- Email: [email protected]
Glioma Cells Induce γ-Glutamyl Transpeptidase Activity in Cultured Blood but Not Lymphatic Endothelial Cells
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
225, 1040–1044 (1996)
1291
Glioma Cells Induce g-Glutamyl Transpeptidase Activity in Cultured Blood but Not Lymphatic Endothelial Cells E. Weber,1 P. Lorenzoni, N. Raffaelli, N. Cavina, and G. Sacchi Istituto di Anatomia Umana Normale, University of Siena, 53100 Siena, Italy Received July 12, 1996 Glial cells have been shown to induce enzyme activities, which are characteristically high in brain capillaries, in cerebral and non cerebral endothelial cells in culture. We evaluated the activity of g-glutamyl transpeptidase (gGT) in freshly isolated (ex vivo) bovine blood (BEC) and lymphatic (LEC) endothelial cells. We also tested the effect of C6 glioma cell conditioned medium (GCM) on gGT activity of BEC and LEC in primary culture. Ex vivo BEC had a high gGT activity, only 9% of which was retained in culture. After exposure to GCM, however, gGT activity of cultured BEC was twofold higher than with control medium. By contrast gGT activity was extremely low in ex vivo LEC and did not significantly increase in cultured LEC exposed to GCM. These data show that the basal levels of gGT are markedly different in BEC and LEC and also that, unlike BEC, LEC are not capable of producing more gGT in response to glial stimulation. q 1996 Academic Press, Inc.
g-Glutamyl transpeptidase (gGT) catalyzes the transfer of the g-glutamyl moiety of glutathione to aminoacids, dipeptides, or glutathione itself with the formation of the respective gglutamyl derivatives. If the nucleophile is water, the overall result is hydrolysis (2). Brain capillary endothelial cells contain high levels of gGT which presumably contributes to selective transport of aminoacids across the blood brain barrier (3). This enzymatic activity markedly decreases in culture and is restored when brain capillary endothelial cells are co-cultured with astrocytes or glioma cells (4–6) or exposed to their conditioned medium (7). Interestingly the effect of glial cells is not limited to cerebral endothelial cells but it also causes differentiation of non cerebral endothelial cells with respect of alkaline phosphatase activity (8,9) and tight junction formation (8,10,11). gGT activity also increases with respect to controls in bovine aortic endothelial cells cultured on glial extracellular matrix (12) and in human umbilical vein endothelial cells co-cultured with astrocytes (13). Lymphatic endothelial cells are similar to blood endothelial cells in many ways and can be expected to have similar basal levels of gGT activity and a similar response to glial stimulation. In this study we evaluated gGT activity in freshly isolated (ex vivo) bovine lymphatic (LEC) and blood (BEC) endothelial cells and the effect of C6 glioma cell conditioned medium on gGT activity of LEC and BEC in primary culture. Our aim was to contribute to determine whether lymphatic vessels arise from the embryonic blood system or from mesenchymal anlages (1). We postulated that if we could demonstrate that lymphatic and blood endothelial cells respond similarly to the same environmental stimuli, this would support the hypothesis of a common origin. MATERIALS AND METHODS Ex vivo cells. Freshly isolated LEC and BEC (ex vivo cells) were used to determine basal levels of gGT. LEC were obtained from bovine thoracic duct by collagenase digestion (collagenase H, Boehringer Mannheim, GmbH,
1 To whom correspondence should be addressed at Istituto di Anatomia Umana Normale, via Aldo Moro, 53100 Siena, Italy. Fax: 577/227069.
1040 0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5221
/
6907$$$$81
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Phase contrast micrographs of confluent monolayers of primary culture BEC (a) and LEC (b) showing polygonal and closely juxtaposed cells. BarÅ70mm.
