Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A ) Characterisation of ΔHS + 44/HS + 48 primary erythroid cells. Top tracks show profiles for ATAC-seq (black) and CTCF ChIP-seq (navy) in primary erythroid cells (Ter119 + ) isolated from WT (Hanssen et al, ) and ΔHS + 44/HS + 48 mice. Profiles show normalised (RPKM) and averaged data from n = 3 biological replicates across the α-globin locus (mm9, chr11:32,000,000–32,330,000) with genes and genomic position, with positioning of genes above or below representing sense and antisense transcription, respectively. The adult α-globin genes are highlighted in red. The individual α-globin superenhancer elements (R1, R2, R3, Rm, and R4) are highlighted in grey. The horizontal grey bars between the tracks represent the ~70 kb α-globin sub-TAD (light grey, chr11:32,136,000–32,202,000) nested within a larger ~165 kb TAD (dark grey, chr11:32,080,000– 32,245,000) (Oudelaar et al, ). The orientation of CTCF motifs is shown under peaks by red (forward) and blue (reverse) arrows. NG Capture-C interaction profiles of the α-globin locus from the viewpoint of the R2 enhancer element with an exclusion zone around the viewpoint, in WT (grey) and ΔHS + 44/HS + 48 (purple) Ter119+ primary erythroid cells. The profiles represent normalised and averaged unique interactions from n = 3 biological replicates with halos representing ± standard deviation, smoothed with a 1D Gaussian filter. ( B ) Reverse transcription qPCR expression analysis. Fold changes in α- (Hba) and β-globin (Hbb) mRNA ratio, Sh3pxd2b mRNA, and Ubtd2 mRNA in ΔHS + 44/HS + 48 (purple) compared to WT (white) Ter119+ erythroid cells, normalised to 18S RNA. Mean of n = 3 biological replicates shown. Error bars display ± standard deviation. ( C ) Characterisation of Δθ1/θ2 primary erythroid cells. As in ( A ) but showing profiles from primary erythroid cells (Ter119 + ) isolated from WT and Δθ1/θ2 mice. NG Capture-C interaction profiles of the α-globin locus from the viewpoint of (i) the R2 enhancer element (ii) Hba-a1/2 genes. ( D ) Differential gene expression WT vs Δθ1/θ2 PolyA + RNA-seq from primary erythroid cells. MAplot of Log2 fold change in gene expression relative to WT against the Log10 read counts; each dot represents a gene. Highlighted in red are Sh3pxd2b, Ubtd2, Hba-a (representing total Hba-a1/2) and the Δθ1/θ2 associated pseudogenes Hbq1a/b . Significant changes plotted with dark red or blue (Wald test P value, Benjamini–Hochberg corrected: P adj <0.01), non-significant changes are in grey and bright red. The θ1/θ2 associated genes are downregulated upon CTCF site deletion, and total Hba-a1/2 is not significantly changed. ( E ) Relative expression of Hba-a1/2 mRNA. From WT, ∆θ1, ∆θ2 and ∆θ1θ2 primary erythroid cells (Ter119 + ). Variant calling analysis performed on Poly(A) + RNA-seq data from n = 3 biological replicates revealed the percentage of reads originating from transcripts of Hba-a1 (light red) or Hba-a2 (dark red). Error bars display ± standard deviation, and each point represents a biological replicate. No significant effect of genotype on the proportion of each α-globin variant was detected by a one-way ANOVA ( P > 0.05) with a Tukey post hoc test. .

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: ChIP-sequencing, Isolation, Capture-C, Standard Deviation, Reverse Transcription, Expressing, Gene Expression, RNA Sequencing, Variant Assay

Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A ) Top tracks show profiles for ATAC-seq (black) and CTCF ChIP-seq (navy) in primary erythroid cells (Ter119 + ) isolated from WT and ΔHS44/48 mice. Profiles show normalised (RPKM) and averaged data from n = 3 biological replicates across the α-globin locus (mm9, chr11:32,000,000–32,330,000) with genes and genomic position, with positioning of genes above or below representing sense and antisense transcription, respectively. The adult α globin genes are highlighted in red. The individual α globin superenhancer elements (R1, R2, R3, Rm, and R4) are highlighted in grey. The horizontal grey bars between the tracks represent the ~70 kb α globin sub-TAD (light grey, chr11:32,136,000–32,202,000) nested within a larger ~165 kb TAD (dark grey, chr11:32,080,000–32,245,000). The orientation of CTCF motifs is shown under peaks by red (forward) and blue (reverse) arrows. NG Capture-C interaction profiles of the α-globin locus from WT (grey) and ΔHS44/HS48 (purple) Ter119+ primary erythroid cells, the following viewpoints: (i) HS-38 CTCF, (ii) HS-39 CTCF, (iii) R1 enhancer element and (iv) Hba-a1/2 genes. The profiles represent normalised and averaged unique interactions from n = 3 biological replicates, smoothed with a 1D Gaussian filter. ( B ) As in ( A ) but showing profiles from primary APH-treated spleen cells isolated from WT and Δθ1/θ2/HS + 44/HS + 48 mice. NG Capture-C interaction profiles of the α-globin locus from WT (grey) and Δθ1/θ2/HS + 44/HS + 48 (purple) from the following viewpoints: (i) HS-38 CTCF, (ii) HS-39 CTCF, (iii) Hba-a1 SNP-specific interactions and (iv) Hba-a2 SNP-specific. The profiles represent normalised and averaged unique interactions from n = 3 biological replicates and halos representing ± standard deviation, smoothed with a 1D Gaussian filter. Differential tracks (ΔCaptureC) show subtractions (v) WT [Hba-a1 – Hba-a2] and (vi) Δθ1/θ2/HS + 44/HS + 48 [Hba-a1 – Hba-a2]) of the mean number of unique interactions per restriction fragment, scaled to a total of 100,000 interactions in cis from SNP-specific counts.

