The relationship between cardiac inflammation and function in mice after myocardial infarction assessed with high frequency speckle tracking echocardiography and MRI

Table of Contents
Chapter 1 Introduction Ошибка! Закладка не определена.
1.1 Ischemic heart diseases and cardiac remodelling Ошибка! Закладка не определена.
1.1.1 Myocardial Infarction and Ischemia Reperfusion Ошибка! Закладка не определена.
1.1.2 Definition and types of cardiac remodelling Ошибка! Закладка не определена.
1.1.3 Characterization of cardiac remodelling Ошибка! Закладка не определена.
1.2 Cardiac dysfunction after myocardial infarction Ошибка! Закладка не определена.
1.2.1 Deterioration of global function of the heart Ошибка! Закладка не определена.
1.2.2 Systolic and diastolic dysfunction Ошибка! Закладка не определена.
1.2.3 Regional cardiac dysfunction after MI Ошибка! Закладка не определена.
1.3 Myocardial inflammation and cardiac repair Ошибка! Закладка не определена.
Regional inflammatory reaction Ошибка! Закладка не определена.
Systemic inflammatory reaction Ошибка! Закладка не определена.
1.3.1 Initiation of the post-infarction inflammatory response: the role of innate immunity Ошибка! Закладка не определена.
The chemokine family in myocardial infarction Ошибка! Закладка не определена.
1.3.2 Chemokines can be united into two types: homeostatic and inducible chemokines. Ошибка! Закладка не определена.
CXC chemokines Ошибка! Закладка не определена.
CC chemokines Ошибка! Закладка не определена.
XC chemokines Ошибка! Закладка не определена.
CX3C chemokines Ошибка! Закладка не определена.
1.3.3 Pro-inflammatory cytokines in myocardial infarction Ошибка! Закладка не определена.
Tumor Necrosis Factor (TNF-α) Ошибка! Закладка не определена.
Interleukin 6 Ошибка! Закладка не определена.
Interleukine 1 (Il-1) Ошибка! Закладка не определена.
Adhesion molecules Ошибка! Закладка не определена.
1.3.4 The cellular basis of the post-infarction inflammatory response Ошибка! Закладка не определена.
Mononuclear cells Ошибка! Закладка не определена.
Mast cells Ошибка! Закладка не определена.
Myofibroblasts Ошибка! Закладка не определена.
1.4 Myocardial inflammation and function Ошибка! Закладка не определена.
1.4.1 Cardiac fibrosis and cardiac dysfunction Ошибка! Закладка не определена.
1.5 Assessment of myocardial inflammation Ошибка! Закладка не определена.
1.6 Hypothesis and aims of dissertation Ошибка! Закладка не определена.
Chapter 2 Ошибка! Закладка не определена.
Materials and methods Ошибка! Закладка не определена.
2.1 Animal husbandry and maintenance Ошибка! Закладка не определена.
2.2 Experimental animal models: myocardial infarction and ischemia-reperfusion Ошибка! Закладка не определена.
Experimental technique Ошибка! Закладка не определена.
Pre-surgical preparation Ошибка! Закладка не определена.
Intubation and ventilation Ошибка! Закладка не определена.
General Surgery Ошибка! Закладка не определена.
Post- operative care Ошибка! Закладка не определена.
Sham models Ошибка! Закладка не определена.
Tissue preparation Ошибка! Закладка не определена.
2.3 Investigation of inflammation in the heart using invasive and non-invasive techniques Ошибка! Закладка не определена.
2.3.1 Immunohistochemistry Ошибка! Закладка не определена.
Staining for macrophages using Mac 3 marker Ошибка! Закладка не определена.
Staining for T cells using CD3 marker Ошибка! Закладка не определена.
Staining for leukocytes using CD 45 marker Ошибка! Закладка не определена.
Picrosirius red staining for fibrosis Ошибка! Закладка не определена.
2.3.3 In vivo detection of myocardial inflammation with MRI Ошибка! Закладка не определена.
Image Analysis Ошибка! Закладка не определена.
2.4 Assessment of cardiac morphology and function after myocardial infarction Ошибка! Закладка не определена.
2.4.1 Assessment of cardiac hypertrophy at whole heart and cellular levels Ошибка! Закладка не определена.
Wheat Germ Agglutinin staining Ошибка! Закладка не определена.
2.4.2 Evaluation of infarct size with MRI Ошибка! Закладка не определена.
2.4.3 Evaluation of cardiac function with MRI Ошибка! Закладка не определена.
2.4.4 Evaluation of cardiac function with conventional echocardiography Ошибка! Закладка не определена.
4.4.5 Evaluation of regional cardiac function with high frequency speckle tracking strain analysis (STE) Ошибка! Закладка не определена.
2.5 Statistical analysis Ошибка! Закладка не определена.
References Ошибка! Закладка не определена.
Chapter 3 Ошибка! Закладка не определена.
Results Ошибка! Закладка не определена.
3.1 No cardiac hypertrophy after 7 days of myocardial infarction Ошибка! Закладка не определена.
3.2 Cardiac function was worse after Permanent Ligation vs Ischemia reperfusion Ошибка! Закладка не определена.
3.3 Increased cardiac inflammation and fibrosis after myocardial infarction Ошибка! Закладка не определена.
3.4 Relationship between global cardiac inflammation and function assessed by MRI Ошибка! Закладка не определена.
3.6 Relationship between cardiac inflammation and function at regional level Ошибка! Закладка не определена.
Relationship between local inflammation and regional dysfunction after MI Ошибка! Закладка не определена.
General Discussion Ошибка! Закладка не определена.
Limitations Ошибка! Закладка не определена.
Clinical applications Ошибка! Закладка не определена.
Conclusion Ошибка! Закладка не определена.

