Search

	COM Home
	What's new
	Terms of reference
	Membership
	FOI
	Meetings
	Publications
	Statements
Alphabetical listing     Numerical listing
	Papers
	Links
	Contact COM

MINUTES

Present:

Chairman: Professor P Farmer	Members: Dr C Allen Dr B Burlinson Dr G Clare Dr J Clements Dr B Elliott Dr D Gatehouse Professor N Gooderham Dr D Lovell Dr I Mitchell Professor D Phillips	Secretariat: Mr J Battershill (HPA secretariat) Mr S Robjohns (HPA minutes) Dr D Benford (FSA secretariat) Dr D Mason (FSA secretariat) Ms Sue Kennedy (HPA Administration)
Assessors: Dr A Smith (HSE)	In attendance : Dr Henegan (PPG - item 4) Dr E Zeiger (Consultant - item 4) Dr D Maroni (PPG - item 4) Dr M Friedman (PPG - item 4) (Polyelectrolyte Producers Group) Miss F Pollitt (HPA COC secretary) Mr K Mistry (DH Administration)	Observers: None

CONTENT

Item		Paragraph
1.	Announcements/Apologies for absence	1
2.	Minutes of the meeting on 4 October 2007 (MUT/MIN/07/3)	5
3.	Matters Arising (not covered by later agenda items)	6
4.	Review of the Genotoxicity of Acrylamide:
	4. 1 Review of additional data from PPG (MUT/08/1) December 2007 and January 2008	8
	4.2 Overview of genotoxicity of acrylamide since 1995 (MUT/08/2)	25
5.	Review of COM guidance	57
	5.1 Discussion paper on scope and process (MUT/08/3)	65
6.	Draft working paper on mutagenicity assessment of chemical Mixtures (MUT/08/4)	71
7.	Draft Annual report 2007	72
8.	Any other business	72
9.	Date of next meeting: 14 February 2008	73

ITEM 1: ANNOUNCEMENTS/APOLOGIES FOR ABSENCE

1. The Chair welcomed Sue Kennedy as the new administration secretary.

2. He also welcomed Dr E Zeiger (Consultant, Chapel Hill NC, USA), Dr D Maroni (Polyelectrolyte Producers Group (PPG)), Dr M Friedman (PPG), Dr M Henegan (PPG), Dr D Mason (FSA), Miss Frances Pollitt (HPA COC secretary), and Stephen Robjohns (HPA secretariat).

3. Apologies for absence were received from Dr E Parry (COM member) Mrs R Glazebrook (COM member), Dr L Hethrington (HPA), Mr B Maycock (FSA), Dr K Burnett (DH Tox unit), Dr D Andrew (PSD), Mr M Hosford (EA), Dr M Simmonds (NPHS Wales), and Dr H Stemplewski (MHRA).

4. Members were reminded of the need to declare any interests before discussion of items.

ITEM 2: MINUTES OF MEETING ON 4 October 2007 (MUT/MIN/07/3)

5. Members agreed the minutes subject to some minor editorial changes.

ITEM 3: MATTERS ARISING (NOT COVERED BY LATER AGENDA ITEMS)

6. The Chair thanked members for their continued hard work and responses to various postal circulations, which had been ongoing between meetings, particularly sodium benzoate and sodium sorbate.

7. Members were also informed that the minutes relating to item 8 on aneuploidy from the last COM meeting had been forwarded to the secretary of the ACP.

ITEM 4: REVIEW OF GENOTOXICITY OF ACRYLAMIDE

4.1 Presentation from PPG (MUT/08/1 additional information tabled)

8. There were no declarations of interest. Members were alerted to tabled comments from one member who had been unable to attend the meeting, a further evaluation of the data reported in Jie and Jia 2001 (Mutagenesis, 16, 145-149, 2001), a graphical representation of the data from Besaratinia and Pfeifer (J Natl Cancer Inst, 95, 889-896, 2003 and 96, 1023-1029, 2004), and a commentary dated 12 February 2008 on dose-response analysis of the PPG study from Dr J Haseman (NIEHS, USA). Members were reminded that the HSE had requested a further evaluation from the COM regarding the information cited by the PPG. The Food Standards Agency had also requested that a consideration be given to all available genotoxicity data on acrylamide by COM. The COM agreed that the ESR review completed by HSE (EU Risk Assessment report 2002) could be used as a basis for the review.

9. The COM undertook a review of a submission of mutagenicity and other data on acrylamide from the PPG at the October 2007 COM meeting. A number of additional pieces of information were requested by COM members which included; additional historical control data regarding the oral 28 day rat MN assay undertaken by PPG, a copy of a paper by Healy L et al (Mutagenesis, 16, 163-168, 2001) which reported on evidence for a time-dependent accumulation of MN in peripheral blood of mice exposed to benzene, a copy of the studies by Witt KL and colleagues (Mutation Research, 649, 101-113, 2008) which presented an oral dose response evaluation for acrylamide induction of MN in bone marrow (microscopy based evaluation) and peripheral blood (flow cytometry) in rats and mice, and additional information on in vivo transgenic mutation assays undertaken with acrylamide (from Lambert IB et al Mutation Research, 2005, 590, 1-280).

