Sodium [[α,α'-(ethylenediimino)bis[2-hydroxybenzene-1-acetato]](4-)]ferrate(1-)

Endpoint summary

Link to relevant study record(s)

Reference

Endpoint:

basic toxicokinetics, other

Remarks:

expert statement based on physico-chemical properties and toxicological data of structurally related substances.

Type of information:

other: expert statement

Adequacy of study:

key study

Study period:

2017

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: expert statement based on physico-chemical properties and toxicological data of structurally related substances.

Objective of study:

absorption

distribution

excretion

metabolism

Qualifier:

according to guideline

Guideline:

other: ECHA guidance on Toxicokinetics (Chapter R. 7C, section R.7.12, 2014)

Deviations:

no

GLP compliance:

no

Type:

absorption

Results:

A moderate absorption potential is considered for oral route of exposure. A limited absorption potential is predicted for dermal and inhalation routes of exposure.

Type:

distribution

Results:

If absorbed, the substance is expected to be distributed into intravasal compartment.

Type:

metabolism

Results:

No extensive metabolism is expected for the chelating agents and polycondensation products.

Type:

excretion

Results:

The target substance is expected to be excreted via the urine or via the bile.

Metabolites identified:

not measured

Conclusions:

O,o-Fe(Na)EDDHA has low absorption potential via all exposure routes. If absorbed, it is expected to be distributed predominantly in intravasal compartment. The substance is not expected to be metabolised and will be excreted mainly unchanged via the faeces or via the urine.

Executive summary:

O,o-Fe(Na)EDDHA is expected to be moderately absorbed after oral exposure, based on its high water solubility and its insolubility in n-octanol. Concerning the absorption after exposure via inhalation, as the chemical is considered to have a low vapour pressure, is highly hydrophilic, insoluble in n-octanol, and has a rather high molecular weight, it is clear, that the substance is poorly available for inhalation and will not be absorbed significantly. The substance is also not expected to be absorbed following dermal exposure into the stratum corneum and into the epidermis, due to its molecular weight and its high water solubility. Concerning its distribution in the body the substance is expected to be distributed mainly in the intravasal compartment, due to its high water solubility. The substance does not indicate a significant potential for accumulation. O,o-Fe(Na)EDDHA is expected not to be significantly metabolised but to be eliminated unchanged via urine and bile.

Description of key information

O,o-Fe(Na)EDDHA is expected to be moderately absorbed after oral exposure, based on its high water solubility and its insolubility in n-octanol. Concerning the absorption after exposure via inhalation, as the chemical is considered to have a low vapour pressure, is highly hydrophilic, insoluble in n-octanol, and has a rather high molecular weight, it is clear, that the substance is poorly available for inhalation and will not be absorbed significantly. The substance is also not expected to be absorbed following dermal exposure into the stratum corneum and into the epidermis, due to its molecular weight and its high water solubility. Concerning its distribution in the body the substance is expected to be distributed mainly in the intravasal compartment, due to its high water solubility. The substance does not indicate a significant potential for accumulation. O,o-Fe(Na)EDDHA is expected not to be significantly metabolised but to be eliminated unchanged via urine and bile.

Key value for chemical safety assessment

Bioaccumulation potential:

no bioaccumulation potential

Absorption rate - oral (%):

100

Absorption rate - dermal (%):

10

Absorption rate - inhalation (%):

100

Additional information

General

There are no ADME studies available for o,o-Fe(Na)EDDHA.The toxicokinetic profile of the test substance was not determined by actual absorption, distribution, metabolism or excretion measurements. Rather, the physical chemical properties of this substance were integrated with the toxicological data on structural analogues to create a prediction of toxicokinetic behaviour.

Toxicological profile of o,o-Fe(Na)EDDHA

There are no studies available for toxicological endpoints for o,o-Fe(Na)EDDHA and therefore read-across from its nearest structural analogue substances UVCB Fe(Na)EDDHA (CAS 84539-55-9) and methylated analogue FeEDDHMANa (CAS 84539-53-7) was performed instead. Additionally, the data on supporting source substance (Fe(3Na)EDDHSA (CAS 84539-54-8) and Fe(3K)EDDHSA (EC 462-490-6) (please refer to the IUCLID files) have been taken into account to predict toxicological behaviour in living organisms. The target substance and the source substances show very similar physical/chemical properties (high water solubility, low Pow, no hydrolysis in water, low vapour pressure) and are thus believed to behave very similar in aqueous solutions (please refer to read-across statement attached to the IUCLID file under section 13; the references cited in this section are listed in the read-across statement).

