Replication of kinetoplast DNA: an update for the new millennium

Invited Review

Replication of kinetoplast DNA: an update for the new millennium

James C. Morris*, Mark E. Drew, Michele M. Klingbeil, Shawn A. Motyka,Tina T. Saxowsky, Zefeng Wang, Paul T. Englund

Department of Biological Chemistry, Johns Hopkins Medical School, Baltimore, MD 21205, USA

Received 2 October 2000; received in revised form 11 December 2000; accepted 11 December 2000

Abstract

In this review we will describe the replication of kinetoplast DNA, a subject that our lab has studied for many years. Our knowledge of

kinetoplast DNA replication has depended mostly upon the investigation of the biochemical properties and intramitochondrial localisation of

replication proteins and enzymes as well as a study of the structure and dynamics of kinetoplast DNA replication intermediates. We will ®rst

review the properties of the characterised kinetoplast DNA replication proteins and then describe our current model for kinetoplast DNA

replication. q 2001 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved.

Keywords: Kinetoplast DNA; Trypanosoma; DNA replication

1. Introduction

Protozoan parasites in the family Trypanosomatidae are

early diverging eukaryotes that cause important tropical

diseases including African sleeping sickness, leishmaniasis,

and Chagas' disease in humans as well as nagana in African

livestock. All of the trypanosomatid parasites have a

remarkable mitochondrial DNA, termed kinetoplast DNA

(kDNA), that has a structure unlike that of any other

known DNA in nature. Within the matrix of each cell's

single mitochondrion the kDNA is a network of a few thou-

sand topologically interlocked DNA circles. There are two

types of circles, maxicircles and minicircles. Each network

contains several dozen maxicircles (in most species they

range in size from about 20 to 40 kb) and several thousand

minicircles (usually 0.5±2.5 kb, although in some species

they are larger). For a more comprehensive review on

kDNA see Shapiro and Englund (1995). Like mitochondrial

DNAs from mammalian cells or yeast, maxicircles encode

ribosomal RNAs and some of the proteins required for mito-

chondrial bioenergetic processes. Some RNA transcripts of

maxicircles are post-transcriptionally modi®ed by the inser-

tion or deletion of uridine residues to form functional open

reading frames, a process termed RNA editing. Editing

speci®city is directed by guide RNAs that are encoded by

the minicircles. For a review on editing see Estevez and

Simpson (1999).

Most studies of kDNA replication in our laboratory, the

Ray laboratory (UCLA) and the Shlomai laboratory

(Hebrew University) have focused on the insect parasite

Crithidia fasciculata. Crithidia fasciculata kDNA networks

puri®ed from non-replicating cells are remarkably homoge-

neous in size and shape, being planar, elliptically-shaped

structures about 10 by 15 mm in size (see EM in Fig. 1

showing a segment of an isolated kDNA network). All of

the minicircles are covalently closed, relaxed, and linked to

an average of three neighbouring minicircles by single inter-

locks (Rauch et al., 1993; Chen et al., 1995). Topologically,

the network has a striking resemblance to the chain mail of

medieval armour. Within the parasite's single mitochon-

drion, the network is condensed in a highly ordered fashion

into a disk-shaped structure about 1 mm in diameter and

0.35 mm thick. (Fig. 2 illustrates how the kDNA is

condensed into a disk.) The kDNA disk is always positioned

near the basal body of the ¯agellum and perpendicular to the

axis of the ¯agellum. Remarkably, there is evidence for a

direct physical linkage between the basal body and the

kDNA network, even though these two structures are sepa-

rated by the double membrane of the mitochondrion (Robin-

son and Gull, 1991).

In this review we describe the replication of kDNA, a

subject that our lab has studied for many years. Our knowl-

edge of kDNA replication has depended mostly upon the

investigation of the biochemical properties and intramito-

chondrial localisation of replication proteins and enzymes

as well as a study of the structure and dynamics of kDNA

replication intermediates. We will ®rst review the properties

International Journal for Parasitology 31 (2001) 453±458

0020-7519/01/$20.00 q 2001 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved.

PII: S0020-7519(01)00156-4

www.parasitology-online.com

* Corresponding author. Tel.: 11-410-955-3458; fax: 11-410-955-7810.

E-mail address: [email protected] (J.C. Morris).

of the characterised kDNA replication proteins and then

describe our current model for kDNA replication.

2. Proteins involved in kDNA replication andmaintenance

Replication proteins have been studied mainly in C. fasci-

culata. This parasite is ideal for enzyme puri®cation and

biochemical studies as it is non-pathogenic, it can be

grown in large quantities (up to 150 L, which yields ,400

g of cells) in inexpensive medium, and there is an ef®cient

method for isolating mitochondria (T. Saxowsky and M.

Klingbeil, unpublished data). In this section we will discuss

the properties of the puri®ed enzymes and proteins. Later we

shall review their intramitochondrial localisation and spec-

ulate on their function in kDNA replication.

2.1. Topoisomerase II

The ®rst mitochondrial replication enzyme puri®ed to

homogeneity from C. fasciculata was a type II topoisome-

rase (topo II) (Melendy and Ray, 1989). This topo II is a

homodimer of 132 kDa subunits. Like other enzymes of this

type, it is ATP-dependent and catalyses catenation and

decatenation of DNA in vitro. A homologue of the C. fasci-

culata enzyme has been cloned from Trypanosoma brucei

(Strauss and Wang, 1990). This topoisomerase, as well as

others characterised from T. brucei, is sensitive to many

conventional topoisomerase inhibitors. These inhibitors,

such as etoposide and VP16, have been valuable in studying

enzyme function (Ray et al., 1992; Shapiro, 1994; Nenortas

et al., 1998). A second topo II that has a distinct intramito-

chondrial localisation has been partially puri®ed from C.

fasciculata (Shlomai et al., 1984).

2.2. Universal minicircle sequence binding protein

Part of the minicircle replication origin, the initiation site

for leading strand synthesis, is a 12 nucleotide sequence

known as the universal minicircle sequence (UMS). This

sequence is `universal' because it is found, with virtually

no variation, in minicircles from all trypanosomatid species

examined. A UMS binding protein (UMSBP) has been puri-

®ed from C. fasciculata and is a homodimer of 13.7 kDa

subunits (Tzfati et al., 1992). Amazingly, this protein also

binds to DNA fragments containing a six nucleotide

sequence (,80 nucleotides from the UMS) that serves as

the initiation site for the ®rst Okazaki fragment (Abu-Elneel

et al., 1999). This origin recognition protein, which likely

plays a role in the initiation of minicircle replication, does

not bind to double-stranded oligonucleotides containing the

UMS dodecamer or the hexameric sequence, although it

binds tightly and speci®cally to these sequences in single-

stranded form (Abeliovich et al., 1993). Surprisingly, it does

bind to these sequences in double-stranded form in cova-

lently-closed intact free minicircles (Avrahami et al., 1995).

Apparently, the minicircle sequence dictates some structural

deformation in the origin region that allows binding (Avra-

hami et al., 1995).

2.3. Primase

A 28 kDa protein that can synthesise small oligoribonu-

cleotides (up to about 10 nucleotides in size) has been puri-

®ed from C. fasciculata mitochondria. The small RNAs that

are products of this enzyme can prime Klenow DNA poly-

merase to initiate DNA synthesis in vitro (Li and Englund,

1997). Further characterisation of this enzyme is ongoing.

2.4. DNA polymerase b

A small (43 kDa) DNA polymerase b (pol b) has been

puri®ed from C. fasciculata mitochondria (Torri and

Englund, 1992). Biochemical studies indicate that the

J.C. Morris et al. / International Journal for Parasitology 31 (2001) 453±458454

Fig. 1. EM showing a segment of a puri®ed C. fasciculata kinetoplast DNA

network. Small loops are the 2.5 kb minicircles, and long strands threading

through the network interior are parts of the 38 kb maxicircles. EM by

David PeÂrez-Morga.

