:9
`
`* F’“‘f3*i':‘_° M
`"
`1
`.
`'~‘:.’.’7’ARY
`
`2'.
`a I:
`j;_
`
`av
`
`ISSN 0014-4827
`
`
`
`F3 ‘K1’330I
`
`
`EXECUTIVE EDITORS
`
`T. CASPERSSON ‘ T. GUSTAFSON - D. MAZIA - NILS R. RINGERTFZ’
`
`ASSISTANT EDITOR
`
`BARBARA CANNON
`
`ADMINISTRATIVE EDITOR
`
`VERA RUNNSTROM-REIO
`
`E D I T O R S
`
` — {.
`
`-BASERGA ' D. BOOTSMA - M, BOUTEILLE ‘ O. I. EPIFANOVA ~ W. W. FRANKE ' P. R. HARRISON
`
`" HELLSTROM - S. A. LATT - R. LEVLMONTALCINI - G. L. NICOLSON -
`
`I. PASTAN ' P. A. PETERSON
`
`I
`
`D. M. PRESCOTT - F. H. RUDDLE - P. O. SEGLEN - K. WEBER
`
`VOLUME 141 - OCTOBER 1982 - NUMBER 2
`
`ECREAL 141 (2) 231-521 (1982)
`
`ACADEMIC PRESS
`
`A Subsidiary of Harcourt Brace Jovanoviclr, Publislwrs
`
`New York London
`
`Paris San Diego San Francisco S210 Paulo Sydney Tokyo Toronto
`BD Exhibit 1008
`Page 1 of 13
`Page 1 of 13
`BD Exhibit 1008
`
`

`
`EXPERIMENTAL CELL RESEARCH
`
`EXECUTIVE EDITORS
`
`T. CASPERSSON, Karolinska Institutet, Department of Tumor Pathology, Karolinska Sjukhuset,
`S-I040! Stockholm 60, Sweden
`T. GUSTAFSON, Wenner-Gren Institute for Experimental Biology, University of Stockholm, S-I I345
`Stockholm, Sweden
`D. MAZIA, Hopkins Marine Station, Stanford University, Department of Biological Sciences, Pacific
`Grove, CA 93950, USA
`NILS R. RINGERTZ, Depanment of Medical Cell Genetics, Medical Nobel Institute, Karolinska
`Institutet, Box 60400, S-l040l Stockholm, Sweden
`
`ASSISTANT EDITOR
`
`BARBARA CANNON, Wenner-Grcn Institute for Experimental Biology, University of Stockholm.
`S-l 1345 Stockholm, Sweden
`
`EDITORS
`
`R. BASERGA, Department of Pathology, Temple University, School of Medicine, 3400 North Broad
`Street, Philadelphia, PA 19140, USA
`D. BOOTSMA. Department of Cell Biology and Genetics, Erasmus University, Postbus I738, Rotter-
`dam, The Netherlands
`M. BOUTEILLE, Laboratoire de Pathologie Cellulaire, Institut National de la Sante et de la Recherche
`Médicale (INSERM) (U I83), 15. rue de I‘Ecole de Médecine, 75270 Paris, Cedex 06. France
`0. 1. EPIFANOVA, Academy of Sciences of USSR. Institute of Molecular Biology, Vavilov ul. 32,
`Moscow B-312, USSR
`W. W. FRANKE, Abteilung fur Membranbiologie und Biochemie, Instilut fur ZelI— und Tumorbiologie.
`Deutsches Krebsforschungszentrum, D—6900 Heidelberg 1, Germany
`I’. R. HARRISON. Wolfson Laboratory for Molecular Pathology, The Beatson Institute for Cancer
`Research, Garscube Estate, Switchback Road, Barsden. Glasgow G61 IBD. UK
`KARL Erux HELLSTROM, Division of Tumor Immunology, Fred Hutchinson Cancer Institute, H24,
`Columbia Street, Seattle, WA 98104, USA
`S. A. LATT, Harvard Medical School, The Children's Hospital Medical Center. 300 Longwood
`Avenue, Boston, MA 02115. USA
`R. LEVI-MONTALCINI, CNRS. Laboratory of Cell Biology, Via G. Romagnosi I8/A. I—00196 Rome,
`Italy
`G. L. NICOLSON, Department of Tumor Biology, University of Texas System Cancer Center, M.D.
