text
string | predicted_class
string | confidence
float16 |
|---|---|---|
For identification and analysis of G. elata infected by A. mellea, hand sections were cut through the infection point of an immature tuber, and the sections were then embedded in agar plates. Images were captured using a Zeiss AX10 fluorescence microscope with ×10 water immersion lenses. Owing to the spontaneous blue fluorescence, both visible and DAPI filters were used to observe the hyphae of A. mellea with fluorescent microscopy (Zeiss AX10).
|
other
| 99.25 |
To analyze tissue and cell structures of G. elata uninfected by A. mellea, paraffin sections (10 μm thickness) were obtained using a Thermo Scientific MicRoM HM 325 sliding microtome. For light microscopic observations of the highly lignified hyphae of A. mellea, sections were stained with Fast Green stain reagent to investigate the infected cells of G. elata. After staining, sections were washed by PBS three times, dehydrated through an alcohol (50%, 80%, 90%, 95%, 100%), cleared in xylene, sealed with neutral gum, and observed using a fluorescence microscope (Zeiss AX10).
|
study
| 99.9 |
About 0.35 g of frozen fresh G. elata were homogenized and ultrasonically extracted for 30 min in four volumes (g mL−1) water. After centrifugation (13,000 rpm, 10 min), 100 μL aliquots of supernatant were mixed with 10 μl of rutin (101.0 μg mL−1) for quantification of p-HA (λ 270 nm, Rt 6.8 min) and S-(p-HA)-glutathione (m/z 412.12, Rt 8.9 min) using an UPLC-PDA-ESI-Q-TOF-MSE method73. The injection volume was 1 μL. The contents were determined by the peak intensity ratios of the analyte to rutin (m/z 609.14, 15.5 min).
|
study
| 100.0 |
Hyphal branching in A. mellea fungi was evaluated in vitro by the paper disk diffusion method74. Primary hyphae were cultured in PDA medium containing 20 g L−1 glucose, 4 g L−1 potato powder and 14 g L−1 agar. The dishes were cultured in the dark for 5–7 days at 23 °C. Secondary hyphae emerge from primary hyphae and grow upward in a negative geotropic manner in the gel; the growth of secondary hyphae was used for the assay. Test samples were first dissolved in acetone then diluted with 70% ethanol in water. The concentration of test sample solutions of natural 5-deoxy-strigol was adjusted with reference to the calibration of synthetic (±)-5-deoxy-strigol in an HPLC analysis. Paper disks (1 cm in length, 8 mm in width) loaded with 15 μL of test sample solution were placed in front of the tips of the secondary hyphae. The control was on the opposite direction of the paper without 5-deoxy-strigol. Hyphal branch patterns were analyzed at 24 h and 48 h after treatment. The sample was scored as positive for hyphal branching if new hyphal branches formed from the treated secondary hyphae. The assay was repeated at least twice, using between three and five dishes for each concentration.
|
study
| 100.0 |
Genome data were deposited in GenBank under accession number PVEL00000000 and transcriptome sequence reads were deposited in the Sequence Read Archive (SRA) under accession number SRX2879747. The standard flowgram format (SFF) files related with bacterial and fungal communities were also deposited in the SRA under study accession SRX2875242 and SRX2876148. Plastid genome data were deposited in GenBank under accession number MF163256. Mitochondrial genome data were deposited in GenBank under accession numbers MF070084-MF070102
|
other
| 99.9 |
Forage legumes are highly valued feed for extensive livestock production. There is an increasing interest worldwide in using annual forage legumes as cover crops to supply soil nitrogen (Sulas, 2005; Piano et al., 2010). Symbiotic nitrogen fixation in legumes leads to high protein fodder content and rejuvenated soils for a sustainable feed system. The clovers in particular are among the most effective to break the ‘infernal circle of the fallow’ a technique known to the Germans as ‘Besömmerung’ (Blanning, 2008). Subterranean clover (Trifolium subterraneum L.) makes the greatest contribution to livestock feed production and soil improvement in terms of total worldwide usage among annual clovers (McGuire, 1985), particularly in Australia, where it is sown over 29 mill. ha. The self‐reseeding ability and grazing tolerance of subterranean clover, even under suboptimal and variable environmental conditions (Nichols et al., 2013), contribute to its widespread distribution.
|
review
| 99.25 |
Subterranean clover is a diploid (2n = 2x = 16), predominantly inbreeding, annual species with a relatively small genome size of 540 Mbp (1C = 0.55 pg DNA; Vižintin et al., 2006) that can be readily hybridized, and exhibits wide diversity for both qualitative and quantitative agronomic and morphological characters (Ghamkhar et al., 2011). Within the genus Trifolium, it is established as a reference species for genetic and genomic studies. We reported a de novo draft genome assembly of the subterranean clover TSUd_r1.1, generated using a combination of long‐ and short‐read sequencing platforms (Figure S1) (Hirakawa et al., 2016). Genetic and genomic analyses of the internationally commercially important perennial legumes white clover and red clover are difficult, as they are outcrossing and have self‐incompatible fertilization, with white clover also being an allotetraploid (2n = 4x = 32) (Abberton and Marshall, 2005; Ghamkhar et al., 2011). As molecular markers developed for white and red clovers were readily transferable to subterranean clover (Ghamkhar et al., 2011), it is likely that subterranean clover QTLs and genes are applicable to these other species. Close synteny with the model legume, Medicago truncatula, also provides opportunities for genomic comparisons and the identification of candidate genes.
|
study
| 99.94 |
A variety of approaches are required for the de novo assemblies to improve draft genomes, and include a method developed by Dovetail Genomics for genome scaffolding using long‐range genomic information obtained by Chicago method (Putnam et al., 2016). In this study, we evaluated genome mapping on nanochannel arrays employing BioNano Irys® system (www.bionanogenomics.com) to validate the initial draft assembly, anchor additional unplaced contigs and to resolve misassemblies, with the objective of improving the overall assembly coverage. BioNano genome mapping is a technique of optical mapping in which specific sequence motifs in single DNA molecules are fluorescently labelled. The labelled DNA molecules are loaded onto the IrysChip where they are electrophoretically linearized in thousands of silicon channels. Fluorescence imaging allows the construction of maps of the physical distances between occurrences of the sequence motifs (Lam et al., 2012). The aim was to advance the pseudomolecule assembly of the eight chromosomes of subterranean clover, based on the direct visualizations of sequence motifs on long single DNA molecules. Gene annotations for the de novo draft genome assembly of the subterranean clover TSUd_r1.1 were conducted using in silico automated Augustus and Maker pipelines only. This was compared with direct transcriptome analysis using whole genome RNA sequencing technology across five different tissues of subterranean clover to generate a valuable resource for the identification and characterization of genes and pathways underlying plant growth and development. This also provided a basis to investigate specific processes, biological functions and gene interactions for key agronomic traits.
|
study
| 100.0 |
The first draft genome assembly of the Australian subterranean clover variety, cv. Daliak, covers 85.4% of the estimated genome in eight pseudomolecules of 401.1 Mb length and was constructed using a linkage map consisting of 35 341 SNPs (Hirakawa et al., 2016). In this study, we evaluated genome mapping via scaffolding using de novo physical maps generated using the BioNano Genomics (BNG) Irys platform to improve the genome assembly. A total of 221.7 Gb (401× genome coverage) of filtered data (molecules >150 kb) was generated on the Irys instrument. After filtering out low‐quality single molecules, a total of 188.5 Gb (341× genome coverage) of data was included in the final BioNano‐based de novo physical map assembly. This physical map assembly consisted of 309 075 individual molecules and 468 consensus maps that spanned 512 Mb (93% genome coverage) with N50 of 1.4 Mb. Multiple contigs (264) from the first draft assembly coalesced into 97 super‐scaffolds (containing 43% of the total genome captured) (Figure 1). In the advanced assembly (Tsub_Refv2.0), the size of sequences longer than >1 Mb increased dramatically from 40 to 189 Mb. This resulted in a 1.4‐fold increase in the N50 with the total percentage of genome captured in pseudomolecules improving from 73 to 80% with a substantial reduction of sequence gaps (Table 1).
|
study
| 100.0 |
(a) In silico map of super‐scaffold 9 aligned to the BioNano optical maps (401× coverage). XMAP alignments for in silico map of sequence scaffolds 103, 143, 1029, 292 and 166 are shown. Consensus genome maps (blue with molecule map coverage shown in dark blue) align to the in silico maps of scaffolds (green with contigs overlaid as translucent coloured squares). An illustration of using BioNano optical maps to assist contig placement, scaffolding and inversion correction. Sequence scaffolds 103, 143, 1029, 292 and 166 were placed within super‐scaffold 9 by the optical maps and among these scaffolds; scaffolds 103, 143 and 166 were reversed in the super‐scaffold. (b) Illustration of misassemblies in the genome examined using BioNano optical maps. The super‐scaffold (colourful bar) contains scaffold Tsud_sc00127.10.1, scaffold Tsud_sc00415.00.1 and a gap between them. Compared with the optical map (blue bar), there are more Bsp QI restriction cut sites (displayed as straight lines) in the super‐scaffold at positions 2; there are some Bsp QI restriction cut sites missing at positions 1. At positions 3, 4 and 5, the super‐scaffold does not consist of BioNano consensus map. All those discordances can be examined in detail in the corresponding contigs or gaps (indicated above the super‐scaffold).
|
study
| 99.94 |
In this study, the BNG Irys platform provided affordable, high‐throughput physical maps of improved contiguity to validate the draft assembly (Hirakawa et al., 2016) generated across a combination of long‐read and short‐read platforms and extended scaffolds. In contrast to alternative, sequencing‐based approaches (e.g. Chicago method of Dovetail Genomics, GemCode Technology of 10× Genomics, or Hi‐C), it enables a highly accurate sizing of gaps in sequence assemblies and provides a real picture of genomic regions intractable to current sequencing technologies, such as long arrays of tandem repeats (Staňková et al., 2016). Thus, combinations of platforms are recommended and no one platform is a perfect technology to use in answering every research question (Chaney et al., 2016).
|
study
| 99.94 |
The identification and annotation of expressed genes within the T. subterraneum genome assembly used high‐throughput whole genome RNA sequencing analyses to predict a total of 32 333 transcripts for 31 272 protein‐coding genes, with evidence for their expression across five different tissue types. The process‐involved annotations that combined evidence from transcriptome alignments obtained from protein homology and in silico gene prediction derived from different tissue types (roots, stem and peduncles, leaf and petioles, flowers and developing seeds) of cv. Daliak (Table S1). Phytozome and TrEMBL were the most informative databases for assigning functional annotations to subterranean clover proteins, with 29 157 (90.2%) and 29 278 (89.9%) proteins annotated, respectively (Table S2). In the draft TSUd_r1.1, the presence of 42 706 genes was predicted from in silico evidence using homology studies, domain searches and ab initio gene predictions. The reduction in genes annotated in the present study was achieved because the latter computational methods are unable to provide definitive evidence about which genes are actually expressed (Guigó et al., 2000; Wheelan and Boguski, 1998).
|
study
| 100.0 |
Repetitive element analysis predicted that 64% of the genome comprises repeat sequences, of which 15.9% and 8.4% are within introns and exons, respectively. About a quarter (23.9%) of repetitive elements were classified in the unknown zone. Like most eukaryotes, transposable elements (TEs) were most commonly long terminal repeats (LTR) retrotransposons (9.3%), with Gypsy‐like elements as the most frequently classified retrotransposons. Other TEs such as DNA, rolling circle (RC) and non‐LTR long retrotransposons such as long interspersed nuclear elements (LINEs) and short interspersed nuclear elements (SINEs) comprised a relatively small proportion (2.2% and 1.9%, respectively) (Figure 2; Table S3). Non‐coding RNA was estimated to comprise 0.12% of the genome, the majority being ribosomal RNA (0.02%) and transfer RNA (0.01%), with predicted snRNA and miRNA representing 0.01% and 0.07%, respectively (Table S4).
|
study
| 99.94 |
To test the quality of the advanced genome assembly and re‐annotation, we conducted a CEGMA analysis (Parra et al., 2007) to identify the presence of core eukaryotic genes. From the core set of 248 eukaryotic genes, there were 240 complete and 247 partial genes present in the advanced assembly, representing 96.8% and 99.6% [in comparison with the 95.6% and 98.0% for the draft assembly TSUd_r1.1] genes of the core set, respectively (Table S5).
|
study
| 100.0 |
The advanced assembly was functionally characterized by assigning gene ontology (GO) terms to proteins by manually transferring the GO terms for Swiss‐Prot IDs using the UniProt‐GOA database (Huntley et al., 2015). All these protein‐coding genes/transcripts were grouped into the three main GO categories: biological processes, molecular function and cellular components. A total of 21 210 (65.6%) of the subterranean clover protein‐coding genes/transcripts were assigned GO terms (Table S1). Of the 131 324 GO terms identified, 5,648 appear only once. There are 1013 GO terms appearing between 50 and 200 times with a total sum of 35 383 (Figure S2).
|
study
| 100.0 |
Among the 5648 GO terms appearing once, the top five most highly represented groups in the biological processes category were genes/transcripts associated with signal transduction, response to abscisic acid, cellular response to DNA damage stimulus, methylation and response to water deprivation (Table S6, for enriched cellular components and molecular functions see Tables S7 and S8, respectively). More detailed classification of the biological process GO category also showed enrichment in comparison with whole UniProt for response to cold and oxidative stress, flower development, flavonoid biosynthesis and embryo development ending in seed dormancy as highly represented groups. Overall, this functional characterization and GO enrichment analysis confirmed gene expression for response to water deprivation, cold and oxidative stress, flavonoid biosynthetic process and embryo development ending in seed dormancy reflecting the adaptation of subterranean clover to the harsh Mediterranean environment.
|
study
| 100.0 |
To investigate the representation of genes among the five tissue types, hierarchical clustering of tissues, based on global gene expression and GO terms, was performed on 100 genes with the highest sum of all transcripts per million (TPM). The biological identity of the tissues was clearly reflected in the analysis by hierarchical clustering of the RNA‐Seq‐based transcriptome (Figure 3; Table S9). Predicted genes showed substantial variation in their expression over tissues, reflecting tissue‐specific biological activities. For instance, in leaf and petiole tissue GO‐based clusters G1 and G2 were enriched and genes‐encoding proteins involved in photosynthesis, binding and chloroplast thylakoid membranes were over‐represented consistent with the specialized biological function of these tissues. Likewise, the root transcriptome profile showed enriched expression for stress and defence responses with genes encoding for cysteine‐type endopeptidase inhibitor activity. However, the G3 cluster indicates commonalities in the transcriptome of all above ground vegetative tissue types and showed enrichments for translational and structural ribosomal pathways.
|
study
| 100.0 |
Heat map showing hierarchical clustering of tissues based on global gene expression and GO terms. Clustering was based on 100 genes with highest sum of all TPM (transcripts per million) of the five tissue types. The dendrogram of the selected genes [vertical axis] is visualized along with the expression patterns that are used to cluster the five tissue samples, i.e. root (RT), developing seed (DSD), leaf and petioles (LF_PT), stem and peduncles (ST_PD) and flower (FL) [horizontal axis]. There are five GO‐based clusters named G1…G5 that contain more than one gene. The GO clusters are separated by a horizontal bar in the heat map. Genes without annotations are omitted from the heat map. Highly expressed genes are shown with red colour and lower expressed genes with white or yellow colour, respectively.
|
study
| 100.0 |
For tissues with predominance of a specific biological activity, a large proportion of reads may have represented abundantly expressed genes and, therefore, deeper sequencing may be needed to detect genes with relatively low expression levels. However, the high correlation between biological replicates and the clustering of tissues based on their biology indicates that the sampling depth in this study is sufficient to draw inferences about the transcriptome.
|
study
| 100.0 |
The tissues sampled in the RNA library preparations allow the determination of dynamic spatial gene expression and characterization. A comparative gene expression analysis between five tissue types (Table S1), using the advanced genome assembly (Tsub_Refv2.0) as the reference, revealed tissue‐specific functional diversification of paralogous genes. Specific gene expression in each of the five different tissues was indicated by the comparison of global gene expression across tissues. To explore gene expression by tissue‐type further, differentially expressed genes were identified in all pairwise tissue‐type comparisons. These differentially expressed genes were then used to identify clusters of co‐expressed genes, which represent spatially enriched expression patterns. Such co‐expression clusters were identified by the Mfuzz package (Futschik and Carlisle, 2005) using the soft clustering approach with fuzzy c‐means algorithm. This analysis generated five clusters, which revealed dynamic gene expression patterns across the five tissue types (Figure 4a; Table S10).
|
study
| 100.0 |
Dynamic tissue‐specific gene co‐expression clusters. (a) Five clusters were generated using Fuzzy C‐means soft clustering algorithm implemented in Mfuzz. Data points on the X‐axis represent root (RT), developing seed (DSD), stem and peduncles (ST_PD), leaf and petioles (LF_PT), flowers (FL) tissues, respectively. The Y‐axis represents gene expression values, where gene expression values were standardized to have zero mean and one standard deviation. (b) The scatterplot generated using REVIGO web browser shows the cluster representatives (i.e. terms remaining after the redundancy reduction) for leaf and petiole tissue‐enriched Cluster 2 biological processes in a two‐dimensional space derived by applying multidimensional scaling to a matrix of the GO terms’ semantic similarities. Bubble colour indicates the P‐value (legend in upper left‐hand corner); size indicates the frequency of the GO term in the wholeUniProt GOA database (bubbles of more general terms are larger).
|
study
| 100.0 |
Cluster 1 included genes with the highest expression in floral tissue (Figure 4b). Among genes in this cluster several are of interest. For example, this cluster includes many genes defining floral organ identity such as floral homeotic protein AGAMOUS, APETALA 2, PMADS 2 and DEFICIENS. In addition, Cluster 1 also included chromatin structure‐remodelling complex protein SYD, MADS‐box protein CMB1 and many chloroplast and carotenoid genes. Leaf‐enriched Cluster 2 included transcripts‐encoding enzymes involved in photosynthesis, Rubisco and oxygenase activity in addition to many chloroplast and plastid‐regulating genes. Cluster 3 contained genes with high expression in stem and peduncle tissue, and included genes/transcripts‐encoding enzymes involved in nutrient transport, structural growth with a high activity of various receptor kinases, expansins and transferases. Developing seed‐enriched Cluster 4 showed genes genes/transcripts involved with embryogenesis and growth. Cluster 5 enriched with root tissue gene expression profiles included different nodulins, binding proteins, MYB‐transcription factors and both stress‐ and defence‐related genes.
|
study
| 100.0 |
Genes within different clusters are potential sources of tissue‐specific promoters. For example, to improve forage nutritional quality by isoflavonoid biosynthesis or digestibility through processes such as cell wall loosening and lignin biosynthesis, promoters specific to leaf tissue are required. Options for transformation have recently broadened to include genome‐editing tools such as CRISPR‐cas9 (Rani et al., 2016).
|
review
| 99.56 |
Genes representing tissue‐enriched clusters were subjected to GO enrichment analysis to further characterize and identify over‐represented functional groups in subterranean clover using REVIGO web server (Supek et al., 2011) to graphically represent the results. The scatterplot depicting leaf‐enriched cluster (Cluster 2) showed enrichment for genes associated with photosynthesis, transport and response to stimulus (Figure 4b; Table S11). The flower‐enriched cluster (1) showed a concentration of genes associated with the processes of nitrogen compound cellular biosynthesis, multicellular organismal organization and development (Table S12). The stem‐ and peduncle‐enriched cluster (3) displayed a high representation of genes associated with response to stimulus, stress, oxidation‐reduction process, transport and signal transduction. The root‐enriched cluster (5) had a high representation of genes associated with response to stress, stimulus, nutrient levels, transport and nitrate metabolism. Clearly, the GO enrichment analyses of tissue‐enriched clusters showed over‐representation of predicted classes of genes in different tissue types and thereby validated our tissue‐enriched gene co‐expression clusters. Additionally, putative genes‐encoding enzymes in subterranean clover were assigned to various pathways in the KEGG database using BLASTKOALA (Kanehisa et al., 2016). This analysis distributed 4,094 genes‐encoding enzymes (12.7%) into 133 different KEGG pathways (Figure S3; Table S13).
|
study
| 100.0 |
Comparative analyses across the Papilionoideae using the published whole genome sequences for 13 legumes and the Tsub_Refv2.0 (Figure 5a and b; Table S14) revealed an overlap of 12 170 orthologous gene clusters between Papilionoideae and Arabidopsis thaliana as the outgroup species. Within the Papilionoideae, the number of orthologous gene clusters specific to galegoid (Lotus japonicus, Trifolium pratense, T. subterraneum, M. truncatula and Cicer arietinum), millettioid (Glycine max, Cajanus cajan, Vigna radiata, V. angularis and Phaseolus vulgaris) and dalbergoid (Arachis ipaensis and A. duranensis) was 10 873, 7976 and 4004, respectively. There were 3171 orthologous gene clusters in common among galegoid, millettioid and dalbergoid species.
|
study
| 100.0 |
(a) Phylogenetic tree using the published predicted protein sequences for 13 legume genomes including the newly annotated Trifolium subterraneum L. and Arabidopsis thaliana. Mash v1.1 (preprint: http://biorxiv.org/content/early/2016/04/19/029827) was used to calculate a distance matrix and pairwise mutation distance estimation using the published whole genome sequences for all the legumes except the transcriptome assembly for Cajanus cajan as a whole genome assembly was not available. The phylogeny was constructed using UPGMA as implemented in the R‐package ‘phangorn’ v2.0.3. (b) Shared and unique gene clusters in A) Papilionoideae species and A. thaliana; B) Millettioid, Galegoid and Dalbergioid clade or and A. thaliana; C) Lotus, Trifoliums, Medicago and Cicers; D) Shared and unique genes in sub‐clover and red clover.
|
study
| 100.0 |
To find Trifolium‐specific genes among the galegoids, we identified 18 131 genes representing 24.3% of the total 74 556 genes using BLAST searches (Table S15). This proportion is higher than reported in chickpea (10%; Garg et al., 2011; Jain et al., 2013) Arabidopsis (4.9%; Lin et al., 2010) and rice (17.4%; Campbell et al., 2007). Trifolium‐specific gene clusters are a source of unique genes controlling important traits as drought tolerance, and disease resistance. Within the genus Trifolium, similar BLAST searches identified 6,325 (19.6%) and 11 806 (28.0%) genes as subterranean‐clover‐specific and red‐clover‐specific genes, respectively (Tables S16 and S17). These candidate subterranean‐clover‐specific genes could be mined further for such key traits as geocarpy and other factors related to the grazing tolerance of subterranean clover. Examining orthologous gene clusters provides an important foundation for comparative biology and functional inference in subterranean clover, because genes with simple orthologous relationships often have conserved functions, whereas genes duplicated more recently relative to speciation often underlie functional diversification. Comparative genomic analysis of the advanced subterranean assembly with the model legume, Medicago truncatula, showed close synteny and extensive collinearity of large sequence blocks (Figure 6). This provides opportunities for genomic comparisons and translation to identify candidate genes for traits of interest in the two species.
|
study
| 100.0 |
Graphical view of the genome structure of Trifolium subterraneum L. A) Syntenic relationship between Medicago truncatula (left) and T. subterraneum (right) with synteny links and synteny density histograms; B) T. subterraneum gene densities C) T. subterraneum SNP densities; D) T. subterraneum QTLs mapped for important traits using the high‐density SNP linkage map E) Names for the QTLs mapped for important traits using the high‐density SNP linkage map [FL Flowering time, LM Leaf marks, CP Calyx pigmentation, SH Stem Hairiness, FO Formononetin, GT Genistein, BCA Biochanin A and HS Hardseededness].
|
study
| 99.8 |
Phenotypic information for eight important traits described by Ghamkhar et al. (2011) (flowering time, hardseededness, leaf marks, calyx tube pigmentation, stem hairiness and the isoflavonoids, formononetin, genistein, and biochanin A) was associated with specific regions in the revised high‐density SNP linkage map (Tables S18 and S19) previously described by Hirakawa et al., 2016;. Significant associations were then mapped onto the advanced assembly (Tsub_Refv2.0) to identify possible candidate genes (Figure 6; Table 2; Table 3). The effective anchoring of sequence scaffolds onto the high‐density SNP linkage map in a high‐quality chromosome‐level genome assembly was a major factor in the identification of loci governing key traits. For example, QTLs for leaf marks (LM), calyx pigmentation (CP) and the isoflavonoids [formononetin (FO), genistein (GT) and biochanin A (BCA)] were found to map adjacently to a region on Chr 5. The linkage between these traits was demonstrated by Francis and Millington (1965), who used mutagenesis on cv. Geraldton, which has a high formononetin content, a prominent leaf mark and anthocyanin pigmentation of leaves, calyx tubes, stipules and stems to produce the low formononetin cv. Uniwager, which concomitantly lost its leaf mark and anthocyanin pigmentation. Co‐localization of the QTLs identified for the traits in the present study explains the observation of breeders that isoflavone content is linked to pigmentation traits (leaf and calyx) (Figure 6; Table 3) (Francis and Millington, 1965). This information illustrates links between the new assembly, the high‐density linkage map and key quantitative traits that can assist future marker‐assisted selection and comparative mapping with other species to improve forage legumes and increase livestock productivity.
|
study
| 99.94 |
To improve the draft assembly of subterranean clover, the BNG Irys® system provided affordable, high‐throughput physical maps of high contiguity to validate the draft and extend existing scaffolds. Effective anchoring of sequence scaffolds to genetic linkage groups, coupled with the use of the BioNano system, resulted in a high‐confidence chromosome‐level genome assembly. Tissue‐specific transcript profiling with RNA‐Seq technology delivered gene expression data of high value for gene annotation of the assembly (Tsub_Refv2.0) and transcriptional dynamics to understand tissue‐specific pathways. This new genomic information is the key to identifying loci governing traits that allow marker‐assisted breeding in subterranean clover for comparative mapping with other species and the identification of tissue‐specific gene promoters for biotechnological improvement of forage legumes.
|
study
| 99.94 |
Suspensions of intact cell nuclei were prepared according to Vrána et al. (2016). Briefly, mature dry seeds of subterranean clover (Trifolium subterraneum L.) cv. Daliak (approx. 20 g) were germinated on moistened paper towels in the dark at 25° ± 0.5 °C until roots achieved 2–3 cm in length. Roots were cut to 1 cm from the apex and transferred into 25 mL formaldehyde fixative (2% v/v) for 20 min at 5 °C, followed by three 5‐min washes in Tris buffer. Finally, the root tips (approx. 40) were cut, transferred into 1 mL IB buffer (Šimková et al., 2003) and homogenized using a mechanical homogenizer (blender) at 13 000 RPM for 18 s. The crude homogenate was filtered through 50‐μm (pore size) nylon mesh and stained with DAPI (2 μg/mL final concentration). A total of fifteen samples were prepared.
|
study
| 100.0 |
For transcriptome work, subterranean clover cv. Daliak plants were sown on 1 May 2015 in the field at Shenton Park, Western Australia (31o57′S, 115o50′E). Five different tissue types (roots, stem and peduncles, leaf and petioles, flowers and developing seeds) were harvested from a single Daliak plant on 24 September, when it was flowering and setting seeds. All samples were taken between 10.00 am and noon to eliminate any diurnal variations.
|
study
| 99.94 |
Cell nuclei were purified by a FACSAria II SORP flow cytometer and sorter (BD Biosciences, San Jose, CA) equipped with UV laser for DAPI excitation. Populations of G1 nuclei were sorted in batches of 700 000 into 1.5‐mL polystyrene tubes containing 660 μL IB buffer. In total, four batches of nuclei were sorted and each batch was used for preparation of one 20‐μL agarose miniplug. DNA embedded in miniplugs was purified by proteinase K (Roche, Basel, Switzerland) treatment according to Šimková et al. (2003). The miniplugs were washed four times in wash buffer (10 mm Tris, 50 mm EDTA, pH 8.0) and five times in TE buffer (10 mm Tris, 1 mm EDTA, pH 8.0) and then melted for 5 min at 70 °C and solubilized with GELase (Epicentre, Madison, WI) for 45 min. A drop dialysis against TE buffer (Merck Millipore, Billerica, MA) was performed for 90 min to purify DNA from any residues and subsequently quantified using a Quant‐iTTM PicoGreen® dsDNA assay (Thermo Fisher Scientific, Waltham, MA).
|
study
| 99.94 |
Total RNA from all the tissue samples (Figure S4) was extracted using the Spectrum™ Plant Total RNA Kit (Sigma‐Aldrich, USA) following the manufacturer's instructions. Aliquots of purified RNA were stored at −80 °C. The concentration of RNA was confirmed using Qubit fluorometer with Qubit RNA assay kit (Life Technologies, Australia). The integrity of total RNA was determined by electrophoretic separation on 1.2% (w/v) denaturing agarose gels. Sequencing libraries were constructed using 500 ng of total RNA with a TruSeq® Stranded Total RNA Sample Prep Kits with Ribo‐Zero (Illumina, San‐Diego, USA) following the manufacturer's instructions. Library concentrations were measured using a Qubit fluorometer with Qubit dsDNA BR assay kit (Life Technologies, USA) and Agilent high‐sensitivity DNA chips (Agilent Technologies, USA). The amplified libraries were pooled in equimolar amounts, and quality was assessed with Agilent high‐sensitivity DNA chips (Agilent Technologies, USA). Paired‐end 100‐bp x 2 sequencing was performed with HiSeq2000 (Illumina, San‐Diego, USA).
