Biochemical and functional characterization of glycosaminoglycans released from degranulating rat peritoneal mast cells:Insights into the physiological role of endogenous heparin
Rebecca Lever a,*,Amir Smailbegovic a,b,Yanira Riffo-Vasquez b,Elaine Gray b,c,
John Hogwood c,Stephen M.Francis d,Neville V.Richardson d,Clive P.Page b,
Barbara Mulloy b,c
a UCL School of Pharmacy,Brunswick Square,London,WC1N1AX,UK
b Sackler Institute of Pulmonary Pharmacology,Institute of Pharmaceutical Science,King's College London,London,SE19NH,UK
c National Institute for Biological Standards an
d Control,Potters Bar,Hertfordshire,EN63QG,UK
d School of Chemistry,University of St.Andrews,St.Andrews,KY169ST,UK
a r t i c l e i n f o
Article history:
Received8June2016
Received in revised form
1October2016
Accepted1November2016 Available online3November2016
Keywords:
Heparin
Mast cells
Dermatan sulphate Glycosaminoglycan a b s t r a c t
The properties of commercially prepared heparin as an anticoagulant and antithrombotic agent in medicine are better understood than is the physiological role of heparin in its native form,where it is uniquely found in the secretory granules of mast cells.In the present study we have isolated and char
acterised the glycosaminoglycans(GAGs)released from degranulating rat peritoneal mast cells. Analysis of the GAGs by NMR spectroscopy showed the presence of both heparin and the gal-actosaminoglycan dermatan sulphate;heparinase digestion profiles and measurements of anticoagulant activity were consistent with thisfinding.The rat peritoneal mast cell GAGs significantly inhibited accumulation of leukocytes in the rat peritoneal cavity in response to IL-1b(p<0.05,n¼6/group),and inhibited adhesion and diapedesis of leukocytes in the inflamed rat cremasteric microcirculation in response to LPS(p<0.001,n¼4/group).FTIR spectra of human umbilical vein endothelial cells(HUVECs) were altered by treatment of the cells with heparin degrading enzymes,and restored by the addition of exogenous heparin.In conclusion,we have shown that rat peritoneal mast cells contain a mixture of GAGs that possess anticoagulant and anti-inflammatory properties.
©2016Published by Elsevier Ltd.
1.Introduction
The glycosaminoglycan(GAG)heparin was discovered a century ago and,as an anticoagulant drug,ranks as one of the most commonly used agents in modern medicine[1].Whilst much is now kno
wn about the nature of commercially prepared pharma-ceutical heparin,both in its unfractionated and low-molecular weight forms,with respect to structure,biological activity and clinical effects[2e4],the physiological role of endogenous heparin is considerably less well understood.It has long been known, however,that heparin possesses additional effects that are both separate to,and separable from,its well-characterized effects on blood coagulation,many of which involve modulation of aspects of immune or inflammatory cell function[5,6].In contrast to the closely related GAG heparan sulphate,the ubiquitous expression of which alone goes some way towards explaining its pivotal role in normal physiology[7,8],mammalian heparin is produced exclu-sively by mast cells.In this respect,heparin has been suggested to be primarily important for the storage of histamine and certain pro-inflammatory granule proteins within the mast cell[9,10].How-ever,it would seem unlikely that a potent anticoagulant molecule having a broad range of biological activities[11]should be bio-synthesized solely for this purpose,and indeed solely within a cell type found outside the vasculature.The localization of mast cells close to vessels of the microcirculation though,as well as their more recent description in pathological tissue sites including tu-mors and atheromatous plaques[12,13],suggests that endogenous heparin may be important in regulation of pathophysiological
*Corresponding author.
E-mail addresses:rebecca.lever@ucl.ac.uk(R.Lever),clive.page@kcl.ac.uk (C.P.
Page).
Contents lists available at ScienceDirect Pulmonary Pharmacology&Therapeutics journal homepa ge:
www.elsevie
r/locate/ypupt
/10.1016/j.pupt.2016.11.002
1094-5539/©2016Published by Elsevier Ltd.
Pulmonary Pharmacology&Therapeutics41(2016)96e102
responses,as well as in normal physiology.It has been suggested that heparin,released from activated mast cells,may be involved in physiological regulation of inflammation[14]through the binding and neutralization of cytotoxic and pro-inflammatory proteins, thus limiting the extent of the inflammatory response and potential tissue damage and remodelling as part of homeostasis.
Many of the non-anticoagulant actions of heparin are mediated through interactions with proteins such as chemokines and growth factors,which often depend upon the binding of heparan sulphate for full activity[15e17].Whilst the structural basis of the antico-agulant activity of heparin is well understood(reviewed in11),the exact structural requirements for the majority of the anti-inflammatory effects of heparin remain to be fully determined. The ability of heparin to interact with a wide variety of proteins can vary from strongly sequence specific,such as the binding of anti-thrombin,to relatively non-specific,in part due to the size and polyanionic nature of the molecule[18,19].In this regard,it is important to consider that commercially-available heparin,which is usually extracted from porcine intestinal mucosa,is standardized only for its anticoagulant activity,which depends heavily on the presence of the high affinity antithrombin-binding pentasaccharide [18].Therefore,any biological activity confined to other poly-saccharide sequences contained within the heparin structure may not necessarily correlate with the total amount of material present in the resultant heterogenous mixture,
and may even be fraction-ated out by the current techniques for preparing heparin commercially as an anticoagulant.A greater understanding of the nature of the GAGs present in mast cells may elucidate the physi-ological role(s)of endogenous heparin and potentially facilitate the design of drugs to mimic specific biological effects of heparin other than anticoagulant activities.
In the present study,therefore,we have sought to examine the nature of GAGs released from peritoneal mast cells of the rat as a product of their degranulation.
2.Experimental procedures
Animals-Male Sprague-Dawley rats(200e250g;Harlan,UK) were housed in an animal unit on a12:12h light:dark cycle,with access to standard laboratory chow and water ad libitum,for at least seven days prior to experimentation.All experiments were per-formed in accordance with local Ethical and UK Home Office approval and guidelines.
Isolation of rat mast cell GAGs(RMCG)-Rats were euthanized by CO2exposure and their peritoneal cavities immediately lavaged with20mL normal saline containing0.05mM EDTA.Recovered cells were washed in modified HBSS(Ca2þ/Mg2þfree)at250Âg for 2min,followed by density-dependent centrifugation to separate mast cells from mononuclear cells.Mast cell pellets were re-suspended in5
mL buffer(PBS containing0.1mg mLÀ1HSA and 5.6mM glucose)and incubated for10min at37 C prior to addition of a further5mL buffer containing5m g mLÀ1compound48/80and incubation for a further20min at37 C.Gross cellular material was removed by centrifugation at150g for10min and discarded.Su-pernatants were then transferred to1mL micro-centrifuge tubes and centrifuged at10,000rpm for15min to sediment intact granules.Supernatants(A)were collected and transferred to a refrigerator and pelleted granules were re-suspended in1mL2M NaCl by vortexing,then incubated at room temperature for30min to facilitate the release of granule contents.The suspension was again centrifuged at10,000rpm for15min,supernatants(B)were collected and dialysed overnight againstfive changes of1.5L dH2O and pellets were discarded.Supernatants A and B were added to poly-L-lysine agarose(Sigma-Aldrich)packed into5ml columns (6mL per columns,pre-washed with3Â3mL2M NaCl),which were capped and placed on a roller for60min at room temperature. Unbound contents were washed with3Â3mL dH2O and bound contents eluted with4Â3mL of1.5M NaCl.Eluents derived from supernatants A and B were combined,dialysed overnight against5 changes of1.5L dH2O and freeze-dried.Average RMCG yield was 0.26mg per106cells,with0.8e1.2Â107cells retrieved per cavity estimated by weight of material.
Heparinase digestion-1mg mLÀ1solutions of the RMCG, unfractionated heparin(5th International Sta
ndard;NIBSC)and heparan sulphate(HS1as previously described[20])were pre-pared,respectively,and each solution treated with10m L heparinase I(approximately0.02IU)from F.heparinum(EC:4.2.2.7)(a kind gift of Leo Pharma,Ballerup,Denmark).Absorbance at234nm was monitored for60min(heparin and heparan sulphate)or120min (mast cell material).
Molecular weight distribution-The molecular weight distribu-tions for RMCG and the USP Heparin Sodium Identification Refer-ence Standard(USP,Rockville,MD,USA)were determined by size exclusion chromatography/gel permeation chromatography(SEC/ GPC)as described in Ref.[21].Briefly,samples were taken up to a concentration of5mg mLÀ1in0.1M ammonium acetate containing 2mg mLÀ1alpha-cyclodextrin as aflowrate marker.Duplicate chromatography runs were performed on a column system con-sisting of TSK SWXL guard column,TSK G4000SWXL and TSK G3000SWXL columns in series,with0.1M ammonium acetate as the mobile phase at0.6mL minÀ1and refractive index detection. The peak molecular weight M p,weight average molecular weight M w,number average molecular weight M n and polydispersity were calculated using Cirrus software(Agilent,Santa Clara,CA,USA).
Anticoagulant activity-Assessment was carried out using two plasma based assays(activated partial thromboplastin time,APTT), using sheep plasma(First Link,UK in accordance with the European Phar
macopoeia(01/2008:20705)or human plasma(NBTS,UK). Purified reagent assays were also carried out to investigate anti-thrombin dependent inhibition of factor Xa and factor IIa activity (USP34NF26)and heparin cofactor II(HCII)dependent inhibition of thrombin.Two heparin preparations,bovine mucosa and porcine mucosa,from the NIBSC panel were included as comparators.All assays used the6th International Standard for Unfractionated Heparin(07/328,NIBSC,UK)as the standard with data analysis carried out using the parallel line bioassay model(Combistats, EDQM).
NMR spectroscopy-RMCG(~5mg)was dissolved in99.8%D2O and transferred to a5mm NMR tube.One dimensional1H and two dimensional TOCSY,and NOESY spectra,were recorded at500MHz, 60 C,using a Varian Unity500NMR spectrometer,with pulse sequences supplied by the manufacturer.Chemical shifts are re-ported relative to deuterated trimethylsilylpropionic acid sodium salt(TSP-d4)(Sigma-Aldrich Ltd.UK)at0ppm.
Effects in in vivo models of inflammation-For peritoneal inflam-matory cell recruitment experiments,rats(as before)were injected i.p.with the RMCG,or an equal volume of vehicle(200m L saline), 30min prior to the administration of20ng rat recombinant interleukin-1b(Sigma-Aldrich Ltd,UK)or vehicle(200m L saline). Animals were euthanized2h later and peritoneal cavities lavaged i
mmediately with20mL saline.Total cells in lavagefluids were counted and differential cell counts were obtained from cytospin preparations,stained using the DiffQuick system(Gamidor,UK).For intravital microscopy of the cremaster muscle,rats were adminis-tered RMCG or saline i.v.immediately prior to s.c injection of25m g LPS to the scrotal sac.Four hours later,animals were anaesthetized with urethane(2mg kgÀ1i.p.).Cremaster muscles were exterior-ized following midline incision and carefully exposed over a transparent viewing area of a heated microscope stage,maintained at37 C,and constantly superfused with Tyrode-HEPES buffer.
R.Lever et al./Pulmonary Pharmacology&Therapeutics41(2016)96e10297
Unbranched,post-capillary venules of30e50m m diameter(!5per animal)were viewed under a Zeiss Axioskop2FS microscope,fitted with a x40water-immersion lens and a x10eye piece.Digital images were captured using an ORCAflash digital camera(Hama-matsu,Japan)attached to an Axio-Workstation computer and im-ages were viewed and recorded for subsequent off-line analysis using IHC acquisition software(Hamamatsu,Japan).Leukocyte rollingflux was quantified as the number of rolling cells passing a fixed point on the venular wall per30s and adherent leukocytes were considered those cells that were stationary for at least30s within a given100m m vessel wall segment.Migrated cells were classed as all cells present within a100m m2area of the surrounding ex
travascular tissue.FTIR spectroscopy of the endothelial glycocalyx-Human umbilical vein endothelial cells(HUVECs;TCS Cellworks Ltd.,U.K.)were cultured to confluency in6-well tissue culture plates(Corning Costar Ltd.,U.K.)at37 C,5%CO2,in medium(MCDB 131)supplemented with fetal bovine serum(2%v/v),hydrocorti-sone(1ng mLÀ1),gentamicin(50m g mLÀ1),amphotericin-B (50ng mLÀ1)and human epidermal growth factor(10ng mLÀ1). Cultures were washed three times with phosphate buffered saline, to remove culture medium,and some wells were incubated with a combination of the enzymes heparinases I,II and III(Sigma-Aldrich Ltd.,U.K.;60min at room temperature,each at0.5IU mLÀ1). Following heparinase treatment,monolayers were washed and some of these wells subsequently received unfractionated heparin (500IU mLÀ1Multiparin®,CP Pharmaceuticals Ltd.,Wrexham,U.K.
e Multiparin is a porcine intestinal mucosal heparin;20min at room temperature)and were washed again.Cells were removed from the plates using a rubber policeman,blotted onto the FTIR crystal and gently dried under nitrogen to remove excess buffer. ATR-FTIR spectra were taken using a6021Galaxy Series spec-trometer(Mattson Instruments Ltd.,U.K.),set at50scans per run.
3.Results
Molecular weight distribution-RMCG was found to be of lower average molecular weight(peak,number and weight averages) than typical porcine unfractionated heparin,but with greater dis-persity(Table1).Moreover,the chromatogram for RMCG exhibits a non-symmetrical peak,which may indicate the presence of more than one distinct population of molecules within the material,in contrast to the symmetrical peak for the heparin standard(Fig.1).
Susceptibility to heparinase digestion-Fig.2shows reaction progress curves for the digestion of equal concentrations of heparin standard,RMCG and heparan sulphate,respectively,by afixed concentration of heparinase I,as measured by the change in absorbance at234nm.RMCG was less susceptible to the actions of this enzyme than standard heparin,both in terms of initial rate and plateau,but more susceptible than heparan sulphate,suggesting the material to be comprised substantially,but not exclusively,of heparin chains.After60min exposure to the enzyme,unsaturated uronic acid generation from the RMCG was approximately half of that from the heparin standard at the same time point,and2.5 times that from heparan sulphate.However,when the reaction time was extended to2h,allowing it to plateau,total generation from the RMCG was increased to approximately60%of that ob-tained from the heparin standard and more than three times that from heparan sulphate.
Anticoagulant activity-The RMCG was found to give a valid po-tency estimation against the porcine unfractionated heparin stan-dard(07/328),although the specific activity of RMCG is lower than porcine and bovine heparin(Table1).The APTT assay using sheep plasma was found to give the highest specific activity of78IU mgÀ1 (Table1).The APTT assay using human plasma and the HCII based
Table1
A:Peak molecular weight M p,number average molecular weight M n,weight average molecular weight M w and polydispersity M w/M n for rat peritoneal mast cell GAGs (RMCG)and for a heparin identity standard,the USP Heparin Sodium Identity Reference Standard(USP ID RS).Results are the means of duplicate determinations. B:Anticoagulant potencies of RMCG,a bovine and porcine heparin sample estimated against6th International Standard for Unfractionated Heparin using a parallel line analysis model.Values are calculated using multiple concentrations of unknown samples against the standard to give IU/mg and95%confidence limits range for the estimated value.
A M p M n M w Polydispersity
(M w/M n)
RMCG7333747112027 1.611
USP ID RS153461271815856 1.247
B Potency in IU/mg
(95%Confidence Limits)
Sheep Plasma APTT Assay Human
Plasma
APTT Assay
anti-Xa
assay
anti-IIa assay HCII assay
RMCG78
76e8161
duplicate part reference57e64
55
54e56
46
44e47
61
54e68
Bovine
Heparin 108
(103e113)
89
(85e93)
94
(91e97)
87
(80e94)
120
(112e129)
Porcine
Heparin 200
(194e206)
185
(175e195)
197
(186e208)
192
(175e211)
199
(186e
214)
Fig.1.A.Size exclusion chromatograms of rat mast cell GAGs(RMCG;peak A)and a
sample of typical unfractionated heparin(the USP Heparin Sodium Identification RS;
peak B).Peak C is theflow-rate marker alpha-cyclodextrin,and peak D is a salt
peak.
Fig.2.Heparinase I digestion profiles of RMCG(A),unfractionated heparin standard
(B)and heparan sulphate(C).In(A)the spectrophotometer was reprogrammed at
60min to extend the data collection period,leading to a gap in the readings at that
time.
R.Lever et al./Pulmonary Pharmacology&Therapeutics41(2016)96e102
98
assay gave similar specific activity,61IU mgÀ1,whilst the two antithrombin dependent assays gave lower activity,anti-Xa55IU mgÀ1and anti-IIa46IU mgÀ1.The ratio of anti-Xa to anti-IIa activity was1.2.
1H-NMR spectra of the RMCG-display signals(Fig.3)that are consistent with the RMCG material being predominantly heparin [22,23],but interestingly containing a substantial proportion, possibly as much as40e50%,of dermatan sulphate[24].Specif-ically,in spite of the rather broad resonances from t
his sample,the TOCSY spectrum of RMCG contains spin systems consistent with the presence of the IdoA residue(2-O-sulfo-a-L-iduronic acid;I2S) and glucosamine residues(a-D-N-sulfoglucosamine-6-O-sulphate; GlcNS,6S)that together make up the major trisulphated disac-charide repeat unit of heparin(Table2).In addition,resonances were apparent in the1H spectrum that correspond to the presence of the galactosamine(b-D-N-acetylgalactosamine-4-O-sulphate; GalNac4S)and non-sulphated IdoA residue(a-L-iduronic acid;I)in the major disaccharide repeat unit of dermatan sulphate(Table2). Comparisons with literature values shown in Table2confirm that the signals attributable to GalNAc4S are characteristic of those in dermatan sulphate[24]rather than chondroitin-4-sulphate[28], and the signals attributable to IdoA are characteristic of those in dermatan sulphate rather than in heparin.
Anti-inflammatory activity-was measured in vivo in the rat in two models of inflammation.Pre-treatment with RMCG inhibited accumulation of leukocytes in the peritoneal cavity in response to the cytokine IL-1b,when both agents were administered i.p. (Fig.4A).In addition,systemic pre-treatment with RMCG inhibited thefirm adhesion and diapedesis of leukocytes in the cremasteric microcirculation,without significantly affecting the number of rolling cells(Fig.4B e D).
FTIR spectroscopy of the endothelial glycocalyx-FTIR spectroscopy of HUVECs yielded spectra indic
ating the presence of sulphated functional groups on the cell surface.Differences were observed in the spectra of untreated endothelial cells when compared with those from enzymatically-treated cells(Fig.5),in that peaks in the window associated with the presence of sulphate groups were abolished following treatment with heparinase enzymes,suggest-ing degradation of heparan sulphate on the endothelial surface. Interestingly,addition of exogenous heparin to enzymatically treated HUVECs led to restoration of the sulphate peak in the spectra of these cells,with introduction of additional peaks found in the spectrum of heparin itself,suggesting the binding of heparin to the cell surface of the endothelium.4.Discussion
A specific physiological role for heparin,over and above that associated with its general characteristics as a GAG,has yet to be defined.Heparin is sometimes considered to be a specialized form of heparan sulphate and,indeed,many of the important protein-binding characteristics of heparan sulphate are known to be shared,and often exceeded,by heparin.In the case of the antico-agulant actions of heparin,this increased potency is known to be due in large part to the relatively frequent occurrence of the antithrombin-binding pentasaccharide sequence that is expressed more rarely in heparan sulphate chains,whereas for other exam-ples of GAG-protein interactions,that are less specific in terms of the GAG structure involved,the significantly greater sulphation density of heparin appears to be the key feature.
In the present study,we have isolated GAGs from degranulating rat peritoneal mast cells in good yield using affinity chromatog-raphy on poly-L-lysine agarose.This technique has the advantage of reducing to a minimum any co-purification of mast cell granule proteins with the GAGs.We have found that endogenous heparin released by degranulating rat peritoneal mast cells is characterised by a relatively low molecular weight profile by comparison with a porcine mucosal heparin reference standard.Free GAGs released from mast cells on degranulation have been depolymerised from their original macromolecular form by mast cell heparanase[25]. The molecular weight distribution of the resulting GAG mixture is characteristic of the source tissue and species;for example,porcine mucosal and bovine lung heparins have consistently different molecular weight profiles[26].Using NMR spectroscopy we have also observed the presence of dermatan sulphate;signals from both IdoA and GalNAc are present,with chemical shift values charac-teristic of dermatan sulphate(Table2),clearly distinguishable from those of unsulphated IdoA in heparin sequences[27]and GalNAc in chondroitin sulphate A[28].The presence of dermatan sulphate in rat peritoneal mast cells has been suggested by comparison of the disaccharide products of digestion with chondroitinase ABC(to which DS is sensitive)with chondroitinase AC(to which DS is not sensitive)[29].Here we confirm the presence of DS by direct spectroscopic observation of the GAG mixture,without degradation or separation of its components.Minor proportions of chondroitin-4-sulphate(CSA),and traces of chondroitin-4,6-sulphate
(CSE) found by Akiyama et al.may be present,but are not visible in our NMR spectra[29].The presence of both heparin and dermatan sulphate chains presumably gives rise to the skewed
molecular Fig.3.A.1H NMR spectrum(500MHz,60 C in D2O)of rat mast cell GAGs.Some of the characteristic resonances of heparin and dermatan sulphate are annotated.
