UNBLENDING OF TRANSCRIPTIONAL CONDENSATES IN HUMAN REPEAT EXPANSION DISEASE
20230168255 · 2023-06-01
Assignee
Inventors
Cpc classification
G01N33/6845
PHYSICS
International classification
Abstract
The present invention relates to a method of determining the capacity of a Cluster 1 mammalian Transcription Factor (TF) for phase separation and/or the capacity for forming a transcriptional condensate in a sample, including a method of determining the presence, localization and/or morphology of a transcriptional condensate comprising said Transcription Factor, and/or of determining the composition of a transcriptional condensate comprising at least one Cluster 1 mammalian TF and/or of determining the transcriptional activity of the TF or a condensate comprising the TF. Further, the present invention relates to an active agent for use in a method of preventing and/or treating a disorder associated with, caused by and/or accompanied with a dysfunction of a biomolecular condensate comprising at least one Cluster 1 mammalian TF.
Claims
1. An in vitro method of determining the capacity of a Cluster 1 mammalian Transcription Factor for phase separation and/or for forming a transcriptional condensate, comprising determining the presence, localization and/or morphology of a transcriptional condensate comprising said Transcription Factor, and/or determining the composition of a transcriptional condensate comprising said Transcription Factor, and/or determining the transcriptional activity of said Transcription Factor or a condensate comprising the Transcription Factor, wherein the Transcription Factor is a Cluster 1 human Transcription Factor selected from the group of human Cluster 1 Transcription Factors shown in Table 1 or an ortholog thereof from a non-human organism and more particularly selected from the group consisting of HOXD13, HOXA13, RUNX2, SOX3, FOXL2, ZIC2, ARX, PHOX2B, and AR.
2. The method of claim 1 wherein an altered capacity for phase separation and/or for forming a transcriptional condensate versus the wild-type is indicative for the presence of a disease-associated mutation in the Cluster 1 Transcription Factor, and wherein the disease mutation comprises an alteration of the number of hydrophobic amino acids and/or an alteration of the number of hydrophilic amino acids and/or an alteration of the overall hydrophobicity in the Transcription Factor, particularly an alteration of an amino acid repeat in an intrinsically disordered region (IDR) of the Transcription Factor.
3. The method of claim 2 wherein the disease-associated mutation comprises an expansion of an amino acid repeat in an IDR of the Transcription Factor.
4. The method of claim 3 wherein the expansion comprises an expansion of at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more amino acid residues, particularly hydrophobic amino acid residues, in an IDR of the Transcription Factor.
5. The method of claim 2, wherein the amino acid repeat is an alanine repeat in an IDR of the Transcription Factor.
6. The method of claim 1, wherein a sample is selected from a body fluid sample, a tissue sample and a cell culture sample.
7. The method of claim 1, wherein the presence, localization and/or morphology of a transcriptional condensate comprising said Transcription Factor is determined by an optical method, including but not limited to high-resolution imaging, confocal imaging, and stochastic optical reconstruction microscopy.
8. The method of claim 1, wherein determining the composition of a transcriptional condensate comprising said Transcription Factor comprises determining the presence and/or amount of at least one transcriptional co-factor such as Mediator or a sub-unit thereof, BRD4.
9. The method of for detecting a genetic disease in a subject wherein an altered capacity for phase separation and/or for forming a transcriptional condensate versus the wild-type is indicative for a genetic disease.
10. The method of claim 1 for screening a test compound whether it shows an effect on the capacity for phase separation and/or for forming a transcriptional condensate.
11. The method of claim 10 comprising determining the effect of the test compound on the capacity for phase separation and/or for forming a transcriptional condensate in the presence of a Transcription Factor comprising an altered capacity versus the wild-type.
12. The method of claim 10 wherein said test compound is an amphiphilic compound comprising (i) a hydrophobic component comprising a hetero-aromatic nitrogen-containing base, e.g. a purine or pyrimidine base, and (ii) a hydrophilic component, e.g. comprising a sugar and/or a phosphate group, and wherein said test compound is particularly a nucleoside diphosphate including a non-hydrolyzable nucleoside diphosphate analog or a nucleoside triphosphate such as ATP including a non-hydrolyzable nucleoside triphosphate analog such as adenosine-5′-[(β,γ)-imido]triphosphate (APPNP).
13. An amphiphilic compound, which is a nucleoside phosphate, particularly a nucleoside diphosphate including a non-hydrolyzable nucleoside diphosphate analog or a nucleoside triphosphate such as ATP including a non-hydrolyzable nucleoside triphosphate analog such as adenosine-5′-[(β,γ)-imido]triphosphate (APPNP), for use in a method of preventing and/or treating a disorder associated with, caused by and/or accompanied with a dysfunction of a transcriptional condensate comprising at least one Cluster 1 mammalian Transcription Factor.
14. An amphiphilic compound for the use of claim 13, wherein the disorder is a genetic disorder, particularly a genetic disorder, more particularly a genetic disorder, which is associated with, caused by and/or accompanied a disease-associated mutation in the Cluster 1 Transcription Factor.
15. In vitro use of an amphiphilic compound, which is a nucleoside phosphate, particularly a nucleoside diphosphate including a non-hydrolyzable nucleoside diphosphate analog or a nucleoside triphosphate such as ATP including a non-hydrolyzable nucleoside triphosphate analog such as adenosine-5′-[(β,γ)-imido]triphosphate (APPNP), for modulating the capacity of a Cluster 1 mammalian Transcription Factor for phase separation and/or for forming a transcriptional condensate, particularly for modulating the presence, localization and/or morphology of a transcriptional condensate comprising said Transcription Factor, and/or for modulating the composition of a transcriptional condensate comprising said Transcription Factor, and/or for modulating the transcriptional activity of said Transcription Factor or a condensate comprising the Transcription Factor.
Description
LEGENDS OF FIGURES
[0050]
[0051] (A) Disease-associated repeat expansions in humans. (a.a: amino acid)
[0052] (B) (left) Hoxd13 whole mount in situ hybridization in an E12.5 mouse embryo. (right) HOXD13 Immunofluorescence (IF) in E12.5 mouse limb bud cells.
[0053] (C) Stochastic optical reconstruction microscopy (STORM) images of E12.5 mouse limb bud cells. The zoomed-in area on the right is highlighted with a red box on the left.
[0054] (D) Graph plotting intrinsic disorder for human HOXD13. The IDR cloned for subsequent experiments is highlighted with a purple bar.
[0055] (E) Scheme of the optoDroplet assay. The optolDR construct consists of the HOXD13 IDR fused to mCherry and the A. thaliana CRY2 PHR domain.
[0056] (F) Representative images of live HEK-293T cells expressing mCherry-CRY2 (top) and HOXD13 IDR-mCherry-CRY2 (bottom) fusion proteins. Cells were stimulated with 488 nm laser every 20 s for 3 minutes.
[0057] (G) Quantification of the fraction of the cytoplasmic area occupied by HOXD13 IDR-mCherry-CRY2 and mCherry-CRY2 droplets in HEK-293T cells over time. Data displayed as mean+/−SEM.
[0058] (H) Fluorescence intensity of HOXD13 IDR-mCherry-CRY2 droplets before, during and after photobleaching. Data displayed as mean+/−SD.
[0059] (I) Time lapse images of a droplet fusion event in HEK-293T cells expressing HOXD13 IDR-mCherry-CRY2 fusion protein.
[0060] (J) (left) Representative images of droplet formation by purified HOXD13-mCherry and mCherry at the indicated concentrations. (right) Phase diagram of HOXD13-mCherry in the presence of different concentrations of PEG-8000. The size of the circles is proportional to the size of droplets detected in the respective buffer conditions.
[0061]
[0062] (A) Amino acid composition of human HOXD13. Ticks represent amino acids indicated on the y-axis at the positions indicated on the x-axis. The IDR cloned for subsequent experiments is highlighted with a purple bar.
[0063] (B) Alanine residues (As) within the HOXD13 IDR sequence are indicated as red ticks. The central alanine repeat consists of 15 As in the wild-type (wt) protein.
[0064] (C) Representative images of live HEK-293T nuclei expressing wt and repeat-expanded HOXD13 IDR-mCherry-CRY2 fusion proteins. Cells were stimulated with 488 nm laser every 20 s for 3 minutes. Arrowheads highlight spontaneously forming IDR condensates present without 488 nm laser stimulation.
[0065] (D) Quantification of the fraction of the nuclear area occupied by HOXD13 wt IDR-mCherry-CRY2 and HOXD13 +7A IDR-mCherry-CRY2 droplets in HEK-293T cells over time. Data displayed as mean+/−SEM.
[0066] (E) Fluorescence intensity of light-induced wt, +7A and +8A HOXD13 IDR droplets before, during and after photobleaching. Data displayed as mean+/−SD.
[0067] (F) Fluorescence intensity of +8A and +9A spontaneously formed HOXD13 IDR condensates before, during and after photobleaching. Data displayed as mean+/−SD.
[0068] (G) Representative images of droplet formation by purified HOXD13 IDR-mCherry fusion proteins in droplet formation buffer.
[0069] (H) Phase diagram of HOXD13 IDR-mCherry fusion proteins. Every dot represents a detected droplet. The inset depicts the projected average size of the droplets as mean+/−SD (middle circle: mean, inner and outer circle: SD). n.d.: not detected.
[0070]
[0071] (A) Representative images of droplet formation by purified MED1-IDR-GFP and HOXD13 IDR-mCherry fusion proteins in droplet formation buffer.
[0072] (B) Quantification of GFP and mCherry fluorescence intensity in HOXD13 IDR-mCherry containing droplets in the indicated MED1 IDR-GFP mixing experiments. Each dot represents one droplet, and the size of the dot is proportional to the size of the droplet.
[0073] (C) Quantification of the ratio of GFP and mCherry fluorescence intensity in HOXD13 IDR-mCherry containing droplets in the indicated MED1 IDR-GFP mixing experiments. P value is from a Welch's t-test.
[0074] (D) Quantification of mCherry fluorescence intensity in MED1 IDR-GFP containing droplets in the indicated MED1 IDR-GFP mixing experiments. P value is from a Welch's t-test.
[0075] (E) Quantification of GFP and mCherry fluorescence intensity in HOXD13 IDR-mCherry containing droplets. Each dot represents one droplet. The size of the dot is proportional to the size of the droplet, and the color of the dot is scaled to the MED1 signal in the droplet. The insets show a simplified phase diagram of HOXD13 IDRs based on the data displayed in
[0076] (F) Representative images of the mixtures in (E).
[0077] (G) Representative images of droplets formed by purified MED1-IDR-GFP and HOXD13 IDR-mCherry fusion proteins.
[0078] (H) Quantification of GFP and mCherry fluorescence intensity in HOXD13 IDR-mCherry containing droplets in the indicated MED1 IDR-GFP mixing experiments. Each dot represents one droplet, and the size of the dot is proportional to the size of the droplet. x-axis is in log.sub.10 scale.
[0079] (I-J) Quantification of the ratio of GFP and mCherry fluorescence in HOXD13 IDR-mCherry containing droplets in the indicated mixing experiments. In (I), the y-axis is in log.sub.10 scale. [0080] (K) Condensate unblending model of the impact of HOXD13 alanine repeat expansions.
[0081]
[0082] (A) (left) Experimental scheme (right) Stochastic optical reconstruction microscopy (STORM) images of wt and spdh E12.5 mouse limb bud cell nuclei. The zoomed-in area on the right is highlighted with a white box in the middle.
[0083] (B) Manders overlap coefficients of the STORM co-localizations. P value is from an unpaired t-test.
[0084] (C) (left) Experimental scheme (right) Representative images of wild type and spdh mouse limb bud cells with or without treatment with 6% 1,6-hexanediol for 1 min.
[0085] (D) Quantification of signal within HOXD13 puncta in mouse limb bud cells [displayed in (C)] with or without treatment with 6% 1,6-hexanediol for 1 min.
[0086] (E) Fluorescence images of ectopically expressed Med1 IDR-YFP in U2OS cells co-transfected with the indicated HOXD13 IDR-LacI-CFP fusion constructs.
[0087] (F) Quantification of the relative MED1 IDR-YFP signal intensity in the HOXD13 IDR foci. P values are from a Dunnett's test.
[0088] (G) Luciferase reporter assays of HOXD13 wt, +7A and +10A mutants co-expressed with a Ra/dh2-luciferase reporter construct.
[0089]
[0090] (A) Scheme of the scRNA-Seq experiment strategy.
[0091] (B) Visualization of the wild-type scRNA-seq data using t-distributed Stochastic Neighbor Embedding (t-SNE).
[0092] (C) Changes in cell type composition in spdh limb buds. Displayed are the relative changes in the proportions of cells that belong to the designated clusters (i.e. cell states) between wt and spdh limb buds.
[0093] (D) Heatmap of differentially up- or downregulated genes in the spdh limb bud relative to wt within the 11 cell clusters. Arrowhead highlights the interdigital mesenchymal cells.
[0094] (E) Gene Ontology (GO) term enrichment analysis of differentially up- or downregulated genes in the spdh limb bud relative to wt within individual cell clusters.
[0095] (F) Profiles of Capture C, HOXD13 ChIP-Seq and scRNA-Seq data at the Msx1 locus. The mean expression value in spdh (red) and wt cells (blue) within each cluster are also displayed. Arrowhead highlights the expression level in interdigital mesenchymal cells, where the expression difference is the most profound.
[0096] (G) Number of HOXD13 peaks in topologically associating domains (TADs) that contain a gene dysregulated in Cluster 4. P value is from a Wilcoxon rank sum test.
[0097] (H) Mean Capture C signal around HOXD13 peaks within topologically associating domains (TADs) that contain a gene dysregulated in Cluster 4.
[0098] (I) ChIP-Seq binding profiles around the Msx2 locus.
[0099] (J) Quantification of the mean H3K27Ac signal at the nearest HOXD13 binding sites around the indicated genes within the same TAD. P value is from a Wilcoxon rank sum test.
[0100]
[0101] (A, J, S) Graphs plotting intrinsic disorder for HOXA13, RUNX2 and TBP. The IDRs cloned for subsequent experiments are highlighted with a purple bar.
[0102] (B, K, T) Representative images of HEK-293T nuclei expressing the indicated TF IDR-mCherry-CRY2 fusion proteins. Cells were stimulated with 488 nm laser every 20 s for 3 minutes.
[0103] (C, L, U) Quantification of the fraction of the nuclear area occupied by droplets of the indicated TF IDR-mCherry-CRY2 fusion proteins in HEK-293T nuclei over time. Data displayed as mean+/−SEM.
[0104] (D, M, V) Fluorescence intensity of droplets of the indicated TF IDR-mCherry-CRY2 fusion proteins before, during and after photobleaching. For the HOXA13 +7A IDR and the RUNX2 +10A IDR the spontaneously formed droplets were bleached, for all other fusion proteins the light-induced droplets were bleached. Data displayed as mean+/−SD.
[0105] (E, N) Representative images of droplet formation by purified TF IDR-mCherry fusion proteins in droplet formation buffer.
[0106] (F, O) Phase diagram of TF IDR-mCherry fusion proteins. Every dot represents a detected droplet. The inset depicts the projected average size of the droplets as mean+/−SD (middle circle: mean, inner and outer circle: SD). n.d.: not detected
[0107] (G, P) Representative images of droplet formation by purified MED1-IDR-GFP and TF IDR-mCherry fusion proteins in droplet formation buffer with 10% PEG-8000.
[0108] (H-Q) Quantification of GFP and mCherry fluorescence intensity in TF IDR-mCherry containing droplets in the indicated Med1 IDR-GFP mixing experiments. Each dot represents one droplet, and the size of the dot is proportional to the size of the droplet.
[0109] (I-R) (left): GAL4 activation assay schematic. The luciferase reporter plasmid, and the expression vector for the GAL4 DBD-TF IDR fusion proteins were transfected into HEK-293T cells. (right): Luciferase reporter activity of the indicated TF IDRs fused to GAL4-DBD. p<10.sup.−3 for both wt/mutant comparisons (Dunnett's test).
[0110]
[0111] (A) Classification of human TF IDRs. The inner circle depicts the clusters of TF IDRs. The outer circle includes the annotation of the DBDs of the TFs whose IDRs were classified in the inner circle.
[0112] (B) Boxplot of PONDR scores (disorder) of human TF DBDs and IDRs.
[0113] (C) Boxplot of phyloP scores (conservation) of human TF DBDs and IDRs
[0114] (D) Enrichment of TFs whose IDRs belong to the seven IDR clusters for the indicated sequence features, functional and phenotypic categories. Red box highlights significant enrichment (q<0.05).
[0115] (E) Representative images of HEK-293T cells expressing the indicated HOXD13 IDR-mCherry-CRY2 fusion proteins. Cells were stimulated with 488 nm laser every 20 s for 3 minutes.
[0116] (F) Quantification of the fraction of the nuclear area occupied by HOXD13 IDR-mCherry-CRY2 droplets in HEK-293T cells over time. Data displayed as mean+/−SEM.
[0117] (G) Plot of the nuclear area occupied by HOXD13 IDR-mCherry-CRY2 droplets versus the Alanine content and Asp/Glu content of the IDR constructs. Data displayed as mean+/−SEM.
[0118] (H) (left): GAL4 activation assay schematic. The luciferase reporter plasmid, and the expression vector for the GAL4 DBD-TF IDR fusion proteins were transfected into HEK-293T cells. (right): Luciferase reporter activity of the indicated TF IDRs fused to GAL4-DBD. P values are from a Dunnett's test.
[0119] (I) Normalized luciferase activity of the indicated HOXA13 IDRs fused to GAL4 DBD. The blue line is a linear regression line, and the grey zones denote the 95% conference interval. P value is from a t-test.
[0120]
[0121] (A) Quantification of GFP and mCherry fluorescence intensity in HOXD13 IDR-mCherry containing droplets in the indicated MED1 IDR-GFP mixing experiments. Each dot represents one droplet, and the size of the dot is proportional to the size of the droplet. x-axis is in log.sub.in scale.
[0122] (B) Quantification of the ratio of GFP and mCherry fluorescence in HOXD13 IDR-mCherry containing droplets in the indicated mixing experiments.
[0123] APPNP: adenosine-5′-[(β,γ)-imido]triphosphate, source: NU-407, Jena Bioscience.
EXAMPLES
1. Experimental Model and Subject Details
Cell Culture
[0124] HEK-293T, Kelly, SH-SY5Y, U2OS-2-6-3, and Cos7 cells were cultured in DMEM with GlutaMAX (ThermoFisher Scientific, 10566-016) supplemented with 10% FBS (Sigma Aldrich, F4135), and 100 U/mL penicillin-streptomycin (Gibco, 15140), at 37° C. with 5% CO2 in a humidified incubator.
Mouse Husbandry
[0125] All animal procedures were conducted as approved by the local authorities (LAGeSo Berlin) under the license number #G0368/08.
2. Method Details
Construct Generation
[0126] All optolDR constructs used in this study were derived from the mCherry-Cry2WT plasmid (Addgene, 101221) described in (Shin et al., 2017). To generate mCherry-Cry2WT-NLS, a SV40 NLS primer pair (Sigma) was annealed and ligated into the mCherry-Cry2 plasmid. To generate optolDR constructs, sequences containing the IDRs for Hoxd13 wt, Hoxd13-7A, Hoxd13-15A, Hoxd13 DEdel, Hoxa13 wt, Hoxa13, Hoxa13 +7A, TBP 38Q, TBP 53Q, Runx2, Runx2 +10A were ordered as synthetic DNA from commercial vendors (Genewiz and IDT). IDR sequences were flanked by appropriate restriction sites for cloning. The IDR fragments were then ligated into mCherry-Cry2WT or mCherry-Cry2WT-NLS. For insertion of the alanine expansions +7A, +8A, +9A, and +14A into the Hoxd13-mCherry-Cry2-NLS construct, expansion oligonucleotides encoding respective alanine expansions were purchased (Sigma) and then inserted into the alanine stretch of Hoxd13-mCherry-Cry2-NLS using Gibson assembly, to generate Hoxd13(+7A)-mCherry-Cry2-NLS, Hoxd13(+8A)-mCherry-Cry2-NLS, Hoxd13(+9A)-Cry2-NLS, Hoxd13(+14A)-Cry2-NLS. All constructs were sequence verified. The nucleotide sequences of the constructs are shown in the attached sequence listing.
