Nascent alt-protein chemoproteomics reveals a pre-60S assembly checkpoint inhibitor

Chemoproteomic profiling of nascent alt-proteins

Motivated by the hypothesis that alt-proteins undergoing regulated synthesis could play important cellular roles, we developed a chemoproteomic approach to identify them (Fig. 1a). We leveraged bio-orthogonal non-canonical amino acid tagging (BONCAT)¹⁷, in which the methionine analogue azidohomoalanine (AHA) is metabolically incorporated into all newly synthesized proteins by the endogenous protein translation machinery. Previous BONCAT workflows derivatized AHA-labeled proteins with biotin-alkyne, requiring column-based removal of excess biotin-alkyne prior to streptavidin enrichment – a step that may eliminate small proteins. We therefore designed a protocol to enrich small proteins from whole proteomes after AHA-labeling with a C8 column¹⁹, followed by copper-free Click chemistry capture directly on dibenzocyclooctyne magnetic beads. On-bead digest was followed by liquid chromatography/tandem mass spectrometry-based alt-protein identification²⁰. Our C8 column/on-bead BONCAT workflow detected more than twice as many canonical small proteins as biotin-streptavidin/peptide gel-based BONCAT capture (Supplementary Table 1). Consistent with a previous report¹⁹, C8 columns afford less stringent small protein enrichment than peptide gels (Extended Data Fig. 1a). The expression level and size distribution of gene products detected in AHA-labeled samples is comparable to unlabeled samples, suggesting that our method is unbiased with respect to protein length and abundance (Extended Data Fig. 1a-b).

We then profiled nascent small proteins in HEK 293T cells under control conditions (DMSO treatment), oxidative stress (sodium arsenite treatment), DNA damage stress (etoposide treatment) and unfolded protein response stress (DTT treatment) (Supplementary Table 2 and Extended Data Fig. 1c). Searching our data against the annotated human proteome database followed by GO analysis revealed gene expression signatures consistent with the stress conditions applied (Extended Data Fig. 1d-e). We then searched the data against a 3-frame RNA-seq translation database with stringent peptide-spectral match parameters as previously reported⁷, leading to identification of a total of 22 unannotated alt-proteins or N-terminal extensions of canonical proteins (Extended Data Fig. 1f, Supplementary Data S1). Nine were only detected under stress, including alt-CNPY2, which was observed after etoposide treatment (Extended Data Fig. 1g, Extended Data Fig. 2a), and 16 were detected with mass spectrometry in this study for the first time^7,19,21-23 (Supplementary Data S1).

We validated translation of six alt-proteins detected under various conditions in our screen (Extended Data Fig. 2b-h, Supplementary Table 3). The cDNA sequence comprising the 5'UTR of the encoding transcript through the stop codon of the putative alt-ORF was over-expressed from a mammalian expression vector with a FLAG-HA epitope tag on the C-terminus of the alt-protein, followed with immunostaining. As shown in Figure 1b and Extended Data Figure 1h, over-expressed alt-DRAP1 (previously validated and included as a positive control⁷), alt-PRR3, alt-PRH1, alt-CNPY2 and alt-CACTIN exhibited nucleocytoplasmic immunofluorescence. Over-expressed MINAS-60 co-localized with a nucleolar protein, fibrillarin (Fig. 1c). Immunoblotting confirmed the sizes of the six translated alt-proteins (Fig. 1d). MINAS-60 and alt-CNPY2 produced two immunoreactive bands, which is due to multiple in-frame start codon for MINAS-60 (vide infra), and may be caused by multiple start sites or post-translational modifications for alt-CNPY2.

To determine whether expression of alt-CNPY2 is induced by DNA damage, we treated HEK 293T over-expressing alt-CNPY2 with etoposide, followed by immunoblotting. As shown in Figure 1e, alt-CNPY2 exhibited dose- and time-dependent upregulation upon etoposide treatment of up to two fold. And the mRNA level of CNPY2 did not change (Fig. 1f), indicating that upregulation of alt-CNPY2 is likely post-transcriptional. Taken together, these results suggest that our chemoproteomic pipeline can detect alt-proteins undergoing active synthesis during normal cell growth and cellular stress.

MINAS-60 is cell-cycle regulated and conserved in mouse

We selected MINAS-60, which was detected in normally growing cells, for further study, because it localizes to the nucleolus, the site of ribosome biogenesis. MINAS-60 is nested within the CDS of human RBM10, so probing its biological role could shed light on the under-explored class of overlapping alt-proteins (Fig. 2a, Supplementary Data S2-4). A different MINAS-60 tryptic peptide was previously detected in colorectal cancer, supporting expression of MINAS-60 in human tissue²². To validate expression of MINAS-60 from the RBM10 genomic locus, we generated two independent Cas9-directed knock-in (KI) HEK 293T cell lines with a 3×GFP11-FLAG-HA tag appended to the 3' end of MINAS-60 alt-ORF²⁴, followed with immunostaining. As shown in Figure 2b, endogenously expressed MINAS-60 co-localizes with fibrillarin, suggesting MINAS-60 functions in the nucleolus.

