The Pennsylvania State University, Graduate School
The Department of Molecular and Cell Biology
DNA REPAIR AND THE Dictyostelium discoideum CELL-CYCLE
Brooks John Kelly
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
I compared the responses of Dictyostelium discoideum radB (HPS517 and HPS521) cells against that of the wild-type (HPS401) by measurments of: cell division lags after UV irradiation, CIPC or aphidicolin treatment; DNA synthesis lags after exposure to aphidicolin or UV irradiation; rates of nuclear DNA repair after UV irradiation with alkaline sucrose gradients. In addition, flow cytometric analysis of nuclei from CIPC- and/or aphidicolin-arrested cells was used to quantitate the D. discoideum cell-cycle. DNA damage/repair and its influence on the D. discoideum cell-cycle was further examined by size distribution analysis of whole-cells after UV irradiation, or during treatment with CIPC or aphidicolin. The incorporation of tritiated thymidine (3H-TdR) into DNA, after a 30 min pulse-labeling, was examined for the non-thymidine incorporating strain AX-2 and the thymidine incorporating strain HPS401 by CsCl gradients of both whole-cell and nuclear lysate.
The wild-type and radB strains undergo the same initial rate of DNA damage/repair after UV. Unless perturbed by UV irradiation or treatment with CIPC or aphidicolin, axenic cells are predominantly in the G1-phase (51-59%) of their cell-cycle, and to a lesser extent in S-phase (23%) and G2/M-phases (18-26%). Wild-type and radB strains become mitotically inhibited during treatment with CIPC. Wild-type cells, but not the radB strains, arrest in S-phase during aphidicolin treatment and in S- and G2-/M-phases after UV irradiation. The wild-type ability to arrest in both S- and G2-/M-phases after UV irradiation could involve a single DNA polymerase.
This thesis also hypothesizes that stationary phase cells arrest mostly in G0/G1-phase (57%), and to a lesser extent in S- (9%), and G2-/M-phases (34%), and that cells with a G0/G1-phase content of DNA transform into spore cells, cells with an S-phase content of DNA transform to stalk adhesion cells, and cells with a G2-/M-phase content of DNA transform into stalk cells.
1. Wild-type (HPS401) and radB (HPS517) whole-cell
uptake of 3H-TdR as a function of short periods
of time (min)........................................................ 37
2. Wild-type (HPS401) and radB (HPS517) nuclear
uptake of 3H-TdR during short periods of
time (min)............................................................. 38
3. Wild-type (HPS401) and radB (HPS517 and HPS521)
whole-cell uptake of 3H-TdR over long periods
of time (h)............................................................. 40
4. Wild-type (HPS401) and radB (HPS517 and HPS521)
nuclear uptake of 3H-TdR as a function of long
periods of time (h)............................................ 41
5. Fractional survival of wild-type (HPS401) and radB
(HPS517 and HPS521) cells after UV.................... 42
6. Wild-type (HPS401) cell division after UV.................... 44
7. radB (HPS521) cell division after UV.............................. 45
8. Wild-type nuclear DNA synthesis after UV.................. 46
9. radB (HPS521) nuclear DNA synthesis after UV.......... 48
10. Wild-type (HPS401) nicking and rejoining of single-
strand DNA as a function of time after UV, as seen
on alkaline sucrose gradients....................... 50
11. radB (HPS521) nicking and rejoining of single-strand
DNA as a function of time after UV irradiation,
as seen on alkaline sucrose gradients................... 51
12. Summary of wild-type and radB single-stranded nuclear
DNA nicking and rejoining, as a function of time after
UV, by integration of the areas under the curves of
the nuclear-main-band fractions from data
which produced Figures 10 and 11....................... 53
13. Standard distribution of wild-type nuclei as seen on a
flow cytometer.................................................... 54
14. Wild-type cell division arrest with increasing amounts of
15. Wild-type cell division arrest with increasing amounts of
16. Summary of nuclear flow cytometer profiles of wild-type
cells that were either mitotically inhibited with CIPC
(3.5 mg/ml), or arrested in S-phase with aphidicolin
(55 mg/ml)............................................................. 59
17. Wild-type cell division arrests during CIPC or aphidicolin
18. Size distributions of axenic wild-type control cells
(Control 1) as a function of time during log phase
19. Size distributions as a function of time for wild-type
control cells (Control 2) that were centrifuged 600
rpm x 3 min, resuspended in PBS, recentrifuged
and resuspended in HL5................................. 68
20. Size distributions of wild-type cells arresting in S-phase
with aphidicolin, as a function of time................. 69
21. Size distributions of wild-type cells as a function of time
during mitotic inhibition by CIPC........................... 70
22. Summary of size distributions of wild-type cells that
were either mitotically inhibited with CIPC or
arrested in S-phase with aphidicolin.................... 72
23. Wild-type cell division after release from aphidicolin
arrest as a function of time........................... 73
24. Size distributions of wild-type cells during the initial
6.25 h after UV irradiation............................ 75
25. Size distributions of wild-type cells during the period
between 7.5 h and 12.5 h after UV ...................... 76
26. Size distributions of wild-type cells during the period
between 13.5 h and 19.5 h after UV..................... 78
27. Cell division in two independently isolated radB mutants
(HPS517 and HPS521) during treatment with aphidicolin
(55 mg/ml HL5), or CIPC (3.5 mg/ml
28. Size distributions of radB (HPS517) control cells as a
function of time.................................................. 80
29. Size distributions of radB (HPS517) cells as a function
of time during aphidicolin treatment................... 82
30. Size distributions of radB (HPS521) cells as a function
of time during aphidicolin treatment................... 83
31. Size distributions of radB (HPS517) cells as a function
of time during mitotic inhibition by CIPC........... 84
32. Size distributions of radB (HPS521) cells as a function
of time during mitotic inhibition by CIPC........... 85
33. Cell division of radB (HPS517 and HPS521) strains as a
function of time after UV irradiation.................... 86
34. Size distributions of the radB (HPS517) control cells
for Figures 34-37, as a function of time. These
control cells underwent centrifugation, resuspension
in PBS, no UV irradiation, then recentrifugation
and resuspension in HL5............................................ 88
35. Size distributions of the radB (HPS521) control cells for
Figures 34-37, as a function of time. These control
cells underwent centrifugation, resuspension in PBS,
no UV irradiation, recentrifugation and resuspension
in HL5..................................................................... 89
36. Size distributions of radB (HPS517) cells as a function
of time after centrifugation, resuspension in PBS,
UV irradiation (50 J/m2), recentrifugation and
resuspension in HL5................................................... 90
37. Size distributions of radB (HPS521) cells as a function
of time after centrifugation, resuspension in
PBS, UV irradiation (50 J/m2), recentrifugation
and resuspension in HL5........................................... 91
38. Cell volume at peak of distribution, as a function of
time after UV, from Figures 24-26 and 36........ 92
39. Division in wild-type (HPS401) and radB (HPS517) cells
as a function of time during aphidicolin treatment,
from one of three in vivo nuclear DNA synthesis
experiments shown in Figures 40 and 41........... 95
40. Fraction of wild-type 3H-TdR incorporation as a function of
time after addition of aphidicolin, data from three
41. Fraction of radB (HPS517) 3H-TdR incorporation as a
function of time after addition of aphidicolin, data
from three experiments.............................................. 98
42. Uptake of 3H-TdR in AX-2 and HPS401 cells after 100
mCi/ml pulses lasting 15 and 55 min.................... 100
43. CsCl ultracentrifugation profiles of whole-cell and nuclear
lysates of AX-2 and HPS401 strains after a 30 min
pulse of 3H-TdR.................................................. 101
44. Expanded view of Figure 43, showing the CsCl profiles of
pulse labeled AX-2 whole-cells and nuclei........ 102
45. This adaptation of a previously published figure depicts
two flow cytometric histograms plotting the number
of cells against the DNA content. The first histogram
was derived from exponentially growing vegetative
cells. The second was derived from growth arrested
cells which had been held at stationary phase for
12 h.......................................................................... 108
46. This adaptation of a previously published figure depicts
the frequency of nuclei versus nuclear diameter,
as derived by averaging the diameters of camera
lucida drawings of 30 nuclei, at each state of the
D. discoideum life cycle................................... 112
47. Same as Figure 46, but corrected to show the proper
nuclear proportion of spore cells to stalk cells
that normally develops from vegetative cells.... 117
1. Phases of the Dictyostelium discoideum cell-cycle,
analyzed from data shown in Figure 13.............. 62
Often, the genomes of many different types of organisms contain billions of base pairs of DNA. One of the crucial requirements of this genetic material is that it be reproduced accurately in only a few hours. Normally, DNA synthesis occurs sequentially on chromosomes during S-phase of the eukaryotic cell-cycle at multiple origins of replication within replicons. Throughout S-phase, the chromosomal DNA is replicated by semiconservative DNA synthesis to form two daughter duplexes. Synthesis of DNA proceeds by a disruption and copying of the double helical parental DNA by multiprotein complexes called replisomes. A replisome exists as a protein complex associated with the particular structure that DNA takes at the replication fork. Because DNA strands are antiparallel while polymerization of DNA occurs only in the 5' to 3' direction, semiconservtive replication of double-stranded DNA is an asymmetric process, i.e. there is a leading strand and a lagging strand produced from, respectively, continuous and discontinuous DNA synthesis at the replication fork. These two different activities at the growing replication fork are due to the actions of different polymerase molecules embedded within the replisome.
Sometimes DNA becomes damaged and must be repaired. When DNA is replicated during DNA repair, there are disruptions of the double helical structure of the parental DNA molecule which are caused both by the DNA damage and by association of multiprotein complexes called repairisomes. These multicomponent DNA repair systems consist of the damaged DNA, and the direct or indirect associations of polymerases, ligases, topoisomerases, DNA helicases, endonucleases, glycosylases, methyltransferases, photolyases, structural proteins and other accessory peptides.
Both the replication and repair complexes employ DNA polymerases, which are able to synthesize new DNA strands in the 5'-3' direction from a template strand. While it is true that DNA replication requires many proteins in addition to the DNA polymerases, we may nonetheless think of the DNA synthesizing activities of these complexes in terms of DNA polymerase enzymes. Five major eukaryotic DNA polymerases (a, d, e, b and g) have been characterized to date (Kornberg and Baker 1992). The nuclear replication fork DNA polymerases a, d and e are active during DNA replication, which occurs during S-phase of the cell-cycle. The lagging strand is synthesized discontinuously because its DNA polymerase, a, must first polymerize a short piece of RNA to initiate the polymerization of DNA (Kaguni and Lehman 1988; Lehman and Kaguni 1989). Due to this, and because the parental template is methylated differently than the newly synthesized daughter strand of DNA, there is some amount of structural distinctness which is exploited by post-replication-repair. DNA polymerase e sometimes replaces DNA polymerase d in the leading strand DNA synthesis, and in conjunction with DNA helicase E performs displacement synthesis at the replication fork (Huang et al. 1993; Turchi et al. 1992). DNA polymerase e is also thought to be involved in excision repair (Wang et al. 1993). The nuclear DNA polymerase b plays a role in constitutive excision repair of some types of DNA damage, for instance single-strand nicks (Popanda and Thielmann 1992). The role of the mitochondrial polymerase g is presumably to replicate mitochondrial DNA (Kornberg and Baker 1992).
Unlike DNA polymerase g and b, the nuclear replication fork polymerases a, d and e, are inhibited by the drug aphidicolin (Spadari et al. 1982; Kornberg and Baker 1992).
Eukaryotic and prokaryotic DNA polymerases share the same fundamental type of synthetic activity, extending a DNA chain by adding nucleotides one at a time to the 3'-OH end. The precursor for DNA synthesis is a nucleoside triphosphate which loses its terminal two phosphate groups in the reaction. The choice of nucleotide added to the chain is dictated by base pairing with the template strand. All of the bacterial DNA polymerases, and some eukayotic DNA polymerases possess a 3'-5' exonucleolytic activity which provides a proofreading function. The 3'-5' exonucleolytic activity proceeds in the reverse direction from DNA synthesis, and is responsible in part for the remarkable fidelity of DNA replication. Esherichia coli is estimated to misincorporate only one in a million nucleotides during DNA replication (Glickman and Radman 1980).
By strictly adhering to sequential cell-cycle progression, cells adjust the frequency at which their cycles of DNA replication are initiated to match the rate of cell growth. Normally, the replication of DNA is confined to a part of the cell-cycle called S-phase. DNA synthesis (S-phase) is preceeded by a state called G1-phase, which often comprises a major part of the cycle (Lewin 1985).
When eukaryotic cells move out of the cell cycle and into quiescence during stationary phase arrest, they arrest predominantly in G0-phase, a subset or alternative of G1-phase (Herget et al. 1993; Short et al. 1993; Ansari et al. 1993; Pignolo et al. 1993; Doyle et al. 1993; Hass et al. 1993; Barlat et al. 1993; Prasad and Rosoff 1992; Kimchi 1992; Clarke et al. 1992; Del-Sal et al. 1992; Sell et al. 1992; Tsutsumi et al 1992; Hass et al. 1992; Evans et al 1992; Rose et al 1992; Bantel et al. 1992; Shaulsky et al. 1991; Oberg et al. 1991; Yonish-Rouach et al. 1991; Resnitzsky and Kimcki 1991; Steinmeyer et al 1990; Riddle and Lehtomaki 1981; Yen and Riddle 1979). This is one reason why G1-phase is often the most variable phase of the eukaryotic cell-cycle (Rossow et al. 1979).
DNA synthesis occurring normally in S-phase is called scheduled synthesis. During S-phase, replicative DNA polymerases utilize parental strands of DNA as templates to quickly and efficiently produce daughter strands. The periods of time preceeding and succeeding S-phase are called G1- and G2-phases to indicate that they represent gaps between DNA synthesis (S) and mitosis (M). After S-phase, the DNA of the cell has been doubled, allowing progression into G2-phase. During M-phase, the chromosomes condense, line up and separate. Nuclei reform after the duplicated chromosome have separated, then the cells divide by binary fission. DNA synthesis which occurrs outside of S-phase, i.e. DNA repair, is sometimes referred to as unscheduled DNA synthesis (UDS) (Davies et al. 1993; Furihata et al. 1993).
A "universal model" for the pathways regulating cell division has emerged from a large number of studies in a wide variety of eukaryotic organisms (Balter 1991; O'Farrell 1992). The embryonic or log phase somatic cell-cycle of many organisms is a rapid alternation of interphase and mitotic states. Driving the cell-cycle is an interplay between the cyclins and p34cdc2 kinase, and other cyclin-dependent kinases (Lewin 1990). Cyclin and p34cdc2 kinase form an active protein kinase complex, which was originally called MPF for maturation promoting factor in frog oocytes. MPF and related kinases regulate the action of many other enzymes to control the timing of DNA synthesis and division. The cyclins, as their name implies, are made and then degraded as the cell-cycle progresses. The p34cdc2 kinase is present throughout the entire cell-cycle in roughly constant amounts (McKinney and Heintz 1991).
The identification of cyclins and p34cdc2 kinase was a watershed in understanding the cell-cycle since these enzymes and their relatives play major roles in most of the events of the cell-cycle, including condensation of chromosomes and reorganization of the cellular structure during breakdown of the membranes surrounding the nucleus. In S. pombe, p34cdc2 kinase is required not only for the initiation of M-phase and the segregation of chromosomes into daughter cells during mitosis, but also for control over the onset of S-phase. This leads to two major control points for the cell-cycle, one involving S-phase and the other M-phase. Both these cell-cycle checkpoints involve p34cdc2 kinase in S. pombe. The phosphorylation of p34cdc2 kinase differs during S- or M-phases. For example, both tyrosine at position 15 and threonine at position 14, are dephosphorylated at the begining of mitosis (Nasheuer et al. 1991). In contrast, as the S-phase progresses, phosphorylation of a serine at position 277 decreases (Krek and Nigg 1991).
The interplay of cyclins with p34cdc2 kinase regulates the kinase activity throughout the cell-cycle. Families of cyclins have been identified in sea urchins, clams, starfish, Xenopus, Drosophilia, S. pombe, S. cerevesiae, and human cells (Murray and Kirschner 1989; Murray et al. 1989). The mRNA of cyclins undergoes periodic variation throughout the cell-cycle. Human PRAD1 is a cyclin that is overexpressed in some tumors. Mouse CYL1 is a G1-phase cyclin induced in macrophages in response to stimulation by the growth factor CSF1. The macrophage receptor for the CSF1 gene product, c-fms, is another oncogene, which may be activated constitutively to signal expression of the CYL1 genes. In yeast, the G1-phase cyclins are the CLN1, CLN2 and CLN4 proteins, which are activated at the level of transcription by the protein products of the SWI4 and SWI6 genes (North 1991).
The largest source of DNA damage is the mispairing of bases during replication, repair and recombination (Friedberg 1985). The error frequency for these processes can be as high as 10-1 to 10-2 in the newly synthesized DNA molecule. Fortunately, many of the same DNA polymerases which synthesize the DNA and allowed the misincorporation also have 3'-5' proofreading exonuclease activities, which reduce the error frequency to . Subsequently, several different types of interacting proteins further reduce the mismatch error frequency to with post-replicative mismatch corrections (Kornberg and Baker 1992).
Aside from its replicational fidelity, DNA is susceptible to a wide variety of chemical attack from agents produced in the environment and arising from metabolism. Estimates have been made for the rate at which different types of lesions form in human cells (Saul and Ames 1985). During one 24 h period, out of the six billion base pairs in the diploid genome of a human cell, 26,000 bases are lost from the backbone, 350 cytosines will deaminate to form uracil, 100,000 single-strand breaks will form, and the methyl group of S-adenosyl methionine will be transferred to form 84,000 7-methylguanines and 840 3-methyladenines. One hour of bright sunlight may cause up to 50,000 pyrimidine dimers (PD) in each cell near the surface of the skin. PD are a major photoproduct in DNA produced by UV radiation at approximately 254 nm (Protic-Sabljic and Kraemer 1986).
Depending upon the type of DNA damage, DNA replication may be directly inhibited , or misreading of the DNA template may occur, leading to heritable alterations or destruction of function (Moore and Strauss 1979; Setlow et al. 1963; Coulondre and Miller 1977; Loveless 1969).
DNA appears to be the only molecule which, when damaged, is actively repaired by the cell. Toward this goal, DNA repair mechanisms exist to eliminate altered or mismatched nucleotides. Many of these activities, including photolysis, dealkylation and nucleotide excision, are conserved across biological kingdoms. Photoreactivation in a wide variety of plants and animals, is catalyzed by enzymes called photolyases. The cleavage of pyrimidine dimers to form monomers is an example of repair of DNA damage by direct reversal of base damage. The energy of visible light (300-600 nm) is used by photolyases to break the cyclobutyl pyrimidine dimer ring in situ (Friedberg 1985).
