key: cord-0008710-t15l124f
authors: Stange, Thoralf; Kettmann, Ursula; Holzhausen, Hans-Jürgen
title: Immunoelectron microscopic single and double labelling of aminopeptidase N (CD 13) and dipeptidyl peptidase IV (CD 26)
date: 2012-02-20
journal: Acta Histochem
DOI: 10.1016/s0065-1281(96)80025-0
sha: 3c2631b105784921d63954a02001fa655010095b
doc_id: 8710
cord_uid: t15l124f

Using ultrathin cryosections and immunogold labelling, aminopeptidase N (CD 13) and dipeptidyl peptidase IV (CD 26) were localized on the luminal side of the brush border membrane of proximal tubular cells in human kidney as well as of enterocytes from rat small intestine. Furthermore, both enzymes could be detected on the cell surface of human T lymphocytes and especially aminopeptidase N on human synovial fibroblasts. Gold labelled vesicular structures were also found in the cytoplasm in the apical part of renal proximal tubular cells and synovial fibroblasts. In human kidney the colocalization of the two membrane antigens was possible by using several double labelling methods.

Aminopeptidase N (microsomal alanyl aminopeptidase, mAAP, EC 3.4.11.2, CD 13) and dipeptidyl peptidase IV (DPP IV, EC 3.4.14.5, CD 26) represent membrane exopeptidases, whose biochemical properties and expression sites were intensely investigated. The two membrane proteases are primarily found in the microvillar membrane of renal proximal tubular cells and small intestinal enterocytes of mammals (McDonald and Barrett, 1986 ), but they are also present in the plasma membranes of various tissues such as liver, parotid and submandibular acinar cells, placenta and blood vessels (Gossrau, 1985) . DPP VI and APN can be detected on the cell surface of human synovial fibroblasts (Bathon et aI., 1992) and especially APN on synaptic membranes of the central nervous system (Gros et aI., 1985) . In the immune system DPP VI occurs as a specific marker for T lymphocytes (Mentlein et aI., 1984) , APN is localized on monocytes, neutrophil granulocytes and in small amounts also on T lymphocytes (Ashmun and Look, 1990; Kunz et aI., 1993) . Riemann et al. (1993) found an expression of aminopeptidase N on synovial fluid T cells from patients with different forms of joint effusions. Correspondence to: H.-l. Holzhausen Both enzymes participate in the final stage of peptide hydrolysis in the renal proximal tubules and small intestine (McDonald and Barrett, 1986) and modify the effect of bioactive peptides like enkephalins (APN), substance P and casomorphins (DPP IV) (De la Baume et aI., 1983; Gros et aI., 1985; Nausch et aI., 1990) . The enzymes take also part in the activation and proliferation of lymphocytes (Stein et aI., 1989) . APN serves as a receptor for coronaviruses, which are RNA viruses that cause respiratory and gastrointestinal diseases (Delmas et ai., 1992) . DPP IV has co receptor function for HIV viruses (Callebaut et aI., 1993) . Both peptidases are involved in the cell adhesion of carcinoma cells in metastatic processes (Johnson et aI., 1993; Saiki et aI., 1993) .

During the last few years several cell surface antigens were differentiated as membrane peptidases. APN corresponds to the cell surface antigen CD 13 (Look et aI., 1989) and OPP IV to CO 26 (Ulmer et aI., 1990) .

Nevertheless many physiological functions of these peptidases are still unknown. In order to receive further informations about the physiological role of membrane proteases it is necessary, to find out their precise cellular location (Gossrau, 1985) .

Compared with the light microscopic visualisation of proteases, the electron microscopic demonstration methods for these enzymes have been less frequently applied until now. This limited application of protease ultracytochemistry is explained by the comparatively strong inhibition of the proteases by aldehydes, the co-reaction of proteases different from those under investigation, as well as the reduced stability and moderate electron density of the osmicated final reaction product (Gossrau, 1993) .

The detection of APN and DPP IV with immunoelectron microscopical methods has similar problems. Because of their position on the external cell surface it is difficult to find out the extact localization of ectoenzymes such as APN and DPP IV with immunoultrastructural methods (Kettmann et aI., 1992 b) . The problem with the preembedding method is the difficulty of penetration of markers and antibodies into the cells. Hence intracellular antigens are often not detectable. To improve the limited penetration pre-treatment with detergents is generally necessary, but a too high concentration of detergents can cause membrane damage, and may affect the distribution of the antigen associated with the cell membrane (Mizuno, 1993) . In the postembedding method, only those antigens exposed at the surface of resin-embedded sections are immunolabelled. Some antigens, especially membrane proteins, are very labile to fixation, dehydratation, embedding and polymerisation (Mizuno, t 993) . Glycosylated enzymes with large extracellular domains bound to membranes via a short hydrophobic peptide such as APN or OPP IV are not labelled, if the tissue is embedded in resins and the results are independent of the mode of dehydration and type of resin. Conventional resin embedding as well as low temperature embedding and freeze-substitution failed to demonstrate the localization of APN by immunogold labelling (Kettmann et aI., 1992b) .

Using cryoultramicrotomy in combination with immunogold labelling APN and OPP IV were detectable both on the luminal side of the brush border membrane of proximal tubular cells in rat kidney and intracellularly as potential transport vesicles operating between the Golgi complex and surface membranes (Kettmann et aI., 1992 a, b) . This method represents a useful compromise between preservation of antigenicity and morphology (Slater, t 993) . Immunogold labelling in combination with cryoultramicrotomy has been proven to be a useful method for the visualisation and localization of cell surface and intracellular proteins (Mizuno, 1993) . Because of their abundant expression on the microvillus membrane of proximal tubular cells of the kidney and small intestinal enterocytes, these cells are especially suitable for methodological investigations to localize and also colocalize APN and OPP IV, and were therefore used for the present investigation. Since double labelling allows to study the distribution of two antigens within one cryosection and colloidal gold markers are very suitable for double labelling methods, because gold particles of different size are available, we also used this technique to receive further information on APN and DPP IV at the ultrastructural level.

Antisera. Monoclonal antibodies (mouse) (mAB) against CO 13 (clone Sl.101) and CD 26 (clone BA5) were purchased from Oianova (Hamburg). Polyclonal mono specific rabbit-anti-rat-(pRAR)APN antiserum (used by Kettmann et aI., 1992b) and polyclonal mono specific rabbit-anti-rat-(pRAR)OPP IV antiserum (gift of Dr. A. Heider, Institute of Pathological Biochemistry, Martin Luther University of Halle, Germany) were used for labelling the rat small intestine. For the double labelling investigations polyclonal rabbit-anti-pig-(pRAP)OPP IV antiserum could be used because of its cross-reactivity with human OPP IV. This specificity was proved by 2-dimensional immunoelectrophoresis according to Laurell (1965) and by Western blot described by Kettmann et al. (1992b) .

Electron microscopy embedding technique. Tissues. Small pieces of fresh human kidney cortex (operation specimens) and sections of small intestine from rats after sacrifice in ether anesthesia were washed with ice-cold PBS (pH 7.4) and fixed in 2070 formaldehyde and 0.02% glutaraldehyde in PBS for 30 min at room temperature, followed by 3 washes with PBS. Cells. After centrifugation and washing twice in PBS cultured human synovial fibroblasts and Ilr2-stimulated T lymphocytes (gift of Dr. O. Riemann, Institute of Medical Immunology, Martin Luther University of Halle, Germany) were fixed with 2070 formaldehyde and 0.02% glutaraldehyde in PBS for 20 min, followed by 2 washes with PBS and subsequently embedded in 7% gelatine (Si&ma, St. Louis, USA) in PBS. Afterwards all fixed samples were incubated with 1.15 M sucrose and 10% polyvinylpyrrolidone (PVP K15, Fluka, Buchs, Switzerland) in PBS for approximately 2 days at 5 0c. Further processing and cryosectioning were done following the methods of Tokuyasu (1973 Tokuyasu ( , 1989 . Ultrathin cryosectioning was performed with a Reichert-lung FC 4E cryouitramicrotome.

