Chromatin supraorganization, mitotic abnormalities and proliferation in cells with increased or down-regulated lox expression: Indirect evidence of a LOX–histone H1 interaction in vivo

Micron 42 (2011) 8–16

Contents lists available at ScienceDirect

Micron journal homepage: www.elsevier.com/locate/micron

Chromatin supraorganization, mitotic abnormalities and proliferation in cells with increased or down-regulated lox expression: Indirect evidence of a LOX–histone H1 interaction in vivo Maria Luiza S. Mello a,∗ , Elenice M. Alvarenga a , Benedicto de Campos Vidal a , Armando Di Donato b a b

Department of Anatomy, Cell Biology and Physiology, Institute of Biology, University of Campinas (UNICAMP), 13083-683 Campinas, SP, Brazil Department of Nephrology, Institute G. Gaslini, Largo G. Gaslini 5, 16147 Genoa, Italy

a r t i c l e

i n f o

Article history: Received 22 June 2010 Received in revised form 2 September 2010 Accepted 3 September 2010 Keywords: Lox Histone H1 Chromatin supraorganization Chromosome abnormalities Cell proliferation Image analysis

a b s t r a c t Lysyl oxidases (LOXs) are enzymes that permit the covalent crosslinking of the component chains of collagen and elastin. These enzymes are present inside the nuclei of certain mammalian cells. Previous studies have proposed LOX binding to histone H1 in vitro, and histone H1 is known to control global chromatin compaction and mitotic chromosome architecture. Therefore, in the present study, we analyzed chromatin supraorganizational changes, mitotic abnormalities, mitotic indices and cell death ratios in COS-7 and NRK-49F cells with high and low lox expression levels, respectively. The objective was to support biochemical data of LOX–H1 interaction, by providing evidence of chromatin remodeling in vivo, under different lox expressions. Chromatin decondensation assessed by image analysis was observed in COS-7 cells with increased lox expression. This decondensation is suggested to be promoted by LOX actions on histone H1, which loosens the DNA–H1 complex. In NRK-49F cells transfected with antisense lox or subjected to treatment with beta-aminopropionitrile (BAPN), chromatin condensation and nuclear phenotypic variability were found, which may be due to reduced LOX–H1 interaction. When lox expression was increased in COS-7 cells, the frequency of irregular chromosome plates was not affected, but cell proliferation decreased and “cell death preceded by multinucleation” increased. In NRK-49F cells there was accelerated proliferation induced by transfection with the antisense lox, and confirmed when cells were treated with BAPN. Apoptosis increased in NRK-49F cells only with BAPN treatment whereas cell death preceded by multinucleation increased only after antisense lox transfection. The data presented herein regarding chromatin remodeling indirectly support the hypothesis that LOX binds to histone H1 in vivo. Cell proliferation in COS-7 and NRK-49F cells and cell death at least in COS-7 cells agree with predicted effects of LOX interference in these processes. © 2010 Elsevier Ltd. All rights reserved.

1. Introduction Lysyl oxidases (LOXs) are enzymes that oxidize the side chain of peptidyl-lysine, converting specific lysine residues to residues of ␣-aminoadipic-␦-semialdehyde. This conversion permits the covalent crosslinking of the component chains of collagen and elastin (Pinnel and Martin, 1968; Kagan and Trackman, 1991; Lucero and Kagan, 2006) and accounts for the stabilization and insolubilization of the fibrous deposits of collagen and elastin in the extracellular matrix. In addition, it may also affect intracellular signal responses involved in tissue development, cell proliferation, cell adhesion, cell migration, and tumorigenesis modulation (Contente et al., 1990;

∗ Corresponding author. Tel.: +55 19 3521 6122; fax: +55 19 3521 6185. E-mail address: [email protected] (M.L.S. Mello). 0968-4328/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.micron.2010.09.001

Mello et al., 1995; Di Donato et al., 1997; Ren et al., 1998; Giampuzzi et al., 2001; Kirschmann et al., 2002; Li et al., 2003; Kaneda et al., 2004; Lucero and Kagan, 2006; Wu et al., 2007). In addition to their effects on the extracellular matrix, certain LOX proteins have also been shown to be present inside the nuclei of fibrogenic and aortic smooth-muscle cells (Li et al., 1997), and mature LOX epitopes have been found in the nuclei of osteoblastic cells (Guo et al., 2007). A 50 kDa LOX propeptide has been recognized by antibodies in nuclei of nontransformed and rastransformed NIH 3T3 cells but not in nuclei of ras-transformed cells (Contente et al., 2009). Homologies between lysine-rich regions of tropoelastin and histone H1 have been reported (Giampuzzi et al., 2003b). Histone H1 has been shown to be a substrate for LOX in vitro (Kagan et al., 1983; Giampuzzi et al., 2003b), although histone H2B may also bind LOX, possibly allowing the carboxy-catalytic region to be available to interact with H1 (Giampuzzi et al., 2003b).

M.L.S. Mello et al. / Micron 42 (2011) 8–16

Histone H1 is involved in the control of global chromatin compaction and the expression of selected genes (Karymov et al., 2001; Fan et al., 2005; Hizume et al., 2005; Happel and Doenecke, 2009). The loss of ␧-amino groups as a consequence of oxidative deamination induced by LOX or non-enzymatic oxidants would be expected to affect DNA–histone interactions and change chromatin condensation (Lucero and Kagan, 2006). Indeed, the introduction of antisense lox cDNA into non-tumorigenic revertants of rastransformed NIH 3T3 cells (RS485 cells) makes chromatin more condensed, concomitant with the recovery of ras oncogene expression (Mello et al., 1995). However, additional studies to determine the effects of LOX on chromatin are required (Lucero and Kagan, 2006). In the present study, chromatin supraorganization was studied using image analysis in interphase cells with high and low expression levels of a lox gene. Results are expected to support previous studies by Giampuzzi et al. (2003b) thus possibly correlating lox-induced histone H1 oxidation with chromatin decondensation in vivo. Analysis of mitotic abnormalities, micronucleus frequency, and mitotic indices was also conducted to obtain additional information on the effects of lox expression on chromatin in dividing cells, because histone H1 is required for the mitotic chromosome architecture of vertebrates (Maresca et al., 2005). In addition, data on cell death ratios were also obtained.

