Mutation Research 777 (2015) 68–72

Contents lists available at ScienceDirect

Mutation Research/Genetic Toxicology and Environmental Mutagenesis journal homepage: www.elsevier.com/locate/gentox Community address: www.elsevier.com/locate/mutres

Short communication

Ames positive boronic acids are not all eukaryotic genotoxins Heather Scott a , Richard M. Walmsley a,b,∗ a b

Gentronix Ltd., BioHub at Alderley Park, Cheshire, SK10 4TG, United Kingdom Faculty of Life Sciences, University of Manchester, M13 9PL, United Kingdom

a r t i c l e

i n f o

Article history: Received 22 September 2014 Received in revised form 4 December 2014 Accepted 5 December 2014 Available online 15 December 2014 Keywords: Boronic acids Eukaryotic genotoxins GADD45a GreenScreen HC BlueScreen HC Ames

a b s t r a c t Boronic acids and their derivatives have been exploited for their pharmacological activity and their utility as intermediates in the synthesis of novel non-boron containing compounds. A recent study reported that boronic acids are bacterial mutagens. Here, results are reported from the testing of nine boronic acids using the pan-mechanistic eukaryotic GADD45a genotoxicity assays, BlueScreen HC and GreenScreen HC. Positive results were produced for one compound in GreenScreen and four compounds in BlueScreen. Only negative results were produced when tested with S9 metabolic activation. These data suggest that there is not a general genotoxic liability in eukaryotes, within this chemical domain. Furthermore, they are not potent eukaryotic genotoxins: positive results were produced only at concentrations between 1 mM and 10 mM. Their presence as low concentration contaminants or impurities would be unlikely to produce misleading positive results for a test material. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Boronic acids are of interest in safety assessment because they are used as intermediates in the synthesis of pharmacologically active molecules. They may therefore be present as contaminants or impurities in compounds taken forward into development. O’Donovan et al. [1] reported the first investigation of genotoxicity in these compounds using the Ames bacterial reverse mutation assay. They found that 12 of 13 compounds tested produced positive results. All but one of these 12 were positive in TA100 and or WP2uvrA (pKM101) without S9 metabolic activation. The exception was also positive in TA1537. There was no evidence for adduct formation in these studies, but attention was drawn to the reactivity of these molecules with carbohydrate moieties found in ribonucleosides. Since these moieties are not restricted to

Abbreviations: GFP, green fluorescent protein; GADD, growth arrest and DNA damage; GSHC, GreenScreen human cell GADD45a-GFP reporter assay; BSHC, BlueScreen human cell GADD45a-Gaussia luciferase reporter assay. ∗ Corresponding author at: Gentronix Ltd., BioHub at Alderley Park, Cheshire, SK10 4TG, United Kingdom. Tel.: +44 1625 238743/1 6032746. E-mail address: [email protected] (R.M. Walmsley). http://dx.doi.org/10.1016/j.mrgentox.2014.12.002 1383-5718/© 2014 Elsevier B.V. All rights reserved.

prokaryotes, it is plausible that such indirect effects might also produce positive results in eukaryotic genotoxicity assays. Whilst there have been subsequent reports of acute toxicity for these compounds in vivo [2], there have not been any reports of eukaryotic genotoxicity. This paper reports an assessment of the genotoxicity of 9 boronic acids, including 8 from the O’Donovan study, using the pan-mechanistic GADD45a genotoxicity reporter assays GreenScreen HC (GFP reporter) and BlueScreen HC (Gaussia luciferase reporter). The GADD45a assays are used to detect genotoxic hazard and have been described in detail elsewhere. Validation studies have demonstrated a high sensitivity and specificity to all classes of genotoxic carcinogens amongst mechanistically diverse compound collections including mutagens, aneugens and clastogens [3]. Promutagens are also identified from incubation with S9 liver extracts [4,5]. These studies have included compounds from a variety of applicability domains including pharmaceuticals [6], pesticides and herbicides [7]. Assay protocols both with and without S9 metabolic activation have been described, and demonstrated to produce robust reproducible results in inter-laboratory ring trials [8,9]. The GSHC assays are described in the ECVAM INVITTOX protocol number 132 (http://ecvam-dbalm.jrc.ec.europa.eu/). This broad mechanistic coverage and applicability provides a comprehensive assessment of genotoxic hazard in eukaryotic cells, and is therefore an appropriate test for this initial assessment of genotoxic liability within the boronic acids.

