Expression of Cell Adhesion Molecules and Doublecortin in Canine Anaplastic Meningiomas

Abstract

Tumor cell invasion into the surrounding nervous tissue is one of the histologic hallmarks of anaplastic meningiomas. To identify other possible markers for aggression in canine meningiomas, the relationship between histologic features and the expression of molecules involved in cell adhesion, cell proliferation, and invasion was examined. Immunohistochemistry for epithelial cadherin (E-cadherin), neural cadherin (N-cadherin), β-catenin, doublecortin (DCX), and Ki-67 was performed for 55 cases of canine meningioma. DCX was preferentially expressed in tumor cells invading the brain parenchyma (12 of 14 cases), suggesting its involvement in the invasion process. Regardless of the histologic type, E-cadherin and N-cadherin expression was observed in 31 of 55 and 44 of 55 cases, respectively. There was a significant positive correlation between DCX and N-cadherin expression and a significant negative correlation between E-cadherin and N-cadherin expression, suggesting that decreased E-cadherin and increased N-cadherin expression induce DCX expression. Typical membranous β-catenin expression was observed in 10 of 55 cases, whereas nuclear translocation was observed in 33 cases. Nuclear β-catenin expression was frequently found in anaplastic meningiomas (12 of 14 cases). The Ki-67 labeling indices were significantly higher in anaplastic meningiomas than in other types. These findings indicate that the expression of N-cadherin and DCX and the nuclear translocation of β-catenin are closely associated with the presence of invasion and anaplasia in canine meningiomas. Notably, granular cell meningiomas were negative for almost all the molecules examined, suggesting that they have a different tumor biology than other meningiomas.

Keywords

dogs cell adhesion molecules immunohistochemistry invasion meningiomas

Meningiomas originate from the meningothelial cells covering the central nervous system¹⁰ and constitute about 40% of all primary intracranial tumors in dogs.^4,22 Canine meningiomas are histologically diverse and can undergo mesenchymal and epithelial differentiation, similar to their human counterparts.^17,23 The World Health Organization (WHO) classification for brain tumors of domestic animals includes 9 histologic types of canine meningiomas: meningothelial, fibrous (fibroblastic), transitional, psammomatous, angiomatous, papillary, granular cell, myxoid, and anaplastic (malignant).¹⁰ All except the anaplastic type are slow growing and have a relatively benign progression. In domestic animals, the assessment of meningioma malignancy is still based on morphologic criteria, such as a high mitotic rate and cellularity, patternless growth, and invasion into the surrounding nervous tissue.^10,14 In human meningiomas, tumor cell invasion into the brain parenchyma is the most crucial,^18,24 although this is not a factor in the WHO criteria.²⁸ Few indicators for malignancy are available for canine meningiomas,¹⁴ and further studies are necessary to identify molecules that may correlate with tumor invasion.

Doublecortin (DCX), which plays a crucial role in neuroblast migration during the development of the cerebral cortex, is highly expressed in invasive brain tumors, including malignant meningiomas in people.³ The expression of DCX protein is stronger at the margin of a tumor lesion than at the center,³ suggesting that DCX would be a good marker for invasion. In human meningiomas, Ki-67 is a standard marker of cell proliferation activity and is widely used to evaluate tumor grade and malignancy during pathologic diagnosis.¹⁸ Epithelial cadherin (E-cadherin) is expressed in normal arachnoidal tissue as well as in epithelial tissue and is the main cadherin protein expressed in human meningioma cells.^1,21 The E-cadherin cell adhesion system includes β-catenin, which indirectly connects cadherin to actin filaments.^2,15,25 Altered expression of β-catenin occurs in several human tumors, including gastric, colon, hepatocellular, and esophageal tumors.^15,25 Suppression of E-cadherin expression may result in cells acquiring a mesenchymal morphology and increased motility.²⁶ The expression of neural cadherin (N-cadherin) is often increased in place of E-cadherin.^8,26 One of the etiologic hallmarks of human meningiomas is the loss of chromosome 22.^16,19 Additional hallmarks in human meningioma include the loss of heterozygosity of the neurofibromatosis type 2 (NF2) gene and the loss of the NF2-encoded protein merlin, which functions as a tumor suppressor factor by controlling cadherin-mediated cell–cell adhesion.^2,5,11,16,19 Although the expression of cadherin molecules are believed to play a role in the morphogenesis of human meningiomas,^2,6 there has been no such finding for canine meningiomas.

