Mechanisms of Mitochondrial Damage in Keratinocytes by Pemphigus Vulgaris Antibodies

The development of nonhormonal treatment of pemphigus vulgaris (PV) has been hampered by a lack of clear understanding of the mechanisms leading to keratinocyte (KC) detachment and death in pemphigus. In this study, we sought to identify changes in the vital mitochondrial functions in KCs treated with the sera from PV patients and healthy donors. PV sera significantly increased proton leakage from KCs, suggesting that PV IgGs increase production of reactive oxygen species. Indeed, measurement of Mitochondria-protecting drugs abolish effects of MtAbs. A , differential effects of mitochondria-protecting drugs against the deleterious effects of PV from different patients. Representative results showing the ability of selective drugs to abolish an increase of OCR and proton leak. Het-1A cells were analyzed in real time with a Seahorse XF analyzer, as detailed under “Materials and Methods.” pretreated mice injection


Introduction
Pemphigus vulgaris (PV) is a lifelong, IgG autoantibody-mediated blistering disease affecting oral/esophageal surfaces and/or skin. Patients develop blisters and nonhealing erosions caused by suprabasal split within the stratified squamous epithelium. Prior to the introduction of therapy with oral corticosteroids in the 1950s, PV had a dismal natural course with a 50% mortality rate at 2 years and 100% mortality rate by 5 years after onset of the disease. Although systemic corticosteroid treatment is lifesaving, the high dose and prolonged courses required for disease control are associated with significant adverse effects, including death (1,2). At present, mortality remains at a relatively high level of up to 12% (3). The optimal therapeutic strategy has not been established, the steroid-sparing drugs are associated with significant adverse effects themselves, and it is unknown which is the preferable steroid-sparing agent (4). The ultimate goal of pemphigus research is to develop an effective treatment modality that would allow patients to achieve and maintain clinical remission without the need for systemic corticosteroids.
The mechanism of detachment of keratinocytes (KCs) in PV, termed acantholysis, remains obscure and is a subject of intensive research. On the cell membrane of KCs, IgG autoantibodies produced by PV patients react with desmoglein (Dsg) 3 ± 1 and other self-antigens and elicit downstream signaling events causing cell shrinkage, detachment from neighboring KCs, and rounding up (reviewed in Ref. 5). Although the pathogenic role of anti-Dsg antibodies is well characterized (reviewed in Ref. 6), our recent studies using proteomic technology identified novel targets of PV autoimmunity (7,8). Various anti-keratinocyte antibodies may concur to cause blistering by acting synergistically with anti-Dsg antibodies, as has been described through the "multiple hit" hypothesis (9). Our focus on mitochondrial targets of PV autoimmunity stems from classic and modern studies placing mitochondrial damage within the pathophysiological loop. The mitochondrial dysfunction in the lesional skin of PV patients had been suggested by an increase of lipid peroxidation and peroxidant-antioxidant balance, reflecting an increased production of reactive oxygen species (ROS) (10,11) and measuring oxidative stress (12), respectively; abnormal activities of the mitochondrial enzymes oxidoreductase, adenosine triphosphatase, and NADHcytochrome c reductase (13)(14)(15); and skewed balance between oxybiotic and anoxybiotic metabolism toward the latter (15).
Studies of mitochondrial antibodies (MtAbs) in pemphigus were pioneered by Geoghegan and Jordon in 1992 (16) and further developed by our group. We became interested in MtAbs because we sought to elucidate the mechanism of intrinsic apoptosis of KCs in PV originally demonstrated by us (17) and confirmed by others (18,19). The direct evidence that MtAbs are critical to disease pathology, rather than a bystander phenomena in PV, was provided by the studies demonstrating that PV IgGs enter KCs and specifically bind to a number of mitochondrial proteins, which is associated with the mitochondrial damage manifested by cytochrome c release (20). Most importantly, adsorption of MtAbs abolished the ability of the IgG fraction of PV serum (PVIgG) to cause keratinocyte detachment (i.e. acantholysis) and skin blistering, thus illustrating their pathophysiological significance. Using a protein microarray approach, we have recently analyzed antigen specificities of autoantibodies of a large cohort of pemphigus patients and identified a number mitochondria-associated proteins targets by MtAbs (8). The most common targeted for MtAbs in PV are listed in Table 1. On the other hand, there is growing evidence that the pharmacological agents that can protect mitochondria, such as minocycline, nicotinamide (also called niacinamide), and cyclosporine A, are therapeutic in PV patients (see Table 2). Thus, taken together, the existing data suggest 4 2 Test Sera and Cells

Analysis of Mitochondrial O Respiration by Extracellular Flux Measurement
strongly that PVIgG binding to KCs causes mitochondrial dysfunction and oxidative stress, triggering apoptosis of KCs and acantholysis (also called apoptolysis (21)), and that correction of mitochondrial function may be therapeutic in PV.
In this study, we employed assays of mitochondrial functions to identify changes in the vital mitochondria functions, such as O respiration, mitochondrial membrane potential (ΔΨ ), and intracellular production of ROS, in KCs treated with the sera from PV patients and healthy donors. The obtained results indicated that MtAbs produced by PV patients can disrupt the electron transfer chain, resulting in a loss of electrochemical gradient across the inner membrane, increase ROS production, and reduce the ability of KCs to respond to stress. Although the MtAbs of individual PV patients elicited unique patterns of mitochondrial damage, mitochondria-protecting drugs exhibited a uniform protective effect. Their therapeutic activity was validated in the passive transfer PV model in neonatal BALB/c mice. The obtained results explain the mechanism of therapeutic action of mitochondria-protecting drugs in PV patients and suggest novel avenues for treatment of this potentially lethal immunoblistering disease.

MATERIALS AND METHODS
We tested six PV patient and six normal serum specimens. This study was approved by the University of California Irvine Human Subjects Review Committee. The diagnosis of PV was made based on the results of comprehensive clinical and histological examinations and immunological studies that included both direct immunofluorescence of skin biopsies and indirect immunofluorescence of the patients' sera on various epithelial substrates. The titer of "intercellular" antibodies determined on monkey esophagus ranged from 1/320 to 1/2560. The presence of anti-Dsg 1 and Dsg 3 antibodies in each PV serum was established using the MESACUP Dsg 1 and Dsg 3 ELISA test system (MBL, Nagoya, Japan). The index values for Dsg 1 antibodies ranged from 26 to 72, and those for Dsg 3 antibodies ranged from 65 to 164, i.e. were unequivocally positive. Patient specimens were deidentified prior to testing. As controls, we used normal human sera purchased from Bioreclamation, Inc. (Westbury, NY). The Het-1A cell line, an established clonal population of SV40-immortalized human esophageal squamous epithelial cells (i.e. KCs) widely used for the studies of apoptosis (22), was purchased from American Type Culture Collection (Manassas, VA; catalogue no. CRL-2692) and propagated in the Clonetics brand bronchial cell medium without retinoic acid (Cambrex Bio Sciences, Walkersville, MD), as detailed by us elsewhere (23).
To measure mitochondrial function in Het-1A cells, we employed a Seahorse Bioscience XF24 Extracellular Flux Analyzer (Seahorse Bioscience, North Billerica, MA) and followed the manufacturer's protocol. Briefly, KCs were plated in a 0.2% gelatin coated 24-well Seahorse XF-24 assay plate at 3 × 10 cells/well and grown for 16 h before being treated in quintuplicate with either PV patient's or normal human sera at a final concentration of 4% for another 24 h in the culture medium. On the day of metabolic flux analysis, the cells were washed once with freshly prepared Krebs-Henseleit buffer (111 mM NaCl, 4.7 mM KCl, 2 mM MgSO , 1.2 mM Na HPO , 2.5 mM glucose, and 0.5 mM carnitine, pH 7.4) and incubated in Krebs-Henseleit buffer at 37°C in a non-CO incubator for 1 h. Three base-line measurements of oxygen consumption rate (OCR) were taken before sequential injection of the following mitochondrial inhibitors and final concentration: oligomycin (1 μg/ml), carbonyl cyanide p-trifluoromethoxyphenylhydrazone (3 μM), and rotenone (0.1 μM). Measurements were taken after addition of each of the three inhibitors. The OCR values were automatically calculated and recorded by the Seahorse XF-24 software. The basal respiration was calculated by averaging the three measurements of ORC before injection of inhibitors. The proton leak was Measurement of Mitochondrial ΔΨ Using the JC-1 Dye

Morphometric Assay of Acantholysis in Vivo
Statistical Analysis calculated using OCR measurement after oligomycin injection minus OCR measurement after rotenone injection.
The changes in ΔΨ induced by test sera were measured using a standard protocol (24). Briefly, Het-1A cells were plated in a 6-well plate at a density of 5 × 10 per well, incubated for 16 h to allow cells to adhere to the dish surface, after which the cells there were either left untreated (negative control) or exposed for 24 h to PV or control sera at a final concentration of 4% of the total volume per well. Next, experimental and control cells were exposed to the fluorescent cationic dye JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbo-cyanine iodide) for 25 min before observation at green and red emission wavelengths using a Zeiss AxioPlan II fluorescence microscope (Carl Zeiss, Thornwood, NY). Intact mitochondrial membrane retains the JC-1 dye and forms J-aggregate (orange-red fluorescence), whereas depolarization of membrane decreases the ability to retain the dye that remains as a monomer (green fluorescence). Increased green fluorescence and deceased orange-red fluorescence thus indicated a loss of ΔΨ . In the absence of apoptosis, the lipophilic JC-1 dye bearing a delocalized positive charge accumulates in the mitochondrial matrix and stains the mitochondria bright red, whereas in mitochondrial apoptotic ΔΨ dissipates, the cells display green fluorescence.
The measurement of intracellular ROS was performed in accordance with the manufacturer's protocol. The Het-1A cells were incubated and treated as described above in the ΔΨ measurement section. Briefly, to measure intracellular ROS, the treatment medium was removed, and 5 μM c-H DCFDA-AM (Invitrogen) in PBS, pH 7, was added to each well and incubated for 15 min at 37 °C. The cells were then photographed using a fluorescence microscope. C-H DCFDA-AM is hydrolyzed by intracellular esterases and oxidized to fluorescent carboxy-DCF (excitation, 488 nm; emission, 525 nm).
For the FACS analysis, the experimental and control Het-1A cells were incubated at 37 °C with either 5 μM of JC-1 dye for 60 min (ΔΨ measurement) or 5 μM of c-H DCFDA-AM for 15 min (ROS measurement), washed three times with PBS, trypsinized, and resuspended in PBS. The fluorescence was measured with a BD LSR II flow cytometer (excitation, 488 nm; emission, 576 nm), and the data were analyzed using the BD FACSDiVa software.
PVIgGs were isolated by FPLC protein G affinity chromatography using the FPLC System purchased from Amersham Biosciences following the manufacturer's protocol, as detailed elsewhere (17). Minocycline, nicotinamide, and cyclosporine A were purchased from Sigma-Aldrich. One-day-old pups delivered by the BALB/c mice purchased from the Jackson Laboratory were used to investigate the effect of mitochondria-protecting drugs on the extent of epidermal acantholysis induced by passive transfer of 1 mg/g of body weight of PVIgG. The pups were injected subcutaneously with PVIgG with or without test drugs and examined 24 h later for the extent of epidermal acantholysis, as detailed elsewhere (20). Briefly, the euthanized animals were snap frozen in liquid nitrogen, crosssectioned at the umbilicus level, embedded into the OCT compound (Miles Scientific, Naperville, IL), and stained by hematoxylin and eosin. Five random microscopic fields in each skin section were captured at magnification 10×, using a Macintosh computer attached to the Axiovert 135 inverted microscope. The images were printed, and the extent of acantholysis was computed directly on the prints by measuring the length of the areas in the epidermis in which suprabasal cell detachment spread along more than four adjacent basal cells.
The data were analyzed using analysis of variance against an α level of 0.05 and

Mitochondria-protecting Drugs Abolish Deleterious Effects of PV Sera on Mitochondrial O Respiration in KCs
presented as the mean ± S.D. The graphs were made using GraphPad Prism 5.

RESULTS
Previously, we had demonstrated that PV patients develop autoantibodies that can find their way to mitochondria in KCs and react with a versatile group of mitochondrial proteins, which contributed to PV-like cell detachment in monolayers of human KCs and epidermis of neonatal mice (20). Most recently, we have tested 264 pemphigus and 138 normal control sera on the multiplexed protein array platform containing 701 human genes encompassing many keratinocyte molecules, including 283 mitochondria-associated proteins (8). A detailed analysis of the supplementary data published by us online (8) identified the most common antigens recognized by MtAbs and the percentage of PV patients and healthy people producing such MtAbs (Table 1). Based on the known functions of mitochondrial proteins most commonly targeted in PV, the following mitochondrial pathways might be subject to dysfunction: oxidative phosphorylation, O respiration, and production/inactivation of ROS.
To identify the direct effect of MtAbs on O respiration, we performed a Seahorse Bioscience XF24-3 extracellular flux analysis of OCR in KCs treated with PV or normal human sera. Although none of four tested normal sera changed basal mitochondrial respiration, all PV sera significantly increased this mitochondrial function (Fig. 1, A and B). The PV sera also significantly increased proton leakage from treated cells (Fig. 1C). These results suggested that MtAbs produced by PV patients increase the ROS production in KCs and reduce the ability of these cells to respond to stress, which can trigger the mitochondria mediated (intrinsic) apoptosis.
Exposure of KCs to PV, but not normal, sera caused dramatic changes in the mitochondrial membrane potential, manifested by disappearance of orange-red fluorescence of JC-1 dye with simultaneous increase in green fluorescence of cultured cells ( Fig. 2A). A significant drop in the mitochondrial membrane potential in KCs treated with PV sera was confirmed by measurements of ΔΨ using FACS (Fig. 2B). These results indicated that binding of MtAbs to mitochondrial target proteins disrupts the electron transfer chain, resulting in a loss of electrochemical gradient across the inner membrane.
When intracellular ROS production was measured by the c-H DCFDA-AM dye labeling of cultured KCs, we observed a drastic increase of cell staining in response to treatment by each individual PV serum, compared with normal sera (Fig. 3A). Similar results were obtained in FACS analysis (Fig. 3B), confirming that treatment with PV sera significantly increases the number of c-H DCFDA-AM-positive KCs, compared with normal controls. These results indicated that binding of MtAbs to keratinocyte mitochondria increases intracellular ROS production.
To evaluate direct effects of the mitochondria-protecting drugs minocycline, nicotinamide, and cyclosporine A, given alone or in combination, on the KCs exposed to PV serum, we measured OCR in real time in an Seahorse XF Efflux Analyzer and calculated proton leak. KCs exposed to different PV sera showed differential responses to test drugs (Fig. 4A). Drug combination regimen showed the highest efficacy in abolishing the PV serum-induced elevation of OCR and proton leak (Fig. 4A). These results indicated that although each PV patient's MtAbs display a unique pattern of mitochondrial damage, a combination of mitochondria-protecting drugs exhibits a uniform protective action. Treatment of Experimental PV in Neonatal Mice by Pharmacologically Protecting the Mitochondria Next, we sought to confirm the therapeutic efficacy of the mitochondria-protecting drugs in the passive transfer model of PV in neonatal BALB/c mice. Coadministration of PVIgG and each of test drugs abolished the acantholytic activity of PVIgG, albeit with different efficacy (Fig. 4B). A decrease of PVIgG-induced epidermal splitting in mice treated with minocycline or cyclosporine A reached statistical significance (p < 0.05). Remarkably, the highest efficacy of mitochondrial protection of the combination of minocycline, nicotinamide, and cyclosporine A observed in the mitochondrial assay was consistent with the ability of this combination to almost completely abolish acantholysis in mouse skin (Fig. 4B). Thus, preventing the MtAb-dependent mitochondrial damage protected KCs from PVIgG-induced detachment and skin blistering.

DISCUSSION
PV is a complex, multifactorial disease (5,25). The fact that adsorption of MtAbs from PVIgGs abolished the ability of PVIgGs to induce acantholysis (20) provided strong evidence of the indispensible role of mitochondrial damage in the disease mechanism. In this study, we demonstrated for the fist time dramatic changes in the mitochondria O respiration, dissipation of ΔΨ , and drastic increases in the intracellular levels of ROS. These changes can induce intrinsic apoptosis, which is consistent with observations that PVIgG binding to KCs is associated with mitochondrial damage manifested by cytochrome c release and activation of caspase 9 (17)(18)(19)(20). Because the mode of mitochondrial damage by MtAbs depends on the biological function of targeted protein, the exact primary mechanisms of mitochondrial damage apparently differ from patient to patient, in keeping with striking variations in the disease severity and response to treatment among different PV patients (26). The differential patterns of mitochondrial damage in PV have been also suggested by the fact that the mitochondrial function in KCs exposed to different PV sera showed differential responses to the same mitochondria-protecting drugs (Fig. 4A).
The structural-functional interrelationships of mitochondria-associated proteins targeted by MtAbs (  Table 1) indicate that autoantibody binding to even a single mitochondrial target antigen can distort the neatly orchestrated cascade of interactions of enzymes and their substrates, producing a domino effect, resulting in the mitochondrial dysfunction. Because MtAbs produced by PV patients can bind their target proteins at both mitochondrial outer and inner membranes, as well as mitochondrial matrix, they can interfere with the biological processes taking place at these locations. Furthermore, because MtAbs recognized both mitochondrial-and nuclear-encoded proteins involved in the regulation or execution of specific mitochondrial functions (Table 1), the protein import into the mitochondria may also be altered, resulting in a pathologic remodeling of mitochondrial proteome.
The results of the present study clearly demonstrated that exposure of KCs to PV serum can disrupt their ΔΨ . This phenomenon can be explained based on the knowledge about involvement of mitochondrial structures in multiple physiological pathways, so that formation of an antigen-antibody complex in one location can influence several biochemical reactions. For example, binding of MtAbs to the complex I proteins, such as NADH dehydrogenase or electron transfer flavoprotein targeted in PV (Table 1), can interfere with electron exchange among the carriers and their access to succinate and NADH and affect their ability to establish a chemiosmotic gradient. Thus, disruption of ΔΨ in KCs exposed to PV sera observed in this study can prevent electron transport (because of MtAb binding one or more electron carriers), which is an early event in apoptosis activated through the mitochondrial pathway (reviewed in Ref. 27). It is well established that a delicate balance exists between the moderate ROS production to modulate physiological signaling and the overproduction of ROS that can lead to oxidative stress. ROS detoxification pathways exist to minimize oxidative damage. The presence of anti-superoxide dismutase MtAb in PV patients (Table 1) is in keeping with our observation that preincubation of KCs with PV sera led to excessive ROS production, thus precipitating changes in mitochondrial homeostasis. However, the mitochondria damaged by other types of MtAbs can also lead to ROS increase, creating a pathological process termed the "vicious cycle." For example, impairment in the mitochondrial electron transport that leads to a loss of ΔΨ also can be responsible for an incomplete O reduction, resulting in an increment in ROS production that further amplifies the generation of ROS. Such abnormalities can, in turn, trigger the cell death signaling cascade, wherein the executioner caspases cleave the cell adhesion molecules, such as desmosomal cadherins (reviewed in Ref. 28), leading to both cell detachment (acantholysis) and death (apoptosis), the unique form of keratinocyte demise in PV that we have tentatively termed apoptolysis (21). Further elucidation of the mitochondrial mechanisms of apoptolysis in PV should therefore improve our understanding of disease pathogenesis and facilitate development of personalized therapies based on the pharmacological correction of mitochondrial abnormalities in individual PV patients.
The results of the present study provide the first experimental evidence that pharmacological protection of mitochondrial function can prevent the skin blistering induced by PVIgG. These results shed light on the mechanism of therapeutic action of minocycline and nicotinamide, which have been empirically shown to be effective steroid-sparing agents in PV, as well as of the professional immunosuppressor cyclosporine A. Importantly, reports in the literature testify that in some patients PV can be controlled by the mitochondriaprotecting drugs alone without the need for prednisone (29)(30)(31)(32)(33)(34)(35). Although these drugs belong to different chemical groups, each exhibiting unique pharmacological effects, they share the mitochondria-protecting activities (Table 2). However, except for empirically chosen combinations of minocycline and nicotinamide, no other combinations of the mitochondria-protecting drugs have been reported in PV. Our results demonstrated synergy among minocycline nicotinamide and cyclosporine A. The combination of drugs that most effectively protected mitochondrial respiration equally effectively antagonized the diseasecausing activity of PVIgG in mouse skin. Thus, their synergistic potential is yet to be fully explored.
Because the mitochondria-protecting drugs used in this study are already in use in the treatment of PV patients, the obtained results translate into clinical practice. There is a strong rationale for the principally new treatment of patients with PV. The existing therapies do not allow one to reliably control acute PV without the systemic corticosteroid prednisone, which, although lifesaving, causes serious adverse events, including death, because of the need for its high dose (e.g., 1-2 mg/kg/day) and prolonged (e.g., 18 months) usage to achieve disease control (1,2). Although the incidence of PV is only 1-16 per million population per year (36,37), this disease represents a significant burden to health care professionals and the health care system because of the hazardous side effects of conventional immunosuppressive therapy requiring prolonged and frequent hospitalizations and high costs of modern treatment regimens (38). By identifying the dose-dependent effect, one should be able to increase drug efficacy, because the mitochondria-protecting drugs are being used in PV patients at their suboptimal doses. For example, whereas nicotinamide is given to PV patients at 1.5-2 g/day, this drug can be safely used at higher doses, like in patients with diabetes or schizophrenia (39). Likewise, cyclosporine A is usually used in PV at 0.5-1.5 mg/kg/day, which is much less compared with the dosages used in transplantology (40). These facts suggest that current suboptimal clinical activity of the mitochondria-protecting drugs used in PV can be improved by identifying their optimal combination through the laboratory screening of patient's serum m 2 prior to treatment. Because the mitochondria-protecting drugs are expected to have different effects in different PV patients, as the mitochondrial pathways targeted by MtAbs are apparently unique to each patient, an individualized regimen will be needed to achieve the optimal therapeutic response.
In conclusion, the integrity of mitochondrial function is fundamental to cell life. Mitochondria are involved in many processes essential for cell survival, such as energy production, redox control, calcium homeostasis, and a number of metabolic and biosynthetic pathways. In addition, mitochondria often play an essential role in physiological cell death mechanisms (reviewed in Ref. 41). The results of the present study indicate that binding of PVIgG to mitochondrial targets can disrupt the electron transfer chain, resulting in a decline in ATP production, loss of electrochemical gradient across the inner membrane, and reduction in O with increased generation of the ROS superoxide, hydrogen peroxide, and hydroxyl radical. These novel findings provide a theoretical background for clinical reports of the efficacy of mitochondria-protecting drugs in PV patients. Pharmacological protection of mitochondria and/or compensation of an altered mitochondrial function may therefore become a novel approach to the development of safer, i.e. nonhormonal, therapies of this severe autoimmune blistering disease.     Effects of PV sera on ROS production in KCs. A, representative images of cultured Het-A1 cells treated with PV versus normal sera and subjected to microscopic evaluation of the c-H DCFDA-AM dye staining via fluorescence microscopy, as described under "Materials and Methods." The images were taken with a Zeiss AxioPlan II fluorescence microscope using the same exposure time (40 ms   The drugs had no effect on control cells (not shown). *, p < 0.05 compared with control; #, p < 0.05 compared with PV sera without drugs. B, treatment of experimental PV in neonatal mice by protecting the mitochondria. Neonatal BALB/c mice (n = 3) received a single subcutaneous injection of 1 mg/g of body weight of PVIgG alone or in combination with test drug(s), and the extent of acantholysis in mouse epidermis was measured 24 h after injection using a morphometric assays described under "Materials and Methods." *, p < 0.05 compared with PVIgG injected alone.  (48) Nicotinamide, Nicotinamide is a precursor of the coenzyme NAD used to Used to treat PV patients both +