The primary mechanism of action in vivo of mycophenolate mofetil (MMF) is believed to be inhibition of lymphocyte proliferation. We used novel assays of lymphocyte functions (pharmacodynamics, PD) in whole blood collected from rat heart allograft recipients treated with MMF to investigate the mechanisms of action of the active metabolite of MMF, mycophenolate acid (MPA) in vivo. Allograft recipients were treated orally once daily with 3 different doses of MMF. Seven days after transplantation, blood was collected 24 h after the penultimate dose and several timepoints after the last dose, after which grafts were removed for microscopic grading of rejection. Lymphocytes in whole blood samples were mitogen stimulated through calcium-dependent and -independent signaling pathways. Inhibition of PD was measured by lymphocyte proliferation and expression of several surface antigens on T cells, and was calculated as area under the time-inhibition of immune function effect curve (AUE0−24 h). We found that inhibition of lymphocyte proliferation and antigen expression by MPA correlated highly with MMF-dose, MPA level and with the histologic severities of graft rejection (p < 0.05). In summary, MPA suppressed lymphocyte proliferation and expression of T-cell surface antigens in whole blood collected from MMF-treated allograft recipients, thus demonstrating the multiple mechanisms of suppression of rejection on peripheral blood T cells after MMF treatment.
Mycophenolate mofetil (MMF) is a prodrug, which is rapidly converted to its active metabolite mycophenolic acid (MPA) after administration (1). The primary mechanism of action of MPA is presumed to be anti-lymphoproliferative, a result of inhibition of inosine 5′-monophosphate dehydrogenase (IMPDH), which is required for the de novo synthesis of guanosine nucleotides which are necessary for DNA and RNA synthesis and for lymphocytes to proliferate maximally after stimulation (2).
Over 10 years ago, mycophenolate mofetil (MMF, research compound number RS61443) was first shown to prolong organ allograft survival (3). Additional early studies showed that MMF prevents acute rejection, reverses acute rejection and increases graft survival in several species and in different animal transplantation studies (4–6). Shortly thereafter, the first human renal allograft recipients were treated with MMF (7).
MMF is approved for renal and heart transplant recipients as doses of 1–1.5 g orally twice a day. This regimen is derived from the three multicenter renal transplant trials of MMF for prevention of renal allograft rejection (8). Since fixed dose regimens do not account for inter-individual differences in MPA pharmacokinetics (PK), recent work has focused on the therapeutic drug monitoring of MPA and its correlation with drug efficacy (9). A randomized MMF concentration-controlled trial (RCCT) showed a significant correlation between prevention of graft rejection and MPA plasma trough levels and MPA-area under the concentration-time curve (AUC0−24 h) (10). A prospective study in renal transplant patients has been designed to maintain MPA trough and AUC0−24 h concentrations within a target range (11). In an uncontrolled study in heart transplant patients, MMF dose adjustment according to MPA trough levels was claimed to reduce acute rejection (12).
Clinical experience, however, has shown that monitoring of drug plasma concentrations alone does not accurately reflect a drug effect on immune cell functions (13). Therefore, reliance on PK values for selecting doses of immunosuppressive drugs risks over- and under- immunosuppression, leading to infection and malignancy on one hand and rejection on the other. The ability to measure the effect of immunosuppressants on diverse immune cell functions (pharmacodynamics, PD) should increase drug efficacy and safety. The relevance of AUC, maximal concentration (Cmax) and trough concentration (Ctrough) values for predicting the efficacies and safety is not fully known for all clinically approved immunosuppressants. Therefore, determination of the relationships between PD and PK may improve utility of PK monitoring (14).
We have used whole blood for PD assays instead of isolated mononuclear cells, because of the advantages of this method we have described previously (15). Initially, we measured the effect of single and multiple doses of orally administered MPA on suppression of lymphocyte proliferation and expression of CD25 and CD134 on T cells in nontransplanted rats, and showed high correlations between MPA PD and PK (16). We also showed that PD predicts the histologic severity of rejection in rat heart allograft recipients treated with MPA (17). To define the mechanisms of action of orally administered MPA more clearly, we developed methods that allowed us to expand our PD assays to detect the expression of additional cell surface molecules: CD71 (transferrin-R), CD11a (leukocyte function antigen-1, LFA-1, α-chain) and CD54 (intercellular adhesion molecule-1, ICAM-1). In addition to our previous use of Con A-induced T cell signaling, we recently optimized conditions for mitogenic signaling of T cells through additional pathways by phorbol 12-myristate 13-acetate (PMA) plus ionomycin and PMA plus anti-CD28 monoclonal antibody (mAb) (18).
In all previous studies, we had treated rats with MPA, but in our most recent work (19), we used our expanded PD assays to define the effect of different dose levels of MMF on lymphocytes in nontransplanted rats. In these studies, we showed that lymphocyte proliferation and expression of multiple T-cell surface antigens are suppressed after addition of MPA to rat blood in vitro and during treatment of nontransplanted rats with MMF in vivo. These studies of MMF in vivo also showed high correlations between the PD and PK of MPA (19).
The present study was designed to use our improved PD assays to determine the relationships between PD and severity of heart allograft rejection in rats treated with MMF.
Materials and Methods
Male Lewis (LEW) rats were recipients of BN (Brown Norway) rat hearts. All allograft groups consisted of 6 rats each and were treated once daily with vehicle or 5, 10 or 20 mg/kg of MMF. Recipients in the isograft (LEW to LEW) control group (n = 3 rats) were treated with vehicle. MMF (a gift from Roche Bioscience, Palo Alto, CA, USA) dissolved in vehicle, or vehicle was administered daily by oral gavage beginning on the day of transplantation (day 0).
On day 7 after heart transplantation, blood (800 μL) was sampled by retro-orbital bleeding under ether anesthesia before (PK trough) the final vehicle or MMF treatment and also 6, 12 and 24 h after the last treatment. The anticoagulated blood (100 U sodium heparin/mL blood) was used for cell blood counts (CBC, Coulter Microdiff II cell counter (Software 4CRplus; Coulter, Miami, FL, USA) and for PD whole-blood mitogen-stimulated lymphocyte function assays. For each time-point, tritium-labeled thymidine ([3H]-TdR) incorporation, proliferating cell nuclear antigen (PCNA/DNA) content and expression of cell-surface antigens (CD25, CD134, CD71, CD11a and CD54) were determined. The remaining plasma was frozen for determination of MPA plasma concentration by high performance liquid chromatography (HPLC)-analysis.
All grafts were removed on day 8 after heart transplantation for histologic examination.
Viral antibody-free adult male LEW (RT1l/CrlBR) rats weighing 300–350 g, 11–15 weeks of age and BN (RT1n) rats weighing 250–350 g, 11–16 weeks of age, both purchased from Charles River Laboratories (Wilmington, MA, USA) were housed in polycarbonate micro-isolation cages. Standard diet and tap water were provided ad libitum and animals were acclimated under a 12-h light/dark cycle for 2 weeks before the study began. The study was approved by the Institutional Animal Care and Use Committee. The animals received humane care in compliance with the ‘Principles of Laboratory Animal Care’, formulated by the National Society for Medical Research, and the ‘Guide for the Care and Use of Laboratory Animals’, prepared by the National Institutes of Health (NIH Publication no. 80–123, revised 1985).
Heterotopic cardiac grafts
Intra-abdominal heterotopic cardiac grafting was performed using a modification of the technique described by Ono and Lindsey (20). Briefly, the donor animals were anesthetized with sodium pentobarbital (50 mg/kg intraperitoneally). After laparotomy, the donor animal was heparinized (1000 I.U. heparin/kg i.v.). The pulmonary artery and aorta were transected 3–4 mm above their origins in the heart. The vena cava and pulmonary vein were ligated with 5–0 vicryl. Cold saline was injected into the aorta and the heart was placed in cold saline. The recipient animal was anesthetized with sodium pentobarbital (50 mg/kg intraperitoneally). After a median laparotomy, the abdominal aorta and vena cava were dissected free over a length of 2 cm and clamped using a curved clamp. The graft was implanted with both anastomosis done end-to-side in a running fashion with a 8–0 ethilon on a BV 130–4 needle (Ethicon Inc., Someville, NJ, USA). Surgery time ranged from 32 to 48 min, with a total ischemia time from 22 to 32 min, with a 100% technical success rate. The grafts were evaluated daily by abdominal palpation and scored on an scale from 0 to 4 (0 = no palpable graft contractility; 1 = barely palpable; 2 = weak contractility; 3 = medium contractility; 4 = strong contractility). On the final day of the experiment, all heart grafts were removed for histological examination. The hearts were cut in 3 transverse sections prior to fixation in 10% neutral, buffered formalin solution. After at least 24 h fixation, transverse sections of paraffin embedded tissue were further processed for histology.
All grafts were analyzed and graded by a pathologist, who was blinded to the study groups, after sectioning and staining with hematoxylin and eosin. Histology findings were scored for acute rejection and graded for severity using a scale from 0 (no histologic signs of acute rejection); 1 (mild lymphocyte infiltration, no necrosis, no hemorrhage); 2 (moderate mononuclear or lymphocyte infiltration, minimal necrosis, no hemorrhage); 2.5 (moderate to severe inflammatory infiltration) to 3 (severe rejection with necrosis hemorrhage and a marked acute inflammatory infiltrate or necrosis with loss of cell morphology or worse).