The objective of the present study is to analyze the expression of a series of cell adhesion and associated molecules in canine meningiomas and compare their expression in benign and anaplastic cases. The molecules examined were relevant to cell adhesion (E-cadherin, N-cadherin, and β-catenin), cell invasion (DCX), and cell proliferation (Ki-67). This data will provide new information regarding the histogenesis and histologic grade of canine meningiomas.

Materials and Methods

Tissue Samples and Histopathology

Fifty-five canine meningiomas were collected from 13 necropsy and 42 surgical biopsy cases during the last 5 years (2005–2009). The tumor samples included 21 meningothelial, 4 fibrous, 10 transitional, 3 psammomatous, 1 angiomatous, 2 granular cell, and 14 anaplastic meningiomas. Table 1 shows the breed, sex, age, tumor location, histologic subtype, and clinical evidence of parenchymal invasion for each case.

Table 1.

Canine Meningiomas Examined in the Present Study

Case	Breed	Sex	Age^a	Tumor Location^b	Histologic Subtype	Clinical Evidence of Parenchymal Invasion^c	Histological Evidence, Degree of Invasiveness^d
1	Mix	Female	13	R temporal lobe	Meningothelial	ND	A
2	Mix	Male	15	R temporal lobe	Meningothelial	A	A
3	Golden Retriever	Female	9	Intracranial	Meningothelial	ND	A
4	Toy Poodle	Spayed	12	Cervical spinal cord	Meningothelial	A	A
5	Pomeranian	Male	6	R frontal lobe	Meningothelial	ND	A
6	Shih Tzu	Male	9	R frontal lobe	Meningothelial	ND	A
7	Mix	Castrated	9	L temporal lobe	Meningothelial	ND	A
8	Shih Tzu	Spayed	11	Forebrain	Meningothelial	A	A
9	Yorkshire Terrier	Castrated	9	Cervical spinal cord	Meningothelial	ND	A
10	Mix	Castrated	16	L frontal lobe	Meningothelial	ND	A
11	Bernese Mountain	Male	ND	L olfactory bulb	Meningothelial	A	A
12	Golden Retriever	Male	11	Falx cerebri	Meningothelial	ND	A
13	Golden Retriever	Spayed	9	R frontal lobe	Meningothelial	A	A
14	Beagle	Male	7	Frontal lobes	Meningothelial	ND	A
15	Australian Kelpie	Male	ND	Cerebral base	Meningothelial	ND	A
16	Toy Poodle	Castrated	9	Intracranial	Meningothelial	ND	A
17	Golden Retriever	Spayed	7	Frontal lobes	Meningothelial	P	P (1b)
18	Golden Retriever	Male	11	Parietal lobes	Meningothelial	ND	A
19	Golden Retriever	Unknown	12	R frontal lobe	Meningothelial	ND	A
20	Shih Tzu	Male	ND	L cerebral hemisphere	Meningothelial	ND	A
21	Golden Retriever	Spayed	7	R temporal lobe	Meningothelial	A	A
22	Siberian Husky	Unknown	10	R temporal lobe	Fibrous	ND	A
23	Mix	Castrated	14	R frontal lobe	Fibrous	A	A
24	Labrador Retriever	Female	8	L temporal lobe	Fibrous	ND	P (1b)
25	Mix	Female	ND	Frontal lobes	Fibrous	ND	A
26	Golden Retriever	Castrated	11	R cerebral hemisphere	Transitional	P	P (1b)
27	Shih Tzu	Unknown	ND	L parietal lobe	Transitional	A	A
28	Miniature Dachshund	Male	8	L olfactory bulb	Transitional	A	A
29	Labrador Retriever	Male	10	Intracranial	Transitional	ND	A
30	Siberian Husky	Spayed	12	Olfactory bulbs	Transitional	ND	A
31	Golden Retriever	Female	10	R parietal lobe	Transitional	ND	A
32	Mix	Castrated	13	R frontal lobe	Transitional	ND	A
33	Maltese	Spayed	13	Frontal lobes	Transitional	ND	A
34	Boxer	Spayed	7	R frontal lobe	Transitional	ND	A
35	Mix	Spayed	ND	R temporal lobe	Transitional	A	A
36	Miniature Dachshund	Castrated	14	R frontal lobe	Psammomatous	A	A
37	Mix	Spayed	12	Intracranial	Psammomatous	ND	A
38	Shetland Sheepdog	Female	ND	R temporal lobe	Psammomatous	P	A
39	Mix	Female	13	L temporal lobe	Angiomatous	ND	A
40	Shih Tzu	Female	11	R frontal lobe	Granular cell	P	P (2)
41	Miniature Schnauzer	Male	11	R cerebral hemisphere	Granular cell	P	P (2)
42	Golden Retriever	Spayed	12	Cervical spinal cord	Anaplastic	ND	P (2)
43	Chihuahua	Male	10	R frontal lobe	Anaplastic	P	A
44	Bulldog	Male	7	L frontal lobe	Anaplastic	P	P (2)
45	Boston Terrier	Castrated	13	L temporal lobe	Anaplastic	ND	A
46	Maltese	Female	13	Falx cerebri	Anaplastic	ND	P (2)
47	Shiba	Spayed	11	L olfactory bulb	Anaplastic	P	P (2)
48	Mix	Castrated	11	Intracranial	Anaplastic	A	P (2)
49	Toy Poodle	Male	14	R frontal lobe	Anaplastic	ND	A
50	Golden Retriever	Spayed	11	L frontal lobe	Anaplastic	ND	A
51	Mix	Castrated	21	R frontal lobe	Anaplastic	P	A
52	Shiba	Spayed	ND	L olfactory bulb	Anaplastic	P	P (2)
53	Golden Retriever	Spayed	11	R frontal lobe	Anaplastic	P	P (2)
54	Mix	Male	12	Frontal lobes	Anaplastic	ND	P (2)
55	Pembroke Welsh Corgi	Spayed	7	L frontal lobe	Anaplastic	P	P (2)

