Abstract
Expression of insulin receptor substrates (IRS)-1 and -2 within the mammary gland was found to be high at mid-lactation and dramatically decreased with mammary involution. This observation supports the hypothesis that these proteins are induced in the mammary gland with lactogenesis and involved in normal milk synthesis. To test this hypothesis, lactation capacity, along with indices of mammary secretory cell glucose metabolism and cell signaling were compared in normal mice and mice carrying targeted mutations in either the Irs1 or Irs2 genes. Mammary IRS-1 and IRS-2 protein levels were increased within 1 day of parturition and reached maximal levels by 5 days post partum. Dams carrying germline mutations of Irs1 or Irs2 displayed reduced lactation capacity as assessed by weight gain of pup litters. The reduction was more dramatic in Irs1−/− versus Irs2−/− dams. Maternal body weight was also reduced in Irs1−/− dams as well as in Irs1+/− Irs2+/− dams. The loss of IRS-1 had little impact on mammary gland expression of milk protein mRNAs, glucose transport, or on the abundance and subcellular localization of hexokinases I and II. The loss of IRS-1 was associated with a compensatory increase in insulin-induced IRS-2 phosphorylation; however, the loss of IRS-1 did also cause a reduction in insulin-dependent mammary gland-specific activation of Akt phosphorylation. These results support the conclusion that IRS-1 is important for insulin-dependent activation of Akt signaling within the lactating mammary gland, but that loss of this protein has only modest impact on normal milk synthesis, since related signaling proteins such as IRS-2 may act in compensatory fashion.
Introduction
Establishment of a normal lactation is the result of a number of highly orchestrated processes occurring both during pregnancy and within the first several days of the post partum period. This process, referred to as lactogenesis, is a two-stage developmental program during which the mammary secretory cell prepares for the process of milk secretion. Lactogenesis I is characterized by the detectable induction of lactation-specific gene products within the mammary epithelium. In rodents, this process begins as early as day 10 of pregnancy and is associated with a variety of biochemical as well as morphological changes within the mammary epithelium ( Bauman et al. 1974, Mellenberger & Bauman 1974b, Slaby & Brown 1974, Rosen & Barker 1976, Nakhasi & Qasba 1979, Schwertfeger et al. 2003). Lactogenesis II, also referred to as secretory activation, is characterized by a switch that occurs within the secretory epithelium from the pre-lactating differentiated state described earlier, to a state of intense metabolic, biosynthetic, and secretory activity. It is most simply described as the onset of ‘copious milk secretion’ ( Neville et al. 2001). From a mechanistic standpoint, secretory activation has best been characterized by the dramatic changes that occur both in milk volume and in milk composition ( Neville & Morton 2001). Key events underlying these dramatic changes in milk composition and volume include closure of the epithelial cell tight junctions, upregulation of milk protein synthesis, upregulation of lactose and lipid synthesis, and increased oxidative metabolism ( Mellenberger & Bauman 1974a, b, Rosen et al. 1978, Nguyen et al. 2001). Along with these, increases also occur in the expression and/or activation of enzymes associated with glucose metabolism such as hexokinase II, Glut 1, Glut 12, and PFK2 ( Sochor et al. 1984, Kaselonis et al. 1999, Nemeth et al. 2000, Macheda et al. 2003). Although the underlying trigger for these events is not completely understood, key events involved with lactogenesis include the drop in blood progesterone that coincides with parturition along with surges in blood prolactin and glucocorticoids ( Neville & Morton 2001). Insulin is also believed to be important for supporting metabolic processes during lactation, despite the fact that its mammary-specific role has yet to be completely defined ( Neville et al. 2002).
The mechanism through which insulin acts within cells and tissues has largely been defined in terms of effects mediated through the insulin receptor substrate proteins (IRS)-1 and -2 ( Thirone et al. 2006). The IRS proteins serve as docking proteins to facilitate the interaction of a number of signal transducers ( Yenush & White 1997). Most importantly, they link activation of the receptors for insulin and insulin-like growth factor-I (IGF-I) with a variety of biological responses in cells. Following activation, IRS proteins bind to effector molecules containing the signature src homology domain (SH2) within their amino acid sequence. Some of these effector molecules serve as docking proteins themselves, while others have intrinsic kinase or phosphatase activity. Among the best characterized of IRS-1 interactions is that which occurs with phosphatidylinositol 3′-kinase ( Ruderman et al. 1990, Araki et al. 1994). This interaction is known to mediate the ability of receptors for insulin and IGF-I to activate the serine threonine kinase Akt ( Datta et al. 1997). The activation of Akt in response to insulin or IGF-I stimulation mediates the ability of these two hormones to inhibit apoptosis ( Datta et al. 1997). However, activation of Akt has also been demonstrated to mediate the effects of insulin on protein synthesis, carbohydrate metabolism, and lipid biosynthesis ( Mendez et al. 1996, Gottlob et al. 2001, Schwertfeger et al. 2003).
Previous studies in our laboratory demonstrated that expression of IRS-1 and IRS-2 within the mammary gland is developmentally regulated and that both proteins are highly expressed during lactation ( Lee et al. 2003b). The goal of the present study was to determine the importance of IRS-1 and IRS-2 to normal lactation and understand the potential mechanisms through which these two proteins might mediate the processes necessary to milk synthesis.
Materials and Methods
Experimental animals
All animals were studied in accordance with procedures outlined in the NIH Guide to Care and Use of Experimental Animals. These experiments were approved by the Baylor College of Medicine Animal Care and Use Committee. For the analysis of IRS expression during secretory activation, mammary tissue samples were harvested from timed-pregnant CD-1 females (Charles River Laboratories) on 17 and 19 days of gestation and on 1, 2, 3, 4, and 5 days post partum. To determine the importance of IRS-1 and IRS-2 to lactation, litter weight gain, milk composition, and mammary development were studied in mice that were either heterozygous or homozygous for targeted germline mutations. For both genes, litter weight gain was compared among wild-type (+/+), heterozygous (+/−), and null (−/−) females. Lactation performance was also compared among dams heterozygous for both IRS-1 and IRS-2. Mice carrying germline mutations in the Irs1 or Irs2 genes were previously described ( Araki et al. 1994, Kubota et al. 2000). All animals were on a mixed genetic background consisting of FVB, SV129, and CBA. All genotype effects were tested for by experiments that used littermate comparisons. Cohorts of (totally eight) lactating females were obtained through the use of timed matings as previously described ( Hadsell et al. 2003). On day 1 post partum, each dam received a cross-fostered litter of ten CD-1 pups. Litter weight was then recorded daily for 10 days. On day 10 post partum, milk and mammary tissue samples were collected and weighed from each dam as previously described ( Hadsell et al. 2003). To determine the effects of Irs1 mutation on insulin signaling capacity of the mammary gland, Irs1+/+ or Irs1−/− lactating dams were given an i.v. injections of bovine insulin (Sigma–Aldrich) or long-R3 IGF-I (Sigma–Aldrich) as previously described ( Lee et al. 2003a, Hadsell et al. 2005). On day 3 post partum, dams were fasted for 8 h and then separated from their litters for 2 h to bring endogenous concentrations of insulin and prolactin to a baseline. Mammary tissue samples were then collected 5 min following tail-vein injection of either insulin or long-R3 IGF-I (0.25 mg/kg). To compare the glucose transport capacity among Irs1+/+ and Irs1−/− cells, primary mammary epithelial cell cultures were established from 16-day pregnant mice of each genotype as previously described ( Aggeler et al. 1991).
Milk and tissue analysis
Milk samples were analyzed for lactose, nitrogen, and water as previously described ( Hadsell et al. 2003). Milk samples were assayed for fat using the creamatocrit assay ( Mandel et al. 2005). Epithelial content of the mammary tissue in Irs1+/+ and Irs1−/− dams was determined by segmentation analysis of images captured from hematoxylin- and eosin-stained mammary tissue as previously described ( Hadsell et al. 2003).
Western and northern blotting
Total tissue protein extracts of mammary tissue were prepared from 50 mg tissue as previously described ( Hadsell et al. 2001). Mammary tissue mitochondrial and post-mitochondrial fractions were prepared as previously described ( Darley-Usmar et al. 1987). Total amounts of IRS-1, IRS-2, Akt, phospho-Akt, extracellular signal regulated kinases 1 and 2 (ERK1/2) and phospho-ERK1/2 were measured by western blotting as previously described ( Hadsell et al. 2001). Phospho-Akt was measured using antibodies to phospho-Thr308 or phospho-Ser473 (Cell Signaling Technology, Beverly, MA, USA). Phospho-ERK1/2 was measured using an antibody that detects dual phosphorylation of Thr202 and Tyr204 (Cell Signaling Technology). Tyrosine phosphorylation of IRS-2 was measured by western blotting of IRS-2 immunoprecipitates (IPs). Each IP was prepared by incubating 0.5 mg mammary protein extract with a rabbit polyclonal antibody to IRS-2 (Upstate Biotechnology, Charlottesville, VA, USA) as previously described ( Hadsell et al. 2001). Mitochondrial and post-mitochondrial fractions were blotted with antibodies to hexokinase I (Chemicon International, Temecula, CA, USA), hexokinase II (Chemicon International), cytochrome c oxidase subunit IV (COXIV; Abcam Inc., Cambridge, MA, USA), and tubulin (Abcam Inc). Western blotting for keratin 8 used the TROMA (trophoblastoma) I antibody (University of Iowa Developmental Studies Hybridoma Bank).
To check for equal loading, the samples were also run on gels which were subsequently stained with coomassie. Analysis of total mammary RNA for milk protein mRNA abundance was conducted by northern blotting as previously described ( Hadsell et al. 2003). Densitometry data were collected using a molecular dynamics personal densitometer SI.
Glucose transport
To determine whether mammary cell glucose transport was impaired in response to loss of IRS-1, primary mammary cell cultures were prepared from Irs1+/+and Irs1−/− female mice at 16 days of pregnancy. Primary cells were prepared as previously described ( Aggeler et al. 1991). All cell culture reagents were purchased from Sigma unless specified otherwise. Cells were suspended in 2 × P media consisting of Dulbecco’s minimum essential medium (DMEM) with antibiotics (100 U/ml penicillin, 100 U/ml streptomycin, and 50 μg/ml gentamicin), plus 2 mM glutamine, 2 μg/ml insulin, 20 ng/ml epidermal growth factor (EGF), 2 μM dexamethasone, and 25 mM glucose. Cells were then plated on six-well tissue plates that had been coated in 1 ml/well serum fetuin media for 24 h at 37 °C prior to plating (DMEM, 20% (v/v) heat-inactivated calf serum, 25 mM glucose, and 2 mg/ml fetuin). Cells were incubated for 24 h at 37 °C. After 24 h in culture, the media were changed to standard dexamethasone insulin prolactin (DIP) media, containing 25 mM glucose, 1 mg/l prolactin, 1 mg/l insulin, and 1 μM dexamethasone. The cells were maintained in this media for an additional 48 h. At 72 h after plating, a portion of the cells were harvested and DNA quantitated using the Hoechst assay. All other cells were incubated in glucose-free DIP media, plus or minus cytochalasin B (1 μg/ml) for 1 h at 37 °C. The media were then aspirated off and the cells were incubated for 15 min in glucose-free DIP, plus or minus cytochalasin B with 0.12 mCi (6.25 nmol/l) deoxy-d-glucose, 2-[3H(G)] (Perkin–Elmer, Waltham, MA, USA). Cells were washed thrice with ice-cold PBS and recovered using 500 μl of 5% TCA solution. Cells were placed in 5 ml Budget Solve Complete Counting Cocktail (Fisher Scientific, Pittsburg, PA, USA) and counted on a Tri-Carb 2500 TR Liquid Scintillation Analyzer (Packard Bioscience).
