Mouse Tmem135 overexpression exhibits protein regulation causing chronic inflammation and fibrosis of the heart
By Erika Henningsen, Tetsuya Takimoto, Giangela Stokes, Sarah Lewis, and Akihiro Ikeda
The overexpression of mitochondrial transmembrane protein Tmem135 in mice causes mitochondrial dysfunction and fibrosis of the heart, but it is unknown which regulatory proteins influence this pathway to cause these effects. Here, we observed the modulation of numerous inflammatory pathways to link the effects of Tmem135 overexpression to inflammation and the immune system, thereby increasing understanding of how mitochondrial dysfunction leads to fibrosis. A culmination of current knowledge of these topics and Tmem135 will be discussed by following the molecular pathway of mitochondrial dysfunction through change in the expression of inflammatory enzymes such as Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K), protein kinase B (PKB or AKT), mitogen-activated protein kinase 8 (MAPK8 or JNK), mitogen-activated protein kinase 14 (MAPK14 or p38), nuclear factor kappa-light-chain-enhancer of activated B cells (Nf-κB), and beta actin (β-actin). The protein concentrations for these inflammatory enzymes have been analyzed through western blotting.
Mitochondrial dysfunction can be caused by the overexpression of transmembrane protein 135 (Tmem135) . Overexpression of Tmem135 in transgenic mice causes mitochondrial dysfunction through the increase in mitochondrial fragmentation. Fragmentation is an indicator that more fission than fusion has occurred in these mitochondrial cells. Fission is comparable to mitochondrial division and fusion is comparable to a mitochondrial union. In contrast to the fission shown with overexpression of Tmem135, it was previously observed that mice with mutated Tmem135 experienced excessive mitochondrial fusion, resulting in elongated mitochondria and mitochondrial dysfunction. Therefore, both excessive mitochondrial fission and fusion lead to mitochondrial dysfunction. This is because both fragmentation and elongation of mitochondria lead to impaired respiration and increase the production of reactive oxygen species (ROS) . ROS are created through incompletely reduced oxygen molecules in a redox reaction and are damaging to cells as they activate redox-sensitive transcription factors, ultimately causing tissue inflammation . This inflammation can become malignant to cells as chronic inflammation is a constant stressor for cells. Excessive fission, excessive fusion, and the subsequent production of ROS contribute to the mitochondrial dysfunction resulting in chronic inflammation.
When damage or inflammation occurs, signals throughout an organism report the disturbance. These signals are proteins and chemical mediators released by injured tissues. They are often cytokines, histamines, bradykinins, and prostaglandins that elicit an immune response by attracting macrophages and phagocytes to clean up and destroy any damaged cells via apoptosis, the process of cell destruction, which is often self-regulated within cells. This process should result in a reduction of cell inflammation in the affected areas as cell tissues are replaced (Figure 1) . This typically allows the regenerative process to end without any long-lasting damage. Some components of inflammation are designed to destroy microbes, but this can harm other healthy tissues surrounding the inflamed region. For this reason, inflammation must be tightly regulated by the immune system. Chronic inflammation can be poisonous to surrounding healthy cell tissue and can be initiated by persistent and progressive damage, prolonged exposure to toxic agents such as ROS, or inappropriate inflammatory responses in autoimmunity. Injury that is initially acute inflammation can become chronic inflammation and both can lead to fibrosis (Figure 1).
