*2.5. Expression of PCNA and Ki-67*

The expression of proteins involved in cell proliferation, PCNA and Ki-67, was determined in lung tissue (Figure 4). The cell proliferation markers, PCNA and Ki-67, were highly expressed (*p* < 0.0001) in the lungs of NNK-injected mice (no-C3G-HB, pre-C3G-HB, conti.-C3G-HB, and post-C3G-HB) relative to the saline-injected control mice (naive and control groups). The expression of PCNA was significantly higher compared to that of Ki-67. The level of PCNA and Ki-67 was significantly (*p* < 0.0001) reduced in the lungs of NNK-injected mice that were fed C3G-HB. As a percentage, the expression of PCNA was decreased by 41% to 64% (Figure 4A) and Ki-67 by 33% to 57% (Figure 4B), respectively. The results indicate a reduction of cell proliferation rate in lung tumors of mice fed with C3G-HB dietary supplement.

**Figure 4.** The expression of PCNA and Ki67 in lung tumors. Tumors were imaged using Zeiss Axioplan II and Axiocam HRC color camera at 200 × magnification. The number of (**A**) PCNA and (**B**) Ki-67 positive cells in each section (2 sections/mouse, *n* = 5) were counted using ImageJ software and presented as the expression in bar graphs. One-way ANOVA with Dunnett's test at α = 0.05 and α = 0.1 was used for data analysis. No-C3G-HB, not given C3G-HB supplemented diet; Pre-C3G-HB, C3G-HB supplemented diet only before NNK injection; Conti.-C3G-HB, C3G-HB supplemented diet continuously before and after NNK injection; Post-C3G-HB, C3G-HB supplemented diet only after NNK injection. \* Indicate statistical difference at *p* ≤ 0.05 with mean ± SD. NS, Results do not significantly different.

#### **3. Discussion**

We administered C3G-HB as a dietary supplement before, during, or after exposure to the pro-carcinogen NNK to evaluate the chemopreventive and chemotherapeutic effect of C3G-HB against NNK-induced lung tumorigenesis in A/JCr mice. The A/JCr mouse is recognized as an *in vivo* model for investigating carcinogen-induced lung tumorigenesis [22]. These mice develop spontaneous lung tumors over time [23]; hence, tumor observation in the naive (2/5 mice) and control (1/5 mice) groups is not surprising. To the best of our knowledge, this is the first study to investigate the chemopreventive effect of C3G-HB against NNK-induced lung tumorigenesis in vivo.

C3G is the most predominant anthocyanin in haskap berry extract. C3G represents about 90% of anthocyanins in haskap berry. Our primary goal is to develop a nutraceutical from haskap berry for use in preventing lung carcinogenesis. Developing pure C3G as a nutraceutical is not practical due to the cost of purification and loss of consumer perception as a natural health product. Hence, we did not test the effect of pure C3G; however, our previous *in vitro* study confirmed that pure C3G has a similar effect as of the extracts of the C3G-HB, and reduced the carcinogen-induced DNA damage and oxidative stress in BEAS-2B normal lung epithelial cells [21].

C3G-HB (6 mg of C3G in 0.2 g of lyophilized whole haskap berry power/mouse/day) significantly (*p* < 0.05) reduced the NNK-induced lung tumor multiplicity, the most sensitive indicator of potency [24], by 38% (pre-C3G-HB) and 35% (post-C3G-HB), respectively. However, administering C3G-HB continuously before and after NNK injection was less effective (only 22%) compared to the pre- and post-supplementation. In contrast, measurement of tumor area in histological samples revealed over 65% reduction upon C3G-HB ingestion. Similarly, Khan and colleagues reported a reduction of benzo(a)pyrene- and N-nitroso-tris-chloroethylurea-induced lung tumors by 53% and 74%, respectively, in A/JCr mice that were fed with 0.2% *w*/*v* C3G-rich pomegranate juice [10]. At tumor initiation stage, NNK is converted into electrophilic metabolites (via CYP450s enzymes) that covalently bind with DNA to form bulky DNA adducts leading to lung tumorigenesis [25,26]. Our previous findings confirmed that ethanolic and aqueous extracts of C3G-HB can attenuate the NNK-induced DNA double-strand breaks, suppress oxidative stress, and induce DNA damage repair proteins in normal lung epithelial BEAS-2B cells [21]. Oral administration of haskap berry attenuates oxidative stress in mice [19] and restores oxidative defense mechanisms by activating catalase, superoxide dismutase, glutathione peroxidase, and glutathione in mice that were exposed to ionizing radiation [27–29]. In addition, flavonoids have been reported to reversibly and irreversibly inhibit cytochrome 450 (CYP450) enzymes and interfere in xenobiotic metabolism [30]. The antioxidant activity of C3G-HB [21] may have attenuated the NNK-related electrophilic metabolites and hence inhibited the formation of DNA adducts that trigger lung tumorigenesis. Therefore, it is crucial to investigate the oxidative defense mechanism of the C3G-HB and its effect on CYP450 enzyme activity in the future.