^a In years. ND, no data.

^b R, right; L, left.

^c A, absent; P, present; ND, no data.

^d A, absent; P, present; 1b, tumor cell penetration of pial membrane only; 2, tumor cell infiltration of the parenchyma.

The tissue samples were routinely fixed in 10% phosphate-buffered formalin solution. Paraffin sections (2 to 4 μm thick) were stained with hematoxylin and eosin. Histologic classification of canine meningiomas was performed according to the current WHO classification guidelines for domestic animal tumors.¹⁰

Immunohistochemistry

Immunohistochemistry was performed using the labeled streptavidin biotinylated antibody method. Table 2 lists the primary antibodies used and the canine tissues with positive reactions. Deparaffinized sections were first autoclaved at 120°C for 10 minutes for antigen retrieval. The tissue sections were then treated with 1% hydrogen peroxide / methanol for 30 minutes and incubated in 8% skimmed milk / Tris-buffered saline at 37°C for 40 minutes to block nonspecific reactions. Sections were then incubated at 4°C overnight with the primary antibodies. After being washed 3 times in Tris-buffered saline, the sections were incubated with biotinylated anti-mouse, anti-rabbit, or anti-goat immunoglobulin G (1:400, KPL, Guildford, UK) at 37°C for 40 minutes and then with horseradish peroxidase–conjugated streptavidin (1:300, Dako Japan, Kyoto, Japan) at room temperature for 40 minutes. Finally, the reaction products were visualized with 0.05% 3-3′-diaminobenzidine and 0.03% hydrogen peroxide in Tris/HCl buffer, before being counterstained with Mayer’s hematoxylin.

Table 2.

Primary Antibodies Used in the Present Study

Antibody		Dilution	Source	Positive Reactions in Canine Tissues	Antigen Retrieval	References
Epithelial cadherin	Mouse anti-human monoclonal, 4A2C7	1:100	Invitrogen, Camarillo, CA	Meningothelial, epithelial cells	Target retrieval solution, pH 9.0 (Dako)	1, 2, 21
Neural cadherin	Mouse anti-human monoclonal, 6G11	1:25	Dako, Kyoto, Japan	Neurons	Target retrieval solution, pH 9.0	26
β-catenin	Rabbit anti-human polyclonal	1:400	Millipore, Billerica, MA	Meningothelial, epithelial cells	Target retrieval solution, pH 9.0	2, 15, 25
Doublecortin	Goat anti-human polyclonal	1:200	Santa Cruz Biotechnology, Santa Cruz, CA	Subventricular zone cells	Citrate buffer, pH 6.0	3
Ki-67	Mouse anti-human monoclinal, MIB-1	Ready to use	Dako, Kyoto, Japan	Proliferating cells	Citrate buffer, pH 6.0	18

The percentage of immunopositive cells was defined as the ratio of strongly labeled tumor cells to the total number of cells examined in 5 high-power (400×) fields. Sampled fields consisted of randomly selected nonoverlapping areas, particularly tumor borders, with highest labeling in foci of invasion into brain parenchyma. Box and whisker plot analyses were performed on the minimum number of counted cells per antigen.