Occasionally, DNA damage occurs by incorrect methylation of bases. Examples include 3-methyladenine, 7-methylguanine, O6-methyladenine, O6-methylguanine, O2-methylthymine, and 02-methylcytosine (Rydberg and Lindahl 1980). This type of DNA damage is repaired, in part, by DNA glycosylases which remove the damaged base, or by methyltransferases which may directly repair the damage (Lawley and Warren 1976). Some methyltransferase enzymes have been called suicidal because their acceptor protein is irreversibly inactivated upon alkylation. For instance, the O6-methylguanine methyltransferase removes the methyl group from O6-methylguanine and adds it to cysteine residue. This enzyme has been detected in Drosophilia, fish and mammals (Green and Deutsch 1983; Nakatsuru et al. 1987; Pegg et al. 1982).
A 3-methyladenine glycosylase enzyme has been isolated from mammals and this gene has been cloned from yeast (Chen et al. 1989). As a player in excision repair, this enzyme recognizes and removes 3-methyladenine from the DNA molecule (Lindahl 1982).
Excision repair removes (1)the damaged base, (2)the fragment containing an apurinic or apyrimidinic (AP) site, (3)the nucleotide lesion and neighboring region of DNA, or (4)an interstrand crosslink. Excision-repair systems that have been characterized from different organisms fall into two general pathways. Depending upon the pathway, excision repair systems make incisions either 3' or 5' to the damage site, then they remove the mispaired or damaged bases from DNA, synthesize new stretches of DNA, and ligate any remaining single-strand nicks (Friedberg 1985; Friedberg and Hanawalt 1981). Studies reveal that excision repair of UV damage in several eukaryotes is similiar to that found in E. coli. At least nine complementation groups have been found in yeast, eight in humans, and six in chinese hamster ovary cell lines (Bootsma et al. 1989; Cleaver and Kraemer 1989; Collins and Johnson 1987; Haynes and Kunze 1981).
The distinction between excision-repair pathways and other DNA repair pathways is sometimes vague. Some Drosophila mutants which are deficient in the ability to excise pyrimidine dimers are also deficient in meiotic recombination and are sensitive to ionizing radiation (Boyd and Setlow 1976). Some yeast mutants that are defective in the incision step of excision repair are also defective in certain types of mitotic recombination (Schiestl and Prakash 1990).
Uracil, normally a base in RNA, sometimes arises in DNA by the deamination of cytosine. If this damage is not repaired, GC->AT transitions may occur. To thwart this possibility, enzymes called uracil glycosylases have evolved which remove uracil from DNA. The genes for several uracil glycosylases have been cloned from organisms as diverse as D. discoideum, yeast, humans, and several mammalian viruses (Guyer and Deering unpublished observations; Mullaney et al. 1989; Olsen et al. 1989; Percival et al. 1989).
At times, adenine in DNA deaminates to form hypoxanthine. Hypoxanthine in DNA also occurs during semiconservative DNA synthesis following the incorporation of dIMP instead of dGMP (Thomas et al. 1978). A hypoxanthine DNA glycosylase activity is present in extracts of E. coli, bovine and human cells. This enzyme catalyzes the release of hypoxanthine from nitrous acid-treated DNA (Karran and Lindahl 1978; Karran and Lindahl 1980).
Occasionally, methylated purine bases exhibit opened imidazole rings. There are DNA glycosylases which exist in E. coli, bovine and human cells to remove this type of damage. These formamidopyrimidine DNA glycosylases have an absolute specificity for duplex DNA as the substrate (Chetsanga et al. 1981).
Often ionizing radiation oxidizes the 5,6 double bond of pyrimidines, resulting in the formation of hydroperoxides and thymine glycols, which further degrade to yield formylpyruvylurea, urea and N-substituted urea derivatives. Urea DNA glycosylases have been identified in E. coli that remove this type of damage (Breimer and Lindahl 1980).
These various DNA glycosylases have no other known function than to release damaged bases from DNA. Once this occurs, incision at the resulting apurinic or apyrimidinic (AP) sites must be completed by an AP endonuclease activity. Some DNA glycosylases have associated AP endonuclease activities and are able to carry out both base excision and DNA incision. In any event, these endonucleases incise DNA on either side of the AP site with a directly acting 5' endonuclease, or a DNA glycosylase acting in concert with a 3' AP endonuclease (Kornberg and Baker 1992).
Micrococcus luteus cells have a pyrimidine dimer DNA glycosylase which acts in concert with a 3' AP endonuclease to incise DNA 3' of the AP site. The T4 phage PD DNA glycosylase-AP endonuclease attacks both sides flanking the damage site at the 5' N-glycosyl and the 3' phosphodiester bonds (Friedberg et al. 1981). AP endonucleases have been purified and/or the gene cloned from D. discoideum, yeast, fruit flies, and cattle (Guyer and Deering in preparation; Doetsch et al. 1986; Kelley et al. 1989; Popoff et al. 1990).
Post replication repair systems have also been characterized. This form of DNA repair is thought of as the conversion of low molecular weight DNA replicated on a template into high molecular weight species (Lehman 1981; Resnick et al. 1981). These assays typically measure both the formation and filling of daughter strand gaps, and synthesis across lesions. The line between postreplication repair and homologous recombination is at times not well defined; for instance, the rad52 mutant in yeast is defective in homologous recombination and also impaired in postreplication repair (Prakash 1981). In several eukaryotic as well as some prokaryotic systems, the repair of double strand breaks depends on recombination using a homologous template (Ho 1975; Krasin and Hutchinson 1977; Resnick and Martin 1976). Some of the proteins responsible for homologous recombination, like DMC1, RAD51 and RAD57 of yeast and UvsX of T4 phage, are structural homologs of E. coli RecA. DMC1 has a meiosis specific function required for meiotic recombination and progress through meiosis. Homologous recombination requires proteins that promote homologous base pairing and strand exchange of DNA molecules (Kornberg and Baker 1992).
In yeast, both RAD51 and RAD57 are required for mitotic repair of DNA double strand breaks, and for full levels of meiotic recombination. Other recombination proteins have additional unique activities not common to the family as a whole, such as the ability of RecA to promote cleavage of certain repressor proteins situated on the DNA molecule. Through phylogenetic studies of nucleotide and amino-acid alignments, it is revealed that these different recombination proteins diverged from a common ancestor which existed prior to the divergence of prokaryotes and eukaryotes (Story et al. 1993).
Transcriptionally-linked DNA repair of pyrimidine dimers also occurs in both prokaryotes and eukaryotes (Hanawalt 1990). In this type of repair, dimers are removed from the genome overall but the transcribed strand of active genes is repaired faster than other regions. In fact, a paradigm for coupling of transcription and DNA repair has emerged from studies of E. coli (Selby and Sancar 1990; Selby and Sancar 1991; Selby et al. 1991; Selby and Sancar 1993; Buratowski 1993). These authors have reconstituted transcription and DNA repair in vitro and have shown that the protein encoded by the mfd (mutation frequency decline) gene, shown previously to be required for DNA repair, functions as a transcription-repair coupling factor (TRCF). TRCF is a helicase, a protein which unwinds nucleic acid duplexes. This transcription factor acts by recognizing and displacing RNA polymerases that have stalled at DNA lesions. TRCF not only makes the region accessible for DNA repair, it actively recruits the repair machinery by interacting with UvrA, a protein which recognizes DNA damage. That this (E. coli) TRCF has a helicase motif is very interesting when one keeps in mind that another (human) transcription initiation factor, TFIIH (alternatively called basic transcription factor 2 or BTF2), from the ERCC-3 gene, is also a DNA dependent ATPase and has helicase motifs. The ERCC-3 gene product is also a protein kinase, and has previously been linked to DNA repair. The human ERCC-3 gene is 66% identical in sequence to the product of the Drosophila haywire gene, mutations in which cause UV sensitivity, neural defects and sterility. Mutations in the ERCC-3 gene are found in humans afflicted with both xeroderma pigmentosum (the XP-B group), and Cockayne's syndrome (CS), two diseases with defective DNA repair phenotypes. It is possible that their neurological problems and sterility as well as UV sensitivity, result directly or indirectly from defective transcription. The S. cerevisiae SSL2 gene, otherwise known as RAD25, is 55% identical to the human ERCC-3 gene, and some SSL2 mutations confer UV sensitivity (Park et al. 1992).
Another connection between DNA replication, transcription and homologous recombination repair is found in topoisomerases. These enzymes are found in the nuclear matrix. During DNA replication they can allow strands of the parental DNA molecules to completely relax, bringing the relative linking number, or DL, to zero (Kim and Wang 1989). In yeast this is achieved by the products of the TOP1 and TOP3 genes (Holm et al. 1985; Kim and Wang 1989). For every 10.5 base pairs unwound by a helicase involved in the replication process, one positive supercoil must be removed. During synthesis of RNA or DNA, topoisomerases relieve the strain caused by unwinding as the polymerase complex translocates along the DNA. This is necessary because translocation forces positive superhelicity to accumulate ahead and negative superhelicity to accumulate behind. When the replication of the duplex is nearly complete, the final interlinks between the two daughter strands are converted into links between the duplex portion of the daughter molecules. Decatenation by a type II typoisomerase removes these remaining intertwines to allow the separation of the daughter molecules (Sundin and Varshausky 1981).
Topoisomerases are also found in the mitotic chromosome scaffolding which arise during G2- and M-phases. At this time of the cell-cycle, these enzymes are involved in homologous recombination and repair of double strand breaks. Since the initial strand invasion is favored by negative superhelicity, the introduction of negative supercoils makes it energetically more feasible for a third DNA strand, driven by a recombinase like E. coli RecA or yeast RAD51, to invade the duplex during homologous recombination. In yeast, increased recombination between repeated sequences can result from a topoisomerase deficiency (Kim and Wang 1989). Z-DNA formation occurs after inhibiting topoisomerase activity in the human cell line HL6 (Wittig et al. 1992). DNA that has a large amount of negative superhelicity changes from the B to the Z form and may undergo increased recombination (Boehm et al. 1989; Kmiec and Holloman 1986; Murphy and Stringer 1986).
DNA damage will cause an inhibition of replication (both initiation and elongation), cell division, and an increase in mutagenesis. The inhibition of replication is seen as an S-phase arrest, and the inhibition of cell division is seen as an M-phase arrest. The S-phase checkpoint arrest is thought to keep damaged DNA from being replicated until after it has been repaired. The M-phase checkpoint arrest helps keep mitotic cells with damaged DNA from dividing until after their damage is repaired (Weinert and Hartwell 1988; Hartwell and Weinert 1989; Hartwell 1992).
Different types of DNA damage are repaired at different places in the cell cycle. Some post-replication-repair occurs during S-phase. Recombination-repair (which has been referred to as a type of post-replication-repair), double-strand-break-repair, and homologous recombination occur during G2- and M-phases. Excision-repair of thymine dimers and other lesions occurs throughout the cell-cycle and during cell-cycle checkpoint arrests in S- and M-phases (Hartwell 1992; Kastan et al. 1992).
In yeast, the RAD9 gene is necessary for inhibition of M-phase in response to DNA damage (Weinert and Hartwell 1988). Mutations have been found in human, Drosophila, mouse, and D. discoideum (the radB strain) which confer radioresistant DNA synthesis. Cells with this phenotype are unable to arrest their DNA synthesis and cell division after DNA damage (Lavin and Schroeder 1988). Such defects may be responsible for cancer in higher organisms. Mutations in the human tumor suppressor gene, p53, eliminate the S-phase cell-cycle checkpoint and enhance the frequency of genomic rearrangements (Hartwell 1992). Although wild-type cells block the initiation and elongation of replication in the presence of single-strand DNA breaks, cells from ataxia telangiectasia (AT) patients continue to initiate new replicons and elongate DNA molecules (Painter et al. 1982). These cells lack certain cell-cycle checkpoints, thereby allowing each damaged DNA molecule to replicate and generate two sister chromatids with double-strand breaks. With checkpoint arrests, cells monitor genomic status and fully correct errors.
over half a century ago in the sylvan mountains of
D. discoideum is conveniently manipulated in the laboratory by growth on either solid agar plates or axenically in shaking liquid media. The organism may be grown as either a diploid or a haploid. Its genome is about a hundred times smaller than that of a mammalian cell. During the last decade, parasexual genetic techniques were developed which allow different strains to be crossed to form diploids, from which recombinant haploid segregants can be isolated (Welker and Deering 1976; Welker and Deering 1979b; Welker and Williams 1982; Bronner et al. 1992).
DNA repair in D. discoideum
This thesis focuses upon a striking characteristic of D. discoideum, namely its high resistance to DNA damaging agents (Deering 1968). The dose of ionizing radiation required to kill 90% of a population of D. discoideum amoebae is 5 times that of yeast, 20 times that for E. coli, and 700 times that of cultured human fibroblasts (Howard-Flanders et al. 1966a; Howard-Flanders et al. 1966b; Shiomi and Sato 1979; Game 1983).
This laboratory has characterized AP-endonuclease and uracil glycosylase activities from D. discoiedeum, and recently cloned both genes (Guyer and Deering 1985; Guyer et al. 1986; Guyer and Deering unpublished results). No photoreactivation has been seen with this organism, but otherwise it shares common characteristics with other organisms in its response to DNA damage caused by UV irradiation. In less than 5 minutes after UV irradiation, D. discoideum cells produce very large numbers of single-strand breaks in their DNA (Guialis and Deering 1976a; Guialis and Deering 1976b). It is possible that the initial nicking of the DNA is the result of a dimer specific endonuclease which cleaves the backbone chain on one side of a dimer. After the initial nicking, an excision repair pathway could then excise the dimer and rejoin the DNA. In the wild-type strains, cell division and DNA synthesis are temporarily arrested after UV (Kielman and Deering 1980). Homologous recombination between plasmids occurs at high frequencies (Katz and Ratner 1988) in contrast with homologous recombination of nuclear DNA, which occurs at very low frequencies (Welker and Williams 1982).
Fifteen mutants, designated rad, representing eleven complementation groups (radA, B, C, D, E, F, G, H, J, K, L) were initially selected on the basis of sensitivity to radiation and other forms of DNA damage. Only one, radC, has an obvious correlation between a mutant phenotype and a repair process. This mutant is sensitive to UV, but not ionizing radiation or other agents, and has a deficiency in its ability to incise DNA after UV (Welker and Deering 1979a).
Because wild-type D. discoideum cells and their derived rad mutants were not efficient at incorporating exogenous labeled thymidine into nuclear DNA, this laboratory produced a strain, HPS401, which possessed the ability to efficiently incorporate exogenous tritiated thymidine (3H-TdR) into nuclear DNA due to tmpA600 and tdrA600 mutations. The tmpA600 mutation causes a lack of thymidylate synthase activity, while the tdrA600 mutation results in enhanced transport of the nucleoside thymidine into chromosomal DNA (Podgorski and Deering 1984; Bronner et al. 1988).
This in turn led to the isolation of damage sensitive mutants from HPS401. Towards this goal, eighteen mutants were isolated according to their sensitivity to g-rays or 4-nitroquinoline-1-oxide (4NQO) (Bronner et al. 1992). These mutants were shown to belong to complementation groups radB, G, J, K and L, and all retain the tmpA600 and tdrA600 mutations. Even though the HPS401 strain is in reality a double mutant, both deficient in thymidylate synthase activity and efficient at incorporating exogenous 3H-TdR into chromosomal DNA, in this thesis (except where noted) the HPS401 strain is referred to as the wild-type because of its DNA repair proficency.
D. discoideum radB mutants
Previous work in this laboratory generated several D. discoideum DNA damage sensitive mutants which could efficiently incorporate exogenous 3H-TdR into nuclear DNA. One of these, radB, is sensitive to DNA damage from 254 nm ultraviolet (UV) light, but undergoes normal rates of single-strand nicking and rejoining (Khoury and Deering 1973; Guialis and Deering 1976b; Welker and Deering 1979a; Kielman and Deering 1980). After UV irradiation, the radB cells undergo radiation-resistant cell division and DNA synthesis. This phenotype is also found in the uvs-6 and mus-9 mutants of Neurospora crassa, the gt mutant of Drosophila, the mouse wst mutant, and cells cultured from humans afflicted with ataxia telangiectasia (Lavin and Schroeder 1988).
Cell-cycle coordination of DNA repair in D. discoideum
Historically, the D. discoideum cell-cycle literature has been ambiguous. Unlike the case for other eukaryotic cells, there are reports that D. discoideum cells have no G0/G1-phase, that the predominant cell-cycle phase is G2, and that cells arrest during stationary phase conditions in G2-phase (Leach and Ashworth 1972; Katz and Bourguignon 1974; Soll et al. 1976; Zada-Hames and Ashworth 1977; Zada-Hames and Ashworth 1978; Durston et al. 1984; Weijer et al. 1984a; Weijer et al. 1984b; McDonald and Durston 1984; Maeda 1986; Ohmori and Maeda 1987; Nellen and Saur 1988; Maeda 1988; Maeda et al. 1989; Weijer and Krefft 1989). Because these D. discoideum cell-cycle papers appear to be incongruent with the rest of the established body of eukaryotic cell-cycle literature, and since my preliminary experimental results indicated that axenic D. discoideum cells spend most of their time in G0/G1-phase, I needed to further characterize the cell-cycle before studying the cell-cycle coordination of DNA repair.
There were two main goals of this thesis work. The first goal was to examine where in the cell-cycle axenic D. discoideum cells spend most of their time. Did the major nuclear flow cytometric control peak seen in this work, and discussed elsewhere in the literature, really represent G2-phase or G0/G1-phase cells? Where are the nuclear flow cytometric profiles of S- and G2-/M-phase-arrested cells relative to their unarrested control? Where are the whole-cell size distribution profiles of S- and G2-/M-phase-arrested cells relative to their unarrested control? Do nuclear flow cytometric profiles concur with whole-cell size distribution profiles?
The second goal was to find out what happens to the D. discoideum cell-cycle after UV irradiation. Are there cell-cycle checkpoint arrests, and if so, how are they manifested? How are cell division and DNA replication involved with the coordination of DNA damage repair?
With these goals in mind, I employed two specific eukaryotic cell-cycle inhibitors and UV light to perturb the cell-cycle. The first inhibitor, CIPC, arrests D. discoideum cells in mitosis (White et al. 1981). The second inhibitor, aphidicolin, is a very specific inhibitor for the binding of dCTP to the eukaryotic replication fork DNA polymerases a, d and e, and therefore blocks the cell cycle in S-phase (Matherly et al. 1989; Sgorbati et al. 1991; Sheaff et al. 1991).
The responses of both wild-type (HPS401) and the DNA damage sensitive radB (HPS517 and/or HPS521) strains were examined by: survival of colony forming ability after UV, CIPC or aphidicolin; Coulter Counter ZM-1 measurements of cell division lags after exposure to UV irradiation, CIPC, or aphidicolin; in vivo measurements of DNA synthesis lags after exposure to aphidicolin or UV irradiation by pulse labeling for 30 min with 3H-TdR; alkaline sucrose gradient analysis of nicking and rejoining of single-strand nuclear DNA after UV irradiation; flow cytometric analysis of nuclei from CIPC- or aphidicolin-arrested cells; size distribution analysis of whole-cells treated with UV, CIPC or aphidicolin; CsCl gradients of whole-cell- and nuclear-lysate from cells pulsed 30 min with 3H-TdR.
This thesis is the story of what I saw.
Strains of Dictyostelium discoideum
are named to indicate ploidy and laboratory or institution of origin. Haploid strains constructed here at
Mutations are given a three letter acronym (rad for radiation, i.e. radB) and a letter or number or both to distinguish independently isolated mutations of a particular complementation group.
