Transcriptome software results show significant variation among different commercial pipelines.

BACKGROUND: We have been documenting the biological responses to low levels of radiation (natural background) and very low level radiation (below background), and thus these studies are testing mild external stimuli to which we would expect relatively mild biological responses. We recently published a transcriptome software comparison study based on RNA-Seqs from a below background radiation treatment of two model organisms, E. coli and C. elegans (Thawng and Smith, BMC Genomics 23:452, 2022). We reported DNAstar-D (Deseq2 in the DNAstar software pipeline) to be the more conservative, realistic tool for differential gene expression compared to other transcriptome software packages (CLC, Partek and DNAstar-E (using edgeR). Here we report two follow-up studies (one with a new model organism, Aedes aegypti and another software package (Azenta) on transcriptome responses from varying dose rates using three different sources of natural radiation. RESULTS: When E. coli was exposed to varying levels of K40, we again found that the DNAstar-D pipeline yielded a more conservative number of DEGs and a lower fold-difference than the CLC pipeline and DNAstar-E run in parallel. After a 30 read minimum cutoff criterion was applied to the data, the number of significant DEGs ranged from 0 to 81 with DNAstar-D, while the number of significant DEGs ranged from 4 to 117 and 14 to 139 using DNAstar-E and the CLC pipelines, respectively. In terms of the extent of expression, the highest foldchange DEG was observed in DNAstar-E with 19.7-fold followed by 12.5-fold in CLC and 4.3-fold in DNAstar-D. In a recently completed study with Ae. Aegypti and using another software package (Azenta), we analyzed the RNA-Seq response to similar sources of low-level radiation and again found the DNAstar-D pipeline to give the more conservative number and fold-expression of DEGs compared to other softwares. The number of significant DEGs ranged 31-221 in Azenta and 31 to 237 in CLC, 19-252 in DNAstar-E and 0-67 in DNAStar-D. The highest fold-change of DEGs were found in CLC (1,350.9-fold), with DNAstar-E (5.9 -fold) and Azenta (5.5-fold) intermediate, and the lowest levels of expression (4-fold) found in DNAstar-D. CONCLUSIONS: This study once again highlights the importance of choosing appropriate software for transcriptome analysis. Using three different biological models (bacteria, nematode and mosquito) in four different studies testing very low levels of radiation (Van Voorhies et al., Front Public Health 8:581796, 2020; Thawng and Smith, BMC Genomics 23:452, 2022; current study), the CLC software package resulted in what appears to be an exaggerated gene expression response in terms of numbers of DEGs and extent of expression. Setting a 30-read cutoff diminishes this exaggerated response in most of the software tested. We have further affirmed that DNAstar-Deseq2 gives a more conservative transcriptome expression pattern which appears more suitable for studies expecting subtle gene expression patterns.

A transcriptome software comparison for the analyses of treatments expected to give subtle gene expression responses.

BACKGROUND: In this comparative study we evaluate the performance of four software tools: DNAstar-D (DESeq2), DNAstar-E (edgeR), CLC Genomics and Partek Flow for identification of differentially expressed genes (DEGs) using a transcriptome of E. coli. The RNA-seq data are from the effect of below-background radiation 5.5 nGy total dose (0.2nGy/hr) on E. coli grown shielded from natural radiation 655 m below ground in a pre-World War II steel vault. The gene expression response to three supplemented sources of radiation designed to mimic natural background, 1952 - 5720 nGy in total dose (71-208 nGy/hr), are compared to this "radiation-deprived" treatment. In addition, RNA-seq data of Caenorhabditis elegans nematode from similar radiation treatments was analyzed by three of the software packages. RESULTS: In E. coli, the four software programs identified one of the supplementary sources of radiation (KCl) to evoke about 5 times more transcribed genes than the minus-radiation treatment (69-114 differentially expressed genes, DEGs), and so the rest of the analyses used this KCl vs "Minus" comparison. After imposing a 30-read minimum cutoff, one of the DNAStar options shared two of the three steps (mapping, normalization, and statistic) with Partek Flow (they both used median of ratios to normalize and the DESeq2 statistical package), and these two programs identified the highest number of DEGs in common with each other (53). In contrast, when the programs used different approaches in each of the three steps, between 31 and 40 DEGs were found in common. Regarding the extent of expression differences, three of the four programs gave high fold-change results (15-178 fold), but one (DNAstar's DESeq2) resulted in more conservative fold-changes (1.5-3.5). In a parallel study comparing three qPCR commercial validation software programs, these programs also gave variable results as to which genes were significantly regulated. Similarly, the C. elegans analysis showed exaggerated fold-changes in CLC and DNAstar's edgeR while DNAstar-D was more conservative. CONCLUSIONS: Regarding the extent of expression (fold-change), and considering the subtlety of the very low level radiation treatments, in E. coli three of the four programs gave what we consider exaggerated fold-change results (15 - 178 fold), but one (DNAstar's DESeq2) gave more realistic fold-changes (1.5-3.5). When RT-qPCR validation comparisons to transcriptome results were carried out, they supported the more conservative DNAstar-D's expression results. When another model organism's (nematode) response to these radiation differences was similarly analyzed, DNAstar-D also resulted in the most conservative expression patterns. Therefore, we would propose DESeq2 ("DNAstar-D") as an appropriate software tool for differential gene expression studies for treatments expected to give subtle transcriptome responses.