`Anderson Hospital and Tumor Institute, Houston, TX 77030, USA
`I. PASTAN, National Institute of Health, Laboratory of Molecular Biology, National Cancer Institute,
`Bldg 37, Room 4B27, Bethesda, MD 20205, USA
`P. A. PETERSON, Wallenberg Laboratory, University of Uppsala, Box 562, S-751 22 Uppsala, Sweden
`D. M. PRESCOTT, University of Colorado, Department of Molecular, Cellular and Developmental
`Biology, Boulder, CO 80302, USA
`F. H. RUDDLE, Department of Biology, Yale University, New Haven, CT 06520. USA
`P. O. SEGLEN, Department of Tissue Culture, Norsk Hydro’s Institute for Cancer Research, The
`Norwegian Radium Hospital, Montebello, Oslo 3. Norway
`K. WEBER, Abt. Biochemie I, Max-Planck-lnstitut fur biophysikalische Chemie, Karl-Friedrich-
`Bonhoeffer-Institut, Postfach 968, D-3400 Gottingen, Germany
`Administrative Editor: VERA RUNNSTROM-REIO, Karolinska Institutet, Box 60400, S-l040l Stock-
`holm, Sweden
`
`Editorial Office: Karolinska Institutet, Box 60400. S-10401 Stockholm, Sweden
`Tel. Nat. 08-33 93 80; Int + 468-33 93-80
`
`Published monthly at S-751 8| Uppsala. Sweden, by
`Academic Press. lnc., Ill Fifth Avenue, New York. NY I000}, USA
`1982: Volumes l37—l42. Price: $528.00 USA; $604.50 outside USA
`All prices include postage and handling.
`All subscription orders should be addressed to the office of the Publishers at
`Ill Fifth Avenue, New York, NY l0003, USA.
`Send notices of change of address to the office of the Publishers at least 6-8 weeks in advance.
`Please include both old and new addresses.
`
`Postmaster: Send changes of address to Experimental Cell Research.
`Ill Fifth Avenue, New York. NY 10003. USA.
`Second class postage paid at Jamaica, NY ll43l, USA. Air freight and mailing in the USA by
`Publications Expediting, lnc., 200 Meacham Avenue, Elmont, NY H003, USA.
`© 1982 by Academic Press, Inc.
`Printed in Sweden by Almqvist & Wiksell. Uppsala 1982
`Page 2 of 13
`Page 2 oi13
`
`

`
`Copyright © I982 by Academic Press, Inc.
`All rights of reproduction in any form reserved
`OOI4-4827/82/100397-1 l$02.00/0
`
`Experimental Cell Research 141 (1982) 397—407
`
`hybridization and the immunoc
`
`IN SITU HYBRIDIZATION OF DNA SEQUENCES IN HUMAN
`
`METAPHASE CHROMOSOMES VISUALIZED BY AN INDIRECT
`
`FLUORESCENT IMMUNOCYTOCHEMICAL PROCEDURE
`
`A. C. Van PROOIJEN-KNEGT," * J. F. M. Van HOEK,‘ J. G. J. BAUMAN,*'*
`P. Van DUIJN,‘ I. G. WOOL’ and M. Van der PLOEG'
`
`‘Department of Histochemistry and Cytochemistry. Leiden University, 2333 AL Leiden,
`The Netherlands, and 2Department ofBiochemistry. University of Chicago.
`Chicago, IL 60637, USA
`
`SUMMARY
`
`In situ hybridization and immunocytochemical procedures are described which allow identification
`and localization of specific DNA sequences in human chromosomes by fluorescence microscopy.
`With this method the genes coding for I85 and 28S ribosomal RNA (IRNA) were localized on
`human metaphase chromosomes by in situ hybridization of 18S or 28S rRNA followed by an
`immunocytochemical incubation with specific anti-RNA—DNA hybrid antiserum. Visualization of
`the immunocytochemically localized RNA-DNA hybrids was achieved by indirect
`immuno-
`fluorescence. The antiserum against RNA—DNA hybrid molecules was raised in a rabbit injected
`with poly(rA)—poly(dT). The specificity of the sera was determined using a model system of
`Sephadex beads to which various nucleic acids had been coupled. To obtain optimal specific
`fluorescence and very low aspecific background staining, several modifications of the in situ
`ytochemical procedures were investigated. The use of aminoalkyl-
`si1ane—treated glass slides, removal of unbound fluorochrome molecules from the fluorochrome-
`labelled antibody solutions and application of a proteinase K treatment during the hybridization
`procedure and the immunocytochemical procedure proved to be essential for optimal results.
`
`f..chromosomes and cell nuclei DNA can
`
`phila polytene chromosomes. Visualization
`
`[ detected cytochemically by staining
`ods, which are based on distinctive
`mica] properties of this nucleic acid.
`lse staining procedures are, however,
`to identify specific base sequences.
`it calization of such specific base se-
`, nces can be achieved by in situ hybridi-
`‘- with complementary nucleotide se-
`_' ces. Until recently, the localization of
`
`molecular hybrids in microscopic prepa-
`
`ons could only be detected by auto-
`'ography [I-4].
`
`.-
`
`first approach to the fluorescent de-
`
`'on of nucleic acid hybrids was de-
`bed [5] for SS rRNA genes in Dr0s0-
`
`was achieved by an immunocytochemical
`
`procedure using an antiserum specific for
`
`RNA—DNA hybrids. Subsequently the hy-
`bridization method was improved [6] by
`
`combining the denaturation and reannealing
`
`processes.
`
`Recently, another approach was advo-
`
`cated [7] in which a fluorescent marker is
`
`attached directly to the complementary
`RNA. This method was used to localize the
`
`kinetoplast DNA in Crithidia luciliae cells.
`
`* To whom offprint requests should be sent.
`** Present
`address:
`Radiobiological
`REPGO—TNO, Rijswijk. The Netherlands.
`
`Institute ,
`
`Exp Cell Res I4] (I982)
`
`Page 3 of 13
`Page 3 of 13
`
`
`
`

`
`II.
`
`water-in-oil emulsion twice intramuscularly over ,
`day period. followed by an intravenous injection 5‘.
`vol of the poIynucleotide—mBSA solution diluted
`an equal volume of PBS, one week later. Each
`munizing dose contained 30-50 pg of poly( '.
`poly(dT). The rabbits were bled one week after that
`travenous injection. The immunoglobulins were
`cipitated with ammonium sulphate [12].
`
`Chromosome preparations
`Metaphase chromosome preparations were -~.:..
`from human blood cell cultures [13]. using I‘.-
`alkylsilane-treated glass slides [14].
`Aminoalkylsilane-treated glass slides were -»:~_{
`following a procedure described [15] as a part '
`coupling procedure of enzymes to porous glass - .»._‘
`Microscope glass slides were cleaned by incub
`overnight in a 10% solution of Extran MAO] (
`lisch. E. Merck. Darmstadt) in deionized water.
`_
`were then rinsed with hot (60°C) tap water and
`deionized water and dried at 80°C. These slides ~.‘
`incubated for 16 h in a 2% (v/v) solution of 3- uf_
`propyltriethoxysilane (Aldrich Europe, Beerse,
`gium)
`in dry acetone. Afterwards the slides
`rinsed in acetone and two changes of deionized ,
`and stored in 0.02% NaN3 in deionized water.
`"J
`to use they were again rinsed in deionized water
`air-dried. Slides could be stored for up to 6
`without losing their properties.
`
`.
`
`
`
`Coupling ofnucleic acids to
`
`Sephadex G10
`
`Sephadex G 10 (Pharmacia Fine Chemicals AB.
`sala, Sweden) was activated with CNBr, nucleic :1-jg
`were coupled to the activated agarose beads and
`remaining active groups were blocked with 2,;
`amine [16]. The following nucleic acids were
`po|y(rA) and yeast RNA (Boehringer Man J
`GmbH), double-stranded and denatured DNA r_:
`calfthymus (BDH Chemical Ltd, Poole).
`-
`Hybridization of poly(dT) and of poly(rU) to L
`dex—poly(rA) was performed by the following p 2/
`dures. A settled volume of 120 ,u.l Sephadex—pol
`was diluted with an equal volume of PBS conta‘
`1 M NaCl. To this solution 60 ,ug of poly(dT)(
`"
`or poly(rU) (Boehringer) in 700 pl PBS containing 2
`NaCl were added. Incubation was performed by -:—<_,
`ing the mixture at room temperature for 2 h on a 1.},
`fied hematocryt microshaker (frequency 50 Hz, . ,
`tude less than 0.5 mm). The beads were then
`fuged and the supernatant was removed and _
`for UV measurements. The beads were washed
`times in PBS. The amounts of poly(dT) or poly
`hybridized to the beads were determined by me
`ing the UV absorbance at 260 nm for the poly
`or poly(rU) solutions before and after hybridizati '‘
`
`Procedure for the immunocytochemical
`
`'
`staining ofSepl1adex beads
`Swollen Sephadex G10 beads with coupled nu I
`acids were stored diluted 1:10 in PBS con
`
`398
`
`Van Prooijen-Knegt er al.
`
`the DNA of adenovirus type 5 in virus-in-
`fected KB cells [8] and a number of genes in
`
`giant chromosomes from salivary glands of
`
`Drosophila hydei and Drosophila melano-
`gaster [9, 10].
`
`The present paper describes modifica-
`
`tions of the in situ hybridization and im-
`
`munocytochemical procedures, permitting
`
`identification of specific DNA sequences in
`
`human chromosomes by fluorescence mi-
`
`croscopy. Antisera specific for RNA—DNA
`
`hybrids were raised in rabbits by injecting
`
`complexes of poly(rA)—poly(dT) and meth-
`
`ylated bovine serum albumin. The specific-
`
`ity of the obtained sera was determined
`
`using a model system of Sephadex beads
`to which various nucleic acids were cou-
`
`pled. The effects of modifications of several
`
`steps in the complete procedure on the final
`
`results were investigated and modifications
`
`are presented which increase the specific
`
`hybridization and immunocytochemical lo-
`
`calization potentials and which results in a
`
`lower non-specific background fluores-
`
`cence. This combination of improved meth-
`ods allows a clear visualization of the 18S
`
`and 28S ribosomal cistrons in human meta-
`
`phase chromosomes.
