1993: the fungus that was to make taxol

In April 1993 Science published a paper by Andrea Stierle, Gary Strobel and Donald Stierle on an endophyte isolated from the Pacific yew, Taxus brevifolia. They named it Taxomyces andreanae and reported that it produced taxol — an anticancer drug obtained until then from yew bark, in quantities that made production expensive and hard on the trees. The prospect of growing the drug in a fermenter rather than felling trees made the report one of the most cited papers on endophytes.

The problem lay at the foundation. The original description did not settle the fungus's taxonomic position, because the strain did not sporulate and showed no salient morphological features. The genus Taxomyces was erected for a single isolate whose affinities could not then be established.

In 2013 a German team sequenced the strain's genome in order to find the genes of the taxol biosynthetic pathway. They did not find them. That prompted Tian Cheng, Miroslav Kolařík, Luis Quijada and colleagues to check what the organism actually is. In a paper published in IMA Fungus in 2022 they compared five phylogenetic markers and placed the strain among wood-destroying Basidiomycota, in the phlebioid clade, closest to Ceriporiopsis gilvescens. Taxomyces was synonymised with Ceriporiopsis.

This story sits in MykoŚwiat to show how correction works. A striking report from thirty years ago was not retracted amid scandal — it was simply checked with a newer method and put right. It is worth remembering whenever a popular article mentions "a fungus that makes a drug": the question is whether anyone has found the genes it would need to make it.

A strong observation, later tested against a genome

In 1993 Taxomyces andreanae was isolated from the phloem, or inner bark, of Pacific yew (Taxus brevifolia). In semi-synthetic liquid culture, the team identified taxol by chromatography, mass spectrometry and taxol-specific monoclonal antibodies. Radiolabelled acetate and phenylalanine were incorporated into a labelled product; no taxol was detected in zero-time cultures or in the agar plugs used as inoculum. By the methods then available, this formed a coherent case for production in culture.

Later checks changed the interpretation. Genome mining of the reference culture CBS 279.92 did not recover the expected taxol-biosynthesis genes. A 2022 analysis of ITS, LSU, rpb1, rpb2 and tef1 placed the strain among wood-decaying Basidiomycota and made the new combination Ceriporiopsis andreanae (MycoBank MB845185). The authors judged fungal taxol production unconvincing, while explicitly noting how difficult it is to prove a negative claim.