J. Cell Sci. 37, 8i-go (1977) 81Printed in Great Britain © Company of Biologists Limited 1977

CONTINUOUS DNA REPLICATION IN THE

NUCLEUS OF THE DINOFLAGELLATE

PROROCENTRUM MICANS (EHRENBERG)

S. A. FILFILAN AND D. C. SIGEE»Cytology Unit, Departments of Botany and Zoology,University of Manchester, Manchester, England

SUMMARYThe uptake of tritiated thymine into cells of a heterogeneous population of Prorocentruni

micans was investigated using light-microscope and electron-microscope autoradiography.Specificity of thymine uptake into DNA was demonstrated by the specific removal of labelfrom wax-embedded material using DNase and by the high degree of localization of nuclearlabel to chromosomes in the electron-microscope autoradiographs.

All nuclei, including both dividing and non-dividing cells, showed a substantial uptake oflabel, indicating that nuclear DNA synthesis in Prorocentrum micans is a continuous process.The level of DNA synthesis does show considerable variation, however, with very high levelsin some interphase nuclei.

The continuous replication of nuclear DNA provides further evidence of dinoflagellateaffinity to the prokaryotes, and indicates that Prorocentrum micans is a very primitive eukaryotecell.

INTRODUCTION

Considerable attention has been given in recent years to the fine structure ofdinoflagellates and the possibility that they may represent a condition intermediatebetween the prokaryotes and eukaryotes. Dodge (1965, 1966) suggested that dino-flagellates should be placed in a separate group - the Mesokaryota, on the basis ofprimitive nuclear features including an absence of nuclear spindle, membrane-attached chromosomes, permanent condensation of chromosomes throughout the cellcycle, and the absence of histochemicaUy detectable histones. Loeblich (1976), ina review that included more recent evidence relating to dinoflagellate evolution,supported the concept that these organisms possess both prokaryote and eukaryotefeatures.

The timing of DNA synthesis is a question of central interest in considering theevolutionary affinities of the dinoflagellates. Dodge (1966) postulated a continuoussynthesis, similar to that occurring in prokaryotes, whilst Loeblich (1976) has quotedrecent evidence in support of discontinuous synthesis, similar to that of eukaryotes.

The present work was carried out to determine the nature of DNA synthesis inProrocentrum micans. The genus Prorocentrum is of particular interest since thearrangement of the surface thecae places it in a primitive position within the dino-flagellates (Loeblich, 1976).

• Requests for reprints should be sent to this author.

82 S. A. Filfilan and D. C. Sigee

MATERIALS AND METHODS

Cell culture

Prorocentrum micans (Ehrenberg), obtained from the Cambridge Culture Centre (strainLB 1136/1), was grown in Erdschreiber medium at 22 °C under fluorescent illumination(400 ft. c. (4-3 x io3 lux)). A high proportion of dividing cells was obtained by giving cultures2 cycles of 32 h dark/16 h light, followed by 8 h dark/16 h light. Treatment with tritiatedthymine and counts of cell population were carried out in the final light period on 2 separateculture samples.

Population counts

Cell population counts were carried out at intervals from samples that had been filtered on toa squared filter membrane (Oxoid). The cell population shows a gradual increase during thefirst 2 h (Fig. 1), followed by a rapid increase to about 145 % (at 4 h) of the initial level. Thismeans that, during the period of thymine labelling (Fig. 1), approximately half the cells wereundergoing cell division and half the cells were in interphase.

50 r

40

30

20

10

. Y////////A

1 2 3 4 5 6 7

Time, h

Fig. 1. Percentage increase in cell population during the final light period. Each valueis the mean of two counts. The period of radioactive labelling is indicated (cross-hatched) with the time of fixation (arrow).

Labelling and fixation

Cells were labelled for 2 h during the final light period (Fig. 1) with tritiated thymine (Radio-chemical Centre, Amersham) which was injected into the culture to give a final concentration of10 fid/ml, 8p. act. 21 Ci/mM. Cells were subsequently washed in culture medium and fixed ineither glutaraldehyde or acetic ethanol. Fixation in 4% glutaraldehyde was carried out for2 h at 4 CC in 0-05 M sodium cacodylate buffer at pH 7-2. Cells were then washed in buffer,postfixed in buffered 2 % osmium tetroxide for 2 h, washed in buffer, dehydrated and embeddedin Spurr resin (Spurr, 1969). Thick (4-/WT1) sections were cut on glass knives and used eitherdirectly for light microscopy or were resectioned for electron microscopy using a resin slidetechnique (Sigee, 1976). Material for enzyme studies was fixed for 30 min in acetic ethanolat 20 °C, hydrated and washed in distilled water, then dehydrated and embedded in wax.

