PRION https://doi.org/10.1080/19336896.2021.1961569 REVIEW Decoding the role of coiled-coil motifs in human prion-like proteins Molood Behbahanipour, Javier García-Pardo , and Salvador Ventura * Institut De Biotecnologia I De Biomedicina (Ibb) and Departament De Bioquímica I Biologia Molecular, Universitat Autónoma De Barcelona, Barcelona, Spain ABSTRACT ARTICLE HISTORY Prions are self-propagating proteins that cause fatal neurodegenerative diseases in humans. Received 3 July 2021 However, increasing evidence suggests that eukaryotic cells exploit prion conformational conver­ Revised 22 July 2021 sion for functional purposes. A recent study delineated a group of twenty prion-like proteins in Accepted 25 July 2021 humans, characterized by the presence of low-complexity glutamine-rich sequences with over­ KEYWORDS lapping coiled-coil (CCs) motifs. This is the case of Mediator complex subunit 15 (MED15), which is Prions; amyloids; prion-like overexpressed in a wide range of human cancers. Biophysical studies demonstrated that the domains; coiled-coils; prion-like domain (PrLD) of MED15 forms homodimers in solution, sustained by CCs interactions. protein aggregation Furthermore, the same coiled-coil (CC) region plays a crucial role in the PrLD structural transition to a transmissible β-sheet amyloid state. In this review, we discuss the role of CCs motifs and their contribution to amyloid transitions in human prion-like domains (PrLDs), while providing a comprehensive overview of six predicted human prion-like proteins involved in transcription, gene expression, or DNA damage response and associated with human disease, whose PrLDs contain or overlap with CCs sequences. Finally, we try to rationalize how these molecular signatures might relate to both their function and involvement in disease. INTRODUCTION Mendelian manner. This atypical mechanism of protein transmission provides increased tolerance to stress and Prusiner and co-workers coined the term prion in 1982. facilitates adaptation to changing environments [5,6]. They reported the purification of infectious particles A less recognized feature of PrDs is the presence of from scrapie-infected hamster brains for the first time protein sequence stretches with a propensity to form and demonstrated that they consisted of a specific type coiled-coils (CCs), often overlapping with the Q/N-rich of pathogenic element that causes fatal neurodegenera­ or polyQ regions [3,7]. This super secondary structure tive diseases in mammals and humans [1,2]. It has been has been associated with the formation of dimers, tri­ assumed that the toxicity of prions relies on their mers, and higher-order oligomeric structures. intrinsic capacity to lose their native conformation Furthermore, it has been proposed that coiled-coil (either in ordered or disordered regions) and acquire (CC) formation is an essential step preceding the phy­ a β sheet-rich secondary structure with significant siological conformational switches to amyloid-like aggregation and self-propagating capacities. However, states in yeast prions [3,7]. almost forty years after this seminal discovery, increas­ Our group recently exploited different bioinfor­ ing evidence indicates that prion-like conformational matics tools to systematically analyse the presence of conversion is not always pathogenic. On the contrary, it prion-like domains (PrLDs) at the proteome level can be exploited for functional roles [3]. The best- [2,3,8]. Using this approach, we identified a subset of characterized examples of non-pathogenic prions are human polypeptides that bear PrLDs with properties those identified in yeast and filamentous fungi. Both resembling those of yeast PrDs [3,8]. Subsequently, we yeast and fungal prions are multi-domain proteins with explored whether these human domains contain regions enriched in Asn (N) and/or Gln (Q) residues regions with the propensity to fold into CCs using the [2,4]. Typically, these long Q/N-rich sequences map COILS [9] and PARCOIL2 [10] algorithms. A total of into unstructured regions of the protein termed prion 22 of these PrLDs were predicted to display sequences domains (PrDs). Upon a conformational switch to with a high propensity to fold into CCs, with 20 of a self-perpetuating conformation, the prionic traits them presenting polyQ tracts of different lengths. can be inherited during cell division in a non- *CONTACT Salvador Ventura Salvador.Ventura@uab.es; Javier García-Pardo javiergarciapardo@msn.com Institut De Biotecnologia I De Biomedicina (Ibb) and Departament De Bioquímica I Biologia Molecular, Universitat Autónoma De Barcelona, 08193 Bellaterra, Barcelona, Spain © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 144 M. BEHBAHANIPOUR ET AL. Interestingly enough, we observed that the majority of low complexity regions to their physiological partners the identified proteins, if not all, participate in tran­ is often accompanied by a local increase in the struc­ scriptional regulation. This is the case of the mediator turation of the binding region in a process known as complex subunit 15 (MED15). MED15 is part of the ‘folding upon binding’. This binding and folding mediator multi-protein complex that regulates enhan­ mechanism has been previously reported for different cer-driven gene transcription [3] and, under normal CCs, such as the GCN4 CC dimerization domain. conditions, is located in the nucleus [3]. Thus, This domain shows a two-state unfolding transition MED15 knockdown has been linked to reduced growth in which the binding appears coupled to folding [12]. and decreased transcriptional activation, while MED15 Similarly, thermodynamic analyses with the CC oli­ is highly overexpressed in different human cancers, gomerization domain (SARAH) from serine/threo­ including head and neck squamous cell carcinomas, nine mammalian sterile 20-like kinase (MST1) hepatocellular carcinoma, breast cancer, renal cell car­ demonstrated that this domain is unstructured and cinoma, and testicular germ cell tumours. Indeed, folds upon binding to different partners [13,14]. patients with MED15 over-expression in tumour tissues However, other structural studies of the GCNA4 CC exhibit bad prognosis, significantly shorter survival indicated that a pre-existing helical content promotes times, and more aggressive phenotypes [11]. MED15 complex formation [14–16], which shows that seg­ contains a disordered N-terminal low-complexity ments of intrinsic helical propensity are important region composed of discontinuous polyQ tracts and drivers of the interaction. As mentioned, a strong propensity to form CCs that map to its PrLD. a significant proportion of human prion-like proteins Recent work has focused on elucidating the role of display sequences with a high propensity to fold into the CC motif for structural transitions of initially solu­ CCs that overlap with low complexity sequences. ble α-helical regions to β-sheet amyloid states. This has Thus, it is tempting to speculate that the presence of been particularly challenging to study in the past in the these preformed helical regions may speed up the context of complete PrLDs, and previous reports dealt binding to their partners. After this initial contact, with only relatively short peptides derived from yeast further local folding may still take place within the prions (i.e., Ure2p PrD). Recently, Batlle and co- complex. workers have experimentally demonstrated, for the first time, that is, in fact, the CC motif of MED15 COILED COILS MEDIATE FUNCTIONAL PrLD that mediates the transition towards a β-sheet INTERACTIONS AND AMYLOID FORMATION amyloid state [3]. They proposed that this behaviour may prevent the establishment of relevant protein- Sequences with Poly-Q repeat motifs are known to have protein interactions in pathological situations. a high propensity to form CCs. These sequences are Alternatively, MED15 self-association might result in frequent in yeast prions [4] and human proteins bear­ phase separation and the consequent formation of tran­ ing PrLDs [3,7]. CC motifs were traditionally consid­ scriptional hubs. A third possibility is that the PrLD of ered as molecular spacers between functional domains MED15 would establish pathological interactions with [17]. However, a growing body of evidence suggests other Q-rich proteins, leading to the formation of cyto­ that they frequently contain interaction regions and plasmatic insoluble aggregates. All in all, these recent act as protein-protein interactors (PPI) and/or catalytic results reinforce the hypothesis that aggregation of effectors [17]. CC forming proteins are widely distrib­ MED15 PrLD into amyloid fibrils involves uted and have been shown to play a variety of biological a progressive conformational switch rather than roles [17,18], such as transcription regulation (i.e., leu­ a simple uncontrolled misfolding mechanism. It is fea­ cine zippers) [19]; modulation of chromatin [20]; chro­ sible that the exact mechanism applies to the rest of the mosome dynamics [21]; kinetochore assembly [22]; cell 19 identified polypeptides. Here we discuss this possi­ cycle progression [23]; organization of the centrosome bility and its potential implications for six of these [24]; vesicle transport [25], organelles structuration and predicted prion-like proteins associated with human activation of cell-signalling cascades [26,27], among disease. other functions. Examples of these functions are depicted in Figure 1. CC motifs are particularly abun­ dant in RNA-binding proteins (RBPs) that are localized LOW COMPLEXITY SEQUENCES IN PRION-LIKE to liquid-liquid phase-separated (LLPS) neuronal gran­ PROTEINS ules [18]. This phase transition has been shown essen­ Prion-like proteins are rich in low complexity tial for appropriate RNA trafficking during local sequences. Recent work has shown that binding of protein synthesis, but it is also associated with PRION 145 Figure 1. Structural features of coiled-coils built up by Poly-Q repeats. a) Helical wheel illustration for a regular heptad pattern composed built up by 21 Gln residues. The Gln residues are found in all seven (a-b-c-d-e-f-g) positions of the heptad repeat due to the particular chemical nature of this amino acid. b) Modelled structure of a canonical CC composed of 42 Gln residues. c) Structural representation of the oligomerization states driven by the supercoiling of the α-helices in a parallel orientation. d,e) Examples of functional protein-DNA and protein-protein interactions mediated by CCs. d) X-ray crystal structure of the leucine zipper Pap1 bZIP dimer bound to DNA. e) Crystal Structure of a diUb chain bound to the NEMO-UBAN domain. PDB codes are: 1GD2 and 2ZVN for d) and e), respectively. misfolding and amyloid formation in various human given appropriate stimulus or induced by changes in neurological disorders [18]. the environmental conditions. In support of this The structure of CCs is now well understood, and hypothesis, Hartmann and co-workers reported that this fold is ubiquitous in all domains of life (comprising insertions of two or six residues strain the supercoil up to 10% of all proteins in a given species) [25]. and lead to the local formation of β-strands (also Naturally occurring CCs consist of at least two super­ known as α-β coiled-coil) [28]. Additionally, experi­ coiled α-helices. In their most prevalent form, they are mental data indicate a possible direct structural conver­ built up from a heptad repeat abcdefg, in which the sion of CCs themselves to cross β-sheet amyloids. This amino acids at positions a and d are mostly conserved structural shift was demonstrated in short peptides that hydrophobic residues (see Figure 1a). The resultant α- form α-helical CCs when subjected to heating or pH helices display a high amphipathic character, with both changes [7], implying a structural shift from the intra­ hydrophobic and polar faces. Yeast PrDs and human molecular hydrogen-boding network that maintains PrLDs possess sequence stretches in which Gln residues CCs to the typical intermolecular-bonding landscape are abundant. Gln is both neutral and polar amino acid, present in the β-sheets amyloids [29]. with a long side chain that can be located on the surface As discussed above, a recent study by our group of α-helices and at the theoretically hydrophobic a/d supports this mechanism being behind the formation positions of the heptad repeats (see Figure 1a and of highly ordered amyloids by the PrLDs of MED15. Figure 1b). This dual behaviour allows for a variety of Other human proteins that may respond to the exact energetically favourable oligomerization states driven mechanism were discovered in the same study but by the supercoiling of the α-helices (Figure 1c). Thus, could not be discussed in detail. In the following sec­ it has been proposed that CC-mediated oligomerization tions, we describe the cases of TATA-box-Binding of Q-rich proteins could drive