Staining Methods, Histochemistry
Paraldehyde is a controlled substance, not that easy to obtain
for laboratory use, and it also has a short shelf life. Is there
a way to make the aldehyde-fuchsine stain without using
When aldehyde fuchsine is made in the traditional way, the
paraldehyde decomposes in the presence of acid, yielding
acetaldehyde. This reacts with pararosaniline to form a new
dye, which is the active component of the stain. It is therefore
possible to use acetaldehyde (obtainable from regular chemical
suppliers) instead of paraldehyde.
Peggy Wenk, bless her heart, commented on this in the Journal of
Histotechnology, vol 10, #4 (December 1996): Acetaldehyde as a
substitute for paraldehyde.
2.5 ml acetaldehyde is used in place of 1.5 ml paraldehyde. The
working solution must be refrigerated. It will stain hepatitis B
for 3 – 4 weeks, but is good for elastin for several months.
Acetaldehyde cost about $30 for 100 ml and is stable in a
refrigerator for about 2 years. (Paraldehyde is stable for only a
few months after opening, and is pricey due to handling/admin
fees.) You need to be aware that acetaldehyde is a flammable
liquid that boils at 21 C. The bottle must be cold when you open
Having struggled trying to get paraldehyde, this substitution has
made aldehyde-fuchsin staining feasible in a research laboratory.
[ With some editing and additional comments by J. A. Kiernan ]
** Phosphatases in decalcified, embedded tissue.
Can acid phosphatase activity still be demonstrated in formalin
fixed, decalcified, paraffin embedded bone sections?
Have a go, I used to stain for acid phosphatase in 1-10 æm
sections of demineralized, glutaraldehyde/osmium-fixed
epoxy-embedded specimens with no bother. The method was nothing
special, just a standard napthol AS-BI phosphate/diazotised
While we’re at it, how about alkaline phosphatase in
ethanol-fixed, methacrylate embedded sections? Try: McGadey,J.
1970. Histochemie 23. 180-184. Tetrazolium method for
non-specific alkaline phosphatase. This is an excellent method;
it has never let me down in any application.
I routinely do acid phosphatase staining on formic
acid-decalcified GMA-embedded bones. Alkaline phosphatase can
also be demonstrated in the GMA and is retained by the alcohol
fixation. The problem that I have found with trying to both from
the same block is that the acid phosphatase stains much better
with formalin fixation and the alkaline phosphatase stains
better with alcohol fixation.
I have had good results with acid phosphatase using formic acid
decalcification and paraffin embedding of rodent skull. An
excellent article is C. Liu et al. “Simultaneous demonstration
of bone alkaline and acid phosphatase activities in plastic
embedded sections and differential inhibition of the
activities.” Histochemistry 86:559-565, 1987,
Skeletech, Inc., Kirkland, WA
** Congo red for amyloid
Why does alkaline Congo red stain amyloid feebly in
sections of some specimens but not others?
Here is one possibility. In developing the alkaline Congo red
method, Dr. Holde Puchlter noticed decreased staining with
prolonged fixation in formalin or NBF. This decrease even
applied to unstained sections stored under conditions where
formaldehyde was present in the ambient air.
Medical College of Georgia, Augusta
** Cartilage staining with safranine
How do you stain cartilage with safranine?
The Safranin O method for Cartilage goes like this;
1. Dewax section and take to water.
2. Stain nuclei with a suitable iron haematoxylin.
3. Blue in running tapwater.
4. Rinse in distilled water.
5. Stain with 1% light green diluted 1 in 5 with distilled
water, for 3 minutes.
6. Rinse in 1% acetic acid.
7. Stain with 0.1% Saffranin O, for 4 – 6 minutes.
8. Rinse in 1% acetic acid and check under microscope. Any
overstaining with Safranin can be modified by re-applying the
light green solution briefly, and vice versa.
9. Dehydrate with alcohol,clear and mount.
(Modified from the method in Lillie’s “Histopathological Technic
and Practical Histochemistry”)
** Stain for Chlamydia (Castaneda’s method).
How do you carry out the Castaneda stain for Chlamydia?
Castaneda’s stain for elementary bodies and Rikettsiae (1930)
Castaneda’s staining solution
Potassium dihydrogen phosphate, anhydrous 1 g
Disodium hydrogen phosphate 25 g
Distilled water 1000 ml
Formalin 1 ml
Dissolve the potassium dihydrogen phosphate in 100 ml distilled
water and the disodium hydrogen phosphate in 900 ml distilled
water. Mix the two solutions to give a buffer pH 7.5, and add
formaldehyde as a preservative.
Methylene blue 1 g
Methanol 100 ml
Solution A 20 ml
Solution B 0.15 ml
Formalin 1 ml
Safranine (0.2% aqueous solution) 1 part
Acetic acid (0.1% aqueous solution) 3 parts
* Prepare films from infected tissue and dry in air
* Apply the stain for 3 min.
* Drain, do not wash
* Counterstain for a 1-2 seconds in safranine-acetic acid
* Wash in running water, blot dry.
Rickettsiae, elementary bodies of psittacosis: blue. Cell nuclei
and cytoplasm: red.
Reference: “Biological stains and staining methods.” BDH
Several modifications of Castaneda’s original technique are
given in: Langeron, M.:”Precis de Microscopie”, 1934 and 1948.
** Which staining method for copper is best?
Which histochemical staining method is best for copper in
human or animal tissues? The choice seems to be between
rubeanic acid (not in catalogs) and some impossibly
long name that ends in “rhodanine.”
This question to the HistoNet listserver elicited
replies that generally favored the “rhodanine”
reagent over “rubeanic acid.” Nomenclature can be
confusing! Don’t confuse rhodaNine with rhodaMine,
and note that in any chemical catalog, p-dimethyl-
is indexed under the letter D, not P. A few
general references for copper histochemistry are
added at the end of this FAQ item.
Rubeanic acid is H2NCSCSNH2 and is listed in catalogs as
Dithiooxamide (by Aldrich, Sigma and other vendors).
I prefer the “rhodanine” method for the demonstration
Fixation: 10% neutral buffered formalin.
Embedding: Paraffin sections cut at 6 microns
Distilled water, preferably deionized, should be used in
all solutions and rinses.
Rhodanine saturated solution (stock) –
p-Dimethylaminobenzylinene-rhodanine 0.2 g
Absolute ethanol 100 ml
Rhodanine solution (working) –
Rhodanine saturated solution (stock) 6 ml
Distilled water 94 ml
Diluted Mayer’s hematoxylin
Mayer’s hematoxylin 50 ml
Distilled water 50 ml
0.5% aqueous sodium borate (borax)
Note: The use of chemically clean glassware is necessary.
Shake stock solution before measuring and mixing
solutions and shake the working solution before
pouring it onto the slides.
1. Hydrate slides to distilled water.
2. Incubate slides in rhodanine working solution
at 37 degree C for 18 hours.
3. Wash slides well in several changes of distilled water.
4. Stain slides in diluted Mayer’s hematoxylin for 10 minutes.
5. Rinse slides with distilled water.
6. Quickly rinse slides in 0.5% sodium borate.
7. Rinse slides with distilled water.
8. Dehydrate slides through 95% alcohol to absolute ethanol,
clear, and coverslip with a synthetic mountant.
Copper – orange/red.
Tissue elements – light blue.
Eric C. Kellar
University of Pittsburgh Medical Center
A few references for copper histochemistry.
Irons,RD; Schenk,EA; Lee,CK (1977): Cytochemical methods for
copper. Archives of Pathology and Laboratory Medicine 101,
Cytochem methods for copper. Comparison of dithiooxamide,
Pearse, AGE (1985) Histochemistry, Theoretical and Applied,
4th ed. Vol. 2.
Metal histochemistry is extensively reviewed in Chapter 20.
Szerdahelyi,P; Kasa,P (1986): A highly sensitive method for the
histochemical demonstration of copper in normal rat tissues.
Histochemistry 85, 349-352.
Highly sensitive histoch method for Cu histochemistry.
Magnesium-dithizone, followed by silver intensification.
Szerdahelyi,P; Kasa,P (1986): Histochemical demonstration of
copper in normal rat brain and spinal cord. Histochemistry
Histochemical demonstration of Cu in normal brain,
John A. Kiernan
** Diastase (amylase) control for glycogen
Which is better as a control for glycogen staining:
alpha-amylase or human saliva?
The bought enzyme (10 mg/ml, in water) takes about 10 minutes
to remove the stainable glycogen from a section of liver. The
enzyme is not very expensive.
Saliva is free, and it takes about 30 minutes, but some people
don’t enjoy spitting, or even dribbling, onto their slides. A
theoretical disadvantage of spit is that it contains plenty of
digestive enzymes additional to amylase (= diastase), notably
ribonuclease and various proteases. However, these are unlikely
to remove substances with the same staining properties as
John A. Kiernan
** Evans blue, trypan blue and eosin as tracers.
Can Evans blue be used as a tissue dye, and will it safely wash
out of the tissue during routine paraffin processing? The
object is to trace a catheter leakage then have the dye wash out
of the tissue during processing. Would eosin be OK for the same
Evans blue is an anionic dye with large molecules, closely
related to trypan blue. It was formerly used (? still is in
some places) to measure blood volume, because it binds to serum
proteins and stays in the circulation for a few hours. When it
leaves the blood, some of it sticks to collagen (the elongated
dye molecule favours this) and some is taken into cells,
including macrophages and neurons. The dye-protein complex is
fluorescent (red emission) and this was the first fluorescent
tracer of neuronal uptake and retrograde axonal transport.
Applied to sections, trypan blue stains everything and can be
washed out completely. Slight alkalinity speeds up the
procedure. In the presence of another anionic dye with smaller
molecules (like picric acid), trypan blue becomes selective for
collagen, but is no match for acid fuchsine or sirius red F3B.
I’m sure Evans blue, which is a VERY similar compound, would
have identical properties as a stain.
So: if you want to get rid of the Evans blue, wash the
specimens in slightly alkaline water.
Eosin could also be used in the same way. If you’re after very
small leaks from your catheters, eosin might be more sensitive,
because it’s quite strongly fluorescent even without binding to
anything (green-yellow emission). You could turn off the lab
lights and use a Woods light to watch for leaks. Eosin is also
removable by slightly alkaline water or by alcohols. Acidic
reagents precipitate the insoluble colour-acid.
Evans blue and trypan blue both can be used to determine cell
vitality – live cells exclude the dye(s), dead cells take then
up – the trypan (Evans) blue exclusion test.
As far as catheter leakage is concerned, a fluorescent dye would
certainly be a good choice. Cavers use them to trace
underground rivers, and fluorescent dyes are used for a similar
purpose in opthalmology.
Russ Allison, Wales
** Gallyas’ stain
What is the Gallyas Stain, and what is it for?
Ferenc Gallyas, in Hungary, has been studying and inventing
silver stains for at least 30 years. They all involve the use of
“physical developers” (an ancient and obsolete term from
photography). A physical developer is a mixture containing
silver ions and a reducing agent, made stable for several
minutes or even a few hours by other additives. Gallyas
introduced silicotungstic acid as a stabilizer. Earlier physical
developers used gum arabic, gum mastic, albumen, albumin (no,
they aren’t the same) and other organic macromolecules.
The name of Gallyas is most often connected with his methods for
Alzheimer’s neurofibrillary tangles because neuropathologists
are, by noble tradition, the biggest users of silver staining.
However, there are several other silver staining methods, for a
range of tissue components, developed by Gallyas. His work
probably forms the rarely acknowledged basis of
immunogold-silver amplification for light microscopy and for
some of the silver methods used to detect minute amounts of
protein in Western blots.
Physical development was discovered, for photography and
histology, by Liesegang (1911), and reintroduced to histological
practice in 1955 by Alan Peters, who went on to become a great
authority on the ultrastructure of nervous tissue, especially
that of the cerebral cortex.
I don’t know if this really answers the question, but it’s
interesting to look at the way someone’s name gets attached to
a method, even if at first there’s doubt about _which_ method.
