Endocytosis and intracellular dissociation rates of human insulin-insulin receptor complexes by quantum dots in living cells

Article abstract of DOI:10.1021/bc300526d

Insulin signaling is involved in glucose metabolism, cellular growth, and differentiation. Its function is altered in diabetes and many cancer types. Insulin binding to insulin receptor (IR) triggers diverse signaling pathways. However, signal transduction by IR is not mediated exclusively at the cell surface. Activated ligand-receptor complexes are internalized into endosomes from which the IR recruits adapters acting on substrates that are distinct from those accessible at the membrane. We report the biotinylation of human-recombinant insulin (rhIns) specifically at the position 29 of the B chain. We combined visible fluorescent proteins fused to IR and biotinylated rhIns conjugated with streptavidin-quantum dots to perform extended, quantitative experiments in real time. Modified rhIns bound to the IR and conjugated with the quantum dots was internalized with a rate constant (k) of 0.009 min^-1. Dissociation of insulin-IR complex in endocytosed vesicles occurred with k = 0.006 min^-1.

Full text of DOI:10.1021/bc300526d

Article
pubs.acs.org/bc
Endocytosis and Intracellular Dissociation Rates of Human
Insulin−Insulin Receptor Complexes by Quantum Dots in Living Cells
Jimena Giudice,†,‡,∥ Elizabeth A. Jares-Erijman,‡,⊥ and Federico Coluccio Leskow*,†,§
†Departamento de Química Biolog
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Intendente Guiraldes 2160, Ciudad Universitaria,
C1428EGA, Buenos Aires, Argentina
§Departamento de Ciencias Bas
́ ́
icas, Universidad Nacional de Lujan, Buenos Aires, Argentina
́
ica, IQUIBICEN-CONICET and ‡Departamento de Química Organ
́
ica, CIHIDECAR-CONICET,
̈
S
* Supporting Information
ABSTRACT: Insulin signaling is involved in glucose metabolism,
cellular growth, and differentiation. Its function is altered in diabetes
and many cancer types. Insulin binding to insulin receptor (IR)
triggers diverse signaling pathways. However, signal transduction by
IR is not mediated exclusively at the cell surface. Activated ligand−
receptor complexes are internalized into endosomes from which the
IR recruits adapters acting on substrates that are distinct from those
accessible at the membrane. We report the biotinylation of human-
recombinant insulin (rhIns) specifically at the position 29 of the B
chain. We combined visible fluorescent proteins fused to IR and
biotinylated rhIns conjugated with streptavidin−quantum dots to
perform extended, quantitative experiments in real time. Modified rhIns
bound to the IR and conjugated with the quantum dots was internalized
with a rate constant (k) of 0.009 min−1. Dissociation of insulin-IR complex in endocytosed vesicles occurred with k = 0.006 min−1.
including gene transcription stimulated by FOS and ELK1.3
INTRODUCTION
■
This last pathway operates independently of IRS, acting via the
interaction of IR with the adaptor protein SHC that associates
with GRB2 and thus engages the activation sequence for RAS
and MAPK.4−7
The insulin receptor (IR) is a member of the insulin tyrosine
kinase receptor family, which also includes the insulin-like
growth factor I (IGF-I) receptor (IGF-IR) and the IR related
receptor. The IR can be activated by different ligands leading to
different biological effects attributable to differences in receptor
interaction, internalization rates, and phosphorylation patterns
of the IR and its substrates. Although the major insulin targets
are the liver, adipose tissue, and skeletal muscle, IRs are also
found in the brain, heart, kidney, pulmonary alveoli, pancreatic
acini, placenta vascular endothelium, monocytes, granulocytes,
erythrocytes, and fibroblasts.1 The observation that IRs are not
restricted to insulin target tissues suggests that they may be
functionally linked to multiple processes in addition to their
classical roles in the insulin-dependent regulation of metabo-
lism.
Insulin binding to the IR promotes receptor autophosphor-
ylation and the phosphorylation of IR substrates (IRS-1, IRS-2,
and IRS-3), leading to association of IRSs with different effector
molecules mediating divergent signaling pathways regulating
either glucose homeostasis and metabolic gene expression or
cell growth and differentiation.1,2 The metabolic effect of
insulin is mediated by recruitment and activation of Akt at the
cell membrane, leading to the translocation of the glucose
transporter 4 (GLUT4) promoting the uptake of glucose and
fatty acids, and synthesis of glycogen and protein. Insulin also
activates the cascade that leads to activation of mitogen-
activated protein kinase (MAPK) and mitogenic responses
Signal transduction by IR is not limited to its activation at the
cell surface. The activated ligand−receptor complex is
internalized into endosomes, and this process is dependent
on kinase autophosphorylation.8 Endocytosis of activated
receptors has the dual effect of concentrating receptors within
endosomes and allowing the IR tyrosine kinase to phosphor-
ylate substrates spatially distinct from those accessible at the
plasma membrane.9−15 Acidification of the endosomal lumen
results in dissociation of insulin from its receptor and signaling
termination. Receptors could be recycled to the plasma
membrane or transported to lysosomes for degradation. The
lifetime of a ligand−receptor complex within the endosomes
impacts the response produced by a particular receptor. Insulin,
which has a relatively short endosomal residence, elicits
primarily acute metabolic effects, whereas the epidermal growth
factor (EGF) and the IGF-I, which have longer endosomal
residence times, elicit mitogenic responses.3 These differences in
biological responses probably reflect distinct actions of intra-
cellular signaling molecules at specific cellular locations. Thus,
Received: September 19, 2012
Revised: January 25, 2013
Published: January 29, 2013
© 2013 American Chemical Society
431
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
differences in ligand residency, along with differences in the
intrinsic activity of receptors, may contribute to the activation of
different signaling pathways.3
EXPERIMENTAL PROCEDURES
■
Materials. rhIns was kindly provided by Laboratorios Beta
(Argentina). The streptavidin-QD655 was from Molecular
Probes (# Q10121MP), Invitrogen (Eugene, OR). Bovine insulin
conjugated with FITC (FITC-Ins) and biotin amido caproyl insulin
(BAC-Ins) were from SIGMA (Saint Louis, MO). Lipofectamine
Reagent 2000 was from Invitrogen (Carlsbad, CA). Dialysis
membrane (<3500 Da) was from Cellu Sep (Seguin, TX). Mouse
monoclonal antihuman influenza hemagglutinin (HA) was from
Covance (Emeryville, CA). Mouse monoclonal anti-PY99, rabbit
polyclonal antiphosphorylated Akt (Ser 473), and rabbit polyclonal
anti-ERK2 were from Santa Cruz Biotechnology (Santa Cruz, CA).
Mouse monoclonal anti-IR-β-subunit, rabbit monoclonal anti-p-Akt
(Ser 473, immunofluorescence experiments), and rabbit mono-
clonal anti-p-MAPK (p44/42) were from Cell Signaling Technol-
ogy (Danvers, MA). For specific IR phosphorylation detection, we
used rabbit monoclonal anti-p-IR-β subunit (Tyr 1361) from Cell
Signaling (cat #3023) which detects transfected levels of IR β
subunit and only slightly cross-reacts with IGF-I receptor.
Plasmid Generation. pcDNA3-IR-B: cDNA of IR-B was
provided by Dr. A. Ullrich (Max Planck Institute for Biophysical
Chemistry, Germany) and amplified by PCR using primers
containing HindIII restriction site (IR-Forward: 5′-aagcttatggcca-
ccgggggccgg-3′) and NheI and XbaI sites (IR-Reverse: 5′-tctagagc-
tagcgaaggattggaccgaggcaaggtc-3′). The product was digested with
HindIII and XbaI enzymes and cloned into pcDNA3 vector.
pcDNA3-IR-B-GFP, pcDNA3-IR-B-SCFP and pcDNA3-IR-B-
SYFP: peGFP-C3 (Clontech), pSCFP-3AC, or pSYFP-2F
(provided by T. W. Gadella Jr., University of Amsterdam, The
Netherlands) were digested with NheI and ApaI, and cloned into
pcDNA3-IR-B.
