Specific multivalent molecules boost CRISPR-mediated transcriptional activation

Chen, Rui; Shi, Xinyao; Yao, Xiangrui; Gao, Tong; Huang, Guangyu; Ning, Duo; Cao, Zemin; Xu, Youxin; Liang, Weizheng; Tian, Simon Zhongyuan; Zhu, Qionghua; Fang, Liang; Zheng, Meizhen; Hu, Yuhui; Cui, Huanhuan; Chen, Wei

doi:10.1038/s41467-024-51694-y

Download PDF

Article
Open access
Published: 22 August 2024

Specific multivalent molecules boost CRISPR-mediated transcriptional activation

Nature Communications volume 15, Article number: 7222 (2024) Cite this article

4363 Accesses
2 Altmetric
Metrics details

Subjects

Abstract

CRISPR/Cas-based transcriptional activators can be enhanced by intrinsically disordered regions (IDRs). However, the underlying mechanisms are still debatable. Here, we examine 12 well-known IDRs by fusing them to the dCas9-VP64 activator, of which only seven can augment activation, albeit independently of their phase separation capabilities. Moreover, modular domains (MDs), another class of multivalent molecules, though ineffective in enhancing dCas9-VP64 activity on their own, show substantial enhancement in transcriptional activation when combined with dCas9-VP64-IDR. By varying the number of gRNA binding sites and fusing dCas9-VP64 with different IDRs/MDs, we uncover that optimal, rather than maximal, cis-trans cooperativity enables the most robust activation. Finally, targeting promoter-enhancer pairs yields synergistic effects, which can be further amplified via enhancing chromatin interactions. Overall, our study develops a versatile platform for efficient gene activation and sheds important insights into CRIPSR-based transcriptional activators enhanced with multivalent molecules.

Comparative cofactor screens show the influence of transactivation domains and core promoters on the mechanisms of transcription

Article 20 May 2024

Disordered C-terminal domain drives spatiotemporal confinement of RNAPII to enhance search for chromatin targets

Article 28 March 2024

gDesigner: computational design of synthetic gRNAs for Cas12a-based transcriptional repression in mammalian cells

Article Open access 16 September 2022

Introduction

Gene transcription is modulated by dynamic interactions between DNA-binding transcription factors (TFs) and cis-regulatory elements. TFs consist of DNA-binding domains (DBDs), which possess well-defined protein structures for recognizing specific DNA motifs, and activation domains (ADs) that recruit co-activators and often contain intrinsically disordered regions (IDRs) lacking conventional functional structures^1,2. The cis-regulatory elements recognized by TFs include proximal promoters, where transcription initiation occurs, and distal enhancers, which potentiate transcription via engaging long-range chromatin loops with target promoter mediated by higher-order TF-coactivator complexes^3,4.

To manipulate gene expression and rewire endogenous transcriptional networks, synthetic TFs based on zinc fingers (ZFs)^5,6 and transcriptional activator-like effectors (TALEs)^7,8,9 have been engineered. More recently, a similar system based on nuclease dead Cas9 (dCas9), terms as “clustered regularly interspaced short palindromic repeat activation (CRISPRa)”, has been developed. The first-generation CRISPRa system utilizes a VP64 AD fused with dCas9 to activate the transcription of target gene via guide RNA (gRNA) with sequence complementary to the promoter^10,11,12. In order to achieve higher efficiency, subsequent improvements involve modifying either the fused AD or the gRNA design^13,14. For example, the VPR system fuses VP64, p65, and Rta activation domains to dCas9 for synergistic activation¹⁵. The SunTag system harbors up to 24 GCN4 peptide repeat docking sites on dCas9, thereby recruiting multiple copies of activators via a fused GCN4 antibody to specific genomic loci¹⁶. The CRISPR/gRNA-directed synergistic activation mediator (SAM) system utilizes modified gRNAs to recruit multiple activation domains via RNA aptamers¹⁷. All these modifications aim to increase the local concentration of transcriptional activators at target genomic sites. However, these approaches often require multiple bulky fusion partners, potentially hindering delivery, particularly in therapeutic applications^18,19.

Liquid–liquid phase separation (LLPS) is a common physical phenomenon where IDRs or modular domains (MDs) in proteins can drive formation of biomolecular condensates or membraneless organelles, compartmentalizing biochemical reactions²⁰. During gene transcription, TFs and their co-activators, often comprising IDRs, modulate the formation of transcriptional condensates through homotypic and/or heterotypic multivalent interactions^21,22. Transcriptional condensates at super enhancers have been shown to drive the transcription of highly active genes²³. In addition, fusion of the IDR of FUS protein to TetR-VP16 enabled phase separation and increased transcriptional activation²⁴. More recently, IDRs of NUP98 and FUS were also reported to enhance the transcriptional activation potential of dCas9-VPR²⁵. However, two recent studies proposed that optimized multivalent interactions, rather than LLPS per se, enhance transcriptional activation^26,27. Moreover, Chong et al. found that while IDRs that augment multivalent scaffolding robustly amplify gene expression, excessive LLPS can inhibit transcription²⁶. Though the intricate role of LLPS in transcriptional regulation remains controversial, harnessing multivalent interactions hold a promise for the further development of enhanced transcriptional activation system.

Here, to utilize multivalent interactions for enhancing the potency of CRISPR-based transcriptional activation, we screen a panel of 12 IDRs. Although all these IDRs are well-known for their capability to induce phase separation, only seven can augment activation. Using dCas9-VP64-FUS as an example, we demonstrate that cells, both with and without dCas9 microscopic foci formed by phase separation, show potent transcriptional activation of the target gene. In addition to IDR, we investigate several MDs, another class of molecules facilitating multivalent interactions, but not yet associated with transcriptional regulation. Interestingly, while MDs alone do not enhance dCas9-VP64 activity, their fusion with dCas9-VP64-IDR results in substantial further enhancement of transcriptional activation. Furthermore, by varying the number of gRNA binding sites and fusing dCas9-VP64 with IDRs and MDs of different multivalent capabilities, we uncover that optimal, rather than maximal, cis–trans cooperativity is crucial for robust activation. Finally, by simultaneously targeting promoter–enhancer pairs, our enhanced activation system shows synergistic effects, which are critically dependent on IDRs and further amplifiable via enhancing chromatin interactions. Thus, our study develops a versatile platform not only for efficient gene activation, but also for studying the underlying molecular mechanisms.

Results

Some but not all IDRs enhance the activation potency of dCas9-VP64

To systematically evaluate the effect of various IDRs, we generated a panel of fusion proteins comprising dCas9-VP64 linked to diverse IDRs (dCas9-VP64-IDR; Fig. 1a). The sources and details of these IDRs are presented in Fig. 1b and Supplementary Data 1. All these IDRs are well-known for their capability to induce phase separation both in vitro and in vivo^{28,29,30,31,32,33,34,35,36}. We then utilized a HEK293T reporter cell line (HEK293R, Fig. 1a) expressing a GFP cassette under control of seven tetracycline operators (7xTetO) plus a minimal CMV promoter³⁷. As expected, co-transfection of dCas9-VP64 and guide RNAs targeting the TetO element (gTetO) resulted in moderate activation of GFP (Fig. 1c and Supplementary Fig. 1a). For dCas9-VP64-IDR fusions, we found that some, but not all IDRs enhanced the activation capacity of dCas9-VP64, as demonstrated by increased proportions of GFP-positive cells (Supplementary Fig. 1a) as well as elevated median GFP intensities compared to dCas9-VP64 alone (Fig. 1c). Specifically, IDRs from human FUS, EWS, TAF15, YTHDF1-3, and yeast NUP49 (active IDRs) significantly boosted activation, while IDRs from human CCNT1, TDP43, Tau, hnRNPA2 and rat Erc2 (inactive IDRs) did not improve or even slightly inhibited dCas9-VP64 activity (Fig. 1c). As indicated by the intensity of co-translated BFP (Supplementary Fig. 1b) and the staining of the Flag tag at the N-terminus of dCas9 activators (Supplementary Fig. 1c), the differential activation capabilities of the two IDR groups were not due to their expression levels. Moreover, to verify whether the active IDRs alone could promote gene transcription on their own, we directly fused these IDRs to dCas9 and analyzed their impacts on the activation of GFP expression. As shown in Supplementary Fig. 1d, dCas9-IDR fusions did not induce GFP expression, suggesting that IDRs act as amplifiers rather than direct activators in our system. Finally, to develop a useful tool for modulating gene expression, in addition to the efficiency, it is important to evaluate its specificity or off-target effects. Using dCas9-VP64-FUS as a representative example, we performed mRNA sequencing to assess the transcriptome of HEK293R cells co-expressing dCas9-VP64-FUS with either gTetO or a scrambled gRNA (gScr). As shown in Fig. 1d, e, dCas9-VP64-FUS significantly enhanced GFP expression, with more than 500-fold increase in the gTetO group. More importantly, GFP was the only differentially expressed gene (|Log2 (fold change)|>1, FDR < 0.05), demonstrating that dCas9-VP64-FUS specifically activates target genes with negligible off-target effects.

