Abstract
Evaluating immunomodulating effects is currently a fundamental parameter when designing pharmaceutical products. Rather than wait for the appearance of immunotoxic effects in patients, pre-clinical assays that characterize these potential events in experimental models not only can facilitate the understanding of how these phenomena arise, but sometimes reveal findings that can lead to modifications in currently-accepted methodologies used for estimating the pre-clinical assay predictive risk values. Pentamidine isethionate, a drug used as a second option in treating leishmaniasis, has been repeatedly shown in animals and humans to exert a toxicity at the renal, cardiac and hepatic level; however, its immunotoxic effect have not been as fully evaluated. The potential immunomodulatory effects of pentamidine on splenocytes from two species of mice (BALB/c and ICR) were evaluated ex vivo in this study. The results here indicated that there were significant differences in subpopulation profiles between the strains even before treatment with the drug, with the variances most apparent regarding the relative percentages of CD8+ and CD19+ cells in splenocyte preparations. The data also showed that the drug preferentially induced toxicity in BALB/c mice CD8+ and CD19+ cells more so than in their ICR counterparts. Conversely, the ICR cells seemed to be more susceptible to drug-induced spontaneous blastogenesis than the cells from BALB/c hosts. The differential results seen here confirm that: any potential immunomodulatory effect of a drug should be studied in at least two different mouse strains during an evaluation of overall toxicologic potential; and, there should be attempts to correlate phenotypical findings with those of functionality before a drug can truly be deemed a safe compound.
| Abbreviations | ||
| 7AAD | = | 7-amino-actinomycin D |
| ISI, | = | inverse stimulation indices |
| MFI, | = | mean fluorescence intensity |
INTRODUCTION
Evaluating the immunotoxic potential of different chemicals, drugs or xenobiotics has become essential in current protocols for the safety assessment of these new molecules (Dean and Murray, Citation1991; Roth et al., Citation2006). Toxicity affecting the immune system is related to a variety of adverse effects including suppression, modulation and increased immune system activity. An immunosuppressive effect is often associated with the reduction in the natural resistance of a host to infections, whilst immunostimulation could manifest itself as an autoimmune disease or as hypersensitivity (Descotes, Citation2003; Germolec, Citation2004). It has been shown that some drugs can promote the appearance of autoimmune diseases or allergic reactions at the pharmacological level (Descotes, Citation2004). Previous reports have supported the importance of additional immunotoxicity studies based on factors such as: (1) findings in conventional toxicity studies; (2) pharmacological properties of the compound; (3) drugs administered via topical or respiratory route; (4) drugs administered to pregnant women or used in HIV treatment; (5) structural homology with recognised immunomodulators; and, (6) information regarding adverse effects in clinical or non-clinical assays (Spanhaak, Citation2006).
Pentamidine isethionate (a diamidine drug used in treament of trypanosomiasis and pneumocystosis, as well as a second option drug for treating leishmaniasis) has been shown to cause renal, cardiac, and hepatic damage (Balslev and Nielsen, Citation1992; Andersen et al., Citation1986; Poola et al., Citation2003; Antoniou and Gough, Citation2005; Eckhardt et al., Citation2005; Olliaro et al., 2005). However, very little is known about its immunotoxic effects in humans. What is known through in vitro model studies is that any immunomodulatory activity of pentamidine is related to a cytokine- and chemokine-mediated anti-inflammatory effect and these effects are dose-dependent (Corsini et al., Citation1992; Van Wauwe et al., Citation1996).
Based on the fact that rodents have frequently been used for studying immunotoxicology due to their easy handling, short life spans and their well-characterized immune system (Lovik, Citation1997), it was decided here to study the immunomodulatory effects of pentamidine using BALB/c and ICR mice. The results derived from these studies have led us to confirm the increasingly-accepted fact that when evaluating the potential immunotoxicity of any formulation, it must be studied in at least two different strains of mice, and attempts should be made to correlate phenotypic findings with functional ones prior to designating the drug as being a safe compound.
