Innovations. DNA detectives.

To understand the many potential causes and resulting consequences of DNA damage, scientists first need methods to detect it. Canadian scientists X. Chris Le and Michael Weinfeld, with help from U.S. molecular biologist Steven Leadon, developed a selective, sensitive technique for measuring DNA damage. The scientists combined a thymine glycol antibody with thymine glycol to selectively tag a specific type of DNA damage. They then added a second antibody with fluorescing properties, and used laser-induced fluorescence to identify the damaged portion of the tagged DNA. The fluorescence can be quantified, with higher levels of fluorescence indicating higher DNA damage. The technique was shown to find 1 damaged base in 1 billion normal bases. This level of sensitivity could allow the measurement of DNA damage resulting from clinical levels of radiation, and may allow scientists to establish a day-to-day baseline for DNA damage. From this baseline, it would be possible to ascertain the levels of damage that a cell can tolerate, as well as how much damaged it is capable of repairing on a daily basis.

therapeutic drugs in vivo.
MANY OF THE DRUGS in current use or in development for human cancer chemotherapy exhibit phase-specific cytotoxicity. In the characterization of these drugs, testing of cell-stage responses is normally carried out with cells grown and synchronized in vitro. While precise information can be obtained from these studies, they do not simulate the complex conditions existing in vivo, which may greatly affect drug action. It would be advantageous, therefore, to be able to measure drug action on synchronized target cells in the living animal.
Centrifugal elutriation has been demonstrated to be an effective and rapid method for the high-resolution bulk separation of viable mammalian cells (Meistrich et al., 1977a). Since separation is based on differences in cell size, this method is effective for synchronizing cells. In this way, populations of tumour cells, derived either from cultures or directly from dissociated solid tumours, have been successfully synchronized , 1978a: Meistrich et al., 1977bSuzuki et al., 1977). Hydroxyurea (HU) is a well studied phase-specific cytotoxic agent that preferentially kills S-phase cells (Bachetti & Whitmore, 1969;Kim et al., 1967;Sinclair, 1967). Its effectiveness has also been extensively demonstrated in vivo (Madoc-Jones & Mauro, 1970;Rajewsky, 1970;Stearns et al., 1963).
In this communication we describe the phase-specific cytotoxicity effect of hydroxyurea on synchronized murine fibrosarcoma cells arrested in the lungs of C3Hf/Bu mice after i.v. injection. The tumour cells were separated and synchronized by centrifugal elutriation and characterized with respect to cell-stage distribution by flow microfluorometry (FMF).
Tumour-cell suspension.-Source material was obtained from 6th-generation isotransplants of tumours stored in liquid N2. Singlecell suspensions were prepared by mincing and trypsinization, by a method described in detail elsewhere (Grdina et al., 1975). Cell viability was determined by phase-contrast microscopy and was found to be routinely >95%. To prepare tumour cells for centrifugal elutriation, 1-5x 107 were seeded into each of 20 32-oz glass culture bottles, and incubated at 37°C in a water-saturated atmosphere of 5% CO2 and air (Grdina et al., 1978a). A modified McCoy's 5A growth medium supplemented with 20% foetal calf serum was used (Humphrey et al., 1970). After 24 h incubation, the supernatant containing floating cells was discarded and 20 ml of fresh medium was added to each culture bottle. The attached cells were incubated for an additional 24 h.
Pulse labelling.-To monitor the effectiveness of centrifugal elutriation in separating DNA-synthesizing cells from an asynchronous population, 10 ,Ci of 3H-thymidine (1.9 Ci/mmol; Schwarz/Mann, Orangeburg, N. J.) were added to one of the culture bottles.
After 10 min incubation at 37°C, the uptake  (Shortman, 1973). With a rotor speed of 1525 rev/min, between 2x 108 and 3x 108 cells, suspended in 20 ml of medium, were introduced into the elutriator chamber at a flow rate of 5-4 ml/min. Throughout the separation the rotor speed was constant and the flow rates were varied by equal increments from 5-4 to 27-4 ml/min. Twelve fractions (f) were collected. With the exception of fl (70 ml), all fractions contained 50 ml. Control cell suspensions which were not separated were kept in ice during the elutriator run.
