Fig 1.
Classification of swimming strategies.
A) Schematic drawing of the water maze pool setting showing the location of platform and cues. Animals were placed at either of 3 entries of the pool. The annulus zone and the large circular area around the platform (PTF) were used to identify thigmotaxis, chaining or focal search as previously descried [54]. B) Swimming strategies used by the rats were classified in 8 categories (based on modified criteria proposed by Garthe et al. [54]) and were given a score from 0 to 7. Thigmotaxis, Random search and Scanning are regarded as spatial memory independent, or non-spatial strategies, while Directed, Focal, and Direct search were considered as spatial memory dependent. Chaining was scored separately. A score of 0 was given to the animals unable to swim. The traces are actual swimming patterns observed in our control rats.
Fig 2.
Experimental design and outcomes.
Twenty-one control rats received saline IP injection and were assigned to behavior test (control rats). A first series of 32 rats received H2S injection, 8 rats did not present a coma. The 24 rats that presented a coma, were separated into a non-treatment (H2S) group (12 rats) and methylene blue treated (H2S-MB) group (12 rats). In the H2S group, 8 died immediately (*1 out of 8 rats died within 12 h) and 4 survived from a coma (one these 4 rats was however unable to swim). Meanwhile 9 survived in the H2S-MB group, they were all able to swim. Therefore, to match the number of surviving rats between the 2 groups, a second series of 10 rats received NaHS: 5 died and 5 survived. The 5 surviving rats were added to the H2S group, one of them could not swim. Out of 9 surviving rats in the H2S group, 7 could therefore complete the behavior test, while the 2 rats unable to swim were euthanized within 48 h. Out of 9 surviving rats in the H2S-MB group, 9 completed the behavior test.
Fig 3.
Immediate outcome in keeping with the number of intra-peritoneal NaHS injections required to produce a coma.
There was no difference in the mortality whether 1, 2 or 3 injections were administered. Of note is that the 2 rats that were unable to swim in the H2S groups belonged to the rats that received 3 IP injections.
Fig 4.
Body weight following H2S induced coma.
Body weight decreased gradually over the 4 days of training in the control group. In the H2S group, body weight significantly dropped at D1 day (significantly different from control group, P<0.05), and then did not change thereafter. Note that the 2 rats unable to eat that were euthanized are not included in this computation. In the H2S-MB group, weight also significantly dropped at D1 (significantly different from control group, P<0.05) then remained below baseline until D4. Values are shown as mean ± SD. *significantly different from control at P<0.05. # Significantly different from baseline (day 0) at P<0.05.
Fig 5.
Latency to reach the platform, distance, and path efficiency during Morris water maze testing.
Data shown here correspond to the results obtained in all the rats that were able to swim and to find the platform the 4 tests. All rats decreased the latency and distance to locate the platform and increased the path efficiency throughout the 4 days of training with no difference between groups. In the H2S group, the changes were significant only in for the distance at D3 and D4, and for the path efficiency at D4, due to large standard deviations. Values are shown as mean ± SD. #significantly different from day 1 at P<0.05.
Fig 6.
Average latency, distance, time spent in the platform quadrant, and number platform crossing are displayed during the probe trial. No significant difference was observed between the 3 groups. Values are shown as mean ± SD.
Fig 7.
Frequency distribution of the strategies used during the MWM test.
Panel A: Coding color showing the search strategy used by the Control group (top), H2S group (middle), and H2S-MB group (bottom) (based on coding proposed by Garthe et al. [54]). Of note, is that the two rats in the H2S group unable to swim (#16 and 38, see Table 1) are identified with the dotted line after D2, as they were euthanized at 48 hours. Also in the H2S-MB group, 2 rats, although able to swim were only showing a thigmotaxis pattern during the 4 days of training are identified. Panel B: Radar chart displaying the relative frequency of the various strategies used at D4 by the 3 groups. In the H2S group, the animals used a significantly less effective pattern consisting in higher occurrence of scanning strategy to find the platform when compared to the treated group.