PCNA is necessary for DNA synthesis (a processivity factor of DNA polymerases) and DNA repair (involved in nucleotide excision repair and base excision repair). PCNA is highly expressed during the active cell cycle; G1 phase, peaks at S-phase and declines during G2/M-phases. Ki-67 is expressed in G1-, S-, and G2-phases, but not in the G0-phase of the cell cycle [31]. Similar to previous findings [3–5], NNK induced lung cell proliferation, which is indicated by the highly expressed PCNA (12.7-fold) and Ki-67 (10-fold) in lung tissues of the no-C3G-HB group comparison to the naive group. The C3G-HB dietary supplementation reduced cell proliferation markers, PCNA (40–60%) and Ki-67 (30–60%), respectively. C3G, as a pure compound (250 and 500 μM) and in fruit extracts (0.2% *w*/*v* C3G-rich pomegranate juice) inhibits cell proliferation through deactivating MAPK and PI3K/AKT signaling pathways, which are activated by NNK at cancer progression [10,13,32]. Thus, it is necessary to study the effect of C3G-HB in cell proliferation pathways to understand its tumor inhibitory mechanism.

Regardless of the feed intake, the C3G-HB supplement reduced the body weight of the mice by 3.5% comparison to the regular diet (control vs. naive). C3G enhances energy metabolism by upregulating brown adipose tissue mitochondrial function [33]. Ingestion of C3G-rich blood orange juice [34] and purple sweet potato [35] results in reduced lipogenesis, including triglycerides through activation of

adenosine monophosphate-activated protein kinase (AMPK) signaling pathways. Therefore, enhanced metabolism and/or reduction of fat synthesis might account for the weight loss effect of C3G-HB in control mice over the naive group. AMPK is also identified as a metabolic tumor suppressor which reprograms cellular metabolism and prevent tumorigenesis [36]. Therefore, C3G-activated AMPK may have regulated energy levels that inhibit cell proliferation. These results suggest the beneficial health effect of C3G-HB against NNK-induced lung tumorigenesis *in vivo*.

In humans, C3G metabolism generates various metabolites such as C3G glucuronides, methylates of the C3G, i.e., peonidin-3-glucoside, and simple phenolic acids including protocatechuic acid, ploroglucinaldehyde, and hippuric acid [37–39]. Even though the metabolites of C3G are similar in humans and mice, the clearance rate of C3G in humans is slower than in mice. In humans, C3G metabolites are present in the blood plasma for ≤ 48 h. In contrast, in mice, a major fraction of C3G metabolites are excreted after 24 h [40,41]. Therefore, when considering C3G-HB as a nutraceutical for use in humans, the absorption, distribution, metabolism, and elimination of C3G should be evaluated using a properly designed human clinical study.

#### **4. Materials and Methods**

#### *4.1. Materials*

This study was performed at Dalhousie University's animal care facility, following the approval of the University Committee on Laboratory Animals (protocol 15–106). The carcinogen, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK, MW. 207.23 g/mol, Cat No. M325750) was purchased from Toronto Research Chemicals Inc., Toronto, ON, Canada. Frozen haskap berry cv. Tundra was obtained from LaHave Natural Farm, Blockhouse, NS, Canada. Female A/JCr albino mice at 3–4 weeks age (n = 50) were purchased from Charles River Laboratories, Inc., Montreal, QC, Canada.

