A novel quantitative histochemical assay to measure endogenoussubstrate concentrations in tissue sections. Fundamental aspects

Acta histochem. (Jena) 97, 409-419 (1995) Gustav Fischer Verlag Jena . Stuttgart· New York

Ada

hisllet.e.ia

A novel quantitative histochemical assay to measure endogenous substrate concentrations in tissue sections. Fundamental aspects 10landa M. Koopdonk-Kool and Comelis 1. F. Van Noorden Academic Medical Center, University of Amsterdam, Laboratory of Cell Biology and Histology, Meibergdreef 15, NL-1105 AZ Amsterdam, The Netherlands Accepted 16 June 1995

Summary A quantitative histochemical assay has been developed for measurement of endogenous substrate concentrations in cryostat sections using a colorimetric visualization technique. Model sections of frozen gelatin solutions with known concentrations of glucose-6phosphate (G6P) were sandwiched with a second cryostat section containing glucose-6phosphate dehydrogenase (G6PDH) and all other compounds (with the exception of G6P) that are necessary for the demonstration of G6PDH activity with a tetrazolium salt method. After 60 min of incubation, G6P was converted with concomittant formazan production. The amount of formazan generated was measured cytophotometrically and used as a parameter of the G6P concentration in the first section. A calibration graph was obtained with a high correlation coefficient, allowing the conversion of mean integrated absorbance values into absolute substrate concentrations. The method was highly reproducible, and the recovery of G6P was 85 ± 4% irrespective section thickness (4 - 20 11m) and G6P concentration (0.08 -1.6 mM) in the sections. The sensitivity of the tetrazolium-linked method appeared to be 100 IlM in 20 Ilm thick sections. This sensitivity enables the measurement of physiological substrate concentrations in tissue sections. Spatial resolution was approximately 150 11m, indicating a relatively high rate of diffusion of G6P during the reaction. The model study shows that the method described here allows the quantitative determination of substrate concentrations in tissue sections. These endogenous substrate concentrations are necessary for the calculation of local metabolic fluxes when determined in combination with local enzyme activities and kinetics, thus giving a more accurate reflection of in situ metabolic heterogeneity of tissues.

Key words: substrate measurements - glucose-6-phosphate tetrazolium salts - quantification - enzyme histochemistry

metabolic fluxes -

Introduction Fluxes are parameters of cellular metabolic activity and metabolic regulation under (patho)physiological conditions. In histochemistry, maximum cellular activities of Correspondence to: C. J. F. Van Noorden

410

1. M. Koopdonk-Kool and C. 1. F. Van Noorden

regulatory enzymes are usually measured to represent metabolic fluxes through a certain pathway (Van Noorden and Jonges, 1995 a, b). However, these measurements only provide quantitative information about maximum flux or capacity of a certain pathway and not about the actual metabolic flux in situ (Newsholme et al., 1980; Pette and Hofer, 1980). Fluxes are not only determined by enzyme activities, but also by concentrations of substrates, cofactors and (allosteric) effectors (Groen et al., 1982; Jungermann et al., 1982) apart from inhibitors. Therefore, concentrations of substrates are essential for the calculation of metabolic fluxes. Determination of in situ enzyme activity and kinetics with quantitative cytochemical methods has been used for some time to analyse metabolic heterogeneity of tissues and cells (for reviews, see Nakae and Stoward, 1995; Van Noorden and Jonges, 1995a, b). Metabolic parameters determined in vitro may not reflect in situ or in vivo metabolic conditions, because enzymes. may act differently in homogenates as used in vitro than in the cytoplasm which can be considered as a concentrated solution (Masters, 1981; Fulton, 1982; Aragon and Sols, 1991; Clegg, 1991). Mapping of substrate concentrations in tissues was introduced by Kogure and Alonso in 1978 who measured in situ ATP concentrations with a quantitative cytochemical method. Since then, several approaches have been developed to measure local substrate and/or cofactor concentrations such as microbiochemical analysis (Teutsch, 1985; Teutsch et al., 1995), the use of microlight guides and miniature oxygen electrodes (Matsumura and Thurman, 1983; Thurman and Kauffman, 1985), positron emission tomography (Schelbert, 1991; Messa et al., 1992), quantitative intravital microscopy (Toth et al., 1992; Piston et al., 1995), nuclear magnetic resonance spectroscopy (Schrader et al., 1993), immunohistochemistry (Ma et al., 1994) and bioluminescence (Paschen et al., 1981; Mueller-Klieser et al., 1988; Mueller-Klieser and Walenta, 1993; Kim et al., 1993). In the present study, it has been investigated whether a relatively simple quantitative histochemical assay using a colorimetrical visualization technique could be used for local substrate measurements in tissue sections. Validity of the method was investigated in a model system of cryostat sections of gelatin solutions containing known amounts of glucose-6-phosphate (G6P) on the basis of the criteria proposed by Stoward (1980; see also Van Noorden and Butcher, 1991). The method is based on conversion of G6P by glucose-6-phosphate dehydrogenase (D-glucose-6-phosphate: NADP+ oxidoreductase; G6PDH; E.C. 1.1.1.49) coupled to tetrazolium salt reduction resulting in production of formazan which in turn could be measured cytophotometrically. Application of the method for measuring endogenous substrate concentrations are published elsewhere (Geerts et al., 1994; Geerts, Jonker, Charles, Van Noorden and Lamers, submitted).