2. Materials and methods 2.1. Cells SV-40-transformed African green monkey kidney fibroblast COS-7 and rat kidney fibroblast NRK-49F cell lines were used. Cells were cultivated as previously reported (Giampuzzi et al., 2000, 2001). Briefly, the cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 1% glutamine, 1% nonessential aminoacids, 100 ␮g/mL streptomycin, and 100 units/mL penicillin under a humidified atmosphere of 5% CO2 at 37 ◦ C. To synchronize the cell cycle, cells were starved before growth on the slides. COS-7 cells were transiently transfected with the pSG5-LOX vector, a pSG5 plasmid carrying high levels of LOX sequences (Giampuzzi et al., 2000). In addition, as a control, COS-7 cells were transfected with empty pSG5 plasmid. NRK-49F cells were stably transfected with the pCL03 vector, a pCDNA3 plasmid carrying fragments of the mouse LOX coding sequence. These fragments ranged from −33 to +985 and were inserted in the antisense orientation (Contente et al., 1993) by subcloning in Kpn I and Xba I restriction sites (Giampuzzi et al., 2001). As a control, NRK-49F cells were transfected with the pCDNA3 empty plasmid only. The analysis was performed 48 h after transfection, when the expression of the recombinant protein was at its maximum level, as tested by Western blotting in time-course experiments. The NRK49F antisense-LOX were stably transfected and therefore previously clone-selected by G418 treatment. The cells were plated on growth to subconfluency in 100 mm dish plates. Then the cells were detached by trypsin treatment and collected. After washing in their growth medium, about 20,000 cells were plated on SlideFlasks (Nunc, InterMed). Finally, the cells were allowed to grow for 24–48 h to reach subconfluency. Where indicated ␤-aminopropionitrile (BAPN) a LOX inhibitor (Tang et al., 1989), was added to the growth medium to a final concentration of 0.1 mM. All the preparations were washed twice in PBS and fixed in a mixture of absolute ethanol–glacial acetic acid (3:1, v/v) for 1 min, rinsed in 70% ethanol for 5 min, and air-dried.

9

2.2. Western blot analysis Total cell lysates were prepared in RIPA buffer (25 mM Tris–HCl, pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate and 0.1% sodium dodecyl sulfate), containing a protease cocktail and the phosphatase inhibitor I and II cocktails (Sigma, MO, USA). The lysates were cleared by 30 min centrifugation at 20,000 × g. Typically, 30 ␮g of the total cell lysate were separated on SDS-PAGE. For the COS-7 cells transfected with human-LOX mature protein, the gel was analyzed by Western blot and probed with Omni-probe M21 antibody (Santa Cruz Biotechnology, USA), which targets the poly (His)6 epitope introduced at the 5 of the human recombinant LOX protein. In the case of the NRK-49F stably transfected with antisense LOX, the Western blot was probed with the antibody anti-LOX (V20) also purchased from Santa Cruz Biotechnology. The proteins recognized by the antibody were detected by using a secondary anti-rabbit antibody coupled to alkaline phosphatase and developing the blot with NBT/BCIP reagents (Roche Diagnostics GmbH, Mannheim, Germany). 2.3. DNA topochemistry The cells prepared for studies on chromatin supraorganization and DNA content analysis were subjected to the Feulgen reaction with hydrolysis performed in 4 M HCl for 60 min at 25 ◦ C (Mello, 1997). The material was then treated with Schiff reagent for 40 min in the dark, rinsed three times (5 min each) in sulfurous water and once in distilled water, air dried, and mounted in Cargille oil (nD = 1.54). All the staining and subsequent steps were performed in parallel for all of the cellular preparations to minimize variations in the experimental conditions and reduce the possibility of systematic errors. Frequencies of certain mitotic abnormalities, micronuclei and mitotic indices were established for fixed cell preparations stained with 0.025% toluidine blue (TB) (Merck, Darmstadt, Germany) solution in McIlvaine buffer at pH 4.0 for 15 min, rinsed rapidly in distilled water, air dried, and then mounted in nujol mineral oil (Lison, 1960; Vidal, 1972a,b). 2.4. Image analysis Two hundred Feulgen-stained interphase nuclei chosen at random from at least three slides under each experimental condition were measured using Carl Zeiss/Kontron equipment and Kontron KS400 software (Oberkochen/Munich, Germany). Microscopy images were obtained with a Zeiss Axiophot 2 microscope equipped with a 40/0.75 Neofluar objective, optovar 2, 0.90 condenser, and light of  = 546 nm. A 100-W halogen illuminator, a voltage regulator for high intensity maintained at point 4, filter wheels 1 and 2 rotated into position 100 (open position), and a luminous-filter diaphragm (transmitted light) at its maximal opening were used. These illumination conditions were maintained for all of the nuclei investigated. Images to be processed were fed from the microscope into a Pentium computer through a Sony CCD IRIS/RGB Hyper HAD color video camera. The threshold low (L) and high (H) levels were defined such that the nuclear images appeared false-colored green and were well separated from each other and from the background (in the present case L = 0 and H = 116–120 gray values). Under the optical conditions used, 1 ␮m corresponded to 7.23 pixels. The minimum area that was measured using this apparatus corresponded to 4 pixels. The software provided the following quantitative information: nuclear area (␮m2 ), nuclear perimeter (␮m), nuclear feret ratio (= minimum feret/maximal feret, a measure of the image ellipticity), mean gray value per nucleus, which was subsequently converted into optical density (OD = absorbance), standard deviation of the total densitometric values per nucleus or absorbance

10

M.L.S. Mello et al. / Micron 42 (2011) 8–16

variability per nucleus (SDtd) (which reflects the variability in the degree of chromatin packing state per nucleus), entropy (the number of bits necessary to store the densitometric values per nucleus image or amount of gray value variability in a nucleus) (Kontron KS400, 1995; Mello et al., 2009) and energy (calculated from the densitometric histogram, such that in contrast to entropy, large values are found in images with large regions of uniform intensity) (Kontron KS400, 1995; Doudkine et al., 1995). For conversion of the mean gray values into absorbances, gray values of 0 and 255 corresponded to light transmittances of 0 and 100%, respectively. The integrated OD (IOD; in this case, Feulgen-DNA content) was equal to the absorbances multiplied by the nuclear area. 2.5. Mitotic abnormalities, micronucleation and mitotic indices The nuclear/chromosome abnormalities investigated were as follows: tripolar/tetrapolar/multipolar metaphases, lagging chromosomes, chromosome bridging, and micronucleation. Mitotic indices were also estimated. Approximately 2000 cells per preparation were examined to determine micronuclei frequency and mitotic indices. 2.6. Cell death ratios Apoptotic cells and cells identified as those that will be suffering “cell death preceded by multinucleation” (denomination recommended in substitution to mitotic catastrophe) (Kroemer et al., 2009) were evaluated in preparations subjected to the Feulgen reaction and counterstained with acid fast green or stained with TB (Mello et al., 2004). Apoptotic cells were defined as cells presenting extremely condensed chromatin or apoptotic bodies. Approximately 2000 cells per preparation were examined to determine the cell death ratios.