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: ChIP-sequencing, Isolation, Capture-C, Standard Deviation

Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A ) Characterisation of Δθ1 primary erythroid cells. Top tracks show profiles for ATAC-seq (black) and CTCF ChIP-seq (navy) in primary erythroid cells (Ter119 + ) isolated from WT and Δθ1 mice. Profiles show normalised (RPKM) and averaged data from n = 3 biological replicates across the α-globin locus (annotations are as in Fig. ). ( B ) Characterisation of Δθ2 primary erythroid cells. Top tracks show profiles for ATAC-seq (black) and CTCF ChIP-seq (navy) in primary erythroid cells (Ter119 + ) isolated from WT and Δθ2 mice. Profiles show normalised (RPKM) and averaged data from n = 3 biological replicates across the α-globin locus (annotations are as in Fig. ). NG Capture-C interaction profiles of the α-globin locus from the combined viewpoint of the Hba genes each with an exclusion zone, in WT (grey) and Δθ1/θ2 (pink) Ter119+ primary erythroid cells. The interaction profiles represent normalised and averaged unique interactions from n = 2 biological replicates and halos representing ± standard deviation, smoothed with a 1D Gaussian filter. ( C – E ) Differential expression (PolyA + RNA-seq) in Δθ1, Δθ2 and Δθ1/θ2 primary erythroid cells. MAplot of Log2 Fold change versus Log10 of normalised counts in the models above vs WT; each dot represents a gene. Genes with an adjusted P value (padj, Benjamini–Hochberg corrected) <0.01 are highlighted in blue. Genes of interest are highlighted in red (3’genes Sh3pxd2b , Ubtd2 , Total Hba ( Hba-a1/2 ) and θ1/θ2 associated genes Hbq1b/a respectively) and those with a significant difference from WT highlighted with dark red. Results from n = 3 biological replicates from each genotype. There is an unexpectedly high number of differential genes in the Δθ2 model further Hbq1a appears unchanged upon Δθ2, which is incongruent with the result in Δθ1/θ2. As the Δθ1/θ2 does not show these differences and is a combinatorial deletion of both θ1/θ2 and we must assume these differences in Δθ2 expression are due to technical error.

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: ChIP-sequencing, Isolation, Capture-C, Standard Deviation, Quantitative Proteomics, RNA Sequencing, Expressing

Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A ) Schematic of the design of the insulator assay. The α-globin locus was edited in Mouse Embryonic Stem Cells (mESCs). The key features of the design are: (1) A single insertion site is located between the globin genes and the enhancer elements; (2) Hba-a1 is tagged with mVenus (red/yellow) to allow readout of α-globin expression by FACS; (3) the cell line is hemizygous for the α-globin locus to allow for increased targeting efficiency and single allele genomics; finally, (4) engineered mESCs are differentiated and erythroid cells are isolated (CD71+ve) via a 7-day in vitro differentiation protocol allowing for gene expression read out (Francis et al, ). ( B ) Insulation strength of α-globin gene fragments in a boundary reporter assay. Left: Schematics of the models: ΔR1R2 = mESC with deleted R1 and R2 enhancers and with αYFP; HS-38 = positive control/mESC with HS-38 CTCF site inserted; WT αYFP = parental αYFP reporter with no insert ( n = 3); αP + G = inserted α-globin promoter and gene body ( n = 3); αP Only = inserted α-globin promoter only ( n = 3); αG Only = inserted α-globin gene body only without the promoter (no expected transcription) ( n = 1). Right: Median YFP fluorescence in the CD71+ fraction of cells measured by FACS and normalised to the parental αYFP reporter. Data acquired from clones (numbers indicated above) across three independent differentiations. Error bars display standard deviation. Statistical analysis was performed using a one-way ANOVA with a Tukey post hoc test, P values are presented above comparisons. ( C ) FACS gating strategy and density plots of YFP fluorescence in the CD71+ fraction of cells. (i) Single cells were gated by FSC-H/A, (ii) then gated for CD71 + /YFP + . (iii) Density plots for each clone from one differentiation shown to the left: ΔR1R2 (dark grey), WT αYFP (grey), αP + G (dark blue), αP Only (blue) and αG Only (teal). ( D ) RPKM normalised PolyA- RNA-seq derived from CD71+ erythroid cells from each of the indicated models, aligned to the Hba-a1 -only custom genome. Bigwigs are visualised on the sense strand as mean of data from n = 3 independent differentiations of a representative clone per genotype. Note that the reference does not have inserted sequences and should be used as an indication for the activity around the inserted sites. ( E ) Quantification of PolyA- transcription from either the left or the right-hand side of the insert, using the PolyA- RNA-seq signal (as in D ) in each of the models. Normalised counts were generated using Featurecounts and DEseq2. *Pseudo count of 1 was assigned to counts of 0 to allow for visualising on a logarithmic scale. For changes in expression in the region right of the insert, WT αYFP-αP Only, P adj = 0.00016 and for WT αYFP-αP + G , P adj= 3.69e-48. ( F ) RPKM normalised ATAC-seq performed on CD71+ erythroid cells derived from each of the indicated engineered models and aligned to a custom mm9 genome with Hba-a1 :: mVenus . Note that the reference does not have inserted sequences in the reference and should be used as an indication for the activity around the insert sites—further the signals over Hba-a1/2 are an amalgamation of reads from the native and inserted variants. The insertion site is highlighted in grey and the Hba-a1 :: mVenus reference in red. Data are from each of the clones shown in ( B , C ) and (where possible) are averages of biological replicates. .

