Light and Electron Microscopic Demonstration of the p75Nerve Growth Factor Receptor in Normal Human CutaneousNerve Fibers: New Vistas

Yong Liang and Olle JohanssonExperimental Dermatology Unit, Department of Neuroscience, Karolinska Institute, Stockholm, Sweden

The nerve growth factor and its receptor are importantin nerve growth, differentiation, maturation, and main-tenance. In order to explore the exact distribution of p75low-affinity nerve growth factor receptor (p75 NGFr)expression in cutaneous nerve fibers, p75 NGFr andneuron-specific enolase double immunofluorescence andimmunoelectron microscopic studies were conducted onnormal human skin samples. After p75 NGFr and neuron-specific enolase immunofluorescence double staining, thedermal nerves were strongly p75 NGFr-immunoreactive(IR); however, very few p75 NGFr-IR nerve fibers werefound in the epidermis. p75 NGFr immunoreactivity wasfound mainly in the peripheral part of cutaneous nervetrunks and fibers, whereas neuron-specific enolaseimmunoreactivity was mainly seen within the axons.After ultrastructural immunostaining, the Schwann cell

In human skin, there is a rich distribution of nerve fibers andfree nerve endings. Most of them occupy the upper dermis,but many of them also extend into the epidermis (cf. Wanget al, 1990; Hilliges et al, 1995). The cutaneous nerve fibers arecomposed of the axons and their terminal branches, which are

surrounded by the cytoplasmic sheath of Schwann cells. Most of themare believed to be sensory nerve fibers. Recently, neurotrophins, suchas nerve growth factor (NGF) and its receptor (NGFr), have beendemonstrated to be of great importance in the normal developmentof the peripheral nervous system (Levi-Montalcini and Angeletti, 1968).They are both essential for the survival of certain nerves, especially thesensory ones (Lee et al, 1992). Their effects include normal neuriteoutgrowth, neuroneal differentiation, survival of large numbers ofneurons, and supporting the rescue of injured and diseased nerves(Gage et al, 1988; Heumann, 1994). It has been reported that thedensity of some nerves is dependent upon the local concentration ofNGF (Korsching and Thoenen, 1983). The NGFr complex is composedof two kinds of transmembrane receptor proteins: those of the Trkfamily (the high-affinity receptors) and the p75 pan-neurotrophinreceptor (the low-affinity type). Both of them can combine withNGF and the two-receptor system can potentially lead to greater

Manuscript received September 22, 1997; revised February 28, 1998; acceptedfor publication March 10, 1998.

Reprint requests to: Dr. Olle Johansson, Experimental Dermatology Unit,Department of Neuroscience, Karolinska Institute, 171 77 Stockholm, Sweden.

Abbreviations: IR, immunoreactive; NGFr, NGF receptor; NSE, neuron-specific enolase.

0022-202X/98/$10.50 · Copyright © 1998 by The Society for Investigative Dermatology, Inc.

114

membrane was strongly p75 NGFr-IR. The Schwann cellmembrane facing the connective tissue was more stronglyp75 NGFr-IR than the part of the membrane close tothe axon; the Schwann cell cytoplasm nearest to themembrane sometimes also showed a high p75 NGFrimmunoreactivity, whereas the rest of the cytoplasm wasgenerally more weakly p75 NGFr-IR; however, the axonitself seldom showed any such immunoreactivity; actu-ally, only a few parts of the axonal membrane revealed aweak staining, leaving most of the membrane unstained.The axoplasm was not p75 NGFr-IR. The results – thatin human cutaneous nerves it is mainly the Schwann cellsthat express p75 NGFr immunoreactivity – further stressthe active role of the glial ensheathment in the controland maintenance of a normal target innervation. Keywords: immunoelectron microscopy/immunofluorescence/neuron-specific enolase/skin. J Invest Dermatol 111:114–118, 1998

responsiveness to NGF (Chao and Hempstead, 1995). The p75 nervegrowth factor receptor (p75 NGFr) has intensively been investigatedin the central and peripheral nervous system (mostly in animal tissue)by many laboratories. In human skin tissue, at the light microscopiclevel, p75 NGFr immunoreactivity has been localized in cutaneousnerves at all levels, as well as in certain basal keratinocytes (Fantini andJohansson, 1992) and in the Meissner and Pacinian corpuscles of thehuman digital skin (Vega et al, 1993). At the ultrastructural level, p75NGFr immunoreactivity has been studied in the lower lip of rats andin primary and permanent canine tooth pulps of cats (Fried andRisling, 1991; Ribeiro-da-Silva et al, 1991); however, the ultrastructuraldistribution of p75 NGFr in human cutaneous nerve fibers is, atpresent, unknown. To explore the exact distribution of p75 NGFrexpression in normal human cutaneous nerve fibers, immunofluores-cence and immunoelectron microscopic studies were performed onnormal human skin samples. For comparative reasons, neuron-specificenolase (NSE) was also employed in the part of the study utilizingimmunofluorescence.

MATERIALS AND METHODS

Immunofluorescence The resulting tissue from biopsies of normal healthyvolunteers (n 5 7) performed under local anesthesia with 1% Xylocaine withoutepinephrine were investigated using immunofluorescence. The specimens wereimmediately immersed for 3 h at 4°C in a 0.1 M phosphate buffer (pH 7.4)containing 4% (wt/vol) paraformaldehyde and 14% (vol/vol) of a saturatedpicric acid solution for fixation and then rinsed in the same buffer containing10% sucrose for at least 24 h. Microm (Heidelberg, Germany) cryosectionsof 14 µm, thawed onto gelatin-coated slides, were processed for indirectimmunofluorescence staining. The primary antibodies [mouse anti-human p75

VOL. 111, NO. 1 JULY 1998 P75 NGFR IN HUMAN CUTANEOUS NERVES 115

Figure 1. p75 NGFr and NSE double-labeling of cutaneous nerve fibers. Immunofluorescence p75 NGFr (A, C, E) and NSE (B, D, F) double-staining ofthe same slides using different excitation lights, respectively. (A) p75 NGFr immunoreactivity in the peripheral part of a nerve fiber; (B) NSE immunoreactivity inthe central part of the same nerve fiber; (C) p75 NGFr-IR and (D) NSE-IR in the same cross-sectioned nerve trunk. In part (E), the arrow indicates a very weakp75 NGFr-IR free nerve ending in the epidermis, whereas in part (F) the same free nerve ending (arrow) shows a strong NSE immunoreactivity. Scale bar, 50 µm.

NGFr (1:200, Amersham, Little Chalfont, U.K.) and rabbit anti-human NSE(1:1200, UltraClone, Isle of Wight, U.K.)] were diluted in a 0.01 M phosphatebuffer containing 0.3% Triton X-100. The sections were incubated with theseantibodies overnight at 4°C in a humid atmosphere followed by incubation for30 min at 37°C with fluorescein-isothiocyanate-conjugated donkey anti-mouseand tetramethylrhodamine-isothiocyanate-conjugated donkey anti-rabbit IgGanti-sera (both 1:160, Jackson, West Grove, PA), which were diluted in thesame buffer as the primary antibodies. All rinses before and after incubationswere performed with the same buffer solution described above. The mountingmedium contained para-phenylenediamine to prevent fading of the fluorescence(Johnson and Nogueira Araujo, 1981). The respective sections were observedand photographed in a Nikon Microphot-FXA fluorescence microscope usingdifferent excitation lights. For quality control of the experiment, in order toavoid nonspecific reactions of the first and secondary antibodies, normal serumfrom the same species as the first antibody, diluted at 1:50, or a phosphate-buffered saline were used on two slides instead of the first antibody, respectively;to avoid a cross-reaction of the secondary antibody in the double-staining, twoslides incubated with mouse p75 NGFr antibody were reacted with thetetramethylrhodamine-isothiocyanate-conjugated donkey anti-rabbit IgG andtwo slides incubated with rabbit NSE antibody were incubated with thefluorescein-isothiocyanate-conjugated donkey anti-mouse IgG. One per centnormal donkey serum was also added to both the first and the secondary anti-bodies.

