Full text of DOI:10.1078/0065-1281-00625

acta histochem. 104(1) 3–14 (2002)
© Urban & Fischer Verlag
http://www.urbanfischer.de/journals/actahist
acta
histochemica
The specificity of the histochemical NADPH diaphorase
reaction for nitric oxide synthase-1
in skeletal muscles is increased in the presence of urea*
Oliver Baum**, Alexander Miethke, Achim Wöckel, Gregor Willerding, and Gerit Planitzer
Department of Anatomy, Universitätsklinikum Benjamin Franklin, Freie Universität Berlin, Germany
Received 22 October 2001 and in revised form 4 December 2001; accepted 7 December 2001
Summary
Nitric oxide synthase-1 (NOS-1) can be demonstrated in the sarcolemma region of myofibers in
rodent skeletal muscles with the use of NADPH diaphorase histochemistry. Since other, especially
intrafibrar enzymes also exhibit NADPH diaphorase activity, we tried to increase the specificity of
the histochemical reaction for NOS-1. A qualitative and quantitative analysis was performed on cryo-
stat sections of fast-twitch oxidative myofiber-rich tongue and fast-twitch glycolytic myofibers-rich
tibialis anterior muscle derived from C57 mice and NOS-1 deficient knockout mice. All myofibers of
both C57 mice and NOS-1 knockout mice contained significant intrafibrar NADPH diaphorase activ-
ity which was inhibited to almost background levels when 2 M urea was added to the incubation
medium. On the other hand, myofibers of C57 mice but not of NOS-1-deficient knockout mice exhib-
ited NADPH diaphorase activity in their sarcolemma region which was only weakly reduced in the
presence of 2 M urea as was demonstrated by image analysis. Quantitative data on the activity of
NADPH diaphorase(s) were obtained in situ by photometric analysis of formazan extracted from
cryostat sections. The catalytic activity in tongue and tibialis anterior muscle was reduced in presence
of 2 M urea to approximately 27% in C57 mice and to 7–17% in NOS-1 knockout mice, respectively.
An in vitro NADPH diaphorase assay performed on homogenates of skeletal muscles also revealed an
inhibitory effect of 2 M urea in both mouse strains and, additionally, indicated an upregulation of
NADPH diaphorase activity in NOS-1 knockout mice. Finally, an immunodepletion analysis demon-
strated that NOS-1 comprises 38% of the total NADPH diaphorase activity in tongue and approxi-
mately 59% in tibialis anterior muscle in C57 mice. In conclusion, we recommend the addition of
2 M urea to the incubation medium to increase the specificity of the NADPH diaphorase reaction to
localise NOS-1 with the use of catalytic histochemistry.
Key words: nitric oxide synthase-1-NADPH diaphorase-skeletal muscle-image analysis-quantitative
catalytic histochemistry
Introduction
skeletal muscle is involved in the establishment of the
steady state in the energy metabolism at many levels
(reviewed in Reid, 1998; Stamler and Meissner, 2001).
NOS-1 is concentrated in the sarcolemma region of
Skeletal muscles of mice, rats, humans and many other
mammalian species are important sources of nitric
oxide synthase-1 (NOS-1) which produces the potent
signalling molecule NO. The NOS-1/NO-system in
*Dedicated to Professor Reinhart Gossrau on the occasion of his 61st birthday
**Correspondence to: Dr. O. Baum, Department of Anatomy, Universitätsklinikum Benjamin Franklin, Freie Universität Berlin, Königin-Luise-
Str. 15, D-14195 Berlin-Dahlem, Germany; tel: +49-30-8445 1947; fax: +49-30-8445 1943; e-mail: olibaum@zedat.fu-berlin.de
0065-1281/02/104/1-003 $ 15.00/0

4
Baum et al.
myofibers where it is particularly clustered in junctional ment and/or are applicable only on cell extracts or
parts such as costameres (Gossrau 1998; Baum et al., homogenates so that the micro-environment of NOS-1
2000), neuromuscular junctions (Kusner and Kaminski, in situ is lost.
1996; Grozdanovic and Gossrau, 1998) and myotendi-
The catalytic histochemical assay based on the
nous junctions (Chang et al., 1996). The targeting of the NADPH diaphorase reaction represents a simple
enzyme to the sarcolemma is primarily mediated by its method that was originally used to directly demonstrate
N-terminal PDZ domain which links NOS-1 to α1-syn- the presence of NOS with high sensitivity and reliabili-
trophin and, thus, integrates the enzyme into the dys- ty in cryostat sections of the brain (Dawson et al., 1991;
trophin glycoprotein complex (Brenman et al., 1996). Hope et al., 1991). Image analysis allows the rapid
Not all myofibers in skeletal muscles contain equal quantification of the catalytic activity (Planitzer et al.,
amounts of NOS-1. In rats, the highest concentrations 2001). All known NOS proteins/isoforms are able to
of NOS-1 are found in fast-twitch oxidative (FOG; type reduce NADPH-dependently water-soluble tetrazolium
IIa) myofibers, whereas fast-twitch glycolytic (FG; type salts such as nitroblue tetrazolium to water-insoluble
IIb) myofibers exhibit significantly lower levels of formazan with their common C-terminal located reduc-
NOS-1 (Planitzer et al., 2001). Slow-twitch oxidative tase domain and are, therefore, identified as NADPH
(SO; type I) myofibers contain the lowest amounts of diaphorases (Bredt et al., 1991). However, the histo-
NOS-1 (Grozdanovic et al., 1995; Kapur et al., 1997; chemical method is not specific for NOS-1 since sever-
Planitzer et al., 2001).
al other enzymes catalyse the NADPH diaphorase reac-
Qualitative and quantitative data on NOS-1 expression tion in many tissue including skeletal muscle (Stoward
in skeletal muscles have been obtained with the applica- et al., 1991; Tracey et al., 1993; Blottner et al., 1995;
tion of various methodological approaches. Immunoblot- Planitzer et al., 2000).
