Introduction
Breast cancer is the most common malignancy for women worldwide, with more than one million women diagnosed each year [
1‐
3]. Although genetic mutations increase the risk for development of breast cancer, this impacts only a small percentage of women. Reproductive and lifestyle factors increasingly are implicated to play a role in breast cancer development [
4]. Obesity is one of the lifestyle factors thought to affect the risk for postmenopausal breast cancer [
5] and is associated with increased mortality in women diagnosed with breast cancer [
6,
7]. Obesity, in turn, impacts levels of serum estrogen, insulin, and/or insulin-like growth factor-1, which have all been implicated as possible links between obesity and breast cancer [
8‐
12]. Leptin, a protein secreted primarily from adipose tissue [
13‐
16], also has been implicated as a mammary tumor (MT) growth factor. For example, increased cell proliferation was observed when human breast cancer cells were incubated with leptin in
in vitro studies [
17‐
19]. Additionally, the leptin receptor was identified in breast cancer cell lines and in human breast cancer tissues [
17‐
19].
Animal models are frequently used to gain insights into human disease processes. In general, higher body weights and/or obesity has been associated with shortened MT latency and/or increased incidence for development of spontaneous and carcinogen-induced MT in animals [
20‐
25]. An exception noted was that genetically obese
Lep
ob
Lep
ob
mice were reported to have a decreased incidence of spontaneous MTs [
26], and recently our laboratory found that when
Lep strain mice were crossbred with transgenic MMTV-TGF-α mice, obese MMTV-TGF-α-
Lep
ob
Lep
ob
mice did not develop oncogene-induced MTs, although their lean littermates did [
27]. The metabolic defect in
Lep
ob
Lep
ob
mice is an absence of leptin [
28]. In a second study,
Lepr mice that exhibit a mutation in the leptin receptor were crossbred with MMTV-TGF-α mice and again no MTs were detected in obese MMTV-TGF-α-
Lepr
db
Lepr
db
mice [
29]. One could argue that MMTV-TGF-α MTs are not affected by obesity; but in an additional experiment, MMTV-TGF-α mice with diet-induced obesity had shortened MT latency [
30]. Thus, in the presence of a functioning leptin axis, MTs develop. These findings led us to hypothesize that leptin is a growth factor for breast/mammary cancer cells.
In the present study, we used a dietary-induced obesity protocol to further explore the relationship of serum leptin levels and body weight with respect to MT development in MMTV-TGF-α mice. Leptin signaling proteins in MT and mammary fat pad (MFP) tissues were characterized. Although six different leptin receptor subtypes have been identified in numerous tissues and cell types, primarily the long isoform (ObRb) is responsible for activating leptin signaling pathways [
31,
32]. In this study, we determined expression of the total leptin receptor (ObR) as well as ObRb. Leptin acts through several signaling pathways, including Jak2/STAT3 (Janus kinase 2/signal transducer and activator of transcription 3) and PI3K/Akt [
33‐
36], and here we determined protein and/or mRNA expression levels for these pathways. Another metabolic action of leptin that has been reported is inhibition of apoptosis in human colon [
37] and prostate [
38] cancer cell lines and in leukemic cells [
39]. Thus, proteins involved in apoptosis (that is, total poly [ADP-ribose]polymerase [PARP], PARP [89 kDa], PARP [24 kDa], Bax, Bcl-2, and Bcl-xL) as well as caspase-3 activity were determined in MT and MFP tissues in the present study.