MATERIALS AND METHODS
Mice
Nine ICR (Pharmacy Department Biotherium, Universidad Nacional de Colombia) and eleven BALB/c (Instituto Nacional de Salud Biotherium) 4-week-old female mice were used in this study. Animals were kept in groups of 3 per cage and 1 cage with 5 individuals (corresponding to a group of BALB/c mice). During the experimentation, mice were fed with standard pellet diet (Rodentina, Purina S.A., Bogotá, Colombia) and water ad libitum. All experiments were approved by the ethical committee of the Fundación Instituto de Inmunologia de Colombia.
Obtaining Splenocytes
The mice were sacrificed by cervical dislocation, according to norms for manipulating Biotherium animals issued by the Pharmacy Department, Science Faculty, Universidad Nacional de Colombia. The spleens were extracted in sterile conditions using a biosafety cabinet. The organs were placed in Petri dishes with RPMI 1640 medium (GIBCO, Scotland, UK) supplemented with fetal bovine serum (FBS, GIBCO). Single cell suspensions of splenocytes were prepared by syringe-mediated disaggregation. Mononuclear splenocytes were then separated by using a Ficoll-Hypaque density gradient with a 1.077 density (ICN Biomedical Inc., Aurora, OH).
Pentamidine Exposure
The tested drug was pentamidine isethionate [Pentacarinate (May and Baker, Dagenham, UK)]. The drug concentration range used here was selected based upon previous reports; a level was used that demonstrated a specific anti-leishmanial effect while having minor effects on target cells (Delgado et al., Citation2001). The results obtained for pentamidine were validated with the use of cyclosporine A (CsA) as a positive control (data not shown) known for immunosuppressive activity (Yang et al., Citation2005; Rezzani et al., Citation2006; Roy et al., Citation2006).
Splenocytes obtained from the naive mice were used to determine the effects of the drug on cell division and upon sub-population drug-dependent mortality. For the spontaneous blastogenesis analysis, aliquots of the harvested splenocytes from each mouse were placed in 96-well dishes at 2 × 105 cells/well and then exposed (in triplicate) to 400, 200, 100, or 50 μ g pentamidine/ml without mitogens; control wells received vehicle only. To assess subpopulation mortality, aliquots of cells were into four subgroups: one for analysis of baseline subpopulation frequencies (i.e., markers measured immediately using duplicate samples from each mouse) and the three remaining for the study of pentamidine-dependent splenocyte expression and mortality after 24, 48, or 72 hr of culture in the presence of one of two different concentrations of the drug. Specifically, 4 × 105 cells/well (in duplicate wells) were placed in the wells of 24-well dishes and then received either 200 or 50 μ g pentamidine/ml. The kinetics of expression of the CD4+, CD8+, CD14+, and CD19+ markers on the exposed splenocytes were subsequently analyzed at different timepoints (0 [basal expression], 24, 48 and 72) during the exposure regimens.
Spontaneous Blastogenesis Assay
Splenocytes (2 × 105 cells/well) were placed into triplicate wells in 96-well dishes containing the drug dilutions. The supernatants were collected after 72 hr of incubation, and cells were pulsed with 1 μ Ci/well [3H]-thymidine (Amersham-Pharmacia, Buckinghamshire, UK) in fresh (drug-free) culture medium. The cultures were harvested 16 hr later and incorporated radiation was quantified on an LS6500 scintillation counter (Beckman, Fullerton, CA). Inverse stimulation indices (ISI) were obtained from the equation:
Flow Cytometry
Splenocytes obtained immediately after spleen perfusion (baseline count), and treated with drug concentrations at different times, were labelled with antibodies specific for CD4 (anti-mouse CD4-fluorescein isothiocyanate [FITC] conjugate), CD8 (anti-mouse CD8-phycoerythrin-[PE]), CD14 (anti-mouse CD14-FITC), and CD19 (anti-mouse CD19-FITC) co-receptors (all DAKOCorp, Carpinteria, CA). 7-amino-actinomycin D (7AAD, Pharmingen, San Diego, CA) was used as a viability marker. Expression of these markers was monitored by flow cytometry (FACScan, Becton Dickinson, San José, CA) every 24 hr. In each analysis, 30,000 events were acquired and the cell population behaviour in R1 was analysed (Pheng et al., Citation2000, Citation2002). The results obtained reflect the means (averages) for all replicates of mice in the same strain.
