- Email: [email protected]
Expression of muscarinic binding sites in primary human brain tumors
DevelopmentalBrain Research, 14 (1984) 61-70 Elsevier
61
BRD 50044
Expression of Muscarinic Binding Sites in Primary Human Brain Tumors DAVID GURWITZ I, NISSIM RAZON 2, MORDECHAI SOKOLOVSKY 1and HERMONA SOREQ 3
1Department of Biochemistry, George S. Wise Facultyfor Life Sciences, Tel A viv University, Tel A viv; 2Department of Neurosurgery, Ichilov Hospital, Tel A viv Medical Center, Tel A viv and 3Department of Neurobiology, Weizmann Institute of Science, Rehovot, (Israel) (Accepted January 17th, 1984)
Key words: muscarinic receptors - - glioblastoma multiforme - - astrocytoma - - meningioma - acetylcholinesterase- - malignancy --binding affinity
The expression of muscarinic binding sites was examined in a collection of primary brain tumors of different cellular origins and various degrees of dedifferentiation, as compared to control specimens. Eleven gliogenous tumors were examined, all of which contained substantial amounts of muscarinic binding sites. Most of the other tumor types examined did not display detectable binding of [3H]Nmethyl-4-piperidyl benzilate ([3H]4NMPB). Scatchard analysis indicated the existence of homogenous antagonist sites in both normal forebrain and gli0blastoma multiforme, with K d values of 1.2 nM and 0.9 nM, respectively. The density of muscarinic binding sites varied between tumors from different patients, and also between specimens prelevated from different areas of the same tumor. This variability, as well as the average density of binding sites, appeared to be larger in highly malignant tumors than in less malignant ones. In contrast, the density of muscarinic receptors from control specimens was invariably high, but within the same order of magnitude. To test whether the muscarinic binding activity in the brain tumors is correlated to other cholinoceptive properties, cholinesterase activity was also examined. Individual data for density of [3H]4NMPB binding sites were then plotted against corresponding values of cholinesterase activity. The pattern of distribution of these values was clearly different in tumor specimens, when compared to that observed in samples derived from non-malignant brain. Our observations indicate that human brain cells of gliogenous origin are capable of expressing muscarinic binding sites, and that, if a correlation exists between muscarinic receptors and cholinesterase levels in gliogenous tumors, it differs from that of non-malignant brain tissue. INTRODUCTION Biochemical studies on binding of muscarinic ligands have been extensively carried out using brain tissue as a m o d e l system (reviewed recently by Soko10vsky et al. 26) and special attempts were m a d e to correlate the density of muscarinic receptors ( m A C h R ) in the h u m a n brain to the functioning of cholinergic circuits. D e c r e a s e s in muscarinic binding were shown in the corpus striatum of H u n t i n g t o n ' s disease patients6,8,30. Biochemical analysis of the muscarinic receptors has then been carried out in normal human brain, as c o m p a r e d with those in the choreic brain. It was found that the n u m b e r of receptors, but not their affinity properties, is changed in the caudate nucleus and p u t a m e n of Huntington chorea patients. This decrease a p p e a r e d to be correlated to r e d u c e d activity of cholineacetyltransferase3t. H o w e v e r , the w e l l - d o c u m e n t e d decreases in
the density of muscarinic receptors in the choreic brain a p p e a r to represent an exceptional case. The regional distribution of muscarinic binding sites was found to remain unchanged in the brain of A l z h e i m er's-type d e m e n t i a patients, in spite of the evidence for extensive loss of enzymes of the cholinergic system4, 29. In the frontal cortex of elderly people, muscarinic receptor binding sites have been r e p o r t e d to remain unchanged31. W h e n morphological analysis was carried out in parallel, a certain decrease in the density of muscarinic receptors was found in specimens with normal morphological evidence, but not in those derived from brains with senile degenerations 32. Even in cholinergic-innervated regions such as the anterior parts of the human h i p p o c a m p u s , the age-related decrease in the density of muscarinic receptors was found to be r a t h e r limited 19. In certain cases, the expression of muscarinic binding properties thus remains u n i m p a i r e d in the human
Correspondence: H. Soreq, Dept. of Neurobiology, Weizmann Institute of Science, Rehovot 76100, Israel.
62 brain, under conditions where an apparent damage has been caused to cholinergic neurons. This indicates that brain cell types other than such neurons may also express muscarinic binding properties. The indication that non-neuronal brain cells contain muscarinic receptors has been supported by experiments with human 1321 NI astrocytoma ceils in culture. Agonist-induced stimulation of muscarinic receptors in these cells was shown to result in a decrease in cyclic AMP accumulation 7, via the activation of a phosphodiesterase 17, affecting both synthesis and degradation of cyclic AMP, and reducing the density of muscarinic receptors~8. However, to the best of our knowledge, there are no reports regarding the level and properties of muscarinic receptors in vivo, in cells of gliogenous origin. Studying these receptors in their primary in vivo tissue source is particularly important since the density of muscarinic binding sites in cloned cells has been shown to decrease significantly with cell passage2L In an earlier study, we attempted to correlate muscarinic receptors to specific in vivo cellular origins in the developing rodent cerebellum. The cell type composition in the cerebellum is relatively simple, and various malformations are available which are deficient in particular cell types. However, in spite of these advantages, we could not, in that study, reveal whether cerebellar cells of non-neuronal origin express muscarinic binding sites 27. To address this issue directly, we therefore examined the in vivo levels and biochemical properties of muscarinic binding sites in human primary brain tumors, which provide a tissue source derived essentially of a single cellular origin 23. Various tumors, of neuroectodermal and mesenchymal origins 15, were screened for this purpose. Our observations reveal that gliogenous intracranial tumors contain substantial levels of muscarinic binding sites, of similar affinity properties to those observed in non-malignant brain tissue. Furthermore, our findings indicate that in gliogenous tumors, the interrelationship between the expression of muscarinic receptors and cholinesterase is different than that of the non-malignant brain. MATERIALS AND METHODS
Chemicals Acetylcholine chloride and the detergent Triton
X-100 were obtained from Sigma. [3H]Acetylcholine iodide (90 mCi/mmol) was obtained from New England Nuclear. The potent muscarinic antagonist [3H]N-methyl-4-piperidyl benzilate ([3H]4NMBP, 69 Ci/mmol) and unlabeled muscarinic ligands were as described 13. All salts and buffers were Analar grade.
Brain tissues Specimens from brain tumors were invariably preievated at the time of surgery. Out of the 16 tumor samples examined, 8 were from the frontal lobe, 5 were in the temporal lobe, 2 in the parietal region and one in the cerebellum (Table I). Control specimens, all from the frontal tip of patients who died from diseases not related to the central nervous system, were generally prelevated 24 h postmortem (after about 24 h at 4 °C). Whenever possible prelevation at surgery was employed for control samples as well. All tissues were frozen in liquid nitrogen following prelevation, and were stored at --70 °C until use. Histological and pathological preparations Tissue excised at the time of prelevation was fixed in 4% buffered formaldehyde for 15 h at room temperature, washed twice with PBS and embedded in paraffin. Five/~m sections were prepared, mounted on glass slides and stained with haematoxilin-eosin. Histological characteristics were defined according to the classification accepted by the WHO 34. Summary of the pathological findings for each of the tumor specimens employed is presented in Table I, as confirmed independently by a senior pathologist. Biochemical determinations Thawed tissue samples, about 50 mg, were suspended in 0.01 M Tris, pH 7.4 and homogenized as described 28. DNA content of these tissue samples was determined as described by Leyva and KelleyTM. Protein content was determined on the solubilized tissue samples according to Lowry et a1.16. Muscarinic binding measurements Tissue samples were weighed and homogenized in 0.32 M sucrose, as previously described 13. Aliquots of homogenates (50/A) were incubated at 25 °C for 30 min in 2 ml buffer, containing 118 mM NaCl, 5 mM KCI, 1 mM MgC12, 10 mM glucose and 25 mM Tris-HCl (pH = 7.4 at 25 °C). For antagonist
63 ~
~
ae
' ~
~ ~us.~
"~
~ ,, .~.
