© Thomas Tobin 2006, 2013



The Mare Reproductive Loss Syndrome (MRLS), the Septic Penetrating Setal Emboli Pathogenesis thereof and Recently Reported Caterpillar Related Abortions in Camels

 
Thomas Tobin, MVB, MSc, PhD, MRCVS, DABT,  
Professor of Veterinary Science and Professor, 
The GraduateCenterfor Toxicology, 
The MaxwellH.Gluck EquineResearchCenter
Universityof Kentucky
Lexington ,Kentucky40546-0099
www.thomastobin.com

and

Dr. Kimberly Brewer, DVM  
1711 Lakefield North Court
Wellington Fla 33414
 



OVERVIEW: THE MARE REPRODUCTIVE LOSS SYNDROME (MRLS)

In three weeks around Derby Day, May 5th, 2001, Central Kentucky lost one third of its in utero foal crop to the Mare Reproductive Loss Syndrome (MRLS). Total losses were estimated at $500 million. MRLS was characterized by minimal clinical signs in affected mares, relatively non-specific bacterial infections in aborted fetuses, and was also associated with a very small number of highly unusual unilateral uveitis cases. Investigations were immediately begun to determine the cause/s of this syndrome.

Infectious causes were ruled out early. Field and epidemiological studies linked the syndrome to a coincident epidemic of Eastern Tent Caterpillars (ETC), suggesting a toxin driven syndrome. Numerous candidate environmental toxins were investigated and ruled out; then, when the seasonal ETCs became available in April of 2002, they were rapidly shown to produce Early and Late Fetal Losses (EFL and LFL) in pregnant mares.  

Working with the late fetal loss model we saw our first fetal losses at three days after our first 50 g/day caterpillar administration. We immediately abandoned the toxin hypothesis and assigned a primary abortifacient role to the bacteria; the question then became how exposure to the caterpillars caused bacteria to enter the fetal membranes.

Electron Micrographic (EM) studies in the summer of 2001 had shown the ETC setae (hairs) to be barbed. We therefore considered the possible outcomes if small numbers of barbed setal fragments penetrated the intestinal tract, entered intestinal blood vessels, and then re-distributed following cardiac output. Penetration of fetal membranes by these distributing/distributed septic setal fragments/emboli would directly introduce bacterial contaminants into the fetus, where they would proliferate unchecked by the immature fetal immune system. This probabilistic model immediately explained the early and late fetal loss events, and also the relatively rapid onset of late fetal losses, and most importantly, the unique unilateral uveitis cases occurring in horses of any age or sex.  This biologically unique model, identified on July 10th, 2002, we named the Septic Penetrating Setal Emboli (SPSE) Model of MRLS and it was immediately communicated to selected colleagues.

Replication of the LFL abortions in 2003 using a higher (100 g/day) dose of caterpillars and a similar dose of irradiated caterpillars enabled us to demonstrate that the LFL abortions follow a unique probabilistic statistical model elucidated by Accelerated Failure Time analysis. Furthermore, this model showed that the initial intensity of the abortion response was exponentially related to the caterpillar dose;  at high doses of caterpillars exposed mares abort very rapidly, while the caterpillars are underfoot, as had happened in Kentucky in 2001.  On the other hand, at lower doses the abortions are delayed and many abortions occur after the caterpillars have dispersed, obscuring both the role of the caterpillars and their relationship to the syndrome. This highly significant probabilistic mathematical analysis underlying MRLS was developed soon after Memorial Day, 2003. 

These unusual mathematical characteristics of MRLS strongly supported our proposed Septic Penetrating Setal Emboli model. Then, in the fall of 2003, colleagues necropsying pigs that had been dosed with ETC observed numerous intestinal microgranulomas encasing setal fragments. These setal fragments are septic penetrating setal fragments that did not enter blood vessels; as such these findings provided substantial histological evidence in direct support of our septic penetrating setal emboli hypothesis.  Similarly, colleagues dosing rats with ETC in 2001 had observed but not communicated their identification of similar intestinal microgranulomas containing setal fragments, as presented here. 

This simple septic penetrating setal mechanism is obviously a defensive mechanism of the caterpillar, and likely a very ancient one. Review of this proposed mechanism immediately suggests that other caterpillar setae or indeed any mechanically equivalent structure should produce similar syndromes.  More recently a syndrome closely related to MRLS named Equine Amnionitis and Fetal Loss (EAFL) has been identified in Australia. It is associated with exposure to processionary caterpillars, and administration of processionary caterpillars has been shown to abort pregnant mares.  Since then, this and apparently similar caterpillar related field and experimental abortions have been shown to be associated with caterpillars, including Eastern Tent Caterpillars (ETC) in Kentucky ( 02, 03) and Florida (06) and other caterpillar species in Australia (March, 05) and Florida (September, 05).  Additionally, recent reports by Volpato and Di Nardo and their colleagues (2013) make clear that exposure to/ingestion of caterpillars by pregnant camels has long been recognized as a cause of camel abortions in camel pastoralist communities in the Western Sahara.     

