C8. Mitogen Activated Protein Kinases - Biology

C8.  Mitogen Activated Protein Kinases - Biology

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

It is often the case that occupied receptors activate protein kinases, which activate other protein kinases, which activate yet other protein kinases to produce phospho-proteins which may act as transcription factors. An example is the mitogen activated protein kinase (MAPK system). A mitogen is an external chemical signal that causes mitosis or cell division. Activated of transcription factors by their phosphorylation through a mitogen activated kinase is required. The sequence of events is:

  • binding of external signal to membrane receptor and activation of receptor kinase
  • phosphorylation of receptor kinase and interaction with an activator GTP binding protein like ras
  • binding of activated G-protein to and activation of a mitogen activated protein kinase kinase kinase (MAPKKK)
  • MKKK phosphorylates and activates another kinase, MAPKK
  • MKK phosphorylates and activates mitogen activated protein kinase, MAPK
  • MAPK phosphorylates inactive transcription factors (or other proteins) and activates them. Unfortunately (from a naming point of view) when the activated proteins are themselves protein kinase, they are called mitogen activated protein kinase activated protein kinases (MAPKAPK)

There are seven types of MAPKs, four conventional and three atypical. Four typical ones are described in the table below.

Activator GTP binding proteinRas:GTP
MSK 1,2
An eventual
Protein Target

MAP Kinase System from Cell Signaling

MAP Kinase System animation from Promega


  • Prof. Henry Jakubowski (College of St. Benedict/St. John's University)

Protein C Jun

D Role of c-Jun in Axon–Schwann Cell Interactions

The transcription factor c-Jun, which belongs to the bZIP family of transcription factors ( 199 ), is exclusively expressed by nonmyelinating Schwann cells in normal nerve and previously myelinating Schwann cells after injury ( 200 ). During PNS development and regeneration, the expression pattern of c-Jun is consistent with its regulation by axon–Schwann cell interactions. Thus, c-Jun is likely to be involved in determining the differentiated state of nonmyelinating and axon-deprived Schwann cells, although the target genes for this transcription factor in such cells remain to be determined.


Mitogen-activated protein kinases (MAPKs) are ubiquitously expressed and regulate a wide variety of functions in virtually all cell types. In vertebrates five families of MAPKs encoded by 11 different genes include extracellular signal-regulated kinase ERK1/2, c-Jun N-terminal kinase JNK1/2/3, p38α/β/δ/γ, ERK5 and ERK7 (Uhlik et al., 2004). Splice variants exist for several of the MAPK proteins. MAPKs phosphorylate specific substrate proteins that regulate cell proliferation, survival, motility, metabolism, transcription and translation. MAPKs contribute to the cellular response to diverse stimuli, including growth factors, cytokines, and stresses such as toxins, numerous drugs, changes in cell adherence, osmolarity, oxygen radicals, ultraviolet light and temperature (Pearson et al., 2001). It is therefore not unexpected that dysregulated MAPK activity is associated with a variety of pathological states, including those arising from inflammation, such as arthritis and inflammatory bowel disease (Johnson and Lapadat, 2002 Hollenbach et al., 2004), as well as syndromes that include the uncontrolled cellular proliferation and tissue remodeling characteristic of cancer (Gollob et al., 2006).

An understanding of how any of the arrays of stimuli that activate MAPKs can induce a specific outcome has been the goal of an intensive research effort. MAPKs are the terminal kinase in a three kinase phosphorelay module, in which MAPKs are phosphorylated and activated by mitogen-activated protein kinase kinase MKKs, which themselves are phosphorylated and activated by mitogen-activated protein kinase kinase kinase (MKKKs). Of the MAPK signaling module components that have been identified in human cells, MKKKs are the largest in number (at least 20 genes), as compared to the MKKs (seven genes) and MAPKs (11 genes) (Uhlik et al., 2004). The numbers of each MKKK, MKK and MAPK are actually larger when splice variants are counted. Interestingly, some MKKKs can activate different MAPK modules, or activate the same module with altered duration or localization to render distinct outcomes (Johnson et al., 2005). Given the variety of stimuli known to activate the different MAPKs, the diversity of MKKKs represents a predominant mechanism by which specificity is achieved in MAPK activation. Restated, MKKKs can route signals originating from widely differing inputs to MAPK signaling modules to provide specific functional responses. In this sense, MKKKs represent organizational nodes or ‘hubs’ that integrate responses to provide specificity in MAPK activation.

Figure 1 illustrates a wiring diagram for four of the 20 MKKKs. MEKKs have significant homology in their kinase domains (see below) and differentially regulate more than one MAPK (Fanger et al., 1997 Yujiri et al., 1998 Abell et al., 2005). The connections in the wiring diagram represent protein interactions based on published data in PubMed. In addition, targeted gene knockouts in mice for MEKK1-4 have been characterized (Yujiri et al., 1998 Garrington et al., 2000 Yang et al., 2000 Abell et al., 2005). The knockout phenotypes for MEKK1, 2, 3 and 4 clearly define specific modules within the connections map that have defined essential physiological roles for each MEKK. For such a connections map a module can be defined as a discrete unit of function that is separable from other components of the system. An example is the MEKK3-OSM-KRIT module that regulates cerebral vascular integrity. Mutation of OSM or KRIT results in the disease cerebral cavernous malformations (CCM) that affect one in 200 people in the United States (Sahoo et al., 1999 Zawistowski et al., 2005). One can imagine the complexity of such a connections map for all 20 MKKKs. In this review, we examine the mechanisms by which MKKKs integrate signals from diverse origins to elicit specific functional outcomes from MAPK activation.

Connections map showing protein interactions for MEKK1, 2, 3 and 4. The connections map illustrates MEKKs as functional signaling hubs for integrating multiple upstream inputs into the control of specific MAPKs. The maps were created from PubMed data base searches and filtering of data based on the phenotypes of mice having targeted gene disruption or kinase-inactivating knockins of each MEKK. Color codes represent different classes of proteins or pathways that feed into the four MEKKs. Green represents scaffold proteins and receptors that utilize the scaffolds in stimulating a response. Pink represents tyrosine kinases. Light blue represents serine–threonine kinases. Red represents receptors and regulators of receptor response. Yellow represents GTPases.



RT-PCR primers were obtained from TIB Molbiol (Berlin, Germany). Secondary antibodies were from Li-Cor Biosciences (Lincoln, NE, USA). Skepinone-L was synthesized as previously described ( 14 ). Materials used: CAY10566, Cayman (Ann Arbor, MI, USA) FlexiTube GeneSolution siRNAs directed against SCD-1 (Qiagen, Hilden, Germany) non-immune goat serum, Invitrogen (Carlsbad, CA, USA) DMEM/high glucose (4.5 g/L) medium, trypsin/EDTA solution (PAA Laboratories, Coelbe, Germany) mouse anti-cyclin D1 (1:2000), rabbit anti-cyclin B1 (1:1000), mouse anti-cyclin E (1:1000), rabbit or mouse anti-β-actin (1:1000), mouse anti-phospho-ERK1/2 (Thr202/Tyr204 1:2000), rabbit anti-phospho-MEK1/2 (Ser217/221 1:1000) rabbit anti-phospho-myristoylated alanine-rich C-kinase substrates (Ser152/156 1:1000), rabbit anti-caspase 3 (1:1000), rabbit anti-phospho-Src family (Tyr416 1:1000) mouse anti-phospho-JNK (Thr183/Tyr185 1:2000), rabbit anti-phospho-Akt (Ser473 1:1000), rabbit anti-phospho-p38 MAPK (Thr180/Tyr182 1:1000), rabbit anti-p38 MAPK (1:1000), mouse anti-IκBα (1:1000), mouse anti-CHOP (1:1000), rabbit anti-ATF-4 (1:1000), rabbit anti-BiP (1:1000: Cell Signaling, Danvers, MA, USA) mouse anti-endoplasmic reticulum/Golgi intermediate compartment (ERGIC) 53, Enzo Life Sciences (Lörrach, Germany) lipid standards (Avanti Polar Lipids, Alabaster, AL, USA) 1, 2- 3 H-2-deoxy-D-glucose (Hartmann Analytics, Braunschweig, Germany) solvents and all other chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA) or Wako Pure Chemicals (Osaka, Japan) unless stated otherwise.

Cells, cell differentiation, and cell cycle synchronization

Mouse NIH-3T3 and 3T3-L1 fibroblasts were grown at 37°C and 5% CO2 in DMEM supplemented with 10% (v/v) heat-inactivated FCS. To synchronize NIH-3T3 cells in G2/M-phase of the cell cycle, cells (70-80% confluent) were treated with nocodazole (0.4 μg/ml) in DMEM supplemented with 10% fetal calf serum (FCS) for 20 hours (37°C, 5% CO2) as described elsewhere ( 15 ). Round mitotic cells were detached by rocking and squirting, washed 3 times with ice-cold PBS pH 7.4 and reseeded at 4 × 10 6 cells/75 cm 2 flask.

3T3-L1 fibroblasts were differentiated into adipocytes as described elsewhere ( 16 ). Postconfluent cells (12-well plate) were treated with DMEM containing 10% FCS, 1 μg/ml insulin, 1 μM dexamethasone, and 0.5 mM 3-isobutyl-1-methylxanthine for 2 days (37°C, 5% CO2) and then cultivated for another 2 days in DMEM plus 10% FCS and 1 μg/ml insulin.

Determination of cell number and cell viability

Total and viable cells were counted after trypan blue staining using a Vi-Cell Series Cell Counter (Beckman Coulter, Krefeld, Germany). Alternatively, cell viability was measured using the colorimetric thiazolyl blue tetrazolium bromide (MTT) dye reduction assay. NIH-3T3 cells (7 × 10 3 /96-well plate) were cultivated for 16 h (37°C, 5% CO2) and then preincubated with vehicle (DMSO) or CAY10566 for 48 hours. MTT was added, the formazan product solubilized with SDS (10%, m/v, in 20 mM HCl) after 3 hours, and the absorbance read at 595 nm as described elsewhere ( 17 ).

Inhibition of SCD-1 by RNA interference

NIH-3T3 cells were grown to approximately 60% confluence in 25 cm 2 dishes and transfected with siRNA duplex oligonucleotides (15 nM) using Lipofectamine RNAiMax transfection reagent (10 μl, Invitrogen) according to the manufacturer's protocol. The 3 SCD-1 siRNAs targeted the sequences 1) 5'-CACAACAGCTTTAAATAATAA-3', 2) 5'-TAGTGAGATTTGAATAATTAA-3', and 3) 5'-CCGGTACAGTATTCTTATAAA-3', respectively. On-TargetPlus nontargeting siRNA #1 (Thermo Scientific, Waltham, MA, USA) was used as scrambled control siRNA.

Extraction of lipids

Lipids were extracted from NIH-3T3 cells (5 × 10 5 in PBS pH 7.4) by successive addition of methanol, chloroform, and saline (final ratio: 14:34:35:17) as previously described ( 18 ). The organic layer was evaporated, and the extracted lipids were dissolved in 100 μl methanol and diluted, and 1,2-dimyristoyl-sn-glycero-3-phosphatidylcholine and 1, 2-dimyristoyl-sn-glycero-3-phosphatidylethanolamine were used as internal standards.

Reversed phase liquid chromatography

Lipids were separated on an Acquity UPLC BEH C8 column (1.7 μm, 1 × 100 mm, Waters, Milford, MA, USA) using an Acquity Ultraperformance LC system (Waters) as previously described ( 15 ). For phospholipid analysis, chromatography was performed at a flow rate of 0.75 ml/min at 45°C using a gradient of 30% mobile phase A (acetonitrile/water, 10/90, 10 mM ammonium acetate)/70% mobile phase B (acetonitrile/water, 95/5, 10 mM ammonium acetate) to 20% mobile phase A/80% mobile phase B within 5 minutes and to 100% mobile phase B within 2 minutes followed by isocratic elution for another 2 minutes. Triacylglycerols and cholesterol ester were separated at a flow rate of 0.75 ml/min at 45°C using a gradient of 100% mobile phase A (acetonitrile/water, 95:5, 10 mM ammonium acetate) to 70% mobile phase A/20% mobile phase B (isopropanol) within 6 minutes and subsequent isocratic elution for 3 minutes.

Mass spectrometry

The chromatography system was coupled to a QTRAP 5500 mass spectrometer (AB Sciex, Darmstadt, Germany) or a Vantage triple stage quadrupole mass spectrometer (Thermo Scientific for cell cycle analysis). Both were equipped with electrospray ionization sources. Quantification of glycerophospholipids using the QTRAP 5500 mass spectrometer was based on the detection of both fatty acid anion fragments through multiple reaction monitoring according to ( 15 ). The most intensive transition was selected for quantification. The ion spray voltage was set to 4500 V in the negative ion mode the heated capillary temperature to 350-700°C the sheath gas pressure to 45-55 psi the auxiliary gas pressure to 75-80 psi the declustering potential to 40-50 V and the collision energy to 46, 38, 20, 62, and 52 V for phosphatidylcholines, ethanolamines, serines, PI, and phosphatidylglycerols, respectively. Sphingomyelins were quantified as [M + H + ] by m/z =184 precursor ion scans (collision energy: 33 V).

Using the Vantage triple stage quadrupole mass spectrometer, cell cycle-dependent changes in the lipid composition were monitored by m/z =184 precursor ion scans in the positive ion mode (collision energy: 35 V phosphatidylcholines and sphingomyelins) or full scans in the negative ion mode (all other phospholipids) as described elsewhere ( 19 ). Phosphatidylethanolamine and -serine headgroups were confirmed by m = 141.0 or m = 87.0 neutral loss scans (collision energy: 25 V, positive and negative ion mode, respectively) and phosphatidylcholine and PI headgroups by m/z = 184 or m/z = 241.0 precursor ion scans (collision energy: 35 V, positive and negative ion mode, respectively) . The fatty acid composition was determined at a collision energy of 40 V by product ion scans. The higher signal intensity of sn-1 than sn-2 fatty acid anions was utilized to estimate the isomeric position of the fatty acids ( 20 ).

