Mitochondria and Antioxidants: The Active Players in Islet Oxidative Stress

the protective effects of polyphenols in vitro on islets or cultured insulin-producing cells against oxidative challengers. It was also shown that polyphenols from olive leaves protected INS-1 cells from H 2 O 2 toxicity Cells pre-incubated with whole leaf extract or individual polyphenol compound oleuropein, followed by peroxide treatment, showed a lower percentage of necrotic and apoptotic death compared to untreated controls. Polyphenols stimulated activity of catalase, which resulted in a lower level of cellular ROS, along with improvement in insulin production. In this study it was suggested that polyphenols act through a redox-modulating mechanism rather than through direct free radical scavenging. Supplementation of cultured media with polyphenols from green tea increased recovery rates of isolated human and nonhuman primate islets. Polyphenols from tea extracts preserved islets by increasing the level of anti-apoptotic Bcl-2 and decreasing level of pro-apoptotic BAX al. 2004). Chemical biology utilizes chemical principles to modulate systems to either investigate the underlying biology or create new function. Over recent years, chemical biology has received particular attention of many scientists in the life sciences from botany to medicine. This book contains an overview focusing on the research area of protein purification, enzymology, vitamins, antioxidants, biotransformation, gene delivery, signaling, regulation and organization. Particular emphasis is devoted to both theoretical and experimental aspects. The textbook is written by international scientists with expertise in synthetic chemistry, protein biochemistry, enzymology, molecular biology, drug discovery and genetics many of which are active chemical, biochemical and biomedical research. The textbook is expected to enhance the knowledge of scientists in the complexities of chemical and biological approaches and stimulate both professionals and students to dedicate part of their future research in understanding relevant mechanisms and applications of chemical biology.


Introduction
Diabetes mellitus (DM) of type 1 (T1D) and type 2 (T2D) are characterized by persistently high glucose (HG) blood levels known as hyperglycemia. The preponderance of evidence points to a significant role of oxidative stress in the development of complications in patients with DM. In particular, the pathogenic increase in reactive oxygen (ROS) and nitrogen species (RNS) as well as accumulation of oxidation and nitration products has been well documented in cases of diabetes. ROS and RNS affect all types of biological molecules: they cause oxidation of membrane lipids, modification of protein amino groups as well as deoxynucleotites. Insulin producing beta-cells, which are part of pancreatic islets of Langerhans, perform the energetically demanding function of sensing blood glucose and releasing insulin to sustain metabolic homeostasis. They are highly specialized endocrinal cells with a complex system of signal transduction and insulin producing capacity. In pathology of diabetes beta-cells are the ones which are most susceptible to oxidative stress.

Sources of oxidative stress in islets
In islets, reactive species can originate from several sources. NAD(P)H-oxidase located in the plasma membrane produces molecule of O 2-. Several isoforms of this enzyme were found in islets (Uchizono et al. 2006;Newsholme et al. 2007;Newsholme et al. 2009) and they are considered as substantial producers of ROS. The enzyme can be activated by exposure to fatty acids and in normal conditions it is believed to participate in glucosestimulated insulin secretion (GSIS) Santos et al. 2011). Islets also possess both types of nitric oxide synthases (NOS): constitutive cNOS (Nakada et al. 2003) and inducible (iNOS) (Darville and Eizirik 1998;Kutlu et al. 2003). Nitric oxide produced by cNOS is part of normal beta-cell physiology, while activation of iNOS is associated with beta-cell destruction, in particular via cytokines produced by immune cells.
There is plentiful evidence that hyperglycemic conditions cause rise in ROS and RNS in beta-cells reviewed in: (Newsholme et al. 2007;Acharya and Ghaskadbi 2010). It was shown that high glucose triggers generation of ROS in rodent and human islets as well as in insulinoma cell lines (Tanaka et al. 1999;Tanaka et al. 2002;Bindokas et al. 2003;Robertson et al. 2003). A definitive role for mitochondria in glucose-induced ROS signal was proposed in a number of papers (Maechler and Wollheim 2001;Brownlee 2003;Fridlyand and

Islet mitochondria as a source of reactive oxygen species
The main role of mitochondria in a cell is the production of ATP molecules for cellular energetic needs using for that energy of metabolite oxidation. ATP production by mitochondria is central for glucose sensing and insulin release in beta-cells, the fact that directs attention to mitochondria in diabetes studies. In the process of metabolite oxidation, mitochondria are dealing with the transfer of electrons along respiratory Complexes I through IV. Eventually four electrons are combined with four protons, H + , and one molecule of oxygen, producing two molecules of water; the reaction takes place within Complex IV (for reference see (Nicholls and Ferguson 2002). However, as a side reaction, a low portion of unpaired electrons leaks from the respiratory complexes and interacts with molecular oxygen producing a molecule of superoxide anion radical, a form of ROS. Currently, it is believed that the sites of ROS production in the mitochondrial respiratory chain are Complex I and Complex III (Turrens 2003;Rigoulet, Yoboue, and Devin 2011).