Data analysis
Litter weight gain and dam weight were analyzed using the repeated measures procedure of SPSS (version 12.01 for Windows, SPSS Inc., Chicago, IL, USA) with genotype as the fixed variable and day post partum as a repeated measure within each dam. Northern and western blotting data, milk composition data, litter size, and glucose transport data were all analyzed as one-way ANOVA designs using genotype as independent variables. Conception rate data among the genotypes were compared using the Kruskal–Wallis test in SPSS. All data are presented as lsmeans ± s.e.m. Differences were considered statistically significant at α = 0.05.
Results
Our previous work showed that IRS-1 and IRS-2 levels were developmentally regulated in the mammary gland ( Lee et al. 2003b). Interestingly, the expression profile of IRSs was suggested to be similar to that of milk proteins such as β-casein and whey acidic protein ( Rosen et al. 1999), with increased expression during late pregnancy, dramatically elevated levels during lactation, and rapid loss of protein expression during involution. However, this initial study had only analyzed samples in mid-lactation and could provide no indication of how rapidly the expression of IRS-1 and IRS-2 increases during lactation. To determine whether increased IRS-1 and IRS-2 protein expression was linked to the process of secretory activation, we measured the abundance of these two proteins by immunoblotting of mammary tissue extracts prepared from mice during the immediate peri-parturient period (Fig. 1 ). These blots demonstrated a twofold increase (P < 0.05) in the abundance of both IRS-1 and IRS-2 by 2 days post partum compared with -1 day post partum (Fig. 1A ). Densitometric analysis demonstrated that both proteins appeared to reach a maximum level at 4 days post partum, while abundance of the epithelial marker keratin 8 did not change appreciably (Fig. 1B ). These results support the conclusion that the expression of both IRS-1 and IRS-2 is induced in the mammary gland during secretory activation.
The observation that IRS-1 and IRS-2 expression was induced during the immediate post partum period, suggested that these proteins might berequired for normal lactation. To test this hypothesis, lactation capacity was compared between wild-type dams and dams which were either heterozygous or null for Irs1 and/or Irs2 (Fig. 2 ). Lactation was established in these mice through the injection of pregnant mare seurm gonadotropin (PMS) and human chorionic gonadotropin (HCG) coupled with timed mating. This approach allowed for the establishment of lactation not only in the wild-type mice, but also in females that were heterozygous for either or both of the Irs mutations or null for either Irs1 or Irs2. However, although fertility was similar among Irs+/+ females and females that were either Irs1+/− , Irs2+/− , or Irs2−/− , fertility was significantly reduced in Irs1−/− females. This reduction was evident not only in conception rates, but also in litter sizes. In comparison with Irs+/+ dams, Irs1−/− dams had lower (P < 0.05) conception rates (52 vs 14% respectively) and smaller litter sizes (10 ± 1 vs 7 ± 1 respectively). This decreased fertility, though relatively minor for the first three cohorts studied, became a significant impediment to further studies of lactation in Irs1−/− dams. Consequently, lactation data from Irs1−/− dams were obtained from only the first three of the eight cohorts that were studied.
To assess lactation capacity, average pup weight during the first 10 days of lactation was measured in litters of ten pups each that were cross-fostered onto Irs+/+ dams, or dams that were heterozygous or null for either Irs1 (Fig. 2A ) or Irs2 (Fig. 2B ), or dams that were double heterozygotes (Irs1+/− Irs2+/− ; Fig. 2C ). Pup’s weight gain was decreased in litters nursed by either Irs1−/− or Irs2−/− dams. However, pup weight was unaffected in litters nursed on Irs1+/− , Irs2+/− , or Irs1+/− -Irs2+/− dams. These data support the conclusion that the loss of IRS-1 or IRS-2 diminishes lactation capacity.
Comparison of the body weights among dams of the different genotypes demonstrated that Irs1−/− dams were significantly smaller than their wild-type littermates (Fig. 2D ). Dams that were Irs2−/− had no discernable difference in body weight from Irs2+/+ dams at the beginning of lactation, but failed to undergo the same weight gain during early lactation as the Irs2+/+ and Irs2+/− dams (Fig. 2E ). Dams that were Irs1+/− Irs2+/− were also significantly smaller than their Irs+/+ dams (Fig. 2F ), yet they had exactly the same lactation capacity, indicating that differences in observed lactation capacity may not have been simply related to decreased body size.
In order to further understand the basis for the reduced pup litter weight gain on the Irs null dams, milk samples and mammary tissue were collected on day 10 post partum. Both the amount of milk recovered after oxytocin injection, and the wet weights of the number 4 mammary glands were significantly lower (P < 0.05) in Irs1−/− mice than their wild-type siblings (Fig. 3A ). However, these differences were proportional to the lowered body weight and were similar among both genotypes when expressed on a body weight basis (data not shown). For Irs2−/− dams, mammary gland and milk weight were unchanged (Fig. 3B ). In addition, comparison of hematoxylin–eosin-stained mammary tissue sections among Irs+/+ or Irs−/− dams, and Irs2+/+ or Irs2−/− dams revealed no alterations due to genotype (data not shown).
