Evidence for linkage of familial Diamond-Blackfan anemia to chromosome 8p23.3-p22 and for non-19q non-8p disease.

Diamond-Blackfan anemia (DBA) is a rare congenital hypoplastic anemia that usually presents early in infancy and is inherited in 10% to 20% of cases. Linkage analysis has shown that DBA in many of both dominant and recessive DBA families mapped to chromosome 19q13.2 leading to the cloning of a gene on chromosome 19q13.2 that encodes a ribosomal protein, RPS19. However, subsequently, mutations of the RPS19 gene have only been identified in 25% of all patients with DBA. This study analyzed 14 multiplex DBA families, 9 of which had 19q13.2 haplotypes inconsistent with 19q linkage. A genome-wide search for linked loci suggested the presence of a second DBA locus in a 26.4-centimorgan (cM) interval on human chromosome 8p. Subsequently, 24 additional DBA families were ascertained and all 38 families were analyzed with additional polymorphic markers on chromosome 8p. In total, 18 of 38 families were consistent with linkage to chromosome 8p with a maximal LOD score with heterogeneity of 3.55 at D8S277 assuming 90% penetrance. The results indicate the existence of a second DBA gene in the 26.4-cM telomeric region of human chromosome 8p23.3-p22, most likely within an 8.1-cM interval flanked by D8S518 and D8S1825. Seven families were inconsistent with linkage to 8p or 19q and did not reveal mutations in the RPS19 gene, suggesting further genetic heterogeneity. (Blood. 2001;97:2145-2150)

Maturation and lineage-specific expression of the coxsackie and adenovirus receptor in hematopoietic cells.

Adenovirus vectors have been used to transfer genes into both hematopoietic progenitor cells and tumor cells, including carcinoma cells that have metastasized to bone marrow (BM). However, the relative susceptibility of different subsets of hematopoietic cells is unknown. In permissive cells adenoviral-mediated gene transfer is mediated by the coxsackievirus and adenovirus receptor (CAR) protein and alpha(v) integrins expressed on the cell surface of the target cells. This prompted us to investigate the expression of CAR on subpopulations of hematopoietic cells, determine whether this protein played a role in adenovirus-mediated gene transfer of hematopoietic cells and whether we could modulate CAR to enhance gene transfer efficiency. In this report we show that CAR is expressed on approximately 40% of all human BM cells, including erythroid and myeloid cells, but not lymphoid cells. Of the CD34(+) cells, 10%-15% expressed CAR, but this did not include most colony-forming progenitor cells, nor the most primitive CD38(-) subpopulation. The presence of CAR correlated well with gene transfer efficiency, but we were unable to induce CAR expression on immature, noncommitted progenitor cells. In conclusion, our results show that primitive hematopoietic progenitor cells lack CAR expression, but that expression is acquired during erythroid and myeloid differentiation.

Diamond-Blackfan anemia.

Diamond Blackfan anemia is a rare congenital hypoplastic anemia that usually presents early in infancy. Congenital anomalies, in particular of the head and upper limbs, are present in about 25% of reported patients. The disease is characterized by a moderate to severe macrocytic anemia, occasional neutropenia or thrombocytosis, a normocellular bone marrow with erythroid hypoplasia, and an increased risk of developing leukemia. Recent genetic studies have led to the identification of mutations in the ribosomal protein RPS19 in approximately 25% of sporadic and familial cases, a second gene on chromosome 8p, and evidence for an additional locus (or loci). The pathogenesis is unknown. The majority of patients respond to prednisone, and often erythropoiesis can be maintained with low doses of the drug. Both remissions and increased resistance to steroid treatment can occur. Patients who do not respond to treatment are usually transfusion dependent, although responses to high dose steroid, androgen, and interleukin-3 have been observed. Bone marrow transplantation can be curative.

Gene dose-dependent control of hematopoiesis and hematologic tumor suppression by CBP.

Mice with monoallelic inactivation of the CBP gene develop highly penetrant, multilineage defects in hematopoietic differentiation and, with advancing age, an increased incidence of hematologic malignancies. The latter are characterized, at least in some cases, by loss of heterozygosity (LOH) at the CBP locus. No such pathology was observed in wild-type or p300 heterozygous null mice of the same age and genetic background. Thus, a full complement of CBP, but not p300, is required for normal hematopoietic differentiation. These results also provide the first experimental evidence for the hypothesis that CBP has tumor-suppressing activity.

