Mass spectrometry identifies protein–protein interactions by converting an enrichment or covalent-capture experiment into LC-MS/MS identifications. The three mainstream routes are affinity purification–MS (AP-MS), crosslinking MS (XL-MS), and proximity labeling with BioID or TurboID. AP-MS reports proteins that remain associated with a bait through capture and washing. XL-MS reports residue pairs that were close enough to be covalently linked. BioID and TurboID report proteins that were biotinylated near a ligase-fused bait in living cells. None of these tables, by itself, proves a binary binding affinity or a diagnostic interaction.

Protein–protein interactions (PPIs) are central to cellular signal transduction, metabolic regulation, and chromatin remodeling. Identifying and characterizing interaction networks supports mechanistic cell biology, disease-mechanism work, and drug-target screening. Traditional yeast two-hybrid (Y2H) and co-immunoprecipitation (Co-IP) assays remain useful, but Y2H is biased toward interactions that can be reconstituted in the yeast nuclear environment, and Co-IP can carry non-specific background. MS-coupled immunoaffinity enrichment, chemical crosslinking, and proximity labeling add throughput, optional quantification, and, for XL-MS, site-level spatial constraints.

Researchers who already have a defined bait, cell system, or project objective can also review the MtoZ Biolabs Protein-Protein Interaction Analysis Service for project-specific feasibility and analysis planning.

What Mass Spectrometry Can Claim for a PPI Experiment

The first decision is the evidence type, not the instrument name. AP-MS asks which proteins co-purify with this bait after lysis under a stated wash stringency. XL-MS asks which residues were close enough for a chosen crosslinker, commonly DSS or BS3, to form a covalent pair. BioID and TurboID ask which proteins entered the labeling radius of a biotin-ligase fusion during a live-cell window. Those three questions are not interchangeable restatements of the same interactome.

A protein recovered with a bait after one purification is weaker evidence than reproducible enrichment against an empty-tag, bead-only, or unrelated-bait control. Enrichment against controls is still weaker than reciprocal Co-IP, domain mapping, or a pairwise binding assay. XL-MS site pairs constrain geometry. They do not reconstruct a complete complex structure by themselves. Proximity lists include neighbors that never form a stable complex.

Figure 1. AP-MS, XL-MS, and BioID or TurboID convert different physical events into LC-MS/MS identifications; the evidence type must be named before the interactome is interpreted.

An MS PPI result supports only the capture chemistry that was designed: co-purification, covalent contact, or live-cell proximity.

How Mass Spectrometry Complements Y2H and Co-IP

Y2H remains a pairwise screening tool when both partners can be expressed as fusions in yeast. Co-IP remains a native-complex check when a suitable antibody exists and the interaction survives lysis. MS does not retire those assays. It changes the readout from one or a few Western bands to a peptide-identified protein list, and it allows the same enrichment chemistry to be quantified across conditions.

MS-based PPI work can identify many co-enriched or proximal proteins in one acquisition after control filtering. That scale is design-dependent, not a guaranteed count of hundreds to thousands of true interactors. Specificity still depends on tag design, wash stringency, and matched controls. Coupling with SILAC or TMT can compare interaction occupancy across treatments, provided the label chemistry matches the enrichment route. XL-MS can add interaction-site and conformational constraints that Y2H and Co-IP do not provide.

Mainstream MS Strategies for Protein–Protein Interaction Analysis

1. Affinity purification–MS (AP-MS)

AP-MS is among the most commonly used MS routes for interaction studies. The protein of interest is fused to a tagged expression construct such as Flag, HA, or Strep-tag, or it is captured with a validated antibody when endogenous pull-down is the design. Cells are lysed, the bait is enriched by immunoaffinity, captured complexes are washed and eluted, and LC-MS/MS identifies co-purified proteins. Stable cell lines and transient expression systems are both used; tag placement at the N or C terminus can change bait folding, localization, and the recovered partner set.

AP-MS is experimentally direct and can be combined with SILAC or TMT to compare interaction strength or occupancy across conditions. It is a poor route for many transient or weak-affinity interactions, because those partners are lost during wash. Tag fusion can also alter structure or function. Empty-tag, bead-only, or unrelated-bait controls belong in the same record as the prey table.