|
study
| 100.0 |
The genome sequence of subterranean clover (TSUd_r1.1) (Hirakawa et al., 2016) was analysed with Nickers software to assess the frequency of recognition sites for four different nicking enzymes (Nt.BbvC1, Nt.BsPQ1, Nb.bsm1 and Nb.BsrD1). The optimal labelling frequency was calculated for endonuclease Nt.BsPQ1 (7.6 sites/100 kb). DNA was then processed using NLRS protocol using the IrysPrep® Reagent Kit (BioNano Genomics, San Diego, CA) following manufacturer's instructions. DNA was nicked using 8U of Nt.BspQ1 (New England BioLabs, Beverly, MA) for two hours at 37 °C in NEBuffer 3. The nicked DNA was labelled with a fluorescent‐dUTP nucleotide analogue using Taq polymerase (New England BioLabs) for one hour at 72 °C. After labelling, the nicks were ligated with Taq ligase (New England BioLabs) in the presence of dNTPs for 30 min at 37 °C. The backbone of the labelled DNA was stained with IrysPrep® DNA Stain (BioNano Genomics). The NLRS DNA concentration was measured again with the Quant‐iTTM PicoGreen® dsDNA assay.
|
study
| 100.0 |
Labelled and stained DNA was loaded on the Irys chip, and four consecutive runs were performed (each run consisting of 30 cycles). A total of 489.8 Gb data were generated, of which 221.7 Gb exceeded 150 kb, the threshold for map assembly. These filtered data (>150 kb), corresponding to 401× coverage of the subterranean clover genome, were compiled from 994 895 molecules with N50 of 215.4 kb. De novo assembly of the BioNano map was performed by a pairwise comparison of all single molecules and graph building (Cao et al., 2014). A P‐value threshold of 1e−9 was used during the pairwise assembly, 1e−10 for extension and refinement steps and 1e−15 for a final refinement.
|
study
| 100.0 |
The complete pipeline of the Stitch algorithm by Shelton et al. (2015) (https://github.com/i5K-KINBRE-script-share/Irys-scaffolding/blob/master/KSU_bioinfo_lab/assemble_XeonPhi/assemble_XeonPhi_LAB.md) was run on the online cluster zythos provided by the Pawsey Supercomputing Center, Western Australia. The first draft genome assembly of T. subterraneum (TSUd_r1.1) (Hirakawa et al., 2016) was used as the reference genome. Super‐scaffolds were generated, and BioNano IrysView was used to examine the new assemblies and the alignments between BioNano consensus maps and the in silico scaffolds. Based on these alignments, misassemblies were identified in TSUd_r1.1 (Figure 1b).
|
study
| 100.0 |
All RNASeq libraries were Illumina, San‐Diego, USA TruSeq adapter trimmed and quality trimmed (sliding window, minimum quality score: 20) using Trimmomatic v 0.36 (Bolger et al., 2014). Trimmed libraries were aligned to the advanced reference sequence using HISAT2 v 2.0.1 (insert size 0 to 1000) (Kim et al., 2015). The resulting SAM files were converted to BAM format using samtools v1.3 (Li et al., 2009).
|
study
| 99.9 |
The resulting predicted proteins were aligned to several databases using blastp v 2.2.31+ (Altschul et al., 1990) (minimum e‐value 1e−10). The database used were Swiss‐Prot and TrEMBL downloaded on March 13 2016 (Boeckmann et al., 2003) and all 2 542 385 predicted proteins downloaded from Phytozome v11 on 14 March 2016 (Goodstein et al., 2012). For each predicted protein, the hit with the highest score and lowest e‐value was chosen as annotation.
|
study
| 100.0 |
GO terms were assigned to proteins by manually transferring the GO terms for Swiss‐Prot IDs using the UniProt‐GOA database downloaded on 13 March 2016 (Huntley et al., 2015). KEGG K numbers were assigned to all predicted proteins using BLASTKOALA (taxonomy group: Plants, KEGG GENES database: family_eukaroytes) (Kanehisa et al., 2016).
|
study
| 99.94 |
Tissue expression was estimated using kallisto v 0.42.4 (Bray et al., 2015). All genes with total transcripts per million (TPM) count below 1 or where more than two tissue types had a TPM of 0 were removed. To make log‐normalization possible, 0.25 was added to the remaining TPM values. Expression was log‐normalized and clustered into 12 clusters using Mfuzz v. 2.30.0 (Futschik and Carlisle, 2005) (m = 1.25).
|
study
| 100.0 |
Clustering was based on 100 genes with highest sum of all TPM (transcripts per million) of the five tissue types. The dendrogram of the selected genes [vertical axis] is visualized along with the expression patterns used to cluster the five tissue samples, i.e. root (RT), developing seed (DSD), leaf and petioles (LF_PT), stem and peduncles (ST_PD) and flower (FL) [horizontal axis]. There are five GO‐based clusters named G1, G2, G3, G4 and G5 that contain more than one gene. The GO clusters are separated by a horizontal bar in the heat map. Genes without annotations are omitted from the heat map. Over‐expressed genes are shown with red colour and under expressed genes with white or yellow colour, respectively (Figure 3).
|
study
| 100.0 |
The linkage map was constructed using MultiPoint 3.3 (http://www.multiqtl.com/) as described in (Hirakawa et al., 2016). The complete set of 188 F2 lines of a biparental population 92S80 (cv. Woogenellup x cv. Daliak) phenotypic data as reported by Ghamkhar et al., 2011 was used for QTL screens for the following morphological and agronomic traits: levels of the oestrogenic isoflavones, formononetin (FO), genistein (GT) and biochanin A (BCA); days to first flowering (FT); leaf marks (LM); pigmentation of calyces (CP); hairiness of stems (SH); and hardseededness (HS). Levels of the isoflavones FO, GT and BCA were measured using the technique of Francis and Millington (1965). FT was measured as the number of days from sowing to appearance of the first flower. The morphological traits, LM, CP and SH, were scored 120 days after sowing, using the rating systems given in Nichols et al., (1996). Hardseededness was measured in the laboratory using the method of Quinlivan (1961).
|
study
| 100.0 |
QTL screens for the traits reported (Table 3) were conducted using an inclusive composite interval mapping (ICIM) approach implemented in QTL IciMapping v4.0 (Wang et al., 2014). Missing phenotypes were deleted using the ‘Deletion’ command in the software. The walking speed was set at 1 cm. A suitable probability for entering marker variables in stepwise regression was chosen so that the variation explained by the model approximated the trait heritability. The regression model was then used for background genetic variation control in the ICIM QTL mapping. The LOD was calculated using 1000 permutations, with a Type 1 error being 0.05, and significant QTLs were defined accordingly.
|
study
| 100.0 |
A phylogenetic tree was constructed, based on published predicted protein sequences for 13 legume genomes, including those of the newly annotated Trifolium subterraneum L. and A. thaliana. Mash v1.1 (Ondov et al., 2016) was used to calculate a distance matrix using k‐mer counting and pairwise mutation distance estimation, using the predicted amino acids from the published whole genome sequences for all the legumes, except Cajanus cajan, for which only the transcriptome assembly was used (as the whole genome assembly is not available). The tree was then constructed using UPGMA as implemented in the R‐package ‘phangorn’ v2.0.3. Shared and unique gene families were called using BLAST and OrthoMCL (Li et al., 2003). OrthomclToVenn (https://github.com/philippbayer/orthomcltovenn) was used to calculate the number of unique and shared genes between: (i) the Papilionoideae species and A. thaliana; (ii) the Millettioid, Galegoid and Dalbergioid clades; (iii) A. thaliana and the Lotus, Trifolium, Medicago and Cicer genera; and (iv) T. subterranean and T. pratense.
|
study
| 100.0 |
P.K. conceived and performed the research and wrote the manuscript with contributions from P.E.B, Z.M, J.V, Y.Y, R.A, D.E, J.B, P.N, J.D and W.E. The BioNano Irys®;System genome mapping experiments were led and designed by J.D and performed and analysed by Z.M, J.V and Y.Y. P.E.B and Y.Y did the bioinformatics analysis and helped with the figure preparations. All authors read the manuscript and approved the content.
|
other
| 99.94 |
Table S14 Distance matrix and pairwise mutation distance estimation calculated by Mash v1.1 (preprint: http://biorxiv.org/content/early/2016/04/19/029827) using the published whole genome sequences for all the legumes except the transcriptome assembly for Cajanus cajan as a whole genome assembly not available.
|
other
| 95.0 |
A globally accepted standard regimen for advanced gastric cancer (AGC) has not been established. S-1, a fourth-generation oral fluorouracil (5-FU), has opened new perspectives with simplicity and convenience over the traditional backbone of chemotherapy for AGC, 5-FU.
|
review
| 95.25 |
S-1 contains tegafur/gimeracil/oteracil. Gimeracil reduces the degradation of 5-FU by inhibiting dihydropyrimidine dehydrogenase (DPD), and oteracil improves its gastrointestinal tolerability by inhibiting orotate phosphoribosyltransferase (OPRT). S-1 plus cisplatin is recommended as standard treatment by the Japanese Gastric Cancer Association and has been approved in Asian and European countries . However, many patients cannot tolerant the toxicity of the widely accepted S-1 plus platinums .
