4,4'-Methylenediphenyl diisocyanate, oligomeric reaction products with 2,4'-diisocyanatodiphenylmethane

Specific investigations: other studies

Additional information

Study records 7.9.3a-g add to the knowledge regarding adducts of MDI with biological molecules. They are relevant to the general Toxicokinetics section 7.1, as well as to the consideration of carcinogenicity section 7.6.

7.9.3a, b:

Vock et al (1995) used sensitive in vitro methods to ascertain potential binding of MDI to nucleotides and then to rat skin. Adduct levels of MDI to skin DNA were only some 20-fold above the limit of detection. The authors concluded that a minute fraction of MDI can reach DNA in vivo in a chemically reactive form. In comparison with the genotoxic skin carcinogen 7,12-dimethylbenz[a]anthracene on the other hand, the DNA-binding potency of MDI was more than 1000-fold lower. In their studies of 1996, Vock et al used tissues obtained from the rats exposed to MDI by inhalation as reported by Hoymann et al (1995), that is to say the bioassay to females with 17h/d exposures to monomeric MDI. Using the P-32 postlabelling techniques, adducts of MDI with DNA were not detected in lung, respiratory epithelium, liver, kidney, bladder or lymphocytes. Using the technique to assess adducts of the diamine MDA with DNA, very low levels, barely above detection limits, were found in olfactory epithelium only, other tissues being negative.  This result is questionable as the control exposure group had higher adduct levels that the MDI exposed groups. Greim (2008) suggested high protein binding might be responsible. Overall the marginal DNA binding of MDI only in olfactory epithelium of rats chronically exposed to MDI aerosol, but not in lung or other organs, indicate that the lung tumours likely arise from mechanisms other than direct DNA binding. 

 

7.9.3c

Sepai et al (1995) report on biomarkersobtained from the rats exposed to MDI by inhalation as reported by Hoymann et al (1995), that is to say the bioassay to females with 17h/d exposures to monomeric MDI, with an additional single exposure group. Blood haemoglobin adducts and urinary metabolites after acid or base hydrolysis were quantified. Both haemoglobin adducts and urinary hydrolysed metabolites were found in dose- and exposure duration – related levels, although at 3 months of exposure adducts seemed to be saturated. The authors discuss presence of “free” MDA and acetylated-MDA as urinary metabolites after base extraction. However the base extraction conditions used were certainly capable of hydrolysing carbamates to the diamine, and the data indicate the presence of MDI-carbamate conjugates in the urine rather than free MDA. Furthermore the chemistry of MDI interations with amino acids and proteins as investigated by Mormann et al (2006, 2008) show that acetylated MDI may be formed without MDA as an intermediate. The acetylated MDA is thus derived from the laboratory hydrolysis of samples and not as a direct metabolic intermediate.

 

7.9.3d, e

The possibility of detection of glutathione-MDI metabolite in urine was investigated by Bartels et al (2009). Animals were exposed to either MDI, or to a bis-glutathione-MDI conjugate. Although MDI metabolites were excreted, no structures were identified that could be directly assigned as MDI-glutathione metabolites. Leng (2005) also investigated blood and urine metabolites from rats exposed by intratracheal instillation of bis-glutathione-MDI. All samples were hydrolysed before analysis. While the results support the GHS-MDI conjugate as an intermediary metabolite, yields were low and MDI-specific markers were not sought in this work. There seems to be a confounding false-positive background level of hydrolysis-derived MDA (also seen by Sepai et al, 1995). As it is clear that glutathione plays an important role in MDI metabolism in vivo, it seems that the transfer process as described by Mormann and others predominates and glutathione MDI-specific metabolites are not useful as exposure biomarkers. 

 

7.9.3f, g, o

Mormann et al 2006 studied reactions of isocyanates, including 4,4’-MDI, with N-acetyl-L-cysteine (AcCys) at different molar ratios in aqueous buffer solutions of pH 5.0 – 7.4. Type and amounts of products formed in these reactions were identified and quantified. Conjugates and ureas were found to be the main products. The ratio of these two compounds varied with the ratio of AcCys to isocyanate. Minor amounts of MDI-conjugates were found apart from high amounts of insoluble material, which proved to be unreacted MDI encapsulated by oligomeric ureas. The reaction of the –SH group with the isocyanate moiety was independent of the pH of the solution in the range studied. No diamine, i.e. 4,4’-MDA, could be detected. Reactions of the isocyanates with an aqueous buffer solution of pH 6.5 in the absence of AcCys gave ureas as main products, while significant amounts of unreacted diisocyanates remained encapsulated in the mixture. No 4,4’-MDA was detected under these conditions.

Mormann and Frank (2004) and Mormann et al. (2008) reported transcarbamoylation of AcCys adducts of MDI with model compounds to assess thiolysis, aminolysis, alcoholysis and hydrolysis in aqueous phosphate buffer solution and in dimethylacetamide (DMAc). In aqueous buffer reaction rate was thiolysis > aminolysis > hydrolysis. No 4,4’-MDA could be detected. Under physiological conditions hydrolysis should compete with thiolysis under homogeneous conditions while ureas and carbamates should be much more stable against hydrolysis. No free isocyanate groups could be detected in any of the reactions. In conclusion the isocyanate moiety in thiocarbamates is readily transferred to sulfhydryl- and amino groups but not to aliphatic hydroxyl groups. Under physiological conditions hydrolysis competes with these transcarbamoylation reactions

Isocyanate groups readily react with active hydrogen containing nucleophiles in hydroxy-, amino-, and sulfhydryl group containing bio-molecules of living organisms. If an aqueous medium contains reactive biological compounds such as amino acids, peptides or proteins, competing reactions of the aromatic isocyanates with water and functional groups (e. g. NH₂, NH, OH, SH, and COOH) in biomolecules can be assumed. Adducts of isocyanates and mercapto groups of cysteine have been identified as key intermediates (Baillie and Slatter, 1991). These thiocarbamates were shown to be reactive intermediates which can transfer the carbamoyl residue to other active hydrogen containing molecules present in the system, e. g. to amine functions in lysine or N-terminal groups in proteins and peptides. This in vitro work shows the chemical feasibility of the reaction sequence apparent in vivo, that is the initial reaction of MDI with lung glutathione and surfactant, followed by transfer to systemic proteins (albumin and globin). 

Study records 7.9.3h-q provide additional data for consideration with respect to the carcinogenic mode of action of MDI (Section 7.6)

Two acute exposure investigative studies to identify lung effects and recovery were conducted using polymeric MDI (Pauluhn, 2000; Kilgour et al., 2002). Pauluhn (2000) exposed rats to pMDI by nose-only inhalation for 6 hours at concentrations of 0, 0.7, 2.4, 8 or 20 mgMDI/m3 and samples of BAL fluid were collected at 0 and 3 hours, 1, 3 and 7 days after exposure. The only significant change at 0.7mg/m3 was increased protein in BAL fluid at 1 and 3 hours after exposure. The conclusion of the study was that the effects, particularly elevation of GSH, were dose-related and transient over the 3 or 7 days post exposure period reflecting a compensatory response to pMDI interacting with the air-blood barrier. There was no evidence of any appreciable acute cytotoxicity to the endothelial cells.

The study of Kilgour et al. (2002), using 6 hour nose-only exposures at concentrations of 0, 10, 30 and 100 mg pMDI/m3, also reported cellular and protein changes which returned to control values 10 days after exposure. Cell replication measured by BrdU incorporation was seen in the terminal bronchiolar and the alveolar regions of the lungs in all groups 3 days after exposure.

Histopathological examination of the lungs one day after treatment showed dose-related accumulation of macrophages in the airways, pneumonitis and cell exudates in the lumen. At day 3 after exposure these effects were resolving but with bronchial hyperplasia and hypertrophy evident. By day 10 all evidence of effects including the bronchial hyperplasia and hypertrophy had resolved. Ultra-structural examination revealed surfactant and cell debris in the lumen together with dose related hyperplasia of Type II cells and increased cellular surfactant. Overall the study showed a distinct pattern of effect and recovery thought to be due in part to stimulation of the Type II cells. In less well detailed studies Reuzel et al. (1994b) reported similar lung effects at similar exposure concentrations.

