Week 4 (Proteomics)

What is the Proteome?

The sum of all the proteins synthesised in an organism, tissue, cell, sub-cellular organelle or simply the sample being studied

• Genomes fexed (largely) - Proteomes dynamic

What is Poteomics?

The qualitative and quantitative comparison of proteomes under different conditions to further unravel biological processes

How well do the genome, transcriptome and proteome correlate?

• Not very

• Protein expression levels vary hugely (not always correlated with significance)

• Anywhere between 30% and 80% of genes expressed at anyone time in a cell/tissue

Why does the mRNA sequence not equal Protein sequence?

Becasue of Post-Translational modifications (PTMs)

• can be permenant or temporary

• Protein Splicing (inteins)

• Additions/deletions (e.g. glycosylation

• Proteomics seeks to identify and quantify all protein components taking into account variations

What are 6 Proteomic analysis methods?

• 1D - Polyacrylamide Gel electrophoresis (PAGE)

• 2D PAGE

• Liquid Chromoatography

• Mass spectrometry

• Protein microassays

• High resoloution microscope and labled antibodies

What are the pros/cons of 1D PAGE?

• Quick

• Easy

• Poor resoloution

What is the difference between 1D and 2D PAGE?

• 1st dimension seperated by isoelectric point

• 2D page then seperated by mass (size)

What is Liquid Chromatography?

• Seperation of whole proteins or peptide fragments in solution

• Column tyoes used:

  • Anion/Cation

  • Reverse phase (RP)

  • Affinity(‘natural’ or tags)

• Leads to Fractionation of samples

What is the Basic principle of Mass Spectrometry?

Seperates samples on the basis of mass(m) : charge (z) ratio (m/z)

What does the ‘Source’ part of a Mass Spectrometer do, what are the two main types?

Produces ionized froms of sample in the gas phase

• MALDI

• ESI

What does the ‘Mass' analyzer’ part of a Mass Spectrometer do?

A divice which seperates out the ions by m/z ratio

What does the ‘Detector’ part of a Mass Spectrometer do?

A device to detect the different ions and generate a signal

How does MALDI generate Ionized samples?

• The disruption of protein or peptide samples (analyates) into the gas phase by mixing them within a light-absorbing matrix, which is dried onto a target plate

• When the laser hits the matrix, it rapidly vaporizes and transfers energy to the analyte molecules, ejecting them into the gas phase and ionizing them - typically by proton tranfer - so they can be analyxed by mass spectrometry

How does ESI generate ionized samples?

• Produces ions by passing a soloution of the analyte through a charged needle, creating a fine mist of highly charged droplets

• As the solvend evaporates, the droplets shrink until they can nolonger hold their charge, causing them to burst (Coulomb fission) and release gas-phase ions that enter the mass spectrometer

What are different devices that seperate out the different ion species(Mass Analysers) in Mass Spec?

• Time-of-flight (TOF)

• Quadrupole

• Ion trap

• Fourier-trasnform ion cyclotron resonance (FT-ICR)

• Ornitrap

What are some factors that may lead to misidentification or faliure when searching databases?

• Post-translational modifications that might mislead identification

• Protein degradation - a fragment created in handling with a chance match to a known protein

• Relative concentrations from single intensities can be misleading if a particular ion is suppressed

• Incomplete database - not all organisms have fully sequenced and annotated genomes

What is Peptide Mass Spectromeny?

• Where proteins are cleaved into smaller peptide fragments to aid mass identification

• Multiple different peptide matches gives confidence to protein assignment

• Fragmentation (Chemical or enzymatic, reasonably predictable formation)

• Match observed sizes against database - ‘peptide mass fingerprint’

What is MS/MS?

Use of two mass analyzer steps to seperate/select ion species linked by an additional fragmentation

• Isolated peptide from 1st MS (MS1) is intentionally fragmented (often by collisions with inert gas), breaking it into smaller, predictable pices

• The second mass spectrometer (MS2) measures the masses of these fragment ions, allowing the peptide sequnces to be deduced from the fragmentation pattern

What is Database Interrogation?

