Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Investigation of the Prognostic Value of TNF-α Gene Polymorphism among Patients Treated with Infliximab, and the Effects of Infliximab Therapy on TNF-α Production and Apoptosis

Investigation of the Prognostic Value of TNF-α Gene Polymorphism among Patients Treated with... Original Paper Received: February 27, 2004 Pathobiology 2004;71:274–280 Accepted: May 28, 2004 DOI: 10.1159/000080062 Investigation of the Prognostic Value of TNF- · Gene Polymorphism among Patients Treated with Infl iximab, and the Effects of Infl iximab Therapy on TNF- · Production and Apoptosis a a d b b Attila Balog Gergely Klausz János Gál Tamás Molnár Ferenc Nagy c a a Imre Ocsovszky Zsófi a Gyulai Yvette Mándi a b Department of Medical Microbiology and Immunobiology, First Department of Internal Medicine, c d Department of Biochemistry, Faculty of Medicine, University of Szeged, Szeged , Department of Rheumatology, County Hospital, Kecskemét , Hungary Key Words more suitable candidates for infl iximab treatment. Al- Tumor necrosis factor-· · Gene polymorphism · though in vitro infl iximab treatment for 48 h resulted in Infl iximab · Apoptosis · Intracellular cytokine · signifi cant (44.2 8 1.17%) apoptosis in PBMCs, in ex vivo Rheumatoid arthritis · Crohn’s disease samples from RA patients who received infl iximab, apoptosis was only 13.3 8 1.6%. Furthermore, infl iximab did not result in irreversible inhibition of the TNF- · -pro- Abstract ducing ability or in the signifi cant apoptosis of PBMCs. Objectives: Infl iximab, a chimeric anti-tumor necrosis Copyright © 2004 S. Karger AG, Basel factor (TNF) antibody, is highly effective for the treat- ment of Crohn’s disease (CD) and rheumatoid arthritis (RA). Our experiments focused on RA and CD patients Introduction receiving infl iximab. Since cytokine production is large- ly determined by genetic factors, the promoter polymor- The central role of tumor necrosis factor- · (TNF- · ) in phisms of TNF- · were examined among these patients. the pathogenesis of rheumatoid arthritis (RA) and Crohn’s Additionally, the changes caused by infl iximab in the disease (CD) has been well documented [1, 2] , and TNF- · -producing ability and apoptosis of peripheral TNF- · is therefore an excellent therapeutic target. One blood mononuclear cells (PBMCs) were investigated. of the most effective biological agents, infl iximab, a chi- Methods: The TNF- · genotypes were analyzed by PCR- meric anti-TNF antibody, has been shown to be highly RFLP. The in vitro TNF- · production of the PBMCs was effective for the treatment of both patients with CD [3] detected by fl ow cytometric analysis. The TNF- · concen- and patients with RA [4, 5] . The experiments reported tration in the supernatant was measured by bioassay. here focus on patients with active RA in whom tradition- Apoptosis was detected by annexin V-fl uorescein iso- al treatment with disease-modifying antirheumatic drugs thiocyanate labeling. Results and Conclusions: 8 of the had not been successful, and on patients with steroid-re- 12 nonresponder patients carried the TNF A allele associ- fractory CD. All of these patients underwent infl iximab ated with high TNF- · production. We suggest that the de- therapy. Despite the generally dramatic effi cacy of infl ix- termination of TNF polymorphism may help identify imab, some of our patients did not display a clinical re- © 2004 S. Karger AG, Basel Prof. Dr. Yvette Mándi 1015–2008/04/0715–0274$21.00/0 Department of Medical Microbiology and Immunobiology Fax +41 61 306 12 34 Faculty of Medicine, University of Szeged E-Mail karger@karger.ch Accessible online at: PO Box 427, Dóm tér 10, HU–6720 Szeged (Hungary) www.karger.com www.karger.com/pat Tel. +36 62 545115 , Fax +36 62 545113, E-Mail yvette@comser.szote.u-szeged.hu of Helsinki. The ethical committee of the participating university sponse. We hypothesized that a genetic predisposition approved the study. Informed written consent was obtained in ad- might be involved in the level of responsiveness. Since vance from each patient. cytokine production is largely determined by the poly- morphism of regions regulating cytokine expression, the DNA Extraction most frequently investigated promoter polymorphism at Genomic DNA from whole blood containing EDTA was ex- tracted using standard techniques (High Pure PCR Template Prep- the –308 position of TNF- · [6] was examined among RA aration Kit, Roche, catalog No. 1796828). and CD patients receiving infl iximab therapy. The less frequent TNF A allele of the –308 TNF- · gene polymor- –308 TNF- · Genotyping phism has been associated with increased TNF transcrip- The G to A transition at position –308 in the promoter region tion and production [7–9] and with more severe, either is associated with elevated expression of TNF- · . Using the method described by Wilson et al. [18] , a PCR was followed by digestion steroid-dependent or fi stulizing CD [10] . We therefore set with the endonuclease Nco I. Upon analysis by acrylamide gel elec- out to investigate the possible association between the – trophoresis, the TNF G allele gives two fragments of 87 and 20 bp, 308 TNF- · genotype and the response to infl iximab. while the TNF A allele gives a single, 107-bp fragment. A total of An effective host response depends in part on the cy- 9 patients with RA and 14 with CD as well as 75 healthy blood do- tokine-producing ability, including TNF- · [11] . Since nors were analyzed. there have been reports of adverse events with TNF in- Detection of TNF- · in Supernatant of Whole Blood Cultures hibitors, including infections and cases of sepsis [11–13] , TNF- · concentrations in the supernatants of whole blood cul- and since the mechanism of action of infl iximab is only ture (WBC) samples of the 9 patients with RA were determined by partly understood [5, 14, 15] , the changes caused by in- bioassay, applying the WEHI 164 mouse fi broblast cell line [19] . fl iximab in the TNF- · -producing ability of peripheral Human recombinant TNF- · (Amersham) was used as a standard in the same assay. The specifi city of the effect of TNF- · in the super- blood mononuclear cells (PBMCs) were investigated, to- natants was confi rmed via the neutralizing effect of monoclonal gether with apoptosis in PBMCs. anti-TNF- · antibody, clone TA-31 (Sigma product No. T1549). The amounts of TNF- · are expressed in U/ml. In control experiments, TNF- · concentrations in the supernatants of WBC samples were Patients and Methods also determined using a TNF- · ELISA kit (Biosource), according to the instructions of the manufacturer. For comparison, the amounts of TNF- · determined in the two assays are expressed in U/ml, where Patients Nine patients with disease-modifying antirheumatic drug-re- 1 unit is equivalent to 34 pg of recombinant human TNF- · . fractory RA were diagnosed according to the revised 1987 Ameri- can College of Rheumatology criteria [16] . The infusion regimen Isolation of Human Mononuclear Cells was as follows: 3 mg/kg infl iximab (Schering-Plough) administered Peripheral white blood cells were prepared by density gradient purifi cation (Ficoll-Paque, Sigma) from heparinized venous blood at 0, 2 and 6 weeks and then every 8 weeks for 1 year. The 14 patients with chronic active CD unresponsive to stan- samples from healthy volunteers and from RA patients treated with dard therapy were diagnosed in accordance with clinical guidelines infl iximab, and 95% cell viability was then confi rmed by trypan blue staining. Subsequently, 5 ! 