Characterization of nuclear DNA in 12 species of Chlorella (chlorococcales, chlorophyta) by DNA reassociation

BioSystems, 24 (1990) 1 4 5 - 1 5 5

145

Elsevier Scientific Publishers Ireland Ltd.

Characterization of nuclear DNA in 12 species of Chlorella (Chlorococcales, Chlorophyta) by DNA reassociation Richard DSrr and Volker A.R. Huss Institut f~r Botanik und Pharmazeutische Biologic der Universit~ Staudtstrasse 5, D-85~0 Erlangen {F.R.G.) {Received January 16th, 1989) {Revision received May l l t h , 1990}

Strains of 12 different species of the genus Chlorella were analyzed for amount, reiteration frequency and kinetic complexity of chromosomal DNA components by Cot analysis. The resulting Cot curves reveal at least two different DNA components consisting of single copy DNA (up to 95%} and of repetitive DNA with complexities of 4.1 x 108 base pairs (bp) to approximately 11.7 x 10s bp and a reiteration frequency of 100-760. The total amount of repetitive DNA is less than 90/0 of the nuclear genome and similar in all strains studied. In contrast, the total kinetic complexity varies in a wide range from 1.26 x 10~ bp to 8.08 x 107 bp which is mainly due to differences in the size of single copy DNA. The genome sizes in Chlorella seem not to be correlated with biochemical and physiological characteristics and therefore are unlikely to be useful as a taxonomical marker. A comparison of thermal denaturation profiles showed that the melting points of repetitive and single copy DNA differ by approximately 7°C which may result from base mismatch and/or from a distinct base composition of the repetitive DNA.

Keywords: Chlorella; Cot analysis; Kinetic complexity; Repetitive DNA; Single copy DNA; Thermal denaturation.

Introduction

The chlorococcal, asexual green alga Chlorella is widely used as a preferred model organism not only for primitive eukaryotes but also for higher plant cells. The usefulness of this model system is complicated by the diversity of biochemical and physiological properties of the various species which are difficult to distinguish by morphological criteria (Kessler, 1982). DNA homology studies on a molecular level clearly showed substantial heterogeneity even within some species (Huss et al., 1986, 1987). Strains of C. sorokiniana, for instance, can be divided into three DNA-homology groups and only 3 out of 17 strains are closely related to the type strain. Chlorella is commonly regarded as an haploid organism, although condensed chromoCorrespondence to: V.A.R, Huss.

somes could not be observed (Kobayashi et al., 1987). Except for the nuclear DNA content of C. pyrenoidosa 211-8b (= C. fusca var. vacuolata) which is 4.8 x 10 -14 g (Bayen and Dalmon, 1975) very little is known about the structure, arrangement and complexity of nuclear DNA in Chlorella and other green algae (Hirano et al., 1986). The research has been focused on the nature and organization of DNA in the chromosomes of animals (Britten and Davidson, 1971) and higher plants (Bendich and McCarthy, 1970; Rimpau et al., 1978). Within the family Gramineae the dispersion of repeated DNA sequences in highly ordered patterns and the genome organization are well characterized (Deshpande and Ranjekar, 1980). Genomes of all eukaryotes and of some prokaryotes (cf. Yates and Holmes, 1987) are composed of single copy (s.c.), repetitive (rep.) and foldback DNA as common elements of their sequence arrangement. In eukaryotic genomes the amount of

0303-2647/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland

146

the rep. DNA components varies over an extremely wide range from 5o/o in some fungi up to approximately 75% in Triticum (Mitra and Bhatia, 1986). In DNA reassociation experiments rep. DNA components react more rapidly than does s.c. DNA. Depending on the portion of rep. DNA this could influence DNA homology values which usually are calculated from initial renaturation rates of DNA mixtures of two different organisms in order to determine their genealogical relationship (cf. Huss et al., 1986, 1989). Therefore, the previously determined DNA homologies within the genus Chlorella may reflect mainly the relationship between rep. DNA components instead of s.c. sequences representing the less conservative part of the genome (Belford and Thompson, 1981). In this case the homology values should be too high and the heterogeneity within Chlorella even greater than previously assumed. The knowledge of the amount of nuclear rep. sequences in Chlorella is a first step towards a characterization of the genome structure in this genus. The experiments described below were carried out to determine the kinetic complexity of total nuclear DNA and the composition, reiteration frequency and size of different DNA components within various species of Chlorella. The analysis of DNA reassociation data allows the comparison with data of the genome organization of other algae, fungi and higher plants. Moreover, defined separation and further investigation of DNA sequences which are represented only once or many times in nuclear DNA will be practicable subsequent to C0t-analysis. The results of this investigation may contribute to a more profound comprehension of the phylogenetic relationships within the genus Chlorella. Materials and methods Strains and culture conditions All 13 Chlorella strains investigated in this

study (Table 1) were grown in the media of Kessler and Czygan (1970) or Kessler and Zweier (1971) with additional trace elements under the conditions described by Huss et al. (1986). Extraction and purification of nuclear D N A After harvesting and washing with T r i s EDTA buffer (0.05 M Tris, 0.03 M EDTA, pH 7.0) the algae were homogenized with glass beads (Kerfln and Kessler, 1978). DNA purification was carried out essentially according to a modified procedure after Marmur as described by Huss et al. (1986). Organelle DNA was separated from nuclear DNA by isopycnic centrifugation in CsC1. The DNA solution was adjusted to a density of 1.619 g/ c m 3, 50 /~g bisbenzimide (Hoechst Dye No.