Culture medium (CM) was prepared using RPMI 1640 supplemented with 100 U/mL of penicillin, 100 mg/mL streptomycin, and l-glutamine (2 mm), all obtained from Sigma Chemical Co. (St. Louis, MO, USA). Con A (Vector Laboratories, Inc.; Burlingame, CA, USA) was diluted in CM, sterile filtered (2 μm, Applied Scientific, San Francisco, CA, USA) and stored at − 70 °C. PMA (Sigma Chemical Co.) was dissolved in dimethyl sulfoxide (DMSO, Sigma Chemical Co.) at a stock solution of 100 mg/mL and stored at − 20 °C. Working solutions of PMA were freshly prepared in phosphate buffered saline (PBS, Coulter, Miami, FL, USA). Purified mouse mAb anti-CD28 (Clone: JJ319; Pharmingen, San Diego, CA, USA) was diluted in CM before each experiment. [3H]-TdR with a specific activity of 6.7 Ci/mmol was purchased from New England Nuclear (Boston, MA, USA). Vehicle was prepared by dissolving sodium carboxymethylcellulose (City Chemical Corp., New York, NY, USA; viscosity of 1300–2200 centipoises at 25C in a 1% weight/volume deionized water solution) in 0.9% benzyl alcohol, 0.4% polysorbate 80, 0.9% sodium chloride in water to produce a 0.2% solution, and was stored at room temperature.
All fluorescein isothiocyanate- (FITC), phycoerythrin- (PE) and biotin-labeled mAb (anti-CD25, anti-CD134 anti-CD71, anti-CD11a, anti-CD54 and anti-PAN-T cell) and the mouse isotype controls IgG2a, IgG2b and IgG1 were purchased from Pharmingen. Streptavidin phycoerythrincyanin5 (PECy5) and anti-PCNA (Clone PC 10) were purchased from Dako Corporation (Carpinteria, CA, USA). RNAse, propidium iodide (PI), saponin and sodium azide were purchased from Sigma Chemical. Red blood cell (RBC) lysis buffer was made daily with 8.29 g NH4CL, 1 g KHCO3, 37.2 mg Na2-EDTA (J.T. Baker Chemical Corp, Phillipsburg, NJ, USA) in 1 liter of deionized H2O (pH 7.2). Permeabilizing buffer contained 1% heat inactivated fetal calf serum (FCS, Hyclone, Logan, UT, USA), 0.1% saponin (volume/volume) and 0.1% sodium azide (weight/volume) in PBS. Pure formaldehyde solution was purchased from Fluka Chemie AG, Switzerland, and methanol was purchased from Fischer Scientific, CA, USA.
Whole blood mitogen-stimulated lymphocyte function assay
One hundred and fifty microliters of diluted blood was added to wells of flat-bottom 96-well tissue culture microtiter plates (Becton Dickinson Labware, Franklin Lakes, NJ, USA). To each well, 50 μL of Con A, or 25 μL of PMA plus 25 μL of anti-CD28, or 50 μL of CM (unstimulated) were added to reach a final volume of 200 μL. The final dilution of blood in the well was 1 : 10. The final concentration in 200 μL blood cultures was 15 μg/mL for Con A, 50 ng/mL for PMA and 10 μg/mL for anti-CD28. Cultures were analyzed by [3H]-TdR incorporation on day 4 or by flow cytometry on day 3 after Con A-stimulation. The PMA + anti-CD28 assay cultures were analyzed after 2 days by [3H]-TdR incorporation or flow cytometry. All cultures were incubated at 37C in a humidified 5% CO2-air, water jacketed incubator.
Assessment of lymphocyte proliferation by [3H]-TdR incorporation
Ten microliters of [3H]-TdR(1 μCi) diluted in RPMI were added to each of 5 replicate wells during the last 16–24 h of incubation. Cells were collected onto glass fiber filters and washed with PBS using a TOMTEC multi-channel cell harvester (Wallac Oy, Turku, Finland). Methanol was used in the last wash to bleach the sample in order to reduce the color-quenching effect (21). Each filtermat was placed in a sample bag and filled with 4.5 mL OptiScint ‘HiSafe’ scintillation fluid (FSA Laboratory Supplies, Loughborough, UK). Beta emissions from each filtermat were quantified as counts per minute (CPM) in a Wallac 1450 Micro Beta Plus liquid scintillation counter (Wallac Oy, Turku, Finland). The background activity defined as CPM in empty wells was below 100 CPM. The mean CPM of five replicates was calculated and the coefficient of variation was not greater than 20%.
Analysis of lymphocyte functions by flow cytometry
After different durations of incubation depending whether Con A or PMA + anti-CD28 was used as mitogen, the contents from 7 wells of a 96-well plate were pooled (1400 μL). Eight hundred microliters were used for measurement of PCNA expression and DNA content and 600 μL were used to detect surface antigens. All samples were analyzed with an Epics XL-MCL flow cytometer equipped with an air-cooled argon laser (488 nm) using system II Coulter software (Coulter Corporation, Miami, FL, USA). Emitted light of the fluorochromes was collected with bandpass filters through 525 nm (FITC), 575 nm (PE) and 675 nm (PECy5 and PI). Unstimulated cultured blood was used as a negative control. Specificity controls included replacement of primary mAbs with isotype mouse immunoglobulins.
