Evaluation of endothelial glycocalyx injury biomarkers in feline hemotropic mycoplasmosis

The present study aimed to investigate endothelial glycocalyx (eGCx) damage in cats with feline hemotropic mycoplasmosis caused by Mycoplasma haemofelis using selected biomarkers and to determine the diagnostic and prognostic significance of these biomarkers. The study included 25 cats with feline hemotropic mycoplasmosis and 10 healthy cats. Clinical examination, blood gas analysis, complete blood count, and biochemical analysis were performed. Hemotropic mycoplasmosis diagnosed by microscopic examination and molecularly confirmed by PCR targeting the Mycoplasma haemofelis 16s rRNA gene. To evaluate endothelial glycocalyx damage, syndecan-1, endothelin-1 (ET-1), asymmetric dimethylarginine (ADMA), and vascular endothelial growth factor-A (VEGF-A) concentrations were measured using cat-specific commercial ELISA kits. Of the cats with feline hemotropic mycoplasmosis, 14 (56%) survived and 11 (44%) died. While syndecan-1 and ET-1 concentrations were significantly higher in cats with hemotropic mycoplasmosis compared to the control group (p < 0.001), no statistically significant difference was found for ADMA and VEGF-A concentrations (p > 0.05). Endothelial glycocalyx biomarkers showed significant correlations with each other and with hematological parameters (p < 0.01). The results of the ROC analysis showed that ET-1 with area under the curve (AUC) of 0.821 (p < 0.01) and VEGF-A with AUC of 0.805 (p < 0.010) were found to be significant prognostic indicators. In conclusion, this study demonstrated that serum syndecan-1 and ET-1 can be used as diagnostic and serum ET-1 and VEGF-A as prognostic biomarkers in cats with hemotropic mycoplasmosis. Our results indicate the development of eGCx damage in feline hemotropic mycoplasmosis and suggest that glycocalyx disruption may contribute to the pathogenesis of the disease.

The significance of vector-borne diseases has recently become more recognized due to ecological and climatic changes.Among these diseases, infections caused by hemotropic mycoplasma species play an essential role.Hemotropic mycoplasmas (hemoplasmas) are uncultivable bacteria that epierythrocytically infect red blood cells and can cause anemia by inducing hemolysis.Mycoplasma haemofelis is one of the most pathogenic feline hemoplasmas and can cause significant hemolytic disease even in immunocompetent cats.Although Mycoplasma haemofelis is less common in cats, it is reported to be responsible for acute clinical infections compared to other feline hemotropic Mycoplasma species 1,2 .
Endothelial glycocalyx (eGCx) is a 1-3 µm thick dynamic layer that covers the apical surface of all endothelial cells and maintains vascular hemostasis, tonus, and permeability 3,4 .This layer regulates leukocyte adhesion/ migration by maintaining endothelial oncotic pressure and plays a vital role in inhibiting intravascular thrombosis 4,5 .It also controls vascular resistance through proper endothelial nitric oxide (NO) production and vasodilation 6 .Endothelial glycocalyx, which has attracted more attention in recent years, consists of glycoproteins containing sialic acid, proteoglycans (syndecan, glypican) which are membrane-bound core proteins,

Cat with hemotropic mycoplasmosis
Twenty-five owned cats of different breeds, ages, and sexes diagnosed with hemotropic mycoplasmosis and exhibiting clinical signs such as anemia, icterus, and depression were enrolled.All cats were owned and except 3 of them had access to the outdoor.The materials to be used in the microscopical, serological, and molecular analysis were collected from cats admitted to the Animal Hospital of the Faculty of Veterinary Medicine, Selcuk University for diagnosis and treatment.The study has not included cats with any disease other than hemotropic mycoplasmosis.

Healthy cats
Ten healthy cats of different breeds, age and sex that were admitted to the Animal Hospital of the Faculty of Veterinary Medicine, Selcuk University for vaccination, preventive antiparasitic drug administration, and general control were enrolled.All the cats were owned and lived indoors.Healthy cats, which constituted the control group, were defined as the absence of hemotropic mycoplasmosis in microscopy and PCR analysis, rectal temperature within normal limits, and no use of antimicrobial or other drugs within 2 weeks before sampling.Healthy cats also had normal CBC, blood gas analysis and biochemistry.

