Both male and female adult (8–12 weeks old) C57BL/6J mice were used. Mice were group-housed in a 12-h light-dark cycle with food and water ad libitum. All experiments were undertaken in accordance with Institutional Animal Care and Use Committee at the University of Texas Health San Antonio (UT Health SA, protocol 20210002AP) and were in compliance with the National Institutes of Health Guide for Care and Use of Laboratory Animals. The experiments were conducted using two groups of mice. The first group were implanted with electroencephalogram (EEG) headmounts and coincidently assessed for effects on abnormal EEG waveforms and behavioral tremors using coincident video-EEG recordings. A second group of mice (without EEG headmounts) were assessed for effects on respiratory function using unrestrained whole animal plethysmography. The combined two groups were used to assay lethality.

Stock solutions of diisopropylfluorophosphate (DFP, 1.06 g/ml density, Sigma D0879) were diluted with PBS (1 μL in 4 mL PBS, 0.265 mg/ml of DFP) and vortexed for 1 min before use. The DFP solution was injected into mice subcutaneously (sc) in the right pelvic area at a volume of 8 μl per gram weight, resulting in a dosage of 2.1 mg/Kg that caused 33% (8/24) lethality without therapeutic cotreatment. A slightly higher concentration of 2.8 mg/Kg was found to cause 80% lethality (8/10) and was not used. Control saline or treatments were injected sc immediately after DFP (within 30 s) in the left pelvic area at 8 μl per gram weight to obtain concentrations of 50 mg/Kg for HI6 (Sigma SML0224), 3 mg/Kg of Atropine sulfate (Sigma A0257) and 2 mg/Kg of dequalinium chloride (Sigma PHR1300). Injections were conducted under transient isoflurane gas anesthesia. All treatment compounds were dissolved as stock solutions in sterile water and stored frozen except dequalinium chloride, which required dissolution in 100% methanol at 5 mM (2.6 mg/ml) and further dilution into saline on the day of the experiment. It is worth noting that unlike other reports (Manetta et al., 1993; Yu et al., 2020) we found that dequalinium chloride did not dissolve in aqueous solutions (even with heating and sonication).

Thirty minutes prior to administration of pilocarpine (300 mg/kg, Tocris Bioscience, Bristol, United Kingdom) EEG-implanted mice received an injection of scopolamine methylnitrate 1 mg/kg (MP Biomedicals, Santa Ana, CA, United States) that does not cross the blood-brain barrier and inhibits peripheral pilocarpine effects. Pilocarpine was delivered via intraperitoneal (I.P.) injection in 0.9% normal saline (pH 7.4). The mice were confirmed to have electrographic and behavioral seizure activity within 30 min following injection.

Mice were implanted with electroencephalogram (EEG) screw-type electrodes under general anesthesia using inhaled isoflurane (5%). A tethered EEG system with a prefabricated head-mount containing the screw electrodes was used (catalog number 8201-EEG; Pinnacle Technology, Inc., Lawrence, Kansas, United States). The head-mount included a ground screw (left frontal), and a reference screw electrode (left parietal) for right frontal (F, channel 1) and the right parietal (P, channel 2) electrodes as previously described (Bugay et al., 2020a). The recording electrodes were implanted at positions + 1.4, 2.3 mm (frontal) and + 1.4, −3.7 mm (parietal) relative to medial and bregma, respectively. Postoperatively, animals were housed individually and carefully monitored for pain and distress. Mice recovered from surgery a minimum of 48 h before exposure to DFP or treatments.

For video-EEG recordings, the head-mount was connected to a pre-amplifier (Gain 25X; Pinnacle Technologies), which in turn was connected via wires to a commutator and extracellular amplifiers of the Stellate Harmonie acquisition hardware. The system collected coincident video images of the animals during EEG activity. The EEG records were scored using a 0.1–35 Hz bandpass filter and a 60 Hz notch noise filter. Preceding chemical exposures, mice were recorded for a minimum of 2 h before they were removed and injected under transient isoflurane anesthesia with DFP alone or with treatments at dosages described above. The mice were quickly returned to the recording chamber and monitored for an additional 24 h. Brain activity was assayed visually by the investigator and also by the computer-assisted EEG stellate harmonie review software.

For each experiment, a period of 1-h EEG recording before drug injection was used as a control. A period of 1-h EEG after drug injection was used for analyzing the drug’s acute effects and after 23-h EEG recording the 1-h EEG period was used for analyzing recovery. After visual inspection of the video-EEGs, 4-s epochs EEG power spectra were calculated by Fast Fourier Transform (FFT). FFT lengths of 5.1 s and Spectral Restart Intervals of 10.24 s were used. Mean square root power values (μV/Hz) were then normalized to the total power to obtain relative values comparable among the different animals.