**Fig. 1: dCas9-VP64-IDR robustly activates reporter gene with high specificity.**

IDR-enhanced CRISPRa works on endogenous genes

Given that transcriptional regulation of endogenous genes is more intricate than that of the reporter, we proceeded to evaluate the potential of dCas9-VP64-IDR fusion proteins to activate expression of endogenous genes. For this purpose, we designed gRNAs targeting the promoters of four different genes, namely NTF3, ASCL1, MYOD1, and IL1RN, and co-transfected them with either dCas9-VP64 or dCas9-VP64-FUS into HEK293T cells. We then measured the expression of these target genes using quantitative reverse transcription PCR (RT-qPCR). As shown in Fig. 2a–d, fusing FUS IDR to dCas9-VP64 significantly enhanced the transcriptional efficiency of these endogenous genes. Moreover, dCas9-VP64-FUS also exhibited superior ability when activating multiple genes using gRNAs targeting promoters of NTF3, ASCL1, and MYOD1 simultaneously (Supplementary Fig. 2), demonstrating that our system can be used for multiplex activation of endogenous genes. Furthermore, to evaluate the generalizability of the FUS IDR in CRISPR-based gene activation system, we substituted the Cas protein or AD in the dCas9-VP64-FUS activator. As shown in Fig. 2e–h, by replacing dCas9 with dCpf1 or VP64 with VPR, the FUS IDR significantly boosted the activation ability of both dCas9-VPR and dCpf1-VP64 on endogenous genes.

**Fig. 2: Endogenous genes activated by dCas9-VP64-IDR.**

The multivalent interactions mediated by IDRs are crucial for enhancing transcriptional activation

IDRs can engage in multivalent interactions between the low-complexity sequences. To investigate whether the ability of IDRs to enhance CRISPRa is through such multivalent interactions, we generated a dCas9-VP64-FUS^27YS fusion protein where all 27 tyrosine residues in the FUS IDR were mutated to serine, thereby abolishing its ability to form multivalent interactions while preserving intrinsically disordered property³⁸. In contrast to the wild-type FUS IDR, the addition of the FUS^27YS mutant failed to amplify dCas9-VP64 activation of the GFP reporter (Fig. 3a) as well as endogenous genes (Fig. 3b, c). Furthermore, we divided the FUS IDR into three parts and reshuffled them to generate the activators containing various scrambled FUS sequences, namely FUS-132, FUS-213, FUS-231, FUS-312, and FUS-321 (Supplementary Fig. 3b). Subsequently, we assessed the activation potential of these activators on the GFP reporter and two endogenous genes. As shown in Supplementary Fig. 3c–e, there was no significant difference in the activation of endogenous genes among these activators. The FUS-321 mutant exhibited a slightly higher activation effect only on the reporter gene compared to the other activators. These results together suggested that both the amino acid composition and the intact multivalent interaction ability are essential for the enhancement effect of FUS IDR.

**Fig. 3: IDR-mediated multivalent interactions and transcriptional activation.**

To check whether IDR-containing dCas9 activators form biomolecular condensates via LLPS, we co-expressed BFP-tagged dCas9-VP64-IDR fusion proteins with a gTetO in the HEK293R cells to visually examine condensate formation and correlate that with transcriptional activation. Active IDR fusions including dCas9-VP64-FUS and dCas9-VP64-YTHDF1 formed multiple distinct puncta and robustly activated GFP expression (Supplementary Fig. 3a). In contrast, the FUS^27YS mutant significantly interfered with puncta formation and GFP activation (Supplementary Fig. 3a). Interestingly, inactive IDR fusions such as dCas9-VP64-TDP43 and dCas9-VP64-CCNT1 also generated puncta in cells yet did not enhance GFP expression (Supplementary Fig. 3a). Subsequently, we assessed the fluidity of dCas9-VP64-FUS and dCas9-VP64-TDP43 puncta through fluorescence recovery after photobleaching (FRAP) analysis. As shown in Fig. 3d, dCas9-VP64-FUS displayed rapid fluorescence recovery, while dCas9-VP64-TDP43 showed a relatively slower recovery rate (Fig. 3e). Moreover, we found that visible puncta were present in only around 46% (Supplementary Fig. 3f) dCas9-VP64-FUS expressing cells with a wide range of GFP expression (Supplementary Fig. 3a). The GFP intensity and puncta number as well as the BFP intensity measured in 200 cells also manifested no correlation (Supplementary Fig. 3g, h). Taking together, our analysis suggested that the ability of IDRs to engage in multivalent interactions, rather than that for inducing phase separation per se, governs their capabilities of transcriptional activation enhancement.

Next, if the activation enhancement was via multivalent interaction, the tandem fusion of multiple active IDRs should further increase activation efficiency. To check this, using dCas9-VP64-FUS as an example, we extended the C-terminus with additional IDRs of FUS, EWS, or TAF15. Indeed, the addition of homologous or heterologous IDRs additively enhanced activation of both the GFP reporter (Fig. 3f) and endogenous genes (Fig. 3g, h), suggesting a positive correlation between IDR-mediated multivalent interaction and transcriptional activation.

Finally, to investigate whether IDRs recruit specific transcriptional or chromatin regulatory factors for transcriptional regulation, we performed co-IP analysis and revealed that both the effective dCas9-VP64-FUS and the non-effective dCas9-VP64-TDP43 could interact with BRG1, the core catalytic subunit of the BAF complex, but not the Mediator component MED1 (Fig. 3i). Intriguingly, only dCas9-VP64-FUS showed an interaction with the RNA polymerase II subunit RPB1 (Fig. 3i). Furthermore, to investigate whether these factors could be recruited to the targeted locus, we conducted ChIP-qPCR analysis at the promoters of NTF3 and IL1RN, which were targeted by the respective gRNAs. As shown in Fig. 3j, k, fusion of either effective IDR FUS or non-effective IDR TDP43 to the dCas9-VP64 activator exhibited similar DNA-binding capabilities. Consistent with the co-IP results, BRG1, but not MED1 was notably enriched at these regions when the regions were targeted by both the effective dCas9-VP64-FUS and non-effective dCas9-VP64-TDP43. However, RPB1 enrichment was observed only when the promoters were targeted by dCas9-VP64-FUS, in consistence with the RNA expression data. In addition, we utilized two distinct shRNAs to knock down the expression of BRG1 (Supplementary Fig. 3i). As shown in Supplementary Fig. 3j–l, a significant reduction in the activation efficiency of dCas9-VP64-FUS on both the GFP reporter and endogenous genes was observed (Supplementary Fig. 3j–l), suggesting the necessity of the BAF complex in the activation of target genes. However, as both the effective dCas9-VP64-FUS and the non-effective dCas9-VP64-TDP43 could recruited BRG1 to the targeted locus, the recruitment of BAF complex is not sufficient for transcriptional activation.

