How Do You Mend a Broken Heart?
Stephen Emerson, M.D., Ph.D.
Kawada H, et al. Nonhematopoietic mesenchymal stem cells can be mobilized and differentiate into cardiomyocytes after myocardial infarction. Blood 2004; 104:3581.
In this paper, Kawada et al. show that some cell in whole bone marrow can be mobilized by G-CSF following myocardial injury and contribute to regenerating myocardial myocytes. However, purified mobilized primitive hematopoietic stem cells (HSCs) were not able to generate recovering cardiac myocytes following infarction. Finally, a stromal cell-derived pre-cardiac cell line could itself be mobilized and contribute to cardiac myocytes. All told, these data suggest that while mobilized HSCs may not be able to improve cardiac repair, stromal precursors mobilized in the same process can indeed produce cardiac myocytes, and perhaps improve cardiac function after myocardial infarction (MI).
Over the past five years, tremendous excitement has arisen over the potential for using bone marrow-derived stem cells to improve the recovery of injured tissues, including myocardium following infarction and brain cells following strokes. Not surprisingly, excitement has waxed and waned, depending on whether the last published paper showed promising or disappointing results. While conflicting data have accumulated from mouse, primate, and human studies, an intriguing hypothesis has arisen: mobilized HSCs may not develop into cardiac myocytes, but mobilized non-hematopoietic stromal elements may themselves improve myocardial function, either by producing functioning myocytes, or increasing fibroblast mass in supportive scar tissue.
To track cell fates following mobilization, the authors of this paper used donor bone marrow (BM), genetically modified to fluoresce bright green, allowing single donor-derived cells to be detected in tissue sections. Using this technology, they showed that some BM-derived cell mobilized by G-CSF form cardiac myocytes, but this cell did not appear to be a primitive HSC. They hypothesized that the cell responsible for cardiac repair in their mice was a mesenchymal stem cell (MSC), but unfortunately they did not have a convenient technique to homogeneously purify these cells to test their idea. As a substitute, they derived from isolated adherent cells a pre-cardiac cell strain, called cardiomygenic or CMG, which they also genetically engineered to fluoresce bright green for detection. First, they injected the CMG cells directly into the mouse BM to create a population of fluorescent adherent stromal cells. After cardiac injury and treatment with G-CSF, green fluorescing cells were mobilized and then found in the myocardium, often in cells costaining with myocardial markers such as α-actinin. Taking these three sets of experiments together, they argue that MSCs mobilized by G-CSF are able to differentiate into cardiac myocytes following injury. If this process were of sufficient magnitude, perhaps stem cells (albeit mesenchymal stem cells) could accelerate or improve recovery for MI patients.
If these results are confirmed, much more work needs to be done until these findings can be productively applied to the clinic. For example, techniques that maximize the mobilization of MSCs need to be developed. In addition, large animal studies need to occur to determine at what point after MI these mobilized MSCs would have maximal effect on cardiac regeneration. But perhaps, at least, we now have a functional cell type to test and track, so that one day, stem cells will help hearts mend.
Return to Top
Hematopoietic Stem Cells Run Home
Robert Lowsky, M.D., FRCPC
Christopherson II KW, Hangoc G, Mantel CR, Broxmeyer HE. Modulation of Hematopoietic Stem Cell Homing and Engraftment by CD26. Science 2004; 305:1000-1003.
The ability of intravenous infusions of hematopoietic stem cells (HSCs) to home to their bone marrow niche and provide short and long-term engraftment is the basis for clinical hematopoietic cell transplantation. It is assumed that only a small fraction of injected HSCs successfully home to hematopoietic sites, emphasizing the importance of cell numbers, particularly when the donor supply is limited (cord blood) or insufficient numbers of cells are collected from patients scheduled for autologous transplants. In this paper, Cristopherson et al. demonstrate that inhibition or deletion of CD26 on donor cells greatly increases their homing efficiency to recipient bone marrow niches, thereby providing an alternate means to enhance engraftment.