Triacylglycerols and cholesterol ester were analyzed as [M + NH4] + adducts by multiple reaction monitoring using a QTRAP 5500 mass spectrometer. Transitions to [M-fatty acid anion] + fragments were measured for lipid identification. The most intensive (and species-specific) transition was selected for quantification. The isomeric positions of the fatty acids in triacylglycerols were not determined. In variation to the settings described and referenced above, the ion spray voltage was set to 5500 V in the positive ion mode, the heated capillary temperature to 350-400°C, the sheath gas pressure to 55-60 psi, the auxiliary gas pressure to 70 psi, the declustering potential to 55-120 V, and the collision energy to 35 V (triacylglycerols) or 22 V (cholesterol ester).

Mass spectra were processed using the Analyst 1.6 (AB Sciex) or Xcalibur 2.0 software (Thermo Scientific) as described elsewhere ( 15 , 20 ). The proportion of lipid species (=relative intensity) is given as percentage of the sum of all species in the respective subclass (=100%). Total phospholipid subclass intensities combine the intensities of all species of the respective subclass and were normalized to the number of cells and the internal standard 1, 2-dimyristoyl-sn-glycero-3-phosphatidylcholine.

Sample preparation, SDS-PAGE, and Western blot

Cells were sonified (2 × 5 s, on ice) in 20 mM Tris-HCl pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% (v/v) Triton X-100, 1 mM phenylmethanesulphonyl fluoride, 60 μg/ml soybean trypsin inhibitor, 10 μg/ml leupeptin, 5 mM sodium fluoride, 1 mM sodium vanadate, and 2.5 mM sodium pyrophosphate. After centrifugation (12,000 g, 5 min, 4°C), the supernatant was taken up in 1× SDS/PAGE sample loading buffer [125 mM Tris-HCl pH 6.5, 25% (m/v) sucrose, 5% SDS (m/v), 0.25% (m/v) bromophenol blue, and 5% (v/v) β-mercaptoethanol] and boiled for 5 minutes at 95°C. Aliquots (10 μg protein) were resolved by 10 or 12% (m/v) SDS-PAGE and transferred to a Hybond ECL nitrocellulose membrane (GE Healthcare, Munich, Germany). Membranes were blocked with 5% (m/v) BSA or skim milk for 1 hour at room temperature and incubated with primary antibodies overnight at 4°C. Immunoreactive bands were stained with IRDye 800CW-labeled (1:10,000, each) and/or IRDye 680LT-labeled anti-rabbit or anti-mouse IgG (1:80,000, each) and visualized by an Odyssey infrared imager (Li-Cor Biosciences). Data from densitometric analysis were background-corrected.

Fatty acid supplementation

The culture medium of NIH-3T3 cells was replaced by DMEM containing 10% (v/v) heat-inactivated FCS plus palmitate, palmitoleate, or oleate at the indicated concentrations. Culture medium supplemented with 400 μM palmitate was sonicated at 40°C for 20 minutes prior use. Cells were incubated at 37°C and 5% CO2 for the indicated times.

Cellular uptake of 2-deoxy-D-glucose

Cellular uptake of 2-deoxy-D-glucose was determined according to Yand and Yao ( 16 ). Differentiated 3T3-L1 adipocytes (25 cm 2 flask) were treated with CAY10566 (3 μM), Skepinone-L (1 μM), or vehicle (DMSO) for 48 hours (37°C, 5% CO2). Cells were washed, supplemented with vehicle (DMSO), Skepinone-L (1 μM), or the control inhibitor cytochalasin B, and stimulated with insulin (100 nM) in 10 mM HEPES/10 mM sodium phosphate buffer (pH 7.4) containing 128 mM sodium chloride, 4.7 mM potassium chloride, 1.25 mM calcium chloride, and 1.25 mM magnesium sulfate for 15 minutes. Then, 3 H-labeled 2-deoxy-D-glucose (0.5 μCi/ml at 0.5 mM) was added. After 10 minutes, cells were washed 3 times with ice-cold PBS pH 7.4 plus 10 mM glucose, immediately lysed in 0.1 N NaOH and mixed with Rotiszint eco plus (3 ml, Carl Roth GmbH, Karlsruhe, Germany) for liquid scintillation counting using a Packard TRI-CARB 2100TR Liquid Scintillation Analyzer.

Immunofluorescence staining and microscopy

NIH-3T3 cells (2 × 10 4 /cm 2 ) were seeded onto coverslips and cultured in presence of CAY10566 (3 μM), Skepinone-L (1 μM), or vehicle (DMSO 37°C, 5% CO2). The culture medium was replaced after 42 hours against culture medium containing palmitate (400 μM) plus CAY10566 (3 μM), Skepinone-L (1 μM), or vehicle (DMSO). This step was omitted for experiments using nonstressed cells. After 6 hours, samples were fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, and blocked with 5% normal goat serum (10 minutes, room temperature). Samples were incubated with mouse anti-ERGIC53 antibody (1:100 Enzo Life Sciences) for 1 hour at room temperature followed by goat anti-mouse IgG Alexa Fluor 555 (1:1000 1 hour, room temperature Invitrogen). DNA was stained with 0.1 μg/ml DAPI for 3 minutes at room temperature. The coverslips were mounted on glass slides with Mowiol containing 2.5% n-propyl gallate (Sigma-Aldrich). The fluorescence was visualized with an Axio Observer.Z1 microscope and a Plan-Apochromat 40X/1.3 Oil DIC M27 objective (both by Carl Zeiss GmbH, Jena Germany). Images were taken at room temperature with an AxioCam MR3 camera and were acquired, cut, linearly adjusted in the overall brightness and contrast, and exported to TIF by the AxioVision 4.8 software (Carl Zeiss GmbH).

Gene Sense primer (5′ → 3′) Anti-sense primer (5′ → 3′)

Quantitative RT-PCR

Total RNA was prepared using E.Z.N.A. Total RNA Kit I (Omega Bio-tek, Norcross, GA, USA). First-strand cDNAs were synthesized from 1 μg RNA using Superscript III (Invitrogen). PCR was conducted in Mx3000P 96-well plates (25 μl) using a Mx3005P qPCR system (Agilent Technologies, Santa Clara, CA, USA). The PCR mix contained cDNA (2.5 μl), Maxima SYBR Green/Rox qPCR Master Mix (1×, Thermo Scientific) and forward and reverse primers (0.5 μM, each). Primer information is provided in Table 1. The cycling program started with 10 minutes at 95°C and included 40 cycles of 15 seconds at 95°C, 30 seconds at 63°C, and 30 seconds at 72°C. cDNA levels were quantified using the MxPro QPCR Software. mRNA expression was normalized to glyceraldehyde-3-phosphate dehydrogenase for knockdown experiments or to the amount of RNA for cell cycle studies. Results are given in arbitrary units.


Data are presented as mean ± SE of n observations. Statistical evaluation of the data was performed by 1-way ANOVAs for independent or correlated samples followed by Tukey honest significant difference (HSD) post hoc tests or by Student's t test for paired and correlated samples. P values <0.05 were considered statistically significant. All statistical calculations were performed using GraphPad InStat 3.10 (GraphPad Software Inc., La Jolla, CA, USA).


Identified phosphopeptides

(a) The aim of this study was to establish, as a routine path, a method for identification and characterization of individual phosphorylated kinases p38 and HuR in vitro using: TiO2, IMAC, SIMAC coupled to MSA and MS3NL on the LTQ ion Trap mass spectrometer (Thermo). Purification and fusion proteins were expressed in Escherichia coli. The kinase assay was carried out incubating with different types of protein phosphatase inhibitors in order to increase the levels of protein-kinases phosphorylation prior to the analysis. In fact, sodium pervanadate, a tyrosine phosphatase inhibitor, coupled to a combination of two phosphatase inhibitor cocktails from Sigma (one cocktail containing serine/threonine phosphatase inhibitors and one containing tyrosine phosphatase inhibitors) was also used. Protein kinases were digested with lysyl endopeptidase and trypsin and subsequently enriched for phosphorylated peptides using TiO2, IMAC and SIMAC phosphoenrichments. The isolated phosphopeptides were desalted, cleaned and analyzed by nano-LC ESI-MS/MS using a Thermo LTQ ion Trap MSMS instrument. All experiments were performed in triplicate. The LC-MS/MS experiments and Mascot database searching resulted in overall significant peptide hits. All the peptides were determined with a mass error of less than 5.5 ppm. A total of 6 phosphopeptides were validated by manual evaluation of the LC-MS/MS data sets obtained from the four triplicated experiments. Of these, 6 were assigned to unique amino acid phosphorylated sequences resulting in the identification of 3 unique proteins across all experiments.

(b) The analysis of the 5 μl (

3 μg) of the sample purified by IMAC and desalted and cleaned by R3/C18 permitted us to obtain 2 unknown phosphorylated peptides when using MSA on the nano-LC-LTQ ion Trap instrument. Both phosphorylated peptides were manually validated and correspond to: R.VLVDQTTGLSR.G and R.SLFSSIGEVESAK.L. Those two phosphopeptides belong to the HuR RNA binding protein gi/1022961 protein. The analysis of 5 μl (

3 μg) of the sample purified by TiO2 and desalted and cleaned by R3/C18 permitted us to obtain 4 unknown phosphorylated peptides when using MSA on the nano-LC-LTQ instrument.

The four phosphorylated peptides were manually validated and correspond to: R.VLVDQTTGLSR.G, R.SLFSSIGEVESAK.L, K.DVEDMFSR.F which belong to the HuR RNA binding protein gi/1022961 and another phosphopeptide: K.DLSSIFR.G which belongs to p38 MAP Kinase gi/1469306 (Table 1).

3 μg) of the sample purified by SIMAC and desalted and cleaned by R3/C18 allowed us to obtain 6 unknown phosphorylated peptides when using MSA on the nano-LC-LTQ ion Trap instrument. The six phosphorylated peptides were manually validated and correspond to: K.DVEDMFSR.F, R.VLVDQTTGLSR.G, K.DANLYISGLPR.T, R.SLFSSIGEVESAK.L which belong to HuR RNA binding protein gi/1022961 and R.TAVINAASGR.Q which belongs to Chain B, Structure Of Appbp1-Uba3-nedd8-Mgatp-Ubc12 (c111a), A Trapped Ubiquitin-Like Protein Activation Complex gi/126031226, and K.DLSSIFR.G which belongs to p38 MAP Kinase gi/1469306 (Table 1).

Therefore, when using MSA by the LTQ Ion Trap instrument, SIMAC (6 phosphopeptides purified, identified and validated) efficiency is higher than TiO2 (4 phosphopeptides purified and identified) and IMAC (2 phosphopeptides purified, identified and validated) for these protein-kinases studied. It has been described that IMAC easily enriches multiple phosphorylated peptides while TiO2 mono-phosphorylated ones. In fact, SIMAC has been optimized to get the best efficiency from IMAC and TiO2 and complement both in just one method (see reference previously mentioned [34]). This supports our data.

In any case, we recommend that in order to study kinase phosphorylated protein kinases, combine the three resins (or ever more phosphoenrichments methods) in order to purify as many as possible phosphopeptides [46]. The reason for this is that each sample needs to be optimized and tested with different complementary strategies. The analysis of the 5 μl (

3 μg) of the sample purified by SIMAC and desalted and cleaned by R3/C18 allowed us to get 5 unknown phosphorylated peptides when using Data Dependent Neutral Loss MS3 (DDNLMS3) on the nano-LC-LTQ ion Trap instrument. The five phosphorylated peptides were manually validated and correspond to: K.DVEDMFSR.F, K.DANLYISGLPR.T, R.SLFSSIGEVESAK.L which belong to HuR RNA binding protein gi/1022961 K.DLSSIFR.G which belongs to p38 MAP Kinase gi/1469306 and R.TAVINAASGR.Q which belongs to Chain B, Structure Of Appbp1-Uba3-nedd8-Mgatp-Ubc12 (c111a), A Trapped Ubiquitin-Like Protein Activation Complex gi/126031226 (Table 1).

(c) SIMAC coupled to MAS and MS3-NL mass spectrometry analysis. The preferred approach for analyzing samples using mass spectrometry is to produce structurally significant product ions using the process of ion dissociation. A method commonly known as Data Dependent Neutral Loss MS3 (DDNLMS3) (developed by Coon and co-workers [47]) analysis enables selective fragmentation by isolating a neutral loss ion fragment from an MS/MS experiment and then subjecting it to further dissociation [48]. Despite of this, DDNLMS3 did not allow us to get as efficient results as when using MSA for our protein-kinases analyses. It is well known that the production of neutral loss ions in MS/MS, is almost always accompanied by partial fragmentation of the precursor ion and these diagnostic fragment ions are subsequently lost when the neutral loss ions are isolated for MS3. Multistage activation (or pseudo MS3) allowed us to get spectra that were the combination of MS/MS and MS3 fragmentation and thus retaining the informative fragments from the precursor ion more efficiently. This is due to the fact that MSA produced more structurally informative ions by eliminating the ion isolation step between MS/MS and MS3 for the study of phosphorylated protein kinases p38 and HuR in vitro. We observed that - in this research study related to the previously phosphorylated proteins after in vitro kinase reaction- multistage activation was a faster route to a more information- rich spectra since the ion-trap does not require refilling for the MS3 scan, as with the traditional neutral loss experiment (DDNLMS3). We concluded during the first tests-analyses of the protein kinases in vitro, that when compared to DDNLMS3, multistage activation generated spectra with increased signal intensity and a greater number of structurally diagnostic ions for phosphorylated peptides. Thus we chose MSA as a routine path for this kind of analysis (p38 and HuR phosphorylated kinases in vitro). Further benefits of using multistage activation are demonstrated in other studies of phosphopeptides, including large scale analysis [49]. The information-rich spectra generated using multistage activation were particularly important for these compounds because there is often a significant loss of sequence informative fragment ions generated in MS/MS. For this study, more ions were identified with multistage activation than with MS/MS or MS3 in the DDNLMS3 method. In addition, the signal intensities were generally higher with multistage activation compared to MS/MS or MS3 of DDNLMS3 method. In fact, multistage activation resulted in more information for the suite of phosphopeptides studied (Table 1) (see an example of the spectrum of an identified phosphorylated peptide when using SIMAC coupled to MSA in the LTQ ion Trap mass spectrometer and Mascot, Figure 2).

