Who is hiding behind these square scales? or The mystery of the origin of square shell plates in testate amoebae

Contributed by Anush Kosakyan


It would be hard to find a testate amoeba lover who does not know the genus Quadrulella. These vase-shaped species are a very unique group within family Hyalospheniidae (Arcellinida) since they are capable of secreting their own square or rectangular shell scales (or plates). These scales are characterisstic only for this genus, while most other members of the family are predators that build their shell by using mainly recycled plates taken or scavenged from other testate amoebae (i.e. kleptosquamy, Lahr et al. 2015).


My story started in the summer of 2013, when Edward Mitchell returned from his holiday in South Africa with a bag of Sphagnum samples. The samples were full of Quadrulella shells, including species that had never been seen since their original description back in 1957. And that was it….”the right time” for single-cell barcoding of this group. From this material, we obtained very interesting results that have been published recently (Kosakyan et al., 2016).

The most surprising result was that organisms presenting square shell plates appeared not to be monophyletic: two different groups of taxa with square scales (genus Quadrulella and newly established genus Mrabella) were indeed placed far from each other on the mtCOI-based phylogenetic tree. We thought of two possible evolutionary scenarios to explain this situation. These two hypotheses depend on whether the square-shaped plates in an organism have an autogenous or exogenous origin. This is a relevant aspect because hyalospheniids are known to engage in the behaviour of kleptosquamy, or recycling of shell plates produced by other organisms (Lahr et al., 2015).

Assuming that all amoebae that present square-shaped plates in their shells have produced them autogenously, the phylogenetic hypothesis presented here indicates homoplasy:

1) either the ability to produce square plates is an ancestral character that has been lost in a number of lineages (reversion); or

2) this character has evolved more than once independently (convergence).

We have assumed that convergence is the most parsimonious alternative (see the detailed discussion in the paper for further detail), with only two steps required – one event in genus Quadrulella and another independent event in genus Mrabella. Although we should not forget another unrelated group of organisms that are capable of producing square plates: the shelled amoeba Paraquadrulla is capable of producing calcareous square plates (the genus has not been sequenced yet, but while it is almost certainly an Arcellinid it is most likely unrelated to hyalosphenids or at best branches in a basal position to the whole group).

Alternatively, considering the fact that majority of hyalosphenids have the ability to scavenge plates from prey and use them to make the shell (kleptosquamy), a possible scenario is that Mrabella is in fact using scales from preyed Quadrulella, or scavenging these scales from the environment to construct the shell. And indeed, this was our first guess when we saw the description of Nebela galeata (current name Gibbocarina galeata) from Africa by Gauthier-Lièvre (1957). It is almost identical to Q. subcarinata in general shape and dimensions of the test (L=180-200, B=98-144, A=31-41 μm) (see Figure). Gauthier-Lièvre, 1957 (in her Fig. 10A) showed that square plates typical for Quadrulella can be integrated in the shell of N. galeata.  This is very easy to explain since Gauthier-Lièvre (1957) documented other Quadrulella species from the same locality where she found N. galeata and Q. subcarinata. Thus, other Quadrulella species could have provided the square plates use by these two species.


Legend: A- Line drawing of Gibbocarina (Nebela) galeata from Congo by Gauthier-Lievre, 1957. B and C– Scanning electron microscopy and light microscopy micrographs of Mrabella subcarinata from South Africa. Scale bars =50 µm (in A) and 20 µm (in B). Modified from Kosakyan et al. 2016.


However,  things are not that easy in our case for two reasons:

1) Mrabella would specifically select these plates, a trait that is not known for any other hyalosphenid presenting kleptosquamy; these species generally use a mixture of plates from different origins to build their shell (see for instance: Apodera vas in Fig. 69 in Meisterfeld 2002);

2) if not specifically selecting, then Quadrulella would have to be the most abundant prey organism, or Quadrulella plates would have to be the most abundant in the environment, both options are likely not true since Quadrulella tends to be present in low abundance in comparison to Euglypha (although Euglypha was not abundant in our sample where we found Mrabella subcarinata), a genus of filose testate amoebae which is generally abundant and produces oval and ornamented siliceous plates.

Additionally, there are a number of other described species in genus Quadrulella that differ from the tear-shaped morphology of Quadrulella s.str. These also present similar shapes to other genera: Quadrulella vas, Q. constricta Apodera vas, Q. lageniformis Padaungiella lageniformis, Q. tubulata P. tubulata. These “mirror” species could either be a result of convergent evolution or alternatively represent cases of the “classical” hyalosphenids (A. vas, Padaungiella ssp.) that live in environments where euglyphids are rare but Quadrulella are abundant enough to provide material for building their shells. We have suggested that at this point these Quadrulella species must be treated as incertae sedis, and their sequencing will certainly illuminate the conundrum of the evolution of square-shaped plates.


Gauther-Lievre L. 1957.  Additions aux Nebela d’Afrique. Bull. Soc. Hist. Nat. l’Afrique du Nord 48, 494-523.

Kosakyan, A., Lahr, D. J. G., Mulot, M., Meisterfeld, R., Mitchell, E. A. D. and Lara, E. 2016. Phylogenetic reconstruction based on COI reshuffles the taxonomy of hyalosphenid shelled (testate) amoebae and reveals the convoluted evolution of shell plate shapes. Cladistics, 32: 606-623. doi:10.1111/cla.12167



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s