in U.S.A. Purification and Properties of Unicellular Blue-Green … · STANIER ET AL. be recognized as a major group of bacteria, distinguished from other photosynthetic bacteria

BACTEROLOGICAL REVIEWS, June 1971, p. 171-205Copyright © 1971 American Society for Microbiology

Vol. 35, No. 2Printed in U.S.A.

Purification and Properties of Unicellular Blue-GreenAlgae (Order Chroococcales)

R. Y. STANIER, R. KUNISAWA, M. MANDEL, AND G. COHEN-BAZIRE

Department of Bacteriology and Immunology, University of California, Berkeley, California 94720, and TheUniversity of Texas, M. D. Anderson Hospital and Tumor Institute, Houston, Texas 77025

INTRODUCTION............................................................ 171

MATERIALS AND METHODS............................................... 173

Sources of Strains Examined.................................................. 173

Media and Conditions of Cultivation........................................... 173

Enrichment, Isolation, and Purification of Unicellular Blue-Green Algae 176

Measurement of Absorption Spectra............................................ 176

Determination of Fatty-Acid Composition....................................... 176

Extraction of DNA.......................................................... 176

Determination of Mean DNA Base Composition................................. 177

ASSIGNMENT OF STRAINS TO TYPOLOGICAL GROUPS 177

DNA BASE COMPOSITION................................................. 181

PHENOTYPIC PROPERTIES................................................ 184

Motility................................................................... 184

Growth in Darkness and Utilization of Organic Compounds........................ 184

Nitrogen Fixation............................................................ 191Maximal Temperature for Growth............................................. 191

Sensitivity to Light ......................................................... 192Penicillin Sensitivity.......................................................... 192Cellular Absorption Spectra................................................... 192

Cellular Fatty-Acid Composition ....... ....................................... 193PROBLEM OF GENERIC NOMENCLATURE IN THE ORDER CHROOCOC-

CALES.................................................................. 194

POSSIBLE CORRESPONDENCES BETWEEN TYPOLOGICAL GROUPS ANDTRADITIONAL FORM GENERA IN CHROOCOCCACEAE 195

MERISMOPEDIA: A GROWTH FORM OF APHANOCAPSA? ................ 197

GENERA CHLOROGLOEA AND CHLOROGLOEOPSIS ...................... 199

SPECIATION IN CHROOCOCCACEAE ....................................... 199

Subdivisions in Typological Group I (Synechococcus) ......... .................... 199Subdivisions in Typological Group II (Aphanocapsa, Gloeocapsa, Microcystis) ... .... 201

CONCLUDING COMMENTS 202

LITERATURE CITED........................................................ 203

To the memory of E. G. Pringsheim(1881-1970).

INTRODUCTIONBlue-green algae occupy an anomalous position

in the biological world. They are treated bybotanists as a division (or class) of algae becausethey are photoautotrophs that use water as anelectron donor and contain the two photopig-ments (chlorophyll a and (-carotene) that arethe chemical hallmarks of plant photosynthesis.However, it has long been recognized that incellular and organismal respects they resemble thebacteria. This was first noted by Cohn (14), whosubsequently proposed (15) to place the twogroups in a common division, the Schizophytae,even though their shared properties then appearedto be negative ones. The problem of defining inpositive terms the properties that are shared bybacteria and blue-green algae remained intractable

for many decades (16, 58, 69, 74) and was solvedonly when the structure of the cell in these twogroups could be studied at the level of resolutionafforded by the electron microscope. Electronmicroscopic investigations, together with analyti-cal data on cell wall composition and ribosomalstructure, have now revealed the common de-nominators, which are fundamental: bacteria andblue-green algae are the only organisms with cellsof the procaryotic type (6, 10, 21, 47, 70).The designation blue-green algae is therefore

misleading, although this common name is nowso firmly established that its use can probablynever be eradicated. These organisms are notalgae; their taxonomic association with eucaryoticalgal groups is an anachronism, formally equiva-lent to classifying the bacteria as a constituentgroup of the fungi or the protozoa. In view oftheir cellular structure, blue-green algae can now

171

on February 29, 2020 by guest

http://mm

br.asm.org/

Dow

nloaded from

http://mmbr.asm.org/

STANIER ET AL.

be recognized as a major group of bacteria,distinguished from other photosynthetic bacteriaby the nature of their pigment system and bytheir performance of aerobic photosynthesis.Their properties are still very poorly known.Much of the available information about themlies in the domain of natural history, not biology.The structural diversity of blue-green algae is

considerable, and for this reason algologists havebeen able to develop elaborate taxonomic treat-ments of the group based solely on structuralproperties, determined through microscopicexamination of field collections (19, 28). Mostalgal taxonomists recognize several differentorders. The unicellular blue-green algae areassigned to a single order, Chroococcales, con-taining two families, Chroococcaceae and Ento-physalidaceae (28). Members of the Chroococ-caceae are coccoid or rod-shaped organisms whichmultiply by binary fission, and they may formloose colonies in which the constituent cells are

held together by a common slime layer or bysheaths. In structural terms, they appear to becounterparts of unicellular true bacteria. Mem-bers of the Entophysalidaceae are coccoid or-

ganisms, distinguished primarily by their colonialgrowth habit. They grow as dense, parenchyma-tous masses of cells, typically found on the sur-face of moist rocks.

In the study of any microbial group, the transi-tion from natural history to biology is largelydependent on the solution of a technical problem:the isolation and maintenance of its members inpure culture. The present paucity of informationabout blue-green algae reflects the difficulty thathas been encountered in the purification of thesemicroorganisms. The total number of strainspurified so far is small, and the pure strains nowavailable are far from representative of the evi-dent biological diversity of the group. Never-theless, the nutrient requirements of these or-

ganisms are simple, and it is easy to obtain growthon solid media. The primary obstacle to obtainingpure cultures seems to lie in the synthesis by manyblue-green algae of a copious extracellular slimelayer, which harbors bacteria.Some filamentous blue-green algae have been

purified by taking advantage of their vigorousgliding movement and inducing phototacticmigration of trichomes or hormogonia over (orthrough) agar plates (41). This does not alwayseliminate bacteria, however, and purification ofsome filamentous strains has been achieved onlyby treatment of trichomes or resting structureswith ultraviolet light (9), toxic chemicals (27),or heat (75).

In principle, the purification of unicellularblue-green algae should be simpler than that of

filamentous forms, since most of the unicellularmembers of the group are immotile and, therefore,will form small, compact colonies on solid media.The most successful pioneer in the purification ofblue-green algae, M. B. Allen (1), isolated oneunicellular strain, the organism known as Ana-cystis nidtilans, by standard plating methods. Herstrain remained for many years the sole repre-sentative of the Chroococcales available in pureculture, and for this reason it has been widely usedin physiological and biochemical investigations(see 33). A few additional unicellular strains havebeen purified from marine sources by Van Baalen(73), and Castenholz (11) has purified severalthermophilic unicellular strains from hot springs.Our work on this problem began about 8 years

ago when M. M. Allen (2) succeeded in purifyingby plating methods some unicellular strains thathad been received as unialgal cultures from otherlaboratories. This convinced us that purificationis not difficult, provided that suitable material isavailable in the form of unialgal cultures or crudecultures that contain a preponderance of uni-cellular blue-green algae. However, these or-ganisms typically occur in nature as minor com-ponents of mixed algal populations; since theirnutritional requirements do not differ signifi-cantly from those of many other algae, theycannot be specifically enriched from such mixedpopulations by nutritional selection. Naturalpopulations that consist largely of unicellularblue-green algae may arise sporadically as waterblooms in lakes or ponds (28) but are of rela-tively constant occurrence only in hot springs,where they are composed of a few physiologicallyhighly specialized thermophilic forms (11).

Allen and Stanier (5) discovered that manymesophilic blue-green algae, both filamentousand unicellular, can be enriched successfullyfrom mixed algal populations by temperatureselection, since their temperature maxima aresignificantly higher than those of nearly alleucaryotic algae from the same environments.When samples of soil or fresh water are placedin a suitable mineral medium and incubated in thelight at 35 C, the population that develops con-sists almost exclusively of blue-green algae. Thisenrichment technique is particularly valuable forthe isolation of unicellular blue-green algae be-cause, if present in the inoculum, they eventuallyfar outnumber filamentous members of the groupas the enrichment culture develops. Experiencewith its use for several years has shown that oneor more kinds of unicellular blue-green algae arepresent in almost every natural sample of freshwater, even if no algae can be detected in them bydirect microscopic examination. M. M. Allen(2) purified several unicellular strains from fresh-

172 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from


UNICELLULAR BLUE-GREEN ALGAE

water enrichments, and we have subsequentlyisolated many more strains by the same method.Our collection of pure unicellular blue-green

algae now consists of more than 40 strains. Inaddition to our own isolates, it includes purecultures previously isolated by M. B. Allen (1),Van Baalen (73), and Castenholz and his col-laborators (57, 65), as well as a number of strainsreceived from other investigators in the unialgalstate and purified by us. Although this collection iscertainly not yet representative of the entire orderChroococcales, its size and diversity are nowsufficient to permit a preliminary comparativesurvey of the properties of unicellular blue-greenalgae.The dependence on structural characters alone

for the classification of blue-green algae hascaused particular difficulties in the Chroococcales,the simplest members of the group in structuralrespects. Our initial goal was, accordingly, toexamine other phenotypic traits, determinableonly with pure cultures, which might prove to beof taxonomic value. At the same time, we deter-mined the deoxyribonucleic acid (DNA) basecomposition of every strain because the one pre-vious study of this genetic character in blue-greenalgae (22) indicated that it might be particularlyvaluable for the subdivision of the Chroococcales.Edelman et al. (22) determined DNA base

compositions for 29 strains of blue-green algae-probably a large fraction of all strains then avail-able in pure culture. The material was biased infavor of filamentous representatives, since 22strains belonged to the Nostocaceae, Oscilla-toriaceae, Scytonemataceae, and Rivulariaceae.Among these strains, the range of DNA basecomposition was small and showed no clear-cuttaxonomic correlations; the extreme limits were39 and 51 moles % guanine and cytosine (GC),and for 20 strains the values lay between 42 and48 moles % GC. Only six strains of the orderChroococcales were analyzed, but their range ofbase composition extended from 35 to 71 moles %GC, a span little narrower than that found in theentire group of procaryotes, about 25 to 75 moles% GC (30).DNA base composition is now recognized to

be a character of major taxonomic significanceamong bacteria; it has been found that thetaxonomic subgroups long recognized on purelyphenotypic grounds are, in most cases, alsocharacterized by a relatively narrow and distinc-tive span of DNA base composition (30). Thesituation which presented itself in Chroococcaleswas somewhat different, since the phenotaxonomyof these organisms has remained at the level whichhad been attained, with respect to the bacteria,in the era of Ferdinand Cohn (15). It seemed

possible, therefore, that the major internaldivergences of DNA base composition in Chro-ococcales revealed by the work of Edelman et al.(22) might provide primary clues for the develop-ment of a more satisfactory system of classifica-tion and actually direct the search for taxonomi-cally useful phenotypic characters.

MATERIALS AND METHODSSources of Strains Examined

Strain histories are summarized in Tables 1and 2. In these tables, the strains are arrangedaccording to typological groups, of which thedistinguishing properties are described in a latersection (see Table 4). We have noted the namesunder which strains from other laboratories werereceived, accompanied by a parenthetic G(Geitler, reference 28) or D (Drouet and Daily,reference 20) to designate the nomenclaturalsystem reflected by each name. The state of eachof these strains when it was received in theBerkeley culture collection is indicated by thesymbols P (pure culture) or U (unialgal culture).Strains found on receipt to be unialgal weresubsequently purified by us.

Media and Conditions of CultivationWith the exception of two marine strains

(7002 and 7003) previously shown by Van Baalen(73) to require vitamin B12, all strains examinedcan grow in strictly mineral media. The me-dium which we have adopted [devised by M. M.Allen (3)] for general use (BG-11) is a slightmodification of medium G-11 of Hughes, Gor-ham, and Zehnder (36). Its composition is shownin Table 3. For the preparation of solid media, itis supplemented with 1 % (w/v) Difco agar,sterilized separately from the mineral base (bothat double strength).Medium BG-11 is neutral after sterilization. It

will support growth either in air or in an at-mosphere slightly enriched with CO2. Forroutine maintenance, cultures are incubated inair, but more rapid growth and higher yields areobtainable if cultures are continuously aeratedwith N2 containing 0.5 to 1.0% (v/v) CO2. Allour unicellular strains can grow well in BG-11.A supplement of vitamin B12 (final concentration1 ,ug/liter) is added for the cultivation of strains7002 and 7003, and the latter strain also requiresthe addition of NaCl (1%, w/v) for satisfactorygrowth.Medium BG-1 1 has a low phosphate content

and is poorly buffered. Several other media whichhave been used to cultivate blue-green algae arebetter buffered and provide a larger phosphatesupply-e.g., medium C of Kratz and Myers (42)

VOL. 35, 1971 173


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

TABLE 1. Pure strains of unicellular blue-green algae: typological group I

Typo- Ber-Reerlogical keleName (if any) and other Isolator Subsequent strain historya Refer-strain designationsgop no.

Anacystis nidulans (D)

Berkeley isolateBerkeley isolateSynechococcus lividus (G)

Agmenellum quadruplicatum(D) PR-6

Coccochloris elabens (D) 17-ASynechococcus sp. (G) Cl 53Synechococcus sp. (G) Y 52Synechococcus cedrorum (G),IUCC 1911

Berkeley isolateGloeocapsa alpicola (D) 1051Coccochloris peniocyslis (D)Berkeley isolateBerkeley isolateBerkeley isolateBerkeley isolateBerkeley isolateBerkeley isolateBerkeley isolateSynechococcus elongatus (G)IUCC 563

Microcystis aeruginosa (G)NRC-i (?)