The nucleolus is the site of ribosome biogenesis in eukaryotic cells. Ribosome biogenesis starts with the transcription of rDNA, which oscillates during the cell cycle, nearly ceasing during M phase, increasing during G1, and maximizing during S and G2 in human cells^25-27. We hypothesized that cell cycle expression of MINAS-60 might be correlated with ribosome biogenesis activity if it functions in this process. Immunostaining of synchronized MINAS-60 KI cells revealed that MINAS-60 increased at early S phase, peaking at the 2 h time point. MINAS-60 decreased by late S phase, and was very low during G2/M. At G1, MINAS-60 intensity again increased (Extended Data Fig. 3a). These results were confirmed with immunoblotting (Fig. 2c). MINAS-60 expression is therefore coordinated with ribosome biogenesis activity during the cell cycle.

We then identified the start codon(s) that initiate MINAS-60 translation. We tested two upstream in-frame non-AUG start codons^6,7,28, A₃₈₃TC and A₃₈₆GG (numbered relative to the first nucleotide of the cDNA), as well as seven internal AUG start codons^5,29 (numbered ATG1 – ATG7) relative to the RBM10 CDS (Extended Data Fig. 3b). The ~20 kDa MINAS-60 isoform initiates at ATG1 and the ~15 kDa MINAS-60 isoform at ATG6 or 7, because the indicated species is abrogated only when the corresponding start codon is deleted or mutated (Extended Data Fig. 3c-d). These results were confirmed by over-expressing truncated MINAS-60 coding sequences starting from ATG1 or ATG6 and comparing their products’ sizes with the wild-type construct (Extended Data Fig. 3e). Both the long (~20 kDa) and short (~15 kDa) isoforms of MINAS-60 localize to the nucleolus when over-expressed, suggesting that both may be functional (Extended Data Fig. 3f). The MINAS-60 smORF is therefore entirely contained within the RBM10 coding sequence (Fig. 2a). We speculate that translation of these two MINAS-60 isoforms from RBM10 transcript variant 1 may occur via leaky scanning, and that the additional MINAS-60 isoform diversity observed in KI cells could be generated by alternative splicing, analogous to PTBP3²⁹.

ClustalW alignment of hypothetical murine, bovine, and primate MINAS-60 homologs revealed significant sequence similarity, suggesting MINAS-60 is conserved in mammals (Fig. 2d). To determine whether mouse MINAS-60 can be translated, an expression vector containing the 5'UTR of mouse RBM10 transcript variant 1 through the stop codon of mouse MINAS-60 was transfected into 3T3 cells, followed with immunostaining. As shown in Figure 2e, over-expressed mouse MINAS-60 appeared in the nucleolus. Together, these results indicate that MINAS-60 is endogenously expressed, cell-cycle regulated and conserved from humans to mouse.

MINAS-60 is associated with nucleolar LSU assembly factors

Because many microproteins and alt-proteins bind to and regulate other proteins³⁰, we performed a co-immunoprecipitation (co-IP) with nuclear lysates from MINAS-60-FLAG KI cells, and HEK 293T cells as a control for non-specific bead-binding proteins³¹. Two specific SDS-PAGE bands were observed in the KI co-IP that, upon label-free quantitative proteomics³², yielded 17 proteins enriched >30-fold over control. GO analysis showed that the top 2 enriched biological processes were ribosome biogenesis and ribosomal large subunit (LSU) biogenesis (Extended Data Fig. 4a-c, and Supplementary Data S5-6). To obtain a more comprehensive picture of MINAS-60-associated proteins, we performed co-IPs and analyzed the entire molecular weight range with quantitative proteomics; while abundant ribosomal proteins limited detection sensitivity for other proteins, we observed enrichment of four LSU assembly factors: GTPBP4, MRTO4, BRIX1 and NOP2 (Fig. 3a and Supplementary Data S7-8). Co-IP followed by immunoblotting confirmed the association of these four factors with MINAS-60 (Fig. 3b). Interaction of MINAS-60 with GTPBP4 and MRTO4 did not depend on RNA, because these associations largely survived RNaseA treatment. In contrast, co-purification of MINAS-60 with BRIX1 and NOP2 was diminished after treatment with RNaseA, suggesting that their association with MINAS-60 was likely indirect. We therefore hypothesized that MINAS-60 interacts with nucleolar, late pre-60S particles containing GTPBP4 and MRTO4 to regulate ribosome biogenesis^15,33. Additional negative controls, including HEK 293T cells stably transformed with empty vector, could be useful in further refining the remaining MINAS-60-enriched proteins listed in Supplementary Data S6 and S8.

To determine whether MINAS-60 associates with pre-60S particles, we performed sucrose gradient fractionation of nuclear extracts of HEK 293T cells stably expressing epitope-tagged MINAS-60, followed by immunoblotting. A subpopulation of over-expressed MINAS-60 co-sedimented with ribosome assembly factors in high-molecular weight fractions coincident with pre-ribosomal particles (Fig. 3c), consistent with a role for MINAS-60 in pre-60S biogenesis.