`
`MATERIALS AND METHODS
`
`Phosphate-buffered saline (PBS) contained 0.137 M
`NaCl, 2.68 mM KCI, 7.98 mM Na2HP0,, and 1.47
`mM KI-LZPO. (pH 7.2); 1><SSC (sodium saline citrate,
`pH 7.0) contained 0.15 M NaCl and 0.015 M sodium
`citrate. All chemicals used were of analytical grade.
`
`Immunization procedure
`
`Antibodies against poly(rA)—poly(dT) were elicited in
`rabbits [11]. A 300 pl solution containing poly(rA)-
`poly(dT) (Miles Laboratories Inc. Research Products.
`Stoke Poges, Slough, Bucks., UK;
`1 mg/ml in PBS)
`was mixed with 30 pl of a solution of methylated
`bovine serum albumin (mBSA, Miles; 10 mg/ml in dis-
`tilled water). PBS was added to the resultant turbid
`suspension for a final volume of 3 ml. This mixture
`was stored at —30°C. Immunization was carried out by
`mixing equal volumes of the poly(rA)—poly(dT)—mBSA
`solution and Freund adjuvant (Bacto: Difco Labora-
`tories. Detroit. Mich.) and injecting the resultant
`
`Exp Cell Res 14] (I982)
`
`Page 4 of 13
`Page 4 of 13
`
`

`
`Fluorescent hybridocyrochemistry of human ribosomal genes
`
`399
`
`ethanol (twice) and 100% ethanol, for 5 min each, and
`air-dried. 18S and 28S rRNA prepared from rat liver
`[18] using freshly distilled phenol instead of m-cresol
`were used for hybridization.
`The hybridization reaction was carried out with 15
`,ul of the RNA solution of 20 p.g/ml in 70% formamide/
`3><SSC (which is a mixture of 7 vol parts formamide
`plus 3 vol parts 10><SSC) placed between the slide and
`a 24x50 mm coverslip. Incubation was for 40 h at 34°C
`in a culture-dish containing paper tissue saturated with
`70% formamide/3><SSC. Coverslips were then re-
`moved and the slides were washed for 60 min in three
`changes of 70% formamide/3><SSC and for 30 min
`in two changes of 3><SSC to remove non-hybridized
`RNA.
`
`Immunocytochemistry
`
`Immunocytochemical visualization of the RNA—DNA
`hybrids was performed by the following procedure,
`each step being carried out at room temperature unless
`otherwise stated: (1) two rinses in 2xSSC for 5 min
`each; (2) incubation in a solution of 15 ,u.g RNase A
`plus 0.15 pg RNase T,/ml
`in ZXSSC for 120 min;
`(3) two rinses in 20 mM Tris-HCI pH 7.4 containing
`2 mM CaCl._, for 5 min each; (4) incubation in a pro-
`teinase K solution of 1 p.g/ml
`in 20 mM Tris-HCI,
`pH 7.4 containing 2 mM CaCl2 for 15 min; (5) three
`rinses in PBS for 15 min each; (6) incubation in non-
`immune goat serum 2% in PBS containing 0.02%
`NaN3 for 30 min; (7) incubation in Ig fraction of anti-
`hybrid serum diluted 1 : 20 in PBS containing 2% non-
`immune goat serum plus 0.02% NaN3 for 60 min (lg
`concentration 300 pig/ml based on absorbance meas-
`urements at 280 nm); (8) three rinses in PBS at 37°C
`for 2 min each;
`(9) one rinse in PBS for 10 min at
`37°C; (10) incubation for 60 min in rhodamine—labe1led
`goat-anti-rabbit IgG (GAR—TRITC, United States Bio-
`chemical Corp., Cleveland, Ohio) diluted 1 : 80 in PBS
`containing 2% non-immune goat serum plus 0.02%
`NaN3. Free rhodamine molecules were removed from
`the GAR—TRITC preparations by incubation overnight
`at 4°C with Affi—Gel
`102 (arnino—agarose; BioRacl
`Laboratories, Richmond, Calif; 1 pl Affi—Gel 102 sus-
`pension/96 ptl GAR—TRITC solution) under continu-
`ous agitation; (11) three rinses in PBS at 37°C for 2 min
`each; (12) one rinse in PBS for 10 min at 37°C; (13)
`dehydration in 70% ethanol
`(twice), 90% ethanol
`(twice), both containing 300 mM ammonium acetate,
`followed by 100% ethanol, for 5 min each, and air-
`drying.
`Incubation in the RNase solution, 2% non-immune
`goat serum, specific anti-RNA—DNA IgG solution and
`GAR—TRITC solution were all performed with 35 pl
`of the solutions layered on the preparation, that was
`then covered with a coverslip and kept
`in a moist
`chamber.
`In some experiments the preparations were hybrid-
`ized and treated immunocytochemically according to
`the procedures of Rudkin & Stollar [5] or of Stuart &
`Porter [6]. These experiments are indicated as such
`in the text. After the in situ hybridization and immuno-
`cytochemical procedures the human metaphase chro-
`mosomes were identified by counterstaining with 4’ ,6-
`diamidino-2-phenyl-indole (DAPI) [19].
