Boysenberry and apple juice concentrate reduced acute lung inflammation and increased M2 macrophage‐associated cytokines in an acute mouse model of allergic airways disease

Abstract Bioactive compounds including anthocyanins and other polyphenols are associated with reduced lung inflammation and improved lung function in asthma and other lung diseases. This study investigated the effects of a Boysenberry and apple juice concentrate, high in cyanidin glycosides, ellagitannins, and chlorogenic acid, on a mouse model of allergic airways inflammation. Male C57BL/6J mice were orally gavaged with 2.5 mg/kg of total anthocyanins (TAC) from BerriQi® Boysenberry and apple juice concentrate (0.2 mg/kg human equivalent dose) or water control 1 hr before an acute intranasal ovalbumin (OVA) challenge and were gavaged again 2 days after the intranasal challenge. Consumption of BerriQi® Boysenberry and apple juice concentrate significantly decreased OVA‐induced infiltrating eosinophils, neutrophils, and T cells in the lung, and mucous production. Quantification of gene expression for arginase (Arg1), chitinase 3‐like 3 (Ym‐1), found in inflammatory zone (Fizz1), which have been associated with an anti‐inflammatory macrophage phenotype (M2), found significantly increased Arg1 expression in the lung in the Boysenberry and apple juice concentrate treatment group. There was also increased production of M2‐associated cytokines C‐X‐C motif chemokine ligand (CXCL) 10 and C‐C motif chemokine ligand (CCL) 4. These results suggest that consumption of BerriQi® Boysenberry and apple juice concentrate promoted a shift toward an anti‐inflammatory environment within the lung leading to reduced immune cell infiltration and tissue damage.


| INTRODUC TI ON
Asthma is a heterogeneous, chronic, inflammatory lung disease characterized by reversible airways obstruction, bronchospasm, and infiltration of immune cells (Agache & Akdis, 2016;Agrawal & Bharadwaj, 2005;Barnes, 1996). It is estimated that 150 million people are affected by asthma worldwide, with a 5%-15% prevalence in children (WHO, 2003), and there is evidence that early life exposure to air pollution caused by vehicle exhaust, environmental dust, and industrial processes increases the severity of asthma in children (Jung et al., 2015;Hsu et al., 2015;Miller & Peden, 2014).
The respiratory symptoms such as cough and wheeze are worsened by exposure to pollution (Hoek et al., 2012). Proinflammatory cytokine production in response to allergens by immune cells is further increased with concomitant pollution exposure (Acciani et al., 2013;Brandt et al., 2015;Carlsten et al., 2016;Kim et al., 2011;van Voorhis et al., 2013). Eosinophils, in particular, produce reactive oxygen species and cytokines, leading to epithelial damage and contribute to mucosal inflammation and the recruitment of other proinflammatory immune cells (Amin et al., 2016;Bossley et al., 2012;Brown et al., 1998;Trivedi & Lloyd, 2007). These repeated acute inflammatory responses lead to tissue damage and remodeling, contributing to airway hyperresponsiveness, mucus cell hyperplasia, fixed airway flow obstruction, and loss of lung function over time (Ahdieh et al., 2001;Al-Muhsen et al., 2011;Bergeron et al., 2009;Brightling et al., 2012).
Previously, we have identified that Boysenberry consumption led to decreased chronic lung inflammation and improved lung tissue repair in an animal model of chronic allergic lung inflammation (Shaw et al., 2016). Boysenberries contain high concentrations of anthocyanins (261 mg/g), ellagitannins, and other polyphenols (241 mg/g) (Cooney et al., 2004;Ghosh et al., 2006;McGhie et al., 2012). Apple contains approximately 120-200 mg/g total polyphenols (Paturi et al., 2014), and we have found that procyanidin-enriched apple extracts suppressed IL-4-mediated cytokine production in cell culture models of lung epithelial allergic inflammation Sawyer et al., 2017).
There is increasing interest in understanding the mechanisms of action that specific plant bioactives have in the human body. This is partially to better understand the benefits of consuming specific fruits and vegetables and partially to add value to specific foods through validated health claims. There is also interest in determining if combining specific plants containing different polyphenols can augment the health benefits above those seen with the individual plant. Use of animal models, where dietary intake can be tightly controlled, is useful for both demonstrating/revealing the efficacy for identified compounds and determining the biological mechanisms of action. The aim of this study was to determine whether the combination of Boysenberries and apple, as found in BerriQi® Boysenberry and apple juice concentrate at a dose of 2.5 mg/kg total anthocyanins (TAC), could reduce allergic airways inflammation in response to acute ovalbumin (OVA) exposure in a mouse model system. We also sought to determine the mechanisms involved in any ameliorating effect.