ting of tissue homogenates has been performed to deter-
Several attempts had been made to increase the
mine the NOS-1 concentration in different skeletal mus- specificity of the histochemical NADPH diaphorase
cles (Kapur et al., 1997; Hussain et al., 1997; Planitzer et reaction for NOS-1. This can be semi-quantitatively
al., 2001). Immunohistochemistry has demonstrated the achieved when the incubation is performed on fresh
presence of NOS-1 in the sarcolemma region (Kobzik et cryostat sections in the presence of formaldehyde
al., 1994; Grozdanovic et al., 1995) and possibly in the (Nakos and Gossrau, 1994), thiol inhibitors, aldehydes,
sarcosol of myofibers (Punkt et al., 2001).
alcohols, oxidising agents and NADPH analogues
Immunological methods such as immunoblotting and (reviewed in Gossrau et al., 1996) or urea and other
immunohistochemistry allow only the characterisation chaotropic agents (Gossrau, 1999). However, the effects
of the expression patterns of enzymes but do not yield of the inhibitory compounds on NOS-1 were not quanti-
information on their catalytic activity in situ. Various fied in these studies, what is required if the histochemi-
assays have been developed to monitor the biological cal NADPH diaphorase reaction is used for the analysis
activity of NOS (reviewed in Feelisch and Stamler, of catalytic activity.
1996). Approaches to directly quantify the NO-produc-
We have performed a thorough quantitative study to
tion in biological samples include methods based on the optimise the specificity of the histochemical NADPH
Gries reaction (Schmidt et al., 1990) or the oxyhe- diaphorase reaction for NOS-1. With the application of
moglobin assay (Murphy and Noack, 1994). Alterna- image analysis to NOS-1-deficient knockout mice and
tively, indirect methods such as ELISA-based or bioas- corresponding control animals, we demonstrate a strong
say-based techniques to determine cGMP-levels pro- inhibitory effect of urea on all non-NOS-1 NADPH
duced by soluble guanylate cyclase(s) in response to diaphorases as compared with only a weak influence on
NO activation have been successfully applied to mea- NOS-1-related NADPH diaphorase activity in NOS-1-
sure NOS-activity (Ishii et al., 1991). The most specific rich FOG and FG myofibers. Consequently, we recom-
and probably most popular method to determine NOS mend the use of 2 M urea in the incubation medium to
activity is the citrulline assay detecting the amount of increase the specificity of the histochemical NADPH
radiolabelled citrulline co-generated with NO from L- diaphorase reaction for NOS-1 in skeletal muscles.
arginine in homogenates of tissues or cultured cells with
sensitivity in the picomolar range (Bredt et al., 1991).
Material and methods
Antibodies
Recently, derivatives of 4,5-diaminofluorescein (DAF)
have been introduced to detect NO generation in intact
cultured cells or sections of neuronal tissues (Kojima et
al., 1998). However, these methods to quantify the
activity of NOS have specific disadvantages (reviewed
in Feelisch and Stamler, 1996). They are either only
semi-quantitative, indirect, require expensive equip-
All immunological methods that were applied to anal-
yse NOS-1 expression in skeletal muscles (immunohis-
tochemistry, immunoprecipitation and immunoblotting)
were performed with a peptide-specific polyclonal anti-
acta histochemica 104, 1 (2002)

The specificity of the histochemical NADPH diaphorase reaction 5
serum raised against the C-terminal amino acids tion of dissolved formazan was obtained by ultrasonica-
1409–1429 of rat NOS-1 (Sigma, Munich, Germany) tion for 30 min. The absorbance of this solution was
cross-reacting with mouse NOS-1.
measured in a standard laboratory photometer at 572
nm. Absorbance and mean protein content (determined
in serial sections) were used to calculate the relative
NADPH diaphorase activity expressed as ∆E/mg pro-
Animals and tissue preparation
NOS-1 knockout mice, which were originally obtained tein/30 min. Control experiments to obtain blank values
from Jackson Laboratories (Bar Harbor ME, USA), and for photometry were performed in the absence of β-
the corresponding C57/Bl6 control strain were bred NADPH.
under standardised conditions. Seven healthy male mice
of both strains (6 months old) were used for this study.
Immunohistochemistry
For preparation of tissue samples, mice were sacri-
ficed under deep ether anesthesia by excision of the The immunohistochemical detection of NOS-1 was per-
heart. The tongue and tibialis anterior (TA) muscles formed as described before (Planitzer et al., 2000,
were quickly removed. To prevent interference of 2001). Briefly, cryostat sections (thickness, 10 µm)
NADPH diaphorase-dependent formazan production in were incubated after formaldehyde fixation in a solution
epithelial cells with formazan production in myofibers, of 5% (w/v) bovine serum albumin to block nonspecific
epithelial layers were pealed off from the tongues as binding and were then incubated with the primary anti-
much as possible. For histochemistry, muscle samples NOS-1 antibody in a dilution of 1:1,000 overnight.
were mounted on cork plates with TissueTek, frozen in Immunopositive sites in skeletal muscle tissues were
liquid nitrogen-cooled methylbutane and stored in visualised with a secondary Cy3-conjugated antibody
closed plastic bags at –40 °C until use. For biochemical (Jackson ImmunoResearch, West Grove PA, USA).
analysis, the samples were transferred to reaction tubes, Control incubations were carried out by replacing the
frozen in liquid nitrogen and stored at –40 °C until use.
primary antibody with an IgG fraction of a pre-immune
serum in the corresponding concentration (Sigma) and
did not produce immunoreactivity.
