A Scalable, Cell-Based Method for the Functional Assessment of Ube3a Variants

Summary

A simple and scalable method was developed to assess the functional significance of missense variants in Ube3a, a gene whose loss and gain of function are linked to both Angelman syndrome and autism spectrum disorder.

Abstract

The increased use of sequencing in medicine has identified millions of coding variants in the human genome. Many of these variants occur in genes associated with neurodevelopmental disorders, but the functional significance of the vast majority of variants remains unknown. The present protocol describes the study of variants for Ube3a, a gene that encodes an E3 ubiquitin ligase linked to both autism and Angelman syndrome. Duplication or triplication of Ube3a is strongly linked to autism, whereas its deletion causes Angelman syndrome. Thus, understanding the valence of changes in UBE3A protein activity is important for clinical outcomes. Here, a rapid, cell-based method that pairs Ube3a variants with a Wnt pathway reporter is described. This simple assay is scalable and can be used to determine the valence and magnitude of activity changes in any Ube3a variant. Moreover, the facility of this method allows the generation of a wealth of structure-function information, which can be used to gain deep insights into the enzymatic mechanisms of UBE3A.

Introduction

Recent technological advances have made the sequencing of exomes and genomes routine in clinical settings^1,2. This has led to the discovery of a large number of genetic variants, including millions of missense variants that typically change one amino acid in a protein. Understanding the functional significance of these variants remains a challenge, and only a small fraction (~2%) of the known missense variants have a clinical interpretation^1,3.

A prominent example of this problem is Ube3a, a gene that encodes an E3 ubiquitin ligase that targets substrate proteins for degradation⁴. Ube3a resides within chromosome 15q11-13 and is expressed exclusively from the maternally inherited allele^5,6,7. Observations from disease genetics strongly suggest that insufficient or excessive activity of the UBE3A enzyme causes distinct neurodevelopmental disorders. Deletion of maternal chromosome 15q11-13 causes Angelman syndrome (AS)⁸, a disorder characterized by severe intellectual disability, motor impairments, seizures, a happy demeanor with frequent smiling, and dysmorphic facial features^8,9,10. In contrast, duplication or triplication of maternal chromosome 15q11-13 causes Dup15q syndrome, a heterogeneous condition recognized as one of the most prevalent syndromic forms of autism^11,12,13. In addition, there are hundreds of missense variants identified in Ube3a, the majority of which are considered variants of uncertain significance (VUS) as their functional and clinical significance are unknown. Thus, there is considerable interest in developing methods that can empirically assess Ube3a variants to determine whether they contribute to neurodevelopmental disease.

The UBE3A enzyme belongs to the HECT (homologous to E6-AP C-terminus) domain family of E3 ubiquitin ligases that all possess the eponymous HECT domain, which contains the biochemical machinery necessary to accept activated ubiquitin from E2 enzymes and transfer it to substrate proteins¹⁴. Historically, the characterization of E3 enzymes has relied on reconstituted in vitro systems that require an ensemble of purified proteins^4,15,16. Such methods are slow and labor-intensive and not amenable to assessing the activity of a large number of variants. In previous work, UBE3A was identified to activate the Wnt pathway in HEK293T cells by modulating the function of the proteasome to slow the degradation of β-catenin¹⁷. This insight allows the use of Wnt pathway reporters as an efficient and rapid method to identify both loss- and gain-of-function variants of Ube3a¹⁸. The protocol below describes in detail a method to generate Ube3a variants as well as a luciferase-based reporter to assess changes in the activity of Ube3a variants.