The source substances are of low acute toxicity via oral and dermal routes of exposure. All animals survived the applications of test substances in good health and LD50 greater than 2000 mg/kg bw were reported in these studies. The substances are not skin or eye irritants and do not fulfil requirements to be classified for these endpoints. No skin sensitisation has been reported for Fe(Na)EDDHMA and for Fe(3K)EDDHSA but Fe(Na)EDDHA possesses a weak sensitising potential. The chelates are not mutagenic in various bacterial and mammalian test systems.

The source substances had similar toxicity effects in the repeated dose and reproductive/developmental toxicity studies (Please see detailed description below).

Toxicokinetic analysis of o,o-Fe(Na)EDDHA

The substance o,o-Fe(Na)EDDHA is at 20 °C is an odourless, dark red solid in a microgranulated form (MW of 435.17 g/mol). The substance is soluble in water (20 g/L at 25°C) and has a negative partition coefficient (could not be determined as the substance does not dissolve in n-octanol). It has a very low vapour pressure and has no melting point up to 600 °C under atmospheric conditions. Its boiling point and vapour pressure were not determined, as the melting point is above 300 °C. Hydrolysis as a function of pH does not apply as the substance forms extremely stable complexes.

 

Absorption

Oral absorption

 

Since oral route is the relevant route of exposure, stability of o,o-Fe(Na)EDDHA chelate complexes in the gastrointestinal (GI) tract has a substantial influence on its absorption behaviour and fate into systemic circulation. In this respect, there are uncertainties whether the intact (chelated) o,o-Fe(Na)EDDHA is well absorbed in small intestines with the subsequent distribution to organs and elimination, or whether the complex is dissociating in the GI tract before absorption and then the separate components (iron and chelator) are absorbed. Another possibility is, that the complex o,o-Fe(Na)EDDHA is absorbed as the chelator-iron complex and then is de-complexed in the liver, contributing to the iron load in the liver while still being able to bind iron from serum or from the GI tract following release from the liver. An attempt to clarify these questions will be undertaken in the following sections.

 

Affinity to iron (chelate’ stability):

Several latter publications report data on behaviour of EDDHA, its derivatives and EDTA ligands in solutions with trivalent metal ions at different pH values. Ma et al. (1994) determined stability constants by potentiometric and spectrophotometric methods. The distribution curves for Fe-EDTA indicate that about 69 % of EDTA are chelated at pH 2. With the increase of pH value, EDTA releases iron. Above pH 5 complex FeEDTA is the predominant form of iron in aqueous solution (Ma et al., 1994). Further, the authors report that iron forms complexes with EDDHA completely even at pH 2 and no free metal present in solutions (Ma et al., 1994). This is also explained by higher stability of Fe-EDDHA complexes. The logK stability constants for Fe(Na)EDDHA and Fe(Na)EDDHMA as mixtures of meso and rac isomers are reported 36.89 and 36.59, respectively. In contrast, the log stability constant for FeEDTA is 25.1 (Yunta et al., 2003b; Hyvönen et al., 2003). Lopez-Rayo et al. (2009) showed also that in solutions EDTA de-complexes at the pH range 6-8 releasing iron while EDDHA remains complexed until pH 12. FeEDDHA is described as a highly stable Fe chelate in a wide pH range (2-11), while EDTA de-complexed at pH range 6-8 (Lopez-Rayo et al., 2009).