10. The Chair noted Dr Haseman had been unable to attend the meeting for PPG . He asked Dr Zeiger to make a short presentation on the new data submitted by PPG.

11. Dr Zeiger briefly outlined the oral 28 day micronucleus assay undertaken by PPG in mice. He reported that data had been presented for the low dose response evaluation of acrylamide using both 20,000 reticulocytes and 200,000 reticulocytes. Dr Zeiger noted that statistical significance for acrylamide induced MN formation in mice was only noted at 4 mg/kg bw/day when compared to the historical control (120 animals from other studies) data in this study. The few excursions beyond the historical control range were within that which would be expected by chance. In addition the concurrent control MN frequency in reticulocytes was statistically marginally lower than the historical control MN frequency (P=0.049; t-test).

12. Dr Zeiger briefly commented on the transgenic mutation assays with acrylamide. He noted that the mutation frequency had been reduced at the low dose level for both acrylamide and glycidamide in Big Blue mice using the cII transgene. A table presenting data from the two studies undertaken by Besaratina and Pfeifer (J Natl Cancer Inst, 95, 889-896, 2003 and 96, 1023-1029, 2004), some of the data had been abstracted from graphs in these publications. The evidence indicated a lowest observed effect level (LOEL) of 3.2 µM and Dr Zeiger suggested there was a threshold for response in this assay for both acrylamide and glycidamide. He noted the mutations in mouse embryonic fibroblasts were not attributed to the cII hotspot. Dr Zeiger noted the solvent control data for the two publications were the same which raised some doubts as to when the experiments using glycidamide were undertaken.

13. Dr Zeiger noted some recent publications which postdated the October 2007 PPG submission. Thus in Mei N et al (Food Chem Toxicol, 46, 628-636, 2008) more than 94% of large and 100% of small colony mutants from acrylamide treated cultures exhibited loss of heterozygosity at the Tk locus which contrasted to the generally held view that small and large colonies in the mouse lymphoma assay represented presumed point mutations and chromosome aberrations respectively. Dr Zeiger cited Koyama (Mutation Research, 603, 151-158, 2006) and Jiang et al (Toxicol In vitro 21, 1486-1492, 2007) in support of acrylamide inducing genotoxicity in mammalian cells mediated through a diversity of mechanisms not involving gene mutation. Overall acrylamide produced gross chromosomal damage and glycidamide, depending on cell mammalian cell line used, produced large deletions, intragenic deletions and point mutations.

14. Dr Zeiger noted the most recent toxicogenomic studies with acrylamide in nematodes and mammalian cells (MCF7 human mammary tumour cells). In MCF7 cells the total number of glycidamide responsive genes was not recorded but 104 gene changes were specified into three categories (detoxication, oxidative stress and cancer progression). There was no mention of induction of genes associated with DNA damage/repair pathways other than a 3.3 fold reduction in the expression of GADD45B (a gene involved in growth arrest and DNA-damage response). Dr Zeiger reported that the authors had also concluded that induction of epoxide hydrolase might be a sensitive marker to exposure to glycidamide. He noted that there had been no gene expression changes reported in experiments using acrylamide and MCF7 cells.

15. Dr Zeiger commented that Hasegawa K et al (Tox Sci, 101, 215-225, 2008) had reported the induction of 409 genes in nematodes exposed to acrylamide which included a number of glucuronyl transferases and sperm related proteins and that 111 genes had been down regulated. Although relatively few of the changes in gene expression had been identified, there had been no mention of any changes in DNA damage or repair genes.

16. Dr Zeiger concluded his presentation by referring to two abstracts to be presented at the Society of Toxicology meeting. Guo et al had reported data from gene array investigations performed on liver tissue from mice given acrylamide via the drinking water at 500 mg/l for 3 weeks. A total of 649 genes had been identified as differentially expressed in acrylamide treated mice. These represented a wide diversity of molecular and cellular functions but there was no mention of effects on DNA damage and repair. Majanatha et al had reported 76 differentially expressed mitochondrial genes in liver tissue from Big Blue mice given 500 mg/l acrylamide for 4 weeks. Gene expression changes involved steroid biosynthesis and oxidative phosphorylation but there was no mention of effects on DNA damage and repair genes. No effects on mitochondrial gene function had been reported in Big Blue mice given 600 mg/l glycidamide in the drinking water for 4 weeks. Dr Zeiger reported that PPG was attempting to obtain full data from these studies.