MtoZ Biolabs uses a low-background magnetic-bead enrichment system with a matched lysis and elution workflow on an Orbitrap-class LC-MS/MS platform. Low-abundance recovery still depends on bait expression, wash stringency, and control filtering, not on the platform name alone.

Figure 2. AP-MS identifies proteins that survive tagged-bait capture and washing; empty-tag or bead-only controls and optional SILAC or TMT comparison belong in the same design.

For evidence levels, control logic, and over-read risks in AP-MS, researchers can refer to Understanding AP-MS: Evidence, Limits, and Interpretation for more detailed guidance.

2. Crosslinking MS (XL-MS)

XL-MS uses chemical crosslinkers to lock spatially proximal amino-acid residues. After proteolytic digestion, enrichment of crosslinked peptides, and LC-MS/MS, the method can identify interacting proteins and assign residue-level contacts. Common reagents include DSS and BS3. The same chemistry can be applied to purified complexes, cells, or tissue when accessibility and quenching are controlled.

XL-MS provides interaction-site information and conformational constraints that AP-MS partner lists do not contain. Downstream analysis is more demanding: mixed spectra, low crosslinking efficiency, and the need for high-resolution MS plus dedicated search algorithms all limit how many confident site pairs are reported. A crosslink between two proteins supports proximity at the time of reaction. It does not, by itself, rank binding affinity or prove a complete architecture.

MtoZ Biolabs integrates DSS and BS3 with a dedicated enrichment workflow for crosslinked peptides and analysis algorithms built for site-level assignment. Identification rate and site accuracy still depend on reagent choice, input amount, and the declared false-discovery rule.

3. Tag-assisted proximity labeling (BioID / TurboID)

BioID and TurboID fuse a mutant biotin ligase to the bait. In living cells, the ligase biotinylates proteins within a short radius. Streptavidin affinity purification followed by MS then identifies labeled proteins. TurboID uses a faster ligase variant and can support shorter labeling windows than classical BioID, which matters when the interaction is time-restricted. Labeling occurs before lysis, so the method is less dependent on the stability of a purified complex.

Proximity labeling is well suited to weak or transient associations and to in situ cellular contexts, including some membrane-proximal neighborhoods. Background biotinylation can be high if expression, biotin dose, or labeling time is not optimized. The list is a neighborhood, not a wash-stable complex. Hard-to-transfect cells and primary lines usually need a stable expression route rather than a single transient pulse.

MtoZ Biolabs has established a TurboID lentiviral platform for generating stable cell lines, supported by biotin enrichment and deep LC-MS/MS detection. Expression level and labeling duration still require project-specific optimization.

How to Choose and Combine Strategies

Research need	MS route that often fits	Reassess before queuing the experiment
Wash-stable complex around a tagged bait	AP-MS with matched empty-tag or bead-only controls	Transient or weak interactions are the main target
Weak, transient, or in situ neighborhood	BioID or TurboID in living cells	A purified, wash-resistant complex is required
Residue-level contacts or conformational constraints	XL-MS with DSS or BS3	Only a partner list is needed
Condition-dependent occupancy	AP-MS or proximity labeling plus SILAC or TMT	Absolute Kd or a diagnostic cutoff
Pairwise binary screen in yeast	Y2H, not an MS interactome	Mammalian in situ context is required
Native complex check with an existing antibody	Co-IP, optionally followed by MS	Tag-fusion AP-MS is the only planned bait

Prefer AP-MS when the claim is co-purification of a stable or moderately stable complex and controls can be matched. Prefer BioID or TurboID when the interaction is weak, transient, or must be labeled before lysis. Prefer XL-MS when the next sentence needs a contact site or a spatial constraint. Prefer Y2H or Co-IP when the question is pairwise reconstitution or a native antibody check, not an open LC-MS/MS inventory.

Combining strategies is often more defensible than expanding one list. TurboID can screen candidate proximal proteins, AP-MS can test which candidates survive affinity capture, and XL-MS can assign sites on a smaller set of validated partners. Orthogonal Western or reciprocal Co-IP checks still belong on priority candidates.

Figure 3. Choose the MS route from the evidence needed, then combine TurboID, AP-MS, and XL-MS when one capture chemistry cannot defend the claim.

When AP-MS versus BioID or TurboID is the main fork, researchers can refer to AP-MS vs BioID and TurboID: How to Choose for a more detailed strategy-selection framework.

Experimental Limits and Design Boundaries

MS PPI tables do not automatically report direct physical binding. Shared contaminants, abundant cytoskeletal proteins, and recurrent bead binders can survive even a careful wash if controls are weak. Overexpression can create non-physiological partners. Incomplete nuclear reconstitution in Y2H and antibody-dependent background in Co-IP remain relevant when those assays are used as orthogonal checks rather than as replacements.

XL-MS coverage is sparse relative to the full interface. BioID and TurboID labeling radii can include proteins that never touch the bait. SILAC and TMT compare relative occupancy under one design. They do not convert an interactome into a Kd. Diagnostic cutoffs and companion-diagnostic claims sit outside ordinary research PPI-MS reporting unless a separately validated assay framework is in place.

Project Planning and Service Scope

A usable PPI-MS plan names the bait, the evidence type, the cell system, the controls, and whether later orthogonal validation is required. Vector construction, tag-expression checks, enrichment chemistry, Orbitrap Exploris or Q Exactive Plus-class LC-MS/MS acquisition, and network-level bioinformatics are then scoped to that plan. Quantitative, crosslinking, and proximity-labeling routes can be combined, but they should not be mixed into one statistical table without a declared bridge.

Once the bait, evidence type, and control set are defined, the project can be scoped as an AP-MS, XL-MS, or BioID/TurboID study. Researchers can review the MtoZ Biolabs Protein-Protein Interaction Analysis Service for project-specific sample evaluation, feasibility assessment, and workflow planning.

Frequently Asked Questions

1. Can mass spectrometry replace Y2H or Co-IP for every PPI question?

No. MS expands the identified partner set and can add quantification or site information. Y2H and Co-IP remain appropriate when the claim is pairwise reconstitution or a native antibody check.

2. Does an AP-MS prey list prove direct binding?

No. AP-MS reports co-purification after capture and washing. Indirect complex members and residual contaminants can appear until controls and orthogonal assays are applied.

3. When should TurboID be chosen over AP-MS?

Choose TurboID when the interaction is weak, transient, or must be labeled in living cells before lysis. Choose AP-MS when a wash-stable complex and control-filtered co-purification are the intended evidence.

4. What does XL-MS add after a partner list already exists?

XL-MS can assign residue-level contacts and conformational constraints. It does not replace control-filtered discovery, and the analysis is more demanding than a standard AP-MS search.

5. Can SILAC or TMT turn a PPI screen into an affinity constant?

No. Those labels support relative comparison of the same bait-associated proteins across conditions. A Kd requires a dedicated binding assay.

Reference

1. Gingras A.C., Gstaiger M., Raught B., Aebersold R. (2007). Analysis of protein complexes using mass spectrometry. Nat. Rev. Mol. Cell Biol., 8, 645-654. https://doi.org/10.1038/nrm2208
2. Roux K.J., Kim D.I., Raida M., Burke B. (2012). A promiscuous biotin ligase fusion protein identifies proximal and interacting proteins in mammalian cells. J. Cell Biol., 196, 801-810. https://doi.org/10.1083/jcb.201112098
3. Branon T.C., Bosch J.A., Sanchez A.D., et al. (2018). Efficient proximity labeling in living cells and organisms with TurboID. Nat. Biotechnol., 36, 880-887. https://doi.org/10.1038/nbt.4201
4. Leitner A., Walzthoeni T., Kahraman A., et al. (2010). Probing native protein structures by chemical cross-linking, mass spectrometry, and bioinformatics. Mol. Cell. Proteomics, 9, 1634-1649. https://doi.org/10.1074/mcp.R000089-MCP201