R.Lever et al./Pulmonary Pharmacology&Therapeutics41(2016)96e10299
weight distribution of the rat GAGs (Fig.1);the mode of depoly-merisation of mast cell DS is not currently known to us.
There have been a few previous attempts to look at the release of endogenous rat heparin [30,31],the first of which [30]character-ized the whole serglycin proteoglycan rather than the GAG.Wang and Kovanen have also previously reported the release of endoge-nous heparin from rat mast cells,and studied the ability of this material to inhibit the proliferation of aortic vascular smooth muscle cells,suggesting that the endogenous heparin may be more potent in this regard than an exogenous heparin preparation [31].
We have demonstrated in this study that rat peritoneal mast cell GAGs,consisting of a mixture of heparin and DS,possess signi ficant anti-in flammatory activity in a number of assays in addition to a clear ability to act as an anti-coagulant.The GAG content of mast cells and of other granule-containing cells such as basophils has not been accurately determined thus far,although it has been proposed that in rodents there are two types of mast cell referred to as connective tissue (containing heparin)and mucosal (containing chondroitin sulphate),but only based on histochemical staining [32,
33].It has been previously suggested that heparin is found in mast cell granules whereas chondroitin sulphate is found in baso-phils [34].However,in the present study,the mast cell granule contents we have examined contain both heparin and dermatan sulphate (chondroitin sulphate B),suggesting that both GAGs may occur in the same granule (though it is possible that our peritoneal mast cell preparation includes two separate cell populations).
The time course pro file of heparinase I digestion of the RMCG indicates that roughly 60%of its constituent material is susceptible
Table 2
Assignment of the 1H NMR spectrum of RMCG:chemical shifts in ppm at 500MHz,60 C,in D 2O relative to TSP at 0ppm.
Heparin Dermatan sulphate
GlcNS,6S IdoA2S GalNAc4S IdoA This study
Ref.[22]This study Ref.[22]This study Ref.[24]Ref [28]This study Ref.[24]Ref [27]H1 5.39 5.39 5.21 5.24 4.65e 4.7 4.7 4.60 4.90 4.90 5.04H2 3.27 3.31 4.30 4.37 4.04 4.05 4.03 3.54 3.54 3.7
8H3 3.69 3.70 4.20 4.23 4.02 3.95
4.01 3.92 3.92 4.12H4 3.76 3.78 4.11 4.13 4.65 4.65e 4.7 4.74 4.10 4.11 4.08H5 4.02 4.06 4.80
4.81
3.85 3.8
3.81
4.72
4.72
4.84
H6n.d. 4.37 3.80
3.75e 3.8
3.82H6’n.d.
4.30 3.77GlcNAc CH
2.05
2.05
2.08 2.08
2.02
1In chondroitin 4sulphate at 60deg.C.
2In the heparin-derived sequence IdoA-GlcNS,6S at 40deg.
C.
Fig.4.A.IL-1b -induced (20ng)accumulation of cells in the peritoneal cavity of the rat and the effect of 10m g kg À1RMCG administered locally.Open bars indicate total cell counts and filled bars neutrophil counts (p 0.05vs .IL-1b ,n ¼6/group).B-D.Rolling (B),firmly adherent (*p <0.05vs LPS)(C)and transmigrated (*p <0.001vs LPS)(D)cells in the cremasteric microcirculation of the rat inresponse to 25m g of LPS and the effect of RMCG administered systemically (1and 10m g Kg À1,n ¼4/group).
R.Lever et al./Pulmonary Pharmacology &Therapeutics 41(2016)96e 102
100
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论