During regeneration of damaged cells, reconstruction attempts to be identical to previous cell and tissue architecture. However, if the restructuring of cells is complex or there is an excessive amount of tissue damage, the formation of scar tissue and continuation of inflammation can occur. This process is known as fibrosis. Scar tissue is the buildup of fibroblasts that become one large fibrosis scar. The scarring continues until the inflammation around these cells can be shut down . Initially, the structure recreated by fibroblasts heals and sustains areas containing damaged tissue, but over time the accumulation of fibroblasts and their byproducts can do more damage than good. This can be seen in cases of acute and chronic fibrosis. When tissue regeneration occurs in the heart, damaged muscle cells are reconstructed with fibroblasts and will have a strong healthy fibrous base, but after some time, numerous fibroblasts lead to contraction difficulties . This stems from the combination of extracellular matrix (ECM) and collagen buildup that stiffen the fibrous muscle tissue in the heart. Mitochondrial dysfunction leads to inflammation and fibrosis according to Lewis, which explains why the amount of collagen found in transgenic mice that overexpress Tmem135 is extreme . The ECM synthesized by fibroblasts holds collagen, a known stiffening agent. When the ECM is improperly reorganized, it can create an even larger fibrotic scar. The increase in collagen causes fibrotic scars to be stiffer than normal tissue and significantly affects myocardial muscle tissue. As noted, scar tissue in the heart is a major problem because it is very dense and prevents efficient heart muscle contraction . A condition consistently found in transgenic mice with Tmem135 overexpression is hypertrophy. Hypertrophy is defined as the enlargement of the heart—specifically in the left ventricle—due to the buildup of fibroblasts that significantly reduces contractility and left ventricular blood volume ejection. This occurs due to the excessive creation of fibroblasts, which cause an enlargement of the left ventricle similar to swelling. The volume of the ventricle within the heart shrinks as the buildup of fibroblasts increases, resulting in reduced contractility and an increased risk of heart failure. This can also cause the organ as a whole to grow .
Because cardiovascular disease and heart failure are the leading causes of death in the United States, it is important to look at the pathways involved in diseased phenotypes of cardiomyopathy. We documented the protein expression of inflammatory pathways affected by Tmem135 in order to explore the connection between mitochondrial dysfunction and inflammation that causes fibrosis of the heart in transgenic mice at a molecular level. Thus, we investigated Tmem135’s connection to cardiovascular complications via expression analysis and tested the causal relationship of Tmem135 overexpression to mitochondrial dysfunction, which leads to chronic inflammation and hypertrophy from fibrosis.
Gene Identification and Analysis Through DAVID
Previously researched and identified functions were expressed through Gene Ontology (GO) terms. A predetermined gene list composed from preliminary data regarding genes related to Tmem135 was analyzed through the Database for Annotation, Visualization, and Integrated Discovery (DAVID) to yield a functional analysis of these genes and to produce their GO terms. Important genes and pathways related to chronic inflammation and fibrosis were identified using the GO terms. Connections were made with existing sample data which showed that gene expression increase or decrease was correlated to the overexpression of Tmem135. The gene expression was represented by two-tailed, unpaired t-tests. These were performed by comparing transgenic (n=4) and wild-type (n=4) in log2 expression values. Those with a p-value of <0.05 were considered differentially expressed.
To find different genes relating to inflammation and fibrosis, the subject areas of interleukins (IL), integrin alphas (Itga) and betas (Itgb), transforming growth factors (Tfg), fibroblast growth factors (Fgf), and toll-like receptors were studied. There are numerous different pathways that genes use to produce inflammation in cells. These pathways often have similar proteins expression that can be experimentally tested to learn how they affect inflammation pathways when over or under-concentrated in comparison to control groups. To observe how Tmem135 overexpression affects expression in these proteins, the findings were then compared to the gene trends observed in the sample data.
Search data from pathways altering Tmem135 expression with GO terms was related to inflammation centered around the integrin (Itg) family. Its pathways to inflammation were analyzed through the Kyoto Encyclopedia of Genes and Genomes (KEGG Pathways). The examined Itg pathways were centered between extracellular matrix accumulation and inflammation in transgenic mice. The pathways were analyzed for both their connections to inflammation as well as their connections to other inflammation pathways. To confirm the change of expression was due to Tmem135, these inflammatory pathways were compared and mutual proteins were chosen for experimentation.
Tissues from 4 wild type mice and 4 transgenic mice overexpressing Tmem135 were pooled and homogenized in RIPA buffer with a protease inhibitor cocktail. Equal amounts of protein were subjected to SDS-PAGE using a 12% Bis-Tris gel and primary antibodies against Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K; Cell Signaling Technology), protein kinase B (AKT; Cell Signaling Technology), mitogen-activated protein kinase 14 (p38; Cell Signaling Technology), phosphorylated mitogen-activated protein kinase 14 (phosphorylated p38; Cell Signaling Technology), mitogen-activated protein kinase 8 (JNK; Cell Signaling Technology), nuclear factor kappa-light-chain-enhancer of activated B cells (Nf-κB; Santa Cruz Biotechnology), and β-actin (β-actin; Abcam). Rabbit (Cell Signaling Technology) or mouse (Cell Signaling Technology) secondary antibodies were used and protein concentrations were detected with a chemiluminescent agent (Amersham ECL Plus Western blotting detection system, General Electric, Buckinghamshire, UK) and exposed to X-ray film (Thermo Scientific, Rockford, IL). Novex Sharp Pre-Stained Protein Standard was chosen as a loading control because it expresses a wide range of molecular weight proteins. Western blot analyses were performed in duplicate experiments for each genotype (total eight mice assayed/genotype) and these X-ray images of protein concentrations were quantified with ImageJ and Fiji.
Gene Identification and Analysis Through DAVID
Inflammatory response was a major GO term found through DAVID analysis (Figure 2). Other important GO terms were cell proliferation and extracellular matrix organization due to their known influence on fibrosis. Analyzed research related to inflammation showed genes within the Itg family, among others, to be statistically significant in expression between wild type and transgenic mice.
Analysis of the DAVID-produced genes yielded many interconnected pathways. When examining the pathway for Itg affecting inflammation, some of the pathways were up-regulated and inhibiting different proteins. Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) was a recurring protein in the examined pathways, making it an excellent experimental protein for western blot analysis. This protein is a component of actin cytoskeleton pathways (Figure 3), focal adhesion pathways (Figure 4), and the PI3K signaling pathways (Figure 5). In the Itg pathway, it has been observed that the enzyme PI3K relies on the expression of focal adhesion kinases (FAK). This then allows for the expression of PI3K activating phosphatidylinositol (3,4,5)-trisphosphate (PIP3), which in turn activates the expression of protein kinase B (AKT). AKT, another protein of interest that is also known as PKB, is connected to both anti-apoptotic and apoptotic pathways. AKT can either activate nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB), causing anti-apoptotic results or inhibits B-cell (Bcl) proteins which causes apoptotic results. NF-kB is activated when AKT phosphorylates IκB kinase (IKK). IKK then indirectly promotes NF-kB. This promotion allows the transcription of pro-survival genes, which are genes that survive apoptosis.
Further analysis of the KEGG pathways led to the examination of mitogen-activated protein kinase 8 (JNK) and mitogen-activated protein kinase 14 (p38). These proteins are related to AKT and are found in the MAPK signaling pathway (Figure 6). Additionally, JNK is also found in the PI3K signaling pathways (Figure 5). JNK and p38 are separately activated through phosphorylation conducted by the actions of the mitogen-activated protein kinase kinase (MKK) family which are in turn activated by AKT. JNK and p38, as a result, both have the ability to phosphorylate various proteins that cause proliferation, differentiation, inflammation, and apoptosis. This ability to up-regulate apoptosis leads us to believe that these protein concentrations will be lower in Tmem135 mice as previous research shows extreme levels of fibrosis may potentially be caused by the inability to complete apoptosis .
PI3K Western Blotting
The experimental western blot analysis analyzed the Itg pathway’s effect on Tmem135 and focused on the expression of PI3K, a protein found within this pathway. The antibody used for western blot analysis was against PI3K- p110α (C73F8). For all western blot experiments, we compared expression between control wild type mice and transgenic Tmem135 mice. PI3K was increased in transgenic mice overexpressing Tmem135 by approximately 28.8% (Figure 7). These results produced a p-value of 0.0003 and are therefore significant.
AKT Western Blotting
The antibody used for western blot analysis was against Akt (Ser473) (D9E). We observed the expression of AKT in both wild type and transgenic Tmem135 mice. Western blot data for AKT produced inconclusive results. Expression of AKT was not seen on the blots as neither the wild type nor transgenic mice produced any of AKT proteins within the heart.
p38 Western Blotting
The antibody used for this blot analysis was against p38α MAP Kinase. We observed the expression of p38 in both wild type and transgenic Tmem135 mice. Western blot data for p38 showed a greater protein concentration for wild type mice and a decreased expression level in transgenic mice overexpressing Tmem135. The data showed an average 43.96% higher concentration of p38 in the wild type mice (Figure 8). This data yielded a significant p-value of 0.0128.
Phosphorylated p38 Western Blotting
Similarly, phosphorylated p38 (Phospho-p38) was examined with western blot analysis. The primary antibody used was against Phospho-p38 MAPK (Thr180/Tyr182) (D3F9) XP. Expression of Phospho-p38 was observed in wild type and transgenic mice (Figure 11). The phosphorylated p38 protein was expressed in greater concentration in wild type mice and not as greatly expressed for transgenic mice overexpressing Tmem135 (Figure 9). The mean concentration difference was 9.87% and the data was shown to be significant with a p-value of 0.0151.
JNK Western Blotting
The antibody used for western blot analysis was a phospho-SAPK/JNK antibody (CST:9251). We observed the expression of JNK in both wild type and transgenic Tmem135 mice. The western blot data showed wild type mice to have a greater mean concentration of JNK by 9.4%. This data was shown to be significant with a p-value of 0.0002.
Nf-κB Western Blotting
The antibody used for western blot analysis was against Nf-κB p65 (A) (sc-109). We observed the expression of Nf-κB in both wild type and transgenic Tmem135 mice. Western blot data for Nf-κB showed a greater protein concentration for transgenic mice who overexpressed Tmem135 compared to wild type mice. The data showed that the concentration of Nf-κB was an average of 25.88% higher in the transgenic mice (Figure 11). This data is significant with a p-value at 0.03.
Beta Actin Western Blotting
A beta-actin antibody (ab8226) was used. We observed the expression of β-actin in both wild type and transgenic Tmem135 mice. The wild type western blot data had a greater mean concentration of β-actin than transgenic mice by 2.85% (Figure 12). This data was shown to be not significant with a p-value of 0.7255. Therefore, there is no significant difference in the concentration of β-actin in wild type mice and transgenic mice.
Our analysis finds MCSA as a more effective treatment for EIB than either ICS or LTRA. The MCSA treatment proved to be a significantly more effective pharmacological solution for EIB symptoms. This finding may be due to the fact that mast cells are the direct cause of inflammation in the lungs. Targeting mast cells may be important through the blocking of calcium ion flow, preventing the release of the inflammatory agent histamine .
When analyzing treatments of ICS for EIB, it was found that the maximal fall in %FEV was lower in the treatment group than in the placebo group. When comparing the effectiveness of ICS therapy to the placebo, there was a statistically significant difference (p-value = 0.0001) (Figure 1). These results were as we had expected, as ICS are used widely to repress the symptoms of inflammation for patients with asthma, as well as being prescribed for individuals with EIB.
In the analysis of LTRA treatments for EIB, it was found that the mean %FEV fall was lower in the treatment group than the placebo group (Figure 2). This was expected as LTRAs, like Montelukast, are used actively in the pharmaceutical industry to treat asthma and EIB, supporting that LTRAs are a viable form of EIB reducer. Differences in the studies’ methodologies may have led to variability, such as the baseline FEV values, as differences among participants may show some importance in the responses to each treatment.
An interesting difference between the three classes of drugs was the difference in duration of the treatment. Studies on ICS tended to focus primarily on long term analysis of its effect on pulmonary inflammation. Since it is a treatment associated more with asthma than EIB, it was not studied with the sudden change in lung function associated with EIB. The MCSA and LTRA studies focused more on shorter term effects, as both are used as fast acting responders after exercise-induced airway obstruction has occurred. Even though these are all viable ways to treat EIB, as seen in the individual analyses for each treatment type, they are used in different ways.
