TEM O F DNA

MI C R O S C O P Y A N D AN A LY S I S • MA R C H 2002 23

A Mazza1, A Casale 1 and A Raudino 2 1International Institute of Genetics and Biophysics, Naples, and2Department of Chemistry, University of Catania, Italy

I N T R O D U C T I O NA one-step release-spreading method hasbeen developed with the aim of producingthe most conservative and fastest electronmicroscope screening of an organelle’sgenomic DNA [1]. The method is based onsimultaneous organelle lysis and DNA release-spreading by a detergent-high salt-ethidiumbromide-cytochrome C suspending solution(DSEC) capable of spreading, at its water-airinterface, the organelle’s disassembled mole-cules. The molecules so spread are prone toabsorb and counter-stain by acidified uranylacetate (UO2) after touching the film-coatedelectron microscope grid [1].

The method, assayed on a variety of isoton-ically and hypotonically treated algalorganelles, or isolated algal DNAs, has provedto be a very efficient size and heterogeneityspectral indicator especially suited for rescu-ing and screening DNA sub-heteromeres insubliminal amounts [1,2]. Its proposed use as afast universal genomic screener, particularlysuited for the diagnosis of human mt-DNApathogenic deletions [3-5], is supported herethrough the cumulative discussion of previousresults [1] and present findings obtained by arefined experimental procedure. The resultsreferring to isotonically or hypotonicallytreated Acetabularia (A) mitochondria will betreated in detail in that they document the fullpower of the method.

T H E R E F I N E D P R O C E D U R EThe improvement of a previous procedure [1]has been possible through the reformulationof the DSEC composition and use of an uncon-ventional absorption sequence. The reformu-lated DSEC (DRSS) is made up of the followingirreplaceable chemicals in the indicated rangeof ad hoc adjustable concentrations: 10-4-10-3 MTris-phosphate-EDTA buffer titrated to pH 8-9with potassium hydroxide (KOH); 0.04-0.1 Mpotassium and calcium chlorides (KCl, CaCl2); 3-4µg/ml ethidium bromide (EtBr); 0.05-0.1%(w/v) cytochrome C (CytC); 0.2-0.3 mg/mlof 4000-8000 polyethylene glycol (PEG); 0.05-0.1%(w/v) sodium deoxycholate (NaDOC); 0.1-0.2mg/ml Brij 58. On occasion, DRSS may beconveniently enriched by: 0.04-0.1 Mmonoacidic potassium phosphate (K2HPO4);0.5-1% (w/v) malic acid; 10-3-10-2 M ammoniumacetate; 10-3 % M Triton X-100; 10-5 -10-8 Moctyl-�-D-glycopyranoside (OGP); and 1-3%(w/v) symperonic acid, which together con-tribute in enhancing DNA’s contrast andincreasing stability of its supporting film.

DNA samples, either free molecules orgenomes encased within the matrix of isoton-

B I O G R A P H YA. Mazza is head ofresearch at the CNR’sInternational Instituteof Genetics and Bio-physics in Naples, Italy.He graduated as M.D.,qualified as a pneumol-ogist and anato-mopathologist, and specialized in the elec-tron microscopy of biomacromolecules. DrMazza has published in the fields of: lungand heart ultrastructure and physiology,experimental embryology, cellular physiol-ogy, ultrastructure and physiology of virusesand extranuclear genomes and physics ofhigh resolution electron microscopy of bio-molecules. His present activities relate thephysics of molecular absorption and the bio-physics/ therapy of the tumour cell.

A B S T R A C TThe one-step electron microscopy of DNA isa conservative procedure which allows holis-tic rescue of DNA genomic heteromeresfrom 0.1µm (330 base pairs) to 100-300µm,thus permitting a quantitative diagnosis ofthe quasi-totality of genomic DNA hetero-geneity. Its technical and semantic preroga-tives, support its widest application as a fastuniversal genome screener.

K E Y W O R D Sdeoxyribonucleic acid, DNA, molecularbio–logy, transmission electron microscopy,specimen preparation

A U T H O R D E TA I L SDr Antonio Mazza, Istituto Internazionale diGenetica e Biofisica, Via Marconi, 10, 80125Napoli, Italy. On leave at the Department ofChemistry, Università della Calabria, 87030Arcavacata di Rende (CS), Italy.Tel: +39 817257111Fax: +39 817257202

Microscopy and Analysis (UK), 88, 23-25, 2002.©2002 Rolston Gordon Communications.

ically or hypotonically treated cytoplasmicorganelles are diluted in 0.1-1.0ml of DNA-preserving solution at concentrations which aspreviously shown [1], may range from 107 -1013 Daltons/ml or 1 - 104 individuals/ml. Thediluted samples are then simply mixed inabout 80ml sterile DRSS in a 10cm Petri dishand left to stand at will under the dish cover.At this point DNA testing may be performedby addition of appropriate probes to the DRSSDNA-spreading suspension.

DNA harvesting by simply touching the 75cm2 DRSS sample surface with a coated EMgrid may start a few (5-10) minutes after thesample’s dilution or manipulation. Howeverbecause of the sterility provided by the dishcover, it can continue for days until the appar-ent exhaustion of the sample, given the time-insensitive efficiency of DRSS and the indefi-nite stability of its DNA monolayer [1].

The DRSS-spread DNA monolayers manifestdifferential absorption and staining behaviourwith relation to the ground substance of theabsorbing film. In fact, they stick promptly tothe touching carbon film and can then becounterstained by subsequent immersion for30-60 sec in an acetone solution containing5x10-4 M uranyl acetate, 2% methanol, 2%ethanol and 0.01-0.1% formaldehyde (acidi-fied UO2), followed by dehydration in 95%ethanol and hexane drying (C-DNA; Fig 1a). Onthe other hand, the DNA monolayers neitherabsorb nor contrast on identically processedFormvar-silicon oxide (SiO)/dioxide(SiO2)-coated grids (FSS) (Touzart and Matignon,Vitry sur Seine, Paris, France) unless the DNA-touched FSS are rubbed dry against the freshlycleaved surface of mica, before UO2 contrast-ing and post-stain drying (FSSM-DNA; Fig 1b).Neither improvements of C-DNA quality byagainst mica rubbing dry nor FSS absorptionby simple pressing of the DNA-touched FSSonto mica has ever been observed.

F E AT U R E S A N D E F F I C I E N C Y O F T H ER E F I N E D M E T H O DBoth the C- and the FSSM-DNAs from DRSS-lysed Acetabularia mitochondria are in theform of mini -(0.1 to1.0µm long) to maxi – (1.1to 92.3 µm long) circular, 10-12nm thick, nearlyround and territorially isolated molecules. Byboth ultrastructural criteria and control exper-iments [1,6,8], these were unequivocally iden-tified as fully relaxed covalently closed DNAcircles (Fig 1a, b).

Both the maxi- and mini- circles of DNAindifferently absorb on either carbon or FSSM,contrary to what Fig 1, limited for brevity toonly exemplary demonstrations, might sug-

One-Step Transmission ElectronMicroscopy of Deoxyribonucleic Acid

P23-24-25 Mazza 13/3/02 12:24 pm Page 1

MI C R O S C O P Y A N D AN A LY S I S • MA R C H 200224

gest. These, and analogous observations onlinear DNAs [1 and unpublished], excludeinvolvement of any form or size factors regu-lating DNA absorption under the conditionsof the one-step procedure.

The DNA circles on carbon were often intouch with background 0.2-3µm kinked RNAsand 1x3µm2 globular or fenestrated matrixand membrane remnants (Fig 1a); on FSSM thecircles stood out because they were much bet-ter contrasted against a much cleaner back-ground ( Fig 1b).

Under either random or exhaustive DRSS-DNA monolayer sampling on carbon or FSSMneither overlapping nor juxtaposing DNAswere ever observed.

Similar results were obtained for Acetabu-laria mt-DNAs at 108 to 1013 Daltons/ml (notshown) and organelles at 1 to 104/ml (Fig. 1a,b). These figures are converted into occupiedareas (occupancy) by taking into account a

conversion factor of 1.9x106 Daltons/µm DNA[1,6,8] plus the ~ 120µm2 of the mean A mt-DNA circle, some 40µm long (Fig 2), and the 3mm2 occupancy of the mean A mitochondrialgenome [1,6], the above observation concernssamples exceeding 10-5 to 3 times the 75cm2

spreading surface but in no case affectingterritorial isolation of spread DNA molecules.

All the evidence indicates that DRSS release-spread DNA molecules neither overcrowd noroverlap on the spreading surface, and whenexceeding its full occupancy, randomly fluctu-ate in the bulk DRSS orderly replacing theirspread partners harvested from the surface.