Materials and Methods Solutions of 0,0.077,0.154,0.308,0.462,0.616,0.770, 1.155 and 1.540 mM G6P (disodium salt; Boehringer, Mannheim, Germany) and 8070 (w/v) gelatin (Merck, Darmstadt, Germany) in 100 mM phosphate buffer (pH 7.45) were chilled in liquid nitrogen (Smith et aI., 1979; Paschen 1985). The material was stored at - 80 DC until further use. Sections (4 - 20 j.1m thick) were cut at a cabinet temperature of - 25°C on a Bright cryostat fitted with a rotary retracting microtome set at a low speed to ensure constant section thickness. The gelatin G6P-containing sections were picked up onto clean glass slides. A solution containing 100 mM phosphate buffer (pH 7.45), 8% (w/v) gelatin, 0.48 mM NADP (Boehringer), 4 mM MgCh, 5 mM sodium azide, 0.1 mM 1-methoxyphenazine methosulphate (1-methoxyPMS; Serva, Heidelberg, Germany), 5 mM tetranitroblue tetrazolium (tetranitro BT; Serva; dissolved in dimethyl formamide and ethanol; final concentration of each solvent, 11170) (Van Noorden and Frederiks, 1992) and 74 IV G6PDH/ml (Serva) was chilled in liquid nitrogen as well. Cryostat sections, 24 j.1m thick,

Measurement of substrates in tissue sections

411

////////////

glass slide Fig. 1. Schematic representation of the incubation method to detect substrate concentrations in tissue sections showing a cryostat section of a frozen gelatin solution with a known concentration of G6P (A) sandwiched between a glass slide and a second section containing G6PDH, NADP (but not G6P) and other compounds necessary for the tetrazolium-linked reaction to visualize G6PDH activity (B).

were cut and placed on top of the G6P section. As soon as the G6PDH-containing section was picked up onto the G6P-containing section (Fig. 1), the following reaction was started: 1. G6P + NADP+

G6PDH

~

6-phosphogluconolactone + NADPH

2. NADPH + I-methoxyPMS + tetranitro BT---> NADP + I-methoxyPMS + formazan L After 60 min of incubation at 37°C, the sections were mounted in glycerol and formazan production was measured with a Vickers M85a scanning and integrating cytophotometer. Preliminary experiments showed that after about 10 min the maximum amount of G6P molecules had been converted. The reaction was allowed to proceed for 60 min to ensure maximum conversion of the substrate. The above mentioned incubation conditions were varied as shown in Table 1 in order to optimize conversion of G6P. Measurements were made with a x 6.3 planachromatic objective (NA 0.20), a band width setting of 65, Table 1. Overview of experiments performed to improve recovery of G6P in gelatin cryostat sections and detected by the tetrazolium-linked method. The standard experimental conditions were: 20 l!m thick model sections containing 0.77 mM G6P covered with 24l!m thick G6PDH-containing gelatin sections, consisting of 100 mM phosphate buffer (pH 7.45), 80/0 gelatin, 0.48 mM NADP, 4 mM MgCh, 5 mM NaN), 0.1 mM I-methoxyPMS, 5 mM tetranitro BT and 74 IV G6PDH/ml. Incubation took place in dry air at 37°C. After 1 h sections were mounted in glycerol and measured cytophotometrically. Variations in incubation conditions