Our findings suggest that MCSA treatment may be a better alternative than ICS and LTRA for the control of EIB symptoms. We believe that this is primarily because MCSA blocks histamine release and consequently shuts down inflammation at its root . Future studies may be able to help affirm our findings through a wider range of testing. Long term MCSA and LTRA treatment need to be further studied, as these studies are less prevalent. This testing could be useful in determining which treatment provides the most beneficial balance between cost, FEV outcome, and convenience for the patient. Additionally, a future study could attempt to negate the differences among individuals by giving each participant each treatment with proper washout periods in between trials. Age has not been studied sufficiently regarding FEV response, so it may be beneficial to study age groups in order to investigate the role of age between treatments and results. Similarly, studies on the differences between well-trained athletes and the general population are lacking, so it may be useful to compare elite athletes who compete in cold-environment sports to non-athletes who work in a low temperature setting.
Our meta-analysis of various pharmacological treatments used to control the symptoms of EIB showed that MCSA provides the greatest relief following periods of strenuous exercise because it causes the maximum fall in %FEV. Due to the drug’s direct impact on cells that cause inflammation, those with EIB may find that MCSA can control their symptoms to a greater extent than their current medication. Our findings will help enrich understanding of not only the biological mechanisms that cause EIB, but also how these three treatment options interact with those mechanisms. Continued research will hopefully lead to an effective and convenient method to prevent or minimize the symptoms of EIB, a condition affecting a large percentage of the general public.
- Boulet LP, O’Byrne PM. Asthma and exercise-induced bronchoconstriction in athletes. New England Journal of Medicine. 2015; 372(7):641-648.
- Parsons JPM, Mastronarde JGM, FCCP. Exercise-induced bronchoconstriction in athletes. Chest Journal. 2005; p. 3996-3974.
- Weiler JM, Brannan JD, Randolph CC, Hallstrand TS, Parsons J, Silvers W, et al. Exercise-induced bronchoconstriction update-2016. Journal of Allergy and Clinical Immunology. 2016; 138(5):1292.
- Hermansen CL, Kirchner JT. Identifying exercise-induced bronchospasm – Treatment hinges on distinguishing it from chronic asthma. Physician and Sportsmedicine. 2005; 33(12):25-30.
- Gerber AN. Measuring safety of inhaled corticosteroids in asthma. Annals of Allergy Asthma & Immunology. 2016; 117(6):577-581.
- Patient Information Singulair [Internet]. 2016. Merck Sharp & Dohme Corp; [cited 2017 Mar 3] Available from http://www.merck.com/product/usa/pi_circulars/s/singulair/singulair_ppi.pdf
- Murphy S. Cromolyn sodium: basic mechanisms and clinical usage. Pediatr Asthma Allergy Immunol 1988; 2: 237-54
- Intal Inhaler package insert (RPR—US), Rev 9/97, Rec 2/99, 2016.
- PDR Physicians’ Desk Reference. 53rd Montvale, NJ: Medical Economics Company Inc; 1999; 53: 2589-91
- Arnold DH, Gebretsadik T, Abramo TJ, Hartert TV. Noninvasive testing of lung function and inflammation in pediatric patients with acute asthma exacerbations. Journal of Asthma. 2012; 49(1):29-35.
- Kazani S, Sadeh J, Bunga S, Wechsler ME, Israel E. Cysteinyl leukotriene antagonism inhibits bronchoconstriction in response to hypertonic saline inhalation in asthma. Respiratory Medicine. 2011; 105(5):667-673.
- Piliponsky AM, Gleich GJ, Bar I, Levi-Schaffer F. Effects of eosinophils on mast cells: a new pathway for the perpetuation of allergic inflammation. Molecular Immunology. 2002; 38(16-18):1369-1372.