SV40 NLS forward primer (SEQ ID NO 1)
SV40 NLS reverse primer (SEQ ID NO 2)
+7A oligonucleotide for Gibson assembly (SEQ ID NO 3)
+8A oligonucleotide for Gibson assembly (SEQ ID NO 4)
+9A oligonucleotide for Gibson assembly (SEQ ID NO 5)
+14A oligonucleotide for Gibson assembly (SEQ ID NO 6)
Hoxd13 wt gene fragment (SEQ ID NO 7)
Hoxa13 wt gene fragment (SEQ ID NO 8)
Hoxa13 +7A gene fragment (SEQ ID NO 9)
TBP 38Q gene fragment (SEQ ID NO 10)
TBP 53Q gene fragment (SEQ ID NO 11)
Runx2 wt gene fragment (SEQ ID NO 12)
Runx2 +10A gene fragment (SEQ ID NO 13)
Multiple cloning site gene fragment (SEQ ID NO 14)
[0127] For the generation of IDR-YFP constructs, a YFP fragment was PCR amplified from an mCitrine vector and cloned into the mCherry-Cry2WT-NLS plasmid, to generate YFP-NLS. Hoxd13-IDR and Hoxd13(+7A)-IDRs were subcloned into YFP-NLS plasmid, to generate Hoxd13-IDR-YFP-NLS and Hoxd13(+7A)-IDRYFP-NLS. For the generation of Med1-IDR-YFP-NLS, the Med1-IDR was PCR amplified from a Med1 IDR expression vector (Boija et al., 2018; Sabari et al., 2018), and ligated into YFP-NLS.
mEYFP forward primer (SEQ ID NO 15)
mEYFP reverse primer (SEQ ID NO 16)
Med1 forward primer (SEQ ID NO 17)
Med1 reverse primer (SEQ ID NO 18)
[0128] For the generation of pET-IDR constructs, pET45b(+), mCherry was subcloned into pET45b(+) (Sigma Cat No. 71327), to generate pET-mCherry. IDRs were then PCR amplified from OptolDR vectors using Q5 polymerase (NEB M0494S), and cloned into pET-mCherry using the NEBuilder HiFi DNA Assembly master mix (NEB E2621L). This pipeline was used to generate pET45-Hoxd13(WT)-IDR, pET45-Hoxd13(+7A)-IDR, pET45-Hoxd13(+10A)-IDR, pET45-Runx2(WT)-IDR, pET45-Runx2(+10A)-IDR, pET45-Hoxa13(WT)-1DR, pET45-Hoxa13(+7A)-IDR, and pET45-TBP37Q-IDR.
[0129] In the Gal4-DBD luciferase system, GAL4 DBD-IDR fusion constructs were expressed from a backbone of GAL4-DBD expression vector [GI-GAL4-DBD (Addgene 42500)], from which GIGAL4 sequence was removed. GAL4-DBD was PCR amplified from pCAG-GAL4-DBD-GBP2 vector (Addgene 49439) while introducing a short multiple cloning site C-terminal to GAL4-DBD allowing IDR cloning. Synthetic DNA fragments for wt and alanine repeat-deletion IDRs of GSX2, HOXD11, TBX1, EOMES, BHLHE41, and MNX1 sequences were flanked by AsiSI and BsiWI sequences for cloning, and purchased from Genewiz. For HEY2 and OLIG2, wt IDR sequences were PCR amplified from human iPSC gDNA and alanine repeat-deletion sequences were ordered from Genewiz. For HOXA13 and HOXD13, IDR sequences were PCR amplified from previously described vectors, except HOXA13 −44A and −58A deletion sequences were ordered from Genewiz. The HOXA13 −15A construct lacks alanines 88-103. In case of RUNX2, sequences were cloned from optolDR constructs with SpeI and BsiWI restriction enzymes and inserted to GAL4-DBD vector with a longer multiple cloning site with NheI and BsiWI restriction sites.
GAL4-DBD fw primer (SEQ ID NO 19)
GAL4-DBD (short MCS) rev primer (SEQ ID NO 20)
GAL4-DBD (long MCS) rev primer (SEQ ID NO 21)
OLIG2 fw primer (SEQ ID NO 22)
OL/G2 rev primer (SEQ ID NO 23)
HEY2 fw primer (SEQ ID NO 24)
HEY2 rev primer (SEQ ID NO 25)
GSX2 wt sequence (SEQ ID NO 26)
GSX2 Ala-del sequence (SEQ ID NO 27)
HOXD11 wt sequence (SEQ ID NO 28)
HOXD11 Ala-del sequence (SEQ ID NO 29)
TBX1 wt sequence (SEQ ID NO 30)
TBX1 Ala-del sequence (SEQ ID NO 31)
EOMES wt sequence (SEQ ID NO 32)
BHLHE41 wt sequence (SEQ ID NO 33)
BHLHE41 Ala-del sequence (SEQ ID NO 34)
HOXA13 −58A (Ala-del) sequence (SEQ ID NO 35)
HOXA13 −44A sequence (SEQ ID NO 36)
MNX1 wt sequence (SEQ ID NO 37)
MNX1 Ala-del sequence (SEQ ID NO 38)
HEY2 Ala-del sequence (SEQ ID NO 39)
OLIG2 Ala-del sequence (SEQ ID NO 40)
[0130] For LacO-LacI experiments (
MCS fw primer (SEQ ID NO 41)
MCS rev primer (SEQ ID NO 42)
Protein Disorder Analysis
[0131] Intrinsically disordered regions (IDRs) in Hoxd13, TBP, Hoxa13, Runx2, and Foxp1-4 (
Isolation of Limb Bud Cells
[0132] Limb buds from E12.5 wild type and spdh homozygous embryos were micro-dissected individually and digested with Trypsin-EDTA 0.05% (Gibco) for 15 minutes at 37° C. and gently dissociated by pipetting after 5, 10, and 15 minutes. Cells were mixed in cell culture media (DMEM, 10% FBS, 2 mM L-Glutamine, 50 U/ml Penicillin/Streptomycin) and a single-cell suspension was obtained using a 40 μm cell strainer (Falcon). About 150,000 limb bud cells from each embryo were seeded in 1000 μl medium onto fibronectin-coated glass coverslips in 12 well plates. After 30-60 minutes, additional cell culture medium was added, and cells were grown for 24 hours. After 24 hours, the cells were rinsed twice with PBS and fixed 15 min at room temperature with 4% PFA/PBS.
Whole Mount In Situ Hybridization (WISH)
[0133] WISH was performed as previously described (Kuss et al., 2009; Villavicencio-Lorini et al., 2010).
Cell Treatments
[0134] Transfection: For transient transfection and live cell imaging, HEK-293T cells were seeded onto chambered coverslips (Ibidi, 80826-90), and transiently transfected 20-24 h later using lipoD293 (Signagen, SL100668) or FuGENE HD (Promega) according to the manufacturer's protocol.
[0135] 1,6-hexanediol treatment: Isolated limb bud cells attached to fibronectin coated glass coverslips were treated with 6% or 0% hexanediol (Sigma, 240117) in 1 mL cell culture media for 1 minute at 37° C. After treatment, cells were washed with 1 mL of PBS, and immediately fixed in 4% paraformaldehyde in PBS for 10 min. Fixed cells were washed and then stored at 4° C. until processing for immunofluorescence and microscopy.
Western Blot
[0136] Cells were washed once with ice-cold PBS, and lysed in RIPA buffer (Thermo, 88900) supplemented with protease inhibitors (Thermo, 87786). Protein concentration was determined using bicinchoninic acid assay (Thermo, 23225), according to the manufacturer's protocol. Lysates containing equal protein amounts were then heated to 95° C. for 10 minutes, and separated on 4-12% Tris-acetate gels (Invitrogen, NP0322BOX). Protein was transferred onto a nitrocellulose membrane. After transfer, the membrane was blocked. The membranes were incubated with 1:750 anti-Hoxd13 (abcam ab229234) and 1:3000 anti-HSP90 (BD Biosciences, 610419) antibody diluted in 5% non-fat milk in TBST overnight at 4° C. with shaking. The next day, membranes were washed, incubated with fluorescent anti-mouse (IRDye 800CW Donkey anti-Mouse, Li-Cor P/N 925-32212) and anti-rabbit (IRDye 680LT Donkey anti-Rabbit, Li-Cor P/N 925-68023) secondary antibodies at 1:10000 dilution according to the manufacturer's protocol, and finally washed in the dark. Membranes were imaged on a LICOR Odyssey Clx imager.
Immunofluorescence
[0137] Isolated limb bud cells, Kelly cells, and SH-SY5Y cells attached to coated glass coverslips were fixed in 4% paraformaldehyde (PFA) (Sigma-Aldrich, P6148) in PBS for 10 min and stored at 4° C. in PBS or processed immediately. Cells were permeabilized in 0.1% Triton X-100 (Thermo Scientific, 85111) in PBS for 10 min at RT. Then, after washing and blocking, cells were incubated with the primary antibody (anti-Hoxd13 Invitrogen PA5-66661 1:250 dilution or anti-Hoxd13 abcam ab19866 1:150 dilution) in blocking solution at 4° C. overnight. After washing, cells were incubated with the secondary antibody (donkey anti-Rabbit IgG Alexa Fluor 568 Invitrogen A10042 1:1000 dilution) in blocking solution in the dark. Nuclei were counterstained with 0.24 mg/mL 4′,6-diamidino-2-phenylindole (DAPI) in PBS in the dark. Coverslips were mounted onto slides with Vectashield (Vector, H-1000) and sealed. Images were acquired at the confocal laser-scanning microscope (Zeiss LSM880, 63×oil objective, NA 1.4, 1 Airy Unit). Raw images (.czi files) were processed in FIJI. The abcam ab19866 antibody was used in
Stochastic Optical Reconstruction Microscopy (STORM)
[0138] Isolated E12.5 limb bud cells attached to fibronectin coated glass coverslips were fixed in 4% paraformaldehyde (PFA) (Sigma-Aldrich, P6148). The fixed samples were washed, treated with permeabilization solution (PBS supplemented with 0.1% Triton X-100) and treated with blocking solution (permeabilization solution supplemented with 10% fetal bovine serum). After blocking, samples were incubated with primary anti-Hoxd13 antibody (Thermo, PA5-66661, 1:250 in blocking solution), BRD4 antibody (Clone A-7, Santa Cruz, sc-518021, 1:250 in blocking solution), or HP1α antibody (clone 15.19s2, Millipore/Merck, 1:250 in blocking solution), and then washed. Stained and washed samples were then incubated with secondary antibody (goat anti-Rabbit IgG Alexa Fluor 647 Invitrogen A32733/goat anti-Rabbit IgG Cy3 Invitrogen A21235, 1:1000 in blocking buffer for Hoxd13 primary and goat anti-Mouse IgG Alexa Fluor 647 Invitrogen A10520, 1:1000 in blocking buffer for BRD4/HP1α primary), after which samples were washed. For imaging, samples were placed in a one-well magnetic chamber, covered in switching buffer consisting of 0.15 M 2-mercaptoethanol/0.2 M Tris, pH 8.0 with 4% (w/v) glucose, 0.25 mg/ml glucose oxidase, 20 μg/ml catalase and 5 nM Sytox Orange (Thermo, S11368). Images were acquired with a Vutara 352 super resolution microscope (Bruker) equipped with a Hamamatsu ORCA Flash4.0 sCMOS for super resolution imaging and a 60×oil immersion TIRF objective with numerical aperture 1.49 (Olympus). Data were acquired with TIRF/HILO illumination at a laser-power density of 62.5 kW/cm.sup.2 using a 532 and 639 nm laser.