TABLE 1 List of Chlorella strains studied. Species

Strains

Origin"

C. minutissima Shih. & Krauss C. vulgaris Beijerinck C. protothecoides Kriiger C. saccharophila var. saccharophila [(Kriiger) Migula] Fott & Nov~kov~ C. zofingiensis DSnz C. homosphaera Skuja C. lobophora Andreyeva C. kessleri Fott & Nov~kov~ C. fusca var. vacuolata Shih. & Krauss C. luteoviridis Chodat C. sorokiniana Shih. & Krauss C. mirabilis Andreyeva C. saccharophila var. eUipsoidea (Gerneck) Fott & Nov~kov~t

C-1.1.9. 211-11b z 211-7a T

Bethesda GSttingen G6ttingen

211-9a z 211-14a T 211/8e T 750-Iz 211-11g T

G6ttingen G6ttingen Cambridge Andreyeva G6ttingen

211-8b T 211-2a T 211-8k T 748-I T

G6ttingen G6ttingen G6ttingen Andreyeva

211-1a z

G6ttingen

•Bethesda, culture collection at Bethesda, Maryland, USA; Giittingen, Algensammlung des Pflanzenphysiologischen Instituts der Universitit GSttingen, FRG; Cambridge, Culture Centre of Algae and Protozoa, Cambridge, UK; Andreyeva, V.M. Andreyeva, Leningrad, USSR. rType strain.

147

33258) per mg DNA were added and the solution filled into Quick Seal Tubes (Beckman). Centrifugation was carried out in a Beckman L8-60M ultracentrifuge in a VTi 50 vertical rotor at 42,000 rev./min {145.000 g) for 30 h at 20 °C. The DNA bands were visualized by UV irradiation (312 rim) and removed from the gradient with a syringe. Bisbenzimide was extracted from DNA solutions with CsCI saturated isopropanol. The DNA solution was concentrated in dialysis tubes over crystalline saccharose and subsequently dialyzed against 20 mM Tris--HC1 (pH 7.0) or 1.62 × SSC (0.243 M NaC1, 0.0243 M trisodium citrate, pH 7.0).

DNA fragment length DNA was fragmented three times in a French Pressure Cell (Aminco) at a pressure of 1.5 x 106 Pa. The molecular weight of the resulting DNA fragments was determined by agarose gel electrophoresis (Southern, 1979).

DNA reassociation DNA reassociation kinetics were monitored by hydroxylapatite (HAP) chromatography (Bernardi, 1965) as well as by determination of optical hypochromicity (Britten et al., 1974). For consistency of the reassociation conditions, each DNA solution was individually incubated at the optimal renaturation temperature Tot = (melting point T - 25°C) in 0.316 M [Na ÷] according to the guanine plus cytosine (G + C) content. The melting points were determined by thermal denaturation from the midpoint of hyperchromic shift or, in the case of HAP chromatography, from the equation of Frank-Kamenetskii (1971) T

= 176 - (2.6 - %G + C/100) × (36 - 7.04 log[Na÷])

The DNA concentration in all reaction mixtures was determined by the diphenylamine method according to Burton (1968).

Hydroxylapatite column chromatography DNA was labelled with deoxy-[l',2'-3H~ guanosine-5'-triphosphate (Amersham) by means of the nick translation procedure (Rigby et al., 1977) according to conditions suggested by the suppliers. Free nucleotides were separated from labelled DNA on an Elutip-d mini column (Schleicher and Schfill, Dassel, F.R.G.). Denaturation of radioactively labelled and sheared unlabelled DNA at a ratio of 1:1000 (w/w) was performed by addition of 0.07 vol. of 2 M Na0H and additional boiling in an oil bath twice for 10 min at 120°C with intermediate cooling on ice. After neutralization with 0.27 vol. of 2 M Tris--HCI (pH 7.0), DNA reassociation was performed at T r in 10% NES (2 M NaC1, 20 mM EDTA, 0.3% SDS) with a final concentration of 0.316 M [Na÷]. Samples were reassociated to different Cot values, diluted to 0.18 M [Na ÷] with ice-cold bidistilled H20 and rapidly frozen in ice/acetone for storing, or poured immediately onto a hydroxylapatite column at 60°C (Britten and Kohne, 1966). Single stranded DNA was eluted with 4--5 vols. of 0.12 M Na-phosphate buffer (pH 6.8; 0.3% SDS) while double stranded DNA was eluted with 4 vols. 0.48 M Na-phosphate buffer (pH 6.8; 0.3% SDS) at temperatures of 60°C. The percentage of reassociation (°AR) was calculated by determination of radioactivity, measured in a Packard Scintillation Counter (Tricarb 1500). The extent of DNA self-reassociation (foldback DNA) was determined in control experiments by reassociation of the labelled DNA alone. The experimental data were corrected for the amount of foldback DNA as follows %R