Fig. 2. Organisation of the kinetoplast DNA network in vivo. The C. fasci-

culata network is a disk 1 mm in diameter and 0.35 mm thick. The pie-

shaped sector shows individual interlocked minicircles stretched out paral-

lel to the disk's axis.

enzyme is non-processive and, because it lacks a 3 0 proof-

reading exonuclease, error prone. However, pol b is ef®-

cient in ®lling small gaps (Torri et al., 1994). Sequence

analysis indicates that this protein is related to mammalian

pol b (33% identical in sequence to the human enzyme).

This is the ®rst b-type polymerase described in mitochon-

dria (Torri and Englund, 1995). Mammalian nuclear b poly-

merases and the yeast pol b homologue (Pol IV) function in

base excision repair. The role of C. fasciculata mitochon-

drial pol b is not fully understood, but the enzyme is prob-

ably not the major replicative polymerase (see below).

2.5. Ribonuclease H

Structure-speci®c endonuclease 1 (SSE1) from C. fasci-

culata mitochondria is a 32 kDa enzyme that has ribonu-

clease H activity and may be involved in primer removal

(Engel and Ray, 1998). This protein has a domain similar to

the 5 0 exonuclease domain of bacterial DNA polymerase I

(Engel and Ray, 1999). In vitro studies have revealed that

SSE1 recognises the structure of its substrate, cleaving a

non-base-paired 5 0 tail on the 3 0 side of its ®rst base-paired

nucleotide (Engel and Ray, 1998). There are two other

detectable RNase H activities (38 and 45 kDa) in C. fasci-

culata, of which the larger is enriched in puri®ed kineto-

plasts (Ray and Hines, 1995; Engel and Ray, 1998).

Interestingly, a single gene (RNH1) encodes both of these

proteins, which are not essential for cell viability as demon-

strated by genetic knockout (Ray and Hines, 1995). These

proteins can complement an Escherichia coli strain defec-

tive in RNase H, an enzyme implicated in the regulation of

RNA priming (Campbell and Ray, 1993).

2.6. p18, p17, and p16

The C. fasciculata genes KAP2, KAP3, and KAP4 encode

the small basic proteins p18, p17, and p16, respectively (Xu

and Ray, 1993; Xu et al., 1996). These proteins associate

tightly with kDNA, as they were recovered from isolated

kDNA networks after reversible cross-linking in vivo with

formaldehyde (Xu and Ray, 1993). These histone-like

proteins can condense kDNA in vitro and can rescue E.

coli that are de®cient in the HU protein, a DNA-binding

protein that plays a role in chromosomal condensation,

replication, and recombination (Xu et al., 1996).

3. Localisation of kDNA replication proteins within themitochondrial matrix

Immunoelectron and immuno¯uorescence microscopy

have revealed speci®c localisation of these enzymes in

distinct sites around the kinetoplast disk (see diagram in

Fig. 3). The kDNA disk is ¯anked by two antipodal sites

containing at least three enzymes involved in replication.

These are topo II (Melendy et al., 1988), pol b (Ferguson et

al., 1992), and SSE1 (Engel and Ray, 1998). Based on ¯uor-

escence in situ hybridisation, there are also free minicircle

replication intermediates in the antipodal sites (Ferguson et

al., 1992). There is preliminary evidence that the second

topo II is localised throughout the kDNA disk (Shlomai,

1994). Primase is localised on the anterior and posterior

faces of the disk (Li and Englund, 1997). The protein loca-

lisation diagrammed in Fig. 3 refers to cells undergoing

kDNA replication. At other non-replicative stages of the

cell cycle some of the proteins alter their location (Johnson

and Englund, 1998).

4. kDNA replication intermediates

The kDNA's network structure complicates its replica-

tion mechanism. The problem is that each network, contain-

ing 5000 covalently-closed minicircles (in the case of C.

fasciculata), must double their minicircle copy number

during each cell cycle. The two progeny networks must be

distributed to the two daughter cells during cell division.

There are serious topological problems that must be over-

come for this process to occur. In this review we will focus

on the replication of minicircles, but see Hajduk et al.

(1984) and Carpenter and Englund (1995) for a description

of maxicircle replication.

