Lonhienne et al. 10.1073/pnas.1001085107
SI Materials and Methods
GST was expressed from expression vector pGEX-KG (1) and
purified by affinity chromatography on a glutathione-agarose col-
umn according to the manufacturer’s instruction (Scientifix).
Mouse monoclonal anti-GFP IgG (Roche; catalog no. 11814460-
001) was used for dot-blots, Western blots, protein degradation,
and protein cross-linking experiments, as an immunochemically
detectable molecule for uptake. Alexa Fluor 680 goat anti-mouse
antibody (Molecular Probes) was used as secondary antibody. DQ
Green BSA was obtained from Invitrogen.
DQ Green BSA Proteolysis Experiment.CellswereincubatedwithDQ
and washed once with IB. The pellet was then resuspended in IB
containing IgG-Cy3 (0.02 μg/μL), and cells were incubated for
a further hour. After washing once and resuspending in IB, cells
were examined by confocal laser scanning microscopy (CLSM).
For control experiments (see Insets), DQBSA (0.02 μg/μL) in
IB was incubated with 0.002 μg/μL of trypsin (positive control) or
without trypsin (negative control) for 1 h, and solutions were ex-
amined by CLSM with the same settings as used for monitoring
DQBSA in Gemmata obscuriglobus cells.
Fractionation Experiments. Cells were collected from two to three
M1-agar plates, washed once, and resuspended in 500 μL of bt-
DMSO buffer [10 mM bis-Tris (pH 7.5), 0.1 mM MgCl2, and 20%
DMSO] supplemented with 10 μL of protease inhibitor mix
(Protease Inhibitor Mixture Set 3; Merck), 10 μg of DNase, and
for 10 min. Supernatant was centrifuged at 100,000 × g in a Beck-
man Coulter tabletop ultracentrifuge (Optima TLX). The pellet
step sucrose gradient (Fig. S9), and centrifuged in an SW60 rotor
on a Beckman Coulter L8-60M ultracentrifuge at 215,000 × g for
fraction were used for dot-blot analysis for presence of GFP.
and ultracentrifuge at 100,000 × g, and the pellet resuspended in
gradient. After centrifugation using the same rotor and centrifuge
for 16 h at 215,000 × g, fractions were collected from the top
(≈400 μL each). GFP-positive fraction 10 from the 20–60% su-
sucrose/bt-DMSO gradient centrifugation using the same con-
ditions. This fractionation was performed in a similar way as the
previous step. A schematic diagram of the fractionation experi-
ments is presented in Fig. S9A. Quality of fractionation experi-
ments was monitored by transmission electron microscopy using
uranyl acetate-stained samples (Fig. S11).
Saturation Experiment. The experiment was carried out in three
replicates using the same batch of G. obscuriglobus cells. Cells
were grown on plates and prepared as described for experiments
testing GFP uptake in the main text (Materials and Methods).
G. obscuriglobus cells were resuspended in 850 μL of IB and ali-
quoted into eight Eppendorf tubes, each containing 100 μL of cell
suspension. IB (1.9 mL) containing different concentrations of
GFP was added to cell suspensions to reach a final concentration
of GFP ranging from 2 to 40 μg/mL (Fig. S2). Samples were con-
tinuously mixed on a vertical rotating wheel. After 2 h of in-
cubation at 28 °C, samples were immediately cooled on ice and
careful removal of the supernatant, cells were resuspended in 200
μL of ice-cold IB and centrifuged again. The supernatant was re-
moved, and cells were resuspended in 100 μL of ice-cold IB. The
and GFP fluorescence was measured using a fluorescence plate
reader, POLARStar OPTIMA (Imgen Technologies) (excitation
filter set to A-405, emission filter set to 520).
Competition Experiment. The experiment was carried out in three
replicates using the same batch of G. obscuriglobus cells. Cells
grown on plates were resuspended in 350 μL of IB and aliquoted
in three Eppendorf tubes, each containing 100 μL of cell suspen-
sion. One of the aliquots was not incubated with GFP and served
to monitor the fluorescence background. GFP (10 μg/mL) alone,
GFP (10 μg/mL) and ovalbumin (100 μg/mL), or GFP (10 μg/mL)
of incubation at 28 °C, samples were immediately cooled on ice.