The gene or a genotype is written in italics (for instance, RADB), and a gene product or phenotype is written in plain type with the first letter of the acronym in uppercase for wild type (i.e. RadB).
Throughout this thesis the term n-DNA (nuclear DNA) means the DNA prepared from nuclei isolated from control, UV or drug treated whole cells.
In general, four D. discoideum haploid strains are referred to in this thesis: AX-2, HPS401, HPS517 and HPS521. The AX-2 strain was the cell line used most often by previous workers to examine the D. discoideum cell-cycle. The other three strains were constructed in the R.A. Deering Lab, and all are 3H-TdR uptake mutants (Hurley and Deering 1988; Bronner et al. 1992). The HPS401 strain was employed as the repair proficient wild type and was contrasted against the two independently isolated radB mutant strains HPS517 and HPS521. Spores were gathered from fruiting bodies on Escherichia coli B/r DUM agar plates and maintained as stocks at -70 °C in PBS with 15% glycerol.
Stocks of cells were grown as per published methods (Bronner et al. 1988; Hurley et al. 1989) at 23 °C on either DUM agar plates (per liter H2O: 10 g Difco Bacto Peptone, 2 g dextrose, 1.5 g KH2PO4, 0.4 g Na2HPO4, 15 g Difco Bacto agar) with an E. coli B/r lawn, or axenically with shaking at about 190 rpm in HL5 (per liter H2O: 14.3 g Difco Protease Peptone, 7.15 g Difco yeast extract, 15.4 g dextrose, 0.51 g Na2HPO4, 0.49 g KH2PO4, 0.1 g streptomycin sulfate and 0.2 MU of penicillin, pH 6.7) in a 23°C room on a New Brunswick G-10 shaker.
A 1215 model LKB Rackbeta Liquid Scintillation Counter was employed to measure 3H-TdR incorporation. Scintillation fluids used included Optiphase Hi Safe 3 (LKB) for aqueous samples, or Ecoscint O Filter Fluor (National Diagnostic) for samples dried on filters. Counts were normalized to the number of cells in the sample.
A Coulter Counter Model ZM-1 was used in conjunction with a Coulter Channelyzer 256 (kindly loaned by the Biotechnology Institute at PSU's Wartik Laboratory) throughout this study (with a 100 micron aperture) to determine both the number of cells per ml and the relative cell size distribution.
Coulter Counter ZM-1 settings include: current, 15.0; full scale, 10; polarity, auto; lower threshold, 10; upper threshold 99.9; alarm threshold, off; attenuation, 16; preset gain,1; count, corrected; data feed, manual; manometer select, 500.
Settings for the Channelyzer 256 include: channels, 64; edit, off; control, timer; time, 30-60 sec; count, 5000 in peak channel; x axis, volume; cal, no; size, 9.85 E; Kc, 10.672; TH units, 43.5;
Hard copies of data were gathered with a Fujitsu DX2300 printer, set at: RS232C serial output, 9600; the device mode, EPSON; screen dump, yes; channel data, yes; overlay mode, no; format, standard; analogue plot, yes.
Experiments measuring the percent survival of cells after irradiation were performed according to published procedures (Deering et al. 1970).
Various experiments were performed to contrast the responses of wild-type and mutant cells. Experimental cultures were derived from 20 ml of log-phase cells, at 2 x 106 cells/ml, which had been irradiated with UV at 23 °C in a fashion similiar to published procedures (Kielman and Deering 1980; Podgorski and Deering 1980).
Briefly, these procedures were as follows. Cells were centrifuged at 1,200 g for 5 min and the supernatant fluid discarded. The cell pellet was resuspended in 20 ml PBS (10 mM KCl, 10 mM NaCl, 16 mM Na2HPO4, 34 mM KH2PO4; pH 6.5) and shaken at 100 rpm in polystyrene petri dishes or multiwell plates of various sizes (Fisher) during UV irradiation. A 15W germicidal lamp (Westinghouse) at an intensity of 4.8 J/m2 sec was used for all irradiation experiments. The lamp was positioned approximately 21 cm above the bottom of the dish or plate. After appropriate exposure times, aliquots were removed, centrifuged at 1,200 g for 5 min and the supernatant fluid discarded. The cell pellet was resuspended in 20 ml HL5 and shaken as previously described for axenic growth conditions.
Isopropyl N-(-3-chlorophenyl) carbamate (CIPC) was made into 20 mg/ml stocks in 70% dimethylsulfoxide (DMSO). Both CIPC and DMSO were purchased from Sigma. Stocks were stored at -20 °C or -70 °C in small aliquots and not reused after thawing once. Cell division and cell size after addition of CIPC, as a function of time, were measured with a Coulter Counter Channelyzer. Survival assays were performed with increasing amounts of CIPC according to published procedures (White et al. 1981).
Aphidicolin (Sigma) was made into 5 mg/ml stocks in 70% ethanol. Stocks were stored at -20°C or -70°C in small aliquots and not reused after thawing once. Cell division and cell size distribution after addition of aphidicolin, as a function of time, were measured with a Coulter Counter and a Channelyzer 256. Survival assays after aphidicolin were performed in a fashion similiar to that described above for CIPC.
Nuclei were isolated by cell lysis in nuclear preparation buffer (30 mM HEPES, 10 mM magnesium acetate, 10% sucrose, 10 mM NaCl, 2% (v/v) Nonidet P40 (Sigma), pH 7.5 with NH4OH) (Hurley et al. 1989). All nuclear isolations were done as quickly as possible.
In a 4 °C room, 1 ml of axenic cells were centrifuged for 3 sec at 12,000 rpm, and all but ~100 ml of the supernatant fluid discarded. The cell pellet was vortexed for 3 sec and to the resuspended cells was added 1 ml PBS buffer. This mixture was vortexed for 30 sec, then centrifuged for 3 sec at 12,000 rpm. All but ~100 ml of the supernatant fluid was removed and the pellets vortexed for 3 sec. After addition of 1 ml ice-cold nuclear preparation buffer, cells were vortexed for 30 sec then incubated at 0-2 oC for 5 min. After this period, nuclei were centrifuged for 7 sec at 12,000 rpm. All but ~50 ml of the supernatant fluid was discarded and nuclei were resuspended with 30 sec of vortexing, diluted to 1 ml with more nuclear preparation buffer and vortexed for 30 sec. After a 5 min incubation, nuclei were centrifuged and resuspended as above to yield a total volume of ~50 ml. Resuspended nuclei were either used immediately for flow cytometry, assays measuring the uptake of 3H-TdR, or frozen at -70 °C for later use.
Thawed or fresh nuclei were lysed after addition of 10 ml of 20% sodium dodecyl sulfate (SDS, from SIGMA), 10 ml of 10 mg/ml proteinase K (SIGMA), 30 sec of vortexing at room temp, and incubation at 37 °C for 15 min. Lysed nuclei were then diluted with 1 ml of ice-cold 10% TCA (SIGMA), vortexed for 30 sec and placed at -20 °C for at least 1 hour to precipitate all DNA larger than about 10 bp. Samples were collected on Millipore HAWP filters and counted as previously described.
Nuclei from 1 ml of 2 x 106 cells/ml were kept at 0-2 °C in 400 ml of nuclear resuspension buffer, and treated with 10 ml of 1 mg/ml propidium iodide and 10 ml of 100 mg/ml RNase A (Sigma).
The tubes were vortexed for 7 sec,
set at 0-2°C for 10 minutes, then run through a EPICS 753 flow cytometer
Alkaline sucrose gradients were used to study the rejoining of single strand breaks in DNA after UV. Wild type HPS401 and the mutant strain HPS521, initially at 1 x 106 cells/ml, were labeled for one cell doubling in HL5 with 20 mCi 3H-TdR/ml HL5. For each experiment, 20 ml of labeled cells at 2 x 106 cells/ml were centrifuged, resuspended in PBS, and (not irradiated or) irradiated with 100 J/m2 UV light by methods previously described. The cells were then transferred to plastic centrifuge tubes, recentrifuged, resuspended in 20 ml fresh HL5 and shaken at 190 rpm. At specific times, 1 ml aliquots of cells were transferred to 0°C Eppendorf tubes. Nuclei were prepared from cells as described, except the pelleted nuclei were not resuspended until just before ultracentrifugation, and then gently resuspended with a micropipet.
Gradients were prepared with 5% and 25% (w/v) sucrose solutions in alkaline buffer (0.3 N NaOH, 0.9 M NaCl, 5 mM EDTA, pH 12.2) as per published methods (Khoury and Deering 1973). Linear gradients were produced at room temperature using a gradient maker (Buchler) to form 12.2 ml gradients simultaneously in 3 polyallomer tubes (Seton). The gradients were overlayered with 0.3 ml of 0.2 M EDTA in 2% sodium lauryl sarcosine just prior to the application of 100 ml to 200 ml of nuclei in nuclear resuspension buffer. Gradients were centrifuged concurrently in L2-65B and L8-80M ultracentrifuges with SW41 rotors (Beckman) at 20 °C and 35,000 rpm for 165 min, then 40 separate 0.3 ml fractions were collected into 5 ml scintillation vials (each of which had 0.3 ml of 1 M HCl). To each vial was added 4 ml of Optiphase Hi-Safe 3 Scintillation Fluid and, after vortexing for 30 sec, all samples were counted on the scintillation counter for 5 minutes.
In vivo DNA synthesis
Experiments measuring the uptake of 3H-TdR into DNA as a function of time were performed with the wild-type HPS401 and the radB strains HPS517 and HPS521. Many of these experiments lasted for at least one cell doubling time (8-16 h, depending on strain) and sometimes two.
These procedures were standardized such that 1 ml aliquots from 20 ml of 2 x 106 cells/ml after treatment with UV or aphidicolin were pulse labeled for 30 min. Pulse labelings of 1 ml aliquots of wild type and radB cells were performed by addition of 100 mCi 3H-TdR to cells in disposable 24 well plastic plates then shaking at 180 rpm for 30 min. Labeled cells were collected into Eppendorf tubes held at 0-2 °C then centrifuged 3 sec at 4 °C. All but a minimum amount of the supernatant liquid was removed, the pellet vortexed 3 sec, and 1 ml of 0-2 °C PBS was added. These were vortexed 3 sec, centrifuged 3 sec at 4 °C, and all but a minimum amount of the supernatant liquid was removed. Cells were either lysed immediately for analysis of whole-cell 3H-TdR incorporation or the nuclei isolated for analysis of nuclear 3H-TdR incorporation.
Whole-cell lysis was achieved by vortexing the washed cells with 1 ml of PBS, 10 ml of 100 mg/ml proteinase K (EM Science) and 10 ml of 20% Sarkosyl (Sigma) for 30 sec then incubation at 37 °C for 15 min. Samples were often frozen at -70 °C at this point for later use. If not frozen, samples were immediately modified by the addition of 1 ml of 10% TCA (Sigma) and placed at -20 °C for at least 1 h, after which samples were thawed and the precipitated DNA was collected by vacuum filtration through 25 mm HAWP filters (Millipore) prewet with 10% TCA. The precipitated DNA was washed once with 10 ml ice-cold 10% TCA then once with 5 ml of 95% ethanol. The filters were dried under IR lamps for five minutes then placed in scintillation vials, covered with Ecoscint O Filter Fluor, capped, vortexed 30 sec and counted for five minutes in the scintillation counter.
Nuclei were purified as described, diluted to 300 ml with PBS then lysed with 10 ml of 1 mg/ml proteinase K, 10 ml of 10% sarkosyl, vortexing for 30 sec and a 15 min incubation at 37 °C. To this nuclear DNA lysate was added 1 ml 0-2 °C 10% TCA and the mixture vortexed 30 sec then placed at -20 °C for 1 hour. The precipitated DNA was collected and counted as described above.
CsCl gradient analysis of uptake of 3H-TdR into DNA
The uptake of tritiated thymidine into 10% TCA precipitable DNA, for both whole cells and nuclei, over a short periods of time (30 min), was examined with a thymidine-incorporating strain (HPS401) and a non-thymidine incorporating strain (AX-2) according to methods described (Kielman and Deering 1980; Clark and Deering 1981).
Large (500 ml) flasks containing 75 ml of either HPS401 or AX2 axenically growing cells at 5 x 106 cells/ml were aliquoted into three 20 ml portions in smaller (250 ml) flasks. Each flask contained 2000 mCi 3H-TdR to label DNA, and was shaken during the course of the experiment.
After 30 min of labeling, one flask from each strain was harvested. Each flask was split into two 10 ml aliquots which were placed in 50 ml polystyrene centrifuge tubes (prewashed with PBS). The aliquots were then centrifuged at 1,200 g for five min at 23 °C then resuspended in 1 ml PBS and transferred to Eppindorf tubes held in ice-water.
To examine whole-cell incorporation of 3H-TdR, each of the tubes was centrifuged for 3 seconds at 12,000 rpm and 4°C, fluid was aspirated away, then the pellets were vortexed 3 sec and resuspended into 1 ml of 0.1 M EDTA, pH 8.0, and centrifuged as above. The pellets of 5 x 107 cells were vortexed and resuspended in 0.77 ml of 0.1 M EDTA (pH 8.0), 0.23 ml 20% sodium lauryl sarkosinate and 0.27 g CsCl. These components were mixed gently and held at 65 °C for 15 min. To 0.45 ml of lysate was added 0.40 ml of a 100 mg/ml solution of netropsin (Lederle Laboratories). The netropsin stock was in 10 mM tris and 1 mM EDTA, pH 8.4. The contents of the tubes were gently mixed, held at 23 °C for 30 min, then 5.35 ml of a stock solution of 56.34 wt% CsCl, density 1.71 g/ml was added. The final density here was 1.62 g/ml. The samples were topped off to 12.5 ml with 6.3 ml of a 1.62 g/ml CsCl solution and centrifuged in a 50Ti rotor at 30,000 rpm and 20 °C for 65 h. Thirty 0.42 ml fractions were collected and analyzed with the scintillation counter.
To examine nuclear uptake of 3H-TdR, the samples to be prepared for nuclei were centrifuged 3 seconds and all but ~100 ml of the supernatant fluid aspirated off. Cells were vortexed into a suspension and nuclei were prepared by methods stated previously. All but ~50 ml of the supernatant fluid covering the nuclear pellet was aspirated away and the nuclear pellets were vortexed for 60 seconds. Nuclei were further resuspended and analyzed on CsCl gradients as detailed above.
Exploiting the tmpA600 and tdrA600 mutations
Previous work in this laboratory resulted in several Dictyostelium discoideum cell lines which could efficiently incorporate exogenous 3H-TdR into nuclear DNA. This ability is due to the tmpA600 and tdrA600 mutations which cause a lack of thymidylate synthase activity and enhanced transport of the tritiated nucleoside thymidine (3H-TdR), respectively (Podgorski and Deering 1984; Bronner et al. 1988).
These mutations are located in both the parental (wild-type) HPS401 strain and the derived radB mutant strains HPS517 and HPS521 used in this work. HPS401 cells exhibited a doubling time of 8-12 h as compared with a 15-20 h doubling time for the radB cells. The basis for the extended cell-cycle in radB cells is unknown. Unlike the wild-type HPS401 cells, radB strains are sensitive to DNA damage from 254 nm ultraviolet (UV) light (Payez et al. 1972; Bronner et al. 1992).
Experiments exploiting these mutations were performed to measure the whole-cell and nuclear uptake of 3H-TdR in wild-type and mutant cells as a function of time. This approach was initially used to quantitate the whole-cell and nuclear uptake of 3H-TdR as shown in Figures 1 and 2, respectively. Thirty minute (min) labeling periods with 100 mCi 3H-TdR per ml of whole-cells were also used for uptake assays after UV, or aphidicolin (shown later).
Figure 1. Wild-type (HPS401) and radB (HPS517) whole-cell uptake of 3H-TdR as a function of short periods of time (min). These points are plotted on the time scale at the end of each 30, 80 or 90 min pulse.
Figure 2. Wild-type (HPS401) and radB (HPS517) nuclear uptake of 3H-TdR as a function of short periods of time (min). These points are plotted on the time scale at the end of each 30, 80 or 90 min pulse.
Long (10-12 hr) labeling periods, with 20 mCi 3H-TdR/ml of whole-cells, sufficiently labeled whole-cells and nuclei as shown in Figures 3 and 4, respectively, for use with alkaline sucrose gradient experiments. Alkaline sucrose gradient results are discussed later in this chapter.
Ultraviolet (UV) light was used as the radiation agent to damage cells, and therefore DNA. The survival of cells in exponentially growing cultures after treatment with increasing fluences of UV are shown in Figure 5. Wild-type cells survived high UV fluences, 50 J/m2, above which an approximately exponential decline in survival was seen.
In contrast, 6 J/m2 was sufficient to kill 98% of the irradiated radB cells. These methods and results are in agreement with published work for survival after UV (Payez et al. 1972; Bronner et al. 1992).
Lags in cell division and nuclear DNA synthesis occur in wild-type cells after UV irradiation for periods of time which increase with fluence (Kielman and Deering 1980; Hurley et al. 1989). After cultures re-enter exponential growth, cells are assumed to have completed nuclear DNA repair (Khoury and Deering 1973; Deering and Jensen 1973).
Figure 3. Wild-type (HPS401) and radB (HPS517 and HPS521) whole-cell uptake of 3H-TdR as a function of long periods of time (h). These points are plotted on the time scale at the end of each pulse.
Figure 4. Wild-type (HPS401) and radB (HPS517 and HPS521) nuclear uptake of 3H -TdR as a function of long periods of time (h). These points are plotted on the time scale at the end of each pulse.
Figure 5. Fractional survival of wild-type (HPS401) and radB (HPS517 and HPS521) cells after UV irradiation.
The HPS401 cells halted their cell division for periods of time increasing with fluence as seen in Figure 6. The HPS401 cells irradiated with 50-100 J/m2 lagged for 4-7 h relative to the control before re-entering log phase growth. The HPS401 cells displayed a dose effect with increasing fluence.
In contrast, Figure 7 shows that the radB cells irradiated with 50-100 J/m2 lagged about 3 h more than the control (which lagged 2 h), before resuming cell division. Thus the radB cells do lag in cell division after UV, about 3 h relative to the control, but they lag less than the wild-type cells. As seen in Figure 7, the radB cells did not display a dose effect with increasing fluence.
Nuclear DNA synthesis rates were measured during aphidicolin treatment or after UV irradiation by pulse labeling for 30 min with 100 mCi 3H-TdR/ml of cells, as a function of time after perturbation.
Examples of wild-type nuclear DNA synthesis after UV irradiation are shown in Figure 8. The HPS401 cells exhibited a lag in nuclear DNA synthesis for periods of time which increased with UV fluence. The HPS401 cells were more resistant to centrifugal manipulations than the radB cells, as evidenced by how quickly the (No UV) Control in Figure 8 resumed nuclear DNA synthesis. These control cells reached pre-UV, steady-state levels of DNA synthesis by about 5 h.
Figure 6. Division of wild-type (HPS401) cells after increasing amounts of UV irradiation.
Figure 7. Division of radB (HPS521) cells after increasing amounts of UV irradiation.
