Perspective

Signaling Pathways Regulating Golgi Structure and Function

See allHide authors and affiliations

Science's STKE  30 Oct 2001:
Vol. 2001, Issue 106, pp. pe38
DOI: 10.1126/stke.2001.106.pe38

Abstract

At the end of the cell division cycle, organelles must be equally distributed to two daughter cells. This includes the many stacks of the Golgi apparatus. Several kinases have been implicated in regulating Golgi disassembly during mitosis, but much about this process remains obscure.

Signal transduction pathways involving protein and lipid kinases regulate almost all aspects of cell function. This includes regulating the structure of the Golgi apparatus in preparation for and during mitosis in animal cells. It has been known for many years that protein transport through the secretory pathway of animal cells is regulated in a cell cycle-dependent fashion. A seminal study on the membrane glycoprotein of vesicular stomatitis virus found that its sugar processing was blocked during mitosis as a result of inhibition of exit from the endoplasmic reticulum (ER) (1). Later, it was found that the Golgi apparatus disassembles during mitosis and that Golgi proteins are redistributed to small vesicular and tubular structures dispersed throughout the cell (2). The first hints that this was due to direct regulation by mitotically active kinases came when it was shown that Cdc2 kinase was responsible for the inhibition of intra-Golgi vesicular transport during mitosis (3). One target of the Cdc2-Cyclin B mitotic kinase is GM130, a large coiled-coil protein of the cis-Golgi. GM130 is phosphorylated on a single serine at position 25, a modification that prevents the binding of a cytosolic factor known as p115 (4, 5). The p115-GM130 interaction is important for the recognition process by which vesicles coming from the ER identify the Golgi apparatus before fusion (6); its abolition is thought to explain the mitotic block in membrane traffic from ER to Golgi (4, 5). However, it is unlikely that the action of Cdc2 kinase on a single substrate protein can explain all the structural changes seen in the Golgi during mitosis.

A number of recent studies have shown that members of the polo-like kinase (Plk) and mitogen-activated protein (MAP) kinase families can phosphorylate structural proteins of the Golgi apparatus (7-9). Interestingly, the two substrates identified in these studies are both members of the Golgi reassembly stacking protein (GRASP) family of proteins identified in a screen for proteins required for the reassembly of the Golgi apparatus upon mitotic exit (10, 11). Lin and co-workers identified GRASP65 in a screen for substrates of mammalian Plk1 during mitosis (7). They also found that GRASP65 is a target for the Cdc2-Cyclin B mitotic kinase, which is noteworthy because the GRASP65 binding protein GM130 is a known Cdc2-Cyclin B substrate during mitosis (4, 7). These findings, together with the observation that a dominant negative form of Plk1 can block fragmentation by mitotic cytosol, suggest a pivotal role for Plk1 in the onset of Golgi breakdown (8). Plk1 is not sufficient for complete Golgi disassembly, implying that other mitotically active kinases are needed. The prime candidate is, of course, Cdc2-Cyclin B, which targets GRASP65 and GM130.

Research from a number of groups has suggested that mitogen-activated protein (MAP) kinase kinase 1 (MEK1) signaling may be important for an early stage of Golgi fragmentation that occurs during prophase and prometaphase of the cell division cycle (12, 13). Recently, it has been found that GRASP55 is a target for the MAP kinase ERK2, and may be responsible for generating an epitope recognized by the mitosis-specific phosphoprotein monoclonal antibody (MPM2) (9). An obvious conclusion to draw from these observations is that MEK1 mediates its effects on Golgi fragmentation through ERK2 and its substrate GRASP55. However, this would seem not to be the case, because MEK1 apparently acts through an unidentified Golgi-associated ERK, rather than ERK2 (12).

Cha and Shapiro (14) recently showed that an antibody specific for tyrosine-phosphorylated ERK stains the Golgi apparatus, as well as the nucleus, during late G2 and early prophase of the cell division cycle. They also found that elevating the level of tyrosine phosphorylation of ERK by overexpression of MEK1 caused a partial disruption of the Golgi. These effects did not seem to rely on the kinase activity of ERK but rather on its ability to be tyrosine-phosphorylated. This raises the obvious question of what recognizes tyrosine-phosphorylated ERK on the Golgi. How these observations relate to those on GRASP55 as an ERK2 substrate is unclear, but they may reflect the existence of more than one pathway for MAP kinase signaling to regulate Golgi structure and function.

