- Email: [email protected]
Light microscopical localization of enzymes by means of cerium-based methods
Acta histochem. 77, 67-73 (1985)
Department of Histochemistry of the Institute of Anatomy, Friedrich Schiller University, J ena, GDR
Light microscopical localization of enzymes by means of cerium-based methods II. A new cerium-lead-technique for alkaline phosphatase By
KARL-JURGE~
HALBHGBER and NORBERT ZIMMERMANN With Plate IV (l{eeei\·el[ Augw.;t, 1:1, 1984)
Summary The use of cerium ions as a primary capture reagent in phosphatase histochemistry on the light and the electron microscopie level is a progress in the field of enzyme localization. :\lany influences of other captures (as of leal[ ions), e.g. enzyme inhibition, diffusion and other artefacts, are restricted when cerium-hased methods arc used. But the broader use of cerium is difficult, hccause cerium ions are at alkaline pH converted to the insoluble cerium hydroxide, which intensively precipitates in the incubation medium. This is an important disadvantage for a successful histochemical detection of alkaline phosphatase. The aim of this paper is to describe a new cerium- based method for the light microscopical detection of alkaline phosphatase, which is free from all these above mentioned problems. It is proposed a eollidine buffer-sucrose containing medium, which holds cerium ions at pH = 9.0 in solution. The histochemical results of this method are excellent. The method is compared with a strontium based technique, the eoupling azo dye technique for alkaline phosphatase as well as with in vitro anel histochemical experiments with several chelator agents. The cerium-based collidine-sucrose technique is superior to all other procedures tested here and is recommonded for a broader use.
Introduction A further development of new enzyme detection methods in phosphatase histochemistry is necessary, because the classical lead based GOMoRI-methods and their numerous modifications lead to marked enzyme inhibition, unspecific lead phosphate, and lead nucleotide deposits (affinity artefacts) as well as lead phosphate diffusion (diffusion artefacts). These disadvantages can be attributed to toxic effects of lead cations and to specific physicochemical properties of the chemically undefined lead phosphates in general (see review HALBHUBER 1973). Therefore new cerium-based methods were described (literature see in ZIMMERMANN and HALBHUBER 1984). Special difficulties are connected with the cerium mediated detection of alkaline phosphatase, because due to the high pH value of = 9.0 in the incubation medium, cerium ions form cerium hydroxide precipitates (HULSTEART et al. 1983), which leads to the following consequences: 1. Variable decrease in the cerium ion concentration 2. Reduction of the capture capacity. 3. Unsatisfactory and unreproducible topochemicall'esuIts. 5*
68
K.·J. HALBHUBER and N. ZIMMERMANN
Our new alkaline cerium method is free from all these prohlems. A fundamental component of our method is a collidine buffer system containing sucrose which is able to chelate cerium ions at higher alkaline pH values. The primary reaction product, ceriulll phosphate, is converted to lead phosphate which can be visualized as lead sulfide. In addition, the efficiency of the cerium method is compared with a strontium-based and with azo dye stuff coupling method for alkaline phosphatase.
Material and Methods 1. A" experimental material, we used the kidneys of adult rats (strain Uje: vYIST) of both sexes after perfusion via Aorta thoracica with 2 % glutaraldehyde in 50 mmol cacodylate buffer (pH = i.4, isosmolaric with NaCl) as previously described (ZIMMERMANN and HALBHl:BER 1984). Before use, the tissue samples were stored in 50 mmol cacodylate buffer for 10 min at 4 DC.
2. Light microscopical cerium-based method for detection of alkaline phosphatase 2.1. 7 p.m thick cryostat sections (Frigocut 2700, Jung, Heidelberg, FRG) of the kidneys were mounted on gelatinized slides 2.2. Preincubation of kidney sections in 1 mmol CeCI 3 (VEB Laborchemie Apolda, GDR) in 0.2 mol s-collidine-l mol/l HCI-buffer (2, 4, 6 collidine; VEB Teerdestillation Chemische Fabrik Erkner, GDR) containing 7.5 % sucrose (VEB Laborchemie Apolda, GDR), (pH = 9.2) for 10 min at 20 DC 2.3. Incubation medium: 1 mmol CeCI 3 , 4 mmol J\IgCI2 (VEB Laborchemie Apolcla, GDR), 10 mmol sodium-{i-glycerophosphatc (Merck, Darmstadt, FRG), 7.5 % sucrose in 0.2 mol collidine1 moll I HCI-buffer (pH = 9.0), 60 to 120 min at 37°C. Technical notes: For preparation of 0.2 mol collidine buffer (pH = 9.2), 2.67 ml s-collidine were solved in 50 ml aqua dest. After that, 1 to 2 drops of 1.25 mol/l HCI were added, and finally aqua dest. to a total volume of 100 m!. Then 750 mg sucrose were solved in 8.5 ml collidine-HClbuffer, then addition of 0.5 ml Ceel 3 solution (30 mg CeCl 3 /5 ml aqua dest.) in form of small drops by pipette during permanent stirring, then adclition of 8 mg MgCl 2 as well as 1 ml Na-{iglycerophosphate solution (30 mglml aqua dest.). It is necessary to use always thoroughly boiled (C0 2 free) distilled water 2.4. Rinsing in aqua dest. for 10 min 2.5. Incubation in 0.5 % lead acetate (VEB Laborchemie Apolda, GDR) in aqua dest. 10 min at 20 DC 2.6. Rinsing in aqua dest. for 10 min 2.7. Treatment with a solution of 2 % ammonium sulfide (VEB Laborehemie Apolda, GDR) or 2.5 % Na 2 S in aqua dest. [(100 ml 2.5 % Na-sulfide contain 10 mil molll HCl); (P.P.H. Polski Ochezynriki Chern., Gliwice, Polancl)] 2 min at 20°C 2.8. Rinsing in aqua dest. for about 10 to 60 min and mounting of sections in glycerol jelly medium 3. Comparing enzyme reactions 3.1. Strontium-based technique 3.1.1. 7/km thick cryostat sections of the rat kidneys were mounted on gelatinized slides. 3.1.2. Preincubation of kidney sections in 70 mmol glycine NaOH buffer (pH = 9.0), (VEB Feinehemie Sebnitz, GDR) 10 min at 20 DC 3.1.3. Incubation of slides in the following medium for 60 to 120 min at 37°C: SrCI 2 1 mmol, MgCl 2 4 mmol, Na-{i-glycerophosphate 10 mmol in 70 mmol glycine-NaOH buffer (pH = 9.07) 3.1.4. Rinsing in aqua dest. for 10 min 3.1.5. Incubation in 0.5 % lead acetate in aqua dest. 10 min at 20 DC 3.1.6. Rinsing in aqua clest. for 10 min 3.1.7. 2 % ammonium- or Na-sulfide (as in 2.7.) 2 min at 20°C 3.1.8. Short rinsing in aqua dest. and mounting in glycerol jelly 3.2. Coupling azo dye method according to PEARSE (1968) with employment of the tetrazotate ofo-dianisidine (Echtrotsalz LTR, Carl Pinnow, Berlin-West)
Light microscopical localization of enzymes. II.
69
4. Control incubations 4.1. Substrate control: Incubation without sodium-f3-glycerophosphate 4.2. Capture control: Incubation without cerium chloride 4.3. Heat inactivation of the enzyme by putting the section into 50 mmol cacodylate buffer at 90°C for 15 min 4.4. Enzyme inhibit'ion: Pretreatment of sections with 10 mmol cysteine hydrochloride (Re!lnal, Budapest, Hungaria) in 0.2 mol collidine buffer (pH = 9.0) for 10 min at 37 DC (GEYER 1973) 5. Optimation procedures 5.1. Variation of the CeCI 3 concentration: 1, :1, and 5 mmol 5.2. Conversion of the cerium phosphate into lead phosphate by freshly prepared alkaline lead citrate (REYNOLDS 1963) instead of lead acetate 5.3. Employment of several chelator agents 5.3.1. In vitro precipitation test: The following chelators were added to 8.0 ml of 0.2 mol collidine buffer (pH = 9.0) containing 1 mmol CeCI 3 , 10 mmol Na-f3-glycerophosphate and 4 mmol MgCI2 in a concentration of 0.5 or 1 % instead of 7.5 % sucrose: dextran 20 and 60 (Serv([, Heidelberg, FRG), glycerol (VEB Laborchemie Apolda, GDR), tiron (VEB Berlin-Chemie, Berlin-.\dlershof, GDR), nitriloacetic acid (Chelaplex I, VEB Berlin-Chemie, Berlin-Adlershof, GDR), ethylendiamintetraacetic acid as Na-salt (Chelaplex III, VEB Berlin-Chemie, Berlin-Adlershof, GDR), sodium-potassium-tartrate (VEB Laborchemie Apolda, GDR), o-phenanthroline (Chemapol, Prahe, CSSR) and dipyridyl (Schonert K.-G., Leipzig, GDR). 15 min after addition of 0.2 ml 67 mmol Na 2 HP0 4 solution (pH = 9.0) the formation of corium phosphate precipitates was checked. 5.3.2. Incubation of kidney sections (see 2.) in the media of 5.3.1. and checking of the histochemical enzyme reaction.
Results When the cerium-based procedure for alkaline phosphatase is employed, a typical distribution pattern of the reaction products can be seen. The epithelial cells of the tubular system in the renal cortex were clearly reactive. Especially the brush border region of the epithelial cells of the main tubules is intensively stained in more or less dark brown colour. These cell areas are sharply marked off from the resting cell body (Plate IV: Fig. 1). Other parts of the tubular epithelial cells are nonreactive. The background of the sections had a homogenous tinge of brown. Only in some cases nuclei of various cell types showed a weak brown staining. The strontium-based method provided nearly identical results. But the cellular brush borders always showed a slightly deeper brown staining in comparison to the cerium-based technique (Plate IV: Fig. 2). Unspecific staining was not observed. Kidney sections, which were tested parallel for alkaline phosphatase activity with a coupling azo dye method, also showed a strong positive reaction. The brush borders were intensively stained in a red brown colour. But in many cases, the azo dye reaction products were not so evidently sharply marked off from the nonreactive parts of the cells. In contrast to the metal techniques, i.e. cerium and strontium-based methods, the region of thc brush border basis revealed a slightly undistinct localization of the reaction products. Their cell bodies contained weakly (Fig. 3). In all cases the control reactions provided a negative result. Without Na-p-glycerophosphate or cerium chloride, no enzyme activity was detectable (Plate IV: Fig. 4). Only the cells were weakly hOl1logenowl and brown-yellow. The heat inactivated sections showed very slight brown staining without any favoured special structures. After cysteine treatment of the sections, a few brush borders showed a more but also only weak tinge of brown, when compared with the bodies of the tubular cells. The intensity of the positive enzyme reaction was unchanged independent of the cerium chloride concentration which ranged from 1 to 5 mmol. In a few cases, a
70
K.-J. HALBHUBER and N. ZIMMERMANN
Table 1. Summary of the results with different chelator agents in stabilizing the cerium containing collidine buffer medium and in the capability of bound cerium ions to react with free inorganic phosphate in both the in vitro test and the tissue section Chelator agent
Cerium containing histochemical colli dine buffer medium after equilibration at pH = 9.0
Cerium phosphate precipitation after addition of inorganic phosphate to the complete collidine buffer mixture in the in vitro test
Histochemical results in the tissue section
glycerol
nearly transparent
evident
positive
sucrose
transparent
evident
strongly positive
dextran 20 and 60
transparent
no
negative
chelaplex I and III
transparent
no
negative
tartrate
transparent
no
negative
dipyridyl
weakly muddy
no
negative
o-phenanthroline
muddy
moderate
weakened