`
`Exp CeHRes14I(I982)
`
`
`
`ii % NaN3. A 20-fold dilution of this suspension was
`" e in PBS containing 1% Triton X-100 (BDH Chem-
`~_-j-
`). Twenty-five microlitre samples were trans-
`" ed to 1.5 ml plastic reaction tubes (Eppendorf
`ratebau Netheler+Hinz GmbH, Hamburg) and 25
`of a 5-fold serial dilution of the Ig solution in PBS
`the anti-hybrid or the normal rabbit serum (initial
`3. concentration 10 mg/ml) were added. Incubation
`f, for 90 min at room temperature, during which the
`-1.’ were constantly agitated, as described before.
`_. er incubation the beads were washed twice in 1 ml
`"PBS containing 0.65 M NaCl followed by subse-
`3-’ sedimentation at 10000 g and once with 1 ml
`ii containing 0.5 % Triton X-100, for at least 10 min
`ch. The final wash buffer was removed to a volume
`. 25 pl and 25 pl from a 1:15 dilution of SwAR—
`C (Swine anti-rabbit serum IgG, heavy and light
`‘n; DAKO lmmunoglobulins, Copenhagen; lg con-
`,.'- tration 0.48 mg/ml; fluorochrome/protein ratio 3.1)
`'_ PBS containing 0.5% Triton X-100, were added.
`,,_‘’u' incubation and two more washings with 1 ml
`;; containing 0.65 M NaCl as in the first step. the
`ads were finally washed in 1 ml PBS containing
`02% NaN3 and fluorescence of the individual beads
`g measured microfluorometrically.
`
`V icrofluorometry
`.11" orescence measurement of the individual Sephadex
`it s embedded in PBS containing 0.02% NaN3 was
`" ormed as described by Bauman et al. [16].
`
`‘
`
`ybridization 0fRNA to human
`taphase chromosomes
`"order to remove possibly present endogenous RNA,
`slides were first treated with a solution of 100 ug
`5 ase A (from bovine pancreas, Boehringer) plus 1 tag
`_'i ase T, (from Aspergillus oryzae, Boehringer) per ml
`‘ ZXSSC for 2 h at room temperature. A few drops
`the RNase solution were layered over the prepara-
`ff: which was then covered with a cover glass. This
`5, bation was performed in a moist chamber. After
`, bation the cover glasses were removed and the
`‘to s were washed three times in 2><SSC for 30 min
`I then dehydrated in 70% ethanol (twice), 90%
`~: ol (twice) and 100% ethanol. Each alcohol de-
`tion was for 5 min after which the slides were
`-dried. The DNA in the preparation was then de-
`,‘=
`red with freshly prepared 0.07 N NaOH for 3 min
`[i wed by rinses in 70% ethanol (twice), 90% etha-
`"is (twice) and I00% ethanol, for 1 min each, and air-
`1 mg.
`this pretreatment, hybridization was per-
`ter
`". ed [8] with the following modifications.
`j
`e preparations were treated with a proteinase K
`Pilzen, chromatographisch gereinigt. Merck)
`tion of 1 ,u.g/ml in 20 mM Tris-HCI pH 7.4 con-
`ng 2 mM CaCl2 for 15 min at 37°C [17] (the pro-
`’ se K solution being preincubated for 4 h at 37°C
`liminate possible traces of RNase or DNase ac-
`, ) and after this treatment the preparations were
`ydrated by incubation in 70% ethanol (twice), 90%
`
`Page 5 of 13
`Page 5 of 13
`
`

`
`
`
`ecules from the fluorochrome-labelled .
`
`«E
`
`’
`rabbit antibody solutions prior to use.
`was accomplished by incubating the e‘
`tions with Affi-Gel 102 overnight at 4°C, 5,
`cording to F. T. Bosman & A. F. P i.
`
`(personal communication).
`de Goeij
`moval of unbound TRITC molecules
`
`_
`
`the conjugated antibody solutions was .
`firmed by measuring the absorbance sf"
`tra (200—700 nm) of the fluorochrome
`belled antibody solutions before and -Ti
`incubation with Affi-Gel 102. This pr,
`
`dure reduced the non—specific backgro
`staining of Sephadex G 10 beads. For Fl 0
`conjugated antibody solutions at our nj
`posal, incubation with Affi—Gel 102 prov"
`
`unnecessary.
`However, even after removal of unbo 7
`
`TRITC molecules with Affi-Gel 102,
`batches of FlTC—labe1led Swine-anti-Ra..."
`
`(SWAR) antibodies proved more suitabl
`
`the experiments with the model system;
`Sephadex
`beads,
`than TRITC-labeu
`Goat-anti-Rabbit
`(GAR) and TRITC:
`
`belled Sheep-anti-Rabbit (ShAR) anti
`
`ies. The latter two gave rise to conside .
