Parasitic worms are small and often similar in appearance. Yet as you will read below, we put great effort into telling one species from another. You might wonder why we are so concerned about proper identification. Does it matter if we group two different species together if they are so similar anyway, or mistakenly separate one species into two? In biology, yes, it does matter.
For instance, proper identification of parasitic worms is important for treatment. Different species might call for different drugs. In order to treat a parasite infection effectively, the species responsible for the infection needs to be properly identified. A more general example is global warming. For models predicting survival in a changing environment, it is important to know who is what. When two species are assumed to be the same, but are in fact not, this might lead to overestimations of the chance of long-term survival. The opposite holds true, as well. When two ‘species’ are actually one, their chances of survival can be greatly increased.
So after discussing the commonly underrepresented role of parasites in food webs, and the significance of parasites in maintaining stable populations, this week’s focus was on the worms themselves: What are the different types of worms out there and how do we know which one is which? Well, as it turns out, parasitic worms are a pretty difficult bunch to tell apart.
Based on morphological characteristics, parasitic worms (helminths) can be split into a few distinct groups. We recognize tapeworms (cestodes), flukes (trematodes) and roundworms (nematodes). In short, tapeworms are characterised by a segmented body structure and a distinct head that features suckers and a rostellum (a retractable, cone-like muscular structure) with or without hooks to attach to the host. Flukes generally share a flat body structure and an oral and ventral sucker. The bodies of both tapeworms and flukes are covered by tegument, a protective layer that allows the worms to secrete waste and absorb nutrients. Roundworms have a cylindrical, radially symmetrical body shape and a ‘head’ that comprises a buccal opening (mouth), which may contain lips and teeth. Instead of having a tegument, their body is protected by a tough cuticle.
Within each of these three groups, variation is plentiful. Taking nematodes as an example, a vast amount of species (>25,000) has been identified (Morand et al., 2006). But estimates suggest that there might be as much as 75,000 to 1,000,000 different species of worms out there, so a large number has not even been described yet (Morand et al., 2006; Dobson et al., 2008). This might be due to the lack of methods to tell species apart, the result of species’ definitions, or simply because we still have to stumble across a number of those 1,000,000 species.
As a result of physical similarities, identifying species can prove quite cumbersome (Figure 1). Distinctions can be made based on the geographic location where the worms are found, and whether they have a direct or indirect life cycle (the latter meaning that more than one type of host is necessary for completion of the life cycle). Other distinguishing factors include the host and type of tissue that the worm infects. More detailed differentiation can be based on, for example, the distance between the oral and ventral sucker in the case of flukes. For tapeworms the shape, size or absence of rostellar hooks as well as total body length are distinctive features, while cuticular expansions, reproductive structures, the number of lips and the shape of the tail can be used to tell nematodes apart.
Figure 1. Drawings of the tail morphology of lungworm larvae of 5 genera. In these drawings differences might appear clear, but imagine 3D worms, in different positions. Would you then still be able to tell apart b from d? And would you see that c(i) and c(ii) are the same species, rather than two similar ones? This figure is taken from van Wyk et al., 2004.
The above are examples of phenotypic features that help us differentiate between species by eye. Some species are virtually impossible to distinguish between (cryptic species) and lead to the question of how different do species have to look for us to be able to tell them apart.
Molecular tools are complementary to morphology. Because of evolutionary history, certain genes are largely similar across related species, with only a few mutations defining species. These unique mutations can be used for identification using Polymerase Chain Reaction (PCR) reactions. This technique amplifies DNA, and can be customized to amplify the DNA of one specific species. At the end of a PCR the DNA is amplified so often that its presence can be detected. If the specific species is not present, no DNA will be amplified, and nothing will be detected after the PCR is completed. Once such PCRs are designed and their reliability verified, one only needs to extract the DNA of the worm of interest, or even a pool of worms, and run PCRs to determine which species are present in a sample.
This week’s articles illustrate the difficulties associated with species identification using both morphological (Kazacos and Turek, 1982) and molecular (Dangoudoubiyam et al., 2009) tools. In the former, scanning electron microscopy was used to describe the nematode species Baylisascaris procyonis while the latter uses PCR to detect this species in a multispecies sample. Kazacos and Turek (1982) address the distinct features of the worm in detail. It extensively describes the three lips and denticles (teeth like structures) that can be used to discriminate among different Baylisascaris species, with some features being unique to B. procyonis (Figure 2A-C).
Figure 2. A and B show the three lips of B. procyonis. C shows a denticular row, with a pit that was not described for ascarids (three lipped worms) before. Adapted from Kazacos and Turek, 1982.
In a background paper from the sixties (Sprent, 1967), the Baylisascaris genus was well characterized based on morphology alone. This characterization was very accurate as in the past 50 years just one species was added to the genus. Other than that, not much has changed. The problem with morphological studies however is that it is very laborious and specialized knowledge and experience is needed to reliably identify closely related species by eye, especially as species in the Baylisascaris genus do not have many external structures that assist in identification.
Molecular tools, once designed and thoroughly tested, allow for relatively quick and unambiguous identification. Dangoudoubiyam et al. (2009) describe the design of a PCR assay that selectively amplifies B. procyonis DNA. The paper shows that the method can be used on both larval and egg stages. This is a clear benefit over morphological methods as eggs, while simple to sample, are so similar that they cannot be used for morphological identification. A pitfall however was the limited specificity: it did not only amplify B. procyonis DNA as planned, but also that of the related B. columnaris (Figure 3). In other words, the current PCR setup cannot discriminate between B. procyonis and B. columnaris.
Figure 3. The results of a PCR. Pieces of DNA of different sizes are separated on a agarose gel. By electrophoresis a current draws the DNA through the gel, separating them by size. The DNA is then visualized using a UV light.. The top white bands are a ‘positive control’ and indicate that the PCR worked. The lower bands in column 2 (B. procyonis) and 9 (B. columnaris) indicate that DNA of both species was amplified. This is not good, as it should only have amplified that of B. procyonis. Adapted from Dangoudoubiyam et al., 2009.
The two studies nicely indicate how morphological (‘jeans’) and molecular (‘genes’) methods can complement each other. Morphology can be used to identify species on its own but it is time consuming and expertise is required. PCR on the other hand provides a fast and easy approach, once the method is well established for the species of interest. The challenges associated with morphological studies are not issues likely to be solved. For PCR the future looks more promising, but effort needs to be put into unique identifying sequences for all (known) worms and high throughput screening.
A subject shortly touched upon during the discussion which I find intriguing in this context, is the concept of species definitions per se. And this is also where I think the main problem behind both techniques lies. Morphology is based on phenotypes alone and phenotypes can diverge, or not, due to different causes. Growing up in alternative environments might result in different phenotypes even when the worms have the same genetic background (phenotypic plasticity). Alternatively, genetically distinct species might appear similar at a phenotypic level when they experience similar environments (genetic canalization).
At a molecular level, too, differentiation based on one or a few genes might give biased results. Currently, only known sequences are used, and sequence similarities for the genes of choice may not be representative of similarities elsewhere in the genome. Or the other way around, some individual genes may differ, but they might not actually result in a reduced reproductive potential between two ‘species’.
Regardless of which method you choose, you might falsely classify two individuals as originating from two distinct species, or vice versa. Neither method is currently ideal for calling species independently. As long as caveats remain in our understanding of genome diversity and function, impartially defining species will, in my mind, be a utopia.