To determine whether the reduced lactational performance in the Irs1−/− dams was associated with alterations in milk composition, we measured the fat, protein, lactose, and water content of samples collected on day 10 post partum (Fig. 4 ). This analysis revealed that milk from Irs1−/− dams had significant, though modest, decreases (P < 0.05) in both water and lactose content. Fat content on the other hand tended to be higher (P < 0.1) in milk from Irs1−/− dams than that from their wild-type siblings. This result suggests that loss of IRS-1 has a modest impact on milk synthesis.
To determine whether the diminished milk synthesis capacity and altered milk lactose in Irs1−/− mice might be linked with alterations in milk protein gene expression or in mammary secretory cell carbohydrate metabolism, we used northern blotting to measure the mammary tissue abundance of the β-casein and α-lactalbumin mRNAs, we compared 2-deoxy-glucose (2DOG) uptake in primary mammary epithelial cells, and we measured the abundance of hexokinase I and II isoforms in mitochondrial and post-mitochondrial tissue fractions isolated from Irs1+/+ or Irs1−/− mice. The loss of Irs1 had no impact on the mRNA abundance of β-casein and α-lactalbumin (data not shown). Uptake of 2DOG was similar among Irs1+/+ and Irs1−/− cells (Fig. 5A ). Specificity of 2DOG uptake was measured by comparing uptake in the presence and absence of cytochalasin B. Non-specific uptake accounted for about 30% of total 2DOG uptake. As an independent method to test for possible alterations in glucose metabolism in response to loss of IRS-1, the abundance of the two enzymes hexokinases I and II was compared among mitochondrial and cytosolic fractions prepared from mammary tissue of Irs1+/+ and Irs1−/− dams. Western blotting was conducted for hexokinases I and II, COXIV, and cytokeratin 8 (Fig. 5B ). Densitometric analysis of these blots (Fig. 5C and D ) demonstrated that both the overall abundance of these enzymes and their distribution between mitochondrial and post-mitochondrial tissue fractions were similar among Irs1+/+ and Irs1−/− mice. This result supports the suggestion that loss of IRS-1 probably had little impact on the expression of milk protein genes, glucose transport, or on the hexokinase activity and subsequent glucose phosphorylation.
To determine whether the loss of IRS-1 had any impact on insulin- or IGF-I-dependent signaling within the lactating mammary gland, we measured the phosphorylation of Akt and ERK1/2 in mammary tissue collected from 3-day lactating, insulin- or IGF-I-stimulated, dams (Fig. 6 ). Immunoblotting of mammary tissue extracts with phosphospecific antibodies demonstrated increased Akt phosphorylation on both Thr308 and Ser473 in response to insulin and IGF-I (Fig. 6A and B ). This induction was greatest with insulin. However, we noted significantly lower induction of Akt phosphorylation in Irs1−/− mammary tissue compared with Irs1+/+ mammary glands. Insulin, but not IGF-I caused a modest increase in the phosphorylation of ERK1/2 on Tyr204 and Thr202. However, this induction was similar among Irs1−/− and Irs1+/+ mammary glands (Fig. 6B ). These results suggest that insulin-dependent activation of Akt was diminished in mammary tissue from Irs1−/− dams.
To determine whether the loss of IRS-1 impacted the ability of signaling pathways to activate IRS-2, we compared the phosphorylation and p85-binding activity of IRS-2 in Irs1+/+ and Irs1−/− mammary glands (Fig. 7A ). Immunoprecipitation of IRS-2 was followed by anti-phosphotyrosine immunoblotting. Densitometry on these blots revealed 20-fold higher IRS-2 tyrosine phosphorylation in mammary tissue from Irs1−/− dams than that in Irs+/+ dams (Fig. 7B ). This data therefore show that the loss of IRS-1 attenuates insulin-induced Akt phosphorylation, but that there is a compensatory increase in insulin-stimulated IRS-2 phosphorylation in IRS-1 null mammary glands.
Discussion
The requirement of insulin for normal lactation has been illustrated in animal models in a variety of ways ( Walters & McLean 1968a, b, Martin & Baldwin 1971, Kunjara et al. 1986, Lau et al. 1993). Thus, a reasonable hypothesis is that key mediators of insulin signaling might be necessary to establish or maintain a normal lactation. To date, there have been only limited attempts at establishing the functional importance of specific insulin signaling molecules to the process of lactogenesis ( Li et al. 2002, Schwertfeger et al. 2003). Our studies make several novel observations concerning the importance of IRS proteins to lactation. First, mammary expression of both IRS-1 and IRS-2 was doubled by 48 h after the onset of lactation. Secondly, the loss of both IRS-1 and IRS-2 caused decreased lactation capacity. Thirdly, the loss of IRS-1 caused a modest decrease in both milk water content and lactose concentration. Fourthly, the loss of IRS-1 decreases insulin-dependent induction of Akt phosphorylation. Fifthly, the loss of IRS-1 resulted in an increase in the phosphorylation of IRS-2 and increased association of IRS-2 with the p85 regulatory subunit of PI3 kinase. Finally, these changes in lactation capacity, milk composition, and mammary cell signaling were associated with little or no change in milk protein gene expression, mammary glucose transport, or in the abundance and subcellular localization of hexokinases I and II. Taken together, the results support our hypothesis that IRS-1 and IRS-2 are important for normal lactation, but they also suggest that compensatory mechanisms within the secretory epithelium may prevent the appearance of a more dramatic phenotype in animals with a single deletion of either IRS-1 or IRS-2.
The induction of IRS protein expression during the first few days of lactation is consistent with previous studies published in our laboratory which demonstrated higher levels of IRS protein expression during mid-lactation compared with late pregnancy ( Lee et al. 2003b). The immunoblot results in the current study extend that observation by demonstrating that this upregulation of IRS-1 and IRS-2 is an early event that is temporally linked with the process of secretory activation. The published results on other components of the insulin signaling pathway have also demonstrated increased expression of both the insulin receptor and Akt with the onset of lactation, suggesting that the coordinated upregulation of multiple insulin signaling pathway components may occur during the early post partum period ( Flint 1982, Schwertfeger et al. 2003).
