The orbuculum, or crystal ball, was invented about 3000 BC according to Wikipedia. Mystical orbs were used in numerous cultures to communicate with the gods or learn of future threats. Wouldn’t it be nice to identify horses that are genetically pre-disposed to get sarcocystosis? A genetic EPM-crystal ball.
Thirteen years ago, we participated in studies that sought to identify cell markers unique to horses with equine protozoal myeloencephalitis, EPM. The idea was that cell markers or the “gene expression signature” unique to EPM would be found in immune cells circulating in the blood. The differences in gene expression between animals with and without clinical evidence of EPM would be analyzed using several statistical measures. Genes that showed statistically significant differences in clinically positive horses were compared to those that were clinically negative, and the genes that showed a significant difference (those significantly up-regulated) would constitute the EPM-gene signature.
Controlled laboratory studies testing the hypothesis that a gene signature could be found and be useful in the diagnosis and treatment of EPM were undertaken. A controlled infection (induced stress) study was used to accurately know the day of exposure. Sarcocystis neurona oocysts were administered to 20 stressed horses to elicit disease. Blood samples were taken 10 times over 28 days to collect RNA, the measure of a turned on gene. The up-regulated genes (identified by the RNA analysis) were assayed on a custom microarray for determining gene expression (the specific array was patented, but not by us). In this blinded study (veterinarians didn’t know which horses were infected) clinical exams were performed,serum and CSF were tested, and post-mortem exams were conducted to ensure that clinically ill horses did get EPM.
This experiment was eventually published with the infection data, but the gene analysis data was not reported. Remarkably, the horses that showed signs didn’t have organisms that could be demonstrated in the brain tissues. Inflammation was considered diagnostic of successful infection. Scientists conducting this study identified a gene signature. Success! There were differences between the up-regulated genes in the clinically ill horses that were infected and those that were not infected, control horses.
Time to test the gene signature. Field cases of suspected EPM were used in a second study. The gene expression from horses with suspected EPM, those that had serum and CSF analysis to be as sure as possible the horses fit the diagnostic criteria at the time, came from clinical cases. It took 6 weeks to process the samples and get a result because the assay is technical. An obvious down side of the endeavor was cost. The hundreds of dollars that the eventual assay would cost, and the six-week turnaround time, made it a clinical non-starter. More importantly, the assay didn’t diagnose chronic disease (disease that was present after 28 days from the initial infection). The cases presented to veterinarians are chronic. The acute gene signature did not identify field cases that the veterinarians diagnosed.
In a third study, 13 animals were used in a merozoite challenge model that did not use stress. The horses were randomly assigned to a group, 8 were challenged while 5 were sham challenged. This study ran 90 days to detect an acute and chronic gene signature. Cells were assayed for gene expression at 28 days (acute) and 90 days (chronic). If the acute gene signature was the same in both models, an accurate marker between the two studies could be identified to identify acute, possibly current, disease. Likewise, the chronic markers, significant expression of genes at 90 days should match the field study and identify horses with long term disease even if the organisms were eliminated.
The cumulative results of the controlled studies identified 31 genes that were highly statistically different at day 28 between animals that developed clinical EPM and those that did not. An EPM index score calculated for the gene signature, identified in the first controlled study, was successfully used to identify some, but not all, of the horses with acute disease in the third, controlled study.
Horses with chronic EPM, day 90 of the third study, were not identified using the signature developed in the first study. Further, chronic EPM could not be identified in clinical field samples using the gene signature developed from acute disease, day 28, in either the stress model or the merozoite model. Because many horses present with suspect EPM after having had clinical signs for weeks or months, the value of a genetic signature was doubtful.
We identified drugs that selectively reduce the expression of some of the upregulated genes that were stimulated during acute and chronic disease. Some drugs returned horses to normal, removal of the drugs allowed the horse to again show signs of disease. Surprisingly, some drugs we tested made horses worse! We found that the innate immune response and the genetic signature of host cells are the key to disease associated with sarcocystosis.
The data was useful. The upregulated genes included MHC Class II receptors, chemokine receptors, IgG molecules, natural killer cells, several interferon-induced proteins and a handful of others.
More studies are undoubtedly in the pipeline. As those studies are completed and in a few years published, they may be compared to the work done in 2005. Or perhaps the EPM-gene signature is already relegated to the cutting room floor, the genes frozen in time, yielding space in the freezer for newer endeavors and lost to analysis. Our experience showed us there are disease-signatures present in the blood samples and these are markers that can effectively direct treatment for horses with disease and identify horses that are resistant to disease. We don’t think there will be a crystal ball that predicts which healthy horse will be come sick when exposed to Sarcocystis, sadly it isn’t that simple.