In order to use the STE (VisualSonics, USA) software v1.1.1, we used semi-automatic technique by placing tracking points on the endocardial and epicardial borders. The frame-by-frame tracking technique through the entire cardiac cycle calculated the global cardiac function such as EF. In addition, the software used to analyse the LV mass automatically after placing the endo and epicardial borders, although in some cases manual modification was essential to optimize the border tracking.
LVEDD and LVESD were obtained from M-mode using parasternal Long-Axis view. Several measurements at different levels were taken and the average was calculated.



4.4.5 Evaluation of regional cardiac function with high frequency speckle tracking strain analysis (STE)
The non-invasive, early, and stable quantification of regional myocardial function was used to analyze each segment of left ventricle after myocardial infarction. This technique is novel and based on two-dimensional LV long-axis images. Strain can be detected from Dopler-based or 2D echo.
Strain is a dimensionless quantity and is produced by application of stress. It represents the fractional or percentage change from the original or unstressed dimension and includes both lengthening, or expansion (positive strains) and shortening, or compression (negative strains). Ideally, one would like to measure Lagrangian strain, defined as (L2L0)/L0, where L0 is the length corresponding to zero stress and L is the instantaneous length.10 Because zero stress lengths are technically difficult to measure, L0 is often replaced by the initial muscle length or the end-diastolic length.

STE is a technique that uses patterns resulting from acoustic shadows within the myocardium to track myocardial motion within one cardiac cycle. Moreover, STE is not dependant on insonation angle, which became its benefit, because in Doppler-based echo it is dependant. In addition, strain was presented in 2 different modes such as radial and longitudinal.
To analyze strain we used Vevo 2100 software, which utilizes semi-automated border tracking technique. Firstly, we placed tracking points on the endocardial borders. The points are used as a guide for border definition and frame-by-frame tracking in one cardiac cycle. Manual adjustments can be made as required in order to optimize border tracking. Thus, when the border set up, we proceeded with next step, which automatically divided left ventricle into six segments (two apical, two middle, two basal). The strain in each segment was represented by a curve, and thus we had six curves. Using this analysis, we obtained data for regional radial and longitudinal strain.

2.5 Statistical analysis
All data is expressed as mean values ± standard error of mean (SEM). The appropriate statistical analyses were carried out depending upon the following criteria:
One-way ANOVA was used when more than two groups were being compared. Repeated measures ANOVA was used for repeated measures in multiple groups.
Liner regression was used to compare the relationship between groups and parameters.
Post-hoc analysis was carried after one-way ANOVAs when significance was reached, using Bonferroni’s multiple comparison tests on selected groups, where there were multiple comparisons with one group. All statistics were analyzed using GraphPad Prism (v4.03 for Windows, San Diego, CA). P<0.05 was considered statistically significant.

References
Protti, A., A. Sirker, et al. (2010). "Late Gadolinium Enhancement of Acute Myocardial Infarction in Mice at 7T: Cine-FLASH Versus Inversion Recovery." Journal of Magnetic Resonance Imaging 32(4): 878-886.
Slavin, G. S., S. D. Wolff, et al. (2001). "First-pass myocardial perfusion MR imaging with interleaved notched saturation: Feasibility study." Radiology 219(1): 258-263.
Chapter 3
Results
3.1 No cardiac hypertrophy after 7 days of myocardial infarction
Post-MI cardiac remodelling is characterized by myocyte death and enhanced inflammation in the infarct and by peri-infarct areas in the early phase followed by hypertrophy of non-infarcted myocardium, fibrosis, ventricular dilatation, and contractile dysfunction at the late stage. In this study, heart structure and function, particularly the relationship between the severity and location of inflammation and cardiac dysfunction were investigated after 7 days of either permanent LAD ligation (PL) or ischemia/reperfusion (IR).
One week of PL or IR did not induce cardiac hypertrophy in terms of heart weight (Fig 3.1A) or heart weight/body weight ratio (Fig 3.1B). The mouse body weights were also unchanged before and after MI in all groups (Fig 3.2.).
To evaluate whether cardiac hypertrophy occurs at the cellular level, myocyte cross-section size in the remote area was analyzed by WGA staining of left ventricular sections. Again, there was no significant difference in myocyte size post-MI in either model (Fig 3.3).
A

B
Fig 3.1 No cardiac hypertrophy after 7 days MI in terms of heart weight (A) or heart weight body weight ratio (B) (Sham PL n=6, PL n=16, Sham IR n=6, IR n=16)
A
B
Fig 3.2: Similar body weights before and after PL (A) or IR (B): (Sham PL n=6, PL n=16, Sham IR n=6, IR n=16)
A
B
Fig 3.3 (A) Representative images show no myocyte hypertrophy after 7 days MI by WGA staining. (Scale bars 20 µm). (B) Mean data are shown below. Sham PL n=6, PL n=16, Sham IR n=6, IR n=16.
3.2 Cardiac function was worse after Permanent Ligation vs Ischemia reperfusion
Global cardiac function was assessed by standard echocardiography (Vevo 2100 system), including left ventricular end diastolic volume (LVEDV), left ventricular end systolic volume (LVESV), stroke volume (SV) and ejection fraction (EF). Cardiac Magnetic Resonance Imaging (CMRI) was also performed in some studies to evaluate the infarct size and cardiac function.
Compared with sham controls, after 1 week of PL, hearts displayed significant LV dilatation and dysfunction as revealed by significant increases in LVEDV and LVESV (Fig 3.4), and marked decreases in EF and SV by echocardiography (Fig 3.5). Similar changes were observed in IR hearts, but with less severity than PL. No changes of heart rate were observed in any group (Fig 3.6).
A
Sham
PL
IR
B
Fig 3.4. (A) Representative M mode echo images of sham, PL, and IR hearts in the parasternal long axis view at the level of the papillary muscles. (B) Mean data of LVESV and LVEDV by standard echo analysis using the 2D area on a parasternal long axis view. (*Sham PL vs. PL P<0.001; ** Sham IR vs. IR P<0.001; # PL vs. IR P<0.001.Sham PL n=6, Sham IR n=6, PL n=16, IR n=11)

