The Worm that killed my Grandfather

The Worm that killed my Grandfather

I have a personal interest in the Echinococcus species of parasites - one of their number killed my grandfather. He probably contracted it when he was working in Persia, as it then was, during the 1930s and early 1940s. His lungs were particularly badly affected, so he probably had an infection with Echinococcus granulosus. I was about six weeks old when he died. He was universally liked, and I didn't get to know him. I only found out about the cause of the terrible disease he suffered a few years ago, long after I had specialised as a parasitologist. Funny old world.

Lipid binding properties of the Antigen B protein of the hydatid disease tapeworm Echinococcus granulosus

See below for our preliminary data on the lipid binding properties of this protein.

Click here (http://www.dpd.cdc.gov/dpdx/html/Echinococcosis.htm) to see the life cycle of the parasite and for a description of how horrible the infection is. 

The Antigen B protein is present at high concentrations in hydatid cyst fluid, and appears to be produced in a number of isoforms by the parasites. It has also proved to be a valuable serodiagnostic tool (e.g. (1-4)).

But, what is the function of Antigen B for the parasite? To answer that question, we first have to have a good idea of what it does biochemically in the test tube.

Here, we examined one isoform of Antigen B from Echinococcus granulosus, derived from clone Eg55 (GenBank accession number Z26336) (5). The full sequence of the protein is -

MLLALALVSFVVVTQADDGLTSTSRSVMKMIGERKYFFERDPLGQKVVDLLKELEEVFQLLRKKLRTALKSHLRELVAEGK

The region in bold is the presumptive signal sequence that is removed posttranslationally, and is not present in the recombinant protein examined here.

The amino acid sequence of E. granulosus Antigen B aligns reasonably with a small fatty acid binding protein found in the tapeworm Moniezia expansa of sheep (6,7). Such a small (~8 kDa) lipid binding/transporter protein has not been reported from other animals, although unrelated lipid binding proteins of this size have been found in plants to be important as lipid storage molecules, and some are also involved in the transmission of emergency signals between distant parts of a plant. In the case of these signaling proteins, either the proteins are themselves the signals, or they carry a hydrophobic signal molecule, or that the protein + ligand complex comprises the signal.

Back to Antigen B. The protein used in our preliminary study was bacterial recombinant Eg55, produced as a fusion protein with thioredoxin, and sent to us from Klaus Brehm's lab (kbrehm@hygiene.uni-wuerzburg.de) in the Institut für Hygiene und Mikrobiologie der Universität Würzburg.

Lipid binding was detected using a range of fluorescent lipid probes that alter their emission spectrum when they interact with a protein's hydrophobic ligand binding site.

Our data were gathered rather an embarrasingly long time ago (2001, more or less). The protein did not exactly behave ideally or as expected, so we cogitated. While we were doing so, another group published on Echinococcus Antigen B's lipid binding activities (8). So, the ship of fate and opportunity sailed without us, but our results are complementary to the published findings, and are here for all to see.

Summary of Results

The structures of all of the fluorescent lipidic probes used are given below.

  • Antigen B bound well to cPnA (cis-parinaric acid) - see Figure 1. This is a natural, intrinsically fluorescent, fatty acid (purified from the seeds of a tree growing on some very nice Pacific islands, would you believe). cPnA has no artificial fluorescence reporter group, so is just about the closest one can get to an unmodified, natural ligand that will give a good fluorescent signal.
  • Antigen B also bound ANS - see Figure 2 - which is used to detect hydrophobic regions on misfolded proteins, and also efficiently detects hydrophobic binding pockets in lipid binding proteins.
  • Arachidonic acid displaced both cPnA and ANS from Antigen B, indicating that a fatty acid could be a natural ligand for Eg55 (Figures 1 and 2). In both cases, though, the competition was not as efficient, and did not go to completion, as we see with other lipid binding proteins. This could be due to a number of reasons, such as a proportion of the recombinant protein being misfolded, the thioredoxin fusion partner interfering with normal binding behaviour, and the formation of higher order aggregates that have unusual properties (Antigen B obtained from hydatid fluid is well known to form multimers), or that the recombinant protein came pre-loaded with a ligand(s) that binds more strongly. Or, of course, that Antigen B's real preference is for hydrophobic ligands different from those we tested.
  • Antigen B did not appear to interact with either of two compounds (DAUDA and DACA) that are commonly used to investigate lipid-binding proteins from other parasites and other organisms(9-14). DAUDA and DACA are fatty acids with an attached fluorescent dansyl reporter group. So, DAUDA binds just fine to the presumptive homologue of Antigen B in M. expansa, but not to Antigen B itself. So, that's nice, we agree with the published paper (8).