Both carbon and FSSM DNAs scatter at thechosen 0.1µm precision limit intervals in a con-tinuous size distribution strongly skewedaround the 0.1-1.3µm modal peaks (Fig 2). Thispattern, which suggests intramolecular recom-bination as the source of Acetabularia mt-DNAheterogeneity [1], closely overlaps the size dis-

tribution of the free molecules and stablesupramolecular aggregates of DNA extrudedfrom A mitochondria by the osmotic shocktechnique [6]. This indicates that the DRSS iscapable of conservative DNA disentanglementand dissociation from in-vivo protein, and pos-sibly RNA, binding constraints [1,6,8] whichcontribute to the stability of the osmoticallyresistant protein-DNA [6] aggregates. Addingto the above are the punctual and selectiverescuing and preservation of such rare andunstable functional intermediates as theapproximately 1% replicating [1] and 10%recombining DNAs retrieved in carbon andFSSM of DRSS-spread A mitochondria or phe-nol-extracted A mt-DNAs (Mazza in prepara-tion). Neither the DRSS chemicals nor the pHcan be freely changed if the efficiency and pre-rogatives of the one-step method are to bepreserved. Only DRSS’s neutralisation or up topH 5 acidification seems to be compatible with

Figure 1: Acetabularia mitochondrial DNA release-spreadby the DRSS method. (a) Montage of images of concentrically arrangedmaxi DNA (72.4; 66.2; 35.4µm) spread on a noisyheterologously charged carbon film. (b) Mini (0.4-1.4 µm) circular DNA on the DNA-selecting, electron translucent amorphous silica ofsilicon oxide-coated grids rubbed dry against mica(FSSM). Scale bars = 1.5µm

P23-24-25 Mazza 13/3/02 12:26 pm Page 2

TEM O F DNA

MI C R O S C O P Y A N D AN A LY S I S • MA R C H 2002 25DIDYOUENJOYTHISARTICLE? DOYOUHAVE ATOPICYOUCOULDWRITEABOUT? CIRCLEREADERENQUIRYNO. 336 ORVISIT OURWEBSITE: www.microscopy-analysis.com

the method although leading to nucleoid-likeDNA reaggregation and exclusive carbonabsorption [1]. On the other hand, the relativeproportions of DRSS chemicals seem to needtentative adjustment within the range of con-centrations previously indicated, in as yet a pri-ori undefinable relation with the variableamounts or compositional patterns of the het-erologous test materials.

The above findings identifying the criticalrole of the pH, cations (organic and inorganic),detergent and DNA elastic status as regulatorsof assembly/disassembly equilibrium and sur-face spreading capacity, indicate number andtypes of physicochemical parameters heuristi-cally testable for the purpose of understand-ing the mechanism and patterns of functionalDNA-protein interaction in living organelles.This, together with the DRSS-DNA compatibil-ity and the possibility of straight-on experi-mental testing, makes both the strength pointand the prerogative of the one-step proce-dure conveniently assayable and adjustableunder whatever experimental conditions [7]devised to rescuing quiescent or functionallystimulated, but morphologically and dimen-sionally identifiable DNAs.

The sensitivity of the method, though lim-ited to maximum ~100µm circular, or some300µm conveniently folded linear DNAs, asallowed by the 83x83µm2 area of maximalmechanical resistance of the routine electronmicroscope film, has proven to be very effi-cient. In fact, it permits the rescuing of downto 10-3 individuals occurring within 0.1µm (or330 base pairs) to ~100µm circular (Figs 1a, b,2) and linear [1, and unpublished] monotonicDNA distributions at 0.1µm intervals with noidentification gaps and from populations 0.1to 10 times occupancy of the spreading inter-face. Coincidentally, there are indications thatby a micro-droplet version of the methodwhole genomes of even individual organellescan be rescued on either carbon or, better,FSSM grids [Mazza, in preparation]. The reso-lution has been presently fixed at ±0.09 µm totest the validity of the method even under the

limiting experimental conditions of the rou-tine electron microscope laboratory but canbe without effort amenable down to ~50 bpsby the use of appropriate internal standardsand optics dictated electron microscope pro-cedural precautions [8]. Both the sensitivityand resolutions however, as they are,favourably compare with the geneticallygrounded biochemical methodologies [7] inDNA both rescuing and sizing within thewhole spectrum of genomic molecular het-erogeneities, as known from indications of themost representative literature [1-9]. In addi-tion they implement applications of the one-step method to samples and molecules of verylimited amounts and size.

