Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jun 14;11(1):12487.
doi: 10.1038/s41598-021-91792-1.

A genome-wide portrait of pervasive drug contaminants

Affiliations

A genome-wide portrait of pervasive drug contaminants

Joseph Uche Ogbede et al. Sci Rep. .

Abstract

Using a validated yeast chemogenomic platform, we characterized the genome-wide effects of several pharmaceutical contaminants, including three N-nitrosamines (NDMA, NDEA and NMBA), two related compounds (DMF and 4NQO) and several of their metabolites. A collection of 4800 non-essential homozygous diploid yeast deletion strains were screened in parallel and the strain abundance was quantified by barcode sequencing. These data were used to rank deletion strains representing genes required for resistance to the compounds to delineate affected cellular pathways and to visualize the global cellular effects of these toxins in an easy-to-use searchable database. Our analysis of the N-nitrosamine screens uncovered genes (via their corresponding homozygous deletion mutants) involved in several evolutionarily conserved pathways, including: arginine biosynthesis, mitochondrial genome integrity, vacuolar protein sorting and DNA damage repair. To investigate why NDMA, NDEA and DMF caused fitness defects in strains lacking genes of the arginine pathway, we tested several N-nitrosamine metabolites (methylamine, ethylamine and formamide), and found they also affected arginine pathway mutants. Notably, each of these metabolites has the potential to produce ammonium ions during their biotransformation. We directly tested the role of ammonium ions in N-nitrosamine toxicity by treatment with ammonium sulfate and we found that ammonium sulfate also caused a growth defect in arginine pathway deletion strains. Formaldehyde, a metabolite produced from NDMA, methylamine and formamide, and which is known to cross-link free amines, perturbed deletion strains involved in chromatin remodeling and DNA repair pathways. Finally, co-administration of N-nitrosamines with ascorbic or ferulic acid did not relieve N-nitrosamine toxicity. In conclusion, we used parallel deletion mutant analysis to characterize the genes and pathways affected by exposure to N-nitrosamines and related compounds, and provide the data in an accessible, queryable database.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Layout of the interactive chemogenomic web application. Selecting a compound of interest from the drop-down list (red circle), will allow one to explore (left panel) its fitness profile, GO enrichment profile, individual gene fitness profile (cofitness) or cluster analysis of all the compounds in our screens. Compounds were saved with their final concentration and the date (yy/mm/dd) when the count matrices were performed e.g. NDMA_674.9µM_190801. To allow for custom visualizations, the user can adjust thresholds for FDR and fitness scores. Image is a screenshot from the web application.
Figure 2
Figure 2
Fitness profiles showing genes (via deletion strains) affected by nitrosamines and related compounds. Genes (arranged alphabetically) are plotted on the horizontal axis versus their corresponding fitness defect (FD) scores on the vertical axis. Sensitive strains have a positive FD score, while resistant strains have a negative FD score. (A) Sensitive strains affected by NDMA include those lacking ARG1, ARG3, GCN4, LIP5, AIM26 etc. (B) Genes affected by NDEA (via effect on their deletion strains) are PEP12, ARG3, SNF1, BRO1, among others. (C) CSG2, SUR1, ILV1, PMT2 etc. have their deletion strains affected by NMBA. (D) DMF was sensitive to strains lacking the following genes ARG3, GCN4, ORT1, SNF8 etc. The concentration at which the compound was screened is indicated. Not all strains (genes) with significant FD scores are labelled. Fitness profiles were generated using the Bioconductor v3.12, https://www.bioconductor.org/.
Figure 3
Figure 3
Gene Ontology enrichment profiles for the nitrosamines and DMF. Genes lacking in the sensitive strains were enriched for certain biological processes including; NDMA: arginine biosynthesis, fatty acid metabolic process, mitochondrial genome maintenance and macroautophagy. NDEA: protein transport, vacuolar transport, arginine biosynthesis, negative regulation of translation and macroautophagy. NMBA: protein transport, protein targeting to vacuole and cell morphogenesis. DMF: arginine biosynthesis, protein transport and iron–sulfur cluster assembly. Each node represents a gene set while the node size represents the number of genes in the gene set. The edge represents overlap between gene sets while the edge width represents the number of genes that overlap between connected gene sets. Not all pathways enriched at significant FD score are shown, e.g. septin ring assembly for NDEA, etc. Enrichment profiles were generated using the Bioconductor v3.12, https://www.bioconductor.org/.
Figure 4
Figure 4
Proposed chemistry of NDMA formation from DMF during sartan production. The active pharmaceutical ingredient in sartan drugs has a tetrazole ring and the conventional method to produce this ring was reported to be slow. In order to accelerate the production process, a synthesis procedure that uses the solvent, dimethylformamide (DMF) and sodium azide (in place of tributyltin azide) was introduced in 2012. During tetrazole synthesis, a small amount of dimethylamine would be formed from DMF. The synthetic process also involves the use of nitrous acid to dispose of the excess sodium azide. This nitrous acid (a nitrosating agent) which is produced from sodium nitrite under acidic conditions can react with dimethylamine to form NDMA. (R = aryl, alkyl or vinyl; DMF = dimethylformamide; NDMA = N-nitrosodimethylamine). Figure adapted from,–.