Collection of blood samples
Blood samples were collected from both healthy and hemotropic mycoplasmosis cats at the time of admission.Blood samples for blood gas analysis, complete blood count (CBC), biochemical and endothelial damage marker analyses were collected from the vena cephalica antebrachium.Plastic syringes containing sodium heparin were used for blood gas measurements.Tubes containing K 3 EDTA were used for CBC analysis.Blood gas and CBC measurements were performed within 5-10 min after sample collection.Non-anticoagulant tubes were used for serum collection.Blood samples collected for biochemical, and biomarker analyzes were kept at room temperature for 15 min and then centrifuged at 20 × g for 10 min.The sera were divided into two parts and one part was used for biochemical analysis and the other part was stored at -80 °C for biomarker analysis.

Microscopic diagnosis of hemotropic mycoplasmosis
In this study, 0.5-1.5 mL blood samples were collected in anticoagulated tubes containing ethylenediamine tetraacetic acid (K 3 -EDTA) from 33 cats with suspected hemotropic mycoplasmosis based on clinical signs such as severe anemia, icterus and depression and hemogram findings.Thin blood smears were immediately prepared from all anticoagulated venous blood samples, dried in the open air, fixed with absolute methanol for 5 min, and stained with 10% Giemsa solution for 30 min.After staining, the thin blood smears were washed under tap water and dried.Immersion oil was dripped on the slides, and the smears were examined under 100× magnification of a light microscope (Leica DM1000) for epieritrocytically localized hemoplasmas.At least 50 microscopic fields were scanned to detect these haemopathogens.

Molecular confirmation of hemotropic mycoplasmosis
The presence of the primary causative agent of feline hemotropic mycoplasmosis, Mycoplasma haemofelis, was also molecularly investigated in the blood samples of all sera to be used in enzyme-linked immunosorbent assays.Accordingly, genomic DNA was first extracted from blood samples using the High Pure PCR Template Preparation Kit (Roche, Germany) following the manufacturer's instructions.The extracted genomic DNA was stored at − 20 °C until molecular analysis.Each DNA sample was then screened by species-specific PCR targeted to amplify the M. haemofelis 16S rRNA gene fragment (170 bp).Molecular screening was performed according to the procedure previously described by Jensen et al. 20 .The final volume of the reaction mixture was 10 µL and consisted of 2 µL 5× One Taq Standard Reaction buffer (BioLabs, New England), 0.2 µL dNTP mix (Deoxynucleotide solution mix, BioLabs, New England), 0.2 µM each primer (Hf-F: 5′-ACG AAA GTC TGA TGG AGC AAT A-3′ and Hf-R: 5′-ACG CCC AAT AAA TCC GRA TAA T-3′), 0.05 µL Taq DNA polymerase (BioLabs, New England), 1 µL DNA template and 6.35 µL double distilled water (Invitrogen, UltrapureTM Distilled Water, DNAse/RNAse-Free).Mycoplasma haemofelis positive genomic DNA sample previously confirmed by sequence analysis (OR979169, https:// www.ncbi.nlm.nih.gov/ nucco re/ OR979 169) was used as positive control and double distilled water (Invitrogen, UltrapureTM Distilled Water, DNAse/RNAse-Free) was used as a negative control.PCR amplicons were electrophoresed on 1.5% agarose gels, stained with ethidium bromide, and visualized under a UV transilluminator (UVP, Upland, CA, USA).Considering the co-infection status in cats, all blood samples included in the study were molecularly confirmed to be negative for some other haemopathogens such as Babesia sp., Rickettsia sp., Bartonella sp. and Anaplasma sp.For this purpose, PCR analyses utilizing universal primers for the amplification of Babesia/Theileria/Hepatozoon sp.18S rRNA, Anaplasma/Ehrlichia sp.16S rRNA, Rickettsia sp.glt A and Bartonella sp.16-23S gene fragments were performed following the conditions described in the relevant literature [20][21][22][23] .The primer pair (HF-F/R) used in this study for the molecular identification of Mycoplasma haemofelis is also used for the identification of Candidatus Mycoplasma haemominutum (CMh), another less pathogenic feline haemotropic Mycoplasma species.This primer can be used to amplify 170 bp fragments of the 16S rRNA gene for Mycoplasma haemofelis and 193 bp fragments of the same gene for CMh 20 .In addition, the molecular negativity of another feline haemoplasma species, Candidatus Mycoplasma turicensis was checked by PCR targeting the amplification of the 488 bp region of the 16S rRNA gene of this species 24 .