`
`r
`
`higher non—specific background stainin
`
`the beads. SwAR—FITC was therefore u
`in all experiments with the Sephadex m L
`
`,
`system.
`When determining the specific fluo
`cence values, a ‘correction was applied
`eliminate the fluorescence value caused
`
`non-specific binding of the immunogl
`lins to the beads, using the fluoresce
`values of the beads treated with normal
`'
`
`bit serum and SwAR—FITC. The results"
`
`400
`
`Van Prooi}'en—Knegt er al.
`
`Fluorescence microscopy and
`
`labelled anti-rabbit antibodies and
`
`ing the fluorescence intensity of the n.—
`vidual beads.
`
`When using TRITC-labelled antibod"
`
`it proved necessary to reduce non-spec
`
`background staining of Sephadex beads"
`the removal of unbound fluorochrome us-
`
`microphotography
`After the hybridization and the immunocytochernical
`and DAPI staining procedures, the preparations were
`inspected using a Dialux microscope (Leitz GmbH,
`Wetzlar, FRG) with epi—illumination from an HBO 200
`W mercury arc (Osram Gmbl-l, Berlin). The DAPI
`fluorescence emission was visualized using a combina-
`tion of an UG l excitation filter, an LP 460 barrier filter
`(both Schott and Gen, Mainz) and a dichroic mirror
`AH 400 (Leitz). The rhodamine fluorescence was ob-
`served with LP 530 (Schott) and SP 560 (Balzers AG,
`Lichtenstein) filters in the excitation beam. a dichroic
`mirror AH 580 (Leitz) and a LP 590 (Schott) as a barrier
`filter.
`Fluorescence photography of the metaphases was
`performed with the same Dialux microscope, using the
`oil-immersion objective (Fluoresz, 63X/1.30, Leitz)
`and high-speed Kodak Tri-X-Pan film (Eastman Ko-
`dak Comp., Rochester, N.Y.). The recording capabili-
`ties of the photographic emulsion were improved by
`pre-exposing the film to homogeneous white light [20].
`Exposure times were 1 sec or less for DAPI fluores-
`cence and up to 3 min for the rhodamine fluorescence.
`
`RESULTS
`
`Specificity of anti-hybrid serum
`
`The specificity of the immunoglobulins for
`
`their respective antigens was tested in a
`
`model system consisting of Sephadex beads
`
`[21].
`The lg fractions, obtained by ammonium-
`
`sulphate precipitation of the sera from rab-
`
`bits after
`
`immunization with poly(rA)—
`
`poly(dT) complexed with mBSA, were in-
`cubated in increasing dilutions with Sepha-
`dex G10 beads to which various nucleic
`
`acids had been coupled. Sephadex beads
`
`which were inactivated following CNBr ac-
`tivation served as controls. In other control
`
`experiments incubations of the different
`
`types of beads were performed with either
`
`normal
`
`rabbit serum or PBS containing
`
`0.5% Triton X-100 (conjugate control), in-
`stead of the serum tested. The immuno-
`
`globulins bound to the Sephadex beads
`
`after incubation and washing procedures
`
`are detected and quantitated by subsequent
`incubation of the beads with fluorochrome-
`
`Exp Cell Res 141 (I982)
`
`Page 6 of 13
`Page 6 of 13
`
`

`
`Fluorescent hybridocytochemistry of human ribosomal genes
`
`401
`
`
`
`bridization of the rodent rRNA to the hu-
`
`man ribosomal genes under the hybridiza-
`tion conditions used [22].
`When the metaphase preparations were
`
`hybridized with 18S or 28S rRNA from rat
`
`liver and subsequently treated immuno-
`
`cytochemically with the specific anti—hybrid
`
`antiserum according to the procedure of
`Rudkin & Stollar [5], no specific chromo-
`
`somal fluorescence was detectable visually
`
`or in microphotographs of individual meta-
`
`phases. In addition there was a relatively
`
`high non-specific background fluorescence.
`
`Application of the method of Stuart & Por-
`ter [6] resulted in a specific fluorescence on
`the short arms of at best one or two acro-
`
`centric chromosomes per metaphase just
`
`visible against a high aspecific background
`
`fluorescence, which was comparable to that
`
`seen after the procedure of Rudkin & Stol-
`
`lar. When hybridization was performed ac-
`cording to the method of Bauman et al. [8]
`we found, after the indirect immunofluo-
`
`rescence procedure, specific fluorescence
`
`on the distal parts of the D and G group
`chromosomes. The results were evaluated
`
`both visually and via microphotographs of
`
`individual metaphases. Although the results
`
`were unmistakably specific,
`
`the fluores-
`
`cence yield was relatively poor.
`Since this low specific fluorescence might
`
`be due to chromosomal proteins covering
`
`the DNA and thus hampering the access of
`
`the large RNA molecules to the denatured
`
`DNA or of the immunoglobulins to the
`
`formed RNA-DNA hybrids, the following
`
`modification was evaluated to improve the
`
`procedure.
