immunohistochemical detection of arteriolar ...

Acta Veterinaria Hungarica 58 (4), pp. 423–430 (2010) DOI: 10.1556/AVet.58.2010.4.3

IMMUNOHISTOCHEMICAL DETECTION OF ARTERIOLAR HYPERPLASIA IN CANINE LIVER BIOPSY SAMPLES USING THE CLAUDIN-5 ANTIBODY Csaba JAKAB1*, Miklós RUSVAI1, Péter GÁLFI2, Ágnes SZABÁRA1, Zoltán SZABÓ4 and Janina KULKA3 1

2

Department of Pathology and Forensic Veterinary Medicine and Department of Pharmacology and Toxicology, István u. 2, H-1078 Budapest, Hungary; 3 2nd Department of Pathology, Semmelweis University, Budapest, Hungary; 4 C. J. Hall Veterinary Surgeons – Exotic Centre, London, United Kingdom (Received 26 January 2010; accepted 21 June 2010)

Claudins are key tight junctional proteins between adjacent epithelial, mesothelial or endothelial cells, which are responsible for the permeability of the paracellular space. This paper describes that the endothelial cells of normal hepatic arterioles, portal venules and portal lymphatics as well as the endothelium of sinusoids from dogs show strong membranous claudin-5 cross-reactivity. In 25 liver biopsy samples taken from dogs with portal vein hypoperfusion, an increased number of arterioles was detected in the portal areas (PAs) by the use of humanised anti-claudin-5 antibody. The increased number of hyperplastic hepatic arterioles per PA was 5–6, 8–12 and 15–20 in the case of small, medium-sized and large PAs, respectively. It is suggested that the claudin-5 marker can improve the detection of hepatic arteriolar proliferation in the PAs of liver samples. Key words: Arteriolar hyperplasia, dog, liver, biopsy, claudin-5, immunohistochemistry

Portal areas (PAs, also called portal triads) are located in the corners of liver lobules. Each PA contains three tubular structures such as a branch of the bile duct (interlobular bile duct), a branch of the portal vein (portal venule) and a branch of the hepatic artery (hepatic arteriole). A typical PA also contains two additional structures, a nerve fibre and a lymphatic channel, but these are inapparent in routine histological specimens. These portal structures are embedded in the connective tissue (Samuelson, 2007). *

Corresponding author; E-mail: [email protected]; Phone: 0036 (1) 478-4181; Fax: 0036 (1) 478-4284 0236-6290/$ 20.00 © 2010 Akadémiai Kiadó, Budapest

424

JAKAB et al.

The main histological hallmark of different liver diseases with portal vein hypoperfusion or inadequate portal vein flow (such as primary hypoplasia of the portal vein, congenital portosystemic shunt, intrahepatic arteriovenous fistula, and portal vein obstruction) is arteriolar hyperplasia, or arteriolar proliferation in the PAs. In some chronic liver diseases associated with portal hypertension such as liver cirrhosis and hepatic fibrosis, arteriolar hyperplasia can be seen in the PAs, too (Rothuizen et al., 2006). Materials and methods Histopathology A total of 25 liver biopsy samples taken from dogs and submitted for diagnosis to the Department of Pathology and Forensic Veterinary Medicine, Faculty of Veterinary Science, Szent István University between 2002 and 2010. The histopathological diagnosis was made according to the WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease (Rothuizen et al., 2006). Five intact liver samples collected during necropsy from succumbed dogs were also examined. The liver biopsy samples were collected from 9 females and 16 males with an average age of 2.7 years (range: 1–6 years). The necropsy samples were collected from 3 females and 2 males with an average age of 8.4 years (range: 6–10 years). Immunohistochemistry All samples were fixed in 8% neutral buffered (in PBS, pH 7.0) formaldehyde solution for 24 h at room temperature, dehydrated in a series of ethanol and xylene and embedded in paraffin. The 3–4 μm thick sections were cut, and routinely stained with haematoxylin and eosin. The slides for the claudin-5 immunohistochemical reaction were deparaffinised in xylene and graded ethanol. The deparaffinised sections were treated with primary antibody claudin-5 (diluted 1:100, mouse monoclonal, Zymed Inc., San Francisco, CA, USA – #18-7364) at room temperature for 60 min after treatment with an appropriate antigen retrieval solution (Target Retrieval Solution, DAKO, Glostrup, Denmark, pH 6; treated in microwave oven for 30 min). Immunohistochemical staining was performed using the streptavidin-peroxidase procedure. Antigen-bound primary antibody was detected using a standard avidin-biotin immunoperoxidase complex (DAKO LSAB2 Kit). The chromogen substrate was aminoethylcarbazole (AEC). Mayer’s haemalaun was used for counter-staining. Negative control was performed by the omission of the primary antibody. The external positive control was the endothelial cell component of tumour-induced microvessels of mammary gland carcinomas (Jakab et al., 2008). The staining pattern was membrane/cytoplasmic. The internal positive controls were the endothelial cells of liver sinusoids. Acta Veterinaria Hungarica 58, 2010

CLAUDIN-5 POSITIVITY OF ARTERIOLAR HYPERPLASIA IN THE CANINE LIVER

425

Histomorphometry During the histopathological examination of intact liver necropsy samples, we analysed and measured some of the histoquantitative parameters of the PAs of the liver, such as the biggest longitudinal diameter of the PA, the number of portal hepatic arterioles, the number of portal venules and the number of portal lymphatics, with the help of a microscopic oculometer.

Results Based on the histoquantitative parameters of the PAs of intact liver samples, three groups were established with regard to the biggest longitudinal diameter of the PAs: (1) small, (2) medium and (3) large PAs. The biggest longitudinal diameter was 90–300 μm in the first group (Fig. 1A), 300–800 μm in the second group (Fig. 1C) and more than 800 μm in the third group. The endothelial cells of the hepatic arterioles, portal venules, portal lymphatics and the endothelium of the sinusoids showed a strong membranous claudin-5 positivity (internal positive control) (Figs 1B and 1D). The biliary epithelial cells, the portal fibroblasts, and the liver cells were negative for claudin-5 protein (internal negative controls). The strong claudin-5 positivity of the endothelial cells helped us to count the number of the different vascular channels in the PAs. The histoquantitative parameters of the small intact PAs included 1 hepatic arteriole/PA, 1 portal venule/ PA and 2–3 portal lymphatic channels/PA. The medium intact PAs included 2–3 hepatic arterioles/PA, 1 portal venule/PA and 2–6 portal lymphatic channels/PA; whereas the large intact PAs included 4–5 hepatic arterioles/PA, 1 portal venule/ PA and 8–11 portal lymphatic channels/PA (Table 1). Pathological proliferation of the hepatic portal arterioles was detected in all biopsy samples (Figs 2A and 2C). The endothelial cells of the hyperplastic hepatic arterioles showed strong membranous claudin-5 positivity, and this helped us to count the increased number of arterioles in the different PAs (Figs 2B and 2D). The number of proliferating hepatic arterioles was 5–6/PA in the case of small PAs, 8–12/PA in the case of medium PAs (Figs 3 and 4) and 15–20 in the case of large PAs. The hepatic arterioles were more tortuous and the endothelial cells were more hypertrophied than in the normal arterioles. The absence or decreased diameter of the portal venule was detected in the biopsy samples.

Discussion Claudins are key components of epithelial, mesothelial and endothelial tight junction intercellular structures which act as a barrier to the paracellular flux of water and solutes and the transmigration of other cells (Tsukita et al., 2001). Acta Veterinaria Hungarica 58, 2010

426

Acta Veterinaria Hungarica 58, 2010

JAKAB et al.

Fig. 1. A. Histopathological picture of the portal triad: hepatic arteriole (black arrow), portal venule (black asterisk), bile duct (blue arrow) and lymphatic channel (green arrow) of the small portal area in the normal liver. Haematoxylin and eosin (HE), × 200; B. The endothelial cells of the hepatic arteriole (black arrow), portal venule (black asterisk) and lymphatic channel (green arrow) showed a strong linear, membranous immunoreactivity for claudin-5 endothel-specific tight junction protein (brown discolouration inside the vascular and lymphatic channels). The biliary epithelial cells were negative for this claudin. Immunohistochemistry (IHC), × 200; C. Histopathological picture of the small portal area in a normal liver sample with three hepatic arterioles (black arrows). HE, × 100; D. The endothelial cells of the hepatic arterioles (black arrows), portal venule (black asterisk) and lymphatic channels (green arrow) of the medium portal area were positive but the biliary epithelial cells and the portal fibroblasts were negative for claudin-5. The periportal endothelial cells of the sinusoids showed strong membranous claudin-5 positivity but the liver cells were negative for this claudin. IHC, × 100

CLAUDIN-5 POSITIVITY OF ARTERIOLAR HYPERPLASIA IN THE CANINE LIVER 427


Fig. 2. A. Histopathological picture of portal hepatic arteriole proliferation of the small portal area in a liver biopsy sample. HE, × 200; B. Five claudin-5-positive hyperplastic arterioles in the small portal area with lymphatic channels, without portal venule. IHC, × 200; C. Histopathological picture of the longitudinal and cross sections of the hyperplastic hepatic arterioles of the small portal area in a liver biopsy sample. HE, × 200; D. The proliferating hepatic arterioles were more tortuous and the claudin-5-positive endothelial cells were more hypertrophied than the normal arterioles in liver samples. IHC, × 200

428

JAKAB et al.

