Elevated circulating IGF-I promotes mammary gland development and proliferation

Abstract

Animal studies have shown that IGF-I is essential for mammary gland development. Previous studies have suggested that local IGF-I rather than circulating IGF-I is the major mediator of mammary gland development. In the present study we used the hepatic IGF-I transgenic (HIT) and IGF-I knockout/HIT (KO-HIT) mouse models to examine the effects of enhanced circulating IGF-I on mammary development in the presence and absence of local IGF-I. HIT mice express the rat IGF-I transgene under the transthyretin promoter in the liver and have elevated circulating IGF-I and normal tissue IGF-I levels. The KO-HIT mice have no tissue IGF-I and increased circulating IGF-I. Analysis of mammary gland development reveals a greater degree of complexity in HIT mice as compared to control and KO-HIT mice, which demonstrate similar degrees of mammary gland complexity. Immunohistochemical evaluation of glands of HIT mice also suggests an enhanced degree of proliferation of the mammary gland, whereas KO-HIT mice exhibit mammary gland proliferation similar to control mice. In addition, HIT mice have a higher percentage of proliferating myoepithelial and luminal cells than control mice, whereas KO-HIT mice have an equivalent percentage of proliferating myoepithelial and luminal cells as control mice. Thus, our findings show that elevated circulating IGF-I levels are sufficient to promote normal pubertal mammary epithelial development. However, HIT mice demonstrate more pronounced mammary gland development when compared to control and KO-HIT mice. This suggests that both local and endocrine IGF-I play roles in mammary gland development and that elevated circulating IGF-I accelerates mammary epithelial proliferation.

Figure 1

Body weight and mammary gland weight of control, HIT, and KO-HIT female mice. A, Body weight of control, HIT, and KO-HIT mice at 4, 8, and 16 wk of age. B, Mammary gland weight of control, HIT, and KO-HIT mice at 4, 8, and 16 wk of age. For A and B, n = 10 for all groups. For both graphs, a, P < 0.05 HIT vs. LL; b, P < 0.05 HIT vs. KO-HIT.

Figure 2

Elevated circulating IGF-I enhances mammary gland development in the presence of local IGF-I and may be sufficient for mammary gland development in the absence of local IGF. Mammary glands were collected from control, HIT, KO-HIT mice at 4, 8, and 16 wk of age. A, Whole mammary gland mounts were prepared at the indicated ages (for all groups, n = 8–10 at 4 wk of age, n = 5 at 8 wk of age and n = 3–4 at 16 wk of age). B, Ductal extension into the fat pad of 4-wk-old mice was measured as the ratio of duct length to fat pad length. C, The number of TEBs per gland of 4 week mice was counted, quantified, and is depicted per total gland area. D and E, Mammary gland complexity as a measure of the extent of ductal branching was quantified by determining the number of intersecting branches along a line drawn equidistant between the leading edge of the ducts and the lymph node. Complexity was measured as the number of branches per unit length. For all graphs, a, P < 0.05 HIT vs. LL; b, P < 0.05 HIT vs. KO-HIT.

Figure 3

Elevated circulating IGF-I leads to mammary gland hyperplasia. Mammary glands of 6- to 8-wk-old virgin control, HIT, and KO-HIT mice were collected and stained with H&E (n = 3 for all groups). Magnification, ×40 (A), ×200 (B).

Figure 4

Cellular proliferation is enhanced by increased circulating IGF-I in the presence of local IGF-I but is normalized by increased circulating IGF-I in the absence of local IGF-I. Mammary glands of 6- to 8-wk-old virgin control, HIT, and KO-HIT mice were collected and serial sections were subjected to immunofluorescence staining to measure BrdU incorporation (n = 3 for all groups). Antibody to cytokeratin 14 (ck14) was used to identify myoepithelial cells and DAPI was used to stain for nuclei. A, BrdU incorporation was quantified and is represented as percent of BrdU-positive cells per total cells per ductal structure evaluated. Ductal elements were classified as TEBs and ducts. TEBs were divided into body cells and cap cells and were also combined for analysis (TEB total). TEBs and ducts were combined for analysis (total). a, P < 0.05 HIT vs. LL; b, P < 0.05 HIT vs. KO-HIT. B, Representative images of TEB caps at ×400 for all groups. C, Representative images of ducts at ×400 for all groups.

Figure 5

Proliferation of myoepithelial and luminal cells is increased by elevated IGF-I in the presence of local IGF-I and normalized by elevated IGF-I in the absence of local IGF-I. Mammary glands of 6- to 8-wk-old virgin control, HIT, and KO-HIT mice were collected and serial sections were subjected to immunofluorescence staining to measure BrdU incorporation (n = 3 for all groups). A, Antibody to cytokeratin 14 (ck14) was used to identify myoepithelial cells and DAPI was used to stain for nuclei. Luminal cells were defined as cells that stained for DAPI but not ck14. The total number of myoepithelial cells and luminal cells per ductal structure were counted. The number of BrdU⁺ myoepithelial and luminal cells were quantified and the percent of BrdU⁺ cells per total cells per ductal structure was calculated. B, The percentage of BrdU⁺ myoepithelial cells is represented graphically for the three groups. C, The percentage of BrdU⁺ luminal cells is represented graphically for the three groups. a, P < 0.05 HIT vs. LL; b, P < 0.05 HIT vs. KO-HIT.

Figure 6

Elevated IGF-I does not affect levels of protein expression and basal phosphorylation of the IGF-IR and IR, nor does it affect expression levels or activation of downstream signaling pathways. Mammary glands were removed from 6–8 week virgin control, HIT, and KO-HIT mice were collected. Protein extraction and Western immunoblotting were performed. A, Western immunoblot was performed to detect protein levels of the phosphorylated IGF-I receptor/insulin receptor (p-IGF-IR/IR), the total IGF-IR, the IR, phosphorylated AKT (p-AKT), total AKT (AKT), phosphorylated ERK/MAPK (p-ERK), or total ERK/MAPK (ERK). β-Actin was used as a loading control. The positive control (+) was IGF-I stimulated NIH 3T3-B3 fibroblasts which overexpress the IGF-IR. B and C, Densitometric analysis was performed on the Western Blots shown in A. B, Graph represents the total levels of AKT, ERK/MAPK, IR, and IGF-IR normalized to β-actin. C, Ratio of phosphorylated protein to total protein level was calculated for all groups. b, P < 0.05 HIT vs. KO-HIT.