Figure 8. Wild-type (HPS401) DNA synthesis after UV irradiation as measured by uptake of 3H-TdR into nuclear, TCA precipitable material during 30 min pulses. These points are plotted on the time scale at the beginning of each 30 minute pulse.
The HPS401 cells in Figure 8 that were irradiated with 50 J/m2 showed almost a 2 h lag in nuclear DNA synthesis. Once DNA synthesis started at 2 h, the cells took 4 more h (6 total) to reach wild-type levels of nuclear DNA synthesis.
The HPS401 cells irradiated with 100 J/m2, as shown in Figure 8, displayed almost a 5 h lag in nuclear DNA synthesis, and eventually took more than 12 h to reach wild-type steady-state rates. The dose effect with increasing fluence, including DNA damage sensitive DNA synthesis in HPS401 cells, is consistent with published work (Hurley et al. 1989).
The radB no UV control in Figure 9 underwent a 2 h lag before resuming any increase in nuclear DNA synthesis. Once begun, DNA synthesis reached steady-state levels by about 7 h.
The radB cells irradiated with 50-100 J/m2 displayed the same 2 h lag before resuming nuclear DNA synthesis, Figure 9, but no additional lag relative to the control. Once DNA synthesis started, it continued for 5 h. During the next 8 h, a sharp decline in nuclear DNA synthesis was seen, leading to almost complete cessation by 15 h. The radB cells did not show a dose effect with increasing fluence. These results show that the radB cells have a reduced lag in nuclear DNA synthesis after UV as compared with wild-type HPS401. With small fluences (up to about 50 J/m2) the radB cells do not display a lag in their nuclear DNA synthesis after UV relative to the control.
Figure 9. radB (HPS521) DNA synthesis after UV irradiation as measured by uptake of 3H-TdR into nuclear, TCA precipitable material during 30 min pulses. These points are plotted on the time scale at the beginning of each 30 minute pulse.
This evidence of DNA damage resistant DNA synthesis in radB cells is consistent with published conclusions for the radB phenotype (Payez et al. 1972; Lavine and Schroeder 1988).
Alkaline sucrose gradients were used to study DNA repair as a function of time after UV irradiation. Wild type and mutant cells (20 ml cultures at 2 x 106/ml) were initially labeled with 20 mCi 3H-TdR/ml HL5 for 12 h. Nuclei were prepared from 1 ml aliquots of labeled cells taken from UV irradiated or unirradiated cultures over time, and run on alkaline sucrose gradients.
Figure 10 shows the alkaline sucrose gradient profiles of wild-type HPS401 cells after treatment with UV. Over many hours, the cells rebuild their DNA molecules toward the normal single-strand molecular weight represented by the "high-molecular-weight" peak at fractions 5-18.
The alkaline sucrose gradient results for radB cells are shown in Figure 11. This demonstrates that irradiated radB cells (like HPS401 cells), undergo rapid nicking of their nuclear DNA (within 5 minutes after UV irradiation) to produce the peak at fractions 27-37. As with wild-type (HPS401), over many hours the radB (HPS521) cells rebuild their DNA molecules back to the normal single-strand "high molecular weight" peak of fractions 5-18.
By integrating the area of the nuclear-main-band region under the curves of alkaline sucrose gradient fractions 5 to 18 (from
Figure 10. Nicking and rejoining of single-strand nuclear DNA of wild-type (HPS401) cells after UV irradiation (50 J/m2), as seen by alkaline sucrose gradient ultracentrifugation.
Figure 11. Nicking and rejoining of single-strand nuclear DNA of radB (HPS521) cells after UV irradiation (50 J/m2), as seen by alkaline sucrose gradient ultracentrifugation.
Figures 10 and 11), the data are summarized as shown in Figure 12 in terms of the amount of "high molecular weight" DNA as a function of time after UV irradiation. The radB (HPS521) strain exhibited a rate of DNA repair indistinguishable from that of the wild-type: 50% by 6 to 7 h. This is one paradox displayed by radB cells: good DNA repair versus poor survival.
To help understand how lags in cell division and nuclear DNA synthesis after UV irradiation (and the reduced lags in the radB) are related to the cell-cycle, I employed a flow cytometer to examine relative nuclear DNA content. Profiles such as that shown in Figure 13 were obtained when using vegetative cells as the source of nuclei. Figure 13 is typical of flow cytometric results obtained at the beginning of the study when the nuclei preparations were performed very gently, quickly and at 0 °C. When nuclei preparations were done less gently, the second (smaller) peak to the right of the first (larger peak) was seen to decrease. This effect became more noticable further into the study, as the diminishing second peak was found to represent lysis of G2-/M-phase nuclei during preparation. For instance the control of Figure 16 (presented later), from an experiment done four months after that for Figure 13, has only the first (larger) peak.
To identify the cell-cycle stages represented by the flow cytometric peaks in Figure 13, two specific inhibitors of the
Figure 12. Summary of the wild-type and radB nuclear DNA nicking and rejoining after UV light by integration of the area under the curve of the nuclear-main-band fractions #5 through #18 from data shown in Figures 10 and 11.
Figure 13. The distribution of wild-type (HPS401) nuclei as seen with a flow cytometer. This figure represents typical data taken during the beginning of the flow cytometric study when nuclei were more gently prepared.
eukaryotic cell-cycle were employed. The first inhibitor, CIPC, causes mitotic inhibition of D. discoideum cells (White et al. 1981). The second inhibitor, aphidicolin, is a specific inhibitor for the binding of dCTP to the eukaryotic replication fork DNA polymerases a, d and e, and therefore blocks the cell cycle in S-phase (Prasad et al. 1989; Matherly et al. 1989; Sgorbati et al. 1991; Sheaff et al. 1991; Baumstark-Khan 1992).
The concentration of CIPC used in subsequent experiments, 3.5 mg/ml of cells, was determined by interpolation of the data shown in Figure 14. Reversibility of a 48 h CIPC arrest in most cells was demonstrated when at least 65% of plated cells formed plaques on lawns of E. coli B/r. Plaque counts were performed as indicated previously under "Survival of colony forming ability."
Likewise, the concentration of aphidicolin used in subsequent experiments, 55 mg/ml of cells, was determined by interpolation of the data shown in Figure 15. Reversibility of the 48 h aphidicolin arrest was demonstrated when at least 95% of plated cells formed plaques on lawns of E. coli B/r. During these initial experiments, aphidicolin was found to be ineffective with the radB strains HPS517 and HPS521, as seen from results presented later.
CIPC caused a noticable decrease in cell division after about 1 h (Figure 14), while the effect of aphidicolin was not seen until after 2.5 h (Figure 15). These results are consistent with the consideration that an M-phase arrest from CIPC would immediately halt cell division, while the S-phase arrest from aphidicolin would
Figure 14. Wild-type (HPS401) cell division arrest with increasing amounts of CIPC. This family of curves was used to interpolate to the concentration of 3.5 mg CIPC per ml of cells used in later cell-cycle arrest experiments.
Figure 15. Wild-type (HPS401) cell division arrest with increasing amounts of aphidicolin. This family of curves was used to interpolate to the concentration of 55 mg aphidicolin per ml of cells used in later cell-cycle experiments.
not be seen in terms of halted cell division until after the unarrested G2- and M-phase cells pass through M-phase. Thus for a 7.5 h doubling, cells spend approximately 2.5 h, or (100(2.5/7.5))= 33% of their time in G2- and M-phases.
Once the optimal dosages for CIPC and aphidicolin were determined, nuclear flow cytometric profiles necessary for analysis of data (such as that shown previously in Figures 13) were obtained. The control of Figure 16 exhibits a profile with a (single) major peak centered around 35 Relative Nuclear DNA Content (RNDC) units. This peak represents the most populated phase of the cell-cycle. The following results suggest that this peak primarily represents G0/G1-phase cells.
When cells were arrested with aphidicolin, the peak shifted to the right to center around 50 RNDC units, Figure 16. This represents approximately 1.5 times the control DNA content of 35 RNDC units. This nuclear flow cytometric peak at 50 RNDC units was therefore derived from cells arrested with aphidicolin in S-phase. If the control peak of nuclei is representative of predominantly G0/G1-phase cells, then the S-phase arrest peak should be to the right of the control peak, and this is seen from nuclei (Figure 16) and whole-cells (presented later in Figure 20).
When cells were mitotically inhibited with CIPC, the flow cytometric curve for nuclei bifurcated. One peak, seen at 85 RNDC units, corresponds to the approximate M-phase nuclear DNA content because it is approximately twice the control DNA content of 35
Figure 16. Summary of nuclear flow cytometer profiles of wild-type (HPS401) cells arrested with CIPC (3.5 mg/ml), or aphidicolin (55 mg/ml) for approximately one cell doubling period. This figure represents typical data taken several months into the flow cytometeric study when nuclei were less gently prepared.
RNDC units. The other peak at 30 RNDC units corresponds to a G0/G1-phase nuclear DNA content.
Towards the end of the flow cytometric portion of this study, (December 3, 1991), the efficiency of the nuclei preparation procedure was estimated by isolating nuclei from a known number of HPS401 cells, placing them in a hemacytometer slide, then counting them using a microscope. An average of 75 nuclei were counted from every 100 cells. Hence, approximately (100-(100(75/100)))= 25% of the nuclei lysed during isolation. It is probable that this is from lysis of fragile G2-/M-phase nuclei during preparation, as mentioned previously. The data for Figure 13 were taken several months before that of Figure 16. It's likely that lysis of fragile G2-/M-phase nuclei during preparation is what caused the control of Figure 16 to show a smaller "upper" G2-/M-peak (fractions 70-100), relative to that shown in Figure 13 (fractions 90-140). Figure 13 is therefore is a more accurate representation of the relative cell-cycle proportions than the control of Figure 16.
As in all of these experiments, stock cells were from cultures recently started from spores. These cultures were never allowed to grow above 8 x 106 cells/ml during normal log phase growth. These cells did not appear to be multinucleate when stained with propidium iodide and viewed under a fluorescent microscope at 488 nm.
Microscopic observations of nuclei isolated from CIPC-treated cells showed that mitotic nuclei are larger and much more fragile than the interphase nuclei isolated from untreated cells. Under the microscope, mitotically inhibited nuclei were seen to be quite prone to bursting. Since the flow cytometer counted an absolute number of nuclei, and because the G0/G1-phase peak of the CIPC-treated cells may be only a part of the total if a significant number of G2-/M-phase nuclei are lysing during isolation, it is possible that the relative heights of the flow cytometric peaks shown in Figure 16 may be an artifact of handling.
The D. discoideum cell-cycle
After examining the positional relationships of the peaks in Figure 16, Figure 13 was interpreted as an approximation of the relative cell-cycle proportions of vegetative cells. Most cells, represented by fractions 20-80, are in G0/G1-phase. A smaller number of cells, represented by fractions 100-150, are in G2-/M-phase. Cells in S-phase are represented by the area between G0/G1-phase and G2-/M-phase peaks. The histogram displayed in Figure 13 is therefore composed of proportional, overlapping G0/G1-, S-, and G2-/M-phase peaks, respectively. This is as it should be if D. discoideum vegetative cells utilize a typical eukaryotic cell-cycle (Darzynkiewicz et al. 1979; Barfod and Barfod 1980a; Barfod and Barfod 1980b; Darzynkiewicz et al. 1981a; Darzynkiewicz et al. 1981b; Dosik et al. 1981; Darzynkiewicz et al. 1982; Darzynkiewicz and Traganos 1982; Darzynkiewicz 1983; Darzynkiewicz 1984; Herget et al. 1993).
An approximation of the phases of the Dictyostelium discoideum cell-cycle are presented in Table 1. These numbers were computer-derived using EPICS Cytologic (S-phase-fitting) Software: Program 1 and data shown in Figure 13: SFIT with constant S-phase; Program 2, SFIT with linear S-phase; Program 3, Multiple Broadened Rectangles with one compartment; Program 4, Multiple Broadened Rectangles with two compartments; Program 5, Multiple Broadened Rectangles with three compartments (EPICS Cytologic User Manual 1988).
G0/G1- S- G2-/M-Phase
Program 1 54 9 37
Program 2 39 29 32
Program 3 61 28 11
Program 4 49 27 24
Program 5 51 24 25
mean 51% 23% 26%
Table 1. Percentages of total cell-cycle. These analyses used different computer algorithms to define the G0/G1- and G2/M-peaks and approximate S-phase.
The S-fitting computer programs employed for Table 1 made no attempt to distinguish between G2- and M-phases. The published D. discoideum mitotic index suggests that M-phase represents only about 1.4% of the total vegetative cell-cycle (Williams 1980). Thus, a corrected approximation of the D. discoideum cell-cycle phases is: G1, 51%; S, 23%; G2, 24.6%; M, 1.4%.
This thesis assumes that the cell volume, 4/3pr3, should double during passage through the cell-cycle, from 4/3p(cell radius during G0/G1-phase)3 to twice this quantity during G2-/M-phase. During the course of one cell-cycle, the radius of each cell, (3V/4p)1/3, should increase by a factor of r2/r1 = (3V2/4p)1/3/(3V1/4p)1/3 = (3(2)/4p)1/3/(3(1)/4p)1/3 = 1.26.
The literature defines the major nuclear diameter peak of distribution of actively growing vegetative cells at about 3.4 mm (Bonner and Frascella 1953; Bonner et al. 1955). This thesis work suggests that this measurement (3.4 mm) represents the diameter of G0/G1-phase nuclei (see Conclusions for discusson). The radius of these nuclei would be about (3.4/2)= 1.7 mm. As with whole-cells, the nuclear volume, 4/3pr3, should double during passage through the cell-cycle, from 4/3p(1.7)3= 20.6 mm3 during G0/G1-phase, to 2(4/3p(1.7)3)= 41.2 mm3 in M-phase. The nuclear radius (like the whole-cell radius) would increase by a factor of 1.26 during the course of one cell-cycle. Therefore it is calculated that during the course of one cell-cycle, the nuclear diameter should increase from 3.4 mm during G1-phase, to about 3.4(1.26)= 4.3 mm in M-phase.
It was found that size distributions from whole-cells processed with the Channelyzer 256 were analogous to the relative distributions of nuclear DNA content obtained from nuclei run through the flow cytometer. One slight difference in settings between the flow cytometer and the channelyzer, which was not noticed until after I completed the flow cytometric section of this work, was that the flow cytometer counted a fixed number of events during a variable amount of time, while the channelyzer counted a variable number of events during a fixed amount of time. This is why the flow cytometer control peak is the same approximate size as the drug arrested profiles (Figure 16) while the Channelyzer results show a much larger control peak relative to the two drug arrested samples (typical examples of this effect are shown later in Figures 20 and 21). The incongruence is noted but does not detract from the positional relationships of the peaks to each other in a significant fashion.
Wild-type (HPS401) cell division
during treatment with CIPC or aphidicolin is shown in Figure 17. The cell size distribution of the first
control, Control 1, as a function of time, is shown in Figure 18. This control from whole-cells (Figure 18) and
the control from isolated nuclei (Figures 16) both display a similiar
distribution. While the control
Figure 17. Wild-type (HPS401) cell division arrests during aphidicolin or CIPC treatment.
Figure 18. Size distribution profiles of wild-type (HPS401) control cells (Control 1) as a function of time during log-phase growth.
Another control examined cell size distribution during a 3.3 h lag in cell division caused by a low speed centrifugation. This second control (typical data are shown in Figure 19) was important in interpreting the results from all of the UV irradiation or drug release experiments. For many years, our laboratory has referred to this type of lag as a "buffer readaptation lag." The peak for the second control centered around 11.5 RV units and, after the lag in division brought on by centrifugations and resuspensions, its height increased as a function of time. Therefore, the horizontal position of the peak was not affected after centrifugation, just the height, which transiently decreased during the 3.3 h lag in cell division.
The size distributions for aphidicolin treated HPS401 cells are shown in Figure 20. By comparison to control profiles with peaks located at 11.5 RV units in Figures 18 and 19, Figure 20 shows S-phase arrested cells continuing to increase in size as a function of time during treatment with aphidicolin to between 15-23 RV units. This is presumably because wild-type cells continue to produce mRNA and proteins despite inhibition of their DNA synthesis.
The distributions of cell size during treatment with aphidicolin for HPS401 seen in Figure 20 are similar to other eukaryotic distributions of cell size during treatment with aphidicolin (Matherly et al. 1989; Sgorbati et al. 1991; Hoffmann et al. 1991). The patterns of size distributions for whole cells during treatment with CIPC, Figure 21, are analogous to the patterns of nuclear DNA content during CIPC treatment shown previously (in
Figure 19. Size distribution profiles of wild-type (HPS401) control cells (Control 2) that were centrifuged 600 rpm x 3 min, resuspended in PBS, recentrifuged and resuspended in HL5.
Figure 20. Size distributions of wild-type (HPS401) cells during treatment with aphidicolin (55 mg/ml).
Figure 21. Size distributions of wild-type (HPS401) cells during treatment with CIPC (3.5 mg/ml).
Figure 16). During the first 12 h of CIPC treatment, the height of the normal distribution peak at 11.5 RV units (which corresponds to the population of cells seen previously in Figure 16 at 28 RNDC units) decreases. The peak which arises at 28 RV units most likely corresponds to the M-phase population of cells seen previously (in Figure 16 at 85 RNDC units). Comparisons of the peak of Control 1 shown in Figure 18, to the CIPC arrest peaks at 14 and 28 RV units in Figure 21 supports the hypothesis that the peak of the normal distribution of the control contains (primarily) G0/G1-phase cells. A summary of cell size distributions after arrest with drugs is shown in Figure 22.
Wild-type (HPS401) cells were held in an S-phase arrest for 12 h with aphidicolin. The arrest was terminated at 12 h after which cells were centrifuged at 1,200 g for 5 min, resuspended in 20 ml PBS, gently shaken, then recentrifuged and resuspended in 20 ml HL5. The aphidicoin-release cell division data are shown in Figure 23. These results show that cells undergo some aspects of synchronous division after release from aphidicolin arrest.
This set of experiments measured the size distributions of HPS401 cells at various times after irradiation with UV light. The initial period of time corresponding to the lag in cell division after
Figure 22. Summary of size distributions of wild-type (HPS401) cells after CIPC (3.5 mg/ml) or aphidicolin (55 mg/ml) treatment.
Figure 23. Wild-type (HPS401) cell division during and after release from arrest with aphidicolin (55 mg /ml).
UV, 0-6.5 h, is shown in Figure 24. It is possible that the lag in cell division is a result of a failure to progress to M-phase due to arrest in S- and/or G2-phases. It is also possible that the lag in cell division is a result of a failure to progress through M-phase due to an arrest in M-phase. Figure 24 represents a combination of peaks from cells arresting as a function of time after UV irradiation. In this figure, the peaks of the distributions move from around 9 RV units at 0 h, down and to the right to 11-17 RV units through the first 6 h. This suggests a mixture of S- and G2-/M-phase arrests. These results are paralleled by results of published work connecting DNA damage to S- and M-phase cell-cycle checkpoint arrests in yeast (Hartwell 1992). Since Figure 12 indicates that 50% of the nicks are repaired by 6 to 7 h, it may be assumed that a great deal of DNA repair occurred during the period illustrated by the cell size distributions shown in Figure 24.