Germany, 0.215 U/mg, 0.2% in phosphate buffered saline, PBS, 15 min. at 377C) as previously described (14). BEC were obtained from bovine thoracic aorta by the same enzymatic treatment. Routinely the yield of 6-7 thoracic ducts was pooled into one sample to achieve the same number of cells as from a segment of thoracic aorta and the cells were split equally into two aliquots, one of which was used to determine protein content, the other gGT activity. Cell cultures. LEC and BEC obtained as ex vivo cells were seeded onto gelatin coated 35mm Costar 6-well plates at a density of 2.51105 cells/plate and kept in medium A: DMEM (GIBCO) containing 15% fetal bovine serum (GIBCO), 5% horse serum (Sigma), 100 mg/ml endothelial cell growth supplement (Sigma), 100 mg/ml heparin (Sigma) 50 U/ml penicillin and 50 mg/ml streptomycin (GIBCO). When the cells were still elongated and confluence was expected in approximately 24 h, medium A was replaced with medium B which contained 4% fetal bovine serum, 15% horse serum, heparin and antibiotics. This medium was designed to slow down cell proliferation. Cells were kept for 72 h in this medium, then washed 3 times in PBS and the medium replaced with a 1:1 mixture of C6 glioma cell conditioned medium and fresh medium B containing 200 mg/ml of heparin (GCM) or control medium (plain medium B). gGT activity in GCM and in control medium was less than 1U/ml. Three days later they were washed three times in PBS without calcium and magnesium and harvested with a 0.05% trypsin-0.02% versene solution (GIBCO). Each well yielded approximately 11106 cells. The content of two wells was pooled into one centrifuge tube and split equally into two aliquots, one of which was used to determine protein content, the other gGT activity. Identification of cells. Cultures of BEC and LEC were identified by cobblestone morphology at confluence (Fig.1) and by the uptake of (Fig.2) acetylated low density lipoproteins labeled with the fluorescent probe 1,1*dioctadecyl3,3,3*,3*-tetramethyl-indocarbocyanine perchlorate (DiI-Ac-LDL, Biomedical Technologies, Stoughton, MA) (15). C6 glioma cell conditioned medium. Cultures of C6 glioma cells, purchased from the American Type Culture Collection (Rockville, MD), were used to condition the medium. They were cultured in 25 cm2 Corning flasks according to the manufacturer’s instructions. At confluence the cells were washed with PBS and the medium replaced with 6ml of medium B without heparin. After 48h the medium was collected and centrifuged at 18001g for 10 min. to remove cell debris. The supernatant was sterile-filtered and used immediately or stored at 0207C until use. Determination of gGT content. Ex vivo (approximately 31106) or cultured (approximately 11106) LEC and BEC were washed twice in PBS, resuspended in 450 ml of 0.1 M Tris buffer pH 7.8, sonicated 10 sec. at 20W and assayed for gGT by the method of Huseby (16) modified by Edwards (17). Briefly, 200 ml of cell suspension was added to 1 ml of substrate (4.6 mM L-g-glutamyl-p-nitroanilide, Sigma, in Tris HCl containing 10 mM magnesium chloride) and preincubated for 10 min. at 377C. The reaction was started by adding the acceptor dipeptide (100 ml of glycylglycine, Sigma, 575 mM in distilled water, pH 7.5) and allowed to proceed for 40 min. at 377C. It was stopped by adding 1ml 8.5% cold trichloroacetic acid (TCA). After 10 min., 0.75 ml of 1M Tris buffer pH 9.6 was added to neutralize the TCA. The absorbance of p-nitroaniline was read at 410 nm in a Perkin-Elmer spectrophotometer. Controls were obtained as above except that glycyl-glycine was added after addition of TCA. Enzyme activity was calculated using a molar absorption coefficient of 10,500 M01cm01. A unit of enzyme activity is defined as the amount of gGT catalyzing formation of 1 nmole of p-nitroaniline per minute at 377C. Protein content was determined by the method of Lowry et al. (18) with bovine serum albumin as standard. The statistical significance of differences between 1041
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Intense staining of BEC (a) and LEC (b) with DiI-Ac-LDL. BarÅ20mm.
sets of experimental data was assessed using Student’s t test. Unless otherwise stated, data are presented as mean and 95% confidence intervals (C.I.).