to amyloid formation, Protein (TBP), cAMP-response element-protein 146 M. BEHBAHANIPOUR ET AL. -(CREB) binding protein (CBP), ataxin-1 (ATXN-1), transcriptional activity regulation [33]. Detailed ataxin-8 (ATXN-8), Lysine Methyltransferase 2D sequence analysis of TBP with PLAAC, a program (KMT2D), and Forkhead box protein P2 (FOXP2), as aimed to identify PrLDs in protein sequences [34], representatives of this set of polypeptides. revealed the presence of a 120-residues N-terminal segment with prion-like amino acid composition (Figure 2a and Figure 2b). According to COILS [9] TATA-BOX-BINDING PROTEIN (TBP) and PARCOIL2 [10], two algorithms intended to iden­ The TATA-box-binding protein (TBP) is an integral tify sequences with a high CC propensity, the component of the transcription initiation complex N-terminal tail of TBP also encodes a sequence of required by all three eukaryotic RNA polymerases about 45 residues with a high propensity to form [30]. TBP contains a highly conserved core domain at a CC that overlaps with the Q-rich and PrLD motifs the C-terminal region that mediates many of its tran­ (see Figure 2a and Figure 2c-d). This region comprises scriptionally relevant interactions in eukaryotes six predicted continuous heptad repeats (Figure 2a and (Figure 2a). This C-t region has a symmetric structure Figure 2e) built mainly by Gln residues (78%). with two well-conserved TBP domains that mediate Although the function of the PrLD domain from binding to the targeted DNA [31]. On the contrary, TBP remains unclear, it may likely play a role in reg­ the N-terminal tail of TBP is evolutionarily divergent ulating TBP function, since the N-terminal region and modulates the DNA-binding ability of its where it resides can repress the ability of the TBP to C-terminal part [32]. Interestingly, the N-terminal seg­ bind the TATA box and at the same time can permit ment contains a low complexity region with a Q-rich cooperative binding with other basal factors. motif (that varies in length from 25 to 42 residues in Detection of the TATA-BOX by TBP is an essential healthy individuals) that is thought to be involved in step for the formation of a transcription initiation com­ Figure 2. TBP contains a Q-rich coiled-coil PrLD. a) Linear representation of human TBP, showing the location of the N-t Q-rich sequence and the pair of C-t TATA-box-binding domains. b) The predictions for the prion-like domain (red) identified using PLAAC (30) predictor. CC per-residue probability of TPB calculated using c) COILS [9] and d) PARCOIL2 [10]. d) Wheel representation of the heptad repeats (a-b-c-d-e-f-g) of TBP. The sequence corresponding to the soft amyloid core (SAC) of TPB predicted by pWALTZ [45] is indicated. PRION 147 plex. In this process, TBP binds to TATA-BOX after the CC motif and connects with the adjacent sequences in two different steps. The first step requires doublet of TBP domains (Figure 2a). The presence of the interaction of the inhibitory DNA-binding (IDB) this SAC might facilitate amyloid formation by decreas­ surface from the C-terminal part of TBP to positively ing the energy barrier for the conformation transition, regulate the formation of the unstable DNA unbent with the adjacent CC region facilitating the subsequent complex. The second step involves a conformational intermolecular contacts, as demonstrated for switch that leads to forming a stable DNA bent com­ MED15 [3]. plex [35]. The N-terminal region inhibits the formation In summary, TPB plays an essential role in tran­ of the bent complex and favours that of the unbent one, scription initiation, and the presence of PrLD in this and conformational changes in this region are neces­ protein might be important for this function, through sary to form the active assembly. its binding to DNA or by its interaction and regulation Recent cryo-EM structural studies have revealed of/by other transcriptional players. The questions are if, many details concerning the early steps of TFIID and how, mutations/expansions at its Q-rich region assembly [36]. TBP interacts with up to 13 different might promote amyloid conversion and which would TBP-associated factors (TAFs) to build up the TFIID be the role of CCs in this putative transition. complex, with six of them (TAF4, TAF5, TAF6, TAF9, Answering them, should help to understand the con­ TAF10, and TAF12) present in two copies. This large tribution of the PrLD identified in TBP to disease onset macromolecular complex of about 1.3 MDa is essential and progression. for recognizing core promoter sequences and recruiting the preinitiation complex (PIC) during basal transcrip­ cAMP-RESPONSE ELEMENT-BINDING PROTEIN (CBP) tion. Our previous analysis of TBP-associated factors (TAFs) revealed an overrepresentation of Q-rich cAMP-response element-protein-(CREB) binding pro­ sequences and overlapping CCs motifs [3]. This trend tein (CBP) is a transcriptional coactivator localized in has been previously described by Fiumara and co- the cell nucleus. CBP has a lysine acetyltransferase workers in yeast, who reported that many of the protein domain (HAT in Figure 3a) located in the central interactors of yeast functional prions contain CCs [7]. region of the protein and catalyzes the acetylation of For instance, human TAF4 contains four glutamine- target proteins [46]. CBP interacts with a variety of cell rich (Q-rich) domains involved in mediating interac­ signalling proteins, especially those with established tions with transcriptional activators. pro-survival effects in neurons, such as the CRE- TBP has also been linked to the key pathological binding protein (CREB) [47] and plays a crucial role features of some neurodegenerative disorders, includ­ in critical biological processes, such as embryogenesis, ing neuronal intranuclear hyaline inclusion disease development, differentiation, and apoptosis [48]. (NIH-ID) [37], spinocerebellar ataxia types 1, 2, 3 Thanks to its HAT motif, CBP is endowed with histone (SCA1, SCA2, SCA3) [38,39], dentatorubral- acetyltransferase activity, promoting the acetylation of pallidoluysian atrophy (DRPLA) [38], Huntington’s histones, which influences chromatin condensation and disease (HD) [40], and Alzheimer’s disease (AD) [41]. is a key mechanism in regulating transcription [46]. On In particular, expansion of the polyQ tract beyond 42 the other hand, the N- and C-terminal domains of CBP residues in TBP has been associated with the develop­ can act as transactivation