Essential role for the p55 tumor necrosis factor receptor in regulating hematopoiesis at a stem cell level.

Hematopoietic stem cell (HSC) self-renewal is a complicated process, and its regulatory mechanisms are poorly understood. Previous studies have identified tumor necrosis factor (TNF)-alpha as a pleiotropic cytokine, which, among other actions, prevents various hematopoietic progenitor cells from proliferating and differentiating in vitro. However, its role in regulating long-term repopulating HSCs in vivo has not been investigated. In this study, mice deficient for the p55 or the p75 subunit of the TNF receptor were analyzed in a variety of hematopoietic progenitor and stem cell assays. In older p55(-/-) mice (>6 mo), we identified significant differences in their hematopoietic system compared with age-matched p75(-/-) or wild-type counterparts. Increased marrow cellularity and increased numbers of myeloid and erythroid colony-forming progenitor cells (CFCs), paralleled by elevated peripheral blood cell counts, were found in p55-deficient mice. In contrast to the increased myeloid compartment, pre-B CFCs were deficient in older p55(-/-) mice. In addition, a fourfold decrease in the number of HSCs could be demonstrated in a competitive repopulating assay. Secondary transplantations of marrow cells from primary recipients of p55(-/-) marrow revealed impaired self-renewal ability of p55-deficient HSCs. These data show that, in vivo, signaling through the p55 subunit of the TNF receptor is essential for regulating hematopoiesis at the stem cell level.

One-day ex vivo culture allows effective gene transfer into human nonobese diabetic/severe combined immune-deficient repopulating cells using high-titer vesicular stomatitis virus G protein pseudotyped retrovirus.

Retrovirus-mediated gene transfer into long-lived human pluripotent hematopoietic stem cells (HSCs) is a widely sought but elusive goal. A major problem is the quiescent nature of most HSCs, with the perceived requirement for ex vivo prestimulation in cytokines to induce stem cell cycling and allow stable gene integration. However, ex vivo culture may impair stem cell function, and could explain the disappointing clinical results in many current gene transfer trials. To address this possibility, we examined the ex vivo survival of nonobese diabetic/severe combined immune-deficient (NOD/SCID) repopulating cells (SRCs) over 3 days. After 1 day of culture, the SRC number and proliferation declined twofold, and was further reduced by day 3; self-renewal was only detectable in noncultured cells. To determine if the period of ex vivo culture could be shortened, we used a vesicular stomatitis virus G protein (VSV-G) pseudotyped retrovirus vector that was concentrated to high titer. The results showed that gene transfer rates were similar without or with 48 hours prestimulation. Thus, the use of high-titer VSV-G pseudotyped retrovirus may minimize the loss of HSCs during culture, because efficient gene transfer can be obtained without the need for extended ex vivo culture.

Phase I/II trial of PIXY321 (granulocyte-macrophage colony stimulating factor/interleukin-3 fusion protein) for treatment of inherited and acquired marrow failure syndromes.

Fourteen paediatric patients with advanced amegakaryocytic thrombocytopenia (AMT) or other bone marrow (BM) failure syndromes were enrolled on one of two phase I/II dose escalation studies of PIXY321. PIXY321 was administered subcutaneously in doses ranging from 250 to 750 mg/m2/d. No dose-limiting toxicity was observed. Peak absolute neutrophil count (ANC) was higher than baseline in all patients. Most transfusion-independent patients demonstrated elevation in haematocrit and/or platelet count. Trilineage haemopoietic responsiveness was evident in the three transfusion-independent patients. In these paediatric populations PIXY321 is well tolerated and merits consideration as a potential therapy.

Identification of distinct elements of the stromal microenvironment that control human hematopoietic stem/progenitor cell growth and differentiation.

