*4.1. Pathology and Specimen Collection*

A thorough gross examination of the deceased patient was performed within 12 h of death. Throughout the procedure, aseptically collected sections of liver, kidney, lung, spleen, cerebral frontal lobe, small intestine, large intestine, gastric contents, feces, urine, bile, and, heart blood were collected for ancillary testing. Fresh tissues and fluids were individually placed into sterile bags or sterile syringes to be held in a −20 ◦C freezer for storage until testing could be performed.

Sections of all abdominal and thoracic organs along with brain, eyes, skin, skeletal muscle, sciatic nerve, and bone marrow were preserved in 10% neutral buffered formalin. Tissues were trimmed, placed into cassettes, and processed via routine paraffin embedding techniques. All tissues were made into 3–5 μm sections and stained with hematoxylin and eosin stain (H&E) for histological review. Microscopic review was performed by a board-certified veterinary pathologist using a Nikon eclipse 80i (Nikon, Minato, Tokyo, Japan) and photomicrographs were captured by an Accu-scope Excelis HD (Commack, NY, USA).

Antemortem sample collection from a subset of the surviving dogs (Table 1) including urine, blood, vomit, and hair was performed by owners or veterinary staff during and after the initial onset of clinical signs. Urine was collected via free catch method and stored in sterile jars or plastic collection containers without preservative. Liver, kidney, free catch urine, blood and hair samples from three control dogs not associated with the HAB event were additionally collected (one deceased and the other living). Tissue, hair, and fluid samples were chilled and/or frozen prior to overnight shipment or transfer. Sample collection dates in relation to exposure can be referenced in Table 1.

#### *4.2. Phycology of Vomitus Sample*

A ca 0.5 g subset of the C-GR#2 vomitus sample was suspended in 5 mL of deionized water (DI) and gently vortex mixed. Wet mounts of un-fixed sample were prepared (3x) and scanned at 100X for the presence of potentially toxigenic cyanobacteria using a Nikon TE200 inverted microscope equipped with phase contrast and epifluorescence (green light excitation, 510–560, FT580, LP590). A Nikon digital sight DS-Fi1 camera was used for micrographs. Higher magnification was used as necessary for identification and micrographs.

#### *4.3. Adda MC*/*NOD Analyses*

#### 4.3.1. Specimen Homogenization

Subsets were taken from liver samples (composited 5 subsets; ca 1 g each) and the entire kidney (2.7 g) was cut into small pieces and placed into 30 mL homogenization vials with 10 mM phosphate buffer at pH 7 (1:3). The vomitus sample was emptied into a glass beaker, mixed with a spatula and subsampled (2 subsets; 1 g each) into individual 7 mL homogenization vials with extractant solution (75% acetonitrile in 100 mM acetic acid; 1:3). Acetonitrile (HPLC grade) and glacial acetic acid (>99%) were both from Thermo Fisher Scientific (Waltham, MA, USA). Ceramic beads (2.8 mm) were added and the samples were homogenized at 5 m/s for 30 seconds (1 cycle for the vomitus and 2 cycles for the tissues) using an Omni Bead Ruptor 24 (Omni International, Inc, Kennesaw, GA, USA). Hair samples were pulverized in 30 mL vials with stainless steel 2.8 mm beads dry at 6.8 m/s (30 seconds) and transferred to glass vials as 50 mg subsets. Aliquots of kidney and liver (400 μL) were transferred to glass vials for subsequent oxidations/extractions. The vomitus vials were centrifuged (1500 g; 10 min) and supernatants retained. The pellets were vortex mixed with additional extractant (1 mL), centrifuged as before and supernatants pooled and saved for extractions.