RESULTS
We evaluated the activity of gGT in ex vivo freshly isolated BEC and LEC. The results obtained are shown in Fig.3. Mean activity of gGT in BEC was 44.1 U gGT/mg protein (C.I. 26.7-72.7). Extensive individual variations ranging from 13 to 146 U gGT/mg protein were observed between different bovines. By contrast the activity of gGT of ex vivo freshly isolated LEC was much lower (mean 0.1 U gGT/mg protein C.I. 0.0-1.5). The difference between gGT activity in ex vivo BEC and LEC was highly significant (Põ0.001, Student’s t test for unpaired data). The effect of GCM on gGT levels in cultured LEC and BEC is shown in Fig.4. The activity of gGT was twofold higher in primary cultures of BEC exposed to GCM (mean 7.3 U gGT/ mg protein, 95% C.I. 6.0-8.6), compared to controls (mean 3.9 U gGT/mg protein, C.I. 2.05.9). This difference was highly statistically significant (Põ0.005, Student’s t test for paired data). On the other hand, the activity of gGT in cultured control LEC was 0.1, C.I. 0.0-0.2 U gGT/mg protein and increased (not significantly) to 0.3, C.I. 0.0-1.0 in cultured LEC exposed to GCM.
FIG. 3. gGT activity in ex vivo BEC and LEC (U/mg protein, mean{SE, *ÅPõ0.001). 1042
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. gGT activity (U/mg protein, mean{SE) in cultured BEC and LEC. gGT activity increased significantly in cultured BEC exposed to GCM with respect to controls. *ÅPõ0.005. The difference between gGT activity in cultured LEC exposed to GCM and controls was not significant.
In summary, gGT activity in cultured BEC was 9% of the value observed in ex vivo cells and showed a twofold increase with respect to controls, when cultured BEC were stimulated with CM of C6 glioma cells; gGT activity in ex vivo LEC was one 440th that of ex vivo BEC and did not significantly increase in culture when stimulated with GCM. DISCUSSION
Our results show that LEC differ substantially from BEC in terms of gGT activity. In fact ex vivo LEC were found to have extremely low levels of gGT activity, and gGT activity of cultured LEC was not influenced by GCM. gGT activity is notoriously present in serum and at least part of the gGT activity found in ex vivo BEC might have been adsorbed from serum. However the induction of gGT in cultured BEC by GCM clearly demonstrates that BEC do produce gGT activity. It seems unlikely that the much lower gGT activity in cultured BEC with respect to ex vivo BEC may depend on the enzyme treatments used: trypsin, used for detaching the cells from cultured surfaces, does not affect gGT activity in endothelial cells (19); collagenase may contain traces of contaminants which destroy gGT, however it was used for the same length of time and at the same concentration in the isolation of both cultured and ex vivo cells. In our opinion, the reduction of gGT activity in cultured BEC to 9% of that observed in ex vivo cells is more plausibly due to cell proliferation and adaptation to culture conditions. Our findings are consistent with those of Mischeck (20) and Meyer (21) who reported that the activities of gGT and alkaline phosphatase decreased in brain capillary endothelial cells in primary culture with increasing cell proliferation. An inverse correlation between proliferation and differentiation was postulated by Maxwell who found that astrocyte conditioned medium, which enhanced gGT activity (7) and glucose analogue uptake (22) by cerebral microvessel endothelial cells, also inhibited mitosis. This suggests that astrocytes signal the endothelial cells to cease mitosis and to begin functioning as a more differentiated cerebral endothelial cell. The fact that LEC do not respond to GCM suggests that the extremely low basal levels of gGT in ex vivo LEC with respect to BEC are not due to different environmental conditions but rather to an intrinsic incapacity of LEC to respond to this stimulus. In conclusion our data suggest that either the origin of lymphatic endothelium is different from that of blood endothelium or that the endothelium of lymphatic vessels differentiates to such an extent that it loses the capacity to respond to the unknown glial factor/s responsible for gGT induction in BEC. 1043