turbidity of the incubation mixture with 5 mmol was observed which was progressive during incubation of sections at 37°C. Such sections always showed unspecific brownish precipitates of variable size over several structures. If instead of lead acetate, alkaline lead citrate was employed for conversion of the cerium phosphate into lead phosphate, the enzyme reaction products were only slightly brown after treatment with ammonium sulfide, i.e. the enzyme reaction was only weakly positive. Unspecific staining of other structures than brush borders was not detected under these conditions. But the background staining was almost absent here_ A modification of the composition of the alkaline collidine buffer system in direction of the content with several chelator agents leads to different results. The employed chelators are listed in Table 1. From this, it can be noted that with the exception of sucrose the agents provided more or less unsatisfactory results. Moreover the table show~ that parallel to the turbidity of the incubation mixture, e.g. in the presence of O-phenanthroline or dipyridyl, the cerium phosphate precipitation in V1:tro was inhibited or decreased. Analogous to these facts, the histochemical reactions in these mediums were also negatively altered. Furthermore, the application of glycerol and dipyridyl as chelators leads to a progressive increase in the turbidity of the incubation solution during incubation of the sections at 37°C. Hence, the histochemically detectable enzyme activity is reduced, and in the structures of the section unspecific precipitates can be seen. In the presence of tiron, dextran, chelaplex I and III as well as tartrate the cerium containing mixtures are clear at pH = 9.0, but after the addition of inorganic phosphate, no cerium phosphate precipitation can be seen and the histochemical procedures are always negative.
Discussion The results presented here, demonstrate a clear, selective detection of alkaline phosphatase activity in the brush borders of renal cortex of the rat. All employed methods, i.e. the ceriumbased method in a collidine-sucrose system, the strontium-based technique, and the technique based on the coupling azo dye provided similar results. The specificity of the enzyme reaction is supported by:
Light microscopical localization of enzymes. II.
71
1. The control experiments, which revealed a completely negative histoehemical reaction in the brush borders. (It is known from these structures that only they contain alkaline phosphatase in a mpmbrane bound form.) 2. Thp laek of the staining of other structures than brush borders by means of the enzyme reo actions, which were tested hero (lack of the coreaction of other enzymes). 3. The almost complete absence of unspecific percipitates over different structures (lack of affinity artefads). Especially distinct localization of alkaline phosphatase can be observed by means of the metal capture techniques, i.e. the cerium- and strontium-based methods. Compared to the coupling azo dye method, the topochemical enzyme (letection by the former techniques is more precise, in the latter case the stained brush borders aro not so clearly marked off from the nonstained negative cell body of the tubular cells. Also a slight coreaction of other enzymes in the cell bodies "-as observed.
In this respect, it must he conelu(led that a detectable diffusion of the enzyme molecules out of the brush border membrane is not responsible for the more difficult enzyme localization by the coupling azo dye method. Rather, it is probable that the hardly soluble azo dye can partly diffuse in the sections. The coreaction of the cell bodies of the main tubular cells can probably be attributed to a weak p-nitrophonylphosphate splitting by membrane bound ATPases located in the J3-cytomemhranes of the cells. In contrast to these findings, a disturbing diffusion of ccrium phosphate and strontium phosphatc seems to bc restricted, i.e. cannot be observed, and the reactions arc very specific for alkaline phosphatase. In this respeet, the metal capture methods for alkaline phosphatase are superior to the standard coupling azo dye methorls. On the other hand, the histochemical enzyme localization by means of the primary phosphate captures cerium and strontium is equally satisfactory. Differences in the light microscopic reaction pattern between both reaction types arc negligible. Only in rare cases, the chromatin of nuclei shows a weak tinge of brown as expression of an unimportant affinity artefact. Such affinity artefacts were not observed when cerium-baseel methods were used at the electron microscopic level. The cerium phosphate precipitates are electron opaque enough for rlirect visualization in contrast to strontium phosphate. By the USE' of strontium techniques, the resulting strontium phosphate has to be converted into lead phosphate because of its bctter electron opacity. This causes a number of affinity artefacts (ZAJIC and SCHACHT 198:3). It seems that the lead phosphate also has a great affinity to chromatin (more details see in HALBHl:BER 197:3). But it is not clear why this artefact cannot always bc seen. Thus, in principle, both metal capture procedures are valid methods also in light microscopic histochemistry of the alkaline phosphatase. Moreover, in this connection, it is necessary to note that the first strontium-based method described by FIRTH (1974) considered higher concentrations of strontium (20 mmol!) and p-nitrophenylphosphate as substrate in comparison to our technique. But the high strontium concentrations and the unusual substrate led to the following disadvantages: 1. A slight decrease in enzyme activity by strontium ions, which is detectable after short incubation times (e.g. 30 min) 2. Slight coreaction of ;\fg 2 + dependent ATPases.