Statistical Analyses

All data analyses were performed with SPSS 15.0. First, a chi-square test was performed to determine the statistical differences between each immunohistochemical stain score and histologic tumor type and between each immunohistochemical stain score and canine meningioma tumor location (frontal lobe, temporal lobe, parietal lobe, olfactory bulb, and cervical spinal cord). Based on these statistical data, comparisons between benign and anaplastic meningiomas were performed with the chi-square test. Then, linear discriminant analyses were performed for association between immunohistochemical parameters and their discriminant values in anaplastic meningiomas. Pearson correlation coefficient analysis was used to assess whether there was any correlation between the immunostaining scores of cadherin molecules and those of nuclear β-catenin, DCX, and Ki-67. A P value of .05 was considered to be significant for all analyses performed.

Results

Histopathology

Table 1 summarizes histologic findings and degree of tumor invasiveness. Histologic evidence of invasive behavior was found in 14 of 55 cases (25%). In the 14 invasive cases, tumor cell infiltration into the parenchyma was observed in 11 cases (79%), and penetration of tumor cells into the pial membrane was observed in 3 cases (21%). Tumor cell dissemination along Virchow-Robin spaces was not observed in any case.

Immunohistochemistry

Table 3 summarizes immunohistochemistry results for each marker molecule. First, the immunoreactivity for E-cadherin, N-cadherin, and β-catenin in normal arachnoidal tissue was examined. Normal arachnoidal samples were collected at necropsy from a 5-year-old male Beagle dog without neurologic disease. E-cadherin was detected along the cell membranes of normal arachnoidal cells, whereas N-cadherin was not (Fig. 1a, 1b). β-Catenin was also distributed along the cell membranes but not in cytoplasm and nuclei of intact arachnoidal cells (Fig. 1c). Arachnoidal cells were completely immunonegative for DCX and Ki-67.

Table 3.

Immunohistochemical Expression of Cell Adhesion and Other Related Molecules in Canine Meningiomas^a

Case	Histologic Type	Epithelial Cadherin	Neural Cadherin	Membranous β-Catenin	Nuclear β-Catenin	Doublecortin	Ki-67
1	Meningothelial	++	–	–	–	–	+
2	Meningothelial	–	+++	–	++	+	+
3	Meningothelial	+	–	–	–	–	+
4	Meningothelial	–	+++	–	–	++	+
5	Meningothelial	++	–	+++	–	–	–
6	Meningothelial	–	+++	++	+	–	–
7	Meningothelial	–	+++	–	–	–	+
8	Meningothelial	+	++	–	–	–	+
9	Meningothelial	+++	–	+++	–	–	–
10	Meningothelial	+++	–	++	–	–	–
11	Meningothelial	–	+++	–	+	–	+
12	Meningothelial	+	+++	–	–	–	+
13	Meningothelial	+	++	–	+	–	+
14	Meningothelial	–	+++	–	++	++	++
15	Meningothelial	+	+++	–	–	–	+
16	Meningothelial	+	–	+++	–	+	+
17	Meningothelial	+	++	–	++	++	–
18	Meningothelial	–	+++	–	+	–	+
19	Meningothelial	–	+++	–	+++	++	+
20	Meningothelial	+	++	–	++	–	+
21	Meningothelial	++	++	–	+	–	+
22	Fibrous	–	+++	–	+	–	++
23	Fibrous	–	+++	–	++	–	+
24	Fibrous	–	+++	–	++	++	+
25	Fibrous	–	+++	–	–	–	+
26	Transitional	–	+++	–	++	++	+
27	Transitional	++	–	–	–	–	+
28	Transitional	–	+++	–	–	–	+
29	Transitional	+	+++	–	–	++	–
30	Transitional	++	–	+++	+	–	–
31	Transitional	–	+++	–	++	+	+
32	Transitional	–	+++	–	++	++	+
33	Transitional	+	+++	–	++	+	+
34	Transitional	++	–	++	–	–	–
35	Transitional	+++	+	++	–	–	–
36	Psammomatous	++	–	–	+	–	+
37	Psammomatous	++	–	++	–	–	+
38	Psammomatous	+	++	–	++	+	+
39	Angiomatous	–	+++	–	+	++	–
40	Granular cell	–	–	–	–	–	–
41	Granular cell	–	–	–	–	–	+
42	Anaplastic	–	+++	–	–	+	+
43	Anaplastic	+	++	–	+	–	+
44	Anaplastic	–	+++	–	+++	+++	++
45	Anaplastic	–	+++	–	++	+	+
46	Anaplastic	–	+++	–	++	++	+
47	Anaplastic	–	+++	–	++	++	+
48	Anaplastic	+	+++	–	++	++	+
49	Anaplastic	+	+++	–	+	+++	++
50	Anaplastic	–	+++	–	++	+	+
51	Anaplastic	+++	–	+++	–	–	+
52	Anaplastic	–	+++	–	++	++	+
53	Anaplastic	++	+++	–	+	++	+
54	Anaplastic	+	+++	–	+++	++	+
55	Anaplastic	++	+	–	++	+++	++