Following infusion of HSCs, local high levels of the chemoattractant CXCL12 (also known as stromal-derived factor-1; SDF-1) in the bone marrow are thought to promote homing into this tissue through the interaction with its receptor CXCR4 expressed on HSCs. CD26 is a membrane bound extracellular enzyme present on most HSCs that cleaves CXCL12 to negatively regulate HSC migration. To illustrate this point, the authors incubated murine HSC-enriched populations with inhibitors of CD26, or used HSCs from CD26 deficient mice, and demonstrated significantly improved homing efficiency using in vitro chemotaxis assays. Using a murine model of transplantation and limiting doses of donor cells, they confirmed that approximately 1-log fold fewer donor HSCs are needed to establish long-term hematopoietic cell engraftment and rescue lethally irradiated recipient animals if CD26-/- donor cells are the source of the graft. Finally, using a competitive repopulating assay, the authors demonstrated that six months after transplantation of equal numbers of CD26-/- donor cells and control competitor cells, there was a significantly increased contribution to chimerism from the experimental cells.
This report is significant as it implies that there are many more cells in the donor graft with HSC potential than previously appreciated, but that entry of these cells into the bone marrow from the circulation is prevented by virtue of their expression of CD26. If this is correct, then emphasis should focus on more detailed analyses of the specific cell types affected by CD26 inhibition. The expression of genes involved in self-renewal needs to be demonstrated in CD26-positive HSCs. In contrast, a recent report suggested that transplanted HSCs engraft with absolute efficiency but that most engrafted cells fail to sustain self-renewal, implying extinction events in the stem cell fraction (Benveniste et al. Nature Immunology July 2003). These differences may reflect that reported purifications are not yet complete. From a clinical perspective, studies to determine if CD26 blockade will translate to enhanced engraftment are needed. If this strategy is realized, it will bring the promise of successful transplantation when limited numbers of HSCs are available. It may offer a way to enable extending cord blood transplants from pediatric to adult patients and to improve outcomes for cancer patients from whom adequate number of HSCs cannot be obtained. In their paper, Christopherson et al. have contributed an important study to better understand how infusions of HSCs migrate and run home.
Return to Top
Does Gastric Cancer Originate in the Bone Marrow?
Andrew Schafer, M.D.
Houghton JM, Stoicov C, Nemura S, et al. Gastric cancer originating from bone marrow-derived cells. Science 2004; 306:1568-1571.
Mutations in the body's small supply of naturally-occurring stem cells may give rise to "cancer stem cells," which have been recently identified in various types of malignancies1. It has been assumed that these cancer stem cells originate in the specific tissues within which the tumor arises. For example, gastric cancers would be derived from cancer stem cells that are native to gastric epithelium. However, bone marrow-derived cells (BMDCs) are known to be recruited to sites of tissue injury and inflammation, and are therefore potential progenitors of solid tumors. In this paper in the November 26 issue of Science, Houghton et al. have shown that chronic gastric infection of mice with the ulcer- and cancer-causing bacterium, Helicobacter, induces repopulation of the stomach with BMDCs, which subsequently transform to epithelial cancer. Before infecting the mice with Helicobacter, the investigators lethally irradiated and transplanted them with bone marrow from mice expressing a genetically engineered marker (nonmammalian beta-galactosidase) that allowed the marrow cells to be distinguished from the recipients' own cells. After 20 weeks of infection, the labeled BMDCs began to seek out and repopulate the damaged gastric mucosa. In the setting of chronic gastric inflammation, these BMDCs then progressed through metaplasia and dysplasia to intraepithelial cancer. The experiment was also done by transplanting female mice with male bone marrow and showing that the resulting gastric cells that transformed to cancer had Y chromosomes.
These exciting and provocative findings offer the novel concept of bone marrow-derived progenitor cells homing to and repopulating chronically inflamed gastric mucosa and then transforming to gastric cancer. They lend credence to the controversial notion of cancers arising from "cancer stem cells," but with an intriguing new twist: that the stem cells originate in a different tissue (bone marrow) than the one in which the tumor arises2. The study should stimulate important lines of further research. It remains to be proven that the transplanted cells actually differentiate into gastric epithelial cells, as these authors suggest, rather than simply fusing with them, which is a tendency of bone marrow cells. This explanation would be a different, but no less interesting, new model of carcinogenesis. In light of the well-studied link between chronic infection, inflammation, and cancer (of which the association of H. pylori with gastric cancer is a prototype), it will be interesting to determine if BMDCs are likewise the source of other inflammation-linked cancers (such as hepatocellular, lung, cervical, esophageal, and ovarian cancers)3. This study also raises the theoretical concern that bone marrow stems that might be used in the future as regenerative therapy to treat heart disease and other conditions could also home in on other tissues that are chronically inflamed, in an attempt to repair them, and then undergo malignant transformation. The results of this study should lead to a radical reassessment of current assumptions about how all cancers originate and open new avenues for their diagnosis and treatment.
