Smallest Things Considered

by Merry

Oddly, coconut palms have
only one known bacterial
disease, one possible viral
disease, but at least two
diseases caused by viroids.
The tree in the foreground
is infected with cadang-
cadang, a viroid disease,
and shows symptoms of yel-
lowing, stunting, and lack
of fruit production. Source.

What do stunted coconut palms, misshapen potato tubers, and peach trees with necrotic branches have in common? They are three of the numerous crops stricken with diseases caused by viroids, an astonishing group of minimalist plant pathogens. There isn't much to a viroid, just one single-stranded, circular RNA molecule. The largest viroid genome so far is 399 nucleotides, the smallest a mere 246—about one tenth the size of the smallest viruses (hepadnavirus) and one hundredth the size of more typical viruses. Being labeled as "subviral," they are even less likely than the viruses to be granted a place on the tree of life. They get by without capsid or membrane shell. They encode no proteins. They don't reverse-transcribe into DNA when they replicate. They never insert into the host genome. Some of them cause disease symptoms, some don't. They simply replicate inside plant cells and then their progeny move on to the next location to repeat the process. Their very existence raises questions, many without answers.

What does a viroid look like? Well, that depends on the viroid. The thirty-plus species known so far fall into two groups. Most belong to the Pospiviroidae (PSTVd), named after the Potato Spindle Tuber ViroiD. Four species, including the Avocado SunBlotch ViroiD, make up their own group, the Avsunviroidae (ASBVd). Although all of them lack protective capsids, they are nevertheless highly structured.

Most of their RNA genome base-pairs with complementary sequences in other parts of the same RNA molecule to form sections of double-helix. The PSTVds form rod-shaped structures, the ASBVds have more elaborate forms. (See figure.) For both groups, the bulges and loops are important for survival, replication, and trafficking. (For a recent review, click here.)

Predicted lowest free energy secondary structures of (a) the potato spindle tuber viroid, a Pospiviroidae, and (b) peach latent mosaic viroid, an Avsunviroidae. Source.

How do they replicate when they encode no proteins? Viroids rely on host cell factors to help with just about everything. Once the viroid RNA enters a cell, it recruits factors for "RNA trafficking" to the compartment where it replicates; that's the nucleus for the PSTVds, the chloroplast for the ASBVds. Then, for copying their RNA, both groups co-opt host DNA-dependent RNA polymerases, redirecting them to use the viroid RNA as template. This works, but is error-prone, giving one of the viroids the fastest known rate of mutation.

Viroids under the electron microscope.
Source.

Both types replicate using a rolling circle mechanism. Although the details differ, they both produce long, multimeric strands of RNA that must be cut into monomers, each monomer then ligated to form an individual circular viroid. For the cutting ASBVds employ a do-it-yourself strategy. They dedicate 40 nucleotides of their sequence to form the smallest known naturally-occurring ribozyme. This hammerhead ribozyme catalyzes the self-cleavage needed to transform the long multimeric strands into individual genomes. After replication, the progeny RNAs use host factors to help them move to neighboring cells through the connecting plasmodesmata or to distant parts of the plant via the phloem.

Next question. How do they cause disease if they don't encode any proteins? Directly, by RNA interference (RNAi). RNAi is a mechanism that degrades double-stranded RNAs (dsRNA). In plants, this defends against RNA viruses by attacking their double-stranded replication intermediates. Not only does this strategy not work against viroids, but it actually causes the symptoms associated with viroid infection. How does this happen? Short lengths of RNA clipped from the viroid genome can base pair with complementary host mRNA to form dsRNAs that are then degraded. Destroying the mRNAs effectively turns off the corresponding genes in the host. You might wonder why host-derived RNAi doesn't degrade the double-stranded viroids. This is likely due to the viroid's complex secondary structure and to the fact that all of their double-stranded regions are too short to activate RNAi.

Where did viroids come from? Viroids show no relationship to viruses, but there is some evidence—all seemingly a bit of a stretch—suggesting evolutionary links to transposable elements, or maybe to mitochondrial plasmids, or even to self-splicing introns. Take your pick. All viroids form a monophyletic group along with the satellite RNAs also found in plants. Are they ancient? Those who picture early life on the planet as based on RNA, not DNA, would like to think so.