Using a novel collection of conditionally immortalized mouse stromal cell clones, we evaluated the role of distinct elements of the hematopoietic microenvironment in supporting and regulating the growth, division, and differentiation of a candidate human stem cell population (CD34+/CD38-). We found functional diversity in the capacity of different stromal cell clones to support the growth of primitive (CD34+/CD38-) and committed (CD34+/CD38+) hematopoietic progenitors and their differentiation into mature hematopoietic cells (CD34-/CD45+). Among the stromal cell clones that supported long-term hematopoiesis, we identified two clones that induced expansion of CD34+ progenitor/stem cells during the first 4 weeks of coculture and that supported the maintenance of this CD34+ population for up to 10 weeks in vitro. However, these two clones appeared to represent two different microenvironments with regard to the signals they provide to the different CD34+ progenitor subpopulations: One stromal clone preserved a pool of undifferentiated, relatively quiescent (CD34+/CD38-) progenitor cells, allowing their differentiation at a low rate into more committed (CD34+/CD38+) progenitors; the other fostered a more extensive and rapid differentiation of all CD34+/CD38- progenitors into CD34+/CD38+ cells, preferentially maintaining this committed population at a higher rate of cell division. These stromal cell clones were also able to support the proliferation and differentiation of CD34+/CD38- cells in conditions in which progenitor-stroma contact was prevented. This collection of stromal cell clones may represent a unique tool for the study of stromal regulators of hematopoiesis as well as for the support of gene transfer into hematopoietic progenitor cells.

Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species.

We previously described a method for isolating murine hematopoietic stem cells capable of reconstituting lethally irradiated recipients, which depends solely on dual-wavelength flow cytometric analysis of murine bone marrow cells stained with the fluorescent DNA-binding dye Hoechst 33342. This method, which appears to rely on the differential ability of stem cells to efflux the Hoechst dye, defines an extremely small and homogeneous population of cells (termed SP cells). We show here that dual-wavelength analysis of Hoechst dye-stained human, rhesus and miniature swine bone marrow cells reveals a small, distinct population of cells that efflux the dye in a manner identical to murine SP cells. Like the murine SP cells, both human and rhesus SP cells are primarily CD34-negative and lineage marker-negative. In vitro culture studies demonstrated that rhesus SP cells are highly enriched for long-term culture-initiating cells (LTC-ICs), an indicator of primitive hematopoietic cells, and have the capacity for differentiation into T cells. Although rhesus SP cells do not initially possess any hematopoietic colony-forming capability, they acquire the ability to form colonies after long-term culture on bone marrow stroma, coincident with their conversion to a CD34-positive phenotype. These studies suggest the existence of a hitherto unrecognized population of hematopoietic stem cells that lack the CD34 surface marker classically associated with primitive hematopoietic cells.

The beta chain of the interleukin-3 receptor functionally associates with the erythropoietin receptor.

Interleukin-3 (IL-3) and granulocyte-macrophage colony-stimulating factor (GM-CSF) receptors share a common beta chain (beta(c)), and both cytokines enhance erythropoietin (Epo)-dependent in vitro erythropoiesis by primary hematopoietic progenitors and factor-dependent cells. These data suggest that the Epo receptor and beta(c) may functionally interact. To determine whether such interactions can be documented, we studied a murine factor-dependent cell line (Ba/F3), which endogenously expresses IL-3R. First, Ba/F3 cells were transfected with murine EpoR, making them responsive to both IL-3 and Epo. Next, the EpoR expressing cells were transfected with murine beta(c). This resulted in an enhanced sensitivity of these cells to Epo, which was especially pronounced at low Epo concentrations. Ba/F3-EpoR were then treated with antisense oligodeoxynucleotides to the murine beta. Control sense and nonsense had no effect on Epo-dependent growth, but the antisense markedly and specifically inhibited Epo-dependent growth. In contrast, the antisense did not affect beta-globin message levels (another Epo-responsive effect in these cells) detectable by Northern blot. Finally, Western blot analysis of proteins immunoprecipitated from cells expressing both receptors with antibody against beta and blotted with antibody against EpoR, or immunoprecipitated with antibody against EpoR and blotted with antibody against beta, showed that EpoR and beta coimmunoprecipitate. These data show that the beta chain functionally and physically associates with the EpoR. This suggests that these cytokine receptors exist as a large supercomplex and offers the first molecular explanation for the synergistic effects of IL-3 and GM-CSF with Epo during erythropoiesis.

Diamond-Blackfan anemia.