The analysis of claudin gene expression using in silico and RT-PCR approaches has demonstrated differences in the expression patterns among the various claudin molecules in human non-neoplastic and neoplastic tissues (Hewitt et al., 2006). A previous veterinary study tested humanised anti-claudin-5 antibodies on different normal and neoplastic soft tissues. The investigations were mainly carried out on paraffin-embedded sections of formalin-fixed tissues (Jakab et al., 2008). Ridyard et al. (2007) detected an increased expression of claudin-2 in the colonic epithelium of dogs with idiopathic lymphocytic-plasmacytic colitis (Ridyard et al., 2007). Table 1 Histoquantitative parameters of the portal area in intact liver samples Histomorphometric parameters of the portal areas

Biggest longitudinal diameter Hepatic arteriole (number) Portal venule (number) Portal lymphatic (number)

Large portal area

Medium portal area

Small portal area

> 800 μm 4–5 1 8–11

300–800 μm 2–3 1 2–6

90–300 μm 1 1 2–3

Fig. 3. Eight claudin-5-positive proliferating hepatic arterioles of the medium portal area in a liver biopsy sample with claudin-5-positive sinusoid endothelial cells. IHC, × 100 Acta Veterinaria Hungarica 58, 2010

CLAUDIN-5 POSITIVITY OF ARTERIOLAR HYPERPLASIA IN THE CANINE LIVER

429

Fig. 4. Higher magnification of the claudin-5-positive hyperplastic hepatic arterioles of the medium portal area in a liver biopsy sample. IHC, × 400

The claudin-5 protein is an endothel-specific claudin (Tsukita et al., 2001). The endothelial barrier function requires the adhesive activity of VE-cadherin and claudin-5, two key components of adherent and tight endothelial junctions (Gavard and Gutkind, 2008). A previous veterinary study, using 8% buffered formalin-fixed and paraffin-embedded tissue samples and humanised anticlaudin-5 antibody, has demonstrated that the claudin-5 tight junction protein is an excellent immunohistochemical marker for examining the endothelium of tumour-induced primary (Jakab et al., 2008). A diagnostically significant histopathological hallmark of different liver diseases characterised by portal vein hypoperfusion is the proliferation of the portal arterioles. In the present study, the claudin-5 (cross-reactive humanised) marker was used for the detection of hyperplastic hepatic arterioles in the liver biopsy samples. To our knowledge, this is the first study that has examined the expression of claudin-5 in the proliferating hepatic arterioles of liver samples. The study demonstrated that the endothelial cells of the hepatic arterioles, portal venules and lymphatic channels in the PAs showed a strong immunoreactivity for claudin-5. Based on the histoquantitative parameters (biggest longitudinal diameter of the PA and the number of hepatic arterioles, portal venules and lymActa Veterinaria Hungarica 58, 2010

430

JAKAB et al.

phatic channels per PA), three histological groups were established in the case of intact PAs. By the use of claudin-5, we could exactly detect the increased number of arterioles in the PAs and differentiate between claudin-5 positive arterioles and claudin-5 negative biliary epithelial cells. During the immunohistochemical analysis of liver tissue, the suitable internal positive controls were the endothelial cells of the sinusoids and the negative controls were the portal biliary epithelial cells and portal fibroblasts. Acknowledgements We would like to thank Renáta Pop and Magdolna Pekár for their assistance with the immunohistochemical reactions.

References Burek, M., Arias-Loza, P. A., Roewer, N. and Förster, C. Y. (2010): Claudin-5 as a novel estrogen target in vascular endothelium. Arterioscler. Thromb. Vasc. Biol. 30, 298–304. Gavard, J. and Gutkind, J. S. (2008): VE-cadherin and claudin-5: it takes two to tango. Nat. Cell Biol. 10, 883–885. Hewitt, K. J., Agarwal, R. and Morin, P. J. (2006): The claudin gene family: expression in normal and neoplastic tissues. BMC Cancer, 6, 1–8. Jakab, Cs., Halász, J., Kiss, A., Schaff, Zs., Szász, A. M., Rusvai, M., Abonyi Tóth, Zs. and Kulka, J. (2008): Evaluation of microvessel density (MVD) in canine mammary tumours by quantitative immunohistochemistry of the claudin-5 molecule. Acta Vet. Hung. 56, 495–510. Ridyard, A. E., Brown, J. K., Rhind, S. M., Else, R. W., Simpson, J. W. and Miller, H. R. P. (2007): Apical junction complex protein expression in the canine colon: differential expression of claudin-2 in the colonic mucosa in dogs with idiopathic colitis. J. Histochem. Cytochem. 55, 1049–1058. Rothuizen, J., Bunch, S. E., Charles, J. A., Cullen, J. M., Desmet, V. J., Szatmári, V., Twedt, D. C., van den Ingh, T. S. G. A. M., van Winkle, T. J. and Washabau, R. J. (2006): WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease. Saunders Elsevier, St. Louis, Missouri, USA. pp. 45–60. Samuelson, D. A. (2007): Textbook of Veterinary Histology. Saunders Elsevier, St. Louis, Missouri, USA. pp. 358–366. Tsukita, S., Furuse, M. and Itoh, M. (2001): Multifunctional strands in tight junctions. Nat. Rev. Mol. Cell Biol. 2, 285–293.
