Skip to content

cartoonBioassays are desired and can be useful tests that assist veterinarians in diagnosing disease and forming a treatment plans.  Generally, multiple tests over time are best and indicate if there is a treatment response or lack of effective therapy.

It is extremely important to understand what is being tested in an assay, how values are reported, and what the value means. This basic understanding gives veterinarians clues on how the assay will change over time in a tested animal.  A common question we get is “Why did the SAG titer (go up, not change)?” on a retested sample. This blog reviews the basics and after reading it you should be able to answer that question.  It is also good to stretch the envelope and understand new tools and what they can reveal about diseases in horses. Understanding new tools requires some background information, that is usually in accessible documents for those which are interested.

Bioassays differ in format and formats are as different as apples and oranges.  For example, enzyme linked immunosorbent assays, ELSIA’s, are different test systems (format) than immunofluorescent assays, IFAT’s.  Unless someone has published data comparing tests in a “head-to-head” experiment, direct comparisons between tests are not valid and this includes ELISA and IFAT testing. ELISA tests are different because they are designed to select different molecules. If you have a tool that detects apples, you won’t count oranges.  Likewise, if your tool detects oranges, you will miss the apples.  And if your experience and understanding is limited to one tool, without some investigation, you will miss what is possible. New tools can differentiate a Granny Smith from a McIntosh!

You may need a few definitions here for clarity.

Disease causing organisms elicit immune responses in infected animals, this response is called an adaptive response.  Adaptive responses can be an antibody or more complicated T-cell responses. We leave out the T-cell responses in this discussion even though they are important in polyneuritis equi pathology.  Antibodies are adaptive because they adapt the immune system to respond to a specific organism. There are also immune defenses, innate immune molecules called cytokines, that are released to fight off  infections before the body has identified the specific organism.  Cytokines are not specific, as opposed to adaptive immunity. A test measures molecules called analytes.  In antibody detecting tests, the protein that elicited the adaptive immune response is called an antigen.  An analyte can be an antibody or a cytokine or vitamin or toxic substance (lead is a good example). An analyte can be air.

Antibodies are very specific for the proteins that induced the response.  Organisms can share antigens, organisms as different as humans and mice can share antigens.  Antibodies to these common proteins don’t clearly differentiate organisms to genus or perhaps species.  Some antigens can be so specific they are used to identify a strain belonging to a species! Tests that differentiate organisms by an antibody response is called serotyping. A good example is S. neurona. There are three specific and dominant Surface AntiGens (SAG’s) of S. neurona, that define the genotype.  Because each SAG is genetically mutually exclusive, each protozoa only displays one of the three possible antigens on its surface.  When S. neurona infects the animal, an antibody is produced and directed against that SAG, the cause of the infection can be serotyped.

Keep in the back of your mind that opossums usually are infected with all three genotypes of S. neurona and horses generally are exposed to more than one genotype. S. neurona also elicits responses to numerous common antigens. Detecting responses to common antigens won’t distinguish between genus or species.  For example, on some testing formats shared antigens between S. neurona and S. fayeri and S. falcatula are detected and a positive result on the test won’t distinguish organisms causing EPM (S. neurona), EMS (S. fayeri) or no horse-disease at all (S. falcatula). Educated guesses are made based on dilutions and probability statistics.  Direct measures are preferred.

Other than antibody molecules, other analytes are measured in body fluids.  Analytes can be cytokines such as IL6, C reactive protein (CRP), vitamins, and neurofilament light (NfL).

Let’s get back to assay format. A Western Blot (WB) is a method to separate all the antigens from an organism by molecular size. The analyte (an antibody in the blood or CSF) reacts with the antigens and a pattern is detected.  The intensity of the reaction and experience with a pattern are factors on deciding a test positive status.  A confounding factor in WB’s are the size of some antigens.  It is no surprise that organisms have lots of similar sized proteins.  It isn’t a surprise that some types of separation gels will inadequately separate proteins that are similar in size, the conditions used to make the antigen preparation and even the conditions to run the assay have large impacts.   My PhD thesis demonstrated that SAG 1 ran on a WB with other similar sized proteins and using detergent in the assay decreased the amount of SAG 1 on the gel.  Not a new concept yet ignored in commercial testing, ultimately clouding the interpretation of SAG 1 infections. It shouldn’t be a big surprise that using S. neurona as the antigen source in a WB will give different results than using S. falcatula for  antigens.

The IFAT format uses a  S. neuronaSAG1 strain S2 in the test. To detect a SAG 5 or SAG 6 strain of S. neurona using the S. neuronaSAG1 strain S2 test, the interpretation must be based on common antigens.  That means S. fayeri antibodies found in horses will be positive on S. neuronaSAG1 strain S2 tests. There are several papers from different authors that discuss the data from serum/CSF analysis.  A statistical probability analysis can sometimes be used with test interpretation.  Be sure and know how positive values were obtained to validate the test, one or 100 or 1000 horses.

ELISA plateThe 2, 4/3 ELISA uses a chimeric (synthetically made antigen) and is a completely different assay than the SAG 1, SAG 5 or SAG 6 ELISA’s. The SAG 1, 5, 6 ELISA  uses recombinant proteins, folded in the native S. neurona proteins conformation (no detergent). The 2, 4/3 ELISA isn’t capable of serotyping S. neurona.  The 2, 4/3 protein will detect the SAG 2, SAG 3, and SAG 4 protein of S. fayeri.  The antigen we use to detect S. fayeri isn’t a surface antigen.  The S. fayeri assay we use employs the toxin from toxin producing strains of S. fayeri that may cause disease. For the skeptics, there are four papers available that discuss this toxin and its relationship to equine disease. ELISA tests can be specific enough to detect the infection stage of an the organism, such as Borrelia (Lyme disease).  The antigen displayed during acute infection is most interesting to veterinarians considering Lyme disease in a horse. Antibody detecting tests detect antibody against myelin protein 2 (MP2).  There are copious papers on this topic in people and horses, the protein, and disease associated with the protein.  The importance in running the test is the source and  native folding of MP2.

Now you know the keys to understanding bioassays are  format of the tests and antibody specificity used in each test. What else is important?

When will antibody levels change? It is expected that antibodies will go down depending on the immune status of the animal (naïve, first infection) or experienced (more than one infection) as well as exposure. It is known that antibodies against S. neurona in a naïve animal will go down in a couple of months.  The longevity of maternal antibodies in a foal are known. The SAG antibody levels in an experienced horse can take 10 months to recede, unless the animal is chronically exposed to the organism, an then the antibody levels will remain elevated. This information gives you the ability to use judicious testing and why, if tested too soon, an animal will remain positive or the antibody levels will go up.

Anti-myelin protein antibody levels can take 5 months to return to normal after the myelin isn’t presented to antibody producing cells. Levels of CRP and NfL will respond to the infection status, NfL responds to effective therapy and goes down in 7-10 days, CRP takes longer. The take home message is that antibody levels take a long time to go down while some of the other analytes decrease in a short time.  Also, understand that antibody levels are a measure of immune response and it is desirable to have protective antibodies.  It is undesirable to have autoimmune antibodies (anti-myelin protein antibodies). One needs to understand the disease pathology to effectively select and interpret tests.  Preventing protective immunity is not a good tactic while treating disease is a good strategy.

Putting it together.  Knowing what the test is designed to tell you and how those values are expected to change over time are fundamental.  It can not be said enough that there are no EPM tests.  None.  The tests that detect antibody against S. neurona do just that. They measure an antibody response to S. neurona but do not tell you the location in the body.  We used clinical disease, experimental infections, and recombinant antigens given as a vaccine to validate our SAG 1, SAG 5, and SAG 6 testing.  We report antibody levels (as a last dilution positive) against S. neurona when we run the SAG ELISAS.  We report a level of anti-toxin in serum or CSF in the S. fayeri ELISA test result.  The level of antibody against MP2 (or the reactive IL6 receptor, MPP protein) is an indication of demyelination.  An animal shouldn’t have these antibodies in their blood or CSF.  The MP2/MPP antibody is a measure of disease and the disease pathology.  Reporting a direct measure (mg/ml or micrograms/ml) for CRP and NfL are informative. CRP is a non-specific, acute phase cytokine response to inflammation and NfL is a measure of the axon damage in demyelinating and non-demyelinating polyneuropathies.

horse menuAn unfortunate legacy of COVID-19 is the critical importance that better diagnostics tools could have played to mitigate the virus’ impact on human health and the world’s economy. And the lost lives.  As widely known in the industry, lab services account for less than 3% of total U.S. healthcare spending today, even though their test results impact a majority of all clinical decisions regarding patient care. Just as importantly, diagnostic testing saves horses lives.