For proper mitochondrial functioning with low levels of ROS production, the availability of end-point electron acceptor, molecular oxygen, is absolutely necessary. Temporal hypoxia, which causes a halt in electron flow followed by reoxygenation and resuming of electron flow, results in an increase of mitochondrial ROS (Selivanov et al. 2009). This phenomenon is related to a widely known phenomenon of ischemia-reperfusion injury and is a critical factor in the process of islet isolation from donor pancreata for transplantation purposes, as donor tissues inevitably become hypoxic when blood circulation stops. Presence of antioxidants in preservation and isolation media detoxifies mitochondria-derived ROS and substantially improves viability of islets and their potency to normalize blood glucose in recipient diabetic animals (Bottino et al. 2002;Bottino et al. 2004;Sklavos et al. 2010).
Manipulation of electron flow in the respiratory chain by specific inhibitors of mitochondrial respiratory Complexes can either increase or decrease ROS generation. In normal conditions, with low levels of glucose and low base level of ROS, inhibitors of Complexes I and III, rotenone and antimycin, respectively, increase ROS production in rat islets (Armann et al. 2007;Leloup et al. 2009). This is consistent with data obtained on isolated mitochondria in vitro (Votyakova and Reynolds 2001;Starkov, Polster, and Fiskum 2002;Starkov and Fiskum 2003;Rigoulet, Yoboue, and Devin 2011).
An important question is whether mitochondrial inhibitors are capable of down regulating ROS signals originating from the respiratory chain under conditions of hyperglycemia and hyperlipidemia. The answer depends on the type of chemical agent and the locus it binds in the respiratory chain. Oxidation of succinate in Complex II results in a high ROS generation because this prosess initiates forward electron flow to Complex III as well as reverse electron flow upstream to Complex I (Selivanov et al. 2011). Thus, inhibition of succinate oxidation would lead to a decrease in overall ROS. This effect was observed in a work by Sakai and collaborators (Sakai et al. 2003) where an inhibitor of Complex II thenoyltrifluoroacetone (TTFA) decreased glucose-stimulated ROS in human islets and MIN-6 cells. The inhibitor of Complex III antimycin A blocks Q-cycle in a way that increases free radical forms of respiratory chain components (Votyakova and Reynolds 2001;Starkov, Polster, and Fiskum 2002;Starkov and Fiskum 2003). Leloup and co-authors observed a similar effect of antimycin A on rat islets (Leloup et al. 2009).
Complex I, or NADH-oxidoreductase, is the largest and the most sophisticated segment in the respiratory chain; its architecture, functioning and mechanisms of ROS generation has been extensively studied (Magnitsky et al. 2002;Vinogradov 2008). In mammals it consists of 47 subunits with a number of redox centers to transfer electrons. Among them, there are flavine mononucleotide at the entrance of the Complex, which binds NADH, several sulfuriron clusters and several Coenzyme Q binding sites in the middle of the Complex. According to the current consensus, all inhibitors, which bind to subunits located in the middle of Complex I and block electron flow within the complex, such as rotenone, piericidin and others, increase ROS (Grivennikova and Vinogradov 2006). The only known inhibitor that decreases ROS generation in Complex I is diphenylene iodonium, which binds at the very entrance at the flavin mononucleotide (FMN) site (Liu, Fiskum, and Schubert 2002). This compound, though, is unspecific and also inhibits NAH(P)H oxidase of the plasma membrane. In other words, if an inhibitor blocks the very entry of Complex I at the site of FMN and prevents the access of electrons into Complex I at all, it would prevent ROS generation. If an inhibitor blocks electron flow somewhere in the middle of Complex I and allows redox centers upstream of the block to be over-reduced, it would result in an increase of ROS (Genova et al. 2003). In accordance with this notion, inhibitory effects of diphenylene iodonium on ROS induced by hyperglycemia in MIN-6 cells were reported by (Tsubouchi et al. 2005), though the researchers attributed this fact solely to the plasma membrane NAD(P)H oxidase inhibition. Regarding the other Complex I inhibitors, there is some controversy in the literature. The same authors reported no effect of rotenone, the effect could be expected if ROS was already increased, while earlier work by Sakai and collaborators reported that rotenone twofold decreased hyperglycemia-induced ROS in human islets and MIN-6 cells (Sakai et al. 2003).
