Nanotechnology: thinking small.

Nanotechnology--building devices on the atomic scale--may unleash some big scientific advances early in the new millennium. Last January in Arlington, Virginia, nearly 100 representatives from academia, industry, and government laid out the general goals for the next decade of nanotechnology research by U.S. government agencies. Some predict that the potentially rich opportunities in this field may trigger a nanotechnology initiative in the federal budget request for Fiscal Year 2001. In the meantime, Congress has asked the NIEHS to explore how nanotechnology might be used to address environmental heath problems.

A PREVIOUS report from our laboratory (Wibe et al., 1979) describes the inhibition of synchronized NHIK 3025 cells by treatment with NY 3170. Data demonstrating the influence of NY 3170 on the traverse of cells through the cell cycle were presented. The metaphase-arresting properties were examined in detail.
The present investigation demonstrates the inactivating effects of NY 3170 on NHIK 3025 cells, with special attention to the age response. The results are compared with the previously reported (Wibe et al., 1979) cell-cycle inhibition by this drug.

MATERIALS AND METHODS
Cell culture.-Information on the human cell line NHIK 3025 and the routines followed in handling stock cultures, as well as the chemical structure and the origin of the mitotic inhibitor, has been reported previously (Wibe et al., 1979;Gacek et al., 1979).
Inactivation of single cells was measured as loss of the ability to form macroscopic colonies after 10-12 days of incubation. The medium was always changed 6-7 days after plating. The colonies were fixed in absolute ethanol and stained with methylene blue. Colonies containing more than 40 cells were scored for calculating surviving fractions. Experiments in our laboratory have shown that the plating efficiency of NHIK 3025 cells in the medium used (E2a supplied with 20% human serum and 10% horse serum) is 85-100%, irrespective of cell density. However, when the number of surviving (colonyforming) cells per dish exceeds 300, counting will be inaccurate owing to overlapping colonies. Therefore, the number of cells plated in each experimental group was kept at a level entailing less than 300 viable cells after treatment.
After trypsinization and plating of NHIK * Fellow of the Norwegian Cancer Society Landsforeningen mot Kreft. 3025 cells in Petri dishes there is a lh lag before cell proliferation re-starts. In the experiments described here. the plated cells wTere allowed to attach for 2 h before treatmeint. Consequently, the cells were in exponential growth Awhen treatment writh NY 3170 began.
The population-doubling time of NHIK 3025 cells plated in Petri dishes w-as estimated as 17-18 h from a growNt,h curve recorded. This fully agrees with the, values for mean generation time of NHIK 3025 cells measured in asynchronous populations cultivated in tissue-culture flasks (Pettersen et al., 1977) and in populations synchronized by mitotic selection (Pettersen et al., 1977;Wibe et al.. 1979, and present paper).
Continuous exposure to N I' 3170. Asynclironously growing cells were trypsinized and gently agitated wTith a pipette to obtain a single-cell suspension. The cells w-ere plated in Petri dishes (200-1000 cells/dish) previously filled w-ith medium. After the attachment period (2 h) the control medium was replaced by medium containing the concentration of NY 3170 to be examined, in whiclh the cells were tested for survival for the required incubation tinme (12 days). l1ariation of exposure time. Plating of single cells (200-500 cells/dish) and addition of NY 3170 was as described above. At set times, the medium containing NY 3170 was remnoved, the dishes w!ere rinsed x 3 wA-ith Hanks' solution, and control medium wias added to allow grow-th into colonies of surviving cells.  (Wibe et (ti., 1979). Accumulation of mitotic cells was allow-ed for 2-5 h. after which the accumulated mitotic cells wA-ere shaken off using a reciprocal shaker (Pettersen et al.. 1977) and removed. Then fresh mediunm containing NY 3170 was added. This shaking procedure Aw-as repeated sshortly afterwards to ensure that no metaphase-arrested cells w-ere left in the flask. Half an hour later the cells newly arrested in metaphase Ax-ere shaken off and transferredl to another culture flask, in which the cells were incubated in the presence of NY 3170 until plating. The arrested cells did not attach to the bottom of this flask.
At set timnes, a certain amount of cell sus-pension w\>as transferred to a centrifuge tube.