IDR and MD cooperatively enhance transcriptional activation

In addition to IDRs, MDs represent another class of protein domains that can self-assemble into higher-order oligomers through multivalent interactions and undergo phase separation. To test whether MDs can enhance activation efficiency, we linked two MDs, the NTD domain of Par3 (Par3 MD)³⁹ and a simplified Shank3 protein containing the PDZ, HBS, CBS and SAM domains (Shank3 MD)⁴⁰, to dCas9-VP64, respectively (Supplementary Data 2). Surprisingly, both MDs failed to enhance the transcriptional activation ability of dCas9-VP64 on the GFP reporter as well as endogenous genes (Fig. 4a, b, Supplementary Fig. 4a).

**Fig. 4: Cooperation of IDR and MD in CRISPRa-based transcriptional activation.**

Inspired by a recent study showing that MD could largely enhance the ability of IDR-induced phase separation in vitro⁴¹, we hypothesized that IDR and MD in combination might further enhance transcriptional activation. Toward this goal, we linked the MDs of Par3 or Shank3 to dCas9-VP64-FUS and observed a much stronger activation efficiency on endogenous genes (Fig. 4a, b). Next, to test whether this enhancement was caused by MD-mediated multivalent interactions, we mutated methionine 1793 in the SAM domain of Shank3 to alanine (Shank3^MA) or glutamic acid (Shank3^ME). These two mutations have been demonstrated to disrupt oligomerization of Shank3 to different extents⁴². As shown in Supplementary Fig. 4b, indeed, the Shank3^ME mutant effectively disrupted the formation of condensates. As predicted, these mutations also abolished the amplification effect to varying degrees in accordance with their attenuated oligomerization propensities (Fig. 4c, d).

Next, to reduce the size and delivery challenge of dCas9-VP64-IDR-MD fusion proteins, we tested seven short synthetic MDs (~30 amino acids), namely HOTag1-7 (Supplementary Data 2), for enhancing dCas9-VP64-FUS activation of endogenous genes. These HOTags are de novo-designed coiled coils, capable of forming tetramer, pentamer, hexamer, and heptamer, respectively⁴³. Similar to natural MDs, HOTags exhibited varying abilities to amplify dCas9-VP64-FUS activation capability, with HOTag3 showing the most significant enhancement (Fig. 4e, f). In contrast, direct fusion of HOTags to dCas9-VP64 did not enhance activation of the same endogenous targets or the GFP reporter (Supplementary Fig. 4c–e), consistent with our previous observation for natural MDs. Finally, we checked whether the order of connecting IDR and MD influenced the activation effect. Interestingly, as shown in Fig. 4g, dCas9-VP64-FUS-HOTag3 induced slightly but statistically significantly higher endogenous gene activation compared to dCas9-VP64-HOTag3-FUS, indicating the importance of optimizing the architecture of fusion protein. Together, the cooperative effects between IDRs and MDs allow substantial enhancement of dCas9-VP64 activator on endogenous genes, dependent on the design and arrangement of the components.

Optimal cis–trans interactions are required for robust transcriptional activation

Whereas fusing the MDs to dCas9-VP64-FUS enhanced activation of endogenous gene targets compared to dCas9-VP64-FUS alone, surprisingly, the addition of the same MDs inhibited dCas9-VP64-FUS activation of a GFP reporter gene under the control of seven tandem TetO repeats (7xTetO) in HEK293R cells (Supplementary Fig. 4f). Considering that the major difference between endogenous genes and the GFP reporter is that the promoter of the reporter contains seven tandem TetO repeats targeted by our gRNA, we hypothesized that the multivalent interactions provided by fused IDR-MDs may exceed the optimal range for activating the 7xTetO GFP reporter. If this is the case, the IDR-MD fusion may still show enhanced activation when the promoter of the reporter gene contains fewer gRNA binding sites. To test this, we compared fusion proteins with or without IDR and MD in parallel on GFP reporters under control of different numbers of gRNA binding sites, namely 1xTetO, 7xTetO, and 14xTetO. As expected, for the same activator, more binding sites always led to higher transcriptional activation. Interestingly, whereas for the 1xTetO reporter, GFP intensity positively correlated with the multivalent capacity of dCas9-based activators, the activator containing IDR-MD with the highest multivalent capacity failed to further enhance GFP expression in the 7xTetO and 14xTetO reporters (Fig. 4h). A similar phenomenon was also observed for dCas9-VP64 fused with 1x, 2x, and 3x FUS IDR (Supplementary Fig. 4g). Collectively, our findings demonstrate that an excess of cis–trans interactions can lead to a decrease in efficiency of transcriptional activation. This suggests that achieving robust transcriptional activation might require optimal cooperativity between trans-acting factors and cis-regulatory DNA elements.

Simultaneously targeting enhancer-promoter synergistically boosts gene activation

Enhancers are distal cis-regulatory elements that can interact with target gene promoters, bringing bound transcriptional co-activators into proximity to regulate transcription. First, to test our IDR/MD-containing activators in modulating enhancer activity, we chose the well-characterized human β-globin locus, which contains β-like globin genes (HBB, HBD, HBE, and HBG1/2 as well as a pseudogene HBBP1) that are developmentally regulated by a shared upstream enhancer cluster known as the locus control region (LCR) (Fig. 5a). The LCR consists of five DNase I hypersensitive sites (HS1-HS5) among which the HS2 enhancer has been extensively studied as an erythroid-specific enhancer playing a critical role in regulating the transcription of β-globin genes within this locus⁴⁴. As shown in Supplementary Fig. 5a–e, compared to the dCas9-VP64, targeting IDR/MD-containing activators to the HS2 enhancer led to higher though still moderate activation, but only for HBG1/2.

**Fig. 5: Synergistic activation by concurrently targeting enhancer-promoter pairs.**

Given that enhancer-promoter communication can be mediated by multivalent protein-protein interactions, we investigated whether concurrently guiding the synthetic activators containing multivalent molecules to endogenous enhancer-promoter pairs could synergistically promote target gene activation (Fig. 5b). Indeed, compared to targeting either enhancer or promoter alone, simultaneous targeting of the HS2 enhancer and the promoter of individual β-globin genes by IDR/MD-containing activators led to substantially higher gene activation. As shown in Fig. 5c–g, the expression of all the β-globin genes (HBE, HBG1/2, HBD, and HBB) and the pseudogene HBBP1 could be synergistically boosted by dCas9-VP64-FUS and dCas9-VP64-FUS-HOTag3, whereas no such synergy was observed for dCas9-VP64. Moreover, we designed gRNAs to target the flanking regions of the HS2 enhancer (2–2.5 kb and 0.5–1 kb upstream or downstream of the HS2 region). As shown in Supplementary Fig. 5f, g, compared to the HS2 enhancer, simultaneous targeting its flanking regions and the promoter of β-globin genes by dCas9-VP64-FUS resulted in a notable decrease in activation to varying degrees. The extent of decrease is correlated with the distance from the HS2 enhancer, which demonstrated the specificity of simultaneous targeting of enhance/promoter pair. Furthermore, this synergistic effect was also observed when the enhancer-promoter pair of MYOD1 was targeted by IDR/MD-containing activators (Supplementary Fig. 5h). Together, our data suggested the superiority of multivalent molecules in activating gene expression by simultaneously targeting enhancer-promoter pairs.

