Nucleoside Analogues Are Potent Inducers of Pol V-Mediated Mutagenesis

biomolecules

Article Nucleoside Analogues Are Potent Inducers of Pol V-mediated Mutagenesis

Balagra Kasim Sumabe 1,2,3 , Synnøve Brandt Ræder 1 , Lisa Marie Røst 4 , Animesh Sharma 5, Eric S. Donkor 6, Lydia Mosi 2,3, Samuel Duodu 2,3, Per Bruheim 4 and Marit Otterlei 1,7,*

1 Department of Clinical and Molecular Medicine, Faculty of Medicine and Health Sciences, NTNU, Norwegian University of Science and Technology, NO-7489 Trondheim, Norway; [email protected] (B.K.S.); [email protected] (S.B.R.) 2 West African Centre for Cell Biology of Infectious Pathogens (WACCBIP), University of Ghana, P.O. BOX LG 54 Accra, Ghana; [email protected] (L.M.); [email protected] (S.D.) 3 Department of Biochemistry, Cell and Molecular Biology, University of Ghana, P.O. BOX LG 54 Accra, Ghana 4 Department of Biotechnology and Food Science, Faculty of Natural Sciences, NTNU Norwegian University of Science and Technology, NO-7481 Trondheim, Norway; [email protected] (L.M.R.); [email protected] (P.B.) 5 Proteomics and Modomics Experimental Core Facility (PROMEC), NTNU Norwegian University of Science and Technology, NO-7481 Trondheim, Norway; [email protected] 6 Department of Medical Microbiology, University of Ghana Medical School, P.O. Box 4236 Accra, Ghana; [email protected] 7 Clinic of Laboratory medicine, St. Olav University Hospital, NO-7006 Trondheim, Norway * Correspondence: [email protected]; Tel.: +47-72573075; Fax: +47-72576400

Abstract: Drugs targeting DNA and RNA in mammalian cells or viruses can also affect bacteria present in the host and thereby induce the bacterial SOS system. This has the potential to increase Citation: Sumabe, B.K.; Ræder, S.B.; mutagenesis and the development of antimicrobial resistance (AMR). Here, we have examined Røst, L.M.; Sharma, A.; Donkor, E.S.; nucleoside analogues (NAs) commonly used in anti-viral and anti-cancer therapies for potential Mosi, L.; Duodu, S.; Bruheim, P.; effects on mutagenesis in Escherichia coli, using the rifampicin mutagenicity assay. To further explore Otterlei, M. Nucleoside Analogues the mode of action of the NAs, we applied E. coli deletion mutants, a peptide inhibiting Pol V (APIM- Are Potent Inducers of Pol peptide) and metabolome and proteome analyses. Five out of the thirteen NAs examined, including V-mediated Mutagenesis. Biomolecules three nucleoside reverse transcriptase inhibitors (NRTIs) and two anti-cancer drugs, increased the 2021, 11, 843. https://doi.org/ 10.3390/biom11060843 mutation frequency in E. coli by more than 25-fold at doses that were within reported plasma concentration range (Pl.CR), but that did not affect bacterial growth. We show that the SOS response Academic Editor: Udi Qimron is induced and that the increase in mutation frequency is mediated by the TLS polymerase Pol V. Quantitative mass spectrometry-based metabolite proﬁling did not reveal large changes in nucleoside Received: 21 April 2021 phosphate or other central carbon metabolite pools, which suggests that the SOS induction is an Accepted: 2 June 2021 effect of increased replicative stress. Our results suggest that NAs/NRTIs can contribute to the Published: 5 June 2021 development of AMR and that drugs inhibiting Pol V can reverse this mutagenesis.

Publisher’s Note: MDPI stays neutral Keywords: NA; NRTIs; β-clamp; SOS; Pol V; TLS; AMR; MDR with regard to jurisdictional claims in published maps and institutional afﬁl- iations. 1. Introduction Antimicrobial resistance (AMR) is on the rise, posing serious global health threats, including high medical costs and mortality rates. In the United States, AMR is reported Copyright: © 2021 by the authors. to account for nearly 3 million infections and 35 thousand deaths yearly [1]. In Europe, a Licensee MDPI, Basel, Switzerland. higher prevalence of AMR infections is reported by countries in the southern and eastern This article is an open access article parts of the continent, compared to countries in the north [2]. Exact data are not available distributed under the terms and for Africa because 40% of the countries do not have records on the incidence of AMR, but conditions of the Creative Commons the fraction of infections with AMR is increasing [3]. Several Asian countries have a very Attribution (CC BY) license (https:// high incidence of AMR [3] and, in the next 30 years, antibiotic resistance is estimated to creativecommons.org/licenses/by/ 4.0/). cause an annual global mortality rate of 10 million deaths and a USD 100 trillion loss in

Biomolecules 2021, 11, 843. https://doi.org/10.3390/biom11060843 https://www.mdpi.com/journal/biomolecules Biomolecules 2021, 11, 843 2 of 13

the global economy, if not addressed. Most of these deaths and losses in the economy are predicted to be from Africa and Asia [4]. Following the outbreak of COVID-19 [5], the increased usage of antibiotics could further accelerate the development of antibiotic resistance [5–7]. Bacteria gain resistance to antibiotics via the development of endogenous mutations and/or transfer of resistance genes. Cellular stress, induced, for example, by antibiotics, activates SOS-dependent DNA translesion synthesis (TLS) and this is a major mechanism through which endogenous mutations lead to antibiotic resistance [8]. The ability of the TLS polymerases to introduce mutations is linked to their more spacious catalytic site, enabling a bypass of DNA lesions, and their lack of 30-5’ exonuclease activity (proofreading activity). There are three TLS polymerases in Escherichia coli, Pol II, Pol IV and Pol V (gene product of polB, dinB and umuDC, respectively). In contrast to Pol II and Pol IV, which are also expressed at low levels in absence of SOS response, Pol V is strictly SOS-regulated and, therefore, vital for bacterial mutagenesis [9]. All three bacterial TLS polymerases have been implicated in the generation of mutations contributing to the development of resistance and virulence in bacteria. For example, all three polymerases contribute to the development of mutations conferring ciprofloxacin and rifampicin resistance [8] and both Pol IV and Pol V generate base substitution mutations in the ampD gene, conferring resistance to ampi- cillin [10]. Pol IV is also shown to be relevant for virulence in uropathogenic E. coli (UPEC) strains in infected mice [11]. All bacterial polymerases interact with a subunit of the DNA polymerase III (Pol III) holoenzyme, the β-clamp, a homodimer with structural similarities to the mammalian DNA sliding clamp, PCNA (proliferating cell nuclear antigen) [12]. Infections with resistant bacteria will certainly be more devastating in immunocom- promised individuals, such as those having HIV/AIDS and cancer, compared to the normal population. Thus, with limited help from the immune system, efficient antibiotics become vital. It is therefore worrying that an increased number of multidrug-resistant (MDR) bacteria have been isolated from HIV-positive individuals compared to HIV-negative individuals in several studies [13–16]. Characterization of MDR bacteria isolated from HIV patients suggests that endogenous mutations contribute more to the development of resistance than uptake of plasmids carrying drug resistance [17]. Nucleoside analogues (NAs) are used not only in anti-cancer therapies, but also in anti-retroviral therapies as they effectively reduce viral load, and thereby improve the health of the patients and reduce the risk of viral transmission [18]. NAs mimic the natural building blocks of DNA, deoxyribonucleotide triphosphates (dNTPs), and are incorporated into DNA. However, elongations and/or base pairing are blocked by these drugs and this leads to termination of replication [19]. Some NAs, e.g., zidovudine, have a much higher affinity for viral reverse transcriptase than for human polymerases and these are therefore defined as nucleoside reverse transcriptase inhibitors (NRTI) [20]. Here, we have investigated the mutagenic activities of some NAs commonly used in anti-viral and anti-cancer therapies on E. coli MG1655. We show that NAs, particularly stavudine, didanosine, 5-fluoro-uracil (5-FU), emtricitabine and fludarabine, increase the mutation frequency by 50–100-fold in E. coli. Deletion mutant analysis showed that the increased mutagenesis was driven by the TLS polymerase Pol V, suggesting that the SOS response is induced in the bacteria upon NA treatment. Furthermore, the cell- penetrating APIM-peptide, previously shown to target the polymerase interaction site on the β-clamp and inhibit Pol V, significantly inhibited the NA-induced increase in mutation frequency. This supports a Pol V-dependent mutagenesis. Induction of SOS by stavudine was verified by proteomic analysis and metabolite profiling and showed that SOS was not induced by nucleotide starvation. Data presented here suggest that stavudine induced SOS by interfering with the bacterial replications machinery at doses far below the minimal inhibitory concentration (MIC). Biomolecules 2021, 11, 843 3 of 13

2. Materials and Methods 2.1. Bacterial Species, Strains and Growth Measurements E. coli K12 (MG1655) wild type (WT) and TLS polymerase deletion strains ∆umuDC (∆Pol V) from the Keio collection and ∆polB/∆dinB (∆Pol II/∆Pol IV) made in-house as previously described [21] were used. Brieﬂy, polB was inactivated in E. coli MG1655 ∆Pol IV using JW0059: polB. A phage stock (T4GT7) diluted in T4 buffer was added to the JW0059::polB culture, incubated and plated on LB agar plates. The lysate was harvested and added to 2 mL T4 buffer and 0.2 mL CHCl3 (T4GT7 lysate). The mixture was centrifuged and supernatant stored at 4 ◦C, over 20 µL CHCl3. An amount of 0.5 mL overnight culture of the recipient cells was re-suspended in 1 mL of T4 buffer and aliquoted. The T4GT7 lysate (10−1, 10−2, 10−3 and 10−4 dilutions) was added to the aliquots, incubated for 25 minutes at room temperature and plated on LB-agar plates with kanamycin. Colonies obtained were re-streaked twice to obtain a pure culture. Kanamycin resistance cassettes replacing the deleted polB genes were removed by an FLP-recombinase expressed by pCP20 plasmid.

