SEARCH

SEARCH BY CITATION

Keywords:

  • horse;
  • inflammation;
  • laminitis;
  • mast cell;
  • RT-qPCR;
  • whole blood biomarker

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Inflammatory damage to the digital laminae, a structure responsible for suspension of the distal skeleton within the hoof capsule, results in a painful and often life-threatening disease in horses called laminitis. There can be many diverse causes of laminitis; however, previous work in the horse has suggested that in each case, the inflammation and resulting tissue damage is consistent with the action of mediators released from mast cells (MC), as well as the downstream consequences of their activation. The recent development of molecular genetics tools to characterize cells based on their transcriptional activity makes a new approach for measuring MCs possible. Healthy thoroughbred horses from a variety of age groups were used to assess the amount of variation in KIT (encoding mast cell growth factor receptor) and TPSB2 (encoding mast cell tryptase beta 2) gene expression present in the population and to establish “normal” values. Horses (n = 9) with a wider range of body condition scores (3–8), because of a more lax management setting that could predispose them to laminitis, had significantly higher KIT expression in circulating peripheral blood cells than horses under individualized management conditions (n = 10) that produced ideal body condition scores (4–6) (mean 2.573-fold, P < 0.0005). Likewise, horses affected with acute laminitis (n = 11) had elevated expression of TPSB2 (2.760-fold, P = 0.0011) relative to control horses (n = 15). These data suggest that investigation of MC-related genes KIT and TPSB2 may be effective to assay MC population and activity. More work is needed to refine the diagnostic criteria to better describe at what point MC activation occurs and illustrate the use of gene expression assays in clinical cases of laminitis. Additionally, MC activation is associated with inflammatory disease in several mammalian species and may prove a valuable therapeutic target in the horse.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Laminitis is one of the most common causes of death in the horse. The precise aetiology of the disease is unknown, although it can occur following a wide variety of inflammatory stimuli. It is characterized by a breakdown of the laminae, a complex structure that suspends the distal phalanx within the hoof capsule. This inflammation destabilizes the foot, leading to rotation and/or displacement of the distal phalanx. Accompanied by bruising and oedema, it is extremely painful and debilitating. Vascular changes are associated with the disease, in the form of both increased regional blood flow and regional hypoperfusion, especially in the laminar capillary bed, leading to vascular damage and poor perfusion of the tissues (Moore et al. 2004). The diverse mechanisms for initiation of laminitis are varied and poorly understood. They can include obesity, equine metabolic syndrome, carbohydrate overload (over-consumption of grain or lush pasture), black walnut toxicity, endotoxemia, sepsis, mechanical injury and glucocorticoid administration, to name but a few (Hood 1999). These varied triggers are frequently grouped in to three broad categories, endocrinopathic, toxic and traumatic laminitis. Regardless of the triggering event, be it physical trauma, obesity/endocrine imbalance or bacterial/toxic assault, the end result is the same, lamellar breakdown.

Mast cells are sensitive to the types of biological situations responsible for the development of laminitis and play a central role in inciting damage caused by the types of triggers mentioned above. Mast cells (MCs) are a specialized subset of immune cells of bone marrow origin. Specifically, connective tissue mast cells act as ‘gatekeepers of the microvasculature’, are located in vascularized connective tissues and are often a first defence against attack. Mast cells respond to a variety of stimuli, just a few of which include endocrine imbalance, bacteria and mechanical injury (Gruber & Kaplan 2005). Once activated, they quickly degranulate, releasing powerful pre-formed mediators including histamine, heparin, superoxide, tryptase and chymase as well as other Matrix metalloproteinases (MMPs), platelet-activating factor, TNF-α and many inflammatory cytokines (Gruber & Kaplan 2005). The mediators produced by MCs lead to rapid inflammation in their resident tissues followed by the recruitment of leucocytes and additional MCs. Although equipped to protect the body from a variety of assaults, MC activation, particularly in a confined area, can cause more damage than good (Lazarus et al. 2000).

