Immunohistological study of neuronai markers in inflamed gingiva obtained from children with Down's syndrome

Monica Barr-Agholme\ Thomas Modeer^ and Johan Luthman^* ^Department of Pedodontics, Karolinska Institutet, Huddinge, Sweden; ^Department of Histology and Neurobiology, Karolinska Institutet, Stockholm, Sweden

Barr-Agholme M, Modeer T and Luthman J: Immunohislological study of neuronai markers in inflamed gingiva obtained from children with Down's syndrome. J Clin Periodontol 1991; 18: 624-633. Abstract. The histological appearance of the gingiva in children with Down's syndrome (DS) was studied with special reference to infiammatory involvement and innervation. A dense infiltration of infiammatory cells was seen in the propria of most of the DS patients, including a few polymorphonuclear leucocytes. A hyperplasia of the epithehum was also found. The innervation of the gingiva was studied using immunohistochemistry. Nerve fibers as well as nerve bundles immunoreactive to neurofilament (NF) were seen in the propria, while occasionally intraepithehal NF fibers were observed. Calcitonin gene-related peptide (CGRP)-immunoreactive fibers and fiber bundles were also visuahzed, but they were less abundant than NF fibers. The density of NF and CGRP fibers and fiber bundles was estimated by semiquantitative evaluation. A higher density of NF and CGRP immunoreactive structures was observed in the propria of DS patients compared to the control subjects, while no obvious alteration was seen in their distribution in the propria. In addition, sparsely distributed fibers immunoreactive to peptide histidine isoleucine amide (PHI) and vasoactive intestinal polypeptide (VIP) fibers as well as neuropetide Y (NPY) and tyrosine hydroxylase (TH) were seen, mainly surrounding blood vessels. A few substance P (SP) fibers were also found, mostly close to the epithelium. No obvious differences of these sparsely distributed fibers were seen in the DS patients compared to controls. Thus, a profound infiammatory involvement of the gingiva of DS patients is seen concomitant with a hyperinnervation of the presumed sensory component of the gingival innervation. In contrast, no alterations were seen in the density of neuronai markers related to autonomic nerve fibers. The sensory hyperinnervation observed is probably not specifically related to DS, but may be due to a sprouting of afferent nerves induced by the infiammatory reaction. However, factors released from the sensory afferents could contribute to the gingival infiammation seen in DS.

There is a high prevalence of chronic not fully understood. Abnormalities in infiammatory periodontal disease in host response (Breg 1977) and different children with Down's syndrome (DS) immunological functions may be of im(Cohen et al. 1961, Johnson & Young portance (Reuland-Bosma et al. 1986b, 1963, Brown 1978, Modeer et al. 1990). 1988a, b). Alterations in connective The mechanisms involved in the gin- tissue and vascularization in DS have gival infiammatory processes in DS are also been suggested to play a role in the pathogenesis of the gingival affections (Cohen et al. 1961, Claycomb et al. 1970), while changes in the morphology *Present address: Departments of Pharmaof the epithelium have also been decology and Psychiatry, University of Colorscribed (Cohen et al. 1961, Reuland-Bosado Health Sciences Center, Denver, USA

Key words: Down's syndrome; gingiva; innervation; inflammation; neurofilament; calcitonin gene-related peptide; neuropeptides. Accepted for publication 30 July 1990

ma et al. 1988a). It has furthermore been suggested that the periodontal affections seen in DS may involve neurodystrophic processes (Cohen et al. 1961). In recent years, the role of the peripheral innervation in the course of infiammatory processes has been highlighted. Thus, the release of presumed pro-infiammatory substances, such as the neuropeptide substance P (SP), from primary sensory afferents seems to be involved in infiammatory reactions.

Gingival innervation of Down's syndrome e.g., neurogenic infiammation (Lembeck & Holzer 1979), while the sympathetic nervous system might also contribute to physiologic changes associated with inflammation (Levine et al. 1985). It has furthermore been shown that in inflammation of more a chronic nature alterations in receptors for sensory neuropetides (Mantyh et al. 1989) and hyperinnervation of presumed sensory fibers containing calcitonin generelated peptide (CGRP) can occur (Kimberly et al. 1987, Kimberly & Byers 1988). The nervous supply of the human gingiva and periodontium has been described in a number of studies (Stewart & Lewinsky 1939, Gairns & Aitchinson 1950, Bernick 1957, Rapp et al. 1957, Dixon 1962, Von Lautenbach 1966, Griffin & Harris 1968, Desjardins et al. 1971, Luzardo-Baptista 1973, Van Steenberghe 1979). It has furthermore been suggested that the innervation and different biologically active peptides (neuropeptides) related to the peripheral nervous system could be involved in the pathogenesis of various pathological processes of the gingiva (Luthman et al. 1988b, 1989). Several neuropeptides have been shown to occur in the human gingiva in both neuronal and non-neuronal structures. In presumed sensory nerve fibers of the human gin-

giva SP and CGRP as well as neurofilament (NF) have been shown to occur (Luthman et al. 1988b). Nerve fibers immunoreactive to vasoactive intestinal polypeptide (VIP), peptide histidine isoleucine amide (PHI) and neuropeptide Y (NPY) have also been demonstrated in human gingiva (Luthman et al. 1988b). VIP and PHI appear to be confined to parasympathetic nerve fibers (see Lundberg & Hokfelt 1986), while NPY coexists with the catecholamine synthetizing enzyme TH, suggesting a relation to sympathetic fibers (Lundberg et al. 1982, Edwall et al. 1985). Thus, immunohistochemical visualization of these substances can be used to study the distribution and density of different components of the peripheral innervation of the human gingiva. The present study was undertaken in order to describe the inflammatory involvement as well as the innervation of the gingiva in children with DS. The distribution of nerve fibers as well as the occurence and distribution of different neuropeptides were studied using immunohistochemistry. The findings were compared to gingival material obtained from control subjects. The findings show a profound infiammatory reaction of the gingiva in DS concomitant with indications of sensory hyperinnervation.