Immunoelectron microscopy For the immunoelectron microscopy, corres-ponding biopsies from healthy volunteers were immediately immersed in afixative containing 4% (wt/vol) paraformaldehyde, 0.2% (vol/vol) saturatedpicric acid, and 0.1% (vol/vol) glutaraldehyde in a 0.1 M phosphate buffer

(pH 7.4) for 3 h at 4°C, followed by a rinse in the same buffer. Vibratomesections of 100 µm were cut and subjected to a freeze-thaw treatment. Pre-embedding immunostaining was performed according to the ABC method(DAKO ABC kit, DAKO, Glostrup, Denmark). Briefly, the sections, afterimmersion in 0.6% hydrogen peroxide for quenching endogenous peroxidaseactivity, were incubated in the mouse anti-p75 NGFr antibody (1:200) overnightat 4°C followed by immersion in a biotinylated goat anti-mouse IgG (1:100)and an avidin-biotin-peroxidase complex (1:100), both for 60 min at roomtemperature. The sections were then incubated in a medium containing 0.06%3,39-diaminobenzidine and 0.03% hydrogen peroxide in 0.01 M phosphate-buffered saline (pH 7.4). Post-fixation was carried out in 1% osmium tetroxidefor 60 min followed by dehydration in gradient ethanol. The tissues, afterimmersion in propylene oxide and a mixture of 50% propylene oxide and 50%Epon 812, were then embedded in Epon 812, and were afterwards polymerized.Ultrathin sections were cut on an LKB III Ultrotome and no counter-stainingwas carried out. The ultrathin sections were observed and photographed in aJEOL JEM-1200 electron microscope. As a means of quality control, sectionswere incubated as outlined above, but without inclusion of the primary antibody.

RESULTS

Different immunofluorescence demonstration of p75 NGFrand NSE in cutaneous nerve fibers After p75 NGFr and NSEimmunofluorescence double-staining, the dermal nerves were stronglyp75 NGFr-immunoreactive (IR), as well as NSE-IR. The p75 NGFrimmunoreactivity was found mainly in the periphery of cutaneous nervefibers; however, in the central part, the p75 NGFr immunoreactivityappeared weak, whereas the NSE immunoreactivity was mainly seen

116 LIANG AND JOHANSSON THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 2. Ultrastructural demonstration of p75 NGFr in normal cutaneous nerve fiber components and basal keratinocyte membrane. Ultrastructuralimmunocytochemistry of p75 NGFr mainly in Schwann cell membranes and the cytoplasm next to it (A–E). In parts (A)–(C), arrows indicate axons, a few parts ofp75 NGFr immunoreactivity in the axonal membrane are seen as well. Part (D) shows that the perineurium is weakly stained (arrowheads). Part (E) demonstrates astrong p75 NGFr immunoreactivity in the Schwann cell membrane and in patches of the cytoplasm close to the membrane. Part (F) shows a p75 NGFr-IR basalkeratinocyte membrane (arrowheads). Scale bars, 500 nm.

within the central part of the axons (Fig 1A, B). In the nerve trunks,this feature was more clearly demonstrated (Fig 1C, D). Very few p75NGFr-IR nerve fibers were found in the epidermis and those foundwere weak; however, NSE-IR free nerve endings clearly extendedinto the epidermis (Fig 1E, F). The double-staining revealed anobviously different staining pattern for the two markers in the samenerve profile.