Fig. 1. Western blots confirming the expression of the recombinant mature LOX with His6 epitope in COS-7 cells (A) and of antisense LOX vector in NRK-49F cells (B) (two independent transfections each).

3.2. Image analysis 2.7. Statistical analysis

The analysis of the Feulgen-DNA values (IOD) especially for the NRK-49F cells transfected with antisense lox revealed increased numbers of cells in the process of doubling these values (Fig. 2). A comparison of the parameters obtained by image analysis was thus performed only for the cells assumed to be in the lower G1 phase of their cell cycle, such that their IOD values after lox or antisense lox transfections (COS-7 and NRK-49F cells, respectively) did not differ from respective controls (Tables 1 and 2). Under these circumstances the nuclear areas were found to increase with lox transfection in COS-7 cells or with antisense lox transfection in NRK-49 cells, although the increase in nuclear areas was much higher in COS-7 cells (29.9%) than in NRK-49 F cells (12.5%) (Tables 1 and 2).

Calculations and statistical analyses were conducted using the Minitab 12TM software (State College, PA, USA). 3. Results 3.1. Western blot analysis Western blots confirmed the expression of the recombinant mature LOX with a His6 epitope in COS-7 cells (A) and LOX downregulation in the NRK-49F cells expressing the antisense LOX vector (B) (Fig. 1).

Table 1 Image analysis parameters for the Feulgen-stained lox-transfected COS-7 cells in the G1 phase based on Fig. 2 (IOD values < 70). Parameter type

Parameters

Geometric

Nuclear area (␮m2 ) Nuclear perimeter (␮m) Nuclear feret OD (absorbances) IOD (AU) SDtd Entropy Energy

Densitometric Textural

Cells transfected with pSG5 plasmid (control)a

Cells transfected with pSG5-LOX vectorb

X

S

Md

X

S

94.92A 50.54 0.76A 0.38A 34.92A 13.71A 5.65A 0.024A

26.26 16.83 0.10 0.04 8.25 3.73 0.41 0.008

92.53 45.96a

123.34B 78.73 0.81B 0.31B 36.94A 9.32B 5.08B 0.036B

26.96 31.53 0.07 0.06 7.53 2.82 0.44 0.012

Md 67.01b

AU, arbitrary units; IOD, integrated optical density or Feulgen-DNA amounts; Md, median; OD, optical density; S, standard deviation; SDtd, standard deviation of total densitometric values per nucleus; X, arithmetic means. Different letters in the same line (a, b; A, B) indicate differences that were significant at P0.05 ; minor letters (a, b), Mann–Whitney test; major letters (A, B), ANOVA. a n, number of nuclei = 127. b n, number of nuclei = 79.

M.L.S. Mello et al. / Micron 42 (2011) 8–16

11

Table 2 Image analysis parameters for the Feulgen-stained antisense lox-transfected NRK-49F cells in the G1 phase based on Fig. 2 (IOD values < 52). Parameter type

Parameters

Geometric

Nuclear area (␮m2 ) Nuclear perimeter (␮m) Nuclear feret OD (absorbances) IOD (AU) SDtd Entropy Energy

Densitometric Textural

Cells transfected with pCDNA3 plasmid (control)a

Cells transfected with antisense lox-containing pCL03 vectorb

X

S

Md

X

S

Md

90.06A 52.61 0.83A 0.40 35.33A 10.99 5.36A 0.028A

23.38 23.13 0.07 0.04 6.63 2.38 0.32 0.007

89.44 45.64a

101.39B 45.35 0.76B 0.39 35.69A 19.23 6.12B 0.016B

36.34 11.15 0.09 0.13 8.26 4.92 0.33 0.006

94.36 43.88b

0.39a 35.28 10.67a

0.43a 34.27 18.66b

AU, arbitrary units; IOD, integrated optical density or Feulgen-DNA amounts; Md, median; OD, optical density; S, standard deviation; SDtd, standard deviation of total densitometric values per nucleus; X, arithmetic means. Different letters in the same line (a, b; A, B) indicate differences that were significant at P0.05 ; minor letters (a, b), Mann–Whitney test; major letters (A, B), ANOVA. a n, number of nuclei = 181. b n, number of nuclei = 123.

As regards nuclear perimeters and feret ratios, they were found to increase with lox up-regulation in COS-7 cells (Table 1) but decrease with the antisense lox transfection in NRK-49F cells (Table 2). These results suggest that COS-7 cell nuclei became less elongated and acquired larger contours with increasing lox expression. On the other hand, NRK-49F cell nuclei acquired a more elongated shape but their contours became smaller after transfection with the antisense lox construction. Analysis of the densitometric parameters revealed that lox transfection in COS-7 cells induced a decrease in nuclear absorbances (OD values, converted from mean gray values (transmittances)). This decrease in nuclear absorbances and the increase in nuclear areas with lox transfection agree with the finding that in the selected cells, the Feulgen-DNA content after transfection with

