Inhibiting exosomal MIC-A and MIC-B shedding of cancer cells to overcome immune escape: new insight of approved drugs

Abstract

Our knowledge of the role of innate immunity in protecting against cancers has expanded greatly in recent years. An early focus was on the adoptive transfer of natural killer (NK) cells and, although this approach has demonstrated promising results in many patients, a few limitations including immune escape of tumors from cytotoxic killing by NK cells have caused treatment failures. Downregulation of the expression of activating ligands on the surface of cancer cells and prevention of the activity of soluble factors are among the mechanisms employed by cancer cells to overcome NK-mediated immunity. It has become evident that a class of small membranous structures of endosomal origin known as exosomes play a key role in regulating the local tumor microenvironment. Here we hypothesize that exosome secretion by cancer cells, which is greater than that of normal cells, is an important escape mechanism employed by cancer cells. Interruption of exosome release by various inhibitory agents in combination with the adoptive transfer of NK cells may overcome, at least in part, the treatment failures that occur with adoptive NK cell transfer. In this regard, repositioning of approved drugs with previously shown effects on exosome release may be a good strategy to bypass the safety issues of newly identified agents and will also dramatically reduce the huge costs of drug approval process.

Introduction

Since the development of the immune surveillance theory by Sir Frank MacFarlane, different approaches, known as immunotherapy, have been undertaken to boost the ability of the immune system to eliminate tumors [1]. Cancer immunotherapy now constitutes a large proportion of therapeutic approaches in use to manage different types of malignancies and is predicted to account for 60% of all cancer treatments by the end of the next decade [1]. Adoptive transfer of immune compartments to cancer patients is commonly used with promising results seen [2]. For example, monoclonal antibodies such as rituximab (against CD20) and trastuzumab (against HER2-neu), have been extensively used as effective agents in the treatment of a wide range of cancers including hematological malignancies and breast cancer [3]. An alternative approach has utilized the adoptive transfer of specific immune cells such as CD8⁺ T cells with the aim of exploiting their cytotoxic effects against cancer cells [4].

The recognition of the key role played by innate immune cells in the response to cancers as well as the recent technical advances have led to the emergence of natural killer [5] cells as potential candidates for targeted cancer therapy [6, 7]. The feature of NK cells that makes them a good option for use in cancers is their ability to recognize cancer cells without any need for previous sensitization [6]. Despite the early promising results both experimentally and clinically from the adoptive transfer of NK cells to cancer patients, there are still several obstacles that limit the efficacy of NK-targeted immunotherapy [2]. The escape of cancers from cytotoxic killing by NK cells is one of the major challenges that limits the effectiveness of NK-based therapies [4].

Exosomes are nano-sized (30–100 nm) membranous structures shed by almost all cells including healthy and cancer cells [8]. Cancer cells release larger quantities of exosomes compared to their non-malignant counterparts [9]. Exosomes derived from cancer cells contribute to tumor growth and progression via multiple mechanisms including immunosuppression, angiogenesis promotion, reprogramming of the tumor microenvironment and induction of drug resistance [10]. Regarding the immunosuppressive role of tumor-derived exosomes (TEX), several mechanisms have been previously suggested; For instance, TEX from a wide range of cancers can interfere with the function of both helper and cytotoxic T cells and induce their apoptosis [11–13]. TEX can also promote the formation and function of regulatory T cells [14, 15], promote myeloid-derived suppressor cell (MDSC) differentiation [16], disturb the adhesive properties of leukocytes [17], and induce immune toleration [9].

Here, we hypothesize that the increased state of exosome release by cancer cells may contribute to the elimination of activating ligands on the surface of cancer cells that are required for the optimal activation of effective cytotoxic responses by NK cells. Reducing or reprogramming exosome release from cancer cells may improve the clinical efficacy of adoptive NK transfer by preventing one of their major escape mechanisms.

Hypothesis

Whether an NK cell is activated, remains inactive or is exhausted is determined by a finely-tuned balance between the signals received from its activating and inhibitory receptors [18]. Under normal physiological conditions, the inhibitory signals outweigh the activating stimuli giving rise to a resting NK cell [18]. NK cells receive some of their most important inhibitory signals via their receptors for HLA class I (killer immunoglobulin receptors-KIRs), a molecule which is normally present on many cell types labeling them as a healthy self-cell which should not be attacked; however, when these cell surface receptors are lacking, a characteristic of many stressed cells including transformed cells, the target cell lacks sufficient inhibitory signals to stop the NK cell from responding [7, 19]. MHC class I polypeptide-related sequence A and B (MICA and MICB) on cancer cells serve as ligands for the NK cell activating receptor NKG2D and their presence reinforces these NK cell stimulatory signals resulting in activation of the NK cell and its cytotoxic anti-tumor response [20].

Any successful NK-targeted cancer immunotherapy consists of two main compartments: 1) ex vivo expansion and activation of NK cells to achieve enough number of cells required for an effective transfer 2) increasing the cytotoxicity of NK cells against cancers both ex vivo and in vivo [21]. During the course of adoptive NK transfer to cancer patients, cancer cells may develop escape mechanisms which help them evade killing by NK cells [4]. Among the most important escape mechanisms employed by cancer cells to avoid NK-mediated cytotoxic killing is the downregulation of MICA and MICB [22].

In a recent study, the proteolytic shedding of these two activating ligands by a set of enzymes mostly belonging to the ADAM (a disintegrin and metalloproteinase) and MMP (matrix metalloproteinase) families, impaired NK-driven immunity. Antibodies specific for the proteolytic site of MICA and MICB which prevented proteolytic shedding of these stimulatory molecules strongly promoted NK-mediated anti-tumor immunity [23]. These findings show the importance of induced stimulatory signals and that impairment of their expression decreases the ability of NK cells to eliminate cancer cells. Hypoxia which is present in the tumor milieu also contributes to the decreased expression of MICA/B on tumor cells enabling cancer cells to evade NK-mediated killing [24].

Exosomes of endosomal origin are derived from late multi-vesicular bodies subsequent to the inward budding of plasma membrane at specific regions [10]. This pathway of exosome biogenesis enables the transfer of many cell surface molecules to the surface of exosomes [25]. Indeed, exosomes share many surface markers with that of the their cell of origin [26] including tetraspanins, integrins and other binding molecules [27]. For example, exosomes released by activated dendritic cells have the co-stimulatory molecule CD86 on their surface [27].

Based on this evidence, we propose that the stimulatory ligands of NKG2D i.e. MICA and MICB are shed from their cell of origin via exosome release. Patients with cervical cancer have higher serum levels of MICA in comparison to normal individuals indicating that cells of these patients have increased their shedding of MICA possibly as an escape mechanism against NK immunity [28]. Importantly, serum exosomes of AML patients expressed greater levels of MICA/B than those of healthy adults [29]. In addition, exosomes released from epithelial ovarian cancer cells express the ligands for both NKG2D and for DNAM-1. The latter being another activating receptor present on NK cells [30]. Moreover, these ligand-bearing exosomes downregulated the expression of their receptor counterparts which can further enhance the escape of cancer cells from NK-induced cell death [30].

These findings suggest that cancer cells may evade NK-driven anti-tumor immunity via exosome-mediated shedding of stimulatory ligands including MICA/B that are expressed on their cell surface. This would indicate that agents that reduce exosome release may overcome immune avoidance [10] (Fig. 1).

Fig. 1.

Schematic illustration of adoptive NK cell transfer to cancer patients and the probable outcomes of parallel use of exosome inhibitors. Without the use of exosome inhibitors, cancer cells evade NK cell cytotoxicity via shedding of their activating NKG2D ligands (MICA/B). Moreover, tumor derived exosomes (TEX) induce diverse effects to the advantage of tumors including reducing the expression of NKG2D on the surface of NK cells. Once exosome inhibitor agents are used in parallel to NK transfer, these effects are reversed and the exosome mediated-shedding of activating ligands on the surface of tumor cells becomes interrupted allowing for a better activation and anti-tumor cytotoxic response of NK cells

Recommended agents

A number of studies have used different pharmacological agents to interfere with exosome release from cancer cells. The goals of these studies range from increasing the sensitivity of cancer cells to chemotherapy to the inhibition of tumor-promoting mechanisms driven by tumor-derived exosomes [10]. Drug repositioning is recently surging as a useful strategy in hopes that introducing new effects for previously approved drugs could bypass the difficulties of the regulations required for a new drug making it more cost-benefit as well. An example of these drugs that might be used herein is sulfisoxazole, which is an antibiotic that has been shown to inhibit the secretion of exosomes by targeting the endothelin receptor A in breast cancer cells [31]. Cell membrane stabilizers such as ketotifen have been successfully used to increase the chemosensitivity of three tumor cell lines including MCF7, HeLa and BT549 by suppressing exosome release [32]. In addition, manumycin A, an antibiotic agent that inhibits Ras/Raf/MEK/ERK1/2 signaling, inhibited exosome release from prostate cancer cells [32]. Calcium chelators such as EGTA and other pharmacological agents like imipramine and chlorpromazine have been also used to successfully inhibit exosome release by cancer cells [10] which makes them potential options for combination with adoptive NK cell transfer. Lipid rafts are not only involved in the process of EX biogenesis, but also are known to be among the key coordinators of cell membrane composition; a finding that gives rise to the idea that agents with the ability to affect lipid raft formation and composition such as omega 3 fatty acids are also potential candidates for induction of changes in EX release and content. On such a basis, long chain poly unsaturated fatty acids (PUFAs) such as docosahexaenoic acid (DHA) and Eicosapentaenoic acids (EPA), which have been shown to possess lipid raft-modifying effects [33], may also be desirable candidates to change exosome release and function [34].