Conclusions

  • E. granulosus Antigen B appears to bind lipids, although with a specificity somewhat different from that of the similar protein from M. expansa.
  • Cis-parinaric acid is a good probe for this binding activity, and is displaceable by arachidonic acid, a fatty acid that the parasite would encounter in its hosts - and a biologically important one at that.
  • ANS binding is potentially very useful in the rapid screening for more natural ligands or drugs.

What, then, might Antigen B be doing for the parasite?

  • It could be a lipid store, keeping compounds like fatty acids in solution and protecting the parasite’s membranes safe from the damaging effects of free fatty acids.
  • The fatty acids are likely to be an energy source for the parasite, and for membrane construction.
  • And/or, Antigen B could be extracting fatty acids from the host proteins, such as serum albumin, that penetrate the parasite's external cyst wall and are present in the hydatid cysts fluid. Lipids bound to Antigen B might then be accessible to the parasite's cells via appropriate receptors.
  • And/or, Antigen B could be transporting important signalling lipids or their precursors (such as arachidonic acid itself) for use by the parasite in manipulating its host tissue environment, inflammation and immune response. The parasites might then be performing the same trick as ticks. Like parasitic worms, ticks cannot manufacture their own polyunsaturated fatty acids. So, ticks (and possibly also nematodes) take in host lipids to be modified and released in the form of important signalling lipids such as prostaglandins and leukotrienes to control the host's immune and inflammatory responses.
  • Might Antigen B's unusual binding characteristics (if that they be) mean that it could be used to target parasite tissue with specific drugs?

Figures of Binding - Figures(in pdf format)

If you have any questions about any of the above, then please feel free to contact me.

Malcolm Kennedy
Institute of Biodiversity, Animal Health and Comparative Medicine 
Graham Kerr Building
University of Glasgow
Glasgow G12 8QQ
Scotland, UK
malcolm.kennedy@glasgow.ac.uk

Data collected May 2001
This file September 2005

References

1. Helbig, M., Frosch, P., Kern, P., and Frosch, M. (1993) Journal of Clinical Microbiology 31, 3211-3215
2. Gonzalez-Sapienza, G., Lorenzo, C., and Nieto, A. (2000) Journal of Clinical Microbiology 38, 3979-3983
3. Ibrahem, M. M., Craig, P. S., McVie, A., Ersfeld, K., and Rogan, M. T. (1996) Research in Veterinary Science 61, 102-106
4. Leggatt, G. R., Yang, W., and McManus, D. P. (1992) Transactions of the Royal Society of Tropical Medicine and Hygiene 86, 189-192
5. Frosch, P., Hartmann, M., Muhlschlegel, F., and Frosch, M. (1994) Molecular and Biochemical Parasitology 64, 171-175
6. Barrett, J., Saghir, N., Timanova, A., Clarke, K., and Brophy, P. M. (1997) European Journal of Biochemistry 250, 269-275
7. Janssen, D., and Barrett, J. (1995) Biochemical Journal 311, 49-57
8. Chemale, G., Ferreira, H. B., Barrett, J., Brophy, P. M., and Zaha, A. (2005) Biochimica Et Biophysica Acta-Proteins and Proteomics 1747, 189-194
9. Garofalo, A., Rowlinson, M. C., Amambua, N. A., Hughes, J. M., Kelly, S. M., Price, N. C., Cooper, A., Watson, D. G., Kennedy, M. W., and Bradley, J. E. (2003) Journal of Biological Chemistry 278, 8065-8074
10. Kennedy, M. W., Heikema, A. P., Cooper, A., Bjorkman, P. J., and Sanchez, L. M. (2001) Journal of Biological Chemistry 276, 35008-35013
11. Kennedy, M. W., Brass, A., McCruden, A. B., Price, N. C., Kelly, S. M., and Cooper, A. (1995) Biochemistry 34, 6700-6710
12. Kennedy, M. W., Scott, J. C., Lo, S., Beauchamp, J., and McManus, D. P. (2000) Biochemical Journal 349, 377-384
13. Prior, A., Jones, J. T., Blok, V. C., Beauchamp, J., McDermott, L., Cooper, A., and Kennedy, M. W. (2001) Biochemical Journal 356, 387-394
14. Suire, S., Stewart, F., Beauchamp, J., and Kennedy, M. W. (2001) Biochemical Journal 356, 369-376