Statistical Analyses
The results were analysed using the template created with CellQuest software (Becton Dickinson). A two-tail Student's t-test was performed in order to compare the obtained results against those from pentamidine-free controls. Results were considered statistically significant at p < 0.05.
RESULTS
Determining Baseline Percentages of CD4+, CD8+, CD14+, and CD19+ Cells among BALB/c and ICR Mice Splenocytes
All results obtained were the average percentages (±SD) of cells that were positive for each cell marker within each set of mouse strains evaluated. These analyses indicated that the baseline percentages of CD4+ cells were equal in the two mice populations (). However, the baseline percentage of CD8+ cells was greater in ICR mice than in the BALB/c ones, while the latter strain had the greater percentage of CD19+ cells. Populations M1 and M2 values were designed according to auto-fluorescence control read in the same fluorescent channels (FL-1 for FITC and FL-2 for PE). With respect to the quantity of molecules (receptors) per cell, analysis via the mean fluorescence intensity (MFI) determinations for each respective antigen () indicated that the only significant differences detected between strains was in the expression of CD14+ co-receptors on splenocytes (i.e., BALB/c levels > ICR levels).
TABLE 1 Baseline relative percentages of positive cells for and mean fluorescence intensity (MFI) of antigens evaluated in murine splenocytes
Functional Assay for Establishing Immunomodulatory Activity of Pentamidine
To establish any immunomodulatory activity of pentamidine, 2 × 105 splenocytes were placed in triplicates in 96-well dishes containing the drug dilutions. The cells were pulsed with 1 μ Ci/well [3H]-thymidine and the incorporated radiation was subsequently quantified. Though this spontaneous blastogenesis assay was not expected to be conclusive about all facets of the potential immunomodulatory activities of the drug, important strain-related differences in mono-nuclear (non-phagocyte) cell responses to the drug were nevertheless apparent. Only at the lowest dose (i.e., 12.5 μ g/ml) tested did cells from BALB/c mice display a greater proliferation than their ICR counterparts (). These results might enable a supposition to be made that cells from BALB/c hosts manifested a resistance to/down-regulation of the apparent proliferation-inducing effects of the pentamidine seen with the ICR cells at the other tested doses. While this finding remains to be more fully explored, it is clear that the pattern observed with the BALB/c cells was not due to any generalized increase in cell mortality. The fact that at doses of drug > 12.5 μ g/ml, the ISI values did not rise to well above 1.0 seemed to negate this as a possible explanation for the strain differences in susceptibility to modulation of at least this one function of these immune system cells.
FIG. 1 Functional assay of blastogenesis with ICR and BALB/c spleen cells treated ex vivo with pentamidine. Splenocytes (2 × 105) were placed into wells of a 96-well plate containing various concentrations of the drug. After 72 hr of incubation, the medium was removed and the cells were pulsed with 1 μ Ci [3H]-thymidine/well in fresh drug-free medium. Cultures were harvested 16 hr later and incorporated radiation was quantified. Data are reported in terms of counts per minute. All results were generated from analysis of triplicate samples for each mouse; total mice analyzed (N) were 9 and 11 for the ICR and BALB/c strains, respectively.