H. Scott, R.M. Walmsley / Mutation Research 777 (2015) 68–72

2. Materials and methods 2.1. Chemicals Table 1 lists the compounds tested in this study. They were chosen to complement the O’Donovan study, and selected on the basis of availability. 2.2. GADD45a assays The GreenScreen HC and BlueScreen HC assays have been fully described elsewhere so are only briefly summarised here. The assay uses human lymphoblastoid TK6 cells. It is a preferred cell line in genetic toxicology because it is p53 wild type, and hence has a properly regulated DNA damage response. Both assays contain plasmids which retain the promoter and the expressed gene sequences, except for a region between translational start and intron 3 which is replaced with a reporter gene: green fluorescent protein in GSHC and Gaussia luciferase in BSHC. Control strains for each assay contain an out-of-frame reporter, which allows identification of any compound-related fluorescence or absorbance interference. Compounds were tested in growth medium containing 1% DMSO in a series of 2 fold dilutions across the rows of 96 well microplates (75 ␮l per well: 9 dilutions for GSHC; 8 dilutions for BSHC). Internal controls included growth media alone (to detect microbial contamination), compound alone (to detect levels of compound colour or fluorescence that might interfere with data collection), and positive controls (methyl methane sulphonate for GSHC, 4-nitroquinoline 1-oxide for BSHC, cyclophosphamide for S9 assays) at two concentrations to confirm proper reporter response. Reporter cells from passage, which had not been allowed to grow beyond one million cells per ml, were collected by centrifugation, re-suspended in double strength assay medium, and 75 ␮l added to each test sample (reducing top dose to 10 mM, and DMSO concentration to 1%). In the metabolic activation protocols, cells and test samples were incubated for 3 h in the presence of 1% S9 liver extracts from Aroclor treated male Sprague–Dawley rats (Moltox), combined with cofactors. They were then washed, re-suspended in fresh assay medium and incubated for a further 45 h before data collection. At the end of the assay, reporter outputs (GFP fluorescence, or GLuc luminescence) were recorded together with a measure of culture density: light absorbance (620 nm) for GSHC; thiazole orange (TO) fluorescence for BSHC. TO is a nucleic acid binding dye, detectable using standard fluorescein (FITC) filters (Excitation: 485 nm; Emission: 535 nm). The fluorescence or luminescent reporter output was divided by the measure of cell density to produce a ‘brightness’ value. In this way the assay discriminates between wells containing low numbers of strongly light-emitting cells and wells containing high numbers of weakly light-emitting cells. A positive result is concluded where exposure within the acceptable toxicity range (≥30% relative suspension growth) produces a significant increase in brightness. This is

69

defined as greater than 3 times the standard deviation derived from studies of toxic and non-toxic non-genotoxins. From the triplicate testing, compounds producing 3/3 or 2/3 positive results were recorded as positive. 3. Results In the data presented below, all assays fulfilled data acceptance criteria: these include positive controls. Exceptions are noted in the text. Table 1 lists the tested compounds along with CAS numbers, purity, MW and top test concentrations. Fig. 1 shows the dose response graphs for the tested compounds. Table 2 summarises the results. 3.1. GreenScreen HC In the absence of S9 2,5-Dimethoxyphenylboronic acid (2,5,DMPBA) was the only compound that produced a positive result in all three repeat GSHC tests (LEC 2500 ␮M), and after 48 h exposure only. 5-Formyl-2-thienylboronic acid produced uninterpretable data due to auto-fluorescence at the wavelength (488 nM) used for GFP detection. This was apparent from fluorescence in the GFP detection channel from both test (GFP expressing) and control (non-GFP expressing) strains. In the presence of S9, the assay produced negative data for all compounds including the fluorescent 2,5,DMPBA. m-Tolylboronic acid produced an isolated positive result in only one of three tests at the maximum tolerated dose (10 mM) and was therefore classified at negative. 3.2. BlueScreen HC results In the absence of S9, 4 compounds produced positive results: 2,5-dimethoxyphenylboronic acid (LEC 5000 ␮M), 3,5difluorophenylboronic acid (LEC 1250 ␮M), m-tolylboronic acid (LEC 2500 ␮M) and p-tolylboronic acid (2500 mM). All produced negative results in the presence of S9. 4. Discussion This study reports the genotoxicity testing of 9 boronic acid derivatives using the GADD45a GreenScreen and BlueScreen assays, with and without S9 metabolic activation. All produce positive results in the Ames test, and the results presented here suggest that this does not translate into a general eukaryotic liability. When tested using the GreenScreen HC assay, 7 of the 9 produced negative results. 2,5-Dimethoxyphenylboronic acid (2,5DMPBA) produced the only positive result, and only without S9. 5-Formyl-2-thienylboronic acid is highly coloured and whilst it did not produce interpretable data in the GSHC assay without S9, it was negative in the S9 protocol, where the wash step to remove

Table 1 Compounds tested in this study. Compound name

CAS no.