lox or the control plasmid shows no difference (Table 1). In the NRK49F cells with no difference in DNA content regardless they were transfected with antisense lox or plasmid only, but with nuclear areas slightly increased after antisense lox transfection, OD values were not obviously affected, although their variability increased after antisense lox transfection (Table 2). Textural parameters such as SDtd and entropy decreased with lox transfection in the Feulgen-stained nuclei of COS-7 cells and increased with the antisense lox transfection in NRK-49F cells (Tables 1 and 2). These data suggest a more homogeneous transparency of the stained chromatin with lox up-regulation and coarser stained chromatin with lox down-regulation. The “energy” parameter values increased with high lox expression in COS-7 cells and decreased after antisense lox transfection in NRK-49F cells (Tables 1 and 2). This finding is in agreement with the nuclear entropy data and the definition of these parameters (Doudkine et al., 1995; Kontron KS400, 1995). The increase in energy values in the chromatin of COS-7 cells suggests that larger regions with uniform low absorbance intensity occurred with increased lox. Conversely, the decrease in energy values in the chromatin of NRK-49F cells suggests that larger areas of heterogeneous absorbance intensity occurred with lox down-regulation. For both cell types the textural changes are suggestive of chromatin remodeling (Oberholzer et al., 1996; Mello et al., 2009). We further treated nontransfected cells with BAPN. COS-7 cells treated with this LOX inhibitor showed opposite results in image analysis OD values and textural characteristics as compared with the changes observed with lox transfection (Tables 1 and 3). Mean nuclear areas and perimeters, and IOD values increased after BAPN treatment (Table 3). The results for the BAPN-treated NRK-49 F cells in the G1 phase showed the same changes in image analysis parameters as those obtained with antisense-lox transfection when compared to nontransfected cells, except for nuclear areas and feret (Tables 2 and 4). 3.3. Mitotic abnormalities, micronucleation and mitotic indices

Fig. 2. Feulgen-DNA content (IOD) distributions for COS-7 and NRK-49F cells with lox up- and down-expressions, respectively (red lines), in comparison with respective controls (black lines). n, 200.

Abnormal mitoses were observed under all experimental conditions analyzed. Examples of mitotic abnormalities such as tri-, tetra- and multi-polar spindles, chromosome bridges and lagging chromosomes, and of a micronucleate cell are shown in toluidine blue-stained COS-7 cells transfected with the pSG5 plasmid (Fig. 3a–g). Abnormal mitosis, abnormal metaphases and chromosome bridging frequencies in COS-7 cells decreased with increased lox expression in comparison to control transfection, but they did not differ from values for nontransfected controls (Table 5). No differences were found between the frequencies for the lox- and plasmid-transfected cells in lagging chromosomes (Table 5), but

12

M.L.S. Mello et al. / Micron 42 (2011) 8–16

Table 3 Image analysis parameters for Feulgen-stained COS-7 cells subjected to BAPN treatment and evaluated in association with IOD values < 53. Nontransfected cellsa

Parameter type

Parameters

Geometric

Nuclear area (␮m2 ) Nuclear perimeter (␮m) Nuclear feret OD (absorbances) IOD (AU) SDtd Entropy Energy

Densitometric Textural

BAPN-treated nontransfected cellsb

X

S

X

S

103.69A 48.94A 0.79A 0.39A 39.72A 10.36A 5.27A 0.031A

21.83 8.74 0.09 0.04 6.85 2.88 0.38 0.009

109.74B 50.32B 0.78A 0.40B 43.91B 9.44B 5.14B 0.033B

15.20 7.13 0.08 0.03 5.52 2.44 0.33 0.008

AU, arbitrary units; BAPN, beta-aminopropionitrile; IOD, integrated optical density or Feulgen-DNA amounts; OD, optical density; S, standard deviation; SDtd, standard deviation of total densitometric values per nucleus; X, arithmetic means. Different letters in the same line (A, B) indicate differences that were significant at P0.05 (ANOVA). a n, number of nuclei = 103. b n, number of nuclei = 142. Table 4 Image analysis parameters for Feulgen-stained NRK-49F cells subjected to BAPN treatment and assumed to be in a G1 phase (IOD values < 55). Parameter type

Parameters

Nontransfected cellsa X

S

X

S

Geometric

Nuclear area (␮m2 ) Nuclear perimeter (␮m) Nuclear feret OD (absorbances) IOD (AU) SDtd Entropy Energy

117.71A 51.87A 0.83A 0.36A 42.59A 7.27A 4.81A 0.042A

21.00 9.36 0.07 0.03 6.31 1.80 0.34 0.010

103.70B 46.58B 0.85B 0.41B 43.93A 7.81B 4.89B 0.039B

20.36 7.58 0.06 0.04 7.58 2.32 0.38 0.011

Densitometric Textural

BAPN-treated nontransfected cellsb

AU, arbitrary units; BAPN, beta-aminopropionitrile; IOD, integrated optical density or Feulgen-DNA amounts; Md, median; OD, optical density; S, standard deviation; SDtd, standard deviation of total densitometric values per nucleus; X, arithmetic means. Different letters in the same line (A, B) indicate differences that were significant at P0.05 (ANOVA). a n, number of nuclei = 164. b n, number of nuclei = 138.

these values were higher than those found in the nontransfected cells. In summary, these results indicate that lox up-regulation was not enough to reduce chromosome abnormalities in COS-7 cells. No differences were found in micronucleus frequency and mitotic indices, when comparing lox-transfected and plasmidtransfected COS-7 cells (Table 6). Micronucleus frequencies in transfected COS-7 cells were even higher than in nontransfected control cells whereas the opposite was found for mitotic indices. The latter indicates a decrease in cell proliferation with lox upregulation in COS-7 cells (Table 6). In NRK-49F cells transfected with antisense lox, abnormal mitoses were decreased in comparison to cells transfected with the plasmid, although no difference was observed in different types of mitotic abnormalities isolated, including abnormal metaphases (Table 5), or micronucleus frequency (Table 6). Abnormal mitoses were more representative in nontransfected NRK-49F cells subjected to BAPN treatment than in antisense lox-transfected cells

but they did not differ in comparison to nontransfected control (Table 5). The frequencies of the different types of chromosomal abnormalities under BAPN treatment did practically not differ from respective controls or from values obtained in antisense-lox transfected cells (Table 5). Although micronuclei were induced by BAPN treatment, their frequency was much lower than that found in transfected cells (Table 6). Cell proliferation increased after antisense-lox transfection in comparison to respective control or after nontransfected cells were treated with BAPN (Table 6). 3.4. Cell death frequencies Cell death morphologies associated with apoptosis and cell death preceded by multinucleation (Fig. 3h and i) were observed in all cell preparations examined. The apoptotic ratios were not affected by lox or plasmid transfections (COS-7 cells). Transfection

Table 5 Mitotic abnormalities in lox-transfected COS-7 cells as well as in antisense lox-transfected and BAPN-treated nontransfected NRK-49F cells. Cells

COS-7

NRK-49F

Transfection

pSG5 plasmid (control) pSG5-LOX vector Nontransfected control pCDNA3 plasmid (control) Antisense lox-containing pCL03 vector Nontransfected control Nontransfected cells + BAPN

Mitosis total no.