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ACKNOWLEDGMENTS This work was financed by the Ministero dell’ Universita` e della Ricerca Scientifica e Tecnologica, Rome, Italy. We thank Dr. Aldo Paolicchi of the University of Pisa for helpful discussion.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
Kotani, M. (1990) Arch. Histol. Cytol. 53(Suppl.), 1–76. Meister, A., Tate, S. S., and Griffith, O. W. (1981) Methods Enzymology 77, 237–253. Orlowski, M., Sessa, G., and Green, J. P. (1974) Science 184, 66–68. De Bault, L. E., and Cancilla, P. A. (1980) Science 207, 653–655. De Bault, L. E. (1981) Brain Res. 220, 432–435. Tontsch, U., and Bauer, H. C. (1991) Brain Res. 539, 247–253. Maxwell, K., Berliner, J. A., and Cancilla, P. A. (1987) Brain Res. 410, 309–314. Tio, S., Deenen, M., and Marani, E. (1990) Eur. J. Morphol. 28, 289–300. Takemoto, H., Kaneda, K., Hosokawa, M., Ide, M., and Fukushima, H. (1994) FEBS Lett. 350, 99–103. Shivers, R. R., Arthur, F. E., and Bowman, P. D. (1988) J. Submicrosc. Cytol. Pathol. 20, 1–14. Tagami, M., Yamagata, K., Fujino, H., Kubota, A., Nara, Y., and Yamori, Y. (1992) Cell. Tissue Res. 268, 225– 232. Mizuguchi, H., Hashioka, Y., Fujii, A., Utoguchi, N., Kubo, K., Nakagawa, S., Baba, A., and Mayumi, T. (1994) Brain Res. 651, 155–159. Hurwitz, A. A., Berman, J. W., Rashbaum, W. K., and Lyman, W. D. (1993) Brain Res. 625, 238–243. Weber, E., Lorenzoni, P., Lozzi, G., and Sacchi, G. (1994) In Vitro Cell. Dev. Biol. 30A, 287–288. Voyta, J. C., Via, D. P., Butterfield, C. E., and Zetter, B. R. (1984) J. Cell. Biol. 99, 2034–2040. Huseby, N. E., and Stromme, J. H. (1974) Scand. J. Clin. Lab. Invest. 34, 357–363. Edwards, A. M. (1982) Cancer Res. 42, 1107–1115. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. Caspers, M. L., and Diglio, C. A. (1984) Biochim. Biophys. Acta 803, 1–6. Mischeck, U., Meyer, J., and Galla, H. J. (1989) Cell. Tissue Res. 256, 221–226. Meyer, J., Mischeck, U., Veyhl, M., Henzel, K., and Galla, H. J. (1990) Brain Res. 514, 305–309. Maxwell, K., Berliner, J. A., and Cancilla, P. A. (1989) J. Neuropathol. Exp. Neurol. 48, 69–80.
1044
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
225, 1040–1044 (1996)
1291
Glioma Cells Induce g-Glutamyl Transpeptidase Activity in Cultured Blood but Not Lymphatic Endothelial Cells E. Weber,1 P. Lorenzoni, N. Raffaelli, N. Cavina, and G. Sacchi Istituto di Anatomia Umana Normale, University of Siena, 53100 Siena, Italy Received July 12, 1996 Glial cells have been shown to induce enzyme activities, which are characteristically high in brain capillaries, in cerebral and non cerebral endothelial cells in culture. We evaluated the activity of g-glutamyl transpeptidase (gGT) in freshly isolated (ex vivo) bovine blood (BEC) and lymphatic (LEC) endothelial cells. We also tested the effect of C6 glioma cell conditioned medium (GCM) on gGT activity of BEC and LEC in primary culture. Ex vivo BEC had a high gGT activity, only 9% of which was retained in culture. After exposure to GCM, however, gGT activity of cultured BEC was twofold higher than with control medium. By contrast gGT activity was extremely low in ex vivo LEC and did not significantly increase in cultured LEC exposed to GCM. These data show that the basal levels of gGT are markedly different in BEC and LEC and also that, unlike BEC, LEC are not capable of producing more gGT in response to glial stimulation. q 1996 Academic Press, Inc.