The following case scenarios illustrate the folly of skipping diagnostic testing

A patients gums are very pale and anemia is a likely diagnosis.  In this case, the course of action is not directed with diagnostic point of care (POC) tests to provide actionable information, treatment commences with an iron supplement.  The intent is to “observe a treatment response”. You can take your pick of species in this example, be it goat, horse, or human.

If you picked a goat the most likely cause of anemia is parasitic and no amount of iron supplement will save the patient. The correct treatment is an effective anti-parasitic agent.  The anti-parasitic drug must be selected based on the parasite target--be it strongyle or coccidia.  An incorrect guess as to the parasite is just as lethal as treating parasites with iron. Picking the diagnostics from the a la carte menu is the best option.

If you selected horse as the example, a cause of anemia might be blood loss that is due to a guttural pouch mycosis. An iron supplement could help with chronic anemia, but supplements won’t prevent the eventual acute onset of lethal exsanguination—these horses often die because they hemorrhage from the nose.  Instituted early, surgery and medical treatment with various antifungal preparations can be effective.

Chronic anemia can be due to an inability of the bone marrow to produce red cells and differentiated from anemia due to blood loss.  Toxins, metabolic diseases-such as renal disease, and cancer can result in a lack of red cell production in humans and animals.

Diagnostic tests are selected and yield enhanced directed information, ultimately improving patient care and outcomes. Diagnostic tests save money.

What drove some people to rely on treatment response in dealing with equine sarcocystosis, a cause of equine protozoal myeloencephalitis (EPM) instead of pursuing diagnostic testing?  Perhaps it was confusing early messaging from leading experts in the field. The Western Blot was unable to distinguish between infection and disease. The Western Blot test failed to distinguish between species such as neurona, fayeri, and falcatula.  Experts failed to get the message out that antibodies vary in response to exposure, infection, and central nervous system disease.  An important lost message was that different tests measure different things.  It wasn’t the testing that failed, it was the messaging on what information the tests could reveal that missed the mark.

It took many years before three serotypes of Sarcocystis neurona were recognized. Once the serotypes were discernable, it was no longer necessary to use non-specific detection of apicomplexans that required reducing data to a probability.  And then there are the confusing experiments that absolved S. falcatula of infecting horses.  Failing to recognize that S. neurona, serotype SAG 6, is antigenically (highly) similar to S. falcatula made these infections indistinguishable with the available testing. And then there is the question of relapses.  Ineffective treatment led to the myth that most horses  with EPM will relapse. However, relapses are often a manifestation of another disease.  A disease that is a bystander to S. neurona…and not treatable with anti-protozoal agents.

The effects of an infection with Sarcocystis are innate immune responses that result in a peripheral neuropathy.  There are other causes of peripheral neuropathies.  The bystander mechanism, inflammation, results in non-demyelinating or demyelinating disease.  The pathophysiology of polyneuritis is a spectrum, from initiation to end-stage disease. Diagnostic tests can indicate the difference between non-demyelinating and demyelinating disease and direct appropriate treatments. With careful interpretation it is possible to stage the disease and that can help with clinical decisions.

The diagnostic menu for equine disease is constantly being developed and refined.  Each test gives a clue to the disease process and most importantly, diagnostics help predict the effect of a drug on the disease process.  A diagnostic test is used in conjunction with patient history, clinical signs, and clinician experience.  A diagnostic test can save money, time, and often lives. Once familiar with testing options, (https://pathogenes.com/w/epm-testing/),  rational treatments can be planned. If you need help reading the menu, give us a call.

