-restricted form of KL are viable but severely anemic, whereas other lineages in these animals are marginally affected or not affected by this defect. The fact that erythropoiesis is abnormal, despite high levels of circulating EPO and the presence of soluble KL, suggests that normal erythroid differentiation and maturation require both a functional membrane-restricted KL/KIT and an EPO signaling pathway. Several studies have investigated the relationship between KIT signaling and erythroid cell fate that may involve cross phosphorylation of EPOR following activation of KIT/KL signaling. Furthermore, data suggest that tyrosine cross-phosphorylation of EPOR is sustained longer when cells are cultured on steel stromal cells engineered to express the membrane-restricted form of KL than cells expressing the soluble form.

KIT is a member of the immunoglobulin-domain containing receptor family and contains 6 major autophosphorylation tyrosines (Y568, Y703, Y721, Y730, and Y900) on its intracellular domain.98 Upon KL/KIT interaction, these tyrosines are sequentially phosphorylated and dock proteins that activate specific signaling pathway. Y568 is the first Y to be phosphorylated and is the docking site for EPOR responsible for the KL-mediated EPOR transactivation discussed above. By interaction with EPOR, Y568 may indirectly activate STAT5. In fact, whether KL activates the STAT5 pathway in erythroid cells has been controversial. KL was found to be unable to activate STAT5 in prospectively isolated human erythroid progenitor cells but more recent single-cell analyses indicate that KL activates STAT5 in bipotent erythroid/megakaryocytic but not in myelomonocytic progenitor cells. This site also docks Src activating the MEK signaling.100 The MEK signaling is also activated directly by binding of MEK to Y703. In human and murine erythroid progenitors, KL induces rapid (within 15 minutes) activation of the MEK downstream effector ERK1/2, which lasts only 1 hour. In human erythroleukemic K562 and myeloid MO7e cells, the rapid KL-dependent ERK activation is associated with proliferation, whereas the late sustained ERK activation is responsible for differentiation. Y721 docks and activates PI3K, which in turn activates the mTOR downstream protein mTORC1 (promoting ribosomal protein biosynthesis) and mTORC2/AKT (promoting survival). Activation of the PI3K/ AKT pathway has been reported in murine erythroid progenitors and in human MO7e cells. Y721 also docks CD63, which prime KIT for endocytosis. By tuning the speed of endocytosis, CD63 regulates the strength of the overall response of erythroid cells to KL. Once in the cytoplasm, KIT is phosphorylated at Y900 which recruits CRKI/ CRKII and c-Cbl, posing the receptor for lysosomal degradation.103 Degradation of KIT triggers de novo KIT gene expression, maturation, and export to the cell surface for new KL interaction. Y730 is an interesting docking site because it engages PLCγ increasing interaction with EPOR and sustaining hyperproliferation of erythroid pro genitor cells. Y730 is phosphorylated upon KIT binding to stromal but not to soluble KL. These complex signaling pathways indicate that regulation of KIT metabolism is at least as important as that of its signaling in determining the extent of the response of the erythron to KL stimulation.

The soluble form of KL is produced by proteolytic cleavage of membrane KL and is released in the circulation. Soluble KL effectively supports erythroid maturation in vitro but is dispensable for steady-state erythropoiesis, because targeted mutant mice expressing exclusively the more stable membrane isoform of KL, KL2, lacking the major proteolytic cleavage site, have normal hematocrit values. However, these mice recover poorly from radiation-induced anemia. In wild-type mice, sublethal radiation induced a transient fourfold increase in KL in the serum (from 2 ng/mL), reaching a peak after 7 days. In contrast, the proteolytic-cleavage mutant KLKL2/KL2 mice did not release soluble KL into the serum after sub lethal radiation, and survival was significantly diminished because of anemia. This phenotype is remarkably similar to that of mice lacking the dimerization domain of glucocorticoid receptor (GR). The observation that soluble KL specifically stabilizes through its ERK1/2 signaling GRα in human erythroid cells suggests that soluble KL initiates the response to stress by priming erythroid cells to respond to glucocorticoids.

