A Mechanism for Cell Size Regulation by the Insulin and Insulin-Like Growth Factor-I Receptors

Cell size plays an important role in cell proliferation, as cells must double in size during the cell cycle (1). Cell size is largely determined by ribosome biogenesis (2), which is dependent on the activity of RNA polymerase I [ref. 3; for a general review on cell growth, see ref. 4]. The type 1 insulin-like growth factor (IGF-I) receptor (IGF-IR) accounts, in a nonredundant way, for ∼50% of mouse embryo growth (5). Deletion of the insulin receptor substrate-1 (IRS-1) in Drosophila (6) or mice (7) also results in smaller animals, the 50% reduction being due to a decrease in both cell size and cell number (6, 8). IRS-1 is also the docking protein of the insulin receptor (InR), and it seems therefore paradoxical that deletion of the InR genes results in mice that are normal in size at birth (9).

The upstream binding factor 1 (UBF1) is a nucleolar protein that participates in the regulation of RNA polymerase I activity at the rDNA promoter (3). The activity of UBF1 is increased by growth factors and depends on its phosphorylation (10, 11) and the levels of the protein. UBF1 levels are increased in hypertrophic cells (12, 13) and overexpression of UBF1 increases cell size in myeloid cells (14). IGF-IR signaling induces nuclear translocation of IRS-1, which binds to UBF1 and activates it (15, 16). In addition, in 32D myeloid cells, expression of IRS-1 stabilizes the levels of UBF1. In parental cells (32D IGF-IR) that do not express IRS-1 (or IRS-2; ref. 17), the UBF1 protein (but not its mRNA) is rapidly degraded when the cells are shifted from interleukin-3 (IL-3) to IGF-I (16). Ectopic expression of IRS-1 (32 IGF-IR/IRS-1 cells) doubles the size of the cells, as measured by fluorescence-activated cell sorting (FACS) analysis (18), and prevents the down-regulation of UBF1 (16).

We have investigated in 32D myeloid cells whether the levels of UBF1 may explain the paradoxical results reported in mice on the respective contributions of the IGF-IR and InR to cell size. We used three 32D-derived cell lines for these studies: 32D IGF-IR cells that do not express IRS-1, 32D IGF-IR/IRS-1 cells, and 32D InR/IRS-1, described in detail in previous articles (19, 20). 32D InR cells not expressing IRS-1 could not be included in these experiments as they die rapidly (within 16 hours) when shifted to insulin.

FACS analysis. Exponentially growing cells were seeded in RPMI 1640 supplemented with the respective growth factors. After 48 hours, the cells were analyzed by FACS for cell size (18).

Giemsa staining. Giemsa staining was done as described previously (17).

Western blots. Western blots were carried out as described in detail in previous reports (16, 18). The antibodies used were mouse monoclonal anti-UBF (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), mouse monoclonal anti-growth factor receptor binding protein 2 (Grb2; Transduction Laboratories), anti-IRS-1 antibody (Upstate, Inc., Charlottesville, VA), and rabbit polyclonal antibodies against the α-subunits of the IGF-IR and InR (Santa Cruz Biotechnology). Secondary antibodies were peroxidase goat anti-rabbit IgG (Oncogene Science, Inc., Manhasset, NY) and peroxidase goat anti-mouse IgG (Oncogene Science). The Western blotting detection reagent was enhanced chemiluminescence (Amersham Biosciences, Piscataway, NJ).

rRNA synthesis. Cells were incubated with the respective growth factors (Life Technologies, Inc., Gaithersburg, MD) for 48 hours. The cells were then labeled for 4 hours with [³²P]orthophosphate at a final concentration of 250 μCi/mL (ICN Biochemicals, Inc., Cleveland, OH) in phosphate-free medium (Life Technologies). After labeling, the cells were washed and incubated in fresh medium for 2 hours. Total RNA was isolated using RNeasy Mini kit (Qiagen, Chatsworth, CA) and separated by electrophoresis on 1% agarose formaldehyde gels. After drying, the ³²P-labeled rRNA was visualized by autoradiography. The amounts of RNA in each lane were monitored by ethidium bromide staining.