Statistical analysis
The data of this study were analyzed using SPSS 25 (IBM Corp. Released 2017.IBM SPSS Statistics for Windows, Version 25.0 Armonk, NY: IBM Corp.) statistical program.One Sample Kolmogorov-Smirnov test was applied to evaluate the prerequisites for normal distribution (parametric or non-parametric) of the data.Since the study data showed non-parametric distribution, they were presented as median and interquartile range (IQR, 25th-75th percentile).For comparison between groups, the Mann-Whitney U test was used.The Spearman rank correlation test was used to determine the correlation between variables.In Spearman rank correlation analysis, the scale ranges from -1 to 1, with values close to 1 indicating a strong correlation and values close to 0 indicating a weaker correlation.In addition, Receiver Operating Characteristic (ROC) analysis was performed to determine the sensitivity, specificity, and cut-off values of selected variables in predicting mortality in surviving and nonsurviving cats.Statistical significance was considered as p < 0.05.www.nature.com/scientificreports/

Microscopic examination findings
Microscopic examination of thin blood smears commonly revealed M. haemofelis-infected erythrocytes as well as some abnormalities including ghost RBCs, erythrocyte agglutination, macrocytosis, anisocytosis, reticulocytosis and Howell-Jolly bodies, with rare presence of spherocytes.

Blood gas and CBC analysis
Venous blood gas and complete blood count parameters of healthy and infected cats are presented in Table 1.While BE and HCO 3 levels of cats infected with M. haemofelis were significantly lower than healthy cats (p < 0.05).
In CBC analysis, WBC, Lym, Mon, Gra, and RDW levels of cats with hemotropic mycoplasmosis were significantly higher, while RBC, Hct, Hb, and PLT levels were lower (p < 0.05).

Biochemical analysis
Biochemical parameters of healthy and cats with hemotropic mycoplasmosis are presented in Table 2. Lactate, TBil, DBil, AST, LDH, and P levels of cats with hemotropic mycoplasmosis were significantly higher than healthy cats, but albumin and A:G ratio were lower (p < 0.05).No statistically difference was detected in other biochemical parameters (p > 0.05).

Biomarker analysis
Biomarker concentrations of healthy and cats with hemotropic mycoplasmosis are presented in Table 3. Syndecan-1 and ET-1 concentrations of cats with hemotropic mycoplasmosis were significantly higher than the healthy ones (p < 0.001).There was no significant difference in ADMA and VEGF-A concentrations between healthy and cats with hemotropic mycoplasmosis (p > 0.05) (Table 3).www.nature.com/scientificreports/

Correlation analysis
Correlations between endothelial glycocalyx biomarker concentrations and some hematologic parameters determined in healthy and infected cats by Spearman correlation analysis are presented in Table 4. ET-1 concentration in blood serum showed a strong and positive correlation (p < 0.01) with syndecan-1.ADMA showed a weak positive correlation with syndecan-1 concentrations (p < 0.05), while ET-1 showed a moderate positive correlation (p < 0.01).There was a strong positive correlation (p < 0.01) between VEGF-A and ADMA concentrations.WBC levels showed a strong positive correlation (p < 0.01) with syndecan-1 and a moderate positive correlation (p < 0.01) with ET-1.RBC levels were strongly and negatively correlated with syndecan-1 (p < 0.01) and ET-1 (p < 0.01) concentrations.However, there was a moderate and negative correlation with WBC (p < 0.01).There was a moderate and negative correlation between PLT levels and syndecan-1 (p < 0.01) and ET-1 (p < 0.01) concentrations, while there was a strong positive correlation with RBC (p < 0.01).Among the biochemical parameters, lactate had a moderate positive correlation with ET-1 (p < 0.01) and WBC (p < 0.01), while A:G ratio had a moderate and negative correlation with VEGF-A (p < 0.01).