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: Expressing, Isolation, In Vitro, Gene Expression, Insulation, Reporter Assay, Positive Control, Fluorescence, Clone Assay, Standard Deviation, RNA Sequencing, Derivative Assay, Activity Assay, Generated

Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A – C ) Differential expression between the WT aYFP reporter and the insertion models. MAplot of Log2 fold change in gene expression relative to WT against the Log10 of read counts; each dot represents a gene. Highlighted in red are Hba-a (representing total Hba-a1/2 and aP+G in the P + G model), the region downstream of the insertion site (ins_right) and other genes in the locus. Significant changes plotted with dark red or blue (Wald test P value, Benjamini–Hochberg corrected: P adj<0.01), non-significant changes are in grey and bright red. Data is n = 5 replicates of WT-aYFP and n = 3 replicates of each other genotype. As these libraries were not globin-depleted, the representation is skewed toward globin genes. ( D ) Hba-x expression as normalised counts in PolyA-Plus RNA-seq. Data is representative of n = 5 replicates of WT-aYFP and n = 3 replicates of each other genotype. Error bars represent ± standard deviation. ( E ) mVenus expression as normalised counts in PolyA-Plus RNA-seq. Data is representative of n = 5 replicates of WT-aYFP and n = 3 replicates of each other genotype. Error bars represent ± standard deviation. ( F ) Exonic SNP-specific count PolyA-Plus RNA-seq. The aP+G gene had its own unique SNP in exon 3 allowing counting of the proportion of transcripts between the inserted or native α-globin copies. SNPs specific counts of each variant of Hba were counted similarly to Fig. in RNA-seq from the CD71+ cells. Error bars represent ± standard deviation. ( G ) Heatmap and summary profile displaying Rad21 ChIP-seq signal in WT in vitro-derived CD71+ erythroid cells at non-redundant transcription start sites (TSS) of inactive (14422) and active (8332) genes, which do not have a CTCF binding site within a 2 kb window around the TSS.

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: Quantitative Proteomics, Gene Expression, Expressing, RNA Sequencing, Standard Deviation, Variant Assay, ChIP-sequencing, In Vitro, Derivative Assay, Binding Assay

Journal: The EMBO Journal

Article Title: A functional overlap between actively transcribed genes and chromatin insulator elements

doi: 10.1038/s44318-026-00730-2

Figure Lengend Snippet: ( A ) Cohesin accumulation across the α-globin locus in CD71+ erythroid cells derived from edited reporter cells. Rad21 ChIP-seq performed on CD71+ erythroid cells derived from each of the indicated engineered models and aligned to a custom genome with Hba-a1::mVenus in its native position, based on the mm9 genome reference. Note that the reference does not have insert sequences in the reference and should be used as an indication of the activity around the insert sites. The insertion site is highlighted in grey and the Hba-a1 reference in red. Top panel shows the locus (custom mm9: 32,020,000–32,250,000) whilst the bottom panel zooms in on the insertion site at the locus (custom mm9: 32,152,000–32,182,000). Bigwigs are visualised from n = 1 (WT-hemizygous, WT αYFP, αP Only ) and as an average of n = 3 (αP + G) independent differentiations of a representative clone per genotype. Where appropriate, error bars display ± standard deviation. ( B ) Rad21 ChIP-seq read counts in each replicate. Read counts performed over regions covering the insert site (chr11:32,171,181–32,174,582), R1 enhancer element (chr11:32,144,849–32,146,726) and HS-38 CTCF site (chr11:32,136,773–32,137,613) from the data in ( A ). Reads were normalised to the number of reads over a selected region in the β-globin locus and normalised to the WT αYFP to exemplify differences between models. ( C ) Capture-C from R2 enhancer in CD71+ erythroid cells derived from edited reporter cells. ATAC-seq in the top panel to show the positions of the elements of interest. CaptureC profiles represent the mean number of normalised unique interactions per restriction fragment from n = 3 (WT αYFP), n = 4 (αP Only and αP + G) independent differentiations, these are represented in dark grey, light blue and dark blue, respectively. Halos represent ± standard deviation. Differential tracks (ΔCaptureC) show subtractions (αPOnly - WT αYFP and αP + G- WT αYFP) of the mean number of unique interactions per restriction fragment, scaled to a total of 100,000 interactions in cis. Note differences represented are not determined to be significant (threshold P adj>0.05). The region around the insert (mm9: Insert region: chr11:32,169,922–32,173,392) is highlighted in grey, and region covering native targets ( Hba-x - upstream of Hba-a1 chr11:32,175,105–32,182,970) is highlighted in red. To note, read coverage over the Hba-a1/2 genes is a combination of both inserted and native sources. ( D ) Schematic to show a dynamic, directional tracking mechanism of chromatin loop extrusion by cohesin from the α-globin enhancers to the promoters. Multi-protein complexes recruited to the actively transcribing genes stall cohesin translocation on chromatin, resulting in cohesin retention at active genes, in addition to CTCF binding sites. .

