Bis(2-ethylhexyl) phthalate

Immunotoxicity

Description of key information

The weight of evidence suggests that, although some phthalates may have an intrinsic ability to modify adaptive immune responses (under strictly defined conditions, and via mechanisms that have yet to be elucidated), there is no evidence to suggest that these chemicals have a consistent and proven ability to enhance allergic sensitisation under conditions of exposure that are relevant for human health. It is premature therefore to implicate phthalates as having contributed to the increasing prevalence of atopic allergy and asthma.

Additional information

The following is quoted from the review article of Kimber and Dearman in 2009 (An assessment of the ability of phthalates to influence immune and allergic responses ).

Effects of phthalates on antibody expression in vivo

The putative ability of certain phthalates to act as adjuvants for the development of immune responses to known protein allergens has been investigated in animal (almost exclusively mouse) models. In considering the conduct and interpretation of such models it is important to appreciate that the immunological response of greatest moment in the development of allergic sensitisation and of atopic allergic disease (and therefore the event of greatest relevance when considering the possible contribution of the putative adjuvant activity of phthalates in driving susceptibility to allergy) is the production of specific IgE antibody. This is the primary effect molecule of atopic allergy. During the acquisition of sensitisation IgE antibody distributes systemically and associates, via specialised plasma membrane receptors, with mast cells. As described above, subsequent encounter with the inducing allergen causes mast cell activation resulting in inflammation and the signs and symptoms of an allergic reaction. Although specific IgE antibody is therefore the most relevant marker of allergic sensitisation, it is produced in very small amounts and for this (and other reasons) is difficult to measure accurately. As a consequence, other antibody read-outs are commonly used in addition to, or as surrogates for, specific IgE. Those most commonly used in mouse models are total plasma levels of IgE immunoglobulin (as opposed to specific IgE antibody), and the induction of IgG1 antibody responses. Total plasma levels of IgE are frequently elevated during immune responses to allergens, and IgG1 antibody production is regulated by mechanisms similar to those that control IgE responses (Snapper et al., 1988). It is of some importance for the interpretation of data derived from murine models that IgG1 is not only considerably less effective at stimulating mast cell degranulation than is IgE (Ovary, 1982), but also IgG1 antibody levels do not always mirror exactly the same pattern as IgE antibody responses. Thus, we have observed very different dose response profiles for IgG1 anti-OVA antibody responses compared with anti-OVA IgE antibody responses following subcutaneous administration of DEHP to mice (Dearman et al., 2008a). A lack of correlation between IgG1 and IgE specific antibody production was recorded also for detergent enzymes and food proteins (Sarlo et al., 2005).

Another issue is the dose metric to be used for comparison between different types of exposures. Although dose/unit body weight is accepted generally for comparisons between species, and for routes of exposure such as oral or intraperitoneal (ip), it is unclear how appropriate this metric is for consideration, for example, of topical exposure. For the current comparisons we have therefore elected to use total dose per application as the dose metric for comparison, given that all experiments have utilised the same experimental animal (the mouse).

Experimental approaches have included examination of the ability of various phthalate monoesters and diesters (administered by a variety of routes of exposure) to influence specific antibody responses to a reference protein allergen, ovalbumin (OVA), or to a reference immunogen (keyhole limpit haemocyanin; KLH) (Larsen et al., 2001a; 2001b; 2002; 2003; 2007; Lee et al., 2004; Larsen and Nielsen, 2007; 2008; Dearman et al., 2008a; 2008b; 2009; Hansen et al., 2007). The impact of phthalates on the total serum concentration of IgE has also been investigated (Butala et al., 2004; Lee et al., 2004; Dearman et al., 2009). In many, but not all, instances the strain of mouse that is most susceptible to the development of IgE-mediated reactions, the BALB/c strain mouse, has been utilised.

