Determining the Amino Acid Sequence of a Protein
How do you find the amino acid sequence? To determine it, researchers isolate a protein, identify or predict the order of its amino acids, and verify the result using methods such as Edman degradation, mass spectrometry, NMR spectroscopy, or DNA sequencing. The chosen approach depends on the protein’s size, purity, availability, and whether the goal is to obtain its complete sequence or confirm only particular regions Most people skip this — try not to..
Understanding amino acid sequence is essential because a protein’s primary structure—the linear order of its amino acids—determines how the molecule folds, where it functions, and which other molecules it interacts with. A single altered residue can cause a protein to malfunction or trigger an immune response.
What Is an Amino Acid Sequence?
Proteins are chains of amino acids linked by peptide bonds. That's why each amino acid contains an amino group, a carboxyl group, a hydrogen atom, and a variable side chain. The side chain gives each amino acid its distinct chemical properties.
The sequence of a protein is normally written from the N-terminus, where the amino group is located, to the C-terminus, where the carboxyl group is located. For example:
Met–Lys–Trp–Ala–Ser–Leu–Val–Glu
What this tells us is methionine is the first residue, followed by lysine, tryptophan, alanine, serine, leucine, and valine, with glutamate at the final position.
The sequence of a protein is normally written from the N-terminus, where the amino group is located, to the C-terminus, where the carboxyl group is located. For example:
Met–Lys–Trp–Ala–Ser–Leu–Val–Glu
What this tells us is methionine is the first residue, followed by lysine, tryptophan, alanine, serine, leucine, and valine, with glutamate at the final position. Determining this order is more difficult than identifying which amino acids are present. A protein may contain several copies of the
Determining this order is more difficult than identifying which amino acids are present. A protein may contain several copies of the same residue, and the chemical properties of neighboring side chains can influence the reactivity of a specific peptide bond. Early methods relied on laborious chemical degradation, but modern workflows combine high-resolution mass spectrometry with bioinformatics to achieve rapid, high-confidence results Worth keeping that in mind..
Classical Chemical Sequencing: Edman Degradation
For decades, Edman degradation was the gold standard for de novo protein sequencing. Day to day, the reaction selectively cleaves the N-terminal amino acid under mildly acidic conditions, converting it to a phenylthiohydantoin (PTH) derivative that can be identified by chromatography. The shortened peptide remains intact, allowing the cycle to repeat.
While Edman sequencing provides unambiguous, direct evidence of the N-terminal sequence, it has significant limitations. It proceeds slowly (roughly 30–60 minutes per cycle), fails if the N-terminus is chemically blocked (e.On top of that, it requires a relatively pure protein sample (typically >10 pmol) and cannot easily handle complex mixtures. , by acetylation or pyroglutamate formation), and becomes unreliable beyond 30–50 residues due to accumulating incomplete reactions. g.Today, it is largely reserved for verifying the N-terminus of recombinant proteins or sequencing short peptides resistant to enzymatic digestion It's one of those things that adds up..
The Mass Spectrometry Revolution
Mass spectrometry (MS) has supplanted chemical degradation as the primary tool for protein sequencing. The standard "bottom-up" workflow involves:
- Proteolytic digestion: The protein is cleaved into predictable peptides using sequence-specific proteases (most commonly trypsin, which cuts C-terminal to lysine and arginine).
- Separation: Peptides are resolved by liquid chromatography (LC) to reduce sample complexity.
- Ionization and analysis: Eluting peptides are ionized (usually via electrospray ionization, ESI) and introduced into the mass spectrometer.
- Tandem MS (MS/MS): Selected precursor ions are fragmented—typically by collision-induced dissociation (CID), higher-energy collisional dissociation (HCD), or electron-transfer dissociation (ETD)—generating sequence-informative fragment ions (b- and y-ions).
The resulting fragmentation spectra are searched against protein databases (e.Even so, , UniProt) using algorithms such as SEQUEST, Mascot, or MaxQuant. g.This database-dependent identification is extremely sensitive (attomole level), high-throughput, and capable of analyzing complex mixtures like whole-cell lysates Nothing fancy..
De Novo Sequencing and Hybrid Approaches
When the organism’s genome is unsequenced, the protein is heavily modified, or a novel variant is suspected, database searching fails. This leads to De novo sequencing algorithms interpret MS/MS spectra directly to deduce the peptide sequence without a reference database. While challenging due to incomplete fragmentation and isobaric residues (Leu/Ile, Gln/Lys), modern tools (e.g., PEAKS, Novor, DeepNovo) leveraging deep learning have dramatically improved accuracy.
A powerful hybrid strategy, sequence tagging, extracts short, high-confidence "tags" (3–5 residues) from spectra de novo and uses them to query databases with relaxed constraints, bridging the gap between known and unknown sequences.
Top-Down and Middle-Down Proteomics
Bottom-up proteomics loses connectivity between distant modifications and struggles with large proteins. On top of that, Top-down proteomics analyzes intact proteins (often >50 kDa) directly, using high-resolution instruments (FT-ICR, Orbitrap) and fragmentation techniques like ETD or ultraviolet photodissociation (UVPD) that preserve labile post-translational modifications (PTMs). This provides a complete view of proteoforms—the exact molecular species arising from alternative splicing, PTMs, and sequence variants.
Middle-down proteomics offers a compromise, using proteases that generate large peptides (3–20 kDa), such as Asp-N or Glu-C, or limited digestion, balancing sequence coverage with the ability to localize combinatorial PTMs Simple as that..
The Genomic Shortcut: DNA Sequencing
In practice, the amino acid sequence of a known gene product is most often inferred from the DNA sequence. Still, sanger sequencing or next-generation sequencing (NGS) of the coding region (cDNA or genomic DNA) provides the primary translation product instantly and cheaply. This approach is standard for recombinant protein expression, antibody engineering, and variant screening.
That said, DNA sequencing cannot reveal:
- Post-translational modifications (phosphorylation, glycosylation, acetylation, etc., selenocysteine insertion, ribosomal frameshifting).
- RNA editing or recoding events (e.g.). Here's the thing — * Proteolytic processing (signal peptide cleavage, propeptide removal, activation cleavage). * Somatic mutations or splicing variants not captured in the reference genome.
Which means, mass spectrometric validation of the actual expressed protein product remains essential for functional characterization.
Specialized Challenges
Disulfide Bond Mapping
Cysteine connectivity dictates tertiary structure. Mapping disulfide bonds requires non-reducing digestion followed by MS/MS analysis of cross-linked peptides. Algorithms must distinguish intra- from inter-chain links, often aided by partial reduction