The fact that the loss of IRS-1 or IRS-2 resulted in decreased lactation capacity, as measured by the ability of the dams to support the growth of cross-fostered litters, also supports the suggestion that these two proteins are important to lactation. However, neither one alone was absolutely necessary to lactation since null animals could still lactate albeit at reduced capacity. This observation coupled with the observation that tyrosine phosphorylation of IRS-2 was increased in the mammary tissue of IRS-1 knockout dams supports the idea that compensatory interactions exist between IRS-1 and IRS-2 within the mammary gland. Although this phenomenon of compensatory interaction between IRS-1 and IRS-2 has been demonstrated in earlier studies, it appears to be tissue specific ( Araki et al. 1994, Yamauchi et al. 1996). In other studies on compensatory interactions between IRS-1 and IRS-2, the phenomenon was clearly indicated by genetic crosses in which both genes were targeted in the same animal ( Withers et al. 1999, Kido et al. 2000, Miki et al. 2001). Although similar crosses were made for the purpose of testing the extent of compensatory interactions between IRS-1 and IRS-2 in the mammary gland, only double heterozygous progeny were fertile.
The fact that body weight was significantly decreased in both the IRS-1 and IRS-2 null dams suggests that at least some of the lactation phenotype may be attributable to the fact that the dams are smaller. In this regard, there is clearly a link between mature body weight and lactation capacity not only in mice but also within other species. More specifically, selection for milk production capacity produces larger animals in both mice and cows ( Nagai & Sarkar 1978, Hansen 2000). However, the difference in body weight and lactation capacity in the Irs1−/− and Irs2−/− mice is somewhat mitigated by the fact that Irs1+/− Irs2+/− mice were significantly smaller than their Irs1+/+Irs2+/+littermates, but both sets of mice had identical lactation capacity. This result would indicate that either there is a threshold for the effect of body weight on lactation capacity or that at least part of the lactation defect in the Irs1−/− dams was intrinsic to the mammary gland. Subsequent analysis of the mammary tissue from these mice was intended to determine whether the lactation defect could be at least partially attributable to mammary-specific effects. Although not all of these measurements supported the presence of a mammary-specific defect in the Irs1−/− mice, one of them did. The phosphorylation of mammary Akt in response to an insulin or IGF-I challenge was decreased in Irs1−/− mice.
The fact that IRS-1 dams displayed reduced lactation capacity along with decreased concentrations of milk lactose suggested that the mammary cells isolated from these mice could have had decreased glucose transport capacity or altered glucose metabolism. In mammary secretory cells, insulin does not acutely stimulate glucose transport, but is required for the maintenance of transport capacity ( Prosser et al. 1987, Nemeth et al. 2000). In addition, the expression of hexokinase II increases dramatically during secretory activation and treatment of lactating rats with neutralizing antisera to insulin has been shown to cause an alteration in the subcellular distribution of mammary hexokinase activity in conjunction with a reduction in glycolytic activity ( Walters & McLean 1968a, b). Consequently, we were somewhat surprised to find that the loss of IRS-1 had no perceptible impact on mammary on glucose uptake or mitochondrial hexokinase abundance. This surprise was even greater in light of the fact that one of the mammary-specific endpoints which was affected by the loss of IRS-1 was insulin-dependent induction of Akt phosphorylation.
The Irs2 gene was originally identified because of a residual substrate protein that was phosphorylated in response to insulin in Irs1−/− mice ( Kadowaki et al. 1996). Subsequent studies have demonstrated numerous instances of compensation between IRS-1 and IRS-2, however, the proteins clearly have both unique and conserved function. We found a compensatory increase in IRS-2 activity in insulin-stimulated IRS-1 null mammary glands. However, despite this induction of IRS-2 phosphorylation, the loss of IRS-1 was clearly associated with decreased insulin-dependent phosphorylation of Akt at both Thr308 and Ser473. The loss of insulin signaling in Irs1−/− mice has previously been reported for a number of tissues ( Valverde et al. 1999, Ueki et al. 2000). However, some tissues such as white adipose tissue or brown preadipocytes are still capable of exhibiting insulin-stimulated Akt phosphorylation even with the loss of IRS-1. Although the basis for this tissue specificity is not clear, it appears that the mammary cells depend on IRS-1 for insulin-dependent Akt phosphorylation.
In summary, we have shown that both Irs1−/− and Irs2−/− mice show reduced lactation capacity. The loss of IRS-1 is associated with minor alterations in milk composition but no change in milk protein mRNAs, glucose transport, or the abundance and subcellular localization of hexokinases I and II. The minor lactation phenotype may in part be explained by an apparent compensatory increase in insulin-induced IRS-2 phosphorylation. Further studies are required to determine the impact of loss of both IRS-1 and IRS-2 on lactation capacity.
IRS-1 and IRS-2 protein levels are rapidly increased during secretory activation. Protein abundance of IRS-1 and IRS-2 was measured by immunoblotting (A) of mammary tissue extracts prepared from mice on days 17 (−3) and 19 (−1) of pregnancy and on days 1, 2, 3, 4, and 5 post partum. Immunoblotting of lysates for cytokeratin 8 (K8) was used to control for loading. Densitometry data collected from the immunoblots were used to quantify changes in abundance with time (B). Each bar represents the lsmeans ± s.e.m. for three animals. *represents P < 0.05.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Lactation performance is reduced in Irs1−/− and Irs2−/− dams, but not in Irs1+/− Irs2+/− dams. Litter weight gain was measured for CD-1 cross-fostered litters (ten pups) that were nursed by dams that were wild-type (+/+), heterozygous (+/−) or null (−/− ) for Irs1 (A) or Irs2 (B), or heterozygous for both Irs1 and Irs2 (C). Body weight was also compared among dams that were wildtype, heterozygous or null for Irs1 (D), Irs2 (E), or heterozygous for both Irs1 and Irs2 (F). Each symbol represents the lsmean ± s.e.m. for 4–13 animals. Asterisks indicate either a significant (P < 0.05) main effect of genotype (D and F), or a significant day-by-genotype interaction (A, B, and E).