Immunolabelling. The cryosections were transferred to formvar-coated nickel grids (Bio Cell Plano, Marburg), washed with PBS, afterwards with PBS containing 50 mM glyCine and finally with 0.5% albumin (Sigma, St. Louis, USA) and 0.2% gelatine (Sigma, St. Louis, USA) in PBS. The sections were incubated in a moist chamber with the specific antiserum (dilution 1 : 100) at 5 °C overnight. The reaction with gold-labelled secondary antibody or protein A was done for 1 h at room temperature. Goat-anti-mouse IgG (GAM, Aurion Immuno Gold, Wageningen, The Netherlands) was labelled with 6 nm and 12 nm gold particles, protein A (Bio Cell Plano, Marburg) with 10 nm gold particles. After washing in PBS and tri-distilled water the sections were embedded in 1.1 % tylose (Fluka, Buchs, Switzerland) containing 0.5% uranyl acetate. Controls were done by replacing the specific serum with the respective preimmunserum. Double labelling. In the first technique APN and OPP IV were successively labelled with antibodies raised in the same species (Geuze et aI., 1981) . The first antigen was marked according to the standard labelling procedure with GAM 6 nm gold. The free binding places of the primary antibody were saturated by incubation with unlabelled goat-anti-mouse IgG (lmmunpnip., Berlin) 0.01070 in PBG for 15 min. The other antigen was labelled with a specific antibody and GAM-conjugated gold particles of different diameter (15 nm, Aurion Immuno Gold, Wageningen, The Netherlands) as used for the first incubation. In the second technique the two antigens were simultaneously labelled with antibodies from different origin (Tapia et aI., 1983). The procedure was carried out as common labelling with the mixed primary (pRAP-OPP IV antiserum and mAB against GO 13) and secondary antibodies [GAM-6 nm gold and goat-anti-rabbit IgG (GAR) 10 nm gold (Bio Cell Plano, Marburg)]. To control successfully double labelling one of the two specific antisera was replaced by the respective preimmunserum. Additionally, the labelling sequence of APN and OPP IV was changed. Furthermore, the double labelling results were compared with the corresponding single labelling.

All figures show good preservation of APN and DPP IV antigenicity as well as ultrastructural morphology. APN and DPP IV were found on the luminal side of the brush border membranes of human kidney proximal tubular cells (Fig. 1 a, b) and enterocytes of rat small intestine (Fig. 5 a, b) . A prominent, regular and diffuse labelling of both peptidases was found on these tissues. Control experiments (data not shown) with the respective preimmunserum did not show non-specific labelling. Fig. 4 . Simultaneous double labelling of ultrathin cryosections of brush border from human kidney proximal tubular cells. a. with CD 13-mAB/GAM-6 nm gold and pRAP-DPP IV antiserum/ GAR-10 nm gold, b. control labelling using preimmunserum/GAR-10 nm gold and CD 13-mAB/GAM-6 nm gold, c. control labelling using preimmunserum/GAM-6 nm gold and pRAP-DPP IV antiserum/GAR-tO nm gold. x 80.000.

The double labelling experiments showed a close spatial relationship between APN and DPP IV (Figs. 2a, 3 a, 4a ). Using the successive or simultaneous method no substantial differences in double labelling were seen. No cross-reactivity occurred in double labelling because control experiments led to identical results (Figs. 2 b, c, 3 b, c, 4 b, c) . Labelling density for APN appeared to be higher than for DPP IV (see also Kenny and Maroux, 1982) . In Fig. 6 (a, b) and Fig. 7 (a, b) the detection of APN (synovial fibroblasts, T lymphocytes) and DPP IV (T lymphocytes) on cultured cells are visible. Furthermore, goldlabelled vesicle-like structures were found in the luminal part of the cytoplasm of renal proximal tubular cells (Fig. 1 b) and of synovial fibroblasts (Fig. 6b ).

Immunogold labelling in combination with cryoultramicrotomy has been proven to be a useful method for the visualization and localization of cell surface and intracellular proteins (Mizuno, 1993) . Especially the membrane preservation (Mizuno, 1993) is responsible for the precise localization of APN (CD 13) and DPP IV (CD 26) on the brush border membranes of several cells from different species (Fig. 1, 5 , Kettmann et aI., 1992a, b) as well as of various cultured cells (Figs. 6, 7) . Cryosections according to Tokuyasu (1973 Tokuyasu ( , 1989 offer a good compromise between antigenicity and ultrastructure preservation and are suitable for the detection both intracellular (Figs. 1 b, 6 b) and extracellular antigens (all figures).

Our observations (Fig. 1 , 5) agree with biochemical data of Kenny and Maroux (1982) , who described the abundant expression of APN and DPP VI on the brush border membrane of kidney tubular cells and small intestinal enterocytes. Bathon et ai. (1992) reported the two peptidases on the cell surface of human synovial fibroblasts. Our immunological detection of APN coincides with these findings (Fig. 6 ) as does the localization of both ectoenzymes on activated T lymphocytes (Fig. 7 a, b) with data of Kunz et ai. (1993) and Mentlein et ai. (1984) . But the expression of CD 13 and CD 26 on T lymphocytes is less then on synovial fibroblastes or brush border membranes.

The gold-labelled vesicle-like structures (Fig. 1 b, 6b ) may be connected with biosynthesis, co-translational glycosylation and intracellular transport of microvillar proteins to the luminal membrane (Danielsen et aI., 1987) . Hansen et aI. (1987) suggest that in pig enterocytes these transport vesicles operate between Golgi complex and cell membranes Kettmann et aI. (1992 b) detected the same vesicular structures also in the luminal part of rat kidney tubular cells and discussed them as transport vesicles.

The two applied double labelling methods resulted in similar findings. The successive technique (see Figs. 2a, 3 a) as well as the simultaneous procedure (see Fig. 4a ) are useful for double labelling of closely related surface antigens. Although in the two methods the same side of the grids were used for the application of both antibodies with their respective gold probe size, no cross-contamination occurred. These results correspond to informations from Geuze et aI. (1981) and Tapia et aI. (1983) .

Immunogold labelling intensity do not necessarily represent the presence of a defined amount of antigen. Furthermore, labelling densities vary markedly according to the size of the gold particles used and densities decrease with increasing gold particle size (Giberson and Demaree 1994) . This creates problems with the quantitative interpretation of the results especially in cases of double labelling. To overcoming this problem silver-enhanced ultrasmall (1 nm) gold probes can be used (Slater, 1993) . Because of the sensitivity of the ultracryosections, in our study double labelling was only carried out with larger gold particles of different size. By changing the label sequence of APN and DPP IV and comparing the double labelling results with the respective single labelling we can evaluate the quantitative relation between both antigens. As has been shown for single labelling of synovial fibroblasts and T lymphocytes double labelling procedures can also be applied to gelatine embedded cells. In conclusion, using the described technique it is possible to investigate localization and colocalization of other cell surface and cytosolic antigens at the ultrastructural level.

Metalloprotease activity of CD 13/aminopeptidase N on the surface of human myeloid cells

Cultured human synovial fibroblasts rapidly metabolize kinins and neuropeptides

T cell activation antigen, CD 26, as a cofactor for entry of HIV in CD 4 + cells

Noren 0, and Sjostrom H (1987) Biosynthesis

Participation of both "enkephalinase" and aminopeptidase activities in the metabolism of endogenous enkephalins

Noren 0, and Laude H (1992) Aminopeptidase N is a major receptor for the enteropathogenic coronavirus TGEV

Use of colloidal gold particles in double-labeling immunoelectron microscopy of ultrathin frozen tissue sections

The influence of immunogold particle size on labeling density

Cytochemistry of membrane proteases

Electron microscopic cytochemistry and immunocytochemistry in biomedicine

Identification of aminopeptidase M as an enkephalin-inactivation enzyme in rat cerebral membranes

Immunomicroscopic localization of aminopeptidase N in the pig enterocyte. Implications for the route of intracellular transport

Lung endothelial dipeptidyl peptidase IV is an adhesion molecule for lung-metastatic rat breast and prostate carcinoma cells

Topology of microvillar membrane hydrolases of kidney and intestine

Ultrastucturallocalization of dipeptidylpeptidase IV in the glomerulum of the rat kidney

Immunoelectron microscopic localization of microsomal alanine aminopeptidase

Aminopeptidase N (CD 13, EC 3.4.11.2.) occurs on the surface of resting and concanavalin A-stimulated lymphocytes

Antigen-antibody crossed electrophoresis

Human myeloid plasma membrane glycoprotein CD 13 (gp 150) is identical to aminopeptidase N

Mammalian proteases. A glossary and bibliography

Dipeptidyl peptidase IV as a new surface marker for a subpopulation of human T-lymphocytes

Nausch I, Mentlein R, and Heymann E (1990) The degradation of bioactive peptides and proteins by dipeptidyl peptidase IV from human placenta

Demonstration of CD 13/Aminopeptidase N on synovial fluid T cells from patients with different forms of joint effusions

Role of aminopeptidase N (CD 13) in tumor-cell invasion and extracellular matrix degradation

Cluster report: CD 26

De Mey 1, and Polak 1M (1983) Double immunogold staining method for the simultaneous ultrastructural localization of regulatory peptides

Use of poly(vinylpyrrolidone) and poly(vinyla