Study of minicircle replication intermediates, mostly in

C. fasciculata, has uncovered the following highlights of the

replication mechanism. (A) Replication occurs during a

discrete phase of the cell cycle, nearly concurrent with the

nuclear S phase (Cosgrove and Skeen, 1970). (B) Prior to

replication, minicircles are covalently closed, and after

replication they are gapped. The presence of gaps is thought

to distinguish newly replicated minicircles from those that

have not been replicated, ensuring that each replicates once

per cell cycle (Englund, 1978). (C) Minicircles do not repli-

cate while linked to the network, but instead they are indi-

vidually released from the network, presumably by a topo II

(Englund, 1979). (D) The covalently-closed free minicircles

replicate unidirectionally as u-structures, forming gapped

progeny (Kitchin et al., 1984; Ntambi and Englund, 1985;

J.C. Morris et al. / International Journal for Parasitology 31 (2001) 453±458 455

Fig. 3. Localisation of kinetoplast DNA replication enzymes.

Birkenmeyer and Ray, 1986; Birkenmeyer et al., 1987). (E)

Reattachment of the replicated gapped free minicircles

occurs at the network periphery (Englund, 1978; Guilbride

and Englund, 1998). This speci®city of free minicircle reat-

tachment leads to the development of two zones in the

replicating network, a peripheral zone of newly replicated

gapped minicircles and a central zone of covalently-closed

minicircles. As replication proceeds the peripheral zone of

gapped minicircles enlarges and the central zone of cova-

lently-closed minicircles shrinks. See Fig. 4 for a diagram of

minicircle release and reattachment and Fig. 5 for evidence

that newly replicated gapped minicircles are localised

around the network periphery. (F) When all the minicircles

have replicated, the minicircle copy number has doubled, to

10 000 in the case of C. fasciculata (PeÂrez-Morga and

Englund, 1993b). At this time the gaps are repaired, the

network undergoes scission, and the two networks, each

containing a complete complement of covalently-closed

minicircles, are distributed to the two daughter cells during

cell division.

5. The current replication model

The diagram in Fig. 6 shows a section through the

network with newly replicated and reattached minicircles

(bold circles) indicated at the edges of the disk. The disk

is ¯anked by the two antipodal sites and it is sandwiched by

the two zones of primase. Covalently-closed minicircles are

released from the kDNA disk, possibly by the topo II that is

thought to reside in this region. Once released from the

network the free minicircles encounter primase and possibly

other proteins such as UMSBP, helicases, and the replica-

tive polymerase. There are two possibilities as to what could

happen next. The free minicircles could assemble into a

replication initiation complex and migrate to the antipodal

sites to complete replication. Alternatively, they could

J.C. Morris et al. / International Journal for Parasitology 31 (2001) 453±458456

Fig. 4. Diagram of a replicating network and free minicircles, not drawn to

scale. Covalently-closed minicircles are released from the network and

undergo replication, forming two progeny containing gaps. These are reat-

tached to the network periphery. The region of the network containing

gapped minicircles is shown by dots.

Fig. 5. Isolated kinetoplast DNA networks visualised by ¯uorescence microscopy. (Left) Networks stained with 4 0,6-diamidino-2-phenylindole (DAPI).

(Right) Same networks in which the gapped minicircles are labelled with ¯uorescein-deoxyuridine triphosphate (dUTP) using terminal transferase. Note

the peripheral localisation of gapped minicircles. Networks with a narrow ring of ¯uorescein ¯uorescence are from early stages of replication. Those labelled

uniformly with ¯uorescein have all minicircles replicated. Images by Lys Guilbride (Guilbride and Englund, 1998).

complete replication before migrating to the antipodal sites

and the gapped progeny could then move to these sites. In

both models, many of the minicircle gaps are repaired at the

antipodal sites. This process could involve primer removal

by SSE1 and gap ®lling by the pol b. The progeny mini-

circles, still containing one or a small number of gaps, can

then be attached to the network periphery by a topo II and

are completely repaired by pol b and a DNA ligase.