 

These results can be relevant for the assessment of the fate of Fe(Na)EDDHA and its derivative complexes at physiological conditions. In the stomach, where low pH values dominate,it seems that, based on the abovementioned data, Fe(Na)EDDHA , if ingested, will not dissociate completely and in the small intestine, where pH raises, is expected to remain in its chelated form.Theoretically, taking only the binding capacity of the ligands to Fe³⁺into account, similar behaviour could apply for EDDHMA, since it has also very high stability constants. Since, no physiological pH values above 11 exist in the gastrointestinal tract, the majority of EDDHA and EDDMA is likely to be in the chelated form before absorption. For comparison with EDTA, administered orally, “iron (primarily Fe3+) remains complexed with EDTA under the acidic conditions prevailing in the stomach. The chelate holds the iron in solution as the pH rises in the upper small intestine, but the strength of the complex is progressively reduced allowing at least partial exchange with other metals and the release of some of the iron for absorption. Further results indicate that iron dissociates from the chelate and is released into the common non-haem iron pool before absorption” (IPCS, 2014; Candela et al., 1984). From studies with EDTA it is also known that absorption of iron (intrinsic and extrinsic) increased at least twice in average in different persons and in animals fed fortified diets with Fe-EDTA comparing to iron from inorganic salts (i.e ferrous sulphate) (Heimbach et al., 2000). The authors concluded that the mechanism by which chelated iron forms a common pool with intrinsic iron differs from the mechanism of absorption for simple iron salts, resulting in greater overall bioavailability. Comparably, it can be assumed that chelated iron from FeEDDDA and FeEDDHMA complexes, will be more available for absorption. However, taking into account the differences in the stability of EDDHA, EDDHMA and Fe-EDTA chelates at pH 6-8 in the gut, it is clear that the absorption mechanism described for Fe-EDTA does not fully apply. While EDTA forms different metal species complexes dependent on their affinity constant, the pH environment and the concentration of competing metals and/or ligands in the gastrointestinal tract (Heimbach et al., 2000), in contrast, EDDHA and EDDHMA seem to be more stable compounds that are not expected to dissociate and/or compete with other metals in the gut because the affinity to iron is much higher than in EDTA (Lucena, 2012). No substitution of iron by other metals is possible, as determined experimentally (Lopez-Rayo et al., 2009). The organic moiety is also stabilised in the chelate impending the direct interaction with living organism (Lucena, 2012). It could mean that the chelating ligands EDDHA and EDDHMA, being systemically available, had so high affinity to iron leaving other metal levels unaffected.

Based on the data on affinity to iron and to other metals, a preliminary conclusion is that Fe(Na)EDDHA, if ingested, is not expected to fully dissociate and, in small intestines, where pH raises, is expected to remain predominantly in its chelated form, unless any enzymatic or other mechanisms exist to de-chelate the complexes.

 

Prediction of absorption behaviour of chelates based the results of long-term toxicity studies with Fe(Na)EDDHA and Fe(Na)EDDHMA

 

Although little differences in the molecule structure could result in a significant difference of toxicological activity of two substances, effects observed in the repeated studies may be relevant to elucidate their absorption potential.

For Fe(Na)EDDHA and Fe(Na)EDDHMA, the established NOAELs are of the same order of magnitude. In the 28-day oral range finding study (gavage) with the source substance UVCB Fe(Na)EDDHA, dose levels of 50, 200 and 1000 mg/kg bw were administered to Sprague Dawley rats. The treatment resulted in impaired body weight development at 200 and 1000 mg/kg bw/day and accordingly correspondent lower food intake (CIBA-GEIGY, 1996a). An anaemia without erythropoietic response was noted at the mid and at the highest dose. The kidney was identified as the target organ by microscopical examination, by blood chemistry data evaluation and by organ weight evaluation. In addition, body weight relative organ weight changes were noted in the heart, adrenals and spleen. The effects in the following-up 90-day study were similar. Lower food intake and impaired body weight development was noted at 200 mg/kg bw/day (the highest dose tested) (Novartis Crop Protection AG, 1998). Reversible effects on red blood cell (normochromic anaemia) and white blood cell parameters, and higher values of platelets and prothrombin activity were noted at 50 and/or 200 mg/kg bw/day. In addition, changes of blood chemistry and urine parameters concerning the liver and kidneys were noted. The body weight relative heart weight was increased in males at 200 mg/kg bw/day. In the 90-day study, the NOAEL of 10 mg/kg bw was estimated from the LOAEL (please refer to IUCLID file for more detailed information).

