Letter pubs.acs.org/chemneuro
Insertable NO/CO Microsensors Recording Gaseous Vasomodulators Reflecting Differential Neuronal Activation Level with Respect to Seizure Focus Yejin Ha,† Youngmi Lee,*,† and Minah Suh*,‡,§,∥ †
Department of Chemistry and Nano Science, Ewha Womans University, Seoul 03760, Republic of Korea Center for Neuroscience Imaging Research, Institute for Basic Science (IBS), Suwon 16419, Republic of Korea § Department of Biomedical Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea ∥ Samsung Advanced Institute of Health Science and Technology (SAIHST), Sungkyunkwan University, Suwon 16419, Republic of Korea ‡
ABSTRACT: Nitric oxide (NO) and carbon monoxide (CO) are important signaling molecules shaping vasomodulation. This paper reports simultaneous in vivo monitoring of NO, CO and dendritic summation of action potential at three different cortical regions: seizure focus and two additional places, vertically and horizontally separated by 1.2 mm from the seizure focus, during epileptic seizure induced by 4-aminopyrindine injection. An amperometric dual microsensor having a high spatiotemporal resolution monitored fast and dynamic changes of NO and CO, and neural changes were recorded with a glass pipet electrode for local field potential (LFP). At all three locations, onsets and offsets of NO and CO changes well synchronized with fast LFP changes, while the patterns and concentrations of NO and CO changes were varied depending on the sensing locations. The insertable NO/CO dual microsensor was successful to measure intimately linked NO and CO in acute seizure events with high sensitivity, selectivity, and spatiotemporal resolution. KEYWORDS: Nitric oxide, carbon monoxide, local field potential, amperometric sensor, epileptic seizure
T
Despite the importance of NO and CO in regulating brain hemodynamics, few studies have been done to show interactive dynamics of two neurotransmitters. Recently, Reis et al. reported antagonistic interaction between CO and NO in neurons during water deprivation. They studied NO and CO indirectly using immunohistochemical analysis of nNOS and HO-1.14 Likewise, indirect and postmortal analyses based on immunohistology have been widely used for NO and CO studies in the brain. Although these methods provide the information regarding whether NOS and HO are enhanced or blunted under a certain condition, it is difficult to observe dynamic in vivo changes of NO and CO gases directly. Various methods have been used for direct measurements of NO and CO such as fluorescence, chemiluminescence, electrochemical methods, and so forth.15,16 Since the half-lives of NO and CO are short and they are easily diffused, analytical techniques for real-time direct measurements of NO and CO require short response time and low detection limits.16,17 Electrochemical method is one of the most powerful tools for real-time and direct analyses of rapidly degradable molecules such as NO and CO.
wo signaling gases, nitric oxide (NO) and carbon monoxide (CO), share common properties and physiological roles. NO and CO are endogenously released from cells via enzyme activity of nitric oxide synthase (NOS) and heme oxygenase (OH), respectively.1,2 Both NOS and HO have a few isoforms: NOS includes endothelial NOS (eNOS), inducible NOS (iNOS), and neuronal NOS (nNOS), and HO includes an inducible form (HO-1) and two constitutive forms (HO-2 and HO-3).1,2 NO is involved in neuroprotection, neurotransmission, vasodilation, immune, and inflammatory actions depending on its concentration.3−6 CO also acts as neurotransmitter, vasomodulator, anti-inflammatory and antiapoptotic factor.2,7,8 Although it is known that NO and CO play similar biological roles, the interactions between the gases are still controversial. The opposite observations, inhibition and up-regulation of HO by NO, have been reported.9 This different behavior seemed to depend upon the kind of HO isoform targeted, cellular concentration of HO compared to NOS, and characteristics of NO donors. On the other hand, the relationship between CO and NOS was relatively clear. Although the underlying mechanisms were different by the type of NOS and HO isoforms, high-concentration CO inhibited NOS and lowconcentration CO induced NO releases by inhibiting HO. In the more recent studies, antagonistic relations between the specific NOS and HO isoforms have been reported.10−13 © XXXX American Chemical Society
Received: April 17, 2017 Accepted: June 29, 2017 Published: June 29, 2017 A
DOI: 10.1021/acschemneuro.7b00141 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Letter
ACS Chemical Neuroscience
Figure 1. Schematic diagram of experimental setup for location dependent measurements of NO and CO at the spot O, V, and H (θ, 45°; ϕ, 90°; l, 1.2 mm; r < 1 mm).
1.2 mm underneath the brain surface. In contrast, NO and CO were measured at three different locations by positioning a NO/CO dual microsensor in a close proximity to the LFP recording electrode and nanoinjector position (O); at the brain surface (denoted as V, 1.2 mm separated vertically from the spot O); and at another spot in the layer 5 (denoted as H, horizontally separated from the spot O by 1.2 mm). The representative responses of NO and CO changes at the spots O, V, and H are shown in Figure 2. Each seizure was divided into three periods:20 seizure initiation, propagation, and termination period. In each period of the seizures, NO and CO changes are tightly correlated with each other. In the seizure initiation period, NO and CO levels started to increase from their basal levels simultaneously as the seizure initiated. The ascended levels were maintained above the basal lines throughout the propagation period. During the termination period, NO and CO levels declined and returned to their basal levels just like the LFP signals. However, the detailed inspection of NO and CO changes revealed slightly different patterns depending on the sensing location. NO and CO of in-tissue measurement at a seizure focal area (spot O, Figure 2a) increased sharply and made peaks through seizure initiation, and they showed steady-state levels in the propagation period. The NO and CO measurements at a cortical surface (spot V, Figure 2b) and in-tissue at a seizure extrafocal area (spot H, Figure 2c) showed the gradual increases followed the decrease during the propagation period. NO and CO decreases were continued to the termination period producing the modest mountain shapes. In the initiation period, peak-shaped changes of NO and CO appeared at the spot V, while they were hardly observed at the spot H. Regardless of the sensor position (spots O, V and H), the onset time of NO and CO concentration changes coincided with that of LFP signal.
In previous research, our group developed an amperometric NO/CO dual microsensor for real-time in vivo detection of NO and CO.16 The sensor with high spatiotemporal resolution was applied for monitoring the levels of NO and CO in a living rat’s cortical region during acute seizures induced by 4aminopyridine (4-AP).18 Spontaneous seizure is one of most powerful events in brain and is manifested in abrupt changes in both neuronal activation and hemodynamics.19 Measurements of NO, CO, and neuronal dendritic potential were concomitantly conducted at seizure focus and they showed tight links one another.16 The levels of NO and CO were increased and decreased as epileptic seizure was progressed and terminated. The onset and offset of NO and CO releases were synchronized with the ones of dendritic potentials. Thus, real-time measurements of NO and CO may effectively represent neuronal activation pattern during acute seizures. To better understand the hypersynchronous and hyperexcitable neuronal events in relation to endogenous NO and CO, we demonstrate location-dependent measurements of NO and CO in current study. Changes of NO and CO levels were monitored at three different regions (i.e., near the 4-AP injection spot at layer 5, a spot at cortical surface, and another spot at layer 5) using a NO/CO dual microsensor previously developed.16
■
RESULTS AND DISCUSSION
NO and CO concentrations are in tight coupling with epileptic seizures. We previously showed measurements of a single dual NO/CO microsensor near epileptic focus closely resembled neuronal dendritic potentials during spontaneous seizures.16 In this study, we further examined whether NO and CO measurements were location-dependent in relation to seizure focus. In Figure 1, a local field potential (LFP) electrode and a nanoinjector were positioned at O in cortical layer 5, which is B
DOI: 10.1021/acschemneuro.7b00141 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Letter
ACS Chemical Neuroscience
Pattern V2. The CO changes in Patterns H1 and H2 were similar to each other, showing a featureless gradual increase followed by the decrease. In contrast, the NO change in Pattern H2 was peculiar: a broad peak, a steady-state region and a concave recovery were observed in initiation, propagation, and termination periods, respectively. Patterns V1 and V2 occupied 75% and 25% each (seizure n = 29; animal n = 3), and Patterns H1 and H2 were 56% and 44%, respectively (seizure n = 34; animal n = 3). One of the most significant differences among three different locations was the levels of NO and CO concentration changes, which were obtained with following equation. ΔNO(or CO)x = |NO(or CO)peak, x − NO(or CO)base, x | (1)
where NO(or CO)peak,x is the peak concentration of NO(or CO) observed in the initiation or propagation period (the average value of 98 to 100% of the highest data), and NO(or CO)base,x is the base concentration of NO(or CO) (the average value of 30 s prior to seizure initiation. ΔNO(or CO)x is the absolute difference between NO(or CO)peak,x and NO(or CO)base,x. The subscripted x is O, V or H, indicating the NO/ CO measuring site. ΔNO and ΔCO values at the spots O, V, and H are shown in Table 1 and Figure 4 with statistical analysis. The measured values of ΔNO or ΔCO at a single measuring site were similar regardless of the different patterns with an exception of ΔNOH. Therefore, ΔNO and ΔCO values were statistically analyzed only depending on the measuring location; but ΔNOH values for H1 and H2 were independently analyzed as shown in Figure 4 and Table 1. NO and CO changes at seizure focus, ΔNOO and ΔCOO, were 12.83-fold and 3.32-fold higher than the ones at cortical surface, ΔNOV and ΔCOV, respectively. The ratio of ΔNOO to ΔNOV was much greater than that of ΔCOO to ΔCOV. It is assumed that the difference is related to the molecular properties of the gases. Although both CO and NO are oxidizable gases, NO is more reactive and has a shorter lifetime than CO.21 Interestingly, NO and CO changes measured at the spot H (in tissue at seizure extrafocal area) were different to each other. As shown in Figure 4a, ΔCOH was as high as ΔCOO, and there was no significant difference between them (p > 0.05, paired t-test). However, ΔNOH was different depending on the observed patterns as presented in Figure 3. NO change in Pattern H1 (ΔNOH1) was small likewise ΔNOV, but that in Pattern H2 (ΔNOH2) was as high as ΔNOO (Figure 4b). According to current and previous works,16 it is obvious that NO and CO are closely linked to robust neuronal activities and their concentrations are simultaneously changed with dendritic potentials changes as presented in LFP responses, of seizures. Spontaneous NO and CO changes during epileptic seizures may reflect complicated neuronal activation patterns in cerebral cortex. Locally induced seizures spread to adjacent areas vertically and horizontally through different brain regions, horizontal and vertical connection of cortical network.22 The horizontal spread via horizontal-connection occurs mainly through layer 5 of the cortex.22,23 As shown in Figure 4 and Table 1, for the vertically distant measurements at cortical surface, the values of ΔCOV and ΔNOV were much smaller than those of on-site measurements (ΔCOO and ΔNOO). On the other hand, for the horizontally distant measurements at the spot H, the value of ΔCOH was observed in a large magnitude,
Figure 2. Representative dynamic responses of LFP, NO and CO at (a) focal seizure area in layer 5, spot O; (b) brain surface, spot V; and (c) extrafocal area in layer 5, spot H.
In fact, a few distinctive patterns of NO and CO changes were observed at each sensing location including the representative examples shown in Figure 2. As shown in Figure 3a, the in-tissue measurement at seizure focal area (spot O) showed three different patterns (denoted as Patterns O1, O2, and O3). Pattern O1 was similar to the representative pattern as in Figure 2a, while Patten O2 had no sharp peak in the initiation period for NO. Pattern O3 showed smooth arch shapes for both of NO and CO during an overall seizure event. The relative observances of Patterns O1, O2, and O3 were 42%, 33% and 25%, respectively, among 32 seizures in 4 animals. Both the cortical surface measurements at the spot V and in-tissue measurements at an extrafocal area (spot H) showed two different patterns each (Figure 3b, denoted as Patterns V1 and V2 and Figure 3c, denoted as Pattern H1 and H2). Patterns V1 and H1 were the major patterns at the spot V and H, respectively. In contrast to Pattern V1, sharp NO and CO peaks of the initiation period were not clearly recognized in C
DOI: 10.1021/acschemneuro.7b00141 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Letter
ACS Chemical Neuroscience
Figure 3. Representative patterns of LFP, NO, and CO at (a) focal seizure area in layer 5, spot O; (b) brain surface, spot V; and (c) extrafocal area in layer 5, spot H.
Table 1. Values of ΔNO and ΔCO (Mean ± SD) Measured at Focal Seizure Area in Layer 5 (spot O), Brain Surface (spot V), and Extrafocal Area in Layer 5 (spot H) NO/CO measuring site O V H1 H2
ΔNO (μM)
ΔCO (μM)
± ± ± ±
1.93 ± 0.76 0.58 ± 0.24
0.246 0.021 0.027 0.217
0.103 0.016 0.007 0.043
population of neurons than the superficial layer. Also, horizontal-connection is more abundant in layer 5 than the superficial layers, therefore seizure propagated mainly is more profound in layer 5 than superficial layers. Our concomitant NO/CO measurements at different locations in cortex may reflect different architectures of cortex and propagation patterns of seizures. The detailed patterns of NO and CO changes, i.e., peaks in initiation period, shapes of the propagation period, might reflect complex activities of a variety of cells related to neurovascular coupling (e.g., neurons, astrocytes, endothelial cells) releasing NO and CO as signaling molecules.24 However, to fully interpret the differences of NO and CO changes among
1.95 ± 0.45
similar to that of ΔCOO. ΔNOH2 value was also high enough to be comparable to ΔNOO, while the magnitude of ΔNOH1 was 9.3-fold smaller than ΔNOO. This might be caused by different neuronal population in cortical layers. Layer 5 has more dense D
DOI: 10.1021/acschemneuro.7b00141 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
Letter
ACS Chemical Neuroscience
Figure 4. Values of (a) ΔNO and (b) ΔCO at focal seizure area in layer 5 (spot O), brain surface (spot V), and extrafocal area in layer 5 (spot H) based on the values in Table 1. ***p value