800 800 800 168 800 800 800

Abnormal mitosis (%)

Abnormal metaphases (%)

Chromosome bridges (%)

Lagging chromosomes (%)

X

S

X

S

Md

X

S

Md

X

S

Md

24.63A 14.63B 11.88B 16.86A 6.75B 11.25A 16.00A

2.75 4.27 2.69 5.94 2.90 4.21 2.80

14.38 9.00 7.00 8.97 4.88 7.00 11.38

0.85 2.27 2.04 1.47 2.93 3.16 1.75

14.25a 8.75b 7.00b 9.20a 4.25a,b 7.50a 11.25a

10.25A 5.63B 2.63 7.89 1.88 1.75 2.13

3.01 2.21 0.25 4.51 0.85 0.96 1.11

10.25 5.50a 2.50b 8.03a 1.75a 2.00a 2.50a

4.13 2.38 1.38 2.33 2.63 3.50 4.88

2.78 0.48 0.63 2.69 0.85 0.58 0.75

3.50a 2.25a 1.50b 2.33a 2.75a 3.50a 5.00a,b

a, b; A, B, different letters in the same column for the same cell type indicate differences that were significant at P0.05 ; minor letters (a, b), Mann–Whitney test; major letters (A, B), Student’s t-test; BAPN, beta-aminopropionitrile; Md, median; n, total number of cell phases counted; S, standard deviation; X, arithmetic means. Number of preparations analyzed per experimental condition, 8.

M.L.S. Mello et al. / Micron 42 (2011) 8–16

13

Fig. 3. Mitotic abnormalities, micronucleation and cell death aspects as seen in toluidine blue-stained COS-7 cells transfected with the pSG5 plasmid. Tripolar (a—arrow), tetrapolar (b—arrow) and multipolar (c) spindles, chromosome bridges (d, e—arrows), a lagging chromosome (f—arrow), a micronucleus-bearing cell (g), early apoptosis (h) and cell death preceded by multinucleation (i) are shown. Bars, 10 ␮m.

with antisense lox did not affect the apoptotic ratios in comparison to respective plasmid transfection (NRK-49F cells), although lower apoptotic ratios were found in nontransfected controls (Table 7). The apoptotic ratios for BAPN-treated and antisense lox-transfected NRK-49F cells did not differ from each other but they were higher than the apoptotic ratio for nontransfected cells (Table 7). In COS-7 cells cell death preceded by multinucleation was found to increase under lox up-regulation in comparison to nontransfected control.

This cell death form was also found to increase under antisense lox down-regulation but not after BAPN treatment in NRK-49 F cells (Table 7). 4. Discussion The present image analysis data indicate that chromatin decondensation is elicited with increased lox expression in COS-7 cells

14

M.L.S. Mello et al. / Micron 42 (2011) 8–16

Table 6 Micronucleus frequency and mitotic index in lox-transfected COS-7 cells as well as in lox antisense-transfected and BAPN-treated nontransfected NRK-49F cells. Cells

Transfection

COS-7

pSG5 plasmid (control) pSG5-LOX vector Nontransfected control pCDNA3 plasmid (control) Antisense lox-containing pCL03 vector Nontransfected control Nontransfected cells + BAPN

NRK-49F

MN (%)a

MI (%)b

X

S

Md

X

S

Md

0.54 0.26 0.06 0.28 0.36 0.06 0.11

0.41 0.27 0.02 0.18 0.29 0.02 0.02

0.51a 0.27a 0.05b 0.29a 0.29a 0.05b 0.10c

2.66A 2.36A 3.46A,B 0.33A 1.49B 2.17Cc 2.57Dc

0.93 0.81 0.73 0.09 0.75 0.36 0.26

2.32 2.16 3.26 0.34 1.19 2.08 2.47

a, b; A, B, different letters in the same column for the same cell type indicate differences that were significant at P0.05 ; minor letters (a, b), Mann–Whitney test; major letters (A, B), Student’s t-test; BAPN, beta-aminopropionitrile; Md, median; MN, micronucleus; MI, mitotic index; S, standard deviation; X, arithmetic means. a Number of preparations analyzed per experimental condition: 4. b Number of preparations analyzed per experimental condition: 8. c Number of preparations analyzed per experimental condition: 7.

and that chromatin condensation is induced with lox downregulation in NRK-49F cells. The chromatin texture characteristics in interphase COS-7 and NRK-49F cells after BAPN treatment support the results obtained with lox and antisense lox transfection assays. We propose that the less condensed chromatin state accompanying the high expression of lox in COS-7 cells is promoted by LOX action on histone H1, which causes the DNA–H1 complex to become loosened (Giampuzzi et al., 2003b). Indeed, histone H1 acts on 3-D folding of chromatin fibers, and the mode of the chromatin fiber compaction changes depending on linker histone H1 (Hizume et al., 2005). Based on the hypothesis that LOX binds to histone H1 (Giampuzzi et al., 2003b), the chromatin thicker fiber would acquire a more unfolded state, and chromatin higher-order architecture would become less condensed. The results for NRK-49F cells with lox down-regulation agree with previous reports for ras-transformed NIH 3T3 cells with low lox expression (Mello et al., 1995). Indeed, the down-regulation of lox in NRK-49F cells has been reported to induce a constitutive activation of the ras oncogene and an up-regulation of cyclin D1, which may contribute to the appearance of an oncogenic phenotype, primarily due to lox down-regulation, highlighting its role as a tumor suppressor (Giampuzzi et al., 2001, 2003a, 2005). This process may contribute to the nuclear phenotype variability, which is defined in terms of heterogeneity in chromatin packing states (larger absorbance variability, increased entropy and SDtd values and decreased energy) that were observed in NRK-49F cells. In favor of the hypothesis of LOX interacting with (and probably modifying) histone H1, experiments in progress (“Lysyl oxidase regulates MMTV promoter: enhancement of the promoter basal activity and of its response to glucocorticoids by histone H1 modification”, Di Donato) suggest that LOX exogenous overexpression regulates the activity of the Mouse Mammary Tumor Virus (MMTV) promoter. This observation is very important to our hypothesis because the MMTV promoter is critically regulated by histone H1

