A Western-type diet accelerates tumor progression in an autochthonous mouse model of prostate cancer

Abstract

Epidemiological studies have provided evidence suggesting an important role for diet and obesity in the development of cancer. Specifically, lipid nutrients of the diet have been identified as important regulators of tumor development and progression. In the present study, we have examined the role of dietary fat and cholesterol in the initiation and progression of prostate cancer using the well-characterized TRAMP mouse model. Consumption of a Western-type diet--that is, enriched in both fat and cholesterol--accelerated prostate tumor incidence and tumor burden compared to mice fed a control chow diet. Furthermore, we also show that this diet increased the extent and the histological grade of prostate tumors. These findings were confirmed by the presence of increased levels of protein markers of advanced tumors in prostates obtained from animals fed a Western-type diet compared to those obtained from control animals. Increased lung metastases in animals fed a Western-type diet were also observed. In addition, we found that with a Western diet, animals bearing tumors presented with reduced plasma cholesterol levels compared with animals fed a control diet. Finally, we show that tumors obtained from animals fed a Western-type diet displayed increased expression of the high-density lipoprotein receptor SR-BI and increased angiogenesis. Taken together, our data suggest that dietary fat and cholesterol play an important role in the development of prostate cancer.

Figure 1

Increased tumor incidence and burden in TRAMP mice fed a Western-type diet. A: Twenty-eight-week-old TRAMP male mice fed either a chow or a Western diet were sacrificed and examined for the presence of gross and bulky prostate tumors. Results are expressed as percentage of mice that developed gross prostate tumors that made it impossible for the individual dissection of the different prostatic lobes (n = 17 and 18 for chow and Western group, respectively). B: When present, prostate tumors were excised and weighed. Results are represented as the average tumor weight in g of tumor per animal ± SE (n = 3 and 6 for chow and Western diet group, respectively). C: Representative images of GU tracts obtained from 28-week-old TRAMP mice fed a chow or a Western diet, demonstrating the presence of prostatic tumoral masses. D: Prostate tumors were fixed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Representative histological images are shown for TRAMP males fed a chow (n = 3) or a Western-type diet (n = 6). Scale bars = 200 μm.

Figure 2

Increased hyperplasia of the GU apparatus obtained from TRAMP mice fed a Western-type diet. A: GU tracts from nontransgenic and TRAMP mice when a gross tumor was not present at 28 weeks of age were carefully dissected en bloc. Representative pictures of GU apparatus obtained from nontransgenic and TRAMP males fed a chow or Western diet are shown. B: After dissection, GU tracts were weighed. This weight was considered to be an indirect measure of the tumor burden in TRAMP mice. Data are expressed in g of GU tract per animal ± SE (n = 14 and 12 for chow and Western diet group, respectively). ***P < 0.001 nontransgenic versus transgenic littermates; *P < 0.001 TRAMP males fed a Western diet versus TRAMP males fed a chow diet.

Figure 3

TRAMP mice fed a Western-type diet show advanced ventral prostate carcinogenic lesions. Ventral prostate lobes of nontransgenic and TRAMP male mice fed a chow or a Western diet until 28 weeks of age were microdissected, fixed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Representative histological images are shown of ventral prostates corresponding to nontransgenic males (A), TRAMP males fed a chow diet (B), and TRAMP males fed a Western diet (C). Images were taken at an original magnification of ×40. Each section was graded as normal (N), prostatic intraepithelial neoplasia (PIN), well-differentiated adenocarcinoma (WD), moderately differentiated adenocarcinoma (MD), and poorly differentiated adenocarcinoma (PD) using a scale that had previously been established for TRAMP mice. The percentage of mice in a given histopathological stage is represented (D). Eight TRAMP males fed a chow diet and 10 TRAMP males fed a Western diet were examined. Scale bars = 50 μm.

Figure 4

TRAMP mice fed a Western-type diet show more advanced anterior prostate carcinogenic lesions. Anterior prostate lobes of nontransgenic and TRAMP male mice fed a chow or a Western diet until 28 weeks of age were microdissected, fixed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Representative histological images are shown of anterior prostates corresponding to nontransgenic littermates (A), TRAMP males fed a chow diet (B), and TRAMP males fed a Western diet (C). Images were taken at an original magnification of ×40. Each section was graded as normal (N), prostatic intraepithelial neoplasia (PIN), well-differentiated adenocarcinoma (WD), moderately differentiated adenocarcinoma (MD), and poorly differentiated adenocarcinoma (PD) using a scale that had been previously established for TRAMP mice. The percentage of mice in a given histopathological stage is represented (D). Nine TRAMP males fed a chow diet and eight TRAMP males fed a Western diet were examined. Scale bars = 50 μm.

