Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jun 1.
Published in final edited form as: Structure. 2015 Nov 19;23(12):2267–2279. doi: 10.1016/j.str.2015.10.010

Pre-Anchoring of Pin1 to Unphosphorylated c-Myc in a Fuzzy Complex Regulates c-Myc Activity

Sara Helander 1,5,7, Meri Montecchio 1,5, Robert Pilstål 2,5, Yulong Su 3,5, Jacob Kuruvilla 4, Malin Elvén 1,6, Javed ME Ziauddin 1, Madhanagopal Anandapadamanaban 1, Susana Cristobal 4, Patrik Lundström 1, Rosalie C Sears 3, Björn Wallner 2, Maria Sunnerhagen 1,*
PMCID: PMC4807222  NIHMSID: NIHMS768189  PMID: 26655473

SUMMARY

Hierarchic phosphorylation and concomitant Pin1-mediated proline isomerization of the oncoprotein c-Myc controls its cellular stability and activity. However, the molecular basis for Pin1 recognition and catalysis of c-Myc and other multisite, disordered substrates in cell regulation and disease is unclear. By nuclear magnetic resonance, surface plasmon resonance, and molecular modeling, we show that Pin1 subdomains jointly pre-anchor unphosphorylated c-Myc1–88 in the Pin1 interdomain cleft in a disordered, or “fuzzy”, complex at the herein named Myc Box 0 (MB0) conserved region N-terminal to the highly conserved Myc Box I (MBI). Ser62 phosphorylation in MBI intensifies previously transient MBI-Pin1 interactions in c-Myc1–88 binding, and increasingly engages Pin1PPIase and its catalytic region with maintained MB0 interactions. In cellular assays, MB0 mutated c-Myc shows decreased Pin1 interaction, increased protein half-life, but lowered rates of Myc-driven transcription and cell proliferation. We propose that dynamic Pin1 recognition of MB0 contributes to the regulation of c-Myc activity in cells.

Graphical Abstract

graphic file with name nihms768189u1.jpg

INTRODUCTION

Among the most fundamental processes in cell biology is the regulation of the cell cycle and, thereby, cell growth. A key player in this process is the peptidyl-prolyl isomerase Pin1 that has been shown to function as a molecular timer by acting as a switch of various cell-signaling processes (Liou et al., 2011; Lu et al., 2007). One of the targets of Pin1 is the proto-oncogenic transcription factor c-Myc, which is a universal regulator of cell growth, apoptosis, and proliferation in both normal and tumor cells (Lin et al., 2012; Meyer and Penn, 2008; Nie et al., 2012). While the pro-proliferative properties of c-Myc are activated by phosphorylation of Ser62, its cellular stability and degradation are controlled by subsequent phosphorylation at Thr58, both sites being located in the conserved Myc Box I (MBI) region of its N-terminal transcriptional activation domain (TAD) (Hann, 2006). The time window of phosphorylation and dephosphorylation at these two sites is affected by Pin1-mediated cis-trans isomerization of the Ser62-Pro motif, thereby regulating specific interactions of c-Myc with modifying kinases, predominantly acting on trans substrates (Brown et al., 1999; Lu et al., 2002a) and phosphatases, which control the pathway of c-Myc activation and ubiquitin-mediated degradation (Figure 1A) (Farrell and Sears, 2014; Sears, 2004), and which have been shown to act on cis substrates (Werner-Allen et al., 2011). Pin1 also facilitates the dynamic binding of c-Myc to target gene promoters, enhancing association with transcriptional co-activators and transcriptional activation of target genes, thus potentiating c-Myc’s oncogenic activity (Farrell et al., 2013). The importance of regulatory coupling c-Myc activation with its subsequent degradation in maintaining normal cell growth is evidenced by cancer-associated mutations at or near Thr58 and Ser62 that result in maintaining c-Myc in its activated, Ser62-phosphory-lated state (Bahram et al., 2000; Wang et al., 2011), and many regulators of c-Myc that bind to this region are themselves either oncogenes or tumor suppressors (Tu et al., 2015). Although Pin1 promotes c-Myc degradation in normal cells, this activity is uncoupled in cancer cells where both proteins have oncogenic activities and are overexpressed, and Pin1 functions only as a c-Myc co-activator (Farrell et al., 2013). Potential uncoupling mechanisms involve reduced Axin1 function, which scaffolds c-Myc degradation (Arnold et al., 2009; Zhang et al., 2012b) and Pin1’s downregulation of Fbw7, an E3 ligase controlling degradation of a number of oncoproteins including c-Myc (Min et al., 2012).

Figure 1. Phosphorylation and Conserved Patterns in c-Myc.

Figure 1

Open in a new tab

(A) A schematic of the biological context of Myc and Pin1 interactions as described in the Introduction. Myc phosphorylations at T58 and S62 annotated with encircled P.

(B) Sequence outline, indicating the here-proposed MB0 (c-Myc15–33) together with previously identified Myc boxes (MBI, MBII, MBIIIa, MBIIIb, and MBIV), the basic region (BR), and the helix-loop-helix leucine zipper (HLH-LZ).

(C) A sequence logo representing c-Myc residues 1–74 based on a Hidden Markov Model of Myc and generated as described in Experimental Procedures. High stacks of few, dominant amino acid letters signify conservation; MB0 and MBI are highlighted. Graphical representations (curl and arrow) respectively indicate the location of previously identified transient secondary structure (Andresen et al., 2012).

(D) Overlaid HSQC NMR spectra of unphosphorylated c-Myc1–88 (black) and pSer62-c-Myc1–88 (red). Peaks that display CSPs >0.05 ppm on Ser62 phosphorylation are labeled.

(E) CSPs on phosphorylation for all assigned c-Myc1–88 amide resonances.

See also Figure S1.

Pin1 comprises two independently folded subdomains: Pin1WW (residues 6–39) and Pin1PPIase (residues 50–163), connected by a flexible linker (Bayer et al., 2003; Jacobs et al., 2003; Ranganathan et al., 1997). The catalytic proline isomerization activity toward pSer/pThr-Pro motifs entirely resides in Pin1PPIase (Lu et al., 1999; Ranganathan et al., 1997). The Pin1PPIase alone binds weakly to native, phosphorylated peptide targets (KD > 390 μM) (De et al., 2012; Verdecia et al., 2000), with significant affinities (Kd range 0.5–90 μM) observed only for peptides selected from library screens (Duncan et al., 2011; Namanja et al., 2011; Verdecia et al., 2000) or designed inhibitors (Namanja et al., 2011; Zhang et al., 2012a). Pin1WW binds pSer/ pThr-Pro-containing peptides with variable affinities (Kd > 7 μM) (Lu et al., 1999; Verdecia et al., 2000), preferably in the extended, trans conformation (De et al., 2012; Namanja et al., 2011; Verdecia et al., 2000; Wintjens et al., 2001). In intact Pin1, the two domains create an interdomain cleft, distant from the Pin1PPIase active site, where phosphorylated target peptides (Kd 5–80 μM) (Verdecia et al., 2000) as well as buffer components such as polyethylene glycol (PEG) bind in well-defined conformations (reviewed in Matena et al., 2013). Pin1WW binding at phosphorylated pSer/pThr-Pro sites has been proposed to increase the local effective concentration of substrate and activity on neighboring phosphorylated sites (Jacobs et al., 2003; Lu et al., 1999), to facilitate substrate transfer (De et al., 2012; Lu et al., 1999; Wintjens et al., 2001), and/or to sequester trans conformations away from the active site, thereby providing directionality to the cis-trans conversion (De et al., 2012; Namanja et al., 2011; Wintjens et al., 2001). Still, the functional reason for binding of targeted peptides to an interdomain cleft distant from the Pin1 active site remains unresolved.

Recent studies suggest that collaborative and possibly allosteric mechanisms jointly involve both Pin1 domains in target binding (reviewed in Peng, 2015). Binding of shorter peptide substrates and small molecules to Pin1 has been shown to affect interdomain mobility and linker dynamics (Jacobs et al., 2003), and increased affinity and isomerization of phosphorylated peptides binding to Pin1PPIase has been shown in the presence of PEG-induced transient domain interactions (Matena et al., 2013). Interactions between the two domains have been shown to allosterically affect the isomerization activity by an internal dynamic circuit through the Pin1PPIase interior (Namanja et al., 2011), as well as through residues in the domain interface (Wilson et al., 2013), both recently supported by molecular simulations (Guo et al., 2015). However, to understand how the dual-domain protein Pin1 acts on its longer, multiply phosphorylated, and often intrinsically disordered substrates (Lu and Zhou, 2007), the interaction with such substrates needs to be studied in structural and dynamic detail, but as yet such studies have not been achieved.

