- Email: [email protected]
Biochemical and functional assessment of equine lymphocyte phosphodiesterases and protein kinase C
Veterinary Immunology and Immunopathology 98 (2004) 153–165
Biochemical and functional assessment of equine lymphocyte phosphodiesterases and protein kinase C K.J. Rickardsa,*,1, C.P. Pageb, A.S. Hamblinc, N.T. Gooded, F.M. Cunninghama a
Department of Veterinary Basic Sciences, The Royal Veterinary College, Hawkshead Campus, North Mymms, Hertfordshire AL9 7TA, UK b Sackler Institute of Pulmonary Pharmacology, Division of Pharmacology and Therapeutics, GKT School of Biomedical Sciences, King’s College London, Guy’s Campus, London Bridge, London SE1 9RT, UK c Department of Pathology and Infectious Disease, The Royal Veterinary College, Camden Campus, Royal College Street, London NW1 0TU, UK d Department of Veterinary Basic Sciences, The Royal Veterinary College, Camden Campus, Royal College Street, London NW1 0TU, UK Received 22 May 2003; received in revised form 24 October 2003; accepted 2 December 2003
Abstract Lymphocytes play an important role in allergic inflammation and have been implicated in the pathogenesis of equine allergic skin and respiratory disease. Targeting intracellular signalling pathways in human lymphocytes has demonstrated a role for both phosphodiesterase and protein kinase C in cell activation. The aim of this study was to measure total cyclic nucleotide hydrolysing phosphodiesterase activity and to identify the phosphodiesterase and protein kinase C isoenzymes present in equine lymphocytes. The functional significance of these isoenzymes was then investigated by examining their role in peripheral blood mononuclear cell proliferation using isoenzyme selective inhibitors. Total cyclic adenosine monophosphate hydrolysing phosphodiesterase activity was double that of cyclic guanosine monophosphate (30 2 pmol/min mg versus 16 3 pmol/ min mg for cyclic adenosine and cyclic guanosine monophosphate phosphodiesterase activity, respectively). Evidence for the presence of PDE1, 3, 4 and 5 was obtained and PKCa, b, d, e, Z, i, y and z were identified. Selective inhibitors of PDE4, PKCd and conventional PKCs a and b caused significant inhibition of mitogen-induced peripheral blood mononuclear cell proliferation. This study demonstrates a functional role for specific signalling isoenzymes and suggests that, in the context of allergic inflammation, targeting inflammatory cells involved in disease pathogenesis with relevant isoenzyme inhibitors may have therapeutic potential. # 2003 Elsevier B.V. All rights reserved. Keywords: Phosphodiesterase; Protein kinase C; Lymphocytes; Allergic inflammation
1. Introduction *
Corresponding author. Tel.: þ44-207-848-6096; fax: þ44-207-848-6097. E-mail address: [email protected] (K.J. Rickards). 1 Present address: Sackler Institute of Pulmonary Pharmacology, Division of Pharmacology and Therapeutics, GKT School of Biomedical Sciences, King’s College, London, Guy’s Campus, London Bridge, London SE1 9RT, UK.
Lymphocytes have been implicated in the pathogenesis of equine allergic skin and respiratory disease. In ponies susceptible to the seasonally recurrent allergic skin disease, sweet itch, increased numbers of intradermal CD3þ T lymphocytes are found at skin
0165-2427/$ – see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.vetimm.2003.12.001
154
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
sites injected with Culicoides antigen. Additionally ponies with sweet itch have increased numbers of circulating CD4þ T cells (McKelvie et al., 1999). In the allergic respiratory disease, recurrent airway obstruction (RAO), perivascular and peribronchiolar accumulation of lymphocytes has been demonstrated (Winder and von Fellenberg, 1988). A change in T-cell phenotype on exposure to antigen has been suggested by McGorum et al. (1993) who found an increase in the CD4þ:CD8þ T lymphocyte ratio after challenge in horses susceptible to RAO. More recently Lavoie et al. (2001) reported a predominant Th2-type cytokine profile in BAL samples from horses with RAO. In human allergic diseases such as asthma and atopic dermatitis, alteration in lymphocyte phenotype resulting in over expression of Th2-type cytokines has been suggested to be important in disease pathogenesis (Chan et al., 1993; Kirman and Le Gros, 1998). Thus interest has centred on intracellular signalling mechanisms involved in lymphocyte activation in order to develop novel compounds capable of attenuating lymphocyte function. There is considerable evidence to suggest that the second messenger protein kinase C (PKC) plays an essential role in T-cell activation having both stimulatory and regulatory effects (Szamel and Resch, 1995). PKC exists as a family of serine/threonine-specific isoenzymes of which 10 have been identified. The isoenzymes can be divided into three groups: (1) conventional (a, bI, bII and g); (2) novel (d, E, Z and y); (3) atypical (i and z) based on their structure, localisation and substrate requirements (Way et al., 2000). Inhibitors have been developed with some degree of isoenzyme specificity and in vitro studies using human cells have shown them to attenuate lymphocyte proliferation (Birchall et al., 1994; Hassan et al., 1997). Another signalling pathway that has been investigated in relation to T-cell function involves the second messengers, cyclic adenosine and guanosine monophosphate (cAMP and cGMP, respectively). These molecules are thought to be important in regulating inflammatory cell function and intracellular levels are maintained by G-protein linked cyclases that increase production and a growing family of enzymes termed phosphodiesterases (PDE) that break down the cyclic nucleotides. Like PKC, PDEs exist in multiple forms with 11 isoenzymes having been characterised based on substrate specificity and sensitivity to endogenous
and exogenous regulators (PDE1–11) (Manganiello et al., 1995; Soderling et al., 1998a,b, 1999; Fawcett et al., 2000). The three predominant isoenzymes present in human T lymphocytes are PDE3, 4 and 7 (Giembycz et al., 1996) and in vitro studies using selective PDE4 and non-selective inhibitors have demonstrated inhibition of IL-2 biosynthesis, proliferation, adhesion, cytokine gene expression and chemotaxis (Giembycz et al., 1996; Essayan et al., 1997; Gonzalez-Amaro et al., 1998; Briggs et al., 1999; Hidi et al., 2000). A PDE4 selective inhibitor has also been shown to reduce IgE synthesis by B-lymphocytes (Cooper et al., 1985), which contain predominantly PDE4 and PDE7 (Gantner et al., 1998) and an antisense oligonucleotide directed against PDE7 has been shown to regulate T-cell function (Li et al., 1999). At present there is no information on the PDE and PKC isoenzyme profiles or functional significance of the isoenzymes identified in equine lymphocytes. In this study, we initially characterised the isoenzymes present in lymphocytes isolated from healthy ponies. Following on from this the effects of selective PDE isoenzyme inhibitors and three PKC inhibitors, one selective for PKCd, one for conventional PKCs and one non-selective, were investigated for their ability to attenuate equine mononuclear cell proliferation.
2. Materials and methods 2.1. Animals Six healthy New-Forest ponies (age range 10–20 years) were used for the studies. The animals were kept at pasture and treated routinely with anthelminthics and tetanus vaccination. 2.2. Lymphocyte purification Venous blood (50 ml) was collected into 0.4 M ethylenediaminetetraacetic acid (EDTA). Peripheral blood mononuclear cells (PBMC) were separated by Percoll density gradient centrifugation (Drummer et al., 1996). PBMCs were harvested from the Percoll–plasma interface, washed twice in Hank’s balanced salt solution without Ca2þ and Mg2þ and re-suspended at 5 106 cells/ml in RPMI 1640 containing 2 mM L-glutamine and 10% heat inactivated
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
foetal calf serum. Cells were cultured in tissue culture flasks for 1 h at 37 8C in 5% CO2 after which nonadherent cells were removed, counted and re-suspended at 8 106 cells/ml in 0.2% bovine serum albumin (BSA) in phosphate buffered saline (PBS). Purity of the lymphocyte population was assessed using flow cytometry as previously described (Siedek et al., 1997).
155
described by Lowry et al. (1951). Aliquots containing approximately 1 106 homogenised cells were assayed in the presence of isoenzyme selective inhibitors for PDE1, 3, 4 and 5 and cGMP, which stimulates PDE2. All compounds were added at a single concentration at least 10-fold higher than the documented IC50 for inhibition of the purified isoenzyme in other species whilst still conferring selectivity (Nicholson and Shahid, 1993; Tenor et al., 1996).
2.3. Flow cytometry 2.5. Identification of PKC isoenzymes Briefly, purified cells were screened using a panel of monoclonal antibodies that recognise equine cells expressing CD3, CD4, CD8, CD13, MHC class I, MHC class II and immunoglobulin. Isotype-matched negative controls were also included. Cells, 4 106 per well, were incubated on ice for 30 min in 96-well microtitre plates with optimal predetermined dilutions of primary antibody (undiluted for CD4 and CD8, 1:2 dilution for CD13, 1:10 dilution for CD3, MHC class I and II, 1:100 dilution for immunoglobulin). Following three washes with 0.2% BSA in PBS, FITC-conjugated rabbit anti-mouse immunoglobulin F(ab)2 was added to each well (1:50 dilution in 0.2% BSA in PBS). After 30 min incubation on ice, cells were washed three times as before and kept in 1% paraformaldehyde in PBS at 4 8C until fluorescence was measured on an FACS analyser (Becton Dickinson, Oxford, UK) with Lysys II software. Cells were gated according to their forward and side scatter and analysed for the proportion of FITC-stained cells after correcting for non-specific staining using the appropriate negative control. 2.4. Measurement of PDE activity Non-adherent lymphocytes were lysed in homogenisation buffer (leupeptin 10 mg/ml, Tris (pH 8.0) 20 mM, MgCl2 2 mM, dithiothreitol 1 mM, EDTA 5 mM, benzamidine 1.3 mM, sucrose 0.25 M, Na-ptosyl-L-lysine chloromethyl ketone 20 mM, triton X-100 1%; v/v), vortexed and stored at 80 8C prior to analysis. Phosphodiesterase activity was measured using a two-step radioactivity assay as previously described (Rickards et al., 2000) and expressed as picamoles hydrolysed cyclic nucleotide per minute per milligram protein. The protein content of the samples was measured using the colorimetric method
Western blotting was carried out as previously described (Laemmli, 1970). Briefly, 1 107 cells were solubilised by boiling for 5 min in 1 ml sodium dodecyl sulphate (SDS; 4%) buffer containing 125 mM Tris pH 6.8, 10% glycerol, 50 mM dithiothreitol and 25 mg/ml bromophenol blue and stored at 20 8C. Cell proteins from 30 ml aliquots were separated by SDS-polyacrylamide gel electrophoresis (PAGE) using a 9% polyacrylamide resolving gel and 5% stacking gel and transferred electrophoretically onto a polyvinylidene difluoride membrane. Membranes were blocked for 2 h with 5% (w/v) skimmed milk powder in Tris-buffered saline, pH 7.5, and then probed with PKC isoenzyme-specific antibodies (1:250 dilution in Tris-buffered saline for b, i and y; 1:200 dilution for all other antibodies), which have previously been validated in equine granulocytes (Greenaway et al., 2000). After washing, membranes were incubated with species-specific horseradish peroxidase-linked anti-IgG secondary antibodies (1:5000 dilution in 0.05% (w/v) milk powder in Tris-buffered saline) and immunoreactive bands visualised using the enhanced chemiluminescence detection system (Amersham International, Buckinghamshire, UK). The amounts of individual PKC isoenzymes were estimated; semi-quantitatively by densitometric analysis corrected for protein content in each sample by densitometric analysis of SDS-PAGE separated proteins stained with Brilliant Blue G-Colloid (Sigma Diagnostics, Dorset, UK). 2.6. PBMC proliferation Proliferation was measured using the method described by McKelvie et al. (1998). Briefly, PBMCs were separated from 50 ml venous blood collected
156
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
into heparin (20 i.u./ml) as described above, counted and re-suspended at 1 106 cells/ml in RPMI 1640 containing 10% autologous horse plasma, 2 mM L-glutamine, 50 mg/ml streptomycin and 50 i.u./ml penicillin. Cells ð5 105 Þ were incubated with a range of concentrations of phytohaemagglutinin (PHA; 3.5–225 mg/ml) for 72 h at 37 8C in 5% CO2 at which point ½3 H-thymidine (0.8 mCi) was added and, after a further 18 h incubation, cells were harvested onto glass fibre filters using a cell harvester (Ilacon, Tonbridge, Kent, UK). The filters were placed in scintillation fluid and radioactivity counted in a liquid scintillation counter (Tri-CarbTM 2500TR, Canberra Packard, Berkshire, UK). Proliferation was then measured in the presence of the PDE4 inhibitor, rolipram (108 to 104 M), the PDE5 inhibitors, zaprinast (107 to 104 M) and sildenafil (108 to 2 106 M), the PDE3 inhibitors, quazinone (108 to 105 M) and milrinone (107 to 104 M), a combination of milrinone (105 M) and zaprinast (105 M), the non-selective PKC inhibitor, Ro-31-8220 (5 109 to 105 M), the conventional PKC inhibitor, Go¨ -6976 (5 1010 to 106 M), or the PKCd inhibitor, rottlerin (108 to 105 M) using the concentration of PHA that produced 90% maximal proliferation (28 mg/ml). All compounds were dissolved in DMSO and serial dilutions were made in RPMI 1640 containing 2 mM L-glutamine, 50 mg/ml streptomycin and 50 i.u./ml penicillin to ensure that the final concentration of DMSO in contact with the cells was 0.1% with the exception of the top concentrations of sildenafil, Ro-31-8220 and Go¨ -6976 and the combination of milrinone and zaprinast where the final concentration of DMSO was 0.2%. Appropriate DMSO controls were included in the study. 2.7. Materials Sodium heparin was obtained from National Veterinary Supplies, Staffordshire, UK. 3 H-cAMP, 3 HcGMP and species-specific horseradish peroxidaselinked secondary antibodies were obtained from Amersham Pharmacia Biotech, Buckinghamshire, UK. 3 H-thymidine was obtained from ICN Pharmaceuticals Ltd., Hampshire, UK. PHA was obtained from Biostat, Derbyshire, UK. RPMI 1640 and penicillin-streptomycin (5000 i.u./ml to 5000 mg/ml) were obtained from Invitrogen, Renfrewshire, UK.
Methanol was obtained from Merck Ltd., Leicestershire, UK. Optiphase Hisafe 8 was obtained from Fisher Scientific, Leicestershire, UK. Ecoscint O was obtained from National Diagnostics, Yorkshire, UK. Rolipram was kindly provided by Schering AG, Berlin, Germany. Zaprinast, milrinone, quazinone, vinpocetine, rottlerin, Ro-31-8220 and Go¨ -6976 were obtained from Calbiochem, Nottinghamshire, UK. Foetal calf serum was obtained from Harlan SeraLab, Leicestershire, UK. Sildenafil was kindly provided by Pfizer, Kent, UK. FACS isotype negative controls and antibodies recognising CD13, MHC class I and II were obtained from Serotec Ltd., Oxfordshire, UK. The antibody recognising B cells was obtained from VMRD Inc., Washington, USA. The antibody recognising CD3 was kindly provided by Prof. J. Stott, University of California, CA. The antibody recognising CD4 was kindly provided by Dr. M. Holmes, Cambridge, UK. The antibody recognising CD8 was kindly provided by S. Mayall, Royal Veterinary College, London, UK. The FITC-conjugated rabbit anti-mouse immunoglobulin F(ab)2 was obtained from DAKO Ltd., Cambridgeshire, UK. Rabbit polyclonal PKC a, d, E, g, Z and z antibodies were obtained from Santa Cruz Biotechnology, California, USA. Mouse monoclonal PKC b, i and y antibodies were obtained from Transduction Laboratories via Affiniti Research Products Ltd., Devon, UK. All other reagents were obtained from Sigma Diagnostics, Dorset, UK. 2.8. Data analysis All data are expressed as mean S:E:M. To calculate lymphocyte purity following separation, the total number of CD3 positive and immunoglobulin positive cells were expressed as a percentage of the number of cells staining positive for MHC class I. PBMC proliferation was expressed as a stimulation index by calculating mean decays per million (dpm) in the presence of PHA and dividing by mean dpm in the presence of medium alone. The PDE profile and PBMC proliferation data were analysed by repeated measures one-way ANOVA (GraphPad Prism3, GraphPad Software, San Diego, USA) and any significant differences further investigated by Dunnett’s post hoc test where the control data was either total PDE activity or PBMC proliferation in the presence of
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
DMSO. Differences were considered significant when a P value of 5% or less was found.
3. Results 3.1. Lymphocyte purification FACS analysis of the cell population following lymphocyte separation demonstrated that 91 5% stained positive for lymphocyte markers. Of the lymphocyte gated population, 7 1% were B cells, 62 3% were CD4þ and 21 3% were CD8þ lymphocytes. Contaminants staining positive for CD13 comprised 7 2% of the cell population. Although no marker for platelets was included, microscopic analysis of a modified Wright’s-stained smear under high power showed no visible evidence of platelets. 3.2. PDE isoenzyme analysis The PDE isoenzyme profile of equine lymphocytes is shown in Fig. 1. All the isoenzyme selective inhi-
157
bitors reduced PDE activity indicating that equine lymphocytes contain PDE1, 3, 4 and 5. Addition of cGMP also inhibited cAMP PDE activity providing further evidence for the presence of PDE3 as this isoenzyme is inhibited by cGMP. Lymphocyte cAMP PDE activity was approximately double that of cGMP PDE activity (30 2 pmol hydrolysed cAMP/min mg versus 16 3 pmol hydrolysed cGMP/min mg). 3.3. PKC isoenzyme analysis Western blot analysis of the PKC isoenzymes present in equine lymphocytes showed that all but PKCg were expressed (Fig. 2). This isoenzyme was identified in the control sample (horse cerebrum) demonstrating the ability of the monoclonal antibody to recognise equine PKCg (data not shown). All the bands found at the correct molecular weight for the relevant PKC isoenzyme appear as a single band except for PKCd and PKCz, which appear as doublets. In addition to these primary bands, all antibodies except those directed against PKCE and Z showed further bands at lower molecular weights although these were always of a lower intensity than the main band with the exception of lanes 3 and 4 for PKCb where all the bands identified were of a similar intensity. Where these bands correspond to a 50 kDa product, which is likely to represent a fragmentation product (Kishimoto et al., 1983), this molecular weight is marked. In all other cases the bands fall between 50 and 75 kDa. 3.4. PBMC proliferation
Fig. 1. cAMP and cGMP PDE isoenzyme profile of equine lymphocytes. Filled columns represent total PDE activity. Vinpocetine (1 104 M) inhibits PDE1, cGMP (5 106 M) stimulates PDE2 and inhibits PDE3, quazinone (6 105 M) inhibits PDE3, rolipram (1 105 M) inhibits PDE4 and zaprinast (1 105 M) inhibits PDE5. Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between responses in the presence of inhibitors and the relevant control are indicated by an asterisk (*P < 0:05).
Equine PBMCs proliferated in response to PHA giving a characteristic bell-shaped curve. Maximum proliferation occurred with 56 mg/ml PHA (mean stimulation index ¼ 87 29), whilst 28 mg/ml resulted in 90% of maximum response. This concentration was used in the subsequent inhibitor studies. Addition of the PDE4 inhibitor, rolipram, resulted in concentration-dependent inhibition of PHA-induced PBMC proliferation with significant effects occurring at concentrations of 107 M and above (Fig. 3a). Following addition of either milrinone or quazinone, both PDE3 inhibitors, proliferation was also significantly attenuated (Fig. 3b and c) with the effect occurring at a lower concentration in the presence of quazinone (106 M). Both inhibitors produced a similar amount of
158
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Fig. 2. Western blot identification of the PKC isoenzymes in equine lymphocytes (approx. 1 107 cells/ml). Arrows indicate molecular weight markers. Lane 1 corresponds to horse brain control and lanes 2–7 correspond to 30 ml lymphocyte aliquots where each lane represents cells from the same horse on consecutive blots.
inhibition at 105 M (24 4 and 16 5% for milrinone and quazinone, respectively). Only milrinone could be investigated at 104 M where inhibition of a similar magnitude to that caused by rolipram occurred (72 3 and 79 3%, respectively). Proliferation was also significantly reduced by zaprinast, a PDE5 inhibitor, when it was added at a concentration of 105 M, producing a similar amount of inhibition to the PDE3 inhibitors at this concentration (Fig. 3d). At 104 M, proliferation was inhibited by 69 2%. Another PDE5 inhibitor, sildenafil, was tested as this compound has been shown to be a more selective and potent inhibitor of PDE5 (Ballard et al., 1998). In the presence of concentrations ranging from 108 to 2 106 M, sildenafil produced a small inhibitory effect that was independent of concentration (12 1% versus 10 5% at 108 M and 2 106 M, respectively; Fig. 3e) and was statistically significant at all concentrations except the highest tested. When a combination of 105 M milrinone and 105 M zaprinast was added the amount of inhibition achieved appeared to be simply additive ð45 3%Þ. Ro-31-8220 (non-selective PKC inhibitor), Go¨ 6976 (inhibitor of conventional PKCs) and rottlerin (inhibitor of PKCd) attenuated PHA-induced PBMC
proliferation in a concentration-dependent manner (Fig. 4a–c). At the two highest concentrations tested, Ro-31-8220 caused complete abrogation of the proliferative response whilst the inhibitory effect of rottlerin and Go¨ -6976 was only partial at the top concentration tested.
4. Discussion The aim of this study was to identify equine lymphocyte PKC and PDE isoenzyme profiles as the initial step in establishing expression and roles for individual isoenzymes in modulating lymphocyte function in the horse. FACS analysis of the cell population isolated from equine blood demonstrated 91% purity with 2% granulocyte and 5% monocyte contamination. Drummer et al. (1996) using an identical cell separation procedure produced a 77% pure population, which may reflect differences in antibody binding affinity during FACS analysis. The percentages of lymphocyte gated cells staining positive for CD4 and CD8 were similar to those documented by others when screening peripheral blood collected from healthy horses
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
159
Fig. 3. Effect of: (a) rolipram (108 to 104 M), (b) quazinone (108 to 105 M), (c) milrinone (106 to 104 M), (d) zaprinast (108 to 104 M) and (e) sildenafil (108 to 2 106 M) on PHA-induced PBMC proliferation. Results are expressed as % of the stimulation index in the presence of PHA (28 mg/ml) alone (mean stimulation index ¼ 76 20 for (a)–(d); and 161 13 and 152 14 for 0.1 and 0.2% DMSO, respectively, for (e)). Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between control and drug treatment are indicated by an asterisk (*P < 0:05).
(McGorum et al., 1993; Watson et al., 1997) indicating that Percoll density gradient centrifugation does not influence the composition of the lymphocyte population. In addition, the separation technique proved to be highly reproducible with an inter-assay coefficient of variation of 4.5%.
In these studies, PDE and PKC isoenzyme expression were measured in the total lymphocyte population. Although it is possible to separate equine lymphocyte subtypes using magnetic cell separation (Kydd et al., 1994), yields of purified equine T and B cells are likely to be low. The data obtained in these
160
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Fig. 4. Effect of: (a) Ro-31-8220 (5 109 to 105 M), (b) Go¨ 6976 (5 1010 to 106 M) and (c) rottlerin (108 to 105 M) on PHA-induced PBMC proliferation. Results are expressed as % of the stimulation index in the presence of PHA (28 mg/ml) alone (mean stimulation index ¼ 157 24 and 158 20 for 0.1 and 0.2% DMSO, respectively). Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between control and drug treatment are indicated by an asterisk (*P < 0:05).
studies of the total lymphocyte population will enable further studies to focus on the relevant isoenzymes when investigating individual lymphocyte subsets. Using isoenzyme selective inhibitors, evidence for the presence of PDE1, PDE3, PDE4 and PDE5 was obtained in equine lymphocytes, which is similar to the isoenzyme profile of human lymphocytes (Tenor et al., 1995; Giembycz et al., 1996; Gantner et al., 1997, 1998). The greatest inhibition of cAMP PDE activity occurred in the presence of the PDE4 inhibitor, rolipram, suggesting that PDE4 is the predominant isoenzyme. This is in agreement with findings in human lymphocytes (Tenor et al., 1995; Giembycz et al., 1996; Gantner et al., 1997) and indeed leucocytes from several species including equine neutrophils (Rickards et al., 2000). It should be noted that there is currently no PDE7 selective inhibitor available, thus it was not possible to investigate PDE7 expression in equine cells. Previously, it has been shown that PDE5 activity in preparations of mononuclear cells originated from platelet contamination (Epstein and Hachisu, 1984). Platelet contamination of Percoll-separated equine lymphocytes was minimal suggesting that the observed PDE5 activity was likely to be of lymphocyte origin. Analysis of PKC isoenzyme expression demonstrated that all but PKCg were present in equine lymphocytes. PKCg was, however, present in the positive control sample of horse cerebrum indicating that, as shown previously (Greenaway et al., 2000), the monoclonal antibody used recognised equine PKCg. Detection of all the other isoenzymes agrees with the profile in human T lymphocytes (Baier et al., 1993; Fulop et al., 1995) and suggests that, as with human cells, individual isoenzymes have distinct functional roles with respect to lymphocyte activation (Wilkinson and Nixon, 1998). In human cells PKCb has been shown to be the predominant isoenzyme (Fulop et al., 1995). Purified equine PKC isoenzymes are not available and in the current study, it was not therefore possible to determine the relative amounts present as antibody avidity and affinity influence the density of the observed bands. However, if data are expressed as ratios of one PKC isotype, for which PKCa was chosen in view of its ubiquitous expression (Nishizuka, 1988), an estimate of the relative amounts of each isoenzyme for individual animals can be calculated (Table 1).
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
161
Table 1 PKC isoenzyme ratios calculated from optical density measurements following correction for protein content of the samplesa a b d E Z i y z OD a
1 0.6 2.2 2.0 1.0 0.8 1.3 1.4 10.0
1 0.7 2.7 8.3 1.0 1.9 3.0 2.8 1.8
1 0.7 2.3 7.1 0.9 2.5 3.5 3.5 1.5
1 0.5 1.7 2.0 0.8 1.1 1.6 1.4 9.1
1 0.5 1.8 1.6 0.6 1.0 1.3 1.4 8.0
1 0.5 1.8 1.6 0.6 1.1 1.4 1.3 9.8
a Each column corresponds to an individual animal. The ratios are expressed in terms of the amount of PKCa that is present in each sample. The bottom row represents the PKCa optical density (OD), measured in arbitrary units, for each animal following correction for protein content of the samples.
Such an approach suggests considerable inter-animal variability in isoenzyme expression. All cell samples were analysed at the same time in the same way and no fractionation was undertaken. Thus changes in expression as a result of translocation or downregulation during sampling or cell processing cannot explain the observed variability. However, it should be noted that, in addition to the predominant band identified at the known molecular weight for the relevant isoenzyme, secondary bands of lower molecular weight were identified for most of the isoenzymes, which are likely to represent fragmentation products (Kishimoto et al., 1983; Dekker et al., 1993). In addition PKCd and PKCz separate out as doublets suggesting that these two isoenzymes are present in different phosphorylation states (England and Rumsby, 2000). As mentioned previously, with the exception of PKCb in two of the six ponies, the relative intensity of the secondary bands and doublets correlates well with that of the corresponding main band for each individual animal suggesting that modification is a consistent phenomenon. However, the possibility remains that some of the inter-animal variability in isoenzyme expression occurs as a result of differences in PKC processing during cell separation. This study confirmed earlier findings that equine PBMCs proliferate in response to the mitogen, PHA, producing a characteristic bell-shaped curve (McKelvie et al., 1998). The PDE4 inhibitor, rolipram, caused concentration related inhibition of the proliferative response, which reached significance at 1 107 M. This concentration is lower than the documented IC50 for inhibition of PDE4 suggesting that the signalling pathway for mitogen-induced PBMC proliferation is
cAMP-dependent and that cAMP levels are controlled by PDE4. The inhibition produced by rolipram was similar to that seen in human PBMCs (Landells et al., 2001) and may be due to a direct effect on T lymphocytes (Betz and Fox, 1991; Giembycz et al., 1996; Essayan et al., 1997) or may involve indirect effects on accessory cells such as monocytes (Banner et al., 1995). Although rolipram has previously been shown to decrease cell viability as inhibition of proliferation increases, Banner et al. (1996) found that at any given concentration of rolipram the anti-proliferative effect was much greater than the reduction in cell viability. Thus the effects of rolipram in the present study are unlikely to be entirely attributable to a reduction in cell viability. Quazinone and milrinone only produced significant inhibitory effects at concentrations 2- and 30-fold higher than their documented IC50’s for PDE3 inhibition, respectively (Calbiochem, Nottinghamshire, UK data sheets). This suggests that PDE3 is either not involved, or is of limited importance in, regulating the cAMP-dependent steps in the signalling pathway for PBMC proliferation, which is in agreement with human studies (Giembycz et al., 1996). It is possible that the small inhibitory effect seen at 1 105 M may have been due to inhibition of monocyte function, although selective inhibition of PDE3 in human monocytes had no effect on TNFa release (Landells et al., 2000). Moreover, the effects of higher concentrations of milrinone may be due to inhibition of PDE4 as the IC50 is 2 105 M (Calbiochem, Nottinghamshire, UK data sheet). As with all the compounds used that produced inhibitory effects, further work is required to identify the precise cellular target for their action.
162
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Involvement of cGMP-dependent protein kinase in inhibition of human PBMC proliferation has been documented using cGMP analogues and direct activators of guanylyl cyclase (Fischer et al., 2001). Therefore, the effect of the PDE5 inhibitor, zaprinast, on equine PBMC proliferation was investigated. Significant inhibitory effects were only seen at concentrations higher than the documented IC50 for PDE5 inhibition (Nicholson and Shahid, 1993). However, it should be noted that this compound has been reported to show poor cell penetration (Nicholson and Shahid, 1993) thus the intracellular concentration may have been lower than required for PDE5 inhibition. In order to further investigate whether inhibition of PDE5 was functionally significant in equine PBMC proliferation, the effect of the more potent and selective PDE5 inhibitor, sildenafil, was investigated (Ballard et al., 1998). An inhibitory effect was observed using sildenafil, which, although small, was significant but unrelated to concentration. Although no data could be found on the effect of sildenafil on PBMC proliferation in other species, the compound has been shown to inhibit proliferation of human cultured smooth muscle cells with an IC50 of 2 106 M (Adolfsson et al., 2002). Solubility precluded investigation of concentrations above 2 106 M in equine PBMC proliferation. Thus, use of sildenafil did not provide conclusive evidence of PDE5 involvement in equine PBMC proliferation and it remains possible that the effect of zaprinast is due to inhibition of PDE1 (Nicholson and Shahid, 1993). Another possibility is that elevating cGMP has an indirect effect via a reciprocal increase in cAMP due to inhibition of PDE3. To test this hypothesis, a combination of zaprinast and milrinone was examined with both inhibitors being added at a concentration that caused a small but significant effect individually (1 105 M). However, the inhibitory effect obtained was additive, rather than synergistic, suggesting that the two inhibitors are working independently. Therefore, it appears that the signalling pathway for equine PBMC proliferation is under both cAMP- and cGMP-dependent regulation. Addition of the non-selective PKC inhibitor, Ro-318220, significantly inhibited proliferation at all concentrations tested except 5 108 M. The lack of effect at this concentration is difficult to explain as a 10-fold lower concentration did cause inhibition. However, as most of the PKC isoenzymes are present
in equine lymphocytes, it is possible that different isoenzymes, some of which may be regulatory, are inhibited at different concentrations of Ro-31-8220. Higher concentrations caused complete inhibition demonstrating the importance of PKC in the signalling pathways involved in lymphocyte proliferation, which is similar to its effect in human lymphocytes (Fraser et al., 1993). However, it is possible that at the higher concentrations some of the inhibitory effect on proliferation may have been a result of increased apoptosis (Han et al., 2000). Although no information is available on the concentration of PKC inhibitors required for selective isoenzyme inhibition in equine cells, the results obtained do suggest that PKCd is of functional importance as approximately 30% inhibition of proliferation was achieved in the presence of rottlerin at the concentration documented to be the IC50 for inhibition of PKCd (Way et al., 2000). However, it should be noted that rottlerin also inhibits PKCy and this isoenzyme has been shown to be involved in human T-cell signalling (Altman et al., 2000). The conventional PKC inhibitor, Go¨ -6976, also caused inhibition of the proliferative response at a concentration similar to the reported IC50’s for PKCa and b (2 109 and 6 109 M, respectively) suggesting a role for one or both of these isoenzymes (Martiny-Baron et al., 1993). This is in agreement with the findings of Hassan et al. (1997) who found that in human CD4þ T-cells, Go¨ 6976 (20 mM) completely inhibited anti-CD2þCD28induced intracellular IL-2 accumulation. As mentioned earlier the use of PBMCs precludes evaluation of the precise target of PKC inhibition in the proliferation signalling cascade. Indeed, rottlerin has been shown to inhibit IL-1b production in PMA- and LPSstimulated human monocytes (Kontny et al., 2000) suggesting that accessory cells may be targeted in addition to lymphocytes. In conclusion, equine lymphocytes have been shown to contain several PKC and PDE isoenzymes. As in other species these isoenzymes appear to have distinct functional roles with PDE4, PKCa and/or b and PKCd and/or y being involved in equine PBMC proliferation. It will be of interest to investigate the importance of PDE and PKC isoenzymes in other lymphocyte responses such as cytokine production of direct relevance to allergic respiratory and skin disease pathogenesis.
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Acknowledgements The authors would like to thank the owners who kindly allowed their animals to be used in this study and the Horserace Betting Levy Board for their financial support.
References Adolfsson, P.I., Ahlstrand, C., Varenhorst, E., Svensson, S.P., 2002. Lysophosphatidic acid stimulates proliferation of cultured smooth muscle cells from human BPH tissue: sildenafil and papaverin generate inhibition. Prostate 51, 50–58. Altman, A., Isakov, N., Baier, G., 2000. Protein kinase Ctheta: a new essential superstar on the T-cell stage. Immunol. Today 21, 567–573. Baier, G., Telford, D., Giampa, L., Coggeshall, K.M., BaierBitterlich, G., Isakov, N., Altman, A., 1993. Molecular cloning and characterization of PKC theta, a novel member of the protein kinase C (PKC) gene family expressed predominantly in hematopoietic cells. J. Biol. Chem. 268, 4997–5004. Ballard, S.A., Gingell, C.J., Tang, K., Turner, L.A., Price, M.E., Naylor, A.M., 1998. Effects of sildenafil on the relaxation of human corpus cavernosum tissue in vitro and on the activities of cyclic nucleotide phosphodiesterase isozymes. J. Urol. 159, 2164–2171. Banner, K.H., Bertin, B., Moodley, I., Page, C.P., 1996. Effect of isoenzyme selective phosphodiesterase inhibitors on the proliferation of murine thymus and spleen cells. Pulm. Pharmacol. 9, 35–41. Banner, K.H., Roberts, N.M., Page, C.P., 1995. Differential effect of phosphodiesterase 4 inhibitors on the proliferation of human peripheral blood mononuclear cells from normals and subjects with atopic dermatitis. Br. J. Pharmacol. 116, 3169–3174. Betz, M., Fox, B.S., 1991. Prostaglandin E2 inhibits production of Th1 lymphokines but not of Th2 lymphokines. J. Immunol. 146, 108–113. Birchall, A.M., Bishop, J., Bradshaw, D., Cline, A., Coffey, J., Elliott, L.H., Gibson, V.M., Greenham, A., Hallam, T.J., Harris, W. et al., 1994. Ro 32-0432, a selective and orally active inhibitor of protein kinase C prevents T-cell activation. J. Pharmacol. Exp. Ther. 268, 922–929. Briggs, W.A., Wu, Q., Orgul, O., Choi, M., Scheel Jr., P.J., Burdick, J., 1999. Lymphocyte suppression by rolipram with other immunosuppressive drugs. J. Clin. Pharmacol. 39, 794–799. Chan, S.C., Li, S.H., Hanifin, J.M., 1993. Increased interleukin-4 production by atopic mononuclear leukocytes correlates with increased cyclic adenosine monophosphate-phosphodiesterase activity and is reversible by phosphodiesterase inhibition. J. Invest. Dermatol. 100, 681–684. Cooper, K.D., Kang, K., Chan, S.C., Hanifin, J.M., 1985. Phosphodiesterase inhibition by Ro 20-1724 reduces hyperIgE synthesis by atopic dermatitis cells in vitro. J. Invest. Dermatol. 84, 477–482.
163
Dekker, L.V., McIntyre, P., Parker, P.J., 1993. Mutagenesis of the regulatory domain of rat protein kinase C-eta. A molecular basis for restricted histone kinase activity. J. Biol. Chem. 268, 19498–19504. Drummer, H.E., Reubel, G.H., Studdert, M.J., 1996. Equine gammaherpesvirus 2 (EHV2) is latent in B lymphocytes. Arch. Virol. 141, 495–504. England, K., Rumsby, M.G., 2000. Changes in protein kinase C epsilon phosphorylation status and intracellular localization as 3T3 and 3T6 fibroblasts grow to confluency and quiescence: a role for phosphorylation at ser-729? Biochem. J. 352, 19– 26. Epstein, P., Hachisu, R., 1984. Cyclic nucleotide phosphodiesterase in normal and leukemic human lymphoblasts. In: Strada, S., Thompson, W. (Eds.), Advances of Cyclic Nucleotide and Protein Phosphorylation Research. Raven, New York, pp. 304– 321. Essayan, D.M., Huang, S.K., Kagey-Sobotka, A., Lichtenstein, L.M., 1997. Differential efficacy of lymphocyte- and monocyte-selective pretreatment with a type 4 phosphodiesterase inhibitor on antigen-driven proliferation and cytokine gene expression. J. Allergy Clin. Immunol. 99, 28–37. Fawcett, L., Baxendale, R., Stacey, P., McGrouther, C., Harrow, I., Soderling, S., Hetman, J., Beavo, J.A., Phillips, S.C., 2000. Molecular cloning and characterization of a distinct human phosphodiesterase gene family: PDE11A. Proc. Natl. Acad. Sci. U.S.A. 97, 3702–3707. Fischer, T.A., Palmetshofer, A., Gambaryan, S., Butt, E., Jassoy, C., Walter, U., Sopper, S., Lohmann, S.M., 2001. Activation of cGMP-dependent protein kinase Ibeta inhibits interleukin 2 release and proliferation of T cell receptor-stimulated human peripheral T cells. J. Biol. Chem. 276, 5967–5974. Fraser, J.D., Straus, D., Weiss, A., 1993. Signal transduction events leading to T-cell lymphokine gene expression. Immunol. Today 14, 357–362. Fulop Jr., T., Leblanc, C., Lacombe, G., Dupuis, G., 1995. Cellular distribution of protein kinase C isozymes in CD3-mediated stimulation of human T lymphocytes with aging. FEBS Lett. 375, 69–74. Gantner, F., Gotz, C., Gekeler, V., Schudt, C., Wendel, A., Hatzelmann, A., 1998. Phosphodiesterase profile of human B lymphocytes from normal and atopic donors and the effects of PDE inhibition on B cell proliferation. Br. J. Pharmacol. 123, 1031–1038. Gantner, F., Tenor, H., Gekeler, V., Schudt, C., Wendel, A., Hatzelmann, A., 1997. Phosphodiesterase profiles of highly purified human peripheral blood leukocyte populations from normal and atopic individuals: a comparative study. J. Allergy Clin. Immunol. 100, 527–535. Giembycz, M.A., Corrigan, C.J., Seybold, J., Newton, R., Barnes, P.J., 1996. Identification of cyclic AMP phosphodiesterases 3, 4 and 7 in human CD4þ and CD8þ T-lymphocytes: role in regulating proliferation and the biosynthesis of interleukin-2. Br. J. Pharmacol. 118, 1945–1958. Gonzalez-Amaro, R., Portales-Perez, D., Baranda, L., Redondo, J.M., Martinez-Martinez, S., Yanez-Mo, M., Garcia-Vicuna, R., Cabanas, C., Sanchez-Madrid, F., 1998. Pentoxifylline inhibits
164
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
adhesion and activation of human T lymphocytes. J. Immunol. 161, 65–72. Greenaway, E.C., Cunningham, F.M., Goode, N.T., 2000. Differential localization of protein kinase C isotypes in equine eosinophils and neutrophils. J. Leukocyte Biol. 68, 575–582. Han, Z., Pantazis, P., Lange, T.S., Wyche, J.H., Hendrickson, E.A., 2000. The staurosporine analog, Ro-31-8220, induces apoptosis independently of its ability to inhibit protein kinase C. Cell. Death Differ. 7, 521–530. Hassan, J., Rainsford, E., Reen, D.J., 1997. Linkage of protein kinase C-beta activation and intracellular interleukin-2 accumulation in human naive CD4 T cells. Immunology 92, 465– 471. Hidi, R., Timmermans, S., Liu, E., Schudt, C., Dent, G., Holgate, S.T., Djukanovic, R., 2000. Phosphodiesterase and cyclic adenosine monophosphate-dependent inhibition of T-lymphocyte chemotaxis. Eur. Resp. J. 15, 342–349. Kirman, J., Le Gros, G., 1998. Which is the true regulator of TH2 cell development in allergic immune responses? Clin. Exp. Allergy 28, 908–910. Kishimoto, A., Kajikawa, N., Shiota, M., Nishizuka, Y., 1983. Proteolytic activation of calcium-activated, phospholipid-dependent protein kinase by calcium-dependent neutral protease. J. Biol. Chem. 258, 1156–1164. Kontny, E., Kurowska, M., Szczepanska, K., Maslinski, W., 2000. Rottlerin, a PKC isozyme-selective inhibitor, affects signaling events and cytokine production in human monocytes. J. Leukocyte Biol. 67, 249–258. Kydd, J., Antczak, D., Allen, W., Barbis, D., Butcher, G., Davis, W., Duffus, W., Edgington, N., Grunig, G., Holmes, M., Lunn, D., McCulloch, J., O’Brien, A., Perryman, L., Tavernor, A., Williamson, S., Zhang, C., 1994. Report of the first international workshop on leucocyte antigens, Cambridge, UK, July 1991. Vet. Immunol. Immunopathol. 42, 3–60. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680– 685. Landells, L.J., Spina, D., Souness, J.E., O’Connor, B.J., Page, C.P., 2000. A biochemical and functional assessment of monocyte phosphodiesterase activity in healthy and asthmatic subjects. Pulm. Pharmacol. Ther. 13, 231–239. Landells, L.J., Szilagy, C.M., Jones, N.A., Banner, K.H., Allen, J.M., Doherty, A., O’Connor, B.J., Spina, D., Page, C.P., 2001. Identification and quantification of phosphodiesterase 4 subtypes in CD4 and CD8 lymphocytes from healthy and asthmatic subjects. Br. J. Pharmacol. 133, 722–729. Lavoie, J.P., Maghni, K., Desnoyers, M., Taha, R., Martin, J.G., Hamid, Q.A., 2001. Neutrophilic airway inflammation in horses with heaves is characterized by a Th2-type cytokine profile. Am. J. Resp. Crit. Care Med. 164, 1410–1413. Li, L., Yee, C., Beavo, J.A., 1999. CD3-and CD28-dependent induction of PDE7 required for T cell activation. Science 283, 848–851. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275.
Manganiello, V.C., Murata, T., Taira, M., Belfrage, P., Degerman, E., 1995. Diversity in cyclic nucleotide phosphodiesterase isoenzyme families. Arch. Biochem. Biophys. 322, 1–13. Martiny-Baron, G., Kazanietz, M.G., Mischak, H., Blumberg, P.M., Kochs, G., Hug, H., Marme, D., Schachtele, C., 1993. Selective inhibition of protein kinase C isozymes by the indolocarbazole Go 6976. J. Biol. Chem. 268, 9194–9197. McGorum, B.C., Dixon, P.M., Halliwell, R.E., 1993. Phenotypic analysis of peripheral blood and bronchoalveolar lavage fluid lymphocytes in control and chronic obstructive pulmonary disease affected horses, before and after natural (hay and straw) challenges. Vet. Immunol. Immunopathol. 36, 207–222. McKelvie, J., Foster, A.P., Cunningham, F.M., Hamblin, A.S., 1999. Characterisation of lymphocyte subpopulations in the skin and circulation of horses with sweet itch (Culicoides hypersensitivity). Equine Vet. J. 31, 466–472. McKelvie, J., Little, S., Foster, A.P., Cunningham, F.M., Hamblin, A., 1998. Equine peripheral blood mononuclear cells proliferate in response to tetanus toxoid antigen. Res. Vet. Sci. 65, 91–92. Nicholson, C.D., Shahid, M., 1993. Inhibitors of cyclic nucleotide phosphodiesterase isoenzymes—their potential utility in the therapy of asthma. Pulm. Pharmacol. 6, 101–117. Nishizuka, Y., 1988. The molecular heterogeneity of protein kinase C and its implications for cellular regulation. Nature 334, 661– 665. Rickards, K., Page, C., Lees, P., Cunningham, F., 2000. Phosphodiesterase activity in neutrophils from horses with chronic obstructive pulmonary disease. Vet. Immunol. Immunopathol. 76, 319–330. Siedek, E., Little, S., Mayall, S., Edington, N., Hamblin, A., 1997. Isolation and characterisation of equine dendritic cells. Vet. Immunol. Immunopathol. 60, 15–31. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1998a. Cloning and characterization of a cAMP-specific cyclic nucleotide phosphodiesterase. Proc. Natl. Acad. Sci. U.S.A. 95, 8991– 8996. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1998b. Identification and characterization of a novel family of cyclic nucleotide phosphodiesterases. J. Biol. Chem. 273, 15553–15558. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1999. Isolation and characterization of a dual-substrate phosphodiesterase gene family: PDE10A. Proc. Natl. Acad. Sci. U.S.A. 96, 7071–7076. Szamel, M., Resch, K., 1995. T-cell antigen receptor-induced signal-transduction pathways—activation and function of protein kinases C in T lymphocytes. Eur. J. Biochem. 228, 1– 15. Tenor, H., Hatzelmann, A., Church, M.K., Schudt, C., Shute, J.K., 1996. Effects of theophylline and rolipram on leukotriene C4 (LTC4) synthesis and chemotaxis of human eosinophils from normal and atopic subjects. Br. J. Pharmacol. 118, 1727–1735. Tenor, H., Staniciu, L., Schudt, C., Hatzelmann, A., Wendel, A., Djukanovic, R., Church, M.K., Shute, J.K., 1995. Cyclic nucleotide phosphodiesterases from purified human CD4þ and CD8þ T lymphocytes. Clin. Exp. Allergy 25, 616–624. Watson, J.L., Stott, J.L., Blanchard, M.T., Lavoie, J.P., Wilson, W.D., Gershwin, L.J., Wilson, D.W., 1997. Phenotypic
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165 characterization of lymphocyte subpopulations in horses affected with chronic obstructive pulmonary disease and in normal controls. Vet. Pathol. 34, 108–116. Way, K.J., Chou, E., King, G.L., 2000. Identification of PKCisoform-specific biological actions using pharmacological approaches. Trends Pharmacol. Sci. 21, 181–187.
165
Wilkinson, S.E., Nixon, J.S., 1998. T-cell signal transduction and the role of protein kinase C. Cell Mol. Life Sci. 54, 1122–1144. Winder, N.C., von Fellenberg, R., 1988. Chronic small airway disease in the horse: immunohistochemical evaluation of lungs with mild, moderate and severe lesions. Vet. Rec. 122, 181–183.
Biochemical and functional assessment of equine lymphocyte phosphodiesterases and protein kinase C K.J. Rickardsa,*,1, C.P. Pageb, A.S. Hamblinc, N.T. Gooded, F.M. Cunninghama a
Department of Veterinary Basic Sciences, The Royal Veterinary College, Hawkshead Campus, North Mymms, Hertfordshire AL9 7TA, UK b Sackler Institute of Pulmonary Pharmacology, Division of Pharmacology and Therapeutics, GKT School of Biomedical Sciences, King’s College London, Guy’s Campus, London Bridge, London SE1 9RT, UK c Department of Pathology and Infectious Disease, The Royal Veterinary College, Camden Campus, Royal College Street, London NW1 0TU, UK d Department of Veterinary Basic Sciences, The Royal Veterinary College, Camden Campus, Royal College Street, London NW1 0TU, UK Received 22 May 2003; received in revised form 24 October 2003; accepted 2 December 2003
Abstract Lymphocytes play an important role in allergic inflammation and have been implicated in the pathogenesis of equine allergic skin and respiratory disease. Targeting intracellular signalling pathways in human lymphocytes has demonstrated a role for both phosphodiesterase and protein kinase C in cell activation. The aim of this study was to measure total cyclic nucleotide hydrolysing phosphodiesterase activity and to identify the phosphodiesterase and protein kinase C isoenzymes present in equine lymphocytes. The functional significance of these isoenzymes was then investigated by examining their role in peripheral blood mononuclear cell proliferation using isoenzyme selective inhibitors. Total cyclic adenosine monophosphate hydrolysing phosphodiesterase activity was double that of cyclic guanosine monophosphate (30 2 pmol/min mg versus 16 3 pmol/ min mg for cyclic adenosine and cyclic guanosine monophosphate phosphodiesterase activity, respectively). Evidence for the presence of PDE1, 3, 4 and 5 was obtained and PKCa, b, d, e, Z, i, y and z were identified. Selective inhibitors of PDE4, PKCd and conventional PKCs a and b caused significant inhibition of mitogen-induced peripheral blood mononuclear cell proliferation. This study demonstrates a functional role for specific signalling isoenzymes and suggests that, in the context of allergic inflammation, targeting inflammatory cells involved in disease pathogenesis with relevant isoenzyme inhibitors may have therapeutic potential. # 2003 Elsevier B.V. All rights reserved. Keywords: Phosphodiesterase; Protein kinase C; Lymphocytes; Allergic inflammation
1. Introduction *
Corresponding author. Tel.: þ44-207-848-6096; fax: þ44-207-848-6097. E-mail address: [email protected] (K.J. Rickards). 1 Present address: Sackler Institute of Pulmonary Pharmacology, Division of Pharmacology and Therapeutics, GKT School of Biomedical Sciences, King’s College, London, Guy’s Campus, London Bridge, London SE1 9RT, UK.
Lymphocytes have been implicated in the pathogenesis of equine allergic skin and respiratory disease. In ponies susceptible to the seasonally recurrent allergic skin disease, sweet itch, increased numbers of intradermal CD3þ T lymphocytes are found at skin
0165-2427/$ – see front matter # 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.vetimm.2003.12.001
154
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
sites injected with Culicoides antigen. Additionally ponies with sweet itch have increased numbers of circulating CD4þ T cells (McKelvie et al., 1999). In the allergic respiratory disease, recurrent airway obstruction (RAO), perivascular and peribronchiolar accumulation of lymphocytes has been demonstrated (Winder and von Fellenberg, 1988). A change in T-cell phenotype on exposure to antigen has been suggested by McGorum et al. (1993) who found an increase in the CD4þ:CD8þ T lymphocyte ratio after challenge in horses susceptible to RAO. More recently Lavoie et al. (2001) reported a predominant Th2-type cytokine profile in BAL samples from horses with RAO. In human allergic diseases such as asthma and atopic dermatitis, alteration in lymphocyte phenotype resulting in over expression of Th2-type cytokines has been suggested to be important in disease pathogenesis (Chan et al., 1993; Kirman and Le Gros, 1998). Thus interest has centred on intracellular signalling mechanisms involved in lymphocyte activation in order to develop novel compounds capable of attenuating lymphocyte function. There is considerable evidence to suggest that the second messenger protein kinase C (PKC) plays an essential role in T-cell activation having both stimulatory and regulatory effects (Szamel and Resch, 1995). PKC exists as a family of serine/threonine-specific isoenzymes of which 10 have been identified. The isoenzymes can be divided into three groups: (1) conventional (a, bI, bII and g); (2) novel (d, E, Z and y); (3) atypical (i and z) based on their structure, localisation and substrate requirements (Way et al., 2000). Inhibitors have been developed with some degree of isoenzyme specificity and in vitro studies using human cells have shown them to attenuate lymphocyte proliferation (Birchall et al., 1994; Hassan et al., 1997). Another signalling pathway that has been investigated in relation to T-cell function involves the second messengers, cyclic adenosine and guanosine monophosphate (cAMP and cGMP, respectively). These molecules are thought to be important in regulating inflammatory cell function and intracellular levels are maintained by G-protein linked cyclases that increase production and a growing family of enzymes termed phosphodiesterases (PDE) that break down the cyclic nucleotides. Like PKC, PDEs exist in multiple forms with 11 isoenzymes having been characterised based on substrate specificity and sensitivity to endogenous
and exogenous regulators (PDE1–11) (Manganiello et al., 1995; Soderling et al., 1998a,b, 1999; Fawcett et al., 2000). The three predominant isoenzymes present in human T lymphocytes are PDE3, 4 and 7 (Giembycz et al., 1996) and in vitro studies using selective PDE4 and non-selective inhibitors have demonstrated inhibition of IL-2 biosynthesis, proliferation, adhesion, cytokine gene expression and chemotaxis (Giembycz et al., 1996; Essayan et al., 1997; Gonzalez-Amaro et al., 1998; Briggs et al., 1999; Hidi et al., 2000). A PDE4 selective inhibitor has also been shown to reduce IgE synthesis by B-lymphocytes (Cooper et al., 1985), which contain predominantly PDE4 and PDE7 (Gantner et al., 1998) and an antisense oligonucleotide directed against PDE7 has been shown to regulate T-cell function (Li et al., 1999). At present there is no information on the PDE and PKC isoenzyme profiles or functional significance of the isoenzymes identified in equine lymphocytes. In this study, we initially characterised the isoenzymes present in lymphocytes isolated from healthy ponies. Following on from this the effects of selective PDE isoenzyme inhibitors and three PKC inhibitors, one selective for PKCd, one for conventional PKCs and one non-selective, were investigated for their ability to attenuate equine mononuclear cell proliferation.
2. Materials and methods 2.1. Animals Six healthy New-Forest ponies (age range 10–20 years) were used for the studies. The animals were kept at pasture and treated routinely with anthelminthics and tetanus vaccination. 2.2. Lymphocyte purification Venous blood (50 ml) was collected into 0.4 M ethylenediaminetetraacetic acid (EDTA). Peripheral blood mononuclear cells (PBMC) were separated by Percoll density gradient centrifugation (Drummer et al., 1996). PBMCs were harvested from the Percoll–plasma interface, washed twice in Hank’s balanced salt solution without Ca2þ and Mg2þ and re-suspended at 5 106 cells/ml in RPMI 1640 containing 2 mM L-glutamine and 10% heat inactivated
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
foetal calf serum. Cells were cultured in tissue culture flasks for 1 h at 37 8C in 5% CO2 after which nonadherent cells were removed, counted and re-suspended at 8 106 cells/ml in 0.2% bovine serum albumin (BSA) in phosphate buffered saline (PBS). Purity of the lymphocyte population was assessed using flow cytometry as previously described (Siedek et al., 1997).
155
described by Lowry et al. (1951). Aliquots containing approximately 1 106 homogenised cells were assayed in the presence of isoenzyme selective inhibitors for PDE1, 3, 4 and 5 and cGMP, which stimulates PDE2. All compounds were added at a single concentration at least 10-fold higher than the documented IC50 for inhibition of the purified isoenzyme in other species whilst still conferring selectivity (Nicholson and Shahid, 1993; Tenor et al., 1996).
2.3. Flow cytometry 2.5. Identification of PKC isoenzymes Briefly, purified cells were screened using a panel of monoclonal antibodies that recognise equine cells expressing CD3, CD4, CD8, CD13, MHC class I, MHC class II and immunoglobulin. Isotype-matched negative controls were also included. Cells, 4 106 per well, were incubated on ice for 30 min in 96-well microtitre plates with optimal predetermined dilutions of primary antibody (undiluted for CD4 and CD8, 1:2 dilution for CD13, 1:10 dilution for CD3, MHC class I and II, 1:100 dilution for immunoglobulin). Following three washes with 0.2% BSA in PBS, FITC-conjugated rabbit anti-mouse immunoglobulin F(ab)2 was added to each well (1:50 dilution in 0.2% BSA in PBS). After 30 min incubation on ice, cells were washed three times as before and kept in 1% paraformaldehyde in PBS at 4 8C until fluorescence was measured on an FACS analyser (Becton Dickinson, Oxford, UK) with Lysys II software. Cells were gated according to their forward and side scatter and analysed for the proportion of FITC-stained cells after correcting for non-specific staining using the appropriate negative control. 2.4. Measurement of PDE activity Non-adherent lymphocytes were lysed in homogenisation buffer (leupeptin 10 mg/ml, Tris (pH 8.0) 20 mM, MgCl2 2 mM, dithiothreitol 1 mM, EDTA 5 mM, benzamidine 1.3 mM, sucrose 0.25 M, Na-ptosyl-L-lysine chloromethyl ketone 20 mM, triton X-100 1%; v/v), vortexed and stored at 80 8C prior to analysis. Phosphodiesterase activity was measured using a two-step radioactivity assay as previously described (Rickards et al., 2000) and expressed as picamoles hydrolysed cyclic nucleotide per minute per milligram protein. The protein content of the samples was measured using the colorimetric method
Western blotting was carried out as previously described (Laemmli, 1970). Briefly, 1 107 cells were solubilised by boiling for 5 min in 1 ml sodium dodecyl sulphate (SDS; 4%) buffer containing 125 mM Tris pH 6.8, 10% glycerol, 50 mM dithiothreitol and 25 mg/ml bromophenol blue and stored at 20 8C. Cell proteins from 30 ml aliquots were separated by SDS-polyacrylamide gel electrophoresis (PAGE) using a 9% polyacrylamide resolving gel and 5% stacking gel and transferred electrophoretically onto a polyvinylidene difluoride membrane. Membranes were blocked for 2 h with 5% (w/v) skimmed milk powder in Tris-buffered saline, pH 7.5, and then probed with PKC isoenzyme-specific antibodies (1:250 dilution in Tris-buffered saline for b, i and y; 1:200 dilution for all other antibodies), which have previously been validated in equine granulocytes (Greenaway et al., 2000). After washing, membranes were incubated with species-specific horseradish peroxidase-linked anti-IgG secondary antibodies (1:5000 dilution in 0.05% (w/v) milk powder in Tris-buffered saline) and immunoreactive bands visualised using the enhanced chemiluminescence detection system (Amersham International, Buckinghamshire, UK). The amounts of individual PKC isoenzymes were estimated; semi-quantitatively by densitometric analysis corrected for protein content in each sample by densitometric analysis of SDS-PAGE separated proteins stained with Brilliant Blue G-Colloid (Sigma Diagnostics, Dorset, UK). 2.6. PBMC proliferation Proliferation was measured using the method described by McKelvie et al. (1998). Briefly, PBMCs were separated from 50 ml venous blood collected
156
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
into heparin (20 i.u./ml) as described above, counted and re-suspended at 1 106 cells/ml in RPMI 1640 containing 10% autologous horse plasma, 2 mM L-glutamine, 50 mg/ml streptomycin and 50 i.u./ml penicillin. Cells ð5 105 Þ were incubated with a range of concentrations of phytohaemagglutinin (PHA; 3.5–225 mg/ml) for 72 h at 37 8C in 5% CO2 at which point ½3 H-thymidine (0.8 mCi) was added and, after a further 18 h incubation, cells were harvested onto glass fibre filters using a cell harvester (Ilacon, Tonbridge, Kent, UK). The filters were placed in scintillation fluid and radioactivity counted in a liquid scintillation counter (Tri-CarbTM 2500TR, Canberra Packard, Berkshire, UK). Proliferation was then measured in the presence of the PDE4 inhibitor, rolipram (108 to 104 M), the PDE5 inhibitors, zaprinast (107 to 104 M) and sildenafil (108 to 2 106 M), the PDE3 inhibitors, quazinone (108 to 105 M) and milrinone (107 to 104 M), a combination of milrinone (105 M) and zaprinast (105 M), the non-selective PKC inhibitor, Ro-31-8220 (5 109 to 105 M), the conventional PKC inhibitor, Go¨ -6976 (5 1010 to 106 M), or the PKCd inhibitor, rottlerin (108 to 105 M) using the concentration of PHA that produced 90% maximal proliferation (28 mg/ml). All compounds were dissolved in DMSO and serial dilutions were made in RPMI 1640 containing 2 mM L-glutamine, 50 mg/ml streptomycin and 50 i.u./ml penicillin to ensure that the final concentration of DMSO in contact with the cells was 0.1% with the exception of the top concentrations of sildenafil, Ro-31-8220 and Go¨ -6976 and the combination of milrinone and zaprinast where the final concentration of DMSO was 0.2%. Appropriate DMSO controls were included in the study. 2.7. Materials Sodium heparin was obtained from National Veterinary Supplies, Staffordshire, UK. 3 H-cAMP, 3 HcGMP and species-specific horseradish peroxidaselinked secondary antibodies were obtained from Amersham Pharmacia Biotech, Buckinghamshire, UK. 3 H-thymidine was obtained from ICN Pharmaceuticals Ltd., Hampshire, UK. PHA was obtained from Biostat, Derbyshire, UK. RPMI 1640 and penicillin-streptomycin (5000 i.u./ml to 5000 mg/ml) were obtained from Invitrogen, Renfrewshire, UK.
Methanol was obtained from Merck Ltd., Leicestershire, UK. Optiphase Hisafe 8 was obtained from Fisher Scientific, Leicestershire, UK. Ecoscint O was obtained from National Diagnostics, Yorkshire, UK. Rolipram was kindly provided by Schering AG, Berlin, Germany. Zaprinast, milrinone, quazinone, vinpocetine, rottlerin, Ro-31-8220 and Go¨ -6976 were obtained from Calbiochem, Nottinghamshire, UK. Foetal calf serum was obtained from Harlan SeraLab, Leicestershire, UK. Sildenafil was kindly provided by Pfizer, Kent, UK. FACS isotype negative controls and antibodies recognising CD13, MHC class I and II were obtained from Serotec Ltd., Oxfordshire, UK. The antibody recognising B cells was obtained from VMRD Inc., Washington, USA. The antibody recognising CD3 was kindly provided by Prof. J. Stott, University of California, CA. The antibody recognising CD4 was kindly provided by Dr. M. Holmes, Cambridge, UK. The antibody recognising CD8 was kindly provided by S. Mayall, Royal Veterinary College, London, UK. The FITC-conjugated rabbit anti-mouse immunoglobulin F(ab)2 was obtained from DAKO Ltd., Cambridgeshire, UK. Rabbit polyclonal PKC a, d, E, g, Z and z antibodies were obtained from Santa Cruz Biotechnology, California, USA. Mouse monoclonal PKC b, i and y antibodies were obtained from Transduction Laboratories via Affiniti Research Products Ltd., Devon, UK. All other reagents were obtained from Sigma Diagnostics, Dorset, UK. 2.8. Data analysis All data are expressed as mean S:E:M. To calculate lymphocyte purity following separation, the total number of CD3 positive and immunoglobulin positive cells were expressed as a percentage of the number of cells staining positive for MHC class I. PBMC proliferation was expressed as a stimulation index by calculating mean decays per million (dpm) in the presence of PHA and dividing by mean dpm in the presence of medium alone. The PDE profile and PBMC proliferation data were analysed by repeated measures one-way ANOVA (GraphPad Prism3, GraphPad Software, San Diego, USA) and any significant differences further investigated by Dunnett’s post hoc test where the control data was either total PDE activity or PBMC proliferation in the presence of
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
DMSO. Differences were considered significant when a P value of 5% or less was found.
3. Results 3.1. Lymphocyte purification FACS analysis of the cell population following lymphocyte separation demonstrated that 91 5% stained positive for lymphocyte markers. Of the lymphocyte gated population, 7 1% were B cells, 62 3% were CD4þ and 21 3% were CD8þ lymphocytes. Contaminants staining positive for CD13 comprised 7 2% of the cell population. Although no marker for platelets was included, microscopic analysis of a modified Wright’s-stained smear under high power showed no visible evidence of platelets. 3.2. PDE isoenzyme analysis The PDE isoenzyme profile of equine lymphocytes is shown in Fig. 1. All the isoenzyme selective inhi-
157
bitors reduced PDE activity indicating that equine lymphocytes contain PDE1, 3, 4 and 5. Addition of cGMP also inhibited cAMP PDE activity providing further evidence for the presence of PDE3 as this isoenzyme is inhibited by cGMP. Lymphocyte cAMP PDE activity was approximately double that of cGMP PDE activity (30 2 pmol hydrolysed cAMP/min mg versus 16 3 pmol hydrolysed cGMP/min mg). 3.3. PKC isoenzyme analysis Western blot analysis of the PKC isoenzymes present in equine lymphocytes showed that all but PKCg were expressed (Fig. 2). This isoenzyme was identified in the control sample (horse cerebrum) demonstrating the ability of the monoclonal antibody to recognise equine PKCg (data not shown). All the bands found at the correct molecular weight for the relevant PKC isoenzyme appear as a single band except for PKCd and PKCz, which appear as doublets. In addition to these primary bands, all antibodies except those directed against PKCE and Z showed further bands at lower molecular weights although these were always of a lower intensity than the main band with the exception of lanes 3 and 4 for PKCb where all the bands identified were of a similar intensity. Where these bands correspond to a 50 kDa product, which is likely to represent a fragmentation product (Kishimoto et al., 1983), this molecular weight is marked. In all other cases the bands fall between 50 and 75 kDa. 3.4. PBMC proliferation
Fig. 1. cAMP and cGMP PDE isoenzyme profile of equine lymphocytes. Filled columns represent total PDE activity. Vinpocetine (1 104 M) inhibits PDE1, cGMP (5 106 M) stimulates PDE2 and inhibits PDE3, quazinone (6 105 M) inhibits PDE3, rolipram (1 105 M) inhibits PDE4 and zaprinast (1 105 M) inhibits PDE5. Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between responses in the presence of inhibitors and the relevant control are indicated by an asterisk (*P < 0:05).
Equine PBMCs proliferated in response to PHA giving a characteristic bell-shaped curve. Maximum proliferation occurred with 56 mg/ml PHA (mean stimulation index ¼ 87 29), whilst 28 mg/ml resulted in 90% of maximum response. This concentration was used in the subsequent inhibitor studies. Addition of the PDE4 inhibitor, rolipram, resulted in concentration-dependent inhibition of PHA-induced PBMC proliferation with significant effects occurring at concentrations of 107 M and above (Fig. 3a). Following addition of either milrinone or quazinone, both PDE3 inhibitors, proliferation was also significantly attenuated (Fig. 3b and c) with the effect occurring at a lower concentration in the presence of quazinone (106 M). Both inhibitors produced a similar amount of
158
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Fig. 2. Western blot identification of the PKC isoenzymes in equine lymphocytes (approx. 1 107 cells/ml). Arrows indicate molecular weight markers. Lane 1 corresponds to horse brain control and lanes 2–7 correspond to 30 ml lymphocyte aliquots where each lane represents cells from the same horse on consecutive blots.
inhibition at 105 M (24 4 and 16 5% for milrinone and quazinone, respectively). Only milrinone could be investigated at 104 M where inhibition of a similar magnitude to that caused by rolipram occurred (72 3 and 79 3%, respectively). Proliferation was also significantly reduced by zaprinast, a PDE5 inhibitor, when it was added at a concentration of 105 M, producing a similar amount of inhibition to the PDE3 inhibitors at this concentration (Fig. 3d). At 104 M, proliferation was inhibited by 69 2%. Another PDE5 inhibitor, sildenafil, was tested as this compound has been shown to be a more selective and potent inhibitor of PDE5 (Ballard et al., 1998). In the presence of concentrations ranging from 108 to 2 106 M, sildenafil produced a small inhibitory effect that was independent of concentration (12 1% versus 10 5% at 108 M and 2 106 M, respectively; Fig. 3e) and was statistically significant at all concentrations except the highest tested. When a combination of 105 M milrinone and 105 M zaprinast was added the amount of inhibition achieved appeared to be simply additive ð45 3%Þ. Ro-31-8220 (non-selective PKC inhibitor), Go¨ 6976 (inhibitor of conventional PKCs) and rottlerin (inhibitor of PKCd) attenuated PHA-induced PBMC
proliferation in a concentration-dependent manner (Fig. 4a–c). At the two highest concentrations tested, Ro-31-8220 caused complete abrogation of the proliferative response whilst the inhibitory effect of rottlerin and Go¨ -6976 was only partial at the top concentration tested.
4. Discussion The aim of this study was to identify equine lymphocyte PKC and PDE isoenzyme profiles as the initial step in establishing expression and roles for individual isoenzymes in modulating lymphocyte function in the horse. FACS analysis of the cell population isolated from equine blood demonstrated 91% purity with 2% granulocyte and 5% monocyte contamination. Drummer et al. (1996) using an identical cell separation procedure produced a 77% pure population, which may reflect differences in antibody binding affinity during FACS analysis. The percentages of lymphocyte gated cells staining positive for CD4 and CD8 were similar to those documented by others when screening peripheral blood collected from healthy horses
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
159
Fig. 3. Effect of: (a) rolipram (108 to 104 M), (b) quazinone (108 to 105 M), (c) milrinone (106 to 104 M), (d) zaprinast (108 to 104 M) and (e) sildenafil (108 to 2 106 M) on PHA-induced PBMC proliferation. Results are expressed as % of the stimulation index in the presence of PHA (28 mg/ml) alone (mean stimulation index ¼ 76 20 for (a)–(d); and 161 13 and 152 14 for 0.1 and 0.2% DMSO, respectively, for (e)). Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between control and drug treatment are indicated by an asterisk (*P < 0:05).
(McGorum et al., 1993; Watson et al., 1997) indicating that Percoll density gradient centrifugation does not influence the composition of the lymphocyte population. In addition, the separation technique proved to be highly reproducible with an inter-assay coefficient of variation of 4.5%.
In these studies, PDE and PKC isoenzyme expression were measured in the total lymphocyte population. Although it is possible to separate equine lymphocyte subtypes using magnetic cell separation (Kydd et al., 1994), yields of purified equine T and B cells are likely to be low. The data obtained in these
160
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Fig. 4. Effect of: (a) Ro-31-8220 (5 109 to 105 M), (b) Go¨ 6976 (5 1010 to 106 M) and (c) rottlerin (108 to 105 M) on PHA-induced PBMC proliferation. Results are expressed as % of the stimulation index in the presence of PHA (28 mg/ml) alone (mean stimulation index ¼ 157 24 and 158 20 for 0.1 and 0.2% DMSO, respectively). Columns represent mean values and bars represent S.E.M. ðn ¼ 6Þ. Data were analysed by repeated measures one-way ANOVA followed by Dunnett’s post hoc test. Significant differences between control and drug treatment are indicated by an asterisk (*P < 0:05).
studies of the total lymphocyte population will enable further studies to focus on the relevant isoenzymes when investigating individual lymphocyte subsets. Using isoenzyme selective inhibitors, evidence for the presence of PDE1, PDE3, PDE4 and PDE5 was obtained in equine lymphocytes, which is similar to the isoenzyme profile of human lymphocytes (Tenor et al., 1995; Giembycz et al., 1996; Gantner et al., 1997, 1998). The greatest inhibition of cAMP PDE activity occurred in the presence of the PDE4 inhibitor, rolipram, suggesting that PDE4 is the predominant isoenzyme. This is in agreement with findings in human lymphocytes (Tenor et al., 1995; Giembycz et al., 1996; Gantner et al., 1997) and indeed leucocytes from several species including equine neutrophils (Rickards et al., 2000). It should be noted that there is currently no PDE7 selective inhibitor available, thus it was not possible to investigate PDE7 expression in equine cells. Previously, it has been shown that PDE5 activity in preparations of mononuclear cells originated from platelet contamination (Epstein and Hachisu, 1984). Platelet contamination of Percoll-separated equine lymphocytes was minimal suggesting that the observed PDE5 activity was likely to be of lymphocyte origin. Analysis of PKC isoenzyme expression demonstrated that all but PKCg were present in equine lymphocytes. PKCg was, however, present in the positive control sample of horse cerebrum indicating that, as shown previously (Greenaway et al., 2000), the monoclonal antibody used recognised equine PKCg. Detection of all the other isoenzymes agrees with the profile in human T lymphocytes (Baier et al., 1993; Fulop et al., 1995) and suggests that, as with human cells, individual isoenzymes have distinct functional roles with respect to lymphocyte activation (Wilkinson and Nixon, 1998). In human cells PKCb has been shown to be the predominant isoenzyme (Fulop et al., 1995). Purified equine PKC isoenzymes are not available and in the current study, it was not therefore possible to determine the relative amounts present as antibody avidity and affinity influence the density of the observed bands. However, if data are expressed as ratios of one PKC isotype, for which PKCa was chosen in view of its ubiquitous expression (Nishizuka, 1988), an estimate of the relative amounts of each isoenzyme for individual animals can be calculated (Table 1).
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
161
Table 1 PKC isoenzyme ratios calculated from optical density measurements following correction for protein content of the samplesa a b d E Z i y z OD a
1 0.6 2.2 2.0 1.0 0.8 1.3 1.4 10.0
1 0.7 2.7 8.3 1.0 1.9 3.0 2.8 1.8
1 0.7 2.3 7.1 0.9 2.5 3.5 3.5 1.5
1 0.5 1.7 2.0 0.8 1.1 1.6 1.4 9.1
1 0.5 1.8 1.6 0.6 1.0 1.3 1.4 8.0
1 0.5 1.8 1.6 0.6 1.1 1.4 1.3 9.8
a Each column corresponds to an individual animal. The ratios are expressed in terms of the amount of PKCa that is present in each sample. The bottom row represents the PKCa optical density (OD), measured in arbitrary units, for each animal following correction for protein content of the samples.
Such an approach suggests considerable inter-animal variability in isoenzyme expression. All cell samples were analysed at the same time in the same way and no fractionation was undertaken. Thus changes in expression as a result of translocation or downregulation during sampling or cell processing cannot explain the observed variability. However, it should be noted that, in addition to the predominant band identified at the known molecular weight for the relevant isoenzyme, secondary bands of lower molecular weight were identified for most of the isoenzymes, which are likely to represent fragmentation products (Kishimoto et al., 1983; Dekker et al., 1993). In addition PKCd and PKCz separate out as doublets suggesting that these two isoenzymes are present in different phosphorylation states (England and Rumsby, 2000). As mentioned previously, with the exception of PKCb in two of the six ponies, the relative intensity of the secondary bands and doublets correlates well with that of the corresponding main band for each individual animal suggesting that modification is a consistent phenomenon. However, the possibility remains that some of the inter-animal variability in isoenzyme expression occurs as a result of differences in PKC processing during cell separation. This study confirmed earlier findings that equine PBMCs proliferate in response to the mitogen, PHA, producing a characteristic bell-shaped curve (McKelvie et al., 1998). The PDE4 inhibitor, rolipram, caused concentration related inhibition of the proliferative response, which reached significance at 1 107 M. This concentration is lower than the documented IC50 for inhibition of PDE4 suggesting that the signalling pathway for mitogen-induced PBMC proliferation is
cAMP-dependent and that cAMP levels are controlled by PDE4. The inhibition produced by rolipram was similar to that seen in human PBMCs (Landells et al., 2001) and may be due to a direct effect on T lymphocytes (Betz and Fox, 1991; Giembycz et al., 1996; Essayan et al., 1997) or may involve indirect effects on accessory cells such as monocytes (Banner et al., 1995). Although rolipram has previously been shown to decrease cell viability as inhibition of proliferation increases, Banner et al. (1996) found that at any given concentration of rolipram the anti-proliferative effect was much greater than the reduction in cell viability. Thus the effects of rolipram in the present study are unlikely to be entirely attributable to a reduction in cell viability. Quazinone and milrinone only produced significant inhibitory effects at concentrations 2- and 30-fold higher than their documented IC50’s for PDE3 inhibition, respectively (Calbiochem, Nottinghamshire, UK data sheets). This suggests that PDE3 is either not involved, or is of limited importance in, regulating the cAMP-dependent steps in the signalling pathway for PBMC proliferation, which is in agreement with human studies (Giembycz et al., 1996). It is possible that the small inhibitory effect seen at 1 105 M may have been due to inhibition of monocyte function, although selective inhibition of PDE3 in human monocytes had no effect on TNFa release (Landells et al., 2000). Moreover, the effects of higher concentrations of milrinone may be due to inhibition of PDE4 as the IC50 is 2 105 M (Calbiochem, Nottinghamshire, UK data sheet). As with all the compounds used that produced inhibitory effects, further work is required to identify the precise cellular target for their action.
162
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Involvement of cGMP-dependent protein kinase in inhibition of human PBMC proliferation has been documented using cGMP analogues and direct activators of guanylyl cyclase (Fischer et al., 2001). Therefore, the effect of the PDE5 inhibitor, zaprinast, on equine PBMC proliferation was investigated. Significant inhibitory effects were only seen at concentrations higher than the documented IC50 for PDE5 inhibition (Nicholson and Shahid, 1993). However, it should be noted that this compound has been reported to show poor cell penetration (Nicholson and Shahid, 1993) thus the intracellular concentration may have been lower than required for PDE5 inhibition. In order to further investigate whether inhibition of PDE5 was functionally significant in equine PBMC proliferation, the effect of the more potent and selective PDE5 inhibitor, sildenafil, was investigated (Ballard et al., 1998). An inhibitory effect was observed using sildenafil, which, although small, was significant but unrelated to concentration. Although no data could be found on the effect of sildenafil on PBMC proliferation in other species, the compound has been shown to inhibit proliferation of human cultured smooth muscle cells with an IC50 of 2 106 M (Adolfsson et al., 2002). Solubility precluded investigation of concentrations above 2 106 M in equine PBMC proliferation. Thus, use of sildenafil did not provide conclusive evidence of PDE5 involvement in equine PBMC proliferation and it remains possible that the effect of zaprinast is due to inhibition of PDE1 (Nicholson and Shahid, 1993). Another possibility is that elevating cGMP has an indirect effect via a reciprocal increase in cAMP due to inhibition of PDE3. To test this hypothesis, a combination of zaprinast and milrinone was examined with both inhibitors being added at a concentration that caused a small but significant effect individually (1 105 M). However, the inhibitory effect obtained was additive, rather than synergistic, suggesting that the two inhibitors are working independently. Therefore, it appears that the signalling pathway for equine PBMC proliferation is under both cAMP- and cGMP-dependent regulation. Addition of the non-selective PKC inhibitor, Ro-318220, significantly inhibited proliferation at all concentrations tested except 5 108 M. The lack of effect at this concentration is difficult to explain as a 10-fold lower concentration did cause inhibition. However, as most of the PKC isoenzymes are present
in equine lymphocytes, it is possible that different isoenzymes, some of which may be regulatory, are inhibited at different concentrations of Ro-31-8220. Higher concentrations caused complete inhibition demonstrating the importance of PKC in the signalling pathways involved in lymphocyte proliferation, which is similar to its effect in human lymphocytes (Fraser et al., 1993). However, it is possible that at the higher concentrations some of the inhibitory effect on proliferation may have been a result of increased apoptosis (Han et al., 2000). Although no information is available on the concentration of PKC inhibitors required for selective isoenzyme inhibition in equine cells, the results obtained do suggest that PKCd is of functional importance as approximately 30% inhibition of proliferation was achieved in the presence of rottlerin at the concentration documented to be the IC50 for inhibition of PKCd (Way et al., 2000). However, it should be noted that rottlerin also inhibits PKCy and this isoenzyme has been shown to be involved in human T-cell signalling (Altman et al., 2000). The conventional PKC inhibitor, Go¨ -6976, also caused inhibition of the proliferative response at a concentration similar to the reported IC50’s for PKCa and b (2 109 and 6 109 M, respectively) suggesting a role for one or both of these isoenzymes (Martiny-Baron et al., 1993). This is in agreement with the findings of Hassan et al. (1997) who found that in human CD4þ T-cells, Go¨ 6976 (20 mM) completely inhibited anti-CD2þCD28induced intracellular IL-2 accumulation. As mentioned earlier the use of PBMCs precludes evaluation of the precise target of PKC inhibition in the proliferation signalling cascade. Indeed, rottlerin has been shown to inhibit IL-1b production in PMA- and LPSstimulated human monocytes (Kontny et al., 2000) suggesting that accessory cells may be targeted in addition to lymphocytes. In conclusion, equine lymphocytes have been shown to contain several PKC and PDE isoenzymes. As in other species these isoenzymes appear to have distinct functional roles with PDE4, PKCa and/or b and PKCd and/or y being involved in equine PBMC proliferation. It will be of interest to investigate the importance of PDE and PKC isoenzymes in other lymphocyte responses such as cytokine production of direct relevance to allergic respiratory and skin disease pathogenesis.
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
Acknowledgements The authors would like to thank the owners who kindly allowed their animals to be used in this study and the Horserace Betting Levy Board for their financial support.
References Adolfsson, P.I., Ahlstrand, C., Varenhorst, E., Svensson, S.P., 2002. Lysophosphatidic acid stimulates proliferation of cultured smooth muscle cells from human BPH tissue: sildenafil and papaverin generate inhibition. Prostate 51, 50–58. Altman, A., Isakov, N., Baier, G., 2000. Protein kinase Ctheta: a new essential superstar on the T-cell stage. Immunol. Today 21, 567–573. Baier, G., Telford, D., Giampa, L., Coggeshall, K.M., BaierBitterlich, G., Isakov, N., Altman, A., 1993. Molecular cloning and characterization of PKC theta, a novel member of the protein kinase C (PKC) gene family expressed predominantly in hematopoietic cells. J. Biol. Chem. 268, 4997–5004. Ballard, S.A., Gingell, C.J., Tang, K., Turner, L.A., Price, M.E., Naylor, A.M., 1998. Effects of sildenafil on the relaxation of human corpus cavernosum tissue in vitro and on the activities of cyclic nucleotide phosphodiesterase isozymes. J. Urol. 159, 2164–2171. Banner, K.H., Bertin, B., Moodley, I., Page, C.P., 1996. Effect of isoenzyme selective phosphodiesterase inhibitors on the proliferation of murine thymus and spleen cells. Pulm. Pharmacol. 9, 35–41. Banner, K.H., Roberts, N.M., Page, C.P., 1995. Differential effect of phosphodiesterase 4 inhibitors on the proliferation of human peripheral blood mononuclear cells from normals and subjects with atopic dermatitis. Br. J. Pharmacol. 116, 3169–3174. Betz, M., Fox, B.S., 1991. Prostaglandin E2 inhibits production of Th1 lymphokines but not of Th2 lymphokines. J. Immunol. 146, 108–113. Birchall, A.M., Bishop, J., Bradshaw, D., Cline, A., Coffey, J., Elliott, L.H., Gibson, V.M., Greenham, A., Hallam, T.J., Harris, W. et al., 1994. Ro 32-0432, a selective and orally active inhibitor of protein kinase C prevents T-cell activation. J. Pharmacol. Exp. Ther. 268, 922–929. Briggs, W.A., Wu, Q., Orgul, O., Choi, M., Scheel Jr., P.J., Burdick, J., 1999. Lymphocyte suppression by rolipram with other immunosuppressive drugs. J. Clin. Pharmacol. 39, 794–799. Chan, S.C., Li, S.H., Hanifin, J.M., 1993. Increased interleukin-4 production by atopic mononuclear leukocytes correlates with increased cyclic adenosine monophosphate-phosphodiesterase activity and is reversible by phosphodiesterase inhibition. J. Invest. Dermatol. 100, 681–684. Cooper, K.D., Kang, K., Chan, S.C., Hanifin, J.M., 1985. Phosphodiesterase inhibition by Ro 20-1724 reduces hyperIgE synthesis by atopic dermatitis cells in vitro. J. Invest. Dermatol. 84, 477–482.