Figure 5
Figure 5
Relationship between nitrosamine/DMF degradation and arginine metabolism as observed in our experiments. Genes labelled in black correspond to deletion strains that displayed growth defects in our experiments, while those in blue did not show fitness defects. The dashed arrow with question marks (in yellow) means the reaction is hypothetical, i.e. depicting NH4 from nitrosamine degradation entering the arginine biosynthetic pathway. In the diagram, only metabolic intermediates of nitrosamine/DMF that we screened are shown.
Figure 6
Figure 6
Fitness profile showing genes whose deletion strains are sensitive to treatment by metabolic intermediates of nitrosamines. Fitness defect scores are plotted on the vertical axis vs genes on the horizontal axis. Strains affected by the compounds lack the following genes; methylamine: BCK1, DAL81, ARG3, FPS1 and GRR1. Ethylamine: DAL81, ARG3, SNF6, ELO3, SNF6 and ELO3. Formamide: ARG3, FPS1, NPL6 and SPT4. Ammonium sulfate: DAL81, ARG3, PEP3, THR1, etc. It could be seen that ARG3 was affected by all the metabolites, in addition to ARG1 being affected by methylamine and formamide. Concentration and date of count matrices analysis are indicated. Not all strains with significant FD scores are labelled. Plots were generated using the Bioconductor v3.12, https://www.bioconductor.org/.
Figure 7
Figure 7
Percent inhibition of the intermediates against nineteen strains lacking arginine and ammonium metabolism genes. Strains lacking CPA1 and CPA2 were sensitive to formamide, but not to methylamine, ethylamine and ammonium sulfate. However, strains that are missing ARG1, ARG3 and ARG5,6 have increased sensitivity to the four compounds. Each bar represents the mean ± standard of inhibition (%).
Figure 8
Figure 8
Percentage inhibition of the 19 arginine/ammonium deletion strains by N-nitrosamines and DMF. All the compounds showed greater  effects on ARG1, ARG3, CPA1 and CPA2 compared to other genes.
Figure 9
Figure 9
Gene ontology enrichment profiles showing pathways affected by the metabolic intermediates of N-nitrosamines. Genes deleted in strains that displayed increased sensitivity were enriched for certain biological processes including; methylamine include: chromatin remodeling, protein targeting to vacuole, microautophagy and arginine biosynthetic process. Ethylamine: cell wall chitin biosynthetic process and conjugation with cellular fusion. Formamide: arginine biosynthetic process, chromatin remodeling, ATP export and protein transport. Ammonium sulfate: protein targeting to vacuole, phosphorylation and cellular amino acid biosynthetic process. Enrichment profiles were generated using the Bioconductor v3.12, https://www.bioconductor.org/.
Figure 10
Figure 10
Genes and pathways whose deletion strains were affected by formaldehyde and 4NQO. Formaldehyde sensitized strains lacking CDC26, RAD5 and RAD18 which are involved in DNA repair process. Similarly, 4NQO induced sensitivity for strains deleted for RAD2, RAD14, RAD10, MMS2 and MUS81, involved in DNA repair. As can be seen from the figure, many affected genes are those involved in cell cycle and DNA repair. Not all pathways at significant FD score are shown, e.g. chromatin remodelling in 4NQO. Plots were generated using the Bioconductor program v3.12, https://www.bioconductor.org/.
Figure 11
Figure 11
Hierarchical cluster analysis of the all 22 chemogenomic screens. Pearson correlation was used to generate the “coinhibitory” square matrix and Ward was used as the distance. Branch colors on the two identical dendrogram indicate the three major clusters. Compounds within each cluster are highly correlated as indicated by the color scale as well as the dendrogram height and may correspond to compounds that have similar chemogenomic profiles and therefore may suggest similar mechanism of action. For example, the fitness profiles for NDEA all cluster tighter as expected (orange dendrogram branch color). Globally, two of the three major clusters (orange and navy branches in dendrogram) that include NDMA, NDEA, DMF and formamide exhibit significant correlation (visualized as darker red region in the heatmap). This finding is consistent with the chemogenomic profile similarity between these compounds, all of which were enriched for the genes involved in the arginine biosynthetic pathway. Figure was generated using the Bioconductor v3.12, https://www.bioconductor.org/.

References

    1. McKinney JD. The molecular basis of chemical toxicity. Environ. Health Perspect. 1985;61:5–10. doi: 10.1289/ehp.85615. - DOI - PMC - PubMed
    1. Sartan recalls beg the question: Is compendial impurity testing enough? PharmTechhttps://www.pharmtech.com/view/sartan-recalls-beg-question-compendial-im....
    1. Sörgel F, et al. The contamination of valsartan and other sartans, part 1: New findings. J. Pharm. Biomed. Anal. 2019;172:395–405. doi: 10.1016/j.jpba.2019.05.022. - DOI - PubMed
    1. Magee PN, Barnes JM. The production of malignant primary hepatic tumours in the rat by feeding dimethylnitrosamine. Br. J. Cancer. 1956;10:114–122. doi: 10.1038/bjc.1956.15. - DOI - PMC - PubMed
    1. Scanlan RA, Issenberg P. N-nitrosamines in foods. C R C Crit. Rev. Food Technol. 1975;5:357–402. doi: 10.1080/10408397509527180. - DOI

Publication types

MeSH terms

LinkOut - more resources