Cyanobacteria are one of the largest sub-groups of Gram-negative photosynthetic prokaryotes known, and contain an extensive number of different genera and species (Lu et al., 1997). Cyanobacteria are also considered the most ancient group of oxygenic photosynthetic organisms (Graham and Wilcox, 2000). Additionally, members of this group possess chlorophyll a and phycobiliproteins such as phycocyanin and phycoerythrin, which are responsible for the blue-green pigmentation often evident in this group (Rudi et al., 1997). As a result of their pigmentation, cyanobacteria were traditionally referred to as blue-green algae. Furthermore, cyanobacteria exhibit 16S and 5S rRNA sequence similarities to other members of the eubacteria (Castenholz et al., 1989).

Cyanobacteria are among the most geographically widespread group of organisms known, and are known to dominate some aquatic and terrestrial environments (Honda et al., 1998). Within these environments, cyanobacteria are often primary producers in a number of complex and diversified food webs (Anagnostidis and Komárek 1985; Honda et al., 1998). Interestingly, cyanobacteria have proven to be useful tools in examining the photosynthetic and endosymbiotic origins of chloroplasts in plants. It is commonly accepted that early endosymbiotic strains of cyanobacteria gave rise to photosynthetic organelles in plants such as chloroplasts (Iteman et al., 2000). The endosymbiotic relationship between early forms of cyanobacteria and eukaryotes allowed primitive plants to undergo oxygenic photosynthesis (Graham and Wilcox, 2000).

Current research involving cyanobacteria has provided useful and practical information in regard to their environmental and ecological significance. In desert soils for instance, certain cyanobacterial strains (e.g. Nostoc) fix atmospheric nitrogen, a vital nutrient for vascular plant physiology (Buttars et al., 1998). Additionally, cyanobacteria play a critical role in the stabilization of desert soils. A vast number of cyanobacteria produce and secrete mucilage, which aids in the prevention of soil erosion by binding soil particles together. For instance, in desert reclamation studies, inoculated cyanobacterial treatments have been applied to disturbed and damaged desert soil crusts (cryptogamic crusts) in an attempt to stabilize and protect desert soils (Buttars et al., 1998). The results from these experiments have shown that inoculation treatments have been successful in re-establishing cyanobacterial biomass in these soils (Buttars et al., 1998).

Cryptogamic crusts are conglomerates of microbiotic organisms, which include cyanobacteria, lichens and mosses. These cryptogamic crusts play an essential role in supporting arid plant communities in the deserts of the Western United States in that they contain nitrogen fixing and mucilage producing cyanobacteria. Consequently, damage to these crusts can be quite a detriment to the fitness and survival of vascular plant communities within these deserts.

Despite their roles as ecologically important entities, the widespread persistence of cyanobacteria also comes with some consequences. For example, members of the genus Microcystis are linked to the occurrence of noxious cyanobacterial blooms in many fresh water systems (Neilan, 1995; Neilan et al., 1997). Certain species of Microcystis respond to various environmental stimuli such as the eutrophication of their habitats, resulting in the formation of blooms (Neilan et al., 1997). These rapidly forming blooms often produce noxious compounds such as microcystins, a family of heptapeptide toxins that cause hepatotoxicity in agricultural and domestic animals (Neilan et al., 1997). The elevated level of microcystins in human drinking water has been linked to the occurrence of liver tumors in humans (Neilan et al., 1997). The widespread occurrence of toxic bloom forming species of Microcystis has prompted many phycologists to study and classify members of this genus. In terms of the overall study and classification of cyanobacteria, the ecological and environmental importance and relevance for studying cyanobacteria has been firmly established. However, the alpha-level taxonomy of this group is presently in a state of disarray

Due to their prokaryotic and phototrophic nature, cyanobacteria were originally grouped with traditional prokaryotic algae (Anagnostidis and Komárek, 1985). Traditional methods of classifying cyanobacteria relied heavily upon morphological and cytological characteristics (Anagnostidis and Komárek, 1985; Castenholz and Waterbury, 1989; Neilen et al., 1995; Lu et al., 1997). These characteristics were used to define and separate cyanobacteria into distinct divisions and classes (Lu et al., 1997). The earliest taxonomic monographs aimed at classifying cyanobacteria considered them algae. (Thuret, 1875; Bornet and Flahault, 1888; Gomont, 1892). Taxonomists who were involved in these classification schemes during this time were limited by the absence of culture strains, advanced light microscopes and biochemical tests. Eventually, it was established that blue-green algae lacked structures such as chloroplasts and membrane-enclosed nuclei that were present in green algae. This observation led to the conclusion that cyanobacteria were more closely related to bacteria (Cohn, 1872; 1875).

Bornet and Flahaut (1888) and Gomont (1892) wrote the first cyanobacterial taxonomic monographs. Bornet and Flahault (1888) classified the heterocystous cyanobacteria, while Gomont was responsible for the classification of nonheterocystous, filamentous blue-green algae (Oscillatoriaceae). Exactly 40 years later, Geitler (1932) provided a comprehensive taxonomic review that recognized 1300 species, 145 genera, 20 families, and 3 orders. The classification system proposed by Geitler (1932) marks the start of the modern era of cyanobacterial systematics (Turner, 1997). Geitler’s work (1932) relied on morphology of field-collected specimens and formed the foundation for numerous subsequent taxonomic revisions (Desikachary, 1959; Fritsch, 1945; Bourelly, 1970; Golubic, 1967). The morphological, botanical approach of these classic monographs is referred to as the "Geitlerian" approach to cyanobacterial systematics (Anagnostidis and Komárek, 1985).

Drouet and Daily (Drouet, 1968, 1973, 1978, 1981; Drouet and Daily, 1956) later revised Geitler’s classification scheme. In this revision, Drouet and Daily drastically reduced the bulk of Geitler’s recognized taxa to 24 genera and 62 species based on their hypothesis that most species of cyanobacteria are actually morphologically variable ecophenes (forms with the same genotype showing phenotypic differences due to environmental stimuli) or a small number of genetically homologous taxa (Anagnostidis and Komárek, 1985). Drouet’s work consisted of examining and classifying large numbers of collected specimens using light microscopy. The relatively simplistic nomenclature used in Drouet’s classification scheme was very popular among phycologists and taxonomists (Anagnostidis and Komárek, 1985). Unfortunately, Drouet’s classification system failed to reflect the genetic diversity among cyanobacteria in both nature and in culture (Anagnostidis and Komárek, 1985). In addition, Drouet failed to incorporate biochemical, genetic, or physiological data in his classification scheme (Anagnostidis and Komárek, 1985). As a result of these deficiencies, Drouet’s classification scheme was deemed unacceptable by many phycologists and shortly thereafter, other taxonomists including Stanier et al. (1978), began devising classification schemes which reflected the plasticity identified by Drouet while recognizing the genetic heterogeneity he missed (Anagnostidis and Komárek, 1985; Turner, 1997).

The classification scheme proposed by Stanier et al. (1971), Stanier et al. (1978), Waterbury and Stanier (1977, 1978), and Rippka et al. (1979), used the strain clone as the basic taxonomic unit. This method of classification is typically popular among bacteriologists, who focused on defining genera based on a polyphasic approach. The polyphasic approach used by bacteriologists evaluated physiological, cytological, and biochemical characteristics, in addition to morphological features of axenic strains (Anagnostidis and Komárek, 1985; Castenholz and Waterbury, 1989). In using this approach, many bacteriologists avoided describing species (Castenholz and Waterbury, 1989). Stanier et al. (1978) and Rippka et al. (1979) also proposed that since cyanobacteria are prokaryotes, the systematic treatment of this group should not be based on botanical criteria, but on the examination of axenic cultures using bacterial criteria.