MINAS-60 inhibits protein synthesis and cell proliferation

We hypothesized that MINAS-60 could regulate ribosome biogenesis via its association with GTPBP4 and MRTO4. Ribosome biogenesis is required for protein synthesis, which promotes cell growth and proliferation, and thus is commonly upregulated in cancer cells^11,34. We reasoned that, if MINAS-60 regulates ribosome biogenesis, its absence should result in changes to protein synthesis and cell proliferation. We required a system to query the function of MINAS-60 independent of RBM10, despite their co-expression from the same transcript. To this end, we knocked down (KD) RBM10 in HEK 293T cells with two different shRNAs, which silence the entire mRNA and both proteins. To deconvolute phenotypic effects specific to MINAS-60, as well as to exclude off-target effects of shRNA, we generated rescue cell lines stably expressing MINAS-60 (Rescue_MINAS-60) or RBM10 (Rescue_RBM10) on the KD background. qRT-PCR and immunoblotting revealed that RBM10 mRNA is efficiently silenced by both shRNAs, and the KD cells express MINAS-60 or RBM10 after rescue (Extended Data Fig. 5a-b). Noting that the RBM10 rescue construct could be subject to leaky translation to produce both RBM10 and MINAS-60, we also rescued the KD with an RBM10 construct bearing an A₃₉₈TG to TAA mutation that eliminates the first MINAS-60 start codon while preserving RBM10 translation (Extended Data Figure 5a-b).

To test the effect of MINAS-60 on cellular protein synthesis, we labeled nascent peptides in the control, RBM10 KD, Rescue_MINAS-60, Rescue_RBM10 and Rescue_RBM10(A₃₉₈TG-TAA) cell lines with puromycin followed by anti-puromycin immunoblotting³⁵. As shown in Figure 3d and Extended Data Figure 5c, KD of RBM10 led to a significant increase in global protein synthesis, which was rescued by reintroduction of MINAS-60. Partial rescue by RBM10 was also observed, but was likely due to leaky MINAS-60 expression, because this effect was not present in the RBM10(A₃₉₈TG-TAA) rescue cells. Similar results were observed for a second shRNA targeting RBM10 (Extended Data Fig. 5d-e). MINAS-60, and not RBM10, may therefore play a negative regulatory role in ribosome biogenesis.

We next sought to determine which regions of MINAS-60 are required for its protein-protein interactions and inhibitory effect on protein translation. We transfected deletion mutants of MINAS-60 into HEK 293T cells, followed by FLAG-IP and immunoblotting. As shown in Extended Data Figure 6a, deletion of the N-terminal or middle region of MINAS-60 maintained association with GTPBP4 and MRTO4, while deletion of the C-terminal region abolished the association. We then examined the effect of the deletion constructs on global protein synthesis in HEK 293T. As shown in Extended Data Figures 6b-c, over-expression of MINAS-60 full-length, N-terminal deletion or middle region deletion reduced puromycylation, while over-expression of MINAS-60 C-terminal deletion did not, compared with empty vector control. The C-terminus of MINAS-60 is therefore required for association with GTPBP4 and MRTO4 and negative regulation of ribosome biogenesis.

Because protein synthesis, cell growth and proliferation are linked^10,11, we asked whether MINAS-60 regulates cell proliferation. As shown in Figure 3e, RBM10 depletion led to a significant increase in cell proliferation, consistent with a published report³⁶. Remarkably, this increase was rescued by reintroduction of MINAS-60 alone, and partially rescued by RBM10. Similar results were observed for a second shRNA targeting RBM10 (Extended Data Fig. 5f). These results suggest that MINAS-60 negatively regulates ribosome biogenesis, subsequently decreasing cellular protein synthesis and cell proliferation.

MINAS-60 downregulates cytoplasmic LSU export and function

Eukaryotic ribosome biogenesis can be divided into sequential processes³⁷: pre-rRNA transcription; chemical modification and processing of the pre-rRNA in the nucleolus; folding, assembly and maturation of the pre-ribosomal subunits in the nucleolus and nucleus; and export and final maturation of ribosome subunits in the cytoplasm. Because MINAS-60 co-purified with proteins involved in various steps of LSU biogenesis, we wished to determine the step in this process that it regulates.

To determine whether MINAS-60 controls pre-rRNA transcription, we performed qRT-PCR targeting the primary pre-rRNA (47S/45S/30S) using previously published primers³⁵ in lysates from control, RBM10 KD, Rescue_MINAS-60 and Rescue_RBM10 cells. No significant differences were observed (Extended Data Fig. 7a). To determine whether MINAS-60 regulates the processing of LSU pre-rRNA, we performed northern blotting analysis with these cell lines using a probe complementary to ITS2, which detects all LSU pre-rRNA processing products³⁵. Again, no significant differences were observed (Extended Data Fig. 7b). Finally, no significant differences were observed in the ratio of the mature 28S/18S rRNAs in whole-cell lysates (Extended Data Fig. 7c). MINAS-60 therefore does not regulate transcription or processing of the pre-rRNA.