NflightPattern recognition is an important process that emphasizes the the identification of data regularities in a given scenario. People are natural pattern-seekers. How many times have you heard things happen in three’s?  Humans are hard wired to recognize sets of three events, even if science proves the events are unrelated!

Students are trained in veterinary school to become observers.  The trained veterinary-observer develops into a diagnostician after years of clinical practice.  The art of practice is a combination of science and observation enhanced by continued questioning of the objective outcomes one effects with treatments.  A clinician adds tools to his/her toolbox over time.

Bioassays are tools that are available in the clinicians toolbox. Our passion is neurodegenerative diseases in horses and people.  We bioassay a lot of samples from laboratory experiments and clinical submissions.  Sometimes, we compare our results to other laboratories by running tests that evaluate similar disease conditions, using different testing platforms, in order understand the differences in case interpretation. Our head-to-head tests (duplicate samples that are run on different platforms) figure into our interpretations.

After results are obtained, recognized patterns are passed along to the field veterinarian. We are not immune to event-bias and overcome that tendency by using algorithms that are coupled with statistical analysis. The systems analysis procedure produces, in a finite number of steps, the answer to questions we pose. For example, is this horse likely to relapse?

Our algorithms sift through the data we get from tests and red flag results to which we should pay attention. A clinician may evaluate a case a year, or perhaps several cases over several years.  Each case presents an individual interpretation of disease that makes field medicine enigmatic.  We enter tens of thousands of results into our algorithm and give the succinct final analysis to the veterinarian. The data to feed the algorithm come from serum bioassays important to the diagnostician.  At the very least, test results offer objective parameters to veterinarians on which to base their treatment decisions. 

There are some new things to consider. And these topics were discussed in our recent Zoom meetings.  We hope you joined in!  To assist with clinical analysis of horses with neuromuscular disease, several bioassays are used.  The term for the assayed molecule is “analyte”. Some of the assays detect antibody against foreign proteins as the analyte and include surface proteins of Sarcocystis neurona, (remember, these are unique to S. neurona and mutually exclusive to serotypes of each neurona species), or Neospora hughesii.  Anti-toxin against Sarcocystis fayeri is an analyte and if disease is stimulated an autoimmune reaction follows.  Two areas of the myelin protein P2 are analytes in our “Sidewinder” panel.  There are bioassays for antigens that include C-reactive protein (CRP) and neurofilaments (NfL) molecules.

The principle difference one should consider between detecting antibodies versus detecting antigens as analytes is time frame for a change in the levels found in the serum.  Some antibodies won’t decrease for months after they are produced against an antigen (foreign agent). Another consideration is that a naïve animal will show a reduction in antibodies much sooner than an animal that is “experienced” with the infection.  That means prior exposure is important information that should be taken into consideration when examining a case.

An animal that is chronically exposed to an organism in the environment will maintain antibodies due to new gut infections and it can be tough to interpret these test results in the face of acute disease. That means prevalence of disease is an important consideration in analysis of these cases. The life-cycle of the organism is an important consideration. Does the organism complete the life-cycle in the host, as in S. fayeri or is it unable to mature, as in S. neurona infections in horses? Does the organism change it’s repertoire of antigens presented during infection, as does S. neurona? Did you consider that one infection, with a particular serotype of S. neurona may not protect against another serotype?  Are antibodies produced against a serotype of S. neurona protective against infection but stimulatory to the inflammation that can become dysregulated?

Antigen molecules, such as CRP, an acute phase protein, are useful. CRP is elevated in inflammation when it is associated with the cytokine IL6.  There are several innate plasma buffer systems that regulate IL6—>CRP, but occasionally the system becomes dysregulated. Chronic dysregulation can lead to an inflammatory condition and chronic inflammation can lead to an autoimmune disease.  The presence of CRP indicates inflammation due to an infective process, however it isn’t specific to one particular organism. CRP is quick to be produced but in our analysis, it doesn’t decline in days.  It can take weeks.  Horses have several conditions that can keep the CRP value elevated; they include encysted parasites or hind gut ulcer disease. 