Enhancing chromatin interactions further amplifies transcriptional activation

To test whether enhancing enhancer-promoter interactions could further amplify gene expression, we engineered inducible dimerization of dCas9 and dCpf1 fusions using the ABA-induced ABI1-PYL1 system, which has been demonstrated to modulate chromatin interactions⁴⁵. Specifically, we appended ABI1 and PYL1 dimerization partners to dCas9-VP64-FUS and dCpf1-VP64-FUS, respectively, and then guided them to enhancer or promoter using specific gRNAs (Fig. 6a). For comparison, we also generated several control fusion proteins lacking VP64, FUS IDR, or both. In the absence of ABA, targeting the HS2 enhancer and HBD or HBG1/2 promoter with ABI1-mCherry-dCas9-VP64-FUS and PYL1-BFP-dCpf1-VP64-FUS, respectively, led to higher expression of HBD or HBG1/2, respectively, confirming synergistic activation by dual targeting as observed above (Supplementary Fig. 6a, b). Strikingly, the addition of ABA to induce ABI1-PYL1 heterodimerization further significantly enhanced the activation of HBD and HBG1/2, but only when both the enhancer and promoter were targeted by VP64-FUS fusions (Fig. 6b, c, Supplementary Fig. S6c, d), again highlighting the essential contribution of FUS IDR for such synergistic effect. Furthermore, we designed gRNA and crRNA to recruit dCas9-VP64-FUS and dCpf1-VP64-FUS to the enhancer and promoter regions of the human MYOD1 gene, respectively. As illustrated in Supplementary Fig. 6e, the presence of ABA notably boosted the activation of MYOD1.

**Fig. 6: Transcriptional activation enhanced via enhancing chromatin interactions.**

To further investigate the specificity of enhancer-promoter interactions in our synthetic system, we utilized the same tethering strategy to target the HS2 enhancer and the promoter regions of UBQLN3 or TRIM22, which are not direct targets of HS2. These promoter regions are located approximately 230 kb and 400 kb away from the HS2 enhancer, respectively. As shown in Supplementary Fig. 6f, g, in HEK293T cells stably co-expressing ABI1-mCherry-dCas9-VP64-FUS with gHS2 and PYL1-BFP-dCpf1-VP64-FUS with crUBQLN3 or crTRIM22, we observed only a slight increase of TRIM22 expression, while there was no statistically significant alteration in UBQLN3 expression following ABA treatment. Taken together, these empirical findings suggested that the amplified activation via tethering distal and proximal elements would work efficiently only for specific enhancer-promoter pairs.

Discussion

While increasing evidence has suggested regulatory roles of multivalent molecules in diverse cellular processes, their precise quantitative contributions to transcriptional control remained to be fully elucidated⁴⁶. In this study, by engineering synthetic CRISPR activators containing IDR/MD, we delineated their roles in the modulation of transcriptional activation. In our engineered system, we found that the induction of transcriptional activation was strictly dependent on the AD, and multivalent molecules themselves could not directly induce gene transcription. This finding aligns well with other studies showing that the presence of an IDR alone is insufficient to activate reporter or endogenous genes in various synthetic systems^24,25,27. Instead, IDRs and MDs act as amplifiers to enhance the activation of gene transcription in our system. By fusing IDRs and MDs with different multivalent capabilities to the activators, our tool could achieve different expression levels of a target gene, enabling quantitative analysis of the effect of different gene expression levels across a broad range. Moreover, our synthetic system has a negligible off-target effect, comparable to other published CRISPRa approaches^10,15,16,17. These findings support incorporating multivalent molecules as an effective strategy to improve CRISPRa efficiency, without compromising specificity.

Although a previous study found that IDRs of FUS and NUP98 could increase the activation capability of dCas9-VPR²⁵, it remains unclear whether all IDRs hold such potential. Our finding that certain IDRs can enhance dCas9-VP64 activation ability while others cannot, raises an important question: what factors contribute to the specificity of IDRs in boosting gene transcription? While IDRs themselves often exhibit high conformational flexibility⁴⁷, it is important to consider physicochemical properties and conformations of IDR fusion proteins within the context of transcriptional activation⁴⁸. However, we observed no significant differences in physicochemical properties between active and inactive IDRs (Supplementary Data 1). Moreover, an IDR’s interaction partners can profoundly impact its capacity to amplify gene activation. For instance, FUS IDR interacts with components of the BAF chromatin remodeling complex⁴⁹ and engages critical co-activators and chromatin regulators more extensively⁵⁰, whereas the IDRs of m6A reader YTHDF proteins may recruit enhancer RNAs to promote gene activation⁵¹. Our co-IP and ChIP-qPCR data demonstrated that the fusion of either FUS IDR or TDP43 IDR to the dCas9-VP64 activator did not affect its DNA-binding capability, and both fusion proteins could interact with BRG1. However, only the dCas9-VP64-FUS fusion could interact with and recruited the RNA polymerase II subunit RPB1 to the target locus, in consistent with the corresponding RNA expression data. It awaits future study how the differential recruitment of RPB1 was achieved.

IDRs and MDs are two classes of molecules facilitating multivalent interactions and associated with the capacity of forming LLPS. However, MDs differ markedly from IDRs in their amino acid composition and can undergo compact folding⁵². For example, FUS IDR is a well-characterized prion-like domain capable of phase separation by itself⁵³, while the SAM domain of Shank3 MD can specifically interact with each other in a head-to-tail manner, forming large polymers⁴². IDR interactions are often weak and transient, while MD interactions tend to be relatively strong and more defined^20,54. Unlike IDRs, the MDs examined in our study did not enhance the activation ability of dCas9-VP64, indicating the role of transient, nonspecific nature of multivalent IDR-IDR interaction in transcriptional activation. However, IDR and MD in combination could cooperatively amplify transcriptional activation. This is likely because the IDR-MD molecule could form a larger but still dynamic molecular network due to the presence of both transient and specific interactions mediated by IDR and MD, respectively. Indeed, it has recently been demonstrated that the IDR-MD chimera of FUS IDR and Shank3 MD can phase separate in vitro at very low concentrations. Interestingly, the Shank3 MD mutants exhibit attenuation of specific interactions for molecular network formation in the condensed phase⁴¹, consistent with our findings on their weak effects on transcriptional activation when combined with FUS IDR. Intriguingly, such synergistic effect of unspecific flexible and specific rigid multivalent interactions on formation of phase separation was often observed in naturally occurring condensates consisting of DNA/TFs or RNA/RNA-binding proteins (RBPs), where IDRs of TFs/RBPs mediate unspecific interactions, whereas DNA/RNA-binding domains facilitate specific interactions with nucleic acids^47,55.

The multivalent ability of synthetic activators is likely the key to enhancing transcriptional activation, where IDRs-mediated multivalent interactions can increase the local concentration of activators and co-activators at target sites, resulting in higher activation efficiency^25,56. However, we found that visible LLPS-like puncta were not a prerequisite for transcriptional activation, as (1) CRISPR activators containing either active IDRs (FUS, YTHDF1) or inactive IDRs (TDP43, CCNT1) could form LLPS-like puncta and (2) GFP could be activated in reporter cells with and without dCas9-VP64-FUS puncta in our experiments. Therefore, the formation of visible LLPS-like puncta in cells was likely neither required nor sufficient to induce transcriptional activation, in line with a recent publication suggesting that multivalent interactions mediated by IDRs are important for gene activation²⁷. However, solely based on current data, we cannot completely exclude the role of forming transcriptional condensates, as some tiny and transient condensates, especially those only at specific transcriptional loci, cannot be observed under the normal confocal microscope.