""Yhen the strontium concentrations were reduced to 1 mmol and J3-g1ycerophosphate was the substrate, these problems would not appear. The employment of strontium ions as primary capture is easier than that of cerium ions, because thc former cannot react at the alkaline pH to an insoluble hydroxide. Under these alkaline conditions, strontium phosphate is precipitable enough. Completely different problems are connected with cerium. Free cerium ions react at pH = 9.0 to insoluble cerium hydroxide. Therefore, the general application of cerium ions at alkaline pH is difficult. HULSTAERT et al. (1983) described an unsatisfactory glycine-NaOH buffer system with 1 mmol CeCl 3 at pH = 9.0 which leads to an intensive formation of unspecific cerium precipitates in the incubation mixture. With this method, the alkaline phosphatase reaction mostly gives weak or negative results and the reaction is not reproducible. The unspecific precipitations in the tissues were removed by incubation of the speeimens after enzyme reaction in a 0.1 mol cacodylate buffer (containing 6.8 % sucrose) at pH = 6.0 to resolve the cerium hydroxide deposits. Another method for alkaline phosphatase has been recently proposed by ROBINSON et al. (1983). They employed a Tris maleate buffer at pH = 8.0 containing 7.5 % sucrose. The disadvantage of this method is that a pH = 8.0 is not high enough to guarantee a reaction only of the alkaline phosphatase. Rather, a coreaction of other
72
K.·J. HALBHUBER and N. ZIMMERMANN
phosphatases is possible (GEYER 1973). Therefore we attempted to find an useful collidine buffersucrose system for alkaline phosphatase histochemistry, which has the following properties: 1. Avoidance of free cerium ions by binding to buffer components 2. Therefore inhibition of spontaneous formation and precipitation of cerium hydroxide 3. Capability of bound cerium ions to react with free inorganic phosphate to hardly soluble cerium phosphate 4. No disturbing depressing effects on enzyme activity.
The comparison of in vitro experiments and histochemical enzyme reactions evidently shows the superiority of the cerium containing collidine buffer with sucrose to all other systems which were tested here (Table 1). Sucrose is an essential ingredient of the collidine buffer to stabilize cerium ions in solution at alkaline pH, because collidine buffer alone has not the capability to inhibit cerium hydroxide precipitation. On the other hand, sucrose in another buffer (e.g. Tris maleate, glycine.NaOH) is also not sufficient to stabilize the cerium salt solution. Thus, only collidine in combination with sucrose is able to do this. The other tested chelators have the following disadvantages: 1. Partial precipitation of cerium hydroxide immediately after application of cerium chloride to the mixture equilibrated at pH = 9.0 (o-phenanthroline, dipyridyl) 2. Progressive cerium hydroxide precipitation during incubation of sections for 2 h at 37 DC in the incubation medium (o.phenanthroline, dipyridyl, glycerol) 3. Strong binding of cerium ions and impossibility to react with free inorganic phosphate dextran 20 and 60, Tiron, chelaplex I and III, tartrate).
The histochemical results are in conformity with these very differentiated in vitro experiments and lead to a more or less unsatisfactory enzyme detection. Therefore it is right to conclude that the cerium· based collidine buffer-sucrose system provides the best method to detect alkaline phosphatase activity in cryostat sections. This procedure is recommended for a broader use in light microscopical phosphatase histochemistry. Experiments for an adaption of this method to the electron microscopical level are in progress.
Acknowledgements The authors thank Mrs. U. -:\10LLER for excellent technical assistance.
References FIRTH,1. A., Problems of specificity in the use of a strontium capture technique for the cytochemicallocalization of ouabain-sensitive, potassium-dependent phosphatase in mammalian renal tubules. J. Histochem. Cytochem. 22, 1163~1168 (1974). GEYER, G., Ultrahistochemie. 2. Auf!., VEB Gustav Fischer Verlag, Jena 1973. HALBHUBER, K.-J., Methodische Bedingungen fUr den histochemischen Nachweis anorganischer Ionen mit besonderer Beriicksichtigung des anorganischen Orthophosphats. Erg. Expel'. :\1ed. (Berlin) 14, 1-102 (1973). HULSTAERT, C. E., KALICHARAX, D., and HARDOXK, M. J., Cytochemical demonstration of phosphatases in the rat liver by a cerium-based method in combination with osmium tetroxide and potassium ferro cyanide postfixation. Histoehemistry 78, 71-79 (1983). PEARSE, A. G. E., Histochemistry. Theoretical and Applied. 2nd Churchill, London 1968. REYNOLDS, E. S., The use of lead citrate at high pH as an electron opaque stain in electron micro· scopy. J. Cell BioI. 17, 208~212 (1963). ROBINSON, J. M., and KARNOVSKY, :\1. J., Ultrastructural localization of several phosphatases with cerium. J. Histochem. Cytochem. 31, 1197-1208 (1983). ZAJIC, G., and SCHACHT, J., Cytochemical demonstration of adenylate cyclase with strontium chloride in the rat pancreas. J. Histochem. Cytochem. 31, 25~28 (1983). ZIMMERMANN; N., and HALBHUBER, K.-J., Light microscopical localization of enzymes by means of cerium-based methods. 1. Detection of acid phosphatase by a new cerium-Icad.teclmique (Ce-Pb-method). Acta histochem. 76, 97-104 (1985). Authors' address: Prof. Dr. K.-J. HALBHUBER and Dr. N. ZIMMERMANN, Institute of Anatomy, Friedrich Schiller University, DDR - 6900 Jena, Teichgraben 7.
Acta histochem. Vol. 77
Plate IV
2
4
3
Halbhub er a nd Zimm erman n VEB GUSTAV FISCHER VERLAG JENA
Light microscopical localization of enzymes. II.
73
Explanation of Plate IV Fig. I. Alkaline phosphatase activity of the brush border of the epithelial cells in the tuhular system in the renal cortex. Only the brush border regions of the cells are stained. X 250. Fig. 2. Alkaline phosphatase demonstrated by means of the strontium-based method in the cortex of the rat kidney. Note the selective deep brown staining of the eellular brush border. X 250. Fig. 3. Coupling azo dye technique for the demonstration of alkaline phosphatase activity In the kidney. The brush borders of the tubular cells are strongly stained, but the staining is not so clearly marked off from the residual cell body as in the case of the cerium or strontium methods. X 250. Fig. 4. Cerium-based method for the demonstration of alkaline phosphatase; incubation without sodium-{J-glyeerophosphate (substrate control). The enzyme reaction is completely negative. X 160.