There is a problem with this model. It predicts that

progeny gapped free minicircles would be linked to the

network only adjacent to the antipodal sites. Yet the ¯uor-

escence images shown in Fig. 5 clearly demonstrate that the

gapped minicircle progeny are distributed uniformly around

the network periphery. How does this uniform distribution

occur? Several years ago we provided evidence that there is

a relative movement of the kDNA disk and the antipodal

sites that could easily account for the uniform distribution of

the minicircles around the replicating network (PeÂrez-

Morga and Englund, 1993a). One possibility, shown in

Fig. 7, is that the kDNA disk actually spins, leading to a

uniform distribution of minicircles around the disk.

6. What will happen in the new millennium?

As discussed in this review, biochemical characterisation

of C. fasciculata replication proteins in our lab and other

labs has been a powerful method for elucidating the kDNA

replication mechanism. However, there could be problems

ahead if we continue to depend exclusively on this

approach. For example, we found clear evidence that C.

fasciculata mitochondria contain DNA polymerase activ-

ities in addition to the well-characterised pol b (M. Kling-

beil, unpublished data), one of which could be the

replicative polymerase. However, attempts to purify the

polymerase were unsuccessful because of its instability.

As an alternative, approaches based on genomics could be

useful in the identi®cation of proteins involved in kDNA

replication. Putative coding regions can be identi®ed in the

rapidly advancing T. brucei genome project using homol-

ogy-based searches. This sequence information is suf®cient

for speci®c inhibition of gene expression utilising the

recently developed technique of RNA interference

(RNAi). This technique works well in T. brucei (Ngo et

al., 1998; Wang et al., 2000). Genes responsible for pheno-

types associated with kDNA can be cloned and recombinant

proteins expressed in order to study their enzymatic proper-

ties and subcellular localisation. This coupling of genomic

and proteomic strategies may provide the next advance in

our understanding of the kDNA replication mechanism.

Stay tuned.

Acknowledgements

We thank Viiu Klein for valuable contributions to this

work. M.M.K. is supported by National Research Service

Award Fellowship 5F32AI09789. T.T.S. is supported by the

Fannie and John Hertz Foundation. Work in our lab is

supported by grant GM 27608 from the National Institutes

of Health.

J.C. Morris et al. / International Journal for Parasitology 31 (2001) 453±458 457

Fig. 7. The spinning kinetoplast. The ellipse represents the top of the

kinetoplast disk. Small circles represent antipodal sites. Kinetoplast spins

in the direction of small arrows. Solid lines represent rows of minicircles

that are attached adjacent to antipodal sites. After nearly half a turn (lower

right) the periphery is almost completely ®lled. Continued spinning of the

kinetoplast results in minicircles being attached in a spiral pattern (PeÂrez-

Morga and Englund, 1993a).

Fig. 6. Kinetoplast DNA replication model. See text for details.

References

Abeliovich, H., Tzfati, Y., Shlomai, J., 1993. A trypanosomal CCHC-type

zinc ®nger protein which binds the conserved universal sequence of

kinetoplast DNA minicircles: isolation and analysis of the complete

cDNA from Crithidia fasciculata. Mol. Cell. Biol. 13, 7766±73.

Abu-Elneel, K., Kapeller, I., Shlomai, J., 1999. Universal minicircle

sequence-binding protein, a sequence-speci®c DNA-binding protein

that recognizes the two replication origins of the kinetoplast DNA

minicircle. J. Biol. Chem. 274, 13419±26.

Avrahami, D., Tzfati, Y., Shlomai, J., 1995. A single-stranded DNA bind-

ing protein binds the origin of replication of the duplex kinetoplast

DNA. Proc. Natl. Acad. Sci. USA 92, 10511±5.

Birkenmeyer, L., Ray, D.S., 1986. Replication of kinetoplast DNA in

isolated kinetoplasts from Crithidia fasciculata. Identi®cation of mini-

circle DNA replication intermediates. J. Biol. Chem. 261, 2362±8.

Birkenmeyer, L., Sugisaki, H., Ray, D.S., 1987. Structural characterization

of site-speci®c discontinuities associated with replication origins of

minicircle DNA from Crithidia fasciculata. J. Biol. Chem. 262,

2384±92.

Campbell, A.G., Ray, D.S., 1993. Functional complementation of an

Escherichia coli ribonuclease H mutation by a cloned genomic frag-

ment from the trypanosomatid Crithidia fasciculata. Proc. Natl. Acad.

Sci. USA 90, 9350±4.

Carpenter, L.R., Englund, P.T., 1995. Kinetoplast maxicircle DNA replica-

tion in Crithidia fasciculata and Trypanosoma brucei. Mol. Cell. Biol.

15, 6794±803.

Chen, J., Rauch, C.A., White, J.H., Englund, P.T., Cozzarelli, N.R., 1995.

The topology of the kinetoplast DNA network. Cell 80, 61±69.

Cosgrove, W.B., Skeen, M.J., 1970. The cell cycle in Crithidia fasciculata.

Temporal relationships between synthesis of deoxyribonucleic acid in

the nucleus and in the kinetoplast. J. Protozool. 17, 172±7.

Engel, M.L., Ray, D.S., 1998. A structure-speci®c DNA endonuclease is

enriched in kinetoplasts puri®ed from Crithidia fasciculata. Nucleic

Acids Res. 26, 4733±8.

Engel, M.L., Ray, D.S., 1999. The kinetoplast structure-speci®c endonu-

clease I is related to the 5 0 exo/endonuclease domain of bacterial DNA

polymerase I and colocalizes with the kinetoplast topoisomerase II and

DNA polymerase beta during replication. Proc. Natl. Acad. Sci. USA

96, 8455±60.

Englund, P.T., 1978. The replication of kinetoplast DNA networks in

Crithidia fasciculata. Cell 14, 157±68.

Englund, P.T., 1979. Free minicircles of kinetoplast DNA in Crithidia

fasciculata. J. Biol. Chem. 254, 4895±900.

Estevez, A.M., Simpson, L., 1999. Uridine insertion/deletion RNA editing

in trypanosome mitochondria ± a review. Gene 240, 247±60.

Ferguson, M., Torri, A.F., Ward, D.C., Englund, P.T., 1992. In situ hybri-

dization to the Crithidia fasciculata kinetoplast reveals two antipodal

sites involved in kinetoplast DNA replication. Cell 70, 621±9.

Guilbride, D.L., Englund, P.T., 1998. The replication mechanism of kine-

toplast DNA networks in several trypanosomatid species. J. Cell Sci.

111, 675±9.

Hajduk, S.L., Klein, V.A., Englund, P.T., 1984. Replication of kinetoplast

DNA maxicircles. Cell 36, 483±92.

Johnson, C.E., Englund, P.T., 1998. Changes in organization of Crithidia

fasciculata kinetoplast DNA replication proteins during the cell cycle.

J. Cell Biol. 143, 911±9.

Kitchin, P.A., Klein, V.A., Fein, B.I., Englund, P.T., 1984. Gapped mini-

circles. A novel replication intermediate of kinetoplast DNA. J. Biol.

Chem. 259, 15532±9.

Li, C., Englund, P.T., 1997. A mitochondrial DNA primase from the trypa-

nosomatid Crithidia fasciculata. J. Biol. Chem. 272, 20787±92.

Melendy, T., Ray, D.S., 1989. Novobiocin af®nity puri®cation of a mito-

chondrial type II topoisomerase from the trypanosomatid Crithidia

fasciculata. J. Biol. Chem. 264, 1870±6.

Melendy, T., Sheline, C., Ray, D.S., 1988. Localization of a type II DNA

topoisomerase to two sites at the periphery of the kinetoplast DNA of

Crithidia fasciculata. Cell 55, 1083±8.

Nenortas, E.C., Bodley, A.L., Shapiro, T.A., 1998. DNA topoisomerases: a

new twist for antiparasitic chemotherapy? Biochim. Biophys. Acta

1400, 349±54.

Ngo, H., Tschudi, C., Gull, K., Ullu, E., 1998. Double-stranded RNA

induces mRNA degradation in Trypanosoma brucei. Proc. Natl.

Acad. Sci. USA 95, 14687±92.

Ntambi, J.M., Englund, P.T., 1985. A gap at a unique location in newly

replicated kinetoplast DNA minicircles from Trypanosoma equiper-

dum. J. Biol. Chem. 260, 5574±9.

PeÂrez-Morga, D., Englund, P.T., 1993a. The attachment of minicircles to

kinetoplast DNA networks during replication. Cell 74, 703±11.

PeÂrez-Morga, D., Englund, P.T., 1993b. The structure of replicating kine-

toplast DNA networks. J. Cell Biol. 123, 1069±79.

Rauch, C.A., PeÂrez-Morga, D., Cozzarelli, N.R., Englund, P.T., 1993. The

absence of supercoiling in kinetoplast DNA minicircles. EMBO J. 12,

403±11.

Ray, D.S., Hines, J.C., 1995. Disruption of the Crithidia fasciculata RNH1

gene results in the loss of two active forms of ribonuclease H. Nucleic

Acids Res. 23, 2526±30.

Ray, D.S., Hines, J.C., Anderson, M., 1992. Kinetoplast-associated DNA

topoisomerase in Crithidia fasciculata: crosslinking of mitochondrial

topoisomerase II to both minicircles and maxicircles in cells treated

with the topoisomerase inhibitor VP16. Nucleic Acids Res. 20, 3353±6.

Robinson, D.R., Gull, K., 1991. Basal body movements as a mechanism for

mitochondrial genome segregation in the trypanosome cell cycle.

Nature 352, 731±3.

Shapiro, T.A., 1994. Mitochondrial topoisomerase II activity is essential for

kinetoplast DNA minicircle segregation. Mol. Cell. Biol. 14, 3660±7.

Shapiro, T.A., Englund, P.T., 1995. The structure and replication of kine-

toplast DNA. Annu. Rev. Microbiol. 49, 117±43.

Shlomai, J., 1994. The assembly of kinetoplast DNA. Parasitol. Today 10,

341±6.

Shlomai, J., Zadok, A., Frank, D., 1984. A unique ATP-dependent DNA

topoisomerase from trypanosomatids. Adv. Exp. Med. Biol. 179, 409±

22.

Strauss, P.R., Wang, J.C., 1990. The TOP2 gene of Trypanosoma brucei: a

single-copy gene that shares extensive homology with other TOP2

genes encoding eukaryotic DNA topoisomerase II. Mol. Biochem. Para-

sitol. 38, 141±50.

Torri, A.F., Englund, P.T., 1992. Puri®cation of a mitochondrial DNA

polymerase from Crithidia fasciculata. J. Biol. Chem. 267, 4786±92.

Torri, A.F., Englund, P.T., 1995. A DNA polymerase beta in the mitochon-

drion of the trypanosomatid Crithidia fasciculata. J. Biol. Chem. 270,

3495±7.

Torri, A.F., Kunkel, T.A., Englund, P.T., 1994. A beta-like DNA polymer-

ase from the mitochondrion of the trypanosomatid Crithidia fascicu-

lata. J. Biol. Chem. 269, 8165±71.

Tzfati, Y., Abeliovich, H., Kapeller, I., Shlomai, J., 1992. A single-stranded

DNA-binding protein from Crithidia fasciculata recognizes the nucleo-

tide sequence at the origin of replication of kinetoplast DNA minicir-

cles. Proc. Natl. Acad. Sci. USA 89, 6891±5.

Wang, Z., Morris, J.C., Drew, M.E., Englund, P.T., 2000. Inhibition of

Trypanosoma brucei gene expression by RNA interference using an

integratable vector with opposing T7 promoters. J. Biol. Chem. 275,

40174±9.

Xu, C., Ray, D.S., 1993. Isolation of proteins associated with kinetoplast

DNA networks in vivo. Proc. Natl. Acad. Sci. USA 90, 1786±9.

Xu, C.W., Hines, J.C., Engel, M.L., Russell, D.G., Ray, D.S., 1996.

Nucleus-encoded histone H1-like proteins are associated with kineto-

plast DNA in the trypanosomatid Crithidia fasciculata. Mol. Cell. Biol.

16, 564±76.

J.C. Morris et al. / International Journal for Parasitology 31 (2001) 453±458458