![FIG. 1 Functional assay of blastogenesis with ICR and BALB/c spleen cells treated ex vivo with pentamidine. Splenocytes (2 × 105) were placed into wells of a 96-well plate containing various concentrations of the drug. After 72 hr of incubation, the medium was removed and the cells were pulsed with 1 μ Ci [3H]-thymidine/well in fresh drug-free medium. Cultures were harvested 16 hr later and incorporated radiation was quantified. Data are reported in terms of counts per minute. All results were generated from analysis of triplicate samples for each mouse; total mice analyzed (N) were 9 and 11 for the ICR and BALB/c strains, respectively.](/cms/asset/cf1873b6-bfec-4d14-b9bc-0664b860acbd/iimt_a_267864_uf0001_b.gif)
Phenotype Evaluation of CD4+, CD8+, CD14+, and CD19+ Splenocytes by Flow Cytometry
For determining the effect of pentamidine isethionate on different spleen cell populations, isolated splenocytes were cultured in the presence of one of two concentrations of the drug and labeled every 24 hr thereafter with monoclonal antibodies specific for CD4, CD8, CD19, and CD14 co-receptors; 7AAD was used as a viability label. The data in , and reflect the percentages of drug-dependent mortality (quantified by 7AAD staining) for each splenocyte subpopulation analyzed. These data were based upon cytometric analyses for each cell population (for both mouse strains) that generated density plots at the different timepoints during the ex vivo drug exposures (the various plots are not shown here). Based on the assumption that the level of antigen recognition by each fluorescent-antibody was unaffected by the exposure/drug itself (i.e., epitopic modification did not occur), cell population shifts observed across a series of plots were used to reflect time-dependent changes in viability for each given subpopulation.
TABLE 2 CD4 cell mortality
TABLE 3 CD8 cell mortality
TABLE 4 CD19 cell mortality
In the plots generated in every assay, each quadrant provides information about the status of cells at each timepoint analyzed. Specifically, subpopulations corresponding to the left-lower (double negatives) quadrant were those viable cells not recognized by the given surface antigen-specific antibody against CD4, CD8, CD14, or CD19). Cells localized to the right-lower quadrant (single positives) were those that were viable and stained with the specific antibody only. Cells which presented themselves in the left-upper (single positives) and right-upper (double positives) were those that were dying/dead and so marked with supravital stain (7AAD) only or with 7AAD and the specific antiboidy, respectively. Based upon these defined segregations, the percentages of cells present in each quadrant were converted to a numeric value using CellQuest software; those in the right-upper quadrant (i.e., double positives) were used to permit estimates to be calculated of any changes in relative numbers of dead cells within each specific splenocyte subpopulation. For example, after 24 hr of exposure, among BALB/c mice cells treated with 200 μ g pentamidine/ml, the 7AAD marker was detected in 11.2% of all CD8+ cells. Thus, it can be said there was a 100% increase (doubling) in mortality compared to that in cells harvested after 24 hr in the absence of pentamidine (i.e., negative control; ≈5.6% mortality among CD8+ cells).
Employing these analytical approaches, a two-tail Student's t-test was used to compare results from all treated cells against those from the untreated controls. Among cells from BALB/c hosts, differences in viabilities were found for treatments within 24 hr of culture. Balb/c strain splenocyte subpopulations bearing CD8 or CD19 markers demonstrated increased cell death (quantified by 7AAD staining) due to a presence of 200 μ g/ml pentamidine isethionate as compared to untreated control cells; treatment with the lower 50 μ g/ml dose had no effect. In contrast, no such effect was evident with the cells from the ICR mice. The viabilities of the CD4+ subpopulations obtained from either strain were not modulated in the presence of any dose of pentamidine (). In contrast, CD14+ splenocytes were susceptible to all treatments (data not shown). By 48 hr of culture (and thereafter), there were no drug-associated effects on viabilities among any of the splenocyte subpopulations, regardless of parent strain. In all of these studies, strain differences (p < 0.05) were only detected with the CD8+ cells treated with 200 μ g/ml pentamidine for 24 hr.
It is important to note that no changes related to the surface antigens themselves were detected in these studies. Among all the cells (live + dead/dying) examined in each experiment, the expressions of each marker on the surface of the cells were unaffected. The only changes observed after drug treatments were those noted above regarding differences within the mortality percentages among some cell subtypes.
DISCUSSION AND CONCLUSIONS
A study was undertaken here to examine the potential adverse effect(s) of pentamidine on the immune system in a murine model (Schulte and Ruehl-Fehlert, Citation2006). The main reason for pursuing these studies was to better understand the mechanisms underlying some of the toxic effects that had been noted in patients exposed to the suggested dose of the drug during treatment of parasitic diseases (such as leishmaniasis) (Lightburn et al., Citation2003; Atsmon and Dolev, Citation2005). Taking advantage of the murine model's reported usefulness in the immunotoxicologic evaluation of drugs (Lang et al., Citation1993; Holladay and Blaylock, Citation2002; Haley, Citation2003), it was decided that splenocytes from adult female TH2 profile BALB/c (Launois et al., Citation1999) and TH1 profile mice would be used for these preliminary screenings of the immunomodulatory activities of this drug.