Mortality analysis
The results of the ROC analysis to determine the relationship between some parameters and mortality in nonsurviving cats with hemotropic mycoplasmosis are presented in Fig. 1.The results of the ROC analysis showed that A:G ratio at the cut-off point of 0.46 g/dL, area under the curve (AUC) of 0.877 (95% confidence interval (Cl): 0.723-1.000,p = 0.001) with 92% sensitivity and 88% specificity (Fig. 1A); ET-1 at the cut-off point of 89.08 ng/L, area under the curve (AUC) 0.821 (95% Cl: 0.649-0.994,p = 0.007) with 90% sensitivity and 72% specificity (Fig. 1B); VEGF-A at the cut-off of 57.44 pg/mL, area under the curve (AUC) 0.805 (95% Cl: 0.630-0.980,p = 0.010) with 90% sensitivity and 72% specificity (Fig. 1C) were found to be significant prognostic indicators for mortality prediction in cats with hemotropic mycoplasmosis.

Discussion
This study evaluated the concentrations of serum syndecan-1, ET-1, ADMA, and VEGF-A, markers of glycocalyx degradation, in cats with feline hemotropic mycoplasmosis.Our results showed that the serum concentrations of syndecan-1 and ET-1 were high in cats with feline hemotropic mycoplasmosis and there were significant correlations with other markers.In addition, the biomarkers ET-1 and VEGF-A were useful in predicting mortality in cats with feline hemotropic mycoplasmosis.Taken together, our results suggest that glycocalyx degradation may contribute significantly to the pathogenesis of feline hemotropic mycoplasmosis.
There is a paucity of information regarding the evaluation of blood gas analysis in cats with feline hemotropic mycoplasmosis.İder et al. 25 reported that metabolic acidosis and hyperlactatemia developed in cats with hemotropic mycoplasmosis.Studies in anemic patients infected with Plasmodium species have reported that systemic hypoxia due to anemia may lead to an increase in anaerobic glycolysis, higher lactate production, and lactic acidosis 26 .It has also been reported that the decreased hepatic clearance of lactic acid in malaria may contribute to the development of lactic acidosis 27 .It has also been reported that Plasmodium-infected erythrocytes produce up to 100 times more lactate than uninfected erythrocytes 28 .In the present study, hyperlactatemia and mild metabolic acidosis in cats with feline hemotropic mycoplasmosis may be attributed to a decrease in tissue perfusion due to anemia 25,26 and an increase in lactate production by infected erythrocytes 26,28 .
Changes in the leukocyte panel in cats with feline hemotropic mycoplasmosis have been reported to be highly variable and of limited diagnostic value 2,29 .On the other hand, it has been reported that infected cats may develop a total leukocyte increase 2,25 , monocytosis, and in some cases lymphocytosis 30 .In the present study, the increase in total leukocyte count, granulocytosis, monocytosis, and lymphocytosis in cats with hemoplasmosis was interpreted as an indicator of inflammation due to infection in consistency with previous studies 2, 25 .In addition, macrocytic and hypo/normochromic regenerative anemia due to hemolysis are among the most common findings in cats with hemotropic mycoplasmosis 1,31 .In this study, low RBC, Htc, and Hb levels in cats with hemotropic mycoplasmosis may indicate the development of anemia due to hemolysis.However, although MCV levels did not differ between the study groups, MCV and RDW levels above the reference range in cats with feline hemotropic mycoplasmosis may indicate that the anemia has a regenerative character 1, 31 .
Platelet (PLT) levels have been reported to be lower in cats with hemotropic mycoplasmosis than in healthy cats 30 .Although there are no data explaining the development of thrombocytopenia in infected animals, it is accepted that low PLT counts in diseases with hemolysis, such as malaria, may be due to immune complex formation and phagocytosis of platelets by splenic macrophages 32,33 .In the present study, thrombocytopenia in cats with hemotropic mycoplasmosis may be related to phagocytosis of platelets by splenic macrophages 32,33 , similar to the pathogenesis of malaria.
