Abstract
Pancreatic cancer is the fourth leading cause of cancer deaths in the United States. Perfluorooctanoic acid (PFOA), a persistent environmental pollutant, has been shown to induce pancreatic acinar cell tumors in rats. Human epidemiologic studies have linked PFOA exposure to adverse chronic health effects including several types of cancer. Previously, we demonstrated that PFOA induces oxidative stress and focal ductal hyperplasia in the mouse pancreas. Here, we evaluated whether PFOA promotes pancreatic cancer using the LSL-KRasG12D;Pdx-1 Cre (KC) mouse model of pancreatic cancer. KC mice were exposed to 5 ppm PFOA in drinking water starting at 8 weeks of age and analyzed at 6 and 9 months of age. At the 6-month time point, PFOA exposure increased pancreatic intraepithelial neoplasia (PanIN) area by 58%, accompanied by a 2-fold increase in lesion number. Although PanIN area increased at 9 months, relative to 6 months, no treatment effect was observed. Collagen deposition was enhanced by PFOA at both the 6- and 9-month time points. PFOA also induced oxidative stress in the pancreas evidenced by elevated antioxidant activity of superoxide dismutase (Sod), catalase and thioredoxin reductase, and a ~3-fold increase in Sod1 mRNA and protein levels at 6 months. Although antioxidant activity was not enhanced by PFOA exposure at the 9-month time point, increased pancreatic oxidative damage was observed. Collectively, these results show that PFOA elicited temporal increases in PanIN lesion area and desmoplasia concomitant with the induction of oxidative stress, demonstrating that it functions to promote pancreatic cancer progression.
We demonstrate that PFOA exposure elicits temporal increases in PanIN lesion area and desmoplasia concomitant with the induction of oxidative stress in the LSL-KRasG12D;Pdx-1Cre mouse model of pancreatic cancer, demonstrating that PFOA exposure acts at the promotion stage of carcinogenesis.
Graphical Abstract
Graphical Abstract.
Introduction
Pancreatic cancer is the fourth leading cause of cancer deaths, and one of the most lethal forms of cancer diagnosed in the United States. Despite advances in the identification of genetic alterations that occur during pancreatic cancer progression, overall prognosis has not improved, with median survival time following diagnosis < 6 months and the 5-year survival rate less than 10% (1). Several risk factors have been identified as potential contributors to pancreatic cancer development and progression, including environmental and lifestyle factors, such as smoking, drinking and diet, medical conditions such as diabetes and pancreatitis, as well as exposure to environmental chemicals (2,3).
One environmental chemical implicated in the pathogenesis of chemically induced pancreatic cancer is perfluorooctanoic acid (PFOA). PFOA is a member of the perfluoroalkyl substances (PFAS) chemical group, consisting of >4700 environmentally persistent chemicals that exhibit widespread ground and surface water contamination due to extensive use since the 1940s (4). PFOA is widely used in the manufacture of an array of industrial and commercial products, such as Teflon and Scotchguard. Humans are exposed to PFOA by drinking water, dust in homes, food products or migration from food packaging and cookware (5). As PFOA does not readily decompose in the environment, and due to continued exposure, detectable levels of PFOA have been identified in 98% of the American population, with mean serum levels measured at 3.9 ng/ml (6), and historically has reached levels up to 22 000 ng/ml in those occupationally exposed (7). PFOA is readily absorbed, but poorly eliminated, exhibiting a predicted half-life of 3.8 yrs in humans (8). In rats, chronic exposure to PFOA induces liver, Leydig and pancreatic acinar cell tumors (9,10). In humans, epidemiologic studies have linked PFOA exposure to adverse chronic health effects including several types of cancer (11–13).
Greater than 85% of human pancreatic cancers are classified as pancreatic ductal adenocarcinoma (PDAC). PDAC progresses through stages characterized by morphological alterations and nuclear atypia (14). The most common precursor lesion for PDAC is pancreatic intraepithelial neoplasia (PanIN), which progresses in severity from stage 1 to stage 3 and coincides with or is preceded by acinar–ductal metaplasia (ADM). ADM is a phenotypic switch whereby acinar cells begin to express ductal markers, such as cytokeratin 19 and Sox9 (15,16). One of the earliest molecular changes detected in pancreatic cancer is mutation of the KRAS gene, which is found in >90% of human PDAC (14). Targeting constitutively active KRas (KRasG12D) to early pancreatic progenitors using the Pdx-1 or Ptf1a gene promoters in mice, results in a histologic progression of pancreatic cancer that closely mimics that observed in human disease (17). KRasG12D mice develop small areas of ADM and low-grade PanINs by 2 months of age. In this model, the number and severity of PanIN lesions increases at 6 months concomitant with an increase in ADM, and after 8–12 months, PDAC develops at low frequency (17).
Inflammatory processes and oxidative stress have been shown to participate in pancreatic cancer progression. When mice were subjected to chronic inflammation in the pancreas in the presence of oncogenic KRasG12V, the incidence of PDAC increased to 100% by 8 months of age (18). Furthermore, both inflammation and oxidative stress are included as enabling characteristics and hallmarks of cancer (19,20). Additionally, the ability of a chemical to induce oxidative stress has been identified as one of the key characteristics of carcinogens used in cancer hazard identification (21-23). In vitro, PFOA, as well as other PFAS, induced oxidative stress in the HepG2 cell line (24-26) and an inflammatory response in human mast cells, evidenced by the induction of pro-inflammatory cytokines and increased production of reactive oxygen species (ROS) (27). In humans, PFAS exposure was associated with activation of oxidative stress, evidenced by modulation of glutathione, α-tocopherol and ascorbate pathways (28). We previously demonstrated that PFOA accumulates, and induces oxidative stress in the mouse pancreas following a 7-day exposure (29), and that oxidative stress can be triggered by endoplasmic reticulum stress in pancreatic acinar cells (30).