=

100 × (% binding to HAP - % foldback) (100 - % foldback)

Spectrophotometric monitoring of DNA reassociation The decrease in UV absorption (260 nm) of

148

reassociating, denatured DNA was followed in a Gilford 2600 or a Gilford Response II spectrophotometer with thermal programming software. Sheared DNA was dialyzed against 1.62 × SSC/15% dimethyl sulphoxide (DMSO) and adjusted to optical densities of 1.0 and 0.5. DMSO was applied to allow complete strand separation at temperatures up to 98°C (Escara and Hutton, 1980). After thermal denaturation and determination of the melting point and hyperchromicity, the temperature was lowered to the optimal renaturation temperature as quickly as possible. The decrease in absorption was surveyed every 2 0 - 6 0 s by the self-registering spectrophotometer considering the storage capacity of 10,000 data points. The reassociation data were corrected for collapse hypochromicity (Bendich and Anderson, 1977).

Cot analysis Amount, frequency and number of different nuclear DNA components were estimated principally according to Britten et al. (1974). Cot values were calculated as follows mg DNA/ml

Cot =

333.5

× s

where 333.5 corresponds to the average molecular weight of a nucleotide (sodium salt) and s is the renaturation time in seconds. Cot values less than 10-2 and greater than 50 were determined by the HAP procedure whereas the middle range was monitored optically. Kinetic complexities of total nuclear DNAs were estimated from the rate constants K of the major algal DNA component in comparison with that of Escherichia coli as a standard

Dot blot hybridization Dot blot hybridizations were carried out as described by Williams and Mason (1986) and Anderson and Young (1986) using rep. and s.c. DNA isolated by HAP-chromatography at a Cot of 2.4. Ribosomal RNA was separated by SDS-polyacrylamide gel electrophoresis (Slater, 1983). DNA was labelled with deoxy[l',2'-3H]-guanosine-5'-triphosphate (Amersham) by means of the multiprime method of Feinberg and Vogelstein (1983) or the nick translation procedure (Rigby et al., 1977). Ribosomal RNA was labelled with biotinhydrazide according to Reisfeld et al. {1987). Hybridized 3H-labelled nucleic acids were detected in a liquid scintillation counter or in the case of biotin hydrazide as described by Leary et al. {1983). Results

E. coli B was used as a standard for the calculation of kinetic complexities and as a control for identical reassociation conditions. Fragment length of both labelled and unlabelled DNA was 500 _+ 100 bp. For comparison of DNA reassociation monitored by the two different methods described above, the monovalent cation concentration was adjusted uniformly to 0.316 M. The optical reassociation data had to be corrected for 8% TABLE 2 Influence of 15% DMSO on Cot~ (observed) values of nuclear DNA of C. luteoviridis 211-2a r

DNA component

Kinetic complexity (bp) =

4.2 × 108 bp ×

Second-order rate plots were calculated after Wetmur and Davidson (1968) and the amount of non-repeated DNA was evaluated as described by Sz~csi (1981).

Ks.c.Chlorella gE co,

Repetitive Single copy

Cot1~

Factor

0% DMSO

15% DMSO

5.5 x 10 -2 45.0

3.5 x 10 -2 27.1

1.57 1.66

149 0

•° ~

. o_o o.. • • • 0 • . . . . . . . . . . o \ 0 0

•

\

\ ••

XX

00

0

0

•

X 0

•

\\

•

0

• 0

\ 0 0o

\

•

\ 000\\

0 \

i

u

-2

-I

i 0

, '~

\ i 2

I~ Eof (tool. I-Is}

Fig. 1. Reassociation curves of s h e a r e d nuclear DNA of Chlorella sorokiniana (solid circles) and E. coli (open circles). The dashed line r e p r e s e n t s t h e single copy DNA fraction of C. sorokiniana. Reassociation was carried out with DNA concentrations of 25, 50 and 500/~g/ml in 0.316 M [Na ÷] at t h e optimal r e n a t u r a t i o n t e m p e r a t u r e . As much more data points w e r e obtained by t h e optical monitoring of DNA reassociation than can be shown in the figure, each circle r e p r e s e n t s t h e mean of 10 values. The Cot~ was found to be 27.3 mol 1-' s in C. sorokiniana and 1.92 mol 1-x s in E. coli.