~
~:.~
•.- ~ .;~ .~
.~.~.~,~.~
&
•c , . "~
~' O
O
O
O
,.,Q
~ E E E E E E E E
c~
~Q <
~ ~"
~ ~'~ O O
~ ~ ~ O
O
64 binding studies, 8 mM [3H]4NMBP were included. At this concentration, muscarinic receptors are nearly saturated ~3. In some experiments, complete binding isotherms for [3H]4NMBP were studied, using various ligand concentrations. For agonist binding studies, the buffer contained 2 nM [3H]4NMBP, with or without 250/.tM carbamylcholine, and the inhibition of binding of the labeled antagonist by the unlabeled agonist was determined. Incubation was terminated by rapid filtration through GF/C filters (Whatman, 25 mm diameter), followed by 4 × 3 ml washes with ice-cold buffer. Radioactivity was determined using a scintillation cocktail (Lumac, Hydroluma) and a scintillation spectrometer (Packard, Tricarb 300). Assays were carried out in triplicates, and specific binding was defined as total minus non-specific, i.e. binding in the presence of 1 ~M unlabeled atropine, Cholinesterase measurements
Tissue samples were weighed and homogenized in 0.9 ml of ice-cold 1 M NaC1-0.05 M Tris chloride, pH 7.5, using 10 strokes of a Teflon-glass homogenizer driven by a Heidolf Motor at half speed (1000 rpm). A 0.1 ml aliquot of aqueous Triton X-100 (10% v/v) was added to each homogenate, and homogenization was repeated using a further 10 strokes. The homogenates were centrifuged for 5 min in an Eppendorf $412 centrifuge and the supernarant was reserved. Cholinesterase activity was performed according to the general procedure of Johnson and RusselP 1 and as described 2s. Reaction mixtures contained 20/~1 of homogenate supernatant, 10 /~1 of 1.2 M NaC1-0.5 M Tris chloride, pH 7.4, 30 ~1 of 10 mM radioactive acetylcholine (made by diluting [3H]acetylcholine into 10 mM acetylcholine chloride so as to yield ca. 80,000 cpm total counts in each sample), and water to a final volume of 100 ~tl, made up in a 5 ml scintillation vial. For BW284C51 inhibition measurements, an aliquot of a stock solution of the inhibitor (1,5-bis(4-allyl dimethylammoniumphenyl pentan-3-one dibromide, Sigma) was added so as to yield a final concentration of 10-5 M 2s. Assays were run at 25 °C, for 30-120 rain, and the reaction was terminated by addition of 100/~1 of stopping mixture (1 M chloroacetic acid-0.5 M NaOH-2.0 M NaC1). Scintillation Cocktail (4 ml of 90% xylene-10% isoamylalcohol con-
taining 10% 'Xyloflour') was then added. After vigorous shaking, samples were counted in a Packard Model 3390 Tricarb Liquid Scintillation Spectrometer. All measurements were in linear range with respect to time and protein concentration. RESULTS Homogenates from primary brain tumors were tested for the presence of specific muscarinic binding sites. Out of the various types of tumors which were examined, substantial levels of [3H]4NMBP binding sites were determined in all of the tumors derived from gliogenous origin, such as glioblastoma multiforme and grade I and |I astrocytomas. In both of these tumor types, binding was inhibited by 20--50% in the presence of 250/~M carbamylcholine. Low levels of muscarinic receptors could also be detected in 2 out of 5 meningiomas that were examined. These were, in both cases, recurrent tumors (Table I). High variability in the density of muscarinic receptors was observed between different glioblastoma tumors. In order to exclude the possibility that the variation in the level of receptors could reflect different numbers of cells in the tissue samples, the amount of DNA was measured in most of the samples tested. The total amount of D N A varied between 0.4 and l.(I mg/g of wet weight tissue, and there was no correlation to the level of receptor. In addition, the content of protein was determined, and was found to range between 70-140 mg/g tissue in 80% of the samples. Moreover, the general size distribution of the total tissue proteins appeared by SDS-polyacrylamide gel analysis to be rather similar in all of the samples that were tested (not shown). Thus, the variations in the level of muscarinic binding sites cannot be explained by selectively high in situ proteolysis in part of the samples. The variability in the density of muscarinic binding sites was most conspicuous in samples derived from the peripheral regions of gliogenous tumors. These displayed an average level of 15.5 + 14.0 (n = 4, see Table I). In samples derived from core regions in such tumors, the average density of receptors was significantly lower, at a level of 5.7 _+ 6.2 (n = 15, Table I). This difference could probably be attributed to the cell type diversity between different areas in gliogenous tumors. Fig. 1 displays such morpho-
65
• 30- / l
;
"S 20i- /
', /
'%.
'/ '~,
\'~
l
\'~ '~
I
",%
35-45 25-35 ~5-z5 545
o-5
0
mAChR, fmol/mg tissue Fig. 1. Top: differences in histological appearance between various areas in glioblastoma multiforme tumors. Prelevation of specimens, fixation and staining were as described in Materials and Methods. Note the different histological appearance of tissue from the periphery (left) and core (right) regions. Bottom: distribution of mAChR-density in gliogenous tumors and in control brain. Tissue samples were prelevated from the core regions of gliogenous tumors and from control brain tissue (Tables I and II). Preparation of homogenates and binding of [3H]4NMBP were as described in Materials and Methods. Percent of specimens which displayed specific mAChR densities is presented.
distribution patterns of individual values of m A C h R densities in control specimens, as compared with those obtained with core samples of gliogenous tumors. It is apparent that the two patterns are rather parallel, with a clear difference in peak values. There appeared to be no correlation between the density of receptors in non-malignant brain samples and the age of the patients, in agreement with Wastek and Yamamura 31. Scatchard analysis of [3H]4NMBP-binding indicated, for preparations from both non-malignant brain and glioblastoma multiforme tumors, the existence of homogenous antagonist sites. The dissociation constants for [aH]4NMBP were found to be 1.2 nM and 0.9 nM for control and glioblastoma samples, respectively (Fig. 2). These values are very close to those detected for [3H]4NMBP-binding in the mouse brain 13. Furthermore, competition experiments indicated similar apparent affinity values for the binding of the agonist carbamylcholine towards the muscarinic receptors in non-malignant brain and gliogenous tumors. The inhibition asserted by 250 ,uM carbamylcholine on the binding of 2 nM [3H]4NMBP was found to be 35.2% _+ 6.4% and 35.5% _+ 9.7% for control (n = 5) and gliogenous (n = 10) samples, respectively (Tables I and II). In an attempt to correlate the muscarinic site properties with the clinical stage of the glioblastoma tu-
[•
nn Q_
~ E
z~ , E IC o
o.
logical differences between the periphery and core areas of a single glioblastoma tumor. In the periphery, hyperplasia of endothelial cells, rosettes of undifferentiated glial cells and pseudopallisades are most conspicuous. In contrast, the core region mainly includes neoplastic, hyperchromatic quiescent cells and some necrotic areas. This figure also reveals the
I
o ~ Control o ~ 1 2 nM
==20 nn ~
I
0
X
Glioblostomo ~ o
X, I0
20
\
50
40
Specific3H-4NMPBBound(fmo[es/mgtissue) Fig. 2. Affinity properties of muscarinic binding sites from nonmalignant brain tissue and from glioblastoma multiforme. Scatchard plot of specific [3H]4NMBP binding to homogenates of non-malignant, control brain tissue (O, samples 3, Table II) and of glioblastoma multiforme (0, samples 6, Table I) is presented. The K d values were 1.2 nM and 0.9 nM for the control and the tumor samples, respectively.