Materials and methods
Materials
Primary antibodies against Ob, ObR, phosphorylated STAT3 (pSTAT3), Jak2, phosphorylated Akt (pAkt), Bcl-2, Bcl-xL, Bax, and alkaline phosphatase-conjugated goat anti-mouse immunoglobulin G (IgG) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). ObRb was purchased from Linco Research, Inc. (now part of Millipore Corporation, Billerica, MA, USA). PARP (#9542) and anti-rabbit IgG conjugated with alkaline phosphatase were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). β-Actin was purchased from Delta Biolabs, LLC (Gilroy, CA, USA). Enhanced chemifluorescence (ECF) substrate was obtained from Amersham Biosciences (now part of GE Healthcare, Little Chalfont, Buckinghamshire, UK). Tris-base solution, Tris/Glycine/SDS buffer, and polyacrylamide gradient gels were purchased from Bio-Rad Laboratories, Inc. (Hercules, CA, USA). Reverse transcriptase (Superscript II) and oligo(dT) were purchased from GibcoBRL (now part of Invitrogen Corporation, Carlsbad, CA, USA). Polyvinylidene difluoride (PVDF) membranes (Immobilon-P) were from Millipore Corporation. Immunoglobulin-alcaline phosphatase-conjugated secondary antibodies were purchased from Amersham Life Science Inc. (now part of GE Healthcare).
Animals and study design
MMTV-TGF-α (C57BL6) female mice were produced and genotyped at the Hormel Institute (Austin, MN, USA) as previously described [
40]. Mice were randomly assigned at 10 weeks of age to either a Low-Fat group fed commercial rodent diet (3.3 kcal/g) (
n = 20) or to a High-Fat group fed a moderately high-fat diet consisting of 48% rodent diet, 44% condensed milk, and 8% corn oil (4.47 kcal/g) (
n = 50). Mice were individually caged, and food intakes and body weights were determined weekly, at which time mice were palpated to identify the presence of MTs. Once MTs were detected, growth was monitored with calipers. Mice fed the high-fat diet with final body weights ± 2 standard deviations of Low-Fat mice were classified as Obesity-Resistant (
n = 17). The remaining High-Fat mice were designated as Obesity-Prone (
n = 33). Mice were euthanized when MT size reached 20 mm or at 85 weeks of age. When possible, blood samples were obtained by cardiac puncture before the mice were euthanized and serum prepared for leptin measurement. Serum leptin levels were determined using a commercial kit in accordance with the manufacturer's protocol (Linco Mouse Leptin Kit; Millipore Corporation). MTs and parametrial and retroperitoneal fat pads were removed and weighed. MFP samples were obtained but not weighed. MTs and tissue samples that appeared abnormal were preserved in formalin and sent to the Department of Pathology and Laboratory Medicine of the Mayo Foundation (Rochester, MN, USA) for histopathological analyses to determine malignancy and/or disease status. The remaining tissue was stored at -80°C until used. All procedures with mice were performed under the guidelines and with approval of the University of Minnesota Institutional Animal Care and Use Committee.
Western blot analysis of mammary tumor and mammary fat pad proteins
Tissue samples were homogenized in extraction buffer with protease inhibitors. Total protein was extracted using a Total Protein Extraction Kit (Chemicon International, Temecula, CA, USA) and quantitated using a Bio-Rad protein assay kit with bovine serum albumin as a standard (Bio-Rad Laboratories, Inc.). Extracted proteins were electrophoresed on 4% to 15% polyacrylamide gradient gels and then transferred to a PVDF membrane. Blots were blocked in Tris-Base solution containing 1% milk concentrate and 0.15% Tween-20. The PVDF membranes were incubated with appropriate primary antibodies against leptin signaling proteins (leptin, long [ObRb] and total [ObR] forms of the leptin receptors, Jak2, pAkt, and pSTAT3) and apoptosis signaling proteins (Bcl-2, Bcl-xL, Bax, and PARP) in MFP and MT samples. YES Membranes were next incubated with a secondary antibody conjugated to alkaline phosphatase. ECF substrate was used to visualize the bands using a Storm 840 Machine Imaging System (Molecular Dynamics, Sunnyvale, CA, USA). Standard molecular weight markers were run simultaneously for comparing molecular weights of the visualized proteins. The intensity of Western blot bands was quantified by densitometric analysis using the program UN-SCAN-IT gel (Silk Scientific, Inc., Orem, UT, USA). Results were expressed as the ratio of intensity of the protein of interest to that of β-actin from the same sample. Although some mice had more than one MT, only one MT per animal was analyzed. For each measurement, 'n' is 3 to 6 for MT and 4 to 6 for MFP samples, as indicated in figure legends. Because no differences were found for MT and MFP samples from Obesity-Prone and Obesity-Resistant mice, these results were combined and designated as High-Fat.