Chemical Modification of rhIns. The rhIns was biotinylated
in a two-step synthesis: first, by generation of N-hydroxysucci-
nimide (NHS) ester of biotin bound to 4 poly(ethylene oxide)
(PEO) groups (PEO4-biotin) from the corresponding acid. One
equivalent (equiv) (0.0042 mmol) PEO4-biotin acid was
reacted with 2 equiv NHS in dimethylformamide (DMF); 2.5
equiv of N,N′-dicyclohexilcarbodiimide (DCC) was added, and
after reaction for 4 h at room temperature (RT), the products
were monitored by TLC in reverse-phase (methanol−water
(3:1 and 1:2) as mobile phase). We modified the conditions
reported by Jensen and Wilkinson27 in order to covalently bind
the PEO4-biotin to rhIns. The product of the first step was
mixed with 0.0045 mmol rhIns dissolved in DMF/triethylamine
(TEA) during 6 h at RT. The reaction was monitored by TLC
in reverse phase (methanol−water (3:1) as mobile phase). The
reaction was stopped by adding 1 M acetic acid, dialyzed
against milliQ water for 3 days (renewed every 12 h),
lyophilized, and analyzed by matrix-assisted laser desorption/
ionization (MALDI-TOF) (Ultraflex II, Bruker, CEQUIBIEM,
FCEN, UBA, Argentina): (i) without any treatment; (ii) after
treatment with 20 mM DTT and 100 mM iodoacetamide for
15 min at RT; (iii) after treatment with trypsin for 1−2 h at
37 °C; and (iv) after treatment with trypsin and subsequently
with dithiothreitol (DTT).28 For MALDI characterization, the
solid sample was dissolved in 2 M urea, 15 mM Tris-HCl pH
8.0, and diluted in 50 mM (NH)4HCO3. Sinapinic acid
(10 mg/mL) dissolved in 70% acetonitrile/0.125% TFA was
used as matrix.
During the past 30 years, many reports have studied
internalization of different receptor−ligand complexes (in particular,
insulin−IR) mostly using radioactive ligands in order to measure
endocytosis kinetics. These biochemical approaches gave a
deep view of cell traffic as an average of cell populations.
However, very little was done in order to study insulin−IR
complexes endocytosis and dissociation in vivo and in
individual cells. In addition, almost all the available data of
internalization rates comes from experiments performed with
hormones from animal origin (bovine, porcine, etc.).
During the past decade, diverse labeling strategies have
provided a spatiotemporal view of the dynamics of proteins in
living cells.16−18 The combination of the visible fluorescent
proteins (VFPs) fused to the molecules of interest and
quantum dots (QDs) bound to different ligands allowed the
study in living cells of the multiple stages of signaling by
fluorescence microscopy. QDs provide unique advantages for
cellular imaging: (i) intense and uniform brilliance, permitting
detection down to the single nanoparticle level, and reliable
quantification of binding and transport phenomena; (ii)
photostability, allowing imaging over prolonged periods; (iii)
broad excitation spectra, allowing the simultaneous excitation of
VFPs and QDs; (iv) narrow, tunable emission bands
conveniently red-shifted to the fluorescence of the VFPs; (v)
biochemical stability and specificity with low background and
adjustable stoichiometry (ligands/QDs) afforded by the
biotin−streptavidin titration; (vi) ligand compliance, the
noninterference of the QD with the biochemical specificity of
an attached ligand; and (vi) steric compatibility, as, for example,
the ability of two QD-ligands to bind to two monomers of a
dimeric receptor.19
Biotinylated ligands in combination with streptavidin-
conjugated QDs provide valuable tools for studying receptor
tyrosine kinase dynamics in living cells, including the
characterization of endocytosis and retrograde transport.19−25
Recently, it was demonstrated that QDs specifically targeted to
EGFR can clearly distinguish low-grade as well as high-grade
glioma tissue from normal brain tissue both at the macroscopic
and the single cell level with high contrast ratios. The strong
and photostable fluorescence and rapid differential binding of
these probes meet some of the criteria required by surgeons to
distinguish tumor cells left in the resection cavity.26
Here, we addressed the study of IR internalization using the
advantages of the QD to track the endocytosis of ligand−
receptor complexes in individual cells. We modified human
recombinant insulin (rhIns) by introducing a biotin group
specifically in the lysine located at the position 29 of the B
chain (B29). We incorporated a linker of ∼18 Å in order to
maintain biological activity and increase accessibility. The
modification allowed subsequent conjugation with streptavi-
din−QDs and thus the visualization of human insulin binding
specifically to the IR-B fused to the super yellow fluorescent
protein (SYFP) in vivo. We imaged endocytosis and trafficking
of the human insulin-QD655-IR-SYFP complex in real time,
performing quantitative assessment of ligand−receptor internal-
ization and colocalization. These procedures led to the
estimations of the rate constants and lifetime of the internalized
insulin-IR complexes.
Cell Culture and Transfection. Cell culture reagents
(Dulbecco’s modified Eagle’s medium (DMEM) Glutamax,
Optimem, fetal bovine serum (FBS), trypsin, supplements, and
432
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
antibiotics) were from GIBCO (Grand Island, NE). HeLa,
CHO, and Cos cells (ATCC-CCL-2) were maintained in
DMEN supplemented with penicillin, streptomycin, and 10%
(v/v) FBS at 37 °C in 5% CO2. HEK 293 cells were maintained
similarly but supplemented with 1 mM sodium pyruvate. For
HepG2 cells, we used a nonessential amino acid supplement
(0.1 mM). Cells were plated at 1 × 105 cells/well in 24 well
plates onto 12 mm glass coverslips one day before transfection in
DMEM/10% (v/v) FBS without antibiotics. Cells were trans-
fected with Lipofectamine Reagent 2000 for 5 h (0.4 μg DNA and
1 μL Lipofectamine per well in DMEM no adds). After
transfection, cells were cultured for 12 h in DMEM/10% (v/v)
FBS and then for 48 h in DMEM/10% (v/v) FBS/streptomycin
and penicillin.
Western Blots. Following stimulation with 100 nM rhIns or
FITC-Ins 10 min, cells were lysed in buffer containing 50 mM
Tris-HCl pH 7.5, 1 mM EDTA, 1 mM EGTA, 150 mM NaCl,
10% (v/v) glycerol, 1% (v/v) NP40, 1 mM MgCl2, 0.1% (v/v)
SDS, 2 mM Na3VO4, 40 mM sodium β-glycerophosphate, 1 mM
DTT, 1 mM NaF, and a protease inhibitor cocktail (Roche),
centrifuging for 10 min at 13 000g at 4 °C. Supernatants were
mixed with loading buffer (50 mM Tris-HCl pH 6.8, 2% (v/v)
SDS, 0.1% (w/v) Bromophenol blue, 10% (v/v) glycerol, 100 mM
β-mercaptoethanol) and heated for 5 min at 100 °C. Following
10% SDS-PAGE and transfer, membranes were blocked in 5%
(w/v) nonfat dried milk in 0.1% (v/v) Tween-TBS buffer (TTBS)
for 1 h, washed and incubated overnight at 4 °C with primary
antibodies diluted in 5% (w/v) BSA/TTBS (anti-IR-β-subunit,
1/500; anti-ERK2, 0.4 μg/mL; anti-p44/42 MAPK, 1/500; anti-p-
Akt, 0.4 μg/mL; anti-HA, 1/1000) or in 2% (w/v) BSA/TTBS
(anti-PY20, 2 μg/mL). The following day, membranes were
incubated with secondary antibodies for 1 h and washed. Blots
were developed by chemiluminescence using a Bio-Imaging
Analyzer Bas-1800II and Image Gauge 3.12, FUJIFILM (FCEN,
UBA. Argentina).
