The structure of kinetoplast DNA

Biochimica et Biophysica Acta, 390 (1975) 155--167 © Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands

BBA 98283 THE S T R U C T U R E OF KINETOPLAST DNA I. PROPERTIES OF THE INTACT MULTI-CIRCULAR COMPLEX F R O M C R I T H I D I A L UCI LI A E

C.M. KLEISEN, P. BORST and P.J. WEIJERS Section for Medical Enzymology, Laboratory of Biochemistry, University of Amsterdam, Eerste Constantijn Huygensstraat 20, Amsterdam (The Netherlands) ( Received November 12th, 1974)

Summary 1. We have developed a modified isolation procedure that yields kinetoplast DNA networks containing more than 90% closed circular DNA, as judged by two criteria: (a) In 0.15 M NaC1/0.015 M sodium citrate (pH 7.0), less than 10% of the intact kinetoplast DNA melts in the temperature region of sonicated kinetoplast DNA. In 7.2 M NaC104 the kinetoplast DNA melts with a Tm 26°C higher than sonicated kinetoplast DNA. Even after complete melting in 7.2 M NaC104 at 90°C, the network remains intact, as judged by regain of hypochromicity on cooling and analysis in CsC1 containing propidium diiodide. (b) In alkaline sucrose gradients more than 90% of the kinetoplast DNA sediments in a single peak. 2. In CsC1 gradients containing ethidium bromide or propidium diiodide intact kinetoplast DNA gives a single uni-modal band showing an extremely restricted dye uptake. From the position of the band relative to the bands of PM2 DNA, the superhelix density of these networks is calculated to be +3.9 twists per 1000 base pairs. The superhelix density of closed mini-circles, efficiently liberated from the networks by shear in a French press, is --0.5 twists per 1000 base pairs. We attribute the high superhelix density (the highest y e t observed in any DNA) of intact networks to their compact, highly catenated structure, leading to an additional constraint on dye uptake, superimposed on the restriction due to closed circularity.

Introduction

The kinetoplast DNA (K-DNA) of the protozoan order Kinetoplastida, which includes the genera Trypanosoma and Crithidia, is a compact mass of A b b r e v i a t i o n s : K-DNA, k i n e t o p l a s t DNA; SSC, 150 mM NaC1/15 mM s o d i u m citrate (pH 7.0).

156 extra-nuclear DNA with a molecular weight up to 4 • 1010 (see ref. 1). This DNA is localized in the single mitochondrion of these organisms and its loss entails the inability to synthesize functional mitochondria (see ref. 1). On this basis K-DNA is now considered to be the equivalent of mtDNA in other organisms, but direct proof for this contention is lacking. The fact that some trypanosomal drugs, like ethidium bromide or hydroxystilbamidine, appear to act by specifically blocking replication and transcription of K-DNA [2,3] adds to the interest of attempts to elucidate its structure and function. The very compact nature of the K-DNA network has recently allowed its isolation and handling w i t h o u t apparent shear-degradation [4--7]. In the lightmicroscope these networks look like cup-like round sheets of DNA [4]. In electron micrographs mini-circles form a prominent c o m p o n e n t of all isolated preparations, intact or partly degraded [4,8--17]. They are often extensively catenated and vary in size from 0.29 pm in Leishmania to 0.75 pm in Crithidia. Each K-DNA network is estimated to contain 10 00(t--25 000 of these circles. Renaturation studies [17] and fingerprint analysis suggest that in Leishmania all circles have the same base sequence, limiting their information content to less than one-tenth of that in animal mtDNA (cf. ref. 18). The question whether more complex DNA forms an integral part of the network becomes, therefore, of central importance. The experimental results on this point are contradictory, however. Linear DNA, much longer than mini-circle length, has been observed in partly degraded K-DNA preparations in several laboratories [4,8,10,13,14,19,20]. Renaturation experiments also give suggestive evidence for the presence of a minor fraction of more complex DNA [17,21]. In a very detailed study, however, Renger and Wolstenholme [12] did not find longer (linear) DNA as an integral part of the network. To resolve this question and obtain further information on the molecular structure and genetic complexity of K-DNA, we have studied isolated K-DNA from Crithidia luciliae in more detail. In this paper we present our experiments on the properties of intact K-DNA, specifically focussing on the question of whether all the DNA in the complex is interlinked by covalent bonds. In a later paper the results of controlled degradation of K-DNA by shear and restriction endonucleases will be presented. Materials and Methods Growth conditions and isolation of K-DNA. Cells were grown aerobically at 24°C in a 20-1 bottle with 6 1 Bon~ medium [22], containing per l: 2.34 g N a 2 H P O 4 . 2 H 2 0 , 4.0 g NaC1, 0.4 g KC1, 2.0 g glucose, 1.0 g oxoid liver infusion, 15.0 g tryptose, 20 mg hemine (pH 7.4). I ml Anti-foam per 1 medium was added. Cells were harvested by centrifugation in the stationary phase (As 46 nm = 0.80--1.00). The pellets were washed twice with 100 mM NaC1/10 mM EDTA/10 mM Tris, pH 8.0. After the washings the pellet was resuspended in 5 ml 100 mM NaC1/250 mM EDTA/10 mM Tris, pH 8.0 per g wet weight of cells*. * The use of such communication).

