Published December 1, 1961
THE USE OF SPECIFIC A N T I B O D Y
IN ELECTRON MICROSCOPY
I. Preparation of Mercury-Labeled Antibody
FRANK
A. P E P E ,
Ph.D.
From the Department of Anatomy, School of Medicine, University of Pennsylvania, Philadelphia
ABSTRACT
INTRODUCTION
The specific reaction of an antibody with its
antigen has found histochemical application in the
fluorescent antibody technique (1, 2). The fluorescence of the fluorescein-labeled antibody in
ultraviolet light reveals the distribution of the specific antigen in a tissue section. Application of this
technique to finer problems of cell structure has
been limited by the resolution of the light microscope. Extension of the technique to electron microscopy would be possible if the combination of
antibody with a specific antigen in cell structures
could be visualized in the electron microscope.
One means of visualizing antibody in combination
with a tissue antigen in the electron microscope is
to increase selectively the intensity with which the
antibody molecules scatter electrons by introducing heavy metal into the antibody molecules. Such
a conjugated antibody may then be visualized as
an increase in density, in a particular region of the
cell; more ideally, individual conjugated antibody
molecules might be resolved. Another means of
detecting antibody in combination with a tissue
antigen is to observe the change in morphology
of a recognizable cell structure due to the adherence of antibody to the antigenic components
of the structure. In this latter case no special heavy
metal tag is necessary for identification of the antibody since visualization depends on a change in
shape rather than an increase in density. This
paper will deal with the introduction of heavy
metal into the antibody molecule; the following
paper will deal with its visualization by electron
microscopy. A subsequent paper will deal with
the visualization of antibody without specific
heavy metal tag.
The following considerations were taken into
account in choosing the heavy metal to be introduced into the antibody molecule: (a) The metal
must be of as high a molecular weight as possible
to be effective in electron scattering. (b) It must be
introduced in sufficient quantity to increase visibly
the electron opacity of the antibody molecules.
(c) It should be covalently linked directly or indirectly to the protein to eliminate the possibility
of dissociation from the antibody and subsequent
binding to other constituents of the tissue. Non-
515
Downloaded from on June 16, 2017
The preparation of antimyosin conjugated with mercury and fluorescein is described. The
mercury was introduced to permit visualization of the antibody in the electron microscope.
An organic mercurial, tetraacetoxymercuriarsanilic acid, was prepared and coupled to the
antibody through the diazonium salt. The fluorescein was coupled through the isocyanate
by a modification of the procedure described by Coons and Kaplan. The antibody conjugate retained its specificity of reaction with the tissue antigen. This was demonstrated by
the staining pattern obtained in fluorescence microscopy.
Published December 1, 1961
METHODS
For fluorescence microscopy a Reichert Zetopan research microscope was used. Dark-field illumination
was obtained using a Zeiss cardioid condenser and a
100 )< achromatic oil immersion objective containing
a built-in ultraviolet filter. The light source was an
Osram HBO-200 lamp. The combination of a 1 mm
Corning 5840 and a 1 mm Schott UG-1 filter was
used for the incident light. Exposures were of the
order of 5 minutes using Kodak type IIa-G spectroscopic plates and Kodak D-19 developer.
(4). ~,-Globulin from normal serum was isolated in
the same way for use in control experiments.
