BRCA1 on Chromosome 17

What linkage analysis actually does

The search for BRCA1 was not a hunt for a known target. In 1990, nobody knew what the gene encoded, what the protein looked like, or even whether a single inherited gene with strong effects on breast cancer risk existed at all. Most researchers in the field were skeptical that it would. The diseases clustered in some families, yes, but cancer is common, families share environments as well as genomes, and the statistical signal might be nothing more than coincidence dressed up as inheritance. Mary-Claire King, then at the University of California, Berkeley, thought otherwise — and she had a method.

Linkage analysis works by tracking DNA markers of known chromosomal position through families where a disease clusters across generations. If a marker and a disease phenotype travel together through a pedigree more often than chance allows, the gene responsible must lie somewhere near that marker on the chromosome. The method requires large, well-documented families, accurate phenotyping of who did and did not develop the disease, careful statistical testing, and the patience to accumulate enough families to reach significance. It requires nothing about the biology of the target gene. You are locating a region, not identifying a molecule.

King began collecting families in the early 1970s. The work was methodical and, for long stretches, unrewarding. Many families were too small to produce usable linkage data on their own; only by pooling many pedigrees did the signal emerge. Her laboratory developed the statistical framework alongside the data collection, building what amounted to a mapping enterprise before mapping was fashionable.

The 1990 result and what followed

At the annual meeting of the American Society of Human Genetics in October 1990, King presented the finding that would reframe hereditary breast cancer as a tractable molecular problem. Analysing data from twenty-three extended families with strong histories of early-onset breast cancer, her group showed statistically significant linkage to a region on the long arm of chromosome 17 — subsequently designated 17q21. The lod score, the standard measure of linkage strength, exceeded what the field required to declare a result credible. There was a gene there. Its sequence was unknown, its protein was unknown, but its approximate address was now on the map.

The skeptics who had doubted a single high-penetrance gene would be found had to revise their position. The implication was clear: in families with early-onset breast cancer clustering across multiple generations, a dominant inherited susceptibility gene at 17q21 was responsible for a substantial fraction of cases. Further analysis suggested the same region was implicated in hereditary ovarian cancer as well, which extended the stakes considerably. Linkage had done what it was designed to do.

What followed 1990 was a race of a different kind — from chromosomal region to gene sequence. Several laboratories, including King's, worked to narrow the interval using additional markers and additional families. The interval shrank, but isolating the gene itself required a level of sequencing and genomic infrastructure that was still being assembled across the field. It was a consortium led by Mark Skolnick at Myriad Genetics, working with the National Institutes of Health and several academic collaborators, that ultimately cloned and sequenced BRCA1 in 1994. The sequence revealed a large, complex gene encoding a protein involved in DNA damage repair and cell-cycle control — roles that explained in molecular terms why loss of the gene's function could initiate cancer. King's address had been correct.

What the gene's biology says

BRCA1 is a tumour suppressor gene. Under the two-hit model formulated by Alfred Knudson in 1971, individuals who inherit one defective copy of a tumour suppressor retain a functional second copy; cancer results when the second copy is lost or inactivated in a somatic cell. For BRCA1 carriers, that somatic second hit is all that stands between a cell and a functional loss of the protein's repair activity. The protein participates in homologous recombination, one of the cell's principal mechanisms for repairing double-strand breaks in DNA. When it is absent, the genome accumulates errors at a rate that drives malignant transformation.

Mutations in BRCA1 are not uniformly distributed across populations. Some communities carry specific founder mutations — variants that originated in a single ancestor and spread through a closed or partially closed population. The most studied example is in Ashkenazi Jewish populations, where three specific BRCA1 and BRCA2 variants occur at substantially elevated frequencies compared with the general population, according to the National Cancer Institute. This makes founder-mutation populations particularly informative for epidemiological studies and has shaped much of what is known about penetrance — the probability that a carrier will develop cancer over a lifetime.

Penetrance estimates have themselves been a matter of ongoing refinement. Early figures, derived from families ascertained precisely because they had extraordinary cancer clustering, were high and may have overstated the lifetime risk for carriers identified in less extreme family contexts. Later, population-based studies produced lower but still substantial estimates. The numbers remain the subject of careful study rather than settled consensus, and they depend in part on what other genetic and environmental factors a carrier also carries.

King's place in the history of the search

It is worth being precise about what King established in 1990 and what she did not. She located a region; she did not identify or sequence the gene. The sequencing credit belongs to the Skolnick-led consortium and its institutional partners. What King contributed was the demonstration, at a moment when informed opinion doubted the premise, that a single dominant locus with major effect existed and could be found by classical linkage in families. That contribution altered the trajectory of breast cancer genetics entirely.

Her 1990 presentation came after seventeen years of data collection — a period during which she was working against the grain of prevailing opinion. The argument that hereditary breast cancer was probably polygenic, or that environmental factors would overwhelm any genetic signal, was not unreasonable given what was known in the early 1970s. What King judged was that the question was worth pursuing with enough rigor to get a real answer. The answer was not what the skeptics expected.

The discovery also opened a debate that the genetics and epidemiology communities continue to work through: the relationship between high-penetrance rare variants like BRCA1 mutations, which explain a small fraction of all breast cancers but confer large individual risks, and the many common low-penetrance variants identified through genome-wide association studies, which individually do almost nothing but collectively account for a substantial share of the population-attributable risk. BRCA1 was the beginning of that story, not its conclusion. Chromosome 17 yielded a gene; the gene opened questions that a single sequence could not close.