** Gram staining of sections (Brown & Hopps method).
I just did a B & H gram stain for the first time. All tissue
stained various shades of purple against a clear background.
There was no yellow or red staining at all. The protocol I used
replaced all acetone differentiation steps with 95% ethanol,
“to avoid over-decolorizing.”
What am I doing wrong? Should I:
1. Use acetone instead of 95% ethanol, or a combination of
2. Use saturated aqueous picric acid?
3. Use 0.1% basic fuchsin (instead of 0.01%)?
The following modifications of Brown & Hopps give consistent
differentiation of Gram negatives with reduced risk of
over-differentiation. Cellosolve is used instead of acetone, and
tartrazine instead of picric acid.
The crystal violet staining is as in the original method.
Modifications are as follows:-
Substitute Lugol’s or Jensen’s iodine for Gram’s to give a
stronger crystal violet-iodine complex.
Use cellosolve (= ethylene glycol monoethyl ether = 2-ethoxyethanol)
as decoloriser. The smell can be unpleasant, but it is slower in
its action and more easily controlled.
Use 0.5% basic fuchsine, for 5 mins, to counterstain the Gram
After rinsing with water apply Gallego’s differentiator
(1% acetic acid with 2% formalin, in water) for 5 mins.
Rinse with water and flood sections with 1.5% tartrazine for
Rinse the slides with water. Now take one slide at a time:
blot with filter paper, flood with cellosolve for 6 – 10
secs, blot again, and then place slide directly in xylene,
2 or 3 changes
Coverslip and mount. Repeat with the remaining slides, one at a
The extra step with the cellosolve seems to remove excess
fuchsine from cytoplasmic elements in the background, thereby
increasing visibility of Gram-negative bacteria.
Lab. Manager, Aust.Biostain.
** Oxidants for hematoxylin
Can a less toxic oxidizing agent be substituted for mercuric
oxide in Harris’s alum hematoxylin?
Yes. Mercuric oxide for the oxidation of hematoxylin in Harris’s
hemalum can be replaced with sodium iodate (NaIO3) or other
According to Hansen (1895), one of the following is, in general,
needed for the oxidation of 1 gm of hematoxylin to hematein:
* KMnO4: 177 mg
* KClO3: 114 mg
* KIO3: 200 mg
* NaIO3: 197 mg
* KCr2O7: 276 mg
It is advisable to use only half of these quantities, to delay
over-oxidation. Vacca (1985) suggested 75 mg NaIO3 per gm
hematoxylin, and P. Bock (1989) suggested 98.5 mg NaIO3 per gm
Bock, P.: Romeis’ Mikroskopische Technik; 1989
Hansen, F.C.C.: Eine schnelle Methode zur Herstellung des
Bohmersen Hematoxylins. Zoolog. Anz. 473; 1895.
Vacca: Laboratory Manual of Histochemistry; 1985.
Almost every hematoxylin can be used regressively, my favorite
for general histology is:
“Mayer’s acid hemalum, modified by Lillie”:
“Dissolve 5gm hematoxylin by holding overnight in 700 ml
distilled water; add 50 gm ammonium alum and 0.25 gm NaIO3.
After these have gone into solution, add 300 ml glycerin C.P.
and 20 ml glacial acetic acid. May be used immediately; stain
for 5 min.”
Procedure. (5-7 æm paraffin sections, fixation: Bouin; manual
* Sections to distilled water.
* Sections to alum-hematoxylin (3 min).
* Sections to acid alcohol (2-3 dips or until differentiated).
* Rinse sections in tap water (about 10 sec, until most of
the acid alcohol has dissapeared from the slide).
* Rinse sections in 1% NaHCO3 in distilled water (1 min).
* Rinse sections in distilled water (1 min).
* Sections in 0.5% eosin Y in distilled water (30 sec).
* Rinse sections in distilled water (a few dips, until most
of the “free” eosin has dissapeared).
* Dehydrate, clear, mount.
** McFaydean’s stain for anthrax bacilli
What is M’Faydean’s stain?
[ This has been put together from three replies
to a question raised on the HistoNet newsgroup. ]
M’Faydean’s stain is a simple stain using any well
polychromed methylene blue (e.g. aged Loefflers). It is
applied to heat-fixed smears for 10-30 seconds.
Polychroming (demethylation) is traditionally achieved by
exposure of Loeffler’s soln. to light and air for several
months until it acquires a purplish tinge. However the
oxidation process can be accelerated by application of heat
as in Unna’s method. (G. Gurr, 1963 p. 88 & 91); also E.
Gurr, 1960, pp. 264-268).
Loeffler’s methylene blue:
Methylene blue 0.5 g
1% w/v Aq. KOH 1.0 ml
Ethanol 30.0 ml
D.water 70.0 ml
Warm the water to 50C., stir in methylene blue and
add other ingredients, cool and filter before use.
Polychrome methylene blue (Unna):
Methylene blue 1.0 g
Pot. carbonate 1.0 g
Ethanol 20.0 ml
D.water 100 ml
Dissolve methylene blue in water, add pot. carb. and
alcohol, place in boiling water bath and evaporate
to 100 ml.
Any other polychrome methylene blue formulation should work
Results: Bacilli appear Navy Blue with Anthax showing a
narrow area (capsule) around and between bacilli that is
reddish purple (metachromatic). A strong word of warning:
many species of bacillus may also be encapsulated, e.g.
Cereus etc. If you produce any positives get them confirmed
at a Reference Microbiology Lab. for Infectious Diseases, or
try the Armed Forces Institite of Pathology.
Gurr doesn’t give any further references in his book as to
McFadyean, whether the method was published or by personal
“Encyclopedia of microscopic stains,” by Edward Gurr.
London: Arnold, 1960. (pp 264-268)
“Biological Staining Methods.” by George T. Gurr.
7th Edition. 1963. (Published by George T. Gurr Ltd.
136-144, New King’s Road, London, S.W.6.)