• A number of program suites used with outputs from MS analysis to interrogate sequence databases for 'hits’ or matches to predictable peptide fragments of all proteins in a given genome(s)

• Within the seach can set tolleracnes on sequnce match:

  • Define how sample prepared

  • Accept possible missed cleavag site

  • Permit inaccuracies in measured values

  • Allow for fixed and variable post-translational modifications

  • Include approaches that involve labeling of samples

How can Post Translational Modifications Complicate Mass Spec?

Some PTMs result in same m/z ratio so indistinguishable without ms/ms sequnecing

What are the two main methods of protein quantitation?

• Label Free - does not require modification of the sample

• Label Based - addition of tags or use of stable isotopes

What 4 things determine the protein abundance level?

• Transcription Rate

• Translation Rate

• RNA decay rate

• Protein decay rate

What are problems faced by MS with protein quantitation, what comparison can be made however?

• Response of different samples will vary based on their own unique properties

• Amount ‘a’ cannot be compared to amount ‘b’, however, amount ‘a’ at timepoint 1 can be compared to amount ‘a’ at timepoint 2

What factors are conisdered when deciding a Protein Quatitation Method?

• Proteome coverage

• Dynamic range of sample abundance being measured

• Quantitve accuracy required

• Number of samples being compared

How does Label-free Protein Quantitation Work?

• Proteins digested into peptides, then analyzed by MS

• Ion intensities are directly proportional to conc of correcponding protein in sample

• Methods:

  • Spectral counting

  • Intensity based methods

What are the pros and cons of Label-Free MS Quantitation?

• Pro

  • Good for High-throughput without need for for faff (labeling)

  • Cheaper as sysntheisising label standards can be expensive

  • Avoids potential biases from labeling techniques

• Cons

  • Ensuring peptide ion intensities are consistent and reproducable

How does Label-based Protien Quantitation work?

• Use of stable isotopic or chemical compounds to compare protein abundance levels by introducing a known mass difference

• In Vitro

  • SILAC-like chemical tags, TMT, iTRAQ

  • Label peptides after extraction and digenstion

• In Vivo

  • SILAC (most common)

  • Introduces heavy isotopes into proteins during cell growth

What are the pros and cons of Label-based MS Quantitation?

• Pros

  • High Accuracy and reproducability

• Cons

  • Additional cost

  • Specialized reagents (faff)

  • Experimental constraints (partucularly for in vivo which is limited to metabolically active systems)

What is ICPL?

Isotope-Coded Protein Labeling

• Labeling Level: Protein or Peptide (lyside residues)

• In vitro, MS1

• Light and heavy isotopic tags are chemically attached; mass differences ar measured in MS

• Pros: Early mixing reduces vairiability, compatible with many ample types

• Cons: Incomplete sampling can occur

What is ICAT

Isotope-Coded Affinity Tags

• Labaling level: Peptides containing cystein residues

• In vitro MS1

• Light/heavy tags bind cysteins and include an affinity handle for enrivhment

• Pros: Reduces sample complexity, improves detection of low-abundance proteins

• Cons: Only cystein-containing proteins quantified; limited proteome coverage

What is iTRAQ?

Isobaric Taga for Relative and Absolute Quantitations

• Labeling level: Peptides (N-termini and lysines)

• In vitro, MS2

• Isobaric tags release reporter ions upon fragmentation

• Pros: High multiplexing (up to 8-plex); samples combined before MS analysis

• Cons: Ratio compression due to co-isolation interferance; costly reagents

What is TMT?

Tamdem Mass Tags

• Labeling level: Peptides (N-termini and lysines)

• In viitro, MS2,MS3-based

• Isobaric tags generate reporter ions ; MS3 improves accuracy

• Pros: Very high multiplexing (up to 18-plex); high throughput

• Cons: Requires advanced instrumnentation; expensive

What is SILAC?