Besides cell–cytokine interactions, (paracrine) cell–cell adhesion and adhesion of cells to the extracellular matrix are important functions of the microenvironment. Perhaps most studied are the β1 integrins VLA4 and VLA5, which mediate the adherence of hematopoietic cells to stromal cells, fibronectin, or other components of the extracellular matrix. In β1 integrin/ko mice, hematopoietic stem cells fail to colonize the fetal liver during embryonic development. Conditional deletion of β1 integrin in adults leads to mild anemia with increased ROS formation and decreased RBC survival at steady state and in impairment in erythroid stress response (Fig.1). Milder responses to stress are seen with conditional deletion of α4 integrin. In addition to intrinsic effects on erythropoiesis, antibodies to VLA4 or to the VCAM-1 (a VLA4 ligand on endothelial cells) influence the retention of stem/progenitor cells (and many erythroblasts in case of anti-α4) in bone marrow and thereby impair their homing and lead to their mobilization in adult mice, in primates, and in MS patients. In in vitro studies, hematopoietic progenitors bind to specific domains of fibronectin in a differentiation-dependent manner (long-term culture initiating cell and day 12 CFU-spleen in mice adhere mainly through the heparin-binding domain and CS-1). BFU-E and other progenitor cells adhere to both the cell-binding (Arg-Gly-Asp-Ser [RGDS]) and heparin-binding domains, whereas CFU-E preferentially bind the RGDS sequence, and reticulocytes fail to adhere to fibronectin. This differential binding could influence the proliferation and especially the maturation and survival of developing erythroid cells, particularly under stress as well as the migration of progenitor cells in and out of the bone marrow cavity. Hematopoietic cytokines/chemokines present in the microenvironment can also modulate the affinity of β1 integrins for ligand, adding complexity to the regulation of erythropoiesis within the marrow microenvironment.

Fig1. STRESS ERYTHROPOIESIS AND THE MACROPHAGE NICHE. (A) Macrophage subsets in WT mice at steady state and during response to stress. Morphology and expression of F4/80, CD11b, and vascular cellular adhesion molecule-1 (VCAM-1) was analyzed in preparations of erythroid islands isolated from spleens of PHZ treated mice on day 4. (i): Hematoxylin staining of an erythroid island: note the different maturation stages of the erythroblasts that adhere to themselves and to the central macrophage engorged in damaged red cells after PHZ treatment; light green stain for F4/80 and magenta nuclear stain (assigned color) for surrounding erythroblasts (ii); light green F4/80 cells and DAPI-stained erythroblasts (iv); and green VCAM-1 positivity on central macrophages (v). (B) Important participating pathways leading to proliferative expansion of erythroid cells (mainly in the spleen) during stress in mice. Following genetic impairment of several pathways (i.e., Kit, GC-R, TR-α, and β1 integrins in a cell intrinsic manner), proliferative erythroid expansion in the spleen is severely compromised even in the presence of adequate levels of erythropoietin (EPO). It is possible that β1 integrins (specifically α5β1) intercept at least some of these known pathways, but the molecular details and cooperating/interacting molecules are missing (illustrated by dashed lines). Interpretations of the influence of these pathways and their signaling intermediates are complicated by the fact that these pathways have not been selectively eliminated in erythroid cells. Consequently, it is not clear at what stage of erythroid differentiation the FAK- or Src-dependent signaling is activated, early or late or both, and future studies should clarify these issues. (Part A from Ulyanova T, Phelps SR, Papayannopoulou T. The macrophage contribution to stress erythropoiesis: when less is enough. Blood. 2016;128:1756–1765. Part B modified from Ulyanova T, et al. Erythroid cells generated in the absence of specific β1-integrin heterodimers accumulate reactive oxygen species at homeostasis and are unable to mount effective antioxidant defenses. Haematologica. 2013;98[11]:1769–1777, Suppl Fig 3B.)