g-Glutamyl transpeptidase (gGT) catalyzes the transfer of the g-glutamyl moiety of glutathione to aminoacids, dipeptides, or glutathione itself with the formation of the respective gglutamyl derivatives. If the nucleophile is water, the overall result is hydrolysis (2). Brain capillary endothelial cells contain high levels of gGT which presumably contributes to selective transport of aminoacids across the blood brain barrier (3). This enzymatic activity markedly decreases in culture and is restored when brain capillary endothelial cells are co-cultured with astrocytes or glioma cells (4–6) or exposed to their conditioned medium (7). Interestingly the effect of glial cells is not limited to cerebral endothelial cells but it also causes differentiation of non cerebral endothelial cells with respect of alkaline phosphatase activity (8,9) and tight junction formation (8,10,11). gGT activity also increases with respect to controls in bovine aortic endothelial cells cultured on glial extracellular matrix (12) and in human umbilical vein endothelial cells co-cultured with astrocytes (13). Lymphatic endothelial cells are similar to blood endothelial cells in many ways and can be expected to have similar basal levels of gGT activity and a similar response to glial stimulation. In this study we evaluated gGT activity in freshly isolated (ex vivo) bovine lymphatic (LEC) and blood (BEC) endothelial cells and the effect of C6 glioma cell conditioned medium on gGT activity of LEC and BEC in primary culture. Our aim was to contribute to determine whether lymphatic vessels arise from the embryonic blood system or from mesenchymal anlages (1). We postulated that if we could demonstrate that lymphatic and blood endothelial cells respond similarly to the same environmental stimuli, this would support the hypothesis of a common origin. MATERIALS AND METHODS Ex vivo cells. Freshly isolated LEC and BEC (ex vivo cells) were used to determine basal levels of gGT. LEC were obtained from bovine thoracic duct by collagenase digestion (collagenase H, Boehringer Mannheim, GmbH,
1 To whom correspondence should be addressed at Istituto di Anatomia Umana Normale, via Aldo Moro, 53100 Siena, Italy. Fax: 577/227069.
1040 0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5221
/
6907$$$$81
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Phase contrast micrographs of confluent monolayers of primary culture BEC (a) and LEC (b) showing polygonal and closely juxtaposed cells. BarÅ70mm.
Germany, 0.215 U/mg, 0.2% in phosphate buffered saline, PBS, 15 min. at 377C) as previously described (14). BEC were obtained from bovine thoracic aorta by the same enzymatic treatment. Routinely the yield of 6-7 thoracic ducts was pooled into one sample to achieve the same number of cells as from a segment of thoracic aorta and the cells were split equally into two aliquots, one of which was used to determine protein content, the other gGT activity. Cell cultures. LEC and BEC obtained as ex vivo cells were seeded onto gelatin coated 35mm Costar 6-well plates at a density of 2.51105 cells/plate and kept in medium A: DMEM (GIBCO) containing 15% fetal bovine serum (GIBCO), 5% horse serum (Sigma), 100 mg/ml endothelial cell growth supplement (Sigma), 100 mg/ml heparin (Sigma) 50 U/ml penicillin and 50 mg/ml streptomycin (GIBCO). When the cells were still elongated and confluence was expected in approximately 24 h, medium A was replaced with medium B which contained 4% fetal bovine serum, 15% horse serum, heparin and antibiotics. This medium was designed to slow down cell proliferation. Cells were kept for 72 h in this medium, then washed 3 times in PBS and the medium replaced with a 1:1 mixture of C6 glioma cell conditioned medium and fresh medium B containing 200 mg/ml of heparin (GCM) or control medium (plain medium B). gGT activity in GCM and in control medium was less than 1U/ml. Three days later they were washed three times in PBS without calcium and magnesium and harvested with a 0.05% trypsin-0.02% versene solution (GIBCO). Each well yielded approximately 11106 cells. The content of two wells was pooled into one centrifuge tube and split equally into two aliquots, one of which was used to determine protein content, the other gGT activity. Identification of cells. Cultures of BEC and LEC were identified by cobblestone morphology at confluence (Fig.1) and by the uptake of (Fig.2) acetylated low density lipoproteins labeled with the fluorescent probe 1,1*dioctadecyl3,3,3*,3*-tetramethyl-indocarbocyanine perchlorate (DiI-Ac-LDL, Biomedical Technologies, Stoughton, MA) (15). C6 glioma cell conditioned medium. Cultures of C6 glioma cells, purchased from the American Type Culture Collection (Rockville, MD), were used to condition the medium. They were cultured in 25 cm2 Corning flasks according to the manufacturer’s instructions. At confluence the cells were washed with PBS and the medium replaced with 6ml of medium B without heparin. After 48h the medium was collected and centrifuged at 18001g for 10 min. to remove cell debris. The supernatant was sterile-filtered and used immediately or stored at 0207C until use. Determination of gGT content. Ex vivo (approximately 31106) or cultured (approximately 11106) LEC and BEC were washed twice in PBS, resuspended in 450 ml of 0.1 M Tris buffer pH 7.8, sonicated 10 sec. at 20W and assayed for gGT by the method of Huseby (16) modified by Edwards (17). Briefly, 200 ml of cell suspension was added to 1 ml of substrate (4.6 mM L-g-glutamyl-p-nitroanilide, Sigma, in Tris HCl containing 10 mM magnesium chloride) and preincubated for 10 min. at 377C. The reaction was started by adding the acceptor dipeptide (100 ml of glycylglycine, Sigma, 575 mM in distilled water, pH 7.5) and allowed to proceed for 40 min. at 377C. It was stopped by adding 1ml 8.5% cold trichloroacetic acid (TCA). After 10 min., 0.75 ml of 1M Tris buffer pH 9.6 was added to neutralize the TCA. The absorbance of p-nitroaniline was read at 410 nm in a Perkin-Elmer spectrophotometer. Controls were obtained as above except that glycyl-glycine was added after addition of TCA. Enzyme activity was calculated using a molar absorption coefficient of 10,500 M01cm01. A unit of enzyme activity is defined as the amount of gGT catalyzing formation of 1 nmole of p-nitroaniline per minute at 377C. Protein content was determined by the method of Lowry et al. (18) with bovine serum albumin as standard. The statistical significance of differences between 1041
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Intense staining of BEC (a) and LEC (b) with DiI-Ac-LDL. BarÅ20mm.
sets of experimental data was assessed using Student’s t test. Unless otherwise stated, data are presented as mean and 95% confidence intervals (C.I.).
RESULTS
We evaluated the activity of gGT in ex vivo freshly isolated BEC and LEC. The results obtained are shown in Fig.3. Mean activity of gGT in BEC was 44.1 U gGT/mg protein (C.I. 26.7-72.7). Extensive individual variations ranging from 13 to 146 U gGT/mg protein were observed between different bovines. By contrast the activity of gGT of ex vivo freshly isolated LEC was much lower (mean 0.1 U gGT/mg protein C.I. 0.0-1.5). The difference between gGT activity in ex vivo BEC and LEC was highly significant (Põ0.001, Student’s t test for unpaired data). The effect of GCM on gGT levels in cultured LEC and BEC is shown in Fig.4. The activity of gGT was twofold higher in primary cultures of BEC exposed to GCM (mean 7.3 U gGT/ mg protein, 95% C.I. 6.0-8.6), compared to controls (mean 3.9 U gGT/mg protein, C.I. 2.05.9). This difference was highly statistically significant (Põ0.005, Student’s t test for paired data). On the other hand, the activity of gGT in cultured control LEC was 0.1, C.I. 0.0-0.2 U gGT/mg protein and increased (not significantly) to 0.3, C.I. 0.0-1.0 in cultured LEC exposed to GCM.
FIG. 3. gGT activity in ex vivo BEC and LEC (U/mg protein, mean{SE, *ÅPõ0.001). 1042
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. gGT activity (U/mg protein, mean{SE) in cultured BEC and LEC. gGT activity increased significantly in cultured BEC exposed to GCM with respect to controls. *ÅPõ0.005. The difference between gGT activity in cultured LEC exposed to GCM and controls was not significant.
