Journal of the American Chemical Society
Article
incubated with biologically relevant metal ions for 30 min at
room temperature and then the premixed solutions were
further incubated with 1 μM of CD649 for 30 min. Fluorescent
gel analysis showed that CD649 selectively labels the protein in
case of Cu(I)/Cu(II) and the labeled band became
significantly diminished upon addition of BCS (structure is
shown at Figure 2b), supporting the notion that the
nucleophilic amino acids on a protein surface were labeled
with CD649 via acyl transfer reaction in the presence of copper
ions (Figure 1e). We further checked the CD649 selectivity
with HEK 293T lysates incubated with Cu(I) and Fe(II)
prepared in the bioglove box, which exhibits the comparable
labeling capability of CD649 toward Cu(I)/Cu(II) labeled
protein (Figure S5). To investigate any oxidation state
specificity of CD probes toward copper, lysozyme in PBS
was premixed with Cu(I)/Cu(II) for 5 min, followed by
incubation with CD649 for 2 h under a N2 atm so that no
oxidation of Cu(I) would occur. Similar fluorescence
intensities of labeled lysozyme were found in the presence of
Cu(I) and Cu(II) ions, respectively (Figure S10), indicating
that the CD probes are responsive toward both Cu(I) and
Cu(II) ions.
In addition, we found a linear increase in fluorescence
intensity of HEK 293T cell lysates labeled by CD649 with
increasing copper ion concentration, and a limit of detection is
found to be 0.864 nM (Figure S8). These data suggest that
protein labeling by the CD probes is highly sensitive and dose-
dependent toward copper. A dose-dependent increase in
fluorescence intensity of lysozyme labeled by CD649 was also
found with increasing copper, whereas changes in fluorescence
intensities with fixed copper ion concentrations and varying
lysozyme concentrations were much smaller and plateaued at
high lysozyme concentrations (Figure S9). In view of the much
higher concentrations of protein compared to CD probe in
biological environments, the effect of changes in protein
concentration on observed fluorescence should be minimal.
Finally, to evaluate the relative copper affinity of the CD
probes, CD433 was employed in a competitive copper binding
assay with the reported copper fluorescent probe, CSR1, whose
emission profile does not overlap with that of CD433.80,81 In
these experiments, CD433 (0−5 equiv) was continuously
added to a pH 7.0 buffer solution (25 mM HEPES) containing
the copper-complexed form of CSR1, which is reported to
have a dissociation constant (Kd) with Cu(I) of 2 × 10−13 M.59
The solution containing the probes was excited at 608 nm,
where CD433 does not absorb, and as the concentration of
CD433 was increased, the observed fluorescence intensity
derived from CSR1 significantly decreased, suggesting that
CD433 can compete with CSR1 for copper binding and is
capable of displacing the copper ion from this Si-rhodol sensor
(Figure S6). As CSR1 is capable of monitoring dynamic
changes in labile copper pools in live cells as shown by
identification of copper fluxes induced by lipolysis,59 this set of
experiments presages that the CD platform should have
sufficient affinity for copper in related cellular assays.
by preincubating with Cu(gtsm) to increase intracellular
copper levels and BCS to decrease intracellular copper levels.
As anticipated, we observed a robust 2-fold increase in
fluorescence intensity following treatment with 2 μM Cu-
(gtsm), with the labeling being distributed evenly across the
cytosol as compared to the punctate staining pattern observed
under basal conditions. In contrast, HEK 293T cells pretreated
with 200 μM BCS for 10 h to induce copper depletion
displayed a significant decrease in fluorescence intensity as
compared to control cells as shown with CD649 imaging
(Figure 2a), establishing that this ABS probe can monitor
changes in labile copper levels in live cells, with fairly low cell
toxicity as indicated by PrestoBlue assay (Figure S11).
Likewise, CD433 was incubated with HEK 293T cell lysates
in the presence of 1 equiv of Cu(I), and fluorescent intensity
was measured at different incubation times up to 120 min.
Those experiments show that the integrated fluorescence
linearly increased and the fluorescence was quenched by
preincubation with BCS (Figure S7), which confirms the
copper-responsiveness of this probe. Next, we sought to
evaluate the ability of CD649 to assay labile copper pools in
genetic models of copper misregulation. In particular, we
focused on matched cell lines with different levels of ATP7A, a
copper-translocating P-type ATPase that regulates cellular
copper pools by mediating cellular copper egress and secretory
function. Indeed, mutations in the copper transporters ATP7A
and ATP7B are at the heart of genetic disorders of copper
homeostasis.26,82−84 In particular, Menkes disease is derived
from a loss of ATP7A function with systemic copper
misregulation leads to hyperaccumulation of liver copper.85
As anticipated, CD649-stained ATP7A knockout (KO) cells
showed significantly higher levels of fluorescence compared to
control ATP7A wild-type (WT) cells, revealing that CD649
can indeed detect elevated levels of labile copper in ATP7A
KO cells, presumably due to the deficiency in cellular copper
export (Figure 2d,e). Furthermore, both WT and KO cells
showed an increase in fluorescence signal following incubation
with Cu(gtsm) in a time- and dose-dependent manner, which
can be attenuated by the addition of the copper chelator BCS
(Figure 2b). These data validate that CD649 can detect
changes in labile copper levels with pharmacological and/or
genetic manipulation.
CD649 Enables Activity-Based Sensing of Copper in
Major Brain Cell Types. After confirming that CD649 can
monitor changes in labile copper pools in mammalian cell line
models, we then moved on to apply this new chemical tool to
probe labile copper pools relevant to the brain and central
nervous system. Indeed, copper homeostasis has profound
effects in the brain as a neuromodulator in healthy
states12,14,86−89 and as a mediator of neuroinflammation and
neurotoxicity in aging and disease states.6,14,90−93 In this
context, previous work from our laboratory using a
combination of X-ray fluorescence microscopy (XRFM) and
a reversible binding-based BODIPY fluorescent probe for
copper (CS3) identified that neuronal activity triggers
significant movements of labile copper pools from the soma
to peripheral neuronal processes.13 Here, we sought to expand
on this observation by providing foundational information on
labile copper pools in three main types of cells in the brain for
comparison: neurons, astrocytes, and microglia. Starting with
neurons, CD649 revealed punctate staining localized primarily
to the soma of cultured primary hippocampal neurons in their
resting state (Figure 3b). Activation and depolarization of
Live-Cell Imaging of Labile Copper Pools with CD649
Under Situations of Copper Elevation or Depletion.
After establishing that CD probes are sensitive to copper with
high metal selectivity and can label a model protein by SDS-
PAGE analysis with a fluorogenic readout, we next tested the
ability of this ABS platform to respond to changes in labile
copper levels in living cells. Specifically, we applied CD649 to
live HEK 293T cells and cellular copper levels were perturbed
E
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX