A Penny For Your Thoughts

So, what should we do about Blastocystis? What do we want to know?

I believe the imminent answer to the latter question is easy: We want to know whether it’s pathogenic, whether we should treat it and how. But I also think that there are many other interesting aspects of Blastocystis which are also of broad interest to the general public, namely: How about the many cases of asymptomatic Blastocystis carriage? What does Blastocystis do in our guts? Could it have any potentially beneficial impact on our health?

Given the fact that Blastocystis has not been implicated in any outbreaks (admittedly: I guess that no one actually ever looked for Blastocystis in outbreak investigations... except for me!), I reckon that the chance of it being involved in acute diarrhoea is small. So, in that respect it's very different from the other intestinal protists such as Giardia, Cryptosporidium, Cyclospora, microsporidia, even Entamoeba histolytica. It's actually more reminiscent of helminth infections, which are are often chronic, and when light hardly give rise to symptoms (depending on species that is!).So I'm more thinking along the lines of co-evolution, adaptation, etc.

Maybe future research will call for a shift in paradigm, but until then I think that we should do what we already can, just at a larger scale and see where it takes us, namely:

Where Are We On Blastocystis Subtypes?

As mentioned, Blastocystis exhibits remarkable intrageneric diversity, which is continuously being explored by us and our colleagues. We are convinced that the genus of Blastocystis comprises multiple species, but for now we call them "ribosomal lineages" or "subtypes" and allocate numbers to each subtype, hence ST1, ST2, etc. While the number of subtypes that can be found in humans remains stable, we and our colleagues are still expanding the subtype universe in non-human hosts (I will be blogging on this shortly).

Barcoding currently represents state-of-the-art in Blastocystis subtyping, and luckily this method appears to gain a foothold in labs across the world.

Nine subtypes have been found in humans, but some of them only on rare occasions. A recent study going out from London School of Hygiene and Tropical Medicine and led by Dr Alfellani and published just now in Acta Tropica looked at 356 Blastocystis sequences from samples from the UK and Libya, but also from sub-Saharan Africa, namely Liberia and Nigeria.

Is This A New Subtype?

To quote one of my colleagues attending the recent IWOP 2012 meeting in Tarrytown, NY, Blastocystis subtyping in humans and animals is becoming 'trendy', and so we keep trying to advocate for a standardisation of the metholodology of Blastocystis subtyping.

We recently changed the title of our page at www.pubmlst.org/blastocystis so that now it is called Blastocystis Subtype (18S) and Sequence Typing (MLST) Databases, and we added some text to front page:

In terms of genetic markers, the barcode region (Scicluna et al., 2006) is by far the best represented in publicly available sequence databases, and the correct subtype can be identified by BLAST analysis in the sequence database at the present site. Blasting against this database has the added advantages, compared to using GenBank, of automatically assigning allele types to the SSU-rDNA as well as using the consensus subtype nomenclature (unlike GenBank where the subtype is included only if one was part of the accession submission and no attempt to impose a standard nomenclature is made). In case the sequence does not match any of the ones in the database despite full coverage of the region, this indicates that the sequence represents a new allele or maybe even a new subtype depending on the amount of variation. If a new subtype is suspected, we suggest doing PCR and sequencing of the complete SSU rRNA gene with subsequent phylogenetic analysis using reference sequences.

Now, the last bit is extremely important. We have seen examples of researchers (including ourselves!) assigning sequences to a new a subtype in the absence of complete SSU rDNA data (in fact complete sequences for ST10-ST14 are not yet publicly available!). Doing so has a least two major limitations/drawbacks: Far from all SSU rDNA regions have been validated as being representative of the whole SSU rRNA gene in terms of phylogenetic analysis, and therefore phylogenetic inferences based on non-validated regions may have little or at least less support than anticipated. Moreover, if someone analyses e.g. position 600-1600, and phylogenetic analysis based on this region reveals a potentially new subtype, this makes it impossible for his/her colleague who has data covering positions 1-600 from a Blastocystis isolate that may also represent a new subtype to ascertain whether it might be same subtype (see example below)!

Obtaining complete SSU rDNA sequences directly from faecal DNA may be a cumbersome task but is sometimes possible by combining sequence-specific primers with low-specificity primers such as the RD5 and the RD3 primers (Clark, 1997). If a cultured isolate is available, obviously this makes complete SSU rDNA sequencing much easier.

While it appears that the number of subtypes occurring in humans stays around 9, our gut feeling is that we are yet to uncover quite a few subtypes colonising non-human mammals, and it's great to see an increasing number of teams exploring the genetic diversity of Blastocystis. For instance, Dr Ronald Fayer and his group recently published exciting data on a new Blastocystis subtype in cattle, which they named ST14 (Fayer et al., 2012).