2.2. Peptide A cell-penetrating peptide targeting the β-clamp named APIM-peptide (Ac-MDRWLVK- GILQWRKI-RRRRRRRR-NH2) (Innovagen, Lund, Sweden) was used. This is the same peptide as RWLVK*, used in Nedal et al. [21], except it has three arginine residues less in its cell-penetrating tail.

2.3. Nucleoside Analogues (NAs) NAs used in this study (listed in Table1) were purchased from Glentham Life Sciences Ltd. (Wiltshire, UK).

2.4. Bacterial Cultivation Luria Bertani (LB) broth Miller/Microbiological agar and Mueller Hinton broth/agar were used for bacterial cultivation in the rifampicin mutagenicity (RifR) and MIC assays, respectively. Bacterial cultures were incubated at 37 ◦C, with or without shaking at 250 rpm. Rifampicin (Tokyo chemicals industry Co Ltd., Tokyo, Japan) was used for the RifR assay.

2.5. MIC Assay The MIC of the nucleoside analogues and APIM-peptide were determined following the protocol used for microtiter broth microdilution assays established by the clinical and laboratory standard institutes [22] with modiﬁcations as described [21]. MIC was determined after 24 h.

2.6. Rifampicin Mutagenicity Assay (RifR) Assay This assay was conducted as described [21]. Brieﬂy, pre-cultures of E. coli MG1655 (WT and mutant strains) were inoculated in LB broth (Miller) and grown at 37 ◦C, for 16 h. The pre-cultures were diluted in fresh LB broth (1:100) and 10 mL cultures were grown in 50 mL conical tubes for one hour at 37 ◦C (shaking incubator at 250 rpm), before they were treated with NAs and/or APIM-peptide. Untreated cultures were used as control. The cultures were further incubated for two hours at 37 ◦C, before the cultures were serially diluted in LB and plated on LB agar. Undiluted cultures (800 µL) were plated in LB soft agar containing rifampicin (100 µg/mL) on LB agar plates and incubated for 48 h, at 37 ◦C. The bacterial colony-forming units per millilitre (CFU/mL) were determined from plates with and without rifampicin and the frequency of rifampicin resistance per 108 CFU (RifR/108) was calculated.

2.7. Protein Extraction When harvesting cells for the RifR assay, 2 × 2 mL of untreated cultures and culture treated with stavudine (Stav) (0.32 µg/mL), APIM-peptide (26 µg/mL), or the combination were pelleted for cell extracts (4000 rpm, 10 min) for each biological replica (n = 3). The Biomolecules 2021, 11, 843 4 of 13

pellets were resuspended in an OmniCleave™ storage buffer (Lucigen, Middleton, WI, USA) and incubated at room temperature in the presence of 200 U OmniCleave™ endonu- clease (Lucigen), 20 µg/µL RNAse (Sigma-Aldrich, St. Louis, MO, USA), 10 U DNAse I (Sigma-Aldrich), 1× phosphatase inhibitor cocktails II and III (Sigma-Aldrich, St. Louis, MO, USA) and 1× cOmplete™ protease inhibitor cocktail (Roche Diagnostics GmbH, Basel, Switzerland). The cells were sonicated on ice at 1.5 output control and 15% duty cycle, for 30 s. The sonication was repeated until a clear phase was observed and cell debris was removed by centrifugation (1400 rpm, 15 min). The supernatants were snap-frozen and stored at −80 ◦C. The total protein concentration was measured using NanoDrop One C (Thermo Scientiﬁc, Waltham, MA, USA) and the Bio-Rad protein assay with a UV-1700 visible spectrophotometer (Pharmaspec, Redmond, WA, USA, 595 nm).

2.8. Multiplexed Inhibitor Beads (MIB) Assay and Mass Spectrometry (MS) Analysis of Proteome Before MS analysis of the protein extract, kinases and other ATP/GTP binding proteins were enriched using the MIB assay as described [23]. A total of 100 µg of the protein extract was analysed per sample.

2.9. Metabolite Profiling When harvesting cells for the RifR assay, 4 × 4 mL of untreated culture and culture treated with stavudine (Stav) (0.32 µg/mL), APIM-peptide (26 µg/mL), or the combination were also harvested for MS -based metabolite profiling of central carbon metabolite pools, including glycolytic, pentose phosphate pathway (PPP) and tricarboxylic acid cycle (TCA) intermediates, and the complete nucleoside phosphate pool. Complete sampling, processing and MS-based metabolite profiling protocols can be found in previous publi- cations [24,25]. In short, cells were harvested by fast filtration using a vacuum filtering unit with controlled pressure 800 psi below the ambient pressure. After a quick water rinse, filters were transferred to tubes containing 50% ice-cold acetonitrile (VWR chemicals, city, France) for quenching of metabolism. The tubes were immediately snap-frozen and stored at −80 ◦C. Metabolite extraction was performed through three cycles of freeze-thaw, followed by centrifugation to remove cell debris before the freeze-drying of the cell-free metabolite containing supernatant. Dried supernatants were reconstituted in distilled water, filtered through 3 kDa spin filters (VWR) and analysed on a capillary ion chromatograph (capIC, Thermo Scientific) coupled to a TQ-XS triple quadrupole MS (Waters Corporation, Milford, PA, USA). Samples and analytical grade standards (Sigma-Aldrich, St. Louis, MO, USA) were spiked with yeast extract cultivated with U13C-glucose (Sigma-Aldrich, St. Louis, MO, USA) as the sole carbon source, as described in [25].

2.10. Data Analysis Data from the MIB assay were analysed as described [23]. CapIC-MS/MS data were processed in the TargetLynx application manager of MassLynx 4.1 (Waters Corporation, Milford, PA, USA). Absolute quantification was performed by interpolation of calibration curves prepared from serial dilutions of corresponding analytical grade standards (Sigma- Aldrich), calculated by least squares regression with 1/x weighting. Response factors of the analytical standard and biological extracts were corrected by the corresponding response factor of the U13C-labeled isotopologue spiked into the samples, a strategy enabling the highest quantitative accuracy and precision [26,27]. Principal component analysis (PCA) was performed on auto-scaled metabolite concentrations normalized to the sum, applying MetaboAnalyst 4.0 [28]. Missing values were replaced by the mean concentration of each metabolite. Unpaired, two-tailed Student’s t-tests were used for comparing mutation frequencies; significance levels are described in figure legends.

2.11. Mass Spectrometry Data Analysis Proteins were quantified by processing MS data using MaxQuant v.1.6.17.0 [29]. Open workflow [30] provided in FragPipe version 14 was used to inspect the raw files to deter- Biomolecules 2021, 11, 843 5 of 13

mine optimal search criteria. Namely, the following search parameters were used: enzyme specified as trypsin with a maximum of two missed cleavages allowed; acetylation of protein N-terminal, oxidation of methionine, deamidation of asparagine/glutamine and phosphorylation of serine/threonine/tyrosine as dynamic post-translational modification. These were imported in MaxQuant, which uses m/z and retention time (RT) values to align each run against each other sample with 1 min window match-between-run function and 20 min overall sliding window, using a clustering-based technique. These were further queried against the E. coli (strain K12) proteome including isoforms downloaded from Uniprot (https://www.uniprot.org/proteomes/UP000000625; accessed on 24 August 2020) and MaxQuant’s internal contaminants database using Andromeda built into MaxQuant. Both protein and peptide identifications false discovery rate (FDR) was set to 1%; only unique peptides with high confidence were used for final protein group identification. Peak abundances were extracted by integrating the area under the peak curve. Each protein group abundance was normalized by the total abundance of all identified peptides for each run and protein by calculated median summing all unique and razor peptide-ion abundances for each protein, using label-free quantification (LFQ) algorithm [31] with minimum peptides ≥1. LFQ values for all samples were combined and log-transformed with base 2 and the transformed control values were subtracted. The resulting values reflecting the change relative to control for each condition were subjected to a two-sided non-parametric Wilcoxon signed rank test [32], as implemented in MATLAB R2020a (Math Works Inc., https://www.mathworks.com/, accessed on 18 March 2020), in order to check the consistency in the directionality of the change, namely, a negative sign reflecting decreased and a positive sign reflecting the increased expression of respective protein groups. The choice of this non-parametric test avoids the assumption of a certain type of null distribution as in the Student’s t-test by working over the Rank of the observation instead of the observation value itself. Further, it also makes it robust to outliers and extreme variations noticed in observed values. Differentially expressed (DE) protein groups were identified at p 0.25. The UniProt accession IDs of these DE were mapped to pathways (www.wikipathways.org/index.php/DownloadPathwaysversionwikipathways- 20201010-gmt-Homo_sapiens.gmt) using R (https://www.R-project.org/, accessed on 19 October 2020) libraries, org.Hs.eg.db and clusterProfiler (www.liebertpub.com/doi/10.108 9/omi.2011.0118). Venn diagrams were built using the R package limma [33] and heatmap using pheatmap (https://cran.r-project.org/web/packages/pheatmap/index.html, accessed on 4 January 2020). The raw data with results have been deposited in the Pro- teomeXchange Consortium via the PRIDE [34] partner repository with the dataset identi- fier, project ID PXD025370 (Reviewer account Username: [email protected] Password: seTUpXp4).

3. Results and Discussion 3.1. NAs Induce Mutagenesis in Bacteria at Concentrations That Do Not Affect Bacterial Growth Antibacterial activities of some anti-viral and anti-cancer NAs have been reported previously [19]. Since doses below the MIC of the different NAs must be used to test the mutagenic potential of these drugs, we ﬁrst determined the MICs of the different NAs on E. coli MG1655. Among the NAs tested (Table1), only stavudine, didanosine and 5-FU inhibited visible bacterial growth at concentrations below 256 µg/mL (maximal concentration examined). The lack of antibacterial activities of several of these NAs, e.g., cytarabine, vidarabine, ﬂudarabine, abacavir, emtricitabine, lamivudine and tenofovir, are in agreement with published data [35,36]. Biomolecules 2021, 11, x 6 of 14

Biomolecules 2021, 11, x 6 of 14

Biomolecules 2021, 11, x tration examined). The lack of antibacterial activities of several of these NAs, e.g., cytara-6 of 14 bine, vidarabine, fludarabine, abacavir, emtricitabine, lamivudine and tenofovir, are in trationagreement examined). with published The lack data of antibacterial [35,36]. activities of several of these NAs, e.g., cytarabine, vidarabine, fludarabine, abacavir, emtricitabine, lamivudine and tenofovir, are in tration examined). The lack of antibacterial activities of several of these NAs, e.g., cytara- agreementTable 1. Nucleoside with published analogues data (NAs) [35,36]. and their minimal inhibitory concentrations (MICs) in Esche- richiabine, colividarabine, MG1655. fludarabine, abacavir, emtricitabine, lamivudine and tenofovir, are in agreement with published data [35,36]. Table 1. Nucleoside analogues (NAs) and their minimal inhibitory concentrations (MICs) in Esche- Biomolecules 2021, 11, x richia coli MG1655. 6 of 14 Nucleoside MIC Table 1. NucleosideReported analogues Plasma (NAs) Concen- and their minimal inhibitory concentrations (MICs) in Esche- Biomolecules 2021, 11, x richiaAnalogue coli MG1655. Structure Use 7 of 15 Biomolecules 2021, 11, x tration (Pl.CR.) (µg/mL) 7 of 15 Nucleoside(NA) (µg/mL)MIC tration examined).Reported The Plasmalack of antibacterial Concen- activities of several of these NAs, e.g., cytara- Biomolecules 2021, 11, x Analogue Structure Use 7 of 15 Nucleosidebine, vidarabine, tration fludarabine, (Pl.CR.) (µg/mL)abacavir, emtricMICitabine, lamivudine and tenofovir, are in (NA) Reported Plasma Concen- (µg/mL) Biomolecules 2021, 11, x agreementAnalogue with published data [35,36]. Structure Use 7 of 15 Stavudine tration (Pl.CR.)≤0.9 (µg/mL) Nucleoside(NA) Reported Plasma MIC(µg/mL) HIV, NRTI. dTTP ana- Nucleoside Reported Plasma MIC 64 AnalogueTable 1. Nucleoside Concentration analogues (Pl.CR.) (NAs) and their minimal inhibitoryStructure concentrations logue (MICs)Use in Esche- AnalogueStavudine(NA)(Stav) Concentration (µg/mL)≤[37] (Pl.CR.)0.9 (µg/mL) Structure Use Biomolecules 2021, 11, 843 Nucleosiderichia coli MG1655. Reported Plasma MIC HIV, NRTI. dTTP ana-6 of 13 (NA) (µg/mL) (µg/mL) 64 NucleosideAnalogueStavudine Concentration Reported Plasma≤ 0.9(Pl.CR.) MIC Structure logueUse (NA)(Stav) (µg/mL)[37] (µg/mL) HIV, NRTI. dTTP ana- AnalogueNucleoside Concentration (Pl.CR.) MIC64 Structure Use StavudineDidanosine Reported≤0.90.05–0.86 Plasma Concen- logue (NA)Analogue(Stav) (µg/mL)[37] (µg/mL) Structure HIV NRTI.HIV,Use dATP NRTI. ana- Table 1. NucleosideStavudine analogues (NAs) andtration their≤0.9 (Pl.CR.) minimal inhibitory(µg/mL)64 concentrations8 (MICs) in Escherichia coli MG1655. (NA) (µg/mL) dTTPHIV,logue analogue NRTI. (Stav)(Dida) [37] [38] 64 Nucleoside Analogue StavudineDidanosineReported Plasma≤0.90.05–0.86 dTTP analogue (Stav) [37] MIC (µg/mL) StructureHIV NRTI.HIV, dATP NRTI. Use ana- (NA) Concentration (Pl.CR.) (µg/mL) 64 8 StavudineDidanosine ≤0.90.05–0.86 dTTPlogue analogue (Stav)(Dida) [37] [38] HIV NRTI.HIV, dATP NRTI. ana- Stavudine ≤0.9 64 8 Stavudine DidanosineEmtricitabine ≤0.9 0.05–1.01–11.50.86 HIV, NRTI.dTTPHIV,logue analoguedTTP NRTI. ana- dTTP (Stav)(Dida) [37] [38] 64 64 HIV NRTI.HIV dCTP NRTI. ana- (Stav) Didanosine [37] 0.05–0.86 8 >256 logueanalogue (Stav) [37] dATPHIVlogue analogue NRTI. (Dida)(Emtri) [38] [39] 8 DidanosineEmtricitabine 0.05–1.01–11.50.86 dATP analogue (Dida) [38] HIV NRTI.HIV dCTP NRTI. ana- Didanosine 0.05–0.86 8 >256 Emtricitabine 1.01–11.5 dATPHIVlogue analogue NRTI. Didanosine (Dida)(Emtri) 0.05–0.86 [38] [39] 8 HIV NRTI.HIV NRTI.dCTP dATPana- (Dida) Didanosine[38 ] 0.05–0.86 8 >256 analogue EmtricitabineTenofovir 1.010.050–0.077–11.5 HIV NRTI.dATPlogue dATPanalogue ana- (Dida)(Emtri) [38] [39] 8 HIV NRTI.HIV dTTP NRTI. ana- Emtricitabine 1.01–11.5 >256 >256 logue (Dida) [38] dCTPHIVlogue analogue NRTI. Emtricitabine(Emtri)Tenofovir(Teno) 1.01[39]0.050–0.077–11.5 [40] >256 HIV NRTI.dCTPHIV dTTP analogueNRTI. ana- Emtricitabine (Emtri) 1.01–11.5 [39] >256>256 >256 HIV NRTI. dCTP (Emtri) EmtricitabineTenofovir [ 39] 1.010.050–0.077–11.5 dCTPlogueanalogue analogue (Emtri)(Teno) [39] [40] HIV NRTI.HIV dTTPNRTI. ana- Emtricitabine 1.01–11.5 >256 >256 Tenofovir 0.050–0.077 HIV NRTI.dCTPlogue dCTPanalogue ana- (Emtri)Abacavir(Teno) [39]0.04–1.13 [40] >256 HIV NRTI. Tenofovir 0.050–0.077 >256 HIV NRTI.logue dGTP ana- Tenofovir (Emtri)0.050–0.077 [39] >256 dTTPHIVHIV analogue NRTI. NRTI. dTTP Tenofovir(Teno) 0.050[40]–0.077 >256>256 logue (Teno) Abacavir(Abaca) [40] 0.04–1.13[41] dTTPHIVanalogue analogueNRTI. (Teno) [40] >256 HIV NRTI. dGTP ana- Tenofovir 0.050–0.077 >256 dTTP analogue (Teno)Abacavir [40]0.04–1.13 HIVlogue NRTI. (Abaca)Tenofovir 0.050–0.077[41] >256 HIV NRTI. dGTP ana- Abacavir 0.04–1.13 >256 HIV NRTI.dTTP analoguedTTP ana- Lamivudine(Teno) [40]0.02–0.11 >256 HIVlogue NRTI. Abacavir (Abaca)0.04–1.13 [41] >256>256 HIV-1 NRTI.HIVlogue NRTI. dTTP dGTP ana- (Abaca) Abacavir(Teno) [41] 0.04–1.13[40] >256 dGTPanalogue analogue (Abaca) [41] HIVlogue NRTI. Abacavir(Lamu) 0.04–1.13[42] >256 Lamivudine 0.02–0.11 dGTPHIV analogueNRTI. (Abaca) [41] >256 HIV-1 NRTI. dTTP ana- Abacavir 0.04–1.13 >256 dGTP analogue (Abaca)Lamivudine [41]0.02–0.11 HIVlogue NRTI. Biomolecules 2021, 11, x (Lamu) [42] >256 HIV-1 NRTI. dTTP7 ana-of 14 Lamivudine Abacavir0.02–0.11 0.04–1.13 >256 >256 dGTPHIV-1 analogue NRTI. dTTP Biomolecules 2021(Lamu), 11, x Lamivudine(Abaca)Acyclovir [ 42] 0.02[41]–0.11 10.0 HIV NRTI. analoguedGTP ana-7 of 14 >256 HIV-1logue NRTI. (Lamu) [42] >256 logue Lamivudine(Abaca) 0.02–0.11[41] >256 dTTP analogue Biomolecules 2021, 11, x (Lamu) [42] HIV-1 NRTI.7 of 14 LamivudineAcyclovir(Acyc) 0.02–0.1110.0[43] >256 Acyclovir 10.0 Herpesviruses.dTTPHIV-1 analogue NRTI. dGTP Lamivudine(Lamu) [43] 0.02[42]–0.11 >256>256 Biomolecules 2021(Acyc), 11, x Vidarabine 0.2–0.4 >256 dTTPanalogue analogue7 of 14 (Lamu)Acyclovir [42] 10.0 Herpesviruses.HIV-1 NRTI.dATP ana- VidarabineLamivudine(Acyc) 0.02–0.110.2–0.4[43] >256 >256 Herpesviruses. dGTP ana- Biomolecules 2021, 11, x Acyclovir 10.0 >256 Herpesviruses.HIV-1 dTTPNRTI.logue analogue dTTP dATP 7ana- ana-of 14 (Lamu)Idoxuridine(Vida) [42]10.0–36.0[45,46] >256>256 HerpesviruseHerpesviruses.Herpesviruses.logues. dUTP dUTPana- Idoxuridine (Acyc)10.0–36.0 [43] >256 logue AcyclovirVidarabine 10.00.2–0.4 >256 >256 logue.analogue. ThymidylatedGTPlogue analogue Thymidylate phos- (Idox) (Acyc)(Vida)(Lamu) [ 44] [43][45,46] [42] Herpesviruses. Herpesviruses. dATP ana- Acyclovir(Idox) 10.0 [44] >256 >256 phosphatase inhibitor. Idoxuridine 10.0–36.0 HerpesvirusephatasedGTPHerpesviruses.logue inhibitor. s.analogue dUTP ana- Acyclovir(Acyc)Vidarabine(Vida) [43]10.00.2–0.4[45,46] >256 Entecavir 0.015–0.036 >256 Herpesviruses.logue. ThymidylatedGTPHerpesviruses. analogue dATP phos- ana- Idoxuridine(Acyc) [43]10.0–36.0 >256 >256 Herpesviruses.Hepatitis virus. dUTPdGTP ana-an- VidarabineAcyclovir(Idox) 0.2–0.4[44]10.0 >256 phatasedGTPHerpesviruses.logue inhibitor. analogue Vidarabine (Acyc)Entecavir 0.2–0.4 [43]0.015–0.036 >256 Herpesviruses.logue. ThymidylateHerpesviruses. dGTPdATP phos- dATPana-ana- (Vida) [45,46] >256 >256>256 Hepatitis aloguevirus. dGTP an- (Vida) (Idox)(Ente) [ 45,46] [44][47] >256 phataselogueloguedUTP inhibitor.analogue Idoxuridine(Vida)(Acyc) 10.0–36.0[45,46][43] Herpesviruses.alogue Entecavir(Ente) 0.015–0.036[47] Herpesviruses.analogue. >256 Hepatitis virus.dUTP dGTP an- Idoxuridine 10.0–36.0 >256 ThymidylatedUTP (Idox) [44] Herpesviruses.alogueanalogue. IdoxuridineEntecavir(Ente) 10.00.015–0.036–36.0[47] Entecavir 5-Fluoro0.015–0.036 Ura- >256 HepatitisHepatitisphosphatase analogue.virus.dUTP virus.dGTP dGTP an- 0.025–8.0 >256>256 Herpesviruses.Thymidylate dUTP ana- (Ente) Idoxuridine(Idox)IdoxuridineEntecavir cil [ 47 ] 10.0[44]0.015–0.036–10.0–36.036.0 >256 Multiple cancers.inhibitor.analogue Thymi- 5-Fluoro Ura- 4 HepatitisThymidylate alogueanalogue.virus. dGTP an- (Idox)(Ente) [44]0.025–8.0 [47] >256 >256>256 logue. Thymidylatephosphatase phos- cil Multipledylate synthaseThymidylatephosphatase cancers. inhibitor Thymi- (Idox)(Idox) [44] [48][44] 4 phatasealogueinhibitor. inhibitor. 5-Fluoro(Ente)(5-FU) Ura- [47] dylate synthaseinhibitor. inhibitor 0.025–8.0[48] phosphataseMultiple cancers. cil 0.025–8.0 Multiple cancers. Thymi- 5-Fluoro Uracil (5-FU) 4 4 Thymidylateinhibitor. synthase (5-FU) 5-Fluoro Ura-[48] dylate synthase inhibitor 0.025–8.0[48] inhibitor 5-FluoroFludarabine(5-FU)cil Ura- 0.4–2.0 Multiple cancers. Thymi- 0.025–8.0 4 Leukaemia and lym- cil >256 Multipledylate synthase cancers. inhibitor Thymi- Fludarabine 0.4–2.0[48] Leukaemia and (5-FU) 0.4–2.0 4 phoma.Leukaemia dATP and analogue lym- Fludarabine (Fluda) [49] >256 dylate synthaselymphoma. inhibitor dATP (Fluda) [49] [48] >256 Fludarabine(5-FU) 0.4–2.0 phoma. dATPanalogue analogue (Fluda) [49] Leukaemia and lym- >256 Fludarabine 0.4–2.0 phoma. dATP analogue Cytarabine(Fluda) ~486.0[49] Leukaemia and lym- Cytarabine ~486.0 Leukaemia.Leukaemia. dCTP ana- dCTP Fludarabine 0.4–2.0 >256 >256>256 (Cyta) Cytarabine[50 ] ~486.0 phoma.Leukaemia dATPlogueanalogue and analogue lym- (Fluda)(Cyta) [49][50] >256 Leukaemia. dCTP ana- >256 phoma. dATP analogue Cytarabine(Fluda) ~486.0[49] logue (Cyta) [50] Leukaemia. dCTP ana- >256 logue Cytarabine(Cyta)To examine the mutagenic~486.0[50] potential of the different NAs, E. coli MG1655 was treated Leukaemia. dCTP ana- with 0.5× MIC or 0.25× of the maximal concentration>256 used in the MIC assay (256 μg/mL). CytarabineTo examine the mutagenic~486.0 potential of the different NAs, E. coli MG1655logue was treated The(Cyta) NRTIs stavudine, didanosine[50] and emtricitabine induced a strongLeukaemia. (45–125×) dCTP increase ana- with 0.5× MIC or 0.25× of the maximal concentration>256 used in the MIC assay (256 μg/mL). in mutation frequency (Figure 1A). Abacavir and lamivudine, two other NRTIs,logue also sig- The (Cyta)NRTIsTo examine stavudine, the mutagenic didanosine[50] potential and emtricitabine of the different induced NAs, aE. strong coli MG1655 (45–125×) was increase treated nificantly increased the mutation frequency (~10–15×). 5-FU and fludarabine, two NAs inwith mutation 0.5× MIC frequency or 0.25× (Figureof the maximal 1A). Abacavir concentration and lamivudine, used in the two MIC other assay NRTIs, (256 alsoμg/mL). sig- commonly used in anticancer therapy, increased the mutation frequency by more than 50 nificantlyThe NRTIsTo examine increased stavudine, the themutagenic didanosine mutation potential frequencyand emtricitabine of the (~10–15×) different induced. 5-FUNAs, aandE. strong coli fludarabine, MG1655 (45–125×) was two increase treated NAs times. The sub-MIC doses used in Figure 1A are in all cases except for 5-FU higher than inwith mutation To0.5× examine MIC frequency or the 0.25× mutagenic (Figureof the maximal 1A). potential Abacavir concentration of the and different lamivudine, used NAs, in the E.two coliMIC other MG1655 assay NRTIs, (256 was alsoμ treatedg/mL). sig- commonlyplasma concentrations used in anticancer obtained therapy, in patients increased (see theTable mutation 1). We frequencytherefore examinedby more than doses 50 nificantlyThewith NRTIs 0.5× MICincreased stavudine, or 0.25× the didanosineof mutation the maximal frequencyand concentrationemtricitabine (~10–15×) inducedused. 5-FU in theaand strong MIC fludarabine, assay(45–125×) (256 two μincreaseg/mL). NAs times.that are The in sub-MIC the human doses plasma used concentrations in Figure 1A arerange in (Pl.CR)all cases (Figure except 1B) for and5-FU found higher that than in commonlyin mutation used frequency in anticancer (Figure therapy, 1A). Abacavir increased and the lamivudine, mutation frequencytwo other byNRTIs, more also than sig- 50 plasmaTheaddition NRTIs concentrations to stavudine, 5-FU, stavudine, didanosineobtained didanosine, in and patients emtricitabine emtricitabine (see Table induced 1). and We fludarabine a therefore strong (45–125×) examinedwere potent increase doses in- times.nificantly The increased sub-MIC dosesthe mutation used in frequencyFigure 1A (~10–15×)are in all .cases 5-FU except and fludarabine, for 5-FU higher two NAsthan thatinducers mutation are inof themutagenesis frequency human plasma (Figure at clinically concentrations 1A). relevantAbacavir doses. rangeand lamivudine, Thus, (Pl.CR) these (Figure twodrugs other1B) may and NRTIs, increase found also thatbacte- sig- in plasmacommonly concentrations used in anticancer obtained therapy, in patients increased (see theTable mutation 1). We frequencytherefore examinedby more than doses 50 additionnificantlyrial mutagenesis to increased 5-FU, andstavudine, the contribute mutation didanosine, to frequency hard-to-treat emtricitabine (~10–15×) bacterial. and5-FU infections fludarabine and fludarabine, when were used potent clinically.two NAs in- thattimes. are The in thesub-MIC human doses plasma used concentrations in Figure 1A rangeare in (Pl.CR)all cases (Figure except 1B) for and5-FU found higher that than in ducerscommonlyIt still ofremains mutagenesis used elusivein anticancer at why clinically thesetherapy, relevantNAs increased are doses.more the mutagenicThus, mutation these than frequencydrugs the may other by increase moreNAs thantested. bacte- 50 additionplasma concentrations to 5-FU, stavudine, obtained didanosine, in patients emtricitabine (see Table 1).and We fludarabine therefore examinedwere potent doses in- rialtimes.Stavudine mutagenesis The sub-MICwas the and most doses contribute potent used inducertoin hard-to-treatFigure of 1A muta are genesis.bacterial in all cases A infections more except than whenfor 50× 5-FU increase used higher clinically. in mu-than ducersthat are of in mutagenesis the human plasma at clinically concentrations relevant doses. range Thus, (Pl.CR) these (Figure drugs 1B) may and increase found thatbacte- in Itplasmatation still remains frequencyconcentrations elusive was detected whyobtained these at in0.016 NAs patients μ areg/mL, more(see a doseTable mutagenic 20-fold 1). We lowerthan therefore thethan other theexamined reported NAs tested. doses up- rialaddition mutagenesis to 5-FU, and stavudine, contribute didanosine, to hard-to-treat emtricitabine bacterial and infections fludarabine when were used potentclinically. in- Stavudinethatper arelevels in was theof Pl. humanthe CR most (0.32 plasma potent μg/mL). concentrations inducer That ofNAs muta increaserangegenesis. (Pl.CR) the A mutation more (Figure than 1B)frequency 50× and increase found in bacteria inthat mu- in ducers of mutagenesis at clinically relevant doses. Thus, these drugs may increase bacte- tationItadditionmay, still frequencyatremains leastto 5-FU, partly, elusive was stavudine, explaindetected why these whydidanosine, at 0.016 a NAs high μ g/mL, arefraction emtricitabine more a dose of mutagenic antibiotic-resistant 20-fold and fludarabine lowerthan thethan bacteriaother the were reported NAs ispotent isolated tested. up- in- rial mutagenesis and contribute to hard-to-treat bacterial infections when used clinically. perStavudineducersfrom levels HIV-positive of mutagenesis ofwas Pl. the CR most patients(0.32 at potent μclinicallyg/mL). [13–16,35]. inducer That relevant NAsof muta doses. increasegenesis. Thus, the A thesemutation more drugs than frequency may50× increase increase in bacteria in bacte- mu- may,tationItrial still mutagenesis at frequencyremains least partly, elusive wasand explain detectedcontribute why whythese at to 0.016a NAshard-to-treathigh μ g/mL,fractionare more a dosebacterialof mutagenic antibiotic-resistant 20-fold infections lowerthan the thanwhen bacteriaother the used reportedNAs is clinically. isolated tested. up- fromperStavudineIt still levels HIV-positive remains ofwas Pl. theelusive CR patientsmost (0.32 why potent μ g/mL).[13–16,35]. these inducer ThatNAs NAsofare muta more increasegenesis. mutagenic the A mutation more than than thefrequency 50× other increase NAs in bacteria intested. mu- may,tationStavudine at frequency least was partly, the was most explain detected potent why at inducer 0.016a high μ ofg/mL,fraction muta a genesis.doseof antibiotic-resistant 20-fold A more lower than than 50× bacteria the increase reported is isolated in mu- up- fromtationper levels HIV-positive frequency of Pl. CRwas patients (0.32 detected μ g/mL).[13–16,35]. at 0.016 That μ NAsg/mL, increase a dose 20-foldthe mutation lower thanfrequency the reported in bacteria up- may,per levels at least of Pl.partly, CR (0.32explain μg/mL). why a That high NAsfraction increase of antibiotic-resistant the mutation frequency bacteria inis bacteriaisolated frommay, HIV-positiveat least partly, patients explain [13–16,35]. why a high fraction of antibiotic-resistant bacteria is isolated from HIV-positive patients [13–16,35]. Biomolecules 2021,, 11,, 843x 7 7of of 13

To examine the mutagenic potential of the different NAs, E. coli MG1655 was treated with 0.50.5×× MICMIC or or 0.25× 0.25× ofof the the maximal maximal concentration concentration used used in in the the MIC MIC assay (256 µμg/mL). The NRTIs stavudine, didanosine andand emtricitabineemtricitabine inducedinduced aa strongstrong (45–125(45–125×)×) increase inin mutation mutation frequency frequency (Figure (Figure 11A).A). AbacavirAbacavir andand lamivudine,lamivudine, two other NRTIs, alsoalso sig-sig- nificantlynificantly increased the mutationmutation frequencyfrequency (~10–15(~10–15×)×).. 5-FU 5-FU and and fludarabine, fludarabine, two NAs commonly used used in in anticancer anticancer therapy, therapy, increa increasedsed the the mutation mutation frequency frequency by bymore more than than 50 times.50 times. The The sub-MIC sub-MIC doses doses used used in inFigure Figure 1A1A are are in in all all cases cases except except for 5-FU higherhigher thanthan plasma concentrations obtained obtained in in patients (see Table 11).). WeWe thereforetherefore examinedexamined dosesdoses thatthat are are in in the the human human plasma plasma concentrations concentrations range range (Pl.CR) (Pl.CR) (Figure (Figure 1B)1B) and and found found that that in additionin addition to 5-FU, to 5-FU, stavudine, stavudine, didanosine, didanosine, emtricitabine emtricitabine and and fludarabine fludarabine were were potent potent in- ducersinducers of ofmutagenesis mutagenesis at atclinically clinically relevant relevant doses. doses. Thus, Thus, these these drugs drugs may may increase increase bacte- bacte- rialrial mutagenesis mutagenesis and and contribute contribute to to hard-to-treat hard-to-treat bacterial bacterial infections infections when when used used clinically. clinically. ItIt still still remains elusive elusive why why these these NAs NAs are are more more mutagenic mutagenic than than the the other other NAs NAs tested. tested. Stavudine was was the the most most potent potent inducer inducer of of muta mutagenesis.genesis. A more A more than than 50× 50increase× increase in mu- in tationmutation frequency frequency was was detected detected at 0.016 at 0.016 μg/mL,µg/mL, a dose a dose 20-fold 20-fold lower lower than than the reported the reported up- perupper levels levels of ofPl. Pl. CR CR (0.32 (0.32 μg/mL).µg/mL). That That NAs NAs increase increase the the mutation mutation frequency frequency in in bacteria may, at least partly, explainexplain whywhy aa highhigh fraction of antibiotic-resistant bacteria is isolated fromfrom HIV-positive patients [[13–16,35].13–16,35].

Figure 1. NucleosideNucleoside analogues (NAs) induce mutagenesis in Escherichia coli MG1655. DataData represents represents the average increase in mutationmutation frequencyfrequency ininE. E. coli coliMG1655 MG1655 cells cells detected detected in in the the Rif RifR assayR assay after after treatment, treatment, compared compared to untreatedto untreated control. control. (A) Mutation(A) Mutation frequencies frequencies determined determined using using 0.5× MIC0.5× MIC or 0.25 or× 0.25×of maximal of maximal examined examined concentration concentration (256 µ (256g/mL) μg/mL) (Table (Table1) of the 1) differentof the different NAs. The NAs. relative The relati increaseve increase in mutation in mutation frequency frequency is given is from given 5 independent from 5 independent biological biological replicas. Stavudinereplicas. Stavu- (Stav, dine (Stav, 32 μg/mL), didanosine (Dida, 4 μg/mL), emitrcitabine (Emtri, 64 μg/mL), tenofovir (Teno, 64 μg/mL), abacavir 32 µg/mL), didanosine (Dida, 4 µg/mL), emitrcitabine (Emtri, 64 µg/mL), tenofovir (Teno, 64 µg/mL), abacavir (Abaca, (Abaca, 64 μg/mL), lamuvudine (Lamu, 64 μg/mL, acyclovir (Acyc, 64 μg/mL), idocuridine (Idox, 64 μg/mL), vidarabine 64 µg/mL), lamuvudine (Lamu, 64 µg/mL, acyclovir (Acyc, 64 µg/mL), idocuridine (Idox, 64 µg/mL), vidarabine (Vida, (Vida, 64 μg/mL), entecavir (Ente, 64 μg/mL), 5-fluro uracil (5-FU, 2 μg/mL), fludarabine, (Fluda, 64 μg/mL) and cytarabine µ µ µ µ 64(Cyta,g/mL), 64 μg/mL). entecavir (B (Ente,) Mutation 64 g/mL), frequency 5-ﬂuro at uracilplasma (5-FU, concentrations 2 g/mL),ﬂudarabine, of Stav (0.32 (Fluda, μg/mL 64 andg/mL) 0.016 and μg/mL), cytarabine Dida (Cyta, (0.5 64μg/mL),µg/mL). Emtri (B )(10 Mutation μg/mL) frequency and Fluda at (1.1 plasma μg/mL). concentrations The relative of increase Stav (0.32 in mutationµg/mL and frequency 0.016 µ g/mL),is given Dida from (0.5a minimumµg/mL), Emtriof 3 independent (10 µg/mL) biological and Fluda replicas, (1.1 µ exceptg/mL). for The 0.32 relative μg/mL increase Stav, which in mutation are from frequency14 independent is given biological from a replicas. minimum Each of 3biological independent replica biological consists replicas,of 2–3 technical except forreplicates. 0.32 µg/mL Unpaired Stav, whichStudent’s are t-test, from 14two-tailed, independent ****p < biological 0.0001, *** replicas.p < 0.001, Each**p < biological0.01 and *p replica< 0.05. consists of 2–3 technical replicates. Unpaired Student’s t-test, two-tailed, **** p < 0.0001, *** p < 0.001, ** p < 0.01 and * p < 0.05. 3.2. Pol V is the TLS Polymerase Responsible for the Mutagenesis Induced by NAs 3.2. PolTo Vdetermine Is the TLS which Polymerase TLS polymerase Responsible is for responsible the Mutagenesis for the Induced induced by NAsmutagenesis, the mutationTo determine frequency which in WT, TLS a strain polymerase with deletion is responsible of Pol forV (Δ thePol induced V) and mutagenesis,a Pol IV and Pol the IImutation double deletion frequency strain in WT, (ΔPol a strain II/ΔPol with IV) deletion of E. coli of MG1655 Pol V ( ∆werePol V)examined and a Pol after IV treatment and Pol II withdouble stavudine. deletion The strain results (∆Pol clearly II/∆Pol indicate IV) of E.that coli theMG1655 main TLS were polymerase examined responsible after treatment for the increase in mutation frequency is Pol V and that the deletion of both the other two TLS Biomolecules 2021, 11, x 8 of 13

polymerases in E. coli., Pol IV and Pol II, did not significantly reduce the mutation fre- Biomolecules 2021, 11, 843 quency (Figure 2A). Pol V is considered to be the main SOS inducible8 TLS of 13 polymerase in bacteria (reviewed in [51]). We have previously shown that peptides containing the PCNA interacting motif withAPIM, stavudine. bind to The the results β-clamp clearly and indicate block that the the mainfunction TLS polymeraseof Pol V [21]. responsible Here forwe show that the theAPIM-peptide increase in mutation was able frequency to reduce is Pol Vthe and muta thattion the deletion frequency of both induced the other by two stavudine when TLS polymerases in E. coli., Pol IV and Pol II, did not signiﬁcantly reduce the mutation these were combined (combination) (Figure 2B), further supporting a Pol V-dependent frequency (Figure2A). Pol V is considered to be the main SOS inducible TLS polymerase inmutagenesis. bacteria (reviewed in [51]).

