Angewandte
Research Articles
Chemie
separate the responses of K+ and Na+ from that of Ca2+. The
main interference obtained at MBAPTA-E against the
determination of Ca2+ was found to be Mg2+ (ca. 16%)
because the structure of 4 carboxyl groups could also be easy
to form a chelate with Mg2+. Fortunately, METH-E demon-
strated no detectable interferent (< ca. 3.0%). Due to the
special molecular structure with two amide groups, METH-E
has a particularly capture capacity for Ca2+. After a compre-
hensive comparison for the three kinds of electrodes from
dynamic linear range, sensitivity, reversibility, and selectivity,
METH-E was chosen as the optimized biosensor because of
its superior reversibility and selectivity, suitable detection
range, and acceptable sensitivity. Next, to test the anti-fouling
ability of METH-E, it was implanted in the mouse brain for
60 days. It was found that the sensitivity of this electrode still
maintained 92 ꢂ 0.6% (n = 8, S.D; Figure S16b, Supporting
Information), compared with that of the control experiment,
proving that METH-E could work steadily for a long period
up to 2 months in a biological environment such as brain.
Finally, the biological damage of METH-E was characterized
by TTC staining experiments (Figure S16c, Supporting In-
formation). We found that the mouse brain remained intact
activity after stained with TTC after the microfiber electrode
was implanted into the mouse brain for 60 days, demonstrat-
ing that our developed microelectrode had quite low damage
and was competent for subsequent in vivo experiments.
As demonstrated above, the developed METH-E showing
high selectivity and reversibility for Ca2+ recognition with
excellent biocontamination resistance established a reliable
and durable approach for real-time tracking and continuously
monitoring the levels of Ca2+ in the live brain. Then, 7
METH-E electrodes with different lengths and internal
reference electrode were constructed into a microelectrode
array (Figure 3a,b) with the same active tip length of 300 mm,
which made it possible to simultaneously monitor the
concentration of Ca2+ in the brain regions with different
depths in the same mouse. Meanwhile, the electrophysiolog-
ical electrode was also integrated with the array to record LFP
signals. Next, the developed METH-E microarray was
implanted into 7 regions of the mouse brain (Figure 3b),
such as the primary motor cortex (M1), primary somatosen-
sory barrel) cortex (S1BF), hippocampal CA1 (CA1) hippo-
campal dentate gyrus (DG), lateral dorsal (LD), caudate-
putamen (CPu), and reticular nuclei (RT). It was the first
observation that the initial change time and decreasing rate of
Ca2+ in each brain area were different upon ischemia
(Figure 3d). Simultaneous measurements of LFP (Figure 3 f)
during ischemia showed a large suppression of discharge,
proving the ischemic state of neurons. Specifically, the
extracellular [Ca2+] changed at ca. 1.11–1.24 min in the
superficial cortical areas (M1, S1BF) and then at ca. 2.44–
2.71 min in the hippocampus (CA1, DG) after ischemia, while
the changes in the deep areas (LD, CPu, RT) response was
much later after ca. 4.44–4.75 min. Meanwhile, the decreasing
rate of [Ca2+] in superficial brain areas (M1, S1BF, CA1, DG)
was 0.25–0.31 mMminꢀ1, which was faster than that of 0.10–
0.11 mMminꢀ1 in deep brain areas (LD, CPu, RT Figure 3d).
At the end of ischemia for 20 min, the extracellular [Ca2+] was
finally decreased to 0.15–0.21 mM in superficial brain regions
(M1, S1BF, CA1, DG), much lower than that of 0.43–0.63 mM
in deep brain regions (LD, CPu, RT). These data indicated
that the sensitivity of each brain area toward ischemia was
different. Since the blood vessel density in the superficial
brain area was lower than that in the deep layer, it was
reasonable that hypoxia and Ca2+ influx firstly occur in the
superficial layer. Then, the duration of ischemia was extended
to 45 and 90 min. It was observed that extracellular [Ca2+] was
decreased to 50–60% and further to 27–45% (vs. 20 min
ischemia) at the end of ischemia for 45 and 90 min,
respectively. These in vivo data indicated that the decrease
in extracellular [Ca2+] became more severe with the pro-
longed ischemia time. The above data were collected by the
same microarray after implanted for 5 days (Figure 3c).
Next, the arterial embolism was removed for reperfusion
using the same electrode. Benefiting from the high reversi-
bility of METH-E microarray, the fluctuation of Ca2+
concentration was accurately and continuously monitored.
Specifically, taking 20 min I/R as an example, the extracellular
[Ca2+] in deep brain regions of LD, CPu and RT was
recovered after ca. 0.71–0.79 min of reperfusion (Figure 4d),
and increased to ca. 97% of the baseline value. While in the
superficial brain regions of M1, S1BF, CA1, and DG it was
slowly recovered after ca. 3.05–3.25 min of reperfusion and
eventually reached ca. 92% of the baseline value. For
reperfusion after 45- and 90-min ischemia, the extracellular
[Ca2+] of each brain area was still recovered more than 70%
of the initial value. Since the middle artery was located closer
to the deep brain area, [Ca2+] was recovered first. It was worth
noting that TTC staining of mice brain slices after 20 min I/R
showed that the infarct area was less than ca. 2% (Figure 3e),
proving the good activity of each brain area. When the
ischemia time was further extended to 45 and 90 min, despite
Ca2+ recovery was > 83% after reperfusion, a large area of
infarction (> 35%) was observed in TTC staining experi-
ments (Figure 3e). The LFP signal after reperfusion further
proved that neuronal discharge was hardly recovered after
prolonged ischemia. Our previous research had shown that
more ROS was produced after reperfusion.[15] Ca2+ was
transferred into cells upon ischemia, which generated a large
amount of ROS in mitochondria.[16] Although, as shown
above, extracellular Ca2+ was almost recovered after reper-
fusion, the produced ROS in live cells changed the intra-
cellular ATP content, which eventually led to mitochondrial
Ca2+ overload and neurons death.[17]
To further analyze the relationship between ROS and
Ca2+ overload and neuronal death, GSH, the scavenger of
ROS, was injected into the mice with 400 and 800 mgkgꢀ1
before MCAO surgery. In the 400 mgkgꢀ1 mice group upon
ischemia for 20, 45, and 90 min, the decreasing rate of [Ca2+
]
was effectively reduced to 53–71% of the control group in
M1, S1BF, CA1, DG, LD, CPu, and RTareas. Moreover, GSH
also significantly increased the recovery rate of extracellular
Ca2+ during reperfusion after ischemia (Figure 3d). After 20,
45, and 90 I/R, the [Ca2+] recovery rate of each brain area was
increased to 132–145%, 156–168%, and 185–193% of the
control group, respectively. The subsequent TTC staining
experiments showed that after ischemia for 90 min, the
infarction area (Figure 3e) for 400 mgkgꢀ1 GSH administra-
&&&&
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 11
These are not the final page numbers!