Animal histological images for in vivo maximum tolerated dose and toxicity
Data files
Sep 02, 2025 version files 24.66 GB
-
1-Kidney.svs
194.55 MB
-
1-Liver.svs
600.44 MB
-
1-Spleen.svs
296.37 MB
-
16-Kidney.svs
427.86 MB
-
16-Liver.svs
704.42 MB
-
16-Spleen.svs
523.58 MB
-
17-Kidney.svs
539.42 MB
-
17-Liver.svs
1.01 GB
-
17-Spleen.svs
445.06 MB
-
18-Kidney.svs
482.07 MB
-
18-Liver.svs
543.24 MB
-
18-Spleen.svs
450.21 MB
-
19_Spleen.svs
501.84 MB
-
19-Kidney.svs
505.81 MB
-
19-Liver.svs
879.49 MB
-
2-Kidney.svs
191.94 MB
-
2-Liver.svs
705.49 MB
-
2-Spleen.svs
294.22 MB
-
20-Kidney.svs
460.39 MB
-
20-Liver.svs
487.05 MB
-
20-Spleen.svs
489.42 MB
-
21-Kidney.svs
450.84 MB
-
21-Liver.svs
978.64 MB
-
21-Spleen.svs
510.81 MB
-
22-Kidney.svs
432.58 MB
-
22-Liver.svs
687.30 MB
-
22-Spleen.svs
158.04 MB
-
23-Kidney.svs
167.23 MB
-
23-Liver.svs
232.03 MB
-
23-Spleen.svs
144.10 MB
-
24-Kidney.svs
192.66 MB
-
24-Liver.svs
260.23 MB
-
24-Spleen.svs
194.62 MB
-
25-Kidney.svs
163.84 MB
-
25-Liver.svs
375.63 MB
-
25-Spleen.svs
206.76 MB
-
26-Kidney.svs
171.95 MB
-
26-Liver.svs
307.52 MB
-
26-Spleen.svs
162.37 MB
-
27-Kidney.svs
173.98 MB
-
27-Liver.svs
372.10 MB
-
27-Spleen.svs
149.39 MB
-
28-Kidney.svs
162.65 MB
-
28-Liver.svs
357.05 MB
-
28-Spleen.svs
138.47 MB
-
29-Kidney.svs
141.69 MB
-
29-Liver.svs
240.02 MB
-
29-Spleen.svs
188.82 MB
-
3-Kidney.svs
198.16 MB
-
3-Liver.svs
621.25 MB
-
3-Spleen.svs
338.53 MB
-
30-Kidney.svs
164.13 MB
-
30-Liver.svs
263.06 MB
-
30-Spleen.svs
167.60 MB
-
31-Kidney.svs
211.61 MB
-
31-Liver.svs
255.37 MB
-
31-Spleen.svs
177.79 MB
-
32-Kidney.svs
200.07 MB
-
32-Liver.svs
491.13 MB
-
32-Spleen.svs
200.91 MB
-
7-Kidney.svs
195.59 MB
-
7-Liver.svs
640.32 MB
-
7-Spleen.svs
239.74 MB
-
8-Kidney.svs
207.15 MB
-
8-Liver.svs
461.44 MB
-
8-Spleen.svs
354.46 MB
-
9-Kidney.svs
184.07 MB
-
9-Liver.svs
408.01 MB
-
9-Spleen.svs
227.24 MB
-
README.md
4.78 KB
Abstract
The rising rate of antibiotic resistance is a widely acknowledged challenge, especially for gram-negative bacterial infections. Some drugs, such as the lipopeptide polymyxin B (PMB), have high antimicrobial activity against gram-negative organisms but also high toxicity owing to low specificity, resulting in a low therapeutic index and poor clinical utility. Due to the high rate of nephrotoxicity, systemic use of PMB is presently restricted to last-line therapy, with adverse clinical tradeoffs. To salvage such antibiotics, we propose specific drug delivery to bacterial cells using engineered phage as a carrier. We replaced the receptor-binding protein of phage M13 with an antibody fragment (scFab) recognizing the core antigen of lipopolysaccharide (LPS), creating a phage that bound a wide range of clinically important gram-negative pathogen species. We then cross-linked thousands of PMB molecules per virion, making ‘PMB-Phage’. PMB-Phage reduced the minimum inhibitory concentration (MIC) by up to 2 orders of magnitude for a variety of species, including the ESKAPEE pathogens Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli, and Acinetobacter baumannii. To test efficacy, immunocompetent mice were infected with multidrug-resistant P. aeruginosa in a pneumonia model. PMB-Phage treatment was effective, causing bacteriostasis and up to 500-fold reduction in bacterial counts compared to a negative control, with a ~90-fold increase in potency compared to PMB. Similarly, P. aeruginosa infection of the cornea (keratitis) was treated effectively by PMB-Phage, with ~10,000-fold reduction in bacterial counts and a >10-fold increase in potency compared to PMB. PMB-Phage was well-tolerated, with no toxic effects observed in vitro or in vivo, including by serum biomarkers and histology. This work demonstrates a proof-of-concept of phage-antibiotic conjugates as a strategy to improve cell targeting for bacteria over mammalian cells, combining engineerable phage-based targeting with large payload capacity. Phage-drug conjugates may represent a next-generation strategy to greatly improve potency and therapeutic index for otherwise toxic molecules.
README: Animal histological images for in vivo maximum tolerated dose and toxicity
Associated article: Yang Y, Vexler S, Jordan MC, et al. “A Synthetic Phage-Peptide Conjugate as a Potent Antibacterial Agent for Pseudomonas aeruginosa Infections.” ACS Central Science. DOI: 10.1021/acscentsci.5c00562
Contact: Irene A. Chen — ireneachen@ucla.edu
License: Data are released by Dryad under CC0 (public domain).
1) Summary
This dataset contains whole-slide histology images from a 7-day in vivo maximum tolerated dose (MTD) / toxicity study in mice exposed to polymyxin B (PMB), engineered phage–peptide conjugate (PMB-M13αLPS), 1× PBS vehicle, or no injection. For each animal, formalin-fixed, paraffin-embedded tissues (kidney, liver, spleen) were H&E-stained and scanned as whole-slide images. Slides were labeled so that the pathologist was blinded to treatment. The dataset is intended to enable independent inspection of tissue morphology for treatment-related toxicity.
2) What’s included
Files are at the top level and follow the naming scheme:
<AnimalNumber>-.svs
<AnimalNumber>: integer ID (see mapping below)
<Tissue>: one of Kidney, Liver, Spleen
Each animal should have up to three slides (kidney, liver, spleen). Animals 4–6 have no slides (animal death after a single PMB dose).
3) Animal numbering → treatment (dose, route, schedule)
Abbreviations:
PMB = polymyxin B (sulfate); PMB-M13αLPS = engineered M13 phage targeted to LPS and conjugated with PMB; IV = intravenous (tail vein); QD = once daily; PBS = phosphate-buffered saline.
All injections were IV, QD for 7 days in 100 µL unless shown as “No injection”.
1–3 PMB 21 mg/kg
4–6 PMB 72 mg/kg (no histology; animals died after first dose)
7–9 PMB-M13αLPS 1.6×10^11 virions/day
16–19 PBS (vehicle control)
20–22 PMB 2.5 mg/kg
23–25 PMB-M13αLPS 6.85×10^9 virions/day
26–28 PMB-M13αLPS 1.37×10^10 virions/day
29–31 PMB-M13αLPS 2.74×10^10 virions/day
32 No injection of substance (untreated control)
4) How to access/open the data (free & open-source)
Recommended: QuPath (v0.4+). QuPath is free, cross-platform, and supports .svs via Bio-Formats.
Quick start:
Install QuPath: https://qupath.github.io
File → Open… → select any .svs file.
If prompted, set Image type = “Brightfield (H&E)”.
Use the zoom/pan tools to inspect tissue; toggle the scale bar (View → Show scale bar).
For annotations or measurements: use the Tools panel; results export via Analyze → Measure… and File → Export….
Alternatives (also free/open):
ASAP (Automated Slide Analysis Platform) — Windows/Linux; reads .svs.
OpenSlide + openslide-viewer — lightweight viewer for .svs.
5) Data structure, variables, and conventions
File format: .SVS (pyramidal TIFF-based whole-slide images).
Tissues: Kidney, Liver, Spleen.
Animal IDs: integers listed in Section 3.
Missing data: No images for animals 4–6 (mortality after first PMB dose).
Quality notes: Some slides may contain scanning artifacts near edges (tissue folds, pen marks). These do not affect central parenchyma review.
Units/scale: Pixel size and magnification are embedded in the SVS metadata and read automatically by QuPath; use the QuPath scale bar for calibrated measurements.
6) Methods (brief, for context)
Male mice received daily IV tail-vein injections for 7 days per Section 3. Twenty-four hours after the last injection, blood was collected for clinical chemistry and animals were euthanized; kidney, liver, and spleen were fixed, processed, H&E-stained, and scanned as whole-slide images. Blinded histopathology was performed to assess any treatment-related lesions.
7) File inventory (as uploaded)
For transparency, this version contains .svs files named for animals:
1, 2, 3; 7, 8, 9; and 16–32 (three tissues per animal where available).
If any filename differs by punctuation (e.g., 19_-Spleen.svs), treat it as the corresponding animal/tissue listed above.
8) How to cite
Please cite both the dataset and the associated article.
Dataset (example, update with the final DOI once published):
“Irene A. Chen Lab (2025). Animal Histological Images for In vivo Maximum Tolerated Dose and Toxicity." Dryad, DOI: 10.5061/dryad.47d7wm3mw
Article:
Yang Y, Vexler S, Jordan MC, et al. “A Synthetic Phage-Peptide Conjugate as a Potent Antibacterial Agent for Pseudomonas aeruginosa Infections.” ACS Central Science. DOI: 10.1021/acscentsci.5c00562
9) Version history
2025-09-02: Initial upload of 24.66 GB of .svs slides and README.
10) Support
Questions or corrections: ireneachen@ucla.edu
