Immunohistochemical characterization of abnormal innervation of colon in Hirschsprung's disease using D7 monoclonal antibody

Immunohistochemical Characterization of Abnormal Innervation of Colon in Hirschsprung's Disease Using D 7 Monoclonal Antibody By Takao Fujimoto, Denis J. Reen, and Prem Puri Dublin, Ireland 9 Innervation p a t t e r n s in normal and aganglionic colon w e r e studied using a panel of antineuronal cell antibodies. One antibody, D 7, w h i c h recognizes a subset of neuronal cells of t h e peripheral and central nervous system reacted s t r o n g l y w i t h nerve fibers in t h e circular muscle of t h e normal colon. Immunohistochemical scanning of t h e entire r e s e c t e d specimen of colon f r o m t h r e e children w i t h Hirschsprung's disease d e m o n s t r a t e d large numbers of D 7 i m m u n o r e a c t i v e nerve fibers in t h e circular muscle of t h e ganglionic colon, f e w fibers in t h e transitional zone, and no i m m u n o r e a c t i v e fibers in t h e aganglionic s e g m e n t of b o w el. While t h e absence of D7 i m m u n o r a a c t i v e fibers paralleled t h e absence of m y e n t e r i c ganglion cells in t h e aganglionic segment, a critical region of colon w a s identified w h e r e i n D 7 reactive fibers w e r e evident ahead of t h e appearance of ganglion cells. These findings indicate t h a t t h e fundamental p a t h o l o g y in Hirschsprung's disease is n o t only t h e absence of ganglion cells of t h e m y e n t e r i c and submucuous plexuses but also t h e absence of D 7 i m m u n o r e a c t i v e fibers in the circular muscle of t h e colon. 9 1 9 8 7 b y Grune & S t r a t t o n , Inc.

sprung's disease using D 7 monoclonal antibody and a range of other antineuronal cell antibodies. MATERIALS AND METHODS The entire resected specimen of colon was obtained from three patients with Hirschsprung's disease following Swenson's operation? One patient had classic rectosigmoid Hirschsprung's disease, one had long-segment aganglionosis, and one had total colonic aganglionosis. The resected colon was cut at 0.5-cm intervals and the most distal specimen was coded number 1 (Fig 1). Full-thickness strips of colon, measuring 0.5 x 0.5 cm were removed, snap frozen, and stored at - 2 0 ~ until used. As a control study, colonic tissue was obtained from four patients during closure of colostomy for imperforate anus. Colonic tissues were also obtained at the time of autopsy from age-matched controls with no evidence of gastrointestinal disease.

Immunocytochemistry Immunocytochemical studies were carried out on tissue sections using the avidin-biotin-peroxidase complex method 7 and the indirect

INDEX WORDS: Hirschsprung's disease; neuronal cells; monoclonal antibody.

L T H O U G H it is now 100 years since Hirschsprung first described congenital megacolon as a distinct clinical entity, the pathophysiologic basis of Hirschsprung's disease is still not fully understood. A major obstacle to the understanding of the pathophysiology of aganglionosis has been the lack of specific markers for various neuronal cell types that may be affected in this condition. Recently, with the availability of polyclonat and in particular monoclonal antibodies it has become possible to identify more precisely the innervation patterns in normal and aganglionic colon. TM We have produced a monoclona] antibody, D 7, which recognizes a subset of neuronal antigens that are distributed throughout the wall of normal rectum and colon. 5 The purpose of this study was to characterize the nature and extent of innervation of the entire resected specimen of colon from patients with Hirsch-

40 39 a8 37 36 35 34 33 32 31 30 29

A

From the Children's Research Centre, Our Lady's Hospital for Sick Children, Dublin. Presented at the 33rd Annual Congress of the British Association of Paediatric Surgeons, Birmingham, England, July 16-18, 1986. Address reprint requests to Prem Purl, Children's Research Centre, Our Lady's Hospital for Sick Children, Crumlin, Dublin 12, Ireland. 9 1987 by Grune & Stratton, Inc. 0022-3468/87/2203~9016503.00/0 246

GANGLIONIC BOWEL

28

'26 25

24 23 22

,

21 2o i~

~

TRANSITIONAL

J

ZONE

18

9

17 i6' 15 14 13 12 11 lo 9 8

7 6

s,

43

AGANGLIONIC

BOWEL

2 1

Fig 1. Diagramatic representation of location of tissue sections examined in patients with Hirschsprung's disease.