10 /ml PBMCs suspended in me- and on the basis of conventional radiological, endoscopic and his- topathologic examinations [17] . Fistulas were a common complica- dium were used in all the experiments. Isolated cells were incu- tion. The infusion regimen was as follows: 5 mg/kg infl iximab ad- bated in RPMI-1640 medium (Gibco) supplemented with 10% fe- ministered at 0, 2 and 6 weeks. tal bovine serum, antibiotics and glutamine. All patients underwent regular control examinations. All pa- Intracellular Staining of TNF- · tients and controls were adults of Hungarian ethnic origin and re- sided in Hungary. PBMCs were incubated in RPMI-1640 medium (Gibco) supple- Following the infl iximab therapy, the RA patients were divided mented with 10% fetal bovine serum, antibiotics and glutamine for into responding and nonresponding groups according to the follow- 24 or 48 h in the presence or absence of infl iximab (100 Ì g/ml). Af- ing parameters: the DAS-28 (disease activity score), a health assess- ter centrifugation, the cells were washed with phosphate-buffered saline (PBS). They were then stimulated with heat-killed (90 ° C for ment questionnaire, a pain visual analog score, the swollen and tender joint index, morning stiffness, the C-reactive protein level, 10 min adjusted to 10 /ml) [20] Staphylococcus aureus (SA) as TNF the erythrocyte sedimentation rate, the physician’s global assess- inducer in the presence of brefeldin A (10 Ì g/ml) for 4 h at 37 ° C in ment, the patient’s global assessment, and the American College of a humidifi ed 5% CO incubator. After centrifugation, the cells were Rheumatology functional capacity score. After treatment with in- washed with PBS and permeabilized with Cytofi x-Cytoperm solu- tion (Becton-Dickinson) for 20 min at room temperature in the fl iximab, the CD patients were categorized as responders if steroid therapy could be stopped, if there were no draining fi stulas or most dark. Thereafter, they were washed twice with PermWash (Becton- of the fi stulas had closed, and if the number of liquid or very soft Dickinson) solution and once with added fl uorescein isothiocyanate (FITC)-labeled human TNF- · specifi c monoclonal antibodies stools had decreased markedly. This study was performed in accordance with the ethical stan- (MAbs) for 30 min at room temperature in the dark. They were next washed twice with PermWash and once with added PBS. Flow cy- dards laid down in the most recent version of the 1964 Declaration Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 275 Polymorphism and TNF- · Production tometric analysis was subsequently performed with a FACStar plus fl uorescence-activated cell sorter (Becton-Dickinson) at 488-nm ex- citation to estimate intracellular TNF- · in the peripheral mono- nuclear cells. Cells were gated for CD14-positive cells as follows. CD14 expression was detected with phycoerythrin-conjugated MAbs. At the same time, a dot plot display was used to identify monocytes, which correlated with the gate applied for CD14-posi- tive cells. In subsequent experiments, this gate was applied. Detection of Apoptosis with Annexin V Cell apoptosis was assessed by annexin V-FITC staining with a fl ow cytometric apoptosis detection kit (PharMingen, Becton-Dick- inson, catalog No. 556420). The PBMCs were incubated for 24–48 h in the presence or ab- sence of infl iximab (100 Ì g/ml), then washed twice in PBS and re- suspended in binding buffer (0.1 M Hepes-NaOH, 1.4 M NaCl, 25 m M CaCl ) at a concentration of 1 ! 10 cells/ml. 5 Ì l of an- Fig. 1. TNF- · concentration in supernatants of whole blood sam- nexin V-FITC and 10 Ì l of propidium iodide (PI) were added to a ples from 9 RA patients treated with infl iximab following heat- 100- Ì l aliquot of each cell solution. This was followed by gentle killed SA stimulation of the cells. A: Before infl iximab therapy; B: mixing and incubation for 15 min at room temperature in the dark. 48 h after the 3rd infusion; C: 48 h after the 4th infusion; D: 48 h Next, 400 Ì l of binding buffer was added to each tube. Flow cyto- after the 4th infusion – infl iximab was washed out from the whole metric analysis was performed within 30 min. blood cells. Data are means 8 SEM. Those cells that were negative for both dyes were considered to be live cells; necrotic cells were positive for both fl uorochromes or only for PI, and apoptotic cells were positive only for annexin V- FITC and negative for PI. Table 1. TNF- · genotypes in patients receiving infl iximab therapy and in control subjects Statistical Analysis The levels of intracellular TNF- · production with the SA in- Group TNF G/ TNF G/ TNF A/ ducer and the experimental results for apoptosis were compared by TNF G TNF A TNF A means of one-way ANOVA. The Student-Newman-Keuls test was used for post hoc pairwise multiple comparisons. In all tests, an · Patients treated with infl iximab (n = 23) level of p ! 0.05 was taken as an indication of statistical signifi cance. Responders 10 110 All statistical calculations were performed with the GraphPad Nonresponders 14 180 Prism4 statistical program. Healthy controls (n = 75) 51 22 2 The variations in the TNF-· genotypes were analyzed by PCR- Results RFLP, using the NcoI restriction enzyme, in 23 patients (9 with RA, 14 with CD) treated with infl iximab and 75 age- and sex-matched healthy blood donors from the South Hungarian population. TNF- · Polymorphism in Patients Treated with Infl iximab Table 1 shows genotypes for –308 TNF- · in patients receiving infl iximab and in the control group. The frequen- cy of TNF A (which is associated with high TNF- · produc- Effect of Infl iximab Therapy on in vitro TNF- · tion) carriers was 32% in healthy blood donors from among Production by WBCs the South Hungarian population (24 of 75), which was The in vitro TNF- · production by WBCs was deter- similar to previously reported data [21] . The frequency of mined at different times after infl iximab therapy. The TNF A carriers in the patient group was higher, i.e. 39.1% TNF- · concentration in the supernatants of the WBCs (9 of 23). Almost all of these 9 patients (8/9) were nonre- before infl iximab treatment was considerable (527.8 8 sponders (4 RA and 4 CD nonresponder patients). Eight 363.2 U/ml; fi g. 1 ). It is noteworthy that those patients of the 12 nonresponder CD or RA patients carried the who carried the TNF A allele exhibited the highest in vi- TNF A allele (66.6%). Statistical analysis was not applied tro TNF production (data not shown). because of the relatively small number of patients receiv- After the 3rd infusion of infl iximab, the in vitro ing infl iximab therapy. However, the high number of TNF TNF- · production was only 33.0 8 12.2 U/ml, and after A carriers in the nonresponding group is very striking. the 4th infusion, it was below the detection limit of the Pathobiology 2004;71:274–280 276 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi Mononuclear cells Mononuclear cells without SA stimulation stimulated with SA for 4 h 40 40 Fig. 2. Flow cytometric analysis of the in- tracellular TNF- · content in human mono- cytes. The basal and heat-killed SA-stimu- lated cytokine responses of the monocytes 10 M1 10 M1 were determined. Monocytes were gated for CD14-positive cells. Intracellular TNF- · expression was detected via the staining of 0 0 0 1 2 3 4 0 1 2 3 4 permeabilized cells with FITC-conjugated 10 10 10 10 10 10 10 10 10 10 MAbs specifi c for human TNF- · , and mea- FL1-Height FL1-Height sured by fl ow cytometric analysis. M1 = Fluorescence intensity in monocytes. assay. When we washed out the infl iximab from the WBCs (two washings at 1,500 rpm for 5 min with PBS, followed by resuspension in RPMI-1640 supplemented with 10% FCS), a marked TNF- · concentration of 319.4 8 141.3 U/ml was measured even after the 4th infusion. It is very likely that the anti-TNF antibody in the superna- tants of the WBCs could neutralize the cytotoxic effect of TNF- · . The reduced TNF could simply be due to TNF- · trapping by infl iximab, which means it is then not recog- nized either by bioassay or by ELISA. Effect of Infl iximab on Intracellular TNF- · Content of Monocytes Following SA stimulation, the in vitro TNF- · produc- Fig. 3. Intracellular TNF- · production in monocytes of healthy tion by monocytes from the healthy controls was in- blood donors following in vitro infl iximab treatment (A–C) and in creased signifi cantly, as revealed by an intracellular TNF- monocytes of patients receiving infl iximab therapy (D). The intra- · staining method. Figure 2 depicts representative ex- cellular TNF- · content of the cells (10 in each sample), expressed periments on the detection of the intracellular TNF- · as mean fl uorescence intensity, was measured by fl ow cytometric content in human monocytes. analysis of permeabilized cells stained with FITC-conjugated hu- man specifi c MAbs against TNF- · , and gated for monocytes. A: Pretreatment of mononuclear cells with 100 Ì g/ml in- Mononuclear cells from healthy blood donors (n = 4) with (fi lled fl iximab for 24 or 48 h did not result in a loss of their bars) or without (open bars) heat-killed SA stimulation; B: mono- TNF-producing ability; the mean fl uorescence intensity nuclear cells from healthy blood donors (n = 4), cultivated for 48 h, data for the control and SA-stimulated cells were signifi - and thereafter incubated with (fi lled bars) or without (open bars) cantly different in each experiment, as assessed by the SA; C: mononuclear cells from healthy blood donors (n = 4), pre- treated in vitro with infl iximab for 48 h, and thereafter incubated ANOVA test ( fi g. 3 ). The same tendency was observed for with (fi lled bars) or without (open bars) SA; D: mononuclear cells the mononuclear cells of the RA patients 48 h after the from patients with RA (n = 6) were isolated 48 h after the 4th infu- 4th infusion of infl iximab: F(7,28) = 65.89, p ! 