domains, and the protein also ment of cerebellar ataxia (SCA17), an autosomal domi­ contains three potential α-helical motifs containing the nant and progressive neurodegenerative disease [42]. sequence LXXLL (amino acids 68–78, 355–365, and Pathological expansion in its polyQ domain has been 2067–2077) called NR boxes, to mediate interactions proposed to induce neurodegeneration. This expansion with nuclear receptors. The C-terminal NR box lies caused TBP aggregation and precipitation within neu­ within the C-terminal low complexity region of CBP rons [43,44], decreased its dimerization, altered its and interacts with other coactivators, including activa­ binding to other transcription factors. This behaviour tor for thyroid hormone and retinoid receptors (ACTR) is compatible with the presence of sequence stretches and the steroid receptor coactivator 1 (SRC-1) [49–51]. able to assemble into highly order amyloid-like struc­ Detailed sequence analysis of CBP showed that this tures. The emerging picture is that these so-called soft protein contains up to six discontinuous segments with amyloid cores (SAC), are necessary for prion conver­ prion-like features (Figure 3a). In fact, the C-terminal sion in yeast as well as in human cells [3]. We identified PrLDs identified in this protein overlap with two pre­ a similar region in TBP that comprises residues 93- dicted CC motifs and with a tract of 18 consecutive Gln QQQQAVAAAAVQQSTSQQATQ-112 by using residues near its carboxy-terminal tail (residues 2199– pWALTZ [45]. This sequence is located immediately 2216). This polyQ motif has been involved in the 148 M. BEHBAHANIPOUR ET AL. Figure 3. Human proteins containing PrLDs and overlapping CC regions. a-e) Linear representation of a) CBP, b) ATXN1, c) ATXN8, d) KMT2D, and e) FOXP2 showing the location of the Pfam domains (green) (3). The prion-like domains (red) based on PLAAC [34] predictions (0.8–1 score) and the CC regions (yellow) using COILS [9] predictor are indicated (0.8–1 score). In a-e), the COILS and PLAAC per-residue probability plots have been included. f) Linear representation of other human prion-like proteins with over­ lapping CCs. Note that the COILS score (> 0.3) has been considered for AAK1 and NFAT5. sequestration of CBP to aggregates of several distinct for MED15, which might facilitate polyQ-mediated polyQ proteins [52]. Indeed, CBP dysfunction has been interactions and CBP aggregation. identified in a number of neurological disorders, such as the Rubinstein–Taybi syndrome (RTS), AD, Amyotrophic Lateral Sclerosis (ALS) and polyQ- ATAXIN-1 (ATXN1) AND ATAXIN-8 (ATXN8) related diseases, in which CBP has been found in the The spinocerebellar ataxias (SCAs) are a group of auto­ inclusion bodies formed by polyglutamine-containing somal dominantly inherited progressive disorders asso­ proteins in HD, DRPLA, SCA3, SCA7, and spinal and ciated with cerebellar degeneration and progressive bulbar muscular atrophy (SBMA) [52–56]. CBP deple­ ataxia [57]. Almost thirty different genetic causes of tion from its normal location and sequestration into SCAs have been reported so far and are numbered aggregates through interaction with expanded polyQ chronologically in order of discovery [57]. Most of proteins has been suggested to be mediated by its them have shown to progress with classic cerebellar C-terminal polyQ tract [52]. As a result, impaired CBP- signs [57]. However, many variants display disabling mediated transcription and the lack of CBP-dependent non-cerebellar features such as brainstem dysfunction, acetylation are associated with increased cell death eye movement abnormalities, and visual loss [57]. rates. It is tempting to propose that the C-terminal Ataxin-1 (ATXN1) is an RNA binding protein with region of CBP and specifically its Q-rich, CC- a role in transcriptional regulation. ATXN1 contains an containing, PrLD might be involved in RNA-binding motif (residues 541–767) comprising a conformational switch, similar to the one described a globular AXH domain (residues 568–689) involved PRION 149 in ATXN1 interactions. Predictions using PLAAC and protein. As shown in Figure 3c and according to COILS suggested that ATXN1 encodes prionic traits at PLAAC and COILS predictions, almost all the 80- the N-terminal region, with five CC heptad repeats residues of ATXN8 map to a PrLD with the potential overlapping with the detected PrLD (Figure 3b). to acquire a CC fold. It remains unclear how the struc­ Mutations in the ataxin-1 gene have been associated tural features of proteins bearing long polyQ stretches with the development of spinocerebellar ataxia type 1 are translated with toxic phenotypes. This might (SCA type 1). This protein is located in the nucleus and depend on the unique capacity of these proteins to interacts with RBM17 (a splicing factor) and several target other cellular proteins through CC-to-CC inter­ transcription regulators, including SMRT, HDAC3 an actions. The concept of CC-mediated aggregation also Capicua [58–60]. Wild-type ATXN1 has a length of 816 provides a plausible mechanism for polyQ-protein residues. The presence of a trinucleotide repeat disor­ deposition in the brain. der caused by the expansion of the CAG repeat in the N-terminal region of ATXN1 leads to an expanded LYSINE METHYLTRANSFERASE 2D (KMT2D) polyQ tract and disease development. This polyQ tract (residues 197–225) is located at the N-terminal region The KMT2D gene (also called MLL4, ALR, or MLL2) is of the protein, mapping at the predicted PrLD and CC a member of the SET family of histone methyltransfer­ motifs. This expansion in ATXN1 favours its binding to ase enzymes. KMT2D gene in human maps to the transcriptional repressor protein Capicua homolog 12q13.12, including over 19 kb pairs in length and via the RNA- binding protein and spliceosome compo­ contains up to 56 exons, encoding a large polypeptide nent RBP7. This aberrant binding leads to a disruption with 5537 amino acids. The transcript of this gene has in gene expression and splicing events in the affected a molecular weight of 593 kDa and catalyzes the methy­ neurons [57]. ATXN1 shuttles between the nucleus and lation (either mono-, di-, and trimethylation) on lysine cytoplasm. However, ATXN1 dynamics are altered by 4 (K4) of the histone H3 protein (H3K4). This is an the Gln expansion and failure in nuclear export upon important histone modification that controls transcrip­ polyQ expansion dramatically reduces the