high

EDTA

concentrations

w a s s u g g e s t e d to u s b y D r L. S i m p s o n

(personal

157

Cells were lysed by adding an equal volume of a detergent mix, containing 4% sodium dodecylsulphate, 2% triisopropylnaphthalene sulphonate, 8% 4-aminosalicylate and 6% butanol-2, made up in 100 mM NaC1/250 mM E D T A / 1 0 mM Tris, pH 8.0. After a b o u t 1 min in ice, the lysate was deproteinised with 1.5 volume of freshly distilled phenol for 10 min at 0°C in the dark. This was repeated three times, the second and third time with 1 volume of phenol. The water phase was dialysed overnight against 3 1 of 100 mM NaC1/100 mM E D T A / 1 0 mM Tris, pH 8.0 with one change and then incubated with 100 pg pancreatic ribonuclease A per ml (pretreated for 15 min at 80°C) for 1 h at 37 ° C. Then sodium dodecylsulphate was added to a final concentration of 0.1% and pronase (preincubated for 2 h at 37°C and 15 min at 80°C) to a final concentration of 2 mg per ml. Incubation for another 1 h at 37°C followed. The lysate was then deproteinised two or three times, with an equal volume of freshly distilled phenol at 0°C in the dark. The water phase was layered directly on top of an ice-cold 20% sucrose cushion made up in 100 mM NaC]/100 mM EDTA/10 mM Tris, pH 8.0, in a SW-27 nitrocellulose tube and spun during 1 h at 19 000 rev./min at 4°C in a Spinco 65-B ultracentrifuge. The (invisible) pellet of K-DNA wa~ taken up in 100 mM NaCI/100 mM EDTA/10 mM Tris, pH 8.0, and dialysed overnight against 100 mM NaC1/100 mM E D T A / 1 0 mM Tris, pH 8.0, with one change. The K-DNA was then spun to equilibrium in a CsC1/ethidium bromide density gradient in a 64-angle rotor at 45 000 rev./min for at least 48 h. The K-DNA was recovered from the gradient by siphoning o u t the band, made visible by ultraviolet illumination. Preparation o f labelled DNA. Labelling of the cells with 3 z p was done by growing them in 5 1 Bon~ medium containing Tris • HC1 instead of Na2 HPO4 and 10 mCi 32 p. The phosphate in the tryptose and liver infusion present in this medium was removed b y addition of 10 g MgSO4 per 1 of a 10% solution of either tryptose or liver infusion, and slowly adjusting the pH to 8.1 with NH4 OH under vigorous stirring. To allow all the phosphate to precipitate, the solution was kept in the cold r o o m for at least 3 h. Then the supernatant was decanted and spun for 10 min at 10 000 rev./min in the Sorvall GSA rotor at 4°C. The supernatant was used for the medium that was inoculated with a dephosphorylated pre-culture. Growth in this medium is slower and the final cell density reached is lower than in normal Bond medium. The isolation procedure of labelled K-DNA was the same as for unlabelled DNA. Melting of the DNA. The melting experiments were all done in a Gilford 2400 spectrophotometer, equipped with a Haake thermostat. The stoppered cuvettes contained a b o u t 0.7 ml of DNA solution, covered with liquid paraffin oil. Before the melting experiments the DNA was extensively dialysed against the buffer to be used. All melting curves were corrected for thermal expansion. For the correction of the melting experiments, done in 7.2 M NaC104, an expansion curve was determined. Isolation o f mini-circles. The 32 P-labelled DNA was extensively dialysed against 1 mM Tris/0.5 mM sodium citrate (pH 7.0). Before shearing in the French press N2 was bubbled through the solution and through the pressure cell for 10 min at 0°C. The DNA was sheared at 3500 lb/inch 2 at a concentration between 5 and 25 pg/ml. The sheared DNA was concentrated in Carbowax