| Mice and materials
C57BL/6J male mice were group housed on 12-hr light/dark cycle in a conventional animal facility at The New Zealand Institute for Plant and Food Research Limited (Palmerston North, New Zealand). Mice were fed Prodiet RMH1800 standard chow for rodents (Lab Diet) and filtered water ad libitum throughout the study, and all attempts to minimize suffering were made. All experimental procedures were approved by the AgResearch Grasslands Animal Ethics Committee (AE approvals #14839, #14731 and #14016) and carried out in accordance with the Animal Welfare Act (1999). A commercially available Boysenberry and apple juice concentrate ingredient (BerriQi®) was supplied by and is available from Anagenix Ltd (Auckland, New Zealand). Legendplex™ 13-plex Th cytokine, proinflammatory cytokine, and proinflammatory chemokine panels, Zombie NIR™ fixable viability dye, and anti-mouse CD3 (clone 17A2), CD4 (clone phase, 100% A, was held for 2 min before being ramped linearly to 88% A at 14 min, returning to 5% A at 15 min, and held for 4 min before resetting to the original conditions. The sample injection volume was 10 µl. The MS data were acquired in the positive mode.
Standards of cyanidin-3-O-glucoside were used to quantitate anthocyanin concentrations with PDA detection at 520 nm, and the results for individual and total anthocyanin concentrations are reported as cyanidin-3-O-glucoside equivalents.
Other phenolic compound separation was achieved using a Hypersil GOLD aQ 1.9 µ C18 175 Å (Thermo Scientific), 150 × 2.1 mm column maintained at 45°C. The solvents were (A) water + 0.1% formic acid and (B) acetonitrile + 0.1% formic acid (flow rate, 200 µl/ min). The initial mobile phase, 95% A/5% B, was ramped linearly to 85% A at 10 min, held for 3.75 min, then ramped linearly to 75% A at 18 min, 67.2% A at 25 min, 50% A at 28 min, and 3% A at 29 min, and held for 4 min before resetting to the original conditions. The sample injection volume was 4 µl. The MS data were acquired in the negative mode. The phenolic acids, gallic acid, ellagic acid, protocatechuic acid, chlorogenic acid (3-p-caffeoylquinic acid) and caffeic acid, the flavan-3-ols, catechin and epicatechin, the procyanidin B2, the nonglycosylated flavonols, quercetin and myricetin, and the chalcone, phloretin-2-O-glucoside were quantified by LC-MS using pure standards of these compounds. Detected derivatives of coumaric acid are expressed as coumaric acid equivalents. Detected flavonol glycosides were quantified by LC-MS using a pure standard of quercetin-3-O-glucoside and are expressed as quercetin-3-O-glucoside equivalents. Other detected chalcones were quantified as phloretin-2-O-glucoside equivalents. Hydrolyzable tannins were quantified by LC-MS using a standard of sanguiin H6 that had been isolated previously (>98% purity by LC-MS). Other detected tannins were quantified as sanguiin H-6 equivalents.

| Ovalbumin-induced airway inflammation model
Allergic airway disease was induced as previously described (Shaw & Harper, 2013;Shaw et al., 2017). For the Boysenberry and apple interventions, mice were randomized into receiving either water (vehicle control) or 2.5 mg/kg TAC in the BerriQi® Boysenberry and apple juice concentrate as previously described .
Briefly, mice were fasted for 4 hr before being orally gavaged with water (control) or at a dose of 2.5 mg/kg body weight TAC in the BerriQi® Boysenberry and apple juice concentrate made up to a total volume of 200 µl in water 1 h before OVA challenge and again 2 days postchallenge. Mice were sacrificed by anesthetic overdose 4 days following intranasal ovalbumin challenge and immune parameters, and gene expression was analyzed.

| Immune parameter analysis
Bronchoalveolar lavage fluid (BALF) and lung tissues were collected as previously described, and immune cells were phenotyped by flow cytometry (Shaw & Harper, 2013). Lung tissue supernatant for cytokine analysis was prepared as previously described . Cytokine production in lung tissue supernatant was measured by Legendplex bead-based multiplex immunoassays as per the manufacturer's instruction. Both cell phenotyping and the cytokine multiplex assays were analyzed using a BD FACSverse (BD Biosciences). H&E and AB-PAS histological staining were performed by Massey IVABS histology unit.