If viroids are ancient, or even just a century old, why the sudden appearance of viroid diseases in our crops in recent decades? At least some viroids were accidentally introduced by humans from wild relatives where often their infections are asymptomatic. We become aware of them in our crops when they cause disease. Once introduced, we have insured that they spread. They used to have to rely primarily on direct transmission to the next generation via seed or pollen. Some were occasionally carried from one host to the next by chewing insects. Nowadays, our practice of mono-cropping makes inadvertent transmission by insects far more likely. Better yet, most often we transport the viroids ourselves, introducing them directly into their next host as we prune or graft.

Why are there no viroids that infect animals? Maybe even bacteria? Not a clue.

Ancient or not, viroids are today expanding their horizons through global commerce, meeting new hosts through modern agriculture. Life is good for viroids today—if indeed it would occur to anyone to describe viroids as alive.

Ding B (2009). The biology of viroid-host interactions. Annual review of phytopathology, 47, 105-31 PMID: 19400635

Comments

[qetzal]

First, a great big "Thank you" to Merry for the Diener retrospective. Very interesting!

Second, in case anyone else is interested, Diener explains that viroids were isolated based on infectivity - basically testing different samples for their ability to infect a host plant (e.g. after treatment with DNase, RNase, or protease, taking different fractions from density centrifugation, etc.).

Third, upon further reading I learned that there is one viroid-like replicon known in animals - hepatitis delta virus. It's a single-stranded RNA satellite virus of Hepatitis B virus. Like viroids, its RNA genome is mostly self-complementary, forming a rod-like structure. Also like viroids, it replicates by a rolling-circle mechanism, and uses an inherent ribozyme activity to cleave off the monomers.

Unlike viroids, HDV has a large genome (1.7kb) that includes a single protein-coding gene. HDV co-opts the HepB surface antigen to form its capsid.

There's an interesting open-access review here (MMC Lai 2005, J Virol 79:7958-7958).

Merry replies:

And thank you, Qetzal, for sharing your finds. As to the HDV, it is another example of where the "critters" refuse to be assigned to neat, non-overlapping (one could also add "monophyletic") categories.

[qetzal]

How hard have we looked for viroids in animals? How were they first detected in plants, and has anyone applied similar techniques to animals?

Merry replies:
Qetzal, you do ask good questions! I don't have many answers here. I can comment that viroids were discovered in plants because they caused disease in some valued crops. When researchers tried to track down the virus responsible, they found it was not a virus but these truly bizarre things now called viroids. This personal perspective by the discoverer, Theodor Diener, might be of interest.

Timeline: Discovering viroids — a personal perspective
Theodor O. Diener

http://www.nature.com/nrmicro/journal/v1/n1/full/nrmicro736.html

Nature Reviews Microbiology 1, 75-80 (October 2003)

[Christopher Taylor]

I'd always assumed that viroids couldn't infect animals because animals don't have plasmodesmata-connected cells allowing easy transmission. Mind you, now I think about it I suppose that if viroids developed from host DNA after the evolution of plants that could also be an explanation.

Merry adds:
So many possibilities! Playing the speculative game with you, it might be that the viral defense mechanisms of animal cells might be better able to recognize and destroy viroids.

Or, if all viroids replicated in chloroplasts, then one could imagine that the viroids could co-opt only the chloroplast-associated DNA-dependent RNA polymerase for its own replication. But there are viroids that replicate just fine in the nucleus. Cross that one off the list.

Or, in addition to vertical inheritance, plants provide the possibility of horizontal transmission via insect vectors. Animals aren't so accommodating.

Or....

[Nathan Myers]

This is the most amazing bit of bio-news I've encountered all year. I place these alongside prions as somehow more accidental and less purposeful than other structures that appear in living things. Does that make me a bad evolutionist?

Attacking chloroplasts seems a clue; we should be looking for these things infecting free-living descendants of the prokaryotic progenitor of chloroplasts. As with viruses, "infection" can mean something very different for a genetically uniform population of single-celled organisms than for one of us. I cannot quite bring myself, though, to suggest that it might, like virus-like structures, have actually been useful to them.