Stable Isotobe Labeling by Amino Acids in Cell Culture?

• Labeling level: Whole proteins (metabolic incorporation)

• In vivo, MS1

• Cells incorporate heavy amino acids during growth

• Pros: excelent quantitative accuracy; minimal sample handling bias

• Cons: Limited to cell culture or model organisms; low multiplexing (2-3 conditions)

What is Antibody Capture?

• Uses Antibodies immobilized on beads or solid supports

• Selectively isolate target protein from a complex mixture

• Then an enrichment step preformed aimed at reducing background from non-target proteins

• Then pull down assay preformed (Western blotting, ELISA or MS)

• Signal intensity correlates with protein amount

What is Epitope tagging?

• Involves tagging with short well-characterized peptide tag to a protein of intrest

• Epitope tags enable more reproducible measurements across different experiments

• Pull down assay then preformed (Western blotting, immunoprecipitation, or MS)

• Signal intensity correlates with protein amount

What is a protein MicroArray?

• Enable high throughput protein quantitation by immobilizing many different proteins on a solid surface

• For quantitation, the intensity of the detected signal at each spot reflects abundance

• Allows simultaneous measuremnt of hundreds to thousands of proteins

• Good for comparing between samples but accurate quntitation can be shaky

What is a Functional protein Micro Array?

• Requires expressed proteins from all ORFs

• Proteins are tagged and attached to a coated glass slide

• Used to investigate, protein-protein, protein-ligand, and protein-cell interactions

What does APEX stand for?

Engineered Ascorbate Peroxidase

What is APEX?

A genetically encoded enzyme used for proximity-dependant labling of proteins inside living cells. Rather than measuring global expression levels, allows identification of proteins within a specific sub cellular compartemnt

How is APEX targeted?

Genetically

• APEX is fused to a targeting sequence or protein that localsies to a specific compartement e.g.:

  • Mitochondria → Mitochondrial targeting sequence

  • Nucleus → Nuclear Localisation Signal (NLS)

  • ER Lumen → Signal peptide + retention sequence

  • Synapes → Fusion to synaptic proteins

How is APEX proximal?

• Biotin-phenol converted into a short -lived biotin radical

• The radical reacts with electron rich amino-acids (mainly tyrosines)

• Only proteins within ~20nm get labled

Reaction takes ~1min giving good temporal resoloution also

What two substrates are added in APEX?

• Biotin-phenol

• Hydrogen Peroxide (H2O2)

What happens in APEX after tagging?

• The cells are lysed

• Biotinylated proteins are captured usiing Streptavidin Beads

• Proteins are identified by LC/MS-MS

What is SRM?

Selective Reaction Monitoring

• An alternative shotgun proteomics that targets specific proteins/peptides

• Targets are determined to be useful as biomarkers or components of particullar cellular networks by proteome analysis

• The approach produces consistent and quantitative measurements that can be applied to detection of biomarkers for diseases such as various forms of cancers

What is a rough Work flow for SRM?

• Protein digesiton → tryptic peptides

• Targeted Peptide Selection, Choose peptides that are:

  • Unique to the protein

  • Chemically stable

  • Well behaved in MS

• Method development, Determine:

  • Optimal precursor m/z

  • Best Fragment ions

  • Colission energies

• Data Aquisition

  • MS monitors only the defined transitions

  • Signal intensities ~ peptide abundance

• Quantification

  • Often uses stable isotobe - labelled peptides as internal standards

  • Enables absoloute quantification

Why is SRM so specific?

3 layers of Specificity

• Precursor ion mass (Q1)

• Fragmentation pattern (Q2)

• Product ion mass (Q3)

Dramatically reduces background noise

What were the main findings from the targeted proteome leukemia prognosis study?