"FABLE II
Control forebrain samples: description, muscarinic receptors and cholinesterase levels MI, myocardial infarct
Sample
Age~Sex
Disease
Muscarinic binding sites (mA ChR) Carbamylcholine Density inhibition (%) (fmol/mg tissue)
Cholinesterase (ChE) specific activity, (nmol/min/g/tissue)
la lb 2 3 4a 4b 5a 5b 6 7 8 9 10 l1 12 13 14 15 16
71/M 71/M 55/F 66/17 66/M 66/M 66/F 66/F 55/M 7 l/M 74/F 64/M 70/M 81/M 1/M 78/M 54/M 47/M 70/M
Neurosyphilis died of MI Neurosyphilis died of MI Sepsis MI Pneumonia Pneumonia Pulmonary embolism Pulmonary embolism MI B rain infarction MI MI Pulmonary edema Pancreatic carcinoma Cardiac malformation Bronchopneunomia Brain edema MI + sepsis MI
14.2 19.0 23.2 35.9 33.0 17.0 14.0 8.5 27.0 21.0 26.0 32.0 27.0 22.0 19.0 37.0 25.0 28.0 29.0
410 n.d. 303 235 172 n.d. 351 n.d. 253 87 n.d. n.d. 367 179 296 68 128 167 175
29 n.d. 37 n.d. 35 n.d. 45 n.d. 30 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
m o r s , we c h e c k e d the density of r e c e p t o r s as r e l a t e d
( T a b l e III). E v e n w h e n c o r e samples o n l y w e r e con-
to the d u r a t i o n of the disease. T h e a v e r a g e density of
s i d e r e d , t h e r e s e e m e d to be a significant 2-fold de-
r e c e p t o r s in t u m o r samples d e r i v e d f r o m p a t i e n t s
crease in m A C h R density in less m a l i g n a n t glioblas-
w h o died within 2-8 m o n t h s f r o m the first s u r g e r y was f o u n d to be 11.8 + 12.9 f m o l / g tissue, w h e r e a s in
tomas. In the m o r e b e n i g n a s t r o c y t o m a g r a d e II, an i n t e r m e d i a r y level of 7.5 + 1.8 f m o l / m g tissue was
patients w h o lived m o r e t h a n 8 m o n t h s , the a v e r a g e
o b s e r v e d . In the single s a m p l e of the e v e n m o r e be-
receptor
nign a s t r o c y t o m a s g r a d e I that was e x a m i n e d , the
density
was
3.5
+
1.9 f m o l / m g tissue
TABLE III
Muscarinic binding sites in human brain and primary brain tumors of various cellular origins Type of tissue
Cellular origin
Average density of mA ChR (frnol/mg tissue)
1 Control (non-malignant brain) 2 Glioblastoma multiforme (2-8 months duration) - - core samples only 3 Glioblastoma multiforme (over 8 months duration) - - core samples only 4 Astrocytoma grade II 5 Astrocytoma grade I 6 Meningioma 7 Oligodendroglioma 8 Neuroblastoma 9 Medulloblastoma 10 Neuroectoderm 11 Lung carcinoma (intracranial metastatis)
neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm mesenchyme neuroectoderm neuroectoderm primitive neuroectoderm primitive neuroectoderm epidermal
(n (n (n (n (n (n (n (n (n (n (n (n (n
= = = = = = = = = = = = =
19) 24.0 +_ 7.7 8) 11.8 + 12.9 6) 6.6 _+ 9.6 7) 3.5 + 1.9 6) 3.6 + 2.1 4) 7.5 + 1.8 1) 0 5) 1.4 + 2.0 1) 0 1) 0 1) 0 1) 0 2) 0
67
density of muscarinic receptors was below detection level (Table III). The values obtained in such small sample populations should certainly be interpreted with special precaution. Taken this way, our data imply that the density of muscarinic receptors in transformed cells of gliogenous origin in the human brain is relatively high in aggressive, highly malignant cells and lower in relatively benign tumoral cells, of a higher differentiation state. Several single samples from various other types of tumors were also tested for the presence of muscarinic binding sites. These included oligodendroglioma, neuroblastoma, medulloblastoma, primitive neuroectodermal tumor and intracranial mestastasis of lung carcinoma. In all of these, the density of muscarinic receptors was below detection level (Table III). The principal acetylcholine-hydrolyzing enzyme in the mammalian brain is acetylcholinesterase (acetylcholine hydrolase; EC 3.1.1.7, AChE)llA8, 32. In general, histochemical evidence suggests that AChE is associated with cholinergic synapses and produced by cholinergic neurons. Moreover, the specific activity of AChE decreases considerably in cases of damage to the cholinergic circuits in the brain 29. We therefore decided to test for AChE levels in the analyzed tumors as an additional measure for cholinoceptive activities, although acetylcholinesterase activity per se does not directly reflect the level of cholinergic communication. The specific activity of cholinesterase in glioblastomas, astrocytomas and meningiomas was found to be close to or even higher than the levels of this enzyme in non-malignant, control tissue samples (Tables I and II). In 85 % of the tumor samples, and in all of the control samples, inhibition by BW284C51 was/>75% (not shown), indicating that the major part of this activity represents 'true' acetylcholinesterase. In order to test whether the muscarinic binding activity in these brain tumors is correlated to the cholinesterase activity, we plotted individual data of cholinesterase activity against corresponding values of [3H]4NMBP binding sites. This was also done for control tissue samples. The pattern of distribution obtained by this analysis for the brain tumors was clearly different from that obtained with control brain samples (Fig. 3A). When cholinesterase specific activity values were divided by muscarinic receptor densities, remarkably different average arbitrary ratios, of 86.5
Tumors
Controls
700 == 6001 50C rE
"6
40~
E
0.
LU tO
5OO 2O0 I00 0
I0 20
50
0
I0 20
30
mAChR, fmol/mg tissue
i
200-
0
Ee -
I00
< E
!
U.I
c.) 2£
io-
: I
Tumors
Controls
Fig. 3. Correlation between the specificactivityof cholinesterase and the density of muscarinicbinding sites in control and tumor tissues. Cholinesterase (ChE) activity was assayed on [3H]acetylcholine and the density of muscarinic receptors (mAChR) was measured by binding of [3H]4NMBP, as described in Materials and Methods. The points represent individual tissue samples derived from tumor (T) or non-malignant, control (C) brain tissue. A) Correlation of cholinesterase specificactivityto the density of muscarinicreceptors. B) Ratio between cholinesteraseactivity and receptor density. Horizontal bars represent the average values.
-+ 66.5 and 10.6 + 8.0 were obtained for the brain tumors and the non-malignant brain tissue samples, respectively (Fig. 3B). This difference indicates that the expression of these two cholinoceptive activities is not similarly inter-related in these two tissue sources. DISCUSSION Human brain tumors of gliogenous origin appear by our findings to express muscarinic binding properties, with similar affinity characteristics for antagonists and agonists to those observed in non-malignant brain tissue. The density of muscarinic receptors in intracranial gliogenous tumors was found to be lower than, but at the same order of magnitude with the levels we observed in non-malignant brain tissue, which corroborate the values reported by others for human frontal cortex4.31 and hippocampus 19. Primary tumor tissue is heterogenous in cell type and stage of cell cycle, and is phenotypically divergent from its cell of origin. Therefore, estimation of the apparent levels of muscarinic binding sites in normal and malignant brain tissue can be affected by several factors, related to differences in the composition of tissue examined. However, the concentration of DNA and of protein appears to be rather consistent in all of the samples tested, indicating that the variations in muscarinic receptor levels are not due to major differences in number of cells per sample. In addition, the relatively high levels of muscarinic binding sites in tumors of gliogenous origin are unlikely to be due to invasion by blood vessels, since in other neuroectodermal-originated tumors, which display an equally high distribution of vascular formations, we could not detect any muscarinic binding activity. Highly malignant glioblastoma tumors, derived from patients who died within 2-8 months from the first operation, displayed levels of muscarinic receptors that were considerably higher than those measured in less malignant tumors of the same type, like astrocytoma grade I or II. The proliferation rates of cells within malignant gliogenous tumors have been reported to be significantly higher than those of cells in less malignant gliomas and astrocytomas 3. It is therefore possible that in the normal brain as well, gliogenous cells in their early development stages transiently express high levels of muscarinic binding
sites. It should be noted, that the high variability in phenotypic properties of cells within the tumor specimens certainly contributes to the variations observed in muscarinic receptor levels between different tumors and between samples derived from a single tumor. For example, the core region of glioblastoma tumors appears to be enriched in cells that are arrested in a quiescent phase of the cell cycle24. However, we have recently found the level of receptors to epidermal growth factor to be relatively high in core specimens of glioblastomas, as compared to periphery samples, and much higher than that observed in control specimenslL This excludes the possibility that the relatively low level of mAChR in the core area is due to the inability of core cells to express functional receptors. In contrast to the core, the peripheral region of glioblastoma tumors has been reported to be enriched in neoplastic, poorly differentiated astrocytes, that do not result from reactive gliosis 24. In addition to these morphological differences, the average proliferation rate of peripheral cells appeared to be higher than that of cells in the core region9, ~0. These [3H]thymidine incorporation experiments thus imply, that the higher variability in mAChR density in the borders of glioblastoma tumors can, perhaps, be explained by differences in the proliferation rates of cells in this region. The concentration of muscarinic binding sites has been reported to be down regulated by agonists in NIE-115 mouse neuroblastoma cellszS, in NG-108-15 hybrid cells, derived from mouse neuroblastoma and rat glioma 12 and in 1321 NI human astrocytoma cells 17,18. However, our measurements show no correlation between the density of receptors and the specific activity of cholinesterase in the human gliomas in vivo. Since the concentration of receptors in cultured cells decreases with cell passage, one cannot exclude the possibility that muscarinic binding sites are regulated differently in vivo and in cultured cells. Further studies will thus be required to unravel the functional role(s) of, and the mechanism(s) regulating the expression of mAChR in gliogenous cells, within different areas in glioblastoma tumors and in the non-malignant human brain. In normal neural tissue, receptors levels vary considerably between regions. Furthermore, within a given region, only a few percent of the cells may have high receptor levels, even in receptor-rich regions.