Journal of Pediatric Surgery, Vol 22, No 3 (March), 1987: pp 246-251

HIRSCHSPRUNG'S DISEASE USING D~ MONOCLONAL ANTIBODY

247

/ ( ..

/, I . ~.qr

!

\

L.,

.Z-

-/

"r

q O

Fig 2. Immunohistochemical localization of D 7 in normal colon using peroxidase method. (A) Myenteric (arrows) and submucosal (arrowheads) ganglia (• (B) Myenteric plexus showing D~-stained fine nerve meshwork surrounding ganglion cells ( • 500). (C) Numerous D 7 immunoreactive nerve fibers in the circular muscle ( x 50),

immunofluorescence method of Coons. 8 The following antisera were used at appropriate dilutions: D 7 monoclonal antibody, antineurofilament protein, 68kD (Dako, Glostrup, Denmark), antineurofilament protein, NF2F11, 70 kD, 200 kD (Sambio, Uden, The Netherlands), anti-neuron-specific enolase (NSE; Dako), and anti-S-100 protein (Dako). Cryostat sections (8 um) were incubated with antisera for 24 hours at 4~ The second and third layers were incubated for one hour each at room temperature. Visualization of

the peroxidase was achieved by the addition of 3, 3'-diaminobenzidine as substrate.

Enumeration o f D 7 Immunoreactive Nerve Fibers in the Circular Muscle The number of D 7 immunoreactive nerve fibers were counted in three visual fields per section using a • 10 objective, The mean

248

FUJIMOTO, REEN, AND PURl

number of immunoreactive fibers per mm 2 was then determined in each case.

RESULTS

Reactivity With D 7 Monoclonal Antibody Controls. D 7 immunoreactive fibers were present in abundance in both myenteric and submucous plexuses (Fig 2A). While the dense network of D 7 immunoreactive fibers in myenteric ganglia sometimes gave the microscopic appearance of staining of the cell cytoplasm, closer examination of the staining pattern showed that D 7 immunoreactivity was confined to neuronal components other than the cell body (Fig 2B). Numerous D 7 immunoreactive fibers were observed throughout the circular muscle, running parallel to smooth muscle cells (Fig 2C). These fibers appeared to be located between muscle cells. In the longitudinal muscle, only a few fibers were seen, which appeared to penetrate into the muscle but not ramify. A few D 7 immunoreactive fibres were also present in the lamina propria, muscularis mucosae, and submucosa. Aganglionic segment in Hirschsprung's disease. The aganglionic segment of bowel was characterized by the absence of D 7 immunoreactive fibers in the circular muscle and submucosa (Fig 3A). A few D 7 reactive fibers could be seen in the aganglionic segment as it approaches the oligoganglionic region.

Hypertrophic nerve trunks showed strongly positive immunoreactivity both in the submucosa and the intermuscular space (Fig 3B). Oligoganglionic segment in Hirschsprung's disease. The oligoganglionic segment contained occasional D 7 immunoreactive ganglia and hypertrophic nerve trunks (Fig 4A). In the circular muscle coat, the number of D 7 immunoreactive fibers had increased compared to the aganglionic bowel (Fig 4B). The number of D 7 immunoreactive fibers in the ganglionic colon in the three cases studied were 123, 118, and 114 fibers per mm2, respectively (Fig 5). While D 7 reactive fibers made their appearance in low numbers in the proximal end of the aganglionic segment, a sudden increase in circular muscle innervation was seen in the oligoganglionic segment. The presence of D 7 immunoreactive fibers preceded the appearance of ganglion cells in all patients. Ganglionic segment in Hirschsprung's disease. The distribution and number of D 7 immunoreactive fibers was similar to that seen in controls. D7