NOELs of 40 mg/kg bw and 20 mg/kg bw were established in the 28-day and in the 90-day oral studies in rats treated with Fe(Na)EDDHMA (Banks, 1987, Schoenmakers, 1996), respectively. Decreased body weight and low food consumption was observed in animals in the highest dose level groups in both studies (1000 mg/kg bw and 500 mg/kg bw are the highest doses in 28 and in 90-day studies, respectively). At 500 mg/kg bw, clinical signs were lethargy, hunched posture and piloerection. Red blood cells, haemoglobin and haematocrit were reduced at the mid and at the highest doses in both studies. Increased creatinine with increased relative kidney weight as well as nephrosis and cortical tubular cell vacuolation point to an adaptive response of kidney were also noted (please refer to section 4.4.5 or to IUCLID file for more details).

Based on the toxicity findings, it is clear that Fe(Na)EDDHA and Fe(Na)EDDHMA are well absorbed into systemic circulation. Based on the effects observed in the studies, it is reasonable to assume that anaemia effects are due to the elimination of considerable amounts of iron. It could mean that there are mechanisms, which are probably involved into de-complexation of chelates before absorption, similarly to those described for EDTA (WHO, 2008). As the result of this de-complexation, chelators free of iron have a certain absorption potential, which is, likely, high enoughto sequester again considerable amounts of systemically available iron leading to anaemia effects. The anaemia symptoms (reduced red blood cells, haemoglobin and haematocrit) and findings in kidneys suggest that systemically absorbed chelates (probably free chelators free of iron) compete with the internal pool of free iron (i.e bound to transferrin), complex this iron and are either excreted in the urine or mobilize and redistribute iron to the kidneys, similarly to the pathway reported for EDTA (Zhu et al. 2006, cited in WHO, 2008). Another explanation for the anaemia effects observed in the studies with Fe(Na)EDDHA and Fe(Na)EDDHMA could be a condition similar to that observed in thalassemia patients. The patients have suppressed erythropoiesis leading to also reduced blood cell levels, low haemoglobin and haematocrit values together with hepatic iron overload as the result of frequent blood transfusion or high absorption of dietary iron (Bergeron et al., 1999; Bergeron and Raymond, 2003; Herschko, 2010). Such a condition could be mimicked by high amounts of iron absorbed in form of Fe(Na)EDDHA and Fe(Na)EDDHMA chelates. In case if the Fe(Na)EDDHA and Fe(Na)EDDHMA compete with transferrin, excess of iron will be transported to the liver, while released EDDHA and EDDHMA chelators would bind further iron resulting in anaemia effects. Thus, it could be assumed, that when Fe(Na)EDDHA and Fe(Na)EDDHMA enter systemic circulation in their chelated form, they probably become de-chelated in plasma or in the liver and then they sequester further iron from GI tract, by entering enterohepatic cycle, or bind iron from the liver, competing with ferritin or bind iron in serum, competing with transferrin.

 

A release of iron from the complexes before mucosal absorption cannot completely be ruled out. In this case, it could mean that their chance to enter systemic circulation in the dissociated form as EDDHA and EDDHMA exists too. Then, the absorption of the released iron from the complex would follow normal physiological pathways as responsible for non-haem-iron. The absorption of the chelating ligands EDDHA and EDDHMA would lead to binding of body-own iron resulting in anaemia effects. There is also possibility that the free EDDHA and EDDHMA will hinder iron absorption in the GI tract. In favour of this hypothesis are well-known facts about phytic acid and polyphenolic compounds that often found in plant-based diets. They inhibit iron absorption by forming stable and insoluble aggregates or by chelating iron with high affinity so that iron is not available for absorption by brush border iron transport proteins even if it is soluble, thereby preventing the iron from entering intestinal epithelial cells (Zhu et al., 2006). The released iron from Fe(Na)EDDHA and Fe(Na)EDDHMA can also interact with the absorption mechanisms known for haem iron. Haem iron enters the mucosa via a pathway that is different for non-haem iron and involves interaction of iron in with a haem receptor (EFSA, 2006) the same as iron in a porphyrin complex does. It cannot be ruled out that haem-oxygenase, the enzyme that cleaves the porphyrin and liberates iron for transfer into the body (EFSA, 2006), also interacts with the chelator-iron complexes, releasing and taking up iron into systemic circulation. Then free chelators would prevent absorption of further dietary non-haem iron in mucosa as described by Zhu et al (2006). Such a scenario could explain the effects related to anaemia in animals observed after administration of Fe(Na)EDDHA and Fe(Na)EDDHMA in the repeated dose toxicity studies.