17. Dr Zeiger gave an overall summary of the genotoxicity data on acrylamide and cited the evidence for aneuploidy in mammalian cells, inhibition of proteins involved in microtubule assembly, binding and disassembly. The studies reporting evidence for aneuploidy in vivo in both somatic and germ cells were noted. Overall acrylamide and glycidamide interfere with spindle related proteins in vitro at concentrations equivalent to those producing chromosome damage. There is evidence for a threshold of activity in the bone marrow MN assay with acrylamide. Acrylamide and glycidamide interfere with the meiotic and mitotic processes through spindle apparatus to produce lagging chromosomes and non disjunction. Gene mutations in vitro attributed to acrylamide have been shown to be primarily multi-locus deletions. Acrylamide induced Big Blue cII mutations in vivo are primarily frameshift mutations at a hot spot. The difference in magnitude of response between Hprt and cII mutations in vivo reflects the ability of the Hprt locus to respond to large deletions which are not recoverable at the cII or lac loci. Dr Zeiger considered that acrylamide induced cII mutations showed evidence for a threshold of response. Studies using gene expression response to glycidamide do not show any evidence for induction of DNA damage/repair pathways in vitro or in vivo. Acrylamide treatment does not induce DNA damage or repair pathways as measured in gene expression profiles. The patterns of genetic damage produced by acrylamide and glycidamide in mammalian cells, and gene expression patterns are consistent with damage resulting from the production of reactive oxygen species.

Questions on PPG presentation

18. Members asked for more information on the dose-response analysis of the PPG study taking into account dose levels of 4 mg/kg bw/day and above. Dr Zeiger recalled that linear, non linear and threshold models fitted the data equally well. PPG referred to the commentary from Dr J Haseman (NIEHS) which reported that the historical control data from 120 animals for the MN NCEs (but not MN-RETs) were directly comparable to the data. Dr Haseman had noted in his commentary that five data points from the historical control animals fell outside the 95% confidence interval for the historical control range which was very close to the number expected. There were no MN-NCE values outside the 95% confidence interval for the concurrent control group and animals dosed up to and including 2.0 mg/kg bw acrylamide/day (3/60 expected). This was consistent with a threshold response at very low doses of acrylamide. It was noted that historical control data were marginally but statistically significantly lower than the concurrent control group for the PPG study. Overall the data were considered by PPG to be consistent with threshold doses of 2 mg/kg bw/day and below. Members considered that the detailed historical control data might be heterogeneous and that historical control data close in time to the concurrent control group was very similar. Members noted the proposal for further dose-response analysis using the historical control data and considered this would not help but that further analyses where the control MN data were subtracted from the test material data would be useful. The available data did not exclude a linear dose-response model which was the simplest to apply in this particular case.

19. Members asked for more information on the size of the hprt gene compared to the cII gene and the influence of this factor with regard to the genotoxicity of acrylamide. Dr Zeiger noted the hprt gene was 45kb compared to 294bp for the cII gene. Dr Zeiger considered that the hprt would respond to large deletions including those which extended outside the coding region. Members considered that the hprt gene generally responded to small deletions, with large deletions and those extending outside the coding region resulting in cell death. Members considered that the observed loss of heterozygosity reported in the mouse lymphoma cells could in part represent recombination deletions. Dr Zeiger considered that it was necessary to re-evaluate the paradigm whereby it was assumed small mutant LY5178Y colonies represented chromosomal changes and large mutant LY5178Y colonies represented gene mutations. Members considered the information for mouse lymphoma cells was still consistent with glycidamide induced DNA adduction and subsequent gene mutations.

20. Members asked for further comments on the mechanism of mutation in transgenic animals at the cII transgene. Dr Zeiger considered that the in vitro studies in Big Blue mouse fibroblasts were consistent with a threshold response. In in vivo studies in Big Blue transgenic mice acrylamide induced frameshift mutations predominantly at a hotspot in the cII transgene at dose levels which were highly toxic to the animals. Dr Zeiger did note that there were other mutations reported by Majanatha et al Environ Mol Mutagen, 47, 6-17, 2006. Mutation responses in transgenic fibroblasts and transgenic Big Blue mice at toxic doses might represent oxidative stress. Members commented that the control data reported by Besaratinia should be followed up by the secretariat but there was no current reason to doubt the data reported in these two publications.

21. Members commented that no P53 related changes in mRNA levels were reported and also commented that upregulation of P53 transcription would not be expected. Furthermore, DNA damage does not routinely lead to upregulation of transcription of DNA damage repair genes. Therefore, lack of a change in DNA damage related gene expression profile cannot be used as an argument for lack of genotoxic response. There was uncertainty that the studies reported could add any relevant information to the evaluation of acrylamide and glycidamide.

