Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2014 Sep 10;9(9):e107428. doi: 10.1371/journal.pone.0107428

MID1 Catalyzes the Ubiquitination of Protein Phosphatase 2A and Mutations within Its Bbox1 Domain Disrupt Polyubiquitination of Alpha4 but Not of PP2Ac

Haijuan Du 1, Kuanlin Wu 1, Alma Didoronkute 1, Marcus V A Levy 1, Nimish Todi 1, Anna Shchelokova 1, Michael A Massiah 1,*
Editor: Adriano Marchese2
PMCID: PMC4160256  PMID: 25207814

Abstract

MID1 is a microtubule-associated protein that belongs to the TRIM family. MID1 functions as an ubiquitin E3 ligase, and recently was shown to catalyze the polyubiquitination of, alpha4, a protein regulator of protein phosphatase 2A (PP2A). It has been hypothesized that MID1 regulates PP2A, requiring the intermediary interaction with alpha4. Here we report that MID1 catalyzes the in vitro ubiquitination of the catalytic subunit of PP2A (PP2Ac) in the absence of alpha4. In the presence of alpha4, the level of PP2Ac ubiquitination is reduced. Using the MID1 RING-Bbox1-Bbox2 (RB1B2) construct containing the E3 ligase domains, we investigate the functional effects of mutations within the Bbox domains that are identified in patients with X-linked Opitz G syndrome (XLOS). The RB1B2 proteins harboring the C142S, C145T, A130V/T mutations within the Bbox1 domain and C195F mutation within the Bbox2 domain maintain auto-polyubiquitination activity. Qualitatively, the RB1B2 proteins containing these mutations are able to catalyze the ubiquitination of PP2Ac. In contrast, the RB1B2 proteins with mutations within the Bbox1 domain are unable to catalyze the polyubiquitination of alpha4. These results suggest that unregulated alpha4 may be the direct consequence of these natural mutations in the Bbox1 domain of MID1, and hence alpha4 could play a greater role to account for the increased amount of PP2A observed in XLOS-derived fibroblasts.

Introduction

MID1 is a member of the tripartite motif (TRIM) protein family consisting of the N-terminal RING, two B-box, and Coiled-Coil (RBCC) domains (Figure 1A). The RING and Bbox are the domains that possess E3 ligase functionality and important for substrate specificity [1], [2]. Recently, we demonstrate that the MID1 Bbox domains are essential for optimal MID1 E3 ligase functionality [1], [2]. The MID1 Bbox domains exhibit weaker E3 ligase compared the RING domain, as evident by autoubiquitination activities. When the RING domain is natively linked to Bbox1 or both Bbox domains, the rate of E3 ligase activity is increased by at least 10-fold, suggesting that the Bbox domains may have, in part, E4 ligase functionality [1], [2].

Figure 1. Natural mutations of MID1 and birth anomalies.

Figure 1

Open in a new tab

A. Schematic representation of MID1 showing the locations of the various types of mutations. MID1 is a member of the TRIM family of proteins, consisting of seven distinct domains. The RING and Bbox domains coordinate two zinc ions each and adopt ββα-RING folds. The coiled-coil (CC) domain is important for MID1 dimerization. The CC contains a region that is required for microtubule association, denoted as the COS box [35]. The C-terminal half contains the fibronectin type III (FNIII) and SPRY/B30.2 domains. Superimposed on the schematic structure are the four types natural mutations observed with patients with X-linked Opitz G Syndrome. B. Summary of all reported mutations, to date, of MID1 and the associated anomalies [27], [28], [29], [30], [31], [32], [33], [34]. The four types of mutations observed are color-coded (red = point mutation, blue = frame shifts, magenta = termination/truncation, and black = deletion). Each column lists the different types of mutations (color coded) associated with each domain and grouped with the reported birth defects. X indicates a truncation mutation. The most common defects are hypertelorism, hyperspadias and cleft lip/palate based on the number of mutations associated with these defects. Other craniofacial anomalies that are not listed included low set ears, widows peak, heart defects and small jaw or chin to name a few.

The Bbox domains, which coordinate two zinc-ions and adopt ββα-RING folds, are required for the efficient ubiquitination of alpha4, an oncogenic protein initially identified as an immunoglobin binding protein (IgBP-1). MID1 RING domain catalyzes only the monoubiquitination of alpha4, but the RING in tandem with Bbox1 or both Bbox domains (RING-Bbox1 and RING-Bbox1-Bbox2) catalyzes the polyubiquitination of alpha4. Neither the individual nor the tandem Bbox domains catalyzes the polyubiquitination of alpha4.

Alpha4 down-regulates protein phosphatase 2A (PP2A) in numerous cellular pathways, including the target of rapamycin (TOR) pathway [3], [4], [5]. PP2A is a major serine/threonine phosphatase that regulates cellular processes associated with metabolism, cell-cycle progression, and apoptosis [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]. PP2A is a heterotrimeric complex consisting of a catalytic subunit (PP2Ac), a scaffolding (PP2A/a, PR65) subunit, and numerous regulatory (PP2A/b) subunits [9], [18], [19], [20]. Alpha4 binds the PP2Ac subunit and displaces the scaffold and regulatory subunits [5], [21]. Alpha4 can sequester PP2Ac in a stable reserve, supposedly protecting it from degradation, and releasing it to counter stressful stimuli to the cell [4], [22].

The Bbox1 domain binds alpha4 and this interaction is hypothesized to recruit PP2Ac to MID1 for ubiquitin-mediated degradation [17], [23], [24], [25]. XLOS-derived fibroblast cells show an increased cellular PP2Ac concentration, suggesting that MID1 and alpha4 are important for PP2Ac homeostasis [17], [24], [25], [26]. However, to date, there has not been any evidence that MID1 directly targets PP2Ac for ubiquitin-mediated degradation.

In the past decade, there have been numerous reports identifying mutations of MID1 from individuals with ventral midline anomalies that affect the brain, face, heart and genitalia (Figure 1B) [27], [28], [29], [30], [31], [32], [33], [34]. Individuals with these defects are diagnosed with X-linked Opitz G Syndrome (XLOS). The MID1 protein from XLOS patients is observed to have point mutations, frame-shifts, deletions, and truncations in all its domains except for the region encompassing the RING domain (Figure 1). Despite the diversity of mutations, there is no pattern or correlation of birth defects associated with specific mutations. The most common anomalies include hypertelorism, hypospadias, clefts of the lip and/or palate (CL/P), and laryngotracheoesophageal (LTE) cleft (Figure 1B). These anomalies are associated with mutations within the MID1 Bbox domains.

In vivo fluorescence studies using MID fused to a green fluorescent protein tag (GFP-MID1) reveal that a number of XLOS-observed mutations within the C-terminal half of MID1 cause MID1 to form cytoplasmic ‘clumps’ instead of normally being associated with the microtubules [27], [35], [36], [37]. Deletion of the SPRY/B30.2 domain (MID1delCTD) results in MID1 aggregation [17], [35], [38]. The overexpression of exogenous MID1delCTD in COS-1 and MDCK cells have been shown to sequester native endogenous MID1 through aggregation and decrease the activity of native MID1 [1], [27], [35], [36], [39]. In contrast, there is no information about the effects of mutations within the N-terminal half of MID1, specifically within the Bbox domains.

In this article, we present data showing that MID1 catalyzes the ubiquitination of PP2Ac in the absence of alpha4. Using the MID1 RB1B2 protein construct, we observed that the RB1B2 proteins with natural mutations within the Bbox1 domain still catalyze the ubiquitination of PP2A but not the polyubiquitination of alpha4. These results suggest that alpha4 may play a greater role to account for the increase of PP2A observed in XLOS-derived fibroblasts.

Methods and Materials

Plasmid construction for recombinant protein production

The DNA encoding the MID1 RING-Bbox1-Bbox2 (RB1B2, residues 1–214) was cloned into an Expresso T7 plasmid (Lucigen Corp., Middleton, WI) containing a sequence for a C-terminal His6-tag. Mutations within the DNA for Bbox1 (A130S/T/V, C142T, C145S) and Bbox2 (C195F) domains were performed using standard PCR mutagenesis protocol using the RB1B2 domain construct. The His6-tagged alpha4 were cloned as previously described [1], [2]. The cDNA of PP2Ac was cloned from human PP2A into the pET151/D-TOPO vector with a N-terminal His6-tagged using the following primers: 5′CACCATGGACGAGAAGGTGTTCACC3′ and 5′TTACAGGAAGTAGTCTGGGGTACGACG3′. DNA sequencing confirmed the integrity of DNA inserts and mutations before the plasmids were transformed into BL21(DE3) cells.

The BL21(DE3) cells expressing the wild type and mutants MID1 RB1B2 constructs, alpha4, and PP2Ac were grown in LB media at 37°C to an OD600 of ∼0.5 and induced with 0.2 mM isopropyl β-D-1-thiogalactopyranoside for 18 hr at 18°C. The RB1B2 protein and the various mutants were purified as previously described [1], [2]. Briefly, harvested cells were stored at −80°C before being resuspended in buffer 50 mM Tris, 200 mM NaCl (pH 7.5), with protease inhibitors and lysed with sarkosyl and sonication. Lysates were clarified by centrifugation at 20,000×g at 4°C for 30 min. The RB1B2, alpha4, and PP2Ac proteins were purified using standard the Ni2+-NTA affinity chromatography approach.

In vitro ubiquitination assay using purified recombinant proteins

Human Ub-activating enzyme E1, human Ub-conjugating enzymes (UbcH5b), UbcH (E2) enzyme, and biotinylated ubiquitin were purchased from BostonBiochem (Bio-Techne Inc., Minneapolis, MN). The autoubiquitination assays of the MID1 E3 ligase domains were performed in a total volume of 30 µL. The mixture contained 1 unit (generates 1 µmol of phosphate per minute) of thermostable inorganic pyrophosphatase (New England Biolabs), 1 mM DTT, 5 mM ATP-Mg2+, 0.125 µM E1, 2.5 µM E2 (UbcH5b/UBE2D2), and 2 µM biotinylated ubiquitin (Ub) in 20 mM Tris–HCl (pH 7.5), and purified recombinant proteins. The ubiquitination assays were incubated at 30°C for 90 minutes and the reaction terminated with 2x SDS-PAGE non-reducing sample loading buffer. The ubiquitination of alpha4 and PP2Ac were performed similarly to the autoubiquitination reaction by adding these proteins to the reaction mixture and using non-biotinylated Ub. The reactions were performed at 30°C for 4 hrs. The reaction products were subsequently evaluated by SDS-PAGE and chemiluminescent western blot (WB) analysis. For the autoubiquitination assay, the antibody used was avidin, which is specific for biotin. For the assays containing the alpha4 and PP2Ac proteins, the antibodies were specific for the C-terminal region of alpha4 (cat #sc-32174) and PP2Ac (cat #sc-376673) and were purchased from Santa Cruz Biotechnology. Control reactions were performed by omitting individual components from the assay. WB digital images were acquired with a Syngene G5:Box chemidoc system.

Results

MID1 catalyzed the ubiquitination of protein phosphatase 2A catalytic subunit

To understand how mutations within the Bbox domains of MID1 might affect the regulation of PP2Ac, we began by determining whether MID1 could directly catalyze the ubiquitination of PP2Ac. To this end, we performed in vitro ubiquitination studies using recombinant full-length MID1 and PP2Ac. MID1 was constructed with a maltose-binding protein (MBP) tag at its N-terminus to enhance solubility.