158

and after dialysis spun to equilibrium in a CsC1/propidium diiodide gradient in a SW-50.1 rotor at 35 000 rev./min. After the run the gradient was dripped out by a hole in the b o t t o m of the tube and the Cerenkov radiation of the 32 p was counted. Peak fractions were collected and dialysed against 1 × 100 mM NaC1/100 mM EDTA/10 mM Tris, pH 8.0 or 5 mM Tris • HC1 (pH 7.2). Superhelix density calculations. For the determination of the superhelix density, the b u o y a n t density m e t h o d of Gray et al. [23] was used. All calculations were done using Eqns 8a and 8b from this reference. For calibration of the gradient we used covalently closed and open circular PM2 DNA, because this DNA has the same density in CsC1 as our K-DNA (Table I) and, therefore, correction factor f-i (Eqn 4 of ref. 23) was assumed to be unity. In the calculation we further assume that molecular weight effects in the superhelix density determination are negligible. For superhelix density determinations, 3 z P-labelled K-DNA and marker 3 H-labelled PM2 DNA were run in the same gradient of CsC1 containing propidium diiodide, in the Spinco SW-50.1 rotor at 35 000 rev./min. A photograph of the banding pattern was scanned (as in Fig. 1), and used to calculate the relative distances of the bands to the center of rotation. In one determination the radioactivity profiles of the DNAs after fractionating the tube were used for calculation. The superhelix density of the isolated mini-circles was also determined in this way (see Fig. 3). The superhelix density of PM2 DNA was assumed to be --5.5 twists per 1000 base pairs [23]. Photography and scanning. CsC1 gradients containing ethidium bromide or propidium diiodide were photographed on Scientia film from Agfa-Gevaert with a Nikon F camera, equipped with a Kodak Wratten filter No. 16. Fluorescence of the DNA bands was induced by illumination at 350 nm with a Universal ultraviolet lamp, type TL-900, from CAMAG. For superhelix density determinations, the photographs were scanned at 600 nm in the gel scan accessory of the Gilford 2400 spectrophotometer. Centrifugation. Preparative equilibrium gradients were run in a Spinco 65-B ultracentrifuge, analytical gradients in the SW-50.1 rotor. When dye was added a final concentration of 350 pg/ml was used for both ethidium bromide and propidium diiodide. Electron microscopy. The surface spreading technique, basically according to Kleinschmidt [24], was used to visualize the K-DNA. The spreading solution contained 1 M a m m o n i u m acetate, 1 mM EDTA (pH 7.5), 0.01% cytochrome c and K-DNA. The hypophase contained 250 mM a m m o n i u m acetate. The DNAprotein film was picked up on carbon-coated grids (200 mesh) and dried in ethanol and isopentane. The grids were rotary shadowed with Pt/Pd alloy (80/20) and the DNA was photographed in a Philips EM-300 electron microscope (60 kV), at a magnification between 5000 and 10 000. Magnification was calibrated with a carbon replica of a diffraction grating (2160 lines/mm, E.F. Fullam) and all length measurements were made relative to PM2 DNA component II (3.02 pm) present as an internal standard. Materials. All growth medium ingredients were obtained from Difco or British Drug Houses. Propidium diiodide A grade was purchased from CalBio-. chem, pancreatic deoxyribonuclease I, pancreatic ribonuclease A, pronase, alkaline protease and hemine equine type III from Sigma. Anti-foam was oh-

159 tained from D o w Coming. Sodium ortho[ 3 z p] phosphate and [Me -3 H] thymidine were purchased from The Radiochemical Center, Amersham. Results

Equilibrium centrifugation o f K-DNA in CsCl containing propidium diiodide In a closed circular duplex DNA the number of turns that one strand makes around the other is fixed. As a consequence, denaturation is impeded and complete strand separation is impossible. Absence of strand separation can be conveniently detected b y sedimentation through alkaline gradients; impeded denaturation by restricted dye uptake and a large increase in Tm, relative to nicked circles. We have verified these properties for intact K-DNA. In all experiments we have used the closed circular duplex DNA of phage PM2 as reference; this DNA has the same density in CsC1 and the same Tm as K-DNA from Crithidia and presumably, therefore, also the same base composition (Table I). When K-DNA is rapidly isolated in high EDTA concentrations to minimize enzymic nicking and centrifuged to equilibrium in CsC1 containing intercalating dyes, all DNA is found in a uni-modal band at an unusually high density. Fig. 1 gives an experiment with propidium diiodide; similar results were obtained with ethidium bromide, but with a smaller separation between closed and open circles. This high density of the K-DNA was reproducibly found for different preparations and several experiments show that this is due to its closed circular structure and n o t to the presence of contaminants or other structural abnormalities: (1) The density of the intact DNA in CsC1 without dye was 1.702 g/era 3 . This is exactly the value calculated from the Tm of the sonicated DNA, assuming that the linear relation between Tm and density found for other DNAs containing only the four standard bases holds for K-DNA (Table I).

