To evaluate a possible DNA damage in control and treated cells, analyses on several levels of DNA organization were carried out in all the considered cell lines: nuclei, metaphases and cellular structures recovered from the culture medium.

Chromosomal suspensions and nuclei of clone-7 astrocytes, isolated in our laboratory, were subjected to the TUNEL assay (Fig. 1a–g). In this cell line, DNA fragmentation was only found at the nuclear level, while no metaphase chromosomes with fragmentation points were observed.

Fluorescent TUNEL assay in astrocytes. Images of representative fields showing nuclei from control and treated cells. DNA fragmentation is shown (a1–f1). Nuclei marked with propidium iodide (a2–f2). Merging of both signals (a3–f3 and g). Control cells are shown(a1–a3). RSV-treated cells are shown (b1–b3). Doxo-treated cells are shown (c1–c3). Doxo + RSV-treated cells (d1–d3). The following controls are shown: positive control (e1–e3), negative control (f1–f3). Enlargement (g) of a particular of control preparation (a3). White arrows indicate two fragmented pyknotic nuclei. Scale bar, 20 μm. Histogram showing TUNEL-positive nuclei, percentage of the total cells (h). Histograms showing the relative area in unfragmented or fragmented nuclei of control (i) and RSV-treated cells (j). Doxo-treated cells (k). Doxo + RSV-treated cells (l). Experiments were performed in triplicate and data are expressed as means ± standard deviation (n = 3 ± SD); there were 100 analyzed cells in each experimental group. Asterisks indicate responses significantly different from those of the controls (h) and significantly different in fragmented versus unfragmented samples (i–l) (*p < 0.05)

An aliquot of 36% of fragmented nuclei was observed in control cells (Fig. 1h).

This percentage decreased to 16% in RSV-treated cells, and it was established to 36% after Doxo treatment. RSV + Doxo cotreatments slightly increased the value to 39%.

In this cell line, a peculiarity about the area of fragmented nuclei was observed. In both the controls and all the treated cells, DNA fragmentation concerned only small-sized nuclei (pyknosis-like nuclei), while all other normal sized nuclei showed unfragmented DNA (Fig. 1g).

The differences between an unfragmented nuclei area and fragmented nuclei area were found significant in all the samples.

The above qualitative and quantitative analyses suggest that in these cells, DNA fragmentation could be related to apoptotic phenomena since the DNA fragmentation signal did not concern the replicating cells (no metaphase chromosomes showed fragmentation points) and all nuclei with characteristics of pyknosis showed fragmented DNA.

Qualitatively, we observed that the culture medium was rich in suspended cellular structures containing DNA. These showed partially fragmented nuclei, as in control cells (Fig. 2a1–a3) and in RSV + Doxo-treated cells (Fig. 2d1–d3). DNA fragmentation and partitioning into small spherical structures were found in RSV-treated cells (Fig. 2b1–b3) and Doxo-treated cells (Fig. 2c1–c3).

Fluorescent TUNEL assay of detached cellular structures, recovered from astrocytes culture medium. Images of representative fields showing fragmented DNA in control and treated cells. DNA fragmentation (a1–d1). DNA marked with propidium iodide (a2–d2). Merging of both signals (a3–d3). Control samples are shown(a1–a3). RSV-treated samples (b1–b3). Doxo-treated samples (c1–c3). Doxo + RSV-treated samples (d1–d3). White arrows indicate spherical corpuscles containing fragmented DNA. Scale bar, 20 μm. Histogram showing TUNEL-positive nuclei, percentage of total cells. Experiments were performed in triplicate and data are expressed as mean ± standard deviation (n = 3 ± SD); there are 100 analyzed cells in each experimental group . Asterisks indicate responses that are significantly different from those of the controls (*p < 0.05)

For this detached population, we detected high fragmentation rates, about 33% in controls, 90% in RSV-treated cells, 100% in Doxo-treated cells and 89% in RSV + Doxo-treated cells. However, in all treatments, the amount of detached cellular structures with fragmented DNA was always greater and with a statistically significant difference compared with control cells.

A completely different scenario was observed in Caco-2 cells. A peculiar feature was that DNA fragmentation affected chromosomes, uncondensed chromatin nuclei, and, only in treated cells, detached cellular structures.