Because of their broad tissue distribution and migratory development, in vivo study of MCs is difficult in large mammals where invasive procedures must be limited. Identification of cells in tissue is traditionally carried out by morphology, histochemical staining and immunological staining. Identification of MCs by histology is often dependent on subjective quantification of heterochromatic staining of the granules. In humans, strategies have been developed to take advantage of the accessibility of whole blood and the transient presence of MC products in the circulation. These techniques measure the presence of mediators including tryptase and histamine by ELISA and have proven useful for conditions ranging from mastocytosis to anaphylaxis and heart attack. The recent completion of the equine genome sequence and the resulting novel molecular genetics tools specific to the horse are useful to characterize cells based on their transcriptional activity and thus make a new approach possible. Using the genome sequence, primers and probes can be quickly and easily generated for our target of interest. We have chosen RT-qPCR for mast cell-specific genes, including that encoding mast cell growth factor receptor (KIT, expressed exclusively by mast cells in the circulation, aka CD117) and TPSB2 (encoding mast cell tryptase beta 2, a mediator produced by maturing and activated MCs) (Liu et al. 2006). The KIT gene has been shown to be expressed specifically by circulating MC progenitors, the precursor cells for mature MCs that appear to be similar to basophils (Schernthaner et al. 2005). Measuring changes in KIT gene expression in blood will illustrate changes in the circulating population of MC precursors. Finally, tryptase (encoded by TPSB2) is a mast cell-specific product that is induced during maturation and activation (Liu et al. 2006). Presence of TPSB2 mRNA in the peripheral circulation will reflect the presence of maturing mast cell precursors as they migrate to their destination tissues (Welker et al. 2005). We therefore hypothesize that the detection of these genes by qRT-PCR could allow their use as biomarkers for determining the potential aetiologies and clinical severity of laminitis.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Horses

Experiment 1  Thirty thoroughbred horses were selected from a single farm to establish reference values for relative gene expression and evaluate changes relative to age and body condition. Fifteen horses were between 6 months of age and 2 years. The remaining 15 individuals were 5 to 23 years. All horses were healthy and on a carefully managed diet designed to maintain optimal body condition. The mean body condition score (BCS) of this group was six (Henneke et al. 1983). As defined by Henneke and coauthors, body condition scores are based on assessment of adiposity at five anatomical regions that is combined in to a single score that ranges from one to ten. Scores between four and six are ideal, while scores above seven are associated with obesity, and scores below three are associated with poor condition. Samples from this group of horses were obtained over a 2-day period in late September.

Experiment 2  Two groups of thoroughbred mares (‘A’ and ‘B’) on two different farms with two different management situations were identified to compare the effect of environment on MC-related genes (Table S1-a). During these months, lush grazing is plentiful in well-managed pastures, thus exposing our study horses to one of the most common risk factors for laminitis. Mature, age-matched Group A mares were selected from among the horses used in Experiment 1. These ten mares resided primarily at pasture but were kept indoors overnight during inclement weather, including early spring. On farm A, a major focus of management was to maintain a BCS of five by restricting access to lush pasture and feeding supplemental grain when necessary. At the time of sampling, group A mares were free from lameness, had body condition scores ranging from five to six and were on average 15.3 years (range 10 to 23 years). At farm A, the incidence of laminitis was reported by management to be approximately 2%. Group ‘B’ was comprised of nine mares residing at pasture, including during the lush spring and fall seasons associated with pasture laminitis. Supplemental hay was used to maintain body condition during winter months. Individuals were removed from pasture and stall-kept only for veterinary treatment (including for development of lameness). At the time of sampling, mares were free from lameness, had body condition scores ranging from three to eight and were on average 16.7 years (range 11 to 23 years). This broad range in body condition was likely due to the lack of individualized nutritional management and free access to lush spring pasture. Incidence of laminitis at farm B was reported by management to be approximately 20%, providing evidence that the environment on this farm was more favourable to the development of laminitis than that on farm A. Samples from farm B were obtained over a 2-day period in late September.