0.001 ( fi g. sion with infl iximab, and incubated for a further 4 h with (fi lled 3 ). Accordingly, we conclude that the infl iximab therapy bars) or without (open bars) SA. In each group, a signifi cant differ- did not inhibit the ability of these cells to respond to the ence was observed between the stimulated and nonstimulated sub- groups [F(7,28) = 65.89, p ! 0.001]. bacterial inducer with TNF- · production. Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 277 Polymorphism and TNF- · Production Counts Counts Untreated PBMCs from a healthy PBMCs from a healthy blood donor after a 48- hour blood donor treated in vitro with incubation period infliximab for 48 h R2 R2 3 3 2 2 Fig. 4. Effect of in vitro infl iximab treat- ment on the apoptosis of PBMCs. PBMCs 1 1 25.46% 44.23% 10 10 were labeled with annexin V-FITC and PI. Surviving cells (low signals for both annex- in V and PI) appear in the lower-left quad- 0 0 10 10 rant, early apoptotic cells (high annexin V 0 1 2 3 4 0 1 2 3 4 signal, but low PI signal) in the lower-right 10 10 10 10 10 10 10 10 10 10 FL1-Height FL1-Height quadrant, and necrotic cells (high signals for both annexin V and PI) in the upper- Annexin V staining Annexin V staining right and upper-left quadrants. tosis of PBMCs isolated from the patients receiving in- fl iximab therapy 48 h after the 4th infusion. Figure 4 shows a representative experiment for the detection of apoptosis of the PBMCs of a healthy blood donor; in vitro infl iximab treatment (100 Ì g/ml) of the PBMCs for 48 h resulted in an increase in the percentage of apoptotic cells (25.46 vs. 44.23%). The kinetic study revealed that spon- taneous apoptosis was slightly elevated following a 48- hour incubation of the PBMCs ( fi g. 5 ), but it was not sig- nifi cant according to the ANOVA test. There was a sig- nifi cant increase in apoptosis, however, as a result of a 48-hour in vitro treatment of the PBMCs with infl iximab [ fi g. 5 , bar D; F(4,17) = 83.53, p ! 0.001]. In comparison Fig. 5. Comparison of in vitro and in vivo effects of infl iximab on with this in vitro infl iximab treatment, in ex vivo samples the apoptosis of PBMCs. The percentage of apoptotic cells was de- investigated directly from the RA patients undergoing termined by annexin V-FITC labeling. A: Untreated PBMCs from infl iximab therapy, the degree of apoptosis was not con- healthy blood donors (n = 4) after a 24-hour incubation; B: PBMCs siderable ( fi g. 5 , bar E). Hence, we conclude that this in- from healthy blood donors (n = 4) treated in vitro with infl iximab fl iximab therapy itself did not result in as high a rate of (100 Ì g/ml) for 24 h; C: untreated PBMCs from healthy blood do- nors (n = 4) after a 48-hour incubation; D: PBMCs from healthy apoptosis as was observed with the in vitro treatment of blood donors (n = 4) treated in vitro with infl iximab (100 Ì g/ml) for PBMCs for 48 h. 48 h; E: PBMCs from RA patients (n = 6) 48 h after the 4th infusion with infl iximab. Data are means 8 SEM. * Signifi cantly higher number of apoptotic cells in D [F(4,17) = 83.53, p ! 0.001]. Discussion The clinical effi cacy of the biological response modifi er Effect of Infl iximab on Apoptosis of PBMCs infl iximab proved to be impressive. One of the crucial The effect of infl iximab on the apoptosis of PBMCs questions that arises is how to select the optimal rate of was investigated at different time points following the in TNF inhibition for each patient [22] . The degree of TNF vitro infl iximab pretreatment of PBMCs from the healthy inhibition required to control the disease in an optimal blood donors. The results were compared with the apop- manner may vary from patient to patient. It may depend Pathobiology 2004;71:274–280 278 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi PI staining FL2-Height PI staining FL2-Height on the patient’s weight or the drug metabolism, but also local infl ammation and appears to be vital in keeping in- on each patient’s innate TNF production. We should learn fections localized. Infections and cases of sepsis have been to titrate our therapy so as to achieve a level of TNF inhi- reported in patients receiving infl iximab [12, 13] . Tuber- bition that counters the disease, but without crippling im- culosis seems to be particularly common [11] . An effective portant immune functions. Determination of the optimal host response depends partly on the ability to produce an degree of TNF inhibition that is safe and effective in each appropriate Th1 cytokine profi le, including TNF- · , which patient is crucial. We hypothesized that the therapeutic is involved in protection against both mycobacterial infec- response of patients treated with infl iximab might be re- tion and pathogenesis of tuberculosis [26] . We have dem- lated to the genetic propensity of the host to produce high onstrated that the neutralizing effect of anti-TNF therapy levels of TNF- · . Knowledge of the TNF- · gene polymor- does not result in irreversible inhibition of the TNF- · - phism in all patients treated with infl iximab may result in producing ability of mononuclear cells. a better understanding of how to optimize TNF inhibition Monocytopenia is commonly observed after treatment in individual patients. It is possible that a higher dose of with infl iximab. Treatment with infl iximab at therapeutic infl iximab is necessary for optimal TNF inhibition in concentrations resulted in apoptosis of monocytes in pa- TNF A carriers, as this allele is associated with a higher tients with chronic active CD who were receiving a dose TNF- · -producing ability [7, 8] . An increased frequency of of 6 5 mg/kg [27] . We investigated the role of infl iximab- the TNF A allele has been found among nonresponder RA induced apoptosis of PBMCs in RA patients receiving patients receiving infl iximab [23] . In contrast, there was 3 mg/kg infl iximab. In our in vitro experiments, a 48-hour no relevant association between the –308 TNF gene poly- treatment of PBMCs with infl iximab resulted in a signifi - morphism and the clinical response to infl iximab in the cant increase in the level of apoptosis ( fi g. 4 ). However, study of Louis et al. [24] in CD patients. In another study, signifi cant apoptosis was not observed in a direct analysis a rather small cohort of patients did display an association of the PBMCs of RA patients treated with infl iximab. between polymorphisms in the lymphotoxin A gene and In our experiments, we investigated the effects of in- the response to infl iximab [25] . These results, and also fl iximab on the TNF- · production and apoptosis of those of our own investigation, should be confi rmed in PBMCs in vitro and in vivo, together with the possible larger populations of patients. Our pilot study draws at- clinical relevance of the –308 TNF- · polymorphism in tention to the connection between the therapeutic re- infl iximab therapy of RA and CD patients. sponse to infl iximab and the –308 TNF- · gene polymor- Treatment of TNF inhibitors may be accompanied by phism in RA and CD patients. There was a strong ten- adverse events, and the precise mechanism of the infl ix- dency for a much higher frequency of carriers of the TNF imab therapy remains unclear. In our study, the