| Real-time qPCR analysis
Mouse lung tissue was snap-frozen in liquid nitrogen and crushed into powder using a mortar and pestle with liquid nitrogen to preserve RNA integrity. The RNA was extracted from the powder using a TRIzol total RNA extraction protocol. RNA was quantified using an LVis plate in a POLARstar Omega plate reader (BMG Labtech), and the quality of the ribosomal RNA bands confirmed by agarose gel electrophoresis (data not shown). cDNA was synthesized from the lung sample RNA templates using the iScript™ cDNA Synthesis kit as per the

| Statistical analysis
All data were analyzed using one-way analysis of variance (ANOVA) with a Tukey's post hoc test and graphed in SigmaPlot 12.5 (Systat Software Inc.).

| Chemical composition of the Boysenberry and apple juice concentrate
The results of the LC-MS analysis showed that cyanidin glycosides, ellagitannins, and chlorogenic acid were the major components in BerriQi® Boysenberry and apple juice concentrate (

| Effect of Boysenberry and apple juice concentrate intervention on ovalbumin-induced allergic airways inflammation
Acute intranasal OVA exposure resulted in an infiltration of immune cells into the lung (Figure 1a) and increased mucous production There was a trend toward an increased percentage of CD206+/ CD14-macrophages in the lungs of mice that consumed 2.5 mg/kg TAC BerriQi® Boysenberry and apple juice concentrate (Figure 3a). To determine whether it was possible that there was increased alternatively activated macrophages, we measured the gene expression of Arg1, Ym-1, Fizz1, and Nos2 in lung tissue. We found that both OVA alone and 2.5 mg/kg TAC BerriQi® Boysenberry and apple juice concentrate consumption led to a significant (p < .01) fold increase in Ym-1 (4.0 ± 2.4 and 4.7 ± 1.8, respectively) and Fizz1 (15.4 ± 11.6 and 23.0 ± 13.6, respectively) gene expression compared to naïve mice (Table 2). 2.5 mg/ kg TAC BerriQi® Boysenberry and apple juice concentrate also led to a significant increase in Arg1 compared to OVA alone (p < .05) and naïve (p < .001), whereas OVA alone did not significantly increase Arg1 compared to naïve mice (Table 2). We found no significant fold change in Nos2 gene expression between any of the treatment.

| D ISCUSS I ON
We evaluated the effects of dietary supplementation with 2.5 mg/ kg TAC BerriQi® Boysenberry and apple juice concentrate, on the immune responses in a mouse model of acute allergic airways inflammation. Our results show that consumption of 2.5 mg/kg TAC BerriQi® Boysenberry and apple juice concentrate reduced granulocyte and local T-cell infiltration into the lung after OVA challenge, but did not alter T-cell activation within the lung draining lymph node or the levels of classical Th-2 and Th-1 cytokines in the lung at four days following OVA challenge. Our current results indicated that BerriQi® Boysenberry and apple juice concentrate had little impact on the Th-2/Th-1 mediated allergic response of mice, but rather targeted innate proinflammatory immune pathways. This is consistent with our previously reported finding in a mouse model of chronic allergic airways inflammation using 10 mg/kg TAC Boysenberry juice concentrate (Shaw et al., 2016). Chemical composition analysis showed that the BerriQi® Boysenberry and apple juice concentrate formulation contained high concentrations of cyanidin glycosides, ellagitannins, and chlorogenic acid. These compounds have been previously shown to reduce inflammatory signaling in vitro (Cassidy et al., 2015;El-Shitany et al., 2014;McGhie et al., 2012) and in vivo animal models of inflammation (Auclair et al., 2008;Denis et al., 2016;Guan et al., 2017;Impellizzeri et al., 2015;Shaw et al., 2017). Our current results suggest that consumption of 2.5 mg/kg TAC BerriQi® Boysenberry and apple juice concentrate, which also contains high levels of ellagitannins and chlorogenic acid, could have broader lung health benefits beyond allergic asthma disease by promoting the resolution of inflammation caused by innate immune cell overactivation. We found that both OVA alone and with BerriQi® Boysenberry and apple juice concentrate consumption resulted in a significant fold increase in Ym-1 and Fizz1 gene expression compared to naïve mice at 4 days following challenge, and we did not see increased Nos2