- Marx J. Mutant stem cells may
seed cancer. Science 2003;
301:1308-1310.
- Marx J. Bone marrow cells: the
source of gastric cancer? Science
2004; 306:1455-1457.
- Balkwill F, Mantovani A.
Inflammation and cancer: back to
Virchow? Lancet 2001; 357:539-545.
Return to Top
Making Cancer More Likely
Nancy Andrews, M.D., Ph.D.
Bond GL, Hu W, Bond EE, et al. A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell 2004; 199:591-602.
The tumor suppressor protein p53 serves a vital function in protecting cells from DNA damage and oncogene activation. Its importance is underscored by the fact that at least half of all human cancers carry mutations in the p53 gene. When a p53 mutation is passed genetically among family members, it leads to a marked, inherited predisposition to a variety of cancers, known as the Li-Fraumeni syndrome. Normally, p53 is controlled by other proteins, which modulate its activity. MDM2 is one of these proteins - it acts to put the brakes on p53. In this report, Bond et al. describe a common variation in the MDM2 gene, called SNP309, which causes more MDM2 to be produced. Under normal circumstances, this variation is inconsequential. But when p53 is already compromised by an inherited or a spontaneous mutation, increased production of MDM2 markedly accelerates the clinical appearance of tumors. Bond and colleagues show that about 12 percent of normal individuals carry two copies of the SNP309 version of the MDM2 gene. It occurs at the same frequency in Li-Fraumeni kindreds with p53 mutations, where this added genetic insult leads, on average, to a nine-year earlier appearance of a first tumor. It also predisposes Li-Fraumeni patients to a greater number of independent cancers over their lifetimes, apparently because the protective p53 pathway is severely disabled. Having two copies of the SNP309 version of MDM2 is also associated with much earlier onset of spontaneous malignancies in patients who have not inherited p53 mutations.
The authors have not yet addressed the implications for the 40 percent of normal individuals who carry one SNP309 copy of the MDM2 gene. However, based on animal studies and the results presented in this paper, it is reasonable to guess that they may also have an increased risk of tumors early in life.
Return to Top
Genetic Thrombophilia from A to Z
Roy Silverstein, M.D.
Van de Water N, Tan T, Ashton F, et al. Mutations within the protein Z-dependent protease inhibitor gene are associated with venous thromboembolic disease: a new form of thrombophilia. British J. Hematology 2004; 127:190-194.
Yin ZF, Huang ZF, Cui J, Fieher R, Lasky N, Ginsburg D, Broze GJ, Jr. Prothrombotic phenotype of protein Z deificiency. Proc Natl Acad Sci (USA) 2000; 97:6734-6738.
In the late 1970s, a novel vitamin K dependent protein was discovered and named Protein Z (PZ). Although genetically related to the proteins of the coagulation cascade, a role for PZ in hemostasis was unknown until recently, when PZ was found to bind to another protein, termed Z-dependent protease inhibitor (ZPI). ZPI is a serine protease inhibitor genetically related to antithrombin. The ZPI-PZ complex becomes tethered to the surface of activated blood and vascular cells by the vitamin K-dependent gla domain of PZ where it functions as a potent inhibitor of coagulation factor Xa. Interestingly, ZPI also binds heparin (independently of PZ) and the complex inhibits factor XIa. Thus ZPI down regulates clot formation at two important steps, and mutations within the ZPI gene are candidates for genetic thrombophilias. Van de Water and colleagues have now identified two common mutations in the ZPI gene predicted to cause loss of protein expression. In a study of 250 patients with venous thromboembolism (VTE) from Aukland Hospital in New Zealand, they found these mutations in 11 patients (4.4 percent) compared to only 2 controls (0.8 percent), representing a relative risk of 5.7 (95 percent confidence intervals of 1.25-26). The prevalence of ZPI mutations in patients with VTE could be even higher, as several additional polymorphisms were detected in the VTE subjects, although the population size of the study was not large enough to determine statistical significance. The study revealed that most of the thrombotic events in the subjects with ZPI mutations occurred in the setting of known additional risk, such as concomitant Factor V Leiden, oral contraceptive use, or surgery. This is consistent with a study in mice where PZ deficiency was shown to enhance thrombotic risk only in the setting of Factor V Leiden.