Preparation of the Organic Mercurial Tetraacetoxymercuriarsanilic Acid
30 gm of mercuric acetate (0.094 mole) was fused
and 3 gm of dry arsanilic acid powder (0.014 mole)
was added to the melt. This mixture was kept molten
by heating gently for 5 minutes and the viscous melt
was then poured into a beaker to cool. These procedures were performed under an efficient exhaust
hood because of the highly poisonous mercury and
arsenic. The hardened material was then scraped
loose and suspended in 100 ml of distilled water. An
additional 50 ml of water was added to the strawcolored suspension. 35 ml of 20 per cent N a O H was
then added with stirring to give a deep orange-red
precipitate. The precipitate was centrifuged out and
the resulting pellet was washed twice by resuspension
in 90 ml of 3.8 per cent NaOH. After the third centrifugation the pellet was suspended in 100 ml of
20 per cent acetic acid, which converted it to a white
flocculent precipitate. This precipitate was centrifuged and the pellet was washed once more with
55 ml of 20 per cent acetic acid by suspension and
centrifugation. The final pellet was dried immediately
by careful application of vacuum. The dry product
was ground to a powder. The rationale of these steps
was as follows: The initial sodium hydroxide washes
removed any unreacted arsanilic acid by dissolving it
as the sodium salt. In addition, any unreacted mercuric acetate was converted to the insoluble mercuric
oxide, giving the deep orange-red color. The acetic
acid wash then removed the mercuric oxide by converting it to the soluble acetate, leaving only the
relatively insoluble reaction product behind.
Analysis of Product
The colorimetric method of Cholak and Hubbard
(5) was used for mercury analysis. The Schwarzkopf
Microanalytical Laboratory, New York, performed
C, H, N, and As analyses.
Tetraacetoxymercuriarsanilic acid,
Preparation of Antibody and Normal Globulin
(NH2) C6(HgC2HaOs)4AsO3H2,
Myosin was extracted from chicken muscle by the
procedure of Mommaerts and Parrish (3) and was
further purified by three reprecipitations at an ionic
strength of 0.05. Approximately 20 mg of myosin in
4 ml of 0.3 M KCI solution was injected into rabbits
intraperitoneally once a week. Serum was collected
weekly during the course of injections once the titer
was developed. The serum from several rabbits was
pooled and the 7-globulin fraction of the serum was
isolated by the ethanol fractionation procedure
516
theoretically contains: C, 13.4 per cent; H, 1.3 per
cent; N, 1.1 per cent; As, 6.0 per cent; Hg, 64.1 per
cent. The compound obtained by the synthesis described above contained: C, 12.3 per cent; H, 1.1 per
cent; N, 1.0 per cent; As, 5.4 per cent; Hg, 63.7 per
cent. This analysis satisfactorily confirms the structure of the synthesized product. The compound
darkened at 229-230°C and did not melt at 265°C.
If it was left in the bright sunlight decomposition
T H E JOURNAL OF BIOPHYSICAL AND BIOCHEMICAL CYTOLOGY * VOLUME 11, 1961
Downloaded from on June 16, 2017
specific binding to cell structures through the
metal must be eliminated. (d) Introduction of the
metal into the antibody molecule must not eliminate the reaction between antibody and antigen.
(e) If the metal is first substituted in an organic
molecule, a functional group of this molecule must
be available for reaction with the antibody. (])
T h e a m o u n t of metal needed to give adequate
electron opacity to the final conjugated antibody
must not result in insolubility of the modified antibody. (g) The metal must be introduced by means
which do not substantially increase the size of the
antibody molecule, since this will be one factor
determining the precision of localization. (h) The
metal should be stable in the electron beam, or
methods for stabilizing it must be available. (i) In
addition, it would be advantageous to be able to
label the antibody with fluorescein as well as
the heavy metal to permit direct comparison of results obtained by fluorescence microscopy and electron microscopy.
T h e metal which showed the greatest promise of
fulfilling these criteria was mercury. Mercury consequently was coupled to the antibody indirectly
through a diazonium link. I n addition, fluorescein
was coupled to the antibody by use of the fluorescein isocyanate. The resulting modified antibody retained its specificity to its tissue antigens in
spite of the extensive coupling.
Published December 1, 1961
occurred and droplets of mercury were formed. It
was therefore stored in the dark.
Preparation of Antibody and Normal Globulin
Conjugates
changes of 25 per cent glycerol in M/60 PO4 buffer
p H 7.4. Analyses showed that no mercury was lost
on storage or extensive dialysis. T h e conjugates were
stored at -- 24°C.
