Cornell University

Mystery liverwort fungus, chapter 4

This is the Friday Afternoon Mycologist

New to the Mushroom Blog? You’ll want to catch up on our earlier FAM installments, because it’s becoming clear that something exciting is happening. Maybe something new is being discovered…

  • chapter 1, in which we meet the FAM and the fungus on his liverwort
  • chapter 2, in which the liverwort fungus is found to be vexing
  • chapter 3, in which we meet Molecule Man and a sheep

Big news! Molecule Man got lucky with the DNA!

As you recall, we took a pinch of dead liverwort tissue with structures of our fungus and sacrificed it to a DNA extraction kit. From there it was mixed with enzymes, shaken up, spun down, extracted, precipitated, suspended, dissolved, electrophoresed and subjected to the most wonderful of biochemical reactions, the polymerase chain reaction (PCR).

Molecule Man used PCR to amplify two adjacent bits of DNA, specifically the ‘small subunit of the nuclear ribosomal DNA’ (SSU, often also called the 18S) and its neighbor, the ‘internal transcribed spacer’ (ITS). He was successful in amplifying DNA from two of the three samples, and both the SSU and ITS were successfully amplified in at least some of the samples.

A diagram helps sometimes.  Good science always rhymes.

The bands on this agarose gel mean PCR successPCR is usually the tricky part, so we’ve done well to have PCR products that we can try to sequence. DNA sequencing involves more enzymes, shaking, spinning, precipitating, suspending, dissolving, and another mysterious biochemical reaction, and then the passage of the resulting solution through a long, thin column in an automated DNA sequencer. The resulting multicolored pattern of peaks and valleys represents the sequence of bases in the DNA fragment. After some editing with software, the DNA sequence emerges as a series of the letters A C T and G.

Output from the DNA sequencer: Peaks of different colours=different bases

Molecule Man was successful in getting two different SSU sequences, but the ITS sequences failed.

One of the great resources of modern biology is a database of DNA sequences called GenBank. On this website, you can cut and paste any DNA sequence and compare it with any DNA sequence that has ever been published. For identification of fungi (or anything else), this is a powerful tool, but the results have to be interpreted intelligently. This kind of analysis is called a BLAST search. Here is the result of the BLAST search with one of our liverwort fungus SSU sequences:

This is what the BLAST results looked like, only they were interactive.

The closest match for our sequence is another unidentified fungus! If you click to the record in the database (GenBank AY941257), you will see this unidentified fungus was a sterile culture isolated from an organ transplant patient, from an investigation not yet published. There are six bases that differ in our sequence and three gaps (i.e. ‘missing’ bases) from this human isolate. This is a lot for an SSU sequence to be considered the same species, but maybe suggestive of the same genus. But we don’t have a genus name here. The next match is 97% similar, with many more base differences, and the similarities fall off from there. This suggests that our fungus might be a member of the ‘taxonomic wonderland’ of the Pleosporales (see chapter 2), if this sequence actually represents the liverwort fungus and not some contaminant. It is hard to imagine two more different ‘hosts’ for apparently closely related fungi. Perhaps the most obvious question this finding suggests is to ask whether the transplant patient was in the habit of eating liverworts.

We also have a second SSU sequence. But it is from the wrong side of the fungal tracks. It is a basidiomycete, most similar to yet another unidentified fungus related to the jelly fungi, in this case a DNA sequence obtained directly from a specimen collected in the Antarctic (GenBank AY250847)! The second closest neighbor has a name, Fibulobasidium inconspicuum, a little yeasty thing first isolated from a tree branch in Louisiana (GenBank D64123). So our second sequence probably represents a yeast that was hiding among those blobby bacterial colonies on the isolation plate in chapter 3.

The two DNA sequences are at the end of this post, so you can try BLAST searches yourself. It will be interesting to see how these results change as more sequences are added to GenBank. These results are accurate now, at the start of 2007, but might be completely different by the end of the year!

So the DNA work is exciting, but unfortunately for now it does not help us identify the liverwort fungus. We have a strong hint that our fungus is related to the order Pleosporales, but the GenBank database doesn’t yet give us a solid species or even genus name. Based on morphology, it makes some sense that this fungus might belong to this order, but we cannot be sure that the DNA sequence we have belongs to the fungus we see. To be certain, we should repeat the DNA isolation and sequencing and see if we get the same answer.

In these four posts, we have gone once around the loop: found a fungus, looked at it, tried to culture it, dived into the literature, sequenced some DNA and we have come up empty. We could either give up… or we could describe this fungus now as a new genus and species. But I am not satisfied. We will go around the loop again… and my Friday afternoon diversion is invading the rest of the week.

Continued in Chapter 5


The world of DNA sequencing might seem exotic, but I was surprised to find a DNA analysis kit for sale for $40 in a local toy store while doing my Christmas shopping. You might want to check these out… a bit of googling leads to what appears to be the most elaborate, the Discovery DNA Explorer Kit, which allows you to send a few samples off to a commercial lab for the high tech part of the process. Who knows, it may not be long before some kind of DNA analysis or sequencing is as routine for the hobbyist as it is for the professional.

p.s. Thanks to Kent Loeffler for his patient help with the images in this post.

GenBank can be your taxonomic sandbox, too. You’re already right there with us on the cutting edge. Here are the sequences of the mystery fungus itself, with helpful instructions for DIY BLASTing.How to do a BLAST search:

  1. Go to the BLAST site.
  2. Under Basic Blast, click on ‘nucleotide blast’.
  3. Paste a DNA sequence into the ‘Enter Query Sequence’ box. For ‘Choose Search Set’ under ‘Database’, select ‘Nucleotide collection’ in the pick list. Click the BLAST button.
  4. Another screen will come up telling you how long the search will take. The results will eventually appear in this window with no more action from you.
  5. The results will be as shown in the blog text above, and will consist of a graphical view of the results, an inconspicuous line that you can click on to get a phylogenetic tree including your sequence, a table of results organized by the probabilistic similarity value known as ‘E’, and then text aligning your ‘query’ sequence with the closest relatives. Play around. You can’t hurt anything.

here you go:

The sequence of ‘Sample 2’, SSU of the liverwort fungus: AGCAATTATA

The SSU sequence of ‘Sample 3’, the basidiomycetous yeast contaminating the liverwort fungus: GCATGTCTAAGTATAAACAAATTCATACGGTGAA

Continued in Chapter 5



4 Responses to “ Mystery liverwort fungus, chapter 4 ”


Most people don't pay much attention to fungi, which include things like mushrooms, molds, yeasts, and mildews. Here at Cornell we think they're pretty fascinating. In fact, even the most disgusting foot diseases and moldy strawberries are dear to our hearts. We'd like to talk to you about fungi, so that like us, you too can tell gross stories at the dinner table. Afterwards, maybe you'll notice some things you would have overlooked before, and we think this could be good for the planet.

Kathie T. Hodge, Editor

Beneath Notice, our book of borescopic mycology.


Entries Comments

Or subscribe by email by entering your address: