Ammonium formate

Endpoint summary

Description of key information

Skin Irritation: Test according to OECD guideline 439, 15 min incubation with Ammonium formate, 42h post-incubation, determination of cell viability for Ammonium Formate: 100% as compared to control

Eye irritation: Test according to OECD guideline 438, ICE, 30 mg applied to cornea, 10 sec incubation with 240 min post-incubation of three chicken corneas, overall ICE classes: twice III (based on the corneal swelling of 19% within 240 minutes and fluorescein retention of 1.7), once IV (based on the corneal opacity score of 2.7).

Test according to OECD guideline 492, Epiocular with reconstructed human keratinocytes, 50 mg applied to tissue, 6h incubation with 42h post-incubation, determination of viability (MTT assay): 2% as compaered with the negative control

Key value for chemical safety assessment

Skin irritation / corrosion

Link to relevant study records

Reference

Endpoint:

skin irritation: in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2019-06-19 to 2019-12-18 (provisional draft final report)

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 439 (In Vitro Skin Irritation: Reconstructed Human Epidermis Test Method)

Version / remarks:

2015

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Test system:

human skin model

Cell type:

non-transformed keratinocytes

Cell source:

other: not specified

Justification for test system used:

The EPISKIN model has been validated for irritation testing in an international trial. After a review of scientific reports and peer reviewed publications on the EPISKIN method, it showed evidence of being a reliable and relevant stand-alone test for predicting rabbit skin irritation, when the endpoint is evaluated by MTT reduction and for being used as a replacement for the Draize Skin Irritation test (OECD TG 404 and Method B.4 of Annex V to Directive 67/548/EEC) for the purposes of distinguishing between skin irritating and
no- skin irritating test substances (STATEMENT OF VALIDITY OF IN-VITRO TESTS FOR SKIN IRRITATION; ECVAM; Institute for Health & Consumer Protection; Joint Research Centre; European Commission; Ispra; 27 April 2007).

Vehicle:

unchanged (no vehicle)

Details on test system:

RECONSTRUCTED HUMAN EPIDERMIS (RHE) TISSUE
- Model used: EpiSkinTM Small Model (EpiSkinTMSM), manufactured by EPISKIN SNC Lyon, France
- Tissue batch number(s):
Supplier:SKINETHIC Laboratories , 4, rue Alexander Fleming, 69366 Lyon Cedex 07 - France
Batch No.:19-EKIN-025
Expiry date: 24 June 2019

TEMPERATURE USED FOR TEST SYSTEM
- Temperature used during treatment / exposure: room temperature/ 37°C
- Temperature of post-treatment incubation (if applicable): 37°C

REMOVAL OF TEST MATERIAL AND CONTROLS
-Volume and number of washing steps: approximately 25 mL PBS 1x solution to remove all of the test material from the epidermal surface
- Observable damage in the tissue due to washing: No

MTT DYE USED TO MEASURE TISSUE VIABILITY AFTER TREATMENT / EXPOSURE
- MTT concentration: 3 mg/mL in saline buffer
- Incubation time: 3h
- Spectrophotometer: Thermo Scientific; Multiscan FC
- Wavelength: 570 ± 10 nm
- Linear OD range of spectrophotometer: 0.2136 – 3.1752

FUNCTIONAL MODEL CONDITIONS WITH REFERENCE TO HISTORICAL DATA
- Viability: Determined by SDS concentration in a MTT test, results should be between 1.5 mg/mL ≤ IC50 ≥ 3.0 mg/mL, results were: 2.1 mg/mL
- Barrier function: Number of cell layers > 4; 7 cell layers were detected
- Morphology: Multi-layered, highly differentiated epidermis consisting of organized basal, spinous and grnular layers, and a multilayered stratum corneum (7 cell layers)
- Contamination: On blood of donors, the absence of HIV-1 and HIV-2 antibodies, hepatitis C antibodies and hepatitis B antigen HBs was verified. On cells from donors, the absence of bacteriafungus and mycoplasma were verified.

NUMBER OF REPLICATE TISSUES: 3 for the treatment cells

CONTROL TISSUES USED IN CASE OF MTT DIRECT INTERFERENCE
- N. of replicates : 3
- Method of calculation used: Negative control:
OD Negative Control (ODNC) = ODNCraw – ODblank mean
–The corrected mean OD of the 3 negative control values is calculated: this corresponds to 100% viability
Positive control
OD Positive Control (ODPC) = ODPCraw – ODblank mean
– The corrected mean OD of the 3 positive control values is calculated
– The % viability for each positive control replicate is calculated relative to the mean negative control:
% Positive Control 1 = (ODPC1 / mean ODNC) ×100
% Positive Control 2 = (ODPC2 / mean ODNC) ×100
% Positive Control 3 = (ODPC3 / mean ODNC) ×100
– The mean value of the 3 individual viability % for positive control is calculated:
Mean PC % = (%PC1 + %PC2 + %PC3) / 3
OD Treated Tissue (ODTT) = ODTTraw – ODblank mean
– The corrected mean OD of the 3 test item values is calculated
– The % viability for each test item replicate is calculated relative to the mean negative control:
% Treated Tissue 1 = (ODTT1 / mean ODNC) ×100
% Treated Tissue 2 = (ODTT2 / mean ODNC) ×100
% Treated Tissue 3 = (ODTT3 / mean ODNC) ×100
– The mean value of the 3 individual viability % for test item is calculated
Mean TT % = (%TT1 + %TT2 + %TT3) / 3



PREDICTION MODEL / DECISION CRITERIA (choose relevant statement)
- The test substance is considered to be corrosive to skin if the viability after 3 minutes exposure is less than 50%, or if the viability after 3 minutes exposure is greater than or equal to 50 %.
- The test substance is considered to be non-corrosive to skin if the viability after 3 minutes exposure is greater than or equal to 50%.
- Justification for the selection of the cut-off point(s) if different than recommended in TG 431 and 439

Control samples:

yes, concurrent negative control

yes, concurrent positive control

yes, concurrent MTT non-specific colour control

Amount/concentration applied:

TEST MATERIAL
- Amount(s) applied (volume or weight with unit): 10 mg

NEGATIVE CONTROL
- Amount(s) applied (volume or weight): 10 µL
- Concentration (if solution): 1 x PBS

POSITIVE CONTROL
- Amount(s) applied (volume or weight): 10µL
- Concentration (if solution):5% SDS (aq.)