Cell size of 32D-derived cells. FACS analysis of a single experiment is shown in Fig. 1. There is a definite increase in the size of 32D IGF-IR IRS-1 cells when compared with either 32D InR IRS-1 cells or 32D IGF-IR cells. The difference can be calculated from the difference between G₁ and G₂ cells (18). 32D IGF-IR IRS-1 cells are about twice as large as the other two cell lines. Figure 1 actually shows no difference in size between 32D IGF-IR and 32D InR IRS-1 cells. In repeated experiments, the InR IRS-1 cells are slightly larger than the 32D IGF-IR cells. The differences persist in separated G₂ cells (Fig. 1, right). The experiments in Fig. 1 were done on cells 48 hours after shifting from IL-3 to the respective ligands (IGF-I for the IGF-IR and insulin for the InR), when the cells are still proliferating (19, 20). Not surprisingly, the difference disappears when the cells are shifted to the wrong ligand (i.e., InR/IRS-1 cells to IGF-I or IGF-IR/IRS-1 cells to insulin) or when the cells are grown in IL-3 (data not shown).

The results of FACS analysis were confirmed by Giemsa staining. Figure 2A shows Giemsa staining of representative fields for 32D IGF-IR IRS-1 and 32D InR IRS-1 cells 48 hours after shifting to the respective ligands (Fig. 2A). 32D IGF-IR IRS-1 cells are significantly larger than 32D InR IRS-1 cells. Figure 2B shows a comparison of G₂ cells from 32D IGF-IR IRS-1 cells versus 32D InR IRS-1 cells and 32D IGF-IR cells. The figures confirm the FACS data that the cells expressing the IGF-IR and IRS-1 are larger than the cells expressing the InR and IRS-1 or the IGF-IR cells without IRS-1. There was no significant difference in the cell cycle distribution of the three cell lines for the first 48 hours after shifting to the respective ligands (data not shown). After 48 hours, 32D IGF-IR cells stop proliferating and begin to differentiate, and a cell cycle analysis at later times would be meaningless.

UBF1 levels. Because UBF1 is stable in 32D IGF-IR IRS-1 cells but down-regulated in 32D IGF-IR cells (16), we investigated whether UBF1 levels could also explain the difference in size between 32D IGF-IR IRS-1 and 32D InR IRS-1 cells. UBF1 is down-regulated in cells expressing IRS-1 and the InR (Fig. 3) but remains stable in 32D IGF-IR IRS-1 cells. UBF1 is also down-regulated in 32D IGF-IR cells, as reported previously (16). Thus, the levels of UBF1 correlate with cell size, as one would expect (3). Figure 3 also shows the levels of IRS-1 and of the respective receptors. An antibody to Grb2 was used to monitor protein loading. After corrections for loading, one can say that receptors and IRS-1 are expressed roughly at similar levels in all cell lines, except, of course, in 32D IGF-IR cells, where IRS-1 is absent. Figure 3 is taken from a single experiment, but the experiment was repeated several times with the same results. In 32D IGF-IR cells, the down-regulation of UBF1 protein was not accompanied by a down-regulation of its mRNA (16). Down-regulation of UBF1 in InR IRS-1 cells was also limited to the protein, with the mRNA levels remaining high (data not shown).

The difference in UBF1 levels is accompanied by a difference in rRNA synthesis (Fig. 4). In this experiment, 32D IGF-IR IRS-1 cells and 32D InR IRS-1 cells were shifted to their respective ligands for 48 hours and then labeled with ³²P (15) for 4 hours. RNA was extracted and Northern blots were carried out followed by autoradiography. The amounts of RNA (ethidium bromide staining) were similar, but the labeling is markedly increased in the cells with the IGF-IR. This was to be expected, as cell size is largely determined by the activity of RNA polymerase I (see Introduction), and, if there is a cell size difference, there should also be a difference in rRNA synthesis.

Taken together, these results indicate that UBF1, a regulator of RNA polymerase I activity, is more stable in myeloid cells expressing the IGF-IR and IRS-1 than in cells expressing the InR and IRS-1. The levels of UBF1 provide one reasonable explanation for the difference in cell size between the two cell lines expressing the two receptors.

As to the significance of these findings, 32D cells are murine myeloid cells that behave in culture like myeloid cells in the bone marrow of animals (i.e., they differentiate into granulocytes with the appropriate growth factors; ref. 17). It would be an extrapolation to propose that the difference in size found in 32D-derived cells is the only explanation for the differences found in animals between the two receptors. However, it is legitimate to suggest that UBF1 levels provide at least one important mechanism to explain the size difference. In this report, we have focused on the difference in cell size. Previous articles (confirmed during the work on this project) have already indicated that the down-regulation of UBF1 is due to the down-regulation of the protein (mRNA levels remain steady) and mostly to degradation of the protein, although a small decrease in synthesis has also been observed (16).

In conclusion, we have shown in myeloid cells that the InR is not as effective as the IGF-IR in regulating the levels of UBF1. Our results provide a molecular mechanism for the observation that deletion of the IGF-IR or IRS-1 causes a 50% decrease in body size, whereas deletion of the InR does not.

Grant support: NIH grant CA089640.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.