Protocol

1. Mutagenesis cloning to generate  Ube3a variants

1. Clone all Ube3a variants into the pCIG2 plasmid (Figure 1A), a bicistronic vector that contains a chicken-β-actin promoter and an internal ribosome entry site (IRES) for EGFP expression¹⁹. Full-length Ube3a constructs contain an N-terminal Myc-tag sequence and are cloned into pCIG2 using a 5' SacI site and a 3' XmaI site. In addition, naturally occurring EcoRI, EcoRV, and PstI sites within the Ube3a coding sequence are also used to subclone fragments.
   NOTE: De-identified variants for Ube3a can be obtained through the literature and in publicly accessible repositories such as the ClinVar database¹. Additional unreported variants can be obtained through personal communication with medical centers.
2. Use an adapted two-step megaprimer mutagenesis method²⁰ to introduce mutations into the Ube3a reading frame. In the first step, design the mutagenic oligonucleotide that contains the desired mutation (the example shown in this protocol is to generate a UBE3A Q588E variant) flanked by at least 30 base pairs (bp) of homology 5′ and 3′ to the mutation (Figure 1B and Table 1).
3. Use the mutagenic oligonucleotide along with a complementary oligonucleotide for the first round PCR to generate a 200-400 bp megaprimer containing the mutation (Figure 1C; Table 2 and Table 3). Use the entire 50 μL of the PCR reaction volume for agarose gel electrophoresis using a 1% gel and subsequent gel purification (Table of Materials). Elute the PCR product in 30 μL of deionized H₂O (diH₂O) for use in the second round PCR step below.
4. Set up the second round PCR as indicated (Figure 1C; Table 2 and Table 3). This step will generate larger DNA fragments (typically ~1-1.5 kb in length) containing the mutation and terminal restriction sites suitable for sub-cloning (Figure 1C).
5. After completion of the PCR reaction, purify the resulting product using a PCR purification kit (Table of Materials). Elute in 30 μL and digest all of the purified product and the pCIG2 Ube3a WT construct with the appropriate restriction enzymes (the example shows digestion with EcoRI and XmaI).
6. After 2 h of digestion at 37 °C, resolve the digestion products through agarose gel electrophoresis (1% gel), and gel-purify the appropriate products. Set up ligations according to the manufacturer's protocol (Table of Materials), transform the ligation products into a DH10B bacterial strain (Table of Materials), and then plate with ampicillin (100 μg/mL) selection.
7. The next day, pick two to four colonies to inoculate 3 mL cultures containing Luria Broth (LB) and 100 μg/mL ampicillin. Let the cultures grow overnight at 37 °C in an orbital incubator with shaking (225 rpm).
8. The following day, use 1-1.5 mL of the bacterial culture to miniprep the plasmid DNA (Table of Materials). Save the remaining culture by placing them at 4 °C. Screen the miniprepped plasmids by Sanger sequencing using an oligonucleotide that binds ~200 bp from the desired mutation.
9. Use the bacterial cultures saved at 4 °C to re-inoculate 50 mL cultures to midiprep the correct plasmid (Table of Materials). Fully sequence the Ube3a coding sequence to validate the correct sequence of the DNA.
10. Create working stocks of Ube3a variant plasmids, as well as the β-catenin activated reporter (BAR)²¹, TK-Renilla luciferase, ligase-dead Ube3a (UBE3A C820A mutation), and pCIG2 empty vector at a concentration of 100 ng/μL using sterile H₂O for the subsequent transfection steps.

2. Preparation and transfection of human embryonic kidney 293T (HEK293T) cells

1. Perform the assays in HEK293T cells grown in Dulbecco's modified Eagle medium (DMEM) pre-supplemented with glutamine, 10% fetal bovine serum (FBS), and antibiotic-antimycotic agent as described previously¹⁷. Grow the cells in a sterile, humidified incubator at 37 °C with 5% CO₂.
2. Using a biosafety cabinet, first plate the suspended HEK293T cells onto tissue culture-treated 96-well flat-bottomed plates at a density of 2.2 × 10⁴ cells per well in 100 μL of culture media and allow them to grow overnight.
3. On the second day, transfect the cells in a biosafety cabinet with Firefly and Renilla luciferase reporter plasmids (Table 4) and Ube3a plasmids in triplicate. Perform the transfections in 10 μL of total volume per well using a 10:1:8 ratio of plasmids (50 ng of BAR, 5 ng of TK-Renilla, and 40 ng of Ube3a plasmid per well, respectively), as previously described¹⁸, and 0.4 μL of transfection reagent (Table 4).
   NOTE: For each experiment, an empty vector (GFP only) and a plasmid encoding ligase-dead Ube3a are also transfected to serve as negative controls.
4. Create a master mix for the transfections for each variant in a 1.5 mL microcentrifuge tube (Table 4) and incubate at room temperature (RT) for 15 min. After the incubation, gently agitate the transfection mixture by tapping the tube, and then add 10 μL of the transfection mixture directly to the existing growth media in the wells.
5. Afterward, return the cells to the incubator and allow the plasmids to express for 48 h. Replacement of the growth media is not necessary.
6. Monitor the transfection efficiency by GFP fluorescence using a fluorescence microscope. HEK293T cells are efficiently transfected by this method, and >80% of cells should express GFP after 48 h.

3. Measurement of luciferase activity

NOTE: Luciferase activity is assessed using a commercially available system that assays both Firefly and Renilla luciferase (Table of Materials) according to the manufacturer's protocol.