poten- A allele among nonresponders ( table 1 ). tially TNF- · -producing PBMCs retained their host de- The low number of patients in our study group is due fense function and their number was not decreased sta- to the high costs of infl iximab therapy. A much larger set tistically by apoptosis. Monocytopenia/leukopenia was of patients on infl iximab therapy will be necessary in the not observed in either our RA patients or our CD pa- future for a reliable statistical analysis in order to confi rm tients. These investigations may promote a better under- our preliminary results. standing of the effects of infl iximab. It is important to Another important feature of our study is the demon- stress that a rigorous follow-up of patients receiving in- stration that the cytokine-producing ability of PBMCs is fl iximab is essential, however, in order for possible toxic not irreversibly impaired by infl iximab therapy, as re- effects and infections to be recognized in time, so that vealed by ex vivo experiments in which washed WBCs alternative medication can be applied [22] . were used ( fi g. 1 ). Stimulation of these white blood cells In view of the high costs of infl iximab therapy, it ap- with killed SA led to TNF- · production, which could be pears necessary to determine the TNF genotype before detected in a bioassay when the anti-TNF antibodies were such therapy is initiated, so as to achieve a better progno- washed out in the supernatant of the cell cultures. The sis. retained TNF- · -producing ability was further proved in experiments in which intracellular TNF- · was detected Acknowledgements in monocytes. TNF- · plays an important part in the pathomechanism We thank Mrs. Györgyi Müller for expert technical assistance. of both RA and CD [1, 2] . Additionally, it has a central This work was supported by Hungarian Research Grants OTKA role in the immune system, is an important mediator of T 042455 and ETT 124/2003. Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 279 Polymorphism and TNF- · Production References 1 Feldmann M, Brennan MF, Maini RN: Role 9 Louis E, Franchimont D, Piron A, Gevaert Y, 19 Espevik T, Nissen-Meyer J: A highly sensitive of cytokines in rheumatoid arthritis. Annu Rev Schaaf-Lafontaine N, Roland S, Mahieu P, cell line, WEHI 164 clone 13, for measuring Immunol 1996; 14: 397–440. Malaise M, De Groote D, Louis R, Belaiche J: cytotoxic factor/tumor necrosis factor from hu- 2 Hampe J, Shaw SH, Saiz R, Leysens N, Lan- Tumour necrosis factor (TNF) gene polymor- man monocytes. Scand J Immunol 1986; 24: termann A, Mascheretti S, Lynch NJ, MacPher- phism infl uences TNF- · production in lipo- 739–743. son AJS, Bridger S, van Deventer S, Stokkers polysaccharide (LPS)-stimulated whole blood 20 Megyeri K, Mándi Y, Degré M, Rosztóczy I: P, Morin P, Mirza MM, Forbes A, Lennard- cell cultures in healthy humans. Clin Exp Im- Induction of cytokine production by different Jones JE, Mathew CG, Curran ME, Schreiber munol 1998; 113: 401–406. Staphylococcal strains. Cytokine 2002; 19: S: Linkage of infl ammatory bowel disease to 10 Louis E, Peeters M, Franchimont D, Seidel L, 206–212. human chromosome 6p. Am J Hum Genet Fontaine F, Demolin G, Croes F, Dupont P, 21 Vatay A, Bene L, Kovacs A, Prohaszka Z, Sza- 1999; 65: 1647–1655. Davin L, Omri S, Rutgeerts P, Belaiche J: Tu- lai C, Romics L, Fekete B, Karadi I, Füst G: 3 D’Haens G, van Deventer S, Van Hogezand R, mour necrosis factor (TNF) gene polymor- Relationship between the tumor necrosis fac- Chalmers D, Kothe C, Baert F, Braakman T, phism in Crohn’s disease (CD): Infl uence on tor alpha polymorphism and the serum C-reac- Schaible T, Geboes K, Rutgeerts P: Endoscop- disease behaviour? Clin Exp Immunol 2000; tive protein levels in infl ammatory bowel dis- ic and histological healing with infl iximab anti- 119: 64–68. ease. Immunogenetics 2003; 55: 247–252. tumor necrosis factor antibodies in Crohn’s 11 Gardam MA, Keystone EC, Menzies R, Man- 22 O’Dell JR: TNF- · inhibition: The need for a disease: A European multicenter trial. Gastro- ners S, Skamene E, Long R, Vinh DC: Anti- tumor necrosis factor thermostat. Mayo Clin enterology 1999; 116: 1029–1134. tumour necrosis factor agents and tuberculosis Proc 2001; 76: 573–575. 4 Lipsky PE, van der Heijde DM, St Clair EW, risk: Mechanism of action and clinical man- 23 Mugnier B, Balandraud N, Darque A, Roudier Furst DE, Breedveld FC, Kalden JR, Smolen agement. Lancet Infect Dis 2003; 3: 148–155. C, Roudier J, Reviron D: Polymorphism at po- JS, Weisman M, Emery P, Feldmann M, Har- 12 Andus T, Stange EF, Hoffl er D, Keller-Stani- sition –308 of the tumor necrosis factor alpha riman GR, Maini RN; Anti-Tumor Necrosis slawski B: Suspected cases of severe side effects gene infl uences outcome of infl iximab therapy Factor Trial in Rheumatoid Arthritis with after infl iximab (Remicade) in Germany (in in rheumatoid arthritis. Arthritis Rheum 2003; Concomitant Therapy Study Group: Infl ix- German). Med Klin (Munich) 2003; 98: 429– 48: 1849–1852. imab and methotrexate in the treatment of 436. 24 Louis E, Vermeire S, Rutgeerts P, De Vos M, rheumatoid arthritis. N Engl J Med 2000; 343: 13 Colombel JF, Loftus EV Jr, Tremaine WJ, Van Gossum A, Pescatore P, Fiasse R, Pelck- 1594–1602. Egan LJ, Harmsen WS, Schleck CD, Zinsmeis- mans P, Reynaert H, D’Haens G, Malaise M, 5 Keystone EC: Tumor necrosis factor-alpha ter AR, Sandborn WJ: The safety profi le of in- Belaiche J: A positive response to infl iximab in blockade in the treatment of rheumatoid ar- fl iximab in patients with Crohn’s disease: The Crohn disease: Association with a higher sys- thritis. Rheum Dis Clin North Am 2001; 27: Mayo clinic experience in 500 patients. Gas- temic infl ammation before treatment but not 427–443. troenterology 2004; 126: 19–31. with –308 TNF gene polymorphism. Scand J 6 Wilson AG, de Vreis N, Poicot F, di Giovine 14 van Deventer SJH: Review article: Targeting Gastroenterol 2002; 37: 818–824. FS, van der Putte LBA, Duff GW: An allelic TNF · as a key cytokine in the infl ammatory 25 Taylor KD, Plevy SE, Yang H, Landers CJ, polymorphism within the human tumor necro- processes of Crohn’s disease – the mechanisms Barry MJ, Rotter JI, Targan SR: ANCA pat- sis factor alpha promoter region is strongly as- of action of infl iximab. Aliment Pharmacol tern and LTA haplotype relationship to clinical sociated with HLA A1, B8, and DR3 alleles. J Ther 1999; 13(suppl 4):3–8, discussion 38. responses to anti-TNF antibody treatment in Exp Med 1993; 177: 557–560. 15 Papadakis KA, Targan SR: Tumor necrosis Crohn’s disease. Gastroenterology 2001; 120: 7 Bouma G, Crusius JBA, Oudrek PM, Kolkman factor: Biology and therapeutic inhibitors. 1347–1355. JJ, von Blomberg BME, Kostense PJ, Giphart Gastroenterology 2000; 119: 1148–1157. 26 Kaplan G, Freedman VH: The role of cyto- MJ, Schreuder GM, Meuwissen SG, Pena AS: 16 Arnett FC, Edworthy SM, Bloch DA, McShane kines in the immune response to tuberculosis. Secretion of tumour necrosis factor alpha and DJ, Fries JF, Cooper NS, Healey LA, Kaplan Res Immunol 1996; 147: 565–572. lymphotoxin alpha in relation to polymor- SR, Liang MH, Luthra HS: The American 27 Lügering A, Schmidt M, Lügering N, Pauels phisms in the TNF genes and HLA-DR alleles. Rheumatism Association 1987 revised criteria HG, Domschike W, Kucharzik T: Infl iximab Relevance for infl ammatory bowel disease. for the classifi cation of rheumatoid arthritis. induces apoptosis in monocytes from patients Scand J Immunol 1996; 43: 456–463. Arthritis Rheum 1988; 31: 315–324. with chronic active Crohn’s disease by using a 8 Wilson AG, Symons JA, McDowell TL, McDe- 17 Podolsky DK: Infl ammatory bowel disease. N caspase-dependent pathway. Gastroenterology vitt HO, Duff GW: Effects of a polymorphism Engl J Med 1991; 325: 928–937. 2001; 121: 1145–1157. in the human tumor necrosis factor · promot- 18 Wilson AG, Giovine FS, Blakemore AIF, Duff er on transcriptional activation. Proc Natl GW: Single base polymorphism in the human Acad Sci USA 1997; 94: 3195–3199. tumour necrosis factor alpha gene detectable by Nco I restriction of PCR product. Hum Mol Genet 1992; 1: 353. Pathobiology 2004;71:274–280 280 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Pathobiology Karger