Article Snippet: The ∆∆Ct method was used for relative quantification of RNA abundance using TaqMan Universal PCR Master Mix (Applied Biosystems, ThermoFisher: 4304437) and the following TaqMan probes: Mm00845395_s1 ( Hba-a1/2 ), Mm01611268_g1 ( Hbb-b1 ), Mm00616672_m1 ( Sh3pxd2b ), Mm00612868_m1 ( Ubtd2 ), and Mm04277571_s1 ( Rn18s ).

Techniques: Derivative Assay, ChIP-sequencing, Activity Assay, Standard Deviation, Capture-C, Translocation Assay, Binding Assay

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Screening of bottleneck reaction steps in the 1w–SmHBA operon by multienzyme catalysis. a, Schematic illustration of the reaction components. The in vitro reaction system consisted of crude cell extracts (CCE) from strains harboring the 1w–SmHBA plasmid as the basic catalytic source, supplemented with multienzyme-expressing strains carrying plasmids encoding 4–5, 2–3, or single heterologously expressed Cob enzymes to alter the enzyme composition in each reaction. b, Comparison of in vitro HBA production using CCE from strains H1, H2, and LvH0. Two-sided unpaired t -test is carried out between H1, H2, and LvH0. Unpaired t -test of data: H1 to H2, ∗∗∗∗, P < 0.0001 (t = 35.12); LvH0 to H2, ∗∗∗∗, P < 0.0001 (t = 30.84). c, In vitro HBA production using CCE from H21–H28 strains, each harboring an additional RcCob enzyme. Red bars indicate values higher than those of the H2 reactant, whereas blue bars indicate values lower than those of H2 reactant. Two-sided unpaired t -test is carried out between H1 to H21-28. Unpaired t -test of data:H2 to H21, ∗, P = 0.0227 (t = 3.605); H24 to H2, ∗∗∗∗, P < 0.0001 (t = 24.60); H27 to H2, ∗, P = 0.5363 (t = 0.6757). d, In vitro HBA production using CCE from H37–H44 strains, in which SmCob enzymes were replaced with the corresponding RcCob enzymes. Two-sided unpaired t -test is carried out between H1 to H37-44. Unpaired t -test of data:H37 to H2, ∗, P = 0.0145 (t = 4.129); H41 to H2, ∗, P = 0.0315 (t = 3.246); H42 to H2, ∗∗∗, P = 0.0003 (t = 11.66); H43 to H2, ∗, P = 0.0229 (t = 3.592); H44 to H2, ∗∗∗∗, P < 0.0001 (t = 15.76). e, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 4–5 Cob enzymes. HBA-A: CCE with pET28a–CobAIGJM; HBA-B: CCE with pACYCDuet-1–CobFKLH. Unpaired t -test of data: HBA-A 45 OD 600 to Control in HBA titer, ns, P = 0.0729 (t = 2.418); HBA-B 45 OD 600 to Control in HBA titer, ∗∗, P = 0.0029 (t = 6.502); HBA-A 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0022 (t = 7.002); HBA-B 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0011 (t = 8.309); f, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 2–3 Cob enzymes. AIG: CCE with pet28a-CobAIG; JM: CCE with pet28a-CobJM; FK: CCE with pet28a-CobFK; LH: CCE with pet28a-CobLH. Unpaired t -test of data: AIG 30 OD600 to Control in Urogen III titer, ∗∗∗, P = 0.0009 (t = 8.801); JM 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 24.39); FK 30 OD600 to Control in Urogen III titer, ∗, P = 0.0304 (t = 3.285); LH 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 15.93); g, Screening of reactions supplemented with CCE from strains heterologously expressing a single Cob enzyme. Unpaired t -test of data: CobA + to ori, ∗∗∗, P = 0.0002 (t = 13.95); CobI + to ori, ∗∗∗, P = 0.0005 (t = 10.57); CobG + to ori, ∗∗∗, P = 0.0003 (t = 12.00); CobJ + to ori, ∗∗∗, P = 0.0001 (t = 15.48); CobM + to ori, ∗∗∗, P = 0.0005 (t = 10.46); CobF + to ori, ∗∗∗∗, P < 0.0001 (t = 16.42); CobK + to ori, ∗∗∗, P = 0.0006 (t = 9.873); CobL + to ori, ∗∗, P = 0.0042 (t = 5.882); CobH + to ori, ∗∗∗, P = 0.001 (t = 8.675).

Article Snippet: Unpaired t -test of data: HBA-A 45 OD 600 to Control in HBA titer, ns, P = 0.0729 (t = 2.418); HBA-B 45 OD 600 to Control in HBA titer, ∗∗, P = 0.0029 (t = 6.502); HBA-A 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0022 (t = 7.002); HBA-B 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0011 (t = 8.309); f, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 2–3 Cob enzymes.

Techniques: In Vitro, Plasmid Preparation, Expressing, Comparison, Control

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Screening of bottleneck reaction steps in the 1w–SmHBA operon by multienzyme catalysis. a, Schematic illustration of the reaction components. The in vitro reaction system consisted of crude cell extracts (CCE) from strains harboring the 1w–SmHBA plasmid as the basic catalytic source, supplemented with multienzyme-expressing strains carrying plasmids encoding 4–5, 2–3, or single heterologously expressed Cob enzymes to alter the enzyme composition in each reaction. b, Comparison of in vitro HBA production using CCE from strains H1, H2, and LvH0. Two-sided unpaired t -test is carried out between H1, H2, and LvH0. Unpaired t -test of data: H1 to H2, ∗∗∗∗, P < 0.0001 (t = 35.12); LvH0 to H2, ∗∗∗∗, P < 0.0001 (t = 30.84). c, In vitro HBA production using CCE from H21–H28 strains, each harboring an additional RcCob enzyme. Red bars indicate values higher than those of the H2 reactant, whereas blue bars indicate values lower than those of H2 reactant. Two-sided unpaired t -test is carried out between H1 to H21-28. Unpaired t -test of data:H2 to H21, ∗, P = 0.0227 (t = 3.605); H24 to H2, ∗∗∗∗, P < 0.0001 (t = 24.60); H27 to H2, ∗, P = 0.5363 (t = 0.6757). d, In vitro HBA production using CCE from H37–H44 strains, in which SmCob enzymes were replaced with the corresponding RcCob enzymes. Two-sided unpaired t -test is carried out between H1 to H37-44. Unpaired t -test of data:H37 to H2, ∗, P = 0.0145 (t = 4.129); H41 to H2, ∗, P = 0.0315 (t = 3.246); H42 to H2, ∗∗∗, P = 0.0003 (t = 11.66); H43 to H2, ∗, P = 0.0229 (t = 3.592); H44 to H2, ∗∗∗∗, P < 0.0001 (t = 15.76). e, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 4–5 Cob enzymes. HBA-A: CCE with pET28a–CobAIGJM; HBA-B: CCE with pACYCDuet-1–CobFKLH. Unpaired t -test of data: HBA-A 45 OD 600 to Control in HBA titer, ns, P = 0.0729 (t = 2.418); HBA-B 45 OD 600 to Control in HBA titer, ∗∗, P = 0.0029 (t = 6.502); HBA-A 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0022 (t = 7.002); HBA-B 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0011 (t = 8.309); f, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 2–3 Cob enzymes. AIG: CCE with pet28a-CobAIG; JM: CCE with pet28a-CobJM; FK: CCE with pet28a-CobFK; LH: CCE with pet28a-CobLH. Unpaired t -test of data: AIG 30 OD600 to Control in Urogen III titer, ∗∗∗, P = 0.0009 (t = 8.801); JM 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 24.39); FK 30 OD600 to Control in Urogen III titer, ∗, P = 0.0304 (t = 3.285); LH 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 15.93); g, Screening of reactions supplemented with CCE from strains heterologously expressing a single Cob enzyme. Unpaired t -test of data: CobA + to ori, ∗∗∗, P = 0.0002 (t = 13.95); CobI + to ori, ∗∗∗, P = 0.0005 (t = 10.57); CobG + to ori, ∗∗∗, P = 0.0003 (t = 12.00); CobJ + to ori, ∗∗∗, P = 0.0001 (t = 15.48); CobM + to ori, ∗∗∗, P = 0.0005 (t = 10.46); CobF + to ori, ∗∗∗∗, P < 0.0001 (t = 16.42); CobK + to ori, ∗∗∗, P = 0.0006 (t = 9.873); CobL + to ori, ∗∗, P = 0.0042 (t = 5.882); CobH + to ori, ∗∗∗, P = 0.001 (t = 8.675).

Article Snippet: Supplementation of HBA-A or HBA-B CCE into a basic H2-derived HBA reaction system at gradient concentrations revealed that addition of HBA-B (downstream enzymes) significantly increased HBA titer by 17.1 ± 3.2 % and reduced Urogen III accumulation by 70.0 ± 10.8 %, while supplementation with HBA-A CCE (upstream enzymes) increased HBA titers by 12.8 ± 8.6 % and reduced Urogen III accumulation by 56.6 ± 10.0 % ( e).