Ultrastructural demonstration of p75 NGFr immunoreactivity innormal cutaneous nerve fiber components At the ultrastructurallevel, many p75 NGFr-IR unmyelinated cutaneous nerve fibers wereobserved in the dermis. In these nerve fibers, the Schwann cellmembrane was strongly p75 NGFr-IR (Fig 2A–E), with the Schwanncell membrane facing the connective tissue being more strongly p75NGFr-IR than the part of the membrane close to the axon. Certainpatches of the Schwann cell cytoplasm nearest to the membrane alsoshowed a high p75 NGFr immunoreactivity. Furthermore, a generalstaining of the whole cytoplasm was demonstrated, sometimes withdifferent organelles exhibiting a somewhat higher immunoreactivityon their outside. The axon itself seldom showed any labeling. Only afew parts of the axon membrane revealed a weak stain, leaving mostof the membrane unstained. The axoplasm itself was not p75 NGFr-IR (Fig 2B, C). Occasionally, the perineuroneal components showed

a weaker p75 NGFr immunoreactivity (Fig 2D). The basal keratinocytemembrane was also p75 NGFr positive (Fig 2F).

Control experiments No specific staining was seen in the controlexperiments (Fig 3).

DISCUSSION

It has previously been reported that the cutaneous nerve fibers and nervetrunks are characterized by an intensive p75 NGFr immunoreactivity(Fantini and Johansson, 1992). In our study, using p75 NGFr and NSEdouble-staining, we noticed that in some small nerve fibers, andespecially in the nerve trunks, p75 NGFr immunoreactivity occursmainly in the peripheral part, whereas NSE immunoreactivity occursin the central part. Because the perineuroneal connective tissue, theSchwann cell, and the axon are closely connected to one fiber profile,it is difficult to exactly discern the different components of a nervefiber at the light microscopic level. NSE is a specific neuroneal marker;thus, it demonstrates axons well (Schmechel et al, 1978). p75 NGFrand NSE immunofluorescence double-staining, together with p75NGFr immunoelectron microscopy, make it easier to see the exactdistribution of p75 NGFr in the cutaneous nerve fibers. Thus, the resultsof this study have clearly demonstrated p75 NGFr immunoreactivity in

VOL. 111, NO. 1 JULY 1998 P75 NGFR IN HUMAN CUTANEOUS NERVES 117

Figure 3. Negative results obtained by p75 NGFr immunofluorescenceand ultrastructural immunocytochemical control experiments. Sectionsprocessed for p75 NGFr immunofluorescence (A) and ultrastructuralimmunocytochemistry (B), but without the primary antibody. No specificimmunoreactive structures are visualized. Note that without the positiveimmunoreaction and parallel counter-staining, the contrast is weak in theelectron microscopic material. The thick arrow indicates a Schwann cell and thethin arrow shows an axon. Scale bars: (A) 50 µm; (B) 500 nm.

the Schwann cell membrane and in some parts of the axonal membraneof the cutaneous nerve fibers in the human adult.

These results show that in human cutaneous nerve fibers p75 NGFrimmunoreactivity mainly occurs in Schwann cells, with the mostprominent p75 expression apparently localized to the Schwann cellmembrane adjacent to the connective tissue. It has previously beenshown that such an immunoreactivity is strongly present at earlydevelopmental stages, but subsequently, with the start of myelination,a downregulation of the p75 NGFr expression takes place, until, inthe adult peripheral nerves, only the perineurium and a few endoneurialcells remain positive (Scarpini et al, 1988); however, our results seemnot to support the view that p75 NGFr immunoreactivity is normallyonly present in perineural cells in adults, but rather supports thepresence of p75 NGFr immunoreactivity in the axons and especiallyin the surrounding Schwann cells. The fact that the most prominentp75 NGFr expression apparently is localized to the Schwann cellmembrane adjacent to the connective tissue, would suggest an activerole of the target tissue, through Schwann cells, in the control andmaintenance of a normal sensory innervation. In addition, it has beenrevealed that, during development, the expression of NGF receptorsin cutaneous sensory fibers does not begin until the cutaneous targetfields are reached by these fibers (Davies et al, 1987). This fact, togetherwith the demonstration that NGF synthesis in the developing skin isactivated only after a sensory innervation has been established, furthersupports an active role for the target surface epithelium in the controland maintenance of normal sensory innervation.