* *

use of 6 and 12l!m thick G6PDH-containing sections enzyme sections containing 18% PVA instead of 8% gelatin * enzyme sections containing 0.8 mM NADP, 5 mM MgCh, 0.32 mM l-methoxyPMS (according to Rieder et aI., 1978) * enzyme sections containing either 37, 148 or 222 IV G6PDH/ml * incubation in moist chambers * incubation in the presence of 18% PVA * incubation in distilled water * incubation of a soluble G6P-containing mixture with a soluble gelatin-containing enzyme mixture, subsequently measured spectrophotometrically * incubation of a soluble G6P-containing mixture with a soluble gelatin-containing enzyme mixture, subsequently frozen and measured cytophotometrically

412

1. M. Koopdonk-Kool and C. 1. F. Van Noorden

a mask with an effective diameter of 190 ltm and a scanning spot with an effective diameter of 3.2ltm. Tetranitro BT-formazan production was measured at the wavelength of 557 nm. For each G6P concentration, 75 absorbance readings (machine units) were taken; 5 cytophotometric measurements were made in each of 5 different model sections incubated in 3 separate experiments. Mean integrated absorbance (MIA) ± standard deviation (SO) was calculated for each G6P concentration using a calibration curve (Van Noorden and Frederiks, 1992). Absorbance values were converted to mM G6P after subtracting absorbance values of sections of gelatin-containing solutions lacking G6P (controls) by recourse of the molar extinction coefficient (k) of tetranitro BT-formazan being 19000lmol- 1 cm- 1 at 557nm (Van Noorden and Frederiks, 1992). To study precision of localization by determining spatial resolution, 8 ltm thick model sections of gelatin solutions containing different G6P concentrations were prepared as shown in Fig. 2. Linescans were made using a x 25 objective (NA 0.50), a mask with an effective diameter of 10 ltm and a scanning spot with an effective diameter of 0.8 ltm. Spatial resolution was calculated from determinations of the length of scans over which changes of absorbance occurred at borders of the gelatin solutions containing different concentrations of G6P. Statistical analysis was performed when appropriate using the Student's t-test. The level of significance was taken as 0.05.

a.

B

A

-B b.

A-

Fig. 2. Schematic representation of two models used to determine spatial resolution of the visualization method to detect in situ G6P concentrations. Measurements were taken along a line as indicated by the bar. a. Cryostat section containing 2 areas with different concentrations of G6P. A, 0.077 mM; B, 7.7 mM. b. Cryostat sections containing either 1.54 mM (A) or 6.16 mM G6P (B) partly covering each other.

Results A linear relationship with a high correlation coefficient (0.999) was found between G6P concentrations in 20 J.l.m thick model sections and formazan production (Fig. 3, Table 2). The intra-experimental variation appeared to be approx. 6070 (data not shown), whereas the inter-experimental variation was approx. 12%. Especially the low intra-experimental variation indicates a high reproducibility of the tetrazolium-linked method. Formazan was homogeneously precipitated throughout the sections. Because absorbance values of sections containing 0, 0.077 and 0.154 mM G6P were found to be signifcantly different, the sensitivity limit of the method was taken as 100 J.l.M. The recovery of G6P incorporated in the gelatin sections was 85 ± 4% irrespective the concentration of G6P (Table 2). The relationship between formazan production

Measurement of substrates in tissue sections

413

MIA X 10 3

60 40 T

T

...

20

0.4

...

0.8

1.2

G6P (mM) Fig. 3. Relationship between mean integrated absorbance (MIA) at 557 nm and the G6P concentration in 20 Jlm thick cryostat sections (Y = 0.032X + 0.0036, r = 0.999). Each value represents the mean of 5 measurements in 5 different model sections in each of 3 separate experiments. Standard deviations are given at all G6P concentrations.