Department of Histochemistry of the Institute of Anatomy, Friedrich Schiller University, J ena, GDR
Light microscopical localization of enzymes by means of cerium-based methods II. A new cerium-lead-technique for alkaline phosphatase By
KARL-JURGE~
HALBHGBER and NORBERT ZIMMERMANN With Plate IV (l{eeei\·el[ Augw.;t, 1:1, 1984)
Summary The use of cerium ions as a primary capture reagent in phosphatase histochemistry on the light and the electron microscopie level is a progress in the field of enzyme localization. :\lany influences of other captures (as of leal[ ions), e.g. enzyme inhibition, diffusion and other artefacts, are restricted when cerium-hased methods arc used. But the broader use of cerium is difficult, hccause cerium ions are at alkaline pH converted to the insoluble cerium hydroxide, which intensively precipitates in the incubation medium. This is an important disadvantage for a successful histochemical detection of alkaline phosphatase. The aim of this paper is to describe a new cerium- based method for the light microscopical detection of alkaline phosphatase, which is free from all these above mentioned problems. It is proposed a eollidine buffer-sucrose containing medium, which holds cerium ions at pH = 9.0 in solution. The histochemical results of this method are excellent. The method is compared with a strontium based technique, the eoupling azo dye technique for alkaline phosphatase as well as with in vitro anel histochemical experiments with several chelator agents. The cerium-based collidine-sucrose technique is superior to all other procedures tested here and is recommonded for a broader use.
Introduction A further development of new enzyme detection methods in phosphatase histochemistry is necessary, because the classical lead based GOMoRI-methods and their numerous modifications lead to marked enzyme inhibition, unspecific lead phosphate, and lead nucleotide deposits (affinity artefacts) as well as lead phosphate diffusion (diffusion artefacts). These disadvantages can be attributed to toxic effects of lead cations and to specific physicochemical properties of the chemically undefined lead phosphates in general (see review HALBHUBER 1973). Therefore new cerium-based methods were described (literature see in ZIMMERMANN and HALBHUBER 1984). Special difficulties are connected with the cerium mediated detection of alkaline phosphatase, because due to the high pH value of = 9.0 in the incubation medium, cerium ions form cerium hydroxide precipitates (HULSTEART et al. 1983), which leads to the following consequences: 1. Variable decrease in the cerium ion concentration 2. Reduction of the capture capacity. 3. Unsatisfactory and unreproducible topochemicall'esuIts. 5*
68
K.·J. HALBHUBER and N. ZIMMERMANN
Our new alkaline cerium method is free from all these prohlems. A fundamental component of our method is a collidine buffer system containing sucrose which is able to chelate cerium ions at higher alkaline pH values. The primary reaction product, ceriulll phosphate, is converted to lead phosphate which can be visualized as lead sulfide. In addition, the efficiency of the cerium method is compared with a strontium-based and with azo dye stuff coupling method for alkaline phosphatase.
Material and Methods 1. A" experimental material, we used the kidneys of adult rats (strain Uje: vYIST) of both sexes after perfusion via Aorta thoracica with 2 % glutaraldehyde in 50 mmol cacodylate buffer (pH = i.4, isosmolaric with NaCl) as previously described (ZIMMERMANN and HALBHl:BER 1984). Before use, the tissue samples were stored in 50 mmol cacodylate buffer for 10 min at 4 DC.
2. Light microscopical cerium-based method for detection of alkaline phosphatase 2.1. 7 p.m thick cryostat sections (Frigocut 2700, Jung, Heidelberg, FRG) of the kidneys were mounted on gelatinized slides 2.2. Preincubation of kidney sections in 1 mmol CeCI 3 (VEB Laborchemie Apolda, GDR) in 0.2 mol s-collidine-l mol/l HCI-buffer (2, 4, 6 collidine; VEB Teerdestillation Chemische Fabrik Erkner, GDR) containing 7.5 % sucrose (VEB Laborchemie Apolda, GDR), (pH = 9.2) for 10 min at 20 DC 2.3. Incubation medium: 1 mmol CeCI 3 , 4 mmol J\IgCI2 (VEB Laborchemie Apolcla, GDR), 10 mmol sodium-{i-glycerophosphatc (Merck, Darmstadt, FRG), 7.5 % sucrose in 0.2 mol collidine1 moll I HCI-buffer (pH = 9.0), 60 to 120 min at 37°C. Technical notes: For preparation of 0.2 mol collidine buffer (pH = 9.2), 2.67 ml s-collidine were solved in 50 ml aqua dest. After that, 1 to 2 drops of 1.25 mol/l HCI were added, and finally aqua dest. to a total volume of 100 m!. Then 750 mg sucrose were solved in 8.5 ml collidine-HClbuffer, then addition of 0.5 ml Ceel 3 solution (30 mg CeCl 3 /5 ml aqua dest.) in form of small drops by pipette during permanent stirring, then adclition of 8 mg MgCl 2 as well as 1 ml Na-{iglycerophosphate solution (30 mglml aqua dest.). It is necessary to use always thoroughly boiled (C0 2 free) distilled water 2.4. Rinsing in aqua dest. for 10 min 2.5. Incubation in 0.5 % lead acetate (VEB Laborchemie Apolda, GDR) in aqua dest. 10 min at 20 DC 2.6. Rinsing in aqua dest. for 10 min 2.7. Treatment with a solution of 2 % ammonium sulfide (VEB Laborehemie Apolda, GDR) or 2.5 % Na 2 S in aqua dest. [(100 ml 2.5 % Na-sulfide contain 10 mil molll HCl); (P.P.H. Polski Ochezynriki Chern., Gliwice, Polancl)] 2 min at 20°C 2.8. Rinsing in aqua dest. for about 10 to 60 min and mounting of sections in glycerol jelly medium 3. Comparing enzyme reactions 3.1. Strontium-based technique 3.1.1. 7/km thick cryostat sections of the rat kidneys were mounted on gelatinized slides. 3.1.2. Preincubation of kidney sections in 70 mmol glycine NaOH buffer (pH = 9.0), (VEB Feinehemie Sebnitz, GDR) 10 min at 20 DC 3.1.3. Incubation of slides in the following medium for 60 to 120 min at 37°C: SrCI 2 1 mmol, MgCl 2 4 mmol, Na-{i-glycerophosphate 10 mmol in 70 mmol glycine-NaOH buffer (pH = 9.07) 3.1.4. Rinsing in aqua dest. for 10 min 3.1.5. Incubation in 0.5 % lead acetate in aqua dest. 10 min at 20 DC 3.1.6. Rinsing in aqua clest. for 10 min 3.1.7. 2 % ammonium- or Na-sulfide (as in 2.7.) 2 min at 20°C 3.1.8. Short rinsing in aqua dest. and mounting in glycerol jelly 3.2. Coupling azo dye method according to PEARSE (1968) with employment of the tetrazotate ofo-dianisidine (Echtrotsalz LTR, Carl Pinnow, Berlin-West)
Light microscopical localization of enzymes. II.
