0023-6837/02/8205-667$03.00/0 LABORATORY INVESTIGATION Vol. 82, No. 5, p. 667, 2002 Copyright © 2002 by The United States and Canadian Academy of Pathology, Inc. Printed in U.S.A.

BRIEF METHOD A Sample-Saving Preparation to Extract DNA-Binding Proteins from Cardiac and Vascular Tissues Anna Cargnoni, Laura Tiberio, Patrizia Martina, Roberta Ardesi, Luisa Schiaffonati, and Roberto Ferrari Cardiovascular Research Center (AC, PM, RA), Salvatore Maugeri Foundation for Care and Research, IRCCS, Gussago, , Department of Biomedical Sciences & Biotechnology (LT, LS), University of Brescia, Brescia, and Department of Clinical & Experimental Medicine (RF), University of , Ferrara,

he analysis of the changes in the DNA-binding a lipopolysaccharide (LPS)-treated group (n ϭ 4) T activity of specific transcription factors is one of (Shames et al, 1998). the approaches to study the ability of a tissue to Hearts and aortas were immediately washed with develop and/or to adapt to external stimuli. Therefore, cold saline plus a mixture of protease inhibitors (1 mM the availability of pure and functional nuclei, together phenylmethylsulfonyl fluoride, 25 ␮g/ml leupeptin, 10 with nuclear subfractions, is of crucial importance. A ␮g/ml aprotinin, and 1 ␮g/ml pepstatin A), injected large number of methods have already been devel- into the aortic root. A portion (30 to 50 mg) of cardiac oped; however, they are commonly related to cultured or aortic tissue was finely minced, transferred into an cells or soft mammalian tissues. When applied to all-glass tissue grinder with a clearance of 0.09 to 0.16 tough tissues, such as heart and vessels, they result in mm and volume of 2 ml (Wheaton), and homogenized lower yield, so that large amounts of sample are ina6ϫ volume of homogenization medium containing always needed (Allfrey et al, 1964; Dignam et al, 1983; 0.25 M sucrose, 10 mM Tris (pH 7.5), and protease Fleischer and Kervina, 1974). The aims of our work inhibitors. In the case of cardiac tissue, 0.1% of were: (a) to prepare good quality nuclei and nuclear Nonidet P-40 was added to the medium. Homogeni- proteins from very small samples (30 to 50 mg) of zation was performed with two cycles of 10 strokes muscle tissues (heart and vessels) by optimizing exis- with 10 minutes “rest” in between. tent methods, in terms of homogenization and purifi- The homogenate was centrifuged at 1000 ϫg for 10 cation conditions; and (b) to verify the suitability of our minutes at 4° C. The supernatant was centrifuged at preparation to study transcription factor activities, in 100,000 ϫg for 60 minutes at 4° C to obtain the particular nuclear factor-␬B (NF-␬B). We have specif- cytosolic fraction. The pellet was gently resuspended ically analyzed NF-␬B activation because it has re- ␮ cently gained interest in the cardiovascular field in 700 to 800 l of 0.25 M sucrose, 10 mM Tris, and 1 (Bourcier et al, 1997; Purcell et al, 2001) and because mM MgCl2 (pH 7.5) and purified on a sucrose cushion ϫ it was crucial to have a pure nuclear fraction avoiding (2.0 M, 4 ml) by centrifugation at 70,000 g for 70 any cytoplasmic contamination. In fact, NF-␬Bis minutes at 4° C. present in the cytoplasm, anchored to an inhibitory Nuclear pellets were lysed on ice, in a buffer con- protein: I␬B␣. After activation, I␬B␣ is degraded and taining 10 mM Tris (pH 7.5), 1 mM MgCl2, 0.3 M NaCl, NF-␬B gains access to nuclei for transcriptional regu- 25% glycerol, and freshly added 50 mM dithiothreitol, lation (Henkel et al, 1993; Lenardo and Baltimore, by vortexing (6 cycles/hour). The nuclear protein frac- 1989). tions were collected after centrifugation at 40,000 ϫg To evaluate NF-␬B under quiescent (negative con- for 30 minutes at 4° C; they were then aliquoted, trol) or activated (positive control) conditions, we have snap-frozen in liquid nitrogen, and stored at Ϫ80° C. studied two groups of rats: a control group (n ϭ 3) and The integrity and purity of the nuclei were confirmed by fluorescence microscopic analysis, after staining with 4',6-diamidino-2-phenylindole and sulforhodam- ine 101, which stain DNA and proteic matrix, respec- Received January 14, 2002. Address reprint requests to: Dr. Anna Cargnoni, Cardiovascular Research tively. The nuclei appeared unbroken and almost com- Center, Salvatore Maugeri Foundation, Via Pinidolo, 23, 25064 Gussago, pletely free from cytoplasmic contaminants (data not Brescia, Italy. E-mail: [email protected] shown).

Laboratory Investigation • May 2002 • Volume 82 • Number 5 667 Cargnoni et al To detect NF-␬B activation, we have studied the teins for cardiac and 2 ␮g for aortic extracts) were cellular fractions by the immunoblotting technique. In incubated in a binding mixture [10 mM Tris, pH 7.8, 5% detail, we have analyzed the presence of p65 (one of glycerol, 1 mM EDTA, 0.5 mM dithiothreitol, 0.5 ␮gof the two proteins that commonly constitutes NF-␬B) in poly (dI-dC)] and 0.1 to 0.5 ng of labeled probe. The the nucleus and the presence of I␬B␣ in the cytosol. samples were loaded onto a 5% polyacrylamide gel in To this end, proteins of both nuclear (30 ␮g for heart or 1ϫ Tris-borate-EDTA. Gels were run at 4° C for 2 8 ␮g for aorta) and cytosolic (30 ␮g for heart or 20 ␮g hours at 20 mA, dried, and autoradiographed at for aorta) extracts were fractionated onto 10% (p65) or Ϫ70° C with Kodak MR film. For competition experi- 12% (I␬B␣) SDS-PAGE electrophoresis and trans- ments, a 100-fold excess of specific unlabeled probe ferred on polyvinylidene difluoride membranes. After was added. Supershift assays were performed by blocking (5% nonfat dry milk), the membranes were incubating the nuclear extracts with 10 ␮gofan immunoblotted with anti-p65 and anti-I␬B␣ antibodies anti-p65 antibody (Santa Cruz Biotechnology). To nor- (Santa Cruz Biotechnology, Santa Cruz, California). malize the loading, we have analyzed DNA-binding Immune complexes were detected by anti-IgG conju- activity versus octamer-1, an ubiquitous transcription gated to horseradish peroxidase (Santa Cruz Biotech- factor whose site is present in many housekeeping nology) and analyzed by the enhanced chemilumines- genes, by using the following probe: 5'-GATCGAA- cence assay (ECLϩplus; Amersham International, TGCAAATCACTAGCT-3'. Arlington Heights, ). The membranes were then The results obtained with ElectroMobility Gel Shift rehybridated for octamer-1 and glyceraldehyde-3- Assay support those obtained by Western blotting phospho-dehydrogenase expressions to perform analysis. No DNA-binding activity was detected in loading controls of nuclear and cytosolic extracts, control hearts and aortas, whereas specific signals respectively. were observed in both tissues from LPS-treated ani- The results obtained by immunoblotting detection mals. Signal specificity was validated by supershift show that, under quiescent conditions (control group), and competition assays (Fig. 2). As expected, the p65 was almost absent in the nuclear fractions and DNA-binding activities of octamer-1 were constant I␬B␣ was present in the cytosolic compartment from both in control and LPS-treated tissues (Fig. 2). both cardiac and aortic tissues (Fig. 1). As expected, Our results demonstrate that our sample-saving when NF-␬B was activated (LPS-treated group), p65 preparation does not produce any artifacts because it markedly accumulated in the nuclear fractions and does not activate NF-␬B per se in samples from the I␬B␣ levels were reduced in the cytosolic compart- control group (Figs. 1 and 2). It allows the identification ment from both tissues (Fig. 1). of NF-␬B activation in samples from the LPS-treated Finally, we have analyzed NF-␬B DNA-binding ac- group as demonstrated by: (a) accumulation of p65 in tivity in nuclear extracts by ElectroMobility Gel Shift the nucleus and concomitant reduction of I␬B␣ levels Assay, using a 32P-radiolabeled double-stranded oli- gonucleotide probe containing NF-␬B consensus binding site 5'-GGATCCTCAACAGAGGGGACTT- TCCGAGGCCA-3'. Samples (15 ␮g of nuclear pro-