22. Members commented on the Witt et al paper (Mutation Research, 649, 101-113, 2008) and noted that the data across all the chemicals tested had shown that the previously high abstraction of MN erythrocytes by the spleen particularly in rats was not evident. The clear dose-response for acrylamide in mice and lack of evidence for MN formation in rats related predominantly to the higher rate and extent of conversion of acrylamide to glycidamide in mice compared to rats. PPG noted that autoradiography studies in male mice indicated that the majority of absorbed acrylamide was bound to proteins in the testes and noted the higher potency of acrylamide in germ cells compared to somatic cells and the specificity of acrylamide for latter stages of spermatogenesis (from pachytene spermatocytes up to spermatozoa) were consistent with a predominant genotoxic mechanisms involving protein binding rather than DNA binding. In answer to a question from a COM member PPG considered that translocations reported in studies with acrylamide might relate to the oxidative stress induced by acrylamide. Members commented that there were only a limited number of studies which had used formamidopyrimidine-DNA-glycosylase (FPG) treatment to further explore the role of oxidative DNA damage in genotoxicity of acrylamide and noted there was evidence for such an effect using the comet assay (Blasiak et al Chem Biol Interact, 149, 137-149, 2004).

23. Members asked if further data using the CYP2E1 proficient strain of S typhimurium YG7108pin3ERb5 with glycidamide were available. PPG indicated that they were still currently seeking a laboratory to undertake this work. PPG questioned whether the N7-GA-Gua adduct would be mutagenic in bacteria. The Chair noted the minutes of the October 2006 meeting should be amended as these minutes had indicated that appropriate data were available. Members noted N7-methyl-GUA was mutagenic in Salmonella typhimurium strains.

24. The Chair thanked Dr Zeiger for his presentation. PPG withdrew from the meeting.

4.2 Review of genotoxicity of acrylamide (MUT/08/2)

25. Members were reminded that the COM secretariat drafted an overview of the EU Risk Assessment of acrylamide and outlined a strategy for the COM review of published literature in MUT/07/17. The COM agreed that the EU risk assessment could form the basis of literature reviewed up to 1995, and that COM secretariat overview would focus on published literature from 1995 onwards. The COM also agreed to review a number of specific research papers published prior to 1995 which had been identified by the secretariat but not included in the EU risk assessment report. A reference list for the current review was subsequently distributed to COM members. A small number of newly published additional references were identified by COM members. The overview presented a systematic review of the genotoxicity literature on acrylamide post 1995. Summaries of all the retrieved references had been provided in Annex 1 table 1 with a concise summary of the results of each study in table 2 of Annex 1. Excerpts from the reviewed references were provided in Annex 2. Copies of relevant toxicokinetic studies had not been provided at this juncture (although cited as Annex 3 to MUT/08/2 in the covering paper). The overview provided in the covering paper had been subdivided into a series of questions on each of the genotoxicity testing approaches used. The secretariat noted that it had not been possible to undertake a review of all the glycidamide data in the time available, although the key studies had been summarised in Annex 1.

26. The secretariat noted that overall the evidence supported at least three possible modes of acrylamide genotoxicity which included metabolism to glycidamide with subsequent DNA adduct formation, oxidative stress induced by acrylamide and glycidamide and inhibition of proteins (such as kinesins) associated with microtubule function by acrylamide and glycidamide. These were not necessarily mutually exclusive and may each play a part in genotoxicity related to acrylamide exposure. The Committee was asked to consider whether it might be appropriate to apply a threshold to each of these modes of action.

27. The Chair asked for general remarks on the conduct of the review. Members agreed the review had been conducted to a high standard and the tabulation of the data had aided considerably with regard to undertaking the review. A number of other general aspects of the review were also raised during the COM discussion. Members noted that redundancy of protein targets for acrylamide indicated that there would be potential threshold for some aspects of acrylamide induced genotoxicity activity. However none of the studies reported could inform on a threshold for genotoxicity and the results of such tests would at most determine a No Observed Effect level for the particular test and test material under consideration.

28. The Chair asked members to comment in the order set out in the covering paper and to advise on the questions raised at the end of each section.

In vitro genotoxicity of acrylamide

In vitro -gene mutation

29. Members commented that the study by Yang et al (J Vet Sci, 6, 103-109, 2005) had reported a positive mutagenic response in Salmonella typhimurium TA 98 both in the absence and presence of exogenous metabolic activation (Rat S-9 Aroclor 1254 induced) and in S. typhimurium TA100 in the presence of exogenous metabolic activation at relatively high doses of 2-5 mg/plate. These data had not been found in several other tests in S. typhimurium strains either in presence of absence of exogenous metabolic activation. Overall the results might represent a stress response in the particular strains used in this laboratory. Members considered that the data from this study did not alter the weight of evidence with regard to the genotoxicity of acrylamide in bacteria. There were no equivalent data from this laboratory to indicate what results would have been obtained with glycidamide. Members reviewed the data provided by Emmert B et al (Toxicology, 228, 66-76, 2006) and agreed no conclusions could be reached given the inconsistent response from test materials which required metabolic activation by CYP2E1. Members commented on the lack of data for glycidamide in S. typhimurium tester strains which metabolically competent for CYP2E1. Overall the available data did not alter the conclusions reached in the EU risk assessment report.

In vitro gene mutation (mammalian)

30. Members reviewed the data from Besaratinia and Pfeifer (J Natl Cancer Inst, 95, 889-896, 2003 and 96, 1023-1029, 2004). The evidence supported a dose-related mutagenic effect at the cII transgene in Big Blue fibroblasts with glycidamide and a dose-related effect for acrylamide which was subsequently reduced at high dose levels by a cytotoxic effect. Members considered it inappropriate to refer to there being a 'threshold' evident in this assay and that it would be better to consider this the limit of detection for mutations in this assay. Members considered that CYP2E1 activity in Big Blue fibroblasts had not been established and the mechanism for the mutagenicity of acrylamide was unclear. Members noted the equivocal results reported by Baum M et al (Mutat Res, 580, 61-69, 2005) and negative results reported by Koyama N et al (Mutat Res, 603, 151-158, 2006) which might be due to a lack of metabolic activation of acrylamide in these test systems. Overall the data from Besaratinia was most likely consistent with a gene mutational effect of acrylamide mediated though metabolism to glycidamide, although no definite conclusions could be drawn from the available data.

31. The COM briefly reviewed the additional summary of the publication from Mei N et al (Food Chem Tox, 46, 628-636, 2008) and asked for a full copy to be made available. The data were consistent with a positive mutagenic effect of acrylamide in LY5178Y cells but no conclusion on the mechanism of action could be derived.

32. Members concluded that the available data did extend the conclusion reached in the EU risk assessment report and that there was evidence to suggest a gene mutational effect of acrylamide in mammalian cells. However, in view of the uncertainties with regard to the mutagenic mode of action, the COM could not support the definite conclusions outlined in the EU risk assessment report.

In vitro cell transformation

33. The COM concluded that no definite conclusion regarding the genotoxicity of acrylamide could be derived from the available cell transformation studies.

In vitro chromosomal aberrations

34. There was evidence for chromosomal aberrations (predominantly breaks) at high acrylamide dose levels in CHL cells (in presence and absence of exogenous metabolic activation) (Yang H et al J Vet Sci, 6, 103-109, 2005) and in V79 Mz cells (reportedly CYP2E1 deficient) (Martins C Toxicol Sci, 95, 383-390, 2007). The data indicated a clastogenic effect of acrylamide in the absence of exogenous metabolic activation but no definite conclusion on the mode of action could be derived. Members considered that differences in the maximum tested concentration could explain some of the differences observed between studies. There was also considered to be conflicting data with regard to the requirement for CYP2E1 activity; possibly indicative of a multiple modes of action. Overall these data did not extend the conclusions reached in the EU risk assessment report.

In vitro micronucleus tests (mammalian cells)

35. Members agreed that the study undertaken by Jie Y and Jai C (Mutagenesis, 16, 145-149, 2001) using FISH analysis of chromosomal changes in NIH3T3 cells indicated a clastogenic and aneugenic mode of action for acrylamide. Members noted that there were no specific data on CYP2E1 activity in NIH3T3 cells. Members noted some limited evidence for an equivocal response of acrylamide in some donors in peripheral blood lymphocytes (Baum M et al Mutat Res, 580, 61-69, 2005) and an equivocal response in TK6 cells at high dose levels (Koyama N et al, Mutat Res, 603, 151-158, 2006). When considering the role of kinesin inhibition in these assays, members questioned whether it would be necessarily appropriate to apply a threshold to the effect. There would need to be data regarding the redundancy of the kinesin target molecules and/or binding site(s) on kinesin. Furthermore, the effects of partial inhibition of the total kinesin pool needed to be investigated, since it might not be an all or nothing response.

36. Overall these micronucleus studies extended the conclusions reached in the EU risk assessment report where no in vitro micronuclei studies in mammalian cells had been reported.

In vitro DNA damage (mammalian cells)

37. Members noted that in vitro DNA damage tests using the comet assay in a number of cell lines had incorporated specialised procedures to investigate the role of metabolism, oxidative DNA damage and free radical formation.

38. Ma X et al 1996 (in abstract form only) investigated comet response in keratinocyte cell line HaCaT cells. A positive response was documented which was reduced by incubation with the pan P450 inhibitor 1-aminobenzotriazole. The authors concluded that the response in this test system was due to oxidative metabolites formed from acrylamide (abstract available only). Blasiak J et al (Chem Biol Interact, 149, 137-149, 2004) reported a dose related increase in DNA damage (alkaline comet assay) in PBLs from a healthy male donor. Post treatment of cell with repair enzymes recognising alkylated bases increased the response indicating DNA alkylation. Pre-treatment with compounds/vitamins inhibiting free radical/oxygen species formation caused a decrease in DNA damage. Overall the mechanism of effects was unclear. It was noted that the extent of acrylamide metabolism by PBLs was unclear. A negative result for DNA damage was reported by Baum using whole blood cultures from three male healthy donors. (Baum M et al 2005). Puppel N et al (Mutat Res 580, 71-80, 2005) reported a positive response for DNA damage in V79 cells and Caco-2 cells at relatively high concentrations (6 mM acrylamide) but not in rat hepatocytes. It was noted that treatment with BSO (to reduce glutathione levels) increased the DNA damage. Members noted that V-79 cells and Caco-2 cells were not expected to have any CYP2E1 activity but specific data were not available. The mechanism for the DNA damage in this study was unclear but was most likely mediated by acrylamide. Evidence for DNA damage was documented in primary thyroid cells from dog, sheep and humans and in rat thyroid cell line PC13and FRTL5 cells using alkaline comet assay. The impact of metabolism in these cells and mechanisms of effects was not studied (Chico G et al Mol Cell Endocrinol, 257-8, 6-14,2006). Negative findings for the comet assay were reported in TK6 cells (at concentrations up to 14 mM). (Koyama N et al Mutat Res, 603, 151-158, 2006).