Botlom

Top

K-DNA

meniscus B

/ i

i

i

i

i

i

i

F i g . 1. Preparative C s C l / p r o p i d i u m d i i o d i d e e q u i l i b r i u m g r a d i e n t o f i n t a c t K - D N A and internal P M 2 D N A m a r k e r . T h e s c a n n i n g w a s o b t a i n e d as d e s c r i b e d in M a t e r i a l s and M e t h o d s . T h e a r r o w s i n d i c a t e t h e c l o s e d and open circular forms of phage PM2 DNA.

160 TABLE I D E N S I T Y I N CsC1 A N D T m FOR, C. L U C I L I A E K - D N A Mol p e r c e n t a g e s (G + C) w e r e c a l c u l a t e d f r o m the d e n s i t y in CsCl a n d t h e T m in 1 X SSC, a c c o r d i n g to the r e f e r e n c e s in s q u a r e b r a c k e t s . DNA

Density in CsCl ( g / c m 3)

Intact K-DNA N i c k e d or sonicated K-DNA PM2 D N A

Tm (° C)

Mol p e r c e n t (G + C) c a l c u l a t e d f r o m : p

1.702

>100

1.702 1.702 [29]

Tm

42.9 [ 3 0 ]

86.8 86.4 [29]

42.9 [30] 42.9 [30]

-42.0 [31] 41.1 [ 3 1 ]

(2) Treatment of the K-DNA with pancreatic deoxyribonuclease I leads to a shift in density in CsC1/propidium diiodide that increases with increasing deoxyribonuclease concentration or reaction time (Fig. 2). This is the result expected for a complex of interlocked closed circles. (3) Drastic treatment of the K-DNA with ribonuclease (150 pg ribonuclease A + 75 units ribonuclease T1 per ml for 1 h at 37°C) or proteases (1 mg alkaline protease per ml for 1 h at 30°C or 2.5 mg pronase per ml for 1 h at 45 ° C) did n o t affect its density in CsC1/propidium diiodide (results not shown). On the other hand, 96% of 32 P-labelled K-DNA was rendered acid soluble by treatment with pancreatic deoxyribonuclease I. These results argue against RNA or protein contributing to the restricted dye uptake or the structure of the complex in general. In early experiments in which the isolation procedure of Laurent et al. [ 5] was used without modification, the K-DNA banded at a lower equilibrium density and always a second band was visible at the position of nicked circles. The same result was obtained in 0.5 M sodium EDTA (pH 8.0). We attribute this to nicking of the complex during isolation. The yield of the modified procedure is only about 40% of that obtained with the original procedure of Laurent et al. (Steinert, M., personal communication). We attribute this to the Top -r

,,n .500 a.

E

300

c

! 1;

'ib

i"s

20

2"~

~o

"~

~o

~'

io

60

Top

I:L

t 60(

20C

10

15

20

25

30

3s froction

/o

~5

60

Top 1000

600

200

,0

,5

20

25

30

35

6o

~0

Top 200C

160C

120C

800

400

45 50 55 60 30 35 40 Iroction Fig. 2. C s C 1 / p r o p i d i u m d i i o d i d e e q u i l i b r i u m g r a d i e n t s o f K - D N A n e t w o r k s a f t e r i n c u b a t i o n w i t h p a n c r e a t i c d e o x y r i b o n u c l e a s e I. I n c u b a t i o n s w e r e d o n e in 0.1 × SSC + 10 m M Tris + 5 m M MgC12 ( p H 7.4) at r o o m t e m p e r a t u r e . R e a c t i o n s w e r e s t o p p e d b y a d d i t i o n of 50 m M s o d i u m E D T A ( p H 7.4). a, c o n t r o l ; b, 30 rain, 0 . 0 5 ng d e o x y r i b o n u c l e a s e p e r m l ; e, 6 0 min, 0 . 0 5 ng d e o x y r i b o n u c l e a s e p e r m l ; d, 4 5 m i n , 0 . 0 6 ng d e o x y r i b o n u c l e a s e p e r m l ( d i f f e r e n t e x p e r i m e n t ) . X X , 32p-labelled K-DNA; o <) 3H-labelled PM2 D N A . 10

15

20

25

162 shorter incubation of the lysate in the detergent/pronase mixture, resulting in a greater loss of DNA in the interphase during the phenol extraction.