Fig 3.5. Ejection Fraction and Stroke Volume measured by ECHO. (*Sham PL vs Pl P<0.001;**Sham IR vs IR P<0.01; #PL vs IR P<0.001: Sham PL n=6, Sham IR n=6, PL n=16, IR n=11)
Fig 3.6. No changes in heart rate: (Sham PL n=6, Sham IR n=6, PL n=16, IR n=11)
CMRI with late gadolinium (Gd) enhancement (LGE) is currently used as the gold standard for the in vivo diagnosis of MI and estimation of infarct size in patients. We recently validated pre-clinical cine fast low shot angle (cine-FLASH) LGE MRI for the quantification of MI size in mice at 7T in vivo (Protti A and Shah, 2010). Seven mice were imaged in this study after 7 days PL. Average infarct area values derived from MRI. Fig 3.7 shows a representative example of images). Cardiac functional (EF and SV) and volumetric (LVESV and LVEDV) parameters were calculated from cine-FLASH acquisitions after Gd injection. In agreement with echo analysis, hearts displayed worse LV dilatation and dysfunction 1 week post PL compared with IR (Fig 3.8 and 3.9). The correlation between infarct size and EF as a marker of cardiac dysfunction was excellent (Fig 3.10.). Due to the weak enhancement of Gd after 2 or 6 days of IR, the infarct size was assessed by MRI only in the PL model in this study. Fig 3.7. Typical MR images at end-diastole in 1mm thick slices 7 days post-MI. Infarct size analysis was performed using an in-house developed computer software program, which was able to delineate the infarct region due to the different signal intensity compared to the remote viable myocardium and blood. (Red arrows indicate the infarcted myocardium, which is grey, black arrows show non-infarcted area, which is dark black).
Fig 3.8. Ejection Fraction and Stroke Volume measured by MRI. (* Sham vs. PL P<0.001; ** Sham vs. IR P<0.001; # PL vs. IR P<0.001: Sham n=6, PL n=7, IR n=12)
Fig 3.9. LVESV and LVEDV by MRI. (* Sham vs PL P<0.001; ** Sham vs IR P<0.001; # PL vs IR P<0.001. Sham n=6, PL n=7, IR n=12)
Fig 3.10. Correlation between infarct size and EF in individual hearts: r²=0.7891; P =0.0002.
3.3 Increased cardiac inflammation and fibrosis after myocardial infarction
Inflammation is an early response to stress in the heart after ischemia. Inflammatory cells are recruited from blood vessels and infiltrate into the myocardium. Many inflammatory cells could be seen in the infarcted myocardium as well as remote areas on histological sections. Several inflammatory cell markers including CD45, MAC3, and CD3 were studied by immunohistochemistry to assess local inflammation in the heart 7 days after myocardial infarction. Cardiac fibrosis was also evaluated to assess the consequence of myocardial inflammatory response.
Firstly, anti-CD45 antibodies were used to check the presence of leukocytes as an evidence of general inflammatory process. The appearance of leukocytes is rare in sham-operated heart, however, the numbers of CD45 positive cells significantly increased in IR hearts, and was even higher in PL hearts (Fig 3.11). Moreover, it appeared that most of leukocytes were infiltrated within the infarcted myocardium (more towards to the edge of the representative image), and some in the remote area (more towards to the centre of the representative image) (Fig 3.11).
Secondly, the presence of macrophages was assessed by staining for the macrophage-specific marker Mac3 in heart sections. In line with the leukocyte infiltration data, we found that there was a significant increase in recruitment of macrophages in the permanent ligation model compared to ischemia reperfusion (Fig. 3.12).
Thirdly, LV sections were stained for the T cell-specific marker CD3. Compared to sham controls, MI led to a significant increase in the numbers of CD3 positive cells. Interestingly, IR hearts had more T cells than PL (Fig 3.13).
A
B
Fig 3.11. (A) Representative histological staining for the pan-leukocyte marker CD45 (red arrows indicate CD45 positive cells with brown colour. Scale bar 25 µm). (B) Mean data expressed as number of CD45 positive cells per LV section (* Sham PL vs. PL p<0.001, n=6/16; ** Sham IR vs. IR p<0.001, n=6/12; # PL vs. IR p<0.001)
A
B
Fig 3.12. (A) Representative histological staining for the macrophage-specific marker Mac3 (red arrows indicate Mac3 positive cells with brown colour. Scale bar 25 µm with 40x magnification). (B) Mean data expressed as the number of Mac3 positive cells per LV section (* Sham PL vs. PL p<0.001, n=6/9; ** Sham IR vs. IR p<0.001, n=6/9; # PL vs. IR p<0.001)
A
B
Fig 3.13. (A) Representative histological staining for the T cell-specific marker CD3 (red arrows indicate CD3 positive cells with brown colour. Scale bar 25 µm with 40x magnification). (B) Mean data expressed as number of CD3 positive cells per LV section (* Sham PL vs PL p<0.001, n=6/16; ** Sham IR vs IR p<0.001, n=6/16; # PL vs IR p<0.001).
Lastly, interstitial fibrosis was evaluated by Picrosirius red staining after 7 days of permanent ligation or ischemia reperfusion. In sham-operated animals, cardiac fibrosis was less than 1%. However, the hearts exhibited significantly increased interstitial fibrosis after 1 week MI. Nevertheless, there was no difference of cardiac fibrosis between PL and IR (Fig 3.14.).
A
B
Fig 3.14. Fibrosis after MI. (A) Representative examples of sections stained with Picrosirius Red. Scale bar 100 µm. (B) Quantification of interstitial fibrosis, expressed as percentage of total myocardial cross-sectional area (* Sham PL vs. PL p<0.05 n=6/16; **Sham IR vs. IR p<0.001, n=6/14).
3.4 Relationship between global cardiac inflammation and function assessed by MRI
To investigate the relationship between cardiac inflammation and function, we analyzed the correlation between the total number of inflammatory cells (CD 45, MAC 3 and CD3) in the heart and cardiac function as assessed by MRI-derived EF in the both the PL and IR models. The numbers of macrophages in the hearts were negatively correlated with cardiac function in the PL group (Fig 3.15.). However, no correlation was found between the numbers of CD45-positive cells (pan-leukocyte marker) or T cells and EF. No relationship was found between any inflammatory cell type and cardiac function in the IR group (Fig 3.16).
A