We performed a simultaneous analysis of PCNA and DNA content expression with the combination of FITC-labeled anti-PCNA mAb and PI to determine the percentage of cells in S or G2/M-phase of the cell cycle. For simplicity, these cells are denoted S/G2M positive. Eight milliliters of lysis buffer were mixed with 800 μL cultured blood and RBCs were lysed for 10 min at room temperature. Leukocytes were pelleted (200 g, 10 min) and again washed with 2 mL PBS (200 g, 5 min). Leukocytes were fixed by adding 1 mL of PBS containing 1% (volume/volume) formaldehyde and then kept on ice for 5 min. After washing with 2 mL PBS (200 g, 5 min), cells were resuspended with 2 mL ice-cold 100% methanol and stored for at least 10 min at 4 °C. Prior to staining, cells were washed with 2 mL PBS (200 g, 5 min) to remove methanol. The cell pellet was then resuspended in a staining mixture containing 107 μL permeabilizing buffer, 10 μL RNAse A (stock solution 100 mg/mL in H2O), 5 μL PI (stock solution 1 mg/mL in H20) and 2.5 μL anti-PCNA FITC mAb. Cells were incubated in the staining mixture overnight at room temperature. After washing with 2 mL PBS, leukocytes were pelleted (200 g, 5 min) and resuspended in 500 μL PBS containing 10 μg/mL PI. Total lymphocytes were collected using forward and side scatter and displayed as dot plots in 2-color PCNA/DNA distributions as PCNA-positive cells with S/G2M-phase DNA content. Five thousand events were analyzed per sample.
Expression of mitogen-stimulated cell surface antigens
We performed 2- and 3-color analysis for detection of cell surface antigen expression on T-lymphocytes. To each 200 μL incubated blood from pooled cultures, 100 μL of mAb (1 : 100 dilution in PBS containing 3% FCS and 0.1% NaN3) was added. After incubation for 30 min in the dark at room temperature, samples containing biotin-labeled mAb were washed with 2 mL PBS. To these samples, 20 μL of streptavidin PECy5 (1 : 20 dilution in PBS containing 3% FCS and 0.1% NaN3) was added and then incubated for 15 min in the dark at room temperature. Following staining, 4 mL lysis buffer was added to each tube and RBCs were lysed at room temperature for 10 min. Leukocytes were pelleted (200 g, 10 min) and again washed with 2 mL PBS (200 g, 5 min). Before analysis, leukocytes were resuspended in 500 μL PBS containing 0.5% (volume/volume) formaldehyde. Forward and side scatter were chosen for differentiating viable lymphocytes from debris, dead cells and other leukocytes, as previously described (22). Data were displayed as histograms and as dual color dot plots to measure per cent positive cells. Five thousand events for each sample were analyzed.
In rats the erythrocyte/plasma ratio is as low as 0.1–0.15 and the bound fraction of MPA in plasma is 98% (23). Therefore, plasma is the matrix of choice for measurement of MPA concentrations (24). Drug quantification in rat plasma was done by gradient HPLC, as described previously (25). To 50 μL plasma, naproxen (Sigma Chemicals) in acetonitrile (single step extraction) was added as an internal standard. The mobile phase consisted of acetonitrile and sulfuric acid adjusted to pH 4.0. A Sphereclone 5 μm ODS (2), 200 × 4.6 mm column from Phenomenex (Torrance, CA, USA) was used. Samples were analyzed on a binary HPLC-system equipped with an autosampler (Shimadzu Kyoto, Japan). The MPA concentrations were calculated using calibration curves and were corrected using the internal standard. The area under the plasma concentration–time curve (AUC0–24 h) was calculated by the trapezoidal rule using the MPA plasma concentrations 6, 12 and 24 h after dosing. Pharmacokinetic data from studies with MPA in rats (25) showed that this ‘abbreviated’ AUC is, on average, 85% lower than the AUC calculated from MPA plasma concentrations analyzed after 0.5, 1, 2, 6, 12 and 24 h.
Calculation of MPA pharmacodynamics
The inhibitory effect of MPA was expressed as per cent inhibition of [3H]-TdR incorporation or inhibition of expression of lymphocyte function analyzed by flow cytometry and calculated as follows:
Per cent inhibition = [1 - (Treatment/Pretreatment)] × 100
‘Pretreatment’ represents the results obtained from stimulated blood without addition of drugs. ‘Treatment’ represents the results from stimulated blood drawn on day 7 and 8 after transplantation. MPA plasma concentrations, which produced a 50% inhibition (IC50) of the maximum inhibitory effect (Imax), were calculated after fitting the concentration – effect curves in an Imax sigmoidal pharmacodynamic model using WinNonlin Software version 1.1 (Scientific Consulting, Inc., Cary, NC, USA). Analogous to the calculation of the AUC0−24 h, the area under the pharmacodynamic-effect time curve (AUE0−24 h) was calculated using the trapezoidal rule:
AUE0 - 24 h = [(E0 h + E6 h)/2 × 6] + [(E6 h + E12 h)/2 × 6] + [(E12 h + E24 h)/2 × 12]
and expressed the inhibitory effect of MPA on lymphocyte function over 24 h after dosing. Furthermore, we observed the maximal and the trough MPA effect on lymphocyte function (Emax and Etrough).