Statistical significance was determined depending on the experimental design. The unpaired t-test, one-way or two-way ANOVA with repeated measures followed by Tukey’s multiple comparison test were performed using GraphPad Prism 9.0. All data are expressed as mean ± standard error of the mean (SEM). Statistical significance was considered at p < 0.05.

To assay tremors, an experimenter blinded to the treatment performed the behavioral classification. Visual inspection of the video recordings (removed of EEG data) for the 1 h period following injection of DFP alone or DFP with treatments were scored. Animals were scored 1 if no tremors were apparent, 2 if rare tremors were apparent, 3 if tremors occurred approximately through 50% of the recording period, 4 if the mice had rare periods without tremors, and 5 if tremors occurred continuously. It is worth noting that control animals having vehicle injections (Figure 1) or anticholinergics in the absence of DFP (data not shown) had no detectable tremors. The mice were not scored on a Racine scale since we did not see any other behaviors that are typical of seizures such as rearing and falling, clonus, head bobbing, running or jumping.

Unrestrained whole animal plethysmography of mice was conducted using the BUXCO system (Buxco Research Systems, Wilmington, NC, United States) according to previous procedures (Lomask, 2006). Plethysmography chambers were calibrated for 1 ml of flow preceding each experiment. Mice were allowed 60 min to acclimate to the chambers. For chemical exposures, the mice were removed from the chambers and injected with DFP alone or DFP with treatments as described above. Injections were conducted during acute isoflurane anesthesia, where the mice recovered approximately 20–30 s after chemical injections before returning to the plethysmography chambers. The mice were recorded for an additional 1 h following injections and then for 1 h the following 22–24 h after injections to assay recovery.

In order to understand the neuronal and respiratory pathophysiological effects of organophosphate toxicity and their therapeutic interventions, we employed a concentration of DFP that was below the LD₅₀ (2.1 mg/Kg, 0.64 LD₅₀) based on previous studies (Tripathi and Dewey, 1989). This allowed us to study both acute (<1 h) and sustained effects (22–24 h) during the recovery period. Using this concentration, we found that 67% (16/24) of animals survived 24 h after DFP exposure. In contrast, standard treatments (Hulse et al., 2019; Alozi and Rawas-Qalaji, 2020; Aman et al., 2021) delivered coincident with DFP almost fully eliminated lethality. This included the muscarinic antagonist Atropine (3 mg/Kg), which promoted 94% (16/17) survival, and the acetylcholinesterase reactivator HI6 (50 mg/Kg), which caused 100% (16/16) survival. We also tested a non-specific anticholinergic, dequalinium chloride (2 mg/Kg), which antagonizes both muscarinic (Bugay et al., 2020b) and nicotinic receptor activation (Dunn, 1993). It was 100% effective (11/11 survived) in eliminating lethality.

Tremors and involuntary or uncontrollable movements are an obvious symptom of organophosphate poisoning (Reji et al., 2016). Behavioral scoring was used to quantify tremor severity (scored in a range from 1 for no tremors to 5 with constant tremors). In the absence of DFP, an observer blind to treatments scored all animals 1 for severity (Figure 1). Treatment with DFP revealed obvious tremors during the first hour of exposure (Figure 1, tremor score 3.5 ± 0.41). The severity of tremors declined following 22–24 h of recovery (Figure 1, 1.91 ± 0.21, P = 0.0004), but was still elevated above control levels (P = 0.007). Both atropine (2.2 ± 0.28, P = 0.014) and dequalinium chloride (1.8 ± 0.30, P = 0.013), but not HI6 (2.9 ± 0.35, P = 0.52), significantly reduced DFP tremors within 1 h following DFP exposure (Figure 1). During the recovery phase (22–24 h after exposure), dequalinium chloride (1.0 ± 0, P = 0.007), but not atropine (1.5 ± 0.16, P = 0.24) or HI6 (1.4 ± 0.52, P = 0.11), showed significantly better recovery than recovery following DFP alone exposure (Figure 1).

Electroencephalogram records across three time periods were analyzed: a control 1 h period immediately before, 1 h immediately after subcutaneous DFP injection (2.1 mg/Kg), and a 1 h period at 22–24 h after DFP injection (recovery period). Example records are shown in Figure 2A with traces for frontal (F) and parietal electrodes (P) taken before (Figure 2A1), immediately following (Figure 2A2), and 22–24 h after injection of DFP (Figure 2A3). Before DFP injections, the EEG traces at the frontal and parietal electrodes are very similar and show normal activity (Figure 2A1). For the frontal electrode, DFP had little effect by visual inspection (F electrode, Figure 2A2). For the parietal electrode, DFP injection (P, Figure 2A2) caused a prominent and sustained synchronization of cortical activity of ∼1–2 Hz tonic spikes. The focal spikes, although abnormal and rhythmic, do not show the evolution of amplitude and frequency seen of typical electrographic seizure activity of other chemoconvulsants (Fisher et al., 2014; Supplementary Figure 1). The EEG pattern was apparent for long durations (average duration of 171 ± 33 min per mouse, N = 11). For periods where abnormal electrographic activity was or was not present, there was no correlation with tremor behavior. The following day, the 1–2 Hz tonic spikes pattern seen in the parietal electrode had dissipated, although the EEG trace showed significantly reduced amplitudes (Figure 2A3). It is worth noting that detection of conventional seizures evoked with muscarinic agonist pilocarpine using the electrodes employed here showed generalized electrographic seizures that appear on both forebrain and caudal parietal electrodes (Supplementary Figure 1). Thus, the focal, parietal-specific spike pattern appears to be unique to the DFP model using sublethal concentrations.

