Journal: eLife

Article Title: De novo-designed transmembrane domains tune engineered receptor functions

doi: 10.7554/eLife.75660

Figure Lengend Snippet: ( a ) Rosetta fold, dock, and design uses backbone fragments from natural MPs to construct symmetric, de novo architectures and an MP energy function to optimize the amino acid sequence. ( b ) Round 1 designs were biased towards the hydrophobic amino acid Leu relative to naturally occurring transmembrane domains (TMDs). In round 2, we incorporated a sequence diversification step that conformed the amino acid propensities to those observed in natural TMDs. ( c ) The programmed membrane proteins (proMPs) strongly self-associate in the E. coli inner membrane as evidenced by high viability in the deep sequencing TOXCAT-β-lactamase (dsTβL) self-association assay . The TMDs from human quiescin sulfhydryl oxidase 2 (QSOX2) and ErbB2 provide positive controls for TMD self-association, whereas the C-terminal portion of human L-selectin (CLS) provides a negative control. ( d–f ) Designed positions that are buried at the interface (orange) are more sensitive to mutation according to dsTβL analysis ( y -axis) than exposed positions (blue). Mutations are predicted to be detrimental or neutral/beneficial using computational mutation scanning of the model structures (Materials and methods). Changes in self-association energies upon mutation are computed according to . ( g ) proMPs produced as free peptides form SDS-stable homo-oligomers. SDS-PAGE samples containing approximately 15, 45, and 135 μg of peptide were heated to 95℃ for 1 min and run under reducing conditions. * indicates the position of a minor contaminant from the fusion protein used to generate proMP peptides (Materials and methods). Molecular weight below each gel is for a monomer of the corresponding peptide sequence with additional N-terminal EPE and C-terminal RRLC flanking sequences (Materials and methods). See additional examples in . ( h, i ) The 2.55 Å resolution structure (blue ribbon) determined from crystals grown in monoolein lipid cubic phase (LCP) shows that proMP 1.2, designed to form a dimer, associates to form a trimer in a lipid bilayer environment. ( i ) Forward-folding ab initio prediction of proMP 1.2 in trimeric (C3) symmetry results in a model structure ( h , gray ribbon) that is very close to the experimentally determined one.

Article Snippet: Recombinant DNA reagent , pMAL-dsTbL(plasmid) , , Addgene:73805 , TOXCAT β-lactamase assays.

Techniques: Construct, Sequencing, Membrane, Histone Association Assay, Negative Control, Mutagenesis, Produced, SDS Page, Molecular Weight

Journal: eLife

Article Title: De novo-designed transmembrane domains tune engineered receptor functions

doi: 10.7554/eLife.75660

Figure Lengend Snippet: ( a ) De novo-designed proMPs exhibited a high proportion of the large, flexible, and hydrophobic amino acid Leu (bold), in accordance with the high lipophilicity of this amino acid according to the deep sequencing TOXCAT-β-lactamase (dsTβL) lipophilicity scale. Forward-folding ab initio calculations in C2 symmetry, however, exhibited a flat energy landscape with multiple low-energy structures that diverged from the design. ( b ) Simulated annealing Monte Carlo simulations starting from the sequence in ( a ) augmented with a potential that biased sequence choices to the propensities observed in natural transmembrane domains (TMDs), resulting in sequences with fewer Leu amino acids. Although the Rosetta energies of the designs prior to sequence diversification were more favorable than after (compare the y-axes of the two plots), the sequence-diversified sequences clearly converged to the designed structure. These results suggest that the sequence composition of natural TMDs encodes negative design principles that ensure folding to a unique conformation. Red one-letter codes indicate positions at the homodimer interface; smooth red and green lines are visual aids. The oligomeric state (dimers) is noted by the pair of black dots at the bottom of the figure.

Article Snippet: Recombinant DNA reagent , pMAL-dsTbL(plasmid) , , Addgene:73805 , TOXCAT β-lactamase assays.