Thesampleswere thencentrifuged andtreated in thesamewayas
for the saturation experiment. The reported fluorescence data
were obtained by subtracting background values corresponding to
autofluorescence of cells.
Cells. Cells grown on plates were resuspended in 350 μL of IB
containing 1 μg/mL of mouse IgG. After 1 h incubation at 28 °C,
cells were washed three times with 1 mL of IB buffer and re-
suspended in 350 μL of IB buffer. Aliquots of 100 μL were sam-
pled immediately after resuspension and after 2 h and 16 h in-
cubation at 28 °C. Incubation was terminated by cooling on ice.
Cells were centrifuged and resuspended in 50 μL of SDS/PAGE
loading buffer, and the lysate was loaded onto a polyacrylamide
gel. Proteins contained in 10 μL of each sample were resolved by
SDS/PAGE (4–20% gradient gel) and characterized by Western
blot analysis using Alexa Fluor 680 goat anti-mouse antibody
(Molecular Probes). Detection was performed by using an Od-
yssey infrared imaging system (Li-COR).
Cross-Linking of Mouse IgG (Anti-GFP Antibody) and Testing Uptake
of Cross-Linked Ig. Mouse IgG (20 μg) was incubated in 100 μL of
10 mM Tris-HCl, pH 7.5, containing 1% formaldehyde for 1 h at
room temperature. Formaldehyde solution was then substituted
with 10 mM Tris-HCl, pH 7.5, using Vivaspin 500 concentrators
(3 kDa exclusion limit; GE Healthcare). Cells grown on plates
IB containing ≈5 μg/mL of cross-linked mouse IgG. After 1 h of
incubation at 28 °C, cells were washed three times with 0.5 mL of
IB buffer and resuspended in 50 μL of SDS/PAGE loading buffer.
The samples were resuspended in sample buffer without mer-
captoethanol and loaded onto the SDS/PAGE gel, without heat-
resolved by SDS/PAGE (4–20% gradient gel) and characterized
by Western blot analysis using Alexa Fluor 680 goat anti-mouse
antibody (Molecular Probes). Detection was performed by using
an Odyssey infrared imaging system (Li-COR).
Mass Spectrometry. Anti-GFP IgG treated with or without formal-
dehyde was digested overnight with trypsin (1:25) in 25 mM ammo-
nium bicarbonate. Samples were desalted and concentrated using
a C18 ZipTip (Millipore) and spotted to a MALDI target plate with
Lonhienne et al. www.pnas.org/cgi/content/short/1001085107 1 of 4
0.5 μL alpha-cyano-4-hydroxycinnamic acid (CHCA) matrix (10
mg/mL CHCA/25 mM diammonium citrate/50% acetonitrile/0.1%
TFA). Spectra were analyzed on a Voyager MALDI-TOF in-
strument in reflectron mode, across the mass range 700–4,000 m/z
using a minimum of 500 shots per spectra. Data were processed
Quantification of gold particles was carried out using iTEM
software from Olympus Soft Imaging.
1. Hakes DJ, Dixon JE (1992) New vectors for high level expression of recombinant
proteins in bacteria. Anal Biochem 202:293–298.
treated with 1 mM sodium azide 15 min before addition of GFP and incubation for 1 h at 28 °C. Top and bottom frames correspond to GFP fluorescence and
bright field, respectively. (B) G. obscuriglobus cells were incubated with GFP (Left), preincubated with sodium azide for 15 min before addition of GFP (Middle),
or preincubated with sodium azide before addition of GFP and ATP (Right). Cells were incubated for 1 h at 28 °C before CLSM. Top and bottom frames
correspond to GFP fluorescence and bright field, respectively. (Scale bars, 50 μm.)
The uptake of GFP by G. obscuriglobus cells is energy dependent. (A) G. obscuriglobus cells incubated with GFP for 1 h at 0 °C, 28 °C, or 37 °C, or cells
trations of GFP. Cells were then washed and analyzed for fluorescence emission by microplate reader assay. Error bars represent the SD from three experi-
Saturation experiment demonstrating concentration limit for protein uptake. G. obscuriglobus cells were incubated for 1 h with different concen-
incubated with mixtures of both Cy3-labeled Ig and GFP at ratios 1:20, 20:1, and 1:1 (1 unit corresponds to 10 μg protein/mL). (Scale bar, 5 μm.)