Diamond-Blackfan anemia (DBA) is a rare, congenital, hypoplastic anemia that usually presents in early infancy. Congenital anomalies, particularly of the head and upper limbs, are present in about a quarter of reported patients. The disease is characterized by a moderate-to-severe macrocytic anemia, occasional neutropenia or thrombocytosis, a normocellular bone marrow with erythroid hypoplasia, and an increased risk of developing leukemia. The pathogenesis is unknown. The majority of patients respond to prednisone, and often erythropoiesis can be maintained with low doses of the drug. Both remissions and increased resistance to steroid treatment can occur. Nonresponders usually are transfusion dependent, although responses to high dose steroid, androgen, and interleukin-3 have been observed. Bone marrow transplantation can be curative.

The beta c component of the granulocyte-macrophage colony-stimulating factor (GM-CSF)/interleukin 3 (IL-3)/IL-5 receptor interacts with a hybrid GM-CSF/erythropoietin receptor to influence proliferation and beta-globin mRNA expression.

BACKGROUND: The interaction of different members of the hematopoietic growth factor receptor family may be relevant to the increased proliferation and the failure of differentiation that characterizes the myeloid leukemias. We recently demonstrated that a chimeric receptor (GMER) that is composed of the extracellular and transmembrane domains of the human granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor alpha-chain (GMR alpha) and the cytoplasmic domain of the murine erythropoietin receptor mEpoR binds hGM-CSF with low affinity (3 nM) and confers both proliferative and differentiation signals to stably transfected murine Ba/F3 cells. MATERIALS AND METHODS: To investigate whether the common beta-subunit of the GM-CSF receptor (beta c) can interact with GMER, either the entire beta-subunit or a mutant, truncated beta-subunit that completely lacks the cytoplasmic domain (beta tr) was introduced into Ba/F3 cells that express GMER, and the binding of GM-CSF as well as proliferation and differentiation responses were measured. RESULTS: Scatchard analysis showed that both GMER + beta c and GMER + beta tr bound hGM-CSF with high affinity (Kd 40 pM to 65 pM). Proliferation assays showed that the maximum growth of cells expressing GMER + beta c was identical to that of cells with GMER alone. However, proliferation of the cells that expressed GMER + beta tr was reduced by 80-95% of GMER. Dose-response curves showed that the concentration of GM-CSF required for half-maximal growth was 0.5-5.0 pM for GMER + beta c and 0.5-5 nM for GMER and GMER + beta tr. The EpoR cytoplasmic domain of GMER also undergoes ligandinducible tyrosine phosphorylation. However, the tyrosine phosphorylation did not correlate with growth in cells expressing beta tr. Coexpression of beta c with GMER in Ba/F3 cells grown in hGM-CSF markedly enhanced beta-globin mRNA expression. CONCLUSIONS: These results indicate that beta c can transduce a unique signal in association with GMER to influence both proliferative and differentiation signal pathways.

A constitutively activated chimeric cytokine receptor confers factor-independent growth in hematopoietic cell lines.

The high-affinity receptor for granulocyte-macrophage colony-stimulating factor (GMR) comprises at least 2 distinct subunits, alpha and beta common (beta c), whereas the normal erythropoietin receptor (nEpoR) comprises only one known subunit. An arginine to cysteine (R129C) mutation of the extracytoplasmic domain of the murine EpoR leads to Epo-independent growth in transduced cells (cEpoR). To investigate the proliferative functions of the cytoplasmic regions of each GMR subunit separately and the potential of the R129C EpoR mutation to induce factor-independent growth through heterologous receptor regions, we constructed four hybrid receptors: the extracellular region of either murine nEpoR or cEpoR linked to the transmembrane and cytoplasmic regions of either the human GMR alpha or beta c subunit (nE alpha, nE beta, cE alpha, and cE beta). We then expressed them in an interleukin-3-dependent murine cell line, Ba/F3. Expression of nE beta led to Epo-dependent growth, whereas expression of cE beta conferred factor-independent growth. Surprisingly, expression of cE alpha also resulted in factor-independent cell growth, whereas nE alpha did not respond to Epo. Furthermore, the functional hybrid receptors showed Epo-dependent (nE beta) or constitutive (cE alpha and cE beta) tyrosine phosphorylation of the cytoplasmic kinases JAK1 and JAK2. We reasoned that the proliferative signal of cE alpha was transduced either through the alpha tail itself or through an accessory protein such as the endogenous murine beta common subunit (mu beta c). To distinguish these possibilities, the chimeric receptor cE alpha was expressed in the interleukin-2-dependent murine cell line, CTLL-2, that does not express mu beta c. cE alpha did not induce cell growth in CTLL-2; however, when mu beta c was coexpressed with cE alpha in CTLL-2, factor-independent growth was reconstituted. In conclusion, the cytoplasmic domain of the GMR alpha subunit requires a beta chain for transduction of a proliferative signal. Furthermore, the R129C EpoR mutation can constitutively activate heterologous receptors to mediate factor-independent proliferation.