As stated previously, the taxonomic differentiation of cyanobacteria is based upon the morphological distinction between different taxa (Ishida et al., 1997). However, the concomitant application of both botanical and bacteriological criteria in cyanobacterial taxonomy has resulted in ongoing controversies in regards to the proper classification of this group (Garcia-Pichel et al., 1998). Anagnostidis and Komárek (1985; 1990; 1999) proposed the most recent re-classification scheme aimed at addressing this issue. Anagnostidis and Komárek developed a formal classification scheme that focuses on objective features and characteristics that occurred across genera and that could be used to distinguish species (Anagnostidis and Komárek 1985, 1988, 1990, Komárek and Anagnostidis, 1986, 1989). The purpose of this classification scheme was to offer a system of compromise to reconcile the differences between the bacteriological and botanical groups (Anagnostidis and Komárek, 1985). Within this system, the taxonomic nomenclature is based mainly upon botanical criteria such as cell morphology and life history. Their system uses morphological characteristics such as cell division, polarity, method of false branching, tapering, and hormogonia formation to define and distinguish genera (Anagnostidis and Komárek, 1985). Furthermore, the system proposed by Anagnostidis and Komárek made use of information obtained from bacteriological methods when deemed appropriate (Anagnostidis and Komárek, 1985, 1988, 1990; Komárek and Anagnostidis, 1986, 1989).

However, Anagnostidis and Komárek (1988) caution that the re-classification of cyanobacteria is difficult for a number of reasons. For instance, traditional taxonomic criteria were developed over a century ago and were predicated on erratic characters such as false branching and sheath characteristics. Also, the morphology of cyanobacteria is heavily influenced by environmental factors, and the importance and relevance of environmentally induced phenotypic plasticity was not observed or well understood by early taxonomists. Similarly, the genetic diversity of collected strains within a culture may be decreased by selective culturing conditions (Lu et al., 1997). Additionally, there is a multitude of different species concepts supported by authors of differing backgrounds. Lastly, and most importantly, a substantial number of taxonomists have been very conservative in their approach to adopting new taxonomic criteria. As a result, these taxonomists insist upon using outdated names and terms that are not in line with current taxonomic systems.

The problem in using botanical criteria to classify cyanobacteria is that culturing conditions and environmental plasticity often induce morphological changes (Casamatta and Vis, personal communication). The most plastic characters observed in cyanobacteria include sheath color, sheath thickness, granulation, false branching, and cell length (Anagnostidis and Komárek, 1985). Less variable characters include pigmentation, cell length to width ratios, tapering, trichome width, and the ability to form calyptra (Anagnostidis and Komárek, 1985). Furthermore, morphological characters such as type of cell division and thylakoid structure are constant, and are not influenced by variations in environmental or culturing conditions (Anagnostidis and Komárek, 1985). Culture and environmentally induced morphological changes among cyanobacteria often lead to inaccurate identification and classification taxa (Nelissen et al., 1992; Nübel et al., 1997). Morphological changes can be problematic in establishing species definitions in that they are usually defined based on cell dimensions and ecology, such that distinct separations are often not evident (Nübel et al., 1997). As a result, additional criteria for classifying cyanobacteria are needed.

Currently, there is an increased interest in applying molecular techniques to resolve many of the issues and problems created by the present state of cyanobacterial taxonomy (Giovannoni et al., 1988; Wilmotte and Golubic 1991). Within the last decade, a number of phycologists began using molecular techniques to answer questions dealing with the taxonomy, population dynamics, and the evolution of cyanobacteria. Among the most popular molecular techniques employed by bacteriologists are DNA-DNA hybridization, sequence determination of small ribosomal subunit ribonucleic acids (16S rRNAs), and to a lesser degree, restriction fragment length polymorphism (RFLP) analysis of DNA (Turner, 1997). DNA-DNA hybridization techniques aid bacteriologists in elucidating evolutionary relationships between various strains of cyanobacteria both within and between genera (Turner, 1997). This technique is also recognized as being the most effective method of determining evolutionary relationships between closely related taxa (Stackenbrant and Goebel, 1994). Similarly, RFLPs have been used to resolve differences among serotypes or strains of the same species (Turner, 1997). However, RFLPs are very time consuming and require relatively large amounts of genomic DNA (Li, 2000). Molecular techniques involving ribosomal RNA sequence analysis are commonly used to investigate evolutionary relationships within different genera of cyanobacteria. However, rRNA data may not be very useful in determining lower level taxonomic relationships (Turner, 1997, Fox et al., 1992). Although rRNA sequence data does not provide the taxonomic resolving power seen with RFLPs and DNA-DNA hybridization, it is nonetheless advantageous in that it is a single-step experiment (Turner, 1997). For instance, genetic analyses involving DNA-DNA hybridization and RFLPs require repetitive examination for subsequent comparisons, whereas subsequent examinations are not required once a sequence for an rRNA gene has been deduced (Turner, 1997). In general, information obtained using molecular techniques is very useful in that it provides researchers with a powerful and independent data set in which hypotheses generated from other data, such as morphology and physiology, can be tested (Li, 2000).

Carl Woese and his colleagues were responsible for the initial use of rRNA sequence data to examine evolutionary relationships among bacteria by comparing ribonuclease T₁-generated oligonucleotides (Turner, 1997). The objective of their study was to implement a system of taxonomic classification in bacteria in which comparisons of homologous genes could be made to examine evolutionary relationships (Woese et al., 1975). Subsequently, a number of authors have constructed and published phylogenetic trees based on 16S rRNA gene sequences (Nelissen et al., 1992; Garcia-Pichel et al., 1998; Honda et al., 1998; Otsuka et al., 1998). However, studies examining the16S rRNA gene and 16S-23S rRNA internal transcribed spacer (ITS) sequence data have indicated that morphological characteristics may not coincide with molecular data in cyanobacteria. For instance, based on 16S rRNA gene sequence data, Otsuka et al. (1998) reported that the variation of phycobilin pigment composition observed in 15 strains in 5 different species of Microcystis could possibly be the result of different ecophenes. The results from this study showed little variation in the 16S rRNA gene sequence among strains differing in their phycobilin pigment composition, suggesting that all strains in their study were likely the same species. Otsuka et al. (1998) concluded that the phenotypic characteristics of Microcystis do not reflect their phylogeny, and that the taxonomy of this group needs further revision. In a study conducted by Palinska et al. (1996), five species of Merismopedia were collapsed into a single species (Merismopedia punctata Meyen) based on the percent similarity of their 16S rRNA gene sequence (96-97%).

rRNA operons in prokaryotes such as cyanobacteria encode products that exhibit highly conserved secondary structures (Wheeler and Honeycutt, 1988). It has been demonstrated that bacteria, including cyanobacteria, often contain multiple copies of rRNA operons (Iteman et al., 2000; Li, 2000; Boyer et al., in press). Ribosomal RNA operons encodes three ribosomal RNAs, 16S, 23S, and 5S, and in some cases, three different tRNAs (Iteman et al., 2000). In cyanobacteria, the 16S-23S (ITS) regions (Fig. 1) of different rRNA operons contain zero, one, or two tRNA genes (Iteman et al., 2000). In instances where only one tRNA is present, it is tRNA^Ile; when two tRNAs are present, they are tRNA^Ile and tRNA^Ala (Iteman et al., 2000). The variations in the sequence and lengths of different rRNA operons illustrates that substantial variations can occur between the same genes within a single strain, as well as between species (Iteman et al., 2000). Escherichia coli and Bacillus subtilis, for example, contain seven and ten unidentical copies of the rRNA operon with its associated ITS regions, respectively (Boyer et al., in press). The presence of multiple operons within a strain could affect the interpretation of phylogenetic data obtained using molecular methods. When multiple operons are suspected or detected in a given genome, the sequence for each copy must be determined and analyzed to resolve genetic variation between strains (Li, 2000).