Measurement of caspase-3 activity in mammary tumor and mammary fat pad
Caspase-3 activity was determined as described in the manufacturer's protocol (R&D Systems, Inc., Minneapolis, MN, USA), using the same protein sample prepared for Western blot analysis. This assay is based on detection of cleavage of the substrate DEVD-AFC. The assay was carried out in 96-well flat-bottom microplates. Briefly, 100 μg of protein for MT and 150 μg of protein for MFP samples were incubated with 200 μM Ac-DEVD-pNA caspase-3 substrate in 50 μL of caspase assay buffer at 37°C for 4 hours. The release of pNA was measured at 400-nm excitation and 505-nm emission using a fluorometric Fluoroskan Ascent plate reader (Labsystem, now part of Thermo Fisher Scientific, Inc. Waltham, MA, USA). The fluorescence value obtained with the substrate, but without sample protein, was used as a background/negative control. Caspase-3 activity levels were calculated by subtracting the background value from the sample value. Each sample was run in duplicate. Values from the Low-Fat group were taken as 100%, and results from the High-Fat group were calculated relative to the Low-Fat group as a percentage.
RT-PCR analysis of mRNA expression in mammary tumor and mammary fat pad samples
mRNA expression levels of leptin and long form (ObRb) and total form (ObR) of leptin receptors and STAT3 were measured in MT and MFP samples using reverse transcription-polymerase chain reaction (RT-PCR) analysis. Briefly, total RNA was extracted from equal amounts of frozen tissue samples using an RNeasy mini kit in accordance with the manufacturer's protocol (Qiagen Inc., Valencia, CA, USA). Isolated total RNA was quantified using a spectrophotometer, and equal amounts of RNA from each tissue were used for RT-PCR. Total RNA was reverse-transcribed using random hexamers, RNase inhibitor, reverse transcriptase (Superscript II), and oligo(dT) primers to synthesize first-strand cDNA from mRNA (Invitrogen Corporation). Total volume was adjusted to 50 μL using DNase- and RNase-free water. PCR primers were designed using nucleotide sequence for mouse ObR (GenBank accession number: U42467), ObRb (GenBank accession number: U58861), leptin (GenBank accession number: NM_008493), and STAT3 (GenBank accession number: U06922). The following primers were used: ObR, forward 5'-CAGATTCGATATGGCTTAAGT-3' and reverse 5'-GTTAAAATTCACAAGGGAAGC-3'; ObRb, forward 5'-ACACTGTTAATTTCACACCAGAG-3' and reverse 5'-TGGATAAACCCTTGCTCTTCA-3'; leptin, forward 5'-CCTGTGGCTTTGGTCCTATCTG-3' and reverse 5'-AGGCAGGCTGGTGAGGACCTG-3'; and STAT3, forward 5'-CAGGGTGTCAGATCACATGG-3' and reverse 5'-TTATTTCCAAACTGCATCAATG-3'.