immersion objective. Excitation and emission filters were as
follows: excitation SYFP, 515 nm; emission SYFP, band-pass
(BP) 535−565 nm; excitation QD655, 405 nm; emission
QD655, BP 585−685 nm. (ii) Zeiss LSM 510 Meta with a Plan-
Apochromat 63× 1.4 NA oil immersion objective. Excitation and
emission filters were as follows: excitation FITC, 488 nm; emission
FITC, 510−563 nm; excitation Cy3, 532 nm; emission Cy3, 563−
606 nm; excitation GFP, 488 nm, emission GFP, BP 500/20 nm;
excitation QD655, 488 and 458 nm; emission QD655, LP 650 nm.
Wide-field microscopy was carried out with an Olympus
IX71 microscope with a 40× 1.15 NA water-immersion objective
or a Zeiss Axiovert S100 with a 63× 1.25 NA oil immersion
objective (Zeiss). A mercury arc lamp excitation and suitable filter
were used. Camera: Hamamatsu Orca CCD C4742−95.
Image Processing. Images were processed with Matlab
(TU Delft, The Netherlands, http://www.ph.tn.tudelft.nl/
DIPlib/index.html) and ImageJ (http://rsb.info.nih.gov,
NIH). Background of each channel (mean of empty region
for ImageJ and image median for Matlab) was subtracted and,
in some cases, a median filter was applied (radius: 1 pixel) only
for presentation. For wide field experiments (Akt activation), z-
stacks (80 frames, 0.2 μm step size) were deconvolved using
Huygens Professional Scientific Volume Imaging v 3.6 software,
applying a quick maximum likelihood estimation algorithm.
Internalization Analysis. For quantitative internalization
experiments, we defined the onset time as the end of labeling
washes, and cells were incubated at 37 °C for 20, 40, 35, 95,
150, 160, and 300 min of incubation at 37 °C.
Segmentation (Cell and Interior). Channel backgrounds
were subtracted. Segmentation was performed for each cell using
VFP signal or DIC image. After cell and interior segmentation, we
measured QD655 signal in both regions with ImageJ or Matlab.
Estimation of the Internalization Ratio. To compute the
relative amount of internalized fluorescence, we estimated
QDtotal as the measurements of the QD655 in the region cell
and the QDinterior as the measurements in the region interior.
Finally, we calculated for each cell the ratio QDinterior/QDtotal. This
“internal calibration” approach was chosen to remove influence of
amplifier gain and zoom factor for each image acquisition condition.
Colocalization Analysis. Manders coefficient and PDM
(differences from the mean) and frequencies plots were performed
with ImageJ plugin ″Intensity correlation analysis”. Manders
coefficients were calculated by eqs 1−3.
FITC-Ins Labeling. Before the experiment, cells expressing
IR-B, starved overnight, were washed with Tyrode’s buffer
(135 mM NaCl, 10 mM KCl, 10 mM MgCl2, 1 mM CaCl2,
10 mM HEPES pH 7.2, 0.1% (w/v) BSA) at RT and incubated
with 50 nM FITC-Ins for 15 min at 15 °C. Cells were either
fixed in 3.7% PFA on ice during 20 min or incubated at 37 °C
for different periods of time before fixation.
Labeling in Vivo with QDs and Biotinylated Insulins
and Internalization. Before the experiment, cells expressing
IR-B or IR-B-VFP were starved overnight, washed with
Tyrode’s buffer, and labeled following a two-step protocol by
incubating with 50 nM FITC-Ins or 50 nM BAC-Ins or rhIns-
PEO4-biotin for 15 min, washing with Tyrode’s buffer, and
subsequently incubating with 1 nM QD655 for 10 min in
darkness. After washing, cells were incubated at 37 °C in DMEM
for different time periods before fixation or live visualization.
Immunofluorescence. After overnight starvation, trans-
fected cells were stimulated with 100 nM rhIns, FITC-Ins, or
BAC-Ins for 10 min, washed with PBS, and immediately fixed
in cold methanol for 30 min at −20 °C, blocked with PBS/0.3%
(v/v) Triton X-100/1% (w/v) BSA for 1 h at 37 °C and
incubated with anti-PY99 (1 μg/mL), anti-p-Akt (Ser 473; Cell
Signaling, 1/200), or anti-p-IR β subunit (0.3 μg/mL) overnight at
4 °C. The following day, samples were incubated with a secondary
antibody conjugated with Cy3 during 1 h at 37 °C and washed.
Microscopy. Confocal laser scanning microscopy was
performed with the following microscopes: (i) Olympus
Fluoview FV 1000 with a UPLSAPO 60× 1.2 NA water
∑ (Imsyfp × Im QD655
∑ (Im syfp)2 × ∑ (Im QD655
)
M =
2
)
(1)
(2)
(3)
∑i Im syfpicoloc
∑i Im syfpi
Msyfp
=
∑i Im QD655icoloc
∑i Im QD655
MQD655
=
where Imsyfp = image from SYFP channel; ImQD655 = image from
QD655 channel; Imsyfpicoloc and Imqd655icoloc = colocalized pixels in
Imsyfp and Imqd655
.
For each pixel, PDM value was calculated by eq 4:
PDM = (Im yfp Im syfp) × (Im QD655 Im QD655)
−
−
(4)
We used the loss of colocalization degree as an estimation of
the dissociation between ligand and receptor. High colocalization
433
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
Figure 1. Expression and function of IR-B chimeras. A. HeLa cells expressing IR-B or IR-B-SCFP were starved overnight in DMEM/0.1% BSA. Then, cells
were stimulated with 100 nM rhIns for 10 min and lysates were assayed by Western blot. B. HeLa cells expressing IR-B-GFP were treated in the same way, but
they were assayed by immunofluorescence using a specific primary antibody against p-IR and a secondary antibody conjugated with Alexa Fluor 555. Imaging
was carried out by confocal microscopy (Zeiss LSM 510 Meta). Scale bars: 10 μm. C. Akt phosphorylation (Ser473) profile in cells coexpressing Akt-HA and
IR-B or EV, starved for 16 h, and stimulated with 100 nM rhIns for 2, 5, and 15 min. D. Immunofluorescence experiments by wide-field microscopy (Olympus
IX71) showing Akt phosphorylation in response to insulin. Deconvolved middle frame of 80 z-stacks with 0.2 μm step size. Scale bars: 10 μm.
Figure 2. Fluorescent and biotinylated insulins in combination with QD. A. HeLa cells expressing IR-B were starved overnight and then labeled with 50 nM
FITC-Ins for 15 min at 15 °C. Cells were fixed directly or after incubation at 37 °C for 30 min. B. HeLa cells expressing IR-B were starved for 18 h,
stimulated with 50 nM FITC-Ins or rhIns for 10 min. After fixation, samples were assayed by immunofluorescence using an anti-p-Tyr (PY99) and a
secondary antibody conjugated with Cy3. C. HeLa cells expressing IR-B-GFP were starved overnight and sequentially incubated for 10 min at 15 °C with
50 nM BAC-Ins followed by 1 nM QD655 and fixed (upper panel) or incubated at 37 °C for 150 min prior to fixation (lower panel). D,E. HeLa cells
expressing IR-B were starved overnight and stimulated with 100 nM rhIns for 10 min. Samples were assayed by immunofluorescense (anti-p-Tyr (PY99);
panel D) or by Western blot (anti-p-Tyr (PY20), IR β-subunit, p-ERK 1/2 and total ERK; panel E). Imaging was performed by confocal microscopy (Zeiss
LSM 510 Meta). Scale bars: 10 μm. F. Cell-by-cell quantification of internalized IR-B-Insulin complexes was performed using Matlab. The same analysis was
performed for IR-B-FITC-Ins, IR-B-BAC-Ins-QD655, and IR-B-GFP-BAC-Ins-QD655 complexes. Results are shown as the mean s.e.m. (n = 9−39 cells.)
coefficient means ligand−receptor association; low coefficients
means lower association.