|
review
| 96.2 |
One of the potential partners of S-1 is leucovorin (LV), known to enhance the efficacy of 5-FU by forming a ternary complex with 5-FU metabolite and thymidylate synthase (TS) which prolongs the inhibition of TS and blocks DNA synthesis. It was proved UFT, a third-generation oral 5-FU, had favorable activity and tolerability when combined with LV in AGC patients . A phase II study of S-1 plus LV (S-1/LV) has demonstrated promising efficacy and acceptable safety in patients with metastatic colorectal cancer (mCRC) . A randomized phase II study reported promising efficacy of S-1/LV and S-1/LV plus oxaliplatin (SOL) for AGC, compared with S-1 plus cisplatin . However, there has been no one-arm phase I/II study for S-1/LV in gastric cancer patients. Interindividual variation in pharmacogenetics of the S-1 metabolic pathway can affect the extent of S-1 metabolism and impact the efficacy and toxicity of S-1-based chemotherapy. Published studies all used “candidate” pharmacogenetic factors to predict the outcomes of AGC patients with S-1 or S-1-based chemotherapy, none of which used S-1 pharmacogenetic pathway approach, which means none integrated CYP2A6 polymorphism, 5-FU metabolic enzymes, and pharmacokinetics at the same time [6–8]. However, those factors do not act in isolation. What is more, the results and methods varied across studies. Finally, predictive values of these candidates might or might not be overcome with drugs combined with S-1.
|
review
| 99.9 |
Preclinical studies have highlighted the importance of TS, the cellular target of 5-FU–folinic acid mechanism of action . Low TS was reported as a predictor of high response and long survival for AGC patients with high-dose 5-FU/LV . In colorectal cancer, both intratumoral TS and DPD gene expressions had been reported to be predictive for the effectiveness of 5-FU/LV or UFT/LV [11, 12]. However, there was no research on the prediction of OPRT, TS, TP, or DPD for efficacy or toxicity of S-1/LV or UFT/LV in gastric cancer. What is more, no available evidence existed on the predictive potential of CYP2A6 polymorphism for 5-FU/LV, UFT/LV, or S-1/LV treatment in gastric cancer. Interestingly, the use of more than single gene expression, such as the combination of low TS with high OPRT, or high OPRT/TS had been reported to be even more predictive of responders to S-1 or S-1 plus cisplatin in gastric cancer patients [13–15].
|
review
| 99.8 |
This current phase II study of S-1/LV is deemed necessary to explore its efficacy and safety as first-line chemotherapy for AGC patients. Considering the frequent grade 3 toxicities, dose reduction, and period prolongation in that previous trial of mCRC , the 4-week S-1/LV was modified as 2 week in this study. As the first one, this study used S-1 pharmacogenetic pathway approach and integrated CYP2A6 polymorphism, DPD, TS, OPRT, thymidine phosphorylase (TP), and 5-FU pharmacokinetics to identify the subset of patients benefiting more, suffering less from S-1/LV.
|
study
| 98.75 |
The eligibility criteria included (1) histologically confirmed metastatic or recurrent gastric cancer; (2) at least one measurable lesion; (3) an age of ≥18; (4) adequate oral intake; (5) no previous antitumor therapy within 5 years (adjuvant chemotherapy without S-1 was allowed if finished ≥6 months before enrollment); (6) an Eastern Cooperative Oncology Group performance status <2; and (7) adequate bone marrow, hepatic, and renal function.
|
other
| 99.6 |
The exclusion criteria included (1) known hypersensitivity to any of the study drugs, or usage of drugs interacting with S-1; (2) serious concomitant conditions; (3) extensive bone, brain, or meningeal metastasis; (4) another synchronous cancer; (5) surgery within 3 months; (6) participating in other clinical studies; (7) pregnant women; (8) subjects with reproductive potential who were unwilling to use an effective method of contraception.
|
study
| 90.5 |
S-1 (20-mg capsules) and LV (25-mg tablets) were provided by DaPeng Co., Ltd, Japan. All patients were orally treated with S-1 in doses of 40 mg (body surface area (BSA) < 1.25 m2), 50 mg (1.25 ≤ BSA < 1.50 m2) and 60 mg (BSA ≥ 1.50 m2) b.i.d. in combination with LV given simultaneously at a fixed dose of 25 mg b.i.d. on days 1–7, followed by a 7-day rest.
|
other
| 99.9 |
Clinical and laboratory examinations were carried out within 7 days before enrollment and each cycle of chemotherapy afterward. Tumor measurement was conducted on the basis of computed tomographic scans, within 15 days before enrollment and every 3 cycles afterward, according to Response Evaluation Criteria in Solid Tumors guidelines (version 1.1). Patients were considered response-assessable if they had overt clinical or radiological evidence of early PD within the first three cycles. All treatment-related adverse events (AEs) were categorized according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events version 4.0 (NCI-CTCAE v4.0).
|
other
| 99.7 |
Peripheral blood was prospectively, anonymously sampled for each patient on the first day of the first cycle at 0 h (pre-dose) and 0.5, 1, 2, 4, 8, 24 h post-S-1/LV morning dosing. The 2 ml plasma at 0 h was separated for measuring baseline protein expression level of DPD, OPRT, TS, and TP. The 4-ml blood cells at 0 h were separated for CYP2A6 gene polymorphism. The 2 ml plasma at 0, 0.5, 1, 2, 4, 8, 24 h was separated for plasma concentrations of 5-FU.
|
study
| 99.94 |
Polymerase chain reaction (PCR)–restriction fragment length polymorphism were used to determine common variant alleles that affect CYP2A6 activity or expression in Asian population (CYP2A6*1A, *1D, *9), *13, and the wild-type allele (CYP2A6*1), as previously described .
|
study
| 100.0 |
Plasma concentrations of 5-FU at 0, 0.5, 1, 2, 4, 8, 24 h were measured using negative ion chemical ionization gas chromatography mass spectrometry. Pharmacokinetic parameters including area under the curve (AUC0–24h), maximum concentration (Cmax), time taken to reach maximum concentration (Tmax), half-time (T1/2), area under the first moment curve (AUMC0–24h), mean resistance time (MRT0–24h), and plasma clearance (CL) were derived with non-compartmental methods using WinNonlin version 3.1 .
|
study
| 99.94 |
The Kaplan–Meier method with two-sided log-rank test was used to estimate the distribution of time to events. PFS was determined from the date of treatment to progression (PD) or death from cancer. TTF was determined from the date of treatment to PD, death, refusal, or interruption due to AEs. OS was calculated from the date of treatment to death from any cause or the last date of follow-up. Receiver operating characteristic (ROC) curve was used for cutoff values of DPD, OPRT, TP, TS, and 5-FU pharmacokinetics in the predictive analyses of response or grade 3–4 AEs. X-Tile software was used for cutoff values of them in the predictive analyses of survival. Logistic regression was used for predictive analyses of response or grade 3–4 AEs, and Cox proportional hazards model was used for predictive analyses of survival. Statistical analyses were performed using the SPSS 19.0.
|
study
| 100.0 |
Between July 2011 and July 2012, a total of 39 eligible patients were enrolled from the Sun Yat-sen University Cancer Center. Clinical cutoff date was March 20, 2014. The median follow-up was 23.13 months. The baseline patient characteristics are summarized in Table 1. The median number of treatment cycles was 6 (range 1–20), with a total of 252. The median treatment period was 3.03 months (range 0.47–12.00). The median relative dose intensity was 91% for S-1 and 100% for LV.Table 1Baseline patient characteristicsCharacteristicsS-1 plus LV (N = 39)No.%Gender (male/female)28/1171.8/28.2Age (years, median, range)55 (21–83)Body surface area (m2, median, range)1.51 (1.33–1.94)ECOG performance status = 139100Primary tumor location Proximal1128.2 Body717.9 Antrum1025.6 Multiple/diffuse1128.2Histology Well differentiated37.7 Moderately differentiated717.9 Poorly differentiated2461.5 Mucinous410.3 Signet-ring cell12.6Lauren classification Diffuse type411.1 Intestinal type1333.3 Mixed type2255.6Her-2 gene type Positive615.4 Negative3384.6Site of metastases Liver1435.9 Lung37.7 Lymph nodes2974.4 Peritoneum1435.9 Bone410.3No. of metastatic/recurrent sites 11846.2 21641.0 3512.8Prior surgery Curative gastrectomy820.5 Palliative gastrectomy/metastectomy512.9 Exploration/bypass410.3 No2256.4Prior adjuvant chemotherapy Yes615.4 No3384.6
|
study
| 96.75 |
All 39 patients were evaluable. No patient had a complete response, 16 had partial response (PR), 14 had stable disease, and 9 had progressive disease. The ORR was 41.0% (95% confidence interval (CI) 24.9–57.2%), and the DCR was 76.9% (95% CI 63.1–90.8%). The median time to response was 1.70 (range 1.40–3.00) months (m).
