Registration Dossier

Administrative data

Key value for chemical safety assessment

Additional information

In the key study for bacterial mutagenicity similar to OECD TG 471, no mutagenic effects were noted in Salmonella typhimurium strains (TA 1535, TA 1537, TA 1538, TA 98, TA 100)

in the presence or absence of metabolic activation for retinyl palmitate (Chételat, 1982).

Equivocal results have been observed in an Ames test according to NTP standart protocols, using Salmonella typhimurium strains TA 1535, TA 1537, TA 97, TA 98, TA 100 (NTP 2003). However, no clear evidence of a mutagenic potential for retinyl palmitate can be derived from this study, since no reproducibility between replicate experiments and no preference to any strain or testing conditions has been observed. Furhermore no data on the test substance are available to proove a retinyl palmitate mediated effect.


In a photomutagenicity test, on the basis of the mouse lymphoma assay in L5178Y TK+/- mouse lymphoma cells without metabolic activation, retinyl palmitate was not mutagenic in the absence of UVA radiation (Mei 2005). Cytotoxicity and an increase in the mutant frequency was observed in the presence of UVA and UVA + retinyl palmitate. A dependency between increasing cytotoxicity and mutation frequency cannot be excluded in this study.



The strucurally and metabolically related structures retinol and retinyl acetate have been investigated in a complete set of genetic in vitro toxicity tests in various mammalian cells or cell-lines.


In a HGPRT gene mutation assay, retinol was found to be non mutagenic with and without metabolic activation with S9 mix prepared from liver homogenate of Aroclor 1254-pretreated rats (Budroe 1988).

In a collaborative study with 10 participating laboratories to evaluate a test protocol of the in vitro micronucleus (MN) test with V79 cells, retinyl acetate has been tested in 3 laboratories without metabolic system (von der Hude 2000). An increase of the frequency of micronuclei has been observed in 1/3 laboratories only at the highest concentration tested (31.6 μg/ml), resulting also in remarkable precipitation, a decrease in proliferation index and an increase in the mitotic index. No information on test substance specification, i.e. impurities is given. 


In a chromosomal abberation test and SCE test in vitro using human fibroblasts (HE2144) without metabolic activation, retinyl acetate was found positive at the highest concentration (65.6 µg/ml) (Sasaki 1980). Furthermore, a chromosomal aberration test and SCE test without  metabolic activation using a fetal Syrian hamster (female) pulmonary epithelial cell line (M3E3/C3) showed a dose dependent increase in the presence of retinol (Mohr 1991). For both studies, no data on cytotoxicity or substance specification are available to confirm the validity of the chosen experimental conditions.


In contrast, a chromosomal abberation test using Chinese hamster fibroblasts without metabolic activation found retiny acetate to be negative (Ishidate 1978).

Furthermore, retinol was found negative in a chromosomal abberration test in vitro using primary human lymphocytes without metabolic activation (Cohen 1970) and both, a chromosomal abberation test and sister chromatid exchange test using Chinese hamster V79 cells showed no genotoxic potenial of retinol in the presence and absence of a metabolic system (Qin 1985).


Retinol and retinyl acetate were negative in a sister chromatid exchange test in vitro in Chinese hamster V79 cells without metabolic activation (Sirianni 1981).

An unscheduled DNA synthesis test in primary rat hepatocytes using retinol and retinyl acetate was negative up to cytotoxic concentrations in the absence of a metabolic system (Budroe 1987, Althaus 1982). 


In in vivo erythrocyte micronucleus tests in mice according to OECD Guideline 474, no statistically significant or biologically relevant differences in the frequency of erythrocytes containing micronuclei were observed between the vehicle control and the 2 dose groups with “vitamin A Rohöl” or “vitamin A raffiniertes Öl” (1000, 2000 mg/kg bw; BASF 26M0378/064145; BASF 26M0532/064147). The test substances represent samples from the technical synthesis process of vitamin A and retinyl acetate is considered to be the main component. However, no detailed information of its composition is available. In conclusion, no chromosome-damaging (clastogenic) effect nor an impairment of chromosome distribution in the course of mitosis (aneugenic activity) in bone marrow cells in vivo was observed for these test samples, beeing indicative for an absence of clastogenicity/ aneugenicity of retinol and the respective ester.

A micronucleus test in ddY mice, scoring MNPCEs and total erythrocytes from bone marrow, is available in literature (Hayashi 1988). A single (37.5, 75, 150, 300 mg/kg bw) injection or 4 daily intraperitoneal injections of retinyl palmitate (300 mg/kg bw/d) did not increase the frequency of MNPCEs in any dose group, providing further evidence for the absence of a clastogenic potential of retinyl palmitate in vivo.

Taken all available information on retinyl palmitate and relevant studies on its respective structural and metabolic analoges retinol and retinyl acetate, retinyl palmitate is not considered to be genotoxic in a weight of evidence.

Endpoint Conclusion:

Justification for classification or non-classification

The present data on genetic toxicity do not fulfill the criteria laid down in regulation 67/548/EEC and 1272/2008/EEC and therefore, a non-classification is warranted.