Evidence for a physical association between the Shc-PTB domain and the beta c chain of the granulocyte-macrophage colony-stimulating factor receptor.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) regulates the growth and function of several myeloid cell types at different stages of maturation. The effects of GM-CSF are mediated through a high affinity receptor that is composed of two chains: a unique, ligand-specific alpha chain and a beta common chain (beta c) that is also a component of the receptors for interleukin 3 (IL-3) and IL-5. Beta c plays an essential role in the transduction of extra cellular signals to the nucleus through its recruitment of secondary messengers. Several downstream signaling events induced by GM-CSF stimulation have been described, including activation of tyrosine kinases and tyrosine phosphorylation of cellular proteins (including beta c) and activation of the Ras/mitogen-activated protein kinase and the JAK/STAT pathways. A region within the beta c cytoplasmic tail (amino acids 517-763) has been reported to be necessary for tyrosine phosphorylation of the adapter protein, Shc, and for the subsequent GM-CSF-induced activation of Ras. In this paper, we describe a physical association between the tyrosine phosphorylated GM-CSF receptor (GMR)-beta c chain and Shc in vivo. Using a series of cytoplasmic truncation mutants of beta c and various mutant Shc proteins, we demonstrate that the N-terminal phosphotyrosine-binding (PTB) domain of Shc binds to a short region of beta c (amino acids 549-656) that contains Tyr577. Addition of a specific phosphopeptide encoding amino acids surrounding this tyrosine inhibited the interaction between beta c and shc. Moreover, mutation of a key residue within the phosphotyrosine binding pocket of the Shc-PTB domain abrogated its association with beta c. These observations provide an explanation for the previously described requirement for Tyr577 of beta c for GM-CSF-induced tyrosine phosphorylation of Shc and have implications for Ras activation through the GM-CSF, IL-3, and IL-5 receptors.

Ligand-dependent repression of the erythroid transcription factor GATA-1 by the estrogen receptor.

High-dose estrogen administration induces anemia in mammals. In chickens, estrogens stimulate outgrowth of bone marrow-derived erythroid progenitor cells and delay their maturation. This delay is associated with down-regulation of many erythroid cell-specific genes, including alpha- and beta-globin, band 3, band 4.1, and the erythroid cell-specific histone H5. We show here that estrogens also reduce the number of erythroid progenitor cells in primary human bone marrow cultures. To address potential mechanisms by which estrogens suppress erythropoiesis, we have examined their effects on GATA-1, an erythroid transcription factor that participates in the regulation of the majority of erythroid cell-specific genes and is necessary for full maturation of erythrocytes. We demonstrate that the transcriptional activity of GATA-1 is strongly repressed by the estrogen receptor (ER) in a ligand-dependent manner and that this repression is reversible in the presence of 4-hydroxytamoxifen. ER-mediated repression of GATA-1 activity occurs on an artificial promoter containing a single GATA-binding site, as well as in the context of an intact promoter which is normally regulated by GATA-1. GATA-1 and ER bind to each other in vitro in the absence of DNA. In coimmunoprecipitation experiments using transfected COS cells, GATA-1 and ER associate in a ligand-dependent manner. Mapping experiments indicate that GATA-1 and the ER form at least two contacts, which involve the finger region and the N-terminal activation domain of GATA-1. We speculate that estrogens exert effects on erythropoiesis by modulating GATA-1 activity through protein-protein interaction with the ER. Interference with GATA-binding proteins may be one mechanism by which steroid hormones modulate cellular differentiation.

Clonality in juvenile chronic myelogenous leukemia.

Juvenile chronic myelogenous leukemia (JCML) is a myeloproliferative disease in which morbidity and mortality are primarily caused by nonhematopoietic organ failure from myelomonocytic infiltration or by failure of the normal bone marrow. Morphologic evidence of maturation arrest, karyotypic abnormalities, and progression to blast crisis are infrequent events. Viral infections and other reactive processes can initially mimic the clinical course of JCML, creating diagnostic problems. Because of the rarity of JCML and technical limitations, formal clonality studies have not been reported previously. Nine female JCML patients were identified by clinical criteria, characteristic 'spontaneous' in vitro cell growth, and negative cultures and titers for various viral agents. Peripheral blood and bone marrow samples were obtained at the time of diagnosis for cell separation and RNA and DNA isolation. To assess clonality, X-chromosome inactivation patterns were evaluated using three different, recently developed polymerase chain reaction-based clonality assays. All nine female JCML patients showed evidence for monoclonal origin of mononuclear cells at the time of diagnosis. Cell separation studies further traced the monoclonal origin back to at least the most primitive myeloid progenitor cell. Reversion to a polyclonal state was demonstrated after bone marrow transplant and also in one patient following treatment with 13-cis retinoic acid. This demonstration of clonality in JCML delineates it from the reactive processes and provides a basis for molecular genetic strategies to identify causally associated mutations.

Expression and function of the human granulocyte-macrophage colony-stimulating factor receptor alpha subunit.

To determine the expression and function of the granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor alpha chain (GMR alpha) during hematopoiesis and on leukemic cells, monoclonal antibodies were raised by immunizing mice with cells expressing high levels of human GMR alpha. A pool of five antibodies isolated from three different mice was used to characterize GMR alpha. This antibody pool (anti-GMR alpha) immunoprecipitated a protein with the expected molecular weight of GMR alpha from COS cells transiently transfected with the GMR alpha gene. In factor-dependent cells, GMR alpha existed as a phosphoprotein. However, its phosphorylation was not stimulated by the presence of GM-CSF. Anti-GMR alpha inhibited the GM-CSF-dependent growth of cell lines and normal bone marrow cells and inhibited the binding of iodinated GM-CSF to its receptor. Cell surface expression of GMR alpha was examined using anti-GMR alpha and flow cytometry. GMR alpha was readily detectable on both blood monocytes and neutrophils. In adherence-depleted normal bone marrow, two separate populations expressed GMR alpha. The most positive cells were predominantly macrophages, whereas the cells that expressed less GMR alpha were largely myelocytes and metamyelocytes. A small population of lin-CD34+ or CD34+CD38- cells also expressed GMR alpha, but they were not capable of significant growth in colony-forming assays. In contrast, the majority of lin-CD34+ and CD34+CD38- cells were GMR alpha-, yet they produced large numbers of myeloid and erythroid colonies in the same assay. Malignant cells from patients with leukemia were also tested for GMR alpha expression. All of the myeloid leukemias and only rare lymphoid leukemias surveyed tested positive for GMR alpha. These results show that anti-GMR alpha is useful for the functional characterization of the GMR alpha and for the detection of myeloid leukemia and that GMR alpha is expressed on certain lineages throughout hematopoietic development; however, progenitors that express the receptor may have a reduced capacity to proliferate in response to hematopoietic growth factors.

Hematopoietic cell proliferation and differentiation.

Most hematopoietic stem cells are quiescent in the G0 stage of the cell cycle. Certain combinations of hematopoietic growth factors can bind to their cognate receptors on the stem cell surface, shorten the G0 stage of the cell cycle, and stimulate cell division. Hematopoietic growth factors also act on committed progenitors to increase their survival and to amplify maturing populations by stimulating proliferation. Although differentiation is probably determined largely by an intrinsic program, there is evidence that certain receptors expressed on more mature cells can also direct differentiation. This view is still controversial, as is the view that hematopoietic growth factors act merely to prevent apoptosis. Different members of the hematopoietic growth factor family of receptors recruit overlapping sets of tyrosine kinases and share common signal transduction pathways such as Ras. They also directly activate a cytoplasmic class of transcription factors that then translocate to the nucleus. The development of committed hematopoietic progenitor cells into mature effector cells is regulated by "master" transcription factors that recognize motifs in the promoters of many genes that encode lineage-specific proteins.