The authors have declared that no competing interest exist.
Conceived and designed the experiments: AL IB JF. Performed the experiments: AL IB PF JF. Analyzed the data: AL IB PF JF. Contributed reagents/materials/analysis tools: PF CO. Wrote the paper: FL JF.
‡ Shared senior authorship
Rheumatoid arthritis (RA)—a widespread chronic inflammatory disease in industrialized countries—is characterized by a persistent and progressive joint destruction. The chronic pro-inflammatory state results from a mutual activation of the innate and the adaptive immune system, while the exact pathogenesis mechanism is still under discussion. New data suggest a role of the innate immune system and especially polymorphonuclear granulocytes (PMNs, neutrophils) not only during onset and the destructive phase of RA but also at the chronification of the disease. Thereby the enzymatic activity of myeloperoxidase (MPO), a peroxidase strongly abundant in neutrophils, may be important: While its peroxidase activity is known to contribute to cartilage destruction at later stages of RA the almost MPO-specific oxidant hypochlorous acid (HOCl) is also discussed for certain anti-inflammatory effects. In this study we used pristane-induced arthritis (PIA) in
Epigallocatechin gallate (EGCG), upon early and continuous oral application, considerably attenuates the symptoms in
Arthritic symptoms are not only dampened in the acute but also in the chronic phase of the disease, which means a lower risk for the development of chronic recurring joint inflammation
The therapeutic effect is comparable to the early injection of methotrexate (MTX) and is not observed upon late oral application or injection of EGCG
Stopped-flow kinetic measurements show that epigallocatechin (EGC) derived from EGCG exhibits a considerable activity with Compounds I and II of myeloperoxidase (MPO)
It can be guessed that the reactivation of the chlorinating MPO activity by EGCG may contribute to the anti-inflammatory effect of the polyphenol
Rheumatoid arthritis (RA) is a common chronic inflammatory autoimmune disease in industrialized countries and is characterized by a persistent inflammation of joints and leads to the progressive destruction of cartilage and bone [
Neutrophils (polymorphonuclear leukocytes, PMNs), the most abundant immune cells in the blood and the first leukocytes recruited during inflammation, are essentially involved in the pathology of RA [
RA was classically induced in small laboratory animals (especially mice and rats) by applying complete Freund's adjuvant (CFA), assuming that the immunogenic part (inactivated
RA is classically treated with steroidal (e.g glucocorticoids) and non-steroidal anti-inflammatory (NSAIDs, e.g. sulfasalazine) drugs to dampen the underlying chronic inflammatory processes [
As shown for several animal RA models, the natural compound epigallocatechin gallate (EGCG) also attenuates arthritic symptoms by suppressing the underlying chronic inflammation [
Female
The rats were earmarked for identification and housed in a climate-controlled room on a 12–12 hours light-dark cycle. Food and water were available ad libitum. Health status of the rats was monitored daily and weight was controlled at least once a week. The animal study took place at a certified laboratory animal facility (Medical-Experimental Centre, University of Leipzig, Leipzig, Germany) and all procedures were approved by the local animal usage committees according to German guidelines on animal care and use (Landesdirektion Sachsen, Referat 24, registration number: TVV13/13). At the end of the study on d155 the animals were sacrificed by carbon dioxide inhalation.
After two weeks of acclimatisation, 150 μl pristane were injected intra-dermally into the tail base of the rats [
As a specific marker for arthritis alpha 1-acid glycoprotein (α1AGP) was quantified in the blood plasma of the rats during the acute arthritic phase. Briefly on d21 plasma was prepared from the blood routinely gained from the animals (see
After sacrifice of the animals on d155 the paws were taken for histopathological analysis and fixed overnight in 4% ice-cold formaldehyde. After rinsing the paws in PBS for 24 hours they were decalcified in Osteosoft® for eight weeks. After dehydration and embedding in paraffin, sections of 6 μm thickness were cut from the metacarpus of the forepaws by using a microtome (Leica SM 2010R, Nussloch, Germany) and mounted on SuperFrost Plus® slides (Menzel-Gläser, Braunschweig, Germany) for histological staining. Deparaffinized and rehydrated histological sections were stained as follows: 5 min Nuclear Fast Red (0.1% w/v), 5 min tungstophosphoric acid solution (5% w/v), 8 min Aniline blue—Orange G (0.1% w/V and 0.3% w/v, respectively). Between the single staining steps the sections were always shortly rinsed in distilled water. After the staining they were shortly differentiated in 96% ethanol, followed by dehydration in isopropyl alcohol and coverslipping with Euparal for analysis. Analysis of slices was carried out by means of a Nikon Eclipse TE2000-E microscope (Nikon Instruments Europe B.V., Düsseldorf, Germany) equipped with a Nikon Digital Sight DS-U1 camera (Nikon Instruments Europe B.V., Düsseldorf, Germany). Pictures were captured and processed with NIS-Elements BR 3.2 software (Nikon Instruments Europe B.V., Düsseldorf, Germany).
Every two weeks about 500 μl of blood were taken from the retrobulbar venous plexus of the animals under anaesthesia (see
The direct interaction between EGCG and active MPO redox intermediates was studied by performing stopped-flow kinetic measurements. Thereby an SX-18MV apparatus from Applied Photophysics, Leatherhead, United Kingdom was used to follow the reaction of epigallocatechin (EGC) with preformed Compound I of human MPO. All measurements were performed in 100 mM phosphate buffer, pH 7.0 at 22°C. Briefly, Compound I of the enzyme was pre-formed in the aging loop by incubating native MPO with a 12.5-fold excess of freshly prepared H2O2 for 50 ms [
As illustrated in
While all data correspond to one experiment the experimental groups are displayed in three diagrams for reasons of comprehensibility. PIA was induced on d0 in all female DA rats except the healthy control group (empty squares; n = 5). The application of pristane led to an acute joint swelling followed by a short intermediate recovery and the subsequent development of chronic arthritic symptoms. For each treatment group the mean score of the rats for each scoring day is given with the SEM (Kruskal-Wallis test followed by Dunn’s Post Hoc test).
As also shown in
Surprisingly, upon permanent oral application of EGCG (
For each individual animal a chronification of the pristane-induced arthritis was assumed if the lowest score observed after the abatement of the acute arthritic phase (usually between d40–d50) increased by ≥ 8 points. The suitability of this definition is proven by the fact, that despite one animal from the MTX i.p. late group all animals with acute arthritic symptoms showed a strong decrease in the scoring values between d28 and d61 (not shown). For the named animal only a moderate decrease of the score was observed during this period, which, however, strongly increased during the chronic phase. Thus the criteria defined for a chronification were also fulfilled in this case. As shown in
The data were taken from the animal experiment displayed in
We also checked whether the early i.p. treatment with MTX or the early p.o. treatment with EGCG had an influence on the onset of PIA. Therefore the time for the first appearance of arthritic symptoms (score of one) was determined individually for each animal. From these data the mean day for the disease onset in the single experimental groups was determined. Yet, as illustrated in
As an endogenous arthritis marker the amount of alpha 1-acid glycoprotein (α1AGP) was determined in the blood plasma of the animals on d21 (average maximum of the acute arthritic phase). As shown in
The acute-phase protein was quantified in blood samples taken on d21 from animals of the healthy negative control (no pristane injection, n = 5), the saline-treated positive control (n = 14) and the rats treated with EGCG p.o. starting on d1 (n = 10). For experimental details see
While the induction of arthritic symptoms in the pristane-treated animals was already extensively addressed by applying a clinical scoring system to quantify the swelling and redness of the paws (see.
The displayed pictures are representative examples for tissue sections obtained from healthy animals (A), from the saline-treated control group (B) and from EGCG treated groups (C: EGCG p.o. early and D: EGCG i.p. late). For experimental details regarding arthritis-induction and treatment see
As a marker for the inflammatory state we repeatedly determined the myeloperoxidase activity in the neutrophils of the rats. For this analysis blood was collected from the animals under short-time anaesthesia via punctuation of the retrobulbar venous plexus. The blood samples were heparinised and neutrophils were isolated within 24 h by using a standardised protocol. By applying the HOCl-specific dye APF only the chlorinating activity of the enzyme was addressed during the subsequent flow cytometry analysis. Thereby by analysing the data obtained on d0 (before pristane injection) we observed strongly differing basal MPO activities within the several groups which, however, also emerges from the different sizes of the experimental groups. The averaged fluorescence value obtained by considering all animals (99.8 RFU) was set to 100%. By considering the relative mean value and the relative standard deviation observed in the single experimental groups this value never exceeded 170.75% (healthy control group) or fell below 62.62% (EGCG p.o. late group). These values were set as the limits for the “physiological range” of the chlorinating MPO activity while following the MPO-derived HOCl production in the blood samples during the course of the disease (dashed lines in
For each animal the geometric mean of the APF-dependent fluorescence intensity distribution was determined as a sign for the HOCl-producing MPO activity of the blood neutrophils.
The data obtained during the repeated analysis of blood samples from the single experimental groups until d117 (
The usability of the chlorinating MPO activity as a marker for the inflammatory events taking place during PIA is also illustrated by the blood analysis data obtained from the experimental groups treated with MTX (
In
For (–)-epicatechin (EC) it has been shown that this flavonoid efficiently reacts with both Compound I and Compound II of MPO following the peroxidase cycle of MPO [
All measurements were performed at 22°C in 100 mM phosphate buffer, pH 7.0. Final concentrations of MPO and H2O2 were 2 μM and 25 μM, respectively. Representative spectra from at least three independent experiments are shown. The indicated time points were selected from 50 to 100 recorded spectra. Compound I was pre-formed by incubating the enzyme with a 12.5-fold H2O2 excess for 50 ms.
In order to determine apparent second-order rate constants kinetic measurements at 456 nm and different flavonoid concentrations were performed. As illustrated in
Experimental conditions are given in
The pristane-induced arthritis (PIA) in female
Also the underlying pathological mechanism, i.e. the development of a permanent pro-inflammatory systemic immune status (as reflected by increased α1AGP levels in the blood) leading to relapsing phases of joint swelling and their subsequent destruction (as evaluated by histology), is well reflected by the model. The comparability of PIA in rats to RA in man is also shown by the fact that an early application of methotrexate (MTX), a classical disease-modifying anti-rheumatic drug (DMARD) [
The flavonoid epigallocatechin gallate (EGCG) was already applied in different animal studies on chronic inflammatory diseases including RA [
In our study we were able to show that upon early and continuous oral application of the polyphenol therapeutic effects were achieved which are similar to those of the early injection of MTX: Both in the acute and in the chronic phase of the disease strongly reduced scoring data as well as a markedly decreased degree of chronification were observed. Even a tendency for the later onset of the arthritic symptoms was found. The orally applied EGCG concentration (0.1% w/v) is well in the range with that used in comparable studies [
The assumption that EGCG, upon early oral application, mainly acts via inhibition of the acute immune response induced by pristane injection is also in line with the determination of α1AGP blood levels determined at d21 of the experiment. As already reported by others both in RA patients and in corresponding animal models the hepatic mRNA expression levels as well as plasma levels of this acute phase protein are known to be markedly elevated [
The obtained results suggest, in line with recent studies, that the innate immune system and neutrophils play an important role especially during the onset of RA [
As illustrated by the analysis of blood samples from d0, the basal MPO activity showed strong individual differences. Still it was possible to define a range for the range of its physiological activity. In fact, by repeated analysis of blood samples from the healthy control group (no pristane injection) this range was never exceeded throughout the experiment. In striking contrast to this in the positive control (pristane injection, NaCl treatment) the MPO activity values clearly followed the course of the disease including onset and abatement of the acute phase and beginning of the chronic phase. Within the latter stadium the MPO activity levels seemed to follow a periodic pattern.Thus the HOCl-producing MPO activity nicely reflects the arthritic symptoms observed during the course of the experiment. In fact, expression and plasma levels of the proteins are elevated both in RA patients and in animal models for the disease [
In line with the therapeutic effect observed upon early and continuous oral application of EGCG in this group the determination of the chlorinating MPO activity almost always yielded values which were in the range of the physiological enzyme activity determined on d0. Only on d28 (acute phase) and on d54 (onset of the chronic phase) the values slightly exceeded this physiological range. All other EGCG treatment groups (late oral EGCG treatment and early/late EGCG injection) yielded higher MPO activity values which, especially during the acute phase (d14 and d28) as well as at the onset of the chronic phase (d54), considerably exceeded the range of physiological enzyme activity determined on d0. Still even in these experimental groups where EGCG showed no therapeutic effect the HOCl production rates were lower as compared to the positive control (pristane injection, NaCl treatment) suggesting a certain anti-inflammatory effect of the flavonoid. Still it has to be clearly stated that while the analysis of the chlorinating MPO activity turned out as a good marker to distinguish between healthy animals and untreated animals with PIA (saline-treated positive control) as well as to detect efficient anti-rheumatic treatments (MTX i.p. early and EGCG p.o. early), the HOCl production by MPO turned out to only partially reflect the clinical scoring data observed in less effective treatment groups.
Given the huge role of the innate immune system not only during the onset of RA but also during the chronification of the disease [
An interesting aspect regarding the immunological role of the HOCl production by MPO comes from the fact that several well-known anti-inflammatory compounds (e.g. the plant-derived polyphenol (–)-epicatechin [
The effect of one-electron donors on the chlorinating MPO activity depends on two main factors: (I) The currently prevailing redox state of the enzyme which strongly depends on the actual (patho-)physiological conditions and (II) the reaction rates of potential substrates with Compound I (
As for any animal model the differences in the organisation of the immune system between rodents and humans also limit the comparability between PIA in rats and RA in humans [
In our animal study the early and continuous oral application of EGCG had a strong therapeutic effect. Yet these results are not directly transferable to humans as the applied polyphenol concentration (0.1% w/v) would mean a daily intake of about 500 mg EGCG in man which is at least five times more than the normal dose [
As already stated neutrophils do not only participate in the onset and solidification of RA [
PIA was induced on d0 in all female DA rats except the healthy control group. The data were taken from the animal experiment displayed in
(TIF)
(DOCX)
This work was made possible by funding from the German Federal Ministry of Education and Research (BMBF, 1315883) as well as by the Sächsische Aufbaubank (SAB) project 100116526 from a funding of the European Regional Development Fund (ERDF).
α 1-acid glycoprotein
analysis of variance
aminophenyl fluorescein
complete Freund's adjuvant
collagen-induced arthritis
Dark Agouti
disease-modifying anti-inflammatory drug
dimethyl sulfoxide
epicatechin
epigallocatechin
epigallocatechin gallate
enzyme-linked immunosorbent assay
green tea polyphenol
immune globulin G
interleukine
inducible nitric oxide synthase
intra-peritoneal
macrophage inflammatory protein 1
methotrexate
myeloperoxidase
nuclear cytosolic factor 1
nuclear extracellular traps
non-steroidal anti-inflammatory drug
phosphate buffered saline
pristane-induced arthritis
polymorphonuclear leukocytes
per os (orally
rheumatoid arthritis
relative fluorescence units
systemic inflammatory response syndrome
systemic lupus erythematosus
tumor necrosis factor α