69 However, decreases in the density of muscarinic receptors have been reported in cholinergic-innervated areas in the elderly braint9, in damaged cholinergic areas in the choreic brain6,8,30 and in the brain of Down syndrome patients 33. The severity of senile dementia of the Alzheimer's type has been shown to be directly correlated with the magnitude of deficit in cholinergic circuits20. One of the pharmacologic approaches to the study of senile dementias, and of aging in general, involves the investigation of agents which affect cholinergic transmissionL These include anti-cholinesterases such as physostigmine, which decrease the breakdown of acetylcholine 5, muscarinic agonists like Arecoline, or combination of direct precursors to acetylcholine, like choline, with agents such as piracetam, facilitating the release of this transmitter from functionally active cholinergic neurons 21. All of these therapeutic treatments are based on the assumption that the remaining functional muscarinic binding sites are located on cholinoceptive neurons, and estimation of the level of these receptors is based on the use of muscarinic ligands in biochemical binding studies. However, the density of m A C h R appears to remain unchanged in several areas of the choreic brain 31 as well as in elderly 3° and demented brain 4, in spite of significant reductions in cholinesterase and in choline acetyltransferase activity29. In major parts of the human brain, the regulation of muscarinic receptors thus appears to be irrespective of that of damage-induced changes in cholinesterase. Our present observations indicate that this
is the case in human gliogenous tumors. It is therefore intriguing to assume that in brain areas where the level of muscarinic receptors remains unchanged in spite of damage to cholinergic neurons, these receptors might be expressed by astrocytes. If indeed part of the receptors are located on glial cells, biochemical measurements alone are insufficient to indicate the abundance of functional receptors on neurons. It therefore appears that new approaches, such as autoradiography with mustard compounds 22, or photo-affinity labeling of muscarinic receptors 1 should be employed on human brain sections, in order to distinguish between muscarinic receptor sites on neurons and glia. A better estimation of the level of functional neuronal receptors may then provide means to develop alternative therapeutic protocols for syndromes involving damage to cholinergic circuits.
REFERENCES
venous physostigmine, Amer. J. Psychiat., 139 (1982) 1421-1424. 6 Enna, S. J., Bennett, J. P., Bylund, D. B., Snyder, S. H., Bird, E. D. and Iversen, L. L., Alterations of brain neurotransmitter receptor binding in Huntington's chorea, Brain Research, 116 (1976) 531-537. 7 Gross, R. A. and Clark, R. B., Regulation of 3,5-monophosphate content in human astrocytoma cells by isoproterenol and carbachol, Molec. Pharmacol., 13 (1977) 242-250. 8 Hiley, C. R. and Bird, E. D., Decreased muscarinic receptor concentration in post-mortem brain in Huntington's chorea, Brain Research, 80 (1974) 355-358. 9 Hoshino, T., Townsend, J. J., Muraoka, I. and Wilson, C. B., An autoradiographic study of human gliomas: growth kinetics of anaplastic astrocytoma and glioblastoma multiforme, Brain, 103 (1980) 967-984. 10 Hoshino, T. and Wilson, C. B., Cell kinetic analyses of hu-
1 Avissar, S., Amitai, G. and Sokolovsky, M., Oligomeric structure of muscarinic receptors is shown by high- and low affinity agonist states, Proc. nat. Acad. Sci., U.S.A., 80 (1983) 156--159. 2 Bartus, A., Strategies for the Development of an Effective Treatment for Senile Dementia, S. Gershon and T. Crook (Eds.), Mark Powley Associates, New Canaan, 1981, pp. 79--90. 3 DeReuck, J., Roels, H. and Van der Eekem, H., Cytophotometric DNA determination in human astroglial tumors, Histopathology, 3 (1979) 107-115. 4 Davies, P. and Verth, A. H., Regional distribution of muscarinic acetylcholine receptor in normal and Alzheimer's type dementia brains, Brain Research, 138 (1978) 385-392. 5 Davis, K. L. and Mobs, R. C., Enhancement of memory processes in Alzheimer's Disease with multiple-dose intra-
ACKNOWLEDGEMENTS We appreciate the help of Prof. Israel Silman and Mrs. Esther Roth in cholinesterase measurements, and we are grateful to Prof. A. Bartal for his continuous support and encouragement. This study was supported by the US Army Medical Research and Development Command (contract no. D A M D 17-82-C2145 to H.S.), by the Hermann and Lilly Schilling Foundation for Medical Research (grant to H.S.) and by the Recanati Fund for Medical Research, Israel (grant to M.S.).
70 man malignant brain tumors (gliomas), Cancer, 44 (1979) 956-962. 11 Johnson, C. D. and Russell, R. L., A rapid, simple radiometric assay for cholinesterase, suitable for multiple determinations, Analyt. Biochem., 64 (1975) 229-238. 12 Klein, W. L., Nathanson, N. M. and Nirenberg, M., Muscarinic acetylcholine receptor regulation by accelerated rate of receptor loss, Biochem. biophys. Res. Commun., 90 (1979) 506-512. 13 Kloog, Y., Egozi, Y. and Sokolovsky, M., Characterization of muscarinic acetylcholine receptors from mice brain: evidence for regional heterogeneity and isomerization, Molec. Pharmacol., 15 (1979) 545-558. 14 Leyva, A. and Kelley, W. N., Measurement of DNA in cultured human cells, Analvt. Biochem., 62 (1974) 173-179. 15 Libermann, T. A., Razon, N., Bartal, A. D., Yarden, Y., Schlessinger, J. and Soreq, H., Expression of epidermal growth factor receptors in human brain tumors, Cancer Res., 44 (1984) 753-760. 16 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the folin phenol reagent, J. biol. Chem., 193 (1951) 265--275. 17 Meeker, R. B. and Harden, T. K., Muscarinic cholinergic receptor-mediated activation of phosphodiesterase, Molec. Pharmacol., 22 (1982) 310-319. 18 Meeker, R. B. and Harden, T. K., Muscarinic cholinergic receptor-mediated control of cyclic AMP metabolism, Molec. Pharmacol., 23 (1983) 384-392. 19 Nordberg, A. and Winblad, B., Cholinergic receptors in human hippocampus: regional distribution and variance with age, Life Sci., 29 (1981) 1937-1944. 20 Perry, E. K., Tomlinson, B. E., Blessed, G., Bergman, K., Bigson, P. H. and Perry, R. H., Correlation of cholinergic abnormalities with senile placques and mental test scores in senile dementia, Brit. Med. J., 2 (1978) 1457-1459. 21 Roberts, E., Potential therapies in aging and senile dementias. In: Alzheimer's disease, Down syndrome and aging. Ann. N. Y. Acad. Sci., 396 (1982) 165-178. 22 Rotter, A., Birdsall, N. J. M., Field, P. M. and Raisman, G., Muscarinic receptors in the central nervous system of the rat. II. Distribution of binding of 3H-propylbenzilylcho-
line mustard in the midbrain and hindbrain, Brain Res. Rev., 1 (1979) 167-183. 23 Russell, D. S. and Rubinstein, L. J., Primary tumors of neuroectodermal origin. In Pathology of the Nervous System, Edward Arnold Ltd., 1963, pp. 93--162. 24 Russell, D. S. and Rubinstein, L. J., Pathology of tumors of the nervous system, (1971) pp. 109-113. 25 Shifrin, G. S. and Klein, W. L., Regulation of muscarinic acetylcholine receptor concentration in cloned neuroblastoma cells, J. Neurochem., 34 (1980) 993-999. 26 Sokolovsky, M., Gurwitz, D. and Kloog, Y., Biochemical characterization of the muscarinic receptors, Advanc. Enzymol., 55 (1983) 137-196. 27 Soreq, H., Gurwitz, D., Eliyahu, D. and Sokolovsky, M., Altered ontogenesis of muscarinic receptors in agranular cerebellar cortex, J. Neurochem., 39 (1982) 756--763. 28 Soreq, H., Parvari, R. and Silman, I., Biosynthesis and secretion of catalytically active acetylcholinesterase in Xenopus oocytes microinjected with mRNA from rat brain and from Torpedo electric organ, Proc. nat. Acad. Sci. U.S.A., 79 (1982) 830-834. 29 Terry, R. D. and Davies, P., Dementia of the Alzheimer type, Ann. Rev. Neurosci., 3 (1980) 77-95. 30 Wastek, G. J., Stern, L. Z., Johnson, P. C. and Yamamura, H. I., Huntington's disease: regional alteration in muscarinic cholinergic receptor binding in human brain~ Life Sci., 19 (1976) 1033-1040. 31 Wastek, G. J. and Yamamura, H. I., Biochemical characterization of the muscarinic cholinergic receptor in human brain: alterations in Huntington's Disease, Molec. Pharmacol., 14 (1978) 768-780. 32 White, P., Hiley, C. R., Goodhart, M. J., Carrasco, L. H., Keet, J. P., Williams, I. E. I. and Bowen, D. M., Neocortical cholinergic neurons in elderly people, Lancet, March 26 (1977) 668--671. 33 Yates, C. M., Simpson, J., Maloney, A. F. J., Gordon, A. and Reid, A. H., Alzheimer-like cholinergic deficiency in Down syndrome, Lancet, 2 (1980) 979. 34 Zulch, K. J,, Principles of the New World Health Organization (WHO) classification of Brain Tumors, Neuroradiology, 19 (1980) 59-66.