|
other
| 99.4 |
The median PFS was 4.13 (95% CI 3.44–4.83) m. The median TTF was 3.70 (95% CI 2.60–4.80) m. The median OS was 11.40 (95% CI 7.76–15.05) m (Fig. 1).Fig. 1Kaplan–Meier curves for the entire population. a Progression-free survival, b time to failure, and c overall survival
|
study
| 89.56 |
Major AEs included myelosuppression (74.4%), gastrointestinal reactions (89.7%), and pigmentation (53.8%); however, they were generally mild and no treatment-related deaths occurred. Anemia (71.8%) was common, and thrombocytopenia was rare (0%). Anorexia (64.1%) was common, and diarrhea was not. Grade 3 AEs occurred in 13 patients (33.3%), and grade 4 AEs occurred in 1 of them (2.6%). Each type of the grade 3–4 AEs occurred in only 1–3 patients, with gastrointestinal reactions in 15.4% and myelosuppression in 10.3% (Table 2).Table 2Adverse eventsAdverse eventsS-1 plus LV (N = 39)Grade 1 (%)Grade 2 (%)Grade 3 (%)Grade 4 (%)All grade (%)Leukopenia6 (15.4)7 (17.9)1 (2.6)014 (35.9)Neutropenia4 (10.3)7 (17.9)1 (2.6)1 (2.6)12 (33.3)Anemia17 (43.6)10 (25.6)1 (2.6)028 (71.8)Thrombocytopenia00000Asthenia9 (23.1)0009 (23.1)Anorexia19 (48.7)5 (12.8)1 (2.6)025 (64.1)Nausea11 (28.2)1 (2.6)0012 (30.8)Vomiting3 (7.7)2 (5.1)1 (2.6)06 (15.4)Diarrhea9 (23.1)1 (2.6)1 (2.6)011 (28.2)Abdominal pain5 (12.8)4 (10.3)2 (5.1)011 (28.2)Skin rash5 (12.8)0005 (12.8)Hand–foot syndrome3 (7.7)1 (2.6)004 (10.3)Pigmentation16 (41.0)5 (12.8)0021 (53.8)Stomatitis6 (15.4)3 (7.7)1 (2.6)010 (25.6)Blurred vision4 (10.3)0004 (10.3)Lacrimation increased4 (10.3)0004 (10.3)Tinnitus1 (2.6)0001 (2.6)ALT elevation4 (10.3)2 (5.1)3 (7.7)09 (23.1)AST elevation4 (10.3)1 (2.6)2 (5.1)07 (17.9)Hypoalbuminemia12 (30.8)2 (5.1)0014 (35.9)Proteinuria1 (2.6)0001 (2.6)
|
other
| 97.25 |
The main reasons for S-1 dose decrease in the 3 patients (7.7%) were grade 3 diarrhea, anorexia, and stomatitis, respectively. The main reasons for dose interruption in the 3 patients (7.7%) were grade 3 vomiting, abdominal pain, and liver enzyme elevation, respectively. The median number of chemotherapy cycles before S-1 dose decrease and interruption was 4 (range 2–6) and 5 (range 2–10), respectively. Course was prolonged by 7 days until the grade 3 liver enzyme elevation decreased to grade 1 in 1 patient. Five patients (12.8%) discontinued treatment before progression not due to AEs, with a median number of treatment cycles as 3 (range 1–6) (supplementary Table S1). Second-line treatment was given to 21 (53.9%) of the 39 patients, among whom 5.1% received palliative surgery while 48.8% received oxaliplatin-based, irinotecan-based, or taxane-based chemotherapy (supplementary Table S2).
|
other
| 97.56 |
For the entire population, baseline plasmic protein expression of DPD, OPRT, TP, TS and their ratios OPRT/DPD, OPRT/TP, OPRT/TS, OPRT/TP + TS, OPRT/DPD + TP, OPRT/DPD + TS, and OPRT/DPD + TS + TP are summarized in supplementary Table S3. The genotypes and allele frequencies of CYP2A6 are shown in supplementary Table S4. Mean plasma concentration–time curve of 5-FU for the entire population is shown in supplementary Figure S1. The AUC0–24h, Cmax, Tmax, T1/2, AUMC0–24h, MRT0–24h, CL were determined for each patient and for the entire population.
|
study
| 100.0 |
There were 16 responders and 23 non-responders. By multivariate logistic regression analysis, high OPRT/TS (>1.246 vs. ≤1.246, odds ratio (OR) 16.962, 95% CI 1.781–161.581, P = 0.014) and peritoneal metastasis (vs. liver metastasis, OR 25.604 (1.852–353.979), P = 0.016) were independently predictive of responding. OPRT/TS differed between responders and non-responders (median ± SD 1.442 ± 0.091 vs. 1.158 ± 0.133, P = 0.037) and response rates differed between patients with high OPRT/TS and low OPRT/TS (>1.246 vs. ≤1.246, 57.1 vs. 22.2%, P = 0.040). Figure 2a shows the ROC curve of OPRT/TS for predicting response.Fig. 2 a ROC curve of OPRT/TS for predicting response and b the ROC curve of OPRT/DPD for predicting grade 3–4 AEs. ROC receiver operating characteristic, OPRT orotate phosphoribosyltransferase, TS thymidylate synthase, DPD dihydropyrimidine dehydrogenase, AEs adverse events
|
study
| 100.0 |
a ROC curve of OPRT/TS for predicting response and b the ROC curve of OPRT/DPD for predicting grade 3–4 AEs. ROC receiver operating characteristic, OPRT orotate phosphoribosyltransferase, TS thymidylate synthase, DPD dihydropyrimidine dehydrogenase, AEs adverse events
|
other
| 99.8 |
Thirteen patients experienced grade 3–4 AEs and 26 patients did not. By univariate logistic regression analysis, high OPRT (accuracy 79.0%), high OPRT/DPD (80.8%), high OPRT/TP (77.8%), high OPRT/TS (71.0%), high OPRT/DPD + TP + TS (78.7%), high OPRT/DPD + TS (77.8%), high OPRT/TP + TS (77.2%), and high OPRT/DPD + TP (80.1%) were all associated with grade 3–4 AEs. OPRT/DPD exhibited the highest accuracy (80.8%). By multivariate logistic regression analysis, high OPRT/DPD [>0.754 vs. ≤0.754, OR 15.566 (1.490–162.605), P = 0.022] was independently predictive of grade 3–4 AEs. The rates of grade 3–4 AEs differed between patients with high OPRT/DPD and low OPRT/DPD (>0.754 vs. ≤0.754, 55.0 vs. 10.5%, P = 0.006). Figure 2b shows the ROC curve of OPRT/DPD for predicting grade 3–4 AEs.
|
study
| 99.94 |
Multivariate analysis with a Cox proportional hazards model demonstrated that high AUC0–24h of 5-FU and metastatic/recurrent sites ≤2 (vs. >3) were significant predictors of prolonged PFS (supplementary Table S5). Similarly, multivariate analysis demonstrated high AUC0–24h of 5-FU was borderline significant predictor of prolonged TTF (supplementary Table S6).
|
study
| 100.0 |
The median PFS differed significantly between patients with high and low AUC0–24h of 5-FU (5.40 vs. 3.70 m, P = 0.022, Fig. 3a), and the median TTF differed borderline between patients with high and low AUC0–24h of 5-FU (4.13 vs. 3.10 m, P = 0.054, Fig. 3b).Fig. 3 a Kaplan–Meier curve of progression-free survival according to AUC0–24h of 5-FU and b the Kaplan–Meier curve of time to failure according to AUC0–24h of 5-FU. AUC areas under the curve
|
study
| 100.0 |
Lower baseline plasmic DPD [>119.200 vs. ≤119.200, harzard ratio (HR) 2.931 (1.155–7.433), P = 0.024] was significantly independent predictor of prolonged OS (supplementary Table S7; Fig. 4).Fig. 4Kaplan–Meier curve of overall survival according to baseline plasmic DPD expression. DPD dihydropyrimidine dehydrogenase
|
study
| 99.94 |
To our knowledge, the current study is the first one-arm and the second phase II trial to evaluate efficacy and toxicity of S-1/LV chemotherapy in AGC patients, which is the first to predict outcomes using S-1 pharmacogenetic pathway approach, on the basis that LV is but a cofactor entering the 5-FU metabolism. The enrollment was between July 2011 and July 2012, which was similar to between October 2011 and December 2012 of that randomized phase II study of S-1/LV versus SOL versus SP for AGC .
|
study
| 87.06 |
We reported S-1/LV regimen yielded promising ORR, PFS, TTF, and OS, without combination with platinum, taxane, irinotecan, or trastuzumab as first-line treatment. In previous phase II studies of S-1 monotherapy in AGC and previous S-1 monotherapy arms of phase III JCOG9912 study, SPIRITS study, SC-101 study, GC0301/TOP-002 study, and START study, conventional dose for 4 weeks followed by 2-week rest was usually given [18–22]. The 2-week S-1/LV here was generally more effective, with a less dose intensity than S-1 monotherapy in these studies. LV can enhance the efficacy of 5-FU by maintaining the plasma 5-FU concentration . Similarly with the previous trial of S-1/LV in mCRC , S-1/LV regimen here also demonstrated better efficacy and safety compared with previously reported UFT/LV in AGC [3, 24]. S-1/LV also showed more potential than 5-FU/LV plus oxaliplatin that we reported in terms of efficacy and safety .
|
study
| 99.9 |
To reduce the duration of effective drug concentration and keeping an appropriate rest period could be effective method of improving safety. So we modified the schedule as 1-week S-1/LV, followed by 1-week rest period, the same as S-1/LV group in that randomized phase II study. The efficacy S-1/LV here was quite comparable with S-1/LV group in that randomized phase II study, in terms of ORR (41.0 vs. 43%), PFS (4.13 vs. 4.2 m), and TTF (3.70 vs. 4.1 m) . Fewer patients in our study received the second-line treatment (53.9 vs. 77%), which helped explain the relatively not so promising OS. Another reason was that there were 5 patients discontinuing treatment before progression not due to adverse events. Meanwhile, S-1/LV in this study saw very satisfactory safety. The frequencies of each type of AEs were generally lower than those in the phase II trial of mCRC and comparable with those in the randomized phase II trial of AGC [4, 5]. Encouragingly, grade 4 AEs occurred in 1 patient (2.6% neutropenia), similarly for the randomized study (2% grade 4 neutropenia and 2% leucopenia). Each type of the grade 3 AEs occurred in only 2.6–7.7% patients, similar to 2–13% for the randomized study . In our study, significantly fewer patients experienced dose reduction, or delayed courses , compared to patients with S-1 monotherapy in previous phase III trials [19, 26]. In our study, gastrointestinal reactions were more common than myelosuppression both in total (89.7 vs. 74.4%) and in grade 3–4 (15.4 vs. 10.3%), and dose reductions or interruptions were chiefly due to diarrhea, stomatitis, anorexia, and vomiting. We also observed dose-limiting toxicity was shifted from hematological to gastrointestinal when S-1 was administered with LV . Probably the capacity of oral intake and gastrointestinal tolerability would be the important indications for this regimen.