All data are expressed as mean ±standard error of mean (SEM). For statistical analysis, SPSS version 8.0 (SPSS Corp., Birmingham, AL, USA) and SigmaStat version 2.0 (Jandel Scientific) were used. The 2-tailed Student's t-test was used to estimate the levels of significance for the differences between 2 groups with equal variance. The Kruskal–Wallis one-way analysis of variance on ranks (anova on ranks) with the Student-Newman-Keuls (SNK) post-hoc test were used to estimate the levels of significance for the differences between two groups. Correlations between PD, PK and histology scores were determined using Pearson product or the Spearman rank order (anova on ranks). A p-value of less than 0.05 was considered significant.
All animals survived until the end of the studies. No morbidity caused by retroorbital bleeding or drug treatment was observed. All animals gained weight during the study period.
Relationships among pharmacokinetics and different dose levels
Within each dose group, inter-rat variations in MPA plasma concentrations were minimal (Figure 1). The pharmacokinetic profiles were distinct for each dose group, demonstrating a close relationship between dose level and MPA plasma concentrations. Twenty-four hours after treatment, no (5 mg/kg) or very little (10, 20 mg/kg) MPA was detected in plasma. The areas under the plasma MPA concentration-time curves (AUC0−24 h) after administration of eight doses MMF were dose-related: 5 mg/kg = 8.2 ± 1.5 mg/L* h, 10 mg/kg = 24.4 ± 2.6 mg/L* h and 20 mg/kg = 41.2 ± 5.1 mg/L* h. There was a high correlation between dose level and MPA AUC0−24 h (r2 = 0.85) (Table 1), and MPA dose level and 6 h plasma concentration (r2 = 0.88), but there was a poor correlation between dose level and MPA trough concentration at time 0 before the last dose (r2 = 0.23) (Table 1).
Table 1. : Correlation between pharmacokinetics (PK), MPA dose and severity of histologic graft rejection. Eight-day treatment studies in rat heart allograft recipients. Correlations between PK and: MPA dose levels and histology scores for graft rejection. Correlation between dose levels and histology scores
MPA potency and maximum inhibition of lymphocyte functions
The effect of MPA on different lymphocyte functions was determined after administration of eight consecutive daily doses of MMF at three dose levels in heart transplanted recipients. Blood was sampled 1 day before transplantation and before start of drug treatment (pretreatment), before (time 0 h, trough) and at three successive times after the eighth dose. Figure 2 shows clear concentration-dependent inhibitions of per cent expression of CD25, CD71, CD11a and CD54 after Con A- or PMA + anti-CD28-stimulation (Figure 2a–h). Similar results were observed for inhibition of [3H]-TdR incorporation, S/G2M positive cells and per cent cells expressing CD134 (data not shown).
For stimulation by either Con A or PMA + anti-CD28, the MPA plasma concentration producing 50% of the maximum inhibition (IC50) of lymphocyte proliferation measured by [3H]-TdR incorporation was lower than the IC50 measured by S/G2M positive (Figure 3a). MPA potency was lowest for inhibition of CD25 expression in both mitogen-stimulated assays (Figure 3a). IC50 values for expression of all other cell-surface antigens in the Con A assay were similar. In the PMA + anti-CD28 assay CD71 expression was inhibited more potently by MPA compared to expression of CD134, CD11a or CD54 (Figure 3a).
The calculated maximum per cent inhibition (Imax) of lymphocyte function by MPA was 100% for both [3H]-TdR incorporation and S/G2M cells, regardless of the mitogen used (Figure 3b). MPA substantially, but incompletely, inhibited expression of cell-surface antigens after stimulation with Con A or PMA + anti-CD28, except for a complete inhibition of CD71 expression after PMA + anti-CD28-stimulation (Figure 3b). The maximal inhibitory effect of MPA on CD25 expression was least (Figure 3b) compared to the other cell surface antigens.
Relationships among MMF dose levels, pharmacokinetics and pharmacodynamics in heart-transplanted recipients
Figure 4 shows the per cent inhibition of lymphocyte functions (pharmacodynamics) in each dose group as a function over time after administration of the last of eight MMF or vehicle doses in heart-transplanted rats. Most of the data show a clear and direct relationship among dose levels and magnitudes and durations of PD effects on lymphocyte proliferation (Figure 4a–d) and on cell-surface antigens (Figure 4e–n) for both mitogen-stimulated assays. Figure 4 shows a low inter-rat variability for PD measurements in the Con A and the PMA + anti-CD28 assays. In both mitogen-stimulated assays, the dynamics of inhibition of lymphocyte proliferation and of surface antigen expression over time were inversely related to MPA plasma concentrations (Figure 1). Maximum inhibitions of all lymphocyte functions occurred 6 h after dosing, which were simultaneous with the highest MPA plasma concentration in all groups. Con A-stimulated lymphocyte proliferation measured by [3H]-TdR incorporation was almost completely inhibited after administration of 5 mg/kg of MMF (96 ± 1%) and remained suppressed 24 h after the last dose (Figure 4a). In the PMA + anti-CD28 assay, inhibition of [3H]-TdR incorporation was less (71 ± 12%) with 5 mg/kg of MMF 6 h after the last dose, and recovered after 12 h (Figure 4b).
Compared to [3H]-TdR incorporation, higher dose levels of MMF were required to inhibit lymphocyte proliferation measured by S/G2M-positive cells in both mitogen assays. In addition, expression of S/G2M-positive cells for all dose groups recovered to pretreatment values regardless of the source of mitogen stimulation (Figure 4c,d). Inhibitions of surface antigen expression measured 6 h after administration of 20 mg/kg of MMF were higher in the Con A assay compared to inhibitions after PMA + anti-CD28-stimulation (Figure 4e–n). Regardless of the mitogen used, expression of cell-surface antigens recovered to pretreatment levels 24 h after the last dose (Figure 4e–n).
Vehicle treated isograft and allograft heart transplanted recipients were used to study the effects of surgery, graft rejection, vehicle treatment and multiple bleeding on lymphocyte function. As shown in Figure 4(a–n) for vehicle treated recipients, none of these manipulations suppressed lymphocyte functions in our whole blood assays (data for isograft recipients not shown).
Relationships among pharmacodynamics and MMF dose levels
We also analyzed the relationships between MMF dose levels and PD effects throughout the dose interval. For analysis of PK, exposures to drugs are often described by areas under the concentration-time curves (AUC0−24 h). Since the dynamics of lymphocyte functions were determined over time (Figure 4), we used these data to calculate the areas under the pharmacodynamic effect-time curves during the 24 h after the last dose (AUE0−24 h). The AUE0−24 h showed statistically significant dose-dependent increases (anova, p < 0.001) among all dose groups for all lymphocyte functions in both mitogen-stimulated assays (Figure 5a,b). Further analysis using the SNK post hoc test showed a significant difference in AUE0−24 h between 5 and 20 mg/kg MMF in both mitogen-stimulated assays (p < 0.05). Significant differences between the other dose groups are shown in Figure 5(a,b) (p < 0.05).
Relationships between pharmacokinetics and pharmacodynamics for each dose level
Table 2 shows detailed data on the coefficients of correlation between different measures of PD and PK. Regardless of the mitogen stimulation, the AUE0−24 h for all lymphocyte functions correlated highly with both MMF dose level and MPA exposure (AUC0−24 h) (Table 2).
Table 2. : Correlation between PD and PK, MPA dose and histologic graft rejection. Eight-day treatment studies in heart allograft recipients treated daily with vehicle or 5, 10 or 20 mg/kg MMF. Correlations between AUE0−24 h for all 7 pharmacodynamic measurements and: MPA dose levels, AUC0−24 h and histology scores for graft rejection
aPharmacodynamic data were derived from results in Figure 4 in which whole blood pharmacodynamic assays were used (see legend for Figure 4). Area under effect-time curve (AUE)0−24 h was calculated using the linear trapezoidal rule.
Dose level groups: vehicle, 5, 10 and 20 mg/kg MMF.
c MPA plasma area under concentration-time curve (AUC)0−24 h was calculated from MPA plasma concentrations assayed on day 7 before drug administration (trough) and 6, 12 and 24 h after drug administration using the linear trapezoidal rule.
Rejection was scored on a scale from 0 (no rejection) to 3 (severe rejection with necrosis of the graft or worse) by a pathologist blinded for the groups.
Phorbol 12-myristate 13-acetate.
Anti-CD28 monoclonal antibody.
Tritium-labeled thymidine measured by counts in min.
i Positive cells in S/G2M-phase of the cell cycle measured by flow cytometric analysis.
α-chain expression measured by flow cytometric analysis.
Ox40 receptor, member of the nerve growth factor/tumor necrosis superfamily, expression measured by flow cytometric analysis.
Transferrin receptor expression measured by flow cytometric analysis.
Leukocyte function antigen-1 receptor
α-chain expression measured by flow cytometric analysis
Intercellular adhesion molecule-1 expression measured by flow cytometric analysis.
Pearson product moment, all correlations were significant (p < 0.05).
Spearman rank order, all correlations were significant (p < 0.05).