Investigation of the Prognostic Value of TNF-α Gene Polymorphism among Patients Treated with Infliximab, and the Effects of Infliximab Therapy on TNF-α Production and Apoptosis

Download PDF
7 pages

Loading...

Page 2

Loading...

Page 3

Loading...

Page 4

Loading...

Page 5

Loading...

Page 6

Loading...

Page 7

 
/lp/karger/investigation-of-the-prognostic-value-of-tnf-gene-polymorphism-among-tN4AwHh70D

References (36)

Publisher
Karger
Copyright
© 2004 S. Karger AG, Basel
ISSN
1015-2008
eISSN
1423-0291
DOI
10.1159/000080062
Publisher site
See Article on Publisher Site

Abstract

Original Paper Received: February 27, 2004 Pathobiology 2004;71:274–280 Accepted: May 28, 2004 DOI: 10.1159/000080062 Investigation of the Prognostic Value of TNF- · Gene Polymorphism among Patients Treated with Infl iximab, and the Effects of Infl iximab Therapy on TNF- · Production and Apoptosis a a d b b Attila Balog Gergely Klausz János Gál Tamás Molnár Ferenc Nagy c a a Imre Ocsovszky Zsófi a Gyulai Yvette Mándi a b Department of Medical Microbiology and Immunobiology, First Department of Internal Medicine, c d Department of Biochemistry, Faculty of Medicine, University of Szeged, Szeged , Department of Rheumatology, County Hospital, Kecskemét , Hungary Key Words more suitable candidates for infl iximab treatment. Al- Tumor necrosis factor-· · Gene polymorphism · though in vitro infl iximab treatment for 48 h resulted in Infl iximab · Apoptosis · Intracellular cytokine · signifi cant (44.2 8 1.17%) apoptosis in PBMCs, in ex vivo Rheumatoid arthritis · Crohn’s disease samples from RA patients who received infl iximab, apoptosis was only 13.3 8 1.6%. Furthermore, infl iximab did not result in irreversible inhibition of the TNF- · -pro- Abstract ducing ability or in the signifi cant apoptosis of PBMCs. Objectives: Infl iximab, a chimeric anti-tumor necrosis Copyright © 2004 S. Karger AG, Basel factor (TNF) antibody, is highly effective for the treat- ment of Crohn’s disease (CD) and rheumatoid arthritis (RA). Our experiments focused on RA and CD patients Introduction receiving infl iximab. Since cytokine production is large- ly determined by genetic factors, the promoter polymor- The central role of tumor necrosis factor- · (TNF- · ) in phisms of TNF- · were examined among these patients. the pathogenesis of rheumatoid arthritis (RA) and Crohn’s Additionally, the changes caused by infl iximab in the disease (CD) has been well documented [1, 2] , and TNF- · -producing ability and apoptosis of peripheral TNF- · is therefore an excellent therapeutic target. One blood mononuclear cells (PBMCs) were investigated. of the most effective biological agents, infl iximab, a chi- Methods: The TNF- · genotypes were analyzed by PCR- meric anti-TNF antibody, has been shown to be highly RFLP. The in vitro TNF- · production of the PBMCs was effective for the treatment of both patients with CD [3] detected by fl ow cytometric analysis. The TNF- · concen- and patients with RA [4, 5] . The experiments reported tration in the supernatant was measured by bioassay. here focus on patients with active RA in whom tradition- Apoptosis was detected by annexin V-fl uorescein iso- al treatment with disease-modifying antirheumatic drugs thiocyanate labeling. Results and Conclusions: 8 of the had not been successful, and on patients with steroid-re- 12 nonresponder patients carried the TNF A allele associ- fractory CD. All of these patients underwent infl iximab ated with high TNF- · production. We suggest that the de- therapy. Despite the generally dramatic effi cacy of infl ix- termination of TNF polymorphism may help identify imab, some of our patients did not display a clinical re- © 2004 S. Karger AG, Basel Prof. Dr. Yvette Mándi 1015–2008/04/0715–0274$21.00/0 Department of Medical Microbiology and Immunobiology Fax +41 61 306 12 34 Faculty of Medicine, University of Szeged E-Mail karger@karger.ch Accessible online at: PO Box 427, Dóm tér 10, HU–6720 Szeged (Hungary) www.karger.com www.karger.com/pat Tel. +36 62 545115 , Fax +36 62 545113, E-Mail yvette@comser.szote.u-szeged.hu of Helsinki. The ethical committee of the participating university sponse. We hypothesized that a genetic predisposition approved the study. Informed written consent was obtained in ad- might be involved in the level of responsiveness. Since vance from each patient. cytokine production is largely determined by the poly- morphism of regions regulating cytokine expression, the DNA Extraction most frequently investigated promoter polymorphism at Genomic DNA from whole blood containing EDTA was ex- tracted using standard techniques (High Pure PCR Template Prep- the –308 position of TNF- · [6] was examined among RA aration Kit, Roche, catalog No. 1796828). and CD patients receiving infl iximab therapy. The less frequent TNF A allele of the –308 TNF- · gene polymor- –308 TNF- · Genotyping phism has been associated with increased TNF transcrip- The G to A transition at position –308 in the promoter region tion and production [7–9] and with more severe, either is associated with elevated expression of TNF- · . Using the method described by Wilson et al. [18] , a PCR was followed by digestion steroid-dependent or fi stulizing CD [10] . We therefore set with the endonuclease Nco I. Upon analysis by acrylamide gel elec- out to investigate the possible association between the – trophoresis, the TNF G allele gives two fragments of 87 and 20 bp, 308 TNF- · genotype and the response to infl iximab. while the TNF A allele gives a single, 107-bp fragment. A total of An effective host response depends in part on the cy- 9 patients with RA and 14 with CD as well as 75 healthy blood do- tokine-producing ability, including TNF- · [11] . Since nors were analyzed. there have been reports of adverse events with TNF in- Detection of TNF- · in Supernatant of Whole Blood Cultures hibitors, including infections and cases of sepsis [11–13] , TNF- · concentrations in the supernatants of whole blood cul- and since the mechanism of action of infl iximab is only ture (WBC) samples of the 9 patients with RA were determined by partly understood [5, 14, 15] , the changes caused by in- bioassay, applying the WEHI 164 mouse fi broblast cell line [19] . fl iximab in the TNF- · -producing ability of peripheral Human recombinant TNF- · (Amersham) was used as a standard in the same assay. The specifi city of the effect of TNF- · in the super- blood mononuclear cells (PBMCs) were investigated, to- natants was confi rmed via the neutralizing effect of monoclonal gether with apoptosis in PBMCs. anti-TNF- · antibody, clone TA-31 (Sigma product No. T1549). The amounts of TNF- · are expressed in U/ml. In control experiments, TNF- · concentrations in the supernatants of WBC samples were Patients and Methods also determined using a TNF- · ELISA kit (Biosource), according to the instructions of the manufacturer. For comparison, the amounts of TNF- · determined in the two assays are expressed in U/ml, where Patients Nine patients with disease-modifying antirheumatic drug-re- 1 unit is equivalent to 34 pg of recombinant human TNF- · . fractory RA were diagnosed according to the revised 1987 Ameri- can College of Rheumatology criteria [16] . The infusion regimen Isolation of Human Mononuclear Cells was as follows: 3 mg/kg infl iximab (Schering-Plough) administered Peripheral white blood cells were prepared by density gradient purifi cation (Ficoll-Paque, Sigma) from heparinized venous blood at 0, 2 and 6 weeks and then every 8 weeks for 1 year. The 14 patients with chronic active CD unresponsive to stan- samples from healthy volunteers and from RA patients treated with dard therapy were diagnosed in accordance with clinical guidelines infl iximab, and 95% cell viability was then confi rmed by trypan blue staining. Subsequently, 5 ! 