Techniques: In Vitro, Plasmid Preparation, Expressing, Comparison, Control

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Assembly of the artificial HBA synthetic operon. a, Schematic illustration of natural HBA biosynthetic pathway. The natural 5-ALA biosynthetic route originating from TCA cycle is showed in gray; the aerobic HBA biosynthetic pathway investigated in this study is highlighted with pink shading; natural anaerobic HBA biosynthetic pathway is showed in purple; the downstream vitamin B12 biosynthetic pathway originating from HBA is in red. Enzymes are depicted in navy and blue, while cofactors, cosubstrates and by-products are indicated in brown. Abbreviation: 5-ALA, 5-Aminolevulinate; PBG, porphobilinogen; HMB, hydroxymethylbilane; PC, precorrin; HBA, hydrogenobyrinic acid; HBAD, hydrogenobyrinate a, c-diamide; CBAD, cob (II)yrinate a, c-diamide; SAM, S-adenosyl- l -methionine; SAH, S-adenosyl- l -homocysteine; L-Gln, l -Glutamine; L-Glu, l -Glutamate; Pi, phosphate. b, Illustration of natural HBA operons from R. capsulatus and S. meliloti , and schematic representation of two basic artificial HBA operons composed of enzymes from different microorganisms. c, Comparison of HBA production between the two artificial HBA operons in E. coli . H1: E. coli MG1655 (DE3), pET28a-1wRcHBA. H2: E. coli MG1655 (DE3), pET28a-1wSmHBA. All results were performed in triplicate (n = 3 biologically independent samples) and data are presented as mean values ± SD. Two-sided unpaired t -test is carried out between H1 and H2. Unpaired t -test of data: HBA/biomass, ∗∗∗∗, P < 0.0001 (t = 18.14); biomass, ns, P = 0.1226 (t = 1.953). d, SDS–PAGE analysis of heterologous expression of Cob enzymes from S. meliloti in E. coli BL21 (DE3) harboring the pET28a plasmid. M, prestained protein ladder; C, pET28a plasmid control; I, pET28a-1w-SmCobI; G, pET28a-1w-SmCobG; J, pET28a-1w-SmCobJ; M, pET28a-1w-SmCobM; F, pET28a-1w-SmCobF; K, pET28a-1w-SmCobK; L, pET28a-1w-SmCobL; H, pET28a-1w-SmCobH; GST-K, pET28a-GST-CobK; K opti , pET28a-CobK opti . Target protein bands are highlighted with red or navy arrows. e, Quantitative analysis of SDS–PAGE results in (d). Target protein bands were quantified using ImageJ based on pixel intensity. f, Comparison of HBA production by the two artificial HBA operons in E. coli . H45: E. coli MG1655 (DE3), pET28a-1wSmHBA-GST-CobK. All results were performed in triplicate (n = 3 biologically independent samples) and data are presented as mean values ± SD. Two-sided unpaired t -test is carried out between H2 and H45. Unpaired t -test of data: HBA/biomass, ∗∗, P = 0.0016 (t = 7.562); biomass, ∗∗, P = 0.0029 (t = 6.501).

Article Snippet: After 12 h of reaction, CCE derived from strain H2 produced a higher HBA titer (3.16 ± 0.14 μM) than those from H1 and LvH0, consistent with the trends observed in vivo ( b, MS identification of synthetic HBA and Urogen III are showed in ).

Techniques: Comparison, SDS Page, Expressing, Plasmid Preparation, Control

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Screening of the bottleneck reaction step in the 1w–SmHBA operon. a, Schematic illustration of the one-plasmid and two-plasmid assemblies of the 1w–SmHBA operon and the corresponding engineered strains. pET28a plasmid was showed in black, pACYC-Duet-1 plasmid was showed in red. b, Comparison of HBA production in strains H2, LvH0, and LvH15. Two-sided unpaired t -test is carried out between H2, LvH0 and LvH15. Unpaired t -test of data: H2 to LvH0, ∗∗∗∗, P < 0.0001 (t = 15.94); H2 to LvH15, ∗∗∗∗, P < 0.0001 (t = 17.87); LvH0 to LvH15, ∗∗∗, P = 0.0003 (t = 12.02). c, Comparison of HBA production between the LvH0 strain and LvH1–LvH8 strains, in which the RBS strength of individual cob genes was enhanced. Two-sided unpaired t -test is carried out between LvH0 and LvH1 to 15. Unpaired t -test of data: LvH1, ns, P = 0.9088 (t = 0.1220); LvH2, ∗∗, P = 0.0022 (t = 6.960); LvH3; ∗∗∗, P = 0.0002 (t = 14.02); LvH4, ∗∗∗, P = 0.0004 (t = 10.78); LvH5, ∗∗∗∗, P < 0.0001 (t = 38.56); LvH6, ∗∗∗, P = 0.0003 (t = 11.46); LvH7, ∗∗, P = 0.0013 (t = 8.015); LvH8, ∗∗∗, P = 0.0004 (t = 11.15); d, Schematic illustration of screening bottleneck enzymes by either introducing additional Cob enzymes from R. capsulatus (Rc) or replacing S. meliloti (Sm) Cob enzymes with the corresponding RcCob enzymes. e, Comparison of HBA production between the H2 strain and H21–H28 strains, each harboring an additional RcCob enzyme. Two-sided unpaired t -test is carried out between H2 and H21 to 28. Unpaired t -test of data:H21, ns, P = 0.0940 (t = 2.187); H22, ns, P = 0.0522 (t = 2.734); H23, ns, P = 0.5933 (t = 0.5795); H24, ∗, P = 0.0121 (t = 4.359); H25, ns, P = 0.0964 (t = 2.164); H26, ∗∗∗, P = 0.0003 (t = 11.48); H27, ns, P = 0.7136 (t = 0.3941); H28, ns, P = 0.1479 (t = 1.790). f, Comparison of HBA production between the H2 strain and H37–H44 strains, in which SmCob enzymes were replaced with the corresponding RcCob enzymes. Two-sided unpaired t -test is carried out between H2 and H37 to 44. Unpaired t -test of data: H37, ∗∗∗∗, P < 0.0001 (t = 17.94); H38, ∗∗∗, P = 0.0001 (t = 14.14); H39, ∗∗∗∗, P < 0.0001 (t = 28.35); H40, ∗∗∗∗, P < 0.0001 (t = 21.08); H41, ns, P = 0.2829 (t = 1.240); H42, ∗∗∗∗, P < 0.0001 (t = 89.48); H43, ∗, P = 0.0252 (t = 3.488); H44, ∗∗∗∗, P < 0.0001 (t = 120.7).

Article Snippet: After 12 h of reaction, CCE derived from strain H2 produced a higher HBA titer (3.16 ± 0.14 μM) than those from H1 and LvH0, consistent with the trends observed in vivo ( b, MS identification of synthetic HBA and Urogen III are showed in ).

Techniques: Plasmid Preparation, Comparison

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Screening of bottleneck reaction steps in the 1w–SmHBA operon by multienzyme catalysis. a, Schematic illustration of the reaction components. The in vitro reaction system consisted of crude cell extracts (CCE) from strains harboring the 1w–SmHBA plasmid as the basic catalytic source, supplemented with multienzyme-expressing strains carrying plasmids encoding 4–5, 2–3, or single heterologously expressed Cob enzymes to alter the enzyme composition in each reaction. b, Comparison of in vitro HBA production using CCE from strains H1, H2, and LvH0. Two-sided unpaired t -test is carried out between H1, H2, and LvH0. Unpaired t -test of data: H1 to H2, ∗∗∗∗, P < 0.0001 (t = 35.12); LvH0 to H2, ∗∗∗∗, P < 0.0001 (t = 30.84). c, In vitro HBA production using CCE from H21–H28 strains, each harboring an additional RcCob enzyme. Red bars indicate values higher than those of the H2 reactant, whereas blue bars indicate values lower than those of H2 reactant. Two-sided unpaired t -test is carried out between H1 to H21-28. Unpaired t -test of data:H2 to H21, ∗, P = 0.0227 (t = 3.605); H24 to H2, ∗∗∗∗, P < 0.0001 (t = 24.60); H27 to H2, ∗, P = 0.5363 (t = 0.6757). d, In vitro HBA production using CCE from H37–H44 strains, in which SmCob enzymes were replaced with the corresponding RcCob enzymes. Two-sided unpaired t -test is carried out between H1 to H37-44. Unpaired t -test of data:H37 to H2, ∗, P = 0.0145 (t = 4.129); H41 to H2, ∗, P = 0.0315 (t = 3.246); H42 to H2, ∗∗∗, P = 0.0003 (t = 11.66); H43 to H2, ∗, P = 0.0229 (t = 3.592); H44 to H2, ∗∗∗∗, P < 0.0001 (t = 15.76). e, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 4–5 Cob enzymes. HBA-A: CCE with pET28a–CobAIGJM; HBA-B: CCE with pACYCDuet-1–CobFKLH. Unpaired t -test of data: HBA-A 45 OD 600 to Control in HBA titer, ns, P = 0.0729 (t = 2.418); HBA-B 45 OD 600 to Control in HBA titer, ∗∗, P = 0.0029 (t = 6.502); HBA-A 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0022 (t = 7.002); HBA-B 45 OD 600 to Control in Urogen III titer, ∗∗, P = 0.0011 (t = 8.309); f, Screening of reactions supplemented with crude cell extracts from strains heterologously expressing 2–3 Cob enzymes. AIG: CCE with pet28a-CobAIG; JM: CCE with pet28a-CobJM; FK: CCE with pet28a-CobFK; LH: CCE with pet28a-CobLH. Unpaired t -test of data: AIG 30 OD600 to Control in Urogen III titer, ∗∗∗, P = 0.0009 (t = 8.801); JM 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 24.39); FK 30 OD600 to Control in Urogen III titer, ∗, P = 0.0304 (t = 3.285); LH 30 OD600 to Control in Urogen III titer, ∗∗∗∗, P < 0.0001 (t = 15.93); g, Screening of reactions supplemented with CCE from strains heterologously expressing a single Cob enzyme. Unpaired t -test of data: CobA + to ori, ∗∗∗, P = 0.0002 (t = 13.95); CobI + to ori, ∗∗∗, P = 0.0005 (t = 10.57); CobG + to ori, ∗∗∗, P = 0.0003 (t = 12.00); CobJ + to ori, ∗∗∗, P = 0.0001 (t = 15.48); CobM + to ori, ∗∗∗, P = 0.0005 (t = 10.46); CobF + to ori, ∗∗∗∗, P < 0.0001 (t = 16.42); CobK + to ori, ∗∗∗, P = 0.0006 (t = 9.873); CobL + to ori, ∗∗, P = 0.0042 (t = 5.882); CobH + to ori, ∗∗∗, P = 0.001 (t = 8.675).