Importantly, caution should be taken to avoid creating confusion in subtype terminology. Confusion can arise when independent researchers assign the same new subtype name (e.g. ST14, ST15, etc.) to novel sequences which in fact belong to different ribosomal lineages, or when incomplete SSU rDNA sequence data are used; this situation was seen recently, when Petrasova et al. (2011), assigned a Colobus sequence to ST5, although it was in fact a ST13 sequence (Clark et al., in press); the situation arose, since Petrasova et al. (2011) did not have data covering the region currently available for ST13 (Parkar et al., 2010), and therefore believed that their sequence was a unique ST5 variant. As for ST14, less than 500 bp are currently available, and these 500 bp are not in the barcode region, making it difficult for all teams using barcoding to compare their data. And so we would like to advocate for making complete SSU rDNA sequences publicly available (Genbank) for potentially new subtypes, for at least two reasons:

1. Phylogenetic inferences based on the complete SSU rDNA will be more robust than those obtained from analysing shorter sequence streches.

2. Complete seqeunces are needed for reference since subtype screening typically includes a single round PCR such as barcoding (Scicluna et al., 2006) amplifying about 550 bp; in the situation where complete SSU rDNAs are available for all known subtypes, it will be quick to analyse, whether a sequence may represent a new subtype, since this will be independent on the SSU rDNA region studied.We therefore hope that complete SSU rDNA sequences will soon be made publicly available for ST10-ST14.

So, when does a complete SSU rDNA sequence represent a new subtype? Well, we have a review paper in press in Advances in Parasitology on recent developments in Blastocystis research, which will be published in less than six months probably, and which also touches on this topic; once the paper is published, I will try and make a summary our thoughts on this...

Further reading:

Clark CG (1997). Extensive genetic diversity in Blastocystis hominis. Molecular and biochemical parasitology, 87 (1), 79-83 PMID: 9233675

Fayer R, Santin M, & Macarisin D (2012). Detection of concurrent infection of dairy cattle with Blastocystis, Cryptosporidium, Giardia, and Enterocytozoon by molecular and microscopic methods. Parasitology research PMID: 22710524

Parkar U, Traub RJ, Vitali S, Elliot A, Levecke B, Robertson I, Geurden T, Steele J, Drake B, & Thompson RC (2010). Molecular characterization of Blastocystis isolates from zoo animals and their animal-keepers. Veterinary parasitology, 169 (1-2), 8-17 PMID: 20089360

Petrášová J, Uzlíková M, Kostka M, Petrželková KJ, Huffman MA, & Modrý D (2011). Diversity and host specificity of Blastocystis in syntopic primates on Rubondo Island, Tanzania. International journal for parasitology, 41 (11), 1113-20 PMID: 21854778
 
Scicluna SM, Tawari B, & Clark CG (2006). DNA barcoding of blastocystis. Protist, 157 (1), 77-85 PMID: 16431158

The Circular Problem of Blastocystis

After submitting stool samples for microbiological analyses, many people with intestinal symptoms are informed by their GPs that they have Blastocystis, and that the clinical significance of this parasite is unknown (which is not entirely wrong). However, some GPs may want to try to eradicate Blastocystis in the absence of other potential causes of the symptoms, prescribing drugs such as Protostat/Flagyl (Metronidazole). During and after treatment, many patients will experience temporary alleviation only "to get back to where they started" after a couple of weeks or so. And often, they will also remain positive for Blastocystis (sometimes Blastocystis may be very difficult to detect during the course of treatment and immediately after treatment, which may be due to a transitory decrease in parasite load for direct and indirect reasons; see below). Anyway, this is the classical scenario.

The problem with Blastocystis is a circular one: There is currently no single 100% successful treatment, and when people with symptoms + Blastocystis cannot get rid of their parasites and thereby get the chance to report on symptom status after permanently cleared infection (+/-clinical improvement), it is - to put it mild - extremely challenging to collect information and data that can assist us in drawing conclusions. It doesn't make it any better that we know that a lot of people have Blastocystis without knowing and without having symptoms.We therefore shouldn't blame health care professionals for being in the dark.

People who do not know a lot about Blastocystis (and who does?) might take to the Internet to get more information, including how to deal with the infection. Not all the advice given on the Internet may be useful and little of it will be based on scientific evidence. Some people may be desperate to try and clear any parasite that they have been diagnosed with, without realising that some parasites might actually be a sign of a healthy gut ecological system and be of potential benefit in terms of maintaining a healthy immune system; we don't know much about this yet. Or maybe the use of antibiotics will damage the general intestinal flora and cause more or more severe symptoms than would the persistence of the parasitic infection! We don't know, but as hinted at in previous posts, our new technologies will assist us in obtaining the information that we have been looking for so long.