As shown in Figure 2A(i, ii), the full-length MID1 protein catalyzed the mono- and di-ubiquitination of PP2Ac as predominant products. The western blot also revealed the presence of a weak intensity smearing pattern of high-molecular weight products indicative of polyubiquitinated PP2Ac (Figure 2A(i)). As part of the control, omission of ATP and MID1 resulted in loss of the mono-, di- and polyubiquitinated PP2Ac bands, confirming that MID1 catalyzes the ubiquitination of PP2Ac (Figure 2A). The intensity of the mono-ubiquitinated PP2Ac band was greater than that of the di-ubiquitinated band, suggesting that Ub addition PP2Ac may be sequential.

Figure 2. MID1 catalyzed the ubiquitination of PP2Ac.

Figure 2

Open in a new tab

A. Western-blot (WB) showing in vitro ubiquitination of PP2Ac by MBP-tagged full-length MID1. (1) Lanes 1 and 2 show the results of ubiquitination assay for control reactions. In lane 1 ATP was omitted (no reaction) and in lane 2, only MBP was used. Lane 3 contains the results for the MPB-tagged MID1-catalyzed reaction. Two strong shifted bands corresponding to mono- and di-ubiquitinated PP2Ac and a faint smearing pattern indicative of poly-ubiquitinated PP2Ac were observed. (ii) Lane 1 shows the Biorad visible marker and lane 2 shows a control reaction in which MBP-MID1 was omitted. Lane 3 shows the ubiquitinated PP2Ac to confirm the molecular weights of the mono- and di-ubiquitinated products. B. (i) Ubiquitination of PP2Ac by MBP-tagged MID1 in the absence and presence of full-length alpha4. (ii) Ubiquitination of PP2Ac with increasing amounts of alpha4. The amounts of alpha4 were present at 1:1, 1:2 and 1:4 molar ratios with PP2Ac. C. Ubiquitination of PP2Ac by the MID1 RB1B2 protein (RING-Bbox1-Bbox2, MID1 residues 1–214). Lanes 1–6 show the results of control experiments in which a specific component of the assay was omitted, including the E1 activating enzyme (E1) (lane 1). Mono- and di-ubiquitinated PP2Ac were observed in lane 7, with all components of the reaction present. Background noise precludes detection of polyubiquitinated products. Antibody was specific for PP2Ac. D. Ubiquitination of PP2Ac by various constructs of the three N-terminal domains. These domains were expressed and purified similarly, and used at roughly the same concentration for the assays. Lane 1 uses the Bbox1-Bbox2 construct that includes the amino acids from the linker region between the RING and Bbox1 domains (LB1B2, residues 71–115). Lane 2 uses a RING domain (residues 1–92) that also includes a large position of the linker region. The RB1 and RB1B2 constructs consist of residues 1–164 and 1–214 respectively. E. Ubiquitination of PP2Ac in the presence of 10 different E2 conjugating enzymes.

Alpha4 down-regulated the level of PP2Ac ubiquitination

The general consensus in the literature is that alpha4, which interacts with PP2Ac and MID1, is required to recruit PP2Ac to the MID1 [5], [15], [17], [40], and therefore may be necessary for efficient ubiquitination of PP2Ac. To test this hypothesis, we performed the ubiquitination assays of PP2Ac in the absence and presence of alpha4 (Figure 2B(i)). The level of PP2Ac ubiquitination was substantially less in the presence of alpha4, as evidence by the decrease of intensities of the mono- and di-ubiquitinated PP2Ac bands (Figure 2B(i, ii)). Given that previous findings suggested that alpha4 was involved in the recruiting of PP2Ac [5], [17], [21], [24], we tested whether targeting of PP2Ac might have a diphasic dependence on alpha4 concentration [41]. We anticipated that the level of ubiquitination of PP2Ac might increase as alpha4 increases but at some point the level should decrease due to competitive binding of the two proteins with MID1. We performed a series of ubiquitination assays, each with increasing amounts of alpha4 compared to PP2Ac. The concentrations of all components of the assays, except alpha4, were kept constant. When the concentrations of alpha4 were less than that of PP2Ac, we did not observe any discernable difference in the intensities of the mono- and di-ubiquitinated PP2Ac bands compared to when alpha4 was absent. At concentrations in which alpha4 was present at equimolar or greater than PP2Ac, we observed roughly the same amount of decrease in mono- and di-ubiquitinated PP2Ac (Figure 2B(ii)).

MID1 RB1B2 E3 domains catalyzed ubiquitination of PP2Ac

We had previously demonstrated that the E3 ligase activity of MID1 was localized to the N-terminal RING-Bbox1-Bbox2 (RB1B2) zinc-binding domains [1], [2]. To determine whether these three domains could catalyze the ubiquitination of PP2Ac, we performed the ubiquitination assay with the RB1B2 protein (MID1 residues 1–214). Similar to results obtained with full-length MID1, the RB1B2 protein catalyzed the mono- and di-ubiquitination of PP2Ac (Figure 2C). The high molecular weight smear pattern was not as obvious, suggesting that domains within the C-terminal half of MID1 may contribute to the polyubiquitination of PP2Ac.

To determine how the RING and Bbox domain contributed to targeting and catalyzing the ubiquitination of PP2Ac, we performed the ubiquitination assay with the following domain constructs: RING (MID1 residues 1–92), RING-Bbox1 (RB1, residues 1–164), RING-Bbox1-Bbox2 (residues 1–214) and Bbox1-Bbox2 (LB1B2, residues 71–214). The LB1B2 protein construct included residues from the linker region between the RING and Bbox1 domains (residues 71–110) but not the RING domain. We previously observed that these four fragments have different levels of E3 ligase activity, as part of the standard autoubiquitination assay and in their ability to catalyze the polyubiquitination of alpha4 [1], [2]. As noted, our goal was to determine if similar results would be observed for the ubiquitination PP2Ac and to identify the role of the RING and Bbox domains. These domains were expressed and purified similarly, and used at roughly the same concentration for the assays. The RING domain only facilitated mono-ubiquitination of PP2Ac. Both the RB1 and RB1B2 proteins catalyzed the mono- and di-ubiquitination of PP2Ac (Figure 2D). No ubiquitination of PP2Ac was observed with the LB1B2 protein. These observations are similar to those obtained with the ubiquitination studies of alpha4 by MID1 [1], [2].