DEVELOPMENT OF THE “SEPTIC PENETRATING SETAL EMBOLI” 
HYPOTHESIS OF MRLS, 2002

The proposed septic penetrating setal emboli based pathogenesis of MRLS, first conceived in July 2002, explains many of the unique clinical and epidemiological characteristics of MRLS. The late term fetus is the largest mass of immunologically poorly protected tissue in the mare; as such it is a large and highly accessible and vulnerable target, explaining the very rapid onset of experimental late term fetal losses. The early fetus is a smaller target, but when accessed by a septic penetrating setal fragment is equally vulnerable. The eye is a very small target indeed, explaining the very small number of affected eyes. Additionally, the uniquely unilateral nature of the eye lesions is consistent with and in fact requires the uniquely quantal nature of the septic penetrating setal emboli hypothesis. This hypothesis was shared on a confidential basis with selected colleagues within hours of conception, then with a somewhat broader group of colleagues and selected administrators the following Monday and soon thereafter at the late July 2002 Bain Fallon Lectures, The Gold Coast, Australia.  

SUPPORTING INFORMATION #1:THE UNIQUE MATHEMATICS OF MRLS, 2003

The following year (May 03) more LFL abortion rate data became available and working with Dr. Marie Gantz of the Dept. of Statistics we showed that experimental LFL closely follows a unique probabilistic mathematical model called Accelerated Failure Time analysis. This unusual mathematical signature is consistent with the unique clinical data, the probabilistic nature of the proposed septic penetrating setal emboli hypothesis, and the unusual field epidemiology of MRLS. As such, these findings were rapidly written up and submitted for publication as a paper on the toxicokinetics of MRLS (Sebastian et al., 2003). Additionally, because the septic penetrating setal emboli hypothesis was viewed with skepticism by some of our colleagues, this first paper was drafted avoiding a direct presentation of our underlying septic penetrating setal emboli hypothesis. However, US Copyright for a document describing the overall hypothesis (Copyright#TXU1111484: 2003) was registered with the Library of Congress, June 17, 2003, and a paragraph describing the SPSE hypothesis was added to the paper prior to publication.
   
SUPPORTING INFORMATION #2:   IDENTIFICATION OF THE INTESTINAL MICROGRANULOMAS CONTAINING SETAL FRAGMENTS, 2003

While this Toxicokinetic paper was under review (Fall 2003) we learned that colleagues performing necropsies on ETC dosed pigs had observed multiple intestinal microgranulomas in these animals, each containing a minute central setal fragment. These setal fragments represent septic penetrating setal fragments that have penetrated the intestine but which did not penetrate blood vessels to become septic penetrating setal emboli. Rather, they became encapsulated in intestinal microgranulomas fully consistent with the first intestinal penetration step of this proposed hypothesis. Colleagues at the State University of New York at Cortland kindly made similar slides available to us from some ETC dosing experiments on pregnant rats, as presented in this communication. 

SUPPORTING INFORMATION #3:  LACK OF CLINCAL SIGNS IN AFFECTED MARES

An unusual aspect of MRLS is the lack of clinical signs in affected mares and the inability to culture bacteria from the bloodstream of MRLS mares. The unique eye data allowed us to estimate the actual number of circulating setal fragments in MRLS mares.  Based on these calculations it appears that the number of setal fragments actually distributing in an MRLS mare is very small, likely less than ten/day in field cases. This finding explains the lack of clinical signs in affected mares and the apparent inability to culture bacteria from the blood of clinical and experimental MRLS mares. 

SUPPORTING INFORMATION #4:THE UNIQUE PLACENTITIS OF MRLS

More recently, the placentitis associated with late fetal loss has been recognized by our colleagues as pathologically unique and we believe entirely consistent with this proposed pathogenesis of MRLS, namely the septic penetrating setal emboli or setal hypothesis of MRLS. 

2013: SUPPORTING INFORMATION #5 CATERPILLARS RECOGNIZED AS THE CAUSE OF OTHER ABORTIONS, INCLUDING CAMEL ABORTIONS 

An immediate (July 10th, 2002 ) conclusion from this proposed pathogenesis was that mechanically and bacteriologically equivalent barbed setal fragments (or mechanically equivalent structures) from any source should reproduce the syndrome. As of January 2013, the world score for caterpillar related equine abortions included Eastern Tent Caterpillars (Malacosoma Americana) from Kentucky, Northern Michigan (02) and Florida (06), Processionary caterpillars (Ochrogaster lunifer) in Australia (March 05), Gypsy Moth Caterpillars (Lymantria dispar) in an experimental challenge in Kentucky (03?) and, in September 05 in Florida, field abortions reportedly associated with Walnut Caterpillars (Datana integerrima).  Finally, and most interestingly, there are recent reports by Volpato and Di Nardo and their colleagues (2013) making clear that exposure to or ingestion of caterpillars has for long been known, and unequivocally ("everybody knows") recognized as a cause of abortions in camels among several groups of nomadic camel pastoralists in the Western Sahara, consistent with an earlier report suggesting caterpillars as a cause of abortions in camels (Bizimana, 1994).  

1.  THE MARE REPRODUCTIVE LOSS SYNDROME (MRLS)

During three weeks around May 5th 2001 ,Central Kentucky abruptly lost 20% to 30% of its in-place foal crop. Of foals conceived in the spring 2001, about 2000 were lost, the so-called Early Fetal Losses (EFLs).  Of foals conceived the previous spring and then close to term, at least 600 were lost, the so-called Late Fetal Losses (LFLs). Based on these overwhelming reproductive losses, the syndrome was named the Mare Reproductive Loss Syndrome, or MRLS. The total economic loss for the 2001 MRLS season to Kentuckyand the breeding and racing industry was initially estimated at greater than $330 million. MRLS, as such, had never previously been recognized.  