DNA replication in Prorocentrum 83

Enzyme treatments

Sections of acetic ethanol-fixed material were hydrated and washed in tap water for 20 minprior to the following enzyme treatments:

DNase II, ex bovine spleen (Koch-light), 10 mg/ml in 0-2 M sodium acetate buffer, pH 5-0,for 1 h.

RNase, ex bovine pancreas (Koch-light), i-o mg/ml in 02 M sodium phosphate buffer,pH 76, for 1 h.

Preparation of autoradiographs

Sections for light-microscope autoradiography were examined, and cells with a clear nucleusin median section were noted and photographed. The slides were then coated with Ilford G-5emulsion, left for 7 days at 4 °C, and processed using Kodak D-19 as developer. Ultrathinsections for electron-microscope autoradiography were coated with a monolayer of Ilford L-4emulsion using a wire loop, left for 3-4 months at 4 °C, and developed with either Ilford D-19or microdol-X.

RESULTS

Light-microscope autoradiography

Under the light microscope chromosomes are always apparent as discrete structures.The phase-contrast appearance of the chromosomes varies according to the stage ofthe cell cycle (Filfilan & Sigee, manuscript in preparation), ranging from dark (cellsundergoing mitosis and early interphase - Fig. 2 A) to pale (middle to late interphase,Fig- 3A)-

The nuclear grain counts for 40 cells are shown in Fig. 4. All of the nuclei examinedshowed a clear uptake of label. Very few silver grains occurred over the cytopiasm,and the background count was not sufficient to make any significant contribution tothe nuclear count (Figs. 2B, 3B). In some cells, with pale chromosomes, the level ofthymine incorporation into the nucleus is exceptionally high (Figs. 3 A, B). Themajority of cells, however, including cells with both pale and dark chromosomes, havenuclear counts in the range 20-100 grains (Figs. 2A, B).

The effect of DNase on the acetic acid-fixed material is to remove almost allradioactivity from the cells, whilst buffer alone has no effect. Treatment with RNasedoes not result in any detectable loss of radioactivity.

Electron-microscope autoradiography

Electron-microscope autoradiographs of a number of cells with varying nuclearstructure consistently show substantial levels of nuclear labelling, in agreement withthe light-microscope results. This correlation can be developed further using theresin slide technique for resectioning 4-/im-thick sections (Sigee, 1976). The auto-radiograph shown in Fig. 5 is derived from a cell with chromosomes which appeareddense under the light microscope, while that in Fig. 6 is derived from a cell with palechromosomes. In these, and all other cells examined, the majority of silver grainsappear to be over or closely associated with chromosomes and are relatively infrequentover nucleoplasm, nucleolus and nuclear envelope.

S. A. Filfilan and D. C. Sigee

2A 2B

Fig. 2. A, cell with dense chromosomes (c) photographed under phase contrast prior tocoating with photographic emulsion. Peripheral chloroplast (p) vacuoles (v). B, auto-radiograph of the cell in A, bright-field illumination. The nucleus contains approxi-mately 20 silver grains. Both x 4200.Fig. 3. A, interphase cell with pale chromosomes in a dense nuclear matrix. Legend asin Fig. 2 A. B, autoradiograph of the cell in A with bright-field illumination. Thenucleus is heavily labelled, with an estimated 110-120 silver grains. Both x 4200.

3B

DNA replication in Prorocentrum 85

The distribution of nuclear silver grains was considered particularly important insupporting the specific uptake of thymine into DNA, since only chromosomes andperipheral nucleolar chromatin contain this molecule. The nuclear grain distributionin relation to the chromosomes is shown diagramatically in Fig. 7 A, and was criticallyinvestigated by measuring the distance from the centre of each grain to the nearestchromosome boundary (Fig. 8A). The results obtained are consistent with thechromosomes being the major source of radioactivity in the nucleus, since the spreadof silver grains from the chromosome boundary approximates to the experimentalgrain distribution obtained by Caro (1962) and the derived data of Salpeter, Bachman& Salpeter (1969). If all the silver grains having centres within 0-3 /im of the boundaryare considered to be derived from the major label source (Sigee & Bell, 1971), thenover 95 % of the nuclear radioactivity can be considered chromosomal.

6 -

5 -

S 4 -

d 3 "2 2 -

1

10 20 30 40 50 60 70 80 90 100 120 140 160 180 200110 130 150 170 190

Nuclear grain count

300

Fig. 4. Light-microscope nuclear grain counts. The nuclear grain counts are shown for40 cells, selected for the possession of a clear nucleus in median section. The resultsare taken from a single autoradiographic batch, with all the slides having the samephotographic processing conditions.