There is a way to modulate mitochondrial ROS generation without interference into the activity of respiratory Complexes. In the cell this function belongs to a special group of proteins located in the inner membrane of mitochondria, called uncoupling proteins (USPs), which in a highly controlled manner modulate membrane potential, basically working as proton conductors (Ricquier and Bouillaud 2000). Islets express an UCP2 isoform of uncoupling protein (Gimeno et al. 1997), and in a number of publications it was shown that overexpression of UCP2 downregulated the levels of ROS (Pi et al. 2009;Affourtit, Jastroch, and Brand 2011).
The same effect can be achieved by using chemical compounds called uncouplers, which, while different in structure, have two properties in common: they can penetrate into mitochondrial membrane and are capable to easily accept or dissociate H + ion. Uncouplers decrease membrane potential by transporting protons into mitochondrial matrix and, in principle, their function is similar to that of UCPs, though they are more powerful modulators (see (Nicholls and Ferguson 2002) for detailed mechanism). Uncouplers were shown to decrease ROS production in glucose-stimulated human and rat islets as well as in MIN-6 cell line (Sakai et al. 2003;Leloup et al. 2009). This is consistent with the data observed on isolated mitochondria in vitro (Votyakova and Reynolds 2001;Selivanov et al. 2008).
Overall, the amount of data on the mechanism of mitochondrial ROS generation in beta-cell is limited and this issue is underexplored, mainly because islets are precious and less available for research compared with other more abundant tissues, like liver, muscle, heart and even brain. Mostly intriguing is the fact that, despite extensive evidence of increased level of free radicals in conditions of high glucose or high lipids, there is still no satisfactory molecular mechanism explaining exactly how high levels of cellular energy metabolites cause beta-cell mitochondria to produce more free radicals.

Antioxidant capacity of insulin producing beta-cells
Antiradical defense systems in the cell consist of a gamut of small antioxidant molecules and enzymes capable of interaction with reactive oxygen and nitrogen species. Endogenous antioxidant molecules are vitamins A, C and E, sulfur-containing compounds like amino acid cysteine and tripeptide glutathione (GSH -reduced form, GSSG -oxidized form). Coenzyme Q may also act as an antioxidant in particular conditions. These compounds possess different efficiency in scavenging harmful oxidants, glutathione being the most potent. The main players in enzymatic antiradical protection are superoxide dismutases (SOD), catalase and glutathione peroxidases (GPx). Within the cell two isoforms of SOD are found: Zn,Cu-SOD (SOD1) which is located in cytoplasm and Mn-SOD (SOD2) located in mitochondrial matrix. They are the first line of defense as they convert superoxide anion radical into O 2 and H 2 O 2 by dismutation. The hydrogen peroxide next can be converted into H 2 O and O 2 by catalase or, to H 2 O, by GPx; the latter will use reduced glutathione molecule as a substrate. Glutathione is a recyclable molecule: its oxidized form is reduced back by glutathione reductase. Mitochondria have both enzymes (about of 10% of total cellular activity) which, working in concert, effectively detoxify peroxides and recycle GSH (for reference see (Halliwell 2001)). The proteins UCPs, which downregulate ROS generation in mitochondria, can also be formally added to antioxidant proteins.