The selection of two differing strains – as opposed to only one type – for analyses was seen to be a wise one. Determinations of the baseline (i.e., without drug) values for splenic mononuclear (non-phagocyte) subpopulation levels at the start of the study clearly indicated that there were significant differences between the BALB/c and ICR mice (re: their respective relative percentages of CD8+ and CD19+ cells). Based on this, we conclude that it is critical in future studies in our laboratories (and other Investigators) that reference values for more than one strain of mouse model initially be established prior to any subsequent analyses of immunomodulation by a given toxicant. To not do so risks unintentionally “selecting in” or “out” for a potential effect. Nevertheless, it is clearly not enough to determine phenotypical parameters (e.g., by quantifying cell subpopulations) of isolated cells to be able to come to conclusions about the immunotoxic potential of any test agent.
Such results could be considered artifactual if they are not accompanied by functional studies that can establish the biological relevance of the cytometry observation (Spanhaak, Citation2006). In the current studies, the isolated splenocytes from each mouse strain were tested for their ability to spontaneously proliferate in the presence of the pentamidine. The use of differing strains (as noted in the Results) led to unequal responses, except at the extremes of the dose ranges tested; whether this is related to some differences between the ICR and BALB/c splenocytes in sensitivity to a triggering of DNA synthesis and/or proliferation by the drug, varying resistance to lethality to the drug, or even remotely to the an induced effect on the ability to take up exogenous nucleotide, remains to be determined.
Comparative functional studies in the literature are mostly lacking. Only a study by Ferrante et al. (Citation1985) has assessed the effects of pentamidine on lymphocyte spontaneous and mitogen-induced proliferation. In those studies that used athymic (nude) and BALB/c mice, the presence of the drug again (albeit at concentrations less than those used in the current studies) led to dose-trend decreases in thymidine incorporation. No information about responses of ICR mice cells is available in the current scientific literature. Nonetheless, the findings here again point out the need for more than one strain of mouse model when performing analyses of immunomodulation by any given toxicant.
As noted by other investigators, in their capacity to provide support to immunotoxicological studies in experimental models, only methodologies should be employed that have had their sensitivity, reproducibility, and predictive values previously determined (Holladay and Blaylock, Citation2002). With respect to pharmaceuticals in particular, assays for evaluating immune system activity in experimental models must also be modified (or complemented) to improve upon their predictive values of the potential immunotoxicities of not only drugs under develop-ment but those already on the market as well (Snodin, Citation2004; Descotes, Citation2005). We believe the critical finding of the studies here – that selection of only one host strain should be avoided as this could unintentionally lead to “selecting in” or “out” for potential effects – should be considered during the development of any ex vivo screening method for the analyses of potential immunotoxicities of pharmacological products. In this study, in spite of not having found statistically-significant quantitative differences, the tendency observed with splenocytes from BALB/c mice to resist pentamidine-induced spontaneous blastogenesis stands in contrast to that with the cells from the ICR hosts (; i.e., ISI values with ICR cells drop and stay below 1.0 over a greater concentration range [i.e., from 25 to 200 μ g/ml] compared to responses of cells from BALB/c counterparts [susceptible to drug only to extreme concentrations]) is something we are interested in exploring further.
In conclusion, pentamidine isethionate appeared to preferentially induce a toxic effect in BALB/c CD8+ and CD19+ splenocytes, but had little capacity to induce spontaneous blastogenesis by the intact splenocyte population. In contrast, splenocytes from ICR mice appeared to be more resistant to toxicity from the drug, but displayed a greater propensity to undergo drug-induced blastogenesis – albeit in a non-linear dose-response manner. These results confirm our hypothesis that immunomodulatory effects should be studied in at least two different mouse strains when evaluating the immunotoxic potential of any formulation, and that phenotypical findings need to be correlated with analyses of functionality.