Anacystis marina (D), 6

Berkeley isolateBerkeley isolateBerkeley isolate

a Berkeley Culture Collection.I Indiana University Algal Culture Collection.

and the medium of Van Baalen (73). Whereasmany of our unicellular strains grow excellentlyin these particular media, some do not; and it isfor this reason that we have adopted mediumBG-1 1 as a general culture medium for unicellularstrains.We routinely use Warm White DeLuxe fluores-

cent tubes (General Electric) as light sources,although other fluorescent tubes or incandescentbulbs are also satisfactory. The light intensity isusually adjusted to lie in the range of 2,000 to3,000 lux at the surface of culture vessels, asmeasured with a Weston illumination meter,model 756. A few strains (6507 and 6909) requirelower light intensities for normal growth (500lux or less).With the exception of thermophilic strains

M. B. Allen

M. M. AllenM. M. AllenD. L. Dyer

C. Van Baalen

C. Van BaalenR. CastenholzR. CastenholzGassner

M. M. AllenG. P. FitzgeraldG. P. FitzgeraldM. M. AllenR. KunisawaR. KunisawaR. KunisawaR. KunisawaR. KunisawaR. KunisawaE. G. Pringsheim

P. R. Gorham

C. Van Baalen

A. NeilsonA. NeilsonA. Neilson

P.L BCCa 1963

D. S. Berns uBCC, 1967P. BCC, 1970

p BCC, 1970P. BCC, 1967P BCC, 1967IUCCb u- BCC,1969

- BCC, 1969-i-*BCC, 1963

IUCC us) BCC,1969R. Safferman R.Haselkorn usBCC, 1969P. BCC, 1970

1, 52,60

22

23

73

735765

2

73

(6715, 6716, and 6717), all strains now in culturegrow satisfactorily at 25 C, although this is con-siderably below the temperature maximum formost mesophilic strains. Largely as a matter ofconvenience, we have incubated plate cultures andstock cultures at room temperature. However,growth is more rapid at higher temperatures;for experimental purposes, mesophilic strains areusually grown in water baths maintained at tem-peratures in the range of 30 to 35 C. The thermo-phils must be incubated at a temperature of 40 C(or higher) to obtain good growth.Liquid cultures of small volume (including

stock cultures) are grown without aeration intest tubes or Erlenmayer flasks of 125-ml capacity,either in water baths provided with lateral il-lumination from banks of fluorescent tubes or

IA

IB

6301

631163126715

7002

7003671667176908

66056910630766036706670767086709671067136907

6911

7001

680269016903

174 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



TABLE 2. Axenic strains of unicellular blue-green algae: typological groups II and III

Typo- Berki Name (if any) and other strain Isolator Subsequent strain history Ref er-lon1strain designations IsltrSbeun tanhsoy encesgroup no.

IIA 6308 Gloeocapsa alpicola (D) 1051 G. P. Fitzgerald u BCC,a 19636701 Berkeley isolate J. Hauxhurst6711 Berkeley isolate R. Kunisawa6804 Berkeley isolate R. Kunisawa6807 Berkeley isolate R. Kunisawa6808 Berkeley isolate R. Kunisawa6714 Berkeley isolate R. Kunisawa6702 Berkeley isolate J. Hauxhurst6803 Berkeley isolate R. Kunisawa6805 Berkeley isolate R. Kunisawa6806 Berkeley isolate R. Kunisawa

IIB 6501 Berkeley isolate M. M. Allen 4, 626909 Gloeocapsa sp. (G) IUCC 795 Markle IUCCb - BCC, 62, 76

1969

1GC 7005 Microcystis aeruginosa (G) G. C. Gerloff - SAUGC u BCC,1969

III 6712 Berkeley isolate R. Kunisawa

Chlorogloea fritschii Mitra A. K. Mitra Botany Dept., Univ. 24, 25,of Calif. (Berkeley) 26, 46

BCC, 1967

a Berkeley Culture Collection.b Indian University Algal Culture Collection.¢ Sammlung von Algenkulturen der Universitat Gottingen.

in a light cabinet. Cultures of larger volume (1.5liters) are grown in Fernbach flasks equipped withmagnetic stirrers and are incubated in water baths.The most convenient light source for such culturesis a circular fluorescent tube suspended aroundthe culture vessel; it can be raised or lowered toprovide the desired light intensity. These flaskcultures are gassed with N2 containing 0.5%(v/v) C02, introduced through a Pasteurpipette positioned in the cotton plug so that thetip of the pipette is slightly above the surface ofthe culture medium. In aerating cultures, it isnot desirable to bubble gas through the medium.This may cause moistening of the cotton plug,which favors contamination, particularly if theculture is incubated for several weeks, oftennecessary with slowly growing strains. Cultureswith a volume of 10 liters are grown in carboysequipped with magnetic stirrers and gassed abovethe level of the medium with N2 containing 1 %

(v/v) CO2. Illumination is provided by two banksof fluorescent tubes on either side of the carboy.

In large flasks and carboys, the growth ratebecomes limited by light at relatively low popu-lation densities, as a result of self shadowing by

TABLE 3. Medium BG-1Ja

AmtCompound (g/fiter)NaNO3 1.5K2HPO4 0.04MgSO4*7 H20 0.075CaCl2*2 H20 0.036Citric acid 0.006Ferric ammonium citrate 0.006EDTA (disodium magnesium salt) 0.001Na2CO3 0.02Trace-metal mix A5b 1 ml/liter

a After autoclaving and cooling, pH of medium7.1.

b H3BO3, 2.86 g/liter; MnCl2.4H20, 1.81 g/liter; ZnSO4*7H20, 0.222 g/liter; Na2MoO4*2 H20,0.39 g/liter; CuSO2*5H20, 0.079 g/liter; Co(NO3)2-6H20, 0.0494 g/liter.

the growing culture. For this reason, the growthyields obtainable in such cultures are relativelylow; growth is virtually arrested when the cellmass reaches a density of 1 to 2 g (wet weight)per liter.We have found it advisable to incubate plate

cultures in clear plastic boxes. (Commercial

VOL. 35, 1971 175


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

vegetable crispers are excellent for this purpose.)It is often necessary to keep plate cultures forseveral weeks, and their enclosure in boxesgreatly reduces external contamination and waterloss by evaporation. Plate cultures of thermophilicstrains are incubated in shallow, gasketed plasticboxes immersed in an illuminated water bath.Two factors of critical importance for the

cultivation of blue-green algae are the cleanlinessof glassware and the purity of the water used toprepare media. Experience has taught us that ifboth factors are not carefully controlled, sporadicfailures to obtain satisfactory growth are almostcertain to occur. We now employ a special system(Millipore Super-Q water purifier) to preparewater for use in media and for the final rinsing ofglassware. All glassware is washed by hand andthen rinsed successively with tapwater, concen-trated HNO3, and purified water.The effects of organic compounds on growth

were determined on plates of medium BG-11supplemented singly with various organic com-pounds. The basal medium was modified byusing NH4NO3 in place of NaNO3 as a nitrogensource. Penicillin sensitivity was determined onplates of medium BG-11 containing gradients ofpenicillin G (0 to 1, 0 to 10, and 0 to 50 units/ml).Temperature maxima were determined by visualestimation of growth in tubes of medium BG-11incubated in water baths at a series of differenttemperatures.

Enrichment, Isolation, and Purification ofUnicellular Blue-Green Algae

Of the 40 strains listed in Tables 1 and 2, 25were isolated at Berkeley from local samples offresh water (ponds, streams, or reservoirs). Innearly every case, the inoculum was subjected toprimary enrichment in a tube of medium BG-11incubated in an illuminated water bath at 35 C.When microscopic examination revealed thepresence of unicellular blue-green algae in anenrichment tube, streaked plates were prepared.Immotile unicellular blue-green algae formcompact, deeply pigmented colonies on plates.Even when they are too small to be seen with thenaked eye, such colonies are therefore easy todetect if plates are examined with a dissectingmicroscope. Early transfer of colonies, performedunder a dissecting microscope, greatly facilitatespurification. Numerous restreakings may berequired to eliminate contaminating bacteria,but given sufficient time and patience this simpletechnique almost never fails. The purification ofimmotile unicellular blue-green algae is tediousbut not difficult.A few strains (notably 6901 and 6903) are very

actively motile, and these were purified by the

technique which has proved effective with somemotile filamentous blue-green algae. A smallpatch of material was placed on the agar surfaceat one side of a petri dish which was illuminatedfrom the opposite side. The blue-green algae re-sponded phototactically by gliding across theplate towards the region of higher light intensity;as soon as some organisms had reached the oppo-site side, they were picked and subjected to arepetition of the same procedure.When the macroscopic and microscopic appear-

ance of the material on plates suggested that apure culture had been obtained, a transfer wasmade to a tube of liquid medium BG-1 1 used forthe maintenance of stock cultures. The purity ofliquid cultures can be controlled in several ways.In our experience, the most reliable control is acareful microscopic examination of ageing cul-tures in which the blue-green algae have becomesenescent and have begun to die. The presence ofcontaminating bacteria is very evident underthese conditions. We have also routinely testedliquid cultures for bacterial contamination bymaking transfers to a complex liquid medium(yeast extract-glucose broth), subsequently in-cubated in the dark at 30 C. The absence ofgrowth under these conditions is a necessarycriterion of purity, but far from a sufficient one.This test will, of course, reveal gross bacterialcontamination; but if only one or a few kinds ofbacteria still contaminate a culture, they will notnecessarily be able to grow in an arbitrarilyselected complex medium, incubated under asingle set of environmental conditions. To devisecultural tests that would reveal all possible nutri-tional categories of contaminating bacteria isscarcely feasible; it is for this reason that we haverelied primarily on the microscopic control ofpurity.

Measurement of Absorption SpectraAbsorption spectra of cell suspensions were

determined in a Cary spectrophotometer, using adiffusing plate of opal glass to minimize the effectsof scattering (66).

Determination of Fatty-Acid CompositionTechniques for determining fatty-acid compo-

sition have been described by Kenyon and Stanier(39).

Extraction of DNACells were harvested by centrifugation, washed

in buffer [0.03 M tris(hydroxymethyl)amino-methane hydrochloride (pH 8.0), containing0.15 M NaCl and 0.001 M ethylenediaminetetra-acetic acid (EDTA)], resuspended in a minimal

176 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



volume of 0.15 M NaCI-0.1 M EDTA (pH 8.0).and stored in the frozen state at -40 C untilextraction.The majority of the unicellular blue-green

algae are refractory to lysis by osmotic shock withor without prior lysozyme treatment (18, 22).Strains 7002, 7003, 7005, and 6707 are sensitive tolysozyme. These were treated with 100 ug oflysozyme per ml at 37 C in the saline-EDTAbuffer for 2 hr; addition of sodium dodecylsulfate (SDS) to a final concentration of 2%(v/v) at 60 C resulted in prompt lysis.Numerous other methods were employed to

disrupt the remaining strains. The cells were in-cubated in lysozyme as above, followed by an ex-cess of cell wall lytic enzymes of Streptomycesalbidoflavus as previously described (64). Alarger measure of success with this method wasachieved when the washed cells were incubated in1 M ,3-mercaptoethanol (29) for 2 hr at 37 C,centrifuged once, and resuspended in lysozymeand then S. albidoflavus enzymes; lysis was thencompleted with SDS as described above.

Cells of strains 6716, 6801, 6501, and 6701 lysepoorly or not at all after enzymatic treatment.These cells were then disrupted in a Hughes pressat -18 C. The frozen paste was thawed in thepresence of SDS in saline-EDTA buffer. For com-parative purposes, one lot of cells of strain 6501was disrupted in a French pressure cell (22).

Nucleic acids were separated from cellulardebris by shaking with an equal volume of saline/EDTA-saturated phenol at pH 8.0 and a drop ofchloroform. The phases were separated bycentrifugation at 8,000 X g at 5 C. Whole cellsappear at the bottom of the centrifuge tube belowthe phenol layer; denatured proteins and much ofthe cell wall and sheath components concentrateat the interface. The aqueous layer was removedwith a wide-mouth pipette and the nucleic acidswere concentrated by layering thereon twovolumes of 95% ethyl alcohol and "spooling"fibers on a thin glass rod. The fibers were washedthree times in small volumes of 70% ethylalcoholand then were dissolved in a small volume ofSSC (0.15 M NaCI, 0.015 M Na3 citrate, pH 7.0).Such partially purified preparations were storedat 5 C over CHC13 and were analyzed within2 weeks.When large amounts of material were available

(strain 7002), the DNA was purified by themethod of Marmur (45).

Determination of Mean DNA BaseComposition

Buoyant densities in CsCl were determined foreach sample in two different Spinco model Eultracentrifuges, using four-cell rotors and optical

masks. One centrifuge was equipped with a photo-multiplier scanner in addition to the usualcamera assembly. Ultraviolet absorbance in thecells was determined at 265.4 nm after centrifuga-tion of 1 to 3 Mg of the DNA for 23 hr at 42,040rev/min at 25 C. Centers of the peaks of theultraviolet-absorbing bands were measured fromscanner tracings or from densitometer tracings ofthe ultraviolet photographs prepared with the aidof a Joyce-Loebl mark III recording densitometer.The densities were calculated with respect to aninternal standard of Bacillus subtilis bacterio-phage 2C DNA of a density of 1.7420 ±i 0.0004g/cm3, which was measured with respect to DNAof Escherichia coli K-12 DNA assumed to be1.7100 g/cm3 (63).The identification of ultraviolet-absorbing

bands as nucleic acid or as polysaccharide cannotbe done by examination by schlieren optics (22)when multicell operation is employed, as in thisreport. This identification was made by examina-tion at 280 nm and at 320 nm. The light absorp-tion by nucleic acids at 280 nm is about half thatat 265.4 nm, and they do not absorb light at 320nm. Polysaccharides appear as ultraviolet-absorbing bands by virtue of light scattering,and their apparent absorption at 320 nm is halfthat at 265.4 nm. Only one strain (7002) yieldedDNA appreciably contaminated with poly-saccharide material. Removal of this material bya phase separation (7) did not alter the recordeddensity of the DNA.

ASSIGNMENT OF STRAINS TOTYPOLOGICAL GROUPS

Classification of Chroococcales is controversial,and even generic identifications require a choicebetween two very different taxonomic treatmentsof the order (20, 28). Since we were not disposedto make this choice on a priori grounds, we pro-visionally assigned all our strains to a series ofunnamed typological groups, each defined by afew salient structural properties (Table 4).We have used the plane or planes of successive

cell divisions as a primary determinative charac-ter. This property is, to some degree, correlatedwith cell shape. Many strains that reproduce byrepeated binary fission in a single plane (group I)have cylindrical or ellipsoidal cells; however, thecells in some strains of this group are spherical(or nearly so) immediately after division. Strainsthat reproduce by successive binary fissions intwo or three planes at right angles to one another(groups II and III) have spherical cells.