In addition to the UbcH5b (UBE2D2) E2 enzyme, our previous studies with MID1 indicated that the RB1B2 protein could interact with multiple E2 enzymes in catalyzing its own polyubiquitination and that of alpha4 [1], [2]. We tested 12 E2 enzymes to determine whether any would be more efficient in facilitating the ubiquitination of PP2Ac. We observed that PP2Ac was ubiquitinated in the presence of UbcH5a-c and UbcH6, similar to results obtained with MID1-catalyzed ubiquitination of alpha4 (Figure 2E) [1].

XLOS-observed MID1 mutations possessed autoubiquitination activities

To investigate how XLOS-observed mutations within the Bbox domains might affect MID1 E3 ligase functionality, we compared the autoubiquitination activity of the mutant RB1B2 constructs with that of wild type RB1B2 protein. Five mutations within the Bbox1 domain (A130T/V/S, C142S, and C145T) and one within the Bbox2 domain (C195F) were designed. The A130S mutation, which is not associated with XLOS, was intended to distinguish the role of the hydroxyl from methyl group of the Thr130 mutation. These mutant RB1B2 proteins were expressed and purified similarly as the wild type RB1B2 protein. All mutant protein constructs were soluble and stable at the experimental conditions used.

As shown with Figure 3, auto-polyubiquitinated products were observed with all of the RB1B2 mutant constructs, similar to the autoubiquitination activities of the wild type RB1B2 protein.

Figure 3. RB1B2 with mutant Bbox domains possessed auto-ubiquitination activity.

Figure 3

Open in a new tab

(i) WB image showing the results of the E3 ligase auto-ubiquitination assay of RB1B2 harboring the C142S, C145T, C195F, and the mutation of Ala130 to Ser, Thr, and Val. Lane 1 does not contain the RB1B2 protein. (ii, iii) The WB images of the bands for the activated E2∼Ub and ubiquitin are shown separately; these bands were part of the same image as (i) but were excised due to their brightness and dynamic range problems for clarity. Antibody used with all four figures was specific for biotinylated ubiquitin. (iv) WB of the stock solutions of the various His-tagged RB1B2 constructs. No His-tagged RB1B2 protein was used in lane 1, which served as control in which the E3 was omitted. This image was taken from a separate gel because the antibody was specific for the His6-tag.

XLOS-observed MID1 mutants catalyze PP2Ac ubiquitination

It is generally acknowledged in the literature that mutations of MID1 are associated with an increase in the amount of PP2Ac, suggesting that MID1 is involved in the regulation of PP2Ac [17], [42]. Therefore, we tested whether mutations within the Bbox domains would disrupt the ubiquitination of PP2Ac. No studies with a mutant RING domain were performed because, to date, no mutation associated with XLOS was reported for the RING domain (Figure 1). We used the RB1B2 construct because the RB1B2 mutants were easier to purify compared to full-length MID1. Moreover, we hypothesize that disruption in the mono- and di-ubiquitination of PP2Ac using the mutant RB1B2 protein would directly correlate with loss of the full-length MID1 to catalyze the ubiquitination of PP2Ac. Qualitatively, all the mutant RB1B2 constructs were observed to catalyze the mono- and di-ubiquitination of PP2Ac, similar to the wild type RB1B2 protein (Figure 4).

Figure 4. RB1B2 with mutant Bbox domains catalyzed the ubiquitination of PP2Ac.

Figure 4

Open in a new tab

WB showing results of the ubiquitination reaction with PP2Ac using RB1B2 containing the various mutations within the Bbox domains. In lane 1, RB1B2 was omitted as control. At the bottom, the ability of the mutant RB1B2 constructs to catalyze the ubiquitination of PP2Ac is qualitatively noted. Antibody was specific against PP2Ac.

XLOS-observed mutations within the Bbox1 domain abolished alpha4 polyubiquitination

In light of these results, we know that mutations within the Bbox domains are associated with physical abnormalities (Figure 1B) and therefore must have some functional effects. As noted, we recently demonstrated that MID1 catalyzed the polyubiquitination of alpha4, leading to its proteasome-mediated degradation [1]. The polyubiquitination was dependent on the interaction of the Bbox1 domain and alpha4. Thus, we anticipate that these mutations might affect the targeting of alpha4.

We performed the ubiquitination assays with full-length alpha4 using the RB1B2 mutants, and the wild type RB1B2 protein as control. The RB1B2 proteins harboring the C142S and C145T mutations were unable to catalyze the polyubiquitination of alpha4. Only a weak band corresponding to mono-ubiquitinated alpha4 was observed (Figure 5A(i, ii)). In contrast, the RB1B2 protein with the C195F mutation within the Bbox2 domain catalyzed the polyubiquitination of alpha4 (Figure 5A(iii)).

Figure 5. XLOS-observed mutations within Bbox1 disrupted polyubiquitination of alpha4.

Figure 5

Open in a new tab

A. Ubiquitination assay of alpha4 was performed with the RB1B2 construct containing (i) C142S, (ii) C145T, and (iii) C195F. B. Ubiquitination of alpha4 by RB1B2 with the (i) A130T, (ii) A130V and (iii) A130S mutations within the Bbox1 domain. The amount of alpha4 used for the assay with the A130T RB1B2 mutant was higher than that used for the A130V mutant assay. Nonetheless, the overall results for the two RB1B2 mutants are very similar. For comparison, wild type RB1B2 was shown to catalyze polyubiquitination of alpha4, as indicated by the smear/ladder pattern designated with the labeled alpha4∼Ubn. Negative controls in which RB1B2 protein is omitted are shown for each reaction, as well. At the bottom of each figure, the ability of the mutant RB1B2 constructs to catalyze the ubiquitination of alpha4 is qualitatively noted. Antibody was specific for the C-terminal region of alpha4.