Australian Biostain P/L
** Microglia with Griffonia lectin.
I have been trying to stain for microglia in paraffin sections
of rat brain using peroxidase-labeled Griffonia simplicifolia
lectin (GSI-B4-HRP) from Sigma. It has been used in various
papers for staining of active and resting microglia but I
cannot seem to get it to work. Are there any tricks that I might
I have not used this lectin for microglia but have used it for
other things. The purity varies considerably because the seeds
of Griffonia, when extracted, may yield just one lectin or
several isolectins (depending on the seeds), and the B4 lectin
is then purified from this mixture. I have found a lot of
variation from batch to batch but more so from manufacturer to
manufacturer. The best luck I had with this lectin was from
Vector Laboratories, Burlingame, California, who specialize in
the production of lectins. I have also had problems with some
lectin-HRP conjugates. In my experience the conjugates
(especially the HRP ones) have only a limited shelf life and
this can lead to background staining. Part of your problem may
be that lectin binding can be significantly altered by fixation
and processing. I would suggest that you first try it on frozen
sections to determine whether the conjugate you have is working.
This lectin usually requires the availability of calcium ions to
bind. If you are using OCT freezing compound, this contains
sufficient calcium if you don’t remove the OCT before staining.
I do not have the latest Vector catalog available at the moment
but believe that they have an antibody against GSI B4. This
might be a better approach if the problem is one of conjugate
breakdown or excessive background staining.
Another point is that the lectin binding can be easily confirmed
with negative (inhibited) controls, inhibitors for GSI B4
Barry R. J. Rittman
Univ. Texas HSC Dental Branch
** Picro-sirius red staining
I have been asked to do a “picrosirius” staining procedure.
What is it?
Picro-sirius red is a solution of sirius red F3B (0.1%) in
saturated aqueous picric acid. It is typically used after an
iron haematoxylin nuclear stain, much as Van Gieson, but for 60
minutes. Rinse in slightly acidified water and dehydrate in
three changes of absolute alcohol. The result is similar to Van
Gieson (Collagen red, cytoplasms & red cells yellow) but sirius
red shows thinner fibres that are often missed by Van Gieson.
The real difference is seen by using a polarizing microscope.
With crossed polars the collagen fibres, even very thin ones,
appear in brilliant orange, yellow and green colours against a
black background. Basement membranes, though stained, do not
exhibit this birefringence because their collagen fibres are not
The dye isn’t one of those certified by the Biological Stain
Commission, and some major American vendors do not have it in
their catalogues. The stuff in my lab was bought from BDH (Gurr)
about 15 years ago. There are many synonyms. The Colour Index
application name is Direct red 80, and the CI number is 35780.
Don’t use a dye that is not CI 35780 even if it has the words
sirius and red in its name.
Puchtler H & Sweat F 1964. Histochemie 4, 29-54
Puchtler H, Sweat FS & Valentine LS 1973.
Beitr. Pathol. 150, 174-187
Junqueira LCU, Bignolas G & Brentain RR 1979.
Histochem. J. 11, 447-455
Lillie RD 1977. Conn’s Biological Stains, 9th ed.
Baltimore: Williams & Wilkins.
Colour Index CD-ROM (1997) Society of Dyers & Colourists,
John A. Kiernan,
** Iron hematoxylin: ripening not needed.
Why does Bancroft and Stevens tell me to ripen my alcoholic
hematoxylin for a month, when the ferric chloride oxidizes it
instantly when you combine the two parts?
Because B & S is wrong (a very unusual thing in that superb
book), and you are right.
For what it’s worth, my experiences and occasional experiments
fully support the conclusions written in the classical works of
Baker, Lillie, Gabe and others. Ferric ions instantly oxidize
hematoxylin to hematein and they also form part of the black
complex that is retained in cell nuclei.
** Enzyme histochemistry on cell cultures
How do you perform enzyme histochemistry (NADH Dehydrogenase,
succinic Dehydrogenase, cytochrome oxidase)on cultured cells
grown on slides? Would you use a detergent (or other means) to
permeabilize membranes prior to application of the reaction
I just take the coverglass from the culture medium, give it a
rinse in buffer, incubate for required time, wash gently, then
mount. No fixing, no detergent; just incubate and mount. It
works, so why complicate matters?
** Malachite green in stain for Cryptosporidium
How do you do a malachite green stain for Cryptosporidium?
The Cryptosporidia are stained by carbol fuchsine; malachite
green is a counterstain for the background.
This is the procedure I use. (I also do the parasitology
here.) It works fairly well but is not the best diagnostic
technique for Cyrptosporidia. There are Meriflour commercial
kits that are better than this stain.
A MODIFIED ZIEHL-NEELSEN TECHNIQUE FOR CRYPTOSPORIDIUM
This is used on fecal smears.
Concentrated carbol fuchsine
10 ml 95% ethyl alcohol
0.3 gm Basic fuchsine
6 ml Liquid Phenol
94 ml Distilled water
Combine in the listed order.
10% Sulfuric Acid
10 ml Sulfuric acid
90 ml Distilled water
5% Malachite Green
95 ml Distilled water
5 gm Malachite green
1. Make a thin smear from the fecal sample.
2. Dry the smear at room temperature.
3. Fix the smear in absolute methanol for 2-5 minutes.
4. Dry at room temperature
5. Fix briefly in a flame.
6. Stain with concentrated carbol fuchsine for 20-30
minutes without heating.
7. Rinse in tap water.
8. Differentiate with 10% sulfuric acid for 20-60 seconds.
(Concentrations from 0.25 to 10% can be used; we use
10% sulfuric acid.)
9. Rinse in tap water.
10. Counterstain with 5% malachite green for 5 minutes.
11. Rinse in tap water.
12. Dry at room temperature.
13. Examine under oil.
14. Cryptosporidia will stain bright red with a
Penn State University
** Confusing dye names (lissamine fast red as an example)
Is there another name for Lissamine Fast Red? I can’t find it
under this name in any dye catalog.
Five or six people identified at least three different dyes
in the answers to this HistoNet query in August 1998. This
emphasizes the importance of identifying dyes by Colour
Index numbers whenever possible. A name like “Lissamine”
has no chemical significance and may be attached to
widely differing compounds! Some opinions follow (mine
is No. 3). Probably all are correct, and there are
different uses for the simlarly named dyes.