In summary, it is likely that a Golgi breakdown is initiated at the end of the G2 phase by the action of Plk1 and a pathway involving ERK, possibly ERK2, targeting the GRASPs and potentially other substrates (Fig. 1). This might correlate to the Golgi-unstacking phase seen in vitro and in living cells, and would fit with the known role of GRASPs in the stacking of cisternae (10, 11, 15). The second phase of Golgi disassembly required for the block in protein transport is probably controlled by Cdc2-Cyclin B phosphorylation of targets such as GM130 (4, 5).

Fig. 1.

(A) Plk1 and Cdc2-Cyclin B activation occur in the very late stages of G2. At the G2/M phase transition, initial phosphorylation of Golgi proteins takes place and the Golgi apparatus starts to disassemble in early prophase (2,5). Later in prophase, additional residues of the respective substrates are likely to become phosphorylated and the Golgi breaks down into small clusters of vesicles and membrane tubules; by metaphase Golgi disassembly is complete, and some proteins may be recycled back to the ER (7, 24). By the end of anaphase or in early telophase, the Golgi apparatus reassembles in the daughter cells before cytokinesis. (B) The current state of knowledge on the regulation of Golgi structure and function by kinases. Red arrows indicate known or putative phosphorylation events of structural Golgi proteins or regulatory pathways. In mammalian cells, Cdc2 is negatively regulated by the Golgi and ER-localized Myt1 kinase, opposed by the Cdc25 phosphatase (25). Green arrows show known interactions among Golgi proteins, cytosolic factors, and vesicles. The effect of phosphorylation on the interactions required for vesicle fusion with the cis-Golgi is indicated by gray lines. The effect of Gβυ on vesicle scission from the trans-Golgi network is indicated by the blue line.

Kinases not only regulate Golgi structure and function during mitosis, they also control vesicular transport to and from the Golgi (Fig. 1). Docking of transport vesicles with the cis-Golgi is mediated by the tethering cytosolic factor p115 to its Golgi receptor GM130 (6). Phosphorylation of p115 in its COOH-terminal domain by a casein kinase II (CKII) or a CKII-like enzyme positively regulates the interaction of p115 with GM130 (16). This regulation by CKII is also needed for the reassembly of the Golgi apparatus from small vesicular fragments at the end of mitosis, where it seems to enhance the p115-promoted interaction of GM130 and another Golgi matrix protein, giantin (16). The scission of transport carriers from the trans-Golgi network is under the regulation of a member of the protein kinase C (PKC) family, PKCμ (also known as PKD) (17). Although the substrate of PKD is not known, its upstream activator is. The Malhotra lab showed that the βυ subunits of heterotrimeric G proteins regulate PKD through its PH domain, potentially by a direct interaction (18). This may explain previous observations that βυ subunits stimulate the formation of trans-Golgi network-derived vesicles in a cell-free assay (19). Whether intra-Golgi transport is regulated by protein kinases is uncertain, but it would seem likely, given that transport into and exit from the Golgi apparatus are controlled by CKII and PKD, respectively.

Despite the identification of a number of kinases and their substrates important for regulating Golgi structure and function, in only one case is it known what the phosphorylation actually regulates: Cdc2-Cyclin B and CKII directly control the interaction of p115 and GM130 (4, 16). Phosphorylation is also only half the story and, again, in only one case is it known what the phosphatase is; protein phosphatase 2A dephosphorylates GM130 late in telophase of the cell division cycle, concurrent with Golgi reassembly (5). Hence, there is still much progress to be made in these two areas. The order of kinase action during mitosis is also a matter for further investigation, because current models rely on a great deal of speculation or extrapolation. Although activation of MAP kinase has been observed in meiosis and during early embryonic divisions, this is not yet shown to be a general feature of the mammalian cell cycle (20, 21). In the case of MAP kinase signaling, the question of upstream activators is also completely open. Simply put, what is the signal for Golgi disassembly? The essential nature of MEK action for Golgi disassembly has also been questioned, but (as discussed above) there is a growing body of evidence to suggest that it is of some importance (22, 23). There are surely many fascinating aspects of the function and regulation of the Golgi apparatus that remain to be discovered, and all are likely to involve protein kinases.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
View Abstract

Stay Connected to Science Signaling

Navigate This Article