RESULTS
AND
DISCUSSION
T h e reactions of m e r c u r y with o r g a n i c c o m p o u n d s
h a v e b e e n studied extensively b u t are n o t clearly
u n d e r s t o o d . M e r c u r y is k n o w n to f o r m derivatives
with m a n y a r o m a t i c o r g a n i c molecules, substituting directly onto the b e n z e n e ring. T h e p r e p a r a tion of heavily s u b s t i t u t e d m e r c u r y derivatives of
aniline a n d acetanilide has b e e n r e p o r t e d (7).
T h e s e c o m p o u n d s are e x t r e m e l y insoluble. T h e
p r o c e d u r e d e s c r i b e d for p r e p a r a t i o n of the tetraTABLE
I
Mercury Content of Conjugates
Experiment no.
Protein
Hg atoms per
molecule "yglobulin
Conjugates of rabbit T-globulin with the diazonium
salt of tetraaeetoxymercuriarsanilic acid:
X-1
X-2
R a b b i t normal T-globulin
R a b b i t antibody
T-globulin
32.4
23.0
Conjugates of mercurated rabbit T-globulin with
fluoreseein isocyanate :
X-1-A
X-2-A
R a b b i t normal T-globulin
R a b b i t antibody
T-globulin
19.5
13.2
a c e t o x y m e r c u r i a r s a n i l i c acid is similar to t h a t rep o r t e d for the p r e p a r a t i o n of t h e m e r c u r y d e r i v a tive of acetanilide. T h e p u r p o s e of t h e arsonic acid
g r o u p was to increase the solubility of the final
c o m p o u n d . T h e presence of the a m i n o g r o u p perm i t t e d d i a z o t i z a t i o n a n d c o u p l i n g of the m e r c u r i a l
to the globulins t h r o u g h the d i a z o n i u m salt. It was
found t h a t the c o u p l i n g r e a c t i o n r e q u i r e d a p H of
9-10. A t this p H , c o u p l i n g occurs p r i m a r i l y at t h e
tyrosine a n d t r y p t o p h a n e residues (8, 9) of t h e
protein. T h e m a x i m u m c o u p l i n g o b t a i n e d in this
w a y was 32 m e r c u r y a t o m s p e r molecule of
T-globulin. S u b s e q u e n t c o u p l i n g of fluorescein
isocyanate to t h e m e r c u r a t e d 3,-globulin b y m e a n s
o f the m o d i f i e d p r o c e d u r e a l r e a d y d e s c r i b e d resulted in loss of s o m e of t h e m e r c u r y , as is s h o w n
by the e x a m p l e s in T a b l e I. T h e v a r i a t i o n in t h e
d e g r e e o f c o u p l i n g is n o t n o w u n d e r s t o o d .
F. A. PEPE
Preparationof Mercury-LabeledAntibody
517
Downloaded from on June 16, 2017
Preparation of the mercury conjugates required
two steps: first, diazotization of the amino group of
tetraacetoxymercuriarsanilic acid to form the diazonium salt; then coupling of this to the antibody.
(1) T h e diazotization reaction was complicated by
the low solubility of the mercurial. T h e procedure
finally used was as follows: To 15 mg of the tetraacetoxymercuriarsanilic acid was a d d e d 3.7 ml of
20 per cent acetic acid and a few drops of an NaNO2
solution (2.73 gm d- 100 ml H~O) to give a slight
excess of NaNO~. T h e excess NaNO~ was detected
by the appearance of free iodine when a drop of the
reaction mixture was mixed with a drop of a solution
of K I on a white plate. T h e reaction mixture was
stirred 5 minutes with a magnetic stirrer at 0°C. T h e
mercury c o m p o u n d did not go into solution completely. T h e solution, while being stirred, gradually
became straw-colored and then a rose-purple color.