To determine whether MINAS-60 regulates nucleocytoplasmic export of the pre-LSU, we quantified the ratio of nuclear vs. cytoplasmic RPL29-GFP³⁸ stably expressed in the control, KD, and rescue cell lines. As shown in Figure 4a and Extended Data Figure 8a, KD of RBM10 decreased the ratio of nuclear to cytoplasmic RPL29-GFP, which can be rescued by reintroduction of MINAS-60, and partially rescued by RBM10 reintroduction, likely due to leaky MINAS-60 translation. Similar results were observed for a second shRNA targeting RBM10 (Fig. 4a, right, and Extended Data Fig. 8b-c). As a control, to determine whether MINAS-60 regulates nucleocytoplasmic export of the pre-40S ribosomal subunit, we quantified the ratio of nuclear vs. cytoplasmic RPS2-GFP³⁸ stably expressed in these cell lines. No significant changes were observed, though RPS2-GFP signal was predominantly cytoplasmic, consistent with prior reports of faster pre-40S maturation and export^14,38 (Extended Data Fig. 9). MINAS-60 therefore may specifically inhibit a step in the LSU biogenesis pathway between rRNA processing and cytoplasmic export.

The observation of faster pre-60S export in the absence of MINAS-60 predicted that the same conditions should increase mature cytoplasmic 60S ribosomal subunits. To test this, we performed cytoplasmic polysome profiling. As shown in Figures 4b-c, knockdown of RBM10 resulted in a small but significant increase in the ratio of cytoplasmic 60S/40S ribosome subunits. This increase was rescued by reintroduction of MINAS-60, and partially rescued by RBM10. Similar results were observed for a second shRNA targeting RBM10 (Fig. 4c, right and Extended Data Fig. 10a). Conversely, over-expression of MINAS-60 led to a small but significant decrease in the ratio of cytoplasmic 60S/40S ribosome subunits as compared to empty vector (Extended Data Figure 10b-c). More significantly, over-expression of MINAS-60 dramatically decreased the levels of 80S monosomes (Extended Data Figure 10b-c), suggesting that fewer non-functional ribosomes are produced when MINAS-60 levels are high. Taken together, these results suggest that MINAS-60 negatively regulates the rate of assembly and nucleocytoplasmic export of pre-60S particles to ensure that they are correctly assembled.

MINAS-60 is a checkpoint inhibitor of pre-60S assembly

We speculated that MINAS-60 is a checkpoint inhibitor that ensures correct pre-60S assembly prior to export from the nucleus, which would suggest that the protein composition of pre-60S particles should change in cells lacking MINAS-60. BRIX1 and MRTO4 are markers of early-to-intermediate, or intermediate-to-late, pre-60S particles, respectively¹⁵. We examined their interactomes as a readout of changes in pre-60S protein composition in the presence or absence of MINAS-60 (Fig. 5a). Affinity purification of MRTO4-FLAG in control or RBM10 KD HEK 293T cells, followed by quantitative proteomics, identified statistically significant increases in several intermediate- and late-stage LSU assembly factors co-purified with MRTO4 in RBM10 KD cells (Fig. 5b, Supplementary Data S9-10). These increases were further confirmed by immunoblotting (Fig. 5c). Similar results were observed for BRIX1 co-IP, though fold changes were smaller (Extended Data Fig. 10d-f, Supplementary Data S11-12). These results suggest that increased interaction between pre-60S assembly factors occurs in the absence of MINAS-60, consistent with its potential function as a checkpoint inhibitor of LSU assembly.

Data analysis.

Two-tailed t-test was performed using Microsoft Excel (version 2013) or GraphPad Prism (version 8.0.0), one-way or two-way ANOVA (Dunnett’s test) was performed using GraphPad Prism (version 8.0.0), and F-tests were performed to evaluate equal variance between samples.

Antibodies.

Primary antibodies for immunoblotting include: anti-FLAG (1:500, Sigma, F3165 or 1:1500, Cell Signaling, 14793); anti-HA (1:3000, Invitrogen, 71-5500); anti-β-actin (1:3000, Invitrogen, MA5-15739); anti-GTPBP4 (1:1000, Abclonal, A4565); anti-MRTO4 (1:1000, ThermoFisher, 20194-1-AP); anti-NSA2 (1:1000, Abclonal, A14475); anti-NOP2 (1:1000, Cell Signaling, 25017); anti-BRIX1 (1:1000, Abclonal, A14481); anti-RPF2 (1:1000, Abclonal, A17224); anti-GNL2 (1:1000, Abclonal, A13191); anti-RPL11 (1:2000, Cell Signaling, 18163); anti-RPL36 (1:1000, Bethyl Laboratories, A305065A-M); anti-RPS2 (1:1000, Invitrogen, PA5-30160); anti-RBM10 (1:1000, Abcam, ab72423); anti-puromycin (1:1000, Kerafast, EQ0001); anti-GFP (1:1000, Abcam, ab183734); anti-cyclin B1 (1:1000, Cell Signaling, 4138). Immunoprecipitation was performed with anti-FLAG M2 affinity gel (Sigma, A2220). Secondary antibodies for immunoblotting were: goat anti-rabbit IgG horseradish peroxidase conjugate (1:8000, Rockland, 611-1302) and goat anti-mouse IgG horseradish peroxidase conjugate (1:4000, Rockland, 610-1319-0500). Primary antibodies for immunostaining were: rabbit anti-HA (1:500, Invitrogen, 71-5500) and mouse anti-Fibrillarin (1:500, Abcam, ab4566). Secondary antibodies for immunostaining were: goat anti-rabbit IgG Alexa fluor 568 (1:500, Invitrogen, A11011) and goat anti-mouse IgG Alexa fluor 647 (1:500, Invitrogen, A21235).

