| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 8, 2002 as doi:10.1096/fj.01-0863fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Groningen University Institute for Drug Exploration, Department of Membrane Cell Biology, University of Groningen, The Netherlands
2Correspondence: University of Groningen, Department of Membrane Cell Biology, A. Deusinglaan 1, 9713 AV Groningen, The Netherlands. E-mail: j.w.kok{at}med.rug.nl
SPECIFIC AIM
The aim of our study was to acquire a complete picture of the changes in sphingolipid metabolism that occur in multidrug-resistant tumor cells and to establish whether the elevated glucosylceramide (GlcCer) level is due to increased expression and activity of glucosylceramide synthase (GCS) or is caused by another mechanism.
PRINCIPAL FINDINGS
1. Differential expression of sphingolipids by 2780AD vs. A2780 cells
Our study revealed remarkable alterations in the expression of sphingolipids by multidrug-resistant (MDR) cells (Table 1
). Compared with drug-sensitive A2780 cells, sphingomyelin (SM) levels of the resistant P-glycoprotein (Pgp) -overexpressing 2780AD cells were almost twice as high whereas GlcCer and galactosylceramide (GalCer) levels were even threefold higher. No significant differences were observed in levels of ceramide (Cer) and sphingosine (So). In contrast to the high levels of GlcCer in 2780AD cells, the expression of its direct product lactosylceramide (LacCer) was only 50% of that measured in wild-type A2780 cells (Table 1)
. Moreover, the levels of higher neutral glycolipids as well as the gangliosides GM3 and GM2 were substantially lower (mass analysis) in MDR cells.
|
2. Differential expression of sphingolipids is not the result of enhanced shedding of complex glycolipids
To examine shedding of glycolipids from 2780 cells, lipids were extracted from culture media in which the cells had grown for 48 h. On radiolabeling with L-[U-14C]serine, no [14C]LacCer was recovered from the medium fraction whereas the recoveries of [14C]Cer, [14C]SM and [14C]GlcCer (and gangliosides) were equal in resistant and parental 2780 cells.
3. Kinetics of sphingolipid metabolism in 2780AD vs. A2780 cells
We performed kinetic analysis of L-[U-14C]serine incorporation into sphingolipid species to obtain insight into how the difference in equilibrium levels of sphingolipids in the two cell types originated. Equilibrium levels for all sphingolipids were reached between 9 and 24 h, with different kinetics for [14C]LacCer vs. [14C]GlcCer and [14C]SM. The ratio of [14C]GlcCer in 2780AD over A2780 increased with time and reached its peak (around 3) only after 24 h. The ratio of [14C]LacCer in 2780AD over A2780 was around 0.5 after 1 h of incubation and remained stable.
4. Reduced formation of LacCer in 2780AD cells, which results in accumulation of GlcCer
The results presented so far suggest that the metabolism of sphingolipids in 2780AD cells is inhibited at the level of LacCer formation. Rigorous evidence came from measurements of sphingolipid conversion in situ. With [14C]n-hexanoic acid (C6)-GlcCer as a precursor, [14C]C6-LacCer formation was three- to fourfold attenuated in 2780AD cells compared with the parental A2780 cells. The biosynthesis of [14C]C6-GlcCer from [14C]C6-Cer was reduced in 2780AD cells. This result strongly suggests that the threefold higher level of GlcCer in 2780AD cells is not a result of changes in GlcCer biosynthesis, but originates from changes in LacCer formation. Indeed, analysis of GCS mRNA (by RT-PCR) and protein expression (by immunoblotting) did not reveal an increase in 2780AD cells.
5. Reduced LacCer formation in 2780AD cells is not the result of changes in LCS expression
The observed reduced formation of LacCer in 2780AD cells could have resulted from reduced expression and/or activity of the enzyme responsible for its biosynthesis, lactosylceramide synthase (LCS). However, the relative mRNA levels corresponding to LCS as determined by RT-PCR were similar in drug-sensitive A2780 and resistant 2780AD cells. Furthermore, the reduced formation of LacCer in 2780AD cells occurred only in intact cells. Relative to intact 2780AD cells, [14C]C6-LacCer formation increased fivefold on cell homogenization either by sonication or a freeze/thaw procedure. In A2780 cells, there was no difference between intact and broken cells. This indicates that an extensive suppression of [14C]C6-LacCer formation in intact 2780AD cells had occurred that is not due to reduced LacCer synthase activity per se.
6. Reduced LacCer formation in 2780AD cells is not the result of Pgp-mediated glucosylceramide translocation
As a potential sphingolipid translocator, Pgp could alter the subcellular distribution of GlcCer, resulting in a decrease of the availability of GlcCer for LacCer formation in the lumen of the Golgi apparatus. However, when Pgp function was effectively inhibited by either 10 µM cyclosporin A, 2 µM PSC833, or 0.2 µM GF120918, [14C]C6-LacCer formation increased only slightly in 2780AD cells (even after GF120918 preincubation) and did not reach nearly the level of A2780 cells. Upon [14C]serine incorporation, the [14C]sphingolipid profile of 2780AD cells did not revert to that of A2780 cells.
7. Reduced LacCer formation in 2780AD cells is the result of uncoupling from glucosylceramide biosynthesis
Biosynthesis of LacCer is not tightly coupled to that of GlcCer. There is a large difference in the rate of biosynthesis between the two glycolipids. The kinetics of GlcCer formation resembled more those of SM formation than those of LacCer formation. This suggests that LacCer biosynthesis in the Golgi apparatus is topologically segregated from GlcCer biosynthesis. By the action of brefeldin A (BFA), a large part of the Golgi apparatus merges with the endoplasmic reticulum, resulting in merging of substrates and enzymes of sphingolipid biosynthetic pathways. When A2780 cells were treated during 48 h with BFA (5 µg/ml), a large increase in sphingolipid biosynthesis occurred, with > ninefold increases in GlcCer and LacCer formation (Table 2
). However, when 2780AD were treated similarly, LacCer formation remained at control levels whereas GlcCer biosynthesis increased 2.5-fold (Table 2)
. Further evidence for a topological segregation between GlcCer and LacCer biosynthesis came from LCS green fluorescence protein (GFP) localization experiments after LCS-GFP cDNA transfection. In A2780 cells, LCS-GFP is found in the cell center in a perinuclear condensed compartment, where it colocalizes with the medial Golgi marker giantin. In resistant 2780AD cells, LCS-GFP is localized in a set of more peripherally located compartments and does not colocalize with giantin.
|
CONCLUSION AND SIGNIFICANCE
The 2780AD human ovarian carcinoma cell line is a well-established cell line that overexpresses Pgp, an ATP binding cassette (ABC) transporter protein that conveys resistance to doxorubicin and other cytotoxic agents. We revealed an extensive and coherent remodeling of sphingolipid metabolism in these cells compared with the parental nonresistant A2780 cell line. Accumulation of all direct biosynthetic products of Cer was observed. In sharp contrast, the levels of complex glycolipids beyond GlcCer, including gangliosides, were reduced. The decreased formation of LacCer we measured in 2780AD cells is a novel mechanism, which explains the decrease in the levels of all complex glycolipids. The threefold increase in the level of GlcCer in the MDR cells is consistent with results obtained in other drug-resistant cell lines such as MCF-7-AdrR, OVCAR-3, and KB-V-1. Enhanced GlcCer levels in MDR cells have been suggested to result from increased GCS activity in MCF-7-AdrR cells. We provide evidence for an alternative mechanism leading to changes in sphingolipid expression in 2780AD cells. Increased expression and activity of GCS were rigorously ruled out at the levels of transcription, translation, and in situ activity of the enzyme (Fig. 1
). Instead, the increase in GlcCer level in 2780AD cells appears to result from decreased conversion to LacCer. Although a contribution of reduced GlcCer breakdown to the enhanced levels of this glycolipid cannot be excluded, our data indicated that GlcCer breakdown activity did not differ between the two cell types.
|
Next we investigated potential mechanisms underlying the relative inhibition of LacCer formation in 2780AD cells. 1) The most straightforward mechanism, i.e., reduced expression or activity of the LCS enzyme (Fig. 1)
, was refuted by results from RT-PCR and cell-free LCS activity assays. 2) Given the observation that reduced LacCer formation occurred only in intact cells, we argued that substrate (GlcCer) availability to the LCS might be hampered in the MDR phenotype. In mitotic HeLa cells, suppression of higher glycosphingolipid synthesis and accumulation of Cer and GlcCer have been shown to occur. This was not the result of reduced LCS activity but was probably caused by vesiculation of membranous organelles. We propose that the extent of uncoupling of GlcCer and LacCer biosynthesis is higher in 2780AD cells (Fig. 1)
relative to A2780 cells, which would explain the high suppression of LacCer formation in resistant cells. In support of this mechanism are results we obtained with BFA. BFA has often been used to map subcellular locations of different glycosyltransferases involved in glycosphingolipid biosynthesis. An increase in biosynthesis of a specific sphingolipid occurs when the enzyme responsible flows to the ER on BFA-induced Golgi-ER fusion, because it was originally located in the cis to trans Golgi compartments prior to the BFA block. In A2780 cells, BFA largely increased the levels of GlcCer and LacCer, indicating that the GCS and LCS were both located prior to the BFA block. However, in 2780AD cells, no increase in LacCer level was observed despite an increased GlcCer level. Thus, in the MDR phenotype, BFA induces merging of GCS but not LCS with the ER, indicating that LCS is located distal to the BFA block. In line with this mechanism are observations on the kinetics of sphingolipid biosynthesis. When pulsed with radiolabeled serine, resistant 2780AD cells incorporate label in GlcCer, which is not readily converted to LacCer due the remote location of the LCS. This results in a gradual accumulation of radiolabeled GlcCer. Further evidence for the segregation of GCS and LCS was obtained with LCS-GFP localization experiments after LCS-GFP cDNA transfection of A2780 and resistant 2780AD cells. The results show that LCS-GFP is localized in 2780AD cells in a compartment distinct from the (medial) Golgi compartment harboring giantin. This localization of LCS-GFP is quite different from that in A2780, where it colocalizes with giantin. 3) It was important to establish whether reduced formation of LacCer was correlated to the activity of overexpressed Pgp in 2780AD cells. There is evidence indicating that sphingolipids can function as substrates for the ABC transporters Pgp and multidrug resistance protein 1. Removal of GlcCer from the cytoplasmic leaflet of the plasma membrane by Pgp may in turn lead to a cellular response in terms of increased flux of GlcCer from Golgi to plasma membrane (Fig. 1)
. However, the use of three different inhibitors of Pgp-mediated efflux activity in 2780AD cells did not reveal a shift in metabolism of (glyco)sphingolipids toward that of A2780 cells.
In conclusion, we have shown that MDR ovarian cancer cells present with multiple alterations in sphingolipid levels compared with parental drug-sensitive cells, revealing a coherent remodeling of sphingolipid metabolism. The apparent block at the level of LacCer formation and the accumulation of GlcCer are due to uncoupling of the GCS and LCS in the Golgi apparatus of MDR cells and not to altered expression or activity of these enzymes.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/fj.01-0863fje; to cite this article, use FASEB J. (May 8, 2002) 10.1096/fj.01-0863fje. 
This article has been cited by other articles:
|
|
 |
|
  M. F. De Rosa, C. Ackerley, B. Wang, S. Ito, D. M. Clarke, and C. Lingwood Inhibition of Multidrug Resistance by AdamantylGb3, a Globotriaosylceramide Analog J. Biol. Chem., February 22, 2008; 283(8): 4501 - 4511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P Y Lee, D. D Mruk, W. Xia, and C Y. Cheng Cellular localization of sphingomyelin synthase 2 in the seminiferous epithelium of adult rat testes J. Endocrinol., January 1, 2007; 192(1): 17 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. J. Hinrichs, K. Klappe, M. van Riezen, and J. W. Kok Drug resistance-associated changes in sphingolipids and ABC transporters occur in different regions of membrane domains J. Lipid Res., November 1, 2005; 46(11): 2367 - 2376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Gouaze, J. Y. Yu, R. J. Bleicher, T.-Y. Han, Y.-Y. Liu, H. Wang, M. M. Gottesman, A. Bitterman, A. E. Giuliano, and M. C. Cabot Overexpression of glucosylceramide synthase and P-glycoprotein in cancer cells selected for resistance to natural product chemotherapy Mol. Cancer Ther., May 1, 2004; 3(5): 633 - 640. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Grazide, A.-D. Terrisse, S. Lerouge, G. Laurent, and J.-P. Jaffrezou Cytoprotective Effect of Glucosylceramide Synthase Inhibition against Daunorubicin-induced Apoptosis in Human Leukemic Cell Lines J. Biol. Chem., April 30, 2004; 279(18): 18256 - 18261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Prinetti, L. Basso, V. Appierto, M. G. Villani, M. Valsecchi, N. Loberto, S. Prioni, V. Chigorno, E. Cavadini, F. Formelli, et al. Altered Sphingolipid Metabolism in N-(4-Hydroxyphenyl)- retinamide-resistant A2780 Human Ovarian Carcinoma Cells J. Biol. Chem., February 14, 2003; 278(8): 5574 - 5583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Paris, A. Morales, O. Coll, A. Sanchez-Reyes, C. Garcia-Ruiz, and J. C. Fernandez-Checa Ganglioside GD3 Sensitizes Human Hepatoma Cells to Cancer Therapy J. Biol. Chem., December 13, 2002; 277(51): 49870 - 49876. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