Previously, Chong et al. found that transcriptional activation requires finely tuned optimal levels of the non-specific interactions mediated by IDRs. Overly high amounts of IDR-IDR interactions can outcompete specific TF-DNA binding, leading to LLPS and transcription repression²⁶. Unlike other persistent protein-nucleic acid condensates like the nucleolus and processing bodies, transcriptional activation is a relatively transient and dynamic process, determined by the kinetics of TFs binding to specific DNA sequences^57,58. By varying the number of gRNA binding sites and applying IDR/MD-containing activators with different multivalent capabilities, we demonstrated that excessive levels of cis–trans interactions constrained activation capability, indicating that optimal cis–trans cooperativities enable robust transcriptional control. A similar finding was also observed in a recent study using chemically controlled TF clustering to induce gene transcription⁵⁹. This may also explain the lack of correlation between LLPS-like puncta and GFP expression observed in our imaging experiments. An optimal level of multivalent interactions mediated by IDR-containing activators might be required for efficient activation of the GFP reporter gene. Either too little or too much transcriptional condensates formation could lead to inefficient activation of gene expression. Of course, whether the conclusions from such synthetic systems could be generalized to the regulation of endogenous genes warrants future investigation.

Compared to robust transcriptional activation by targeting promoters, the activation induced by guiding our synthetic activators to enhancers was substantially less efficient, consistent with previous studies using other dCas9-based activators^60,61,62. However, by simultaneously targeting enhancer-promoter pairs, we observed a synergistic increase in transcriptional activation. Notably, this synergistic effect was only achieved when employing the dCas9-VP64 activator fused with multivalent molecules, aligning well with emerging evidence that highlights the crucial role of IDR-containing chromatin regulators in organizing long-range chromatin interactions^63,64. Furthermore, by employing a chemically controlled tethering system to artificially enhance chromatin interactions between the HS2 enhancer and the promoters of two β-globin genes, we could further achieve additional transcriptional enhancement. However, when the same tethering system was applied to the HS2 enhancer and the promoters of UBQLN3 or TRIM22, no efficient activation amplification was observed, likely due to the specificity of enhancer-promoter interactions and the substantial genomic distance between HS2 enhancer and the promoter of TRIM22 and UBQLN3. In the future, our adjustable synthetic system can be employed to systematically dissect the molecular mechanisms underlying productive long-range chromatin interactions in transcriptional control, although the insights gained using the synthetic systems should be extended to endogenous systems always with caveat.

Methods

Cell culture and transfection

The HEK293T cell line was obtained from the American Tissue Collection Center (ATCC, CRL-3216). The HEK293R reporter cells were generously provided by Dr. Zhou Songyang at Sun Yat-sen University³⁷. These cells were cultured in DMEM medium with addition of 10% FBS and 1% PS at 37 °C and 5% CO₂. For CRSIPRa transfection, 800 ng of dCas9-fusion protein expression plasmids and 200 ng individual or equimolar pooled gRNA expression plasmids were transfected into cells in 12-well plate by PEI (Sigma, 408727). Following a 24-h culture post-transfection, the transfected cells underwent selection for additional 48 h using puromycin (Yeasen, 60209ES) at a final concentration of 2 μg/μl. Subsequently, these cells were utilized for the following experiments, including flow cytometry, RT-qPCR, and mRNA-seq analyses.

For ABI1-PYL1 inducible double CRSIPRa transfection, 400 ng of PYL1-BFP-dCpf1-based activator, 400 ng of ABI1-mCherry-dCas9-based activator, 100 ng enhancer gRNAs and 100 ng promoter crRNAs expression plasmids were transfected into cells using the same method. After 24 h of culture, puromycin (Yeasen, 60209ES) was added into the medium at a final concentration of 2 μg/μl to select positive cells, while abscisic acid (ABA, MCE, HY-100560) was added at a final concentration of 100 μM to induce the ABI1-PYL1 interaction. Finally, the next experiments were performed three days after transfection.

Plasmids construction

The dCas9-fusion protein expression plasmids were generated by incorporating VP64/VPR, IDRs and/or MDs, P2A-BFP into pB-CAGGS-dCas9 (Addgene, 110823) using Gibson assembly with the Hieff Clone® Universal One Step Cloning Kit (Yeasen, 10922ES). The IDRs and MDs, except synthetic HOTag1-7, were amplified by PCR from human, yeast, or mouse cDNA and the specific details were listed in Supplementary Data 1 and 2. For the dCpf1-VP64-FUS fusion protein expression plasmids, a similar assembly process was followed, with the additional step of replacing dCas9. The synthetic gRNAs or shRNAs, listed in Supplementary Data 3, were inserted into Lenti-guide Puro plasmid (Addgene, 52963) or pLKO.1 puro plasmid (Addgene, 8453), respectively, and expressed under control of a U6 promoter. The 1xTetO-GFP, 7xTetO-GFP, and 14xTetO-GFP reporters were constructed by modifying TetO spacer sequence and replacing APEX2 with GFP in the RAR3-APEX2-FLAG plasmid⁶⁵. To construct the PYL-BFP-dCpf1-VP64-FUS and ABI-mCherry-dCas9-VP64-FUS expression plasmids, the Cas9-P2A-Blast in the lentiCas9-Blast plasmid (Addgene, 52962) was replaced with sequence encoding PYL1/ABI1, BFP/mCherry, and CRISPRa proteins.

Virus package and stable cell line construction

For each virus packaging, HEK293T cells were seeded one day prior and transfected with the lentiviral plasmid, pMD2.G (Addgene, 12259), and psPAX2 (Addgene, 12260), at a ratio of 1.64:0.72:1.3 using PEI (Sigma, 408727). The medium was refreshed 12 h after transfection. After 48 h, the supernatant was collected, followed by centrifugation and filtration through a 0.45-μm filter. The lentivirus particles were concentrated using PEG8000 (Promega, V3011) precipitation and then applied to the targeting cells for transduction.

Flow cytometry analysis

HEK293R cells transfected with dCas9-VP64-IDR/MD and gTetO were dissociated to single-cell suspensions by trypsin. The fluorescence intensities of BFP (excited with 405 nm laser), GFP (excited with 488 nm laser), and mCherry (excited with 561 nm laser) were measured using a flow cytometer (Beckman Coulter, CytoFLEX). To exclude cells without dCas9 fusion expression, only BFP-positive cells were included in the calculation of GFP intensity.

For flow cytometry analysis of Flag-tagged dCas9 activators, transfected cells were dissociated into single-cell suspensions using trypsin. After centrifugation, the cell pellets were fixed and permeabilized with the Transcription Factor Fixation/Permeabilization working solution (Invitrogen, 00-5123 and 00-5223) in the dark for 45 min. Subsequently, the treated cells were washed with Permeabilization Buffer (Invitrogen, 00-8333) and then incubated with an anti-Flag antibody (Proteintech, 20543-1-AP) in the dark for 45 min. Following the removal of unbound antibodies with Permeabilization Buffer, CoraLite594-conjugated secondary antibody was applied to label the Flag-tagged dCas9 activators with 30 min incubation. Finally, the labeled cells underwent two washes with Permeabilization Buffer before being loaded onto a flow cytometer for measurement.

Confocal imaging

HEK293R cells were cultured on a Glass Bottom Cell Culture Dish (biosharp BS-20-GJM) and transfected with dCas9-VP64-IDR fusions with gTetO. After 48 h of transfection, living cell imaging was performed using a Zeiss LSM980 Confocal Microscope. The images were captured using a 63× oil objective with a slice interval of 0.2 μm on the ZEISS ZEN 3.6 Software. The fluorescence proteins BFP and GFP expressed in our cells were excited by 405 nm and 488 nm lasers, respectively. The acquired data was subsequently analyzed using the ZEISS ZEN 3.6 software.

FRAP assay

Fluorescence recovery after photobleaching (FRAP) experiments were conducted using a Zeiss LSM980 Confocal Microscope equipped with a ×63 oil objective. A condensate with a diameter of 0.5–4 μm was selected for analysis. A round region measuring 0.7 μm on each side was chosen for FRAP analysis. The BFP signal was bleached using a 405 nm laser at 100% laser power. Fluorescence intensity changes over time were recorded for pre-bleaching frames and within 30 s after bleaching. For quantitative analysis, the average intensity of the frames before photobleaching was normalized to 100%. Image analysis was performed using the ZEISS ZEN 3.6 software.