We recently characterized the structural and dynamic properties of the most N-terminal part of the c-Myc TAD domain (c-Myc1–88), by nuclear magnetic resonance (NMR) and surface plasmon resonance (SPR), detailing also its dynamic and multivalent interactions with the tumor suppressor Bin1 (Andresen et al., 2012). We found that c-Myc1–88 contains two transiently ordered regions: the well-characterized MBI region and a less studied, more N-terminal region conserved in c-, N-, and L-Myc (Cowling and Cole, 2006; Legouy et al., 1987) (Figures 1B and 1C). While MBI is a well-known c-Myc interaction site, and comprises the phosphorylation sites directing c-Myc stability (Meyer and Penn, 2008), so far no clear functional role has been attributed to the transiently ordered region N-terminal to MBI.

Here, we show for the first time how a longer Pin1 substrate, c-Myc1–88, interacts with both domains of Pin1, and how this interaction is affected by phosphorylation. By SPR, NMR, and cellular assays, we show that the transiently ordered, unphosphorylated c-Myc region comprising c-Myc residues 13–32, which we henceforth refer to as “Myc Box 0” (MB0; Figures 1B and 1C), serves as a dynamic anchoring site for Pin1 on c-Myc1–88, both in the absence and presence of Ser62 phosphorylation. Molecular simulations, restricted by experimental data, show how Pin1 WW and PPIase domains jointly bind the MB0 region in a dynamic complex, thus facilitating MBI interaction with the Pin1 active site. Our results suggest that Pin1 binding to MB0 affects the cellular activity window of the c-Myc oncoprotein.

RESULTS

Intrinsic Disorder Is Retained in pSer62-c-Myc1–88

To study the Ser62 phosphorylated state of c-Myc, we performed phosphorylation of c-Myc1–88 in vitro with active CDK2 kinase in complex with CyclinA2, which is known to phosphorylate Ser62-Pro63 in vivo (Hydbring et al., 2010) and in trans (Brown et al., 1999). Specific and near-complete pSer62-c-Myc1–88 phosphorylation was confirmed by mass spectrometry and NMR, in agreement with the extended properties of the MBI region (Andresen et al., 2012) (Figures S1A–S1D). Chemical-shift perturbations (CSPs) were confined to the well-conserved MBI region (Figures 1D and 1E). The transverse relaxation rates for c-Myc1–88 are concentration dependent (Figures S1H and S1I), which agrees with monomer exchange with a small amount of higher molecular weight oligomer not observed by sample inspection or in gel-filtration experiments (Andresen et al., 2012). However, the overall decreased R2 relaxation rates for pSer62-c-Myc1–88 compared with c-Myc1–88 at the same concentration suggest that the additional negative charge introduced by phosphorylation reduces internal interactions (Figure S1F). Both 15N-R1 and {1H}-15N-nuclear Overhauser effect (NOE) relaxation of pSer62-c-Myc1–88 correspond well to results obtained for c-Myc1–88 at similar concentrations (Figures S1E–S1G), suggesting that the intrinsic disorder with transiently structured regions identified for c-Myc1–88 (Andresen et al., 2012) is maintained upon Ser62 phosphorylation.

Intact Pin1 Binds Unphosphorylated c-Myc1–88

We used SPR to investigate the binding of Pin1 and its subdomains to c-Myc1–88 and pSer62-c-Myc1–88 (Figures 2, S2A, and S2B). Since Pin1 is believed to primarily bind at phosphorylated protein sites (Liou et al., 2011; Lu and Zhou, 2007), we were surprised to find that intact Pin1 binds c-Myc1–88 with an apparent KD of 4 μM (Figures 2A and S2B). This Pin1 binding affinity is of similar or higher affinity as previously observed Pin1 binding to phosphorylated peptides, but several orders of magnitude higher than for unphosphorylated Ser/Thr-Pro motifs (Verdecia et al., 2000). Lack of significant Pin1 binding to c-Myc46–69 (Figure S2A) indicated that the primary binding site for Pin1 to c-Myc1–88 is located distant from the Ser62-Pro motif. Both Pin1WW and Pin1PPIase subdomains were consistently found to bind c-Myc1–88 with more than 10-fold weaker affinities compared with intact Pin1 (Figures 2B, 2C, and S2B), suggesting both Pin1 subdomains jointly contribute to the higher c-Myc1–88 affinity of the intact protein. SPR measurements of Pin1 affinities to pSer62-c-Myc1–88 showed difficulties in reaching saturation and anomalous binding effects at higher concentrations, which limited the accuracy (Figures 2D–2F). Notably, however, the phosphorylated substrate may bind the active Pin1 enzyme in several modes with different affinities, due to cis-trans isomerization of pSer62-Pro (De et al., 2012; Jacobs et al., 2003; Namanja et al., 2011). In agreement with this, isothermal titration calorimetry measurements indicate multiple binding events for intact Pin1 binding to pSer62-cMyc1–88, with Kds ranging from 10 to 100 μM (Figures S2C–S2E), i.e. in the same range as measured for other phosphorylated peptides (Verdecia et al., 2000). The higher pSer62-c-Myc1–88 apparent affinities for intact Pin1 compared with its subdomains alone indicate joint subdomain binding contributions also for phosphorylated c-Myc1–88. Finally, while full kinetic analysis was not feasible due to the rapid on- and off-rates in binding, we consistently observed elevated off-rates by visual inspection for Pin1PPIase binding to pSer62-c-Myc1–88 compared with unphosphorylated c-Myc1–88 (Figures 2C and 2F), in agreement with increased isomerization turnover of phosphorylated substrates.

Figure 2. Evaluation of c-Myc1–88-Pin1 Affinities by Surface Plasmon Resonance.

Figure 2

Open in a new tab

Sensorgrams and 1:1 Langmuir fits from steady-state evaluation show binding of Pin1 (red), Pin1WW (green), and Pin1PPIase (blue) to both unphosphorylated (A–C) and phosphorylated c-Myc1–88 (D–F). For all measurements c-Myc1–88 was immobilized and Pin1, Pin1WW, or Pin1PPIase were injected over the surface. SPR experiments are further detailed in Figure S2.

Pin1 Binds c-Myc1–88 in a Dynamic Complex Anchored at the MB0 Region

By NMR, we further investigated c-Myc-Pin1 interactions on a per-residue level by titrating 15N-labeled c-Myc1–88 and pSer62-c-Myc1–88 with unlabeled Pin1. We found that while for specific c-Myc residues resonance intensities were significantly reduced and not recovered on saturating the binding equilibrium (Figure 3), CSPs were very small or nonexistent (Figure S3). While peak intensities in general are lowered upon interaction due to slower molecular tumbling in the complex state, further line broadening leading to reduced intensities can occur as a result of chemical exchange between bound and free states, and/or due to chemical exchange between multiply bound states (Bozoky et al., 2013a, 2013b; Lukhele et al., 2013; Mittag et al., 2008). In the latter case, the signal is not recovered in the bound state due to continued chemical exchange (Bozoky et al., 2013b), which is in agreement with our observations for Pin1/ c-Myc1–88 complexes. Furthermore, we detected no 15N Carr-Purcell-Meiboom-Gill dispersions, showing that line broadening due to chemical exchange occurs on a faster timescale than can be probed by these experiments. These observations are in agreement with the formation of dynamically disordered (Forman-Kay and Mittag, 2013), or “fuzzy” (Fuxreiter and Tompa, 2012), complexes.

Figure 3. NMR Analysis of c-Myc1–88 Per-Residue Interactions with Pin1 and Its Subdomains.

Figure 3

Open in a new tab

HNCO peak intensity ratios between Pin1-bound and free states were derived for c-Myc1–88 when unphosphorylated (A–C) or Ser62-phosphorylated (D–F) at a c-Myc1–88/Pin1, c-Myc1–88/Pin1WW, or c-Myc1–88/Pin1PPIase ratio of 1:2, with estimated saturation levels ranging from 58% to 96% (see Experimental Procedures). Prolines (dark gray; give no signal in HNCO) and residues lacking data due to missing assignment or overlap (light gray) are represented by solid histogram bars. Binding to Pin1 (A and D), Pin1WW (B and E), and Pin1PPIase (C and F) is color-coded as in Figure 2. Gain/loss of interactions in the bound state leads to decreased/ increased peak intensity ratios, respectively. Graphical representations (curl and arrow, respectively) indicate the location of previously identified transient secondary structure (Andresen et al., 2012). The phosphorylation site at S62 is indicated with encircled P. See also Figure S3.

Since c-Myc1–88 exhibits low chemical-shift dispersion (Andresen et al., 2012) and very minor CSPs in the bound state (Figure S3), HNCO intensity ratios (Mittag et al., 2008) were primarily used to characterize the interaction of c-Myc1–88 to intact Pin1 and its subdomains (Figure 3). Binding of intact Pin1 to non-phosphorylated c-Myc1–88 results in major loss of signal intensity primarily in the conserved, transiently ordered MB0 region, but also affects the MBI region (Figures 3A and 3D). Intact Pin1 binding to pSer62-c-Myc1–88 more clearly affects the phosphorylation site and flanking residues, and perturbations extend into the transiently helical and conserved MBI region around c-Myc1–88-Trp50, while the C-terminal part remains comparatively unperturbed (Figure 3D). Binding of either Pin1WW or Pin1PPIase resulted in drastically lowered HNCO intensity ratios in the MB0 region of c-Myc1–88, suggesting that both Pin1 subdomains target this region (Figures 3B and 3C). Pin1WW primarily affects c-Myc residues 16-YDSVQPYFY-23 (Figure 3B), while Pin1PPIase affects the aromatic residues in the 20-PYFY-23 motif as well as conserved residues 29-ENFY-32 slightly C-terminal to the Pin1WW binding motif (Figure 3C). The proline-aromatic 20-PYFY-23 pattern affected by both Pin1 subdomains is also found in Pin1 binding peptides selected from library screens, both in forward (CTGIPWLYC; Duncan et al., 2011) and reverse sequence orientation (Pintide: WFYpSPFLE; Lu et al., 1999; Verdecia et al., 2000).

On phosphorylation, the binding pattern of Pin1 subdomains to pSer62-c-Myc1–88 alters significantly (Figures 3E and 3F). While interactions to MB0 are maintained, both Pin1WW and Pin1PPIase binding now also leads to severely reduced HNCO peak intensity ratios at c-Myc1–88-pSer62 and adjacent residues (Figures 3E and 3F). Furthermore, binding of both c-Myc1–88 and pSer62-c-Myc1–88 to Pin1 results in more extensive effects than would be predicted from Pin1 subdomain binding alone (Figures 3A and 3D). This suggests that both Pin1 domains jointly bind c-Myc, which agrees with the increased affinity for intact Pin1 compared with its subdomains as observed by SPR (Figure 2). Furthermore, the NMR results suggest that the conserved, transiently structured MB0 region in c-Myc1–88 is central for Pin1 interactions with c-Myc1–88 irrespective of the phosphorylation state of Ser62.

To further investigate c-Myc1–88 dynamics upon Pin1 binding, we evaluated 15N-R1, 15N-R1ρ, and {1H}-15N-NOE relaxation experiments for the saturable unphosphorylated c-Myc1–88-Pin1 complex. Although excessive line broadening in the MB0 region limited the evaluation, the residues that showed sufficient signal intensity for analysis had 15N-R1 relaxation rates similar to those of free c-Myc1–88, indicating retained mobility of flanking regions (Figure 4A). With retained R1 rates, the c-Myc1-88:Pin1 complex displays higher 15N-R2 relaxation rates compared with free c-Myc1–88, (Figure 4B), suggesting transient interactions between Pin1 and regions flanking the MB0 anchor site in agreement with decreased HNCO intensity ratios in the same regions (Figure 3). NOE relaxation rates in the c-Myc1–88-Pin1 complex are similar to those in free c-Myc1–88 and correspond well to earlier characterized transient secondary elements (Andresen et al., 2012), suggesting that c-Myc1–88 transient structure is retained upon Pin1 binding (Figure 4C). Taken together, NMR relaxation measurements show retained intrinsic disorder in c-Myc1–88 when binding Pin1, signifying a fuzzy complex (Fuxreiter and Tompa, 2012).

Figure 4. NMR Relaxation Analysis of c-Myc1–88 Binding to Pin1.

Figure 4

Open in a new tab

Relaxation parameters of apo c-Myc1–88 (open circles) and c-Myc1–88 (filled circles) in the presence of 2 mol equivalents of Pin1 (98% saturation). Uncertainties in R1 and R2 were estimated using the jackknife method, and in the heteronuclear NOE as the standard deviation of all permutations of peak ratios in duplicate experiments (Ahlner et al., 2013).

(A) R1 relaxation rates.

(B) R2 relaxation rates.

(C) {1H}-15N-NOE.

Pin1 Recognition Pattern Is Altered by Myc1–88 Phosphorylation

To obtain more detailed molecular insight on the extent of Pin1 that interacts with c-Myc, we mapped spectral changes in 15N, 13C-labeled Pin1 on addition of unlabeled c-Myc1–88 and pSer62-c-Myc1–88 by NMR. For both unphosphorylated and phosphorylated c-Myc, small but significant CSPs were observed (Figures 5, S4A, and S4B). While CSPs only suggest that the chemical environment around the concerned backbone NH groups has changed, the most likely reasons are a direct binding with an interaction partner or an indirect conformational change due to this interaction.

Figure 5. NMR Mapping of Pin1 Interactions with c-Myc1–88 and pSer62-c-Myc1–88.

Figure 5

Open in a new tab

Surface, secondary structure representations, and charts of the CSPs of Pin1, interacting with c-Myc1–88 (A) and pSer62-c-Myc1–88 (B), respectively. Pin1 residues affected by Myc binding are color coded through (A) and (B) as follows: residues in Pin1WW (blue), subdomain interface (orange), Pin1PPIase catalytic loop and active site (red), Pin1PPIase substrate recognition and binding (green). The sulfate group from the crystal structure 1PIN, represented by yellow spheres, indicates the proposed Pin1 active site. See also Figure S4.

In the unphosphorylated c-Myc1–88:Pin1 interaction, a near-continuous surface of CSP-displaying residues is formed by Pin1WW residues centered around Trp34 (Figure 5A, blue), and, extending toward the Pin1PPIase active site, by residues facing the cleft between the two domains (Figure 5A, orange). This suggests a Myc binding surface involving the interdomain cleft, which agrees with Pin1WW and Pin1PPIase jointly binding the c-Myc MB0 region (Figure 3). In the WW domain, affected residues correspond well to those observed in crystal structures to be involved in PEG and phosphopeptide binding (Matena et al., 2013; Namanja et al., 2011; Ranganathan et al., 1997; Verdecia et al., 2000). In addition, a clear pattern of CSPs is observed for Pin1PPIase residues involved in substrate recognition and proline ring placement (Figure 5A; green) while weak CSPs are observed for residues in the active-site loop (Figure 5A; red). Interestingly, the Pin1 interdomain interface and PPIase core region affected by c-Myc1–88 includes many residues in the “Path 1” proposed to mediate allosteric interdomain regulation (Guo et al., 2015).

The pSer62-c-Myc1–88:Pin1 interaction results in a more extensive CSP pattern, which increasingly affects Pin1 residues involved in domain interactions and catalysis (Figure 5B). The effects of Pin1 interaction with pSer62-c-Myc1–88 now extends from the Trp34 region to involve WW residues His27 and Ile28 lining the interdomain cleft, as well as Ala140 facing these residues from the PPIase domain. Notably, Pin1-Ile28 was recently shown to be a key residue in regulating substrate binding affinity and isomerase activity by means of its interdomain contact position (Wilson et al., 2013). Furthermore, facing the Pin1WW domain, the Pin1PPIase helix 1 (residues 82–98), which was not affected by unphosphorylated c-Myc1-88, now shows significant CSPs (Figure 5, orange). Helix1 connects to the active-site loop (Figure 5, red) via highly affected Pin1-His64 (Figure 5), which was previously found to be highly perturbed in the binding of both cis- and trans-restricted inhibitors (Namanja et al., 2011). Notably, Pin1 helix1, the interdomain interface, and the WW pocket are also fundamental parts of the second allosteric pathway (“Path 2”) identified by Guo et al. (2015). Pin1 residues proposed to be involved in substrate recognition and proline ring placement (Namanja et al., 2011) are differently affected: residues 115–117 and 130–131 at the edge of the Pin1PPIase phosphate binding groove are involved in recognition of phosphorylated c-Myc while unphosphorylated c-Myc affects the surface-exposed β strand and loop structure including residues 120–130 (Figure 5; green). Specifically, Pin1 residues I78 and A116, which show enhanced flexibility on ligand binding in previous studies as a possible reflection of substrate recognition (Namanja et al., 2011), are both highly affected by pSer62-c-Myc1–88 binding but not by binding to unphosphorylated c-Myc1–88. Finally, residues 102–105, close to the hydrophobic patch where a second PEG molecule was observed in the Pin1 crystal structure deposited as 1PIN (Ranganathan et al., 1997), now show CSPs, suggesting altered interactions at this site (Figure 5, gray). The interaction of Pin1 with c-Myc could be further modulated by phosphorylation, since two of three important phosphorylation sites in Pin1 (Ser16 and Ser71; Hariharan and Sussman, 2014; Lee et al., 2011; Lu et al., 2002b) are involved in c-Myc interactions with both phosphorylated and unphosphorylated states (Figure 5).