Catalytic histochemistry
NADPH diaphorase activity was visualised in cryostat
sections as reported earlier (Planitzer et al., 2001) using
Image analysis
an incubation medium containing 1 mg/ml β-NADPH As equipment, an Axioskop microscope (Zeiss,
(Biomol, Hamburg, Germany), 0.25 mg/ml nitroblue Oberkochen, Germany) in combination with a F10
tetrazolium and 0.3% (v/v) Triton X-100 in 0.1 M Tris- CCD camera (Panasonic, Hamburg, Germany) was
HCl buffer, pH 7.4. Variation of the concentrations of used which was connected via a PCI digitizer card
the compounds in the incubation medium (NADPH, (Miro, Braunschweig, Germany) to a G3 Desktop 233
nitroblue tetrazolium) and the addition of polyvinyl computer (Apple, Cupertino CA, USA). The public
alcohol as diffusion protectant did not improve the dis- domain software Object Image 1.62p2 (written by N.
crimination between NOS-1 and non-NOS-1 NADPH Fischer, University of Amsterdam, Netherlands, avail-
diaphorase activity (G. Planitzer, thesis, Free University able from the web at http://simon.bio.uva.nl/object-
Berlin, in press). When indicated, this medium was sup- image.html) was applied. Further details of image anal-
plemented with urea. Cryostat sections were incubated ysis to determine initial reaction rates (Vinit) of NADPH
for 1 h in a moist chamber at 37 °C, rinsed several times diaphorase activity in cryostat sections have been
in phosphate buffered saline, pH 7.4 (PBS) and then described previously (Planitzer et al., 2001). Briefly,
cover-slipped in glycerol jelly. Controls performed in cryostat sections of tongues and TA muscle (thickness,
the absence of β-NADPH did not produce formazan.
10 µm) were incubated to demonstrate NADPH
To obtain quantitative data on NADPH diaphorase diaphorase activity at room temp on the scanning stage
activity in skeletal muscles in situ, formazan was of the microscope. A digital image was captured after
extracted from cryostat sections as described before every min during the first 10 min of the incubation, and
(Planitzer et al., 2001). Briefly, 40 cryostat sections of then after every 5 min up to a total of 40 min. Tongues
either tongue or TA muscles (thickness, 10 µm) were were used for the analysis of FOG myofibers, and TA
incubated to detect NADPH diaphorase activity in the muscles for the investigation of FG myofibers.
absence or presence of 2 M urea as described above.
For the determination of fiber types, serial sections of
Subsequently, the sections were scraped off the glass the skeletal muscles were incubated for myosin heavy
slides, transferred to reaction tubes and digested for chain immunohistochemistry and succinate dehydroge-
18 h with a solution of 1% (w/v) collagenase (Merck, nase (SDH) histochemistry. By semiquantitative evalu-
Darmstadt, Germany). An almost homogeneous solu- ation of SDH-dependent formazan production, individ-
acta histochemica 104, 1 (2002)

6
Baum et al.
ual myofibers were identified to be either SO, FOG or BCA method (Pierce, Rockford IL, USA). Combination
FG fiber types as described earlier (Planitzer et al., of both values allowed the calculation of the relative
2001).
NADPH diaphorase activity in the solubilisates
For analysis, 5 FOG myofibers were selected in each expressed as ∆E/mg protein/30 min.
section and their area marked using the computer
mouse. For the determination of NADPH diaphorase
activity in the sarcolemma, a macro was written in the
Immunoprecipitation and immunoblot analysis
built-in macro language to select the area precisely. The For immunoprecipitation of NOS-1, 10 µg of the poly-
time-dependent formazan generation in defined pixels clonal anti-NOS-1 antibody were coupled to 15 mg pro-
was expressed in grey levels that was recorded and con- tein A-sepharose 4B (Sigma, Munich, Germany) in
verted into optical density values (absorbance) with the buffer A (0.5 M Tris-HCl, pH 7.4 containing 0.5% (v/v)
use of a grey level standard (Zeiss). The data obtained Triton X-100, 1 mM EDTA, 1 mM PMSF) for 1 h at
were described as a quadratic function using a least room temp. Subsequently, the isolated antibody-
square fitting method according to Jonker et al. (1995). sepharose-conjugate was added to 0.1 ml of solubilised
The first derivative of this function at t0 = 0 represented tissue (tongue or TA muscle) derived from C57 mice for
Vinit and was taken as a measure of NADPH diaphorase 16 h at 4 °C. The antigen/antibody-sepharose complex-
activity.
es were then pelleted in a standard laboratory centrifuge
(1 min at 1.000 g) to obtain NOS-1-depleted extracts in
the supernatant. Aliquots of the original solubilised tis-
sues, the NOS-1-depleted extracts and the antigen/anti-
In vitro assay of NADPH diaphorase activity
Tissue samples of tongue and TA muscle were body-sepharose pellet were incubated to determine the
homogenised with a micro-douncer (Braun, Melsungen, NADPH diaphorase activity in the in vitro assay. Other
Germany) in a 10-fold volume of solubilisation buffer aliquots were analysed by immunoblotting of NOS-1
(0.1 M Tris-HCl, pH 7.4 containing 0.5% (v/v) Triton which was performed as described before (Planitzer et
X-100, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluo- al., 2001) using an ECL detection kit (Amersham,
ride (PMSF) and a standard protease inhibitor mixture). Braunschweig, Germany).
Homogenates were then centrifuged in a standard labo-
ratory centrifuge (1 min at 1.000 g) to obtain a solubil-
isate fraction in the supernatant in which NOS-1 is Results
found (Planitzer et al., 2001). An aliquot of the super-
natant (500 µl) was mixed and incubated with 500 µl of The histochemical analysis of NADPH diaphorase
a 2-fold concentrated NADPH diaphorase medium (20 activity in the absence of urea in cryostat sections of the
mg/ml β-NADPH, 0.5 mg/ml nitroblue tetrazolium and tongue (used as a muscle which consists exclusively of
0.5% (v/v) Triton X-100 in 0.1 M Tris-HCl buffer, pH FOG myofibers) from C57 mice showed strong intra-
7.4) at 37 °C for 30 min. This medium was supplement- cellular and sarcolemma-associated formazan produc-
ed with 4 M urea when required (resulting in a 2 M final tion in all myofibers (Fig. 1A). In cross sections, intra-
concentration). Absorbance of the formazan produced cellular formazan-positive sites appeared as dots that
was measured at 575 nm immediately after the reaction were significantly larger than formazan granules con-
had been stopped on ice. Blanks for photometry were nected by fine lines. These structures were identified as
obtained from incubations in the absence of NADPH. mitochondria and endoplasmic reticulum on the basis of
An additional aliquot of the solubilisate was used for immunohistochemistry with antibodies against defined
determination of the protein concentration using the markers (data not shown; described in Planitzer et al.,
Fig. 1. Histochemical localisation of NADPH diaphorase activity in cryostat sections of tongue derived from C57 mice and NOS-1 knockout
mice in the absence or presence of urea. Serial cryostat sections of tongue from either C57 mice (A–D) or NOS-1-deficient knockout mice
(E–H) were incubated to demonstrate NADPH diaphorase activity in the absence (A, E) or presence of 2 M (B, F) or 4 M (C, G) urea, respec-
tively. In C57 mice (A), the diaphorase-dependent formazan production in the absence of urea is present in the sarcolemma (arrowheads) and
inside the myofibers as dots and fine lines representing mitochondria and the endoplasmic reticulum. In NOS-1 knockout mice (E), mitochon-
dria and endoplasmic reticulum but not the sarcolemma showed formazan precipitation in the absence of urea. Addition of 2 M urea to the
incubation medium inhibited almost completely the non-sarcolemma diaphorase activity in both mouse strains (B, F). In the presence of 4 M
urea (C, G), NADPH diaphorase-produced formazan was not detectable. Immunohistochemistry with a polyclonal antibody demonstrated
NOS-1 expression in the sarcolemma region (arrowheads) of myofibers of C57 mice (D) but not of NOS-1 knockout mice (H). The numbers
1–3 were added to allow the identification of defined myofibers in serial sections. C and G were photographed with double refraction. Bar in
D, 20 µm.
acta histochemica 104, 1 (2002)

The specificity of the histochemical NADPH diaphorase reaction 7
acta histochemica 104, 1 (2002)

8
Baum et al.
2000). Outside myofibers, NADPH diaphorase activity their low levels of formazan production in C57 mice
was detected in neurons only. Other cell types in tongue (Fig. 2A) and NOS-1 knockout mice (Fig. 2E).
(satellite cells, fibroblasts, pericytes, endothelial cells)
did not express any significant NADPH diaphorase the intrafibrar NADPH diaphorase activity was almost
activity.
completely inhibited in TA muscles of both mouse
When 2 M urea was added to the incubation medium,
The addition of 2 M urea to the NADPH diaphorase strains. Accordingly, only very low levels of formazan
incubation medium resulted in an almost exclusive sar- were present in NOS-1 knockout mice (Fig. 2F), where-
colemma-associated staining since only low levels of as a strong sarcolemma-associated staining was found
formazan were observed in the myofibers (Fig. 1B). in C57 mice (Fig. 2B). The presence of 4 M urea in the
Higher concentrations of urea such as 4 M completely incubation medium abolished formazan production in
abolished the generation of formazan in the tongue (Fig. all myofibers of both mouse strains (Fig. 2C,G). Specif-
1C). Immunohistochemistry revealed that the sarcolem- ic NOS-1 immunoreactivity was restricted to the sar-
ma-associated NADPH diaphorase-positive sites cor- colemma in C57 mice and was absent in NOS-1 knock-
responded with those that were positive for NOS-1 out mice (Fig. 2D,H).
(Fig. 1D).
Although catalytic histochemistry demonstrated that
When tongue from NOS-1 knockout mice was incu- the addition of at least 2 M urea to the incubation medi-
bated to demonstrate NADPH diaphorase activity in the um inhibited almost completely the non-sarcolemma
absence of urea, only mitochondria and endoplasmic NADPH diaphorase activity in both FOG and FG
reticulum were formazan-positive in myofibers whereas myofibers, it remained unclear whether the catalytic
the sarcolemma was formazan-negative (Fig. 1E). The activity was also affected in the sarcolemma region. To
presence of 2 M or 4 M urea in the incubation medium address this question, we applied image analysis to
abolished formazan production in all myofibers (Fig. monitor differences in the amount of NADPH
1F-G). As expected, NOS-1 immunohistochemistry diaphorase-dependent generation of formazan (ex-
showed that NOS-1 was not present (Fig. 1H).
pressed as absorbance) in the sarcolemma region of
The histochemical analysis was performed on cryo- individual myofibers in either the absence or presence
stat sections of TA muscle as well (Fig. 2). This type of of 2 M urea. Fig. 3 shows the nonlinear reaction as
skeletal muscle consists of approximately 50% FG exemplified for tongue of C57 mice. Therefore, Vinit
myofibers, which can be identified in serial sections by were taken as parameter for the velocity of the reaction
the combination of myosin heavy chain immunohisto- (Table 1). In C57 mice, the Vinit of NADPH diaphorase
chemistry and succinate dehydrogenase histochemistry activity was higher in the absence than in the presence
(data not shown; described in Planitzer et al., 2001). In of 2 M urea in the sarcolemma region (in FOG
the absence of urea, strong NADPH diaphorase activity myofibers, 31% inhibition and in FG myofibers, 26%
was present in the sarcolemma region of TA muscle inhibition was found). In the non-sarcolemma region,
from C57 mice. Additionally, mitochondria and endo- the inhibition rates were 81% and 85%, respectively.
plasmic reticulum in the myofibers showed formazan These results demonstrate a strong inhibitory effect of
production. However, FG myofibers contain less mito- urea on all non-sarcolemma-associated NADPH
chondria as compared with FOG and SO fibers and, diaphorases (to almost background levels) as compared
consequently, contained only low intracellular NADPH with a remarkably lower effect on the sarcolemma-asso-
diaphorase activity. Therefore, a variable pattern of ciated NADPH diaphorase activity.
