PREreview of Central Amygdala Neuronal Ensembles Coordinate Visceral Pain and Its Affective Behaviors

In this preprint, Samineni et al. identify specific neuronal populations in the central amygdala (CeA) that are recruited during cyclophosphamide (CYP)-induced interstitial cystitis. The CeA is a key hub for integrating chronic pain and emotional processing in bladder pain syndromes. However, the precise neuronal subtypes within the CeA that encode or modulate the sensory and affective components of cystitis-associated pain have remained unclear.

In this study, the authors used single-nuclei RNA + ATAC sequencing to profile transcriptional and epigenetic changes in CeA neurons and identified specific neuronal ensembles that are activated following CYP treatment. These findings were validated using FosTRAP genetics, which enables activity-dependent neuronal labeling, in combination with RNAscope to confirm the transcriptional identities of CYP-activated CeA neurons. The authors further identified cell type-specific enhancer regions associated with inflammatory activation, which could serve as valuable tools for future efforts to target defined CeA neuronal subtypes in functional studies.

To test the functional relevance of these cells, the authors employed optogenetic approaches to manipulate the cystitis-FosTRAPed CeA neurons. Optogenetic activation of these neurons induced referred abdominal pain, bladder hypersensitivity, anxiety-like behavior, and anhedonia, while inhibition produced opposing effects. These results suggest a causal role for CeA neurons activated during cystitis in driving both nocifensive and affective components of bladder pain.

Overall, this work addresses an important gap in our understanding of how bladder inflammation engages central circuits to produce both sensory and affective symptoms. The authors take a comprehensive and multidisciplinary approach, combining multiomics, electrophysiology, activity-dependent labeling, and circuit-level manipulations to define CeA neuronal populations that link bladder inflammation to emotional states. While the study is impressive in scope and conceptually strong, a few areas could benefit from clarification. In particular, some experimental details require further explanation. We have outlined a few specific suggestions below that could help strengthen the manuscript and further increase its impact.

Major Comments-

1. Figure S3C: There appears to be batch variation between samples 1 (Female 1 and Male 1) and samples 2 (Female 2 and Male 2). Were these two sets of samples prepared, processed, or sequenced under different conditions (e.g., different days, sequencing runs, kits, or personnel)? If so, batch effects could be influencing the clustering results. The authors may consider applying batch-correction or integration methods such as Harmony, Seurat Integration, or Mutual Nearest Neighbors (MNN) to harmonize the datasets across samples.

2. Multimodal Cell Clustering: It would be informative to perform joint clustering using both RNA and ATAC modalities (e.g., Weighted Nearest Neighbor (WNN) analysis for multiome data). The authors could then compare these integrated clusters to those shown in Figures 2B and 2C. Multimodal integration often provides finer resolution and may reveal additional or distinct neuronal subpopulations within the CeA.

3. Experimental Design (Timing): Could the authors clarify the rationale for performing multiome sequencing of the CeA from CYP- and saline-treated mice at 1 hour, rather than 4 hours, which corresponds to the peak of abdominal referred pain sensitization shown in Figure 1? Aligning the molecular analyses with the behavioral peak might yield additional mechanistic insights.

4. Figure 3H (Green-Yellow Pie Chart): The green-yellow pie chart in Figure 3H demonstrates that TRAPed neurons are a small percentage of total activated CeA neurons upon the second cystitis induction. Nothing is written about this, and including an explanation in the main text or figure legend would help readers interpret its meaning and relevance. Do the authors interpret this to indicate the limitation of coverage by the TRAP method, and that they are underestimating the bladder-pain-responsive ensemble, or that new ensembles are recruited with subsequent initiation of bladder pain in the CeA?

5. In Figure S6, the difference between 40% (CFA) and 30% (CQ/home cage) of tdt+cFos neurons seems subtle, and given the spread in numbers seen in S6A, it is perhaps within the range of normal variability. Do the authors see large variability between animals, and how are they getting a total %? Is it the average among animals? These stimuli are located in different areas- is it possibly a difference in salience rather than modality? The animals can easily scratch their nape, so perhaps this causes less distress overall. The interpretation here can be softened, or more experiments should be done to address the modality specificity of the activated neurons in the CeA. For instance, do the CFA activated cFos⁺ neurons also co-express Pde1c and Prkcd?

6. Tail Suspension Test: The tail suspension test was conducted only under conditions of CeA neuronal inhibition. Does optogenetic activation of these neurons produce opposing behavior during the tail suspension test?

7. Baseline cFos Expression: Some baseline cFos activity is observed even in saline-treated mice. Were there any neuronal cell types that showed a reduction in immediate early gene expression following CYP treatment? Identifying such populations could highlight neuronal subtypes that are suppressed during inflammation and offer clues to potential inhibitory gating mechanisms in the CeA.

8. Chromatin Changes and Temporal Dynamics: The authors report increased IEG expression in specific neuronal clusters at 1 hour. It would be valuable to examine corresponding chromatin accessibility changes across neuronal subtypes in the CeA following CYP treatment. Performing GREAT analysis on differential chromatin regions could provide mechanistic insight into the molecular pathways engaged during inflammation. Additionally, do the 1-hour chromatin accessibility changes predict potential downstream transcriptional responses at 4 hours?

Minor Comments-

1. Consider switching the order of Figures 1D and 1E, as Figure 1E is referenced earlier in the main text.

2. Page 3, Line 23: Please include a reference to Figure 1F at this point in the text.

3. After Figure S3D, Figure S4 is mentioned before S3E in the text (Page 5, Lines 1, 8, and 19). It would be helpful to adjust the figure order for consistency.

4. Page 5, Line 19: The figure reference should be Figure S3G, not S3E.

5. Page 5, Line 23: The figure reference should be Figure S3H, not S3F.

6. The color coding of clusters in Figures 2B–C does not appear to match the color scheme in Figure S3E.

7. In Figure 4B, it would be helpful to include in the schematic the time point at which CeA neurons are optically activated to assess behavior after 4 weeks.

8. Check the alignment of Fig4 G,H,I  with E & F.

9. Figures 4R–T are mentioned before Figures 4P–Q in the text. Reordering the figure references would improve the logical flow.

10. Page 8, line 10, Figure number is missing

11. Page 12, Line 3: The text should read “CYP-treated mice” instead of “saline.”

Competing interests

The authors declare that they have no competing interests.

Use of Artificial Intelligence (AI)

The authors declare that they did not use generative AI to come up with new ideas for their review.