In conclusion,the normochromic anaemia findings in the repeated dose toxicity studies with the target substance UVCB Fe(Na)EDDHA and Fe(Na)EDDHMA point to the sequestration of dietary or systemically available iron. The dissociation of complexes can take place either before the mucosal uptake in the gut or after uptake in the blood or liver.

The amount of the absorbed free of iron chelators depends, however, on further factors like molecule size, molecular weight and the solubility in lipids to pass cell membrane. Therefore, further data on absorption of Fe-EDTA and EDTA (as the best investigated chelate) have been taken into account. Additionally, absorption of free iron and the absorption based solely on physico-chemical properties will be discussed in the next sections to gain deeper inside absorption behaviour of o,o-Fe(Na)EDDHA.

 

Absorption of Fe-EDTA complexes (animal and human studies)

There is data available on absorption of intact (complexed iron EDTA).

After EDTA-FeNa has been ingested, the absorption of iron from ferric sodium EDTA is regulated through the same physiological mechanisms as other forms of iron (Zhu et al., 2006, Yeung et al., 2005). The total amount of iron in the human body is ca. 4 g. In general, 1 mg iron per day will be lost. These losses are replenished via the food intake. A normal diet contains ca. 10 to 15 mg Fe per day. Following oral administration, the iron from ferric sodium EDTA is separated from the iron EDTA complex in the lumen of the gut by the intestinal cells of the duodenum and small intestine. Since iron uptake is directly regulated to the body’s need in humans, the mucosa cells will only pick up the amount of non-complexed ferric ions that is needed by the body, and this amount will be transported to the blood plasma where it will be coupled to transferrin, like all other absorbable iron in food (Candela et al., 1984, Zhu et al., 2006; WHO, 2008). This iron joins the general non-haem iron pool that is finally incorporated into the circulating haemoglobin (Candela et al., 1984). The iron component of ferric sodium EDTA is subsequently handled systemically like any other source of iron; the safety and maximum tolerable intake of which has been reviewed and evaluated by a number of distinguished scientific committees such as JECFA/WHO, UK EVM, SCF, IOM and EFSA. Non-absorbed iron will be excreted via the faeces (Candela et al., 1984).

Only a very small fraction of the sodium iron EDTA complex (less than 1–2%) is absorbed intactand is rapidly and completely excreted via the kidneys in the urine whereby mostiron originating from the ingested EDTA-FeNa is released to the physiological mucosal uptake system before absorption(WHO, 2008). In another literature source, intestinal absorption EDTA-FeNa was estimated to be 5 % (Candela et al., 1984, EU Risk Assessment 2004a).

 

Absorption of EDTA as edetic acid

Oral absorption of EDTA free of iron in described in an EU RAR (2004) on edetic ecid. There is mentioned, that “There are no oral toxicokinetic studies or skin absorption studies with EDTA itself or its tetrasodium salt available. According to the dissociation equilibrium of edetic acid administration of different sodium salts will result in dependence on the intestinal pH-value to the formation of various anionic species of EDTA. It can be assumed that the oral and dermal absorption of sodium salts of EDTA and of the free acid is comparable to the measured low absorption of CaNa2EDTA. It is poorly absorbed from the gastrointestinal tract (a maximum of 5% was detected in the urine).”“Following oral administration of the calcium salt of 14C-EDTA (50 mg/kg bw) to rats the chelate was poorly absorbed from the gastrointestinal tract (2 to 18% within 24 hours)”(Foreman et al., 1953).

 

Prediction of absorption of dietary iron in the presence of Fe(Na)EDDHA by oral route of exposure:

As described above, the affinity to iron of the chelating ligand EDDHA and EDDHMA are high that no or little dissociation of the complexes could be expected in comparison to EDTA at the same pH conditions. However, in case, if some quantities of iron are released from the chelator-iron complexes under very acidic conditions of stomach, absorption of iron is expected to underlie normal physiological pathways responsible for iron uptake, although the released free chelators EDDHA and EDDHMA can further sequester luminal or mucosal iron lowering the absorption. According to literature data, absorption of nutritional (especially non-haem) iron is very low. Ferric (Fe3+) food iron is precipitated from solution above pH 3.5 (Heimbach et al., 2000). This insoluble precipitate is poorly absorbed in the small intestines by humans, where non-heme iron is absorbed, unless suitable complexing agents are present (Conrad and Schade, 1968; MacPhail et al., 1981, cited in Heimbach et al., 2000).“The absorption of non-haem iron can be increased substantially by the presence of ligands, such as ascorbate, citrate and fumarate, as well as the presence of amino acids (e.g. cysteine) and oligopeptides resulting from meat digestion (Mulvihill et al, 1998). In contrast, very stable complexes, for example with phytates, phosphates and oxalates, impair non-haem iron absorption. Depending on the concentration of supportive or inhibitory ligands in the intestinal lumen the absorption of non-haem iron can vary by a factor of 10 in single-meal studies, but the effects are less pronounced in more long-term studies (Hallberg and Rossander, 1984; Rossander, 1987; Hunt and Roughead 2000”, cited in EFSA, 2006). Based on this data,in case if the complexes dissociate before absorption,it can be assumed that the absorption of the dietary or released iron from Fe(Na)EDDHA and Fe(Na)EDDHMA, will be limited.

In conclusion, in case of absorption, ingested Fe(Na)EDDHA and Fe(Na)EDDHMA, being very soluble molecules, would easily pass through GI epithelium by passive diffusion process. It cannot be ruled out that other transport mechanisms are involved into the absorption process. The amount of intact substances absorbed in form of iron-chelate complex is expected to be higher than in case of EDTA, because they are very stable iron-complexes at the pH values dominating in the gut. Fe-EDDHA and Fe-EDDHMA release iron either before or after absorption but the amount of ligands is sufficient to sequester body own iron leading to anaemia effects.

 

Oral absorption based on physico-chemical properties

The absorption of the originally chelated iron in form of Fe(Na)EDDHA and Fe(Na)EDDHMA complexes or re-chelated iron will follow common pattern of absorption according to the physico-chemical properties of the complex. O,o-Fe(Na)EDDHA has very high water solubility (20 g/L) and the negative LogPow (not determined as it is assumed to be the same as for the source substance UVCB Fe(Na)EDDHA. The source substances UVCB Fe(Na)EDDHA and Fe(Na)EDDHMA are also water-soluble chemicals: 150-203 and 116 g/L with negative logPow, respectively. Thus, their absorption through the GI epithelium into systemic circulation is expected by passive diffusion through aqueous pores or by carriage across membranes with the bulk passage of water. However, absorption of very hydrophilic substances by passive diffusion may be limited by the rate at which the substance partitions out of the gastrointestinal fluid. The rate of this absorption is not possible to predict too. Nevertheless, the molecular weight of 435.17 g/mol of the main constituent o,o-Fe(Na)EDDHA might hinder the absorption because it is near 500 g/mol, designated as the upper level favourable for absorption (ECHA guidance on Toxicokinetics, Chapter R. 7C, section R.7.12, 2014).

The absorption of other positional isomers is considered to be very similar because they have the same coordination geometry and the complexing efficiency with iron even though their stability constants with iron or species formation at different pH values can differ. The absorption of condensation products can be different depending on their molecular weight, ionisation stage, and the number of phenolate groups (the affinity of iron to phenolate groups is the highest). The absorption of minor amounts of benzol, ethylenediamine or glyoxylic acid is considered to be extensive because these are small water soluble molecules.

In conclusion, a moderate absorption potential is considered for oral route. Moreover, the absorption behaviour is expected to be very similar for the target and the source substance UVCB Fe(Na)EDDHA since they have identical components. The absorption of Fe(Na)EDDHMA could be even higher because it is slightly lipophilic chemical passing easier the epithelium of GI tract.  

For the purposes of hazard assessment (DNEL derivation), 100% oral absorption (in a worst case) is considered appropriate since the findings in the long-term toxicity studies are suggestive of absorption from the gastro-intestinal tract.