T h e insoluble material remaining after stirring became intensely colored. (2) T h e diazonium salt
formed was coupled to the antibody as follows: T h e
solution and undissolved material obtained above
were a d d e d rapidly to 4.0 ml of a solution of the protein (in M/60 PO4 buffer p H 7.4, 0.1 M KCI) with
vigorous stirring at 0°C. Immediately 8.4 ml of cold
6 per cent N a O H solution was added with rapid
stirring to bring the p H to 9-10. This mixture was
stirred on a magnetic stirrer at 0°C for 1 hour. T h e
soluble material was then dialyzed for 2 days at 10°C
against 100 volumes of M/60 P04 buffer p H 7.4,
0.3 M KCI, which was changed twice daily.
T h e resulting globulin-mercury conjugates were
further conjugated with fluorescein. T h e procedure
of Coons and K a p l a n (6) for the conjugation of
fluorescein isocyanate to protein resulted in complete
loss of mercury. Therefore the following modified
procedure was used: A solution of the globulinmercury conjugate was brought to p H 9-10 by dropwise addition of 6 per cent N a O H solution at 0°C,
with constant stirring. For each 50 m g of protein, 1 ml
of fluorescein isocyanate in acetone solution (Sylvana
Chemical Co., Orange, New Jersey) was added
slowly, with stirring at 0°C. T h e p H of the solution
was readjusted to 9-10 after addition of the isocyanate. T h e reaction mixture was stirred for 1 hour on
a magnetic stirrer and the p H was readjusted to 9-10
if necessary. It was then stirred overnight and was
finally dialyzed at 10°C against M/60 PO4 buffer
p H 7.4, 0.3 M KC1, until no fluorescence was detectable in the dialysate. Finally, the conjugated globulins
were dialyzed for 2 days at 10°C against several
Published December 1, 1961
Downloaded from on June 16, 2017
I~IGURE ]
T h e specificity of the staining reaction of the doubly conjugated antibody. X 2300.
a. N o r m a l untreated fibril in phase contrast microscopy.
b. Same fibril as in I a after treatment with doubly conjugated n o r m a l "r-globulin.
c. Same fibril as in 1 b in fluorescence microscopy.
d. Same fibril as in 1 c after treatment with doubly conjugated antimyosin.
e. Same fibril as in 1 d in phase contrast microscopy.
518
THE JOURNAL OF BIOPHYSICAL AND BIOCHEMICAL CYTOLOGY • VOLUME 11, 1961
Published December 1, 1961
1e. Therefore, the specificity of the staining reaction of the antibody after double conjugation, with
both the organic mercurial and the fluorescent
labels, has been established.
Another method for labeling serum globulin, to
permit electron microscopy of antibody staining,
has been described recently by Singer (13). In
Singer's method, the iron-containing protein,
ferritin, was coupled to the globulin molecule
through a ureide link. The use of ferritin conjugates for the staining of tissues has not yet been
reported, but it is obvious that a comparison of
the method with that reported in this series of
studies would be desirable. One may anticipate
two major problems in the use of such ferritin
conjugates for studies on cells or tissues: First, the
large size of the ferritin molecule limits the precision of localization available with the ferritinantibody conjugate. R a b b i t antibody molecules
have been shown, by shadow casting, to be rods
with dimensions of approximately 40 by 250 A (14).
The ferritin molecule itself is a sphere with a
diameter of approximately 100 A (15). I n the
case of the conjugate labeled with organic mercurial and fluorescein, the size of the conjugate
molecule should not be appreciably increased by
addition of these smaller molecules. Secondly,
because of the large size of the ferritin-antibody
conjugate, adequate penetration of the antibody
into the tissue will be difficult. This has been a
problem even with unconjugated antibody, as is
pointed out in paper I I I of this series.
This investigation was supported by Research Grant
C-1957 from the National Cancer Institute of the
National Institutes of Health, United States Public
Health Service.