Competition experiment using microscopy detection demonstrating that GFP competes with other proteins for uptake. G. obscuriglobus cells were
cells were incubated with GFP (10 μg/mL) only, with both GFP (10 μg/mL) and ovalbumin (100 μg/mL), or with both GFP (10 μg/mL) and GST (100 μg/mL) for 90 min.
Cells were then washed and analyzed for fluorescence emission by microplate reader assay. Error bars represent the SD from three experimental replicates.
Competition experiment using plate reader fluorescence detection demonstrating that GFP competes with other proteins for uptake. G. obscuriglobus
stained the cells. (Scale bar, 10 μm.)
G. obscuriglobus cells do not uptake DNA. Cells were incubated with Cy3-labeled DNA (Left) or Cy3 dye alone (Right) before CLSM; only Cy3 dye alone
Lonhienne et al. www.pnas.org/cgi/content/short/10010851072 of 4
cubated with DQBSA for 3 h at 28 °C, washed, and incubated for another hour with IgG-Cy3 (to indicate position of the paryphoplasm by the IgG-Cy3 taken up
by cells). Frames correspond to DQBSA, IgG-Cy3, and merged fluorescence. The merged image shows colocalization of fluorescent DQBSA degradation product
and IgG-Cy3. Insets correspond to controls for DQBSA proteolysis in absence of cells: Inset 1 illustrates absence of fluorescence for DQBSA alone; Inset 2 il-
lustrates fluorescence of DQBSA digested by trypsin. (Scale bar, 2 μm.) (B) Mouse IgG (anti-GFP antibody) is proteolyzed in G. obscuriglobus cells. Cells were
preincubated with mouse IgG for 1 h at 28 °C, washed, and further incubated at 28 °C. After the indicated incubation times, cells were washed, and the lysates
were analyzed by SDS/PAGE under nonreducing conditions to better visualize degradation products and via Western blot using an anti-mouse antibody. Lane
1, mouse IgG (control); lane 2, 0 h of incubation; lane 3, 2 h of incubation; lane 4, 16 h of incubation. The band (arrow) representing IgG heavy chain (the
sample was run in nonreducing conditions) is markedly decreased after 2 h and totally disappears after 16 h.
Degradation of proteins in paryphoplasm of G. obscuriglobus cells. (A) DQBSA is proteolyzed in G. obscuriglobus cells. Before CLSM, cells were in-
cerevisiae. Protein profiles were aligned with the hhalign program available from the HHPred package (1). Secondary structures (α-helices and β-strands) were
predicted with the PSI-PRED program (2) and are represented as pink cylinders (α-helices) and cyan arrows (β-strands) above and below the aligned sequences,
respectively. Where regions containing either β-sheets or α-helices conform to domains corresponding to β-propellor or α-solenoid domains, the predicted (for
gp4978) and known (for clathrin) domains of these β-propellers and α-solenoids are highlighted by a gray shading of the secondary structures.
Alignment and secondary structure comparison of gp4978 and yeast clathrin heavy chain (Uniprot ID: P22137- CLH_YEAST) from Saccharomyces
G. obscuriglobus cells were sonicated, and the insoluble material was separated by differential centrifugation (step 1). The insoluble material from step 1 was
further fractionated via sucrose density gradient centrifugations (steps 2, 3, and 4). Dot-blots from successive fractionations were performed with mouse IgG to
detect GFP-positive fractions. A Western blot with mouse IgG was carried out to confirm the dot-blot results of the last fractionation (step 4). As an example
a Western blot with fractions 2 and 9 is shown; M, molecular mass marker. Confirming dot-blot results, GFP was present in fraction 9 and largely absent in
fraction 2. (B) Dot-blot and Western blot of fractions from the 40–70% sucrose density gradient centrifugation run corresponding to step 4, using anti-gp4978
antibody. Western blot result shows a reacting band of ≈125 kDa corresponding to the expected length of the MC-like protein gp4978.
GFP cofractionates with membrane coat (MC)-like protein gp4978. (A) Diagram of fractionation of G. obscuriglobus cells preincubated with GFP.