#### *4.2. Preparation of C3G-HB and Analysis*

Frozen haskap berries were lyophilized, ground to a fine powder, and stored at −80 ◦C. A representative sample was analyzed to determine the nutrient composition (Harlow Institute, Department of Agriculture, Truro, NS, Canada). C3G was quantified by high-performance liquid chromatography and mass spectrophotometry (HPLC/MS/MS, Waters Limited, Mississauga, ON, Canada) after extraction (1 mg/mL) using methanol containing 1% acetic acid and filtered through a 0.22 μm nylon filter [20].

#### *4.3. Preparation of Dietary Supplement*

Ingestion of 1.5 g polyphenols/day for a healthy adult of 70 kg body weight is considered to be a health-promoting dose [42,43]. A health-promoting C3G-HB dose, equivalent to the animal dose, was calculated as follows [44]:

$$\text{Human equivalent dose (mg/kg)} = \text{Animal dose (mg/kg)} \times \frac{\text{Animal Km Factor}}{\text{Human Km Factor}} \tag{1}$$

Km factor = body weight (kg)/body surface area (m2). The Km factors of mouse and adult human are 3 and 37, respectively [44].

Accordingly, the experimental diet/mouse/day consisted of 0.2 g C3G-HB (equivalent to 6 mg of C3G/mouse/day) and 5% Splenda®mixed into regular mouse chow (Prolab®RMH 3000 from LabDiet, St. Louis, MO, USA) and formed into a 2 g (dry weight) pellet. The control diet consisted of regular mouse chow containing 5% Splenda®. C3G-HB powder was mixed thoroughly for 20 min to obtain a homogeneous preparation for use in making pellets. Pellets were prepared every two days and stored in sealed containers in the dark at 4 ◦C.

#### *4.4. Experimental Plan and Procedure*

The experiment was designed to evaluate the early and late intervention of C3G-HB against NNK-induced lung tumorigenesis (Figure 5). Mice were housed individually in filter-topped plastic cages and maintained under 12-h light-dark cycles. After one week of adaptation, mice were randomly divided into six groups, n = 5 for saline-injected naive and control groups and n = 10 for NNK injected no-C3G-HB, pre-C3G-HB, conti.-C3G-HB and post-C3G-HB groups). Prolab® RMH 3000 diet and distilled water were provided ad libitum. The C3G-HB or control pellets were given daily as a dietary supplement. Briefly, mice in naive and no-C3G-HB groups were given control pellets, while the diet of control, pre-C3G-HB, conti.-C3G-HB and post-C3G-HB groups was supplemented with C3G-HB (Figure 5). Mice in pre-C3G-HB were fed the C3G-HB supplemented diet for three weeks, prior to NNK injection and then switching to the control diet after NNK injection. Post-C3G-HB group was given the control diet until the NNK injection and then switched to C3G-HB supplemented diet until the end of the experiment. Three weeks after the start of dietary supplementation, a single dose of NNK (100 mg/kg body weight in 0.2 mL saline) was injected into the peritoneal cavity of mice to induce lung tumors. Mice in naive and control groups, received an equivalent volume of saline. Once a week, body weight was measured until the end of the experiment.

#### *4.5. Lung Tumor Assays*

4.5.1. Positron Emission Tomography-Magnetic Resonance Imaging (PET-MRI)

Before euthanizing mice, lung tumorigenesis was confirmed by PET-MRI at the Biomedical Translational Imaging Center (BIOTIC) at the IWK Children's Hospital, Halifax, NS, Canada. Briefly, three mice from naive and no-C3G-HB groups were randomly selected and fasted for six hours and then injected with 18F-fluorodeoxyglucose 100 μCi via the tail vein. After 30 min, the mice were

anesthetized, and a PET-MRI scan was performed. Breathing pattern, heart rate, and body temperature were monitored throughout the scanning period.