Hyperbilirubinemia is associated with severe hemolytic anemia and liver damage in cats with feline hemotropic mycoplasmosis 1, 34 .It has been reported that during severe erythrocyte destruction, the capacity of the liver to metabolize bilirubin is exceeded and hyperbilirubinemia develops 35 .In the present study, the increased total and direct bilirubin concentrations in cats with feline hemotropic mycoplasmosis were interpreted as excessive bilirubin production due to intravascular and extravascular hemolysis and exceeding hepatic bilirubin clearance Table 4. Correlations between biomarker concentrations and some hematologic parameters in healthy and infected cats.Syndecan-1 (Syn-1), Endothelin-1 (ET-1), asymmetric dimethylarginine (ADMA), vascular endothelial growth factor-A (VEGF-A), Total leukocytes (WBC), erythrocytes (RBC), platelets (PLT), Lactate (Lac), albumin globulin ratio (A:G).**Correlation is significant at the 0.01 level, *Correlation is significant at the 0.05 level.capacity 35 .In addition, the high AST and LDH enzyme activities determined in cats with hemotropic mycoplasmosis may be related to hemolysis, and the high phosphorus levels may be related to the release of excessive amounts of phosphorus from erythrocytes as a result of intravascular hemolysis 25 .Albumin, total protein, and A:G ratio show changes associated with inflammatory response in various diseases.A low A:G ratio with hyperglobulinemia and hypoalbuminemia has been reported in cats with feline hemotropic mycoplasmosis compared to healthy cats 25,36 .A low A:G ratio indicates active inflammation in cats and has been described as a significant diagnostic parameter 37 .In addition, low A:G ratio has been reported to be an independent predictor of prognosis in some chronic diseases, cancers, and critically ill patients [38][39][40] .In the present study, albumin and A:G ratio were significantly lower in cats with feline hemotropic mycoplasmosis compared to healthy cats and were found to be good prognostic indicators for predicting mortality with 92% sensitivity and 88% specificity at a cut-off point of 0.46 ng/mL.Lower serum albumin concentration and A:G ratio in cats with feline hemotropic mycoplasmosis compared to healthy cats can be interpreted as an inflammatory response due to infection and acute phase response 25,37 .It can be concluded that the A:G ratio can be used to predict mortality in cats with feline hemotropic mycoplasmosis.
During the onset of pathological conditions, the eCGx layer is damaged and endothelial glycocalyx components such as heparan sulfate, syndecan, and hyaluronic acid are incorporated into the bloodstream.Recently, there has been increasing interest in studies to monitor and diagnose diseases by measuring the concentrations of biomarkers of glycocalyx damage 7,9,18 .Syndecan-1, which is bound to the endothelial cell surface, has been reported to be targeted by microbial pathogens, particularly in the early stages of infection, and to be involved in the initiation of acute inflammatory responses 41 .The release of syndecan-1 into the circulation has also been reported to reflect more severe eCGx damage and greater functional impairment 8,9 .Syndecan-1 concentrations were found to be elevated in severe malaria cases due to eCGx damage, and this was negatively correlated with RBC and PLT 9 .Researchers have noted that the glycocalyx stabilizer sphingosine-1-phosphate (S-1-P) is decreased in malaria cases due to anemia and/or thrombocytopenia, and this may contribute to glycocalyx deterioration 9 .In this study, significant correlations between high syndecan-1 concentrations and hematological parameters (WBC, RBC, PLT) in cats with hemotropic mycoplasmosis suggest the development of an acute inflammatory response 41 and eCGx damage 8,9 in these cats.
Heparanase is another critical factor contributing to endothelial glycocalyx degradation 10 .Endothelin-1, released during glycocalyx degradation, is the primary heparanase activator and keeps it in a vicious cycle to continue glycocalyx degradation 6,7,42 .In addition, ET-1 is thought to mediate endothelial dysfunction both by interfering with the expression and activity of endothelial NO synthase 43 and by decreasing NO bioavailability through the formation of reactive oxygen species 44 .In this study, the significantly increased serum ET-1 concentrations in cats with feline hemotropic mycoplasmosis compared to the control group and the positive correlations between syndecan-1 and ADMA may reflect the development of glycocalyx damage by heparanase activation and altered NO bioavailability in cats with feline hemotropic mycoplasmosis 6,42,43 .