to 11% collapse hypochromicity due to intrastrand duplexes formed during the cooling process to the optimal renaturation temperature. Self-reassociation was found to be approximately 2% and subtracted from the HAP data. To compensate for the fundamen-

tal differences of the optical and the HAP method (Britten et al., 1974) as well as for the influence of DMSO in our reaction mixtures, data obtained with HAP had to be divided by 3.2. DMSO increased the reassociation rate by approximately 60% over the complete range of Cot values as shown for C. luteoviridis (Table 2). This effect could be ascertained with C. luteoviridis because its low G + C content (45.2 moP/o; Huss et al., 1989) allows complete DNA strand dissociation in 1.62 × SSC without addition of DMSO. The analysis of reassociation kinetics is demonstrated in detail with C. sorokiniana. Figure 1 shows typical Cot curves for E. coli and nuclear DNA of C. sorokiniana. The Cotm of the algal DNA corresponds to 500/0 reassociation of the slowly reassociating DNA fraction which is assumed to represent the single copy or non-repeated DNA. The observed Cotl~ values of E. coli and C. sorokin/aria were determined to be 1.92 mol 1-1 s and 27.30 tool 1-1 s, respectively. Thus, the nuclear DNA of C. sorokiniana reassociates about 14.2 times slower than the DNA of E. coli indicating a genome size of 5.97 x 107 bp which corresponds to a nuclear DNA content of 5.4 x 10-14 g. The multicomponent DNA reassociation is clearly demonstrated by the initial part of the Cot curve up to a log Cot of 0. After Britten et al. {1974) the amount of different DNA components is characterized by

TABLE 3 Kinetic analysis of nuclear DNA from Chlorella sorokiniana 211-8k T DNA comp.

Fraction size (%)

Cot m o b s . '

K obs."

C0t~ pure b

Copy number c

Complexity (bp)d

rep. s.c. Total"

5.0 95.0 100.0

0.057 27.3 .

17.34 0.0367 .

2.88 × 10 -s 25.89

477 1

6.30 x 103 5.66 x 107 5.97 × 107

.

.

(0.054 pg) "Observed Cot value mol 1-1 s and reaction constant K (1 mol -~ s-9 at 50% reassociation. fraction size/100 × /fob," cNumber of copies = Ko~ (rep.)/Kob,. (s.c.). dKinetic complexity of one copy is e x p r e s s e d as base pairs relative to E. coli (C0tl~ p~, = 1.92; kin. compl. = 4.2 x 106 bp; f r a g m e n t length = 500 bp): Cotl~ p~ x 4.2 x 106/1.92. "Kinetic complexity of total nuclear DNA is e x p r e s s e d as base pairs relative to E. coli using C6t~ oh,. of the s.c. fraction: Cotl~,~ X 4.2 X 106/1.92, bC0tll2 pure =

150

the position of the respective plateau of the Cot curve. Table 3 illustrates the characteristics of the s.c. and the rep. DNA component which represents only 5% of the nuclear genome of C. sorokiniana. Ninety-five percent of the nuclear genome is composed of a single class of sequences with a pure Cotl~ of 25.89 mol 1-1 s regarding the relative concentration of s.c. DNA in the reaction mixture. The copy number of the recognizable rep. DNA component is approximately 477 compared to the slowly reassociating s.c. DNA. The kinetic complexity of the rep. sequences was determined to be 6.3 × l0 s bp. Considering the repetition frequency, a total complexity of about 3.0 × 106 bp was calculated. The second-order rate plotting confirmed the multicomponent DNA reassociation (Fig. 2) typical of genomes of eukaryotes. In contrast, DNA of E. coli revealed a single reaction rate demonstrating that no significant fraction of repeated DNA is present. By the procedure of Sz6csi (1981)

/

/o /o

115 .¢ r

~

/

110

i

/

/

/ 0

0

©

©

105

30

60

90

time {rain)

Fig. 2. Second-order rate plot for the DNA reassociation of Chlorella sorokiniana (solid circles) and E. coli (open circles). According to the method of Sz6csi (1981) the amount of repetitive DNA was determined to be 5.5%. The dashed line represents the extrapolation of the slowly reassociating DNA component. A~ = absorbance of denatured DNA corrected for 10% collapse hypochromicity; A = absorbance of native DNA; A t = absorbance of renaturing DNA at time t.

TABLE 4 Characterization of nuclear DNA components of different Chlorella species I. Species

Genome size (bp × 107)

Fraction/total amount (%)~/(bp x 108) of rep. sequences

Compl. (bp x 103)/copy number of the rep. component ~

C. minutissima C. vulgaris C. protothecoides

1.26 1.40 1.95

7.0/0.88 8.3/1.16 6.4/1.25

9.0/100 11.7/100 6.2/200

C. C. C. C. C.

sacch, var. sacch. zofingiensis homosphaera lobophora kessleri

3.94 4.13 4.18 4.26 4.44

7.2/2.84 5.0/2.07 7.3/3.05 5.5/2.34 4.8/2.13

8.3/340 7.5/280 6.0/510 11.21210 5.2/410

C. fusca var. vacuoL C. luteoviridis C. sorokiniana C. mirabilis C. sacch, var. ellips.

5.21 5.93 5.97 6.02 8.08

6.3/3.28 5.3/3.14 5.0/2.99 4.8/2.89 6.0/4.84

4.8/680 4.1/760 6.3/480 4.1/700 11.5/420

aAverage value of three to four experiments; the relative standard deviation for the genome size and the fraction of rep, sequences is approximately 10%. bDetermined by C0t-analysis (see Materials and methods). ~The complexity and copy number are calculated as described in Table 3.