MRLS  equine abortion



Figure 1.Early (left panel) and late Fetal Losses (right panel) of MRLS.

 

2.  THE ANCILLARY SYNDROMES 
 

Soon three ancillary syndromes were recognized and included in the case definition. These syndromes included about 60 coincident cases of fibrinous pericarditis (epicarditis? Fig 2.), and a slightly smaller number of what can only be described as a unique unilateral panopthalmitis, along with three cases of Actinobacillus encephalitis. These ancillary syndromes, occurring in both male and female horses, were included in the case definition of MRLS. 

MRLS  mare reproductive loss syndrome




Figure 2. The Associated Unilateral Uveitis (left) and Pericarditis (right) Cases: 

 

3.  FIRST OBSERVATIONS:EARLY FETAL LOSS (EFL) OBSERVATIONS

MRLS was first identified on April the 26th 2001 by Dr. Thomas Little, who on that day observed an unusual number of in utero deaths in 60 day-old fetuses in clinically normal mares that he was ultrasounding for sex determination.  These early fetal losses were followed by an overwhelming sequence of early and late fetal losses and, somewhat later, the coincident pericarditis, uveitis and encephalitis syndromes, occurring in very much smaller numbers in horses of all ages and sexes were identified.  

4.  THE PATHOLOGY OF THE LATE TERM FETUSES

The abortions, and particularly the late-term abortions, showed characteristic pathology. There were an increased number of "red bag" presentations with weak and stillborn foals. Necropsy evaluation of aborted foals showed inflammation of the amnion, the umbilicus [funisitis], and the fetal lungs. Bacterial isolates included a high proportion of alpha streptococci, actinobacillus and many other isolates, including Serratia. The Early and Late abortion syndromes peaked on about May 5th 2001, Derby Day, and were essentially over by the end of May.  

5.  THE EXCLUSION OF INFECTIOUS CAUSES

Within the first week extensive bacteriological and virological work, the essential lack of clinical symptoms in affected mares, as well as the essentially simultaneous appearance of the syndrome across the Bluegrass largely excluded an infectious cause.  Work then focused on identifying an environmental toxin or agent as the cause of the syndrome. 

6.  THE INITIAL TOXICOLOGICAL EVALUATIONS

Within two weeks toxicological investigations had ruled out nitrate/nitrite toxicity; work then focused on toxins with reproductive associations; additionally, because of a coincident unusually intense infestation (plague?) of Eastern Tent Caterpillar (ETC, Malacosoma americana), the possibility of an association of the syndrome with the black cherry tree/Eastern Tent Caterpillar biological system was very carefully evaluated.  

7.  CLUE #1: THE TIME AND PLACE ASSOCIATIONS WITH THE CATERPILLAR

Within three weeks preliminary field investigations by Dr. Jimmy Henning of the Univ. of Kentucky College of Agriculture pointed to a close association between the presence of black cherry trees/Eastern Tent Caterpillars and MRLS. Additionally, preliminary toxicological evaluations suggested increased concentrations of cyanide in the hearts of some LFL fetuses. The association of MRLS with presence of the caterpillar was later confirmed by a rigorous epidemiological survey by Dr. Roberta Dwyer of the Gluck Equine Research Center and her associates (Dwyer et al., 2003). These early findings were critically important, because they focused attention squarely on the caterpillars, which were present in enormous numbers at many locations in central Kentucky at the time of the MRLS outbreak.     

caterpillars  MRLS  mare reproductive loss syndrome

Figure 3a.The large numbers of caterpillars during the outbreak.Here they cover a water bucket.
  

  Figure 3b.The Eastern Tent Caterpillar
mare reproductive loss syndrome     MRLS



8.  CLUE #2: THE BARBED SETAE 
    
ON THE CATERPILLARS

The early focus on the caterpillar led Mrs. Pat Van Meter and me to examine them under the Electron Microscope at about the time of the 2001 outbreak (Fig 3c)).  We noted, in passing, the barbed nature of the very fine setae that cover the exterior of the caterpillar and we also noted that in some caterpillar species these setae are venomous and extremely effective offensive weapons.  Although not specifically recognized as such at the time, these were critically important observations. 




Figure 3c.Barbed setae
MRLS    mare reproductive loss

9.  THE ENVIRONMENTAL TOXIN HYPOTHESIS
  

The search for the cause of MRLS at first focused on candidate environmental toxins, at least in part because Eastern Tent Caterpillars were unavailable, being out of season.  Over the next nine months rigorous evaluation of the possible roles of plant estrogens, mycotoxins, ergot alkaloids and Hemlock alkaloids failed to produce any evidence suggestive of a role in MRLS. The possible role of cyanide in MRLS was very carefully examined because of the close association of the Eastern Tent Caterpillar with cyanogenic Black Cherry trees. Rigorous experimental evaluations failed to show any association between cyanide and MRLS (Camargo et al., 2002; Dirikolu et al., 2003). Additionally, vigorous efforts to induce abortions with a variety of toxins, including cyanide and mandelonitrile (an intermediate in the release of cyanide from cherry leaves) and a number of other candidate toxins failed, each and all. 


10.  SPRING 2002: RETURN OF THE CATERPILLARS AND 
        EXPERIMENTAL AND FIELD ABORTIONS


Return of the Eastern Tent caterpillar in spring 2002 showed that exposure of early pregnant mares to Eastern Tent Caterpillars rapidly produced early and late fetal losses, and the abortions closely resembled field MRLS (Webb et al., 2004). By May 2002 it was therefore clear that MRLS was very closely associated with the Caterpillar itself; the question then became how the caterpillar, or factors associated with the caterpillar, produced the abortions.


11.  CLUE #3: RAPID ABORTIONS IN THE 2002 LATE FETAL LOSS
        EXPERIMENT:THE BACTERIAL ABORTION THEORY

Towards the end of the 2002 ETC season we administered 50 g/day of Northern Michigan ETC by intubation to Late Term pregnant mares.  The first abortions occurred very rapidly, within 72 hours, suggesting to us that the primary driving event in the abortions was most likely bacterial proliferation in the fetus and/or fetal membranes.  We therefore abandoned the toxin hypothesis and review of this changed etiology for the abortions and the well established tissue penetrating capabilities of barbed caterpillar setal fragments led us virtually immediately (July 10th, 2002) to the Setal/Septic Penetrating Setal Emboli Hypothesis of MRLS.  

Figure 4. CLUE #2: Very rapid time course of abortions following dosing of mares with 50 (02) and 100 (03, nonirradiated or radiated) grams of Eastern Tent Caterpillars (ETCs) per day for 10 days. The solid lines are the best fit regression for the data points. The calculated x-axis intercepts (apparent lag times) are 20 (100g), 37 (50g), and 193 hours (100g irradiated) after the first dose of ETCs. mathematics of MRLS  mare reproductive loss

12.  CLUE #3, 2002:  BARBED SETAL FRAGMENTS: THE SEPTIC 
        PENETRATING SETAL EMBOLI (SPSE) HYPOTHESIS OF MRLS

Considering the barbed setae of the caterpillars, we proposed that MRLS results from movement driven intestinal penetration by barbed setal fragments, with their associated bacteria (septic penetrating setae), followed by blood vessel penetration and hematogenous redistribution of a proportion of these setal fragments, now forming "Septic Penetrating Setal Emboli" (Tobin et al., 2003, 2004).  Systemic distribution of such embolic setal fragments would follow cardiac output, as is well described and understood with regard to drug distribution. These septic setal fragments would then randomly lodge in distant tissues, and their retained structurally based ability to penetrate moving tissues would allow them to distribute their bacterial contaminants in any moving distant tissue in which they lodged. Significant pathology would result only if the penetrated tissue was poorly protected immunologically. A reduced local immune response was, and is, a critical part of the proposed pathogenesis.

Figure 5.Schematic Construct of Intestinal wall and Intestinal Venule Penetration by Barbed Setal Fragment.
mrls    mare reproductive loss syndrome

13.  THE CRITICAL ROLE OF TARGET TISSUE MASS AND 
        REDUCED IMMUNE RESPONSES

The late term fetus is the largest mass of immunologically poorly protected tissue in the mare. As such, it is a large and vulnerable target, explaining the unusually rapid onset of our experimental late term fetal losses. The early fetus is a much smaller target and therefore less likely to be "hit," but an equally vulnerable target once hit.  The eye is a very small target indeed, explaining the very small number of affected eyes.  Additionally, the uniquely unilateral nature of the uveitis cases is entirely consistent with and requires the discrete, quantal, and very small  number, nature of the proposed triggering septic penetrating setal emboli insults. 
 

13.1.  INITIAL RECEPTION OF THE SPSE /
MRLS HYPOTHESIS

This "septic penetrating setal emboli" hypothesis as a proposed mechanism for MRLS was conceived on or about July 10th 2002. It was communicated to selected colleagues within hours and within days to selected administrators; it would be fair to say that its reception was unenthusiastic, to the point that its presentation at many levels was perceived as being discouraged.  

14.  THE CLINICAL SIGNATURES OF THIS PROBABILISTIC MECHANISM  

14.1.  THE EFFECT OF TARGET SIZE

This proposed pathogenesis depends on the statistical probability of a distant tissue penetration event occurring for any given poorly immune protected tissue. As set forth above, it was immediately clinically obvious that the actual size (target size) of the immunologically poorly protected target would influence the probability and therefore the rate at which any clinically effective septic setal “hit” (or clinically observable result) would occur. These factors immediately explained the rapid onset of the LFL experimental abortions, given the large size of the Late Fetus, which on a weight basis receives at least 15% of the mare’s cardiac output. The early term fetus is much smaller and receives proportionally much less cardiac output. The early term fetus is therefore statistically much less likely to receive a setal “hit,” thereby explaining the much slower rate of occurrence of the EFL experimental abortions. Indeed, very early fetuses were perceived as being protected from MRLS, likely a result of their very small target size. The single eye is a very small target indeed, explaining the relative rarity of the single eye events and the complete lack of any double eye events.  We were soon, however, to be presented with additional and very compelling statistical evidence of the probabilistic mathematics underlying MRLS, namely a unique mathematical signature called Accelerated Failure Time (AFT).  