625

Material and Methods Material

•

Interdental papillae from the incisorpremolar region were obtained under local anesthesia (2% Xylocain-Adrenaline, Astra Lakemedel AB) prior to periodontal treatment of DS children. The biopsies were taken from 13 patients (mean age 15 years, range 11-21) of either sex. The control biopsies were taken from the incisor-premolar region and consisted of excess tissue from surgical orthodontic treatment from 10 patients (mean age 12 years, range 7-15). Methods

Clinical examination The chnical status of the gingiva was examined before the biopsies were taken. Diagnosis of gingivitis was based on bleeding upon probing. Periodontitis diagnosis was based on roentgenological examination. A distance of more than 2 mm between the cemento-enamel junction to the alveolar crest in the site of the biopsy was regarded as presence of periodontitis (see Modeer et al. 1990). Histology and immunohistochemistry The gingival tissue was fixed by immersion for 90 min in a cold ( + 8°C) picric acid/formalin solution, rinsed for at

Table 1. Chnical and histoiogical data of material studied Age (years)

Subject

Sex

DS-l DS-2 DS-3 DS-4 ; DS-5 DS-6 DS-7 DS-8 DS-9 DS-IO DS-!1 DS-l 2 DS-l 3

M

M M M F F M

18

C-l C-2 C-3 C-4 C-5 C-6 C-7 C-8 C-9

F

15

M M F F

15 14 13 13

F

12 12 10

C-10

F F F F M M

M M M M

21 19 19 19 19 19 18 17 16 14 14

12

9 7

Region

Clinical status

Cellinfiltrate

PMNcells

Epitheliumhyperplasia

NF density

CGRP density

33-32 35-34 33-32 32-31 24-23 43-42 24-23 24-23 43-42 32-31 45-44 34-33 32-31

gingivitis/periodontitis gingivitis gingivitis gingivitis healthy healthy gingivitis healthy gingivitis gingivitis/periodontitis gingivitis healthy healthy

2 3 3 2

1 1 2 2 n,d.

2 1 0 2 4 3

2 1 0 1 2 ,0

1

1 n.d. 1 1 1.7 1 2 2 1 2 1

2

n.d.

2 1 1 1 I 1 1 1 1

n,d.

n,d.

I

1

1

0

0

14-13 24-23 34-33 44-43 14-13 15-14 13-12 55-54 55-54 36-75

healthy healthy healthy healthy healthy mild gingivitis healthy mild gingivitis healthy healthy

2 0

0 0

n,d. n,d.

n,d, n,d.

n,d, 0 n.d. n.d. 0 # 0 0 0 0

0

0

n,d. n,d. n,d.

n,d. n,d.

0 1

n,d 0 0

a

0 1 1 1,5 1 n.d. n.d. 1 0 1 1

2 0

n.d. 0.5 1 0.5 1 0.5

0,5 0.5 0 1 n.d. n.d. n.d.

1 . • • • I ' - '

t

1

•

•

0 1

DS denotes Down's syndrome patients and C denotes control subjects. 0 == nothing found/no alteration; l=low density/minor alteration; 2 = moderate density/moderate alteration; 3= high density/large alteration; 4 = very high density/extensive alteration, n.d. = not determined.

626

Barr-Agholme et al.

least 24 h in cold phosphate buffer containing 10% sucrose and bacitracine before sectioning. The gingival tissue was sectioned perpendicularly to the buccal epithelium at 14 //m on a cryostat (20-50 sections per sample). A number sections (3-4 sections/subject) were directly stained with hematoxylin-eosin for general histological evaluation. In most samples Z-A sections per antiserum studied could be obtained. However, some samples were small or partly damaged and consequently it was not possible to obtain a complete set of sections (requiring more than 0.5 mm undamaged tissue) in these biopsies. The infiltrate of infiammatory cells and the relative density of polymorphonuclear leucocytes (PMN cells) as well as the degree of epithelial hyperplasia (both buccal epithelium and gingival sulcus epithelium) were semiquantitatively es-

timated on a blind basis by a five-grade scale (0-4). The remaining sections were processed for indirect immunohistochemistry according to Coons (1958). The sections were incubated with antisera, diluted in 0.01 M phosphatebuffered saline (PBS) containing 0.3% Triton X100, overnight at 4°C in a humid atmosphere. The following primary polyclonal rabbit antisera were used: NF (1:1000), CGRP (1:400), SP (1:200), VIP (1:400), PHI (1:800), TH (1:400) and NPY (1:400). After incubation with primary antisera the sections were rinsed in PBS for 30 min with three changes, then incubated for 30 min at 37°C with tetramethylrhodamine-isothiocyanate isomer R (TRITC) labelled anti-rabbit serum, diluted 1:80, rinsed as before and mounted in glycerol containing 0.1% phenylene diamine as an anti-fading

agent (Johnson and de la NogueiraAraugo 1981). Control sections were incubated with antisera adsorbed for 2 h during active shaking with 50 /zg of the respective substance (NF, CGRP, SP, VIP, PHI, TH and NPY) per 1 ml of serum diluted 1:10. As an additional control, certain sections were incubated with the second antiserum only. The antisera used have been shown not to possess cross-reactivity with other substances in radioimmunological or immunohistochemical tests (see Luthman et al. 1988a). However, cross-reactivity with known, not tested substances, or with unknown peptides or proteins present in the sectioned tissue cannot be excluded, and therefore terms such as "NF-immunoreactive" have been considered appropriate in the text. The sections were examined in a Nikon Microphot-FX fiuorescence microscope. The density of the NF and CGRP fibers and fiber bundles was semiquantitatively estimated by a fivegraded scale (see also Table 1). 3-5 sections/subject for each substance were studied (NF, CGRP). The determinations from different sections obtained from one subject were averaged to constitute one observation and the data compared by the Mann-Whitney [/-test. Results Clinical examination

9 of the DS patients demonstrated gingivitis in the biopsy region and in two of these DS patients alveolar bone loss was diagnosed (the distance from the cemento-enamel junction to the alveolar crest exceeded 2 mm on X-ray), while 5 DS patients did not show any gingival or periodontal affections upon clinical examination. Two of the control subjects demonstrated a mild gingivitis, while the other subjects did not have any gingival affections observed clinically (see Table 1). Histopathological evaluation

Fig. 1. Micrographs of human gingiva from a clinically healthy control subject (A) and from a gingivitis affected DS subject (B). A dense infiltration of inflammatory cells can be seen in the propria of the DS patient. In the propria of DS subjects abundant mononucleated cells with an morphology resembling lymphocytes can be seen (C), while occasionally PMN cells (arrows) are found in the propria close to the epithelium in the upper part of the sulcus (D). Scale bar in A and B = 200 /zm and in C and D = 25 //m.