151

1 22E

1150

1 075

60

J 65

, 70

, 75

, 80

,

B5

f~emperahJre (Q[)

Fig. 3. Thermal denaturation profile (0.1 x SSC) of reassociated repetitive (a), reassociated single copy (b), and sheared native DNA (c) of Chlorella sorokiniana. Repetitive and single copy DNA fractions were separated at a Cot of 2.4. The increase in temperature was 0.5°C/ min. The melting points (midpoints of hyperchromicshift) are 74.5°C (a), 81.0°C (b) and 81.5°C (c).

the non-repeated fraction of C. sorokiniana was estimated to be 94.5%. The total repeated fraction consequently would be 5.5% of the nuclear genome which is in good accordance with the value determined by Cot graph analysis. The genome sizes and fractions, total amounts, complexities and reiteration frequencies of the rep. DNA components of all Chlorella species examined are listed in Table 4. The nuclear genome sizes vary in a wide range from 1.26 × 107 bp to 8.08 × 107 bp corresponding to nuclear DNA contents of 1.1 X 10 -14 g to 7.3 × 10 -14 g and thus are 3--19 times more complex than DNA of E. coli. We could ascertain that most of the strains studied fall in one of th r ee clusters of about 1.5 × 107 bp, 4.2 × 107 bp and 6.0 × 107 bp. The total amount of rep. sequences in Chlorella varies only within the range of 4.8-8.3O/o and can be divided into four groups r e pr es e nt i ng a multiple of approximately 1 × 10e bp (Table 4). The highest content in bp of rep. DNA was found in C. saccharophila var. ellipsoidea which also has the largest nuclear genome

(8.08 × 107 bp). The Cot analysis reveals the presence of at least two different DNA components consisting of s.c. DNA and of rep. DNA with complexities of 4.1--11.7 kbp and reiteration frequencies of approximately 100 - 760. Dot blot hybridizations of nylon filterbound rep. and s.c. DNA with biotin-labelled ribosomal RNA of C. sorokiniana as well as reciprocal tests with filter-bound rRNA and 3H- or biotin-labelled rep. and s.c. DNA were carried out to examine the e x t e n t of reassociation of rep. and s.c. DNA with rRNA. The results (not shown) indicate that rep. DNA in C. sorokiniana contains up to 500/0 ribosomal RNA genes. Control experiments with 5S rRNA of E. coil and rep. DNA of C. sorokin/ana yielded more than 50% reassociation compared to the homologous system with both rRNA and rep. DNA from C. sorokin/ana. The melting behaviour of reassociated rep. and s.c. DNA as well as of sheared native DNA is illustrated in Fig. 3. Single stranded and reassociated duplexes were separated by HAP chromatography at a Cot of 2.4 which was assumed to allow complete reassociation of rep. elements. The thermal stability of reassociated s.c. and rep. DNA was compared with that of sheared native nuclear DNA. Reassociated s.c. and native DNA dissociate within a range of 1 2 - 1 7 ° C with melting points of 81.0°C and 81.5°C and hyperchromicities of 25% and 27%, respectively. The thermal denaturation profile of rep. DNA demonstrates a reduced thermal stability compared to native DNA. The maximum increase in optical density is between 59°C and 82 °C with a melting point of 74.0 °C and a hyperchromicity of about 14%. Discussion

Because of the complexity of eukaryotic genomes a considerable renaturation time is generally required for an almost complete reassociation of dissociated DNA strands with reasonable DNA concentrations. Reassociation

152 experiments taking more than 110 h at temperatures of about 65°C were not performed to avoid a significant influence of strand scissions and depurination of DNA (Britten et al., 1974). To cover nevertheless a large range of Cot values, DNA reassociation was followed by two different methods. The advantage of the optical monitoring of DNA reassociation is that only real base pairing is recorded as a decrease of absorption at 260 nm and that the results are well reproducible. It is limited, however, to DNA concentrations of about 25 ~g/ml to 50 ~g/ml. On the other hand, HAP chromatography is time consuming and may give variable results but it allows the determination of a broad range of Cot values. Therefore, a combination of the two methods is convenient in detecting different DNA components in the nuclear genome. Bayen and Dalmon (1975) found the nuclear DNA content of C. pyrenoidosa 211-8b (= C. fusca var. vacuolata) to be 4.8 × 10-14 g which is in excellent agreement with our data (4.7 x 10-14 g). The kinetic complexities of the nuclear genomes in the Chlorella species studied vary over a wide range by a factor of about 6.5. The nuclear genome size of C. sorokiniana and C. vulgaris differ by a factor of about 4. This is surprising, because it has been suggested that these species may be closely related (Kessler, 1982). In some closely related species of higher plants, differences in genome sizes are generally correlated with the content of rep. DNA (Hutchinson et al., 1980; Walbot and Cullis, 1985). A simultaneous change in the content of rep. and s.c. sequences was observed in higher plants and animals (Ginatulina et al., 1982). Considering the rather low amounts of 4.8% to 8.3% rep. DNA sequences in all Chlorella strains studied, it has to be emphasized that the variations of the kinetic complexities in the genus Chlorella are mainly due to variations of the content of s.c. DNA. With regard to the complexity of s.c. DNA, a capacity of about 8500 and 56,000 genes of average size can be evaluated for C. minutissima and C.

saccharophila

var. eUipsoidea, respectively. As only 5 - 1 0 % of the s.c. DNA information is transcribed (Britten, 1986), C. minutissima may contain less than 1000 and C. saccharophila var. ellipsoidea more than 3000 structural genes. The significance of such a different protein equipment of these phenotypically rather similar organisms is unclear. For Nanochlorum eucaryotum, however, a very small green alga with "minimal eukaryotic features" and closely related to C. minutissima according to 18S rRNA analysis (Huss and Sogin, unpublished data), it has been reported that for instance tubulin-like proteins, histones and nucleosome-like features are apparently absent (Zahn, 1984). The 18S rRNA analysis of N. eucaryotum also clearly showed that this is not a primitive character but instead apparently due to evolutionary reduction (Sargent et al., 1988). All but two species (C. fusca var. vacuolata and C. sacch, var. eUipsoidea) can be assigned to three groups of different genome sizes (Table 4). A similar situation occurs in cyanobacteria. The genome size of 128 defined strains varies by a factor of 5.4 and can be divided into distinct groups. As an explanation, Herdman et al. (1979) assume a line of duplications of the smallest original genome during evolution. After Hutchinson et al. (1980), however, a change in kinetic complexity cannot satisfactorily be explained alone by amplification of s.c. DNA because duplicated sequences would not increase the kinetic complexity. An increase in genome size of Chlorella species by duplication of only rep. DNA and subsequent mutations is unlikely because of the low amount of rep. DNA. Based on a comparison of kinetic complexities with species specific characters of ChloreUa strains and considering a possible duplication of DNA during evolution we conclude that the genome size is not a useful distinctive mark for separation or identification of different Chlorella species. In green algae the amount of rep. DNA sequences varies from insignificant quantities in the case of Chlamydomonas reinhardtii

153 (Howell and Walker, 1976) up to 75% in for example Dictyosphaeria cavernosa (Olsen et al., 1987). Repeated DNA in almost all eukaryotic organisms is organized in highly ordered patterns and may provide an important function. Hirano et al. (1986) showed that the HaeIII family of rep. sequences in Chlorella eUipsoidea is organized in tandem arrays. The very low amount of rep. DNA sequences in Chlorella suggests that the rep. DNA could be comprised considerably of ribosomal genes as shown for Neurospora crassa (Krumlauf and Marzluf, 1979) and Aspergillus nidulans (Timberlake 1978). This was confirmed by dot blot hybridizations with labelled Chlorella and E. coli rRNA. The Cot analyses have shown that the bulk of nuclear DNA (up to 95%) in Chlorella reassociates slowly and represents the s.c. DNA. The rest of the nuclear genome is composed of an intermediate renaturing DNA component with variations in size and reiteration frequency between different species. Rep. DNA usually contains families of non-identical but similar rep. units. Ranjekar et al. (1974) showed that rep. sequences of Secale cereale are quite heterogeneous with a wide range of base sequence divergency and with a different degree of repetition. This heterogeneity results from base rearrangements, substitutions and mutations of DNA sequences during speciation (Narayan, 1988}. Further characterization of the rep. and s.c. DNA components in C. sorokiniana by thermal denaturation showed that there is a negligibly low difference in the melting points of sheared native and reassociated s.c. DNA but a difference of 7.5°C between sheared native and rep. DNA (Fig. 3). A reduction of 1 °C is generally explained as 1% base mispairing during reassociation reactions (Britten et al., 1978} which occurs prevalently during the reassociation of rep. DNA. Therefore, a base mismatch of 7.5% could be estimated for rep. DNA of C. sorokiniana. However, this interpretation of the reduction of the melting point neglects a possibly different G + C content of s.c. and rep. DNA fractions. The latter

apparently considerably consists of ribosomal RNA genes (rDNA). When excluding base mismatch during the reassociation reaction, the melting point of 74.5°C for the rep. DNA fraction implies a G + C content of 49 mol%. This is very close to the approximately 50 mol% G + C usually found for rDNA. Only an exact analysis of the base composition of rep. and s.c. DNA with methods like HPLCanalysis can clarify to what extent the reduction of the thermal stability is caused by base mismatch or by a different base composition of the rep. DNA. The results of this paper demonstrate that the different species of the genus Chlorella have small genomes with a low content of rep. DNA such in comparison with more highly evolved eukaryotes. The observed heterogeneity within this genus (Huss et al., 1989) was confirmed both by variations in kinetic complexity and in rep. DNA components. With respect to the influence of rep. sequences on the quantitative determination of DNA homologies for taxonomical purposes (cf. Huss et al., 1986) we assume that the amount of repeated sequences in ChloreUa is too low to exert a significant effect on DNA homologies.