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Mammary gland wet weight, and milk volume recovered following oxytocin administration, is decreased in Irs1−/− dams. Wet weight of the number 4 mammary gland and weight of milk recovered from all ten glands following oxytocin treatment were compared among Irs+/+and Irs1−/− dams (A), Irs2+/+and Irs2−/− dams (B), and Irs1+/+Irs2+/+and Irs1+/− Irs2+/− dams (C). Each bar represents the mean ± s.e.m. for 4–13 mice. Asterisks indicate a significant decrease (P < 0.05) in comparison with wildtype.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Water and lactose are modestly reduced in milk from Irs1−/− dams. Water, fat, lactose, and protein were measured in milk samples collected from Irs1+/+and Irs1−/− dams (A), and from Irs2+/+and Irs2−/− dams (B) on day 10 post partum (A). Each bar represents the mean ± s.e.m. for 5 and 11 Irs1−/− and Irs1+/+ dams, and 5 and 11 Irs2+/+ and Irs2−/− dams respectively. Asterisks indicate a statistically significant (P < 0.05) decrease.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Mammary cell glucose uptake and mammary abundance and localization of hexokinases (HKs) I and II are unaffected by loss of the Irs1 gene. Mammary epithelial cells were isolated from mice at 16 days of pregnancy and then plated in serum fetuin-coated plastic six-well plates. Total glucose uptake was measured by incubating the cells with 2DOG for 15 min (A). Cytochalasin B was included in some of the wells to provide a measurement of non-specific uptake. Western blotting (B) of mammary mitochondrial and cytoplasmic fractions was used to determine whether the loss of IRS-1 had any impact on HK localization. Purity of the fractions was demonstrated by probing the blots for cytochrome c oxidase subunit IV (COXIV) and tubulin. Densitometry was used to quantify the mitochondrial (mito) and cytoplasmic (cyto) abundance of HKI (C) and HKII (D) in fractions prepared from Irs1+/+ and Irs1−/− mammary tissue. In (A), each bar represents the mean ± s.e.m. for three wells. In (C) and (D), each bar represents the mean ± s.e.m. for five mice.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Insulin- and LR3-IGF-I-dependent phosphorylation of Akt is decreased in mammary tissue from Irs1−/− dams. Protein phosphorylation was measured by western blotting of mammary tissue extracts prepared from 3-day lactating Irs1+/+(+/+) or Irs1−/− (−/−) mice. On day 3 post partum, the mice were fasted for 8 h, separated from their litters for 2 h, and then given a tail-vein injection of either saline, insulin, or LR3–IGF-I (LR3). Tissue was harvested at 5 min following injection. Phosphorylation of Akt was measured with antibodies specific for Ser473 (A) and Thr308 (B). Phosphorylation of MAPK4/3, ERK 1/2 (B) was measured with an antibody that detects dual phosphorylation at Thr202 and Tyr204. Densitometry (C, D, and E) was used to quantitate the signal intensities in each of the blots. Each bar represents the mean ± s.e.m. for two to four mice. * indicate a statistically significant (P < 0.05) genotype effect.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
Increased tyrosine phosphorylation and p85 binding of IRS-1 in mammary tissue from lactating Irs1−/− females. Tyrosine phosphorylation (pY) of IRS-2 and association of IRS-2 with p85 were measured by western blotting of IRS-2 immunoprecipitates prepared from insulin-injected Irs1+/+ or Irs1−/− mice (A). Each lane is loaded with an extract from one animal. Densitometry analysis (B) demonstrated that tyrosine phosphorylation, but not p85 association, was increased for IRS-2 in the Irs1−/− mammary tissue. The normalized ratio is the ratio of integrated optical density for phosphotyrosine to that of total IRS-2. This ratio was 20-fold greater in Irs1−/− mammary tissue than in Irs+/+ mammary tissue. Each bar represents the mean ± s.d. for two mice.
Citation: Journal of Endocrinology 194, 2; 10.1677/JOE-07-0160
The authors thank Dr C Ronald Kahn for providing the Irs1 null mice. They also thank Dr Xiaojiang Cui for help with the immunoprecipitation analysis.
Funding This work was supported by USDA/ARS cooperative agreement no. 58-6250-6001 (DLH), NIH grant no. DK52197 (DLH), and NIH grant no. CA9411B-01 (AVL). This work is a publication of the United States Department of Agriculture/Agricultural Research Service Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine Texas Children’s Hospital, Houston, TX, USA. The contents of this publication do not necessarily reflect the views or policies of the United States Department of Agriculture nor does the mention of any trade names, commercial products or organizations imply endorsement by the United States Government. There is no conflict of interest that would prejudice the impartiality of this publication.
References
Aggeler J, Ward J, Blackie LM, Barcellos-Hoff MH, Streuli CH & Bissell MJ 1991 Cytodifferentiation of mouse mammary epithelial cells cultured on a reconstituted basement membrane reveals striking similarities to development in vivo. Journal of Cell Science 99 407–417.
Araki E, Lipes MA, Patti ME, Brunning JC, Haag B, III, Johnson RS & Kahn CR 1994 Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature 372 186–190.