Glucose sensor function and production of insulin on demand are the two dominating functions in beta-cell physiology and, apparently, this comes at a cost of downregulating some other functions. High susceptibility of islets to oxidative insults was well established long ago, and this feature was utilized to specifically target these cells, thus, creating animal models of diabetes. Compared to cells from other tissues, islet cells have profoundly lower activity of enzymes involved into antiradical defense. As early as in 1979 Grankvist and co-authors shown that alloxan effectively destroys rat islet cells through ROS-mediated mechanisms (Grankvist et al. 1979). Two years later they reported that beta-cell super-sensitivity to oxidative agent is due to a deficiency of anti-radical defense capacity (Grankvist, Marklund, and Taljedal 1981). The activity of the main antioxidant enzymes was found to be about 30% for both types of SODs and only in single percentage range for catalase and glutathione peroxidase (1.2% and 1.8%, respectively), as compared to liver. This data was confirmed by estimating the levels of mRNA of the respective genes by Lenzen and co-workers (Lenzen, Drinkgern, and Tiedge 1996). It should be noted, though, that a more appropriate comparison of islet enzymes' activities would be with that of tissues with similar functions and/or intensity of metabolism. As Table 1 shows, when compared with hypophysis or brain tissues, islets' activities of antioxidant enzymes were only 50% or lower.  Table 1. The levels of anti-oxidant gene expression in islets compared to brain and pituitary gland. *Data adopted from Lenzen & al. (1996). **In (Tiedge et al. 1997) it was determined to be around 5%.
Expression of anti-oxidant enzymes in islets may vary within the same species reflecting genome variations. Zraika and co-authors have found that mRNA levels of Mn-SOD in islets of a diabetes-prone DBA/2 mice were twofold higher than in islets of C57BL/6 mice (Zraika et al. 2006), while the level of catalase was the same. This means that at the same level of superoxide anion radical beta-cells of DBA/2 mice would produce H 2 O 2 twice as fast compared to that of C57BL/6 mice, and consequently, having the same level of catalase, DBA/2 islets would deal with higher levels of peroxide.
Among species the differences in activities of antioxidant enzymes can also vary substantially. It was shown that human islets have more active catalase and SOD than rodent ones and, consequently, are more resistant to oxidative stress (Welsh et al. 1995).
The expression of antiradical enzymes can be changed in diabetes conditions. In islets of Goto-Kakizaki/Paris rats, a model of T2D, expression of the whole spectrum of antiradical defense genes is increased along with an increased level of reduced glutathione compared with normal healthy animals (Lacraz et al. 2010). In patients with T2D expression of Cu,Zn-SOD was found reduced (Sakuraba et al. 2002). Diabetes is a complex and dynamic disease, in which epigenetic and environmental factors can differently affect the expression of antioxidant enzymes at particular time points of its development. Thus, it is often difficult to compare data on diabetes-related oxidative stress in short-lived rodents to long-lived humans.
It is important to note that beta-cells possess a substantial activity of enzymes dealing with superoxide anion radicals, while the activity of the enzymes decomposing H 2 O 2 is very low, especially that of catalase. This feature is in line with increasing evidence that suggests a signaling role for H 2 O 2 molecule in the process of insulin secretion (Pi et al. 2009;Affourtit, Jastroch, and Brand 2011).

Application of chemical compounds with antioxidative properties as a strategy to offset oxidative stress
Low level of antioxidant defense in beta-cells suggested that supplementation of antioxidants could be beneficial. A number of chemical compounds with antioxidant properties were employed to prevent or counteract oxidative stress in islets. The studies were conducted on diabetic animals and on isolated islets. Both preventive and curing actions of antioxidants were studied; the antioxidative agents were administered either prior to induction of diabetes of oxidation stress, or in the course of developing processes.
Historically, the first to study were vitamins A, C and E which are naturally present in the body and possess the ability to scavenge free radicals. Unfortunately they offered very limited or no protection (Kaneto et al. 2001). Hence, a broad spectrum of chemical compounds was tested in search for agents capable to counteract oxidative burden experiencing by islets. They differ in chemical nature and by the mechanism through which they protect from oxidative stress. The most important aspect in search of these compounds is to pay attention to their potential side effects. We will focus on the most prominent ones.  (Czubkowski, Socha, and Pawlowska 2011) and diabetes (Kaneto et al. 2001). There are several mechanisms by which NAC can modulate oxidative stress: (i) it is a precursor in the synthesis of glutathione, an important component of the cellular antiradical defense system in the cell; (ii) NAC, as a thiol-containing compound, is able to directly reduce free radicals as well as S-S bonds in proteins, thus modulating redox signaling (Parasassi et al. 2010) It was reported that NAC is protective or partially protective in animal models of diabetes. It reduces levels of oxidative stress markers and preserves beta-cell mass and function in STZ-treated hamsters (Takatori et al. 2004). Intravenous co-infusion of NAC with high glucose into Wistar rats offered only partial protection, as it quenched ROS, but did not restore beta-cell functions ). However, the same group reported that NAC, co-infused with free fatty acids, not only decreased beta-cells ROS caused by prolonged exposure to fatty acids, but preserved their insulin and C-peptide responses to hyperglycemic clamps ). These two studies show that effectiveness of particular antioxidant molecules depends on the type of oxidative stimuli and the metabolic pathways that are intervened. A study by Kaneto and co-workers also demonstrated that NAC can be protective in T2D-like metabolic deregulations. Using the db/db diabetic mouse model, the researchers found that NAC improved glucosestimulated insulin response in these animals and, on a molecular level, NAC increased expression of pancreatic and duodenal homeobox factor-1 (PDX-1) in islets, a beta-cellspecific transcription factor (Kaneto et al. 2001).