A comparison between HNCO peak intensity ratios for Pin1, in the presence of either c-Myc1–88 or pSer62-c-Myc1–88, reveals a general and uniform reduction of peak intensity ratios for both folded domains of Pin1, in contrast to the residue-specific effects observed for c-Myc1–88 (Figures S4C and S4D). This suggests that the two Pin1 subdomains, which in the free state move independently of each other (Bayer et al., 2003; Jacobs et al., 2003), experience an increased τc as an effect of joint tumbling of the two domains on binding either phosphorylated or unphosphorylated c-Myc1–88, which again supports joint c-Myc binding by Pin1WW and Pin1PPIase subdomains.

Ensemble Models Describe the Fuzzy Myc-Pin1 Complex

To outline the possibilities for deriving a structural model for the c-Myc-Pin1 interaction, we jointly assessed all our experimental data (Figures 2, 3, 4, and 5). SPR measurements showed significant binding of Pin1 to unphosphorylated c-Myc outside of the Pin1-targeted MBI (Figures 2, S2A, and S2B), and, in agreement NMR mapping of HNCO intensity ratios, suggests that Pin1 primarily binds to the MB0 region (c-Myc10–35) (Figure 3). We focused on investigating how MB0 may interact with Pin1, and how this might affect binding of the unphosphorylated MBI region, since binding to pSer62-c- Myc1-88 seems structurally less resolvable due to the presence of multiple interconverting bound states (Figures 2 and S2C–S2E).

To visualize the structural space accessible to a c-Myc1–88- Pin1 dynamic complex that satisfies our experimental data, we first modeled c-Myc10–35 in complex with Pin1 and then extended this ensemble to include the entire c-Myc10–70 region, using Rosetta docking and loop modeling protocols (see Experimental Procedures). The docking was guided by experimental constraints derived from HNCO intensity data from c-Myc1–88 binding to Pin1, Pin1WW and Pin1PPIase, as well as from Pin1 CSPs on c-Myc1–88 binding (see Experimental Procedures). The resulting conformations were clustered using a 3-Åroot-mean-square deviation radius cutoff, and the largest clusters, representing a majority of the docked conformations, were sorted based on combined physical and constraint energies (Figures S6). From each of these clusters, the lowest-energy representative that best satisfied the experimental constraints was included in the ensemble describing the bound complexes (Figure 6 and Movie S1).

Figure 6. Ensemble Model of the Interaction Between c-Myc and Pin1.

Figure 6

Open in a new tab

(A) Filtered and rescaled signals used to guide the Rosetta Monte Carlo simulation from HNCO ratios for c-Myc1–88 versus Pin1WW (top), c-Myc1–88 versus Pin1PPIase (middle), and c-Myc1–88 versus full Pin1 (bottom).

(B) Plot of rescaled signals from Pin1 versus c-Myc1–88 ΔCSPs.

(C) A structural representation of the MB0-anchored Myc-Pin1 fuzzy complex, as presented by the ensemble of lowest-energy c-Myc10–35-Pin1 complex conformations from the 15 top-scoring clusters, visualized as ribbon representations of the c-Myc10–35 backbone superimposed on PDB:1PIN. The cross-section area of the ribbon is scaled to the relative size of the cluster. The black arrow indicates the N- to C-terminal c-Myc10–35 binding direction as defined in Figure S5B.

(D) Bottom view of the complex ensemble with SO4 in the superimposed 1PIN structure bound to the active site, represented by yellow spheres.

(E and F) Extended ensemble model showing Myc-Pin1 fuzzy complexes anchored at MB0 with transient MBI-Pin1 interactions. The c-Myc10–70 ensemble comprises the 17 top-scoring clusters, visualized in the same manner as the c-Myc10–35 model.

(G) Relative position of the modeled Myc fragments on c-Myc1–88 with the same coloring gradient of c-Myc10–35 and c-Myc10–70 as in (C) to (F); transient secondary structure elements and the Ser62 phosphorylation site are indicated.

(H) The experimental constraint score versus the Rosetta standard score is plotted as dots for all c-Myc10–35-Pin1 models; red dot if the model belongs to a cluster and gray otherwise. The lowest-energy representative is shown for each cluster irrespective of size (black squares), with the 15 highest-scoring clusters highlighted (gray circle).

(I) Scatterplot of c-Myc10–70 clusters and -Pin1 models; same visualization as in (H) but with the 17 highest-scoring clusters highlighted.

See also Figure S5.

The resulting models show a wide ensemble of c-Myc10–35 conformations favorably interacting with both Pin1WW and Pin1PPIase by means of the MB0 region (Figures 6C, 6D, 6G, and 6H; Figure S5A; Movie S1). A preferred direction of binding of MB0 in the interdomain cleft is evident and is introduced by the experimental constraints, since without that contribution to the energy function the bias almost disappears (Figure S5B). In the full c-Myc10–70 model, the extent of the linker between MB0 and MBI generously allows for unphosphorylated MBI to interact in a multitude of ways with the Pin1PPIase active-site region (Figures 6E–6G, 6I, and S6), in full agreement with experimental data (Figures 3, 4, and 5). Notably, our NMR relaxation data for unphosphorylated c-Myc1–88 show that interactions between Pin1 and regions flanking the MB0 anchor site are dynamically transient, in contrast to the multistate, fuzzy Pin1 binding of MB0 (Figures 3 and 4). Thus, the entire ensemble of Pin1-bound c-Myc1–70 will also include Myc conformations that are anchored only at MB0, leaving the MBI region disordered and accessible to phosphatases and kinases.

Mutations in MB0 Affect Critical Myc Regulatory Functions

The biophysical and computational data suggest that the conserved MB0 region forms a primary interaction site for Pin1 to unphosphorylated c-Myc1–88. To evaluate the importance of this interaction in cells and for c-Myc’s cellular functions, we generated point mutations in the MB0 coding region, targeting c-Myc-20-PYFY-23 in human corresponding to c-Myc-20-PYFI-23 in murine, and created murine alanine c-Myc mutants 20-PAAA, 20-PAAI, and 20-AAAI. Consistent with the hypothesis that these residues are important for Pin1 to anchor on c-Myc (Figure 3), our co-immunoprecipitation (coIP) experiments reproducibly showed more than 50% reduction in Pin1’s ability to bind to MB0 mutated c-Myc compared with wild-type (WT), with AAAI the lowest affinity observed (Figure 7A). The cellular data agree well with SPR experiments performed with three corresponding human c-Myc1–88 mutants, c-Myc-20-PYFY/AAAA-23, -PYFY/PAAY, and -PYFY/AAFY. Compared with WT c-Myc1–88, the c-Myc-20-AAAA-23 mutant displays a lower affinity toward human Pin1, 15 μM, compared with 4 μM for the WT construct (Figures S7A and S7D). Measurements were also performed with the other two mutants c-Myc-20-PYFY/PAAY and c-Myc-20-PYFY/AAFY, which were also able to bind Pin1, but showed reduced affinity (Figures S7B–S7D). In addition, the mutants are phosphorylated at S62 with similar or higher levels compared with WT c-Myc, suggesting that the reduced affinity is not caused by lack of c-Myc S62 phosphorylation, and consistent with a role for Pin1 in facilitating PP2A-mediated S62 dephosphorylation (Yeh et al., 2004) (Figure 7B). We also observed an increase in protein stability in the MB0 mutants reflected by a prolonged protein half-life following inhibition of protein synthesis by cycloheximide treatment (Figure 7C) and by a higher steady-state level when transfecting equal amounts of plasmids to the WT (Figure 7D, western blot). The increased stability of the MB0 mutants is likely due to downregulation of Pin1’s facilitating c-Myc degradation caused by reduced interaction (Yeh et al., 2004).

Figure 7. Mutations in MB0 Decrease Pin1 Affinity and Regulate c-Myc activity.

Figure 7

Open in a new tab

Reduced binding affinity with Pin1, less transcriptional activity, and reduced pro-proliferation ability is observed for c-Myc-20-PYFI mutants.

(A) CoIP of cell lysates from HEK293s transfected with indicated plasmids using anti-V5 antibody for immunoprecipitation and anti-Flag antibody for detection of Co-immunoprecipitated Pin1 (bottom band). Co-immunoprecipitated Pin1 band densities were quantified after normalization to input, immunoprecipitated c-Myc level, and controls.

(B) Western analysis of lysates from HEK293 expressing indicated plasmids, V5-immunoprecipitated c-Myc were used for S62 phosphorylation analysis, S62A serves as negative control.