The process of carcinogenesis is broadly divided into three stages: initiation, where DNA mutations become fixed into the genome creating initiated cell populations; promotion, where exogenous agents promote the expansion of the initiated cell populations; and progression where the neoplasia progresses to malignancy (31). Here we extend the initial findings of rat 2-year chronic bioassays to specifically address whether exposure to PFOA can act at the promotion stage of carcinogenesis in the LSL-KRasG12D;Pdx-1Cre mouse model of pancreatic cancer. We demonstrate that exposure to PFOA elicited temporal increases in PanIN area and desmoplasia concomitant with the induction of oxidative stress.
Materials and methods
Conditional KRasG12D model and treatment design
LSL-KRasG12D and Pdx-1 Cre mice were obtained from the Jackson Laboratory and maintained as heterozygous lines. Offspring of LSL-KRasG12D and Pdx-1 Cre mice were genotyped by PCR analysis to confirm the presence of both alleles (genotyping primers listed in Supplementary Table I). Male and female LSL-KRasG12D;Pdx-1 Cre (KC) mice were either untreated (control), receiving tap water, or treated with 5 ppm PFOA (Sigma 171468, 96% purity) in drinking water beginning at adulthood (8 weeks). Mice were euthanized by CO2 asphyxiation at 6 and 9 months of age (4 and 7 months of PFOA exposure), at which time serum and tissues were collected. Pancreata were divided in half along the longitudinal axis, from the head to tail of the pancreas. The top section was used for histology for all pancreata. Pancreas sections were formalin fixed for 48 h and then embedded in paraffin, sectioned and stained with H&E (AML Laboratories, St. Augustine, FL). The remaining pancreata were snap frozen in liquid nitrogen for further biochemical and gene expression analysis. The experiments were approved by the Indiana University Bloomington Institutional Animal Care and Use Committee, and mice were maintained in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
Histopathology evaluation
PanIN grade, inflammation score and stroma evaluation were performed by pathologists blinded to the experimental groups on H&E-stained slides using light microscopy to evaluate each section. Murine PanIN lesions were classified by previously identified criteria (32) and assigned a score based on the highest PanIN lesion present (PanIN 1–4; scores 1–4). Inflammation was scored in the tumor microenvironment area only and assigned a score of 0–4 where 0 = none, 1 = mild, 2 = moderate, and 3 = severe inflammation based on the number of infiltrating leukocytes (33). Stromal density was assessed and assigned a score of 1–3 where 1 = minimal (loose), 2 = moderate and 3 = strong (dense) stroma (34,35). A composite histopathology severity score was derived incorporating PanIN lesion stage, inflammation and stromal density = [PanIN stage (1–4) × inflammation score (1–3) × stromal density (1–3)]. Individual scores for each mouse are provided in Supplementary Table II.
Quantitation of PanIN lesions and collagen deposition
Alcian Blue (AB) staining of pancreatic lesions on paraffin sections was performed according to manufacturer’s instructions [Alcian Blue (pH 2.5) staining kit, Vector Labs, Burlingame, CA]. Stained sections were dehydrated before mounting with Histomount (Thermo Fisher Scientific), then scanned using a Motic EasyScan whole-slide imaging system at 20× magnification (Motic, British Columbia). AB lesion areas of digitized images were then assessed using Aperio ImageScope software (Leica Biosystems, Buffalo Grove, IL) by quantifying the area of AB-positive lesions relative to the total area of the pancreatic section. The number and size of PanIN lesions was also quantified using Aperio ImageScope software. Collagen deposition in pancreatic sections was assessed using a Masson’s Trichrome Stain kit (Sigma–Aldrich) per manufacturer’s instructions. Stained sections were dehydrated before mounting with Histomount and digitized using a Motic EasyScan whole-slide imaging system at 20× magnification (Motic, British Columbia). Images were converted to TIFF files, and the collagen content relative to the total area of each section was quantified using ImageJ Software (NIH, Bethesda, MD).
PFOA quantitation
Acetonitrile (240 µl) was added to 10 µl of serum or pancreatic tissue homogenates prepared in RIPA buffer (50 mM Tris pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate) plus protease inhibitors (Pierce protease inhibitor mini tablets, EDTA free). Samples were vortexed for 15 s, sonicated for 5 min and centrifuged at 16 000g for 5 min. A 200 µl volume of supernatant was then dried under N2 gas and reconstituted with mobile phase consisting of LC–MS grade water containing 3.5 mM ammonium bicarbonate and methanol (75:25%). PFOA was quantified using an Agilent 1260 UPLC system coupled with an Agilent 6430 triple quadrupole mass spectrometer. The analytes were separated on an Agilent Zorbax Eclipse XBD-C18 column at 38°C. Analytes were eluted using isocratic conditions in mobile phase at a flow rate of 0.4 ml/min, with a 20 µl injection volume. PFOA was detected by mass spectrometry with an electrospray ion source operating in the negative-ion mode (ESI−) using MRM. Precursor ion (m/z 413) and product ions, m/z 368.9 (quantifier) and m/z 169 (qualifier) were monitored at a fragmentation voltage of 69, cell acceleration voltage of 7 and collision energy of 4 and 12 for each product ion. PFOA levels were quantified from a standard curve prepared at final concentrations between 0 and 0.4 ng/µl using Agilent Mass Hunter (v B.04.00) software.
Enzyme activity measurements and lipid peroxidation
Amylase and lipase activities in serum were measured spectrophotometrically, using kits from Pointe Scientific Inc. (Canton, MI) according to the manufacturer’s instructions. The enzymatic activities of antioxidants were determined using commercially available kits for SOD (Sigma–Aldrich), catalase (Cat), glutathione peroxidase (GPx) and thioredoxin reductase (TrxR) (Cayman Chemical) according to the manufacturer’s instructions. Enzyme activity was quantified in pancreatic tissue homogenates prepared in Tris buffer (100 mM Tris, pH 7.5, 1 mM EDTA) plus protease inhibitors. Results were expressed as activity U/min/µg protein for SOD, nmol/min/µg protein for CAT, and µmol/min/mg protein for TrxR. Malondialdehyde levels were assessed in pancreatic tissue homogenates prepared in RIPA buffer according to manufacturer’s instructions (Cayman Chemical). Results were expressed as µM malondialdehyde (MDA)/mg protein.