The RB1B2 construct containing the A130T or A130V mutations was unable to catalyze the polyubiquitination of alpha4 (Figure 5B(i, ii)). With both mutants, a strong mono-ubiquitinated alpha4 band was observed. A weak band corresponding to di-ubiquitinated alpha4 was also observed. Even though we added slightly greater amount of alpha4 in the ubiquitination assay with A130T (Figure 5B(i)), the similarity in results suggests that the mutations of A130 to threonine or valine might have similar structural and functional effects. The RB1B2 protein containing the A130S mutation was capable of catalyzing the polyubiquitination of alpha4. The intensity of the polyubiquitinated alpha4 bands was reduced when comparing the ratio of the mono- and di-ubiquitinated bands to the smear pattern for the mutant and wild type RB1B2 protein (Figure 5B(iii)).

Discussion

In this study, we provide the first direct evidence that MID1 could catalyze the ubiquitin-modification of PP2Ac and therefore could regulate the function of PP2A (Figure 2A). Previous description of the regulation of PP2A included phosphorylation of the various subunits, carboxymethylation of PP2Ac, and association with other protein complexes, including inhibitors I1 PP2A and I2 PP2A (see reviews [43], [44]). Indication that PP2A could be regulated by ubiquitination was suggested from studies involving alpha4 and the protestin-inducible EDD (E3 identified by differential display) E3 ubiquitin ligase [45]. However, it is unclear whether EDD is directly involved in the ubiquitination of PP2Ac.

The role alpha4 plays in regulating the ubiquitination of PP2Ac by MID1 appears to be complex. On one hand, alpha4 is shown to be a potent regulator of PP2Ac [4], [46]. Alpha4 was shown to bind and induce a conformational change that inactivated the PP2Ac subunit, and this complex apparently protected the PP2Ac subunit from degradation [4], [21]. On the other hand, alpha4 is postulated to facilitate a MID1-alpha4-PP2Ac complex that promoted the ubiquitination of PP2Ac by MID1 [17], [24], [37]. Our ubiquitination data suggest that MID1 could interact with and target PP2Ac in the absence of alpha4 (Figure 2A). The mechanism associated with the reduction in the level of ubiquitination of PP2Ac in the presence of alpha4 is unclear. A first speculation for the reduction might be that MID1 catalyzes the ubiquitination of alpha4, decreasing the available ubiquitin [1]. However, the amount of ubiquitin present was at saturating concentration and sufficiently high to account for the polyubiquitination of both the PP2Ac and alpha4 proteins. Results of the titration of alpha4 from sub-molar equivalents to molar excess compared to the amount of PP2Ac protein present were inconclusive in confirming a diphasic dependence of alpha4 concentration, as was demonstrated for arrestin3 and JNK3α2 in phosphorylating MAP kinases [41]. Further studies are warranted to understand the mechanism of regulation.

Consistent with our previous studies, the RB1B2 protein construct of MID1 is sufficient for catalyzing the ubiquitination of PP2Ac (Figure 2C). It appears that C-terminal domains (CC, FNIII and SPRY/B30.2) may also be important for efficient targeting of PP2Ac. The Bbox domains are required for at least the diubiquitination of PP2Ac (Figure 2D). Without the Bbox domains, the RING domain only facilitated the mono-ubiquitination of PP2Ac. Even though the Bbox domains possess RING E3-like folds [47], [48] they are unable to catalyze the ubiquitination of PP2Ac, indicating that the Bbox domains may function, in part, as E4 ligases, similar to the roles of the RING domains of BARD1 and MDMX [49], [50], [51], [52].

In addition to understanding how natural mutations within the Bbox domains of MID1 affect its E3 ligase activity, these mutations also allows us to gain insights into the mechanism of Bbox domain function. The Cys142 and Cys145 residues are highly conserved (Figure 6A) in Bbox1 domains throughout TRIM proteins. Mutations of these zinc-binding cysteine residues to serine or threonine, residues typically not associated with coordinating zinc ions, should disrupt the coordination of the zinc ion and cause significant structural changes of the Bbox1 domain [48], [53], [54], [55], [56]. Similarly, the C195F mutation should disrupt the structure of the Bbox2 domain.

Figure 6. XLOS-observed mutations of highly conserved amino acids destabilized the Bbox structure.

Figure 6

Open in a new tab

A. Sequence alignment of the MID1 and MID2 Bbox1 domains from human, rat, mouse, chicken and zebrafish. Conserved A130, C142 and C145 residues are highlighted in red font and dark shading. At the bottom, the sequences of the various Bbox1 mutations are shown. B. Ribbon representation of the tertiary structure of the Bbox1-Bbox2 domains in tandem showing the coordinated zinc ions (red spheres) and their ligands. The structure shows the locations of the amino acids that were mutated, specifically residues Ala130, Cys142, Cys145 and Cys195. On the right, a zoomed view showing the relative proximity of Ala130 to the zinc binding Cys119 and Cys142 residues. Space filling representation of the side chain atoms of these three amino acids suggests that mutation of Ala130 to Thr and Val will induce steric clashes with the side chain atoms of Cys119 and Cys142, which can disrupt zinc-coordination.