10 /ml PBMCs suspended in me- and on the basis of conventional radiological, endoscopic and his- topathologic examinations [17] . Fistulas were a common complica- dium were used in all the experiments. Isolated cells were incu- tion. The infusion regimen was as follows: 5 mg/kg infl iximab ad- bated in RPMI-1640 medium (Gibco) supplemented with 10% fe- ministered at 0, 2 and 6 weeks. tal bovine serum, antibiotics and glutamine. All patients underwent regular control examinations. All pa- Intracellular Staining of TNF- · tients and controls were adults of Hungarian ethnic origin and re- sided in Hungary. PBMCs were incubated in RPMI-1640 medium (Gibco) supple- Following the infl iximab therapy, the RA patients were divided mented with 10% fetal bovine serum, antibiotics and glutamine for into responding and nonresponding groups according to the follow- 24 or 48 h in the presence or absence of infl iximab (100 Ì g/ml). Af- ing parameters: the DAS-28 (disease activity score), a health assess- ter centrifugation, the cells were washed with phosphate-buffered saline (PBS). They were then stimulated with heat-killed (90 ° C for ment questionnaire, a pain visual analog score, the swollen and tender joint index, morning stiffness, the C-reactive protein level, 10 min adjusted to 10 /ml) [20] Staphylococcus aureus (SA) as TNF the erythrocyte sedimentation rate, the physician’s global assess- inducer in the presence of brefeldin A (10 Ì g/ml) for 4 h at 37 ° C in ment, the patient’s global assessment, and the American College of a humidifi ed 5% CO incubator. After centrifugation, the cells were Rheumatology functional capacity score. After treatment with in- washed with PBS and permeabilized with Cytofi x-Cytoperm solu- tion (Becton-Dickinson) for 20 min at room temperature in the fl iximab, the CD patients were categorized as responders if steroid therapy could be stopped, if there were no draining fi stulas or most dark. Thereafter, they were washed twice with PermWash (Becton- of the fi stulas had closed, and if the number of liquid or very soft Dickinson) solution and once with added fl uorescein isothiocyanate (FITC)-labeled human TNF- · specifi c monoclonal antibodies stools had decreased markedly. This study was performed in accordance with the ethical stan- (MAbs) for 30 min at room temperature in the dark. They were next washed twice with PermWash and once with added PBS. Flow cy- dards laid down in the most recent version of the 1964 Declaration Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 275 Polymorphism and TNF- · Production tometric analysis was subsequently performed with a FACStar plus fl uorescence-activated cell sorter (Becton-Dickinson) at 488-nm ex- citation to estimate intracellular TNF- · in the peripheral mono- nuclear cells. Cells were gated for CD14-positive cells as follows. CD14 expression was detected with phycoerythrin-conjugated MAbs. At the same time, a dot plot display was used to identify monocytes, which correlated with the gate applied for CD14-posi- tive cells. In subsequent experiments, this gate was applied. Detection of Apoptosis with Annexin V Cell apoptosis was assessed by annexin V-FITC staining with a fl ow cytometric apoptosis detection kit (PharMingen, Becton-Dick- inson, catalog No. 556420). The PBMCs were incubated for 24–48 h in the presence or ab- sence of infl iximab (100 Ì g/ml), then washed twice in PBS and re- suspended in binding buffer (0.1 M Hepes-NaOH, 1.4 M NaCl, 25 m M CaCl ) at a concentration of 1 ! 10 cells/ml. 5 Ì l of an- Fig. 1. TNF- · concentration in supernatants of whole blood sam- nexin V-FITC and 10 Ì l of propidium iodide (PI) were added to a ples from 9 RA patients treated with infl iximab following heat- 100- Ì l aliquot of each cell solution. This was followed by gentle killed SA stimulation of the cells. A: Before infl iximab therapy; B: mixing and incubation for 15 min at room temperature in the dark. 48 h after the 3rd infusion; C: 48 h after the 4th infusion; D: 48 h Next, 400 Ì l of binding buffer was added to each tube. Flow cyto- after the 4th infusion – infl iximab was washed out from the whole metric analysis was performed within 30 min. blood cells. Data are means 8 SEM. Those cells that were negative for both dyes were considered to be live cells; necrotic cells were positive for both fl uorochromes or only for PI, and apoptotic cells were positive only for annexin V- FITC and negative for PI. Table 1. TNF- · genotypes in patients receiving infl iximab therapy and in control subjects Statistical Analysis The levels of intracellular TNF- · production with the SA in- Group TNF G/ TNF G/ TNF A/ ducer and the experimental results for apoptosis were compared by TNF G TNF A TNF A means of one-way ANOVA. The Student-Newman-Keuls test was used for post hoc pairwise multiple comparisons. In all tests, an · Patients treated with infl iximab (n = 23) level of p ! 0.05 was taken as an indication of statistical signifi cance. Responders 10 110 All statistical calculations were performed with the GraphPad Nonresponders 14 180 Prism4 statistical program. Healthy controls (n = 75) 51 22 2 The variations in the TNF-· genotypes were analyzed by PCR- Results RFLP, using the NcoI restriction enzyme, in 23 patients (9 with RA, 14 with CD) treated with infl iximab and 75 age- and sex-matched healthy blood donors from the South Hungarian population. TNF- · Polymorphism in Patients Treated with Infl iximab Table 1 shows genotypes for –308 TNF- · in patients receiving infl iximab and in the control group. The frequen- cy of TNF A (which is associated with high TNF- · produc- Effect of Infl iximab Therapy on in vitro TNF- · tion) carriers was 32% in healthy blood donors from among Production by WBCs the South Hungarian population (24 of 75), which was The in vitro TNF- · production by WBCs was deter- similar to previously reported data [21] . The frequency of mined at different times after infl iximab therapy. The TNF A carriers in the patient group was higher, i.e. 39.1% TNF- · concentration in the supernatants of the WBCs (9 of 23). Almost all of these 9 patients (8/9) were nonre- before infl iximab treatment was considerable (527.8 8 sponders (4 RA and 4 CD nonresponder patients). Eight 363.2 U/ml; fi g. 1 ). It is noteworthy that those patients of the 12 nonresponder CD or RA patients carried the who carried the TNF A allele exhibited the highest in vi- TNF A allele (66.6%). Statistical analysis was not applied tro TNF production (data not shown). because of the relatively small number of patients receiv- After the 3rd infusion of infl iximab, the in vitro ing infl iximab therapy. However, the high number of TNF TNF- · production was only 33.0 8 12.2 U/ml, and after A carriers in the nonresponding group is very striking. the 4th infusion, it was below the detection limit of the Pathobiology 2004;71:274–280 276 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi Mononuclear cells Mononuclear cells without SA stimulation stimulated with SA for 4 h 40 40 Fig. 2. Flow cytometric analysis of the in- tracellular TNF- · content in human mono- cytes. The basal and heat-killed SA-stimu- lated cytokine responses of the monocytes 10 M1 10 M1 were determined. Monocytes were gated for CD14-positive cells. Intracellular TNF- · expression was detected via the staining of 0 0 0 1 2 3 4 0 1 2 3 4 permeabilized cells with FITC-conjugated 10 10 10 10 10 10 10 10 10 10 MAbs specifi c for human TNF- · , and mea- FL1-Height FL1-Height sured by fl ow cytometric analysis. M1 = Fluorescence intensity in monocytes. assay. When we washed out the infl iximab from the WBCs (two washings at 1,500 rpm for 5 min with PBS, followed by resuspension in RPMI-1640 supplemented with 10% FCS), a marked TNF- · concentration of 319.4 8 141.3 U/ml was measured even after the 4th infusion. It is very likely that the anti-TNF antibody in the superna- tants of the WBCs could neutralize the cytotoxic effect of TNF- · . The reduced TNF could simply be due to TNF- · trapping by infl iximab, which means it is then not recog- nized either by bioassay or by ELISA. Effect of Infl iximab on Intracellular TNF- · Content of Monocytes Following SA stimulation, the in vitro TNF- · produc- Fig. 3. Intracellular TNF- · production in monocytes of healthy tion by monocytes from the healthy controls was in- blood donors following in vitro infl iximab treatment (A–C) and in creased signifi cantly, as revealed by an intracellular TNF- monocytes of patients receiving infl iximab therapy (D). The intra- · staining method. Figure 2 depicts representative ex- cellular TNF- · content of the cells (10 in each sample), expressed periments on the detection of the intracellular TNF- · as mean fl uorescence intensity, was measured by fl ow cytometric content in human monocytes. analysis of permeabilized cells stained with FITC-conjugated hu- man specifi c MAbs against TNF- · , and gated for monocytes. A: Pretreatment of mononuclear cells with 100 Ì g/ml in- Mononuclear cells from healthy blood donors (n = 4) with (fi lled fl iximab for 24 or 48 h did not result in a loss of their bars) or without (open bars) heat-killed SA stimulation; B: mono- TNF-producing ability; the mean fl uorescence intensity nuclear cells from healthy blood donors (n = 4), cultivated for 48 h, data for the control and SA-stimulated cells were signifi - and thereafter incubated with (fi lled bars) or without (open bars) cantly different in each experiment, as assessed by the SA; C: mononuclear cells from healthy blood donors (n = 4), pre- treated in vitro with infl iximab for 48 h, and thereafter incubated ANOVA test ( fi g. 3 ). The same tendency was observed for with (fi lled bars) or without (open bars) SA; D: mononuclear cells the mononuclear cells of the RA patients 48 h after the from patients with RA (n = 6) were isolated 48 h after the 4th infu- 4th infusion of infl iximab: F(7,28) = 65.89, p ! 