In the biopsies from the gingivitis- and periodontitis-affected sites in the DS patients a dense infiltration of cells was seen in the propria as well as close to and partly within the sulcus- and buccal-epithelium (Fig. IB, Table 1). The cells observed were mainly mononucleated, with an apparent lymphocyte morphology (Fig. lC). However, occasionally also PMN cells could be seen, mostly in the superficial portions of the

Gingival innervation of Down's syndrome

627

sulcus region (Fig, ID, Table 1), The epithelium showed an obvious hyperplasia of rete pegs in the biopsies studied from DS patients (Fig IB, Table 1), In the few cases of DS biopsies obtained from clinically healthy sites, a less pronounced infiltrate of inflammatory cells was found without any obvious epithehal hyperplasia. In the control biopsies obtained from cUnically healthy sites or sites with mild gingivitis no proliferation of the epithehum could be seen and only occasionally was a mild infiltration of inflammatory cells observed (Fig, lA, Table 1), Immunohistochemical evaluation

Losely and densely packed NF-immunoreactive fiber bundles were observed in the deeper parts of the propria, while single NF-immunoreactive fibers were mainly seen in the more superficial portions of the gingiva in both DS (Fig, 2A, B, C) and control subjects, NF-immunoreactive fibers and bundles were occasionally seen in close proximity to blood vessels, while in a few instances intra-epithehal NF-immunoreative fibers were found (Fig, 2D), In comparison to gingiva obtained from clinically healthy control subjects, gingiva from DS subjects showed an increase in NF-immunoreactive fibers as well as fiber bundles. The density of NF-immunoreactive structures as estimated by semiquantitative evaluation was shown to be significantly (p < 0,05) higher in the DS group compared to the controls (see Table 1). However, no obvious difference in the distribution of the NF immunoreactive structures was seen, i,e, the fiber bundles were mainly localized in the deeper parts of the propria, while single fibers occured in the more superficial parts, Futhermore, no marked difference in the detailed morphology of the NF-immunoreactive structures between DS and control subjects was observed. Fig, 2. Immunofluorescence micrographs of human gingiva after incubation with antiserum to NR (A) Gingival tissue obtained from region 31-32 in a 16-year-old male with DS, NFimmunoreactive bundles can be seen deep in the propria, while NF-immunoreactive nerve fibers occur in the superficial propria, close to the epithelial ridges, (B) Gingiva from region 24-23 in a 19-year-old female with DS, Several NF-immunoreactive fibers occur in close apposition to, or partly within the epithelium, (C) Propria from region 31-32 in a 12-year-old male with DS, Numerous NF-immunoreactive nerve bundles as well as a few single NFimmunoreactive fibers can be seen, (D) Intraepithelial NF-immunoreactive fiber in 12-yearold control subject, region 13-12, e = epithelium and p = connective tissue papilla. Scale bars represent 100 ^m in A and 50 im\ in D, Same magnification in A, B and C,

Single CGRP-immunoreactive nerve fibers were seen mainly close to the epithelium (Fig, 3A), while CGRP-immunoreactive fiber bundles, losely or densely packed, were observed in the deeper parts of the propria (Fig, 3B), Occasionally CGRP-immunoreactive fibers were also seen within the basal parts of the epithehum (Fig, 3A), In comparison to NF-immunoreactive structures, the CGRP-immunoreactive structures were less abundant, although their distribution was similar. In comparison to gingiva obtained from clin-

628

Barr-Agholme et al.

ically healthy control subjects, gingiva from DS subjects showed an increase in CGRP-immunoreactive fibers and fiber bundles. In agreement with this observation, the density of CGRP fibers and fibers bundles, as estimated by semiquantitative evaluation, was shown to be significantly (/7 = O.O5) higher in the DS group compared to the controls (Table 1). On the other hand, no obvious difference in the distribution of the CGRP fibers and fiber bundles was seen and the detailed morphology of the CGRP-immunoreactive structures did not differ between DS and control subjects.

Single nerve fibers immunoreactive to SP were also seen in the superficial layers of the propria (Fig. 3D) of which some were found in the epithelium, while no SP-immunoreactive fiber bundles were found. The distribution of the SP-immunoreactive fibers closely resembled that of CGRP fibers, even if the SP fibers were much less frequently observed. Due to the sparse distribution of the SP-immunoreactive fibers no estimation of the relative density in DS subjects compared to control subjects was performed. ; In gingiva of both DS subjects and control subjects, single VlP-immuno-

reactive fibers could be observed mainly close to blood vessels (Fig. 4A), or occasionally close to the epithelium. PHIimmunoreactive fibers were also found in the vicinity of blood vessels (Fig. 4B) and as single fibers in the propria. The TH-immunoreactive and NPY-immunoreactive fibers seen were also distributed mainly surrounding blood vessels (Fig. 5A, C), although some single fibers without any close proximity to vessels could be observed (Fig. 5B). No discernable difference in the occurrence of VIP, PHI, TH and NPY immunoreactive fibers was seen between gingiva obtained from DS subjects compared to control subjects. No immunoreactive structures were seen when antisera preadsorbed with the different immunogenes were used, or when incubations were performed in the presence of second antisera only. Discussion

Fig. 3. Immunofluorescence micrographs of human gingiva. The sections were incubated with CGRP and SP antisera, (A) CGRP-immunoreactive nerve fibers close to and in the epithelium. The sample was obtained from region 32-33 in a 21-year-old DS male patient with periodontitis, (B) CGRP-immunoreactive fibers and fiber bundles in a dense cluster in the propria of region 42-41 in a 19-year-old chnically healthy DS female patient, (C) CGRP-immunoreactive fibers in the propria close to the epithelium of a control male subject of 7 years. (D) SPimmunoreactive fiber in a connective tissue papilla in region 42-41 of a DS female subject. e = epithelium. Scale bar represents 100 /.im. Same magnification in A, B, C and D.