In histologic sections of normal marrow, islands of maturing erythroblasts (erythroblastic islands) often surround a central macrophage, termed a nurse cell. It has been suggested that the central macrophage supports terminal erythroid maturation by producing specific growth factors, assuring proper stage-specific iron supply, by establishing nanotubules which restrain mitochondria numbers before the process of mitophagy is in place, by providing a hub that promotes enucleation, and finally by engulfing and degrading the pyrenocytes. For this reason, the proteins that mediate the interaction between the two cell types have been the subject of intense investigation. Adhesion may be mediated through the binding of VLA4 (on erythroid cells) to VCAM-1 (on central macrophages), or through several other molecules, such as EMP discussed earlier. These molecular interactions have been identified as being critical for erythroblastic island integrity. In addition, the erythroblast ICAM-4 links erythroblasts to macrophages by interacting with αv integrin expressed in macro phages. Mice with targeted deletion of EMP are severely anemic and die at an embryonic stage, and ICAM-4null mice have markedly reduced erythroblastic islands. Of interest, RB-deficient macrophages do not bind RBnull erythroblasts, and failure of this interaction may mediate the defect in fetal liver erythropoiesis observed in RBnull mice. RB normally stimulates macrophage differentiation by counteracting inhibition of Id2 (a helix-loop-helix protein) on PU.1, a transcription factor crucial in macrophage differentiation. In addition to the aforementioned pathways, macrophage CD163 can serve as an erythroblast adhesion receptor in erythroblastic islands, promoting erythroid proliferation and/or survival. Whether a specific number or type of macrophages is required for optimal responses is not clear. This hypothesis is supported by recent data on single cell expression profiling of macrophages that have identified that at least three different macrophage populations are capable of interacting with erythroid cells and may promote their maturation. However, these studies are still at their infancy and have yet to generate definite conclusions regarding the existence on specialized “erythroid niches” and on their complex function. Of note, tissue macrophages express RNA for EPO and may also influence erythropoiesis through this mechanism. The picture of the role of macrophages in erythropoiesis is further complicated by the fact that during stress erythropoiesis, activation of GR induces stress-specific erythroid progenitor cells to generate specific types of macrophages that are capable of promoting their proliferation even in the presence of maximal concentrations of growth factors.

اخر الاخبار

اخبار العتبة العباسية المقدسة

سماحة السيد الصافي يؤكد ضرورة تنمية الموارد المالية لدائرة التقاعد الوطنية وتعظيمها

بالتزامن مع اليوم الوطني للقرآن الكريم المَجمَع العلميّ ينظم دورة تطويرية لأساتذة مشروع حفظ القرآن الكريم في كربلاء

ضمن فعاليات الملتقى الأكاديمي التربوي.. عقد جلسة حوارية عن فاعلية فلسفة التربية الإسلامية في تحديات العصر الرقمي

الملتقى الأكاديمي التربوي يشهد إقامة محاضرة تناقش مراحل تطور الإنسان من منظور إسلامي

المزيد

الآخبار الصحية

العلاج بالماء البارد.. باحثون يكشفون فوائده الصحية

من دون صالة رياضية.. عادة بسيطة لتحسين الصحة وخسارة الوزن

لصحة القلب وتفادي السرطان.. استهلك هذا النوع من الطعام

دراسة تحذر من خطر شرب القهوة في أكواب البلاستيك

المزيد

الآخبار التكنلوجية

إنجاز طبي تاريخي.. جراحة قلب بلا شق في الصدر

اليابان.. مراحيض توفر فحص للحالة الصحية عبر "تحليل البراز" !

اختراق واسع لمنصة "إنستغرام".. ونشر بيانات على "الدارك ويب"

علماء يحذرون: هكذا سيفنى كوكب الأرض

المزيد

مواضيع ذات صلة

Ontogeny of Erythropoiesis

Cellular Dynamics in Erythropoiesis: Steady State and Stress Erythropoiesis

Erythropoiesis and Iron Regulation

Intrinsic Control of Erythropoiesis:Erythropoietin

Nuclear Organization

Unfolded Protein Response

Autophagy

Nonapoptotic forms of cell death

The Cell Division Cycle

Stem Cells

Proliferation and the Cell Cycle

Growth Factors and Receptors