FigureFigure 2. 2.Stavudine-induced Stavudine-induced mutagenesis mutagenesis in Escherichia in Escherichia coli MG1655 coli is MG1655 driven by is the driven translesion by the translesion R synthesissynthesis (TLS) (TLS) Pol V.Pol (A V.) Data (A) represent Data represent the average the increase average in mutationincrease frequency in mutation using frequency the Rif using the µ assayRifR assay after treatment after treatment of stavudine of stavudine (Stav, 0.32 (Stav,g/mL), 0.32 compared μg/mL), to untreatedcompared control to untreated for E. coli control for E. coli MG1655 wild type (WT) and TLS polymerase deletion strains (∆Pol V and ∆Pol II/Pol IV). (B) MG1655 wild type (WT) and TLS polymerase deletion strains (ΔPol V and ΔPol II/Pol IV). (B) Data Data represent the average increase in mutation frequency using the RifR assay after treatment of represent the average increase in mutation frequency using the RifR assay after treatment of stavu- stavudine (Stav, 0.32 µg/mL), APIM-peptide (26 µg/mL) and a combination (Stav + APIM-peptide), compareddine (Stav, to untreated0.32 μg/mL), control APIM-peptide for E. coli MG1655 (26 wild μg/mL) type (WT).and a The combination mutation frequency (Stav + fromAPIM-peptide), thecompared stavudine-treated to untreated WT cells control (WT andfor E. Stav) coli are MG1655 from 14 wild independent type (WT). biological The replicas.mutation The frequency from mutationthe stavudine-treated frequencies from WT the remainingcells (WT samples and Stav) (∆Pol ar Ve andfrom∆Pol 14 II/Pol independent IV, APIM-peptide biological and replicas. The combination)mutation frequencies are from 5 independent from the remaining biological replicas, samples where (ΔPol 3 technical V and replicatesΔPol II/Pol were IV, plated APIM-peptide and fromcombination) each biological are replicate.from 5 independent Unpaired Student’s biologicalt-test, two-tailed, replicas, **** wherep < 0.0001 3 technical and * p < replicates 0.05. were plated from each biological replicate. Unpaired Student’s t-test, two-tailed, ****p < 0.0001 and *p < 0.05 We have previously shown that peptides containing the PCNA interacting motif APIM, bind to the β-clamp and block the function of Pol V [21]. Here we show that the APIM- peptide3.3. Stavudine was able toTreatment reduce the Induces mutation only frequency Minor induced Changes by in stavudine Central when Carbon these Metabolite were Pools combinedStavudine (combination) is a thymidine (Figure2B), furthernucleoside supporting anal aogue, Pol V-dependent which when mutagenesis. incorporated into DNA

3.3.by Stavudinereverse Treatmenttranscriptase, Induces terminates only Minor Changes the replic in Centralation Carbon of the Metabolite virus. PoolsPyrimidine synthesis is strictly regulated and some NAs, such as 5-FU, are also thymidylate synthase inhibitors. Stavudine is a thymidine nucleoside analogue, which when incorporated into DNA by reverseThus, transcriptase,NAs may affect terminates the nucleotide the replication pools of the and virus. th Pyrimidineis may both synthesis directly is strictly and indirectly stress regulatedthe cells and and some increase NAs, such the as mutation 5-FU, are also frequency. thymidylate To synthase explore inhibitors. the effects Thus, of NAs stavudine on the maynucleotide affect the pools, nucleotide quantitative pools and this MS-based may both directlymetabolite and indirectly profiling stress of theE. coli cells treated and with stavu- increasedine and the APIM-peptide mutation frequency. alone To explore and the the combin effects ofation stavudine of stavudine on the nucleotide and APIM-peptide pools, (com- quantitative MS-based metabolite profiling of E. coli treated with stavudine and APIM- peptidebination) alone were and theperformed. combination PCA of stavudine did not and reveal APIM-peptide significant (combination) differences were in the metabolite performed.pool composition PCA did not of reveal the treated significant cells differences and the in theuntreated metabolite control pool composition (95% confidence of interval, theFigure treated 3A). cells However, and the untreated the combination control (95% confidence treatment interval, clustered Figure closer3A). However, to the APIM-peptide thesingle combination treatment treatment than the clustered stavudine closer to single the APIM-peptide treatment single(Figure treatment 3A), indicating than the a larger sim- stavudine single treatment (Figure3A), indicating a larger similarity between the two. Thisilarity is also between evident the from two. inspecting This is fold also changes evident on thefrom level inspecting of individual fold metabolites; changes on the level of theindividual central carbon metabolites; metabolite the pool central in APIM-peptide carbon me andtabolite combination-treated pool in APIM-peptideE. coli were and combina- moretion-treated similar and, E. morecoli were specifically, moremore similar down-regulated, and, more comparedspecifically, to untreated more down-regulated, control compared to untreated control than stavudine single treatment (Figure 3B). Especially noteworthy is the slight increase in pyrimidine nucleoside pools in stavudine single treated cells, as opposed to down-regulated pools in both in the APIM-peptide- and the combination-treated group. Down-regulation of nucleoside pools after treatment with APIM- Biomolecules 2021, 11, 843 9 of 13

than stavudine single treatment (Figure3B). Especially noteworthy is the slight increase in Biomolecules 2021, 11, x pyrimidine nucleoside pools in stavudine single treated cells, as opposed to down-regulated9 of 13 pools in both in the APIM-peptide- and the combination-treated group. Down-regulation of nucleoside pools after treatment with APIM-peptide has also been also detected in mammalian cancer cells [52]. However, the increased mutation frequency observed by peptide has also been also detected in mammalian cancer cells [52]. However, the in- stavudine as a single agent cannot be explained by cellular stress induced by the small creased mutation frequency observed by stavudine as a single agent cannot be explained increase in nucleotide pools detected. by cellular stress induced by the small increase in nucleotide pools detected.

Figure 3. Stavudine treatment of Escherichia coli MG1655 leads to small changes in glycolytic, PPP and TCAFigure intermediates 3. Stavudine and treatment in nucleoside of Escherichia phosphate coli MG1655 pools. (A leads) Scores to small plot (PC1 changes and in PC2) glycolytic, from PCA PPP of intracellularand TCA intermediates metabolite concentrations, and in nucleoside as measured phosphate by pools. capIC-MS/MS, (A) Scores for plot the (PC1 four and treatment PC2) from groups: PCA of intracellular metabolite concentrations, as measured by capIC-MS/MS, for the four treat- stavudine (Stav, 0.32 µg/mL), APIM-peptide (26 µg/mL) and combination (Stav + APIM-peptide) ment groups: stavudine (Stav, 0.32 μg/mL), APIM-peptide (26 μg/mL) and combination (Stav + and untreated control harvested two hours after treatment, with 95% confidence regions indicated. APIM-peptide) and untreated control harvested two hours after treatment, with 95% confidence (B) Log fold change of treated cultures vs. untreated control. Data are averages of four independent regions2 indicated. (B) Log2 fold change of treated cultures vs. untreated control. Data are averages biologicalof four independent replicas normalized biological by replicas sum. normalized by sum. 3.4. Stavudine Treatment Activates the SOS System 3.4. Stavudine Treatment Activates the SOS System When harvesting samples for mutation frequency and metabolite profiling, cells wereWhen also harvested harvesting for samples proteome for analysis mutation using frequency the MIB and assay metabolite [23]. This profiling, is an assay cells wherewere also activated harvested signalling for proteome proteins, analysis and proteins using the binding MIB assay to these, [23]. This are pulledis an assay down where via theiractivated affinity signalling for ATP/GTP proteins, using and kinase proteins inhibitors binding (also to these, NAs) are immobilized pulled down on via sepharose their af- beads.finity for The ATP/GTP proteins using attached kinase are trypsinizedinhibitors (also and NAs) identified immobilized by MS/MS. on sepharose Approximately beads. 1000The proteins wereattached pulled are down trypsinized from each and sample. identified During by MS/MS. analysis, Approximately only proteins that1000 were pro- similarlyteins were changed pulled down relative from to untreated each sample. control During in all analysis, three biological only proteins replicas that (significant were simi- change,larly changed Wilcoxon relative test [ 23to]) untreated were included control in downstreamin all three biological analysis. Morereplicas proteins (significant were similarlychange, Wilcoxon changed intestE. [23]) coli treatedwere included with stavudine in downstream and the analysis. combination, More than proteins in APIM- were peptide-treatedsimilarly changedE. coli in comparedE. coli treated to either with of stavudine the two groups and the (Figure combination,4A). This than suggests in APIM- that thepeptide-treated APIM-peptide E. induces coli compared a different to responseeither ofthan the two stavudine. groups When (Figure examining 4A). This the suggests change that the APIM-peptide induces a different response than stavudine. When examining the change in proteins known to be related to the bacterial SOS response [53], more SOS proteins were found to change in stavudine and combination-treated cells, than in the APIM- Biomolecules 2021, 11, x 10 of 13

peptide-treated cells (Figure 4B). Heatmaps of the log2 fold change in pull-down of proteins found to be significantly changed in at least one treatment group showed that most of the changed SOS regulated proteins were increased relative to control (Figure 4C). This was expected as the mutation frequencies increased. Interestingly, RecA was increased in both stavudine and the combination, but not in APIM-peptide-treated cells. The opposite Biomolecules 2021, 11, 843 was the case for LexA, which increased significantly only in APIM-peptide-treated10 cells. of 13 A trend towards a reduction in LexA was detected in stavudine and combination-treated bacteria, albeit not significant. These data support that stavudine is inducing an SOS re- insponse proteins in bacteria known toat doses be related far below to the MIC. bacterial APIM-peptide SOS response is not [53 inducing], more SOS an proteinsSOS response, were foundnor reducing to change the inSOS stavudine response and induced combination-treated by stavudine when cells, used than in combination. in the APIM-peptide- This was treatedexpected cells as (Figurethe APIM-peptide4B). Heatmaps is believed of the log to2 inhibitfold change the interactions in pull-down between of proteins the β-clamp found toand be Pol significantly V and thereby changed inhibit in at leastthe increase one treatment in mutation group showedfrequency, that but most not of to the inhibit changed the SOSSOS regulatedinduction proteins [21]. were increased relative to control (Figure4C). This was expected as the mutationWe next frequenciesexamined which increased. pathways Interestingly, were changed RecA upon was increasedtreatment inusing both information stavudine andfrom the stress combination, responses but previously not in APIM-peptide-treated described in ciprofloxacin-treated cells. The opposite bacteria was the [54]. case forWe LexA,foundwhich that stavudine increased treatment significantly led only to a inredu APIM-peptide-treatedced pull-down of several cells. Aproteins trend towards involved a reductionin general in metabolism, LexA was detected while APIM-peptide in stavudine andtreatment combination-treated led to an increased bacteria, pull-down albeit not of significant.more proteins These involved data support in nucleotide that stavudine metabolism is inducing (Figure 4D). an SOS The response latter result in bacteria is interest- at dosesing as farAPIM-peptide below MIC. also APIM-peptide affected the is nucleotide not inducing pools an more SOS response,than stavudine. nor reducing the SOS responseIn summary, induced the bymost stavudine likely explanation when used infor combination. the strong increase This was in expected mutation as thefre- APIM-peptidequency is that isstavudine believed toinduces inhibit SOS the interactionsat doses below between MIC, thelikelyβ-clamp via interference and Pol V andwith therebynormal inhibitbacterial the replications increase in mutationleading to frequency, replication but stress. not to inhibit the SOS induction [21].