0.001 ( fi g. sion with infl iximab, and incubated for a further 4 h with (fi lled 3 ). Accordingly, we conclude that the infl iximab therapy bars) or without (open bars) SA. In each group, a signifi cant differ- did not inhibit the ability of these cells to respond to the ence was observed between the stimulated and nonstimulated sub- groups [F(7,28) = 65.89, p ! 0.001]. bacterial inducer with TNF- · production. Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 277 Polymorphism and TNF- · Production Counts Counts Untreated PBMCs from a healthy PBMCs from a healthy blood donor after a 48- hour blood donor treated in vitro with incubation period infliximab for 48 h R2 R2 3 3 2 2 Fig. 4. Effect of in vitro infl iximab treat- ment on the apoptosis of PBMCs. PBMCs 1 1 25.46% 44.23% 10 10 were labeled with annexin V-FITC and PI. Surviving cells (low signals for both annex- in V and PI) appear in the lower-left quad- 0 0 10 10 rant, early apoptotic cells (high annexin V 0 1 2 3 4 0 1 2 3 4 signal, but low PI signal) in the lower-right 10 10 10 10 10 10 10 10 10 10 FL1-Height FL1-Height quadrant, and necrotic cells (high signals for both annexin V and PI) in the upper- Annexin V staining Annexin V staining right and upper-left quadrants. tosis of PBMCs isolated from the patients receiving in- fl iximab therapy 48 h after the 4th infusion. Figure 4 shows a representative experiment for the detection of apoptosis of the PBMCs of a healthy blood donor; in vitro infl iximab treatment (100 Ì g/ml) of the PBMCs for 48 h resulted in an increase in the percentage of apoptotic cells (25.46 vs. 44.23%). The kinetic study revealed that spon- taneous apoptosis was slightly elevated following a 48- hour incubation of the PBMCs ( fi g. 5 ), but it was not sig- nifi cant according to the ANOVA test. There was a sig- nifi cant increase in apoptosis, however, as a result of a 48-hour in vitro treatment of the PBMCs with infl iximab [ fi g. 5 , bar D; F(4,17) = 83.53, p ! 0.001]. In comparison Fig. 5. Comparison of in vitro and in vivo effects of infl iximab on with this in vitro infl iximab treatment, in ex vivo samples the apoptosis of PBMCs. The percentage of apoptotic cells was de- investigated directly from the RA patients undergoing termined by annexin V-FITC labeling. A: Untreated PBMCs from infl iximab therapy, the degree of apoptosis was not con- healthy blood donors (n = 4) after a 24-hour incubation; B: PBMCs siderable ( fi g. 5 , bar E). Hence, we conclude that this in- from healthy blood donors (n = 4) treated in vitro with infl iximab fl iximab therapy itself did not result in as high a rate of (100 Ì g/ml) for 24 h; C: untreated PBMCs from healthy blood do- nors (n = 4) after a 48-hour incubation; D: PBMCs from healthy apoptosis as was observed with the in vitro treatment of blood donors (n = 4) treated in vitro with infl iximab (100 Ì g/ml) for PBMCs for 48 h. 48 h; E: PBMCs from RA patients (n = 6) 48 h after the 4th infusion with infl iximab. Data are means 8 SEM. * Signifi cantly higher number of apoptotic cells in D [F(4,17) = 83.53, p ! 0.001]. Discussion The clinical effi cacy of the biological response modifi er Effect of Infl iximab on Apoptosis of PBMCs infl iximab proved to be impressive. One of the crucial The effect of infl iximab on the apoptosis of PBMCs questions that arises is how to select the optimal rate of was investigated at different time points following the in TNF inhibition for each patient [22] . The degree of TNF vitro infl iximab pretreatment of PBMCs from the healthy inhibition required to control the disease in an optimal blood donors. The results were compared with the apop- manner may vary from patient to patient. It may depend Pathobiology 2004;71:274–280 278 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi PI staining FL2-Height PI staining FL2-Height on the patient’s weight or the drug metabolism, but also local infl ammation and appears to be vital in keeping in- on each patient’s innate TNF production. We should learn fections localized. Infections and cases of sepsis have been to titrate our therapy so as to achieve a level of TNF inhi- reported in patients receiving infl iximab [12, 13] . Tuber- bition that counters the disease, but without crippling im- culosis seems to be particularly common [11] . An effective portant immune functions. Determination of the optimal host response depends partly on the ability to produce an degree of TNF inhibition that is safe and effective in each appropriate Th1 cytokine profi le, including TNF- · , which patient is crucial. We hypothesized that the therapeutic is involved in protection against both mycobacterial infec- response of patients treated with infl iximab might be re- tion and pathogenesis of tuberculosis [26] . We have dem- lated to the genetic propensity of the host to produce high onstrated that the neutralizing effect of anti-TNF therapy levels of TNF- · . Knowledge of the TNF- · gene polymor- does not result in irreversible inhibition of the TNF- · - phism in all patients treated with infl iximab may result in producing ability of mononuclear cells. a better understanding of how to optimize TNF inhibition Monocytopenia is commonly observed after treatment in individual patients. It is possible that a higher dose of with infl iximab. Treatment with infl iximab at therapeutic infl iximab is necessary for optimal TNF inhibition in concentrations resulted in apoptosis of monocytes in pa- TNF A carriers, as this allele is associated with a higher tients with chronic active CD who were receiving a dose TNF- · -producing ability [7, 8] . An increased frequency of of 6 5 mg/kg [27] . We investigated the role of infl iximab- the TNF A allele has been found among nonresponder RA induced apoptosis of PBMCs in RA patients receiving patients receiving infl iximab [23] . In contrast, there was 3 mg/kg infl iximab. In our in vitro experiments, a 48-hour no relevant association between the –308 TNF gene poly- treatment of PBMCs with infl iximab resulted in a signifi - morphism and the clinical response to infl iximab in the cant increase in the level of apoptosis ( fi g. 4 ). However, study of Louis et al. [24] in CD patients. In another study, signifi cant apoptosis was not observed in a direct analysis a rather small cohort of patients did display an association of the PBMCs of RA patients treated with infl iximab. between polymorphisms in the lymphotoxin A gene and In our experiments, we investigated the effects of in- the response to infl iximab [25] . These results, and also fl iximab on the TNF- · production and apoptosis of those of our own investigation, should be confi rmed in PBMCs in vitro and in vivo, together with the possible larger populations of patients. Our pilot study draws at- clinical relevance of the –308 TNF- · polymorphism in tention to the connection between the therapeutic re- infl iximab therapy of RA and CD patients. sponse to infl iximab and the –308 TNF- · gene polymor- Treatment of TNF inhibitors may be accompanied by phism in RA and CD patients. There was a strong ten- adverse events, and the precise mechanism of the infl ix- dency for a much higher frequency of carriers of the TNF imab therapy remains unclear. In our study, the poten- A allele among nonresponders ( table 1 ). tially TNF- · -producing PBMCs retained their host