B

C
Fig 3.15. Correlation between different leukocyte subsets and ejection fraction (EF) measured by MRI after PL. (A) Numbers of macrophages, (B) total leukocytes, (C) T cells.(EF vs. macrophages, P =0.0078, r²=0.4078, n=16; EF vs. T cells, P=0.1093, n=8; EF vs. leukocytes, P =0.8192, n=16)
A
B

C

Fig 3.16. Correlation between inflammatory cell subtypes and EF in the IR group: (A) macrophages, (B), total leukocytes, C T cells: (EF vs. macrophages, P=0.7697, n=12; EF vs. leukocytes, P=0.7731, n=11; EF vs. T cells, P=0.8776, n=12).
3.5 Detection of cardiac inflammation after MI in vivo by non-invasive MRI-based assessment
Recent advances suggest that MRI may be used to assess regional inflammation in the heart by injection of VSOP (very small iron oxide nano-particles) and quantification of uptake of these particles by macrophages (Sosnovik and Nahrendorf et al., 2007). In current study, this latest technique was applied to detect inflammation in the remodelling mouse heart after MI by non-invasive MRI-based methods. As shown in Fig 3. 17, VSPOP that are taken up by macrophages after 7 days of IR could be successfully detectable by MRI, although it was more difficult in the PL model. Importantly, the reliability and accuracy of MRI-based inflammation assessment was further confirmed by post-mortem histology with Mac 3 staining (Fig 3.18.), and there was a good correlation between the non-invasive technique and histology.
Fig 3.17. Representative MR images of cardiac inflammation. The iron nano-particles that detect inflammation in vivo were injected iv 7 days after IR and an MRI scan was performed 2 days later. MRI images were taken in sequence 1mm to 4 mm from apex to base. The black areas indicate areas of increased iron signal, i.e. cardiac inflammation (with arrows).
B
Fig 3.18. (A) Cardiac inflammation detected by MRI (with arrows) was confirmed by histology (MAC 3 staining) on the same heart. (The heart was cut in sequence 1mm from apex to base, in the same pattern as it was imaged by MRI). (B) Good correlation between the numbers of macrophages and quantity of the black signal was measured by MRI. The black signal was quantified as a percentage of the whole LV. The number of macrophages was counted as the mean number of positive cells out of 4 sections. Liner regression was used to introduce the relationship between in vivo and in vitro study. (P=0.0030; r²=0.9115, n=6 hearts).
3.6 Relationship between cardiac inflammation and function at regional level
To assess further the relationship between the severity and location of inflammation and regional deformation and cardiac dysfunction, high frequency echo with segmental strain analysis (apical, mid and basal) and histology were performed in both ischemic models.
Assessment of regional cardiac function and deformation by a high frequency micro-ultrasound system with speckle tracking
Peak radial and longitudinal strains were similar in both sets of sham-operated mice. However, hearts exhibited a considerable reduction in peak radial strain after 1 week MI, in the apex, mid and basal regions (Fig 3.19). Interestingly, basal peak radial strain was significantly lower in the PL compared to IR group, suggesting different regional dysfunction in these models (Fig 3.19).
Similarly, compared to sham controls, MI hearts displayed contractile dysfunction as revealed by a significant decrease in peak longitudinal strain at all three segments. However, there was no statistical difference in longitudinal strain between the two ischemic models (Fig 3.20).
We found similar results when assessing radial and longitudinal strain rates (maximum rate of development of strain). The radial stratin rate was reduced in all segments of PL or IR hearts as compared to sham controls, and the basal radial strain rate was lower in the PL compared to IR model (Fig ***). Results for the longitudinal strain rate are shown in (Fig. ***).
A

B
C
Fig 3.19 Hearts exhibited different regional dysfunction after PL and IR by radial strain analysis. Basal strains (C) decreased significantly, after 1 w PL than IR compared to the apical (A) and middle segments (B). (* Sham PL vs. PL, P<0.001, n=6/11; ** Sham IR vs. IR, P<0.001, n=6/7; # PL vs. IR, P<0.001).
A