J. A. Kiernan
1. Another name for Lissamine Fast Red is Acid Red 37. You can
try BDH with next Cat no 341772K and it comes in 25 gram
2. I suspect that the dye you’re looking for is Sulforhodamine B,
also known as Lissamine rhodamine B 200, Acid rhodamine B.
The dyers assoc. refer to it as C.I.Acid Red 52. Its C.I.Number
is C.I. 45100.
3. The nearest entry in Conn’s Biological Stains (9th ed,, 1977)
is amidonaphthol red 5B (C.I. 18055, Acid violet 7). Synonyms
include lissamine red 6B and many others. The Colour Index
number (or application name) is the most reliable identifier
of a dye. It should be mentioned in the published instructions
for a method. If it isn’t, your best bet is to find another,
properly explained staining technique for the job.
4. My assumption has been that the lissamine fast red referred
to is the same that Lendrum used in his published method for
muscle fibres. The dye name has the synonym Acid red 37,
Colour Index no. 17045. It appears in Floyd Green’s excellent
reference book “The Sigma Aldrich Handbook of Stains, Dyes and
Indicators” with the further synonyms anthranal red G and
fast light red B. The dye synonyms list I refer to most
frequently as an easy-to-use first stop was published as a
“give away” by Difco in 1974.
** Mayer’s and Gill’s hematoxylins
I would like to know the differences between two types of
hematoxylin: Mayer’s and Gill’s.
Haematoxylin dye concentration for Mayer is 1 gm/L compared
with 2 gm/L for Gill-I. The preservative for Mayer’s is chloral
hydrate and for Gill it is ethylene glycol. The acidifying agent
for Mayer’s is citric acid, whereas for Gill it is acetic acid.
Both have very good shelf lives of two years or more under
correct storage conditions. They both are used mainly as
progressive stains, and are well suited to use as counterstains
as well. Gill-I has some some strong adherents for progressive
It is possible to make either of these in a non-toxic formulation
without compromising performance or shelf life.
Both stains are hemalums: they are solutions containing hematein
(from oxidized hematoxylin), an aluminium salt (the “mordant,”
which forms dye-metal complexes with hematein), an organic acid
to adjust the pH, and a hydrophilic compound (glycerol, ethylene
glycol or chloral hydrate). The last ingredient is variously said
to modify the solubilities of other ingredients, retard the
oxidation of hematoxylin, “preserve” the solution or do nothing
at all. In most hemalums the hematein is generated by adding
enough of an oxidizing agent (most often the iodate ion) to
oxidize about half the hematoxylin. The unoxidized hematoxylin
provides a reservoir from which more hematein is slowly produced
by atmospheric oxidation. This compensates for the atmospheric
over-oxidation of hematein to trioxyhematein (which is useless),
thereby prolonging the life of the solution.
The compositions of Mayer’s and Gill’s hematoxylins are set out
below. Mayer’s recipe was published in 1863, that of Gill, Frost
and Miller in 1974. Gill’s hematoxylin closely resembles
“haematal-16,” a mixture published by J. R. Baker in 1962 that
contained ethylene glycol but no organic acid.
Hematoxylin 1 g Hematoxylin 2 g
Potassium alum 50 g (0.09M) Al sulfate 17.6 g (0.03M)
Sodium iodate 0.2 g Sodium iodate 0.2 g
Citric acid 1 g Acetic acid 40 ml
Chloral hydrate 50 g Ethylene glycol 250 ml
Water to make 1000 ml Water to make 1000 ml
Molar ratio of Al ions to haematein molecules in
the freshly made solution:
A high ratio of aluminium:dye slows down staining and increases
the selectivity for nuclei. Both these hemalums are used
progressively; in principle, Gill’s should stain more quickly
than Mayer’s. The effect of excess aluminium is seen most
strikingly with Ehrlich’s hematoxylin, which is saturated with
alum and relies on atmospheric oxidation (slow) to provide a low
concentration of hematein from an initially large (6 to 7 g/L)
reservoir of hematoxylin. Ehrlich’s hematoxylin is the slowest of
the progressive hemalum stains (up to 30 minutes, compared with 3
to 10 minutes for Mayer’s or Gill’s). Hemalums for regressive
nuclear staining (e.g. Delafield’s, Harris’s) have lower
aluminium:dye ratios than the progressive stains. Acid-alcohol
extracts the dye-metal complex more slowly from nuclei than from
other components of tissues.
Some references. These are for practical, rather than chemical
or theoretical (i.e. speculative) aspects of hemalum staining.
Baker, J.R. (1962). Experiments on the action of mordants. 2.
Aluminium-haematein. Quarterly Journal of Microscopical
Science 103: 493-517.
Bancroft, J.D. & Cook, H.C. (1984). Manual of Histological
Techniques. Edinburgh: Churchill-Livingstone.
Bancroft, J.D. & Stevens, A., eds. (1996). Theory and Practice of
Histological Techniques, 4th ed. London: Churchill-Livingstone.
Ehrlich, P. (1886). Die von mir herruhrende Hamatoxylinlosung.
Zeitschrift f�r wissenschaftliche Mikroskopie 3: 150.
Kiernan, J.A. (1999). Histological and Histochemical Methods:
Theory and Practice, 3rd ed. Oxford: Butterworth-Heinemann.
Llewellyn, Bryan. Stains File. http://www.netbistro.com/~bryand/
(This Web site has a splendid, possibly comprehensive,
collection of hematoxylin stain formulations.)
J. A. Kiernan
Department of Anatomy,
The University of Western Ontario,
LONDON, Canada N6A 5C1
** Effects of pH on staining by dyes
Many stains are acidified, but some are adjusted to a
neutral or even an alkaline pH. Why? Are different dyes
differently affected by pH changes?
For a full answer to your question you will need to
refer to a textbook of histological techniques. Here
is a simplified answer. It applies to basic (cationic)
and acid (anionic) dyes with fairly small molecules.