These primers gave final product sizes of 475, 446, 244, and 600 base pairs (bp) for ObR, ObRb, leptin, and STAT3, respectively. PCR was performed under the following conditions: 94°C for 4 minutes for denaturing, 35 cycles of 94°C for 30 seconds, 62°C for 30 seconds, 72°C for 1 minute, and final extension at 72°C for 10 minutes. Reactions were carried out in a final volume of 25 μL in a thermocycler in the presence of Taq DNA polymerase. Water was used as the negative control, and mouse β-actin primers that produce a product of approximately 600 bp were used as internal controls. PCR products were separated on 1% agarose gel and stained with ethidium bromide, and product size was determined by concurrently separating 100-bp DNA ladder on the same gel. The DNA gel was scanned, the intensity of bands was quantified by densitometry using the program UN-SCAN-IT gel (Silk Scientific, Inc.), and results were expressed as the ratio of intensity of the gene of interest to that of β-actin in samples from individual mice. DNA obtained from tissue samples was run on agarose gel and subsequently extracted. The samples were sent to the Advanced Genetic Analysis Center of the University of Minnesota (St. Paul, MN, USA) for sequencing to confirm that the bands analyzed are the genes of interest (that is, ObR, ObRb, leptin, and STAT3).
Statistical analysis
Results are presented as mean ± standard error of the mean (SEM). Tumor incidence data were analyzed by the chi-square test and Fisher exact test, and latency data were analyzed by the proportional hazard model using the Low-Fat group as the reference. Comparisons among three groups were made by analysis of variance (ANOVA) followed by the Newman-Keuls test to determine differences between specific groups. When only High-Fat versus Low-Fat groups were compared, the Student t test was used. The Pearson correlation test was used to determine whether the correlation between two parameters was significant.
Discussion
In the present study, we demonstrate that mice fed a high-fat diet from 10 to 85 weeks of age and designated Obesity-Prone had significantly higher body and fat pad weights and higher serum leptin levels in comparison with mice fed the same high-fat diet and designated as Obesity-Resistant or Low-Fat. In addition, MT incidence and MT weights were higher in Obesity-Prone mice compared with both Obesity-Resistant and Low-Fat mice, and palpable MT latency was shortened in Obesity-Prone mice compared with Low-Fat mice. Interestingly, there were also differences in comparisons between the Low-Fat and Obesity-Resistant mice despite their similar body weights, such as a trend toward higher palpable MT incidence and higher fat pad weights and serum leptin levels in the Obesity-Resistant mice.
The findings of higher MT weights and presence of high-grade adenocarcinomas in the Obesity-Prone mice indicate similarities to the relationship of obesity with human breast cancer. For example, obesity has been associated with bigger tumors, more advanced disease, poorer prognosis, and/or increased mortality in women with breast cancer [
6]. The shortened MT latency reported for Obesity-Prone mice is consistent with human studies relating higher body weight/body mass index with increased risk of breast cancer. For example, the presence of obesity would promote the development of breast cancer sooner in susceptible women, who if they had stayed lean might eventually die of causes other than breast cancer. Also, if breast cancer did develop at a later age, its consequences may have less impact on shortening life expectancy and/or affecting quality of life. Furthermore, our results suggest that body fat independent of body weight and obesity may be an important factor in MT development as indicated by differences between Obesity-Resistant and Low-Fat mice. This was seen by the many intermediate measurements for the Obesity-Resistant mice and by the fact that gene expression data were impacted more by consumption of the high-fat diet that increased body fat levels.
A major focus of the present study was to determine the impact of diet and/or body weight on expression of leptin signaling proteins in MFP and MT tissues. This was of interest as a consequence of earlier studies indicating that the absence of leptin and its receptor were associated with a lack of MT development in transgenic MMTV-TGF-α mice [
27,
29]. The fact that the leptin receptor was identified in human breast tumors and in human breast cancer cell lines and that the addition of leptin to breast cancer cell lines increased cell proliferation [
17‐
19,
41‐
43] further supported the involvement of leptin in mammary tumorigenesis. Also, while this long-term mouse experiment was under way, additional publications further supported that leptin stimulates proliferation and affects signal transduction in breast cancer cell lines [
41‐
45].
