Worm Breeder's Gazette 14(1): 76 (October 1, 1995)

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.

C. elegans makes a fatty acid/retinoid-binding polyprotein Wormy serum albumin?

Alan Cooper1, Dong Ma2, Larry McReynolds2, Malcolm Kennedy3

1 Department of Chemistry, University of Glasgow, Glasgow G12 8QQ, Scotland, UK
2 New England Biolabs, Tozer Road, Beverly, MA.
3 Division of Infection and Immunity, University of Glasgow, Glasgow G12 8QQ, Scotland, UK (Tel.: (+44) 141 330 5819 FAX: (+44) 141 330 4600 e-mail: gbzc02@udcf.gla.ac.uk).

        It was recently reported in the Gazette that the C. elegans genome
encodes a homolog of the nematode polyprotein allergens (NPAs) which have
been characterised from several parasitic species [Volume 13 (1), 84-85].
These are approximately 15kDa proteins synthesised as large precursor
polyproteins which are subsequently cleaved proteolytically to multiple
copies of the 15kDa functional units.  NPAs can be very abundant in some
parasites, particularly in large nematodes such as Ascaris, Toxocara and
Anisakis, and appear to be secreted in many cases.
        The NPA of Ascaris (the ABA-1 allergen purified directly from the
parasite) binds fatty acids, retinol (Vitamin A) and retinoic acid, as
does recombinant NPA from Dictyocaulus viviparus and Brugia malayi
(characterised in collaboration with Collette Britton and Judi Allen,
respectively).  To check out the C. elegans homolog, DNA encoding one unit
of the polyprotein was produced by PCR, and the recombinant polypeptide
expressed as a fusion protein with maltose binding protein.  This was used
in a spectrofluorimetric assay in which the protein was added to a
dansylated fatty acid.  This probe has a low fluorescence in water which
is enhanced and shifted to a shorter wavelength when the ligand enters a
binding protein.  Figure 1 shows the dramatic effect of addition of the C.
elegans protein, and the extent of the blue-shift in emission wavelength
is indicative of a highly apolar binding site.  Maltose binding protein
itself had no effect.  Figure 1 also shows that this change in
fluorescence is reversed by competition with natural fatty acids such as
oleic acid.  No competition was observed with another hydrophobic ligand,
cholesterol.
        Binding of retinol was indicated by an increase in its intrinsic
fluorescence upon addition of the protein (Figure 2).  This is reversible
upon addition of a fatty acid competitor, so the fatty acid and retinol
binding sites are therefore probably the same, or at least overlap.
        The NPA of C. elegans thus appears to have similar characteristics
to that of parasitic nematodes, and presumably functions to protect and
distribute insoluble and chemically labile hydrophobic compounds within
the organism after absorption by the gut.  While these results indicate
that fatty acids and retinoids are likely to be important ligands, it is
quite possible that we have missed the major binding function and that
different hydrophobic ligands are more important in vivo.
        One puzzle about the NPA of C. elegans (and also that of D.
viviparus) is the extreme variability in the 15kDa polypeptide sequences
encoded in the NPA array, which might be reflected in different binding
activities of the individual 15kDa polypeptides.  The fact that
recombinant NPA units are functional means that they can now be tested in
isolation for gross or subtle differences in their binding function using
fluorescence methods.
        So, it seems that these proteins may be the wormy equivalents of
our serum albumin.  If so, then they are produced in a way that is so far
unique to nematodes, and, to our knowledge, there are no other examples of
fatty acid binding proteins being produced as polyproteins.

Figure 1.  Binding of the fluorescence-labelled fatty acid probe
dansyl-undecanoic acid to the recombinant C. elegans protein.
Fluorescence emission spectra (excitation wavelength 345nm) of 1=B5M
dansyl-undecanoic acid is shown in the absence or presence of the protein.
Also shown is the progressive reversal of these changes in fluorescence by
addition of increasing amounts of oleic acid.


[See WBG for Figures.]

Figure 2.  Binding of retinol to the C. elegans NPA.  Fluorescence
emission spectra (excitation wavelength 350nm) of 2=B5M retinol in the
absence or presence of the protein.  This enhancement of the fluorescence
was reversed upon addition of oleic acid, and further addition of the
competitor completely cancelled the effect. The sharp peak on the left is
the Raman fluorescence from water.

Acknowledgements - We are indebted to Fiona McMonagle for technical help
and to the Wellcome Trust for support through a grant to MWK.  AC is
grateful to his more conservative chemistry colleagues who could not
believe the existence of such a journal and thereby inspired this
publication.