Attraction of opposite electric charges plays a major
part in staining by such dyes.
The structural macromolecules in a section of a tissue
have numerous side-chains that can form either positive
or negative ions.
Acid dyes (attracted to positive sites in tissue).
The positive ions are associated mainly with proteins.
The side chain of the amino acid arginine (a guanidino
group) is a strong base. That means it always carries a
positive charge, even at a high pH. It can therefore always
attract a negatively charged dye ion. At pH 9 or above, all
staining by a simple basic dye (biebrich scarlet is commonly
used) is due to arginine.
The other organic group that can form positive ions is the
amino group, which occurs at the N-terminus of every chain
of amino acids and on the end of the side-chain of lysine.
Amino groups are weak acids: at high pH they are not
ionized, but at low pH an amino group collects a hydrogen
ion (proton) from the solvent and becomes positively
charged. The amino group of lysine can collect a proton even
when there are not many around, as in a neutral or slightly
alkaline medium. Consequently, lysine behaves as a cation
and binds acid dyes at pH about 8 or below. N-terminal amino
groups are weaker acids: they cannot be protonated much
above pH 6, so they are not stained by neutral or alkaline
solutions of acid dyes. More and more amino groups become
protonated (ionized) as the pH is lowered. Staining with an
acid dye therefore occurs more rapidly and more strongly
from the more acid solutions. At a pH around 2, these dyes
The foregoing remarks apply to a “typical” acid dye with
sulfonic acid side-chains. Sulfonic acids are strong acids;
they exist in solution only as sulfonate anions. (Eosin is
not “typical” in this way because it is a salt of a weak
acid. Moreover eosin solutions must not be acidified too
much or insoluble unionized eosin will be precipitated,
leaving a colorless solution.)
Basic dyes (attracted to negative sites in tissue).
The three negatively charged chemical groups present
in a section are:
1. Sulfate (actually half-sulfate) of many carbohydrate
components (glycoproteins in some mucus, heparin in mast
cell granules, chondroitin sulfates in cartilage matrix,
etc.) These are strong acids: ionized even at low pH.
Sulphate groups therefore bind cationic (basic) dyes
at any pH. They are the only things stained at pH 1.
2. Phosphate groups, associated with DNA and RNA. These are
weak acids, so they become protonated (not ionized) if
the concentration of protons (hydrogen ions) is high
enough. Typically this occurs below about pH 2.5. The
phosphates of nucleic acids are fully ionized at pH 3.5
to 4. A basic dye at pH 3 to 4 stains nuclei, cytoplasm
that is rich in RNA and. of course, all the sites of
3. Carboxyl groups. These occur as parts of amino acids
(C-terminal and the side-chains of glutamic and aspartic
acid), sialic acids (mucus and other glycoproteins),
glycosaminoglycans of extracellular matrix carbohydrates
(hyaluronic acid, chondroitin sulfates etc) and free
fatty acids (frozen sections only). Carboxyl groups
ionize over quite a wide range of pH, from 5 up to
about 8. The higher the pH, the stronger and more rapid
the staining by a basic dye. At or above pH 8 it stains
Alkaline solutions of basic dyes are used for staining
semi-thin plastic sections. With anything thicker the color
is too dark to show structural details. For more selective
staining, basic dyes are applied as acidic solutions. At pH
1 only the sulfated materials are displayed. As the pH rises
from 2.5 to 4.5, nuclei and RNA stain with increasing speed
Remember that these simplified arguments do not apply to all
dyes, or even to those most commonly used in routine work.
Horobin, R.W. (1982). Histochemistry: An Explanatory Outline
of Histochemistry and Biophysical Staining. Stuttgart:
Kiernan, J.A. (1999). Histological and Histochemical
Methods: Theory and Practice, 3rd ed. Oxford:
Horobin, R.W. (1988). Understanding Histochemistry:
Selection, Evaluation and Design of Biological Stains.
Chichester: Ellis Horwood.
Lyon, H. (1991). Theory and Strategy in Histochemistry. A
Guide to the Selection and Understanding of Techniques.
John A. Kiernan
Department of Anatomy & Cell Biology
University of Western Ontario
** Histochemical stain for arsenic
Is there a staining method for showing the presence of
arsenic in tissues?
Fix in 10% formalin containing 2.5% copper sulfate for
5 days. Wash for 24 hours in running water. Process and
embed in parffin wax. Deparaffinized sections show green
granules of Scheele’s green (CuHAsO3) which, though
insoluble in water, is dissolved by acids and by ammonium
hydroxide. By substituting copper acetate for the sulfate,
the green granular paris green or cupric acetoarsenite is
produced. Its solubilities are similar (Castel’s method,
Bull. Histol. Appliq. 13: 106, 1936). A light safranine
counterstain gives good contrast.
Source: R. D. Lillie 1965. Histopathologic Technic and
Practical Histochemistry, 3rd ed. p. 445.
** Giemsa staining of blood smears: several hints
My methanol-fixed blood smears are not staining reliably
with Giemsa. Some advice is needed, please.
Fixation of well dried (at RT) PB smears can vary from 1-10
minutes; automated systems tend to use about 1-2 minutes and use
the methanol only once. For manual staining, most labs would fix
for about ten minutes. Precautions must be taken against
absorption of water from humid air. The methanol is usually
replaced twice daily, but more frequently at those times of
the year when humidity is high.
The first sign of unacceptable water content in the fixing
methanol will be the appearance of clear refractive spaces on
the biconvave surfaces of erythrocytes: perhaps only a few cells
per high-power field, but this will increase further as the
water content increases, and eventually the films will lose all
diagnostic value. Replacement of the methanol when you see more
than say 1-2/HPF might not be a bad idea. This artifact may also
be seen in some automated systems where the stain pack is not
turned over very quickly. Rather than replacing the stain pack,
economy of reagent can be maintained by manually fixing the
slides before thay go on the machine. This is particularly so
for the older Hematek grey models.
Caution. Longer fixation times are required for bone marrow
smears: 15-20 minutes, and always use fresh methanol for these.