It should be emphasized that the planes of suc-cessive cell divisions cannot always be determinedwith certainty by simple microcscopic examina-tion because in many strains the separation of

177VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

TABLE 4. Key to the typological groups ofunicellular blue-green algae

A. Reproduction by repeated binary fission in asingle plane, frequently resulting in the formationof short chains of cells. Group I

1. Cells cylindrical, ellipsoidal, or spherical im-mediately after division; cells do not containrefractile polar granules. Group IA2. Cells cylindrical or ellipsoidal after division;cells always contain refractile polar granules.

Group IBB. Reproduction by binary fission in two or threesuccessive planes at right angles to one another;cells spherical.

1. Cell divisions occur at regular intervals;never form parenchymatous masses of cells.

Group IIa. Cells not ensheathed; do not contain gasvacuoles. Group IIAb. Cells ensheathed; do not contain gasvacuoles. Group IIBc. Cells not ensheathed; contain gas vacuoles.

Group IIC2. Cell division irregular; growth leads to theformation of parenchymatous, tightly packedmasses of cells. Group III

daughter cells occurs shortly after the completionof cell division. When necessary, we determinedthis character by observing the growth of strainsin slide cultures (4).Most strains that divide successively in a

single plane form short chains of cells. The extentof chain formation is frequently variable withina single strain, being dependent both on the con-ditions of cultivation and the stage of growth.This raises a fundamental determinative question:what is the distinction between a unicellularblue-green alga which can form short chains ofcells and a filamentous blue-green alga in whicha trichome is the vegetative unit of structure?There is no clear-cut answer to this question, as

shown by the controversy which has alreadysurrounded the correct taxonomic assignment ofstrain 6301. A typical representative of our typo-logical group I, this strain was purified in 1952by M. B. Allen (1) who submitted it to FrancisDrouet for identification (Allen, personal com-

munication). He initially identified it as a memberof the Chroococcaceae, under the name Anacystisnidulans. However, according to Silva (67),Drouet subsequently reidentified this strain as

Phormidium mucicola (viz., a member of theOscillatoriaceae). Pringsheim (60) also inter-preted the chains of cells produced by strain 6301

as trichomes, though of an unusual kind. Hedisagreed with its assignment to Phormidium andproposed the new genus Lauterbornia for it. Onthe other hand, Allen and Stanier (4), Padmajaand Desikachary (52), and Kunisawa andCohen-Bazire (43) argued that chain formation by6301 and similar strains is insufficient ground forexcluding them from the Chroococcales. A similarphenomenon is characteristic of many eubacteria(e.g., Streptococcus and Lactobacillus) which areconventionally regarded as unicellular organisms.We shall adopt this view here and consider allmembers of group I to be representatives of theChroococcales, even though the great majorityof them can form short chains of cells. Thestrains for which this interpretation may be mostquestionable are 6901 and 6903 (group IB) whichtypically occur as chains of at least two to fourcells. We might have been inclined to interpretthese two strains as filamentous blue-greenalgae which produce short trichomes were it notfor their numerous similarities to the third strainof group IB, which is unicellular under the samegrowth conditions.The microscopic appearance of each strain is

shown in Fig. 1-5 and 7-9. These photomicro-graphs were selected to illustrate the typical size,shape, and arrangement of cells from youngcultures in the course of active growth, as seenwith phase-contrast illumination. With the excep-tion of Fig. 7 and 9, all were printed at the samemagnification, indicated by the scale markeraccompanying each group of prints. The strainswere arranged in these figures according to theirassignments to typological groups and subgroups;in the two largest typological subgroups (IA andIIA), strains of markedly different DNA basecomposition were placed in separate figures (Fig.1 and 2 for strains of group IA; Fig. 4 and 5 forstrains of group IIA).

Within group I, we have distinguished twotypological subgroups. The larger of these, IA(Fig. 1 and 2), comprises strains without anydistinguishing intracellular structures visible inthe light microscope. The cells may be cylindrical,ellipsoidal, or spherical, and the size range isconsiderable. Group lB (Fig. 3) comprises threestrains with rod-shaped cells which have con-spicuous refractile granules at each pole. This is aconstant and regular structural property, whichpermits a ready differentiation from all strains ofgroup IA. The polar granules arise in the courseof septum formation, as can be seen from the

178 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



671.6

6716.. ..

6715 6717

epdo

*

.06301

'4

* 6311'4

6908

ee2

7002' 7003 :

10

FIG. 1. Photomicrographs of strains belonging to typological group IA. All strains illustrated have DNA oflow GC content (48 to 56 moles %).

6312

179VOL. 35X 1971


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

-- -

1 0 A

FIG. 2. Photomicrographs of strains belonging to

typological group IA. All strains illustrated have DNAof high GC content (66 to 71 moles %).

appearance of dividing cells of strain 6903(arrows) in Fig. 3.Within group II, we have distinguished three

subgroups. The strains of group 11A (Fig. 4and 5) do not produce sheaths or gas vacuoles,although the cells may be surrounded by adiffuse slime layer, demonstrable in India-inkpreparations (Fig. 6). The two strains of groupIIB (Fig. 7) form conspicuous, sharply defined,multilaminate sheaths. Geitler (28) expressedsome doubts about the distinctness of sheathsand slime layers in Chroococcaceae. However, wehave never found any difficulty in making aclear distinction between these two kinds ofextracellular structures. Furthermore, the prop-erty of sheath formation has remained constantin strain 6501 through the period of more thanfive years, during which this strain has been keptin pure culture. We believe, therefore, that thisstructural character is a significant and taxonomi-cally useful one. The single strain assigned togroup IIC (7005) forms gas vacuoles, visible inthe light microscope as small refractile bodies(Fig. 8). It is not ensheathed, and it can be dis-tinguished typologically from the strains ofgroup IIA only by the property of gas-vacuoleformation.A single strain with spherical cells that divide

in more than one successive plane (strain 6712)differs markedly in growth habit from the strainsof group II, and has been assigned to a separatetypological group III. The size of the cells in thisstrain is highly variable, and it grows as dense,irregular, parenchymatous masses in which manycells have a polygonal shape as a result of closepacking. The characteristic growth habit ofstrain 6712 is best revealed by successive photo-micrographs of a developing slide culture (Fig.9), which show how the parenchymatous aggre-gates develop from single cells. The colony struc-ture of this strain is also unusual: colonies onplates are hard, dry, and coherent, in contrast tothe butyrous or slimy colonies formed by otherunicellular strains.One other blue-green alga of uncertain taxo-

nomic position which has been isolated in pureculture has a somewhat similar mode of growth.This is the organism originally described byMitra (46) under the name Chlorogloea fritschii,a strain of which we have compared with 6712.As shown by its development in a slide culture(Fig. 10), C. fritschii closely resembles 6712 invegetative growth habit, except for the occasionalformation of short chains of cells, a characterthat we have not observed in 6712. However, aswill be discussed, the two strains differ in other

180 BACTERIOL. REV.

1%, -V

tos


http://mm

br.asm.org/

Dow

nloaded from



,,1O

FIG. 3. Photomicrographs of strains belonging to typological group lB. Note the refractile, bipolar granuleswhich distinguish strains ofgroup IBfrom those of group IA. Arrows on the photomicrograph ofstrain 6903 pointto newly formed polar granules which are developing at the site ofseptum formation in two dividing cells.

phenotypic respects and probably are not closelyrelated taxonomically.

DNA BASE COMPOSITIONThe buoyant densities and calculated mean

DNA base compositions of the strains are listedin Table 5, which also includes the values previ-ously published (22) for six of these strains. Theagreements between the independent determina-tions are as close as can be expected. It should be

noted that the value previously published forstrain 6604 was in fact determined on strain 6603,an error that we have corrected in Table 5.The present, much more extensive analyses

have not extended the over-all range of base com-position for the Chroococcales; the limits remainat approximately 35 and 70 moles % GC. How-ever, they do reveal significant facts about thedistribution of DNA base composition within theorder (Fig. 11). The strains of typological groups

181VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

9.\'?2;