Techniques: Sequencing

Journal: eLife

Article Title: De novo-designed transmembrane domains tune engineered receptor functions

doi: 10.7554/eLife.75660

Figure Lengend Snippet: Each design and a set of single-point mutants were introduced in selected positions (bold) and assayed using the TOXCAT-β-lactamase (TβL) selection scheme. Amino acids that are positioned at the interface are in orange and those that are lipid-facing are in blue. The mutations were generated using DNA oligos with the degenerate codon NYS, encoding the amino acid identities: Ala, Val, Ile, Leu, Phe, Met, Ser, Thr, and Pro (mutations to Pro were excluded from the analysis). We note that programmed membrane protein (proMP) 1.3, whose migration on SDS-PAGE indicates an oligomer larger than a dimer, also shows poor segregation of mutations at predicted interface versus lipid-exposed positions in TβL selection, consistent with the formation of a structure that is different from the design model in this case.

Article Snippet: Recombinant DNA reagent , pMAL-dsTbL(plasmid) , , Addgene:73805 , TOXCAT β-lactamase assays.

Techniques: Selection, Generated, Membrane, Migration, SDS Page

Journal: eLife

Article Title: De novo-designed transmembrane domains tune engineered receptor functions

doi: 10.7554/eLife.75660

Figure Lengend Snippet: ( a ) Ab initio structure predictions for proMP-1 in C2, C3, and C4 symmetries exhibit flat energy landscapes, suggesting that the design would not form homo-oligomers. ( b ) The design model of proMP-1 shows no flat surfaces that are prone to self-association; yet, the sequence is highly hydrophobic with apparent insertion energy computed using the deep sequencing TOXCAT-β-lactamase (dsTβL) scale of –18.7 kcal/mol (compared to –5.9 kcal/mol for a reference monomer [the human CLS sequence]). ( c ) The design is extremely sensitive to chloramphenicol in the dsTβL assay (compared to the weak natural homodimer transmembrane domain [TMD] from human quiescin sulfhydryl oxidase 2 [QSOX2], which served as a reference), verifying that it does not form homo-oligomers.

Article Snippet: Recombinant DNA reagent , pMAL-dsTbL(plasmid) , , Addgene:73805 , TOXCAT β-lactamase assays.

Techniques: Sequencing

Journal: eLife

Article Title: De novo-designed transmembrane domains tune engineered receptor functions

doi: 10.7554/eLife.75660

Figure Lengend Snippet:

Article Snippet: Recombinant DNA reagent , pMAL-dsTbL(plasmid) , , Addgene:73805 , TOXCAT β-lactamase assays.

Techniques: Flow Cytometry, Fluorescence, Microscopy, Recombinant, Sequencing, Concentration Assay, Enzyme-linked Immunosorbent Assay, Modification, Saline, Software

Journal: bioRxiv

Article Title: Carbohydrate adaptation drives liver-brain axis maturation

doi: 10.1101/2025.11.05.685548

Figure Lengend Snippet: a. Top-ranked functional groups of differential genes significantly associated with developmental progression. b. Transmission electron microscopy (TEM) images of hepatic mitochondria across postnatal stages. Scale bar, 0.2 um. c. Comparison of ETC complex activity measured as oxygen consumption rate in isolated liver mitochondria across postnatal stages ( N = 4-6 per group). Data are means ± SEM and analyzed by two-tailed unpaired t-test, * P < 0.05, **** P < 0.0001. d. Waterfall plot of nuclear-encoded mitochondrial genes involved in mtDNA gene regulation and ETC function, ranked by their correlation with developmental progression ( P < 0.05). e. Quantitative RT-PCR of Lactb2 expression in developing livers (mean ± SEM, N = 4 per stage). Two-tailed unpaired t-test compared of each stage to P0. * P < 0.05, ** P < 0.01. f. Immunoblot analysis of LACTB2, ETC complex proteins and mitochondrial VDAC in developing livers with vinculin as loading control. g. Immunohistochemical (IHC) staining of LACTB2 in major mouse organs using mouse tissue microarrays. Scale bar, 50 µm. h. scRNA-seq scatter plot showing Lactb2 expression levels across distinct hepatic cell populations during postnatal development.