GFP or Cy3-labeled Ig and GFP, at equimolar concentrations (10 μg protein/mL). Cy5-labeled streptavidin and GFP (Upper) or Cy3-labeled Ig and GFP (Lower)
were coincubated with G. obscuriglobus cells for 1 h at 28 °C. Streptavidin-Cy5 and IgG-Cy3 indicate the signals for Cy5 and Cy3 only, GFP indicates signal for
GFP fluorescence, and “merge” indicates merger of both Cy5 and GFP or Cy3 and GFP signals, respectively; the merge images show colocalization of the
corresponding proteins. (Scale bar, 2 μm.)
Internalized proteins colocalize in G. obscuriglobus cells. CLSM of G. obscuriglobus cells incubated with mixtures of both Cy5-labeled streptavidin and
1. Söding J (2005) Protein homology detection by HMM-HMM comparison. Bioinformatics 21:951–960.
2. Jones DT (1999) Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol 292:195–202.
Lonhienne et al. www.pnas.org/cgi/content/short/10010851073 of 4
G. obscuriglobus cells. Histogram of the average density of anti-GFP (A) and anti-gp4978 (B) gold labeling (gold particles/μm2) in the paryphoplasm and in the
nonparyphoplasm remaining region (“rest of cell”) of the G. obscuriglobus cells. Error bars shown are 95% confidence intervals of the mean.
Distribution of gold particles detecting GFP and gp4978 MC-like protein in the paryphoplasm and the nonparyphoplasm regions of the
step gradient (negative for reaction with anti-GFP antibody via Western blot) contains mostly flagella as well as membranous debris; fraction 9 (Middle) from
the 20–60% continuous gradient (negative for reaction with anti-GFP antibody via Western blot) contains mainly membrane sheets covered with characteristic
pore-like structures; fraction 10 (Bottom) from the 20–60% continuous gradient (positive for reaction with anti-GFP antibody via Western blot) contains mostly
membrane vesicles and small pieces of broken membranes. When this fraction 10 was further subjected to another sucrose density gradient fractionation (30–
70%) it resulted in only one visible band, corresponding to fraction 9 positive for anti-GFP antibody reactivity. (Scale bars, 50 nm.)
Transmission electron micrographs of negatively stained preparations of fractions from density gradient centrifugations. Fraction 7 (Top) from the
Cells were incubated with mouse IgG for 1 h. After washing, cell lysates were electrophoretically separated on an SDS gel (Left), then blotted and probed with
an anti-mouse antibody (Right). Lane 1, non–cross-linked mouse IgG (anti-GFP antibody); lane 2, cross-linked mouse IgG; lane 3, E. coli DH5α cells preincubated
with cross-linked mouse IgG; lane 4, G. obscuriglobus cells preincubated with cross-linked mouse IgG; lane 5, G. obscuriglobus cells not incubated with cross-
linked mouse IgG. Note that the whole cross-linked IgG molecule of 150 kDa seen in the lane 2 control is also present in lane 4, showing the protein in-
corporated by G. obscuriglobus cells. (B) Colocalization of cross-linked mouse IgG with MC-like protein gp4978 in the same step gradient membrane fraction
derived from lysed G. obscuriglobus cells that had been incubated with mouse IgG (Upper). Dot-blot of fractions of a sucrose step gradient centrifugation using
anti-gp4978 antibody (anti-MC-like protein) and anti-mouse antibody (to detect mouse IgG) (Lower). Below is a Western blot of fraction 3 from the sucrose
step gradient centrifugation run seen in the upper panel, using anti-mouse antibody, showing reacting bands corresponding to cross-linked mouse IgG. Lane A,
mouse IgG; lane B, cross-linked mouse IgG; lane C, fraction 3. (C) Confirmation of successful cross-linking of IgG by mass spectrometry. MALDI-TOF spectra of
IgG treated with or without formaldehyde and digested with trypsin. Upper: Addition of 12 Da to mass to mass at 1649 m/z to give 1,661 m/z, indicating the
addition of formaldehyde to the protein. Lower: Presence of a new peak (arrow) in the presence of formaldehyde at 2,313 m/z, due to the cross-linking of two
or more peptides by formaldehyde.
Uptake of cross-linked mouse IgG (anti-GFP antibody) by G.obscuriglobus cells. (A) Western blot analysis showing uptake of cross-linked mouse IgG.
Lonhienne et al. www.pnas.org/cgi/content/short/10010851074 of 4