Superhelix density of intact K-DNA and isolated mini-circles Gray et al. [23] have shown that the distance between open and closed circular DNA under standard conditions in CsC1/propidium diiodide (or CsC1/ethidium bromide) decreases linearly with the increase of the number of negative super-twists per unit length of DNA, the superhelix density. We have used this m e t h o d to determine the overall superhelix density of the intact K-DNA. PM2 DNA was included as an internal reference to calibrate the superhelix density against band distance (Fig. 1). From four different K-DNA preparations we calculated a mean value for the superhelix density of +3.9 superhelical twists per 1000 base pairs. For comparison the superhelix density of isolated mini-circles was determined. Sonication of networks, the method used by Simpson and Da Silva [9] to isolate mini-circles from K-DNA of Leishmania tarentolae, gave only low yields (3% maximum) with Crithidia. Presumably the circles from Crithidia do n o t survive sonication, because they are nearly 3-fold larger than those from Leishmania. Shear degradation in a French press, however, yielded a mixture of DNA molecules a b o u t 20% of which showed restricted dye uptake (Fig. 3). The DNA in the higher density peak was shown by electron microscopy to contain nearly exclusively m o n o m e r circles (Fig. 4) with an occasional catenated oligomet. The average c o n t o u r length of these circles is 0.76 pm, in good agreement with the 0.75 pm reported by Renger and Wolstenholme [12]. The superhelix density of the mini-circles, calculated from Fig. 3, was --0.5 twists per 1000 base pairs. The very high superhelix density found for the intact K-DNA can, therefore, n o t be attributed to the mini-circles present in it.

Top lOOC

l i

i

h c

3

/

"~60C

I

200

It

!

10

20

30 z,O IrQclion

50

60

Fig. 3. CsCl/propidium diiodide density gradient of shea~-degraded K - D N A . T h e arrows indicate the position in the s a m e gradient of 3 H-labelled marker closed and open circular P M 2 D N A . Experiment performed by M r K. Fonck.

163

Fig. 4. E l e c t r o n m i c r o g r a p h o f D N A f r o m t h e m i n i - c h ' c l e p e a k ( b r a c k e t e d f r a c t i o n s ) f r o m Fig. 3.

Melting of K-DNA Fig. 5a shows the melting profiles of intact and sonicated K-DNA in 1 X SSC. Whereas sonicated DNA undergoes a sharp thermal transition with a Tm of 86.8°C, the intact complex is hardly denatured even at 100°C. To obtain complete denaturation the analysis was repeated in 7.2 M NaC104, a chaotropic solvent in which the Tm of linear DNA is 40°C below the Tm in 1 X SSC (Fig. 5b). In this solvent a maximum hyperchromicity of 43% was obtained for the intact complex. The broad thermal transition, spanning a b o u t 60 ° C, is strikingly similar to that reported for the closed circular duplex form of p o l y o m a DNA [25]. Also the ATm between the intact and sonicated form of K-DNA (about 26°C) is the same as the ATm found for p o l y o m a DNA (25°C). The bizarre melting curve reported by Steinert and Van Assel [21] for intact K-DNA was never observed by us. When the intact heat-denatured K-DNA was cooled d o w n to 50°C it regained its hypochromicity, whereas the sonicated DNA remained hyperchromic (Fig. 5c}. After the heating and cooling cycle the DNA was run to equilibrium in CsC1/propidium diiodide. The insert in Fig. 5c shows that the complex still forms a single hyper-sharp band at a density clearly below that of the closed PM2 DNA circles run in a parallel tube. This shows that the K-DNA complex is held together by bonds that will survive heating to 90°C in a chaotropic solvent.

164

E

~ 14C c

130

/

!

/

12C .Q (3

o

lO E c ot~q

b

~13q o

~'12C o

//r 30

50 70 90 temperature(=C)

11o c

~ 140

~ i

i

~0

12C

flC,

~tlc

~b

13C

;0 40 9o t;0

'

3'0 ~'o ;o

lb

temperaturePC)

~'o';" ~o

tempemture(~C)

10

Fig. 5. Melting p r o f i l e s of K - D N A in v a r i o u s salt c o n c e n t r a t i o n s , a, 1 X SSC; b, 0.1 X SSC, 7.2 M NaClO4 ( p H 7.0); c, 1 m M Tris, 1 m M E D T A , 7.2 M N a C 1 0 4 ( p H 7.0). X --×, sonicated K-DNA; © ~), i n t a c t K - D N A . T h e i n s e r t in p a n e l (c) s h o w s t h e f l u o r e s c e n t b a n d s in t w o i d e n t i c a l p r o p i d i u m diiodide equilibrium gradients containing either intact K-DNA after the heating-cooling cycle depicted by the open circles (left t u b e ) or c o n t r o l PM2 D N A ( r i g h t t u b e ) .

Band sedimentation o f K-DNA in neutral and alkaline gradients Fig. 6a shows intact K-DNA sedimenting through neutral sucrose. When the intact K-DNA is sedimented through alkaline sucrose the bulk of the DNA again sediments in a single peak about two times faster than in neutral gradients 180

9or

160

80

140

70

120

60

o

b

x

c

i i

50

100

40 ..~ "~ 6O o .o "o 2 4O

3o I

/";.°

2C

2°1/ ~Ol

";o

= 1'5

fraction

20

25

30

g

tb

~'s fraction

~0

2's

d0

165 9C ell •r

7C

~ E

60

£r

SO

~

40

.g

e

20 10 0

5

10

;'~

fraction

~b

~

t0

Fig. 6. S e d i m e n t a t i o n p r o f i l e s of i n t a c t K - D N A . a, n e u t r a l 5 - - 2 1 % l i n e a r s u c r o s e g r a d i e n t ; b, alkaline 5 - - 2 1 % l i n e a r s u c r o s e g r a d i e n t ; c, r e s e d i m e n t a t i o n of f r a c t i o n 28 f r o m g r a d i e n t in b o n an alkaline 5 - - 2 1 % l i n e a r s u c r o s e g r a d i e n t . N e u t r a l g r a d i e n t s c o n t a i n e d 1 M NaC1, 10 m M Tris, 10 m M E D T A ( p H 7.4). Alkaline g r a d i e n t s c o n t a i n e d 0.2 M N a O H , 0 . 8 M NaC1 a n d 10 m M E D T A . T h e g r a d i e n t s w e r e r u n in t h e SW-41 r o t o r at 4°C a n d a t 10 0 0 0 r e v . / m i n ; a a n d b for 13 m i n , c f o r 10 rain.

(Fig. 6b), showing that the complex is held together by alkali-stable bonds. A small and variable fraction of the DNA stays, however, at the top of the gradient. Electron microscopy of this fraction showed intact K-DNA complexes only and on resedimentation of this fraction more than half of it now co-sedimented with the bulk of the unfractionated DNA (Fig. 6c). This suggests that some complexes, although intact, remain at the top of the gradient. We have no explanation for this curious artefact. It was n o t observed with the added marker DNAs and it was present whether the gradient was dripped from the b o t t o m , siphoned o u t from the top or extruded by pumping heavy sucrose under the gradient. Discussion

Our results show by t w o methods, absence of strand separation in alkaline gradients and melting profiles, that virtually all DNA in the K-DNA network is present in the form of closed circles, interlocked by bonds that are stable to pronase, ribonuclease, alkali and heating to 90 ° C in 7.2 M NaC10+. The technical problem of avoiding all nicking of isolated closed circles, makes it difficult to judge whether the small amounts of DNA, behaving as open circles in some experiments, are real. In the alkaline gradients a maximum of 10% of the input DNA remained at the top of the gradient, b u t resedimentation shows that at least part of this material consists of intact networks, staying at the meniscus for unclear reasons. In the melting experiments about 10--20% of the hyperchromicity of the intact complex falls in the region where sonicated DNA melts, b u t this is hardly more than observed with closed circular p o l y o m a DNA [25]. It seems likely, therefore, that non-replicating K-DNA in vivo contains