Phospho-site assignment & manual validation of the phosphorylated peptide VLVDQ TTph GLSR obtained by Mascot analysis. The monoisotopic mass of neutral peptide Mr (calc) resulted was 1267.6173. Fixed modifications chosen were: Carbamidomethyl (C), while for variable modifications T7: Phospho (ST), with neutral losses 97.9769 (shown in table) was selected. The y5 ion and b7 are those which allowed identification of the treonine (5) as phosphorylated (ph) amino acid (T in red colour). In addition the phosphate fingerprint of the neutral loss (NL) from the parent ion is also a positive signal of phosphorylation. Six b ions and 8 y ions were continuously matched respectively.

Nevertheless, it must be pointed out that Jiang and co-workers developed a specific classification filtering strategy for their studies (using different samples) which significantly improved the coverage of the phosphoproteome analysis when using NLMS3 (see reference previously mentioned [48]). In fact, Jiang and co-workers obtained a higher coverage of the phosphopeptide identifications when processing and filtering specific methods which they developed for the spectra from NLMS3, compared with MS2 and MSA strategies. In relation to this, we should say that just one more phosphopeptide was identified and validated when we used SIMAC coupled to MSA (new 6 identified phoshopeptides) compared to when we coupled SIMAC to DDNLMS3 (5 new identified phosphopeptides). In addition, those 5 new phosphorylated peptides identified and their phospho-site assignments in each specific amino acid are the same ones following both strategies (see Figure 3 and Table 1). Moreover, the 6 new phosphopeptides and phospho-site assignments showed high reproducibility in all cases during the four triplicate experiments we carried out.

The efficiency and reproducibility of the phosphopeptide purification and identification when using

3 μg of protein kinases per each resin and/or phosphoenrichment method (SIMAC, TiO 2and IMAC) coupled to R3/C18 and MSA-LTQ ion Trap mass spectrometer is illustrated. [A] Four triplicate experiments were carried out in order to identify the phosphopeptides. The phospho-site identifications were carried out from pooled and non-pooled assays (inter- and intra-assays) confirming a high reproducibility. The 6 phosphorylated peptides identified were isolated and validated in the four triplicate analyses, not only by Mascot (at least 4 continuously -y and -b ions matched)but also by manual inspection of all the spectra. SIMAC allowed the purification of 3 phosphorylated proteins: HuR RNA binding, p38 MAP Kinase and Trapped Ubiquitin-Like Protein Activation Complex, and 6 phosphorylated peptides related to those previously mentioned proteins. TiO2 and IMAC allowed the isolation of 2 phoshorylated proteins: HuR RNA binding and p38 MAP Kinase, and 1 phosphopeptide related to the protein kinase HuR RNA binding. [B] SIMAC coupled to MSA allowed the identification of one more phosphopeptide compared to SIMAC coupled to DDNLMS3. Nevertheless, both strategies (SIMAC coupled to MSA and SIMAC coupled to DDNLMS3) allowed the identification of the same number of phosphorylated proteins (3). [C] and [D] Three phosphorylated proteins and six phosphopeptides were identified when using SIMAC coupled to MSA. From those three phosphoproteins identified, six phosphopeptides were identified: (a) TiO2 coupled to MSA allowed the identification of two equal/same phosphorylated proteins and four equal/same phosphopeptides as SIMAC and (b) IMAC allowed the identification of one equal/same protein and two equal/same phosphopeptides. Thus, SIMAC is more efficient than the other tested resins for this study, while TiO2 and IMAC corroborate the reproducibility of the phosphorylated proteins and phosphopeptides identified.

All our MS analyses were carried out by CID. We hypothesize that combining CID with ETD or ECD fragmentation, it is probable that more and/or complementary data would be obtained according to the methodological study of Navajas and co-workers [50]. ECD occurs only on the peptide backbone - which is an advantage -, and labile phosphate groups are left intact on the resulting c- and z- fragment ions, thus, complementary identification of other specific phosphorylation sites would be enabled [51, 52]. As a result, we recommend using CID to start with, and would recommend switching to ETD, in the event you were not able to determine the phosphorylation site, if you have the possibility of the required instrument [53–57]. The phosphopeptides purified, identified and validated, including also the site-assignments of the phosphate group are illustrated in Table 1.

The efficiency and reproducibility of the phosphopeptide purification and identification when using

3 μg of protein kinases per each resin and or phosphoenrichment method (SIMAC, TiO2 and IMAC) coupled to R3/C18 and MSA-LTQ ion Trap mass spectrometer is illustrated in Figure 3.

An example of a phospho-site assignment and manual validation of the phosphorylated peptide (VLVDQTTphGLSR) obtained by Mascot analysis is illustrated in Figure 2.

Bioinformatic modelling and molecular dynamics simulations

To study the potential functional effect of serine phosphorylation in the above indicated sequence locations, 3D structural models for the phosphorylated state of both MAP kinase p38beta (p38B) and HuR were generated using bioinformatics procedures. As shown in figure 4, phosphorylated Ser-279 of p38B is located in a loop placed on the external surface of the protein structure, far away from the active site of the kinase. It is conceivable that the phosphorylation of this residue does not affect p38B structure stability or folding, but external contacts to accompanying proteins, modulate the nature of the putative interaction (see reference previously mentioned [39]).

Location of phosphorylated Ser-279 in the protein structure of human MAP kinase p38beta (p38B). A model for phosphorylated serine was located in the structural position of residue Ser-279 in the 3D crystallographic coordinates of p38B (Protein Data Bank code: 3GC8). Position of the ATP binding site is indicated. Plot was generated using PyMOL (DeLano Scientific, San Carlos, CA).

In the case of HuR, only one of the four phosphorylated residues found (Ser-48) fall into a structure domain of the protein previously crystallized: the dimmerized first RNA recognition motif (see reference previously mentioned [40]). Thus, only the putative structural effect of the phosphorylated and non-phosphorylated state of Ser-48 could be analyzed through structural bioinformatic tools including Molecular Dynamics (MD) simulation.

As shown in figure 5A, Ser-48 is located in the dimmerization surface, surrounded by residues Glu-47 and Lys-50 that form a pair of saline bonds potentially implicated in the stabilization of the dimmer. To test the effect of the presence of a phosphorylated Ser in the maintenance of the quaternary structure, two unrestricted MD computer simulations were performed in presence or absence of a phosphorylated Ser in position 48, as indicated under "Materials and Methods". Results obtained after 10ns of MD showed that, in the case of the unphosphorylated dimmer structure (non-phosphorylated Ser-48), both monomers experimented a significant displacement from their initial relative positions, resulting in a complete disorganization of the quaternary structure (Figure 5B -left-). Measurement of root mean square deviation (RMSD) values of the monomers and the dimmer, as well as the distances between the C atoms of the contact residues (Glu47A-Lys50B, Ser48A-Ser48B and Lys50A-Glu48B) indicated a clear and irreversible displacement from the initial values during the first steps of the MD simulation (Figure 5C, upper plots). In contrast, when the same MD simulation was performed in presence of a phosphorylated Ser in position 48 of both monomers, a clear stabilization of the dimmerized structure was obtained, showing no displacement from their initial position (Figure 5B -right-) and exhibiting constant values of RMSD values and invariable RMSD values and distances between contact residues (Figure 5C, lower plots).

Effect of the phosphorylation of Ser-48 in the stability of the dimmer structure of the first RNA recognition motif of HuR. [A] Crystal structure of HuR dimmer (Protein Data Bank code: 3HI9, chains B and D) indicating the position of Ser-48, Glu-47 and Lys-50 in the dimmerization surface. [B] Relative spatial position of the two HuR monomers after 10ns of unrestricted Molecular Dynamics (MD) simulation of both non-phosphorylated (red structure, left) and phosphorylated (green structure, right) states of Ser-48. Position of initial dimmer structure, prior to the MD simulation, is included for comparison (in gray). Note the large displacement of the non-phosphorylated state in contrast to the stability exhibited by the dimmer in presence of phosphoSer48. Respective positions of Ser-48 and phosphoSer-48 are indicated. [C] Left: RMSD values measured for HuR dimmer (red) monomer A (blue) and monomer B (green) during the 10 ns trajectory of the unrestricted MD simulation of the dimmer in presence of non-phosphorylated Ser-48 (HUR plot, top panel) or phosphorylated pSer-48 (HURP plot, lower panel). Right: continuous measurement of Cα Cα distances between residues E47(A)-K50(B) (red), S48(A)-S48(B) (blue) and E47(A)-K50(B) (green), during the MD trajectory. Distortion of the HuR dimmer in the presence of Ser-48 (HUR plot, top panel) when compared to the phosphorylated state of the protein (HURP plot, lower panel) is patent in both RMSD and Cα-Cα measurements. Structure plots were generated as in Figure 4.

These results suggest that the phosphorylation of Ser48 in the first protein RNA recognition motif of HuR has potentially a stabilizing effect, exerting a regulatory role on the biological function of protein by regulating the maintenance of the dimmerized quaternary state, a requirement of the complex prior to RNA binding activity [58].


Myelin basic protein (MBP) binds to negatively charged lipids on the cytosolic surface of oligodendrocyte membranes and is most likely responsible for adhesion of these surfaces in the multilayered myelin sheath. It can also polymerize actin, bundle F-actin filaments, and bind actin filaments to lipid bilayers through electrostatic interactions. MBP consists of a number of posttranslationally modified isomers of varying charge, some resulting from phosphorylation at several sites by different kinases, including mitogen-activated protein kinase (MAPK). Phosphorylation of MBP in oligodendrocytes occurs in response to various extracellular stimuli. Phosphorylation/dephosphorylation of MBP also occurs in the myelin sheath in response to electrical activity in the brain. Here we investigate the effect of phosphorylation of MBP on its interaction with actin in vitro by phosphorylating the most highly charged unmodified isomer, C1, at two sites with MAPK. Phosphorylation decreased the ability of MBP to polymerize actin and to bundle actin filaments but had no effect on the dissociation constant of the MBP−actin complex or on the ability of Ca 2+ -calmodulin to dissociate the complex. The most significant effect of phosphorylation on the MBP−actin complex was a dramatic reduction in its ability to bind to negatively charged lipid bilayers. The effect was much greater than that reported earlier for another charge isomer of MBP, C8, in which six arginines were deiminated to citrulline, resulting in a reduction of net positive charge of 6. These results indicate that although average electrostatic forces are the primary determinant of the interaction of MBP with actin, phosphorylation may have an additional effect due to a site-specific electrostatic effect or to a conformational change. Thus, phosphorylation of MBP, which occurs in response to various extracellular signals in both myelin and oligodendrocytes, attenuates the ability of MBP to polymerize and bundle actin and to bind it to a negatively charged membrane.

This work was supported by a grant to JMB from the Canadian Institutes of Health Research.

To whom correspondence should be addressed: Division of Structural Biology and Biochemistry, Hospital for Sick Children, 555 University Ave., Toronto, ON, Canada M5G 1X8. Tel.: (416) 813-5919. Fax: (416) 813-5022. E-mail: [email protected]

Division of Structural Biology and Biochemistry, Research Institute, Hospital for Sick Children.

Department of Paediatric Laboratory Medicine, Hospital for Sick Children.

Material and Methods

Materials availability

Further information and requests for resources and reagents should be directed to and will be fulfilled by Chiara Francavilla by email at [email protected] .

Reagents and Tools table

This information is provided in a separate Reagents and Tools Table.

