I wish I could say that I’ve never cited a paper I haven’t read, but sadly, it wouldn’t be true. There are certain landmark papers that are just too easy to cite, following the lead of previous papers, without ever reading much more than the title. It is a mistake, of course. Papers often become landmarks not only for the data reported, but also for the unique, forward-thinking and insightful interpretation of those data. They can also be useful reminders of the power of traditional experimental approaches. This is the case for the seminal paper by Cox and Walter, reporting the identification of HAC1 as a key regulator of the unfolded protein response (UPR).

As a newcomer to the UPR, I cited Cox and Walter several times before I ever read beyond the abstract. When I finally did — feeling, guiltily, that I should probably know a bit more about the history of my new field — I was so impressed by its elegance and insight that I’ve never taken it for granted since.

the HAC1 mRNA was spliced upon UPR induction

By 1996, when the paper was published, the basic mechanism of the UPR had been laid out. This was a transcriptional response, activating a conserved promoter sequence to increase the expression of genes that included the protein chaperone BiP, and that required the kinase IRE1. The ‘missing piece’ was a transcription factor.

Cox and Walter searched for this transcription factor using a sensitive and elegant yeast genetics approach — a good reminder of the power of classical yeast genetics. They screened, in an ire1 deletion background, for genes that could activate HIS3 (a gene necessary for cell survival) under the control of four copies of the UPR-responsive element (UPRE). They identified three genes: IRE1, SWI4 (encoding a transcription factor that is part of the general transcription machinery) and HAC1. HAC1 encoded another putative transcription factor that proved to be necessary to activate the UPR. It could also rescue ire1 mutants, and bound directly and specifically to the UPRE, fulfilling all the criteria for a UPR transcription factor. But how was HAC1 regulated?

The HAC1 protein seemed to appear only following UPR induction, despite the constitutive presence of its mRNA. Surprisingly, though, a new, shorter HAC1 RNA species was detected upon endoplasmic reticulum stress. Using primer extension, then cloning and sequencing (a reminder of how laborious such things used to be!) they found that this RNA was missing an internal sequence, identical in all the clones they tested. This finding provided a key insight: the HAC1 mRNA was spliced upon UPR induction! Splicing created a stable protein that was capable of inducing the UPR.

Confusingly, however, the HAC1 RNA did not contain consensus splice sequences. Undiscouraged, the Walter laboratory pursued their splicing theory. By the following year, they had determined that IRE1 could splice HAC1 directly, in the cytosol, through its C-terminal tail domain.

This remains a unique activation mechanism, and re-reading these papers gives me a powerful sense of how exciting the discoveries must have felt at the time. For me, they represent a reminder of the continued relevance of classical genetics, the insight it takes to spot a novel mechanism and the value of knowing your history.