Despite the increasing evidence that leptin impacts human breast cancer cells
in vitro, the role of leptin in the development of human breast cancer
per se is not clear. Some studies have reported that breast cancer subjects have higher serum leptin levels than controls, but others have not [
46‐
50]. Since some of these studies focused on younger subjects or combined pre- and postmenopausal women and subject numbers were small, it is not surprising that inconclusive results have been obtained so far. Additionally, as in the present study, serum was obtained at only one time point, usually at diagnosis. Several recent studies have suggested that serum leptin levels in association with expression of leptin and/or leptin receptor isoforms in breast tumors or with gene polymorphisms for leptin and its receptor may be more definitive in identifying the impact of serum leptin on prognosis with respect to disease recurrence and death [
51,
52]. Although as indicated leptin and leptin receptor expression have been identified in human breast tumors, proteins associated with leptin signaling have not been characterized. In the present study, we not only found that leptin receptors ObR and ObRb are present in MTs obtained from MMTV-TGF-α mice, but we detected the presence of Jak2 and pSTAT3 proteins, which are located in the downstream pathway of leptin signaling. These proteins were also shown to be expressed in MFP. Additionally, the presence of leptin itself was detected in both MFP and MT samples, indicating the potential for paracrine and/or autocrine action of leptin. Furthermore, we found that consumption of a high-fat diet was associated with lower protein expression levels of ObRb and Jak2 in MFP and with lower ObRb, Jak2, and pSTAT3 protein expression levels in MT samples.
Our findings of a reduction in ObRb and/or Jak2 and pSTAT3 protein expression levels are similar to previous reports for effects of diet-induced obesity and consumption of high-fat diets on these proteins in the hypothalamus. For example, in male (Fischer 344 × Brown Norway) rats with diet-induced obesity as a consequence of consumption of a high-fat diet, ObRb protein and mRNA expressions were reported to be reduced in the hypothalamus compared with chow controls [
53]. In addition, 30 days of calorie restriction after the 105 days of consuming the high-fat diet resulted in increased ObRb and pSTAT3 protein expression in diet-induced obese rats. In other studies, the reduction of leptin receptor expression levels in the hypothalamus of high-fat-fed Osborne-Mendel rats [
54] and C57BL6 mice [
55] in comparison with low-fat counterparts also parallels our findings in MFP and MT tissues. In adipose tissue from male C57BL6 mice fed a high-fat diet for 8 weeks, higher serum leptin levels were associated with reduced expression levels of the short form of the leptin receptor, although expression levels of the long form of the leptin receptor were similar to those of low-fat control mice [
56]. This reduction in the leptin receptor and/or components of the signaling pathway has been termed leptin resistance and is associated with genetic and diet-induced obesities as well as with aging [
57‐
60].
In the present study, we also assessed effects of low-fat versus high-fat diets on expression levels of apoptosis-related proteins in MTs and MFP. All of the proteins, total PARP, PARP-89, PARP-24, Bax, Bcl-2, Bcl-xL, and caspase-3 activity, were detected in both tissues. Furthermore, we found that diet affected most of these proteins more than body weight status, with the consumption of a high-fat diet being associated with reduced apoptosis in both MTs and MFP. Leptin has been reported to inhibit apoptosis in human colon [
37] and prostate [
38] cancer cell lines and in leukemic cells [
39]. Since mice fed the high-fat diet had higher serum leptin levels compared with those fed the low-fat diet, the lower expression of these apoptotic proteins is consistent with
in vitro results. We also found that the consumption of a high-fat diet resulted in significantly decreased cleaved PARP protein expression levels in MT and MFP samples. In addition, the activity of caspase-3, which cleaves total PARP, was lower in MFP and MT tissues of the mice fed the high-fat diet. Both caspase-3 and cleaved PARP products are involved in DNA breakage and lead to apoptosis. Although in the present study cell survival protein expression levels either did not change (pAkt and Bcl-2) or were decreased (Bcl-XL) in high-fat diet mice that also had higher serum leptin levels, in an
in vitro study with HTB-26 and ZR75-1 breast cancer cell lines, it was shown that leptin increased the expression levels of pERK (phosphorylated extracellular signal-regulated kinase) and pAkt [
41]. This difference may well be a consequence of the different study designs (
in vitro versus
in vivo). In particular, in the present study, tissues were removed after long-term diet maintenance and exposed to many other hormones and cytokines in addition to leptin in the body, whereas in the other study, cells were maintained with leptin for only 48 hours after serum deprivation for 20 hours
in vitro.