1. Carefully aspirate the culture media from transfected HEK293T cells, and then wash the cells with cold phosphate-buffered saline (PBS). Aspirate the PBS and lyse the cells by adding 25 μL of non-denaturing lysis buffer and incubate for 15 min with gentle rocking on ice.
2. Afterward, add 20 μL of the resulting lysate (no centrifugation is necessary) into a new 96-well assay plate containing 100 μL of a Firefly luciferase substrate reagent. Mix the plate gently by tapping.
3. Load the plate onto a plate reader and measure the luminescence using a top detector with a read height of 1 mm and an integration time of 1 s.
   NOTE: The gain of the detector may need to be adjusted based on the strength of the signal.
4. Immediately after reading the Firefly luciferase luminescence, add 100 µL of Renilla luciferase substrate to the wells. This step simultaneously quenches the Firefly luciferase activity while assessing Renilla luciferase activity. Mix the samples by gentle agitation of the plate and measure the luminescence again to assess the Renilla luciferase activity.
   NOTE: While other plates can be used for this assay, using plates with black frames and white wells will provide the most sensitive results as this amplifies the luciferase signal while minimizing noise.
5. Quantify the reporter responses as the Firefly luciferase to Renilla luciferase ratio (Firefly/Renilla), and average the values of the triplicate measurements. Afterward, normalize the Firefly/Renilla value for each variant against the values for wild-type (WT) Ube3a expressing cells to compare the relative activity of variant proteins.

Results

Large-scale functional screening of Ube3a missense variants identifies a broad landscape of loss- and gain-of-function mutations
Previous work with Ube3a mutants suggested that the Wnt response can serve as a reporter of cellular UBE3A protein activity. These observations were expanded, and additional validation experiments were performed to investigate whether the BAR assay is suitable to report a range of UBE3A activities in the cell. First, HEK293T cells were transfected with var...

Discussion

The protocol described here provides an efficient and scalable method to assess the enzymatic activity of Ube3a variants. There are several technical details that warrant careful consideration when using this assay. One consideration is the choice of Wnt reporter plasmids used in this assay. The protocol described here specifically uses the β-catenin activated reporter (BAR)²¹, a reporter that contains a concatemer of 12 T-cell factor (TCF) response elements separated by specifically...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.05% Trypsin-EDTA (1x), phenol red	Gibco	25300-054
1 Kb DNA ladder	Lambda Biotech	M108-S
100 bp DNA Ladder	Lambda Biotech	M107
10x Buffer for T4 DNA Ligase with 10 mM ATP	New England BioLabs	B0202A
5x Phusion HF Reaction Buffer	New England BioLabs	B0518S
Antibiotic-Antimycotic Solution	Corning	30004CI
Black/White Isoplate-96 Black Frame White Well plate	PerkinElmer	6005030
Carbenicillin Disodium Salt	Midwest Scientific	KCC46000-5
Countess cell counting chamber  slides	Invitrogen by Thermo Fisher Scientific	C10283
Countess II Automated Cell Counter	life technologies		Cell counting machine
Custom DNA oligos	Integrated DNA Technologies (IDT)
Deoxynucleotide (dNTP) Solution Mix	New England BioLabs	N0447S
DMEM, high glucose, GlutaMAX Supplement, pyruvate	Gibco	10569044	Basal medium for supporting the growth of HEK293T cell line
DPBS (1x)	Gibco	14190-136
Dual-Luciferase Reporter Assay System	Promega	E1910
EcoRI-HF	New England BioLabs	R3101S	Restriction enzyme
Fetal Bovine Serum, qualified, heat inactivated	Gibco	16140071	Fetal bovine serum
Fisherbrand Surface Treated Tissue Culture Dishes	Fisherbrand	FB012924
FuGENE 6 Transfection Reagent	Promega	E2691
Gel Loading Dye Purple (6x)	New England BioLabs	B7024A
HEK293T cells	ATCC	CRL-3216
High Efficiency ig 10B Chemically Competent Cells	Intact Genomics	1011-12	E. coli DH10B cells
HiSpeed Plasmid Midi Kit	Qiagen	12643	Midi prep
pCIG2 plasmid
pGL3 BAR plasmid
Phusion HF DNA Polymerase	New England BioLabs	M0530L	DNA polymerase
ProFlex 3 x 32 well PCR System	Applied biosystems by life technologies		Thermocycler
pTK Renilla plasmid
QIAprep Spin Miniprep Kit (250)	Qiagen	27106	Mini prep
QIAquick Gel Extraction Kit (250)	Qiagen	28706	Gel purification
QIAquick PCR Purification Kit (250)	Qiagen	28106	PCR purification
rCutSmart Buffer	New England BioLabs	B6004S
SacI-HF	New England BioLabs	R3156S	Restriction enzyme
Synergy HTX Multi-Mode Reader	BioTek		Plate reader runs Gen5 software v3.08 (BioTek)
T4 DNA Ligase	New England BioLabs	M0202L	Ligase
TAE Buffer, Tris-Acetate-EDTA, 50x Solution, Electrophoresis	Fisher Scientific	BP13324
Tissue Culture Plate 96 wells, Flat Bottom	Fisherbrand	FB012931
UltraPure Ethidium Bromide Solution	Invitrogen by Thermo Fisher Scientific	15585011
XmaI	New England BioLabs	R0180S	Restriction enzyme