 

Absorption by inhalation

Concerning the absorption after exposure via inhalation, as EDDHA analogues has really low vapour pressures (<1E-6 hPa and < 1E-5 hPa for UVCB Fe(Na)EDDHA, and Fe(Na)EDDHSA, respectively), the substances are marginally available for inhalation. The substances are manufactured in granular form. In the target substance, only 25 % particles are smaller than 125 µm (Valagro, 2016). In the source substance UVCB Fe(Na)EDDHA only 2.7 % of the particle showed a diameter lower than 100 µm. No particles were found less than 10 µm. Particles are mainly less than 500 µm (95%) were determined for Fe(3Na)EDDHSA (Valagro, 2012a). In contrast, Fe(K)EDDHMA contains particles that are mainly smaller than 20 µm (SEPC 2002c). Thus, it is very unlikely, that big amounts of the UVCB Fe(Na)EDDHA and Fe(Na)EDDHMA reach the lung. Nevertheless, if the substances reached the lung, it is not very likely that they are taken up rapidly. The source substance UVCB Fe(Na)EDDHA showed no toxicity after inhalation administration, in an acute inhalation toxicity study when applied at a dose of 4200 mg/m³. Together, this indicates low systemic availability after inhalation and if bioavailable, no toxicity effects via this route of administration. However, for the purpose of hazard assessment (DNEL derivation), 100 % inhalation absorption is considered appropriate since no substance specific data is available for the registered substance.

Dermal absorption

The substances are not expected to be absorbed significantly into the stratum corneum following dermal exposure, based on their physico-chemical properties: very high water solubility (20 g/L, 150-203 and 116 g/L, for o,o-Fe(Na)EDDHA, UVCB Fe(Na)EDDHA and Fe(Na)EDDHMA, respectively) and the negative LogPow (see table 5). The supporting source substance Fe(3Na)EDDHSA has similar values.

According to ECHA guidance on Toxicokinetics (Chapter R. 7C, section R.7.12, 2014), these substances are not likely to be sufficiently lipophilic to cross the stratum corneum. Very high water solubilities above 10,000 mg/L together with the log P value below 0 further indicate that the substance may be too hydrophilic to cross the lipid rich environment of the stratum corneum. Dermal uptake for these substances will be low. Accordingly, their systemic toxicity via the skin has been proven to be low (no mortality after dermal application of 2000 mg/kg bw to Fe(3K)EDDHSA, Fe(Na)EDDHA and Fe(K)EDDHMA in rats (please refer to section 4.4.2.). In a dermal repeated dose toxicity study (28 day)(CIBA-GEIGY, 1996b) beside local effects, only slight systemic effects on body and adrenal weight were observed at the limit dose of 1000 mg/kg bw/day, supporting the limited bioavailability via this route compared with the toxic effects noted after oral application. NOEL of 100 mg/kg bw is established for dermal route of exposure. Based on these data, 10 % of dermal absorption is considered for o,o-Fe(Na)EDDHA, due to a negative logPow of the nearest analogues ( -4.0 for Fe(3K)EDDHSA and -4.2 for Fe(Na)EDDHA).

 

Distribution and Bioaccumulation

If absorbed, the target and the source substances are not expected to bear an accumulative potential based on its high water solubility and the low logPow value. The EDDHA main constituents, its positional isomers and condensation products are all the constituents that are very soluble in water and, if absorbed, are expected to be distributed to vascular system and will not be distributed into the cells, as the cell membranes require a substance to be soluble also in lipids to be taken up. Based on their very low BCF values the substances are very unlikely to bioaccumulate in the human body. As it is known that “substances with LogPow values of 3 or less would be unlikely to accumulate with the repeated intermittent exposure patterns normally encountered in the workplace” (TGD, Part 1), no enhanced risk for accumulation will be associated with the substance, due to its negative LogPow.

 

Metabolism and excretion

Free of metal EDDHA and EDDHMA moiety as well as their positional isomers are not expected to be metabolised in the body but are rather excreted chelated with iron. No data is available to the registrant on metabolites of the organic moiety of the target and the source substances. The only reason to consider so is due to the unique property of chelating agents chelate metals. Therefore, several synthetic chelates are used in chelation therapy in patients in course of therapy with iron overload. In case if chelates are de-chelated, based on the structure of the molecules and their nature, metabolism in the human body will mainly consist on phase-II metabolising steps, leading to an even better water solubility for excretion. Glucuronidation is one of such pathways leading to a better water solubility for excretion and it is most probable to occur also for EDDHA and its derivatives. This is in compliance with the results obtained in the genotoxic tests with the source substances showing no effects with and without metabolising system. Metabolic activation leading to more toxic metabolites is thus not very likely. In case of formation of glucoronid derivates, there is a possibility of entero-hepatic recycling, and the risk of a re-entering into system, but it does not bear any risk for the organism.

Additionally, extensive experimental data exist on the metabolism of EDTA compounds. Neither the iron nor the EDTA moiety of EDTA-FeNa undergoes biotransformation. Evidence for this conclusion is based on a number of absorption and metabolism studies in animals and humans which indicated that both EDTA and iron are excreted unchanged following ingestion of NaFeEDTA (EU Risk Assessment, 2004a; IPCS, 2014; Heimbach et al., 2000).

Based on the water solubility and the logPow value, excretion via the urine is likely.Urine parameters (acidic, reddish urine) were impaired in the repeated dose toxicity studies with these chelates indicating an extensive urinary excretion of the chelate complexes or their metabolites. Further, due to the high stability constant of the iron chelate complex Fe(EDDHA (logK = 36.89, as mixture of meso and rac isomers) (Yunta et al., 2003b), it is clear that, if chelated, it exerts a low reactivity in the organism. Therefore, it is assumed that most of this very water soluble iron fraction will be excreted unchanged in the chelated form mainly in the faeces and in the urine. As the substances have molecular weights above 300 g/mol the excretion of a considerable amount via the bile is also possible, especially if phase-II conjugation takes place e. g. with formation of glucoronid derivates.

 

Conclusion on ADME

The substances are not expected to be readily absorbed after oral exposure, based on their rather high molecular weight, their very high water solubility and their negative LogPow values. This agrees with the LD50 > 2000 mg/kg bw, determined in rats after oral exposure to the chelates Fe(3K)EDDHSA, Fe(Na)EDDHA, and Fe(K)EDDHMA. In case of absorption through the GI epithelium, it is expected to be by passive diffusion through aqueous pores or by carriage across membranes with the bulk passage of water. In case of long-term exposures, even though the intestinal absorption is not expected to be extended, the amount of the absorbed fraction will be sufficient to produce adverse effects related to anaemia effects. The effects observed in the long-tern studies with these substances are similar in nature and in the amplitude of the strength of effects. Therefore, it can be concluded that the different percentages in the number of positional isomers, the nature of their impurities and differences in their composition have minor influence on the main effect related to chelation ability of dietary or body own iron. The absorption will consist of the intact chelator-iron complexes or released iron ions and chelator moieties. The same behaviour is assumed for the positional isomers and the condensation products of these chelates. Since they have a weaker complexing ability, they will not contribute significantly to the effects produced by the main ortho-ortho constituents. In contrast to Fe-EDTA, Fe-EDDHA and Fe-EDDHMA are not expected to dissociated extensively releasing iron before absorption because the affinity of iron to EDDHA and EDDHMA chelates are much higher than that to EDTA. Fe-EDDHA and Fe-EDDHMA release iron either before or after absorption but the amount of the ligands is sufficient to sequester body own iron leading to anaemia effects.

In addition, if any mechanism exists promoting or supporting the dissociation of the chelates in the GI tract, in plasma or in the liver, the toxicity is expected to be driven by iron deficiency, resulting from sequestering and excretion of iron from liver and blood by EDDHA or its analogues.

Concerning the absorption after exposure via inhalation, as the compounds have really low vapour pressures, it is clear, that they have a low availability for inhalation. The chelates are not expected to be absorbed significantly following dermal exposure into the stratum corneum, due to their rather high molecular weight, negative LogPow values and high water solubility. Accordingly, their systemic toxicity via the skin has been proven to be low (no mortality after dermal application of 2000 mg/kg bw to Fe(3K)EDDHSA, Fe(Na)EDDHA and Fe(K)EDDHMA in rats.

Absorbed chelates into systemic circulation are expected to be distributed throughout the body (mainly into intravascular compartment), although elimination is expected rather than the distribution to organs. It would be rapidly cleared by the kidneys. Thus, the substances are not expected to bear accumulative potential. The chelated complexes or their chelating agents are not expected to be extensively metabolised but to be eliminated mainly via the bile (in cases as glucuronic acid conjugates) or, due to their high water-solubility to a smaller extent, oxidised or unchanged via the urine.Not absorbed in the gut chelated complexes would be excreted unchanged predominantly in the faces.