One very dynamic marker in neurodegenerative disease is neurofilament light (NfL).  Neurofilaments are cytoplasmic neuronal proteins highly expressed in large myelinated axons. The levels of NfL expressed in body fluids are in proportion to the degree of axonal damage (inflammatory, neurodegenerative, traumatic, and cerebrovascular diseases). The utility of NfL is based on the rapid decline of levels, within days, of effective therapy! The difficulty with another measurable neurofilament antigen, heavy chain, is that heavy chains can clump in some cells and clumps aren’t detected in live bioassays.

Our algorithm for suspect cases of EPM first evaluates levels of surface antigens from S. neurona, 1, 5, and 6, as well as CRP. If there is supporting history from bioassay, we can determine if the horse is experienced or naïve.  Further analysis can determine if there is chronic exposure to the parasite in the environment.  Our algorithm gives less attention to N. hughesii and Borrelia infections unless there is a high prevalence of disease (the algorithm uses zip code for the regional association). There is an association with S. fayeri and anti-myelin protein P2 antibody.  For example, 786 horses with circulating S. fayeri anti-toxin and of those, 610 also had circulating anti-myelin protein P2 antibody. This is important to evaluate on a case-by-case basis, but points out that some equine muscular sarcocystosis can result in a demyelinating polyneuropathy.  There are treatment implications to these data.

It may be useful for a clinician to distinguish between demyelinating, (has antibody against MP2), and non-demyelinating polyneuropathy because treatment and prognosis will vary between these presentations.  While NfL responds quickly to successful treatment, this marker can be present in both demyelinating and non-demyelinating polyneuropathies. A panel of assays are useful to determine the pathogenesis of disease.  Take advantage of our neurodegenerative disease panel by downloading our submission form.  The data will be submitted into our algorithm and our interpretation returned to the veterinarian.

 

 

 

 

horseontrailFinding biomarkers that reflect the amount of peripheral nerve damage (peripheral neuropathies) and that the biomarker will quickly drop in value in response to to effective treatment are desired goals.  The tools we need for developing biomarkers for equine neurodegenerative diseases are not available.  These tools include a laboratory model for each neurodegenerative disease, putative treatments, and a money bin.

There is an alternate path leading to biomarker development and that is the horizon we are chasing.  The biomarker quest project identifies natural cases of disease with neurodegeneration followed by evaluating the data from those cases. Sifting through the data is a process of eliminating the negative, selecting the positive, and interpreting the in-between.  I hear a jingle in there somewhere! Generally diseases follow a typical course, or pathogenesis.  Interpreting enough cases points toward the direction we should take and where to concentrate our assets. Often clues to a direction come from previous researchers.

Our assay to detect anti-myelin protein antibodies was described in the literature over 25 years ago.  Researchers used data from horses with polyneuritis equi (PNE) to study human neurodegenerative diseases. Since then novel developments in molecular biology, such as learning how to fold proteins and identifying sequences (genetic code) that are key elements in inflammation, refined the bioassay’s.  The anti-myelin protein antibody tests affirm a specific disease pathogenesis for PNE in humans and horses that involves myelin degeneration, exposure of the immune system to the myelin protein, and production of anti-myelin protein antibodies.  A boon to our work is being able to glean information from human literature and apply it to our horse cases.  An appropriate application from new human neurodegenerative research is an assay for serum or plasma neurofilament light, NfL.

Neurofilaments are the major cytoskeletal proteins of neurons in both the central nervous system (CNS) and peripheral nervous system (PNS).  Neurofilaments form a structure (lattice) composed of light, medium, and heavy chains. Damage to nerves releases fragments of the neurofilament proteins into the central nervous system fluid (CSF) or circulatory system (plasma or serum). The elevation of neurofilaments in the CNS was observed in patients with amyotrophic lateral sclerosis over 20 years ago. Other neurodegenerative diseases also result in the release of NfL.  Abnormal levels of neurofilaments are associated with the disease process and are not necessarily specific for the etiology.