Implementing methodologies such as those proposed in this study has emerged as a necessity, supported by immunotoxicological studies in experimental models, meaning that their sensitivity, reproducibility and predictive values must be previously determined (Holladay and Blaylock, Citation2002). Assays for evaluating immune system activity in experimental models must also be modified or complemented to lead to improved predictive immunomodulatory values, not just for drugs being developed but also for those already being sold (Snodin, Citation2004; Descotes, Citation2005). The tools used in this study could become a method of ex vivo screening proving useful for analysis of potential immunotoxicity of pharmacological products.
We would like to thank the Pharmacy Department from the Universidad Nacional de Colombia and the Instituto Nacional de Salud (INS) for donating the mice used in the study. We would also like to thank Jason Garry for translating the manuscript. This research was performed with support from the Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnología “Francisco José de Caldas” “Colciencias” Contract #060/2006. We would also like to thank Becton Dickinson Colombia and Equimed Ltda. for supplying some of the materials and reagents needed for carrying out the project. The authors have no conflicts of interest that are directly relevant to the content of this study. Authors Plaza and Marino are also affiliated with the Biology Department, Universidad Nacional de Colombia, Bogota, Columbia; Authors Marino and Delgado have dual affiliation with Pharmacy Department, Universidad Nacional de Columbia, Bogota, and Author Plaza is also affiliated with the Fundacion Instituto de Immunologia de Colombia, Bogota.
REFERENCES
- Andersen R., Boedeker M., Ma M., Goldstein J. C. Adverse reactions associated with pentamidine isethionate in AIDS patients: Recommendations for monitoring therapy. Drug. Intell. Clin. Pharm. 1986; 20: 862–868
- Antoniou T., Gough K. A. Early-onset pentamidine-associated second-degree heart block and sinus bradycardia: Case report and review of the literature. Pharmacotherapy 2005; 25: 899–903
- Atsmon J., Dolev E. Drug-induced hypomagnesaemia: Scope and management. Drug. Saf. 2005; 28: 763–788
- Balslev U., Nielsen T. L. Adverse effects associated with intravenous pentamidine isethionate as treatment of Pneumocystis carinii pneumonia in AIDS patients. Dan. Med. Bull. 1992; 39: 366–368
- Corsini E., Dykstra C., Craig W. A., Tidwell R. R., Rosenthal G. J. Pneumocystis carinii induction of tumor necrosis factor-α by alveolar macrophages: Modulation by pentamidine isethionate. Immunol. Lett. 1992; 34: 303–308
- Dean J. H., Murray M. J. Toxic responses of the immune system. Toxicology: The Basic Science of Poisons, 4th Edition, M. O. Amdur, J. Doull, C. D. Klaasen. Pergamon Press, New York 1991; 280–333
- Delgado G., Puentes F., Moreno A., Patarroyo M. E. Flow cytometry, a useful tool for detecting the lethal effect of pentamidine on Leishmania viannia complex promastigote forms. Pharmacol. Res. 2001; 44: 281–286
- Descotes J. From clinical to human toxicology: Linking animal research and risk assessment in man. Toxicol. Lett. 2003; 11: 140–141
- Descotes J. Importance of immunotoxicity in safety assessment: A medical toxicologist's perspective. Toxicol. Lett. 2004; 149: 103–108
- Descotes J. Immunotoxicology: Role in the safety assessment of drugs. Drug. Saf. 2005; 28: 127–136
- Eckhardt L. L., Rajamani S., January C. T. Protein trafficking abnormalities: A new mechanism in drug-induced long QT syndrome. Br. J. Pharmacol. 2005; 145: 3–4
- Ferrante A., Secker K., Thong Y. H. Selective inhibition of the in vitro murine B lymphocyte response by Pentamidine. Int. J. Immunopharmac 1985; 7: 281–285
- Germolec D. R. Sensitivity and predictivity in immunotoxicity testing: Immune endpoints and disease resistance. Toxicol. Lett. 2004; 149: 109–114
- Haley P. J. Species differences in the structure and function of the immune system. Toxicology. 2003; 188: 49–71
- Holladay S. D., Blaylock B. L. The mouse as a model for developmental immunotoxicology. Hum. Exp. Toxicol. 2002; 21: 525–531
- Lang D. S., Meier K. L., Luster M. I. Comparative effects of immunotoxic chemicals on in vitro proliferative responses of human and rodent lymphocytes. Fundam. Appl. Toxicol. 1993; 21: 535–545
- Launois P., Himmelrich H., Tacchini-Cottier F., Milon G., Louis J. A. New insight into the mechanisms underlying TH2 cell development and susceptibility to Leishmania major in Balb/c mice. Microbes Infect 1999; 1: 59–64
- Lightburn E., Morand J. J., Meynard J. B., Kraemer P., Chaudier B., Pages F., Garnotel E., Patte J. H., Banzet S., Dampierre H., Lepage J., Morillon M., Boutin J. P., Hovette P., Chouc C. Management of American cutaneous leishmaniasis. Outcome apropos of 326 cases treated with high-dose pentamidine isethionate. Med. Trop. 2003; 63: 35–44
- Lovik M. Mutant and transgenic mice in immunotoxicology: An introduction. Toxicology. 1997; 119: 65–76
- Olliario P. L., Guerin P. J., Gerstl S., Haaskjold A. A., Rottingen J. A., Sundar S. Treatment options for visceral leishmaniasis: A systematic review of clinical studies done in India, 1980–2004. Lancet Infect. Dis. 2005; 5: 763–774
- Pheng S. R., Augera C., Chakrabarti S., Massicotte E., Lamontagne L. Sensitivity to methylmercury-induced autoimmune disease in mice correlates with resistance to apoptosis of activated CD4+ lymphocytes. J. Autoimmun. 2002; 20: 147–160
- Pheng S. R., Chakrabarti S., Lamontagne L. Dose-dependent apoptosis induced by low concentrations of methylmercury in murine splenic Fas+ T-cell subsets. Toxicology 2000; 149: 115–128
- Poola N. R., Kalis M., Plakogiannis F. M., Taft D. R. Characterization of pentamidine excretion in the isolated perfused rat kidney. J. Antimicrob. Chemother. 2003; 52: 397–404
- Rezzani R., Pompili E., Agostini C., Buffoli B., Bonomini F., Nori S., Fumagalli L., Bianchi R. Expression of non-muscle myosin heavy chain in rat heart after immunosuppressive treatment. Int. Immunopharmacol. 2006; 6: 962–967
- Roth D. R., Roman D., Ulrich P., Mahl A., Junker U., Perenes E. Design and evaluation of immunotoxicity studies. Exp. Toxicol. Pathol. 2006; 57: 367–371
- Roy M. K., Takenaka M., Kobori M., Nakahara K., Isobe S., Tsushida T. Apoptosis, necrosis and cell proliferation-inhibition by cyclosporine A in U937 cells (a human monocytic cell line). Pharmacol. Res. 2006; 53: 293–302
- Schulte A., Ruehl-Fehlert C. Regulatory aspects of immunotoxicology. Exp. Toxicol. Pathol. 2006; 57: 385–389
- Snodin D. J. Regulatory immunotoxicology. Does the published evidence support mandatory non-clinical immune function screening in drug development?. Regul. Toxicol. Pharmacol. 2004; 40: 336–355
- Spanhaak S. The ICH S8 immunotoxicity guidance. Immune function assessment and toxicological pathology: Autonomous or synergistic methods to predict immunotoxicity?. Exp. Toxicol. Pathol. 2006; 57: 373–376
- Van Wauwe J., Aerts F., Van Genechten H., Blockx H., Deleersnijder W., Walter H. The inhibitory effect of pentamidine on the production of chemotactic cytokines by in vitro stimulated human blood cells. Inflamm. Res. 1996; 45: 357–363
- Yang Z., Peng Y., Wang S. Immunosuppressants: Pharmacokinetics, methods of monitoring and role of high performance liquid chromatography/mass spectrometry. Clin. Applied. Immunol. Rev. 2005; 5: 405–430