Skip to main content

Differential Proteomic Expression of Equine Cardiac and Lamellar Tissue During Insulin-Induced Laminitis

Cite this dataset

Campolo, Allison et al. (2019). Differential Proteomic Expression of Equine Cardiac and Lamellar Tissue During Insulin-Induced Laminitis [Dataset]. Dryad.


Endocrinopathic laminitis is pathologically similar to the multi-organ dysfunction and peripheral neuropathy found in human patients with metabolic syndrome.  Similarly, endocrinopathic laminitis has been shown to partially result from vascular dysfunction. However, despite extensive research, the pathogenesis of this disease is not well elucidated and laminitis remains without an effective treatment. Here, we sought to identify novel proteins and pathways underlying the development of equine endocrinopathic laminitis. Healthy Standardbred horses (n=4/group) were either given an electrolyte infusion, or a 48-hour euglycemic-hyperinsulinemic clamp. Cardiac and lamellar tissues were analyzed by mass spectrometry (FDR=0.05). All hyperinsulinemic horses developed laminitis despite being previously healthy. We identified 538 and 737 unique proteins in the cardiac and lamellar proteomes, respectively. In the lamellar tissue, we identified 14 proteins which were significantly upregulated and 13 proteins which were significantly downregulated in the hyperinsulinemic group as compared to controls. These results were confirmed via real-time reverse-transcriptase PCR. A STRING analysis of protein-protein interactions revealed that these upregulated proteins were primarily involved in coagulation and complement cascades, platelet activity, and ribosomal function, while downregulated proteins were involved in focal adhesions, spliceosomes, and cell-cell matrices. Novel significant differentially expressed proteins associated with hyperinsulinemia-induced laminitis include talin -1, vinculin, cadherin-13, fibrinogen, alpha-2-macroglobulin, and heat shock protein 90. In contrast, no proteins were found to be significantly differentially expressed in the heart of hyperinsulinemic horses compared to controls. Together, these data indicate that while hyperinsulinemia induced, in part, microvascular damage, complement activation, and ribosomal dysfunction in the lamellae, but a similar effect was not seen in the heart. In brief, this proteomic investigation of a unique equine model of hyperinsulinemia identified novel proteins and signaling pathways, which may lead to the discovery of molecular biomarkers and/or therapeutic targets for endocrinopathic laminitis.


Proteomic Analysis

Tissues were homogenized in RIPA buffer containing protease inhibitor cocktail, and protein contents were quantified by bicinchoninic assay, as previously described (Maria, Campolo et al. 2015).  Aliquots (25 micrograms) of each lysate were precipitated with TCA/acetone, redissolved and reduced for 1 hr in 8 M urea, 5 mM TECP, 100 mM Tris pH 8.5, and then alkylated for 20 min by adding 10 mM IAA.  Reactions were then diluted to 2 M urea, trypsin was added to 4 micrograms / ml, and digestions were incubated overnight at 37 degrees C.  After digestion, peptides were desalted by solid phase extraction on monolithic C18 pipet tips, and lyophilized to dryness.

For LC-MS/MS, one microgram of peptides was injected onto a 0.15 x 50 mm vented trap column, followed by separation on a 0.075 x 400 mm Picofrit column/emitter (New Objective).  Both columns were packed with 3-micron Magic AQ C18 particles (Bruker).  Peptides were separated at room temperature using a 5-40% ACN/0.1% formic acid gradient performed over 100 min at a flow rate of 250 nL/min, eluting directly into a New Objective PV-550 nanoelectrospray ion source.  Peptides were analyzed uing an LTQ OrbitrapXL mass spectrometer (Thermo) via a Top Six data-dependent acquisition method as previously described (Voruganti, Lacroix et al. 2013)

Proteins were identified and quantified by using MaxQuant (Cox and Mann 2008) to search the raw instrument files against a database of  22,718 reference equine protein sequences downloaded from UniProt on 12/27/15.  Searches used the fixed modification carbamidomethyl (C) and the following variable modifications:  oxidized (M), acetyl (protein N-terminus) and glutamine cyclization to glutamate).  Searches were performed using default MaxQuant settings, with the addition of label-free quantitation via the LFQ algorithm and match between runs to transfer MS/MS identifications between LC-MS/MS data files. 

Statistical analyses were performed using the Perseus software package (Max Planck Institute of Biochemistry) (Tyanova, Temu et al. 2016) to analyze log(2)-transformed MaxQuant LFQ protein intensities. Samples were identified as being significantly differentially expressed via Student’s t-test using the Benjamini-Hochberg multiple tests correction, with a false discovery rate of 0.05 (Benjamini and Hochberg 1995).  Protein-protein interactions and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways were identified via STRING 10.5 (Search Tool for the Retrieval of Interacting Genes/Proteins).


Animal Health Foundation

Oklahoma State University