Inflammation and cancer : The role of the human neutrophil

Chronic inflammation of both infective and non-infective origin has been implicated in the aetiology of approximately 30% of all human epithelial malignancies. The primary carcinogens are reactive oxygen species (ROS) derived from activated, infiltrating cells of the innate immune system, especially neutrophils, which inflict oxidative damage on the DNA of bystander epithelial cells. The consequence is gene modifications which initiate cellular transformation. The process of tumourigenesis is exacerbated by the sustained generation of pro-proliferative ROS, as well as by the release of neutrophil-derived cytokines and proteases, all of which contribute to tumour promotion and progression. It is now well recognised that, in addition to inflammation causing cancer, many cancers per se induce an inflammatory response, with a high magnitude of neutrophil influx being indicative of a poor prognosis. In this setting, CXC chemokines produced by tumours not only promote neutrophil influx and hyperreactivity, but also cause autocrine activation of the proliferation of the chemokine-producing tumour cells. These various mechanisms of inflammation-mediated tumourigenesis are the primary focus of this review, together with a consideration of neutrophil-targeted anti-inflammatory strategies with potential as adjunctive cancer therapy.


Introduction
The link between cancer and inflammation was described 150 years ago by the distinguished German pathologist Rudolph Virchow. 1,24][5][6] In this setting, reactive oxygen and nitrogen species (ROS/RNS) released by activated phagocytes at sites of inflammation inflict oxidative damage on neighbouring bystander cells, especially epithelial cells, initiating tumourigenesis. 7,83][4][5][6][7][8][9][10][11][12][13][14][15][16][17] Although inflammation is a major and primary cause of cancer, many cancers per se also activate an inflammatory response, resulting in infiltration by various types of myeloid cells of the innate immune system.9][20][21][22][23] Although this tumour-associated inflammatory response is potentially protective, at least in the case of neutrophils, monocytes/macrophages and dendritic cells, 19,20 it may also contribute to tumour progression and metastasis through the mechanisms alluded to above. 9 this review, we focus on the pro-tumourigenic potential of the neutrophil, which, amongst other types of immune and inflammatory cells, is abundant in tumours, appearing to be an important, independent predictor of poor outcome in many, 3,21 but not all, types 24 of malignancy.The major themes addressed here are the roles of neutrophil-derived/-associated ROS, chemokines/cytokines, proteinases and adhesion molecules in tumour initiation, promotion, progression and metastasis, as well as the potential role of anti-inflammatory strategies in cancer prevention and therapy.

Neutrophils and carcinogenesis
As described by Weitzman and Gordon in their seminal review 7 and, more recently, by Knaapen et al. 10 , the propensity for cancers to develop at sites of inflammation is well recognised, the association being supported by compelling epidemiological and experimental evidence.Examples of inflammation-associated cancers, primarily epithelial, of both infective and non-infective origin are shown in Tables 1 and 2, respectively.In the setting of inflammation-associated cancer, phagocyte-derived ROS -produced and released extracellularly by infiltrating neutrophils -have been identified as the primary offenders. 7,10These indiscriminate, toxic agents are potent carcinogens, posing the potential hazard of oxidative damage to the DNA of bystander, host structural cells at sites of inflammation and resulting in the gene modifications which precede cellular transformation.
Convincing evidence demonstrating the carcinogenic potential of ROS was derived from experiments in which eukaryotic structural cells and lymphocytes were exposed to activated neutrophils, to cell-free enzymatic ROS-generating systems, or to the relatively stable, cell-permeant ROS hydrogen peroxide (H 2 O 2 ) in vitro, which resulted in severe oxidative stress and damage to the genetic material of these cells. 7,10In all of these systems, direct oxidative damage to DNA appears to involve intracellular conversion of H 2 O 2 to a highly potent and reactive ROS -hydroxyl radical -probably by Fenton-type mechanisms involving electron donation by heavy metals.The types of ROS-mediated damage include: (1) gross chromosomal damage (sister chromatid exchanges), (2) single-and double-DNA strand breaks and (3) oxidative damage to the bases in DNA. 7,10In the case of the latter, the signature of oxidative damage is conversion of guanosine to 8-hydroxydeoxyguanosine, although the other DNA bases are also vulnerable to oxidative damage. 7RNS produced predominantly by macrophages result in the formation of reactive aldehydes and malondialdehydes which also induce point mutations. 8n addition to the direct, DNA-damaging activities of phagocyte-derived ROS, these oxidants also inhibit the activities of several DNA repair enzymes, thereby exacerbating oxidative damage to genetic material. 10n this context, it is noteworthy that hypochlorous acid generated via the H 2 O 2 -dependent oxidation of chloride ions by myeloperoxidase (MPO), the neutrophil/monocyte primary granule enzyme, has been reported to interfere with the base excision repair enzyme poly (ADP-ribose) polymerase. 25Other DNA repair enzymes which are susceptible to oxidative inactivation include the glycolase Ogg1 and topoisomerase II, which are inactivated by nitric oxide and H 2 O 2 , respectively, compromising repair of 8-hydroxydeoxyguanosine moieties and strand scission/ligation. 10,26,27Exposure to sunlight Sources [1][2][3][4][5][6][7][8][9] Clearly, ROS-mediated direct damage to DNA, together with oxidative dysfunction of DNA repair enzymes, predisposes to gene modifications, which, particularly in the case of mutations to tumour suppressor and promoter genes, may lead to cellular transformation.However, these mechanisms are not the only ones by which phagocyte-derived ROS contribute to carcinogenesis.Other mechanisms include: (1) oxidative conversion of pre-carcinogenic chemicals/xenobiotics to complete carcinogens 10,28 , (2) redox activation of intracellular signalling mechanisms, which not only promote aberrant cellular proliferation, but also intensify inflammation-related oxidative stress [29][30][31][32][33][34] and (3) oxidative suppression of the proliferative activity of anti-tumour T lymphocytes 35,36 .
In addition to these mechanisms, several neutrophil-derived chemokines/ cytokines, proteinases and adhesion molecules also contribute to tumourigenesis via their pro-proliferative, pro-angiogenic and prometastatic activities.

Neutrophil-mediated oxidative activation of pre-carcinogens
Neutrophil-derived ROS, specifically those generated by the MPO/H 2 O 2 /halide system, have been implicated in the transformation to carcinogens of chemical pollutants generated by industrial, motor vehicle and household combustive processes, as well as those present in cigarette smoke. 10Examples of the former include aromatic and heterocyclic amines, especially polycyclic aromatic hydrocarbons, while benzo(a)pyrene in cigarette smoke undergoes oxidative conversion to BPDE (bay-region diol expoxides), which is mutagenic via formation of covalent adducts with guanine. 10In addition, MPO-derived ROS have been reported to convert the anti-cancer drug etoposide to its potentially mutagenic phenoxy radical, which may explain the increased frequency of secondary myeloid leukaemia in cancer patients treated with this agent. 28

Redox activation of cellular proliferation
0][31][32] However, when structural cells, especially epithelial cells, are subjected to intense oxidative stress, whether directly as a consequence of protracted activation of Nox enzymes or indirectly because of influx of extracellular H 2 O 2 as a result of proximity to activated phagocytes, or both, then cell proliferation as a consequence of dysregulated intracellular signalling may ensue.Although disputed by those who believe that over-exposure to H 2 O 2 is more likely to drive the cells into apoptosis, [29][30][31][32] this scenario is countered to some extent by the following lines of evidence from experimental sources: (1) exposure of a Barrett's oesophagus adenocarcinoma cell line to low concentrations of H 2 O 2 resulted in cell proliferation which was associated with sequential activation of extracellular regulated kinase 2, MAPK, the transcription factor, nuclear factor kappa B (NFκB) and Nox 5-S 33 ; and (2) exposure of human oral cancer cells to the H 2 O 2 -producing microorganism Enterococcus faecalis resulted in catalase-inhibitable activation of the epidermal growth factor receptor and cell proliferation, underscoring the association between infection with this bacterial pathogen and oral carcinogenesis. 34

ROS-mediated inactivation of tumour-targeted T cells
Although H 2 O 2 at low concentrations can trigger the proliferation of epithelial cells via intracellular, redox signalling mechanisms, at higher concentrations this ROS can also promote the oxidative inactivation of the protective activities of T lymphocytes. 35,36In the setting of murine models of experimental tumourigenesis, infiltrating phagocytes, most