Quantitative RT–PCR analysis
Total RNA was isolated from pancreatic tissues (Trizol, Invitrogen) following a modification of manufacturer’s instructions, including a 20-fold Trizol volume to tissue weight ratio (36). Twenty micrograms of total RNA was reverse transcribed with Superscript II reverse transcriptase (Invitrogen) using random hexamers (Promega) for priming. Real-time PCR was performed using KAPA Biosystems 2X SYBR Green Master Mix and gene-specific primers on an Applied Biosystems QuantStudio3 Real-time PCR System. Primer pairs for specific genes were designed using the Primer Express program (Applied Biosystems), with β-actin amplification used as the endogenous control. Samples were measured in triplicate and analyzed by the threshold cycle (Ct) comparative method. The 2−ΔΔCt value was calculated, where ΔCt = Cttarget−Ctβ-actin, and ΔΔCt = ΔCtsample−ΔCtreference. Relative quantitation for each gene is shown, where control levels in the pancreas were set to 1.0. Sequences of primers used for quantitation are included in Supplementary Table I.
Western analysis
Pancreatic tissue homogenates were prepared in RIPA buffer containing protease inhibitors. Equal amounts of protein (50 µg) were resolved by SDS/PAGE, transferred to PVDF membrane (Immobilon; Millipore) and subjected to Western blot analysis. Immunoblots were visualized utilizing enhanced chemiluminescence (WesternBright ECL; Advansta) and autoradiography film. The primary antibodies used for Western blot analysis were as follows: SOD1 (37385) and SOD2 (13141) from Cell Signaling Technology (Danvers, MA); and β-actin (A5316) from Sigma–Aldrich. Secondary antibodies against rabbit and mouse species conjugated to horseradish peroxidase were obtained from Cell Signaling Technology. Autoradiography films were scanned (Hewlett-Packard ScanJet 6300C), and pixel intensity was quantified to determine relative protein levels using ImageJ software.
Statistical analysis
The data were analyzed by one-way ANOVA followed by a Dunnett’s two-tailed test for comparison against controls when the overall model indicated a statistically significant effect. For all studies, treatment groups were considered significantly different from control values when P < 0.05. Statistical analysis was provided by the Biostatistical Consulting Center, Indiana University School of Public Health-Bloomington.
Results
Characterization of long-term exposure to PFOA in the LSL-KRasG12D;Pdx-1Cre mouse model
The LSL-KRasG12D;Pdx-1Cre mouse model was used to determine whether PFOA exposure can promote the growth of pancreatic lesions. In this model, LSL-KRasG12D mice are crossed with mice harboring Cre recombinase expression driven by the Pdx-1 gene promoter, resulting in the expression of KRasG12D in early pancreatic progenitors. Although Cre recombinase is expressed in a stochastic manner, this model has been shown to faithfully recapitulate the full spectrum of histologic pathology progression observed in human pancreatic cancer development (17). As lesion severity progresses in a time-dependent manner, with PDAC observed only at late time points of 16–24 months (17), this model is ideal to study the effects of PFOA on the promotion stage of carcinogenesis. Male and female LSL-KRasG12D;Pdx-1Cre (KC) mice were provided control tap water or tap water containing 5 ppm PFOA, beginning at adulthood (8 weeks). Exposure to PFOA in drinking water was chosen, as it represents a significant route of human exposure. Mice were sacrificed at 6 and 9 months of age, (4 and 7 months of PFOA exposure), at which time serum and tissues were collected (Figure 1A). As expected, serum levels of PFOA were significantly elevated in mice exposed to PFOA, with control (CTL) KC mice of both age groups exhibiting PFOA levels of 0.003 µg/ml, whereas serum levels in PFOA-treated 6- and 9-month-old KC mice were 41.96 ± 16.45 and 26.35 ± 17.53 µg/ml PFOA, respectively (Figure 1B). PFOA levels in pancreatic tissue were also elevated in mice exposed to PFOA, with CTL KC mice at 6 and 9 months exhibiting levels of 0.039 ± 0.011 and 0.033 ± 0.009 ng/mg protein, respectively, whereas KC mice exposed to PFOA exhibited levels of 2.85 ± 1.65 ng/mg protein at 6 months and 3.05 ± 2.67 ng/mg protein at 9 months (Figure 1C). Although the serum PFOA levels at 9 months of age, corresponding to 7 months of PFOA exposure, were reduced relative to the 6-month time point, PFOA levels in the pancreas continued to accumulate. As shown in Table 1, PFOA exposure did not significantly alter body weights at either time point evaluated. Both absolute and relative pancreas weights increased in CTL KC mice at 9 months when compared with the 6-month time point. In PFOA-exposed 6-month KC mice, a greater variability in absolute and relative pancreas weight was observed, which precluded observation of statistical significance. At the 9-month time point, both absolute and relative pancreas weights were significantly decreased in PFOA-treated KC mice versus CTL KC mice (Table 1). Both absolute and relative liver weights were increased at both time points in KC mice exposed to PFOA. Serum amylase and lipase levels were measured as markers of pancreatic function. Although no significant change in serum amylase levels was observed at either time point between treated or untreated KC mice, lipase levels were elevated in PFOA-exposed mice at both time points, which approached statistical significance at the 9-month time point (Table 1).
Figure 1.
Long-term PFOA exposure in the KC mouse pancreatic cancer model. (A) Experimental design for PFOA administration in KC mice. Exposure to 5 ppm PFOA in drinking water was initiated in KC mice at 2 months of age and continued until time of sacrifice at 6 and 9 months of age, corresponding to 4 and 7 months of PFOA exposure, respectively. (B and C) Serum (B) and pancreatic tissue (C) levels of PFOA were determined at sacrifice in mice exposed to tap water (CTL KC) or 5 ppm PFOA (PFOA KC). Plotted are individual values; horizontal bars represent the mean ± SD (n = 9–11 mice per group). ∗∗P < 0.01, ∗∗∗P < 0.001.
Table 1.
Characteristics of Control (CTL) and PFOA-exposed KC mice at 6 and 9 months of age
Body weight (g) | Pancreas weight (g) | Relative pancreas weighta | Liver weight (g) | Relative liver weighta | Amylase (U/L) | Lipase (U/L) | ||
---|---|---|---|---|---|---|---|---|
6 Months | CTL | 25.8 ± 4.9 | 0.38 ± 0.11 | 1.50 ± 0.39 | 1.40 ± 0.28 | 5.43 ± 0.47 | 326.0 ± 113.5 | 85.7 ± 28.5 |
PFOA | 24.7 ± 2.3 | 0.60 ± 0.53 | 2.42 ± 2.01 | 2.06 ± 0.32∗∗∗ | 8.31 ± 0.85∗∗∗ | 269.0 ± 76.9 | 98.3 ± 27.8 | |
9 Months | CTL | 28.9 ± 6.4 | 0.66 ± 0.15‡ | 2.32 ± 0.42 | 1.55 ± 0.33 | 5.40 ± 0.57 | 281.3 ± 71.3 | 64.6 ± 20.8 |
PFOA | 25.1 ± 4.6 | 0.44 ± 0.14∗ | 1.71 ± 0.33∗ | 2.35 ± 0.52∗ | 9.50 ± 2.03∗∗∗ | 269.5 ± 40.3 | 80.3 ± 17.5 |
Significantly different from age-matched CTL KC: ∗P < 0.05; ∗∗∗P < 0.001. Significantly different from 6 Mo CTL KC mice: ‡P < 0.05.
Relative organ weight = (organ weight/body weight) × 100.
Exposure to PFOA leads to increased PanIN lesion area and number
Pathological evaluation of H&E-stained pancreatic sections was performed to determine PanIN grade, inflammation and stromal density (Figure 2A). The mean PanIN grade, scored as the most advanced lesion present in the section, was not significantly different between CTL KC or PFOA-treated KC mice at either time point (Figure 2B). PFOA exposure did not alter the inflammation grade at the 6-month time point; however, PFOA exposure significantly increased inflammation in 9-month PFOA-treated KC mice, compared with 9-month CTL KC mice (Figure 2B). Stromal density, evaluated as loose, moderate or dense based on previously established criteria (34,35), was significantly increased by PFOA treatment at 6 months. At 9 months of age, stromal density was significantly higher in both CTL and PFOA-exposed KC mice relative to the 6-month CTL KC group (Figure 2B). A composite histopathology severity score was derived for each animal that incorporated PanIN grade, inflammation and stromal density (Figure 2C), with scores for individual mice provided in Supplementary Table II. At 6 months, the mean histopathological severity score increased from 8.0 ± 2.6 (range 4–12) in CTL KC mice, to 19.4 ± 15.9 (range 8–64) in PFOA KC mice. At 9 months, the histopathological severity score increased from 19.9 ± 9.4 (range 12–36) in CTL KC mice to 25.0 ± 10.8 (range 12–48) in PFOA KC mice (Figure 2C).
Figure 2.
Histologic characterization of PanIN lesion development in KC mice following exposure to PFOA. (A) Representative H&E and Alcian Blue stained sections of the pancreata of 6-month-old wild-type mice (6 Mo Control), and 6- and 9-month-old KC mice treated with tap water (Control KC) or 5 ppm PFOA (PFOA KC) (scale bar, 300 µm). (B) H&E-stained sections of individual pancreata from 6- and 9-month-old CTL KC and PFOA KC mice were evaluated for PanIN lesion grade, peri-pancreatic inflammation and peri-pancreatic stromal density as described in Materials and methods. ∗∗∗Significantly different from 6 Mo CTL KC, P < 0.001; #Significantly different from 9 Mo CTL KC; P < 0.05. (C) A composite histopathology severity score was derived as described in Materials and methods. Shown are individual values of 6- and 9-month-old CTL KC and PFOA KC mice where horizontal bars represent the mean ± SD (n = 9–11 mice per group). ∗P < 0.05.
To evaluate whether PFOA promotes the growth of pancreatic lesions, lesion area was quantified following Alcian Blue (AB) staining of pancreatic sections (Figure 2A). At the 6-month time point, PFOA resulted in a 58% increase in AB area (4.87 ± 1.65% CTL KC versus 7.71 ± 3.01% PFOA KC, Figure 3A). Similarly, a 2-fold increase in the number of lesions number per mm2 of pancreatic tissue was observed in 6-month-old PFOA KC mice compared with CTL KC (8.73 ± 3.19 CTL KC versus 16.72 ± 6.32 PFOA KC, Figure 3B). Further analysis of individual PanIN lesions was performed to evaluate whether PFOA caused a change in the size distribution of PanIN lesions. At the 6-month time point, PFOA exposure significantly increased the number of lesions less than 10 000 µm in size compared with CTL KC, with the largest number of lesions observed in the 1000–2499 µm size range (Figure 3C). When lesion size distribution was evaluated in 6-month-old KC mice, PFOA exposure significantly increased the percentage of total lesions in the 1000–2499 µm size range, while the percentage of larger lesions was decreased as compared with CTL KC (Figure 3D). At the 9-month time point, PanIN lesion area was significantly increased relative to the 6-month time point, however, significant differences in the area occupied by PanIN lesions was not observed between CTL KC and PFOA KC mice (Figure 3A). Although the number of PanIN lesions was also increased in 9-month-old CTL KC and PFOA KC mice relative to 6-month-old KC mice, no significant treatment effect was seen (Figure 3B). In contrast to the 6-month time point, at 9 months, the number of small lesions was lower PFOA KC mice; however, the percentage of larger lesions > 10 000 µm was increased compared with CTL KC (Figure 3E and F).