The ability of phthalates to display adjuvant properties under these experimental conditions appears to be influenced very markedly by the route through which exposure to the phthalate is effected. Subcutaneous exposure to some phthalates appears to be particularly effective at causing augmentation of antibody responses to OVA, with relatively low doses of phthalate (0.1 mg-40 mg phthalate per dose) impacting on the vigour of IgG1 and/or IgE antibody responses. When the adjuvant effects of various monophthalates were studied , for some chemicals (mono-2-ethylhexyl phthalate [MEHP], mono-n-octyl phthalate [MNOP] and mono-iso-nonyl phthalate [MINP]), both inhibitory “immunosuppressive” effects (statistically significant decreases in antibody production) and adjuvant effects (statistically significant increases in antibody production) were observed at different doses within the same experiment (Larsen et al., 2001a). Thus, for example, subcutaneous treatment of mice with MINP at 20 mg inhibited anti-OVA IgE antibody production; whereas 2 mg of the same chemical enhanced IgE antibody responses. Moreover, in the same experiment 0.2 mg was found to inhibit anti-OVA IgG1 antibody production. Only inhibitory effects on IgE and IgG1 antibody levels were recorded following exposure to 2 mg mono-iso-decyl phthalate (MIDP), with administration of either 20 mg or 0.2 mg of phthalate being without impact on antibody production.

Using an identical experimental system, and similar concentration ranges, subcutaneous injection of mice with monobenzyl phthalate (MBzP ), or mono-n-butyl phthalate (MBP ), was without effect on either IgG1 or IgE antibody responses (Larsen et al., 2001a). The same investigators have also examined the immunomodulatory effects of various diester phthalates (Larsen et al., 2001b; 2002; 2003). Inhibitory activity was not seen with any of these compounds. However, treatment with 40 mg DEHP was reported to cause increased expression of IgG1 anti-OVA antibody, although it was without effect on IgE anti-OVA antibody levels (Larsen et al., 2001b). Other diester phthalates such as di-n-octyl phthalate (DNOP), DINP and DIDP augmented both IgG1 and IgE anti-OVA antibody responses at certain doses (Larsen et al., 2002). It is not clear, however, why in a subsequent publication (Larsen and Nielsen, 2008) the same investigators reported effects with the same compounds on IgG1 antibody production, in the absence of changes in IgE antibody. In the same experimental system, subcutaneous injection of phthalates that are structurally similar to DEHP, including BBP, bis-(2-ethylhexyl) terephthalate (DOTP) and trioctyl trimellitate (TOTM), was found not to influence either IgG1 or IgE antibody responses (Larsen et al., 2003; Larsen and Nielsen, 2008). The authors speculated that the length of the carbon side chains is important in dictating the biological activity of phthalates in this respect; the suggestion being that a total of 16 carbon atoms distributed between the 2 ester groups confers maximum adjuvant activity (Larsen and Nielsen, 2008).

Intraperitoneal (ip) route of administration of phthalates has also been reported to have adjuvant effects at relatively low doses. Thus, in 129/Sv strain mice ip injection of DEHP (100 mg) with the reference allergen OVA resulted in a significant increase in IgG1 anti-OVA antibody, but was without effect on IgE antibody responses (Larsen and Nielsen, 2008). These data are comparable with previous observations that subcutaneous injection of DEHP with OVA in BALB/c strain mice enhanced specific IgG1 antibody without impacting on specific IgE responses (Larsen et al., 2001b).

Using a somewhat different protocol, other investigators have shown that ip administration of either DEHP or DINP (40 to 125 mg) to BALB/c strain mice enhanced the total serum concentration of IgE, and also increased levels of specific IgE antibody provoked by subcutaneous injection of the immunogen KLH (Lee et al., 2004). Di-iso-nonyl phthalate proved more effective than did DEHP for the augmentation of IgE anti-KLH antibody responses. In our own laboratory, and using a similar protocol, we also found that ip injection of DINP, but not of DEHP, was associated with an increase in IgG1 anti-KLH antibody production (Dearman et al., 2009).

In contrast, the general experience has been that exposure of mice to phthalates via routes that more closely reflect conditions of human contact (that is either oral or topical exposure), is without impact on antibody responses, or is considerably less effective, at mediating adjuvant-like effects.  Oral exposure of BALB/c strain mice to 50 mg per dose DEHP or DINP was without influence on the total serum concentration of IgE, or on levels of IgG1 anti-KLH antibody induced by concurrent subcutaneous administration of the antigen (Dearman et al., 2009). The doses of phthalates used in those experiments, and that were without effect on antibody production or IgE levels, did, however, display significant systemic effects, measured as a function of increases in liver weight (Dearman et al., 2009). Topical exposure of B6C3F1 mice to high doses (up to 100 mg) of DEHP, BBP, DINP or di-iso-hexyl phthalate (DIHP) also failed to alter the total serum concentration of IgE antibody, despite being associated with significant elevations in liver weight (Butala et al., 2004). Using a similarly high dose (50 mg) of DEHP delivered topically (with concurrent subcutaneous injection of the reference allergen OVA) we did not see any impact on the development of anti-OVA IgG1 or IgE antibody responses. However, in the same series of experiments a high dose of BBP (100 mg) significantly enhanced IgG1 anti-OVA antibody responses in the absence of any effect on IgE antibody production (Dearman et al., 2008a; 2008b). In interpreting these data it is important to appreciate that topical administration of chemicals to mice in the absence of occlusion will inevitably result in significant oral exposure as the result of grooming.

There is also some information available about the impact of inhaled phthalates on immune function in mice. Inhalation exposure to DEHP (up to 13 mg/m3) together with OVA had similar effects to that of simultaneous subcutaneous or ip administration of the two materials (Larsen et al., 2007). That is, increased IgG1 anti-OVA antibody levels were found in the absence of effects on IgE antibody production. Airway exposure to MEHP (the major metabolite of DEHP), had apparently identical effects to those seen with DEHP, but at considerably lower doses than those required for the parent compound (0.03 mg/m3) (Hansen et al., 2007). The authors calculated that, with respect to extrapolation to human airways, there was a margin-of-safety of approximately 50 (Larsen et al., 2007 ). They therefore drew the conclusion that “realistic” DEHP exposure levels likely to be encountered in the environment would not be expected to cause adjuvant effects in humans, or to result in allergic inflammation of the lung (Larsen et al., 2007).

Taken together, these experimental studies reveal that the route of administration of phthalate is of considerable importance, and appears to have a decisive influence on whether immunomodulatory effects are induced. Comparatively low doses of phthalates affect antibody parameters when administered only by routes of exposure (subcutaneous or ip) that do not reflect, and are not generally relevant for, opportunities for human contact with phthalates . Although these are among the routes are favoured for experimental immunisation (van Zijverden and Granum, 2000) and, in the case of subcutaneous administration, for the purposes of vaccination, neither are routes by which individuals would normally be expected to encounter phthalates in the domestic or occupational environment. 

This interpretation is consistent with recent investigations in which potential routes of exposure to the most commonly used phthalates have been estimated employing scenario-based approaches (Wormuth et al., 2006). It was found that with respect to DEHP and DIBP, infants and toddlers receive the majority of their exposure orally through the ingestion of both dust and food, while in older children and in adults exposure is primarily through the diet. Similarly for DINP and DIDP, it is estimated that exposure of children results primarily from oral contact and ingestion (mouthing as well as ingestion of dust). For adults, exposure to DIDP is also largely by the oral route (from food and to a lesser extent dust), with dermal and inhalation pathways making smaller contributions. In contrast, in the case of DINP, it is the skin and inhalation pathways that make the largest contributions to adult exposure patterns, with lower levels encountered via oral ingestion. For BBP, oral ingestion is the most important route of exposure for children, with inhalation exposure assuming greater important among adults (Wormuth et al., 2006). The data summarised above from studies in mice indicate clearly that the oral exposure to phthalates is substantially less effective at causing adjuvant-like effects than are either subcutaneous or ip exposures. 