Figure 5

Increased cellular proliferation in the ventral prostate of TRAMP mice fed a Western-type diet. Ventral prostate lobes from 28-week-old TRAMP male mice fed a chow or a Western diet were microdissected, fixed, embedded in paraffin, sectioned at 5 μm, and subjected to immunohistochemical analysis for cyclin D1 (A) and PCNA (B). Representative images are shown for each dietary condition. Four TRAMP males fed a chow diet and four TRAMP males fed a Western diet were examined. Original magnification, ×40. Scale bars = 50 μm.

Figure 6

Increased lung metastasis in TRAMP mice fed a Western-type diet. A: After removal of the GU tract, 2 ml of 10% buffered formalin were injected through the trachea into the lungs of 28-week-old TRAMP male mice until the lungs were completely inflated and filled with formalin. Lungs were then excised and placed into formalin for 24 hours. The left lung of each animal was paraffin-embedded, sectioned every 50 μm, and stained with hematoxylin and eosin. Lung sections were then examined under a microscope using a ×10 objective for the presence of metastatic foci (defined as a cluster of a minimum of 10 cells). Results are expressed as percentage of mice that developed at least one lung metastatic foci (n = 7 and 9 for chow and Western group, respectively). B: The number of metastatic foci was scored for each mouse. Results are represented as the average number of metastatic foci per lung ± SE (n = 7 and 9 for chow and Western diet group, respectively). *P < 0.05 Western diet-fed versus chow diet-fed animals.

Figure 7

TRAMP mice fed a Western-type diet showed decreased plasma cholesterol levels and body and epididymal fat weights after tumor development. A: Total plasma cholesterol content was determined from fasting plasma samples obtained from TRAMP male mice and their non-transgenic littermates using a colorimetric assay. Plasma from 8, 22, and 28 weeks of age mice were analyzed. Total serum cholesterol is expressed in mg of cholesterol/dL ± SE (n = 19 and 17 for nontransgenic and TRAMP mice fed a chow diet, respectively, and n = 17 and 18 for nontransgenic and TRAMP mice fed a Western diet, respectively). ***P < 0.001 Western diet-fed versus chow diet-fed animals; *P < 0.05 TRAMP versus nontransgenic males fed a Western diet. B and C: Fasting plasma samples isolated from 10 mice on each experimental condition at 28 weeks of age were pooled and loaded atop two Superose 6 columns. Fractions were collected and then analyzed for their cholesterol content using a colorimetric assay. B: Lipoprotein profiles obtained from 28-week-old nontransgenic and TRAMP male mice fed a chow diet. C: Lipoprotein profiles obtained from 28-week-old nontransgenic and TRAMP male mice fed a Western diet. D: At the time of sacrifice, total body weight was determined for each mouse. Data are expressed in g of body weight ± SE (n = 19 and 17 for nontransgenic and TRAMP mice fed a chow diet, respectively, and n = 17 and 18 for nontransgenic and TRAMP mice fed a Western diet, respectively). ***P < 0.001 Western diet-fed versus chow diet-fed animals; *P < 0.05 TRAMP versus nontransgenic males fed a Western diet. E: At the time of sacrifice, epididymal fat was dissected and weighed for each mouse. Data are expressed in g of fat ± SE (n = 19 and 17 for nontransgenic and TRAMP mice fed a chow diet, respectively, and n = 17 and 18 for nontransgenic and TRAMP mice fed a Western diet, respectively). ***P < 0.001 Western diet-fed versus chow diet-fed animals; ^#P < 0.001 versus TRAMP versus nontransgenic males fed a Western diet.

Figure 8

TRAMP mice fed a Western-type diet showed increased expression of SR-BI in the ventral prostate. Ventral prostate lobes from 28-week-old TRAMP male mice fed a chow or a Western diet were microdissected, formalin-fixed, paraffin-embedded, sectioned at 5 μm, and subjected to immunohistochemical analysis for SR-BI. Representative images are shown for each dietary condition. Four TRAMP males fed a chow diet and four TRAMP males fed a Western diet were examined. Original magnification, ×40. Scale bars = 50 μm.

Figure 9

Increased plasma cholesterol is associated with increased tumor angiogenesis in TRAMP mice. Ventral prostate lobes from 28-week-old TRAMP male mice fed a chow or a Western diet were microdissected, formalin-fixed, paraffin-embedded, sectioned at 5 μm, and subjected to immunohistochemical analysis for CD31. A: Representative images are shown for each dietary condition. Three TRAMP males fed a chow diet and three TRAMP males fed a Western diet were examined. Original magnification, ×40. Scale bars = 50 μm. Arrows indicate the presence of CD31-positive blood vessels. B: Microvessel quantification was performed by counting the number of CD31-positive vessels in three representative fields for each section. Results are expressed as average ± SE. Significant difference *P < 0.05.