The present results demonstrate that patients with DS have profound infiammatory affections of the gingiva, as evidenced histologically. This finding is in agreement with previous studies describing severe gingivitis and periodontitis in patients with DS (Cohen et al. 1961, Johnson & Young 1963, Brown 1978, Modeer et al. 1990, see also Reuland-Bosma et al. 1986a). The biopsies from DS children showed a dense infiltration of inflammatory cells in the propria and close to the junctional epithelium. The different inflammatory cells observed had a similar distribution as that seen in other periodontal lesions (Attstrom 1970, Schroeder et al. 1973). Thus, the few PMN cells found were mainly found in the junctional epithelial region, while mononuclear lymphocytelike cells were seen throughout the propria. In addition, hyperplasia of both the sulcus and the buccal epithelium was found in the biopsies from the DS children, even in those DS subjects where no clinical signs of infiammation were seen. The observed epithelial hyperplasia is in agreement with findings from studies on the gingival morphology under clinically normal conditions (Cohen et al. 1961) as well as after experimentally induced gingivitis (Reuland-Bosma et al. 1988a). Thus, DS patients appear to have profound infiammatory involvement of the gingiva. In agreement with previous studies, nerve fiber-like structures immunoreactive to several different neuronal

Gingival innervation of Down's syndrome markers were found in the gingiva (Luthman et al. 1988a, b, 1989). A higher density of both NF-immunoreactive and CGRP-immunoreactive structures in the gingiva of DS patients was furthermore found compared to gingiva obtained from control subjects, while the general distribution of the NF and CGRP fibers and fiber bundles did not differ between DS and control subjects. The density of the fibers and fibers bundles was furthermore estimated by semiquantitative evaluation. Semiquantitative estimations of the innervation density were used in the present study, since both N F and CGRP occurred in single fibers as well as in losely or densely packed fibers bundles making direct quantitation of the nerve fiber density impossible to perform. The validity of semiquantitative estimations of nerve fiber density has been studied in different experimental models. In general a good correlation has been found between semiquantitative methods to estimate innervation density in relation to biochemical techniques (e.g., see Seiger & Olson 1977). Thus, semiquantitative estimation of nerve fiber density seems to provide a resonably accurate method to study differences in innervation density in cases when neither biochemical or automated image analysis techniques can be employed. The gingival material

investigated in the present study was obtained from the same regions in the oral cavity of both the DS patients and the control subjects (see Table 1). It is therefore not plausible that differences related to regional variations in the normal distribution of nerve fibers in the gingiva affected the estimation of nerve density. Moreover, it does not seem likely that differences in the age range of the subjects studied affected the results, since the material was taken from children of approximately the same age. The distribution of the neuronal specific intermediate filament N F seems to be confined to sensory nerve fibers in the peripheral nervous system (Seiger et al. 1984, see also Schlaepfer & Lynch 1977) and NF immunohistochemistry appears to label a specific population of intermediate and large sensory neurons (Lawson et al. 1984). Furthermore, N F immunohistochemistry seems to be a useful method to visualize sensory nerve fibers in the human skin (Dalsgaard et al. 1984) as well as in the gingiva and periodontium of rat and man (Maeda et al. 1987, Maeda 1987, Luthman et al. 1988b). CGRP and SP are found in small sensory neurons (Hokfelt et al. 1976, Rosenfelt et al. 1983), where they appear to co-exist (see Lundberg & Hokfelt 1986). CGRP and SP have been demonstrated to occur in the trigeminal

'g- 4, Immunofluorescence micrographs of human gingiva obtained from DS patients. The ctions were incubated with VIP and PHI antisera. (A) Thin VIP-immunoreactive fibers close ' a blood vessel (v) in the deeper parts of the propria of region 42-41 in a 19-year-old female th DS. (B) PHI-immunoreactive fiber in the vicinity of a blood vessel of region 42-43 in a -year-old male with DS. Scale bar represents 50 fim. Same magnification in A and B.

629

system, including oral regions, and possess similar physiological functions (Uddman et al. 1985, Wakisaka et al. 1985, Franco-Cereceda et al. 1987, Gazehus et al. 1987). Thus, the higher density of NF and CGRP fibers in the gingiva of DS patients most probably reflects a sensory hyperinnervation, which seems to involve sensory neurons of different sizes. No alteration in the distribution and occurrance of the other neuronal markers studied was observed. In contrast to CGRP, SP-immunoreactive fibers were very sparsely distributed and the density of the SP innervation was therefore not feasible to evaluate. VIP and PHI fibers as well as the TH and NPY fibers occurred mainly in the vicinity of blood vessels, even if some single PHI, VIP and NPY fibers were seen in the propria. These findings are in accordance with previous studies (Luthman et al. 1988) and indicate autonomic functions of these fibers. In fact, VIP and PHI appear to coexist in mainly parasympathetic nerve fibers in the peripheral nervous system (see Lundberg & Hokfelt 1986), while NPY is found in fibers expressing immunoreactivity to the catecholamine synthetizing enzyme TH, indicating a relation to the sympathetic system (Lundberg et al. 1982, Edwall et al. 1985). Although the infiammatory reactions in the DS gingiva may involve alteration of the blood vessel density, no obvious changes in the occurrance of the nerve fibers in the vicinity of vessels were observed. Connective tissue and vascularization changes have been suggested to play a role in the pathogenesis of the gingival reaction observed in DS (Cohen et al. 1961, Claycomb et al. 1970). It would therefore be of importance to study the morphology and density of blood vessels in the gingiva of patients with DS, concomitant with a more detailed study of the vessel innervation. Although further studies are needed to describe the involvement of the nervous systems in the processes of inflammation, including possible morphological alterations, several studies have suggested that sensory nerve fibers can mediate infiammatory reactions as well as undergo hyperinnervation in chronically inflamed tissue. The occurrence of plasma extravasation, edema, weal and erythema formation has been shown to be related to the sensory innervation, i.e. the neurogenic infiammation response (Bruce 1910, Celander & Folkow

630

Barr-Agholme et al.