(Archer et al., 1991; Banks et al., 2001; Bhattacharjee et al., 2001; Belikov et al., 2007). In fact, the glucocorticoid response of the promoter requires the presence of histone H1 in its phased nucleosomal arrangement to open the promoter structure and allow access of the transcription factor NF-1, the bona fide promoter activator. Interestingly, overexpression of LOX was able to activate the promoter even in the absence of the hormonal stimulation, and, in the presence of the hormone significantly increased its response. These results are compatible with the removal of histone H1 from the MMTV active promoter region via LOX–histone H1 interaction. The increase in the nuclear area in lox-transfected COS-7 cells assumed to be in G1, favors the hypothesis that a chromatin decondensation process is involved. Regarding NRK-49F cells transfected with antisense lox, the increase in nuclear areas may be due to an increase in nuclear proteins, including nuclear matrix proteins, as demonstrated for the rat adrenal medulla pheochromocytoma PC12 cells treated with BAPN by Saad et al. (2010). The nuclear feret ratio data demonstrating that with increased lox expression COS-7 cell nuclei changed their shape from ellipsoidal to a rather spheroidal shape, and the inverse with decreased lox expression in NRK-49F cells may be related to changes in the nuclear microenvironment, possibly in association with the nuclear matrix (He et al., 2008). The effect of the nature, quantity and/or physical properties of extracellular matrix proteins indirectly on the nuclear size, shape and architecture is disregarded because in the present case the expressed recombinant LOX is the mature protein deprived of the pre-peptide with the hydrophobic signal sequence. Therefore it cannot be exported outside the cell. All the effects are thus due to intracellular interactions (enzymatic and/or chaperon-like activities) (Giampuzzi et al., 2001, 2003a,b). The different changes in nuclear shapes depending on which method to affect lox up-regulation was used (antisense lox transfection or BAPN) may indicate that different chemical pathways were involved, thus leading to different microenvironmental responses.

Table 7 Cell death ratios in lox-transfected COS-7 cells as well as in antisense lox-transfected and BAPN-treated nontransfected NRK-49F cells. Cells

Transfection

COS-7

pSG5 plasmid (control) pSG5-LOX vector Nontransfected control pCDNA3 plasmid (control) Antisense lox-containing pCL03 vector Nontransfected control Nontransfected cells + BAPN

NRK-49F

Apoptotic ratio (%)

Ratio of cell death preceded by multinucleation (%)

X

S

Md

X

S

Md

0.15 0.07 0.11 0.26 0.26 0.09 0.16

0.16 0.04 0.05 0.15 0.24 0.03 0.07

0.12a 0.07a 0.10a 0.20a 0.17a 0.10b 0.17a

0.21A 0.14B 0.07 0.04 0.08 0.03 0.04

0.05 0.06 0.03 0.02 0.04 0.04 0.03

0.20a 0.12b 0.07c 0.05a 0.07b 0.02a 0.05a

a, b; A, B, different letters in the same column for the same cell type indicate differences that were significant at P0.05 ; minor letters (a, b), Mann–Whitney test; major letters (A, B), Student’s t-test; BAPN, beta-aminopropionitrile; Md, median; S, standard deviation; X, arithmetic means. Number of preparations analyzed per experimental condition, 8.

M.L.S. Mello et al. / Micron 42 (2011) 8–16

In fact, it is not known whether LOX might also have a chaperon-like activity separated from its enzymatic role (Kagan and Li, 2003). Regarding mitotic abnormalities, although chromatin organization was affected presumably by LOX–H1 interactions in lox-transfected COS-7 cells, it was not effective enough to reduce the abnormal signals responsible for original chromosome-mitotic spindle aberrant relationships and micronucleus generation that occurred in nontransfected controls. In these cells, which are SV40 virus-transformed fibroblasts, an extra folding of chromatin promoted by elevated linking of histone H1 may occur as a consequence of low lox expression and of a fewer H1–LOX interactions. Irregular metaphase plates are generated because chromosome recruitment of factors that are essential for productive interactions with spindle microtubules would be disturbed (Maresca et al., 2005). When lox expression is experimentally increased in lox-transfected COS-7 cells, an increase in H1–LOX interactions, less H1 linking to chromatin and diminishment in frequency of irregular chromosome plates would be expected. However, lox upregulation in COS-7 cells was not effective to promote decrease in chromosome abnormalities. As expected, when lox expression increased, the proliferative status of these cells, as assessed by the mitotic index, decreased, and cell death preceded by multinucleation increased. Cell death preceded by multinucleation is the currently recommended denomination for mitotic catastrophe (Kroemer et al., 2009). The increase in mitotic index found in antisense lox-transfected NRK-49F cells compared to the plasmid-transfected cells and confirmed in nontransfected cells treated with BAPN is in agreement with LOX well-described tumor suppressor role, favoring the differentiation processes over the proliferative ones. In fact, LOX down-regulation is associated with D1 cyclin up-regulation with consequent increase of proliferation rate (Giampuzzi et al., 2003a, 2005). The accelerated proliferation induced by transfection with the antisense lox and reported activation of a ras oncogene in these cells (Giampuzzi et al., 2001) might be expected to be associated to increase in abnormal mitoses, if considering reports for NIH-3T3 mouse cells transfected with human mutated K-ras (Nigro et al., 1996). Micronuclei, which were more evident in BAPN-treated NRK-49F cells, are usually derived from incorrectly aligned chromosomes during metaphase as well as from lagging chromosomes and broken chromosome bridges during later mitotic stages. However, it has recently been demonstrated that not all mitoses with chromosome and spindle abnormalities give rise to micronuclei, which can originate from other sources such as the mother cells (Rao et al., 2008). The fact that results for transfectants obtained only with the plasmids differ in comparison with nontransfected controls have also been observed in other cell systems (Mello et al., 1994). These findings have been assumed to result from complex DNA rearrangements attained after transfection and a response of the host genome to disrupt the integrity of the foreign DNA, when these vectors are incorporated into mammalian cell DNA (Calos et al., 1983). In conclusion, chromatin remodeling as found in COS-7 and NRK-49F cells under different lox expression levels is suggested to support the hypothesis that LOX binds to histone H1 in vivo. Cell proliferation in COS-7 and NRK-49F cells and cell death at least in COS-7 cells agree with an expected interference of LOX in these processes.

Acknowledgments This investigation was supported by the São Paulo State Research Foundation (FAPESP, grant no. 06/66-8) and the Brazilian National Council for Research and Development (CNPq, grant no.

15

306611/2006-6). MLSM received a research fellowship from CNPq (PQ-1A). References Archer, T.K., Cordingley, M.G., Wolford, R.G., Hager, G.L., 1991. Transcription factor access is mediated by accurately positioned nucleosomes on the mouse mammary tumor virus promoter. Mol. Cell. Biol. 11, 688–698. Banks, G.C., Deterding, L.J., Tomer, K.B., Archer, T.K., 2001. Hormone-mediated dephosphorylation of specific histone H1 isoforms. J. Biol. Chem. 276, 36467–36473. Belikov, S., Astrand, C., Wrange, Ö., 2007. Mechanism of histone H1-stimulated glucocorticoid receptor DNA binding in vivo. Mol. Cell. Biol. 27, 2398–2410. Bhattacharjee, R.N., Banks, G.C., Trotter, K.W., Lee, H.-L., Archer, T.K., 2001. Histone H1 phosphorylation by Cdk2 selectively modulates Mouse Mammary Tumor Virus transcription through chromatin remodeling. Mol. Cell. Biol. 21, 5417–5425. Calos, M.P., Lebkowski, J.S., Botchan, M.R., 1983. High mutation frequency in DNA transfected into mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 80, 3015–3019. Contente, S., Kenyon, K., Rimoldi, D., Friedman, R.M., 1990. Expression of gene rrg is associated with reversion of NIH 3T3 transformed by LTR-c-H-ras. Science 249, 796–798. Contente, S., Csiszar, K., Kenyon, K., Friedman, R.M., 1993. Structure of the mouse lysyl oxidase gene. Genomics 16, 395–400. Contente, S., Yeb, T.J.A., Friedman, R.M., 2009. Tumor suppressive effect of lysyl oxidase proenzyme. Biochim. Biophys. Acta 1793, 1272–1278. Di Donato, A., Lacal, J., Di Duca, M., Giampuzzi, M., Ghiggeri, G., Gusmano, R., 1997. Microinjection of recombinant lysyl oxidase blocks oncogenic p21-Ha-ras and progesterone effects on Xenopus laevis oocyte maturation. FEBS Lett. 419, 63–68. Doudkine, A., MaCaulay, C., Poulin, N., Palcic, B., 1995. Nuclear texture measurements in image cytometry. Pathologica 87, 286–299. Fan, Y.H., Nikitina, T., Zhao, J., Fleury, T.J., Bhattacharyya, R., Bouhassira, E.E., Stein, A., Woodcock, C.L., Skoultchi, A.I., 2005. Histone H1 depletion in mammals alters global chromatin structure but causes specific changes in gene regulation. Cell 123, 1199–1212. Giampuzzi, M., Botti, G., Di Luca, M., Arata, L., Ghiggeri, G., Gusmano, R., Ravazzolo, R., Di Donato, A., 2000. Lysyl oxidase activates the transcription activity of human collagen III promoter. Possible involvement of Ku antigen. J. Biol. Chem. 275, 36341–36349. Giampuzzi, M., Botti, G., Cilli, M., Gusmano, R., Borel, A., Sommer, P., Di Donato, A., 2001. Down-regulation of lysyl oxidase-induced tumorigenic transformation in NRK-49F cells characterized by constitutive activation of ras proto-oncogene. J. Biol. Chem. 276, 29226–29232. Giampuzzi, M., Oleggini, R., Di Donato, A., 2003a. Altered adhesion features and signal transduction in NRK-49F cells transformed by down-regulation of lysyl oxidase. Biochim. Biophys. Acta: Proteins Proteomics 1647, 239–244. Giampuzzi, M., Oleggini, R., Di Donato, A., 2003b. Demonstration of in vitro interaction between tumor suppressor lysyl oxidase and histones H1 and H2: definition of the regions involved. Biochim. Biophys. Acta: Proteins Proteomics 1647, 245–251. Giampuzzi, M., Oleggini, R., Albanese, C., Pestell, R., Di Donato, A., 2005. Beta-catenin signaling and regulation of cyclin D1 promoter in NRK-49F cells transformed by down-regulation of the tumor suppressor lysyl oxidase. Biochim. Biophys. Acta: Mol. Cell. Res. 1745, 370–381. Guo, Y., Pischon, N., Palamakumbura, A.H., Trackman, P.C., 2007. Intracellular distribution of the lysyl oxidase polypeptide in osteoblastic cells. Am. J. Physiol.: Cell Physiol. 292, C2095–C2102. Happel, N., Doenecke, D., 2009. Histone H1 and its isoforms: contribution to chromatin structure and function. Gene 431, 1–12. He, S.H., Dunn, K.L., Espino, P.S., Drobic, B., Li, L., Yu, J., Sun, J.M., Chen, H.Y., Pritchard, S., Davie, J.R., 2008. Chromatin organization and nuclear microenvironments in cancer cells. J. Cell. Biochem. 104, 2004–2015. Hizume, K., Yoshimura, S.H., Takeyasu, K., 2005. Linker histone H1 per se can induce three-dimensional folding of chromatin fiber. Biochemistry 44, 12978–12989. Kagan, H.M., Li, W., 2003. Lysyl oxidase: properties, specificity, and biological roles inside and outside of the cell. J. Cell. Biochem. 88, 660–672. Kagan, H.M., Trackman, P.C., 1991. Properties and function of lysyl oxidase. Am. J. Resp. Cell Mol. Biol. 5, 206–210. Kagan, H.M., Williams, M.A., Calaman, S.D., Berkowitz, E.M., 1983. Histone H1 is a substrate for lysyl oxidase and contains endogenous sodium borotritidereducible residues. Biochem. Biophys. Res. Commun. 115, 186–192. Kaneda, A., Wakazono, K., Tsukamoto, T., Watanabe, N., Yagi, Y., Tatematsu, M., Kaminishi, M., Sugimura, T., Ushijima, T., 2004. Lysyl oxidase is a tumor suppressor gene inactivated by methylation and loss of heterozygosity in human gastric cancers. Cancer Res. 64, 6410–6415. Karymov, M.A., Tomschik, M., Leuba, S.H., Caiafa, P., Zlatanova, J., 2001. DNA methylation-dependent chromatin fiber compaction in vivo and in vitro: requirement for linker histone. FASEB J. 15, 2631–2641. Kirschmann, D.A., Seftor, E.A., Fong, S.F.T., Nieva, D.R.C., Sullivan, C.M., Edwards, E.M., Sommer, P., Csiszar, K., Hendrix, M.J.C., 2002. A molecular role for lysyl oxidase in breast cancer invasion. Cancer Res. 62, 4478–4483. Kontron Elektronic Imaging System KS400, 1995. User’s Guide, vol. 1. Eching/Munich. Kroemer, G., Galluzzi, L., Vandenabeele, P., Abrams, J., Alnemri, E.S., Baehrecice, E.H., Blagosklonny, M.V., El-Deiry, W.S., Golstein, P., Green, D.R., Hengartner, M.,