Statistical Analysis. Results were expressed as the mean of
Graphical Fitting. Curves for internalization rates and
colocalization data were fitted by Origin 8.6. For internalization
data, we used a Sigmoidea logistic curve, and for colocalization,
we applied a dose response model for fitting. In both cases,
R2 = 0.99 allowing us to conclude that fitting was appropriate.
at least 3 independent experiments
s.e.m. p values were
estimated using two-tailed Student’s t tests.
434
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
a
Table 1. Internalization of Bovine Insulin-IR Complexes
IR-B + FITC-Ins
IR-B + BAC-Ins + QD
s.e.m. (%)
IR-B-GFP + BAC-Ins + QD
time (min)
av (%)
s.e.m. (%)
n (cells)
av (%)
n (cells)
av (%)
s.e.m. (%)
n (cells)
20
46
60
75
1
4
3
12
12
15
48
64
81
2
3
2
17
39
23
44
67
87
3
3
3
21
22
9
40
150
a
Cell-by-cell quantification of internalized IR-Insulin complexes was performed using Matlab. The same analysis was performed for IR-B-FITC-Ins,
IR-B-BAC-Ins-QD655, and IR-B-GFP-BAC-Ins-QD655 complexes.
Figure 3. Modification of rhIns and fragmentation analysis. A. Biotinylation of rhIns. B. Amino acid sequence of human insulin and scheme of the
consequences of different treatments used for MALDI-TOF characterization. S−S indicates disulfide bonds. C. The table shows the mass (Da) for A
and B chains of rhIns, the dipeptide with the disulfide bonds, and the peptides obtained after trypsin treatment.
Therefore, we calculated the first derivative as a measurement of
the rate of internalization or dissociation (k). The maximum
ligand−receptor internalization or dissociation rates were obtained
by determining the zero of the second derivate of the curve.
generic antibody against p-Tyr, PY20) or by immunofluor-
escence using a specific antibody against the phosphorylated
IR-β subunit (Figure 1A,B). Both approaches confirmed
specific activation of the chimeras since nontransfected cells
did not show phosphorylation signal. Activation signal was
majorly localized at the plasma membrane at 10 min; however,
some endocytosed signal is detected close to the membrane
coming from activated IRs inside the endosomes (Figure 1B).
We evaluated Akt activation (phosphorylation of Ser 473) in
cells expressing IR-B or IR-B-SCFP and Akt-HA. We
performed these experiments by Western blot (Figure 1C)
and immunofluorescence (Figure 1D). Both approaches
confirmed that IR-B and IR-B-VFP were able to activate Akt
signaling in response to rhIns.
Visualization of Internalized IR-B with FITC-Ins and BAC-
Ins-QD655. Commercially available FITC-Ins bound to the
receptor in cells overexpressing IR-B (Figure 2A). Immunofluor-
escence using an antibody against phosphorylated Tyr (PY99)
showed autophosphorylation in transfected cells, whereas nontrans-
fected cells did not show any signal (Figure 2B). To assess the inter-
nalization of the insulin-IR complexes, cells were incubated at 37 °C
for 30 min after stimulation with FITC-Ins. We confirmed FITC-
ins-IR internalization (Figure 2A), although we discovered numerous
RESULTS
■
Generation and Expression of IR-B Chimeras. We
generated vectors expressing IR-B wild-type or fused with the
visible fluorescent proteins green (eGFP), cyan (SCFP-3AC),
and yellow (SYFP-2F). Different human cell lines (HeLa, HEK
293, HepG2) were transfected with these plasmids showing
correct localization of IR-B at the plasma membrane. However,
nonhuman cells (CHO y Cos7) showed protein accumulation
at the Golgi probably due to failures in protein folding and post-
translational modifications (Figure S1). We decided to continue our
experiments in HeLa cells given their low levels of endogenous IR
and their morphology and reduced autofluorescence.
It is known that, after insulin binding, IR changes its
conformation and is autophosphorylated in different residues of
its kinase domain. These events lead to Akt pathway activation
regulating different cellular processes such as glucose
metabolism, growth, gene expression, and differentiation. We
evaluated IR phosphorylation either by Western blot (with a
435
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
a
Table 2. Predicted Molecular Masses
a
Predicted molecular masses (Da) are shown for each treatment (DTT, trypsin, and trypsin + DTT) of insulin, mono-biotinylated insulin (A1, B1,
and B29), di-biotinylated insulin (A1B1, A1B29, and B1B29), and tri-biotinylated insulin (A1B1B29). B indicates PEO4-biotin.
disadvantages: FITC-Ins background signal, fluorophore stability is
low, and, most important, its fluorescence is affected by the low pH
of the endosomes. This scenario led us to design a strategy taking
advantage of commercially available streptavidin conjugated QD in
combination with biotinylated human recombinant insulin. To verify
whether the conjugation of the ligand with nanoparticles of ∼20 nm
or fusion with VFPs of the receptor affects endocytosis dynamics, we
used commercially available bovine insulin (BAC-Ins). HeLa cells
expressing IR-B-GFP were treated with 50 nM BAC-Ins and after
washes subsequently labeled with streptavidin QD655. Cells were
directly fixed or previously incubated at 37 °C for 150 min. Labeling
was specific: nontransfected cells did not show QD655 signal (Figure
2C). Additionally, internalization was not affected, since after 150
min, almost all the QD655 signal was detected intracellularly. We
confirmed by Western blot and immunofluorescence that ligand
biotinylation does not affect the biological activity: BAC-Ins triggered
IR-B and ERK 1/2 phosphorylation at similar levels that rhIns
(Figure 2D,E).
To evaluate if VFP size (∼27 kD) or QD size (∼18 nm) affects
internalization rate, we quantified endocytosis in a cell by cell
analysis. Cells were incubated at 37 °C after labeling (either with
FITC-Ins or with BAC-Ins+QD655) for 20, 40, or 150 min, fixed
on ice, and imaged by confocal microscopy. Using Matlab
routines, each cell was segmentated into “membrane” and
“interior”. Intensity of red signal was quantified in both areas,
and we estimated the ratio QDmembrane/QDmembrane+interior. These
results are shown in Figure 2F and suggest that internalization is
not affected by the VFP or by the QD size (Table 1).
436
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
Figure 4. Characterization of the reaction product by MALDI-TOF. A−D. MALDI-TOF (positive mode) of the reaction product: without treatment
(A), after DTT reduction and carbamidomethylation (B), after trypsin treatment for 1−2 h at 37 °C (C) or after treatment with trypsin for 1−2 h at
37 °C and subsequently with 20 mM DTT and 100 mM iodoacetamide for 15 min at RT) (D). The matrix used was sinapinic acid (10 mg/mL)
dissolved in 70% acetonitrile, 0.125% TFA. E. Experimental peaks (Da) obtained for the reaction product treated with DTT, trypsin, or both. In bold
and underscore, we show the predicted peaks which are exclusively present when B29 residue was modified, and in gray, the ones which are shared
with the unmodified insulin molecule. We show predicted molecular masses, experimental peaks, and assignment.