Further supporting evidence

Inhibitors of exosome release may not only attenuate the shedding of MICA/B by cancer cells but may also hinder the tumor-promoting effects of tumor-derived exosomes particularly those involved in the suppression of NK activity [10]. This would enhance the efficacy and overall outcome of adoptive NK transfer to cancer patients. Several studies have shown that exosomes released from cancer cells can suppress NK-mediated immunity by a variety of different mechanisms [35]. The presence of MICA/B on the surface of tumor-derived exosomes downregulates NKG2D expression on NK cells [30, 36]. Tumor-derived exosomes also increase the release of TGF-β from regulatory T cells which, in turn, decreases the expression of NKG2D on NK cells. TGF-β also directly decreases the surface expression of MICA/B on malignant cells [37, 38]. Tumor-derived exosomes also inhibit IL-2-mediated activation of NK cells as well as key functions including the release of the cytolytic protein perforin [39].

Obstacles and limitations

The dual role of TEX in mediating the anti-tumor immune response makes it difficult to design exosome-based treatment strategies. It has been shown that TEX can exert beneficial effects which may boost anti-tumor immunity [10]. These include chemo-sensitization of cancer stem cells that resist conventional chemotherapy [5] and serving as a rich source of tumor-associated antigens for cross antigen presentation by dendritic cells [40]. On this basis, there is the probability that interfering with TEX secretion from cancer cells may diminish some of their known beneficial effects and consequently end in detrimental outcomes. However, to the best of our knowledge there are no reports linking these possible undesired effects with any of the known or recently identified exosome inhibitors. On the contrary, recent studies on the use of exosome release inhibitors support their beneficial roles rather than any detrimental effects. For instance, it has been recently shown that inhibiting exosome and microvesicle release from cancer cells can sensitize them to chemotherapy and reduce their growth in vivo [41]. In line with this, the inhibition of microvesicle secretion from cancer cells using calpeptin, a calpain inhibitor, resulted in a 100-fold reduction in the required dose of fluorouracil and docetaxel to induce the expected regression in tumor size [42]. Another study showed that knockdown of Rab27a, a regulator of exosome release [34], could diminish the chemotactic effects of exosomes derived from cancer cells [43]. All this evidence, along with many other findings, support the beneficial effects of using exosome release inhibitors as parts of therapeutic strategies used in cancer treatment. Another limitation is that any attempt to specifically target the release of exosomes from cancer cells may result in effects on other cell types including immune cells. Since no differences in exosome biogenesis pathways between cancer and healthy cells have been described that would increase specificity, cell selectivity needs to be achieved by alternative approaches which has proved difficult in the past. Among the few studies focusing on selective inhibition of exosome release from cancer cells is a high-throughput study which identified several potent inhibitors of exosomes release from investigated prostate cancer cells including neticonazole, ketoconazole, tipifarnib, climbazole, and triademenol. These compounds were found to induce their effect either via ESCRT-dependent or ESCRT-independent pathways. It was finally concluded that repositioning of drugs can be of potential utility for selective inhibition of exosome release form cancer cells in cancer therapy [44]. Another approach which to our knowledge has not been tested so far, is the specific targeting of exosome release from cancer cells through conjugation of non-selective inhibitors with monoclonal antibodies against cancer surface markers. This approach is similar to what is done for specific targeting of immunotoxins to cancer cells, which remains to be tested.

Use in combination therapy

It is evident that the most effective therapeutic approaches for cancer treatment involve combination therapies [4]. Combining adoptive NK cell transfer with other strategies of immunotherapy is not an exception. The use of IL-15 in combination with NK therapy results in much better outcomes [45]. The use of monoclonal antibodies in conjunction with NK transfer has also shown promise [46]. Exosome release inhibitors should be used in combination with passive transfer of activated NK cells to achieve better outcomes. For instance, TEX can diminish the efficacy of monoclonal antibody therapy with the anti-HER2 mAb trastuzumab. Exosomes released from HER2⁺ cancer cells also express HER2 on their surfaces which acts as a decoy molecule to trap the administered trastuzumab in breast cancer patients limiting its efficacy [47].

Conclusion

Cancer cells not only produce higher amounts of exosomes in comparison to normal cells but also produce  exosomes which contain NK cell activating ligands that are normally expressed on the surface of cancer cells. There are also greater levels of soluble MICA and MICB in the sera of cancer-bearing patients. This indicates that exosome secretion by cancer cells may be an escape mechanism employed to evade NK-mediated cytotoxic killing. In addition to reducing cancer cell surface expression of MICA or MICB, TEX can induce a significant reduction of the activating receptor NKG2D on the surface of NK cells, which further dampens NK-mediated cytotoxicity. These mechanisms may play an important role in the failure of adoptive NK transfer in some cancer patients. The use of pharmacological agents to interfere with different stages of exosome biogenesis and release such as ketotifen, a membrane stabilizer, may be an effective approach to reverse these escape mechanisms and to increase the efficacy of NK transfer. In vivo animal studies are warranted to optimize the use of combined TEX inhibitor and adoptive transfer regimes as a prelude to their use in future clinical trials.

Compliance with ethical standards

The authors have no conflicts of interest to declare. This research is based on a review of the literature and involves no animal/human experiments.