Most persons using Giemsa prefer to stain the smear first with
May Grunwald or Jenner stain, either using it neat or diluting
1:2 with buffer. This pre-step improves the granule definition
and clarity, and also changes the traditional reddish purple of
nuclei with plain Giemsa to a blue purple as seen with Wright’s
The selection of Sorensen’s buffer will vary form 6.4-7.2, with
the lower pH being most popular with Wright’s rather than
Giemsa. The aim is to select a pH that produces a colour balance
that readily allows the user to differentiate between
normochromic and polychromic red cells and to distinguish toxic
granulation when present, this is usually pH 6.8. If looking for
malarial parasites, then a pH of 7.2 is preferable because it
allows better contrast to detect chromatin dots, trophozioites
Dilution of the Giemsa solution is best done immediately before
use and will vary from 1:8 to 1:12 depending upon your protocol.
As a general rule of thumb the higher dilutions require longer
staining times of about 20 minutes, and the less dilute stains
need between 6 and 12 minutes, depending upon tthe quality of
the Giemsa. It was frequently claimed that the longer times gave
better definition, but I must admit that I’ve seen short timed
smears that are every bit as good.
For many years good quality Giemsa would be stable after
dilution for 6 to 8 hours. For the last 2 or 3 yrs, however, the
best you can hope for is 3 to 4 hours. After dilution the
solution starts to deteriorate, with the appearance of floccules
and a subsequent loss of staining ability or strength. As the
time progresses you may need to compensate by increasing the
staining time, but after 3 hours you will need to replace it.
Recipes for Giemsa vary, whether it be that of Hayhoe or of
Dacie & Lewis, and measurements may be by weight or volume.
Stock solutions that have a 50% by volume content of glycerol
(Analar or USP) are the most stable. Under no circumstances ever
heat your glycerol to more than 45C, even though most texts say
56C. Above these temperatures there is a risk of oxidation, even
in the stock solution, I use 45C as a cut-off point to give me a
safety margin. Dye content will also vary from 0.45 to 0.8%.
Lillie’s comments should considered here. After standing for up
to 5 days, filtration to remove undissolved material is
Differentiation, by giving the slides two rinses in buffer of
two minutes each, is fairly standard, but you can overdo it. A
single rinse of three quick dips may in fact suffice. It will
depend upon your Giemsa solution and tastes. If overstaining is
a problem then consider adding methanol to your buffer rinse,
starting at 5% and adjusting according to results, followed by a
water rinse to remove solvent.
Mike Rentsch, “Histomail,” Downunder
** Automated H & E staining problems
We are having a problem with our H & E being inconsistent
(sometimes from day to day, sometimes from batch to batch). We
have an automated stainer and use bought solutions of
hematoxylin and eosin. We do not change program times or
reagents, yet sometimes our stain is light and sometimes it is
dark (preferred). We have not changed any processes, vendors, or
manufacturers, but our stain is continually changing.
The same hematoxylin, eosin, alcohol, and xylene are on our
manual stain line. We stain those following the same times as
on the auto stainer and they come out perfect every time.
Is your manual stain set-up absolutely identical to your
automatic stainer set-up, in time values as well as reagent
set-up? If so, the times on the machine may be too short, as
You commented that when you stain from your manual set-up the
staining results are fine. I would recommend that you “manually”
stain using your automatic stainer set-up. If you are able to
acheive the desired results, then we can identify the mechanical
differences between human and machine staining. It would be
helpful to compare your stain programs (Manual procedure and
Analyzing the stain, is the nuclear stain OK but the
counterstain is too light? Is the nuclear stain too light but
the counterstain OK? Is the nuclear stain too light and the
counterstain too light? Are the stains consistant in their
lightness throughout the specimen and throughout all sections on
the slide? Do you notice an improvement in the stain after the
new reagents have become somewhat diluted?
One of the biggest differences between hand and machine staining
is how the surface tension of the reagent currently on the slide
is broken and then replaced by the next reagent. When we stain
by hand we exert much more and varied force than a machine does
when plunging the slides into the reagent. We also knock off
more reagent, so less of the reagent clings to the slide with
each move. A stainer (machine, not human) simply lowers the
slides slowly, in a single plane, into the reagent. Even the
agitation of the machine staining is in that single plane (up
and down) movement. When we stain by hand we cause the reagent
in the dish to bombard the slide from several angles and with
greater force that breaks the surface tension in less time than
it takes a machine can accomplish. Therefore longer exposure
times (of tissues to stain) may be required on a machine to
yield the same results as hand staining.
When programming the machines I find it necessary to watch the
hand staining carefully in order to make an accurate translation
of a “dip” to a time value that the machine could reproduce. A
“dip” in acid alcohol in manual staining may not be able to be
reproduced by a machine. I may be able to use 1% acid alcohol in
hand-staining but have to use 0.5% acid alcohol on the staining
machine with a 2-second timing value to get the same results.
Ten “dips” in a manual stain may require 30 seconds on a
machine. Ten “dips” in a manual alcohol step may require 1
minute on a machine for the same results.
One of the things we need to remember is that the machine will
move the slides exactly the same way for the programmed time. We
humans (consciously or unconsciously) adjust our handling of the
slides based on how the sections or even the reagents look.
Sakura Finetek USA, Inc.
Torrance, CA 90501
** Verhoeff’s stain for myelin and elastin
Can Verhoeff’s elastic tissue stain (iron hematoxylin with
iodine) be used to stain myelin sheaths?
H. Puchtler and F. S. Waldrop published “On the Mechanism of
Verhoeff’s Elastica Stain: A Convenient Stain for Myelin Sheath”
in Histochemistry 62:233-247 (1979).
They stated: “Verhoeff’s elastica stain is definitely not
specific for elastin and is inferior to orcein and
resorcin-fuchsin because of the required differentiation with
its inherent bias to produce patterns which conform to
expectations. However, Verhoeff’s elastica stain is far superior
to other metal-hematein technics for myelin sheaths. The
combined Verhoeff-picro-Sirius Red F3BA stain can be performed
in 30 min and does not require differentiation. It is therefore
suggested to reclassify Verhoeff’s elastica stain as a method
for myelin sheaths.”