Antioxidant properties of N-acetyl
NAC protects against oxidative stress stimuli in vitro. LDL oxidation level, which is relevant to pathology of T2D, significantly changes the expression of genes involved in the production and secretion of insulin and in cell survival mechanisms. This effect was offset by NAC (Favre et al. 2011). NAC counteracts the damaging effect of human amylin (hA), a small fibrillogenic protein, which accumulates in beta-cells in most subjects with T2D (Konarkowska et al. 2005). It is believed that NAC offers protection as a reagent capable of reducing protein SH groups, rather than a general ROS scavenger.
NAC is also protective against oxidative damage caused by direct short exposure of islets to H 2 O 2 and free fatty acids, but is inefficient or partially efficient against long exposure islets to high glucose, or cytokines (Khaldi et al. 2006;Oprescu et al. 2007;Michalska et al. 2010).
In some cases NAC, while decreasing ROS, did not restore beta-cell functions completely (Wang et al. 2004;Tang et al. 2007).
ALA is an essential co-factor in several mitochondrial oxidative complexes; the most important of these are pyruvate dehydrogenase (PDH) complex, 2-oxoglutarate dehydrogenase (OGDH) complex, and the complex for oxidation of branched chain amino acids (BCDH) (Nelson 2005).
Administration of alpha-lipoic acid to non-obese diabetic (NOD) mice decreased incidence of diabetes induced by cyclophosphamide from 60% to 30%. It also reduced severe intraislet infiltration and increases the percentage of islets with mild per-insular and periductular infiltrates (from 8.4 to 29.6 and 25.9%, respectively, P < 0.01) (Faust et al. 1994). The authors concluded that the anti-inflammatory action of lipoic acid may be due to its ability to scavenge oxygen radicals and to suppress nitric oxide production.
In the alloxan-induced diabetic mouse ALA lowered blood glucose, increased insulin release and prevented loss of beta cells and their dysfunction (Zhang et al. 2009). ALA prevented development of diabetes mellitus in obese Otsuka Long-Evans Tokushima Fatty (OLETF) rats, a T2D animal model. It diminished glycosuria, reduced body weight and protected pancreatic beta-cells from destruction. ALA also reduced triglyceride accumulation in skeletal muscle and pancreatic islets (Song et al. 2005).
In vitro studies provided a deeper insight into the mechanism of ALA action on cellular and organellar levels. ALA counteracted oxidative stress induced by proinflamatory cytokines IL-1 , IL-6 and IFN-by preventing NF-κB activation (Zhang et al. 2009) and restoring insulin secretion (Schroeder et al. 2005). When a direct oxidative insult was applied in a form of H 2 O 2 (Lee, Kwon, et al. 2009) or xantine-xantine oxidase (Burkart et al. 1993), pretreatment with ALA decreased cellular ROS and c-JNK activation, stabilized mitochondrial membrane potential and induced Akt phosphorylation, altogether offering protection to beta-cells from oxidative stress. In MIN6 cells and rat islets, this compound also offsets the deleterious actions of free fatty acids, which feature an in vitro model for conditions of T2D (Shen et al. 2008). In particular, it decreased levels of ROS, restored mitochondrial membrane potential, glucose-induced ATP and glucose stimulated insulin secretion.
It is worth noting that ALA, like many other redox active chemicals (Skulachev et al. 2009), has an optimum concentration for anti-radical actions, and, consequently, for protective activity. Optimum protective concentration for isolated mouse islets against oxidative stress induced by cytokine IL-6 was found to be 10 -9 M; lower and higher concentrations were not effective (Schroeder et al. 2005). In the case of chemically-induced oxidative stress in INS-1 cells or isolated islets, ALA prevented apoptotic cell death in 150-300 μM concentrations, while higher concentrations caused apoptosis. Discrepancies in the effective concentration ranges reported in these studies may lie in different nature of oxidative stress stimuli and in different protective mechanisms by which ALA acted in each case. Concentration of 2 mM, which is not physiologically relevant, impaired functions of isolated rat islets and MIN6 cells (Targonsky et al. 2006).