:; S"7

pAmk _dIMUL, tJ. i

~~~~~~.."At ...J~sLIW m

*a_

.I

108

FIG. 4. Photomicrographs of strains belonging to typological group IIA. All strains illustrated have DNA oflow GC content (35 to 37 moles %)).

I and II show ranges that are almost entirely dis-tinct. The range for group I extends from 45 to72 moles % GC; for group II, from 35 to 47moles % GC. The value for the one strain of

group III falls within the range characteristicfor group II. In Fig. 11, the ranges established byEdelman et al. (22) for the families Nostocaceae,Oscillatoriaceae, and Scytonemataceae have also

182 BACTERIOL. REV.

4

'Iwd'V..NI"..:,A 1,00. A

.403


http://mm

br.asm.org/

Dow

nloaded from


171..

i '.8'0

10o

FIG. 5. Photomicrographs of strains belonging to typological group IIA. All strains illustrated have DNA ofhigh GC content (46 to 47 moles %).

.e srt-----L

-

-

w

_k~~~,44, -<4

lt~~ s

.k . AP11

r 4

tOAFIG. 6. Photomicrographs (wet mounts in India ink) of three strains belonging to typological group IIA, to

show the slime layers formed by some representatives of this typological group.

183

0

0o0

680

VI.-

114 4 -tr-

1, .I

-

.

4s-


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

lOl

FIG. 7. Photomicrographs of strains belonging totypological group JIB. Note the thin, multilaminatesheaths which unite the cells into microcolonies, en-tirely different in structure from the slime layers (Fig.6) produced by some strains of group IIA.

been included. They are comparatively narrowand cover the region of overlap between theranges for groups I and II of the Chroococcales.The strains of group I fall into two widely

separated base compositional clusters, each ofwhich has a comparatively narrow range. Theseclusters are not well correlated with typologicalsubdivisions. The 14 strains with values in therange of 45 to 56 moles % GC include 11 strainsof group IA and all 3 strains of group IB; the 11strains with values in the range of 66 to 73 moles% GC are all of group IA. A comparison of thephotomicrographs of individual strains of groupIA in Fig. 1 (low-GC cluster) and 2 (high-GCcluster) shows, furthermore, that many strainsdiffering by over 10 moles % GC in base compo-sition cannot be distinguished by cell size or

shape. The base compositional data thereforeprovide prima facie evidence that the existing

classification of unicellular blue-green algae onthe basis of structural characters has very poortaxonomic resolving power.Group II comprises three separate base compo-

sitional clusters, again poorly correlated withtypological subdivisions. The strains of groupIIA fall into two clusters at the extreme ends ofthe base compositional span (35 to 37 and 46 to47 moles % GC). A comparison of the photo-micrographs of the individual strains of thesetwo clusters (Fig. 4 and 5, respectively) does,however, reveal that they are distinguishablefrom one another by cell size: strains of high GCcontent have smaller cells than strains of lowGC content. The strains of group IIB, charac-terized typologically by multilaminate sheaths,are intermediate in base composition (40 to 42moles % GC) between the strain clusters of groupIIA. The single gas-vacuolate strain of group IICis indistinguishable in base composition from thehigh-GC cluster of group IIA. Strain 6712(typological group III) is similar in DNA basecomposition (40.8 moles % GC) to Chlorogloeafritschii (42.9 moles % GC).

PHENOTYPIC PROPERTIESMotility

Gliding movement is widespread, perhapsuniversal, among filamentous blue-green algae.In some of these organisms the trichomes aremotile; in others motility is confined to hormo-gonia (37). There have been a number of reports,summarized by Pringsheim (61), that certainmembers of the Chroococcales are motile, but theextent to which gliding movement occurs in thisorder is still unclear.A majority of the unicellular strains that we

have examined appear to be permanently im-motile. However, we have observed gliding move-ment in some representatives of both typologicalgroups I and II. All strains of group IB are vigor-ous gliders. Movement is particularly rapid instrains 6901 and 6903, which regularly grow asshort chains of cells. Rapid gliding movement isalso characteristic of one strain in group IA,6910. Motile strains of group I show a positivephototactic response on plates exposed to uni-lateral illumination (Fig. 12).More restricted motility occurs in several

strains of group IIA. The movement of thesestrains is always very slow, and we have not beenable to recognize it with certainty by microscopicexamination of wet mounts. Motility was firstdiscovered on plate cultures of strain 6711 (Fig.13a). As can be seen, the phototactic response isnegative. During maintenance in liquid culture,strain 6711 gives rise to nonmotile variants

184 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



FIG. 8. Photomicrograph of strain 7005, the sole representative of typological group JIC. Arrows point tosmall gas vacuoles.

(Fig. 1 3b). We have been able to preserve a motilestock of this strain by repeated selection ofmotile clones on plates.

Other strains of group 11A in which colonymovements have been occasionally observed are6701, 6803, 6804, and 6808. Movement of thesestrains on plates is both sporadic and of limitedextent, and we do not understand the factors thatdetermine its occurrence.

Accordingly, our experience shows that glidingmovement is rare in typological group I, but whenit does occur in strains of this group it is rapidand easily recognized. In typological group IIA,the potential capacity to perform slow glidingmovement may be fairly common, but it is diffi-cult to detect for the reasons given above. Sincemotility seems to be an unstable character instrain 6711, it could have escaped detection inother strains of group IIA as a result of its lossduring their maintenance in liquid culture. Wehave not detected movement in strains of groupsHIB, TIC, and III.The marked differences between motile strains

of groups I and II with respect to the regularityand the rate of movement may perhaps indicatethat the mechanisms of movement are dissimilar.Pringsheim (61) suggested, on the basis of micros-

copic observations, that two types of movementoccur in blue-green algae. Gliding movement(Gleitbewegung) is seen characteristically inmembers of the Oscillatoriaceae; it is relativelysmooth and rapid, and accompanied by axialrotation of the filament. Jerking movement(Ruckbewegung) is slow, irregular, and fre-quently interrupted. The movement of strains ofgroup I can clearly be described as gliding move-ment. It is possible that the comparatively slowmigration of colonies on plates we have observedin strains of group IIA reflects the occurrence ofjerking movement, although, as already men-tioned, we have not been able to detect it micro-scopically.

Growth in Darkness and Utilization ofOrganic Compounds

Pure cultures are essential for a study of theutilization of organic compounds by blue-greenalgae; consequently, information about thisaspect of their physiology is still very limited.Reviewing the information then available, Holm-Hansen (33) concluded that most blue-green algaeare obligate phototrophs, although there are well-authenticated reports that a few strains can growin the dark at the expense of organic compounds.

185VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


lop

FIG. 9. Two series ofphotomicrographs to illustrate the growth of strain 6712 (typological group ill) iln a slideculture. The niumber in the top left corner of each print indicates the time (in hours) after the preparation of theslide culture at which the photograph was taken. In the series on the right, the Ibur cells initially present andtheir progeny have been numbered. A comparison of the development of each cell reveals the irregular timing ofcell division in this strain. The parenchymatous aggregates of compressed cells produced after 120 hr are typicalfor strain 6712.

186


http://mm

br.asm.org/

Dow

nloaded from


UNICELLULAR BLUE-GREEN ALGAE 187

10/A

FIG. 10. Series of photomicrographs to illustrate the growth of Chlorogloea fritschii in a slide culture. Thenumber in the top left corner of each print indicates the time (in hours) after the preparation of the slide cultureat which the photograph was taken. One cell in the initial group (arrow), distinguishable by its relatively thickwall, eventually gave rise (arrows at 163 and 194 hr) to a chain of6 small cells.

VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

TABLE 5. Buoyant densities and guanine-plus-cytosine (GC) contents of deoxyribonucleic acid fromstrains of unicellular blue-green algae

Typological Cell disruption Mean dMensi Mean density GC (molgroup Stanmethod5 tycm G (mls %) tyg/cm(moles b

IA

IB

IIA

IIB

IIC

III

6715671667176301631169086312700270036605691063076603670667076708670967106713690769117001

680269016903

63086701671168046807680867146702680368056806

65016909

7005

6712Chlorogloeafritschii

BDBBBCAAABCBBBA, CBBBCCCB

DCC

BBC, DCCCBBBCC

C, D, EC

A

BD

1.71251.71251.7111.7141.7141.71451.7091.7081.70851.7061.7071.72851.72451.7291.72631.7281.72651.7271.7281.7301.7251.728

1.7051.7051.704

1.6941.6951.6961.69551.6951.6951.70651.70651.70651.7071.7055

1.7001.7005

1.7045

1.7001.702

53.653.65255.155.155.650.249.049.546.948.069.765.770.267.769.467.968.469.471.466.369.4

45.945.944.9

34.735.736.736.235.735.747.447.447.448.046.4

40.841.3

45.4

40.842.9

1.7151.715

1.7301.722

1.728

1.694

5656

7163

69

35

a A, lysozyme-sodium dodecyl sulfate (SDS); B, lysozyme, Streptomyces albidoflavus enzyme, SDS;C, B, after j3-mercaptoethanol; D, Hughes press; E, French pressure cell.

b From data of Edelman et al. (22).

In analyzing this question, a clear distinctionmust be made between the utilization of organiccompounds in the light and in the dark. Providedthat it is permeable to organic compounds, evenan obligate phototroph may well be able to

assimilate them in the light, using photochemicalreactions for the generation of adenosine triphos-phate (ATP) and reducing power. The photo-synthetic growth rate may be increased by theprovision of organic substrates, if they can serve

188 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



manLEG TCmRe OFA

hi,

z

IL

GROUP N

#OSTCAA E

HLVA

rcYocWVTrA

40

so so

70

U

70FIG. 11. Ranges of mean DNA base composition among the blue-green algae. Data for the Chroococales,

taken from Table 5, show the compositional ranges for the subgroups defined in Table 4. Data for filamentousfamilies are taken from Edelman et al. (1967). Numbers in parentheses under each bar denote the number of strainsanalyzed.

as general sources of cellular carbon and if therate of photosynthetic growth is normally limitedby the rate of CO2 assimilation. However, togrow with organic compounds in the dark, aphotosynthetic organism must be able to use themas sources of ATP and reducing power, generatedthrough respiration or fermentation, and notmerely as sources of assimilable carbon.

Smith, London, and Stanier (68) studied theutilization of organic compounds in the light bythree of the unicellular strains in our collection:6301 and 6307 (typological group IA) and 6308(typological group IIA). All three strains couldassimilate organic acids, amino acids, and glucose,although none of the compounds tested increasedthe rate of photosynthetic growth. The uptake ofthese compounds by cells was light dependent.Even acetate, the most rapidly assimilated com-pound that was examined, contributed only some10% of the carbon incorporated into newlysynthesized cell material. An investigation of thepatterns of incorporation of '4C-labeled acetateand other substrates into the cellular amino acidssuggested that the tricarboxylic acid cycle wasnot operative, and this inference was supportedby the failure to detect a key enzyme of the cycle,a-ketoglutarate dehydrogenase, in cell-free ex-tracts.With respect to the photoassimilation of ace-

tate, essentially similar results were reported byPearce and Carr (55) and Hoare et al. (31) for

several unicellular and filamentous blue-greenalgae. Hoare and his associates (32) subsequentlyextended such studies to many other strains,including several in which growth in the dark hadpreviously been reported. The absence of afunctional tricarboxylic acid cycle has beenconfirmed for every blue-green alga so farexamined.

Smith et al. (68) were unable to demonstratereduced nicotinamide adenine dinucleotide oxi-dase activity in extracts of the unicellular strainsthat they studied; however, low levels of thisactivity were subsequently reported in blue-greenalgae by some workers (35), but not by others(8, 44). The mechanisms and physiological sig-nificance of respiratory electron transport inthese organisms still remain to be elucidated.

Since none of the blue-green algae appears tohave a functional tricarboxylic acid cycle, it isevident that a normal respiratory metabolism oforganic compounds in the dark cannot occur;growth in the dark, therefore, probably takesplace at the expense either of glycolysis or of thepentose phosphate cycle (56). It should be notedthat carbohydrates are the only organic sub-strates that have been reported to support thegrowth of blue-green algae in the dark (24, 40).This has been confirmed (R. Rippka, unpublishedobservations) for several of the filamentous blue-green algae in our collection: only glucose,fructose, and sucrose permit growth in the dark.It would be surprising if blue-green algae that

I-A ~~~~~~~I-A(II) (N)

C) ) U)

(I

(10)

(5)

VOL. 35, 1971 189


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

Jr

6711

Iii'

13a

6910

a6\

e 0 to *0_me

-ii,t,~~~~~~~~~~~~~~~~~~~~~~~~~~ 0 *~~~~~~~-

IbaLF~C

mm GM 'A

.1

11 i 0.1

"I :i / d'I.f.O

FIG. 12. Phototactic migration on plates of strains6903 (typological group IB) and 6910 (typologicalgroup IA). Each plate was inoculated with a singlehorizontal streak slightly below the center. Photo-graphs were taken after incubation at 30 C for 6 days.Illumination was unidirectional (arrow). Both strainsare positively phototactic.

possessed an effective machinery of respirationwere unable to grow at the expense of such com-

pounds as succinate and pyruvate, known to beassimilable in the light (68). The strict de-pendence on carbohydrates as energy-yieldingorganic substrates suggests a very limited re-spiratory capacity.

FIG. 13. Phototactic migration on plates of strain6711 (typological group hIA). Figure 13a shows a

small group of colonies grown for about 2 weeks on a

plate with unidirectional illumination (arrow); allhave migrated from the points of initial growth (attop) away from the light source. Figure 13b shows a

plate streaked from a liquid stock culture of the same

strain afier incubation for 10 days with unidirectionalillumination (arrow). The heterogeneity with respectto motility is evident; many colonies show no evidenceof movement. Phototaxis is, again, invariably negative.

We have screened all strains of groups I, II,

and III qualitatively for growth in the dark at theexpense of a series of organic compounds. Thecompounds tested were glucose (0.25 %, w/v);

6903 IF-, 1.

.. . I,

. ,

I

1.I4

I.-,i - P, j

f i t

190 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



acetate, succinate, pyruvate, and glutamate(furnished as sodium salts at a concentration of0.1%,, w/v). The experiment was conducted byusing plates of mineral agar to which the organiccompounds were added singly. NH4 NO3 wasused in place of NaNO3 as a nitrogen source topreclude the possibility that nitrate is not autilizable nitrogen source in the dark, as might bethe case if its assimilatory reduction were alight-dependent process. The plates were inocu-lated by placing small drops of liquid cultures onthe agar surface. One set of plates was incubatedat 30 C in the dark and one set was incubatedat the same temperature in the light. Growth ofmost strains in the set of plates incubated in thelight was clearly evident after 6 days, and themore slowly developing strains showed growthwithin 2 weeks. However, after a month nostrains showed detectable growth on the set ofplates incubated in the dark.The growth of nostocaceous blue-green algae

in the dark at the expense of sugars is sometimesextraordinarily slow and may become clearlyevident only after incubation periods of severalmonths. Hence, the results of this experimentwith unicellular blue-green algae do not rigor-ously exclude the possibility that some strainsmay be capable of comparably slow growth inthe dark. Even if this phenomenon could bedemonstrated, it would obviously be of littlephysiological or ecological significance. For allpractical purposes, the unicellular blue-greenalgae that we have examined can be consideredobligate phototrophs.Even among filamentous blue-green algae, we

have not yet encountered a strain able to grow inthe dark with sugars at a rate remotely compar-able to its growth rate in a mineral medium in thelight. Hence, although it is formally correct todescribe some filamentous blue-green algae asfacultative chemoheterotrophs, the potential forchemoheterotrophy appears to be very limited inall blue-green algae tested so far.

Nitrogen FixationUntil recently, the capacity for nitrogen fixa-

tion among blue-green algae appeared to be in-variably associated with the ability to form hetero-cysts. Heterocysts are not produced by unicellularblue-green algae, with the exception of Chlor-ogloea fritschii, an organism whose assignmentto the Chroococcales is controversial (26). In1969, however, Wyatt and Silvey (76) reportedthat a Gloeocapsa species can grow well with N2as a nitrogen source under aerobic conditions,and they demonstrated by appropriate assaymethods the presence of nitrogenase in thisstrain.We have screened most strains of groups I, II,

and III for the ability to grow well throughrepeated transfers in liquid medium BG-11devoid of nitrate, and we have found only twothat possess this property. One is 6909, the strainstudied by Wyatt and Silvey (76); the other is6501, a closely similar strain isolated at Berkeley.The formation of nitrogenase by strain 6501 hasbeen confirmed by the technique of acetylenereduction (62). Both nitrogen-fixing strainsbelong to group IIB, characterized typologicallyby enclosure of the cells in multilaminate sheaths.Since these are the only representatives of groupIIB so far isolated in pure culture, it is not yetpossible to know whether nitrogen fixation is auniversal character in this typological group. Itshould be noted that strain 6712 (typologicalgroup III), which resembles Chlorogloea fritschiiin some structural respects, does not share itsability to grow at the expense of N2.

Maximal Temperature for GrowthWe have determined approximate temperature

maxima for all strains by visual estimation ofgrowth in medium BG-11, incubated in illumi-nated water baths at various temperatures (Fig.14). Excluding 3 thermophilic isolates from hotsprings, for which the temperature maximum is53 C or higher, 35 of the 37 strains examined hadtemperature maxima between 35 and 43 C. Moststrains isolated at Berkeley had been preselectedin the course of their isolation for the ability togrow at 35 C. However, 11 of 12 nonthermo-philic strains received from other laboratorieshad temperature maxima in the range of 35 to43 C, even though, to the best of our knowledge,they had been isolated at temperatures below30 C. A relatively high temperature maximum

12

U)

Z 8

0 6

zXI

TEMPERATURE MAXIMUM. C

FIG. 14. Frequency distribution of the temperaturemaximum for growth of 40 unicellular strains of blue-green algae.

191VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


TABLE 6. Maximal temperatures of growth for unicellular blue-green algae

Strains belonging to typological groupsMaximal

temperature IA, low GCa IA, high GC IBIA, low GC IIA, high GC IIB IG IIIcluster cluster cluster cluster

<35C 6605 700535 C 6307, 6603, 6909

6710, 671337 C 6312 6707, 6708, 6901, 6903 6308, 6711, 6806

6709, 6706, 68046907, 6911

39 C 7003, 6910 6802 6701, 6808 6702, 6803, 6501 67126805, 6714

41 C 7001 680743 C 6908, 6301,

6311, 700253 C (or 6715, 6716,

higher 6717

a Guanine plus cytosine.

therefore appears to be characteristic of manyunicellular blue-green algae isolated from non-thermal habitats. It should be noted that thetemperature maxima of three strains isolated byVan Baalen from marine sources (73) lay between39 and 43 C, even though organisms from thishabitat might have been expected to possesslower temperature ranges than fresh-waterisolates.

Table 6 shows the distribution of temperaturemaxima among strains belonging to differenttypological groups. Within typological group IA,there is a fair correlation between DNA basecomposition and temperature maximum. Of the11 strains of high GC content (66 to 72 moles %),10 have relatively low temperature maxima (35 to37 C). Of the 11 strains of low GC content (48to 56 moles %), 7 have temperature maxima inexcess of 40 C. In typological group II, only 1 of14 strains can grow at temperatures in excess of40 C.

Sensitivity to LightFluorescent light intensities of 2,000 to 3,000

lux are not deleterious to most unicellular strains.However, the representatives of group IIB (6501and 6909) are unusually light sensitive, and theybecome partly bleached through destruction ofphycobiliproteins if exposed to light intensities inthis range. The highest fluorescent light intensityat which these two strains show normal growthand pigmentation is approximately 500 lux.Several strains of group IIA (6702, 6308, 6714.6711, 6804, 6805, and 6808) bleach when ex-posed to 5,000 lux.

Penicillin SensitivitySemiquantitative experiments with gradient

plates show that most unicellular blue-green algae

are extremely sensitive to penicillin G. The onlycomparatively resistant strain is 6712 (typologicalgroup III), which can grow in the presence of atleast 50 units/ml. Five strains (6301, 6311, 6803,6805, and 6806) can grow in the presence of 1unit/ml, but not of 10 units/ml. The remaining34 strains are inhibited by penicillin concentra-tions of 1 unit/ml or less. Therefore, the inhibi-tory concentrations for the majority of unicellularblue-green algae are similar to those for gram-positive eubacteria and far below the levelsrequired to inhibit growth of most gram-negativeeubacteria.

Cellular Absorption Spectra

The photopigments characteristic of blue-greenalgae are chlorophyll a, phycobiliproteins (C-phycocyanin, allo-phycocyanin, and in somestrains phycoerythrin), and a variable array ofcarotenoids (51, 72). In intact cells, the mainvisible absorption maxima are attributable tochlorophyll a, C-phycocyanin (which alwayspredominates over al/o-phycocyanin), and phyco-erythrin, if this phycobiliprotein is present.

Table 7 shows the positions of the in vivoabsorption maxima for the unicellular strains inour collection. The position of the chlorophyll apeak is essentially constant, at 680 to 683 nm.The position of the C-phycocyanin peak shows agreater variation, extending from 627 to 638 nm.This probably reflects strain differences withrespect to the intracellular state of molecularaggregation of C-phycocyanin. The absorptionmaximum of the isolated pigment from severaldifferent strains lies at 622 to 625 nm, and break-age of the cells typically causes an immediateshift of the peak position to a shorter wavelength.A few strains also contain phycoerythrin, whose

192 STANIER ET AL. BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



TABLE 7. Positions and relative heights of chlorophylla and phycobiliproteiii peaks in intacl cells of

unicellular strainIs4

Strain

Low GC cluster67156716671763016311690863127002

700366056910

High GC cluster6307

66036706

6707

6708

6709

6710

6713

6907

6911

7001

6802

6901

6903

Low GC cluster6308

6711

6701

6804

6807

6808

High GC cluster6714

6702

6803

6805

6806

6501

6909

7005

Peak positions (nm)

Phyco- Phyco- Co-eryth- cyanin

5b

577

580

570570

573577

574

634

628

635

630630629636634

635

629

625

637

636

636637

637

638

637

638

635

630

632

632

633

630

629

627

627

625

628

629

628

628

627

628

630

627

628

682

680680683682

680682

682

680683680

683

683

681

680683

683

682

683

680681

681

682

682

681

683682

682

680

683

680

682

680681

681

682

681

680

Relativepeak

heights,phycocy-anin/chloro-phyll a

0.830.71

0.760. 801 080.920.880.961.170.870.78

1.201.251.261 321.191.281.211 221.57I 39I 03

0. 790.980.98

0.790.840.680.870.740.72

The last column of Table 7 shows the ratio ofthe peak heights of C-phycocyanin (627 to 638nm) and chlorophyll a (680 to 683 nm) as meas-ured from the in vivo absorption spectra. Inmost strains, this ratio does not exceed 1.0.However, an exceptionally large ratio (1.19 to1.36), indicative of a very high phycocyanin con-tent, is characteristic of many strains of groupIA. Almost all the strains which show thisproperty have DNA of high GC content (65 toto 72 moles %7c GC).

Cellular Fatty-Acid CompositionNearly all bacteria, including the photosyn-

thetic purple and green bacteria, contain onlysaturated and monounsaturated fatty acids (38)Eucaryotes also contain polyunsaturated fattyacids. With respect to cellular fatty-acid compo-sition, the blue-green algae are heterogeneous(34, 49, 50, 54). The most extensive comparativeanalysis so far published (39) showed that themajority of filamentous blue-green algae (15 of17 strains examined) contain large amounts ofpolyunsaturated fatty acids, whereas the majority

TABLE 8. Fatty-acid composition of unicellularblue-green algae of typological group I

Group

IA

0.920.940.980.98I 02

0.80

0.73

a Spectra measured behind opal glass.b Relatively low concentration, so that its presence is not

necessarily evident from the color of the cells.

presence is revealed by a peak or shoulder in thein vivo spectrum at approximately 570 nm. Somestrains contain a relatively low concentration ofphycoerythrin, so that its presence is not neces-

sarily evident from the color of the cells asjudged by visual inspection.

IB

Mleandeoxyribo-

nucleicacid basecompo-sition

(moles %,GC)

48-56

66-71

45-46 I

Percentage(by

weight) ofStrain polyunsat-

uratedfattyacidsa

6605691063016311631267156716671770027003

6307660367066707670867097001

680269016903

581220

0

110

2431

10

0

0

0

0

23

434536

Presence ofb

Linoleic -Lino-acid

+

+

+4

+4

+-

+-

y-Lino-lenicacid

4+

aContent of linoleic (18:2), a-linolenic (18:3a),and -y-linolenic (18:3-) acids, as percentage (byweight) of the total fatty acids of the cell.

I Key: + indicates major fatty acid; + indi-cates minor fatty acid.

VOL. 35, 1971 193

Typo-logicalgroup

IA

l

IIB

1IC


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL. BACTERIOL. REV.

TABLE 9. Fatty-acid composition of lalgae of typological groups j

Group

IIA

IIBIIC

III

Aleandeoxy-ribonu-

cleic acidbase

comI)o-sition

moles ,GC

46-48

35-37

41

454143

Strain

67026714680368056806

630867016711680468076808

650170056712

Chlorogloeafritschiii

Per-centage

(byweight)

ofpolyun-

saturatedfattyacidsa

4755

454342

4

423

453

28

unicellular blue-green constituent. In most strains of group I, a-linolenicand III acid is the predominant triply unsaturated fatty

Presence ofb acid. Fatty-acid composition is of particulartaxonomic significance in group IIA, where it isperfectly correlated with the internal subdivision

Lino- a-Lin- 7-Linof the group in terms of DNA base composition.

leic olenic nei Strains of high GC content (46 to 48 moles c%)acid acid acid have a high polyunsaturated fatty-acid content,

whereas strains of low GC content (35 to 37moles %) have a low polyunsaturated fatty-acid

+ + content. Strain 7005 (group IIC) resembles the+ + high GC cluster of group IIA with respect both+ + to DNA base composition and fatty-acid compo-+ + sition. It should also be noted (Table 9) that

fatty-acid composition distinguishes strain 6712- (typological group III) from Chlorogloeafritschii,

_ which it resembles in both DNA base composi-_ _ tion and growth habit.

A:-

+

+ +

a Content of linoleic (18:2), a-linolenic (18:3a), and -y-

linolenic (18:3y) acids, as percentage (by weight) of the totalfatty acids of the cell.

Key: + indicates major fatty acid; A+ indicates minorfatty acid.

of unicellular ones (14 of 20 strains examined)do not. Nearly all the unicellular strains in our

collection have now been analyzed for fatty-acidcomposition. The detailed data will be publishedelsewhere by C. Kenyon, but we have abstractedfrom them the information which appears to beof particular taxonomic significance (Tables 8and 9).

It is evident that unicellular blue-green algaefall into two general groups in terms of fatty-acidcomposition. Some have a high content of poly-unsaturated fatty acids, which represent between12 and 50' by weight of the total fatty acids ofthe cells. Such strains always contain significantamounts of triply unsaturated fatty acids: eitherthe a- or the y-isomer of linolenic acid. Otherstrains either do not contain detectable poly-unsaturated fatty acids or have a very low content,ranging from 1 to 4% of the total fatty-acidcontent of the cells. Furthermore, such strainsnever contain either isomer of linolenic acidthe only polyunsaturated fatty acid present isthe doubly unsaturated linolenic acid.

Strains of high polyunsaturated fatty-acidcontent occur in both typological groups I andII. In strains of group II, y-linolenic acid isalways the predominant triply unsaturated fattyacid; a-linolenic acid, if present at all, is a minor

PROBLEM OF GENERIC NOMENCLATUREIN THE ORDER CHROOCOCCALES

Taxa in the Chroococcales are defined by struc-tural properties, determined in the first instanceon field collections; type materials have usuallybeen conserved as dried herbarium specimens.Over the course of time, many genera and specieshave been proposed. The traditional taxonomy ofthe Chroococcales has been excellently epitomizedby Geitler (28). His treatment is a critical onethat points out the numerous uncertainties:differences between named species are often small,and in some instances generic boundaries are

unclear. Despite its limitations, this taxonomictreatment of the group remains the one mostwidely employed.

Drouet and Daily (20) proposed a majortaxonomic revision of the Chroococcales whichinvolved a drastic reduction in the number ofgenera and species. However, this so-calledrevision did not embody any extension ofknowledge about the properties of the group. Itwas based in large part on a systematic compara-tive study of published descriptions and survivingherbarium specimens, some dating from the 18thcentury. The generic and specific descriptions ofDrouet and Daily convey a minimum of factualinformation and consist largely of lists of syno-nyms. Furthermore, it is very difficult for othertaxonomists to evaluate their allegations con-cerning synonymy, since the evidence presentedconsists of a series of photomicrographs ofherbarium specimens, for the most part of poortechnical quality.The danger of attempting to base the classifica-

tion of organisms as structurally simple as themembers of the Chroococcales on the examina-tion of herbarium specimens scarcely needs to beemphasized. These specimens almost always

194


http://mm

br.asm.org/

Dow

nloaded from



TABLE 10. Generic and specific names previouslyassigned to 10 strains of group Ia

Strain No. Proposed name

System of Drouet and Daily (20)6301 Anacyslis nidulans7002 Agmenellum quadruplicatum7003 Coccochloris elabens6307 Coccochloris peniocystis7001 Anacyslis marina

System of Geitler (28)6715 Synechococcus lividus6716 Synechococcus sp.6717 Synechococcus sp.6908 Synechococcus cedrorum6907 Synechococcus elongatus

a Short rods or cocci, 0.8 to 2.0 urm wide, re-producing by binary fission in a single plane.