The 16S rRNA gene has been an effective tool in deducing phylogenetic relationships between different genera within orders proposed by Komárek and Anagnostidis (Turner, 1997). Additionally, the 16S rRNA gene has been useful in identifying and classifying strains that belong to a single clade (Palinska et al., 1996; Otsuka et al., 1998). For instance, Nelissen et al. (1992) found that the 16S rRNA sequences of five strains of Pseudanabaena were nearly identical, and therefore concluded that Pseudanabaena was a single monophyletic taxon. In a more recent study, Honda et al. (1998) found that trees constructed using 16S rRNA sequence data resulted in the clustering of members of the genus Synechococcus, indicating that these strains were closely related.

Although 16S rRNA sequence analysis is quite useful in determining evolutionary relationships among organisms at the genus level, it is not, however, very informative when applied to lower levels of taxonomic treatments such as species delineation (Turner, 1997). The sequence of 16S rRNA genes is highly conserved and are more susceptible to intense selective pressure over evolutionary time than most protein-encoding genes (Vinuesa, 1998). This is due to the fact that protein-encoding genes are more tolerable to variations in amino acid codon sequences (silent mutations in the third nucleotide in a codon) than rRNA genes. Moreover, the level of degeneracy in non-protein-coding genes, such as the 16S rRNA gene, is much less than what is observed in protein-coding genes ( ). Due to its conservative nature, the 16S rRNA gene possesses insufficient sequence divergence for comparing species within generic clades (Bolch et al., 1996).

There have been instances in which various protein-encoding genes have been used in the classification of cyanobacterial strains. Bolch et al. (1996) investigated the phylogenetic relationship of three morphospecies of cyanobacteria: Anabaena circinalis, Microcystis aeruginosa, and Nodularia spugmigena. In their study, phycobilisome subunit gene fragments (cpcBA-IGS) were used to classify members of the above species (Bolch et al., 1996). The results from this study provide evidence that DNA fragments produced by RFLP’s provide adequate polymorphic molecular markers that can be used in classifying cyanobacterial strains at the sub-generic level (Bolch et al., 1996).

With the advent of numerous molecular techniques such as polymerase chain reaction (PCR) and automated nucleic acid sequencing technology, large amounts of DNA sequence data have become available. As a result, of our understanding of bacterial evolutionary relationships has been greatly enhanced (Turner, 1997). For example, there is a vast and increasing number of complete or nearly complete 16S rRNA sequence data available in the EMBL sequence database (Wilmotte, 1994). However, 16S rRNA sequence data, in and of itself, is not sufficient in establishing evolutionary relationships among cyanobacteria (Wilmotte and Golubic, 1991; Komárek, 1994). Among distantly related taxa, long branch attraction frequently occurs and can contribute to erroneous phylogenetic conclusions. Among closely related taxa, the 16S rRNA sequence is not sufficiently different to provide clear separation or phylogenetic comparison. As stated earlier, due to environmental influences on morphological characteristics, the strict use of morphological data is also insufficient. The most effective method of determining evolutionary relationships among cyanobacteria is to analyze morphological data in conjunction with molecular data.

Members of the cyanobacteria are generally classified into five orders: Chroococcales, Nostocales, Oscillatoriales, Pleurocapsales, and Stigonematales (Anagnostidis and Komárek, 1988). Members of the order Oscillatoriales are filamentous and lack heterocysts and akinetes (Albertano and Kovacik, 1994; Turner, 1997). Gomont (1892) suggested several criteria for the classification of genera within this group. The Oscillatoriales were separated into 15 genera within 2 tribes, with sheath type and deposition within sheath as the diagnostic criteria (Gomont, 1892). The two tribes included the Lyngbyaceae (filaments with a single trichome) and Vaginariaceae (filaments with multiple trichomes). The order Oscillatoriales includes the families Borziacaea, Homoeotrichaceae, Oscillatoriaceae, Phormidiaceae, Pseudanabaenaceae, and Schizothrichaceae as discussed in Anagnostidis and Komárek, 1988 (Table B-1). Oscillatoria, Lyngbya, Phormidium, Schizothrix, and Plectonema are representative genera in some of these families.

Within the Oscillatoriales, the most problematic group in terms of taxonomic classification is what Drouet called Schizothrix calcicola, a single variable species. Within this species, Drouet (1932) compiled 700 cyanobacterial taxa which included nonheterocystous, nontapering filamentous forms which appeared to be unconstricted to slightly constricted at the crosswalls, with cell dimensions of 0.2-3.5 m m wide and 0.2-6.0 m m long. These strains were found in a number of varying habitats, which included fresh and marine waters, soils, and hot springs.

Later, Baker and Bold (1970) examined 52 isolates within what Drouet considered S. calcicola. In their study, Baker and Bold concentrated on plant mass characteristics and from this work identified and characterized 20 varieties. However, their work was not widely accepted among phycologists due to several flaws inherent in their approach. Included among these flaws was the questionable validity of Drouet’s classification of S. calcicola. Drouet’s treatment of this group was tested and later disproved by a number of workers (Anagnostidis and Komárek, 1985). Also, since the time of the study conducted by Baker and Bold (1970), there had been a shift away from using descriptions of varieties as taxonomic units, as a number of varieties are now considered unique species.

Many of the species within the genera Oscillatoria, Lyngbya, Phormidium, Schizothrix, and Plectonema can be included in Drouet’s treatment of Schizothrix calcicola. These genera were originally classified by Gomont (1892) based on sheath

characteristics and the presence or absence of false branching. However, the sheath characteristics frequently used in identifying and classifying members of this group are influenced by culturing and environmental conditions (Albertano and Kovacik, 1994). This led to the eventual transfer of a number of strains within the Oscillatoriales to a new group designated by Rippka (1979) as LPP-group B (Lyngbya, Phormidium, and Plectonema).

Anagnostidis and Komárek (1988) reassigned many different taxa from the LPP-B group into a new genus Leptolyngbya, thus creating over 75 new combinations (Albertano and Kovacik, 1994; Turner, 1997). This newly established group is typified by the presence of a number of distinguishing morphological characteristics such as thin sheaths, immobility of filaments, thin uniseriate trichomes, arrangement of thylakoids, and cell wall constrictions (Albertano and Kovacik, 1994; Turner, 1997). Anagnostidis and Komárek’s revision of the LPP-B group is based on more of a modern approach than that used by Geitler (1932). However, their view that certain morphological characteristics can be useful in determining evolutionary relationships, such as sheath characteristics and the number of trichomes within a filament, should be reviewed and tested. Interestingly, they retained Schizothrix, defining it narrowly and retaining it as a genus distinct from Leptolyngbya (Anagnostidis and Komárek, 1988).

In reviews of the literature on microbiotic (cryptogamic) crusts (Johansen 1993, Evans and Johansen, 1999), over 30 taxa belonging to Leptolyngbya, Schizothrix, and Phormidium have been reported to occur in desert soils across the western United States. The reviewers conclude that these taxa are possibly multiple ecophenes of the same species. Members of the genus Leptolyngbya, as well as other members within the Oscillatoriales, are in need of subsequent taxonomic evaluation to resolve the present state of disorder and confusion.

The purpose of the present study is to examine the phylogenetic relationship of various strains of Leptolyngbya isolated from arid and semi-arid soils located in the western United States. This thesis addresses the following questions. (1) Are the desert soil strains with thin (1.5-3.5 µm wide) constricted trichomes in one genus, or multiple genera? (2) Is the 16S rRNA gene sequence data consistent with the morphology of this group? (3) How many species do we likely have? (4) Is the genetic diversity (genospecies) higher or lower than the number of taxa recognized by morphology? (5) Is there evidence for multiple 16S rRNA operons in this group? (6) What are the phylogenetic relationships of some Leptolyngbya taxa from desert soils? To answer these questions, 17 different isolates were characterized using various morphological characters. In addition, a portion of the 16S rRNA gene (bp 369-1350) from each isolate was sequenced and analyzed to determine phylogenetic relationships between these isolates. The morphological characteristics examined and described in this study were compared with the resulting molecular data. The results of these analyses are reported in this thesis.