Quantitative reverse transcription-PCR

RNA was extracted from transfected HEK293T cells using the RNA isolater Total RNA Extraction Reagent (Vazyme, R401). Subsequently, cDNA synthesis was performed using the HiScript II Q Select RT SuperMix for qPCR (Vazyme, R233). Real-time PCR was conducted using the ChamQ SYBR qPCR Master Mix (Vazyme, Q311) on the CFX Connect Thermal Cycler (BIO-RAD). The fold changes of the target genes relative to the control cells were calculated using the ΔΔCt method, with normalization to GAPDH expression. Please refer to Supplementary Data 4 for a list of all qPCR primers used.

Chromatin immunoprecipitation (ChIP)

ChIP experiments were carried out according to the standard protocol provided by the SimpleChIP Plus Sonication Chromatin IP Kit (CST, 56383). HEK293R cells were transfected with Flag-tagged dCas9 activators together with gTetO. Sonication was performed using the Bioruptor pico (Diagenode) by applying 10 cycles of 30 s ON and 30 s OFF to obtain chromatin fragments of ~100–500 bp. The anti-Flag antibody (Proteintech, 20543-1-AP), anti-BRG1 antibody (Abcam, ab110641), anti-RNA Polymerase II RPB1 antibody (BioLegend, 664906), anti-MED1 antibody (Abcam, ab64965) and Normal Rabbit IgG (Cell Signaling, 2729S) were used. ChIP DNA was purified using the Zymo ChIP DNA cleanup kit (Zymo, D5205). The ChIP enrichment was analyzed by quantitative PCR.

Co-immunoprecipitation (co-IP) and western blotting (WB)

HEK293T cells transfected with Flag-tagged dCas9 activators along with gRNAs targeting NTF3 or IL1RN promoter were harvested and homogenized in lysis buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, protease inhibitor cocktail (TargetMol, C0001) with 2 h of rotation. Following centrifugation, the protein lysates in the supernatant were collected, and their concentrations were determined using the BCA Protein Quantification Kit (Vazyme, E112-01). For immunoprecipitation, the protein lysates were incubated overnight with rotation using an anti-Flag antibody (Proteintech, 20543-1-AP) and Normal Rabbit IgG (Cell Signaling, 2729S). Subsequently, ChIP-Grade Protein G Magnetic Beads (Cell Signaling, 9006S) were introduced to each reaction. After an additional 2 h of rotation, the beads were washed four times with lysis buffer and eluted by heating at 95 °C for 10 min. The eluted proteins were loaded onto an 8% PAGE gel for separation and transferred to PVDF membranes using wet/tank blotting systems. Afterwards, the PVDF membranes were blocked using 5% skim milk powder/TBST and then incubated overnight at 4 °C with primary antibodies, including anti-Flag antibody (Proteintech, 20543-1-AP), anti-BRG1 antibody (Abcam, ab110641), anti-RNA Polymerase II RPB1 antibody (BioLegend, 664906), and anti-MED1 antibody (Abcam, ab64965). Following three washes with TBST, the membranes were incubated with secondary antibodies for 1 h at room temperature. Subsequently, the membranes were washed three times with TBST and incubated with Clarity Western ECL substrate (Bio-Rad, 170-5061). The protein bands were detected using a ChemiDoc XRS+ imaging system (Bio-Rad).

mRNA-seq library construction and data processing

RNA was isolated from HEK293R cells transfected with dCas9-VP64 or dCas9-VP64-FUS, along with gTetO and gScr, using the TRIzol™ reagent (Invitrogen, 15596018). Subsequently, mRNA-seq libraries were prepared using the Hieff NGS® Ultima Dual-mode mRNA Library Prep Kit (Yeasen, 12309), with 2 μg of total RNA input. Briefly, mRNA was captured and purified from the total RNA using oligo(dT) RNA capture beads. Following fragmentation, strand-specific ds-cDNAs were synthesized through a two-step linear amplification process and ligated to sequencing adapters. The cDNA fragments, ranging from 350 to 450 bp, were selected and purified using the Hieff NGS® DNA Selection Beads (Yeasen, 12601). Finally, libraries were amplified with indexed primers for 12 cycles of PCR and sequenced on the NovaSeq 6000 platform (Illumina) in a 2 × 150 nt manner.

For mRNA-seq data analysis, clean reads were aligned to human reference genome (hg38) supplemented with GFP sequence by HISAT2⁶⁶. Aligned reads were inputted into featureCounts⁶⁷ for counting the read number of coding genes and GFP. The genome annotation gtf file was obtained from Ensemble v105 and only coding genes and additional GFP annotation were used for count. The differential expression genes were identified using DEseq2⁶⁸.

Statistics and reproducibility

All experiments were independently repeated at least three times with consistent results. The number of replicates in each experiment is indicated in the figure legends and/or methods. Bar plot data are expressed as means ± SD using Prism software (GraphPad). Statistical comparisons were performed using one-way ANOVA followed by Welch’s test or a two-sided Welch’s t-test. Boxplot data are presented as the median fluorescence intensity, with the 25th and 75th quartiles, as well as the 5th and 95th percentiles. Statistical significance was determined by a two-sided Wilcoxon rank-sum test. A P-value of <0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The RNA-seq data generated from this study have been submitted to the NCBI under the accession number GSE248523. All materials, protocols and described in this study are available from the corresponding author upon request. Source data are provided with this paper.