Normal hemostasis is maintained by a tightly orchestrated set of enzymes, cofactors, and inhibitors acting in a coordinated manner in response to vascular insults. Genetic mutations or polymorphisms that influence circulating levels or activities of these proteins contribute significantly to individual thrombotic risks. Well-known examples include null mutations in protein C and antithrombin, the prothrombin G20210A polymorphism, Factor V Leiden, and still uncharacterized genetic changes that lead to subtle but persistent increases in coagulation factors VIII, IX, and XI. The relative risk for VTE associated with single genetic thrombophilias is generally low (ranging from two - five), but is often synergistically increased in the setting of additional acquired risks, such as oral contraceptive use, cigarette smoking, and immobility. Further elucidation of genetic influences on clinical thrombotic risk is important as we move toward the goal of developing individualized, evidence-based strategies for primary and secondary prevention. The study reviewed here has now identified mutations in an obscure natural anticoagulant system as potential important contributors to VTE risk. Since age is a clear risk factor for VTE and the Aukland study excluded subjects older than age 60, it will be important to study the contribution of PZI mutations to thrombophilia in older individuals.
Return to Top
Are Routine Blood Transfusions Deadly?
J. Douglas Rizzo, M.D.
Rao SV, Jollis JG, Harrington RA, et al. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004; 292:1555-1562.
Transfusion to support a hemoglobin concentration greater than 8-10 mg/dl has become common medical practice, especially in patients with cardiac or critical illness. This practice derives largely from the belief that transfusions will increase oxygen delivery to ischemic tissue and improve clinical outcome. This paper may provoke re-consideration of this practice. Using data collected prospectively in three recent large interventional trials for acute coronary syndrome (ACS), the investigators evaluated the association between blood transfusions and mortality among patients who developed anemia during their initial hospital course. More than 24,000 patients had complete data regarding bleeding, transfusion, and outcomes; among these 10 percent (2401) were transfused at least one unit of whole blood or packed red cells. Transfused patients were more often older, female, black, and of lower median body weight, had more medical comorbidities, and presented with higher risk ACS. The median nadir hematocrit at transfusion was 29 percent. Using sophisticated modeling techniques to adjust for patients' baseline characteristics, bleeding and transfusion propensity, and nadir hematocrit, patients who received a transfusion had a 3.94-fold risk of 30-day mortality (95 percent CI: 3.26-4.75) compared to patients who were not transfused. Landmark analysis demonstrated a trend association between transfusion and increased 30-day mortality. Logistic regression models revealed an interaction between transfusion and nadir hematocrit such that patients transfused with a nadir hematocrit over 25 percent had significantly higher odds of death within 30 days.
How can we explain this adverse association between blood transfusion and short-term survival that opposes current wisdom? The findings derive from post-hoc analysis of prospective trial data and may be explained by confounding factors not measured in the clinical trials. The authors acknowledge they were unable to evaluate transfusion indications or appropriateness. Stored red cells are generally deplete of nitric oxide (NO) and may bind NO from the circulation, leading to vasoconstriction, platelet aggregation, and diminished oxygen delivery. Blood transfusion may transiently increase immunosuppression, exacerbate pulmonary edema, or increase inflammatory mediators that worsen myocardial ischemia. In a previously published randomized trial1 comparing restrictive and liberal transfusion strategies (Hgb < 7 vs. < 10 g/dL respectively) for patients with critical illness, there were significantly higher rates of myocardial infarction and pulmonary edema in the liberally transfused group, with no difference in 30-day all-cause mortality. Interestingly, a few studies of erythropoietic stimulants in cancer patients have also demonstrated adverse outcomes, possibly by affecting tumor biology2,3. While speculative, it could be hypothesized that transfusions or erythropoietic agents to improve anemia have an adverse affect on outcome through disruption of undefined physiologic mechanisms of anemia. Taken together, these data suggest a cautious approach to blood transfusion for coronary syndrome and other critically ill patients with hemoglobin concentrations greater than 7-8 g/dL until randomized clinical trials can clarify the benefits and harms more clearly.