Power spectrum analysis of a 15 min sampling of regions of abnormal activity provides a qualitative assessment of the EEG abnormalities observed during DFP exposure (blue trace, Figure 2B). For comparison, the graph also includes power analysis of the EEGs before (black trace) and 22–24 h following (brown trace) DFP exposure (Figure 2B). Two major changes are clearly discernable in the parietal electrode (Figure 2B, Parietal). DFP causes an enhancement in the delta region (1–4 Hz) spike patterns with highest power at 1.4 Hz that declines the following day after recovery (Figure 2B, Parietal). In addition, there is a dramatic reduction of the theta (4–8 Hz) through alpha range (8–13 Hz) following DFP exposure that is not recovered the following day (Figure 2B, Parietal). In the frontal electrode, acute DFP did not cause an increase in the delta frequencies (Figure 2B, Frontal) but did cause a decrease in power across the 2–11 Hz frequencies that recovered the following day (Figure 2B, Frontal). We also conducted power spectra analysis across a longer 1 h period following DFP injection, which is less biased for abnormal events than a 15 min analysis. The power spectra for DFP indeed shows a significant increase in the 1.4 Hz peak and decline in theta frequencies (Supplementary Figure 2).

Effects of organophosphate toxicity on respiration has been well-documented (Hulse et al., 2014), and indeed respiratory failure is proposed to be a cause of organophosphate mortality (Hulse et al., 2019; Aman et al., 2021). Therefore, we examined the effect of DFP and treatments on mouse respiratory mechanics using identical timelines and drug concentrations as used in the EEG studies. The mice were studied using non-invasive whole body plethysmography, where others have documented effects of organophosphate on respiratory mechanics (Villa et al., 2007; Houze et al., 2008, 2018, 2019). For these studies, we used breathing rate and tidal volumes as reporters for DFP respiratory toxicity.

Similar to EEG studies, we assayed breathing mechanics (Figure 3) for 1 h periods before treatments (control, black traces), immediately after treatments (acute effects, blue traces), and 22–24 h following treatments (recovery phase, brown traces). Preceding DFP injections, control measurements show a decline in breathing rate and tidal volumes as mice acclimate to the chamber (Figures 3A–C). For comparison, we therefore selected the average of the last 5 min of each 1 h period where tidal volumes approach a plateau (Figure 3B). These are plotted as summary data in Figure 4. Transient removal of the mice from the chambers and DFP injection caused a dramatic reduction in tidal volumes and breathing rate (Figures 3D–F, respectively) that was well below values for mice injected with saline alone (Figures 3A–C, summarized in Figures 4C,D). The following day, mice did not fully recover and had depressed tidal volumes and breathing rates that were approximately intermediate to that of the control and DFP injection (Figures 4C,D). Atropine treatment did not rescue acute effects of DFP, although it promoted recovery with regard to tidal volumes (Figure 4F) but not breathing rates (Figure 4E). Both HI6 and dequalinium had the most protective effects on respiratory mechanics. They had acute protective effects with regard to breathing rates, but not tidal volumes, and promoted recovery in both parameters (Figures 4G–J).

As controls, we conducted plethysmography studies for effects of the treatments in the absence of organophosphate (Supplementary Figure 3). In all cases, the treatments alone lacked significant effect during or following recovery relative to pretreatment controls. In order to compare plethysmography findings between treatment groups, we also normalized the plethysmography data to control values before DFP exposure (Supplementary Figure 4). Comparisons between treatments groups (Supplementary Figure 4) yielded results that were consistent to non-normalized data (Figure 4). Both HI6 and dequalinium were significantly improved compared to DFP alone (p < 0.05) for breathing frequency acutely and following recovery, and improved recovery of tidal volumes (Supplementary Figure 4). However all treatments were not significantly different than DFP alone for acute effects on tidal volume (Supplementary Figure 4). In summary, although none of the treatments appear to protect fully to acute DFP exposure, some treatments such as HI6 and dequalinium improved recovery of respiration.