References

Soto, L. F. et al. Compendium of human transcription factor effector domains. Mol. Cell 82, 514–526 (2022).
Article CAS PubMed Google Scholar
Lambert, S. A. et al. The human transcription factors. Cell 175, 598–599 (2018).
Article CAS PubMed Google Scholar
Lee, T. I. & Young, R. A. Transcriptional regulation and its misregulation in disease. Cell 152, 1237–1251 (2013).
Article CAS PubMed PubMed Central Google Scholar
Hafner, A. & Boettiger, A. The spatial organization of transcriptional control. Nat. Rev. Genet. 24, 53–68 (2023).
Article CAS PubMed Google Scholar
Beerli, R. R. & Barbas, C. F. 3rd Engineering polydactyl zinc-finger transcription factors. Nat. Biotechnol. 20, 135–141 (2002).
Article CAS PubMed Google Scholar
Beerli, R. R., Dreier, B. & Barbas, C. F. 3rd. Positive and negative regulation of endogenous genes by designed transcription factors. Proc. Natl Acad. Sci. USA 97, 1495–1500 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Maeder, M. L. et al. Robust, synergistic regulation of human gene expression using TALE activators. Nat. Methods 10, 243–245 (2013).
Article CAS PubMed PubMed Central Google Scholar
Perez-Pinera, P. et al. Synergistic and tunable human gene activation by combinations of synthetic transcription factors. Nat. Methods 10, 239–242 (2013).
Article CAS PubMed PubMed Central Google Scholar
Zhang, F. et al. Efficient construction of sequence-specific TAL effectors for modulating mammalian transcription. Nat. Biotechnol. 29, 149–153 (2011).
Article PubMed PubMed Central Google Scholar
Gilbert, L. A. et al. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442–451 (2013).
Article CAS PubMed PubMed Central Google Scholar
Maeder, M. L. et al. CRISPR RNA-guided activation of endogenous human genes. Nat. Methods 10, 977–979 (2013).
Article CAS PubMed PubMed Central Google Scholar
Perez-Pinera, P. et al. RNA-guided gene activation by CRISPR-Cas9-based transcription factors. Nat. Methods 10, 973–976 (2013).
Article CAS PubMed PubMed Central Google Scholar
Dominguez, A. A., Lim, W. A. & Qi, L. S. Beyond editing: repurposing CRISPR-Cas9 for precision genome regulation and interrogation. Nat. Rev. Mol. Cell Biol. 17, 5–15 (2016).
Article CAS PubMed Google Scholar
Chavez, A. et al. Comparison of Cas9 activators in multiple species. Nat. Methods 13, 563–567 (2016).
Article CAS PubMed PubMed Central Google Scholar
Chavez, A. et al. Highly efficient Cas9-mediated transcriptional programming. Nat. Methods 12, 326–328 (2015).
Article CAS PubMed PubMed Central Google Scholar
Tanenbaum, M. E., Gilbert, L. A., Qi, L. S., Weissman, J. S. & Vale, R. D. A protein-tagging system for signal amplification in gene expression and fluorescence imaging. Cell 159, 635–646 (2014).
Article CAS PubMed PubMed Central Google Scholar
Konermann, S. et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 517, 583–588 (2015).
Article ADS CAS PubMed Google Scholar
Liao, H. K. et al. In vivo target gene activation via CRISPR/Cas9-mediated trans-epigenetic modulation. Cell 171, 1495–1507.e1415 (2017).
Article CAS PubMed PubMed Central Google Scholar
Zhou, H. et al. In vivo simultaneous transcriptional activation of multiple genes in the brain using CRISPR-dCas9-activator transgenic mice. Nat. Neurosci. 21, 440–446 (2018).
Article CAS PubMed Google Scholar
Banani, S. F., Lee, H. O., Hyman, A. A. & Rosen, M. K. Biomolecular condensates: organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 18, 285–298 (2017).
Article CAS PubMed PubMed Central Google Scholar
Boija, A. et al. Transcription factors activate genes through the phase-separation capacity of their activation domains. Cell 175, 1842–1855.e1816 (2018).
Article CAS PubMed Google Scholar
Cho, W. K. et al. Mediator and RNA polymerase II clusters associate in transcription-dependent condensates. Science 361, 412–415 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Sabari, B. R. et al. Coactivator condensation at super-enhancers links phase separation and gene control. Science https://doi.org/10.1126/science.aar3958 (2018).
Schneider, N. et al. Liquid-liquid phase separation of light-inducible transcription factors increases transcription activation in mammalian cells and mice. Sci. Adv. https://doi.org/10.1126/sciadv.abd3568 (2021).
Liu, J. et al. CRISPR-assisted transcription activation by phase separation proteins. Protein Cell, https://doi.org/10.1093/procel/pwad013 (2023).
Chong, S. et al. Tuning levels of low-complexity domain interactions to modulate endogenous oncogenic transcription. Mol. Cell 82, 2084–2097.e2085 (2022).
Article CAS PubMed Google Scholar
Trojanowski, J. et al. Transcription activation is enhanced by multivalent interactions independent of phase separation. Mol. Cell 82, 1878–1893.e1810 (2022).
Article CAS PubMed Google Scholar
Xiang, S. et al. The LC domain of hnRNPA2 adopts similar conformations in hydrogel polymers, liquid-like droplets, and nuclei. Cell 163, 829–839 (2015).
Article CAS PubMed PubMed Central Google Scholar
Wei, M. T. et al. Nucleated transcriptional condensates amplify gene expression. Nat. Cell Biol. 22, 1187–1196 (2020).
Article CAS PubMed Google Scholar
Tziortzouda, P., Van Den Bosch, L. & Hirth, F. Triad of TDP43 control in neurodegeneration: autoregulation, localization and aggregation. Nat. Rev. Neurosci. 22, 197–208 (2021).
Article CAS PubMed Google Scholar
Lu, H. et al. Phase-separation mechanism for C-terminal hyperphosphorylation of RNA polymerase II. Nature 558, 318–323 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Liang, M. et al. Oligomerized liprin-alpha promotes phase separation of ELKS for compartmentalization of presynaptic active zone proteins. Cell Rep. 36, 109476 (2021).
Article CAS PubMed Google Scholar
Gao, Y. et al. Multivalent m(6)A motifs promote phase separation of YTHDF proteins. Cell Res. 29, 767–769 (2019).
Article CAS PubMed PubMed Central Google Scholar
Fu, Y. & Zhuang, X. m(6)A-binding YTHDF proteins promote stress granule formation. Nat. Chem. Biol. 16, 955–963 (2020).
Article CAS PubMed PubMed Central Google Scholar
Celetti, G. et al. The liquid state of FG-nucleoporins mimics permeability barrier properties of nuclear pore complexes. J. Cell Biol. https://doi.org/10.1083/jcb.201907157 (2020).
Ambadipudi, S., Biernat, J., Riedel, D., Mandelkow, E. & Zweckstetter, M. Liquid-liquid phase separation of the microtubule-binding repeats of the Alzheimer-related protein Tau. Nat. Commun. 8, 275 (2017).
Article ADS PubMed PubMed Central Google Scholar
Chen, Y. et al. Repurposing type I-F CRISPR-Cas system as a transcriptional activation tool in human cells. Nat. Commun. 11, 3136 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Bracha, D. et al. Mapping local and global liquid phase behavior in living cells using photo-oligomerizable seeds. Cell 175, 1467–1480.e1413 (2018).
Article CAS PubMed PubMed Central Google Scholar
Liu, Z. et al. Par complex cluster formation mediated by phase separation. Nat. Commun. 11, 2266 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Zeng, M. et al. Reconstituted postsynaptic density as a molecular platform for understanding synapse formation and plasticity. Cell 174, 1172–1187.e1116 (2018).
Article CAS PubMed Google Scholar
Shen, Z. et al. Biological condensates form percolated networks with molecular motion properties distinctly different from dilute solutions. Elife, https://doi.org/10.7554/eLife.81907 (2023).
Baron, M. K. et al. An architectural framework that may lie at the core of the postsynaptic density. Science 311, 531–535 (2006).
Article ADS CAS PubMed Google Scholar
Zhang, Q. et al. Visualizing dynamics of cell signaling in vivo with a phase separation-based kinase reporter. Mol. Cell 69, 334–346.e334 (2018).
Article MathSciNet CAS PubMed PubMed Central Google Scholar
Carter, D., Chakalova, L., Osborne, C. S., Dai, Y. F. & Fraser, P. Long-range chromatin regulatory interactions in vivo. Nat. Genet 32, 623–626 (2002).
Article CAS PubMed Google Scholar
Morgan, S. L. et al. Manipulation of nuclear architecture through CRISPR-mediated chromosomal looping. Nat. Commun. 8, 15993 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Wagh, K., Garcia, D. A. & Upadhyaya, A. Phase separation in transcription factor dynamics and chromatin organization. Curr. Opin. Struct. Biol. 71, 148–155 (2021).
Article CAS PubMed PubMed Central Google Scholar
Moses, D., Ginell, G. M., Holehouse, A. S. & Sukenik, S. Intrinsically disordered regions are poised to act as sensors of cellular chemistry. Trends Biochem. Sci. 48, 1019–1034 (2023).
Article CAS PubMed Google Scholar
Lyons, H. et al. Functional partitioning of transcriptional regulators by patterned charge blocks. Cell 186, 327–345.e328 (2023).
Article CAS PubMed PubMed Central Google Scholar
Linden, M. et al. FET family fusion oncoproteins target the SWI/SNF chromatin remodeling complex. EMBO Rep. https://doi.org/10.15252/embr.201845766 (2019).
Kim, Y. J. et al. Light-activated macromolecular phase separation modulates transcription by reconfiguring chromatin interactions. Sci. Adv. 9, eadg1123 (2023).
Article CAS PubMed PubMed Central Google Scholar
Lee, J. H. et al. Enhancer RNA m6A methylation facilitates transcriptional condensate formation and gene activation. Mol. Cell 81, 3368–3385.e3369 (2021).
Article CAS PubMed PubMed Central Google Scholar
van der Lee, R. et al. Classification of intrinsically disordered regions and proteins. Chem. Rev. 114, 6589–6631 (2014).
Article PubMed PubMed Central Google Scholar
Kato, M. et al. Cell-free formation of RNA granules: low complexity sequence domains form dynamic fibers within hydrogels. Cell 149, 753–767 (2012).
Article CAS PubMed PubMed Central Google Scholar
Wright, P. E. & Dyson, H. J. Intrinsically disordered proteins in cellular signalling and regulation. Nat. Rev. Mol. Cell Biol. 16, 18–29 (2015).
Article CAS PubMed PubMed Central Google Scholar
Lafontaine, D. L. J., Riback, J. A., Bascetin, R. & Brangwynne, C. P. The nucleolus as a multiphase liquid condensate. Nat. Rev. Mol. Cell Biol. 22, 165–182 (2021).
Article CAS PubMed Google Scholar
Ma, S. et al. Phase-separated DropCRISPRa platform for efficient gene activation in mammalian cells and mice. Nucleic Acids Res. 51, 5271–5284 (2023).
Article CAS PubMed PubMed Central Google Scholar
Wan, Y. et al. Dynamic imaging of nascent RNA reveals general principles of transcription dynamics and stochastic splice site selection. Cell 184, 2878–2895.e2820 (2021).
Article CAS PubMed PubMed Central Google Scholar
Donovan, B. T., Chen, H., Jipa, C., Bai, L. & Poirier, M. G. Dissociation rate compensation mechanism for budding yeast pioneer transcription factors. Elife https://doi.org/10.7554/eLife.43008 (2019).
Wu, J. et al. Modulating gene regulation function by chemically controlled transcription factor clustering. Nat. Commun. 13, 2663 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Klann, T. S. et al. CRISPR-Cas9 epigenome editing enables high-throughput screening for functional regulatory elements in the human genome. Nat. Biotechnol. 35, 561–568 (2017).
Article CAS PubMed PubMed Central Google Scholar
Li, K. et al. Interrogation of enhancer function by enhancer-targeting CRISPR epigenetic editing. Nat. Commun. 11, 485 (2020).
Article ADS PubMed PubMed Central Google Scholar
Hilton, I. B. et al. Epigenome editing by a CRISPR-Cas9-based acetyltransferase activates genes from promoters and enhancers. Nat. Biotechnol. 33, 510–517 (2015).
Article CAS PubMed PubMed Central Google Scholar
Ahn, J. H. et al. Phase separation drives aberrant chromatin looping and cancer development. Nature 595, 591–595 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, J. et al. Phase separation of OCT4 controls TAD reorganization to promote cell fate transitions. Cell Stem Cell 28, 1868–1883.e1811 (2021).
Article CAS PubMed Google Scholar
Ke, M. et al. Integrated and quantitative proteomic approach for charting temporal and endogenous protein complexes. Anal. Chem. 90, 12574–12583 (2018).
Article CAS PubMed Google Scholar
Kim, D., Langmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12, 357–360 (2015).
Article CAS PubMed PubMed Central Google Scholar
Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
Article CAS PubMed Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This work was supported by the National Key R&D Program of China (Grant No. 2021YFF1201000 and 2022YFC3400400 to W.C.), the Shenzhen Science and Technology Program (Grant No. KQTD20180411143432337 to W.C.) and the Shenzhen Key Laboratory of Gene Regulation and Systems Biology (Grant No. ZDSYS20200811144002008 to W.C.). We thank Dr. Zhou Songyang (SYSU) for providing the GFP reporter cells, Dr. Guanhua Bai (SUSTech) for the technical support, Dr. Lin Deng (SZBL), Dr. Yang Yu (IBP), and Dr. Cong Yu (SUSTech) for sharing plasmids. We also thank Dr. Mingjie Zhang (SUSTech) for helpful discussion on the project and review of the manuscript. We further thank the Center for Computational Science and Engineering of SUSTech for the support on computational resource.