-
Hebert PC, Wells G, Blajchman MA, et
al. A multicenter, randomized, controlled
clinical trial of transfusion requirements
in critical care. N Engl J Med
1999;340:409-417.
-
Henke M, Laszig R, Rube C, et al.
Erythropoietin to treat head and neck
cancer patients with anaemia undergoing
radiotherapy: randomized, double-blind,
placebo-controlled trial. Lancet
2003;362:1255-60.
-
Leyland-Jones B; BEST Investigators and
Study Group. Breast cancer trial with
erythropoietin terminated unexpectedly.
Lancet Oncol 2003;5: 206-7.
Return to Top
Immune Response and Clinical Course in Follicular Lymphoma
Peter Lee, M.D.
Dave SS, et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med. 2004; 351(21):2159-69
Non-Hodgkin's lymphoma (NHL) and chronic lymphocytic leukemia (CLL) are both diseases with heterogeneous clinical courses: some patients progress slowly over many years, while others have a much more aggressive course, often leading to early death. Clearly, biological heterogeneity exists within both diseases, neither of which are apparent by routine laboratory examinations. Microarray (gene expression) analysis has made significant contributions in recent years to the subclassification of certain cancers with correlation to clinical outcome (see the Mini-Review by Dr. Connors on page 7). Examples include the subclassification of diffuse large B cell lymphoma (DLBCL) and CLL into subgroups with different gene expression patterns which correlate with clinical course. In this paper by Dave et al., the authors tackled the heterogeneity in follicular lymphoma (FL). They analyzed samples from a total of 191 untreated FL patients using microarrays. Ninety-five specimens were used as a training set, from which a molecular predictor of survival was constructed, and then validated in an independent test set of the remaining 96 specimens. Two gene expression signature sets were generated based on survival of subjects in the training set - signature one positively correlated with survival, and signature two negatively correlated with survival. Signature one includes genes encoding T cell markers (e.g., CD7, CD8B1, ITK, LEF1, and STAT4) and genes highly expressed in macrophages (e.g., ACTN1 and TNFSF13B). Signature two includes genes preferentially expressed in macrophages, dendritic cells, or both (e.g., TLR5, FCGR1A, SEPT10, LGMN, and C3AR1). Together, these two signature sets successfully stratified subjects in the test set into four quartiles with widely disparate median survival of 13.6, 11.1, 10.8, and 3.9 years, independent of clinical prognostic variables (age, stage, number of extranodal sites, LDH, performance status, sex, B symptoms, tumor grade, and IPI score). While these genes all relate to immune responses, intriguingly, they do not arise from the lymphoma cells (CD19+), but rather nonmalignant immune cells (CD19-) within the biopsy specimens. This suggests that differences in biology of the host immune response - rather than the tumor cells themselves - may underlie differences in clinical course and outcome in FL.
As with DLBCL and CLL, this study demonstrates that microarray analysis is a useful tool to stratify FL patients into subgroups with different clinical courses, and thus may help identify patients for early treatment. Moreover, this study went further to identify the cells from which the gene expression differences arise. Tumor specimens consist of both malignant cells and nonmalignant cells, which include stromal and immune cells. Thus, microarray analyses of tumor specimens represent the composite gene expression of diverse cell types. Intriguingly, the cells which appear to underlie differences in clinical behavior of FL are the nonmalignant immune cells. This is reminiscent of a report last year which demonstrated that transition from MGUS to multiple myeloma is accompanied by loss of function of NKT cells in patients. Together, these reports highlight the interplay between the host immune response and tumor cells as a critical determinant in clinical outcome. As such, strategies which specifically modulate the functional status of nonmalignant immune cells in FL may hold significant promise.