Acknowledgements We thank Mrs. G. Huss for excellent technical assistance and Professor Dr. E. Kessler for the critical reading of the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft.

References Anderson, M.L.M.and Young, B.D., 1986, Quantitative filter hybridisation,in: Nucleic Acid Hybridisation - A Practical Approach, B.D. Hames and S.J. Higgins (eds.) (IRL Press, Oxford,U.K.),pp. 73-110. Bayen, M. and Dalmon, J., 1975, Determination physicochimique de la ploidie chez l'algue unicellulaire Chlorella pyrenoidosa (souche 211/8b). Biochim. Biophys. Acta 395, 213-219. Belford, H.S. and Thompson, W.F., 1981, Single copy DNA homologies in Atriplex. II. Hybrid thermal stabilities and molecular phylogeny.Heredity 46, 109122.

154 Bendich, A.J. and Anderson, R.S., 1977, Characterization of families of repeated DNA sequences from four vascular plants. Biochemistry 16, 4655-4663. Bendich, A.J. and McCarthy, B.J., 1970, DNA comparison among barley, oats, rye and wheat. Genetics 65, 5 4 5 565. Bernardi, G., 1965, Chromatography of nucleic acids on hydroxyapatite. Nature 206, 779-783. Britten, R.J., 1986, Rates of DNA sequence evolution differ between taxonomic groups. Science 231, 1393-1398. Britten, R.J. and Davidson, E.H., 1971, Repetitive and nonrepetitive sequences and a speculation on the origins of evolutionary novelty. Q. Rev. Biol. 46, 1 1 1 138. Britten, R.J. and Kohne, R.E., 1966, Nucleotide sequence repetition in DNA. Carnegie Inst. Washington Yearb. 65, 78-106. Britten, R.J., Graham, D.E. and Neufeld, B.R., 1974, Analysis of repeating DNA sequences by reassociation. Methods Enzymol. 29, 363--406. Britten, R.J., Cetta, A. and Davidson, E.H., 1978, The single copy DNA sequence polymorphism of the sea urchin Strongylocentrotus purpuratus. Cell 15, 11751186. Burton, K., 1968, Determination of DNA concentration with diphenylamine. Methods Enzymol. 12, 163--166. Deshpande, V.G. and Ranjekar, P.K., 1980, Repetitive DNA in three Graminaceae species with low DNA content. Hoppe-Seyler's Z. Physiol. Chem. 361, 12231233. Escara, J.F. and Hutton, J.R., 1980, Thermal stability and renaturation of DNA in dimethyl sulfoxide solutions: acceleration of the renaturation rate. Biopolymers 19, 1315-1327. Feinberg, A.P. and Vogelstein, B., 1983, A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6-13. Frank-Kamenetskii, M.D., 1971, Simplification of the empirical relationship between melting temperature of DNA, its GC content and concentration of sodium ions in solution. Biopolymers 10, 2623-2624. Ginatulina, L.K., Ginatulin, A.A., Lyapunova, E.A. and Vorontsov, N.N., 1982, Genome analysis of ground squirrels of the genus Citellus (Rodentia, Sciuridae). I. DNA reassociation kinetics and genome size of eight species. Genetica 59, 211-221. Herdman, M., Janvier, M., Rippka, R. and Stanier, R.Y., 1979, Genome size of cyanobacteria. J. Gen. Microbiol. 111, 73-85. Hirano, H., Takeda, Y. and Iwamura, T., 1986, Unique repetitive sequences of 170 bp in Chlorella. Plant Mol. Biol. 7, 311--317. Howell, S.H. and Walker, L.L., 1976, Informational complexity of the nuclear and chloroplast genomes of Chlamydomonas reinhardi. Biochim. Biophys. Acta 418, 249-- 256.

