To systematically examine how PTBP1 regulates protein expression through 3’UTR motifs, we employed a massively parallel reporter assay (MPRA) with a library of 467 synthetic 3’UTRs fused to GFP, each containing six occurrences of a 6-7 nucleotide oligomer (Nicolet, 2025). After retroviral transduction of the MPRA library into activated CD8+ T cells, we deleted PTBP1 or used non-targeting RNPs as control. After 6 days of rest, we reactivated PTBP1 KO and control T cells for 16h with a-CD3/a-CD28, or left them nonactivated. The 15% top (GFPhi) and 15% bottom (GFPlo) GFP-expressing cells were FACS-sorted, and motif enrichment for high or low protein expression was determined by sequencing.

MPRA screen

The retroviral library was produced by transfecting 4.5 μg of the library plasmid DNA into FLYRD18 retroviral packaging cells, as described above. CD8+ T cells were activated with α-CD3/α-CD28 for 48h and transduced as described above. 24h after transduction, cells were washed and nucleofected with Cas9 RNPs, as described above. Transduction efficiencies were kept between 8-12% to minimize multiple transgene integrations. After expansion for 7 days, T cells were harvested and stained with Near-IR live/dead marker (Life Technologies) for dead cell exclusion, in addition to α-CD8 and α-CD4. The top and bottom 15% GFP-expressing CD8+ T cells, in addition to the full GFP-positive fraction, were sorted on an Aria II (BD Biosciences). Fluorescence-activated cell sorted cells were subsequently pelleted and resuspended in 1 μl of DirectPCR Lysis Reagent (Viagen) per 5,000 cells, supplemented with proteinase K (0.5 mg/ml; Viagen). The obtained lysate was incubated at 55°C for 3 hours, followed by 85°C for 45 min and 95°C for 5 min.

3′UTR reporter assay library amplification and sequencing

An initial preamplification of the synthetic 3′UTRs was performed on 10 μl of lysate (±50,000 cells) in a 50-μl reaction volume using the NEBnext master mix (NEB) and 10 μM forward (5′-AAAGACCCCAACGAGAAGC-3′) and reverse (5′-AGTCTATAGCTACTAGGCG-3′) primers. This PCR amplifies three (out of five) of the motif sequences. An initial touchdown stage was performed by reducing the annealing temperature 1°C per cycle from 58° to 50°C (eight cycles), followed by six cycles at 50°C. Extension time was kept at 20 s. PCR products were purified using Cytiva Sera-Mag Select at a 1.5:1 Sera-Mag:PCR-product ratio, following the manufacturer’s protocol. Each of these samples were subsequently used as template for a second PCR, with (forward: 5′-ACACTCTTTCCCTACACGACGCTCTTCCGATCTAGACCCCAACGAGAAGCGCGATCAC-3′, reverse: 5′-GACTGGAGTTCAGACGTGTGCTCTTCCGATCTAGTCTATAGCTACTAGGCGATA-3′) primers to amplify the product and attach partial Illumina adapters. This second PCR was performed using the NEBnext (NEB) and consisted of 10-15 cycles with a constant annealing temperature of 65°C and extension time of 30 s. Correct products (±275 base pairs) were purified using Cytiva Sera-Mag Select at a 1.5:1 Sera-Mag:PCR-product ratio, following the manufacturer’s protocol. Purified products were indexed with custom Unique Dual Index, pooled, purified, and sequenced at the genomics core facility of the Netherlands Cancer institute (paired-end sequencing, 300 cycles). Raw data are available at GEO (see below).

3′UTR parallel reporter analysis

The obtained read 2 data contained the three motif sequences. First, the sequencing reads were trimmed using cutadapt to remove constant sequences and PCR handles. Second, trimmed reads were aligned to the synthetic UTR library using bowtie2, which was performed using the ‘local’ alignment and ‘very-sensitive’ options. Third, aligned reads were quantified using htseq-count, using a custom gtf file constructed from the synthetic UTR library. Motifs with less than 20 read counts across all samples were filtered out, and library size of each sample was scaled to 10,000 to allow comparison between samples.