Review Article
Neutrophils and cancer Page 2 of 6

Neutrophil-derived cytokines in tumourigenesis
Although originally believed to have a very short lifespan and an extremely limited biosynthetic capacity, the survival time of neutrophils in the circulation of healthy humans has recently been reported to be 5.4 days. 37Following extravasation to sites of infection, tissue injury or cancer, this time may be considerably longer because of exposure to anti-apoptotic cytokines, especially granulocyte colony-stimulating factor (G-CSF) and granulocyte/macrophage colony-stimulating factor (GM-CSF). 38Extended survival of neutrophils is associated with acquisition of the capacity, albeit limited, to synthesise cytokines or chemokines, 14,39 some of which are already stored in pre-synthesised, rapidly mobilisable form in cytoplasmic secondary and tertiary granules. 40,41][41] With respect to tumour promotion/progression, the most prominent of these are VEGF, which promotes tumour neovascularisation, and IL-8, which not only sustains neutrophil influx and activation, but also promotes tumour cell proliferation by the autocrine and paracrine mechanisms described under 'Chemokines and tumourigenesis'.
Although the evidence is somewhat less compelling than that for VEGF and IL-8, several other neutrophil-derived cytokines have been implicated in tumour promotion/progression and angiogenesis.Because these have been extensively reviewed recently, 14 they are considered only briefly here.APRIL (also known as 'a proliferation-inducing ligand') and BAFF (B cell activation factor, BLyS), both of which belong to the TNF ligand family, interact with several receptors, especially BMCA (B cell maturation antigen), TAC1 and BAFF receptor, inducing B cell proliferation and survival.Both of these cytokines are produced by tumour-infiltrating neutrophils, and have been implicated in tumour promotion in malignancies such as diffuse large B cell lymphoma and multiple myeloma. 14costatin M (OSM) and hepatocyte growth factor (HGF) are cytokines which are both stored and synthesised by tumour-infiltrating neutrophils.OSM appears to mediate tumour progression via induction of detachment of tumour cells and activation of synthesis of pro-angiogenic VEGF and fibroblast growth factor by endothelial cells, while HGF induces an invasive phenotype. 14ese various cytokines and their reported roles in tumourigenesis are summarised in Table 3, with the exception of IL-8 which is discussed in detail below.

Chemokines and tumourigenesis
Chemokines are a group of low molecular weight chemotactic cytokines which promote the receptor-mediated migration of cells of the innate and adaptive immune systems.In the case of neutrophils, these cells are attracted to sites of tissue injury or infection by members of the sub-family of CXC/ELR-motif-positive chemokines (CXC denotes the presence of an intervening amino acid, X, between the first two conserved cysteine residues, while the ELR motif is a glu-leu-arg sequence preceding the first conserved cysteine residue).The various members of this chemokine sub-family are shown in Table 4, with the potent neutrophil chemoattractant IL-8 (CXCL8) predominating.Notwithstanding production by cells of the innate and adaptive immune systems, CXC/ELR + chemokines, especially IL-8, are also produced by various types of structural cells, including epithelial and endothelial cells, fibroblasts and smooth muscle cells.The major counter-receptor for these CXC/ELR + chemokines is CXCR2 (IL-8 also interacts with CXCR1), which is expressed not only on neutrophils and mast cells, but also on epithelial and endothelial cells. 12portantly, and aside from their primary role in neutrophil mobilisation, CXC/ELR + chemokines and CXCR2 are also expressed by a diverse range of human cancers, including cancers of the breast, bladder, cervix, colon, liver, lymphatics, oesophagus, ovary, prostate and skin. 12,13,42In this setting, these chemokines drive tumour expansion via both autocrine and paracrine pro-proliferative interactions with CXCR2expressing tumour cells. 12,13,42,43In the case of oesophageal squamous epithelial cells, and probably other tumour cell types, CXCR2-mediated proliferation results from activation of the transcription factor early growth response-1. 11In addition, tumour neovascularisation is mediated via the pro-angiogenic activities of these chemokines, especially IL-8, 12 while the chronic influx of inflammatory cells exacerbates ROS-mediated oxidative damage to DNA and immunosuppression.

Neutrophil-derived proteases in tumour angiogenesis and metastasis
Neutrophil-derived proteases -specifically elastase and matrix metalloproteinase-9 (MMP-9) stored in primary and secondary/tertiary granules, respectively -have also been implicated in inflammation-associated tumour neovascularisation and invasion.Elastase has been reported to degrade the intercellular adhesion molecule cadherin, 44 while MMP-9 is a potent inducer of angiogenesis and tumour metastasis. 16,17

Neutrophil adhesion molecules in tumour metastasis
Notwithstanding the expression of counter-receptors for endothelial adhesion molecules by some types of tumours, 2 pro-adhesive interactions between circulating neutrophils, albeit in an animal model  45 In this setting, neutrophil/tumour cell adhesion is mediated via interactions of the β2-integrin Mac-1 on neutrophils, with its counter-receptor, intercellular adhesion molecule-1 (ICAM-1), on tumour cells. 45e aforementioned mechanisms of neutrophil/inflammation-mediated tumourigenesis are summarised in Figures 1 and 2. Proposed mechanism by which chronic inflammation leads to oxidative damage to the DNA of bystander tissue cells as a result of the sustained release of reactive oxygen species (ROS) from infiltrating neutrophils.Tumour initiation is followed by promotion and progression as a result of ongoing exposure to neutrophil-derived pro-proliferative and pro-angiogenic mediators.