In human medicine, a significant correlation between ET-1 concentrations and morbidity and mortality has been reported in sepsis 45,46 , and chronic renal failure 47 .On the other hand, in veterinary medicine, ET-1 concentrations were found to be higher in non-surviving premature calves with respiratory distress syndrome than in survivors and were also significant for mortality with 87% sensitivity and 82% specificity 48 .Similarly, ROC analyses showed that ET-1 was a promising prognostic indicator for predicting mortality in cats with feline hemotropic mycoplasmosis with 90% sensitivity and 72% specificity in the present study.The results indicate that ET-1 concentrations can be used as diagnostic and prognostic biomarkers in cats with feline hemotropic mycoplasmosis.
The present study showed a negative correlation between serum ET-1 concentrations and RBC and PLT, and a positive correlation with WBC and lactate.Although the reasons for the relationships between ET-1 concentrations and hematological parameters are not fully elucidated, ET-1 has been shown to play a role in the pathogenesis of hypoxia 48 and many inflammatory diseases, leading to increased concentrations 45,46,49 .Furthermore, strong correlations between ET-1 and anemia and thrombocytopenia have been demonstrated in patients with chronic renal failure 47 .These findings may reflect the development of an acute inflammatory response in cats with hemotropic mycoplasmosis 45,49 and a potential link between ET-1, anemia and thrombocytopenia 47 .
Another heparanase activator is asymmetric dimethylarginine (ADMA), an endogenous endothelial NO synthase (eNOS) inhibitor 42 .This biomarker reduces NO production by inhibiting eNOS function 50 .ADMA concentrations were found to decrease and then gradually increase in cases of acute severe malaria and sepsis [50][51][52] .In the present study, there was no significant difference between ADMA concentrations of healthy and infected cats.The cats with hemoplasmosis included in our study were in the acute clinical stage of the disease and therefore we expected ADMA concentrations to be low 50,52 .The release of free ADMA into the circulation due to erythrocyte lysis during disease 53 may have contributed to the no statistical difference.
In the present study, ADMA concentrations were positively correlated with serum syndecan-1, ET-1 and VEGF-A concentrations.Two of these biomarkers (ET-1 and VEGF-A) are associated with reduced endothelial NO bioavailability 6,42,54,55 and syndecan-1 is directly related to glycocalyx degradation 8,9 .Taken together, these findings suggest that in cats with feline hemotropic mycoplasmosis, microvascular reactivity is impaired and endothelial dysfunction develops due to decreased endothelial NO bioavailability.
VEGF-A concentrations have been reported to be elevated in malaria and inflammatory diseases as a result of endothelial dysfunction 13,14 .In a recent study, there was no significant difference in VEGF-A concentrations between children with malaria and healthy ones 12 .Low concentrations have been associated with severe organ damage and mortality in sepsis, but it has not been a good indicator of prognosis 56 .Also, no association was found between low VEGF-A concentrations and disease severity or mortality in severe malaria 54,55 .In these studies [54][55][56] low VEGF-A concentrations were associated with decreased NO bioavailability.In the present study, VEGF-A was not a diagnostic marker because VEGF-A concentrations were similar in healthy and infected cats.This may be related to the fact that the levels of PLT, which is the main source of VEGF-A, are low in cats with hemotropic mycoplasmosis 57 .However, VEGF-A concentrations were found to be an important indicator of mortality with a sensitivity of 90% and a specificity of 72%.Surprisingly, serum VEGF-A concentrations were not low in non-surviving infected cats, and those above the cut-off of 57.44 pg/mL died.Therefore, our results do not support previous studies [54][55][56] .In non-surviving cats, VEGF-A concentrations may be elevated due to protection or repair of endothelial damage 11 .In our study, the deterioration of circulating VEGF-A concentrations and its positive correlation with serum ADMA concentrations may reflect the impaired endothelial NO production and endothelial dysfunction described in severe malaria 55 and sepsis 56 .In light of these findings, VEGF-A may play a role in the pathogenesis of hematropic mycoplasmosis.

Figure 1 .
Figure 1.Receiver operating characteristic (ROC) curve analysis to discriminate between surviving and non-surviving cats with hemotropic mycoplasmosis based on serum (A) A:G ratio, (B) ET-1 and (C) VEGF-A concentrations.