So, how do we move on from here? There is no doubt that scientific studies are key. Pilot data are available showing that at least one of the genetic variants (subtypes) of Blastocystis is more common in patients with symptoms than in the background population, but this still needs confirmation.

The genetic diversity of Blastocystis found in humans is huge. If the genetic diversity of Blastocystis was visible, different subtypes of Blastocystis would probably be as different as these marble balls!

We need substantial funding for carrying out large-scale studies aiming to confirm these data. Once epidemiological association has been sufficiently demonstrated, the next step is to find out whether those strains/subtypes associated with disease are characterised by having effector proteins not seen or - maybe more convincingly - not expressed in strains/subtypes seen in healthy individuals. If we have proof of both epidemiological association and expression of virulence genes, then next step could  include a randomised control treatment (RCT) study in order to identify the drug(s) that lead to microbiological and/or clinical improvement, i.e. parasite eradication and alleviation of symptoms, respectively.

It may be so that different subtypes of Blastocystis respond to different antibiotics. And if successful treatment is dependent on other factors as well such as complex microbial interspecies interactions, it may be perplexing to realise, that different individuals may respond differently to any given treatment. As Pepper and Rosenfield suggest in their paper about microbiome multistability: A key consequence of multistability is that different instances of the same type of system, such as different individual gut microbiomes, may show very different responses to the same perturbation.

And so, how does this relate to Blastocystis treatment? Well, since none of the treatments used for treating Blastocystis are specific for this parasite (metronidazole for instance is a broad-spectrum antibiotic used to eradicate a range of anaerobic bacteria, including Clostridium), there will probably be a mixture of direct and indirect effects on Blastocystis upon treatment. The direct effect on Blastocystis will depend on its susceptibility to the antibiotic, while the indirect effect will depend on the bacterial flora and how it responds during treatment. Hence, drugs may directly affect Blastocystis and/or perturb the intestinal flora to an extent which makes it an unsuitable habitat for Blastocystis. We hope soon to be able to investigate the interaction between Blastocystis and gut bacteria by metagenomic approaches. It should be kept in mind though that metronidazole is absorbed from the proximal part of the intestine, while Blastocystis is a parasite of the colon; hence, it may very well be so that metronidazole does not reach Blastocystis in its niche. When treating intestinal amoebiasis, metronidazole is given together with a luminal drug to ensure targeting both invasive and the luminal Entamoeba histolytica.

So, while we should keep pursuing the role of Blastocystis in disease, we should also try to explore whether there are some good sides to Blastocystis colonisation and whether we can learn to see the parasite as a proxy for something (clinical condition, enterotype, etc.). I have expanded a bit on this in my recent paper "Thinking Blastocystis Out Of The Box", available in the journal Trends in Parasitology. To this end, learning about the bacterial communities that may influence Blastocystis growth and establishment may improve our ability to understand Blastocystis in an ecological context.

For those who are interested in this, may I suggest some further reading (including papers on (unpredictable) antibiotics-associated changes in gut flora and individualised responses to perturbations in the gut microbiome and a couple of studies on Blastocystis subtypes where links to disease phenotypes have been identified):

Pepper, J., & Rosenfeld, S. (2012). The emerging medical ecology of the human gut microbiome Trends in Ecology & Evolution, 27 (7), 381-384 DOI: 10.1016/j.tree.2012.03.002

Dethlefsen, L., & Relman, D. (2010). Colloquium Paper: Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation Proceedings of the National Academy of Sciences, 108 (Supplement_1), 4554-4561 DOI: 10.1073/pnas.1000087107

Stensvold, C., Christiansen, D., Olsen, K., & Nielsen, H. (2011). Blastocystis sp. Subtype 4 is Common in Danish Blastocystis-Positive Patients Presenting with Acute Diarrhea American Journal of Tropical Medicine and Hygiene, 84 (6), 883-885 DOI: 10.4269/ajtmh.2011.11-0005

Domínguez-Márquez, M., Guna, R., Muñoz, C., Gómez-Muñoz, M., & Borrás, R. (2009). High prevalence of subtype 4 among isolates of Blastocystis hominis from symptomatic patients of a health district of Valencia (Spain) Parasitology Research, 105 (4), 949-955 DOI: 10.1007/s00436-009-1485-y

Stensvold, C., (2012). Thinking Blastocystis Out Of The Box Trends in Parasitology DOI: 10.1016/j.pt.2012.05.004