What is interesting is that human patients with demyelinating and axonal forms of an inherited neuropathy exhibit a slowly progressive, axonal degeneration at a constant rate.  Patients with the inherited disease were monitored for NfL over time and the plasma concentration of NfL increased when values were compared to age matched, healthy controls. It was noted that plasma NfL (pNfL) concentration increases with advancing age in some normal subjects.  An increase in disease severity was correlated with pNfL and pNfL discriminated between patients with the disease versus healthy controls. There are sub-types of this inherited, human neurodegenerative disease and NfL was elevated in all forms of disease. In humans, NfL is elevated in several other neurodegenerative diseases.  Because NfL isn’t specific to etiology it wont’ be useful by itself for a diagnosis; however, because it is a dynamic measure of axonal damage, NfL promises utility for monitoring  a response to treatment. There’s a path we intend to follow!

NfL may be an important biomarker when evaluating a disease known to show no CNS involvement.  In these cases,  changes in concentration of NfL would be attributed to peripheral neuropathies. One caveat is that NfL may be elevated in a T-cell proliferative disorder because neurofilaments are expressed in human T lymphocytes. In human studies, NfH (the heavy chain) was not correlated with disease severity. One proposed reason for the lack of correlation between disease and NfH concentration is that NfH aggregates form and these protein clumps lower detected levels of NfH in fluids.

In humans, and horses with polyneuritis equi,  the gold standard for measuring disease severity for patients is a clinical score. There are several limitations to the clinical score including the scale and a ceiling effect for the most severely affected individuals. It is worth mentioning that a therapeutic benefit for neurodegenerative diseases is to stop progression of disease and is most useful early in the disease process.

A blood biomarker may be more sensitive to multifocal peripheral nerve disease, such as polyneuritis equi.  If proven, that means that when neurofilaments are detected in the serum and disease is supported by the clinical exam, treatment can begin before severe irreparable damage occurs. The most useful interpretation of NfL bioassay is a change in an individual over time or with treatment because intrasubject variability is not expected. The intersubject variability in NfL concentration is a factor we are examining between treated and not treated horses.  Our data may indicate how one horse responds is more meaningful than how groups of horses respond.  That’s why we have our biostatistician on instant redial!

Our goal is to determine the responsiveness of NfL concentration in the clinical course of PNE and relate the assay to demyelination using a second bioassay that detects antibodies against exposed myelin proteins. We have some data to discuss if it relates to a specific case you are working with, the data we have will be at least a year from publication.  If we can help give us a call, we can guide your test selection.

Sometimes we are asked what leads us down a specific road or what inspires an idea.  It is generally the need to find a solution to a problem that is an “outlier”.  An outlier is a case or situation that doesn’t fit the rest of the data.  Everything isn’t a bell-shaped curve, although that is where we start.  As the Covid19 Pandemic sweeps the country, everyone is becoming savvy with interpreting clinical disease and seropositive test results. We thought it would be fun to let you interact with our data as we develop our newest idea.

Our recent work with amyotrophic lateral sclerosis (ALS)  led us to neurofilament light, NfL, a possible biomarker for disease in equine neurodegenerative disease. Serum NfL is a validated marker for ALS as well as a potential pharmacodynamic biomarker that is relevant to ALS therapy development.  It is possible that we could measure NfL in horses and use the result to select candidates for a field study as well as determine a response to treatment.  NfL can change in days with an effective therapy.

We started the analysis of banked serum from some of the cases on which we consulted and will look for patterns.  Our objective is to identify disease conditions and clinically validate serum NfL as a prognostic in our work. When we are satisfied we will send all the data to our biostatistician, the real number cruncher. Of course, he always says we have a biased data set-we look at horses that veterinarians suspect equine protozoal myeloencephalitis or polyneuritis equi.  We have some “gold standard” negative sera and even sera from other countries in which some diseases aren’t present.

What does early, exciting work look like? Here is the result.  Nine sera were selected and examined against many controls.  We detected NfL in horses. A high value in our clinical case analysis was 46.04 ng/ml and the lowest 2.78 ng/ml.  We will select more cases and update our graph so you can follow along! Most likely the graph will get more complicated as we add horses with and without disease, and before and after treatment.

image

If we selected a sample from a horse that you submitted you already got the result by email.  If you would like us to add your sample to our early analysis, please give us a call at the lab.  We are all here hunkered down and continuing to offer you cutting edge work.