In reconciling effects seen in experimental models with potential hazards to human health it is relevant to consider the nature of adjuvant effects observed with phthalates in mice. Particularly in more recent studies, evaluations of effects on antibody production have been confined only to consideration of IgG1 responses. As discussed previously, although IgG1 antibody production in the mouse is often used as a surrogate for IgE antibody responses, IgG1 is considerably less effective at inducing murine mast cell degranulation than is IgE, and in humans the class of antibody that mediates allergic and anaphylactic reactions is (almost exclusively) IgE (Lehrer and Vaughan 1976; Ovary, 1982; Oshiba et al., 1996). With respect to human health a strong case can be made, therefore, that an induced elevation in mice of IgG1 antibody production, in the absence of effects on IgE antibody responses, is of lesser concern than would be a significant elevation in IgE levels .

One other aspect of the impact of phthalates on immune responses requires consideration. As summarised above, it is clear that under some circumstances, and at some doses, certain phthalates were reported to display what were described as “immunosuppressive” effects (inhibition of IgG1 and/or IgE antibody production), (Larsen et al., 2001a). This inhibitory activity was not exclusively a consequence of generalised systemic adverse effects as depression of antibody responses was not observed exclusively at maximal doses. One can, of course, speculate in general terms about variable dose-related effects of chemicals on the orchestration of immune responses, and in particular on immunoregulatory elements. It is nevertheless difficult to develop a plausible and coherent hypothesis to explain why in some circumstances certain phthalates appear to inhibit antibody production at lower doses, while augmenting the same responses at higher doses. It might be argued that at different dose levels such phthalates are acting selectively on distinct cellular vectors of immune function resulting in unexpected dose-related influences on immune responses. While it is not possible presently to discount such dose-selective effects, an alternative, and to our minds more plausible, explanation is that the readout of antibody production that has been used for evaluation of the potential impact on phthalates on the immune system is subject to natural fluctuations and perturbations that might not necessarily be treatment-related. If this is indeed the case then there is clearly need for some caution in defining the putative immunomodulatory properties of phthalates.

Effects of phthalates on other immune and inflammatory parameters in vivo

Although the majority of studies investigating the potential of phthalates to impact on immune function have focused on consideration of induced changes in antibody responses, less commonly, other readouts have been used. One such is the activity of discrete functional sub-populations of T lymphocytes that collectively dictate the quality and vigour of adaptive immune responses. A seminal observation, made over two decades ago, is that the induction and maintenance of IgE antibody responses is controlled by functional subpopulations of T helper (Th) cells. It quickly became clear that the successful maturation of activated B lymphocytes into IgE producing cells, and a sustained IgE antibody response, requires the preferential development of the Th2 sub-population of Th cells. These cells produce an array of cytokines including interleukins 4, 5 and 13 (IL-4, IL-5 and IL-13) that promote IgE antibody production and facilitate IgE-mediated allergic inflammation (Finkelman et al., 1988a; 1988b; Mosmann et al., 1991). The other major functional sub-population of Th cells is designated Th1 cells. These cells display a different pattern of cytokine production, including, importantly, interferon-γ (IFN-γ) that antagonises the genesis of IgE antibody responses. Moreover, Th1 cells and their cytokine products serve to constrain and inhibit immediate-type allergic responses effected by IgE antibody (Finkelman et al., 1988a; 1988b; Mosmann et al., 1991).

Against this background it is legitimate therefore to consider whether phthalates have any potential to impact on Th cell function, and in particular to influence or perturb the balance between Th1 and Th2 cells and their soluble cytokine products. Although relatively few studies of phthalates have sought to examine Th1- and Th2-type responses as a function of differential cytokine expression, there is available some information.

Exposure of BALB/c strain mice to DEHP or to DINP (40 to 125 mg) by ip injection was reported to increase production of the Th2-type cytokine IL-4 by KLH-primed lymph node cells (LNC) (Lee et al., 2004). Some evidence for increased levels of IL-4 production by draining LNC associated with exposure to phthalates was also described by Maruyama et al (2007). They sensitised CD-1 strain mice topically to the experimental contact allergen fluorescein isothiocyanate (FITC). The FITC was formulated either in a 1:1 mixture of acetone and various low molecular weight phthalate esters (DEP, DPP and DBP), or with acetone alone. Compared with mice that had been exposed to FITC in the absence of phthalates, those that had received phthalates topically were reported to produce increased amounts of IL-4 by draining LNC (Maruyama et al., 2007). The impact of phthalates on cytokine expression has also been investigated following inhalation exposure. In one series of experiments BALB/c strain mice were exposed by inhalation to OVA with or without DEHP. In this instance mice that received DEHP displayed not only elevated levels of IL-5 (a type 2 cytokine), but also increased amount of the Th1 product IFN-γ (Larsen et al., 2007).  

Finally, the impact of phthalates has been considered also following oral administration. Dietary exposure of rats to 12,000 ppm of DEHP was found to alter the balance of Th1/Th2 cytokines in the liver towards a preferential Th2 type phenotype. In the context of these experiments, however, the phthalate-induced Th2 response had an ameliorating effect on the tissue damaging Th1 response that was being provoked by the intraperitoneal administration of Mycobacterium bovis purified protein derivative (PPD) (Badr et al., 2007).

It is important to acknowledge, however, that in other studies exposure of rodents to relatively high doses of DEHP and various other phthalates either by topical treatment, or via gavage, have failed to impact on cytokine expression (Butala et al., 2004; Piepenbrink et al., 2005). In the study by Piepenbrink et al (2005), other immune parameters (lymphoid organ weights, thymus histology, antibody levels) were also found to be unaffected by oral administration of DEHP. In contrast, other authors have demonstrated that dietary exposure of C57BL/6 strain mice to relatively low levels of DEHP was immunosuppressive, resulting in a marked atrophy of both thymus and spleen (Yang et al., 2000).

In other investigations attention has focused instead on examination of whether exposure to phthalates is able to influence inflammatory processes. There are several studies which suggest that exposure to DEHP can enhance ongoing inflammatory responses. In a rat model of OVA-induced allergic asthma, gavage exposure to DEHP was associated with enhanced inflammatory cell infiltration in OVA-challenged airways (Yang et al., 2008). Similar general conclusions were drawn from the use of another rodent model, in this case of atopic dermatitis. Mice of NC/Nga strain develop signs and symptoms of atopic dermatitis. Using these animals it was found that ip exposure to DEHP exacerbated skin lesions induced by intradermal injection of house dust mite allergen. These changes were associated with increased expression of cutaneous chemokines such as macrophage inflammatory protein-1a (MIP-1a) (Takano et al., 2006). Furthermore, the same group reported that maternal ip exposure to DEHP during the neonatal (but not the fetal) period aggravated skin lesions in mite allergen-sensitised male offspring, but not in female offspring (Yanagisawa et al., 2008). The biological significance of these observations are as yet uncertain.

It can be concluded, therefore, that investigations of the effects of phthalates upon various immune and inflammatory responses in experimental animals have yielded conflicting, and somewhat variable, results ranging from potentiation of immune or inflammatory responses, to the absence of any effect, to inhibitory or immunosuppressive activity. Although the available evidence suggests that certain phthalates, when delivered at appropriate doses, and via an appropriate route, can impact on immune and inflammatory function in rodents, it is clear that as yet no consistent pattern has emerged. There is clearly a case to be made for the design of more definitive animal studies that will allow development of a more detailed understanding of whether and to what extent, and under what conditions, certain phthalates are able to effect meaningful changes in immune function. In advance of that and access to more detailed information it is appropriate to question whether there are any other relevant data available.