Fig 8.
Daily evolution of the spatial dependent and non-dependent strategies used during the MWM test.
Data of the Control group (top), H2S group (middle), and H2S-MB group (bottom) are displayed. All groups showed a progression towards more spatial patterns during the training phase, but spatial strategy was used much less often in the H2S group.
Fig 9.
Examples of swimming strategies in 2 intoxicated rats treat with MB.
The recordings in the upper panels where obtained from a rat with no brain lesions while those in the lower panels were obtained in the rat #92, (Table 1) that was later found to have neuronal cortical necrosis. The latter rat was unable to find the platform even though its behavior was normal during open field test.
Fig 10.
Sections of frontal cortex (panels a and b) and thalamus (c) from one rat that presented a coma but with no neurological deficit (a1, b1, c1) and from rat (#16, Table 1) that was unable to swim after H2S exposure (a2, b2, c2). In contrast to rat with no symptom, the brain of rat #16 showed diffuse and extended neuronal necrosis and neuropil edema affecting the outer frontal cortex (motor agranular cortex) and the cingulate gyrus (anterior limbic area). Neurons are hypereosinophilic with karyolytic or pyknotic nuclei and peri-nuclear edema, Bregma 0.0. 400x. Panel c1 shows normal thalamus at same level and magnification in the intoxicated rat with no deficit. Panel c2: Extensive neuronal necrosis in the lateral posterior nucleus of the thalamus. Bregma -4.8. 400x.
Fig 11.
Histopathology of the hippocampus and piriform cortex in the rat # 38 that displayed severe neuronal necrosis in the frontal, temporal, parietal and occipital cerebral cortices.
The same pattern was found in all the rats presenting cortical lesions. a) Note the diffuse acute neuronal necrosis and neuropil edema affecting the outer retrosplenial and occipital cortex, while the hippocampus was completely unaffected. Bregma -4.5. 40x; b) Note the sharp demarcation at the rhinal fissure between the necrotic temporal cortex (top) and the unaffected piriform cortex (bottom). Bregma -4.5. 100x.
Fig 12.
GFAP staining by DAB chromagen with hematoxylin counterstain in one rat exposed to saline (control) and one rat intoxicated with H2S with normal recovery.
There was no difference in astrocyte density or morphology between the 2 animals.
Table 1.
Summary of location of the various brain lesions.
Fig 13.
Histology of the frontal cerebral cortex in one rat of the H2S-MB group (#98, see Table 1) that could swim and find the platform.
Marked asymmetry of neuronal necrosis in the left hemisphere was found, while the right hemisphere was nearly normal (Bregma 0.0. 40x). All other lesions in this rat (Table 1) were ipsilateral.
Fig 14.
Average swimming strategy score in all the intoxicated rats with and without brain lesions following H2S induced coma in non-treated animals (red symbols) and in animals receiving MB (blue symbols).
The swimming strategy was scored according to the swimming pattern described in the Fig 1 (the rats unable to swim were scored as zero), the score was averaged over the last 2 days of training. The score of the rats reaching the platform was significantly higher in the treated group than in the non-treated animals (*P<0.01, see text for further details). The ID numbers of the rats presenting with brain lesions (corresponding to the ID numbers in Table 1) are shown. In all rats but one, lesions were highly predictable from the clinical picture.
Fig 15.
Example of the histopathology of the lung in a control rat and a rat exposed to H2S.
Hematoxylin and eosin stained sections of perfused lungs from control and H2S-exposed rats, 100X magnification. Lungs of both rats are normal (top panels) with no evidence of lesions that would be expected to occur during inhaled sulfide exposure. As illustrated on the lower panels, both unexposed controls and H2S exposed rats displayed mild expansion of the peribronchial, peribronchiolar and perivascular interstitium by low to moderate numbers of lymphocytes and macrophages with fewer plasma cells consistent with chronic bronchus associated lymphoid tissue hyperplasia (BALT, see text for additional comments).