In control cells, we observed several metaphases rich in fragmented chromosomes and nuclei with uncondensed chromatin but reporting defined points of fragmentation (Fig. 3a1–a3 and g).

Fluorescent TUNEL assay in Caco-2 cells. Images of representative fields showing nuclei and metaphases from control and treated cells. DNA fragmentation is shown (a1–f1). DNA marked with propidium iodide (a2–f2). Merging of both signals (a3–f3 and g). Control cells are shown (a1–a3). RSV-treated cells (b1–b3). Doxo-treated cells (c1–c3). Doxo + RSV-treated cells (d1–d3). Positive control (e1–e3) and negative control (f1–f3) are shown. Enlargement of a representative fragmented metaphase of control cells (g). Scale bar, 20 μm. Histogram showing TUNEL-positive nuclei, percentage of total cells (h). Histograms showing the relative area in unfragmented or fragmented nuclei of control (i); RSV-treated cells (j). Doxo-treated cells (k). Doxo + RSV-treated cells (l). Experiments were performed in triplicate and data are expressed as mean ± standard deviation (n = 3 ± SD). There are 100 analyzed cells in each experimental group. Asterisks indicate responses significantly different from those of the controls (h), and which are significantly different in fragmented versus fragmented samples (i–l) (*p < 0.05)

In RSV-treated cells, on the other hand, no DNA fragmentation signals were found in cells adherent to the substrate (Figs. 3b1–b3). In Doxo-treated cells, many nuclei containing fragmented DNA were found with no distinction between normal or pyknotic nuclei. Specifically, the nuclear fragmentation did not concern the whole area but only some points that appeared as fragmented spots (Figs. 3c1–c3). Cells exposed to the combined treatment (RSV + Doxo), however, showed nuclei with more diffuse fragmented DNA (Figs. 3d1–d3).

As reported in the histogram (Fig. 3h), a high percentage of nuclei containing fragmented DNA (66%) was found only in Doxo-treated cells, for which, compared to control, a statistically significant difference was found.

Analyzing the area of fragmented and unfragmented nuclei, no statistically significant difference was found between the two parameters (Figs. 3i–l), except for Doxo-treated cells and excluding RSV-treated cells, which had no nuclei containing fragmented DNA. Consequently, since metaphase chromosomes represent a clear sign of cell replication, the presence of DNA fragmentation, also in chromosomes, suggested an unlikely relationship with the ongoing death phenomena.

In parallel, an analysis on detached cellular structures was performed. No fragmented or unfragmented nuclear structures were detected in the medium from control cells (Figs. 4a1–a3). Large clusters of nuclei with a high number of fragmented DNA spots were found in the medium of RSV-treated cells (Figs. 4b1–b3). Groups of nuclei with a lower fragmentation rate than the RSV-treated ones were identified in the culture medium recovered from Doxo-treated cells (Figs. 4c1–c3). A high number of nuclei was found in the culture medium of cells treated with RSV + Doxo. All these nuclei exhibited diffuse fragmentation, albeit with a very slight intensity (Figs. 4d1–d3).

Fluorescent TUNEL assay in detached cellular structures, recovered from Caco-2 cells’ culture medium. Images of representative fields showing fragmented DNA in control and treated cells. DNA fragmentation is shown (a1–d1). DNA stained with propidium iodide (a2–d2). Merging of both signals (a3–d3). Control samples are shown (a1–a3). RSV-treated samples (b1–b3). Doxo-treated samples (c1–c3). Doxo + RSV-treated samples (d1–d3). Scale bar, 20 μm. Histogram showing TUNEL-positive nuclei, percentage of total cells. Experiments were performed in triplicate and data are expressed as mean ± standard deviation (n = 3 ± SD). There are 100 analyzed cells in each experimental group. Asterisks indicate responses that are significantly different from those of the controls (*p < 0.05)

Considering the DNA damage, MDA–MB–231 cells showed a different behavior compared with the other two cell lines. The control cells did not show signals of DNA fragmentation, neither at the nuclei level nor at the chromosomes level (Fig. 5a1–a3). Following treatment with RSV, many nuclei containing fragmented DNA were observed (Fig. 5b1–b3). The nuclei fragmentation rate was increased following treatment with Doxo (Fig. 5c1–c3) and greatly reduced in RSV + Doxo cotreated cells. Surprisingly, only 1% of the analyzed metaphases showed some fragmented chromosomes (Fig. 5g). The increase in nuclear fragmentation reached statistical significance, especially for cells exposed to RSV or Doxo (Fig. 5h).

Fluorescent TUNEL assay in MDA-MB-231 cells. Images of representative fields showing nuclei and metaphases from control and treated cells. DNA fragmentation is shown (a1–f1). DNA stained with propidium iodide (a2–f2). Merging of both signals (a3–f3 and g). Control cells are shown(a1–a3). RSV-treated cells (b1–b3). Doxo-treated cells (c1–c3). Doxo + RSV-treated cells (d1–d3). Positive controls (e1–e3) and negative controls (f1–f3) are shown. Enlargements of a representative fragmented metaphase of control cells (g). White arrows indicate chromosomes with fragmented points. Scale bar, 20 μm. Histogram showing the percentage of fragmented nuclei (h). Histogram showing TUNEL-positive nuclei, percentage of total cells (i); RSV-treated cells (j). Doxo-treated cells (k). Doxo + RSV-treated cells (l). Experiments were performed in triplicate and data are expressed as mean ± standard deviation (n = 3 ± SD). There are 100 analyzed cells in each experimental group. Asterisks indicate responses significantly different from those of the controls (h) and that are significantly different in fragmented versus unfragmented samples (i–l) (*p < 0.05)

A relationship was observed between the reduction of the nuclei area and the DNA fragmentation. As reported in the histograms (Fig. 5i–l), the differences between unfragmented nuclei area and fragmented nuclei areas in all the samples were found to be significant.

These data suggest that in MDA–MB–231 cells, DNA fragmentation could be related to apoptotic phenomena, although some cells showed fragmented chromosomes, indicative of DNA damage in replicating cells.

Except for control cells, numerous cellular structures (apoptotic body-like corpuscles) were observed in all treatments in the suspended cell population belonging to the MDA–MB–231 cell line (Fig. 6).

Fluorescent TUNEL assay of detached cellular structures, recovered from MDA-MB-231 medium culture. Images of representative fields showing fragmented DNA from control and treated cells. DNA fragmentation is shown (a1–d1). DNA stained with propidium iodide (a2–d2). Merging of both signals (a3–d3). Control samples are shown (a1–a3). RSV-treated samples (b1–b3). Doxo-treated samples (c1–c3). Doxo + RSV-treated samples (d1–d3). Scale bar, 20 μm. Histogram showing TUNEL-positive nuclei, percentage of total cells. Experiments were performed in triplicate and data are expressed as mean ± standard deviation (n = 3 ± SD). There are 100 analyzed cells in each experimental group. Asterisks indicate responses that are significantly different from those of the controls (*p < 0.05)

As shown in the histogram, 100% of fragmented spots for RSV-treated cells, 58% for Doxo-treated cells, and 88% for combined RSV + Doxo-treated cells was observed.

Since the culture medium recovered from RSV-treated cells showed a greater variety of detached cellular structures, a more detailed analysis was performed.

In all cases, a breakdown of fragmented DNA within the nuclear structure was observed, highlighting the simultaneous presence of intact DNA and fragmented DNA. The latter appeared in spherical body structures clearly detached from the portions of unfragmented DNA (Figs. 7a1–d3).

Fluorescent TUNEL assay of detached cellular structures, recovered from MDA-MB-231 cells’ culture medium. Images of representative fields showing fragmented DNA from RSV-treated cells. DNA fragmentation is shown (a1–d1). DNA stained with propidium iodide (a2–d2). Merging of both signals (a3–d3). Nuclear structure with unfragmented central body and fragmented peripheral DNA islands (a1–a3). Nuclear structure with fragmented central body and two agglomerates of unfragmented peripheral DNA (b1–b3). Nuclear structure divided into two compartments of fragmented and unfragmented DNA (c1–c3). Nuclear structure was divided into three compartments, one with unfragmented DNA and two with fragmented DNA (d1–d3). Scale bar, 5 μm

By considering these morphological aspects, it seems that this cell line is very active in eliminating cells with fragmented DNA (which we found in detached cellular structures), but at the same time, although in a small percentage of the replicating cells showed fragmented DNA as chromosomal evidence.