Protein networks identification and analysis

Question 1

Predicting protein sequences limitations

Answer 1

•Reading frames, splice variants and annotated function must be correct.
•Predicting localization depends on recognisable signal peptides or motifs. In the cell many determinants of membrane association are transient.
•Post translational factors increase complexity
-These factors include localisation, modifications, interactions, abundance and turnover at the protein level.

Question 2

Alternative splicing

Answer 2

another source of variation
“A homologue of the Downs Syndrome Cell Adhesion Molecule (Dscam) protein in Drosophila melanogaster has 38,016 isoforms
Note the entire Drosophila melanogaster genome only has 15,016 genes!”

Question 3

changes of location

Answer 3

-During production/breakdown
-During Function
We cannot predict all the complexity and information content of the functioning cell from genomics/transcriptomics alone (or by any single technique).

Question 4

Proteomics

Answer 4

the study of the ‘entire’ protein content of an organism, tissue or cell.
Proteomics seeks to provide information on the identity, amount, modifications, and subcellular location of proteins.

Question 5

Challenges

Answer 5

•	Proteins are dynamic in their location, abundance, splicing, modifications and interactions.
–	Often cannot distinguish splice variants or closely related proteins using peptide-based methods
•	Huge range of abundance. Low-abundance proteins may be the most important class of proteins but are the most difficult to study with current methodologies.
•	Rapid temporal changes. Proteins can interact, move and be modified very quickly (seconds to minutes)
•	Tools: Proteins cannot be studied with the scale, speed, sensitivity and reliability that is currently achievable with nucleic acids. We cannot yet do ‘protein PCR!’: no amplification. Possible new methods with RNA aptamers. An aptamer is a sequence of single strand nucleic acid with a variable region of around 40 bases.

Question 6

How can proteins be measured?

Answer 6

• Immunological approaches
• Biochemical approaches
• Mass spectrometry-based methods
• Fluorescent or other imaging methods
• Emerging technologies

Question 7

Immunological approaches

Answer 7

-Use antibodies to detect (i.e. localise) or to physically isolate protein
e.g. immunohistology ’Human Atlas’ project.
– Enzyme Linked Immunosorbent Assay (ELISA)
– Western blotting

Question 8

immunohistology ’Human Atlas’ project.

Answer 8

importance in human histology- the human protein atlas. Measurements of protein location. Changes of location indicates disease. 11,274 antibodies corresponding to 8,489 protein-coding genes.

Question 9

Human atlas examples

Answer 9

CD44
a cell-surface glycoprotein involved in cell-cell interactions, cell adhesion and migration.
Alternative splicing is the basis for the structural and functional diversity of this protein and may be related to tumour metastasis. However, the full-length nature of some of these variants has not been determined. CD44 has 30 splice variants in humans.
DSCAM
a member of the immunoglobulin superfamily of cell adhesion molecules involved in human central and peripheral nervous system development. This gene is a candidate for Down syndrome and congenital heart disease (DSCHD). DSCAM is very highly spliced in Drosophila but has only 2 isoforms in humans

Question 10

Limitations of immunological approaches

Answer 10

•need one antibody per protein = laborious and expensive.
• The binding specificity depends on the epitope (binding sites)
– could be non-specific
– inhibited by modifications/interactions/(un)folding
– not all antibodies are suitable for all tasks

Question 11

uses of antibodies in biochemistry

Answer 11

–Detection: histology (confocal and electron microscopy), enzyme-linked immunosorbent assay (ELISA), Western blotting.
–Isolation of protein and interacting partners by co- immunoprecipitation

Question 12

Overcoming the limitation of a specific antibody for each protein

Answer 12

•Introduce a well characterized ‘TAG’ to the protein of interest using genetic modification. This relieves one limitation and some of the expense
Tags include:
•Small proteins: GST,GFP
•Peptides: FLAG.Myc,6xHis
•Tandem affinity tags (TAP)
-larance paper: specific ab or tagged protein immunoprecipitation advantage- high sensitivity can be achieved for proteins in low abundance.

Question 13

Fusion proteins

Answer 13

• These fusion proteins help to visualise where a protein is located (with various limitations depending on the tag and organism or cell type)
• Fusion can be used to enrich the modified (tagged) protein and interacting partners using co-immunoprecipitation or by fractionation of cell extracts and testing for presence of tag.

Question 14

Biochemical approaches

Answer 14

Use the physical properties of different proteins as purification tools: size, charge, affinity binding, hydrophobicity.
- You can find out where your protein is during the purification:
• by following protein’s activity (if known/traceable)
• by following presence of protein (fluorescence, epitope tag) Epitope tagging is a technique in which a known epitope is fused to a recombinant protein using genetic engineering. Epitope tags make it possible to detect proteins when no antibody is available.

Question 15

Emerging tech

Answer 15

RNA aptamers and nanopores

Question 16

Detecting proteins in complicated mixtures

Answer 16

• Polyacrylamide gels (denaturing, SDS-PAGE) (native no SDS)
• Western blotting (always denaturing)
• Enzyme Linked Immunosorbent Assay (ELISA)
• Fuse fluorescent tag
• Mass spectrometry to identify proteins

Question 17

Mass spec

Answer 17

4 stages: sample, ionisation source, flight path, detector
Proteins can be prepared for MS from gel slices or from solution digests e.g total cellular proteomics. Then we digest the thousands of proteins into 20-50 peptides from each protein

Question 18

LC/MS-MS

Answer 18

1. protein
2. tyrptic digest
3. nanoflow LC of peptides
4. MS to select intact peptide and fragment peptide
5. peptide mass, retention time, fragmentation, intensity
6. assign peptides through database search (larance reading database for raw ms available)

Question 19

LC

Answer 19

To deal with very complicated mixtures of peptides. LC separate a mixture of molecules by a physical property.
1.Temporarily bind the mixture of peptides to the column
2. Wash peptides off by applying solvent gradient
3. Peptides elute off column on basis of their hydrophobicity.
The beads are covered with long carbon chains (C18). The amino acid sequence of a peptide determines its hydrophobicity (but hydrophobicity is not enough to identify a peptide).
• Liquid chromatography (LC) gives a retention time for the peptide.

Question 20

Concept C18(reverse phase) chromatography

Answer 20

Organic modifiers such as acetonitrile is used to elute the hydrophobic peptides. A gradient of the organic solvent will release the peptides in the order of their hydrophobic interaction strengths.

Question 21

Time of Flight MS

Answer 21

Ions mass to charge ratio (m/z) is determined. as ions traverse the analyser, they separate in space. A detector is positioned at the end of the analyser to measure the arrival time of ions. Ions of lesser m/z arrive first, followed by ions of greater m/z.

Question 22

Quad MS

Answer 22

The quadrupole scans a range of frequencies that lets ions of a specific mass to charge ratio reach the detector. all the other ions go to waste

Question 23

Ion trap- collision induced dissociation

Answer 23

MS- scans ions out over full mass range
CID-selected ion is retained, excited to fragment
MS2-fragment ions ejected

Question 24

Fragment to sequence

Answer 24

Can use predicted protein sequences where you match experimental fragment mass spectra to predicted fragment masses from genome. -confidence score.

Question 25

Mass spec proteomics main stages

Answer 25

• Sample preparation with separations(LC)/fractionation(SDS page)
• Ionisation (ESI,MALDI)
• Measure intact peptides and fragment ions (MS/MS)
• Data analysis and protein identification

Question 26

MS quantitative-fragments and peptides

Answer 26

– Can make standards using stable isotopes of nitrogen (N15) and carbon (C14) to create mass difference without altering biochemical properties of the peptides.
– quantification at precursor ion level (e.g. stable isotope labelling of amino acids in cell culture SILAC)
– Or quantification at fragment ion level (e.g selected reaction monitoring SRM)

Question 27

Protein identities are inferred from peptides.

Answer 27

It is important to have the correct sequence in the database.
Protein inference problem-because peptide sequences are inferred from MS, conserved peptides are a problem
Western blotting may be necessary to resolve unique protein identities

Question 28

MS not enough

Answer 28

SERK problem- one experiment tried to find what interacts with FLS2 used GFP coIP to find interactors found that a SERK protein interacts with FLS2. But not sure which one. In second experiment each of the 5 SERKs were expressed with 2 tagged proteins, immunoprecipiated then blotted to detect the other

Question 29

Current status of proteomics

Answer 29

– Integration of genetic methods, imaging, mass spectrometry and biochemical approaches.
– Identification of proteins by MS widely ‘routine’
• Total numbers comparable to transcriptomics
• Identification of protein from gel bands, affinity purification, cell lysates
– Quantification more common and reliable; not quite routine
• spectrum counting, peak areas, selected reaction monitoring
• Stable isotopes can be used to greatly improve accuracy (depends on cell types)

Question 30

Generating and navigating proteome maps using Mass Spec by Ahrens

Answer 30

• there has been improvement in proteome coverage
• ms used to generate partial proteome maps
• 2 complimentary approaches taken
• 1 systemic fractionation
• ms data generated from a particular species
• difficulty in knowing completeness
• assess completeness if protein product for all annotated ORFs located in the genome of the species is conclusively identified
• tight control over protein false discovery needed