Mobile resistome of human gut and pathogen drives anthropogenic bloom of antibiotic resistance.

BACKGROUND: The impact of human activities on the environmental resistome has been documented in many studies, but there remains the controversial question of whether the increased antibiotic resistance observed in anthropogenically impacted environments is just a result of contamination by resistant fecal microbes or is mediated by indigenous environmental organisms. Here, to determine exactly how anthropogenic influences shape the environmental resistome, we resolved the microbiome, resistome, and mobilome of the planktonic microbial communities along a single river, the Han, which spans a gradient of human activities. RESULTS: The bloom of antibiotic resistance genes (ARGs) was evident in the downstream regions and distinct successional dynamics of the river resistome occurred across the spatial continuum. We identified a number of widespread ARG sequences shared between the river, human gut, and pathogenic bacteria. These human-related ARGs were largely associated with mobile genetic elements rather than particular gut taxa and mainly responsible for anthropogenically driven bloom of the downstream river resistome. Furthermore, both sequence- and phenotype-based analyses revealed environmental relatives of clinically important proteobacteria as major carriers of these ARGs. CONCLUSIONS: Our results demonstrate a more nuanced view of the impact of anthropogenic activities on the river resistome: fecal contamination is present and allows the transmission of ARGs to the environmental resistome, but these mobile genes rather than resistant fecal bacteria proliferate in environmental relatives of their original hosts. Video abstract.

A novel sulfonamide resistance mechanism by two-component flavin-dependent monooxygenase system in sulfonamide-degrading actinobacteria.

Sulfonamide-degrading bacteria have been discovered in various environments, suggesting the presence of novel resistance mechanisms via drug inactivation. In this study, Microbacterium sp. CJ77 capable of utilizing various sulfonamides as a sole carbon source was isolated from a composting facility. Genome and proteome analyses revealed that a gene cluster containing a flavin-dependent monooxygenase and a flavin reductase was highly up-regulated in response to sulfonamides. Biochemical analysis showed that the two-component monooxygenase system was key enzymes for the initial cleavage of sulfonamides. Co-expression of the two-component system in Escherichia coli conferred decreased susceptibility to sulfamethoxazole, indicating that the genes encoding drug-inactivating enzymes are potential resistance determinants. Comparative genomic analysis revealed that the gene cluster containing sulfonamide monooxygenase (renamed as sulX) and flavin reductase (sulR) was highly conserved in a genomic island shared among sulfonamide-degrading actinobacteria, all of which also contained sul1-carrying class 1 integrons. These results suggest that the sulfonamide metabolism may have evolved in sulfonamide-resistant bacteria which had already acquired the class 1 integron under sulfonamide selection pressures. Furthermore, the presence of multiple insertion sequence elements and putative composite transposon structures containing the sulX gene cluster indicated potential mobilization. This is the first study to report that sulX responsible for both sulfonamide degradation and resistance is prevalent in sulfonamide-degrading actinobacteria and its genetic signatures indicate horizontal gene transfer of the novel resistance gene.

Polymorphism of antibiotic-inactivating enzyme driven by ecology expands the environmental resistome.