166 only closed circular DNA and that open circles found after isolation are either due to nicking during isolation or derived from replicating networks (cf. ref. 26). It follows that the longer linear DNA molecules that have been observed in partly degraded K-DNA by us (unpublished results) and in other laboratories must represent degradation products of larger circles, if contamination with nuclear DNA is rigorously excluded. Suggestive evidence that most of the K-DNA consists of closed circles has also been reported for two other organisms. Burnett [6,7] has briefly reported that K-DNA from Tr~panosoma brucei sediments homogeneously through alkaline/NaC1 with an s20.w = 4 • 103 and does not show hyperchromicity when heated in 0.2 M Na*. Tbe fact that no melting was observed either in 7.2 M NaC104 (contrast Fig. 5} in these experiments, makes it difficult to attribute Burnett's results to the closed circular nature of the K-DNA. Results analogous to ours have been reported by Simpson and Berliner [27] for L. tarentolae. These results suggest that our finding, that virtually all K-DNA in Crithidia is in a closed circular duplex form, may apply to K-DNA in general. We have found that the density of intact K-DNA in CsC1 containing ethidium bromide or propidium diiodide is much higher than that of isolated closed mini-circles. Actually, the difference in density between nicked and intact networks is greater than previously observed for any other circular DNA in nature and the complex is the only natural DNA described up till now with a positive superhelix density. This unusual result can be explained in two ways: (1) Whereas the mini-circles in the complex have a superhelix density around 0, the rest of the DNA has an extremely high positive superhelix density, leading to an overall density of the complex of about +4 twists per 1000 base pairs. (2) The c o m p a c t tertiary structure of the network and the high degree of catenation of the molecules lead to an additional restriction on dye uptake, superimposed on the restriction due to closed circularity. Whereas no biochemical mechanism for the introduction of positive super-twists is known, previous experiments in this laboratory have shown [28] that a catenated DNA network has a greatly restricted uptake of ethidium, even though this network did not contain circles. We therefore prefer hypothesis 2. The very high density of intact K-DNA in CsC1/propidium diiodide has been of practical use in t w o respects: (i) it provides a simple method to completely remove contaminating nuclear DNA and (ii) it provides a sensitive index for the degree of nicking of the circles in the complex. It should be realised, however, that this index is essentially pragmatic because there is no way to know what the dye uptake of intact networks should be. Restricted dye uptake or the existence of only one sharp band in CsC1/ethidium bromide equilibrium gradients in itself, is no p r o o f for complete absence of nicking (see Fig. 2), in contrast to the two other parameters, melting profile and sedimentation through alkali, used in our experiments. The very high density in CsC1/ethidium bromide has not been observed with K-DNA previously isolated from Crithidia [12] or from other organisms [9,26]. In very recent work with L. tarentolae [27] networks were found to band at the same position as isolated closed mini-circles in ethidium/CsC1 gradients

