Protective effect of 14-3-3 antibodies on stressed neuroretinal cells via the mitochondrial apoptosis pathway

Reagents

The used reagents were purchased from Sigma Aldrich, St. Louis, MO, US if not stated
otherwise. L-alanyl-L-glutamine was purchased from Biochrom AG (Berlin, Germany).
The rabbit anti-14-3-3 abs for cell incubation and the secondary antibody goat-anti-rabbit
IgG-HL (FITC) were purchased from Abcam (Cambridge, UK), H
₂
O
₂
and paraformaldehyde from Carl Roth GmbH (Karlsruhe, Germany), Staurosporine from
Calbiochem (San Diego, CA), 70 % ethanol, formic acid, trifluoroacetic acid (TFA)
and acetonitrile (ACN) from Merck (Darmstadt, Germany), 14-3-3 antibodys for immunohistochemical
staining from Lifespan (Seattle, WA), the BCA Pierce Protein Assay kit from Fisher
scientific (Waltham, MA), Trypsin from Promega (Mannheim, Germany), wheat germ agglutinin
conjugate with tetramethylrhodamin (WGA) from Invitrogen (Karlsruhe, Germany), vectashield
mounting medium with 4’,6-diamidino-2-phenylindole (DAPI) from Vector Laboratories
(California, USA) and HPLC H
₂
O from Applichem (Darmstadt, Germany).

Serum samples

The serum was collected from patients suffering from POAG according to the classification
of the guidelines of the European glaucoma society as well as from healthy volunteers
after given written informed consent. Furthermore, patients suffering from autoimmune
diseases or other neurodegenerative diseases were excluded from the study. The studies
were performed in accordance with the Declaration of Helsinki on medical research
involving human subjects. The samples were age matched. Ethics approval: No: 837.219.07
(5754); Ethics committee of the Landesärztekammer Rhineland-Palatinate.

Cell culture

RGC-5 cells, a neuroretinal cell line of mouse origin, provided by Dr. Neeraj Agarwal,
were used as model for retinal cells of neuronal origin 27]. They were grown in 75 culture flasks in Dulbecco’s modified eagle medium supplemented
with 10 % fetal calf serum (FCS), 100 U/ml penicillin, 100 U/ml streptomycin and 4 % L-alanyl-L-glutamine.
The cells were cultivated in a humidified incubator at 37 °C with 95 % air and 5 %
CO
₂
and were passaged when they reached a confluence of 80 %.

Cell treatment with different serum types

250 000 RGC-5 cells were transferred to culture dishes with an 82.7 mm inner diameter.
After 24 h the cells were incubated with medium containing 5 % FCS and either 5 %
serum from healthy people or 5 % serum from POAG patients (n in each experimental
group: 8). Each individual serum was used for an individual run. The cells were incubated
with the different serum types for 24 h. After 24 h the medium was discarded and the
cells were washed twice with phosphate buffered saline (PBS). Cell lysis was performed
and the detached cells were transferred to an Eppendorf tube with lysis buffer [Urea
9.5 M, Chaps 2 %, DTT 1 %?+?proteinase inhibitor mix (P 1860)]. Furthermore an ultrasonic
pulse echo instrument [Labsonic®M (Sartorius)] was used to perform additional cell
lysis. Protein concentration of the samples was determined with the method of Lowry
28]. After the protein concentration was determined, a sample-pool for each experimental
group was created with a total amount of 80 ?g protein. The pooled samples were separated
with a 1D SDS gel. Each lane was divided into 15 pieces and digested with trypsin
in order to measure the peptide profile with Maldi- Orbitrap MSMS.

Protein profiling with Maldi- Orbitrap MSMS

The protein profiles were analysed with Maldi- LTQ Orbitrap XL using Maldi-steel targets.
The samples were dried in a concentrator and acidified with 0.1 % TFA. C-18 ZipTips
(Millipore, Billerica, MA) were used to purify the samples and the peptides were eluted
directly onto a Maldi Target with 40 % ACN and 60 % ACN. Measurements were performed
according to the manufacturers’ protocol.

Protein profiling with capillary LC-ESI-MSMS

The protein profiles were analysed with capillary LC-ESI-MSMS using a C-18 pre-column
(30 mm x 0.5 mm) and a C18 analytical column (150 mm x 0.5 mm, both Thermo Scientific,).
A Rheos Allegro HPLC Pump (Thermo Scientific) was the solvent delivery system. The
pump flow rate was 200 ?l/min, and reduced to a column flow of 10 ?l/min (M-472 graduated
microsplit valve (Upchurch, Scientific, USA). With two running buffers (A (98 % H
₂
O, 1.94 % ACN, 0.06 % methanol, 0.05 % TFA)?+?B (95 % ACN, 3 % methanol, 2 % H
₂
O, 0.05 % TFA) a linear gradient of 80 min was performed (0–47 min: 0–100 % B, 47–49 min:
100 % B, 49–58: 100 %–0 % B, 58–80 min: 0 % B). Mass spectra were obtained using an
LTQ OrbitrapXL.

Cell treatment with 14-3-3 antibodies and different stress factors

RGC-5 cells were seeded in 24 well plates. Depending on the stress factor and therefore
the overall incubation time, the cells were either seeded with 45000 cells per well
(for the experiments with H
₂
O
₂
and staurosporine) or 40000 cells per well (experiments with the stress factor glutamate).
The cells were preincubated with different concentrations of chicken polyclonal anti
14-3-3 sigma antibodies (0.005, 0.1, 0.5, 1, 5 and 10 ?g/ml) for 3 h. As an additional
control group the cells were incubated with a non-retina specific antibody against
myoglobin. To induce apoptosis the cells were stressed with staurosporine (1.5 ?M
for 5 h) or glutamate (20 mM for 24 h). Furthermore oxidative stress was induced by
incubating the cells with 50 ?M H
₂
O
₂
for 1 h (n in each experimental group?=?4). We used this amount of H
₂
O
₂
because test showed that we were able to detect a rise in ROS without a loss of viability
of the cells using this concentration. Subsequently cell viability tests and ROS measurements
were performed. Figure 1 shows an overview of the experimental setup.

Fig. 1. Experimental setup for RGC-5 cells incubated with 14-3-3 ab. The cells were seeded
in 24 well plates and were let to rest for 24 h. Then antibody preincubation of the
cells with different antibody concentrations was performed for 3 h. The control cells
not incubated with the antibody were incubated with normal medium for the same amount
of time. After 3 h the medium containing the antibodies or the control medium was
replaced with medium containing one of the stress factors (glutamate, staurosporine
or H
₂
O
₂
). Depending on the stress factor, the incubation time varied. After stressing the
cells, viability tests with crystal violet staining was performed with all cells,
and ROS level measurements were performed in the cells stressed with H
₂
O
₂

Cell viability test

Cell viability was assessed with crystal violet staining. The cells were fixed with
3 % paraformaldehyde (15 min) and rinsed with PBS. Subsequently the cells were stained
with 0.1 % crystal violet solution for 20 min. Excess stain was removed by washing
the plates with distilled water three times. The bound stain of the viable cells was
resolved in 70 % ethanol for 3 h. The supernatants were read with the Multiscan ascent
plate reader (Thermo scientific) at 570 nm. The absorption was expressed as a percentage
of the control cells, which were only treated with the stress factor. An unpaired
student t-test was used to compare the data obtained and was realized with Statistica (Statsoft,
Tulsa, Oklahoma, USA). A p value??0.05 was significant and a p value??0.01 was declared
highly significant.

ROS-test

To quantify ROS we used 2’,7’-dichlorodihydrofluorescein-diacetate (DCFH-DA). Intracellular
esterases’ and ROS convert the non-fluorescence stain 2’, 7’ dichlorodihydrofluorescein
(DCFH) to the fluorescent stain dichlorofluorescein (DCF). Cells were loaded with
10 mM DCFH-DA in the incubation chamber for 15 min. Then the culture medium was replaced,
to remove the unbound DCFH-DA. To generate ROS, 50 ?M H
₂
O
₂
was added. The fluorescence was measured by using the microplate reader fluoroscan
ascent (Thermo scientific) with excitation/emission wavelengths of 485/538 nm. The
absorption was expressed as a percentage of the control cells, which were only treated
with 50 ?M H
₂
O
₂
. The ROS- level was normalised by measuring the viability of the cells in the same
well. An unpaired student t-test was used to compare the data obtained and was realized with Statistica.