The nature of protective effects exhibited by ALA and NAC suggests that these compounds can act by multiple mechanisms, working as direct scavengers of free radicals, or by controlling of the thiol-disulfide level of reduction in signaling protein molecules (Pietta 2000;Pandey and Rizvi 2009;Parasassi et al. 2010). The antioxidant properties of ALA can be boosted by mitochondrial thioreductase (Trx), which restores its reduced form (Packer, Witt, and Tritschler 1995).

Coenzyme Q (CoQ)
Coenzyme Q (CoQ), a quinone, is an essential component of mitochondrial respiratory complexes I and III. In general, quinones can undergo two reversible one-electron redox processes, converting them into semiquinones and hydroquinones. High stability of semiquinone radicals renders the redox reversibility and is the basis for antioxidant properties of the quinone-related systems; both quinones and hydroquinones can serve as radical-protecting agents.
In cells, exogenous CoQ can exhibit either antioxidant or pro-oxidant properties, depending on conditions. As it was shown (Schroeder et al. 2005), a very low concentration of 10 -12 M CoQ10 restored insulin production by mouse islets which were impaired by exposure to cytokine IL-1 . However, higher concentrations showed no effect or were even harmful.
Overdoses of CoQ (50-200μM), although able to stimulate insulin release, were toxic to human islets and INS-1 cells. At these concentrations, CoQ, being a strong electrophile, covalently binds to E2 components of pyruvate dehydrogenase and -ketoglutarate dehydrogenase complexes in mitochondria causing a substantial inhibition of the complexes and eventually triggering cell apoptosis (MacDonald et al. 2004). This circumstance put into a question the possibility to utilize CoQ10 as antioxidant to counteract diabetes-related oxidative burden as it would be difficult to find the right dose, while overdose would greatly outweigh potential benefits.
Some synthetic and plant-derived compounds may possess profound antioxidant properties, actively reacting with free radicals and ROS. These compounds belong to different chemical types; the most known of them are stable nitroxyl free radicals, metalloporphyrins, and plant polyphenols.
Superoxide anion radical is the first product resulting from the passing of unpaired electrons to molecular oxygen in the chain of chemical reactions causing oxidative stress. Thus, its detoxification would seem strategically advantageous. This notion prompted the introduction of a number of compounds which belong to different classes, but can all act as superoxide scavengers as they are able to effectively disproportionate superoxide to dioxygen and hydrogen peroxide. These compounds have a common name of SOD-mimetics.
Application of TEMPOL in diabetic animal models and in vitro on isolated islets was found to be protective against a variety of oxidative stress stimuli. Intravenous co-infusion of TEMPOL with high glucose or with free fatty acids into rats prevented islet dysfunction caused by hyperglycemia  or by hyperlipidemia ). In Zuker rats, an animal model for T2D that features hyperglycemia, hyperinsulinemia as well as renal oxidative stress and high blood pressure, TEMPOL administered in drinking water reduced blood glucose, insulin secretion, renal oxidative stress and blood pressure which implies normalization of islet function (Banday et al. 2005).

Metalloporphyrins
Metalloporphyrins represent another well-known class of organic redox-active compounds. Manganese metalloporphyrins feature high chemical stability in different oxidation states and coordination of substrates at the central metal atom (Patel and Day 1999). Manganese ion can coordinate additional water or H 2 O 2 in its axial sites, thus, performing its catalytic action. The stability of the porphyrin ring enables a variety of reversible catalytic metalcentered redox processes. Depending on substitutents, Mn-porphyrin compounds vary in their overall electric charge, redox potential and lipophic-hydrophylic properties. Eventually these physico-chemical properties translate into different catalytic activities and different abilities to penetrate into cells and cellular compartments. These compounds are capable of catalyzing not only the reaction of O 2 -˙ dismutation, but also the reduction or disproportionation of other reactive oxygen and nitrogen species, like H 2 O 2 , HO˙, NO and ONOO - (Patel and Day 1999).
Manganese complexes of porphyrins were employed in studies of a number of biological models of oxidative stress (Batinic-Haberle et al. 2011). In an animal model of STZ-diabetic rats, Mn-porphyrin MnTM-2-PyP 5+ , administered after the dose of STZ, counteracted the oxidative stress as judged by decreased levels of lipid peroxidation in blood plasma and erythrocytes (MDA products). Though it did not normalize blood glucose, it still decreased mortality of STZ-treated animals and increased their life span (Benov and Batinic-Haberle 2005). The data suggested that application of this antioxidant after a major STZ-induced oxidative stress did not restore islet function, but rather ameliorated following hyperglycemia-related complications. In the other experimental settings, the animals were preconditioned with Mn-porphyrin (MnTE-2PyP 4+ , FBC-007), possessing a higher catalytic activity and a better ability to penetrate into cells. The animals were also given regular injections after STZ administration. Such treatment prevented development of diabetes in mice as monitored up to 120 days (Sklavos et al. 2010). It appears, therefore, that a preventive measure may play an important role in successful diabetes treatment.
In vitro, Mn-porphyrins showed a protective action also on the models of cultured isolated human and rodent islets as well as on insulin producing cell lines (INS-1 cells) subjected to oxidative insult. Bottino and co-workers demonstrated that the presence of MnTE-2PyP 5+ (AEOL10113) and MnTDE-2-ImP 5+ (AEOL10150) as supplement to media during islet isolation resulted in up to three-fold increase of the viable mass of human islets, the fact of vital importance for transplantation medicine (Bottino et al. 2002). In a consequent publication a cascade of stressful events triggered by the procedure of islet isolation from the whole pancreas was studied in detail (Bottino et al. 2004). The islet isolation is a lengthy procedure that causes, apart of hypoxia-reoxygenation, mechanical and chemical stress to the cells. This results in activation of stress-related signals NF-κB and poly(ADP-ribose) polymerase (PARP) as well as increased levels of proinflamatory cytokines (Bottino et al. 2004). Mn-porphyrin effectively decreased NF-κB binding to DNA, PARP activation and release of cytokines and chemokines in islet cells, eventually resulting in higher survival and better insulin release (Bottino et al. 2004). Compound MnTE-2PyP 5+ protected human islets from STZ-induced cell death and ensured better islet function after transplantation into immunodeficient diabetic mice. This holds true for transplantation of islets from the same mice strain (syngeneic), different strain (allogeneic) or in case of transplantation of human islets into mice (xenogeneic) (Sklavos et al. 2010).
Due to the complexity of metabolic and signal pathways, which vary in different cell types and physiological conditions, a possible protective effect of a Mn-porphyrin may depend on the cell type and the type of oxidative insult. Thus, it was reported that MnTMPyP preserved INS-1 cell viability and insulin secretion upon exposure to both NO and O 2 -, while human islets were protected by this compound only from NO, but not from superoxide anion radical (Moriscot et al. 2007).

Polyphenolic compounds
Polyphenolic compounds, naturally occurring in plants, were extensively studied in recent years as potential remedies against many diseases like cancer, cardio-vascular disorders and diabetes. They can be found in numerous dietary and medicinal plants and comprise an important part of the human diet, though they are generally viewed as nonnutrients.
Particularly rich in polyphenols are red grapes, berries, tea leaves and some spices. These polyphenols comprise several types of compounds, i.e. phenolic acids, stilbenes, lignans and flavonoids, the latter are oxygen heterocycles (chromenes, for structures and more detailed classification see review by (Pandey and Rizvi 2009)). Flavonoids are formed in plants from aromatic amino acids phenylalanine and tyrosine, and malonate (Pietta 2000).
Polyphenols act as mild reductants in alkali and neutral pH by reducing common inorganic and organic oxidants and react with radicals both in reversible and irreversible ways, depending on their particular structure, thus, exerting direct antioxidant capacity (Pietta 2000;Pandey and Rizvi 2009). There are also other ways how polyphenols can chemically intervene into oxidative processes as shown in experiments in vitro: polyphenols can chelate transition metals like copper and iron, which catalyze propagation of radical chain reactions (Afanas'ev et al. 1995;Korkina and Afanas'ev 1997;Brown et al. 1998) and inhibit the enzymes responsible for superoxide anion production, such as xanthine oxidase (Arimboor et al. 2011) and protein kinase C (Ursini et al. 1994). However, interaction of flavonoids with isolated beef heart mitochondria in vitro caused an additional production of ROS due to inhibition of respiratory complex I (Hodnick et al. 1986;Hodnick et al. 1988;Hodnick, Duval, and Pardini 1994).
Several papers reported the protective effects of polyphenols in vitro on islets or cultured insulin-producing cells against oxidative challengers. It was also shown that polyphenols from olive leaves protected INS-1 cells from H 2 O 2 toxicity (Cumaoglu et al. 2011). Cells preincubated with whole leaf extract or individual polyphenol compound oleuropein, followed by peroxide treatment, showed a lower percentage of necrotic and apoptotic death compared to untreated controls. Polyphenols stimulated activity of catalase, which resulted in a lower level of cellular ROS, along with improvement in insulin production. In this study it was suggested that polyphenols act through a redox-modulating mechanism rather than through direct free radical scavenging. Supplementation of cultured media with polyphenols from green tea increased recovery rates of isolated human and nonhuman primate islets. Polyphenols from tea extracts preserved islets by increasing the level of antiapoptotic Bcl-2 and decreasing level of pro-apoptotic BAX (Zhang et al. 2004).
In animal models of both T1D and T2D, polyphenols of different origin were reported to lower blood glucose levels (Al-Awwadi et al. 2004;Su, Hung, and Chen 2006;Ciocoiu et al. 2009;Dixit and Kar 2010;Ong et al. 2011). The hypoglycemic effect can be caused by multiple mechanisms. Polyphenols can decrease glucose absorption from intestine as they inhibit amylase, a polysaccharide-hydrolyzing enzyme (Ong et al. 2011). On the other hand, some studies show that polyphenols enhance glucose uptake by muscle cells via increasing expression of glucose transporter Glut4 (Cao et al. 2007;Ong et al. 2011). The third possible mechanism of lowering blood glucose by these compounds could be an inhibition of gluconeogenesis in liver by inhibiting glucose-6-phosphatase, a key enzyme in this process (Ong et al. 2011). Lowering blood glucose itself alleviates excessive metabolic burden on islets and undoubtedly plays a positive role in preserving their mass and function, a notion which was confirmed by (Coskun et al. 2005;Hahm, Park, and Son 2011).
Administration of polyphenols to diabetic animals resulted in a lower level of markers of oxidative stress as judged by lower levels of products of lipid peroxidation in pancreatic homogenates (Coskun et al. 2005), kidneys (Lee, Wang, et al. 2009) and blood plasma (Ciocoiu et al. 2009;Hininger-Favier et al. 2009) of STZ-rats. This may be a result of direct antioxidant activity of polyphenols, as well as a result of their ability to upregulate expression of antioxidant-defense enzymes: glutathione peroxidase (GSHPx) superoxide dismutase and catalase activities (Ciocoiu et al. 2009;Lee, Wang, et al. 2009;Dixit and Kar 2010). It is well known that oxidative stress plays a substantial role in the destruction of beta-cells by infiltrated self macrophages and lymphocytes. The fact that administration of polyphenols to NOD mice, a model for autoimmune diabetes, significantly decreased incidence of the disease is evidence of a direct redox modulation by polyphenols (Zunino, Storms, and Stephensen 2007).
Numerous publications were devoted to study effects of polyphenols on a number of health-related functions. However, it should be noted that the effective therapeutic dose of natural polyphenols is rather high varying from 15 to 500mg/kg of animal body weight. This translates into 1 to 20g of polyphenols a day for a human patient of 65 kg of weight; the amount seems to be unfeasible to get as a part of regular diet or acute therapeutic treatment. However, studies on the effect of polyphenol-rich diets with much lower doses on human patients mostly of T2D report improvement in several metabolic responses, like blood glucose and lipid content, insulin sensitivity and markers of oxidative stress (Banini et al. 2006;Dembinska-Kiec et al. 2008;Stote and Baer 2008;Zunino 2009;Fenercioglu et al. 2010).

Concluding remarks
According to the current hypothesis, the lack of anti-ROS defense capacity in pancreatic beta-cells is related to the signaling role of oxidants in glucose sensing and insulin release. In normal physiological conditions the levels of oxidants are low and mitochondria play an essential role in the chain of signaling events by releasing ROS in a controlled manner. Redox reactions in beta-cells are in fine balance, but this balance can be destroyed by persistent hyperglycemia and hyperlipidemia. In conditions of diabetes beta-cells are imposed to a burden of free radicals not considered in evolution and, therefore, antioxidants administered in a proper way can alleviate the oxidative burden and offset the destruction of beta-cells.