FigureFigure 4. 4.Stavudine Stavudine inducesinduces SOSSOS ininEscherichia Escherichia colicoli.. ResultsResults ofof thethe MIBMIB assayassay ofof extractextract fromfromE. E. colicoliMG1655 MG1655treated treatedwith with stavudinestavudine (Stav, (Stav, 0.32 0.32µ μg/mL),g/mL), APIM-peptideAPIM-peptide (26(26µ μg/mL)g/mL) andand thethe combinationcombination (Stav(Stav ++ APIM-peptide)APIM-peptide) forfor 22 h.h. DataData fromfrom

threethree independent independent biological biological replicas replicas are are shown. shown. Protein Protein concentrations concentrations are are given given as as log log2 2fold fold change change relative relative to to untreated untreated control.control. ( A(A) Venn) Venn diagrams diagrams displaying displaying the the number number of proteins of proteins signiﬁcantly significantly changed changed in pull-downs in pull-downs from treated from cellstreated relative cells to untreated control according to the Wilcoxon signed-rank test. (B) Venn diagram showing an overlap in the different

treatment groups with proteins involved in SOS [53]. (C) Heatmaps displaying log2 fold change in protein concentration of SOS proteins relative to untreated control in the different treatment groups. The grey boxes represent no data. (D) Heatmaps

displaying log2 fold change in proteins concentration of proteins involved in general metabolism, nucleotide metabolism, lipid biosynthesis and DNA metabolism relative to untreated control in the different treatment groups. * Proteins that are signiﬁcantly changed according to the Wilcoxon signed-rank test (i.e., changed the same way in all three replicas). Biomolecules 2021, 11, 843 11 of 13

We next examined which pathways were changed upon treatment using information from stress responses previously described in ciproﬂoxacin-treated bacteria [54]. We found that stavudine treatment led to a reduced pull-down of several proteins involved in general metabolism, while APIM-peptide treatment led to an increased pull-down of more proteins involved in nucleotide metabolism (Figure4D). The latter result is interesting as APIM- peptide also affected the nucleotide pools more than stavudine. In summary, the most likely explanation for the strong increase in mutation frequency is that stavudine induces SOS at doses below MIC, likely via interference with normal bacterial replications leading to replication stress.

4. Conclusions Our findings presented herein indicate that anti-viral and anti-cancer NAs are capable of inducing bacterial mutagenesis and contributing to high AMR in HIV and cancer patients at clinically relevant concentrations. Stavudine induces mutagenesis by activating the E. coli SOS system and, thereby, the TLS polymerase Pol V. Thus, the combination of NAs/NRTIs and antibiotic treatment in already vulnerable patients may drive the development of and selection for, antibiotic-resistant bacteria. APIM-peptide treatment significantly reduced the stavudine-induced mutagenesis suggesting that this peptide could be a good drug candidate in the fight against AMR.

Author Contributions: Conceptualization, B.K.S., E.S.D., L.M., S.D. and M.O.; methodology, L.M.R., P.B. and M.O.; validation, B.K.S., S.B.R., L.M.R., P.B. and M.O.; formal analysis, B.K.S., A.S., P.B., L.M.R. and M.O.; investigation, B.K.S., S.B.R. and L.M.R.; resources, M.O. and P.B.; writing—original draft preparation, B.K.S., S.B.R., P.B. and M.O.; writing—review and editing, B.K.S., L.M.R., A.S., P.B., M.O., L.M. and S.D.; visualization, B.K.S., S.B.R., A.S., P.B. and M.O.; supervision, M.O.; project administration, M.O.; funding acquisition, M.O. and P.B. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by WACCBIP-DELTAS Africa grant (DEL-15-007) and Welcome Trust 107755/Z/15/Z (awarded to Gordon Akanzuwine Awandare), the Joint Research Committee between St. Olavs and Faculty of Medicine and Health Science, NTNU, the program NTNU Health, at Norwegian University of Science and Technology (NTNU) and the Trond Mohn foundation. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The raw data is deposited in PRIDE, project ID PXD025370 (Reviewer account Username: [email protected] Password: seTUpXp4). Acknowledgments: We would like to thank Siri Bachke for technical assistance and NorStore/Notur Projects NS/NN9036K for data storage/computation services. Conﬂicts of Interest: No conﬂict of interest to declare.

References 1. CDCP. Antibiotic Resistance Threats in the United States; US Department of Health and Human Service: Atlanta, GA, USA, 2019. 2. European Centre for Disease Prevention and Control. Surveillance of Antimicrobial Resistance in Europe 2018; European Centre for Disease Prevention and Control: Stockholm, Sweden, 2019. 3. Tadesse, B.T.; Ashley, E.A.; Ongarello, S.; Havumaki, J.; Wijegoonewardena, M.; González, I.J.; Dittrich, S. Antimicrobial resistance in Africa: A systematic review. BMC Infect. Dis. 2017, 17, 1–17. [CrossRef] 4. O’Neill, J. Review on Antimicrobial Resistance. Antimicrobial Resistance: Tackling a Crisis for the Health and Wealth of Nations. 2014. Available online: https://amr-review.org (accessed on 4 January 2020). 5. Getahun, H.; Smith, I.; Trivedi, K.; Paulin, S.; Balkhy, H.H. Tackling antimicrobial resistance in the COVID-19 pandemic. Bull. World Health Organ. 2020, 98, 442–442A. [CrossRef] 6. World Health Organization. Clinical Management of COVID-19: Interim Guidance, 27 May 2020; World Health Organization: Geneva, Switzerland , 2020. 7. Rawson, T.M.; Moore, L.S.; Zhu, N.; Ranganathan, N.; Skolimowska, K.; Gilchrist, M.; Satta, G.; Cooke, G.; Holmes, A. Bacterial and fungal co-infection in individuals with coronavirus: A rapid review to support COVID-19 antimicrobial prescribing. Clin. Infect. Dis. 2020, 71, 2459–2468. Biomolecules 2021, 11, 843 12 of 13