^a–, 0% positive cells; +, 1 to 10% positive cells; ++, 11 to 50% positive cells; +++, 51 to 100% positive cells.

Figure 1. Normal arachnoidal tissue. Epithelial cadherin (E-cadherin) (a) and β-catenin (c) were detected in the cell membrane of normal arachnoidal cells, but neural cadherin (N-cadherin) (b) was not.

Figure 2. Meningothelial meningioma; dog No. 9. Immunopositive tumor cells for E-cadherin (a), N-cadherin (b), and β-catenin (c). Labeled streptavidin biotin method. Counterstained with Mayer’s hematoxylin.

Figure 3. Meningothelial meningioma; dog No. 19. Immunopositive tumor cells for E-cadherin (a), N-cadherin (b), and β-catenin (c). Inset: Nuclear expression of β-catenin. Labeled streptavidin biotin method. Counterstained with Mayer’s hematoxylin.

In canine meningiomas, regardless of the histologic features, the single expression of E-cadherin and N-cadherin was observed in 12 of 55 cases (22%) and 24 of 55 cases (44%), respectively (Figs. 2, 3), whereas their combined expression was observed in 17 cases (31%). The expression of β-catenin along the cell membrane was observed in only 10 cases (18%), with cytoplasmic expression in most cases. The translocation of β-catenin into the nucleus was detected in 33 cases (62%). DCX expression in tumor cells was detected in 26 cases (47%) and was more frequently observed in anaplastic meningiomas than in other types. DCX was preferentially expressed in the tumor cells at foci of invasion into the brain parenchyma (Fig. 4a), and DCX-positive cells were always detected at tumor borders when the histologic samples contained the marginal regions. Tumor cells invading the subpial nervous tissue tended to exhibit more intense immunoreactivity for DCX (Fig. 4b). The number of cells with Ki-67-positive nuclei varied from 0 to 18.1%. Figure 5 presents the results of box and whisker plot analyses for the minimum number of counted cells per antigen. Notably, granular cell meningiomas had a different immunohistochemical profile from other types of meningiomas, showing no immunoreactivity to any of the examined molecules.

Figure 4.

Anaplastic meningioma; dog No. 46. Doublecortin-immunopositive tumor cells are mainly observed at the tumor margin adjacent to the brain parenchyma (a). Tumor cells invading the subpial nervous tissue tend to exhibit more intense immunoreactivity for doublecortin (b). Labeled streptavidin biotin method. Counterstained with Mayer’s hematoxylin.

Figure 5.

Box plots of quantitative scores for each molecule examined in canine meningiomas. No significant differences are detected in the expression of epithelial cadherin (E-cadherin) (a) and neural cadherin (N-cadherin) (b) between benign and anaplastic meningiomas, but significant differences are found for nuclear β-catenin (c), doublecortin (DCX) (d), and Ki-67 (e). Values in box plots reflect the 25th, 50th (median), and 75th percentiles. The whiskers represent approximately 90 to 95% of the values. °, moderate outliers; *, extreme outliers.