Huss, V.A.R., D6rr, R., Grossmann, U. and Kessler, E., 1986, Deoxyribonucleic acid reassociation in the taxonomy of the genus Chlorella. I. Chlorella sorokiniana. Arch. Microbiol. 145, 329--333. Huss, V.A.R., Schwarzw~lder, E. and Kessler, E., 1987, Deoxyribonucleic acid reassociation in the taxonomy of the genus Chlorella. II. Chlorella saccharophila. Arch. Microbiol. 147, 2 2 1 - 224. Huss, V.A.R., Huss, G. and Kessler, E., 1989, Deoxyribonucleic acid reassociation and interspecies relationships of the genus Chlorella (Chlorophyceae). Plant Syst. Evol. 168, 71--82. Hutchinson, J., Narayan, R.K.J. and Rees, H., 1980, Constraints upon the composition of supplementary DNA. Chromosoma 78, 137-145. Kerfin, W. and Kessler, E., 1978, Physiological and biochemical contributions to the taxonomy of the genus Chlorella XI. DNA hybridization. Arch. Microbiol. 116, 97-103. Kessler, E., 1982, Chemotaxonomy in the Chlorococcales, in: Progress in Phycological Research, Vol. 1, F.E. Round and D.J. Chapman (eds.) {Elsevier, Amsterdam), pp. 111 - 135. Kessler, E. and Czygan, F.C., 1970, Physiologische und biochemische Beitrage zur Taxonomie der Gattung Chlorella. IV. Verwertung organischer Stickstoffverbindungen. Arch. Mikrobiol. 70, 211--216. Kessler, E. and Zweier, I., 1971, Physiologische und biochemische Beitriige zur Taxonomie der Gattung Chlorella. IV. Verwertung organischer Stickstoffverbindungen. Arch. Mikrobiol. 70, 211--216. Kobayashi, M., Hatta, K. and Sugai, M., 1987, Endopolyploidy in Chlorella. Arch. Microbiol. 146, 309-314. Krumlauf, R. and Marzluf, G.A., 1979, Characterization of the sequence complexity and organization of the Neurospora crassa genome. Biochemistry 18, 3705--3713. Leary, J.J., Brigati, D.J. and Ward, D.C., 1983, Rapid and sensitive colorimetric method for visualizing biotinlabeled DNA probes hybridized to DNA or RNA immobilized on nitrocellulose: Bio-blots. Proc. Natl. Acad. Sci. USA 80, 4045-4049. Mitra, R. and Bhatia, R.C., 1986, Repeated DNA sequences and polyploidy in cereal crops, in: DNA Systematics, Vol. III: Plants, S.K. Dutta (ed.) (CRC Press, Boca Raton, FL) pp. 21--43. Narayan, R.K.J., 1988, Constraints upon the organization and evolution of chromosomes in AUium. Theor. Appl. Genet. 75, 319-329. Olsen, J.L., Stam, W.T., Bot, P.V.M. and van den Hoek, C., 1987, scDNA-DNA hybridization studies in Pacific and Carribean isolates of Dictyosphaeria cavernosa (Chlorophyta) indicate a long divergence. Helgoliinder Meeresunters. 41, 377-- 383. Ranjekar, P.K., Lafontaine, J.G. and Pallotta, D., 1974, Characterization of repetitive DNA in rye (Secale cereale). Chromosoma 48, 427--440. Reisfeld, A., Rothenberg, J.M., Bayer, E.A. and Wilchek, M., 1987, Nonradioactive hybridization probes pre-

155 pared by the reaction of biotin hydrazide with DNA. Biochem. Biophys. Res. Commun. 142, 519-526. Rigby, P.W.J., Dieckmann, M., Rhodes, C. and Berg, P., 1977, Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J . M o l . Biol. 113, 237--251. Rimpeau, J., Smith, D.B. and Flavell, R.B., 1978, Sequence organisation in barley and oats chromosomes revealed by interspecies DNA/DNA hybridization. Heredity 44, 131 - 149. Sargent, M., Zahn, R., Waiters, B., Gupta, R. and Kaine, B., 1988, Nucleotide sequence of the 18S rDNA from the microalga Nanochlorum euca~yotum. Nucl. Acids Res. 16, 4156. Slater, R.J., 1983, The extraction and fractionation of RNA, in: Techniques in Molecular Biology, J.M. Walker and W. Gaastra (eds.) (Groom Helm Ltd, Kent), pp. 113-- 134. Southern, E.M., 1979, Gel electrophoresis of restriction fragments. Methods Enzymol. 68, 152--177.

Sz~csi, A., 1981, Kinetic complexity and repetitiveness of Fusarium graminearum nuclear DNA. Exp. Mycol. 5, 323 - - 329. Timberlake, W.E., 1978, Low repetitive DNA content in AspergiUus nidulans. Science 202, 973-975. Walbot, V. and Cullis, C.A., 1985, Rapid genomic change in higher plants. Annu. Rev. Plant Physiol. 36, 367396. Wetmur, J.G. and Davidson, N., 1968, Kinetics of renaturation of DNA. J. Mol. Biol. 31,349-370. Williams, J.G. and Mason, P.J., 1986, Hybridisation in the analysis of RNA, in: Nucleic Acid Hybridisation - A Practical Approach, B.D. Haines and S.J. Higgins (eds.) (IRL Press, Oxford, U.K.), pp. 139-159. Yates, J.R. and Holmes, D.S., 1987, Two families of repeated DNA sequences in Thiobacillus ferrooxidans. J. Bacteriol. 169, 1861-1870. Zahn, R.K., 1984, A green alga with minimal eukaryotic features: Nanochlorum eucaryotum. Origins Life 13, 289 - 303.