The localization of radioactivity is shown by comparison with random distributions(Figs. 7B, 8B) obtained by plotting points from random number tables on to a tracingof the nucleus superimposed over a lined grid. Less than 70 % of the random pointsoccurred within 0-3 //.m of the chromosome boundary.

In all of the autoradiographs examined the silver grains appear to be dispersed overthe whole chromosome population, and are not restricted to a few profiles in section(Figs. 5, 6). This demonstrates that chromosomal DNA replication is of generaloccurrence, and that the continuous uptake of thymine cannot be accounted for interms of asynchrony of replication throughout the chromosome complement.

DISCUSSIONThymine, rather than thymidine, was used in this experiment, since a previous

attempt to label cells with the nucleoside only achieved a low level of incorporation(Sigee, unpublished observations), and there is some evidence that thymidine is nottaken up by other dinoflagellates (Allen et al. 1975). Radioactive thymine has been

86 S. A, Filfilan and D. C. Sigee

DNA replication in Prorocentrum

Fig. 7. Localization of nuclear silver grains. Tracings taken from a single nucleus toshow oudines of nuclear membrane and chromosomes, with distributions of auto-radiographic silver grains (A) and random points (B).

Fig. 5. Electron-microscope autoradiograph showing details of nucleus. The ultrathinsection has been resectioned from a cell with dense chromosomes (light microscopy).Silver grains occur largely over or close to the chromosomes (c) with very few over thenucleoli. The surrounding cytoplasm includes chloroplasts (p), vacuoles (i>) andmitochondria (m). x 10000.Fig. 6. Electron-microscope autoradiograph showing part of nucleus and surroundingcytoplasm. The ultrathin section has been resectioned from a cell with pale chrom-osomes (light microscopy). The chromosomes are heavily labelled throughout, x 9600.

S. A. Filfilan and D. C. Sigee

r ' ' ' 10-6Interna

0-6

IL % 10-4 0-2

grains

0-4 0-2

- 40

- 30

/7- 10

/ /

- 30

- 20

- 10

grai

nsilv

er

r-r-H 7 7 1\ •'—I 1

0-2. 0-4

0-2 0-4

n0-6

0-6

A

I I 1 i 1 ' T 10-8 1-0

External grains

B

I 1

0-8 1 0

Distance from boundary of nearest chromosome, /im

Fig. 8. Localization of nuclear silver grainsA, autoradiograph. The distribution of nuclear silver grains in relation to the

boundaries of chromosomes. The results are taken from 3 separate autoradiographsof the same nucleus.

B, random distribution. The distributions of random points have been plotted innumbers equal to the silver grains over a tracing of the 3 nuclei.

used as a specific precursor in several previous instances (Werner, 1971; Numan &Wilson, 1975). The specificity of labelling in Prorocentrum is shown both by the local-ization to the chromosomes, which are known to contain DNA but not RNA or basicproteins (Dodge, 1964a, b), and also by the specific action of DNase in removing thelabel from the cells.

Within the population of Prorocentrum cells used in this study, all the cells shownuclear uptake of label irrespective of the stage of cell cycle. This result has beenconfirmed by repeat experiments, and must be interpreted as continuous DNAsynthesis throughout a large part of the cell cycle. This synthesis is variable, however,reaching a high level during interphase.

The continuous DNA synthesis found in Prorocentrum has not been found in anyother dinoflagellate so far investigated. Studies on the uptake of thymidine in anendozoic dinoflagellate of Anthopleura (Franker, 1971) and 32P incorporation intoalkali-insoluble material of Cryptothecodinium cohnii (Franker, Sakhrani, Pritchard &Lamden, 1974) both reveal a discrete phase of DNA synthesis. A similar conclusionwas reached for Cachonina niei (Loeblich, 1976) on the basis of fluorimetric analysis,though it must be noted that the precise timing of 5-phase differs for each of these3 species.

DNA replication in Prorocentrum 89

Prorocentrum resembles these dinoflagellates, and other eukaryotes, in having thehighest levels of nuclear DNA synthesis during interphase, but is very unusual in thesynthesis being continuous over most, if not all, of the cell cycle. The implications ofcontinuous DNA synthesis in Prorocentrum are a matter for speculation, but may relateto 2 unusual and quite separate characteristics of dinoflagellate chromosomes -polyteny and prokaryote affinities.

Several authors have proposed that chromosomes in dinoflagellates are polytenic(Grasse & Dragesco, 1927; Haapala & Soyer, 1973). In other polytene cells, such asthose of dipteran salivary glands or pulvilli, the synthesis of nuclear DNA occurs asa discrete but prolonged S-phase (Pavan, 1963; Roberts, Whitten & Gilbert, 1974).This comparison goes some way to account for the observations on Prorocentrum, butwas not considered likely in view of the continuation of replication during cell division.It would also not explain why other dinoflagellates do not have a similar DNAreplication pattern.