39. The COM noted that no comet assays had been reported in the EU risk assessment report and agreed the data reviewed in MUT/08/2 extended the conclusions reached in the EU risk assessment report. The data were consistent with a complex mode of action involving a number of possible genotoxic actions of acrylamide and glycidamide.

Comments on summary of in vitro genotoxicity

40. Members commented that the reference to the bacterial mutagenicity reported by Yang et al needed to be placed into context with all the data available on acrylamide. More information on the potential range of mechanisms of acrylamide genotoxicity in vitro should be incorporated.

DNA adduct formation

41. The Committee reviewed all the available in vitro and in vivo DNA adduct data published since 1995. There were adequate methods for N7-GA-Gua and N3-GA-Ade, but it was noted that the method was not able to detect the N1-GA-Ade adduct. In cultured cells acrylamide adducts were detected at dose levels (ca 2 mM) substantially higher than required for the detection of DNA adducts following treatment with glycidamide (ca 1 µM) (Martins C et al Tox Sci, 95, 383-390, 2007). The N3-GA-Ade were present at levels of approximately 60-100 fold lower than N7-GA-Gua. Members noted a similar pattern of results from in vivo studies with acrylamide and noted that the levels of N7-GA-Gua were proportional to the internal exposure to glycidamide (AUC) in both rats and mice. (Doerge et al Tox Appl Pharmacol, 202, 258-267, 2005, Tox Appl Pharmacol, 208, 199-209, 2005, Mutat Res, 580, 131-141, 2005) Members noted the study by Ghanayem B et al (Tox Sci, 88, 311-318, 2005) where CYP2E1 null and wild type mice had been used. The finding of low but detectable levels of N7-GA-Gua in tissues (liver, testes, lung), from CYP2E1 null mice dosed with 50 mg/kg bw acrylamide indicated metabolism to glycidamide via other pathways other than through CYP2E1. Members discussed whether it would be possible to identify metabolic pathways for the formation of glycidamide which did not involve CYP2E1 but could not derive any conclusions.

42. Members noted that PPG had indicated that N7-GA-Gua was of very limited mutagenic potential and agreed that the mutagenic potential of this adduct could not be completely discounted. Members also considered that the potential formation of O6-GA-Gua adducts was possible.

43. Members agreed the studies reviewed did expand the conclusions reached in the EU risk assessment report.

In vivo genotoxicity of acrylamide

In vivo gene mutagenicity

44. The Chair recalled that the COM position for in vivo mutagens was to presume a non-threshold mechanism unless compound specific data were available to conclude a threshold existed. The COM had previously reviewed transgenic mutation studies for a number of compounds and in the absence of a clear rationale to disregard such results and to suggest a clear non-genotoxic mechanism, the conclusion reached when there was evidence of in vitro mutagenic activity was that positive data in in vivo transgenic mutagenicity studies indicated in vivo mutagenic potential by a gene mutational non-threshold mechanism. Members considered that the results reported by Majanatha M et al (Environ Mol Mut, 47, 6-17, 2006) needed to be carefully considered. The majority of mutations were consistent with N7-GA-Gua adducts (A-G transitions), but A-C and A-T transversions were consistent with N1-GA-Ade and N3-GA-Ade adducts. Overall the data provided evidence for a gene mutational response of acrylamide mediated by metabolism to glycidamide. Members noted the PPG proposal regarding evidence of a frame shift mutagenic effect at certain hotspots in the cII transgene. Members concluded there were insufficient compound specific data for any change to current COM approach to in vivo mutagens. Members noted that data using Big Blue rats and subsequent cell lines would assist in determining the role of metabolism in the observed mutagenic responses in the cII transgene. Members agreed the hprt mutagenicity data from transgenic animals was supportive of the conclusions reached for the cII transgene.

45. Members agreed these data expanded the conclusions reached in the EU risk assessment report.

In vivo chromosomal aberrations

46. Members agreed the data reported post the completion of the EU risk assessment report was consistent with the conclusions of that report. Members noted the evidence for hyperploidy reported by Gassner and Adler (Mutat Res, 367, 195-202, 1996) was supporting information regarding a potential aneugenic effect.

In vivo micronucleus assays.