Correlations between histology scores, MMF dose levels, pharmacokinetics and pharmacodynamics
Another important part of this study was to determine the correlation between the histologic grade of rejection and the following: dose level, measurements of PK and measurements of PD. The median histology score for vehicle treated isografted hearts was 0 (no rejection). For allograft hearts in different treatment groups, the median histology scores were: vehicle = 3, MMF 5 mg/kg = 2.5, MMF 10 mg/kg = 1.5 and MMF 20 mg/kg = 1 (Figure 5). There were significant differences in histology scores, among the treatment groups (p < 0.05, anova, SNK post hoc), except between 10 and 20 mg/kg MMF. Paired comparisons showed a decrease of histologic severity of rejection with increasing dose. The correlation between dose level and rejection score was high (r = 0.88, Table 1). As was the correlation between PK and histology score (Table 1). We found that both AUC0−24 h and C6 h MPA plasma levels correlated well with histology score (r2 = 0.83 and r2 = 0.85, respectively). For trough MPA plasma concentrations, this correlation was lower (Table 1). Furthermore, we investigated the relationships between PD measurements of immune suppression and histology scores in MMF-treated rats. Table 2 shows that AUE0−24 h measured by flow cytometric analysis of T-lymphocyte functions correlated more highly with the histology score than PD measured by [3H]-TdR incorporation in the Con A assay. MPA PD 24 h (Etrough) or the maximal PD (Emax) effect after the last dose correlated less highly with histology score (Etrough: r2 = 0.14–0.38 Con A, r2 = 0.13–0.69 PMA + anti-CD28; Emax: r2 = 0.62–0.83 Con A, r2 = 0.36–0.79 PMA + anti-CD28) compared with the AUE0−24 h correlations.
In earlier studies, we demonstrated that MPA (not MMF) treatment inhibits expression of CD25 and CD134 on lymphocytes (16) and that these PD measurements in MPA-treated rat heart allograft recipients correlate highly with histologic graft rejection (17). Next, we improved upon our whole blood assays (18) and used these assays to quantitate the PD effects of MMF administered to nontransplanted rats (19). In the present study, we gained further insights into the mechanisms of action of MMF and showed correlations of MMF PD with severities of histopathologic graft rejection in MMF treated rat heart allograft recipients.
An important part of our study was to determine the correlation of PD with MMF dose, PK and histologic graft rejection. We chose 6 h after dosing with MMF as the earliest time to assess PD of MMF. After oral administration of MPA in nontransplanted rats, our PK studies showed a first peak of MPA plasma concentration at 0.5 h and a second peak between 6 and 12 h (25). These findings were confirmed by previous results of MMF PK studies in nontransplanted rats (23). Furthermore, we found no significant differences in the PD of MMF between 6 h after dosing and earlier timepoints (unpublished data). Due to the small blood volume of rats, we minimize the risk of bleeding and due to repetitive anesthesia required for multiple timepoints, blood was collected 6, 12 and 24 h after dosing.
Our study confirmed the inhibitory effect of MMF on lymphocyte proliferation measured by [3H-TdR]-incorporation (Figure 4a,b) and its inhibition of the flow cytometric detection of S/G2M cells (Figure 4c,d). Figure 4 clearly shows the difference in sensitivities between these two methods to detect the inhibition by MMF on the proliferative response of stimulated lymphocytes. Proliferation measured by [3H]-TdR incorporation did not distinguish between different MMF dose groups when MPA plasma concentrations were above 1 mg/L, because MPA plasma concentrations below 1 mg/L already caused complete inhibition of proliferation measured with this method. Using the flow cytometric analysis for proliferation (biphasic S/G2M-DNA detection), however, we showed that an increasing inhibitory effect is found throughout the 0–5 mg/L MPA plasma concentration range. Therefore, the coefficients of correlation between the AUE0−24 h and dose levels or plasma concentrations (AUC0−24 h) are higher for S/G2M expression than the coefficients of correlation between AUE0−24 h and dose levels or AUC0−24 h[3H]-TdR incorporation (Table 2). The difference between [3H]-TdR incorporation and flow cytometric measurements of lymphocyte proliferation can be explained by factors inherent to each technique as described previously (16).
A dose-dependent increase in the inhibitory effect of MMF on the expression of all measured T-cell surface antigens was observed, as reflected by the high coefficients of correlation between the respective AUE0−24 h and dose levels or plasma concentrations (AUC0−24 h) (Table 2).
Furthermore, we observed that there was an inverse relationship between PD and PK (Figures 1 and 4) and of the inhibitory effect of MMF measured by its antiproliferative effect within the first 24 h after oral administration was reversible (Figure 4). Studies of inhibition of IMPDH also show a trend toward an inverse relationship between the enzyme activity and MPA concentrations (26). A similar reversibility and PD-PK inverse relationship were observed for the inhibition of the expression of T-cell surface antigens (Figure 4).
Since PD measures inhibition of immune functions in blood of MMF-treated rats, it was important to determine whether differences in PD predicted the effects of MMF on the histologic severity of graft rejection.