At the ultrastructural level our results have demonstrated that p75

NGFr expression is stronger in Schwann cells than in axons, which issimilar to the results of Wakabayashi et al (1995), who demonstratedthe p75 NGFr mainly in Schwann cells in the human urinary bladder.The function of the p75 NGFr is currently not fully understood. It isrelated to both neuroneal cell growth [e.g., by increasing the rate ofNGF binding to the high-affinity NGFr (Mahadeo et al, 1994), bymodulating Trk tyrosine kinase activity (Verdi et al, 1994), and throughretrograde transport of neurotrophins (Johnson et al, 1987)] and cellapoptosis (Rabizadeh et al, 1993). Many biologic functions appear tobe realized only through the p75 NGFr in Schwann cells. For instance,in the absence of TrkA, the NGF, through binding to p75 NGFr, mayactivate the transcription factor ‘‘nuclear factor kappa B’’ in the Schwanncell (Carter et al, 1996), which is a potent gene expression regulator(Baeuerle and Henkel, 1994). And the p75 NGFr, binding with NGF,promotes Schwann cell migration (Anton et al, 1994), which mightbe critical in Schwann cell-guided axonal growth, branching, andregeneration. Finally, the labeling of certain organelles may indicate aninternalization process; however, this speculation needs to be clarifiedin future studies.

It is well accepted that the target tissue may play a maintenance rolefor the skin sensory nervous system under normal circumstances. Ithas also been found that the keratinocytes can produce the NGF aswell as its receptor (Davies et al, 1987; DiMarco et al, 1991; Fantiniand Johansson, 1992), both of which are thought to support cutaneousinnervation and maintain normal sensory functions. Also, an over-expression of epidermal NT-3 may increase the fiber number as wellas enhance innervation (Albers et al, 1996).

Our results, therefore, suggest that target maintenance and regulationmechanisms actually may be realized, at least partly, through Schwanncells. The Schwann cells may play a role in the regulation of axonalbranching, axonal growth, sensory function maintenance, and injuredaxon repair.

This work was supported by the Swedish Work Environment Fund (proj. no. 94–0375, 96–0841, and 97–1056), the Cancer and Allergy Foundation, Vårdalstiftelsen,Svenska Industritjanstemannaforbundet (SIF), Sveriges Civilingenjorsforbunds Miljofond,and the Medical Faculty of the Karolinska Institute. We thank Ms. Gunilla Holmkvistand Ms. Eva-Karin Johansson, respectively, for skilful technical and secretarial assistance.

REFERENCES

Albers KM, Perrone TN, Goodness TP, Jones ME, Green MA, Davis BM: Cutaneousoverexpression of NT-3 increases sensory and sympathetic neuron number andenhances touch dome and hair follicle innervation. J Cell Biol 134:487–497, 1996

Anton ES, Weskamp G, Reichardt LF, Matthew WD: Nerve growth factor and its low-affinity receptor promote Schwann cell migration. Proc Natl Acad Sci USA 91:2795–2799, 1994

Baeuerle PA, Henkel T: Function and activation of NF-kappa B in the immune system.Ann Rev Immunol 12:141–179, 1994

Carter BD, Kaltschmidt C, Kaltschmidt B, Offenhauser N, Bohm-Matthaei R, BaeuerlePA, Barde Y-A: Selective activation of NF-κ B by nerve growth factor through theneurotrophin receptor p75. Science 272:542–545, 1996

Chao MV, Hempstead BL: p75 and Trk: a two-receptor system. TINS 18:321–326, 1995Davies AM, Bandtlow C, Heumann R, Korsching S, Rohrer H, Thoenen H: Timing and

site of nerve growth factor synthesis in developing skin in relation to innervationand expression of the receptor. Nature 326:353–358, 1987

DiMarco E, Marchisio PC, Bondanza S, Franzi AT, Cancedda R, De Luca M: Growth-regulated synthesis and secretion of biologically active nerve growth factor by humankeratinocytes. J Biol Chem 266:21718–21722, 1991