61
BRD 50044
Expression of Muscarinic Binding Sites in Primary Human Brain Tumors DAVID GURWITZ I, NISSIM RAZON 2, MORDECHAI SOKOLOVSKY 1and HERMONA SOREQ 3
1Department of Biochemistry, George S. Wise Facultyfor Life Sciences, Tel A viv University, Tel A viv; 2Department of Neurosurgery, Ichilov Hospital, Tel A viv Medical Center, Tel A viv and 3Department of Neurobiology, Weizmann Institute of Science, Rehovot, (Israel) (Accepted January 17th, 1984)
Key words: muscarinic receptors - - glioblastoma multiforme - - astrocytoma - - meningioma - acetylcholinesterase- - malignancy --binding affinity
The expression of muscarinic binding sites was examined in a collection of primary brain tumors of different cellular origins and various degrees of dedifferentiation, as compared to control specimens. Eleven gliogenous tumors were examined, all of which contained substantial amounts of muscarinic binding sites. Most of the other tumor types examined did not display detectable binding of [3H]Nmethyl-4-piperidyl benzilate ([3H]4NMPB). Scatchard analysis indicated the existence of homogenous antagonist sites in both normal forebrain and gli0blastoma multiforme, with K d values of 1.2 nM and 0.9 nM, respectively. The density of muscarinic binding sites varied between tumors from different patients, and also between specimens prelevated from different areas of the same tumor. This variability, as well as the average density of binding sites, appeared to be larger in highly malignant tumors than in less malignant ones. In contrast, the density of muscarinic receptors from control specimens was invariably high, but within the same order of magnitude. To test whether the muscarinic binding activity in the brain tumors is correlated to other cholinoceptive properties, cholinesterase activity was also examined. Individual data for density of [3H]4NMPB binding sites were then plotted against corresponding values of cholinesterase activity. The pattern of distribution of these values was clearly different in tumor specimens, when compared to that observed in samples derived from non-malignant brain. Our observations indicate that human brain cells of gliogenous origin are capable of expressing muscarinic binding sites, and that, if a correlation exists between muscarinic receptors and cholinesterase levels in gliogenous tumors, it differs from that of non-malignant brain tissue. INTRODUCTION Biochemical studies on binding of muscarinic ligands have been extensively carried out using brain tissue as a m o d e l system (reviewed recently by Soko10vsky et al. 26) and special attempts were m a d e to correlate the density of muscarinic receptors ( m A C h R ) in the h u m a n brain to the functioning of cholinergic circuits. D e c r e a s e s in muscarinic binding were shown in the corpus striatum of H u n t i n g t o n ' s disease patients6,8,30. Biochemical analysis of the muscarinic receptors has then been carried out in normal human brain, as c o m p a r e d with those in the choreic brain. It was found that the n u m b e r of receptors, but not their affinity properties, is changed in the caudate nucleus and p u t a m e n of Huntington chorea patients. This decrease a p p e a r e d to be correlated to r e d u c e d activity of cholineacetyltransferase3t. H o w e v e r , the w e l l - d o c u m e n t e d decreases in
the density of muscarinic receptors in the choreic brain a p p e a r to represent an exceptional case. The regional distribution of muscarinic binding sites was found to remain unchanged in the brain of A l z h e i m er's-type d e m e n t i a patients, in spite of the evidence for extensive loss of enzymes of the cholinergic system4, 29. In the frontal cortex of elderly people, muscarinic receptor binding sites have been r e p o r t e d to remain unchanged31. W h e n morphological analysis was carried out in parallel, a certain decrease in the density of muscarinic receptors was found in specimens with normal morphological evidence, but not in those derived from brains with senile degenerations 32. Even in cholinergic-innervated regions such as the anterior parts of the human h i p p o c a m p u s , the age-related decrease in the density of muscarinic receptors was found to be r a t h e r limited 19. In certain cases, the expression of muscarinic binding properties thus remains u n i m p a i r e d in the human
Correspondence: H. Soreq, Dept. of Neurobiology, Weizmann Institute of Science, Rehovot 76100, Israel.
62 brain, under conditions where an apparent damage has been caused to cholinergic neurons. This indicates that brain cell types other than such neurons may also express muscarinic binding properties. The indication that non-neuronal brain cells contain muscarinic receptors has been supported by experiments with human 1321 NI astrocytoma ceils in culture. Agonist-induced stimulation of muscarinic receptors in these cells was shown to result in a decrease in cyclic AMP accumulation 7, via the activation of a phosphodiesterase 17, affecting both synthesis and degradation of cyclic AMP, and reducing the density of muscarinic receptors~8. However, to the best of our knowledge, there are no reports regarding the level and properties of muscarinic receptors in vivo, in cells of gliogenous origin. Studying these receptors in their primary in vivo tissue source is particularly important since the density of muscarinic binding sites in cloned cells has been shown to decrease significantly with cell passage2L In an earlier study, we attempted to correlate muscarinic receptors to specific in vivo cellular origins in the developing rodent cerebellum. The cell type composition in the cerebellum is relatively simple, and various malformations are available which are deficient in particular cell types. However, in spite of these advantages, we could not, in that study, reveal whether cerebellar cells of non-neuronal origin express muscarinic binding sites 27. To address this issue directly, we therefore examined the in vivo levels and biochemical properties of muscarinic binding sites in human primary brain tumors, which provide a tissue source derived essentially of a single cellular origin 23. Various tumors, of neuroectodermal and mesenchymal origins 15, were screened for this purpose. Our observations reveal that gliogenous intracranial tumors contain substantial levels of muscarinic binding sites, of similar affinity properties to those observed in non-malignant brain tissue. Furthermore, our findings indicate that in gliogenous tumors, the interrelationship between the expression of muscarinic receptors and cholinesterase is different than that of the non-malignant brain. MATERIALS AND METHODS
Chemicals Acetylcholine chloride and the detergent Triton
X-100 were obtained from Sigma. [3H]Acetylcholine iodide (90 mCi/mmol) was obtained from New England Nuclear. The potent muscarinic antagonist [3H]N-methyl-4-piperidyl benzilate ([3H]4NMBP, 69 Ci/mmol) and unlabeled muscarinic ligands were as described 13. All salts and buffers were Analar grade.
Brain tissues Specimens from brain tumors were invariably preievated at the time of surgery. Out of the 16 tumor samples examined, 8 were from the frontal lobe, 5 were in the temporal lobe, 2 in the parietal region and one in the cerebellum (Table I). Control specimens, all from the frontal tip of patients who died from diseases not related to the central nervous system, were generally prelevated 24 h postmortem (after about 24 h at 4 °C). Whenever possible prelevation at surgery was employed for control samples as well. All tissues were frozen in liquid nitrogen following prelevation, and were stored at --70 °C until use. Histological and pathological preparations Tissue excised at the time of prelevation was fixed in 4% buffered formaldehyde for 15 h at room temperature, washed twice with PBS and embedded in paraffin. Five/~m sections were prepared, mounted on glass slides and stained with haematoxilin-eosin. Histological characteristics were defined according to the classification accepted by the WHO 34. Summary of the pathological findings for each of the tumor specimens employed is presented in Table I, as confirmed independently by a senior pathologist. Biochemical determinations Thawed tissue samples, about 50 mg, were suspended in 0.01 M Tris, pH 7.4 and homogenized as described 28. DNA content of these tissue samples was determined as described by Leyva and KelleyTM. Protein content was determined on the solubilized tissue samples according to Lowry et a1.16. Muscarinic binding measurements Tissue samples were weighed and homogenized in 0.32 M sucrose, as previously described 13. Aliquots of homogenates (50/A) were incubated at 25 °C for 30 min in 2 ml buffer, containing 118 mM NaCl, 5 mM KCI, 1 mM MgC12, 10 mM glucose and 25 mM Tris-HCl (pH = 7.4 at 25 °C). For antagonist
63 ~
~
ae
' ~
~ ~us.~
"~
~ ,, .~.