intrafibrar formazan production was observed: FOG
Quantitative data on the activity of NADPH
and SO myofibers showed strong NADPH diaphorase diaphorase(s) in skeletal muscles of both C57 mice and
activity whereas FG myofibers were characterised by NOS-1 knockout mice in situ were obtained by extrac-
Fig. 2. Histochemical localisation of NADPH diaphorase activity in cryostat sections of tibialis anterior (TA) muscle from C57 mice and NOS-
1 knockout mice in the absence or presence of urea. Serial cryostat sections of TA muscle from either C57 (A–D) or NOS-1-deficient knock-
out mice (E–H) were incubated to demonstrate NADPH diaphorase activity in the absence (A, E) or presence of 2 M (B, F) or 4 M (C, G) urea,
respectively. Immunohistochemistry with a polyclonal antibody showed NOS-1 expression in the sarcolemma region of myofibers (arrow-
heads) of C57 mice (D) but not NOS-1 knockout mice (H). The numbers 1–3 were added to allow identification of defined myofibers in serial
sections: numbers 1 and 3 represent FOG myofibers and number 2 a FG myofiber in A–D, while number 1 is a FOG myofiber, number 2 a FG
myofiber and number 3 a SO myofiber in E-H. Note the lower amount of sarcolemma-associated formazan in myofibers of C57 mice with the
lowest non-sarcolemma NADPH diaphorase activity (e.g. number 2) which are FG myofibers.The asterisk in E–H marks a vessel that was non-
specifically positive for NOS-1 (H) with the secondary antibody as demonstrated in control experiments (data not shown).
C and G were photographed with double refraction. Bar in D, 20 µm.
acta histochemica 104, 1 (2002)

The specificity of the histochemical NADPH diaphorase reaction 9
acta histochemica 104, 1 (2002)

10 Baum et al.
tion of formazan from cryostat sections and subsequent NADPH diaphorase activity that was half of the activity
photometric measurement (Fig. 4). Formazan produc- in tongue.
tion was specific for the catalytic activity in myofibers
An in vitro assay was established to measure
since other cell types in skeletal muscles contain only NADPH diaphorase activity in homogenates of skeletal
negligible NADPH diaphorase activity as was demon- muscles. The values obtained in this assay (Fig. 5)
strated by histochemistry. Amounts of formazan could be directly compared with those determined in the
extracted from sections of tongue derived from C57 in situ assay since both were performed under identical
mice and NOS-1 knockout mice were reduced to 27% experimental conditions and were, in general, several-
and 7%, respectively, in the presence of 2 M urea. fold higher. Again, the presence of urea significantly
Residual NADPH diaphorase activity in TA muscle was decreased NADPH diaphorase activity, in both muscles:
27% for C57 mice and 17% for NOS-1 knockout mice tongue from C57 mice and NOS-1 knockout mice
in the presence of 2 M urea. When NADPH diaphorase exhibited 65% and 57% residual catalytic activity,
activity in NOS-1 knockouts was compared with that of respectively, whereas in TA muscles of C57 mice and
C57 mice, tongue expressed 63% and TA muscles 45%, NOS-1 knockout mice 31% and 6% rest activity,
respectively. In both mouse strains, TA muscles showed respectively, was found. In the in vitro assay, NADPH
Table 1. Determination of initial velocities (Vinit) of NADPH diaphorase activity in FOG and FG myofibers of C57 mice and NOS-1-deficient
knockout mice. Cryostat sections of tongue and tibialis anterior muscle were analysed histochemically for NADPH diaphorase activity either
in the absence or presence of 2 M urea. Formazan production in the sections was recorded using image analysis exclusively in FOG myofibers
in the tongue and FG myofibers in tibialis anterior muscle as identified in serial sections on the basis of succinate dehydrogenase activity and
myosin-heavy chain fiber-typing (data not shown). Absorbance values are plotted against incubation time. Standard deviations of the Vinit val-
ues were less than 10% in all cases and omitted for reasons of clarity.
Mouse strain
Fiber type (muscle)
Vinit without urea (∆E/min)
Vinit with 2 M urea (∆E/min)
Sarcolemma region Non-sarcolemma
region
Sarcolemma region Non-sarcolemma
region
C57/Bl6
FOG (tongue)
FG (tibialis anterior)
3.75 · 10–2
2.72 · 10–2
1.76 · 10–2
0.67 · 10–2
2.59 · 10–2
2.01 · 10–2
0.33 · 10–2
0.31 · 10–2
NOS-1–/–-knockout
FOG (tongue)
FG (tibialis anterior)
<0.1 · 10–2
<0.1 · 10–2
1.91 · 10–2
0.71 · 10–2
<0.1 · 10–2
<0.1 · 10–2
<0.1 · 10–2
<0.1 · 10–2
Fig. 4. Determination of NADPH diaphorase activity in cryostat sec-
tions of tongue and tibialis anterior (TA) muscle in situ. Fourty cryo-
stat sections of tongue and TA muscle from either C57 mice or NOS-
1-deficient knockout mice were incubated to detect NADPH
diaphorase activity histochemically in the absence or presence of 2
M urea. Subsequently, the sections were transferred to reaction
tubes and treated with collagenase to dissolve the formazan pro-
duced in the tissue sections. Photometric determination of both for-
mazan and protein concentrations allowed the calculation of for-
mazan produced per mg protein per unit time. Each bar is the mean
value ± SD of 3 measurements.
Fig. 3. The presence of 2 M urea in the incubation medium reduces
the NADPH diaphorase activity in the sarcolemma region and the
non-sarcolemma compartment of tongue myofibers of C57 mice.