Question 31

Protein turnover

Answer 31

The length of time that a protein remains available to perform its function is significantly influenced by its turnover rate.
Protein turnover is the balance between protein synthesising and protein degradation

Question 32

Isotopes

Answer 32

Pulse chase labelling used to be measured with radionucleotides 14C,35S,32P limited sensitivity of scintillation counter or photographic film
Now we tend to use stable isotopes instead Or fluorescent labels

Question 33

Stable C13

Answer 33

We use naturally occurring stable C13 isotopes to determine charge state

Question 34

Stable 15N

Answer 34

Unlabelled light peptide = all 14N

Fully labelled “heavy peptide” = All 15N

Question 35

Stable isotope labelling of amino acids in cell culture (SILAC)

Answer 35

Separately metabolically label haploid or diploid yeast with stable isotopes in cell culture (SILAC).
After treatment mix light and heavy cultures. Proteins and peptides are chemically identical, so they stay together throughout the fractionation and identification steps in LC- Ms/MS.
But their mass (m/z) differs and the peak area of each is quantitative

Question 36

Yeast protein turnover (degradation rate) by Claydon & Beynon

Answer 36

made estimations in exponentially growing cells

Question 37

Global analysis of protein localisation in yeasts Huh et al.

Answer 37

Systematically GFP-tag each (6234) yeast ORF in its chromosomal location
– by homologous recombination
– GFP added at C-terminus
Transform yeast → 6,029 strains with chromosomally GFP-tagged ORFs
Analyse yeast with fluorescence microscope → 4,156 (i.e. 75% of yeast proteome) showed GFP fluorescence
Strains initially classified into 12 localisation categories by eye. Refined by co-localising with known markers
Good correlation between localisation and transcriptional co-regulation
Good correlation between protein localisation and protein-protein interaction data
Chromosomal integration (of GFP-fusion proteins) solves the problems related to overexpression artefacts
• The resulting database (http://yeastgfp.yeastgenome.org/ ) is available

Question 38

usual problem

Answer 38

The C-terminal fusion of GFP causes mis-localisation of proteins in a number of compartments

Question 39

organelle proteomics

Answer 39

by density fractionation
• Physical separation of organelles by their density
• Identification of proteins by mass spectrometry
• Label-free quantitation of peptides by mass spectrometry.(applies two quantification strategies: (1) spectral counting or (2) spectrometric signal intensity to measure the protein expression. )
• Correlation of protein abundance profiles with known proteins established as markers for an organelle.

Question 40

protein correlation profiles

Answer 40

Protein correlation profiles translate directly into fluorescent staining patterns
Function and predictive results from Foster
• Mapped 1404 proteins to 10 subcellular locations
–39% of organelle proteins had multiple locations
• Comparison of localisation with Huh 2003 results 74% agreement between yeast and mouse cell lines
• Identified assumed trans-acting transcription factors and cis- promoter motifs of organelle related genes.

Question 41

Examples of databases and resources for protein localisation:

Answer 41

UNIPROT, DAVID, illuminated plant cell

Question 42

Methods to identify protein-protein interactions

Answer 42

• In vivo imaging methods: FRET, BiFC, FRAP
• Yeast 2-hybrid (Y2H) and derivatives
• Phage display
• Protein chips/peptide arrays/substrate arrays
• Affinity methods
– Co-sedimentation, size exclusion filtration
– Affinity chromatography
– Co-immunoprecipitation
– Proximity labelling
– Interactions can be stabilized by cross-linking

Question 43

membrane yeast two hybrid(MYTH)

Answer 43

does not require that interactions occur in the nucleus, thereby allowing the use of full length membrane proteins, in their natural context of the cellular membrane
In MYTH, a membrane bait protein is fused, at a cytosolic terminus, to a tag consisting of Cub linked to an artificial transcription factor.
However, a major drawback is that some interactions of proteins from other organisms may not occur in the yeast milieu because of the possible lack of associating factors or protein modifications

Question 44

LUMIER

Answer 44

LUMIER provides a powerful method to study mammalian PPIs in their natural environment in an automated

Question 45

mammalian protein-protein interaction trap

Answer 45

PPIs take place in the cytosol of intact mammalian cells.

a novel screening method for anti-HIV compounds in intact human cells.

Question 46

Phage display

Answer 46

identify and optimize polypeptides with novel functions
Phage libraries, which can consist of artificial or naturally existing peptides, proteins and protein variants presented on phages
particularly suited for rapid screening and optimization of molecular interactions, making it of special interest in drug development.