Figure 1. Figure 2. Nuclear factor-␬B (NF-␬B) activation was evaluated as nuclear p65 and NF-␬B binding activity was evaluated by ElectroMobility Gel Shift Assay in cytosolic I␬B␣ myocardial and aortic contents in cellular extracts from control myocardial and aortic nuclear extracts from control and LPS-treated groups. A and LPS-treated groups. A representative radiogram is shown. Octamer-1 and representative autoradiogram is shown. Competition: a 100-fold excess of glyceraldehyde-3-phospho-dehydrogenase (GAPDH) expressions were ana- specific unlabeled double-stranded probe was added. Supershift assays: 10 ␮g lyzed as loading controls for the nuclear and the cytosolic extracts, of an anti-p65 antibody was added to the nuclear extracts. Octamer-1 binding respectively. activity was analyzed as loading control.

668 Laboratory Investigation • May 2002 • Volume 82 • Number 5 Brief Method in the cytosol (Fig. 1); and (b) induction of NF-␬B Bourcier T, Sukhova G, and Libby P (1997). The nuclear DNA-binding activity in the nucleus (Fig. 2). Moreover, factor-␬B signaling pathway participates in dysregulation of our preparation was useful to detect the expression vascular smooth muscle cells in vitro and in human athero- and the activity of DNA-binding proteins other than sclerosis. J Biol Chem 272:15817–15824. NF-␬B, as shown by our findings on octamer-1 (Fig. 2). Dignam GD, Lebovitz RM, and Roeder RG (1983). Accurate Our preparation offers important advantages: (a) It is transcription initiation by RNA polymerase II in a soluble easy to perform and not time consuming. (b) It avoids extract from isolated mammalian nuclei. Nucleic Acids Res pooling samples. (c) It offers the possibility of studying 11:1475–1489. a regional physiologic/pathologic process that in- Fleischer S and Kervina M (1974). Subcellular fractionation of volves only a portion of the organ of interest. (d) It can rat liver. In: Fleischer S and Packer L, editors. Methods in be used on human biopsy specimens, thus overcom- enzymology. : Academic Press, 6–41. ing the limitations of a morphologic study. Henkel T, Machleidt T, Alkalay I, Kronke M, Ben-Neriah Y, and Baeuerle PA (1993). Rapid proteolysis of I␬B␣ is neces- sary for activation of transcription factor NF-␬B. Nature Acknowledgements 365:182–185. We thank Dr. Alessandro Bettini for editorial help. Lenardo MJ and Baltimore D (1989). NF-␬B: A pleiotropic We are also indebted to Prof. Piergiovanni Grigolato mediator of inducible and tissue-specific gene control. Cell and Dr. Moris Cadei for microscopy analysis. This 58:227–229. study was supported by Cofinanziamento MURST Purcell NH, Tang G, Yu C, Mercurio F, DiDonato J, and Lin A 2000. (2001). Activation of NF-␬B is required for hypertrophic growth of primary rat neonatal ventricular cardiomyocytes. Proc Natl Acad Sci USA 98:6668–6673. References Shames BD, Meldrum DR, Selzman CH, Pulido EJ, Cain BS, Allfrey VG, Littau VC, and Mirsky AE (1964). Methods for the Banerjee A, Harken AH, and Meng X (1998). Increased levels ␬ ␣ purification of thymus nuclei and their application to studies of myocardial I B- protein promote tolerance to endotoxin. of nuclear protein synthesis. J Cell Biol 21:213–231. Am J Physiol 275:H1084–H1091.

Laboratory Investigation • May 2002 • Volume 82 • Number 5 669