Cell counting and volume analysis.-Cell counts were made using a haemocytometer and/or a Coulter counter (Model ZBI; Coulter Electronics, Hialeah, FL) fitted with a 70,um-diam aperture. The volume distribution of cells was determined with a Coulter counter and a multichannel analyser (Channelyzer II; Coulter Electronics) and XY recorder. The modal cell volume was designated as the volume corresponding to the modal channel number of the volume distribution of each sample (Grdina et al., 1978a).
fluorometer with a laser wave-length setting of 457 ,um was used to measure the DNA content of individual cells in suspension (Steinkamp et al., 1973). Cells were stained with mithramycin (Grdina et al., 1978a). The resultant histograms of DNA fluorescence were computer analysed (Johnston et al., 1978).
Lung colony assay.-The colony-forming efficiency (CFE) of FSa cells was determined by a lung colony assay (Hill & Bush, 1969).
To maximize CFE, recipient mice, with their hind legs shielded, were whole-body irradiated with 1000 rad (Grdina et al., 1978b). After 24 h these mice were injected with 8 x 103 viable FSa cells from each of the elutriator fractions or an unseparated control population (USC), along with 106 heavily irradiated (HIR; 10,000 rad) FSa tumour cells. HIR cells were not separated by centrifugal elutriation. After 14 days the mice were killed, their lungs removed, the lobes separated and fixed in Bouin's solution, and the colonies of tumour cells counted.
Hydroxyurea testing in vivo.-Hydroxyurea (manufactured by Ben Venue Laboratories, Inc., Bedford, OH) was obtained from the Division of Cancer Treatment, National Cancer Institute, NIH, Bethesda, MD. HU was made up at 100 mg/ml in sterile water. Twenty minutes after the injection of viable synchronized FSa cells, half of the 20 animals from each group were injected i.p. with HU at a concentration of 1I0 mg/g body wt, a dose which is nontoxic to C3H mice (Withers et al., 1974). Under the conditions described, no drug toxicity was observed in the recipient animals.

RESULTS
The recovery of cells following centrifugal elutriation was routinely greater than 90%. In each of 3 experiments, with the exception of f 1, the viability of cells in each fraction exceeded 95%. Fl was discarded because it contained subcellular debris and damaged cells. Since fll and fl2 contained both large and small cells, as well as small clumps of cells washed out of the rotor at the end of the run, they were also discarded. Data from a representative experiment are presented in Fig. 1. The percentages of cells recovered in the various fractions are plotted along with their respective modal cell volumes. Starting with f3, the average cell volume increased with increasing fraction number from 1090 to 2100 ,um3. Also plotted in Fig. 1 is the percentage of radioactivity recovered in each fraction due to labelled cells incorporating [3H]TdR during the 10-min pulse before elutriation and the net ct/min per 106 cells per fraction. The maximum radioactivity was found in f7 and f8. DNA histograms illustrating the effectiveness of centrifugal elutriation in synchronizing tumour cells are presented in Fig. 2. The percentages of G1, 5, and G2+M, and the coefficients of variation (CV) of the G1 fluorescence peaks, as calculated from these histograms, are contained in the Table. Based on this analysis, f8 was found to contain the greatest percentage of cells with a DNA content characteristic of S phase. No peak of fluorescence representing G1 normal nontumour cells was observed in either the fractions of cells recovered after elutriation or in the unseparated control population,  lung-colony number was found in any of the elutriated control groups (see Fig. 3). FSa cells with modal volumes less than 800 ,um3 are known to be relatively deficient in clonogenic ability (Grdina et al., 1978b). HU reduced lung colonies to some since these cells are eliminated by the short-term culture conditions (Grdina et al., 1978b). The cell-killing effect of HU was tested in vivo on an unseparated control (USC) population and synchronized elutriated FSa populations lodged in the lungs of test animals. USC colony number was reduced by 35% after HU, agreeing well with the 33% S-phase cells determined by FMF analysis (see Table). With the exception of f2 cells, which were relatively small, i.e. with a modal volume of only 580 jUm3, no appreciable difference in degree in all but the groups of n with f2 and f3 cells. These each contained 9000 G1 cells. mum reduction of lung colon in mice injected with f 8, the found to contain the largest p S-phase cells (see Fig. 2). Percentages of surviving with the percentage of cells in phases of the cell cycle for elutriated fractions tested, are and presented in Fig. 4 for co DISCUSSION Centrifugal elutriation is effective method for synchroni cells. Relatively large numbers be separated on the basis of s loss of viability. To improve and to reduce significantly the presence of non-tumour cells in the suspension, FSa cells can be incubated for 48 h in vitro without any observable effect on CFE (Grdina et al., 1978a).
A lung colony assay with WBI mice and HIR cells was used to determine the CFE of various separated FSa populations. This method has 2 advantages: first, cells from the various elutriator fractions are equally retained in the lungs of recipient animals, and over 95%o remain in the lungs 20 min after injection; and second, the CFE is maximized and is independent of either cell size or position within the division cycle (Grdina et al., * GI 1978b). Untreated mice could, if necessary, * s be substituted for WBI animals in the A G2 IM assay if the drug to be studied were toxic to irradiated animals. Under this condi-/< tion, however, tumour-cell retention in the lungs and CFE would be reduced and would vary for cells from the different ..L...... elutriator fractions (Grdina et al., 1978b). The maxi-fractions. The reasons for the discrepancy -ies occurred in the determinations of S-phase cells in cell fraction flO is unclear. Since cells were fixed for ercentage of FMF analysis after elutriation, it may be that a fraction of the cells which took up cells, along label in late S progressed into G2 during the various the separation procedure. each of the HU was chosen for the study because its summarized action on S-phase cells has been extensively mparison.
characterized and documented (Bacchetti & Whitmore., 1969;Pfeiffer & Tolmach, 1967;Sinclair, 1967). It has also been used effectively in vivo to determine the propora fast and tion of clonogenic cells in S phase in solid zing tumour tumours, and has been suggested as more 3 of cells can effective for this type of determination ,ize, without than even suicide-labelling with [3H]synchrony TdR (Rockwell et al., 1976). Following i.p. injection, HU essentially disappears from the femoral marrow within 3 h (Madoc-Jones & Mauro, 1970). During this period of time, cell killing is limited to S-phase cells, and cells in G1 are prevented from entering S. A strong correlation between cell killing by HU and relative percentage of cells in S in the elutriated cell fractions, as determined by FMF analysis, is evident (see Fig. 4). These data are also in good agreement with data generated from studies performed in vitro on Chinese hamster ovary cells synchronized by centrifugal elutriation (Meyn et al., unpublished).
Finally, while HU was administered 20 min after tumour-cell injection in these experiments the timing of drug administration could be adjusted to accommodate drugs requiring bioactivation in vivo. The requirement of the assay is simply that the synchronized tumour cells arrested in the lungs be exposed to the active cytotoxic agent before synchrony is lost due to cell progression.
We have described the effect of an S-phase-specific cytotoxic agent in vivo on populations of synchronized tumour cells lodged in the lungs of test animals. Previous studies involving the characterization of chemotherapeutic agents for the treatment of malignant diseases have been limited to either in vitro or complex and heterogeneous in vivo systems. Using the method characterized with HU and described here, it is possible to study and evaluate in a relatively short time the effectiveness of chemotherapeutic agents in vivo against not only G1, S, and/or G2 populations of tumour cells, but also against selected normal cell populations in the tumour-bearing animals, such as marrow. In this manner, drugs can be characterized in vivo with respect to their phase specificity (or lack of it) in cell killing and in toxicity to the host animal.