~
~:.~
•.- ~ .;~ .~
.~.~.~,~.~
&
•c , . "~
~' O
O
O
O
,.,Q
~ E E E E E E E E
c~
~Q <
~ ~"
~ ~'~ O O
~ ~ ~ O
O
64 binding studies, 8 mM [3H]4NMBP were included. At this concentration, muscarinic receptors are nearly saturated ~3. In some experiments, complete binding isotherms for [3H]4NMBP were studied, using various ligand concentrations. For agonist binding studies, the buffer contained 2 nM [3H]4NMBP, with or without 250/.tM carbamylcholine, and the inhibition of binding of the labeled antagonist by the unlabeled agonist was determined. Incubation was terminated by rapid filtration through GF/C filters (Whatman, 25 mm diameter), followed by 4 × 3 ml washes with ice-cold buffer. Radioactivity was determined using a scintillation cocktail (Lumac, Hydroluma) and a scintillation spectrometer (Packard, Tricarb 300). Assays were carried out in triplicates, and specific binding was defined as total minus non-specific, i.e. binding in the presence of 1 ~M unlabeled atropine, Cholinesterase measurements
Tissue samples were weighed and homogenized in 0.9 ml of ice-cold 1 M NaC1-0.05 M Tris chloride, pH 7.5, using 10 strokes of a Teflon-glass homogenizer driven by a Heidolf Motor at half speed (1000 rpm). A 0.1 ml aliquot of aqueous Triton X-100 (10% v/v) was added to each homogenate, and homogenization was repeated using a further 10 strokes. The homogenates were centrifuged for 5 min in an Eppendorf $412 centrifuge and the supernarant was reserved. Cholinesterase activity was performed according to the general procedure of Johnson and RusselP 1 and as described 2s. Reaction mixtures contained 20/~1 of homogenate supernatant, 10 /~1 of 1.2 M NaC1-0.5 M Tris chloride, pH 7.4, 30 ~1 of 10 mM radioactive acetylcholine (made by diluting [3H]acetylcholine into 10 mM acetylcholine chloride so as to yield ca. 80,000 cpm total counts in each sample), and water to a final volume of 100 ~tl, made up in a 5 ml scintillation vial. For BW284C51 inhibition measurements, an aliquot of a stock solution of the inhibitor (1,5-bis(4-allyl dimethylammoniumphenyl pentan-3-one dibromide, Sigma) was added so as to yield a final concentration of 10-5 M 2s. Assays were run at 25 °C, for 30-120 rain, and the reaction was terminated by addition of 100/~1 of stopping mixture (1 M chloroacetic acid-0.5 M NaOH-2.0 M NaC1). Scintillation Cocktail (4 ml of 90% xylene-10% isoamylalcohol con-
taining 10% 'Xyloflour') was then added. After vigorous shaking, samples were counted in a Packard Model 3390 Tricarb Liquid Scintillation Spectrometer. All measurements were in linear range with respect to time and protein concentration. RESULTS Homogenates from primary brain tumors were tested for the presence of specific muscarinic binding sites. Out of the various types of tumors which were examined, substantial levels of [3H]4NMBP binding sites were determined in all of the tumors derived from gliogenous origin, such as glioblastoma multiforme and grade I and |I astrocytomas. In both of these tumor types, binding was inhibited by 20--50% in the presence of 250/~M carbamylcholine. Low levels of muscarinic receptors could also be detected in 2 out of 5 meningiomas that were examined. These were, in both cases, recurrent tumors (Table I). High variability in the density of muscarinic receptors was observed between different glioblastoma tumors. In order to exclude the possibility that the variation in the level of receptors could reflect different numbers of cells in the tissue samples, the amount of DNA was measured in most of the samples tested. The total amount of D N A varied between 0.4 and l.(I mg/g of wet weight tissue, and there was no correlation to the level of receptor. In addition, the content of protein was determined, and was found to range between 70-140 mg/g tissue in 80% of the samples. Moreover, the general size distribution of the total tissue proteins appeared by SDS-polyacrylamide gel analysis to be rather similar in all of the samples that were tested (not shown). Thus, the variations in the level of muscarinic binding sites cannot be explained by selectively high in situ proteolysis in part of the samples. The variability in the density of muscarinic binding sites was most conspicuous in samples derived from the peripheral regions of gliogenous tumors. These displayed an average level of 15.5 + 14.0 (n = 4, see Table I). In samples derived from core regions in such tumors, the average density of receptors was significantly lower, at a level of 5.7 _+ 6.2 (n = 15, Table I). This difference could probably be attributed to the cell type diversity between different areas in gliogenous tumors. Fig. 1 displays such morpho-
65
• 30- / l
;
"S 20i- /
', /
'%.
'/ '~,
\'~
l
\'~ '~
I
",%
35-45 25-35 ~5-z5 545
o-5
0
mAChR, fmol/mg tissue Fig. 1. Top: differences in histological appearance between various areas in glioblastoma multiforme tumors. Prelevation of specimens, fixation and staining were as described in Materials and Methods. Note the different histological appearance of tissue from the periphery (left) and core (right) regions. Bottom: distribution of mAChR-density in gliogenous tumors and in control brain. Tissue samples were prelevated from the core regions of gliogenous tumors and from control brain tissue (Tables I and II). Preparation of homogenates and binding of [3H]4NMBP were as described in Materials and Methods. Percent of specimens which displayed specific mAChR densities is presented.
distribution patterns of individual values of m A C h R densities in control specimens, as compared with those obtained with core samples of gliogenous tumors. It is apparent that the two patterns are rather parallel, with a clear difference in peak values. There appeared to be no correlation between the density of receptors in non-malignant brain samples and the age of the patients, in agreement with Wastek and Yamamura 31. Scatchard analysis of [3H]4NMBP-binding indicated, for preparations from both non-malignant brain and glioblastoma multiforme tumors, the existence of homogenous antagonist sites. The dissociation constants for [aH]4NMBP were found to be 1.2 nM and 0.9 nM for control and glioblastoma samples, respectively (Fig. 2). These values are very close to those detected for [3H]4NMBP-binding in the mouse brain 13. Furthermore, competition experiments indicated similar apparent affinity values for the binding of the agonist carbamylcholine towards the muscarinic receptors in non-malignant brain and gliogenous tumors. The inhibition asserted by 250 ,uM carbamylcholine on the binding of 2 nM [3H]4NMBP was found to be 35.2% _+ 6.4% and 35.5% _+ 9.7% for control (n = 5) and gliogenous (n = 10) samples, respectively (Tables I and II). In an attempt to correlate the muscarinic site properties with the clinical stage of the glioblastoma tu-
[•
nn Q_
~ E
z~ , E IC o
o.
logical differences between the periphery and core areas of a single glioblastoma tumor. In the periphery, hyperplasia of endothelial cells, rosettes of undifferentiated glial cells and pseudopallisades are most conspicuous. In contrast, the core region mainly includes neoplastic, hyperchromatic quiescent cells and some necrotic areas. This figure also reveals the
I
o ~ Control o ~ 1 2 nM
==20 nn ~
I
0
X
Glioblostomo ~ o
X, I0
20
\
50
40
Specific3H-4NMPBBound(fmo[es/mgtissue) Fig. 2. Affinity properties of muscarinic binding sites from nonmalignant brain tissue and from glioblastoma multiforme. Scatchard plot of specific [3H]4NMBP binding to homogenates of non-malignant, control brain tissue (O, samples 3, Table II) and of glioblastoma multiforme (0, samples 6, Table I) is presented. The K d values were 1.2 nM and 0.9 nM for the control and the tumor samples, respectively.
"FABLE II
Control forebrain samples: description, muscarinic receptors and cholinesterase levels MI, myocardial infarct
Sample
Age~Sex
Disease
Muscarinic binding sites (mA ChR) Carbamylcholine Density inhibition (%) (fmol/mg tissue)
Cholinesterase (ChE) specific activity, (nmol/min/g/tissue)
la lb 2 3 4a 4b 5a 5b 6 7 8 9 10 l1 12 13 14 15 16
71/M 71/M 55/F 66/17 66/M 66/M 66/F 66/F 55/M 7 l/M 74/F 64/M 70/M 81/M 1/M 78/M 54/M 47/M 70/M
Neurosyphilis died of MI Neurosyphilis died of MI Sepsis MI Pneumonia Pneumonia Pulmonary embolism Pulmonary embolism MI B rain infarction MI MI Pulmonary edema Pancreatic carcinoma Cardiac malformation Bronchopneunomia Brain edema MI + sepsis MI
14.2 19.0 23.2 35.9 33.0 17.0 14.0 8.5 27.0 21.0 26.0 32.0 27.0 22.0 19.0 37.0 25.0 28.0 29.0
410 n.d. 303 235 172 n.d. 351 n.d. 253 87 n.d. n.d. 367 179 296 68 128 167 175
29 n.d. 37 n.d. 35 n.d. 45 n.d. 30 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.