Serial cryostat sections of tongue from C57 mice were incubated to
detect NADPH diaphorase activity histochemically in the absence or
presence of 2 M urea. Generation of formazan was recorded with
the use of image analysis in a number of selected myofibers and,
subsequently, photometrically quantified in both sarcolemma and
non-sarcolemma regions in time as indicated.The data are the mean
values of 20–30 myofibers analysed.
acta histochemica 104, 1 (2002)

The specificity of the histochemical NADPH diaphorase reaction 11
diaphorase activities as measured in homogenates of
tongue and TA muscle from NOS-1 knockout mice were
higher than in those from C57 mice.
Discussion
Although catalytic profiles of the different NADPH
diaphorases have not been characterised, various
reports suggest that the histochemical NADPH
diaphorase reaction is not specific for a defined enzyme
(Stoward et al., 1991; Tracey et al., 1993; Blottner et
al., 1995). However, characterisation of their catalytic
profiles is important since it finally may lead to the
development of (a) method(s) that is (are) specific for
single NADPH diaphorases and allow quantification
studies. We have performed a qualitative and quantita-
tive characterisation of the histochemical NADPH
diaphorase reaction in skeletal muscle derived from
C57 mice and NOS-1 knockout mice. We have deter-
mined Vinit’s of the NADPH diaphorase(s) in the sar-
colemma and the intrafibrar compartment applying
image analysis and quantified specific NADPH
diaphorase activities using an in situ assay. In summa-
ry, our results reveal a strong inhibitory effect of 2 M
urea on intrafibrar NADPH diaphorase activity as com-
pared with a significant weaker inhibition of the activi-
ty in the sarcolemma region. This compartment-selec-
tive inhibitory effect of urea was, in principle, observed
in all myofibers investigated albeit FOG myofibers
contained significantly more non-sarcolemma (intrafi-
brar) NADPH diaphorase activity than FG fibers in the
absence of urea.
The proportion of NOS-1 in the pool of NADPH
diaphorases in tongue and TA muscle was determined in
an immunodepletion assay (Fig. 6). After immunopre-
cipitation with anti-NOS-1 antibodies, detergent
extracts of tongue derived from C57 mice contained
approximately 62% NADPH diaphorase activity as
compared with complete homogenate. The immunopre-
cipitation fraction comprised 38% of the NADPH
diaphorase activity. Depletion of NOS-1 from TA mus-
cle homogenates reduced NADPH diaphorase activity
to approximately 41% whereas NOS-1 comprised
approximately 59%.
The in vitro assay was applied as an alternative
methodological approach to quantify NADPH dia-
phorase activity in tissues. The values determined in
this assay were, in general, higher than those mea-
sured with the in situ assay. This discrepancy could be
the result of the improved availability of compounds
of the incubation medium to the solubilised enzymes
or, alternatively, may be caused by the destruction of
Fig. 5. Determination of NADPH diaphorase activity of tongue and
tibialis anterior (TA) muscle in vitro. Detergent extracts of tongue
and TA from either C57 mice or NOS-1 knockout mice were incubat-
ed to detect NADPH diaphorase activity in vitro either in the absence
or presence of 2 M urea. Photometric determination of both for-
mazan and protein concentrations allowed calculation of formazan
produced per mg protein per unit time. Each bar is the mean value ±
SD of 6 measurements.
Fig. 6. Immunodepletion analysis to determine the proportion of NOS-1 in NADPH diaphorase activity in homogenates of tongue and tibialis
anterior (TA) muscle. Detergent extracts of tongue and TA muscle of C57 mice were subjected to immunoprecipitation with an anti-NOS-1
antibody. Subsequently, aliquots of the original solubilisates (S), the NOS-1-depleted extracts (D) and the NOS-1-containing immunoprecipi-
tates (IP) were incubated in vitro to determine NADPH diaphorase activity (A) or analysed by immunoblotting to control the effectiveness of
immunoprecipitation as shown for one tongue preparation (B). Bars in A represent mean values ± SD of 3 immunoblotting measurements.
acta histochemica 104, 1 (2002)

12 Baum et al.
the microenvironment of the enzymes in the intact tis- C57 mice all NADPH diaphorase activity present in
sue. In both assays, FOG myofiber-rich tongue con- this site is related to NOS-1. Since, in addition, the
tained approximately twice as much NADPH intrafibrar Vinit values were almost identical in skele-
diaphorase activity than FG myofiber-rich TA muscle tal muscles of C57 and NOS-1 knockout mice, one
and, again, the addition of 2 M urea inhibited signifi- may speculate that comparable amounts of formazan
cantly the generation of formazan in all samples are produced in the non-sarcolemma compartment in
investigated. However, the inhibition rates of the cat- both mouse strains. Consequently, the higher catalytic
alytic activity in vitro were significantly lower in activities as determined histochemically with the in
tongue than in TA muscle (of both mouse strains) as situ assay in skeletal muscles of C57 mice should
compared with the respective values determined in the reflect NOS-1-specific NADPH diaphorase activity.
in situ assay. We suggest that the mitochondria (which Calculations revealed that the proportion of NOS-1 in
are rich in NADPH diaphorase activity; Lind et al., total NADPH diaphorase activity is approximately
1990; Olausson et al., 1995) of FOG myofibers are 37% in tongue and 52% in TA muscle. These values
less sensitive to the treatment with urea under the in correspond very well with those obtained in the
vitro assay conditions (where they are solubilised) immunodepletion assay of muscle of C57 mice show-
than under in situ assay conditions. The most striking ing that in tongue 38% and in TA muscle 59% of the
difference between both assays applied was the higher total NADPH diaphorase activity is produced by
NADPH diaphorase activity in vitro in both tongue NOS-1.
and TA muscle homogenates of NOS-1 knockout mice
The fact, that NOS-1-specific NADPH diaphorase
which contained even more activity than the activity was always restricted to the sarcolemma allows
homogenates of C57 mice. This observation can only the conclusion that previously reported intrafibrar
be explained by an upregulation of the activity of the NADPH diaphorase activity (Kobzik et al., 1995;
non-NOS-1 NADPH diaphorase(s) in NOS-1 knock- Frandsen et al., 1996) and NOS-1 immunoreactivity
out mice. The molecular basis for this upregulation is (Frandsen et al., 1996; Punkt et al., 2001) were either
not known yet.
not related to NOS-1, were caused by methodological
To our knowledge, there is only one report that inves- variations or reflect species-specific differences in
tigated the effect of urea on the reductase activity of NOS-1 expression.