69
4. Control incubations 4.1. Substrate control: Incubation without sodium-f3-glycerophosphate 4.2. Capture control: Incubation without cerium chloride 4.3. Heat inactivation of the enzyme by putting the section into 50 mmol cacodylate buffer at 90°C for 15 min 4.4. Enzyme inhibit'ion: Pretreatment of sections with 10 mmol cysteine hydrochloride (Re!lnal, Budapest, Hungaria) in 0.2 mol collidine buffer (pH = 9.0) for 10 min at 37 DC (GEYER 1973) 5. Optimation procedures 5.1. Variation of the CeCI 3 concentration: 1, :1, and 5 mmol 5.2. Conversion of the cerium phosphate into lead phosphate by freshly prepared alkaline lead citrate (REYNOLDS 1963) instead of lead acetate 5.3. Employment of several chelator agents 5.3.1. In vitro precipitation test: The following chelators were added to 8.0 ml of 0.2 mol collidine buffer (pH = 9.0) containing 1 mmol CeCI 3 , 10 mmol Na-f3-glycerophosphate and 4 mmol MgCI2 in a concentration of 0.5 or 1 % instead of 7.5 % sucrose: dextran 20 and 60 (Serv([, Heidelberg, FRG), glycerol (VEB Laborchemie Apolda, GDR), tiron (VEB Berlin-Chemie, Berlin-.\dlershof, GDR), nitriloacetic acid (Chelaplex I, VEB Berlin-Chemie, Berlin-Adlershof, GDR), ethylendiamintetraacetic acid as Na-salt (Chelaplex III, VEB Berlin-Chemie, Berlin-Adlershof, GDR), sodium-potassium-tartrate (VEB Laborchemie Apolda, GDR), o-phenanthroline (Chemapol, Prahe, CSSR) and dipyridyl (Schonert K.-G., Leipzig, GDR). 15 min after addition of 0.2 ml 67 mmol Na 2 HP0 4 solution (pH = 9.0) the formation of corium phosphate precipitates was checked. 5.3.2. Incubation of kidney sections (see 2.) in the media of 5.3.1. and checking of the histochemical enzyme reaction.
Results When the cerium-based procedure for alkaline phosphatase is employed, a typical distribution pattern of the reaction products can be seen. The epithelial cells of the tubular system in the renal cortex were clearly reactive. Especially the brush border region of the epithelial cells of the main tubules is intensively stained in more or less dark brown colour. These cell areas are sharply marked off from the resting cell body (Plate IV: Fig. 1). Other parts of the tubular epithelial cells are nonreactive. The background of the sections had a homogenous tinge of brown. Only in some cases nuclei of various cell types showed a weak brown staining. The strontium-based method provided nearly identical results. But the cellular brush borders always showed a slightly deeper brown staining in comparison to the cerium-based technique (Plate IV: Fig. 2). Unspecific staining was not observed. Kidney sections, which were tested parallel for alkaline phosphatase activity with a coupling azo dye method, also showed a strong positive reaction. The brush borders were intensively stained in a red brown colour. But in many cases, the azo dye reaction products were not so evidently sharply marked off from the nonreactive parts of the cells. In contrast to the metal techniques, i.e. cerium and strontium-based methods, the region of thc brush border basis revealed a slightly undistinct localization of the reaction products. Their cell bodies contained weakly (Fig. 3). In all cases the control reactions provided a negative result. Without Na-p-glycerophosphate or cerium chloride, no enzyme activity was detectable (Plate IV: Fig. 4). Only the cells were weakly hOl1logenowl and brown-yellow. The heat inactivated sections showed very slight brown staining without any favoured special structures. After cysteine treatment of the sections, a few brush borders showed a more but also only weak tinge of brown, when compared with the bodies of the tubular cells. The intensity of the positive enzyme reaction was unchanged independent of the cerium chloride concentration which ranged from 1 to 5 mmol. In a few cases, a
70
K.-J. HALBHUBER and N. ZIMMERMANN
Table 1. Summary of the results with different chelator agents in stabilizing the cerium containing collidine buffer medium and in the capability of bound cerium ions to react with free inorganic phosphate in both the in vitro test and the tissue section Chelator agent
Cerium containing histochemical colli dine buffer medium after equilibration at pH = 9.0
Cerium phosphate precipitation after addition of inorganic phosphate to the complete collidine buffer mixture in the in vitro test
Histochemical results in the tissue section
glycerol
nearly transparent
evident
positive
sucrose
transparent
evident
strongly positive
dextran 20 and 60
transparent
no
negative
chelaplex I and III
transparent
no
negative
tartrate
transparent
no
negative
dipyridyl
weakly muddy
no
negative
o-phenanthroline
muddy
moderate
weakened
turbidity of the incubation mixture with 5 mmol was observed which was progressive during incubation of sections at 37°C. Such sections always showed unspecific brownish precipitates of variable size over several structures. If instead of lead acetate, alkaline lead citrate was employed for conversion of the cerium phosphate into lead phosphate, the enzyme reaction products were only slightly brown after treatment with ammonium sulfide, i.e. the enzyme reaction was only weakly positive. Unspecific staining of other structures than brush borders was not detected under these conditions. But the background staining was almost absent here_ A modification of the composition of the alkaline collidine buffer system in direction of the content with several chelator agents leads to different results. The employed chelators are listed in Table 1. From this, it can be noted that with the exception of sucrose the agents provided more or less unsatisfactory results. Moreover the table show~ that parallel to the turbidity of the incubation mixture, e.g. in the presence of O-phenanthroline or dipyridyl, the cerium phosphate precipitation in V1:tro was inhibited or decreased. Analogous to these facts, the histochemical reactions in these mediums were also negatively altered. Furthermore, the application of glycerol and dipyridyl as chelators leads to a progressive increase in the turbidity of the incubation solution during incubation of the sections at 37°C. Hence, the histochemically detectable enzyme activity is reduced, and in the structures of the section unspecific precipitates can be seen. In the presence of tiron, dextran, chelaplex I and III as well as tartrate the cerium containing mixtures are clear at pH = 9.0, but after the addition of inorganic phosphate, no cerium phosphate precipitation can be seen and the histochemical procedures are always negative.