Inflammation-targeted chemotherapy and immunotherapy in cancer
7][48] Although the underlying mechanism is presumed to be anti-inflammatory in origin, other mechanisms, such as attenuation of prostaglandin E2-mediated inhibition of tumour-targeted T lymphocytes, have also been proposed. 2In the case of therapy, the potential of NSAIDs as adjuncts to conventional anti-cancer therapies remains largely unknown, a possible exception being the use of aspirin in the treatment of colorectal cancer associated with PIK3CA gene mutations. 49,50her potential pharmacological strategies include the use of inhibitors of MMP-9, although these have proved disappointing in phase II/III clinical trials in various types of malignancy, 51 and, perhaps the most promising strategy albeit unproven in the clinical setting, the use of pharmacological antagonists of CXCR2 43 and possibly dual antagonists of CXCR1/CXCR2.In addition to these, other categories of pharmacological agent which target the pro-inflammatory activities of neutrophils include 14/15-membered macrolide antibiotics and inhibitors of type 4 phosphodiesterase (PDE), the predominant PDE in human neutrophils.Unlike corticosteroids, which have limited efficacy in controlling neutrophilic inflammation, macrolides and PDE4 inhibitors possess a range of neutrophil-targeted anti-inflammatory activities which have recently been described in detail elsewhere. 52Although untested with respect to their adjunctive potential as anti-inflammatory agents in cancer chemotherapy, it is noteworthy that novel macrolides and PDE4 inhibitors are currently under investigation for their direct antitumour activities. 53,546][57] Although monoclonal antibodies which target neutrophil-mobilising cytokines such as TNF, IL-8 and, more recently, IL-17A 58,59 have been proposed as adjunctive anti-inflammatory strategies in the therapy of cancer, inhibitors of CXCR2 appear to be a superior option.

Conclusions
Although they are key players in innate host defence, human neutrophils are also inadvertent participants in the aetiology of inflammation-related cancers via the release of carcinogenic ROS and other mediators which contribute to tumour promotion and progression.Other types of cancer, which are not inflammatory in origin, also utilise inflammatory mechanisms to enhance their proliferative and invasive potential.The most significant of these mechanisms is the production of CXC/ELR + chemokines.These chemokines not only recruit pro-tumourigenic neutrophils, but are also pro-proliferative and pro-metastatic via their autocrine interactions with CXCR2 expressed on tumour cells.These important insights into inflammation-associated mechanisms of tumourigenesis have enabled identification of potential anti-inflammatory adjunctive strategies to complement conventional anti-cancer therapies.However, given the range of neutrophil-and tumour-derived inflammatory mediators which contribute to tumourigenesis, selective targeting of a single mediator is unlikely to be successful.Although unproven in the clinical setting, selective antagonists of CXCR2, which target both neutrophils and tumour cells, represent a possible exception, as do NSAIDs, particularly aspirin.Preventive strategies include routine implicated in the aetiologies of diffuse large B cell lymphoma and multiple myeloma *B cell activation factor (BAFF) Tumour promotion; also implicated in the aetiology of B cell malignancies *Oncostatin M (OSM) Tumour progression *Hepatocyte growth factor (HGF) Tumour progression*Recently reviewed by Tecchio et al.14 Figure 1:Proposed mechanism by which chronic inflammation leads to oxidative damage to the DNA of bystander tissue cells as a result of the sustained release of reactive oxygen species (ROS) from infiltrating neutrophils.Tumour initiation is followed by promotion and progression as a result of ongoing exposure to neutrophil-derived pro-proliferative and pro-angiogenic mediators.

Table 4 :
Examples of neutrophil-targeted CXC/ELR + chemokines experimental liver metastasis, have also been reported to mediate delivery of tumour cells to distant sites.
4 Volume 110 | Number 1/2 January/February 2014 South African Journal of Science http://www.sajs.co.za of Proposed mechanism by which tumour-derived CXC/ELR + chemokines exacerbate promotion and progression via the autocrine induction of proliferation and metastasis, as well as by recruitment of pro-tumourigenic neutrophils.
of low-dose NSAIDs, immunisation against cancer-causing viral pathogens, early aggressive antimicrobial chemotherapy to eradicate chronic inflammation caused by microbial pathogens, and avoidance of pro-inflammatory aspects of lifestyle such as cigarette smoking and excessive exposure to ultraviolet radiation.