Sanger Chain-Termination Sequencing: Reading DNA One Base at a Time

What It Is and Where It Happens

Sanger sequencing is an in vitro enzymatic reaction — it happens in a test tube, not inside a cell — that reconstructs the base order of a single-stranded DNA template. It exploits the normal chemistry of DNA replication but deliberately corrupts it. In a living cell, DNA polymerase extends a strand by forming a phosphodiester bond between the 3'-hydroxyl (3'-OH) of the growing chain and the incoming nucleotide's 5'-phosphate. Sanger's insight was to spike the reaction with 2',3'-dideoxynucleotides (ddNTPs), which lack the 3'-OH group entirely.

• Template: the DNA to be read (a plasmid insert, a PCR amplicon, or the original φX174 genome).
• Primer: a short oligonucleotide (~18–24 nt) that anneals adjacent to the region of interest and provides a free 3'-OH for the polymerase to start from.
• The reaction mix: DNA polymerase, buffer with Mg²⁺, all four dNTPs, and a small fraction of the four ddNTPs.

Because ddNTP incorporation is random and rare, each template molecule terminates at a different position, producing a population of fragments of every possible length.

The Mechanism, Step by Step

The power of the method comes from one chemical fact: a chain cannot grow past a dideoxynucleotide because there is no 3'-OH to attack the next phosphate.

1. Denature and anneal. The double-stranded template is heated to ~95 °C to separate strands, then cooled so the primer hybridizes to its complementary site.
2. Extend. DNA polymerase adds dNTPs one at a time, reading the template 3'→5' and synthesizing the new strand 5'→3'.
3. Terminate. At each position, there is a small probability (set by the dNTP:ddNTP ratio, often ~100:1) that a ddNTP is inserted instead of a dNTP. Once incorporated, elongation stops permanently at that base.
4. Build the ladder. Across billions of template copies, termination occurs at every position, generating nested fragments differing by exactly one nucleotide.
5. Separate and read. Fragments are resolved by electrophoresis; the shortest migrates farthest. Reading from small to large recovers the sequence 5'→3'.

In the classic four-lane protocol, each ddNTP (ddATP, ddCTP, ddGTP, ddTTP) was run in a separate reaction and lane.

Key Molecules and Characteristic Numbers

The molecular actors are precise and worth naming:

• ddNTPs: the terminators. A dideoxyribose is missing the oxygen at both the 2' and 3' carbons; the absent 3'-OH is what blocks further bond formation.
• DNA polymerase: early protocols used the Klenow fragment of E. coli Pol I; modern kits use a modified T7 DNA polymerase (Sequenase) or a thermostable enzyme in cycle sequencing, which mimics PCR with a single primer to linearly amplify signal.
• Fluorescent dyes: in dye-terminator chemistry each of the four ddNTPs carries a different fluorophore (e.g., emitting near 540, 570, 595, and 625 nm), collapsing the reaction into a single tube and lane.

Characteristic performance: a single high-quality read spans roughly 500–1,000 bp, with the first ~20 bases and the tail beyond ~700 bases typically noisier. Per-base accuracy in the mid-read window exceeds 99.99% (Phred Q40). A capillary run takes on the order of 1–3 hours for 96 samples in parallel on a modern genetic analyzer.

How the Ladder Is Observed

Reading the ladder requires separating fragments that differ by a single base out of hundreds. The medium is a denaturing polyacrylamide or a linear-polymer gel that keeps DNA single-stranded and sieves by length.

• Original method (1977–1990s): radioactive (³²P or ³⁵S) labels and slab-gel electrophoresis, with four lanes exposed to X-ray film. The sequence was read by eye, bottom to top.
• Automated method (1990s–present): Leroy Hood and Applied Biosystems introduced fluorescent dye-terminators and capillary array electrophoresis. Fragments migrate through a thin glass capillary filled with polymer; as each passes a laser-plus-detector window, its fluorophore is excited and the color identifies the terminating base.

Software converts the four-color signal into a chromatogram (electropherogram) — a series of colored peaks — and assigns a Phred quality score to each base, where Q20 = 1% error and Q30 = 0.1% error. Heterozygous positions appear as overlapping double peaks, which is why Sanger remains the gold standard for confirming single point mutations.

How It Compares to Related Methods

Sanger sequencing is one of two methods published in 1977; the other was Maxam–Gilbert chemical sequencing, which broke end-labeled DNA at specific bases using reagents like dimethyl sulfate and hydrazine. Maxam–Gilbert needed no polymerase or primer but used hazardous chemicals and gave shorter reads, so Sanger's cleaner enzymatic approach won out.

• Versus PCR: both use polymerase and primers, but PCR uses two primers to exponentially amplify a fragment, whereas cycle sequencing uses one primer and ddNTPs to linearly build a terminated ladder.
• Versus next-generation sequencing (NGS): Illumina's sequencing-by-synthesis uses reversible terminators that are imaged then chemically unblocked, letting the same strand be read base-by-base across billions of clusters in parallel. Sanger reads one long fragment at high accuracy; NGS reads many short fragments at massive scale.
• Versus long-read platforms (PacBio, Oxford Nanopore): these read tens of kilobases per molecule but historically at lower per-base accuracy than Sanger's mid-read Q40.

Significance, Applications, and Open Questions

Sanger's method powered the Human Genome Project, which produced its first draft in 2001 largely on automated dye-terminator capillary sequencers — a ~3.2-billion-base genome assembled from millions of overlapping ~700 bp reads. It remains indispensable where accuracy and read length matter more than throughput.

• Clinical validation: confirming a pathogenic variant (e.g., in BRCA1, CFTR, or TP53) flagged by NGS, since regulators often require an orthogonal method.
• Plasmid and clone verification: checking that a cloned insert or CRISPR edit has the intended sequence.
• Microbial identification and forensics: 16S rRNA gene sequencing for bacteria; short-tandem-repeat and mitochondrial analysis in forensic labs.

Its limits are structural: it reads only ~1 kb per reaction, struggles with GC-rich or homopolymer stretches, and cannot economically sequence a whole human genome (which would need millions of separate reactions). The open frontier is not Sanger itself but hybrid workflows — using cheap NGS for discovery and high-accuracy Sanger or long-read confirmation for the variants that matter clinically.

Sanger sequencing versus the earlier Maxam–Gilbert method and modern next-generation sequencing (NGS)

Feature	Sanger (chain-termination)	Maxam–Gilbert (chemical)	Illumina NGS (SBS)
Year introduced	1977	1977	2006
Basis	Enzymatic ddNTP termination	Chemical base cleavage	Sequencing-by-synthesis, reversible terminators
Read length	500–1,000 bp	~100–200 bp	50–300 bp
Throughput	1 read per capillary	1 read per gel lane	Billions of reads per run
Raw accuracy	>99.99% (Q40+)	~99%	~99.9% (Q30)
Typical use today	Validation, clonal plasmids, forensics	Obsolete	Whole genomes, RNA-seq, variant calling