STORM Image Analysis
[0139] STORM images were analyzed using Vutara SRX software (Bruker) or the Picasso software package (Schnitzbauer et al., 2017).
[0140] For Hoxd13 localization and cluster analysis (
[0141] For Co-IF STORM analysis (
Live Cell Imaging
[0142] All live cell imaging experiments were performed on an LSM880 confocal microscope (Zeiss) equipped with an incubation chamber with a heated stage at 37° C. Images were acquired with either a Plan-Apochromat 40×0.95 Korr M27 or a 63×1.40 oil DIC objective. OptoDroplet assay: The optolDR assay was adapted from (Shin et al., 2017). Briefly, approximately 20,000 cells were seeded per well on chambered coverslips one day before transfection. The following day, cells were transfected with 200 ng of the respective optolDR construct. 48 hours post-transfection, culture medium was refreshed, and cells were imaged on a Zeiss LSM 880 confocal microscope. Droplet formation was induced with scans with the 488 nm laser every 20 seconds for the duration of imaging (
[0143] For FRAP experiments of light induced droplets (
[0144] LacO-LacI assays (
Optodroplet Assay Analysis
[0145] All data analysis on images was performed in Zen Blue 2.5 (Zeiss) Arivis Vision4D or ImageJ. For optoDroplet analysis in Zen Blue 2.5, nuclear and cytoplasmic mCherry signals were automatically thresholded and size filtered to define primary regions of interest. Within these primary regions, optoDroplets were detected using a second fixed pixel intensity threshold. For optoDroplet analysis in Arivis Vision4D, images were first converted to an appropriate file format as multiple scope files in Arivis SIS converter. Nuclear mCherry signals were automatically thresholded and size-filtered to define primary regions of interest. OptoDroplets were detected as secondary regions of interest. The pipeline was then used to analyze at least 50-200 cells per genotype (indicated by in each figure panel). The phase-shifted fraction was calculated as the total area of detected optoDroplets within a nucleus divided by the area of the corresponding cell's nucleus (
[0146] For phase-shifted fraction comparison in the Hoxd13 deletion series in
[0147] For FRAP analysis, mean pixel intensity of regions of interests were measured using ImageJ, and normalized to pre-bleaching intensity. Captured intensity traces were averaged over multiple replicates for each genotype (indicated as n in figure panels), and values were plotted against time.
Confocal/Fluorescence Image Analysis
[0148] Hoxd13 puncta detected in fixed cell immunofluorescence (
In Vitro Droplet Assay Analysis
[0149] For image analysis, droplets were first automatically thresholded in either the mEGFP or mCherry channel using Zen Blue (Version 3.0), and then filtered by size, pixel intensity, and pixel deviation. Mean fluorescence intensities, area, and diameter of these primary regions were measured on both channels. For quantification and data visualization (
LacO-LacI Tethering Image Analysis
[0150] For LacO-LacI image analysis (
Statistical Analysis
[0151] Statistical analyses of microscopy image quantifications were carried out in Graphpad Prism and R. Overlap analysis of super-resolution images was carried out using Vutara SRX analysis software.
Protein Purification and In Vitro Droplet Formation Assays
[0152] For protein expression, plasmids were transformed into BL21(DE3) (NEB M0491S) cells, and grown in automatic-induction medium (AIM), as described in (Studier, 2005). For protein purification from the cultivated cells, pellets were resuspended in 30 mL of Buffer A (50 mM Tris pH 7.5, 300 mM NaCl) supplemented with 8 M Urea and cOmplete protease inhibitors (Sigma, 11697498001). This denatured suspension was sonicated and clarified by centrifugation. Supernatants containing fusion proteins were loaded onto a His GraviTrap column (GE HealthCare, 11003399) pre-equilibrated in Buffer A supplemented with 8 M Urea. The loaded column was washed with 6% Buffer B (50 mM Tris pH 7.5, 300 mM NaCl, 500 mM imidazole) in Buffer A supplemented with 8 M Urea, and 6% Buffer B. Proteins were eluted in 50% Buffer B, immediately diluted 1:2 in storage buffer (50 mM Tris pH 7.5, 125 mM NaCl, 1 mM DTT, 10% Glycerol), and then concentrated. The resulting fraction was then diluted 1:1000 in storage buffer, re-concentrated, and then stored at −20° C. All fusion proteins were purified in the same manner.
[0153] For in vitro droplet formation, recombinant mCherry or mEGFP fusion proteins were measured for concentration, and then diluted or mixed to suitable final concentrations in storage buffer. For microscopy, these solutions were further diluted 1:2 in either 20% or 10% PEG-8000 in deionized water (w/v). 10 μL of this suspension was pipetted onto chambered coverslips, and immediately imaged using a LSM880 confocal microscope equipped with a 63×1.40 oil DIC objective and AiryScan detector. All images were acquired from within the solution interface, and performed before droplets settled onto the bottom of the coverslip, as described in (Sabari et al., 2018). For droplet assays using preformed GFP-Med1 condensates (
Luciferase Reporter Assays
[0154] The murine proximal enhancer of Aldh1a2 (Kuss et al., 2009) was cloned into the pGL2-enhancer luciferase reporter vector and co-transfected into Cos7 cells with pcDNA3.1 (wild type or mutant Hoxd13) expression plasmids and pRL-CMV (Promega) using FuGENE HD (Promega) according to the manufacturer's instructions. 24 hours post transfection, reporter activity in 2.5 μl of cell lysate was measured using the Dual-Glo Luciferase Assay System (Promega) according to the manufacturer's instructions (
[0155] Measurements were performed similarly in Gal4-DBD luciferase assays (
H3K27Ac ChIP-Seq with S2 Spike-In Control (ChIP-RX)
[0156] Isolated E12.5 limb bud cells (hand plate) from 2 wild-type, 2 homozygous spdh, and 2 heterozygous spdh mice (approximately 3.5 million cells each) across 2 separate het×het mouse crosses were fixed in 1% formaldehyde on ice and quenched with 2.5 M glycine. Each pool was spiked in with 1 million fixed fly S2 cells (Orlando et al., 2014), after which cell suspensions were lysed in lysis buffer (50 mM HEPES, 140 mM NaCl, 1 mM EDTA, 10% Glycerin, 0.5% NP-40, 0.25 Triton X-100, pH 7.5 supplemented with Protease inhibitors and Na-Butyrate). After shearing, 10-15 μg chromatin was incubated with 4 μL of H3K27Ac antibody (Diagenode C15410174) in a total of 1.2 mL of buffer (10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 0.1% Na-DOC, 0.5% NLaroylsarcosine, pH 8.0 supplemented with Protease inhibitors and Na-Butyrate) overnight at 4° C. After incubation, 30 μL of Protein G beads were added to the chromatin and antibody suspensions, and allowed to incubate overnight at 4° C. Samples were then washed with 1 mL of RIPA buffer (to 50 mM HEPES-KOH, 1 mM EDTA, 1% NP-40, 0.7% Na-DOC, 500 mM LiCl, pH 7.55 supplemented with Protease inhibitors and Na-Butyrate). Beads were then washed with TE buffer (supplemented with protease inhibitors and Na-butyrate) and then centrifuged down to remove TE buffer. Chromatin was then eluted using 210 μL of elution buffer (50 mM Tris-HCl, 10 mM EDTA, 1% SDS, pH 8.0) and heated to 65° C. Eluates were then treated with 5 μL of Proteinase K overnight at 65° C. The next day, 4 μL of RNAse A was added to the samples, vortexed, spun down and incubated at 37° C. for 30 minutes. Chromatin was then extracted using phenol-chloroform, and precipitated using 70% ice cold EtOH, with centrifugation at max speed for 10 minutes. Supernatant was removed, and the resulting pellet was dissolved in deionized water. Total yield was assessed by Qubit, and then sent for sequencing. Respective ChIP samples and input controls were paired-end sequenced using Illumina high-throughput sequencing, to a depth of around 25 million reads.
[0157] Cluster Identification To identify cell populations in the wild type scRNA-Seq data the following Seurat functions were used: FindVariableGenes, RunPCA, RunTSNE and FindClusters (Butler et al., 2018). Clusters were identified by first building a shared nearest neighbor graph and then running the Louvain algorithm on it (Blondel et al., 2008).
Assignment of Cell Types to Clusters
[0158] The marker genes in the 11 clusters of the wild type reference map were identified by running the FindAllMarkers functions of the Seurat package (Butler et al., 2018). The clusters were then assigned to cell types using a GO term enrichment analysis, inspection of localized gene expression data from whole mount in situ hybridization [e.g. using the Gene eXpression Database (GXD)], and literature data on marker genes of cell types in the developing limb.
[0159] Cluster 1: Proliferating cells (S phase) of the distal proliferating mesenchyme; Cluster 2: Proliferating cells (Cytokinesis) of the distal proliferating mesenchyme; Cluster 3: Perichondrium (Proliferating mesenchyme); Cluster 4: Interdigital mesenchyme; Cluster 5: Perichondrium; Cluster 6: Proximal chondrocytes (carpal progenitors); Cluster 7: Distal chondrocytes (phalanx progenitors); Cluster 8: Proliferating cells; Cluster 9: Myoblasts; Cluster 10: Hematopoietic cells; Cluster 11. Erythrocytes.