Figure 3.
Effect of PFOA exposure on PanIN lesion area and size distribution in KC mice. (A–F) Individual pancreata sections were stained with Alcian Blue (AB) and digitized. Aperio ImageScope software was then used to calculate (A) area occupied by AB+ PanIN lesions relative to total area of the pancreas and (B) total PanIN lesion number/mm2 of pancreas. (C–F) PanIN lesion size was further stratified and represented as PanIN lesion number/mm2 (C, E) or represented as % of total PanIN lesions in the designated size range (D, F) of 6- and 9-month-old CTL KC and PFOA KC mice. Plotted is the mean ± SD (n = 9–11 mice per group). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.
PFOA exposure leads to enhanced desmoplasia
PDAC is characterized by a strong desmoplastic reaction, which is driven by activated pancreatic stellate cells, fibroblasts and leukocytes (37). To evaluate whether PFOA exposure led to changes in stromal composition, we determined the overall collagen content in the pancreas of 6- and 9-month KC mice using Masson’s Trichrome staining (Figure 4A). Quantitative analysis of the area occupied by collagen demonstrated that PFOA exposure increased collagen deposition by 63% in 6-month-old KC mice (9.64 ± 5.34% CTL KC versus 15.76 ± 6.28% PFOA KC), which was accompanied by increased mRNA expression of Col1a2 (Figure 4C). At the 9-month time point, collagen deposition was increased relative to the 6-month time point in both KC and PFOA KC mice, remaining significantly elevated in PFOA KC mice (14.27 ± 3.06% CTL KC versus 19.70 ± 6.83% PFOA KC) (Figure 4B).
Figure 4.
PFOA exposure increases desmoplasia in KC mice exposed to PFOA. (A, B) Individual pancreata sections were stained with Masson’s Trichrome (MT) to visualize collagen content, then were digitized. Shown are representative sections (A) of 6-month-old wild-type mice (6 Mo Control) and 6- and 9-month-old CTL KC and PFOA KC mice. Collagen content relative to the total area of each section (B) was quantified using ImageJ Software and represented as % area occupied by MT staining. Individual values of 6- and 9-month-old CTL KC and PFOA KC mice are plotted, where horizontal bars represent the mean ± SD (n = 9–11 mice per group). (C) qRT–PCR analysis was performed to determine mRNA expression of collagen 1A2 (Col1a2) in pancreata of CTL KC and PFOA KC mice. Shown is the average fold-induction ± SD of measurements performed in triplicate with 6-month-old CTL KC values set to 1.0 (n = 9–11 mice per group). ∗P < 0.05.
Exposure to PFOA leads to oxidative stress in the pancreas
Oxidative stress results from an imbalance between production of ROS and their enzymatic or non-enzymatic removal by antioxidant defense mechanisms. A common response to oxidative stress is the upregulation of antioxidant expression and enzymatic activity, which serve to neutralize excess ROS species. Superoxide dismutase (Sod) is an enzyme that catalyzes the dismutation of superoxide anion (O2·−) to hydrogen peroxide (H2O2), while Cat, peroxiredoxin (Prx) and GPx convert H2O2 to water. The thioredoxin (Trx) and TrxR system reduce Prx, and glutathione reductase (GSR) reduces GPx that is oxidized in the conversion of H2O2 to H2O (38). We have previously shown that short-term (7-day) exposure to PFOA resulted in oxidative stress evidenced by increased mRNA expression of Sod1, Sod2 and Cat in the pancreas (29). To determine whether long-term exposure to PFOA elicited a similar response in KC mice, we first examined activity levels of major antioxidant enzymes. As shown in Figure 5A, 5 ppm PFOA led to a significant increase in Sod enzyme activity, with a 33% and 25% increase in activity observed at the 6- and 9-month time points, respectively, compared with activity in CTL KC mice. PFOA also elicited significant increases in pancreatic Cat and TrxR activity at the 6-month time point compared with CTL KC mice (Figure 5B and C). Although both pancreatic Cat and TrxR activities were increased at the 9-month time point in both CTL KC and PFOA KC mice, PFOA exposure did not further enhance this effect. We further probed whether the increased Sod activity was due to selective upregulation of the intracellular isoforms of Sod; Sod1 and Sod2, at the protein and mRNA expression levels. Although Sod2 protein and mRNA levels were similar at the 6- and 9-month time points in both CTL and PFOA KC mice, PFOA exposure elicited an ~3-fold increase in Sod1 protein and mRNA levels at the 6-month time point (Figure 5D–G). Prolonged oxidative stress can cause oxidative damage to DNA, protein and lipid, which can contribute to disease pathology. As a direct marker of oxidative damage, we determined the level of MDA, a stable product of oxidized lipids, in the pancreas. As shown in Figure 5H, similar MDA levels were observed in CTL and PFOA KC mice at 6 months; however, PFOA exposure resulted in a 1.5-fold increase in MDA at the 9-month time point. Together these results demonstrate that PFOA exposure increased PanIN area and number in the KC mouse model of pancreatic cancer at the 6-month time point, indicating promotion of early lesion formation in this model. At the 9-month time point, increased desmoplasia and inflammation were observed, indicating that PFOA exposure also increased disease severity later in pancreatic cancer progression.
Figure 5.