Table 2. Mean integrated absorbance at 557 nm (MIA ± SD), recovered G6P concentration (±SD) and percentage recovery (± SD) in 20 Jlm thick sections containing different substrate concentrations, as visualized with the tetrazolium-linked method. Each value represents the mean of 5 measurements in 5 different model sections in each of 3 separate experiments G6P concentration in gelatin solution (mM)

MIA ± SDxl0 J

0.000 0.077 0.154 0.308 0.462 0.616 0.770 1.155 1.540

4.79 6.21 8.65 12.47 17.68 24.03 28.58 40.05 54.61

Mean 0,70 recovery

± ± ± ± ± ± ± ± ±

0.66 0.11 1.29 1.26 2.58 2.82 2.89 6.63 9.34

G6P concentration recovered ± SD (mM)

o

0.070 0.134 0.234 0.372 0.539 0.658 0.960 1.343

± ± ± ± ± ± ± ±

0.001 0.020 0.024 0.054 0.063 0.067 0.159 0.230

0,70

Recovery ± SD

o 91 87 76 80 87 86 83 87

± ± ± ± ± ± ±

2 13

8 12 10

9

14 ± 15

85 ± 4

and section thickness showed proportionality, also independently of the G6P concentration (Fig. 4). Fig. 5 shows the relationship between the G6P concentration dissolved in the gelatin solutions and the amount of G6P which could be recovered using the tetrazolium-linked method. The curves are straight lines with similar slopes for all section thicknesses except for 4 l!m thick sections. Approximately 87 ± 40/0 recovery could be achieved, irrespective of section thickness. In the case of 4l!m thick sections, formazan was not produced in significant amounts when G6P concentrations ::5 0.6 mM were used.

414

1. M. Koopdonk-Kool and C. 1. F. Van Noorden

MIA x 10 3

60

40

20

4

8

20

12

thickness (urn) Fig. 4. Relationship between mean integrated absorbance (MIA) at 557 nm and thickness of cryostat sections containing different concentrations of G6P. • , 0 mM G6P; b., 0.077 mM G6P; 0, 0.308 mM G6P; 0,0.616 mM G6P; _, 1.540 mM G6P. Each value represents the mean of 5 measurements in 5 different model sections.

G6P (mM) recovered

1.5

0.6 0.3 0.3 0.6

1.5

G6P (mM) dissolved Fig. 5. Relationship between G6P concentration which was recovered using the tetrazolium-linked method and the amount of G6P dissolved in the gelatin solutions of sections with various thicknesses. +, 4 11m; b., 811m (Y = 0.81X- 0.016); 0, 12l1m (Y = 0.91X - 0.052); +, 20 11m (Y = 0.89X - 0.047). The slopes represent the percentage of G6P recovered.

Measurement of substrates in tissue sections

415

To investigate the reason of the constant difference of approx. 15010 between G6P concentrations dissolved and G6P concentrations recovered, experiments were performed as presented in Table 1. All variations did not result in a recovery higher than 85%. In a final experiment, the reaction was performed in solution and again only 85 % of the G6P was recovered. Precision of localization as determined in model sections with two different G6P concentrations (see Fig. 2a) is demonstrated in Figs. 6 and 7. As is shown in Fig. 6, there is

Fig. 6. Photomicrograph of a model cryostat section (8 Ilm thick) containing 2 areas with different G6P concentrations as schematically shown in Fig. 2a and visualized with the tetrazolium salt method. Precipitation of tetranitro BT-formazan at the border of the 2 areas is diffuse.

MIA

X

103

80 60

40 20

-500

0

500

Fig. 7. Profile of mean integrated absorbance (MIA) at 557 nm along a scan line in a model section containing 0.077 and 7.7 mM G6P, respectively, as shown in Fig. 2a.

1. M. Koopdonk-Kool and C. 1. F. Van Noorden

416

a diffuse border between the areas with different G6P concentrations. Fig. 7 illustrates that approx. 150 ~m of the area containing the highest concentration of G6P shows diminished formazan precipitate, whereas a zone of 150 ~m of the area containing the lowest G6P concentration demonstrates increased formazan precipitate. Because thawing and mixing may have occurred at the borders when the second G6P-containing gelatin solution was poured into the first frozen gelatin solution, another model was used as well. Two cryostat sections containing different concentrations of G6P were attached to a glass slide at a temperature of - 25°C as shown in Fig. 2 b, with a section of a frozen enzyme solution on top. In this way, diffusion of G6P molecules was prevented, because thawing and condensing of one section on top of the other did not occur. Upon thawing at room temperature, the reaction was initiated, thus allowing the G6P molecules and intermediates to be converted immediately. The results obtained with this model are demonstrated in Fig. 8. Again, a zone of approx. 150 ~m in each area showed effects of diffusion. Therefore, it was concluded that the spatial resolution in the gelatin sections was 150 ~m.

MIA

X

10 3

120

80

40

-500

o

500

thickness ( urn) Fig. 8. Profile of mean integrated absorbance (MIA) at 557 nm along a scan line in model sections shown in Fig. 2b.

Discussion In the present study, the method originally developed by Kogure and Alonso (1978) has been adapted to measure substrate concentrations in tissue sections using a histochemical colorimetric instead of a luminescence visualization technique. There are several advantages of the technique based on tetrazolium salt reduction. First of all, the method is relatively simple and only requires a cryostat and a cytophotometer or image analysing system. Second, the method can be applied to a large range of substrates and third, sections can be kept after staining without loss of staining quality. The calibration procedure showed a linear relationship between mean integrated absorbance and G6P concentration (see Fig. 3). In Thble 2 it is shown that in 20 ~m thick sections approximately 85070 of the G6P dissolved in the gelatin solutions could be

Measurement of substrates in tissue sections

417

demonstrated. Reasons for the fact that not all G6P could be measured are not really clear yet. Fig. 4 shows a linear relationship between mean integrated absorbance and thickness of model sections for every G6P concentration, indicating a recovery of constant proportion for every thickness. This is further illustrated in Fig. 5, where, except for 4 ~m thick sections, the G6P concentration recovered is proportional to the concentration dissolved in the gelatin solutions. As the slopes of the curves represent the percentage of G6P which could be recovered, it can be concluded from Thble 2 and Fig. 5 that the percentage recovery is 85 ± 4070, independent of section thickness up to 20 ~m. Apparently, diffusion of G6P into the enzyme-containing sections is not a limiting factor, irrespective of the distance between substrate and enzyme molecules. Several experiments were performed to optimize the incubation conditions for the enzyme reaction (see Table 1), but the maximum amount of G6P that could be demonstrated always appeared to be approx. 85%. In solution, only 85% was recovered as well. Nevertheless, the procedure used for calibration is valid and can be used for converting absorbance values into absolute units of substrate concentrations, taking into account 85 % of recovery. This is in agreement with studies of Ochoa et al. (1950) and Slater (1953) who found a recovery of approx. 80% when determining substrate concentrations spectrophotometrically in solutions. The sensitivity of the method is high. A G6P concentration of 100 ~M could be detected in 20 J.!m thick sections. Thus this technique should enable for example the determination of G6P concentrations in liver sections to discriminate between periportal and pericentral zones of liver lobules, where the physiological G6P concentration as biochemically determined can be as low as 100 J.!M (Jungermann et al., 1982). The spatial resolution as determined in gelatin sections was at best 150 J.!m, which is considerably lower than that found by Mueller-Klieser et al. (1988), which was between 10 -100 J.!m. However, the models used for estimation of the spatial resolution may not be fully appropriate. In contrast to tissue sections, diffusion in gelatin model sections occurs freely as there are no barriers, such as cell membranes. Thus the spatial resolution is probably underestimated and localization could be better in tissue sections. Experiments to determine glutamate concentrations in rat liver cryostat sections proved that this is indeed the case (Geerts et al., 1994; Geerts, Jonker, Charles, Van Noorden and Lamers, submitted). The present model study indicates that in situ measurements of substrate concentrations with the quantitative histochemical analysis can be generally applied. So far, only a limited number of substrates could be analyzed in situ. In principle, all substrates which can be directly linked to a dehydrogenation reaction, and which are present in tissues at concentrations higher than 100 J.!M can be measured in situ. For substrates present in lower concentrations, enzymatic cycling according to Lowry and Passonneau (1972) has to be performed to increase formazan production. Also substrates which are not directly converted by a dehydrogenase may be measured by using a multistep reaction (Lojda et al., 1979; Frederiks et al., 1988; Van Noorden and Frederiks, 1992). The measurement of in situ substrate concentrations enables a further investigation of concepts such as metabolic heterogeneity, which are largely based on the distribution of enzyme concentrations or the capacity of enzymes (Jungermann and Katz, 1989; Van Noorden and Jonges, 1995 b). There is evidence that the distribution of enzymes is not the true reflection of metabolic activity and in situ analysis of metabolic fluxes by the determination of kinetic parameters of enzymes and the actual local concentrations of their substrates might yield further information on metabolic heterogeneity. For example, glucose-6-phosphatase has been shown to have a higher Vmax in periportal areas than in pericentral areas of the rat liver lobules, whereas the KM values were higher as well (Jonges et al., 1992). Therefore, it also depends on local substrate concentrations whether the glucose production by glucose-6-phosphatase is different in the two areas of such a metabolically heterogeneous tissue as the liver (Van Noorden and Jonges, 1995b).

418

1. M. Koopdonk-Kool and C. 1. F. Van Noorden

In conclusion, it appears possible to measure substrate concentrations in sections with a quantitative histochemical method using a colorimetric visualization technique. The calibration procedure allows conversion of the absorbance as measured cytophotometrically into absolute substrate concentrations.

Acknowledgements

The authors gratefullly acknowledge the editorial criticism of Prof. Dr. Jan James and Dr. Wilma Frederiks, the preparation of the manuscript by Mrs. Trees Pierik and the preparation of the figures by Mr. Jan Peeterse.

References Aragon JJ, and Sols A (1991) Regulation of enzyme activity in the cell: effect of enzyme concentration. FASEB J 5: 2945 - 2950 Clegg JS (1991) The organization of aqueous compartments in cultured animal cells. In Tigyi J, Kellermayer M, Hazlewood CF (Eds) The physical aspect of the living cell. Akademiai Kiad6, Budapest, pp129-144 Frederiks WM, Marx F, and Van Noorden CJF (1988) Quantitative histochemistry of creatine kinase in rat myocardium and skeletal muscle. Histochem J 20: 624-628 Fulton AB (1982) How crowded is the cytoplasm? Cell 30: 345 - 347 Geerts WJC, Jonker A, Charles R, Lamers WH, and Van Noorden CJF (1994) In situ measurement of glutamate concentrations in rat liver. Proc Roy Micr Soc 29: 242 Groen AK, Van der Meer R, Westerhoff H, Wanders RJA, Akerboom TPM, and Tager JM (1982) Control of metabolic fluxes. In Sies H (Ed) Metabolic compartmentation, Academic Press, New York, pp9-37 Jonges GN, Van Noorden CJF, and Lamers WH (1992) In situ kinetic parameters of glucose-6-phosphatase in the rat liver lobulus. J Bioi Chern 267: 4878 - 4881 Jungermann K, and Katz N (1989) Functional specialization of different hepatocyte populations. Physiol Rev 69: 708 -764 Jungermann K, Heilbronn R, Katz N, and Sasse D (1982) The glucose/glucose-6-phosphate cycle in the periportal and perivenous zone of rat liver. Eur J Biochem 123: 429-436 Kim JK, Haselgrove JC, and Shapiro 1M (1993) Measurement of metabolic events in the avian epiphyseal growth cartilage using a bioluminescence technique. J Histochem Cytochem 41: 693 -702 Kogure K, and Alonso OF (1978) A pictorial representation of endogenous brain ATP by a bioluminescent method. Brain Res 154: 273 - 284 Lojda Z, Gossrau R, and Schiebler TH (1979) Enzyme histochemistry. A laboratory manual. Springer, Berlin, Heidelberg, New York Lowry OH, and Passonneau JV (1972) A flexible system of enzymatic analysis. Academic Press, New York, London Ma N, Aoki E, and Semba R (1994) An immunohistochemical study of aspartate, glutamate, and taurine in rat kidney. J Histochem Cytochem 42: 621-626 Masters CJ (1981) Interactions between soluble enzymes and subcellular structure. CRC Crit Rev Biochem 11: 105 -143 Matsumura T, and Thurman RG (1983) Measuring rates of O2 uptake in periportal and pericentral regions of liver lobule: stop-flow experiments with perfused liver. Am J Physiol 244: G656 - G659 Messa C, Choi Y, Hoh CK, Jacobs EL, Glaspy JA, Rege S, Nitzsche E, Huang SC, Phelps ME and Hawkins RA (1992) Quantification of glucose utilization in liver metastases: parametric imaging of FDG uptake with PET. J Comp Ass Tomography 16: 684-689 Mueller-Klieser W, and Walenta S (1993) Geographical mapping of metabolites in biological tissue with quantitative bioluminescence and single photon imaging. Histochem J 25: 407 -420 Mueller-Klieser W, Walenta S, Paschen W, Kallinowski F, and Vaupel P (1988) Metabolic imaging in microregions of tumors and normal tissues with bioluminescence and photon counting. J Nat Cancer Inst 80: 842 - 848