Twenty-two weeks after NNK treatment, mice were anesthetized with isoflurane. Blood samples were collected by cardiac puncture, and a higher dose of isoflurane was used to sacrifice mice. Dissected lungs were perfused and washed in phosphate-buffered saline (PBS) before being fixed in 10% (*v*/*v*) acetate-buffered formalin. Peripheral lung tumors were enumerated using a dissecting microscope. Lungs were embedded in paraffin, and paraffin-embedded tissues were stored for histopathological examination.

#### 4.5.2. Tumor Histology and Tumor Area

Paraffin-embedded lungs were cut into 5 μm thick sections (50 tissue sections/lung) using a microtome (Leica Rm 2255, Leica Biosystems, Concord, ON, Canada). Three sections representing three areas of the lungs were selected at predetermined depths and stained with hematoxylin and eosin (H and E). The H and E-stained lung sections were imaged under bright field microscopy (AxioPlan 11MOT AxioCam HRc, Carl Zeiss Canada Ltd., Toronto, ON, Canada), and lung tumor area was quantified using ImageJ software [45].

#### 4.5.3. Immunohistochemistry

The expression of proliferating cell nuclear antigen (PCNA) and Ki-67 was evaluated by immunohistochemistry (IHC). Briefly, paraffin sections were deparaffinized in xylene, rehydrated through an ethanol solution gradient and washed carefully in running tap water. Antigen retrieval was carried out by heating sections in 0.01 M citrate buffer (pH 6.0) for 30 min in the decloaking chamber. Endogenous peroxidase activity was quenched by incubating the section in 3% H2O2 in Tris-buffered saline (TBS) for 10 min. Non-specific binding sites were blocked by incubation with rodent block (M) from Biocare Medical, Pacheco, CA, USA. The sections were incubated overnight at room temperature in a humid chamber with monoclonal antibodies against PCNA (1:6000 dilution) or Ki-67 (1:50 dilution). After several washes with TBS, the slides were incubated with anti-rabbit horseradish peroxidase-conjugated secondary antibody (EnVision, Dako North America Inc.Carpinteria, CA, USA) for 30 min, then washed three times with TBS and incubated with chromogen 3-diaminobenzidine (DAB Chromogen kit, Biocare Medical, Pacheo, Ca, USA) for 3 min. The slides were carefully rinsed under running tap water and counterstained with hematoxylin. The slides were observed under bright field microscopy (AxioPlan 11MOT AxioCam HRc, Carl Zeiss Canada Ltd., North York, ON, Canada) at 200× magnification.

#### *4.6. Statistical Analysis*

The observed differences in the tumor multiplicity, tumor area, PCNA and Ki-67 expression were tested for statistical significance using one-way Analysis of Variance (ANOVA). Tukey's pairwise comparison and Dunnett's test with a 95% confidence interval was used for comparisons among multiple groups. Minitab statistical software was used for data analysis.

#### **5. Conclusions**

In summary, we have demonstrated that dietary supplementation of C3G-HB can inhibit the NNK-induced lung tumorigenesis in A/JCr mice. C3G-HB may be a promising dietary supplement to suppress lung cancer development among high-risk populations such as smokers, possibly via effects on critical cellular signaling pathways that regulate cell proliferation. Future studies of the effects of C3G-HB on phase I and phase II metabolic enzymes and cell signaling pathways will elucidate the mode of action of C3G-HB against lung carcinogenesis.

**Author Contributions:** M.A. performed all the experiments, analyzed the data, and drafted the manuscript. H.P.V.R., the principal investigator, designed the study. All the authors made intellectual contributions to the manuscript, have read, edited, and approved the final manuscript.

**Funding:** This study was supported by the Discovery Grant of Natural Sciences and Engineering Research Council (NSERC) of Canada (HPVR; Grant number RGPIN2016-05369) and the Cancer Research Training Program of the Beatrice Hunter Cancer Research Institute (BHCRI) supported by the Saunders-Matthey cancer prevention foundation (MA; Grant number CRTP2018). APC was sponsored by MDPI.

**Acknowledgments:** The authors would like to recognize the intellectual contribution of Dr. Michael Johnston, former Director of the BHCRI, at the early stage of this project designing.

**Conflicts of Interest:** The authors declare that there is no conflict of interest.