16

M.L.S. Mello et al. / Micron 42 (2011) 8–16

˜ Knight, R.A., Kumar, S., Lipton, S.A., Maloni, W., Nunez, G., Peter, M.E., Tschopp, J., Yuan, J., Piacentini, M., Zhivotovsky, B., Melino, G., 2009. Classification of cell death: recommendations of the Nomenclature Committee on Cell Death 2009. Cell Death Differ. 16, 3–11. Li, W., Nellaiappan, K., Strassmaier, T., Graham, L., Thomas, K.M., Kagan, H.M., 1997. Localization and activity of lysyl oxidase within nuclei of fibrogenic cells. Proc. Natl. Acad. Sci. U.S.A. 94, 12817–12822. Li, W., Nugent, M.A., Zhao, Y., Chau, A.N., Li, S.J., Chou, L.N., Liu, G., Kagan, H.M., 2003. Lysyl oxidase oxidizes basic fibroblast growth factor and inactivates its mitogenic potential. J. Cell. Biochem. 88, 152–164. Lison, L., 1960. Histochimie et Cytochimie Animales. Gauthier-Villars, Paris. Lucero, H.A., Kagan, H.M., 2006. Lysyl oxidase: an oxidative enzyme and effector of cell function. Cell. Mol. Life Sci. 63, 2304–2316. Maresca, T.J., Freedman, B.S., Heald, R., 2005. Histone H1 is essential for mitotic chromosome architecture and segregation in Xenopus laevis egg extracts. J. Cell Biol. 169, 859–869. Mello, M.L.S., 1997. Cytochemistry of DNA, RNA, and nuclear proteins. Braz. J. Genet. 20, 257–264. Mello, M.L.S., Barbisan, L.F., Lareef, M.H., Russo, J., Vidal, B.C., 2004. Cell death evaluation in benzo[a]pyrene-transformed human breast epithelial cells after microcell-mediated transfer of chromosomes 11 and 17. Mutat. Res. 546, 39–43. Mello, M.L.S., Contente, S., Vidal, B.C., Planding, W., Schenck, U., 1995. Modulation of ras transformation affecting chromatin supraorganization as assessed by image analysis. Exp. Cell Res. 220, 374–382. Mello, M.L.S., Lin, T.Y., Russo, J., 1994. Scanning microphotometry image analysis of Ha-ras-transformed human breast epithelial cells. Anal. Cell. Pathol. 7, 301–309. Mello, M.L.S., Russo, P., Russo, J., Vidal, B.C., 2009. Entropy of Feulgen-stained 17-beta-estradiol-transformed human breast epithelial cells as assessed by restriction enzymes and image analysis. Oncol. Rep. 21, 1483–1487.

Nigro, S., Geido, E., Infusini, E., Orecchia, R., Giaretti, W., 1996. Transfection of human mutated K-ras in mouse NIH-3T3 cells is associated with increased cloning efficiency and DNA aneuploidization. Int. J. Cancer 67, 871–875. Oberholzer, M., Östreicher, M., Christen, H., Brühlmann, M., 1996. Methods in quantitative image analysis. Histochem. Cell Biol. 105, 333–355. Pinnel, S.R., Martin, G.R., 1968. The cross-linking of collagen and elastin: enzymatic conversion of lysine in peptide linkage to alpha-aminoadipic-deltasemialdehyde (allysine) by an extract from bone. Proc. Natl. Acad. Sci. USA 61, 708–716. Rao, X.T., Zhang, Y.Y., Yi, Q.Y., Hou, H.L., Xu, B., Chu, L., Huang, Y., Zhang, W.R., Fenech, M., Shi, Q.H., 2008. Multiple origins of spontaneously arising micronuclei in HeLa cells: direct evidence from long-term live cell imaging. Mut. Res.: Fund. Mol. Mech. Mutagen. 646, 41–49. Ren, C., Yang, G., Timme, T.L., Wheeler, T.M., Thompson, T.C., 1998. Reduced lysyl oxidase messenger RNA levels in experimental and human prostate cancer. Cancer Res. 58, 1285–1290. Saad, F.A., Torres, M., Wang, H., Graham, L., 2010. Intracellular lysyl oxidase: effect of a specific inhibitor on nuclear mass in proliferating cells. Biochem. Biophys. Res. Commun. 396, 944–949. Tang, S.S., Chichester, C.O., Kagan, H.M., 1989. Comparative sensitivities of purified preparations of lysyl oxidase and other amine oxidases to active site-directed enzyme inhibitors. Connect. Tissue Res. 19, 93–103. Vidal, B.C., 1972a. Birefringence measurements on toluidine blue-stained chromosomes during mitosis. Ann. Histochim. 17, 145–150. Vidal, B.C., 1972b. Doppelbrechungsdispersion und Lineardichroismus von Euund Heterochromatin nach Färbung mit Toluidinblau. Nachweis eines CottonEffektes. Beitr. Pathol. 145, 269–285. Wu, G.J., Guo, Z.M., Chang, X.F., Kim, M.S., Nagpal, J.K., Liu, J.W., Maki, J.M., Kivirikko, K.I., Ethier, S.P., Trink, B., Sidransky, D., 2007. LOXL1 and LOXL4 are epigenetically silenced and can inhibit Ras/extracellular signal-regulated kinase signaling pathway in human bladder cancer. Cancer Res. 67, 4123–4129.