Chemical Modification of rhIns. We generated rhIns co-
valently bound to biotin through a linker of PEO4 with a length
of ∼18 Å. The linker was introduced in order to separate the
active part of the hormone from the modified portion of the
molecule and minimize interference with insulin function. The
synthesis was performed in two steps: (i) generation of NHS
ester of PEO4-biotin; and (ii) reaction between the previous
product and rhIns (Figure 3A).
(ii) treatment with DTT plus alkylation with iodoacetamide (A and
B chains of insulin were obtained after the reduction of disulfide
bonds); (iii) treatment with trypsin (this enzyme cleaves after the
last Arg located at the position 22 of the B chain (B22) and in some
cases can digest after Lys 29 of B chain (B29)); or (iv) treatment
with trypsin and subsequently with DTT plus iodoacetamide.
Figure 3B shows schematically the consequences of each treatment
on insulin molecule and Figure 3C summarizes predicted molecular
mass for the products of each treatment. Taking this into account
and considering the mass of each PEO4-biotin group (473.2 Da)),
Characterization of reaction products was performed by MALDI-
TOF after different treatments: (i) the sample without treatment;
437
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
Figure 5. Kinetics of internalization of human insulin in living cells. A. HeLa cells transiently expressing IR-B-SYFP were starved overnight and either
not stimulated or stimulated with 100 nM rhIns for 10 min at 37 °C. Lysates were assayed by Western blot. B. Cells expressing IR-B-SYFP labeled
with rhIns-PEO4-biot and QD655 were visualized in vivo by confocal microscopy (Olympus Fluoview FV 1000). Scale bar: 10 μm. C. Cells were
only treated with QD655 without rhIns-PEO4-biot. Acquisition parameters were the same than in panel B. Scale bar: 10 μm. D. PDM graph for cell
shown in panel B performed with ImageJ. E. Linear trajectories crossing the cell of the two signals using ImageJ. F. HeLa cells expressing IR-B-SYFP
labeled with 50 nM rhIns-PEO4-biot and 1 nM QD655 were incubated at 37 °C for 35, 95, 160, and 300 min and visualized in living cells at RT by
confocal microscopy (Olympus Fluoview FV 1000). Scale bars: 10 μm. G. Quantification of internalization ratio in individual cells is shown as the
mean s.e.m. (n = 6−7 cells.)
we estimated the expected molecular masses for each modified
peptide (Table 2).
which is important for binding to the IR, remained unmodified.29,30
The yield of the reaction estimated by MALDI-TOF was 15%.
rhIns-PEO4-biot Binds IR at the Plasma Membrane.
HeLa cells were transfected with the plasmid encoding IR-B
fused to SYFP (pcDNA3-R-B-SYFP). We evaluated the correct
expression of the fusion protein first by Western blot. IR-
B-SYFP was correctly expressed exhibiting the correct size of
the β subunit of IR fused with SYFP. This band appeared
∼25 kDa higher than that the endogenous IR. It was also
confirmed that the IR-B-SYFP was activated after stimulation
with rhIns (Figure 5A).
After 48 h of expression, cells were labeled with biotinylated
rhIns (rhIns-PEO4-biotin) and subsequently with different
streptavidin−QD655 concentrations (0.5, 1.0, 2.0, 3.0, 4.0 nM)
as we described above. Binding was quantified by confocal
microscopy imaging. Since QD655 binding was shown to be
concentration dependent in the range 0.5−3.0 nM, we decided
continue to work with 1 nM, a low unsaturated concent-
ration that does not compromise detection, to ensure single
In Figure 4A−D, the spectra obtained after the different
treatments are shown. The sample without treatment showed a
major peak for insulin (m/z = 5805.6 Da) and a minor one of
m/z = 6279.9 Da corresponding to monobiotinylated insulin
(see predicted molecular masses, experimental peaks, and
assignment in Figure 4E). Analysis showed that we did not obtain
insulin modified at the amino terminal of A chain: trypsin treatment
followed by DTT showed a single peak (m/z = 2400.8 Da)
corresponding to A chain + Na+ (Figure 4D). DTT treatment
showed a minor peak of m/z = 3902.1 Da, corresponding to B
chain modified with the PEO4-biotin group (Figure 4B). After both
treatments, we observed a peak (m/z = 2485.0 Da) corresponding
to unmodified B chain without the C-terminus (GFFYTPKT)
(Figure 4D). Altogether, the evidence leads to the conclusion that
the PEO4-biotin group has been introduced at Lys located at
position 29 of B chain and that the amino terminal of A chain,
438
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
Figure 6. Colocalization of rhIns-QD655 and IR-B-SYFP over the time in vivo. A. Frequencies and PDM graph for cells incubated at 37 °C for 35,
160, and 300 min. IR-B-SYFP signal is the vertical axis and Ins-QD655 the horizontal one in the frequencies plot. M: Manders coefficient. MQD655
:
Manders coefficient in the red channel. MSYFP: Manders coefficient in the green channel. B. Representation of the results performed in individual
cells for Manders coefficients. Results are shown as the mean s.e.m. (n = 6 cells.)
rhIns-PEO4-biotin binding to each nanoparticle (data not
shown). Only transfected cells showed QD655 labeling,
demonstrating that rhIns-PEO4-biotin was bound to the cells
through the IR-B (Figure 5B). Control transfected cells showed
no QD655 signal when treated without previous incubation with
rhIns-PEO4-biotin, confirming that QD655 binding was specific
(Figure 5C). As expected, whereas SYFP signal was present at the
membrane, as well as in the perinuclear region, rhIns-PEO4-biotin-
QD655 was only detected at the membrane.
To analyze the colocalization between IR and its ligand, we
calculated Manders coefficients for each cell using the ImageJ
plugin ″Intensity correlation analysis”. The PDM (product
differences of the mean) graph showed that major contributions to
the colocalization coefficient were localized at the plasma membrane
(Figure 5D). We also evaluated colocalization across the cell
body by superimposing the two signals over a linear trajectory
(Figure 5E). The intracellular intensity of SYFP was considerable,
reflecting receptor processing in the biosynthetic pathway.
Real Time Imaging of Human Insulin Internalization
through IR-B. We next addressed whether modified insulin
was capable of being internalized through the IR. Experiments
were performed by incubating the cells at 37 °C for different
times (35, 95, and 160 min) before imaging by confocal micro-
scopy. Whereas at the onset (no incubation at 37 °C) QD655
signal was localized exclusively at the plasma membrane, after
incubation at 37 °C, QDs was detected inside the cell. The
fraction of intracellular QD655 increased over time, leading to an
accumulation of the signal in perinuclear regions (Figure 5F). To
quantify these observations, we performed a segmentation of the
cell images into total and interior compartments, and determined
the total QD655 signal in both regions. The degree of
internalization was obtained from the ratio QDinterior/QDtotal
(Figure 5G): 0.09
0.02 (onset); 0.22
0.06 (35 min);
0.72 0.09 (95 min); 0.89 0.03 (160 min); and 0.94 0.02
(300 min) (n = 6−7 cells for each time point). The data were fit to
a sigmoid curve (R2 = 0.99) and calculation of the first derivative
was taken as a measure of the internalization rate constant (k), the
maximal value of which was 0.009 0.001 min−1.
Endosomal Insulin-IR-B Complex Lifetime. It is
generally proposed that IR endocytosis is crucial for turning off
the metabolic signaling at the membrane and that signals from the
endosomes preferentially target the mitogenic cascade. Thus,
internalization on the activated receptor appears to be critical for
maintaining signaling balance. We hypothesize that IR remain
active in endosomes as long as insulin remains bound to it, and
thus determined the kinetics of colocalization as a measurement of
insulin−IR complex integrity inside the cell.