Reactivity With Other AntiNeuronal Cell Antibodies The pattern of immunoreactivity with antineurofilament protein (Dako), antineurofilament protein (Sanbio), anti-NSE, and anti-S- 100 protein in normal colon and aganglionic colon was similar to that reported by other investigators. In the normal colon, both antineu-

Fig 3. Immunohistochemical localization of D 7 in aganglionie colon using peroxidase method. (A) Staining of circular muscle and intermuscular space showing presence of hypertrophic nerve trunks and absence of D~ reactive nerve fibers in the circular muscle ( x 125). (B) Hypertrophic nerve trunk in the intermuscular sulcus (x 500).

HIRSCHSPRUNG'S DISEASE USING D7 MONOCLONAL ANTIBODY

249

Fig 4. Immunohistochemical localization of D7 in oligoganglionic colon. (A) Immunofluorescent localization of D? showing both ganglia (arrow) end hypertrophic nerve trunks in the intermuscular space (• 500). (B) D7 reactive nerve fibers in the circular muscle by peroxidase method ( • 50).

OLIGOGANGLIONIC A G A N G L I O N I C9~~

L

GANGLIONIC

~' C a s e I

OLIGOGANGLIONIC r

AGANGLIONIC

~

GANGLIONIC

~) C a s e 2

OLIGOGANGLIONIC

#

AGANGLIONIC

Case3 ~GANGLIONIC~

E w

==

125

[--

..Q '*"

100

(~

75

! E

Case1

~ ~

Case 2

Case 3

50,

E_ "6 d Z

]

i 25

2

4

6

8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66

Length of resected specimen(cm) Fig 5. disease.

Frequency and distribution of D7 immunoreactive fibers in the circular muscle of colon in three patients with Hirschsprung's

250

FUJIMOTO, REEN, AND PURl

Fig 6. Immunohistochemical localization of antineurofilament protein and anti-NSE in ganglionic and aganglionic colon. (A) Antineurofilament protein-positive nerve fibers in the circular muscle of ganglionic bowel (• 50). (B) Staining of aganglionic colon with antineurofilament protein showing positive nerve fibers in the circular muscle (• (C) Anti-NSE positive nerve fibers in the circular muscle of the ganglionic bowel (• (D) Staining of aganglionic colon with anti-NSE showing NSE positive nerve fibers in the circular muscle (xS0).

HIRSCHSPRUNG'S DISEASE USING D7 MONOCLONAL ANTIBODY

rofilament antibodies stained axons in the myenteric and submucous plexuses, but weak immunoreactivity was seen in ganglion cells (Fig 6A). In the aganglionic segment, hypertrophic nerve trunks were heavily stained as well as reduced numbers of nerve fibers in the circular muscle (Fig 6B). Anti-NSE antibody strongly stained ganglion cell bodies as well as numerous nerve fibers in the normal colon (Fig 6C). In the aganglionic segment N S E immunoreactive ganglion cells were absent, with a concomitant reduction in the number of immunoreactive nerve fibers (Fig 6D). Hypertrophic nerve trunks showed a positive reactivity with anti-NSE antibody. The level of immunoreactivity with S- 100 protein was generally weaker than that seen with other antineuronal cell antibodies. However, its distribution was similar to that of N S E in normal and aganglionic colon. Anti-S-100 antibody protein did not stain the ganglion cell body in myenteric and submucous plexus in the normal colon. DISCUSSION