• Suggests a set of transcriptional and metabolic regulators that could be considered as predictors of prognisis and therapeutic targets in AML

• Therapy against the factors casusing kinase up regulation rather than their activity might be beneficial

• The study presents markers that could help predict AML relapse and directed therapeutic stratergies

What type of Mass Spectronemy did the 2014 Human Proteome studies use?

The samples of different tissues were fractionated, digested and analysed on the high-resoloution and high-accuracy Orbitrap mass analyser

What was found from the Establishment of the ProteonimicsDB?

• The Bulk of protein mass is contributed by only a small numbe rof genes, only 2350 ‘house keeping genes ‘ account for ~75% of proteome mass

• 10,000 - 12,000 Ubiqutious proteins - for general control and maintinence

What can be predicted from well from the RNA transcript level for a given translation rate?

Protein level

What can expression profiling predict?

• Compositions of large complexes

• Reveal unexpected distributions

What were the main findings from the 2015 Antibody and Transcriptome analysis of the Human Proteome study?

• Large numbers of genes with elevated expression in male tissue, brain, and liver, Relatively few in lung, pancrease ,and adipose tissue

• For most tissues, only 10% of the transcripts are encoded by genes with a specific tissue elevated expression

• More than 70% of the transcripts from the pancreas ~60% from the salivary gland and ~40% of the transcripts in liver encode secreted proteins

• Of the most abundant genes, the predictions of the localization of the corresponding proteins revealed that many (53%) are secreted proteins

What was found from the expresison of FDA approved drug targets?

• 30% of the approved drugs target proteins expressed in all analysed tissues

• This ubiquitous expression may have implications for treatments using these proteins as drug targets

What was found from analysing the Cancer Proteome?

• 525 genes implicated in malignant transformation

• Expression analysis based on transcriptomics data shows that a majority (60%) of these expressed in all tissues, with only a fraction of genes expressed in a tissue or group-enriched manner

• Lack of tissue specificity for many of these genes unsurprising because of involvement in normal growth regulation and cell cycle control

• However, it enmphasizes the possible adverse effects of treatment with drugs targeting proteins expression in all tissues

What was the difference between Tissue vs Cell Line protein expression?

• Many tissue-enriched genes identified in normal tissues are down regulated or completely ‘turned off’ in the corresponding cell lines

• In contrast, the house keeping proteins are expressed at the same level in both tissues and corresponding cell lines

• Thus cell lines are ‘de-differentiated’, with shared charecteristics and lack of tissue-specific features due to down-regulation of tissue enriched genes

• Implies that conclusions from cell line studies should only be confered on the corresponding tissue with caution

What happened in the 2017 Sub cellular map of the Human Proteome study?

• The subcellular locations of 12,003 proteins were determined by Immunoflorecence (IF) microscopy using ~14,000 antibodies in cell lines of various origins

  • Enabled precise definition of location

  • Revealed single cell variation in expression patterns

• High resolution IF images mapped proteins to distinct subcellular structures

  • Defined the proteomes of 13 major cellular organells

    • Revealed multi-localising protiens

    • Expression variability on a single-cell level

• Smaller organelles showed larger diversity than previously recognised

• Approx half of all proteins were localized to multiple compartments

  • Suggests a shared pool of proteins, even among functionally unrelated organelles

What two Organelles have the largest Proteome?

• Nucles (and its sub-structures) - 6245 proteins

• Cytosol - 4279 proteins

What was found from the 2023 Global detection of Human Variants and Isoforms by deep proteome sequencing Study?

• Attempted to find the sequence variants in the Human Proteome

• Typically proteins identified by 2 or 3 peptides - misses variants and isoforms in the proteome

• This study allowed a much higher (median 80%) coverage of the proteome

• Allowed allele specific expression studies and mutation effects on protein expression and stability

• Generated a set of ~25,000 peptides suggesting ~5,000 splice events can be detected in the proteome - this was previously undetected becasue of the low peptide coverage used for large scale proteomics studies

• Work suggests that de novo proteome assembly is possible, althoigh currently limited to high abundance proteomes

What is Structural proteomics/genomisc (SP/G) and why was it developed?