Statistical Analyses

The relationship between the histologic type of canine meningioma and the expression of each molecule was examined. Significant differences were found between fibrous and psammomatous types for both E-cadherin and N-cadherin expression (both P = .030) and between psammomatous and anaplastic types for N-cadherin expression (P = .027). Granular cell meningiomas were differentially classified in the present study because they did not express any cell adhesion molecules. No significant differences were observed on the basis of tumor location. Significant differences in the expression of DCX, nuclear β-catenin, and Ki-67 were found between benign and anaplastic meningiomas (Table 4 ). However, no significant differences were found between benign and anaplastic meningiomas for E-cadherin or N-cadherin expression.

Table 4.

Relationship Between Immunohistochemical Scores of Examined Molecules and Histologic Malignancy in Canine Meningiomas

No. Cases Given Each Score^a
Molecules	–	+	++	+++	P
Epithelial cadherin
Benign	17	11	8	3
Anaplastic	7	4	2	1	.958
Neural cadherin
Benign	10	0	7	22
Anaplastic	1	0	2	11	.278
Nuclear β-catenin
Benign	19	8	12	0
Anaplastic	2	3	7	2	.021
Doublecortin
Benign	23	6	10	0
Anaplastic	2	4	5	3	.003
Ki-67
Benign	16	22	1	0
Anaplastic	0	11	3	0	.003

^a–, 0% positive cells; +, 1 to 10% positive cells; ++, 11 to 50% positive cells; +++, 51 to 100% positive cells.

Table 5 presents results of linear discriminant analyses between immunohistochemical parameters and their discriminant values. In terms of their expression, nuclear β-catenin, DCX, and Ki-67 appeared to be reliable single predictive molecules, and accuracy scores for predicting the histologic grade were 73.6%, 71.7%, and 77.4%, respectively. Adding the results of E-cadherin and N-cadherin to that of nuclear β-catenin did not increase the predictive value of histologic grade (73.6%), whereas combining all results increased the predictive accuracy to 84.9% .

Table 5.

Linear Discriminant Analyses Between Immunohistochemical Parameters and Their Discriminant Values

Molecules	Probability of Histologic Grade (%)^a
Epithelial cadherin	43.4
Neural cadherin	52.8
Nuclear β-catenin	73.6
Doublecortin	71.7
Ki-67	77.4
Combination of epithelial cadherin, neural cadherin, and nuclear β-catenin	73.6
Combination of all molecules examined	84.9

^a The number represents the diagnostic concordance rate of histologic grade.

Table 6 summarizes the correlation of the percentage of positive cells for cadherin molecules in relation to all other molecules examined. A significant negative correlation was found between E-cadherin and N-cadherin and between E-cadherin and nuclear β-catenin, as shown by Pearson correlation coefficients. A significant positive correlation was found between N-cadherin and nuclear β-catenin, between N-cadherin and DCX, and between N-cadherin and Ki-67.

Table 6.

Correlations Between Percentages of Immunopositive Cells for Cadherin Molecules and Other Molecules Examined in the Present Study

	Neural Cadherin	Nuclear β-Catenin	Doublecortin	Ki-67
Epithelial cadherin
Pearson r	–.665	–.269	–.229	–.244
P (2-tailed)	.000	.047	.092	.076
Neural cadherin
Pearson r	—	.395	.370	.292
P (2-tailed)	—	.003	.005	.032

Discussion

There are several novel findings concerning the expression of molecules involved in cell adhesion, cell invasion, and cell proliferation in these canine meningioma cases. The percentages of nuclear β-catenin-positive, DCX-positive, and Ki-67-positive cells were significantly higher in anaplastic meningiomas than in benign meningiomas. By linear discriminant analyses, these 3 molecules appeared to be of value for predicting the histologic grade of canine meningiomas in terms of single-factor analyses. Although combining the results from the cell adhesion molecules (E-cadherin, N-cadherin, and nuclear β-catenin) did not increase the predictive value of the histologic grade, the use of all 5 parameters was the most reliable predictor. These results indicate that the expression patterns of these molecules can be used for assessing the anaplastic nature of canine meningiomas. However, granular cell meningiomas showed a distinct cell adhesion molecule expression pattern, and so this tumor is considered to be very different from other types of meningioma. With respect to subtype-related differences, fibrous and psammomatous types had statistically significant differences in the expression of both cadherin molecules, whereas psammomatous and anaplastic types had statistically significant differences in N-cadherin. No significant differences were observed in respect to tumor locations. However, findings must be viewed in relation to the small sample size for each histologic subtype and in each site.