Return to Top
MicroRNAs: A New Player in Human Malignancies?
Josef Prchal, M.D.
Calin GA, Liu CG, Sevignani C, et al. MicroRNA profiling reveals distinct signatures in B cell chronic lymphocytic leukemias. Proc Natl Acad Sci USA 2004; 101(32):11755-60.
Dr. Carlo Croce has been at the forefront of efforts to find chromosome defects associated with disease. His and other groups followed an observation that >50 percent of patients with chronic lymphocytic leukemia (CLL) have a deletion at the 13q14 locus. However, their efforts failed to find a deleted tumor suppressor gene or an aberrantly expressed proto-oncogene. They asked if this region might contain a gene with a non-conventional function, perhaps not even encoding a protein. Indeed, they found two sequences that encoded for microRNA (miRNA), a newly described class of 22-nucleotide non-coding RNAs whose presumed function is to bind and inactivate messenger RNAs to modulate gene expression1. In the present paper, they have followed this observation by investigating the possible involvement of miRNAs in human cancers. They mapped 186 miRNAs and found that these were frequently localized to the sites of nonrandom genetic alteration seen in various human malignancies. Here, they report that analyses of differential expression of 254 miRNA in normal and CLL tissue allowed them to distinguish CLL cells from normal lymphocytes and to separate the cells expressing Zap-70, a marker of CLL aggressivity.
The human genome project, while providing a valuable roadmap to our genetic heritage, also generated unexpected surprises. One of these was that although there are only ~30,000 genes that code for proteins, the diversity of existing proteins would exceed 100,000 genes. This situation is largely created by the alternative splicing of messenger RNAs. Some of these alternatively-processed gene transcripts are tissue specific, a process now beginning to be explored. The other unexpected finding was the discovery of genes not coding for proteins but only generating RNA transcripts, and specifically those coding for miRNAs. These constitute about 1 percent of all genes. What are miRNAs? miRNAs are 22 nucleotide non-coding RNAs whose presumed mechanism is to bind and inactivate messenger RNAs of as yet largely unidentified genes that modulate organismal development and lineage specific differentiation. Growing evidence suggests that miRNAs or their machinery play a role in some neurological diseases and non-hematopoietic cancers2. Clearly, the yet-to-be defined role of these molecules in human disease and the potential for their manipulation for therapeutic purposes will be of far-reaching importance. In the meantime, studies of miRNAs will rapidly increase our experimental armamentarium for better understanding of tissue differentiation and proliferation, human development, and yet another mechanism for the malignant transformation of normal cells.
-
Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions
and down-regulation of micro-RNA genes miR15 and miR16
at13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci
U S A 2002;99(24):15524-9.
-
Gong H, Liu CM, Liu DP, Liang CC. The role of small RNAs in
human diseases: Potential troublemaker and therapeutic
tools. Med Res Rev 2005; [Epub ahead of print].
Return to Top
Elegant Mechanisms to Protect Against Red Cell Oxidation
Kenneth Kaushansky, M.D.
Kong Y, Zhou S, Kihm AJ, Katein AM, Yu X, Gell DA, Mackay JP, Adachi K, Foster-Brown L, Louden CS, Gow AJ, Weiss MJ. Loss of alpha-hemoglobin-stabilizing protein impairs erythropoiesis and exacerbates beta-thalassemia. J Clin Invest 2004;114:1457-66.
Feng L, Gell DA, Zhou S, Gu L, Kong Y, Li J, Hu M, Yan N, Lee C, Rich AM, Armstrong RS, Lay PA, Gow AJ, Weiss MJ, Mackay JP, Shi Y. Molecular mechanism of AHSP-mediated stabilization of alpha-hemoglobin. Cell 2004;119:629-40.
Two years ago α-hemoglobin stabilizing protein (AHSP; also known as erythroid associated factor [ERAF]) was first identified as an erythroid specific protein in a screen of GATA-1 response genes. The protein was found to bind to α-globin but not β-globin or hemoglobin (Hgb)A, and it was shown to stabilize the former against oxidation and unfolding in vitro. Based on these earlier studies, in these two papers many of the same investigators have now demonstrated the in vivo relevance of the protein and the molecular mechanisms by which it acts.
