| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Program in Cell Dynamics, Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, Worcester, Massachusetts, USA
1Correspondence: Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, 377 Plantation St., Worcester, MA 01605, USA. E-mail: thoru.pederson{at}umassmed.edu
ABSTRACT
The inherited bone marrow failure syndromes are clinically distinct but share some common features. Difficult to treat and typified by a poor prognosis, their pathogenesis is unknown. Recent findings that some patients with the erythroblastopenia Diamond-Blackfan anemia (DBA) have mutations in ribosomal proteins have led to the idea that this and perhaps other bone marrow failure disorders result from an inadequate supply of normally functioning ribosomes. According to this hypothesis, an insufficiency of the protein synthetic capacity limits the replicative potential of cells, with the DBA disease phenotype in particular arising from a block of one or more of the two to four critical, temporally compressed cell divisions in the differentiation program of the erythroid lineage in the fetal liver and the postnatal bone marrow. Here I propose an alternative (but not mutually exclusive) hypothesis centered on nucleoli: the specialized intranuclear domains within which ribosomes are assembled. It was recently discovered that the nucleoli contain cell cycle machinery in close proximity to nascent ribosomes. Although mutations in ribosomal proteins might be expected to negatively influence the cell’s protein synthetic capacity, I suggest it is also possible that the DBA mutations directly affect the nucleolus to destabilize or otherwise deregulate the coresident cell cycle machinery. This hypothesis envisions that the ribosomal protein mutations discovered in DBA act upstream from ribosome assembly by interfering with the staging of cell cycle progression machinery in the nucleolus, in a pretranslational mode of pathogenesis.—Pederson, T. Ribosomal protein mutations in Diamond-Blackfan anemia: might they operate upstream from protein synthesis?
Key Words: pediatric hematology • nucleolus • cell cycle control
THE INHERITED BONE MARROW FAILURE syndromes include Fanconi anemia (FA), Diamond-Blackfan anemia (DBA), dyskeratosis congenita, and Shwachman-Diamond syndrome (1
, 2)
. Each presents as a distinct pediatric clinical entity, and yet they have some overlapping features. The standard of care for the bone marrow failure syndromes consists of steroid and/or cytokine administration together with transfusion therapy or hematopoietic stem cell transplantation. In DBA, in which the average age of diagnosis is 2 months, stem cell transplantation from a matched sibling produces a good outcome. In FA there is a significant predisposition to acute myeloid leukemia, myelodysplastic syndrome, as well as lymphomas, multiple carcinomas, and osteogenic sarcomas. There is also a high incidence of congenital anatomical abnormalities in the inherited bone marrow failure syndromes, although it is rare that these are major factors in the mortality.
GENES
The first molecular breakthrough in DBA was the discovery in 1999 that some patients have a mutation in the gene for a protein constituent of the cell’s protein synthesis machine—the ribosome. This machine consists of a small and large subunit, each built on a scaffold of RNA onto which various proteins are organized. The protein initially found to be mutated in DBA resides in the small subunit and is called RPS19 (3)
. Subsequently, mutations in a second small ribosomal subunit protein, RPS24, were found in other DBA patients (4)
.
Since many DBA patients do not display mutations in either of these ribosomal proteins, it is impossible at this time to know whether the disease will turn out to have a ribosome-centered mutation in all cases. Nonetheless, the reported ribosomal protein mutations are the most exciting molecular development to date in DBA and could represent an enabling insight for all the bone marrow failure disorders if, indeed, they have a common pathogenic trigger (although this is by no means certain).
The first reported RPS19 mutation associated with DBA was due to a balanced translocation between one of the X chromosomes in this female patient and autosome 19, with the break point found to reside in the third intron of the RPS19 gene (3)
. Numerous splice site, frameshift, or nonsense mutations in RPS19 were also described (3
, 5
6
7)
, and in all cases the cell’s haplosufficiency or insufficiency would depend on whether the wild-type RPS19 allele can generate enough product, either in a cell type that can get by with less or by compensatory overproduction from the wild-type allele. In other patients, the RPS19 gene has missense mutations so that a normal length protein is produced with a single amino acid alteration (3
, 6
, 7)
. Here the sequellae include competition with wild-type RPS19 in the ribosome biosynthesis pathway. It seems possible that these missense mutant RPS19 proteins retain sufficient resemblance to their normal counterparts to make their way into ribosomes, but ones that are hypofunctional, again causing the cell to be limited in its protein synthetic output. The pedigree studies of DBA inheritance so far do not provide a clear picture of whether the ribosomal protein mutations always act as dominant inhibitors of ribosome production or function, but the envisioned scenarios are all plausible. Recent experiments have used siRNA technology to knock down RPS19 in either HeLa (8)
or, more appropriately, CD34+ erythroid progenitor cells (9)
. While these sorts of experiments are potential guideposts, it is difficult to know the extent to which the observed effects represent a true phenocopy of the disease, especially because in DBA the mutant ribosomal proteins may not be limiting.
WHAT WOULD A RIBOSOME DEFICIENCY MEAN FOR THE ORGANISM?