Consumption of BerriQi®
gene expression in any of the groups. These results suggested that 4 days post-OVA challenge the infiltrating monocyte/macrophages were more M2-like rather than M1-like. This is consistent with other studies that have shown that lung macrophages express increased M2-associated genes following OVA challenge (Siddiqui et al., 2013).
Alternatively activated macrophage Fizz1 expression, in particular, has been associated with regulating Th-2-mediated lung inflammation by modulating IL-4 and IL-5 (Nair et al., 2009). It has also been shown that depletion of alternatively activated macrophages does not ameliorate allergic airways inflammation (Nieuwenhuizen et al., 2012).
However, BerriQi® Boysenberry and apple juice concentrate led to a significant increase in Arg1 compared to both OVA alone and naïve and OVA alone did not significantly increase Arg1 compared to naïve mice. Previous studies have shown that changes in Fizz1 and Ym-1 gene expression can be constitutive in M2-like macrophages (Wong et al., 2010), and arginase activity regulation has been shown to be regulated independently of Fizz1 and Ym-1 gene expression in lung macrophages (Raes et al., 2002). Arginase expression, particularly by M2-like macrophages, has been associated with lung remodeling (Martinez et al., 2009), and increased arginase activity is associated with lower iNOS activity through substrate competition, leading to reduced inflammation (Hey et al., 1997;Johann et al., 2007;Mori & Gotoh, 2000). The increase in Arg1 gene expression is similar to our previously reported study showing increased arginase protein expression by alternatively activated macrophages as a result of chronic Boysenberry consumption (Shaw et al., 2016). Further, research looking at an animal model Th-2-mediated inflammation has identified M2 macrophage-derived Fizz1 as a key limiting factor for Th2-mediated pulmonary inflammation (Nair et al., 2009).

F I G U R E 4
BerriQi® Boysenberry and apple juice concentrate does not alter classical Th-1/Th-2 cytokines. Lung tissue production of (a) IL-5, (b) IL-9, (c) IL-10 (d) IFNγ, and (e) TNFα were determined 4 days post-OVA challenge by Legendplex. Data presented as mean ± SEM for two experimental replicates with n = 9-10 per treatment groups Consistent with Boysenberry and apple juice concentrate polyphenols resulting in a greater shift to an anti-inflammatory environment, mice that consumed 2.5 mg/kg TAC BerriQi® Boysenberry and apple juice concentrate showed increased levels of the cytokines CXCL10 and CCL4, which are produced by M2 macrophages, compared to OVA alone and naïve mice. CXCL10 and CCL4 are chemokines that attract monocytes/macrophages, and CXCL10 has been shown to also inhibit the infiltration of eosinophils in response to allergic airways inflammation (Su et al., 2008).
IL-17A was also increased in the BerriQi® Boysenberry and apple juice concentrate group compared to naïve mice only, and the levels of IL-17F were not affected by any of the treatments at the time point measured. High IL-17 expression, including IL-17A and IL-17F, has been implicated in asthma pathogenesis (Gurczynski & Moore, 2018;Wang & Liu, 2008). However, there is also some evidence that elevated IL-17A (Linden & Dahlen, 2014) increases the abundance of MMP-9, an important tissue remodeling protein in asthma (Shaw et al., 2016) and late-stage increases in IL-17A concentration can induce apoptosis of neutrophils and eosinophils (Linden & Dahlen, 2014;Wang & Liu, 2008  Anthocyanins have also been shown to inhibit proinflammatory proteins (Esposito et al., 2014;Fu et al., 2014) and activate anti-inflammatory pathways in models of inflammation (Chen et al., 2016;Edirisinghe et al., 2011;Khanna et al., 2001;Koh et al., 2015;Liu et al., 2015). Previously, we reported that 10 mg/kg TAC Boysenberry juice concentrate can increase the abundance of alternatively activated (M2) macrophages, which promote tissue repair in a chronic model of airways inflammation (Shaw et al., 2016). It is possible that the combination of the different polyphenols in the BerriQi® Boysenberry and apple juice concentrate acts on a number of different immune pathways to regulate the immune responses to OVA.
We found that mice that consumed BerriQi® Boysenberry and apple juice concentrate had reduced immune cell infiltration in response to acute OVA challenge, and this could be as a result of a shift toward an anti-inflammatory environment within the lung. These results highlight the potential of anthocyanin-rich Boysenberry and apple dietary supplementation to modulate innate immune pathways during acute allergic lung inflammation. Further work is needed to determine whether these pathways are also altered in other lung inflammatory conditions, such as air pollution exposure, and to determine the underlying molecular mechanisms the mediate the reported effects as well as clinical studies to show if these findings are translatable to human health.

ACK N OWLED G M ENTS
The authors acknowledge the efforts of TC Chadderton and September 2016 (PFR reference #33609).

CO N FLI C T O F I NTE R E S T
None of the other authors declare any other Competing Interest.

E TH I C A L R E V I E W
All animal experimental procedures were approved by the AgResearch Grasslands Animal Ethics Committee (AE approvals #14839, #14731 and #14016) and carried out in accordance with the New Zealand Animal Welfare Act (1999).