PFOA exposure induces oxidative stress in KC mice. (A–C) Enzymatic activity of Sod (A), Cat (B) and TrxR (C) in pancreatic tissues of 6- and 9-month-old CTL KC and PFOA KC mice was determined as described in Materials and methods. Plotted are the means ± SD (n = 9–11 mice per group). (D, E) Pancreatic tissue lysates of 6- and 9-month-old CTL KC and PFOA KC mice were subjected to Western blot analysis (D) for expression of Sod1 and Sod2, where β-actin blot is used as a loading control. Blots were scanned and pixel intensity of Sod1 and Sod2 expressed relative to β-actin expression (E). Shown is the mean ± SD (n = 3 mice per group). (F, G) qRT–PCR analysis was performed to determine Sod1 and Sod2 mRNA expression in pancreata of CTL KC and PFOA KC mice. Shown is the average fold-induction ± SD of measurements performed in triplicate with 6-month-old CTL KC values set to 1.0 (n = 9–11 mice per group). (H) MDA levels were determined in pancreatic tissues of 6- and 9-month-old CTL KC and PFOA KC mice. Shown is the mean ± SD (n = 9–11 mice per group). ∗P < 0.05; ∗∗P < 0.01. (I) Model for promotion of pancreatic cancer by PFOA. PFAS exposure triggers activation of the UPR resulting in O2− production and upregulation of SOD1 through activation of PERK. SOD1 can stimulate cell proliferation by increasing production of H2O2, which acts as a second messenger in mitogen signaling or through its elimination of ROS, thereby preventing ROS-stimulated apoptosis.
Discussion
Carcinogenesis is a multistage process that is broadly divided into three stages: initiation, promotion and progression. During the initiation phase, DNA mutations become fixed in the genome creating an initiated cell that can then be propagated by cell division. Mutation of the KRAS gene, leading to its constitutive activation, has been identified as a key mutational event in pancreatic cancer initiation. KRAS mutations have been observed in early PanIN lesions and in >90% of PDAC cases (14). The promotion stage requires sustained exposure to an exogenous or endogenous agent that mediates changes in gene expression leading to the expansion of initiated cell populations into preneoplastic lesions. Continued exposure to these same chemical promoters, or exposure to additional agents drives the progression stage in which full malignancy is acquired. PFOA exposure resulted in an increased incidence of pancreatic acinar cell tumors in rats through an undefined mechanism (9,10). PFOA is not a direct mutagen thus not likely functioning as an initiator of carcinogenesis; however, PFOA exposure induced oxidative stress and inflammation in both animal and human studies, both of which are key characteristics possessed by carcinogenic agents that can contribute to carcinogenesis (23). In the KRasG12V mouse model of pancreatic cancer, high-grade PanIN lesions and PDAC were observed only after treatment with cerulein which initiates a mild form of pancreatitis (18,39). In humans, chronic pancreatitis leads to a 40-fold increase in risk for development of pancreatic cancer (40). In the present study, we used the LSL-KRasG12D;Pdx-1Cre mouse model of pancreatic cancer, as PanIN lesions develop slowly over time, making this model ideal to study the early acceleration or promotion of PanIN lesions driven by exogenous or endogenous agents. We show here that PFOA exposure resulted in a significant increase in both the number and area of PanIN lesions at 6 months (Figure 3), supporting that PFOA can drive early lesion expansion, which is a characteristic of agents that act at the promotion stage of carcinogenesis (31).
ROS, such as O2·− and H2O2, play important roles in normal physiology where they participate in redox reactions and act as second messengers in signal transduction (41,42). Elevated ROS can also induce oxidative stress leading to increased expression of antioxidant enzymes such as Sod1. Increased Sod1 has been postulated to enhance cancer cell growth by protecting cells from ROS-mediated apoptosis and by increasing H2O2 levels (43). Sod1 was shown to protect cancer cells from ROS-induced damage in oncogene-driven lung and mammary cancer (44,45) and protected cells from ROS and ischemia-induced oxidative DNA damage and cell death (46). Additionally, inhibition of Sod1 activity decreased non-small cell lung carcinoma cell line growth by inducing superoxide-mediated apoptosis (44). We have previously shown that a 1-week exposure to PFOA in mice induces oxidative stress and resulted in the induction of Sod1, Sod2 and Cat mRNA levels in the pancreas (29). In the present study, we show that Sod enzyme activity is induced at the 6- and 9-month time points in PFOA KC mice compared with CTL KC mice, while mRNA and protein levels of Sod1 are significantly increased at the 6-month time point in PFOA KC mice (Figure 5). We have shown that PFOA stimulates endoplasmic reticulum stress and activates the unfolded protein response (UPR), leading to Nrf2 activation and upregulation of Sod1 (30). Importantly, activation of the UPR can stimulate production of ROS, including O2·− (47,48) and Sod1 expression was shown to be required for cell survival during endoplasmic reticulum stress in yeast (47). Increased SOD1 activity was also seen in pancreatic cancer patients, compared with patients with chronic pancreatitis or control subjects (49). Taken together, these results support the following model for PFOA-mediated promotion of early PanIN expansion in pancreatic cancer (Figure 5I). PFOA exposure triggers activation of the UPR resulting in O2·− production and upregulation of Sod1 through activation of PERK (30). Sod1 then stimulates cell proliferation by increasing production of H2O2, which acts as a second messenger in mitogen signaling or through its elimination of ROS thereby preventing ROS-stimulated apoptosis.
A significant increase in MDA was only observed at the 9-month time point in PFOA KC mice (Figure 5H), suggesting that prolonged PFOA exposure results in sustained oxidative stress, leading to failure of compensatory antioxidant defense mechanisms and cellular damage. We also observed increased inflammation and desmoplasia at the 9-month time point, supporting that prolonged exposure to PFOA is increasing the severity of PanIN lesions (Figures 2 and 4). As the abundant desmoplastic reaction is considered one reason for the aggressive nature and chemoresistance of pancreatic cancer, these results also suggest that PFOA exposure may worsen chemotherapeutic efficacy (35,50).