|
study
| 99.94 |
Japan Gastric Cancer Association guideline states that S-1 alone could be considered for patient who is not suitable for S-1 plus cisplatin therapy . Previous studies showed S-1 monotherapy could be a reasonable option in the treatment of elderly patients. In our study, the ORR, PFS, OS were sound for patients of age >70 (47.1%, 6.00, 11.40 m) without difference compared with those of age ≤70 (36.4%, 4.00, 11.20 m) and seemed better than previously reported S-1 monotherapy (ORR 14.3–26.3%) and UFT/LV (ORR 22%) for elderly AGC patients [28–30]. The oral convenience makes the S-1/LV regimen extremely useful clinically, especially for elderly patients.
|
study
| 99.94 |
Genotyping the peripheral blood for CYP2A6 polymorphism, quantifying plasmic protein expression of DPD, TS, TP, and OPRT with ELISA, is more optimal, convenient, quicker than evaluating genes, mRNAs, or proteins in tumor tissue, especially when 5-FU plasmic pharmacokinetics were to be integrated at the same time. Thus, there may be more clinical accessibility and prospects. What is more, the cutoff levels were determined by standard statistic analysis, not the simple median or mean. The plasmic expression of DPD in this study was consistent with previously reported with ELISA; however, no report for plasmic OPRT, TP, or TS was available . The frequencies of CYP2A6 alleles in this population were compatible with other Asian population [31–33]. We especially examined the CYP2A6*13, because its function in Asian patients was unclear. The 0% of CYP2A6*13 again proved it is rare in Asia.
|
study
| 100.0 |
In clinical setting, numerous studies have reported a low TS, low TP, or a high OPRT expression contributed to a high sensitivity to UFT, S-1, or S-1-based treatment in gastric cancer patients, with or without influencing the PFS or OS [34–38]. There were also evidences that low TS was a predictor of high response for AGC patients with 5-FU/LV, or 5-FU/LV plus cisplatin/oxaliplatin [10, 39, 40]. However, most studies failed to find the DPD expression related to either response, PFS, or OS and less studies did demonstrate high DPD mRNA was predictor of poor OS, not ORR or PFS in AGC [35, 41, 42]. These clinical findings reflect theoretical roles of TS and OPRT. We found high OPRT/TS alone significantly predicted responding. The resultant high OPRT/TS here revealed preferential use of the OPRT pathway versus TS pathway during 5-FU metabolism. In humans, the preferential use of the OPRT pathway was revealed to correlate with a higher sensitivity to 5-FU . Ichikawa et al. and Tanemura et al. [13, 14] both reported the combination of high OPRT and low TS was more predictive of responders to S-1 or S-1 based chemotherapy in gastric cancer patients than either alone, while the low TP was not. Apart from the above PCR or ELISA methods, the quantitative double-fluorescence immunohistochemistry method, reported by Hashiguchi K, was used to access the protein expressions and their ratios quantitatively and found a significant correlation between OPRT/TS, OPRT/DPD, or OPRT/(TS + DPD) and response to S-1 in the AGC patients, among which OPRT/TS showed the strongest correlation with the clinical response . These three studies generally agree with our finding.
|
study
| 99.94 |
In this current study, low baseline plasmic DPD was not related to response; however, it was related to long OS, compatible with previous study that high intratumoral DPD mRNA was predictor of poor OS, not ORR or PFS . Previous studies reported clinical response to S-1 in gastric cancer was not influenced by intratumoral DPD expression . It can be explained that S-1 has antitumor activity even in tumor with high expression of DPD because of the inhibition of DPD by CDHP . Even though it had no prediction of response, low DPD did relate to long OS. Firstly, S-1 enabled high 5-FU concentrations to be maintained in blood for long periods of time by inhibiting of DPD and 5-FU maintenance was a reason of long survival. Secondly, low levels of intratumoral DPD have been generally shown to predict long survival in gastric cancer patients treated with 5-FU-based chemotherapy . Many patients here received second-line chemotherapy comprising of 5-FU with platinum or other drugs.
|
study
| 99.94 |
We found plasmic OPRT (P = 0.024), TS (P = 0.044) expression significantly inversely, while DPD (P = 0.073), TP (P = 0.080) borderline inversely correlated with AUC0–24h of 5-FU. The number of CYP2A6 gene variants (P = 0.889) did not correlate with AUC0–24h. This was consistent with that CYP2A6 gene correlated with tegafur pharmacokinetics, but not with 5-FU pharmacokinetics . AUC0–24h of 5-FU, not OPRT/TS, was predictive of PFS revealed that other factors may also influence PFS by AUC0–24h of 5-FU. We did not demonstrate the number of CYP2A6 gene variants correlated with efficacy. Although some studies demonstrated patients having fewer CYP2A6 variants had better PFS in AGC patients with S-1 plus cisplatin, or S-1 plus docetaxel [6, 46], divergences on the relation between CYP2A6 genetic polymorphisms and response existed for both gastric and colorectal cancer patients with S-1 or S-1-based chemotherapy [47, 48]. In our study, second-line treatment excluded S-1 and tegafur. CYP2A6 converts enzymatically tegafur, the effector molecule of S-1, to 5-FU, and this role produces no meaning on second-line treatment.
|
study
| 100.0 |
That high OPRT predicted grade 3–4 AEs as well as affected response can be theoretically understood, and in animal models, oteracil in S-1 was found to inhibit the OPRT by 70% in the small intestine; however, the inhibition was limited to 0–20% in tumor regions without affecting the antitumor activity of 5-FU. Besides, high OPRT/DPD, OPRT/TP, OPRT/TS, OPRT/DPD + TP + TS, OPRT/DPD + TS, OPRT/TP + TS, and OPRT/DPD + TP were all associated with grade 3–4 AEs, among which, high OPRT/DPD exhibited the highest accuracy. Cui et al. reported lower baseline plasmic DPD correlated with higher grade of toxicities in AGC patients with S-1 plus docetaxel by ELISA. Further studies were warranted to decide whether OPRT or OPRT/DPD better predicts grade 3–4 AEs. The literature shows severe diarrhea was dose-limiting toxicity for S-1 in Caucasians and severe neutropenia in East Asians perhaps due to CYP2A6 gene polymorphism , while this study did not see its relation to severe diarrhea, neutropenia, or total grade 3–4 AEs and no relation of CYP2A6 gene polymorphism to AUC0–24h or Cmax of 5-FU here helped explain.
|
study
| 99.94 |
The 2-week S-1/LV regimen demonstrated promising efficacy and satisfactory safety as first-line chemotherapy for AGC. To balance both the efficacy and toxicity, S-1 pharmacogenetic pathway may help find an optimal subset of patients with high OPRT/TS, high AUC0–24h of 5-FU, low DPD that may benefit more from S-1/LV, which awaits validation in another large and well-defined population.
|
other
| 99.44 |
Below is the link to the electronic supplementary material. Fig. S1Mean plasma concentration–time curve of 5-FU for the entire population. 5-FU: fluorouracil, AUC0-24h: areas under the curve, Cmax: maximum concentration, Tmax: time taken to reach maximum concentration, T1/2: half-time, AUMC0-24h: area under the first moment curve, MRT0-24h: mean resistance time, CL: plasma clearance (TIFF 163 kb) Supplementary material 2 (DOCX 19 kb)
|
other
| 99.3 |
Mean plasma concentration–time curve of 5-FU for the entire population. 5-FU: fluorouracil, AUC0-24h: areas under the curve, Cmax: maximum concentration, Tmax: time taken to reach maximum concentration, T1/2: half-time, AUMC0-24h: area under the first moment curve, MRT0-24h: mean resistance time, CL: plasma clearance (TIFF 163 kb)
|
other
| 99.9 |
Hepatocellular carcinoma (HCC) is the third most common cause of cancer deaths and the fourth most common cancer worldwide 1. Chronic liver damage, such as that caused by chronic hepatitis, liver cirrhosis, and fatty liver disease, is closely associated with the occurrence of HCC 2. Recent advances in sequencing technologies have enabled the identification of multiple driver genetic alterations and pathways implicated in hepatocarcinogenesis and tumor progression 3. This may help us to develop new targets and biomarkers that ultimately improve clinical decision‐making and patient outcomes. Of note, among the recurrent oncogenic mutations identified in HCC, the most prevalent is telomerase reverse transcriptase (TERT) promoter (TERTp) mutations, which occurs in up to 60% of patients, highlighting the importance of telomere biology in HCC molecular pathogenesis 4.
|
review
| 99.75 |
Telomeres are specialized structures located at the ends of chromosomes, playing a critical role in maintaining chromosomal integrity and stability. In normal cells, continuous telomere shortening with each cell division triggers DNA damage responses and initiates irreversible cellular senescence or apoptosis 5. Alternatively, chromosome instability and DNA damage‐induced genetic mutations due to shortened telomeres may result in neoplastic transformation 6, 7. It has been proposed that permanent proliferation of cancer cells depends on the maintenance of telomere length 8. To counteract telomere shortening, up to 90% of human cancers, including HCC, reactivate telomerase 9. Data have accumulated that tumor telomeres are shorter than normal tissues in the majority of human cancers 10, 11, 12, while longer telomeres in sarcomas and gliomas were also observed 13. Likewise, there are conflicting results regarding the impact of telomere length on cancer susceptibility and survival. For example, short telomeres predicted a poor prognosis in chronic lymphocytic leukemia and colorectal cancer, but a reduced death risk in esophageal cancer 14. Intrinsic biological features in each cancer demand that the clinical significance of telomere length needs cancer‐specific investigation. Although HCC has the second highest frequency of TERT promoter mutations among 31 cancer types 13, the clinical relevance of telomere attrition or elongation in HCC remains unknown.