163
Dekker, L.V., McIntyre, P., Parker, P.J., 1993. Mutagenesis of the regulatory domain of rat protein kinase C-eta. A molecular basis for restricted histone kinase activity. J. Biol. Chem. 268, 19498–19504. Drummer, H.E., Reubel, G.H., Studdert, M.J., 1996. Equine gammaherpesvirus 2 (EHV2) is latent in B lymphocytes. Arch. Virol. 141, 495–504. England, K., Rumsby, M.G., 2000. Changes in protein kinase C epsilon phosphorylation status and intracellular localization as 3T3 and 3T6 fibroblasts grow to confluency and quiescence: a role for phosphorylation at ser-729? Biochem. J. 352, 19– 26. Epstein, P., Hachisu, R., 1984. Cyclic nucleotide phosphodiesterase in normal and leukemic human lymphoblasts. In: Strada, S., Thompson, W. (Eds.), Advances of Cyclic Nucleotide and Protein Phosphorylation Research. Raven, New York, pp. 304– 321. Essayan, D.M., Huang, S.K., Kagey-Sobotka, A., Lichtenstein, L.M., 1997. Differential efficacy of lymphocyte- and monocyte-selective pretreatment with a type 4 phosphodiesterase inhibitor on antigen-driven proliferation and cytokine gene expression. J. Allergy Clin. Immunol. 99, 28–37. Fawcett, L., Baxendale, R., Stacey, P., McGrouther, C., Harrow, I., Soderling, S., Hetman, J., Beavo, J.A., Phillips, S.C., 2000. Molecular cloning and characterization of a distinct human phosphodiesterase gene family: PDE11A. Proc. Natl. Acad. Sci. U.S.A. 97, 3702–3707. Fischer, T.A., Palmetshofer, A., Gambaryan, S., Butt, E., Jassoy, C., Walter, U., Sopper, S., Lohmann, S.M., 2001. Activation of cGMP-dependent protein kinase Ibeta inhibits interleukin 2 release and proliferation of T cell receptor-stimulated human peripheral T cells. J. Biol. Chem. 276, 5967–5974. Fraser, J.D., Straus, D., Weiss, A., 1993. Signal transduction events leading to T-cell lymphokine gene expression. Immunol. Today 14, 357–362. Fulop Jr., T., Leblanc, C., Lacombe, G., Dupuis, G., 1995. Cellular distribution of protein kinase C isozymes in CD3-mediated stimulation of human T lymphocytes with aging. FEBS Lett. 375, 69–74. Gantner, F., Gotz, C., Gekeler, V., Schudt, C., Wendel, A., Hatzelmann, A., 1998. Phosphodiesterase profile of human B lymphocytes from normal and atopic donors and the effects of PDE inhibition on B cell proliferation. Br. J. Pharmacol. 123, 1031–1038. Gantner, F., Tenor, H., Gekeler, V., Schudt, C., Wendel, A., Hatzelmann, A., 1997. Phosphodiesterase profiles of highly purified human peripheral blood leukocyte populations from normal and atopic individuals: a comparative study. J. Allergy Clin. Immunol. 100, 527–535. Giembycz, M.A., Corrigan, C.J., Seybold, J., Newton, R., Barnes, P.J., 1996. Identification of cyclic AMP phosphodiesterases 3, 4 and 7 in human CD4þ and CD8þ T-lymphocytes: role in regulating proliferation and the biosynthesis of interleukin-2. Br. J. Pharmacol. 118, 1945–1958. Gonzalez-Amaro, R., Portales-Perez, D., Baranda, L., Redondo, J.M., Martinez-Martinez, S., Yanez-Mo, M., Garcia-Vicuna, R., Cabanas, C., Sanchez-Madrid, F., 1998. Pentoxifylline inhibits
164
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165
adhesion and activation of human T lymphocytes. J. Immunol. 161, 65–72. Greenaway, E.C., Cunningham, F.M., Goode, N.T., 2000. Differential localization of protein kinase C isotypes in equine eosinophils and neutrophils. J. Leukocyte Biol. 68, 575–582. Han, Z., Pantazis, P., Lange, T.S., Wyche, J.H., Hendrickson, E.A., 2000. The staurosporine analog, Ro-31-8220, induces apoptosis independently of its ability to inhibit protein kinase C. Cell. Death Differ. 7, 521–530. Hassan, J., Rainsford, E., Reen, D.J., 1997. Linkage of protein kinase C-beta activation and intracellular interleukin-2 accumulation in human naive CD4 T cells. Immunology 92, 465– 471. Hidi, R., Timmermans, S., Liu, E., Schudt, C., Dent, G., Holgate, S.T., Djukanovic, R., 2000. Phosphodiesterase and cyclic adenosine monophosphate-dependent inhibition of T-lymphocyte chemotaxis. Eur. Resp. J. 15, 342–349. Kirman, J., Le Gros, G., 1998. Which is the true regulator of TH2 cell development in allergic immune responses? Clin. Exp. Allergy 28, 908–910. Kishimoto, A., Kajikawa, N., Shiota, M., Nishizuka, Y., 1983. Proteolytic activation of calcium-activated, phospholipid-dependent protein kinase by calcium-dependent neutral protease. J. Biol. Chem. 258, 1156–1164. Kontny, E., Kurowska, M., Szczepanska, K., Maslinski, W., 2000. Rottlerin, a PKC isozyme-selective inhibitor, affects signaling events and cytokine production in human monocytes. J. Leukocyte Biol. 67, 249–258. Kydd, J., Antczak, D., Allen, W., Barbis, D., Butcher, G., Davis, W., Duffus, W., Edgington, N., Grunig, G., Holmes, M., Lunn, D., McCulloch, J., O’Brien, A., Perryman, L., Tavernor, A., Williamson, S., Zhang, C., 1994. Report of the first international workshop on leucocyte antigens, Cambridge, UK, July 1991. Vet. Immunol. Immunopathol. 42, 3–60. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680– 685. Landells, L.J., Spina, D., Souness, J.E., O’Connor, B.J., Page, C.P., 2000. A biochemical and functional assessment of monocyte phosphodiesterase activity in healthy and asthmatic subjects. Pulm. Pharmacol. Ther. 13, 231–239. Landells, L.J., Szilagy, C.M., Jones, N.A., Banner, K.H., Allen, J.M., Doherty, A., O’Connor, B.J., Spina, D., Page, C.P., 2001. Identification and quantification of phosphodiesterase 4 subtypes in CD4 and CD8 lymphocytes from healthy and asthmatic subjects. Br. J. Pharmacol. 133, 722–729. Lavoie, J.P., Maghni, K., Desnoyers, M., Taha, R., Martin, J.G., Hamid, Q.A., 2001. Neutrophilic airway inflammation in horses with heaves is characterized by a Th2-type cytokine profile. Am. J. Resp. Crit. Care Med. 164, 1410–1413. Li, L., Yee, C., Beavo, J.A., 1999. CD3-and CD28-dependent induction of PDE7 required for T cell activation. Science 283, 848–851. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275.
Manganiello, V.C., Murata, T., Taira, M., Belfrage, P., Degerman, E., 1995. Diversity in cyclic nucleotide phosphodiesterase isoenzyme families. Arch. Biochem. Biophys. 322, 1–13. Martiny-Baron, G., Kazanietz, M.G., Mischak, H., Blumberg, P.M., Kochs, G., Hug, H., Marme, D., Schachtele, C., 1993. Selective inhibition of protein kinase C isozymes by the indolocarbazole Go 6976. J. Biol. Chem. 268, 9194–9197. McGorum, B.C., Dixon, P.M., Halliwell, R.E., 1993. Phenotypic analysis of peripheral blood and bronchoalveolar lavage fluid lymphocytes in control and chronic obstructive pulmonary disease affected horses, before and after natural (hay and straw) challenges. Vet. Immunol. Immunopathol. 36, 207–222. McKelvie, J., Foster, A.P., Cunningham, F.M., Hamblin, A.S., 1999. Characterisation of lymphocyte subpopulations in the skin and circulation of horses with sweet itch (Culicoides hypersensitivity). Equine Vet. J. 31, 466–472. McKelvie, J., Little, S., Foster, A.P., Cunningham, F.M., Hamblin, A., 1998. Equine peripheral blood mononuclear cells proliferate in response to tetanus toxoid antigen. Res. Vet. Sci. 65, 91–92. Nicholson, C.D., Shahid, M., 1993. Inhibitors of cyclic nucleotide phosphodiesterase isoenzymes—their potential utility in the therapy of asthma. Pulm. Pharmacol. 6, 101–117. Nishizuka, Y., 1988. The molecular heterogeneity of protein kinase C and its implications for cellular regulation. Nature 334, 661– 665. Rickards, K., Page, C., Lees, P., Cunningham, F., 2000. Phosphodiesterase activity in neutrophils from horses with chronic obstructive pulmonary disease. Vet. Immunol. Immunopathol. 76, 319–330. Siedek, E., Little, S., Mayall, S., Edington, N., Hamblin, A., 1997. Isolation and characterisation of equine dendritic cells. Vet. Immunol. Immunopathol. 60, 15–31. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1998a. Cloning and characterization of a cAMP-specific cyclic nucleotide phosphodiesterase. Proc. Natl. Acad. Sci. U.S.A. 95, 8991– 8996. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1998b. Identification and characterization of a novel family of cyclic nucleotide phosphodiesterases. J. Biol. Chem. 273, 15553–15558. Soderling, S.H., Bayuga, S.J., Beavo, J.A., 1999. Isolation and characterization of a dual-substrate phosphodiesterase gene family: PDE10A. Proc. Natl. Acad. Sci. U.S.A. 96, 7071–7076. Szamel, M., Resch, K., 1995. T-cell antigen receptor-induced signal-transduction pathways—activation and function of protein kinases C in T lymphocytes. Eur. J. Biochem. 228, 1– 15. Tenor, H., Hatzelmann, A., Church, M.K., Schudt, C., Shute, J.K., 1996. Effects of theophylline and rolipram on leukotriene C4 (LTC4) synthesis and chemotaxis of human eosinophils from normal and atopic subjects. Br. J. Pharmacol. 118, 1727–1735. Tenor, H., Staniciu, L., Schudt, C., Hatzelmann, A., Wendel, A., Djukanovic, R., Church, M.K., Shute, J.K., 1995. Cyclic nucleotide phosphodiesterases from purified human CD4þ and CD8þ T lymphocytes. Clin. Exp. Allergy 25, 616–624. Watson, J.L., Stott, J.L., Blanchard, M.T., Lavoie, J.P., Wilson, W.D., Gershwin, L.J., Wilson, D.W., 1997. Phenotypic
K.J. Rickards et al. / Veterinary Immunology and Immunopathology 98 (2004) 153–165 characterization of lymphocyte subpopulations in horses affected with chronic obstructive pulmonary disease and in normal controls. Vet. Pathol. 34, 108–116. Way, K.J., Chou, E., King, G.L., 2000. Identification of PKCisoform-specific biological actions using pharmacological approaches. Trends Pharmacol. Sci. 21, 181–187.
165
Wilkinson, S.E., Nixon, J.S., 1998. T-cell signal transduction and the role of protein kinase C. Cell Mol. Life Sci. 54, 1122–1144. Winder, N.C., von Fellenberg, R., 1988. Chronic small airway disease in the horse: immunohistochemical evaluation of lungs with mild, moderate and severe lesions. Vet. Rec. 122, 181–183.