Since the ribosomal proteins display extraordinary evolutionary conservation, as does the ribosome itself, it is likely that most mutations would be lethal to the embryo when present in a homozygous condition, as has indeed been found in the case of the RPS19 mutation (10)
. For a heterozygous embryo, the issue comes down to the balance between the output of the wild-type allele against the negative effects of the mutant gene, as outlined above. A surprising finding has been that, in the only study of its kind to date, mice heterozygous for a RPS19 mutation displayed no obvious erythropoietic abnormalities (10)
. In patients with this mutation, the inheritance pattern is that of a dominant trait, so the mouse result is confounding. It has subsequently been reconsidered in light of new findings (11
, 12)
.
In the mammalian fetal liver, there are two to four essential cell cycles that erythroblasts undergo before disgorging the nucleus, simplifying their cytoplasm and mounting a massive synthesis of globin, heme, transferrin receptor, etc., as professional reticulocytes. We know that some cell types have reserves of ribosomes that can be recruited—oocytes being the gold standard, where maternal stockpiles of ribosomes allow the fertilized egg to go from a translationally dormant to hyperactive state in a few minutes or hours without any new ribosome synthesis, based on post-transcriptional activation of stored mRNAs. But the mammalian fetal liver’s erythropoietic cells or, postnatally, the bone marrow’s erythroblasts may enjoy no such luxury, and a husbandry of translational machinery may dominate these critical hours. The general notion that certain erythroid progenitor cells lie right at the cusp of being underequipped for protein synthesis has a certain appeal not only for the hematological axis of bone marrow failures, but if applied to other cell lineages in the developing fetus, could also explain the congenital craniofacial abnormalities seen in DBA and other bone marrow failure disorders as resulting from stalled divisions of stem cells leading to blocked differentiation pathways or hypocellularity. Defects in ribosome production have been linked to the classical minute mutation in Drosophila (13)
and, more recently, to a similar phenotype in the mouse (14)
.
A second scenario is worth mentioning. Let us assume that mutant ribosomal proteins in DBA do make their way into ribosomes and even get into polysomes (i.e., they work to some degree). One consequence might be a slowed rate of translational elongation. At first blush, one might suppose this would simply mean a reduction of a few percent in the amount of a given protein produced per minute, which perhaps would be tolerable. But slowed translation could have a more adverse effect due to the kinetics of nascent vs. completed protein folding. Most proteins adopt their functional configuration after translation, assisted by ATP-mediated chaperones. However, some proteins acquire their proper folding as they are being synthesized, and recent evidence suggests that even a minimal slowing of translational elongation can change this nascent protein folding. This possibility of a cotranslational, protein folding abnormality in DBA is completely open. Although centered on the ribosome and thus differing from the central hypothesis to be advanced in this article, the possibility of altered cotranslational protein folding when the nascent polypeptide’s elongation has been slowed on mutant ribosomes is an idea that deserves exploration.
ANOTHER IDEA
The nucleolus is formed by the transcriptional action of the genes for the two large RNA components of the ribosome. These genes are present as 
400 copies in human cells, distributed in tandem blocks on five of the autosomes, and the cytological manifestation of the nucleolus arises from the focal transcriptional activity of these genes and the colocalized rRNA processing and ribosome assembly events, all of which involve high concentrations of RNA and proteins and thus present a strong phase-contrast or histochemical image. Ribosome assembly in the nucleolus is one of the most well-defined cases of the congruence of form and function in cell biology. Over the past decade, however, this classical view of a monofunctional nucleolus has been revised (15)
.
We now know that the nucleolus has other functions, and one of the most conspicuous is that it plays a role in cell cycle progression. At first, the evidence that the nucleolus plays a role in cell cycle progression was indirect, based on the findings that it contains telomerase components and mitogenic growth factors (15
, 16)
. When I published these speculative reviews in 1998 I thought I was onto something new. But there is rarely anything new under the sun, and as I discovered in the library recently, it had been found half a century ago that focal UV microbeam irradiation of single nucleoli in grasshopper neuroblasts blocked subsequent entry into mitosis (17)
. This finding could not be readily attributed to a protein synthesis defect given the presence of unirradiated nucleoli in the same cell together with the existing supply of ribosomes (17)
. Subsequently, proteomic analyses of highly purified nucleoli from a human tumor cell line revealed numerous proteins that have no obvious relationship to ribosome biosynthesis (18
19
20)
. Then, in 2002, the author’s group discovered that there are sites within the nucleoli that are relatively deficient in nascent ribosomes (21)
; in a subsequent study these sites were found to contain the stem cell and tumor cell-specific protein nucleostemin (22)
. In more recent work, these nucleostemin-rich sites with nucleoli have been found to also contain the tumor suppressor protein ARF (23)
. In further support of this mosaic nucleolar landscape of nascent ribosomes interspersed with sites containing cell cycle regulatory proteins, an electron microscopy method that distinguishes RNA-rich structures from protein-rich, RNA-deficient ones revealed that nascent ribosomes in the nucleolus are situated immediately adjacent to protein-rich particles (22)
.