Quantification of Cell State Proportions
[0160] Spdh cells were assigned to the identified 11 clusters of the wild type reference cell state map using the nearest neighbor method. Spdh cells were assigned to the wild type cluster that had the most similar mean expression profile. For each cluster and condition we normalized the cell counts by dividing by the total number of wild type or spdh cells used. Last, we determined the log 2 ratio of these normalized cell state percentages in the spdh and wild type samples per cluster (
Differential Expression Analysis
[0161] Differential expression analysis was performed within clusters between wild type and assigned spdh cells by using the FindMarkers function of the Seurat package. To display differentially expressed genes in
Gene Ontology (GO) Term Enrichment Analysis
[0162] Gene Ontology (GO) term enrichment was performed on differentially expressed genes per individual cluster using the enrichGO function of the clusterProfiler R package (version 3.10.1) (Yu et al., 2012). The cut-off value for significance was set at q<0.05. Redundant terms were removed from the graphical display of the results, and terms displayed in
ChIP-Seq Data (Previously Published)
[0163] The Hoxd13 ChIP-Seq data was downloaded from GEO (accession number GSE81358), and was described in a previous study (Sheth et al., 2016). Previously published murine FLAGHoxd13 and FLAG-Hoxd13 Q317R ChIP-Seq data in chicken micromass were obtained from GEO (accession number GSE44799).
ChIP-Seq Data Processing
[0164] The Hoxd13 ChIP-Seq data was aligned to the mm9 genome assembly using bowtie version 1.0.0 (Langmead et al., 2009) with parameters ‘-n 2 -e 70 -m 1 -k 1 --best -l 200’.
[0165] The murine FLAG-Hoxd13 transfected into chicken micromass ChIP-Seq datasets were aligned to the chicken genome (assembly gg3). The paired-end sequencing reads of the murine H3K27Ac ChIP-Seq data with D. melanogaster spike-ins were first adapter- and quality trimmed Then reads were aligned to the mouse genome mm9 or fly (D. melanogaster) genome.
[0166] Genome wide rpm/bp normalized coverage profiles were created by binning the mouse genome into 3 mb regions and subsequently using bamToGFF_turbo.py resulting in 50 bp sized bins. Hoxd13 and spdh rpm/bp values for each bin were finally normalized by spike-in factors obtained from the D. melanogaster spike-in ChIP-Seq reads (Orlando et al., 2014).
H3K27Ac ChIP-Seq analysis
[0167] In
Capture C Data
[0168] Capture C profiles for hindlimb and midbrain tissues from mouse embryos at the developmental stage E11.5 were obtained from GEO (accession number GSE84795).
Topologically Associating Domains (TADs)
[0169] Topologically associating domains (TADs) in mouse E12.5 limb bud cells were described in a previous study (Kraft et al., 2019).
Enrichment of Hoxd13 ChIP-Seq Peaks Around Cluster 4 Dysregulated Genes
[0170] The number of Hoxd13 ChIP-Seq peaks within TADs that contain a Cluster 4 dysregulated gene was calculated as follows. First, TADs that contain the transcription start site (TSS) of at least one gene dysregulated in Cluster 4 (
Enrichment of Capture C Signal at Hoxd13 Peaks
[0171] The mean contact frequency between Cluster 4 dysregulated genes and Hoxd13 ChIP-Seq peaks (
Mean Expression Value of Individual Genes
[0172] For genes depicted in
Repeat Expansions in Hoxd13, Hoxa13, Runx2 and TBP
[0173] The repeat expansion information displayed in
IDR Classification: Data Retrieval and Preprocessing
[0174] Protein sequences for all human RefSeq gene models were downloaded from UCSC (Karolchik et al., 2004). For each protein, the PONDR (disorder) scores (VSL2) were downloaded from PONDR (Peng et al., 2006). An intrinsically disordered region (IDR) was defined as a sequence of at least 50 consecutive amino acids with a PONDR (VSL2) score above 0.5. Two IDRs in the same protein were fused if less than 40 consecutive amino acids with a score below 0.5 separated them. For the classification of TF IDRs (
IDR/DBD Clustering and Annotation
[0175] The amino acid composition, aliphatic index, isoelectric point (pI), PONDR, GRAVY and instability scores for each IDR of all TFs was used as input. Principal component analysis (PCA) was performed to identify the most variable features, and the first ten PCs (that explain 80% of the variability of the input data) were subsequently used. The transformed data were then subject to K-means clustering. We determined the Bayesian information criterion (BIC) for various values of “k” and identified k=7 as an optimal cluster number as the infliction point on the BIC plot (Schwarz, 1978). For each TF, the DBD annotation of a previous study was used (Lambert et al., 2018). The presence of a homopolymeric repeat was defined using a minimum repeat length of ten amino acids in the IDR (
[0176]
[0177] The enrichment analyses of TFs whose IDRs comprise the 7 IDR clusters for categories displayed in
3. Results
3.1 the Hoxd13 IDR Drives Phase Separation
[0178] Disease-associated repeat expansions are significantly enriched in transcription factors (TFs) (p<10-4, Fisher's test;
[0179] Intrinsically disordered regions (IDRs) in proteins are known to drive phase separation (Banani et al., 2017; Shin and Brangwynne, 2017), and the Hoxd13 N-terminus has sequence features predictive of an IDR (
[0180] If the Hoxd13 IDR can drive phase separation, the purified IDR should form liquid-like droplets in vitro. Purified recombinant Hoxd13 IDR-mCherry fusion protein indeed formed droplets in the presence of 10% PEG-8000, while an mCherry control did not (
3.2 Synpolydactyly-Associated Repeat Expansions Enhance Hoxd13 IDR Phase Separation
[0181] Expansions of an alanine repeat in the IDR of Hoxd13 cause synpolydactyly, a congenital limb malformation (
[0182] Phase separated condensates deep beyond the phase boundary can transition to a gel-like state characterized by arrested, yet reversible dynamics (Shin et al., 2017). Consistent with this notion, the light-induced droplets formed by repeat-expanded Hoxd13 IDRs exhibited considerably slower FRAP rates than droplets formed by the wild type IDR (
[0183] To further probe the effect of alanine repeat expansion on Hoxd13 phase separation, we purified various recombinant Hoxd13 IDR-mCherry fusion proteins, and investigated their phase separation capacity in droplet formation buffer in vitro. The +7A and +10A IDR mutants formed more, and more concentrated droplets compared to the wt IDR at similar concentrations (
3.3 Synpolydactyly-Associated Repeat Expansions Alter the Composition of Hoxd13-Containing Condensates
[0184] Recent studies indicate that TFs can form heterotypic condensates with the IDR of the Med1 subunit of the Mediator co-activator in vitro, and that co-condensation of TFs with the Med1 IDR requires the TFs' IDR (Boija et al., 2018). We thus hypothesized that the alanine repeat expansion of the Hoxd13 IDR alters its ability to co-condense with transcriptional co-activators. To test this model, we purified recombinant Med1 IDR-GFP fusion protein and mixed it with various purified Hoxd13 IDR-mCherry fusion proteins. Of note, the condensation behavior of the Med1 IDR is similar to that of purified partial Mediator complex, and the Med1 IDR has thus been a useful surrogate for Mediator condensates in vitro (Boija et al., 2018; Guo et al., 2019; Sabari et al., 2018). We found that heterotypic Med1 IDR-Hoxd13 IDR co-condensates had dramatically altered composition when Hoxd13 IDRs containing synpolydactyly-associated repeat expansions (+7A, +10A) were used in the mixing experiments. The Med1 IDR droplets incorporated the wild type Hoxd13 IDR at several concentrations, and the wild type Hoxd13 IDR exclusively occurred in co-condensates with the Med1 IDR (
[0185] We next tested whether various Hoxd13 IDRs unblend from the Med1 IDR even at concentrations at which they form comparable homotypic droplets alone. To this end, we preassembled Med1 IDR-GFP droplets for 30 minutes, and mixed the preassembled Med1 IDR-GFP droplets with various concentrations of Hoxd13 IDR-mCherry fusion proteins. These included concentrations where the Hoxd13 proteins alone formed droplets similar in number, size, and protein content (wt: 5 μM, +7A: 1 μM, +10A: 0.2 μM; see
[0186] The unblending of repeat-expanded Hoxd13 IDRs from Med1 IDR-containing co-condensates indicates that the alanine repeat expansion increases the preference for homotypic Hoxd13 IDR-IDR interactions over heterotypic interactions with the Med1 IDR. If this model is true, disrupting weak hydrophobic interactions could revert the composition of repeat-expanded Hoxd13 IDR-containing Med1 co-condensates to be more similar to the composition of Hoxd13 wt IDR-Med1 IDR co-condensates (i.e. “reblend” them). To test this notion, we treated Hoxd13 IDR-Med1 IDR co-condensates with ATP, a small hydrotropic molecule known to solubilize hydrophobic molecules at mM concentrations (Patel et al., 2017). ATP reblended Hoxd13 +7A IDR-Med1 IDR co-condensates in a dose dependent manner (
[0187] In a further experiment, it was shown that APPNP (adenosine-5′-[(β,γ)-imido]triphosphate) a non-hydrolyzable analog of ATP was capable of reblending by reverting the composition of co-condensates formed by repeat-expanded Hoxd13 with Med1 IDR in vitro (
3.4 Altered Composition and Properties of Repeat-Expanded Hoxd13-Condensates In Vivo
[0188] Next we sought to investigate the effect of disease-associated repeat expansions on endogenous Hoxd13-containing condensates in vivo. To this end, we isolated limb bud cells from homozygous spdh mouse embryos and wild type littermates. Spdh mice carry +7A repeat expanded Hoxd13 alleles, and homozygous spdh mice exhibit synpolydactyly (Bruneau et al., 2001). We first tested whether transcriptional co-activators are associated with Hoxd13-containing condensates in limb bud cells using antibody staining and STORM imaging. As no high quality Mediator antibodies were available to co-stain with Hoxd13 in STORM, Brd4 was visualized instead of Mediator. Brd4 is a ubiquitous co-activator that co-purifies with Mediator (Jiang et al., 1998), co-condenses with Mediator (Sabari et al., 2018), and Brd4 chemical inhibition dissolves Mediator condensates in vivo (Cho et al., 2018). We found that Brd4 more frequently overlapped Hoxd13 puncta in wild type limb bud cells than in spdh limb bud cells harboring Hoxd13 +7A alleles [1.5-fold difference in Manders co-efficient, p-value<0.001, twotailed t-test)] (
[0189] To investigate the biophysical properties of Hoxd13-containing condensates in vivo, wt and spdh limb bud cells were treated with 1,6-hexanediol (1,6-HD), a short-chain aliphatic alcohol that dissolves various intracellular membraneless organelles (Boehning et al., 2018; Chong et al., 2018; Sabari et al., 2018). Hoxd13 was subsequently visualized with immunofluorescence. Both wt and +7A Hoxd13 localized within discrete nuclear puncta (
[0190] To test whether repeat-expansion impairs the ability of Hoxd13 condensates to recruit Mediator in vivo, various Hoxd13 IDRs were tethered to a LacO array in U2OS cells expressing an ectopic Med1 IDR-YFP fusion protein (
[0191] The Mediator co-activator plays key roles in recruiting RNA Polymerase II to TF-bound genes (Levine et al., 2014), so a reduction of Med1 content in mutant Hoxd13-containing condensates would be expected to reduce the transcriptional activity of repeat-expanded Hoxd13. Consistent with this idea, repeat-expanded Hoxd13 alleles displayed reduced activity in luciferase reporter assays (
[0192] To rule out that the repeat expansion affects DNA binding of Hoxd13, we performed Chromatin Immunoprecipitation followed by sequencing (ChIP-Seq) on FLAG-tagged Hoxd13 proteins expressed in a chicken transgenic cell system (Ibrahim et al., 2013).
[0193] Taken together, these results suggest that repeat expansions alter the composition and biophysical properties of Hoxd13-containing condensates in disease-relevant limb bud cells and transgenic cell systems, and reduce Hoxd13-dependent transcriptional activity without affecting DNA binding.
3.5 Hoxd13 Repeat Expansion Alters the Transcriptional Program of Several Cell Types in a Cell-Specific Manner
[0194] The condensate unblending model predicts that the Hoxd13 repeat expansion alters the composition and properties of Hoxd13-containing transcriptional condensates, which leads to deregulated gene expression programs. To comprehensively assess the impact of Hoxd13 repeat expansion on gene expression in disease-relevant cells, we performed single cell RNASeq (scRNA-Seq) on limb buds from mouse embryos that carry either wild type or +7A repeat-expanded Hoxd13 alleles (spdh mice). In total, 9,655 single cell transcriptomes were captured, with on average >60,000 transcripts and .sup.˜3,500 genes detected per cell. The single cell transcriptomes of 4,464 wild type cells were clustered to generate a reference map of cell states in the limb bud (
[0195] Examination of dysregulated genes within individual cell states revealed cell type-specific changes in the transcriptional program of several cell types in the spdh limb. For example, genes associated with mesenchyme differentiation (e.g. Msx1, Msx2, Tgfb2) and digit morphogenesis (e.g. Hoxd12, Hoxd13) were downregulated in the spdh interdigital mesenchymal cells (Cluster 4; whose abundance was lower in the spdh limbs) (
3.6 Disease-Associated Repeat Expansions Alter the Phase Separation Capacity of Other TFs
[0196] Amino acid repeat expansions in TFs occur in various diseases (
[0197] Hoxa13 is a homeobox TF involved in vertebrate limb- and urogenital tract development (Albrecht and Mundlos, 2005), and alanine repeat expansions in the Hoxa13 IDR cause hand-foot-genital syndrome (HFGS) (
[0198] Runx2 is a RUNT family TF that controls bone morphogenesis, and expansions of a short alanine and glutamine repeat in the Runx2 IDR are associated with cleidocranial dysplasia (CCD), a disorder of severe skeletal defects (
[0199] The TATA-box binding factor TBP is a highly conserved general transcription factor that plays a key role in transcription initiation. The TBP N-terminus contains a polymorphic polyglutamine (Q) repeat, whose typical size ranges between 24-42 glutamines, and repeats consisting of >46 glutamines are linked to spinocerebellar ataxia type 17 (SCA17), a progressive neurodegenerative disease (Nakamura et al., 2001). A TBP IDR with wild type glutamine repeat length (38Q) facilitated phase separation over the Cry2 control in the optoDroplet system (
3.7 A Catalog of Human Transcription Factor IDRs
[0200] Transcription factors (TFs) typically consist of highly structured DNA binding domains (DBD) and intrinsically disordered activation domains (ADs/IDRs) (Lambert et al., 2018; Mitchell and 10 Tjian, 1989; Staby et al., 2017). TF IDRs are low complexity protein sequences, several of which are known to drive phase separation (Boija et al., 2018; Chong et al., 2018). TF IDRs frequently contain homopolymeric repeats (proline, serine, alanine, glycine, glutamine, histidine), but only expansions of alanine and glutamine repeats have been linked to human pathologies to date. We thus reasoned that various molecular features of TF IDRs dictate their phase separation capacity. To gain insights into those features, we created a catalog of IDRs in .sup.˜1,500 human TFs. We first identified IDRs and DBDs in human TFs, and clustered them based on amino acid composition, hydrophobicity, aliphatic index, stability, isoelectric point and disorder. The clustering algorithm separated DBDs and IDRs with 96% efficiency, and could even separate DBD families previously annotated based on structural homology (Lambert et al., 2018). The TF IDRs were broadly distributed in seven major clusters based on various features (
[0201] One key feature of cluster 1 IDRs was high alanine content, and this cluster included 32 of the 33 poly-alanine-containing IDRs (p<10-16, Fisher's exact test) (
4. Discussion
[0202] We found that repeat expansions in intrinsically disordered regions (IDRs) of transcription factors (TFs) alter the phase separation capacity of those TFs, and their capacity to form transcriptional condensates with components of the transcription machinery. The IDR of the HOXD13 TF drives phase separation and co-condensation with the Mediator co-activator. Synpolydactyly-associated alanine repeat expansions facilitate homotypic Hoxd13 phase separation, and the mutant Hoxd13 IDRs co-condense less readily with Mediator than the wild type HOXD13 IDR, a phenomenon we term “condensate unblending”. In a mouse synpolydactyly model, HOXD13 repeat expansion led to cell-type specific transcriptional changes of HOXD13 targets in disease-relevant cells. Thus, disease-associated mutations in TF IDRs alter the features that drive the TF's condensation behavior, and present a molecular classification of TF IDRs as a framework to dissect the pathomechanism of diseases associated with transcription factor dysfunction. The results presented here support a model that disease-associated amino acid repeat expansions in TF IDRs alter the TFs' phase separation capacity, and their ability to co-condense with transcriptional coactivators. For the Hoxd13 TF, synpolydactyly-associated alanine repeat expansions were found to enhance the phase separation capacity of the Hoxd13 IDR, and the mutant Hoxd13 IDR was able to co-condense with dramatically less Mediator than the wild type IDR, a phenomenon we term “condensate unblending” (
[0203] Several lines of evidence indicate that altered phase separation underpins the effect of Hoxd13 repeat expansions. The saturation concentration Csat (at which condensates are observed) is lower for purified mutant Hoxd13 IDRs than the wt IDR (
[0204] Repeat expansion diseases include severe, incurable neurodegenerative and developmental diseases typically associated with the presence of large protein aggregates (Albrecht and Mundlos, 2005; Orr and Zoghbi, 2007; Zoghbi and Orr, 2000). The condensate model presented here may explain several observations of the pathology of repeat expansion diseases that are less readily explained by toxicity of aggregates. For example, for poly-alanine expansions that occur in TFs, aggregates or nuclear inclusions were only described in overexpression systems but not in primary tissue to date (Albrecht and Mundlos, 2005; Albrecht et al., 2004; Villavicencio-Lorini et al., 2010). Furthermore, the short +7A repeat expansion of Hoxd13 ecapitulates the human synpolydactyly phenotype in mice, yet without any evidence of protein aggregation in the limb bud (Albrecht et al., 2004; Villavicencio-Lorini et al., 2010). For polyglutamine expansion diseases, e.g. Huntington's disease, previous studies have established a correlation between aggregate formation by the mutant huntingtin (htt) protein and degeneration of specific neurons (Davies et al., 1997; Ross and Poirier, 2004). However, manipulation of the Ubiquitin-proteasome pathway was reported to decrease aggregate formation without decreasing toxicity of the repeat-expanded htt protein in primary cell models, suggesting that the soluble fraction of the protein may be responsible for its cytotoxic effects (Saudou et al., 1998; Truant et al., 2008). Changes in the phase separation capacity, and miscibility of the repeat expanded proteins in heterotypic condensates are consistent with these diverse observations.
[0205] Disease-associated repeat expansions occur not just in transcription factors but in other nuclear proteins e.g. the ataxin genes, several of which have been linked to chromatin regulation and transcription (Albrecht and Mundlos, 2005; Darling and Uversky, 2017; La Spada and Taylor, 2010; Orr and Zoghbi, 2007). It is thus plausible that disease-associated changes may alter the phase separation capacity and miscibility of various nuclear proteins within heterotypic condensates, which in turn perturbs cellular gene expression programs. Mutations in the IDR of cellular proteins, and repeat expansions in RNA have indeed been recently linked to altered phase separation (Jain and Vale, 2017; Meyer et al., 2018; Molliex et al., 2015; Patel et al., 2015). Dysregulated phase separation may thus underlie a wide spectrum of diseases.
[0206] The results presented here also provide insights into the function of TF activation domains (ADs). TF ADs are typically intrinsically disordered sequences that interact with components of the transcription machinery, and recent studies suggest that various TF ADs (i.e. IDRs) have the capacity to phase separate and to co-condense with the Mediator co-activator in vitro, though the functional importance of these observation has been unclear (Boija et al., 2018; Chong et al., 2018). The results presented here provide evidence for the importance of biochemical characteristics that drive TF phase separation for normal TF function in vivo. Eukaryotic genomes tend to encode hundreds of TFs, whose ADs have limited sequence-level similarity, but many ADs are nevertheless functionally interchangeable (Lambert et al., 2018; Mitchell and Tjian, 1989; Staby et al., 2017; Stampfel et al., 2015). A key outstanding question is whether the interchangeability of TF IDRs is determined by their phase separation capacity, and whether the phase separation capacity of interchangeable TF IDRs has a shared molecular basis. The phase separation of the IDR of the FUS RNA-binding protein, for example, is predominantly driven by cation-pi interactions between tyrosine and arginine residues (Wang et al., 2018), while the phase separation capacity of the Oct4 and Gcn4 TFs' IDRs is predominantly driven by negatively charged residues (Boija et al., 2018). Our draft catalog of human TF IDRs indicates that at least for a subset of TFs, e.g. Hoxd13, the phase separation capacity may be determined by alanine residues, and that different TF IDR clusters correlate with various biological functions and phenotypic effects. Future work into the molecular basis of TF phase separation, and how developmental signaling, post-translational modifications and genetic variants impact the miscibility of TFs in heterotypic condensates with components of the transcriptional machinery may thus open new condensate-directed therapeutic avenues for diseases associated with transcriptional dysregulation.