Confocal images obtained showed that QD and SYFP signals
colocalized at the plasma membrane at the onset of the
experiments. This condition was maintained in early endosomal
compartments after endocytosis, whereas at latter times, QD
signal associated with insulin was distributed in a distinct
manner from that of the SYFP (receptor signal) and was
probably accumulated in lysosomes (Figure 6A, last panel).
We quantified these observations by calculating Manders
coefficients in individual cells for each time point, and generated
PDM and frequency plots (Figure 6A). PDM showed differences
between the early internalization steps (onset and 35 min) and the
late ones (95 and 300 min). At the beginning and up to 35 min,
the QD and SYFP frequencies mapped to a line close to 45°; in
contrast, after 95 and 300 min the plot showed two populations of
pixels for each signal (Figure 6A). Accordingly, colocalization
coefficients decreased significantly (p < 0.0001) over time:
0.81 0.02 (onset); 0.74 0.06 (35 min); 0.41 0.04 (95 min);
439
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
0.34 0.02 (160 min); and 0.28 0.02 (300 min) (n = 6 cells
for each time point) (Figure 6B). From this analysis, it was
possible to estimate the maximum ligand−receptor dissociation
rate: kmax = (0.006 0.001) min−1. In addition, we calculated the
half-time of colocalization in endocytosed vesicles of 68 min and
estimated that after 115 min the degree of colocalization in
endocytosed vesicles was reduced by more than 90%. We interpret
this value as a characterized dissociation time of the intracellular
pool of ligand−receptor complexes, and that mitogenic signaling
terminated during the ensuing process of receptor dephosphor-
ylation.
using HeLa cells, we calculated k = 0.009 min−1 and a half-time of
79 min using the human IR-B and human insulin. The methods
employed allowed determination of the endocytosis rate constant
at later times than those available using radiolabeled ligands.
Internalization of insulin−IR constitutes the major mecha-
nism of insulin degradation and down-regulation of cell surface
receptors.43,44 However, signal transduction by IR is not limited
to its activation at the plasma membrane. The activated ligand−
receptor complex internalized into endosomes would activate
substrates that are spatially distinct from those accessible at the
plasma membrane. IR internalization is required for the
activation of SHC/MAPK pathway but not for IRS-1 and Akt
phosphorylation, suggesting that the latter molecules are
activated by IR at the membrane. From this information and
the data reported here, we infer that the dynamics of the
activated IR depends on the cellular context and it is crucial for
the correct balance of its signaling.
It has been postulated that the balance between the metabolic
and mitogenic cascade activation depends on the degree of kinase
activation.45,46 Insulin analogues with slower dissociation constants
induce a greater degree of mitogenic compared to metabolic
signaling than unmodified insulin.46−48 Moreover, different
reports demonstrated a blockage of mitogenic pathway by
internalization inhibition.49−51 Combined internalization and
colocalization experiments allowed a comparative and quantita-
tive analysis of dynamics of IR signaling and its termination. We
propose that internalization of activated receptor plays a crucial
role in the signaling balance. On one hand, it would turn off
metabolic signaling by displacing IR from the effector molecules
localized at the membrane; on the other hand, mitogenic signaling
would be promoted from endosomes, although endosomal
trafficking would finally lead to dissociation of insulin−receptor
complexes in late endosomes and lysosomes switching off the
signaling pathways.
DISCUSSION
■
In order to study the endocytosis of insulin-IR complexes in
living cells, we used biotinylated rhIns conjugated with
streptavidin−QD. As compare with conventional fluorescent
ligands (FITC-Ins), fluorescent nanoparticles have a high
brilliance and stability allowing long-term experiments. In
addition, they are photostable and are not affected by pH
changes during trafficking and can be combined with different
VFPs. To achieve this, we covalently modified rhIns using a
biotin bound to a linker constituted by four PEO groups.
Reaction product contained unmodified insulin and mono-
biotinylated insulin (15%), allowing the imaging of receptor−
ligand complex using streptavidin−QDs. Human recombinant
biotinylated insulin bound properly to IR and led to its
internalization. Between the hormone and the biotin, we
introduced a linker of ∼18 Å corresponding to PEO groups.
The presence of the linker allowed insulin binding to the IR
while conjugated with the nanoparticle.
Biotinylation of human insulin was used in the past to purify the
IR by avidin−Sepharose affinity chromatography.31−34 In order to
be used as bifunctional ligands for affinity chromatography of the
receptor, insulin analogues should bind simultaneously to avidins
and the receptor and bonds must be stable to resist exhaustive
washing procedures to remove contaminants from affinity
columns. The cited publications addressed two issues: (i) finding
a spacer arm long enough to separate the biotin group from
insulin active site so that receptor and avidin binding can occur
simultaneously; and (ii) selecting chemical reaction conditions
avoiding modification of amino terminal of A chain. In our work,
we improved both conditions in order to generate labeled insulin
allowing imaging of binding and internalization of insulin−IR
complex in living cells.
ASSOCIATED CONTENT
* Supporting Information
Figure S1 and Figure S2. This material is available free of charge
via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: federico@fbmc.fcen.uba.ar. Tel.: (5411) 4576-3300.
■
Present Address
∥Department of Pathology and Immunology, Baylor College of
Medicine, One Baylor Plaza, Houston, TX 77030, United
States.
Conjugation of modified human insulin with streptavidin−QD
combined the established mode of chemical insulin modification
with new nanotechnology for studying IR dynamics and allowed
quantification of internalization and dissociation kinetics of
insulin−receptor complexes. Although biotinylated insulin, BAC-
Ins, and FITC-Ins are commercially available and were successfully
applied in cellular studies by microscopy,35−39 they are of bovine
or porcine origin. Here, we applied the modification of rhIns to
study the kinetics of the ligand receptor endocytosis and
dissociation in a completely human context. In the 1980s and
1990s, numerous papers were published measuring internalization
kinetics of IR using traditional radioactive experiments and insulins
of different animal origins.40−43 Some of these reports are
discordant, depending on the cells used or the method of analysis.
For example, while Draznin and colleagues39 reported k = 0.049
min−1 and a half-time of 14.5 min in rat-isolated hepatocytes using
Author Contributions
J.G., E.A.J.E., and F.C.L. conceived and designed the experi-
ments; J.G. performed the experiments; J.G., E.A.J.E., and
F.C.L. analyzed the data; J.G. and F.C.L.wrote the paper.
Notes
The authors declare no competing financial interest.
⊥Deceased on September 29, 2011.
ACKNOWLEDGMENTS
■
Elizabeth A. Jares-Erijman died during manuscript preparation
and we dedicate it to her memory. We thank Laboratorios
BETA (Argentina) for providing the rhIns and Dr. M. Cristina
Matulewickz (FCEN, UBA) for her help with sample
lyophilization. We especially thank Dr. Thomas Jovin for his
help and support and Dr. Silvia Moreno for her kindly help
42
pork radiolabeled insulin, Knuston derived k = 0.0023 min−1
and a half-time of 4.9 h using embryonic Swiss mouse fibroblastic
3T3-C2 cells and bovine radiolabeled insulin. In the present work,
440
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
(11) Grimes, M. L., Zhou, J., Beattie, E. C., Yuen, E. C., Hall, D. E.,
Valletta, J. S., Topp, K. S., LaVail, J. H., Bunnett, N. W., and Mobley,
W. C. (1996) Endocytosis of activated TrkA: Evidence that nerve
growth factor induces formation of signaling endosomes. J. Neurosci.