1953). It has furthermore been suggested that release of putative sensory transmitters, Hke SP, may be mediators of neurogenic inflammation (Lembeck and Holzer 1979, Levine et al. 1984). A number of different neuropetides including SP and CGRP have also been shown to affect the prohferation of various cell types involved in the tissue responses in inflammation (Nilsson et al. 1985, see also Haegerstrand 1990). A role of sensory neurons in chronically

infiammatory processes is also indicated from studies on experimentally (adjuvant) induced arthritis in rats, where enhanced levels of SP have been observed (Colpaert et al. 1983, Levine et ai. 1984). Moreover, it has been shown that a significant group of afferent sensory nerve fibers are activated during infiammatory states, while these fibers are not normally responsive in healthy tissue to even severely noxious stimuli, (see McMahon & Koltzenburg 1990).

Fig. 5. Immunofluorescence micrograph of human gingiva from DS patients. The sections were incubated with TH or NPY antisera. (A-B) TH-immunoreactive fibers close to the lumen of a blood vessels (v) deep in the propria of region 42-41 of a 19-year-old female with DS. (C) NPY-immunoreactive fibers in the blood vessel wall of gingiva obtained from region 4241 in the same subject as in (A). Scale bar represent 50 /^m. Same magnification in A, B and

Thus, major alterations in the physiological activity of sensory nerves occurs in inflammatory reactions. However, there is also evidence suggesting morphological alterations of sensory nerve fibers in inflamed tissue. An apparent hyperinnervation of NF-immunoreactive fibers has been observed in connection with the surface of superficial skin wounds in humans (Hermansson et al. 1987). Furthermore, experimentally induced infiammation of rat pulp and periodontium has recently been shown to induce axonal sprouting with increased density of CGRP nerve fibers (Kimberly et al. 1987, Kimberly & Byers 1988), while a rich innervation of nerve fibers has also been shown surrounding periapical lesions in humans (Martinelli & Rulh 1967). There are also indications that alteration in the gingival NF-immunoreactive innervation occurs in periodontitis of man (Luthman et al. 1989). It therefore seems hkely that a sensory hyperinnervation occurs in the gingiva of DS patients as a consequence of the profound infiammatory involvement observed, although it cannot be completely excluded that DS patients have a relatively higher density of sensory nerve fibers per se in the gingiva. It has for instance been suggested that neurodystrophic effects could be engaged in the histological alterations observed in DS (Cohen et al. 1961). However, presently no evidence has been presented demonstrating a role of the nervous system in the morphological changes seen in DS. Thus, the observed hyperinnervation of sensory nerve fibers in the gingiva of DS patients is probably due to mechanisms related to general infiammatory processes and not specifically connected to DS. On the other hand, factors released from the sensory afferents may contribute to the infiammatory reactions seen in the gingiva of DS patients, an effect that could be potentiated by the increased density of the sensory innervation. The factors that could be involved in initiating sprouting of nerve fibers in infiammatory processes are presently unknown. A degeneration of nerve fibers has been shown to occur within heavily infiamed tissued, while stimulated growth is seen in the immediately adjacent regions (Kimberly & Byer: 1988), indicating an action of diffusibl' growth factors. It has been shown tha immune cells, such as macrophages ar' a major source for neuronotrophic fac tors (Nathan et al. 1980) and growt i

Gingival innervation of Down's syndrome factors released from immune cells may induce sprouting of nerve fibers in at least the central nervous system (see Churchill-Bohn & Kanuicki 1990). In the present study, an infiltrate of infiammatory cells was observed in the gingiva of DS patients, although no obvious infiltrate of macrophages could be seen. It is therefore possible that factors produced locally in the gingiva by different infiammatory cells could stimulate nerve growth. In conclusion, the present study shows a profound infiammatory involvement of the gingiva of DS patients during normal conditions. A higher density of NF- and CGRP-immunoreactive nerve fibers was furthermore observed in the gingiva of DS patients, indicating a sensory hyperinnervation. In contrast, no obvious alterations were seen in the distribution of neuronai markers related to the autonomic innervation. The sensory hyperinnervation is probably not specifically related to DS, but is due to a sprouting of afferent nerves in response to the infiammatory reaction. Thus, the sensory hyperinnervation of the gingiva of DS patients seems to present an example of structural alteration of nerve fibers induced by general infiammatory processes which could lead to production of neuronotrophic factors. On the other hand, the increased density of the sensory innervation could thereafter potentiate the infiammatory reaction by enhanced release of neuronai factors that contribute to infiammatory processes.

Acknowledgements This study was approved by the Regional Ethical Committee of Karolinska Institute (Huddinge Hospital). Support for this work was obtained by grants from the Swedish Medical Research Council (14X-06836), the Swedish Dental Society and the Karolinska Institute Research Funds. The authors are grateful to Miss Diana Blomquist for valuable clinical assistance. Miss Susie Tandberg for technical assistance and Mrs Ida Engqvist for expert secretarial ; ^sistance. For generous gifts of NF, SP, • HI, VIP and NPY antisera the authors i e grateful to Drs. D. Dahl, Boston, L. ' Ison, Stockholm, C. Cuello, Montreal, f Terenius, Uppsala and J. Fahrenk! g, Copenhagen. CGRP antiserum was 1 rchased from Peninsula Labora^ ies. Inc., Belmont.