Albertano, D. and Kovacik, L. 1994. Is the genus Leptolyngbya (Cyanophyte) a homogeneous taxon? Algological Studies. 75: 37-51.

Anagnostidis, K. and J. Komárek. 1985. Modern approach to the classification system of the Cyanophytes 1: Introduction. Arch. Hydrobiol. 71, Algological Studies. 38/39: 291-302.

Anagnostidis, K. and J. Komárek. 1988. Modern approach to the classification system of the Cyanophytes 3: Oscillatoriales. Arch. Hydrobiol. 80, Algological Studies. 50-53:327-472.

Anagnostidis, K. and J. Komárek. 1990. Modern approach to the classification system of the Cyanophytes 5: Stigonematales. Algological Studies. 59: 1-73.

Anagnostidis, K. and Komárek. 1999. Cyanoprokaryota. 1. Teil Chroococcales. In: Ettl, H., Gärtner, H. Heynig, D. Mollenhauser (eds.). Süsswasserflora von Mitteleuropa 19/1:1-548.

Baker, A. F. & H. C. Bold. 1970. Taxonomic studies in the Oscillatoriaceae. Phycological studies. 10:1-105. University of Texas Publication No. 7004.

Bolch, C. J. S., Orr, P. T., Jones, G. J. and Blackburn, S. L. 1999. Genetic, morphological, and toxicological variation among globally distributed strains of Nodularia (cyanobacteria). J. Phycol. 35:339-355.

Bornet, E. and Flahault, C. 1888. Revision des Nostocacées. Ann. Sci. Nat., Bot. Ser.7 3:323-381, 4:343-373, 5:51-129, 7:177-262.

Bourrelly, P. 1970. Les algues d’eau douce III. pp 512. N. Boubée and Cie., Paris

Boyer, S. L., Flechtner, V. R., and Johansen, J. R. (in press). Is the 16S-23S rRNA internal transcribed spacer (ITS) region a good tool for use in molecular systematics and population genetics? A case study in cyanobacteria. Mol. Biol. Evol. 18:1057-1069.

Buttars, S. M., St. Clair, L. L., Johansen, J. R., Sray, J. C., Payne, M. C., Webb, B. L., Terry, R. E., Pendleton, B. K., and Warren, S. D. 1998. Pelletized cyanobacterial soil amendments: laboratory testing for survival, escapability, and nitrogen fixation. Arid Soil and Rehabilitation. 12: 165-178.

Castenholz, R. W. and Waterbury, J. B. 1989. Cyanobacteria. Staley, J. T., Bryant, M. P., Pfennig, N., and Holt, J. H. (eds.) Bergey’s Manual of Systematica Bacteriology, Vol. 3. Section 19:Oxygenic photosynthetic bacteria. Williams and Wilkins, Baltimore.

Cohn, F. 1872 Grundzüge einer nuen natürlichen Anordnung der Kryptogamischen Pflanzen. Jarhb. Schles. Ges. Vaterl. KuH. 49:83-89.

Cohn, F. 1875. Untersuchungen über Bakterien. II. Beitr. Biol. Pflanz. 1:141-207.

Desikachary, T. V. 1959. Cyanophyta. Indian Council Agricultural Research, New Delhi. 686 pp.

Drouet, F. 1963. Ecophenes of Schizothrix calcicola (Oscillatoriaceae). Proc. Acad. Nat. Sci. Phila. 115(9):261-281.

Drouet, F. 1968. Revision of the Classification of Oscillatoriaceae. Monographs of the Academy of Natural Sciences, Philadelphia 15:1-370.

Drouet, F. 1973. Revision of the Nostocaceae with Cylindrical Trichomes. Hafner Press, Macmillan, New york. 292 pp.

Drouet, F. 1978. Revision of the Nostocaceae with Constricted Trichomes. Beiheft Nova Hedwigia 57:1-258.

Drouet, F. 1981. Revision of the Stigonemataceae with a summary of the classification of the blue-green algae. Beiheft Nova Hedwigia 66:1-221.

Drouet, F. and Daily, W. A. 1956. Revision of the coccoid Myxophyceae. Butler University Botanical Studies 12:1-218.

Evans, R. D. and Johansen, J. R. 1999. Microbiotic crusts and ecosystem processes. Critical Reviews in Plant Sciences 18:183-225.

Fritsch, C. 1945. The structure and reproduction of algae, vol II. Cambridge Univ. Press, Cambridge.

Fox, G. E., Wisotzkey, J. D., Jurtshuk, J. R. 1992. How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity. Int. J. Syst. Bacter. 42: 166-170.

Garcia-Pichel, F., Nübel, U., and Muyzer, G. 1998. The phylogeny of unicellular, extremely halotolerant cyanobacteria. Arch. Microbiol. 169:469-482.

Geitler, L. 1932. Cyanophyceae. In: Rabenhorst’s Kryptogamenflora von Deutschland, Österreich und der Schweiz 14:1-1196. Reprint by Koeltz Scientific Books, Königstein, Germany.

Giovannoni, S. J., Turner, S., Olsen, G. J., Barns, S., Lane, D. J., and Pace, N. R. 1988. Evolutionary relationships among cyanobacteria and green chloroplasts. Journal of Bacteriology 170 No. 8: 3584-3592.

Golubic, S. 1967. Algenvegetation der Felsen. –Die binnengewässer 23:183 pp., Stuttgart.

Gomont, M. 1892. Monographie des Oscillariées. Ann. Sci. Nat. Ser. Bot. 15:263-368. 16:91-264.

Graham, L.E. and Wilcox, L.W. 2000. Algae. Prentice Hall, Upper Saddle River, NJ. 640 pp. + appendices.

Honda, D., Yokota, A., and Sugiyama, J. 1998. Detection of seven major evolutionary lineages in cyanobacteria based on the 16S rRNA gene sequence analysis with new sequences of five marine Synechococcus strains. J. Mol. Evol. 48: 723-739.

Iteman, I., Rippka, R, Tandeau de Marsac, N, and Herdman, M. 2000. comparison of conserved structural and regulatory domains within divergent 16S rRNA-23S rRNA spacer sequences of cyanobacteria. Microbiology 146:1275-1286.

Johansen, J. R. 1993. Cryptogamic crusts of semiarid and arid lands of North America. Journal of Phycology 29:140-147.

Komárek, J. 1994. Current trends and species delimitation in the cyanoprokaryote taxonomy. In: Cronberg, G., O.M. Skulberg, and T. Willen (eds.), Cyanobacteria, Morphology, Taxonomy, Ecology, Toxicology: Proceedings of the 12^th symposium of the International Association for Cyanophyte Research (IAC), Sjöarp (Sweden) 1992. Algological Studies 75:11-29.

Komárek, J. and Anagnostidis, K. 1986. Modern approach to the classification system of cyanophytes. 2:Chroococcales. Algological Studies 43:157-226.

Komárek, J. and Anagnostidis, K. 1989. Modern approach to the classification system of cyanophytes. 4:Nostocales. Algological Studies 56:247-345.

Li, X. 2000. Phenotypic and genotypic characterization of Nostoc species from 6 different sites in the Mojave desert. Master’s Thesis, John Carroll University.

Lu, W., Evans, E. H., McColl, S. M. and Saunders, V. A. 1997. Identification of cyanobacteria by polymorphisms of PCR-amplified ribosomal DNA spacer regions. FEMS Micorbiology Letters 153:147-149.