First, we showed that the histologic graft rejection score decreased with increasing MMF dose. We also showed that the correlation between histology scores and trough MPA plasma concentrations was poor (r2 = 0.57), but higher for Cmax (r2 = 0.85) or AUC0−24 h (r2 = 0.83). For the PD values, the coefficients of correlation were similarly high (r2 between 0.70 and 0.85) for each mitogen and for both suppression of lymphocyte proliferation and suppression of T-cell surface antigen expression. The high correlations between graft histology scores and the PD values suggest that PD can be an alternative to or a useful addition to PK monitoring of the immunosuppressive effects of MMF.
MMF is a prodrug, which is rapidly converted to MPA. MPA is an effective and reversible uncompetitive inhibitor of IMPDH, causing depletion of intracellular GTP pools. The resulting inhibition of lymphocyte proliferation is thought to largely explain the immunosuppressive action of MMF in vivo (27).
Studies with isolated mononuclear human cells show that MPA eliminates the p27Kip1 inhibitory activity and inhibits the induction of cyclin D/CDK6 kinase, which leads to cell cycle arrest in early to mid-G1-phase (28). Other studies of Con A-stimulated human T lymphocytes show that MPA induces GTP depletion and decreases ATP pools, resulting in a lack of pyrimidine synthesis (29).
In our present study, we observed that MMF inhibited expression of surface antigens on peripheral blood lymphocytes after stimulation with both calcium-dependent (Con A) and calcium-independent stimulation (PMA + anti-CD28) in allograft recipients. Remarkably, the IC50s for the expression of surface antigens and the IC50s for proliferation were similar. In general, IC50s were lower for Con A-stimulation than for stimulation by PMA + anti-CD28. This difference between the two mitogens might be due to differences in the modes of action of Con A and PMA + anti-CD28 (18). We showed for the first time that suppression of both calcium-dependent and -independent stimulation of expression of T-cell surface antigens by MMF treatment correlated highly with the severity of histologic graft rejection.
Previous studies of MPA's biochemical effects may explain the mechanisms of action by which MMF suppressed expression of specific T-cell surface activation antigens. Proliferating lymphocytes depend on both the purine and the pyrimidine pathways to supply ribonucleotides necessary for DNA, RNA and protein synthesis as well as for ribonucleotide intermediates, which are required for proper lipid and protein glycosylation and membrane synthesis (30,31). In addition, studies in human cell lines show that MPA suppresses the synthesis of RNA-primed DNA intermediates (32). In other studies, MPA has been shown to suppress guanine nucleotide protein (G-protein) levels (33), which are required for activation of some T-cell surface antigens (34).
MPA also blocks the glycosylation of adhesion molecules (35,36); necessary for proper folding and regulation of other T-cell surface antigens (37,38). Recent studies, however, failed to show that MPA decreases glycosylation on Con A-stimulated lymphocytes (39).
Brequinar, an inhibitor of the de novo pyrimidine synthesis, decreases ATP pools, as does MPA in human lymphocytes (30). Besides its antiproliferative effect, brequinar suppresses IL-2 R expression in mitogen-stimulated peripheral blood mononuclear cells in vitro. Inhibition of mRNA synthesis and subsequent protein translation by brequinar may explain its suppression of IL-2 R expression (40). Similar mechanisms may be responsible for the inhibition by MMF of expression of T-cell surface antigens in our study. The importance of mRNA for regulation of receptor functions has been shown by others (41). In studies with human mixed lymphocyte reactions, MPA inhibits IL-2 R and transferrin R expression after 6 days of culture, but ineffectively inhibits IL-2 R or transferrin R expression after 3 days of culture (42). Since IL-2 R, LFA-1 or ICAM-1 on T cells are targets for therapeutic antibodies used to suppress rejection, suppression of their expression by MMF treatment emphasizes their important role in allograft rejection (43–47).
Our data in this study confirm that a substantial immunosuppressive mechanism of action of MMF is its antiproliferative effect. However, our data also suggest that MMF mediates its suppression of graft rejection by inhibition of the expression of several T-cell surface antigens. Our study has also demonstrated that our novel whole blood PD assays offer the opportunity to uncover new mechanisms of action of any immunosuppressive drug. This knowledge may be exploited to use immunosuppressive drugs more effectively and to combine different immunosuppressants in treatment regiments more intelligently. Although our results with MMF monotherapy in rat heart allograft recipients cannot predict the value of PD for therapeutic drug monitoring in patients, we have successfully used our whole blood flow cytometric PD assays to quantitate immunosuppression by MMF administered to nonhuman primates (48) and humans (49).
This work was supported by Roche Pharmaceutical Company, NJ, USA and by the Hedco Foundation and the Ralph and Marian Falk Trust. Markus J. Barten was supported by the Novartis study grant of the European Society of Transplantation. Teun van Gelder was supported by the Foundation ‘Vereniging Trustfonds Erasmus Universiteit Rotterdam’ in the Netherlands and by the Dutch Kidney Foundation. Jan F. Gummert was supported by Deutsche Forschungsgemeinschaft grant Gu 472/1–1 (J.F.G.).