167

and this was interpreted to mean that these networks were intact and contained less than 5% open circular or linear DNA. No melting curves of these networks were presented, however, and in alkaline gradients less than 50% of the DNA sedimented in a broad peak at 2800 S, the remainder being spread over the gradient. In view of this, the possibility remains that these networks were partly nicked and that intact K-DNA networks from L. tarentolae will turn out to have the same unusual high density in ethidium/CsC1 as intact K-DNA from C. luciliae. Acknowledgements We thank Mrs H.J. Kwant-den Harink for expert technical assistance and Mrs F. Fase-Fowler for providing both the labelled and the unlabelled PM2 DNA. This work was supported (in part) by a grant from the Netherlands Foundation for Chemical Research (S.O.N.) with financial aid from the Netherlands Organization for the Advancement of Pure Research (Z.W.O.). References 1 S i m p s o n , L. ( 1 9 7 2 ) I n t . Rev. C y t o l . 3 2 , 1 3 9 - - 2 0 7 2 N e w t o n , B.A. ( 1 9 7 2 ) in C o m p a r a t i v e B i o c h e m i s t r y o f P a r a s i t e s ( V a n d e n B o s s c h e , H., e d . ) , p p . 1 2 7 - - 1 3 8 , A c a d e m i c Press, N e w Y o r k 3 S t e i n e r t , M., V a n Assel, S. a n d S t e i n e r t , G. ( 1 9 6 9 ) E x p . Cell Res. 56, 6 9 - - 7 4 4 S i m p s o n , L. ( 1 9 7 3 ) J. P r o t o z o o l . 2 0 , 2 - - 8 5 L a u r e n t , M., V a n Assel, S. a n d S t e i n e r t , M. ( 1 9 7 1 ) B i o c h e m . B i o p h y s . Res. C o m m u n . 4 3 , 2 7 8 - - 2 8 4 6 B u r n e t t , J . K . ( 1 9 7 2 ) T r a n s . R. S o c . T r o p . Med. H y g . 6 6 , 3 5 4 7 B u r n e t t , J . K . ( 1 9 7 3 ) T r a n s . R. Soc. T r o p . Med. H y g . 6 7 , 2 5 4 - - 2 5 5 8 R i o u , G. a n d Delain, E. ( 1 9 6 9 ) P r o c . N a t l . A c a d . Sci. U.S. 6 2 , 2 1 0 - - 2 1 7 9 S i m p s o n , L. a n d Da Silva, A, ( 1 9 7 1 ) J. Mol. Biol. 5 6 , 4 4 3 - - 4 7 3 1 0 L a u r e n t , M. a n d S t e i n e r t , M. ( 1 9 7 0 ) P r o c . N a t l . A c a d . Sci. U.S. 6 6 , 4 1 9 - - 4 2 4 11 R e n g e r , H.C. a n d W o l s t e n h o l m e , D . R . { 1 9 7 0 ) J . Cell Biol. 4 7 , 6 8 9 - - 7 0 2 1 2 R e n g e r , H . C . a n d W o l s t e n h o l m e , D . R . ( 1 9 7 2 ) J. Cell Biol. 5 4 , 3 4 6 - - 3 6 4 13 Delain, E., B r a c k , Ch., L a c o m e , A. a n d R i o u , G. ( 1 9 7 2 ) in C o m p a r a t i v e B i o c h e m i s t r y o f P a r a s i t e s ( V a n d e n B o s s c h e , H., e d . ) , p p . 1 6 7 - - 1 8 4 , A c a d e m i c Press, N e w Y o r k 1 4 B r a c k , Ch., Delain, E., R i o u , G. a n d F e s t y , B. ( 1 9 7 2 ) J. U l t r a s t r u c t . Res. 3 9 , 5 6 8 - - 5 7 9 1 5 W e s l e y , R . D . a n d S i m p s o n , L. ( 1 9 7 3 ) B i o c h i m . B i o p h y s . A c t a 3 1 9 , 2 3 7 - - 2 5 3 16 W e s l e y , R . D . a n d S i m p s o n , L. ( 1 9 7 3 ) B i o c h i m . B i o p h y s . A c t a 3 1 9 , 2 5 4 - - 2 6 6 17 W e s l e y , R . D . a n d S i m p s o n , L. ( 1 9 7 3 ) B i o c h i m . B i o p h y s . A c t a 3 1 9 , 2 6 7 - - 2 8 0 18 B o r s t , P. ( 1 9 7 2 ) A n n u . Rev. B i o c h e m . 4 1 , 3 3 3 - - 3 7 6 19 O z e k i , Y., O n o , T., O k u b o , S. a n d I n o k i , S. ( 1 9 7 0 ) B i k e n J. 1 3 , 3 8 7 - - 3 9 3 2 0 O n o , T., O z e k i , Y., O k u b o , S. a n d I n o k i , S. ( 1 9 7 1 ) B i k e n J. 1 4 , 2 0 3 - - 2 1 5 21 S t e i n e r t , M. a n d V a n Assel, S. ( 1 9 7 2 ) in C o m p a r a t i v e B i o c h e m i s t r y o f P a r a s i t e s ( V a n d e n B o s s c h e , H., e d . ) , p p , 1 5 9 - - 1 6 6 , A c a d e m i c Press, N e w Y o r k 2 2 B o n ~ , G. a n d S t e i n e r t , M. ( 1 9 5 6 ) N a t u r e 1 7 8 , 3 0 8 2 3 G r a y , J r , H . B . , U p h o l t , W.B. a n d V i n o g r a d , J. ( 1 9 7 1 ) J. Mol. Biol. 6 2 , 1 - - 1 9 2 4 K l e i n s c h m i d t , A . K . ( 1 9 6 8 ) in M e t h o d s in E n z y m o l o g y ( C o l o w i c k , S.P. a n d K a p l a n , N . O . , e d s ) , Vol. 1 2 B , p p . 3 6 1 - - 3 7 7 , A c a d e m i c Press, N e w Y o r k 2 5 V i n o g r a d , J., L e b o w i t z , J. a n d W a t s o n , R . ( 1 9 6 8 ) J. Mol. Biol. 3 3 , 1 7 3 - - 1 9 7 2 6 S i m p s o n , L., S i m p s o n , A.M. a n d W e s l e y , R . D . ( 1 9 7 4 ) B i o c h i m . B i o p h y s . A c t a 3 4 9 , 1 6 1 - - 1 7 2 2 7 S i m p s o n , L. a n d Berliner, J. ( 1 9 7 4 ) J . P r o t o z o o l . 2 1 , 3 8 2 - - 3 9 3 28 H o l l e n b e r g , C.P., B o r s t , P., Flavell, R . A . , V a n Kreijl, C . F . , V a n B r u g g e n , E . F . J . a n d A r n b e r g , A.C. (1972) Biochim. Biophys. Acta 277, 44--58 29 E s p e j o , R . T . , C a n e l o , E,S. a n d S i n s h e i m e r , R . L . ( 1 9 6 9 ) P r o c . N a t l . A c a d . Sci. U.S. 6 3 , 1 1 6 4 - - 1 1 6 8 3 0 S c h i l d k r a u t , C . L . , M a r m u r , J. a n d D o t y , P. ( 1 9 6 2 ) J. Mol. Biol. 4, 4 3 0 - - 4 4 3 31 F r a n k - K a m e n e t s k i i , M.D. ( 1 9 7 1 ) B i o p o l y m e r s 1 0 , 2 6 2 3 - - 2 6 2 4