B
C

Fig 3.20 MI induced regional cardiac dysfunction by longitudinal strain analysis. There were no differences in apical (A), (B) and (C). (* Sham PL vs. PL, P<0.001, n=6/11; ** Sham IR vs. IR, P<0.001, n=6/7).
Relationship between local inflammation and regional dysfunction after MI
Finally, we evaluated the relationship between local inflammation and regional dysfunction in both ischemic models by correlating the LV segmental strains with the local numbers of MAC3 positive cells in each segment of the left ventricle. In the PL model, there were good correlations between the total numbers of local macrophages and regional cardiac dysfunction in the apical, middle and basal segments respectively (all P<0.05) (Fig 3. 21). In line with this, we also found a good negative correlation of local inflammation with longitudinal segmental strain in all three segments (Fig 3.22)
In the ischemia/reperfusion model, the numbers of macrophages were negatively correlated with apical and basal radial strain, but not middle segment strain (Fig 3.23). However, there were good correlations between the local macrophage numbers and apical and middle longitudinal strain, but not basal longitudinal strain (Fig 3.24). These results indicate that not only the global cardiac dysfunction but also the subtle changes of regional function and deformation may be related to local inflammation in the heart.
A

B

C

Fig 3.21. Relationship between local macrophages and regional dysfunction by radial apical, mid and basal strain analysis in the PL model (A) P=0.0232, r²=0.7201, n=6, (B) P=0.0036, r²=0.9040, n=6) and (C) P=0.0155, r²=0.8036, n=6.
A

B
C
Fig 3.22. Relationship between local macrophages and regional dysfunction by longitudinal apical, mid and basal strain analysis in the PL model (A) P=0.0019, r²=0.9293, n=6, (B) P<0.0001, r²=0.9882, n=6 and (C) P=0.0327, r²=0.7201, n=6.
A

B
C
Fig 3.23. Relationship between local macrophages and regional dysfunction by radial apical (A) P=0.0312, r²=0.6383, n=7, (B) P= 0.8634, n=7) and (C) P=0.0186, r²=0.7023,n=7.
A

B
C
Fig 3.24. Relationship between local macrophages and regional dysfunction by longitudinal apical (A) P=0.0040, r²=0.8347, n=7, (B) P=0.0172, r²=0.7107, n=7 and (C) P =0.1678, r²=0.3420, n=7.
General Discussion
Myocardial infarction causes significant morbidity and mortality. In the early phase after acute MI, inflammation is a key mechanism that is necessary for repair of the infarct and clearance of damaged cells and debris (Aguor and Arslan et al.,2012). This repair process results in formation of a scar. However, inflammation may also have damaging consequences for the heart. It can promote excessive fibrosis and can contribute to contractile dysfunction (Aguor and Arslan et al.,2012). The amount of inflammation may be affected by whether or not the blocked coronary artery is re-opened - i.e. whether there is permanent ischemia or ischemia-reperfusion. Establishing the relationship between inflammation and contractile function after MI, and how this is affected by reperfusion, would be useful in better understanding the pathophysiology of cardiac recovery after acute MI. In this work, a mouse model of experimental MI was used to study the relationship between inflammation and contractile function and cardiac remodelling, as well as to compare permanent coronary occlusion versus ischemia-reperfusion. The relationship between cardiac inflammation and cardiac function was assessed at regional and global levels, using a combination of state-of-the-art in vivo non-invasive methods (high-resolution echocardiography and MRI) as well as in vitro histology. The hypothesis was that inflammation affects contractile function at both a regional and a global ventricular level.
In order to establish PL and IR models we used similar surgical techniques to ligate the LAD, however, the only difference between them was the duration of the ligation of the artery within the IR model. The survival rate was not significantly different between the two models. Following the surgical procedure, animals were checked for the basic physiological changes such as weight loss. Thereafter animals were assessed for quantification of changes in function, inflammation, and fibrosis.
Sham (control) animals were used to compare the models changes. To check the absence or presence of the heart hypertrophy we weighted the hearts after a week of surgical procedure and there was no change. The heart weight/body weight ratio was also checked with no changes(Zaino and Tabor 1963). A better method to analyze developed hypertrophy in the remote myocardium is an assessment of myocyte size. However, histological analysis established that there were no alterations in myocyte size, thus proving the absence of hypertrophy. This was because it was an early time point of one week.
Non-invasive techniques such as ECHO and MRI were applied to assess the global and regional cardiac function to know whether those models were different. Although echocardiography is commonly used to evaluate cardiac function after MI, CMR may provide an accurate functional assessment (Asanuma and Fukuta et al., 2012). The studies by Asanuma et al and Antoni et al were compared in order to see if there was a difference between them in deterioration of cardiac function after MI, such as EF, SV, LVEDV, and LVESV (Gardner, Bingham et al. 2009). In the current study, we found no significant change between global parameters assessed by MRI or ECHO. Nevertheless, we used MRI for analysis of global function when investigating the relationship between global cardiac function and global inflammation in the cardiac tissue.
It is well known that that cardiac function decreases after severe cardiac insult (Litwin, Katz et al. 1994).In order to confirm that, we used ECHO basic parameters such as EF, SV, LVEDV, and LVESV and found that cardiac function indeed deteriorated in ischemia permanent ligation and ischemia reperfusion models. All the parameters worsened in comparison to the control group. Usually this occurred due to the thinning of the infracted myocardium and tissue loss resulting from necrosis, therefore remodelling of the non-infarcted segments and dilatation of LV cavity(Pfeffer and Braunwald 1990). In addition, we observed that Ischemia Permanent Ligation model was worse in function with to a more dilated left ventricle in end diastole and end systole. We concluded that the function of the heart was worse in IPL when compared IR. We found that, STE analyses were not significantly different to MRI-derived measures, in terms of global cardiac function (Hoffmann, von Bardeleben et al. 2005).
The regional cardiac function was assessed with STE derived ECHO system using strain imaging. This method is more sensitive and subtle on the regional level and detects relatively minor changes in heart (Litwin and Katz et al. 1994). The apical, mid and basal segments were careful investigated by using STE software. It was predicted that in myocardial infarction model, there should be a decrease in strain, especially in the apical area after the LAD ligation (Jia, Olafsson et al. 2009) (Antoni, Mollema et al. 2010). Our studies showed that peak radial and longitudinal strains were similar in both sham-operated mice. However, hearts demonstrated an extensive reduction in peak radial strain and strain rate after a week of MI. Interestingly, basal peak radial strain in PL was significantly decreased compared to IR, suggesting different regional dysfunction in these models. The regional function is expected to be different because IR will have smaller infarcts.
Similarly, compared to sham controls, permanent ligation and ischemia reperfusion models hearts displayed obvious contractile dysfunction as revealed by significant decrease in peak longitudinal strain at all three segments. Nevertheless, there was no statistical difference in longitudinal strain between two ischemic models.
It was vital for our project to investigate each heart segment with the number of positive infiltrated inflammatory cells to the changes in strain. Ischemia permanent ligation model showed a good correlation between basal, mid and apical radial strains. Same as radial strain there was a good relationship between longitudinal basal, mid and apical segments and local macrophages infiltration. It was found that local infiltration of macrophages correlated well with strain in all three cardiac segments in the ischemia-reperfusion model.
Cardiac Fibrosis in sham-operated animals was less than 1%. However, the hearts exhibited significantly increased interstitial fibrosis after 1 week of MI. Nevertheless, there was no difference of cardiac fibrosis between PL and IR.
Nowadays a large number of studies are concentrating on non-invasive methods of investigation and possibly treatments. One of those is MRI with using different types of contrast agents. For our project iron nanoparticles contrast agent was applied to assess the presence of inflammation e.g. macrophages(Grundmann, Bode et al. 2011). It showed how inflammation activated and regulated after myocardial insult within 7 days. In addition, it correlated with the number of macrophages assessed by histology.
Our results confirmed that inflammation of the heart could be detected in the early stages post MI using non-invasive techniques. All ECHO and MRI studies were confirmed by histology. We also found a good correlation between global and regional cardiac function with inflammation in the heart. Hence, it could be said that if there was more inflammation in the heart after myocardial infarction, then the global and regional function was worse.
Moreover, after comparing the whole data between those parameters and inflammatory cell inflammation, it was found that the number of macrophages rather than leukocytes and T-cells were negatively correlated with EF in permanent ligation group. On the other hand, there was no relationship between global cardiac function (EF) and inflammatory cells in IR model.
Our studies continued to investigate the STE-based quantification of parameters of global LV structure and function, i.e. volumes, EF, SV, that are automatically calculated by the software.
Another aim of this project was to quantify the amount of fibrosis and inflammation in the two models 7 days after MI. The first week post MI is associated with inflammatory response to stress, wound healing and creating new blood vessels and scar formation(Frangogiannis, Smith et al. 2002). A large number of studies showed the presence of different inflammatory cells in different models(Timmers, Pasterkamp et al. 2012); however, none of them have compared the extent of inflammation in IPL and IR in the early phase after MI (Timmers, Pasterkamp et al. 2012). The number of leukocytes dramatically increased in PL compared to IR. Leukocytes clear the cardiac tissue from dead and necrotic cells (Arslan, de Kleijn et al. 2011). Other inflammatory cells assessed were macrophages, which remove necrotic cardiac myocytes and apoptotic neutrophils; secrete cytokines, chemokines and growth factors; and modulate phases of the angiogenic response(Lambert, Lopez et al. 2008). We found more macrophages in permanent ligation than in ischemia reperfusion. In terms of T-cells, we found controversial data. The Ischemia Reperfusion model had elevated numbers of T-cells compared to PL. Some researchers hypothesized that T (Treg) cells modulate inflammatory responses, attenuate ventricular remodelling, and subsequently improve cardiac function after MI (Tang, Yuan et al. 2012). They also found that Treg cells serve to protect against adverse ventricular remodelling and contribute to improve cardiac function after myocardial infarction via inhibition of inflammation and direct protection of cardiomyocytes. Therefore, we could consider that the cardiac function was improved after IR compare to PL due to T cells infiltration.
The amount of infiltrate and necrosis in the recipient rats appeared directly related to the size of the infarct from the donor rats. This suggests that larger infarcts lead to a greater inflammatory response. In the IR model by the meaning, there was less necrosis. None of the other organs (kidney, liver, lung, or brain) had evidence of infiltrates. Therefore, Alan Maisen proved direct evidence of autoimmune myocardial injury produced by adoptive transfer of concanavalin A–activated splenocytes after myocardial infarction. They proposed that neurohumoral activation early in the post-infarction period triggers a series of specific inflammatory and immunological events that lead to formation of specific clones of T cells. When these are activated and transferred into normal rats, cardiac-specific cellular infiltration occurs, occasionally accompanied by myocardial necrosis (Maisel and Cesario et al. 1998) .
There was no significant difference in developing cardiac fibrosis after neither PL nor IR. It was assessed by other studies in both models and all of them confirmed the presence of scar and fibrosis (Tsuda, Gao et al. 2003; Aguor, Arslan et al. 2012).
A relationship between global cardiac function and inflammation was only evident when studying macrophages and only in PL. The cardiac function was worse when the number of infiltrated macrophages was increased. We could say that there was a relationship between cardiac function and inflammation. However, the global data, such as EF and global inflammation is not sufficient to make this certain conclusion. Therefore, regional cardiac deformation and regional inflammation was checked and compared. Strain application in ECHO was used as a novel technique in our project.
We found a significant reduction in all segments (basal, mid and apical) in both models after MI and especially in the apical segments in PL, confirming the severity of this model and the presence of scar tissue (due to loss of contractile function on ECHO visually) in the apical area. It was confirmed by radial strain (Asanuma, Fukuta et al. 2012). In longitudinal strain, there was also a reduction in all 6 segments (Zaliaduonyte-Peksiene, Vaskelyte et al. 2012).
Since only macrophages showed a positive relationship between EF and global inflammation in the heart, we checked them with 6 different segment of the left ventricle. An interesting and significant fact was that in the PL model there was a good negative correlation between the apical, mid, basal radial strains, and the number of macrophages in the same segments respectively. Thus, increased numbers of inflammatory cells on the seventh day after MI was detrimental to the cardiac function. The heart function post Mi depends on the quantity of inflammation. Seven days are crucial for patients and depend for future prognosis. While the wound is repairing due to complex inflammatory process and the heart function compensates, most probably inflammatory stage is dangerous for the future prognosis and survival rate.
Lastly, we checked whether we could find a good correlation between regional inflammation and regional dysfunction. Some studies showed a good correlation between infract size and global strain and concluded that global strain is an excellent predictor of myocardial infarct size in chronic ischemic heart disease(Gjesdal, Hopp et al. 2007).
In the longitudinal strain, there was the same negative correlation between macrophages and all LV segments. Regional function got worth with the higher number of macrophages.
In the IR model in the radial strain there was only correlation in apical and basal sections, but no relationship in the middle segment of LV between cardiac deformation and inflammation. In addition, longitudinal strains had strong negative correlation with all 6 segments.
Using different types of contrast agents in this project helped us to visualize the inflammation in the heart and to evaluate the infarct size in vivo.
VSOP contrast agent was used to assess an inflammation in the two models; however, we could evaluate the uptake by macrophages only in the Ischemia Reperfusion model (Thompson, Emani et al. 2003; van den Bos, Wagner et al. 2003; van den Bos and Baks et al. 2006).
Some researches already carried out studies with PL and there was no evidence of iron nanoparticle accumulation in ischemia permanent occlusion; however, the contrast agent was detected by MRI in the hearts of the sham-operated mice (Sosnovik, Nahrendorf et al. 2007).
After checking the black signal of the heart tissue and presenting it as a percentage of the LV, we checked the relationships between the number of macrophages and VSOP uptake. We found a good correlation between them meaning that MRI and contrast agents can be used to assess inflammation in the heart tissue after MI without the need for histology (which was confirmed by histology itself).
Late Gadolinium contrast agent was used to assess the infarct size in both models(Ordovas and Higgins 2011) , but the uptake by this contrast agent was only visualized in ischemia reperfusion(Voelkl, Haubner et al. 2011). However, recently, Price carried out studies with Late Gadolinium in 2 different models and showed that there was an uptake in both models (Lightfoot, D'Agostino et al. 2010).
We could also confirm that when the infarct size increases the cardiac function is decreased as a compensatory process (Yang, Berr et al. 2004).
Limitations
Although the research has reached its aims, there were some unavoidable limitations.
Firstly, because of the time limit, this project could include more experiments and a larger number of samples in each group for better statistical analysis. It was also could be valuable to study the same models, but at the different time points (such as 3, 7 and 14 days) for the short and long time prognosis of adverse cardiac remodelling.
Secondly, it would be motivating to study human samples as the one of the subjects of the research project, which directly refers to clinical scenarios. However, it is not possible because of failure to access the human's heart tissue.
Thirdly, there were some limitations from technical point of view. To maintain a better ECHO image for the precise cardiac function was challenging due to the post surgical suture on the chest. Although it wasn’t vital for the global cardiac function, but for strain images it was essential to place the tracking points exact on the border of endo and pericardium due to very sensitive VEVO 2100 software
Clinical applications
The findings from this thesis have several practical implications for the understanding, assessment, and treatment of individuals with myocardial infarction.
Firstly, it has been proved that using non-invasive methods of investigations of cardiovascular diseases bring harmless diagnostic for the patients. MRI, together with contrast agents have became a very important application and uses to diagnose most of the cardiovascular diseases inclusive inflammatory (Hunold, Schlosser et al. 2005) also after either surgical or medical treatment, and to assess the infarct borders which are very essential to identify for the future prognosis and survival rate (O'Regan, Ahmed et al. 2009). However, it is an expensive procedure and cannot be used for each patient nowadays.
ECHO technique is widely used nowadays in all clinics. A novel parameter such as strain and strain rate gives a better understanding of LV function. It also provides strong prognostic information in patients after AMI. This technique is subtle and enables angle-independent quantification of myocardial deformation that may detect any little change on early stage.
Knowing the relationship between local inflammation and function in the heart could bring a new looks on patient’s management and future prognosis in adverse cardiac remodelling.
Conclusion
In conclusion, the studies performed in this thesis have highlighted that inflammation of the heart can be detected in the early stages post MI using non-invasive techniques. All ECHO and MRI studies were confirmed by histology. A good correlation had been found between global and regional cardiac function together with inflammation (macrophages) in the heart.