Measurement of substrates in tissue sections

419

Nakae Y, and Stoward PJ (1995) The kinetics of enzymes in situ, with special reference to lactate and succinate dehydrogenase. Histol Histopathol 10: 463 -479 Newsholme EA, Crabtree B, and Zammit VA (1980) Use of enzyme activities as indices of maximum rates of fuel utilization. In Evered D, O'Connor M (Eds) Trends in enzyme histochemistry and cytochemistry. Excerpta Medica, Amsterdam, pp 245 - 258 Ochoa S, Salles JBV, and Ortiz PJ (1950) Biosynthesis of dicarboxylic acids by carbon dioxide fixation. III. Enzymatic synthesis of I-malic acid by reductive carboxylation of pyruvic acid. J Bioi Chern 187: 863 -874 Paschen W (1985) Regional quantitative determination of lactate in brain sections. A bioluminescent approach. J Cereb Blood Flow Metab 5: 609-612 Paschen W, Niebuhr I, and Hossmann K-A (1981) A bioluminescence method for the demonstration of regional glucose distribution in brain slices. J Neurochem 36: 513 - 517 Pette D, and Hofer HW (1980) The constant proportion enzyme group concept in the selection of reference enzymes in metabolism. In Evered D, O'Connor M (Eds) Trends in enzyme histochemistry and cytochemistry. Excerpta Medica, Amsterdam, pp 231 - 244 Piston DW, Masters BR, and Webb WW (1995) Three-dimensionally resolved NAD(P)H cellular metabolic redox imaging of the in situ cornea with two-photon excitation laser scanning microscopy. J Microsc 178: 20-27 Rieder H, Teutsch HF, and Sasse D (1978) NADP-dependent dehydrogenases in rat liver parenchyma. I. Methodological studies on the qualitative histochemistry of G6PDH, 6PGDH, malic enzyme and ICDH. Histochemistry 56: 283 - 298 Schelbert HR (1991) Positron emission tomography for the assessment of myocardial viability. Circulation 84: 1122-1131 Schrader MC, Eskey CJ, Simplaceanu V, and Ho C (1993) A carbon-13 nuclear magnetic resonance investigation of the metabolic fluxes associated with glucose metabolism in human erythrocytes. Biochim Biophys Acta 1182: 162 - 178 Slater EC (1953) Spectrophotometric determination of fructose-l ,6-diphosphate, hexosemonophosphates, adenosinetriphosphate and adenosinediphosphate. Biochem J 53: 157 -167 Smith MT, Loveridge N, Wills ED, and Chayen J (1979) The distribution of glutathione in the rat liver lobule. Biochem J 182: 103 - 108 Stoward PJ (1980) Criteria for the validation of quantitative histochemical enzyme techniques. In Evered D, O'Connor M (Eds) Trends in enzyme histochemistry and cytochemistry. Excerpta Medica, Amsterdam, pp 11 - 31 Teutsch HF (1985) Quantitative histochemical assessment of regional differences in hepatic glucose uptake and release. Histochemistry 82: 159 -164 Teutsch HF, Goellner A, and Mueller-Klieser W (1995) Glucose levels and succinate and lactate dehydrogenase activity in EMT6/Ro tumor spheroids. Eur J Cell Bioi 66: 302 - 307 Thurman RG, and Kauffman FC (1985) Sublobular compartmentation of pharmacologic events (SCOPE): Metabolic fluxes in periportal and pericentral regions of the liver lobule. Hepatology 5: 144-151 Toth A, Tischler ME, Pal M, Koller A, and Johnson PC (1992) A multipurpose instrument for quantitative intravital microscopy. J Appl Physiol 73: 296 - 306 Van Noorden CJF, and Butcher RG (1991) Quantitative enzyme histochemistry. In Stoward PJ, Pearse AGE (Eds) Histochemistry. Theoretical and applied. Vol. 3, 4th ed, Churchill Livingstone, Edinburgh, London, Melbourne, New York, Tokyo. pp 355 -432 Van Noorden CJF, and Frederiks WM (1992) Enzyme Histochemistry. A laboratory manual of current methods. Oxford University Press, Oxford Van NoordenCJF, and JongesGN (1995a) Analysis of enzyme reactions in situ. Histochem J 27: 101-118 Van Noorden CJF, and Jonges GN (1995 b) Heterogeneity of kinetic parameters of enzymes in situ in rat liver lobules. Histochemistry 103: 93 - 101