Spin-trapping nitrones such as α-phenyl-N-tert-butylnitrone (PBN) have traditionally been used to trap and stabilize free radicals for detection by electron paramagnetic resonance (EPR) spectroscopy. Unlike classical antioxidants, these agents have never been evaluated therapeutically in allograft transplantation. In the present study, we examined potential mechanisms of action of treatment with PBN in a rat model of acute cardiac allograft transplantation. Graft rejection was determined by histological examination and graft function determined by in situ sonomicrometry. DNA binding for nuclear factor (NF)-κB and activator protein (AP-1) were determined by gel shift assays. Western blot and reverse transcriptase-polymerase chain reaction (RT-PCR) analysis was performed for inducible nitric-oxide synthase (iNOS) and inflammatory cytokines. Histological rejection scores were elevated in untreated allografts and decreased by treatment with PBN. In situ sonomicrometry revealed decreased heart rate and distended end diastolic and end systolic segment lengths with rejection. Although PBN did not alter heart rate, it did normalize the distention of both diastolic and systolic cardiac dimension. EPR spectroscopy revealed nitrosylation of myocardial heme protein in untreated allografts that was decreased by treatment with PBN. PBN also decreased iNOS protein and iNOS mRNA. RT-PCR analysis revealed enhanced cytokine gene expression for interferon-γ, interleukin-6, and interleukin-10 in untreated allografts. Expression for these genes was potently inhibited or abolished in recipients treated with PBN. PBN treatment also decreased DNA binding of transcription factors, NF-κB and AP-1. Thus, PBN retains significant anti-inflammatory properties through its action to down-regulate cytokine gene expression that contribute to protection against acute alloimmune activation in cardiac allografts.
The complex array of mediators of alloimmune activation in acute allograft rejection is incompletely understood. The interaction of antigen-presenting cells and lymphocytes results in immune activation. This process causes the release or synthesis of mediators such as reactive oxygen species, nitric oxide (NO) derived from iNOS, lymphokines, and cytokines that are prime candidates for mediators of acute inflammatory responses in cardiac allografts.
Reactive oxygen may be an antecedent signal that mediates proinflammatory gene expression in alloimmune activation owing to activation of redox-sensitive transcription factors. Activation and nuclear translocation of transcription factors such as NF-κB and AP-1 are known to require reactive oxygen as a signal. These transcription factors are important in the expression of proinflammatory mediators such as cytokines and iNOS. The importance of activation of these transcription factors are supported by our studies showing that antioxidants such as dimethylthiourea and dithiocarbamate-based derivatives that inhibit activation of NF-κB and/or AP-1 also decrease cardiac allograft rejection (Cooper et al., 1998; Pieper et al., 2001, 2003). Although these antioxidants have been protective in cardiac transplantation, it has been shown in other experimental in vitro models that certain antioxidants in cytokine-stimulated cell cultures can also cause paradoxically increased activation of these transcription factors (Meyer et al., 1994; Schenk et al., 1994; Sen and Packer, 1996). Thus, it cannot be predicted with certainty that a given antioxidant will inhibit DNA binding of these transcription factors.
Nitrones are a class of agents known as spin traps that have been developed over the years for trapping and stabilizing free radicals for unequivocal detection in biological systems using electron paramagnetic resonance (EPR) spectroscopy (Degray and Mason, 1994). One of these agents, α-phenyl-N-tert-butylnitrone (PBN), has been shown to display anti-inflammatory actions in cytokine-stimulated cells in culture (Kotake et al., 1998) and following a single, acute injection to offset inflammatory actions of lipopolysaccharide in vivo (Sang et al., 1999). The anti-inflammatory potential of chronic administration with PBN in vivo is unknown. In the present study, we investigated the potential efficacy of PBN during alloimmune activation in cardiac transplants and to determine whether the benefit of PBN could be explained by effects on DNA binding of redox-sensitive transcription factors and down-regulation of proinflammatory genes such as iNOS and cytokines.
Materials and Methods
Transplantation. Donor and recipient rats were anesthetized with an i.p. injection of 50 mg/kg sodium pentobarbital. Heterotopic cardiac transplantation was performed under sterile conditions as previously described (Cooper et al., 1998). Cardiac graft function was also quantitated in situ at postoperative day 6 (POD6) using sonomicrometry from crystals placed on the external surface at the mid-heart level to determine short-axis dimension/graft function (Sonometrics Corp., London, Ontario, Canada). This time was chosen since all grafts at POD6 have functional activity, albeit decreased, whereas full cessation of function typically occurs on POD7 in this model.