Fantini F, Johansson O: Expression of growth-associated protein 43 and nerve growthfactor receptor in human skin: A comparative immunohistochemical investigation.J Invest Dermatol 99:734–742, 1992

Fried K, Risling M: Nerve growth factor receptor-like immunoreactivity in primary andpermanent canine tooth pulps of the cat. Cell Tiss Res 264:321–328, 1991

Gage FH, Armstrong DM, Williams LR, Varon S: Morphological response of axotomizedseptal neurons to nerve growth factor. J Comp Neurol 269:147–155, 1988

Heumann R: Neurotrophin signalling. Curr Opin Neurobiol 4:668–679, 1994Hilliges M, Wang L, Johansson O: Ultrastructural evidence for nerve fibers within all vital

layers of the human epidermis. J Invest Dermatol 104:134–137, 1995Johnson GD, Nogueira Araujo GM, C: A simple method for reducing the fading of

immunofluorescence during microscopy. J Immunol Meth 43:349–350, 1981Johnson EM Jr, Taniuchi M, Clark HB, Springer JE, Koh S, Tayrien MW, Loy R:

Demonstration of the retrograde transport of nerve growth factor receptor in theperipheral and central nervous system. J Neurosci 7:923–929, 1987

Korsching S, Thoenen H: Nerve growth factor in sympathetic ganglia and corresponding

118 LIANG AND JOHANSSON THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

target organs of the rat: correlation with density of sympathetic innervation. ProcNatl Acad Sci USA 80:3513–3516, 1983

Lee KF, Li E, Huber LJ, Landis SC, Sharpe AH, Chao MV, Jeanisch R: Targeted mutationof the gene encoding the low affinity NGF receptor p75 leads to deficits in theperipheral sensory nervous system. Cell 69:737–749, 1992

Levi-Montalcini R, Angeletti PU: Nerve growth factor. Physiol Rev 48:534–569, 1968Mahadeo D, Kaplan L, Chao MV, Hempstead BL: High affinity nerve growth factor

binding displays a faster rate of association than p140trk binding. Implications formulti-subunit polypeptide receptors. J Biol Chem 269:6884–6891, 1994

Rabizadeh S, Oh J, Zhong L-T, Yang J, Bitler CM, Butcher LL, Bredesen DE: Inductionof apoptosis by the low-affinity NGF receptor. Science 261:345–348, 1993

Ribeiro-da-Silva A, Kenigsberg RL, Cuello AC: Light and electron microscopic distributionof nerve growth factor receptor-like immunoreactivity in the skin of the rat lowerlip. Neuroscience 43:631–646, 1991

Scarpini E, Ross AH, Rosen JL, Brown MJ, Rostami A, Koprowski H, Lisak RP: Expression

of nerve growth factor receptor during human peripheral nerve development. DevelBiol 125:301–310, 1988

Schmechel D, Marangos PJ, Zis AP, Brightman M, Goodwin FK: Brain endolases asspecific markers of neuronal and glial cells. Science 199:313–315, 1978

Vega JA, Del Valle ME, Haro JJ, Calzada B, Suarez-Garnacho S, Malinovsky L: Nervegrowth factor receptor immunoreactivity in Meissner and Pacinian corpuscles of thehuman digital skin. Anat Rec 236:730–736, 1993

Verdi JM, Birren SJ, Ibanez CF, et al: p75LNGFR regulates Trk signal transduction andNGF-induced neuronal differentiation in MAH cells. Neuron 12:733–745, 1994

Wakabayashi Y, Tomoyoshi T, Tooyama I, Kitahama K, Kim SU, Maeda T: Low-affinitynerve growth factor receptor immunoreactivity in the human urinary bladder.Neurosci Lett 186:9–12, 1995

Wang L, Hilliges M, Jernberg T, Wiegleb-Edstrom D, Johansson O: Protein gene product9.5-immunoreactive nerve fibres and cells in human skin. Cell Tiss Res 261:25–33, 1990