Immunocytochemical staining

RGC-5 cells were grown in ?-slide IV (Ibidi GmbH, Munich, Germany) and subsequently
washed with PBS. Then the cells were fixed with 3 % paraformaldehyde for 15 min and
incubated with 0.25 % Triton-X-100 for 12 min. After 3 wash steps with PBS, the cells
were treated with 1 % bovine serum albumin for 20 min. Afterwards, the cells were
incubated with 2 ?g/ml rabbit polyclonal anti 14-3-3 sigma antibodies overnight, then
gently washed 3 times with PBS and incubated with Goat polyclonal secondary antibody
to rabbit IgG-HL conjugated with FITC for 1.5 h. After 3 washing steps with PBS the
cells were visualized with a fluorescence microscope (Leica Microsystem, Heidelberg,
Germany). To investigate the antibody uptake in living cells, the cells were preincubated
with 10 ?g/ml rabbit polyclonal anti 14-3-3 sigma abs for 3 h and then washed with
PBS to remove unbound antibodies. Controls were preincubated with medium not containing
the polyclonal anti 14-3-3 sigma abs. The cells then were treated as described above
and visualized with a Leica fluorescence microscope and using Lucia G/F software.

Cell lysate preparation

For proteomic analysis the cells incubated with anti-14-3-3 antibody were grown in
60 x 15 mm cell culture dishes and incubated with 0.5 ?g/ml chicken polyclonal anti
14-3-3 sigma antibodies. Cells were detached using cell dissociation solution (CDS)
and lysed by freezing at ?80 °C after adding 0.1 % Dodecyl-D-?- Maltosid with proteinase
inhibitor. Additionally, the cells were treated with a sonication bath for 1 min.
After centrifugation, the supernatant was used for determining the protein concentration
by BCA Pierce Protein Assay kit.

SDS PAGE separation and in-gel digestion

Protein separation was performed with a denaturing gel electrophoresis. Each lane
was cut into 17 pieces, incubated with acetonitrile (ACN) and ammonium bicarbonate
(AB) and dried in a concentrator. The pieces were digested with trypsin (0.7 ?g Trypsin
in 80 % HPLC H
₂
O, 10 % ACN, 10 % AB) over night. The supernatant was collected and the remaining
proteins were dissolved with an extraction buffer (38 % HPLC H
₂
O, 2 % formic acid, 60 % ACN) for 30 min. The supernatants then were pooled, dried
in a concentrator and acidified with 0.1 % trifluoroacetic acid. C-18 ZipTips were
used to purify the samples. The samples were measured with capillary LC-ESI-MSMS as
described above.

Data processing

The obtained mass spectra measured with Maldi- Orbitrap MSMS were used for an identification
and quantification of the proteins. Using Mascot search engine, the spectra were transferred
to SwissProt database. The identification of the proteins was performed using Mus
musculus as taxonomy and trypsin as digesting enzyme. This information is necessary
for database to calculate the theoretical mass. Furthermore one missed cleavage was
allowed. As MALDI was used, the charge state was set to 1+. The error window of the
mass was set at 100 ppm and 0.8 Da. The normalisation and quantification of the peptides
was performed with PSP (former P2M), our in-house proteomics pipeline software and
transferred to Statistica software for quantification, as described before 29].

The obtained mass spectra measured with LC-ESI-MSMS were used for identification and
quantification with Maxquant (Max Planck Institute of Biochemistry, Martinsried, Germany).
The tolerance in mass precision for MS/MS was 20 ppm and 0.5 Da. The protein and peptide
false discovery rate were set to 0.01 and the minimum peptide length was 6 amino acids.
The evaluation was implemented with Ingenuity Pathway Analysis (IPA) Software. Only
proteins with a 2-fold changed expression upon 14-3-3 sigma antibody treatment were
included in the analysis. The statistical significance of each pathways was calculated
by IPA using a Fisher Exact test p??0.05.