The prokaryote affinities of dinoflagellate chromosomes are now widely appreciated,and have been summarized by Soyer & Haapala (1974). The replication of nuclearDNA in Prorocentrum clearly shows some similarity to prokaryotes, where replicationcan occur continuously throughout the cell cycle, depending on the generation timeand the duration of replication (Holland, 1970). This similarity extends also to mito-chondria and plastids, where autoradiographic studies have generally revealed con-tinuous but variable DNA synthesis (Sigee, 1972).

It seems most likely that the pattern of DNA synthesis in Prorocentrum relates to itsprimitive prokaryote characteristics rather than to any secondarily derived feature,such as chromosome polyteny. This implies a uniquely primitive position for thegenus within the dinoflagellates, and is supported by the conclusions reached byLoeblich (1976), who described Prorocentrum as a 'living fossil' on the basis of itssurface morphology.

REFERENCES

ALLEN, J. R., TUTTLE, R. C, HEDBERG, M. F., KLOTZ, L. C. & LOEBLICH, A. R. (1975).Characterisation of repeated-sequence DNA in Cryptothecodinium. J. Phycol. 11, Suppl. 15.

CARO, L. (1962). High resolution autoradiography. II. The problem of resolution. J.CellBiol.15, 180-199-

DODGE, J. D. (1964a). Chromosome structure in the Dinophyceae II. Cytochemical studies.Arch, mikrobiol. 48, 66-80.

DODGE, J. D. (19646). Cytochemical staining of sections from plastic-embedded flagellates.Stain Technol. 39, 381-386.

DODGE, J. D. (1965). Chromosome structure in the dinoflagellate and the problem of themesokaryotic cell. Exerpta med. int. Congr. Ser. 91, 339.

DODGE, J.D. (1966). The Dinophyceae. inThe Chromosomes of the Algae (ed.M.B.E. Godward),pp. 96-115. New York: St Martin's Press.

FRANKER, C. K. (1971). Division synchrony in primary cultures of an endozoic dinoflagellate.J. Phycol. 7, 165-169.

FRANKER, C. K., SAKHRANI, L. M., PRICHARD, C. D. & LAMDEN, C. A. (1974). DNA synthesisin Cryptothecodinium cohnii. J. Phycol. 10, 91-94.

GRASSE, P. P. & DRAGESCO, J. (1957). L'ultrastructure du chromosome des p6ridiniens et sesconsequences g6n6tiques. C. r. hebd. Stanc. Acad. Set., Paris 245, 2447-2452.

90 S. A. Filfilan and D. C. Sigee

HAAPALA, O. K. & SOYER, M. O. (1973). Structure of dinoflagellate chromosomes. Nature, NewBiol. 344, 195-197-

HOLLAND, I. B. (1970). DNA replication in bacteria. Sci. Prog., Lond. 58, 71-98.LOEBLICH, A. R. (1976). Dinoflagellate evolution: speculation and evidence. J. Protozool. 23,

13-28.NUMAN, H. N. & WILSON, C. M. (1975). Thymidine and thymine uptake in human dental

plaque. Caries Res. 9, 705-717.PAVAN, C. (1963). Synthesis. In Genetics Today, Proc. XI int. Cong. Genetics, The Hague,

vol. 2 (ed. S. J. Geerts), pp. 335-342. Oxford: Pergamon.ROBERTS, B., WHITTEN, J. M. & GILBERT, L. I. (1974). DNA synthesis patterns in the giant

footpad nuclei of Sarcophaga bullata. Chromosome 47, 193-201.SALPETER, M. M., BACHMAN, L. & SALPETER, E. E. (1969). Resolution in electron microscope

radioautography. J. Cell Biol. 41, 1-20.SIGEE, D. C. (1972). Pattern of cytoplasmic DNA synthesis in somatic cells of Pteridium

aquilimtm. Expl Cell Res. 73, 481-486.SIGEE, D. C. (1976). A resin-slide technique to select fixed embedded cells for transmission

electron microscopy. J. Microscopy 108, 325-329.SIGEE, D. C. & BELL, P. R. (1971). The cytoplasmic incorporation of tritiated thymidine during

oogenesis in Pteridium aquilinum. J. Cell Sci. 8, 467-487.SOYER, M. O. & HAAPALA, O. K. (1974). Division and function of dinoflagellate chromosomes.

jf. Microscopie 19, 137-146.SPURR, A. R. (1969). A low viscosity resin embedding medium for electron microscopy.

J. Ultrastruct. Res. 26, 31-43.WERNER, R. (1971). Mechanism of DNA replication. Nature, Lond. 330, 570-572.

(Received 12 February 1977)