Experiment 3  Eleven additional thoroughbred mares diagnosed by the attending veterinarian with chronic or acute laminitis were obtained from farm ‘B’ (Table S1-b). The majority of these mares developed laminitis as a result of obesity and/or a sudden change in pasture/diet. Disease severity among affected horses ranged from mild lameness treated with supportive shoes to Obel grade IV lameness and eventual euthanasia (individual scores provided in Table S1-b). These samples were collected in the summer months (from May to August). A set of 15 older, laminitis-free mares from Experiment 1 were used as controls, as they matched this group in breed as well as mean age.

RNA isolation

Blood was collected by jugular venipuncture into RNA-stabilizing PAXgene blood collection tubes. Total RNA was isolated and DNase-treated using the PAXgene Blood RNA kit (PreAnalytiX/Qiagen). RNA samples were quantified using a NanoDrop spectrophotometer (NanoDrop Technologies).

RT-qPCR

Reverse transcription was carried out using the TaqMan Reverse Transcription reagents and random hexamers [Applied Biosystems (ABI)] with incubation for 45 m at 45 °C in a thermocycler. Exon-spanning primers and MGB-binding probes were designed using the Primer Express software program (ABI) for the following genes: β-actin (ACTB, accession #AF035774), mast cell growth factor receptor (KIT, #AY874543) and mast cell tryptase beta 2 (TPSB2, aka equine MTC-1, accession #AJ515902) (Table 1). Exon boundaries were determined by BLAT search (UCSC genome browser, genome.ucsc.edu) and alignment of the mRNA sequence against the equine genome (EqCabv2.0). All primers produced a single band of the correct size when products were visualized by electrophoresis on an agarose gel. Although there are several known actin pseudogenes, DNase treatment eliminated detectable transcripts in non-RT controls. The primers and probes were used in a reaction including the equivalent of 40 ng cDNA from the RT step and TaqMan Gene Expression Master Mix (ABI) as per the manufacturers’ protocol. Thermocycling and detection were performed on the 7500 Fast Real-Time PCR System using the standard parameters (60°C annealing, ABI). Appropriate negative and non-RT controls were included for each run.

Table 1.   Primers and probes used in this study.
GeneForward Primer 5′–3′Reverse Primer 5′–3′Probe 5′–3′Amplicon Size BPs
ACTBCGA GGC CCC CTG AACGGT CTC AAA CAT GAT CTG GGT CAT CFAM-CAA GGC CAA CCG C-MGBNFQ61
KITGCG TCC TGC TTC TCC TGT TGGA CTC ACA GAT GGT TGA GAA GAVIC-CGC GTC CAG ACA GG-MGBNFQ59
TPSB2CGG CGG CGC ACT GTGCA GCT GCA CCC TGA TGT CTC TVIC-CGG ACA TTG AAG ATT T-MGBNFQ60

Data analysis

Assay efficiency was calculated for each sample using the LinRegPCR software (Ramakers et al. 2003). Individual wells were excluded and replaced if their calculated efficiency (E) exceeded two standard deviations from the mean or if the R2 for the linear fit estimating efficiency for that well was below 0.99. Relative target gene expression was adjusted for mean reaction efficiency and calculated relative to a reference gene using the following equation: relative expression (RE) = E(Reference gene)Ct (Reference)/E(Target gene)Ct (Target) (Schefe, 2006). There was no significant difference in raw ACTB Ct among groups, indicating that multiple references genes were not necessary. Fold-change is calculated as a ratio of relative gene expression in one group versus the other group in the same experiment.

Graphs, descriptive statistics and tests for significance were calculated using the JMP 7.0 software package (SAS Institute Inc., 2007). Coefficient of fit was calculated for the data from experiment 1, including linear and polynomial fits. A Shapiro–Wilk test for goodness-of-fit to the normal distribution revealed that the data from experiments 2 and 3 were not normally distributed. For these experiments, a Wilcoxon test was used to calculate P-values.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Experiment 1

No trend in expression of either gene (KIT and TPSB2) was observed owing to age among the group I horses (best fit R= 0.050 and 0.053 for a polynomial curve, Fig. 1a and b, respectively). Additionally, no effect was observed based on BCS category, although the limited amount of variation within this group is not a good measure of this type of association. Mean values were established for the group as a whole of 2.303–3 for KIT/ACTB and 3.70e-6 for TPSB2/ACTB.

image

Figure 1. KIT/ACTB (a) and TPSB2/ACTB (b) relative gene expression as a function of age (n = 30). The mean (line) is superimposed.