Duration of treatment / exposure:

15 min ± 0.5 min

Duration of post-treatment incubation (if applicable):

42h ± 1h

Number of replicates:

Triplicates

Irritation / corrosion parameter:

% tissue viability

Run / experiment:

mean of three replicates

Value:

100

Vehicle controls validity:

not examined

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

no indication of irritation

Remarks:

determination of viability in the negative control

Irritation / corrosion parameter:

% tissue viability

Run / experiment:

mean of three replicates

Value:

21

Vehicle controls validity:

not examined

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

positive indication of irritation

Remarks:

Determination of viability in the positive control

Irritation / corrosion parameter:

% tissue viability

Run / experiment:

mean of three replicates

Value:

100

Vehicle controls validity:

not examined

Negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

no indication of irritation

Remarks:

Determination of the viability in the treated cells

Other effects / acceptance of results:

- OTHER EFFECTS:
- Visible damage on test system: Not reported
- Direct-MTT reduction: No
- Colour interference with MTT: No

ACCEPTANCE OF RESULTS:
- Acceptance criteria met for negative control: Yes, the mean OD value of the three negative control tissues should be between 0.6 and 1.5 and the standard deviation value (SD) of the % viability should be ≤ 18.
- Acceptance criteria met for positive control: Yes, the acceptable mean percentage viability range for positive controls is 0-40% and the standard deviation value (SD) of the % viability should be ≤ 18.
- Acceptance criteria met for variability between replicate measurements: For test chemicals, the standard deviation value (SD) of the % viability should be ≤ 18.
- Range of historical values if different from the ones specified in the test guideline: No

Interpretation of test results
According to the United Nations Globally Harmonized System (UN GHS) of Classification and Labelling of Chemicals (7th revised edition; 2017) and as implemented in the European Commission Regulation (EC) No 1272/2008 on classification, labelling and packaging of substances and mixtures (EU CLP), the irritancy potential of test substances is predicted for distinguishing between irritant or corrosive (Category 2 or Category 1) and non-irritant (No Category) substances.
In the present study, the irritancy potential of test substances is predicted by mean tissue viability of tissues exposed to the test substance. The test chemical is identified as requiring classification and labelling according to UN GHS (Category 2 or Category 1), if the mean relative viability after 15 minutes exposure and 42 hours post incubation is less or equal (≤) to 50% of the negative control. However, this test method (OECD 439) cannot resolve between UN GHS Categories 1 and 2, further information on skin corrosion (OECD 431) will be required to decide on its final classification. In case the test chemical is found to be non-corrosive, and shows tissue viability after exposure and post-treatment incubation is less than or equal (≤) to 50 %, the test chemical is considered to be irritant to skin in accordance with UN GHS Category 2.
Depending on the regulatory framework in member countries, the test chemical may be considered as non-irritant to skin in accordance with UN GHS No Category if the tissue viability after exposure and post-treatment incubation is more than (>) 50 %.
The prediction model (PM) is described below:

Criteria for In Vitro
interpretation Classification
Mean tissue viability % is ≤ 50 % Category 2 or Category 1
Mean tissue viability % is > 50 % No Category

OD values and viability percentages of the controls and test item:

Substance	Optical Density (OD)		Viability (%)
Negative Control: 1x PBS	1	1.002	105
	2	0.899	94
	3	0.952	100
	mean	0.951	100
	standard deviation (SD)		5.46
Positive Control: SDS (5 % aq.)	1	0.238	25
	2	0.197	21
	3	0.178	19
	mean	0.204	21
	standard deviation (SD)		3.20
Test Item: Ammonium Formate	1	0.966	102
	2	0.890	94
	3	1.002	105
	mean	0.953	100
	standard deviation (SD)		6.05

Interpretation of results:

GHS criteria not met

Conclusions:

The results obtained from this in vitro skin irritation test, using the EPISKIN model, indicated that the test item reveals no skin irritation potential under the utilised testing conditions. The test item Ammonium Formate is considered to be non-irritant to skin and is therefore not classified (UN GHS No Category).

Executive summary:

In a dermal irritation study performed in accordance with OECD Guideline 439 (In Vitro Skin Irritation) (adopted July 28, 2015), Ammonium formate was applied to the three-dimensional human epidermis model tissue for an exposure period of 15 minutes in triplicates. 5μL of deionised water were topically applied to the epidermal surface in order to improve further contact between the solid and the epidermis. Each approximately 10 mg of the test item were applied to the wetted tissues. The test item was spread to match the surface of the tissue. After 15 minutes exposure at room temperature, the tissues were washed with phosphate buffered saline to remove any residual test material. Subsequently the tissue constructs were incubated for 42 h at 37°C. Cytotoxicity (irritancy) was expressed as the reduction of mitochondrial dehydrogenase activity measured by formazan production from MTT at the end of the treatment.

The positive (5% SDS) and negative (deionised water) control gave responses that were within the acceptance criteria and as such demonstrated the validity of the study.

The relative mean tissue viability obtained after 15 minutes treatment with Ammonium Formate compared to the negative control tissues was 100%.

Since the mean relative tissue viability for the test substance was above 50%, Ammonium Formate is identified to be not irritating.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (not irritating)

Eye irritation

Link to relevant study records

Reference

Reference 1

Endpoint:

eye irritation: in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2019-06-27 to 2019-12-18 (provisional draft final report)

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 438 (Isolated Chicken Eye Test Method for Identifying i) Chemicals Inducing Serious Eye Damage and ii) Chemicals Not Requiring Classification for Eye Irritation or Serious Eye Damage)

Version / remarks:

2018

Deviations:

no

Qualifier:

according to guideline

Guideline:

EU method B.48 (Isolated chicken eye test method for identifying occular corrosives and severe irritants)

Version / remarks:

2017

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Species:

chicken

Strain:

other: ROSS 308

Details on test animals or tissues and environmental conditions:

SOURCE OF COLLECTED EYES
- Source: TARAVIS KFT. 9600 Sárvár, Rábasömjéni út 129. Hungary
- Storage, temperature and transport conditions of ocular tissue (e.g. transport time, transport media and temperature, and other conditions): Head collection was performed by a slaughter house technician. Heads were removed immediately after sedation of the chickens (sedation was performed by electric current). The heads were transported to the testing facility at the earliest convenience for use approximately within 2 hours from collection. The ambient temperature was (19.3 ºC to 20.4 ºC) during the transport which is considered optimal. All eyes used in the assay were from the same groups of eyes collected on one specific day.
After collection, the heads were inspected for appropriate quality and wrapped with paper moistened with saline, then placed in a plastic box that can be closed (4-5 heads/box).
- Time interval prior to initiating testing: minimum 3h
- indication of any existing defects or lesions in ocular tissue samples: No, after removing the head from the plastic box, it was put on soft paper. The eyelids were carefully cut away with scissors, avoiding damaging the cornea. One small drop of fluorescein solution 2 (w/v) % was applied onto the cornea surface for a few seconds and subsequently rinsed off with 20 mL isotonic saline. Then the fluorescein-treated cornea was examined with hand-held slit lamp or slit lamp microscope, with the eye in the head, to ensure that the cornea was not damaged. If the cornea was in good condition, i.e. not obvious damage on the head or eye, the eyeball was carefully removed from the orbit.
- Indication of any antibiotics used: No

Vehicle:

unchanged (no vehicle)

Controls:

yes, concurrent positive control

yes, concurrent negative control

Amount / concentration applied:

TEST MATERIAL
- Amount(s) applied (volume or weight with unit): 0.03 g

Duration of treatment / exposure:

10 sec

Duration of post- treatment incubation (in vitro):

240 min

Number of animals or in vitro replicates:

3 eyes for treatment, positive and negative control

Details on study design:

SELECTION AND PREPARATION OF ISOLATED EYES
After removing the head from the plastic box, it was put on soft paper. The eyelids were carefully cut away with scissors, avoiding damaging the cornea. One small drop of fluorescein solution 2 (w/v) % was applied onto the cornea surface for a few seconds and subsequently rinsed off with 20 mL isotonic saline. Then, and prior to excision from the head, the fluorescein-treated cornea was examined with a hand-held slit lamp or a slit lamp microscope to ensure that it was not damaged (i.e., fluorescein retention and corneal opacity scores ≤ 0.5). If the cornea was in good condition, the eyeball was carefully removed from the orbit.

The eyeballs were carefully removed from the orbit by holding the nictitating membrane with surgical forceps, while cutting the eye muscles with bent scissors without cutting off the optical nerve too short, i.e. could not be held with surgical forceps. The procedure avoided pressure on the eyes in order to prevent distortion of the cornea and subsequent corneal opacity. Once removed from the orbit, the eyes were placed onto paper moisted with saline and the nictitating membrane and connective tissue were cut away. The prepared eyes were kept on moisted paper in a closed box to maintain humidity and to prevent the eyes from becoming dry.

EQUILIBRATION AND BASELINE RECORDINGS
The prepared eyes were placed in steel clamps with the cornea positioned vertically with the eyes in the correct relative position (same position as in the chicken head), again avoiding too much pressure on the eye by the clamp. Because of the relatively firm sclera of the chicken eyeball, only slight pressure was applied to fix the eyes properly. The clamp with the eyeball was transferred to a chamber of the superfusion apparatus. The clamp holding the eye was positioned in such a way that the entire cornea was supplied with saline solution dripping from a stainless steel tube, at a rate of approximately 3 to 4 drops/minutes. The door of the chamber was kept closed except for manipulations and examinations, to maintain temperature and humidity.
The appropriate numbers of eyes were selected and after being placed in the superfusion apparatus, the selected eyes were examined again with the slit lamp microscope to assess their condition. For this purpose, the focus was adjusted to see clearly the isotonic saline which was flowing on the cornea surface.
Eyes with a high baseline fluorescein staining (i.e., > 0.5) or corneal opacity score (i.e., > 0.5) were rejected and replaced.
The cornea thickness was measured using the depth measuring device on the slit lamp microscope (Haag-Streit BQ 900) with the slit-width set at 9½, equaling 0.095 mm. Any eye with cornea thickness deviating more than 10 % from the mean value of all eyes, or eyes that showed any other signs of damage, were rejected and replaced.
For the eyes selected for the test because they satisfied the suitability criteria, acclimatization started and was maintained for approximately 45 to 60 minutes. The temperature was verified to be in the range of 32 ± 1.5 °C in all chambers during the acclimatization and treatment periods.

NUMBER OF REPLICATES
3
NEGATIVE CONTROL USED
physiological saline (NaCl 0.9%)

POSITIVE CONTROL USED
Imidazole (0.03g)

APPLICATION DOSE AND EXPOSURE TIME
0.03 g of the test item for 10 sec

OBSERVATION PERIOD
240 min

REMOVAL OF TEST SUBSTANCE
- Volume and washing procedure after exposure period: The time of application was monitored and, after an exposure period of 10 seconds from the end of the application, the cornea surface was rinsed thoroughly with 20 mL saline solution at ambient temperature, while taking care not to damage the cornea but attempting to remove all the residual test item if possible. The eye in the holder was then returned to its chamber. The time while the eye was out of the chamber was limited to the minimum.
- Indicate any deviation from test procedure in the Guideline : No differences reported

METHODS FOR MEASURED ENDPOINTS:
The control and test eyes were evaluated pre-treatment and at approximately 30, 75, 120, 180 and 240 minutes after the post-treatment rinse. Minor variations within ± 5 minutes were considered acceptable. The cornea thickness and cornea opacity were measured at all time points. Fluorescein retention was determined at baseline (t=0) and 30 minutes after the post-treatment rinse.
- Corneal opacity:
Corneal opacity was measured according to the following formulae:
ΔCO at time t = CO at time t – CO at t=0
Mean CO(at time t) = FEΔCO (at time t)+ SEΔCO(at time t) + TEΔCO(at time t)/3
with
CO = Cornea opacity
ΔCO = Difference between cornea opacity and cornea opacity reference value
FEΔCO = Difference between first eye cornea opacity and first eye cornea opacity reference value
SEΔCO = Difference between second eye cornea opacity and second eye cornea opacity reference value
TEΔCO = Difference between third eye cornea opacity and third eye cornea opacity reference value
Mean CO = The mean corneal opacity value
at time t = Observation time at 30, 75, 120, 180 or 240 minutes after the post-treatment rinse
at t=0 = Reference value

