Michael J. Kuchenreuther, Jose M. Sanchez, Angela C. Hirbe, Jason D. Weber, Anthony J. Apicelli, Alexander P. Miceli, Leonard B. Maggi, Katherine N. Weilbaecher, Anthony J. Saporita, Crystal L. Schulte-Winkeler, and Mary E. Olanich
Cellular growth (i.e., macromolecular synthesis) is an essential function during the early parts of the cell cycle. For cells to transit the G1 restriction point, they must duplicate nearly their entire protein content; failure to do so would result in smaller daughter cells (12). Only recently has an emphasis been placed on the fundamental control of cell growth and its link to the cell cycle. Developments in the understanding of how the cell senses environmental nutritional cues has led to a flurry of research on understanding the mechanisms underlying growth control (40). Not surprisingly, several of these pathways converge on the synthesis of new ribosomes in the cell nucleolus and the regulation of translation. Approximately half of the cell's energy expenditure is directed toward ribosome biogenesis (26). The nucleolus, long recognized as a marker for active cellular growth, was first described in the early 1960s as the center of ribosomal DNA (rDNA) transcription and ribosome biogenesis (6, 32). This organelle is composed of three regions, on the basis of morphology at the ultrastructural level: the fibrillar centers, the dense fibrillar compartment, and the granular zone. rDNA transcription occurs in the junction region between the fibrillar centers and the surrounding dense fibrillar component, and the resulting rRNA is further processed in the periphery of the dense fibrillar component. Further posttranscriptional modifications and assembly into subunits occur in the surrounding granular region (18). While the primary mechanisms regulating these processes have been well studied in Saccharomyces cerevisiae (13), multicellular organisms demand more complex regulatory mechanisms, in that proliferative capacity is determined not only by the relative abundance of nutrients but also by complicated extracellular signals and growth factors. Indeed, previous studies have demonstrated convergence between the growth and the proliferation pathways via regulation of the tumor suppressor gene products Rb and p53 (9, 17, 43, 48). Both products are known to negatively regulate the activity of polymerase I in rDNA transcription. Oncogenes such as c-Myc also regulate the transcription of rDNA and the genes that encode ribosomal proteins, implying that an intricate network exists within the nucleolus to ensure the proper synthesis of ribosomes (7, 15, 16). The tumor suppressor p19ARF represents an attractive candidate for coupling proliferation to growth. Given its nucleolar localization (39, 44, 45) and potent induction by hyperproliferative signals (19, 20, 31, 50), ARF represents a potential nucleolar integrator of growth signals coming into the cell. It has been regarded classically as an activator of p53 through its ability to sequester Mdm2, the E3 ubiquitin ligase for p53, in the nucleolus (39, 44, 45). However, recent data have demonstrated a role for ARF in binding to and affecting the function of the ribosomal chaperone nucleophosmin (NPM), independent of its ability to regulate p53 (4, 8, 21). Furthermore, these data are consistent with those from a growing number of studies with mice and humans that describe p53-independent functions for ARF tumor suppression (35). Given ARF's nucleolar localization, its role in suppressing cellular growth and proliferation, and its ability to bind to a protein involved in ribosome biogenesis, we were inclined to explore the functional and physiological consequences of ARF disruption of growth and ribosome biogenesis. Through in vitro and in vivo assays, we utilized targeted Arf knockout mice and selective ARF knockdown via lentiviral RNA interference. Cells derived from Arf-null mice displayed significant alterations in gross nucleolar morphology and abundance and had a marked increase in basal protein synthesis levels compared to that in wild-type cells. Furthermore, this increase in protein synthesis was correlated to increased ribosome biogenesis and cytoplasmic ribosome content, implying a regulatory role for ARF in these processes. Importantly, though ARF levels are nearly undetectable in low-passage mouse embryonic fibroblasts (19), the knockdown of endogenous ARF via short hairpin RNA (shRNA) constructs mimicked the Arf-null nucleolar and ribosomal phenotype, implying an important ribosome homeostatic role for basal ARF proteins in wild-type cells. The progrowth phenotype of the Arf loss was not limited to proliferating cells, as fully differentiated osteoclasts from Arf-null mice exhibited tremendous gains in protein synthesis and overall activity in vivo. Mechanistically, all of the ribosome gains exhibited by the loss of Arf were reversed by the removal of the nucleolar NPM proto-oncogene, indicating that NPM, when untethered from ARF, promotes unrestrained ribosome biogenesis. Taken together, these data strongly argue for a moment-to-moment “thermostat”-like role for basal ARF molecules in controlling NPM-directed ribosome biogenesis and protein synthetic rates.