If, as is now clear, the nucleolus contains cell cycle machinery in close proximity to nascent ribosomes, we can entertain another idea for how mutant ribosomal proteins could act. Either by their absence (failure to localize in the nucleolus) or presence (nucleolar localization and assembly into mutant ribosomes or negative effects on wild-type ribosome assembly), the molecular contours of the nascent ribosomes would be altered. To the extent that these nascent ribosomes normally are so closely juxtaposed to cell cycle regulatory machinery in the nucleolus, it is possible that the latter molecules would display altered nucleolar affinity or intranucleolar relocations. There are various consequences that can be envisioned, including increased release of ARF into the nucleoplasm and/or an increased nucleolar retention of the normally shuttling nucleostemin, either of which would be predicted to delay or arrest cell cycle progression. Such envisioned impairments of cell cycle progression emanating from a molecular reconfiguration of the cell cycle machinery in nucleoli would seem to be as compatible with the features of bone marrow failure syndromes as the scenarios based on mutant ribosomal proteins acting to impair protein synthesis. The hypothesis being presented here is that a disruption of normal ribosome assembly in the nucleolus triggers a direct negative effect on cell cycle progression upstream from any protein synthetic deficiencies arising from the subsequent emergence of mutant ribosomes to the cytoplasm. Of course, this new hypothesis of a disruption of cell cycle machinery staging in the nucleolus resulting from the presence of misshaped nascent ribosomes and the current, more popular view that the effect is downstream, at the level of translation, are not mutually exclusive ideas.
CONCLUDING PERPSECTIVE
Whether the idea for DBA pathogenesis advanced here has any merit can only be determined by future work. Looking over the DBA field, the idea of cell cycle disorder centered in the nucleolus is a viable idea and could explain much. I have been interested in eryrthropoiesis ever since I was a graduate student (24)
and will continue thinking about how ribosome deficiencies might operate, both in the nucleolus as argued here and at the translational level. But I wish to close on a broader note, one of concern.
In thinking about DBA as a case in point, I have been struck by how unable we are to draw plausible scenarios for the cellular phenotype, much less the pathology, when any given gene is mutated, be it a ribosomal protein, a cytokine, or a receptor. It is one thing to predict what might happen if a tumor cell pumps out a chemotherapeutic drug (viz., survives the attack), but it is quite another to predict what a cell would do if a mutant succinate dehydrogenase were operating in the mitochondrion with a kcat 20% of normal. We would all agree that oxidative phosphorylation would be compromised, but who among us would predict a defect in hair follicle development or some such thing in these mice? And yet such findings are being reported all the time now. It seems obvious there is some profound principle about cellular regulation during development and differentiation that we still have not discovered.
This humility is a good thing, reinforced each month when many papers appear that take us by surprise, describing phenotypes none of us would have predicted when a single gene is knocked out or down. It is a good thing because it keeps us on our toes and keeps our vision on the endless frontier, to borrow Warren Weaver’s enduring term. We must keep informing the public and Congress that these constant surprises are the true frontier. We should admit, indeed broadcast, that organisms are more complex than we know. But, unlike the possible case of the cosmos, it is unlikely that organisms are more complex than we can know.
ACKNOWLEDGMENTS
Work from the author’s laboratory cited in this article was supported by grants from the U.S. National Institutes of Health (GM-21595) and the U.S. National Science Foundation (MCB-0445841). Constructive comments on the manuscript were received from several pediatric hematology and DBA experts, who are gratefully acknowledged: Benjamin Ebert (Dana Farber Cancer Institute, Boston, MA, USA), Steven Ellis (University of Louisville, Louisville, KY, USA), Karen Gripp (Du Pont Hospital, Wilmington, DE, USA), Fabrizio Loreni (University of Rome, Rome, Italy), Peter Newburger (University of Massachusetts Medical School, Worcester, MA, USA), Akiko Shimamura (Dana Farber Cancer Institute), and Colin Sieff (Whitehead Institute, Cambridge, MA, USA). The author also thanks David G. Nathan (Dana Farber Cancer Institute) for his encouragement. The idea set forth in this article occurred to the author while attending the Eighth Annual Diamond Blackfan Anemia International Consensus Conference sponsored by the Daniella Maria Arturi Foundation, held in New York City March 17–19, 2007, an invitation to which he thanks Steven Ellis and Marie and Emanuel Arturi.
Received for publication May 8, 2007. Accepted for publication May 17, 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  T. Pederson and R. Y.L. Tsai In search of nonribosomal nucleolar protein function and regulation J. Cell Biol., March 23, 2009; 184(6): 771 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Farrar, M. Nater, E. Caywood, M. A. McDevitt, J. Kowalski, C. M. Takemoto, C. C. Talbot Jr, P. Meltzer, D. Esposito, A. H. Beggs, et al. Abnormalities of the large ribosomal subunit protein, Rpl35a, in Diamond-Blackfan anemia Blood, September 1, 2008; 112(5): 1582 - 1592. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