Neilan, B. A. 1995. Identification and phylogenetic analysis of toxigenic cyanobacteria by multiplex randomly amplified polymorphic DNA PCR. Applied and Environmental Microbiology. 61: 2286-2291.

Neilen, B. A., Stuart, J. L., Goodman, A. E., Cox, P.T., and Hawkins, P. R. 1997. Specific amplification and restriction polymorphisms of the cyanobacterial rRNA operon spacer region. System. Appl. Microbiol. 20, 612-621.

Nelissen, B., Wilmotte, R., De Baere, R., Haes, F., Van De Peer, Y., Neefs, J. M., and De Wachter, R. 1992. Phylogenetic study of cyanobacteria on the basis of 16S ribosomal RNA sequences. Belg. Journ. Bot. 125: 210-213.

Nübel, U., Garcia-Pichel, F., and Muyzer, G. 1997. PCR primers to amplify 16S rRNA genes from cyanobacteria. Applied and Environmental Microbiology. 63: 3327-3332.

Otsuka, S., Suda, S., Li, R., Watanabe, M., Oyaizu, H., Matsumoto, S., and Watanabe, M. M. 1998. Phylogenetic relationships between toxic and non-toxic strains of the genus Microcystis based on the 16S to 23S internal transcribed spacer sequence.  FEMS Microbiology Letters 172 (1): 15-21.

Palinska, K. A., Liesack, W., Rhiel, E., and Krumbein, W. E. 1996. Phenotype variability of identified genotypes: the need for a combined approach in cyanobacterial taxonomy demonstrated on Merismopedia-like isolates. Arch Microbiol. 166:224-233.

Rippka, R., Deruelles, J., Waterbury, J. B., Herdmann, M. and Stanier, Y. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology 111:1-61.

Rudi, K., Skulberg, O.M., Larsen, F. and Jakobsen, K. S. 1997. Strain characterization and classification Oxyphotobacteria in clone cultures on the basis of 16S rRNA sequences from variable regions V6, V7 and V8. Applied and Environmental Microbiology Vol. 63 (7): 2593-2599.

Stanier, R. Y., Sistrom, W. R., Hansen, T. A., Whitton, B. A., Castenholz, R. W., Pfennig, N., Gorlenko, V.N., Kondratieva, E. N., Eimhjellen, K. E., Whittenbury, R., Gherna, R. L. and Truper, H. G. 1978. Proposal to place the nomenclature of the cyanobacteria (blue-green algae) under the rules of the International Code of Nomenclature of Bacteria. Int. J. Syst. Bacter. 28:335-336.

Thuret, G. 1875. Essay de classification des Nostochinées.- Ann. Sci. Nat. 6, Bot. 1:372-382.

Turner, S. 1997. Molecular systematics of oxygenic photosynthetic bacteria. Plant Syst. Evol., Suppl. 11: 13-52.

Vinuesa, P., Rademaker, J. L. W., de Bruijn, F. J. and Werner, D. 1998. Genotypic characterization of Bradyrhizobium strains nodulating endemic woody legumes of Canary Islands by PCR-restriction fragment length polymorphism analysis of genes encoding the 16S rRNA (the 16S rRNA gene) and the 16S-23S rDNA intergenic spacers, repetitive extragenic palindromic PCR genomic fingerprinting and partial 16S rRNA Gene sequencing. Applied and Environmental Microbiology 64:2096-2104.

Wheeler, W. C. and Honeycutt, R. L. 1988. Paired sequence difference in ribosomal RNAs: Evolutionary and phylogenetic implications. Mol. Biol. Evol. 5(1):90-96.

Wilmotte, A. 1994. Molecular evolution and taxonomy of the cyanobacteria. In: Bryant, D. A. (ed.) The Molecular Biology of Cyanobacteria. Kluwer, Dordrecht, pp. 1-25.

Wilmotte, A., and Golubic, S. 1991. Morphological and genetic criteria in the taxonomy of Cyanophyta/Cyanobacteria. Arch. Hydrobiol. Suppl. 64:1-24. Woese, C. R., Sogin, M. L., Bonen, L., and Stahl, D. 1975. Sequence studies on 16S ribosomal RNA from a blue-green alga. J. Molec. Evol. 4:307-315.

CHAPTER 2: TAXONOMIC RESOLUTION OF LEPTOLYNGBYA UTILIZING THE 16S rRNA GENE SEQUENCE

rRNA operons in prokaryotes such as cyanobacteria encode products that exhibit highly conserved secondary structures (Wheeler and Honeycutt, 1988). It has been demonstrated that bacteria, including cyanobacteria, often contain multiple copies of rRNA operons (Iteman et al., 2000; Li, 2000; Boyer et al., in press). Ribosomal RNA operons encodes three ribosomal RNAs, 16S, 23S, and 5S, and in some cases, three different tRNAs (Iteman et al., 2000). In cyanobacteria, the 16S-23S (ITS) regions (Fig. 1) of different rRNA operons contain zero, one, or two tRNA genes (Iteman et al., 2000). In instances where only one tRNA is present, it is tRNA^Ile; when two tRNAs are present, they are tRNA^Ile and tRNA^Ala (Iteman et al., 2000). The variations in the sequence and lengths of different rRNA operons illustrates that substantial variations can occur between the same genes within a single strain, as well as between species (Iteman et al., 2000). Escherichia coli and Bacillus subtilis, for example, contain seven and ten unidentical copies of the rRNA operon with its associated ITS regions, respectively (Boyer et al., in press). The presence of multiple operons within a strain could affect the interpretation of phylogenetic data obtained using molecular methods. When multiple operons are suspected or detected in a given genome, the sequence for each copy must be determined and analyzed to resolve genetic variation between strains (Li, 2000).

Over 30 taxa belonging to Leptolyngbya, Schizothrix, and Phormidium have been reported to occur in desert soils across the western United States (Johansen 1993, Evans and Johansen, 1999). The reviewers conclude that these taxa are possibly multiple ecophenes of the same species. Members of the genus Leptolyngbya, as well as other members within the Oscillatoriales, are in need of taxonomic evaluation to resolve the present state of disorder and confusion.

The aim of the present study is to examine the phylogenetic relationships of 17 different strains of Leptolyngbya isolated from arid and semi-arid soils located in the western United States. The 17 different desert soil isolates were characterized using various morphological characters. In addition, a portion of the 16S rRNA gene (bp 369-1350) from each isolate was sequenced and analyzed to determine phylogenetic relationships among these isolates. The morphological characteristics examined and described in this study were compared with the resulting molecular data.

The cyanobacterial strains used in this study were collected, isolated and identified by Jocelyn Muller and Marianne Kingsley from semi-arid deserts of the Western United States. Strains were initially grown in liquid Z-8 media (Carmichael, 1986) and were incubated at 20^o C under fluorescent light (200 µE?s^-1?cm^-2) with a 16 hr light/8 hr dark photoperiod. Growing cultures were then transferred to both Z-8 agar plates and slants and were maintained in the above conditions.

All cyanobacterial isolates were examined using a high-resolution Olympus photomicroscope equipped with Nomarski DIC optics to study cellular features. Colony morphology was examined using a stereomicroscope. For each of the 17 strains examined in this study, morphological characteristics such as sheath type, presence or absence of false branching, cell and trichome dimensions, presence of peripheral thylakoids, constrictions to crosswalls, meristematic zones, and shape of end cells were noted. In addition to taking descriptive notes on morphological characteristics, pictures of cells were taken at 400 X for future references.

Total genomic DNA was extracted from cultures using the CTAB method as modified by Cullings (1992) for the isolation and purification of DNA from mucilaginous organisms (Doyle and Doyle, 1987). DNA pellets were re-suspended in 50 m L of TE buffer and the resulting genomic DNA was checked using 1% agarose/ethidium bromide gels. Extracted DNA samples were stored at –20⁰ C.

Gene amplification was accomplished by means of PCR as described in Wilmotte et al. (1993). The PCR primers used in this study included the following:

Primer 1 5’ CTC TGT GTG CCT AGG TAT CC 3’ (after Wilmotte, 1994)

Primer 2 5’ GGG GGA TTT TCC GCA ATG GG 3’ (after Nübel et al., 1997)

Primer 6 5’ GAC GGG CCG GTG TGT ACA 3’ (after Wilmotte, 1994).

The relative locations and position of these primers with respect to the 16S rRNA and 23S rRNA genes are shown in Fig. 1.

Initially, DNA samples were amplified using primers 1 and 2, which are cyanobacterial-specific primers. The resulting amplified products were analyzed on 1% agarose/ethidium bromide gels and were determined to be approximately 1600 base pairs in length ("long PCR"). This product was then used as a template for reamplification using primers 2 and 6 (Fig. 1), which resulted in a product of approximately 900 base pairs in length ("short PCR").

All PCR reactions were performed in a total volume of 100 m L containing 10.0 m L of 10 X Taq polymerase buffer (20 mM Tris HCl, 3 mM MgCl₂, 100 mM KCl); 0.5 m L primer mixture (1.2 m L primer 1 or 6, 1.2 m L primer 2, 7.6 m L dH₂O); 0.5 m L of a stock solution of dNTPs [(10 mM in each dNTP); dATP, dCTP, dGTP, and dTTP]; 0.5 m L Taq polymerase; 1.0 m L of extracted genomic DNA (50 ng), and the appropriate amount of dH₂O to bring the volume to 100m L. The reactions were overlaid with mineral oil, and thermal cycling was conducted using an Amplitron thermal cycler using the following parameters: 94^o C for 60 s, 55^o for 45 s, and 72^o C for 4 minutes repeated for 35 cycles (primer pair 1 and 2), and 94^o C for 60 s, 55^o for 45 s, and 72^o C for 2 minutes repeated for 20 cycles (primer pair 2 and 6). After amplification, a 7-minute/72^o extension step was included for primer pair 1 and 2, whereas primer pair 2 and 6 received no such extension. PCR products were analyzed on 1% agarose/ethidium bromide gels in 1X TBE buffer.

Amplified PCR products were cloned into pCR 4-TOPO plasmids containing sites for universal primers M13 forward and reverse using the TOPO^TM TA cloning kit (Invitrogen®). After transformation, E. coli cells (Invitrogen®) were plated onto Luria Broth plates containing 100 mg/L of ampicillin. Additionally, 40 m L of Xgal (20 mg/ml) and 4 m L of IPTG (200 mg/ml) were spread onto the plates prior to the plating of transformed cells. Plates were incubated at 37^o C overnight. On the following day, plates were refrigerated at 4^oC for at least 2 to 4 hours to enhance the detection of blue colonies. Clones containing the cloned fragment appeared as white colonies, whereas clones lacking the gene insert appeared as blue colonies. Colonies containing the gene insert were selected and cultured in 4 ml of LB broth containing 8 µL of ampicillin (50 mg/ml) at 37^oC overnight.

Cultured E. coli cells containing cloned inserts were harvested by centrifugation at 17000 rpm for 10 minutes. Plasmids were isolated according to the instructions provided in the QIAprep Mini-prep kit. To verify the presence of a cloned insert in clones, plasmid DNA was digested using EcoR I. Two restriction sites for EcoR I flank both sides of the cloning site in the pCR-4-TOPO vector. Digests were resolved on 1 % agarose/ethidium bromide gels to detect plasmid inserts.

Two replicate plasmid samples were isolated from each cloning plate and sequenced by Cleveland Genomics. Automated sequencing was performed using universal infrared (IR) primers M13IR forward and reverse.

Reverse sequences were converted to their respective complementary forward sequence using Omiga^TM. Forward and reverse sequences were aligned using the CLUSTAL W Multiple Sequence Alignment Program (Thompson et al., 1994). The resulting sequence alignments were checked by eye for ambiguities and PCR errors by the examination of chromatograms, with corrections made where appropriate. Corrected sequences were aligned with published Leptolyngbya sequences (GenBank X84810 and GenBank X84808, referred to in this paper as LEBO GB and LEFO GB, respectively) to compare sequence identity. Three strains each of Microcoleus vaginatus and Microcoleus steenstrupii were used as outgroups. All aligned sequences were analyzed using PAUP* 4.0b (Swofford 1998) to conduct heuristic searches (100 replicates each) for most parsimonious trees (Swofford 1998). The following analyses were completed: 1) neighbor-joining using logdet distance metric and equal substitution rates, 2) neighbor-joining using HKY85 distance metric and equal substitution rates, 3) neighbor-joining using HKY85 distance metric and gamma substitution rates, 4) neighbor-joining using Jukes-Cantor distance metric and gamma substitution rates, 5) maximum parsimony using logdet distance metric and equal substitution rates, 6) maximum likelihood using HKY85 distance metric and equal substitution rates. Support for each tree was subsequently obtained by running bootstrap analyses with 1000 replicates each. All trees were unrooted. A similarity matrix for all strains was constructed using MEGA to compare sequence identity. The tree reported has the maximum likelihood topology, with branch lengths adjusted for distance using an unweighted group average algorithm (Pielou 1984) for averaging distance among clusters.

All sequences were submitted to GenBank. A full alignment of these sequences is given in Appendix A. An abbreviated alignment was generated from the full alignment by removing all invariant base pairs from the 1042 bp alignment. This was done to allow easier recognition of consistent patterns within the 16S rRNA gene.

The morphological data obtained in this study provided insufficient basis for the proper placement of these strains within presently established species reported in the literature. The morphological characters used in describing these strains were very similar across strains. There were relatively few characters, and many of them, including cell length, trichome width, presence of meristematic zones, and degree of constriction at the crosswalls varied continuously (Table 1). This overlap would make it very difficult to write a key that would reliably separate the morphologically similar strains.

Despite the high degree of overlap of morphological characteristics in this study, there were minor differences observed in all strains (Appendix B). Therefore, these strains could not be combined into the same morphospecies. The species descriptions of Lyngbya, Oscillatoria, Phormidium, Plectonema, and Schizothrix in Geitler (1932) are incomplete by today’s standards, and it is possible to put most of the strains under study into species in this treatise if a generous standard of fit is used. However, in using the most current taxonomic treatment of Leptolyngbya by Komárek (unpublished), only strains SEV4-3-c3 (Phormidium diguetti), SEV1-1-c2 (Leptolyngbya crispata var. 1), and SEV4-3-c6 (Leptolyngbya crispata var. 2), properly matched current morphospecies descriptions. Based on subsequent molecular analysis that indicated that SEV4-3-c3 was not a Phormidium, it was called Leptolyngbya sp. 7 in this treatment to reflect its correct generic position.

Recently, there has been emphasis placed upon using biotope as an important taxonomic character. Thus, specimens isolated from a sulfur hot spring would not be given the same name as a species originally described from alpine lakes, even if the morphology were indistinguishable. None of the cyanobacterial taxa in Geitler (1932) were originally described from deserts, and thus it is very likely that all strains isolated from American desert soils in this study are new to science, and will eventually require descriptions as new species.

Partial 16S rRNA sequences from a total of 17 strains were obtained. Sequence data for some strains were not obtained (not reported here) and in some instances only one clone in a strain was successfully sequenced. In some cases, sequencing yielded ambiguous results. Finally, some bacterial 16S rRNA sequences were recovered from the cloned PCR product. All sequences were used in BLAST search of GenBank to ensure that they were Oscillatorialian before subsequent analyses were undertaken. Only one strain, Leptolyngbya sp. X, contained an 11 pb insert between base pair positions 93 and 104. This 11 bp insert has been observed in Microcoleus vaginatus (Boyer, unpublished data) (Fig. 2).

Maximum likelihood analysis of the 16S rRNA gene sequence data resulted in a single tree illustrating phylogenetic relationships among all strains in this study (Fig. 3). The topology and branch support observed in the maximum likelihood tree was nearly identical to trees generated using distance and parsimony as optimality criteria (Appendix C). The majority of the strains described as members of the genus Leptolyngbya are clustered into a highly supported clade (90% bootstrap value) consisting of L. crispata and Leptolyngbya spp. 1, 2, 3, 4, 5, 6, 7, 8. Aquatic strains obtained from GenBank, L. boryanum and L. foveolarum, were closely related to taxa within the Leptolyngbya clade (Fig. 3). Leptolyngbya sp. X, was included in the clade containing Microcoleus vaginatus strains obtained from Boyer (unpublished data) (Fig. 3).

The maximum likelihood tree (Fig. 3) also shows that clones from the same strain have conserved sequence homologies. In most cases, clones obtained from the same strain are adjacent to each other within the tree. However, there are instances where this does not occur. For example, the multiple clones from Leptolyngbya spp. 2, 4, 5, and 6 come out near, yet not directly adjacent to each other. The initial hypothesis generated to explain this phenomenon revolved around the idea of PCR error. To address this issue, an attempt was made to construct consensus sequences for each strain by correcting ambiguous PCR errors. However, in examining the sequence alignment data (Fig. 2), it became evident that the observed nucleotide differences were not the result of random PCR amplification errors. The base pair differences depicted in the alignment appear to follow distinct and consistent patterns, as opposed to random variation. For example, at bp 510-512, the sequences have either ATC or TGT, and this is a pattern consistent throughout most strains. In the one strain for which four clones were obtained (Leptolyngbya sp. 12), one of the three clones was substantially different from all other strains in one short segment (bp 496-504). These patterns suggest the presence of multiple operons.

To compare sequence similarity of the strains used in this study, a distance matrix was constructed (Appendix D). Strains within the Leptolyngbya clade (Fig. 3, L) all have 16S rRNA sequence similarity values above 95%, suggesting that this particular clade represents one genus. They differ markedly from the Microcoleus vaginatus clade (86.2-90.7% similarity). The strains at the base of the clade, Leptolyngbya spp. 12, 13, 14 are distinct from Microcoleus vaginatus (89.9-90.2% similar), Microcoleus steenstrupii (88.6-91.1%), and the upper Leptolyngbya clade (86.5-91.4%).

The aim of this study was to resolve the taxonomic relationship among 17 strains of Leptolyngbya isolated from three desert sites in western North America using 16S rRNA sequence data. Based on the 16S rRNA sequence data obtained in this study, there is strong evidence that Leptolyngbya is not monophyletic, although a major clade of morphologically and genetically similar strains does exist within this group (Fig. 3).

Based on the 16S rRNA sequence data, the resulting cladogram (Fig. 3) shows that Leptolyngbya sp. X, although morphologically similar to other members of Leptolyngbya, clusters with the Microcoleus vaginatus clade. This finding suggests that Leptolyngbya sp. X is a different species within the genus Microcoleus. The overall sequence similarity among strains in the Microcoleus clade is 98.8-99.3%, while Leptolyngbya sp. X shares 97.6-98.2% similarity with those strains. Consequently, Leptolyngbya sp. X requires further examination, and should be described as a new species of Microcoleus. These results provide a vivid example of how morphological and molecular data can offer contradictory results and conclusions in regard to the phylogenetic relationships of cyanobacteria.

The cladogram (Fig. 3) shows that Leptolyngbya sp. 13 and 14 are not included within the major Leptolyngbya clade, but rather are clustered together at the bottom of the tree. Their position indicates that they have a great deal of genetic distance from all other strains examined in this study. Based on their overall similarity to species within the major Leptolyngbya clade (86.5-91.4%), Leptolyngbya spp 13 and 14 likely represent a different genus. Moreover, these species are very similar to each other (98.0-98.2%).

In the case of Leptolyngbya sp. 12, however, the 16S rRNA sequence data was congruent with the morphological data. Leptolyngbya sp. 12 was the only strain in which the morphological character of multiple trichomes in a wide, closed sheath was observed. This is the diacritical character of the genus Schizothrix. Based on the cladogram (Fig. 3), Leptolyngbya sp. 12 is paraphyletic to the main Leptolyngbya clade, and should therefore be placed in a separate genus. Although sheath characteristics have been largely dismissed by many microbiologists (Drouet, 1968; Castenholz and Waterbury, 1989), Komárek and Anagnostidis (1988) claim that multiple trichomes in a large closed sheath is a valid character, and define not only the genus Schizothrix, but the whole family Schizothricaceae. The results presented here strengthen the claim that Schizothrix is a valid genus, and not just an assemblage of Leptolyngbya and Phormidium species. Currently, there are no recognized Schizothrix in the PCC or ATCC culture collections, and no sequence data for the genus are available in GenBank. In the future, we plan to examine the thylakoid structure for our putative Schizothrix strain in the transmission electron microscope. This new information would provide researchers with data that can be used to make comparisons with members of the Pseudanabaenaceae and Phormidiaceae, which differ in thylakoid structure, and would help to resolve the phylogentic placement and potential validity of Schizothrix.

The 16S rRNA sequence data obtained in this study provide evidence for multiple operons in Leptolyngbya. There have been other reports of multiple ribosomal operons in cyanobacteria in the literature. For instance, Li (2000) reported the presence of four distinct 16S-23S internal transcribed spacer (ITS) regions representing four different operons in 15 strains of Nostoc. Li (2000) found that the ITS regions contained no tRNA genes, both tRNA^Ile and tRNA^Ala genes, or fragments of both tRNA genes. Iteman et al. (2000) conducted a study that examined the 16S-23S ITS region of Nostoc PCC 7120. They found two sequence patterns (no tRNA genes, and two tRNA genes), also demonstrating multiple operons.

Although the operon variability observed in this study was relatively minor, the occurrence of multiple operons in phylogenetic analyses can be problematic. For instance, Boyer et al. (in press), caution that any work consisting of PCR amplification of the 16S-23S ITS region is subject to preferential amplification within some operons more than others. In interpreting sequence data of this nature, it becomes difficult to assess sequence homology. This posses serious problems in determining phylogenetic relationships among taxa using this data. More importantly, it is inappropriate to conduct phylogenetic analyses with non-homologous traits.

The information provided by the cladogram (Fig. 3), indicates that there are possibly 12-13 different species represented in this study. The morphological data do not support the species diversity represented in the cladogram. Within the top clade of the tree (Fig. 3), there is evident morphological variability among the four strains making up the clade. This supports the claim that the 16S rRNA gene is too conservative to separate genetically similar species (Fox et al., 1992). The cladogram presented in this study shows a substantial number of monophyletic taxa based on 16S rRNA sequences. Additionally, there is a distinct possibility that these monophyletic clades likely harbor multiple species, which may or may not differ morphologically.

In terms of morphology, the data obtained in this study were insufficient in delineating species. The morphological characters used in describing the 17 strains in this study did not offer sufficient means for accurate identification at the species level. Taxa described in Geitler (1932) were originally described from temperate humid climates, mostly from aquatic habitats. They were described from field material without culturing. Finally, the descriptions are often fairly short and ambiguous or silent on a number of morphological features considered useful today (such as type of cell division or thylakoid structure). Geitler’s taxonomic treatment of many strains of cyanobacteria consequently consisted of limited, simple, and overlapping characteristics, such as cell dimensions, false branching, and the presence of a sheath. As a result of the limited characteristics used in Geitler’s taxonomic treatments, many cyanobacterial strains from diverse habitats and parts of the world can be placed within established European taxa if biogeography and ecology are ignored. In the past, the majority of species names applied to desert soil cyanobacteria were derived from morphologically similar aquatic species or terrestrial species from temperate European climates. However, if ecology, physiology and biogeography are considered, along with the fact that species often lose sheath characteristics in culture, Geitler’s (1932) taxonomy quickly becomes inadequate. Many different strains of Leptolyngbya, as well as other cyanobacteria, may be inaccurately and incompletely classified. As a result, their alpha level taxonomy is in need of revision.

Komárek (unpublished) has provided a more recent treatment of the genus Leptolyngbya. This treatment includes a larger variety of morphological characteristics to describe and classify species. In using the morphological criteria set forth by Komárek, it was not possible to confidently identify most of the strains in this study at the species level.

Additional work involving other character sets needs to be done to resolve some of the current confusion in the alpha-level taxonomy of Leptolyngbya. The 16S-23S ITS sequence is more variable than the 16S rRNA gene and could possibly provide more of an accurate depiction of the phylogenetic relationship among strains and species of Leptolyngbya. However, using the highly variable 16S-23S ITS region to determine phylogenetic relationships is difficult due to the multiple operon dilemma. Likewise, examining the thylakoid structure of members of Leptolyngbya could elucidate the presence of species in genera which resemble Leptolyngbya, but which are phylogenetically distinct. The thylakoid structure of cyanobacteria is not sensitive to varying environmental and culturing conditions, and therefore provides a stable character that can be used to determine taxonomic relationships among thin-trichomed members of the Oscillatoriales.

Albertano, D. and Kovacik, L. 1994. Is the genus Leptolyngbya (Cyanophyte) a homogeneous taxon? Algological Studies. 75: 37-51.

Anagnostidis, K. and J. Komárek. 1988. Modern approach to the classification system of the Cyanophytes 3: Oscillatoriales. Arch. Hydrobiol. 80, Algological Studies. 50-53:327-472.

Boyer, S. L., Flechtner, V. R., and Johansen, J. R. (in press). Is the 16S-23S rRNA internal transcribed spacer (ITS) region a good tool for use in molecular systematics and population genetics? A case study in cyanobacteria. Mol. Biol. Evol.

Castenholz, R. W. and Waterbury, J. B. 1989. Cyanobacteria. Staley, J. T., Bryant, M. P., Pfennig, N., and Holt, J. H. [Eds.] Bergey’s Manual of Systematica Bacteriology, Vol. 3. Section 19:Oxygenic photosynthetic bacteria. Williams and Wilkins, Baltimore.

Cullings, K. W. 1992. Simplified Doyle and Doyle extraction procedure. Molecular Ecology 1:233-240.

Doyle, J. J. and Doyle, J. L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19:11-15.

Drouet, F. 1963. Ecophenes of Schizothrix calcicola (Oscillatoriaceae). Proc. Acad. Nat. Sci. Phila. 115(9):261-281.

Drouet, F. 1968. Revision of the Classification of Oscillatoriaceae. Monographs of the Academy of Natural Sciences, Philadelphia 15:1-370.

Evans, R. D. and Johansen, J. R. 1999. Microbiotic crusts and ecosystem processes. Critical Reviews in Plant Sciences 18:183-225.

Fox, G. E., Wisotzkey, J. D., Jurtshuk, J. R. 1992. How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity. Int. J. Syst. Bacter. 42: 166-170.

Garcia-Pichel, F., Nübel, U., and Muyzer, G. 1998. The phylogeny of unicellular, extremely halotolerant cyanobacteria. Arch. Microbiol. 169:469-482.

Geitler, L. 1932. Cyanophyceae. In: Rabenhorst’s Kryptogamenflora von Deutschland, Österreich und der Schweiz 14:1-1196. Reprint by Koeltz Scientific Books, Königstein, Germany.

Giovannoni, S. J., Turner, S., Olsen, G. J., Barns, S., Lane, D. J., and Pace, N. R. 1988. Evolutionary relationships among cyanobacteria and green chloroplasts. Journal of Bacteriology 170 No. 8: 3584-3592.

Gomont, M. 1892. Monographie des Oscillariées. Ann. Sci. Nat. Ser. Bot. 15:263-368. 16:91-264.

Honda, D., Yokota, A., and Sugiyama, J. 1998. Detection of seven major evolutionary lineages in cyanobacteria based on the 16S rRNA gene sequence analysis with new sequences of five marine Synechococcus strains. J. Mol. Evol. 48: 723-739.

Iteman, I., Rippka, R, Tandeau de Marsac, N, and Herdman, M. 2000. Comparison of conserved structural and regulatory domains within divergent 16S rRNA-23S rRNA spacer sequences of cyanobacteria. Microbiology 146:1275-1286.

Johansen, J. R. 1993. Cryptogamic crusts of semiarid and arid lands of North America. Journal of Phycology 29:140-147.

Li, X. 2000. Phenotypic and genotypic characterization of Nostoc species from 6 different sites in the Mojave desert. Master’s Thesis, John Carroll University.

Nelissen, B., Wilmotte, R., De Baere, R., Haes, F., Van De Peer, Y., Neefs, J. M., and De Wachter, R. 1992. Phylogenetic study of cyanobacteria on the basis of 16S ribosomal RNA sequences. Belg. Journ. Bot. 125: 210-213.

Nübel, U., Garcia-Pichel, F., and Muyzer, G. 1997. PCR primers to amplify 16S rRNA genes from cyanobacteria. Applied and Environmental Microbiology. 63: 3327-3332.

Otsuka, S., Suda, S., Li, R., Watanabe, M., Oyaizu, H., Matsumoto, S., and Watanabe, M. M. 1998. Phylogenetic relationships between toxic and non-toxic strains of the genus Microcystis based on the 16S to 23S internal transcribed spacer sequence.  FEMS Microbiology Letters 172 (1): 15-21.

Palinska, K. A., Liesack, W., Rhiel, E., and Krumbein, W. E. 1996. Phenotype variability of identified genotypes: the need for a combined approach in cyanobacterial taxonomy demonstrated on Merismopedia-like isolates. Arch Microbiol. 166:224-233.

Pielou, E. C. 1984. The Interpretation of Ecological Data: A Primer on Classification and Ordination. John Wiley and Sons, New York. 1-237 pp.

Rippka, R., Deruelles, J., Waterbury, J. B., Herdmann, M. and Stanier, Y. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology 111:1-61.

Swofford, D.L. 1998. PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, Massachesetts.

Thompson, J. D., Higgins, D. G., and Gibson, T. J. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignments through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.

Turner, S. 1997. Molecular systematics of oxygenic photosynthetic bacteria. Plant Syst. Evol., Suppl. 11: 13-52.

Wheeler, W. C. and Honeycutt, R. L. 1988. Paired sequence difference in ribosomal RNAs: Evolutionary and phylogenetic implications. Mol. Biol. Evol. 5(1):90-96.

Wilmotte, A. 1994. Molecular evolution and taxonomy of the cyanobacteria. In: Bryant, D. A. (ed.) The Molecular Biology of Cyanobacteria. Kluwer, Dordrecht, pp. 1-25.

Wilmotte, A., Van der Auwera, G. and De Wachter, R. 1993. Structure of the 16S ribosomal RNA of the thermophilic cyanobacterium Chlorogloeopsis HTF (‘Mastigocladus laminosus HTF’) strain PCC7518, and phylogenetic analysis.  FEBS Microbio. Let. 317:96-100.

Woese, C. R., Sogin, M. L., Bonen, L., and Stahl, D. 1975. Sequence studies on 16S ribosomal RNA from a blue-green alga. J. Molec. Evol. 4:307-315.