Intra- and interspecies differences in growth and toxicity of Pseudo-nitzschia while using different nitrogen sources

Abstract

Clonal cultures of plankton are widely used in laboratory experiments and have contributed greatly to knowledge of microbial systems. However, many physiological characteristics vary drastically between strains of the same species, calling into question our ability to make ecologically relevant inferences about populations based on studying one or a few strains. This study included 19 non-axenic strains of three species of the diatom Pseudo-nitzschia isolated primarily from the mid-Atlantic coastal region of the United States. Toxin (domoic acid) production and growth rates were measured in cultures using different nitrogen sources (NH₄⁺, NO₃⁻ and urea) and growth irradiances. The strains exhibited broad differences in growth rate and toxin content even between strains isolated from the same water sample. The influence of bacteria on toxin production was not investigated. Both P. multiseries clones produced toxin, yet preferentially used different nitrogen sources. Only two of nine P. calliantha and two of five P. fraudulenta isolates were toxic and domoic acid content varied by orders of magnitude. All three species had variable intraspecies growth rates on each nitrogen source, but P. fraudulenta strains had the broadest range. Light-limited growth rate and maximum growth rate in P. fraudulenta and P. multiseries varied with species. These findings show the importance of defining intra- and interspecies variability in ecophysiology and toxicity. Ecologically relevant functional diversity in the form of ecotypes or cryptic species appears to be present in the genus Pseudo-nitzschia.

Introduction

Much of what we know about plankton physiology comes from experiments using laboratory cultures. Most Pseudo-nitzschia culture studies focus on P. multiseries with the results extrapolated to other species and strains (Bates, 1998). However, broad species and strain differences have been well documented (Gallagher, 1980, Gallagher, 1982, Elrifi and Turpin, 1985, Goldman and Dennett, 1985, Holmes et al., 1991, Wood and Leatham, 1992, Larsen and Bryant, 1998, Burkholder et al., 2005) and draw into question the validity of making conclusions about microalgal species physiology based on one or a few strains.

Such high variability among strains casts doubt on the species concept in microalgae in general. While the creation of a species is the result of a biological reality, secondary characteristics are used (frustule morphology, ribosomal genes) to detect and define them. Manhart and McCourt (1992) stated that:

“Practicing phycologists often seem to strive to delineate biological species while basing descriptions solely upon morphological data. The assumption is that morphological species closely approximate biological species of algae, but only rarely is this hypothesis tested. If species assignment is a hypothesis of relationship, then many (perhaps most) implicitly described biological species of algae represent untested hypotheses.”

This implies a basic fault in microalgal species definitions which could explain high strain variability. However, Wood and Leatham (1992) argue that difficulties arise when culture studies involving only a few clones are used in an attempt to define interspecies differences without assessing within species variation, not necessarily as the result of a fault in the species definition. The number of isolates necessary to accurately define within species variation using statistical methods is difficult to define and restricts the number of studies able to include such an analysis (Lakeman and Cattolico, 2007).

The alternative, making conclusions about a species based on one strain, would likely lead to erroneous results. Toxin production in 17 strains of Alexandrium tamarense showed broad differences, the lowest approximating zero mouse units (MU) per 10⁴ cells and the highest at 1.1 MU per 10⁴ cells (Ogata et al., 1987). No one strain was representative of the species. Taking the analysis one step further, 15 sub-strains taken from one strain had a 0.6 MU per 10⁴ cells range in toxin production. An analysis of PSP toxin composition in two strains of A. tamarense showed that one strain, SB31, produced mostly the sulfocarbamoyl derivative C2 while the other strain, SB32, produced mostly GTX3 and GTX4 (Cembella et al., 2002). Studies such as these show broad differences among strains and argue that finding one “representative” strain is highly unlikely if not impossible. Yet, an analysis of recent publications shows only 40% of studies that use culture experiments consider the possibility of significant strain differences when making conclusions (Burkholder and Glibert, 2006).