Residue Ala130 is also highly conserved in Bbox1 domains (Figure 6A). Even thought Ala130 is not involved in metal coordination the structure of the Bbox1 domain shows the side chain methyl group in close proximity to methylene atoms of Cys119 and Cys142 (Figure 6B). We postulate that the larger side chains of valine or threonine would steric hinder the position of side chain of Cys119 and Cys142 and their coordination of the zinc ion. Thus, the A130T/V mutation could have similar structural effects as the mutations of the Cys142 and Cys145 residues. In contrast, the A130S mutation is equated to replacing of one of the methyl protons of alanine with a hydroxyl moiety. The hydroxyl group is flexible and provides less steric clashes with the cysteine side chains, allowing the Bbox1 domain to maintain native function.

Perturbation of the Bbox1 domain structure could disrupt the interaction of RB1B2 protein with the alpha4 protein. In a non-biased functional screen, we previously identified a L146Q substitution that abolished the MID1-alpha4 interaction and the subsequent ubiquitination of alpha4 [1]. Thus, the loss of polyubiquitination of alpha4 indicates that the A130T/V, C142S and C145T mutations destabilize the interaction necessary for ubiquitination of alpha4. In these instances, the Bbox2 domain was unable to compensate for the loss of function of the Bbox1 domain.

Qualitatively, the mutations within the Bbox1 domain did not affect the ability of the RB1B2 protein to catalyze the di-ubiquitination of PP2Ac. Given that the both RING domain and folded Bbox1 domain are required for catalyzing the diubiquitination of PP2Ac (Figure 2D), the diubiquitination of PP2Ac by the RB1B2 with mutations within the Bbox1 domain suggests that the Bbox2 domain compensated for the loss of function of the mutant Bbox1 domain, at least in the targeting PP2Ac.

Interestingly, the RB1B2 protein with the C195F mutation was observed to catalyze the ubiquitination of PP2Ac and alpha4 (Figures 4, 5B). These results were somewhat unexpected because the C195F mutation is associated with XLOS and should have an effect on the targeting of PP2Ac or alpha4. Thus, it is possible that the Bbox2 domain may have other roles that could not be distinguished from the current studies.

It is intriguing to note that the different mutations within the Bbox1 domain, which should have resulted in same loss of structure and function, do not correlate with the same set of birth defects (Figure 1B). The truncation MID1 mutation that yields just the RING-Bbox1 domains (Figures 1A and B) is associated with overlapping defects as Bbox1 domain mutations. Similarly, the C195F mutation is associated with overlapping birth defects as the A130T/V mutations, with two additional defects (anal malformation and developmental delays) (Figure 1B). These results suggest that there may be other factors that contribute to XLOS, one of which may be the role the highly homologous MID2 [38].

In conclusion, MID1 could regulate the function of PP2A by catalyzing the ubiquitination PP2Ac and alpha4. It is possible that many of XLOS-observed mutations, including mutations of the Bbox1 domain, result in an increase of unmodified or monoubiquitinated alpha4, which could subsequently bind and protect PP2Ac from ubiquitin-mediated degradation. Thus, alpha4 may have a significantly greater role in contributing to X-linked Opitz syndrome that includes protecting PP2Ac.

Acknowledgments

We thank Dr. Timothy Cox (University of Washington, Seattle) for helpful discussions.

Data Availability

The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper.

Funding Statement

Funding provided by Award 1052520, nsf.gov. The National Science Foundation had no role in the design, data collection, analysis, decision to publish or preparation of this manuscript.