A summary of the results of protein and mRNA expression levels is presented in Table
3. The effects of high-fat diet relative to low-fat diet were similar on the expression levels of leptin, ObR, ObRb, Jak2, total PARP, caspase-3, Bcl-2, Bcl-Xl, and pAkt protein both in MT and MFP tissues. Overall changes that did occur were primarily in the direction of reduced apoptosis in the MFP and MT tissues obtained from mice fed the high-fat diet. The present study is the first to show that consumption of a high-fat diet (for 75 weeks) has effects on pro-apoptotic and survival protein expression levels in MTs and MFP. However, several previous publications have studied the effects of calorie restriction on similar protein expression levels. For example, levels of pro-apoptotic Bax, caspase-3 activity, and cleaved PARP products were all higher in MTs from energy-restricted rats (40% reduction) compared with MTs from
ad libitum-fed rats, whereas Bcl-2 and pAkt were reduced [
61]. In another study, using MTs obtained from energy-restricted MMTV-TGF-α mice, caspase-3 activity and DNA breakage were elevated in mice subjected to 50% restriction but not to 25%, but pAkt was not affected [
62]. In the present study, consumption of a high-fat diet resulted in little change on cell survival proteins (Bcl-2, Bcl-xL, and pAkt) in MT as well as MFP tissues, with the primary effect of consumption of the high-fat diet resulting in reduced apoptosis. Thus, calorie restriction appears to promote apoptosis in MTs whereas consumption of the moderately high-fat diet, even without effects on caloric intake, did not.
Table 3
Summary of diet effects on leptin and apoptosis signaling protein and mRNA expression levels
Leptin signaling molecules | | | | |
Leptin | NC | | NC | NC |
ObR | NC | | NC | |
ObRb | | | | |
Jak2 | | NM | | NM |
pSTAT3 | | | NC | NC |
Apoptosis signaling molecules | | | | |
PARP (total) | | | | |
PARP (89 kDa) | | | NC | |
PARP (24 kDa) | | | NC | |
Caspase-3 activity | | | | |
Bax | NC | | | |
Bcl-2 | NC | | NC | |
Bcl-XL | | | | |
pAkt | NC | | NC | |
Although it is documented that consumption of high-fat diets shortens MT latency, the molecular mechanism mediating this effect is unknown. In the present study, we report that mice that were fed a high-fat diet and that became obese had higher MT incidence and serum leptin levels, which were correlated with total fat pad and body weights. We also demonstrated that expression levels of leptin and apoptotic-associated proteins were modulated by long-term high-fat diet consumption. Leptin resistance appears to be associated with high-fat diet consumption in both MTs and MFP, and an anti-apoptotic protein profile characterizes the MTs from mice fed the high-fat diet. At this point, it is not clear what would happen at time points prior to the development of leptin resistance and how this would relate to MT development.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
MPC is the principal investigator of the present study and contributed to the analysis and interpretation of the data and drafting of the manuscript. SD contributed to the analysis and interpretation of the data and drafting of the manuscript. XH contributed to animal dissection and tissue processing. YZ performed statistical analysis. NJM provided assistance in data interpretation and manuscript preparation. JPG read the pathological slides from the tissues taken from the animals to decide whether the tissues were tumors or not, in addition to classifying tumor grades. All authors contributed to revisions of the manuscript and approved the submitted manuscript.