In the first study, Kong and colleagues genetically eliminated AHSP to demonstrate its essential role in vivo in both health and disease. They found that AHSP null mice display modest erythrocyte poikilocytosis and a mild anemia characterized by Heinz bodies, a two-fold shortened erythrocyte survival, enhanced erythroblast apoptosis, splenic erythrophagocytosis, and an excess of oxidized globin chains attached to erythrocyte membranes. The mice could partially respond to the insult; the marrow and spleen displayed enhanced erythropoiesis characterized by a shift in erythroid precursor maturity, and erythrocyte polychromasia was seen in the peripheral blood. The erythrocytes from null mice were more sensitive to an oxidative insult (phenylhydrazine) than normal cells, consistent with the predicted effect of AHSP. When the AHSP null allele was crossed into a murine model of β-thalassemia intermedia, the severity of oxidative damage increased and the resting hemoglobin level fell substantially below that of the thalassemic control mice. In the second study, using x-ray crystallography to solve the tertiary structure of a AHSP-Fe(II) α-hemoglobin co-crystal, Feng and colleagues have elucidated the elegant mechanism by which AHSP protects α-Hgb from oxidation, and why it "gives up" its cargo to β-Hgb to form HgbA. AHSP binds to the G and H helices of α-Hgb, in a hydrophobic protein-protein interaction that mimics the interface between the α-and β-globin chains in HgbA. Moreover, binding of AHSP was found to facilitate the conversion of oxidized (Fe(III))α-Hgb to the reduced (Fe(II)) form. And since the area of contact between AHSP and α-Hgb is less than that between β-Hgb and α-Hgb, β-Hgb can displace AHSP from α-Hgb when present.
Multiple mechanisms exist in erythroid progenitors to quantitatively balance synthesis of the three components of HgbA, α-globin, β-globin, and heme. For example, Bach1 inhibits globin gene transcription under conditions of low heme availability, and the Heme Regulated Kinase inhibits globin synthesis under similar circumstances. And while α-globin genes are transcribed at levels in excess of those for β-globin, translation of the latter is more efficient. The reason for the tight coupling of α and β-globin chains is their inherent susceptibility to oxidation and, once oxidized, their potential to act as a powerful erythrocyte oxidants. This finding is the molecular explanation for the pathogenesis of the severe hemolytic state that characterizes the β-thalassemic syndromes; modest to severe reduction in β-globin chain synthesis leads to an excess of α-Hgb, which denatures and precipitates as inclusions, oxidizing erythroid precursor and red cell membrane proteins and lipids, leading to further cellular damage. Despite the elegant mechanisms in place to balance globin chain synthesis, under physiological conditions a slight excess of α-Hgb is produced and must be protected against oxidation; with the two cited studies, it is now clear that AHSP provides that function during erythropoiesis. Together, these two studies provide an elegant molecular explanation by which a highly unstable and potentially toxic but essential erythrocyte protein, α-Hgb monomer, can be transiently chaperoned to preclude it from attack, thereby preventing erythrocyte destruction until it can form the highly stable tetrameric HgbA. It should be apparent that manipulation of AHSP could provide novel approaches to ameliorating oxidative erythrocyte damage and improve the ineffective erythropoiesis that characterizes the thalassemias.
Return to Top
A NOTCH Forward in T-ALL
Michael Williams, M.D.
Weng AP, et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 2004; 306:269-271.
Weng AP, et al. Gain-of-function NOTCH1 mutations occur frequently in human T cell acute lymphoblastic leukemia. Blood 2004; 104(Part 1):5a. Abstract #4, American Society of Hematology 2004 Annual Meeting, Plenary Session presentation.
NOTCH1 encodes a transmembrane receptor essential for the differentiation of T lymphocytes from progenitor cells. The gene was originally identified as an activated component of the rare t(7;9) chromosomal translocation in human T-ALL. Given the importance of this receptor in T cell development, and its ability to induce T-ALL in a murine model, Weng and colleagues postulated a more general role for its deregulated expression in human T-ALL. They confirmed this by demonstrating that gain-of-function mutations were indeed present in five human T-ALL cell lines, primarily localized to the NOTCH1 heterodimerization and PEST domains. Turning their attention to clinical samples from newly diagnosed and untreated T-ALL patients, they sequenced these domains and found mutations in 58 percent of pediatric (n=187) and 62 percent of adult (n=37) cases. All subtypes of T-ALL were found to carry these mutations, whereas none were present in B-ALL. No clear correlates have been identified as yet between the presence of the mutations and the clinical presentation or patient outcome.