(C) Western analysis of lysates from HEK293 transfected with equal amounts of V5-Myc plasmids (top: WT; bottom: AAAI mutant). Four independent experiments were used to quantify half-life.

(D) 4xE-box driven luciferase signal detected from lysates of HEK293s co-transfected with indicated V5 plasmids or empty vector as control. Luciferase signals were adjusted based on expression levels shown in the immunoblots below.

(E) DNA binding affinity for WT and mutant V5-Myc shown by quantitative ChIP at indicated promoters. ChIP levels were adjusted by respective inputs and protein levels. Lysates of HEK293s transfected empty vector were used as control for normalization.

(F) Colony formation assays were performed in REF52 cells co-transfected with H-Ras and either WT or mutant c-Myc as indicated for up to 3 weeks.

Statistical significance relative to WT was calculated using a two-tailed t test: *p < 0.05; **p < 0.01; ***p < 0.001; n.s., not significant. All quantifications were based on at least three independent experiments.

See also Figure S7.

Since in addition to facilitating c-Myc Ser62 dephosphorylation and degradation, Pin1 initially stimulates c-Myc DNA binding and target gene activation (Farrell et al., 2013), we tested the effects of the c-Myc-20-PYFI mutations on c-Myc transcriptional activity. We measured the luciferase signal driven by the canonical c-Myc binding sequence, 4xE-Box-luc, following expression of c-Myc WT or MB0 mutants. Mutants with disruption of c-Myc-20-PYFI consistently showed decreased ability to drive luciferase signal compared with WT c-Myc (Figure 7D). The c-Myc-20-PYFI mutants also have significantly decreased interaction with endogenous targets genes, E2F2 and Nucleolin, measured by quantitative ChIP (Figure 7E). Interestingly, the degree to which the mutants display impaired activities coordinates well with their abilities to interact with Pin1 (Figure 7A), suggesting that Pin1 is the main regulator of the c-Myc-20-PYFI domain and its effects on c-Myc activity. To test the role of c-Myc-20-PYFI in cell proliferation, we conducted colony formation assays in REF52 cells. As shown in Figure 7F, the ability of cells to form colonies with the AAAI mutant and Ras was dramatically decreased compared with cells with WT Myc and Ras, highlighting the importance of the c-Myc-20-PYFI domain for c-Myc’s pro-proliferative function. We find that several pro-proliferative transcription factors that are Pin1 targets (Liou et al., 2011) contain a PYFY-type motif N-terminal to their pSer-Pro sites, suggesting a more general role of this motif in Pin1-regulated pro-proliferative targets (Figure S7E).

DISCUSSION

The Pin1 enzyme is well known for its capability of cis-trans isomerization of phosphorylated (p)Ser/(p)Thr-Pro target sequences, and specifically acts on intrinsically disordered substrates in cell regulation to direct their regulatory activities (Liou et al., 2011; Lu et al., 2007). However, limited light has been shed on how Pin1 interacts with longer substrates, since the present molecular knowledge predominantly has been gained from Pin1 studies of interactions with small molecules or peptides. The present work significantly advances our understanding of Pin1-substrate interactions by showing, using a wide spectrum of biophysical and cellular methods, that Pin1 recognition of its well-known target c-Myc involves pre-anchoring to an unphosphorylated conserved motif distal from the phosphorylated target site for Pin1-mediated proline isomerization, which is biologically critical and structurally significant. Specifically, we find that Pin1 binds unphosphorylated c-Myc in a Pin1 interdomain-anchored dynamic, or fuzzy, complex at a well-conserved region here designated MB0 (Figure 1), and we show that such anchoring is structurally compatible with concomitant transient binding to the Pin1PPIase active site region of the MBI (p)Ser/(p)Thr-Pro Pin1 target site(s) for cis/trans isomerization (Figure 6), in full agreement with our NMR data (Figures 3, 4, and 5). Importantly, disturbing Pin1 binding to its unphosphorylated MB0 anchor site results in impaired c-Myc transcriptional activity and reduced c-Myc function in cellular proliferation assays.

A compelling advantage of Pin1 pre-anchoring to unphosphorylated, distal site(s) would be to increase the local concentration of substrate presented to the active site and thereby further increase the catalytic efficiency of Pin1. Our study substantiates this, since we in our NMR experiments directly observe a significant recognition of the unphosphorylated MBI region by intact Pin1 on MB0 binding (Figure 3), indicating that this recognition is facilitated by an increase in local concentration due to Pin1-MB0 pre-anchoring. When c-Myc1–88 is phosphorylated at Ser62, we find that Pin1 distinctly recognizes the phosphorylation site in MBI but still affects the N-terminal, non-phosphorylated MB0 (Figure 3), suggesting a maintained role of this interaction in the phosphorylated state. Furthermore, the mere localization of MB0 binding to the interdomain cleft (Figures 5 and 6) may also mediate increased catalytic activity. Recent work has shown enhanced interdomain interactions in Pin1 by small-molecule binding at the Pin1 subdomain interface (Matena et al., 2013), and critical interdomain contact residues such as Ile28 have been shown to affect Pin1 catalytic activity and substrate binding affinity (Wilson et al., 2013). Notably, however, simultaneous tight binding of designed bivalent Pin1 binders to both WW and PPIase creates inhibition of Pin1 activity (Daum et al., 2007), suggesting that intrinsic dynamics in binding both Pin1 sites is required for native substrate activity.

Our results show that mutations in the MB0 Pin1-anchoring region dramatically decrease Ras-dependent transformation and simultaneously show prolonged c-Myc half-life. At first glance this may appear confusing, since several stabilizing c-Myc mutations have been shown to increase cell proliferation (Chang et al., 2000; Salghetti et al., 1999). However, our results are compatible with recent data demonstrating that Pin1 isomerization at pSer62-Pro63 in c-Myc promotes c-Myc DNA binding and target gene activation (Figure 1; Farell et al., 2013) even though it can also facilitate S62-dephosphorylation by protein phosphatase 2A and increase c-Myc turnover (Figure 1; Yeh et al., 2004; Arnold and Sears, 2006). Thus, a Myc mutant with deficient Pin1 interaction would indeed be more stable and Ser62 phosphorylated, but would not promote transcriptional and/or oncogenic activity. Since our biophysical results indicate that Pin1 binding to the MB0 site N-terminal to the c-Myc pSer62-Pro motif serves to allosterically activate Pin1 and increase the frequency of encounters between the MBI target sequence and the PPIase active site, it is indeed a tenable hypothesis that with disturbed MB0 pre-recognition, both the Pin1 isomerase activity and the affinity between Myc and Pin1 will be too low to be biologically efficient. This would result in a higher prevalence of Myc isomers that are less active in transcriptional activity in the Myc-accessible cellular activity window, with concomitant lowering of cell growth, as shown in our MB0 mutants.

We suggest that the biological requirement for both Pin1 subdomains to reach full Pin1 functionality is based on the biophysical requirements for joint subdomain interaction to longer substrates, which is necessary to allosterically elevate enzymatic activity and specificity to the levels required for efficient biological regulation. Recent studies show that interdomain dynamics is affected by binding of small molecules or substrate peptides/analogs to the active site and/or to the interdomain region (Bayer et al., 2003; Jacobs et al., 2003; Matena et al., 2013; Namanja et al., 2007; Wilson et al., 2013). Our data support and extend these findings by showing how Pin1 integrates recognition to both these sites in binding to a longer substrate. Importantly, we show that not only key residues, such as Pin1WW-Ile28 and its counterpart Ala140 in Pin1PPIase, but a wide range of interdomain residues in both Pin1 subdomains are differentially affected with respect to the phosphorylation state of the binding peptide (Figure 5). A recent study performed by molecular dynamics simulations proposes that the two Pin1 domains are allosterically regulated through two pathways: a first quiescent state called Path 1, and a second state (Path 2) which is sequentially activated upon substrate binding to the WW pocket (Guo et al., 2015). Interestingly, in our experimental work with a long substrate, we find that the CSPs in Pin1 created by the binding with c-Myc1–88 resembles the Path 1 pattern, while with pS62-cMyc1–88 a pattern comparable with Path 2 emerges (Figure 5). Notably, while the WW pocket is involved in binding both phosphorylated and unphosphorylated c-Myc1–88, it is only in the presence of pS62-c-Myc1–88 that the second pathway becomes active. Therefore, we propose that the presence of the phosphate in the active site represents the real trigger factor for the activation of Path 2, given joint binding of the substrate protein to the WW pocket. Thus, the allosteric communication between the two Pin1 domains is not a one- way signal initiated by Pin1WW binding, but it is a multistep process whereby Path 1 is activated by substrate binding to the WW domain and Path 2 by the phosphate binding the PPIase active site. Taken together, a comparison between results relating to Pin1 allostery in our experimental data and in the molecular dynamics simulations (Guo et al., 2015) reveals interesting similarities, but also different implications concerning the activation mechanism of allosteric pathways.