8. Cirz, R.T.; Chin, J.K.; Andes, D.R.; De Crécy-Lagard, V.; Craig, W.A.; Romesberg, F.E. Inhibition of Mutation and Combating the Evolution of Antibiotic Resistance. PLoS Biol. 2005, 3, e176. [CrossRef][PubMed] 9. Merrikh, H.; Kohli, R.M. Targeting evolution to inhibit antibiotic resistance. FEBS J. 2020, 287, 4341–4353. [CrossRef][PubMed] 10. Petrosino, J.F.; Galhardo, R.S.; Morales, L.D.; Rosenberg, S.M. Stress-induced beta-lactam antibiotic resistance mutation and se- quences of stationary-phase mutations in the Escherichia coli chromosome. J. Bacteriol. 2009, 191, 5881–5889. [CrossRef][PubMed] 11. Gawel, D.; Seed, P.C. Urinary tract infection drives genome instability in uropathogenic Escherichia coli and necessitates translesion synthesis DNA polymerase IV for virulence. Virulence 2011, 2, 222–232. [CrossRef] 12. Altieri, A.S.; Kelman, Z. DNA Sliding Clamps as Therapeutic Targets. Front. Mol. Biosci. 2018, 5, 87. [CrossRef] 13. Campos, P.E.; Suarez, P.G.; Sánchez, J.; Zavala, D.; Arevalo, J.; Ticona, E.; Nolan, C.M.; Hooton, T.M.; Holmes, K.K. Multidrug- resistant Mycobacterium tuberculosis in HIV-Infected Persons, Peru. Emerg. Infect. Dis. 2003, 9, 1571–1578. [CrossRef] 14. Marami, D.; Balakrishnan, S.; Seyoum, B. Prevalence, Antimicrobial Susceptibility Pattern of Bacterial Isolates, and Associated Factors of Urinary Tract Infections among HIV-Positive Patients at Hiwot Fana Specialized University Hospital, Eastern Ethiopia. Can. J. Infect. Dis. Med Microbiol. 2019, 2019, 1–8. [CrossRef][PubMed] 15. Romana, K.; Rieger, A.; Presterl, E.; Wrba, T.; Diab-Elschahawi, M. Epidemiology of Urinary Tract Infections in HIV Positive Patients at a Tertiary Care University Hospital in Central Europe (2011–2016). Infect. Disord. Drug Targets 2018, 18, 199–206. 16. Ssengooba, W.; Lukoye, D.; Meehan, C.J.; Kateete, D.P.; Joloba, M.L.; De Jong, B.C.; Cobelens, F.G.; Van Leth, F. Tuberculosis resistance-conferring mutations with fitness cost among HIV-positive individuals in Uganda. Int. J. Tuberc. Lung Dis. 2017, 21, 531–536. [CrossRef] 17. Adeyemi, F.M.; Ako-nai, K.A.; Adejuyigbe, E.; Ebhodaghe, B.I.; Osho, P.O.; Oyeniyi, T.T.; Kassim, O.O. Molecular characterization and antibiotic resistance profiles of bacterial isolates cultured from HIV seropositive patients. Arch. Clin. Microbiol. 2015, 6, e2. 18. Jordheim, L.P.; Durantel, D.; Zoulim, F.; Dumontet, C. Advances in the development of nucleoside and nucleotide analogues for cancer and viral diseases. Nat. Rev. Drug Discov. 2013, 12, 447–464. [CrossRef] 19. Thomson, J.M.; Lamont, I.L. Nucleoside Analogues as Antibacterial Agents. Front. Microbiol. 2019, 10, 952. [CrossRef] 20. Furman, P.A.; Fyfe, J.A.; Clair, M.H.S.; Weinhold, K.; Rideout, J.L.; Freeman, G.A.; Lehrman, S.N.; Bolognesi, D.P.; Broder, S.; Mitsuya, H. Phosphorylation of 30-azido-30-deoxythymidine and selective interaction of the 50-triphosphate with human immunodeficiency virus reverse transcriptase. Proc. Natl. Acad. Sci. USA 1986, 83, 8333–8337. [CrossRef] 21. Nedal, A.; Ræder, S.B.; Dalhus, B.; Helgesen, E.; Forstrøm, R.J.; Lindland, K.; Sumabe, B.K.; Martinsen, J.H.; Kragelund, B.B.; Skarstad, K.; et al. Peptides containing the PCNA interacting motif APIM bind to the β-clamp and inhibit bacterial growth and mutagenesis. Nucleic Acids Res. 2020, 48, 5540–5554. [CrossRef] 22. CLSI. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard—Ninth Edition; CLSI Document M07-A9; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2012. 23. Petrovic, V.; Olaisen, C.; Sharma, A.; Nepal, A.; Bugge, S.; Sundby, E.; Hoff, B.H.; Slupphaug, G.; Otterlei, M. On-column trypsinization allows for re-use of matrix in modified multiplexed inhibitor beads assay. Anal. Biochem. 2017, 523, 10–16. [CrossRef][PubMed] 24. Røst, L.M.; Thorfinnsdottir, L.B.; Kumar, K.; Fuchino, K.; Langørgen, I.E.; Bartosova, Z.; Kristiansen, K.A.; Bruheim, P. Absolute Quantification of the Central Carbon Metabolome in Eight Commonly Applied Prokaryotic and Eukaryotic Model Systems. Metabolites 2020, 10, 74. [CrossRef][PubMed] 25. Stafsnes, M.H.; Røst, L.M.; Bruheim, P. Improved phosphometabolome profiling applying isotope dilution strategy and capillary ion chromatography-tandem mass spectrometry. J. Chromatogr. B 2018, 1083, 278–283. [CrossRef] 26. Røst, L.M.; Shafaei, A.; Fuchino, K.; Bruheim, P. Zwitterionic HILIC tandem mass spectrometry with isotope dilution for rapid, sensitive and robust quantification of pyridine nucleotides in biological extracts. J. Chromatogr. B 2020, 1144, 122078. [CrossRef] 27. Lien, S.K.; Kvitvang, H.F.N.; Bruheim, P. Utilization of a deuterated derivatization agent to synthesize internal standards for gas chromatography–tandem mass spectrometry quantification of silylated metabolites. J. Chromatogr. A 2012, 1247, 118–124. [CrossRef] 28. Chong, J.; Wishart, D.S.; Xia, J. Using MetaboAnalyst 4.0 for Comprehensive and Integrative Metabolomics Data Analysis. Curr. Protoc. Bioinform. 2019, 68, e86. [CrossRef] 29. Tyanova, S.; Temu, T.; Cox, J. The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat. Protoc. 2016, 11, 2301–2319. [CrossRef][PubMed] 30. Geiszler, D.J.; Kong, A.T.; Avtonomov, D.M.; Yu, F.; Leprevost, F.D.V.; Nesvizhskii, A.I. PTM-Shepherd: Analysis and Summariza- tion of Post-Translational and Chemical Modifications From Open Search Results. Mol. Cell. Proteom. 2021, 20, 100018. [CrossRef] 31. Cox, J.; Hein, M.; Luber, C.A.; Paron, I.; Nagaraj, N.; Mann, M. Accurate Proteome-wide Label-free Quantification by Delayed Nor- malization and Maximal Peptide Ratio Extraction, Termed MaxLFQ. Mol. Cell. Proteom. 2014, 13, 2513–2526. [CrossRef][PubMed] 32. Gibbons, J.D.; Chakraborti, S. Nonparametric Statistical Inference, 6th ed.; CRC Press: Boca Raton, FL, USA, 2021. 33. Ritchie, M.E.; Phipson, B.; Wu, D.; Hu, Y.; Law, C.W.; Shi, W.; Smyth, G.K. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015, 43, e47. [CrossRef][PubMed] 34. Perez-Riverol, Y.; Csordas, A.; Bai, J.; Bernal-Llinares, M.; Hewapathirana, S.; Kundu, D.J.; Inuganti, A.; Griss, J.; Mayer, G.; Eisenacher, M.; et al. The PRIDE database and related tools and resources in 2019: Improving support for quantification data. Nucleic Acids Res. 2019, 47, D442–D450. [CrossRef][PubMed] Biomolecules 2021, 11, 843 13 of 13

35. Sandrini, M.P.B.; Clausen, A.R.; On, S.L.W.; Aarestrup, F.M.; Munch-Petersen, B.; Piškur, J. Nucleoside analogues are activated by bacterial deoxyribonucleoside kinases in a species-specific manner. J. Antimicrob. Chemother. 2007, 60, 510–520. [CrossRef][PubMed] 36. Shilaih, M.; Angst, D.C.; Marzel, A.; Bonhoeffer, S.; Günthard, H.F.; Kouyos, R.D. Short communication Antibacterial effects of antiretrovirals, potential implications for microbiome studies in HIV. Antivir. Ther. 2018, 23, 91–94. [CrossRef] 37. Hurst, M.; Noble, S. Stavudine. Drugs 1999, 58, 919–949. [CrossRef] 38. Cruciani, M.; Liuzzi, G.; Chirianni, A.; Audagnotto, S.; Bonora, S.; Di Biagio, A.; Sinicco, A.; Bassetti, M.; Gatti, G. Penetration of didanosine in semen of HIV-1-infected men. J. Antimicrob. Chemother. 2006, 57, 1244–1247. [CrossRef] 39. Bang, L.M.; Scott, L.J. Emtricitabine. Drugs 2003, 63, 2413–2424. [CrossRef] 40. Calcagno, A.; De Requena, D.G.; Simiele, M.; D’Avolio, A.; Tettoni, M.C.; Salassa, B.; Orofino, G.; Bramato, C.; Libanore, V.; Motta, I.; et al. Tenofovir Plasma Concentrations According to Companion Drugs: A Cross-Sectional Study of HIV-Positive Patients with Normal Renal Function. Antimicrob. Agents Chemother. 2013, 57, 1840–1843. [CrossRef][PubMed] 41. Capparelli, E.V.; Letendre, S.L.; Ellis, R.J.; Patel, P.; Holland, D.; McCutchan, J.A. Population Pharmacokinetics of Abacavir in Plasma and Cerebrospinal Fluid. Antimicrob. Agents Chemother. 2005, 49, 2504–2506. [CrossRef] 42. Yuen, G.J.; Lou, Y.; Bumgarner, N.F.; Bishop, J.P.; Smith, G.A.; Otto, V.R.; Hoelscher, D.D. Equivalent Steady-State Pharmacokinetics of Lamivudine in Plasma and Lamivudine Triphosphate within Cells following Administration of Lamivudine at 300 Milligrams Once Daily and 150 Milligrams Twice Daily. Antimicrob. Agents Chemother. 2004, 48, 176–182. [CrossRef][PubMed] 43. Garré, B.; Shebany, K.; Gryspeerdt, A.; Baert, K.; Van Der Meulen, K.; Nauwynck, H.; Deprez, P.; De Backer, P.; Croubels, S. Pharmacokinetics of Acyclovir after Intravenous Infusion of Acyclovir and after Oral Administration of Acyclovir and Its Prodrug Valacyclovir in Healthy Adult Horses. Antimicrob. Agents Chemother. 2007, 51, 4308–4314. [CrossRef] 44. Lerner, A.M.; Bailey, E.J. Concentrations of Idoxuridine in Serum, Urine, and Cerebrospinal Fluid of Patients with Suspected Diagnoses of Herpesvirus hominis Encephalitis. J. Clin. Investig. 1972, 51, 45–49. [CrossRef] 45. McEvoy, G.K. American Hospital Formulary Service-Drug Information 92; American Society of Hospital Pharmacists Inc.: Bethesda, MD, USA, 1993. 46. National Center for Biotechnology Information. In PubChem Compound Summary for CID 21704, Vidarabine; National Center for Biotechnology Information: Bethesda, MD, USA, 2020. 47. Robinson, D.M.; Scott, L.J.; Plosker, G.L. Entecavir. Drugs 2006, 66, 1605–1622. [CrossRef][PubMed] 48. Kosovec, J.E.; Egorin, M.J.; Gjurich, S.; Beumer, J.H. Quantitation of 5-fluorouracil (5-FU) in human plasma by liquid chromatography/electrospray ionization tandem mass spectrometry. Rapid Commun. Mass Spectrom. 2007, 22, 224–230. [CrossRef] 49. Gandhi, V.; Plunkett, W. Cellular and Clinical Pharmacology of Fludarabine. Clin. Pharmacokinet. 2002, 41, 93–103. [CrossRef][PubMed] 50. Hande, K.R.; Stein, R.S.; McDonough, D.A.; Greco, F.A.; Wolff, S.N. Effects of high-dose cytarabine. Clin. Pharmacol. Ther. 1982, 31, 669–674. [CrossRef] 51. Sutton, M.D.; Smith, B.T.; Godoy, V.G.; Walker, G.C. The SOS response: Recent insights into umuDC-dependent mutagenesis and DNA damage tolerance. Annu. Rev. Genet. 2000, 34, 479–497. [CrossRef][PubMed] 52. Røst, L.M.; Olaisen, C.; Sharma, A.; Nedal, A.; Petrovic, V.; Kvitvang, H.F.N.; Bruheim, P.; Otterlei, M. PCNA has specific functions in regulation of metabolism in haematological cells. bioRxiv 2020.[CrossRef] 53. Simmons, L.A.; Foti, J.J.; Cohen, S.E.; Walker, G.C. The SOS Regulatory Network. EcoSal Plus 2008, 3.[CrossRef] 54. Cirz, R.T.; Jones, M.B.; Gingles, N.A.; Minogue, T.D.; Jarrahi, B.; Peterson, S.N.; Romesberg, F.E. Complete and SOS-Mediated Response of Staphylococcus aureus to the Antibiotic Ciprofloxacin. J. Bacteriol. 2007, 189, 531–539. [CrossRef][PubMed]