47. Members noted that there were 11 additional studies postdating the EU risk assessment report retrieved by the secretariat. Members agreed the evaluation of Abramasson-Zetterberg (Mutat Res, 535, 215-222, 2003) and considered the most simple modelling of the dose-response data (ie linear) was acceptable in a situation where other approaches gave only marginal improvement in dose-response analysis. The data from other dose-response studies (Paulsson et al 2002, Davies and Rieco 2007) were also consistent with a linear dose-response. It was agreed that a NOEL could be determined from these studies but this would reflect the study design and sensitivity of the genotoxicity assay under consideration rather than a biological threshold for genotoxic activity. The COM noted the evidence for both clastogenic and aneugenic effects of acrylamide reported by Schriever-Schwem G et al (Mutagenesis, 12, 201-207, 1997) and considered the evidence suggested a more potent clastogenic response than aneugenic response.

48. The COM agreed these data did expand the conclusions reached in the EU risk assessment report particularly with regard to dose-response assessment in mice.

In vivo human biomonitoring

49. The COM agreed no useful conclusions could be derived from the one study retrieved.

In vivo DNA damage assays

50. Members noted the evidence from Ghanayem et al (Mut Res, 578, 284-297, 2005) where no DNA damage was reported in CYP2E1 null mice suggested that metabolism of acrylamide was required for the DNA damaging effects in mice dosed with acrylamide. It was likely that the dose level was too low for any metabolism via pathways other than CYP2E1 to form glycidamide in appreciable amounts for detection in the comet assay.

51. The Committee agreed these new data did expand the conclusions reached in the EU risk assessment report and hence COM would need to provide updated conclusions for this aspect of acrylamide genotoxicity.

In vivo DNA synthesis

52. These data were noted by members but no conclusions were reached with regard to the genotoxicity of acrylamide.

Comments on conclusion on In vivo genotoxicity of acrylamide

53. The Committee agreed the conclusion was adequate but asked for more information on dose-response analysis to be included.

In vivo germ cell mutagenicity of acrylamide

54. The Committee only agreed a number of general comments on the reviewed data and agreed to consider the conclusion on germ cell mutagenicity of acrylamide in more detail at the 12 June 2008 meeting. Members commented on the evidence from Ghanayem et al (Biol Reprod, 72, 157-163, 2005) and noted evidence for dominant lethal effects (reduced implantation and live births) of acrylamide in CYP2E1 proficient mice but no evidence for dominant lethal effects in CYP2E1 null mice. Members noted that Adler et al (Mutagenesis, 15, 133-136, 2000) there was some limited evidence that pre treatment with 1-aminobenzoreiazole (ABT) to inhibit metabolism of acrylamide partially offset the dominant lethal effects of acrylamide in mice. However no definite conclusions could be drawn since treatment with ABT alone also had an effect on sperm motility. Members agreed that there was evidence for effects on pachytene spermatocytes, late spermatids and early spermatozoa Marchetti F et al (Environ Mol Mutagen, 30, 410-417, 1997). Members noted that there was evidence for a range of effects on micronuclei formation in mouse embryos including both clastogenicity and aneugenicity. Titenko-Holland N et al (Environ Mol Mutaqen, 31, 206-217, 1998).

55. The Chair noted that there was a need to consult a member who was not present at the meeting with regard to the interpretation of some of these data.

Concluding remarks

56. The Chair thanked members for their contributions and noted that the COM had made good progress with the review. He suggested that the secretariat liaise with PPG regarding any other additional data which they may have, and that a further review paper on glycidamide should be drafted. A further consideration of the germ cell data on acrylamide should also be undertaken at the next meeting. The secretariat should prepare a first draft working paper for members' consideration.

ITEM 5: REVIEW OF COM GUIDANCE: DISCUSSION PAPER ON SCOPE AND PROCESS

57. The Committee heard that the terms of reference for the COM included provision of generic advice on testing strategies and risk assessment of mutagens. The COM had previously published generic guidance documents, and this had been accomplished generally at around 10 year intervals. The COM guidance documents (1989, and 2000) had been influential both with regard to national and international approaches to mutagenicity test strategies.

58. Members also heard that the objective of the COM guidance had always been to aim for the highest level of scientific advice possible. It was noted that the operation of regulatory and advisory decision making in genotoxicity within the UK and EU was very different to the situation which pertained in 2000. Thus most guidance on chemical risk assessment was now based at an international level (eg WHO IPCS, EU levels) with relatively little at the national level. Thus it was suggested that revision of the COM guidance would not significantly influence the current approach used by REACH or the revision of the International Conference on the Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) review of their guidance on mutagenicity testing (currently ongoing). However, members were informed that the COM was well placed, as an independent scientific advisory committee, to provide 'gold' standard advice, which could subsequently influence regulatory thinking. MUT/08/3 provided a copy of the existing guidance (Annex 2), information on the timeline for the previous consideration of COM guidance (Annex 1), and a discussion on potential scope for a revised guidance document. It was suggested that there was a need to include aspects of generic COM guidance which was not included in current guidance (eg approach to risk assessment of mutagens and approach to identification of compounds acting through potential threshold mechanisms).