retro- tionally active promoters and enhancers [70], increas­ transport into the cytoplasm [61]. Regardless of the ing the transcription of the gene packaged around the polyQ expansion, subtle increases in wild type histones [71]. ATXN1 levels could also lead to SCA1 hallmarks [62], KMT2D contains a cluster of conserved domains at whereas decreasing ATXN1 expression helps to reduce the C-terminal end, comprising a PHD (plant home­ ATXN1’s accumulation and mitigates cerebellar SCA1 odomain)-zinc-finger like domain, two phenylalanine, pathogenesis [63–66]. Conversely, it also has been and tyrosine (FY)-rich motifs (F/Y-rich N-terminus shown that loss of the protein increases the levels of (FYRN) and F/Y-rich C-terminus (FYRC)) and BACE1 and Aβ pathology, indicating that ATXN1 a catalytic SET domain (SET: Su(var)3–9, Enhancer-of- levels is a potential factor for AD development [67]. zeste (E(z)), and Trithorax) that are needed for its Co-regulation by microRNAs (miRNAs) together with enzymatic function [72]. KMT2D is a major regulator the RNA-binding protein PUMILIO1 (PUM1) are of cell-type-specific gene expression in cell differentia­ shown to fine-tune posttranscriptionally ATXN1 tion during tissue development and embryogenesis mRNA levels [62,63,68]. Posttranslational modifica­ [73,74]. In addition, KMT2D has also been related to tions such as phosphorylation, transglutamination, ubi­ tumour suppression and immune signalling [74–76]. quitination, and sumoylation are also required for the In humans, KMT2D contains a high predicted dis­ ATXN1 function [60]. Furthermore, the N-terminal α- order content of (55%), a common characteristic of helical CC domain in the PrLD of MED15 has been nuclear proteins involved in transcription and chroma­ shown to efficiently promote spontaneous ATXN1 tin organization, with a series of polyQ tracts at the aggregation in vitro, implying that CC regions in modi­ C terminus end of the highly conserved central regions. fiers are essential to promote aggregation and toxicity These regions are part of a large low complexity seg­ effects on polyQ diseases [69]. ment, consisting of four discontinuous regions with Much less is known about ATXN8. Expansions in predicted prion-like features (Figure 3d), with the lar­ this protein have been linked to spinocerebellar ataxia gest prion-like region containing a set of three conse­ type 8 (SCA8) in humans. These polyQ expansions cutive CC motifs that overlap with the polyQ repeats. showed a reduced penetrance compared to the expan­ Most KMT2D mutations are assumed to cause trun­ sions observed for other SCA-associated disease genes cated proteins that do not perform the function prop­ [57]. A difference with ATXN1 is that the expansion on erly due to loss of the catalytic SET domain [72]. CAG repeats in ATXN8 leads to a nearly pure polyQ Mutated KMT2D has been linked to developmental 150 M. BEHBAHANIPOUR ET AL. disorders including Kabuki syndrome [77], congenital Q230) adjacent to it, with both Gln-rich regions dis­ heart disease [78], and multiple cancer types such as playing a high propensity to form CCs, according to medulloblastoma, lymphoma, hepatocellular carci­ COILS. Thus, FOXP2 shares all the sequential features noma, gastric cancer, breast cancer and prostate cancer shown to promote a CC to amyloid transition in the [73,74]. This pleiotropic effect results from the fact that case of MED15 and it can be considered a potential its pathogenic variants cause the interruption of histone functional human prion-like protein. methylation and abnormal enhancer regulation, leading to changes in transcription, thus affecting normal OTHER PREDICTED PRION-LIKE HUMAN growth and development [74]. The exon 39 (3581– PROTEINS WITH COILED-COILS 4510 aa) constitutes a mutational hotspot [79] and, interestingly enough, it contains the most extended As previously mentioned, a total of 20 human prion- prion-like domain, with missense disease-associated like domains (PrLDs) containing high CC propensity mutations clustering at the predicted CC and Q-rich sequences that, in the majority of cases, overlapped motifs comprising residues 3897–3975 [70]. Whether with polyQ stretches were identified in the human such mutations predispose to disease by disrupting proteome [3]. These predicted prion-like proteins normal protein-protein interactions, mediated by work in the regulation of transcription, and they com­ KMT2D during chromatin remodelling, or promoting prise key transcription coactivators, including MED12, a conformational transition towards an aggregated state MED15, MAML3 and MAML2, CBP, protein kinase remains to be elucidated. and ATP binding activities like BMP2K, AAK1 or DNA-binding transcription factors such as TBP, NFAT5, NCOA3, NCOA2 and FOXP2 (Figure 3f). FORKHEAD BOX PROTEIN P2 (FOXP2) For most of them, it is unknown whether they can FOXP2 is a multifunctional transcription factor access a prionic state and, in this case, if α-helical CC expressed in various brain regions and peripheral within PrLDs would be self-sufficient mediators of organs during embryonic development and adulthood functional prions aggregation or just intermediates or [80,81]. In humans, FOXP2 has 715 residues (80kDa) facilitators in the β-sheet formation process. In any and is one of the most highly conserved proteins in case, we foresee an increasing interest in studying the vertebrate genomes [82]. It has been shown to impact verisimilitude of these transitions, provided the func­ gene regulation in multiple aspects of neuronal devel­ tional relevance of this prion-like proteins subset. opment [83–85].Indeed, Forkhead-box protein P2 (FOXP2) was one of the first genes to be linked to CONCLUSIONS human language disorder, characterized by a broader cognitive dysfunction and primary motor impairment. Recent observations suggest that the coincidence of low Thus, this protein is essential for the normal develop­ complexity regions and CC-forming sequences in the ment of speech and language [86–88]. Human onco­ PrLDs of prion-like proteins is a frequent feature that genesis has also been linked to dysregulated FOXP2 might be critical for their function. In humans, these function [89]. proteins are key functional mediators in transcription FOXP2 contains different sequence motifs, including and are associated with divergent disorders such as neu­ a single C2H2 zinc finger (C2H2-ZF) motif and rodegenerative diseases or cancer. Although only a leucine