In summary, gGT activity in cultured BEC was 9% of the value observed in ex vivo cells and showed a twofold increase with respect to controls, when cultured BEC were stimulated with CM of C6 glioma cells; gGT activity in ex vivo LEC was one 440th that of ex vivo BEC and did not significantly increase in culture when stimulated with GCM. DISCUSSION
Our results show that LEC differ substantially from BEC in terms of gGT activity. In fact ex vivo LEC were found to have extremely low levels of gGT activity, and gGT activity of cultured LEC was not influenced by GCM. gGT activity is notoriously present in serum and at least part of the gGT activity found in ex vivo BEC might have been adsorbed from serum. However the induction of gGT in cultured BEC by GCM clearly demonstrates that BEC do produce gGT activity. It seems unlikely that the much lower gGT activity in cultured BEC with respect to ex vivo BEC may depend on the enzyme treatments used: trypsin, used for detaching the cells from cultured surfaces, does not affect gGT activity in endothelial cells (19); collagenase may contain traces of contaminants which destroy gGT, however it was used for the same length of time and at the same concentration in the isolation of both cultured and ex vivo cells. In our opinion, the reduction of gGT activity in cultured BEC to 9% of that observed in ex vivo cells is more plausibly due to cell proliferation and adaptation to culture conditions. Our findings are consistent with those of Mischeck (20) and Meyer (21) who reported that the activities of gGT and alkaline phosphatase decreased in brain capillary endothelial cells in primary culture with increasing cell proliferation. An inverse correlation between proliferation and differentiation was postulated by Maxwell who found that astrocyte conditioned medium, which enhanced gGT activity (7) and glucose analogue uptake (22) by cerebral microvessel endothelial cells, also inhibited mitosis. This suggests that astrocytes signal the endothelial cells to cease mitosis and to begin functioning as a more differentiated cerebral endothelial cell. The fact that LEC do not respond to GCM suggests that the extremely low basal levels of gGT in ex vivo LEC with respect to BEC are not due to different environmental conditions but rather to an intrinsic incapacity of LEC to respond to this stimulus. In conclusion our data suggest that either the origin of lymphatic endothelium is different from that of blood endothelium or that the endothelium of lymphatic vessels differentiates to such an extent that it loses the capacity to respond to the unknown glial factor/s responsible for gGT induction in BEC. 1043
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC
Vol. 225, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ACKNOWLEDGMENTS This work was financed by the Ministero dell’ Universita` e della Ricerca Scientifica e Tecnologica, Rome, Italy. We thank Dr. Aldo Paolicchi of the University of Pisa for helpful discussion.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
Kotani, M. (1990) Arch. Histol. Cytol. 53(Suppl.), 1–76. Meister, A., Tate, S. S., and Griffith, O. W. (1981) Methods Enzymology 77, 237–253. Orlowski, M., Sessa, G., and Green, J. P. (1974) Science 184, 66–68. De Bault, L. E., and Cancilla, P. A. (1980) Science 207, 653–655. De Bault, L. E. (1981) Brain Res. 220, 432–435. Tontsch, U., and Bauer, H. C. (1991) Brain Res. 539, 247–253. Maxwell, K., Berliner, J. A., and Cancilla, P. A. (1987) Brain Res. 410, 309–314. Tio, S., Deenen, M., and Marani, E. (1990) Eur. J. Morphol. 28, 289–300. Takemoto, H., Kaneda, K., Hosokawa, M., Ide, M., and Fukushima, H. (1994) FEBS Lett. 350, 99–103. Shivers, R. R., Arthur, F. E., and Bowman, P. D. (1988) J. Submicrosc. Cytol. Pathol. 20, 1–14. Tagami, M., Yamagata, K., Fujino, H., Kubota, A., Nara, Y., and Yamori, Y. (1992) Cell. Tissue Res. 268, 225– 232. Mizuguchi, H., Hashioka, Y., Fujii, A., Utoguchi, N., Kubo, K., Nakagawa, S., Baba, A., and Mayumi, T. (1994) Brain Res. 651, 155–159. Hurwitz, A. A., Berman, J. W., Rashbaum, W. K., and Lyman, W. D. (1993) Brain Res. 625, 238–243. Weber, E., Lorenzoni, P., Lozzi, G., and Sacchi, G. (1994) In Vitro Cell. Dev. Biol. 30A, 287–288. Voyta, J. C., Via, D. P., Butterfield, C. E., and Zetter, B. R. (1984) J. Cell. Biol. 99, 2034–2040. Huseby, N. E., and Stromme, J. H. (1974) Scand. J. Clin. Lab. Invest. 34, 357–363. Edwards, A. M. (1982) Cancer Res. 42, 1107–1115. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. Caspers, M. L., and Diglio, C. A. (1984) Biochim. Biophys. Acta 803, 1–6. Mischeck, U., Meyer, J., and Galla, H. J. (1989) Cell. Tissue Res. 256, 221–226. Meyer, J., Mischeck, U., Veyhl, M., Henzel, K., and Galla, H. J. (1990) Brain Res. 514, 305–309. Maxwell, K., Berliner, J. A., and Cancilla, P. A. (1989) J. Neuropathol. Exp. Neurol. 48, 69–80.
1044
AID
BBRC 5221
/
6907$$$$82
08-06-96 11:00:20
bbrca
AP: BBRC