59. It was also suggested that that two COM members could work on an initial draft, which could be discussed at later COM meetings regarding revisions. The drafts and revisions would be published on the internet to allow a fully open process. Relevant professional groups could attend meetings during the drafting process and a consultation document would be produced.

60. The committee considered the potential scope of a new guidance document. Members recalled that previous guidance had been based on the use of well validated methods as the primary approaches to genotoxicity testing. Members agreed that the rationale for revising the COM genotoxicity test strategy and new test methods would have to be evaluated in light of recent testing developments eg proposals to update the ICH mutagenicity testing guidance. Members agreed that it would be important to be aware of the new approaches recommended by other authoritative bodies such as the US EPA, OECD and UKEMS and that their views could be sought during an open process of producing a new guidance document.

61. Members noted a number of initial aspects which could be considered including proposals made by the ICH, eg an option to omit an in vitro mammalian test when two in vivo end points have appropriate data and the weight of evidence suggested that an acceptable in vitro could not be undertaken, the proposal for reduction of the top dose in in vitro mammalian cell assays from 10 mM to 1 mM, the need for duplicate tests with bacterial mutagenicity assays, the integration of genotoxicity endpoints into routine toxicology studies, updating advice on the choice of a 2nd in vivo assay to include the comet assay (with a decreasing emphasis on the UDS assay), and the proposal that a concurrent positive control could be omitted as a routine part of in vivo genotoxicity assays. Members noted that although these were appropriate aspects to review, much of the ICH process and data evaluation was not open to scrutiny.

62. The Committee agreed that it was important to emphasise in any new COM guidance document that a testing strategy had been developed with the 'three Rs' in mind (Reduction, Refinement and Replacement of animal tests).

63. Members agreed that there was a need to include advice on the use of QSAR data in the revision of the COM testing strategy particularly in the light of the proposals for greater use of QSAR information as part of the REACH strategy for chemical risk assessment. The committee also recommended that advice on the use of human biomonitoring data which had been the subject of extensive COM consideration over the past few years should be included. This would include inclusion of biomonitoring data within a genotoxicity testing strategy and also the use of biomonitoring data in the risk assessment process.

64. The Chair suggested the secretariat circulate a proposed scoping paper between meetings and that the scope of the COM guidance document was considered at the June 2008 COM meeting. It was noted that the proposals to include more information on consideration of the evaluation of potential thresholds for genotoxicity and biomonitoring would make the proposed revised COM guidance more similar to the current COC guidance (Guidance for the risk assessment of chemical carcinogens) which was published in 2004.

ITEM 6: DRAFT WORKING PAPER ON CHEMICAL MIXTURES (MUT/08/4)

65. The COM had considered the evaluation of the mutagenicity of mixtures at its previous meeting in October 2007 and at other earlier meetings. In response to the committees' earlier considerations a draft working paper on the mutagenicity assessment of mixtures had been prepared (MUT/08/4).

66. Members were reminded that a COT Working Group had undertaken a substantive review of the risk assessment of mixtures and that the terminology adopted by COT should be used in the COM draft working paper.

67. The main focus of the COM work had been to provide advice on specific aspects of genotoxicity of mixtures which had not been covered in the COT review or in the generic guidance from Interdepartmental Group on Health Risk from Chemicals (IGHRC). This included the mutagenicity testing of whole mixtures, fractions and the evaluation of interactive studies of genotoxicity using combinations of chemicals.

68. One key conclusion from the systematic review was that there was no example of an interaction (ie synergy or antagonism) with regard to mutagenicity identified. Essentially, there was no appropriate independent confirmation of the results in separate tests, or within an appropriate genotoxicity testing strategy for the identification of interactions and no definite conclusions could be reached.

69. The COM had also agreed that available studies had raised a number of potential hypotheses regarding interaction between ultimate DNA reactive chemicals and DNA structure (eg different mechanisms of DNA alkylation), the effect of covalent binding to DNA of one chemical on the potential for other reactive metabolites and chemicals to bind to DNA and possible epigenetic mechanisms, which could potentially result in a mutagenic response that resulted from an interactive effect between chemicals (ie synergistic or antagonistic). However, the COM concluded that there was a need for further research for the confirmation of such mechanisms. But, if such mechanisms were confirmed in an appropriate mutagenicity testing strategy these might be potentially significant for public health.

70. Members agreed the document had been well drafted. The Chair asked for any comments to be provided to the secretariat in writing.

ITEM 7: DRAFT ANNUAL REPORT 2007 (MUT/08/5)

71. The Chair asked for any comments to be provided to the secretariat in writing.

ITEM 8: ANY OTHER BUSINESS

72. Members requested information regarding a recent COT workshop on epigenetics and were informed that abstracts from the meeting would be made available to the COM.

ITEM 9: DATE OF NEXT MEETING

73. 12 June 2008.

ACTIONS

Top

copyright: © | updated: 2 July 2008