The highly dynamic association between non-phosphorylated c-Myc1–88 and Pin1 shows the characteristics of a fuzzy complex (Fuxreiter and Tompa, 2012): alternate conformations are allowed in the bound state, and large parts of the bound c-Myc1–88 peptide have dynamic properties closely resembling the unbound state. The unphosphorylated Myc-Pin1 complex must therefore be considered as an ensemble of conformations where, although one orientation of c-Myc in the interdomain cleft is prevalent, a wide variety of chain-wrapping modes around Pin1 are allowed in the bound ensemble (Figures 6 and S5). Redistribution among multiple states within the bound ensemble could well occur on Myc phosphorylation, which could also increase the propensities for MBI binding to the Pin1 interdomain cleft, as suggested by Pin1WW binding to both MB0 and MBI upon Ser62 phosphorylation (Figure 3E) as well as by previous crystal structures showing phosphorylated short peptides or peptide analogs bound in the interdomain cleft (Matena et al., 2013; Verdecia et al., 2000). The juggling of substrates on and off the active site, as well as the enzymatic efficiency (De et al., 2012), may thus be controlled by dynamically shifting the ensemble of interdomain Pin1 bound states, as indicated by the requirement of both domains for efficient catalysis and by the multiple modes of Myc-Pin1 recognition identified here. Furthermore, the fuzzy complex with a flexible 30-residue (>60 Å) linker between the Myc MB0 and MBI binding sites (Figure 6), together with the dynamics of the bound complex (Figure 4), will allow for facile kinase/phosphatase access to Ser/Thr-Pro sites in MBI with maintained MB0 association of Pin1 throughout the regulatory cycle (Figure 1). Our findings, together with the possible presence of distal motifs in other Pin1-dependent pro-proliferative transcription factors (Figure S7E), suggest that dynamic pre-recognition of distal motifs by Pin1 could play a more general role in the timing of cellular events in growth and differentiation.

EXPERIMENTAL PROCEDURES

Bioinformatics Sequence Evaluation

A Profile Hidden Markov Model logo, describing sequence conservation, was constructed for c-Myc1–88 as described in Supplemental Experimental Procedures.

Protein Expression, Purification, and Phosphorylation

Human Pin1 constructs for Escherichia coli overexpression of full-length Pin1 (1–163), Pin1WW (6–39), and Pin1PPIase (50–163) (Bayer et al., 2003) were kindly provided by Prof. Bayer (University of Duisburg-Essen). Pin1 and Myc proteins were prepared according to previously published protocols (Andresen et al., 2012; Bayer et al., 2003) with only minor modifications (see Supplemental Experimental Procedures). Phosphorylation of c-Myc1–88 in vitro was performed using CDK2/CyclinA2 kinase (Promega) and analyzed by liquid chromatography-tandem mass spectrometry as detailed in Supplemental Experimental Procedures.

NMR Spectroscopy

All experiments were performed at 15°C using a 600-MHz Varian INOVA spectrometer equipped with a cryoprobe. c-Myc1–88 and pSer62-c-Myc1–88 concentrations were as low as permitted by the sensitivity of the experiments to limit possible effects of self-association: 80–115 μM in HNCO titrations, 180 μM in Pin1 titrations, and 250 μM in relaxation measurements. Samples for HNCOs and relaxation experiments of the Myc-Pin1 complex contained 1 mol equivalent of c-Myc1–88 (unphosphorylated or phosphorylated) and 2 mol equivalents of Pin1, Pin1WW, or Pin1PPIase. Details concerning recorded spectra, data processing, and relaxation analysis are described in the Supplemental Experimental Procedures. Assignments are deposited in the BMRB database with access codes BMRB: 26662 (c-Myc1–88) and BMRB: 26663 (pSer62-c-Myc1–88).

Surface Plasmon Resonance Experiments

SPR measurements were performed at 25°C using a Biacore 3000 instrument. The sensorgrams were all adequately fitted by a 1:1 Langmuir model, which was thus employed to derive apparent Kds for interactions of intact Pin1 and its subdomains with c-Myc1–88, pSer62-c-Myc1–88, or c-Myc1–88 mutants. Details on measurement conditions and sensorgram evaluations are given in Supplemental Experimental Procedures and Figure S2.

Structural Modeling Guided by Experimental Constraints

As detailed in Supplemental Experimental Procedures, Rosetta FlexPepDock (Raveh et al., 2011) was used to dock c-Myc10–35 and c-Myc40/45–70 fragments to Pin1 in a Monte Carlo approach. First, c-Myc10–35 was modeled using constraints from HNCO experiments involving either of the separate Pin1WW and Pin1PPIase domains. The ensemble model was then extended to include MBI by first modeling c-Myc40–70 and c-Myc45–70 separately using the same procedure, but with constraints from HNCO experiments involving both Pin1 domains. By joining these docked ensembles with those of c-Myc10–35 by loop modeling (Wang et al., 2007; Qian et al., 2007) and subsequent refinement, a joint c-Myc10–70 MB0+MBI ensemble was obtained. The docking was guided by experimental constraints derived from significant signals in CSP and HNCO experiments. The threshold for Pin1 ΔCSP signals was set using a σ filter with 2σ cutoff (Schumann et al., 2007), whereas the c-Myc1–88 HNCO ratio thresholds for WW/PPIase and intact Pin1 binding were set at below 0.4 and 0.2, respectively, to qualitatively reflect the observed binding pattern. Above (Pin1 ΔCSP) and below (c-Myc1–88 HNCOs) the thresholds, each set of signals was linearly rescaled between 0.0 and 1.0.

Cell Culture, Transfection, Immunoprecipitations, and Biological Assays

Cell culture, Myc immunoprecipitation, and luciferase assays were performed as previously described (Yeh et al., 2004; Arnold et al., 2009; Farrell et al., 2013). ChIP experiments were performed based on previous studies (Farrell et al., 2013) with slight modifications. The immunoprecipitated DNA were purified with the QIAquick PCR purification kit and used for qPCR analysis with specified primers as described previously (Farrell et al., 2013). Antibodies used for western blotting were anti-V5 mouse monoclonal antibody (Invitrogen), anti-FLAG M2 mouse monoclonal antibody (Sigma), anti-β-actin antibody (ab8227), and anti-c-Myc (phosphor S62) antibody (ab78318). For colony formation assays, expression vectors containing c-Myc (WT or mutant) and H-rasG12V were used for transfections in REF52 cells. 72 hr after transfection, cells were replated at low density. Following replating, cells were maintained in DMEM containing 4% fetal bovine serum, and the colonies were visualized by staining with crystal violet blue 2 weeks later.

Detailed protocols are provided in Supplemental Experimental Procedures.

Supplementary Material

Supplemental

Highlights.

  • Pin1 binds unphosphorylated Myc distant from the isomerization target site

  • Binding site mutations trap Myc in a stable, transcriptionally less active state

  • A Myc-Pin1 “fuzzy complex” is described by NMR and molecular simulations

  • The first structure of Pin1 in complex with a long, disordered target peptide

Acknowledgments

We acknowledge Drs. Cecilia Andresen, Veronika Csizmok, Julie Forman-Kay, and Linda Penn for critical discussion, Dr. Peter Bayer for sharing human Pin1, Pin1WW, and Pin1PPIase expression constructs, Marie Roth for protein production, the Swedish NMR Center (SNC) and SWEDSTRUCT for access to NMR spectrometers, and the Knut and Alice Wallenberg Foundation for equipment grants to LiU and SNC. Project funding is acknowledged from the Swedish Cancer Foundation (M.S.), the Swedish Child Cancer Foundation (M.S.), the Carl Trygger foundation (M.S.), the LiU Cancer Research Network (M.S.), the Swedish Research Council (S.C., P.L., M.S., B.W.), and the NCI R01s CA129040 and CA100855 (R.S.).

Footnotes

AUTHOR CONTRIBUTIONS

Conceptualization: M.S., R.S., and S.H. Methodology: M.A., P.L., R.P., M.S., R.S., Y.S., and B.W. Investigation: S.H., M.M., R.P., Y.S., J.M., M.E., J.K., and M.S. Writing—original draft: M.S., S.H., M.M., R.P., R.S., Y.S., and B.W. Reviewing and editing: M.S., M.A., M.M., R.S., R.P., and B.W. Supervision: M.S., M.A., S.C., P.L., R.S., and B.W. Project administration: M.S. Co-first authors equally contributed Myc NMR and SPR (S.H.), Pin1 NMR and isothermal titration calorimetry (M.M.), modeling (R.P.), and cell biology (Y.S.).