Structural proteomics/genomics aims to detemine representative protein structures across an organims or all species (the ’protein universe’). It emerged in the late 1990s in response to the Human Genome Project, using a large-scale ‘omics’ approach to provide structural biology resources for the wider bioscience community

How did the goals and methods of structural proteomics/genomics evolve?

Early estimates suggested ~16,000 carefully selected protein structures would cover ~90% of protein folds, but this proved insufficient as protein sequences grew exponentially. The focus shifted to biologically important protein families, requiring high-throughput pipelines for target selection, protein production, sample preperation, crystalization and structure determination using x-ray crystalography and NMR

What were the 5 main initiatives that determined ~95% of structures via SP/G?

• NIH Funding for Protein Structure Initiative 2000-2015

• Protein 3000 (Japan) 2002-2007

• Structural Genomics Consortium (UK/Canada) 2003-2011

• Enzyme Function Initiative 2010-1015

• National Institute for Allergy and Infectious disease 2008-2017

What are the Key Steps and Features of X-ray crystalography in protein structure determination?

• Proteins are purified to high concentration (~10mg/ml) and crystalised, then cryo-cooled crystals are used to collect X-ray diffraction data

• Phase information is obtained by molecular replacement or experimental methods, allowing calculation of 3D electron density maps

• Atomic models are built and refined, interpreted, and deposited in databases

• Structural proteomics relies on high-flux synchtron sources (e.g. Diamond) with focused beams, robotic handling, and advanced detectors for high-throughput data collection

What are the Key steps and features of NMR spectroscopy in Protein Structure Determination?

• Highly purified, concentrated proteins (~10 mg/ml; typically <30-35 kDA) are isotopically labled(15N,13C) and studied in soloution using high-field NMR spectrometers

• Nuclear spins are alinged and preturbed using multidimensional experiments, producing spects from which peaks are asigned to specific residues

• Distance constraints derived from these data are used to iteratively build and refine 3D structural models

• Structural genomics NMR shared production pipelines with X-ray methods, though automation was limited in reliability; the most successful effort was the RIKEN Protein 3000 initiative

What are the key steps in protein structure setermination by Cryo-EM?

• Moderately purified, low-concentration samples (<1mg/ml; typically large proteins or complexes) are applied to grids and stained or plunge frozen

• Grids are imaged with an electron beam, transmitted electrons are detected, and particle images are identified and classified by orientation

• Multiple views are averaged and combined to generate a 3D electron density map, into which atomic model are fitted and refined, often using high-resoloution domains from X-ray stuctures

What are the advantages and limitations of Cryo-EM, particularly for structural proteomics?

• Cryo-EM requires less concentrated protein and avoids crystalisation, but typically yields moderate resoloution and works best for large particles

• Historically iy was limited by detector technology, microscope stability, and image contrast, restriciting its use in structural proteomics

• Although automated particle selection and classification now improve throught, ample sample preperation on grids remains a major bottleneck

How is mass spectronemy used to obtain structural infromation on proteins?

• Beyond identification, MS provides structural insights via hydrogen-deuterium exchange (HDX-MS) to probe surface exposure and binding interfaces, cross-linking MS to genrate distance constraints for structural modelling, ion mobility MS to asses protein folding and overall chape, and direct MS of intact proteions (e.g. antibodies) to evaluate stabillity, interactions and specificity

How do AI-based methods predict proteion structures and why are they important?

• AI-based structure prediction used genome-derived protein sequences and multiple sequence alignements to identify homologous structures and residue covariation

• Models are generated by combining experimantal structure data with statistical covariance information and are scored for reliability

• Although not a direct structural proteomics techniwue, these methods rely heavily on data from experimental structure determination and have been greatly advanced by it