Despite significant advances in the understanding of the biology of pancreatic cancer, overall prognosis has not significantly improved. Moreover, a clear understanding of how environmental exposures influence the promotion and progression of pancreatic cancer is lacking. Here, we demonstrate that PFOA exposure results in a temporal increase in PanIN development and severity in a mouse model of pancreatic cancer. Due to the widespread human exposure to PFOA and its biological persistence, these results suggest that PFOA may be a prominent agent that acts at the promotion stage of carcinogenesis. There is some epidemiologic evidence demonstrating an association between exposure to PFOA and increased risk for development of pancreatic cancer. A study in occupationally exposed workers demonstrated an increased hazard ratio (1.36) for pancreatic cancer in the higher exposure quartiles (12) and a positive trend was seen for pancreatic cancer and increased PFOA plasma levels in a general population study (13). PFOA production has been phased out in the United States; however, significant environmental contamination still exists. In addition, other PFAS, with similar biological persistence, have been introduced to replace PFOA and are now emerging as significant environmental contaminants. Our results, coupled with the available epidemiologic studies, indicate that future research is needed to address data gaps related to the involvement of the >4700 member PFAS family in pancreatic cancer.
Supplementary Material
Glossary
Abbreviations
- ADM
acinar–ductal metaplasia
- CTL
control
- MDA
malondialdehyde
- PanIN
pancreatic intraepithelial neoplasia
- PDAC
pancreatic ductal adenocarcinoma
- PFAS
perfluoroalkyl substances
- PFOA
perfluorooctanoic acid
- ROS
reactive oxygen species
- UPR
unfolded protein response
Contributor Information
Lisa M Kamendulis, Department of Environmental and Occupational Health, Indiana University School of Public Health, Bloomington, IN 47405, USA.
Jessica M Hocevar, Department of Environmental and Occupational Health, Indiana University School of Public Health, Bloomington, IN 47405, USA.
Mikayla Stephens, Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, IN 46202, USA.
George E Sandusky, Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, IN 46202, USA.
Barbara A Hocevar, Department of Environmental and Occupational Health, Indiana University School of Public Health, Bloomington, IN 47405, USA.
Funding
National Institutes of Health (ES026370 to B.A.H.)
Conflict of Interest Statement
None declared.
Data availability
The data underlying this article are available in the article and in its online supplementary material.
References
- 1. Siegel, R.L., et al. (2021) Cancer Statistics, 2021. CA Cancer J Clin, 71, 7–33. [DOI] [PubMed] [Google Scholar]
- 2. Yadav, D., et al. (2013) The epidemiology of pancreatitis and pancreatic cancer. Gastroenterology, 144, 1252–1261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Barone, E., et al. (2016) Environmental risk factors for pancreatic cancer: an update. Arch. Toxicol., 90, 2617–2642. [DOI] [PubMed] [Google Scholar]
- 4. Pan, Y., et al. (2018) Worldwide distribution of novel perfluoroether carboxylic and sulfonic acids in surface waters. Environ. Sci. Technol., 52, 7621–7629. [DOI] [PubMed] [Google Scholar]
- 5. Schecter, A., et al. (2010) Perfluorinated compounds, polychlorinated biphenyls, and organochlorine pesticide contamination in composite food samples from Dallas, Texas, USA. Environ. Health Perspect.., 118, 796–802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Calafat, A.M., et al. (2007) Polyfluoroalkyl chemicals in the U.S. population: data from the National Health and Nutrition Examination Survey (NHANES) 2003-2004 and comparisons with NHANES 1999-2000. Environ. Health Perspect., 115, 1596–1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Steenland, K., et al. (2010) Epidemiologic evidence on the health effects of perfluoroctanoic acid (PFOA). Environ. Health Perspect., 118, 1100–1108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lau, C., et al. (2007) Perfluoroalkyl acids: a review of monitoring and toxicological findings. Toxicol. Sci., 99, 366–394. [DOI] [PubMed] [Google Scholar]
- 9. Biegel, L.B., et al. (2001) Mechanisms of extrahepatic tumor induction by peroxisome proliferators in male CD rats. Toxicol. Sci., 60, 44–55. [DOI] [PubMed] [Google Scholar]
- 10. Caverly Rae, J.M., et al. (2015) Evaluation of chronic toxicity and carcinogenicity of ammonium 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propanoate in Sprague-Dawley rats. Toxicol. Rep., 2, 939–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Steenland, K., et al. (2021) PFAS and cancer, a scoping review of the epidemiologic evidence. Environ. Res., 194, 110690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Raleigh, K.K., et al. (2014) Mortality and cancer incidence in ammonium perfluorooctanoate production workers. Occup. Environ. Med., 71, 500–506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Eriksen, K.T., et al. (2009) Perfluooctanoate and perfluooctanesulfanate plasma levels and risk of cancer in the general Danish population. J. Natl. Cancer Inst.., 101, 605–609. [DOI] [PubMed] [Google Scholar]
- 14. Hezel, A.F., et al. (2006) Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev., 20, 1218–1249. [DOI] [PubMed] [Google Scholar]
- 15. Kopp, J.L., et al. (2012) Identification of Sox9-dependent acinar-to-ductal reprogramming as the principal mechanism for initiation of pancreatic ductal adenocarcinoma. Cancer Cell, 22, 737–750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jain, R., et al. (2010) The use of cytokeratin 19 (CK19) immunohistochemistry in lesions of the pancreas, gastrointestinal tract and liver. Appl. Immunohistochem. Mol. Morphol., 18, 9–15. [DOI] [PubMed] [Google Scholar]
- 17. Hingorani, S.R., et al. (2003) Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell, 4, 437–450. [DOI] [PubMed] [Google Scholar]
- 18. Guerra, C., et al. (2007) Chronic pancreatitis is essential for induction of pancreatic ductal adenocarcinoma by K-Ras oncogenes in adult mice. Cancer Cell, 11, 291–302. [DOI] [PubMed] [Google Scholar]