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures, seven figures, and one movie and can be found with this article online at http://dx.doi.org/10.1016/j.str.2015.10.010.

References

  1. Ahlner A, Carlsson M, Jonsson BH, Lundström P. PINT: a software for integration of peak volumes and extraction of relaxation rates. J Biomol NMR. 2013;56:191–202. doi: 10.1007/s10858-013-9737-7. [DOI] [PubMed] [Google Scholar]
  2. Andresen C, Helander S, Lemak A, Farès C, Csizmok V, Carlsson J, Penn LZ, Forman-Kay JD, Arrowsmith CH, Lundström P, Sunnerhagen M. Transient structure and dynamics in the disordered c-Myc transactivation domain affect Bin1 binding. Nucleic Acids Res. 2012;40:6353–6366. doi: 10.1093/nar/gks263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arnold HK, Sears RC. Protein phosphatase 2A regulatory subunit B56alpha associates with c-myc and negatively regulates c-myc accumulation. Mol Cell Biol. 2006;26:2832–2844. doi: 10.1128/MCB.26.7.2832-2844.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Arnold HK, Zhang X, Daniel CJ, Tibbitts D, Escamilla-Powers J, Farrell A, Tokarz S, Morgan C, Sears RC. The Axin1 scaffold proteinpromotes formation of a degradation complex for c-Myc. EMBO J. 2009;28:500–512. doi: 10.1038/emboj.2008.279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bahram F, von der Lehr N, Cetinkaya C, Larsson LG. c-Myc hot spot mutations in lymphomas result in inefficient ubiquitination and decreased proteasome-mediated turnover. Blood. 2000;95:2104–2110. [PubMed] [Google Scholar]
  6. Bayer E, Goettsch S, Mueller JW, Griewel B, Guiberman E, Mayr LM, Bayer P. Structural analysis of the mitotic regulator hPin1 in solution: insights into domain architecture and substrate binding. J Biol Chem. 2003;278:26183–26193. doi: 10.1074/jbc.M300721200. [DOI] [PubMed] [Google Scholar]
  7. Bozoky Z, Krzeminski M, Chong PA, Forman-Kay JD. Structural changes of CFTR R region upon phosphorylation: a plastic platform for intramolecular and intermolecular interactions. FEBS J. 2013a;280:4407–4416. doi: 10.1111/febs.12422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bozoky Z, Krzeminski M, Muhandiram R, Birtley JR, Al-Zahrani A, Thomas PJ, Frizzell RA, Ford RC, Forman-Kay JD. Regulatory R region of the CFTR chloride channel is a dynamic integrator of phospho-dependent intra- and intermolecular interactions. Proc Natl Acad Sci USA. 2013b;110:E4427–E4436. doi: 10.1073/pnas.1315104110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brown NR, Noble MEM, Endicott JA, Johnson LN. The structural basis for specificity of substrate and recruitment peptides for cyclin-dependent kinases. Nat Cell Biol. 1999;1:438–443. doi: 10.1038/15674. [DOI] [PubMed] [Google Scholar]
  10. Chang DW, Claassen GF, Hann SR, Cole MD. The c-Myc transactivation domain is a direct modulator of apoptotic versus proliferative signals. Mol Cell Biol. 2000;20:4309–4319. doi: 10.1128/mcb.20.12.4309-4319.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cowling VH, Cole MD. Mechanism of transcriptional activation by the Myc oncoproteins. Semin Cancer Biol. 2006;16:242–252. doi: 10.1016/j.semcancer.2006.08.001. [DOI] [PubMed] [Google Scholar]
  12. Daum S, Lücke C, Wildemann D, Schiene-Fischer C. On the benefit ofbivalencyinpeptide ligand/pin1 interactions. J Mol Biol. 2007;374:147–161. doi: 10.1016/j.jmb.2007.09.019. [DOI] [PubMed] [Google Scholar]
  13. De S, Greenwood AI, Rogals MJ, Kovrigin EL, Lu KP, Nicholson LK. Complete thermodynamic and kinetic characterization of the isomer-specific interaction between Pin1-WW domain and the amyloid precursor protein cytoplasmic tail phosphorylated at Thr668. Biochemistry. 2012;51:8583–8596. doi: 10.1021/bi3008214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Duncan KE, Dempsey BR, Killip LE, Adams J, Bailey ML, Lajoie GA, Litchfield DW, Brandl CJ, Shaw GS, Shilton BH. Discovery and characterization of a nonphosphorylated cyclic peptide inhibitor of the peptidylprolyl isomerase, Pin1. J Med Chem. 2011;54:3854–3865. doi: 10.1021/jm200156c. [DOI] [PubMed] [Google Scholar]
  15. Farrell AS, Sears RC. MYC degradation. Cold Spring Harb Perspect Med. 2014;4 doi: 10.1101/cshperspect.a014365. http://dx.doi.org/10.1101/cshperspect.a014365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Farrell AS, Pelz C, Wang X, Daniel CJ, Wang Z, Su Y, Janghorban M, Zhang X, Morgan C, Impey S, et al. Pin1 regulates the dynamics of c-Myc DNA binding to facilitate target gene regulation and oncogenesis. Mol Cell Biol. 2013;33:2930–2949. doi: 10.1128/MCB.01455-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Forman-Kay JD, Mittag T. From sequence and forces to structure, function, and evolution of intrinsically disordered proteins. Structure. 2013;21:1492–1499. doi: 10.1016/j.str.2013.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fuxreiter M, Tompa P. Fuzzy complexes: a more stochastic view of protein function. Adv Exp Med Biol. 2012;725:1–14. doi: 10.1007/978-1-4614-0659-4_1. [DOI] [PubMed] [Google Scholar]
  19. Guo J, Pang X, Zhou HX. Two pathways mediate interdomain allosteric regulation in pin1. Structure. 2015;23:237–247. doi: 10.1016/j.str.2014.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hann SR. Role of post-translational modifications in regulating c-Myc proteolysis, transcriptional activity and biological function. Semin Cancer Biol. 2006;16:288–302. doi: 10.1016/j.semcancer.2006.08.004. [DOI] [PubMed] [Google Scholar]
  21. Hariharan N, Sussman MA. Pin1: a molecular orchestrator in the heart. Trends Cardiovasc Med. 2014;24:256–262. doi: 10.1016/j.tcm.2014.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hydbring P, Bahram F, Su Y, Tronnersjö S, Högstrand K, von der Lehr N, Sharifi HR, Lilischkis R, Hein N, Wu S, et al. Phosphorylation by Cdk2 is required for Myc to repress Ras-induced senescence in cotransformation. Proc Natl Acad Sci USA. 2010;107:58–63. doi: 10.1073/pnas.0900121106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jacobs DM, Saxena K, Vogtherr M, Bernado P, Pons M, Fiebig KM. Peptide binding induces large scale changes in inter-domain mobility in human Pin1. J Biol Chem. 2003;278:26174–26182. doi: 10.1074/jbc.M300796200. [DOI] [PubMed] [Google Scholar]
  24. Lee TH, Chen CH, Suizu F, Huang P, Schiene-Fischer C, Daum S, Zhang YJ, Goate A, Chen RH, Zhou XZ, et al. Death-associated protein kinase 1 phosphorylates Pin1 and inhibits its prolyl isomerase activity and cellular function. Mol Cell. 2011;42:147–159. doi: 10.1016/j.molcel.2011.03.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Legouy E, DePinho R, Zimmerman K, Collum R, Yancopoulos G, Mitsock L, Kriz R, Alt FW. Structure and expression of the murine L-myc gene. EMBO J. 1987;6:3359–3366. doi: 10.1002/j.1460-2075.1987.tb02657.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lin CY, Lovén J, Rahl PB, Paranal RM, Burge CB, Bradner JE, Lee TI, Young RA. Transcriptional amplification in tumor cells with elevated c-Myc. Cell. 2012;151:56–67. doi: 10.1016/j.cell.2012.08.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Liou YC, Zhou XZ, Lu KP. Prolyl isomerase Pin1 as a molecular switch to determine the fate of phosphoproteins. Trends Biochem Sci. 2011;36:501–514. doi: 10.1016/j.tibs.2011.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lu KP, Zhou XZ. The prolyl isomerase PIN1: a pivotal new twist in phosphorylation signalling and disease. Nat Rev Mol Cell Biol. 2007;8:904–916. doi: 10.1038/nrm2261. [DOI] [PubMed] [Google Scholar]
  29. Lu PJ, Zhou XZ, Shen M, Lu KP. Function of WW domains as phosphoserine- or phosphothreonine-binding modules. Science. 1999;283:1325–1328. doi: 10.1126/science.283.5406.1325. [DOI] [PubMed] [Google Scholar]
  30. Lu KP, Liou YC, Zhou XZ. Pinning down proline-directed phosphorylation signaling. Trends Cell Biol. 2002a;12:164–172. doi: 10.1016/s0962-8924(02)02253-5. [DOI] [PubMed] [Google Scholar]
  31. Lu PJ, Zhou XZ, Liou YC, Noel JP, Lu KP. Critical role of WW domain phosphorylation in regulating phosphoserine binding activity and Pin1 function. J Biol Chem. 2002b;277:2381–2384. doi: 10.1074/jbc.C100228200. [DOI] [PubMed] [Google Scholar]
  32. Lu KP, Finn G, Lee TH, Nicholson LK. Prolyl cis-trans isomerization as a molecular timer. Nat Chem Biol. 2007;3:619–629. doi: 10.1038/nchembio.2007.35. [DOI] [PubMed] [Google Scholar]
  33. Lukhele S, Bah A, Lin H, Sonenberg N, Forman-Kay JD. Interaction of the eukaryotic initiation factor 4E with 4E-BP2 at a dynamic bipartite interface. Structure. 2013;21:2186–2196. doi: 10.1016/j.str.2013.08.030. [DOI] [PubMed] [Google Scholar]
  34. Matena A, Sinnen C, van den Boom J, Wilms C, Dybowski JN, Maltaner R, Mueller JW, Link NM, Hoffmann D, Bayer P. Transient domain interactions enhance the affinity of the mitotic regulator pin1 toward phosphorylated peptide ligands. Structure. 2013;21:1769–1777. doi: 10.1016/j.str.2013.07.016. [DOI] [PubMed] [Google Scholar]
  35. Meyer N, Penn LZ. Reflecting on 25 years with MYC. Nat Rev Cancer. 2008;8:976–990. doi: 10.1038/nrc2231. [DOI] [PubMed] [Google Scholar]
  36. Min SH, Lau AW, Lee TH, Inuzuka H, Wei S, Huang P, Shaik S, Lee DY, Finn G, Balastik M, et al. Negative regulation of the stability and tumor suppressor function of Fbw7 by the Pin1 prolyl isomerase. Mol Cell. 2012;46:771–783. doi: 10.1016/j.molcel.2012.04.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mittag T, Orlicky S, Choy WY, Tang X, Lin H, Sicheri F, Kay LE, Tyers M, Forman-Kay JD. Dynamic equilibrium engagement of a polyvalent ligand with a single-site receptor. Proc Natl Acad Sci USA. 2008;105:17772–17777. doi: 10.1073/pnas.0809222105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Namanja AT, Peng T, Zintsmaster JS, Elson AC, Shakour MG, Peng JW. Substrate recognition reduces side-chain flexibility for conserved hydrophobic residues in human Pin1. Structure. 2007;15:313–327. doi: 10.1016/j.str.2007.01.014. [DOI] [PubMed] [Google Scholar]
  39. Namanja AT, Wang XJ, Xu B, Mercedes-Camacho AY, Wilson KA, Etzkorn FA, Peng JW. Stereospecific gating of functional motions in Pin1. Proc Natl Acad Sci USA. 2011;108:12289–12294. doi: 10.1073/pnas.1019382108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Nie Z, Hu G, Wei G, Cui K, Yamane A, Resch W, Wang R, Green DR, Tessarollo L, Casellas R, et al. c-Myc is a universal amplifier of expressed genes in lymphocytes and embryonic stem cells. Cell. 2012;151:68–79. doi: 10.1016/j.cell.2012.08.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Peng JW. Investigating dynamic interdomain allostery in Pin1. Biophys Rev. 2015;7:239–249. doi: 10.1007/s12551-015-0171-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Qian B, Raman S, Das R, Bradley P, McCoy AJ, Read RJ, Baker D. High resolution protein structure prediction and the crystallographic phase problem. Nature. 2007;450:259–264. doi: 10.1038/nature06249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ranganathan R, Lu KP, Hunter T, Noel JP. Structural and functional analysis of the mitotic rotamase Pin1 suggests substrate recognition is phosphorylation dependent. Cell. 1997;89:875–886. doi: 10.1016/s0092-8674(00)80273-1. [DOI] [PubMed] [Google Scholar]
  44. Raveh B, London N, Zimmerman L, Schueler-Furman O. Rosetta FlexPepDock ab-initio: simultaneous folding, docking and refinement of peptides onto their receptors. PLoS One. 2011;6:e18934. doi: 10.1371/journal.pone.0018934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Salghetti SE, Kim SY, Tansey WP. Destruction of Myc by ubiquitin-mediated proteolysis: cancer-associated and transforming mutations stabilize Myc. EMBO J. 1999;18:717–726. doi: 10.1093/emboj/18.3.717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schumann FH, Riepl H, Maurer T, Gronwald W, Neidig KP, Kalbitzer HR. Combined chemical shift changes and amino acid specific chemical shift mapping of protein-protein interactions. J Biomol NMR. 2007;39:275–289. doi: 10.1007/s10858-007-9197-z. [DOI] [PubMed] [Google Scholar]
  47. Sears RC. The life cycle of c-myc: from synthesis to degradation. Cell Cycle. 2004;3:1133–1137. [PubMed] [Google Scholar]
  48. Tu WB, Helander S, Pilstål R, Hickman KA, Lourenco C, Jurisica I, Raught B, Wallner B, Sunnerhagen M, Penn LZ. Myc and its interactors take shape. Biochim Biophys Acta. 2015;1849:469–483. doi: 10.1016/j.bbagrm.2014.06.002. [DOI] [PubMed] [Google Scholar]
  49. Verdecia MA, Bowman ME, Lu KP, Hunter T, Noel JP. Structural basis for phosphoserine-proline recognition by group IV WW domains. Nat Struct Biol. 2000;7:639–643. doi: 10.1038/77929. [DOI] [PubMed] [Google Scholar]
  50. Wang C, Bradley P, Baker D. Protein-protein docking with backbone flexibility. J Mol Biol. 2007;373:503–519. doi: 10.1016/j.jmb.2007.07.050. [DOI] [PubMed] [Google Scholar]
  51. Wang X, Cunningham M, Zhang X, Tokarz S, Laraway B, Troxell M, Sears RC. Phosphorylation regulates c-Myc’s oncogenic activity in the mammary gland. Cancer Res. 2011;71:925–936. doi: 10.1158/0008-5472.CAN-10-1032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Werner-Allen JW, Lee CJ, Liu P, Nicely NI, Wang S, Greenleaf AL, Zhou P. cis-Proline-mediated Ser(P)5 dephosphorylation by the RNA polymerase II C-terminal domain phosphatase Ssu72. J Biol Chem. 2011;286:5717–5726. doi: 10.1074/jbc.M110.197129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wilson KA, Bouchard JJ, Peng JW. Interdomain interactions support interdomain communication in human Pin1. Biochemistry. 2013;52:6968–6981. doi: 10.1021/bi401057x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wintjens R, Wieruszeski JM, Drobecq H, Rousselot-Pailley P, Buée L, Lippens G, Landrieu I. 1H NMR study on the binding of Pin1 Trp-Trp domain with phosphothreonine peptides. J Biol Chem. 2001;276:25150–25156. doi: 10.1074/jbc.M010327200. [DOI] [PubMed] [Google Scholar]
  55. Yeh E, Cunningham M, Arnold H, Chasse D, Monteith T, Ivaldi G, Hahn WC, Stukenberg PT, Shenolikar S, Uchida T, et al. A signalling pathway controlling c-Myc degradation that impacts oncogenic transformation of human cells. Nat Cell Biol. 2004;6:308–318. doi: 10.1038/ncb1110. [DOI] [PubMed] [Google Scholar]
  56. Zhang M, Wang XJ, Chen X, Bowman ME, Luo Y, Noel JP, Ellington AD, Etzkorn FA, Zhang Y. Structural and kinetic analysis of prolyl-isomerization/phosphorylation cross-talk in the CTD code. ACS Chem Biol. 2012a;7:1462–1470. doi: 10.1021/cb3000887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zhang X, Farrell AS, Daniel CJ, Arnold H, Scanlan C, Laraway BJ, Janghorban M, Lum L, Chen D, Troxell M, et al. Mechanistic insight into Myc stabilization in breast cancer involving aberrant Axin1 expression. Proc Natl Acad Sci USA. 2012b;109:2790–2795. doi: 10.1073/pnas.1100764108. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental
NIHMS768189-supplement-Supplemental.pdf (3.7MB, pdf)

RESOURCES