The expression levels of E-cadherin and N-cadherin in canine meningiomas were not significantly different between benign and anaplastic cases, which is similar to findings in human meningiomas.^6,7,21 However, a significant negative correlation between E-cadherin and N-cadherin expressions was noted. These results suggest a complementary expression in canine meningiomas regardless of the histologic type. N-cadherin expression in tumor cells is regarded as a sign of invasive behavior.^1,9,12,27 Like E-cadherin, the N-cadherin molecule generally interacts with other N-cadherin molecules expressed in nearby cells in a homophilic manner.²⁶ Consequently, N-cadherin expressed on a tumor cell may increase its affinity for stromal cells, such as fibroblasts that normally express N-cadherin. This association seems to help tumor cells invade into stroma.²⁶ The shift of the expression from E-cadherin to N-cadherin in canine meningiomas may facilitate invasion into the surrounding nervous tissue that normally expresses N-cadherin.

The cytoplasmic tail of the E-cadherin molecule binds to β-catenin to form the cadherin–actin filament complex.² Such complex formation is necessary for functional cell–cell adhesion.² β-Catenin also plays a decisive role in the Wnt signaling pathway, which has been implicated in the pathogenesis of various human cancers.¹³ The activation of the Wnt pathway leads to the stabilization of β-catenin and its translocation into the nucleus. Such nuclear accumulation of β-catenin is thought to induce the transcription of some oncogenes through the formation of a complex containing β-catenin and a transcription factor such as Tcf/Lef1.²⁰ The present study showed the more frequent nuclear localization of β-catenin in anaplastic meningiomas compared to their benign counterparts, as well as a significant positive correlation between β-catenin nuclear localization and the expression of N-cadherin. Therefore, increased expression of N-cadherin and nuclear localization of β-catenin may be involved in the progression to more anaplastic canine meningiomas.

Tumor cell invasion into the surrounding nervous tissue, which is a hallmark of anaplasia in the WHO criteria,¹⁰ was present in most canine anaplastic meningiomas examined. DCX expression was significantly increased in anaplastic meningiomas compared to benign meningiomas. DCX expression was particularly observed at the tumor margin adjacent to normal nervous tissue, and the tumor cells invading to the subpial nervous tissue tended to exhibit more intense immunoreactivity for DCX. Whereas DCX expression could be an accurate predictor of anaplastic meningiomas, its utility may be limited when the border area between the tumor and surrounding nervous tissue is not obtained surgically. In addition, DCX expression had a significant positive correlation with N-cadherin. In tumors, invasive behaviors are a sign of progression.^1,9,12,27 Therefore, the mechanism by which canine meningiomas invade into the surrounding nervous tissue might be related to the expression of DCX and N-cadherin.

The present study revealed the expression of a series of cell adhesion and related molecules in canine meningiomas by immunohistochemistry. All these findings indicate that the expression of N-cadherin and DCX and the nuclear translocation of β-catenin are closely associated with progression of canine meningiomas.

Footnotes

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

The authors declared that they received no financial support for their research and/or authorship of this article.

References

Akat

Bleck

Lee

Haselmann-Weiss

Kartenbeck

: Characterization of a novel type of adherens junction in meningiomas and the derived cell line HBL-52. Cell Tissue Res 331:401–412, 2008.

Brunner

Romeike

Jung

Comtesse

Meese

: Altered expression of beta-catenin/E-cadherin in meningiomas. Histopathology 49:178–187, 2006.

Daou

Smith

Litofsky

Hsieh

Ross

: Doublecortin is preferentially expressed in invasive human brain tumors. Acta Neuropathol 5:472–480, 2005.

Dickinson

Surace

Cambell

Higgins

Leutenegger

Bollen

LeCouteur

Gutmann

: Expression of the tumoreuppressor genes NF2, 4.1B, and TSLC1 in canine meningiomas. Vet Pathol 46:884–892, 2009.

Dumanski

Rouleau

Nordenskjold

Collins

: Molecular genetic analysis of chromosome 22 in 81 cases of meningioma. Cancer Res 50:5863–5867, 1990.

Figarella-Branger

Pellissier

Bouillot

Bianco

Mayan

Grisoli

Rougon

: Expression of neural cell-adhesion molecule isoforms and epithelial cadherin adhesion molecules in 47 human meningiomas: correlation with clinical and morphological data. Mod Pathol 7:752–761, 1994.

Figarella-Branger

Roche

Daniel

Dufour

Bianco

Pellissier

: Cell-adhesion molecules in human meningiomas: correlation with clinical and morphological data. Neuropathol Appl Neurobiol 23:113–122, 1997.