References

  • 1. Du H, Huang Y, Zaghlula M, Walters E, Cox TC, et al. (2013) The MID1 E3 Ligase Catalyzes the Polyubiquitination of Alpha4 (alpha4), a Regulatory Subunit of Protein Phosphatase 2A (PP2A): Novel Insights into MID1-mediated regulation of PP2A. The Journal of biological chemistry 288: 21341–21350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Han X, Du H, Massiah MA (2011) Detection and Characterization of the In Vitro E3 Ligase Activity of the Human MID1 Protein. J Mol Biol 407: 505–520. [DOI] [PubMed] [Google Scholar]
  • 3. Kong M, Bui TV, Ditsworth D, Gruber JJ, Goncharov D, et al. (2007) The PP2A-associated protein alpha4 plays a critical role in the regulation of cell spreading and migration. The Journal of biological chemistry 282: 29712–29720. [DOI] [PubMed] [Google Scholar]
  • 4. Kong M, Ditsworth D, Lindsten T, Thompson CB (2009) Alpha4 is an essential regulator of PP2A phosphatase activity. Molecular cell 36: 51–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. McConnell JL, Watkins GR, Soss SE, Franz HS, McCorvey LR, et al. (2010) Alpha4 is a ubiquitin-binding protein that regulates protein serine/threonine phosphatase 2A ubiquitination. Biochemistry 49: 1713–1718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Schweiger S, Schneider R (2003) The MID1/PP2A complex: a key to the pathogenesis of Opitz BBB/G syndrome. Bioessays 25: 356–366. [DOI] [PubMed] [Google Scholar]
  • 7. Garcia A, Cayla X, Guergnon J, Dessauge F, Hospital V et al. (2003) Serine/threonine protein phosphatases PP1 and PP2A are key players in apoptosis. Biochimie 85: 721–726. [DOI] [PubMed] [Google Scholar]
  • 8. Janssens V, Goris J (2001) Protein phosphatase 2A: a highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling. Biochem J 353: 417–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Sontag E (2001) Protein phosphatase 2A: the Trogan Horse of cellular signaling. Cellular Signalling 13: 7–11. [DOI] [PubMed] [Google Scholar]
  • 10. Lechward K, Awotunde OS, Swiatek W, Muszynska G (2001) Protein phosphatase 2A: variety of forms and diversity of functions. Acta Biochim Pol 48: 921–933. [PubMed] [Google Scholar]
  • 11. Hunter T (1995) Protein kinases and phosphatases: the yin and yang of protein phosphorylation and signaling. Cell 80: 225–236. [DOI] [PubMed] [Google Scholar]
  • 12. Millward TA, Zolnierowicz S, Hemmings BA (1999) Regulation of protein kinase cascades by protein phosphatase 2A. Trends Biochem Sci 24: 186–191. [DOI] [PubMed] [Google Scholar]
  • 13. Shenolikar S (1994) Protein serine/threonine phosphatases–new avenues for cell regulation. Annu Rev Cell Biol 10: 55–86. [DOI] [PubMed] [Google Scholar]
  • 14. Wera S, Hemmings BA (1995) Serine/threonine protein phosphatases. Biochem J 311 (Pt 1): 17–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. LeNoue-Newton M, Watkins GR, Zou P, Germane KL, McCorvey LR, et al. (2011) The E3 ubiquitin ligase- and protein phosphatase 2A (PP2A)-binding domains of the Alpha4 protein are both required for Alpha4 to inhibit PP2A degradation. The Journal of biological chemistry 286: 17665–17671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Liu E, Knutzen CA, Krauss S, Schweiger S, Chiang GG (2011) Control of mTORC1 signaling by the Opitz syndrome protein MID1. Proceedings of the National Academy of Sciences of the United States of America 108: 8680–8685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Trockenbacher A, Suckow V, Foerster J, Winter J, Krauss S, et al. (2001) MID1, mutated in Opitz syndrome, encodes an ubiquitin ligase that targets phosphatase 2A for degradation. Nat Genet 29: 287–294. [DOI] [PubMed] [Google Scholar]
  • 18. Turowski P, Favre B, Campbell KS, Lamb NJ, Hemmings BA (1997) Modulation of the enzymatic properties of protein phosphatase 2A catalytic subunit by the recombinant 65-kDa regulatory subunit PR65alpha. Eur J Biochem 248: 200–208. [DOI] [PubMed] [Google Scholar]
  • 19. Kremmer E, Ohst K, Kiefer J, Brewis N, Walter G (1997) Separation of PP2A core enzyme and holoenzyme with monoclonal antibodies against the regulatory A subunit: abundant expression of both forms in cells. Mol Cell Biol 17: 1692–1701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Mayer-Jaekel RE, Hemmings BA (1994) Protein phosphatase 2A–a ‘menage a trois’. Trends Cell Biol 4: 287–291. [DOI] [PubMed] [Google Scholar]
  • 21. Watkins GR, Wang N, Mazalouskas MD, Gomez RJ, Guthrie CR, et al. (2012) Monoubiquitination promotes calpain cleavage of the protein phosphatase 2A (PP2A) regulatory subunit alpha4, altering PP2A stability and microtubule-associated protein phosphorylation. The Journal of biological chemistry 287: 24207–24215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Jiang L, Stanevich V, Satyshur KA, Kong M, Watkins GR, et al. (2013) Structural basis of protein phosphatase 2A stable latency. Nature communications 4: 1699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Du H, Massiah MA (2011) NMR studies of the C-terminus of alpha4 reveal possible mechanism of its interaction with MID1 and protein phosphatase 2A. PloS one 6: e28877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Aranda-Orgilles B, Aigner J, Kunath M, Lurz R, Schneider R, et al. (2008) Active transport of the ubiquitin ligase MID1 along the microtubules is regulated by protein phosphatase 2A. PloS one 3: e3507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Aranda-Orgilles B, Rutschow D, Zeller R, Karagiannidis AI, Kohler A, et al. (2011) Protein phosphatase 2A (PP2A)-specific ubiquitin ligase MID1 is a sequence-dependent regulator of translation efficiency controlling 3-phosphoinositide-dependent protein kinase-1 (PDPK-1). The Journal of biological chemistry 286: 39945–39957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Buchner G, Montini E, Andolfi G, Quaderi N, Cainarca S, et al. (1999) MID2, a homologue of the Opitz syndrome gene MID1: similarities in subcellular localization and differences in expression during development. Hum Mol Genet 8: 1397–1407. [DOI] [PubMed] [Google Scholar]
  • 27. Cox TC, Allen LR, Cox LL, Hopwood B, Goodwin B, et al. (2000) New mutations in MID1 provide support for loss of function as the cause of X-linked Opitz syndrome. Hum Mol Genet 9: 2553–2562. [DOI] [PubMed] [Google Scholar]
  • 28. Fontanella B, Russolillo G, Meroni G (2008) MID1 mutations in patients with X-linked Opitz G/BBB syndrome. Human mutation 29: 584–594. [DOI] [PubMed] [Google Scholar]
  • 29. So J, Suckow V, Kijas Z, Kalscheuer V, Moser B, et al. (2005) Mild phenotypes in a series of patients with Opitz GBBB syndrome with MID1 mutations. Am J Med Genet A 132: 1–7. [DOI] [PubMed] [Google Scholar]
  • 30. De Falco F, Cainarca S, Andolfi G, Ferrentino R, Berti C, et al. (2003) X-linked Opitz syndrome: novel mutations in the MID1 gene and redefinition of the clinical spectrum. Am J Med Genet 120A: 222–228. [DOI] [PubMed] [Google Scholar]
  • 31. Ferrentino R, Bassi MT, Chitayat D, Tabolacci E, Meroni G (2007) MID1 mutation screening in a large cohort of Opitz G/BBB syndrome patients: twenty-nine novel mutations identified. Human mutation 28: 206–207. [DOI] [PubMed] [Google Scholar]
  • 32. Hu CH, Liu YF, Yu JS, Ng YY, Chen SJ, et al. (2012) A MID1 gene mutation in a patient with Opitz G/BBB syndrome that altered the 3D structure of SPRY domain. American journal of medical genetics Part A 158A: 726–731. [DOI] [PubMed] [Google Scholar]
  • 33. Ruiter M, Kamsteeg EJ, Meroni G, de Vries BB (2010) A MID1 mutation associated with reduced penetrance of X-linked Opitz G/BBB syndrome. Clinical dysmorphology 19: 195–197. [DOI] [PubMed] [Google Scholar]
  • 34. Ji X, Xing Y, Xu Y, Liu Y, Chen Y, et al. (2014) A novel mutation in MID1 in a patient with X-linked Opitz G/BBB syndrome. Gene 537: 140–142. [DOI] [PubMed] [Google Scholar]
  • 35. Short KM, Cox TC (2006) Subclassification of the RBCC/TRIM superfamily reveals a novel motif necessary for microtubule binding. The Journal of biological chemistry 281: 8970–8980. [DOI] [PubMed] [Google Scholar]
  • 36. Schweiger S, Foerster J, Lehmann T, Suckow V, Muller YA, et al. (1999) The Opitz syndrome gene product, MID1, associates with microtubules. Proc Natl Acad Sci U S A 96: 2794–2799. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Liu J, Prickett TD, Elliott E, Meroni G, Brautigan DL (2001) Phosphorylation and microtubule association of the Opitz syndrome protein mid-1 is regulated by protein phosphatase 2A via binding to the regulatory subunit alpha 4. Proc Natl Acad Sci U S A 98: 6650–6655. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Perry J, Short KM, Romer JT, Swift S, Cox TC, et al. (1999) FXY2/MID2, a gene related to the X-linked Opitz syndrome gene FXY/MID1, maps to Xq22 and encodes a FNIII domain-containing protein that associates with microtubules. Genomics 62: 385–394. [DOI] [PubMed] [Google Scholar]
  • 39. Short KM, Hopwood B, Yi Z, Cox TC (2002) Mid1 and Mid2 homo- and heterodimerise to tether the rapamycin-sensitive PP2A regulatory subunit, Alpha 4, to microtubules: implications for the clinical variability of X-linked Opitz GBBB syndrome and other developmental disorders. BMC Cell Biology 3: 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Everett AD, Brautigan DL (2002) Developmental expression of alpha4 protein phosphatase regulatory subunit in tissues affected by Opitz syndrome. Dev Dyn 224: 461–464. [DOI] [PubMed] [Google Scholar]
  • 41. Zhan X, Kaoud TS, Kook S, Dalby KN, Gurevich VV (2013) JNK3 enzyme binding to arrestin-3 differentially affects the recruitment of upstream mitogen-activated protein (MAP) kinase kinases. The Journal of biological chemistry 288: 28535–28547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Aranda-Orgilles B, Trockenbacher A, Winter J, Aigner J, Kohler A, et al. (2008) The Opitz syndrome gene product MID1 assembles a microtubule-associated ribonucleoprotein complex. Human genetics 123: 163–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Lechward K, Awotunde OS, Swiatek W, Muszynska G (2001) Protein phosphatase 2A: variety of forms and diversity of functions. Acta biochimica Polonica 48: 921–933. [PubMed] [Google Scholar]
  • 44. Sontag E (2001) Protein phosphatase 2A: the Trojan Horse of cellular signaling. Cellular signalling 13: 7–16. [DOI] [PubMed] [Google Scholar]
  • 45. McDonald WJ, Thomas LN, Koirala S, Too CK (2014) Progestin-inducible EDD E3 ubiquitin ligase binds to alpha4 phosphoprotein to regulate ubiquitination and degradation of protein phosphatase PP2Ac. Molecular and cellular endocrinology 382: 254–261. [DOI] [PubMed] [Google Scholar]
  • 46. Kong M, Fox CJ, Mu J, Solt L, Xu A, et al. (2004) The PP2A-associated protein alpha4 is an essential inhibitor of apoptosis. Science 306: 695–698. [DOI] [PubMed] [Google Scholar]
  • 47. Massiah MA, Matts JA, Short KM, Simmons BN, Singireddy S, et al. (2007) Solution structure of the MID1 B-box2 CHC(D/C)C(2)H(2) zinc-binding domain: insights into an evolutionarily conserved RING fold. J Mol Biol 369: 1–10. [DOI] [PubMed] [Google Scholar]
  • 48. Massiah MA, Simmons BN, Short KM, Cox TC (2006) Solution structure of the RBCC/TRIM B-box1 domain of human MID1: B-box with a RING. J Mol Biol 358: 532–545. [DOI] [PubMed] [Google Scholar]
  • 49. Mallery DL, Vandenberg CJ, Hiom K (2002) Activation of the E3 ligase function of the BRCA1/BARD1 complex by polyubiquitin chains. Embo J 21: 6755–6762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Xia Y, Pao GM, Chen HW, Verma IM, Hunter T (2003) Enhancement of BRCA1 E3 ubiquitin ligase activity through direct interaction with the BARD1 protein. J Biol Chem 278: 5255–5263. [DOI] [PubMed] [Google Scholar]
  • 51. Wang X, Jiang X (2012) Mdm2 and MdmX partner to regulate p53. FEBS letters 586: 1390–1396. [DOI] [PubMed] [Google Scholar]
  • 52. Linke K, Mace PD, Smith CA, Vaux DL, Silke J, et al. (2008) Structure of the MDM2/MDMX RING domain heterodimer reveals dimerization is required for their ubiquitylation in trans. Cell death and differentiation 15: 841–848. [DOI] [PubMed] [Google Scholar]
  • 53.Massiah MA, Blake PR, Summers MF (1998) Nucleic Acid Interactive Protein Domains That Require Zinc: Oxford University Press.
  • 54. Berg JM (1988) Proposed structure for the zinc-binding domains from transcription factor IIIA and related proteins. Proc Natl Acad Sci U S A 85: 99–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Berg JM (1990) Zinc fingers and other metal-binding domains. Elements for interactions between macromolecules. J Biol Chem 265: 6513–6516. [PubMed] [Google Scholar]
  • 56. Klug A (2010) The discovery of zinc fingers and their development for practical applications in gene regulation and genome manipulation. Quarterly Reviews of Biophysics 43: 1–21. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper.

Articles from PLoS ONE are provided here courtesy of PLOS

RESOURCES