Partially synchronized progression through G2- and M-phases as a function of time after UV is described in Figure 25. This figure shows that the peaks moved left to smaller size from around 18 to 12 RV units as cells exited from the lag in division. Once the peaks centered around 12 RV units at 8.5 h, G0/G1-phase entry was evidenced by the increase in peak heights which continued through approximately one cell doubling. At 12.5 h a cell doubling was reached and the curve shifted to the right from 12 RV units indicating the end of G0/G1-phase.
Figure 24. Wild-type (HPS401) cell size distributions as a function of time during the first 6.25 h after UV irradiation.
Figure 25. Size distributions of wild-type (HPS401) cells during the period between 7.5 h and 12.5 h after UV irradiation.
Another synchronized passage through G2-, M-, and more G0/G1-phase entry as a function of time after UV is shown in Figure 26. From 13.5-15 h the peaks shifted through 13 RV units to 15 RV units, suggesting that the cells had progressed through S-, G2- and M-phases. At the end of 15 h, some G0/G1-phase entry is displayed by the small peak seen emerging at 11 RV units. By 16.75 h, the profile has bifurcated. The "upper" peak, at about 17 RV units, corresponds to G2-/M-phase cells while the "lower" peak, centered at about 10 RV units, corresponds to passage into G0/G1-phase. By 19 h, the peak at 10 RV units is dominant.
radB size distribution during CIPC or aphidicolin
Cell-cycle arrests were examined for radB (HPS517 and HPS521) strains during treatment with either 3.5 mg CIPC/ml of HL5, or 55 mg aphidicolin/ml HL5. Growth curve results are summarized in Figure 27. Through comparisons with the controls it is seen that both strains of radB treated with aphidicolin underwent control rates of exponential cell division for more than 26 h. Both had an 18 h doubling time. The radB cells treated with CIPC remained in a halted cell division condition throughout the entire 26 h.
Cell size distribution studies were performed on the radB untreated control cells with the channelyzer. The results are shown in Figure 28. These results are analogous with previously shown control profiles.
Figure 26. Size distributions of wild-type (HPS401) cells during the period between 13.5 h and 19.5 h after UV irradiation.
Figure 27. Cell division in two independently isolated radB mutants (HPS517 and HPS521) after treatment with aphidicolin or CIPC.
Figure 28. Cell size distributions of radB (HPS517) control cells as a function of time.
Size distribution analysis after aphidicolin treatment indicates that radB (HPS517 and HPS521) strains continue cell-cycle progression with increasing peak heights centering between 10-15 RV units, as shown in Figures 29 and 30. The radB cells, although resistant to aphidicolin treatment, did become slightly smaller as a function of time during treatment with aphidicolin. As will be discussed later, this lack of S-phase arrest after aphidicolin could be related to the lack of arrest after UV irradiation (Figures 36 and 37).
Both radB (HPS517 and HPS521) strains mitotically arrested during CIPC treatment (Figures 31 and 32). These figures show that the major control peaks, at 14-16 RV units, decreased in height and shifted over to the right to about 34 RV units. From Figure 31, the peak at 16 RV units represents G0/G1-phase cells, and the peak at 32 RV units, G2-/M-phase cells. Likewise for Figure 32, the peak at 14 RV units represents G0/G1-phase cells, and the peak at 29 RV units, G2-/M-phase cells.
As measured by these methods, the radB cells undergo the normal arrest response after CIPC but are resistant to treatment with aphidicolin.
radB size distribution after UV irradiation
More examples of the reduced lag in cell division after UV for the radB (HPS517 and HPS521) mutants are shown in Figure 33 from another radB post-UV cell size distribution experiment. The unirradiated radB controls underwent a cell division lag until about
Figure 29. Size distributions of radB (HPS517) cells as a function of time after addition of aphidicolin (55 mg/ml HL5).
Figure 30. Size distributions of radB (HPS521) cells as a function of time after addition of aphidicolin (55 mg/ml HL5).
Figure 31. Size distributions of radB (HPS517) cells arresting as a function of time during treatment with CIPC (3.5 mg/ml).
Figure 32. Size distributions of radB (HPS521) cells arresting as a function of time during CIPC treatment (3.5 mg/ml).
Figure 33. Cell division of radB (HPS517 and HPS521) strains as a function of time after UV irradiation (50 J/m2).
3.5 h, after which time they began exponential growth. The lag in the division of the unirradiated control cells is due to experimental manipulations, including centrifugations and resuspensions. Both radB strains irradiated with UV (50 J/m2) showed cell division lags until about 5.5 h, after which time they began exponential cell division until 10 h. Thus, the irradiated radB strains lagged about (5.5 - 3.5 h), or 2 h longer than the unirradiated radB controls.
The cell size distributions for these radB control cultures are shown in Figures 34 and 35. Profiles of increasing height at 11 RV units are seen after a 3-4 h lag. These results, showing increasing numbers of cells of the same size distributions, are consistent with control profiles described previously.
Figures 36 and 37 show that peak height increases transiently at 11 RV units as both irradiated radB strains exhibited an increase
in cell size followed by a decrease. These results should be contrasted with the post irradiation S- and G2-/M-phase arrests of HPS401 described in Figures 24, 25, and 26, and compared with the radB growth after aphidicolin shown in Figures 29 and 30. The radB cells do not arrest in S- or G2-/M-phases after UV irradiation and they do not arrest in S-phase during aphidicolin treatment. Figure 38, which is derived from Figures 24, 25, 26, and 36, describes changes in the position of the peak of the distribution as a function of time after UV irradiation for both wild-type (HPS401) and radB (HPS517) cells. In Figure 38, there are two wild-type curves shown, one representing an "upper" peak and the second, a
Figure 34. Size distributions of radB (HPS517) control cells which had been centrifuged, resuspended in PBS, recentrifuged and resuspended in HL5.
Figure 35. Size distributions of radB (HPS521) control cells which were centrifuged, resuspended in PBS, recentrifuged and resuspended in HL5.
Figure 36. Size distributions of radB (HPS517) cells which were centrifuged, resuspended in PBS, UV irradiated with 50 J/m2, recentrifuged and resuspended in HL5.
Figure 37. Size distributions of radB (HPS521) cells which were centrifuged, resuspended in PBS, UV irradiated with 50 J/m2, recentrifuged and resuspended in HL5.
Figure 38. Cell volume at the peak of the size distribution(s). These data were taken from Figures 24, 25, 26 and 36.
"lower" peak. These results suggest that cells undergo division during the period when two different peaks are visible in the same timepoint, for instance at 8.5 h in Figures 25, or around 16.75 h in Figure 26. During division, the "upper" peak (to the right) represents M-phase cells, and the "lower" peak, G0/G1-phase cells. From Figure 38, the wild-type lag response included an increase in relative volume at the peak of the distribution, which lasted almost 7.5 h. This was 4.2 h longer than the wild-type control, which lagged 3.3 h (Figure 19). After 7.5 h the wild-type cells came out of the lag and the relative volume at the peak of the distribution decreased. The resulting cell division continued for 6 more h and caused an approximate doubling in the number of cells. Between 13.5-15 h after UV irradiation another increase is seen in relative volume at the peak of the distribution. Between 15-19 h after UV irradiation the relative volume at the peak of the distribution decreased. The wild-type peak volume increase-decrease-increase-decrease response shown in Figure 38 suggests that some synchronized cell division occurred after the UV irradiation lag. This is not too surprising since the total lag after UV approximated a cell-cycle period.
In contrast, the radB response shown in Figure 38 includes an increase in relative volume at the peak of the distribution which lasted almost 5.5 h. This was 2 h longer than the radB control, which lagged only 3.5 h (Figure 33). After 5.5 h the radB cells came out of the lag and the relative volume at the peak of the distribution decreased. Cell division continued almost 5 more hours and caused an approximate doubling in the number of cells. Therefore, radB cells do undergo a lag in their cell division after UV irradiation, but they lag less than wild-type cells: 2 h as compared to 4.2 h.
Aphidicolin inhibits the nuclear replication fork DNA polymerases a, d and e, but not the mitochondrial DNA polymerase g, nor the constitutive nuclear DNA repair polymerase b (Kornberg and Baker 1992). The experiments presented next describe overall synthesis of nuclear DNA after addition of aphidicolin to the culture using uptake of 3H-TdR into DNA. The purpose of this experiment was to further characterize the response of wild-type and/or radB cells to the drug aphidicolin. If the radB cells are replication fork DNA polymerase mutants, they might be resistant to aphidicolin (Kornberg and Baker 1992).
Data showing cell division for HPS401 and HPS517 during treatment with 55 mg/ml aphidicolin are presented in Figure 39.
The wild-type (HPS401) rates of synthesis of nuclear DNA during aphidicolin treatment are summarized in Figure 40 in terms of fraction of control cpm incorporated during a 30 min pulse of label, for three aphidicolin experiments. During 1.5-5.5 h of exposure to aphidicolin, cultures of HPS401 cells decrease their nuclear DNA synthesis 30-70% relative to that of the control.
Figure 39. Cell division in wild-type (HPS401) and radB (HPS517) cells as a function of time during aphidicolin during one of the three in vivo nuclear DNA synthesis experiment shown in Figures 40 and 41.
Figure 40. Fraction of wild-type (HPS401) control 3H-TdR incorporation as a function of time during aphidicolin, (55 mg/ml). These points are plotted on the time scale at the beginning of each 30 min pulse.
Relative amounts of nuclear DNA synthesis for a radB (HPS517) strain during aphidicolin treatment are presented in Figure 41, from 3 experiments. These results indicate that in the presence of aphidicolin, the radB cells do not significantly decrease their nuclear DNA synthesis relative to the control.
CsCl gradient analysis of 3H-TdR uptake
This analysis of the D. discoideum cell-cycle, indicating that most of the cells in a randomly growing axenic population are in G0/G1-phase is in contrast to previous reports that most cells, including stationary phase cells, are in G2-phase. However, many of the previous conclusions were based upon studies of uptake of 3H-TdR into the DNA of D. discoideum, and the assumption that this label is incorporated into nuclear chromosomal DNA (Leach and Ashworth 1972; Zada-Hames and Ashworth 1977; Zada-Hames and Ashworth 1978; Weijer et al. 1984a; Weijer et al. 1984b; McDonald and Durston 1984; Durston 1984; Sharp and Watts 1985; Maeda et al. 1986; McDonald 1986; Aerts et al. 1985; Ohmori and Maeda 1987; Nellen and Saur 1988; Maeda et al. 1989; Weijer and Krefft 1989). The following results suggest that this may not be the case.
AX-2 cells were chosen to compare with the HPS401 cells because AX-2 was the strain used in the majority of previous labeling experiments characterizing the D. discoideum cell-cycle (Leach and Ashworth 1972; Weijer and Krefft 1989).
Figure 41. Fraction of radB (HPS517) control 3H-TdR incorporation as a function of time during aphidicolin, (55 mg/ml). These points are plotted on the time scale at the beginning of each 30 min pulse.
Figures 42, 43 and 44 describe the uptake of 3H-TdR into both whole-cell and nuclear DNA for HPS401 and/or the non-thymidine incorporating strain AX-2. Figure 42 shows that after 55 min of labeling with 100 mCi 3H-TdR/ml, the HPS401 strain incorporated almost 5,000 cpm/106 nuclei, while the AX-2 incorporated 43 cpm/106 nuclei.
CsCl profiles of 30 min pulse labeled HPS401 and AX-2 cells are shown in Figure 43. These profiles consist of nuclear main band DNA peaks at fractions 31-37, and nuclear ribosomal and mitochondrial DNA peaks at fractions 25-30. It is seen from Figure 43 that after a 30 min pulse HPS401 cells incorporated more 3H-TdR nuclear main band chromosomal DNA than mitochondrial and nuclear ribosomal DNA: 3885 cpm>1390 cpm. The AX-2 cells did not incorporate enough 3H-TdR to register distinct peaks on the graph shown in Figure 43, therefore the vertical scale had to be expanded. Figure 44 shows that after a 30 min pulse AX-2 cells incorporated more 3H-TdR into mitochondrial and nuclear ribosomal DNA than main band chromosomal DNA: 65 cpm > 32 cpm.
One way or another, all previously published D. discoideum cell-cycle papers have based their conclusions upon autoradiographic analysis of mitotic nuclei isolated from AX-2 cells
which had been pulse labeled with 3H-TdR. These results show that the published labeling conditions produced an average AX-2 nuclear incorporation of about ((32+43)/2), or about 38 cpm per 106 nuclei. In theory this should produce about 102 disintegrations per minute
Figure 42. Uptake of 3H-TdR into TCA precipitable material for AX-2 and HPS401 cells. These points are plotted on the time scale at the end of the 15 and 55 min pulses.
Figure 43. CsCl profiles of whole-cell and nuclear lysate
from AX-2 and HPS401 cells after a 30 min pulse of 3H-TdR. This figure shows the uptake of 3H-TdR into nuclear chromosomal DNA (Fractions 31-37), and nuclear ribosomal and mitochondrial DNAs (Fractions 25-30).
Figure 44. Same as Figure 43 but expanded to detail the AX-2 uptake of 3H-TdR into nuclear chromosomal (fractions 31-37), nuclear ribosomal and mitochondrial DNA (fractions 26-30).
per 106 nuclei. Optimally, this should be about 1 disintegration per week per nucleus. In reality, 50% of these disintegrations will go upwards and not be counted on the autoradiograph. The remaining 50% which goes down towards the autoradiograph will be counted with probably no more than 40% efficiency, resulting in less than 0.2 disintegrations per week per nucleus. If the autoradiographs are left exposed for an average of 3 weeks, then one would expect to see less than 1 exposed grain per nucleus.
Therefore, it seems that the amount of 3H-TdR incorporated into AX-2 nuclear chromosomal DNA during a 30 min pulse might not be enough to produce sufficient disintegrations per week per nuclei on an autoradiograph for accurate analysis of the cell-cycle. This exposes the opportunity for a number of published D. discoideum cell-cycle experiments to be redone, possibly with the 3H-TdR-incorporating strains HPS401, HPS517 and HPS521. A critical few of these experiments are mentioned later in the Conclusions.
It was confirmed that the major aspects of the phenotype of the thymidine requiring D. discoideum radB cells isolated by Bronner and used in this work are the same as the non-thymidine requiring radB strains characterized earlier. These include (1) damage-resistant cell division and DNA synthesis, i.e., the "impatient" phenotype, (2) same rate of nicking and rejoining of single-strand breaks after UV induced DNA damage, (3) reduced growth rate relative to the parental non-UV sensitive strain, and (4) the cells undergo transient early DNA synthesis after UV before repair is complete. This replicated damaged DNA will not support another round of division or DNA synthesis.
In a non-synchronous, exponentially growing population, about 59% of the wild-type D. discoideum (HPS401) cells are in G0/G1-phase, (calculations discussed shortly). This contrasts with literature reports suggesting that most cells are in G2-phase and that G0/G1-phase is very short or nonexistent (Leach and Ashworth 1972; Zada-Hames and Ashworth 1977; Zada-Hames and Ashworth 1978; Weijer et al. 1984a; Weijer et al. 1984b; McDonald and Durston 1984; Durston 1984; Sharp and Watts 1985; Maeda et al. 1986; McDonald 1986; Aerts et al. 1985; Ohmori and Maeda 1987; Nellen and Saur 1988; Maeda et al. 1989; Weijer and Krefft 1989). Evidence for this conclusion comes from the observation that vegetative cells have a nuclear distribution consisting primarily of a "lower" peak of G0/G1-phase nuclei, and to a smaller extent an "upper" peak of G2-/M-phase nuclei (see Figure 13). The G0/G1-phase nuclei (or whole-cells) are thought to be about one-half of the volume (or DNA content), and approximately (100(1/1.26))= 79% the diameter of G2-/M-phase nuclei (or whole-cells). Thus the "universal model" for regulation of eukaryotic cell division prevails: D. discoideum has a typical eukaryotic cell-cycle (Loskutoff and Paul 1978; Darzynkiewicz et al. 1979; Barfod and Barfod 1980a; Barfod and Barfod 1980b; Darzynkiewicz et al. 1981a; Darzynkiewicz et al. 1981b; Dosik et al. 1981; Yancheva and Djondjurov 1982; Darzynkiewicz et al. 1982; Darzynkiewicz and Traganos 1982; Darzynkiewicz 1983; Darzynkiewicz 1984; Drewinko 1984; Luk et al. 1985; Herget 1993).
After UV induced DNA damage, wild-type (HPS401) cells pause in S- and G2-/M-phases for periods of time which increase with fluence, in contrast with the radB cells. This is similiar to what other eukaryotic systems do after DNA damage. For instance, human, Chinese hamster ovary, and yeast cells transiently arrest in S-, and/or G2-/M-phases after DNA damage (Weinert and Hartwell 1988; Musk et al. 1988; Sorenson and Eastman 1988; Downes et al. 1990; Schiano et al. 1991; Jimenez et al. 1992; Bennett et al. 1993; Kessis et al. 1993).
All cells tested were seen to be sensitive to mitotic inhibition by CIPC. Both whole-cells and isolated nuclei displayed a bifurcated distribution during CIPC treatment, whereby a first "lower" arrest peak was seen to correspond to G0/G1-phase arrest, and the second "upper" peak, G2-/M-phase arrest.
Aphidicolin blocks wild-type DNA synthesis but does not block radB DNA synthesis, or is at least much less effective at it. While it is possible that radB cells could possess high dCTP pools and/or low uptake of aphidicolin and thus appear unaffected by the drug, these results suggest that the defect in radB which leads to replication over and past DNA damage, rather than blockage by UV-induced lesions, may be due to an altered replication fork DNA polymerase which also confers insensitivity to aphidicolin. The D. discoideum radB situation could be analogous to that of the mutant herpes virus whose DNA polymerase has altered substrate recognition and resistance to aphidicolin (Hall and Woodward 1989; Hall et al. 1989).
In light of these thesis results, which basically reinterpreted the D. discoideum cell-cycle, I was given a mandate by my Ph.D. Thesis Committee to reinterpret related, previously published, D. discoideum cell-cycle literature. Towards this objective I have focused upon two other previously published conclusions. The first concerns the cell-cycle position during stationary phase growth conditions. The second concerns the relationship between cell-cycle position and prestalk and prespore differentiation.
Normally, if cultured on a solid surface, D. discoideum cells aggregate and form a slug when they run out of food. The slug then proceeds through development and forms a fruiting body. However, when vegetative cells cultured axenically in liquid media run out of food, they can not undergo development. When vegetative cells grow exponentially in a limited amount of liquid media, dividing up to about 5 x 106 cells/ml and then finially stopping at about 1 to 2 x 107 cells/ml, the cells are then said to be in a stationary phase arrest (Durston et al. 1984). The cell-cycle position of stationary phase D. discoideum cells has been reported in the literature as being predominantly in G2-phase (Soll et al. 1976; Zada-Hames 1977; Zada-Hames 1978; Durston et al. 1984). This reinterpretation hypothesizes that stationary phase D. discoideum cells arrest mostly in G0/G1-phase, and to a lesser extent in S-, and G2-/M-phases.
Evidence for this hypothesis is presented in Figure 45. This figure was adapted from a previous study which measured relative DNA content (i.e. fluorescence intensity) of both vegetative and stationary phase D. discoideum cells (Durston et al. 1984).
From Figure 45 it is seen that both exponential and stationary vegetative cells exhibit the same basic bimodal fluorescence intensity distribution (the similarity of distribution between vegetative and stationary phase cells, data not shown here, may also be seen in my June through July 1991 flow cytometric notes). The top half of Figure 45 is exactly parallel to results previously
Figure 45. This adaptation, from Durston et al. 1984, depicts two flow cytometric histograms. The first was derived from exponentially growing vegetative cells. The second was from growth arrested cells which had been held in stationary phase conditions for 12 h. Boxed inserts are part of the reinterpretation.
presented for vegetative cells (in Figure 13). There is a larger "lower" peak at a fluorescence intensity of 75 units, and a smaller "upper" peak at a fluorescence intensity of about 150 units. The authors (Durston et al. 1984) even state that "stationary phase cells gave a bimodal distribution similiar to that of exponential cells, in which the modal intensities of the single and double intensity peaks showed no consistent differences from those of exponential cells." Durston et al. conclude that the double intensity peak is due to multinuclearity. This thesis work directly disputes their multinuclearity theory, and reinterprets the peak seen at 80-90 units for both vegetative and stationary phase cells as being representative of G0/G1-phase nuclei. Conversely, this reinterpretation defines the peak seen at 150-160 units for both vegetative and stationary phase cells to be representative of G2-/M-phase nuclei. The percentages of the vegetative cell-cycle phases were calculated from data shown in Figure 45 by measuring the areas under the curve, and are as follows: G0/G1, 56%; S, 10%; G2/M, 34%. Likewise, the percentages of the stationary phase cell-cycle phases were calculated, and are as follows: G0/G1, 57%; S, 9%; G2/M, 34%.
Therefore, stationary phase D. discoideum cells arrest predominantly in G0/G1-phase, and to a lesser extent in S-, G2-/M-phases. This is similiar to what is seen for other eukaryotic systems in the literature. Stationary phase mouse cells arrest predominantly in G0/G1-phase and to a lesser extent in S-, and G2-/M-phases (Loskutoff and Paul 1978; Herget 1993). More than 95% of stationary phase mouse cells display a G0/G1-phase content of DNA, the rest display S- and G2-/M-phase contents of DNA (Luk et al. 1985). Stationary phase mouse erythroleukemia cells display distributions in G0/G1-, S-, and G2-/M-phases which are almost identical to their exponentially growing control (Yancheva and Djondjurov 1982). Over 90% of the cells in three different Chinese hamster ovary cells lines displayed a G0/G1-phase DNA content after being arrested with stationary phase contitions, the rest display S- and G2-/M-phase contents of DNA (Baisch 1988). The non-proliferating stationary phase of quiescent human cells is traditionally located to a large extent in G0/G1-phase, and to a lesser extent in S-, and G2-/M-phases (Drewinko et al. 1984).
As mentioned previously, vegetative D. discoideum cells cultured on a solid surface aggregate and form a slug when they run out of food. The slug goes on to further differentiate into a fruiting body composed of stalk and spore cells. This transformation of vegetative cells into either spore or stalk cells is non-random. It has been reported that during development the cells comprising the anterior third of the migrating slug will become stalk cells, and those comprising the posterior two-thirds differentiate into spore cells (Bonner 1947; Shaffer 1953). More recently, it has been reported that a population of vegetative D. discoideum cells will transform itself into 75% spore cells and 25% stalk cells (Weijer et al 1984). Therefore a more accurate estimation of the vegetative transformation ratio of spore cells to stalk cells is ((67+75)/2):((33+25)/2), or 2.5:1. This ratio will be mentioned again shortly.
The cell-cycle phase at the time of differentiation has long been thought to somehow control the decision for prestalk or prespore differentiation. It has been reported that cells which are in the early part of the cell-cycle become stalk cells, and that cells late in the cycle become spore cells (Weijer et al. 1984; Sharpe et al. 1984). This reinterpretation of the published information hypothesizes just the opposite: cells late in the cycle (G2-/M-phase) become stalk cells, and cells in the early part of cycle (G0/G1-phase) become spore cells.
The evidence for this hypothesis lies in a previous study which measured nuclear volume and diameter throughout all the different stages of the D. discoideum life cycle. These published data are adapted from Bonner et al. 1955, in Figure 46, as a series of graphs on nuclear size. In this figure, the boxed inserts are part of my reinterpretation of spore and stalk cell differentiation. The solid curves represent prespore or spore cells, while the non-solid curves represent prestalk or stalk cells.
As seen in Figure 46, vegetative cells have two peaks of nuclear distribution. Again, this is exactly parallel to results presented previously (Figure 13). It appears that the larger "lower"
Figure 46. This adaptation, from Bonner and Franscella 1953, depicts nuclear frequency plotted against diameter for all stages of development. Each curve is from 30 observations. Veg., agg., mig., and culm. are abbreviations for vegetative, aggregation, migration, and culmination stages, respectively.
peak on the left corresponds to G0/G1-phase of the cell-cycle, and has a measured nuclear volume at the peak of distribution of 20 mm3, and a diameter of 3.4 mm. This same nuclear diameter measurement of 3.4 mm was previously employed in the Results and Discussions section (Analysis of size distribution for wild-type cells), for calculations of mean nuclear radius and volume during the different phases of the cell-cycle. In Figure 46, vegetative nuclei are seen to have a smaller "upper" peak which corresponds to G2-/M-phase of the cell-cycle, and a measured nuclear volume at the peak of distribution of 50 mm3, and a diameter of 4.6 mm. This thesis work assumes that nuclear volume doubles during the cell-cycle, from one volume in G0/G1-phase to two volumes in G2-M-phase, and that this is reflected in the G0/G1-, S-, and G2-/M-phase contents of DNA. Thus for vegetative cells, if the G2-/M-phase nuclei are observed to have a nuclear diameter of 4.6 mm, then the calculated G0/G1-phase diameter should be 4.6(1/1.26)= 3.65 mm. The actual measured diameter for the G0/G1-phase nuclei is 3.4 mm. Therefore the measured diameter of G0/G1-phase nuclei varies from the calculated diameter by about 100-(100(3.65/3.4)), or 7%. The percentages of the vegetative cell-cycle phases were calculated from data shown in Figure 46 by measuring the areas under the curve, and are as follows: G0/G1, 64%; S, 12%; G2/M, 24%.
Starvation conditions produce progressively smaller cells and nuclei. On a solid surface, the continuation of starvation conditions causes cells to aggregate into a slug. During this aggregation, as seen from Figure 46, the nuclear distribution remains similiar, but smaller overall, to that of vegetative cells. Thus aggregating cells also produce nuclei which have two peaks of distribution. The larger "lower" peak appears to correspond to G0/G1-phase nuclei, and has a measured nuclear volume at the peak of distribution of 6.5 mm3, and a nuclear diameter of 2.2 mm. The smaller "upper" peak likewise corresponds to G2-/M-phase nuclei, and has a measured nuclear volume at the peak of distribution of 13 mm3, and a diameter of 2.9 mm. Since the G2-/M-phase nuclei are observed to have a nuclear diameter of 2.9 mm, then the calculated G0/G1-phase diameter should be (2.9(1/1.26))= 2.3 mm. The actual measured diameter for the G0/G1-phase nuclei is 2.2 mm. Therefore the measured diameter of G0/G1-phase nuclei varies from the calculated diameter by about 100-(100(2.3/2.2)), or 5%. The percentages of the aggregating cell's cell-cycle phases were calculated from Figure 46, and are as follows: G0/G1, 68%; S, 3%; G2/M, 29%.
During migration, the slug differentiates into prespore and prestalk cells as the nuclear distribution changes from that seen in vegetative and aggregating cells. As seen from Figure 46, by the end of migration and into the culmination stage, spore cells have arisen from the population having a nuclear volume at the peak of distribution of about 3 to 5 mm3, a diameter of 1.8 to 2.2 mm, and a preponderance of G0/G1-phase DNA. Conversely, during migration, pre-stalk cells arise from the population having preponderance of G2-/M-phase DNA, a nuclear volume at the peak of distribution of 12 mm3, and a diameter of 3 mm. The percentages of stalk cell cell-cycle phases at the end of migration were calculated, and are as follows: G0/G1, 38%; S, 1%; G2/M, 61%.
By culmination, stalk cells possess bimodal peaks of nuclear distribution. The "lower" peak shows a nuclear volume of 6 mm3, a diameter of 2.4 mm, and is representative of G0/G1-phase. The "upper" peak shows a nuclear volume of 12 mm3, a diameter of 2.9 mm, and is representative of G0/G1-phase. The G2-/M-phase volume (12 mm3) is exactly twice that of the G0/G1-phase volume (6 mm3). The diameter of the stalk cell G2-/M-phase nuclei, 2.9 mm, should in theory be offset by G0/G1-phase nuclei which are (1/1.26(2.9))= 2.3 mm in diameter. The actual diameter of stalk cell G0/G1-phase nuclei, 2.4 mm, therefore varies from the calculated diameter by (100-(100(2.3/2.4)))= 4%. The percentages of stalk cell cell-cycle phases during culmination were calculated, and are as follows: G0/G1, 63%; S, 2%; G2/M, 35%.
Perhaps a D. discoideum cell's fate is dependent upon its cell-cycle position. It was mentioned that vegetative cells produce a ratio of 71% spore cells to 29% stalk cells or 2.5:1. This is somewhat approximated by the ratio 51%:26% of G0/G1-phase nuclei to G2-/M-phase nuclei from vegetative D. discoideum cells, see Table 1. However, it is important to keep in mind that Table 1 was devised from data shown in Figure 13, with a series of different S-phase fitting computer programs, some of which may have exaggerated the S-phase and therefore underestimated the G0/G1- and G2-/M-phases. When the areas under the curve of Figure 13 were calculated manually for purposes of this reinterpretation, G0/G1-phase was seen to be almost 65% of the cell-cycle, and G2-/M-phases were seen to be about 25% of the cycle, and S-phase was about 10%. Hence, it is perhaps more accurate to calculate the vegetative G0/G1-phase to G2/M-phase ratio from data shown in Figure 13 (using both the computer and manual calculations) and from the vegetative data shown in Figures 45 and 46. The ratio of G0/G1-phase cells to G2-/M-phase cells is ((51+65+56+64)/2:((26+25+34+24)/2)=59:27, or about 2.2:1. Therefore, the the ratio of spore cells to stalk cells, and the ratio of G0/G1-phase cells to G2-/M-phase cells differs by (100-100(2.2/2.5))= 12%. Thus it is possible that there is a correlation between cell-cycle position during the end of vegetative phase, and the resulting proportions of spore to stalk cells produced during development.
Each curve shown in Figure 46 was derived from 30 observations. This resulted in the apparent proportion of differentiating spore to stalk cells shown in this figure to be about 1:1. In reality, since vegetative cells will produce 71% spore cells and 29% stalk cells, the correct ratio should be 2.5:1, and not 1:1 as is shown in Figure 46. Therefore, a proportional approximation of the corrected spore to stalk cell ratio during the migratory and culmination stages (from data shown in Figure 46) is shown in Figure 47. In this figure, the solid curves represent prespore or
Figure 47. Corrected nuclear proportions (from Figure 46), depicting nuclear frequency plotted against diameter for the beginning and end of migration, and the culmination stages respectively. Boxed inserts are part of the reinterpretation.
spore cells, while the non-solid curves represent prestalk or stalk cells. The percentages of the pre-spore cell-cycle phases at the beginning of migration were calculated, and are as follows: G0/G1, 36%; S, 3%; G2/M, 61%. Likewise, the percentages of the pre-spore cell-cycle phases during the end of migration are as follows: G0/G1, 87%; S, 1%; G2/M, 12%. By culmination, almost 100% of the spores are in G0/G1-phase.
It has been reported that D. discoideum spore cells have a G0/G1-phase content of DNA (Sharpe et al. 1984). Perhaps differentiating cells use cell-cycle-dependent gene regulation for their transformation into spores. We know that each cell begins its life history with the germination of a spore. From each G0/G1-phase spore emerges a vegetative amoeba that spends most of its time in G0/G1-phase of the cell cycle. Perhaps this arrangement allows new amoeba to hit the ground running after emerging from G0/G1-phase spores because they are automatically in G0/G1-phase of the cell-cycle (a phase that this thesis work has concluded comprises the majority of the vegetative cell-cycle, about 59%). Thus after emerging from a spore, each cell would begin its cell-cycle from the beginning of the cell-cycle, G0/G1, ready to feed by phagocytosis during the majority of its cell-cycle (G0/G1-phase). Phagocytosis during G0/G1-phase would then eventually result in the accumulation of enough biomolecules to allow the cell to progress past G0/G1-phase and replicate its DNA during S-phase. Cells would then proceed through G2-phase, rapidly divide in M-phase, and start the cycle anew (in G0/G1-phase).
Why would D. discoideum cells possessing a G2-/M-phase content of DNA differentiate into stalk cells? Perhaps being in G2-/M-phase during some critical period of development renders prestalk cells, in a sense, infertile. Prestalk cells might also rely on some amount of differential gene expression to produce stalk cells during their last cell-cycle progression. Thus, the (non-viable) prestalk cell's seemingly altruistic act of "giving up its life" to lift (viable) spore cells during development might be an inevitable consequence of its cell-cycle position.
Given this hypothetical model of development, it is not too far out to speculate upon the possibility that some (small) amount of differentiating cells, which are in a minor (<10%) part of the cell-cycle during a critical time period (perhaps S-phase), differentiate into attachment cells at the base of the stalk. Again, this would require some amount of cell-cycle dependent differential gene expression. An experiment to test this hypothesis is discussed in the next section.
1. By pulse labeling wild-type (HPS401) cells at representative times after release from aphidicolin arrest, and then quantitating the counts of TCA-precipitable material, we should be able to estimate the relative lengths of the phases of the cell-cycle and test the basic premise of this thesis.
2. By pulse labeling wild-type (HPS401) cells and quantitating labeled mitotic nuclei as a function of time on autoradiographs, we should be able to estimate the relative lengths of the phases of the cell-cycle from another method, thereby allowing a crosscheck with the results from #1 (Zada-Hames and Ashworth 1977).
3. We could test the hypothesis of cell-cycle dependent differentiation by examining whether or not vegetative/aggregating G0/G1-phase cells differentiate into spore cells, S-phase cells differentiate into substrate attachment cells (at the base of the stalk), and/or G2-/M-phase cells differentiate into stalk cells. This experiment would be initiated by continuously labeling a flask of vegetative HPS401 cells with bromodeoxyuridine (BrdU), and plating out aliquots of these cells on plates (without bacteria) for development, as a function of increasing amounts of label incorporation (perhaps 60, 180, and 480 min). After culmination, one would fix sectioned fruiting bodies on slides, and stain with an anti-BrdU antibody (Weijer and Krefft 1989; Rudolph and Latt 1989; Dolbere et al. 1990). If this hypothesis of cell-cycle dependent differentiation is correct, than progressive staining with the anti-BrdU antibody will be seen, starting first with the stalk attachment cells, then the stalk cells, and then finially the spore cells.
4. The wild-type (HPS401) cell-cycle dependent transformation efficiency could be approximated by measuring the relative amounts of transformation by electroporation of plasmids into cells as a function of time, after release from an aphidicolin block. This experiment would show where in the cell-cycle the maximum transformation of cells with exogenous DNA occurs. Historically, transformation efficiency of D. discoideum is rather low, i.e. about 50 to 100 transformants per 106 cells (Nellen and Saur 1988). Perhaps this because most (51-59%) vegetative cells are in a non-transformable G0/G1-phase. Phagocytosis during G0/G1-phase of the cell-cycle could therefore result in the digestion of, and not transformation with, exogenous DNA; maybe transformation efficiency would increase during an aphidicolin-induced S-phase arrest. Conversely, it is possible that transformation efficiency might increase during a CIPC-induced mitotic arrest.
5. Double-strand break formation and rejoining could be examined with pulsed field gel electrophoresis. We might see evidence that some types of DNA damage repair are cell-cycle dependent in HPS401 cells, but not the radB strains. Cell-cycle dependence of some types of DNA damage repair has been shown in yeast (Weinert and Hartwell 1988; Hartwell and Weinert 1989; Terleth et al. 1990; Hanawalt 1990; Hartwell 1992).
6. Replication fork DNA polymerase studies might show that caffeine produces the rad phenotype in wild-type D. discoideum after UV irradiation (Freim and Deering 1970) because this drug abolishes S-, and/or G2-/M-phase arrests after UV induced DNA damage (Downes et al. 1990; Hartwell and Weinert 1989) by targetting the replication fork DNA polymerases (Solberg et al. 1978; Richardson et al. 1981; Hatayama et al. 1984; Das 1987; Musk et al. 1988).
Aerts, R. J., A. J. Durston, and W. H. Moolenaar (1985). Cytoplasmic pH and the regulation of the Dictyostelium cell cycle. Cell 43, 653-657.
Ansari, B., R. Dover, C. Gillmore, and P. Hall (1993). Expression of the nuclear membrane protein statin in cycling cells. Journal of Pathology 169, 391-396.
Baisch, H. (1988). Different quiescence states of three culture cell lines detected by acridine orange staining of cellular RNA. Cytometry 9, 325-331.
Balter, M. (1991). Cell cycle research: down to the nitty gritty. Nature (News and Views) 350, 1253-1254.
Bantel-Schaal, U., and M. Stoehr (1992). Influence of adeno-associated virus on adherence and growth properties of normal cells. Journal of Virology 66, 773-779.
Barfod, I. J. and N. M Barfod (1980a). Cell-production rates extimated by the use of vincristine sulphate and flow cytometry. I. An in vitro study using murine tumour cell lines. Cell Tissue Kinet., 13, 1-8.
Barfod, I. J. and N. M Barfod (1980b). Cell-production rates extimated by the use of vincristine sulphate and flow cytometry. II. Correlation between the cell-production rates of aging ascites tumours and the number of S phase tumour cells. Cell Tissue Kinet., 13, 9-19.
Barlat, I., D. Fesquet, C. Rechot, B. Henglein, A. Dangeac, A. Vie, and J. lanchard (1993). Loss of the G-1-S control of cyclin A expression during tumoral progression of Chinese hamster lung fibroblasts. Cell Growth Differentiation 4, 105-113.
Baumstark-Khan, C. (1992). Effect of aphidicolin on DNA synthesis, PLD-recovery and DNA repair of human diploid fibroblasts. International Journal of Radiation Biology 61, 191-197.
Bennett, C. B., A. L. Lewis, K. K. Baldwin, and M. A. Resnick (1993). Lethality induced by a single site-specific double-strand break in a dispensable yeast plasmid. Proceedings of the National Academy of Sciences USA 90, 5613-5617.
Boehm, T., L. Mengle-Gaw, U. R. Kees, N. Spurr, I. Lavenir, A. Forster and T. H. Rabbitts (1989). Alternating purine-pyrimidine tracts may promote chromosomal translocations seen in a variety of human lymphoid tumours. The EMBO Journal 8, 2621-2631.
Bonner, J. T. and E. B. Franscella (1953). Variations in cell size during the development of the slime mole Dictyostelium discoideum. Biol. Bull. 104, 297-300.