16, 7950−7964.
(12) Pol, A., Calvo, M., and Enrich, C. (1998) Isolated endosomes
from quiescent rat liver contain the signal transduction machinery.
Differential distribution of activated Raf-1 and Mek in the endocytic
compartment. FEBS Lett. 441, 34−38.
with MALDI-TOF analysis, interpretation, and conclusions. We
also thank Dr. Anabella Srebrow for pCMV-Akt-HA plasmid
and Dr. Patricia Cuasnicu for anti-PY20 antibody. This work
was supported by Max-Planck Society (E.A.J.E., Partner Group
grant), University of Buenos Aires (X106), CONICET
(Argentina), and ANPCyT (PICT-2006-01365) (Argentina).
J.G. was a recipient of a CONICET fellowship and was awarded
a short-term EMBO fellowship.
(13) Rizzo, M. A., Shome, K., Watkins, S. C., and Romero, G. (2000)
The recruitment of Raf-1 to membranes is mediated by direct
interaction with phosphatidic acid and is independent of association
with Ras. J. Biol. Chem. 275, 23911−23918.
(14) Sorkin, A., McClure, M., Huang, F., and Carter, R. (2000)
Interaction of EGF receptor and grb2 in living cells visualized by
fluorescence resonance energy transfer (FRET) microscopy. Curr. Biol.
10, 1395−1398.
(15) Jensen, M., and De Meyts, P. (2009) Molecular mechanisms of
differential intracellular signaling from the insulin receptor. Vitamins
and Hormones 80, 51−75.
(16) Jares-Erijman, E. A., and Jovin, T. M. (2003) FRET imaging.
Nat. Biotechnol. 21, 1387−1395.
(17) George, N., Pick, H., Vogel, H., Johnsson, N., and Johnsson, K.
J. (2004) Specific labeling of cell surface proteins with chemically
diverse compounds. J. Am. Chem. Soc. 126, 8896−8897.
(18) Marks, K., and Nolan, G. (2006) Chemical labeling strategies for
cell biology. Nat. Methods 3, 591−596.
(19) Lidke, D. S., Nagy, P., Heintzmann, R., Arndt-Jovin, D. J., Post,
J., Grecco, H. E., Jares-Erijman, E. A., and Jovin, T. M. (2004)
Quantum dot ligands reveal EGFR dynamics in living cells. Nat.
Biotechnol. 22, 198−203.
ABBREVIATIONS
■
BAC-Ins, Biotin amido caproyl insulin; BP, band-pass; DCC,
N,N′-dicyclohexylcarbodiimide; DMF, dimethylformamide;
DTT, dithiothreitol; EGF, epidermal growth factor; EV,
empty vector; FBS, fetal bovine serum; FITC-Ins, Insulin
conjugated with FITC; GFP, enhanced green fluorescent
protein; GLUT4, glucose transporter 4; HA, human influenza
hemagglutinin; IGF, insulin like growth factor; IGF-IR, IGF-I
receptor; IR, insulin receptor; IR-B, insulin receptor isoform B;
IRS, insulin receptor substrates; M, Manders coefficient;
MALDI, matrix-assisted laser desorption/ionization; MAPK,
mitogen-activated protein kinase; NHS, N-hydroxysuccinimide;
PDM, differences from the mean; PEO, polyethylene oxide;
PFA, paraformaldehyde; QD, quantum dot; rhIns, recombinant
human insulin; RT, room temperature; SCFP, super cyan
fluorescent protein; SYFP, super yellow fluorescent protein;
TEA, triethylamine; TOF, time-of-flight; VFP, visible fluo-
rescent protein.
(20) Lidke, D. S., Lidke, K., Rieger, B., Jovin, T. M., and Arndt-Jovin,
D. J. (2005) Reaching out for signals: filopodia sense EGF and
respond by directed retrograde transport of activated receptors. J. Cell
Biol. 170, 619−626.
(21) Lidke, D. S., Nagy, P., Heintzmann, R., Jovin, T. M., and Arndt-
Jovin, D. J. (2007) Biotin-ligand complexes with streptavidin quantum
dots for in vivo cell labeling of membrane receptors. Methods Mol. Biol.
374, 69−80.
(22) Cambi, A., Lidke, D. S., Arndt-Jovin, D. J., Figdor, C. G., and
Jovin, T. M. (2007) Ligand-conjugated quantum dots monitor antigen
uptake and processing by dendritic cells. Nano Lett. 7, 970−977.
(23) Echarte, M. M., Bruno, L., Arndt-Jovin, D. J., Jovin, T. M., and
Pietrasanta, L. I. (2007) Quantitative single particle tracking of NGF-
receptor complexes: transport is bidirectional but biased by longer
retrograde run lengths. FEBS Lett. 581, 2905−29013.
(24) Andrews, N. L., Pfeiffer, J. R., Martinez, A. M., Haaland, D. M.,
Davis, R. W., Kawakami, T., Oliver, J. M., Wilson, B. S., and Lidke, D.
S. (2009) Small, mobile Fc epsilon RI receptor aggregates are signaling
competent. Immunity 31, 469−479.
(25) Sigot, V., Arndt-Jovin, D. J., and Jovin, T. M. (2010) Targeted
cellular delivery of quantum dots loaded on and in biotinylated
liposomes. Bioconjugate Chem. 21, 1465−1472.
(26) Kantelhardt, S., Caarls, W., de Vries, A. H. V., Hagen, G. M.,
Jovin, T. M., and Arndt-Jovin, D. J. (2010) Specific visualization of
glioma cells in living low-grade tumor tissue. Plos One 5, e11323
DOI: 10.1371/journal.pone.0011323.
(27) Jensen, P. E., and Wilkinson, K. D. (1989) Probing the structure
of processed antigen by using biotin and avidin. Mhc-dependent
inhibition of responses to selected biotinyl-insulin derivatives. J.
Immunol. 143, 3423−3429.
(28) Tsai, Y. J., Rottero, A., Chow, D. D., Hwang, K. J., Lee, V. H. L.,
Zhu, G., and Chan, K. K. (1997) Synthesis and purification of NB1-
palmitoyl insulin. J. Pharm. Sci. 86, 1264−1268.
(29) Pullen, R. A., Lindsay, D. G., Wood, S. P., Tickle, I. J., Blundell,
T. L., Wollmer, A., Krail, G., Brandenburg, D., Zahn, H., Gliemann, J.,
and Gammeltoft, S. (1976) Receptor binding region of insulin. Nature
259, 369−373.
REFERENCES
■
(1) Kaplan, S. A. (1984) The insulin receptor. J. Pediatr. 104, 327−
336.
(2) Belfiore, A., Frasca, F., Pandini, G., Sciacca, L., and Vigneri, R.
(2009) Insulin receptor isoforms and Insulin receptor/Insulin-like
growth factor receptor hybrids in physiology and disease. Endocrine
Reviews 30, 1−38.
(3) Bevan, P. (2001) Insulin signaling. J. Cell Sci. 114, 1429−1430.
(4) Kouhara, H., Hadari, Y. R., Spivak-Kroizman, T., Schilling, J., Bar-
Sagi, D., Lax, I., and Schlessinger, J. (1997) A lipid-anchored Grb2-
binding protein that links FGF-receptor activation to the Ras/MAPK
signaling pathway. Cell 89, 693−702.
(5) Gotoh, N., Toyoda, M., and Shibuya, M. (1997) Tyr
phosphorylation sites at amino acids 239 and 240 of Shc are involved
in epidermal growth factor-induced mitogenic signaling that is distinct
from Ras/mitogen-activated protein kinase activation. Mol. Cell. Biol.
17, 1824−1831.