Lewis (Lew:RT11) and Wistar-Furth (WF:RT1u) rat strains were chosen to represent genetic disparity at both the major and minor histocompatibility loci for donor-to-recipient combination of Lew→Lew (for isografts) or WF→Lew (for allografts). Some allograft recipients received i.p. injections of 150 mg/kg PBN daily beginning the day of surgery until harvesting of tissue for analysis on POD4 or POD6. This dose was chosen since it is in the range that has been reported to decrease activation of transcription factors and limit cytokine expression in a model of endotoxic shock (Sang et al., 1999) or to protect against drug-induced insulin-dependent diabetes mellitus (Tabatabaie et al., 1997). At POD4 and POD6, grafts were arrested and flushed with a cold University of Wisconsin solution, minced, and frozen in liquid N2. Samples were either stored under liquid N2 for EPR spectroscopy or at -80°C for Western blot, RT-PCR, and electrophoretic mobility shift assay.
Histological Rejection Scoring. Tissue from grafts at POD6 was fixed in 4% phosphate-buffered formalin with paraffin-embedded sections stained with hematoxylin and eosin. Rejection scoring was based upon a six-point graded criteria established by the International Society for Heart and Lung Transplantation (ISHLT) and described previously (Pieper et al., 2002).
EPR Spectroscopy. X-band EPR spectroscopy was performed on a Varian E-109 spectrometer (Varian, Inc., Palo Alto, CA). Samples from each group were analyzed at 77°K on the same day under similar instrument settings consisting of: 1000 gauss scan range, 4-min scan time, 0.25-s time constant, 2 gauss modulation amplitude, 100-KHz modulation frequency, and 5 mW of microwave power. The magnetic field was calibrated with Fremy's salt giving a g value of 2.0055 ± 0.0001.
Western Blotting. Frozen tissue was processed as previously described in our laboratory (Slakey et al., 1993). Samples were electrophoresed on 7.5% SDS-polyacrylamide gel electrophoresis and transferred to membranes. Blots were probed with 1:2000 dilution of rabbit anti-iNOS (Santa Cruz Biotechnology, Inc. Santa Cruz, CA) and visualized using 1:5000 dilution of donkey anti-rabbit IgG horse-radish peroxidase-conjugate and enhanced chemiluminescence.
Gene Expression. Total RNA was purified using the Promega SV total RNA isolation system (Promega, Madison, WI) according to the manufacturer's directions. RNA concentration was determined spectrophotometrically. cDNA was synthesized from 1 μg of total RNA and oligo(dT) primers using the Invitrogen Superscript first-strand synthesis system for RT-PCR (Invitrogen, Carlsbad, CA) according to the manufacturer's directions. One microliter of cDNA was mixed with 25 pmol of specific sense and antisense primers and Invitrogen PCR supermix to a volume of 25 μl, and the reaction was incubated in a Bio-Rad iCycler (Bio-Rad, Hercules, CA) under the following conditions: 95°C (30s), 60°C (30 s), and 72°C (60 s) for 35 cycling times for interferon-γ, interleukin-6, and interleukin-10. One microliter of the PCR product was resolved by 1% agarose gel electrophoresis. Ethidium bromide-stained specific bands were visualized under UV light and densitometric analysis of specific bands was made using an Alpha Imager (Alpha Innotech, San Leandro, CA) and expressed as a ratio to β-actin gene controls.
Electrophoretic Mobility Shift Assay for Nuclear Proteins, NF-κB or AP-1. Extraction of nuclear protein from homogenates of cardiac allografts was performed as described (Pieper et al., 2003). Double-stranded NF-κB or AP-1 oligonucleotides (Promega) were end-labeled with [γ-32P]ATP as described (Pieper et al., 2003). DNA binding reactions were performed at room temperature using 12 μg of nuclear extract, 0.5 ng of labeled oligonucleotide, and 3 μg of poly(dI-dC) (Pfizer, Inc., Kalamazoo, MI). After incubation for 30 min, the reactions were electrophoresed on 4% nondenaturing polyacrylamide gel in 0.5× Tris-borate EDTA at 10 V/cm. Specificity for NF-κB or AP-1 binding activity was verified by competition with 100-fold excess of unlabeled mutant or wild-type competitor oligonucleotides. Gels were dried and exposed to autoradiographic film. Intensity of NF-κB or AP-1 binding activity was determined by AlphaImager 2000 image analysis system (Alpha Innotech).
Data Analysis. EPR spectra were processed for presentation by using SUMSPEC and Grapher programs (Golden Software, Golden, CO). Statistics were performed by analysis of variance for multiple group means or by Student's t test for comparisons between two group means. Statistical significance was set at the level of P < 0.05.
Histological and Functional Rejection. There was a significant (P < 0.001) increase in the histological rejection score at POD6 in untreated allografts compared with isografts (scores = 1.5 ± 0.5, isografts; 4.8 ± 0.4, untreated allografts, n = 6–9 each). There was a significant (P < 0.01) inhibition in histological rejection scores at POD6 in allografts treated with PBN (score = 3.2 ± 0.4, treated allografts). In situ functional analysis using sonomicrometry revealed significant myocardial dysfunction in untreated allografts compared with isograft controls (Fig. 1). Both heart rate and percentage of short-axis fractional shortening were decreased in untreated allografts, and these changes were not altered by treatment with PBN. Both end systolic and end diastolic lengths were increased in untreated allografts but were normalized in allografts treated with PBN.
EPR Spectroscopy. In allografts examined at POD6, we detected EPR signals for nitrosylation of myocardial heme protein in untreated allografts (not shown) and allografts treated with saline vehicle (Fig. 2). In every sample examined, nitrosylation of heme protein was markedly inhibited in allografts treated with PBN compared with untreated allografts (i.e., 52 ± 9% inhibition versus untreated).
iNOS Expression. Western blot analysis revealed induction of iNOS in untreated allografts compared with isograft controls (Fig. 3A). The increase in iNOS protein was inhibited by 50% in allografts treated with PBN. Furthermore, RT-PCR analysis indicated that decreased iNOS protein by PBN was associated with decreased iNOS mRNA measured in allografts (Fig. 3B).
Nuclear Binding Activities. We also examined nuclear binding activity for transcription factors, NF-κB and AP-1 in nuclear extracts obtained in grafts at POD4. This time was previously determined in our laboratory as the time showing maximum DNA binding activity (Cooper et al., 1998). Nuclear binding for both NF-κB and AP-1 was significantly increased in allografts compared with isograft controls (Fig. 4). The increased nuclear binding activity for both transcription factors was significantly inhibited in allografts receiving chronic treatment with PBN (Fig. 4).
Inflammatory Cytokine Gene Expression. To determine the impact of inhibition of DNA binding of transcription factors, we also examined gene expression for inflammatory cytokines, interferon-γ, interleukin-6, and interleukin-10. In samples derived from POD4, we found increased gene expression for each cytokine in allografts compared with isografts (Fig. 5). Gene expression was inhibited by treatment with PBN. Similarly, in samples derived from POD6, we found increased gene expression for each cytokine in allografts (Fig. 6). In this case, treatment with PBN blocked the increased gene expression for each inflammatory cytokine (Fig. 6).
In the present study, we found that treatment with the free radical spin-trapping agent, PBN, caused a profound inhibition of gene expression for inflammatory cytokines and iNOS. This effect appears to be explained, at least in part, by regulating transcription as we also observed a significant inhibition of transcription factors, NF-κB and AP-1.
Traditionally, nitrones such as PBN have been used as agents to trap and stabilize free radicals for detection by advanced biophysical techniques such as EPR spectroscopy (Degray and Mason, 1994). Indeed, in the setting of solid organ transplantation, this agent has been used to detect free radical production upon reperfusion after long-term cold storage of liver grafts (Connor et al., 1992). In contrast, nitrones have not been evaluated in a therapeutic modality in organ preservation or for alloimmune activation in transplantation. We have previously shown that diverse antioxidants such as dimethylthiourea or vitamins C and E can also provide acute protection in acute cardiac transplants (Slakey et al., 1993; Pieper et al., 2001). Thus, in the absence of any information on the use of nitrones, we hypothesized that PBN could be used therapeutically as an agent to limit alloimmune activation and rejection. PBN has been used to prolong life spans in mice (Edamatsu et al., 1995), to prevent age-related changes in brain protein oxidation and brain function (Carney et al., 1991), and to protect against the onset of drug-induced and autoimmune diabetes mellitus (Tabatabaie et al., 1997; Iovino et al., 1999).
In our study, we showed that PBN significantly limited histological rejection and normalized the distention in end diastolic and end systolic segment lengths indicative of improved graft function. The lack of action of PBN to improve the decrease in heart rate is unknown. Interestingly, this action of PBN appears to be shared by other antioxidants based on recent findings in our laboratories in which heart function was improved by vitamin E or a metalloporphyrin with superoxide dismutase mimetic actions but did not improve the decrease in heart rate (unpublished observations). Thus, in acute cardiac rejection, diastolic and systolic dysfunction versus heart rate may arise via different mechanisms.
Early studies in cultured cells indicated that PBN was able to inhibit iNOS gene expression in vitro in macrophage and insulinoma cells stimulated with cytokines (Kotake et al., 1998; Tabatabaie et al., 2000). A limited number of studies show that PBN may be able to inhibit iNOS gene expression in models of sepsis in vivo (Sang et al., 1999; Zhang et al., 2003). In our study, we found that PBN was able to significantly inhibit iNOS protein and mRNA. Interestingly, we have noted that a variety of antioxidants have shown variable efficacy to inhibit iNOS gene expression in acute cardiac allografts. For example, no change was achieved using vitamin C (Nguyen et al., 2003), whereas partial inhibition was achieved with either vitamin E (unpublished findings) or an iron-diethyldithiocarbamate complex (Pieper et al., 2003). Other than the immunosuppressant agent, cyclosporine, the current study demonstrates the most potent inhibition of iNOS gene expression that we have observed for any intervention to limit free radical activity.
Previously, we have noted marked increases in nitrosylation of myocardial heme protein in untreated allografts and that agents such as cyclosporine or agents that scavenge NO (Pieper et al., 2000, 2002) or inhibit NO production (Nakanishi et al., 1998; Pieper et al., 2004) each individually inhibited nitrosylation. This suggests that protein modification by nitrosylation may be a marker of graft rejection. In the present study, we found significant inhibition of nitrosylation of myocardial heme protein. We conclude that this is likely to occur secondarily by the action to limit iNOS gene expression.
In our study, we showed that iNOS expression was blocked by PBN but that nitrosylation was not completely blocked but inhibited by 52%. This could be explained if PBN was also acting as a NO donor. Indeed, in vitro and in vivo degradation of PBN to form NO has been reported (Chamulitrat et al., 1993; Saito and Yoshioka, 2002). Consistent with this potential action are additional studies showing that plasma NO metabolites at POD6 in allograft recipients treated with PBN revealed levels indistinguishable from that in untreated allografts (isografts = 10 ± 2 μM; untreated allografts = 43 ± 3 μM; PBN-treated = 46 ± 2 μM).
In addition to decreased iNOS gene expression, we found that PBN was a potent inhibitor of gene expression for inflammatory cytokines. In fact, mRNA for all three cytokines, interferon-γ, interleukin-6, and interleukin-10, were completely abrogated on POD6. These findings are consistent with our recent findings of significant inhibition of interferon-γ gene expression using a diethyldithiocarbamate-iron complex (Pieper et al., 2003). Decreased inflammatory gene expression can account for the significantly decreased histological rejection scores indicative of decreased inflammatory cell infiltration into the graft.
Since PBN has been shown to partially inhibit activation of NF-κB and AP-1 in a model of sepsis (Sang et al., 1999), we examined whether PBN might have similar actions in our in vivo model of cardiac alloimmune activation. We found that treatment with PBN abrogated the increased DNA binding activities for transcription factors, NF-κB and AP-1. As there are binding sites for these transcription factors in the promoter region of iNOS and various inflammatory cytokines, we conclude that the abrogation of gene expression for iNOS and inflammatory cytokines could be explained by decreased transcription of these genes.
- Received August 11, 2004.
- Accepted September 3, 2004.
Supported in part by National Institutes of Health Grant HL-64637 (to G.M.P.), Grant AI-41703 (to A.K.K.), and for the National Institutes of Health, Electron Paramagnetic Resonance (EPR) Center Grant EB001981.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: NO, nitric oxide; iNOS, inducible nitric-oxide synthase; NF, nuclear factor; AP-1, activator protein-1; EPR, electron paramagnetic resonance; PBN, α-phenyl-N-tert-butylnitrone; POD, postoperative day; RT-PCR, reverse transcriptase-polymerase chain reaction; IL, interleukin; IFN-γ, interferon-γ.
- The American Society for Pharmacology and Experimental Therapeutics