Leaving aside mechanistic interpretations,detailed and theoretically justified elsewhere(Mazza, in preparation), but stressing the phe-nomenological evidence, it must be pointedout that the one-step procedure worksthrough a two-phase separation pathway: theDRSS-dependent organelle disassembly withDNA still reversible to in-vivo nucleoidal orga-nization [1,8] by DRSS neutralisation/acidifi-cation [1], and the film-dependent physicalseparation of DNA from non-DNA moleculeswhich are trapped on close contiguous andrarely overlapping binding sites in carbon, butare selectively absorbed on, respectively,amorphous SiO/SiO2 and crystalline micaunder the FSSM procedure.

The above indicates that the DRSS methoddoes not definitely destroy the functionalinteractions of DNA-proteins and may permithighly conservative electron microscope analy-sis of in-vivo like morpho-functional repat-terning of DNA through pertinent chemical orenzymatic [10] testing of DRSS-suspendedsamples.

One last consideration on the rapidity of theone-step procedure. It ranges from the fewdays of the routine electron microscope labo-ratory to the possible few hours of the bestequipped and most sophisticated one.

All the above support the authors’ con-tention for the generalised use of the one-step

procedure when dealing with the problem ofthe organelle’s � 100µm circular, or 300µm lin-ear, DNA rescuing and sizing, and warrantsuseful applications in speeding the diagnosisof the 0.3-2.3µm human mt-DNA’s pathogenicdeletions [3-5]. Indeed, their rescuing on evenindividual cytoplasm of single manually enu-cleated and, if needed, protease/detergent-treated, DRSS-suspended cells, is a distinct per-spective of the one-step method.

R E F E R E N C E S1. Mazza A. and Casale A. Molecular heterogeneities of green

algae extranuclear genomes. Preliminary characterization oftwo coenocytic algae extranuclear genomes andphylogenetic implication. Intern. Mar. Biol. OceanographyXVII, Suppl., 509-528, 1991.

2. Small I. et. al. Evolution of plant mitochondrial genomes viasubstoichiometric intermediates. Cell 58, 69-76, 1989.

3. Tanaka-Yamamoto T. et. al. Specific amplification of deleted mitochondrial DNA from a myopathic patient and analysis ofdeleted region with S1 nuclease. Biochimica Biophysica Acta949, 297-304, 1989.

4. Schon E. A. et. al. A direct repeat is a hotspot for large-scaledeletion of human mitochondrial DNA. Science 244, 346-349,1989.

5. Shoffner J. M. et. al. Spontaneous Kearns-Sayre/chronicexternal ophtalmoplegia plus syndrome associated with amitochondrial DNA deletion: a slip-replication model andmetabolic therapy. Proc. Natl. Acad. Sci. U.S.A. 86, 7952-7956,1989.

6 Mazza A. et. al. Structural features of the mithochondrialgenome of Acetabularia Acetabulum. J. Submicrosc. Cytol.Pathol. 22, 627-634, 1990.

7. Sambrook J. et. al. 1989. Molecular cloning: a laboratorymanual. C.S.H. Laboratory Press. Cold Spring Harbor, N.Y.

8. Mazza A. and Casale A. 1992. The chloroplast genome: theheuristic potential of the ultrastructural genome analysisGiorn. Bot. Ital. 126, 1-89.

9. Kiyama R. et. al. Nature of recombination involved in excisionrearrangement of human repetitive DNA. J. Mol. Biol. 198,589-598, 1987.

10. Zimmerman S. B. and Trach S. O. Macromolecular crowdingextends the range of conditions under which DNApolymerase is functional. Biochimica Biophysica Acta 949,297-304, 1988.

©2002 Rolston Gordon Communications.

Figure 2:The left-kurtotic (0.1-1.3µm), 0.4µm modal skewedpopulation of the approximately 1700 DNA circulars of0.1 to 92.3µm contour length, extruded from Acetab-ularia mitochondria under the condition of one-steprelease-spreading electron microscopy. The 0.1µm sizeintervals reflect the chosen measurement’s standardprecision limit of ± 0.09µm (1).

P23-24-25 Mazza 13/3/02 12:27 pm Page 3