Author information

Authors and Affiliations

Shenzhen Key Laboratory of Gene Regulation and Systems Biology, School of Life Sciences, Southern University of Science and Technology, Shenzhen, China
Rui Chen, Xinyao Shi, Xiangrui Yao, Tong Gao, Guangyu Huang, Duo Ning, Zemin Cao, Youxin Xu, Weizheng Liang, Simon Zhongyuan Tian, Qionghua Zhu, Liang Fang, Meizhen Zheng, Huanhuan Cui & Wei Chen
Innovative Center for RNA Therapeutics (ICRT), School of Life Sciences, Southern University of Science and Technology, Shenzhen, China
Rui Chen, Simon Zhongyuan Tian, Qionghua Zhu, Liang Fang, Meizhen Zheng, Huanhuan Cui & Wei Chen
Department of Systems Biology, School of Life Sciences, Southern University of Science and Technology, Shenzhen, China
Rui Chen, Xinyao Shi, Xiangrui Yao, Tong Gao, Guangyu Huang, Duo Ning, Zemin Cao, Youxin Xu, Weizheng Liang, Simon Zhongyuan Tian, Qionghua Zhu, Liang Fang, Meizhen Zheng, Huanhuan Cui & Wei Chen
Guangming Advanced Research Institute, Southern University of Science and Technology, Shenzhen, China
Rui Chen, Qionghua Zhu, Liang Fang, Huanhuan Cui & Wei Chen
Central Laboratory, The First Affiliated Hospital of Hebei North University, Zhangjiakou, Hebei, China
Weizheng Liang
Department of Pharmacology, School of Medicine, Southern University of Science and Technology, Shenzhen, China
Yuhui Hu
Joint Laboratory of Guangdong-Hong Kong Universities for Vascular Homeostasis and Diseases, School of Medicine, Southern University of Science and Technology, Shenzhen, China
Yuhui Hu

Authors

Rui Chen
View author publications
You can also search for this author in PubMed Google Scholar
Xinyao Shi
View author publications
You can also search for this author in PubMed Google Scholar
Xiangrui Yao
View author publications
You can also search for this author in PubMed Google Scholar
Tong Gao
View author publications
You can also search for this author in PubMed Google Scholar
Guangyu Huang
View author publications
You can also search for this author in PubMed Google Scholar
Duo Ning
View author publications
You can also search for this author in PubMed Google Scholar
Zemin Cao
View author publications
You can also search for this author in PubMed Google Scholar
Youxin Xu
View author publications
You can also search for this author in PubMed Google Scholar
Weizheng Liang
View author publications
You can also search for this author in PubMed Google Scholar
Simon Zhongyuan Tian
View author publications
You can also search for this author in PubMed Google Scholar
Qionghua Zhu
View author publications
You can also search for this author in PubMed Google Scholar
Liang Fang
View author publications
You can also search for this author in PubMed Google Scholar
Meizhen Zheng
View author publications
You can also search for this author in PubMed Google Scholar
Yuhui Hu
View author publications
You can also search for this author in PubMed Google Scholar
Huanhuan Cui
View author publications
You can also search for this author in PubMed Google Scholar
Wei Chen
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

W.C. and H.C. conceived the study. W.C., H.C., and R.C. designed the experiments. R.C., X.S., X.Y., T.G., G.H., D.N., Z.C., Y.X., and W.L. performed experiments. R.C. and S.Z.T. analyzed the NGS data. W.C., H.C., and R.C. wrote the manuscript. Y.H., M.Z., L.F., and Q.Z. discussed the results and revised the manuscript. W.C. and H.C. coordinated and supervised the study.

Corresponding authors

Correspondence to Huanhuan Cui or Wei Chen.

Ethics declarations

Competing interests

R.C., X.S., H.C., and W.C. have submitted a patent application related to this work to the China National Intellectual Property Administration (application number 202210903472.2). The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Reporting Summary

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Chen, R., Shi, X., Yao, X. et al. Specific multivalent molecules boost CRISPR-mediated transcriptional activation. Nat Commun 15, 7222 (2024). https://doi.org/10.1038/s41467-024-51694-y

Download citation

Received: 14 January 2024
Accepted: 15 August 2024
Published: 22 August 2024
DOI: https://doi.org/10.1038/s41467-024-51694-y

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.