Cloning and genetic constructs.

The 5'UTR of human DRAP1, PRR3 transcript variant (tv) 2, PRH1 tv 1, CNPY2 tv 1, CACTIN tv 1, or RBM10 tv 1 through the stop codon of each alt-ORF was PCR amplified with a 3' FLAG-HA epitope tag from reverse-transcribed HEK 293T cDNAs, then cloned into pcDNA3. MINAS-60 deletions or mutations bearing a FLAG-HA tag were generated by ligating PCR products into BamHI and EcoRI sites in the pcDNA3 vector. For generation of HEK 293T cells stably expressing MINAS-60, the FLAG-HA tagged coding sequence was PCR amplified and cloned into pLJM1. The RPS2 cDNA clone was purchased from Addgene (a gift from Thomas Tuschl), and the coding sequences of MRTO4, BRIX1 and RPL29 were amplified by PCR from HEK 293T cDNAs, then subcloned into pJLM1. RNAi constructs were made by synthesizing oligonucleotides encoding a 21 bp short hairpin RNA that targets RBM10 (shRNA1, CTTCGCCTTCGTCGAGTTTAG; shRNA2, TCCAACGTGCGCGTCATAAAG), then cloned into pLKO.1. Empty pLKO.1 vector control was purchased from Sigma (SHC001). qPCR primer sequences are provided in Supplementary Table 4.

Cell culture, lentivirus production and stable cell line generation.

HEK 293T and 3T3 cells were purchased from ATCC and early-passage stocks were established. Cells cultured as previously described⁶ up to only 10 passages. To produce lentivirus and generate stable cell lines, HEK 293T cells were co-transfected using polyethyleneimine (Polysciences, 23966) with expression construct in pLJM1, along with pMD2.G and psPAX2, and growth media were replaced after 7-8 h. 48 h post-transfection, media containing viruses was harvested, filtered through a 0.45-μm filter, and infection was performed by mixing with two volumes of fresh media containing suspended HEK 293T cells. 24 h post-infection, the growth media was replaced. 48 h post-infection, stable cells were selected with 4 μg/mL puromycin for 2 days. Early stocks of stable cell lines were established after selection. Stable cell lines were released from puromycin for 2 days prior to use in experiments.

Immunostaining and live-cell imaging.

HEK 293T cells were plated on glass coverslips and, when necessary, transfected the next day. Forty-eight hours later, the cells were fixed in 10% formalin for 15 min at room temperature (RT), permeabilized with PBS containing 0.2% (v/v) Triton X-100, then incubated with primary antibodies for 18 h at 4°C. After washing with PBS, the cells were incubated with secondary antibodies and DAPI for 1 h at RT, washed with PBS and mounted with Mowiol (Sigma, 81381) before viewing.

HEK 293T cells stably expressing RPL29-GFP or RPS2-GFP were grown to 70% confluency on coverslips in 12-well plates. Coverslips were inverted and imaged in pre-warmed DMEM with 10% FBS, 1% penicillin-streptomycin in MatTek imaging dishes. Confocal imaging was performed on a Leica SP8 LS confocal microscope with 63× oil immersion objective with atmosphere-controlled stage at 37°C. Nuclear/cytoplasmic ratios of RPL29 and RPS2 were measured using the ImageJ (version ImageJ2) Intensity Ratio Nuclei Cytoplasm Tool (RRID:SCR_018573; https://github.com/MontpellierRessourcesImagerie/imagej_macros_and_scripts/wiki/Intensity-Ratio-Nuclei-Cytoplasm-Tool).

Immunoprecipitation and proteomics.

Control or MINAS-60 KI HEK 293T cells were grown to 80-90% confluency in 15 cm dishes. Cells were harvested and suspended in 1 mL nuclear isolation buffer (10 mM HEPES-KOH pH 7.4, 100 mM KCl, 5 mM MgCl₂ with 0.5% NP40 and Roche Complete protease inhibitor cocktail tablets (Roche, 11873580001)), and incubated on ice for 10 min, followed by centrifugation at 3,000 g, 4°C, 3 min. Nuclear pellets were suspended in 1 mL lysis buffer [Tris-buffered saline (TBS) with 1% Triton X-100 and Roche Complete protease inhibitor cocktail tablets], followed with sonication and immunoprecipitation as previously described⁶. Elution was in 40 μL of 3× FLAG peptide (Sigma, F4799), at a final concentration of 100 μg/mL in lysis buffer at 4°C for 1 h. Eluted proteins were subjected to SDS-PAGE prior to LC-MS/MS.

Gel slices (protein bands or entire lanes) were digested with trypsin at 37°C for 14-16 h. Resulting peptide mixtures were extracted from the gel, dried, and residual detergent was removed with ethyl acetate followed by de-salting with peptide cleanup C18 spin column (Agilent Technologies, 5188-2750). Peptides were re-suspended in 35 μL 0.1% formic acid (FA), followed by centrifugation at 21,130 g, 4°C, 30 min. 5 μL of each sample was injected on a pre-packed column attached to a nanoAcquity UPLC (Waters) in-line with a Thermo Scientific™ Q Exactive™ Plus Hybrid QuadrupoleOrbitrap™ mass spectrometer and a 130-min gradient was used to further separate the peptide mixtures as follows (solvent A: 0.1% FA; solvent B: acetonitrile (ACN) with 0.1% FA): Isocratic flow was maintained at 0.1 μL/min at 1% B for 40 min, followed by linear gradients from 1% B to 6% B over 2 min, 6% B to 24% B over 48 min, 24% B to 48% B over 5 min, 48% B to 80% B over 5 min. Isocratic flow at 80% B was maintained for 5 min, followed by a gradient from 80% B to 1% B over 5 min, and isocratic flow at 1% B was maintained for 10 min. The full MS was collected over the mass range of 300-1,700 m/z with a resolution of 70,000 and the automatic gain control (AGC) target was set as 3 x 10⁶. MS/MS data was collected using a top 10 high-collisional energy dissociation method in data-dependent mode with a normalized collision energy of 27.0 eV and a 1.6 m/z isolation window. MS/MS resolution was 17,500 and dynamic exclusion was 90 seconds.

For alt-protein and microprotein identification, ProteoWizard MS Convert was used for peak picking and files were analyzed using Mascot (version 2.5.1). Oxidation of methionine and N-terminal acetylation were set as variable modifications, and a previously reported three-frame translation of assembled transcripts from HEK 293T mRNA-seq was used as the database exactly as previously described⁷. Identified peptides were searched against human UniProt protein database (version 2019) using a string-matching algorithm. Peptides found to be identical to annotated proteins were discarded. The remaining peptides were searched against non-redundant human protein sequences using on-line BLAST (Basic Local Alignment Search Tool). Any peptides found to be less than two amino acids different from the nearest protein match were discarded. For co-IP proteomics searches and quantitative analysis, files were analyzed using MaxQuant (version 1.6.8.0), oxidation of methionine and N-terminal acetylation were set as variable modifications, and human UniProt plus MINAS-60 was used as the search database. For all analyses, a mass deviation of 20 p.p.m. was set for MS1 peaks, MS/MS tolerance was 0.6 Da, and a maximum of two missed cleavages were permitted. Maximum false discovery rates (FDR) were set to 1% both on peptide and protein levels. Minimum required peptide length was five amino acids. Protein quantitation was calculated from the LFQ intensity ratio of pulldown to negative control samples using MaxQuant with standard parameters. P-values (t-test) were calculated using Perseus (version 1.5.8.5) with standard parameters.

Dibenzocyclooctyne (DBCO) bead construction.

NHS-activated beads (Pierce, 88826) were conjugated to dibenzocyclooctyne-amine (Sigma, 761540) by 90 min incubation at room temperature with rotation in a saturated 100 mM sodium bicarbonate (pH 8.0) solution, then washed and blocked for 2 h with 3 M ethanolamine. After blocking, the beads were re-suspended in PBS prior to immediate protein conjugation.

BONCAT (bio-orthogonal non-canonical amino acid tagging) and streptavidin pulldown.

HEK 293T cells were grown to 80-90% confluency in 15 cm dishes, treated with methionine aminopeptidase inhibitor TNP470 (50 nM) for 2 h, then immersed in methionine-free DMEM (Corning, 17-204-CI) for 30 min before addition of 4 mM AHA (Click Chemistry tools, 1066-1000) in methionine-free DMEM with 10% FBS. For stress conditions, cells were treated with 4 mM AHA with simultaneous exposure to 500 μM sodium arsenite, 20 μM etoposide (Sigma, E1383), or 1 mM DTT. After a 2 h incubation, the cells were washed twice with cold PBS, harvested, and flash frozen.