Reagent or Resource Source Identifier
Rabbit anti Phospho-EGF Receptor (Tyr1068) Antibody Cell Signaling Technology 2234S
Mouse monoclonal Phospho-p38 MAPK (Thr180/Tyr182) (28B10) Cell Signaling Technology 9216S
Rabbit polyclonal CDK1 (phospho T161) Abcam ab47329-100ug
Rabbit polyclonal CDK1 Abcam ab131450-100ug
Mouse monoclonal FIP1/RCP antibody Bio Techne NBP2-20033
Mouse monoclonal ERK 1/2 Santa Cruz Biotechnology sc-135900
Mouse monoclonal γ-Tubulin Sigma-Aldrich T5326
Mouse monoclonal Vinculin Sigma Aldrich V9264-200UL
Rabbit polyclonal pEGFR Thr669 Cell Signaling Technology 3056s
Rabbit monoclonal pEGFR Thr669 Cell Signaling Technology 8808s
Rabbit monoclonal p44/42 MAPK (Erk1/2) (137F5) Cell Signaling Technology 4695S
Mouse monoclonal GAPDH Abcam ab8245-100ug
Mouse monoclonal EGFR (Ab-1) Merck GR01L-100UG
Rabbit polyclonal EGFR millipore 06-847
Rabbit monoclonal FGFR1 antibody D8E4 Cell Signaling Technology 9740
Rabbit polyclonal SH3BP4 Abcam PLC ab106609-100ug
Rabbit monoclonal FGF Receptor 2 (D4L2V) Cell Signaling Technology 23328S
Rabbit monoclonal P38 Cell Signaling Technology 9212
Rabbit monoclonal GFP Cell Signaling Technology 2956
Peroxidase-AffiniPure F(ab')2 Fragment Goat Anti-Mouse IgG (H + L) (min X Hu, Bov, Hrs Sr Prot) Stratech 115-036-062-JIR-0.5ml
Peroxidase-AffiniPure F(ab')2 Fragment Goat Anti-Rabbit IgG (H + L) (min X Hu Sr Prot) Stratech 111-036-045-JIR-0.5m
Mouse monoclonal to EEA1 BD Bioscience 610457
Goat anti-Rabbit IgG (H + L) Secondary Antibody, Alexa Fluor® 488 conjugate Invitrogen A11034
Goat anti-Mouse IgG (H + L) Secondary Antibody, Alexa Fluor® 488 conjugate Invitrogen A11001
Goat anti-Rabbit IgG (H + L) Secondary Antibody, Alexa Fluor® 568 conjugate Invitrogen A11011
Donkey Anti-Mouse IgG (H + L) Secondary Antibody, Alexa Fluor® 647 conjugate Invitrogen A31571
Donkey Anti-Rabbit IgG (H + L) Secondary Antibody, Alexa Fluor® 647 conjugate Invitrogen A31573
Bacterial and Virus Strains
NEB® 10-beta Competent E. coli (High Efficiency) New England Biolabs Cat. No: C3019H
Biological Samples
Chemicals, Peptides, and Recombinant Proteins
Trypsin porcine pancreas (proteomics grade) Sigma-Aldrich T6567
Lysyl Endopeptidase FUJIFILM Wako Chemicals 2541
TiO beads “Titanspheres” GL Sciences 5020-75000
Pre-cast gradient gel: Nu-PAGE 4-12% Bis-Tris Gel 1.0mm 10 well Invitrogen NP0321BOX
Sep-Pak Classic C18 cartridges Waters WAT051910
Solid Phase Extraction Disk “Empore” C18 (Octadecyl) 3 M Agilent Technologies 2215
Solid Phase Extraction Disk “Empore” C8 (Octyl) 3 M Agilent Technologies 2214
L-ARGININE:HCL Cambridge Isotope Laboratories CLM-2265-H-0.25
L-ARGININE:HCL Cambridge Isotope Laboratories CNLM-539-H-0.5
L-ARGININE:HCL Sigma-Aldrich A6969
L-LYSINE:2HCL Cambridge Isotope Laboratories DLM-2640-0.5
L-LYSINE:2HCL Cambridge Isotope Laboratories CNLM-291-H-0.5
L-LYSINE:2HCL Sigma-Aldrich L8662
2,5-Dihydroxybenzoic acid Sigma-Aldrich 85707
RPMI 1640 Medium for SILAC ThermoFisher Scientific 88365
TRIzol™ Reagent ThermoFisher Scientific Cat. No. 15596026
DIHYDROETHIDIUM Cambridge Bioscience 12013-5mg-CAY
Hoechst 33342 New England Biolabs 4082S
Lipofectamine RNAiMAX Transfection Reagent ThermoFisher Scientific 10601435
Lipofectamine Transfection Reagent Life Technologies 18324020
FuGENE HD Transfection Reagent Promega UK E2311
Sodium Pyruvate solution 100mM (100ml) Sigma-Aldrich S8636-100ML
Crystal violet solution Sigma-Aldrich V5265-250ML
Carestream Kodak BioMax MR Film Kodak Z350370-50EA
Xtra-Clear Flat 8-Strip Caps Star labs I1400-0900-C
96-Well PCR Plate Non-Skirted Low Profile Natural Star labs E1403-0200-C
RPMI 1640 Medium Glutamax Supplement (500ml) Gibco 61870010
ReliaPrep RNA Cell Miniprep System NEB Z6011
Color Prestained Protein Standard Broad Range NEB P7712S
Prestained Protein Standard Broad Range Sigma-Aldrich SDS7B2
T4 DNA Ligase 20,000 u NEB? M0202S
DMEM High glucose HEPES w/o Glutamine and Sodium pyruvate Sigma-Aldrich D6171-6X500ML
DMEM AQ medium Sigma-Aldrich D0819-500ml
RPMI 1640 w/L-Glutamine-Bicarbonate Sigma-Aldrich R8758-6X500ML
Q5 High Fidelity 2x mastermix NEB M0492S
Nutrient Mix F12 HAM Sigma-Aldrich N6658-500ML
Human EGF (Animal Free) PeproTech AF-100-15-1000
MG132 Fisher Sientific 15465519
HYPERFILM ECL 18X24CM VWR International Ltd 28-9068-37
Albumin, Bovine Fraction V (BSA), 100 Grams Cat No: A30075-100.0 Melford Biolaboratories Ltd A30075-100.0
Bradford Reagent Bio-Rad 5000205
Clarity ECL Bio-Rad 1705061
GoScript Reverse Transcription Mix, Random Primers Promega A2801
Pierce Protease Inhibitor Tablets-20 tablets Life Tehnologies A32963
qPCRBIO SyGreen Mix Separate-ROX pcr biosystems PB20.14-50
ProLong Diamond Antifade Mountant-2 mL Life Technologies P36965
ExoSAP-IT Life Technologies 78250.40.ul
DMSO Sigma-Aldrich 276855-250ml
ACETONITRILE VWR International Ltd 1.00030.2500
Escort IV SLS L3287-1ML
Penicillin-Streptomycin (10,000 U/mL) Life Technologies Ltd 15140122
Human EGF Sigma-Aldrich E9644-.2MG
Human TGFα Pepro Tech Limited 100-16A
Human FGF1 Pepro Tech Limited 100-17A-50
Human FGF7 Francavilla et al 2013 PI: Prof Olsen
Human FGF3 Bio Techne 1206-F3-025
Human FGF10 Francavilla et al 2013 PI: Prof Olsen
Enkamin-E Pepro Tech Limited A14-529EP
PD173074 Selleckchem S1264
AG1478 Cell Signalling Technologies 9842
U2106 Cell Signalling Technologies 9903
BMS582949 Selleck Chem S8124
Collagen I, HC, Rat Tail, 100 mg Corning 354249
DMEM powder, high glucose Thermo Fisher 52100021
Fetal Bovine Serum, South American origin Life Technologies 10270106
TW PC MEMBRANE,6.5MM,8.0UM Transwell Inserts Sigma Aldrich CLS3422-48EA
Calcein AM cell permanent Dye Fisher Scientific C1430
Glacial Acetic Acid (HPLC Grade) Fisher Scientific UK 10060000
Formic Acid (HPLC Grade) Sigma-Aldrich 5438040250
Trifluoracetic Acid (Spectroscopy Grade) Sigma-Aldrich 1082621000
Dispase Stem Cell Technologies 7913
Matrigel Corning 354230
DAPI (4',6-Diamidino-2-Phenylindole, Dihydrochloride) Life Technology D1306
Transferrin From Human Serum, Alexa Fluor™ 647 Conjugate Invitrogen T23366
Transferrin From Human Serum, Tetramethylrhodamine Conjugate Invitrogen T2872
Critical Commercial Assays
Click-iT EdU Alexa Fluor 488 Imaging Kit-1 kit Life Technologies C10337
Venor®GeM Classic Mycoplasma PCR Detection Kit(100 tests) Cambridge Bioscience 11-1100
ProtoScript II First Strand cDNA Synthesis Kit New England Biolabs E6560L
ReliaPrep RNA Cell Miniprep System Promega Z6011
Tumor dissociation kit Miltenyi Biotec 130-095-929
Isolate II PCR and Gel kit Bioline BIO-52059
Isolate II plasmid mini kit Bioline BIO-52056
Deposited Data
Raw data (MS) This paper (dataset identifier PXD018184)
Experimental Models: Cell Lines
HCC1937 LGC ATCC® CRL-2336
Experimental Models: Organisms/Strains
BB6RC37 Eyre et al ( 2016 ) PI: R. Clarke
Recombinant DNA
EGFR (pRK5-EGFR) Addgene Plasmid #65225
EGFRT693A Mutagenesis of above
eGFP-Rab11 Addgene Plasmid #12674
eGFP-Rab11_S52N Mutagenesis of above
Dynamin_K44a-eGFP Mutagenesis of Addgene plasmid Plasmid # 34680
HA-FGFR1c Francavilla et al ( 2009 ) PI: Dr Cavallaro
HA-FGFR2b Francavilla et al ( 2013 ) PI: Prof Olsen
HA_FGFR2b_Y656F/Y657F Francavilla et al ( 2013 ) PI: Prof Olsen
Software and Algorithms
Fiji- Image J version: 1.52p Schindelin et al ( 2012 )
GraphPad Prism version 8.0.0 GraphPad Software
MaxQuant version Cox and Mann ( 2008 )
WebGestalt 2019 Liao et al (2019)
Perseus versions or Tyanova et al (2016)
Cytoscape version 3.7.2 Shannon et al ( 2003 )
STRING version 11 Szklarczyk et al ( 2019 )
R framework R Core Team ( 2018 )
Confocal Microscope Leica Sp8 Inverted Lecia
Mx3000P qPCR machine Agilent
UltiMate® 3000 Rapid Separation LC Dionex
QE-HF LC-MS/MS Thermo Fisher Scientific

Methods and Protocols

Experimental models

Cell culture and SILAC labelling

Human breast cancer cell lines were purchased from ATCC, authenticated through short tandem repeat (STA) analysis of 21 markers by Eurofins Genomics, checked monthly for mycoplasma via a PCR-based detection assay (Venor®GeM—Cambio) and grown in the indicated media supplemented with 2 mM L-glutamine and 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% foetal bovine serum. MCF-7 was grown in DMEM/F12. MDA-MB-415 and BT20 were grown in DMEM. HCC1937, T47D and BT549 were grown in RPMI. 1 mM sodium pyruvate was added to T47D.

For quantitative mass spectrometry, BT549 or T47D cells were labelled in SILAC RPMI (PAA Laboratories GmbH, Germany) supplemented with 10% dialyzed foetal bovine serum (Sigma), 2 mM glutamine (Gibco), 100 U/ml penicillin and 100 μg/ml streptomycin for 15 days to ensure complete incorporation of amino acids, which was verified by MS analysis. Three cell populations were obtained: one labelled with natural variants of the amino acids (light label Lys0, Arg0), the second one with medium variants of amino acids (medium label L-[13C6] Arg (+6) and L-[2H4]Lys (+4) Lys4/Arg6) and the third one with heavy variants of the amino acids (heavy label L-[13C6,15N4]Arg (+10) and L- [13C6,15N2]Lys (+8) Lys8/Arg10). The light amino acids were from Sigma, whilst their medium and heavy variants were from Cambridge Isotope Labs (Massachusetts, US).

Breast cancer organoid culture and protein isolation

Organoids were generated from a triple-negative breast cancer PDX tumour, BB6RC37 (Eyre et al, 2016 ). The tumours were minced and digested using a tumour dissociation kit (Miltenyi Biotec) on an orbital shaker at 37°C for 1–2 h. The cells were sequentially strained through 100-µm and 40-µm meshes. 50,000 cells were resuspended in 50 µl cold growth factor-reduced Matrigel (Corning 354230), set as domes in a 24-well plate for 30 min and cultured at 37°C in media as defined by (Sachs et al, 2018 ). The organoids were cultured in media with or without FGF7/10 for 14 days, and EGF/Heregulin were removed from the media 24 h before lysates were obtained. Lysates were prepared by mechanically disaggregating the domes and digesting the Matrigel for 1 h using dispase at 37°C (Stem Cell Technologies, 7913). Cells were washed in PBS and resuspended in lysis buffer as previously described (Santiago-Gomez et al, 2019 ).

Quantitative phosphoproteomics

Experimental design and sample preparation

TPA1: for each cell line and each stimulus, we analysed duplicates for each time point, considering both 1- and 8-min. time points as representative of early signalling. Therefore, we compared four label-free samples for each stimulus in each cell line (Datasets EV1 and EV2). The cell pellet was dissolved in denaturation buffer (6 M urea, 2 M thiourea in 10 mM HEPES pH 8). We obtained 1 mg of proteins from each sample. Cysteines were reduced with 1 mM dithiothreitol (DTT) and alkylated with 5.5 mM chloroacetamide (CAA). Proteins were digested with endoproteinase Lys-C (Wako, Osaka, Japan) and sequencing grade modified trypsin (modified sequencing grade, Sigma) followed by quenching with 1% trifluoroacetic acid (TFA). Peptides were purified using reversed-phase Sep-Pak C18 cartridges (Waters, Milford, MA) and eluted with 50% acetonitrile (ACN). After removing ACN by vacuum concentrator at 60°C, peptides were suspended in phosphopeptide immunoprecipitation buffer (50 mM MOPS pH 7.2, 10 mM sodium phosphate, 50 mM NaCl) and dissolved overnight. Clarified peptides were transferred in a new tube containing immobilized phosphorylated tyrosine antibody beads (pY100-AC, Cell Signalling Technologies) and incubated for two hours at 4°C. After five washes with immunoprecipitation buffer followed by two washes with 50 mM NaCl, the enriched peptides were eluted from the beads three times with 50 μL of 0.1% TFA, loaded on C18 STAGE-tips, and eluted from STAGE-tips with 20 μL of 40% ACN followed by 10 μL 60% ACN and reduced to 5 μL by SpeedVac and 5 μL 0.1% formic acid (FA) 5% ACN added. Peptides from the supernatant were purified using reversed-phase Sep-Pak C18 cartridges (Waters, Milford, MA) and eluted with 50% ACN and further enriched for phosphorylated serine- and phosphorylated threonine-containing peptides, with Titansphere chromatography. Six mL of 12% TFA in ACN was added to the eluted peptides and subsequently enriched with TiO2 beads (5 μm, GL Sciences Inc., Tokyo, Japan). The beads were suspended in 20 mg/mL 2,5-dihydroxybenzoic acid (DHB), 80% ACN, and 6% TFA and the samples were incubated in a sample to bead ratio of 1:2 (w/w) in batch mode for 15 min with rotation. After 5-min centrifugation, the supernatant were collected and incubated a second time with a twofold dilution of the previous bead suspension. The beads were washed with 10% ACN, 6% TFA followed by 40% ACN, 6% TFA and collected on C8 STAGE-tips and finally washed by 80% ACN, 6% TFA. Elution of phosphorylated peptides was done with 20ul 5% NH3 followed by 20 μL 10% NH3 in 25% ACN, which were evaporated to a final volume of 5 μL in a sped vacuum. The concentrated phosphorylated peptides were acidified with addition of 20 μL 0.1% TFA, 5% ACN and loaded on C18 STAGE-tips. Peptides were eluted from STAGE-tips with 20 μL of 40% ACN followed by 10 μL 60% ACN and ACN and reduced to 5 μL by SpeedVac and 5 μL 0.1% FA, 5% ACN added.

A small amount of the eluted peptides (1%) was taken for proteome analysis before enrichment of phosphorylated peptides: after evaporation in a speed vacuum, 40 μl of 0.1% TFA, 5% ACN were added followed by MS analysis.

TPA2: we analysed label-free triplicates for each condition, T47D depleted or not of TTP or RCP and stimulated or not with FGF10. Cells were washed with PBS and lysed at 4°C in ice-cold 1% triton lysis buffer supplemented with Pierce protease inhibitor tablet (Life Technologies) and phosphatase inhibitors: 5 nM Na3VO4, 5 mM NaF and 5 mM β-glycerophosphate. Proteins were precipitated overnight at −20°C in fourfold excess of ice-cold acetone. The acetone-precipitated proteins were solubilized in denaturation buffer (10 mM HEPES, pH 8.0,6 M urea, 2 M thiourea), and 5 mg of proteins was reduced, alkylated and digested, as described above. All the steps were performed at room temperature. The peptide mixture was desalted and concentrated on a C18-Sep-Pak cartridge, eluted and enriched with TiO2 beads, as described above.

TPA3: we analysed duplicates of SILAC-labelled BT549, transfected and treated as described in Fig 2A. We followed the same procedure described for TPA2 with the only difference that 5 mg of each SILAC-labelled lysates was mixed in equal amount before digestion and TiO2 chromatography.

EGFR- and EGFR_T693A-expressing T47D cells: we analysed duplicates of SILAC-labelled T47D transfected and treated as described in Fig 6A. We followed the same procedure described for TPA1 with the only difference that 5 mg of each SILAC-labelled lysates was mixed in equal amounts before digestion and phosphorylated tyrosine enrichment followed by TiO2 chromatography and peptides purification.

Mass spectrometry

Purified peptides were analysed by LC-MS/MS using an UltiMate® 3000 Rapid Separation LC (RSLC, Dionex Corporation, Sunnyvale, CA) coupled to a QE-HF (Thermo Fisher Scientific, Waltham, MA) mass spectrometer. Mobile phase A was 0.1% FA in water, and mobile phase B was 0.1% FA in ACN and the column was a 75 mm x 250 μm inner diameter 1.7 μM CSH C18, analytical column (Waters). A 1 μl aliquot of the sample (for proteome analysis) or a 3 μl aliquot was transferred to a 5 μl loop and loaded on to the column at a flow of 300 nl/min at 5% B for 5 and 13 min, respectively. The loop was then taken out of line and the flow was reduced from 300 nl/min to 200nl/min in 1 min., and to 7% B. Peptides were separated using a gradient that went from 7% to 18% B in 64 min., then from 18% to 27% B in 8 min. and finally from 27% B to 60% B in 1 min. The column was washed at 60% B for 3 min. and then re-equilibrated for a further 6.5 min. At 85 min, the flow was increased to 300nl/min until the end of the run at 90min. Mass spectrometry data were acquired in a data directed manner for 90 min in positive mode. Peptides were selected for fragmentation automatically by data-dependent analysis on a basis of the top 8 (phosphoproteome analysis) or top 12 (proteome analysis) with m/z between 300 and 1750Th and a charge state of 2, 3 or 4 with a dynamic exclusion set at 15 s. The MS resolution was set at 120,000 with an AGC target of 3e6 and a maximum fill time set at 20ms. The MS2 resolution was set to 60,000, with an AGC target of 2e5, and a maximum fill time of 110 ms for Top12 methods, and 30,000, with an AGC target of 2e5, and a maximum fill time of 45 ms for Top8 analysis. The isolation window was of 1.3Th (2.6 Th for SILAC-labelled samples), and the collision energy was of 28.

Raw files analysis

Raw data were analysed by the MaxQuant software suite (Cox & Mann, 2008 ) ( version using the integrated Andromeda search engine (Cox et al, 2011 ). Proteins were identified by searching the HCD-MS/MS peak lists against a target/decoy version of the human UniProt Knowledgebase database that consisted of the complete proteome sets and isoforms (v.2016 https.// supplemented with commonly observed contaminants such as porcine trypsin and bovine serum proteins. Tandem mass spectra were initially matched with a mass tolerance of 7 ppm on precursor masses and 0.02 Da or 20 ppm for fragment ions. Cysteine carbamidomethylation was searched as a fixed modification. Protein N-acetylation, N-pyro-glutamine, oxidized methionine and phosphorylation of serine, threonine and tyrosine were searched as variable modifications. Protein N-acetylation, oxidized methionine and deamidation of asparagine and glutamine were searched as variable modifications for the proteome experiments. For the quantification of SILAC-labelled samples, labelled lysine and arginine were specified as fixed or variable modification, depending on prior knowledge about the parent ion (MaxQuant SILAC triplet identification). In all the other experiments, label-free parameters were used as described (Cox et al, 2014 ). False discovery rate was set to 0.01 for peptides, proteins and modification sites. Minimal peptide length was six amino acids. Site localization probabilities were calculated by MaxQuant using the PTM scoring algorithm (Olsen et al, 2006 ). The datasets were filtered by posterior error probability to achieve a false discovery rate below 1% for peptides, proteins and modification sites. Only peptides with Andromeda score > 40 were included.

Data and statistical analysis

All statistical and bioinformatics analyses were done using the freely available software Perseus, version or (Tyanova & Cox, 2018 ), R framework (R Core Team, 2018), Bioconductor R-package LIMMA (Bolstad et al, 2003 ), WebGestalt (Liao et al, 2019 ), STRING (Szklarczyk et al, 2019 ), Cytoscape (version 3.7.2) (Shannon et al, 2003 ). All measured peptide intensities were normalized using the “normalizeQuantiles” function from the Bioconductor R-package LIMMA, which normalizes the peptide intensities such that each quantile for each sample is set to the mean of that quantile across the dataset, resulting in peptide intensity distributions that are empirically identical. Each dataset was normalized individually. Subsequent data analysis was performed using Microsoft Office Excel, R and Perseus. For the SILAC datasets, we used the normalized SILAC ratios from MaxQuant output txt files. Only peptides with localization probabilities higher than 0.75 (class I, shown in Datasets EV1, EV3–EV6 Olsen et al, 2006 ) were included in the downstream bioinformatics analysis. Pearson correlation was calculated in R. For TPA1, we impute missing values using Perseus default settings, we subtracted the control from log intensity values in order to be able to compare all the cell lines against each other and we used the median for each condition. Hierarchical clustering based on correlation was performed after multi-sample ANOVA test with default parameters in Perseus. For TPA2, we calculated the median and then considered only rows with four valid values, followed by hierarchical clustering based on Euclidean distance in Perseus. For TPA3 and the EGFR/EGFR_T693A T47D dataset, we imputed missing values using Perseus default settings and then calculated the median, followed by hierarchical clustering based on Euclidean distance in Perseus. Clusters used in the follow-up analysis were defined by Perseus and manually checked.

The enrichment of KEGG or GO terms was performed in WebGestalt using the ORA default parameters, and significantly over-represented terms within the data were represented in bar plots. The relation of genes to other diseases was based on the database DISEASES (Pletscher-Frankild et al, 2015 ).

All the protein interaction networks were obtained using the STRING protein interaction database using high confidence, and interactions derived from the Experiments and Databases evidence channels. Data visualization was performed using the software Cytoscape. The Venn diagram was created using the web tool

Biochemical assays

RNA isolation and real-time PCR analysis

RNA from cell lines was isolated with TRIZOL® (Invitrogen). After chloroform extraction and centrifugation, 5 µg RNA was DNase treated using RNase-Free DNase Set (Qiagen) and 1 µg of DNase treated RNA was then taken for cDNA synthesis using the Protoscript I first strand cDNA synthesis kit (New England Biolabs). Selected genes were amplified by quantitative real-time PCR (RT–qPCR) using Sygreen (PCR Biosystems). Relative expression was calculated using the delta-delta CT methodology, and beta-actin was used as reference housekeeping gene. Sequences for primers used can be found in the accompanying Reagent Table. qPCR machine used was Applied Biosystems MX300P.

Transfection and RNA interference

All transfections were carried out in Gibco opti-MEM glutamax reduced serum media (Thermo Fisher Scientific). For RNA interference, all cells were transfected using Lipofectamine RNAiMax (Thermo Fisher Scientific), according to manufacturer instructions. Validated double-stranded stealth siRNA oligonucleotides were used for RNA interference. siRNA Universal Negative Control #2 (Sigma-Aldrich) was used as a control in all RNA interference experiments. BT549 and BT20 cells were transfected using Lipofectamine 3000 (Thermo Fisher Scientific) according to the manufacturer’s instructions, 24 h after RNA interference transfection where indicated. T47D cells were transfected using Escort IV according to manufacturer instructions, same as above. Assays were performed 36 h after transfection, as previously described (Francavilla et al, 2016 ). Where assays were performed more than 36 h after transfection, RNAi and expression were assessed at time of assay to confirm expression.

Cell lysis, protein immunoprecipitation and western blotting

Cells were serum starved overnight in serum-free medium and stimulated for the indicated time points with 100 ng/ml of FGF7, FGF10, EGF or TGFα. Ligands were replenished every 24 h for long-term (24–72 h) stimulation. Where indicated, cells were pre-incubated for 2 h with 100 nM PD173074, 500 nM AG1478, 20 μM U1206, 1 μM MEK162 or 10 μM BMS582949. Control cells were pre-incubated with DMSO alone. After stimulation, cell extraction and immunoblotting were performed as previously described (Francavilla et al, 2016 ). Proteins were resolved by SDS–PAGE and transferred to nitrocellulose membranes (Protran, Biosciences). Proteins of interest were visualized using specific antibodies, followed by peroxidase-conjugated secondary antibodies and by an enhanced chemiluminescence kit (Amersham Biosciences). Blots were visualized either using film exposure or the Universal Hood II Gel Molecular Imaging System (Bio-Rad). Each experiment was repeated at least three times and produced similar results.

Immunoprecipitation of FGFR2 from cell extracts was performed as previously described (Francavilla et al, 2016 ), using anti BEK (Santa Cruz Biotechnology, sc-121). Each experiment was repeated at least three times and produced similar results.

Biotinylation assays

Biotinylation pull down experiments were performed as described previously (Lobingier et al, 2017 ). Briefly biotinylation experiments were performed by transfecting GFP-Rab11-APEX2 constructs in to 2 million cells plated in 10-cm dishes. Cells were pre-incubated (40 min) with biotin phenol (Iris Biotech), after stimulation with ligands, hydrogen peroxide (Sigma-Aldrich) was added for 1 min before quenching with Trolox (Sigma-Aldrich) and sodium ascorbate (VWR) during ice-cold lysis. A 2-hour RT pull down with streptavidin beads was then performed running the supernatant against the bound proteins.

Proliferation assays

Incucyte cell proliferation assay

Indicated cell lines were seeded into 24-well plates at a density of 15,000–20,000 cells per well, depending on growth rate and the design of the experiment. After plating cells were starved and stimulated with indicated ligands every 24 h and imaged every hour using the Incucyte ZOOM (Essen Bioscience), phase-contrast images were analysed to detect cell proliferation based on cell confluence. And average confluency value over 4 h was used to determine the starting confluency from which a relative growth change was calculated. Statistical analysis was performed at the endpoint across repeats, as indicated in the Fig legends.

Crystal Violet

Indicated cells were stained after experimentation by being fixed with 0.5% w/v crystal violet (Sigma) in 4% w/v paraformaldehyde/PBS for 30 min. Fixed cells were then solubilized in 2% w/v SDS/PBS and absorbance measured at 595 nm using Synergy H1 microplate reader (BioTek). Statistical analysis was performed at the endpoint across repeats, as indicated in the Fig legends.

EdU incorporation

Indicated cells were labelled with 20 µM 5-ethynyl-2'-deoxyuridine (EdU) for 4 h and processed following the manufacturer's protocol (Click-iT® EdU Alexa Fluor® 488 Imaging Kit, Thermo Fisher). Prior to imaging, cells were then stained with 5 ng/ml Hoecsht 3342 for 15 min. Stained cells were analysed using a using a Leica microscope system. Statistical analysis was performed at the endpoint across repeats, as indicated in the Fig legends.

Invasion assay

Rat tail-derived collagen I (Corning) was supplemented with 25 µg/ml human fibronectin (Sigma) in DMEM and polymerized in 8-µm Transwell inserts (Corning) for 30 min at room temperature followed by 30 min at 37°C/5% CO2. 5 x 10 4 BT20 cells were seeded on the reverse of each insert and incubated for 6 h at 37°C/5% CO2. Inserts were gently washed and placed in serum-free DMEM and the upper chamber filled with DMEM supplemented with 10% FCS (Life Technologies) and either PBS or 100 ng/ml FGF10 (PeproTech). After 72 h, cells were stained with 500 ng/ml Calcein AM (Thermo Fisher) for 1 h and visualized by Leica Sp8 inverted confocal microscopy in serial sections of 20 µm. Fluorescence intensity of each section was determined using ImageJ v. 1.52p (Schindelin et al, 2012 ) and proportion of invading cells estimated by comparing the total intensity beyond 40 µm with the total overall intensity per insert using GraphPad PRISM version 8.0.0. Statistical analysis was performed at the endpoint across repeats, as indicated in the Fig legends.


Immunofluorescence staining was performed as previously described (Francavilla et al, 2016 ). To detect HA-FGFR2b or endogenous FGFR2, we incubated cells with 10 μg/ml of anti-HA (Covance) or anti-FGFR2 antibody (Cell Signalling) for 45 minutes with gentle agitation. The binding of the antibody did not activate receptor signalling in untreated cells nor induced receptor internalization (see control cells in Fig 1), as previously reported (Francavilla et al, 2009 ). After stimulation, cells were incubated at 37°C for different time points. When indicated, each inhibitor was added prior stimulation. At each time point, non-permeabilized cells were either fixed to visualize the receptor on the cell surface (plasma membrane) or acid-washed in ice-cold buffer (50 mM glycine, pH 2.5) to remove surface-bound antibody. Acid-washed cells were then fixed and permeabilized to visualize the internalized receptor (cytoplasm). Finally, to detect FGFR2b cells were stained with AlexaFluor488-conjugated donkey anti-mouse or anti-rabbit (Jackson ImmunoResearch Laboratories). Nuclei were stained with DAPI. Coverslips were then mounted in mounting medium (Vectashield Vector Laboratories).

For co-localization experiments, cells were acid-washed, fixed, permeabilized with 0.02% saponin (Sigma), treated with a primary antibody against FGFR2, EGFR, TTP, RCP, phosphorylated T693 EGFR, EEA1 for 60 min at 37 °C and stained with AlexaFluor488 (or 568 or 647)-conjugated donkey anti-mouse or anti-rabbit. Samples either expressing GFP-tagged proteins or treated with TRITC-transferrin or Alexa 647-transferrin (to stain transferrin receptor, Tf-R), added to the medium at a final concentration of 50 μg/mL, were kept in the dark. Nuclei were stained with DAPI. Coverslips were then mounted in mounting medium (Vectashield Vector Laboratories).

All the images were acquired at room temperature on a Leica TCS SP8 AOBS inverted confocal using a 100x oil immersion objective and 2.5x confocal zoom. The confocal settings were as follows: pinhole, 1 airy unit, format, 1,024 × 1,024. Images were collected using the following detection mirror settings: FITC 494-530nm Texas red 602-665nm Cy5 640-690nm. The images were collected sequentially. Raw images were exported as.lsm files, and adjustments in image contrast and brightness were applied identical for all images in a given experiment using the freely available software ImageJ v. 1.52p (Schindelin et al, 2012 ).

Quantification of the recycling assay

Quantification of recycling was performed as described (Francavilla et al, 2016 ). For each time point and each treatment, the presence (total) and the localization (cell surface versus internalized) of HA-FGFR2 or endogenous FGFR2 were assessed in at least seven randomly chosen fields. Approximately 100 cells per condition (both acid-washed and not) were analysed from three independent experiments. The results are expressed as the percentage of receptor-positive cells (green) over total cells (corresponding to DAPI-stained nuclei) and referred to the values obtained at time zero. Statistical analysis was performed across repeats, as indicated in the Fig legends.

Quantification of expression fraction, overlap fraction and co-localization

Images were pre-processed using an “À trous” wavelet band-pass filter to reduce the contribution of high-frequency speckled noise to the co-localization calculations. Pixel intensities were then normalized from the original 8-bit range [0,255] to [0,1]. To ensure that co-localization was only computed in well-determined regions of interest (ROI), we used the Fiji/ImageJ (Schindelin et al, 2012 ) built-in ROI manager to create and record these regions.

Finally, to quantify the actual level of co-localization between two markers (e.g. R and G), we used the Manders co-localization coefficients (MCC) M1 and M2 (Manders et al, 1996 ). M1 measures the fraction of the R marker in compartments that also contain the G marker, and M2, the fraction of the G marker in compartments that also contain the R marker. Lower-bound thresholds for pixel intensities IR and IG were automatically determined using the Costes method (Costes et al, 2004 ).

To measure the simultaneous overlap of our three, red, far-red and green markers (R, F, G), we first used the overlap image between marker R and marker F as defined above (i.e. e. IF,R = IF × IR). We then measured the MCC co-localization parameter of this combined image against a Green marker using the MCC formulae above, together with the Costes method to determine the TFR and TG thresholds.

The scripts for the quantification of co-localization were written in the Python language, and the code for Costes-adjusted MCC was taken verbatim from the CellProfiler (McQuin et al, 2018 ) code base.

Student’s t-test was subsequently used to determine the difference in pixel overlap fraction or Manders (Costes) coefficient between different experimental conditions in Figs 1 and Figs 5, and Appendix Fig S4 and S6.


Alzheimer’s disease (AD) is a common progressive neurodegenerative disorder characterized by severe neuronal loss leading to cognitive dysfunction. At present, no effective drug or treatment has been available for the prevention or cure of this disease [1,2,3]. The major neuropathological feature of AD is the aggregation and deposition of amyloid β (Aβ) peptides which are considered one of the major risk factors and causes of AD [3, 4]. Aggregated Aβ peptides play critical roles in neuronal degeneration, neuroinflammation, and oxidative stress [5, 6]. Thus, therapeutic approaches for the effective removal of Aβ deposits may provide neuroprotective benefits in AD.

According to recently published studies, AD may be exacerbated by neuroinflammation as well as Aβ peptides. Neuroinflammation is mainly mediated by microglial cells which are resident immune cells in the central nervous system (CNS) [6, 7]. Under physiologic conditions, microglial cells are involved in various functions, including the regulation of brain development, the maintenance of homeostasis, and the clearance of old synapses or other debris such as Aβ peptides. In AD, microglia has been known to be chronically activated, resulting in impairment of Aβ clearance, overexpression of pro-inflammatory signals, and consequently, neurotoxicity [8, 9]. Therefore, the regulation of microglial activation may be an important therapeutic strategy of AD by lowering excessive pro-inflammatory immune chemotaxis and enhancing neuroprotective function, leading to the modulation of neuroinflammation [10, 11].

Microglial activation plays an important role in several well-known pro-inflammatory signal cascades [12], including those mediated by mitogen-activated protein kinase (MAPK). The MAPK family is a family of serine/threonine protein kinases regulating cell properties in response to extracellular stimuli such as growth factors and inflammatory cytokines [13]. P38 MAPKs, one of the three MAPKs in mammalian cells, are primarily activated by inflammatory cytokines and environmental stresses [14]. P38 MAPKs have four isoforms α, β, γ, and δ which can be divided into two subgroups one is p38α and p38β, and the other is p38γ and p38δ [15]. Among these, p38α and p38β MAPKs are highly expressed in adult mouse brain [16]. SB203580, a traditional p38α/β MAPK inhibitor, had shown therapeutic effects in the LPS-induced depression model and AD mouse model [17, 18]. Recently, two selective p38α MAPK inhibitors, neflamapimod (VX-745) and MW150, also showed therapeutic effects in aged rats and AD mouse model respectively [19, 20]. The phase 2a clinical trials of neflamapimod showed improvements in episodic memory in early AD patients [21, 22]. These results suggest that p38α/β MAPKs are potentially important targets in AD therapeutics, and thus, their inhibitors may be promising drugs.

In our previous study, a novel, selective p38α/β MAPK inhibitor, NJK14047, could successfully ameliorate microglia-mediated neuroinflammation [23]. It reduced inflammatory responses mediated by lipopolysaccharide (LPS) in the BV2 cells and the mouse model. Based on this finding, we intended to investigate the potential of NJK14047 as a therapeutic agent for AD. In this study, we used five familial Alzheimer’s disease (5XFAD) transgenic mice which have been commonly used as AD mouse models. 5XFAD mice overexpress human amyloid precursor protein (hAPP) with three FAD mutations [Swedish (K670N, M671L) Florida (I716V) and London (V717I)] and human presenilin 1 (PS1) with two FAD mutations (M146L and L286V) under the control of the neuron-specific Thy1 promoter [24]. Here, the administration of NJK14047 to the 5XFAD mouse model was suggested to ameliorate memory loss, Aβ deposition, neuroinflammation, and neuronal degeneration via the selective inhibition of p38α/β MAPKs.


Complement activates p38 in GEC. We and others reported previously that complement C5b-9 activates kinases such as protein kinase C, ERK (3), and JNK (31) in GEC. In the current study, we examined if complement activates another MAPK, p38. When antibody-sensitized GEC were exposed to NS to form complement C5b-9, p38 phosphorylation increased by 2.2 ± 0.5-fold, compared with control (cells exposed to HIS) [no treatment 0.7 ± 0.1, HIS 1.0, NS 2.2 ± 0.5 (P < 0.01 vs. HIS), anisomycin 4.6 ± 1.4 (P < 0.001 vs. HIS), n = 6 Fig. 1A]. The enzyme activity of p38 quantified by immune-complex kinase assay also increased in complement-treated cells by 2.4 ± 0.3-fold, correlating to the increase in phosphorylation [no treatment 0.4 ± 0.3, HIS 1, NS 2.4 ± 0.3 (P < 0.01 vs. HIS), H2O2 4.5 ± 1.1 (P < 0.05 vs. HIS), n = 3–4 Fig. 1B]. To verify if C5b-9 assembly was actually required for p38 activation, we incubated antibody-treated GEC with C8D, with or without reconstitution with purified C8. C8D without C8 forms C5b-7, which was previously shown to be biologically inactive in GEC (2). GEC incubated with C8D did not show significant p38 phosphorylation, compared with HIS. However, when C8D was reconstituted with purified C8, phosphorylation of p38 was evident, indicating that formation of C5b-9 is required for p38 activation (Fig. 1C). Densitometric analysis showed that the relative p38 phosphorylation was HIS: 1.0, C8D: 0.9, C8D + C8: 2.4 (average of 2 experiments Fig. 1C).

Fig. 1.Complement C5b-9 stimulates p38 in glomerular epithelial cells (GEC) in culture. A: GEC-cytosolic phospholipase A2 (cPLA2) (GEC that stably overexpress cPLA2, see materials and methods ) were incubated with anti-GEC antiserum (5% vol/vol, 22°C, 40 min) and normal human serum (NS 2.5% vol/vol, to assemble C5b-9, 37°C, 40 min) or heat-inactivated serum (HIS 2.5% vol/vol, control). Anisomycin (1 μM) was used as positive control. Top: immunoblot. Bottom: densitometry. *P < 0.01, **P < 0.001 vs. HIS n = 6 for each group. B: GEC-cPLA2 were stimulated as in A, and p38 activity in cell lysates was evaluated as in materials and methods . H2O2 (1 mM) was used as positive control. Top: immunoblot. Bottom: densitometry. +P < 0.01, ++P < 0.05 vs. HIS n = 3–4 for each group. C: antibody-sensitized GEC-cPLA2 were incubated with HIS (2.5% vol/vol), human C8-deficient serum (C8DS 2.5% vol/vol), or C8DS supplemented with recombinant human C8 (80 μg/ml in undiluted serum) for 40 min at 37°C. H2O2 (1 mM) was used as positive control. Top: immunoblot. Bottom: densitometry (mean of 2 experiments).

To assess if complement-mediated p38 activation occurs in vivo, we next studied p38 activation in glomeruli from rats with PHN. PHN was induced by a single injection of anti-Fx1A antiserum ( materials and methods ), and glomeruli were isolated 14 days later, when rats developed significant proteinuria (∼160 mg/day, control rat: <10 mg/day). p38 was activated by approximately sixfold in glomeruli from rats with PHN, compared with control (control 13 ± 1, PHN 80 ± 22, P < 0.05, n = 3 and 6 Fig. 2A). Phosphorylation of p38 was also higher in PHN (2.9 ± 0.2-fold), compared with control (control 1 ± 0.1, PHN 2.9 ± 0.2, P < 0.01, n = 3 and 6 Fig. 2B). These results indicate that p38 is activated by complement in GEC in vitro and in vivo.

Fig. 2.p38 is activated in glomeruli from rats with passive Heymann nephritis (PHN). A: PHN was induced by intravenous injection of sheep anti-Fx1A antiserum. On day 14, rats were killed and glomeruli were prepared by differential sieving. Glomerular lysates were analyzed for p38 activity (A) or p38 phosphorylation (B) as in Fig. 1. Top: immunoblot. Bottom: densitometry. *P < 0.05, **P < 0.01 vs. control n = 3–6 rats for each group.

Arachidonic acid contributes to complement-induced p38 activation. We next addressed the mechanisms of complement-induced p38 activation. It was reported previously that complement activates cPLA2 in a [Ca 2 + ]i- and protein kinase C-dependent manner, leading to a liberation of arachidonic acid from the membrane phospholipid pool (16). We also reported that complement-induced JNK activation was, at least in part, mediated by arachidonic acid release (17). We first studied if liberation of arachidonic acid contributes to complement-induced p38 activation. GEC-cPLA2 were stimulated with antibody and complement, and p38 activation was compared with control GEC, which were transfected with vector only (GEC-Neo). p38 was activated by complement 1.4 ± 0.1 times in GEC-Neo, whereas the stimulation was 2.0 ± 0.1 times in GEC-cPLA2 (n = 3, P < 0.02), suggesting that liberation of arachidonic acid, at least partially, contributes to complement-induced p38 activation (Fig. 3A). When GEC-Neo were stimulated with exogenous arachidonic acid (30 μM), p38 was clearly activated (2.8 ± 0.6-fold of control, P < 0.05, n = 5), supporting the role of arachidonic acid in p38 activation (Fig. 3B).