- Damage to epithelium based on fluorescein retention:
Fluorescein retention was calculated according to the following formulae:
ΔFR at time t = FR at time t – FR at t=0

Mean FR = FEΔFR (at time t) + SEΔFR(at time t) + TEΔFR at time t)/3
with:
FR = Fluorescein retention
ΔFR = Difference between fluorescein retention and fluorescein retention reference value
FEΔFR = Difference between first eye fluorescein retention and first eye fluorescein retention reference value
SEΔFR = Difference between second eye fluorescein retention and second eye fluorescein retention reference value
TEΔFR = Difference between third eye fluorescein retention and third eye fluorescein retention reference value
Mean FR = The mean fluorescein retention value
at time t = Observation time at 30 minutes after the post-treatment rinse
at t=0 = Reference value
- Swelling: measured with optical pachymeter on a slit-lamp microscope; slit-width setting:
Cornea swelling was calculated according to the following formulae
CS at time t = CT at time t – CT at t=0 X100/CT at t=0
Mean CS at time t = FECS(at time t)+ SECS (at time t) + TECS (at time t)/3
with
CS = Cornea swelling
CT = Cornea thickness
FECS = First eye cornea swelling
SECS = Second eye cornea swelling
TECS = Third eye cornea swelling
Mean CS = The mean percentage of corneal swelling
at time t = Observation time at 30, 75, 120, 180 or 240 minutes after the post-treatment rinse
at t=0 = Reference value
- Others (e.g, histopathology): After consultation with the Sponsor no histopathology evaluation was performed. Corneas are discarded 2 months after the final report.

DECISION CRITERIA: Decision criteria from OECD Guideline 438 was used.

Irritation parameter:

percent corneal swelling

Remarks:

Mean maximum corneal swelling at up to 75 min

Run / experiment:

Mean of three treated eyes

Value:

17

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

3% (Mean maximum corneal swelling at up to 75 min)

Positive controls validity:

valid

Remarks:

30% (Mean maximum corneal swelling at up to 75 min)

Irritation parameter:

percent corneal swelling

Remarks:

Mean maximum corneal swelling at up to 240 min

Run / experiment:

Mean of three treated eyes

Value:

19

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

3% (Mean maximum corneal swelling at up to 240 min)

Positive controls validity:

valid

Remarks:

37% (Mean maximum corneal swelling at up to 240 min)

Irritation parameter:

cornea opacity score

Run / experiment:

Mean of three treated eyes

Value:

2.7

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

0.5 (Mean maximum corneal opacity)

Positive controls validity:

valid

Remarks:

4.0 (Mean maximum corneal opacity)

Irritation parameter:

fluorescein retention score

Remarks:

Mean fluorescein retention

Run / experiment:

Mean of three treated eyes

Value:

1.7

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

0.0 (Mean fluorescein retention)

Positive controls validity:

valid

Remarks:

2.8 (Mean fluorescein retention)

Irritation parameter:

morphological effects

Run / experiment:

In all three replicates

Vehicle controls validity:

not examined

Negative controls validity:

valid

Positive controls validity:

valid

Remarks:

Corneal opacity score 4 was observed in three eyes at 30 minutes after the post-treatment rinse.

Remarks on result:

other: no effects observed in the eyes treated with the test item

Other effects / acceptance of results:

OTHER EFFECTS:
- Visible damage on test system: Corneal opacity score 4 was observed in three eyes at 30 minutes after the post-treatment rinse in the3 positive control.

DEMONSTRATION OF TECHNICAL PROFICIENCY: Prior to routine use of the method the test facility demonstrated the technical proficiency in a separate study using the ten Proficiency Chemicals according to OECD Test Guideline No. 438.

ACCEPTANCE OF RESULTS:
- Acceptance criteria met for negative control: Yes
- Acceptance criteria met for positive control: Yes
According to OECD guideline 438 A test is considered acceptable if the concurrent negative or vehicle/solvent control and the concurrent positive control are identified as GHS Non-Classified and GHS Category 1, respectively.
- Range of historical values if different from the ones specified in the test guideline: please refer to any other information on results incl. tables for the historical data range

Interpretation of results:

study cannot be used for classification

Conclusions:

In this ICET, Ammonium Formate did not cause ocular corrosion or severe irritation in the enucleated chicken eyes. The overall ICE classes were twice III (based on the corneal swelling of 19% within 240 minutes and fluorescein retention of 1.7) and once IV (based on the corneal opacity score of 2.7).
Positive and negative controls showed the expected results. The experiment was considered to be valid.
According to the guideline OECD 438, Ammonium Formate's overall in vitro classification is neither UN GHS Classification Category I (an ocular corrosive or severe eye irritant) nor No Category. Thus, according to the guideline OECD 438, test item has been categorized as “No prediction can be made”.

Executive summary:

This in vitro study was performed to assess the corneal irritation and damage potential of Ammonium Formate by means of the ICE assay using fresh chicken corneas according to OECD guideline 438, adopted 25 June 2018.

The corneas were incubated with the test substance and controls for 10 sec. After rinsing with saline, the corneas were incubated for another 2 h. The test was performed in triplicates. Corneal opacity and swelling as well as fluorescein retention were determined.The overall ICE classes were once II (based on the fluorescein retention of 1.3) and twice III (based on the corneal swelling of 13% within 75 minutes and corneal opacity score of 1.7). 

The positive control was classified as corrosive/severely irritating, UN GHS Classification: Category 1 and the negative control had no significant effects on the chicken eye in this study.

Since the overall ICE classes of the test substance were neither specified for Category 1 nor for Non-Classification, Ammonium Formate could not be classified asUN GHS Classification Category I (an ocular corrosive or severe eye irritant) nor No Categoryunder the experimental conditions described in this report.On the basis of the results obtained in this experiment, the test item is categorized as “No prediction can be made”.

Reference 2

Endpoint:

eye irritation: in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2019-10-15 to 2019-12-19 (provisional draft final report)

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 492 (Reconstructed Human Cornea-like Epithelium (RhCE) Test Method for Identifying Chemicals Not Requiring Classification and Labelling for Eye Irritation or Serious Eye Damage)

Version / remarks:

18 June 2019

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Species:

human

Details on test animals or tissues and environmental conditions:

- Justification of the test method and considerations regarding applicability
The eye irritation potential of a chemical may be predicted by measuring its cytotoxic effect, as reflected in the MTT assay, on the EpiOcular™ (OCL-200-EIT) tissue. The EpiOcular™ model can be used to identify chemicals that do not require classification for eye irritation or serious eye damage according to the UN GHS classification system. This test method can be used to partly replace the in vivo Rabbit eye test (OECD 405).
The EpiOcular™ Eye Irritation Test (EIT) was validated by the European Union Reference Laboratory for Alternatives to Animal Testing (EURL ECVAM) and Cosmetics Europe between 2008 and 2013. From this validation study and its independent peer review it was concluded that the EpiOcular™ EIT is able to correctly identify chemicals (both substances and mixtures) not requiring classification and labelling for eye irritation or serious eye damage according to UN GHS, and the test method was recommended as scientifically valid for that purpose. The EpiOcular™ EIT is recommended to identify chemicals that do not require classification for eye irritation or serious eye damage according to UN GHS (UN GHS No Category) without further testing, however, a drawback of this test is the inability to distinguish between Category 1 (corrosive to eye) and Category 2 (eye irritant). Thus, in case of a positive result from the EpiOcular™ EIT further testing is required.
- Description of the cell system used, incl. certificate of authenticity and the mycoplasma status of the cell live
After the test kit arrival, the tissues were equilibrated to room temperature for about
15 minutes. The Assay Medium was pre-warmed to 37±1 °C. The appropriate number of an assay plate wells (6-well plates) was filled with the pre-warmed medium (1 mL per well). The insert was transferred aseptically into the 6-well plates and pre-incubated at 37±2°C in an incubator with 5±1% CO2, in a ≥95% humidified atmosphere for one hour in the Assay Medium. After one hour, the Assay Medium was replaced by 1 mL of fresh Assay Medium at 37°C and the EpiOcular™ tissues was incubated at standard culture conditions overnight, i.e. for 16 - 24 hours.


The EpiOcular™ (OCL-200-EIT) kits are manufactured according to defined quality assurance procedures. All biological components of the EpiOcular™ tissue and the kit culture medium have been tested for the presence of viruses, bacteria and mycoplasma. The quality of the final product is assessed by undertaking an MTT cell viability test and a cytotoxicity test with Triton X-100 (100 μL of 0.3% (v/v) Triton X-100).
The kit contained Reconstructed ocular tissue model containing normal human kerationcytes.Lot Number: 30630; Keratinocyte strain: 4F1188
The tissue was free of HIV-1, Hepatitis B virus, Hepatitis C virus, Bacteria, yeast, and other fungi.
Tissue viability, barrier function and sterility were tested as follows:
Tissue viability: MTT QC assay, 1h, n = 3; Acceptance criteria: OD (540-570 nm)[1.1-3.0]; Result: 2.176 ± 0.055
Barrier function: ET-50 assay, 100µL 0.3% Triton X-100, 3 time points, n = 2, MTT assay; Acceptance criteria: ET-50 [12.2-37.5 min]; Result: 21.89 min
Sterility: Long term antibiotic and antimycotic free culture; Acceptance criteria: No contamination; Result: Sterile

Vehicle:

unchanged (no vehicle)

Controls:

yes, concurrent positive control

yes, concurrent negative control

Amount / concentration applied:

TEST MATERIAL
- Amount(s) applied (volume or weight with unit): The test item was applied in its original form (approximately 50 mg on top of 0.6 cm² tissue; approx. 83.3 mg/cm²), no formulation was required.

Duration of treatment / exposure:

6h ± 15 min

Duration of post- treatment incubation (in vitro):

18h ± 15 min

Number of animals or in vitro replicates:

Each treatement and the controls were measured in duplicate.

Details on study design:

- Details of the test procedure used
The EpiOcular™ reconstructed ocular tissue model containing normal human keratinocytes (MatTek Corporation) is used in this assay. This three-dimensional model allows the identification of test items with the potential to induce eye irritation or serious eye damage by assessing cell viability after treatment.
The model is composed of stratified human keratinocytes in a three-dimensional structure, consisting of at least three viable layers of cells. Test items can be applied topically to the model so that also water insoluble materials may be tested.
Prior to use, each plate (6, 12, and 24-well) and its cover will be uniquely identified with a permanent marker by a plate number and/or test item number.
The cytotoxicity of the test item and, thus, the ocular irritation potential is evaluated by the relative viability of the tissues treated with test item in comparison to the negative control tissues. Viability is determined by the NAD(P)H-dependent microsomal enzyme reduction of MTT and, to a lesser extent, by the succinate dehydrogenase reduction of MTT, in control tissues and tissues treated with test item (Berridge, et al., 1996). Data for relative MTT conversion are presented in the form of relative survival rates.

- RhCE tissue construct used, including batch number
Reconstructed ocular tissue model containing normal human kerationcytes.Lot Number: 30630; Keratinocyte strain: 4F1188

- Doses of test chemical and control substances used
approximately 50 mg on top of 0.6 cm² tissue; approx. 83.3 mg/cm²

- Duration and temperature of exposure, post-exposure immersion and post-exposure incubation periods (where applicable)
Exposure 6h ± 15 min at 37°C, post-exposure immersion: 25min ± 2 min at room temperature, 18h ± 15 min at 37°C

- Number of tissue replicates used per test chemical and controls (positive control, negative control) Duplicates

- Wavelength and band pass (if applicable) used for quantifying MTT formazan, and linearity range of measuring device (e.g. spectrophotometer)
The OD (Absorbance / Optical Density) of the samples in a 96-well plate spectrophotometer (Thermo Scientific; Multiscan FC) at the wavelength of 570 nm ((±10nm; Read out range: 0-3.5 Abs, Linearity range: 0.2908 – 2.6589) using isopropanol solutions as the blank (8×200 µL).