de- The low number of patients in our study group is due fense function and their number was not decreased sta- to the high costs of infl iximab therapy. A much larger set tistically by apoptosis. Monocytopenia/leukopenia was of patients on infl iximab therapy will be necessary in the not observed in either our RA patients or our CD pa- future for a reliable statistical analysis in order to confi rm tients. These investigations may promote a better under- our preliminary results. standing of the effects of infl iximab. It is important to Another important feature of our study is the demon- stress that a rigorous follow-up of patients receiving in- stration that the cytokine-producing ability of PBMCs is fl iximab is essential, however, in order for possible toxic not irreversibly impaired by infl iximab therapy, as re- effects and infections to be recognized in time, so that vealed by ex vivo experiments in which washed WBCs alternative medication can be applied [22] . were used ( fi g. 1 ). Stimulation of these white blood cells In view of the high costs of infl iximab therapy, it ap- with killed SA led to TNF- · production, which could be pears necessary to determine the TNF genotype before detected in a bioassay when the anti-TNF antibodies were such therapy is initiated, so as to achieve a better progno- washed out in the supernatant of the cell cultures. The sis. retained TNF- · -producing ability was further proved in experiments in which intracellular TNF- · was detected Acknowledgements in monocytes. TNF- · plays an important part in the pathomechanism We thank Mrs. Györgyi Müller for expert technical assistance. of both RA and CD [1, 2] . Additionally, it has a central This work was supported by Hungarian Research Grants OTKA role in the immune system, is an important mediator of T 042455 and ETT 124/2003. Pathobiology 2004;71:274–280 Infl iximab Therapy, TNF- · Gene 279 Polymorphism and TNF- · Production References 1 Feldmann M, Brennan MF, Maini RN: Role 9 Louis E, Franchimont D, Piron A, Gevaert Y, 19 Espevik T, Nissen-Meyer J: A highly sensitive of cytokines in rheumatoid arthritis. Annu Rev Schaaf-Lafontaine N, Roland S, Mahieu P, cell line, WEHI 164 clone 13, for measuring Immunol 1996; 14: 397–440. Malaise M, De Groote D, Louis R, Belaiche J: cytotoxic factor/tumor necrosis factor from hu- 2 Hampe J, Shaw SH, Saiz R, Leysens N, Lan- Tumour necrosis factor (TNF) gene polymor- man monocytes. Scand J Immunol 1986; 24: termann A, Mascheretti S, Lynch NJ, MacPher- phism infl uences TNF- · production in lipo- 739–743. son AJS, Bridger S, van Deventer S, Stokkers polysaccharide (LPS)-stimulated whole blood 20 Megyeri K, Mándi Y, Degré M, Rosztóczy I: P, Morin P, Mirza MM, Forbes A, Lennard- cell cultures in healthy humans. Clin Exp Im- Induction of cytokine production by different Jones JE, Mathew CG, Curran ME, Schreiber munol 1998; 113: 401–406. Staphylococcal strains. Cytokine 2002; 19: S: Linkage of infl ammatory bowel disease to 10 Louis E, Peeters M, Franchimont D, Seidel L, 206–212. human chromosome 6p. Am J Hum Genet Fontaine F, Demolin G, Croes F, Dupont P, 21 Vatay A, Bene L, Kovacs A, Prohaszka Z, Sza- 1999; 65: 1647–1655. Davin L, Omri S, Rutgeerts P, Belaiche J: Tu- lai C, Romics L, Fekete B, Karadi I, Füst G: 3 D’Haens G, van Deventer S, Van Hogezand R, mour necrosis factor (TNF) gene polymor- Relationship between the tumor necrosis fac- Chalmers D, Kothe C, Baert F, Braakman T, phism in Crohn’s disease (CD): Infl uence on tor alpha polymorphism and the serum C-reac- Schaible T, Geboes K, Rutgeerts P: Endoscop- disease behaviour? Clin Exp Immunol 2000; tive protein levels in infl ammatory bowel dis- ic and histological healing with infl iximab anti- 119: 64–68. ease. Immunogenetics 2003; 55: 247–252. tumor necrosis factor antibodies in Crohn’s 11 Gardam MA, Keystone EC, Menzies R, Man- 22 O’Dell JR: TNF- · inhibition: The need for a disease: A European multicenter trial. Gastro- ners S, Skamene E, Long R, Vinh DC: Anti- tumor necrosis factor thermostat. Mayo Clin enterology 1999; 116: 1029–1134. tumour necrosis factor agents and tuberculosis Proc 2001; 76: 573–575. 4 Lipsky PE, van der Heijde DM, St Clair EW, risk: Mechanism of action and clinical man- 23 Mugnier B, Balandraud N, Darque A, Roudier Furst DE, Breedveld FC, Kalden JR, Smolen agement. Lancet Infect Dis 2003; 3: 148–155. C, Roudier J, Reviron D: Polymorphism at po- JS, Weisman M, Emery P, Feldmann M, Har- 12 Andus T, Stange EF, Hoffl er D, Keller-Stani- sition –308 of the tumor necrosis factor alpha riman GR, Maini RN; Anti-Tumor Necrosis slawski B: Suspected cases of severe side effects gene infl uences outcome of infl iximab therapy Factor Trial in Rheumatoid Arthritis with after infl iximab (Remicade) in Germany (in in rheumatoid arthritis. Arthritis Rheum 2003; Concomitant Therapy Study Group: Infl ix- German). Med Klin (Munich) 2003; 98: 429– 48: 1849–1852. imab and methotrexate in the treatment of 436. 24 Louis E, Vermeire S, Rutgeerts P, De Vos M, rheumatoid arthritis. N Engl J Med 2000; 343: 13 Colombel JF, Loftus EV Jr, Tremaine WJ, Van Gossum A, Pescatore P, Fiasse R, Pelck- 1594–1602. Egan LJ, Harmsen WS, Schleck CD, Zinsmeis- mans P, Reynaert H, D’Haens G, Malaise M, 5 Keystone EC: Tumor necrosis factor-alpha ter AR, Sandborn WJ: The safety profi le of in- Belaiche J: A positive response to infl iximab in blockade in the treatment of rheumatoid ar- fl iximab in patients with Crohn’s disease: The Crohn disease: Association with a higher sys- thritis. Rheum Dis Clin North Am 2001; 27: Mayo clinic experience in 500 patients. Gas- temic infl ammation before treatment but not 427–443. troenterology 2004; 126: 19–31. with –308 TNF gene polymorphism. Scand J 6 Wilson AG, de Vreis N, Poicot F, di Giovine 14 van Deventer SJH: Review article: Targeting Gastroenterol 2002; 37: 818–824. FS, van der Putte LBA, Duff GW: An allelic TNF · as a key cytokine in the infl ammatory 25 Taylor KD, Plevy SE, Yang H, Landers CJ, polymorphism within the human tumor necro- processes of Crohn’s disease – the mechanisms Barry MJ, Rotter JI, Targan SR: ANCA pat- sis factor alpha promoter region is strongly as- of action of infl iximab. Aliment Pharmacol tern and LTA haplotype relationship to clinical sociated with HLA A1, B8, and DR3 alleles. J Ther 1999; 13(suppl 4):3–8, discussion 38. responses to anti-TNF antibody treatment in Exp Med 1993; 177: 557–560. 15 Papadakis KA, Targan SR: Tumor necrosis Crohn’s disease. Gastroenterology 2001; 120: 7 Bouma G, Crusius JBA, Oudrek PM, Kolkman factor: Biology and therapeutic inhibitors. 1347–1355. JJ, von Blomberg BME, Kostense PJ, Giphart Gastroenterology 2000; 119: 1148–1157. 26 Kaplan G, Freedman VH: The role of cyto- MJ, Schreuder GM, Meuwissen SG, Pena AS: 16 Arnett FC, Edworthy SM, Bloch DA, McShane kines in the immune response to tuberculosis. Secretion of tumour necrosis factor alpha and DJ, Fries JF, Cooper NS, Healey LA, Kaplan Res Immunol 1996; 147: 565–572. lymphotoxin alpha in relation to polymor- SR, Liang MH, Luthra HS: The American 27 Lügering A, Schmidt M, Lügering N, Pauels phisms in the TNF genes and HLA-DR alleles. Rheumatism Association 1987 revised criteria HG, Domschike W, Kucharzik T: Infl iximab Relevance for infl ammatory bowel disease. for the classifi cation of rheumatoid arthritis. induces apoptosis in monocytes from patients Scand J Immunol 1996; 43: 456–463. Arthritis Rheum 1988; 31: 315–324. with chronic active Crohn’s disease by using a 8 Wilson AG, Symons JA, McDowell TL, McDe- 17 Podolsky DK: Infl ammatory bowel disease. N caspase-dependent pathway. Gastroenterology vitt HO, Duff GW: Effects of a polymorphism Engl J Med 1991; 325: 928–937. 2001; 121: 1145–1157. in the human tumor necrosis factor · promot- 18 Wilson AG, Giovine FS, Blakemore AIF, Duff er on transcriptional activation. Proc Natl GW: Single base polymorphism in the human Acad Sci USA 1997; 94: 3195–3199. tumour necrosis factor alpha gene detectable by Nco I restriction of PCR product. Hum Mol Genet 1992; 1: 353. Pathobiology 2004;71:274–280 280 Balog/Klausz/Gál/Molnár/Nagy/ Ocsovszky/Gyulai/Mándi

Journal

Pathobiology – Karger

Published: Sep 1, 2004

Keywords: Crohn’s disease; Tumor necrosis factor-α; Gene polymorphism; Infliximab; Apoptosis; Intracellular cytokine; Rheumatoid arthritis

There are no references for this article.