For gel-based selection of AHA-labeled small proteins, cells from one 15 cm dish were lysed in 2 mL 1% (w/v) SDS in PBS by boiling for 10 min and then brought up to 0.1% SDS by adding 18 mL PBS with 0.2% TritonX-100. In a separate tube, 200 μL of 70 mM CuSO₄ was combined with 20 μL of 100 mM Tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA, Click Chemistry tools, 1061-100) and 70 μL of 500 mM Tris(2-carboxyethyl)phosphine hydrochloride, then incubated for 10 min at RT. 20 μL of 25 mM biotin-alkyne (BALK, Click Chemistry tools, 1266-5) was added to the mixture and then combined with 20 mL cell lysate, followed with rotation at RT for 6 h. Labeled lysates were then passed through a 3 kDa Amicon Ultra 15 mL column, and diluted 1000 times with RIPA buffer (10 mM Tris-HCl (pH=7.4), 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 140 mM NaCl) to remove excess BALK, followed by concentration to 1 mL. A 50 μL aliquot of streptavidin magnetic beads (NEB) was washed twice with 1 mL RIPA before gentle rotation with 1 mL of biotinylated proteins for 1 h at RT. After binding, beads were washed twice with 1 mL RIPA, once with 1 M KCl, once with 0.1 M sodium carbonate, once with 2 M urea, and twice with RIPA. Bound proteins were eluted by boiling in 1×SDS loading buffer containing 20 mM DTT and 2 mM biotin for 10 min. After elution, proteins were electrophoresed on a 16% tricine gel, and gel bands corresponding in the 2-15 kDa size range were excised for proteomic analysis. As a control, unlabeled cell lysates were size-selected on a tricine gel as described above.

DBCO bead enrichment of AHA-labeled proteins and ERLIC (electrostatic repulsion hydrophilic interaction chromatography) fractionation.

AHA-labeled cells from one 15 cm dish were lysed by boiling in 2 mL 50 mM HCl with 0.01% 2-mercaptoethanol and 0.05% TritonX-100 for 10 min. Lysates were centrifuged at 21,100 g, 4°C for 30 min, and passed through a 5 μm filter before size selection with a C8 column (Agilent Technologies, 12102100) essentially as previously reported²⁰. The C8 column was pre-conditioned with 1 bed volume of methanol and 2 bed volumes of 0.25 M triethylammonium formate (TEAF, pH 3.0). Up to 2 bed volumes of cell lysate were loaded on the column, followed by two washes with 2 bed volumes of TEAF and elution with 2 bed volumes of ACN:TEAF (3:1). The sample was dried and reconstituted in 300 μL PBS. Size-selected lysates were then incubated with 50 μL DBCO beads for 1 h at RT. The beads were washed twice with 1 mL RIPA, once with 1 M KCl, once with 0.1 M sodium carbonate, once with 2 M urea, twice with RIPA, and 6 times with PBS. Proteins conjugated to the beads were subjected to reduction, alkylation and on-bead trypsin digestion according to standard protocols⁵⁰.

For method validation, digested peptides were directly analyzed via 1D LC-MS/MS as described above. For subsequent analysis of control and stress conditions, digested peptides were pre-fractionated using ERLIC on an Agilent 1100 HPLC. Peptides were re-suspended in 55 μL of 85% ACN/0.1% FA and 50 μL was loaded onto a polyWAX LP column (150×1.0 mm; 5 μm 300 Å; PolyLC). Samples were run on an 80 min gradient protocol as follows (Solvent A: 80% ACN 0.1% FA; Solvent B: 30% ACN 0.1% FA): Isocratic flow was maintained at 100% A at a flow rate of 0.3 mL/min for 5 min, followed by a 17 min linear gradient to 8% B, and a 25 min linear gradient to 45% B. Finally, a 10 min gradient to 100% B was followed by a 5 min hold at 100% B before a 10 min linear gradient back to 100% A, followed by an 8 min hold at 100% A. The digested peptides were separated into 15-18 fractions which were dried and re-suspended in 7 μL of 3:8 70% FA : 0.1% TFA before LC-MS/MS analysis.

Generation of knock-in cell lines.

MINAS-60 3xGFP11-FLAG-HA KI HEK 293T cells were generated using CRISPR-Cas9. Guide RNAs (gRNAs) were designed with the guide design tool from the Zhang lab (crispr.mit.edu) to target the RBM10 genomic region: gRNA1, 5'- TGTCGGCCAGGATTCCTACG-3'; gRNA2, 5'- CCCGATAGTCGCCGTCTCGG-3'. Double-stranded DNA oligonucleotides corresponding to the gRNAs were inserted into pSpCas9(BB)-2A-GFP vector (Addgene, as a gift from F. Zhang, MIT, Cambridge, MA). A donor plasmid containing 300 bp left-arm homology and 300 bp right-arm homology around the stop codon of MINAS-60, which are separated by 3xGFP11-FLAG-HA tag and BamHI / NotI restriction sites, was synthesized by GenScript, and a DNA sequence containing pGK promoter and hygromycin B resistance gene were subcloned into the donor plasmid using the BamH1 and NotI restriction sites. An equal mixture of the gRNA and donor plasmids were transfected into HEK 293T cells using polyethyleneimine, and hygromycin B selection was performed 2 days post-transfection. MINAS-60-3xGFP11-FLAG-HA KI cells were validated by genomic PCR and sequencing.