- Description of the method used to quantify MTT formazan
Formazan Extraction
Inserts were removed from the 24-well plate after 3 hours ± 15 minutes; the bottom of the insert was blotted on absorbent material, and then transferred to a pre-labeled 6-well plate containing 2 mL isopropanol in each well so that no isopropanol was flowing into the insert. The plate was sealed with parafilm.
To extract the MTT, the plates were placed on an orbital plate shaker and agitated at 120 rpm for approximately 2 hours at room temperature. At the end of the extraction period, the tissue was not pierced. The corresponding negative, positive, and colorant controls were treated identically.

- Description of evaluation criteria used including the justification for the selection of the cut-off point for the prediction model
Data Calculation for Normal Test Items

The calculations were performed using Microsoft Excel software.
Blank:
– The mean of the 8 blank OD values was calculated
Negative control:
– Individual negative control OD values were corrected with the mean blank OD:
OD Negative Control (ODNC1) = ODNCraw1 – ODblank mean
OD Negative Control (ODNC2) = ODNCraw2 – ODblank mean
Mean OD Negative Control (mean ODNC) = [(ODNC1) + (ODNC2)] / 2
– The corrected mean OD of the 2 negative control values was calculated: this corresponds to 100% viability.
Positive control:
– Individual positive control OD values were corrected with the mean blank OD:
OD Positive Control (ODPC) = ODPCraw – ODblank mean
– The corrected mean OD of the 2 positive control values will be calculated
– The % viability for each positive control replicate will be calculated relative to the mean negative control:
% Positive Control 1 = (ODPC1 / mean ODNC) ×100
% Positive Control 2 = (ODPC2 / mean ODNC) ×100
– The mean value of the 2 individual viability % for positive control was calculated:
Mean PC % = (%PC1 + %PC2) / 2
Test item:
– Individual test item OD values were corrected with the mean blank OD:
OD Treated Tissue (ODTT) = ODTTraw – ODblank mean
– The corrected mean OD of the 2 test item values will be calculated
– The % viability for each test item replicate wase calculated relative to the mean negative control:
% Treated Tissue 1 = (ODTT1 / mean ODNC) ×100
% Treated Tissue 2 = (ODTT2 / mean ODNC) ×100
– The mean value of the 2 individual viability % for test item was calculated
Mean TT % = (%TT1 + %TT2) / 2
Assay Acceptance Criteria

The mean OD value of the two negative control tissues should be between 0.8 and 2.8.
The acceptable percentage viability for positive control (mean of two tissues) is:
- 30 minute exposure: below 50% of control viability (liquids)
- 6 hours exposure: below 50% of control viability (solids)
The difference of viability between the two relating tissues of a single chemical is
< 20% in the same run (for positive and negative control tissues and tissues of single chemicals). This applies also to the killed controls (single chemicals and negative killed control) and the colorant controls which are calculated as percent values related to the viability of the relating negative control.

- Reference to historical positive and negative control results demonstrating suitable run acceptance criteria
Please refer to 'Any other information on materilas and methods incl. tables'


- Positive and negative control means and acceptance ranges based on historical data
- Acceptable variability between tissue replicates for positive and negative controls
The difference of viability between the two relating tissues of a single chemical is < 20% in the same run (for positive and negative control tissues and tissues of single chemicals). This applies also to the killed controls (single chemicals and negative killed control) and the colorant controls which are calculated as percent values related to the viability of the relating negative control. Yes: Please refer to 'Any other information on materilas and methods incl. tables'

- Acceptable variability between tissue replicates for the test chemical; es: Please refer to 'Any other information on materilas and methods incl. tables'

Irritation parameter:

other: viability

Remarks:

%

Run / experiment:

#1

Value:

2

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

97% viability

Positive controls validity:

valid

Remarks:

3% viability

Remarks on result:

positive indication of irritation

Irritation parameter:

other: viability

Remarks:

%

Run / experiment:

#2

Value:

2

Vehicle controls validity:

not examined

Negative controls validity:

valid

Remarks:

103% viability

Positive controls validity:

valid

Remarks:

6% viability

Remarks on result:

positive indication of irritation

Other effects / acceptance of results:

OTHER EFFECTS:
- Visible damage on test system: Not reported

ACCEPTANCE OF RESULTS:
- Acceptance criteria met for positive control:
Yes: The acceptable percentage viability for positive control (mean of two tissues) is for - 6 hours exposure: below 50% of control viability (solids) mean viability 3%

OD values and viability percentages of the test item:

Test Item	Optical Density (OD)		Viability (%)	Δ%
Ammonium Formate	1	0.048	2	0.2
	2	0.043	2
	mean	0.045	2

OD values and viability percentages of the controls:

Controls	Optical Density (OD)		Viability (%)	Δ%[MW1]
Negative Control: Sterile deionized water	1	2.467	97	5.6
	2	2.609	103
	mean	2.538	100
Positive Control: Methyl acetate	1	0.088	3	2.1
	2	0.142	6
	mean	0.115	5

 

Interpretation of results:

study cannot be used for classification

Conclusions:

In the present study Ammonium Formate was tested to evaluate the eye irritation potential in the EpiOcular™ model. Human keratinocytes were incubated with 50 mg of the test item for 6h. After a 18h post incubation period the cell viability was determined. The test item reduced the cell viability significantly as compared to the negative control (viability after post-incubation period: 2%). Thus, Ammonium Formate indicated to be Irritant (UN GHS Category 2) and/or Corrosive (UN GHS Category 1) according to the Globally Harmonized System for Classification and Labelling of Chemicals. However, this test method (OECD 492) cannot resolve between UN GHS Categories 1 and 2.

Executive summary:

In this study conducted according to OECD test guideline 492 (adopted July 28, 2015), the potential of the test item to induce eye irritation was analysed by using the three-dimensional human corneal epithelium model EpiOcular, consisting of normal, human-derived epidermal keratinocytes mimicking characteristics of the corneal epithelium.

The test item was applied topically to the EpiOcular tissue for 6 h followed by 25 min post-soaking incubation after removal of the test item. After an 18 h post-treatment period cytotoxic effects were determined via MTT reduction assay.

Ocular irritation potential of the test item was predicted from the relative mean tissue viabilities compared to the negative control tissues concurrently treated with Aqua dest.

The test item showed no non-specific reduction of MTT and no relevant colouring after mixture with isopropanol or with Aqua dest.

The test item showed irritant effects. The mean relative tissue viability (% negative control) was ≤ 60% (1.8).