Genetic variability in field populations and strains of Pseudo-nitzschia has been widely documented; however, physiological variability has not been as thoroughly investigated (Evans et al., 2004, Orsini et al., 2004). Many Pseudo-nitzschia culture studies present results from one strain (Bates et al., 1991, Douglas and Bates, 1992, Hillebrand and Sommer, 1996, Pan et al., 1996a, Pan et al., 1996b, Fehling et al., 2004, Armstrong Howard et al., 2007). Others compare one strain of multiple species (Jackson et al., 1992, Hargraves et al., 1993, Wang et al., 1993, Maldonado et al., 2002, Fehling et al., 2005). Most studies comparing multiple cultures of one species have focused on toxin production in P. australis, P. seriata or P. multiseries (Bates et al., 1989, Bates et al., 1999, Garrison et al., 1992, Douglas et al., 1993, Villac et al., 1993, Lundholm et al., 1994) with one study investigating toxicity in P. pseudodelicatissima (Pan et al., 2001). Only four studies used multiple strains of the same species to investigate other physiological processes in addition to toxicity. With four strains of P. multiseries and two strains of P. pungens, Bates et al. (1993) analyzed the effect of NO₃⁻ and NH₄⁺ on growth and toxin production. Lundholm et al. (2004) employed two strains of P. multiseries to study the effect of pH on growth and toxin production. Bates et al. (1995) used three strains of P. multiseries to examine the role of bacteria in domoic acid (DA) production. Thessen et al. (2005) studied two strains of P. delicatissima, two strains of P. multiseries and three strains of P. pseudodelicatissima to assess the effect of salinity on growth rate.

Bacterial communities associated with algal cells are increasingly recognized as an important factor in cell physiology that can be highly variable between laboratory strains (Kaczmarska et al., 2005). In Pseudo-nitzschia cultures, biomass and growth rates are not affected by absence or reintroduction of bacteria, but DA production can be 2- to 95-fold higher in the presence of bacteria (Bates et al., 1995). Studies have verified that Pseudo-nitzschia is the source of DA and bacteria are incapable of producing DA autonomously (Douglas and Bates, 1992, Bates et al., 2004). Bacteria play an as yet undefined role in DA production and can have a differential effect among strains and throughout the life of a culture (Stewart, 2008). It has been proposed that the amount of DA measured in a culture is the result of competitive interactions between Pseudo-nitzschia production rate and utilization of DA by extracellular bacteria (Stewart, 2008). This, instead of genetic variability, could explain measured differences in DA production between Pseudo-nitzschia strains (Stewart, 2008).

The meaning of high intraspecific genetic diversity in natural populations is controversial (Fenchel, 2005, Foissner, 2006). High genetic diversity in some protist taxa has been considered an indicator of cryptic species (species that are identical morphologically, but reproductively isolated) and functional diversity (Dolan, 2005, Foissner, 2006, Scheckenbach et al., 2006). Others argue that variation in rRNA is an accumulation of neutral mutations that does not correlate with physiology or show biogeographic patterns (Fenchel, 2005). However, there have been populations of microalgae comprised of distinct physiological or genetic groupings which showed dynamic seasonal abundances. The presence of multiple ecotypes has been demonstrated in populations of Skeletonema costatum in Narragansett Bay (Gallagher, 1980, Gallagher, 1982). These ecotypes have different physiological characteristics, making them better adapted to different environmental conditions and resulting in a succession of ecotypes throughout the year. Similar results have been found with Ditylum brightwellii in Puget Sound (Rynearson et al., 2006). Reproductively isolated cryptic species within P. delicatissima and P. pseudodelicatissima have been identified using morphology, genetic sequences and mating experiments (Amato et al., 2007). The ecological significance of this diversity is not well understood.

This paper is a presentation of strain differences between three species of Pseudo-nitzschia: P. multiseries, P. fraudulenta and P. calliantha for growth rate, toxin production, nitrogen use and saturating growth irradiance. It is also the first presentation of genetic and toxin data from Chesapeake Bay area Pseudo-nitzschia strains.