Maggie Pui Yu Lam

SUMMARY: Endoplasmic reticulum (ER) stress is associated with many human diseases, but current understanding of how ER stress responses reshape the stressed proteome over time is still emerging. Specifically, how organellar protein quality control and clearance pathways coordinate to maintain proteostasis in early and prolonged ER stress is unclear. Here we describe a spatiotemporal proteomic strategy termed simultaneous proteome localization and turnover analysis with time resolution (SPLAT-TR) to interrogate the synthesis, degradation, and localization changes of over 4,000 proteins in early (1-4 hours) and prolonged (16-24 hours) ER stress. As ER stress progresses, each time point is distinguished by distinct protein translocation and clearance regulations, coupled to differential organellar proteostasis and usage of protein degradation pathways. Mitochondria show a bimodal response, with a shift from early activation of respiratory protein towards protein quality control pathways in prolonged stress, coinciding with energetics decline. In the Golgi, the increased clearance of collagen proteins is paradoxically coupled to an increased synthesis of secretory pathway components, suggesting removal by secretion. In the ER, the proteome remodels under prolonged stress via increased synthesis of UFMylation and ER-phagy-related proteins coupled to the selective degradation of ER membrane and microdomain proteins. Inhibition of UFMylation alters ER-phagy receptor usage and synergistically induces stress-induced cell death. These results present a systematic delineation of cell compartment proteostasis under unfolded protein response and highlight the utility of SPLAT-TR to investigate time-dependent cellular events.

Summary: Coordination of biological function requires the partition of cellular components including into biomolecular condensates, but an overall landscape of how protein compartmentalize into higher-order assemblies under stress is still emerging. We apply proteome-wide solubility profiling to compare the compositions of NP-40-insoluble proteins and their phosphorylation status, using a new mass spectrometry-based hybrid bottom-up and chemical middle-down proteomics approach to analyze the solubility behavior of 8,740 proteins and 31,647 phosphopeptides under normal and ER stress conditions. Cell stress induces a pervasive differential partition of proteins in and out of detergent- insoluble cellular compartments. This differential partition is partially orthogonal to stress-induced abundance changes and comprises both phosphorylation-dependent and phosphorylation-independent mechanisms. Whereas phosphorylation-independent partition changes involve largely secretory pathway proteins and implicate higher-order assemblies of chaperones and clients, phosphorylation-based partitions suggest a dynamic rearrangement of biomolecular condensate compositions across cytoplasmic and nuclear ribonucleoprotein assemblies. The accumulation of serine/arginine rich (SR) proteins and other annotated nuclear speckle members in the condensate-rich proteome fractions emerges as a central feature of stress-induced remodeling. Our results establish global solubility dynamics as an integral component of proteome stress response and implicates broad involvements of splice factor spatial reorganization as a prominent facet of ER stress response.

Alternative splicing is an important regulatory layer in gene expression, but knowledge on the isoform protein molecules continue to lag their canonical counterparts. An open question is whether alternative protein isoforms feature different half-life than the canonical counterpart, which could indicate differential usage and functional diversification. Here we combined a proteogenomics approach with heavy water-based protein turnover analysis to survey 24 pairs of canonical-alternative protein isoforms in the mouse heart. The results provide a reference on their numerical half-life and also reveal widespread differences in isoform stability.

N-glycosylation plays essential roles in the folding, trafficking, and maturation of proteins in the secretory pathways, but how individual protein and residue glycosylation rewires under endoplasmic reticulum (ER) stress is unknown. Particularly, intact glycopeptide data that retain the connectivity between glycosylation sites and the attached glycans are needed to reveal the micro- and macro- heterogeneity of N-glycosylation sites and their permutations in stressed cells. Here, we developed an optimized magnetic polyethyleneimine boronic acid-containing scaffold (mPBA) enrichment workflow to achieve sensitive and broad enrichment of intact glycoproteins for mass spectrometry analysis, quantifying 13759 unique protein-, site-, and glycoform combinations, termed glycopeptidoforms, in normal and stressed cells while requiring only 0.1 to 0.5 mg total peptide input. The data reveals a systems-level shift in the fate of hundreds of glycoproteins. N-glycosylation changes are highly dynamic, with magnitude far exceeding protein expression changes, and showing complex protein-, site-, and glycan- specific granularity. Individual glycoform reconfigurations can be observed that suggest lesions within specific steps in protein maturation and trafficking pathways. Mannose trimming is disrupted across multiple proteins and cell states, suggesting a central feature of ER stress mediated glycoproteome remodeling. Together, these results reveal molecular details into the remodeling of protein secretory pathways upon ER stress and highlight the utility of mPBA for sensitive N-glycoproteomics studies.

We describe a bioinformatics strategy to find associations between the secretome-coding genes and functional pathways of two organs. This approach may be used to find cross-talk signals between the heart, adipose, liver, and other tissues. Applied to Genotype-Tissue Expression data, it suggests a potential role of circulating galectin-3 in regulating cardiac protein synthesis and homeostasis pathways.

Pathological cardiac remodeling is associated with the reactivation of fetal genes, yet the extent of the heart’s fetal gene program and its impact on proteome compositions remain incompletely understood. Here, using a proteome-wide protein ratio quantification strategy with mass spectrometry, we identified pervasive isoform usage shifts in fetal and postnatal mouse hearts, involving 145 pairs of highly homologous paralogs and alternative splicing–derived isoform proteins. Proteome-wide ratio comparisons readily rediscovered hallmark fetal gene signatures in muscle contraction and glucose metabolism pathways, while revealing what we believe to be previously undescribed isoform usage in mitochondrial and gene-expression-regulating proteins, including PPA1/PPA2, ANT1/ANT2, and PCBP1/PCBP2 switches. Paralogs with differential fetal usage tend to be evolutionarily recent, consistent with functional diversification. Alternative splicing adds another rich source of fetal isoform usage differences, involving PKM M1/M2, GLS1 KGA/GAC, PDLIM5 long/short, and other spliceoforms. When comparing absolute protein proportions, we observed a partial reversion toward fetal gene usage in pathological hearts. In summary, we present a ratiometric catalog of paralogs and spliceoform pairs in the cardiac fetal gene program. More generally, the results demonstrate the potential of applying the proteome-wide ratio test concept to discover new regulatory modalities beyond differential gene expression.

The synthesis and degradation rates of proteins form an essential component of gene expression control. Heavy water labeling has been used in conjunction with mass spectrometry to measure protein turnover rates, but the optimal analytical approaches to derive turnover rates from the mass isotopomer patterns of deuterium-labeled peptides continue to be a subject of research. Here, we describe a method that comprises (1) a nearest lookup of numerically approximated peptide isotope envelopes, coupled to (2) the selection of optimal mass isotopomer pairs based on peptide sequence rules, to calculate the molar fraction of new peptide synthesis in heavy water labeling mass spectrometry experiments. We validated our approach using an experimental calibration standard comprising mixtures of fully unlabeled and fully labeled proteomes. We then reanalyzed 17 proteome-wide turnover experiments from four mouse organs across multiple data sets and showed that the combined nearest-lookup and rule-based mass isotopomer ratio selection method increases the coverage of well-fitted peptides in protein turnover experiments by up to 58 ± 13%. The workflow is implemented in the Riana software tool for protein turnover analysis and may avail ongoing efforts to study the synthesis and degradation kinetics of proteins in animals on a proteome-wide scale.