The environmental resistome has been recognized as the origin and reservoir of antibiotic resistance genes and considered to be dynamic and ever expanding. In this study, a targeted gene sequencing approach revealed that the polymorphic diversity of the aminoglycoside-inactivating enzyme AAC(6')-Ib was ecological niche-specific. AAC(6')-Ib-cr, previously known as a clinical variant, was prevalent in various soils and the intestines of chickens and humans, suggesting that this variant might not have arisen from adaptive mutations in the clinic but instead originated from the environment. Furthermore, ecologically dominant polymorphic variants of AAC(6')-Ib were characterized and found to display different substrate specificities for quinolones and aminoglycosides, conferring the altered resistance spectra. Interestingly, a novel variant with the D179Y substitution showed an extended resistance spectrum to the recently developed fluoroquinolone gemifloxacin. Our results suggest that soil and animal microbiomes could be major reservoirs of antibiotic resistance; polymorphic diversity expands the antibiotic resistome in the environment, resulting in the potential emergence of novel resistance.

Unique Features of Aeromonas Plasmid pAC3 and Expression of the Plasmid-Mediated Quinolone Resistance Genes.

A highly fluoroquinolone-resistant isolate of Aeromonas species was isolated from a wastewater treatment plant and found to possess multiple resistance mechanisms, including mutations in gyrA and parC, efflux pumps, and plasmid-mediated quinolone resistance (PMQR) genes. Complete sequencing of the IncU-type plasmid, pAC3, present in the strain revealed a circular plasmid DNA 15,872 bp long containing two PMQR genes [qnrS2 and aac(6')-Ib-cr]. A mobile insertion cassette element containing the qnrS2 gene and a typical miniature inverted-repeat transposable element (MITE) structure was identified in the plasmid. The present study revealed that this MITE sequence appears in other Aeromonas species plasmids and chromosomes. Plasmid pAC3 was introduced into Escherichia coli, and its PMQR genes were expressed, resulting in the acquisition of resistance. Proteome analysis of the recipient E. coli strain harboring the plasmid revealed that aac(6')-Ib-cr expression was constitutive and that qnrS2 expression was dependent upon fluoroquinolone stress through regulation by regulator of sigma D (Rsd). To the best of our knowledge, this is the first report to characterize a novel MITE sequence upstream of the PMQR gene within a mobile insertion cassette, as well as the regulation of qnrS2 expression. Our results suggest that this mobile element may play an important role in qnrS2 dissemination. IMPORTANCE In the present study, plasmid pAC3 isolated from a highly fluoroquinolone-resistant isolate of Aeromonas species was sequenced and found to contain two fluoroquinolone resistance genes, aac(6')-Ib-cr and qnrS2. Comparative analyses of plasmid pAC3 and other Aeromonas sp. IncU-type plasmids revealed a mobile insertion cassette element with a unique structure containing a qnrS2 gene and a typical miniature inverted-repeat transposable element (MITE) structure. This study also revealed that this MITE sequence appears in other Aeromonas species plasmids and chromosomes. Our results also demonstrate that the fluoroquinolone-dependent expression of qnrS2 is associated with rsd in E. coli DH5α harboring plasmid pAC3. Our findings suggest that the mobile element may play an important role in qnrS2 dissemination and that Aeromonas species constitute an important reservoir of fluoroquinolone resistance determinants in the environment.

Stakelama sediminis sp. nov., isolated from tidal flat sediment.

A novel bacterial strain designated CJ70(T) was isolated from tidal flat sediment in Korea. A polyphasic approach was used to identify this strain taxonomically. The isolate was Gram-stain-negative, strictly aerobic, yellow-pigmented, rod-shaped and non-motile. Phylogenetic analysis of the 16S rRNA gene sequence revealed that strain CJ70(T) was related most closely to Stakelama pacifica JLT832(T) with 95.7 % similarity and formed an independent phyletic line from recognized species of the genus Sphingomonas, comprising a clade with Stakelama pacifica, which is the only recognized species of the genus Stakelama. The predominant cellular fatty acids of strain CJ70(T) were C(18 : 1)ω7c (60.0 %), C(16 : 0) (21.2 %) and C(14 : 0) 2-OH (5.8 %). The major isoprenoid quinone was ubiquinone-10. The G+C content of the genomic DNA was 61.4 mol%. The results obtained from this study suggested that strain CJ70(T) represents a novel species of the genus Stakelama, for which the name Stakelama sediminis sp. nov. is proposed. The type strain is CJ70(T) ( = KACC 16559(T) = JCM 18079(T)).