5. References
[0207] Alberti, S., Gladfelter, A., and Mittag, T. (2019). Considerations and Challenges in Studying Liquid-Liquid Phase Separation and Biomolecular Condensates. Cell 176, 419-434. [0208] Albrecht, A., and Mundlos, S. (2005). The other trinucleotide repeat: polyalanine expansion disorders. Curr Opin Genet Dev 15, 285-293. [0209] Albrecht, A.N., Kornak, U., Boddrich, A., Suring, K., Robinson, P. N., Stiege, A. C., Lurz, R., Stricker, S., Wanker, E. E., and Mundlos, S. (2004). A molecular pathogenesis for transcription factor associated polyalanine tract expansions. Human molecular genetics 13, 2351-2359. [0210] Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., Duvaud, S., Flegel, V., Fortier, A., Gasteiger, E., et al. (2012). ExPASy: SIB bioinformatics resource portal. Nucleic acids research 40, W597-603. [0211] Banani, S. F., Lee, H. O., Hyman, A. A., and Rosen, M. K. (2017). Biomolecular condensates: organizers of cellular biochemistry. Nature reviews Molecular cell biology. [0212] Bi, W., Deng, J. M., Zhang, Z., Behringer, R. R., and de Crombrugghe, B. (1999). Sox9 is required for cartilage formation. Nature genetics 22, 85-89. [0213] Blondel, V. D., Guillaume, J. L., Lambiotte, R., and Lefebvre, E. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics. [0214] Boehning, M., Dugast-Darzacq, C., Rankovic, M., Hansen, A. S., Yu, T., Marie-Nelly, H., McSwiggen, [0215] D. T., Kokic, G., Dailey, G. M., Cramer, P., et al. (2018). RNA polymerase II clustering through carboxyterminal domain phase separation. Nature structural & molecular biology 25, 833-840. [0216] Boija, A., Klein, I. A., Sabari, B. R., Dall'Agnese, A., Coffey, E. L., Zamudio, A. V., Li, C. H., Shrinivas, K., Manteiga, J. C., Hannett, N. M., et al. (2018). Transcription Factors Activate Genes through the Phase-Separation Capacity of Their Activation Domains. Cell 175, 1842-1855 e1816. [0217] Bruneau, S., Johnson, K. R., Yamamoto, M., Kuroiwa, A., and Duboule, D. (2001). The mouse Hoxd13(spdh) mutation, a polyalanine expansion similar to human type II synpolydactyly (SPD), disrupts the function but not the expression of other Hoxd genes. Developmental biology 237, 345-353. [0218] Buniello, A., MacArthur, J. A. L., Cerezo, M., Harris, L. W., Hayhurst, J., Malangone, C., McMahon, A., Morales, J., Mountjoy, E., Sollis, E., et al. (2019). The NHGRI-EBI GWAS Catalog of published genomewide association studies, targeted arrays and summary statistics 2019. Nucleic acids research 47, D1005-D1012. [0219] Butler, A., Hoffman, P., Smibert, P., Papalexi, E., and Satija, R. (2018). Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nature biotechnology 36, 411-420. [0220] Cho, W. K., Spille, J. H., Hecht, M., Lee, C., Li, C., Grube, V., and Cisse, II (2018). Mediator and RNA polymerase II clusters associate in transcription-dependent condensates. Science. [0221] Choi, J. M., Dar, F., and Pappu, R. V. (2019). LASSI: A lattice model for simulating phase transitions of multivalent proteins. PLoS computational biology 15, e1007028. [0222] Chong, S., Dugast-Darzacq, C., Liu, Z., Dong, P., Dailey, G. M., Cattoglio, C., Heckert, A., Banala, S., Lavis, L., Darzacq, X., et al. (2018). Imaging dynamic and selective low-complexity domain interactions that control gene transcription. Science 361. [0223] Darling, A. L., and Uversky, V. N. (2017). Intrinsic Disorder in Proteins with Pathogenic Repeat Expansions. Molecules 22. [0224] Davies, S. W., Turmaine, M., Cozens, B. A., DiFiglia, M., Sharp, A. H., Ross, C. A., Scherzinger, E., Wanker, E. E., Mangiarini, L., and Bates, G. P. (1997). Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell 90, 537-548. [0225] Dolle, P., Dierich, A., LeMeur, M., Schimmang, T., Schuhbaur, B., Chambon, P., and Duboule, D. (1993). Disruption of the Hoxd-13 gene induces localized heterochrony leading to mice with neotenic limbs. Cell 75, 431-441. [0226] Eden, E., Navon, R., Steinfeld, I., Lipson, D., and Yakhini, Z. (2009). GOrilla: a tool for discovery and visualization of enriched GO terms in ranked gene lists. BMC Bioinformatics 10, 48. [0227] Endesfelder, U., Malkusch, S., Fricke, F., and Heilemann, M. (2014). A simple method to estimate the average localization precision of a single-molecule localization microscopy experiment. Histochem Cell Biol 141, 629-638. [0228] Fabricius, V., Lefebre, J., Geertsema, H., Marino, S. F., and Ewers, H. (2018). Rapid and efficient C terminal labeling of nanobodies for DNA-PAINT. Journal of Physics D: Applied Physics. [0229] Galili, T. (2015). dendextend: an R package for visualizing, adjusting and comparing trees of hierarchical clustering. Bioinformatics 31, 3718-3720. [0230] Goodman, F. R., Bacchelli, C., Brady, A. F., Brueton, L. A., Fryns, J. P., Mortlock, D. P., Innis, J. W., Holmes, L. B., Donnenfeld, A. E., Feingold, M., et al. (2000). Novel HOXA13 mutations and the phenotypic spectrum of hand-foot-genital syndrome. American journal of human genetics 67, 197-202. [0231] Goodman, F. R., Mundlos, S., Muragaki, Y., Donnai, D., Giovannucci-Uzielli, M. L., Lapi, E., Majewski, F., McGaughran, J., McKeown, C., Reardon, W., et al. (1997). Synpolydactyly phenotypes correlate with size of expansions in HOXD13 polyalanine tract. Proceedings of the National Academy of Sciences of the United States of America 94, 7458-7463. [0232] Gu, Z., Gu, L., Eils, R., Schlesner, M., and Brors, B. (2014). circlize Implements and enhances circular visualization in R. Bioinformatics 30, 2811-2812. [0233] Guo, Y. E., Manteiga, J. C., Henninger, J. E., Sabari, B. R., Dall'Agnese, A., Hannett, N. M., Spille, J. H., Afeyan, L. K., Zamudio, A. V., Shrinivas, K., et al. (2019). Pol II phosphorylation regulates a switch between transcriptional and splicing condensates. Nature 572, 543-548. [0234] Hnisz, D., Shrinivas, K., Young, R. A., Chakraborty, A. K., and Sharp, P. A. (2017). A Phase Separation Model for Transcriptional Control. Cell 169, 13-23. [0235] Ibrahim, D. M., Hansen, P., Rodelsperger, C., Stiege, A. C., Doelken, S. C., Horn, D., Jager, M., Janetzki, C., Krawitz, P., Leschik, G., et al. (2013). Distinct global shifts in genomic binding profiles of limb malformation-associated HOXD13 mutations. Genome research 23, 2091-2102. [0236] Innis, J. W., Mortlock, D., Chen, Z., Ludwig, M., Williams, M. E., Williams, T. M., Doyle, C. D., Shao, Z., Glynn, M., Mikulic, D., et al. (2004). Polyalanine expansion in HOXA13: three new affected families and the molecular consequences in a mouse model. Human molecular genetics 13, 2841-2851. [0237] Jain, A., and Vale, R. D. (2017). RNA phase transitions in repeat expansion disorders. Nature 546, 243-247. [0238] Janicki, S. M., Tsukamoto, T., Salghetti, S. E., Tansey, W. P., Sachidanandam, R., Prasanth, K. V., Ried, T., Shav-Tal, Y., Bertrand, E., Singer, R. H., et al. (2004). From silencing to gene expression: real-time analysis in single cells. Cell 116, 683-698. [0239] Jiang, Y. W., Veschambre, P., Erdjument-Bromage, H., Tempst, P., Conaway, J. W., Conaway, R. C., and Kornberg, R. D. (1998). Mammalian mediator of transcriptional regulation and its possible role as an endpoint of signal transduction pathways. Proceedings of the National Academy of Sciences of the United States of America 95, 8538-8543. [0240] Karolchik, D., Hinrichs, A. S., Furey, T. S., Roskin, K. M., Sugnet, C. W., Haussler, D., and Kent, W. J. (2004). The UCSC Table Browser data retrieval tool. Nucleic acids research 32, D493-496. [0241] Kohler, S., Doelken, S. C., Mungall, C. J., Bauer, S., Firth, H. V., Bailleul-Forestier, I., Black, G. C., Brown, D. L., Brudno, M., Campbell, J., et al. (2014). The Human Phenotype Ontology project: linking molecular biology and disease through phenotype data. Nucleic acids research 42, D966-974. [0242] Kraft, K., Magg, A., Heinrich, V., Riemenschneider, C., Schopflin, R., Markowski, J., Ibrahim, D. M., Acuna-Hidalgo, R., Despang, A., Andrey, G., et al. (2019). Serial genomic inversions induce tissue specific architectural stripes, gene misexpression and congenital malformations. Nature cell biology 21, 305-310. [0243] Kuss, P., Villavicencio-Lorini, P., Witte, F., Klose, J., Albrecht, A. N., Seemann, P., Hecht, J., and Mundlos, S. (2009). Mutant Hoxd13 induces extra digits in a mouse model of synpolydactyly directly and by decreasing retinoic acid synthesis. The Journal of clinical investigation 119, 146-156. [0244] Kwon, I., Kato, M., Xiang, S., Wu, L., Theodoropoulos, P., Mirzaei, H., Han, T., Xie, S., Corden, J. L., and McKnight, S. L. (2013). Phosphorylation-regulated binding of RNA polymerase II to fibrous polymers of low-complexity