Discussion The results presented here, demonstrate a clear, selective detection of alkaline phosphatase activity in the brush borders of renal cortex of the rat. All employed methods, i.e. the ceriumbased method in a collidine-sucrose system, the strontium-based technique, and the technique based on the coupling azo dye provided similar results. The specificity of the enzyme reaction is supported by:
Light microscopical localization of enzymes. II.
71
1. The control experiments, which revealed a completely negative histoehemical reaction in the brush borders. (It is known from these structures that only they contain alkaline phosphatase in a mpmbrane bound form.) 2. Thp laek of the staining of other structures than brush borders by means of the enzyme reo actions, which were tested hero (lack of the coreaction of other enzymes). 3. The almost complete absence of unspecific percipitates over different structures (lack of affinity artefads). Especially distinct localization of alkaline phosphatase can be observed by means of the metal capture techniques, i.e. the cerium- and strontium-based methods. Compared to the coupling azo dye method, the topochemical enzyme (letection by the former techniques is more precise, in the latter case the stained brush borders aro not so clearly marked off from the nonstained negative cell body of the tubular cells. Also a slight coreaction of other enzymes in the cell bodies "-as observed.
In this respect, it must he conelu(led that a detectable diffusion of the enzyme molecules out of the brush border membrane is not responsible for the more difficult enzyme localization by the coupling azo dye method. Rather, it is probable that the hardly soluble azo dye can partly diffuse in the sections. The coreaction of the cell bodies of the main tubular cells can probably be attributed to a weak p-nitrophonylphosphate splitting by membrane bound ATPases located in the J3-cytomemhranes of the cells. In contrast to these findings, a disturbing diffusion of ccrium phosphate and strontium phosphatc seems to bc restricted, i.e. cannot be observed, and the reactions arc very specific for alkaline phosphatase. In this respeet, the metal capture methods for alkaline phosphatase are superior to the standard coupling azo dye methorls. On the other hand, the histochemical enzyme localization by means of the primary phosphate captures cerium and strontium is equally satisfactory. Differences in the light microscopic reaction pattern between both reaction types arc negligible. Only in rare cases, the chromatin of nuclei shows a weak tinge of brown as expression of an unimportant affinity artefact. Such affinity artefacts were not observed when cerium-baseel methods were used at the electron microscopic level. The cerium phosphate precipitates are electron opaque enough for rlirect visualization in contrast to strontium phosphate. By the USE' of strontium techniques, the resulting strontium phosphate has to be converted into lead phosphate because of its bctter electron opacity. This causes a number of affinity artefacts (ZAJIC and SCHACHT 198:3). It seems that the lead phosphate also has a great affinity to chromatin (more details see in HALBHl:BER 197:3). But it is not clear why this artefact cannot always bc seen. Thus, in principle, both metal capture procedures are valid methods also in light microscopic histochemistry of the alkaline phosphatase. Moreover, in this connection, it is necessary to note that the first strontium-based method described by FIRTH (1974) considered higher concentrations of strontium (20 mmol!) and p-nitrophenylphosphate as substrate in comparison to our technique. But the high strontium concentrations and the unusual substrate led to the following disadvantages: 1. A slight decrease in enzyme activity by strontium ions, which is detectable after short incubation times (e.g. 30 min) 2. Slight coreaction of ;\fg 2 + dependent ATPases.