Puromycin incorporation.

SUnSET was used to measure protein synthesis⁵¹. HEK 293T cells were grown to 80-90% confluency in 6 well plates, then growth media was replaced with media containing 1 μM puromycin and cells were incubated at 37°C for 1 h according to a previously published protocol⁴³. Cells were then washed once with PBS, harvested and analyzed with immunoblotting.

Sucrose gradient profiling.

HEK 293T cells were seeded in 15 cm dishes at 2.5×10⁷ cells per dish and cultured 24 h. Cells were then treated with 100 μg/mL cycloheximide for 5 min, washed with cold PBS containing 100 μg/mL cycloheximide twice, harvested and flash frozen.

For cytoplasmic polysome profiling, after thawing on ice, cells were lysed in polysome lysis buffer on ice for 10 min (20 mM HEPES-KOH pH 7.4, 100 mM KCl, 5 mM MgCl₂, 100 μg/mL cycloheximide, 1 mM DTT with 1% TritonX-100, Roche Complete protease inhibitor cocktail tablets and ribonuclease inhibitors (Promega N2511)), followed by centrifugation at 21,130 g, 4°C, 10 min. Supernatants were normalized according to absorbance (A260) and layered onto 12 mL 10-50% sucrose gradients (20 mM HEPES-KOH pH 7.4, 100 mM KCl, 5 mM MgCl₂, 100 μg/mL cycloheximide, 1 mM DTT with Roche Complete protease inhibitor cocktail tablets and ribonuclease Inhibitors), followed by centrifugation in an SW-41Ti rotor at 252,878 g, 4°C, 3 h, then sampled using a Biocomp Gradient profiler (model 251) with constant monitoring of optical density at 254 nm using standard parameters. Data were analyzed with Excel (version 2013) and ImageJ2.

For sucrose gradient sedimentation analysis of nuclear lysates, thawed cells were suspended in 1 mL nuclear isolation buffer (10 mM HEPES-KOH pH 7.4, 100 mM KCl, 5 mM MgCl2 with 0.5% NP40, Roche Complete protease inhibitor cocktail tablets and ribonuclease inhibitors), incubated on ice for 10 min, followed by centrifugation at 3,000 g, 4°C, 3 min. Nuclear pellets were suspended in 1 mL nuclear lysis buffer (20 mM HEPES-KOH pH 7.4, 300 mM KCl, 5 mM MgCl₂, 1 mM DTT with 1% TritonX-100, Roche Complete protease inhibitor cocktail tablets and ribonuclease inhibitors), followed with sonication (50% intensity, 5 s pulse with 25 s rest, 5×, MICROSON XL 2000), and centrifugation at 21,130 g, 4°C, 10 min. The supernatants were normalized to A260 and layered onto 12 mL 10-50% sucrose gradients, centrifuged and analyzed as described above.

Crystal violet staining.

4×10⁴ HEK 293T cells were seeded in 12-well plates, and fixed in 10% formalin for 15 min at RT every 24 h. After washing with ddH₂O, cells were stained with 0.1% crystal violet in methanol for 30 min at RT in the dark, followed with three ddH₂O washes and dried. Cells were then immersed in 1 mL 10% acetic acid with shaking for 20 min. 20 μL solution was combined with 80 μL ddH₂O in a 96-well plate, and optical density at 590 nm was monitored with Synergy™ HT.

Cell cycle synchronization.

Following a protocol for cell cycle synchronization⁵², 1.5×10⁵ HEK 293T cells were seeded in 12-well plates and cultured overnight. Cells were then treated with 2 mM thymidine for 16 h, followed by two PBS washes, and incubated with fresh media for 9 h before the second 2 mM thymidine block for 14 h. Cells were washed with PBS, then incubated with fresh media to release from the G1/S boundary, and fixed in 10% formalin for immunostaining or harvested for immunoblotting.

Northern blot.

Total RNA was extracted from HEK 293T cells using TRIzol. To determine changes in levels of LSU pre-rRNA intermediates, 3 μg of total RNA was run on a 1% agarose/1.25% formaldehyde gel in a 1.5 M tricine/ 1.5 M triethanolamine buffer, followed by overnight transfer to a Hybond XL nylon membrane (GE Healthcare, RPN 303S) in 10× saline-sodium citrate transfer buffer after a 15 min soak in 0.5 M sodium hydroxide solution. Membranes were exposed to UV (254 nm) to immobilize the RNA, followed by incubation with denatured yeast tRNA for 1 h at 42°C, and hybridized overnight at 37°C with 5' end radiolabeled oligonucleotide probe (P4 5'-CGGGAACTCGGCCCGAGCCGGCTCTCTCTTTCCCTCTCCG-3') in a solution of 7.5× Denhardt's solution, 5× sodium chloride-sodium phosphate-EDTA buffer with 0.1% SDS, as previously reported⁴⁶. Membranes were also hybridized with a 7SL probe (7SL 5'-TGCTCCGTTTCCGACCTGGGCCGGTTCACCCCTCCTT-3') as a loading control.