Article Snippet: After 12 h of reaction, CCE derived from strain H2 produced a higher HBA titer (3.16 ± 0.14 μM) than those from H1 and LvH0, consistent with the trends observed in vivo ( b, MS identification of synthetic HBA and Urogen III are showed in ).

Techniques: In Vitro, Plasmid Preparation, Expressing, Comparison, Control

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Optimization of the artificial HBA operon. a, In vitro HBA production using CCE from various strains under different modifications. promoter replacement (H2 vs. H46 and H47, trc or tac promoters): Two-sided unpaired t -test is carried out between H2 to H46 and H47. Unpaired t -test of data: H46 to H2, ∗∗∗∗, P < 0.0001 (t = 23.77); H47 to H2, ∗∗∗∗, P < 0.0001 (t = 33.93); chassis variation (H2 vs. H48–H52, different E. coli strains harboring the SmHBA plasmid): Two-sided unpaired t -test is carried out between H2 to H48-52, but no significant differences were observed; culture medium effects (H2 grown in different media): Unpaired t -test of data: 2YT to LB, ∗, P = 0.0168 (t = 3.954); and operon composition (H2 vs. H53 and H54, carrying the FKLHIGJM or FKLH + operon): Two-sided unpaired t -test is carried out between H2 to H53 and H54. Unpaired t -test of data: H53 to H2, ∗∗∗∗, P < 0.0001 (t = 44.32). b, Schematic illustration of the ASmHBA (H2 strain), FKLHIGJM (H53 strain), and FKLH + (H54 strain) operons. c, In vitro HBA synthesis using CCEs from H57 cultivated in 2YT medium, supplemented with gradient amounts of either CCE or CobA enzyme. d. Chromatogram of synthetic HBA in CCE3 group, CCE1 group in (c) with standards. UroIII-STD, uroporphyrinogen III standard; HBA-STD, hydrogenobyrinate acid standard; CCE3 and CCE1, reactants using 1-fold H57 CCE and 3-fold H57 CCE in (c).

Article Snippet: After 12 h of reaction, CCE derived from strain H2 produced a higher HBA titer (3.16 ± 0.14 μM) than those from H1 and LvH0, consistent with the trends observed in vivo ( b, MS identification of synthetic HBA and Urogen III are showed in ).

Techniques: In Vitro, Plasmid Preparation

Journal: Synthetic and Systems Biotechnology

Article Title: Reconstructing the hydrogenobyrinic acid synthetic toolkit by combining cell-free systems and metabolic engineering

doi: 10.1016/j.synbio.2026.01.012

Figure Lengend Snippet: Optimization of the HBA synthetic system. a, HBA titers obtained from screening single Cob enzyme supplementation in the H53∗ CCE reaction: the control H53 CCE reaction, H53–CobA + (H53 CCE with 4 g/L CobA added), and H53–CCE + (H53 CCE with threefold CCE input). ∗ indicates that H53 CCE was prepared from cultures grown in LB medium, whereas unmarked H53 CCE was prepared from cultures grown in 2YT medium. Statistical Significance without bracket denotes unpaired t -test comparisons between each supplemented group (A+, I+, G+, J+, M+, F+, K+, L+, H+) and the corresponding Ori group under the same CCE introduction condition. Unpaired t -test of data: H53-CobA + to H53∗, ∗∗∗, P = 0.0004 (t = 10.67), H53-CCE + to H53∗, ∗∗∗∗, P < 0.0001 (t = 31.05). b, Urogen III accumulation under the same conditions as in (a). Unpaired t -test of data: H53-CobA + to H53∗, ns, P = 0.2509 (t = 1.341), H53-CCE + to H53∗, ∗∗∗∗, P < 0.0001 (t = 61.82). c, Schematic illustration of HBA biosynthesis with SAM supplementation. Abbreviation: 5-ALA, 5-Aminolevulinate; PBG, porphobilinogen; HMB, hydroxymethylbilane; L-Met, l -methionine; SAM, S-adenosyl- l -methionine; SAH, S-adenosyl- l -homocysteine; SRH, S-ribosyl- l -homocysteine. d, Orthogonal combinations of PpK, MetK, and MtnN enzymes used to enhance HBA production via SAM supplementation. The color intensity of the bar corresponds to the magnitude of the values for enhanced visual clarity. e, HBA titers in the optimized SAM supplementation system by adjusting ATP synthesis through varying AMP and SHMP inputs.

Article Snippet: After 12 h of reaction, CCE derived from strain H2 produced a higher HBA titer (3.16 ± 0.14 μM) than those from H1 and LvH0, consistent with the trends observed in vivo ( b, MS identification of synthetic HBA and Urogen III are showed in ).

Techniques: Control