|
review
| 99.6 |
The unique signature of the liver microenvironment, characterized by a chronic inflammatory state and dysregulated immune response, was associated with the biological behavior of HCC 15. Within the HCC microenvironment, cancer‐associated fibroblasts (CAFs) and tumor‐infiltrating lymphocytes (TILs) are of paramount importance. Experimental and clinical evidence demonstrated that interactions between CAFs or TILs and tumor cells could promote HCC progression and metastasis through various mechanisms 16, 17. A recent study in prostate cancer suggested that investigation of telomere length in cancer‐associated stromal cells is feasible and is significant for predicting cancer behavior 18. Large‐scale prospective studies suggested that telomere attrition in peripheral blood leukocytes correlated not only to poor prognosis of a group of human cancers 19, 20 but also with increased mortality in the general population 21. Thus, it is rational to speculate that telomere attrition or elongation in CAFs or TILs would harbor significant clinical value in HCC.
|
review
| 98.8 |
Based on the above information, we evaluated telomere lengths in tumor cells, CAFs, and TILs in a large cohort of HCCs using telomere‐specific fluorescence in situ hybridization (FISH) and qPCR. The recently developed FISH‐based method enables accurate measurement of telomere length at single‐cell resolution, greatly facilitating such analysis 18. We found that shortened telomeres in tumor cells and CAFs, rather than in TILs, were independently and significantly associated with the clinical outcome of HCC patients.
|
study
| 100.0 |
A cohort of 257 HCC patients who received curative hepatectomy at Zhongshan Hospital (Fudan University, China) from January to December 2007 was enrolled. This study was conducted after obtaining informed consent forms from patients and ethical approval from Zhongshan Hospital Research Ethics Committee. The inclusion and exclusion criteria for patients, therapeutic modalities, and postoperative surveillance according to a uniform guideline have been described previously 22. The clinicopathologic features are provided in the supplementary material, Table S1, and patient follow‐up is included in the Supplementary materials and methods. Tissue microarrays were constructed as described previously 23. Details are included in the supplementary material, Supplementary materials and methods.
|
study
| 100.0 |
Isolation of CD45+ leukocytes and α‐SMA+ fibroblasts from peritumoral liver and HCC tissues using MicroBeads and an MS Column (Miltenyi Biotec, Bergisch Gladbach, Germany) was performed according to the manufacturer's instructions. Phenotypic characterization of the isolated cells was conducted using flow cytometry as previously described 24. Details are included in the supplementary material, Supplementary materials and methods.
|
study
| 100.0 |
Genomic DNA was extracted from paired peritumoral liver and HCC tissues from 24 patients, as well as isolated leukocytes and fibroblasts from ten patients. mRNA was extracted from paired peritumoral liver and HCC tissues from another 64 patients. Relative telomere length and mRNA expression were measured by real‐time qPCR as previously described 25.
|
study
| 100.0 |
The assessment of telomere length was conducted by telomere FISH for telomeric DNA as previously described 18, 26. Sections were imaged by an Olympus BX51 fluorescence microscope equipped with a UIS2 optical system and Kohler illuminator (OLYMPUS, Tokyo, Japan) using a 40X/1.42 NA UPLFLN lens with correction collar. Quantification of the digitized fluorescent telomere captures was performed applying the open source JAVA‐based image analysis software package ImageJ as previously described 15. Details are reported in the supplementary material, Supplementary materials and methods.
|
study
| 99.94 |
Data are expressed as the mean ± SEM, and error bars in the figures refer to SEM, median, and interquartile range (IQR). The analysis of association between variables was conducted using the Mann–Whitney U‐test, Student t‐test, chi‐square test, or one‐way ANOVA test when appropriate. Univariate analysis was based on the Kaplan–Meier method using the log‐rank test. Cox proportional hazards regression was used for multivariate analyses. SPSS software (IBM, Armonk, NY, USA) and Graph Pad Prism 6.0 (Graph Pad Software Inc, San Diego, CA, USA) were applied for data analyses. Statistical tests were two‐sided, with p < 0.05 considered significant.
|
study
| 99.94 |
To test the specificity and validity of the Cy3‐labeled PNA probe, and to develop a series of positive controls for telomere length assessment in paraffin‐embedded tissue sections, we first evaluated telomere length in seven cell lines. Each cell line was fixed in formalin and embedded in paraffin to imitate standard slides of patients' tumor tissue. Representative images of the hybridization reaction localizing the fluorescent telomeric PNA probe in each cell line are shown in Figure 1A–C. Procedures for measuring telomere length by ImageJ are shown in the supplementary material, Figure S1A, B, as previously described 18. The highly metastatic cell lines (HCC‐LM3, MHCC‐97H, and MHCC‐97 L) (Figure 1A) showed minimal or no intensities of telomere FISH, while cells with low metastatic capability (Huh7 and SMMC‐7721) (Figure 1B) and the normal liver cell line L‐02 (Figure 1C) showed strong intensities, indicating the involvement of telomere length in HCC aggressiveness. In addition, Hela cells showed a high telomere signal (Figure 1C). Quantitative analysis of telomere length in cell lines is summarized in Figure 1D.
|
study
| 100.0 |
Telomere‐specific FISH in HCC cell lines. Representative FISH images of (A) MHCC‐97 L, MHCC‐97H, and HCC‐LM3; (B) Huh7 and SMMC‐7721; and (C) L‐02 and Hela cells. Left panels: 4',6‐diamidino‐2‐phenylindole (DAPI) (blue) was used to identify nuclei. Middle panels: Cy3‐PNA telomere‐probe fluorescence. Right panels: merged images (original magnification ×40). (D) Statistics of FISH telomere signal intensity.
|
study
| 99.9 |
Quantification of specific telomere‐FISH signals in cells and tissues is linearly proportional to telomere length, and differences of telomere length among cells and tissues can be evaluated via quantitative image analysis 18. First, telomere‐specific FISH was applied on whole slides from 30 HCC samples to elucidate the sub‐location and microanatomic distribution of telomere signals in peritumor, tumor margin, and intratumor areas (Figure 2A, B). Telomere signals were higher in peritumor areas (median 25.5; IQR 24.3–26.9) than in intratumor areas (median 11.8; IQR 10.7–13.0) and tumor margins (median 22.6; IQR 20.0–24.4) (Figure 2C).
|
study
| 100.0 |
Telomere‐specific FISH in HCC tissues. (A) Left: representative image of telomere‐specific FISH on whole slides to elucidate the sub‐location and micro‐anatomic distribution of telomere signals (original magnification ×40). Right: the same section is divided into 48 grids, colored according to the relative telomere density. (B) Top: representative H&E staining of HCC with (left) or without (right) a tumor capsule (original magnification ×20). Bottom: FISH staining in the case with a tumor capsule. White dotted lines highlight the margin area (original magnification ×40). (C) Telomere‐FISH intensity quantification (n = 30). Lines indicate the 25th, 50th, and 75th percentiles. ***p < 0.001. (D) Representative examples of telomere length variation in tumor cells and CAFs in TMA. (a) Strong telomere signals in cancer cells; (b) weak telomere signals in cancer cells; (c) extremely short telomeres in CAFs; (d) long telomeres in CAFs. Asterisks indicate cancer cells and arrows indicate CAFs (original magnification ×40).
|
study
| 100.0 |
FISH was then performed on TMAs containing 257 HCC samples to elucidate telomere lengths in paired peritumor and intratumor areas. We calculated eight cell types separately along with H&E staining for each patient, i.e. tumor cells, peritumor liver cells, CAFs, non‐tumoral fibroblasts (NTFs), TILs, peritumor‐infiltrating lymphocytes (PTILs), peritumor bile duct epithelial cells (P‐BDECs), and tumor bile duct epithelial cells (T‐BDECs) (supplementary material, Figure S1C, D). Figure 2D shows representative images of telomere signals in tumor cells and CAFs at single‐cell resolution. Consistent with telomere shortening in other solid tumors 10, 11, 25, telomere signals were less intense (i.e. shorter telomere length) in tumor cells than in adjacent liver cells (p < 0.001; Figure 3A). In total, 164 out of 257 samples (63.81%) displayed significantly less intense telomere signals among tumor cells compared with corresponding normal liver cells (supplementary material, Figure S2A).
|
study
| 100.0 |
Quantitative results of telomere‐specific FISH in 257 HCCs. (A) Telomere signals in peritumor liver cells and HCC tumor cells (n = 257). (B) Telomere signals in NTFs and CAFs (n = 257). (C) Telomere signals in PTILs and TILs (n = 257). (A–C) Lines indicate the 25th, 50th, and 75th percentiles, with their respective values provided. ***p < 0.001; NS, p = 0.587. (D) The telomere signal of tumor cells correlated with that of CAFs (n = 257; r = 0.2, p < 0.01).
|
study
| 100.0 |
Likewise, 157 of 257 (61.08%) exhibited telomere signal diminution in CAFs compared with paired NTFs (p < 0.001; Figure 3B and supplementary material, Figure S2B). Diffuse or aggregated infiltration of lymphocytes was found in peritumor areas and tumor margins, with less abundant lymphocytes in intratumor areas (supplementary material, Figure S2C). Nevertheless, no obvious differences were found in telomere lengths in TILs and PTILs (p = 0.587; Figure 3C). Intrahepatic bile ducts were detected in intratumor areas of only 92 cases, but in the peritumor area of 168 cases (supplementary material, Figure S2D). Differences in telomere length were hard to find in BDECs between paired intratumor and peritumor areas (data not shown). The telomere length in tumor cells, CAFs, and TILs had a wide dynamic IQR.
|
study
| 100.0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.