** Acridine orange method for DNA and RNA
Can acridine orange be used to stain DNA and RNA in different
fluorescent colors in sections as well as in smears of cells?
In the late sixties, early seventies, I used to use the original
method (Bertalanffy F.D. A new method for cytological
diagnosis of pulmonary cancer. Ann. New York Acad. Sci. 84:
225-238) for screening cytology slides fixed in alcohol for
malignant cells, and I thought it worked quite well, as did my
pathologist at the time. The DNA of the nucleus fluoresces
brilliant green, and RNA in the cytoplasm of malignant cells is
brilliant orange. However, I have never met a cytotechnologist
who liked the method, so, when I was forced to hire one because
of work load, she quickly relegated this technique to the
garbage bin of history.
I don’t know of anyone who is currently using the technique.
However, as we found it very useful at the time, I worked out a
method for using it on paraffin sections, that gives very similar
results to the alcohol fixed smears.
1. Bring paraffin sections to water in usual manner.
2. Stain sections in acridine orange stain for 30 minutes.
3. Rinse sections briefly in 0.5% acetic acid in 100% alcohol.
4. Rinse sections in two additional changes of 100% alcohol.
5. Rinse sections in two changes of xylene.
6. Mount sections in a non-fluorescent resinous medium.
Results: DNA brilliant green. RNA brilliant orange. Most gram
positive microorganisms brilliant orange. Most gram negative
microorganisms (including helicobacter) green to pale orange.
Acridine orange stain
Acridine orange (C.I. 46005) 0.05 gm
Distilled water 500.0 ml
Acetic acid 5.0 ml
(Note, some batches of the acridine orange dye
work better than others.)
Kelowna General Hospital
Kelowna B.C. Canada
** Quickly finding something in a newly cut section
Is there any way to quickly stain paraffin sections so that I
can evaluate whether or not I need to cut further into the
We used to use a cotton ball moistened with dilute methylene
blue to wipe over the surface of the block. This gave us a good
idea of the tissue at that level and helped greatly in the
orientation. If you prefer you can always place a cut section
on a slide and add several drops of dilute aqueous methylene
blue (say 0.05-0.1%), this also works well. No need to mount the
If the structure is fairly large you can use a
pseudo-interference contrast illumination method to see
structure in the section. Just move the objective of the
microscope slightly to one side of its normal position and you
can see 3D structure without doing any deparaffinizing or
staining. You will be surprised how much detail you can make
out. This is a great method for finding glomeruli in kidney
Centers for Disease Control
I have used the following technique when searching for glomeruli
in kidney biopsies.
Mount the section on the slide as usual.
Place the slide on the microscope stage, under a 10x objective.
Close the condenser aperture down, and lower the entire condenser
away from the microscope stage.
What should result is a slightly out of focus image of the
unstained tissue section. You may have to adjust the settings
of the aperture and condenser. This works well for large
structures such as the glomerulus in the nephron of a kidney.
Patrick M. Haley
** Fluorescent lectins: general method
Can anybody give me a working concentration range for staining
with lectins conjugated with TRITC?
The general rule of thumb when staining with fluorescent protein
conjugates is to bracket around 10 micrograms per mL. When using
a good fluorescent IgG conjugate, I found that 5 micrograms/mL
was a bit dim, whereas 20 micrograms/mL often had a bit too much
background. This rule of thumb depends somewhat on the
fluorophore (some yield a higher background, etc), but for TRITC
conjugates, 10 micrograms/mL usually works well.
Although the molecular weight of your lectin is probably is a
bit less than IgG, a 2-3 fold difference in molecular weight
prabably won’t make that much of a difference. I used to use a
TRITC conjugate of wheat germ agglutinin at 10 micrograms per mL
and it stained beautifully.
Karen Larison, in Oregon
** Methyl blue and methylene blue
A method calls for methyl blue, in a mixture with eosin Y. The
nearest name I can find on a bottle is methylene blue. Will it
be OK to use it instead?
No! The only thing these two dyes have in common is a blue
color. Otherwise they have opposite staining properties.
Methyl blue, an acid triphenylmethane dye, is one of the
components of aniline blue. Aniline blue is a generic name that
includes methyl blue (C.I. 42780; Acid blue 93) and water blue
or ink blue (C.I. 42755; Acid blue 22). Most dyes that are sold
under these names are mixtures of both dyes, but some are mostly
methyl blue. A contaminant known as sirofluor is also present in
these dyes, and is exploited in fluorescent stains for callose
in plants. In staining applications any dyes sold as aniline
blue, methyl blue and water blue are interchangeable, provided
that the batch meets the Biological Stain Commission’s standards
in respect of content of reducible blue dye and performance in
standardized staining procedures.
Methyl blue (aniline blue) is used in Mann’s eosin-methyl blue
method and in various trichrome stains such as Mallory’s,
Gomori’s, Cason’s and Heidenhain’s AZAN. It colors collagen
fibers and a few other materials.
Methylene blue (C.I. 52015; Basic blue 9) is a basic thiazine
dye. It may have more scientific uses than any other dye. As a
simple stain, applied from a mildly acidic solution (pH 3 to 4)
it colors nucleic acids and acidic carbohydrates. At neutral or
alkaline pH is colors everything. Methylene blue is used in
conjunction with eosin and other dyes in stains for blood cells
and parasites, and it is also extensively used in bacteriology.
Products of degradation (demethylation or “polychroming”) of
methylene blue are essential components of the commonly used
Romanowsky-Giemsa stains for blood cells. The purple coloration
of leukocyte nuclei and magenta color of malaria parasites seen
with Wright’s and Giemsa’s stains, are due to one of these
products, the dye known as azure B (C.I. 52010).
Methylene blue (and some other thiazine dyes) can provide
beautiful and selective staining of the living neurons and their
cytoplasmic extensions, and has been much used to demonstrate
the innervation of peripheral tissues. Methyl (aniline) blue
cannot be used in this way.
Reference: Conn’s Biological Stains. Entries under the various