Purity(%)

MW

Top test conc. (micromolar)

2-Cyanophenylboronic acid 2-Fluoro-6-methoxyphenylboronic acid 2,5-Dimethoxyphenylboronic acid 3,5-Difluorophenylboronic acid 5-Fluoro-2-methoxyphenylboronic acid 5-Formyl-2-thienylboronic acid m-Tolylboronic acid o-Tolylboronic acid p-Tolylboronic acid

138642-62-3 78495-63-3 107099-99-0 156545-07-2 179897-94-0 4347-33-5 17933-03-8 16419-60-6 5720-05-8

≥95.0 ≥95.0 ≥95.0 ≥95.0 ≥95.0 ≥95.0 97 ≥95.0 97

146.94 169.95 181.98 157.91 169.95 155.97 135.96 135.96 135.96

5000 10000 5000 5000 1250 5000 10000 5000 5000

All compounds were obtained from Sigma–Aldrich.

70

H. Scott, R.M. Walmsley / Mutation Research 777 (2015) 68–72

Fig. 1. Dose response GreenScreen HC (GS) and BlueScreen HC (BS) data in the presence (+S9) and absence of S9 metabolic activation. Data points 5000–2500 ␮M invalidated for 5-formyl-2-thienylboronic acid (6) in GS due to high auto-fluorescence.

------

Threshold defining a positive result.

Table 2 Lowest Effective Concentration values for test compounds. Compound name

MW

Amesa −S9

1. 2. 3. 4. 5. 6. 7. 8. 9. a b c

2-Cyanophenylboronic acidc 2-Fluoro-6-methoxyphenylboronic acid 2,5-Dimethoxyphenylboronic acid 3,5-Difluorophenylboronic acid 5-Fluoro-2-methoxyphenylboronic acid 5-Formyl-2-thienylboronic acid m-Tolylboronic acid o-Tolylboronic acid p-Tolylboronic acid

146.94 169.95 181.98 157.91 169.95 155.97 135.96 135.96 135.96

2.0 1.7 4.1 2.0 4.5 1.9 2.6 1.8

Ames data refer to fold increase over background in TA100 in the O’Donovan 2011 paper. GSHC/BSHC data refer to lowest concentration producing a positive result (mM). The isomer 4-cyanophenylboronic acid tested positive in Ames +/−S9.

