Despite the wide use of antibodies as probes to identify microalgal species, the genetic basis of antibody binding sites is unknown. Crosses of parents having a different antibody binding pattern were carried out and the antibody binding patterns of the offspring cells were compared with those of the parents, to analyse the inheritance of antibody binding sites. Results show that in dinoflagellates: the antibody binding pattern is not affected by environmental conditions; the asexual cell division cycle has no influence on the binding activity of antibodies; and the genes for antibody binding sites lie in the nucleus.

Antibody probes have great potential to advance the pace of microalgal research. An examination of the list of references on microalgae for the last ten years clearly indicates an increasing use of antibodies in algal identification. As an example, antibodies have been used to recognise unicellular eukaryotic algae that lack a characteristic morphology, or fall into small size classes (Shapiro et al. 1989; Campbell et al. 1994). Antibodies have also been successfully employed to distinguish among morphologically similar cyanobacteria (Lopez-Rodas & Costas 1997b) and to characterise morphospecies and strains of the intricate dinoflagellate genus Prorocentrum Ehrenberg (Lopez-Rodas & Costas 1999). Immunological procedures have been successfully employed in the identification of harmful algal species (reviewed by Vrieling & Anderson 1996). Indirect immunofluorescence assays using antibodies against cell surface epitopes is the technique most commonly used to identify unicellular algae (Vrieling & Anderson 1996), but the utilisation of immunobeads to separate target microalgae has been progressing rapidly in recent years (Aguilera et al. 1996; Lopez-Rodas & Costas 1999).

Despite the wide use of antibodies as probes in marine and freshwater phytoplankton there is a lack of knowledge regarding the genetic basis of the cell surface antibody binding sites. It is a priority to elucidate if antibody binding pattern has a genetic basis, scarcely or not affected by environmental effects, or if antibody binding sites are affected by environmental, stochastic and other interacting factors.

In this work, the genetic basis of the antibody binding sites was analysed by crosses of parents having different antibody binding patterns. The antibody binding patterns of the offspring cells were compared with those shown by the parents. The genetic basis for lectin binding patterns has been elucidated using a similar methodological approach (Lopez-Rodas & Costas 1997a).

Three clonal cultures of Prorocentrum lima (Ehrenberg) Dodge (P1 2V, Pl SV and Pl 7V) were used for the production of antibodies and cross-reaction studies. Clones were grown in eight culture conditions, involving permutations of: f/2 or k medium (Sigma), 15C or 20C, and 50 or 120 l,mol photons m-2 s-' from fluorescent tubes under 12: 12 h light: dark cycles. Clones grew exponentially at one doubling each three to four days with normal morphology. The cells used in immunisation of mice were harvested by centrifugation at the midlog growth phase (day 12), washed in phosphate-buffered saline (PBS; 0.02 M phosphate, 0.15 M NaCI, pH 7.5), and fixed in 4% formaldehyde buffered in PBS. The experiments were performed with cells harvested in exponential growth phase (day 12) as well as in stationary phase (day 45).

Mice were immunised to produce separate polyclonal antibodies against cell surface antigens of the three clones. About 1.5 x 106 cells of each strain were removed in 0.5 ml complete Freund's adjuvant and subcutaneously injected into each mouse. A schedule of four booster injections was carried out at 10-day Intervals with incomplete Freund's adjuvant (Sigma). When an acceptable titre was achieved, mice were anaesthetised and production bleeds were obtained infra-heart. Blood obtained was allowed to clot (30 min) and then was centrifuged twice to remove red blood cells, yielding 1 ml of serum per bleed. Four mice were inoculated with each strain and the four sera obtained were mixed, tested, divided into 0.5 ml aliquots; dissolved in 50% PBS pH 7.5 and 0.1% sodium azide, and stored at 4C (working stock) and, -40C (frozen stock). To obtain clone-specific antisera, cross-reacting antibodies were preadsorbed (24 h, 20C) on cell pellets (108 cells to 100 lil antisera) from different strains, according to Costas et al. (1995), Costas & Lopez-Rodas (1996), and L6pez-Rodas & Costas ( 1997b).

For each immunoassay, about 5 x 105 cells were collected by centrifugation and rinsed 3 times with PBS to remove the preservative. Incubation with the primary test preadsorbed antiserum was performed at 20C for 1 h. After preliminary tests to obtain a constant primary antibody: cell ratio, an optimal dilution of 1 : 300 for polyclonal antisera or 1 : 25 for preadsorbed antisera, was used in all the experiments. After rinsing three times with PBS, the binding activity of antiserum was gauged using a secondary fluorescein isothiocyanate (FITC)conjugated goat antimouse antibody (Sigma Chemical; USA) employed at 1.6 x 10-2 tg ml-' for 1 h at 20C. After rinsing four times in PBS the samples were examined for the quality of antibody binding as previously described (Costas & Lopez-- Rodas 1994; Costas et al. 1995; L6pez-Rodas & Costas 1997b). The quality of stain was estimated by epifluorescence in a Zeiss Axiovert microscope with a filter set (450-495 nm excitation, 520-560 nm emission), using the following scale: +++ = bright stain, 100% of cells stained; ++ = medium stain, 100% of cells stained; + = low stain but obviously different from controls, less than 100% of cells stained; - no detectable reaction. All tests were read `blind', that is, the person reading the test did not know the identity of the tested material.