I m m u n o h i s t o c h e m i c a l scanning of the entire resected specimen of colon from three children with Hirschsprung's disease demonstrated large numbers of D 7 immunoreactive nerve fibers in the circular muscle of the ganglionic colon, few fibers in the transitional zone, and no immunoreactive fibers in the aganglionic segment of bowel. While the absence of D 7 immunoreactive fibers paralleled the absence of myenteric ganglion cells in the aganglionic segment, a critical region of colon was identified wherein D 7 reactive fibers were evident ahead of the appearance of ganglion cells. In the three patients studied, D 7 reactive fibers were present in low numbers in the immediate preoligoganglionic region. A sudden increase in D 7 fibers, approaching normal colonic levels, occurred in the oligoganglionic region corresponding with the appearance of ganglion cells in the myenteric and submucous plexuses. Comparative studies with other antineuronal cell antibodies showed that while neurofilament N S E and S-100 positive fibers in the circular muscle were present in abundance in aganglionic tissue and reduced in number in aganglionic bowel, D 7 monoclonal antibody alone identified a neuronal cell antigen in the circular muscle that is present in abundance in the ganglionic segment but is completely absent in the distal aganglionic bowel. We have previously shown by the use of immunofluorescence techniques that D 7 monoclonal antibody reacts with ganglion cells of the myenteric and submucous plexuses of the gastrointestinal tract, as well as recognizing an axonal antigen that is distributed throughout the normal colon and rectum. However, using immunoperoxidase staining together with nuclear

251

counterstaining in this study, it is clear that the antigen recognized by this antibody is not in the ganglion cell body but is located in the supporting neuronal cell infrastructure. Furthermore, staining of the central nervous system also showed that D 7 does not react with neuronal cell bodies but with neuronal components associated with them (unpublished data). This antibody, therefore, appears to recognize a unique neuronal antigen present in the peripheral and central nervous system in humans. There is no clear understanding at present of the pathophysiology of Hirschsprung's disease. The most important histologic finding in this disease is the absence of ganglion cells in the myenteric and submucous plexuses. Various other innervation abnormalities that have been described in this disease include denervation of the adrenergic nerve supply, 9 adrenergic nerve hyperactivity, ~~ cholinergic nerve hyperplasia, H the absence of nonadrenergic inhibitory system, ~2 and abnormal peptidergic innervation) '3 Our findings indicate that the fundamental pathology in Hirschsprung's disease is not only the absence of ganglion cells of the myenteric and submucous plexuses but also the absence of D 7 immunoreactive fibers in the circular muscle of the colon. REFERENCES

1. Bishop AE, Polak JM, Lake BD, et al: Abnormality of the colonic regulatory peptides in Hirschsprung's disease. Histopathology 5:679-688, 1981 2. Taguchi T, Tanaka K, Ikeda K: lmmunohistochemicalstudy of neuron specificenolase and S-100 protein in Hirschsprung's disease. Virchows Arch [A] 405:399-409, 1985 3. Tsuto T, Okamura H, Fukui K: An immunohistochemical investigation of gut hormones in the colon of patients with Hirschsprung's disease. J Pediatr Surg 20:266-270, 1985 4. Kluck P, van Muijen GNP, van der Kamp AWM, et al: Hirschsprung's disease studied with monoclonal antineurofilament antibodies on tissue sections. Lancet 1:652-654, 1984 5. Scallen C, Purl P, Reen DJ: Identification of rectal ganglion cells using monoclonalantibodies. J Pediatr Surg 20:37-40, 1985 6. Swenson O, Bill AH: Resection of rectum and rectosigmoid with preservationof the sphincter for benign spastic lesions producing megacolon.Surgery 24:212-220, 1948 7. Hsu S-M, Raine L, Fanger H: Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques. J Histochem Cytochem 29:577-580, 1981 8. Coons AH: Histochemistry with labelled antibody. Int Rev Cytol 5:1-23, 1956 9. Ehrenpreis T: Some newer aspects on Hirschsprung's disease and allied disorders. J Pediatr Surg 1:329-337, 1966 10. Bennett A, Garrett JR, Howard ER: Adrenergic myenteric nerves in Hirschsprung's disease. Br Med J 1:487-489, 1968 11. Kamijo K, Hiatt RB, Koelle B: Congenital megacolon. A comparisonof the spastic and hypertrophic segmentswith respect to cholinesterase activities and sensitivities to acetylcholine DFP and barium ion. Gastroenterology24:173-188, 1953 12. Frigo GM, Del Tacca M, Lecchini S, et al: Some observations on the intrinsic nervous mechanism in Hirschsprung's disease. Gut 14:35-40, 1973