Download figure to PowerPoint

Experiment 2

Expression of KIT, a cell surface marker for MCs, was increased 2.573-fold (P = 0.0005) in those horses from farm ‘B’, the facility with the higher rate of laminitis, relative to farm ‘A’ (KIT/ACTB,Fig. 1a). An increase of similar magnitude was also observed in TPSB2/ACTB expression between farm ‘B’ and ‘A’ (2.633 fold, P = 0.0071, Fig. 1b). There was no significant difference in age between groups A and B, (mean 15.3 and 16.7 years, P = 0.76). Although the means were not significantly different for BCS (5.4 and 5.56, P = 0.49), the range of BCS values was slightly different between groups (P = 0.004 by a two-sided F-test) (Table S1). However, the sample size was insufficient to assess any direct correlation between this broad range in BCS and gene expression.

Experiment 3

KIT/ACTB expression did not change significantly between groups with laminitis and control horses (Fig. 3a). However, a mean 2.760-fold increase (P = 0.0011) in TPSB2/ACTB was found in the group with laminitis relative to the group with no previous diagnosis of the disease (Fig. 3b). The range of expression within the laminitic group was quite large (0.8442–12.27-fold increase over the control group) in contrast to the more narrow range among control horses. No direct association with Obel score or total days lame and TPSB2 expression was found, although this may in part be attributable to the small sample size (Table S1-b).

image

Figure 3. KIT/ACTB (a) and TPSB2/ACTB (b) relative gene expression in horses with (n = 11) and without laminitis (n = 15). (*P = 0.0011).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

The KIT and TPSB2 genes are expressed almost exclusively in whole blood by mast cells and their precursors. Consequently, these data demonstrate that mast cell gene expression can be measured in whole blood despite the transient nature of their presence in circulation. Although this could also be accomplished using flow cytometry and antibodies for these mast cell products, antibody development can be a lengthy process. Additionally, blood samples collected in PAXgene tubes for RNA analysis are far more stable and easy to handle than fresh blood for flow cytometry. Side-by-side comparisons of traditional flow cytometry and RT-qPCR using the same targets for the quantification of human peripheral blood cell populations have shown that RT-qPCR performs well at this task and is exceptionally sensitive (Pennington et al. 2001).

Age did not have an effect on gene expression for TPSB2 or KIT (Fig. 1). However, statistically significant differences were observed for management (Fig. 2) and the presence of lameness (Fig. 3). Horses in a more relaxed management setting, with unrestricted access to spring pasture, had 2.573-fold higher expression (P = 0.0005) of KIT. This difference in KIT/ACTB expression should, as supported by previous comparisons of flow cytometry and gene expression methods of quantifying cellular populations (Pennington et al. 2001), reflect an increase in the proportion of MC precursors in circulation. This increase could contribute to a predisposition to inflammatory conditions. A proportionate increase in TPSB2/ACTB expression is likely due to this change in the population of TPSB2-carrying cells rather than an increase in the transcription of TPSB2 per cell. This agrees with our hypothesis, given that both groups were free from disease or lameness and therefore MC activation, at the time of sampling. Further work is needed to determine specifically which conditions contribute to these changes in circulating, KIT-expressing cells and what their implications are for disease. As certain breeds are commonly believed to be more predisposed to laminitis, future work should also include the use of these assays in both predisposed and resistant breeds.

image

Figure 2. KIT/ACTB (a) and TPSB2/ACTB (b) relative gene expression in two groups of horses under different management conditions. Groups A (n = 10) and B (n = 9) had a farm-wide incidence of laminitis of 2% and 20%, respectively. (a. *P = 0.0005, b. *P = 0.0071).

Download figure to PowerPoint

In contrast, changes in KIT/ACTB expression between the laminitic and healthy groups were not noted. Theoretically, once MC activation has occurred, all suitably mature MCs have been recruited to their target tissues, leaving behind a baseline level of MC precursors. This may explain why a difference was not detected following an inflammatory event. However, significant changes (2.760-fold on average, P = 0.0011) were noted in TPSB2/ACTB expression in the laminitic group. This likely reflects ongoing activation and maturation of MCs (Welker et al. 2005). The broad range of up-regulation observed in the laminitic group correlates well with the broad diagnostic criteria used in this study. This group contained everything from reasonably sound, Obel grade 0–1, chronically affected individuals to extremely severe acute cases. No relationship could be found between TPSB2 expression and the severity or duration of lameness or acute vs. chronic status. Thus, additional work on larger numbers of horses, including well-documented chronic, acute and developmental cases, with more detailed clinical histories is needed to investigate gene expression changes relative to more detailed diagnoses and a variety of trigger factors. Although not feasible for this study, it would be advantageous in future work if examination and diagnosis could be performed by a single clinician. However, objective quantification of the damage within the hoof remains a difficult task and is not always well correlated with lameness.

Previously, plasma histamine levels in chronically laminitic horses were reported as nearly three times that of unaffected horses (Rautschka et al. 1991). More recently, chronically laminitic horses were shown to have significantly higher responses to intradermal allergen tests compared to unaffected horses (Wagner et al. 2003). MCs are vital to the immediate response to allergens, and their degranulation and release of histamine is responsible for wheal formation (Gruber & Kaplan 2005). Thus, the exaggerated response to allergens may reflect an increase in the number or activity of MCs.

Current research in laminitis has provided additional evidence that MCs could be a pivotal part of the disease process. Emigration of leucocytes is a significant early characteristic of the development of laminitis (Black et al. 2006; Hurley et al. 2006). MCs release several chemotactic factors that direct the infiltration of leucocytes into the site of inflammation (Gruber & Kaplan 2005). Notably, pharmacologic stabilization of MCs reduces the expression of adhesion molecules on circulating leucocytes that is necessary for their movement through vessel walls (Zhao et al. 2005). Additionally, neutrophil infiltration into wounds is significantly reduced in mast cell-deficient mice, although wound healing was otherwise normal (Egozi et al. 2003).

Recent work in humans has provided new insight into the importance of the mast cell in vascular and inflammatory disease, particularly when secondary to obesity or metabolic syndrome (Chaldakov et al. 2001; Deliargyris et al. 2005; Kounis 2006). Although the vascular component of the pathological course of laminitis is well documented and preliminary work has suggested systemic mast cell involvement (Rautschka et al. 1991; Wagner et al. 2003), the specific role of the mast cell in laminitis has yet to be investigated. Our work examines the relationship between laminitis in the horse and circulating mast cells. If a method can be devised to reduce the number or reactivity of mast cells in tissues, then it may be possible to prevent laminitis in susceptible individuals and reduce the recurring cycle of tissue damage and inflammation, potentially allowing the effective pharmacological treatment of early and recurrent cases of the disease. Current treatments focus on reducing pain and damage and can do little to prevent reoccurrence. Although removing known trigger factors can help, there is a desperate need for a way to gain specific control of the inflammatory processes within the hoof. Several drugs are known to be effective in blocking mast cells in humans, and adapting these compounds for equine use may prove beneficial. However, we cannot devise an effective treatment protocol in the absence of a method to assess mast cell number and function. This work has shown that RT-qPCR with mast cell-specific genes could fill that role.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

The authors thank the many horse owners and clinicians who supplied samples and clinical information for this study. S. Brooks was supported by the Paul Mellon Post-Doctorial Fellowship.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. References
  10. Supporting Information

Table S1 Relative expression data for individual horses.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
AGE_2093_sm_TableS1.pdf48KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.