contain a diversity of microbial corpses, and theyvary greatly in their state of dilapidation. Anyassertions about synonymy on such grounds aredebatable, and there is no obvious way to resolvea difference of opinion about the identity or non-identity of two herbarium specimens. Therefore,archeological explorations through the herbariaof the world cannot be expected to contributeusefully to the classification of the Chroococcales.Unfortunately, however, nomenclatural priorityrests with these herbarium specimens, no matterhow uninformative the original published de-scriptions may have been or how unrecognizablethe preserved material may have become throughthe vicissitudes of time. By their assertions ofsynonymy, Drouet and Daily succeeded inbringing back from limbo many ancient genericand specific names which had been mercifullyforgotten by other taxonomists. For this reason,some of the genera that they recognize are noteven mentioned in Geitler's treatment ofthe Chro-ococcales.

Fortunately, we can now evaluate pragmaticallythe relative determinative merits of the treatmentsof Geitler (28) and Drouet and Daily (20). Of thestrains assigned to typological group IA, 10 werenamed strains received from other laboratories.Five had been named in accordance with thesystem of Drouet and Daily; these had all beenidentified by F. Drouet, on the basis of theexamination of cultures submitted to him (M. B.Allen, C. Van Baalen, and G. P. Fitzgerald,personal communications). Five had been namedby other investigators who used the system ofGeitler. The proposed generic and specific as-signments of these 10 strains are shown in Table10.

Detailed comment is unnecessary. Even appliedby one of its creators, the system of Drouet and

Dailey (20) leads to three different generic assign-ments for strains that are very similar in grossstructural respects; therefore, this taxonomicsystem is a determinatively worthless one.Identification through the system of Geitler (28)resulted in a consistent generic assignment forfive other strains belonging to the same typo-logical group. Despite its shortcomings, thissystem provides the only reasonable point ofdeparture for future taxonomic work.

POSSIBLE CORRESPONDENCES BETWEENTYPOLOGICAL GROUPS AND TRADI-

TIONAL FORM GENERA IN THECHROOCOCCACEAE

In Table 11, we have summarized what appearto be the most important determinative charac-ters of some of the principal form genera in theChroococcaceae, as described by Geitler (28).Apart from Microcystis, which is very looselydefined, a primary differentiation is made on thebasis of cell shape and successive planes ofdivision. Aphanocapsa, Gloeocapsa, and Chro-ococcus possess spherical cells which divide intwo or three successive planes; Aphanothece,Gloeothece, and Synechococcus possess ellipsoidalor cylindrical cells which divide transversely inone plane. Broadly speaking, the three formergenera correspond to our typological group II,and the three latter to our typological group I.One character emphasized in most of Geitler's

generic descriptions is colony structure. In uni-cellular blue-green algae that produce multi-laminate sheaths, the formation of colonies isdetermined by the continued enclosure of cells,after division, in a common sheath layer. Thistype of colony structure is readily recognizablein the sheath-forming strains of typologicalgroup IIB (see Fig. 7). However, various more-or-less distinctive colony structures are alsodescribed by Geitler for genera and species whichdo not form sheaths (e.g., Microcystis). In suchcases, the colony is produced by the enclosureof many cells in a common and abundant slimelayer. A similar phenomenon occurs in the bac-terial genus Zoogloea, in which it seems to betaxonomically significant, since the characteristiczoogloeal colonies are formed in pure culture aswell as in natural populations (17). However,none of the unicellular blue-green algae withoutsheaths that we have studied regularly producesdistinctive colonial aggregates in pure culture.This is true even of our strain of Microcystisaeruginosa, a species for which Geitler (28) givesa precise description of colony form: "Youngcolonies spherical or elongated, compact; oldercolonies with net-like interruptions, and an

VOL. 35, 1971 195


http://mm

br.asm.org/

Dow

nloaded from


TABLE 11. Comparison of the properties ofcertain genera of the Chroococcaceae, as defined by Geitler (28)

Genus Cell shape Planes of Sheaths Slime Gas Colony structure and other characteristicsdivision layers vacuoles n tutr n te hrceitc

Aphanocapsa Spherical 2 or 3 _ + - Many cells in a slime layer of irreg-ular form

Gloeocapsa Spherical 3 + Va - 2-8 cells enclosed by multilaminatesheaths; cells spherical afterdivision

Chroococcus Spherical 2 or 3 + V - 2-16 or more cells enclosed by mul-tilaminate sheaths; cells hemi-spherical after division

Microcystis Spherical or 1 or + V Many cells in slime layer; colonycylindrical many can assume a number of different

formsAphanothece Ellipsoidal or 1 + - Many cells in a slime layer, usually

cylindrical of irregular formGleothece Ellipsoidal or 1 + V - As in Gloeocapsa

cylindricalSynechococcus Ellipsoidal or 1 _ _ - Never forms colonies

cylindrical

a Character that is variable within the genus.

indistinctly delineated slime layer." This characteris, accordingly, undeterminable for most of thestrains in our collection. Unless it could beshown (as has been done for Zoogloea) that someunicellular blue-green algae regularly producecolonies of distinctive form held together by aslime layer even in pure culture, this type ofcolony structure must be considered a characterof dubious taxonomic value.Among the form genera with spherical cells

that divide in two or three planes, Aphanocapsacan be distinguished from Gloeocapsa andChroococcus by its inability to form sheaths.This distinction probably appeared less significantto Geitler than it does to us, since he doubtedwhether slime layers and sheaths were reallydifferent kinds of cellular envelopes. Our experi-ence suggests that the distinction is both signifi-cant and readily determinable. If one excludesfrom the generic definition of AphanocapsaGeitler's qualifications concerning colony form(in any case very indefinite), the strains of typo-logical group IIA fall into this form genus.The distinction between Gloeocapsa and

Chroococcus, both of which produce sheaths, isless clear cut. Geitler evidently felt that his verbaldefinitions were unsatisfactory, since he com-mented (28): "the comparison of illustrations (ofChroococcus) with those of Gloeocapsa revealsthe generic differences better than many words.The distinction is often difficult for small-celledspecies, whose generic assignment is frequentlypurely conventional." If one follows Geitler'sadvice, the most striking difference revealed byhis illustrations concerns the shape of cells afterdivision. In most species of Gloeocapsa, a gradual

constriction occurs in the plane of division, sothat daughter cells are spherical after separation.In Chroococcus, on the other hand, there is littleor no constriction in the plane of division, whichappears to be occupied by a cell plate: thedaughter cells are hemispherical after division.This form is maintained through the followingdivision, with the result that in ensheathedpackets of four cells each cell has the shape of aquarter sphere. If this is indeed the basic distinc-tion between the two genera, our ensheathedstrains of group IIB are assignable to Gloeocapsa.The form genus Microcystis is clearly an un-

satisfactory one, as Geitler admitted. It includesboth rods and cocci, dividing in a single plane orin more than one plane. The strain of Microcystisaeraginosa in our collection (typological groupII) can be distinguished from the strains oftypological group IIA (Aphanocapsa) by onlyone major character, the formation of gasvacuoles. However, Geitler records this as avariable property of Microcystis. Evidently, thisgenus needs to be defined in more restrictiveterms, if it is to be maintained at all. The ma-jority of the described species possess sphericalcells that divide in two or three planes. The genusshould probably be restricted to gas-vacuolateunicellular blue-green algae of this type.Of the genera defined by the possession of

ellipsoidal or cylindrical cells that divide in oneplane, Gloeothece is distinguished from Aphano-thece and Synechococcus by the presence ofsheaths. The validity of the genus Gloethece seemshighly questionable to us. In ensheathed coccoidstrains of group IIB (= Gloecapsa), there is a

196 STANIER ET AL. BACTrERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



marked tendency to synchronous division of thecells within a single colony. Consequently, at acertain stage of development, all the cells in acolony may be ellipsoidal. Observed in a fieldcollection, such a colony would almost certainlybe identified as Gloeothece. Hence, this genusmay be synonymous with Gloeocapsa.The distinction between Aphanothece and

Synechococcus seems to rest entirely on theproperty of colony formation, determined by thepresence of an extensive slime layer. This ischaracteristic of Aphanothece and absent fromSynechococcus. In terms of this distinction, allstrains of typological group I appear to be assign-able to Synechococcus.

Table 12 shows the probable correspondencesbetween our typological groups I and It and formgenera of the family Chrococcaceae recognizedby Geitler. Even these proposed correspondencescan be maintained only by making certain revi-sions of the Geitlerian definitions-notably, theomission of colony structure as a generic charac-ter for organisms that do not form sheaths andthe restriction of Microcystis to spherical organ-isms that divide in two or three planes andproduce gas vacuoles. Although these revisionslead to genera that appear satisfactory for de-terminative purposes, it must be emphasized thatboth Synechococcus and Aphanocapsa as now de-fined encompass organisms that differ widely inDNA base composition and in certain pheno-typic respects. Eventually, a subdivision of eachof these genera into at least two genera shouldprobably be undertaken. However, before anynew taxonomic proposals with respect to the Chro-ococcales are put forward, one essential and long-overdue action must be taken, the establishmentof a reasonable nomenclatural date of departurefor the order.We formally propose that the starting date

should be C. Nigeli's monograph GattungenEinzelliger Algen (48), published in 1849. It con-tains good descriptions of most of the majorgenera of unicellular blue-green algae now recog-nized, accompanied by excellent line drawings toillustrate their salient structural properties. In-ternational agreement to accept Nageli's mono-graph as the oldest source of valid generic andspecific names in the Chroococcales would largelyprevent nomenclatural instability of the kind in-troduced by the monograph of Drouet and Daily(20), which has been a source of great confusionduring the past 15 years.

MERISMOPEDIA: A GROWTH FORM OFAPHANOCAPSA?

One of the oldest form genera in the Chro-ococcaceae is Merismopedia. It comprises orga-

nisms with spherical cells not enclosed by sheaths,which divide synchronously in two planes to formflat, rectangular colonies that contain 16, 32, 64,or even more cells in regular two-dimensionalarray (28). Unicellular blue-green algae of thisdistinctive type have been observed frequently infield collections, and many species have beennamed.

Pringsheim (61) was the first to describe cul-tured material of Merismopedia. Two of the threestrains that he examined were unquestionablyauthentic, since he had established them by directisolation of organisms with the characteristiccolony form. In culture, however, the cells oc-curred mostly as pairs or irregular aggregates;Pringsheim's sketch shows cell arrangementsthat are not significantly different from thosecharacteristic of young, healthy cultures of ourstrains of group IIA, which we believe to be as-signable to the genus Aphanocapsa.By chance, we recently found that under special

circumstances some strains of group IIA canproduce Merismopedia-like colonies. Liquid cul-tures of a number of group IIA strains had beengrown with continuous illumination for a week at37 C and subsequently left undisturbed for over amonth at room temperature on a table, wherethey were exposed to a diurnal light-dark cycle.Wet mounts showed that two of these strains,6807 and 6808, had produced aggregates with thetypical form and perfectly synchronized divisioncharacteristic of Merismopedia. These aggregateswere interspersed with single cells, pairs of cells,and irregular clumps, many obviously dead orsenescent. The Merismopedia-like colonies wereparticularly well developed in the culture ofstrain 6808 (Fig. 15). Both strains had been inpure culture for over 2 years, and repeated micro-scopic examinations of material grown undermore favorable conditions had not previously re-vealed such aggregates.

This experience suggests that the ability ofAphanocapsa strains to produce Merismopedia-like growth forms may be relatively common.Binary fission in two planes at right angles to oneanother is clearly a necessary condition, but nota sufficient one. The history of the cultures inwhich these colonies developed points to twoother factors that may be of importance: im-posed synchrony of division (by a diurnal light-dark cycle) and minimal disturbance of the me-dium, so that the slowly growing aggregates arenot physically disrupted.Coupled with the report of Pringsheim (61), our

observations cast some doubt on the validity ofthe genus Merismopedia. Most (if not all) of thedescribed species may well be growth forms ofAphanocapsa species, which have developed in

VOL. 35, 1971 197


http://mm

br.asm.org/

Dow

nloaded from


198 STANIE]

TABLE 12. Probable correspondences between form generaand typological groups

Typo- ofne s~ Gaslogical Cell shape div- vac- Genusgroup sion Cin uoles

IIA Spherical 2 or _ _ Aphano-3 capsa

IIB Spherical 2 or + - Gloeocapsa3

IIC Spherical 2 or - + Microcys-3 UiS

I Cylindrical, el- 1 _ Synecho-lipsoidal, or coccusspherical

R ET AL. BACTERIOL. REV.

nature under conditions that favor the formationof Merismopedia-like colonies. The mainten-ance of Merismopedia as a separate genus wouldbe justified only if it could be shown that the for-mation of such colonies by some coccoid blue-green algae is a constant property, of regular oc-currence in cultured material. The possible exist-ence of such blue-green algae certainly cannot bediscounted, since the same mode ofgrowth is char-acteristic of the bacterium Lampropedia hyalina,and is maintained in pure cultures of this organism(59). L. hyalina has a very complex wall structure,and the two outer layers of the wall enclose therectangular tablets of cells (13, 53). Its distinctive

''''em~~~~~#*'~~*WW

loyFIG. 15. Merismopedia-like aggregates in an old culture of strain 6808. (See text for growth conditions.)

Representative fields, also containing many single cells and cell pairs, are shown in Fig. I5a and J5b. Figures15c and 15d show two particularly regular, 16-celled colonies of which the cells are undergoing division in almostperfect synchrony.

are'*$


http://mm

br.asm.org/

Dow

nloaded from



growth habit is therefore probably determined bythe special properties of the cell wall.

GENERA CHLOROGLOEA ANDCHLOROGLOEOPSIS

One of our unicellular isolates, strain 6712(typological group III) does not appear to beassignable to the family Chroococcaceae. Its struc-ture and development (Fig. 9) correspond well tothe descriptions and illustrations by Geitler (28)of the genus Chlorogloea (family Entophysalida-ceae). However, as shown in Table 13, strain 6712differs in a number of respects from Chlorogloeafritschi Mitra (46), the only other blue-green algaavailable in culture that has been placed in thisgenus.The formation of heterocysts by C. fritschii,