Received for publication, January 10, 1961.
BIBLIOGRAPItY
1. Coons, A. H., General Cytoehemical Methods,
(J. F. Danielli, editor), New York, Academic
Press, Inc., 1958, 1,400.
2. Coons, A. H., Internat. Rev. Cytol., 1956, 5, 1.
3. MOMMAERTS,W. F. H. M., and PARRISH,R. G.,
J. Biol. Chem., 1951, 188, 545.
4. DEUTSCH, H. F., Meth. Med. Research, 1952, 5.
284.
5. CHOLAK, J., and HUBBARD, D. M., Ind. Eng.
Chem., Anal. Ed., 1946, 18, 149.
6. Coons, A. H., and KAPLAN, M. H.. J. Exp. Med.,
1950, 91, 1.
7. WHITMORE, F. C., Organic Compounds of
Mercury, American Chemical Society Monograph, 1921, 234.
8. PUTNAM,F. W., The Proteins, (N. Neurath and
K. Bailey, editors), 1953, 1B, 947.
9. EAGLE,H., and VICKERS, P., J. Biol. Chem., 1936,
114, 193.
10. FINCK, H., HOLTZER, H., and MARSHALL,
J. M., JR., J. Biophysic. and Biochem. Cytol.,
1956, 2, No. 4, suppl., 175.
1l. MARSHALL,J. M., JR., HOLTZER, H., FINCK, H.,
F. A. PEPE Preparation of Mercury-Labeled Antibody
519
Downloaded from on June 16, 2017
The specific antibody used in this work was that
prepared against the muscle protein myosin. The
distribution of myosin in the glycerinated myofibril
has been studied previously by fluorescence microscopy (10-12). In spite of extensive coupling, the
doubly labeled antibody remained specific in its
reaction with the antigen.
This was demonstrated by staining a homogenized fibril preparation of glycerinated chicken
neck muscle. The homogenization was done in
25 per cent glycerol, M/60 PO4 buffer pH 7.4, and
it was stained with antibody in the same solvent.
Staining was done under the microscope by drawing the solutions under the coverslip with filter
paper. Photomicrographs were taken at each step
in the procedure by phase contrast and fluorescence microscopy. No visible change occurred
in the fibril as observed by phase contrast microscopy after treatment with conjugated normal
globulin and wash solution, as isshown in Figs. 1 a
and b. Fluorescence microscopy after treatment
with normal globulin also showed no visible nonspecific staining (Fig. 1 c). The faint image visible
in Fig. 1 c is the dark-field image due to a combination of some unfiltered visible light coming
through the fluorescence optical system and
some autofluorescence of the proteins of the
fibril. This photograph was purposely overexposed
to bring out any faint staining which might have
occurred. This same fibril was then treated with
doubly conjugated antibody globulin followed by
wash solution. The fluorescence micrograph of
this fibril, Fig. 1 d, revealed intense alternating
bright and dark areas corresponding to fluorescent
A bands and non-fluorescent I bands respectively.
This is the same pattern of staining obtained with
antibody labeled with fluorescein alone (10-12).
The corresponding phase micrograph is seen in Fig.
Published December 1, 1961
and PEPE, F., Exp. Cell Research, 1959, 7,
Suppl., 219.
12. HOLTZER, H., MARSHALL, J. M., JR., and
FINCK, H., J. Biophysic. and Biochem. Cytol.,
1957, 3, 705.
13. SINGER, S. J., Nature, 1959, 183, 1523.
14. HALL, C. E., NlsogoFr, A., and SLAYTER, H. S.,
J. Biophysic. and Biochem. Cytol., 6, 407.
15. FARRANT,J. L., Biochim. et Biophysica Acta, 1954,
13, 569.
Downloaded from on June 16, 2017
520
THE JOURNAL OF BIOPHYSICAL AND BIOCHEMICAL
CYTOLOGY • VOLUME
11, 1961