Zusammenfassung Immunohistologische Studie neuronaler Marker in der entziindeten Gingiva von Kindern mit dem Down's Syndrom Das histologische Bild der Gingiva von Kindern mit dem Down's Syndrom (DS) wurde vor allem hinsichtlich seiner Entziindungsbeteiligung und Innervation untersucht. In der Propria der meisten DS-Patienten wurden aus Entzundungszellen und einigen wenigen polymorphkernigen Leukozyten bestehende, dichte Infiltrationen gesehen. Eine Epithelhyperplasie wurde ebenfalls beobachtet. Die Innervation der Gingiva wurde mit immunohistochemischen Methoden studiert. In der Propria wurden sowohl Nervenfasern als auch Nervenbiindel gesehen, die dem Neurofilament (NF) gegeniiber immunoreaktiv waren. Gelegendlich wurden intraepitheliale NF-Fasern beobachtet. Auf Calcitonin genenbezogene Peptide (CGRP)-immunoreaktive Fasern und Faserbiindel wurden gleichfalls sichtbar gemacht, doch kamen sie weniger reichlich vor als die NF-Fasern. Die Dichte der NF- und CGRP-Fasern und Faserbundel wurden mit semiquantitativen Verfahren beurteilt. Im Vergleich zu Kontrollprobanden wurde in der Propria von DS Patienten eine hohere Dichte der NF und CGRP immunoreaktiven Strukturen beobachtet, ein offensichtlicher Unterschied ihrer Verteilung in der Propria wurde jedoch nicht gesehen, ZusatzHch wurden sparlich verteilte Fasern festgestellt, die sich gegeniiber Peptid Histidin Isoleukin Amiden (PHI) und vasoaktiven intestinal Polypeptiden (VIP), wie auch Neuropeptid Y (NPY) und Tyrosin Hydroxylase (TH) immunoreaktiv verhielten und die hauptsachlich in der Umgebung von Blutgefassen gefunden wurden. Einige wenige Substanz P (SP) Fasern wurden ebenfalls entdeckt, meistens nahe dem Epithel. Zwischen Kontrollprobanden und DS Patienten wurden keine ins Auge fallendenden Unterschiede der sparsam verteilten Fasern gesehen. Eine profunde Entziindungsbeteiligung der Gingiva von DS Patienten wird festgestellt, bei gleichzeitiger Hyperinnervation vermuteter sensibler Komponenten der Zahnfleischinnervation, Dagegen wurden keine Veranderungen der Densitat der neuronalen Marker in Bezug auf die autonomen Nervenfasern gesehen. Die beobachtete sensorische Hyperinnervation ist moglicherweise nicht spezifisch fiir das DS, kann aber mit dem Spriessen afferenter Nervenendungen erklart werden, das durch die entziindliche Reaktion induziert wurde, Stoffe, die von den sensorischen Afferenten abgesondert werden, konnen zu der bei dem DS beobachteten Zahnfleischentziindung beitragen.

631

fants ayant le syndrome de Down (DS) a ete etudiee en se penchant specialement sur l'atteinte inflammatoire et l'innervation, Un infiltrat dense de cellules inflammatoires a ete apergu dans la propria de la plupart des enfants DS, y compris quelques leucocytes polymorphonucleaires, Une hyperplasie epitheliale a egalement ete notee. L'innervation gingivale a ete etudiee par immunohistochimie. Des fibres nerveuses ainsi que des faisceaux nerveux reagissant immunologiquement au neurofilament (NF) ont ete apergues dans la propria. Des fibres NF intraepitheliales ont aussi ete observees occasionnellement, Des fibres et des faisceaux reagissant immunologiquement au peptide en rapport genetique avec la calcitonine (CGRP) ont egalement ete mis en evidence mais ils etaient moins abondants que les fibres NF. La densite des fibres et faisceaux NF et CGRP a ete estimee semiquantitativement. Une densite superieure de structures reagissant immunologiquement au NF et CGRP a ete observee dans la propria des enfants DS lorsqu'ils etaient compares aux controles, tandis qu'aucune alteration marquante n'a ete aperfue dans Ieur repartition dans la propria. Le plus souvent situees autour des vaisseaux, des fibres clairsemees reagissaient immunologiquement au peptide amide histidine isoleucine (PHI), au polypeptide intestinal vasoactif (VIP), au neuropeptide Y (NPY) et a l'hydroxylase tyrosine (TH). Quelques fibres substance P (SP) ont egalement ete trouvees, surtout le long de I'epithelium. Aucune difference concernant ces fibres n'a ete remarquee entre les deux groupes d'enfants. Une atteinte inflammatoire profonde de la gencive des enfants DS est done observee simultanement avec une hyperinnervation du composant sensoriel presume de l'innervation gingivale. Par contre, aucune alteration n'a ete trouvee dans la densite des marqueurs neuraux en relation avec les fibres nerveuses autonomes. L'hyperinnervation sensorielle observee n'est probablement pas liee specifiquement au DS mais pourrait etre due a une augmentation du nombre de nerfs afferents induite par la reaction inflammatoire. Cependant, des facteurs liberes par les afferents sensoriels pourraient contribuer a l'inflammation gingivale apergue chez les enfant DS.

References

Attstrom, R. (1970) Presence of leucocytes in cervices of healthy and cronically inflamed gingivae. Journal of Periodontal Research 5, 42-47. Bernick, S. (1957) Innervation of teeth and periodontium after enzymatic removal of collagenous elements. Ora! Surgery, Ora! Medicine and Oral Pat!io!ogy 10, 323332. Breg, W. R. (1977) Down's syndrome: A reResume view of recent progress in research. Pathobiological Annual 1, 257-303. Etude invnunohistologique des marqueurs neuBrown, R. H. (1978) A longitudinal study of raux dans !a gencive enflammee d'enfants periodontal disease in Down's syndrome. ayant le syndrome de Down New Zealand Dental Journal 74, 137-144. L'apparence histologique de la gencive d'en-

632

Barr-Agholme et al.

Bruce, A. N. (1910) tJber die Beziehung der sensiblen Nervendigungen zum Enzundungsvorgang. Naunyn Schmiedebergs !.: Archives of Experimental Pathology and Pharmacology 63, 424-433. Churchill-Bohn, M. & Kanuicki, M, (1990) Bilateral recovery of striatal dopamine after unilateral adrenal grafting into the striatum of the l-methyl-4-(2'-methyl. phenyl)-l,2,3,6-tetrahydropyridine (2'CH3MPTP)-treated mouse. Journal of Neuroscience Research 25, 281-286, Celander, O. & Folkow, B. (1953) The nature and distribution of afferent fibers provided with axon reflex arrangement. : Acta Physiologica Scandinavia 29, 359-370. Claycomb, C. K., Summers, G. W,, Hall, W, B. & Hart, R, W, (1970) Gingival collagen biosyntesis in mongolism. Journal of Periodontal Research 5, 30-35. Cohen, M. M., Winner, R. H., Schwartz, S. & Shklar, G. (1961) Oral aspects of mongolism Part L Periodontal disease in mongolism. Oral surgery. Oral Medicine and Oral Pathology 14, 92-107, Colpaert, D, H., Donnerer, J. & Lembeck, F. (1983). Effects of capsaicin on inflammation and the substance P content of nervous tissue in rat with adjuvant arthritis. Life Sciences 32, 18271834, Coons, A, H, (1958), Fluorescent antibody methods. In: General Cytocheinical Methods, ed, Danielli, J, K, pp, 399-422, New York: Academic Press, Dalsgaard, C,-J,, Bjorklund, H,, Jonsson, C.-E., Hermanson, A, & Dahl, D, (1984) Distribution of neurofilament-immunoreactive nerve fibers in human digital skin, Histochemistry 81, 111-114, Desjardins, R, P, Winkelmann, R, K. & Gonzalez, J. B. (1971) Comparison of nerve endings in normal gingiva with those in mucosa covering endentulous alveolar ridges. Journal of Dental Research 50, 867-879, Dixon, A, D, (1962) The position, incidence and origin of sensory nerve terminations in oral mucous membrane. Archives of Oral Biology 1, 39^8, Edwall, B,, Gazelius, B,, Fazekas, A,, Theodorsson-Norheim, E, & Lundberg, J, M, (1985) Neuropep6tide Y (NPY) and sympathetic control of blood flow in oral mucosa and dental pulp in the cat, Acta Physiolocica Scandinavica 125, 253264, Franco-Cereceda, A,, Henke, H,, Lundberg, J. M, Petermann, J, B, Hokfelt, T. & Fischer J. A. (1987) Calcitonin gene-related peptide (CGRP) in capsaicin-sensitive substance P-immunoreactive sensory neurons in animal and man: Distribution and release by capsaicin. Peptides 8, 399410. Gairns, F. W. & Aitchinson, J. (1950). A preliminary study of the multiplicity of nerve endings in the human gum. Dental •:;: Records 70, 180-194.

Gazelius, B., Edwall, B., Olgart, L., Lundberg, J. M., Hokfelt, T. & Fischer, J. A, (1987) Vasodilatory effects and coexistence of calcitonin gene-related peptide (CGRP) and substance P in sensory nerves of cat dental pulp, Acta Physiologica Scandinavica 130, 33-40, Griffin, C, J, & Harris, R, (1968) Unmyelinated nerve endings in the periodontal membrane of human teeth. Archives of Oral Biology 13, 1207-1212. Haegerstrand, A. (1990) Sensory neuropetides in inflammation and cell proliferation. Thesis, Karolinska Institute, Stockholm, Sweden. Hermanson, A., Dalsgaard, C.-J., Bjorklund, H. & Lindblom, U, (1987) Sensory reinnervation and sensibility after superficial skin wounds in human patients, Neuroscience Letters 74, 377-382, Hokfelt, T,, Kellert, J, O,, Nilsson, G, & Pernow, B, (1976) Substance P: Localization in the central nervous system and in some primary sensory neurons. Science 190, 889-890, Johnson, G, D, & De la Nogueira-Araujo, G, M. (1981) A simple method of reducing the fading in immunofluorescence during microscopy. Journal of Immunological Methods 43, 349-350. Johnson, N. R & Young, M. A. (1963) Periodontal disease in Mongols. Journal of Periodontology 34, 41-47. Kimberly, C. L. & Byers, M, R, (1988) Inflammation of rat molar pulp and periodontium causes increased calcitonin generelated peptide and axonal sprouting. Anatomical Records 111, 289-300, Kimberly, C, L,, Byers, M, R, & Mecifi, K, B, (1987) Demonstration of inflammatory changes to nerve tissue in rat dental pulp using CGRP immunohistochemistry. Journal of Endodontics 13, 421, Lawson, S. N,, Harper, A, A,, Harper, E, I,, Garson, J, A, and Anderton, B, H, (1984) A monoclonal antibody against neurofilament protein specifically labels a subpopulation of rat sensory neurons. Journal of Comparative Neurology 228, 263-272, Lembeck, F, & Hoizer, P (1979) Substance P as a neurogenic mediator of antidromic vasodilatation and neurogenic plasma extravasation. Archives of Pharmacology 310, 175-183, Levine, J, D,, Clark, R,, Devor, M,, Helms, C , Moskowitz, M, A, & Basbaum, A, I, (1984) Intraneuronal substance P contributes to the severity of experimental arthritis. Science 226, 547-549, Levine, J, D,, Dradick, S, J., Roizen, M. F,, Helms, C, & Basbaum, A, I, (1986) Contribution of sensory afferents and sympathetic efferents to joint injury in experimental arthritis. The Journal of Neuroscience 6, 3423-3429, Levine, J, D,, Moskowitz, M, A. & Basbaum, A. I. (1985) The contribution of neurogenic inflammation in experimental arthritis. The Journal of Immunology 135, 843-847. -. .