which was discovered (25, 26) subsequent to itsoriginal description (46), excludes this strain bydefinition from the Chroococcales. Fay, Kumar,and Fogg (26) proposed its transfer to the genusNostoc (order Nostocales), whereas Mitra andPandey (46a) suggested the creation for it of anew genus Chlorogloeopsis, which they assign tothe order Stigonematales. The vegetative growthof this organism (Fig. 10) does not resemble thatof Nostoc species, and for this reason the taxo-nomic solution suggested by Mitra and Pandey(46a) appears to be the more appropriate one.

Strain 6712 fully conforms to the description ofthe form genus Chlorogloea. It does not fix nitro-gen or develop heterocysts, and it has not beenobserved to form chains of cells by successive di-visions in a single plane. This strain is apparentlythe first true representative of the genus Chloro-gloea to have been obtained in pure culture.

SPECIATION IN THECHROOCOCCACEAE

Structural characters, DNA base composition,and physiological and chemical properties permitthe subdivision of typological groups I and IIinto series of strain clusters, most of which

TABLE 13. Comparison of the properties of strain6712 and Chlorogloea fritschii

Determination Strain C. fritschii6712

Deoxyribonucleic acid base 40.8 42.9compositiona

Formation of short filaments - +Formation of heterocysts _ +Nitrogen fixation _ +Content of polyunsaturated Low High

acids

a Expressed as moles % guanine plus cytosine.

probably deserve recognition as separate species.However, the problem of assigning specificnames is difficult. In the genera to which thesestrains appear to belong, many nomenspecieshave already been proposed, but the informationcontent associated with existing specific names isso small that a single specific epithet could fre-quently be applied to several different strains orstrain-clusters, readily distinguishable from oneanother if properties that have hitherto not beenconsidered in the taxonomy of the Chroococcaceaeare taken into account. Eventually, a solution tothis problem must be found. We can see no al-ternative to deciding arbitrarily that an existingspecific name be attached to a particular strain,which would thereafter constitute the neotypefor the species in question. Other strains of similarstructure which differ markedly from the neotypein DNA base composition or in physiological andchemical respects could then be assigned newspecific names. This is the solution which wasadopted in bacteriological nomenclature (71) tofix the type of B. subtilis, of which the originaldescription, based solely on structural criteria, waslater found to apply to more than one specificentity.At the present time, we are unwilling to make

formal proposals to this effect. We shall indicate,therefore, for groups I and II the strains and strainclusters which in our j udgement merit specificrecognition, but without attaching specific namesto them except in the rare cases in which identifi-cation with an existing nomenspecies can bemade without ambiguity.

Subdivisions in Typological Group I(Synechococcus)

Most of the strains of typological group I areimmotile short rods or cocci, 2.0 ,um or less inwidth, without polar granules. Structurally speak-ing, they present few distinguishing charactersand might well be placed in a single species by ataxonomist who considered only structural prop-erties. Nevertheless, these 20 strains can be sub-divided into a series of distinct clusters whenadditional characters are taken into consideration(Table 14). Clusters 1 and 2 are distinguished bythe GC content (66 to 71 moles %) of their DNA,at least 10 moles % higher than the GC content ofthe DNA of all other strains in group I. The 10strains of cluster 1 are all very small rods or cocci,ranging in width from 0.8 to 1.5 jm. Althoughthey show slight variations in cell size and in DNAbase composition, they appear very similar in allother respects. One strain of high GC content,7001, has been placed in a separate cluster 2,since it differs from the strains of cluster 1 withrespect to fatty-acid composition, phycocyanin:

VOL. 35, 1971 199


http://mm

br.asm.org/

Dow

nloaded from


STANIER ET AL.

TABLE 14. Strain clusters of typological group I (Synechococcus)

Cell width Polar(ILM) grn-

0.81.5

1.0

1.51.52.033

2.5-3.0

1.5

Mo-tility

+

Tempmaxi-

mum (C)

Contentof poly-unsat-uratedfattyacids

Requirement for

Vit- 1%amin NaCiB 12

Phyco- Peak ratio ofertri chlorophyll a

35-37 Low - _ - 1.19-1.57 66-71

41>53

43

374339

<354137

37

HighLow

Low

LowHighHighHighHighHigh

High

+


chlorophyll a ratio, and temperature maximumfor growth. It should also be noted that this strainis of marine origin (although it grows well in me-um BG-1 1 without added NaCl), whereas thestrains of cluster 1 are all isolates from freshwater.

Strain 6907 (cluster 1) was received under thename Synechococcus elongatus. According toGeitler (28), S. elongatus has the smallest cellsof any named Synechococcus species, 1.4 to 2 Amin width. These dimensions fit moderately wellfor 6907 and other strains of cluster 1. However,the same name could be applied with equalcogency to clusters 2, 4, or 5. Strain 7001 (cluster2) was identified by Drouet (Van Baalen, per-

sonal communication) as Anacystis marina. Forreasons already discussed, we do not accept thisproposal.Among the strains that are similar in cell struc-

ture to those of clusters 1 and 2 but contain DNAof lower GC content, clusters 3, 4, and 5 resembleone another in most phenotypic respects but canbe distinguished by DNA base composition, aswell as by one phenotypic trait-the temperaturemaximum for growth. The strains of cluster 3(DNA base composition: 52 to 54 moles % GC)were isolated from thermal habitats, and theyhave a temperature maximum of 53 C or higher.The cells are relatively long rods, and this struc-tural property also serves to distinguish them fromstrains of clusters 1, 2, 4, and 5 (See Fig. 1 and 2).One of these strains was received under the name

Synechococcus lividus. The remaining two strainswere received from R. W. Castenholz as Synecho-coccus species. However, Castenholz (12) hasalso applied the specific name lividus to thermo-philic Synechococcus strains of this type, and we

believe that it can be considered the correctspecific epithet.The three strains of cluster 4 and the single

strain of cluster 5 are all isolates from fresh water;they can be construed as nonthermophiliccounterparts of S. lividus. The temperature maxi-mum for the strains of cluster 4 is unusually high(43 C), a character that distinguishes them fromother strains of similar structure from nonthermalhabitats. The single strain of cluster 5 has a muchlower temperature maximum (37 C) and differsin GC content by 5 moles % from the strains ofcluster 4.One strain of cluster 4 (6908) was received

under the name S. cedrorum. However, the widthof the cells in cluster 4 is considerably less thanthe range of 3 to 4.3 j.m given by Geitler (28) forS. cedrorum, and this name is therefore clearlyincorrect. Cluster 4 also includes 6301, the classicisolate by M. B. Allen (1952), to which Drouetapplied the name Anacystis nidulans. Padmajaand Desikachary (52) carefully compared the di-mensions of A. nidulans and of a number ofSynechococcus strains and concluded that Allen'sstrain corresponded in size to Synechococcuselongatus, an opinion with which we concur. Un-fortunately, however, this name could be ap-

Cluster

1 0.8-1.5 _

Deoxyri-bonucleicacid basecompo-sitiona

23

4

56a6b789a

9b

Strains included

6307, 6603,6706, 6707,6708, 6709,6710, 6713,6907, 6911

70016715, 6716,6717

6301, 6311,6908

631270027003660569106802, 6903

6901

+

1.030.76-0.83

0.80-1.08

0.880.961.170.870.78

0.79-0.98

0.98

6952-54

55-56

504949.5474846

45

+ or

200 BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



plied with equal justification to strain clusters 1and 5, which cannot be clearly distinguished fromcluster 4 either by structural properties or byhabitat. The only solution to this problem is todesignate a neotype strain of S. elongatus. Onhistorical grounds we believe that M. B. Allen'sstrain (6301) would be the most appropriateneotype. If this suggestion is accepted by othertaxonomists, the name S. elongatus will becomefixed to strain cluster 4. Other specific epithetscan then be proposed for strain clusters that aredistinguishable from cluster 4 only by nonstruc-tural characters.

Cluster 6 contains two strains of marine origindistinguished by their requirement for vitaminB12 (73). Both show a pronounced tendency tochain formation (see Fig. 1). Although virtuallyidentical in DNA base composition (49 to 49.5moles % GC), they differ from one another byseveral phenotypic characters and have beenplaced in separate subclusters, 6a and 6b. Strain7002 (subcluster 6a) has cylindrical cells 1.5 jsmwide, whereas the cells of strain 7003 (subcluster6b) are 2.0 jsm wide and spherical immediatelyafter division. The temperature maxima differmarkedly (Table 14). Strain 7003 shows a saltrequirement and will not grow satisfactorilyunless the culture medium is supplemented with1% (w/v) NaCI. The other two strains of marineorigin in the collection (7001 and 7002) appearto grow equally well with or without the additionof NaCl.Drouet assigned the two strains of cluster 6 to

different genera, identifying 7002 as Agmenellumquadruplicatum and 7003 as Coccochloris elabens(Van Baalen, personal communication). In Geit-lerian terms, both should be assigned to Synecho-coccus, but we have not been able to identify themwith named species of this genus.Two strains of group I (6605 and 6910) can be

distinguished from the strain clusters so far dis-cussed by their relatively large, ellipsoidal cells,3 Mum in width. They differ from one another withrespect to motility, temperature maximum, andthe presence of phycoerythrin, and they have beenassigned to separate clusters, 7 and 8. In terms ofcell size, they fit the specifications given byGeitler (28) for Synechococcus cedrorum; hereagain, however, a single specific epithet could beapplied with equal justification to two organismsthat are clearly representatives of different species.This problem can be solved only by the designa-tion of a neotype strain for S. cedrorum.The three strains of group IB (6802, 6901, and

6903) are collectively distinguishable from othermembers of group I by a structural character, theregular presence of refractile polar granules. Theyalso differ from all other strains of group I, except

6910, by their rapid gliding movement. To thebest of our knowledge, unicellular blue-greenalgae of this type have not been previously de-scribed. Although their assignment to a new genusmight be appropriate, we shall regard them pro-visionally as members of the genus Synechococcusand assign them to a special cluster 9. Strains 6802and 6903 (subcluster 9a) can be distinguishedfrom strain 6901 (subcluster 9b) by their con-siderably larger cells, 2.5 to 3.0 Mm in width; thecells of strain 6901 are 1.5 Mum in width. The twostrains assigned to cluster 9a differ with respect tochain formation (absent from strain 6802) andthe presence of phycoerythrin (absent from strain6903).

Subdivisions in Typological Group II

(Aphanocapsa, Gloeocapsa, Microcystis)Table 15 summarizes the distinguishing proper-

ties of strain clusters in typological group II.Eleven strains are assignable to Aphanocapsa, andthey can be subdivided into five clusters. Cluster1 comprises five strains with DNA of high GCcontent (46 to 48 moles %) and of high polyun-saturated fatty-acid content, and this cluster is in-ternally homogeneous in all phenotypic respects.The remaining six strains assigned to Aphano-

capsa, all with DNA of low GC content (35 to 37moles %) and of low polyunsaturated fatty-acidcontent, are subdivisible into four clusters by dif-ferences in cell size and temperature maximum andby the presence or absence of phycoerythrin.None of these clusters can be identified satisfac-torily with existing nomenspecies.The one strain (7005, group IIC) assigned to

the genus Microcystis can be identified by cellsize and gas-vacuole formation as Microcystisaeruginosa, but in our hands it has never formedthe distinctive colonial aggregates described ascharacteristic of this species in nature. Zehnderand Gorham (77), who studied a crude culture ofanother strain of M. aeruginosa, reported that itsability to form colonies was rapidly lost after iso-lation. Hence, this character is evidently an un-stable one.The somewhat tenuous nature of the generic

distinction between Microcystis and Aphano-capsa must be emphasized; it can be maintainedonly by taking into account the property of gas-vacuole formation. With respect to both DNAbase composition and polyunsaturated fatty-acidcontent, M. aeruginosa resembles the strains as-signed to cluster 1 of Aphanocapsa, although itcan be distinguished from them at the specificlevel by its larger cell size and lower tempera-ture maximum, as well as by the property of gas-vacuole formation. It is conceivable that a more

201VOL. 35, 1971


http://mm

br.asm.org/

Dow

nloaded from


TABLE 15. Differential properties of strains of group II (Aphanocapsa, Gloeocapsa, Microcystis).I

Sheaths GasvacuolesDiameter Temp N2of cells maxi- fixation

(am) mum(C)

_ _ 2-3 37-39 - _

+

4-53-46-7

7

46

3737-39

4137

<3535-39


satisfactory generic arrangement of nonen-

sheathed strains with spherical cells that divide intwo or three planes could be achieved by therecognition of two genera, distinguished pri-marily by DNA base composition and fatty-acidcontent. In this event, M. aeruginosa and cluster1 of Aphanocapsa would comprise two species inone of these genera, and the remaining strainclusters of Aphanocapsa a series of species in theother genus.The two strains assignable to Gloeocapsa are

readily distinguishable from all other strains ofgroup II by their distinctive, multilaminatesheaths, as well as by one physiological property,the capacity for nitrogen fixation. Apart from a

difference in temperature maximum (35 C for6909, 39 C for 6501) they appear phenotypicallyidentical and also have nearly identical DNA basecompositions. We consider them to be representa-tives of a single species, but we have been unableto identify it with certainty with any of the nu-merous nomenspecies of Gloeocapsa describedby Geitler (28).

CONCLUDING COMMENTSMost of the taxonomic conclusions which have

been presented here are tentative ones. This isascribable primarily to the difficulties encoun-tered in assigning strains characterized on thebasis of pure-culture studies to taxa originallyproposed on the basis of an entirely differentmethodology. In fact, the principal conclusion tobe drawn from our work is a methodological one.We have shown beyond any question that themembers of the Chroococcales, like most otherunicellular bacteria, cannot be adequately char-acterized solely by gross structural properties;many good species remain unrecognizable interms of these criteria. We believe accordinglythat all future taxonomic work on the Chroococ-

cales should be based on the isolation and compari-son of pure strains.The isolation of unicellular blue-green algae

in pure culture is admittedly time consuming andit demands a certain degree of microbiologicalcompetence. It might be argued, therefore, thatfuture taxonomic work would be conducted more

expeditiously and conveniently on impure (so-called unialgal) cultures. This hope is probablyillusory, since properties now recognizable as

being of taxonomic importance (the entire rangeof physiological properties, fatty-acid composi-tion, DNA base composition) cannot be deter-mined reliably on such material. Furthermore,when one deals with microorganisms as small andstructurally unspecialized as are many membersof the Chroococcaceae, the unialgal state of an im-pure culture can never be taken for granted, un-

less its blue-green algal component has beencloned by single-cell isolation or by plating. Thedanger of assuming that a culture is unialgal onthe basis of inadequate evidence can be specifi-cally documented from our experience with twoimpure cultures received from other laboratories.The first case history concerns a supposedly

unialgal culture, labeled Gloeocapsa alpicola1051, which we received in 1963 and from whicha pure culture (strain 6308, typological group IIA)was then isolated. Some years later, fearing thatstrain 6308 had been lost, we again obtained aculture of G. alpicola 1051 from the originaldonor, and again proceeded to isolate a pure cul-ture from it. On this occasion, a totally differentorganism (strain 6910, typological group IA) wasisolated. The properties of strains 6308 and6910 are compared in Table 16. The two purestrains do not differ very markedly in size andmicroscopic appearance (see Fig. 1 and 4), al-though the cells of 6910 are definitely ellipsoidal,whereas those of 6308 are spherical. The differ-

Phyco-erythrinGenus

Aphanocapsa

MicrocystisGloeocapsa

Cluster

1

2345

11

Polyun-saturated

fattyacids

High

LowLowLowLow

HighLow

Deoxyri-bonucleicacid basecompo-sitiona

46-48

3536-37

3636

4541+

+

Strains included

6702, 6714,6803, 6805,6806

63086701, 67116807, 68086804

70056501, 6909+

202 STANIER ET AL. BACTERIOL. REV.


http://mm

br.asm.org/

Dow

nloaded from



TABLE 16. Properties of the two strains isolated from a sup-posedly unialgal culture, Gloeocapsa alpicola 1051

Planes Deoxy-Poynof ribonu- Polyun-

Cell ofes Nl- cleic rate-Strain width Cell shape succe tilo- acid rated(JAm) div- ti base atcid

vision compo- acidensitiona otn

6308 4-5 Spherical 2 _ 35 Low6910 3 Ellipsoidal 1 + 48 High

aExpressed as moles % guanine plus cytosine.

ence between them with respect to planes ofsuccessive division is also not easy to recognizeby simple microscopic examination, since in bothstrains daughter cells usually separate soon afterthe completion of division. If one did not suspectthat G. alpicola 1051 contained two differentspecies of unicellular blue-green algae, its micro-scopic appearance might well be interpreted as

reflecting clonal variation of a single strain. How-ever, the many differences between the two purestrains isolated from G. alpicola 1051 show thatit was bialgal, not unialgal.The second case history concerns a culture re-

ceived with the label Microcystis aeraginosaNRC-1. This culture was unquestionably unialgalwhen we received it; however, the blue-greenalga that it contained, strain 6911, was a small-celled Synechococcus of high GC content (Fig.2), completely different in microscopic appear-

ance from M. aeruginosa (Fig. 8). Although theprevious strain history of 6911 is complex (seeTable 1), its strain designation (NRC-1) and his-tory indicate that it is derived from a culture thatonce contained M. aeruginosa, and was thenbelieved to be unialgal (36, 77). The methodsused to isolate this strain of M. aeruginosa did notinclude cloning on plates. As a result of the smallcell size of 691 1, its pigmentation is not detectablemicroscopically. Hence, its presence in a crudeculture of M. aeruginosa that also contained bac-teria could easily have escaped notice. Further-more, if M. aeruginosa had died out during latertransfers, this fact would not have been evidentmacroscopically, since the culture would stillhave appeared blue-green. We suspect that strain6911 is derived from a culture that was once bial-gal and also contained M. aeruginosa, althoughthis inference cannot be definitely proven.

ACKNOWLEDGMENTS

We thank all those who have contributed to the assembly ofour collection of unicellular blue-green algae. They include col-leagues in other institutions who have sent us pure or unialgalmaterial: Mary B. Allen, D. S. Berns, R. W. Castenholz, G. P.Fitzgerald, R. Haselkorn, W. Koch (Sammlung von Algenkulturender Universitat Gottingen), R. C. Starr (Indiana University cul-ture collection), and C. Van Baalen. Even more important, they

include past or present members of our research group who havehelped with the often tedious work of purification: Mary M. Allen,J. Hauxhurst, A. Neilson, and R. Rippka. We are also grateful toChristine Kenyon for permission to include some of her unpub-lished data on the cellular fatty-acid composition of unicellularblue-green algae.

The work described here, which has extended over almost adecade, has been supported by several grants from the NationalScience Foundation (GB-1419, GB-4112, GB-5748), as well as bygrant AI-1808 from the National Institutes of Allergy and Infec-tious Diseases.

LITERATURE CITED

1. Allen, M. B. 1952. The cultivation of Myxophyceae. Arch.Mikrobiol. 17:34-53.

2. Allen, M. M. 1966. Studies on the properties of some blue-green algae. Ph.D. Dissertation, University of California,Berkeley.

3. Allen, M. M. 1968. Simple conditions for growth of unicellu-lar blue-green algae on plates. J. Phycol. 4:1-3.

4. Allen, M. M., and R. Y. Stanier. 1968. Growth and divisionof some unicellular blue-green algae. J. Gen. Microbiol.51:199-202.

5. Allen, M. M., and R. Y. Stanier. 1968. Selective isolation ofblue-green algae from water and soil. J. Gen Microbiol.51:203-209.

6. Allsopp, A. 1969. Phylogenetic relationships of the Procaryotaand the origin of the eucaryotic cell. New Phytol. 68:591-612.

7. Bellamy, A. R., and R. K. Ralph. 1968. Recovery and puri-

fication of nucleic acids by means of cetyltrimethylammo-nium bromide, p. 156-160. In L. Grossman and K. Moldave(ed.), Methods in enzymology, vol. 12, part B. AcademicPress Inc., New York.

8. Biggins, J. 1969. Respiration in blue-green algae. J. Bacteriol.99:570-575.

9. Bortels, H. 1940. Ueber die Bedeutung des Molybdins furstickstoffbindende Nostocaceen. Arch. Mikrobiol. 11:155-186.

10. Carr, N. G., and I. W. Craig. 1970. The relationship betweenbacteria, blue-green algae and chloroplasts, p. 119-143. InJ. B. Harborne (ed.), Phytochemical phylogeny. AcademicPress Inc., London.

11. Castenholz, R. W. 1969. Thermophilic blue-green algae andthe thermal environment. Bacteriol. Rev. 33:476-504.

12. Castenholz, R. W. 1969. The thermophilic cyanophytes ofIceland and the upper temperature limit. J. Phycol. 5:360-368.

13. Chapman, J. A., R. G. E. Murray, and M. R. J. Salton. 1963.The surface anatomy of Lampropedia hyalina. Proc. Roy.Soc. London Ser. B. 158:498-513.

14. Cohn, F. 1853. Untersuchungen uber die Entwickelungsge-schichte mikroskopischer Algen und Pilze. Nov. Act. Acad.Leop. Carol. 24:103-256.

15. Cohn, F. 1875. Untersuchungen iiber Bakterien. II. Beitr. Biol.Pflanz. 1(3):141-207.

16. Copeland, H. F. 1938. The kingdoms of organisms. Quart.Rev. Biol. 13:383-420.

17. Crabtree, K., and E. McCoy. 1967. Zoogloea ramigera Itzig-sohn, identification and description. Int. J. Syst. Bacteriol.17:1-10.

18. Crespi, H. L., S. E. Mandeville, and J. J. Katz. 1962. Theaction of lysozyme on several blue-green algae. Biochem.Biophys. Res. Commun. 9:569-573.

19. Desikachary, T. V. 1959. Cyanophyta. Indian Council of Ag-ricultural Research, New Delhi.

20. Drouet, F., and W. A. Daily. 1956. Revision of the coccoid

Myxophyceae. Butler Univ. Bot. Stud. 12:1-122.

21. Echlin, P., and I. Morris. 1965. The relationship between blue-

green algae and bacteria. Biol. Rev. (Cambridge) 40:143-187.

22. Edelman, M., D. Swinton, J. A. Schiff, H. T. Epstein, and

VOL. 35, 1971 203


http://mm

br.asm.org/

Dow

nloaded from


204 STANIER ET AL.

B. Zeldin. 1967. Deoxyribonucleic acid of the blue-greenalgae (Cyanophyta). Bacteriol. Rev. 31:315-331.

23. Edwards, M. R., D. S. Berns, W. C. Chiorse, and S. C. Holt.1968. Ultrastructure of the thermophilic blue-green alga,Synechococcus lividus Copeland. J. Phycol. 4:283-298.

24. Fay, P. 1965. Heterotrophy and nitrogen fixation in Chloro-gloea fritschii. J. Gen. Microbiol. 39:11-20.

25. Fay, P., and G. E. Fogg. 1962. Studies in nitrogen fixation byblue-green algae. III. Growth and fixation in Chlorogloeafritschii Mitra. Arch. Mikrobiol. 42:310-321.

26. Fay, P., H. D. Kumar, and 0. E. Fogg. 1964. Cellular factorsaffecting nitrogen fixation in the blue-green alga Chlorogloeafritschii. J. Gen. Microbiol. 35:351-360.

27. Fogg, G. E. 1942. Studies on nitrogen-fixing blue-green algae.I. Nitrogen fixation by Anabaena cylindrica Lemm. J. Exp.Biol. t9:78-87.

28. Geitler, L. 1932. Cyanophyceae, p. 1-1 196. In R. Kolkwitz (ed.),Kryptogamenfiora von Deutschland, Osterreich und derSchweiz. Vol. 14. Akademische Verlagsgesellschaft, Leipzig.

29. Gould, G. W., and A. D. Hitchins. 1963. Sensitization of bac-terial spores to lysozyme and hydrogen peroxide with rea-gents which rupture disulphide bonds. Nature 197:622.

30. Hill, L. R. 1966. An index to deoxyribonucleic acid base com-positions of bacterial species. J. Gen. Microbiol. 44:419-437.

31. Hoare, D. S., S. L. Hoare and R. B. Moore. 1967. The photo-assimilation of organic compounds by autotrophic blue-green algae. J. Gen. Microbiol. 49:351-370.

32. Hoare, D. S., S. L. Hoare, and A. J. Smith. 1969. Assimilationof organic compounds by blue-green algae and photosyn-thetic bacteria. Prog. Photosynthesis Res. 3:1570-1573.

33. Holm-Hansen, 0. 1968. Ecology, physiology and biochemis-try of blue-green algae. Annu. Rev. Microbiol. 22:47-70.

34. Holton, R. W., H. H. Blecker, and T. S. Stevens. 1968. Fattyacids in blue-green algae: possible relation to phylogeneticposition. Science 160:545-547.

35. Horton, A. A. 1968. NADH oxidase in blue-green algae.Biochemn. Biophys. Res. Commun. 32:839-845.

36. Hughes, E. D., P. R. Gorham, and A. Zehnder. 1958. Toxicityof a unialgal culture of Microcystis aeruginosa. Can. J.Microbiol. 4:225-236.

37. Jarosch, R. 1962. Gliding, p. 573-581. In R. A. Lewin (ed.),Physiology and biochemistry of algae. Academic Press Inc.,New York.

38. Kates, M. 1964. Bacterial lipids. Adv. Lipid Res. 2:17-90.39. Kenyon, C., and R. Y. Stanier. 1970. Possible evolutionary

significance of polyunsaturated fatty acids in blue-greenalgae. Nature 227:1164-1166.

40. Kiyohara, T., Y. Fujita, A. Hattori, and A. Watanabe. 1960.Heterotrophic culture of a blue-green alga, Tolypothrixtenuis. J. Gen. Appl. Microbiol. 6:176-182.

41. Koch, W. 1965. Cyanophyceenkulturen. Anreicherungs- undIsolierverfahren, p. 415-431. In H. Schlegel (ed.), Anreich-erungskultur und Mutantenauslese. Fischer, Stuttgart.

42. Kratz, W. A., and J. Myers. 1955. Nutrition and growth ofseveral blue-green algae. Amer. J. Bot. 42:282-287.

43. Kunisawa, R., and G. Cohen-Bazire. 1970. Mutations ofAnacystis ntidulants that affect cell division. Arch. Mikrobiol.70:49-59.

44. Leach, C. K., and N. G. Carr. 1969. Oxidative phosphoryla-tion in an extract of Anabaena variabilis. Biochem. J.112:125-126.

45. Marmur, J. 1961. A procedure for the isolation of deoxyribo-nucleic acid from microorganisms. 3. Mol. Biol. 3:208-218.

46. Mitra, A. K. 1950. Two new algae from Indian soils. Ann.Bot. (London) 14:457-464.

46a. Mitra, A. K., and D. C. Pandey. 1966. On a new genus ofthe blue-green alga Chlorogloeopsis with remarks on theproduction of heterocysts in the alga. Phykos 5:106-114.

47. Murray, R. G. E. 1962. Fine structure and taxonomy of bac-teria, p. 119-144. In G. C. Ainsworth and P. H. A. Sneath(ed.), Microbial classification. University Press, Cambridge.

48. Ndgeli, C. 1849. Gattungen einzelliger Algen, physiologisch

BACTERIOL. REV.

und systemnatisch bearbeitet. Neue Denkschr. Allgem.Schweiz. Gesellsch. 10:1-139.

49. Nichols, B. W., R. V. Harris, and A. T. James. 1965. Thelipid metabolism of blue-green algae. Biochem. Biophys.Res. Commun. 20:256-262.

50. Nichols, B. W., and B. J. B. Wood. 1968. The occurrenceand biosynthesis of gamma-linolenic acid in a blue-greenalga, Spirulina platensis. Lipids 3:46-50.

51. OhEocha, C. Phycobilins, p. 421-435. In R. A. Lewin (ed.),Physiology and biochemistry of algae. Academic Press Inc.,New York.

52. Padmaja, T. D., and T. V. Desikachary. 1968. Studies oncoccoid blue-green algae. I. Synechococcus elongatus andAnacystis nidulans. Phykos 7:62-89.

53. Pangborn, J., and M. P. Starr. 1966. Ultrastructure of Lampro-pedia hyalina. 3. Bacteriol. 91:2023-2030.

54. Parker, P. L., C. VanBaalen, and L. Maurer. 1967. Fattyacids in eleven species of blue-green algae: geochemicalsignificance. Science 155:707-708.

55. Pearce, 3., and N. G. Carr. 1967. The metabolism of acetateby the blue-green algae, Anabaena variabilis and Anacystisnidulans. 3. Gen. Microbiol. 49:301-313.

56. Pearce, J., and N. G. Carr. 1968. The incorporation andmetabolism of glucose by Anabaena variabilis. J. Gen.Microbiol. 54:451-462.

57. Peary, 3. A., and R. W. Castenholz. 1964. Temperature strainsof a thermophilic blue-green alga. Nature 202:720-721.

58. Pringsheim, E. G. 1949. The relationship between bacteriaand myxophyceae.. Bacteriol. Rev. 13:47-98.

59. Pringsheim, E. G. 1955. Lampropedia hyalina. Schroeter 1886and Vannielia aggregata n.g., n.s.p., with remarks on natu-ral and organized colonies of bacteria. 3. Gen. Microbiol.13:285-291.

60. Pringsheim, E. G. 1968. Kleine Mitteilungen uiber Flagellatetiund Algen. XVI. Lauterbornia (Anacystis) nidulans (Rich-ter) nov. gen. nov. comb. Arch. Mikrobiol. 63:1-6.

61. Pringsheim, E. G. 1968. Cyanophyceen-Studien. Arch. Mi-krobiol. 63:331-355.

62. Rippka, R., A. Neilson, G. Cohen-Bazire, and R. Kunisawa.1971. Nitrogen fixation by unicellular blue-green algae.Arch. Mikrobiol., 76:341-348.

63. Schildkraut, C. S., J. Marmur, and P. Doty. 1962. Determi-nation of the base composition of deoxyribonucleic acidfrom its buoyant density in CsCI. J. Mol. Biol. 4:430-443.

64. Schuster, M. L., A. K. Vidaver, and M. Mandel. 1968. Apurple-pigment-producing bean wilt bacterium, Coryne-bacterium flaccumfaciens var. violaceum, n. var. Can. J.Microbiol. 14:423-427.

65. Sheridan, R. P., and R. W. Castenholz. 1968. Production ofhydrogen sulphide by a thermophilic blue-green alga. Na-ture 217:1063-1064.

66. Shibata, K., A. A. Benson, and M. Calvin. 1954. The absorp-tion spectra of suspensions oul iving microorganisms. Bio-chim. Biophys. Adta 15:461-470.

67. Silva, P. 1962. Classification of algae, p. 834 (footnote 1). InR. A. Lewin (ed.), Physiology and biochemistry of algae.Academic Press Inc., New York.

68. Smith, A. 3., J. London, and R. Y. Stanier. 1967. Biochemicalbasis of obligate autotrophy in blue-green algae and thio-bacilli. 3. Bacteriol. 94:972-983.

69. Stanier, R. Y., and C. B. van Niel. 1941. The main outlinesof bacterial classification. 3. Bacteriol. 42:437-466.

70. Stanier, R. Y., and C. B. van Niel. 1962. The concept of abacterium. Arch. Mikrobiol. 42:17-35.

71. St. John Brooks, R., and R. S. Breed. 1937. Actions taken bythe second international microbiological congress in Lon-don, 1936, regarding bacteriological nomenclature. 3. Bac-teriol. 33:445-447.

72. Stransky, H., and A. Hager. 1970. Die Carotinoidmuster unddie Verbreitung des lichtinduzierten Xanthophyllcyclus; inverschiedenen Algenklassen. IV. Cyanophyceae und Rho-dophyceae. Arch. Mikrobiol. 72:84-96.


http://mm

br.asm.org/

Dow

nloaded from



73. Van Baalen, C. 1962. Studies on marine blue-green algae.Bot. Mar. 4:129-139.

74. van Niel, C. B. 1953. Classification and taxonomy of thebacteria and blue-green algae, p. 89-114. itt A century ofprogress in the natural sciences-1853-1953. Calif. Acad.Sci., San Francisco.

75. Wieringa, K. T. 1968. A new method for obtaining bacteria-

free cultures of blue-green algae. Antonie van LeeuwenhoekJ. Microbiol. Serol. 34:54-56.

76. Wyatt, J. T., and J. K. G. Silvey. 1969. Nitrogen fixation byGloeocapsa. Science 165:908-909.

77. Zehnder, A., and P. R. Gorham. 1969. Factors influencing thegrowth Microcystis aeruginosa Kutz. emend. Elenkin. Can.J. Microbiol. 6:645-660.

VOL. 35, 1971 205


http://mm

br.asm.org/

Dow

nloaded from


Documents

in U.S.A. Purification and Properties of Unicellular Blue-Green … · STANIER ET AL. be recognized as a major group of bacteria, distinguished from other photosynthetic bacteria