Figure 6. The Mathematical model of MRLS -- Actual abortion rates following dosing with Eastern Tent Caterpillars

mathematics of MRLS


Accelerated Failure Time (AFT) analysis of MRLS
 

Assumed that time until abortion has a log-normal distribution a survival model that well fits the data is an Accelerated Failure Time Model. The model takes the form:
                         


This model includes an intercept term (β0) and an error term (σεi). Ti is the time that abortion occurs for individual i, and β1 through βk are coefficients for covariates that might affect abortion time.
       

14.2.  THE UNUSUAL PROBABILISTIC MATHEMATICS UNDERLYING 
           MRLS:ACCELERATED FAILURE TIME ANALYSIS

Further evidence for the unique probabilistic mathematics underlying MRLS became apparent as follows. On Memorial Day weekend of 2003 we
plotted the individual time courses of each of the 2002 and 2003 LFL experiments (Figure 4).  The plotted time courses of the three individual sets of MRLS abortions in Figure 4 yielded a clear cut family of time course curves. Each curve clearly fit a smoothly defined exponential rate from the first to the last abortion; however, and most unusually, each curve appeared to commence after a specified “lag time,” and this lag time appeared to be related to the rate at which the abortions occurred. We took Figure 4 to Dr. Marie Gantz of the Dept of Statistics, suggesting that we had half of the equation for MRLS, and requesting that she provide a factor describing the lag times and we would have the complete equation for MRLS. Dr. Gantz pulled the required equation "off the shelf," a probabilistic mathematical analysis called Accelerated Failure Time analysis (Fig 6). As set forth in Figure 6, the time course of MRLS LFL abortions closely follows this very unusual probabilistic mathematical model, entirely consistent with the probabilistic nature of our proposed Septic Penetrating Setal Emboli model/hypothesis of MRLS (Figs 4 and 6,  P: <.0001 .0052 <.0001).

In lay terms, what this model says is that it is the rate at which the setal distribution events occur that determines the speed of onset of an MRLS event. If exposure to caterpillars is high, and the rate of setal penetration/ distribution events is high, then the lag time is minimal and the onset of occurrence of MRLS abortions is very rapid. On the other hand, if exposure to the caterpillars is low, then the rate of setal penetration (or entry into affected tissues) is low, the lag time is long and the onset of the first MRLS event is much slower, taking days to weeks or longer, as set forth in Fig 6 above.

14.3.  THE EXPONENTIAL DOSE / TIME RESPONSE 
           RELATIONSHIP FOR MRLS

The second important mathematical characteristic of this type of mechanism is that the initial intensity of the response, in this case the rate at which the abortions occur, is exponentially related the dose. Thus when exposure to the caterpillars is very high, the lag time is very short, the abortions onset very rapidly and are closely linked in time and place to the presence of the caterpillars, as occurred in Kentucky in 2001, and which constituted clue #1, above. 

O
n the other hand, when exposure to the caterpillars is low, the onset of the abortions is delayed, their number is greatly reduced, and the delayed abortions may occur out of phase with the presence of the caterpillars. Additionally, horses in Kentucky are unlikely to be exposed to caterpillars for more than the short ETC caterpillar migration season, about two weeks or so in Kentucky. These unusual characteristics of the Kentucky caterpillar driven abortions explain both the exceptional intensity of the 2001 Kentucky ETC related abortion storm and the fact that these abortions, although previously identified as a discrete unexplained class of abortions, had never previously been recognized as being related to (or caused by) the caterpillars.   

In sum, these data show that experimental and field cases of MRLS follow a virtually unique probabilistic mathematical model called Accelerated Failure Time analysis, which analysis is entirely consistent with the Septic Penetrating Setal Emboli hypothesis and our proposed underlying probabilistic mathematics of MRLS (Fig 6).  These findings in the summer of 2003 enabled us to submit for publication our first formal toxokinetic/statistical analysis of MRLS (Sebastian et al., 2003) and then to add, while this manuscript was in press, a summary description of its septic penetration setal emboli pathogenesis.  Soon thereafter, this publication was followed by the full septic penetrating setal emboli hypothesis paper (Tobin et al., 2004).  

15.  Fall O3: IDENTIFICATION OF INTESTINAL MICROGRANULOMAS 
       CONTAINING SETAL FRAGMENTS

Figure 5 is a construct from our 2004 paper designed to communicate our visualization of blood vessel penetration by a septic penetrating setal fragment.  What Fig 5 does not address is what happens to the many septic penetrating setal fragments (SPSF) that enter the intestinal tract and do not enter a blood vessel. Our hypothesis might suggest that at least some if not all of these SPSF should work their way through the intestine to its peritoneal surface and cause peritonitis, which we were fully aware was not occurring. This question was the next MRLS conundrum to be answered. 

In we understand the late summer of 2003 a colleague working on a pig model of MRLS necropsied some pigs which had been dosed for several weeks with ETC. Necropsy of these ETC dosed pigs showed evidence of a very large number of intestinal setal penetrations in the form of microgranulomas containing setal fragments in the intestinal tract, fully consistent with our setal hypothesis of MRLS. Discussing these rumored findings with another colleague, Dr. Terry Fitzgerald of Cortland University, he noted that they had made similar observations on ETC-dosed rats (Fig 7), which slides he immediately shared.   

These setal microgranulomas are interesting, because it is the microgranulomas, small enough in and of themselves, which had caught the attention of the pathologists. In the center of each of these microgranulomas was a much smaller setal fragment. These are, of course, our postulated “septic penetrating setal fragments”, that drive MRLS, each caught and immobilized in a little connective tissue microgranuloma capsule and presumably being slowly degraded. In any event, they are no longer “penetrating” and present little or no further risk to the animal [or to us].  And this, of course, is presumably how our intestinal tracts protect us from having small barbed fragments fully penetrate our intestinal tracts and induce peritonitis, presumably an anciently evolved digestive tract protective mechanism. The answer to our intestinal penetration conundrum is that the barbed penetrating fragments are isolated and walled off in connective tissue capsules, and that effectively ends their penetrating ability. 

mrls setal fragment


Figure 7.Intestinal Microgranuloma in an ETC dosed Rat containing an ETC Setal Fragment at its Center. Courtesy of Dr. Terry Fitzgerald and Colleagues, Cortland University, Cortland, New York. 

 

  
16.  SUPPORTING INFORMATION:THE PERICARDITIS CASES  
  
The pericarditis cases are also best explained by the septic penetrating setal emboli hypothesis. Of all tissues in the body, the contracting heart is one through which one might expect septic penetrating setal emboli to migrate/be driven fastest. The central role of the heart in the circulatory system and its ongoing contractile activity may suggest that at least 50% of setal fragments entering cardiac muscle will wind up in the pericardial fluid. Bacteria cultured from the pericarditis cases are those associated with MRLS, although no bacteria were cultured from some pericarditis cases. This may suggest loss of septic contaminants during passage of the septic material through the cardiac musculature. The pericarditis cases are clearly most consistent with and best explained by the septic penetrating setal emboli hypothesis.  
  
There is an interesting contrast between the potential outcomes of intestinal and cardiac tissue penetration by septic penetrating setal fragments. In the intestine, penetrating setal fragments are rapidly “walled off” and thereby neutralized. This is presumably an ancient intestinal defensive strategy designed to prevent the potentially lethal outcome of peritonitis resulting from intestinal penetration by barbed structures. Given the importance of preventing setal penetration, it seems likely that this local “walling off” process is accompanied by a reflex local reduction in intestinal contractility in the affected area. On the other hand, such a strategy is not available in the heart, where fragment movement through cardiac tissue would presumably be much faster than through the intestinal wall and a local reduction or cessation in contractility is a much less viable option.
  
17.  SUPPORTING INFORMATION:  LACK OF CLINICAL SIGNS
        IN AFFECTED MARES:

An unusual aspect of MRLS was the lack of clinical signs in affected mares and the failure to culture bacteria from the bloodstream of MRLS mares.The unique eye data allowed us to estimate the number of circulating setal fragments in MRLS mares.It appears that the number of setal fragments actually distributing in an MRLS mare is very small, likely less that ten/day in field cases.This finding explains the essentially complete lack of clinical signs in affected mares and the practical inability to culture bacteria from the blood of clinical and experimental MRLS mares. In lay terms, in MRLS at least, there are simply too few setal fragments circulating to produce an observable clinical response in the mare herself.  

18.  SUPPORTING INFORMATION: UNIQUE PLACENTITIS OF MRLS:

More recently, the placentitis associated with MRLS has been formally recognized by pathologists as unique and clearly distinguishable from the classic and well described ascending and hematogenous placentites, and also, we believe, fully consistent with the unique SPSE mechanism underlying MRLS.  

19.  SUPPORTING INFORMATION:  OTHER CATERPILLAR DRIVEN
       ABORTION EVENTS

An obvious and early prediction from this proposed pathogenesis is that mechanically and bacteriologically equivalent barbed setal fragments or analogous structures from any source should reproduce the syndrome. As of January 2013, the world score for caterpillar related abortions is, to our knowledge, as follows:

19.1. KENTUCKY

Caterpillars have produced numerous field abortions, MRLS in central Kentucky and adjacent states (2002, 03) and in Florida (2006). Also, Eastern Tent Caterpillars from Kentucky and Northern Michigan have been shown to be fully abortigenic in experimental situations. 

In a challenge experiment in Kentucky, reported in 2004, one of four pregnant mares exposed to Gypsy Moth Caterpillars aborted, although we must note that the investigators chose to distinguish between this single "GMC" related abortion and classic MRLS abortions.  

19.2.   FLORIDA AND NEW JERSEY, 2006

In Florida, one of two pregnant mares exposed to high densities of Walnut Caterpillars (Datana integerrima) aborted within one week, and the second aborted three months later.  

19.3. AUSTRALIA, 2005

Processionary caterpillars (Ochrogaster lunifer) were seen in association with field abortions in the Australian Hunter Valley area (March 2005), and soon thereafter (August 2006) these caterpillars were reported to be abortigenic in experimental tests. In point of fact, the principal investigator in this matter was reportedly at first reluctant to consider the possibility of caterpillar involvement in these equine abortions. Following the suggestion of an Australian horse farmer who had suffered significant abortion related losses, the Australian investigator contacted our research group and our suggested approach to the described abortion matter was unequivocal. In our opinion, the described syndrome was remarkably similar to MRLS, and as such, our recommendation was, caterpillar experiments first, any and all other experimental approaches, second. The suggested caterpillar administration experiments were performed, and the caterpillar administrations rapidly reproduced abortions. This syndrome, which the Australians named Equine Abortion and Fetal Loss (EAFL) is now recognized as being closely related to and broadly similar to MRLS (Cawdell-Smith et al., 2012; Todhunter, et al., 2009).

19.5.  AFRICA, 2013, CATERPILLARS AND CAMEL ABORTIONS

It is said there is nothing new under the Sun. Just recently, researchers investigating the causes of abortions in camels reported that the Sahrawi (literally “people of the desert”) of the north-western Sahara have long known of the association between caterpillar ingestion and camel abortion (
Volpato et al., 2013). Though mentioned as a possible cause of camel abortions by Bizimana in 1994, African caterpillars have until now received little or no recognition by the scientific community as abortifacients. The Sahrawi pastoralists are nomadic people who are dependent on camels for their livelihood. Among these cultures, it has long been traditional knowledge that a caterpillar-borne reproductive loss syndrome exists. The Sahrawi people call this “duda,” which is an Arabic name for “caterpillar.” Duda in camels presents many similarities to Mare Reproductive Loss Syndrome (MRLS), suggesting a similar pathogenesis of the disease. 

Caterpillars affecting mares and camels are both of the family Lasiocampidae. The caterpillar of MRLS (the Eastern Tent Caterpillar) is a hairy caterpillar with barbed setae. The caterpillar affecting Sahara camels is known by the Sahrawi as shedbera, and is described as a hairy caterpillar. Unfortunately because of the absence of an identified adult moth, the exact species of the Lasiocampidae caterpillar has yet to be classified. Family Lasicampidae larva  are generally densely hairy, feed on foliage from trees, and may build communal tents that contain the caterpillar’s hairs and setae. Female moths of these species lay a large number of eggs which, if the environmental conditions are favorable, may lead to outbreaks of the caterpillars.

Both Duda and MRLS outbreaks are associated with large numbers of caterpillars in the environment. When conditions are such, both horses and camels are likely to ingest caterpillars while foraging. With Duda, the caterpillars are associated with Acacia trees which Western Sahara camels feed on, as these trees may be the only green pasture available. In the US, MRLS has been associated with environmental conditions that cause greatly increased numbers of Eastern Tent Caterpillars in trees and pastures where mares graze. Sahrawi pastoralists differentiate duda as the cause of abortions in camels from other causes based on their sudden and episodic patterns of abortion storms associated with an abundance of shedbera caterpillars on Acacia foliage. 

Like MRLS, abortion is the primary clinical sign of Duda. The clinical signs associated with Duda include, in pregnant camels, abortions and uterine prolapse. Calves that are born with Duda show weakness, red eyes, hair falling out, astasis and incoordination, joint effusion, swelling of lymph nodes, and diarrhea. Sahrawi people are clear that duda is “born with the calf”, meaning it is not caused by an external agent after birth. These clinical signs of duda share similarities with the clinical signs and presentations of MRLS. Adult camels who are not pregnant do not show clinical signs of duda, even when ingesting the caterpillars, again similar to MRLS.

Sahrawi herders are well aware that it is difficult for them to prevent duda, as the only means of prevention of this syndrome would be to stop the ingestion of foliage. As caterpillars may outbreak suddenly on the main foraging source for camels, it is difficult for the herders to prevent ingestion of caterpillars by grazing camels. The Sahrawi people also do not have an effective treatment for weak affected calves born during a duda outbreak, and most of the affected calves die. Photographs of the duda caterpillar reveal a remarkable similarity to the Eastern Tent Caterpillar. 

mare reproductive loss syndrome     MRLS
Eastern Tent Caterpillar  

Duda
Caterpillar among the thorns of the Acacia tree.
(Volpato et al., 2013)

20.  CONCLUSIONS, JANUARY 2013

We believe that only satisfactory hypothesis offered to date to explain the unique constellation of clinical syndromes that comprise MRLS
 
 1. the etiology of each of the four MRLS syndromes
 2. the unique probabilistic mathematics underlying MRLS
 3. the unique epidemiological presentation
 4. the unique pathogenesis 
 5. the gross histopathological and bacteriological characteristics of MRLS 
 6. the predicted and similar now-emerging caterpillar-related abortions in 
 horses and most recently in camels
  
is the Setal/Septic Penetrating Setal Emboli hypothesis of MRLS.

We also respectfully note that this hypothesis is
without precedent in the biomedical literature.

A detailed presentation of this hypothesis is available for viewing in a web slide show (Tobin 2005)  which was first presented at the July 2002 Bain-Fallon symposium in Gold Coast, Australia. While the presentation has since been expanded and updated, it retains substantial portions of the first public presentation of the Septic Penetrating Setal Emboli Hypothesis of MRLS made in Australia in 2002.  
  
   
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
REFERENCES


Adkins, P.  (2005)  Tracking the source of mare reproductive loss syndrome.  Equus, January, pp 46-9.

Biziman
a, N.  (1994)  Traditional Veterinary Practice in Africa.  Deutsche Gesellschaft fur Technische Zusammernarbeit.  pp 348-9.

Camargo, F., Dirikolu, L, Sebastian, M., Hughes, C., Crutchfield, J., Harkins, J.D., . Boyles, J., Troppmann, A., McDowell, K., Harrison, L., Tobin, T. (2002). The toxicokinetics of cyanide and mandelonitrile in the horse and their relevance to the Mare Reproductive Loss Syndrome.  Proc 4th International Conference of Racing Analysts and Veterinarians Orlando, FL.

Cawdell-Smith AJ, Todhunter KH, Anderson ST, Perkins NR, Bryden WL.(2012) Equine amnionitis and fetal loss: mare abortion following experimental exposure to Processionary caterpillars (Ochrogaster lunifer). Equine Vet J.44(3):282-8.  
doi: 10.1111/j.2042-3306.2011.00424.x. Epub 2011 Aug 5.

Dirikolu L, Hughes C, Harkins D, Boyles J, Bosken J, Lehner F, Troppmann A, McDowell K, Tobin T, Sebastian MM, Harrison L, Crutchfield J, Baskin SI, Fitzgerald TD. (2003) The toxicokinetics of cyanide and mandelonitrile in the horse and their relevance to the mare reproductive loss syndrome.  Toxicol Mech Methods. 13(3):199-211.

Dwyer, R.M., L.P. Garber, J.L. Traub-Dargatz, B.J. Meade, D. Powell, M.P. Pavlick and A.J. Kane. (2003). Case-control study of factors associated with excessive proportions of early fetal losses associated with Mare Reproductive Loss Syndrome in central Kentucky during 2001. J. Am. Vet. Med. Assoc. 222: 613-619.

Neundorf, F.  (2007)  Caterpillars are aborting our mares. Australian Performance Horse, January, pp 56-8

Sebastian M, Gantz M, Tobin T, Harkins J, Bosken J, Hughes C, Harrison L, Bernard WV, Richter D and Fitzgerald TD. (2003) The Mare Reproductive Loss Syndrome and the eastern tent caterpillar: A toxicokinetic/statistical analysis with clinical, epidemiologic, and mechanistic implications. Vet Ther,4(4):324-339.

Tobin, T. (2002)  MRLS and associated syndromes: toxicological hypotheses.  Proceedings of the First Workshop on Mare Reproductive Loss Syndrome, eds Powell, DG, Troppman, A., Tobin, T., University of Kentucky Agricultural Experiment Station. Lexington, Kentucky, p 75.

Tobin T, Harkins JD, VanMeter PW, Fuller TA: (2004) The Mare Reproductive Loss Syndrome and the Eastern Tent Caterpillar II: A Toxicokinetic/Clinical Evaluation and a Proposed Pathogenesis: Septic Penetrating Setae.Int. J Applied Res in Vet Med, 2(2):142-158.  http://www.jarvm.com/articles/Vol2Iss2/TOBINJARVMVol2No2.pdf
    
Tobin, T (2005) Mare reproductive loss syndrome (MRLS) and the eastern tent caterpillar (ETC): a toxicological / clinical investigation and a proposed pathogenesis; septic penetrating setae. Available for viewing on-line as a powerpoint presentation at http://thomastobin.com/mrlspp.htm 

Todhunter KH, Perkins NR, Wylie RM, Chicken C, Blishen AJ, Racklyeft DJ, Muscatello G, Wilson MC, Adams PL, Gilkerson JR, Bryden WL, Begg AP. (2009) Equine amnionitis and fetal loss: the case definition for an unrecognised cause of abortion in mares. Aust Vet J. 87(1):35-8.

Volpato G, Di Nardo A, Rossi D, Saleh SM, Broglia A (2013) 'Everybody knows', but the rest of the world: the case of a caterpillar-borne reproductive loss syndrome in dromedary camels observed by Sahrawi pastoralists of Western Sahara.Journal of Ethnobiology and Ethnomedicine.9(5)   (10 January 2013) http://www.ethnobiomed.com/content/pdf/1746-4269-9-5.pdf 

Webb, B.A., W.E. Barney, D.L. Dahlman, S.N. DeBorde, C. Weer, N.M. Williams, J.M. Donahue and K.J. McDowell.  (2004). Eastern tent caterpillars (Malacosoma americanum) cause Mare Reproductive Loss Syndrome. J. Insect Physiol. 50:185-193.


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

FURTHER
READING  

1. Proceedings of the First Workshop on Mare Reproductive Loss Syndrome on the Web: http://www.ca.uky.edu/agc/pubs/sr/sr2003-1/sr2003-1.htm. A one-page summary of the hypothesis is presented on page 75.

2. For more on MRLS – including caterpillar control, disease prevention, additional research and archives – go to www.uky.edu/ag/vetscience/mrls/index.htm.

3. A complete listing of scientific articles addressing MRLS and published by the Tobin research group is located at http://equinetoxicology.com/mrlspubs.htm

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SUPPORT CREDITS

From the Equine Pharmacology, Therapeutics and Toxicology Program of the
Maxwell H. Gluck Equine Research Center, University of Kentucky, Lexington, KY.

This research was supported by grants from the USDA Agriculture Research Service Specific Cooperative Agreement #58-6401-2-0025 for Forage-Animal Production Research, the Kentucky Department of Agriculture, the Kentucky Thoroughbred Association Foundation, the Horsemen’s Benevolent and Protection Association, and Mrs. John Hay Whitney.