NADPH diaphorase(s) so far (Narayanasami et al.,
Preliminary investigations (data not shown) indicate
1997). Urea (in concentrations up to 4 M) and guanidine that the addition of urea to the NADPH diaphorase
hydrochloride were found to mimic the binding of incubation medium also inhibits the intrafibrar staining
calmodulin to NOS-1 in an in vitro assay. As a conse- in skeletal muscles of rats, guinea pigs, gerbils and
quence, the NADPH-dependent cytochrome c reductase hamsters. The urea concentration needed for complete
activity of NOS-1 (corresponding to NADPH suppression of intrafibrar activity of NADPH dia-
diaphorase activity) was temporarily increased. These phorase(s) seemed to be dependent on the species inves-
results are apparently in contradiction with our results tigated. Therefore, we conclude that urea as supplement
showing an inhibition of Vinit’s and activity of all to the incubation medium increases the specificity of
NADPH diaphorases by urea in catalytic histochem- the NADPH diaphorase reaction associated with NOS-1
istry. Interestingly, when potassium thiocyanate, which in skeletal muscle not only of mice (as shown here) but
is another chaotropic agent, is applied in concentrations also of rodents in general. In contrast, human quadri-
higher than 200 mM, it causes an irreversible loss of the ceps muscles did not contain intrafibrar NADPH
FAD cofactor and denaturation of DT diaphorase diaphorase activity as demonstrated with catalytic histo-
(Byron et al., 1997). Therefore, it seems reasonable that chemistry (even in absence of urea). Thus, further
the presence of high concentrations of chaotropic agents research should be performed to clarify the molecular
such as urea may either activate or inhibit individual basis of the different species-specific profiles of the
NADPH diaphorases leading to selective staining pat- NADPH diaphorase activity in presence of urea (or
terns in catalytic histochemistry. With respect to the his- other chaotropic reagents) as applied in catalytic histo-
tochemical NADPH diaphorase reaction in cryostat sec- chemistry.
tions of skeletal muscle, it could mean that the cell sur-
face of myofibers provides a microenvironment protect-
ing NADPH diaphorases of the sarcolemma more effi-
Acknowledgement
ciently from denaturation than that/those present inside The skilful technical support of Martina Gutsmann and
the myofibers.
Heidrun Richter is gratefully acknowledged. We would
In NOS-1 knockout mice, NADPH diaphorase- like to thank Prof. Reinhart Gossrau for teaching us the
dependent generation of formazan in the sarcolemma first lessons in catalytic histochemistry and giving us
region was completely abolished suggesting that in the freedom to set our own footsteps in science.
acta histochemica 104, 1 (2002)

The specificity of the histochemical NADPH diaphorase reaction 13
Hussain SN, El-Dwairi Q, Abdul-Hussain MN, and Sakkal D
(1997) Expression of nitric oxide synthase isoforms in nor-
mal ventilatory and limb muscles. J Appl Physiol 83:
348–353
Ishii K, Sheng H, Warner TD, Förstermann U, and Murad F
(1991) A simple and sensitive bioassay method for detec-
tion of EDRF with RFL-6 rat lung fibroblasts. Am J Physi-
ol 261: H598–603
Jonker A, Geerts WJ, Charles R, Lamers WH, and Van Noor-
den CJF (1995) Image analysis and image processing as
tools to measure initial rates of enzyme reactions in sec-
tions: distribution patterns of glutamate dehydrogenase
activity in rat liver lobules. J Histochem Cytochem 43:
1027–1034
Kapur S, Bedard S, Marcotte B, Cote CH, and Marette A
(1997) Expression of nitric oxide synthase in skeletal mus-
cle: a novel role for nitric oxide as a modulator of insulin
action. Diabetes 46: 1691–1700
Kobzik L, Reid MB, Bredt DS, and Stamler JS (1994) Nitric
oxide in skeletal muscle. Nature 372: 546–548
Kobzik L, Stringer B, Balligand JL, Reid MB, and Stamler JS
(1995) Endothelial type nitric oxide synthase in skeletal
muscle fibers: mitochondrial relationships. Biochem Bio-
phys Res Commun 211: 375–381
Kojima H, Nakatsubo N, Kikuchi K, Urano Y, Higuchi T,
Tanaka J, Kudo Y, and Nagano T (1998) Direct evidence of
NO production in rat hippocampus and cortex using a new
fluorescent indicator: DAF-2 DA. Neuroreport 9:
3345–3348
References
Baum O, Planitzer G, Richter H, and Gossrau R (2000) Irreg-
ular costameres represent nitric oxide synthase-1-positive
sarcolemma invaginations enriched in contracted skeletal
muscle fibres. Histochem J 32: 743–751
Blottner D, Grozdanovic Z, and Gossrau R (1995) Histo-
chemistry of nitric oxide synthase in the nervous system.
Histochem J 27: 785–811
Bredt DS, Hwang PM, Glatt CE, Lowenstein C, Reed RR,
and Snyder SH (1991) Cloned and expressed nitric oxide
synthase structurally resembles cytochrome P-450 reduc-
tase. Nature 351: 714–718
Brenman JE, Chao DS, Gee SH, McGee AW, Craven SE, San-
tillano DR, Wu Z, Huang F, Xia H, Peters MF, Froehner SC,
and Bredt DS (1996) Interaction of nitric oxide synthase
with the postsynaptic density protein PSD-95 and alpha1-
syntrophin mediated by PDZ domains. Cell 84: 757–767
Byron O, Mistry P, Suter D, and Skelly J (1997) DT
diaphorase exists as a dimer-tetramer equilibrium in solu-
tion. Eur Biophys J 25: 423–430
Chang WJ, Iannaccone ST, Lau KS, Masters BS, McCabe TJ,
McMillan K, Padre RC, Spencer MJ, Tidball JG, and Stull
JT (1996) Neuronal nitric oxide synthase and dystrophin-
deficient muscular dystrophy. Proc Natl Acad Sci USA 93:
9142–9147
Dawson TM, Bredt DS, Fotuhi M, Hwang PM, and Snyder
SH (1991) Nitric oxide synthase and neuronal NADPH
diaphorase are identical in brain and peripheral tissues.
Proc Natl Acad Sci USA 88: 7797–7801
Kusner LL, and Kaminski HJ (1996) Nitric oxide synthase is
concentrated at the skeletal muscle endplate. Brain Res
730: 238–242
Feelisch M, and Stamler, JS (1996) Methods in nitric oxide
research. Wiley, Chichester
Lind C, Cadenas E, Hochstein P, and Ernster L (1990) DT-
diaphorase: purification, properties, and function. Methods
Enzymol 186: 287–301
Frandsen U, Lopez-Figueroa M, and Hellsten Y (1996) Local-
ization of nitric oxide synthase in human skeletal muscle.
Biochem Biophys Res Commun 227: 88–93
Murphy ME, and Noack E (1994) Nitric oxide assay using
hemoglobin method. Methods Enzymol 233: 240–250
Nakos G, and Gossrau R (1994) When NADPH diaphorase
(NADPHd) works in the presence of formaldehyde, the
enzyme appears to visualize selectively cells with constitu-
tive nitric oxide synthase (NOS). Acta Histochem 96:
335–343
Narayanasami R, Nishimura JS, McMillan K, Roman LJ,
Shea TM, Robida AM, Horowitz PM, and Masters BS
(1997) The influence of chaotropic reagents on neuronal
nitric oxide synthase and its flavoprotein module. Urea
and guanidine hydrochloride stimulate NADPH-
cytochrome c reductase activity of both proteins. Nitric
Oxide 1: 39–49
Gossrau R (1998) Nitric oxide synthase I (NOS I) is a
costameric enzyme in rat skeletal muscle. Acta Histochem
100: 451–462
Gossrau R (1999) Chaotropic reagents for discrimination
between nitric oxide synthase I (NOS I) and non-NOS I
diaphorases in skeletal muscle fibers. 5. NO Forum der
deutschsprachigen Länder. Erlangen, 23–24 April, Abstract
D2
Gossrau R, Nakos, G, Christova, C, and Grozdanovic Z
(1996) Selective visualization of the NADPH diaphorase
activity of nitric oxide synthase in mammalian striated
muscle fibres. In: Moncada S, Stamler J, Gross S, and
Higgs EA (Eds) The biology of nitric oxide, Part 5. Port-
land Press, London, pp 304–307
Olausson T, Fjellstrom O, Meuller J, and Rydstrom J (1995)
Molecular biology of nicotinamide nucleotide transhydro-
genase-a unique proton pump. Biochim Biophys Acta
1231: 1–19
Planitzer G, Baum O, and Gossrau R (2000) Skeletal muscle
fibres show NADPH diaphorase activity associated with
mitochondria, the sarcoplasmic reticulum and the NOS-1-
containing sarcolemma. Histochem J 32: 303–312
Grozdanovic Z, and Gossrau R (1998) Co-localization of
nitric oxide synthase I (NOS I) and NMDA receptor sub-
unit 1 (NMDAR-1) at the neuromuscular junction in rat
and mouse skeletal muscle. Cell Tissue Res 291: 57–63
Grozdanovic Z, Nakos G, Dahrmann G, Mayer B, and Gossrau
R (1995) Species-independent expression of nitric oxide
synthase in the sarcolemma region of visceral and somatic
striated muscle fibers. Cell Tissue Res 281: 493–499
Hope BT, Michael GJ, Knigge KM, and Vincent SR (1991) Planitzer G, Miethke A, and Baum O (2001) Nitric oxide syn-
Neuronal NADPH diaphorase is a nitric oxide synthase.
thase-1 is enriched in fast-oxidative myofibers. Cell Tissue
Proc Natl Acad Sci USA 88: 2811–2814
Res 306: 325–333
acta histochemica 104, 1 (2002)

14 Baum et al.
Punkt K, Zaitsev S, Park JK, Wellner M, and Buchwalow IB Stamler JS, and Meissner G (2001) Physiology of nitric oxide
(2001) Nitric oxide synthase isoforms I, III and protein
kinase-Ctheta in skeletal muscle fibres of normal and strep-
tozotocin-induced diabetic rats with and without Ginkgo
biloba extract treatment. J Histochem 33: 213–219
Reid MB (1998) Role of nitric oxide in skeletal muscle: syn-
thesis, distribution and functional importance. Acta Physiol
Scand 162: 401–409
Schmidt HHHW, Zernikow B, Baeblich S, and Böhme E
(1990) Basal and stimulated formation and release of L-
arginine-derived nitrogen oxides from cultured endothelial
cells. J Pharmacol Exp Ther 254: 591–597
in skeletal muscle. Physiol Rev 81: 209–237
Stoward PJ (1991) Dehydrogenases. In: Stoward PJ and
Pearse AGE (Eds), Histochemistry: theoretical and applied.
Churchill Livingstone, Edinburgh, pp 27–71
Tracey WR, Nakane M, Pollock JS, and Förstermann U
(1993) Nitric oxide synthases in neuronal cells,
macrophages and endothelium are NADPH diaphorases,
but represent only a fraction of total cellular NADPH
diaphorase activity. Biochem Biophys Res Commun 195:
1035–1040
acta histochemica 104, 1 (2002)