m o r s , we c h e c k e d the density of r e c e p t o r s as r e l a t e d
( T a b l e III). E v e n w h e n c o r e samples o n l y w e r e con-
to the d u r a t i o n of the disease. T h e a v e r a g e density of
s i d e r e d , t h e r e s e e m e d to be a significant 2-fold de-
r e c e p t o r s in t u m o r samples d e r i v e d f r o m p a t i e n t s
crease in m A C h R density in less m a l i g n a n t glioblas-
w h o died within 2-8 m o n t h s f r o m the first s u r g e r y was f o u n d to be 11.8 + 12.9 f m o l / g tissue, w h e r e a s in
tomas. In the m o r e b e n i g n a s t r o c y t o m a g r a d e II, an i n t e r m e d i a r y level of 7.5 + 1.8 f m o l / m g tissue was
patients w h o lived m o r e t h a n 8 m o n t h s , the a v e r a g e
o b s e r v e d . In the single s a m p l e of the e v e n m o r e be-
receptor
nign a s t r o c y t o m a s g r a d e I that was e x a m i n e d , the
density
was
3.5
+
1.9 f m o l / m g tissue
TABLE III
Muscarinic binding sites in human brain and primary brain tumors of various cellular origins Type of tissue
Cellular origin
Average density of mA ChR (frnol/mg tissue)
1 Control (non-malignant brain) 2 Glioblastoma multiforme (2-8 months duration) - - core samples only 3 Glioblastoma multiforme (over 8 months duration) - - core samples only 4 Astrocytoma grade II 5 Astrocytoma grade I 6 Meningioma 7 Oligodendroglioma 8 Neuroblastoma 9 Medulloblastoma 10 Neuroectoderm 11 Lung carcinoma (intracranial metastatis)
neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm neuroectoderm mesenchyme neuroectoderm neuroectoderm primitive neuroectoderm primitive neuroectoderm epidermal
(n (n (n (n (n (n (n (n (n (n (n (n (n
= = = = = = = = = = = = =
19) 24.0 +_ 7.7 8) 11.8 + 12.9 6) 6.6 _+ 9.6 7) 3.5 + 1.9 6) 3.6 + 2.1 4) 7.5 + 1.8 1) 0 5) 1.4 + 2.0 1) 0 1) 0 1) 0 1) 0 2) 0
67
density of muscarinic receptors was below detection level (Table III). The values obtained in such small sample populations should certainly be interpreted with special precaution. Taken this way, our data imply that the density of muscarinic receptors in transformed cells of gliogenous origin in the human brain is relatively high in aggressive, highly malignant cells and lower in relatively benign tumoral cells, of a higher differentiation state. Several single samples from various other types of tumors were also tested for the presence of muscarinic binding sites. These included oligodendroglioma, neuroblastoma, medulloblastoma, primitive neuroectodermal tumor and intracranial mestastasis of lung carcinoma. In all of these, the density of muscarinic receptors was below detection level (Table III). The principal acetylcholine-hydrolyzing enzyme in the mammalian brain is acetylcholinesterase (acetylcholine hydrolase; EC 3.1.1.7, AChE)llA8, 32. In general, histochemical evidence suggests that AChE is associated with cholinergic synapses and produced by cholinergic neurons. Moreover, the specific activity of AChE decreases considerably in cases of damage to the cholinergic circuits in the brain 29. We therefore decided to test for AChE levels in the analyzed tumors as an additional measure for cholinoceptive activities, although acetylcholinesterase activity per se does not directly reflect the level of cholinergic communication. The specific activity of cholinesterase in glioblastomas, astrocytomas and meningiomas was found to be close to or even higher than the levels of this enzyme in non-malignant, control tissue samples (Tables I and II). In 85 % of the tumor samples, and in all of the control samples, inhibition by BW284C51 was/>75% (not shown), indicating that the major part of this activity represents 'true' acetylcholinesterase. In order to test whether the muscarinic binding activity in these brain tumors is correlated to the cholinesterase activity, we plotted individual data of cholinesterase activity against corresponding values of [3H]4NMBP binding sites. This was also done for control tissue samples. The pattern of distribution obtained by this analysis for the brain tumors was clearly different from that obtained with control brain samples (Fig. 3A). When cholinesterase specific activity values were divided by muscarinic receptor densities, remarkably different average arbitrary ratios, of 86.5
Tumors
Controls
700 == 6001 50C rE
"6
40~
E
0.
LU tO
5OO 2O0 I00 0
I0 20
50
0
I0 20
30
mAChR, fmol/mg tissue
i
200-
0
Ee -
I00
< E
!
U.I
c.) 2£
io-
: I
Tumors
Controls
Fig. 3. Correlation between the specificactivityof cholinesterase and the density of muscarinicbinding sites in control and tumor tissues. Cholinesterase (ChE) activity was assayed on [3H]acetylcholine and the density of muscarinic receptors (mAChR) was measured by binding of [3H]4NMBP, as described in Materials and Methods. The points represent individual tissue samples derived from tumor (T) or non-malignant, control (C) brain tissue. A) Correlation of cholinesterase specificactivityto the density of muscarinicreceptors. B) Ratio between cholinesteraseactivity and receptor density. Horizontal bars represent the average values.
-+ 66.5 and 10.6 + 8.0 were obtained for the brain tumors and the non-malignant brain tissue samples, respectively (Fig. 3B). This difference indicates that the expression of these two cholinoceptive activities is not similarly inter-related in these two tissue sources. DISCUSSION Human brain tumors of gliogenous origin appear by our findings to express muscarinic binding properties, with similar affinity characteristics for antagonists and agonists to those observed in non-malignant brain tissue. The density of muscarinic receptors in intracranial gliogenous tumors was found to be lower than, but at the same order of magnitude with the levels we observed in non-malignant brain tissue, which corroborate the values reported by others for human frontal cortex4.31 and hippocampus 19. Primary tumor tissue is heterogenous in cell type and stage of cell cycle, and is phenotypically divergent from its cell of origin. Therefore, estimation of the apparent levels of muscarinic binding sites in normal and malignant brain tissue can be affected by several factors, related to differences in the composition of tissue examined. However, the concentration of DNA and of protein appears to be rather consistent in all of the samples tested, indicating that the variations in muscarinic receptor levels are not due to major differences in number of cells per sample. In addition, the relatively high levels of muscarinic binding sites in tumors of gliogenous origin are unlikely to be due to invasion by blood vessels, since in other neuroectodermal-originated tumors, which display an equally high distribution of vascular formations, we could not detect any muscarinic binding activity. Highly malignant glioblastoma tumors, derived from patients who died within 2-8 months from the first operation, displayed levels of muscarinic receptors that were considerably higher than those measured in less malignant tumors of the same type, like astrocytoma grade I or II. The proliferation rates of cells within malignant gliogenous tumors have been reported to be significantly higher than those of cells in less malignant gliomas and astrocytomas 3. It is therefore possible that in the normal brain as well, gliogenous cells in their early development stages transiently express high levels of muscarinic binding
sites. It should be noted, that the high variability in phenotypic properties of cells within the tumor specimens certainly contributes to the variations observed in muscarinic receptor levels between different tumors and between samples derived from a single tumor. For example, the core region of glioblastoma tumors appears to be enriched in cells that are arrested in a quiescent phase of the cell cycle24. However, we have recently found the level of receptors to epidermal growth factor to be relatively high in core specimens of glioblastomas, as compared to periphery samples, and much higher than that observed in control specimenslL This excludes the possibility that the relatively low level of mAChR in the core area is due to the inability of core cells to express functional receptors. In contrast to the core, the peripheral region of glioblastoma tumors has been reported to be enriched in neoplastic, poorly differentiated astrocytes, that do not result from reactive gliosis 24. In addition to these morphological differences, the average proliferation rate of peripheral cells appeared to be higher than that of cells in the core region9, ~0. These [3H]thymidine incorporation experiments thus imply, that the higher variability in mAChR density in the borders of glioblastoma tumors can, perhaps, be explained by differences in the proliferation rates of cells in this region. The concentration of muscarinic binding sites has been reported to be down regulated by agonists in NIE-115 mouse neuroblastoma cellszS, in NG-108-15 hybrid cells, derived from mouse neuroblastoma and rat glioma 12 and in 1321 NI human astrocytoma cells 17,18. However, our measurements show no correlation between the density of receptors and the specific activity of cholinesterase in the human gliomas in vivo. Since the concentration of receptors in cultured cells decreases with cell passage, one cannot exclude the possibility that muscarinic binding sites are regulated differently in vivo and in cultured cells. Further studies will thus be required to unravel the functional role(s) of, and the mechanism(s) regulating the expression of mAChR in gliogenous cells, within different areas in glioblastoma tumors and in the non-malignant human brain. In normal neural tissue, receptors levels vary considerably between regions. Furthermore, within a given region, only a few percent of the cells may have high receptor levels, even in receptor-rich regions.
