Dinoflagellates are a group of single celled algae that compose a highly diversified phylum that displays an amazing range of ecological adaptation. Different species employ autotrophy, heterotrophy or mixotrophy, many are known to be symbiotic or parasitic, and bioluminescence is common. They are found at all latitudes and are often a significant component of marine and freshwater phytoplankton communities. Dinoflagellates are also notable for their unusual genome content and organization (reviewed in [1, 2]). Estimates of dinoflagellate DNA content range from 3 to 250 pg·cell-1 [3, 4], corresponding to approximately 3000–215,000 Mb (in comparison, the haploid human genome is 3180 Mb and hexaploid Triticum wheat is 16,000 Mb). It has been suggested that polyploidy or polyteny may account for this large cellular DNA content , but studies of DNA reassociation kinetics do not support this hypothesis. Dinoflagellates have many chromosomes (up to 325) that are permanently condensed and attached to the nuclear envelope during cell division . Dinoflagellates are the only eukaryotes with DNA that contains 5-hydroxymethylmuracil, which replaces 12–70% of the thymidine .
The unique physical features of the dinoflagellate chromosomes are likely to affect both gene transcription and regulation. While there is an increasing amount of expressed sequence tag information available for dinoflagellates, very few genes have been well characterized with respect to their gene structure and regulation. The few nuclear genes that have been isolated from genomic DNA seem to uniformly lack typical eukaryotic transcriptional elements (e.g. TATA boxes) and polyadenylation sites [8–10]. Studies of dinoflagellate gene expression indicate that these organisms employ both transcriptional (e.g. pcp ; Sahh, Map and Haf  and post-transcriptional (e.g. lbp ; GAPDH ) regulation, with the iron superoxide dismutase of Lingulodinium polyedrum exhibiting both modes, depending upon the stimulus . Recent results from microarray analysis of the dinoflagellate Pyrocystis lunula indicate that approximately 3% of the transcripts included on the array exhibit transcriptional regulation [16, 17].
Together, all of the above data suggests that the organization and regulation of dinoflagellate genes may be different from that of most other eukaryotes. Early microscopic observations of the unusual dinoflagellate nuclear structure led to the hypothesis that dinoflagellates were "mesokaryotes", an intermediate between prokaryotic and eukaryotic microorganisms . However, molecular phylogenetic evidence has since clearly identified them as eukaryotes, and their phylogenetic placement supports Loeblich's (1976)  evolutionary interpretation that the unusual properties of dinoflagellate nuclei are derived and not representative of a mesokaryotic ancestral state. As such, our basic knowledge of eukaryotic genetics and gene expression could only be increased by understanding how (and why) dinoflagellates structure their genes and regulate transcription within the sheer quantities of DNA in their cells. To date, most of the data of gene regulation mechanisms in dinoflagellates has emerged sporadically, from studies of specific genes that are of interest for a particular function. The advent of genomic technologies, in particular global gene expression profiling methods, provides the ability to learn about many genes or transcripts simultaneously, even in uncharacterized systems like dinoflagellates. The application of transcriptional profiling to dinoflagellates, in conjunction with laboratory-based gene characterization, holds tremendous potential for understanding gene regulation in this unique and understudied group. In addition, the availability of broad-based gene expression data has the potential to greatly accelerate the pace of research and discovery for dinoflagellates, algae in general and eukaryotic systems as a whole.
This report describes a global and quantitative analysis of the transcriptome of a dinoflagellate. As a model we have used Alexandrium fundyense, a species that is capable of producing potent neurotoxins, called saxitoxins, which are the causative agent of paralytic shellfish poisoning. The genus comprises approximately 30 different species that are found worldwide, and 10 of which are known to be toxic and cause so-called "red tides" or harmful algal blooms. This study examined gene expression in nutrient-stressed Alexandrium cells using Massively Parallel Signature Sequencing (MPSS), a proprietary technology developed by Solexa, Inc . The method is similar to the well-known Serial Analysis of Gene Expression (SAGE)  in that it acquires a short DNA sequence from a defined position in each gene transcript. However, the depth of sampling with MPSS is much greater, with the resulting data set containing at least 1 million, 17-nucleotide 'signature' sequences, making the technology sensitive to genes expressed at low levels.
The MPSS method is "global" in that it provides quantitative expression information for the entire transcriptome. For uncharacterized organisms like dinoflagellates, MPSS can provide a broader view of the transcriptome than microarray expression profiling, which generally includes only a portion of the transcripts present in a cell. Statistical methods for the analysis of quantitative expression data have demonstrated that the MPSS data are robust . Accepting an estimate of 300,000 mRNA molecules in an average eukaryotic cell, MPSS constitutes a three-fold sampling of a single cell, allowing the identification, comparison and quantification of even rare transcripts. The resulting Alexandrium MPSS data provide a quantitative assessment of the magnitude of transcriptional regulation in dinoflagellates. Comparison of the Alexandrium results to those of other eukaryotes indicates that the distribution and abundance of signature sequences is quite similar in Alexandrium, humans and Arabidopsis. Finally, identification of MPSS signatures via sequencing provides insight into one mechanism that may contribute to the observed signature number in Alexandrium.