GSHCb

BSHC

+S9

−S9

+S9

−S9

+S9

2.1 2.2 2.6 2.2 4.2 1.9 – 2.6

– – 2.5 – – – – – –

– – – – – – – – –

– – 5.0 1.25 – – 2.5 – 2.5

– – – – – – – – –

H. Scott, R.M. Walmsley / Mutation Research 777 (2015) 68–72

S9 also removed confounding colour interference. There is clearly no general genotoxic liability in this class detectable by the GSHC assay. When tested using the BlueScreen HC assay, 4 compounds produced positive results without S9: 2,5-DMPBA (the GSHC positive); 3,5-difluorophenylboronic acid; m-tolylboronic acid; ptolylboronic acid. All produced negative BlueScreen results in the S9 arm of the BSHC assay. In this relatively small compound collection, incubation with S9 effectively neutralises the apparent in vitro genotoxicity of boronic acids. It is not clear why the BSHC produces more positive results than GHSC. However, because the reporter constructs are identical except for their reporter genes, we hypothesize that these compounds are not eukaryotic genotoxins, and that the differences between results from the two assays might instead reflect the different properties of the respective reporter proteins. The green fluorescent protein gene first isolated from Aequora victoria and subsequently genetically modified to increase stability and brightness, is a thermostable, protease-resistant intracellular protein, which continuously emits green light (507 nm) when illuminated with blue light (488 nm). The luciferase gene first isolated from Gaussia princeps luciferase, is secreted from the cell and produces a flash of light, only when provided with its coelenterazine substrate. The Gaussia luciferase protein produced in BSHC is reported to be thermo-tolerant, and luciferase activity persists for at least 7 days in growth medium (https://www.neb.com/products/ e3300-biolux-gaussia-luciferase-assay-kit), but we have found no data relating to protease resistance. It has however been recognised for some time that boronic acids can be potent inhibitors of serine proteases, including chymotrypsin and subtilisin [10], and leukocyte elastase, pancreatic elastase and cathepsin G [11]: those findings led to the development of very specific peptidyl-boronic acid drugs. It may be that compounds which only produce positive results in BlueScreen results are indeed protease inhibitors. This could lead to a dose dependant increase in the amount of luciferase available for secretion from the cell, and an apparently positive genotoxicity result. There are no published studies reporting protease inhibitory activity with any of these 4 compounds, however, unfolded firefly luciferase has been shown to be protease sensitive [12], so Gaussia luciferase might have a similar liability. The generation of negative BSHC results with S9, for compounds producing positive BSHC results without S9 was a little surprising, not least because cells can generally tolerate higher doses of a test substance in the shorter 3 h exposure time used in the S9 assay (c.f. 48 h without S9). For example, EDTA produced a negative GSHC result at the highest tolerated dose (2.5 mM) without S9 (48 h exposure), but a positive result at the higher dose (5 mM) tolerated in the 3 hour S9 exposure; p-nitrophenol showed a similar pattern [13]. Reversal of a genotoxicity positive by S9 is also not without precedent. For example, 2,4-diaminotoluene tested positive in GSHC and BSHC without S9, and negative with S9 [4,5]. In the present study, all 4 compounds which produced positive BSHC results without S9, produced negative results with S9. A plausible explanation for this is that arylboronic acids can be converted by oxygenation to phenols using alkaline hydrogen peroxide [14]. This conversion might also be achieved by the mono-oxygenase cytochrome P450s in S9 extracts. A published example of metabolic de-boronation is provided by Bortezomib, a peptidyl boronic acid that can be metabolized by a spectrum of human P450 enzymes: CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4 [15]. All the compounds were tested to a top concentration of 10 mM unless limited by solubility or toxicity. The boronic acids can be classified as weak genotoxins, since none produce positive results below 1 mM. In the most recent revision of genotoxicity test guidance for pharmaceuticals (ICHS2R), the top testing dose was reduced from 10 mM to 1 mM. If that limit had been applied in

71

this study, none of the compounds would have produced positive results. The new guidance reflected the finding that misleading positive in vitro results for non-carcinogens, were generally produced at relatively high concentrations (≥1 mM). In the published GreenScreen HC validation data (161 compounds), performance is largely unaffected using a 1 mM limit. Five compounds positive ≤10 mM were found negative ≤1 mM: EDTA and p-nitrophenol (non-carcinogens, see above); the nucleoside analogues valacyclovir (non-carcinogen) and azidothymidine (female-specific rodent carcinogen); ofloxacin (non-carcinogen, gyrase inhibitor, which produces equivocal GSHC results). In the context of testing dose, the Ames test data are not directly comparable due to dose unit definition (␮g/plate c.f. mM). However, all these compounds have relatively low molecular weights (135.96–181.98), and if freely diffused through a 30 ml Ames plate their concentrations would have been in the range 0.88–1.17 mM i.e. in a comparable range. It can be concluded from this study that the Ames liability identified in the O’Donovan study does not reflect a general genotoxic liability in this class of compounds. 5. Conclusions Boronic acids produce positive results in the Ames test but weak or negative results in the eukaryotic GADD45a expression assays. This study suggests when such compounds are intended for human exposure, an Ames positive result alone should not exclude development, though additional appropriate in vitro/in vivo data would be required. Conflict of interest RMW is the founder/CSO, and HS are employees of Gentronix Ltd., the company which developed and sells BlueScreen and GreenScreen assay kits and services. Funding This study was funded by Gentronix Ltd. Acknowledgements Michael Ingeleson (University of Manchester) and Steve Beasley (Gentronix Ltd.) are thanked for their advice and critical reading of the manuscript. References [1] M.R. O’Donovan, C.D. Mee, S. Fenner, A. Teasdale, D.H. Phillips, Boronic acids – a novel class of bacterial mutagen, Mutat. Res. 724 (2011) 1–6. [2] Marvin A. Soriano-Ursua, Eunice D. Farfann-Garcıa, Yessica Lopez-Cabrera, Enrique Querejeta Jose´ı G Trujillo-Ferrara, Boron-containing acids: preliminary evaluation of acute toxicity and access to the brain determined by Raman scattering spectroscopy, Neurotoxicol 40 (2014) 8–15. [3] P.W. Hastwell, L.L. Chai, K.J. Roberts, T.W. Webster, J.S. Harvey, R.W. Rees, R.M. Walmsley, High-specificity and high-sensitivity genotoxicity assessment in a human cell line: validation of the GreenScreen HC GADD45a-GFP genotoxicity assay, Mutat. Res. 607 (2006) 160–175. [4] C. Jagger, M. Tate, P.A. Cahill, C. Hughes, A.W. Knight, N. Billinton, R.M. Walmsley, Assessment of the genotoxicity of S9-generated metabolites using the GreenScreen HC GADD45a-GFP assay, Mutagenesis 24 (2009) 35–50. [5] C. Hughes, A. Rabinowitz, M. Tate, L. Birrell, J. Alsop, N. Billinton, R.M. Walmsley, Development of a high-throughput Gaussia luciferase reporter assay for the activation of the GADD45a Gene by Mutagens, Promutagens, Clastogens, and Aneugens, J. Biomol. Scr. 17 (2012) 1302–1316. [6] P.W. Hastwell, T.W. Webster, M. Tate, N. Billinton, A.M. Lynch, J.S. Harvey, R.W. Rees, R.M. Walmsley, Analysis of 75 marketed pharmaceuticals using the GADD45a-GFP ‘GreenScreen HC’ genotoxicity assay, Mutagenesis 24 (2009) 455–463. [7] A.W. Knight, S. Little, K. Houck, D. Dix, R. Judson, A. Richard, N. McCarroll, G. Akerman, C. Yang, L. Birrell, R. Walmsley, Evaluation of high-throughput genotoxicity assays used in profiling the US EPA ToxCastTM chemical, Reg. Toxicol. Pharmacol. 55 (2009) 188–199.

72

H. Scott, R.M. Walmsley / Mutation Research 777 (2015) 68–72

[8] N. Billinton, P.W. Hastwell, D. Beerens, L. Birrell, P. Ellis, S. Maskell, T.W. Webster, S. Windebankd, F. Woestenborghs, A.M. Lynch, A.D. Scott, D.J. Tweats, J.V. Gompel, R.W. Rees, R.M. Walmsley, Inter-laboratory assessment of the GreenScreen HC GADD45a-GFP genotoxicity screening assay: an enabling study for independent validation as an alternative method, Mutat. Res. 653 (2008) 23–33. [9] N. Billinton, S. Bruce, J.R. Hansen, P.W. Hastwell, C. Jagger, C. McComb, M.L. Klug, K. Pant, A. Rabinowitz, R. Rees, M. Tate, A.M. Vinggaard, R.M. Walmsley, A pre-validation transferability study of the GreenScreen HC GADD45a-GFP assay with a metabolic activation system (S9), Mutat. Res. 700 (2010) 44–50. [10] M. Philipp, M. Bender, Inhibition of serine proteases by arylboronic acids, PNAS U. S. A 68 (1971) 478–480. [11] B.A. Kettner, A.B. Shenvi, Inhibition of the serine proteases, leukocyte elastase, pancreatic elastase, cathepsin G, and chymotrypsin by peptide boronic acids, J. Biol. Chem. 259 (1984) 15106–15114.

[12] J. Frydman, H. Erdjument-Bromage, P. Tempst, U. Hartl, Co-translational domain folding as the structural basis for the rapid de novo folding of firefly luciferase, Nat. Struct. Biol. 6 (1999) 697–705. [13] L. Birrell, P. Cahill, C. Hughes, M. Tate, R.M. Walmsley, GADD45a-GFP GreenScreen HC assay results for the ECVAM recommended lists of genotoxic and non-genotoxic chemicals for assessment of new genotoxicity tests, Mutat. Res. 695 (2010) 87–95. [14] A.D. Ainley, F. Challenger, Studies of the boron–carbon linkage. Part I. The oxidation and nitration of phenylboric acid, Chem. Soc. (1930) 2171–2180. [15] T. Pekol, J.S. Daniels, J. Labutti, I. Parsons, D. Nix, E. Baronas, F. Hsieh, L.S. Gan, G. Miwa, Human metabolism of the proteasome inhibitor bortezomib: identification of circulating metabolites, Drug. Metab. Dispos. 33 (2005) 771–777.

Ames positive boronic acids are not all eukaryotic genotoxins.

Boronic acids and their derivatives have been exploited for their pharmacological activity and their utility as intermediates in the synthesis of nove...
840KB Sizes 0 Downloads 11 Views