Two short term, repeat dose investigative inhalation studies were undertaken by Pauluhn et al. (1999) and Kilgour et al. (2002). In the study reported by Pauluhn et al. (1999), rats received repeated exposures over two weeks while Kilgour et al. (2002) exposed animals for four weeks followed by a one month recovery period

The major findings of these studies can be summarised as:

• Increased numbers of cells, (in particular polymorphonuclear neutrophils and lymphocytes), in BAL fluid collected from animals exposed for 4 weeks to pMDI at both the mid- and high dose levels.
• Macrophages from animals exposed for either 2 or 4 weeks to the two highest concentrations of MDI appeared ”foamy” with biochemical analysis showing increased phospholipid content, considered to be indicative of perturbation of pulmonary surfactant homeostasis.
• BAL lavage fluid obtained from animals in the high dose groups exposed for 2 or 4 weeks showed increased levels of phosphatidylcholine and protein and increased activities of γ-GT, LDH, ALP and NAG.
• Dose related increases in cell proliferation, measured by BrdU incorporation, were seen in the terminal bronchioloalveolar regions at all concentrations in both studies. The cell division at lower exposure concentrations was without any apparent inflammatory response and damage to Type I pneumocytes was not seen at any exposure level.
• Electron microscopy of lung tissue showed evidence of dose-related increases of surfactant in the lumen, in alveolar macrophages and in Type II cells suggesting activation of Type II cells.
• Following the 30-day recovery period, virtually all observed responses had returned to normal.

Measurement of biomarkers in blood and urine after intratracheal exposure of rats to MDI-GSH bis-adduct provided evidence that this adduct was a potential intermediary metabolite for inhaled MDI (Pauluhn et al., 2006). MDI adducts with the N-terminal valine of haemoglobin (Sabbioni et al., 2000) and with N-lysine of albumin (Kumar et al., 2009) have been quantified in rats exposed by the protocol of Hoymann et al. (1995). MDI-albumin adducts have also been identified in exposed humans (Sabbioni et al., 2010, cfr. study records 7.1.1r and 7.10.5e; Wisnewski et al., 2010, cfr. study records 7.1.1q and 7.10.5f).

Pauluhn (2002a) investigated the time course of exposure biomarkers (protein adducts) against lung effects at a single exposure concentration of 12.9 mg/m3 , 6h/d for 10 exposure days monitoring lung lavage and urine for a following five weeks. While haemoglobin adducts were reflective of cumulative dose, plasma protein adducts and cellular indices reached a plateau during the exposure period, and resolved rapidly on cessation of exposure.

Early pulmonary responses following inhalation exposures to MDI are characterized by extravasation of serum and cell proteins into the lining fluids of the lung evidenced by increases in protein, cell content, GSH and phospholipids in BAL fluid. Such elevations can be viewed as part of the normal adaptive response to chemically reactive particulates. The MDI-reaction products can then be removed either by phagocytic cells such as alveolar macrophages, (as indicated by increased number of foamy macrophages seen in the lungs of MDI exposed animals), or mechanically by the mucociliary escalator; an interaction between mucous secretion and cilia providing a mechanism for movement of particles to the oesophagus where they are swallowed (Pauluhn, 2011).

MDI exposure also resulted in histopathological changes in the lung, characterized by hypertrophy and hyperplasia, occurred exclusively in Type II pneumocytes, confined to the terminal bronchioles of the lung. There was no evidence of Type I cell damage and the effects occurred without any apparent inflammatory response. Using MDI exposure concentrations similar to those used in the long term studies, (i.e. about 1 or about 5 mg MDI/m3), cell proliferation in the bronchioloalveolar region measured by BrdU incorporation was increased 4 to 9-fold or 4 to 6-fold at 2 or 4 weeks exposure respectively. There was no evidence of cytotoxicity at these exposure concentrations. Ultrastructural examination of lungs from animals exposed to MDI described enlarged Type II cells, increased biosynthesis and secretion of intracellular surfactant phospholipids. Pauluhn et al. (1999) described these findings as due to persistent stimulation and proliferation of Type II cells.

Comparison of dose response curves for cell proliferation and markers of effect measured in the BAL fluid, (such as protein and cell counts), from the acute, short term and chronic studies carried out on MDI showed similar NOAELs of about 0.5 mg/m3 (Pauluhn 2002b, Greim 2008, Pauluhn 2011). While there were differences in the slopes of the dose response it was considered that such differences reflected adaptation and tolerance of repeated exposures (Pauluhn, 2011). As the NOAELs from the acute, repeat and chronic exposure studies, (i.e. about 0.2- 0.5 mg/m3) are similar, it is unlikely the effects are caused by accumulated particles leading to lung overload and inflammation.