(6) Weng, L. P., Smith, W. M., Brown, J. L., and Eng, C. (2001)
PTEN inhibits insulin-stimulated MEK/MAPK activation and cell
growth by blocking IRS-1 phosphorylation and IRS-1/Grb-2/Sos
complex formation in a breast cancer model. Hum. Mol. Genet. 10,
605−616.
(7) Biedi, C., Panetta, D., Segat, D., Cordera, R., and Maggi, D.
(2003) Specificity of insulin-like growth factor I and insulin on Shc
phosphorylation and Grb2 recruitment in caveolae. Endocrinology 144,
5497−5503.
(8) Giudice, J., Barcos, L. S., Guaimas, F., Penas, A., Giordano, L.,
Jares-Erijman, E. A., and Coluccio Leskow, F. (2012) Insulin and
insulin like growth factor II endocytosis and signaling via insulin
receptor B. Cell Communication and Signaling,.
(9) Di Guglielmo, G. M., Baass, P. C., Ou, W. J., Posner, B. I., and
Bergeron, J. J. (1994) Compartmentalization of SHC, GRB2 and
mSOS, and hyperphosphorylation of Raf-1 by EGF but not insulin in
liver parenchyma. EMBO J. 13, 4269−4277.
(10) Bergeron, J. J., Di Guglielmo, G. M., Baass, P. C., Authier, F.,
and Posner, B. I. (1995) Endosomes, receptor Tyr kinase internal-
ization and signal transduction. Biosci. Rep. 15, 411−418.
441
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442

Bioconjugate Chemistry
Article
(30) Markussen, J., Halstrom, J., Wiberg, F. C., and Schaffer, L.
(1991) Immobilized insulin for high capacity affinity chromatography
of insulin receptors. J. Biol. Chem. 266, 18814−18818.
(31) Hofmann, K., Finn, F. M., Friesen, H. J., Diaconescu, C., and
Zahn, H. (1977) Biotinyl insulins as potential tools for receptor
studies. Proc. Natl. Acad. Sci. U.S.A. 74, 2697−2700.
(32) Hofmann, K., Titus, G., Montibeller, J. A., and Finn, F. M.
(1982) Avidin binding of carboxyl-substituted biotin and analogues.
Biochemistry 21, 978−984.
(33) Hofmann, K., Zhang, W. J., Romovacek, H., Finn, F. M.,
Bothner-By, A. A., and Mishra, P. K. (1984) Syntheses of biotinylated
and dethiobiotinylated insulins. Biochemistry 23, 2547−2553.
(34) Finn, F. M., Titus, G., and Hofmann, K. (1984) Ligands for
insulin receptor isolation. Biochemistry 23, 2554−2558.
(35) Hofert, J. F., Mahowald, T. N., Mahowald, N. A., and Heidrick,
M. L. (1988) The binding of FITC-insulin to ANAE-positive cells in
rat thymus. Experientia 44, 37−38.
(49) Biener, Y., Feinstein, R., Mayak, M., Kaburagi, Y., Kadowaki, T.,
and Zick, Y. (1996) Annexin II is a novel player in insulin signal
transduction. Possible association between annexin II phosphorylation
and insulin receptor internalization. J. Biol. Chem. 271, 29489−29496.
(50) Ceresa, B. P., Kao, A. W., Santeler, S. R., and Pessin, J. E. (1998)
Inhibition of clathrin mediated endocytosis selectively attenuates
specific insulin receptor signal transduction pathways. Mol. Cell. Biol.
18, 3862−3870.
(51) Hamer, I., Foti, M., Emkey, R., Cordier-Bussat, M., Philippe, J.,
De Meyts, P., Maeder, C., Kahn, C. R., and Carpentier, J. L. (2002) An
arginine to cysteine (252) mutation in insulin receptors from a patient
with severe insulin resistance inhibits receptor internalization but
preserves signaling events. Diabetologia 45, 657−667.
̈
(36) Ziegler, O., Cantin, C., Germain, L., Dupuis, M., Sekaly, R. P.,
Drouin, P., and Chiasson, J. L. (1994) Insulin binding to human
cultured lymphocytes measured by flow cytometry using three ligands.
Cytometry 16, 339−345.
(37) Rajagopal, J., Anderson, W. J., Kume, S., Martinez, O. I., and
Melton, D. A. (2003) Insulin staining of ES cell progeny from insulin
uptake. Science 299, 363.
(38) Luo, Y., Xu, H., Huang, K., Zhang, Z., Luo, Q., and Liu, Q.
(2005) Imaging on the binding of FITC-insulin with insulin receptors
in cortical neurons of rat. Conf. Proc. IEEE Eng. Med. Biol. Soc. 6,
5706−5708.
(39) Wang, H., Liu, Z. H., Li, G., and Barrett, E. J. (2006) The
vascular endothelial cell mediates insulin transport into skeletal
muscle. Am. J. Physiol. Endocrinol. Metab. 291, E323−E332.
(40) Draznin, B., Trowbridge, M., and Ferguson, L. (1984)
Quantitative studies of the rate of insulin internalization in isolated
rat hepatocytes. Biochem. J. 218, 307−312.
(41) Hachiya, H. L., Takayama, S., White, M. F., and King, G. L.
(1987) Regulation of insulin receptor internalization in vascular
endothelial cells by insulin and phorbol ester. J. Biol. Chem. 262,
6417−6424.
(42) Knutson, V. P. (1992) Ligand-independent internalization and
recycling of the insulin receptor. Effects of chronic treatment of 3T3-
C2 fibroblasts with insulin and dexamethasone. J. Biol. Chem. 267,
931−937.
(43) Formisano, P., Najjar, S. M., Gross, C. N., Philippe, N., Oriente,
F., Kern-Buell, C. L., Accili, D., and Gorden, P. (1995) Receptor-
mediated internalization of insulin. Potential role of pp120/HA4, a
substrate of the insulin receptor kinase. J. Biol. Chem. 270, 24073−
24077.
(44) Yamada, K., Carpentier, J. L., Cheatam, B., Goncalves, E.,
Shoelson, S. E., and Kahn, C. R. (1995) Role of the transmembrane
domain and flanking amino acids in internalization and down-
regulation of the insulin receptor. J. Biol. Chem. 270, 3115−3122.
(45) Wilden, P. A., Backer, J. M., Kahn, C. R., Cahill, D. A.,
Schroeder, G. J., and White, M. F. (1990) The insulin receptor with
phenylalanine replacing Tyr-1146 provides evidence for separate
signals regulating cellular metabolism and growth. Proc. Natl. Acad. Sci.
U.S.A. 87, 3358−3362.
(46) Wilden, P. A., Siddle, K., Haring, E., Backer, J. M., White, M. F.,
and Kahn, C. R. (1992) The role of insulin receptor kinase domain
auto-phosphorylation in receptor-mediated activities. Analysis with
insulin and antireceptor antibodies. J. Biol. Chem. 267, 13719−13727.
(47) Hansen, B. F., Danielsen, G. M., Drejer, K., Sørensen, A. R.,
Wiberg, F. C., Klein, H. H., and Lundemose, A. G. (1996) Sustained
signaling from the insulin receptor after stimulation with insulin
analogues exhibiting increased mitogenic potency. Biochem. J. 315
(Pt1), 271−279.
(48) Jensen, M., Hansen, B., De Meyts, P., Schaffer, L., and Ursø, B.
̈
(2007) Activation of the insulin receptor by insulin and a synthetic
peptide leads to divergent metabolic and mitogenic signaling and
responses. J. Biol. Chem. 282, 35179−35186.
442
dx.doi.org/10.1021/bc300526d | Bioconjugate Chem. 2013, 24, 431−442