                   

    

Effects on the elicitation of clinical symptoms in asthmatics

There has been a single study in which the influence of inhalation exposure to phthalates on immune responses in humans has been examined (Deutschle et al., 2008). Patients allergic to house dust mite and healthy (non-allergic) controls were challenged (nose-only) with atmospheres containing low (0.41 mg/g) or high (2.09 mg/g) levels of DEHP. Clinical symptoms were scored, nasal secretions monitored for cytokine and chemokine expression, and a transcriptional analysis was performed on nasal biopsy samples. Nasal exposure to the dust was without effect on clinical symptoms for either group. House dust mite allergic patients, but not nonallergic individuals, responded to challenge with the DEHPlow dust with up-regulation of a number of markers of inflammation including IL-6 and granulocyte-colony stimulating factor (G-CSF). Exposure to the DEHPhigh dust attenuated the response, down-regulating levels of inflammatory cytokines. These are interesting observations, but as the authors acknowledge, a short term exposure protocol (3 hrs) was used and it is therefore difficult to extrapolate from these data to normal environmental exposure patterns. Moreover, as the subjects were not challenged with house dust mite allergen alone, it was not possible in this study to distinguish between the effects due to low dose DEHP, and those due to house dust mite allergen alone.

In vitro effects

Finally, attempts have also been made to characterise in vitro the ability of various phthalates to impact on immune function. There are various investigations in which the impact of phthalates on primary immune cells, or immune cell lines, has been characterised in culture. In common with other aspects of toxicology, it is particularly important when using in vitro cell culture systems to demonstrate that any observed effects are not due to non-specific toxicity or other artefacts, and that biological significance is ascertained before extrapolation to definition of human health risks. In some in vivo experiments, inhibitory effects have been recorded, although in others immunostimulatory effects have been seen.  There is some evidence that in vitro exposure of allergen-primed LNC to DEHP, or to DINP, was able to enhance the expression of IL-4, but was without effect on the type 1 cytokine IFN-g (Lee et al., 2004). Similar phthalate-induced effects on IL-4 expression by a murine thymoma line (EL4) were reported by the same authors (Lee et al., 2004). In contrast, we failed to see any impact of phthalates on cytokine expression using a very similar experimental system (Dearman et al., 2009). In a rather different approach the impact of culture of the human monocytic cell line THP-1, or peripheral blood mononuclear cells (MNC) from allergic and non-allergic donors, with 6 different monophthalates on cytokine mRNA expression levels was examined (Glue et al., 2002). There was no effect of the monophthalates on cytokine expression by the THP-1 cells, or on MNC derived from non-allergic individuals, although increased expression of IL-4 in MNC from allergic individuals was reported for MNBP only (Glue et al., 2002). There are other reports that certain monophthalates promote the production by epithelial cell lines of inflammatory cytokines such as IL-6 and IL-8 (Jepsen et al., 2004). In these experiments, high doses of phthalate resulted in inhibition of cytokine production. Some phthalates, particularly DEHP and MEHP, have been shown to augment histamine release by human basophils isolated from peripheral blood MNC; results interpreted as suggesting that phthalates may have an adjuvant effect on the elicitation of allergic responses (Glue et al., 2005). The physiological significance of this finding is, however, called into question by the observation that concentrations of DEHP reflective of likely human exposure levels failed to provoke clinical symptoms in allergic individuals (Deustchle et al., 2008).

A variety of other authors have demonstrated that culture of inflammatory cells such as macrophages with phthalates has various inhibitory effects, such as decreasing nitric oxide or tumour necrosis factor-a production, consistent with potentially immunosuppressive effects (Hong et al., 2004; Ohnishi et al., 2008). There is also literature that suggests that phthalates such as MEHP induce apoptosis in B cells (the cells that are responsible for the production of immunoglobulin), indicative of down-regulation of antibody responses (Schlezinger et al., 2004; 2007; Bissonnette et al., 2008). The general biological relevance of these observations is uncertain, as is their relationship, if any, to the immunomodulatory properties of phthalates.