The medium containing NY 3170 was removed bv centrifugation, and the cells wN-ere plated for colony formation in Petri dishes (250 cells/dish) previously filled wAith control medium. The first sample of arrested cells was plated immediately after shake-off. The number of cells plated wN-as determined by haemacytometer counting. The wNhole experiment wNas performed in a 37°C room.
Age response studies. A synchronized populationi of cells was obtained by mitotic siiake-off. After diluting + ith medium to an appropriate concentration, the new-ly selected cells wvere plated in 25 cm2 Nunclon (A/S NUNC. Roskilde, Denmark) flasks (250 cells/ flask). Every 2 h the medium in 2 parallel flasks wN-as replaced by medium containing 2mM NY 3170. The medium in a third flask w%,as simultaneously replaced by fresh mediumn as a control. After the appointed time (1 or 3 h) the 3 flasks wN-ere rinsed x 3 wN-itlh Hanks' solution before addition of control medium for coloniy formation. The variation in the plating efficiency in the control flasks was less than 15%.
The onset and duration of the differenit phases of the cell cycle w%as determined by pulsed incorporation of [3H]thymidine, and registration of time of entry into mitosis as described in a previous report (XWibe et al.. 1979). The generation time wAas about 17 h. The experiments wsere performed in a, 37°C room.
To obtain single-cell surviving fractions, cell multiplicity wNas corrected for by means of the followNing formula (Gillespie et al., 1975) valid for populations of singlets and doublets: f= (N-(X2 -4S (N1T 1))1/2)/2 (N-1) f-=single-cell surviving fraction S = microcolony surviving fraction N =mean multiplicity. Fig. 1 shows the fraction of NHIK 3025 cells forming macroscopic colonies after continuous exposure to different concentrations of NY 3170 for 12 days. The highest concentration of NY 3170 allowing some colony formation after continuous exposure, wvas OlmM. At this concentration the spread in the colony size even of surviving colonies was great, and very few colonies reached the average The survival of asynchronously growing NHIK 3025 cells exposed to NY 3170 for different times is shown in Fig. 2 at a concentration of 01 mM had little effect on the survival of NHIIK 3025 cells when the drug was removed after 24 h, which is more than one generation time.

RES ULTS
From Fig. 1 it can be seen that a small fraction of the cells could form colonies in the presence of OlmM NY 3170, even when continuously exposed for 12 days. When asynchronous populations were exposed to 2mM NY 3170, most of the cells were inactivated after relatively short exposure times. This demonstrates an effect of exposure time per se, and not only inactivation of a particular stage of the cell cycle.
In these experiments, and in those to be described, where the exposure time to various concentrations of NY 3170 was 1-10 h, surviving cells formed colonies of normal control size, suggesting that they had managed to recover completely from (Surviving fractions after exposures beginning 18 or 20 h after selection were not corrected for a multiplicity higher than 2). Approximate distribution of cells among the different phases is indicated at the top. the damage caused by a relatively short exposure to NY 3170. Fig. 3 demonstrates the fractions of I NHIBITION: Moderate prolongation of mitosis.
metaphase-arrested NHIK 3025 populations surviving exposures to 0-2 or 0 4 mm NY 3170, when the exposure time was varied. NY 3170 was added in early G2 (3 h before metaphase) and removed at different times after the moment of entry into metaphase arrest. The reversibility of metaphase arrest caused by NY 3170 was highly dependent on the duration of the arrest and on drug concentration. When the metaphase-arresting agent was immediately removed, the daughter cells were viable. Consequently, metaphase arrest was reversible if the drug was removed shortly after the onset of mitosis. However, damage was irreparable when the metaphase arrest lasted for several hours. When 0 4mM NY 3170 was present for 8 h during mitosis, not a single surviving cell could be seen in the dishes. In Fig. 4 age response curves are shown. When a synchronous cell population was exposed to 2mM NY 3170 for 1 h at different stages of the cell cycle, a lethal effect was found for cells exposed in or close to mitosis. However, a 3h exposure to 2mM NY 3170 induced lethal effects on cells in late S and G2, in addition to mitosis. This emphasizes once again that exposure time is an important parameter for the lethal effects of this drug.  (Wibe et al., 1979)  From comparison of the steepness of the survival curves of asynchronously growing cells in Fig. 2 and of metaphase-arrested cells in Fig. 3 (0.4mM) it seems that mitotic cells are particularly sensitive. This was also the impression from the time-lapse experiments reported in our previous paper (Wibe et al., 1979). At 0 4mM NY 3170, cells burst when accumulated in mitosis, while cells which were severely delayed in interphase and did not reach mitosis, did not disintegrate during the time of filming.
The shape of the age response curves (Fig. 4) also confirms selective inactivation of mitotic cells. Previous experiments (Wibe et al., 1979) have shown that the presence of NY 3170 (0-2mM) during mitosis is a necessary and sufficient condition for metaphase arrest. Thus, both inactivation and metaphase arrest are primarily induced in mitosis. This suggests a connection between the mitotic inhibitory and the inactivating effects of NY 3170.
To facilitate a general survey of the effects of NY 3170 on NHIK 3025 cells, data on inactivation in this report and data on cell-cycle inhibition in our previous report (Wibe et al., 1979) are summarized in Fig. 5.
Reversibility of metaphase arrest caused by treatment with a mitotic inhibitor may be measured in two different ways: (1) The ability of arrested cells to escape from metaphase after removal of the drug.
(2) The ability of arrested cells to form macroscopic colonies after removal of the drug. If the metaphase arrest is found reversible after the second criterion, the first criterion for reversibility is obviously fulfilled too. The following reported results show that statements on the reversibility of mitotic inhibitors are dependent on the techniques used.
Metaphase arrest in Earle's L cells by treatment with vinblastine is reported by Krishan (1968) to be reversible, as measured by the ability to escape mitotic arrest after removal of the drug. However, multipolar chromosome formations, multipolar divisions, and aberrant cytokinesis were seen in many cells released from the mitotic arrest. Bruchovsky et al. (1965) (Earle's L cells) and Mauro & Madoc-Jones (1970) (HeLa cells) reported extensive loss of colony-forming ability for cells exposed to vinblastine during mitosis. George et al. (1965) reported that HeLa cells exposed to 0-1 jig/ml vincristine were irreversibly arrested in mitosis despite repeated washing of the cells with medium. Irreversible metaphase arrest after treatment with vincristine (0.016 Htg/ml) was also indicated by results obtained in our laboratory on NHIK 3025 cells. In this experiment, however, the cells were only followed for 90 min after removal of vincristine (Dahl et al., 1976). Observations reported by Mauro & Madoc-Jones (1970) indicate that mitotic HeLa cells exposed to 0 1 utg/ml vincristine for 3 h are relatively resistant in terms of colony-forming ability. However, when HeLa cells were exposed to 0-1 jug/ml vincristine in S, these workers also observed irreversible metaphase arrest when the cells reached mitosis (Madoc-Jones & Mauro, 1968). In the present study, reversibility was measured by colony-forming ability. The curves in Fig. 3 demonstrate that both exposure time in mitosis and drug concentration are relevant to the reversibility of metaphase arrest caused by NY 3170. In these experiments the cells were exposed to NY 3170 in G2 (from 3 h before shake-off, see Materials and Methods). This G2 exposure per se seemed not to affect the survival of the cells, because all the arrested cells survived if the drug was removed immediately after the onset of mitosis.
The age response of NHIK 3025 cells treated with NY 3170 is similar to the age response reported by Barranco & Humphrey (1971) for CHO cells after treatment with the anticancer drug bleomycin. The only difference was a greater sensitivity in early S than in late S after bleomycin administration.
Madoc-Jones & Mauro (1968) indicated mitosis (S) and early G1 as stages sensitive to a high concentration of vinblastine, whilst S was the most sensitive stage when HeLa or Chinese hamster cells were treated with vincristine. For vinblastine, correlation between cytotoxicity and mitotic-spindle dissolution in proliferating Chinese hamster fibroblasts has recently been demonstrated (Tucker et al., 1977). However, the principle behind the oncolytic effect of the Vinca alkaloids is unknown (Marsden, 1972). The different shapes of the age response curves indicate that the inactivating mechanism of NY 3170 is different from those initiated by treatment with the established mitotic inhibitors vincristine and vinblastine.