References
Aguor, E. N., F. Arslan, et al. (2012). "Quantitative T (2) (*) assessment of acute and chronic myocardial ischemia/reperfusion injury in mice." MAGMA.
Antoni, M. L., S. A. Mollema, et al. (2010). "Prognostic importance of strain and strain rate after acute myocardial infarction." Eur Heart J 31(13): 1640-1647.
Arslan, F., D. P. de Kleijn, et al. (2011). "Innate immune signaling in cardiac ischemia." Nat Rev Cardiol 8(5): 292-300.
Asanuma, T., Y. Fukuta, et al. (2012). "Assessment of Myocardial Ischemic Memory Using Speckle Tracking Echocardiography." Jacc-Cardiovascular Imaging 5(1): 1-11.
Frangogiannis, N. G., C. W. Smith, et al. (2002). "The inflammatory response in myocardial infarction." Cardiovasc Res 53(1): 31-47.
Gardner, B. I., S. E. Bingham, et al. (2009). "Cardiac magnetic resonance versus transthoracic echocardiography for the assessment of cardiac volumes and regional function after myocardial infarction: an intrasubject comparison using simultaneous intrasubject recordings." Cardiovascular Ultrasound 7.
Gjesdal, O., E. Hopp, et al. (2007). "Global longitudinal strain measured by two-dimensional speckle tracking echocardiography is closely related to myocardial infarct size in chronic ischaemic heart disease." Clinical Science 113(5-6): 287-296.
Grundmann, S., C. Bode, et al. (2011). "Inflammasome Activation in Reperfusion Injury Friendly Fire on Myocardial Infarction?" Circulation 123(6): 574-576.
Hoffmann, R., S. von Bardeleben, et al. (2005). "Assessment of systolic left ventricular function: a multi-centre comparison of cineventriculography, cardiac magnetic resonance imaging, unenhanced and contrast-enhanced echocardiography." Eur Heart J 26(6): 607-616.
Hunold, P., T. Schlosser, et al. (2005). "Myocardial late enhancement in contrast-enhanced cardiac MRI: Distinction between infarction scar and non-infarction-related disease." American Journal of Roentgenology 184(5): 1420-1426.
Jia, C. X., R. Olafsson, et al. (2009). "Two-Dimensional Strain Imaging of Controlled Rabbit Hearts." Ultrasound in Medicine and Biology 35(9): 1488-1501.
Lambert, J. M., E. F. Lopez, et al. (2008). "Macrophage roles following myocardial infarction." Int J Cardiol 130(2): 147-158.
Lightfoot, J. C., R. B. D'Agostino, et al. (2010). "Novel Approach to Early Detection of Doxorubicin Cardiotoxicity by Gadolinium-Enhanced Cardiovascular Magnetic Resonance Imaging in an Experimental Model." Circulation-Cardiovascular Imaging 3(5): 550-558.
Litwin, S. E., S. E. Katz, et al. (1994). "Serial Echocardiographic Assessment of Left-Ventricular Geometry and Function after Large Myocardial-Infarction in the Rat." Circulation 89(1): 345-354.
O'Regan, D. P., R. Ahmed, et al. (2009). "Cardiac MRI of myocardial salvage at the peri-infarct border zones after primary coronary intervention." American Journal of Physiology-Heart and Circulatory Physiology 297(1): H340-H346.
Ordovas, K. G. and C. B. Higgins (2011). "Delayed Contrast Enhancement on MR Images of Myocardium: Past, Present, Future." Radiology 261(2): 358-374.
Pfeffer, M. A. and E. Braunwald (1990). "Ventricular remodeling after myocardial infarction. Experimental observations and clinical implications." Circulation 81(4): 1161-1172.
Sosnovik, D. E., M. Nahrendorf, et al. (2007). "Fluorescence tomography and magnetic resonance imaging of myocardial macrophage infiltration in infarcted myocardium in vivo." Circulation 115(11): 1384-1391.
Tang, T. T., J. Yuan, et al. (2012). "Regulatory T cells ameliorate cardiac remodeling after myocardial infarction." Basic Res Cardiol 107(1): 232.
Thompson, R. B., S. M. Emani, et al. (2003). "Comparison of intracardiac cell transplantation: Autologous skeletal myoblasts versus bone marrow cells." Circulation 108(10): 264-271.
Timmers, L., G. Pasterkamp, et al. (2012). "The innate immune response in reperfused myocardium." Cardiovasc Res 94(2): 276-283.
Tsuda, T., E. Gao, et al. (2003). "Post-ischemic myocardial fibrosis occurs independent of hemodynamic changes." Cardiovasc Res 59(4): 926-933.
van den Bos, E. J., T. Baks, et al. (2006). "Magnetic resonance imaging of haemorrhage within reperfused myocardial infarcts: possible interference with iron oxide-labelled cell tracking?" Eur Heart J 27(13): 1620-1626.
van den Bos, E. J., A. Wagner, et al. (2003). "Improved efficacy of stem cell labeling for magnetic resonance imaging studies by the use of cationic liposomes." Cell Transplant 12(7): 743-756.
Voelkl, J. G. J., B. J. Haubner, et al. (2011). "Cardiac Imaging Using Clinical 1.5 T MRI Scanners in a Murine Ischemia/Reperfusion Model." Journal of Biomedicine and Biotechnology.
Yang, Z. Q., S. S. Berr, et al. (2004). "Simultaneous evaluation of infarct size and cardiac function in intact mice by contrast-enhanced cardiac magnetic resonance imaging reveals contractile dysfunction in noninfarcted regions early after myocardial infarction." Circulation 109(9): 1161-1167.
Zaino, E. C. and S. H. Tabor (1963). "Cardiac Hypertrophy in Acute Myocardial Infarction. A Study Based on 100 Autopsied Cases." Circulation 28: 1081-1083.
Zaliaduonyte-Peksiene, D., J. J. Vaskelyte, et al. (2012). "Does Longitudinal Strain Predict Left Ventricular Remodeling after Myocardial Infarction?" Echocardiography-a Journal of Cardiovascular Ultrasound and Allied Techniques 29(4): 419-427.
40
Day 0
Operation
Day 1
Day 2
Day 7
Ischemia Permanent Ligation model
Late Ga
Injection
MRI
ECHO
Day 7
Harvesting
Day 0
Operation
Day 1
Day 2