Bonner, J. T., A. D. Chiquoine, and M. Q. Kolderie (1955). A histochemical study of differentiation in the cellular slime molds. J. Exp. Zoo. 130, 133-158.
Bootsma, D. W. Keijzer, E. G. Jung and E. Bognert (1989). Xeroderma pigmentation complementation group XP-1 withdrawn. Mutation Research 218, 149-152.
Boyd, J. B., and R. B. Setlow (1976). Characterization of postreplication repair in mutagen-sensitive strains of Drosophila melanogaster. Genetics 84, 507-526.
Breimer, L., and T. Lindahl (1980). A DNA glycosylase from E. coli that releasese freee urea from a polydeoxyribonucleotide containing fragments of base residues. Nucleic Acids Res. 8, 6199-6203.
Bronner, C. E., D. L. Welker and R. A. Deering (1992). Mutations affecting sensitivity of the cellular slime mold Dictyostelium discoideum to DNA-damaging agents. Mutation Research, DNA Repair 274, 187-200.
Bronner, C. E., D. L. Welker and R. A. Deering (1988). Genetic analysis of two mutations affecting thymidine metabolism in Dictyostelium discoideum. Genetics 120, 959-964.
Buratowski, S. (1993). DNA repair and transcription: the Helicase connection. Science 260, 37-38.
Chen, J., B. Derfler, A. Maskati and L. Samson (1989). Cloning a eukaryotic DNA glycosylase repair gene by the suppression of a DNA repair defect in Escherichia coli. Proceedings of the National Academy of Sciences USA 86, 7961-7965.
Chetsanga, C. J., M. Lozon, C. Makaroff, and L. Savage (1981). Purification and characterization of E. coli formamidopyrimidine-DNA glycosylase that excises damaged 7-methylguanine from deoxyribonucleic acid. Biochemistry 20, 5201-5205.
Clark, J. M., and R. A. Deering (1981). Excision of pyrimidine dimers from nuclear deoxyribonucleic acid in ultraviolet-irradiated Dictyostelium discoideum. Molecular and Cellular Biology 1, 121-127.
Clarke, A. R., E. R. Maandag, M. Van-Roon, and N. M. T. Van-der-Lugt, M. Van-Der-Valk, M. L. Hooper, A. Berns, and H. Te-Riele (1992). Requirement for a functional RB-1 gene in murine development. Nature (Lond.) 359, 328-330.
Cleaver, J. E., and K. H. Kraemer (1989). Xeroderma pigmentosum, in: The Metabolic Basis of Inherited Disease 6th ed. (eds. C. R. Scrivey, A. L. Beaudet, W. S. Sly and D. Valle) McGraw-Hill, Inc, New York, 2949-2971.
Collins, A. and R. T. Johnson (1987). DNA repair mutants in higher eukaryotes, in: Molecular Biology of DNA Repair. (eds. A. Collins, R. T. Johnson and J. M. Boyle) The Company of Biologists Limited, Cambridge, 61-82.
Coulondre, C. and J. H. Miller (1977). Genetic studies of the lac repressor. IV. Mutagenic specificity in the lacI gene of Escherichia coli. Journal of Molecular Biology 117, 577-606.
Darzynkiewicz, Z., F. Traganos, M. Andreeff, T. Sharpless and M. R. Melamed (1979). Different sensitivity of chromatin to acid denaturation in quiescent and cycling cells as revealed by flow cytometry. J. Histochem. Cytochem. 27, 478-485.
Darzynkiewicz, Z., F. Traganos, and M. R. Melamed (1981a). New cell cycle compartmentsidentified by multiparameter flow cytometry. Cytometry 1:98-108.
Darzynkiewicz, Z., F. Traganos, S. B. Xue, L. Stainano-Coico and M. R. Melamed (1981b). Rapid analysis of drug effects on the cell cycle. Cytometry 1, 279-286.
Darzynkiewicz, Z., H. Crissman, F. Traganos, and J. Steinkamp (1982). Cell heterogeneity during the cell cycle. J. Cell Physiol. 113, 465-474.
Darzynkiewicz, Z. and F. Traganos (1982). RNA content and chromatin structure in cycling and noncycling cell populations studied by flow cytometry. In: (G. M. Padilla and K. S. McCarty, eds.), Genetic Expression in the Cell Cycle. New York: Academic Press.
Darzynkiewicz, Z. (1983). Molecular interactions and cellular changes during the cell cycle. Paramac. Therap. 21, 143-188.
Darzynkiewicz, Z. (1984). Metabolic and kinetic compartments of the cell cycle distinguished by multiparameter flow cytometry. In: (P. Skehan and S. Friedman, eds.), Growth, Cancer and the Cell Cycle. New Jersey: Humana Press.
Das, S. K. (1987). Abrogation of the effects of aphidicolin on NIH3T3 and V79 cells by caffeine. Biochim Biophys Acta 928, 29-35.
Davies, J. E., K. Mynett, A. Gescher, J. K. Chipman (1993). DNA modifications and repair by 2 nitropropane is extensive in hepatocytes of rats compared to those of humans and mice. Mutation Research 287, 157-164.
Deering, R. A. (1968). Dictyostelium discoideum: a gamma-ray resistant organism. Science 162, 1289-1290.
Deering, R. A., M. S. Smith, B. K. Thompson, and A. C. Adolf (1970). Gamma-ray-resistant and -sensitive strains of the slime mold (Dictyostelium discoideum). Radiation Research 43, 711-728.
Deering, R. A. and D. S. Jensen (1973). Nuclear and mitochondrial DNA synthesis in gamma ray-resistant and -sensitive slime mold amoebas. Biophysical Journal 13, 780-794.
Dell-Sal, G., M. E. Ruaro, L. Philipson, and C. Schneider (1992).The growth arrest-specific gene GAS1 is involved in growth supression. Cell 70, 595-607.
Doetsch, P. W., D. E. Helland and W. A. Haseltine (1986). Mechanism of action of a mammalian DNA repair endonuclease. Biochemistry 25, 2212-2220.
Dolbeare, F., K. Wen-Lin, W. Beisker, M. Vanderlaan, and J. W. Gray (1990). Using monoclonal antibodies in bromodeoxyuridine-DNA analysis. Methods in Cell Biology 33, 207-216.
Dosik, G. M., B. Barlogie, A. R. White, W. Gohde, and B. Drewinko (1981). A rapid automated stathmokinetic method for determination of in vitro cell cycle transit times. Cell Tissue Kinet. 14, 121-134.
Downes, C. S., S. R. Musk, J. V. Watson, and R. T. Johnson (1990). Caffeine overcomes a restriction point associated with DNA replication but does not accelerate mitosis. Journal of Cell Biology 110, 1855-1860.
Doyle, J. W., P. F. Ward-Bailey, A. A. Kandutsch (1993). Effects of growth factors on cell cycle arrest in dolichyl phosphate-depleted cultures. Journal of Cell Physiology 155, 171-178.
Drewinko, B., L. Y. Yang, B. Barlogie, and J. M. Trujillo (1984). Cultured human tumor cells may be arrested in all stages of the cycle during stationary phase demonstration of quiescent cells in G1-, S-, and G2-phase. Cell Tissue Kinetics 17, 453-464.
Durston, A. J., C. J. Weijer, J. F. Jongkind, A. Verkerk, A. Timmermans, and W. T. Kulve (1984). A flow fluorometric analysis of the cell-cycle during growth and differentiation in Dictyostelium discoideum. Wilhelm Roux's Archives of Developmental Biology 194, 18-24.
EPICS Cytologic User Manual (1988). EPICS Cytologic Software Version 2.0. Section 3-3. Copyright 1987, by Coulter Electronics, Inc., Hialeah, FL.
Evans, G. A., D. Linnekin, S. Grove, and W L. Farrar (1992). Specific protein kinases modulated during T cell mitogenesis activity of a 55 KDa serine kinase is associated with growth arrest in human T cells. Journal of Biological Chemistry 267, 10313-10317.
Freim, J. O. and R. A. Deering (1970). Ultraviolet irradiation of the vegetative cells of Dictyostelium discoideum. Journal of Bacteriology 102, 36-42.
Friedberg, E. C. (1985). DNA repair, W. H. Freeman and company, San Francisco, 614 pages.
Friedberg, E. C., and P. C. Hanawalt (1981). DNA Repair: A Laboratory Manual of Research Proceedures, vol 1. Marcel Dekker, Inc., New York, 637 pages.
Friedberg, E. C., T. Bonura, E. H. Radany, and J. D. Love (1981). Enzymes that incise damaged DNA. The enzymes. 3rd ed. P. B. Boyer, ed., vol 19, Nucleic acids, part A, p. 251. New York: Academic Press.
Furihata, C., S. Oguchi, and T. Matsushima (1993). Possible tumor-initiating and promoting activity of P methylcatechol and methylhydroquinone in the pyloric mucosa of rat stomach. Japanese Journal of Cancer Research 84, 223-229.
Game, J.C. (1983). Radiation-sensitive mutants and repair in yeast, in: Yeast Genetics (eds. J. F. T. Spencer, D. M. Spencer and A. R. W. Smith) Springer-Verlag, Inc., New York, 109-137.
Glickman, B. W., and M. Radman (1980). Escherichia coli mutator mutants deficient in methylation-instructed DNA mismatch correction. Proceedings of the National Academy of Sciences, USA 77, 1063-1067.
Green, D. A., and W. A. Deutsch (1983). Repair of alkylated DNA: Drosophila have DNA methyltransferase but not DNA glycosylases. Molecular and General Genetics 192, 322-325.
Guialis, A., and R. A. Deering (1976a). Inhibition of UV-induced DNA strand breakage in Dictyostelium discoideum by 2,4-dinitrophenol. Photochemistry and Photobiology 24, 331-336.
Guialis, A., and R. A. Deering (1976b). Repair of deoxyribonucleic acid in ultraviolet light-sensitive and -resistant Dictyostelium discoideum strains. Journal of Bacteriology 127, 59-66.
Guyer, R. B., J. M. Nonnemaker and R. A. Deering (1986). Uracil DNA glycosylase activity from Dictyostelium discoideum. Biochimica et Biophysica Acta 868, 262-264.
Guyer, R. B. and R. A. Deering (1985). Apurinic/apyrimidinic-specific endonuclease activities from Dictyostelium discoideum. Biochimica et Biophysica Acta 824, 304-312.
Haynes, R. H., and B. A. Kunze (1981). DNA Repair and Mutagenesis in Yeast, in: The Molecular Biology of the Yeast Saccharomyces. (eds. J. N. Strathern, E. W. Jones and J. R. Broach) Cold Spring Harbor Press, Cold Spring Harbor, 371-414.
Hall, J., and S. Woodward (1989). Aphidicolin resistance in herpes simples virus type 1 appears to alter substrate specificity in the DNA polymerase. Journal of Virology 63, 2874-2876.
Hall, J. D., Y. Wang, J. Pierpoint, M. S. Berlin, S. E. Rundlett, and S. Woodward (1989). Aphidicolin resistance in herpes simplex virus type I reveals features of the DNA polymerase dNTP binding site. Nucleic Acids Research 17, 9231-9243.
Hanawalt, P. C. (1990). Selective DNA repair in expressed genes in mammalian cells, in: Mutation and Environment Part A: Basic Mechanism. (eds. M. L. Mendelsohn and R. J. Albertini) Wiley-Liss, Inc, New York, 213-222.
Hartwell, L. (1992). Defects in the cell cycle may be responsible for the genomic instability of cancer cells. Cell 71, 543-546.
Hartwell, L., and T. A. Weinert (1989). Checkpoints: controls that ensure the order of cell cycle events. Science 246, 629-634.
Hass, R., H. Gunji, R. Datta, S. Kharbanda, and A. Hartmann (1992). Differentiation and retrodifferentiation of human hyeloid leukemia cells is associated with reversible induction of cell cycle-regulatory genes. Cancer Research 52, 1445-1450.
Hass, R., H. Gunji, M. Hirano, R. Weichselbaum, and D. Kufe (1993). Phorbol ester-induced monocytic differentiation is associated with G-2 delay and down-regulation of cdc25 expression. Cell Growth Differentiation 4, 159-166.
Hatayama, T., S. L. Qi, K. Kim, K. Ichiba, and M. Yukioka (1984). Caffeine-induced reorganization of DNA replicating system occurs on or near nuclear matrix in HELA cells. Biochem Int 9, 651-658.
Herget, T. H., S. F. Brooks, S. Broad and E. Rozengurt (1993). Expression of the major protein kinase C substrate, the acidic 80-kilodalton myristolated alanine-rich C kinase substrate, increases sharply when Swiss 3T3 cells move out of cycle and enter G0. Proceedings of the National Academy of Sciences USA 90, 2945-2949.
Ho, K. S. Y. (1975). Induction of DNA double strand breaks by X-rays in a radiation sensitive strain of Saccharomyces cerevisiae. Mutation Research 30, 327 334.
Hoffman, B., H. M. Hanauske-Abel, A. Flint, and M. Lalande (1991). A new class of reversible cell cycle inhibitors. Cytometry 12, 26-32.
Holm, C., T. Goto, J. C. Wang, and D. Botstein (1985). DNA topoisomeraase II is required at the time of mitosis in yeast. Cell 41, 553-563
Howard-Flanders, P., R. P. Boyce and R. Theriot (1966a). Three loci in Escherichia coli K-12 that control the excision of thymine dimers and certain other mutagen products from host or phage DNA. Genetics 53, 1119-1136.
Howard-Flanders, P., R. P. Boyce and R. Theriot (1966b). DNA repair and genetic recombination: studies on mutants of Escherichia coli defective in these processes. Genetics 53, 1137-1150.
Huang, L., J. J. Turchi, A. F. Wahl, R. A. Bambara (1993). Effects of the anticancer drug cis diamminedichloroplatinum-II on the activities of calf thymus DNA polymerase. Biochemistry 32, 841-848
Hurley, D. L., A. M. Skantar, and R. A. Deering (1989). Nuclear DNA synthesis is blocked by UV irradiation in Dictyostelium discoideum. Mutation Research, 217, 25-32.
Hurley, D. L. and R. A. Deering (1988). Enhanced thymidine uptake causes the lowered thymidine requirement of Dictyostelium discoideum. Mutation Research 217, 25-32.
Jimenez, G., J. Yucel, R. Rowley, and S. Subramani (1992). The RAD3 positive gene of Schizosaccharomyces-pombe is involved in multiple checkpoint functions and in DNA repair. Proceedings of the National Academy of Sciences 89, 4952-4956.
Kaguni, L.S., and I. R. Lehman (1988). Eukaryotic DNA polymerase-primase alpha structure, mechanism, and function. Biochem Biophys Acta 950:87-101.
Karran, P., and T. Lindahl (1978). Enzymatic excision of free hypoxanthine from polydeoxynucleotides and DNA containing deoxyinosine monophosphate residues. J. Biol. Chem. 253, 5877-5882.
Karran, P., and T. Lindahl (1980). Hypoxanthine in DNA: generation of heat-induced hydrolysis of adenine residues and release in free form by a DNA glycosylase from calf thymus. Biochemistry 19, 6005-6009.
Kastan, M. B., Q. Zhan, W. S. El-Deiry, F. Carrier, T. Jacks, W. V. Walsh, B. S. Plunkett, B. Vogelstein, and A. J. Fornace, Jr. (1992). A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 71, 587-597.
Katz, K. S. and D. I. Ratner (1988). Homologous recombinatin and the repair of double-strand breaks during cotransformation of Dictyostelium discoideum. Molecular and Cellular Biology 8, 2779-2786.
Katz, E. R., and L. Y. W. Bourguignon (1974). The cell cycle and its relationship to aggregation in the cellular slime mold, Dictyostelium discoideum. Developmental Biology 36, 82-87.
Kessis, T. D., R. J. Slebos, W. G. Nelson, M. B. Kastan, and B. S. Plunkett (1993). Human papillomavirus 16 E6 expression disrupts the p53-mediated cellular response to DNA damage. Proceedings of the National Academy of Sciences 90, 3988-3992.
Kelley, M. R., S. Venugopal, J. Harless and W. A. Deutsch (1989). Antibody to a human DNA repair protein allows the cloning of a Drosophilia cDNA that encodes an apurinic endonuclease. Molecular and Cellular Biology 9, 965-973.
Khoury, A. T., and R. A. Deering (1973). Sedimentation of DNA of Dictyostelium discoideum lysed on alkaline sucrose gradients: role of single-strand breaks in gamma ray lethality of sensitive and resistant strains. Journal of Molecular Biology 79, 267-284.
Kielman, J. K., and R. A. Deering (1980). Ultraviolet light-induced inhibition of cell division and DNA synthesis in axenically grown repair mutants of Dictyostelium discoideum. Photochemistry and Photobiology 32, 149-156.
Kim, R. A. and J. C. Wang (1989). A subthreshold level of DNA topoisomerase leads to the excision of yeast ribosomal DNA as extrachromosomal rings. Cell 57, 975-979.
Kim, R. A. and J. C. Wang (1989). Identification of the TOP3 gene product as a single strand-specific DNA topoisomerase. Journal of Biological Chemistry 267, 17178-17185.
Kim, R. A. and J. C. Wang (1989). Function of DNA topoisomerase as replicon swivels in Saccharomyces cerevisiae. Journal of Molecular Biology 208, 257-268.
Kimchi, A. (1992). Cytokine triggered molecular pathways that control cell cycle arrest. Journal of Cell Biochemistry 50, 1-9.
Kmiec, E. B. and W. K. Holloman (1986). Homologous pairing of DNA molecules by Ustilago Rec1 protein is promoted by sequences of Z-DNA. Cell 44 545-554.
Kornberg, A. and T. Baker (1992). in DNA Replication (Second Edition), pg 199, and all of Ch 14.
Krasin, F. and F. Hutchinson (1977). Repair of double strand breaks in E. coli which requires recA function and the presence of a duplicate genome. Journal of Molecular Biology 116, 81-98.
Krek, W., and E. A. Nigg (1991). Differential phosphorylation of vertebrate p34CDC-2-kinase at the G1-S and G2-M transitions of the cell cycle: indentification of major phosphorylation sites. EMBO Journal 10, 305-316.
Lavin, M. R. and A. L. Schroeder (1988). Damage-resistant DNA synthesis in eukaryotes. Mutation Research 193, 193-206.
Lawley, P. D., and W. Warren (1976). Removal of mutagenic and lethal lesions from alkylated DNA. In: Progress in clinical and biological research. E. A. Mirard, W. B. Hutchinson and E. Mihich, eds., vol. 132B, p. 241. New York. Alan R. Liss publisher.
Leach, C. K., and J. M. Ashworth (1972). Characterization of DNA from Dictyostelium discoideum after growth of the amoebae in different media. J. Mol. Biol. 68, 35-48.
Lehman, I. R. (1981). Measurment of postreplication repair in mammalian cells. In: DNA Repair: A Laboratory Manual of Research Procedures. vol 1. (eds. E. C. Freidberg and P. C. Hanawalt) Marcel Dekker, Inc., New York, 471-486.
Lehman, I. R., and L.S. Kaguni (1989). DNA polymerase alpha. Journal of Biological Chemistry 264:4265-4268.
Lewin, B. (1985). "..the G1 period often comprises the major part of the cycle.." in GENES II, p521. John Wiley and Sons.
Lewin, B. (1990). Driving the Cell cycle: M phase kinase, its partners and substrates. Cell 61, 743-752.
Lindhal, T. (1982). DNA repair enzymes. Annual Reviews of Biochemistry 51, 61-87.
Loomis, W. F. (1982). The Development of Dictyostelium discoideum. Academic Press, New York, 551 pages.
Loskutoff, D. J., and D. Paul (1978). Intra cellular plasminogen activator activity in growing and quiescent cells. Journal of Cellular Physiology 97, 9-16.
Loveless, A. (1969). Possible relevance of O-6 alkylation of deoxyguanosine to the mutagenicity and carcinogenicity of nitrosamines and nitrosamides. Nature 223, 206-207.
Luk, C. K., P. C. Keng, and R. M. Sutherland (1985). Regrowth and radiation sensitivity of quiescent cells isolated from EMT-6-RO-Fed plateau monolayers. 45, 1020-1025.
Maeda, Y. (1986). A new method for inducing synchronous growth of Dictyostelium discoideum cells using temperature shifts. Journal of General Microbiology 132, 1189-1196.
Maeda, Y. (1988). Changes of endocytotic activities during the cell cycle of Dictyostelium discoideum. Developmental Growth and Differentiation 30, 15-24.
Maeda, Y., T. Ohmori, T. Abe, F. Abe, and A. Amagai (1989). Transition of starving Dictyostelium cells to differentiation phase at a particular position of the cell cycle. Differentiation 41, 169-175.
Matherly, L. H., J. D. Schuetz, E. Westin, and I. D. Goldman (1989). A method for the synchronizatin of cultured cells with aphidicolin: application to the large-scale synchronization of L1210 cells and the study of the cell cycle regulation of thymidylate synthase and dihydrofolate reductase. Analytical Biochemistry 182, 338-345.
McDonald, S. A., and A. J. Durston (1984). The cell cycle and sorting behavior in Dictyostelium discoideum. Journal of Cell Science 66, 195-204.
McDonald, S. A. (1986). Cell-cycle regulation of center initiation in Dictyostelium discoideum. Developmental Biology 117, 546-549.
McKinney, J. D., and N. Heintz (1991). Transcriptional regulation in the eukaryotic cell cycle. Trends in Biochemical Sciences 16, 37-40.
Moore, P., and B. S. Strauss (1979). Sites of inhibition of in vitro DNA synthesis in carcinogen- and UV-treated fX174 DNA. Nature 278, 664-666.
Mullaney, J., W.W. M. Moss, and D. J. McGeoch (1989). Gene UL2 of herpes simplex virus type 1 encodes a uracil-DNA glycosylase. Journal of General Virology 70, 449-454.
Murphy, K. and J. R. Stringer (1986). RecA independent recombination of poly[d(GT)-d(CA)] in pBR322 Nucleic Acids Research 14, 7325-7329.
Murray, A. W., and M. W. Kirschner (1989). Cyclin synthesis drives the early cell cycle. Nature 339, 275-280.
Murray, A. W., M. J. Solomon, and M. W. Kirschner (1989). The role of cyclin synthesis and degradation in the control of maturation promoting factor activity. Nature 339, 2280-339.
Musk, S. R., C. S. Downes, and R. T. Johnson (1988). Caffeine induces uncoordinated expression of cell cyucle functions after UV irradiation accelerated cycle transit sister chromatid exchanges and premature chromosome condensation in a transformed indian muntjac cell line. Journal of Cell Science 90, 591-600.
Nakatsuru, Y., N. Nemoto, K. Nakagawa, P. Masahito and T. Ishikawa (1987). O-6-methylguanine transferase activity in liver from various fish. Carcinogenesis 8, 1123-1127.
Nasheuer, H. P., A. Moore, A. F. Wahl, and T. S. F. Wang (1991). Cell cycle dependent phosphorylation of human DNA polymerase alpha. Journal of Biological Chemistry 266, 7893-7903.
Nellen, W., and U. Saur (1988). Cell-cycle dependent transformation competence in Dictyostelium discoideum. Biochemical and Biophysical Research Communications 154, 54-59.
North, G. (1991). Starting and stopping. Nature 351, 604-605.
Oberg, F., L. G. Larrson, R. Anton, and K. Nilsson (1991). Interferon gamma abrogates the differentiation block in v-myc-expressing U-937 monoblasts. Proceedings of the National Academy of Sciences USA 88, 5567-5571.
O'Farrell, P. H. (1992). Cell cycle control: many ways to skin a cat. Trends in Biology 2, 159-163.
Ohmori, T., and Y. Maeda (1987). The developmental fate of Dictyostelium discoideum cells depends greatly on the cell-cycle position at the onset of starvation. Cell Differentiation 22, 11-18.
Olsen, L. C., R. Aasland, C. U. Wittwer, H. E. Krokan and D. E. Helland (1989). Molecular cloning of human uracil-DNA glycosylase, a highly conserved DNA repair enzyme. The EMBO Journal 8, 3121-3125.
Park, E., S. Guzder, M. H. Koken, I. Jaspers-Dekker, G. Weeda, J. H. J. Hoeijmakers, S. Prakash, and L. Prakash (1992). RAD25 SSL2 the yeast homolog of the human xeroderma pigmentosum group B DNA repair gene is essential for viability. Proceedings of the National Academy of Sciences USA 89, 11416-11420.
Payez, J., R. A. Deering and J. O. Freim (1972). Lethal effect of alkylating agents or ultraviolet light on the gamma-ray resistant parent strain and a gamma-ray sensitive strain of the cellular slime mold Dictyostelium discoideum. Mutation Research 15, 82-85. Pegg, A. E., M. Roberfroid, C. von Bahr, R. S. Foote, S. Mitra, H. Bresil, A. Likhachev & R. Montesano (1982). Removal of O-6-methylguanine from DNA by human liver fractions. Proceedings of the National Academy of Sciences USA 79, 5162-5165.
Percival, K. H., M. B. Klein and P. M. J. Burgers (1989). Molecular cloning and primary structure of the uracil DNA-glycosylase from Saccharomyces cerevisiae. Journal of Biological Chemistry 264, 2593-2598.
Pignolo, R. J., V. J. Cristofalo, and M. O. Rotenberg (1993). Senescent WI-38 cells fail to express EPC-1 gene induced in young cell upon entry into the G-O state. Journal of Biological Chemistry 268, 89949-8957.
Podgorski, G., and R. A. Deering (1980). Effect of methyl methanesulfonate on survival of radiation-sensitive strains of Dictyostelium discoideum. Mutation Research 73, 415-418.
Podgorski, G., and R. A. Deering (1984). Thymidine-requiring mutants of Dictyostelium discoideum. Mol. Cell Biol. 4, 2784-2791.
Popanda, O., and H. W. Thielmann (1992). The function of DNA polymerases in DNA repair synthesis of ultraviolet irradiated human fibroblasts. Biochim Biophys Acta 1129, 155-160.
Popoff, S. C., A. Spira, A. W. Johnson, and B. Demple (1990). Yeast structural gene (APN1) for the major apurinic endonuclease: Homology to Escherichia coli endonuclease IV. Proceedings of the National Academy of Sciences USA 87, 4193-4197.
Prakash, L. (1981). Characterization of postreplication repair in Saccharomyces cerevisiae and effects of rad6, rad18, rev3 and rad52 mutations. Molecular and General Genetics 184, 471-478.
Prasad, G., R. A. Edelson, P. D. Gorycki and T. L. Macdonald (1989). Structure-activity relationships for the inhibition of DNA polymerase a by aphidicoin derivatives. Nucleic Acids Research 17, 6339-6348.
Prasad, K. R., and P. M. Rosoff (1992). Characterization of the energy dependent mating factor-activated calcium influx in Saccharomyces-cerevisiae. Cell Calcium 13, 615-626.
Protic-Sablic, M., and K. H. Kraemer (1986). Reduced repair of non-dimer photoproducts in a gene transfected into zeroderma pigmentosum cells. Photochemistry Photobiology 43, 509-514.
Raper, K. B. (1935). Dictyostelium discoideum, a new species of slime mold from decaying forest leaves. Journal of Agricultural Research, Washington 50, 135-147.
Resnick, M. A., J. Boyce and B. Cox (1981). Postreplication repair in Saccharomyces cerevisiae. Journal of Bacteriology 146, 285-290.
Resnick, M. A., and P. Martin (1976). The repair of double-strand breaks in the nuclear DNA of Saccharomyces cerevisiae and its genetic control. Molecular and General Genetics 143, 119-129.
Resnitzky, D., and A. Kimchi (1991). Deregulated c-myc expression abrogates the interferon and interleukin 6 mediated G0-G1 cell cycle arrest but not other inhibitory responses in M1 myeloblastic cells. Cell Growth and Differentiation 2, 33-42.
Richardson, C. L., A. D. Grant, and G. E. Schulman (1981). The interaction of caffeine and other xanthine analogs with DNA as measured by competitive fluorescence polarization. Environ Mutagen 3, 343.
Riddle, V. G. H., and D. M. Lehtomaki (1981) Growth arrest states of RNA virus transformed and chemically transformed mouse cells. Cancer Research 41, 1778-1783.
Rose, T. M., S. Tremblay, and E. W. Khandjian (1992). Repression of a G0 associated 65 kilodalton protein in actively proliferating and SV40-transformed mouse kidney cells. Biochem Cell Biology 70, 149-155.
Rossow, P. W., V. G. H. Riddle, and A. B. Pardee (1979). Synthesis of labile serum dependent protein in early G1 controls animal cell growth. Proceedings of the National Academy of Sciences 76, 4446-4450.
Rudolph, N. S., and S. A. Latt (1989). Flow cytometric analysis of X-ray sensitivity in ataxia telangiectasia. Mutation Research 211, 31-42.
Rydberg, B., and T. Lindahl (1980). Nonenzymatic methylation of DNA by the intracellular methyl group donor S-adenosyl-L-methionine is a potentially mutagenic reaction. The EMBO J. 1, 211-215.
Saul, R. L. and B. N. Ames (1985). Background levels of DNA damage in the population, in: Mechanisms of DNA Damage and Repair. (eds. M. G. Simic, L. Grossman and A. C. Upton) Plenum Press, New York, 529-535.
Schiano, M. A., B. U. Sevin, J. Perras, R. Ramos, and E. H. Wolloch (1991). In-vitro enhancement of cisplatinum antitumor activity by caffeine and pentoxifylline in a human ovarian cell line. Gynecol Oncol 43, 37-45.
Schiestl, R. H. and S. Prakash (1990). Rad10 an excision repair gene of Saccharomyces cerevisiae is involved in the RAD1 pathway of mitotic recombination. Molecular and Cellular Biology 10, 2485-2491.
Selby, C. P., and A. Sancar (1990). Transcription preferentially inhibits nucleotide excision repair of the template DNA strand in vitro. Journal of Biological Chemistry 265, 21330-21336.
Selby, C. P., and A. Sancar (1991). Gene-specific and strand-specific repair in-vitro partial purification of a transport repair coupling factor. Proceedings of the National Academy of Sciences U.S.A. 88, 8232-8238.
Selby, C. P., and A. Sancar (1993). Molecular mechanism of transcription-repair coupling. Science 260, 53-57.
Selby, C. P., E. M. Witkin, A. Sancar (1991). Eschericia coli MFD mutant deficient in mutation frequency decline lacks strand specific repair in vitro complemenion with purified coupling factor. Proceedings of the National Academy of Sciences U.S.A. 88, 11574-11579.
Sell, C., C. D. Chang, J. Koneicki, H. M. Chen, and R. Baserga (1992). A cryptopromoter is activated in the proliferating cell nuclear antigen gene of growth arrested cells. Journal of Cell Physiology 152, 177-184.
Setlow, R. B., P. A. Swenson, and W. L. Carrier (1963). Thymine dimers and inhibition of DNA synthesis by ultraviolet irradiation of cells. Science 142, 1464-1466.
Sgorbati, S. E. Sparvoli, M. Levi, M. Galli, S. Citterio, and D. Chiatante (1991). Cell cycle kinetic analysis with flow cytometry in pea root meristem synchronized with aphidicolin. Physiologia Plantarum 81, 507-512.
Shaffer, B. M. (1953). Aggregation in cellular slime molds: in vitro isolation of acrasin. Nature 171, 975-978.
Sharpe, P. T., G. M. Knight, and D. J. Watts (1984). Changes in the DNA content of Amoebae of Dictyostelium discoideum during growth and development. 217, 839-844.
Sharp, P. T., and D. J. Watts (1985). The role of the cell cycle in differentiation of the cellular slime mould Dictyostelium discoideum. Molecular and Cellular Biochemistry 67, 3-9.
Shaulsky, G., N. Goldfinger, A. Peled, and V. Rotter (1991). Involvement of wild-type P53 in pre-B-cell differentiation in-vitro. Proceedings of the National Academy of Sciences USA 88, 5567-5571.
Sheaff, R., D. Ilsley, and R. Kuchta (1991). Mechanism of DNA polymerase a inhibition by aphidicolin. Biochemistry 30, 8590-8597.
Shiomi, T. and K. Sato (1979). Isolation of UV-sensitive variants of human FL cells by a viral suicide method. Somatic Cell Genetics 5, 193-201.
Short, A. D., J. Bian, T. K. Ghosh, R. T. Waldron, S. L. Rybak, and D. L. Gill (1993). Intracellular calcium pool content is linked to control of cell growth. Proceedings of the National Academy of Science USA 90, 4986-4990.
Solberg, K. A., S. Ovrebo, R. K. Kleppe, and K. Kleppe (1978). Effect of caffeine on DNA polymerase I from Escherichia-coli studies in-vitro and in-vivo. Mutation Research 51, 1-10.
Soll, D. R., J. Yarger, and M. Mirick (1976). Stationary phase and the cell cycle of Dictyostelium discoideum in liquid nutrient medium. Journal of Cell Science 20, 513-523.
Sorenson, C. M., and A. Eastman (1988). Influence of cis diamminedichloroplatinum-II on DNA synthesis and cell cycle progression in excision repair proficient and deficient chinese hamster ovary cells. Cancer Research 48, 6703-6707.
Spadari, S., F. Sala, and G. P. Pedrali-Noy (1982). Aphidicolin: a specific inhibitor of nuclear DNA replication in eukaryotes. TIBS January Issue, 29-31.
Steinmeyer, K., H. Maacke, and W. Deppert (1990). Cell cycle control by P53 in normal 3T3 and chemically transformed Meth A mouse cells. Oncogene 5, 1691-1700.
Story, R. M., D. K. Bishop, N. Kleckner, and T. A. Steitz (1993). Structural relationships of bacterial Rec A proteins to recombination proteins from bacterophage T4 and yeast. Science 259, 1892-1896.
Sundin, O. and A. Varshausky (1981). Arrest of segregation leads to accumulation of highly intertwined catenated dimers, dissection of the final stages of SV-40 DNA replication. Cell 25 659-672.
Szymkowski, D. E., B. Kelly, and R. A. Deering (1989). A Dictyostelium discoideum cDNA coding for a protein with homology to the rat ribosomal protein L7. Nucleic Acids Research 17, 5393.
Terleth, C., R. Waters, J. Brouwer and P. van de Putte (1990). Differential repair of UV damage in Saccharomyces cerevisiae is cell cycle dependent. The EMBO Journal 9, 2899-2904.
Tsutsumi, A., J. Freire-Moar, and J. T. Ransom (1992). Transient down-regulation of PKC-zeta RNA following crosslinking of membrane IGM on WEHI-231 B lymphoma cells. Cell Immunology 142, 303-312.
Turchi, J. J., G. Siegal, and R. A. Bambara (1992). DNA helicase E and DNA polymerase epsilon functionally interact for displacement synthesis. Biochemistry 31, 9008-9015.
Wang, Z., X. Wu, and E. C. Friedberg (1993). DNA repair synthesis during base excision repair in-vitro is catalyzed by DNA polymerase epsilon and is influenced by DNA polymerases alpha and delta in saccharomyces cerevisiae. Molecular and Cellular Biology 13, 10051-1058.
Weijer, C. J., G. Duschl and C. N. David (1984a). A revision of the Dictyostelium discoideum cell-cycle. Journal of Cell Science 70, 111-132.
Weijer, C. J., G. Duschl and C. N. David (1984b). Dependence of cell-type proportining and sorting on cell cycle phase in Dictyostelium discoideum. Journal of Cell Science 70, 133-145.
Weijer, C. J. and M. Krefft (1989). Expression of a cell surface antigen in Dictyostelium discoideum in relation to the cell cycle. Journal of Cell Science 93, 199-204.
Weinert, T. A., and L. H. Hartwell (1988). The RAD9 gene controls the cell-cycle response to DNA damage in Saccharomyces cerevisiae. Science 24, 317-322.
Welker, D. L., and R. A. Deering (1976). Genetic analysis of radiation-sensitive mutations in the slime mould Dictyostelium discoideum. Journal of General Microbiology 97, 1-10.
Welker, D. L., and R. A. Deering (1979a). In vivo nicking and rejoining of nuclear DNA in ultraviolet-irradiated radiation-resistant sensitive strains of Dictyostelium discoideum. Molec. General Genetics 167, 259-263.
Welker, D. L., and R. A. Deering (1979b). Interactions between radiation-sensitive mutations in double-mutant haploids of Dictyostelium discoideum. Molec. Gen. Gen. 167, 265-270.
Welker, D. L., and K. L. Williams (1982). A genetic map of Dictyostelium discoideum based on mitotic recombination. Genetics 102, 691-710.
White, E., D. Scandella and E. Katz (1981). Inhibition by CIPC of mitosis and development in Dictyostelium discoideum and the isolation of resistant mutants. Developmental Gen. 2, 99-105.
Williams, K. L. (1980). Examination of the chromosomes of Polysphondylium pallidum following metaphase arrest by benzimidazole derivatives and cholchacine. J. of Gen. Micro. 116, 405-409.
Wittig, B., S. Wolfl, T. Dorbic, W. Varhson and A. Rich (1992). Transcription of human c-myc in permeabilized nuclei is associated with formation of Z-DNA in three discrete regions of the gene. The EMBO Journal 11, 4653-4663.
Yen, A., and V. G. H. Riddle (1979). Plasma and platelet associated factors act in G1 to maintain proliferation and to stabilize arrested cells in a viable quiescent state a temporal map of control points in the G1-phase. Experimental Cell Research 120, 349-358.
Yonish-Rouach, E., A. Kimchi, and M. Rubinstein (1991). The antiproliferative effect of cyclosporine in hematopoietic and lymphoblastoid cell lines common mechanistic elements with nterferon-alpha. Transplantation (Baltimore) 51, 1276-1272.
Zada-Hames, I. M., and J. M. Ashworth (1977). The cell-cycle and its relationship to development in Dictyostelium discoideum. Developmental Biology 63, 307-320.
Zada-Hames, I. M., and J. M. Ashworth (1978). The cell cycle during the vegetative stage of Dictyostelium discoideum and its response to temperature change. Journal of Cell Science 32, 1-20.
ã 1993-2007 Brooks John Kelly