`
`The procedures to hybridize 18S and 28S
`
`rRNA to the human metaphase chromo-
`
`some preparations were varied in several
`
`ways. Proteinase K treatment was either
`
`omitted or applied before or after the hy-
`
`bridization procedure or applied both be-
`
`Exp Cell Res I41 (I982)
`
`
`
`
`
`dilution of anti-serum! Ig
`
`. 1. Specificity test of rabbit serum against RNA-
`" hybrids. cc represents conjugate control, for fur-
`? details see text. 0-0, Seph-poly(rA}-poly(dT);
`O, Seph-inactivated; A---A.
`Seph-poly(rA);
`r-A, Seph-poly(rA)—poly(rU); V---V, Seph-ss-
`V---V, Seph—ds-DNA; El---El, Seph-yeast
`
`
`
`‘A.
`
`lg fraction from one serum are illus-
`
`ed in fig. 1. The presence of IgG in the
`
`, m reacting specifically with poly(rA)—
`: (dT) is evident. No significant cross-
`
`poly(rA),
`is observed with
`-: tivity
`" (rA)—poly(rU),
`single-stranded
`and
`ible-stranded DNA from calf thymus, or
`;' st RNA [11].
`finly one of the three injected rabbits pro-
`
`d antiserum of this specificity. This
`
`“4. m was used in further experiments.
`
`in: hybridization and immuno-
`‘chemical hybrid detection
`
`situ hybridization to human lymphocyte
`
`» phase chromosomes was performed
`ti 18S and 28S rRNA prepared from rat
`1‘. Rat 18S and 28S rRNAs have suf-
`
`nt sequence homology (98%) to human
`and 28S rRNAs to ensure efficient hy-
`
`Page 7 of 13
`Page 7 of 13
`
`

`
`Page 8 of 13
`
`

`
`Fluorescent hybridocytochemistry of human ribosomal genes
`
`403
`
`It was shown that preparations on amino-
`
`alkylsilane-treated glass slides exhibit con-
`
`siderably less background staining of the
`
`glass slides than did preparations on rou-
`
`tinely cleaned glass
`
`slides, when using
`
`GAR—TRITC (which produced the highest
`
`specific fluorescence).
`
`After using the optimal procedures (as
`
`described in Materials and Methods), red
`
`fluorescent spots derived from rhodamine
`could be detected on the short arms of the
`
`acrocentric
`
`chromosomes,’ which were
`
`identified with the aid of the DAPI—staining
`
`pattern of the same metaphase (see figs
`2, 3).
`
`Fluorescence specificity was confirmed
`
`in experiments in which E. coli tRNA was
`
`used, instead of the 18S or 28S rRNA, for
`
`hybridization, or in which the specific anti-
`
`serum was replaced by non-immune rabbit
`serum.
`
`In such experiments no specific fluores-
`
`cence was detectable. Chromosome prepa-
`rations obtained from different blood cell
`
`donors sometimes differed in the number of
`
`chromosomes
`
`showing specific
`
`fluores-
`
`cence and in the emission intensity of the
`
`fluoresceing spots. These results agree with
`autoradiographic grain counting data [24,
`
`25].
`
`DISCUSSION
`
`The autoradiographic procedure for the de-
`
`tection of hybrids of radioactively labelled
`
`CRNA and DNA in chromosome prepara-
`
`tions has several disadvantages. The proce-
`
`dure is time—consuming and the topological
`
`resolution is often insufficient
`
`to clearly
`
`locate hybrids within defined bands of the
`
`eukaryotic metaphase chromosomes. When
`
`flu0rochrome—labelled CRNA is used for hy-
`
`bridization, these disadvantages are over-
`
`come. Attaching a fluorochrome molecule
`
`Exp Cell Res 141 ([982)
`
`
`
`re and after the hybridization procedure.
`
`fecific fluorescence after the immuno-
`ochemical procedure was
`evaluated
`1"sually.
`
`‘ Compared with the results obtained after
`
`A procedure without proteinase K, treat-
`; ent either before or after the hybridization
`,action gave rise to an increase in specific
`
`. orescence and a reduction of non-specific
`
`ckground fluorescence, the overall result
`
`ing comparable for both procedures. The
`sults from the procedure with proteinase
`
`’
`
`treatment applied before the hybridiza-
`
`'on reaction were only slightly better.
`
`l
`
`Optimal results were obtained when pro-
`inase K was applied both before and after
`"a hybridization reaction. A considerable
`crease in specific fluorescence yield was
`
`served and the background fluorescence
`
`2: the glass slides was reduced to very low
`":2 1.165.
`
`The optimal temperature for and duration
`
`..‘ the hybridization reaction were deter-
`
`'ned by visual inspection of specific fluo-
`’ cence for several combinations of tem-
`
`ratures (20°C, 34°C (i.e. Tm —25°C) and
`C) and incubation times (18, 40 and 66 h)
`
`‘ hybridization with both 18S rRNA and
`rRNA.
`
`Hybridization for 40 h at 34°C gave opti-
`
`.: I results,
`
`i.e. highest specific fluores-
`
`gnce, low non-specific fluorescence and a
`
`.n d preservation of chromosome mor-
`ology.
`I" As it is known that during immunocyto-
`emical staining IgG molecules behave
`le cationic dyes [23], the influence of 3-
`
`’: inopropyltriethoxysilane
`treatment of
`glass slides on the non-specific back-
`und fluorescence of the glass slides after
`5-: unocytochemistry was studied. This
`
`atment of glass slides reduces the non-
`
`_cific binding of cationic dyes to the
`gative groups of untreated glass [14].
`
`Page 9 of 13
`Page 9 of 13
`
`

`
`404 Van Prootjen-Knegt et al.
`
`.“l
`
`
`
`complexed to mBSA, was tested usi
`model system of Sephadex G10 beads
`which various nucleic acids had been r‘
`
`pled. This proved to be a reliable o‘
`
`system. The specific and non-specific r
`tions of nucleic acids and proteins are 1'
`
`ited to the surface of the spheres beca i
`
`the exclusion limit of the Sephadex G
`
`is 700 D, permitting a direct comparison f
`
`these two types of reaction, provided 5'
`
`free fluorochrome molecules are pre -
`[16].
`
`g
`
`It
`
`is likely that the higher non-spec‘.
`
`background staining of the Sephadex
`in the model experiments with the TR] A
`labelled GAR and ShAR antibodies u
`resulted from interaction of protein-bo z"
`rhodamine with Sephadex beads and if
`
`from an inferior quality of the GAR
`ShAR antibodies. The fact that the G '.
`
`TRITC, when applied_to human chro wt
`some preparations which had been hybn',
`ized with 28S rRNA and treated with an
`
`RNA—DNA antibodies, gave rise to a co
`
`siderably higher specific fluorescence yie:
`
`than SwAR—FITC (despite the fact that
`
`'
`
`fluorochrome/antibody ratio was only 1 -:
`
`GAR—TRITC compared to 3.1 for Sw '
`
`FITC) is in accordance with this hypot -»:i
`sis. FITC-labelled SwAR antibodies we
`
`therefore used in all experiments with ’
`
`Sephadex model system, whereas GA
`
`TRITC was used for chromosome prep
`tions.
`
`Hybridization and immunocytochemical
`
`to RNA does not affect the hybridization
`
`properties of the RNA and the use of fluoro-
`
`chrome-labelled CRNA [10] renders the pro-
`cedure for detection of RNA—DNA hybrids
`
`the resolving
`very rapid and improves
`power which is now limited to that of the
`
`optical system. In addition the use of radio-
`active material is avoided.
`
`The sensitivity of any microscopical
`method for detecting fluorescence, depends
`on the number of fluorochrome molecules
`
`per unit of area in the object being detect-
`
`able. In our fluorescence microscope sys-
`tem the minimal amount of tetramethyl-
`
`rhodamine detectable is 250 molecules per
`0.25 ;;:.m2 against a black background [26].
`It was expected that the fluorescence de-
`
`tection of RNA—DNA hybrids using anti-
`bodies against these hybrids in an indirect
`
`immunocytochemical technique would give
`
`rise to a high sensitivity [26]. For the im-
`munofluorescence method it has been as-
`
`sumed that, when nearly all antibodies are
`
`assumed to be bound monovalently (which
`
`might be the case when high antibody con-
`
`centrations are used), one antibody per 10
`nucleotides can be accommodated [27], and
`
`that the indirect method, using conjugates
`with one fluorochrome per antibody, will
`result in about one fluorochrome molecule
`
`per four nucleotides.
`
`immunocyto-
`The use of the indirect
`chemical method requires the availability of
`
`specific anti-RNA—DNA antibodies. Our
`
`data show that not every immunization re-
`
`sults in the production of the desired spe-
`
`drawback may be solved by the production
`
`of monoclonal antibodies against poly(rA)—
`poly(dT) [28].
`
`Spec:fic‘iI_v testing of sera
`
`The specificity of the sera obtained from
`
`rabbits
`
`injected with poly(rA)—poly(dT)
`
`Exp Cell Res I4] (1982)
`
`Page 10 of 13
`Page 10 of 13
`
`ciflc antibodies but, on the other hand, this
`
`procedures
`
`_
`
`The procedure described by Rudkin & Sto
`
`lar [5] for Drosophila polytene chrome
`
`somes and the improved procedure d
`
`scribed by Stuart & Porter [6] for hybridi
`
`=_
`
`tion and immunocytochemical detection e’
`
`the formed hybrids, proved in our hands
`
`be insufficient for the unequivocal iden
`
`
`
`

`
`Fluorescent hybridocytochemistry of human ribosomal genes
`
`405
`
`
`
`' A—DNA hybrids for the antibodies, or
`th, but optimal results were observed
`I.'t