The controls confirmed the validity of the study. The mean absolute OD570 of the negative control tissues was > 0.8 and < 2.8 (2.538). The mean relative tissue viability (% negative control) of the positive control was < 50% (4.5%). The maximum inter tissue difference of replicate tissues of all dose groups was < 20% (4.9%).

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (irritating)

Justification for classification or non-classification

In accordance with Regulation (EC) No. 1907/2006 (REACH) Annex VII section 8.2 and 8.2.1 two in vitro tests were conducted to identify Ammonium formate as serious eye irritant/corrosive or for no classification. The first in vitro test used was an organotypic in vitro test, the ICE assay (Isolated Chicken Eye, OECD guideline 438, 2018). The endpoints evaluated are corneal opacity, swelling, fluorescein retention, and morphological effects (e.g., pitting or loosening of the epithelium). The scores for determination of each parameter observed are:

Corneal opacity:

Score	Observation
0	No opacity
0.5	Very faint opacity
1	Scattered or diffuse areas; details of the iris are clearly visible
2	Easily discernible translucent area; details of the iris are slightly obscured
3	Severe corneal opacity; no specific details of the iris are visible; size of the pupil is barely discernible
4	Complete corneal opacity; iris invisible

Corneal swelling:

For the calculation of Maximum Swelling, small negative numbers for swelling (0 to -5%) following application are counted as zero. Large negative numbers (> -12 %) are probably due to erosion and indicate a severe effect (scored as class IV). Cases of values of -5 % to -12 % are evaluated on a case by case basis but in the absence of other findings do not indicate a severe effect.

Fluorescein retention:

Score	Observation
0	No fluorescein retention
0.5	Very minor single cell staining
1	Single cell staining scattered throughout the treated area of the cornea
2	Focal or confluent dense single cell staining
3	Confluent large areas of the cornea retaining fluorescein

The results for Ammonium formate from this assay are:

Observation	Value	ICE Class^1
Mean maximum corneal swelling at up to 75 min	17%	III
Mean maximum corneal swelling at up to 240 min	19%	III
Mean maximum corneal opacity	2.7	IV
Mean fluorescein retention	1.7	III
Other Observations	None
Overall ICE Class^1	2xIII, 1xIV

 

Based on the results from this assay, Ammonium formate could neither be classified as eye corrosive (Category I) nor as not classified according to Regulation (EC) No 1272/2008 (CLP) and the Globally Harmonized System for Classification and Labelling of Chemicals (GHS).

For this reason, a second in vitro assay was performed based on the recommendations of REACH Guidance: ‘How to use new or revised in vitro test methods to address Serious eye damage/Eye irritation’, Feb 2018 and according to Regulation (EC) No. 1907/2006 (REACH) Annex VII section 8.2.1.

The second in vitro test used was the EpiOcular™ assay (OECD guideline 492): The eye irritation potential of a chemical may be predicted by measuring its cytotoxic effect, as reflected in the MTT assay, on the EpiOcular™ (OCL-200-EIT) tissue. The EpiOcular™ model can be used to identify chemicals that do not require classification for eye irritation or serious eye damage according to the UN GHS classification system. EpiOcular™ OCL-200, a three-dimensional RhCE tissue model, is produced using primary human epidermal keratinocytes. Relative tissue viability for tissue exposed to the test item is determined against tissues treated with the negative control by the reduction of the vital dye MTT (3-[4,5 - dimethylthiazol-2-yl] - 2,5 - diphenyltetrazolium bromide).

The evaluation criteria for this assay are:

Depending on the regulatory framework in member countries, the test item is identified as not requiring classification and labelling according to UN GHS (No Category) if the mean percent tissue viability after exposure and post-exposure incubation is more than
(>) 60%. In this case no further testing with the help of other test methods is required.

The prediction model (PM) is described below:

Criteria forIn Vitro interpretation	Classification
Mean tissue viability % is ≤ 60 %	Category 2 or Category 1
Mean tissue viability % is > 60 %	No Category

 

The results for Ammonium formate of this assay are:

Treatment with the test item Ammonium formate showed significantly reduced cell viability in comparison to the negative control (mean tissue viability: 1.8 %). All obtained test item viability results were below 60% when compared to the viability values obtained from the negative control.

Based on the results obtained from the second assay conducted with Ammonium formate also revealed no clear classification according to Regulation (EC) No 1272/2008 (CLP) and the Globally Harmonized System for Classification and Labelling of Chemicals (GHS).

In order meet the requirements for registration of Ammonium formate according to Regulation (EC) No. 1907/2006 (REACH) Annex VII and due to the recommendations outlined in REACH Guidance: ‘How to use new or revised in vitro test methods to address Serious eye damage/Eye irritation’, Feb 2018 stating that an ‘in vivo eye irritation study shall only be considered at Annex VIII level’ and . ‘at Annex VIII, if neither of these conclusions can be made, other in vitro study(ies) for this endpoint shall be considered. If the in vitro studies are not suitable for the substance, or the results are not adequate for classification and risk assessment, a further test conducted in vivo to assess the eye irritation potential shall be considered, i.e. none of the in vitro methods described above can be used for the direct identification of eye irritants (Cat. 2 of CLP)’. Furthermore. it is recommended that in case of inconclusive results an assessment of all available information, e.g. from QSAR(s) or skin corrosion studies should take place in order to adequately classify the substance.

Other available information:

In the ‘Guidance Document on an Integrated Approach on Testing and Assessment (IATA) for Serious Eye’, 2017 an additional test is mentioned to conclude that a substance does not need to be classified as Category I (Causes serious eye damage), the Fluorescein Leakage Test (OECD guideline 460). Based on the previously conducted tests and their results and due to the false-positive results which were reported in the Guidance document, this test was not considered to classification purposes. In vivo testing is not considered in terms of animal welfare and because according to REACH and the respective guidances it is not foreseen for registrations in the tonnage band of 1-10 t/a (Annex VII).

For the test item Ammonium formate a skin irritation in vitro study is available conducted according to OECD guideline 439 (Reconstructed Human Epidermis Test Method). Based on the test results Ammonium formate was not classified as skin irritant according to Regulation (EC) No 1272/2008 (CLP) and the Globally Harmonized System for Classification and Labelling of Chemicals (GHS). Like the EpiOcular Test method the RhE Test is based on the measurement of cell viability after treatment with the test item. After a 15min treatment with the test item and subsequently post-incubation time of 42h the cell viability was (mean value) 100% of the cell viability of the control, thus Ammonium formate was not classified as skin irritant.

The Performance of a QSAR method is not recommended for ionic structures because the estimations are based on neutral organics equations using the minimum toxicity principle assuming a simple non-polar narcosis model. This might underestimate the actual toxicity of the test item if used without comparison to empirical data from structurally similar substances.

Since Ammonium formate is a salt constituted of formic acid and Ammonium ions via a ionic bond and it is expected that the substance dissociates when it is dissolved in aqueous solutions, any test conducted in medium, aqueous solutions or water is considered to be exposed only the free ions, thus, it is admissible to use data generated from formic acid acid and other ammonium salts for the evaluation of the eye irritation potential.

Based on the publicly available information formic acid exhibits a very low pH, it is known to be corrosive to metals, therefore, testing of formic acid is not justified in terms of animal welfare. However, ammonium formate is used as food additive (JEFCA[1]) and an evaluation by EFSA[2]about its safety when used as a technological additive for all animal species stated that ‘Formic acid and its salts dissociate at physiological pH to formate, an important constituent of intermediary one-carbon metabolism. The safety assessment of formate is thus independent of whether the anion derives from the free acid or its salts. However, in the case of the salts, any impact of the accompanying cation is also taken into consideration’. Furthermore, it was reported that ‘Formic acid is corrosive and ammonium formate/formic acid and sodium formate/formic acid are assumed to have similar properties. The pure calcium and sodium salts of formic acid are not corrosive. Calcium formate was examined for skin and eye irritancy following OECD protocols (OECD Guidelines 404 and 405). It was found to be non-irritating to skin but irritating to the eye. Sodium formate was similarly examined, but using different protocols. The results essentially paralleled those seen for the calcium salt.’

Similarly to Calcium formate ammonium formate was not irritating to skin in the current in vitro test.

Data for eye irritation for ammonium ions are lacking. However, ammonium ions are basically generated from ammonia following the equilibrium:

NH₃+ H₃O⁺ NH₄⁺+ H₂O

The fraction of ammonia in this equilibrium increases with increasing pH and temperature. Under normal physiological conditions at pH 7.4 and 25°C the predominant chemical species is ammonium.

Ammonia is an integral part of the intermediate metabolism and is endogenously formed by every organism capable of nitrogen fixation. Ammonia is an important source of nitrogen for mammals and plants due to its use in the synthesis of amino acids, DNA, RNA and proteins. It is produced endogenously in all mammalian species. Bacterial degradation of nucleic acids and amino acids in the gut produces ammonia in amounts of 3-4 g/day (ATSDR, 2004[3], EFSA, 2011[4]). Most of this ammonia is transported to the liver for detoxification where it is used in the synthesis of urea and glutamine. This process is an important part of the regulation of systemic pH due to the neutralisation of excess bicarbonate, created from the breakdown of amino acids, by hydrogen ions released from ammonium during the synthesis of urea (Häussinger, 2007[5]). Thus, ammonium is ubiquitously present in organisms and the environment.

Data on eye irritation for Ammonium ions are not available, however, there are some information on the eye irritation potential for ammonium salts like ammonium sulfate or ammonium chloride.

Ammonium chloride was found to be severely irritating to eyes and therefore was classified in Category II ‘Causes serious eye irritation’ Although it was reported that in studies with rabbits according to OECD 405 the effects caused by Ammonium chloride were fully reversible within 8 days and thus ammonium chloride was considered to be moderately irritant to skin and eyes (OECD SIDS, 2003[6]).

Ammonium sulfate was also evaluated by the OECD[7]and it was reported that also in studies conducted with rabbits according to OECD guideline 405, ammonium sulfate only caused slight redness of the eyes which was also found if the rabbits were treated the same way with talcum powder. However, the effects seen were fully reversible within 8 days. Based on these observations it was considered that ammonium sulfate is not irritating to skin and eyes. 

As a consequence, ammonium ions per se are not considered to be irritating in fact the counter ion might be responsible for the irritating effect.

Based on the information presented here, it is assumed that ammonium formate is not corrosive to the eye. The test item Ammonium formate showed in two in vitro tests performed inconclusive results, i.e. in two in vitro test which are validated to distinguish between Category I (Causes serious eye damage/ corrosive) and no classification no prediction could be made. Thus, it was concluded that based on the irritating properties which were shown for the formic acid salts and for some of the ammonium salts to a similar extent, the irritancy potential of Ammonium formate is lowered in comparison to the irritancy potential of formic acid as shown for the other formic acid salts. Since Ammonium salts are not corrosive it is expected that they are reasonably for the reduced irritancy potential presumably due to a certain buffer capacity.

Moreover, since the substance Ammonium formate is currently voluntary classified as Category II’ Causes serious eye irritation’ this classification will be retained.

 

[1]TRS 884-JECFA 49/29:   COMPENDIUM ADDENDUM 8/FNP 52 Add.8/130 (FLAVOUR) (2000); FAS 40-JEC FA 49/147 (FLAVOUR)

[2]EFSA Journal 2015;13(5):4056; Scientific Opinion on the safety and efficacy of ammonium formate, calcium formate and sodium formate when used as a technological additive for all animal species.

[3]ATSDR (Agency for Toxic Substances and Disease Registry), 2004. Toxicological profile for ammonia. U.S. Department of Health and Human Services, Atlanta, Georgia, 269 pp.

[4]EFSA Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids, 2011. Scientific Opinion on Flavouring Group Evaluation 46 Revision 1 (FGE.46Rev1). Ammonia and three ammonium salts from chemical group 30. EFSA Journal 2011; 9(2):1925, 35 pp.

[5]Häussinger D, 2007. Ammonia, urea production and pH regulation. In: The Textbook of Hepatology:

from basic science to clinical practice, 3rd Edition. Eds Rodes J, Benhamou J-P, Blei A, Reichen J and Rizzetto M. Wiley-Blackwell, 181-192. 

[6]OECD SIDS, AMMONIUM CHLORIDE CAS N°: 12125-02-9, 2003

[7]OECD SIDA, AMMONIUM SULFATE CAS N° 7783-20-2, 2004