Sexuality was induced as described previously (Lopez-Rodas & Costas 1997a), by culturing two strains together (about 1000 cells each) in the same flask with f/2 medium supplemented with 5% soil extract. We detected sexual reproduction by direct observation (under an inverted microscope: Zeiss Axiovert 35) of pairing gametes forming a fertilization bridge, according to Bhaud & Soyer-Gobillard (1988) and Faust (1993). Newly formed zygotes were spherical, with a smooth surface and a large nucleus, and were enclosed by a three-- layered wall as described by Faust (1993). Zygotes were observed only in PI 2V X Pl 7V and in PI SV X Pl 7V matings. Zygotes were isolated with a Zeiss-Eppendorff micromanipulator-microinjector and deposited in multidish plates (Nunclon Nalgen NuncInternational; Denmark) with 2 ml of f/2 medium supplemented with 5% fetal bovine serum (Fluka, Riedel-de Haen; Germany). The zygotes of P. lima exhibited a short resting period (about one week), as described by Faust (1993). The protoplast emerged from cyst and divided, resulting in two daughter cells. The second cell division generated tetrads. Each cell in the tetrads was capable of undergoing vegetative (asexual) division. The vegetative cells from each zygote were grown as described above and collected by centrifugation; aliquots of about 103 cells were treated with each preadsorbed antibody, again as described above. The cells were observed by epifluorescence to count the percentage of positive or negative binding. At least 500 cells were counted from the offspring of each zygote. Autofluorescence controls consisted of: (1) 105 103 cells incubated in culture medium with only the secondary FITC-labelled antibody; and (2) 105 103 cells incubated identically as in the standard labelling experiments but with primary serum from nonimmunized mice. All the experiments were performed in triplicate.

Despite the wide use of antibodies in dinoflagellates there are two important questions to be resolved: what roles do genotype and environment play in determining antibody binding patterns, and what is the genetic basis of the antibody binding pattern? Our results showed that the antibody binding pattern was not affected by environmental conditions. The visual evaluation of antibody binding activity by optical staining quality did not show detectable change with different environmental conditions (Table 1). Neither the culture medium, nor temperature or irradiance influenced antibody binding, suggesting that antigenic differences in cell surfaces have a purely genetic basis. Cells from mid-log phase of exponentially growing cultures showed binding patterns similar to those of stationary phase cultures. Microscopical examination of cells from exponentially growing cultures showed that all phases of the cell division cycle were present, including cells in cytokinesis. However, all the cells of a strain showed similar antibody binding, suggesting that the asexual cell division cycle has no influence on the binding activity of antibodies. These properties make antibodies a useful tool for identifying dinoflagellate species.

Since crosses between parents having different antibody binding patterns are needed to analyse the genetic basis of antibody binding pattern, strain-specific antibodies are required. During almost 30 years in which immunology has been applied to phytoplankton characterization, strain-specific antibodies have only rarely been reported (Vrieling & Anderson 1996). However, our tests for specificity and cross reactivity of preadsorbed antisera against Prorocentrum lima strains (Table 2) showed immunological differences between strains Pl 2V, Pl SV, and Pl 7V It was possible to obtain strain-- specific antibodies against strain Pl 2V by preadsorbtion of anti-Pl 2V onto cells from Pl SV or Pl 7V strains. Strain-- specific antibodies against Pl SV could also be obtained by preadsorbing anti-Pl SV onto Pl 7V cells. Preadsorbtion is a simple procedure to obtain strain-specific antibodies in dinoflagellates (L6pez-Rodas & Costar 1999).

Direct studies through experimental breeding are scarce in dinoflagellates. Ishida et al. (1993) showed that PSP toxin composition of Alexandrium spp. is inherited in a Mendelian pattern when producing offspring, and Lopez-Rodas & Costar (1997a) revealed the genetic basis for lectin binding patterns in Prorocentrum lima. Experimental breeding showed that when crosses were made between parents having different antibody binding patterns, the antibody binding patterns were inherited in proportions of 50% + and SO% - in the F, (Table 1). Since dinoflagellates are haploid, the expected inheritance of a Mendelian character (nuclear genes) is in a 1 : 1 ratio. We suggest, therefore, that genes for antibody binding sites are coded by nuclear genes in P. lima. Antibody binding sites exhibit the same inheritance pattern as lectin binding sites (Lopez-Rodas & Costar 1997a). Lectin binding sites on the cell surface have been shown also to be antibody binding sites because both are membrane proteoglycans (Slifkin & Doyle 1990). Consequently, the molecular basis of the Mendelian inheritance of antibody binding patterns could be the same as was proposed for lectin binding sites (L6pez-Rodas & Costar 1997a).

ACKNOWLEDGEMENTS

This work was supported by D.G.E.S.-PB 96-0576-C03-01. We thank Dr J. Juste and Cristina Fenoy for expert technical assistance.

REFERENCES

AGUILERA A., GONZALEZ-GIL S., KEAFER B.A. BC ANDERSON D.M. 1996. Immunomagnetic separation of cells of the toxic dinoflagellate Alexandrium fundyense from natural plankton samples. Marine Ecology Progress Series 143: 255-269.

BHAUD Y R, SOYER-GOBILLARD M.O. 1988. Transmission of gametic nuclei through a fertilization tube during mating in a primitive dinoflagellate Prorocentrum micans Ehrenberg. Journal of Cell Science 89: 197-206.

CAMPBELL L., SHAPIRO L.P cPL HAUGEN E. 1994. ImmUnOChemiCal characterization of eukaryotic ultraplankton from the Atlantic and Pacific Oceans. Journal of Plankton Research 16: 35-51.

COSTAS E. R LOPEZ-RODAS V 1994. Identification of marine dinoflagellates using fluorescent lectins. Journal of Phycology 30: 987990.

COSTAS E. & LOPEZ-RODAS V. 1996. Enumeration and separation of the toxic dinoflagellate Alexandrium minutum from natural samples using immunological procedures with blocking antibodies. Journal of Experimental Marine Biology and Ecology 198: 81-87.

COSTAS E" ZARDOYA R., BAUTISTA J.M., GARRIDO A., ROJO C. BC

LOPEZ-RODAS V 1995. Morphospecies vs genospecies in toxic marine dinoflagellates: an analysis of Gymnodinium catenaturr/Gyrodinium impudicum and Alexandrium minutum/Alexandrium lusitanicum using antibodies, lectins, and gene sequences. Journal of Phycology 31: 801-807.

FAUST M.A. 1993. Alternate sexual reproduction of Prorocentrum lima in culture. In: Toxic phytoplankton blooms in the sea (Ed. by TJ. Smayda & Y. Shimizu), pp. 115-120. Elsevier Science Publishers, New York.

ISHIDA Y., KIM C.H., SAKO Y., HIROOKA N. & UCHIDA A. 1993. PSP toxin production chromosome dependent in Alexandrium spp. In: Toxic phytoplankton blooms in the sea (Ed. by TJ. Smayda & Y. Shimizu), pp. 881-887. Elsevier Science Publishers, New York.

LOPEZ-RODAS V & CosTAS E. 1997a. The genetic basis for lectin binding patterns in Prorocentrum lima (Dinophyceae). Phycologia 36: 406-409.

LOPEZ-RODAS V. & COSTAS E. 1997b. Characterization of morphospecies and strains of Microcystis (Cyanobacteria) from natural populations and laboratory clones using cell probes (lectins and antibodies). Journal of Phycology 33: 446-454.

LOPEZ-RODAS V. & COSTAS E. 1999. Immunogenetical characterization of morphospecies and strains of Prorocentrum (Dinophyceae). Journal of Experimental Marine Biology and Ecology 238: 293308.

SHAPIRO L.P, CAMPBELL L. HC HAUGEN E.M. 1989. Immunochemical recognition of phytoplankton species. Marine Ecology Progress Series 57: 219-224.

SLIFKIN M. & DoYLE R.J. 1990. Lectins and their application to clinical microbiology. Clinical Microbiological Reviews 3: 197-21$. VRIELING E.G. & ANDERSON D.M. 1996. Immunofluorescence in phy

toplankton research: applications and potential. Journal of Phycology 32: 1-6.

Accepted 6 February 2000

VICTORIA LOPEZ-RODAS* AND EDUARDO COSTAS

laboratorio de Genetics de Microalgas, Facultad de Veterinaria (Pabellon; de Zootecnia), Universidad Complutense, 28040, Madrid Spain

V LOPEZ-RODAS AND E. COSTAS. 2000. The genetic basis for antibody binding patterns in Prorocentrum lima (Dinophyceae). Phycologia 39: 160-162.

* Corresponding author (vlrodas@eucmax.sim.ucm.es).

Copyright International Phycological Society Mar 2000
Provided by ProQuest Information and Learning Company. All rights Reserved