""Yhen the strontium concentrations were reduced to 1 mmol and J3-g1ycerophosphate was the substrate, these problems would not appear. The employment of strontium ions as primary capture is easier than that of cerium ions, because thc former cannot react at the alkaline pH to an insoluble hydroxide. Under these alkaline conditions, strontium phosphate is precipitable enough. Completely different problems are connected with cerium. Free cerium ions react at pH = 9.0 to insoluble cerium hydroxide. Therefore, the general application of cerium ions at alkaline pH is difficult. HULSTAERT et al. (1983) described an unsatisfactory glycine-NaOH buffer system with 1 mmol CeCl 3 at pH = 9.0 which leads to an intensive formation of unspecific cerium precipitates in the incubation mixture. With this method, the alkaline phosphatase reaction mostly gives weak or negative results and the reaction is not reproducible. The unspecific precipitations in the tissues were removed by incubation of the speeimens after enzyme reaction in a 0.1 mol cacodylate buffer (containing 6.8 % sucrose) at pH = 6.0 to resolve the cerium hydroxide deposits. Another method for alkaline phosphatase has been recently proposed by ROBINSON et al. (1983). They employed a Tris maleate buffer at pH = 8.0 containing 7.5 % sucrose. The disadvantage of this method is that a pH = 8.0 is not high enough to guarantee a reaction only of the alkaline phosphatase. Rather, a coreaction of other
72
K.·J. HALBHUBER and N. ZIMMERMANN
phosphatases is possible (GEYER 1973). Therefore we attempted to find an useful collidine buffersucrose system for alkaline phosphatase histochemistry, which has the following properties: 1. Avoidance of free cerium ions by binding to buffer components 2. Therefore inhibition of spontaneous formation and precipitation of cerium hydroxide 3. Capability of bound cerium ions to react with free inorganic phosphate to hardly soluble cerium phosphate 4. No disturbing depressing effects on enzyme activity.
The comparison of in vitro experiments and histochemical enzyme reactions evidently shows the superiority of the cerium containing collidine buffer with sucrose to all other systems which were tested here (Table 1). Sucrose is an essential ingredient of the collidine buffer to stabilize cerium ions in solution at alkaline pH, because collidine buffer alone has not the capability to inhibit cerium hydroxide precipitation. On the other hand, sucrose in another buffer (e.g. Tris maleate, glycine.NaOH) is also not sufficient to stabilize the cerium salt solution. Thus, only collidine in combination with sucrose is able to do this. The other tested chelators have the following disadvantages: 1. Partial precipitation of cerium hydroxide immediately after application of cerium chloride to the mixture equilibrated at pH = 9.0 (o-phenanthroline, dipyridyl) 2. Progressive cerium hydroxide precipitation during incubation of sections for 2 h at 37 DC in the incubation medium (o.phenanthroline, dipyridyl, glycerol) 3. Strong binding of cerium ions and impossibility to react with free inorganic phosphate dextran 20 and 60, Tiron, chelaplex I and III, tartrate).
The histochemical results are in conformity with these very differentiated in vitro experiments and lead to a more or less unsatisfactory enzyme detection. Therefore it is right to conclude that the cerium· based collidine buffer-sucrose system provides the best method to detect alkaline phosphatase activity in cryostat sections. This procedure is recommended for a broader use in light microscopical phosphatase histochemistry. Experiments for an adaption of this method to the electron microscopical level are in progress.
Acknowledgements The authors thank Mrs. U. -:\10LLER for excellent technical assistance.
References FIRTH,1. A., Problems of specificity in the use of a strontium capture technique for the cytochemicallocalization of ouabain-sensitive, potassium-dependent phosphatase in mammalian renal tubules. J. Histochem. Cytochem. 22, 1163~1168 (1974). GEYER, G., Ultrahistochemie. 2. Auf!., VEB Gustav Fischer Verlag, Jena 1973. HALBHUBER, K.-J., Methodische Bedingungen fUr den histochemischen Nachweis anorganischer Ionen mit besonderer Beriicksichtigung des anorganischen Orthophosphats. Erg. Expel'. :\1ed. (Berlin) 14, 1-102 (1973). HULSTAERT, C. E., KALICHARAX, D., and HARDOXK, M. J., Cytochemical demonstration of phosphatases in the rat liver by a cerium-based method in combination with osmium tetroxide and potassium ferro cyanide postfixation. Histoehemistry 78, 71-79 (1983). PEARSE, A. G. E., Histochemistry. Theoretical and Applied. 2nd Churchill, London 1968. REYNOLDS, E. S., The use of lead citrate at high pH as an electron opaque stain in electron micro· scopy. J. Cell BioI. 17, 208~212 (1963). ROBINSON, J. M., and KARNOVSKY, :\1. J., Ultrastructural localization of several phosphatases with cerium. J. Histochem. Cytochem. 31, 1197-1208 (1983). ZAJIC, G., and SCHACHT, J., Cytochemical demonstration of adenylate cyclase with strontium chloride in the rat pancreas. J. Histochem. Cytochem. 31, 25~28 (1983). ZIMMERMANN; N., and HALBHUBER, K.-J., Light microscopical localization of enzymes by means of cerium-based methods. 1. Detection of acid phosphatase by a new cerium-Icad.teclmique (Ce-Pb-method). Acta histochem. 76, 97-104 (1985). Authors' address: Prof. Dr. K.-J. HALBHUBER and Dr. N. ZIMMERMANN, Institute of Anatomy, Friedrich Schiller University, DDR - 6900 Jena, Teichgraben 7.
Acta histochem. Vol. 77
Plate IV
2
4
3
Halbhub er a nd Zimm erman n VEB GUSTAV FISCHER VERLAG JENA
Light microscopical localization of enzymes. II.
73
Explanation of Plate IV Fig. I. Alkaline phosphatase activity of the brush border of the epithelial cells in the tuhular system in the renal cortex. Only the brush border regions of the cells are stained. X 250. Fig. 2. Alkaline phosphatase demonstrated by means of the strontium-based method in the cortex of the rat kidney. Note the selective deep brown staining of the eellular brush border. X 250. Fig. 3. Coupling azo dye technique for the demonstration of alkaline phosphatase activity In the kidney. The brush borders of the tubular cells are strongly stained, but the staining is not so clearly marked off from the residual cell body as in the case of the cerium or strontium methods. X 250. Fig. 4. Cerium-based method for the demonstration of alkaline phosphatase; incubation without sodium-{J-glyeerophosphate (substrate control). The enzyme reaction is completely negative. X 160.