69 However, decreases in the density of muscarinic receptors have been reported in cholinergic-innervated areas in the elderly braint9, in damaged cholinergic areas in the choreic brain6,8,30 and in the brain of Down syndrome patients 33. The severity of senile dementia of the Alzheimer's type has been shown to be directly correlated with the magnitude of deficit in cholinergic circuits20. One of the pharmacologic approaches to the study of senile dementias, and of aging in general, involves the investigation of agents which affect cholinergic transmissionL These include anti-cholinesterases such as physostigmine, which decrease the breakdown of acetylcholine 5, muscarinic agonists like Arecoline, or combination of direct precursors to acetylcholine, like choline, with agents such as piracetam, facilitating the release of this transmitter from functionally active cholinergic neurons 21. All of these therapeutic treatments are based on the assumption that the remaining functional muscarinic binding sites are located on cholinoceptive neurons, and estimation of the level of these receptors is based on the use of muscarinic ligands in biochemical binding studies. However, the density of m A C h R appears to remain unchanged in several areas of the choreic brain 31 as well as in elderly 3° and demented brain 4, in spite of significant reductions in cholinesterase and in choline acetyltransferase activity29. In major parts of the human brain, the regulation of muscarinic receptors thus appears to be irrespective of that of damage-induced changes in cholinesterase. Our present observations indicate that this
is the case in human gliogenous tumors. It is therefore intriguing to assume that in brain areas where the level of muscarinic receptors remains unchanged in spite of damage to cholinergic neurons, these receptors might be expressed by astrocytes. If indeed part of the receptors are located on glial cells, biochemical measurements alone are insufficient to indicate the abundance of functional receptors on neurons. It therefore appears that new approaches, such as autoradiography with mustard compounds 22, or photo-affinity labeling of muscarinic receptors 1 should be employed on human brain sections, in order to distinguish between muscarinic receptor sites on neurons and glia. A better estimation of the level of functional neuronal receptors may then provide means to develop alternative therapeutic protocols for syndromes involving damage to cholinergic circuits.
REFERENCES
venous physostigmine, Amer. J. Psychiat., 139 (1982) 1421-1424. 6 Enna, S. J., Bennett, J. P., Bylund, D. B., Snyder, S. H., Bird, E. D. and Iversen, L. L., Alterations of brain neurotransmitter receptor binding in Huntington's chorea, Brain Research, 116 (1976) 531-537. 7 Gross, R. A. and Clark, R. B., Regulation of 3,5-monophosphate content in human astrocytoma cells by isoproterenol and carbachol, Molec. Pharmacol., 13 (1977) 242-250. 8 Hiley, C. R. and Bird, E. D., Decreased muscarinic receptor concentration in post-mortem brain in Huntington's chorea, Brain Research, 80 (1974) 355-358. 9 Hoshino, T., Townsend, J. J., Muraoka, I. and Wilson, C. B., An autoradiographic study of human gliomas: growth kinetics of anaplastic astrocytoma and glioblastoma multiforme, Brain, 103 (1980) 967-984. 10 Hoshino, T. and Wilson, C. B., Cell kinetic analyses of hu-
1 Avissar, S., Amitai, G. and Sokolovsky, M., Oligomeric structure of muscarinic receptors is shown by high- and low affinity agonist states, Proc. nat. Acad. Sci., U.S.A., 80 (1983) 156--159. 2 Bartus, A., Strategies for the Development of an Effective Treatment for Senile Dementia, S. Gershon and T. Crook (Eds.), Mark Powley Associates, New Canaan, 1981, pp. 79--90. 3 DeReuck, J., Roels, H. and Van der Eekem, H., Cytophotometric DNA determination in human astroglial tumors, Histopathology, 3 (1979) 107-115. 4 Davies, P. and Verth, A. H., Regional distribution of muscarinic acetylcholine receptor in normal and Alzheimer's type dementia brains, Brain Research, 138 (1978) 385-392. 5 Davis, K. L. and Mobs, R. C., Enhancement of memory processes in Alzheimer's Disease with multiple-dose intra-
ACKNOWLEDGEMENTS We appreciate the help of Prof. Israel Silman and Mrs. Esther Roth in cholinesterase measurements, and we are grateful to Prof. A. Bartal for his continuous support and encouragement. This study was supported by the US Army Medical Research and Development Command (contract no. D A M D 17-82-C2145 to H.S.), by the Hermann and Lilly Schilling Foundation for Medical Research (grant to H.S.) and by the Recanati Fund for Medical Research, Israel (grant to M.S.).
70 man malignant brain tumors (gliomas), Cancer, 44 (1979) 956-962. 11 Johnson, C. D. and Russell, R. L., A rapid, simple radiometric assay for cholinesterase, suitable for multiple determinations, Analyt. Biochem., 64 (1975) 229-238. 12 Klein, W. L., Nathanson, N. M. and Nirenberg, M., Muscarinic acetylcholine receptor regulation by accelerated rate of receptor loss, Biochem. biophys. Res. Commun., 90 (1979) 506-512. 13 Kloog, Y., Egozi, Y. and Sokolovsky, M., Characterization of muscarinic acetylcholine receptors from mice brain: evidence for regional heterogeneity and isomerization, Molec. Pharmacol., 15 (1979) 545-558. 14 Leyva, A. and Kelley, W. N., Measurement of DNA in cultured human cells, Analvt. Biochem., 62 (1974) 173-179. 15 Libermann, T. A., Razon, N., Bartal, A. D., Yarden, Y., Schlessinger, J. and Soreq, H., Expression of epidermal growth factor receptors in human brain tumors, Cancer Res., 44 (1984) 753-760. 16 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the folin phenol reagent, J. biol. Chem., 193 (1951) 265--275. 17 Meeker, R. B. and Harden, T. K., Muscarinic cholinergic receptor-mediated activation of phosphodiesterase, Molec. Pharmacol., 22 (1982) 310-319. 18 Meeker, R. B. and Harden, T. K., Muscarinic cholinergic receptor-mediated control of cyclic AMP metabolism, Molec. Pharmacol., 23 (1983) 384-392. 19 Nordberg, A. and Winblad, B., Cholinergic receptors in human hippocampus: regional distribution and variance with age, Life Sci., 29 (1981) 1937-1944. 20 Perry, E. K., Tomlinson, B. E., Blessed, G., Bergman, K., Bigson, P. H. and Perry, R. H., Correlation of cholinergic abnormalities with senile placques and mental test scores in senile dementia, Brit. Med. J., 2 (1978) 1457-1459. 21 Roberts, E., Potential therapies in aging and senile dementias. In: Alzheimer's disease, Down syndrome and aging. Ann. N. Y. Acad. Sci., 396 (1982) 165-178. 22 Rotter, A., Birdsall, N. J. M., Field, P. M. and Raisman, G., Muscarinic receptors in the central nervous system of the rat. II. Distribution of binding of 3H-propylbenzilylcho-
line mustard in the midbrain and hindbrain, Brain Res. Rev., 1 (1979) 167-183. 23 Russell, D. S. and Rubinstein, L. J., Primary tumors of neuroectodermal origin. In Pathology of the Nervous System, Edward Arnold Ltd., 1963, pp. 93--162. 24 Russell, D. S. and Rubinstein, L. J., Pathology of tumors of the nervous system, (1971) pp. 109-113. 25 Shifrin, G. S. and Klein, W. L., Regulation of muscarinic acetylcholine receptor concentration in cloned neuroblastoma cells, J. Neurochem., 34 (1980) 993-999. 26 Sokolovsky, M., Gurwitz, D. and Kloog, Y., Biochemical characterization of the muscarinic receptors, Advanc. Enzymol., 55 (1983) 137-196. 27 Soreq, H., Gurwitz, D., Eliyahu, D. and Sokolovsky, M., Altered ontogenesis of muscarinic receptors in agranular cerebellar cortex, J. Neurochem., 39 (1982) 756--763. 28 Soreq, H., Parvari, R. and Silman, I., Biosynthesis and secretion of catalytically active acetylcholinesterase in Xenopus oocytes microinjected with mRNA from rat brain and from Torpedo electric organ, Proc. nat. Acad. Sci. U.S.A., 79 (1982) 830-834. 29 Terry, R. D. and Davies, P., Dementia of the Alzheimer type, Ann. Rev. Neurosci., 3 (1980) 77-95. 30 Wastek, G. J., Stern, L. Z., Johnson, P. C. and Yamamura, H. I., Huntington's disease: regional alteration in muscarinic cholinergic receptor binding in human brain~ Life Sci., 19 (1976) 1033-1040. 31 Wastek, G. J. and Yamamura, H. I., Biochemical characterization of the muscarinic cholinergic receptor in human brain: alterations in Huntington's Disease, Molec. Pharmacol., 14 (1978) 768-780. 32 White, P., Hiley, C. R., Goodhart, M. J., Carrasco, L. H., Keet, J. P., Williams, I. E. I. and Bowen, D. M., Neocortical cholinergic neurons in elderly people, Lancet, March 26 (1977) 668--671. 33 Yates, C. M., Simpson, J., Maloney, A. F. J., Gordon, A. and Reid, A. H., Alzheimer-like cholinergic deficiency in Down syndrome, Lancet, 2 (1980) 979. 34 Zulch, K. J,, Principles of the New World Health Organization (WHO) classification of Brain Tumors, Neuroradiology, 19 (1980) 59-66.