zipper (LZ) motif that mediates FOXP2 a reduced set of functional prions have been deeply dimerization through CC formation. This domain has characterized, compared to the vast number of patholo­ a critical role in DNA binding and promotes hetero­ gic amyloid proteins, some specific traits can already be typic and homotypic protein interactions with FOXP1/ delineated. A feature common to almost all prion-like 2/4 family members [90,91]. In addition, FOXP2 has proteins is the spatial separation of the active globular a strongly conserved forkhead DNA-binding domain and prionic domains. The prionic traits are concentrated (FHD, 100-aa or ‘winged-helix”) at the C-terminus in low complexity regions, which, as described here, that can form a domain-swapped dimer, and two might display a high propensity to form CCs and mediate nuclear localization signals (NLS) [92,93]. As shown protein-protein interactions, either directly or through in Figure 3e, a long and disordered predicted PrLD a folding-upon-binding mechanism. Many human prion- lies near the N-terminus in human FOXP2. It overlaps like proteins interact with DNA and/or RNA, a property with a glutamine-rich region, that contains a long per­ mediated by their globular domains, and that for some of fect polyQ repeat made of up to 40 CAG/CCG repeats them is associated with their liquid-liquid phase separa­ (p.Q152-Q191) and a shorter imperfect repeat (p.Q200- tion and the formation of membraneless compartments. PRION 151 Although it is assumed that pathological conditions result ORCID from the inherent toxicity acquired by the aggregates of Javier García-Pardo http://orcid.org/0000-0001-9179-6371 these proteins, our results with MED15 suggest an alter­ Salvador Ventura http://orcid.org/0000-0002-9652-6351 native explanation in which cytotoxicity is associated with the sequestration of the protein in the wrong com­ partment, its inactivation upon clustering with References a consequent loss of crucial protein-protein interactions [1] Prusiner SB. Novel proteinaceous infectious particles or conversely the establishment of undesired intermole­ cause scrapie. Science. 1982;216(4542):136–144. cular interactions with dysregulate signalling pathways. [2] Gil-Garcia M, Iglesias V, Pallares I, et al. Prion-like The CCs would be important players in these processes, proteins: from computational approaches to proteome- and indeed a significant number of pathological muta­ wide analysis. FEBS open bio. 2021 May 31.1-12. doi: 10.1002/2211-5463.13213 tions map into these domains, either decreasing or [3] Batlle C, Calvo I, Iglesias V, et al. MED15 prion-like increasing their helical propensity. domain forms a coiled-coil responsible for its amyloid Despite progress in the last decades in the field of conversion and propagation. Commun Biol. 2021;4 prions and amyloids, many questions remain to be (1):414. answered. For example, we still need to address how [4] Batlle C, Iglesias V, Navarro S, et al. Prion-like proteins human prion-like proteins participate in transcrip­ and their computational identification in proteomes. Expert Rev Proteomic. 2017;14(4):335–350. tional mechanisms under specific environmental con­ [5] Wickner RB, Edskes HK, Shewmaker F, et al. Prions of ditions and how dynamic are the stimuli that trigger fungi: inherited structures and biological roles. Nature their amyloid conversion in vivo. Another crucial Rev Microbiol. 2007;5(8):611–618. question is how the conformational changes in [6] Gemayel R, Chavali S, Pougach K, et al. Variable PrLDs occur and how they lead to amyloid forma­ glutamine-rich repeats modulate transcription factor activity. Mol Cell. 2015;59(4):615–627. tion. The structural flexibility of low-complexity [7] Fiumara F, Fioriti L, Kandel ER, et al. Essential role of regions and the CC-mediated transition model pro­ coiled coils for aggregation and activity of Q/N-rich vides a plausible mechanism to explain the functional prions and PolyQ proteins. Cell. 2010;143 to pathogenic conversions of at least a fraction of (7):1121–1135. these polypeptides. Providing molecular evidence for [8] Iglesias V, Paladin L, Juan-Blanco T, et al. In silico characterization of human prion-like proteins: beyond such conformational transition in the proteins dis­ neurological diseases. Front Physiol. 2019;10:314. cussed in this review might uncover an additional [9] Lupas A, Van Dyke M, Stock J. Predicting coiled coils layer of transcription regulation and help to under­ from protein sequences. Science. 1991;252 stand why these proteins are indefectibly associated (5009):1162–1164. with human disease. [10] McDonnell AV, Jiang T, Keating AE, et al. Paircoil2: improved prediction of coiled coils from sequence. Bioinformatics. 2006;22(3):356–358. [11] Wang KY, Duan CX, Zou XJ, et al. Increased mediator Disclosure statement complex subunit 15 expression is associated with poor prognosis in hepatocellular carcinoma. Oncol Lett. No potential conflict of interest was reported by the 2018;15(4):4303–4313. author(s). [12] Dragan AI, Privalov PL. Unfolding of a leucine zipper is not a simple two-state transition. J Mol Biol. 2002;321(5):891–908. Funding [13] Constantinescu Aruxandei D, Makbul C, Koturenkiene A, et al. Dimerization-induced folding This work was funded by the Spanish Ministry of Science and of MST1 SARAH and the influence of the intrinsically Innovation (PID2019-105017RB-I00) to S.V., by unstructured inhibitory domain: low thermodynamic ICREA, ICREA-Academia 2015 and 2020 to S.V and by stability of monomer. Biochemistry. 2011;50 EU (PhasAge /H2020-WIDESPREAD-2020-5) to SV. J.G.-P. (51):10990–11000. was supported by the Spanish Ministry of Science, Innovation [14] Walavalkar NM, Gordon N, Williams DC Jr. Unique and Universities with a postdoctoral grant Juan de la features of the anti-parallel, heterodimeric coiled-coil Cierva Incorporacion (IJC2019-041039-I). interaction between methyl-cytosine binding domain 2 (MBD2) homologues and GATA zinc finger domain containing 2A (GATAD2A/p66alpha). J Biol Chem. AUTHOR CONTRIBUTIONS 2013;288(5):3419–3427. [15] Myers JK, Oas TG. Reinterpretation of GCN4-p1 fold­ M.B. contributed to the preparation and writing of the manu­ ing kinetics: partial helix formation precedes dimeriza­ script. J.G.P and SV contributed to the conceptualization, tion in coiled coil folding. J Mol Biol. 1999;289 writing, and revision of the manuscript. (2):205–209. 