domains. Cell 155, 1049-1060. [0245] La Spada, A. R., and Taylor, J. P. (2010). Repeat expansion disease: progress and puzzles in disease pathogenesis. Nature reviews Genetics 11, 247-258. [0246] Lambert, S. A., Jolma, A., Campitelli, L. F., Das, P. K., Yin, Y., Albu, M., Chen, X., Taipale, J., Hughes, T. R., and Weirauch, M. T. (2018). The Human Transcription Factors. Cell 175, 598-599. [0247] Langmead, B., Trapnell, C., Pop, M., and Salzberg, S. L. (2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome biology 10, R25. [0248] Levine, M., Cattoglio, C., and Tjian, R. (2014). Looping back to leap forward: transcription enters a new era. Cell 157, 13-25. [0249] Li, J., Dong, A., Saydaminova, K., Chang, H., Wang, G., Ochiai, H., Yamamoto, T., and Pertsinidis, A. (2019). Single-Molecule Nanoscopy Elucidates RNA Polymerase II Transcription at Single Genes in Live Cells. Cell 178, 491-506 e428. [0250] Lu, H., Yu, D., Hansen, A. S., Ganguly, S., Liu, R., Heckert, A., Darzacq, X., and Zhou, Q. (2018). Phaseseparation mechanism for C-terminal hyperphosphorylation of RNA polymerase II. Nature 558, 318-323. [0251] Maltecca, F., Filla, A., Castaldo, I., Coppola, G., Fragassi, N. A., Carella, M., Bruni, A., Cocozza, S., Casari, G., Servadio, A., et al. (2003). Intergenerational instability and marked anticipation in SCA-17. Neurology 61, 1441-1443. [0252] Mastushita, M., Kitoh, H., Subasioglu, A., Kurt Colak, F., Dundar, M., Mishima, K., Nishida, Y., and Ishiguro, N. (2015). A Glutamine Repeat Variant of the RUNX2 Gene Causes Cleidocranial Dysplasia. Mol Syndromol 6, 50-53. [0253] Mathelier, A., Zhao, X., Zhang, A. W., Parcy, F., Worsley-Hunt, R., Arenillas, D. J., Buchman, S., Chen, C. Y., Chou, A., Ienasescu, H., et al. (2014). JASPAR 2014: an extensively expanded and updated open access database of transcription factor binding profiles. Nucleic acids research 42, D142-147. [0254] Meyer, K., Kirchner, M., Uyar, B., Cheng, J. Y., Russo, G., Hernandez-Miranda, L. R., Szymborska, A., Zauber, H., Rudolph, I. M., Willnow, T. E., et al. (2018). Mutations in Disordered Regions Can Cause Disease by Creating Dileucine Motifs. Cell 175, 239-253 e217. [0255] Mitchell, P. J., and Tjian, R. (1989). Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245, 371-378. [0256] Molliex, A., Temirov, J., Lee, J., Coughlin, M., Kanagaraj, A. P., Kim, H. J., Mittag, T., and Taylor, J. P. (2015). Phase separation by low complexity domains promotes stress granule assembly and drives pathological fibrillization. Cell 163, 123-133. [0257] Mundlos, S., Otto, F., Mundlos, C., Mulliken, J. B., Aylsworth, A. S., Albright, S., Lindhout, D., Cole, W. G., Henn, W., Knoll, J. H., et al. (1997). Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 89, 773-779. [0258] Muragaki, Y., Mundlos, S., Upton, J., and Olsen, B. R. (1996). Altered growth and branching patterns in synpolydactyly caused by mutations in HOXD13. Science 272, 548-551. [0259] Nakamura, K., Jeong, S. Y., Uchihara, T., Anno, M., Nagashima, K., Nagashima, T., Ikeda, S., Tsuji, S., and Kanazawa, I. (2001). SCA17, a novel autosomal dominant cerebellar ataxia caused by an expanded polyglutamine in TATA-binding protein. Human molecular genetics 10, 1441-1448. [0260] Nihongaki, Y., Yamamoto, S., Kawano, F., Suzuki, H., and Sato, M. (2015). CRISPR-Cas9-based photoactivatable transcription system. Chem Biol 22, 169-174. [0261] Orlando, D. A., Chen, M. W., Brown, V. E., Solanki, S., Choi, Y. J., Olson, E. R., Fritz, C. C., Bradner, J. E., and Guenther, M. G. (2014). Quantitative ChIP-Seq normalization reveals global modulation of the epigenome. Cell reports 9, [0262] Orr, H. T., and Zoghbi, H. Y. (2007). Trinucleotide repeat disorders. Annu Rev Neurosci 30, 575-621. [0263] Patel, A., Lee, H. O., Jawerth, L., Maharana, S., Jahnel, M., Hein, M. Y., Stoynov, S., Mahamid, J., Saha, S., Franzmann, T. M., et al. (2015). A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell 162, 1066-1077. [0264] Patel, A., Malinovska, L., Saha, S., Wang, J., Alberti, S., Krishnan, Y., and Hyman, A. A. (2017). ATP as a biological hydrotrope. Science 356, 753-756. [0265] Peng, K., Radivojac, P., Vucetic, S., Dunker, A. K., and Obradovic, Z. (2006). Length-dependent prediction of protein intrinsic disorder. BMC Bioinformatics 7, 208. [0266] Riback, J. A., Zhu, L., Ferrolino, M. C., Tolbert, M., Mitrea, D. M., Sanders, D. W., Wei, M. T., Kriwacki, R. W., and Brangwynne, C. P. (2019). Composition dependent phase separation underlies directional flux through the nucleolus. BioRxiv, http://dx.doi.org/10.1101/809210. [0267] Ross, C. A. (2002). Polyglutamine pathogenesis: emergence of unifying mechanisms for Huntington's disease and related disorders. Neuron 35, 819-822. [0268] Ross, C. A., and Poirier, M. A. (2004). Protein aggregation and neurodegenerative disease. Nat Med 10 Suppl, S10-17. [0269] Sabari, B. R., Dall'Agnese, A., Boija, A., Klein, I. A., Coffey, E. L., Shrinivas, K., Abraham, B. J., Hannett, N. M., Zamudio, A. V., Manteiga, J. C., et al. (2018). Coactivator condensation at super-enhancers links phase separation and gene control. Science. [0270] Saudou, F., Finkbeiner, S., Devys, D., and Greenberg, M. E. (1998). Huntingtin acts in the nucleus to induce apoptosis but death does not correlate with the formation of intranuclear inclusions. Cell 95, 55-66. [0271] Schnitzbauer, J., Strauss, M. T., Schlichthaerle, T., Schueder, F., and Jungmann, R. (2017). Superresolution microscopy with DNA-PAINT. Nature protocols 12, 1198-1228. [0272] Schwarz, G. E. (1978). Estimating the dimension of a model. Annals of Statistics. [0273] Sheth, R., Barozzi, I., Langlais, D., Osterwalder, M., Nemec, S., Carlson, H. L., Stadler, H. S., Visel, A., Drouin, J., and Kmita, M. (2016). Distal Limb Patterning Requires Modulation of cis-Regulatory Activities by HOX13. Cell reports 17, 2913-2926. [0274] Shibata, A., Machida, J., Yamaguchi, S., Kimura, M., Tatematsu, T., Miyachi, H., Matsushita, M., Kitoh, H., Ishiguro, N., Nakayama, A., et al. (2016). Characterisation of novel RUNX2 mutation with alanine tract expansion from Japanese cleidocranial dysplasia patient. Mutagenesis 31, 61-67. [0275] Shin, Y., Berry, J., Pannucci, N., Haataja, M. P., Toettcher, J. E., and Brangwynne, C. P. (2017). Spatiotemporal Control of Intracellular Phase Transitions Using Light-Activated optoDroplets. Cell 168, 159-171 e114. [0276] Shin, Y., and Brangwynne, C. P. (2017). Liquid phase condensation in cell physiology and disease. Science 357. [0277] Staby, L., O'Shea, C., Willemoes, M., Theisen, F., Kragelund, B. B., and Skriver, K. (2017). Eukaryotic transcription factors: paradigms of protein intrinsic disorder. Biochem J 474, 2509-2532. [0278] Stampfel, G., Kazmar, T., Frank, O., Wienerroither, S., Reiter, F., and Stark, A. (2015). Transcriptional regulators form diverse groups with context-dependent regulatory functions. Nature 528, 147-151. [0279] Studier, F. W. (2005). Protein production by auto-induction in high density shaking cultures. Protein Expr Purif 41, 207-234. [0280] Truant, R., Atwal, R. S., Desmond, C., Munsie, L., and Tran, T. (2008). Huntington's disease: revisiting the aggregation hypothesis in polyglutamine neurodegenerative diseases. FEBS J 275, 4252-4262. [0281] Villavicencio-Lorini, P., Kuss, P., Friedrich, J., Haupt, J., Farooq, M., Turkmen, S., Duboule, D., Hecht, J., and Mundlos, S. (2010). Homeobox genes d11-d13 and a13 control mouse autopod cortical bone and joint formation. The Journal of clinical investigation 120, 1994-2004. [0282] Wang, J., Choi, J. M., Holehouse, A. S., Lee, H. O., Zhang, X., Jahnel, M., Maharana, S., Lemaitre, R., Pozniakovsky, A., Drechsel, D., et al. (2018). A Molecular Grammar Governing the Driving Forces for Phase Separation of Prion-like RNA Binding Proteins. Cell 174, 688-699 e616. [0283] Wheeler, J. R., Lee, H. O., Poser, I., Pal, A., Doeleman, T., Kishigami, S., Kour, S., Anderson, E. N., Marrone, L., Murthy, A. C., et al. (2019). Small molecules for modulating protein driven liquid-liquid phase separation in treating neurodegenerative disease. BioRxiv, http://dx.doi.org/10.1101/721001. [0284] Yu, G., Wang, L. G., Han, Y., and He, Q. Y. (2012). clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 16, [0285] Zamudio, A. V., Dall'Agnese, A., Henninger, J. E., Manteiga, J. C., Afeyan, L. K., Hannett, N. M., Coffey, E. L., Li, C. H., Oksuz, O., Sabari, B. R., et al. (2019). Mediator Condensates Localize Signaling Factors to Key Cell Identity Genes. Molecular cell. [0286] Zoghbi, H. Y., and Orr, H. T. (2000). Glutamine repeats and neurodegeneration. Annu Rev Neurosci 23, 217-247.