|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 19, 2004 as doi:10.1096/fj.04-1675fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

* Laboratory of Immunology and Signal Transduction, Department of Experimental Medicine and Biochemical Sciences, University of Rome Tor Vergata, Rome, Italy; and
Department of Molecular Genetics, Chiba University Graduate School of Medicine, Chiba, and Laboratory of Cell Signaling, RIKEN Center for Allergy and Immunology, Yokohama, Japan
1 Correspondence: Department of Experimental Medicine and Biochemical Sciences, University of Rome Tor Vergata, via Montpellier 1, 00133 Rome, Italy. E-mail: roberto.testi{at}flashnet.it
SPECIFIC AIMS
Stress-induced intracellular GD3 accumulation can be followed by adaptation and differentiation or mitochondrial damage and apoptosis. We asked whether this could be dictated by subcellular localization of the GD3 synthase, which is in turn controlled by calnexin.
PRINCIPAL FINDINGS
1. Calnexin retains the GD3 synthase (ST8) in the endoplasmic reticulum (ER)
Chicken ST8 was shown to be normally expressed in the Golgi. Accordingly, the enforced expression of the human ST8 in Chinese hamster ovary (CHO) cells results in its accumulation at the Golgi level (Fig. 1
A). Since the ST8, via its multiple N-linked glycans, interacts with calnexin in the ER, we asked whether increasing levels of calnexin could affect localization of the ST8. Calnexin was therefore transiently coexpressed together with the ST8 in CHO cells and the localization of ST8 in relation to the Golgi was investigated. Figure 1A
shows that overexpression of calnexin prevents the accumulation of the ST8 in the Golgi. The localization of coexpressed ST8 and calnexin was then analyzed at higher resolution in CHO cells. Figure 1B
shows that coexpressed ST8 and calnexin colocalize to the same subcellular compartment.
|
2. ST8 retained in the ER by calnexin remains functional
Since calnexin is essentially confined to the ER, the above findings suggested that its overexpression results in almost complete retention of the ST8 in the ER. To investigate whether calnexin-induced ER retention affects the activity of the ST8, the accumulation of de novo synthesized GD3 was measured in CHO cells transiently overexpressing only ST8 or both ST8 and calnexin. We found that coexpression of calnexin has no effect on the ability of the ST8 to de novo synthesize GD3 (not shown). Earlier studies indicated that, although ST8 is rapidly degraded if retained in the ER by the inhibition of glycan-trimming enzymes with castanospermine, which prevents the interaction with calnexin, ER-retained ST8 binds to GM3 and enables the synthesis of GD3. Thus, when retained in the ER by the interaction with calnexin, ST8 remains functional.
3. ST8 retention in the ER prevents the relocalization of de novo synthesized GD3 to mitochondria
The intracellular accumulation of de novo synthesized GD3, which occurs during ceramide-mediated stress responses, may be followed by its relocation to the mitochondria, causing triggering of the apoptotic program. We asked whether retention of ST8 in the ER could affect the rate of mitochondrial relocalization of de novo synthesized GD3. ST8 was transiently expressed, alone or together with calnexin, in CHO cells and the accumulation of GD3 in relation to mitochondria was investigated. We observed that de novo synthesized GD3 relocates to the mitochondria when cells express only ST8, but when calnexin is simultaneously expressed with ST8, GD3 fails to colocalize with the mitochondria (not shown).
4. Calnexin specifically suppresses ST8-induced apoptosis
The above results suggested that high levels of calnexin, by preventing the relocalization of GD3 to the mitochondria, could affect apoptosis. To directly investigate the possible interference of calnexin in the apoptotic program, apoptosis induction was measured in CHO cells transiently coexpressing ST8 and calnexin. We observed that calnexin overexpression substantially abrogates ST8-induced apoptosis. A calnexin mutant (
N2-CNX) that lacks the P domain, and therefore is unable to bind glycoproteins, has no inhibitory effect. To exclude a generic anti-apoptotic effect of calnexin, apoptosis induction was measured in CHO cells transiently transfected with the proapoptotic bcl2 family member bax. We found that calnexin cannot protect from bax-induced apoptosis (not shown). Our observations were verified in another cell type. Human T lymphoma cells HuT78 were transiently cotransfected with ST8 and calnexin, and apoptosis induction was quantitated. We found that calnexin almost completely suppresses ST8-induced apoptosis in HuT78 cells. A calnexin mutant that lacks the C-terminal ER-retention sequence (
C-CNX), and therefore cannot be retained in the ER, has no effect (not shown). Together, these data indicate that calnexin prevents ST8-induced apoptosis; this requires the capacity of calnexin to interact with glycoproteins and to be retained in the ER.
5. Loss of calnexin sensitizes to apoptosis
We then investigated whether the absence of calnexin could affect ST8-induced apoptosis. As a model for calnexin deficiency, we took advantage of the CEM-NKR cell line, a variant of the human T cell leukemia CEM line known to be defective in calnexin expression at the mRNA level. We stably reconstituted CEM-NKR cells with wild type calnexin and the
N2-CNX mutant. Figure 2
A shows that calnexin is absent in CEM-NKR cells, as opposed to the parental CEM cells, and that calnexin and the
N2-CNX mutant are expressed in the reconstituted CEM-NKR at appreciable levels. No calnexin is detectable in CEM-NKR cells stably expressing the empty vector.
|
CEM-NKR cells endogenously express a functional ST8 (not shown). Similar to CEM cells or other hemopoietic cells, they undergo apoptosis upon ST8 overexpression. As shown in Fig. 2B
, ST8 expression drives cellular apoptosis in CEM-NKR stably expressing the empty vector cells very efficiently. However, the stable reconstitution of calnexin in CEM-NKR provides resistance to ST8 expression, which cannot be achieved by the binding-defective
N2-CNX mutant. We then extended our analysis to ceramide-induced responses. Similar to what we observed with ST8, we found that CEM-NKR undergo apoptosis in response to ceramide exposure (Fig. 2C
). Stable reconstitution with calnexin, but not with its binding-defective mutant, confers ceramide resistance to CEM-NKR, at least to levels similar to those observed in the parental CEM cells (here also shown for comparison).
CONCLUSIONS AND SIGNIFICANCE
ST8-induced GD3 accumulation can mediate stress-induced apoptosis in different cellular systems. In some cell types, however, GD3 accumulation is not lethal. A critical question therefore is how can some normal or transformed cells survive acute GD3 accumulation, a capacity that substantially raises their stress resistance threshold and confers survival advantage. We have recently revealed a molecular mechanism that allows tumor cells to escape GD3 accumulation, which is the ability to acetylate and inactivate GD3. Here we show that, when retained in the ER by calnexin, ST8 activation does not result in apoptosis induction, since de novo synthesized GD3 does not reach the mitochondria. This observation raises the question as to why GD3 generated in the ER fails to reach the mitochondria. Alternative and not mutually exclusive possibilities are that GD3 ganglioside might be modified, then degraded in the ER, or critical glycolipid transporters might be missing in the ER. Calnexin-mediated ER retention of ST8 also prevents ceramide-induced apoptosis, suggesting that this mechanism may be more generally exploited to enhance cellular stress resistance. Calnexin levels can be up-regulated by growth factors and have been found to be elevated in breast cancer cells. Moreover, gene expression profiling has recently revealed that calnexin is highly expressed in multiple myeloma cells. Our data reveal a potential strategy for normal and transformed cells to enhance their survival and uncover a previously unrecognized role for the molecular chaperone calnexin in the control of cellular stress resistance (Fig. 3
).
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-1675fje;
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |