PKCeta is required for beta1gamma2/beta3gamma2- and PKD-mediated transport to the cell surface and the organization of the Golgi apparatus

Abstract

Protein kinase D (PKD) binds to a pool of diacylglycerol (DAG) in the TGN and undergoes a process of activation that involves heterotrimeric GTP-binding protein subunits betagamma to regulate membrane fission. This fission reaction is used to generate transport carriers at the TGN that are en route to the cell surface. We now report that PKD is activated specifically by G protein subunit beta1gamma2 and beta3gamma2 via the Golgi apparatus-associated PKCeta. Compromising the kinase activity of PKCeta-inhibited protein transport from TGN to the cell surface. Expression of constitutively activated PKCeta caused Golgi fragmentation, which was inhibited by a kinase inactive form of PKD. Our findings reveal that betagamma, PKCeta, and PKD act in series to generate transport carriers from the TGN and their overactivation results in complete vesiculation of the Golgi apparatus.

Figure 1.

β1γ2 and β3γ2 expression fragments the Golgi membranes in intact cells. (A) Various combinations of tagged βγ (FLAG-β and HA-γ) were coexpressed in HeLa cells and the organization of the Golgi apparatus monitored by fluorescence microscopy using antibodies to the early (GM130) and late (TGN46) Golgi cisternae. 200 cells were counted in four different experiments. The percentage of cells with fragmented Golgi apparatus is >60% (+++), with β1γ2 and β3γ2 between 20 and 60% (+) and other βγ combinations <20% (−). (B) HeLa cells were transfected with FLAG-β1 and HA-γ2 (a and b), or FLAG-β4 and HA-γ2 (c and d). The cells were visualized by fluorescence microscopy with anti-FLAG (a and c) and anti-TGN46 (b and d) antibodies, respectively. Expression of β1γ2 and β3γ2 fragments the Golgi apparatus.

Figure 2.

Effect of γ2 prenylation on β1γ2 mediated Golgi fragmentation in intact cells. HeLa cells were transfected with FLAG-β1 and HA-γ2C68S (a and b) or HA-γ2 (c and d). The sequence of the carboxy-terminal 20 amino acids of γ2 in which a specific cysteine is replaced with serine to generate γ2-C68S is shown. The site at which geranylgeranyl transferase-I adds the geranylgeranyl residue to the γ2 subunit is highlighted with an arrow. The cells expressing the wild-type and the mutant form of γ2 were probed by fluorescence microscopy with anti-HA (a and c) and anti-TGN46 (b and d) antibodies, respectively.

Figure 3.

Purified β1γ2 subunits cause Golgi fragmentation in semipermeabilized NRK cells. Purified recombinant His-tagged β1γ2 (a and b), β4γ2 (c and d), and β1γ2C68S (e and f) were added to permeabilized NRK cells. 30-min after incubation, the cells were fixed and visualized for immunofluorescence with anti-TGN38 (TGN marker) (a, c, and e), and anti-GM130 (cis-medial Golgi marker) (b, d, and f) antibodies.

Figure 4.

Effect of PH domains on the β1γ2-dependent Golgi fragmentation and PKD phosphorylation. (A) β1γ2-mediated Golgi fragmentation is inhibited by specific PH domains in intact cells. HeLa cells were transfected with FLAG-β1HA-γ2 and GFP-PH domains of proteins listed. The organization of the Golgi apparatus was monitored by fluorescence microscopy using anti-TGN46 and -GM130 (late and early Golgi specific markers) antibodies. 200 cells expressing FLAG-β1 were counted to determine the percentage of cells with fragmented Golgi membranes. Percentage of cells transfected and levels of protein expression for each PH domain was similar in all experiments as monitored by immunofluorescence and Western blotting with anti-GFP antibody, respectively. (B and C) Effect of β1γ2 on PKD phosphorylation in the activation loop. HeLa cells were cotransfected with the constructs listed. GST-PKD was immunoprecipitated and analyzed by Western blotting (B) with antibodies against GST,phospho-PKD (Ser916), phospho-PKD (Ser744–748), respectively, and quantitated by densitometric scan (C). Anti-FLAG antibodies were used to monitor the expression level of β1γ2 and β4γ2 in the respective cell extracts (B). Values are means (±SD, vertical bars) of three separate experiments.

Figure 5.

Activation of PKCη by β1γ2, and subsequent hyper phosphorylation of PKD in the activation loop. HeLa cells were cotransfected with the constructs listed. The cells were lysed and the extracts analyzed by Western blotting to monitor the phosphorylation status of FLAG-PKCη, GFP-PKCɛ, and GST-PKD, respectively. The blots were quantitated as described in Materials and methods. (A) For PKCη, the antibody used recognizes threonine 655 (T655). (B) Similar experimental procedure was used to monitor the effect of β1γ2 expression on the phosphorylation status of PKCɛ phosphorylation (Ser729). (C) Coexpression of β1γ2 and PKCη caused a fourfold increase in the phosphorylation of Ser744/748 (in the activation loop) of PKD without any appreciable change in the autophosphorylation of Ser916 (lane 2, shown in the Western blot and the bar graph). Values are means (±SD, vertical bars) of three separate experiments.

Figure 6.

PKCη is required for protein transport from TGN to the cell surface. HeLa cells were transfected with GFP-tsO45 VSV-G and FLAG-PKCη-wt or FLAG-PKCη-kinase dead. Following the procedure described in Materials and methods, immunofluorescence microscopy was used to detect GFP-tsO45 VSV-G localization. VSV–G protein is retained in the Golgi apparatus in PKCη kinase dead–expressing cells after 120 min incubation at the permissive temperature C (e–h), compared with cells expressing the wild-type PKCη (a–d).

Figure 7.

Expression of constitutively activated PKCη fragments the Golgi apparatus. HeLa cells were transfected with tagged versions of FLAG-PKCη wt (a and b), FLAG-PKCη constitutive active (c and d) and FLAG-PKCη constitutive active + GST-PKD kinase dead (e–h). The localization of the respective proteins was monitored by fluorescence microscopy using specific anti-tag antibodies. PKCη-wt is localized to the Golgi apparatus (a and b). The Golgi apparatus in cells expressing PKCη constitutive active is fragmented (c and d). The PKCη constitutive active–mediated Golgi fragmentation is inhibited upon expression of PKD kinase dead (e and f). Interestingly, PKD kinase dead and PKCη constitutive active colocalize both at the level of Golgi cisternae (g–h) and the emanating tubules (insets in g and h). CA denotes constitutively activated, and KD a kinase-dead form of the respective kinase.

Figure 8.

PKD on the Golgi membranes does not colocalize with AKAP-Lbc. HeLa cells were transfected with GFP-AKAP-Lbc and either wild-type (d, h, and l) or kinase-dead form of GST-PKD (a–c, e–g, and i–k). The location of AKAP-Lbc and PKD was monitored by confocal microscopy using GFP and anti-GST antibodies, respectively. PKD wild type and kinase dead is localized to the TGN (d) and the TGN-derived tubes, in the case of kinase-dead PKD (c and inset). The Golgi-associated form of PKD is in an activated state but it does not colocalize with AKAP-Lbc. Treatment with PdBu activates both the wild-type and the kinase-dead form of PKD. However, the PdBu-activated form of PKD is not sufficient to induce Golgi fragmentation. The PdBu-activated PKD translocates to the plasma membrane and partially colocalizes with AKAP-Lbc (g and h). Forskolin/IBMX treatment retains activated PKD on the plasma membrane, although it is not in complex with AKAP-Lbc (k and l).

Figure 9.

Generation of transport carriers from the TGN: a working hypothesis. (A) We propose that through a cargo-dependent involvement of G protein coupled receptor (GPCR), a trimeric G protein is activated at the TGN. The Gβγ subunits are involved in the production of DAG. The exact role of Gα in this process remains unclear. DAG activates PKCη and recruits PKD to the TGN. PKCη activates the TGN-bound PKD. The identity of Golgi-associated GPCR and the mechanism by which βγ results in the production of DAG are currently not known. Targets of PKCη, in addition to PKD, and the downstream targets of PKD are not known. It is possible that these kinases activate a diacylglycerol kinase that converts DAG into phosphatidic acid (PA) and a phospholipase, which convert PA into lyso-phosphatidic acid (LPA). LPA in our model would destabilize the neck of the budding transport carrier and promote fission. Green arrows denote activation; thin black arrows indicate involvement of additional components; and blue dotted line is suggestive of a hypothetical connection. (B) The scheme outlined above is used to regulate the formation of transport carriers from the TGN (a). Inactivation of PKCη or PKD does not affect cargo sorting and packaging, however membrane fission is inhibited, thus generating cargo filled transport carriers as large tubules attached to the TGN (b). Overactivation of PKCη and its downstream target PKD over activates the fission reaction, and the entire TGN is converted into small vesicles; The vesiculated TGN is an inactive acceptor for transport carriers from the preceding Golgi cisternae, and this combination generates a vesiculated Golgi apparatus (c).