Hatta

Takeichi

: Expression of N-cadherin adhesion molecules associated with early morphogenetic events in chick development. Nature 320:447–449, 1986.

Hazan

Phillips

Qiao

Norton

Aaronson

: Exogenous expression of N-cadherin in breast cancer cells induces cell migration, invasion, and metastasis. J Cell Biol 148:779–790, 2000.

10.

Koestner

Bilzer

Fatzer

Schulman

Summers

Van Winkle

: Histological Classification of Tumors of the Nervous System of Domestic Animals, 2nd ser., vol. 5, pp. 17–33. Armed Forces Institute of Pathology, Washington, DC, 1999.

11.

Lallemand

Curto

Saotome

Giovannini

McClatchey

: NF2 deficiency promotes tumourigenesis and metastasis by destabilising adherens junctions. Genes Dev 17:1090–1100, 2003.

12.

Satyamoorthy

Herlyn

: N-cadherin-mediated intercellular interactions promote survival and migration of melanoma cells. Cancer Res 61:3819–3825, 2001.

13.

Lustig

Behrens

: The Wnt signaling pathway and its role in tumour development. J Cancer Res Clin Oncol 129:199–221, 2003.

14.

Mandara

Pavone

Mandrioli

Bettini

Falzone

Baroni

: Matrix metalloproteinase-2 and matrix metalloproteinase-9 expression in canine and feline meningioma. Vet Pathol 46:836–845, 2009.

15.

Maruyama

Ochiai

Akimoto

Nakamura

Baba

Moriya

Hirohashi

: Cytoplasmic beta-catenin accumulation as a predictor of hematogenous metastasis in human colorectal cancer. Oncology 59:302–309, 2000.

16.

Meese

Blin

Zang

: Loss of heterosygosity and the origin of meningioma. Hum Genet 77:349–351, 1987.

17.

Montoliu

Anor

Vidal

Pumarola

: Histological and immunohistochemical study of 30 cases of canine meningioma. J Comp Pathol 135:200–207, 2006.

18.

Panagopoulos

Lancelotti

CLP

Veiga

JCE

de Aguiar

PHP

Colquhoun

: Expression of cell adhesion proteins and proteins related to angiogenesis and fatty acid metabolism in benign, atypical, and anaplastic meningiomas. J Neurooncol 89:73–87, 2008.

19.

Ruttledge

Sarrazin

Rangaratnam

Phelan

Twist

Merel

Delattre

Thomas

Nordenskjöld

Collins

Dumanski

Rouleau

: Evidence for the complete inactivation of the NF2 gene in the majority of sporadic meningiomas. Nat Genet 6:180–184, 1994.

20.

Schlosshauer

Ellenson

Soslow

: Beta-catenin and E-cadherin expression patterns in high-grade endometrial carcinoma are associated with histological subtype. Mod Pathol 15: 1032–1037, 2002.

21.

Shimada

Ishizawa

Hirose

: Expression of E-cadherin and catenins in meningioma: ubiquitous expression and its irrelevance to malignancy. Pathol Int 55:1–7, 2005.

22.

Snyder

Shofer

Van Winkle

Massicotte

: Canine intracranial primary neoplasia: 173 cases (1986–2003). J Vet Intern Med 20:669–675, 2006.

23.

Summers

Cummings

de Lahunta

: Tumors of the central nervous system. In: Veterinary Neuropathology, pp. 351–401. Mosby-Year Book, Inc, St Louis, MO, 1995.

24.

Utsuki

Oka

Sato

Kawano

Tsuchiya

Kobayashi

Fujii

: Invasive meningioma is associated with a low expression of E-cadherin and beta-catenin. Clin Neuropathol 24:8–12, 2005.

25.

Van Aken

De Wever

Correia da Rocha

Mareel

: Defective E-cadherin/catenin complexes in human cancer. Virchows Arch 439:725–751, 2001.

26.

Weinberg

: Moving out: invasion and metastasis. In: The Biology of Cancer, pp. 587–654. Garland Science, New York, NY, 2007.

27.

Wheelock

Johnson

: Cadherins as modulators of cellular phenotype. Annu Rev Cell Dev Biol 19:207–235, 2003.

28.

Whittle

Smith

Navoo

Collie

: Meningiomas. Lancet 363:1535–1543, 2004.