152 M. BEHBAHANIPOUR ET AL. [16] Zitzewitz JA, Ibarra-Molero B, Fishel DR, et al. [33] Reid SJ, Rees MI, van Roon-Mom WM, et al. Preformed secondary structure drives the association Molecular investigation of TBP allele length:: reaction of GCN4-p1, a model coiled-coil system. J Mol a SCA17 cellular model and population study. Biol. 2000;296(4):1105–1116. Neurobiol Dis. 2003;13(1):37–45. [17] Truebestein L, Leonard TA. Coiled-coils: the long and [34] Lancaster AK, Nutter-Upham A, Lindquist S, et al. short of it. BioEssays. 2016;38(9):903–916. PLAAC: a web and command-line application to iden­ [18] Ford LK, Fioriti L. Coiled-coil motifs of RNA-binding tify proteins with prion-like amino acid composition. proteins: dynamicity in RNA regulation. Front Cell Bioinformatics. 2014;30(17):2501–2502. Dev Biol. 2020;8:1383. [35] Zhao X, Herr W. A regulated two-step mechanism of [19] Oshea EK, Rutkowski R, Kim PS. Evidence That the TBP binding to DNA: a solvent-exposed surface of Leucine Zipper Is a Coiled Coil. Science. 1989;243 TBP inhibits TATA box recognition. Cell. 2002;108 (4890):538–542. (5):615–627. [20] Miyaji-Yamaguchi M, Okuwaki M, Nagata K. Coiled- [36] Patel AB, Louder RK, Greber BJ, et al. Structure of coil structure-mediated dimerization of template acti­ human TFIID and mechanism of TBP loading onto vating factor-I is critical for its chromatin remodeling promoter DNA. Science. 2018;362(6421):6421. activity. J Mol Biol. 1999;290(2):547–557. [37] Takahashi J, Fukuda T, Tanaka J, et al. Neuronal intra­ [21] Matityahu A, Onn I. A new twist in the coil: functions nuclear hyaline inclusion disease with of the coiled-coil domain of structural maintenance of polyglutamine-immunoreactive inclusions. Acta chromosome (SMC) proteins. Curr Genet. 2018;64 Neuropathol. 2000;99(5):589–594. (1):109–116. [38] Uchihara T, Fujigasaki H, Koyano S, et al. Non- [22] Petrovic A, Keller J, Liu YH, et al. Structure of the expanded polyglutamine proteins in intranuclear inclu­ MIS12 complex and molecular basis of its interaction sions of hereditary ataxias–triple-labeling immuno­ with CENP-C at human kinetochores. Cell. 2016;167 fluorescence study. Acta Neuropathol. 2001;102 (4):1028-+. (2):149–152. [23] Patzke S, Hauge H, Sioud M, et al. Identification of [39] Perez MK, Paulson HL, Pendse SJ, et al. Recruitment a novel centrosome/microtubule-associated coiled-coil and the role of nuclear localization in protein involved in cell-cycle progression and spindle polyglutamine-mediated aggregation. J Cell Biol. organization. Oncogene. 2005;24(7):1159–1173. 1998;143(6):1457–1470. [24] Salisbury JL. Centrosomes: coiled-coils organize the [40] van Roon-Mom WM, Reid SJ, Jones AL, et al. cell center. Curr Biol. 2003;13(3):R88–90. Insoluble TATA-binding protein accumulation in [25] Rose A, Schraegle SJ, Stahlberg EA, et al. Coiled-coil Huntington’s disease cortex. Mol Brain Res. 2002;109 protein composition of 22 proteomes–differences and (1–2):1–10. common themes in subcellular infrastructure and traf­ [41] Reid SJ, van Roon-Mom WM, Wood PC, et al. TBP, fic control. BMC Evol Biol. 2005;5(1):66. a polyglutamine tract containing protein, accumulates [26] Maculins T, Garcia-Pardo J, Skenderovic A, et al. in Alzheimer’s disease. Mol Brain Res. 2004;125(1– Discovery of protein-protein interaction inhibitors by 2):120–128. integrating protein engineering and chemical screening [42] Nakamura K, Jeong S-Y, Uchihara T, et al. SCA17, platforms. Cell Chem Biol. 2020;27(11):1441–1451 e7. a novel autosomal dominant cerebellar ataxia caused [27] Rahighi S, Ikeda F, Kawasaki M, et al. Specific recogni­ by an expanded polyglutamine in TATA-binding tion of linear ubiquitin chains by NEMO is important protein. Hum Mol Genet. 2001;10(14):1441–1448. for NF-kappaB activation. Cell. 2009;136(6):1098–1109. [43] van Roon-Mom W, Reid S, Faull R, et al. TATA- [28] Hartmann MD, Mendler CT, Bassler J, et al. alpha/beta binding protein in neurodegenerative disease. coiled coils. eLife. 2016;5:e11861. Neuroscience. 2005;133(4):863–872. [29] Dong H, Hartgerink JD. Role of hydrophobic clusters [44] Nolte D, Sobanski E, Wissen A, et al. Spinocerebellar in the stability of alpha-helical coiled coils and their ataxia type 17 associated with an expansion of 42 conversion to amyloid-like beta-sheets. glutamine residues in TATA-box binding protein Biomacromolecules. 2007;8(2):617–623. gene. J Neurol Neurosurg. 2010;81(12):1396–1399. [30] Huang S, Jj L, Yang S, et al. Neuronal expression of TATA [45] Sabate R, Rousseau F, Schymkowitz J, et al. What box-binding protein containing expanded polyglutamine makes a protein sequence a prion? PLoS Comput in knock-in mice reduces chaperone protein response by Biol. 2015;11(1):e1004013. impairing the function of nuclear factor-Y transcription [46] Chan HM, La Thangue NB. p300/CBP proteins: hATs factor. Brain. 2011;134(7):1943–1958. for transcriptional bridges and scaffolds. J Cell Sci. [31] Purrello M, Di Pietro C, Mirabile E, et al. Physical 2001;114(13):2363–2373. mapping at 6q27 of the locus for the TATA [47] Mantamadiotis T, Lemberger T, Bleckmann SC, et al. box-binding protein, the DNA-binding subunit of Disruption of CREB function in brain leads to TFIID and a component of SL1 and TFIIIB, strongly neurodegeneration. Nat Genet. 2002;31(1):47–54. suggests that it is single copy in the human genome. [48] Giordano A, Avantaggiati ML. p300 and CBP: partners Genomics. 1994;22(1):94–100. for life and death. J Cell Physiol. 1999;181(2):218–230. [32] Lescure A, Lutz Y, Eberhard D, et al. The N-terminal [49] Heery DM, Kalkhoven E, Hoare S, et al. A signature domain of the human TATA-binding protein plays a role motif in transcriptional co-activators mediates bind­ in transcription from TATA-containing RNA polymer­ ing to nuclear receptors. Nature. 1997;387 ase II and III promoters. EMBO J. 1994;13(5):1166–1175. (6634):733–736. PRION 153 [50] Plevin MJ, Mills MM, Ikura M. The LxxLL motif: [68] Lee Y, Samaco RC, Gatchel JR, et al. miR-19, miR-101 a multifunctional binding sequence in transcriptional and miR-130 co-regulate ATXN1 levels to potentially regulation. Trends Biochem Sci. 2005;30(2):66–69. modulate SCA1 pathogenesis. Nat Neurosci. 2008;11 [51] Sheppard HM, Harries JC, Hussain S, et al. Analysis of (10):1137–1139. the steroid receptor coactivator 1 (SRC1)-CREB bind­ [69] Petrakis S, Raskó T, Russ J, et al. Identification of ing protein interaction interface and its importance for human proteins that modify misfolding and proteo­ the function of SRC1. Mol Cell Biol. 2001;21(1):39–50. toxicity of pathogenic ataxin-1. PLoS Genet. 2012;8 [52] Nucifora FC, Sasaki M, Peters MF, et al. Interference (8):e1002897. by huntingtin and atrophin-1 with cbp-mediated tran­ [70] Cuvertino S, Hartill V, Colyer A, et al. A restricted scription leading to cellular toxicity. Science. 2001;291 spectrum of missense KMT2D variants cause (5512):2423–2428. a multiple malformations disorder distinct from [53] McCampbell A, Taylor JP, Taye AA, et al. CREB- Kabuki syndrome. Genet Med. 2020;22(5):867–877. binding protein sequestration by expanded [71] Prives C, Lowe SW. Mutant p53 and chromatin polyglutamine. Hum Mol Genet. 2000;9(14):2197–2202. regulation. Nature. 2015;525(7568):199–200. [54] Steffan JS, Bodai L, Pallos J, et al. Histone deacetylase [72] Zhang J, Dominguez-Sola D, Hussein S, et al. inhibitors arrest polyglutamine-dependent neurode­ Disruption of KMT2D perturbs germinal center B cell generation in Drosophila. Nature. 2001;413 development and promotes lymphomagenesis. Nat (6857):739–743. Med. 2015;21(10):1190. [55] Rouaux C, J-p L, Boutillier A-L. Targeting [73] Lee J-E, Wang C, Xu S, et al. H3K4 mono-and CREB-binding protein (CBP) loss of function as di-methyltransferase MLL4 is required for enhancer acti­ a therapeutic strategy in neurological disorders. vation during cell differentiation. eLife. 2013;2:e01503. Biochem Pharmacol. 2004;68(6):1157–1164. [74] Boniel S, Szymańska K, Śmigiel R, et al. Kabuki syn­ [56] McCampbell A, Taye AA, Whitty L, et al. Histone drome—clinical review with molecular aspects. Genes deacetylase inhibitors reduce polyglutamine toxicity. (Basel). 2021;12(4):468. Proc Natl Acad Sci USA2001;98(26):15179–15184. [75] Ortega-Molina A, Boss IW, Canela A, et al. The histone [57] Klockgether T, Mariotti C, Paulson HL. Spinocerebellar lysine methyltransferase KMT2D sustains a gene ataxia. Nat Rev Dis Primers. 2019;5(1):1–21. expression program that represses B cell lymphoma [58] Tsai -C-C, Kao H-Y, Mitzutani A, et al. Ataxin 1, development. Nat Med. 2015;21(10):1199. a SCA1 neurodegenerative disorder protein, is func­ [76] Cheon CK, Sohn YB, Ko JM, et al. Identification of tionally linked to the silencing mediator of retinoid and KMT2D and KDM6A mutations by exome sequencing thyroid hormone receptors. Proc Natl Acad Sci USA . in Korean patients with Kabuki syndrome. J Hum 2004;101(12):4047–4052. Genet. 2014;59(6):321–325. [59] Orr HT, Zoghbi HY. Trinucleotide repeat disorders. [77] Ng SB, Bigham AW, Buckingham KJ, et al. Exome Annu Rev Neurosci. 2007;30(1):575–621. sequencing identifies MLL2 mutations as a cause of [60] Rocha S, Vieira J, Vázquez N, et al. ATXN1 N-terminal Kabuki syndrome. Nat Genet. 2010;42(9):790–793. region explains the binding differences of wild-type and [78] Zaidi S, Choi M, Wakimoto H, et al. De novo muta­ expanded forms. BMC Med Genomics. 2019;12(1):1–14. tions in histone-modifying genes in congenital heart [61] Irwin S, Vandelft M, Pinchev D, et al. RNA association disease. Nature. 2013;498(7453):220–223. and nucleocytoplasmic shuttling by ataxin-1. J Cell Sci. [79] Hannibal MC, Buckingham KJ, Ng SB, et al. Spectrum of 2005;118(1):233–242. MLL2 (ALR) mutations in 110 cases of Kabuki syndrome. [62] Gennarino VA, Singh RK, White JJ, et al. Pumilio1 Am J Med Genet Part A. 2011;155(7):1511–1516. haploinsufficiency leads to SCA1-like neurodegenera­ [80] Fisher SE. A molecular genetic perspective on speech and tion by increasing wild-type Ataxin1 levels. Cell. language. In G. Hickok & S. Small (Eds.), Neurobiology 2015;160(6):1087–1098. of language: Elsevier. Amsterdam. 2016. p. 13–24. [63] Nitschke L, Tewari A, Coffin SL, et al. miR760 regu­ [81] Marcus GF, Fisher SE. FOXP2 in focus: what can genes lates ATXN1 levels via interaction with its 5′ untrans­ tell us about speech and language? Trends Cogn Sci. lated region. Genes Dev. 2020;34(17–18):1147–1160. 2003;7(6):257–262. [64] Park J, Al-Ramahi I, Tan Q, et al. RAS–MAPK–MSK1 [82] Fisher SE, Scharff C. FOXP2 as a molecular window into pathway modulates ataxin 1 protein levels and toxicity speech and language. Trends Genet. 2009;25(4):166–177. in SCA1. Nature. 2013;498(7454):325–331. [83] Tsui D, Vessey JP, Tomita H, et al. FoxP2 regulates [65] Friedrich J, Kordasiewicz HB, O’Callaghan B, et al. neurogenesis during embryonic cortical development. Antisense oligonucleotide–mediated ataxin-1 reduction J Neurosci. 2013;33(1):244–258. prolongs survival in SCA1 mice and reveals disease- [84] Chiu YC, Li MY, Liu YH, et al. Foxp2 regulates associated transcriptome profiles. JCI Insight. 2018;3(21):21. neuronal differentiation and neuronal subtype [66] Keiser MS, Boudreau RL, Davidson BL. Broad thera­ specification. Dev Neurobiol. 2014;74(7):723–738. peutic benefit after RNAi expression vector delivery to [85] Clovis YM, Enard W, Marinaro F, et al. Convergent repres­ deep cerebellar nuclei: implications for spinocerebellar sion of Foxp2 3′UTR by miR-9 and miR-132 in embryonic ataxia type 1 therapy. Mol Ther. 2014;22(3):588–595. mouse neocortex: implications for radial migration of neu­ [67] Suh J, Romano DM, Nitschke L, et al. Loss of ataxin-1 rons. Development. 2012;139(18):3332–3342. potentiates Alzheimer’s pathogenesis by elevating cerebral [86] Mukamel Z, Konopka G, Wexler E, et al. Regulation of BACE1 transcription. Cell. 2019;178(5):1159–1175. e17. MET by FOXP2, genes implicated in higher cognitive 154 M. BEHBAHANIPOUR ET AL. dysfunction and autism risk. J Neurosci. 2011;31 is modulated by heterotypic and homotypic protein (32):11437–11442. interactions. Mol Cell Biol. 2004;24(2):809–822. [87] Lai CS, Fisher SE, Hurst JA, et al. A forkhead-domain [91] Häußermann K, Young G, Kukura P, et al. Dissecting gene is mutated in a severe speech and language FOXP2 oligomerization and DNA binding. Angew disorder. Nature. 2001;413(6855):519–523. Chem. 2019;131(23):7744–7749. [88] Lieberman P. FOXP2 and human cognition. Cell. [92] Stroud JC, Wu Y, Bates DL, et al. Structure of the 2009;137(5):800–802. forkhead domain of FOXP2 bound to DNA. [89] Herrero MJ, Gitton Y. The untold stories of the Structure. 2006;14(1):159–166. speech gene, the FOXP2 cancer gene. Genes [93] Mizutani A, Matsuzaki A, Momoi MY, et al. Cancer. 2018;9(1–2):11. Intracellular distribution of a speech/language disorder [90] Li S, Weidenfeld J, Morrisey EE. Transcriptional associated FOXP2 mutant. Biochem Biophys Res and DNA binding activity of the Foxp1/2/4 family Commun. 2007;353(4):869–874.