Fig. 3.Arachidonic acid (AA) contributes to complement-induced p38 activation. A: GEC-Neo (vector-transfected cells) and GEC-cPLA2 (GEC that stably overexpress cPLA2) were stimulated with complement for 40 min, and p38 phosphorylation was studied by immunoblotting. Top: immunoblot. Bottom: densitometry. *P < 0.05 vs. GEC-Neo n = 3 each. B: GEC-Neo were stimulated with AA (30 μM) or vehicle [ethanol, control (cntl)] for 20 min, and p38 phosphorylation was studied by immunoblotting. Top: immunoblot. Bottom: densitometry. **P < 0.05 vs. control n = 5 each.

Unstimulated cultured rat GEC express COX-1, whereas COX-2 expression is induced by complement. Arachidonic acid liberated by cPLA2 is further converted by COX-1 and -2 into biologically active eicosanoids such as PGE2, PGF2α, and TXA2 (29). These eicosanoids act via respective specific cell-surface receptors in an autocrine fashion and are known to activate p38 in some systems. To study if eicosanoids contribute to complement-induced p38 activation, we stimulated GEC with PGE2, PGF2α, and U-46619 (stable analog of TXA2). When individually tested, these three eicosanoids failed to activate p38 consistently. Only when three eicosanoids were tested simultaneously was there a modest increase in p38 phosphorylation (∼20%, not shown). Furthermore, a nonselective COX inhibitor, indomethacin, failed to inhibit complement-induced p38 activation (not shown). Taken together, these results indicate that arachidonic acid contributes to complement-induced p38 activation but not through conversion into COX metabolites. Instead, arachidonic acid itself and/or other mediators stimulated by arachidonic acid, such as reactive oxygen species (ROS) (see Role of ROS in the activation of p38 by complement), may be responsible for p38 activation.

Role of ROS in the activation of p38 by complement. We previously showed that complement stimulates ROS generation in GEC in an NADPH oxidase-dependent manner, which contributes to JNK activation (17). We also showed that arachidonic acid stimulates ROS generation in GEC (17). To test whether ROS contribute to complement-mediated p38 activation, we examined complement-mediated p38 activation in GEC treated with anti-oxidants, GSH, and NAC. As shown in Fig. 4, GSH (10 mM) and NAC (10 mM) completely abolished complement-induced p38 phosphorylation [control 3.7 ± 0.3-fold of HIS, GSH 0.9 ± 0.03-fold of HIS (P < 0.01 vs. control), NAC 1.0 ± 0.07-fold of HIS (P < 0.01 vs. control), n = 3]. Thus ROS are likely to be responsible for complement-induced p38 activation. One mechanism of ROS generation might be arachidonic acid released by cPLA2 (17).

Fig. 4.Complement-mediated p38 activation is dependent on reactive oxygen species. GEC-cPLA2 were stimulated with antibody and complement (NS). Cells were preincubated with glutathione (GSH 10 mM) or N-acetyl cysteine (NAC 10 mM) for 30 min before the stimulation. p38 activation was evaluated by immunoblotting for phospho-p38. Top: immunoblot. Bottom: densitometry. *P < 0.01 vs. control n = 3 each.

Impact of p38 activation on GEC function. We next studied the impact of p38 activation on complement-mediated GEC injury. We initially hypothesized that p38 activation contributes to complement-mediated cell injury. However, to our surprise, when GEC were pretreated with p38 inhibitors (PD-169316 and FR-167653, 10 μM each), complement-induced cytotoxicity quantified by specific BCECF release (see materials and methods ) was significantly increased, compared with vehicle-treated cells [Fig. 5A: DMSO: NS 2.5 13 ± 3%, NS 2.75 24 ± 5%, NS 3 38 ± 5%, PD-169316: NS 2.5 20 ± 3% (P < 0.05 vs. DMSO), NS 2.75 29 ± 4% (P < 0.05 vs. DMSO), NS 3 46 ± 4% (P < 0.05 vs. DMSO), n = 3–4 Fig. 5B: DMSO: NS 2.5 18 ± 5%, NS 2.75 36 ± 7%, NS 3 49 ± 9%, FR-167653: NS 2.5 26 ± 7% (P < 0.05 vs. DMSO), NS 2.75 45 ± 7%, NS 3 58 ± 8% (P < 0.05 vs. DMSO), n = 3–4]. The concentration of FR-167653 used (10 μM) inhibited complement-induced p38 activity by 90% in GEC (data not shown). Similar results were obtained when LDH release was used to quantify cytotoxicity [DMSO: NS 5 25 ± 9%, NS 7.5 36 ± 10%, FR-167653: NS 5 29 ± 11%, NS 7.5 52 ± 10% (P < 0.05 vs. DMSO), n = 6]. These results suggest that p38 activation may protect cells from complement-mediated injury.

Fig. 5.p38 Inhibitors augment complement-induced GEC injury. GEC were preincubated with p38 inhibitors PD-169316 (A10 μM) or FR-167653 (B 10 μM) for 30 min at 37°C before stimulation with antibody and complement. Vehicle (DMSO) was used as control. Inhibitors were also included in incubations with anti-GEC antiserum and HIS/NS. After 40 min of incubation with HIS/NS, cytotoxicity was quantified by 2′,7′-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF)-release assay as in materials and methods . Numbers next to NS indicate concentration of NS (vol/vol) used. *P < 0.05 vs. vehicle n = 3–4 for each group.

We and others previously reported that complement C5b-9 disrupts actin microfilaments in cultured GEC (30, 33). One possible mechanism for p38-mediated cytoprotection could be via modulation of the actin cytoskeleton. Thus we next studied the impact of FR-167653 on complement-induced actin depolymerization. When GEC-cPLA2 were incubated with antibody and complement for 3 h, F-actin content decreased by 9 ± 3%. FR-167653 (10 μM) significantly augmented the reduction to 18 ± 4% (P < 0.05, n = 5 each). Thus p38 activation, at least partly, prevents complement-induced actin depolymerization.

To further validate these results, we next generated subclones of GEC that inducibly express a constitutively active mutant of TAK1, a kinase upstream of p38 ( materials and methods ). When one such clone (Fig. 6A, #1) was stimulated with an insect hormone, ponasterone A, expression of TAK1 was induced after 2 h, peaking at 6 h. Phosphorylation of p38 was also observed and peaked at 6 h (Fig. 6A). Ponasterone A has no known impacts on mammalian cells. When this inducible clone was stimulated with ponasterone A for 6 h and exposed to antibody and increasing concentrations of complement (NS), complement-mediated cytotoxicity was attenuated, compared with controls (incubated with ethanol in the place of ponasterone A). Conversely, when cells were preincubated with the p38 inhibitor FR-167653 (10 μM, without stimulation with ponasterone A), cytotoxicity was augmented, compared with control, consistent with the previous results (Fig. 6B). Of interest, when cells were stimulated with ponasterone A in the presence of FR-167653, the impact of ponasterone A (TAK1 induction) was neutralized, but cytotoxicity did not reach the level of FR-167653 treatment alone [control 47 ± 1%, ponasterone A 33 ± 3% (P < 0.02 vs. control), ponasterone A plus FR-167653 43 ± 3%, FR-167653 alone 57 ± 2% (P < 0.05 vs. control) Fig. 6C]. These results indicate that the cytoprotective effect of TAK1 induction is, at least in part, mediated by p38 activation. However, there might be additional pathways downstream of TAK1, which contribute to cytoprotection.

Fig. 6.Induction of constitutively active transforming growth factor-β-activated kinase 1 (TAK1) ameliorates complement-induced GEC injury. A subclone of GEC that overexpresses TAK1 (constitutively active mutant) in an inducible manner was established as in materials and methods . A: time course of TAK1 induction. GEC were incubated with ponasterone A (pona 4 μM) for the indicated times and cell lysates were analyzed for expression of TAK1 and phosphorylation of p38 by immunoblotting using antibodies for HA (TAK1) and phospho-p38, respectively. In clone #1, induction of TAK1 peaked at 6 h when phosphorylation of p38 was also observed. In clone #3, induction of TAK1 was weak and phosphorylation of p38 was not detected. B: clone #1 was incubated with vehicle (control), ponasterone A (4 μM), or FR-167653 (10 μM) for 6 h and stimulated with anti-GEC antiserum and NS (or HIS). Complement-mediated cytotoxicity was quantified as in materials and methods using BCECF release into the medium. *P < 0.05, **P = 0.07, ***P = 0.06 vs. control n = 3 each. C: clone #1 was incubated with vehicle (control), ponasterone A (4 μM), ponasterone A (4 μM) + FR-167653 (10 μM), or FR-167653 (10 μM) for 6 h and stimulated with anti-GEC antiserum and NS (or HIS 3.5% vol/vol). Cytotoxicity was quantified as in materials and methods . +P < 0.02, ++P < 0.05 vs. control n = 3 each. (B and C: independent experiments.)

Inhibition of p38 activation augments proteinuria in PHN. The above results suggest that p38 activation is cytoprotective for GEC against complement-mediated cell injury in vitro. We next addressed whether p38 activation is also cytoprotective in vivo using the PHN rat model of membranous nephropathy. In PHN, it is known that complement C5b-9 causes GEC injury, which leads to proteinuria (21, 32). As shown in Fig. 2, p38 was significantly activated in glomeruli from rats with PHN, compared with control rats. We anticipated that if p38 is cytoprotective for GEC, inhibition of p38 would lead to augmented proteinuria in PHN. Thus we examined the impact of a specific p38 inhibitor, FR-167653, on proteinuria. First, to confirm the effect of FR-167653, rats were treated with FR-167653 from day 7 to 14 ( materials and methods ) and p38 activity in glomeruli was studied on day 14. In FR-167653-treated rats, p38 activity in glomeruli was markedly inhibited to a level comparable to that of control rats [control 13 ± 1, PHN 80 ± 22 (P < 0.05 vs. control), PHN-FR-167653 12 ± 5 (not significant from control), n = 3–6 Fig. 7A]. Phosphorylation of p38 was not affected by this inhibitor in a consistent manner, in agreement with a previous report (27) (Fig. 7A). In rats treated with vehicle, urinary protein excretion on day 14 was 161 ± 33 mg/day (n = 7), significantly higher than normal rats (∼10 mg/day). Consistent with the in vitro cytoprotective effect of p38 activation, rats treated with FR-167653 showed augmented proteinuria (288 ± 54 mg/day, P < 0.05 vs. vehicle, n = 9 Fig. 7B). To verify whether complement activation was influenced by FR-167653, we quantified C3 deposition in glomeruli (see materials and methods ). Glomerular C3 deposition was 91 ± 4 in rats with PHN and 93 ± 3 in rats with PHN treated with FR-167653. Thus FR-167653 did not affect complement activation in glomeruli. These results support a cytoprotective role for p38 in complement-mediated GEC injury in vivo.

Fig. 7.p38 inhibitor, FR-167653, augments proteinuria in PHN. PHN was induced by a single injection of anti-Fx1A antiserum, and rats were treated with vehicle or FR-167653 as in materials and methods . A: on day 14, glomerular p38 activity was measured as in materials and methods . Top: immunoblot. Bottom: densitometric analysis. *P < 0.05 vs. control n = 3–6 rats in each group. Data from densitometric analysis for control and PHN are shared with Fig. 2. B: on day 14, 24-h urine was collected in metabolic cages and urine protein was quantified. **P < 0.05 vs. vehicle n = 7 (vehicle), 9 (FR-167653).

HSP27 overexpression protects GEC from injury. The above results indicate cytoprotective effects of the p38 MAPK pathway in GEC. HSP27 is one of the molecules downstream of p38 MAPK. It is known that when the p38 MAPK pathway is activated, MAPKAPK-2 is phosphorylated/activated and in turn phosphorylates HSP27 (5). It was reported previously that in the rat puromycin aminonucleoside (PAN) nephrosis model, glomerular expression and phosphorylation of HSP27 were increased (24). Moreover, overexpression of HSP27 in mouse podocytes provided protection against PAN (25). We also demonstrated that glomerular expression of HSP27 is increased by ∼1.6-fold in rats with PHN, compared with control rats, similar to the increase seen in PAN nephrosis (24) (Fig. 8A). When treated with FR-167653, glomerular expression of HSP27 showed an upward trend, although the difference was not statistically significant [control 52 ± 2, PHN 84 ± 10 (P < 0.01 vs. control), FR-167653 117 ± 11 (P < 0.01 vs. control), n = 5–8 rats Fig. 8A]. We also studied phosphorylation of HSP27 using phospho-HSP27-specific antibody. Phosphorylation of HSP27 in glomeruli was significantly increased in PHN (Fig. 8B). In contrast to protein expression, phosphorylation of HSP27 was markedly inhibited by FR-167653 [control 15 ± 2, PHN 64 ± 11 (P < 0.01 vs. control), FR-167653 34 ± 11, n = 6 rats each Fig. 8B]. These results suggest that glomerular phosphorylation of HSP27 in PHN is, at least in part, mediated by p38 MAPK.

Fig. 8.Glomerular expression of heat shock protein (HSP27) is increased in rats with PHN. PHN was induced by a single injection of anti-Fx1A antiserum and rats were treated with vehicle or FR-167653 as in materials and methods . On day 14, glomerular lysates were prepared and analyzed by immunoblotting for HSP27 (A) and phospho-HSP27 (B). Top: immunoblot. Bottom: densitometric analysis. *P < 0.01 vs. control n = 5–8 rats each.

Thus we next studied whether the cytoprotective effect of the p38 MAPK pathway is mediated via MAPKAPK-2 and HSP27 in rat GEC in culture. To test this hypothesis, we first studied if MAPKAPK-2 is phosphorylated by complement in GEC. When GEC were exposed to antibody and complement, phosphorylation of p38 was observed for 20–60 min (Fig. 9A). Phosphorylation of MAPKAPK-2 was observed in a similar time course (Fig. 9A). We next studied the impact of overexpression of HSP27 on complement-mediated GEC injury. The plasmid encoding the wild-type HSP27 was stably transfected into GEC, and its expression was verified by immunoblotting. Two subclones (HSP27WT1 and 2) that overexpressed the wild-type HSP27 were selected for further study (Fig. 9B). Control 1 is a subclone of GEC that was transfected with vector alone and control 2 was transfected with the wild-type HSP27 but the expression level was minimal. It should be noted that control cells also expressed HSP27 when the film was exposed longer (Fig. 9B). When HSP27WT1 and 2 were simulated with antibody and complement, specific LDH release was markedly attenuated, compared with the two control cells [NS 2.5: control 1 30 ± 4%, control 2 24 ± 3%, HSP27WT1 6 ± 1% (P < 0.001 vs. both controls), HSP27WT2 7 ± 1% (P < 0.001 vs. both controls), NS 5: control 1 56 ± 8%, control 2 48 ± 5%, HSP27WT1 10 ± 1% (P < 0.001 vs. both controls), HSP27WT2 16 ± 1% (P < 0.001 vs. both controls), NS 10: control 1 73 ± 6%, control 2 66 ± 2%, HSP27WT1 25 ± 1% (P < 0.001 vs. both controls), HSP27WT2 30 ± 1% (P < 0.001 vs. both controls), n = 6–12 Fig. 9C]. To study if phosphorylation of HSP27 is important in cytoprotection, we next established subclones of GEC that overexpress a non-phosphorylatable mutant of HSP27. In this mutant, two Ser residues that are known to be phosphorylated by MAPKAPK-2 are mutated to Ala (10). Two subclones (HSP27mut1 and 2) were chosen for further studies (Fig. 9B). Control 3 is a subclone of GEC that was transfected with vector alone, and control 4 was transfected with HSP27mut, but the expression level was minimal. When HSP27mut1 and 2 were stimulated with complement, specific LDH release was slightly attenuated, compared with control cells, but the attenuation was much smaller than in cells overexpressing the wild-type HSP27 [NS 2.5: control 3 25 ± 4%, control 4 29 ± 2%, HSP27mut1 19 ± 2%, HSP27mut2 31 ± 5%, NS 5: control 3 60 ± 3%, control 4 48 ± 2%, HSP27mut1 41 ± 2% (P < 0.05 vs. control 3), HSP27mut2 48 ± 2%, NS 10: control 3 66 ± 2%, control 4 62 ± 1%, HSP27mut1 53 ± 1% (P < 0.05 vs. both controls), HSP27mut2 50 ± 3% (P < 0.05 vs. control 3), n = 4–12 Fig. 9D]. These results indicate that overexpression of HSP27 induces cytoprotection in GEC and suggest that MAPKAPK-2-mediated phosphorylation of HSP27 is important in this cytoprotection.

Fig. 9.Overexpression of the wild-type HSP27, but not a non-phosphorylatable mutant, protects GEC against complement-mediated cell injury. A: GEC-cPLA2 were stimulated with antibody and complement for the indicated times. Cells lysates were analyzed by immunoblotting for phosphorylation of MAPKAPK-2 and p38. B: subclones of GEC that stably overexpress the wild-type hamster HSP27 (HSP27WT) or a non-phosphorylatable mutant (HSP27mut) were established as in materials and methods . Expression of HSP27 was verified with immunoblotting. Note that control GEC also express HSP27 when exposed longer. C: GEC clones that do not overexpress HSP27 (cntl 1 and 2) and GEC clones that overexpress the wild-type HSP27 (HSP27WT1 and 2) were sensitized with anti-GEC antiserum and stimulated with increasing concentrations of NS (or HIS) for 40 min. Specific lactate dehydrogenase (LDH) release was quantified as in materials and methods as marker of complement-mediated cytotoxicity. *P < 0.001 vs. cntl 1 and 2 n = 6–12 per group. D: GEC clones that do not overexpress HSP27 (cntl 1 and 2) and GEC clones that overexpress a non-phosphorylatable mutant of HSP27 (HSP27mut1 and 2) were stimulated with complement and LDH release was quantified. +P < 0.05 vs. cntl 3. ++P < 0.05 vs. cntl 4 n = 4–12 per group.


This study elucidates the mechanisms of C5b-9-mediated activation of the ERK cascade. ERK activation proceeded via Ras and was dependent on cytoskeletal remodelling. Activated ERK phosphorylated two distinct substrates, and chronic constitutive activation of the ERK pathway exacerbated complement-dependent cytotoxicity. In keeping with earlier results, we demonstrate that in cultured GEC, C5b-9 induced ERK threonine 202 /tyrosine 204 phosphorylation (Fig. 1), which correlates with activation (10). Furthermore, we show that ERK is phosphorylated in glomeruli of rats with PHN (i.e., C5b-9-dependent GEC injury in vivo), and the localization of phospho-ERK is consistent with GEC (Figs. 1 and 2). In previous studies, we demonstrated that assembly of C5b-9 in the GEC plasma membrane (in culture and in vivo) induced transactivation of receptor tyrosine kinases, including EGF-R (13). EGF-R transactivation resulted in activation of phospholipase C-γ1 and PKC, as well as binding of the adaptor protein, Grb2, which links receptor tyrosine kinases to Ras (11, 13). To determine the role of Ras and PKC in complement-mediated ERK activation, we monitored ERK phosphorylation in GEC that had been stably transfected with a dominant-inhibitory mutant of Ras, and in GEC depleted of PKC. These experiments showed that the ERK pathway was activated predominantly via Ras (Fig. 3). Our results are distinct from those in K562 and COS-7 cells, where complement-induced ERK activation occurred via PKC (21).

The protocol employed in the present study did not result in quantitative changes in F-actin following incubation with complement, but in another study (44), GEC injury by complement was associated with loss of actin stress fibers and focal contacts. In vivo, GEC contain F-actin as a layer at the base of the foot processes, and the actin cytoskeleton is important in the maintenance of cell architecture. Previously, we showed that drugs that disassemble the actin cytoskeleton (cytochalasin D and latrunculin B) did not affect complement-mediated activation of EGF-R but inhibited activation of phospholipase C-γ1 and downstream signaling, including activation of PKC and cPLA2 (11). The present study demonstrates that complement-induced ERK activation was inhibited by cytochalasin D and latrunculin B (Fig. 7, A and B). Both drugs also inhibited ERK activation by EGF (which occurs via the Ras and Raf) and PKC (which occurs via Raf, but not Ras) (Figs. 3 and 7D). Thus an intact actin cytoskeleton is required for sequential activation of Raf, MEK, and ERK. Furthermore, cytochalasin D and latrunculin B also blocked ERK phosphorylation in GEC that stably overexpress R4F-MEK (Fig. 7F). This result implies that efficient phosphorylation of ERK by MEK is dependent on an intact actin cytoskeleton. The actions of cytochalasin D and latrunculin B on ERK phosphorylation were similar, although cytochalasin D and latrunculins induce actin cytoskeleton depolymerization through different mechanisms. Cytochalasin D binds to the barbed (growing end) of actin filaments and prevents actin filament formation or leads to disruption of actively turning over actin stress fibers. Latrunculins sequester G-actin monomers preventing actin polymerization and effectively disrupt both actin stress fibers, as well as cortical actin filaments, which are more resistant to cytochalasin D (23).

Similar to the actin-depolymerizing drugs, condensation and stabilization of actin filaments at the cell periphery near the plasma membrane by calyculin A inhibited the complement-induced increase in ERK phosphorylation (Fig. 7C). Stable expression of L 63 RhoA also attenuated ERK phosphorylation by complement (Fig. 9), and this effect is associated with enhanced stress fiber formation (11). L 63 RhoA may have acted by a mechanism analogous to calyculin A, as constitutively active RhoA was reported to induce phosphorylation of ezrin-radixin-moesin proteins (4). Similar to RhoA activation, inhibition of Rho-associated kinase (a downstream effector of RhoA) reduced ERK activation (Fig. 9). Together, the results suggest that both disassembly and stabilization of the actin cytoskeleton can reduce ERK phosphorylation, implying that ERK activation is dependent on cytoskeletal remodelling. Finally, the cytoskeleton-altering drugs had disparate effects on the activation of the JNK pathway in GEC (Fig. 8). Thus there are major differences in the role of the cytoskeleton in facilitating activation of pathways by complement, and pharmacological disassembly of the actin filament network does not exert a general inhibitory effect on all signaling pathways.

Our study, which revealed an important role for the actin cytoskeleton in complement signaling, is in keeping with studies in other systems. For example, in response to insulin treatment, actin filament disassembly blocked activation of ERK and p38 kinase, but not insulin receptor autophosphorylation, phoshatidylinositol 3-kinase, or S6 kinase (42, 45). Moreover, binding of Shc to the insulin receptor was not affected, but binding of Grb2 to Shc was disrupted (45). However, cytochalasin D did not affect ERK activation by lysophosphatidic acid, bombesin, or platelet-derived growth factor (39). In addition, the actin cytoskeleton may be important in cell cycle progression, transcription of serum-inducible genes, and induction of nitric oxide synthase (51). It may seem unusual that calyculin A and L 63 RhoA expression (which facilitate actin polymerization) should produce effects similar to those of cytochalasin D or latrunculin B (which depolymerize the cytoskeleton) and inhibition of Rho-associated kinase. All of these treatments were inhibitory to ERK activation in the present study, and parallel effects of increasing actin polymerization and depolymerization have been reported in other systems. For example, F-actin polymerization with the drug jasplakinolide and depolymerization with latrunculin B or cytochalasin D both inhibited insulin-stimulated glucose uptake in adipocytes (20), lipopolysaccharide-mediated production of reactive oxygen species in monocytes (36), accumulation of phosphatidylinositol 3,4,5-trisphosphate in response to a chemotactic stimulus in neutrophils (48), and apoptosis in airway epithelial cells (49).

To address the functional role of complement-mediated ERK activation in GEC, we examined substrates of ERK, which may point to potential involvement in cellular functions (31). Complement did not induce serine 383 phosphorylation of the transcription factor Elk-1, a potential nuclear target of ERK (Fig. 4). As expected, EGF, a well-known activator of ERK, did actually induce Elk-1 phosphorylation. Perhaps EGF was able to activate/recruit additional factors that are required for ERK to phosphorylate nuclear transcription factors, while complement was not effective. However, complement induced phosphorylation of MAPKAPK-2, which was, in part, dependent on the ERK pathway (as well as p38 kinase Fig. 6). MAPKAPK-2 is more typically a substrate of p38 kinase, but by analogy to GEC, ERK-dependent activation of MAPKAPK-2 has been reported in neutrophils (7). Under basal conditions, MAPKAPK-2 is believed to be located in the nucleus, whereas on activation, MAPKAPK-2 can phosphorylate nuclear transcription factors, or be exported into the cytoplasm, where it can activate other substrates, including Hsp27 (27). Our result supports a role for complement-mediated ERK activation in transcription, as well as in cytoplasmic pathways. Moreover, complement stimulated phosphorylation of cPLA2, a cytoplasmic ERK target on serine 505 (Fig. 5), a well-defined ERK phosphorylation site (18, 22). Interestingly, although cPLA2 is a substrate for ERK in GEC, the complement-induced increase in cPLA2 catalytic activity in these cells did not require ERK activation (10) but was dependent on PKC. Phosphorylation of the cPLA2 serine 505 site contributes to EGF-mediated release of arachidonic acid in GEC (10) and correlates with agonist-stimulated cPLA2 activity in other cells (22). Further study will be required to elucidate additional ERK substrates.

In earlier studies, we demonstrated that complement-induced activation of JNK and p38 mitogen-activated protein kinases was cytoprotective in GEC, i.e., that JNK and p38 limited or restricted complement cytotoxicity (1, 32). The mechanism of protection by p38 was, at least in part, dependent on Hsp27, a substrate of MAPKAPK-2 (1). This study suggests that the role of ERK in GEC injury may be more complex. Inhibition of the complement-induced activation of the ERK cascade in GEC did not affect complement cytotoxicity significantly (Fig. 11). However, a cytoprotective role for ERK via MAPKAPK-2 and Hsp27 cannot be excluded because MAPKAP-2 activation was only partially dependent on ERK (Fig. 6). Chronic (but not brief) incubation with complement in the presence of cytochalasin D or latrunculin B increased complement lysis (Fig. 10), but the combination of ERK inhibition and cytoskeletal disassembly did not increase complement lysis beyond that observed with cytoskeletal disassembly alone (data not shown). In contrast, complement cytotoxicity was enhanced in GEC that stably express constitutively active MEK (Fig. 11). The kinetics of ERK activation are likely to be different between the two conditions. Previously, we demonstrated that during chronic incubation with complement, ERK phosphorylation was enhanced within 20 min and returned toward basal levels by ∼5 h (13). In contrast, in the GEC that stably express R4F-MEK, ERK remains activated continuously (25). Thus complement-dependent cytotoxicity was exacerbated by chronic, constitutive ERK activation. In another study, ERK inhibition in K562 and COS-7 cells enhanced complement lysis (measured as trypan blue uptake) (21). A recent study in GEC (33) showed that complement induced DNA damage and increased expression of the cell cycle-regulatory genes, p21 and GADD45 (growth-arrest DNA damage-45). In this study, inhibition of MEK exacerbated DNA damage and reduced levels of p21 and GADD45. The authors concluded that ERK activation protects GEC from DNA damage via regulation of p21 and GADD45 genes. However, although these authors showed that complement did not induce apoptosis, cytolytic injury (e.g., LDH release) was not examined. Furthermore, the same investigators demonstrated earlier that C5b-9 can induce DNA synthesis (although not proliferation) in GEC (40). Thus the effect of C5b-9 and ERK on DNA regulation and cell injury will require additional study. With the advent of drugs that modulate protein kinase pathways, a better understanding of the activation of protein kinases by C5b-9 will provide insights into novel targets for therapy of glomerular diseases.

Watch the video: Πρωτεΐνες (May 2022).