Bauman DE, Mellenberger RW & Ingle DL 1974 Metabolic adaptations in fatty acid and lactose biosynthesis by sheep mammary tissue during cessation of lactation. Journal of Dairy Science 57 719–723.
Darley-Usmar VM, Rickwood D & Wilson MT 1987 Mitochondria: A Practical Approach., Washington, DC: Oxford Press.
Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y & Greenberg ME 1997 Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91 231–241.
Flint DJ 1982 Insulin binding to rat mammary gland at various stages of cell isolation and purification. Molecular and Cellular Endocrinology 26 281–294.
Gottlob K, Majewski N, Kennedy S, Kandel E, Robey RB & Hay N 2001 Inhibition of early apoptotic events by Akt/PKB is dependent on the first committed step of glycolysis and mitochondrial hexokinase. Genes and Development 15 1406–1418.
Hadsell DL, Alexeenko T, Klemintidis Y, Torres D & Lee AV 2001 Inability of overexpressed des(1–3)human insulin-like growth factor I (IGF-I) to inhibit forced mammary gland involution is associated with decreased expression of IGF signaling molecules. Endocrinology 142 1479–1488.
Hadsell DL, Bonnette S, George J, Torres D, Klementidis Y, Gao S, Haney PM, Summy-Long J, Soloff MS, Parlow AF et al.2003 Diminished milk synthesis in upstream stimulatory factor 2 null mice is associated with decreased circulating oxytocin and decreased mammary gland expression of eukaryotic initiation factors 4E and 4G. Molecular Endocrinology 17 2251–2267.
Hadsell DL, Torres DT, Lawrence NA, George J, Parlow AF, Lee AV & Fiorotto ML 2005 Overexpression of des(1–3) insulin-like growth factor 1 in the mammary glands of transgenic mice delays the loss of milk production with prolonged lactation. Biology of Reproduction 73 1116–1125.
Hansen LB 2000 Consequences of selection for milk yield from a geneticist’s viewpoint. Journal of Dairy Science 83 1145–1150.
Kadowaki T, Tamemoto H, Tobe K, Terauchi Y, Ueki K, Kaburagi Y, Yamauchi T, Satoh S, Sekihara H, Aizawa S et al.1996 Insulin resistance and growth retardation in mice lacking insulin receptor substrate-1 and identification of insulin receptor substrate-2. Diabetic Medicine 13 S103–S108.
Kaselonis GL, McCabe ER & Gray SM 1999 Expression of hexokinase 1 and hexokinase 2 in mammary tissue of nonlactating and lactating rats: evaluation by RT-PCR. Molecular Genetics and Metabolism 68 371–374.
Kido Y, Burks DJ, Withers D, Bruning JC, Kahn CR, White MF & Accili D 2000 Tissue-specific insulin resistance in mice with mutations in the insulin receptor, IRS-1, and IRS-2. Journal of Clinical Investigation 105 199–205.
Kubota N, Tobe K, Terauchi Y, Eto K, Yamauchi T, Suzuki R, Tsubamoto Y, Komeda K, Nakano R, Miki H et al.2000 Disruption of insulin receptor substrate 2 causes type 2 diabetes because of liver insulin resistance and lack of compensatory beta-cell hyperplasia. Diabetes 49 1880–1889.
Kunjara S, Sochor M, Salih N, McLean P & Greenbaum AL 1986 Phosphoribosyl pyrophosphate and phosphoribosyl pyrophosphate synthetase in rat mammary gland. Changes in the lactation cycle and effects of diabetes, insulin and phenazine methosulphate. Biochemical Journal 238 553–559.
Lau C, Sullivan MK & Hazelwood RL 1993 Effects of diabetes mellitus on lactation in the rat. Proceedings of the Society for Experimental Biology and Medicine 204 81–89.
Lee AV, Taylor ST, Greenall J, Mills JD, Tonge DW, Zhang P, George J, Fiorotto ML & Hadsell DL 2003a Rapid induction of IGF-IR signaling in normal and tumor tissue following intravenous injection of IGF-I in mice. Hormone and Metabolic Research 35 651–655.
Lee AV, Zhang P, Ivanova M, Bonnette S, Oesterreich S, Rosen JM, Grimm S, Hovey RC, Vonderhaar BK, Kahn CR et al.2003b Developmental and hormonal signals dramatically alter the localization and abundance of insulin receptor substrate proteins in the mammary gland. Endocrinology 144 2683–2694.
Li G, Robinson GW, Lesche R, Martinez-Diaz H, Jiang Z, Rozengurt N, Wagner KU, Wu DC, Lane TF, Liu X et al.2002 Conditional loss of PTEN leads to precocious development and neoplasia in the mammary gland. Development 129 4159–4170.
Macheda ML, Williams ED, Best JD, Wlodek ME & Rogers S 2003 Expression and localisation of GLUT1 and GLUT12 glucose transporters in the pregnant and lactating rat mammary gland. Cell and Tissue Research 311 91–97.
Mandel D, Lubetzky R, Dollberg S, Barak S & Mimouni FB 2005 Fat and energy contents of expressed human breast milk in prolonged lactation. Pediatrics 116 e432–e435.
Martin RJ & Baldwin RL 1971 Effects of insulin and anti-insulin serum treatments on levels of metabolites in rat mammary glands. Endocrinology 88 868–871.
Mellenberger RW & Bauman DE 1974a Metabolic adaptations during lactogenesis. Fatty acid synthesis in rabbit mammary tissue during pregnancy and lactation. Biochemical Journal 138 373–379.
Mellenberger RW & Bauman DE 1974b Metabolic adaptations during lactogenesis. Lactose synthesis in rabbit mammary tissue during pregnancy and lactation. Biochemical Journal 142 659–665.
Mendez R, Myers MG, Jr, White MF & Rhoads RE 1996 Stimulation of protein synthesis, eukaryotic translation initiation factor 4E phosphorylation, and PHAS-I phosphorylation by insulin requires insulin receptor substrate 1 and phosphatidylinositol 3-kinase. Molecular and Cellular Biology 16 2857–2864.
Miki H, Yamauchi T, Suzuki R, Komeda K, Tsuchida A, Kubota N, Terauchi Y, Kamon J, Kaburagi Y, Matsui J et al.2001 Essential role of insulin receptor substrate 1 (IRS-1) and IRS-2 in adipocyte differentiation. Molecular and Cellular Biology 21 2521–2532.
Nagai J & Sarkar NK 1978 Relationship between milk yield and mammary gland development in mice. Journal of Dairy Science 61 733–739.
Nakhasi HL & Qasba PK 1979 Quantitation of milk proteins and their mRNAs in rat mammary gland at various stages of gestation and lactation. Journal of Biological Chemistry 254 6016–6025.
Nemeth BA, Tsang SW, Geske RS & Haney PM 2000 Golgi targeting of the GLUT1 glucose transporter in lactating mouse mammary gland. Pediatric Research 47 444–450.
Neville MC & Morton J 2001 Physiology and endocrine changes underlying human lactogenesis II. Journal of Nutrition 131 3005S–3008S.
Neville MC, Morton J & Umemura S 2001 Lactogenesis: the transition from pregnancy to lactation. Pediatric Clinics of North America 48 35–52.
Neville MC, McFadden TB & Forsyth I 2002 Hormonal regulation of mammary differentiation and milk secretion. Journal of Mammary Gland Biology and Neoplasia 7 49–66.
Nguyen DA, Parlow AF & Neville MC 2001 Hormonal regulation of tight junction closure in the mouse mammary epithelium during the transition from pregnancy to lactation. Journal of Endocrinology 170 347–356.
Prosser CG, Sankaran L, Hennighausen L & Topper YJ 1987 Comparison of the roles of insulin and insulin-like growth factor I in casein gene expression and in the development of alpha-lactalbumin and glucose transport activities in the mouse mammary epithelial cell. Endocrinology 120 1411–1416.
Rosen JM & Barker SW 1976 Quantitation of casein messenger ribonucleic acid sequences using a specific complementary DNA hybridixation probe. Biochemistry 15 5272–5280.
Rosen JM, O’Neal DL, McHugh JE & Comstock JP 1978 Progesterone-mediated inhibition of casein mRNA and polysomal casein synthesis in the rat mammary gland during pregnancy. Biochemistry 17 290–297.
Rosen JM, Wyszomierski SL & Hadsell D 1999 Regulation of milk protein gene expression. Annual Review of Nutrition 19 407–436.
Ruderman NB, Kapeller R, White MF & Cantley LC 1990 Activation of phosphatidylinositol 3-kinase by insulin. PNAS 87 1411–1415.
Schwertfeger KL, McManaman JL, Palmer CA, Neville MC & Anderson SM 2003 Expression of constitutively activated Akt in the mammary gland leads to excess lipid synthesis during pregnancy and lactation. Journal of Lipid Research 44 1100–1112.
Slaby F & Brown C 1974 Changes in the ribosome content, principal microsomal protein composition, and secretory character of mammary epithelial rough endoplasmic reticulum during differentiation. Evidence that messenger RNAs specific for milk proteins are incorporated into rough endoplasmic reticulum formed de novo after parturition. Journal of Cell Biology 61 613–632.
Sochor M, Greenbaum AL & McLean P 1984 Fructose 2,6-bisphosphate, sugar phosphates and adenine nucleotides in the regulation of glucose metabolism in the lactating rat mammary gland. FEBS Letters 169 12–16.
Thirone AC, Huang C & Klip A 2006 Tissue-specific roles of IRS proteins in insulin signaling and glucose transport. Trends in Endocrinology and Metabolism 17 72–78.
Ueki K, Yamauchi T, Tamemoto H, Tobe K, Yamamoto-Honda R, Kaburagi Y, Akanuma Y, Yazaki Y, Aizawa S, Nagai R et al.2000 Restored insulin-sensitivity in IRS-1-deficient mice treated by adenovirus-mediated gene therapy. Journal of Clinical Investigation 105 1437–1445.
Valverde AM, Kahn CR & Benito M 1999 Insulin signaling in insulin receptor substrate (IRS)-1-deficient brown adipocytes: requirement of IRS-1 for lipid synthesis. Diabetes 48 2122–2131.
Walters E & McLean P 1968a Effect of alloxan-diabetes and treatment with anti-insulin serum on pathways of glucose metabolism in lactating rat mammary gland. Biochemical Journal 109 407–417.
Walters E & McLean P 1968b The effect of anti-insulin serum and alloxan-diabetes on the distribution and multiple forms of hexokinase in lactating rat mammary gland. Biochemical Journal 109 737–741.
Withers DJ, Burks DJ, Towery HH, Altamuro SL, Flint CL & White MF 1999 Irs-2 coordinates Igf-1 receptor-mediated beta-cell development and peripheral insulin signalling. Nature Genetics 23 32–40.
Yamauchi T, Tobe K, Tamemoto H, Ueki K, Kaburagi Y, Yamamoto-Honda R, Takahashi Y, Yoshizawa F, Aizawa S & Akanuma Y 1996 Insulin signalling and insulin actions in the muscles and livers of insulin-resistant, insulin receptor substrate 1-deficient mice. Molecular and Cellular Biology 16 3074–3084.
Yenush L & White MF 1997 The IRS-signalling system during insulin and cytokine action. Bioessays 19 491–500.