- 19. Hanahan, D., et al. (2011) Hallmarks of cancer: the next generation. Cell, 144, 646–674. [DOI] [PubMed] [Google Scholar]
- 20. Luo, J., et al. (2009) Principles of cancer therapy: oncogene and non-oncogene addiction. Cell, 136, 823–837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Guyton, K.Z., et al. (2018) Application of the key characteristics of carcinogens in cancer hazard identification. Carcinogenesis, 39, 614–622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Smith, M.T., et al. (2016) Key characteristics of carcinogens as a basis for organizing data on mechanisms of carcinogenesis. Environ. Health Perspect., 124, 713–721. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Temkin, A.M., et al. (2020) Application of the key characteristics of carcinogens to per and polyfluoroalkyl substances. Int. J. Environ. Res. Public Health, 17, 1668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Yao, X., et al. (2005) Genotoxic risk and oxidative DNA damage in HepG2 cells exposed to perfluooctanoic acid. Mutat. Res., 587, 38–44. [DOI] [PubMed] [Google Scholar]
- 25. Eriksen, K.T., et al. (2010) Genotoxic potential of the perfluorinated chemicals PFOA, PFOS, PFBS, PFNA and PFHxA in human HepG2 cells. Mutat. Res., 700, 39–43. [DOI] [PubMed] [Google Scholar]
- 26. Wielsoe, M., et al. (2015) Perfluoroalkylated substances (PFAS) affect oxidative stress biomarkers in vitro. Chemosphere, 129, 239–245. [DOI] [PubMed] [Google Scholar]
- 27. Singh, T.S., et al. (2012) Perfluorooctanoic acid induces mast cell-mediated allergic inflammation by the release of histamine and inflammatory mediators. Toxicol. Lett., 210, 64–70. [DOI] [PubMed] [Google Scholar]
- 28. Wang, X., et al. (2017) Serum metabolome biomarkers associate low-level environmental perfluorinated compound exposure with oxidative/nitrosative stress in humans. Environ. Pollut., 229, 168–176. [DOI] [PubMed] [Google Scholar]
- 29. Kamendulis, L.M., et al. (2014) Perfluorooctanoic acid exposure triggers oxidative stress in the mouse pancreas. Toxicol. Rep., 1, 513–521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hocevar, S.E., et al. (2020) Perfluorooctanoic acid activates the unfolded protein response in pancreatic acinar cells. J. Biochem. Mol. Toxicol., 34, e22561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Pitot, H.C. (1993) The molecular biology of carcinogenesis. Cancer, 72, 962–970. [DOI] [PubMed] [Google Scholar]
- 32. Hruban, R.H., et al. (2006) Pathology of genetically engineered mouse models of pancreatic exocrine cancer: consensus report and recommendations. Cancer Res., 66, 95–106. [DOI] [PubMed] [Google Scholar]
- 33. Agbaje, M., et al. (2018) Novel inflammatory cell infiltration scoring system to investigate healthy and footrot affected ovine interdigital skin. PeerJ, 6, e5097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Wang, L.M., et al. (2016) The prognostic role of desmoplastic stroma in pancreatic ductal adenocarcinoma. Oncotarget, 7, 4183–4194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Sinn, M., et al. (2014) a-Smooth muscle actin expression and desmoplastic stromal reaction in pancreatic cancer: results from the CONKO-001 study. Br. J. Cancer, 111, 1917–1923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Azevedo-Pouly, A.C.P., et al. (2014) RNA isolation from mouse pancreas: a ribonuclease-rich tissue. J. Vis. Exp., 90, e51779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Noy, R., et al. (2014) Tumor-associated macrophages: from mechanisms to therapy. Immunity, 41, 49–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Ying, H., et al. (2011) PTEN Is a major tumor suppressor in pancreatic ductal adenocarcinoma and regulates an NF-κB–cytokine network. Cancer Discov., 1, 158–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Guerra, C., et al. (2011) Pancreatitis-induced inflammation contributes to pancreatic cancer by inhibiting oncogene-induced senescence. Cancer Cell, 19, 728–739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Raimondi, S., et al. (2009) Epidemiology of pancreatic cancer: an overview. Nat. Rev. Gastroenterol. Hepatol., 6, 699–708. [DOI] [PubMed] [Google Scholar]
- 41. Juarez, J.C., et al. (2008) Superoxide dismutase 1 (SOD1) is essential for H2O2-mediated oxidation and inactivation of phosphatases in growth factor signaling. Proc. Natl Acad. Sci. USA, 105, 7147–7152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Veal, E.A., et al. (2007) Hydrogen peroxide sensing and signaling. Mol. Cell 26, 1–13. [DOI] [PubMed] [Google Scholar]
- 43. Che, M., et al. (2016) Expanding roles of superoxide dismutases in cell regulation and cancer. Drug Discov. Today, 21, 143–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Glasauer, A., et al. (2014) Targeting SOD1 reduces experimental non-small-cell lung cancer. J. Clin. Invest., 124, 117–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Gomez, M.L., et al. (2019) SOD1 is essential for oncogene-driven mammary tumor formation but dispensable for normal development and proliferation. Oncogene, 38, 5751–5765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Tsang, C.K., et al. (2018) SOD1 phosphorylation by mTORC1 couples nutrient sensing and redox regulation. Mol. Cell, 70, 502–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Tan, S.-X., et al. (2009) Cu, Zn superoxide dismutase and NADP(H) homeostasis are required for tolerance of endoplasmic reticulum stress in Saccharomyces cerevisiae. Mol. Biol. Cell, 20, 1493–1508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Zeeshan, H.M.A., et al. (2016) Endoplasmic reticulum stress and associated ROS. Int. J. Mol. Sci., 17, 327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Kodydkova, J., et al. (2013) Antioxidant status and oxidative stress markers in pancreatic cancer and chronic pancreatitis. Pancreas, 42, 614–621. [DOI] [PubMed] [Google Scholar]
- 50. Provenzano, P.P., et al. (2012) Enzymatic targeting of the stroma ablates physical barriers to treatment of pancreatic ductal carcinoma. Cancer Cell, 21, 418–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article are available in the article and in its online supplementary material.