The characterization of these frequent NOTCH1 mutations provides important new information for T-ALL pathogenesis and identifies potential therapeutic targets. For example, the complex NOTCH1 signaling pathway includes a cleavage step mediated by gamma-secretase; pharmacologic inhibitors of this enzyme have already been developed due to the proposed involvement of gamma-secretase in the pathogenesis of Alzheimer's disease. Clinical trials to test these agents in T-ALL are now being initiated and offer the hope for novel therapeutic approaches for patients with newly diagnosed and relapsed or refractory disease. The work by Dr. Weng and colleagues exemplifies once again the important insights to be gained via detailed molecular analysis of chromosomal translocations in hematologic malignancy.
Return to Top
The Benefit of a Double-Edged Sword in the Treatment of Antiphospholipid Syndrome
Lilli Petruzzelli, M.D., Ph.D.
Girardi G, Redecha P, Salmon JE. Heparin prevents antiphospholipid antibody-induced fetal loss by inhibiting complement activation. Nature Medicine 2004; 10(11):1222-1226.
Treatment for antiphospholipid syndrome that is defined by recurrent thrombosis and fetal loss in the setting of antiphospholipid antibodies relies heavily on the use of anticoagulant therapy. In this report, the investigators revisit the role of heparin in the treatment of the syndrome and shed light on why efficacy may not reflect anticoagulant dosing. In a mouse model of antiphospholipid syndrome that is induced by the passive transfer of human antiphospholipid (aPL) antibodies, complement activation was shown to be an essential component of pregnancy loss and diminished fetal growth.1 Treatment of mice that received the aPL antibodies to induce antiphospholipid syndrome with either unfractionated or low-molecular-weight heparin resulted in a marked reduction of fetal resorption or loss. In contrast, agents with minimal or low antithrombin binding, such as fondaparinux sodium or the direct thrombin inhibitor hirudin that acts independently of antithrombin, were not able to quell the fetal loss of mice in which the antiphospholipid syndrome was induced. This correlated with inhibition of complement activation induced by antiphospholipid antibodies. 
In our era of rapid progress toward the development of new anticoagulants, this report demonstrates that a critical understanding of the underlying disease, as well as analysis of the underlying mechanism of how tried and true agents work, requires continued probing. Heparin is proposed to interfere with complement activation at several different levels of the classical pathway, stimulation of the alternative pathway, and in the generation of secondary inflammatory mediators that fuel the process. Although this study uses a mouse model, the investigators demonstrate both clinical efficacy and in vitro evidence that the effects of heparin may well be on the modulation of the complement cascade in the induction of fetal demise that is associated with antiphospholipid antibody syndrome. The findings reveal decreased deposition of C3 in the deciduas of mice but show no effect on antiphospolipid antibody binding. This supports a model where complement deposition and activation are the critical mediators of fetal loss. Doses of heparin below those required for anticoagulation result in reduced fetal demise. Although a dose response course was not assessed, the inhibition of complement might explain why patients on doses of heparin that do not measurably affect the aPTT are protected from fetal loss. Although C3 deposition was measured, it is possible that other steps in the complement cascade - such as the activation of C1, C2, and C4, stimulation of the alternative pathway, and/or the activation of cells such as neutrophils and macrophages - are modulated by heparin in this system. Nonetheless, the report brings forth two important lessons: 1) the treatment of thrombotic complications may only be the tip of the iceberg in vascular damage, and the inflammatory component is a more general component that should be assessed; 2) some of clinical agents may be efficacious not because of their narrow spectrum but because of the broad nature of the biologic processes that they modulate. This should be kept in mind as we move toward more targeted antithrombotic therapy.
- Giarardi G, et. al. J Clin Invest 2003; 112(11)
1644-654.
Return to Top
Return to Table of Contents
|