Article Snippet: After fixation with fresh warm 4% paraformaldehyde in PBS for 15 minutes at room temperature, cells were permeabilized in 0.1% (v/v) Triton X-100 in PBS for 10 minutes, pre-blocked with filtered 1% BSA (w/v) in PBS for 1 hour and then incubated with primary antibody against LACTB2 (1:200, Proteintech, 16783-1-AP), and HPS60 (1:100, ThermoFisher, MA3-012) diluted in the blocking buffer for 1 hour at room temperature.

Techniques: Functional Assay, Transmission Assay, Electron Microscopy, Comparison, Activity Assay, Isolation, Two Tailed Test, Quantitative RT-PCR, Expressing, Western Blot, Control, Immunohistochemical staining, Immunohistochemistry

Journal: bioRxiv

Article Title: Carbohydrate adaptation drives liver-brain axis maturation

doi: 10.1101/2025.11.05.685548

Figure Lengend Snippet: a. Northern blot analysis of mt-RNAs in the indicated isogenic H2.35 hepatocytes. Nuclear 18S rRNA is used as a loading control. b. Blue-Native PAGE analysis of ETC supercomplexes in mitochondria isolated from control and LACTB2-depleted H2.35 hepatocytes. c. ETC complex activity assessed by oxygen consumption rates in the presence of complex-specific substrates in the indicated H2.35 hepatocytes. Results are means ± SEM ( N = 6) and p values are calculated by two-tailed unpaired t-test, * P < 0.05, ** P < 0.0, **** P < 0.0001, ns, not significant. d. Immunoblot analysis of ETC complex proteins in LACTB2-depleted hepatocytes with or without LACTB2 re-expression. The cells were treated with tigecycline at 50 uM for 72h, followed by drug withdraw for the indicated hours. “-” indicates DMSO treatment. e. Survival of WT, Het and KO male and female mice. N numbers are indicated for each group; P value by log-rank Mantel-Cox test. f. Body weight gain up to 6 weeks postpartum. Data are mean ± SD ( N = 4-6 per group) and *** P < 0.001 by mixed-effects model. g. TEM of WT and KO male livers at P35 ( N = 3 per group). Scale bar: 0.5 µm. h. ETC complex activity measured by OCR in isolated liver mitochondria from WT, Het and KO males. Results are mean ± SEM ( N = 3-4 mice per group) and analyzed by two-tailed unpaired t-test. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, n.s. not significant. i. Relative abundance of mitochondrial rRNAs and mRNAs in WT, Het and KO male livers at P35 as assessed by strand-specific RNA-seq ( N = 4 per group). Results are mean ± SD and p values are calculated by two-tailed unpaired t-test, * P < 0.05, ** P < 0.01. j. Blue-Native PAGE analysis of ETC supercomplexes using mitochondria isolated from WT and KO male livers at P35. k. Immunoblot analysis of ISR markers in WT and KO male mouse livers at P35 ( N = 3 per group). l. Survival of KO male mice treated with AAV8-LACTB2, AAV8-NDI1 and ISRIB. N numbers are indicated for each group. P values by log-rank Mantel-Cox test: P < 0.0001 for AAV8-LACTB2 and AAV8-NDI1; P = 0.0022 for ISRIB.

Article Snippet: After fixation with fresh warm 4% paraformaldehyde in PBS for 15 minutes at room temperature, cells were permeabilized in 0.1% (v/v) Triton X-100 in PBS for 10 minutes, pre-blocked with filtered 1% BSA (w/v) in PBS for 1 hour and then incubated with primary antibody against LACTB2 (1:200, Proteintech, 16783-1-AP), and HPS60 (1:100, ThermoFisher, MA3-012) diluted in the blocking buffer for 1 hour at room temperature.

Techniques: Northern Blot, Control, Blue Native PAGE, Isolation, Activity Assay, Two Tailed Test, Western Blot, Expressing, RNA Sequencing