Lundberg, J. M. & Hokfelt, T. (1986) Multiple co-existence of peptides and classical transmitters in peripheral autonomic and sensory neurons - functional and pharmacological implications. In: Progress in brain research, vol. 68, eds. Hokfelt, T., Fuxe, K,, Pernow, B,, pp, 241-262. Amsterdam: Elsevier, Lundberg, J, M,, Terenius, L,, Hokfelt, T,, Martling, C , Tatemoto, K,, Mutt, V,, Polak, J, M., Bloom, S,, & Goldstein, M, (1982) Neuropeptide Y (NPY)-like immunoreactivity in peripheral noradrenergic neurons and effects of NPY on sympathetic function, Acta Physiologica Scandinavia 116, 488-480, Luthman, J,, Dahllof, G,, Modeer, T, & Johansson, O, (1988a) Immunohistochemical study of neuronal markers in human gingiva with phenytoin-induced overgrowth, Scandinavian Journal of Dental Research 96, 339-346, Luthman, J,, Friskopp, J., Dahllof, G.,Ahlstrom, U,, Sjostrom, L. & Johansson, O, (1989) Immunohistochemical study of neurochemical markers in gingiva obtained from periodontitis-affected sites. Journal of Periodontal Research 24, 267-278, Luthman, J,, Johansson, O,, Ahlstrom, U, & Kvint, S, (1988b) Immunohistochemical studies of the neurochemical markers, CGRP, enkephalin, galanin, yMSH, NPY, PHI, proctolin, PTH, somatostatin, SP, VIP, tyrosine hydroxylase and neurofilament in nerves and cells of the human attached gingiva. Archives of Oral Biology 33, 149-158, Luzardo-Baptista, M. (1973) Intraepithelia! nerve fibers in the human oral mucosa. Oral Surgery 35, 372-376, Maeda, T. (1987) Sensory innervation of the periodontal ligament in the incisor and molar of the monkey, Macaca fuscata. An immunohistochemical study for neurofilament protein and glia-specific S100 protein, Archivum Histologicum Japonicum 50, 437-454, Maeda, T,, Iwanaga, T,, Fujita, T,, Takahashi, Y, & Kobayashi, S, (1987) Distribution of nerve fibers immunoreactive to neurofilament protein in rat molars and periodontium. Cell and Tissue Research 249, 13-23, Mantyh, P W,, Cotton, M, D,, Boehmer, C, G,, Welton, M, L,, Passaro, E, P,, Maggio, J, E, & Vigna, S, R, (1989) Receptors for sensory neurpetides in human inflammatory diseases: Implications for the effector role of sensory neurons, Peptides 10, 627-645, Martinelli, C, & Rulli, M, A, (1967) The irnervation of chronic inflammatory himan periapical lesions. Archives of Or I Biology 12, 593-600. McMahon, S, B, & Koltzenburg, M, (199' ) Novel classes of nociceptors: beyond S errington. Trends in Neuroscience 1 s 199-201, Modeer, T,, Barr, M, & Dahllof, G. (199 ') Periodontal disease in children wi ii

Gingival innervation of Down's syndrome Down's syndrome. Scandinavian Journal of Dental Research 98, 228-234. Nathan, C. R, Murray, H. W. & Cohn, Z. A. (1980) The macrophage as an effector cell. New Eng!and Journa! of Medicine 303, 622-626. Nilsson, J., Von Euler, A. M. & Dalsgaard, C.-J. (1985) Stimulation of connective tissue cell growth by substance P and substance K. Nature 315, 61-63. Rapp, R., Kristine, W. D. & Avery, J. K. (1957) A study of neuronai endings in the human gingiva and periodontal membrane. Journal of Canadian Dental Association 23, 637-643. Reuland-Bosma, W. & Van Dijk, L. J. (1986a) Periodontal disease in Down's syndrome. A review. Journal of Clinical Periodontology 13, 63-74. Reuland-Bosma, W., Van Dijk, L. J: & Van der Weele, L. (1986b) Experimental gingivitis around deciduous teeth in children with Down's syndrome. Journal of Clinical Periodontology 13, 294-300. Reuland-Bosma, W., Liem, R. S. B., Jansen, H. W. B., Van Dijk, L. J. & Van der Weele, L. T. (1988a) Morphological aspects of the gingiva in children with Down's syndrome during experimental gingivitis. Journal of Clinical Periodontology 15, 293302. Reuland-Bosma, W., Liem, R. S. B., Jansen,

H. W. B., Van Dijk, L. J. & Van der Weele, L. T. (1988b) Cellular aspects of and effects on the gingiva in children with Down's syndrome during experimental gingivitis. Journal of Clinica! Periodontology 15, 303-311. ' Rosenfelt, M. G., Mermod, J. J., Amara, S. G., Swanson, L. W., Sawchenko, P. E., River, J., Vale, W. W. & Evans, R. M. (1983) Production of a novel neuropeptide encoded by the calcitonin gene via tissue specific RNA processing. Nature 304, 129-135. Schlaepfer, W. W. & Lynch, R. G. (1977) Immunofluorescence studies of neurofilament in the rat and human peripheral and central nervous system. Journal of Cell Biology 74, 241-250. Schroeder, H. E., Miintzel-Pedrazzolo, S. & Page, R. (1973) Correlated morphometric and biochemical analysis of gingival tissue in early gingivitis in man. Archives of Oral Biology 18, 899-923. Seiger, A. & Olson, L. (1977) Quantitation of fiber growth in transplanted central monoamine neurons. Cell and Tissue Research 179, 285-316. Seiger, A., Dahl, D., Ayer-LeLievre, C. & Bjorklund, H. (1984) Appearance and distribution of neurofilament immunoreactivity in iris nerves. Journa! of Comparative Neuroiogy 223, 457-470.

633

Stewart, D. & Lewinsky, W. (1939) A comparative study of the innervation of gum. Proceedings of the Royal Society of Medicine 32, 1054-1062. Uddman, R., Edvinsson, L., Ekman, R., Kingman, T. & McCulloch, J. (1985) Innervation of the feline cerebral vasculature by nerve fibers containing calcitonin generelated peptide: trigeminal origin and coexistance with substance P. Neuroscience Letters 62, 131-136. Van Steenberghe, D. (1979) The structure and function of periodontal innervation. Journal of Periodontal Research 14, 185-203. Von Lautenbach, E. (1966) Uber Nervstrukturen im Weichgewebe des marginalen Parodonts. Schweizerische Monatschriftfiir Zahnheilkunde 76, 165-170. Wakisaka, S., Nishikawa, S,, Ichikawa, H., Matsou, S., Takano, Y. & Akai, M. (1985) The distribution and origin of substance P-like immunoreactivity in rat molar pulp and periodontal tissues. Archives of Oral Biology 30, 813-818. Address: Johan Luthman Department of Histology and Neurobiology Karolinska Institute Box 60 400 S-104 01 Stockholm Sweden